diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,350034 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.4340240102082447, + "eval_steps": 0, + "global_step": 50000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 8.680480204164894e-06, + "grad_norm": 15.5625, + "learning_rate": 0.0, + "loss": 1.2891, + "step": 1 + }, + { + "epoch": 1.736096040832979e-05, + "grad_norm": 22.625, + "learning_rate": 2e-06, + "loss": 1.6406, + "step": 2 + }, + { + "epoch": 2.6041440612494685e-05, + "grad_norm": 10.75, + "learning_rate": 4e-06, + "loss": 1.1016, + "step": 3 + }, + { + "epoch": 3.472192081665958e-05, + "grad_norm": 10.6875, + "learning_rate": 6e-06, + "loss": 0.9922, + "step": 4 + }, + { + "epoch": 4.3402401020824474e-05, + "grad_norm": 18.125, + "learning_rate": 8e-06, + "loss": 1.4375, + "step": 5 + }, + { + "epoch": 5.208288122498937e-05, + "grad_norm": 10.1875, + "learning_rate": 1e-05, + "loss": 1.0, + "step": 6 + }, + { + "epoch": 6.076336142915426e-05, + "grad_norm": 12.9375, + "learning_rate": 1.2e-05, + "loss": 1.2344, + "step": 7 + }, + { + "epoch": 6.944384163331915e-05, + "grad_norm": 15.8125, + "learning_rate": 1.4e-05, + "loss": 1.1875, + "step": 8 + }, + { + "epoch": 7.812432183748404e-05, + "grad_norm": 13.3125, + "learning_rate": 1.6e-05, + "loss": 1.0469, + "step": 9 + }, + { + "epoch": 8.680480204164895e-05, + "grad_norm": 11.375, + "learning_rate": 1.8e-05, + "loss": 0.9844, + "step": 10 + }, + { + "epoch": 9.548528224581384e-05, + "grad_norm": 11.125, + "learning_rate": 2e-05, + "loss": 1.0391, + "step": 11 + }, + { + "epoch": 0.00010416576244997874, + "grad_norm": 12.0625, + "learning_rate": 2.2e-05, + "loss": 1.0469, + "step": 12 + }, + { + "epoch": 0.00011284624265414363, + "grad_norm": 7.25, + "learning_rate": 2.4e-05, + "loss": 0.8203, + "step": 13 + }, + { + "epoch": 0.00012152672285830852, + "grad_norm": 18.5, + "learning_rate": 2.6e-05, + "loss": 1.0312, + "step": 14 + }, + { + "epoch": 0.0001302072030624734, + "grad_norm": 7.4375, + "learning_rate": 2.8e-05, + "loss": 0.9141, + "step": 15 + }, + { + "epoch": 0.0001388876832666383, + "grad_norm": 4.5625, + "learning_rate": 3e-05, + "loss": 0.7305, + "step": 16 + }, + { + "epoch": 0.0001475681634708032, + "grad_norm": 4.15625, + "learning_rate": 3.2e-05, + "loss": 0.7031, + "step": 17 + }, + { + "epoch": 0.0001562486436749681, + "grad_norm": 4.6875, + "learning_rate": 3.4000000000000007e-05, + "loss": 0.5977, + "step": 18 + }, + { + "epoch": 0.000164929123879133, + "grad_norm": 3.75, + "learning_rate": 3.6e-05, + "loss": 0.625, + "step": 19 + }, + { + "epoch": 0.0001736096040832979, + "grad_norm": 2.65625, + "learning_rate": 3.8e-05, + "loss": 0.5898, + "step": 20 + }, + { + "epoch": 0.00018229008428746277, + "grad_norm": 2.25, + "learning_rate": 4e-05, + "loss": 0.582, + "step": 21 + }, + { + "epoch": 0.00019097056449162767, + "grad_norm": 2.65625, + "learning_rate": 4.2000000000000004e-05, + "loss": 0.5977, + "step": 22 + }, + { + "epoch": 0.00019965104469579258, + "grad_norm": 2.09375, + "learning_rate": 4.4e-05, + "loss": 0.4551, + "step": 23 + }, + { + "epoch": 0.00020833152489995748, + "grad_norm": 1.625, + "learning_rate": 4.6e-05, + "loss": 0.4844, + "step": 24 + }, + { + "epoch": 0.00021701200510412235, + "grad_norm": 1.7421875, + "learning_rate": 4.8e-05, + "loss": 0.5156, + "step": 25 + }, + { + "epoch": 0.00022569248530828726, + "grad_norm": 3.703125, + "learning_rate": 5e-05, + "loss": 0.4688, + "step": 26 + }, + { + "epoch": 0.00023437296551245216, + "grad_norm": 1.296875, + "learning_rate": 5.2e-05, + "loss": 0.457, + "step": 27 + }, + { + "epoch": 0.00024305344571661704, + "grad_norm": 2.0625, + "learning_rate": 5.4e-05, + "loss": 0.5938, + "step": 28 + }, + { + "epoch": 0.0002517339259207819, + "grad_norm": 1.2890625, + "learning_rate": 5.6e-05, + "loss": 0.4648, + "step": 29 + }, + { + "epoch": 0.0002604144061249468, + "grad_norm": 1.125, + "learning_rate": 5.800000000000001e-05, + "loss": 0.4805, + "step": 30 + }, + { + "epoch": 0.0002690948863291117, + "grad_norm": 1.265625, + "learning_rate": 6e-05, + "loss": 0.5234, + "step": 31 + }, + { + "epoch": 0.0002777753665332766, + "grad_norm": 1.0078125, + "learning_rate": 6.2e-05, + "loss": 0.4297, + "step": 32 + }, + { + "epoch": 0.0002864558467374415, + "grad_norm": 1.0078125, + "learning_rate": 6.4e-05, + "loss": 0.3555, + "step": 33 + }, + { + "epoch": 0.0002951363269416064, + "grad_norm": 0.984375, + "learning_rate": 6.6e-05, + "loss": 0.3633, + "step": 34 + }, + { + "epoch": 0.00030381680714577133, + "grad_norm": 0.78125, + "learning_rate": 6.800000000000001e-05, + "loss": 0.3398, + "step": 35 + }, + { + "epoch": 0.0003124972873499362, + "grad_norm": 0.87890625, + "learning_rate": 7.000000000000001e-05, + "loss": 0.3574, + "step": 36 + }, + { + "epoch": 0.0003211777675541011, + "grad_norm": 0.6015625, + "learning_rate": 7.2e-05, + "loss": 0.3145, + "step": 37 + }, + { + "epoch": 0.000329858247758266, + "grad_norm": 0.6875, + "learning_rate": 7.4e-05, + "loss": 0.3906, + "step": 38 + }, + { + "epoch": 0.0003385387279624309, + "grad_norm": 0.73046875, + "learning_rate": 7.6e-05, + "loss": 0.3477, + "step": 39 + }, + { + "epoch": 0.0003472192081665958, + "grad_norm": 0.71484375, + "learning_rate": 7.8e-05, + "loss": 0.3477, + "step": 40 + }, + { + "epoch": 0.0003558996883707607, + "grad_norm": 0.59765625, + "learning_rate": 8e-05, + "loss": 0.3711, + "step": 41 + }, + { + "epoch": 0.00036458016857492554, + "grad_norm": 0.5546875, + "learning_rate": 8.2e-05, + "loss": 0.3398, + "step": 42 + }, + { + "epoch": 0.00037326064877909044, + "grad_norm": 0.5703125, + "learning_rate": 8.400000000000001e-05, + "loss": 0.3145, + "step": 43 + }, + { + "epoch": 0.00038194112898325535, + "grad_norm": 0.55078125, + "learning_rate": 8.599999999999999e-05, + "loss": 0.332, + "step": 44 + }, + { + "epoch": 0.00039062160918742025, + "grad_norm": 0.671875, + "learning_rate": 8.8e-05, + "loss": 0.4297, + "step": 45 + }, + { + "epoch": 0.00039930208939158515, + "grad_norm": 0.68359375, + "learning_rate": 8.999999999999999e-05, + "loss": 0.3945, + "step": 46 + }, + { + "epoch": 0.00040798256959575005, + "grad_norm": 0.482421875, + "learning_rate": 9.2e-05, + "loss": 0.3066, + "step": 47 + }, + { + "epoch": 0.00041666304979991496, + "grad_norm": 0.5703125, + "learning_rate": 9.400000000000001e-05, + "loss": 0.2988, + "step": 48 + }, + { + "epoch": 0.0004253435300040798, + "grad_norm": 0.4609375, + "learning_rate": 9.6e-05, + "loss": 0.3203, + "step": 49 + }, + { + "epoch": 0.0004340240102082447, + "grad_norm": 0.6015625, + "learning_rate": 9.800000000000001e-05, + "loss": 0.3828, + "step": 50 + }, + { + "epoch": 0.0004427044904124096, + "grad_norm": 0.48828125, + "learning_rate": 0.0001, + "loss": 0.2773, + "step": 51 + }, + { + "epoch": 0.0004513849706165745, + "grad_norm": 0.392578125, + "learning_rate": 0.000102, + "loss": 0.3672, + "step": 52 + }, + { + "epoch": 0.0004600654508207394, + "grad_norm": 0.5859375, + "learning_rate": 0.000104, + "loss": 0.3965, + "step": 53 + }, + { + "epoch": 0.0004687459310249043, + "grad_norm": 0.70703125, + "learning_rate": 0.000106, + "loss": 0.6133, + "step": 54 + }, + { + "epoch": 0.00047742641122906917, + "grad_norm": 0.357421875, + "learning_rate": 0.000108, + "loss": 0.3184, + "step": 55 + }, + { + "epoch": 0.00048610689143323407, + "grad_norm": 0.5703125, + "learning_rate": 0.00011, + "loss": 0.4648, + "step": 56 + }, + { + "epoch": 0.000494787371637399, + "grad_norm": 0.408203125, + "learning_rate": 0.000112, + "loss": 0.3496, + "step": 57 + }, + { + "epoch": 0.0005034678518415638, + "grad_norm": 0.4375, + "learning_rate": 0.000114, + "loss": 0.3008, + "step": 58 + }, + { + "epoch": 0.0005121483320457287, + "grad_norm": 0.86328125, + "learning_rate": 0.00011600000000000001, + "loss": 0.3867, + "step": 59 + }, + { + "epoch": 0.0005208288122498936, + "grad_norm": 0.455078125, + "learning_rate": 0.000118, + "loss": 0.3359, + "step": 60 + }, + { + "epoch": 0.0005295092924540585, + "grad_norm": 0.412109375, + "learning_rate": 0.00012, + "loss": 0.3926, + "step": 61 + }, + { + "epoch": 0.0005381897726582234, + "grad_norm": 0.361328125, + "learning_rate": 0.000122, + "loss": 0.375, + "step": 62 + }, + { + "epoch": 0.0005468702528623883, + "grad_norm": 0.51171875, + "learning_rate": 0.000124, + "loss": 0.3574, + "step": 63 + }, + { + "epoch": 0.0005555507330665532, + "grad_norm": 0.3125, + "learning_rate": 0.000126, + "loss": 0.2695, + "step": 64 + }, + { + "epoch": 0.0005642312132707181, + "grad_norm": 0.408203125, + "learning_rate": 0.000128, + "loss": 0.3594, + "step": 65 + }, + { + "epoch": 0.000572911693474883, + "grad_norm": 0.3984375, + "learning_rate": 0.00013000000000000002, + "loss": 0.3438, + "step": 66 + }, + { + "epoch": 0.000581592173679048, + "grad_norm": 0.369140625, + "learning_rate": 0.000132, + "loss": 0.3047, + "step": 67 + }, + { + "epoch": 0.0005902726538832129, + "grad_norm": 0.306640625, + "learning_rate": 0.000134, + "loss": 0.2773, + "step": 68 + }, + { + "epoch": 0.0005989531340873778, + "grad_norm": 0.443359375, + "learning_rate": 0.00013600000000000003, + "loss": 0.3242, + "step": 69 + }, + { + "epoch": 0.0006076336142915427, + "grad_norm": 0.380859375, + "learning_rate": 0.00013800000000000002, + "loss": 0.3203, + "step": 70 + }, + { + "epoch": 0.0006163140944957075, + "grad_norm": 0.314453125, + "learning_rate": 0.00014000000000000001, + "loss": 0.3047, + "step": 71 + }, + { + "epoch": 0.0006249945746998724, + "grad_norm": 0.5703125, + "learning_rate": 0.00014199999999999998, + "loss": 0.3887, + "step": 72 + }, + { + "epoch": 0.0006336750549040373, + "grad_norm": 0.3828125, + "learning_rate": 0.000144, + "loss": 0.332, + "step": 73 + }, + { + "epoch": 0.0006423555351082022, + "grad_norm": 0.35546875, + "learning_rate": 0.000146, + "loss": 0.3203, + "step": 74 + }, + { + "epoch": 0.0006510360153123671, + "grad_norm": 0.369140625, + "learning_rate": 0.000148, + "loss": 0.3555, + "step": 75 + }, + { + "epoch": 0.000659716495516532, + "grad_norm": 0.4765625, + "learning_rate": 0.00015, + "loss": 0.377, + "step": 76 + }, + { + "epoch": 0.0006683969757206969, + "grad_norm": 0.376953125, + "learning_rate": 0.000152, + "loss": 0.2656, + "step": 77 + }, + { + "epoch": 0.0006770774559248618, + "grad_norm": 0.484375, + "learning_rate": 0.000154, + "loss": 0.4102, + "step": 78 + }, + { + "epoch": 0.0006857579361290267, + "grad_norm": 0.46875, + "learning_rate": 0.000156, + "loss": 0.2949, + "step": 79 + }, + { + "epoch": 0.0006944384163331916, + "grad_norm": 0.306640625, + "learning_rate": 0.000158, + "loss": 0.3906, + "step": 80 + }, + { + "epoch": 0.0007031188965373565, + "grad_norm": 0.294921875, + "learning_rate": 0.00016, + "loss": 0.332, + "step": 81 + }, + { + "epoch": 0.0007117993767415214, + "grad_norm": 0.65234375, + "learning_rate": 0.000162, + "loss": 0.3164, + "step": 82 + }, + { + "epoch": 0.0007204798569456863, + "grad_norm": 0.296875, + "learning_rate": 0.000164, + "loss": 0.3535, + "step": 83 + }, + { + "epoch": 0.0007291603371498511, + "grad_norm": 0.400390625, + "learning_rate": 0.00016600000000000002, + "loss": 0.3438, + "step": 84 + }, + { + "epoch": 0.000737840817354016, + "grad_norm": 0.314453125, + "learning_rate": 0.00016800000000000002, + "loss": 0.3066, + "step": 85 + }, + { + "epoch": 0.0007465212975581809, + "grad_norm": 0.318359375, + "learning_rate": 0.00017, + "loss": 0.3262, + "step": 86 + }, + { + "epoch": 0.0007552017777623458, + "grad_norm": 0.404296875, + "learning_rate": 0.00017199999999999998, + "loss": 0.334, + "step": 87 + }, + { + "epoch": 0.0007638822579665107, + "grad_norm": 0.4296875, + "learning_rate": 0.000174, + "loss": 0.3242, + "step": 88 + }, + { + "epoch": 0.0007725627381706756, + "grad_norm": 0.310546875, + "learning_rate": 0.000176, + "loss": 0.2539, + "step": 89 + }, + { + "epoch": 0.0007812432183748405, + "grad_norm": 0.271484375, + "learning_rate": 0.000178, + "loss": 0.3184, + "step": 90 + }, + { + "epoch": 0.0007899236985790054, + "grad_norm": 0.34375, + "learning_rate": 0.00017999999999999998, + "loss": 0.3398, + "step": 91 + }, + { + "epoch": 0.0007986041787831703, + "grad_norm": 0.287109375, + "learning_rate": 0.000182, + "loss": 0.3516, + "step": 92 + }, + { + "epoch": 0.0008072846589873352, + "grad_norm": 0.328125, + "learning_rate": 0.000184, + "loss": 0.2852, + "step": 93 + }, + { + "epoch": 0.0008159651391915001, + "grad_norm": 0.357421875, + "learning_rate": 0.000186, + "loss": 0.3184, + "step": 94 + }, + { + "epoch": 0.000824645619395665, + "grad_norm": 0.404296875, + "learning_rate": 0.00018800000000000002, + "loss": 0.2871, + "step": 95 + }, + { + "epoch": 0.0008333260995998299, + "grad_norm": 0.267578125, + "learning_rate": 0.00019, + "loss": 0.2637, + "step": 96 + }, + { + "epoch": 0.0008420065798039947, + "grad_norm": 0.318359375, + "learning_rate": 0.000192, + "loss": 0.2559, + "step": 97 + }, + { + "epoch": 0.0008506870600081596, + "grad_norm": 0.36328125, + "learning_rate": 0.000194, + "loss": 0.4199, + "step": 98 + }, + { + "epoch": 0.0008593675402123245, + "grad_norm": 0.5703125, + "learning_rate": 0.00019600000000000002, + "loss": 0.3438, + "step": 99 + }, + { + "epoch": 0.0008680480204164894, + "grad_norm": 0.38671875, + "learning_rate": 0.00019800000000000002, + "loss": 0.3398, + "step": 100 + }, + { + "epoch": 0.0008767285006206543, + "grad_norm": 0.625, + "learning_rate": 0.0002, + "loss": 0.3555, + "step": 101 + }, + { + "epoch": 0.0008854089808248192, + "grad_norm": 0.3359375, + "learning_rate": 0.000202, + "loss": 0.249, + "step": 102 + }, + { + "epoch": 0.0008940894610289841, + "grad_norm": 0.6015625, + "learning_rate": 0.000204, + "loss": 0.3535, + "step": 103 + }, + { + "epoch": 0.000902769941233149, + "grad_norm": 0.314453125, + "learning_rate": 0.000206, + "loss": 0.3887, + "step": 104 + }, + { + "epoch": 0.0009114504214373139, + "grad_norm": 0.255859375, + "learning_rate": 0.000208, + "loss": 0.2383, + "step": 105 + }, + { + "epoch": 0.0009201309016414788, + "grad_norm": 0.314453125, + "learning_rate": 0.00021, + "loss": 0.3086, + "step": 106 + }, + { + "epoch": 0.0009288113818456437, + "grad_norm": 0.375, + "learning_rate": 0.000212, + "loss": 0.3809, + "step": 107 + }, + { + "epoch": 0.0009374918620498086, + "grad_norm": 0.333984375, + "learning_rate": 0.000214, + "loss": 0.2969, + "step": 108 + }, + { + "epoch": 0.0009461723422539735, + "grad_norm": 0.291015625, + "learning_rate": 0.000216, + "loss": 0.3691, + "step": 109 + }, + { + "epoch": 0.0009548528224581383, + "grad_norm": 0.271484375, + "learning_rate": 0.000218, + "loss": 0.332, + "step": 110 + }, + { + "epoch": 0.0009635333026623032, + "grad_norm": 0.400390625, + "learning_rate": 0.00022, + "loss": 0.3828, + "step": 111 + }, + { + "epoch": 0.0009722137828664681, + "grad_norm": 0.263671875, + "learning_rate": 0.000222, + "loss": 0.4023, + "step": 112 + }, + { + "epoch": 0.0009808942630706332, + "grad_norm": 0.333984375, + "learning_rate": 0.000224, + "loss": 0.3145, + "step": 113 + }, + { + "epoch": 0.000989574743274798, + "grad_norm": 0.3203125, + "learning_rate": 0.00022600000000000002, + "loss": 0.3945, + "step": 114 + }, + { + "epoch": 0.000998255223478963, + "grad_norm": 0.23046875, + "learning_rate": 0.000228, + "loss": 0.2969, + "step": 115 + }, + { + "epoch": 0.0010069357036831276, + "grad_norm": 0.32421875, + "learning_rate": 0.00023, + "loss": 0.3438, + "step": 116 + }, + { + "epoch": 0.0010156161838872925, + "grad_norm": 0.59375, + "learning_rate": 0.00023200000000000003, + "loss": 0.6211, + "step": 117 + }, + { + "epoch": 0.0010242966640914575, + "grad_norm": 0.35546875, + "learning_rate": 0.00023400000000000002, + "loss": 0.3145, + "step": 118 + }, + { + "epoch": 0.0010329771442956224, + "grad_norm": 0.30859375, + "learning_rate": 0.000236, + "loss": 0.3477, + "step": 119 + }, + { + "epoch": 0.0010416576244997873, + "grad_norm": 0.265625, + "learning_rate": 0.00023799999999999998, + "loss": 0.3789, + "step": 120 + }, + { + "epoch": 0.0010503381047039522, + "grad_norm": 0.21875, + "learning_rate": 0.00024, + "loss": 0.2422, + "step": 121 + }, + { + "epoch": 0.001059018584908117, + "grad_norm": 0.345703125, + "learning_rate": 0.000242, + "loss": 0.2988, + "step": 122 + }, + { + "epoch": 0.001067699065112282, + "grad_norm": 0.318359375, + "learning_rate": 0.000244, + "loss": 0.2949, + "step": 123 + }, + { + "epoch": 0.0010763795453164469, + "grad_norm": 0.2392578125, + "learning_rate": 0.000246, + "loss": 0.2451, + "step": 124 + }, + { + "epoch": 0.0010850600255206118, + "grad_norm": 0.296875, + "learning_rate": 0.000248, + "loss": 0.291, + "step": 125 + }, + { + "epoch": 0.0010937405057247767, + "grad_norm": 0.30078125, + "learning_rate": 0.00025, + "loss": 0.3516, + "step": 126 + }, + { + "epoch": 0.0011024209859289416, + "grad_norm": 0.296875, + "learning_rate": 0.000252, + "loss": 0.3301, + "step": 127 + }, + { + "epoch": 0.0011111014661331065, + "grad_norm": 0.251953125, + "learning_rate": 0.000254, + "loss": 0.252, + "step": 128 + }, + { + "epoch": 0.0011197819463372714, + "grad_norm": 0.228515625, + "learning_rate": 0.000256, + "loss": 0.2539, + "step": 129 + }, + { + "epoch": 0.0011284624265414363, + "grad_norm": 0.76953125, + "learning_rate": 0.00025800000000000004, + "loss": 0.3535, + "step": 130 + }, + { + "epoch": 0.0011371429067456012, + "grad_norm": 0.296875, + "learning_rate": 0.00026000000000000003, + "loss": 0.3242, + "step": 131 + }, + { + "epoch": 0.001145823386949766, + "grad_norm": 0.380859375, + "learning_rate": 0.000262, + "loss": 0.3184, + "step": 132 + }, + { + "epoch": 0.001154503867153931, + "grad_norm": 0.427734375, + "learning_rate": 0.000264, + "loss": 0.5039, + "step": 133 + }, + { + "epoch": 0.001163184347358096, + "grad_norm": 0.27734375, + "learning_rate": 0.000266, + "loss": 0.2656, + "step": 134 + }, + { + "epoch": 0.0011718648275622608, + "grad_norm": 0.314453125, + "learning_rate": 0.000268, + "loss": 0.3652, + "step": 135 + }, + { + "epoch": 0.0011805453077664257, + "grad_norm": 0.255859375, + "learning_rate": 0.00027, + "loss": 0.3086, + "step": 136 + }, + { + "epoch": 0.0011892257879705906, + "grad_norm": 0.275390625, + "learning_rate": 0.00027200000000000005, + "loss": 0.3145, + "step": 137 + }, + { + "epoch": 0.0011979062681747555, + "grad_norm": 0.353515625, + "learning_rate": 0.00027400000000000005, + "loss": 0.4375, + "step": 138 + }, + { + "epoch": 0.0012065867483789204, + "grad_norm": 0.2578125, + "learning_rate": 0.00027600000000000004, + "loss": 0.3203, + "step": 139 + }, + { + "epoch": 0.0012152672285830853, + "grad_norm": 0.2275390625, + "learning_rate": 0.00027800000000000004, + "loss": 0.2949, + "step": 140 + }, + { + "epoch": 0.0012239477087872502, + "grad_norm": 0.322265625, + "learning_rate": 0.00028000000000000003, + "loss": 0.3613, + "step": 141 + }, + { + "epoch": 0.001232628188991415, + "grad_norm": 0.337890625, + "learning_rate": 0.00028199999999999997, + "loss": 0.3828, + "step": 142 + }, + { + "epoch": 0.0012413086691955798, + "grad_norm": 0.21484375, + "learning_rate": 0.00028399999999999996, + "loss": 0.3125, + "step": 143 + }, + { + "epoch": 0.0012499891493997447, + "grad_norm": 0.248046875, + "learning_rate": 0.00028599999999999996, + "loss": 0.2773, + "step": 144 + }, + { + "epoch": 0.0012586696296039096, + "grad_norm": 0.296875, + "learning_rate": 0.000288, + "loss": 0.293, + "step": 145 + }, + { + "epoch": 0.0012673501098080745, + "grad_norm": 0.2333984375, + "learning_rate": 0.00029, + "loss": 0.2461, + "step": 146 + }, + { + "epoch": 0.0012760305900122394, + "grad_norm": 0.2255859375, + "learning_rate": 0.000292, + "loss": 0.2461, + "step": 147 + }, + { + "epoch": 0.0012847110702164043, + "grad_norm": 0.27734375, + "learning_rate": 0.000294, + "loss": 0.3887, + "step": 148 + }, + { + "epoch": 0.0012933915504205692, + "grad_norm": 0.2373046875, + "learning_rate": 0.000296, + "loss": 0.3105, + "step": 149 + }, + { + "epoch": 0.0013020720306247341, + "grad_norm": 0.1806640625, + "learning_rate": 0.000298, + "loss": 0.2402, + "step": 150 + }, + { + "epoch": 0.001310752510828899, + "grad_norm": 0.255859375, + "learning_rate": 0.0003, + "loss": 0.2734, + "step": 151 + }, + { + "epoch": 0.001319432991033064, + "grad_norm": 0.259765625, + "learning_rate": 0.000302, + "loss": 0.3125, + "step": 152 + }, + { + "epoch": 0.0013281134712372288, + "grad_norm": 0.251953125, + "learning_rate": 0.000304, + "loss": 0.3223, + "step": 153 + }, + { + "epoch": 0.0013367939514413937, + "grad_norm": 0.23828125, + "learning_rate": 0.000306, + "loss": 0.2695, + "step": 154 + }, + { + "epoch": 0.0013454744316455586, + "grad_norm": 0.19921875, + "learning_rate": 0.000308, + "loss": 0.2441, + "step": 155 + }, + { + "epoch": 0.0013541549118497235, + "grad_norm": 0.244140625, + "learning_rate": 0.00031, + "loss": 0.375, + "step": 156 + }, + { + "epoch": 0.0013628353920538884, + "grad_norm": 0.3125, + "learning_rate": 0.000312, + "loss": 0.3242, + "step": 157 + }, + { + "epoch": 0.0013715158722580533, + "grad_norm": 0.24609375, + "learning_rate": 0.000314, + "loss": 0.293, + "step": 158 + }, + { + "epoch": 0.0013801963524622182, + "grad_norm": 0.2421875, + "learning_rate": 0.000316, + "loss": 0.3027, + "step": 159 + }, + { + "epoch": 0.0013888768326663832, + "grad_norm": 0.21484375, + "learning_rate": 0.00031800000000000003, + "loss": 0.2217, + "step": 160 + }, + { + "epoch": 0.001397557312870548, + "grad_norm": 0.265625, + "learning_rate": 0.00032, + "loss": 0.3691, + "step": 161 + }, + { + "epoch": 0.001406237793074713, + "grad_norm": 0.29296875, + "learning_rate": 0.000322, + "loss": 0.2891, + "step": 162 + }, + { + "epoch": 0.0014149182732788779, + "grad_norm": 0.2353515625, + "learning_rate": 0.000324, + "loss": 0.3301, + "step": 163 + }, + { + "epoch": 0.0014235987534830428, + "grad_norm": 0.2421875, + "learning_rate": 0.000326, + "loss": 0.334, + "step": 164 + }, + { + "epoch": 0.0014322792336872077, + "grad_norm": 0.23828125, + "learning_rate": 0.000328, + "loss": 0.2988, + "step": 165 + }, + { + "epoch": 0.0014409597138913726, + "grad_norm": 0.2177734375, + "learning_rate": 0.00033, + "loss": 0.2578, + "step": 166 + }, + { + "epoch": 0.0014496401940955375, + "grad_norm": 0.2734375, + "learning_rate": 0.00033200000000000005, + "loss": 0.3008, + "step": 167 + }, + { + "epoch": 0.0014583206742997022, + "grad_norm": 0.25, + "learning_rate": 0.00033400000000000004, + "loss": 0.293, + "step": 168 + }, + { + "epoch": 0.001467001154503867, + "grad_norm": 0.255859375, + "learning_rate": 0.00033600000000000004, + "loss": 0.2832, + "step": 169 + }, + { + "epoch": 0.001475681634708032, + "grad_norm": 0.25390625, + "learning_rate": 0.00033800000000000003, + "loss": 0.2676, + "step": 170 + }, + { + "epoch": 0.0014843621149121969, + "grad_norm": 0.2578125, + "learning_rate": 0.00034, + "loss": 0.2734, + "step": 171 + }, + { + "epoch": 0.0014930425951163618, + "grad_norm": 0.251953125, + "learning_rate": 0.000342, + "loss": 0.3125, + "step": 172 + }, + { + "epoch": 0.0015017230753205267, + "grad_norm": 0.333984375, + "learning_rate": 0.00034399999999999996, + "loss": 0.3008, + "step": 173 + }, + { + "epoch": 0.0015104035555246916, + "grad_norm": 0.2890625, + "learning_rate": 0.000346, + "loss": 0.332, + "step": 174 + }, + { + "epoch": 0.0015190840357288565, + "grad_norm": 0.26953125, + "learning_rate": 0.000348, + "loss": 0.2539, + "step": 175 + }, + { + "epoch": 0.0015277645159330214, + "grad_norm": 0.478515625, + "learning_rate": 0.00035, + "loss": 0.3203, + "step": 176 + }, + { + "epoch": 0.0015364449961371863, + "grad_norm": 0.302734375, + "learning_rate": 0.000352, + "loss": 0.3066, + "step": 177 + }, + { + "epoch": 0.0015451254763413512, + "grad_norm": 0.330078125, + "learning_rate": 0.000354, + "loss": 0.3047, + "step": 178 + }, + { + "epoch": 0.001553805956545516, + "grad_norm": 0.265625, + "learning_rate": 0.000356, + "loss": 0.3184, + "step": 179 + }, + { + "epoch": 0.001562486436749681, + "grad_norm": 0.251953125, + "learning_rate": 0.000358, + "loss": 0.3105, + "step": 180 + }, + { + "epoch": 0.001571166916953846, + "grad_norm": 0.3671875, + "learning_rate": 0.00035999999999999997, + "loss": 0.3125, + "step": 181 + }, + { + "epoch": 0.0015798473971580108, + "grad_norm": 0.201171875, + "learning_rate": 0.000362, + "loss": 0.3223, + "step": 182 + }, + { + "epoch": 0.0015885278773621757, + "grad_norm": 0.3671875, + "learning_rate": 0.000364, + "loss": 0.4199, + "step": 183 + }, + { + "epoch": 0.0015972083575663406, + "grad_norm": 0.248046875, + "learning_rate": 0.000366, + "loss": 0.3281, + "step": 184 + }, + { + "epoch": 0.0016058888377705055, + "grad_norm": 0.228515625, + "learning_rate": 0.000368, + "loss": 0.2891, + "step": 185 + }, + { + "epoch": 0.0016145693179746704, + "grad_norm": 0.23828125, + "learning_rate": 0.00037, + "loss": 0.3555, + "step": 186 + }, + { + "epoch": 0.0016232497981788353, + "grad_norm": 0.294921875, + "learning_rate": 0.000372, + "loss": 0.3457, + "step": 187 + }, + { + "epoch": 0.0016319302783830002, + "grad_norm": 0.2578125, + "learning_rate": 0.000374, + "loss": 0.2988, + "step": 188 + }, + { + "epoch": 0.0016406107585871651, + "grad_norm": 0.25390625, + "learning_rate": 0.00037600000000000003, + "loss": 0.3242, + "step": 189 + }, + { + "epoch": 0.00164929123879133, + "grad_norm": 0.27734375, + "learning_rate": 0.000378, + "loss": 0.2949, + "step": 190 + }, + { + "epoch": 0.001657971718995495, + "grad_norm": 0.2578125, + "learning_rate": 0.00038, + "loss": 0.3672, + "step": 191 + }, + { + "epoch": 0.0016666521991996598, + "grad_norm": 0.19921875, + "learning_rate": 0.000382, + "loss": 0.3047, + "step": 192 + }, + { + "epoch": 0.0016753326794038245, + "grad_norm": 0.16796875, + "learning_rate": 0.000384, + "loss": 0.2715, + "step": 193 + }, + { + "epoch": 0.0016840131596079894, + "grad_norm": 0.2734375, + "learning_rate": 0.000386, + "loss": 0.3086, + "step": 194 + }, + { + "epoch": 0.0016926936398121543, + "grad_norm": 0.28515625, + "learning_rate": 0.000388, + "loss": 0.2617, + "step": 195 + }, + { + "epoch": 0.0017013741200163192, + "grad_norm": 0.2412109375, + "learning_rate": 0.00039000000000000005, + "loss": 0.3262, + "step": 196 + }, + { + "epoch": 0.0017100546002204841, + "grad_norm": 0.2373046875, + "learning_rate": 0.00039200000000000004, + "loss": 0.252, + "step": 197 + }, + { + "epoch": 0.001718735080424649, + "grad_norm": 0.228515625, + "learning_rate": 0.00039400000000000004, + "loss": 0.3047, + "step": 198 + }, + { + "epoch": 0.001727415560628814, + "grad_norm": 0.201171875, + "learning_rate": 0.00039600000000000003, + "loss": 0.332, + "step": 199 + }, + { + "epoch": 0.0017360960408329788, + "grad_norm": 0.2421875, + "learning_rate": 0.000398, + "loss": 0.2695, + "step": 200 + }, + { + "epoch": 0.0017447765210371437, + "grad_norm": 0.287109375, + "learning_rate": 0.0004, + "loss": 0.3867, + "step": 201 + }, + { + "epoch": 0.0017534570012413086, + "grad_norm": 0.203125, + "learning_rate": 0.000402, + "loss": 0.2598, + "step": 202 + }, + { + "epoch": 0.0017621374814454735, + "grad_norm": 0.2431640625, + "learning_rate": 0.000404, + "loss": 0.3203, + "step": 203 + }, + { + "epoch": 0.0017708179616496384, + "grad_norm": 0.251953125, + "learning_rate": 0.00040600000000000006, + "loss": 0.3438, + "step": 204 + }, + { + "epoch": 0.0017794984418538033, + "grad_norm": 0.1923828125, + "learning_rate": 0.000408, + "loss": 0.248, + "step": 205 + }, + { + "epoch": 0.0017881789220579682, + "grad_norm": 0.216796875, + "learning_rate": 0.00041, + "loss": 0.2852, + "step": 206 + }, + { + "epoch": 0.0017968594022621332, + "grad_norm": 0.287109375, + "learning_rate": 0.000412, + "loss": 0.3652, + "step": 207 + }, + { + "epoch": 0.001805539882466298, + "grad_norm": 0.265625, + "learning_rate": 0.000414, + "loss": 0.3359, + "step": 208 + }, + { + "epoch": 0.001814220362670463, + "grad_norm": 0.29296875, + "learning_rate": 0.000416, + "loss": 0.2773, + "step": 209 + }, + { + "epoch": 0.0018229008428746279, + "grad_norm": 0.2392578125, + "learning_rate": 0.00041799999999999997, + "loss": 0.3047, + "step": 210 + }, + { + "epoch": 0.0018315813230787928, + "grad_norm": 0.24609375, + "learning_rate": 0.00042, + "loss": 0.3613, + "step": 211 + }, + { + "epoch": 0.0018402618032829577, + "grad_norm": 0.2470703125, + "learning_rate": 0.000422, + "loss": 0.2676, + "step": 212 + }, + { + "epoch": 0.0018489422834871226, + "grad_norm": 0.2431640625, + "learning_rate": 0.000424, + "loss": 0.3633, + "step": 213 + }, + { + "epoch": 0.0018576227636912875, + "grad_norm": 0.5234375, + "learning_rate": 0.000426, + "loss": 0.4062, + "step": 214 + }, + { + "epoch": 0.0018663032438954524, + "grad_norm": 0.306640625, + "learning_rate": 0.000428, + "loss": 0.3223, + "step": 215 + }, + { + "epoch": 0.0018749837240996173, + "grad_norm": 0.2490234375, + "learning_rate": 0.00043, + "loss": 0.293, + "step": 216 + }, + { + "epoch": 0.0018836642043037822, + "grad_norm": 0.22265625, + "learning_rate": 0.000432, + "loss": 0.3086, + "step": 217 + }, + { + "epoch": 0.001892344684507947, + "grad_norm": 0.1904296875, + "learning_rate": 0.00043400000000000003, + "loss": 0.2656, + "step": 218 + }, + { + "epoch": 0.0019010251647121118, + "grad_norm": 0.1982421875, + "learning_rate": 0.000436, + "loss": 0.248, + "step": 219 + }, + { + "epoch": 0.0019097056449162767, + "grad_norm": 0.1728515625, + "learning_rate": 0.000438, + "loss": 0.3379, + "step": 220 + }, + { + "epoch": 0.0019183861251204416, + "grad_norm": 0.216796875, + "learning_rate": 0.00044, + "loss": 0.3301, + "step": 221 + }, + { + "epoch": 0.0019270666053246065, + "grad_norm": 0.1474609375, + "learning_rate": 0.000442, + "loss": 0.2812, + "step": 222 + }, + { + "epoch": 0.0019357470855287714, + "grad_norm": 0.1240234375, + "learning_rate": 0.000444, + "loss": 0.2188, + "step": 223 + }, + { + "epoch": 0.0019444275657329363, + "grad_norm": 0.21484375, + "learning_rate": 0.000446, + "loss": 0.3242, + "step": 224 + }, + { + "epoch": 0.0019531080459371012, + "grad_norm": 0.193359375, + "learning_rate": 0.000448, + "loss": 0.2773, + "step": 225 + }, + { + "epoch": 0.0019617885261412663, + "grad_norm": 0.224609375, + "learning_rate": 0.00045000000000000004, + "loss": 0.293, + "step": 226 + }, + { + "epoch": 0.001970469006345431, + "grad_norm": 0.259765625, + "learning_rate": 0.00045200000000000004, + "loss": 0.3281, + "step": 227 + }, + { + "epoch": 0.001979149486549596, + "grad_norm": 0.2412109375, + "learning_rate": 0.00045400000000000003, + "loss": 0.3184, + "step": 228 + }, + { + "epoch": 0.001987829966753761, + "grad_norm": 0.2333984375, + "learning_rate": 0.000456, + "loss": 0.3164, + "step": 229 + }, + { + "epoch": 0.001996510446957926, + "grad_norm": 0.1796875, + "learning_rate": 0.000458, + "loss": 0.2734, + "step": 230 + }, + { + "epoch": 0.0020051909271620906, + "grad_norm": 0.265625, + "learning_rate": 0.00046, + "loss": 0.4395, + "step": 231 + }, + { + "epoch": 0.0020138714073662553, + "grad_norm": 0.25, + "learning_rate": 0.000462, + "loss": 0.291, + "step": 232 + }, + { + "epoch": 0.0020225518875704204, + "grad_norm": 0.2119140625, + "learning_rate": 0.00046400000000000006, + "loss": 0.332, + "step": 233 + }, + { + "epoch": 0.002031232367774585, + "grad_norm": 0.275390625, + "learning_rate": 0.00046600000000000005, + "loss": 0.3457, + "step": 234 + }, + { + "epoch": 0.00203991284797875, + "grad_norm": 0.197265625, + "learning_rate": 0.00046800000000000005, + "loss": 0.3086, + "step": 235 + }, + { + "epoch": 0.002048593328182915, + "grad_norm": 0.283203125, + "learning_rate": 0.00047, + "loss": 0.4199, + "step": 236 + }, + { + "epoch": 0.00205727380838708, + "grad_norm": 0.26171875, + "learning_rate": 0.000472, + "loss": 0.3477, + "step": 237 + }, + { + "epoch": 0.0020659542885912447, + "grad_norm": 0.23046875, + "learning_rate": 0.000474, + "loss": 0.3145, + "step": 238 + }, + { + "epoch": 0.00207463476879541, + "grad_norm": 0.2197265625, + "learning_rate": 0.00047599999999999997, + "loss": 0.2617, + "step": 239 + }, + { + "epoch": 0.0020833152489995745, + "grad_norm": 0.1845703125, + "learning_rate": 0.00047799999999999996, + "loss": 0.3809, + "step": 240 + }, + { + "epoch": 0.0020919957292037396, + "grad_norm": 0.1748046875, + "learning_rate": 0.00048, + "loss": 0.2715, + "step": 241 + }, + { + "epoch": 0.0021006762094079043, + "grad_norm": 0.1923828125, + "learning_rate": 0.000482, + "loss": 0.2578, + "step": 242 + }, + { + "epoch": 0.0021093566896120694, + "grad_norm": 0.1943359375, + "learning_rate": 0.000484, + "loss": 0.2578, + "step": 243 + }, + { + "epoch": 0.002118037169816234, + "grad_norm": 0.33203125, + "learning_rate": 0.000486, + "loss": 0.2793, + "step": 244 + }, + { + "epoch": 0.0021267176500203992, + "grad_norm": 0.255859375, + "learning_rate": 0.000488, + "loss": 0.3828, + "step": 245 + }, + { + "epoch": 0.002135398130224564, + "grad_norm": 0.177734375, + "learning_rate": 0.00049, + "loss": 0.2695, + "step": 246 + }, + { + "epoch": 0.002144078610428729, + "grad_norm": 0.1748046875, + "learning_rate": 0.000492, + "loss": 0.2656, + "step": 247 + }, + { + "epoch": 0.0021527590906328937, + "grad_norm": 0.138671875, + "learning_rate": 0.000494, + "loss": 0.2432, + "step": 248 + }, + { + "epoch": 0.002161439570837059, + "grad_norm": 0.1650390625, + "learning_rate": 0.000496, + "loss": 0.3125, + "step": 249 + }, + { + "epoch": 0.0021701200510412235, + "grad_norm": 0.15234375, + "learning_rate": 0.000498, + "loss": 0.2695, + "step": 250 + }, + { + "epoch": 0.0021788005312453887, + "grad_norm": 0.1845703125, + "learning_rate": 0.0005, + "loss": 0.3184, + "step": 251 + }, + { + "epoch": 0.0021874810114495533, + "grad_norm": 0.2412109375, + "learning_rate": 0.0005020000000000001, + "loss": 0.3027, + "step": 252 + }, + { + "epoch": 0.0021961614916537185, + "grad_norm": 0.2158203125, + "learning_rate": 0.000504, + "loss": 0.2988, + "step": 253 + }, + { + "epoch": 0.002204841971857883, + "grad_norm": 0.193359375, + "learning_rate": 0.000506, + "loss": 0.293, + "step": 254 + }, + { + "epoch": 0.0022135224520620483, + "grad_norm": 0.1826171875, + "learning_rate": 0.000508, + "loss": 0.2656, + "step": 255 + }, + { + "epoch": 0.002222202932266213, + "grad_norm": 0.21875, + "learning_rate": 0.00051, + "loss": 0.3066, + "step": 256 + }, + { + "epoch": 0.002230883412470378, + "grad_norm": 0.185546875, + "learning_rate": 0.000512, + "loss": 0.3027, + "step": 257 + }, + { + "epoch": 0.0022395638926745428, + "grad_norm": 0.2275390625, + "learning_rate": 0.000514, + "loss": 0.2969, + "step": 258 + }, + { + "epoch": 0.0022482443728787074, + "grad_norm": 0.451171875, + "learning_rate": 0.0005160000000000001, + "loss": 0.3887, + "step": 259 + }, + { + "epoch": 0.0022569248530828726, + "grad_norm": 0.1689453125, + "learning_rate": 0.000518, + "loss": 0.2695, + "step": 260 + }, + { + "epoch": 0.0022656053332870373, + "grad_norm": 0.294921875, + "learning_rate": 0.0005200000000000001, + "loss": 0.2812, + "step": 261 + }, + { + "epoch": 0.0022742858134912024, + "grad_norm": 0.158203125, + "learning_rate": 0.000522, + "loss": 0.3105, + "step": 262 + }, + { + "epoch": 0.002282966293695367, + "grad_norm": 0.1845703125, + "learning_rate": 0.000524, + "loss": 0.3516, + "step": 263 + }, + { + "epoch": 0.002291646773899532, + "grad_norm": 0.236328125, + "learning_rate": 0.000526, + "loss": 0.2578, + "step": 264 + }, + { + "epoch": 0.002300327254103697, + "grad_norm": 0.23828125, + "learning_rate": 0.000528, + "loss": 0.252, + "step": 265 + }, + { + "epoch": 0.002309007734307862, + "grad_norm": 0.2080078125, + "learning_rate": 0.0005300000000000001, + "loss": 0.3516, + "step": 266 + }, + { + "epoch": 0.0023176882145120267, + "grad_norm": 0.224609375, + "learning_rate": 0.000532, + "loss": 0.2969, + "step": 267 + }, + { + "epoch": 0.002326368694716192, + "grad_norm": 0.1796875, + "learning_rate": 0.0005340000000000001, + "loss": 0.3203, + "step": 268 + }, + { + "epoch": 0.0023350491749203565, + "grad_norm": 0.68359375, + "learning_rate": 0.000536, + "loss": 0.3242, + "step": 269 + }, + { + "epoch": 0.0023437296551245216, + "grad_norm": 0.2138671875, + "learning_rate": 0.0005380000000000001, + "loss": 0.2832, + "step": 270 + }, + { + "epoch": 0.0023524101353286863, + "grad_norm": 0.1748046875, + "learning_rate": 0.00054, + "loss": 0.3145, + "step": 271 + }, + { + "epoch": 0.0023610906155328514, + "grad_norm": 0.2197265625, + "learning_rate": 0.0005420000000000001, + "loss": 0.2637, + "step": 272 + }, + { + "epoch": 0.002369771095737016, + "grad_norm": 0.34765625, + "learning_rate": 0.0005440000000000001, + "loss": 0.4414, + "step": 273 + }, + { + "epoch": 0.002378451575941181, + "grad_norm": 0.232421875, + "learning_rate": 0.000546, + "loss": 0.2695, + "step": 274 + }, + { + "epoch": 0.002387132056145346, + "grad_norm": 0.302734375, + "learning_rate": 0.0005480000000000001, + "loss": 0.3809, + "step": 275 + }, + { + "epoch": 0.002395812536349511, + "grad_norm": 0.1904296875, + "learning_rate": 0.00055, + "loss": 0.3242, + "step": 276 + }, + { + "epoch": 0.0024044930165536757, + "grad_norm": 0.291015625, + "learning_rate": 0.0005520000000000001, + "loss": 0.3281, + "step": 277 + }, + { + "epoch": 0.002413173496757841, + "grad_norm": 0.224609375, + "learning_rate": 0.000554, + "loss": 0.2695, + "step": 278 + }, + { + "epoch": 0.0024218539769620055, + "grad_norm": 0.2451171875, + "learning_rate": 0.0005560000000000001, + "loss": 0.3359, + "step": 279 + }, + { + "epoch": 0.0024305344571661706, + "grad_norm": 0.21484375, + "learning_rate": 0.000558, + "loss": 0.2969, + "step": 280 + }, + { + "epoch": 0.0024392149373703353, + "grad_norm": 0.1767578125, + "learning_rate": 0.0005600000000000001, + "loss": 0.2363, + "step": 281 + }, + { + "epoch": 0.0024478954175745004, + "grad_norm": 0.166015625, + "learning_rate": 0.0005620000000000001, + "loss": 0.3203, + "step": 282 + }, + { + "epoch": 0.002456575897778665, + "grad_norm": 0.1376953125, + "learning_rate": 0.0005639999999999999, + "loss": 0.2402, + "step": 283 + }, + { + "epoch": 0.00246525637798283, + "grad_norm": 0.173828125, + "learning_rate": 0.000566, + "loss": 0.2949, + "step": 284 + }, + { + "epoch": 0.002473936858186995, + "grad_norm": 0.171875, + "learning_rate": 0.0005679999999999999, + "loss": 0.3457, + "step": 285 + }, + { + "epoch": 0.0024826173383911596, + "grad_norm": 0.2021484375, + "learning_rate": 0.00057, + "loss": 0.3047, + "step": 286 + }, + { + "epoch": 0.0024912978185953247, + "grad_norm": 0.181640625, + "learning_rate": 0.0005719999999999999, + "loss": 0.3555, + "step": 287 + }, + { + "epoch": 0.0024999782987994894, + "grad_norm": 0.1640625, + "learning_rate": 0.000574, + "loss": 0.2539, + "step": 288 + }, + { + "epoch": 0.0025086587790036545, + "grad_norm": 0.1728515625, + "learning_rate": 0.000576, + "loss": 0.3066, + "step": 289 + }, + { + "epoch": 0.0025173392592078192, + "grad_norm": 0.2138671875, + "learning_rate": 0.000578, + "loss": 0.3379, + "step": 290 + }, + { + "epoch": 0.0025260197394119843, + "grad_norm": 0.2275390625, + "learning_rate": 0.00058, + "loss": 0.2832, + "step": 291 + }, + { + "epoch": 0.002534700219616149, + "grad_norm": 0.259765625, + "learning_rate": 0.0005819999999999999, + "loss": 0.3867, + "step": 292 + }, + { + "epoch": 0.002543380699820314, + "grad_norm": 0.283203125, + "learning_rate": 0.000584, + "loss": 0.3398, + "step": 293 + }, + { + "epoch": 0.002552061180024479, + "grad_norm": 0.201171875, + "learning_rate": 0.0005859999999999999, + "loss": 0.3203, + "step": 294 + }, + { + "epoch": 0.002560741660228644, + "grad_norm": 0.134765625, + "learning_rate": 0.000588, + "loss": 0.25, + "step": 295 + }, + { + "epoch": 0.0025694221404328086, + "grad_norm": 0.201171875, + "learning_rate": 0.00059, + "loss": 0.3105, + "step": 296 + }, + { + "epoch": 0.0025781026206369738, + "grad_norm": 0.1708984375, + "learning_rate": 0.000592, + "loss": 0.3027, + "step": 297 + }, + { + "epoch": 0.0025867831008411384, + "grad_norm": 0.16796875, + "learning_rate": 0.000594, + "loss": 0.3047, + "step": 298 + }, + { + "epoch": 0.0025954635810453036, + "grad_norm": 0.203125, + "learning_rate": 0.000596, + "loss": 0.457, + "step": 299 + }, + { + "epoch": 0.0026041440612494682, + "grad_norm": 0.1513671875, + "learning_rate": 0.000598, + "loss": 0.2793, + "step": 300 + }, + { + "epoch": 0.0026128245414536334, + "grad_norm": 0.177734375, + "learning_rate": 0.0006, + "loss": 0.3379, + "step": 301 + }, + { + "epoch": 0.002621505021657798, + "grad_norm": 0.1591796875, + "learning_rate": 0.000602, + "loss": 0.2676, + "step": 302 + }, + { + "epoch": 0.002630185501861963, + "grad_norm": 0.232421875, + "learning_rate": 0.000604, + "loss": 0.3223, + "step": 303 + }, + { + "epoch": 0.002638865982066128, + "grad_norm": 0.1630859375, + "learning_rate": 0.000606, + "loss": 0.2617, + "step": 304 + }, + { + "epoch": 0.002647546462270293, + "grad_norm": 0.189453125, + "learning_rate": 0.000608, + "loss": 0.3066, + "step": 305 + }, + { + "epoch": 0.0026562269424744577, + "grad_norm": 0.1767578125, + "learning_rate": 0.00061, + "loss": 0.2637, + "step": 306 + }, + { + "epoch": 0.002664907422678623, + "grad_norm": 0.15625, + "learning_rate": 0.000612, + "loss": 0.3105, + "step": 307 + }, + { + "epoch": 0.0026735879028827875, + "grad_norm": 0.1962890625, + "learning_rate": 0.000614, + "loss": 0.3164, + "step": 308 + }, + { + "epoch": 0.002682268383086952, + "grad_norm": 0.1953125, + "learning_rate": 0.000616, + "loss": 0.2539, + "step": 309 + }, + { + "epoch": 0.0026909488632911173, + "grad_norm": 0.19140625, + "learning_rate": 0.0006180000000000001, + "loss": 0.2969, + "step": 310 + }, + { + "epoch": 0.002699629343495282, + "grad_norm": 0.1845703125, + "learning_rate": 0.00062, + "loss": 0.2598, + "step": 311 + }, + { + "epoch": 0.002708309823699447, + "grad_norm": 0.21484375, + "learning_rate": 0.000622, + "loss": 0.2734, + "step": 312 + }, + { + "epoch": 0.0027169903039036118, + "grad_norm": 0.1435546875, + "learning_rate": 0.000624, + "loss": 0.3359, + "step": 313 + }, + { + "epoch": 0.002725670784107777, + "grad_norm": 0.2333984375, + "learning_rate": 0.000626, + "loss": 0.3359, + "step": 314 + }, + { + "epoch": 0.0027343512643119416, + "grad_norm": 0.2109375, + "learning_rate": 0.000628, + "loss": 0.3008, + "step": 315 + }, + { + "epoch": 0.0027430317445161067, + "grad_norm": 0.1865234375, + "learning_rate": 0.00063, + "loss": 0.3086, + "step": 316 + }, + { + "epoch": 0.0027517122247202714, + "grad_norm": 0.27734375, + "learning_rate": 0.000632, + "loss": 0.3105, + "step": 317 + }, + { + "epoch": 0.0027603927049244365, + "grad_norm": 0.1884765625, + "learning_rate": 0.000634, + "loss": 0.248, + "step": 318 + }, + { + "epoch": 0.002769073185128601, + "grad_norm": 0.2294921875, + "learning_rate": 0.0006360000000000001, + "loss": 0.3008, + "step": 319 + }, + { + "epoch": 0.0027777536653327663, + "grad_norm": 0.1806640625, + "learning_rate": 0.000638, + "loss": 0.2617, + "step": 320 + }, + { + "epoch": 0.002786434145536931, + "grad_norm": 0.16796875, + "learning_rate": 0.00064, + "loss": 0.248, + "step": 321 + }, + { + "epoch": 0.002795114625741096, + "grad_norm": 0.32421875, + "learning_rate": 0.000642, + "loss": 0.3457, + "step": 322 + }, + { + "epoch": 0.002803795105945261, + "grad_norm": 0.2001953125, + "learning_rate": 0.000644, + "loss": 0.3047, + "step": 323 + }, + { + "epoch": 0.002812475586149426, + "grad_norm": 0.1845703125, + "learning_rate": 0.000646, + "loss": 0.2969, + "step": 324 + }, + { + "epoch": 0.0028211560663535906, + "grad_norm": 0.1787109375, + "learning_rate": 0.000648, + "loss": 0.248, + "step": 325 + }, + { + "epoch": 0.0028298365465577557, + "grad_norm": 0.220703125, + "learning_rate": 0.0006500000000000001, + "loss": 0.3164, + "step": 326 + }, + { + "epoch": 0.0028385170267619204, + "grad_norm": 0.17578125, + "learning_rate": 0.000652, + "loss": 0.25, + "step": 327 + }, + { + "epoch": 0.0028471975069660855, + "grad_norm": 0.357421875, + "learning_rate": 0.0006540000000000001, + "loss": 0.3594, + "step": 328 + }, + { + "epoch": 0.00285587798717025, + "grad_norm": 0.130859375, + "learning_rate": 0.000656, + "loss": 0.2471, + "step": 329 + }, + { + "epoch": 0.0028645584673744153, + "grad_norm": 0.2470703125, + "learning_rate": 0.0006580000000000001, + "loss": 0.3359, + "step": 330 + }, + { + "epoch": 0.00287323894757858, + "grad_norm": 0.26171875, + "learning_rate": 0.00066, + "loss": 0.3184, + "step": 331 + }, + { + "epoch": 0.002881919427782745, + "grad_norm": 0.2734375, + "learning_rate": 0.000662, + "loss": 0.3438, + "step": 332 + }, + { + "epoch": 0.00289059990798691, + "grad_norm": 0.166015625, + "learning_rate": 0.0006640000000000001, + "loss": 0.2617, + "step": 333 + }, + { + "epoch": 0.002899280388191075, + "grad_norm": 0.173828125, + "learning_rate": 0.000666, + "loss": 0.3105, + "step": 334 + }, + { + "epoch": 0.0029079608683952396, + "grad_norm": 0.2099609375, + "learning_rate": 0.0006680000000000001, + "loss": 0.3359, + "step": 335 + }, + { + "epoch": 0.0029166413485994043, + "grad_norm": 0.2099609375, + "learning_rate": 0.00067, + "loss": 0.3301, + "step": 336 + }, + { + "epoch": 0.0029253218288035694, + "grad_norm": 0.158203125, + "learning_rate": 0.0006720000000000001, + "loss": 0.3008, + "step": 337 + }, + { + "epoch": 0.002934002309007734, + "grad_norm": 0.6015625, + "learning_rate": 0.000674, + "loss": 0.4102, + "step": 338 + }, + { + "epoch": 0.0029426827892118992, + "grad_norm": 0.2255859375, + "learning_rate": 0.0006760000000000001, + "loss": 0.2773, + "step": 339 + }, + { + "epoch": 0.002951363269416064, + "grad_norm": 0.2060546875, + "learning_rate": 0.0006780000000000001, + "loss": 0.2969, + "step": 340 + }, + { + "epoch": 0.002960043749620229, + "grad_norm": 0.19921875, + "learning_rate": 0.00068, + "loss": 0.3203, + "step": 341 + }, + { + "epoch": 0.0029687242298243937, + "grad_norm": 0.166015625, + "learning_rate": 0.0006820000000000001, + "loss": 0.3066, + "step": 342 + }, + { + "epoch": 0.002977404710028559, + "grad_norm": 0.1796875, + "learning_rate": 0.000684, + "loss": 0.3789, + "step": 343 + }, + { + "epoch": 0.0029860851902327235, + "grad_norm": 0.1455078125, + "learning_rate": 0.0006860000000000001, + "loss": 0.3242, + "step": 344 + }, + { + "epoch": 0.0029947656704368887, + "grad_norm": 0.1826171875, + "learning_rate": 0.0006879999999999999, + "loss": 0.2988, + "step": 345 + }, + { + "epoch": 0.0030034461506410533, + "grad_norm": 0.1552734375, + "learning_rate": 0.00069, + "loss": 0.2695, + "step": 346 + }, + { + "epoch": 0.0030121266308452185, + "grad_norm": 0.25, + "learning_rate": 0.000692, + "loss": 0.2871, + "step": 347 + }, + { + "epoch": 0.003020807111049383, + "grad_norm": 0.13671875, + "learning_rate": 0.000694, + "loss": 0.2812, + "step": 348 + }, + { + "epoch": 0.0030294875912535483, + "grad_norm": 0.357421875, + "learning_rate": 0.000696, + "loss": 0.4082, + "step": 349 + }, + { + "epoch": 0.003038168071457713, + "grad_norm": 0.1650390625, + "learning_rate": 0.0006979999999999999, + "loss": 0.2793, + "step": 350 + }, + { + "epoch": 0.003046848551661878, + "grad_norm": 0.19921875, + "learning_rate": 0.0007, + "loss": 0.2891, + "step": 351 + }, + { + "epoch": 0.0030555290318660428, + "grad_norm": 0.181640625, + "learning_rate": 0.0007019999999999999, + "loss": 0.3535, + "step": 352 + }, + { + "epoch": 0.003064209512070208, + "grad_norm": 0.1689453125, + "learning_rate": 0.000704, + "loss": 0.2656, + "step": 353 + }, + { + "epoch": 0.0030728899922743726, + "grad_norm": 0.27734375, + "learning_rate": 0.0007059999999999999, + "loss": 0.2383, + "step": 354 + }, + { + "epoch": 0.0030815704724785377, + "grad_norm": 0.2265625, + "learning_rate": 0.000708, + "loss": 0.3086, + "step": 355 + }, + { + "epoch": 0.0030902509526827024, + "grad_norm": 0.1904296875, + "learning_rate": 0.00071, + "loss": 0.3086, + "step": 356 + }, + { + "epoch": 0.0030989314328868675, + "grad_norm": 0.1611328125, + "learning_rate": 0.000712, + "loss": 0.291, + "step": 357 + }, + { + "epoch": 0.003107611913091032, + "grad_norm": 0.16796875, + "learning_rate": 0.000714, + "loss": 0.291, + "step": 358 + }, + { + "epoch": 0.0031162923932951973, + "grad_norm": 0.1875, + "learning_rate": 0.000716, + "loss": 0.2988, + "step": 359 + }, + { + "epoch": 0.003124972873499362, + "grad_norm": 0.16015625, + "learning_rate": 0.000718, + "loss": 0.3398, + "step": 360 + }, + { + "epoch": 0.0031336533537035267, + "grad_norm": 0.1611328125, + "learning_rate": 0.0007199999999999999, + "loss": 0.2793, + "step": 361 + }, + { + "epoch": 0.003142333833907692, + "grad_norm": 0.1396484375, + "learning_rate": 0.000722, + "loss": 0.3047, + "step": 362 + }, + { + "epoch": 0.0031510143141118565, + "grad_norm": 0.185546875, + "learning_rate": 0.000724, + "loss": 0.3281, + "step": 363 + }, + { + "epoch": 0.0031596947943160216, + "grad_norm": 0.154296875, + "learning_rate": 0.000726, + "loss": 0.2969, + "step": 364 + }, + { + "epoch": 0.0031683752745201863, + "grad_norm": 0.181640625, + "learning_rate": 0.000728, + "loss": 0.3652, + "step": 365 + }, + { + "epoch": 0.0031770557547243514, + "grad_norm": 0.1279296875, + "learning_rate": 0.00073, + "loss": 0.2344, + "step": 366 + }, + { + "epoch": 0.003185736234928516, + "grad_norm": 0.1455078125, + "learning_rate": 0.000732, + "loss": 0.2891, + "step": 367 + }, + { + "epoch": 0.003194416715132681, + "grad_norm": 0.1884765625, + "learning_rate": 0.000734, + "loss": 0.3438, + "step": 368 + }, + { + "epoch": 0.003203097195336846, + "grad_norm": 0.181640625, + "learning_rate": 0.000736, + "loss": 0.3164, + "step": 369 + }, + { + "epoch": 0.003211777675541011, + "grad_norm": 0.1669921875, + "learning_rate": 0.000738, + "loss": 0.2832, + "step": 370 + }, + { + "epoch": 0.0032204581557451757, + "grad_norm": 0.18359375, + "learning_rate": 0.00074, + "loss": 0.2695, + "step": 371 + }, + { + "epoch": 0.003229138635949341, + "grad_norm": 0.22265625, + "learning_rate": 0.000742, + "loss": 0.2559, + "step": 372 + }, + { + "epoch": 0.0032378191161535055, + "grad_norm": 0.2060546875, + "learning_rate": 0.000744, + "loss": 0.2949, + "step": 373 + }, + { + "epoch": 0.0032464995963576706, + "grad_norm": 0.1494140625, + "learning_rate": 0.000746, + "loss": 0.2832, + "step": 374 + }, + { + "epoch": 0.0032551800765618353, + "grad_norm": 0.1669921875, + "learning_rate": 0.000748, + "loss": 0.2832, + "step": 375 + }, + { + "epoch": 0.0032638605567660004, + "grad_norm": 0.279296875, + "learning_rate": 0.00075, + "loss": 0.2812, + "step": 376 + }, + { + "epoch": 0.003272541036970165, + "grad_norm": 0.48828125, + "learning_rate": 0.0007520000000000001, + "loss": 0.4453, + "step": 377 + }, + { + "epoch": 0.0032812215171743302, + "grad_norm": 0.1630859375, + "learning_rate": 0.000754, + "loss": 0.3223, + "step": 378 + }, + { + "epoch": 0.003289901997378495, + "grad_norm": 0.150390625, + "learning_rate": 0.000756, + "loss": 0.2695, + "step": 379 + }, + { + "epoch": 0.00329858247758266, + "grad_norm": 0.236328125, + "learning_rate": 0.000758, + "loss": 0.3164, + "step": 380 + }, + { + "epoch": 0.0033072629577868247, + "grad_norm": 0.25390625, + "learning_rate": 0.00076, + "loss": 0.291, + "step": 381 + }, + { + "epoch": 0.00331594343799099, + "grad_norm": 0.220703125, + "learning_rate": 0.000762, + "loss": 0.3047, + "step": 382 + }, + { + "epoch": 0.0033246239181951545, + "grad_norm": 0.1923828125, + "learning_rate": 0.000764, + "loss": 0.2832, + "step": 383 + }, + { + "epoch": 0.0033333043983993197, + "grad_norm": 0.2021484375, + "learning_rate": 0.0007660000000000001, + "loss": 0.291, + "step": 384 + }, + { + "epoch": 0.0033419848786034843, + "grad_norm": 0.173828125, + "learning_rate": 0.000768, + "loss": 0.2305, + "step": 385 + }, + { + "epoch": 0.003350665358807649, + "grad_norm": 0.25390625, + "learning_rate": 0.0007700000000000001, + "loss": 0.3105, + "step": 386 + }, + { + "epoch": 0.003359345839011814, + "grad_norm": 0.1611328125, + "learning_rate": 0.000772, + "loss": 0.3145, + "step": 387 + }, + { + "epoch": 0.003368026319215979, + "grad_norm": 0.171875, + "learning_rate": 0.0007740000000000001, + "loss": 0.373, + "step": 388 + }, + { + "epoch": 0.003376706799420144, + "grad_norm": 0.1533203125, + "learning_rate": 0.000776, + "loss": 0.2158, + "step": 389 + }, + { + "epoch": 0.0033853872796243086, + "grad_norm": 0.2060546875, + "learning_rate": 0.000778, + "loss": 0.2676, + "step": 390 + }, + { + "epoch": 0.0033940677598284738, + "grad_norm": 0.21484375, + "learning_rate": 0.0007800000000000001, + "loss": 0.3438, + "step": 391 + }, + { + "epoch": 0.0034027482400326384, + "grad_norm": 0.1962890625, + "learning_rate": 0.000782, + "loss": 0.25, + "step": 392 + }, + { + "epoch": 0.0034114287202368036, + "grad_norm": 0.1865234375, + "learning_rate": 0.0007840000000000001, + "loss": 0.2832, + "step": 393 + }, + { + "epoch": 0.0034201092004409682, + "grad_norm": 0.1484375, + "learning_rate": 0.000786, + "loss": 0.293, + "step": 394 + }, + { + "epoch": 0.0034287896806451334, + "grad_norm": 0.1943359375, + "learning_rate": 0.0007880000000000001, + "loss": 0.3242, + "step": 395 + }, + { + "epoch": 0.003437470160849298, + "grad_norm": 0.1376953125, + "learning_rate": 0.00079, + "loss": 0.293, + "step": 396 + }, + { + "epoch": 0.003446150641053463, + "grad_norm": 0.146484375, + "learning_rate": 0.0007920000000000001, + "loss": 0.2266, + "step": 397 + }, + { + "epoch": 0.003454831121257628, + "grad_norm": 0.1640625, + "learning_rate": 0.0007940000000000001, + "loss": 0.3711, + "step": 398 + }, + { + "epoch": 0.003463511601461793, + "grad_norm": 0.21484375, + "learning_rate": 0.000796, + "loss": 0.3184, + "step": 399 + }, + { + "epoch": 0.0034721920816659577, + "grad_norm": 0.2021484375, + "learning_rate": 0.0007980000000000001, + "loss": 0.334, + "step": 400 + }, + { + "epoch": 0.003480872561870123, + "grad_norm": 0.1953125, + "learning_rate": 0.0008, + "loss": 0.3398, + "step": 401 + }, + { + "epoch": 0.0034895530420742875, + "grad_norm": 0.1689453125, + "learning_rate": 0.0008020000000000001, + "loss": 0.2578, + "step": 402 + }, + { + "epoch": 0.0034982335222784526, + "grad_norm": 0.12890625, + "learning_rate": 0.000804, + "loss": 0.2734, + "step": 403 + }, + { + "epoch": 0.0035069140024826173, + "grad_norm": 0.2197265625, + "learning_rate": 0.0008060000000000001, + "loss": 0.332, + "step": 404 + }, + { + "epoch": 0.0035155944826867824, + "grad_norm": 0.1806640625, + "learning_rate": 0.000808, + "loss": 0.2969, + "step": 405 + }, + { + "epoch": 0.003524274962890947, + "grad_norm": 0.205078125, + "learning_rate": 0.0008100000000000001, + "loss": 0.249, + "step": 406 + }, + { + "epoch": 0.003532955443095112, + "grad_norm": 0.244140625, + "learning_rate": 0.0008120000000000001, + "loss": 0.3672, + "step": 407 + }, + { + "epoch": 0.003541635923299277, + "grad_norm": 0.1484375, + "learning_rate": 0.0008139999999999999, + "loss": 0.2617, + "step": 408 + }, + { + "epoch": 0.003550316403503442, + "grad_norm": 0.1806640625, + "learning_rate": 0.000816, + "loss": 0.3086, + "step": 409 + }, + { + "epoch": 0.0035589968837076067, + "grad_norm": 0.13671875, + "learning_rate": 0.0008179999999999999, + "loss": 0.2363, + "step": 410 + }, + { + "epoch": 0.003567677363911772, + "grad_norm": 0.12158203125, + "learning_rate": 0.00082, + "loss": 0.3008, + "step": 411 + }, + { + "epoch": 0.0035763578441159365, + "grad_norm": 0.138671875, + "learning_rate": 0.0008219999999999999, + "loss": 0.2852, + "step": 412 + }, + { + "epoch": 0.003585038324320101, + "grad_norm": 0.1708984375, + "learning_rate": 0.000824, + "loss": 0.3027, + "step": 413 + }, + { + "epoch": 0.0035937188045242663, + "grad_norm": 0.1455078125, + "learning_rate": 0.000826, + "loss": 0.2451, + "step": 414 + }, + { + "epoch": 0.003602399284728431, + "grad_norm": 0.1416015625, + "learning_rate": 0.000828, + "loss": 0.3164, + "step": 415 + }, + { + "epoch": 0.003611079764932596, + "grad_norm": 0.1201171875, + "learning_rate": 0.00083, + "loss": 0.2734, + "step": 416 + }, + { + "epoch": 0.003619760245136761, + "grad_norm": 0.12060546875, + "learning_rate": 0.000832, + "loss": 0.2734, + "step": 417 + }, + { + "epoch": 0.003628440725340926, + "grad_norm": 0.1904296875, + "learning_rate": 0.000834, + "loss": 0.3262, + "step": 418 + }, + { + "epoch": 0.0036371212055450906, + "grad_norm": 0.11669921875, + "learning_rate": 0.0008359999999999999, + "loss": 0.2969, + "step": 419 + }, + { + "epoch": 0.0036458016857492557, + "grad_norm": 0.1591796875, + "learning_rate": 0.000838, + "loss": 0.2539, + "step": 420 + }, + { + "epoch": 0.0036544821659534204, + "grad_norm": 0.1083984375, + "learning_rate": 0.00084, + "loss": 0.2363, + "step": 421 + }, + { + "epoch": 0.0036631626461575855, + "grad_norm": 0.142578125, + "learning_rate": 0.000842, + "loss": 0.3066, + "step": 422 + }, + { + "epoch": 0.00367184312636175, + "grad_norm": 0.1328125, + "learning_rate": 0.000844, + "loss": 0.2969, + "step": 423 + }, + { + "epoch": 0.0036805236065659153, + "grad_norm": 0.1923828125, + "learning_rate": 0.000846, + "loss": 0.2988, + "step": 424 + }, + { + "epoch": 0.00368920408677008, + "grad_norm": 0.12890625, + "learning_rate": 0.000848, + "loss": 0.2393, + "step": 425 + }, + { + "epoch": 0.003697884566974245, + "grad_norm": 0.134765625, + "learning_rate": 0.00085, + "loss": 0.2812, + "step": 426 + }, + { + "epoch": 0.00370656504717841, + "grad_norm": 0.1474609375, + "learning_rate": 0.000852, + "loss": 0.2422, + "step": 427 + }, + { + "epoch": 0.003715245527382575, + "grad_norm": 0.26953125, + "learning_rate": 0.000854, + "loss": 0.334, + "step": 428 + }, + { + "epoch": 0.0037239260075867396, + "grad_norm": 0.1328125, + "learning_rate": 0.000856, + "loss": 0.3047, + "step": 429 + }, + { + "epoch": 0.0037326064877909047, + "grad_norm": 0.228515625, + "learning_rate": 0.000858, + "loss": 0.3066, + "step": 430 + }, + { + "epoch": 0.0037412869679950694, + "grad_norm": 0.19140625, + "learning_rate": 0.00086, + "loss": 0.2852, + "step": 431 + }, + { + "epoch": 0.0037499674481992346, + "grad_norm": 0.17578125, + "learning_rate": 0.000862, + "loss": 0.2754, + "step": 432 + }, + { + "epoch": 0.0037586479284033992, + "grad_norm": 0.1796875, + "learning_rate": 0.000864, + "loss": 0.2402, + "step": 433 + }, + { + "epoch": 0.0037673284086075644, + "grad_norm": 0.12109375, + "learning_rate": 0.000866, + "loss": 0.2734, + "step": 434 + }, + { + "epoch": 0.003776008888811729, + "grad_norm": 0.193359375, + "learning_rate": 0.0008680000000000001, + "loss": 0.2871, + "step": 435 + }, + { + "epoch": 0.003784689369015894, + "grad_norm": 0.142578125, + "learning_rate": 0.00087, + "loss": 0.3008, + "step": 436 + }, + { + "epoch": 0.003793369849220059, + "grad_norm": 0.1669921875, + "learning_rate": 0.000872, + "loss": 0.2715, + "step": 437 + }, + { + "epoch": 0.0038020503294242235, + "grad_norm": 0.16015625, + "learning_rate": 0.000874, + "loss": 0.3086, + "step": 438 + }, + { + "epoch": 0.0038107308096283887, + "grad_norm": 0.11767578125, + "learning_rate": 0.000876, + "loss": 0.2275, + "step": 439 + }, + { + "epoch": 0.0038194112898325533, + "grad_norm": 0.189453125, + "learning_rate": 0.000878, + "loss": 0.375, + "step": 440 + }, + { + "epoch": 0.0038280917700367185, + "grad_norm": 0.1572265625, + "learning_rate": 0.00088, + "loss": 0.3242, + "step": 441 + }, + { + "epoch": 0.003836772250240883, + "grad_norm": 0.1123046875, + "learning_rate": 0.000882, + "loss": 0.2852, + "step": 442 + }, + { + "epoch": 0.0038454527304450483, + "grad_norm": 0.11962890625, + "learning_rate": 0.000884, + "loss": 0.3105, + "step": 443 + }, + { + "epoch": 0.003854133210649213, + "grad_norm": 0.1376953125, + "learning_rate": 0.0008860000000000001, + "loss": 0.2891, + "step": 444 + }, + { + "epoch": 0.003862813690853378, + "grad_norm": 0.1669921875, + "learning_rate": 0.000888, + "loss": 0.3516, + "step": 445 + }, + { + "epoch": 0.0038714941710575428, + "grad_norm": 0.1689453125, + "learning_rate": 0.0008900000000000001, + "loss": 0.3945, + "step": 446 + }, + { + "epoch": 0.003880174651261708, + "grad_norm": 0.181640625, + "learning_rate": 0.000892, + "loss": 0.209, + "step": 447 + }, + { + "epoch": 0.0038888551314658726, + "grad_norm": 0.1376953125, + "learning_rate": 0.000894, + "loss": 0.2715, + "step": 448 + }, + { + "epoch": 0.0038975356116700377, + "grad_norm": 0.1396484375, + "learning_rate": 0.000896, + "loss": 0.3066, + "step": 449 + }, + { + "epoch": 0.0039062160918742024, + "grad_norm": 0.12060546875, + "learning_rate": 0.000898, + "loss": 0.2559, + "step": 450 + }, + { + "epoch": 0.0039148965720783675, + "grad_norm": 0.1416015625, + "learning_rate": 0.0009000000000000001, + "loss": 0.2617, + "step": 451 + }, + { + "epoch": 0.003923577052282533, + "grad_norm": 0.1640625, + "learning_rate": 0.000902, + "loss": 0.2988, + "step": 452 + }, + { + "epoch": 0.003932257532486697, + "grad_norm": 0.1494140625, + "learning_rate": 0.0009040000000000001, + "loss": 0.3164, + "step": 453 + }, + { + "epoch": 0.003940938012690862, + "grad_norm": 0.267578125, + "learning_rate": 0.000906, + "loss": 0.3789, + "step": 454 + }, + { + "epoch": 0.003949618492895027, + "grad_norm": 0.15234375, + "learning_rate": 0.0009080000000000001, + "loss": 0.2773, + "step": 455 + }, + { + "epoch": 0.003958298973099192, + "grad_norm": 0.1123046875, + "learning_rate": 0.00091, + "loss": 0.2051, + "step": 456 + }, + { + "epoch": 0.0039669794533033565, + "grad_norm": 0.142578125, + "learning_rate": 0.000912, + "loss": 0.2422, + "step": 457 + }, + { + "epoch": 0.003975659933507522, + "grad_norm": 0.1513671875, + "learning_rate": 0.0009140000000000001, + "loss": 0.2441, + "step": 458 + }, + { + "epoch": 0.003984340413711687, + "grad_norm": 0.1162109375, + "learning_rate": 0.000916, + "loss": 0.2676, + "step": 459 + }, + { + "epoch": 0.003993020893915852, + "grad_norm": 0.15625, + "learning_rate": 0.0009180000000000001, + "loss": 0.3477, + "step": 460 + }, + { + "epoch": 0.004001701374120016, + "grad_norm": 0.173828125, + "learning_rate": 0.00092, + "loss": 0.2578, + "step": 461 + }, + { + "epoch": 0.004010381854324181, + "grad_norm": 0.201171875, + "learning_rate": 0.0009220000000000001, + "loss": 0.3086, + "step": 462 + }, + { + "epoch": 0.004019062334528346, + "grad_norm": 0.1611328125, + "learning_rate": 0.000924, + "loss": 0.2227, + "step": 463 + }, + { + "epoch": 0.004027742814732511, + "grad_norm": 0.2138671875, + "learning_rate": 0.0009260000000000001, + "loss": 0.4023, + "step": 464 + }, + { + "epoch": 0.004036423294936676, + "grad_norm": 0.1416015625, + "learning_rate": 0.0009280000000000001, + "loss": 0.2969, + "step": 465 + }, + { + "epoch": 0.004045103775140841, + "grad_norm": 0.177734375, + "learning_rate": 0.00093, + "loss": 0.2773, + "step": 466 + }, + { + "epoch": 0.004053784255345006, + "grad_norm": 0.2138671875, + "learning_rate": 0.0009320000000000001, + "loss": 0.3027, + "step": 467 + }, + { + "epoch": 0.00406246473554917, + "grad_norm": 0.1328125, + "learning_rate": 0.000934, + "loss": 0.2617, + "step": 468 + }, + { + "epoch": 0.004071145215753335, + "grad_norm": 0.173828125, + "learning_rate": 0.0009360000000000001, + "loss": 0.2539, + "step": 469 + }, + { + "epoch": 0.0040798256959575, + "grad_norm": 0.306640625, + "learning_rate": 0.0009379999999999999, + "loss": 0.3242, + "step": 470 + }, + { + "epoch": 0.0040885061761616655, + "grad_norm": 0.12158203125, + "learning_rate": 0.00094, + "loss": 0.2334, + "step": 471 + }, + { + "epoch": 0.00409718665636583, + "grad_norm": 0.1298828125, + "learning_rate": 0.000942, + "loss": 0.2891, + "step": 472 + }, + { + "epoch": 0.004105867136569995, + "grad_norm": 0.19921875, + "learning_rate": 0.000944, + "loss": 0.4492, + "step": 473 + }, + { + "epoch": 0.00411454761677416, + "grad_norm": 0.1474609375, + "learning_rate": 0.000946, + "loss": 0.3164, + "step": 474 + }, + { + "epoch": 0.004123228096978325, + "grad_norm": 0.173828125, + "learning_rate": 0.000948, + "loss": 0.3027, + "step": 475 + }, + { + "epoch": 0.004131908577182489, + "grad_norm": 0.1728515625, + "learning_rate": 0.00095, + "loss": 0.2754, + "step": 476 + }, + { + "epoch": 0.0041405890573866545, + "grad_norm": 0.1376953125, + "learning_rate": 0.0009519999999999999, + "loss": 0.2275, + "step": 477 + }, + { + "epoch": 0.00414926953759082, + "grad_norm": 0.2373046875, + "learning_rate": 0.000954, + "loss": 0.2891, + "step": 478 + }, + { + "epoch": 0.004157950017794985, + "grad_norm": 0.1943359375, + "learning_rate": 0.0009559999999999999, + "loss": 0.332, + "step": 479 + }, + { + "epoch": 0.004166630497999149, + "grad_norm": 0.16796875, + "learning_rate": 0.000958, + "loss": 0.3027, + "step": 480 + }, + { + "epoch": 0.004175310978203314, + "grad_norm": 0.1318359375, + "learning_rate": 0.00096, + "loss": 0.2295, + "step": 481 + }, + { + "epoch": 0.004183991458407479, + "grad_norm": 0.1181640625, + "learning_rate": 0.000962, + "loss": 0.2285, + "step": 482 + }, + { + "epoch": 0.004192671938611644, + "grad_norm": 0.1494140625, + "learning_rate": 0.000964, + "loss": 0.2969, + "step": 483 + }, + { + "epoch": 0.004201352418815809, + "grad_norm": 0.2099609375, + "learning_rate": 0.000966, + "loss": 0.3027, + "step": 484 + }, + { + "epoch": 0.004210032899019974, + "grad_norm": 0.1640625, + "learning_rate": 0.000968, + "loss": 0.3477, + "step": 485 + }, + { + "epoch": 0.004218713379224139, + "grad_norm": 0.17578125, + "learning_rate": 0.0009699999999999999, + "loss": 0.2617, + "step": 486 + }, + { + "epoch": 0.004227393859428304, + "grad_norm": 0.208984375, + "learning_rate": 0.000972, + "loss": 0.4824, + "step": 487 + }, + { + "epoch": 0.004236074339632468, + "grad_norm": 0.1806640625, + "learning_rate": 0.000974, + "loss": 0.293, + "step": 488 + }, + { + "epoch": 0.004244754819836633, + "grad_norm": 0.326171875, + "learning_rate": 0.000976, + "loss": 0.4102, + "step": 489 + }, + { + "epoch": 0.0042534353000407985, + "grad_norm": 0.138671875, + "learning_rate": 0.000978, + "loss": 0.2637, + "step": 490 + }, + { + "epoch": 0.004262115780244963, + "grad_norm": 0.11279296875, + "learning_rate": 0.00098, + "loss": 0.2578, + "step": 491 + }, + { + "epoch": 0.004270796260449128, + "grad_norm": 0.11962890625, + "learning_rate": 0.000982, + "loss": 0.1992, + "step": 492 + }, + { + "epoch": 0.004279476740653293, + "grad_norm": 0.1533203125, + "learning_rate": 0.000984, + "loss": 0.252, + "step": 493 + }, + { + "epoch": 0.004288157220857458, + "grad_norm": 0.177734375, + "learning_rate": 0.0009860000000000001, + "loss": 0.2598, + "step": 494 + }, + { + "epoch": 0.004296837701061622, + "grad_norm": 0.142578125, + "learning_rate": 0.000988, + "loss": 0.2422, + "step": 495 + }, + { + "epoch": 0.0043055181812657875, + "grad_norm": 0.1279296875, + "learning_rate": 0.00099, + "loss": 0.2578, + "step": 496 + }, + { + "epoch": 0.004314198661469953, + "grad_norm": 0.1455078125, + "learning_rate": 0.000992, + "loss": 0.2676, + "step": 497 + }, + { + "epoch": 0.004322879141674118, + "grad_norm": 0.169921875, + "learning_rate": 0.000994, + "loss": 0.2852, + "step": 498 + }, + { + "epoch": 0.004331559621878282, + "grad_norm": 0.1357421875, + "learning_rate": 0.000996, + "loss": 0.2754, + "step": 499 + }, + { + "epoch": 0.004340240102082447, + "grad_norm": 0.1962890625, + "learning_rate": 0.000998, + "loss": 0.2754, + "step": 500 + }, + { + "epoch": 0.004348920582286612, + "grad_norm": 0.1669921875, + "learning_rate": 0.001, + "loss": 0.3984, + "step": 501 + }, + { + "epoch": 0.004357601062490777, + "grad_norm": 0.11865234375, + "learning_rate": 0.001002, + "loss": 0.2656, + "step": 502 + }, + { + "epoch": 0.004366281542694942, + "grad_norm": 0.2451171875, + "learning_rate": 0.0010040000000000001, + "loss": 0.3008, + "step": 503 + }, + { + "epoch": 0.004374962022899107, + "grad_norm": 0.1630859375, + "learning_rate": 0.001006, + "loss": 0.2676, + "step": 504 + }, + { + "epoch": 0.004383642503103272, + "grad_norm": 0.123046875, + "learning_rate": 0.001008, + "loss": 0.2461, + "step": 505 + }, + { + "epoch": 0.004392322983307437, + "grad_norm": 0.126953125, + "learning_rate": 0.00101, + "loss": 0.2793, + "step": 506 + }, + { + "epoch": 0.004401003463511601, + "grad_norm": 0.11669921875, + "learning_rate": 0.001012, + "loss": 0.3027, + "step": 507 + }, + { + "epoch": 0.004409683943715766, + "grad_norm": 0.1630859375, + "learning_rate": 0.001014, + "loss": 0.2793, + "step": 508 + }, + { + "epoch": 0.004418364423919931, + "grad_norm": 0.142578125, + "learning_rate": 0.001016, + "loss": 0.2559, + "step": 509 + }, + { + "epoch": 0.0044270449041240965, + "grad_norm": 0.1298828125, + "learning_rate": 0.001018, + "loss": 0.2812, + "step": 510 + }, + { + "epoch": 0.004435725384328261, + "grad_norm": 0.158203125, + "learning_rate": 0.00102, + "loss": 0.2793, + "step": 511 + }, + { + "epoch": 0.004444405864532426, + "grad_norm": 0.142578125, + "learning_rate": 0.0010220000000000001, + "loss": 0.2891, + "step": 512 + }, + { + "epoch": 0.004453086344736591, + "grad_norm": 0.146484375, + "learning_rate": 0.001024, + "loss": 0.334, + "step": 513 + }, + { + "epoch": 0.004461766824940756, + "grad_norm": 0.0986328125, + "learning_rate": 0.001026, + "loss": 0.252, + "step": 514 + }, + { + "epoch": 0.00447044730514492, + "grad_norm": 0.150390625, + "learning_rate": 0.001028, + "loss": 0.2432, + "step": 515 + }, + { + "epoch": 0.0044791277853490855, + "grad_norm": 0.1650390625, + "learning_rate": 0.00103, + "loss": 0.3223, + "step": 516 + }, + { + "epoch": 0.004487808265553251, + "grad_norm": 0.15234375, + "learning_rate": 0.0010320000000000001, + "loss": 0.2148, + "step": 517 + }, + { + "epoch": 0.004496488745757415, + "grad_norm": 0.1357421875, + "learning_rate": 0.001034, + "loss": 0.3164, + "step": 518 + }, + { + "epoch": 0.00450516922596158, + "grad_norm": 0.2734375, + "learning_rate": 0.001036, + "loss": 0.3379, + "step": 519 + }, + { + "epoch": 0.004513849706165745, + "grad_norm": 0.12890625, + "learning_rate": 0.001038, + "loss": 0.2656, + "step": 520 + }, + { + "epoch": 0.00452253018636991, + "grad_norm": 0.142578125, + "learning_rate": 0.0010400000000000001, + "loss": 0.293, + "step": 521 + }, + { + "epoch": 0.0045312106665740745, + "grad_norm": 0.1318359375, + "learning_rate": 0.001042, + "loss": 0.2656, + "step": 522 + }, + { + "epoch": 0.00453989114677824, + "grad_norm": 0.1376953125, + "learning_rate": 0.001044, + "loss": 0.3262, + "step": 523 + }, + { + "epoch": 0.004548571626982405, + "grad_norm": 0.15625, + "learning_rate": 0.001046, + "loss": 0.293, + "step": 524 + }, + { + "epoch": 0.00455725210718657, + "grad_norm": 0.12890625, + "learning_rate": 0.001048, + "loss": 0.2207, + "step": 525 + }, + { + "epoch": 0.004565932587390734, + "grad_norm": 0.1474609375, + "learning_rate": 0.0010500000000000002, + "loss": 0.3086, + "step": 526 + }, + { + "epoch": 0.004574613067594899, + "grad_norm": 0.2236328125, + "learning_rate": 0.001052, + "loss": 0.2676, + "step": 527 + }, + { + "epoch": 0.004583293547799064, + "grad_norm": 0.2236328125, + "learning_rate": 0.001054, + "loss": 0.332, + "step": 528 + }, + { + "epoch": 0.0045919740280032295, + "grad_norm": 0.171875, + "learning_rate": 0.001056, + "loss": 0.2715, + "step": 529 + }, + { + "epoch": 0.004600654508207394, + "grad_norm": 0.1513671875, + "learning_rate": 0.0010580000000000001, + "loss": 0.3008, + "step": 530 + }, + { + "epoch": 0.004609334988411559, + "grad_norm": 0.2080078125, + "learning_rate": 0.0010600000000000002, + "loss": 0.2969, + "step": 531 + }, + { + "epoch": 0.004618015468615724, + "grad_norm": 0.1572265625, + "learning_rate": 0.001062, + "loss": 0.2246, + "step": 532 + }, + { + "epoch": 0.004626695948819889, + "grad_norm": 0.146484375, + "learning_rate": 0.001064, + "loss": 0.249, + "step": 533 + }, + { + "epoch": 0.004635376429024053, + "grad_norm": 0.1552734375, + "learning_rate": 0.001066, + "loss": 0.2832, + "step": 534 + }, + { + "epoch": 0.0046440569092282185, + "grad_norm": 0.17578125, + "learning_rate": 0.0010680000000000002, + "loss": 0.2578, + "step": 535 + }, + { + "epoch": 0.004652737389432384, + "grad_norm": 0.0888671875, + "learning_rate": 0.00107, + "loss": 0.2383, + "step": 536 + }, + { + "epoch": 0.004661417869636549, + "grad_norm": 0.138671875, + "learning_rate": 0.001072, + "loss": 0.2734, + "step": 537 + }, + { + "epoch": 0.004670098349840713, + "grad_norm": 0.1650390625, + "learning_rate": 0.001074, + "loss": 0.291, + "step": 538 + }, + { + "epoch": 0.004678778830044878, + "grad_norm": 0.1591796875, + "learning_rate": 0.0010760000000000001, + "loss": 0.2676, + "step": 539 + }, + { + "epoch": 0.004687459310249043, + "grad_norm": 0.10205078125, + "learning_rate": 0.0010780000000000002, + "loss": 0.2285, + "step": 540 + }, + { + "epoch": 0.0046961397904532074, + "grad_norm": 0.15234375, + "learning_rate": 0.00108, + "loss": 0.2988, + "step": 541 + }, + { + "epoch": 0.0047048202706573726, + "grad_norm": 0.1630859375, + "learning_rate": 0.001082, + "loss": 0.3203, + "step": 542 + }, + { + "epoch": 0.004713500750861538, + "grad_norm": 0.138671875, + "learning_rate": 0.0010840000000000001, + "loss": 0.2793, + "step": 543 + }, + { + "epoch": 0.004722181231065703, + "grad_norm": 0.189453125, + "learning_rate": 0.0010860000000000002, + "loss": 0.293, + "step": 544 + }, + { + "epoch": 0.004730861711269867, + "grad_norm": 0.1875, + "learning_rate": 0.0010880000000000002, + "loss": 0.25, + "step": 545 + }, + { + "epoch": 0.004739542191474032, + "grad_norm": 0.134765625, + "learning_rate": 0.00109, + "loss": 0.1982, + "step": 546 + }, + { + "epoch": 0.004748222671678197, + "grad_norm": 0.1591796875, + "learning_rate": 0.001092, + "loss": 0.2773, + "step": 547 + }, + { + "epoch": 0.004756903151882362, + "grad_norm": 0.1435546875, + "learning_rate": 0.0010940000000000001, + "loss": 0.2891, + "step": 548 + }, + { + "epoch": 0.004765583632086527, + "grad_norm": 0.1484375, + "learning_rate": 0.0010960000000000002, + "loss": 0.2285, + "step": 549 + }, + { + "epoch": 0.004774264112290692, + "grad_norm": 0.1611328125, + "learning_rate": 0.001098, + "loss": 0.2949, + "step": 550 + }, + { + "epoch": 0.004782944592494857, + "grad_norm": 0.220703125, + "learning_rate": 0.0011, + "loss": 0.3047, + "step": 551 + }, + { + "epoch": 0.004791625072699022, + "grad_norm": 0.1103515625, + "learning_rate": 0.0011020000000000001, + "loss": 0.2793, + "step": 552 + }, + { + "epoch": 0.004800305552903186, + "grad_norm": 0.181640625, + "learning_rate": 0.0011040000000000002, + "loss": 0.3809, + "step": 553 + }, + { + "epoch": 0.004808986033107351, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011060000000000002, + "loss": 0.2539, + "step": 554 + }, + { + "epoch": 0.0048176665133115165, + "grad_norm": 0.17578125, + "learning_rate": 0.001108, + "loss": 0.2559, + "step": 555 + }, + { + "epoch": 0.004826346993515682, + "grad_norm": 0.1416015625, + "learning_rate": 0.00111, + "loss": 0.3125, + "step": 556 + }, + { + "epoch": 0.004835027473719846, + "grad_norm": 0.12353515625, + "learning_rate": 0.0011120000000000001, + "loss": 0.2441, + "step": 557 + }, + { + "epoch": 0.004843707953924011, + "grad_norm": 0.130859375, + "learning_rate": 0.0011140000000000002, + "loss": 0.2598, + "step": 558 + }, + { + "epoch": 0.004852388434128176, + "grad_norm": 0.1279296875, + "learning_rate": 0.001116, + "loss": 0.2852, + "step": 559 + }, + { + "epoch": 0.004861068914332341, + "grad_norm": 0.1064453125, + "learning_rate": 0.001118, + "loss": 0.2754, + "step": 560 + }, + { + "epoch": 0.0048697493945365055, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011200000000000001, + "loss": 0.248, + "step": 561 + }, + { + "epoch": 0.004878429874740671, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011220000000000002, + "loss": 0.2832, + "step": 562 + }, + { + "epoch": 0.004887110354944836, + "grad_norm": 0.11376953125, + "learning_rate": 0.0011240000000000002, + "loss": 0.2275, + "step": 563 + }, + { + "epoch": 0.004895790835149001, + "grad_norm": 0.177734375, + "learning_rate": 0.0011259999999999998, + "loss": 0.457, + "step": 564 + }, + { + "epoch": 0.004904471315353165, + "grad_norm": 0.169921875, + "learning_rate": 0.0011279999999999999, + "loss": 0.3164, + "step": 565 + }, + { + "epoch": 0.00491315179555733, + "grad_norm": 0.1552734375, + "learning_rate": 0.00113, + "loss": 0.2754, + "step": 566 + }, + { + "epoch": 0.004921832275761495, + "grad_norm": 0.1591796875, + "learning_rate": 0.001132, + "loss": 0.2559, + "step": 567 + }, + { + "epoch": 0.00493051275596566, + "grad_norm": 0.1591796875, + "learning_rate": 0.001134, + "loss": 0.25, + "step": 568 + }, + { + "epoch": 0.004939193236169825, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011359999999999999, + "loss": 0.2363, + "step": 569 + }, + { + "epoch": 0.00494787371637399, + "grad_norm": 0.435546875, + "learning_rate": 0.001138, + "loss": 0.8203, + "step": 570 + }, + { + "epoch": 0.004956554196578155, + "grad_norm": 0.1455078125, + "learning_rate": 0.00114, + "loss": 0.2598, + "step": 571 + }, + { + "epoch": 0.004965234676782319, + "grad_norm": 0.1455078125, + "learning_rate": 0.001142, + "loss": 0.2969, + "step": 572 + }, + { + "epoch": 0.004973915156986484, + "grad_norm": 0.205078125, + "learning_rate": 0.0011439999999999998, + "loss": 0.293, + "step": 573 + }, + { + "epoch": 0.0049825956371906495, + "grad_norm": 0.1748046875, + "learning_rate": 0.0011459999999999999, + "loss": 0.291, + "step": 574 + }, + { + "epoch": 0.004991276117394815, + "grad_norm": 0.1689453125, + "learning_rate": 0.001148, + "loss": 0.252, + "step": 575 + }, + { + "epoch": 0.004999956597598979, + "grad_norm": 0.126953125, + "learning_rate": 0.00115, + "loss": 0.2773, + "step": 576 + }, + { + "epoch": 0.005008637077803144, + "grad_norm": 0.1611328125, + "learning_rate": 0.001152, + "loss": 0.2578, + "step": 577 + }, + { + "epoch": 0.005017317558007309, + "grad_norm": 0.205078125, + "learning_rate": 0.0011539999999999999, + "loss": 0.25, + "step": 578 + }, + { + "epoch": 0.005025998038211474, + "grad_norm": 0.111328125, + "learning_rate": 0.001156, + "loss": 0.2461, + "step": 579 + }, + { + "epoch": 0.0050346785184156384, + "grad_norm": 0.1162109375, + "learning_rate": 0.001158, + "loss": 0.2168, + "step": 580 + }, + { + "epoch": 0.0050433589986198036, + "grad_norm": 0.08544921875, + "learning_rate": 0.00116, + "loss": 0.2559, + "step": 581 + }, + { + "epoch": 0.005052039478823969, + "grad_norm": 0.412109375, + "learning_rate": 0.0011619999999999998, + "loss": 0.4023, + "step": 582 + }, + { + "epoch": 0.005060719959028134, + "grad_norm": 0.177734375, + "learning_rate": 0.0011639999999999999, + "loss": 0.3223, + "step": 583 + }, + { + "epoch": 0.005069400439232298, + "grad_norm": 0.130859375, + "learning_rate": 0.001166, + "loss": 0.2754, + "step": 584 + }, + { + "epoch": 0.005078080919436463, + "grad_norm": 0.09716796875, + "learning_rate": 0.001168, + "loss": 0.2891, + "step": 585 + }, + { + "epoch": 0.005086761399640628, + "grad_norm": 0.11328125, + "learning_rate": 0.00117, + "loss": 0.2578, + "step": 586 + }, + { + "epoch": 0.005095441879844793, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011719999999999999, + "loss": 0.2871, + "step": 587 + }, + { + "epoch": 0.005104122360048958, + "grad_norm": 0.111328125, + "learning_rate": 0.001174, + "loss": 0.2617, + "step": 588 + }, + { + "epoch": 0.005112802840253123, + "grad_norm": 0.1171875, + "learning_rate": 0.001176, + "loss": 0.3145, + "step": 589 + }, + { + "epoch": 0.005121483320457288, + "grad_norm": 0.10595703125, + "learning_rate": 0.001178, + "loss": 0.2734, + "step": 590 + }, + { + "epoch": 0.005130163800661453, + "grad_norm": 0.09716796875, + "learning_rate": 0.00118, + "loss": 0.2578, + "step": 591 + }, + { + "epoch": 0.005138844280865617, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011819999999999999, + "loss": 0.3203, + "step": 592 + }, + { + "epoch": 0.005147524761069782, + "grad_norm": 0.12109375, + "learning_rate": 0.001184, + "loss": 0.2637, + "step": 593 + }, + { + "epoch": 0.0051562052412739475, + "grad_norm": 0.158203125, + "learning_rate": 0.001186, + "loss": 0.2695, + "step": 594 + }, + { + "epoch": 0.005164885721478112, + "grad_norm": 0.380859375, + "learning_rate": 0.001188, + "loss": 0.3398, + "step": 595 + }, + { + "epoch": 0.005173566201682277, + "grad_norm": 0.154296875, + "learning_rate": 0.0011899999999999999, + "loss": 0.2734, + "step": 596 + }, + { + "epoch": 0.005182246681886442, + "grad_norm": 0.1396484375, + "learning_rate": 0.001192, + "loss": 0.3418, + "step": 597 + }, + { + "epoch": 0.005190927162090607, + "grad_norm": 0.162109375, + "learning_rate": 0.001194, + "loss": 0.2695, + "step": 598 + }, + { + "epoch": 0.005199607642294771, + "grad_norm": 0.1552734375, + "learning_rate": 0.001196, + "loss": 0.2656, + "step": 599 + }, + { + "epoch": 0.0052082881224989365, + "grad_norm": 0.16015625, + "learning_rate": 0.001198, + "loss": 0.3262, + "step": 600 + }, + { + "epoch": 0.005216968602703102, + "grad_norm": 0.158203125, + "learning_rate": 0.0012, + "loss": 0.2891, + "step": 601 + }, + { + "epoch": 0.005225649082907267, + "grad_norm": 0.228515625, + "learning_rate": 0.001202, + "loss": 0.3535, + "step": 602 + }, + { + "epoch": 0.005234329563111431, + "grad_norm": 0.1796875, + "learning_rate": 0.001204, + "loss": 0.2656, + "step": 603 + }, + { + "epoch": 0.005243010043315596, + "grad_norm": 0.1318359375, + "learning_rate": 0.001206, + "loss": 0.2539, + "step": 604 + }, + { + "epoch": 0.005251690523519761, + "grad_norm": 0.1572265625, + "learning_rate": 0.001208, + "loss": 0.2812, + "step": 605 + }, + { + "epoch": 0.005260371003723926, + "grad_norm": 0.12109375, + "learning_rate": 0.00121, + "loss": 0.2598, + "step": 606 + }, + { + "epoch": 0.005269051483928091, + "grad_norm": 0.248046875, + "learning_rate": 0.001212, + "loss": 0.293, + "step": 607 + }, + { + "epoch": 0.005277731964132256, + "grad_norm": 1.828125, + "learning_rate": 0.001214, + "loss": 0.7344, + "step": 608 + }, + { + "epoch": 0.005286412444336421, + "grad_norm": 0.12158203125, + "learning_rate": 0.001216, + "loss": 0.2539, + "step": 609 + }, + { + "epoch": 0.005295092924540586, + "grad_norm": 0.1396484375, + "learning_rate": 0.001218, + "loss": 0.3047, + "step": 610 + }, + { + "epoch": 0.00530377340474475, + "grad_norm": 0.1220703125, + "learning_rate": 0.00122, + "loss": 0.2188, + "step": 611 + }, + { + "epoch": 0.005312453884948915, + "grad_norm": 0.1484375, + "learning_rate": 0.001222, + "loss": 0.2852, + "step": 612 + }, + { + "epoch": 0.0053211343651530804, + "grad_norm": 0.23828125, + "learning_rate": 0.001224, + "loss": 0.2969, + "step": 613 + }, + { + "epoch": 0.005329814845357246, + "grad_norm": 0.302734375, + "learning_rate": 0.001226, + "loss": 0.3613, + "step": 614 + }, + { + "epoch": 0.00533849532556141, + "grad_norm": 0.265625, + "learning_rate": 0.001228, + "loss": 0.2617, + "step": 615 + }, + { + "epoch": 0.005347175805765575, + "grad_norm": 0.08935546875, + "learning_rate": 0.00123, + "loss": 0.2539, + "step": 616 + }, + { + "epoch": 0.00535585628596974, + "grad_norm": 0.134765625, + "learning_rate": 0.001232, + "loss": 0.2988, + "step": 617 + }, + { + "epoch": 0.005364536766173904, + "grad_norm": 0.1396484375, + "learning_rate": 0.001234, + "loss": 0.3164, + "step": 618 + }, + { + "epoch": 0.005373217246378069, + "grad_norm": 0.16796875, + "learning_rate": 0.0012360000000000001, + "loss": 0.2598, + "step": 619 + }, + { + "epoch": 0.0053818977265822346, + "grad_norm": 0.12451171875, + "learning_rate": 0.001238, + "loss": 0.3203, + "step": 620 + }, + { + "epoch": 0.0053905782067864, + "grad_norm": 0.11474609375, + "learning_rate": 0.00124, + "loss": 0.2891, + "step": 621 + }, + { + "epoch": 0.005399258686990564, + "grad_norm": 0.1162109375, + "learning_rate": 0.001242, + "loss": 0.3086, + "step": 622 + }, + { + "epoch": 0.005407939167194729, + "grad_norm": 0.1005859375, + "learning_rate": 0.001244, + "loss": 0.3008, + "step": 623 + }, + { + "epoch": 0.005416619647398894, + "grad_norm": 0.07958984375, + "learning_rate": 0.001246, + "loss": 0.2578, + "step": 624 + }, + { + "epoch": 0.005425300127603059, + "grad_norm": 0.12890625, + "learning_rate": 0.001248, + "loss": 0.2832, + "step": 625 + }, + { + "epoch": 0.0054339806078072235, + "grad_norm": 0.1318359375, + "learning_rate": 0.00125, + "loss": 0.3555, + "step": 626 + }, + { + "epoch": 0.005442661088011389, + "grad_norm": 0.07373046875, + "learning_rate": 0.001252, + "loss": 0.2354, + "step": 627 + }, + { + "epoch": 0.005451341568215554, + "grad_norm": 0.142578125, + "learning_rate": 0.0012540000000000001, + "loss": 0.2852, + "step": 628 + }, + { + "epoch": 0.005460022048419719, + "grad_norm": 0.11181640625, + "learning_rate": 0.001256, + "loss": 0.3789, + "step": 629 + }, + { + "epoch": 0.005468702528623883, + "grad_norm": 0.11376953125, + "learning_rate": 0.001258, + "loss": 0.2852, + "step": 630 + }, + { + "epoch": 0.005477383008828048, + "grad_norm": 0.146484375, + "learning_rate": 0.00126, + "loss": 0.3027, + "step": 631 + }, + { + "epoch": 0.005486063489032213, + "grad_norm": 0.1298828125, + "learning_rate": 0.001262, + "loss": 0.2871, + "step": 632 + }, + { + "epoch": 0.0054947439692363785, + "grad_norm": 0.11474609375, + "learning_rate": 0.001264, + "loss": 0.2988, + "step": 633 + }, + { + "epoch": 0.005503424449440543, + "grad_norm": 0.39453125, + "learning_rate": 0.001266, + "loss": 0.4531, + "step": 634 + }, + { + "epoch": 0.005512104929644708, + "grad_norm": 0.09716796875, + "learning_rate": 0.001268, + "loss": 0.252, + "step": 635 + }, + { + "epoch": 0.005520785409848873, + "grad_norm": 0.1474609375, + "learning_rate": 0.00127, + "loss": 0.291, + "step": 636 + }, + { + "epoch": 0.005529465890053038, + "grad_norm": 0.1875, + "learning_rate": 0.0012720000000000001, + "loss": 0.293, + "step": 637 + }, + { + "epoch": 0.005538146370257202, + "grad_norm": 0.2060546875, + "learning_rate": 0.001274, + "loss": 0.3555, + "step": 638 + }, + { + "epoch": 0.0055468268504613675, + "grad_norm": 0.34765625, + "learning_rate": 0.001276, + "loss": 0.375, + "step": 639 + }, + { + "epoch": 0.005555507330665533, + "grad_norm": 0.1416015625, + "learning_rate": 0.001278, + "loss": 0.2236, + "step": 640 + }, + { + "epoch": 0.005564187810869698, + "grad_norm": 0.10400390625, + "learning_rate": 0.00128, + "loss": 0.3066, + "step": 641 + }, + { + "epoch": 0.005572868291073862, + "grad_norm": 0.08740234375, + "learning_rate": 0.0012820000000000002, + "loss": 0.2891, + "step": 642 + }, + { + "epoch": 0.005581548771278027, + "grad_norm": 0.1279296875, + "learning_rate": 0.001284, + "loss": 0.3242, + "step": 643 + }, + { + "epoch": 0.005590229251482192, + "grad_norm": 0.146484375, + "learning_rate": 0.001286, + "loss": 0.3398, + "step": 644 + }, + { + "epoch": 0.0055989097316863565, + "grad_norm": 0.07763671875, + "learning_rate": 0.001288, + "loss": 0.2148, + "step": 645 + }, + { + "epoch": 0.005607590211890522, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012900000000000001, + "loss": 0.293, + "step": 646 + }, + { + "epoch": 0.005616270692094687, + "grad_norm": 0.11279296875, + "learning_rate": 0.001292, + "loss": 0.3281, + "step": 647 + }, + { + "epoch": 0.005624951172298852, + "grad_norm": 0.091796875, + "learning_rate": 0.001294, + "loss": 0.3105, + "step": 648 + }, + { + "epoch": 0.005633631652503016, + "grad_norm": 0.12109375, + "learning_rate": 0.001296, + "loss": 0.3555, + "step": 649 + }, + { + "epoch": 0.005642312132707181, + "grad_norm": 0.14453125, + "learning_rate": 0.0012980000000000001, + "loss": 0.3477, + "step": 650 + }, + { + "epoch": 0.005650992612911346, + "grad_norm": 0.10205078125, + "learning_rate": 0.0013000000000000002, + "loss": 0.2988, + "step": 651 + }, + { + "epoch": 0.0056596730931155114, + "grad_norm": 0.0908203125, + "learning_rate": 0.001302, + "loss": 0.2539, + "step": 652 + }, + { + "epoch": 0.005668353573319676, + "grad_norm": 0.11572265625, + "learning_rate": 0.001304, + "loss": 0.3086, + "step": 653 + }, + { + "epoch": 0.005677034053523841, + "grad_norm": 0.11083984375, + "learning_rate": 0.001306, + "loss": 0.3906, + "step": 654 + }, + { + "epoch": 0.005685714533728006, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013080000000000001, + "loss": 0.4121, + "step": 655 + }, + { + "epoch": 0.005694395013932171, + "grad_norm": 0.11572265625, + "learning_rate": 0.0013100000000000002, + "loss": 0.3281, + "step": 656 + }, + { + "epoch": 0.005703075494136335, + "grad_norm": 0.1396484375, + "learning_rate": 0.001312, + "loss": 0.3574, + "step": 657 + }, + { + "epoch": 0.0057117559743405, + "grad_norm": 0.0849609375, + "learning_rate": 0.001314, + "loss": 0.2422, + "step": 658 + }, + { + "epoch": 0.0057204364545446655, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013160000000000001, + "loss": 0.2676, + "step": 659 + }, + { + "epoch": 0.005729116934748831, + "grad_norm": 0.1328125, + "learning_rate": 0.0013180000000000002, + "loss": 0.3477, + "step": 660 + }, + { + "epoch": 0.005737797414952995, + "grad_norm": 0.11572265625, + "learning_rate": 0.00132, + "loss": 0.3008, + "step": 661 + }, + { + "epoch": 0.00574647789515716, + "grad_norm": 0.134765625, + "learning_rate": 0.001322, + "loss": 0.2852, + "step": 662 + }, + { + "epoch": 0.005755158375361325, + "grad_norm": 0.11279296875, + "learning_rate": 0.001324, + "loss": 0.2812, + "step": 663 + }, + { + "epoch": 0.00576383885556549, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013260000000000001, + "loss": 0.3086, + "step": 664 + }, + { + "epoch": 0.0057725193357696545, + "grad_norm": 0.08154296875, + "learning_rate": 0.0013280000000000002, + "loss": 0.2188, + "step": 665 + }, + { + "epoch": 0.00578119981597382, + "grad_norm": 0.12158203125, + "learning_rate": 0.00133, + "loss": 0.2891, + "step": 666 + }, + { + "epoch": 0.005789880296177985, + "grad_norm": 0.1181640625, + "learning_rate": 0.001332, + "loss": 0.332, + "step": 667 + }, + { + "epoch": 0.00579856077638215, + "grad_norm": 0.1953125, + "learning_rate": 0.0013340000000000001, + "loss": 0.3223, + "step": 668 + }, + { + "epoch": 0.005807241256586314, + "grad_norm": 0.119140625, + "learning_rate": 0.0013360000000000002, + "loss": 0.3359, + "step": 669 + }, + { + "epoch": 0.005815921736790479, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013380000000000002, + "loss": 0.2461, + "step": 670 + }, + { + "epoch": 0.005824602216994644, + "grad_norm": 0.103515625, + "learning_rate": 0.00134, + "loss": 0.2871, + "step": 671 + }, + { + "epoch": 0.005833282697198809, + "grad_norm": 0.10546875, + "learning_rate": 0.001342, + "loss": 0.2656, + "step": 672 + }, + { + "epoch": 0.005841963177402974, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013440000000000001, + "loss": 0.2559, + "step": 673 + }, + { + "epoch": 0.005850643657607139, + "grad_norm": 0.10400390625, + "learning_rate": 0.0013460000000000002, + "loss": 0.2656, + "step": 674 + }, + { + "epoch": 0.005859324137811304, + "grad_norm": 0.13671875, + "learning_rate": 0.001348, + "loss": 0.2793, + "step": 675 + }, + { + "epoch": 0.005868004618015468, + "grad_norm": 0.111328125, + "learning_rate": 0.00135, + "loss": 0.3008, + "step": 676 + }, + { + "epoch": 0.005876685098219633, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013520000000000001, + "loss": 0.2451, + "step": 677 + }, + { + "epoch": 0.0058853655784237985, + "grad_norm": 0.1142578125, + "learning_rate": 0.0013540000000000002, + "loss": 0.2812, + "step": 678 + }, + { + "epoch": 0.005894046058627964, + "grad_norm": 0.1044921875, + "learning_rate": 0.0013560000000000002, + "loss": 0.373, + "step": 679 + }, + { + "epoch": 0.005902726538832128, + "grad_norm": 0.11083984375, + "learning_rate": 0.001358, + "loss": 0.2168, + "step": 680 + }, + { + "epoch": 0.005911407019036293, + "grad_norm": 0.0966796875, + "learning_rate": 0.00136, + "loss": 0.2793, + "step": 681 + }, + { + "epoch": 0.005920087499240458, + "grad_norm": 0.12890625, + "learning_rate": 0.0013620000000000001, + "loss": 0.293, + "step": 682 + }, + { + "epoch": 0.005928767979444623, + "grad_norm": 0.0927734375, + "learning_rate": 0.0013640000000000002, + "loss": 0.249, + "step": 683 + }, + { + "epoch": 0.0059374484596487875, + "grad_norm": 0.10205078125, + "learning_rate": 0.001366, + "loss": 0.2461, + "step": 684 + }, + { + "epoch": 0.005946128939852953, + "grad_norm": 0.1357421875, + "learning_rate": 0.001368, + "loss": 0.3066, + "step": 685 + }, + { + "epoch": 0.005954809420057118, + "grad_norm": 0.095703125, + "learning_rate": 0.0013700000000000001, + "loss": 0.2598, + "step": 686 + }, + { + "epoch": 0.005963489900261283, + "grad_norm": 0.14453125, + "learning_rate": 0.0013720000000000002, + "loss": 0.2314, + "step": 687 + }, + { + "epoch": 0.005972170380465447, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013740000000000002, + "loss": 0.3086, + "step": 688 + }, + { + "epoch": 0.005980850860669612, + "grad_norm": 0.185546875, + "learning_rate": 0.0013759999999999998, + "loss": 0.2734, + "step": 689 + }, + { + "epoch": 0.005989531340873777, + "grad_norm": 0.1142578125, + "learning_rate": 0.0013779999999999999, + "loss": 0.1924, + "step": 690 + }, + { + "epoch": 0.0059982118210779424, + "grad_norm": 0.1787109375, + "learning_rate": 0.00138, + "loss": 0.2891, + "step": 691 + }, + { + "epoch": 0.006006892301282107, + "grad_norm": 0.1904296875, + "learning_rate": 0.001382, + "loss": 0.3047, + "step": 692 + }, + { + "epoch": 0.006015572781486272, + "grad_norm": 0.1357421875, + "learning_rate": 0.001384, + "loss": 0.2773, + "step": 693 + }, + { + "epoch": 0.006024253261690437, + "grad_norm": 0.10693359375, + "learning_rate": 0.0013859999999999999, + "loss": 0.21, + "step": 694 + }, + { + "epoch": 0.006032933741894601, + "grad_norm": 0.1279296875, + "learning_rate": 0.001388, + "loss": 0.2793, + "step": 695 + }, + { + "epoch": 0.006041614222098766, + "grad_norm": 0.134765625, + "learning_rate": 0.00139, + "loss": 0.3359, + "step": 696 + }, + { + "epoch": 0.006050294702302931, + "grad_norm": 0.12890625, + "learning_rate": 0.001392, + "loss": 0.2432, + "step": 697 + }, + { + "epoch": 0.0060589751825070965, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013939999999999998, + "loss": 0.3105, + "step": 698 + }, + { + "epoch": 0.006067655662711261, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013959999999999999, + "loss": 0.3457, + "step": 699 + }, + { + "epoch": 0.006076336142915426, + "grad_norm": 0.12109375, + "learning_rate": 0.001398, + "loss": 0.3086, + "step": 700 + }, + { + "epoch": 0.006085016623119591, + "grad_norm": 0.162109375, + "learning_rate": 0.0014, + "loss": 0.249, + "step": 701 + }, + { + "epoch": 0.006093697103323756, + "grad_norm": 0.1337890625, + "learning_rate": 0.001402, + "loss": 0.2812, + "step": 702 + }, + { + "epoch": 0.00610237758352792, + "grad_norm": 0.12890625, + "learning_rate": 0.0014039999999999999, + "loss": 0.3145, + "step": 703 + }, + { + "epoch": 0.0061110580637320855, + "grad_norm": 0.10302734375, + "learning_rate": 0.001406, + "loss": 0.2617, + "step": 704 + }, + { + "epoch": 0.006119738543936251, + "grad_norm": 0.126953125, + "learning_rate": 0.001408, + "loss": 0.2539, + "step": 705 + }, + { + "epoch": 0.006128419024140416, + "grad_norm": 0.0908203125, + "learning_rate": 0.00141, + "loss": 0.2432, + "step": 706 + }, + { + "epoch": 0.00613709950434458, + "grad_norm": 0.12158203125, + "learning_rate": 0.0014119999999999998, + "loss": 0.3047, + "step": 707 + }, + { + "epoch": 0.006145779984548745, + "grad_norm": 0.1318359375, + "learning_rate": 0.001414, + "loss": 0.4062, + "step": 708 + }, + { + "epoch": 0.00615446046475291, + "grad_norm": 0.107421875, + "learning_rate": 0.001416, + "loss": 0.2832, + "step": 709 + }, + { + "epoch": 0.006163140944957075, + "grad_norm": 0.0830078125, + "learning_rate": 0.001418, + "loss": 0.2578, + "step": 710 + }, + { + "epoch": 0.00617182142516124, + "grad_norm": 0.099609375, + "learning_rate": 0.00142, + "loss": 0.2949, + "step": 711 + }, + { + "epoch": 0.006180501905365405, + "grad_norm": 0.11279296875, + "learning_rate": 0.0014219999999999999, + "loss": 0.2949, + "step": 712 + }, + { + "epoch": 0.00618918238556957, + "grad_norm": 0.09228515625, + "learning_rate": 0.001424, + "loss": 0.252, + "step": 713 + }, + { + "epoch": 0.006197862865773735, + "grad_norm": 0.1416015625, + "learning_rate": 0.001426, + "loss": 0.3047, + "step": 714 + }, + { + "epoch": 0.006206543345977899, + "grad_norm": 0.1328125, + "learning_rate": 0.001428, + "loss": 0.25, + "step": 715 + }, + { + "epoch": 0.006215223826182064, + "grad_norm": 0.1513671875, + "learning_rate": 0.00143, + "loss": 0.3164, + "step": 716 + }, + { + "epoch": 0.0062239043063862295, + "grad_norm": 0.10009765625, + "learning_rate": 0.001432, + "loss": 0.2812, + "step": 717 + }, + { + "epoch": 0.006232584786590395, + "grad_norm": 0.115234375, + "learning_rate": 0.001434, + "loss": 0.2852, + "step": 718 + }, + { + "epoch": 0.006241265266794559, + "grad_norm": 0.10888671875, + "learning_rate": 0.001436, + "loss": 0.3008, + "step": 719 + }, + { + "epoch": 0.006249945746998724, + "grad_norm": 0.1376953125, + "learning_rate": 0.001438, + "loss": 0.2246, + "step": 720 + }, + { + "epoch": 0.006258626227202889, + "grad_norm": 0.12890625, + "learning_rate": 0.0014399999999999999, + "loss": 0.2949, + "step": 721 + }, + { + "epoch": 0.006267306707407053, + "grad_norm": 0.09716796875, + "learning_rate": 0.001442, + "loss": 0.2383, + "step": 722 + }, + { + "epoch": 0.0062759871876112185, + "grad_norm": 0.1044921875, + "learning_rate": 0.001444, + "loss": 0.2734, + "step": 723 + }, + { + "epoch": 0.006284667667815384, + "grad_norm": 0.0947265625, + "learning_rate": 0.001446, + "loss": 0.207, + "step": 724 + }, + { + "epoch": 0.006293348148019549, + "grad_norm": 0.11328125, + "learning_rate": 0.001448, + "loss": 0.3086, + "step": 725 + }, + { + "epoch": 0.006302028628223713, + "grad_norm": 0.11083984375, + "learning_rate": 0.00145, + "loss": 0.2188, + "step": 726 + }, + { + "epoch": 0.006310709108427878, + "grad_norm": 0.1396484375, + "learning_rate": 0.001452, + "loss": 0.3105, + "step": 727 + }, + { + "epoch": 0.006319389588632043, + "grad_norm": 0.10546875, + "learning_rate": 0.001454, + "loss": 0.3066, + "step": 728 + }, + { + "epoch": 0.006328070068836208, + "grad_norm": 0.12158203125, + "learning_rate": 0.001456, + "loss": 0.293, + "step": 729 + }, + { + "epoch": 0.0063367505490403726, + "grad_norm": 0.10302734375, + "learning_rate": 0.001458, + "loss": 0.2402, + "step": 730 + }, + { + "epoch": 0.006345431029244538, + "grad_norm": 0.11669921875, + "learning_rate": 0.00146, + "loss": 0.2773, + "step": 731 + }, + { + "epoch": 0.006354111509448703, + "grad_norm": 0.11865234375, + "learning_rate": 0.001462, + "loss": 0.2559, + "step": 732 + }, + { + "epoch": 0.006362791989652868, + "grad_norm": 0.126953125, + "learning_rate": 0.001464, + "loss": 0.2988, + "step": 733 + }, + { + "epoch": 0.006371472469857032, + "grad_norm": 0.09423828125, + "learning_rate": 0.001466, + "loss": 0.2246, + "step": 734 + }, + { + "epoch": 0.006380152950061197, + "grad_norm": 0.1318359375, + "learning_rate": 0.001468, + "loss": 0.3301, + "step": 735 + }, + { + "epoch": 0.006388833430265362, + "grad_norm": 0.115234375, + "learning_rate": 0.00147, + "loss": 0.2422, + "step": 736 + }, + { + "epoch": 0.0063975139104695275, + "grad_norm": 0.1591796875, + "learning_rate": 0.001472, + "loss": 0.2539, + "step": 737 + }, + { + "epoch": 0.006406194390673692, + "grad_norm": 0.08984375, + "learning_rate": 0.001474, + "loss": 0.2773, + "step": 738 + }, + { + "epoch": 0.006414874870877857, + "grad_norm": 0.07568359375, + "learning_rate": 0.001476, + "loss": 0.2539, + "step": 739 + }, + { + "epoch": 0.006423555351082022, + "grad_norm": 0.1474609375, + "learning_rate": 0.001478, + "loss": 0.3477, + "step": 740 + }, + { + "epoch": 0.006432235831286187, + "grad_norm": 0.12890625, + "learning_rate": 0.00148, + "loss": 0.3477, + "step": 741 + }, + { + "epoch": 0.006440916311490351, + "grad_norm": 0.16796875, + "learning_rate": 0.001482, + "loss": 0.2773, + "step": 742 + }, + { + "epoch": 0.0064495967916945165, + "grad_norm": 0.12060546875, + "learning_rate": 0.001484, + "loss": 0.2754, + "step": 743 + }, + { + "epoch": 0.006458277271898682, + "grad_norm": 0.115234375, + "learning_rate": 0.0014860000000000001, + "loss": 0.3145, + "step": 744 + }, + { + "epoch": 0.006466957752102847, + "grad_norm": 0.1259765625, + "learning_rate": 0.001488, + "loss": 0.3066, + "step": 745 + }, + { + "epoch": 0.006475638232307011, + "grad_norm": 0.0908203125, + "learning_rate": 0.00149, + "loss": 0.2393, + "step": 746 + }, + { + "epoch": 0.006484318712511176, + "grad_norm": 0.10791015625, + "learning_rate": 0.001492, + "loss": 0.2461, + "step": 747 + }, + { + "epoch": 0.006492999192715341, + "grad_norm": 0.10546875, + "learning_rate": 0.001494, + "loss": 0.2617, + "step": 748 + }, + { + "epoch": 0.0065016796729195055, + "grad_norm": 0.08935546875, + "learning_rate": 0.001496, + "loss": 0.249, + "step": 749 + }, + { + "epoch": 0.006510360153123671, + "grad_norm": 0.11865234375, + "learning_rate": 0.001498, + "loss": 0.2656, + "step": 750 + }, + { + "epoch": 0.006519040633327836, + "grad_norm": 0.12109375, + "learning_rate": 0.0015, + "loss": 0.2578, + "step": 751 + }, + { + "epoch": 0.006527721113532001, + "grad_norm": 0.1162109375, + "learning_rate": 0.001502, + "loss": 0.2559, + "step": 752 + }, + { + "epoch": 0.006536401593736165, + "grad_norm": 0.0888671875, + "learning_rate": 0.0015040000000000001, + "loss": 0.2285, + "step": 753 + }, + { + "epoch": 0.00654508207394033, + "grad_norm": 0.1591796875, + "learning_rate": 0.001506, + "loss": 0.3125, + "step": 754 + }, + { + "epoch": 0.006553762554144495, + "grad_norm": 0.12353515625, + "learning_rate": 0.001508, + "loss": 0.2451, + "step": 755 + }, + { + "epoch": 0.0065624430343486605, + "grad_norm": 0.11181640625, + "learning_rate": 0.00151, + "loss": 0.3027, + "step": 756 + }, + { + "epoch": 0.006571123514552825, + "grad_norm": 0.1015625, + "learning_rate": 0.001512, + "loss": 0.2734, + "step": 757 + }, + { + "epoch": 0.00657980399475699, + "grad_norm": 0.12451171875, + "learning_rate": 0.001514, + "loss": 0.3281, + "step": 758 + }, + { + "epoch": 0.006588484474961155, + "grad_norm": 0.1201171875, + "learning_rate": 0.001516, + "loss": 0.2305, + "step": 759 + }, + { + "epoch": 0.00659716495516532, + "grad_norm": 0.1025390625, + "learning_rate": 0.001518, + "loss": 0.2314, + "step": 760 + }, + { + "epoch": 0.006605845435369484, + "grad_norm": 0.140625, + "learning_rate": 0.00152, + "loss": 0.2773, + "step": 761 + }, + { + "epoch": 0.0066145259155736495, + "grad_norm": 0.12255859375, + "learning_rate": 0.0015220000000000001, + "loss": 0.2227, + "step": 762 + }, + { + "epoch": 0.006623206395777815, + "grad_norm": 0.1123046875, + "learning_rate": 0.001524, + "loss": 0.2617, + "step": 763 + }, + { + "epoch": 0.00663188687598198, + "grad_norm": 0.1005859375, + "learning_rate": 0.001526, + "loss": 0.25, + "step": 764 + }, + { + "epoch": 0.006640567356186144, + "grad_norm": 0.08740234375, + "learning_rate": 0.001528, + "loss": 0.207, + "step": 765 + }, + { + "epoch": 0.006649247836390309, + "grad_norm": 0.09814453125, + "learning_rate": 0.0015300000000000001, + "loss": 0.2598, + "step": 766 + }, + { + "epoch": 0.006657928316594474, + "grad_norm": 0.09130859375, + "learning_rate": 0.0015320000000000002, + "loss": 0.2578, + "step": 767 + }, + { + "epoch": 0.006666608796798639, + "grad_norm": 0.10009765625, + "learning_rate": 0.001534, + "loss": 0.2617, + "step": 768 + }, + { + "epoch": 0.0066752892770028036, + "grad_norm": 0.10693359375, + "learning_rate": 0.001536, + "loss": 0.3008, + "step": 769 + }, + { + "epoch": 0.006683969757206969, + "grad_norm": 0.1279296875, + "learning_rate": 0.001538, + "loss": 0.3613, + "step": 770 + }, + { + "epoch": 0.006692650237411134, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015400000000000001, + "loss": 0.2422, + "step": 771 + }, + { + "epoch": 0.006701330717615298, + "grad_norm": 0.1982421875, + "learning_rate": 0.001542, + "loss": 0.3379, + "step": 772 + }, + { + "epoch": 0.006710011197819463, + "grad_norm": 0.1357421875, + "learning_rate": 0.001544, + "loss": 0.2695, + "step": 773 + }, + { + "epoch": 0.006718691678023628, + "grad_norm": 0.1318359375, + "learning_rate": 0.001546, + "loss": 0.3477, + "step": 774 + }, + { + "epoch": 0.006727372158227793, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015480000000000001, + "loss": 0.252, + "step": 775 + }, + { + "epoch": 0.006736052638431958, + "grad_norm": 0.09130859375, + "learning_rate": 0.0015500000000000002, + "loss": 0.2637, + "step": 776 + }, + { + "epoch": 0.006744733118636123, + "grad_norm": 0.099609375, + "learning_rate": 0.001552, + "loss": 0.2559, + "step": 777 + }, + { + "epoch": 0.006753413598840288, + "grad_norm": 0.44921875, + "learning_rate": 0.001554, + "loss": 0.4492, + "step": 778 + }, + { + "epoch": 0.006762094079044453, + "grad_norm": 0.1044921875, + "learning_rate": 0.001556, + "loss": 0.2188, + "step": 779 + }, + { + "epoch": 0.006770774559248617, + "grad_norm": 0.244140625, + "learning_rate": 0.0015580000000000001, + "loss": 0.2656, + "step": 780 + }, + { + "epoch": 0.006779455039452782, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015600000000000002, + "loss": 0.2793, + "step": 781 + }, + { + "epoch": 0.0067881355196569475, + "grad_norm": 0.130859375, + "learning_rate": 0.001562, + "loss": 0.25, + "step": 782 + }, + { + "epoch": 0.006796815999861113, + "grad_norm": 0.146484375, + "learning_rate": 0.001564, + "loss": 0.2217, + "step": 783 + }, + { + "epoch": 0.006805496480065277, + "grad_norm": 0.146484375, + "learning_rate": 0.0015660000000000001, + "loss": 0.2422, + "step": 784 + }, + { + "epoch": 0.006814176960269442, + "grad_norm": 0.1201171875, + "learning_rate": 0.0015680000000000002, + "loss": 0.2812, + "step": 785 + }, + { + "epoch": 0.006822857440473607, + "grad_norm": 0.1259765625, + "learning_rate": 0.00157, + "loss": 0.2188, + "step": 786 + }, + { + "epoch": 0.006831537920677772, + "grad_norm": 0.134765625, + "learning_rate": 0.001572, + "loss": 0.2305, + "step": 787 + }, + { + "epoch": 0.0068402184008819365, + "grad_norm": 0.1162109375, + "learning_rate": 0.001574, + "loss": 0.2812, + "step": 788 + }, + { + "epoch": 0.006848898881086102, + "grad_norm": 0.091796875, + "learning_rate": 0.0015760000000000001, + "loss": 0.2168, + "step": 789 + }, + { + "epoch": 0.006857579361290267, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015780000000000002, + "loss": 0.252, + "step": 790 + }, + { + "epoch": 0.006866259841494432, + "grad_norm": 0.126953125, + "learning_rate": 0.00158, + "loss": 0.3359, + "step": 791 + }, + { + "epoch": 0.006874940321698596, + "grad_norm": 0.1416015625, + "learning_rate": 0.001582, + "loss": 0.3125, + "step": 792 + }, + { + "epoch": 0.006883620801902761, + "grad_norm": 0.09765625, + "learning_rate": 0.0015840000000000001, + "loss": 0.293, + "step": 793 + }, + { + "epoch": 0.006892301282106926, + "grad_norm": 0.0693359375, + "learning_rate": 0.0015860000000000002, + "loss": 0.2617, + "step": 794 + }, + { + "epoch": 0.0069009817623110915, + "grad_norm": 0.119140625, + "learning_rate": 0.0015880000000000002, + "loss": 0.3008, + "step": 795 + }, + { + "epoch": 0.006909662242515256, + "grad_norm": 0.1025390625, + "learning_rate": 0.00159, + "loss": 0.2949, + "step": 796 + }, + { + "epoch": 0.006918342722719421, + "grad_norm": 0.1455078125, + "learning_rate": 0.001592, + "loss": 0.3262, + "step": 797 + }, + { + "epoch": 0.006927023202923586, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015940000000000001, + "loss": 0.2637, + "step": 798 + }, + { + "epoch": 0.00693570368312775, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015960000000000002, + "loss": 0.25, + "step": 799 + }, + { + "epoch": 0.006944384163331915, + "grad_norm": 0.1455078125, + "learning_rate": 0.001598, + "loss": 0.2148, + "step": 800 + }, + { + "epoch": 0.0069530646435360804, + "grad_norm": 0.10107421875, + "learning_rate": 0.0016, + "loss": 0.3086, + "step": 801 + }, + { + "epoch": 0.006961745123740246, + "grad_norm": 0.10302734375, + "learning_rate": 0.0016020000000000001, + "loss": 0.2891, + "step": 802 + }, + { + "epoch": 0.00697042560394441, + "grad_norm": 0.09326171875, + "learning_rate": 0.0016040000000000002, + "loss": 0.2207, + "step": 803 + }, + { + "epoch": 0.006979106084148575, + "grad_norm": 0.10986328125, + "learning_rate": 0.0016060000000000002, + "loss": 0.2852, + "step": 804 + }, + { + "epoch": 0.00698778656435274, + "grad_norm": 0.10546875, + "learning_rate": 0.001608, + "loss": 0.2793, + "step": 805 + }, + { + "epoch": 0.006996467044556905, + "grad_norm": 0.14453125, + "learning_rate": 0.00161, + "loss": 0.291, + "step": 806 + }, + { + "epoch": 0.007005147524761069, + "grad_norm": 0.115234375, + "learning_rate": 0.0016120000000000002, + "loss": 0.3086, + "step": 807 + }, + { + "epoch": 0.0070138280049652345, + "grad_norm": 0.103515625, + "learning_rate": 0.0016140000000000002, + "loss": 0.2373, + "step": 808 + }, + { + "epoch": 0.0070225084851694, + "grad_norm": 0.0966796875, + "learning_rate": 0.001616, + "loss": 0.209, + "step": 809 + }, + { + "epoch": 0.007031188965373565, + "grad_norm": 0.1611328125, + "learning_rate": 0.001618, + "loss": 0.3398, + "step": 810 + }, + { + "epoch": 0.007039869445577729, + "grad_norm": 0.080078125, + "learning_rate": 0.0016200000000000001, + "loss": 0.1973, + "step": 811 + }, + { + "epoch": 0.007048549925781894, + "grad_norm": 0.162109375, + "learning_rate": 0.0016220000000000002, + "loss": 0.3281, + "step": 812 + }, + { + "epoch": 0.007057230405986059, + "grad_norm": 0.1845703125, + "learning_rate": 0.0016240000000000002, + "loss": 0.2832, + "step": 813 + }, + { + "epoch": 0.007065910886190224, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016259999999999998, + "loss": 0.2793, + "step": 814 + }, + { + "epoch": 0.007074591366394389, + "grad_norm": 0.08154296875, + "learning_rate": 0.0016279999999999999, + "loss": 0.2275, + "step": 815 + }, + { + "epoch": 0.007083271846598554, + "grad_norm": 0.0771484375, + "learning_rate": 0.00163, + "loss": 0.2295, + "step": 816 + }, + { + "epoch": 0.007091952326802719, + "grad_norm": 0.11474609375, + "learning_rate": 0.001632, + "loss": 0.2988, + "step": 817 + }, + { + "epoch": 0.007100632807006884, + "grad_norm": 0.1884765625, + "learning_rate": 0.001634, + "loss": 0.4414, + "step": 818 + }, + { + "epoch": 0.007109313287211048, + "grad_norm": 0.134765625, + "learning_rate": 0.0016359999999999999, + "loss": 0.3105, + "step": 819 + }, + { + "epoch": 0.007117993767415213, + "grad_norm": 0.10205078125, + "learning_rate": 0.001638, + "loss": 0.2227, + "step": 820 + }, + { + "epoch": 0.0071266742476193785, + "grad_norm": 0.10205078125, + "learning_rate": 0.00164, + "loss": 0.2852, + "step": 821 + }, + { + "epoch": 0.007135354727823544, + "grad_norm": 0.10498046875, + "learning_rate": 0.001642, + "loss": 0.208, + "step": 822 + }, + { + "epoch": 0.007144035208027708, + "grad_norm": 0.1181640625, + "learning_rate": 0.0016439999999999998, + "loss": 0.3652, + "step": 823 + }, + { + "epoch": 0.007152715688231873, + "grad_norm": 0.11572265625, + "learning_rate": 0.001646, + "loss": 0.2832, + "step": 824 + }, + { + "epoch": 0.007161396168436038, + "grad_norm": 0.099609375, + "learning_rate": 0.001648, + "loss": 0.2715, + "step": 825 + }, + { + "epoch": 0.007170076648640202, + "grad_norm": 0.10791015625, + "learning_rate": 0.00165, + "loss": 0.2305, + "step": 826 + }, + { + "epoch": 0.0071787571288443675, + "grad_norm": 0.111328125, + "learning_rate": 0.001652, + "loss": 0.252, + "step": 827 + }, + { + "epoch": 0.007187437609048533, + "grad_norm": 0.09423828125, + "learning_rate": 0.0016539999999999999, + "loss": 0.25, + "step": 828 + }, + { + "epoch": 0.007196118089252698, + "grad_norm": 0.107421875, + "learning_rate": 0.001656, + "loss": 0.2734, + "step": 829 + }, + { + "epoch": 0.007204798569456862, + "grad_norm": 0.10302734375, + "learning_rate": 0.001658, + "loss": 0.21, + "step": 830 + }, + { + "epoch": 0.007213479049661027, + "grad_norm": 0.10400390625, + "learning_rate": 0.00166, + "loss": 0.2412, + "step": 831 + }, + { + "epoch": 0.007222159529865192, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016619999999999998, + "loss": 0.252, + "step": 832 + }, + { + "epoch": 0.007230840010069357, + "grad_norm": 0.087890625, + "learning_rate": 0.001664, + "loss": 0.2734, + "step": 833 + }, + { + "epoch": 0.007239520490273522, + "grad_norm": 0.0966796875, + "learning_rate": 0.001666, + "loss": 0.208, + "step": 834 + }, + { + "epoch": 0.007248200970477687, + "grad_norm": 0.0888671875, + "learning_rate": 0.001668, + "loss": 0.2422, + "step": 835 + }, + { + "epoch": 0.007256881450681852, + "grad_norm": 0.07373046875, + "learning_rate": 0.00167, + "loss": 0.2539, + "step": 836 + }, + { + "epoch": 0.007265561930886017, + "grad_norm": 0.09765625, + "learning_rate": 0.0016719999999999999, + "loss": 0.3086, + "step": 837 + }, + { + "epoch": 0.007274242411090181, + "grad_norm": 0.0859375, + "learning_rate": 0.001674, + "loss": 0.2324, + "step": 838 + }, + { + "epoch": 0.007282922891294346, + "grad_norm": 0.0927734375, + "learning_rate": 0.001676, + "loss": 0.2734, + "step": 839 + }, + { + "epoch": 0.0072916033714985114, + "grad_norm": 0.14453125, + "learning_rate": 0.001678, + "loss": 0.293, + "step": 840 + }, + { + "epoch": 0.0073002838517026766, + "grad_norm": 0.0908203125, + "learning_rate": 0.00168, + "loss": 0.2246, + "step": 841 + }, + { + "epoch": 0.007308964331906841, + "grad_norm": 0.1298828125, + "learning_rate": 0.001682, + "loss": 0.2734, + "step": 842 + }, + { + "epoch": 0.007317644812111006, + "grad_norm": 0.1279296875, + "learning_rate": 0.001684, + "loss": 0.3281, + "step": 843 + }, + { + "epoch": 0.007326325292315171, + "grad_norm": 0.10693359375, + "learning_rate": 0.001686, + "loss": 0.3008, + "step": 844 + }, + { + "epoch": 0.007335005772519336, + "grad_norm": 0.115234375, + "learning_rate": 0.001688, + "loss": 0.25, + "step": 845 + }, + { + "epoch": 0.0073436862527235, + "grad_norm": 0.12890625, + "learning_rate": 0.0016899999999999999, + "loss": 0.3027, + "step": 846 + }, + { + "epoch": 0.0073523667329276655, + "grad_norm": 0.0927734375, + "learning_rate": 0.001692, + "loss": 0.2207, + "step": 847 + }, + { + "epoch": 0.007361047213131831, + "grad_norm": 0.1259765625, + "learning_rate": 0.001694, + "loss": 0.2793, + "step": 848 + }, + { + "epoch": 0.007369727693335995, + "grad_norm": 0.11376953125, + "learning_rate": 0.001696, + "loss": 0.2031, + "step": 849 + }, + { + "epoch": 0.00737840817354016, + "grad_norm": 0.130859375, + "learning_rate": 0.001698, + "loss": 0.2871, + "step": 850 + }, + { + "epoch": 0.007387088653744325, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017, + "loss": 0.2949, + "step": 851 + }, + { + "epoch": 0.00739576913394849, + "grad_norm": 0.11572265625, + "learning_rate": 0.001702, + "loss": 0.2285, + "step": 852 + }, + { + "epoch": 0.0074044496141526545, + "grad_norm": 0.1044921875, + "learning_rate": 0.001704, + "loss": 0.2275, + "step": 853 + }, + { + "epoch": 0.00741313009435682, + "grad_norm": 0.12060546875, + "learning_rate": 0.001706, + "loss": 0.3008, + "step": 854 + }, + { + "epoch": 0.007421810574560985, + "grad_norm": 0.1171875, + "learning_rate": 0.001708, + "loss": 0.3184, + "step": 855 + }, + { + "epoch": 0.00743049105476515, + "grad_norm": 0.10546875, + "learning_rate": 0.00171, + "loss": 0.2695, + "step": 856 + }, + { + "epoch": 0.007439171534969314, + "grad_norm": 0.09130859375, + "learning_rate": 0.001712, + "loss": 0.2637, + "step": 857 + }, + { + "epoch": 0.007447852015173479, + "grad_norm": 0.12890625, + "learning_rate": 0.001714, + "loss": 0.3008, + "step": 858 + }, + { + "epoch": 0.007456532495377644, + "grad_norm": 3.140625, + "learning_rate": 0.001716, + "loss": 0.6523, + "step": 859 + }, + { + "epoch": 0.0074652129755818095, + "grad_norm": 0.1474609375, + "learning_rate": 0.001718, + "loss": 0.3145, + "step": 860 + }, + { + "epoch": 0.007473893455785974, + "grad_norm": 0.10791015625, + "learning_rate": 0.00172, + "loss": 0.291, + "step": 861 + }, + { + "epoch": 0.007482573935990139, + "grad_norm": 0.11279296875, + "learning_rate": 0.001722, + "loss": 0.3359, + "step": 862 + }, + { + "epoch": 0.007491254416194304, + "grad_norm": 0.09423828125, + "learning_rate": 0.001724, + "loss": 0.2363, + "step": 863 + }, + { + "epoch": 0.007499934896398469, + "grad_norm": 0.099609375, + "learning_rate": 0.001726, + "loss": 0.2734, + "step": 864 + }, + { + "epoch": 0.007508615376602633, + "grad_norm": 0.11962890625, + "learning_rate": 0.001728, + "loss": 0.2812, + "step": 865 + }, + { + "epoch": 0.0075172958568067985, + "grad_norm": 0.0888671875, + "learning_rate": 0.00173, + "loss": 0.2695, + "step": 866 + }, + { + "epoch": 0.007525976337010964, + "grad_norm": 0.09521484375, + "learning_rate": 0.001732, + "loss": 0.2734, + "step": 867 + }, + { + "epoch": 0.007534656817215129, + "grad_norm": 0.271484375, + "learning_rate": 0.001734, + "loss": 0.2832, + "step": 868 + }, + { + "epoch": 0.007543337297419293, + "grad_norm": 0.1044921875, + "learning_rate": 0.0017360000000000001, + "loss": 0.2988, + "step": 869 + }, + { + "epoch": 0.007552017777623458, + "grad_norm": 0.0966796875, + "learning_rate": 0.001738, + "loss": 0.2539, + "step": 870 + }, + { + "epoch": 0.007560698257827623, + "grad_norm": 0.126953125, + "learning_rate": 0.00174, + "loss": 0.2578, + "step": 871 + }, + { + "epoch": 0.007569378738031788, + "grad_norm": 0.1416015625, + "learning_rate": 0.001742, + "loss": 0.248, + "step": 872 + }, + { + "epoch": 0.007578059218235953, + "grad_norm": 0.12353515625, + "learning_rate": 0.001744, + "loss": 0.3066, + "step": 873 + }, + { + "epoch": 0.007586739698440118, + "grad_norm": 0.09423828125, + "learning_rate": 0.001746, + "loss": 0.2148, + "step": 874 + }, + { + "epoch": 0.007595420178644283, + "grad_norm": 0.125, + "learning_rate": 0.001748, + "loss": 0.3125, + "step": 875 + }, + { + "epoch": 0.007604100658848447, + "grad_norm": 0.11669921875, + "learning_rate": 0.00175, + "loss": 0.2441, + "step": 876 + }, + { + "epoch": 0.007612781139052612, + "grad_norm": 0.1474609375, + "learning_rate": 0.001752, + "loss": 0.2617, + "step": 877 + }, + { + "epoch": 0.007621461619256777, + "grad_norm": 0.103515625, + "learning_rate": 0.0017540000000000001, + "loss": 0.2715, + "step": 878 + }, + { + "epoch": 0.0076301420994609424, + "grad_norm": 0.78515625, + "learning_rate": 0.001756, + "loss": 0.6016, + "step": 879 + }, + { + "epoch": 0.007638822579665107, + "grad_norm": 0.08251953125, + "learning_rate": 0.001758, + "loss": 0.1846, + "step": 880 + }, + { + "epoch": 0.007647503059869272, + "grad_norm": 0.1376953125, + "learning_rate": 0.00176, + "loss": 0.3008, + "step": 881 + }, + { + "epoch": 0.007656183540073437, + "grad_norm": 0.12060546875, + "learning_rate": 0.0017620000000000001, + "loss": 0.2715, + "step": 882 + }, + { + "epoch": 0.007664864020277602, + "grad_norm": 0.1318359375, + "learning_rate": 0.001764, + "loss": 0.293, + "step": 883 + }, + { + "epoch": 0.007673544500481766, + "grad_norm": 0.130859375, + "learning_rate": 0.001766, + "loss": 0.2695, + "step": 884 + }, + { + "epoch": 0.007682224980685931, + "grad_norm": 0.1259765625, + "learning_rate": 0.001768, + "loss": 0.2812, + "step": 885 + }, + { + "epoch": 0.0076909054608900965, + "grad_norm": 0.150390625, + "learning_rate": 0.00177, + "loss": 0.375, + "step": 886 + }, + { + "epoch": 0.007699585941094262, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017720000000000001, + "loss": 0.2578, + "step": 887 + }, + { + "epoch": 0.007708266421298426, + "grad_norm": 0.095703125, + "learning_rate": 0.001774, + "loss": 0.2637, + "step": 888 + }, + { + "epoch": 0.007716946901502591, + "grad_norm": 0.1376953125, + "learning_rate": 0.001776, + "loss": 0.2656, + "step": 889 + }, + { + "epoch": 0.007725627381706756, + "grad_norm": 0.07421875, + "learning_rate": 0.001778, + "loss": 0.2393, + "step": 890 + }, + { + "epoch": 0.007734307861910921, + "grad_norm": 0.15234375, + "learning_rate": 0.0017800000000000001, + "loss": 0.3184, + "step": 891 + }, + { + "epoch": 0.0077429883421150855, + "grad_norm": 0.0859375, + "learning_rate": 0.0017820000000000002, + "loss": 0.2676, + "step": 892 + }, + { + "epoch": 0.007751668822319251, + "grad_norm": 0.1328125, + "learning_rate": 0.001784, + "loss": 0.3086, + "step": 893 + }, + { + "epoch": 0.007760349302523416, + "grad_norm": 0.09814453125, + "learning_rate": 0.001786, + "loss": 0.293, + "step": 894 + }, + { + "epoch": 0.007769029782727581, + "grad_norm": 0.072265625, + "learning_rate": 0.001788, + "loss": 0.1904, + "step": 895 + }, + { + "epoch": 0.007777710262931745, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017900000000000001, + "loss": 0.3086, + "step": 896 + }, + { + "epoch": 0.00778639074313591, + "grad_norm": 0.08935546875, + "learning_rate": 0.001792, + "loss": 0.2305, + "step": 897 + }, + { + "epoch": 0.007795071223340075, + "grad_norm": 0.11083984375, + "learning_rate": 0.001794, + "loss": 0.3105, + "step": 898 + }, + { + "epoch": 0.0078037517035442405, + "grad_norm": 0.11865234375, + "learning_rate": 0.001796, + "loss": 0.3223, + "step": 899 + }, + { + "epoch": 0.007812432183748405, + "grad_norm": 0.19140625, + "learning_rate": 0.0017980000000000001, + "loss": 0.3691, + "step": 900 + }, + { + "epoch": 0.00782111266395257, + "grad_norm": 0.16796875, + "learning_rate": 0.0018000000000000002, + "loss": 0.3047, + "step": 901 + }, + { + "epoch": 0.007829793144156735, + "grad_norm": 0.1318359375, + "learning_rate": 0.001802, + "loss": 0.2031, + "step": 902 + }, + { + "epoch": 0.0078384736243609, + "grad_norm": 0.10986328125, + "learning_rate": 0.001804, + "loss": 0.2422, + "step": 903 + }, + { + "epoch": 0.007847154104565065, + "grad_norm": 0.51953125, + "learning_rate": 0.001806, + "loss": 0.3965, + "step": 904 + }, + { + "epoch": 0.007855834584769229, + "grad_norm": 0.138671875, + "learning_rate": 0.0018080000000000001, + "loss": 0.2441, + "step": 905 + }, + { + "epoch": 0.007864515064973394, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018100000000000002, + "loss": 0.2656, + "step": 906 + }, + { + "epoch": 0.007873195545177559, + "grad_norm": 0.1435546875, + "learning_rate": 0.001812, + "loss": 0.2188, + "step": 907 + }, + { + "epoch": 0.007881876025381724, + "grad_norm": 0.10498046875, + "learning_rate": 0.001814, + "loss": 0.3027, + "step": 908 + }, + { + "epoch": 0.007890556505585889, + "grad_norm": 0.51171875, + "learning_rate": 0.0018160000000000001, + "loss": 0.4023, + "step": 909 + }, + { + "epoch": 0.007899236985790054, + "grad_norm": 0.08251953125, + "learning_rate": 0.0018180000000000002, + "loss": 0.2598, + "step": 910 + }, + { + "epoch": 0.00790791746599422, + "grad_norm": 0.09375, + "learning_rate": 0.00182, + "loss": 0.2715, + "step": 911 + }, + { + "epoch": 0.007916597946198384, + "grad_norm": 0.0791015625, + "learning_rate": 0.001822, + "loss": 0.2275, + "step": 912 + }, + { + "epoch": 0.007925278426402548, + "grad_norm": 0.0908203125, + "learning_rate": 0.001824, + "loss": 0.2891, + "step": 913 + }, + { + "epoch": 0.007933958906606713, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018260000000000001, + "loss": 0.293, + "step": 914 + }, + { + "epoch": 0.007942639386810878, + "grad_norm": 0.083984375, + "learning_rate": 0.0018280000000000002, + "loss": 0.2402, + "step": 915 + }, + { + "epoch": 0.007951319867015043, + "grad_norm": 0.08984375, + "learning_rate": 0.00183, + "loss": 0.2598, + "step": 916 + }, + { + "epoch": 0.007960000347219208, + "grad_norm": 0.080078125, + "learning_rate": 0.001832, + "loss": 0.3125, + "step": 917 + }, + { + "epoch": 0.007968680827423373, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018340000000000001, + "loss": 0.2578, + "step": 918 + }, + { + "epoch": 0.007977361307627539, + "grad_norm": 0.333984375, + "learning_rate": 0.0018360000000000002, + "loss": 0.4102, + "step": 919 + }, + { + "epoch": 0.007986041787831704, + "grad_norm": 0.28515625, + "learning_rate": 0.0018380000000000002, + "loss": 0.7422, + "step": 920 + }, + { + "epoch": 0.007994722268035867, + "grad_norm": 0.10009765625, + "learning_rate": 0.00184, + "loss": 0.2617, + "step": 921 + }, + { + "epoch": 0.008003402748240032, + "grad_norm": 0.0966796875, + "learning_rate": 0.001842, + "loss": 0.2441, + "step": 922 + }, + { + "epoch": 0.008012083228444197, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018440000000000002, + "loss": 0.2227, + "step": 923 + }, + { + "epoch": 0.008020763708648362, + "grad_norm": 0.0693359375, + "learning_rate": 0.0018460000000000002, + "loss": 0.2061, + "step": 924 + }, + { + "epoch": 0.008029444188852528, + "grad_norm": 0.087890625, + "learning_rate": 0.001848, + "loss": 0.252, + "step": 925 + }, + { + "epoch": 0.008038124669056693, + "grad_norm": 0.09814453125, + "learning_rate": 0.00185, + "loss": 0.2559, + "step": 926 + }, + { + "epoch": 0.008046805149260858, + "grad_norm": 0.07666015625, + "learning_rate": 0.0018520000000000001, + "loss": 0.2197, + "step": 927 + }, + { + "epoch": 0.008055485629465021, + "grad_norm": 0.0859375, + "learning_rate": 0.0018540000000000002, + "loss": 0.2617, + "step": 928 + }, + { + "epoch": 0.008064166109669186, + "grad_norm": 0.099609375, + "learning_rate": 0.0018560000000000002, + "loss": 0.2773, + "step": 929 + }, + { + "epoch": 0.008072846589873351, + "grad_norm": 0.119140625, + "learning_rate": 0.001858, + "loss": 0.2676, + "step": 930 + }, + { + "epoch": 0.008081527070077517, + "grad_norm": 0.09033203125, + "learning_rate": 0.00186, + "loss": 0.252, + "step": 931 + }, + { + "epoch": 0.008090207550281682, + "grad_norm": 0.0859375, + "learning_rate": 0.0018620000000000002, + "loss": 0.2363, + "step": 932 + }, + { + "epoch": 0.008098888030485847, + "grad_norm": 0.07275390625, + "learning_rate": 0.0018640000000000002, + "loss": 0.2188, + "step": 933 + }, + { + "epoch": 0.008107568510690012, + "grad_norm": 0.1845703125, + "learning_rate": 0.001866, + "loss": 0.2637, + "step": 934 + }, + { + "epoch": 0.008116248990894177, + "grad_norm": 0.080078125, + "learning_rate": 0.001868, + "loss": 0.2314, + "step": 935 + }, + { + "epoch": 0.00812492947109834, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018700000000000001, + "loss": 0.2969, + "step": 936 + }, + { + "epoch": 0.008133609951302505, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018720000000000002, + "loss": 0.3047, + "step": 937 + }, + { + "epoch": 0.00814229043150667, + "grad_norm": 0.111328125, + "learning_rate": 0.0018740000000000002, + "loss": 0.2773, + "step": 938 + }, + { + "epoch": 0.008150970911710836, + "grad_norm": 0.078125, + "learning_rate": 0.0018759999999999998, + "loss": 0.2422, + "step": 939 + }, + { + "epoch": 0.008159651391915, + "grad_norm": 0.10107421875, + "learning_rate": 0.001878, + "loss": 0.3203, + "step": 940 + }, + { + "epoch": 0.008168331872119166, + "grad_norm": 0.083984375, + "learning_rate": 0.00188, + "loss": 0.2773, + "step": 941 + }, + { + "epoch": 0.008177012352323331, + "grad_norm": 0.07958984375, + "learning_rate": 0.001882, + "loss": 0.2637, + "step": 942 + }, + { + "epoch": 0.008185692832527496, + "grad_norm": 0.09228515625, + "learning_rate": 0.001884, + "loss": 0.2383, + "step": 943 + }, + { + "epoch": 0.00819437331273166, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018859999999999999, + "loss": 0.3105, + "step": 944 + }, + { + "epoch": 0.008203053792935825, + "grad_norm": 0.09375, + "learning_rate": 0.001888, + "loss": 0.252, + "step": 945 + }, + { + "epoch": 0.00821173427313999, + "grad_norm": 0.09814453125, + "learning_rate": 0.00189, + "loss": 0.293, + "step": 946 + }, + { + "epoch": 0.008220414753344155, + "grad_norm": 0.107421875, + "learning_rate": 0.001892, + "loss": 0.332, + "step": 947 + }, + { + "epoch": 0.00822909523354832, + "grad_norm": 0.078125, + "learning_rate": 0.0018939999999999999, + "loss": 0.2168, + "step": 948 + }, + { + "epoch": 0.008237775713752485, + "grad_norm": 0.0947265625, + "learning_rate": 0.001896, + "loss": 0.2676, + "step": 949 + }, + { + "epoch": 0.00824645619395665, + "grad_norm": 0.08642578125, + "learning_rate": 0.001898, + "loss": 0.2266, + "step": 950 + }, + { + "epoch": 0.008255136674160815, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019, + "loss": 0.2578, + "step": 951 + }, + { + "epoch": 0.008263817154364979, + "grad_norm": 0.08203125, + "learning_rate": 0.001902, + "loss": 0.2324, + "step": 952 + }, + { + "epoch": 0.008272497634569144, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019039999999999999, + "loss": 0.3008, + "step": 953 + }, + { + "epoch": 0.008281178114773309, + "grad_norm": 0.11474609375, + "learning_rate": 0.001906, + "loss": 0.3066, + "step": 954 + }, + { + "epoch": 0.008289858594977474, + "grad_norm": 0.1142578125, + "learning_rate": 0.001908, + "loss": 0.3516, + "step": 955 + }, + { + "epoch": 0.00829853907518164, + "grad_norm": 0.068359375, + "learning_rate": 0.00191, + "loss": 0.2148, + "step": 956 + }, + { + "epoch": 0.008307219555385804, + "grad_norm": 0.11328125, + "learning_rate": 0.0019119999999999999, + "loss": 0.291, + "step": 957 + }, + { + "epoch": 0.00831590003558997, + "grad_norm": 0.14453125, + "learning_rate": 0.001914, + "loss": 0.5, + "step": 958 + }, + { + "epoch": 0.008324580515794133, + "grad_norm": 0.107421875, + "learning_rate": 0.001916, + "loss": 0.2656, + "step": 959 + }, + { + "epoch": 0.008333260995998298, + "grad_norm": 0.10693359375, + "learning_rate": 0.001918, + "loss": 0.3086, + "step": 960 + }, + { + "epoch": 0.008341941476202463, + "grad_norm": 0.0830078125, + "learning_rate": 0.00192, + "loss": 0.2246, + "step": 961 + }, + { + "epoch": 0.008350621956406628, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019219999999999999, + "loss": 0.2422, + "step": 962 + }, + { + "epoch": 0.008359302436610793, + "grad_norm": 0.09814453125, + "learning_rate": 0.001924, + "loss": 0.2637, + "step": 963 + }, + { + "epoch": 0.008367982916814959, + "grad_norm": 0.10205078125, + "learning_rate": 0.001926, + "loss": 0.2578, + "step": 964 + }, + { + "epoch": 0.008376663397019124, + "grad_norm": 0.11376953125, + "learning_rate": 0.001928, + "loss": 0.2402, + "step": 965 + }, + { + "epoch": 0.008385343877223289, + "grad_norm": 0.11962890625, + "learning_rate": 0.00193, + "loss": 0.2617, + "step": 966 + }, + { + "epoch": 0.008394024357427452, + "grad_norm": 0.091796875, + "learning_rate": 0.001932, + "loss": 0.2148, + "step": 967 + }, + { + "epoch": 0.008402704837631617, + "grad_norm": 0.12109375, + "learning_rate": 0.001934, + "loss": 0.332, + "step": 968 + }, + { + "epoch": 0.008411385317835782, + "grad_norm": 0.10986328125, + "learning_rate": 0.001936, + "loss": 0.2139, + "step": 969 + }, + { + "epoch": 0.008420065798039948, + "grad_norm": 0.1044921875, + "learning_rate": 0.001938, + "loss": 0.2812, + "step": 970 + }, + { + "epoch": 0.008428746278244113, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019399999999999999, + "loss": 0.25, + "step": 971 + }, + { + "epoch": 0.008437426758448278, + "grad_norm": 0.1025390625, + "learning_rate": 0.001942, + "loss": 0.2656, + "step": 972 + }, + { + "epoch": 0.008446107238652443, + "grad_norm": 0.083984375, + "learning_rate": 0.001944, + "loss": 0.2422, + "step": 973 + }, + { + "epoch": 0.008454787718856608, + "grad_norm": 0.0966796875, + "learning_rate": 0.001946, + "loss": 0.2617, + "step": 974 + }, + { + "epoch": 0.008463468199060771, + "grad_norm": 0.1044921875, + "learning_rate": 0.001948, + "loss": 0.3086, + "step": 975 + }, + { + "epoch": 0.008472148679264936, + "grad_norm": 0.0810546875, + "learning_rate": 0.00195, + "loss": 0.2832, + "step": 976 + }, + { + "epoch": 0.008480829159469102, + "grad_norm": 0.095703125, + "learning_rate": 0.001952, + "loss": 0.2617, + "step": 977 + }, + { + "epoch": 0.008489509639673267, + "grad_norm": 0.078125, + "learning_rate": 0.001954, + "loss": 0.2461, + "step": 978 + }, + { + "epoch": 0.008498190119877432, + "grad_norm": 0.080078125, + "learning_rate": 0.001956, + "loss": 0.2168, + "step": 979 + }, + { + "epoch": 0.008506870600081597, + "grad_norm": 0.07958984375, + "learning_rate": 0.001958, + "loss": 0.2266, + "step": 980 + }, + { + "epoch": 0.008515551080285762, + "grad_norm": 0.10107421875, + "learning_rate": 0.00196, + "loss": 0.2373, + "step": 981 + }, + { + "epoch": 0.008524231560489925, + "grad_norm": 0.08984375, + "learning_rate": 0.001962, + "loss": 0.2734, + "step": 982 + }, + { + "epoch": 0.00853291204069409, + "grad_norm": 0.23046875, + "learning_rate": 0.001964, + "loss": 0.3574, + "step": 983 + }, + { + "epoch": 0.008541592520898256, + "grad_norm": 0.08935546875, + "learning_rate": 0.001966, + "loss": 0.2324, + "step": 984 + }, + { + "epoch": 0.00855027300110242, + "grad_norm": 0.1484375, + "learning_rate": 0.001968, + "loss": 0.2676, + "step": 985 + }, + { + "epoch": 0.008558953481306586, + "grad_norm": 0.083984375, + "learning_rate": 0.00197, + "loss": 0.2139, + "step": 986 + }, + { + "epoch": 0.008567633961510751, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019720000000000002, + "loss": 0.2793, + "step": 987 + }, + { + "epoch": 0.008576314441714916, + "grad_norm": 0.12353515625, + "learning_rate": 0.001974, + "loss": 0.3164, + "step": 988 + }, + { + "epoch": 0.008584994921919081, + "grad_norm": 0.07666015625, + "learning_rate": 0.001976, + "loss": 0.2207, + "step": 989 + }, + { + "epoch": 0.008593675402123245, + "grad_norm": 0.07861328125, + "learning_rate": 0.001978, + "loss": 0.2324, + "step": 990 + }, + { + "epoch": 0.00860235588232741, + "grad_norm": 0.08544921875, + "learning_rate": 0.00198, + "loss": 0.2295, + "step": 991 + }, + { + "epoch": 0.008611036362531575, + "grad_norm": 0.1259765625, + "learning_rate": 0.001982, + "loss": 0.3203, + "step": 992 + }, + { + "epoch": 0.00861971684273574, + "grad_norm": 0.123046875, + "learning_rate": 0.001984, + "loss": 0.2656, + "step": 993 + }, + { + "epoch": 0.008628397322939905, + "grad_norm": 0.10302734375, + "learning_rate": 0.001986, + "loss": 0.2637, + "step": 994 + }, + { + "epoch": 0.00863707780314407, + "grad_norm": 0.08935546875, + "learning_rate": 0.001988, + "loss": 0.2383, + "step": 995 + }, + { + "epoch": 0.008645758283348235, + "grad_norm": 0.07666015625, + "learning_rate": 0.00199, + "loss": 0.2178, + "step": 996 + }, + { + "epoch": 0.0086544387635524, + "grad_norm": 0.1083984375, + "learning_rate": 0.001992, + "loss": 0.2832, + "step": 997 + }, + { + "epoch": 0.008663119243756564, + "grad_norm": 0.0908203125, + "learning_rate": 0.001994, + "loss": 0.252, + "step": 998 + }, + { + "epoch": 0.008671799723960729, + "grad_norm": 0.10791015625, + "learning_rate": 0.001996, + "loss": 0.2734, + "step": 999 + }, + { + "epoch": 0.008680480204164894, + "grad_norm": 0.2236328125, + "learning_rate": 0.001998, + "loss": 0.5977, + "step": 1000 + }, + { + "epoch": 0.00868916068436906, + "grad_norm": 0.10546875, + "learning_rate": 0.002, + "loss": 0.2871, + "step": 1001 + }, + { + "epoch": 0.008697841164573224, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019999999995078867, + "loss": 0.2734, + "step": 1002 + }, + { + "epoch": 0.00870652164477739, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019999999980315473, + "loss": 0.2891, + "step": 1003 + }, + { + "epoch": 0.008715202124981555, + "grad_norm": 0.076171875, + "learning_rate": 0.001999999995570981, + "loss": 0.2578, + "step": 1004 + }, + { + "epoch": 0.008723882605185718, + "grad_norm": 0.07861328125, + "learning_rate": 0.001999999992126188, + "loss": 0.2578, + "step": 1005 + }, + { + "epoch": 0.008732563085389883, + "grad_norm": 0.1923828125, + "learning_rate": 0.001999999987697169, + "loss": 0.2949, + "step": 1006 + }, + { + "epoch": 0.008741243565594048, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019999999822839236, + "loss": 0.3164, + "step": 1007 + }, + { + "epoch": 0.008749924045798213, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019999999758864516, + "loss": 0.2676, + "step": 1008 + }, + { + "epoch": 0.008758604526002379, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019999999685047527, + "loss": 0.248, + "step": 1009 + }, + { + "epoch": 0.008767285006206544, + "grad_norm": 0.0849609375, + "learning_rate": 0.001999999960138828, + "loss": 0.2754, + "step": 1010 + }, + { + "epoch": 0.008775965486410709, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019999999507886767, + "loss": 0.3027, + "step": 1011 + }, + { + "epoch": 0.008784645966614874, + "grad_norm": 0.08984375, + "learning_rate": 0.0019999999404542988, + "loss": 0.2285, + "step": 1012 + }, + { + "epoch": 0.008793326446819037, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019999999291356943, + "loss": 0.2617, + "step": 1013 + }, + { + "epoch": 0.008802006927023202, + "grad_norm": 0.111328125, + "learning_rate": 0.001999999916832864, + "loss": 0.2617, + "step": 1014 + }, + { + "epoch": 0.008810687407227367, + "grad_norm": 0.080078125, + "learning_rate": 0.001999999903545807, + "loss": 0.3086, + "step": 1015 + }, + { + "epoch": 0.008819367887431533, + "grad_norm": 0.08984375, + "learning_rate": 0.0019999998892745235, + "loss": 0.2383, + "step": 1016 + }, + { + "epoch": 0.008828048367635698, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019999998740190135, + "loss": 0.3086, + "step": 1017 + }, + { + "epoch": 0.008836728847839863, + "grad_norm": 0.10205078125, + "learning_rate": 0.001999999857779278, + "loss": 0.2539, + "step": 1018 + }, + { + "epoch": 0.008845409328044028, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019999998405553156, + "loss": 0.2002, + "step": 1019 + }, + { + "epoch": 0.008854089808248193, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019999998223471267, + "loss": 0.332, + "step": 1020 + }, + { + "epoch": 0.008862770288452356, + "grad_norm": 0.091796875, + "learning_rate": 0.0019999998031547117, + "loss": 0.2363, + "step": 1021 + }, + { + "epoch": 0.008871450768656522, + "grad_norm": 0.091796875, + "learning_rate": 0.0019999997829780705, + "loss": 0.25, + "step": 1022 + }, + { + "epoch": 0.008880131248860687, + "grad_norm": 0.0849609375, + "learning_rate": 0.001999999761817203, + "loss": 0.2412, + "step": 1023 + }, + { + "epoch": 0.008888811729064852, + "grad_norm": 0.068359375, + "learning_rate": 0.0019999997396721093, + "loss": 0.2246, + "step": 1024 + }, + { + "epoch": 0.008897492209269017, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999999716542789, + "loss": 0.2324, + "step": 1025 + }, + { + "epoch": 0.008906172689473182, + "grad_norm": 0.08837890625, + "learning_rate": 0.001999999692429243, + "loss": 0.3105, + "step": 1026 + }, + { + "epoch": 0.008914853169677347, + "grad_norm": 0.10009765625, + "learning_rate": 0.001999999667331471, + "loss": 0.2422, + "step": 1027 + }, + { + "epoch": 0.008923533649881512, + "grad_norm": 0.09716796875, + "learning_rate": 0.001999999641249473, + "loss": 0.2539, + "step": 1028 + }, + { + "epoch": 0.008932214130085676, + "grad_norm": 0.103515625, + "learning_rate": 0.0019999996141832482, + "loss": 0.2812, + "step": 1029 + }, + { + "epoch": 0.00894089461028984, + "grad_norm": 0.076171875, + "learning_rate": 0.001999999586132798, + "loss": 0.2539, + "step": 1030 + }, + { + "epoch": 0.008949575090494006, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019999995570981212, + "loss": 0.2891, + "step": 1031 + }, + { + "epoch": 0.008958255570698171, + "grad_norm": 0.10693359375, + "learning_rate": 0.001999999527079219, + "loss": 0.25, + "step": 1032 + }, + { + "epoch": 0.008966936050902336, + "grad_norm": 0.06689453125, + "learning_rate": 0.0019999994960760905, + "loss": 0.2266, + "step": 1033 + }, + { + "epoch": 0.008975616531106501, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019999994640887363, + "loss": 0.2754, + "step": 1034 + }, + { + "epoch": 0.008984297011310666, + "grad_norm": 0.076171875, + "learning_rate": 0.001999999431117156, + "loss": 0.209, + "step": 1035 + }, + { + "epoch": 0.00899297749151483, + "grad_norm": 0.091796875, + "learning_rate": 0.00199999939716135, + "loss": 0.249, + "step": 1036 + }, + { + "epoch": 0.009001657971718995, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019999993622213177, + "loss": 0.2422, + "step": 1037 + }, + { + "epoch": 0.00901033845192316, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019999993262970603, + "loss": 0.2363, + "step": 1038 + }, + { + "epoch": 0.009019018932127325, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019999992893885766, + "loss": 0.2871, + "step": 1039 + }, + { + "epoch": 0.00902769941233149, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019999992514958677, + "loss": 0.2734, + "step": 1040 + }, + { + "epoch": 0.009036379892535655, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019999992126189326, + "loss": 0.3496, + "step": 1041 + }, + { + "epoch": 0.00904506037273982, + "grad_norm": 0.09716796875, + "learning_rate": 0.001999999172757772, + "loss": 0.2598, + "step": 1042 + }, + { + "epoch": 0.009053740852943986, + "grad_norm": 0.1337890625, + "learning_rate": 0.001999999131912386, + "loss": 0.2422, + "step": 1043 + }, + { + "epoch": 0.009062421333148149, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019999990900827747, + "loss": 0.2305, + "step": 1044 + }, + { + "epoch": 0.009071101813352314, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019999990472689376, + "loss": 0.2324, + "step": 1045 + }, + { + "epoch": 0.00907978229355648, + "grad_norm": 0.0732421875, + "learning_rate": 0.001999999003470875, + "loss": 0.2129, + "step": 1046 + }, + { + "epoch": 0.009088462773760644, + "grad_norm": 0.103515625, + "learning_rate": 0.0019999989586885875, + "loss": 0.2891, + "step": 1047 + }, + { + "epoch": 0.00909714325396481, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019999989129220745, + "loss": 0.2891, + "step": 1048 + }, + { + "epoch": 0.009105823734168975, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999998866171336, + "loss": 0.2285, + "step": 1049 + }, + { + "epoch": 0.00911450421437314, + "grad_norm": 0.10498046875, + "learning_rate": 0.001999998818436372, + "loss": 0.2402, + "step": 1050 + }, + { + "epoch": 0.009123184694577305, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019999987697171834, + "loss": 0.2363, + "step": 1051 + }, + { + "epoch": 0.009131865174781468, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019999987200137693, + "loss": 0.2012, + "step": 1052 + }, + { + "epoch": 0.009140545654985633, + "grad_norm": 0.14453125, + "learning_rate": 0.00199999866932613, + "loss": 0.252, + "step": 1053 + }, + { + "epoch": 0.009149226135189798, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999998617654266, + "loss": 0.3047, + "step": 1054 + }, + { + "epoch": 0.009157906615393964, + "grad_norm": 0.09912109375, + "learning_rate": 0.001999998564998177, + "loss": 0.2266, + "step": 1055 + }, + { + "epoch": 0.009166587095598129, + "grad_norm": 0.08349609375, + "learning_rate": 0.001999998511357863, + "loss": 0.2539, + "step": 1056 + }, + { + "epoch": 0.009175267575802294, + "grad_norm": 0.11279296875, + "learning_rate": 0.001999998456733324, + "loss": 0.2422, + "step": 1057 + }, + { + "epoch": 0.009183948056006459, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019999984011245604, + "loss": 0.2266, + "step": 1058 + }, + { + "epoch": 0.009192628536210622, + "grad_norm": 0.091796875, + "learning_rate": 0.0019999983445315723, + "loss": 0.3203, + "step": 1059 + }, + { + "epoch": 0.009201309016414787, + "grad_norm": 0.10400390625, + "learning_rate": 0.001999998286954359, + "loss": 0.3066, + "step": 1060 + }, + { + "epoch": 0.009209989496618953, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019999982283929216, + "loss": 0.3086, + "step": 1061 + }, + { + "epoch": 0.009218669976823118, + "grad_norm": 0.068359375, + "learning_rate": 0.0019999981688472593, + "loss": 0.2227, + "step": 1062 + }, + { + "epoch": 0.009227350457027283, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019999981083173727, + "loss": 0.248, + "step": 1063 + }, + { + "epoch": 0.009236030937231448, + "grad_norm": 0.0703125, + "learning_rate": 0.0019999980468032616, + "loss": 0.2451, + "step": 1064 + }, + { + "epoch": 0.009244711417435613, + "grad_norm": 0.1083984375, + "learning_rate": 0.001999997984304926, + "loss": 0.2676, + "step": 1065 + }, + { + "epoch": 0.009253391897639778, + "grad_norm": 0.0859375, + "learning_rate": 0.0019999979208223666, + "loss": 0.2402, + "step": 1066 + }, + { + "epoch": 0.009262072377843942, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019999978563555827, + "loss": 0.2871, + "step": 1067 + }, + { + "epoch": 0.009270752858048107, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019999977909045744, + "loss": 0.2773, + "step": 1068 + }, + { + "epoch": 0.009279433338252272, + "grad_norm": 0.103515625, + "learning_rate": 0.0019999977244693425, + "loss": 0.2285, + "step": 1069 + }, + { + "epoch": 0.009288113818456437, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019999976570498866, + "loss": 0.248, + "step": 1070 + }, + { + "epoch": 0.009296794298660602, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019999975886462063, + "loss": 0.2852, + "step": 1071 + }, + { + "epoch": 0.009305474778864767, + "grad_norm": 0.154296875, + "learning_rate": 0.001999997519258303, + "loss": 0.3711, + "step": 1072 + }, + { + "epoch": 0.009314155259068932, + "grad_norm": 0.07373046875, + "learning_rate": 0.001999997448886175, + "loss": 0.2314, + "step": 1073 + }, + { + "epoch": 0.009322835739273097, + "grad_norm": 0.08837890625, + "learning_rate": 0.001999997377529824, + "loss": 0.3281, + "step": 1074 + }, + { + "epoch": 0.00933151621947726, + "grad_norm": 0.09814453125, + "learning_rate": 0.001999997305189249, + "loss": 0.2773, + "step": 1075 + }, + { + "epoch": 0.009340196699681426, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019999972318644503, + "loss": 0.2266, + "step": 1076 + }, + { + "epoch": 0.009348877179885591, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019999971575554287, + "loss": 0.2715, + "step": 1077 + }, + { + "epoch": 0.009357557660089756, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019999970822621835, + "loss": 0.332, + "step": 1078 + }, + { + "epoch": 0.009366238140293921, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999997005984715, + "loss": 0.2432, + "step": 1079 + }, + { + "epoch": 0.009374918620498086, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019999969287230234, + "loss": 0.2949, + "step": 1080 + }, + { + "epoch": 0.009383599100702252, + "grad_norm": 0.076171875, + "learning_rate": 0.0019999968504771084, + "loss": 0.2461, + "step": 1081 + }, + { + "epoch": 0.009392279580906415, + "grad_norm": 0.087890625, + "learning_rate": 0.0019999967712469708, + "loss": 0.1953, + "step": 1082 + }, + { + "epoch": 0.00940096006111058, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019999966910326096, + "loss": 0.3008, + "step": 1083 + }, + { + "epoch": 0.009409640541314745, + "grad_norm": 0.1025390625, + "learning_rate": 0.001999996609834026, + "loss": 0.2852, + "step": 1084 + }, + { + "epoch": 0.00941832102151891, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019999965276512196, + "loss": 0.2441, + "step": 1085 + }, + { + "epoch": 0.009427001501723075, + "grad_norm": 0.1171875, + "learning_rate": 0.0019999964444841903, + "loss": 0.252, + "step": 1086 + }, + { + "epoch": 0.00943568198192724, + "grad_norm": 0.08935546875, + "learning_rate": 0.001999996360332939, + "loss": 0.2578, + "step": 1087 + }, + { + "epoch": 0.009444362462131406, + "grad_norm": 0.12890625, + "learning_rate": 0.0019999962751974646, + "loss": 0.3242, + "step": 1088 + }, + { + "epoch": 0.00945304294233557, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019999961890777677, + "loss": 0.248, + "step": 1089 + }, + { + "epoch": 0.009461723422539734, + "grad_norm": 0.1064453125, + "learning_rate": 0.001999996101973849, + "loss": 0.2324, + "step": 1090 + }, + { + "epoch": 0.0094704039027439, + "grad_norm": 0.08203125, + "learning_rate": 0.0019999960138857077, + "loss": 0.2402, + "step": 1091 + }, + { + "epoch": 0.009479084382948064, + "grad_norm": 0.1083984375, + "learning_rate": 0.001999995924813345, + "loss": 0.3281, + "step": 1092 + }, + { + "epoch": 0.00948776486315223, + "grad_norm": 0.134765625, + "learning_rate": 0.0019999958347567594, + "loss": 0.2031, + "step": 1093 + }, + { + "epoch": 0.009496445343356395, + "grad_norm": 0.076171875, + "learning_rate": 0.0019999957437159517, + "loss": 0.1943, + "step": 1094 + }, + { + "epoch": 0.00950512582356056, + "grad_norm": 0.0751953125, + "learning_rate": 0.001999995651690923, + "loss": 0.2305, + "step": 1095 + }, + { + "epoch": 0.009513806303764725, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019999955586816726, + "loss": 0.2969, + "step": 1096 + }, + { + "epoch": 0.00952248678396889, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019999954646882, + "loss": 0.2031, + "step": 1097 + }, + { + "epoch": 0.009531167264173053, + "grad_norm": 0.1279296875, + "learning_rate": 0.001999995369710506, + "loss": 0.2988, + "step": 1098 + }, + { + "epoch": 0.009539847744377218, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019999952737485907, + "loss": 0.2891, + "step": 1099 + }, + { + "epoch": 0.009548528224581384, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019999951768024543, + "loss": 0.2773, + "step": 1100 + }, + { + "epoch": 0.009557208704785549, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019999950788720964, + "loss": 0.1904, + "step": 1101 + }, + { + "epoch": 0.009565889184989714, + "grad_norm": 0.0859375, + "learning_rate": 0.0019999949799575176, + "loss": 0.1973, + "step": 1102 + }, + { + "epoch": 0.009574569665193879, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019999948800587175, + "loss": 0.2344, + "step": 1103 + }, + { + "epoch": 0.009583250145398044, + "grad_norm": 0.083984375, + "learning_rate": 0.001999994779175697, + "loss": 0.2412, + "step": 1104 + }, + { + "epoch": 0.00959193062560221, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019999946773084556, + "loss": 0.2578, + "step": 1105 + }, + { + "epoch": 0.009600611105806373, + "grad_norm": 0.1357421875, + "learning_rate": 0.001999994574456993, + "loss": 0.248, + "step": 1106 + }, + { + "epoch": 0.009609291586010538, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019999944706213103, + "loss": 0.209, + "step": 1107 + }, + { + "epoch": 0.009617972066214703, + "grad_norm": 0.06884765625, + "learning_rate": 0.001999994365801407, + "loss": 0.2578, + "step": 1108 + }, + { + "epoch": 0.009626652546418868, + "grad_norm": 0.10009765625, + "learning_rate": 0.001999994259997284, + "loss": 0.3359, + "step": 1109 + }, + { + "epoch": 0.009635333026623033, + "grad_norm": 0.08349609375, + "learning_rate": 0.00199999415320894, + "loss": 0.2891, + "step": 1110 + }, + { + "epoch": 0.009644013506827198, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999994045436376, + "loss": 0.2441, + "step": 1111 + }, + { + "epoch": 0.009652693987031363, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019999939366795927, + "loss": 0.252, + "step": 1112 + }, + { + "epoch": 0.009661374467235527, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999993826938589, + "loss": 0.2305, + "step": 1113 + }, + { + "epoch": 0.009670054947439692, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019999937162133658, + "loss": 0.2402, + "step": 1114 + }, + { + "epoch": 0.009678735427643857, + "grad_norm": 0.083984375, + "learning_rate": 0.0019999936045039224, + "loss": 0.252, + "step": 1115 + }, + { + "epoch": 0.009687415907848022, + "grad_norm": 0.09423828125, + "learning_rate": 0.00199999349181026, + "loss": 0.2451, + "step": 1116 + }, + { + "epoch": 0.009696096388052187, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019999933781323785, + "loss": 0.3516, + "step": 1117 + }, + { + "epoch": 0.009704776868256352, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019999932634702775, + "loss": 0.2422, + "step": 1118 + }, + { + "epoch": 0.009713457348460517, + "grad_norm": 0.126953125, + "learning_rate": 0.0019999931478239573, + "loss": 0.2305, + "step": 1119 + }, + { + "epoch": 0.009722137828664682, + "grad_norm": 0.21875, + "learning_rate": 0.0019999930311934183, + "loss": 0.7617, + "step": 1120 + }, + { + "epoch": 0.009730818308868846, + "grad_norm": 0.0751953125, + "learning_rate": 0.00199999291357866, + "loss": 0.2266, + "step": 1121 + }, + { + "epoch": 0.009739498789073011, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019999927949796836, + "loss": 0.2773, + "step": 1122 + }, + { + "epoch": 0.009748179269277176, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019999926753964882, + "loss": 0.2578, + "step": 1123 + }, + { + "epoch": 0.009756859749481341, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019999925548290745, + "loss": 0.2734, + "step": 1124 + }, + { + "epoch": 0.009765540229685506, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019999924332774425, + "loss": 0.2617, + "step": 1125 + }, + { + "epoch": 0.009774220709889671, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999992310741592, + "loss": 0.2158, + "step": 1126 + }, + { + "epoch": 0.009782901190093837, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999992187221524, + "loss": 0.2188, + "step": 1127 + }, + { + "epoch": 0.009791581670298002, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999992062717238, + "loss": 0.2412, + "step": 1128 + }, + { + "epoch": 0.009800262150502165, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019999919372287334, + "loss": 0.248, + "step": 1129 + }, + { + "epoch": 0.00980894263070633, + "grad_norm": 0.06494140625, + "learning_rate": 0.001999991810756012, + "loss": 0.1885, + "step": 1130 + }, + { + "epoch": 0.009817623110910495, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019999916832990727, + "loss": 0.2188, + "step": 1131 + }, + { + "epoch": 0.00982630359111466, + "grad_norm": 0.07421875, + "learning_rate": 0.001999991554857916, + "loss": 0.2637, + "step": 1132 + }, + { + "epoch": 0.009834984071318826, + "grad_norm": 0.10302734375, + "learning_rate": 0.001999991425432542, + "loss": 0.2695, + "step": 1133 + }, + { + "epoch": 0.00984366455152299, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999991295022951, + "loss": 0.2158, + "step": 1134 + }, + { + "epoch": 0.009852345031727156, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019999911636291432, + "loss": 0.2637, + "step": 1135 + }, + { + "epoch": 0.00986102551193132, + "grad_norm": 0.197265625, + "learning_rate": 0.0019999910312511185, + "loss": 0.2559, + "step": 1136 + }, + { + "epoch": 0.009869705992135484, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019999908978888766, + "loss": 0.2344, + "step": 1137 + }, + { + "epoch": 0.00987838647233965, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019999907635424186, + "loss": 0.2256, + "step": 1138 + }, + { + "epoch": 0.009887066952543815, + "grad_norm": 0.08984375, + "learning_rate": 0.0019999906282117444, + "loss": 0.2363, + "step": 1139 + }, + { + "epoch": 0.00989574743274798, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999990491896854, + "loss": 0.2324, + "step": 1140 + }, + { + "epoch": 0.009904427912952145, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019999903545977475, + "loss": 0.3125, + "step": 1141 + }, + { + "epoch": 0.00991310839315631, + "grad_norm": 0.0673828125, + "learning_rate": 0.001999990216314425, + "loss": 0.2109, + "step": 1142 + }, + { + "epoch": 0.009921788873360475, + "grad_norm": 0.08203125, + "learning_rate": 0.0019999900770468863, + "loss": 0.2109, + "step": 1143 + }, + { + "epoch": 0.009930469353564638, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019999899367951325, + "loss": 0.25, + "step": 1144 + }, + { + "epoch": 0.009939149833768804, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999989795559163, + "loss": 0.2119, + "step": 1145 + }, + { + "epoch": 0.009947830313972969, + "grad_norm": 0.1279296875, + "learning_rate": 0.001999989653338978, + "loss": 0.3125, + "step": 1146 + }, + { + "epoch": 0.009956510794177134, + "grad_norm": 0.087890625, + "learning_rate": 0.0019999895101345784, + "loss": 0.2715, + "step": 1147 + }, + { + "epoch": 0.009965191274381299, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019999893659459634, + "loss": 0.2129, + "step": 1148 + }, + { + "epoch": 0.009973871754585464, + "grad_norm": 0.11181640625, + "learning_rate": 0.001999989220773134, + "loss": 0.2539, + "step": 1149 + }, + { + "epoch": 0.00998255223478963, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019999890746160895, + "loss": 0.2363, + "step": 1150 + }, + { + "epoch": 0.009991232714993794, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019999889274748303, + "loss": 0.3008, + "step": 1151 + }, + { + "epoch": 0.009999913195197958, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019999887793493566, + "loss": 0.2637, + "step": 1152 + }, + { + "epoch": 0.010008593675402123, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019999886302396693, + "loss": 0.2695, + "step": 1153 + }, + { + "epoch": 0.010017274155606288, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019999884801457676, + "loss": 0.2539, + "step": 1154 + }, + { + "epoch": 0.010025954635810453, + "grad_norm": 0.078125, + "learning_rate": 0.0019999883290676523, + "loss": 0.2578, + "step": 1155 + }, + { + "epoch": 0.010034635116014618, + "grad_norm": 0.0859375, + "learning_rate": 0.0019999881770053226, + "loss": 0.2334, + "step": 1156 + }, + { + "epoch": 0.010043315596218783, + "grad_norm": 0.09912109375, + "learning_rate": 0.00199998802395878, + "loss": 0.2949, + "step": 1157 + }, + { + "epoch": 0.010051996076422948, + "grad_norm": 0.0947265625, + "learning_rate": 0.001999987869928024, + "loss": 0.2422, + "step": 1158 + }, + { + "epoch": 0.010060676556627112, + "grad_norm": 0.10205078125, + "learning_rate": 0.001999987714913055, + "loss": 0.2832, + "step": 1159 + }, + { + "epoch": 0.010069357036831277, + "grad_norm": 0.08349609375, + "learning_rate": 0.001999987558913872, + "loss": 0.2422, + "step": 1160 + }, + { + "epoch": 0.010078037517035442, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019999874019304767, + "loss": 0.2432, + "step": 1161 + }, + { + "epoch": 0.010086717997239607, + "grad_norm": 0.130859375, + "learning_rate": 0.001999987243962869, + "loss": 0.2969, + "step": 1162 + }, + { + "epoch": 0.010095398477443772, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019999870850110485, + "loss": 0.2812, + "step": 1163 + }, + { + "epoch": 0.010104078957647937, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019999869250750158, + "loss": 0.2109, + "step": 1164 + }, + { + "epoch": 0.010112759437852102, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019999867641547707, + "loss": 0.2559, + "step": 1165 + }, + { + "epoch": 0.010121439918056268, + "grad_norm": 0.087890625, + "learning_rate": 0.0019999866022503135, + "loss": 0.2363, + "step": 1166 + }, + { + "epoch": 0.010130120398260431, + "grad_norm": 0.123046875, + "learning_rate": 0.0019999864393616448, + "loss": 0.3125, + "step": 1167 + }, + { + "epoch": 0.010138800878464596, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019999862754887642, + "loss": 0.2441, + "step": 1168 + }, + { + "epoch": 0.010147481358668761, + "grad_norm": 0.08984375, + "learning_rate": 0.0019999861106316723, + "loss": 0.2236, + "step": 1169 + }, + { + "epoch": 0.010156161838872926, + "grad_norm": 0.076171875, + "learning_rate": 0.001999985944790369, + "loss": 0.2812, + "step": 1170 + }, + { + "epoch": 0.010164842319077091, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019999857779648546, + "loss": 0.3379, + "step": 1171 + }, + { + "epoch": 0.010173522799281257, + "grad_norm": 0.091796875, + "learning_rate": 0.0019999856101551292, + "loss": 0.2324, + "step": 1172 + }, + { + "epoch": 0.010182203279485422, + "grad_norm": 0.1337890625, + "learning_rate": 0.001999985441361193, + "loss": 0.4375, + "step": 1173 + }, + { + "epoch": 0.010190883759689587, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019999852715830465, + "loss": 0.2617, + "step": 1174 + }, + { + "epoch": 0.01019956423989375, + "grad_norm": 0.078125, + "learning_rate": 0.0019999851008206896, + "loss": 0.1914, + "step": 1175 + }, + { + "epoch": 0.010208244720097915, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019999849290741225, + "loss": 0.252, + "step": 1176 + }, + { + "epoch": 0.01021692520030208, + "grad_norm": 0.083984375, + "learning_rate": 0.0019999847563433454, + "loss": 0.2031, + "step": 1177 + }, + { + "epoch": 0.010225605680506246, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999984582628358, + "loss": 0.1982, + "step": 1178 + }, + { + "epoch": 0.01023428616071041, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999984407929162, + "loss": 0.2598, + "step": 1179 + }, + { + "epoch": 0.010242966640914576, + "grad_norm": 0.1279296875, + "learning_rate": 0.001999984232245756, + "loss": 0.2656, + "step": 1180 + }, + { + "epoch": 0.010251647121118741, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019999840555781404, + "loss": 0.2246, + "step": 1181 + }, + { + "epoch": 0.010260327601322906, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019999838779263166, + "loss": 0.1943, + "step": 1182 + }, + { + "epoch": 0.01026900808152707, + "grad_norm": 0.11767578125, + "learning_rate": 0.001999983699290283, + "loss": 0.2344, + "step": 1183 + }, + { + "epoch": 0.010277688561731235, + "grad_norm": 0.111328125, + "learning_rate": 0.0019999835196700413, + "loss": 0.332, + "step": 1184 + }, + { + "epoch": 0.0102863690419354, + "grad_norm": 0.154296875, + "learning_rate": 0.001999983339065591, + "loss": 0.3359, + "step": 1185 + }, + { + "epoch": 0.010295049522139565, + "grad_norm": 0.095703125, + "learning_rate": 0.001999983157476933, + "loss": 0.332, + "step": 1186 + }, + { + "epoch": 0.01030373000234373, + "grad_norm": 0.09765625, + "learning_rate": 0.0019999829749040663, + "loss": 0.293, + "step": 1187 + }, + { + "epoch": 0.010312410482547895, + "grad_norm": 0.09228515625, + "learning_rate": 0.001999982791346992, + "loss": 0.2217, + "step": 1188 + }, + { + "epoch": 0.01032109096275206, + "grad_norm": 0.1494140625, + "learning_rate": 0.00199998260680571, + "loss": 0.2539, + "step": 1189 + }, + { + "epoch": 0.010329771442956224, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019999824212802203, + "loss": 0.252, + "step": 1190 + }, + { + "epoch": 0.010338451923160389, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019999822347705233, + "loss": 0.2422, + "step": 1191 + }, + { + "epoch": 0.010347132403364554, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019999820472766197, + "loss": 0.2949, + "step": 1192 + }, + { + "epoch": 0.010355812883568719, + "grad_norm": 0.1181640625, + "learning_rate": 0.001999981858798509, + "loss": 0.2852, + "step": 1193 + }, + { + "epoch": 0.010364493363772884, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019999816693361916, + "loss": 0.1719, + "step": 1194 + }, + { + "epoch": 0.010373173843977049, + "grad_norm": 0.08203125, + "learning_rate": 0.0019999814788896676, + "loss": 0.2383, + "step": 1195 + }, + { + "epoch": 0.010381854324181214, + "grad_norm": 0.109375, + "learning_rate": 0.0019999812874589382, + "loss": 0.2754, + "step": 1196 + }, + { + "epoch": 0.01039053480438538, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019999810950440023, + "loss": 0.2734, + "step": 1197 + }, + { + "epoch": 0.010399215284589543, + "grad_norm": 0.10595703125, + "learning_rate": 0.00199998090164486, + "loss": 0.2422, + "step": 1198 + }, + { + "epoch": 0.010407895764793708, + "grad_norm": 0.0859375, + "learning_rate": 0.001999980707261513, + "loss": 0.2539, + "step": 1199 + }, + { + "epoch": 0.010416576244997873, + "grad_norm": 0.08251953125, + "learning_rate": 0.00199998051189396, + "loss": 0.2715, + "step": 1200 + }, + { + "epoch": 0.010425256725202038, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019999803155422023, + "loss": 0.3281, + "step": 1201 + }, + { + "epoch": 0.010433937205406203, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019999801182062392, + "loss": 0.4082, + "step": 1202 + }, + { + "epoch": 0.010442617685610368, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019999799198860716, + "loss": 0.2559, + "step": 1203 + }, + { + "epoch": 0.010451298165814533, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019999797205816996, + "loss": 0.252, + "step": 1204 + }, + { + "epoch": 0.010459978646018699, + "grad_norm": 0.0791015625, + "learning_rate": 0.001999979520293123, + "loss": 0.2266, + "step": 1205 + }, + { + "epoch": 0.010468659126222862, + "grad_norm": 0.10205078125, + "learning_rate": 0.001999979319020343, + "loss": 0.2344, + "step": 1206 + }, + { + "epoch": 0.010477339606427027, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019999791167633583, + "loss": 0.2695, + "step": 1207 + }, + { + "epoch": 0.010486020086631192, + "grad_norm": 0.09814453125, + "learning_rate": 0.00199997891352217, + "loss": 0.2246, + "step": 1208 + }, + { + "epoch": 0.010494700566835357, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019999787092967788, + "loss": 0.2188, + "step": 1209 + }, + { + "epoch": 0.010503381047039522, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019999785040871842, + "loss": 0.2344, + "step": 1210 + }, + { + "epoch": 0.010512061527243688, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019999782978933865, + "loss": 0.2695, + "step": 1211 + }, + { + "epoch": 0.010520742007447853, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019999780907153865, + "loss": 0.2852, + "step": 1212 + }, + { + "epoch": 0.010529422487652016, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019999778825531838, + "loss": 0.2598, + "step": 1213 + }, + { + "epoch": 0.010538102967856181, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019999776734067787, + "loss": 0.2891, + "step": 1214 + }, + { + "epoch": 0.010546783448060346, + "grad_norm": 0.080078125, + "learning_rate": 0.0019999774632761713, + "loss": 0.3008, + "step": 1215 + }, + { + "epoch": 0.010555463928264511, + "grad_norm": 0.07958984375, + "learning_rate": 0.001999977252161362, + "loss": 0.2324, + "step": 1216 + }, + { + "epoch": 0.010564144408468677, + "grad_norm": 0.09033203125, + "learning_rate": 0.001999977040062352, + "loss": 0.2734, + "step": 1217 + }, + { + "epoch": 0.010572824888672842, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019999768269791393, + "loss": 0.2539, + "step": 1218 + }, + { + "epoch": 0.010581505368877007, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019999766129117267, + "loss": 0.1826, + "step": 1219 + }, + { + "epoch": 0.010590185849081172, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019999763978601126, + "loss": 0.2695, + "step": 1220 + }, + { + "epoch": 0.010598866329285335, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019999761818242975, + "loss": 0.2637, + "step": 1221 + }, + { + "epoch": 0.0106075468094895, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019999759648042827, + "loss": 0.2471, + "step": 1222 + }, + { + "epoch": 0.010616227289693666, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019999757468000673, + "loss": 0.3125, + "step": 1223 + }, + { + "epoch": 0.01062490776989783, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019999755278116518, + "loss": 0.2871, + "step": 1224 + }, + { + "epoch": 0.010633588250101996, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999975307839037, + "loss": 0.2949, + "step": 1225 + }, + { + "epoch": 0.010642268730306161, + "grad_norm": 0.07421875, + "learning_rate": 0.0019999750868822225, + "loss": 0.2178, + "step": 1226 + }, + { + "epoch": 0.010650949210510326, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019999748649412088, + "loss": 0.4844, + "step": 1227 + }, + { + "epoch": 0.010659629690714491, + "grad_norm": 0.10205078125, + "learning_rate": 0.001999974642015996, + "loss": 0.2461, + "step": 1228 + }, + { + "epoch": 0.010668310170918655, + "grad_norm": 0.07373046875, + "learning_rate": 0.001999974418106585, + "loss": 0.2158, + "step": 1229 + }, + { + "epoch": 0.01067699065112282, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019999741932129745, + "loss": 0.2246, + "step": 1230 + }, + { + "epoch": 0.010685671131326985, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019999739673351668, + "loss": 0.2393, + "step": 1231 + }, + { + "epoch": 0.01069435161153115, + "grad_norm": 0.078125, + "learning_rate": 0.00199997374047316, + "loss": 0.2773, + "step": 1232 + }, + { + "epoch": 0.010703032091735315, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019999735126269565, + "loss": 0.2539, + "step": 1233 + }, + { + "epoch": 0.01071171257193948, + "grad_norm": 0.083984375, + "learning_rate": 0.0019999732837965548, + "loss": 0.1895, + "step": 1234 + }, + { + "epoch": 0.010720393052143645, + "grad_norm": 0.068359375, + "learning_rate": 0.001999973053981956, + "loss": 0.2236, + "step": 1235 + }, + { + "epoch": 0.010729073532347809, + "grad_norm": 0.10205078125, + "learning_rate": 0.00199997282318316, + "loss": 0.2451, + "step": 1236 + }, + { + "epoch": 0.010737754012551974, + "grad_norm": 0.10498046875, + "learning_rate": 0.001999972591400168, + "loss": 0.2598, + "step": 1237 + }, + { + "epoch": 0.010746434492756139, + "grad_norm": 0.126953125, + "learning_rate": 0.001999972358632979, + "loss": 0.25, + "step": 1238 + }, + { + "epoch": 0.010755114972960304, + "grad_norm": 0.11083984375, + "learning_rate": 0.001999972124881594, + "loss": 0.2441, + "step": 1239 + }, + { + "epoch": 0.010763795453164469, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019999718901460126, + "loss": 0.2988, + "step": 1240 + }, + { + "epoch": 0.010772475933368634, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019999716544262356, + "loss": 0.1875, + "step": 1241 + }, + { + "epoch": 0.0107811564135728, + "grad_norm": 0.06787109375, + "learning_rate": 0.001999971417722263, + "loss": 0.2295, + "step": 1242 + }, + { + "epoch": 0.010789836893776964, + "grad_norm": 0.08056640625, + "learning_rate": 0.001999971180034095, + "loss": 0.248, + "step": 1243 + }, + { + "epoch": 0.010798517373981128, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019999709413617327, + "loss": 0.2734, + "step": 1244 + }, + { + "epoch": 0.010807197854185293, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999970701705175, + "loss": 0.2246, + "step": 1245 + }, + { + "epoch": 0.010815878334389458, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019999704610644234, + "loss": 0.2051, + "step": 1246 + }, + { + "epoch": 0.010824558814593623, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019999702194394777, + "loss": 0.3223, + "step": 1247 + }, + { + "epoch": 0.010833239294797788, + "grad_norm": 0.11181640625, + "learning_rate": 0.001999969976830338, + "loss": 0.2441, + "step": 1248 + }, + { + "epoch": 0.010841919775001953, + "grad_norm": 0.076171875, + "learning_rate": 0.001999969733237004, + "loss": 0.209, + "step": 1249 + }, + { + "epoch": 0.010850600255206119, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999969488659477, + "loss": 0.209, + "step": 1250 + }, + { + "epoch": 0.010859280735410284, + "grad_norm": 0.08544921875, + "learning_rate": 0.001999969243097757, + "loss": 0.2207, + "step": 1251 + }, + { + "epoch": 0.010867961215614447, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019999689965518445, + "loss": 0.2871, + "step": 1252 + }, + { + "epoch": 0.010876641695818612, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019999687490217387, + "loss": 0.2441, + "step": 1253 + }, + { + "epoch": 0.010885322176022777, + "grad_norm": 0.076171875, + "learning_rate": 0.0019999685005074415, + "loss": 0.2539, + "step": 1254 + }, + { + "epoch": 0.010894002656226942, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019999682510089514, + "loss": 0.2324, + "step": 1255 + }, + { + "epoch": 0.010902683136431108, + "grad_norm": 0.0791015625, + "learning_rate": 0.00199996800052627, + "loss": 0.2266, + "step": 1256 + }, + { + "epoch": 0.010911363616635273, + "grad_norm": 0.07763671875, + "learning_rate": 0.001999967749059397, + "loss": 0.2285, + "step": 1257 + }, + { + "epoch": 0.010920044096839438, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019999674966083326, + "loss": 0.2734, + "step": 1258 + }, + { + "epoch": 0.010928724577043603, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019999672431730777, + "loss": 0.2109, + "step": 1259 + }, + { + "epoch": 0.010937405057247766, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019999669887536316, + "loss": 0.2773, + "step": 1260 + }, + { + "epoch": 0.010946085537451931, + "grad_norm": 0.1572265625, + "learning_rate": 0.001999966733349996, + "loss": 0.2598, + "step": 1261 + }, + { + "epoch": 0.010954766017656097, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999966476962169, + "loss": 0.2188, + "step": 1262 + }, + { + "epoch": 0.010963446497860262, + "grad_norm": 0.072265625, + "learning_rate": 0.0019999662195901535, + "loss": 0.2324, + "step": 1263 + }, + { + "epoch": 0.010972126978064427, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019999659612339477, + "loss": 0.3516, + "step": 1264 + }, + { + "epoch": 0.010980807458268592, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999965701893553, + "loss": 0.2305, + "step": 1265 + }, + { + "epoch": 0.010989487938472757, + "grad_norm": 0.087890625, + "learning_rate": 0.001999965441568969, + "loss": 0.2617, + "step": 1266 + }, + { + "epoch": 0.01099816841867692, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019999651802601968, + "loss": 0.2578, + "step": 1267 + }, + { + "epoch": 0.011006848898881086, + "grad_norm": 0.3671875, + "learning_rate": 0.0019999649179672356, + "loss": 0.3828, + "step": 1268 + }, + { + "epoch": 0.01101552937908525, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019999646546900863, + "loss": 0.3105, + "step": 1269 + }, + { + "epoch": 0.011024209859289416, + "grad_norm": 0.111328125, + "learning_rate": 0.0019999643904287496, + "loss": 0.2461, + "step": 1270 + }, + { + "epoch": 0.011032890339493581, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019999641251832252, + "loss": 0.2539, + "step": 1271 + }, + { + "epoch": 0.011041570819697746, + "grad_norm": 0.09814453125, + "learning_rate": 0.001999963858953514, + "loss": 0.2891, + "step": 1272 + }, + { + "epoch": 0.011050251299901911, + "grad_norm": 0.11376953125, + "learning_rate": 0.001999963591739615, + "loss": 0.2793, + "step": 1273 + }, + { + "epoch": 0.011058931780106076, + "grad_norm": 0.111328125, + "learning_rate": 0.00199996332354153, + "loss": 0.293, + "step": 1274 + }, + { + "epoch": 0.01106761226031024, + "grad_norm": 0.1484375, + "learning_rate": 0.001999963054359258, + "loss": 0.25, + "step": 1275 + }, + { + "epoch": 0.011076292740514405, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019999627841928006, + "loss": 0.2266, + "step": 1276 + }, + { + "epoch": 0.01108497322071857, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999962513042157, + "loss": 0.2637, + "step": 1277 + }, + { + "epoch": 0.011093653700922735, + "grad_norm": 0.0732421875, + "learning_rate": 0.001999962240907328, + "loss": 0.2383, + "step": 1278 + }, + { + "epoch": 0.0111023341811269, + "grad_norm": 0.150390625, + "learning_rate": 0.0019999619677883137, + "loss": 0.3633, + "step": 1279 + }, + { + "epoch": 0.011111014661331065, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019999616936851147, + "loss": 0.2578, + "step": 1280 + }, + { + "epoch": 0.01111969514153523, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999961418597731, + "loss": 0.2598, + "step": 1281 + }, + { + "epoch": 0.011128375621739395, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019999611425261634, + "loss": 0.2832, + "step": 1282 + }, + { + "epoch": 0.011137056101943559, + "grad_norm": 0.099609375, + "learning_rate": 0.0019999608654704118, + "loss": 0.2578, + "step": 1283 + }, + { + "epoch": 0.011145736582147724, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019999605874304756, + "loss": 0.3125, + "step": 1284 + }, + { + "epoch": 0.011154417062351889, + "grad_norm": 0.416015625, + "learning_rate": 0.001999960308406357, + "loss": 0.6797, + "step": 1285 + }, + { + "epoch": 0.011163097542556054, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019999600283980546, + "loss": 0.1934, + "step": 1286 + }, + { + "epoch": 0.01117177802276022, + "grad_norm": 0.09130859375, + "learning_rate": 0.00199995974740557, + "loss": 0.291, + "step": 1287 + }, + { + "epoch": 0.011180458502964384, + "grad_norm": 0.09326171875, + "learning_rate": 0.001999959465428903, + "loss": 0.2852, + "step": 1288 + }, + { + "epoch": 0.01118913898316855, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019999591824680536, + "loss": 0.1953, + "step": 1289 + }, + { + "epoch": 0.011197819463372713, + "grad_norm": 0.09423828125, + "learning_rate": 0.001999958898523022, + "loss": 0.2324, + "step": 1290 + }, + { + "epoch": 0.011206499943576878, + "grad_norm": 0.095703125, + "learning_rate": 0.0019999586135938095, + "loss": 0.2422, + "step": 1291 + }, + { + "epoch": 0.011215180423781043, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019999583276804154, + "loss": 0.2656, + "step": 1292 + }, + { + "epoch": 0.011223860903985208, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019999580407828402, + "loss": 0.252, + "step": 1293 + }, + { + "epoch": 0.011232541384189373, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019999577529010854, + "loss": 0.209, + "step": 1294 + }, + { + "epoch": 0.011241221864393539, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019999574640351494, + "loss": 0.2393, + "step": 1295 + }, + { + "epoch": 0.011249902344597704, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019999571741850337, + "loss": 0.2217, + "step": 1296 + }, + { + "epoch": 0.011258582824801869, + "grad_norm": 0.1484375, + "learning_rate": 0.0019999568833507383, + "loss": 0.2539, + "step": 1297 + }, + { + "epoch": 0.011267263305006032, + "grad_norm": 0.09619140625, + "learning_rate": 0.001999956591532264, + "loss": 0.2617, + "step": 1298 + }, + { + "epoch": 0.011275943785210197, + "grad_norm": 0.08203125, + "learning_rate": 0.00199995629872961, + "loss": 0.1953, + "step": 1299 + }, + { + "epoch": 0.011284624265414362, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019999560049427777, + "loss": 0.2246, + "step": 1300 + }, + { + "epoch": 0.011293304745618528, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019999557101717668, + "loss": 0.248, + "step": 1301 + }, + { + "epoch": 0.011301985225822693, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019999554144165778, + "loss": 0.1865, + "step": 1302 + }, + { + "epoch": 0.011310665706026858, + "grad_norm": 0.142578125, + "learning_rate": 0.0019999551176772116, + "loss": 0.3086, + "step": 1303 + }, + { + "epoch": 0.011319346186231023, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019999548199536674, + "loss": 0.25, + "step": 1304 + }, + { + "epoch": 0.011328026666435188, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019999545212459465, + "loss": 0.1973, + "step": 1305 + }, + { + "epoch": 0.011336707146639351, + "grad_norm": 0.09716796875, + "learning_rate": 0.001999954221554049, + "loss": 0.3066, + "step": 1306 + }, + { + "epoch": 0.011345387626843517, + "grad_norm": 0.09521484375, + "learning_rate": 0.001999953920877975, + "loss": 0.2637, + "step": 1307 + }, + { + "epoch": 0.011354068107047682, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019999536192177245, + "loss": 0.2031, + "step": 1308 + }, + { + "epoch": 0.011362748587251847, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019999533165732985, + "loss": 0.3164, + "step": 1309 + }, + { + "epoch": 0.011371429067456012, + "grad_norm": 0.08984375, + "learning_rate": 0.001999953012944697, + "loss": 0.2266, + "step": 1310 + }, + { + "epoch": 0.011380109547660177, + "grad_norm": 0.09033203125, + "learning_rate": 0.001999952708331921, + "loss": 0.252, + "step": 1311 + }, + { + "epoch": 0.011388790027864342, + "grad_norm": 0.078125, + "learning_rate": 0.0019999524027349697, + "loss": 0.2246, + "step": 1312 + }, + { + "epoch": 0.011397470508068505, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019999520961538437, + "loss": 0.2617, + "step": 1313 + }, + { + "epoch": 0.01140615098827267, + "grad_norm": 0.10302734375, + "learning_rate": 0.001999951788588544, + "loss": 0.3242, + "step": 1314 + }, + { + "epoch": 0.011414831468476836, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019999514800390704, + "loss": 0.248, + "step": 1315 + }, + { + "epoch": 0.011423511948681, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019999511705054234, + "loss": 0.2227, + "step": 1316 + }, + { + "epoch": 0.011432192428885166, + "grad_norm": 0.111328125, + "learning_rate": 0.0019999508599876036, + "loss": 0.2812, + "step": 1317 + }, + { + "epoch": 0.011440872909089331, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999950548485611, + "loss": 0.2598, + "step": 1318 + }, + { + "epoch": 0.011449553389293496, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019999502359994456, + "loss": 0.2617, + "step": 1319 + }, + { + "epoch": 0.011458233869497661, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019999499225291087, + "loss": 0.2988, + "step": 1320 + }, + { + "epoch": 0.011466914349701825, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019999496080746, + "loss": 0.2539, + "step": 1321 + }, + { + "epoch": 0.01147559482990599, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019999492926359194, + "loss": 0.2148, + "step": 1322 + }, + { + "epoch": 0.011484275310110155, + "grad_norm": 0.06689453125, + "learning_rate": 0.001999948976213068, + "loss": 0.2422, + "step": 1323 + }, + { + "epoch": 0.01149295579031432, + "grad_norm": 0.07177734375, + "learning_rate": 0.001999948658806046, + "loss": 0.2109, + "step": 1324 + }, + { + "epoch": 0.011501636270518485, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019999483404148535, + "loss": 0.332, + "step": 1325 + }, + { + "epoch": 0.01151031675072265, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019999480210394914, + "loss": 0.2793, + "step": 1326 + }, + { + "epoch": 0.011518997230926815, + "grad_norm": 0.302734375, + "learning_rate": 0.0019999477006799595, + "loss": 0.457, + "step": 1327 + }, + { + "epoch": 0.01152767771113098, + "grad_norm": 0.06201171875, + "learning_rate": 0.0019999473793362583, + "loss": 0.2266, + "step": 1328 + }, + { + "epoch": 0.011536358191335144, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999947057008388, + "loss": 0.3301, + "step": 1329 + }, + { + "epoch": 0.011545038671539309, + "grad_norm": 0.1015625, + "learning_rate": 0.001999946733696349, + "loss": 0.2354, + "step": 1330 + }, + { + "epoch": 0.011553719151743474, + "grad_norm": 0.0888671875, + "learning_rate": 0.001999946409400142, + "loss": 0.2246, + "step": 1331 + }, + { + "epoch": 0.01156239963194764, + "grad_norm": 0.09619140625, + "learning_rate": 0.001999946084119768, + "loss": 0.2383, + "step": 1332 + }, + { + "epoch": 0.011571080112151804, + "grad_norm": 0.11669921875, + "learning_rate": 0.001999945757855225, + "loss": 0.2656, + "step": 1333 + }, + { + "epoch": 0.01157976059235597, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019999454306065157, + "loss": 0.2422, + "step": 1334 + }, + { + "epoch": 0.011588441072560135, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019999451023736394, + "loss": 0.2559, + "step": 1335 + }, + { + "epoch": 0.0115971215527643, + "grad_norm": 0.107421875, + "learning_rate": 0.0019999447731565965, + "loss": 0.2754, + "step": 1336 + }, + { + "epoch": 0.011605802032968463, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019999444429553877, + "loss": 0.2129, + "step": 1337 + }, + { + "epoch": 0.011614482513172628, + "grad_norm": 0.08984375, + "learning_rate": 0.0019999441117700134, + "loss": 0.2324, + "step": 1338 + }, + { + "epoch": 0.011623162993376793, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019999437796004733, + "loss": 0.2217, + "step": 1339 + }, + { + "epoch": 0.011631843473580959, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019999434464467686, + "loss": 0.2598, + "step": 1340 + }, + { + "epoch": 0.011640523953785124, + "grad_norm": 0.1357421875, + "learning_rate": 0.001999943112308899, + "loss": 0.2539, + "step": 1341 + }, + { + "epoch": 0.011649204433989289, + "grad_norm": 0.091796875, + "learning_rate": 0.001999942777186865, + "loss": 0.3066, + "step": 1342 + }, + { + "epoch": 0.011657884914193454, + "grad_norm": 0.1015625, + "learning_rate": 0.0019999424410806674, + "loss": 0.2969, + "step": 1343 + }, + { + "epoch": 0.011666565394397617, + "grad_norm": 0.076171875, + "learning_rate": 0.001999942103990306, + "loss": 0.2051, + "step": 1344 + }, + { + "epoch": 0.011675245874601782, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019999417659157816, + "loss": 0.2539, + "step": 1345 + }, + { + "epoch": 0.011683926354805948, + "grad_norm": 0.09375, + "learning_rate": 0.0019999414268570947, + "loss": 0.2354, + "step": 1346 + }, + { + "epoch": 0.011692606835010113, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999941086814245, + "loss": 0.3164, + "step": 1347 + }, + { + "epoch": 0.011701287315214278, + "grad_norm": 0.1015625, + "learning_rate": 0.001999940745787233, + "loss": 0.2832, + "step": 1348 + }, + { + "epoch": 0.011709967795418443, + "grad_norm": 0.0654296875, + "learning_rate": 0.00199994040377606, + "loss": 0.2266, + "step": 1349 + }, + { + "epoch": 0.011718648275622608, + "grad_norm": 0.07861328125, + "learning_rate": 0.001999940060780725, + "loss": 0.248, + "step": 1350 + }, + { + "epoch": 0.011727328755826773, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019999397168012295, + "loss": 0.2188, + "step": 1351 + }, + { + "epoch": 0.011736009236030936, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019999393718375734, + "loss": 0.2236, + "step": 1352 + }, + { + "epoch": 0.011744689716235102, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999939025889757, + "loss": 0.2539, + "step": 1353 + }, + { + "epoch": 0.011753370196439267, + "grad_norm": 0.08203125, + "learning_rate": 0.0019999386789577808, + "loss": 0.207, + "step": 1354 + }, + { + "epoch": 0.011762050676643432, + "grad_norm": 0.10498046875, + "learning_rate": 0.001999938331041645, + "loss": 0.2363, + "step": 1355 + }, + { + "epoch": 0.011770731156847597, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019999379821413507, + "loss": 0.2461, + "step": 1356 + }, + { + "epoch": 0.011779411637051762, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019999376322568977, + "loss": 0.25, + "step": 1357 + }, + { + "epoch": 0.011788092117255927, + "grad_norm": 0.1015625, + "learning_rate": 0.001999937281388286, + "loss": 0.2598, + "step": 1358 + }, + { + "epoch": 0.011796772597460092, + "grad_norm": 0.08984375, + "learning_rate": 0.0019999369295355166, + "loss": 0.2539, + "step": 1359 + }, + { + "epoch": 0.011805453077664256, + "grad_norm": 0.13671875, + "learning_rate": 0.00199993657669859, + "loss": 0.2285, + "step": 1360 + }, + { + "epoch": 0.01181413355786842, + "grad_norm": 0.08056640625, + "learning_rate": 0.001999936222877506, + "loss": 0.2812, + "step": 1361 + }, + { + "epoch": 0.011822814038072586, + "grad_norm": 0.0966796875, + "learning_rate": 0.001999935868072265, + "loss": 0.2637, + "step": 1362 + }, + { + "epoch": 0.011831494518276751, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999935512282868, + "loss": 0.2695, + "step": 1363 + }, + { + "epoch": 0.011840174998480916, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999935155509315, + "loss": 0.2422, + "step": 1364 + }, + { + "epoch": 0.011848855478685081, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019999347977516066, + "loss": 0.2559, + "step": 1365 + }, + { + "epoch": 0.011857535958889246, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019999344390097423, + "loss": 0.2158, + "step": 1366 + }, + { + "epoch": 0.01186621643909341, + "grad_norm": 0.095703125, + "learning_rate": 0.0019999340792837238, + "loss": 0.3008, + "step": 1367 + }, + { + "epoch": 0.011874896919297575, + "grad_norm": 0.068359375, + "learning_rate": 0.001999933718573551, + "loss": 0.2207, + "step": 1368 + }, + { + "epoch": 0.01188357739950174, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019999333568792243, + "loss": 0.1973, + "step": 1369 + }, + { + "epoch": 0.011892257879705905, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019999329942007437, + "loss": 0.2295, + "step": 1370 + }, + { + "epoch": 0.01190093835991007, + "grad_norm": 0.1142578125, + "learning_rate": 0.00199993263053811, + "loss": 0.2734, + "step": 1371 + }, + { + "epoch": 0.011909618840114235, + "grad_norm": 0.09521484375, + "learning_rate": 0.001999932265891323, + "loss": 0.252, + "step": 1372 + }, + { + "epoch": 0.0119182993203184, + "grad_norm": 0.1328125, + "learning_rate": 0.0019999319002603844, + "loss": 0.2402, + "step": 1373 + }, + { + "epoch": 0.011926979800522566, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019999315336452933, + "loss": 0.2812, + "step": 1374 + }, + { + "epoch": 0.011935660280726729, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999931166046051, + "loss": 0.2432, + "step": 1375 + }, + { + "epoch": 0.011944340760930894, + "grad_norm": 0.10498046875, + "learning_rate": 0.001999930797462657, + "loss": 0.2148, + "step": 1376 + }, + { + "epoch": 0.01195302124113506, + "grad_norm": 0.0859375, + "learning_rate": 0.0019999304278951126, + "loss": 0.2324, + "step": 1377 + }, + { + "epoch": 0.011961701721339224, + "grad_norm": 0.0927734375, + "learning_rate": 0.001999930057343418, + "loss": 0.2402, + "step": 1378 + }, + { + "epoch": 0.01197038220154339, + "grad_norm": 0.0859375, + "learning_rate": 0.0019999296858075727, + "loss": 0.2852, + "step": 1379 + }, + { + "epoch": 0.011979062681747555, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019999293132875783, + "loss": 0.2344, + "step": 1380 + }, + { + "epoch": 0.01198774316195172, + "grad_norm": 0.142578125, + "learning_rate": 0.0019999289397834344, + "loss": 0.3809, + "step": 1381 + }, + { + "epoch": 0.011996423642155885, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999928565295142, + "loss": 0.2051, + "step": 1382 + }, + { + "epoch": 0.012005104122360048, + "grad_norm": 0.0908203125, + "learning_rate": 0.001999928189822701, + "loss": 0.2402, + "step": 1383 + }, + { + "epoch": 0.012013784602564213, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019999278133661126, + "loss": 0.2012, + "step": 1384 + }, + { + "epoch": 0.012022465082768378, + "grad_norm": 0.0888671875, + "learning_rate": 0.001999927435925376, + "loss": 0.2363, + "step": 1385 + }, + { + "epoch": 0.012031145562972544, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019999270575004925, + "loss": 0.25, + "step": 1386 + }, + { + "epoch": 0.012039826043176709, + "grad_norm": 0.34375, + "learning_rate": 0.0019999266780914623, + "loss": 0.3281, + "step": 1387 + }, + { + "epoch": 0.012048506523380874, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999926297698286, + "loss": 0.252, + "step": 1388 + }, + { + "epoch": 0.012057187003585039, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019999259163209636, + "loss": 0.2715, + "step": 1389 + }, + { + "epoch": 0.012065867483789202, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999925533959496, + "loss": 0.2148, + "step": 1390 + }, + { + "epoch": 0.012074547963993367, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019999251506138834, + "loss": 0.2402, + "step": 1391 + }, + { + "epoch": 0.012083228444197533, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019999247662841257, + "loss": 0.2324, + "step": 1392 + }, + { + "epoch": 0.012091908924401698, + "grad_norm": 0.13671875, + "learning_rate": 0.0019999243809702243, + "loss": 0.3066, + "step": 1393 + }, + { + "epoch": 0.012100589404605863, + "grad_norm": 0.09326171875, + "learning_rate": 0.001999923994672179, + "loss": 0.3828, + "step": 1394 + }, + { + "epoch": 0.012109269884810028, + "grad_norm": 0.0673828125, + "learning_rate": 0.00199992360738999, + "loss": 0.2207, + "step": 1395 + }, + { + "epoch": 0.012117950365014193, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019999232191236583, + "loss": 0.252, + "step": 1396 + }, + { + "epoch": 0.012126630845218358, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019999228298731844, + "loss": 0.1924, + "step": 1397 + }, + { + "epoch": 0.012135311325422522, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019999224396385676, + "loss": 0.2207, + "step": 1398 + }, + { + "epoch": 0.012143991805626687, + "grad_norm": 0.0693359375, + "learning_rate": 0.00199992204841981, + "loss": 0.2363, + "step": 1399 + }, + { + "epoch": 0.012152672285830852, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019999216562169107, + "loss": 0.3672, + "step": 1400 + }, + { + "epoch": 0.012161352766035017, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019999212630298704, + "loss": 0.3086, + "step": 1401 + }, + { + "epoch": 0.012170033246239182, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019999208688586904, + "loss": 0.25, + "step": 1402 + }, + { + "epoch": 0.012178713726443347, + "grad_norm": 0.09033203125, + "learning_rate": 0.00199992047370337, + "loss": 0.2695, + "step": 1403 + }, + { + "epoch": 0.012187394206647512, + "grad_norm": 0.08544921875, + "learning_rate": 0.00199992007756391, + "loss": 0.2324, + "step": 1404 + }, + { + "epoch": 0.012196074686851677, + "grad_norm": 0.07177734375, + "learning_rate": 0.001999919680440311, + "loss": 0.2598, + "step": 1405 + }, + { + "epoch": 0.01220475516705584, + "grad_norm": 0.06640625, + "learning_rate": 0.0019999192823325737, + "loss": 0.2637, + "step": 1406 + }, + { + "epoch": 0.012213435647260006, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999918883240698, + "loss": 0.1895, + "step": 1407 + }, + { + "epoch": 0.012222116127464171, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019999184831646847, + "loss": 0.2578, + "step": 1408 + }, + { + "epoch": 0.012230796607668336, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019999180821045335, + "loss": 0.2598, + "step": 1409 + }, + { + "epoch": 0.012239477087872501, + "grad_norm": 0.06494140625, + "learning_rate": 0.001999917680060246, + "loss": 0.1953, + "step": 1410 + }, + { + "epoch": 0.012248157568076666, + "grad_norm": 0.1044921875, + "learning_rate": 0.001999917277031822, + "loss": 0.2676, + "step": 1411 + }, + { + "epoch": 0.012256838048280832, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019999168730192615, + "loss": 0.334, + "step": 1412 + }, + { + "epoch": 0.012265518528484997, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999916468022566, + "loss": 0.2871, + "step": 1413 + }, + { + "epoch": 0.01227419900868916, + "grad_norm": 0.099609375, + "learning_rate": 0.001999916062041735, + "loss": 0.3027, + "step": 1414 + }, + { + "epoch": 0.012282879488893325, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019999156550767694, + "loss": 0.2539, + "step": 1415 + }, + { + "epoch": 0.01229155996909749, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019999152471276696, + "loss": 0.2441, + "step": 1416 + }, + { + "epoch": 0.012300240449301655, + "grad_norm": 0.07421875, + "learning_rate": 0.0019999148381944355, + "loss": 0.2598, + "step": 1417 + }, + { + "epoch": 0.01230892092950582, + "grad_norm": 0.0888671875, + "learning_rate": 0.001999914428277069, + "loss": 0.2734, + "step": 1418 + }, + { + "epoch": 0.012317601409709986, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019999140173755686, + "loss": 0.2578, + "step": 1419 + }, + { + "epoch": 0.01232628188991415, + "grad_norm": 0.080078125, + "learning_rate": 0.0019999136054899367, + "loss": 0.2266, + "step": 1420 + }, + { + "epoch": 0.012334962370118314, + "grad_norm": 0.062255859375, + "learning_rate": 0.0019999131926201723, + "loss": 0.2217, + "step": 1421 + }, + { + "epoch": 0.01234364285032248, + "grad_norm": 0.078125, + "learning_rate": 0.0019999127787662767, + "loss": 0.2832, + "step": 1422 + }, + { + "epoch": 0.012352323330526644, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019999123639282495, + "loss": 0.2715, + "step": 1423 + }, + { + "epoch": 0.01236100381073081, + "grad_norm": 0.091796875, + "learning_rate": 0.001999911948106092, + "loss": 0.2617, + "step": 1424 + }, + { + "epoch": 0.012369684290934975, + "grad_norm": 0.07177734375, + "learning_rate": 0.001999911531299804, + "loss": 0.2109, + "step": 1425 + }, + { + "epoch": 0.01237836477113914, + "grad_norm": 0.056396484375, + "learning_rate": 0.0019999111135093864, + "loss": 0.1924, + "step": 1426 + }, + { + "epoch": 0.012387045251343305, + "grad_norm": 0.06787109375, + "learning_rate": 0.00199991069473484, + "loss": 0.1992, + "step": 1427 + }, + { + "epoch": 0.01239572573154747, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019999102749761644, + "loss": 0.2578, + "step": 1428 + }, + { + "epoch": 0.012404406211751633, + "grad_norm": 0.10009765625, + "learning_rate": 0.00199990985423336, + "loss": 0.25, + "step": 1429 + }, + { + "epoch": 0.012413086691955798, + "grad_norm": 0.08984375, + "learning_rate": 0.0019999094325064285, + "loss": 0.2891, + "step": 1430 + }, + { + "epoch": 0.012421767172159964, + "grad_norm": 0.10986328125, + "learning_rate": 0.001999909009795369, + "loss": 0.2812, + "step": 1431 + }, + { + "epoch": 0.012430447652364129, + "grad_norm": 0.14453125, + "learning_rate": 0.001999908586100183, + "loss": 0.3008, + "step": 1432 + }, + { + "epoch": 0.012439128132568294, + "grad_norm": 0.08349609375, + "learning_rate": 0.00199990816142087, + "loss": 0.2559, + "step": 1433 + }, + { + "epoch": 0.012447808612772459, + "grad_norm": 0.08056640625, + "learning_rate": 0.001999907735757431, + "loss": 0.2217, + "step": 1434 + }, + { + "epoch": 0.012456489092976624, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999907309109867, + "loss": 0.2676, + "step": 1435 + }, + { + "epoch": 0.01246516957318079, + "grad_norm": 0.115234375, + "learning_rate": 0.0019999068814781774, + "loss": 0.3359, + "step": 1436 + }, + { + "epoch": 0.012473850053384953, + "grad_norm": 0.09912109375, + "learning_rate": 0.001999906452862363, + "loss": 0.2041, + "step": 1437 + }, + { + "epoch": 0.012482530533589118, + "grad_norm": 0.1298828125, + "learning_rate": 0.001999906023262425, + "loss": 0.25, + "step": 1438 + }, + { + "epoch": 0.012491211013793283, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019999055926783627, + "loss": 0.2021, + "step": 1439 + }, + { + "epoch": 0.012499891493997448, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019999051611101775, + "loss": 0.2031, + "step": 1440 + }, + { + "epoch": 0.012508571974201613, + "grad_norm": 0.068359375, + "learning_rate": 0.0019999047285578697, + "loss": 0.2324, + "step": 1441 + }, + { + "epoch": 0.012517252454405778, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019999042950214394, + "loss": 0.2539, + "step": 1442 + }, + { + "epoch": 0.012525932934609943, + "grad_norm": 0.0859375, + "learning_rate": 0.001999903860500887, + "loss": 0.2715, + "step": 1443 + }, + { + "epoch": 0.012534613414814107, + "grad_norm": 0.0625, + "learning_rate": 0.0019999034249962135, + "loss": 0.1484, + "step": 1444 + }, + { + "epoch": 0.012543293895018272, + "grad_norm": 0.06494140625, + "learning_rate": 0.001999902988507419, + "loss": 0.2256, + "step": 1445 + }, + { + "epoch": 0.012551974375222437, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019999025510345045, + "loss": 0.2207, + "step": 1446 + }, + { + "epoch": 0.012560654855426602, + "grad_norm": 0.08837890625, + "learning_rate": 0.00199990211257747, + "loss": 0.2422, + "step": 1447 + }, + { + "epoch": 0.012569335335630767, + "grad_norm": 0.058837890625, + "learning_rate": 0.001999901673136316, + "loss": 0.2314, + "step": 1448 + }, + { + "epoch": 0.012578015815834932, + "grad_norm": 0.08203125, + "learning_rate": 0.0019999012327110433, + "loss": 0.208, + "step": 1449 + }, + { + "epoch": 0.012586696296039097, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019999007913016516, + "loss": 0.2715, + "step": 1450 + }, + { + "epoch": 0.012595376776243263, + "grad_norm": 0.28125, + "learning_rate": 0.001999900348908142, + "loss": 0.3086, + "step": 1451 + }, + { + "epoch": 0.012604057256447426, + "grad_norm": 0.0703125, + "learning_rate": 0.001999899905530515, + "loss": 0.2344, + "step": 1452 + }, + { + "epoch": 0.012612737736651591, + "grad_norm": 0.11279296875, + "learning_rate": 0.001999899461168771, + "loss": 0.332, + "step": 1453 + }, + { + "epoch": 0.012621418216855756, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019998990158229106, + "loss": 0.2891, + "step": 1454 + }, + { + "epoch": 0.012630098697059921, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999898569492934, + "loss": 0.2969, + "step": 1455 + }, + { + "epoch": 0.012638779177264086, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999898122178842, + "loss": 0.2051, + "step": 1456 + }, + { + "epoch": 0.012647459657468252, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019998976738806345, + "loss": 0.2539, + "step": 1457 + }, + { + "epoch": 0.012656140137672417, + "grad_norm": 0.09033203125, + "learning_rate": 0.001999897224598313, + "loss": 0.2344, + "step": 1458 + }, + { + "epoch": 0.012664820617876582, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019998967743318774, + "loss": 0.252, + "step": 1459 + }, + { + "epoch": 0.012673501098080745, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999896323081328, + "loss": 0.2324, + "step": 1460 + }, + { + "epoch": 0.01268218157828491, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019998958708466654, + "loss": 0.3281, + "step": 1461 + }, + { + "epoch": 0.012690862058489075, + "grad_norm": 0.08935546875, + "learning_rate": 0.00199989541762789, + "loss": 0.2461, + "step": 1462 + }, + { + "epoch": 0.01269954253869324, + "grad_norm": 0.10205078125, + "learning_rate": 0.001999894963425003, + "loss": 0.3086, + "step": 1463 + }, + { + "epoch": 0.012708223018897406, + "grad_norm": 0.06640625, + "learning_rate": 0.001999894508238004, + "loss": 0.1689, + "step": 1464 + }, + { + "epoch": 0.01271690349910157, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999894052066894, + "loss": 0.2852, + "step": 1465 + }, + { + "epoch": 0.012725583979305736, + "grad_norm": 0.0625, + "learning_rate": 0.0019998935949116737, + "loss": 0.2129, + "step": 1466 + }, + { + "epoch": 0.0127342644595099, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019998931367723426, + "loss": 0.3242, + "step": 1467 + }, + { + "epoch": 0.012742944939714064, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019998926776489028, + "loss": 0.2441, + "step": 1468 + }, + { + "epoch": 0.01275162541991823, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999892217541353, + "loss": 0.2256, + "step": 1469 + }, + { + "epoch": 0.012760305900122395, + "grad_norm": 0.09765625, + "learning_rate": 0.001999891756449695, + "loss": 0.2539, + "step": 1470 + }, + { + "epoch": 0.01276898638032656, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999891294373929, + "loss": 0.2578, + "step": 1471 + }, + { + "epoch": 0.012777666860530725, + "grad_norm": 0.078125, + "learning_rate": 0.001999890831314055, + "loss": 0.2383, + "step": 1472 + }, + { + "epoch": 0.01278634734073489, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019998903672700744, + "loss": 0.1992, + "step": 1473 + }, + { + "epoch": 0.012795027820939055, + "grad_norm": 0.08935546875, + "learning_rate": 0.001999889902241987, + "loss": 0.2656, + "step": 1474 + }, + { + "epoch": 0.012803708301143218, + "grad_norm": 0.095703125, + "learning_rate": 0.0019998894362297935, + "loss": 0.252, + "step": 1475 + }, + { + "epoch": 0.012812388781347384, + "grad_norm": 0.119140625, + "learning_rate": 0.0019998889692334947, + "loss": 0.3223, + "step": 1476 + }, + { + "epoch": 0.012821069261551549, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019998885012530903, + "loss": 0.25, + "step": 1477 + }, + { + "epoch": 0.012829749741755714, + "grad_norm": 0.08984375, + "learning_rate": 0.001999888032288582, + "loss": 0.1875, + "step": 1478 + }, + { + "epoch": 0.012838430221959879, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999887562339969, + "loss": 0.1943, + "step": 1479 + }, + { + "epoch": 0.012847110702164044, + "grad_norm": 0.13671875, + "learning_rate": 0.001999887091407253, + "loss": 0.2422, + "step": 1480 + }, + { + "epoch": 0.01285579118236821, + "grad_norm": 0.0732421875, + "learning_rate": 0.001999886619490434, + "loss": 0.2129, + "step": 1481 + }, + { + "epoch": 0.012864471662572374, + "grad_norm": 0.10888671875, + "learning_rate": 0.001999886146589512, + "loss": 0.2676, + "step": 1482 + }, + { + "epoch": 0.012873152142776538, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019998856727044883, + "loss": 0.2246, + "step": 1483 + }, + { + "epoch": 0.012881832622980703, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019998851978353634, + "loss": 0.2988, + "step": 1484 + }, + { + "epoch": 0.012890513103184868, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019998847219821377, + "loss": 0.2344, + "step": 1485 + }, + { + "epoch": 0.012899193583389033, + "grad_norm": 0.140625, + "learning_rate": 0.001999884245144811, + "loss": 0.2539, + "step": 1486 + }, + { + "epoch": 0.012907874063593198, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019998837673233846, + "loss": 0.2324, + "step": 1487 + }, + { + "epoch": 0.012916554543797363, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999883288517859, + "loss": 0.2461, + "step": 1488 + }, + { + "epoch": 0.012925235024001528, + "grad_norm": 0.08203125, + "learning_rate": 0.0019998828087282343, + "loss": 0.2432, + "step": 1489 + }, + { + "epoch": 0.012933915504205694, + "grad_norm": 0.06689453125, + "learning_rate": 0.0019998823279545113, + "loss": 0.25, + "step": 1490 + }, + { + "epoch": 0.012942595984409857, + "grad_norm": 0.076171875, + "learning_rate": 0.0019998818461966906, + "loss": 0.2148, + "step": 1491 + }, + { + "epoch": 0.012951276464614022, + "grad_norm": 0.1083984375, + "learning_rate": 0.001999881363454773, + "loss": 0.3281, + "step": 1492 + }, + { + "epoch": 0.012959956944818187, + "grad_norm": 0.10986328125, + "learning_rate": 0.001999880879728758, + "loss": 0.2969, + "step": 1493 + }, + { + "epoch": 0.012968637425022352, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019998803950186475, + "loss": 0.2285, + "step": 1494 + }, + { + "epoch": 0.012977317905226517, + "grad_norm": 0.058349609375, + "learning_rate": 0.001999879909324441, + "loss": 0.1855, + "step": 1495 + }, + { + "epoch": 0.012985998385430682, + "grad_norm": 0.5703125, + "learning_rate": 0.001999879422646139, + "loss": 0.5859, + "step": 1496 + }, + { + "epoch": 0.012994678865634848, + "grad_norm": 0.103515625, + "learning_rate": 0.001999878934983743, + "loss": 0.2129, + "step": 1497 + }, + { + "epoch": 0.013003359345839011, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019998784463372524, + "loss": 0.1992, + "step": 1498 + }, + { + "epoch": 0.013012039826043176, + "grad_norm": 0.08203125, + "learning_rate": 0.0019998779567066685, + "loss": 0.1768, + "step": 1499 + }, + { + "epoch": 0.013020720306247341, + "grad_norm": 0.10546875, + "learning_rate": 0.0019998774660919916, + "loss": 0.248, + "step": 1500 + }, + { + "epoch": 0.013029400786451506, + "grad_norm": 0.220703125, + "learning_rate": 0.001999876974493222, + "loss": 0.332, + "step": 1501 + }, + { + "epoch": 0.013038081266655671, + "grad_norm": 0.09716796875, + "learning_rate": 0.001999876481910361, + "loss": 0.2617, + "step": 1502 + }, + { + "epoch": 0.013046761746859837, + "grad_norm": 0.1044921875, + "learning_rate": 0.001999875988343408, + "loss": 0.252, + "step": 1503 + }, + { + "epoch": 0.013055442227064002, + "grad_norm": 0.08349609375, + "learning_rate": 0.001999875493792364, + "loss": 0.2129, + "step": 1504 + }, + { + "epoch": 0.013064122707268167, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019998749982572304, + "loss": 0.2002, + "step": 1505 + }, + { + "epoch": 0.01307280318747233, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019998745017380066, + "loss": 0.2734, + "step": 1506 + }, + { + "epoch": 0.013081483667676495, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019998740042346938, + "loss": 0.2461, + "step": 1507 + }, + { + "epoch": 0.01309016414788066, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019998735057472922, + "loss": 0.2148, + "step": 1508 + }, + { + "epoch": 0.013098844628084826, + "grad_norm": 0.07421875, + "learning_rate": 0.0019998730062758025, + "loss": 0.21, + "step": 1509 + }, + { + "epoch": 0.01310752510828899, + "grad_norm": 0.076171875, + "learning_rate": 0.001999872505820225, + "loss": 0.2461, + "step": 1510 + }, + { + "epoch": 0.013116205588493156, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019998720043805603, + "loss": 0.2539, + "step": 1511 + }, + { + "epoch": 0.013124886068697321, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019998715019568093, + "loss": 0.2656, + "step": 1512 + }, + { + "epoch": 0.013133566548901486, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019998709985489726, + "loss": 0.2236, + "step": 1513 + }, + { + "epoch": 0.01314224702910565, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019998704941570503, + "loss": 0.1973, + "step": 1514 + }, + { + "epoch": 0.013150927509309815, + "grad_norm": 0.1181640625, + "learning_rate": 0.001999869988781043, + "loss": 0.2598, + "step": 1515 + }, + { + "epoch": 0.01315960798951398, + "grad_norm": 0.134765625, + "learning_rate": 0.0019998694824209517, + "loss": 0.2695, + "step": 1516 + }, + { + "epoch": 0.013168288469718145, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019998689750767764, + "loss": 0.2578, + "step": 1517 + }, + { + "epoch": 0.01317696894992231, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999868466748518, + "loss": 0.2363, + "step": 1518 + }, + { + "epoch": 0.013185649430126475, + "grad_norm": 0.09228515625, + "learning_rate": 0.001999867957436177, + "loss": 0.293, + "step": 1519 + }, + { + "epoch": 0.01319432991033064, + "grad_norm": 0.06787109375, + "learning_rate": 0.001999867447139754, + "loss": 0.1914, + "step": 1520 + }, + { + "epoch": 0.013203010390534804, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019998669358592494, + "loss": 0.2451, + "step": 1521 + }, + { + "epoch": 0.013211690870738969, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019998664235946636, + "loss": 0.2334, + "step": 1522 + }, + { + "epoch": 0.013220371350943134, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019998659103459978, + "loss": 0.2617, + "step": 1523 + }, + { + "epoch": 0.013229051831147299, + "grad_norm": 0.06591796875, + "learning_rate": 0.001999865396113252, + "loss": 0.168, + "step": 1524 + }, + { + "epoch": 0.013237732311351464, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019998648808964266, + "loss": 0.248, + "step": 1525 + }, + { + "epoch": 0.01324641279155563, + "grad_norm": 0.0771484375, + "learning_rate": 0.001999864364695523, + "loss": 0.2334, + "step": 1526 + }, + { + "epoch": 0.013255093271759794, + "grad_norm": 0.05859375, + "learning_rate": 0.0019998638475105406, + "loss": 0.1562, + "step": 1527 + }, + { + "epoch": 0.01326377375196396, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019998633293414813, + "loss": 0.2109, + "step": 1528 + }, + { + "epoch": 0.013272454232168123, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019998628101883446, + "loss": 0.1924, + "step": 1529 + }, + { + "epoch": 0.013281134712372288, + "grad_norm": 0.10595703125, + "learning_rate": 0.001999862290051132, + "loss": 0.2324, + "step": 1530 + }, + { + "epoch": 0.013289815192576453, + "grad_norm": 0.0771484375, + "learning_rate": 0.001999861768929843, + "loss": 0.2256, + "step": 1531 + }, + { + "epoch": 0.013298495672780618, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019998612468244787, + "loss": 0.2188, + "step": 1532 + }, + { + "epoch": 0.013307176152984783, + "grad_norm": 0.0693359375, + "learning_rate": 0.00199986072373504, + "loss": 0.1934, + "step": 1533 + }, + { + "epoch": 0.013315856633188948, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019998601996615265, + "loss": 0.2559, + "step": 1534 + }, + { + "epoch": 0.013324537113393113, + "grad_norm": 0.07080078125, + "learning_rate": 0.00199985967460394, + "loss": 0.209, + "step": 1535 + }, + { + "epoch": 0.013333217593597279, + "grad_norm": 0.09375, + "learning_rate": 0.00199985914856228, + "loss": 0.2441, + "step": 1536 + }, + { + "epoch": 0.013341898073801442, + "grad_norm": 0.0849609375, + "learning_rate": 0.001999858621536548, + "loss": 0.2598, + "step": 1537 + }, + { + "epoch": 0.013350578554005607, + "grad_norm": 0.080078125, + "learning_rate": 0.001999858093526744, + "loss": 0.2539, + "step": 1538 + }, + { + "epoch": 0.013359259034209772, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019998575645328687, + "loss": 0.1963, + "step": 1539 + }, + { + "epoch": 0.013367939514413937, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019998570345549226, + "loss": 0.2266, + "step": 1540 + }, + { + "epoch": 0.013376619994618102, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019998565035929065, + "loss": 0.3203, + "step": 1541 + }, + { + "epoch": 0.013385300474822268, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019998559716468208, + "loss": 0.2109, + "step": 1542 + }, + { + "epoch": 0.013393980955026433, + "grad_norm": 0.0888671875, + "learning_rate": 0.001999855438716666, + "loss": 0.2598, + "step": 1543 + }, + { + "epoch": 0.013402661435230596, + "grad_norm": 0.3671875, + "learning_rate": 0.0019998549048024427, + "loss": 0.3496, + "step": 1544 + }, + { + "epoch": 0.013411341915434761, + "grad_norm": 0.08203125, + "learning_rate": 0.001999854369904152, + "loss": 0.2373, + "step": 1545 + }, + { + "epoch": 0.013420022395638926, + "grad_norm": 0.07275390625, + "learning_rate": 0.001999853834021794, + "loss": 0.209, + "step": 1546 + }, + { + "epoch": 0.013428702875843091, + "grad_norm": 0.1298828125, + "learning_rate": 0.001999853297155369, + "loss": 0.2148, + "step": 1547 + }, + { + "epoch": 0.013437383356047257, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019998527593048784, + "loss": 0.1836, + "step": 1548 + }, + { + "epoch": 0.013446063836251422, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019998522204703224, + "loss": 0.3203, + "step": 1549 + }, + { + "epoch": 0.013454744316455587, + "grad_norm": 0.07861328125, + "learning_rate": 0.001999851680651701, + "loss": 0.1914, + "step": 1550 + }, + { + "epoch": 0.013463424796659752, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019998511398490156, + "loss": 0.2812, + "step": 1551 + }, + { + "epoch": 0.013472105276863915, + "grad_norm": 0.244140625, + "learning_rate": 0.0019998505980622664, + "loss": 0.3457, + "step": 1552 + }, + { + "epoch": 0.01348078575706808, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019998500552914546, + "loss": 0.2598, + "step": 1553 + }, + { + "epoch": 0.013489466237272246, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019998495115365797, + "loss": 0.2246, + "step": 1554 + }, + { + "epoch": 0.01349814671747641, + "grad_norm": 0.12255859375, + "learning_rate": 0.001999848966797643, + "loss": 0.293, + "step": 1555 + }, + { + "epoch": 0.013506827197680576, + "grad_norm": 0.080078125, + "learning_rate": 0.0019998484210746455, + "loss": 0.2305, + "step": 1556 + }, + { + "epoch": 0.013515507677884741, + "grad_norm": 0.091796875, + "learning_rate": 0.0019998478743675865, + "loss": 0.2129, + "step": 1557 + }, + { + "epoch": 0.013524188158088906, + "grad_norm": 0.21875, + "learning_rate": 0.001999847326676468, + "loss": 0.3477, + "step": 1558 + }, + { + "epoch": 0.013532868638293071, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019998467780012897, + "loss": 0.1992, + "step": 1559 + }, + { + "epoch": 0.013541549118497235, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019998462283420527, + "loss": 0.2812, + "step": 1560 + }, + { + "epoch": 0.0135502295987014, + "grad_norm": 0.0732421875, + "learning_rate": 0.001999845677698757, + "loss": 0.2383, + "step": 1561 + }, + { + "epoch": 0.013558910078905565, + "grad_norm": 0.271484375, + "learning_rate": 0.001999845126071404, + "loss": 0.543, + "step": 1562 + }, + { + "epoch": 0.01356759055910973, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019998445734599937, + "loss": 0.2324, + "step": 1563 + }, + { + "epoch": 0.013576271039313895, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999844019864527, + "loss": 0.2422, + "step": 1564 + }, + { + "epoch": 0.01358495151951806, + "grad_norm": 0.1455078125, + "learning_rate": 0.001999843465285004, + "loss": 0.2891, + "step": 1565 + }, + { + "epoch": 0.013593631999722225, + "grad_norm": 0.0673828125, + "learning_rate": 0.001999842909721426, + "loss": 0.209, + "step": 1566 + }, + { + "epoch": 0.01360231247992639, + "grad_norm": 0.109375, + "learning_rate": 0.0019998423531737935, + "loss": 0.3027, + "step": 1567 + }, + { + "epoch": 0.013610992960130554, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999841795642107, + "loss": 0.2891, + "step": 1568 + }, + { + "epoch": 0.013619673440334719, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019998412371263668, + "loss": 0.2422, + "step": 1569 + }, + { + "epoch": 0.013628353920538884, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019998406776265735, + "loss": 0.3398, + "step": 1570 + }, + { + "epoch": 0.013637034400743049, + "grad_norm": 0.0966796875, + "learning_rate": 0.001999840117142728, + "loss": 0.2324, + "step": 1571 + }, + { + "epoch": 0.013645714880947214, + "grad_norm": 0.169921875, + "learning_rate": 0.001999839555674831, + "loss": 0.3418, + "step": 1572 + }, + { + "epoch": 0.01365439536115138, + "grad_norm": 0.09375, + "learning_rate": 0.001999838993222883, + "loss": 0.3398, + "step": 1573 + }, + { + "epoch": 0.013663075841355544, + "grad_norm": 0.072265625, + "learning_rate": 0.0019998384297868848, + "loss": 0.2129, + "step": 1574 + }, + { + "epoch": 0.013671756321559708, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019998378653668368, + "loss": 0.2637, + "step": 1575 + }, + { + "epoch": 0.013680436801763873, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019998372999627395, + "loss": 0.2422, + "step": 1576 + }, + { + "epoch": 0.013689117281968038, + "grad_norm": 0.0859375, + "learning_rate": 0.001999836733574594, + "loss": 0.2285, + "step": 1577 + }, + { + "epoch": 0.013697797762172203, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019998361662023996, + "loss": 0.2246, + "step": 1578 + }, + { + "epoch": 0.013706478242376368, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019998355978461586, + "loss": 0.2578, + "step": 1579 + }, + { + "epoch": 0.013715158722580533, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019998350285058706, + "loss": 0.2188, + "step": 1580 + }, + { + "epoch": 0.013723839202784699, + "grad_norm": 0.0927734375, + "learning_rate": 0.001999834458181537, + "loss": 0.2598, + "step": 1581 + }, + { + "epoch": 0.013732519682988864, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019998338868731573, + "loss": 0.3125, + "step": 1582 + }, + { + "epoch": 0.013741200163193027, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019998333145807333, + "loss": 0.3594, + "step": 1583 + }, + { + "epoch": 0.013749880643397192, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019998327413042654, + "loss": 0.2178, + "step": 1584 + }, + { + "epoch": 0.013758561123601357, + "grad_norm": 0.07373046875, + "learning_rate": 0.001999832167043753, + "loss": 0.1699, + "step": 1585 + }, + { + "epoch": 0.013767241603805522, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019998315917991983, + "loss": 0.2402, + "step": 1586 + }, + { + "epoch": 0.013775922084009688, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019998310155706013, + "loss": 0.1895, + "step": 1587 + }, + { + "epoch": 0.013784602564213853, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019998304383579625, + "loss": 0.2227, + "step": 1588 + }, + { + "epoch": 0.013793283044418018, + "grad_norm": 0.06640625, + "learning_rate": 0.0019998298601612828, + "loss": 0.2012, + "step": 1589 + }, + { + "epoch": 0.013801963524622183, + "grad_norm": 0.078125, + "learning_rate": 0.001999829280980562, + "loss": 0.2109, + "step": 1590 + }, + { + "epoch": 0.013810644004826346, + "grad_norm": 0.072265625, + "learning_rate": 0.001999828700815802, + "loss": 0.2383, + "step": 1591 + }, + { + "epoch": 0.013819324485030511, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999828119667003, + "loss": 0.2539, + "step": 1592 + }, + { + "epoch": 0.013828004965234677, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999827537534165, + "loss": 0.2617, + "step": 1593 + }, + { + "epoch": 0.013836685445438842, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019998269544172897, + "loss": 0.2422, + "step": 1594 + }, + { + "epoch": 0.013845365925643007, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019998263703163766, + "loss": 0.2656, + "step": 1595 + }, + { + "epoch": 0.013854046405847172, + "grad_norm": 0.061767578125, + "learning_rate": 0.0019998257852314274, + "loss": 0.2168, + "step": 1596 + }, + { + "epoch": 0.013862726886051337, + "grad_norm": 0.076171875, + "learning_rate": 0.001999825199162442, + "loss": 0.2598, + "step": 1597 + }, + { + "epoch": 0.0138714073662555, + "grad_norm": 0.1044921875, + "learning_rate": 0.001999824612109421, + "loss": 0.248, + "step": 1598 + }, + { + "epoch": 0.013880087846459666, + "grad_norm": 0.0927734375, + "learning_rate": 0.001999824024072366, + "loss": 0.2871, + "step": 1599 + }, + { + "epoch": 0.01388876832666383, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019998234350512762, + "loss": 0.2871, + "step": 1600 + }, + { + "epoch": 0.013897448806867996, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019998228450461533, + "loss": 0.25, + "step": 1601 + }, + { + "epoch": 0.013906129287072161, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019998222540569977, + "loss": 0.2168, + "step": 1602 + }, + { + "epoch": 0.013914809767276326, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019998216620838102, + "loss": 0.2793, + "step": 1603 + }, + { + "epoch": 0.013923490247480491, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019998210691265913, + "loss": 0.2207, + "step": 1604 + }, + { + "epoch": 0.013932170727684656, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019998204751853414, + "loss": 0.291, + "step": 1605 + }, + { + "epoch": 0.01394085120788882, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019998198802600614, + "loss": 0.2441, + "step": 1606 + }, + { + "epoch": 0.013949531688092985, + "grad_norm": 0.07861328125, + "learning_rate": 0.001999819284350752, + "loss": 0.2559, + "step": 1607 + }, + { + "epoch": 0.01395821216829715, + "grad_norm": 0.08935546875, + "learning_rate": 0.001999818687457413, + "loss": 0.252, + "step": 1608 + }, + { + "epoch": 0.013966892648501315, + "grad_norm": 0.06640625, + "learning_rate": 0.0019998180895800465, + "loss": 0.1865, + "step": 1609 + }, + { + "epoch": 0.01397557312870548, + "grad_norm": 0.115234375, + "learning_rate": 0.0019998174907186524, + "loss": 0.2539, + "step": 1610 + }, + { + "epoch": 0.013984253608909645, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019998168908732317, + "loss": 0.2324, + "step": 1611 + }, + { + "epoch": 0.01399293408911381, + "grad_norm": 0.08984375, + "learning_rate": 0.0019998162900437843, + "loss": 0.2402, + "step": 1612 + }, + { + "epoch": 0.014001614569317975, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019998156882303116, + "loss": 0.2812, + "step": 1613 + }, + { + "epoch": 0.014010295049522139, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019998150854328134, + "loss": 0.2246, + "step": 1614 + }, + { + "epoch": 0.014018975529726304, + "grad_norm": 0.09375, + "learning_rate": 0.0019998144816512917, + "loss": 0.2637, + "step": 1615 + }, + { + "epoch": 0.014027656009930469, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019998138768857463, + "loss": 0.2656, + "step": 1616 + }, + { + "epoch": 0.014036336490134634, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019998132711361778, + "loss": 0.25, + "step": 1617 + }, + { + "epoch": 0.0140450169703388, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019998126644025865, + "loss": 0.2578, + "step": 1618 + }, + { + "epoch": 0.014053697450542964, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999812056684974, + "loss": 0.2812, + "step": 1619 + }, + { + "epoch": 0.01406237793074713, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019998114479833407, + "loss": 0.1621, + "step": 1620 + }, + { + "epoch": 0.014071058410951293, + "grad_norm": 0.08984375, + "learning_rate": 0.0019998108382976868, + "loss": 0.2383, + "step": 1621 + }, + { + "epoch": 0.014079738891155458, + "grad_norm": 0.06591796875, + "learning_rate": 0.001999810227628014, + "loss": 0.2246, + "step": 1622 + }, + { + "epoch": 0.014088419371359623, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019998096159743214, + "loss": 0.2314, + "step": 1623 + }, + { + "epoch": 0.014097099851563788, + "grad_norm": 0.08984375, + "learning_rate": 0.001999809003336611, + "loss": 0.2637, + "step": 1624 + }, + { + "epoch": 0.014105780331767953, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019998083897148832, + "loss": 0.2246, + "step": 1625 + }, + { + "epoch": 0.014114460811972119, + "grad_norm": 0.08984375, + "learning_rate": 0.001999807775109138, + "loss": 0.2324, + "step": 1626 + }, + { + "epoch": 0.014123141292176284, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019998071595193766, + "loss": 0.2129, + "step": 1627 + }, + { + "epoch": 0.014131821772380449, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019998065429455997, + "loss": 0.2207, + "step": 1628 + }, + { + "epoch": 0.014140502252584612, + "grad_norm": 0.07373046875, + "learning_rate": 0.001999805925387808, + "loss": 0.2344, + "step": 1629 + }, + { + "epoch": 0.014149182732788777, + "grad_norm": 0.099609375, + "learning_rate": 0.0019998053068460016, + "loss": 0.25, + "step": 1630 + }, + { + "epoch": 0.014157863212992942, + "grad_norm": 0.1171875, + "learning_rate": 0.001999804687320182, + "loss": 0.2207, + "step": 1631 + }, + { + "epoch": 0.014166543693197108, + "grad_norm": 0.078125, + "learning_rate": 0.0019998040668103498, + "loss": 0.2695, + "step": 1632 + }, + { + "epoch": 0.014175224173401273, + "grad_norm": 0.14453125, + "learning_rate": 0.001999803445316505, + "loss": 0.4668, + "step": 1633 + }, + { + "epoch": 0.014183904653605438, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019998028228386485, + "loss": 0.2412, + "step": 1634 + }, + { + "epoch": 0.014192585133809603, + "grad_norm": 0.083984375, + "learning_rate": 0.0019998021993767818, + "loss": 0.2314, + "step": 1635 + }, + { + "epoch": 0.014201265614013768, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019998015749309044, + "loss": 0.2754, + "step": 1636 + }, + { + "epoch": 0.014209946094217931, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019998009495010177, + "loss": 0.293, + "step": 1637 + }, + { + "epoch": 0.014218626574422097, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999800323087122, + "loss": 0.3281, + "step": 1638 + }, + { + "epoch": 0.014227307054626262, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019997996956892185, + "loss": 0.2246, + "step": 1639 + }, + { + "epoch": 0.014235987534830427, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019997990673073078, + "loss": 0.2578, + "step": 1640 + }, + { + "epoch": 0.014244668015034592, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019997984379413894, + "loss": 0.1748, + "step": 1641 + }, + { + "epoch": 0.014253348495238757, + "grad_norm": 0.17578125, + "learning_rate": 0.0019997978075914657, + "loss": 0.875, + "step": 1642 + }, + { + "epoch": 0.014262028975442922, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019997971762575366, + "loss": 0.2168, + "step": 1643 + }, + { + "epoch": 0.014270709455647087, + "grad_norm": 0.111328125, + "learning_rate": 0.0019997965439396024, + "loss": 0.2344, + "step": 1644 + }, + { + "epoch": 0.01427938993585125, + "grad_norm": 0.08837890625, + "learning_rate": 0.001999795910637665, + "loss": 0.2314, + "step": 1645 + }, + { + "epoch": 0.014288070416055416, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019997952763517236, + "loss": 0.3125, + "step": 1646 + }, + { + "epoch": 0.014296750896259581, + "grad_norm": 0.09765625, + "learning_rate": 0.00199979464108178, + "loss": 0.2949, + "step": 1647 + }, + { + "epoch": 0.014305431376463746, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019997940048278344, + "loss": 0.2158, + "step": 1648 + }, + { + "epoch": 0.014314111856667911, + "grad_norm": 0.064453125, + "learning_rate": 0.001999793367589887, + "loss": 0.1777, + "step": 1649 + }, + { + "epoch": 0.014322792336872076, + "grad_norm": 0.08251953125, + "learning_rate": 0.00199979272936794, + "loss": 0.2109, + "step": 1650 + }, + { + "epoch": 0.014331472817076241, + "grad_norm": 0.09765625, + "learning_rate": 0.0019997920901619927, + "loss": 0.2637, + "step": 1651 + }, + { + "epoch": 0.014340153297280405, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019997914499720465, + "loss": 0.2773, + "step": 1652 + }, + { + "epoch": 0.01434883377748457, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999790808798102, + "loss": 0.1914, + "step": 1653 + }, + { + "epoch": 0.014357514257688735, + "grad_norm": 0.126953125, + "learning_rate": 0.001999790166640159, + "loss": 0.2402, + "step": 1654 + }, + { + "epoch": 0.0143661947378929, + "grad_norm": 0.091796875, + "learning_rate": 0.0019997895234982197, + "loss": 0.2461, + "step": 1655 + }, + { + "epoch": 0.014374875218097065, + "grad_norm": 0.0791015625, + "learning_rate": 0.001999788879372284, + "loss": 0.248, + "step": 1656 + }, + { + "epoch": 0.01438355569830123, + "grad_norm": 0.11279296875, + "learning_rate": 0.001999788234262353, + "loss": 0.2812, + "step": 1657 + }, + { + "epoch": 0.014392236178505395, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019997875881684266, + "loss": 0.2793, + "step": 1658 + }, + { + "epoch": 0.01440091665870956, + "grad_norm": 0.091796875, + "learning_rate": 0.0019997869410905062, + "loss": 0.2295, + "step": 1659 + }, + { + "epoch": 0.014409597138913724, + "grad_norm": 0.091796875, + "learning_rate": 0.001999786293028592, + "loss": 0.2402, + "step": 1660 + }, + { + "epoch": 0.014418277619117889, + "grad_norm": 0.059814453125, + "learning_rate": 0.0019997856439826858, + "loss": 0.1855, + "step": 1661 + }, + { + "epoch": 0.014426958099322054, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999784993952787, + "loss": 0.2949, + "step": 1662 + }, + { + "epoch": 0.01443563857952622, + "grad_norm": 0.09814453125, + "learning_rate": 0.001999784342938897, + "loss": 0.209, + "step": 1663 + }, + { + "epoch": 0.014444319059730384, + "grad_norm": 0.0615234375, + "learning_rate": 0.0019997836909410164, + "loss": 0.25, + "step": 1664 + }, + { + "epoch": 0.01445299953993455, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019997830379591456, + "loss": 0.2656, + "step": 1665 + }, + { + "epoch": 0.014461680020138715, + "grad_norm": 0.072265625, + "learning_rate": 0.001999782383993286, + "loss": 0.2041, + "step": 1666 + }, + { + "epoch": 0.01447036050034288, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019997817290434376, + "loss": 0.2422, + "step": 1667 + }, + { + "epoch": 0.014479040980547043, + "grad_norm": 0.0615234375, + "learning_rate": 0.0019997810731096012, + "loss": 0.1904, + "step": 1668 + }, + { + "epoch": 0.014487721460751208, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999780416191778, + "loss": 0.2539, + "step": 1669 + }, + { + "epoch": 0.014496401940955373, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019997797582899687, + "loss": 0.2754, + "step": 1670 + }, + { + "epoch": 0.014505082421159539, + "grad_norm": 0.08203125, + "learning_rate": 0.0019997790994041734, + "loss": 0.2207, + "step": 1671 + }, + { + "epoch": 0.014513762901363704, + "grad_norm": 0.0712890625, + "learning_rate": 0.001999778439534393, + "loss": 0.2285, + "step": 1672 + }, + { + "epoch": 0.014522443381567869, + "grad_norm": 0.1025390625, + "learning_rate": 0.001999777778680629, + "loss": 0.3398, + "step": 1673 + }, + { + "epoch": 0.014531123861772034, + "grad_norm": 0.08837890625, + "learning_rate": 0.001999777116842881, + "loss": 0.1904, + "step": 1674 + }, + { + "epoch": 0.014539804341976197, + "grad_norm": 0.10986328125, + "learning_rate": 0.001999776454021151, + "loss": 0.2734, + "step": 1675 + }, + { + "epoch": 0.014548484822180362, + "grad_norm": 0.091796875, + "learning_rate": 0.0019997757902154387, + "loss": 0.2617, + "step": 1676 + }, + { + "epoch": 0.014557165302384528, + "grad_norm": 0.09375, + "learning_rate": 0.001999775125425745, + "loss": 0.3008, + "step": 1677 + }, + { + "epoch": 0.014565845782588693, + "grad_norm": 0.57421875, + "learning_rate": 0.0019997744596520705, + "loss": 0.6094, + "step": 1678 + }, + { + "epoch": 0.014574526262792858, + "grad_norm": 0.06201171875, + "learning_rate": 0.001999773792894416, + "loss": 0.207, + "step": 1679 + }, + { + "epoch": 0.014583206742997023, + "grad_norm": 0.091796875, + "learning_rate": 0.001999773125152783, + "loss": 0.2314, + "step": 1680 + }, + { + "epoch": 0.014591887223201188, + "grad_norm": 0.076171875, + "learning_rate": 0.001999772456427171, + "loss": 0.1758, + "step": 1681 + }, + { + "epoch": 0.014600567703405353, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019997717867175826, + "loss": 0.2109, + "step": 1682 + }, + { + "epoch": 0.014609248183609517, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019997711160240164, + "loss": 0.2617, + "step": 1683 + }, + { + "epoch": 0.014617928663813682, + "grad_norm": 0.0673828125, + "learning_rate": 0.001999770444346474, + "loss": 0.2305, + "step": 1684 + }, + { + "epoch": 0.014626609144017847, + "grad_norm": 0.0732421875, + "learning_rate": 0.001999769771684956, + "loss": 0.1992, + "step": 1685 + }, + { + "epoch": 0.014635289624222012, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019997690980394635, + "loss": 0.2969, + "step": 1686 + }, + { + "epoch": 0.014643970104426177, + "grad_norm": 0.09326171875, + "learning_rate": 0.001999768423409997, + "loss": 0.2988, + "step": 1687 + }, + { + "epoch": 0.014652650584630342, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019997677477965577, + "loss": 0.1943, + "step": 1688 + }, + { + "epoch": 0.014661331064834507, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019997670711991455, + "loss": 0.2129, + "step": 1689 + }, + { + "epoch": 0.014670011545038672, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019997663936177613, + "loss": 0.1992, + "step": 1690 + }, + { + "epoch": 0.014678692025242836, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019997657150524067, + "loss": 0.2539, + "step": 1691 + }, + { + "epoch": 0.014687372505447, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019997650355030815, + "loss": 0.2246, + "step": 1692 + }, + { + "epoch": 0.014696052985651166, + "grad_norm": 0.06787109375, + "learning_rate": 0.001999764354969787, + "loss": 0.2275, + "step": 1693 + }, + { + "epoch": 0.014704733465855331, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019997636734525237, + "loss": 0.2148, + "step": 1694 + }, + { + "epoch": 0.014713413946059496, + "grad_norm": 0.12060546875, + "learning_rate": 0.001999762990951292, + "loss": 0.3242, + "step": 1695 + }, + { + "epoch": 0.014722094426263661, + "grad_norm": 0.064453125, + "learning_rate": 0.001999762307466093, + "loss": 0.2383, + "step": 1696 + }, + { + "epoch": 0.014730774906467826, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019997616229969276, + "loss": 0.248, + "step": 1697 + }, + { + "epoch": 0.01473945538667199, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019997609375437967, + "loss": 0.2988, + "step": 1698 + }, + { + "epoch": 0.014748135866876155, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019997602511067003, + "loss": 0.2617, + "step": 1699 + }, + { + "epoch": 0.01475681634708032, + "grad_norm": 0.078125, + "learning_rate": 0.0019997595636856397, + "loss": 0.2891, + "step": 1700 + }, + { + "epoch": 0.014765496827284485, + "grad_norm": 0.07763671875, + "learning_rate": 0.001999758875280616, + "loss": 0.2002, + "step": 1701 + }, + { + "epoch": 0.01477417730748865, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019997581858916293, + "loss": 0.2656, + "step": 1702 + }, + { + "epoch": 0.014782857787692815, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019997574955186804, + "loss": 0.3457, + "step": 1703 + }, + { + "epoch": 0.01479153826789698, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019997568041617703, + "loss": 0.2344, + "step": 1704 + }, + { + "epoch": 0.014800218748101146, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019997561118208994, + "loss": 0.1973, + "step": 1705 + }, + { + "epoch": 0.014808899228305309, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999755418496069, + "loss": 0.2695, + "step": 1706 + }, + { + "epoch": 0.014817579708509474, + "grad_norm": 0.07666015625, + "learning_rate": 0.00199975472418728, + "loss": 0.2354, + "step": 1707 + }, + { + "epoch": 0.01482626018871364, + "grad_norm": 0.10986328125, + "learning_rate": 0.001999754028894532, + "loss": 0.1934, + "step": 1708 + }, + { + "epoch": 0.014834940668917804, + "grad_norm": 0.07080078125, + "learning_rate": 0.001999753332617827, + "loss": 0.208, + "step": 1709 + }, + { + "epoch": 0.01484362114912197, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019997526353571654, + "loss": 0.2617, + "step": 1710 + }, + { + "epoch": 0.014852301629326135, + "grad_norm": 0.0703125, + "learning_rate": 0.0019997519371125474, + "loss": 0.2393, + "step": 1711 + }, + { + "epoch": 0.0148609821095303, + "grad_norm": 0.10888671875, + "learning_rate": 0.001999751237883975, + "loss": 0.2715, + "step": 1712 + }, + { + "epoch": 0.014869662589734465, + "grad_norm": 0.1455078125, + "learning_rate": 0.001999750537671447, + "loss": 0.2852, + "step": 1713 + }, + { + "epoch": 0.014878343069938628, + "grad_norm": 0.083984375, + "learning_rate": 0.0019997498364749656, + "loss": 0.2266, + "step": 1714 + }, + { + "epoch": 0.014887023550142793, + "grad_norm": 0.07421875, + "learning_rate": 0.001999749134294532, + "loss": 0.2285, + "step": 1715 + }, + { + "epoch": 0.014895704030346959, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019997484311301454, + "loss": 0.252, + "step": 1716 + }, + { + "epoch": 0.014904384510551124, + "grad_norm": 0.0771484375, + "learning_rate": 0.001999747726981808, + "loss": 0.2578, + "step": 1717 + }, + { + "epoch": 0.014913064990755289, + "grad_norm": 0.07080078125, + "learning_rate": 0.00199974702184952, + "loss": 0.2168, + "step": 1718 + }, + { + "epoch": 0.014921745470959454, + "grad_norm": 0.1083984375, + "learning_rate": 0.001999746315733282, + "loss": 0.2305, + "step": 1719 + }, + { + "epoch": 0.014930425951163619, + "grad_norm": 0.08056640625, + "learning_rate": 0.001999745608633095, + "loss": 0.2852, + "step": 1720 + }, + { + "epoch": 0.014939106431367784, + "grad_norm": 0.0693359375, + "learning_rate": 0.00199974490054896, + "loss": 0.2275, + "step": 1721 + }, + { + "epoch": 0.014947786911571947, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999744191480877, + "loss": 0.2969, + "step": 1722 + }, + { + "epoch": 0.014956467391776113, + "grad_norm": 0.171875, + "learning_rate": 0.0019997434814288477, + "loss": 0.3086, + "step": 1723 + }, + { + "epoch": 0.014965147871980278, + "grad_norm": 0.08349609375, + "learning_rate": 0.001999742770392872, + "loss": 0.2559, + "step": 1724 + }, + { + "epoch": 0.014973828352184443, + "grad_norm": 0.10791015625, + "learning_rate": 0.001999742058372951, + "loss": 0.3418, + "step": 1725 + }, + { + "epoch": 0.014982508832388608, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019997413453690864, + "loss": 0.2539, + "step": 1726 + }, + { + "epoch": 0.014991189312592773, + "grad_norm": 0.087890625, + "learning_rate": 0.0019997406313812774, + "loss": 0.3047, + "step": 1727 + }, + { + "epoch": 0.014999869792796938, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019997399164095263, + "loss": 0.2402, + "step": 1728 + }, + { + "epoch": 0.015008550273001102, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019997392004538327, + "loss": 0.2021, + "step": 1729 + }, + { + "epoch": 0.015017230753205267, + "grad_norm": 0.1015625, + "learning_rate": 0.001999738483514198, + "loss": 0.2539, + "step": 1730 + }, + { + "epoch": 0.015025911233409432, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019997377655906227, + "loss": 0.2344, + "step": 1731 + }, + { + "epoch": 0.015034591713613597, + "grad_norm": 0.103515625, + "learning_rate": 0.0019997370466831076, + "loss": 0.3164, + "step": 1732 + }, + { + "epoch": 0.015043272193817762, + "grad_norm": 0.337890625, + "learning_rate": 0.001999736326791654, + "loss": 0.4219, + "step": 1733 + }, + { + "epoch": 0.015051952674021927, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019997356059162615, + "loss": 0.2715, + "step": 1734 + }, + { + "epoch": 0.015060633154226092, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019997348840569322, + "loss": 0.2334, + "step": 1735 + }, + { + "epoch": 0.015069313634430257, + "grad_norm": 0.080078125, + "learning_rate": 0.0019997341612136665, + "loss": 0.1836, + "step": 1736 + }, + { + "epoch": 0.01507799411463442, + "grad_norm": 0.11181640625, + "learning_rate": 0.001999733437386465, + "loss": 0.2773, + "step": 1737 + }, + { + "epoch": 0.015086674594838586, + "grad_norm": 0.138671875, + "learning_rate": 0.001999732712575328, + "loss": 0.2461, + "step": 1738 + }, + { + "epoch": 0.015095355075042751, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999731986780257, + "loss": 0.2461, + "step": 1739 + }, + { + "epoch": 0.015104035555246916, + "grad_norm": 0.0703125, + "learning_rate": 0.0019997312600012526, + "loss": 0.1787, + "step": 1740 + }, + { + "epoch": 0.015112716035451081, + "grad_norm": 0.08203125, + "learning_rate": 0.0019997305322383163, + "loss": 0.2168, + "step": 1741 + }, + { + "epoch": 0.015121396515655246, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019997298034914474, + "loss": 0.2324, + "step": 1742 + }, + { + "epoch": 0.015130076995859412, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019997290737606478, + "loss": 0.2422, + "step": 1743 + }, + { + "epoch": 0.015138757476063577, + "grad_norm": 0.0751953125, + "learning_rate": 0.001999728343045918, + "loss": 0.2354, + "step": 1744 + }, + { + "epoch": 0.01514743795626774, + "grad_norm": 0.10009765625, + "learning_rate": 0.001999727611347259, + "loss": 0.2451, + "step": 1745 + }, + { + "epoch": 0.015156118436471905, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019997268786646713, + "loss": 0.3359, + "step": 1746 + }, + { + "epoch": 0.01516479891667607, + "grad_norm": 0.103515625, + "learning_rate": 0.0019997261449981553, + "loss": 0.2656, + "step": 1747 + }, + { + "epoch": 0.015173479396880235, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019997254103477128, + "loss": 0.2559, + "step": 1748 + }, + { + "epoch": 0.0151821598770844, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019997246747133443, + "loss": 0.2168, + "step": 1749 + }, + { + "epoch": 0.015190840357288566, + "grad_norm": 0.06884765625, + "learning_rate": 0.00199972393809505, + "loss": 0.208, + "step": 1750 + }, + { + "epoch": 0.01519952083749273, + "grad_norm": 0.076171875, + "learning_rate": 0.0019997232004928312, + "loss": 0.2227, + "step": 1751 + }, + { + "epoch": 0.015208201317696894, + "grad_norm": 0.111328125, + "learning_rate": 0.001999722461906689, + "loss": 0.3418, + "step": 1752 + }, + { + "epoch": 0.01521688179790106, + "grad_norm": 0.07275390625, + "learning_rate": 0.001999721722336623, + "loss": 0.2227, + "step": 1753 + }, + { + "epoch": 0.015225562278105224, + "grad_norm": 0.080078125, + "learning_rate": 0.0019997209817826356, + "loss": 0.2188, + "step": 1754 + }, + { + "epoch": 0.01523424275830939, + "grad_norm": 0.10595703125, + "learning_rate": 0.001999720240244727, + "loss": 0.2363, + "step": 1755 + }, + { + "epoch": 0.015242923238513555, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019997194977228972, + "loss": 0.2207, + "step": 1756 + }, + { + "epoch": 0.01525160371871772, + "grad_norm": 0.142578125, + "learning_rate": 0.001999718754217148, + "loss": 0.21, + "step": 1757 + }, + { + "epoch": 0.015260284198921885, + "grad_norm": 0.12255859375, + "learning_rate": 0.00199971800972748, + "loss": 0.2637, + "step": 1758 + }, + { + "epoch": 0.01526896467912605, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019997172642538933, + "loss": 0.2266, + "step": 1759 + }, + { + "epoch": 0.015277645159330213, + "grad_norm": 0.0771484375, + "learning_rate": 0.00199971651779639, + "loss": 0.1797, + "step": 1760 + }, + { + "epoch": 0.015286325639534378, + "grad_norm": 0.083984375, + "learning_rate": 0.00199971577035497, + "loss": 0.2559, + "step": 1761 + }, + { + "epoch": 0.015295006119738544, + "grad_norm": 0.068359375, + "learning_rate": 0.0019997150219296343, + "loss": 0.2344, + "step": 1762 + }, + { + "epoch": 0.015303686599942709, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999714272520384, + "loss": 0.2578, + "step": 1763 + }, + { + "epoch": 0.015312367080146874, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019997135221272192, + "loss": 0.1914, + "step": 1764 + }, + { + "epoch": 0.015321047560351039, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999712770750142, + "loss": 0.2695, + "step": 1765 + }, + { + "epoch": 0.015329728040555204, + "grad_norm": 0.08203125, + "learning_rate": 0.0019997120183891516, + "loss": 0.2197, + "step": 1766 + }, + { + "epoch": 0.01533840852075937, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019997112650442504, + "loss": 0.2773, + "step": 1767 + }, + { + "epoch": 0.015347089000963533, + "grad_norm": 0.0888671875, + "learning_rate": 0.001999710510715438, + "loss": 0.3086, + "step": 1768 + }, + { + "epoch": 0.015355769481167698, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019997097554027158, + "loss": 0.2812, + "step": 1769 + }, + { + "epoch": 0.015364449961371863, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019997089991060845, + "loss": 0.1846, + "step": 1770 + }, + { + "epoch": 0.015373130441576028, + "grad_norm": 0.07080078125, + "learning_rate": 0.001999708241825545, + "loss": 0.2354, + "step": 1771 + }, + { + "epoch": 0.015381810921780193, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019997074835610977, + "loss": 0.2305, + "step": 1772 + }, + { + "epoch": 0.015390491401984358, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019997067243127443, + "loss": 0.2148, + "step": 1773 + }, + { + "epoch": 0.015399171882188523, + "grad_norm": 0.1591796875, + "learning_rate": 0.001999705964080485, + "loss": 0.334, + "step": 1774 + }, + { + "epoch": 0.015407852362392687, + "grad_norm": 0.0947265625, + "learning_rate": 0.001999705202864321, + "loss": 0.2178, + "step": 1775 + }, + { + "epoch": 0.015416532842596852, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019997044406642526, + "loss": 0.1758, + "step": 1776 + }, + { + "epoch": 0.015425213322801017, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019997036774802813, + "loss": 0.1992, + "step": 1777 + }, + { + "epoch": 0.015433893803005182, + "grad_norm": 0.1689453125, + "learning_rate": 0.001999702913312407, + "loss": 0.2871, + "step": 1778 + }, + { + "epoch": 0.015442574283209347, + "grad_norm": 0.095703125, + "learning_rate": 0.0019997021481606312, + "loss": 0.2256, + "step": 1779 + }, + { + "epoch": 0.015451254763413512, + "grad_norm": 0.09765625, + "learning_rate": 0.001999701382024955, + "loss": 0.2598, + "step": 1780 + }, + { + "epoch": 0.015459935243617677, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019997006149053793, + "loss": 0.2578, + "step": 1781 + }, + { + "epoch": 0.015468615723821843, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019996998468019035, + "loss": 0.2188, + "step": 1782 + }, + { + "epoch": 0.015477296204026006, + "grad_norm": 0.1337890625, + "learning_rate": 0.00199969907771453, + "loss": 0.2969, + "step": 1783 + }, + { + "epoch": 0.015485976684230171, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019996983076432592, + "loss": 0.2266, + "step": 1784 + }, + { + "epoch": 0.015494657164434336, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019996975365880916, + "loss": 0.2168, + "step": 1785 + }, + { + "epoch": 0.015503337644638501, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019996967645490283, + "loss": 0.2324, + "step": 1786 + }, + { + "epoch": 0.015512018124842666, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019996959915260706, + "loss": 0.2695, + "step": 1787 + }, + { + "epoch": 0.015520698605046832, + "grad_norm": 0.072265625, + "learning_rate": 0.001999695217519218, + "loss": 0.1855, + "step": 1788 + }, + { + "epoch": 0.015529379085250997, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019996944425284733, + "loss": 0.1904, + "step": 1789 + }, + { + "epoch": 0.015538059565455162, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019996936665538354, + "loss": 0.2578, + "step": 1790 + }, + { + "epoch": 0.015546740045659325, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019996928895953067, + "loss": 0.2324, + "step": 1791 + }, + { + "epoch": 0.01555542052586349, + "grad_norm": 0.07470703125, + "learning_rate": 0.001999692111652887, + "loss": 0.2695, + "step": 1792 + }, + { + "epoch": 0.015564101006067655, + "grad_norm": 0.166015625, + "learning_rate": 0.0019996913327265777, + "loss": 0.2695, + "step": 1793 + }, + { + "epoch": 0.01557278148627182, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999690552816379, + "loss": 0.2773, + "step": 1794 + }, + { + "epoch": 0.015581461966475986, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019996897719222924, + "loss": 0.1982, + "step": 1795 + }, + { + "epoch": 0.01559014244668015, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999688990044319, + "loss": 0.2656, + "step": 1796 + }, + { + "epoch": 0.015598822926884316, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019996882071824586, + "loss": 0.2305, + "step": 1797 + }, + { + "epoch": 0.015607503407088481, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019996874233367133, + "loss": 0.1904, + "step": 1798 + }, + { + "epoch": 0.015616183887292644, + "grad_norm": 0.06201171875, + "learning_rate": 0.0019996866385070832, + "loss": 0.2578, + "step": 1799 + }, + { + "epoch": 0.01562486436749681, + "grad_norm": 0.072265625, + "learning_rate": 0.001999685852693569, + "loss": 0.2266, + "step": 1800 + }, + { + "epoch": 0.015633544847700975, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019996850658961724, + "loss": 0.2285, + "step": 1801 + }, + { + "epoch": 0.01564222532790514, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019996842781148934, + "loss": 0.2041, + "step": 1802 + }, + { + "epoch": 0.015650905808109305, + "grad_norm": 0.0791015625, + "learning_rate": 0.001999683489349733, + "loss": 0.2832, + "step": 1803 + }, + { + "epoch": 0.01565958628831347, + "grad_norm": 0.0859375, + "learning_rate": 0.0019996826996006925, + "loss": 0.248, + "step": 1804 + }, + { + "epoch": 0.015668266768517635, + "grad_norm": 0.064453125, + "learning_rate": 0.001999681908867773, + "loss": 0.1719, + "step": 1805 + }, + { + "epoch": 0.0156769472487218, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999681117150974, + "loss": 0.2578, + "step": 1806 + }, + { + "epoch": 0.015685627728925965, + "grad_norm": 0.056884765625, + "learning_rate": 0.0019996803244502976, + "loss": 0.207, + "step": 1807 + }, + { + "epoch": 0.01569430820913013, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019996795307657446, + "loss": 0.2363, + "step": 1808 + }, + { + "epoch": 0.015702988689334296, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999678736097315, + "loss": 0.2617, + "step": 1809 + }, + { + "epoch": 0.015711669169538457, + "grad_norm": 0.103515625, + "learning_rate": 0.0019996779404450105, + "loss": 0.2354, + "step": 1810 + }, + { + "epoch": 0.015720349649742622, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999677143808832, + "loss": 0.3027, + "step": 1811 + }, + { + "epoch": 0.015729030129946787, + "grad_norm": 0.083984375, + "learning_rate": 0.00199967634618878, + "loss": 0.2637, + "step": 1812 + }, + { + "epoch": 0.015737710610150953, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019996755475848553, + "loss": 0.293, + "step": 1813 + }, + { + "epoch": 0.015746391090355118, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999674747997059, + "loss": 0.2422, + "step": 1814 + }, + { + "epoch": 0.015755071570559283, + "grad_norm": 0.0888671875, + "learning_rate": 0.001999673947425392, + "loss": 0.3125, + "step": 1815 + }, + { + "epoch": 0.015763752050763448, + "grad_norm": 0.056884765625, + "learning_rate": 0.001999673145869855, + "loss": 0.1943, + "step": 1816 + }, + { + "epoch": 0.015772432530967613, + "grad_norm": 0.072265625, + "learning_rate": 0.001999672343330449, + "loss": 0.2061, + "step": 1817 + }, + { + "epoch": 0.015781113011171778, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019996715398071744, + "loss": 0.1641, + "step": 1818 + }, + { + "epoch": 0.015789793491375943, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019996707353000334, + "loss": 0.248, + "step": 1819 + }, + { + "epoch": 0.01579847397158011, + "grad_norm": 0.080078125, + "learning_rate": 0.0019996699298090253, + "loss": 0.2539, + "step": 1820 + }, + { + "epoch": 0.015807154451784274, + "grad_norm": 0.064453125, + "learning_rate": 0.001999669123334152, + "loss": 0.2168, + "step": 1821 + }, + { + "epoch": 0.01581583493198844, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999668315875414, + "loss": 0.2168, + "step": 1822 + }, + { + "epoch": 0.015824515412192604, + "grad_norm": 0.0751953125, + "learning_rate": 0.001999667507432812, + "loss": 0.209, + "step": 1823 + }, + { + "epoch": 0.01583319589239677, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019996666980063474, + "loss": 0.2578, + "step": 1824 + }, + { + "epoch": 0.015841876372600934, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999665887596021, + "loss": 0.2207, + "step": 1825 + }, + { + "epoch": 0.015850556852805096, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019996650762018333, + "loss": 0.2656, + "step": 1826 + }, + { + "epoch": 0.01585923733300926, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999664263823785, + "loss": 0.2383, + "step": 1827 + }, + { + "epoch": 0.015867917813213426, + "grad_norm": 0.095703125, + "learning_rate": 0.001999663450461878, + "loss": 0.2324, + "step": 1828 + }, + { + "epoch": 0.01587659829341759, + "grad_norm": 0.0654296875, + "learning_rate": 0.001999662636116112, + "loss": 0.2578, + "step": 1829 + }, + { + "epoch": 0.015885278773621756, + "grad_norm": 0.099609375, + "learning_rate": 0.001999661820786489, + "loss": 0.2324, + "step": 1830 + }, + { + "epoch": 0.01589395925382592, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019996610044730094, + "loss": 0.332, + "step": 1831 + }, + { + "epoch": 0.015902639734030086, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019996601871756737, + "loss": 0.1855, + "step": 1832 + }, + { + "epoch": 0.01591132021423425, + "grad_norm": 0.095703125, + "learning_rate": 0.0019996593688944827, + "loss": 0.2676, + "step": 1833 + }, + { + "epoch": 0.015920000694438417, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019996585496294384, + "loss": 0.2598, + "step": 1834 + }, + { + "epoch": 0.015928681174642582, + "grad_norm": 0.087890625, + "learning_rate": 0.001999657729380541, + "loss": 0.2256, + "step": 1835 + }, + { + "epoch": 0.015937361654846747, + "grad_norm": 0.06494140625, + "learning_rate": 0.001999656908147791, + "loss": 0.1865, + "step": 1836 + }, + { + "epoch": 0.015946042135050912, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019996560859311904, + "loss": 0.2246, + "step": 1837 + }, + { + "epoch": 0.015954722615255077, + "grad_norm": 0.08935546875, + "learning_rate": 0.001999655262730739, + "loss": 0.248, + "step": 1838 + }, + { + "epoch": 0.015963403095459242, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019996544385464383, + "loss": 0.2168, + "step": 1839 + }, + { + "epoch": 0.015972083575663407, + "grad_norm": 0.09423828125, + "learning_rate": 0.001999653613378289, + "loss": 0.2129, + "step": 1840 + }, + { + "epoch": 0.01598076405586757, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019996527872262913, + "loss": 0.1846, + "step": 1841 + }, + { + "epoch": 0.015989444536071734, + "grad_norm": 0.05810546875, + "learning_rate": 0.0019996519600904475, + "loss": 0.1777, + "step": 1842 + }, + { + "epoch": 0.0159981250162759, + "grad_norm": 0.09033203125, + "learning_rate": 0.001999651131970758, + "loss": 0.2715, + "step": 1843 + }, + { + "epoch": 0.016006805496480064, + "grad_norm": 0.1357421875, + "learning_rate": 0.001999650302867223, + "loss": 0.3555, + "step": 1844 + }, + { + "epoch": 0.01601548597668423, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019996494727798444, + "loss": 0.1914, + "step": 1845 + }, + { + "epoch": 0.016024166456888395, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019996486417086226, + "loss": 0.291, + "step": 1846 + }, + { + "epoch": 0.01603284693709256, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019996478096535584, + "loss": 0.2422, + "step": 1847 + }, + { + "epoch": 0.016041527417296725, + "grad_norm": 0.201171875, + "learning_rate": 0.0019996469766146528, + "loss": 0.2559, + "step": 1848 + }, + { + "epoch": 0.01605020789750089, + "grad_norm": 0.0712890625, + "learning_rate": 0.001999646142591907, + "loss": 0.1982, + "step": 1849 + }, + { + "epoch": 0.016058888377705055, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019996453075853218, + "loss": 0.2227, + "step": 1850 + }, + { + "epoch": 0.01606756885790922, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019996444715948978, + "loss": 0.2715, + "step": 1851 + }, + { + "epoch": 0.016076249338113385, + "grad_norm": 0.07373046875, + "learning_rate": 0.001999643634620636, + "loss": 0.2227, + "step": 1852 + }, + { + "epoch": 0.01608492981831755, + "grad_norm": 0.059326171875, + "learning_rate": 0.001999642796662538, + "loss": 0.2012, + "step": 1853 + }, + { + "epoch": 0.016093610298521716, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019996419577206033, + "loss": 0.3027, + "step": 1854 + }, + { + "epoch": 0.01610229077872588, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999641117794834, + "loss": 0.2227, + "step": 1855 + }, + { + "epoch": 0.016110971258930042, + "grad_norm": 0.076171875, + "learning_rate": 0.001999640276885231, + "loss": 0.2344, + "step": 1856 + }, + { + "epoch": 0.016119651739134207, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019996394349917944, + "loss": 0.2539, + "step": 1857 + }, + { + "epoch": 0.016128332219338373, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019996385921145264, + "loss": 0.1982, + "step": 1858 + }, + { + "epoch": 0.016137012699542538, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019996377482534265, + "loss": 0.2344, + "step": 1859 + }, + { + "epoch": 0.016145693179746703, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019996369034084964, + "loss": 0.2676, + "step": 1860 + }, + { + "epoch": 0.016154373659950868, + "grad_norm": 0.07080078125, + "learning_rate": 0.001999636057579737, + "loss": 0.2129, + "step": 1861 + }, + { + "epoch": 0.016163054140155033, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999635210767149, + "loss": 0.2422, + "step": 1862 + }, + { + "epoch": 0.016171734620359198, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019996343629707335, + "loss": 0.2334, + "step": 1863 + }, + { + "epoch": 0.016180415100563363, + "grad_norm": 0.0849609375, + "learning_rate": 0.001999633514190491, + "loss": 0.2051, + "step": 1864 + }, + { + "epoch": 0.01618909558076753, + "grad_norm": 0.062255859375, + "learning_rate": 0.001999632664426424, + "loss": 0.1699, + "step": 1865 + }, + { + "epoch": 0.016197776060971694, + "grad_norm": 0.07421875, + "learning_rate": 0.001999631813678531, + "loss": 0.2266, + "step": 1866 + }, + { + "epoch": 0.01620645654117586, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999630961946815, + "loss": 0.2539, + "step": 1867 + }, + { + "epoch": 0.016215137021380024, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019996301092312756, + "loss": 0.2031, + "step": 1868 + }, + { + "epoch": 0.01622381750158419, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019996292555319144, + "loss": 0.2168, + "step": 1869 + }, + { + "epoch": 0.016232497981788354, + "grad_norm": 0.087890625, + "learning_rate": 0.001999628400848732, + "loss": 0.2305, + "step": 1870 + }, + { + "epoch": 0.01624117846199252, + "grad_norm": 0.0771484375, + "learning_rate": 0.00199962754518173, + "loss": 0.2432, + "step": 1871 + }, + { + "epoch": 0.01624985894219668, + "grad_norm": 0.09033203125, + "learning_rate": 0.001999626688530908, + "loss": 0.2275, + "step": 1872 + }, + { + "epoch": 0.016258539422400846, + "grad_norm": 0.06591796875, + "learning_rate": 0.001999625830896268, + "loss": 0.209, + "step": 1873 + }, + { + "epoch": 0.01626721990260501, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019996249722778114, + "loss": 0.2285, + "step": 1874 + }, + { + "epoch": 0.016275900382809176, + "grad_norm": 0.080078125, + "learning_rate": 0.001999624112675538, + "loss": 0.1973, + "step": 1875 + }, + { + "epoch": 0.01628458086301334, + "grad_norm": 0.0791015625, + "learning_rate": 0.001999623252089449, + "loss": 0.252, + "step": 1876 + }, + { + "epoch": 0.016293261343217506, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019996223905195455, + "loss": 0.2354, + "step": 1877 + }, + { + "epoch": 0.01630194182342167, + "grad_norm": 0.053955078125, + "learning_rate": 0.0019996215279658286, + "loss": 0.1934, + "step": 1878 + }, + { + "epoch": 0.016310622303625837, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019996206644282994, + "loss": 0.2812, + "step": 1879 + }, + { + "epoch": 0.01631930278383, + "grad_norm": 0.07861328125, + "learning_rate": 0.001999619799906958, + "loss": 0.2012, + "step": 1880 + }, + { + "epoch": 0.016327983264034167, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019996189344018067, + "loss": 0.25, + "step": 1881 + }, + { + "epoch": 0.016336663744238332, + "grad_norm": 0.06787109375, + "learning_rate": 0.001999618067912845, + "loss": 0.1758, + "step": 1882 + }, + { + "epoch": 0.016345344224442497, + "grad_norm": 0.0703125, + "learning_rate": 0.0019996172004400747, + "loss": 0.2031, + "step": 1883 + }, + { + "epoch": 0.016354024704646662, + "grad_norm": 0.083984375, + "learning_rate": 0.0019996163319834967, + "loss": 0.2334, + "step": 1884 + }, + { + "epoch": 0.016362705184850827, + "grad_norm": 0.080078125, + "learning_rate": 0.001999615462543111, + "loss": 0.2412, + "step": 1885 + }, + { + "epoch": 0.016371385665054992, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019996145921189206, + "loss": 0.2285, + "step": 1886 + }, + { + "epoch": 0.016380066145259154, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019996137207109244, + "loss": 0.2773, + "step": 1887 + }, + { + "epoch": 0.01638874662546332, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019996128483191246, + "loss": 0.2129, + "step": 1888 + }, + { + "epoch": 0.016397427105667484, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999611974943521, + "loss": 0.2402, + "step": 1889 + }, + { + "epoch": 0.01640610758587165, + "grad_norm": 0.1298828125, + "learning_rate": 0.001999611100584116, + "loss": 0.2852, + "step": 1890 + }, + { + "epoch": 0.016414788066075815, + "grad_norm": 0.103515625, + "learning_rate": 0.0019996102252409094, + "loss": 0.2324, + "step": 1891 + }, + { + "epoch": 0.01642346854627998, + "grad_norm": 0.08544921875, + "learning_rate": 0.001999609348913903, + "loss": 0.2158, + "step": 1892 + }, + { + "epoch": 0.016432149026484145, + "grad_norm": 0.07421875, + "learning_rate": 0.001999608471603097, + "loss": 0.2236, + "step": 1893 + }, + { + "epoch": 0.01644082950668831, + "grad_norm": 0.08935546875, + "learning_rate": 0.001999607593308492, + "loss": 0.2227, + "step": 1894 + }, + { + "epoch": 0.016449509986892475, + "grad_norm": 0.07763671875, + "learning_rate": 0.001999606714030091, + "loss": 0.1816, + "step": 1895 + }, + { + "epoch": 0.01645819046709664, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019996058337678926, + "loss": 0.2051, + "step": 1896 + }, + { + "epoch": 0.016466870947300805, + "grad_norm": 0.08203125, + "learning_rate": 0.001999604952521899, + "loss": 0.2207, + "step": 1897 + }, + { + "epoch": 0.01647555142750497, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019996040702921114, + "loss": 0.2617, + "step": 1898 + }, + { + "epoch": 0.016484231907709136, + "grad_norm": 0.076171875, + "learning_rate": 0.0019996031870785297, + "loss": 0.2168, + "step": 1899 + }, + { + "epoch": 0.0164929123879133, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999602302881156, + "loss": 0.2246, + "step": 1900 + }, + { + "epoch": 0.016501592868117466, + "grad_norm": 0.08251953125, + "learning_rate": 0.00199960141769999, + "loss": 0.2656, + "step": 1901 + }, + { + "epoch": 0.01651027334832163, + "grad_norm": 0.1494140625, + "learning_rate": 0.001999600531535034, + "loss": 0.2715, + "step": 1902 + }, + { + "epoch": 0.016518953828525793, + "grad_norm": 0.0703125, + "learning_rate": 0.0019995996443862886, + "loss": 0.1953, + "step": 1903 + }, + { + "epoch": 0.016527634308729958, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019995987562537545, + "loss": 0.2168, + "step": 1904 + }, + { + "epoch": 0.016536314788934123, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999597867137432, + "loss": 0.2246, + "step": 1905 + }, + { + "epoch": 0.016544995269138288, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019995969770373236, + "loss": 0.1924, + "step": 1906 + }, + { + "epoch": 0.016553675749342453, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999596085953429, + "loss": 0.2119, + "step": 1907 + }, + { + "epoch": 0.016562356229546618, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019995951938857497, + "loss": 0.2227, + "step": 1908 + }, + { + "epoch": 0.016571036709750783, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019995943008342867, + "loss": 0.2129, + "step": 1909 + }, + { + "epoch": 0.01657971718995495, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019995934067990407, + "loss": 0.2734, + "step": 1910 + }, + { + "epoch": 0.016588397670159113, + "grad_norm": 0.11181640625, + "learning_rate": 0.001999592511780013, + "loss": 0.3398, + "step": 1911 + }, + { + "epoch": 0.01659707815036328, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019995916157772046, + "loss": 0.2266, + "step": 1912 + }, + { + "epoch": 0.016605758630567444, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999590718790616, + "loss": 0.248, + "step": 1913 + }, + { + "epoch": 0.01661443911077161, + "grad_norm": 0.0732421875, + "learning_rate": 0.001999589820820249, + "loss": 0.1787, + "step": 1914 + }, + { + "epoch": 0.016623119590975774, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019995889218661035, + "loss": 0.2422, + "step": 1915 + }, + { + "epoch": 0.01663180007117994, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019995880219281816, + "loss": 0.2324, + "step": 1916 + }, + { + "epoch": 0.016640480551384104, + "grad_norm": 0.10107421875, + "learning_rate": 0.001999587121006483, + "loss": 0.252, + "step": 1917 + }, + { + "epoch": 0.016649161031588266, + "grad_norm": 0.08349609375, + "learning_rate": 0.00199958621910101, + "loss": 0.1973, + "step": 1918 + }, + { + "epoch": 0.01665784151179243, + "grad_norm": 0.099609375, + "learning_rate": 0.001999585316211763, + "loss": 0.2617, + "step": 1919 + }, + { + "epoch": 0.016666521991996596, + "grad_norm": 0.064453125, + "learning_rate": 0.001999584412338743, + "loss": 0.2158, + "step": 1920 + }, + { + "epoch": 0.01667520247220076, + "grad_norm": 0.09765625, + "learning_rate": 0.001999583507481951, + "loss": 0.2715, + "step": 1921 + }, + { + "epoch": 0.016683882952404926, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999582601641388, + "loss": 0.2285, + "step": 1922 + }, + { + "epoch": 0.01669256343260909, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999581694817055, + "loss": 0.1934, + "step": 1923 + }, + { + "epoch": 0.016701243912813257, + "grad_norm": 0.115234375, + "learning_rate": 0.0019995807870089527, + "loss": 0.2656, + "step": 1924 + }, + { + "epoch": 0.01670992439301742, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999579878217083, + "loss": 0.1914, + "step": 1925 + }, + { + "epoch": 0.016718604873221587, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019995789684414456, + "loss": 0.1963, + "step": 1926 + }, + { + "epoch": 0.016727285353425752, + "grad_norm": 0.08984375, + "learning_rate": 0.0019995780576820424, + "loss": 0.2246, + "step": 1927 + }, + { + "epoch": 0.016735965833629917, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019995771459388745, + "loss": 0.2246, + "step": 1928 + }, + { + "epoch": 0.016744646313834082, + "grad_norm": 0.083984375, + "learning_rate": 0.001999576233211942, + "loss": 0.2422, + "step": 1929 + }, + { + "epoch": 0.016753326794038247, + "grad_norm": 0.068359375, + "learning_rate": 0.0019995753195012466, + "loss": 0.25, + "step": 1930 + }, + { + "epoch": 0.016762007274242412, + "grad_norm": 0.087890625, + "learning_rate": 0.0019995744048067893, + "loss": 0.2559, + "step": 1931 + }, + { + "epoch": 0.016770687754446578, + "grad_norm": 0.078125, + "learning_rate": 0.0019995734891285707, + "loss": 0.2441, + "step": 1932 + }, + { + "epoch": 0.01677936823465074, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999572572466592, + "loss": 0.2295, + "step": 1933 + }, + { + "epoch": 0.016788048714854904, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019995716548208546, + "loss": 0.2031, + "step": 1934 + }, + { + "epoch": 0.01679672919505907, + "grad_norm": 0.095703125, + "learning_rate": 0.001999570736191359, + "loss": 0.3203, + "step": 1935 + }, + { + "epoch": 0.016805409675263235, + "grad_norm": 0.0625, + "learning_rate": 0.0019995698165781064, + "loss": 0.168, + "step": 1936 + }, + { + "epoch": 0.0168140901554674, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019995688959810977, + "loss": 0.2812, + "step": 1937 + }, + { + "epoch": 0.016822770635671565, + "grad_norm": 0.1162109375, + "learning_rate": 0.001999567974400334, + "loss": 0.2539, + "step": 1938 + }, + { + "epoch": 0.01683145111587573, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019995670518358163, + "loss": 0.3047, + "step": 1939 + }, + { + "epoch": 0.016840131596079895, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999566128287546, + "loss": 0.2295, + "step": 1940 + }, + { + "epoch": 0.01684881207628406, + "grad_norm": 0.138671875, + "learning_rate": 0.001999565203755523, + "loss": 0.2695, + "step": 1941 + }, + { + "epoch": 0.016857492556488225, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999564278239749, + "loss": 0.1846, + "step": 1942 + }, + { + "epoch": 0.01686617303669239, + "grad_norm": 0.095703125, + "learning_rate": 0.0019995633517402253, + "loss": 0.2812, + "step": 1943 + }, + { + "epoch": 0.016874853516896555, + "grad_norm": 0.0859375, + "learning_rate": 0.001999562424256953, + "loss": 0.2246, + "step": 1944 + }, + { + "epoch": 0.01688353399710072, + "grad_norm": 0.087890625, + "learning_rate": 0.001999561495789932, + "loss": 0.2197, + "step": 1945 + }, + { + "epoch": 0.016892214477304886, + "grad_norm": 0.10693359375, + "learning_rate": 0.001999560566339165, + "loss": 0.2656, + "step": 1946 + }, + { + "epoch": 0.01690089495750905, + "grad_norm": 0.09912109375, + "learning_rate": 0.001999559635904651, + "loss": 0.2168, + "step": 1947 + }, + { + "epoch": 0.016909575437713216, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019995587044863926, + "loss": 0.2422, + "step": 1948 + }, + { + "epoch": 0.016918255917917378, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019995577720843907, + "loss": 0.2363, + "step": 1949 + }, + { + "epoch": 0.016926936398121543, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019995568386986452, + "loss": 0.1846, + "step": 1950 + }, + { + "epoch": 0.016935616878325708, + "grad_norm": 0.080078125, + "learning_rate": 0.0019995559043291585, + "loss": 0.2422, + "step": 1951 + }, + { + "epoch": 0.016944297358529873, + "grad_norm": 0.103515625, + "learning_rate": 0.0019995549689759305, + "loss": 0.21, + "step": 1952 + }, + { + "epoch": 0.016952977838734038, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999554032638963, + "loss": 0.25, + "step": 1953 + }, + { + "epoch": 0.016961658318938203, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019995530953182566, + "loss": 0.2461, + "step": 1954 + }, + { + "epoch": 0.01697033879914237, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019995521570138125, + "loss": 0.2227, + "step": 1955 + }, + { + "epoch": 0.016979019279346533, + "grad_norm": 0.0849609375, + "learning_rate": 0.001999551217725632, + "loss": 0.2344, + "step": 1956 + }, + { + "epoch": 0.0169876997595507, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999550277453715, + "loss": 0.1963, + "step": 1957 + }, + { + "epoch": 0.016996380239754864, + "grad_norm": 0.0849609375, + "learning_rate": 0.001999549336198064, + "loss": 0.2832, + "step": 1958 + }, + { + "epoch": 0.01700506071995903, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019995483939586793, + "loss": 0.2441, + "step": 1959 + }, + { + "epoch": 0.017013741200163194, + "grad_norm": 0.6796875, + "learning_rate": 0.0019995474507355617, + "loss": 0.3984, + "step": 1960 + }, + { + "epoch": 0.01702242168036736, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019995465065287127, + "loss": 0.21, + "step": 1961 + }, + { + "epoch": 0.017031102160571524, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019995455613381332, + "loss": 0.207, + "step": 1962 + }, + { + "epoch": 0.01703978264077569, + "grad_norm": 0.0634765625, + "learning_rate": 0.001999544615163824, + "loss": 0.1768, + "step": 1963 + }, + { + "epoch": 0.01704846312097985, + "grad_norm": 0.07421875, + "learning_rate": 0.0019995436680057864, + "loss": 0.2637, + "step": 1964 + }, + { + "epoch": 0.017057143601184016, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019995427198640217, + "loss": 0.2441, + "step": 1965 + }, + { + "epoch": 0.01706582408138818, + "grad_norm": 0.10546875, + "learning_rate": 0.00199954177073853, + "loss": 0.2275, + "step": 1966 + }, + { + "epoch": 0.017074504561592346, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019995408206293134, + "loss": 0.2285, + "step": 1967 + }, + { + "epoch": 0.01708318504179651, + "grad_norm": 0.20703125, + "learning_rate": 0.0019995398695363724, + "loss": 0.3086, + "step": 1968 + }, + { + "epoch": 0.017091865522000677, + "grad_norm": 0.11376953125, + "learning_rate": 0.001999538917459708, + "loss": 0.2314, + "step": 1969 + }, + { + "epoch": 0.01710054600220484, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019995379643993213, + "loss": 0.2314, + "step": 1970 + }, + { + "epoch": 0.017109226482409007, + "grad_norm": 0.058837890625, + "learning_rate": 0.0019995370103552137, + "loss": 0.1719, + "step": 1971 + }, + { + "epoch": 0.017117906962613172, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019995360553273856, + "loss": 0.2227, + "step": 1972 + }, + { + "epoch": 0.017126587442817337, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019995350993158387, + "loss": 0.2617, + "step": 1973 + }, + { + "epoch": 0.017135267923021502, + "grad_norm": 0.08935546875, + "learning_rate": 0.001999534142320573, + "loss": 0.293, + "step": 1974 + }, + { + "epoch": 0.017143948403225667, + "grad_norm": 0.1044921875, + "learning_rate": 0.001999533184341591, + "loss": 0.2559, + "step": 1975 + }, + { + "epoch": 0.017152628883429832, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999532225378893, + "loss": 0.2109, + "step": 1976 + }, + { + "epoch": 0.017161309363633998, + "grad_norm": 0.1142578125, + "learning_rate": 0.00199953126543248, + "loss": 0.3164, + "step": 1977 + }, + { + "epoch": 0.017169989843838163, + "grad_norm": 0.07177734375, + "learning_rate": 0.001999530304502353, + "loss": 0.2129, + "step": 1978 + }, + { + "epoch": 0.017178670324042328, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999529342588513, + "loss": 0.2314, + "step": 1979 + }, + { + "epoch": 0.01718735080424649, + "grad_norm": 0.142578125, + "learning_rate": 0.0019995283796909614, + "loss": 0.3008, + "step": 1980 + }, + { + "epoch": 0.017196031284450655, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019995274158096992, + "loss": 0.2109, + "step": 1981 + }, + { + "epoch": 0.01720471176465482, + "grad_norm": 0.083984375, + "learning_rate": 0.0019995264509447275, + "loss": 0.2441, + "step": 1982 + }, + { + "epoch": 0.017213392244858985, + "grad_norm": 0.076171875, + "learning_rate": 0.0019995254850960465, + "loss": 0.2041, + "step": 1983 + }, + { + "epoch": 0.01722207272506315, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019995245182636585, + "loss": 0.2578, + "step": 1984 + }, + { + "epoch": 0.017230753205267315, + "grad_norm": 0.08544921875, + "learning_rate": 0.001999523550447564, + "loss": 0.2354, + "step": 1985 + }, + { + "epoch": 0.01723943368547148, + "grad_norm": 0.09423828125, + "learning_rate": 0.001999522581647764, + "loss": 0.2441, + "step": 1986 + }, + { + "epoch": 0.017248114165675645, + "grad_norm": 0.0703125, + "learning_rate": 0.0019995216118642595, + "loss": 0.2324, + "step": 1987 + }, + { + "epoch": 0.01725679464587981, + "grad_norm": 0.0966796875, + "learning_rate": 0.001999520641097052, + "loss": 0.2656, + "step": 1988 + }, + { + "epoch": 0.017265475126083975, + "grad_norm": 0.0947265625, + "learning_rate": 0.001999519669346142, + "loss": 0.1973, + "step": 1989 + }, + { + "epoch": 0.01727415560628814, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019995186966115307, + "loss": 0.2383, + "step": 1990 + }, + { + "epoch": 0.017282836086492306, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019995177228932194, + "loss": 0.2295, + "step": 1991 + }, + { + "epoch": 0.01729151656669647, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999516748191209, + "loss": 0.2676, + "step": 1992 + }, + { + "epoch": 0.017300197046900636, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019995157725055004, + "loss": 0.2188, + "step": 1993 + }, + { + "epoch": 0.0173088775271048, + "grad_norm": 0.07470703125, + "learning_rate": 0.001999514795836095, + "loss": 0.2324, + "step": 1994 + }, + { + "epoch": 0.017317558007308963, + "grad_norm": 0.53515625, + "learning_rate": 0.0019995138181829936, + "loss": 0.4336, + "step": 1995 + }, + { + "epoch": 0.017326238487513128, + "grad_norm": 0.0908203125, + "learning_rate": 0.001999512839546198, + "loss": 0.293, + "step": 1996 + }, + { + "epoch": 0.017334918967717293, + "grad_norm": 0.078125, + "learning_rate": 0.001999511859925708, + "loss": 0.1982, + "step": 1997 + }, + { + "epoch": 0.017343599447921458, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019995108793215257, + "loss": 0.2363, + "step": 1998 + }, + { + "epoch": 0.017352279928125623, + "grad_norm": 0.109375, + "learning_rate": 0.0019995098977336517, + "loss": 0.2559, + "step": 1999 + }, + { + "epoch": 0.01736096040832979, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019995089151620873, + "loss": 0.2012, + "step": 2000 + }, + { + "epoch": 0.017369640888533953, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019995079316068335, + "loss": 0.2598, + "step": 2001 + }, + { + "epoch": 0.01737832136873812, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019995069470678914, + "loss": 0.3105, + "step": 2002 + }, + { + "epoch": 0.017387001848942284, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019995059615452613, + "loss": 0.2422, + "step": 2003 + }, + { + "epoch": 0.01739568232914645, + "grad_norm": 0.08056640625, + "learning_rate": 0.001999504975038946, + "loss": 0.2021, + "step": 2004 + }, + { + "epoch": 0.017404362809350614, + "grad_norm": 0.07080078125, + "learning_rate": 0.001999503987548945, + "loss": 0.248, + "step": 2005 + }, + { + "epoch": 0.01741304328955478, + "grad_norm": 0.1044921875, + "learning_rate": 0.00199950299907526, + "loss": 0.2773, + "step": 2006 + }, + { + "epoch": 0.017421723769758944, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019995020096178918, + "loss": 0.1973, + "step": 2007 + }, + { + "epoch": 0.01743040424996311, + "grad_norm": 0.09765625, + "learning_rate": 0.0019995010191768427, + "loss": 0.248, + "step": 2008 + }, + { + "epoch": 0.017439084730167274, + "grad_norm": 0.078125, + "learning_rate": 0.0019995000277521118, + "loss": 0.2002, + "step": 2009 + }, + { + "epoch": 0.017447765210371436, + "grad_norm": 0.078125, + "learning_rate": 0.0019994990353437016, + "loss": 0.2676, + "step": 2010 + }, + { + "epoch": 0.0174564456905756, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019994980419516125, + "loss": 0.1768, + "step": 2011 + }, + { + "epoch": 0.017465126170779766, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019994970475758463, + "loss": 0.2109, + "step": 2012 + }, + { + "epoch": 0.01747380665098393, + "grad_norm": 0.068359375, + "learning_rate": 0.001999496052216403, + "loss": 0.2539, + "step": 2013 + }, + { + "epoch": 0.017482487131188097, + "grad_norm": 0.06884765625, + "learning_rate": 0.001999495055873285, + "loss": 0.2285, + "step": 2014 + }, + { + "epoch": 0.01749116761139226, + "grad_norm": 0.064453125, + "learning_rate": 0.0019994940585464924, + "loss": 0.2168, + "step": 2015 + }, + { + "epoch": 0.017499848091596427, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019994930602360264, + "loss": 0.2188, + "step": 2016 + }, + { + "epoch": 0.017508528571800592, + "grad_norm": 0.087890625, + "learning_rate": 0.0019994920609418885, + "loss": 0.2695, + "step": 2017 + }, + { + "epoch": 0.017517209052004757, + "grad_norm": 0.09423828125, + "learning_rate": 0.00199949106066408, + "loss": 0.2812, + "step": 2018 + }, + { + "epoch": 0.017525889532208922, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999490059402601, + "loss": 0.2139, + "step": 2019 + }, + { + "epoch": 0.017534570012413087, + "grad_norm": 0.0703125, + "learning_rate": 0.0019994890571574534, + "loss": 0.25, + "step": 2020 + }, + { + "epoch": 0.017543250492617252, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019994880539286383, + "loss": 0.2383, + "step": 2021 + }, + { + "epoch": 0.017551930972821417, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019994870497161564, + "loss": 0.2393, + "step": 2022 + }, + { + "epoch": 0.017560611453025583, + "grad_norm": 0.10302734375, + "learning_rate": 0.001999486044520009, + "loss": 0.2422, + "step": 2023 + }, + { + "epoch": 0.017569291933229748, + "grad_norm": 0.07421875, + "learning_rate": 0.0019994850383401974, + "loss": 0.2109, + "step": 2024 + }, + { + "epoch": 0.017577972413433913, + "grad_norm": 0.080078125, + "learning_rate": 0.001999484031176722, + "loss": 0.2363, + "step": 2025 + }, + { + "epoch": 0.017586652893638074, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019994830230295846, + "loss": 0.7656, + "step": 2026 + }, + { + "epoch": 0.01759533337384224, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999482013898786, + "loss": 0.2129, + "step": 2027 + }, + { + "epoch": 0.017604013854046405, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019994810037843275, + "loss": 0.2246, + "step": 2028 + }, + { + "epoch": 0.01761269433425057, + "grad_norm": 0.0810546875, + "learning_rate": 0.00199947999268621, + "loss": 0.2461, + "step": 2029 + }, + { + "epoch": 0.017621374814454735, + "grad_norm": 0.057861328125, + "learning_rate": 0.0019994789806044347, + "loss": 0.1875, + "step": 2030 + }, + { + "epoch": 0.0176300552946589, + "grad_norm": 0.10107421875, + "learning_rate": 0.001999477967539003, + "loss": 0.2891, + "step": 2031 + }, + { + "epoch": 0.017638735774863065, + "grad_norm": 0.17578125, + "learning_rate": 0.0019994769534899155, + "loss": 0.293, + "step": 2032 + }, + { + "epoch": 0.01764741625506723, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019994759384571736, + "loss": 0.2148, + "step": 2033 + }, + { + "epoch": 0.017656096735271395, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999474922440778, + "loss": 0.3027, + "step": 2034 + }, + { + "epoch": 0.01766477721547556, + "grad_norm": 0.06396484375, + "learning_rate": 0.00199947390544073, + "loss": 0.2227, + "step": 2035 + }, + { + "epoch": 0.017673457695679726, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019994728874570314, + "loss": 0.2617, + "step": 2036 + }, + { + "epoch": 0.01768213817588389, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019994718684896826, + "loss": 0.2051, + "step": 2037 + }, + { + "epoch": 0.017690818656088056, + "grad_norm": 0.072265625, + "learning_rate": 0.0019994708485386848, + "loss": 0.1777, + "step": 2038 + }, + { + "epoch": 0.01769949913629222, + "grad_norm": 0.087890625, + "learning_rate": 0.0019994698276040394, + "loss": 0.25, + "step": 2039 + }, + { + "epoch": 0.017708179616496386, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999468805685747, + "loss": 0.2305, + "step": 2040 + }, + { + "epoch": 0.017716860096700548, + "grad_norm": 0.0673828125, + "learning_rate": 0.001999467782783809, + "loss": 0.2383, + "step": 2041 + }, + { + "epoch": 0.017725540576904713, + "grad_norm": 0.154296875, + "learning_rate": 0.001999466758898227, + "loss": 0.2041, + "step": 2042 + }, + { + "epoch": 0.017734221057108878, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019994657340290014, + "loss": 0.25, + "step": 2043 + }, + { + "epoch": 0.017742901537313043, + "grad_norm": 0.41015625, + "learning_rate": 0.0019994647081761335, + "loss": 0.4082, + "step": 2044 + }, + { + "epoch": 0.01775158201751721, + "grad_norm": 0.107421875, + "learning_rate": 0.0019994636813396244, + "loss": 0.2031, + "step": 2045 + }, + { + "epoch": 0.017760262497721373, + "grad_norm": 0.07421875, + "learning_rate": 0.0019994626535194755, + "loss": 0.2734, + "step": 2046 + }, + { + "epoch": 0.01776894297792554, + "grad_norm": 0.0634765625, + "learning_rate": 0.0019994616247156877, + "loss": 0.2168, + "step": 2047 + }, + { + "epoch": 0.017777623458129704, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019994605949282622, + "loss": 0.1797, + "step": 2048 + }, + { + "epoch": 0.01778630393833387, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019994595641572004, + "loss": 0.2754, + "step": 2049 + }, + { + "epoch": 0.017794984418538034, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999458532402503, + "loss": 0.2695, + "step": 2050 + }, + { + "epoch": 0.0178036648987422, + "grad_norm": 0.0654296875, + "learning_rate": 0.001999457499664171, + "loss": 0.2207, + "step": 2051 + }, + { + "epoch": 0.017812345378946364, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999456465942206, + "loss": 0.2188, + "step": 2052 + }, + { + "epoch": 0.01782102585915053, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019994554312366087, + "loss": 0.1982, + "step": 2053 + }, + { + "epoch": 0.017829706339354694, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019994543955473808, + "loss": 0.2334, + "step": 2054 + }, + { + "epoch": 0.01783838681955886, + "grad_norm": 0.07421875, + "learning_rate": 0.001999453358874523, + "loss": 0.2383, + "step": 2055 + }, + { + "epoch": 0.017847067299763025, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019994523212180366, + "loss": 0.2559, + "step": 2056 + }, + { + "epoch": 0.017855747779967186, + "grad_norm": 0.0888671875, + "learning_rate": 0.001999451282577922, + "loss": 0.2158, + "step": 2057 + }, + { + "epoch": 0.01786442826017135, + "grad_norm": 0.0712890625, + "learning_rate": 0.001999450242954182, + "loss": 0.209, + "step": 2058 + }, + { + "epoch": 0.017873108740375517, + "grad_norm": 0.08837890625, + "learning_rate": 0.001999449202346816, + "loss": 0.2275, + "step": 2059 + }, + { + "epoch": 0.01788178922057968, + "grad_norm": 0.06201171875, + "learning_rate": 0.001999448160755826, + "loss": 0.2188, + "step": 2060 + }, + { + "epoch": 0.017890469700783847, + "grad_norm": 0.07080078125, + "learning_rate": 0.001999447118181213, + "loss": 0.2129, + "step": 2061 + }, + { + "epoch": 0.017899150180988012, + "grad_norm": 0.1875, + "learning_rate": 0.001999446074622978, + "loss": 0.168, + "step": 2062 + }, + { + "epoch": 0.017907830661192177, + "grad_norm": 0.08642578125, + "learning_rate": 0.001999445030081123, + "loss": 0.2471, + "step": 2063 + }, + { + "epoch": 0.017916511141396342, + "grad_norm": 0.06396484375, + "learning_rate": 0.001999443984555648, + "loss": 0.2217, + "step": 2064 + }, + { + "epoch": 0.017925191621600507, + "grad_norm": 0.054931640625, + "learning_rate": 0.001999442938046554, + "loss": 0.1797, + "step": 2065 + }, + { + "epoch": 0.017933872101804672, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019994418905538436, + "loss": 0.2695, + "step": 2066 + }, + { + "epoch": 0.017942552582008837, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019994408420775166, + "loss": 0.1992, + "step": 2067 + }, + { + "epoch": 0.017951233062213003, + "grad_norm": 0.099609375, + "learning_rate": 0.001999439792617575, + "loss": 0.2334, + "step": 2068 + }, + { + "epoch": 0.017959913542417168, + "grad_norm": 0.072265625, + "learning_rate": 0.001999438742174019, + "loss": 0.1738, + "step": 2069 + }, + { + "epoch": 0.017968594022621333, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019994376907468505, + "loss": 0.1914, + "step": 2070 + }, + { + "epoch": 0.017977274502825498, + "grad_norm": 0.1171875, + "learning_rate": 0.001999436638336071, + "loss": 0.2852, + "step": 2071 + }, + { + "epoch": 0.01798595498302966, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019994355849416805, + "loss": 0.2539, + "step": 2072 + }, + { + "epoch": 0.017994635463233825, + "grad_norm": 0.08203125, + "learning_rate": 0.0019994345305636807, + "loss": 0.2461, + "step": 2073 + }, + { + "epoch": 0.01800331594343799, + "grad_norm": 0.08349609375, + "learning_rate": 0.001999433475202073, + "loss": 0.2129, + "step": 2074 + }, + { + "epoch": 0.018011996423642155, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019994324188568583, + "loss": 0.1943, + "step": 2075 + }, + { + "epoch": 0.01802067690384632, + "grad_norm": 0.06591796875, + "learning_rate": 0.001999431361528038, + "loss": 0.2256, + "step": 2076 + }, + { + "epoch": 0.018029357384050485, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019994303032156127, + "loss": 0.2656, + "step": 2077 + }, + { + "epoch": 0.01803803786425465, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019994292439195847, + "loss": 0.2754, + "step": 2078 + }, + { + "epoch": 0.018046718344458815, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019994281836399536, + "loss": 0.1924, + "step": 2079 + }, + { + "epoch": 0.01805539882466298, + "grad_norm": 0.078125, + "learning_rate": 0.0019994271223767214, + "loss": 0.2393, + "step": 2080 + }, + { + "epoch": 0.018064079304867146, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019994260601298897, + "loss": 0.1875, + "step": 2081 + }, + { + "epoch": 0.01807275978507131, + "grad_norm": 0.06201171875, + "learning_rate": 0.001999424996899459, + "loss": 0.208, + "step": 2082 + }, + { + "epoch": 0.018081440265275476, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019994239326854304, + "loss": 0.2539, + "step": 2083 + }, + { + "epoch": 0.01809012074547964, + "grad_norm": 0.06640625, + "learning_rate": 0.0019994228674878054, + "loss": 0.1719, + "step": 2084 + }, + { + "epoch": 0.018098801225683806, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019994218013065852, + "loss": 0.1914, + "step": 2085 + }, + { + "epoch": 0.01810748170588797, + "grad_norm": 0.08349609375, + "learning_rate": 0.001999420734141771, + "loss": 0.2676, + "step": 2086 + }, + { + "epoch": 0.018116162186092133, + "grad_norm": 0.197265625, + "learning_rate": 0.0019994196659933634, + "loss": 0.2852, + "step": 2087 + }, + { + "epoch": 0.018124842666296298, + "grad_norm": 0.0830078125, + "learning_rate": 0.001999418596861364, + "loss": 0.2266, + "step": 2088 + }, + { + "epoch": 0.018133523146500463, + "grad_norm": 0.11962890625, + "learning_rate": 0.001999417526745774, + "loss": 0.2812, + "step": 2089 + }, + { + "epoch": 0.01814220362670463, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019994164556465946, + "loss": 0.2207, + "step": 2090 + }, + { + "epoch": 0.018150884106908793, + "grad_norm": 0.062255859375, + "learning_rate": 0.001999415383563827, + "loss": 0.2158, + "step": 2091 + }, + { + "epoch": 0.01815956458711296, + "grad_norm": 0.083984375, + "learning_rate": 0.001999414310497472, + "loss": 0.2061, + "step": 2092 + }, + { + "epoch": 0.018168245067317124, + "grad_norm": 0.06787109375, + "learning_rate": 0.001999413236447531, + "loss": 0.2012, + "step": 2093 + }, + { + "epoch": 0.01817692554752129, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019994121614140053, + "loss": 0.2363, + "step": 2094 + }, + { + "epoch": 0.018185606027725454, + "grad_norm": 0.0947265625, + "learning_rate": 0.001999411085396896, + "loss": 0.1875, + "step": 2095 + }, + { + "epoch": 0.01819428650792962, + "grad_norm": 0.060546875, + "learning_rate": 0.0019994100083962044, + "loss": 0.1924, + "step": 2096 + }, + { + "epoch": 0.018202966988133784, + "grad_norm": 0.055908203125, + "learning_rate": 0.001999408930411932, + "loss": 0.1816, + "step": 2097 + }, + { + "epoch": 0.01821164746833795, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019994078514440784, + "loss": 0.1846, + "step": 2098 + }, + { + "epoch": 0.018220327948542114, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019994067714926463, + "loss": 0.2422, + "step": 2099 + }, + { + "epoch": 0.01822900842874628, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019994056905576364, + "loss": 0.2461, + "step": 2100 + }, + { + "epoch": 0.018237688908950445, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019994046086390504, + "loss": 0.2158, + "step": 2101 + }, + { + "epoch": 0.01824636938915461, + "grad_norm": 0.0576171875, + "learning_rate": 0.0019994035257368883, + "loss": 0.2168, + "step": 2102 + }, + { + "epoch": 0.01825504986935877, + "grad_norm": 0.0859375, + "learning_rate": 0.001999402441851153, + "loss": 0.2402, + "step": 2103 + }, + { + "epoch": 0.018263730349562936, + "grad_norm": 0.06640625, + "learning_rate": 0.001999401356981844, + "loss": 0.209, + "step": 2104 + }, + { + "epoch": 0.0182724108297671, + "grad_norm": 0.059326171875, + "learning_rate": 0.0019994002711289636, + "loss": 0.1914, + "step": 2105 + }, + { + "epoch": 0.018281091309971267, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999399184292512, + "loss": 0.2285, + "step": 2106 + }, + { + "epoch": 0.018289771790175432, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019993980964724913, + "loss": 0.1816, + "step": 2107 + }, + { + "epoch": 0.018298452270379597, + "grad_norm": 0.078125, + "learning_rate": 0.0019993970076689024, + "loss": 0.2227, + "step": 2108 + }, + { + "epoch": 0.018307132750583762, + "grad_norm": 0.0947265625, + "learning_rate": 0.001999395917881746, + "loss": 0.1885, + "step": 2109 + }, + { + "epoch": 0.018315813230787927, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019993948271110245, + "loss": 0.2402, + "step": 2110 + }, + { + "epoch": 0.018324493710992092, + "grad_norm": 0.07421875, + "learning_rate": 0.001999393735356738, + "loss": 0.248, + "step": 2111 + }, + { + "epoch": 0.018333174191196257, + "grad_norm": 0.107421875, + "learning_rate": 0.001999392642618888, + "loss": 0.2305, + "step": 2112 + }, + { + "epoch": 0.018341854671400423, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019993915488974753, + "loss": 0.2266, + "step": 2113 + }, + { + "epoch": 0.018350535151604588, + "grad_norm": 0.1044921875, + "learning_rate": 0.001999390454192502, + "loss": 0.2402, + "step": 2114 + }, + { + "epoch": 0.018359215631808753, + "grad_norm": 0.07177734375, + "learning_rate": 0.001999389358503969, + "loss": 0.1973, + "step": 2115 + }, + { + "epoch": 0.018367896112012918, + "grad_norm": 0.0771484375, + "learning_rate": 0.001999388261831877, + "loss": 0.1836, + "step": 2116 + }, + { + "epoch": 0.018376576592217083, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019993871641762273, + "loss": 0.2305, + "step": 2117 + }, + { + "epoch": 0.018385257072421245, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019993860655370217, + "loss": 0.2441, + "step": 2118 + }, + { + "epoch": 0.01839393755262541, + "grad_norm": 0.09716796875, + "learning_rate": 0.001999384965914261, + "loss": 0.2344, + "step": 2119 + }, + { + "epoch": 0.018402618032829575, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999383865307946, + "loss": 0.2275, + "step": 2120 + }, + { + "epoch": 0.01841129851303374, + "grad_norm": 0.09228515625, + "learning_rate": 0.001999382763718079, + "loss": 0.208, + "step": 2121 + }, + { + "epoch": 0.018419978993237905, + "grad_norm": 0.0830078125, + "learning_rate": 0.00199938166114466, + "loss": 0.2256, + "step": 2122 + }, + { + "epoch": 0.01842865947344207, + "grad_norm": 0.072265625, + "learning_rate": 0.0019993805575876907, + "loss": 0.1816, + "step": 2123 + }, + { + "epoch": 0.018437339953646235, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999379453047172, + "loss": 0.25, + "step": 2124 + }, + { + "epoch": 0.0184460204338504, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019993783475231062, + "loss": 0.2266, + "step": 2125 + }, + { + "epoch": 0.018454700914054566, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019993772410154937, + "loss": 0.2285, + "step": 2126 + }, + { + "epoch": 0.01846338139425873, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019993761335243354, + "loss": 0.2344, + "step": 2127 + }, + { + "epoch": 0.018472061874462896, + "grad_norm": 0.130859375, + "learning_rate": 0.0019993750250496332, + "loss": 0.2539, + "step": 2128 + }, + { + "epoch": 0.01848074235466706, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999373915591388, + "loss": 0.2344, + "step": 2129 + }, + { + "epoch": 0.018489422834871226, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019993728051496003, + "loss": 0.2559, + "step": 2130 + }, + { + "epoch": 0.01849810331507539, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019993716937242727, + "loss": 0.2109, + "step": 2131 + }, + { + "epoch": 0.018506783795279556, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019993705813154054, + "loss": 0.1807, + "step": 2132 + }, + { + "epoch": 0.01851546427548372, + "grad_norm": 0.09375, + "learning_rate": 0.0019993694679230002, + "loss": 0.2383, + "step": 2133 + }, + { + "epoch": 0.018524144755687883, + "grad_norm": 0.10400390625, + "learning_rate": 0.001999368353547058, + "loss": 0.2441, + "step": 2134 + }, + { + "epoch": 0.018532825235892048, + "grad_norm": 0.080078125, + "learning_rate": 0.00199936723818758, + "loss": 0.1758, + "step": 2135 + }, + { + "epoch": 0.018541505716096213, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019993661218445677, + "loss": 0.207, + "step": 2136 + }, + { + "epoch": 0.01855018619630038, + "grad_norm": 0.138671875, + "learning_rate": 0.0019993650045180217, + "loss": 0.2871, + "step": 2137 + }, + { + "epoch": 0.018558866676504544, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999363886207944, + "loss": 0.2305, + "step": 2138 + }, + { + "epoch": 0.01856754715670871, + "grad_norm": 0.1943359375, + "learning_rate": 0.001999362766914335, + "loss": 0.3477, + "step": 2139 + }, + { + "epoch": 0.018576227636912874, + "grad_norm": 0.078125, + "learning_rate": 0.001999361646637197, + "loss": 0.1924, + "step": 2140 + }, + { + "epoch": 0.01858490811711704, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019993605253765304, + "loss": 0.2129, + "step": 2141 + }, + { + "epoch": 0.018593588597321204, + "grad_norm": 0.07421875, + "learning_rate": 0.0019993594031323366, + "loss": 0.2168, + "step": 2142 + }, + { + "epoch": 0.01860226907752537, + "grad_norm": 0.06640625, + "learning_rate": 0.001999358279904617, + "loss": 0.1826, + "step": 2143 + }, + { + "epoch": 0.018610949557729534, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019993571556933725, + "loss": 0.2129, + "step": 2144 + }, + { + "epoch": 0.0186196300379337, + "grad_norm": 0.08935546875, + "learning_rate": 0.001999356030498605, + "loss": 0.1885, + "step": 2145 + }, + { + "epoch": 0.018628310518137865, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019993549043203144, + "loss": 0.2441, + "step": 2146 + }, + { + "epoch": 0.01863699099834203, + "grad_norm": 0.0703125, + "learning_rate": 0.0019993537771585035, + "loss": 0.2402, + "step": 2147 + }, + { + "epoch": 0.018645671478546195, + "grad_norm": 0.072265625, + "learning_rate": 0.001999352649013172, + "loss": 0.1748, + "step": 2148 + }, + { + "epoch": 0.018654351958750356, + "grad_norm": 0.1064453125, + "learning_rate": 0.001999351519884323, + "loss": 0.2637, + "step": 2149 + }, + { + "epoch": 0.01866303243895452, + "grad_norm": 0.06298828125, + "learning_rate": 0.001999350389771956, + "loss": 0.1865, + "step": 2150 + }, + { + "epoch": 0.018671712919158687, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019993492586760733, + "loss": 0.2188, + "step": 2151 + }, + { + "epoch": 0.018680393399362852, + "grad_norm": 0.0625, + "learning_rate": 0.0019993481265966756, + "loss": 0.1885, + "step": 2152 + }, + { + "epoch": 0.018689073879567017, + "grad_norm": 0.34375, + "learning_rate": 0.0019993469935337643, + "loss": 0.3555, + "step": 2153 + }, + { + "epoch": 0.018697754359771182, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019993458594873407, + "loss": 0.2539, + "step": 2154 + }, + { + "epoch": 0.018706434839975347, + "grad_norm": 0.2099609375, + "learning_rate": 0.001999344724457406, + "loss": 0.2129, + "step": 2155 + }, + { + "epoch": 0.018715115320179512, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019993435884439613, + "loss": 0.2109, + "step": 2156 + }, + { + "epoch": 0.018723795800383677, + "grad_norm": 0.08984375, + "learning_rate": 0.001999342451447008, + "loss": 0.2285, + "step": 2157 + }, + { + "epoch": 0.018732476280587843, + "grad_norm": 0.12109375, + "learning_rate": 0.0019993413134665474, + "loss": 0.2695, + "step": 2158 + }, + { + "epoch": 0.018741156760792008, + "grad_norm": 0.130859375, + "learning_rate": 0.0019993401745025804, + "loss": 0.2168, + "step": 2159 + }, + { + "epoch": 0.018749837240996173, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019993390345551085, + "loss": 0.2383, + "step": 2160 + }, + { + "epoch": 0.018758517721200338, + "grad_norm": 0.091796875, + "learning_rate": 0.001999337893624133, + "loss": 0.2021, + "step": 2161 + }, + { + "epoch": 0.018767198201404503, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019993367517096555, + "loss": 0.2227, + "step": 2162 + }, + { + "epoch": 0.018775878681608668, + "grad_norm": 0.111328125, + "learning_rate": 0.0019993356088116765, + "loss": 0.2852, + "step": 2163 + }, + { + "epoch": 0.01878455916181283, + "grad_norm": 0.3359375, + "learning_rate": 0.001999334464930198, + "loss": 0.2988, + "step": 2164 + }, + { + "epoch": 0.018793239642016995, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019993333200652203, + "loss": 0.2207, + "step": 2165 + }, + { + "epoch": 0.01880192012222116, + "grad_norm": 0.11376953125, + "learning_rate": 0.001999332174216745, + "loss": 0.2773, + "step": 2166 + }, + { + "epoch": 0.018810600602425325, + "grad_norm": 0.068359375, + "learning_rate": 0.001999331027384774, + "loss": 0.2002, + "step": 2167 + }, + { + "epoch": 0.01881928108262949, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019993298795693082, + "loss": 0.2441, + "step": 2168 + }, + { + "epoch": 0.018827961562833655, + "grad_norm": 0.0859375, + "learning_rate": 0.0019993287307703486, + "loss": 0.2852, + "step": 2169 + }, + { + "epoch": 0.01883664204303782, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019993275809878966, + "loss": 0.2236, + "step": 2170 + }, + { + "epoch": 0.018845322523241986, + "grad_norm": 0.07177734375, + "learning_rate": 0.001999326430221953, + "loss": 0.1709, + "step": 2171 + }, + { + "epoch": 0.01885400300344615, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019993252784725204, + "loss": 0.2197, + "step": 2172 + }, + { + "epoch": 0.018862683483650316, + "grad_norm": 0.078125, + "learning_rate": 0.0019993241257395987, + "loss": 0.1836, + "step": 2173 + }, + { + "epoch": 0.01887136396385448, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019993229720231902, + "loss": 0.2344, + "step": 2174 + }, + { + "epoch": 0.018880044444058646, + "grad_norm": 0.12890625, + "learning_rate": 0.0019993218173232946, + "loss": 0.3477, + "step": 2175 + }, + { + "epoch": 0.01888872492426281, + "grad_norm": 0.058349609375, + "learning_rate": 0.001999320661639915, + "loss": 0.1797, + "step": 2176 + }, + { + "epoch": 0.018897405404466976, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999319504973051, + "loss": 0.2363, + "step": 2177 + }, + { + "epoch": 0.01890608588467114, + "grad_norm": 0.107421875, + "learning_rate": 0.0019993183473227057, + "loss": 0.2275, + "step": 2178 + }, + { + "epoch": 0.018914766364875307, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019993171886888788, + "loss": 0.2344, + "step": 2179 + }, + { + "epoch": 0.018923446845079468, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019993160290715725, + "loss": 0.2656, + "step": 2180 + }, + { + "epoch": 0.018932127325283633, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019993148684707873, + "loss": 0.2197, + "step": 2181 + }, + { + "epoch": 0.0189408078054878, + "grad_norm": 0.10107421875, + "learning_rate": 0.001999313706886525, + "loss": 0.1885, + "step": 2182 + }, + { + "epoch": 0.018949488285691964, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019993125443187867, + "loss": 0.2617, + "step": 2183 + }, + { + "epoch": 0.01895816876589613, + "grad_norm": 0.07958984375, + "learning_rate": 0.001999311380767574, + "loss": 0.2344, + "step": 2184 + }, + { + "epoch": 0.018966849246100294, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019993102162328877, + "loss": 0.2148, + "step": 2185 + }, + { + "epoch": 0.01897552972630446, + "grad_norm": 0.11474609375, + "learning_rate": 0.001999309050714729, + "loss": 0.2734, + "step": 2186 + }, + { + "epoch": 0.018984210206508624, + "grad_norm": 0.080078125, + "learning_rate": 0.0019993078842131, + "loss": 0.1836, + "step": 2187 + }, + { + "epoch": 0.01899289068671279, + "grad_norm": 0.08642578125, + "learning_rate": 0.001999306716728001, + "loss": 0.209, + "step": 2188 + }, + { + "epoch": 0.019001571166916954, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019993055482594338, + "loss": 0.249, + "step": 2189 + }, + { + "epoch": 0.01901025164712112, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019993043788073994, + "loss": 0.1807, + "step": 2190 + }, + { + "epoch": 0.019018932127325285, + "grad_norm": 0.06201171875, + "learning_rate": 0.001999303208371899, + "loss": 0.1836, + "step": 2191 + }, + { + "epoch": 0.01902761260752945, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019993020369529347, + "loss": 0.2734, + "step": 2192 + }, + { + "epoch": 0.019036293087733615, + "grad_norm": 0.06494140625, + "learning_rate": 0.001999300864550507, + "loss": 0.2031, + "step": 2193 + }, + { + "epoch": 0.01904497356793778, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019992996911646172, + "loss": 0.2305, + "step": 2194 + }, + { + "epoch": 0.01905365404814194, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019992985167952672, + "loss": 0.2266, + "step": 2195 + }, + { + "epoch": 0.019062334528346107, + "grad_norm": 0.08203125, + "learning_rate": 0.0019992973414424578, + "loss": 0.2227, + "step": 2196 + }, + { + "epoch": 0.019071015008550272, + "grad_norm": 0.0732421875, + "learning_rate": 0.00199929616510619, + "loss": 0.2363, + "step": 2197 + }, + { + "epoch": 0.019079695488754437, + "grad_norm": 0.103515625, + "learning_rate": 0.001999294987786466, + "loss": 0.291, + "step": 2198 + }, + { + "epoch": 0.019088375968958602, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019992938094832856, + "loss": 0.1934, + "step": 2199 + }, + { + "epoch": 0.019097056449162767, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019992926301966515, + "loss": 0.3066, + "step": 2200 + }, + { + "epoch": 0.019105736929366932, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019992914499265646, + "loss": 0.2695, + "step": 2201 + }, + { + "epoch": 0.019114417409571097, + "grad_norm": 0.0625, + "learning_rate": 0.001999290268673026, + "loss": 0.1787, + "step": 2202 + }, + { + "epoch": 0.019123097889775263, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019992890864360367, + "loss": 0.2461, + "step": 2203 + }, + { + "epoch": 0.019131778369979428, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999287903215599, + "loss": 0.2578, + "step": 2204 + }, + { + "epoch": 0.019140458850183593, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019992867190117133, + "loss": 0.2695, + "step": 2205 + }, + { + "epoch": 0.019149139330387758, + "grad_norm": 0.0869140625, + "learning_rate": 0.001999285533824381, + "loss": 0.2363, + "step": 2206 + }, + { + "epoch": 0.019157819810591923, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019992843476536034, + "loss": 0.1836, + "step": 2207 + }, + { + "epoch": 0.019166500290796088, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019992831604993826, + "loss": 0.1533, + "step": 2208 + }, + { + "epoch": 0.019175180771000253, + "grad_norm": 0.06494140625, + "learning_rate": 0.001999281972361719, + "loss": 0.1768, + "step": 2209 + }, + { + "epoch": 0.01918386125120442, + "grad_norm": 0.06787109375, + "learning_rate": 0.001999280783240614, + "loss": 0.2354, + "step": 2210 + }, + { + "epoch": 0.01919254173140858, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999279593136069, + "loss": 0.2168, + "step": 2211 + }, + { + "epoch": 0.019201222211612745, + "grad_norm": 0.1376953125, + "learning_rate": 0.001999278402048085, + "loss": 0.25, + "step": 2212 + }, + { + "epoch": 0.01920990269181691, + "grad_norm": 0.07275390625, + "learning_rate": 0.001999277209976664, + "loss": 0.207, + "step": 2213 + }, + { + "epoch": 0.019218583172021075, + "grad_norm": 0.07275390625, + "learning_rate": 0.001999276016921807, + "loss": 0.2305, + "step": 2214 + }, + { + "epoch": 0.01922726365222524, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019992748228835153, + "loss": 0.2383, + "step": 2215 + }, + { + "epoch": 0.019235944132429406, + "grad_norm": 0.10546875, + "learning_rate": 0.00199927362786179, + "loss": 0.3027, + "step": 2216 + }, + { + "epoch": 0.01924462461263357, + "grad_norm": 0.103515625, + "learning_rate": 0.0019992724318566328, + "loss": 0.2266, + "step": 2217 + }, + { + "epoch": 0.019253305092837736, + "grad_norm": 0.05859375, + "learning_rate": 0.0019992712348680447, + "loss": 0.1816, + "step": 2218 + }, + { + "epoch": 0.0192619855730419, + "grad_norm": 0.06787109375, + "learning_rate": 0.001999270036896027, + "loss": 0.2158, + "step": 2219 + }, + { + "epoch": 0.019270666053246066, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019992688379405813, + "loss": 0.2734, + "step": 2220 + }, + { + "epoch": 0.01927934653345023, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019992676380017086, + "loss": 0.1787, + "step": 2221 + }, + { + "epoch": 0.019288027013654396, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019992664370794104, + "loss": 0.2559, + "step": 2222 + }, + { + "epoch": 0.01929670749385856, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999265235173688, + "loss": 0.2441, + "step": 2223 + }, + { + "epoch": 0.019305387974062727, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019992640322845424, + "loss": 0.2432, + "step": 2224 + }, + { + "epoch": 0.01931406845426689, + "grad_norm": 0.076171875, + "learning_rate": 0.001999262828411975, + "loss": 0.1992, + "step": 2225 + }, + { + "epoch": 0.019322748934471053, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019992616235559877, + "loss": 0.2344, + "step": 2226 + }, + { + "epoch": 0.01933142941467522, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019992604177165815, + "loss": 0.2402, + "step": 2227 + }, + { + "epoch": 0.019340109894879384, + "grad_norm": 0.1455078125, + "learning_rate": 0.001999259210893757, + "loss": 0.2617, + "step": 2228 + }, + { + "epoch": 0.01934879037508355, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019992580030875166, + "loss": 0.1934, + "step": 2229 + }, + { + "epoch": 0.019357470855287714, + "grad_norm": 0.099609375, + "learning_rate": 0.001999256794297861, + "loss": 0.1885, + "step": 2230 + }, + { + "epoch": 0.01936615133549188, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999255584524792, + "loss": 0.2188, + "step": 2231 + }, + { + "epoch": 0.019374831815696044, + "grad_norm": 0.125, + "learning_rate": 0.0019992543737683104, + "loss": 0.2617, + "step": 2232 + }, + { + "epoch": 0.01938351229590021, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019992531620284175, + "loss": 0.1914, + "step": 2233 + }, + { + "epoch": 0.019392192776104374, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999251949305115, + "loss": 0.209, + "step": 2234 + }, + { + "epoch": 0.01940087325630854, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019992507355984044, + "loss": 0.209, + "step": 2235 + }, + { + "epoch": 0.019409553736512705, + "grad_norm": 0.11181640625, + "learning_rate": 0.001999249520908286, + "loss": 0.2656, + "step": 2236 + }, + { + "epoch": 0.01941823421671687, + "grad_norm": 0.515625, + "learning_rate": 0.0019992483052347623, + "loss": 0.625, + "step": 2237 + }, + { + "epoch": 0.019426914696921035, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019992470885778343, + "loss": 0.2559, + "step": 2238 + }, + { + "epoch": 0.0194355951771252, + "grad_norm": 0.0927734375, + "learning_rate": 0.001999245870937503, + "loss": 0.248, + "step": 2239 + }, + { + "epoch": 0.019444275657329365, + "grad_norm": 0.11083984375, + "learning_rate": 0.00199924465231377, + "loss": 0.2148, + "step": 2240 + }, + { + "epoch": 0.019452956137533527, + "grad_norm": 0.08984375, + "learning_rate": 0.0019992434327066365, + "loss": 0.207, + "step": 2241 + }, + { + "epoch": 0.019461636617737692, + "grad_norm": 0.1318359375, + "learning_rate": 0.001999242212116104, + "loss": 0.3105, + "step": 2242 + }, + { + "epoch": 0.019470317097941857, + "grad_norm": 0.080078125, + "learning_rate": 0.0019992409905421732, + "loss": 0.2178, + "step": 2243 + }, + { + "epoch": 0.019478997578146022, + "grad_norm": 0.083984375, + "learning_rate": 0.0019992397679848466, + "loss": 0.2598, + "step": 2244 + }, + { + "epoch": 0.019487678058350187, + "grad_norm": 0.083984375, + "learning_rate": 0.0019992385444441247, + "loss": 0.2109, + "step": 2245 + }, + { + "epoch": 0.019496358538554352, + "grad_norm": 0.1044921875, + "learning_rate": 0.001999237319920009, + "loss": 0.2383, + "step": 2246 + }, + { + "epoch": 0.019505039018758517, + "grad_norm": 0.07275390625, + "learning_rate": 0.001999236094412501, + "loss": 0.2383, + "step": 2247 + }, + { + "epoch": 0.019513719498962682, + "grad_norm": 0.06640625, + "learning_rate": 0.0019992348679216017, + "loss": 0.207, + "step": 2248 + }, + { + "epoch": 0.019522399979166848, + "grad_norm": 0.125, + "learning_rate": 0.001999233640447313, + "loss": 0.2656, + "step": 2249 + }, + { + "epoch": 0.019531080459371013, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019992324119896356, + "loss": 0.207, + "step": 2250 + }, + { + "epoch": 0.019539760939575178, + "grad_norm": 0.07666015625, + "learning_rate": 0.001999231182548571, + "loss": 0.2188, + "step": 2251 + }, + { + "epoch": 0.019548441419779343, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999229952124121, + "loss": 0.1982, + "step": 2252 + }, + { + "epoch": 0.019557121899983508, + "grad_norm": 0.1015625, + "learning_rate": 0.0019992287207162866, + "loss": 0.2812, + "step": 2253 + }, + { + "epoch": 0.019565802380187673, + "grad_norm": 0.1064453125, + "learning_rate": 0.001999227488325069, + "loss": 0.2598, + "step": 2254 + }, + { + "epoch": 0.01957448286039184, + "grad_norm": 0.1767578125, + "learning_rate": 0.00199922625495047, + "loss": 0.2266, + "step": 2255 + }, + { + "epoch": 0.019583163340596003, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019992250205924903, + "loss": 0.248, + "step": 2256 + }, + { + "epoch": 0.019591843820800165, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999223785251132, + "loss": 0.1768, + "step": 2257 + }, + { + "epoch": 0.01960052430100433, + "grad_norm": 0.21875, + "learning_rate": 0.001999222548926396, + "loss": 0.3125, + "step": 2258 + }, + { + "epoch": 0.019609204781208495, + "grad_norm": 0.083984375, + "learning_rate": 0.0019992213116182835, + "loss": 0.2324, + "step": 2259 + }, + { + "epoch": 0.01961788526141266, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019992200733267964, + "loss": 0.248, + "step": 2260 + }, + { + "epoch": 0.019626565741616826, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019992188340519357, + "loss": 0.2402, + "step": 2261 + }, + { + "epoch": 0.01963524622182099, + "grad_norm": 0.061767578125, + "learning_rate": 0.0019992175937937023, + "loss": 0.1953, + "step": 2262 + }, + { + "epoch": 0.019643926702025156, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019992163525520985, + "loss": 0.2158, + "step": 2263 + }, + { + "epoch": 0.01965260718222932, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019992151103271254, + "loss": 0.2285, + "step": 2264 + }, + { + "epoch": 0.019661287662433486, + "grad_norm": 0.091796875, + "learning_rate": 0.0019992138671187836, + "loss": 0.249, + "step": 2265 + }, + { + "epoch": 0.01966996814263765, + "grad_norm": 0.7109375, + "learning_rate": 0.0019992126229270756, + "loss": 0.5078, + "step": 2266 + }, + { + "epoch": 0.019678648622841816, + "grad_norm": 0.08203125, + "learning_rate": 0.001999211377752002, + "loss": 0.2207, + "step": 2267 + }, + { + "epoch": 0.01968732910304598, + "grad_norm": 0.10205078125, + "learning_rate": 0.001999210131593564, + "loss": 0.2002, + "step": 2268 + }, + { + "epoch": 0.019696009583250147, + "grad_norm": 0.150390625, + "learning_rate": 0.0019992088844517637, + "loss": 0.2061, + "step": 2269 + }, + { + "epoch": 0.01970469006345431, + "grad_norm": 0.142578125, + "learning_rate": 0.001999207636326602, + "loss": 0.2207, + "step": 2270 + }, + { + "epoch": 0.019713370543658477, + "grad_norm": 0.0927734375, + "learning_rate": 0.00199920638721808, + "loss": 0.2227, + "step": 2271 + }, + { + "epoch": 0.01972205102386264, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019992051371262, + "loss": 0.2539, + "step": 2272 + }, + { + "epoch": 0.019730731504066804, + "grad_norm": 0.125, + "learning_rate": 0.0019992038860509624, + "loss": 0.2539, + "step": 2273 + }, + { + "epoch": 0.01973941198427097, + "grad_norm": 0.051025390625, + "learning_rate": 0.001999202633992369, + "loss": 0.1719, + "step": 2274 + }, + { + "epoch": 0.019748092464475134, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019992013809504213, + "loss": 0.2051, + "step": 2275 + }, + { + "epoch": 0.0197567729446793, + "grad_norm": 0.0859375, + "learning_rate": 0.00199920012692512, + "loss": 0.2393, + "step": 2276 + }, + { + "epoch": 0.019765453424883464, + "grad_norm": 0.0732421875, + "learning_rate": 0.001999198871916467, + "loss": 0.2148, + "step": 2277 + }, + { + "epoch": 0.01977413390508763, + "grad_norm": 0.080078125, + "learning_rate": 0.001999197615924464, + "loss": 0.2236, + "step": 2278 + }, + { + "epoch": 0.019782814385291794, + "grad_norm": 0.083984375, + "learning_rate": 0.001999196358949112, + "loss": 0.2012, + "step": 2279 + }, + { + "epoch": 0.01979149486549596, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019991951009904123, + "loss": 0.2383, + "step": 2280 + }, + { + "epoch": 0.019800175345700124, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019991938420483666, + "loss": 0.2207, + "step": 2281 + }, + { + "epoch": 0.01980885582590429, + "grad_norm": 0.0966796875, + "learning_rate": 0.001999192582122975, + "loss": 0.2461, + "step": 2282 + }, + { + "epoch": 0.019817536306108455, + "grad_norm": 0.056640625, + "learning_rate": 0.0019991913212142414, + "loss": 0.1719, + "step": 2283 + }, + { + "epoch": 0.01982621678631262, + "grad_norm": 0.080078125, + "learning_rate": 0.001999190059322165, + "loss": 0.2109, + "step": 2284 + }, + { + "epoch": 0.019834897266516785, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019991887964467474, + "loss": 0.6211, + "step": 2285 + }, + { + "epoch": 0.01984357774672095, + "grad_norm": 0.08984375, + "learning_rate": 0.001999187532587991, + "loss": 0.2656, + "step": 2286 + }, + { + "epoch": 0.019852258226925115, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019991862677458962, + "loss": 0.2461, + "step": 2287 + }, + { + "epoch": 0.019860938707129277, + "grad_norm": 0.06640625, + "learning_rate": 0.001999185001920465, + "loss": 0.1777, + "step": 2288 + }, + { + "epoch": 0.019869619187333442, + "grad_norm": 0.09228515625, + "learning_rate": 0.001999183735111699, + "loss": 0.3047, + "step": 2289 + }, + { + "epoch": 0.019878299667537607, + "grad_norm": 0.087890625, + "learning_rate": 0.0019991824673195983, + "loss": 0.2637, + "step": 2290 + }, + { + "epoch": 0.019886980147741772, + "grad_norm": 0.1015625, + "learning_rate": 0.0019991811985441655, + "loss": 0.2539, + "step": 2291 + }, + { + "epoch": 0.019895660627945937, + "grad_norm": 0.054931640625, + "learning_rate": 0.001999179928785402, + "loss": 0.1865, + "step": 2292 + }, + { + "epoch": 0.019904341108150102, + "grad_norm": 0.07373046875, + "learning_rate": 0.001999178658043309, + "loss": 0.2363, + "step": 2293 + }, + { + "epoch": 0.019913021588354268, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019991773863178874, + "loss": 0.2871, + "step": 2294 + }, + { + "epoch": 0.019921702068558433, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019991761136091387, + "loss": 0.2344, + "step": 2295 + }, + { + "epoch": 0.019930382548762598, + "grad_norm": 0.08642578125, + "learning_rate": 0.001999174839917065, + "loss": 0.2422, + "step": 2296 + }, + { + "epoch": 0.019939063028966763, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019991735652416668, + "loss": 0.2324, + "step": 2297 + }, + { + "epoch": 0.019947743509170928, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999172289582946, + "loss": 0.25, + "step": 2298 + }, + { + "epoch": 0.019956423989375093, + "grad_norm": 0.078125, + "learning_rate": 0.001999171012940904, + "loss": 0.2422, + "step": 2299 + }, + { + "epoch": 0.01996510446957926, + "grad_norm": 0.080078125, + "learning_rate": 0.001999169735315542, + "loss": 0.1914, + "step": 2300 + }, + { + "epoch": 0.019973784949783423, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019991684567068615, + "loss": 0.252, + "step": 2301 + }, + { + "epoch": 0.01998246542998759, + "grad_norm": 0.0712890625, + "learning_rate": 0.001999167177114864, + "loss": 0.1748, + "step": 2302 + }, + { + "epoch": 0.01999114591019175, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019991658965395503, + "loss": 0.2441, + "step": 2303 + }, + { + "epoch": 0.019999826390395915, + "grad_norm": 0.083984375, + "learning_rate": 0.001999164614980923, + "loss": 0.25, + "step": 2304 + }, + { + "epoch": 0.02000850687060008, + "grad_norm": 0.080078125, + "learning_rate": 0.001999163332438982, + "loss": 0.1992, + "step": 2305 + }, + { + "epoch": 0.020017187350804246, + "grad_norm": 0.08544921875, + "learning_rate": 0.00199916204891373, + "loss": 0.2949, + "step": 2306 + }, + { + "epoch": 0.02002586783100841, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019991607644051683, + "loss": 0.1953, + "step": 2307 + }, + { + "epoch": 0.020034548311212576, + "grad_norm": 0.189453125, + "learning_rate": 0.0019991594789132974, + "loss": 0.3125, + "step": 2308 + }, + { + "epoch": 0.02004322879141674, + "grad_norm": 0.080078125, + "learning_rate": 0.0019991581924381193, + "loss": 0.1992, + "step": 2309 + }, + { + "epoch": 0.020051909271620906, + "grad_norm": 0.08349609375, + "learning_rate": 0.001999156904979635, + "loss": 0.2217, + "step": 2310 + }, + { + "epoch": 0.02006058975182507, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019991556165378462, + "loss": 0.2041, + "step": 2311 + }, + { + "epoch": 0.020069270232029236, + "grad_norm": 0.0634765625, + "learning_rate": 0.0019991543271127552, + "loss": 0.2021, + "step": 2312 + }, + { + "epoch": 0.0200779507122334, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019991530367043614, + "loss": 0.2109, + "step": 2313 + }, + { + "epoch": 0.020086631192437567, + "grad_norm": 0.08251953125, + "learning_rate": 0.001999151745312668, + "loss": 0.2422, + "step": 2314 + }, + { + "epoch": 0.02009531167264173, + "grad_norm": 0.068359375, + "learning_rate": 0.0019991504529376755, + "loss": 0.1924, + "step": 2315 + }, + { + "epoch": 0.020103992152845897, + "grad_norm": 0.076171875, + "learning_rate": 0.0019991491595793857, + "loss": 0.2559, + "step": 2316 + }, + { + "epoch": 0.020112672633050062, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019991478652378, + "loss": 0.2119, + "step": 2317 + }, + { + "epoch": 0.020121353113254224, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019991465699129196, + "loss": 0.2344, + "step": 2318 + }, + { + "epoch": 0.02013003359345839, + "grad_norm": 0.1083984375, + "learning_rate": 0.001999145273604746, + "loss": 0.2109, + "step": 2319 + }, + { + "epoch": 0.020138714073662554, + "grad_norm": 0.06689453125, + "learning_rate": 0.0019991439763132808, + "loss": 0.1826, + "step": 2320 + }, + { + "epoch": 0.02014739455386672, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999142678038525, + "loss": 0.2305, + "step": 2321 + }, + { + "epoch": 0.020156075034070884, + "grad_norm": 0.08984375, + "learning_rate": 0.001999141378780481, + "loss": 0.209, + "step": 2322 + }, + { + "epoch": 0.02016475551427505, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019991400785391487, + "loss": 0.2598, + "step": 2323 + }, + { + "epoch": 0.020173435994479214, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019991387773145306, + "loss": 0.2148, + "step": 2324 + }, + { + "epoch": 0.02018211647468338, + "grad_norm": 0.07763671875, + "learning_rate": 0.001999137475106627, + "loss": 0.2188, + "step": 2325 + }, + { + "epoch": 0.020190796954887544, + "grad_norm": 0.08203125, + "learning_rate": 0.0019991361719154414, + "loss": 0.2461, + "step": 2326 + }, + { + "epoch": 0.02019947743509171, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019991348677409733, + "loss": 0.1836, + "step": 2327 + }, + { + "epoch": 0.020208157915295875, + "grad_norm": 0.12109375, + "learning_rate": 0.0019991335625832254, + "loss": 0.252, + "step": 2328 + }, + { + "epoch": 0.02021683839550004, + "grad_norm": 0.052490234375, + "learning_rate": 0.001999132256442198, + "loss": 0.1572, + "step": 2329 + }, + { + "epoch": 0.020225518875704205, + "grad_norm": 0.09228515625, + "learning_rate": 0.001999130949317893, + "loss": 0.2188, + "step": 2330 + }, + { + "epoch": 0.02023419935590837, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019991296412103124, + "loss": 0.2363, + "step": 2331 + }, + { + "epoch": 0.020242879836112535, + "grad_norm": 0.0703125, + "learning_rate": 0.0019991283321194567, + "loss": 0.2119, + "step": 2332 + }, + { + "epoch": 0.0202515603163167, + "grad_norm": 0.0751953125, + "learning_rate": 0.001999127022045328, + "loss": 0.1855, + "step": 2333 + }, + { + "epoch": 0.020260240796520862, + "grad_norm": 0.0693359375, + "learning_rate": 0.001999125710987928, + "loss": 0.2188, + "step": 2334 + }, + { + "epoch": 0.020268921276725027, + "grad_norm": 0.099609375, + "learning_rate": 0.0019991243989472564, + "loss": 0.2559, + "step": 2335 + }, + { + "epoch": 0.020277601756929192, + "grad_norm": 0.068359375, + "learning_rate": 0.001999123085923317, + "loss": 0.1797, + "step": 2336 + }, + { + "epoch": 0.020286282237133357, + "grad_norm": 0.09814453125, + "learning_rate": 0.001999121771916109, + "loss": 0.2402, + "step": 2337 + }, + { + "epoch": 0.020294962717337522, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999120456925636, + "loss": 0.2617, + "step": 2338 + }, + { + "epoch": 0.020303643197541688, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999119140951898, + "loss": 0.252, + "step": 2339 + }, + { + "epoch": 0.020312323677745853, + "grad_norm": 0.09521484375, + "learning_rate": 0.001999117823994897, + "loss": 0.2207, + "step": 2340 + }, + { + "epoch": 0.020321004157950018, + "grad_norm": 0.0859375, + "learning_rate": 0.0019991165060546335, + "loss": 0.2539, + "step": 2341 + }, + { + "epoch": 0.020329684638154183, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019991151871311106, + "loss": 0.1953, + "step": 2342 + }, + { + "epoch": 0.020338365118358348, + "grad_norm": 0.060546875, + "learning_rate": 0.0019991138672243282, + "loss": 0.165, + "step": 2343 + }, + { + "epoch": 0.020347045598562513, + "grad_norm": 0.08447265625, + "learning_rate": 0.001999112546334289, + "loss": 0.2002, + "step": 2344 + }, + { + "epoch": 0.02035572607876668, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019991112244609932, + "loss": 0.252, + "step": 2345 + }, + { + "epoch": 0.020364406558970843, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019991099016044432, + "loss": 0.2441, + "step": 2346 + }, + { + "epoch": 0.02037308703917501, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019991085777646403, + "loss": 0.2188, + "step": 2347 + }, + { + "epoch": 0.020381767519379174, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019991072529415856, + "loss": 0.2236, + "step": 2348 + }, + { + "epoch": 0.020390447999583335, + "grad_norm": 0.09765625, + "learning_rate": 0.0019991059271352806, + "loss": 0.3047, + "step": 2349 + }, + { + "epoch": 0.0203991284797875, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999104600345727, + "loss": 0.248, + "step": 2350 + }, + { + "epoch": 0.020407808959991666, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999103272572926, + "loss": 0.2344, + "step": 2351 + }, + { + "epoch": 0.02041648944019583, + "grad_norm": 0.0634765625, + "learning_rate": 0.0019991019438168793, + "loss": 0.1533, + "step": 2352 + }, + { + "epoch": 0.020425169920399996, + "grad_norm": 0.1015625, + "learning_rate": 0.001999100614077588, + "loss": 0.252, + "step": 2353 + }, + { + "epoch": 0.02043385040060416, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019990992833550537, + "loss": 0.2246, + "step": 2354 + }, + { + "epoch": 0.020442530880808326, + "grad_norm": 0.1416015625, + "learning_rate": 0.001999097951649278, + "loss": 0.2539, + "step": 2355 + }, + { + "epoch": 0.02045121136101249, + "grad_norm": 0.06689453125, + "learning_rate": 0.001999096618960263, + "loss": 0.1904, + "step": 2356 + }, + { + "epoch": 0.020459891841216656, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019990952852880087, + "loss": 0.2373, + "step": 2357 + }, + { + "epoch": 0.02046857232142082, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019990939506325175, + "loss": 0.2617, + "step": 2358 + }, + { + "epoch": 0.020477252801624986, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019990926149937908, + "loss": 0.1689, + "step": 2359 + }, + { + "epoch": 0.02048593328182915, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019990912783718296, + "loss": 0.2246, + "step": 2360 + }, + { + "epoch": 0.020494613762033317, + "grad_norm": 0.07568359375, + "learning_rate": 0.001999089940766636, + "loss": 0.2129, + "step": 2361 + }, + { + "epoch": 0.020503294242237482, + "grad_norm": 0.06298828125, + "learning_rate": 0.001999088602178211, + "loss": 0.1787, + "step": 2362 + }, + { + "epoch": 0.020511974722441647, + "grad_norm": 0.095703125, + "learning_rate": 0.001999087262606556, + "loss": 0.2598, + "step": 2363 + }, + { + "epoch": 0.020520655202645812, + "grad_norm": 0.057373046875, + "learning_rate": 0.001999085922051673, + "loss": 0.1963, + "step": 2364 + }, + { + "epoch": 0.020529335682849974, + "grad_norm": 0.06103515625, + "learning_rate": 0.0019990845805135634, + "loss": 0.2051, + "step": 2365 + }, + { + "epoch": 0.02053801616305414, + "grad_norm": 0.08984375, + "learning_rate": 0.001999083237992228, + "loss": 0.25, + "step": 2366 + }, + { + "epoch": 0.020546696643258304, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019990818944876686, + "loss": 0.2402, + "step": 2367 + }, + { + "epoch": 0.02055537712346247, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019990805499998866, + "loss": 0.2227, + "step": 2368 + }, + { + "epoch": 0.020564057603666634, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019990792045288842, + "loss": 0.2383, + "step": 2369 + }, + { + "epoch": 0.0205727380838708, + "grad_norm": 0.072265625, + "learning_rate": 0.001999077858074662, + "loss": 0.2148, + "step": 2370 + }, + { + "epoch": 0.020581418564074964, + "grad_norm": 0.078125, + "learning_rate": 0.001999076510637222, + "loss": 0.2266, + "step": 2371 + }, + { + "epoch": 0.02059009904427913, + "grad_norm": 0.1005859375, + "learning_rate": 0.001999075162216565, + "loss": 0.2598, + "step": 2372 + }, + { + "epoch": 0.020598779524483295, + "grad_norm": 0.083984375, + "learning_rate": 0.001999073812812693, + "loss": 0.2148, + "step": 2373 + }, + { + "epoch": 0.02060746000468746, + "grad_norm": 0.09033203125, + "learning_rate": 0.001999072462425608, + "loss": 0.2422, + "step": 2374 + }, + { + "epoch": 0.020616140484891625, + "grad_norm": 0.11083984375, + "learning_rate": 0.00199907111105531, + "loss": 0.2402, + "step": 2375 + }, + { + "epoch": 0.02062482096509579, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019990697587018015, + "loss": 0.2285, + "step": 2376 + }, + { + "epoch": 0.020633501445299955, + "grad_norm": 0.080078125, + "learning_rate": 0.001999068405365084, + "loss": 0.2041, + "step": 2377 + }, + { + "epoch": 0.02064218192550412, + "grad_norm": 0.09619140625, + "learning_rate": 0.001999067051045159, + "loss": 0.2227, + "step": 2378 + }, + { + "epoch": 0.020650862405708285, + "grad_norm": 0.06689453125, + "learning_rate": 0.0019990656957420277, + "loss": 0.1934, + "step": 2379 + }, + { + "epoch": 0.020659542885912447, + "grad_norm": 0.08056640625, + "learning_rate": 0.001999064339455692, + "loss": 0.2324, + "step": 2380 + }, + { + "epoch": 0.020668223366116612, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019990629821861525, + "loss": 0.2061, + "step": 2381 + }, + { + "epoch": 0.020676903846320777, + "grad_norm": 0.0576171875, + "learning_rate": 0.0019990616239334113, + "loss": 0.1826, + "step": 2382 + }, + { + "epoch": 0.020685584326524942, + "grad_norm": 0.083984375, + "learning_rate": 0.0019990602646974697, + "loss": 0.2363, + "step": 2383 + }, + { + "epoch": 0.020694264806729108, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019990589044783296, + "loss": 0.2305, + "step": 2384 + }, + { + "epoch": 0.020702945286933273, + "grad_norm": 0.099609375, + "learning_rate": 0.001999057543275992, + "loss": 0.2617, + "step": 2385 + }, + { + "epoch": 0.020711625767137438, + "grad_norm": 0.0703125, + "learning_rate": 0.0019990561810904585, + "loss": 0.21, + "step": 2386 + }, + { + "epoch": 0.020720306247341603, + "grad_norm": 0.08642578125, + "learning_rate": 0.001999054817921731, + "loss": 0.2354, + "step": 2387 + }, + { + "epoch": 0.020728986727545768, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019990534537698106, + "loss": 0.2393, + "step": 2388 + }, + { + "epoch": 0.020737667207749933, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019990520886346984, + "loss": 0.2344, + "step": 2389 + }, + { + "epoch": 0.020746347687954098, + "grad_norm": 0.09130859375, + "learning_rate": 0.001999050722516397, + "loss": 0.2266, + "step": 2390 + }, + { + "epoch": 0.020755028168158263, + "grad_norm": 0.08642578125, + "learning_rate": 0.001999049355414907, + "loss": 0.2617, + "step": 2391 + }, + { + "epoch": 0.02076370864836243, + "grad_norm": 0.09375, + "learning_rate": 0.00199904798733023, + "loss": 0.2197, + "step": 2392 + }, + { + "epoch": 0.020772389128566594, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019990466182623675, + "loss": 0.2539, + "step": 2393 + }, + { + "epoch": 0.02078106960877076, + "grad_norm": 0.142578125, + "learning_rate": 0.0019990452482113214, + "loss": 0.3516, + "step": 2394 + }, + { + "epoch": 0.02078975008897492, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019990438771770925, + "loss": 0.2188, + "step": 2395 + }, + { + "epoch": 0.020798430569179086, + "grad_norm": 0.10302734375, + "learning_rate": 0.001999042505159683, + "loss": 0.2656, + "step": 2396 + }, + { + "epoch": 0.02080711104938325, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019990411321590944, + "loss": 0.25, + "step": 2397 + }, + { + "epoch": 0.020815791529587416, + "grad_norm": 0.0791015625, + "learning_rate": 0.001999039758175328, + "loss": 0.1895, + "step": 2398 + }, + { + "epoch": 0.02082447200979158, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019990383832083846, + "loss": 0.2148, + "step": 2399 + }, + { + "epoch": 0.020833152489995746, + "grad_norm": 0.455078125, + "learning_rate": 0.001999037007258267, + "loss": 0.3008, + "step": 2400 + }, + { + "epoch": 0.02084183297019991, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019990356303249755, + "loss": 0.1992, + "step": 2401 + }, + { + "epoch": 0.020850513450404076, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019990342524085123, + "loss": 0.2246, + "step": 2402 + }, + { + "epoch": 0.02085919393060824, + "grad_norm": 0.08154296875, + "learning_rate": 0.001999032873508879, + "loss": 0.1934, + "step": 2403 + }, + { + "epoch": 0.020867874410812406, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019990314936260767, + "loss": 0.5469, + "step": 2404 + }, + { + "epoch": 0.02087655489101657, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019990301127601065, + "loss": 0.2227, + "step": 2405 + }, + { + "epoch": 0.020885235371220737, + "grad_norm": 0.0810546875, + "learning_rate": 0.001999028730910971, + "loss": 0.208, + "step": 2406 + }, + { + "epoch": 0.020893915851424902, + "grad_norm": 0.095703125, + "learning_rate": 0.0019990273480786714, + "loss": 0.2344, + "step": 2407 + }, + { + "epoch": 0.020902596331629067, + "grad_norm": 0.072265625, + "learning_rate": 0.0019990259642632085, + "loss": 0.1797, + "step": 2408 + }, + { + "epoch": 0.020911276811833232, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019990245794645847, + "loss": 0.2227, + "step": 2409 + }, + { + "epoch": 0.020919957292037397, + "grad_norm": 0.058349609375, + "learning_rate": 0.001999023193682801, + "loss": 0.1953, + "step": 2410 + }, + { + "epoch": 0.02092863777224156, + "grad_norm": 0.09326171875, + "learning_rate": 0.001999021806917859, + "loss": 0.1855, + "step": 2411 + }, + { + "epoch": 0.020937318252445724, + "grad_norm": 0.064453125, + "learning_rate": 0.0019990204191697603, + "loss": 0.209, + "step": 2412 + }, + { + "epoch": 0.02094599873264989, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019990190304385066, + "loss": 0.2246, + "step": 2413 + }, + { + "epoch": 0.020954679212854054, + "grad_norm": 0.0908203125, + "learning_rate": 0.001999017640724099, + "loss": 0.2695, + "step": 2414 + }, + { + "epoch": 0.02096335969305822, + "grad_norm": 0.0751953125, + "learning_rate": 0.001999016250026539, + "loss": 0.2148, + "step": 2415 + }, + { + "epoch": 0.020972040173262384, + "grad_norm": 0.09716796875, + "learning_rate": 0.001999014858345829, + "loss": 0.291, + "step": 2416 + }, + { + "epoch": 0.02098072065346655, + "grad_norm": 0.10009765625, + "learning_rate": 0.001999013465681969, + "loss": 0.248, + "step": 2417 + }, + { + "epoch": 0.020989401133670715, + "grad_norm": 0.0849609375, + "learning_rate": 0.001999012072034962, + "loss": 0.2344, + "step": 2418 + }, + { + "epoch": 0.02099808161387488, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019990106774048087, + "loss": 0.1865, + "step": 2419 + }, + { + "epoch": 0.021006762094079045, + "grad_norm": 0.064453125, + "learning_rate": 0.0019990092817915106, + "loss": 0.2285, + "step": 2420 + }, + { + "epoch": 0.02101544257428321, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019990078851950697, + "loss": 0.207, + "step": 2421 + }, + { + "epoch": 0.021024123054487375, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019990064876154873, + "loss": 0.2168, + "step": 2422 + }, + { + "epoch": 0.02103280353469154, + "grad_norm": 0.08740234375, + "learning_rate": 0.001999005089052765, + "loss": 0.2773, + "step": 2423 + }, + { + "epoch": 0.021041484014895705, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019990036895069043, + "loss": 0.2305, + "step": 2424 + }, + { + "epoch": 0.02105016449509987, + "grad_norm": 0.056640625, + "learning_rate": 0.0019990022889779064, + "loss": 0.1992, + "step": 2425 + }, + { + "epoch": 0.021058844975304032, + "grad_norm": 0.078125, + "learning_rate": 0.001999000887465773, + "loss": 0.2676, + "step": 2426 + }, + { + "epoch": 0.021067525455508197, + "grad_norm": 0.08544921875, + "learning_rate": 0.001998999484970506, + "loss": 0.252, + "step": 2427 + }, + { + "epoch": 0.021076205935712362, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019989980814921066, + "loss": 0.1914, + "step": 2428 + }, + { + "epoch": 0.021084886415916528, + "grad_norm": 0.0478515625, + "learning_rate": 0.0019989966770305764, + "loss": 0.168, + "step": 2429 + }, + { + "epoch": 0.021093566896120693, + "grad_norm": 0.2080078125, + "learning_rate": 0.001998995271585917, + "loss": 0.3555, + "step": 2430 + }, + { + "epoch": 0.021102247376324858, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019989938651581297, + "loss": 0.2266, + "step": 2431 + }, + { + "epoch": 0.021110927856529023, + "grad_norm": 0.06787109375, + "learning_rate": 0.001998992457747216, + "loss": 0.2041, + "step": 2432 + }, + { + "epoch": 0.021119608336733188, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019989910493531785, + "loss": 0.1973, + "step": 2433 + }, + { + "epoch": 0.021128288816937353, + "grad_norm": 0.171875, + "learning_rate": 0.001998989639976017, + "loss": 0.3125, + "step": 2434 + }, + { + "epoch": 0.021136969297141518, + "grad_norm": 0.083984375, + "learning_rate": 0.0019989882296157346, + "loss": 0.2207, + "step": 2435 + }, + { + "epoch": 0.021145649777345683, + "grad_norm": 0.06201171875, + "learning_rate": 0.0019989868182723317, + "loss": 0.1973, + "step": 2436 + }, + { + "epoch": 0.02115433025754985, + "grad_norm": 0.1591796875, + "learning_rate": 0.001998985405945811, + "loss": 0.2695, + "step": 2437 + }, + { + "epoch": 0.021163010737754014, + "grad_norm": 0.060546875, + "learning_rate": 0.0019989839926361727, + "loss": 0.1904, + "step": 2438 + }, + { + "epoch": 0.02117169121795818, + "grad_norm": 0.08740234375, + "learning_rate": 0.001998982578343419, + "loss": 0.2012, + "step": 2439 + }, + { + "epoch": 0.021180371698162344, + "grad_norm": 0.0751953125, + "learning_rate": 0.001998981163067552, + "loss": 0.2188, + "step": 2440 + }, + { + "epoch": 0.02118905217836651, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019989797468085726, + "loss": 0.2441, + "step": 2441 + }, + { + "epoch": 0.02119773265857067, + "grad_norm": 0.0791015625, + "learning_rate": 0.001998978329566482, + "loss": 0.2324, + "step": 2442 + }, + { + "epoch": 0.021206413138774836, + "grad_norm": 0.0732421875, + "learning_rate": 0.001998976911341282, + "loss": 0.2285, + "step": 2443 + }, + { + "epoch": 0.021215093618979, + "grad_norm": 0.061279296875, + "learning_rate": 0.001998975492132975, + "loss": 0.208, + "step": 2444 + }, + { + "epoch": 0.021223774099183166, + "grad_norm": 0.0693359375, + "learning_rate": 0.001998974071941562, + "loss": 0.1875, + "step": 2445 + }, + { + "epoch": 0.02123245457938733, + "grad_norm": 0.078125, + "learning_rate": 0.001998972650767044, + "loss": 0.2266, + "step": 2446 + }, + { + "epoch": 0.021241135059591496, + "grad_norm": 0.09716796875, + "learning_rate": 0.001998971228609423, + "loss": 0.2617, + "step": 2447 + }, + { + "epoch": 0.02124981553979566, + "grad_norm": 0.07958984375, + "learning_rate": 0.001998969805468701, + "loss": 0.2012, + "step": 2448 + }, + { + "epoch": 0.021258496019999826, + "grad_norm": 0.0703125, + "learning_rate": 0.0019989683813448787, + "loss": 0.2373, + "step": 2449 + }, + { + "epoch": 0.02126717650020399, + "grad_norm": 0.08056640625, + "learning_rate": 0.001998966956237958, + "loss": 0.2637, + "step": 2450 + }, + { + "epoch": 0.021275856980408157, + "grad_norm": 0.125, + "learning_rate": 0.0019989655301479413, + "loss": 0.3086, + "step": 2451 + }, + { + "epoch": 0.021284537460612322, + "grad_norm": 0.07861328125, + "learning_rate": 0.001998964103074829, + "loss": 0.2002, + "step": 2452 + }, + { + "epoch": 0.021293217940816487, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019989626750186226, + "loss": 0.2178, + "step": 2453 + }, + { + "epoch": 0.021301898421020652, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019989612459793246, + "loss": 0.2441, + "step": 2454 + }, + { + "epoch": 0.021310578901224817, + "grad_norm": 0.0751953125, + "learning_rate": 0.001998959815956936, + "loss": 0.209, + "step": 2455 + }, + { + "epoch": 0.021319259381428982, + "grad_norm": 0.07421875, + "learning_rate": 0.001998958384951458, + "loss": 0.2334, + "step": 2456 + }, + { + "epoch": 0.021327939861633144, + "grad_norm": 0.072265625, + "learning_rate": 0.0019989569529628936, + "loss": 0.2012, + "step": 2457 + }, + { + "epoch": 0.02133662034183731, + "grad_norm": 0.09521484375, + "learning_rate": 0.001998955519991243, + "loss": 0.2227, + "step": 2458 + }, + { + "epoch": 0.021345300822041474, + "grad_norm": 0.058837890625, + "learning_rate": 0.001998954086036508, + "loss": 0.1768, + "step": 2459 + }, + { + "epoch": 0.02135398130224564, + "grad_norm": 0.099609375, + "learning_rate": 0.00199895265109869, + "loss": 0.2168, + "step": 2460 + }, + { + "epoch": 0.021362661782449804, + "grad_norm": 0.0966796875, + "learning_rate": 0.001998951215177791, + "loss": 0.2227, + "step": 2461 + }, + { + "epoch": 0.02137134226265397, + "grad_norm": 0.10986328125, + "learning_rate": 0.001998949778273813, + "loss": 0.2363, + "step": 2462 + }, + { + "epoch": 0.021380022742858135, + "grad_norm": 0.09814453125, + "learning_rate": 0.001998948340386757, + "loss": 0.2285, + "step": 2463 + }, + { + "epoch": 0.0213887032230623, + "grad_norm": 0.0654296875, + "learning_rate": 0.001998946901516624, + "loss": 0.1904, + "step": 2464 + }, + { + "epoch": 0.021397383703266465, + "grad_norm": 0.09423828125, + "learning_rate": 0.001998945461663416, + "loss": 0.2402, + "step": 2465 + }, + { + "epoch": 0.02140606418347063, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019989440208271355, + "loss": 0.2021, + "step": 2466 + }, + { + "epoch": 0.021414744663674795, + "grad_norm": 0.08056640625, + "learning_rate": 0.001998942579007783, + "loss": 0.1895, + "step": 2467 + }, + { + "epoch": 0.02142342514387896, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019989411362053605, + "loss": 0.1699, + "step": 2468 + }, + { + "epoch": 0.021432105624083125, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019989396924198696, + "loss": 0.208, + "step": 2469 + }, + { + "epoch": 0.02144078610428729, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019989382476513114, + "loss": 0.2051, + "step": 2470 + }, + { + "epoch": 0.021449466584491456, + "grad_norm": 0.1181640625, + "learning_rate": 0.001998936801899688, + "loss": 0.248, + "step": 2471 + }, + { + "epoch": 0.021458147064695617, + "grad_norm": 0.07861328125, + "learning_rate": 0.001998935355165001, + "loss": 0.1963, + "step": 2472 + }, + { + "epoch": 0.021466827544899782, + "grad_norm": 0.05859375, + "learning_rate": 0.0019989339074472515, + "loss": 0.1738, + "step": 2473 + }, + { + "epoch": 0.021475508025103947, + "grad_norm": 0.06005859375, + "learning_rate": 0.0019989324587464416, + "loss": 0.1709, + "step": 2474 + }, + { + "epoch": 0.021484188505308113, + "grad_norm": 0.057373046875, + "learning_rate": 0.0019989310090625725, + "loss": 0.1768, + "step": 2475 + }, + { + "epoch": 0.021492868985512278, + "grad_norm": 0.1171875, + "learning_rate": 0.001998929558395646, + "loss": 0.252, + "step": 2476 + }, + { + "epoch": 0.021501549465716443, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019989281067456636, + "loss": 0.2109, + "step": 2477 + }, + { + "epoch": 0.021510229945920608, + "grad_norm": 0.109375, + "learning_rate": 0.001998926654112627, + "loss": 0.2852, + "step": 2478 + }, + { + "epoch": 0.021518910426124773, + "grad_norm": 0.0947265625, + "learning_rate": 0.001998925200496538, + "loss": 0.2227, + "step": 2479 + }, + { + "epoch": 0.021527590906328938, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019989237458973978, + "loss": 0.2383, + "step": 2480 + }, + { + "epoch": 0.021536271386533103, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019989222903152074, + "loss": 0.2051, + "step": 2481 + }, + { + "epoch": 0.02154495186673727, + "grad_norm": 0.0908203125, + "learning_rate": 0.00199892083374997, + "loss": 0.2266, + "step": 2482 + }, + { + "epoch": 0.021553632346941434, + "grad_norm": 0.06884765625, + "learning_rate": 0.001998919376201686, + "loss": 0.2275, + "step": 2483 + }, + { + "epoch": 0.0215623128271456, + "grad_norm": 0.09814453125, + "learning_rate": 0.001998917917670357, + "loss": 0.2188, + "step": 2484 + }, + { + "epoch": 0.021570993307349764, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019989164581559853, + "loss": 0.2793, + "step": 2485 + }, + { + "epoch": 0.02157967378755393, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019989149976585717, + "loss": 0.2168, + "step": 2486 + }, + { + "epoch": 0.021588354267758094, + "grad_norm": 0.06396484375, + "learning_rate": 0.001998913536178118, + "loss": 0.1738, + "step": 2487 + }, + { + "epoch": 0.021597034747962256, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019989120737146267, + "loss": 0.2461, + "step": 2488 + }, + { + "epoch": 0.02160571522816642, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019989106102680982, + "loss": 0.2715, + "step": 2489 + }, + { + "epoch": 0.021614395708370586, + "grad_norm": 0.1015625, + "learning_rate": 0.0019989091458385345, + "loss": 0.2617, + "step": 2490 + }, + { + "epoch": 0.02162307618857475, + "grad_norm": 0.060546875, + "learning_rate": 0.0019989076804259372, + "loss": 0.1895, + "step": 2491 + }, + { + "epoch": 0.021631756668778916, + "grad_norm": 0.08544921875, + "learning_rate": 0.001998906214030308, + "loss": 0.2344, + "step": 2492 + }, + { + "epoch": 0.02164043714898308, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019989047466516486, + "loss": 0.1641, + "step": 2493 + }, + { + "epoch": 0.021649117629187246, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019989032782899607, + "loss": 0.2266, + "step": 2494 + }, + { + "epoch": 0.02165779810939141, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019989018089452454, + "loss": 0.2441, + "step": 2495 + }, + { + "epoch": 0.021666478589595577, + "grad_norm": 0.080078125, + "learning_rate": 0.0019989003386175043, + "loss": 0.1973, + "step": 2496 + }, + { + "epoch": 0.021675159069799742, + "grad_norm": 0.07421875, + "learning_rate": 0.00199889886730674, + "loss": 0.2148, + "step": 2497 + }, + { + "epoch": 0.021683839550003907, + "grad_norm": 0.0908203125, + "learning_rate": 0.001998897395012953, + "loss": 0.248, + "step": 2498 + }, + { + "epoch": 0.021692520030208072, + "grad_norm": 0.08203125, + "learning_rate": 0.0019988959217361454, + "loss": 0.2578, + "step": 2499 + }, + { + "epoch": 0.021701200510412237, + "grad_norm": 0.08984375, + "learning_rate": 0.0019988944474763188, + "loss": 0.2422, + "step": 2500 + }, + { + "epoch": 0.021709880990616402, + "grad_norm": 0.060546875, + "learning_rate": 0.0019988929722334742, + "loss": 0.1768, + "step": 2501 + }, + { + "epoch": 0.021718561470820567, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019988914960076144, + "loss": 0.2207, + "step": 2502 + }, + { + "epoch": 0.02172724195102473, + "grad_norm": 0.080078125, + "learning_rate": 0.00199889001879874, + "loss": 0.2021, + "step": 2503 + }, + { + "epoch": 0.021735922431228894, + "grad_norm": 0.083984375, + "learning_rate": 0.0019988885406068534, + "loss": 0.2197, + "step": 2504 + }, + { + "epoch": 0.02174460291143306, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019988870614319554, + "loss": 0.2148, + "step": 2505 + }, + { + "epoch": 0.021753283391637224, + "grad_norm": 0.076171875, + "learning_rate": 0.001998885581274048, + "loss": 0.2227, + "step": 2506 + }, + { + "epoch": 0.02176196387184139, + "grad_norm": 0.1572265625, + "learning_rate": 0.001998884100133133, + "loss": 0.2471, + "step": 2507 + }, + { + "epoch": 0.021770644352045555, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019988826180092115, + "loss": 0.2031, + "step": 2508 + }, + { + "epoch": 0.02177932483224972, + "grad_norm": 0.0849609375, + "learning_rate": 0.001998881134902286, + "loss": 0.2197, + "step": 2509 + }, + { + "epoch": 0.021788005312453885, + "grad_norm": 0.09375, + "learning_rate": 0.0019988796508123574, + "loss": 0.1982, + "step": 2510 + }, + { + "epoch": 0.02179668579265805, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019988781657394273, + "loss": 0.2402, + "step": 2511 + }, + { + "epoch": 0.021805366272862215, + "grad_norm": 0.09619140625, + "learning_rate": 0.001998876679683498, + "loss": 0.2246, + "step": 2512 + }, + { + "epoch": 0.02181404675306638, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019988751926445702, + "loss": 0.2207, + "step": 2513 + }, + { + "epoch": 0.021822727233270545, + "grad_norm": 0.091796875, + "learning_rate": 0.0019988737046226462, + "loss": 0.2383, + "step": 2514 + }, + { + "epoch": 0.02183140771347471, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019988722156177273, + "loss": 0.2305, + "step": 2515 + }, + { + "epoch": 0.021840088193678876, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019988707256298156, + "loss": 0.2383, + "step": 2516 + }, + { + "epoch": 0.02184876867388304, + "grad_norm": 0.07421875, + "learning_rate": 0.001998869234658912, + "loss": 0.207, + "step": 2517 + }, + { + "epoch": 0.021857449154087206, + "grad_norm": 0.07421875, + "learning_rate": 0.0019988677427050187, + "loss": 0.1738, + "step": 2518 + }, + { + "epoch": 0.021866129634291367, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019988662497681373, + "loss": 0.2002, + "step": 2519 + }, + { + "epoch": 0.021874810114495533, + "grad_norm": 0.07568359375, + "learning_rate": 0.001998864755848269, + "loss": 0.1963, + "step": 2520 + }, + { + "epoch": 0.021883490594699698, + "grad_norm": 0.068359375, + "learning_rate": 0.001998863260945416, + "loss": 0.2061, + "step": 2521 + }, + { + "epoch": 0.021892171074903863, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019988617650595793, + "loss": 0.208, + "step": 2522 + }, + { + "epoch": 0.021900851555108028, + "grad_norm": 0.09423828125, + "learning_rate": 0.001998860268190761, + "loss": 0.2334, + "step": 2523 + }, + { + "epoch": 0.021909532035312193, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019988587703389624, + "loss": 0.1836, + "step": 2524 + }, + { + "epoch": 0.021918212515516358, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019988572715041856, + "loss": 0.1738, + "step": 2525 + }, + { + "epoch": 0.021926892995720523, + "grad_norm": 0.06884765625, + "learning_rate": 0.001998855771686432, + "loss": 0.1689, + "step": 2526 + }, + { + "epoch": 0.02193557347592469, + "grad_norm": 0.07421875, + "learning_rate": 0.001998854270885703, + "loss": 0.2305, + "step": 2527 + }, + { + "epoch": 0.021944253956128854, + "grad_norm": 0.07763671875, + "learning_rate": 0.001998852769102001, + "loss": 0.2363, + "step": 2528 + }, + { + "epoch": 0.02195293443633302, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019988512663353265, + "loss": 0.2188, + "step": 2529 + }, + { + "epoch": 0.021961614916537184, + "grad_norm": 0.08203125, + "learning_rate": 0.001998849762585682, + "loss": 0.1953, + "step": 2530 + }, + { + "epoch": 0.02197029539674135, + "grad_norm": 0.08984375, + "learning_rate": 0.0019988482578530693, + "loss": 0.2236, + "step": 2531 + }, + { + "epoch": 0.021978975876945514, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019988467521374893, + "loss": 0.2285, + "step": 2532 + }, + { + "epoch": 0.02198765635714968, + "grad_norm": 0.0791015625, + "learning_rate": 0.001998845245438944, + "loss": 0.1895, + "step": 2533 + }, + { + "epoch": 0.02199633683735384, + "grad_norm": 0.087890625, + "learning_rate": 0.001998843737757435, + "loss": 0.2246, + "step": 2534 + }, + { + "epoch": 0.022005017317558006, + "grad_norm": 0.07421875, + "learning_rate": 0.001998842229092964, + "loss": 0.2158, + "step": 2535 + }, + { + "epoch": 0.02201369779776217, + "grad_norm": 0.061767578125, + "learning_rate": 0.0019988407194455327, + "loss": 0.2119, + "step": 2536 + }, + { + "epoch": 0.022022378277966336, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019988392088151428, + "loss": 0.2051, + "step": 2537 + }, + { + "epoch": 0.0220310587581705, + "grad_norm": 0.06689453125, + "learning_rate": 0.0019988376972017957, + "loss": 0.1777, + "step": 2538 + }, + { + "epoch": 0.022039739238374666, + "grad_norm": 0.0986328125, + "learning_rate": 0.001998836184605493, + "loss": 0.248, + "step": 2539 + }, + { + "epoch": 0.02204841971857883, + "grad_norm": 0.10302734375, + "learning_rate": 0.001998834671026237, + "loss": 0.2188, + "step": 2540 + }, + { + "epoch": 0.022057100198782997, + "grad_norm": 0.06494140625, + "learning_rate": 0.001998833156464029, + "loss": 0.1963, + "step": 2541 + }, + { + "epoch": 0.022065780678987162, + "grad_norm": 0.078125, + "learning_rate": 0.00199883164091887, + "loss": 0.2041, + "step": 2542 + }, + { + "epoch": 0.022074461159191327, + "grad_norm": 0.08984375, + "learning_rate": 0.0019988301243907625, + "loss": 0.2246, + "step": 2543 + }, + { + "epoch": 0.022083141639395492, + "grad_norm": 0.08447265625, + "learning_rate": 0.001998828606879708, + "loss": 0.2109, + "step": 2544 + }, + { + "epoch": 0.022091822119599657, + "grad_norm": 0.57421875, + "learning_rate": 0.001998827088385708, + "loss": 0.2988, + "step": 2545 + }, + { + "epoch": 0.022100502599803822, + "grad_norm": 0.30078125, + "learning_rate": 0.001998825568908764, + "loss": 0.2324, + "step": 2546 + }, + { + "epoch": 0.022109183080007987, + "grad_norm": 0.09130859375, + "learning_rate": 0.001998824048448878, + "loss": 0.1836, + "step": 2547 + }, + { + "epoch": 0.022117863560212152, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019988225270060516, + "loss": 0.2402, + "step": 2548 + }, + { + "epoch": 0.022126544040416314, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019988210045802863, + "loss": 0.1641, + "step": 2549 + }, + { + "epoch": 0.02213522452062048, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019988194811715837, + "loss": 0.2129, + "step": 2550 + }, + { + "epoch": 0.022143905000824644, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019988179567799456, + "loss": 0.2041, + "step": 2551 + }, + { + "epoch": 0.02215258548102881, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019988164314053737, + "loss": 0.2637, + "step": 2552 + }, + { + "epoch": 0.022161265961232975, + "grad_norm": 0.1015625, + "learning_rate": 0.00199881490504787, + "loss": 0.2656, + "step": 2553 + }, + { + "epoch": 0.02216994644143714, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019988133777074355, + "loss": 0.2402, + "step": 2554 + }, + { + "epoch": 0.022178626921641305, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019988118493840727, + "loss": 0.2227, + "step": 2555 + }, + { + "epoch": 0.02218730740184547, + "grad_norm": 0.087890625, + "learning_rate": 0.001998810320077782, + "loss": 0.2148, + "step": 2556 + }, + { + "epoch": 0.022195987882049635, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019988087897885665, + "loss": 0.1885, + "step": 2557 + }, + { + "epoch": 0.0222046683622538, + "grad_norm": 0.11083984375, + "learning_rate": 0.001998807258516427, + "loss": 0.2285, + "step": 2558 + }, + { + "epoch": 0.022213348842457965, + "grad_norm": 0.078125, + "learning_rate": 0.001998805726261365, + "loss": 0.1758, + "step": 2559 + }, + { + "epoch": 0.02222202932266213, + "grad_norm": 0.42578125, + "learning_rate": 0.001998804193023383, + "loss": 0.2637, + "step": 2560 + }, + { + "epoch": 0.022230709802866296, + "grad_norm": 0.1083984375, + "learning_rate": 0.001998802658802482, + "loss": 0.2324, + "step": 2561 + }, + { + "epoch": 0.02223939028307046, + "grad_norm": 0.068359375, + "learning_rate": 0.001998801123598664, + "loss": 0.2256, + "step": 2562 + }, + { + "epoch": 0.022248070763274626, + "grad_norm": 0.1318359375, + "learning_rate": 0.001998799587411931, + "loss": 0.2344, + "step": 2563 + }, + { + "epoch": 0.02225675124347879, + "grad_norm": 0.087890625, + "learning_rate": 0.001998798050242284, + "loss": 0.2246, + "step": 2564 + }, + { + "epoch": 0.022265431723682953, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019987965120897245, + "loss": 0.209, + "step": 2565 + }, + { + "epoch": 0.022274112203887118, + "grad_norm": 0.0625, + "learning_rate": 0.001998794972954255, + "loss": 0.1699, + "step": 2566 + }, + { + "epoch": 0.022282792684091283, + "grad_norm": 0.0615234375, + "learning_rate": 0.001998793432835877, + "loss": 0.1943, + "step": 2567 + }, + { + "epoch": 0.022291473164295448, + "grad_norm": 0.259765625, + "learning_rate": 0.0019987918917345917, + "loss": 0.2676, + "step": 2568 + }, + { + "epoch": 0.022300153644499613, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019987903496504014, + "loss": 0.2266, + "step": 2569 + }, + { + "epoch": 0.022308834124703778, + "grad_norm": 0.0625, + "learning_rate": 0.001998788806583307, + "loss": 0.1748, + "step": 2570 + }, + { + "epoch": 0.022317514604907943, + "grad_norm": 0.0986328125, + "learning_rate": 0.001998787262533311, + "loss": 0.2275, + "step": 2571 + }, + { + "epoch": 0.02232619508511211, + "grad_norm": 0.08984375, + "learning_rate": 0.001998785717500415, + "loss": 0.2656, + "step": 2572 + }, + { + "epoch": 0.022334875565316274, + "grad_norm": 0.06640625, + "learning_rate": 0.00199878417148462, + "loss": 0.2295, + "step": 2573 + }, + { + "epoch": 0.02234355604552044, + "grad_norm": 0.08447265625, + "learning_rate": 0.001998782624485928, + "loss": 0.2412, + "step": 2574 + }, + { + "epoch": 0.022352236525724604, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019987810765043413, + "loss": 0.2061, + "step": 2575 + }, + { + "epoch": 0.02236091700592877, + "grad_norm": 0.06982421875, + "learning_rate": 0.001998779527539861, + "loss": 0.2061, + "step": 2576 + }, + { + "epoch": 0.022369597486132934, + "grad_norm": 0.1005859375, + "learning_rate": 0.001998777977592489, + "loss": 0.2188, + "step": 2577 + }, + { + "epoch": 0.0223782779663371, + "grad_norm": 0.0703125, + "learning_rate": 0.0019987764266622267, + "loss": 0.2217, + "step": 2578 + }, + { + "epoch": 0.022386958446541264, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019987748747490757, + "loss": 0.2256, + "step": 2579 + }, + { + "epoch": 0.022395638926745426, + "grad_norm": 0.1171875, + "learning_rate": 0.0019987733218530387, + "loss": 0.2891, + "step": 2580 + }, + { + "epoch": 0.02240431940694959, + "grad_norm": 0.083984375, + "learning_rate": 0.001998771767974116, + "loss": 0.2246, + "step": 2581 + }, + { + "epoch": 0.022412999887153756, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019987702131123108, + "loss": 0.3809, + "step": 2582 + }, + { + "epoch": 0.02242168036735792, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019987686572676237, + "loss": 0.1729, + "step": 2583 + }, + { + "epoch": 0.022430360847562086, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019987671004400563, + "loss": 0.2168, + "step": 2584 + }, + { + "epoch": 0.02243904132776625, + "grad_norm": 0.10107421875, + "learning_rate": 0.001998765542629611, + "loss": 0.2207, + "step": 2585 + }, + { + "epoch": 0.022447721807970417, + "grad_norm": 0.0869140625, + "learning_rate": 0.001998763983836289, + "loss": 0.2314, + "step": 2586 + }, + { + "epoch": 0.02245640228817458, + "grad_norm": 0.087890625, + "learning_rate": 0.0019987624240600924, + "loss": 0.2715, + "step": 2587 + }, + { + "epoch": 0.022465082768378747, + "grad_norm": 0.10400390625, + "learning_rate": 0.001998760863301023, + "loss": 0.2617, + "step": 2588 + }, + { + "epoch": 0.022473763248582912, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019987593015590817, + "loss": 0.1777, + "step": 2589 + }, + { + "epoch": 0.022482443728787077, + "grad_norm": 0.072265625, + "learning_rate": 0.001998757738834271, + "loss": 0.1631, + "step": 2590 + }, + { + "epoch": 0.022491124208991242, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019987561751265925, + "loss": 0.1875, + "step": 2591 + }, + { + "epoch": 0.022499804689195407, + "grad_norm": 0.0751953125, + "learning_rate": 0.001998754610436048, + "loss": 0.2021, + "step": 2592 + }, + { + "epoch": 0.022508485169399572, + "grad_norm": 0.05712890625, + "learning_rate": 0.001998753044762638, + "loss": 0.1973, + "step": 2593 + }, + { + "epoch": 0.022517165649603738, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019987514781063657, + "loss": 0.2188, + "step": 2594 + }, + { + "epoch": 0.022525846129807903, + "grad_norm": 0.11767578125, + "learning_rate": 0.001998749910467232, + "loss": 0.2969, + "step": 2595 + }, + { + "epoch": 0.022534526610012064, + "grad_norm": 0.0771484375, + "learning_rate": 0.001998748341845239, + "loss": 0.209, + "step": 2596 + }, + { + "epoch": 0.02254320709021623, + "grad_norm": 0.0859375, + "learning_rate": 0.001998746772240389, + "loss": 0.1836, + "step": 2597 + }, + { + "epoch": 0.022551887570420395, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019987452016526825, + "loss": 0.2061, + "step": 2598 + }, + { + "epoch": 0.02256056805062456, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019987436300821218, + "loss": 0.1992, + "step": 2599 + }, + { + "epoch": 0.022569248530828725, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019987420575287083, + "loss": 0.2207, + "step": 2600 + }, + { + "epoch": 0.02257792901103289, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019987404839924444, + "loss": 0.1787, + "step": 2601 + }, + { + "epoch": 0.022586609491237055, + "grad_norm": 0.1396484375, + "learning_rate": 0.001998738909473331, + "loss": 0.2402, + "step": 2602 + }, + { + "epoch": 0.02259528997144122, + "grad_norm": 0.0859375, + "learning_rate": 0.001998737333971371, + "loss": 0.1992, + "step": 2603 + }, + { + "epoch": 0.022603970451645385, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019987357574865645, + "loss": 0.1963, + "step": 2604 + }, + { + "epoch": 0.02261265093184955, + "grad_norm": 0.0986328125, + "learning_rate": 0.001998734180018914, + "loss": 0.2012, + "step": 2605 + }, + { + "epoch": 0.022621331412053716, + "grad_norm": 0.083984375, + "learning_rate": 0.001998732601568422, + "loss": 0.1973, + "step": 2606 + }, + { + "epoch": 0.02263001189225788, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019987310221350895, + "loss": 0.2178, + "step": 2607 + }, + { + "epoch": 0.022638692372462046, + "grad_norm": 0.0888671875, + "learning_rate": 0.001998729441718918, + "loss": 0.1875, + "step": 2608 + }, + { + "epoch": 0.02264737285266621, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019987278603199096, + "loss": 0.3105, + "step": 2609 + }, + { + "epoch": 0.022656053332870376, + "grad_norm": 0.275390625, + "learning_rate": 0.0019987262779380655, + "loss": 0.3203, + "step": 2610 + }, + { + "epoch": 0.022664733813074538, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019987246945733883, + "loss": 0.2656, + "step": 2611 + }, + { + "epoch": 0.022673414293278703, + "grad_norm": 0.09521484375, + "learning_rate": 0.001998723110225879, + "loss": 0.2188, + "step": 2612 + }, + { + "epoch": 0.022682094773482868, + "grad_norm": 0.08740234375, + "learning_rate": 0.00199872152489554, + "loss": 0.207, + "step": 2613 + }, + { + "epoch": 0.022690775253687033, + "grad_norm": 0.095703125, + "learning_rate": 0.0019987199385823723, + "loss": 0.2266, + "step": 2614 + }, + { + "epoch": 0.022699455733891198, + "grad_norm": 0.0859375, + "learning_rate": 0.001998718351286378, + "loss": 0.2168, + "step": 2615 + }, + { + "epoch": 0.022708136214095363, + "grad_norm": 0.103515625, + "learning_rate": 0.001998716763007559, + "loss": 0.2891, + "step": 2616 + }, + { + "epoch": 0.02271681669429953, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019987151737459164, + "loss": 0.1689, + "step": 2617 + }, + { + "epoch": 0.022725497174503693, + "grad_norm": 0.1875, + "learning_rate": 0.0019987135835014525, + "loss": 0.2578, + "step": 2618 + }, + { + "epoch": 0.02273417765470786, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019987119922741698, + "loss": 0.2334, + "step": 2619 + }, + { + "epoch": 0.022742858134912024, + "grad_norm": 0.1015625, + "learning_rate": 0.001998710400064068, + "loss": 0.2734, + "step": 2620 + }, + { + "epoch": 0.02275153861511619, + "grad_norm": 0.09375, + "learning_rate": 0.0019987088068711507, + "loss": 0.2949, + "step": 2621 + }, + { + "epoch": 0.022760219095320354, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019987072126954187, + "loss": 0.1973, + "step": 2622 + }, + { + "epoch": 0.02276889957552452, + "grad_norm": 0.0791015625, + "learning_rate": 0.001998705617536874, + "loss": 0.1865, + "step": 2623 + }, + { + "epoch": 0.022777580055728684, + "grad_norm": 0.1015625, + "learning_rate": 0.0019987040213955186, + "loss": 0.2891, + "step": 2624 + }, + { + "epoch": 0.02278626053593285, + "grad_norm": 0.07763671875, + "learning_rate": 0.001998702424271354, + "loss": 0.2266, + "step": 2625 + }, + { + "epoch": 0.02279494101613701, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019987008261643817, + "loss": 0.2227, + "step": 2626 + }, + { + "epoch": 0.022803621496341176, + "grad_norm": 0.08984375, + "learning_rate": 0.0019986992270746035, + "loss": 0.25, + "step": 2627 + }, + { + "epoch": 0.02281230197654534, + "grad_norm": 0.13671875, + "learning_rate": 0.0019986976270020213, + "loss": 0.25, + "step": 2628 + }, + { + "epoch": 0.022820982456749506, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019986960259466375, + "loss": 0.1924, + "step": 2629 + }, + { + "epoch": 0.02282966293695367, + "grad_norm": 0.060546875, + "learning_rate": 0.0019986944239084527, + "loss": 0.1797, + "step": 2630 + }, + { + "epoch": 0.022838343417157837, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019986928208874694, + "loss": 0.2559, + "step": 2631 + }, + { + "epoch": 0.022847023897362, + "grad_norm": 0.06591796875, + "learning_rate": 0.001998691216883689, + "loss": 0.1719, + "step": 2632 + }, + { + "epoch": 0.022855704377566167, + "grad_norm": 0.083984375, + "learning_rate": 0.0019986896118971134, + "loss": 0.209, + "step": 2633 + }, + { + "epoch": 0.022864384857770332, + "grad_norm": 0.5546875, + "learning_rate": 0.001998688005927744, + "loss": 0.5664, + "step": 2634 + }, + { + "epoch": 0.022873065337974497, + "grad_norm": 0.06787109375, + "learning_rate": 0.001998686398975584, + "loss": 0.2236, + "step": 2635 + }, + { + "epoch": 0.022881745818178662, + "grad_norm": 0.0908203125, + "learning_rate": 0.001998684791040633, + "loss": 0.2129, + "step": 2636 + }, + { + "epoch": 0.022890426298382827, + "grad_norm": 0.087890625, + "learning_rate": 0.0019986831821228943, + "loss": 0.2383, + "step": 2637 + }, + { + "epoch": 0.022899106778586992, + "grad_norm": 0.1748046875, + "learning_rate": 0.001998681572222369, + "loss": 0.2266, + "step": 2638 + }, + { + "epoch": 0.022907787258791158, + "grad_norm": 0.078125, + "learning_rate": 0.001998679961339059, + "loss": 0.2344, + "step": 2639 + }, + { + "epoch": 0.022916467738995323, + "grad_norm": 0.11669921875, + "learning_rate": 0.001998678349472966, + "loss": 0.252, + "step": 2640 + }, + { + "epoch": 0.022925148219199488, + "grad_norm": 0.0703125, + "learning_rate": 0.001998676736624092, + "loss": 0.1807, + "step": 2641 + }, + { + "epoch": 0.02293382869940365, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019986751227924386, + "loss": 0.2344, + "step": 2642 + }, + { + "epoch": 0.022942509179607815, + "grad_norm": 0.062255859375, + "learning_rate": 0.001998673507978008, + "loss": 0.1982, + "step": 2643 + }, + { + "epoch": 0.02295118965981198, + "grad_norm": 0.1005859375, + "learning_rate": 0.001998671892180801, + "loss": 0.2734, + "step": 2644 + }, + { + "epoch": 0.022959870140016145, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019986702754008203, + "loss": 0.2051, + "step": 2645 + }, + { + "epoch": 0.02296855062022031, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019986686576380667, + "loss": 0.2305, + "step": 2646 + }, + { + "epoch": 0.022977231100424475, + "grad_norm": 0.1357421875, + "learning_rate": 0.001998667038892543, + "loss": 0.2363, + "step": 2647 + }, + { + "epoch": 0.02298591158062864, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019986654191642508, + "loss": 0.1924, + "step": 2648 + }, + { + "epoch": 0.022994592060832805, + "grad_norm": 0.0859375, + "learning_rate": 0.001998663798453191, + "loss": 0.1934, + "step": 2649 + }, + { + "epoch": 0.02300327254103697, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019986621767593663, + "loss": 0.2188, + "step": 2650 + }, + { + "epoch": 0.023011953021241136, + "grad_norm": 0.11572265625, + "learning_rate": 0.001998660554082778, + "loss": 0.2539, + "step": 2651 + }, + { + "epoch": 0.0230206335014453, + "grad_norm": 0.158203125, + "learning_rate": 0.0019986589304234284, + "loss": 0.2852, + "step": 2652 + }, + { + "epoch": 0.023029313981649466, + "grad_norm": 0.95703125, + "learning_rate": 0.0019986573057813183, + "loss": 0.4863, + "step": 2653 + }, + { + "epoch": 0.02303799446185363, + "grad_norm": 0.15625, + "learning_rate": 0.0019986556801564505, + "loss": 0.2031, + "step": 2654 + }, + { + "epoch": 0.023046674942057796, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019986540535488263, + "loss": 0.1895, + "step": 2655 + }, + { + "epoch": 0.02305535542226196, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019986524259584474, + "loss": 0.2363, + "step": 2656 + }, + { + "epoch": 0.023064035902466123, + "grad_norm": 0.6484375, + "learning_rate": 0.001998650797385316, + "loss": 0.5977, + "step": 2657 + }, + { + "epoch": 0.023072716382670288, + "grad_norm": 0.11669921875, + "learning_rate": 0.001998649167829433, + "loss": 0.293, + "step": 2658 + }, + { + "epoch": 0.023081396862874453, + "grad_norm": 0.072265625, + "learning_rate": 0.0019986475372908014, + "loss": 0.2168, + "step": 2659 + }, + { + "epoch": 0.023090077343078618, + "grad_norm": 0.09765625, + "learning_rate": 0.001998645905769422, + "loss": 0.2051, + "step": 2660 + }, + { + "epoch": 0.023098757823282783, + "grad_norm": 0.062255859375, + "learning_rate": 0.001998644273265297, + "loss": 0.1738, + "step": 2661 + }, + { + "epoch": 0.02310743830348695, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019986426397784283, + "loss": 0.209, + "step": 2662 + }, + { + "epoch": 0.023116118783691113, + "grad_norm": 0.060546875, + "learning_rate": 0.0019986410053088174, + "loss": 0.2031, + "step": 2663 + }, + { + "epoch": 0.02312479926389528, + "grad_norm": 0.953125, + "learning_rate": 0.001998639369856466, + "loss": 0.7188, + "step": 2664 + }, + { + "epoch": 0.023133479744099444, + "grad_norm": 0.09375, + "learning_rate": 0.0019986377334213763, + "loss": 0.1875, + "step": 2665 + }, + { + "epoch": 0.02314216022430361, + "grad_norm": 0.1455078125, + "learning_rate": 0.00199863609600355, + "loss": 0.3125, + "step": 2666 + }, + { + "epoch": 0.023150840704507774, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019986344576029885, + "loss": 0.2158, + "step": 2667 + }, + { + "epoch": 0.02315952118471194, + "grad_norm": 0.06640625, + "learning_rate": 0.0019986328182196936, + "loss": 0.168, + "step": 2668 + }, + { + "epoch": 0.023168201664916104, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019986311778536683, + "loss": 0.2891, + "step": 2669 + }, + { + "epoch": 0.02317688214512027, + "grad_norm": 0.0625, + "learning_rate": 0.0019986295365049126, + "loss": 0.1934, + "step": 2670 + }, + { + "epoch": 0.023185562625324434, + "grad_norm": 0.08984375, + "learning_rate": 0.0019986278941734295, + "loss": 0.2559, + "step": 2671 + }, + { + "epoch": 0.0231942431055286, + "grad_norm": 0.06640625, + "learning_rate": 0.0019986262508592204, + "loss": 0.2383, + "step": 2672 + }, + { + "epoch": 0.02320292358573276, + "grad_norm": 0.083984375, + "learning_rate": 0.001998624606562287, + "loss": 0.2402, + "step": 2673 + }, + { + "epoch": 0.023211604065936926, + "grad_norm": 0.0947265625, + "learning_rate": 0.001998622961282631, + "loss": 0.2178, + "step": 2674 + }, + { + "epoch": 0.02322028454614109, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019986213150202546, + "loss": 0.2578, + "step": 2675 + }, + { + "epoch": 0.023228965026345257, + "grad_norm": 0.0751953125, + "learning_rate": 0.00199861966777516, + "loss": 0.2363, + "step": 2676 + }, + { + "epoch": 0.02323764550654942, + "grad_norm": 0.0810546875, + "learning_rate": 0.001998618019547348, + "loss": 0.2617, + "step": 2677 + }, + { + "epoch": 0.023246325986753587, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019986163703368206, + "loss": 0.2246, + "step": 2678 + }, + { + "epoch": 0.023255006466957752, + "grad_norm": 0.236328125, + "learning_rate": 0.0019986147201435803, + "loss": 0.2207, + "step": 2679 + }, + { + "epoch": 0.023263686947161917, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019986130689676283, + "loss": 0.2227, + "step": 2680 + }, + { + "epoch": 0.023272367427366082, + "grad_norm": 0.0888671875, + "learning_rate": 0.001998611416808966, + "loss": 0.248, + "step": 2681 + }, + { + "epoch": 0.023281047907570247, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019986097636675963, + "loss": 0.2002, + "step": 2682 + }, + { + "epoch": 0.023289728387774412, + "grad_norm": 0.080078125, + "learning_rate": 0.0019986081095435208, + "loss": 0.2402, + "step": 2683 + }, + { + "epoch": 0.023298408867978578, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019986064544367404, + "loss": 0.2168, + "step": 2684 + }, + { + "epoch": 0.023307089348182743, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019986047983472574, + "loss": 0.1953, + "step": 2685 + }, + { + "epoch": 0.023315769828386908, + "grad_norm": 0.095703125, + "learning_rate": 0.001998603141275074, + "loss": 0.2168, + "step": 2686 + }, + { + "epoch": 0.023324450308591073, + "grad_norm": 0.1298828125, + "learning_rate": 0.001998601483220192, + "loss": 0.2236, + "step": 2687 + }, + { + "epoch": 0.023333130788795235, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019985998241826126, + "loss": 0.2109, + "step": 2688 + }, + { + "epoch": 0.0233418112689994, + "grad_norm": 0.107421875, + "learning_rate": 0.0019985981641623377, + "loss": 0.248, + "step": 2689 + }, + { + "epoch": 0.023350491749203565, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019985965031593697, + "loss": 0.1982, + "step": 2690 + }, + { + "epoch": 0.02335917222940773, + "grad_norm": 0.08740234375, + "learning_rate": 0.00199859484117371, + "loss": 0.2285, + "step": 2691 + }, + { + "epoch": 0.023367852709611895, + "grad_norm": 0.0966796875, + "learning_rate": 0.00199859317820536, + "loss": 0.2021, + "step": 2692 + }, + { + "epoch": 0.02337653318981606, + "grad_norm": 0.19140625, + "learning_rate": 0.0019985915142543224, + "loss": 0.3047, + "step": 2693 + }, + { + "epoch": 0.023385213670020225, + "grad_norm": 0.08154296875, + "learning_rate": 0.001998589849320599, + "loss": 0.1816, + "step": 2694 + }, + { + "epoch": 0.02339389415022439, + "grad_norm": 0.0859375, + "learning_rate": 0.0019985881834041906, + "loss": 0.2598, + "step": 2695 + }, + { + "epoch": 0.023402574630428555, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019985865165050997, + "loss": 0.2002, + "step": 2696 + }, + { + "epoch": 0.02341125511063272, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019985848486233286, + "loss": 0.2188, + "step": 2697 + }, + { + "epoch": 0.023419935590836886, + "grad_norm": 0.09765625, + "learning_rate": 0.0019985831797588783, + "loss": 0.2617, + "step": 2698 + }, + { + "epoch": 0.02342861607104105, + "grad_norm": 0.1142578125, + "learning_rate": 0.001998581509911751, + "loss": 0.2314, + "step": 2699 + }, + { + "epoch": 0.023437296551245216, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019985798390819483, + "loss": 0.252, + "step": 2700 + }, + { + "epoch": 0.02344597703144938, + "grad_norm": 0.0751953125, + "learning_rate": 0.001998578167269472, + "loss": 0.2246, + "step": 2701 + }, + { + "epoch": 0.023454657511653546, + "grad_norm": 0.07568359375, + "learning_rate": 0.001998576494474325, + "loss": 0.248, + "step": 2702 + }, + { + "epoch": 0.023463337991857708, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019985748206965076, + "loss": 0.2754, + "step": 2703 + }, + { + "epoch": 0.023472018472061873, + "grad_norm": 0.056640625, + "learning_rate": 0.0019985731459360224, + "loss": 0.2021, + "step": 2704 + }, + { + "epoch": 0.023480698952266038, + "grad_norm": 0.1083984375, + "learning_rate": 0.001998571470192871, + "loss": 0.332, + "step": 2705 + }, + { + "epoch": 0.023489379432470203, + "grad_norm": 0.09130859375, + "learning_rate": 0.001998569793467055, + "loss": 0.2695, + "step": 2706 + }, + { + "epoch": 0.02349805991267437, + "grad_norm": 0.09765625, + "learning_rate": 0.0019985681157585772, + "loss": 0.248, + "step": 2707 + }, + { + "epoch": 0.023506740392878533, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019985664370674385, + "loss": 0.1719, + "step": 2708 + }, + { + "epoch": 0.0235154208730827, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019985647573936413, + "loss": 0.2168, + "step": 2709 + }, + { + "epoch": 0.023524101353286864, + "grad_norm": 0.73828125, + "learning_rate": 0.0019985630767371866, + "loss": 0.498, + "step": 2710 + }, + { + "epoch": 0.02353278183349103, + "grad_norm": 0.111328125, + "learning_rate": 0.0019985613950980778, + "loss": 0.165, + "step": 2711 + }, + { + "epoch": 0.023541462313695194, + "grad_norm": 0.07958984375, + "learning_rate": 0.001998559712476315, + "loss": 0.208, + "step": 2712 + }, + { + "epoch": 0.02355014279389936, + "grad_norm": 0.0791015625, + "learning_rate": 0.001998558028871901, + "loss": 0.2246, + "step": 2713 + }, + { + "epoch": 0.023558823274103524, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019985563442848375, + "loss": 0.252, + "step": 2714 + }, + { + "epoch": 0.02356750375430769, + "grad_norm": 0.1328125, + "learning_rate": 0.0019985546587151263, + "loss": 0.1953, + "step": 2715 + }, + { + "epoch": 0.023576184234511854, + "grad_norm": 0.16796875, + "learning_rate": 0.001998552972162769, + "loss": 0.2656, + "step": 2716 + }, + { + "epoch": 0.02358486471471602, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019985512846277674, + "loss": 0.293, + "step": 2717 + }, + { + "epoch": 0.023593545194920185, + "grad_norm": 0.076171875, + "learning_rate": 0.0019985495961101244, + "loss": 0.2227, + "step": 2718 + }, + { + "epoch": 0.023602225675124346, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019985479066098404, + "loss": 0.25, + "step": 2719 + }, + { + "epoch": 0.02361090615532851, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019985462161269184, + "loss": 0.2168, + "step": 2720 + }, + { + "epoch": 0.023619586635532677, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019985445246613596, + "loss": 0.209, + "step": 2721 + }, + { + "epoch": 0.02362826711573684, + "grad_norm": 0.09375, + "learning_rate": 0.001998542832213166, + "loss": 0.2246, + "step": 2722 + }, + { + "epoch": 0.023636947595941007, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019985411387823393, + "loss": 0.2246, + "step": 2723 + }, + { + "epoch": 0.023645628076145172, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019985394443688817, + "loss": 0.2393, + "step": 2724 + }, + { + "epoch": 0.023654308556349337, + "grad_norm": 0.158203125, + "learning_rate": 0.0019985377489727947, + "loss": 0.25, + "step": 2725 + }, + { + "epoch": 0.023662989036553502, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019985360525940805, + "loss": 0.2246, + "step": 2726 + }, + { + "epoch": 0.023671669516757667, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019985343552327405, + "loss": 0.2402, + "step": 2727 + }, + { + "epoch": 0.023680349996961832, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019985326568887772, + "loss": 0.1738, + "step": 2728 + }, + { + "epoch": 0.023689030477165997, + "grad_norm": 0.1044921875, + "learning_rate": 0.001998530957562192, + "loss": 0.2344, + "step": 2729 + }, + { + "epoch": 0.023697710957370163, + "grad_norm": 0.08837890625, + "learning_rate": 0.001998529257252987, + "loss": 0.2461, + "step": 2730 + }, + { + "epoch": 0.023706391437574328, + "grad_norm": 0.14453125, + "learning_rate": 0.0019985275559611633, + "loss": 0.2168, + "step": 2731 + }, + { + "epoch": 0.023715071917778493, + "grad_norm": 0.0849609375, + "learning_rate": 0.001998525853686724, + "loss": 0.2656, + "step": 2732 + }, + { + "epoch": 0.023723752397982658, + "grad_norm": 0.0859375, + "learning_rate": 0.00199852415042967, + "loss": 0.2109, + "step": 2733 + }, + { + "epoch": 0.02373243287818682, + "grad_norm": 0.158203125, + "learning_rate": 0.0019985224461900033, + "loss": 0.2871, + "step": 2734 + }, + { + "epoch": 0.023741113358390985, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019985207409677266, + "loss": 0.2949, + "step": 2735 + }, + { + "epoch": 0.02374979383859515, + "grad_norm": 0.068359375, + "learning_rate": 0.0019985190347628404, + "loss": 0.1895, + "step": 2736 + }, + { + "epoch": 0.023758474318799315, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019985173275753475, + "loss": 0.2598, + "step": 2737 + }, + { + "epoch": 0.02376715479900348, + "grad_norm": 0.059814453125, + "learning_rate": 0.0019985156194052495, + "loss": 0.2012, + "step": 2738 + }, + { + "epoch": 0.023775835279207645, + "grad_norm": 0.154296875, + "learning_rate": 0.0019985139102525486, + "loss": 0.2246, + "step": 2739 + }, + { + "epoch": 0.02378451575941181, + "grad_norm": 0.0732421875, + "learning_rate": 0.001998512200117246, + "loss": 0.1914, + "step": 2740 + }, + { + "epoch": 0.023793196239615975, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019985104889993443, + "loss": 0.2207, + "step": 2741 + }, + { + "epoch": 0.02380187671982014, + "grad_norm": 0.12890625, + "learning_rate": 0.001998508776898845, + "loss": 0.2969, + "step": 2742 + }, + { + "epoch": 0.023810557200024306, + "grad_norm": 0.130859375, + "learning_rate": 0.00199850706381575, + "loss": 0.2275, + "step": 2743 + }, + { + "epoch": 0.02381923768022847, + "grad_norm": 0.06884765625, + "learning_rate": 0.001998505349750061, + "loss": 0.2275, + "step": 2744 + }, + { + "epoch": 0.023827918160432636, + "grad_norm": 0.12109375, + "learning_rate": 0.0019985036347017803, + "loss": 0.1523, + "step": 2745 + }, + { + "epoch": 0.0238365986406368, + "grad_norm": 0.181640625, + "learning_rate": 0.001998501918670909, + "loss": 0.2197, + "step": 2746 + }, + { + "epoch": 0.023845279120840966, + "grad_norm": 0.0859375, + "learning_rate": 0.00199850020165745, + "loss": 0.209, + "step": 2747 + }, + { + "epoch": 0.02385395960104513, + "grad_norm": 0.05859375, + "learning_rate": 0.001998498483661405, + "loss": 0.1914, + "step": 2748 + }, + { + "epoch": 0.023862640081249296, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019984967646827748, + "loss": 0.2754, + "step": 2749 + }, + { + "epoch": 0.023871320561453458, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019984950447215627, + "loss": 0.1992, + "step": 2750 + }, + { + "epoch": 0.023880001041657623, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019984933237777694, + "loss": 0.2051, + "step": 2751 + }, + { + "epoch": 0.02388868152186179, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019984916018513975, + "loss": 0.2773, + "step": 2752 + }, + { + "epoch": 0.023897362002065953, + "grad_norm": 0.0634765625, + "learning_rate": 0.0019984898789424488, + "loss": 0.1494, + "step": 2753 + }, + { + "epoch": 0.02390604248227012, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019984881550509244, + "loss": 0.1758, + "step": 2754 + }, + { + "epoch": 0.023914722962474284, + "grad_norm": 0.08203125, + "learning_rate": 0.001998486430176828, + "loss": 0.167, + "step": 2755 + }, + { + "epoch": 0.02392340344267845, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019984847043201595, + "loss": 0.2217, + "step": 2756 + }, + { + "epoch": 0.023932083922882614, + "grad_norm": 0.173828125, + "learning_rate": 0.001998482977480922, + "loss": 0.2119, + "step": 2757 + }, + { + "epoch": 0.02394076440308678, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019984812496591166, + "loss": 0.2021, + "step": 2758 + }, + { + "epoch": 0.023949444883290944, + "grad_norm": 0.1171875, + "learning_rate": 0.001998479520854746, + "loss": 0.2363, + "step": 2759 + }, + { + "epoch": 0.02395812536349511, + "grad_norm": 0.068359375, + "learning_rate": 0.0019984777910678118, + "loss": 0.2432, + "step": 2760 + }, + { + "epoch": 0.023966805843699274, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019984760602983153, + "loss": 0.1758, + "step": 2761 + }, + { + "epoch": 0.02397548632390344, + "grad_norm": 0.0751953125, + "learning_rate": 0.001998474328546259, + "loss": 0.21, + "step": 2762 + }, + { + "epoch": 0.023984166804107605, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019984725958116446, + "loss": 0.2148, + "step": 2763 + }, + { + "epoch": 0.02399284728431177, + "grad_norm": 0.06103515625, + "learning_rate": 0.001998470862094475, + "loss": 0.2246, + "step": 2764 + }, + { + "epoch": 0.02400152776451593, + "grad_norm": 0.1142578125, + "learning_rate": 0.00199846912739475, + "loss": 0.2031, + "step": 2765 + }, + { + "epoch": 0.024010208244720097, + "grad_norm": 0.07470703125, + "learning_rate": 0.001998467391712473, + "loss": 0.1953, + "step": 2766 + }, + { + "epoch": 0.02401888872492426, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019984656550476455, + "loss": 0.2559, + "step": 2767 + }, + { + "epoch": 0.024027569205128427, + "grad_norm": 0.1787109375, + "learning_rate": 0.00199846391740027, + "loss": 0.3066, + "step": 2768 + }, + { + "epoch": 0.024036249685332592, + "grad_norm": 0.083984375, + "learning_rate": 0.0019984621787703474, + "loss": 0.2041, + "step": 2769 + }, + { + "epoch": 0.024044930165536757, + "grad_norm": 0.099609375, + "learning_rate": 0.0019984604391578803, + "loss": 0.2324, + "step": 2770 + }, + { + "epoch": 0.024053610645740922, + "grad_norm": 0.10302734375, + "learning_rate": 0.00199845869856287, + "loss": 0.2559, + "step": 2771 + }, + { + "epoch": 0.024062291125945087, + "grad_norm": 0.1279296875, + "learning_rate": 0.001998456956985319, + "loss": 0.2109, + "step": 2772 + }, + { + "epoch": 0.024070971606149252, + "grad_norm": 0.177734375, + "learning_rate": 0.001998455214425229, + "loss": 0.2656, + "step": 2773 + }, + { + "epoch": 0.024079652086353417, + "grad_norm": 0.173828125, + "learning_rate": 0.001998453470882602, + "loss": 0.3633, + "step": 2774 + }, + { + "epoch": 0.024088332566557583, + "grad_norm": 0.234375, + "learning_rate": 0.0019984517263574395, + "loss": 0.2559, + "step": 2775 + }, + { + "epoch": 0.024097013046761748, + "grad_norm": 0.111328125, + "learning_rate": 0.0019984499808497437, + "loss": 0.2363, + "step": 2776 + }, + { + "epoch": 0.024105693526965913, + "grad_norm": 0.1142578125, + "learning_rate": 0.001998448234359517, + "loss": 0.2871, + "step": 2777 + }, + { + "epoch": 0.024114374007170078, + "grad_norm": 0.146484375, + "learning_rate": 0.00199844648688676, + "loss": 0.2188, + "step": 2778 + }, + { + "epoch": 0.024123054487374243, + "grad_norm": 0.0712890625, + "learning_rate": 0.001998444738431476, + "loss": 0.2158, + "step": 2779 + }, + { + "epoch": 0.024131734967578405, + "grad_norm": 0.10009765625, + "learning_rate": 0.001998442988993666, + "loss": 0.248, + "step": 2780 + }, + { + "epoch": 0.02414041544778257, + "grad_norm": 0.1328125, + "learning_rate": 0.0019984412385733326, + "loss": 0.2324, + "step": 2781 + }, + { + "epoch": 0.024149095927986735, + "grad_norm": 0.08642578125, + "learning_rate": 0.001998439487170477, + "loss": 0.1807, + "step": 2782 + }, + { + "epoch": 0.0241577764081909, + "grad_norm": 0.271484375, + "learning_rate": 0.0019984377347851017, + "loss": 0.2051, + "step": 2783 + }, + { + "epoch": 0.024166456888395065, + "grad_norm": 0.068359375, + "learning_rate": 0.001998435981417208, + "loss": 0.1943, + "step": 2784 + }, + { + "epoch": 0.02417513736859923, + "grad_norm": 0.1044921875, + "learning_rate": 0.001998434227066799, + "loss": 0.3086, + "step": 2785 + }, + { + "epoch": 0.024183817848803395, + "grad_norm": 0.1982421875, + "learning_rate": 0.001998432471733875, + "loss": 0.2559, + "step": 2786 + }, + { + "epoch": 0.02419249832900756, + "grad_norm": 0.0849609375, + "learning_rate": 0.001998430715418439, + "loss": 0.2383, + "step": 2787 + }, + { + "epoch": 0.024201178809211726, + "grad_norm": 0.1083984375, + "learning_rate": 0.001998428958120493, + "loss": 0.2207, + "step": 2788 + }, + { + "epoch": 0.02420985928941589, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019984271998400387, + "loss": 0.1689, + "step": 2789 + }, + { + "epoch": 0.024218539769620056, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019984254405770773, + "loss": 0.2246, + "step": 2790 + }, + { + "epoch": 0.02422722024982422, + "grad_norm": 0.158203125, + "learning_rate": 0.001998423680331612, + "loss": 0.1982, + "step": 2791 + }, + { + "epoch": 0.024235900730028386, + "grad_norm": 0.10546875, + "learning_rate": 0.0019984219191036432, + "loss": 0.2324, + "step": 2792 + }, + { + "epoch": 0.02424458121023255, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019984201568931746, + "loss": 0.1924, + "step": 2793 + }, + { + "epoch": 0.024253261690436716, + "grad_norm": 0.078125, + "learning_rate": 0.0019984183937002066, + "loss": 0.2246, + "step": 2794 + }, + { + "epoch": 0.02426194217064088, + "grad_norm": 0.08056640625, + "learning_rate": 0.001998416629524742, + "loss": 0.1826, + "step": 2795 + }, + { + "epoch": 0.024270622650845043, + "grad_norm": 0.068359375, + "learning_rate": 0.001998414864366782, + "loss": 0.2188, + "step": 2796 + }, + { + "epoch": 0.02427930313104921, + "grad_norm": 0.07666015625, + "learning_rate": 0.00199841309822633, + "loss": 0.2021, + "step": 2797 + }, + { + "epoch": 0.024287983611253373, + "grad_norm": 0.08447265625, + "learning_rate": 0.001998411331103386, + "loss": 0.2637, + "step": 2798 + }, + { + "epoch": 0.02429666409145754, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019984095629979534, + "loss": 0.1973, + "step": 2799 + }, + { + "epoch": 0.024305344571661704, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019984077939100334, + "loss": 0.1992, + "step": 2800 + }, + { + "epoch": 0.02431402505186587, + "grad_norm": 0.1435546875, + "learning_rate": 0.001998406023839628, + "loss": 0.1934, + "step": 2801 + }, + { + "epoch": 0.024322705532070034, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019984042527867395, + "loss": 0.1943, + "step": 2802 + }, + { + "epoch": 0.0243313860122742, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019984024807513695, + "loss": 0.2363, + "step": 2803 + }, + { + "epoch": 0.024340066492478364, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019984007077335202, + "loss": 0.2383, + "step": 2804 + }, + { + "epoch": 0.02434874697268253, + "grad_norm": 0.07470703125, + "learning_rate": 0.001998398933733193, + "loss": 0.2236, + "step": 2805 + }, + { + "epoch": 0.024357427452886694, + "grad_norm": 0.0634765625, + "learning_rate": 0.0019983971587503907, + "loss": 0.1865, + "step": 2806 + }, + { + "epoch": 0.02436610793309086, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019983953827851144, + "loss": 0.2402, + "step": 2807 + }, + { + "epoch": 0.024374788413295025, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019983936058373666, + "loss": 0.2832, + "step": 2808 + }, + { + "epoch": 0.02438346889349919, + "grad_norm": 0.1357421875, + "learning_rate": 0.001998391827907149, + "loss": 0.2578, + "step": 2809 + }, + { + "epoch": 0.024392149373703355, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019983900489944635, + "loss": 0.2139, + "step": 2810 + }, + { + "epoch": 0.024400829853907516, + "grad_norm": 0.07666015625, + "learning_rate": 0.001998388269099312, + "loss": 0.2383, + "step": 2811 + }, + { + "epoch": 0.02440951033411168, + "grad_norm": 0.146484375, + "learning_rate": 0.0019983864882216974, + "loss": 0.2412, + "step": 2812 + }, + { + "epoch": 0.024418190814315847, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019983847063616204, + "loss": 0.1748, + "step": 2813 + }, + { + "epoch": 0.024426871294520012, + "grad_norm": 0.10888671875, + "learning_rate": 0.001998382923519083, + "loss": 0.2266, + "step": 2814 + }, + { + "epoch": 0.024435551774724177, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019983811396940875, + "loss": 0.1875, + "step": 2815 + }, + { + "epoch": 0.024444232254928342, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019983793548866364, + "loss": 0.1807, + "step": 2816 + }, + { + "epoch": 0.024452912735132507, + "grad_norm": 0.115234375, + "learning_rate": 0.0019983775690967306, + "loss": 0.2129, + "step": 2817 + }, + { + "epoch": 0.024461593215336672, + "grad_norm": 0.0810546875, + "learning_rate": 0.001998375782324373, + "loss": 0.2031, + "step": 2818 + }, + { + "epoch": 0.024470273695540837, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019983739945695655, + "loss": 0.2148, + "step": 2819 + }, + { + "epoch": 0.024478954175745003, + "grad_norm": 0.08544921875, + "learning_rate": 0.001998372205832309, + "loss": 0.291, + "step": 2820 + }, + { + "epoch": 0.024487634655949168, + "grad_norm": 0.173828125, + "learning_rate": 0.0019983704161126064, + "loss": 0.25, + "step": 2821 + }, + { + "epoch": 0.024496315136153333, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019983686254104595, + "loss": 0.25, + "step": 2822 + }, + { + "epoch": 0.024504995616357498, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019983668337258697, + "loss": 0.2148, + "step": 2823 + }, + { + "epoch": 0.024513676096561663, + "grad_norm": 0.078125, + "learning_rate": 0.00199836504105884, + "loss": 0.2383, + "step": 2824 + }, + { + "epoch": 0.024522356576765828, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019983632474093716, + "loss": 0.1787, + "step": 2825 + }, + { + "epoch": 0.024531037056969993, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019983614527774667, + "loss": 0.1934, + "step": 2826 + }, + { + "epoch": 0.024539717537174155, + "grad_norm": 0.1884765625, + "learning_rate": 0.001998359657163127, + "loss": 0.2148, + "step": 2827 + }, + { + "epoch": 0.02454839801737832, + "grad_norm": 0.123046875, + "learning_rate": 0.001998357860566355, + "loss": 0.2637, + "step": 2828 + }, + { + "epoch": 0.024557078497582485, + "grad_norm": 0.11669921875, + "learning_rate": 0.001998356062987152, + "loss": 0.2793, + "step": 2829 + }, + { + "epoch": 0.02456575897778665, + "grad_norm": 0.072265625, + "learning_rate": 0.0019983542644255205, + "loss": 0.2344, + "step": 2830 + }, + { + "epoch": 0.024574439457990815, + "grad_norm": 0.10107421875, + "learning_rate": 0.001998352464881462, + "loss": 0.2227, + "step": 2831 + }, + { + "epoch": 0.02458311993819498, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019983506643549793, + "loss": 0.2041, + "step": 2832 + }, + { + "epoch": 0.024591800418399146, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019983488628460733, + "loss": 0.208, + "step": 2833 + }, + { + "epoch": 0.02460048089860331, + "grad_norm": 0.1328125, + "learning_rate": 0.001998347060354747, + "loss": 0.249, + "step": 2834 + }, + { + "epoch": 0.024609161378807476, + "grad_norm": 0.0859375, + "learning_rate": 0.001998345256881001, + "loss": 0.1973, + "step": 2835 + }, + { + "epoch": 0.02461784185901164, + "grad_norm": 0.0732421875, + "learning_rate": 0.001998343452424839, + "loss": 0.2285, + "step": 2836 + }, + { + "epoch": 0.024626522339215806, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019983416469862617, + "loss": 0.1758, + "step": 2837 + }, + { + "epoch": 0.02463520281941997, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019983398405652715, + "loss": 0.2266, + "step": 2838 + }, + { + "epoch": 0.024643883299624136, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019983380331618705, + "loss": 0.252, + "step": 2839 + }, + { + "epoch": 0.0246525637798283, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019983362247760605, + "loss": 0.1973, + "step": 2840 + }, + { + "epoch": 0.024661244260032467, + "grad_norm": 0.083984375, + "learning_rate": 0.001998334415407843, + "loss": 0.1562, + "step": 2841 + }, + { + "epoch": 0.024669924740236628, + "grad_norm": 0.07275390625, + "learning_rate": 0.001998332605057221, + "loss": 0.1953, + "step": 2842 + }, + { + "epoch": 0.024678605220440793, + "grad_norm": 0.11328125, + "learning_rate": 0.0019983307937241957, + "loss": 0.2617, + "step": 2843 + }, + { + "epoch": 0.02468728570064496, + "grad_norm": 0.05908203125, + "learning_rate": 0.0019983289814087695, + "loss": 0.1523, + "step": 2844 + }, + { + "epoch": 0.024695966180849124, + "grad_norm": 0.263671875, + "learning_rate": 0.0019983271681109447, + "loss": 0.1855, + "step": 2845 + }, + { + "epoch": 0.02470464666105329, + "grad_norm": 0.0791015625, + "learning_rate": 0.001998325353830722, + "loss": 0.1553, + "step": 2846 + }, + { + "epoch": 0.024713327141257454, + "grad_norm": 0.205078125, + "learning_rate": 0.0019983235385681044, + "loss": 0.2148, + "step": 2847 + }, + { + "epoch": 0.02472200762146162, + "grad_norm": 0.3671875, + "learning_rate": 0.001998321722323094, + "loss": 0.2148, + "step": 2848 + }, + { + "epoch": 0.024730688101665784, + "grad_norm": 0.0908203125, + "learning_rate": 0.001998319905095692, + "loss": 0.2344, + "step": 2849 + }, + { + "epoch": 0.02473936858186995, + "grad_norm": 0.3125, + "learning_rate": 0.001998318086885901, + "loss": 0.2598, + "step": 2850 + }, + { + "epoch": 0.024748049062074114, + "grad_norm": 0.1845703125, + "learning_rate": 0.001998316267693723, + "loss": 0.25, + "step": 2851 + }, + { + "epoch": 0.02475672954227828, + "grad_norm": 0.0791015625, + "learning_rate": 0.00199831444751916, + "loss": 0.1719, + "step": 2852 + }, + { + "epoch": 0.024765410022482445, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019983126263622138, + "loss": 0.3223, + "step": 2853 + }, + { + "epoch": 0.02477409050268661, + "grad_norm": 0.12060546875, + "learning_rate": 0.001998310804222886, + "loss": 0.1758, + "step": 2854 + }, + { + "epoch": 0.024782770982890775, + "grad_norm": 0.197265625, + "learning_rate": 0.001998308981101179, + "loss": 0.2363, + "step": 2855 + }, + { + "epoch": 0.02479145146309494, + "grad_norm": 0.21875, + "learning_rate": 0.001998307156997095, + "loss": 0.2148, + "step": 2856 + }, + { + "epoch": 0.0248001319432991, + "grad_norm": 0.1435546875, + "learning_rate": 0.001998305331910636, + "loss": 0.1699, + "step": 2857 + }, + { + "epoch": 0.024808812423503267, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019983035058418032, + "loss": 0.1963, + "step": 2858 + }, + { + "epoch": 0.024817492903707432, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019983016787905998, + "loss": 0.2734, + "step": 2859 + }, + { + "epoch": 0.024826173383911597, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019982998507570267, + "loss": 0.1289, + "step": 2860 + }, + { + "epoch": 0.024834853864115762, + "grad_norm": 0.279296875, + "learning_rate": 0.0019982980217410867, + "loss": 0.2139, + "step": 2861 + }, + { + "epoch": 0.024843534344319927, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019982961917427815, + "loss": 0.209, + "step": 2862 + }, + { + "epoch": 0.024852214824524092, + "grad_norm": 0.126953125, + "learning_rate": 0.0019982943607621127, + "loss": 0.1885, + "step": 2863 + }, + { + "epoch": 0.024860895304728257, + "grad_norm": 0.177734375, + "learning_rate": 0.001998292528799083, + "loss": 0.2246, + "step": 2864 + }, + { + "epoch": 0.024869575784932423, + "grad_norm": 0.1767578125, + "learning_rate": 0.001998290695853694, + "loss": 0.2031, + "step": 2865 + }, + { + "epoch": 0.024878256265136588, + "grad_norm": 0.419921875, + "learning_rate": 0.0019982888619259477, + "loss": 0.1895, + "step": 2866 + }, + { + "epoch": 0.024886936745340753, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019982870270158462, + "loss": 0.2227, + "step": 2867 + }, + { + "epoch": 0.024895617225544918, + "grad_norm": 0.0712890625, + "learning_rate": 0.001998285191123392, + "loss": 0.2129, + "step": 2868 + }, + { + "epoch": 0.024904297705749083, + "grad_norm": 0.2265625, + "learning_rate": 0.001998283354248586, + "loss": 0.2354, + "step": 2869 + }, + { + "epoch": 0.024912978185953248, + "grad_norm": 0.1015625, + "learning_rate": 0.001998281516391431, + "loss": 0.207, + "step": 2870 + }, + { + "epoch": 0.024921658666157413, + "grad_norm": 0.087890625, + "learning_rate": 0.001998279677551929, + "loss": 0.2637, + "step": 2871 + }, + { + "epoch": 0.02493033914636158, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019982778377300816, + "loss": 0.2656, + "step": 2872 + }, + { + "epoch": 0.02493901962656574, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019982759969258915, + "loss": 0.2266, + "step": 2873 + }, + { + "epoch": 0.024947700106769905, + "grad_norm": 0.0703125, + "learning_rate": 0.0019982741551393597, + "loss": 0.1846, + "step": 2874 + }, + { + "epoch": 0.02495638058697407, + "grad_norm": 0.1689453125, + "learning_rate": 0.001998272312370489, + "loss": 0.2266, + "step": 2875 + }, + { + "epoch": 0.024965061067178235, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019982704686192813, + "loss": 0.1787, + "step": 2876 + }, + { + "epoch": 0.0249737415473824, + "grad_norm": 0.08203125, + "learning_rate": 0.0019982686238857387, + "loss": 0.2227, + "step": 2877 + }, + { + "epoch": 0.024982422027586566, + "grad_norm": 0.185546875, + "learning_rate": 0.0019982667781698626, + "loss": 0.2197, + "step": 2878 + }, + { + "epoch": 0.02499110250779073, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019982649314716555, + "loss": 0.2227, + "step": 2879 + }, + { + "epoch": 0.024999782987994896, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019982630837911196, + "loss": 0.3633, + "step": 2880 + }, + { + "epoch": 0.02500846346819906, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019982612351282566, + "loss": 0.2285, + "step": 2881 + }, + { + "epoch": 0.025017143948403226, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019982593854830683, + "loss": 0.2168, + "step": 2882 + }, + { + "epoch": 0.02502582442860739, + "grad_norm": 0.1015625, + "learning_rate": 0.0019982575348555572, + "loss": 0.2832, + "step": 2883 + }, + { + "epoch": 0.025034504908811556, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019982556832457256, + "loss": 0.2178, + "step": 2884 + }, + { + "epoch": 0.02504318538901572, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019982538306535748, + "loss": 0.1826, + "step": 2885 + }, + { + "epoch": 0.025051865869219887, + "grad_norm": 0.0849609375, + "learning_rate": 0.001998251977079107, + "loss": 0.2012, + "step": 2886 + }, + { + "epoch": 0.02506054634942405, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019982501225223243, + "loss": 0.2422, + "step": 2887 + }, + { + "epoch": 0.025069226829628213, + "grad_norm": 0.08154296875, + "learning_rate": 0.001998248266983229, + "loss": 0.2578, + "step": 2888 + }, + { + "epoch": 0.02507790730983238, + "grad_norm": 0.08984375, + "learning_rate": 0.001998246410461823, + "loss": 0.1514, + "step": 2889 + }, + { + "epoch": 0.025086587790036544, + "grad_norm": 0.1953125, + "learning_rate": 0.0019982445529581074, + "loss": 0.2266, + "step": 2890 + }, + { + "epoch": 0.02509526827024071, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019982426944720856, + "loss": 0.2227, + "step": 2891 + }, + { + "epoch": 0.025103948750444874, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019982408350037594, + "loss": 0.1748, + "step": 2892 + }, + { + "epoch": 0.02511262923064904, + "grad_norm": 0.1875, + "learning_rate": 0.00199823897455313, + "loss": 0.1689, + "step": 2893 + }, + { + "epoch": 0.025121309710853204, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019982371131202, + "loss": 0.1709, + "step": 2894 + }, + { + "epoch": 0.02512999019105737, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019982352507049717, + "loss": 0.1875, + "step": 2895 + }, + { + "epoch": 0.025138670671261534, + "grad_norm": 0.11767578125, + "learning_rate": 0.001998233387307447, + "loss": 0.2207, + "step": 2896 + }, + { + "epoch": 0.0251473511514657, + "grad_norm": 0.080078125, + "learning_rate": 0.0019982315229276275, + "loss": 0.2324, + "step": 2897 + }, + { + "epoch": 0.025156031631669865, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019982296575655153, + "loss": 0.2344, + "step": 2898 + }, + { + "epoch": 0.02516471211187403, + "grad_norm": 0.1484375, + "learning_rate": 0.0019982277912211125, + "loss": 0.1992, + "step": 2899 + }, + { + "epoch": 0.025173392592078195, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019982259238944216, + "loss": 0.2266, + "step": 2900 + }, + { + "epoch": 0.02518207307228236, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019982240555854445, + "loss": 0.1924, + "step": 2901 + }, + { + "epoch": 0.025190753552486525, + "grad_norm": 0.150390625, + "learning_rate": 0.001998222186294183, + "loss": 0.2119, + "step": 2902 + }, + { + "epoch": 0.02519943403269069, + "grad_norm": 0.0732421875, + "learning_rate": 0.001998220316020639, + "loss": 0.2324, + "step": 2903 + }, + { + "epoch": 0.025208114512894852, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019982184447648148, + "loss": 0.1973, + "step": 2904 + }, + { + "epoch": 0.025216794993099017, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019982165725267124, + "loss": 0.1895, + "step": 2905 + }, + { + "epoch": 0.025225475473303182, + "grad_norm": 0.220703125, + "learning_rate": 0.001998214699306334, + "loss": 0.1426, + "step": 2906 + }, + { + "epoch": 0.025234155953507347, + "grad_norm": 0.228515625, + "learning_rate": 0.001998212825103681, + "loss": 0.1875, + "step": 2907 + }, + { + "epoch": 0.025242836433711512, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019982109499187568, + "loss": 0.2119, + "step": 2908 + }, + { + "epoch": 0.025251516913915677, + "grad_norm": 0.08251953125, + "learning_rate": 0.001998209073751562, + "loss": 0.1943, + "step": 2909 + }, + { + "epoch": 0.025260197394119843, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019982071966020993, + "loss": 0.2383, + "step": 2910 + }, + { + "epoch": 0.025268877874324008, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019982053184703706, + "loss": 0.25, + "step": 2911 + }, + { + "epoch": 0.025277558354528173, + "grad_norm": 0.13671875, + "learning_rate": 0.0019982034393563786, + "loss": 0.2402, + "step": 2912 + }, + { + "epoch": 0.025286238834732338, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019982015592601246, + "loss": 0.2539, + "step": 2913 + }, + { + "epoch": 0.025294919314936503, + "grad_norm": 0.123046875, + "learning_rate": 0.0019981996781816107, + "loss": 0.2266, + "step": 2914 + }, + { + "epoch": 0.025303599795140668, + "grad_norm": 0.06494140625, + "learning_rate": 0.001998197796120839, + "loss": 0.2266, + "step": 2915 + }, + { + "epoch": 0.025312280275344833, + "grad_norm": 0.111328125, + "learning_rate": 0.001998195913077812, + "loss": 0.2676, + "step": 2916 + }, + { + "epoch": 0.025320960755549, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019981940290525312, + "loss": 0.1758, + "step": 2917 + }, + { + "epoch": 0.025329641235753163, + "grad_norm": 0.1484375, + "learning_rate": 0.0019981921440449988, + "loss": 0.2148, + "step": 2918 + }, + { + "epoch": 0.025338321715957325, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019981902580552173, + "loss": 0.2363, + "step": 2919 + }, + { + "epoch": 0.02534700219616149, + "grad_norm": 0.10546875, + "learning_rate": 0.001998188371083188, + "loss": 0.1895, + "step": 2920 + }, + { + "epoch": 0.025355682676365655, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019981864831289144, + "loss": 0.1787, + "step": 2921 + }, + { + "epoch": 0.02536436315656982, + "grad_norm": 0.158203125, + "learning_rate": 0.0019981845941923967, + "loss": 0.2471, + "step": 2922 + }, + { + "epoch": 0.025373043636773986, + "grad_norm": 0.08447265625, + "learning_rate": 0.001998182704273638, + "loss": 0.2109, + "step": 2923 + }, + { + "epoch": 0.02538172411697815, + "grad_norm": 0.078125, + "learning_rate": 0.00199818081337264, + "loss": 0.1895, + "step": 2924 + }, + { + "epoch": 0.025390404597182316, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019981789214894054, + "loss": 0.1602, + "step": 2925 + }, + { + "epoch": 0.02539908507738648, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019981770286239355, + "loss": 0.2637, + "step": 2926 + }, + { + "epoch": 0.025407765557590646, + "grad_norm": 0.185546875, + "learning_rate": 0.0019981751347762327, + "loss": 0.2266, + "step": 2927 + }, + { + "epoch": 0.02541644603779481, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019981732399462987, + "loss": 0.2119, + "step": 2928 + }, + { + "epoch": 0.025425126517998976, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019981713441341365, + "loss": 0.2578, + "step": 2929 + }, + { + "epoch": 0.02543380699820314, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019981694473397474, + "loss": 0.1826, + "step": 2930 + }, + { + "epoch": 0.025442487478407307, + "grad_norm": 0.068359375, + "learning_rate": 0.0019981675495631336, + "loss": 0.1553, + "step": 2931 + }, + { + "epoch": 0.02545116795861147, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019981656508042977, + "loss": 0.2324, + "step": 2932 + }, + { + "epoch": 0.025459848438815637, + "grad_norm": 0.08740234375, + "learning_rate": 0.001998163751063241, + "loss": 0.21, + "step": 2933 + }, + { + "epoch": 0.0254685289190198, + "grad_norm": 0.083984375, + "learning_rate": 0.001998161850339966, + "loss": 0.209, + "step": 2934 + }, + { + "epoch": 0.025477209399223964, + "grad_norm": 0.11767578125, + "learning_rate": 0.001998159948634475, + "loss": 0.1865, + "step": 2935 + }, + { + "epoch": 0.02548588987942813, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019981580459467693, + "loss": 0.1689, + "step": 2936 + }, + { + "epoch": 0.025494570359632294, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019981561422768514, + "loss": 0.2227, + "step": 2937 + }, + { + "epoch": 0.02550325083983646, + "grad_norm": 0.09326171875, + "learning_rate": 0.001998154237624724, + "loss": 0.2363, + "step": 2938 + }, + { + "epoch": 0.025511931320040624, + "grad_norm": 0.3359375, + "learning_rate": 0.001998152331990388, + "loss": 0.1934, + "step": 2939 + }, + { + "epoch": 0.02552061180024479, + "grad_norm": 0.302734375, + "learning_rate": 0.0019981504253738466, + "loss": 0.2188, + "step": 2940 + }, + { + "epoch": 0.025529292280448954, + "grad_norm": 0.46484375, + "learning_rate": 0.001998148517775101, + "loss": 0.1982, + "step": 2941 + }, + { + "epoch": 0.02553797276065312, + "grad_norm": 0.4765625, + "learning_rate": 0.001998146609194154, + "loss": 0.2383, + "step": 2942 + }, + { + "epoch": 0.025546653240857285, + "grad_norm": 0.1474609375, + "learning_rate": 0.001998144699631007, + "loss": 0.1973, + "step": 2943 + }, + { + "epoch": 0.02555533372106145, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019981427890856628, + "loss": 0.1992, + "step": 2944 + }, + { + "epoch": 0.025564014201265615, + "grad_norm": 0.25, + "learning_rate": 0.0019981408775581228, + "loss": 0.1396, + "step": 2945 + }, + { + "epoch": 0.02557269468146978, + "grad_norm": 0.35546875, + "learning_rate": 0.0019981389650483897, + "loss": 0.207, + "step": 2946 + }, + { + "epoch": 0.025581375161673945, + "grad_norm": 0.5703125, + "learning_rate": 0.0019981370515564654, + "loss": 0.2988, + "step": 2947 + }, + { + "epoch": 0.02559005564187811, + "grad_norm": 0.283203125, + "learning_rate": 0.0019981351370823514, + "loss": 0.1934, + "step": 2948 + }, + { + "epoch": 0.025598736122082275, + "grad_norm": 0.337890625, + "learning_rate": 0.0019981332216260504, + "loss": 0.2061, + "step": 2949 + }, + { + "epoch": 0.025607416602286437, + "grad_norm": 0.09765625, + "learning_rate": 0.001998131305187565, + "loss": 0.1904, + "step": 2950 + }, + { + "epoch": 0.025616097082490602, + "grad_norm": 0.271484375, + "learning_rate": 0.0019981293877668962, + "loss": 0.1719, + "step": 2951 + }, + { + "epoch": 0.025624777562694767, + "grad_norm": 0.251953125, + "learning_rate": 0.001998127469364047, + "loss": 0.2266, + "step": 2952 + }, + { + "epoch": 0.025633458042898932, + "grad_norm": 0.0908203125, + "learning_rate": 0.001998125549979019, + "loss": 0.2129, + "step": 2953 + }, + { + "epoch": 0.025642138523103097, + "grad_norm": 0.17578125, + "learning_rate": 0.0019981236296118137, + "loss": 0.3711, + "step": 2954 + }, + { + "epoch": 0.025650819003307263, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019981217082624347, + "loss": 0.2148, + "step": 2955 + }, + { + "epoch": 0.025659499483511428, + "grad_norm": 0.2734375, + "learning_rate": 0.001998119785930883, + "loss": 0.1963, + "step": 2956 + }, + { + "epoch": 0.025668179963715593, + "grad_norm": 0.1337890625, + "learning_rate": 0.001998117862617161, + "loss": 0.1826, + "step": 2957 + }, + { + "epoch": 0.025676860443919758, + "grad_norm": 0.1376953125, + "learning_rate": 0.001998115938321271, + "loss": 0.2451, + "step": 2958 + }, + { + "epoch": 0.025685540924123923, + "grad_norm": 0.296875, + "learning_rate": 0.0019981140130432146, + "loss": 0.2471, + "step": 2959 + }, + { + "epoch": 0.025694221404328088, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019981120867829947, + "loss": 0.1826, + "step": 2960 + }, + { + "epoch": 0.025702901884532253, + "grad_norm": 0.296875, + "learning_rate": 0.0019981101595406125, + "loss": 0.2139, + "step": 2961 + }, + { + "epoch": 0.02571158236473642, + "grad_norm": 0.08544921875, + "learning_rate": 0.00199810823131607, + "loss": 0.2061, + "step": 2962 + }, + { + "epoch": 0.025720262844940583, + "grad_norm": 0.0830078125, + "learning_rate": 0.001998106302109371, + "loss": 0.2119, + "step": 2963 + }, + { + "epoch": 0.02572894332514475, + "grad_norm": 0.185546875, + "learning_rate": 0.0019981043719205158, + "loss": 0.1816, + "step": 2964 + }, + { + "epoch": 0.02573762380534891, + "grad_norm": 0.15625, + "learning_rate": 0.001998102440749507, + "loss": 0.2363, + "step": 2965 + }, + { + "epoch": 0.025746304285553075, + "grad_norm": 0.0791015625, + "learning_rate": 0.001998100508596347, + "loss": 0.2324, + "step": 2966 + }, + { + "epoch": 0.02575498476575724, + "grad_norm": 0.1884765625, + "learning_rate": 0.001998098575461038, + "loss": 0.2363, + "step": 2967 + }, + { + "epoch": 0.025763665245961406, + "grad_norm": 0.255859375, + "learning_rate": 0.0019980966413435815, + "loss": 0.2188, + "step": 2968 + }, + { + "epoch": 0.02577234572616557, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019980947062439806, + "loss": 0.2266, + "step": 2969 + }, + { + "epoch": 0.025781026206369736, + "grad_norm": 0.134765625, + "learning_rate": 0.0019980927701622364, + "loss": 0.248, + "step": 2970 + }, + { + "epoch": 0.0257897066865739, + "grad_norm": 0.2734375, + "learning_rate": 0.0019980908330983513, + "loss": 0.2041, + "step": 2971 + }, + { + "epoch": 0.025798387166778066, + "grad_norm": 0.083984375, + "learning_rate": 0.001998088895052328, + "loss": 0.2305, + "step": 2972 + }, + { + "epoch": 0.02580706764698223, + "grad_norm": 0.1044921875, + "learning_rate": 0.001998086956024168, + "loss": 0.2422, + "step": 2973 + }, + { + "epoch": 0.025815748127186396, + "grad_norm": 0.15234375, + "learning_rate": 0.001998085016013873, + "loss": 0.2539, + "step": 2974 + }, + { + "epoch": 0.02582442860739056, + "grad_norm": 0.19140625, + "learning_rate": 0.0019980830750214464, + "loss": 0.209, + "step": 2975 + }, + { + "epoch": 0.025833109087594727, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019980811330468896, + "loss": 0.1992, + "step": 2976 + }, + { + "epoch": 0.02584178956779889, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019980791900902047, + "loss": 0.2188, + "step": 2977 + }, + { + "epoch": 0.025850470048003057, + "grad_norm": 0.1376953125, + "learning_rate": 0.001998077246151394, + "loss": 0.1963, + "step": 2978 + }, + { + "epoch": 0.025859150528207222, + "grad_norm": 0.091796875, + "learning_rate": 0.001998075301230459, + "loss": 0.2383, + "step": 2979 + }, + { + "epoch": 0.025867831008411387, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019980733553274033, + "loss": 0.1953, + "step": 2980 + }, + { + "epoch": 0.02587651148861555, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019980714084422272, + "loss": 0.2363, + "step": 2981 + }, + { + "epoch": 0.025885191968819714, + "grad_norm": 0.2255859375, + "learning_rate": 0.001998069460574934, + "loss": 0.1855, + "step": 2982 + }, + { + "epoch": 0.02589387244902388, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019980675117255257, + "loss": 0.248, + "step": 2983 + }, + { + "epoch": 0.025902552929228044, + "grad_norm": 0.08984375, + "learning_rate": 0.001998065561894004, + "loss": 0.208, + "step": 2984 + }, + { + "epoch": 0.02591123340943221, + "grad_norm": 0.236328125, + "learning_rate": 0.001998063611080371, + "loss": 0.2031, + "step": 2985 + }, + { + "epoch": 0.025919913889636374, + "grad_norm": 0.05908203125, + "learning_rate": 0.00199806165928463, + "loss": 0.168, + "step": 2986 + }, + { + "epoch": 0.02592859436984054, + "grad_norm": 0.0859375, + "learning_rate": 0.001998059706506782, + "loss": 0.2246, + "step": 2987 + }, + { + "epoch": 0.025937274850044705, + "grad_norm": 0.0927734375, + "learning_rate": 0.001998057752746829, + "loss": 0.1826, + "step": 2988 + }, + { + "epoch": 0.02594595533024887, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019980557980047737, + "loss": 0.1592, + "step": 2989 + }, + { + "epoch": 0.025954635810453035, + "grad_norm": 0.083984375, + "learning_rate": 0.001998053842280618, + "loss": 0.1836, + "step": 2990 + }, + { + "epoch": 0.0259633162906572, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019980518855743645, + "loss": 0.209, + "step": 2991 + }, + { + "epoch": 0.025971996770861365, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019980499278860146, + "loss": 0.2168, + "step": 2992 + }, + { + "epoch": 0.02598067725106553, + "grad_norm": 1.046875, + "learning_rate": 0.001998047969215571, + "loss": 0.2324, + "step": 2993 + }, + { + "epoch": 0.025989357731269695, + "grad_norm": 0.11376953125, + "learning_rate": 0.001998046009563035, + "loss": 0.2197, + "step": 2994 + }, + { + "epoch": 0.02599803821147386, + "grad_norm": 0.330078125, + "learning_rate": 0.00199804404892841, + "loss": 0.2109, + "step": 2995 + }, + { + "epoch": 0.026006718691678022, + "grad_norm": 0.2578125, + "learning_rate": 0.0019980420873116976, + "loss": 0.1963, + "step": 2996 + }, + { + "epoch": 0.026015399171882187, + "grad_norm": 0.482421875, + "learning_rate": 0.0019980401247128993, + "loss": 0.2305, + "step": 2997 + }, + { + "epoch": 0.026024079652086352, + "grad_norm": 0.185546875, + "learning_rate": 0.001998038161132018, + "loss": 0.1777, + "step": 2998 + }, + { + "epoch": 0.026032760132290517, + "grad_norm": 0.2216796875, + "learning_rate": 0.001998036196569056, + "loss": 0.2021, + "step": 2999 + }, + { + "epoch": 0.026041440612494682, + "grad_norm": 0.08935546875, + "learning_rate": 0.001998034231024015, + "loss": 0.2012, + "step": 3000 + }, + { + "epoch": 0.026050121092698848, + "grad_norm": 0.4765625, + "learning_rate": 0.0019980322644968973, + "loss": 0.2773, + "step": 3001 + }, + { + "epoch": 0.026058801572903013, + "grad_norm": 0.59765625, + "learning_rate": 0.001998030296987705, + "loss": 0.2471, + "step": 3002 + }, + { + "epoch": 0.026067482053107178, + "grad_norm": 0.119140625, + "learning_rate": 0.00199802832849644, + "loss": 0.2334, + "step": 3003 + }, + { + "epoch": 0.026076162533311343, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019980263590231046, + "loss": 0.1885, + "step": 3004 + }, + { + "epoch": 0.026084843013515508, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019980243885677016, + "loss": 0.2188, + "step": 3005 + }, + { + "epoch": 0.026093523493719673, + "grad_norm": 0.087890625, + "learning_rate": 0.001998022417130232, + "loss": 0.2051, + "step": 3006 + }, + { + "epoch": 0.02610220397392384, + "grad_norm": 0.2060546875, + "learning_rate": 0.001998020444710699, + "loss": 0.2539, + "step": 3007 + }, + { + "epoch": 0.026110884454128003, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019980184713091044, + "loss": 0.1709, + "step": 3008 + }, + { + "epoch": 0.02611956493433217, + "grad_norm": 0.44140625, + "learning_rate": 0.00199801649692545, + "loss": 0.2354, + "step": 3009 + }, + { + "epoch": 0.026128245414536334, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019980145215597383, + "loss": 0.2109, + "step": 3010 + }, + { + "epoch": 0.026136925894740495, + "grad_norm": 0.8203125, + "learning_rate": 0.001998012545211972, + "loss": 0.2617, + "step": 3011 + }, + { + "epoch": 0.02614560637494466, + "grad_norm": 0.1025390625, + "learning_rate": 0.001998010567882152, + "loss": 0.1914, + "step": 3012 + }, + { + "epoch": 0.026154286855148826, + "grad_norm": 0.1484375, + "learning_rate": 0.001998008589570281, + "loss": 0.2363, + "step": 3013 + }, + { + "epoch": 0.02616296733535299, + "grad_norm": 0.1845703125, + "learning_rate": 0.001998006610276362, + "loss": 0.1699, + "step": 3014 + }, + { + "epoch": 0.026171647815557156, + "grad_norm": 0.1669921875, + "learning_rate": 0.001998004630000396, + "loss": 0.1719, + "step": 3015 + }, + { + "epoch": 0.02618032829576132, + "grad_norm": 0.453125, + "learning_rate": 0.001998002648742386, + "loss": 0.2461, + "step": 3016 + }, + { + "epoch": 0.026189008775965486, + "grad_norm": 0.1708984375, + "learning_rate": 0.001998000666502333, + "loss": 0.2539, + "step": 3017 + }, + { + "epoch": 0.02619768925616965, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019979986832802405, + "loss": 0.2109, + "step": 3018 + }, + { + "epoch": 0.026206369736373816, + "grad_norm": 0.1005859375, + "learning_rate": 0.00199799669907611, + "loss": 0.252, + "step": 3019 + }, + { + "epoch": 0.02621505021657798, + "grad_norm": 0.302734375, + "learning_rate": 0.001997994713889944, + "loss": 0.2461, + "step": 3020 + }, + { + "epoch": 0.026223730696782147, + "grad_norm": 0.216796875, + "learning_rate": 0.0019979927277217445, + "loss": 0.1895, + "step": 3021 + }, + { + "epoch": 0.02623241117698631, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019979907405715137, + "loss": 0.2188, + "step": 3022 + }, + { + "epoch": 0.026241091657190477, + "grad_norm": 0.14453125, + "learning_rate": 0.0019979887524392533, + "loss": 0.1807, + "step": 3023 + }, + { + "epoch": 0.026249772137394642, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019979867633249664, + "loss": 0.1992, + "step": 3024 + }, + { + "epoch": 0.026258452617598807, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019979847732286542, + "loss": 0.2207, + "step": 3025 + }, + { + "epoch": 0.026267133097802972, + "grad_norm": 0.296875, + "learning_rate": 0.00199798278215032, + "loss": 0.1924, + "step": 3026 + }, + { + "epoch": 0.026275813578007134, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019979807900899647, + "loss": 0.1953, + "step": 3027 + }, + { + "epoch": 0.0262844940582113, + "grad_norm": 0.09912109375, + "learning_rate": 0.001997978797047591, + "loss": 0.1982, + "step": 3028 + }, + { + "epoch": 0.026293174538415464, + "grad_norm": 0.197265625, + "learning_rate": 0.0019979768030232016, + "loss": 0.2441, + "step": 3029 + }, + { + "epoch": 0.02630185501861963, + "grad_norm": 0.1962890625, + "learning_rate": 0.001997974808016798, + "loss": 0.1455, + "step": 3030 + }, + { + "epoch": 0.026310535498823794, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019979728120283827, + "loss": 0.2168, + "step": 3031 + }, + { + "epoch": 0.02631921597902796, + "grad_norm": 0.1640625, + "learning_rate": 0.0019979708150579577, + "loss": 0.1914, + "step": 3032 + }, + { + "epoch": 0.026327896459232124, + "grad_norm": 0.53125, + "learning_rate": 0.0019979688171055257, + "loss": 0.2119, + "step": 3033 + }, + { + "epoch": 0.02633657693943629, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019979668181710885, + "loss": 0.2148, + "step": 3034 + }, + { + "epoch": 0.026345257419640455, + "grad_norm": 0.10546875, + "learning_rate": 0.001997964818254648, + "loss": 0.2852, + "step": 3035 + }, + { + "epoch": 0.02635393789984462, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019979628173562064, + "loss": 0.2383, + "step": 3036 + }, + { + "epoch": 0.026362618380048785, + "grad_norm": 0.1865234375, + "learning_rate": 0.001997960815475767, + "loss": 0.2051, + "step": 3037 + }, + { + "epoch": 0.02637129886025295, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019979588126133306, + "loss": 0.2109, + "step": 3038 + }, + { + "epoch": 0.026379979340457115, + "grad_norm": 0.296875, + "learning_rate": 0.0019979568087689, + "loss": 0.1953, + "step": 3039 + }, + { + "epoch": 0.02638865982066128, + "grad_norm": 0.498046875, + "learning_rate": 0.001997954803942477, + "loss": 0.2246, + "step": 3040 + }, + { + "epoch": 0.026397340300865445, + "grad_norm": 0.1953125, + "learning_rate": 0.0019979527981340644, + "loss": 0.1875, + "step": 3041 + }, + { + "epoch": 0.026406020781069607, + "grad_norm": 0.373046875, + "learning_rate": 0.0019979507913436643, + "loss": 0.2051, + "step": 3042 + }, + { + "epoch": 0.026414701261273772, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019979487835712784, + "loss": 0.1816, + "step": 3043 + }, + { + "epoch": 0.026423381741477937, + "grad_norm": 0.12158203125, + "learning_rate": 0.001997946774816909, + "loss": 0.2197, + "step": 3044 + }, + { + "epoch": 0.026432062221682102, + "grad_norm": 0.392578125, + "learning_rate": 0.001997944765080559, + "loss": 0.2539, + "step": 3045 + }, + { + "epoch": 0.026440742701886268, + "grad_norm": 0.287109375, + "learning_rate": 0.0019979427543622297, + "loss": 0.1699, + "step": 3046 + }, + { + "epoch": 0.026449423182090433, + "grad_norm": 0.166015625, + "learning_rate": 0.001997940742661924, + "loss": 0.3574, + "step": 3047 + }, + { + "epoch": 0.026458103662294598, + "grad_norm": 0.185546875, + "learning_rate": 0.001997938729979644, + "loss": 0.2422, + "step": 3048 + }, + { + "epoch": 0.026466784142498763, + "grad_norm": 0.2119140625, + "learning_rate": 0.001997936716315391, + "loss": 0.2422, + "step": 3049 + }, + { + "epoch": 0.026475464622702928, + "grad_norm": 0.296875, + "learning_rate": 0.0019979347016691688, + "loss": 0.2119, + "step": 3050 + }, + { + "epoch": 0.026484145102907093, + "grad_norm": 0.0703125, + "learning_rate": 0.001997932686040978, + "loss": 0.168, + "step": 3051 + }, + { + "epoch": 0.02649282558311126, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019979306694308217, + "loss": 0.2578, + "step": 3052 + }, + { + "epoch": 0.026501506063315423, + "grad_norm": 0.13671875, + "learning_rate": 0.001997928651838702, + "loss": 0.1875, + "step": 3053 + }, + { + "epoch": 0.02651018654351959, + "grad_norm": 0.384765625, + "learning_rate": 0.001997926633264621, + "loss": 0.1797, + "step": 3054 + }, + { + "epoch": 0.026518867023723754, + "grad_norm": 0.17578125, + "learning_rate": 0.001997924613708581, + "loss": 0.1797, + "step": 3055 + }, + { + "epoch": 0.02652754750392792, + "grad_norm": 0.369140625, + "learning_rate": 0.001997922593170584, + "loss": 0.1846, + "step": 3056 + }, + { + "epoch": 0.026536227984132084, + "grad_norm": 0.29296875, + "learning_rate": 0.001997920571650632, + "loss": 0.2051, + "step": 3057 + }, + { + "epoch": 0.026544908464336246, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019979185491487282, + "loss": 0.2207, + "step": 3058 + }, + { + "epoch": 0.02655358894454041, + "grad_norm": 0.17578125, + "learning_rate": 0.001997916525664874, + "loss": 0.2031, + "step": 3059 + }, + { + "epoch": 0.026562269424744576, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019979145011990713, + "loss": 0.1768, + "step": 3060 + }, + { + "epoch": 0.02657094990494874, + "grad_norm": 0.265625, + "learning_rate": 0.001997912475751323, + "loss": 0.252, + "step": 3061 + }, + { + "epoch": 0.026579630385152906, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019979104493216314, + "loss": 0.1807, + "step": 3062 + }, + { + "epoch": 0.02658831086535707, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019979084219099983, + "loss": 0.209, + "step": 3063 + }, + { + "epoch": 0.026596991345561236, + "grad_norm": 0.1162109375, + "learning_rate": 0.001997906393516426, + "loss": 0.2266, + "step": 3064 + }, + { + "epoch": 0.0266056718257654, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019979043641409166, + "loss": 0.2246, + "step": 3065 + }, + { + "epoch": 0.026614352305969566, + "grad_norm": 0.41015625, + "learning_rate": 0.001997902333783473, + "loss": 0.2324, + "step": 3066 + }, + { + "epoch": 0.02662303278617373, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019979003024440966, + "loss": 0.2129, + "step": 3067 + }, + { + "epoch": 0.026631713266377897, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019978982701227897, + "loss": 0.208, + "step": 3068 + }, + { + "epoch": 0.026640393746582062, + "grad_norm": 0.81640625, + "learning_rate": 0.0019978962368195547, + "loss": 0.8203, + "step": 3069 + }, + { + "epoch": 0.026649074226786227, + "grad_norm": 0.16796875, + "learning_rate": 0.0019978942025343943, + "loss": 0.2061, + "step": 3070 + }, + { + "epoch": 0.026657754706990392, + "grad_norm": 0.41796875, + "learning_rate": 0.0019978921672673096, + "loss": 0.167, + "step": 3071 + }, + { + "epoch": 0.026666435187194557, + "grad_norm": 0.408203125, + "learning_rate": 0.0019978901310183043, + "loss": 0.1836, + "step": 3072 + }, + { + "epoch": 0.02667511566739872, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019978880937873795, + "loss": 0.2285, + "step": 3073 + }, + { + "epoch": 0.026683796147602884, + "grad_norm": 0.60546875, + "learning_rate": 0.001997886055574538, + "loss": 0.2734, + "step": 3074 + }, + { + "epoch": 0.02669247662780705, + "grad_norm": 0.27734375, + "learning_rate": 0.0019978840163797813, + "loss": 0.2891, + "step": 3075 + }, + { + "epoch": 0.026701157108011214, + "grad_norm": 0.1953125, + "learning_rate": 0.0019978819762031122, + "loss": 0.2373, + "step": 3076 + }, + { + "epoch": 0.02670983758821538, + "grad_norm": 0.09375, + "learning_rate": 0.0019978799350445337, + "loss": 0.2188, + "step": 3077 + }, + { + "epoch": 0.026718518068419544, + "grad_norm": 0.345703125, + "learning_rate": 0.0019978778929040466, + "loss": 0.2734, + "step": 3078 + }, + { + "epoch": 0.02672719854862371, + "grad_norm": 0.337890625, + "learning_rate": 0.0019978758497816535, + "loss": 0.2324, + "step": 3079 + }, + { + "epoch": 0.026735879028827875, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019978738056773567, + "loss": 0.2422, + "step": 3080 + }, + { + "epoch": 0.02674455950903204, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019978717605911595, + "loss": 0.1973, + "step": 3081 + }, + { + "epoch": 0.026753239989236205, + "grad_norm": 0.15625, + "learning_rate": 0.0019978697145230625, + "loss": 0.3438, + "step": 3082 + }, + { + "epoch": 0.02676192046944037, + "grad_norm": 0.1259765625, + "learning_rate": 0.001997867667473069, + "loss": 0.1914, + "step": 3083 + }, + { + "epoch": 0.026770600949644535, + "grad_norm": 0.185546875, + "learning_rate": 0.001997865619441181, + "loss": 0.2412, + "step": 3084 + }, + { + "epoch": 0.0267792814298487, + "grad_norm": 0.279296875, + "learning_rate": 0.0019978635704274, + "loss": 0.2197, + "step": 3085 + }, + { + "epoch": 0.026787961910052865, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019978615204317295, + "loss": 0.2266, + "step": 3086 + }, + { + "epoch": 0.02679664239025703, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019978594694541707, + "loss": 0.1797, + "step": 3087 + }, + { + "epoch": 0.026805322870461192, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019978574174947267, + "loss": 0.2383, + "step": 3088 + }, + { + "epoch": 0.026814003350665357, + "grad_norm": 0.2255859375, + "learning_rate": 0.001997855364553399, + "loss": 0.2227, + "step": 3089 + }, + { + "epoch": 0.026822683830869522, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019978533106301904, + "loss": 0.2285, + "step": 3090 + }, + { + "epoch": 0.026831364311073688, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019978512557251027, + "loss": 0.2578, + "step": 3091 + }, + { + "epoch": 0.026840044791277853, + "grad_norm": 0.20703125, + "learning_rate": 0.001997849199838138, + "loss": 0.2695, + "step": 3092 + }, + { + "epoch": 0.026848725271482018, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019978471429693, + "loss": 0.168, + "step": 3093 + }, + { + "epoch": 0.026857405751686183, + "grad_norm": 0.1962890625, + "learning_rate": 0.001997845085118589, + "loss": 0.1953, + "step": 3094 + }, + { + "epoch": 0.026866086231890348, + "grad_norm": 0.134765625, + "learning_rate": 0.0019978430262860085, + "loss": 0.3535, + "step": 3095 + }, + { + "epoch": 0.026874766712094513, + "grad_norm": 0.11962890625, + "learning_rate": 0.00199784096647156, + "loss": 0.2109, + "step": 3096 + }, + { + "epoch": 0.02688344719229868, + "grad_norm": 0.1015625, + "learning_rate": 0.001997838905675246, + "loss": 0.1875, + "step": 3097 + }, + { + "epoch": 0.026892127672502843, + "grad_norm": 0.142578125, + "learning_rate": 0.0019978368438970696, + "loss": 0.1914, + "step": 3098 + }, + { + "epoch": 0.02690080815270701, + "grad_norm": 0.119140625, + "learning_rate": 0.0019978347811370318, + "loss": 0.1973, + "step": 3099 + }, + { + "epoch": 0.026909488632911174, + "grad_norm": 0.30078125, + "learning_rate": 0.0019978327173951357, + "loss": 0.25, + "step": 3100 + }, + { + "epoch": 0.02691816911311534, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019978306526713826, + "loss": 0.2129, + "step": 3101 + }, + { + "epoch": 0.026926849593319504, + "grad_norm": 0.1083984375, + "learning_rate": 0.001997828586965776, + "loss": 0.2061, + "step": 3102 + }, + { + "epoch": 0.02693553007352367, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019978265202783172, + "loss": 0.1846, + "step": 3103 + }, + { + "epoch": 0.02694421055372783, + "grad_norm": 0.2314453125, + "learning_rate": 0.001997824452609009, + "loss": 0.2891, + "step": 3104 + }, + { + "epoch": 0.026952891033931996, + "grad_norm": 0.1318359375, + "learning_rate": 0.001997822383957853, + "loss": 0.1865, + "step": 3105 + }, + { + "epoch": 0.02696157151413616, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019978203143248526, + "loss": 0.1494, + "step": 3106 + }, + { + "epoch": 0.026970251994340326, + "grad_norm": 0.1171875, + "learning_rate": 0.0019978182437100094, + "loss": 0.1797, + "step": 3107 + }, + { + "epoch": 0.02697893247454449, + "grad_norm": 0.369140625, + "learning_rate": 0.0019978161721133252, + "loss": 0.1855, + "step": 3108 + }, + { + "epoch": 0.026987612954748656, + "grad_norm": 0.1572265625, + "learning_rate": 0.001997814099534803, + "loss": 0.1543, + "step": 3109 + }, + { + "epoch": 0.02699629343495282, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019978120259744447, + "loss": 0.2539, + "step": 3110 + }, + { + "epoch": 0.027004973915156986, + "grad_norm": 0.50390625, + "learning_rate": 0.0019978099514322526, + "loss": 0.1816, + "step": 3111 + }, + { + "epoch": 0.02701365439536115, + "grad_norm": 0.1640625, + "learning_rate": 0.001997807875908229, + "loss": 0.1924, + "step": 3112 + }, + { + "epoch": 0.027022334875565317, + "grad_norm": 0.33203125, + "learning_rate": 0.0019978057994023764, + "loss": 0.249, + "step": 3113 + }, + { + "epoch": 0.027031015355769482, + "grad_norm": 0.400390625, + "learning_rate": 0.0019978037219146967, + "loss": 0.2275, + "step": 3114 + }, + { + "epoch": 0.027039695835973647, + "grad_norm": 0.283203125, + "learning_rate": 0.0019978016434451925, + "loss": 0.1699, + "step": 3115 + }, + { + "epoch": 0.027048376316177812, + "grad_norm": 0.185546875, + "learning_rate": 0.0019977995639938657, + "loss": 0.1953, + "step": 3116 + }, + { + "epoch": 0.027057056796381977, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019977974835607188, + "loss": 0.2236, + "step": 3117 + }, + { + "epoch": 0.027065737276586142, + "grad_norm": 0.0947265625, + "learning_rate": 0.001997795402145754, + "loss": 0.1846, + "step": 3118 + }, + { + "epoch": 0.027074417756790304, + "grad_norm": 0.1142578125, + "learning_rate": 0.001997793319748974, + "loss": 0.2363, + "step": 3119 + }, + { + "epoch": 0.02708309823699447, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019977912363703804, + "loss": 0.1865, + "step": 3120 + }, + { + "epoch": 0.027091778717198634, + "grad_norm": 0.13671875, + "learning_rate": 0.0019977891520099756, + "loss": 0.2031, + "step": 3121 + }, + { + "epoch": 0.0271004591974028, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019977870666677625, + "loss": 0.1836, + "step": 3122 + }, + { + "epoch": 0.027109139677606964, + "grad_norm": 0.119140625, + "learning_rate": 0.001997784980343743, + "loss": 0.1865, + "step": 3123 + }, + { + "epoch": 0.02711782015781113, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019977828930379193, + "loss": 0.1719, + "step": 3124 + }, + { + "epoch": 0.027126500638015295, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019977808047502935, + "loss": 0.1934, + "step": 3125 + }, + { + "epoch": 0.02713518111821946, + "grad_norm": 0.14453125, + "learning_rate": 0.001997778715480868, + "loss": 0.1611, + "step": 3126 + }, + { + "epoch": 0.027143861598423625, + "grad_norm": 0.11279296875, + "learning_rate": 0.001997776625229645, + "loss": 0.1836, + "step": 3127 + }, + { + "epoch": 0.02715254207862779, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019977745339966276, + "loss": 0.2148, + "step": 3128 + }, + { + "epoch": 0.027161222558831955, + "grad_norm": 0.255859375, + "learning_rate": 0.0019977724417818174, + "loss": 0.2207, + "step": 3129 + }, + { + "epoch": 0.02716990303903612, + "grad_norm": 0.212890625, + "learning_rate": 0.0019977703485852165, + "loss": 0.1709, + "step": 3130 + }, + { + "epoch": 0.027178583519240285, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019977682544068272, + "loss": 0.2012, + "step": 3131 + }, + { + "epoch": 0.02718726399944445, + "grad_norm": 0.099609375, + "learning_rate": 0.0019977661592466525, + "loss": 0.209, + "step": 3132 + }, + { + "epoch": 0.027195944479648616, + "grad_norm": 0.271484375, + "learning_rate": 0.001997764063104694, + "loss": 0.3164, + "step": 3133 + }, + { + "epoch": 0.02720462495985278, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019977619659809543, + "loss": 0.2012, + "step": 3134 + }, + { + "epoch": 0.027213305440056942, + "grad_norm": 0.166015625, + "learning_rate": 0.001997759867875435, + "loss": 0.2305, + "step": 3135 + }, + { + "epoch": 0.027221985920261108, + "grad_norm": 0.3046875, + "learning_rate": 0.0019977577687881397, + "loss": 0.1934, + "step": 3136 + }, + { + "epoch": 0.027230666400465273, + "grad_norm": 0.07080078125, + "learning_rate": 0.00199775566871907, + "loss": 0.2041, + "step": 3137 + }, + { + "epoch": 0.027239346880669438, + "grad_norm": 0.19140625, + "learning_rate": 0.001997753567668228, + "loss": 0.2246, + "step": 3138 + }, + { + "epoch": 0.027248027360873603, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019977514656356163, + "loss": 0.2178, + "step": 3139 + }, + { + "epoch": 0.027256707841077768, + "grad_norm": 0.123046875, + "learning_rate": 0.0019977493626212365, + "loss": 0.1943, + "step": 3140 + }, + { + "epoch": 0.027265388321281933, + "grad_norm": 0.88671875, + "learning_rate": 0.001997747258625092, + "loss": 0.2129, + "step": 3141 + }, + { + "epoch": 0.027274068801486098, + "grad_norm": 0.51953125, + "learning_rate": 0.001997745153647185, + "loss": 0.1885, + "step": 3142 + }, + { + "epoch": 0.027282749281690263, + "grad_norm": 0.099609375, + "learning_rate": 0.001997743047687517, + "loss": 0.2021, + "step": 3143 + }, + { + "epoch": 0.02729142976189443, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019977409407460904, + "loss": 0.2266, + "step": 3144 + }, + { + "epoch": 0.027300110242098594, + "grad_norm": 0.1640625, + "learning_rate": 0.0019977388328229076, + "loss": 0.2432, + "step": 3145 + }, + { + "epoch": 0.02730879072230276, + "grad_norm": 0.0859375, + "learning_rate": 0.001997736723917972, + "loss": 0.1826, + "step": 3146 + }, + { + "epoch": 0.027317471202506924, + "grad_norm": 0.166015625, + "learning_rate": 0.0019977346140312843, + "loss": 0.2207, + "step": 3147 + }, + { + "epoch": 0.02732615168271109, + "grad_norm": 1.4921875, + "learning_rate": 0.0019977325031628477, + "loss": 0.375, + "step": 3148 + }, + { + "epoch": 0.027334832162915254, + "grad_norm": 0.19921875, + "learning_rate": 0.0019977303913126646, + "loss": 0.2539, + "step": 3149 + }, + { + "epoch": 0.027343512643119416, + "grad_norm": 0.58984375, + "learning_rate": 0.001997728278480737, + "loss": 0.1807, + "step": 3150 + }, + { + "epoch": 0.02735219312332358, + "grad_norm": 0.359375, + "learning_rate": 0.001997726164667067, + "loss": 0.2012, + "step": 3151 + }, + { + "epoch": 0.027360873603527746, + "grad_norm": 0.349609375, + "learning_rate": 0.001997724049871657, + "loss": 0.1797, + "step": 3152 + }, + { + "epoch": 0.02736955408373191, + "grad_norm": 0.337890625, + "learning_rate": 0.0019977219340945097, + "loss": 0.207, + "step": 3153 + }, + { + "epoch": 0.027378234563936076, + "grad_norm": 0.50390625, + "learning_rate": 0.001997719817335627, + "loss": 0.1836, + "step": 3154 + }, + { + "epoch": 0.02738691504414024, + "grad_norm": 0.486328125, + "learning_rate": 0.0019977176995950117, + "loss": 0.1953, + "step": 3155 + }, + { + "epoch": 0.027395595524344406, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019977155808726657, + "loss": 0.1982, + "step": 3156 + }, + { + "epoch": 0.02740427600454857, + "grad_norm": 0.0947265625, + "learning_rate": 0.001997713461168591, + "loss": 0.2246, + "step": 3157 + }, + { + "epoch": 0.027412956484752737, + "grad_norm": 0.255859375, + "learning_rate": 0.001997711340482791, + "loss": 0.207, + "step": 3158 + }, + { + "epoch": 0.027421636964956902, + "grad_norm": 0.08935546875, + "learning_rate": 0.001997709218815267, + "loss": 0.2109, + "step": 3159 + }, + { + "epoch": 0.027430317445161067, + "grad_norm": 0.2734375, + "learning_rate": 0.001997707096166022, + "loss": 0.1387, + "step": 3160 + }, + { + "epoch": 0.027438997925365232, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019977049725350576, + "loss": 0.2559, + "step": 3161 + }, + { + "epoch": 0.027447678405569397, + "grad_norm": 0.64453125, + "learning_rate": 0.0019977028479223765, + "loss": 0.207, + "step": 3162 + }, + { + "epoch": 0.027456358885773562, + "grad_norm": 0.5078125, + "learning_rate": 0.001997700722327981, + "loss": 0.1836, + "step": 3163 + }, + { + "epoch": 0.027465039365977727, + "grad_norm": 0.11865234375, + "learning_rate": 0.001997698595751874, + "loss": 0.1729, + "step": 3164 + }, + { + "epoch": 0.02747371984618189, + "grad_norm": 0.0791015625, + "learning_rate": 0.001997696468194057, + "loss": 0.2021, + "step": 3165 + }, + { + "epoch": 0.027482400326386054, + "grad_norm": 0.109375, + "learning_rate": 0.001997694339654533, + "loss": 0.1885, + "step": 3166 + }, + { + "epoch": 0.02749108080659022, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019976922101333037, + "loss": 0.1895, + "step": 3167 + }, + { + "epoch": 0.027499761286794384, + "grad_norm": 0.13671875, + "learning_rate": 0.0019976900796303716, + "loss": 0.2031, + "step": 3168 + }, + { + "epoch": 0.02750844176699855, + "grad_norm": 0.11376953125, + "learning_rate": 0.001997687948145739, + "loss": 0.2051, + "step": 3169 + }, + { + "epoch": 0.027517122247202715, + "grad_norm": 0.19140625, + "learning_rate": 0.0019976858156794085, + "loss": 0.1621, + "step": 3170 + }, + { + "epoch": 0.02752580272740688, + "grad_norm": 0.1708984375, + "learning_rate": 0.001997683682231382, + "loss": 0.207, + "step": 3171 + }, + { + "epoch": 0.027534483207611045, + "grad_norm": 0.26953125, + "learning_rate": 0.0019976815478016624, + "loss": 0.252, + "step": 3172 + }, + { + "epoch": 0.02754316368781521, + "grad_norm": 0.1328125, + "learning_rate": 0.0019976794123902518, + "loss": 0.2754, + "step": 3173 + }, + { + "epoch": 0.027551844168019375, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019976772759971524, + "loss": 0.1816, + "step": 3174 + }, + { + "epoch": 0.02756052464822354, + "grad_norm": 0.125, + "learning_rate": 0.001997675138622367, + "loss": 0.249, + "step": 3175 + }, + { + "epoch": 0.027569205128427705, + "grad_norm": 0.125, + "learning_rate": 0.001997673000265897, + "loss": 0.209, + "step": 3176 + }, + { + "epoch": 0.02757788560863187, + "grad_norm": 0.291015625, + "learning_rate": 0.0019976708609277453, + "loss": 0.1602, + "step": 3177 + }, + { + "epoch": 0.027586566088836036, + "grad_norm": 0.390625, + "learning_rate": 0.0019976687206079147, + "loss": 0.1875, + "step": 3178 + }, + { + "epoch": 0.0275952465690402, + "grad_norm": 0.232421875, + "learning_rate": 0.0019976665793064066, + "loss": 0.1885, + "step": 3179 + }, + { + "epoch": 0.027603927049244366, + "grad_norm": 0.380859375, + "learning_rate": 0.001997664437023224, + "loss": 0.2207, + "step": 3180 + }, + { + "epoch": 0.027612607529448528, + "grad_norm": 0.13671875, + "learning_rate": 0.001997662293758369, + "loss": 0.2051, + "step": 3181 + }, + { + "epoch": 0.027621288009652693, + "grad_norm": 0.1728515625, + "learning_rate": 0.001997660149511844, + "loss": 0.2246, + "step": 3182 + }, + { + "epoch": 0.027629968489856858, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019976580042836514, + "loss": 0.2207, + "step": 3183 + }, + { + "epoch": 0.027638648970061023, + "grad_norm": 0.1328125, + "learning_rate": 0.0019976558580737935, + "loss": 0.2168, + "step": 3184 + }, + { + "epoch": 0.027647329450265188, + "grad_norm": 0.267578125, + "learning_rate": 0.001997653710882273, + "loss": 0.2334, + "step": 3185 + }, + { + "epoch": 0.027656009930469353, + "grad_norm": 0.46484375, + "learning_rate": 0.0019976515627090916, + "loss": 0.1797, + "step": 3186 + }, + { + "epoch": 0.027664690410673518, + "grad_norm": 0.2734375, + "learning_rate": 0.0019976494135542514, + "loss": 0.252, + "step": 3187 + }, + { + "epoch": 0.027673370890877683, + "grad_norm": 0.357421875, + "learning_rate": 0.001997647263417756, + "loss": 0.2656, + "step": 3188 + }, + { + "epoch": 0.02768205137108185, + "grad_norm": 0.10546875, + "learning_rate": 0.001997645112299607, + "loss": 0.1729, + "step": 3189 + }, + { + "epoch": 0.027690731851286014, + "grad_norm": 0.7421875, + "learning_rate": 0.0019976429601998064, + "loss": 0.2598, + "step": 3190 + }, + { + "epoch": 0.02769941233149018, + "grad_norm": 0.181640625, + "learning_rate": 0.001997640807118357, + "loss": 0.1943, + "step": 3191 + }, + { + "epoch": 0.027708092811694344, + "grad_norm": 0.1953125, + "learning_rate": 0.0019976386530552613, + "loss": 0.2305, + "step": 3192 + }, + { + "epoch": 0.02771677329189851, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019976364980105214, + "loss": 0.1855, + "step": 3193 + }, + { + "epoch": 0.027725453772102674, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019976343419841397, + "loss": 0.1699, + "step": 3194 + }, + { + "epoch": 0.02773413425230684, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019976321849761187, + "loss": 0.1875, + "step": 3195 + }, + { + "epoch": 0.027742814732511, + "grad_norm": 0.15625, + "learning_rate": 0.0019976300269864606, + "loss": 0.2305, + "step": 3196 + }, + { + "epoch": 0.027751495212715166, + "grad_norm": 0.1201171875, + "learning_rate": 0.001997627868015167, + "loss": 0.2168, + "step": 3197 + }, + { + "epoch": 0.02776017569291933, + "grad_norm": 0.376953125, + "learning_rate": 0.0019976257080622424, + "loss": 0.1689, + "step": 3198 + }, + { + "epoch": 0.027768856173123496, + "grad_norm": 0.2138671875, + "learning_rate": 0.001997623547127687, + "loss": 0.1768, + "step": 3199 + }, + { + "epoch": 0.02777753665332766, + "grad_norm": 0.1318359375, + "learning_rate": 0.001997621385211504, + "loss": 0.207, + "step": 3200 + }, + { + "epoch": 0.027786217133531826, + "grad_norm": 0.134765625, + "learning_rate": 0.001997619222313696, + "loss": 0.2363, + "step": 3201 + }, + { + "epoch": 0.02779489761373599, + "grad_norm": 0.08251953125, + "learning_rate": 0.001997617058434265, + "loss": 0.1689, + "step": 3202 + }, + { + "epoch": 0.027803578093940157, + "grad_norm": 0.08984375, + "learning_rate": 0.0019976148935732137, + "loss": 0.1895, + "step": 3203 + }, + { + "epoch": 0.027812258574144322, + "grad_norm": 0.1533203125, + "learning_rate": 0.001997612727730544, + "loss": 0.1924, + "step": 3204 + }, + { + "epoch": 0.027820939054348487, + "grad_norm": 0.142578125, + "learning_rate": 0.0019976105609062584, + "loss": 0.1875, + "step": 3205 + }, + { + "epoch": 0.027829619534552652, + "grad_norm": 0.328125, + "learning_rate": 0.0019976083931003597, + "loss": 0.2236, + "step": 3206 + }, + { + "epoch": 0.027838300014756817, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019976062243128494, + "loss": 0.2637, + "step": 3207 + }, + { + "epoch": 0.027846980494960982, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019976040545437307, + "loss": 0.1719, + "step": 3208 + }, + { + "epoch": 0.027855660975165147, + "grad_norm": 0.412109375, + "learning_rate": 0.0019976018837930057, + "loss": 0.1992, + "step": 3209 + }, + { + "epoch": 0.027864341455369313, + "grad_norm": 0.28515625, + "learning_rate": 0.001997599712060677, + "loss": 0.248, + "step": 3210 + }, + { + "epoch": 0.027873021935573478, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019975975393467463, + "loss": 0.209, + "step": 3211 + }, + { + "epoch": 0.02788170241577764, + "grad_norm": 0.0751953125, + "learning_rate": 0.001997595365651217, + "loss": 0.1797, + "step": 3212 + }, + { + "epoch": 0.027890382895981804, + "grad_norm": 0.416015625, + "learning_rate": 0.0019975931909740905, + "loss": 0.2158, + "step": 3213 + }, + { + "epoch": 0.02789906337618597, + "grad_norm": 0.1806640625, + "learning_rate": 0.001997591015315369, + "loss": 0.2148, + "step": 3214 + }, + { + "epoch": 0.027907743856390135, + "grad_norm": 0.099609375, + "learning_rate": 0.0019975888386750563, + "loss": 0.207, + "step": 3215 + }, + { + "epoch": 0.0279164243365943, + "grad_norm": 0.26171875, + "learning_rate": 0.001997586661053154, + "loss": 0.2217, + "step": 3216 + }, + { + "epoch": 0.027925104816798465, + "grad_norm": 0.1494140625, + "learning_rate": 0.001997584482449664, + "loss": 0.2031, + "step": 3217 + }, + { + "epoch": 0.02793378529700263, + "grad_norm": 0.70703125, + "learning_rate": 0.0019975823028645892, + "loss": 0.2041, + "step": 3218 + }, + { + "epoch": 0.027942465777206795, + "grad_norm": 0.1357421875, + "learning_rate": 0.001997580122297932, + "loss": 0.167, + "step": 3219 + }, + { + "epoch": 0.02795114625741096, + "grad_norm": 0.1640625, + "learning_rate": 0.0019975779407496947, + "loss": 0.2129, + "step": 3220 + }, + { + "epoch": 0.027959826737615125, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019975757582198794, + "loss": 0.1836, + "step": 3221 + }, + { + "epoch": 0.02796850721781929, + "grad_norm": 0.25, + "learning_rate": 0.0019975735747084886, + "loss": 0.1992, + "step": 3222 + }, + { + "epoch": 0.027977187698023456, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019975713902155253, + "loss": 0.1768, + "step": 3223 + }, + { + "epoch": 0.02798586817822762, + "grad_norm": 0.177734375, + "learning_rate": 0.001997569204740991, + "loss": 0.2471, + "step": 3224 + }, + { + "epoch": 0.027994548658431786, + "grad_norm": 0.25390625, + "learning_rate": 0.001997567018284889, + "loss": 0.1914, + "step": 3225 + }, + { + "epoch": 0.02800322913863595, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019975648308472207, + "loss": 0.1719, + "step": 3226 + }, + { + "epoch": 0.028011909618840113, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019975626424279893, + "loss": 0.1963, + "step": 3227 + }, + { + "epoch": 0.028020590099044278, + "grad_norm": 0.3515625, + "learning_rate": 0.0019975604530271967, + "loss": 0.2031, + "step": 3228 + }, + { + "epoch": 0.028029270579248443, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019975582626448455, + "loss": 0.1992, + "step": 3229 + }, + { + "epoch": 0.028037951059452608, + "grad_norm": 0.26953125, + "learning_rate": 0.001997556071280938, + "loss": 0.2031, + "step": 3230 + }, + { + "epoch": 0.028046631539656773, + "grad_norm": 0.0908203125, + "learning_rate": 0.001997553878935477, + "loss": 0.2227, + "step": 3231 + }, + { + "epoch": 0.028055312019860938, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019975516856084643, + "loss": 0.1826, + "step": 3232 + }, + { + "epoch": 0.028063992500065103, + "grad_norm": 0.07958984375, + "learning_rate": 0.001997549491299902, + "loss": 0.1973, + "step": 3233 + }, + { + "epoch": 0.02807267298026927, + "grad_norm": 0.1044921875, + "learning_rate": 0.001997547296009794, + "loss": 0.1758, + "step": 3234 + }, + { + "epoch": 0.028081353460473434, + "grad_norm": 0.205078125, + "learning_rate": 0.0019975450997381417, + "loss": 0.2812, + "step": 3235 + }, + { + "epoch": 0.0280900339406776, + "grad_norm": 0.1328125, + "learning_rate": 0.001997542902484947, + "loss": 0.1875, + "step": 3236 + }, + { + "epoch": 0.028098714420881764, + "grad_norm": 0.1298828125, + "learning_rate": 0.001997540704250213, + "loss": 0.165, + "step": 3237 + }, + { + "epoch": 0.02810739490108593, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019975385050339423, + "loss": 0.2041, + "step": 3238 + }, + { + "epoch": 0.028116075381290094, + "grad_norm": 0.396484375, + "learning_rate": 0.0019975363048361366, + "loss": 0.2168, + "step": 3239 + }, + { + "epoch": 0.02812475586149426, + "grad_norm": 0.140625, + "learning_rate": 0.0019975341036567993, + "loss": 0.1777, + "step": 3240 + }, + { + "epoch": 0.028133436341698424, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019975319014959316, + "loss": 0.25, + "step": 3241 + }, + { + "epoch": 0.028142116821902586, + "grad_norm": 0.6953125, + "learning_rate": 0.0019975296983535365, + "loss": 0.2344, + "step": 3242 + }, + { + "epoch": 0.02815079730210675, + "grad_norm": 0.21875, + "learning_rate": 0.0019975274942296167, + "loss": 0.1836, + "step": 3243 + }, + { + "epoch": 0.028159477782310916, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019975252891241742, + "loss": 0.1631, + "step": 3244 + }, + { + "epoch": 0.02816815826251508, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019975230830372114, + "loss": 0.1992, + "step": 3245 + }, + { + "epoch": 0.028176838742719246, + "grad_norm": 0.259765625, + "learning_rate": 0.001997520875968731, + "loss": 0.1719, + "step": 3246 + }, + { + "epoch": 0.02818551922292341, + "grad_norm": 0.169921875, + "learning_rate": 0.0019975186679187353, + "loss": 0.1641, + "step": 3247 + }, + { + "epoch": 0.028194199703127577, + "grad_norm": 0.267578125, + "learning_rate": 0.0019975164588872264, + "loss": 0.2168, + "step": 3248 + }, + { + "epoch": 0.028202880183331742, + "grad_norm": 0.09765625, + "learning_rate": 0.0019975142488742074, + "loss": 0.209, + "step": 3249 + }, + { + "epoch": 0.028211560663535907, + "grad_norm": 0.16796875, + "learning_rate": 0.0019975120378796798, + "loss": 0.2168, + "step": 3250 + }, + { + "epoch": 0.028220241143740072, + "grad_norm": 0.5234375, + "learning_rate": 0.001997509825903647, + "loss": 0.1943, + "step": 3251 + }, + { + "epoch": 0.028228921623944237, + "grad_norm": 0.138671875, + "learning_rate": 0.001997507612946111, + "loss": 0.1807, + "step": 3252 + }, + { + "epoch": 0.028237602104148402, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019975053990070736, + "loss": 0.166, + "step": 3253 + }, + { + "epoch": 0.028246282584352567, + "grad_norm": 0.10888671875, + "learning_rate": 0.001997503184086538, + "loss": 0.1924, + "step": 3254 + }, + { + "epoch": 0.028254963064556732, + "grad_norm": 0.6640625, + "learning_rate": 0.0019975009681845067, + "loss": 0.2021, + "step": 3255 + }, + { + "epoch": 0.028263643544760898, + "grad_norm": 0.5546875, + "learning_rate": 0.0019974987513009814, + "loss": 0.1738, + "step": 3256 + }, + { + "epoch": 0.028272324024965063, + "grad_norm": 0.287109375, + "learning_rate": 0.001997496533435965, + "loss": 0.207, + "step": 3257 + }, + { + "epoch": 0.028281004505169224, + "grad_norm": 0.11376953125, + "learning_rate": 0.00199749431458946, + "loss": 0.1924, + "step": 3258 + }, + { + "epoch": 0.02828968498537339, + "grad_norm": 0.4609375, + "learning_rate": 0.001997492094761469, + "loss": 0.1689, + "step": 3259 + }, + { + "epoch": 0.028298365465577555, + "grad_norm": 0.11328125, + "learning_rate": 0.001997489873951994, + "loss": 0.1875, + "step": 3260 + }, + { + "epoch": 0.02830704594578172, + "grad_norm": 0.08935546875, + "learning_rate": 0.001997487652161037, + "loss": 0.1953, + "step": 3261 + }, + { + "epoch": 0.028315726425985885, + "grad_norm": 0.279296875, + "learning_rate": 0.0019974854293886017, + "loss": 0.2129, + "step": 3262 + }, + { + "epoch": 0.02832440690619005, + "grad_norm": 0.1357421875, + "learning_rate": 0.00199748320563469, + "loss": 0.2539, + "step": 3263 + }, + { + "epoch": 0.028333087386394215, + "grad_norm": 0.134765625, + "learning_rate": 0.0019974809808993035, + "loss": 0.1582, + "step": 3264 + }, + { + "epoch": 0.02834176786659838, + "grad_norm": 0.255859375, + "learning_rate": 0.0019974787551824452, + "loss": 0.1719, + "step": 3265 + }, + { + "epoch": 0.028350448346802545, + "grad_norm": 1.1171875, + "learning_rate": 0.0019974765284841178, + "loss": 0.7461, + "step": 3266 + }, + { + "epoch": 0.02835912882700671, + "grad_norm": 0.439453125, + "learning_rate": 0.0019974743008043237, + "loss": 0.2002, + "step": 3267 + }, + { + "epoch": 0.028367809307210876, + "grad_norm": 0.78125, + "learning_rate": 0.001997472072143065, + "loss": 0.2227, + "step": 3268 + }, + { + "epoch": 0.02837648978741504, + "grad_norm": 0.341796875, + "learning_rate": 0.0019974698425003446, + "loss": 0.2246, + "step": 3269 + }, + { + "epoch": 0.028385170267619206, + "grad_norm": 0.259765625, + "learning_rate": 0.0019974676118761645, + "loss": 0.1758, + "step": 3270 + }, + { + "epoch": 0.02839385074782337, + "grad_norm": 0.259765625, + "learning_rate": 0.0019974653802705272, + "loss": 0.1641, + "step": 3271 + }, + { + "epoch": 0.028402531228027536, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019974631476834355, + "loss": 0.166, + "step": 3272 + }, + { + "epoch": 0.028411211708231698, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019974609141148914, + "loss": 0.1543, + "step": 3273 + }, + { + "epoch": 0.028419892188435863, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019974586795648975, + "loss": 0.1553, + "step": 3274 + }, + { + "epoch": 0.028428572668640028, + "grad_norm": 0.267578125, + "learning_rate": 0.0019974564440334566, + "loss": 0.1904, + "step": 3275 + }, + { + "epoch": 0.028437253148844193, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019974542075205702, + "loss": 0.1621, + "step": 3276 + }, + { + "epoch": 0.028445933629048358, + "grad_norm": 0.0859375, + "learning_rate": 0.001997451970026242, + "loss": 0.1553, + "step": 3277 + }, + { + "epoch": 0.028454614109252523, + "grad_norm": 0.15625, + "learning_rate": 0.0019974497315504735, + "loss": 0.1973, + "step": 3278 + }, + { + "epoch": 0.02846329458945669, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019974474920932675, + "loss": 0.1729, + "step": 3279 + }, + { + "epoch": 0.028471975069660854, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019974452516546264, + "loss": 0.2031, + "step": 3280 + }, + { + "epoch": 0.02848065554986502, + "grad_norm": 0.126953125, + "learning_rate": 0.0019974430102345526, + "loss": 0.2656, + "step": 3281 + }, + { + "epoch": 0.028489336030069184, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019974407678330485, + "loss": 0.1855, + "step": 3282 + }, + { + "epoch": 0.02849801651027335, + "grad_norm": 0.140625, + "learning_rate": 0.001997438524450117, + "loss": 0.2324, + "step": 3283 + }, + { + "epoch": 0.028506696990477514, + "grad_norm": 0.09619140625, + "learning_rate": 0.00199743628008576, + "loss": 0.2031, + "step": 3284 + }, + { + "epoch": 0.02851537747068168, + "grad_norm": 0.1337890625, + "learning_rate": 0.00199743403473998, + "loss": 0.1875, + "step": 3285 + }, + { + "epoch": 0.028524057950885844, + "grad_norm": 0.30078125, + "learning_rate": 0.00199743178841278, + "loss": 0.2246, + "step": 3286 + }, + { + "epoch": 0.02853273843109001, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019974295411041617, + "loss": 0.1992, + "step": 3287 + }, + { + "epoch": 0.028541418911294174, + "grad_norm": 0.193359375, + "learning_rate": 0.001997427292814129, + "loss": 0.1855, + "step": 3288 + }, + { + "epoch": 0.028550099391498336, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019974250435426818, + "loss": 0.1855, + "step": 3289 + }, + { + "epoch": 0.0285587798717025, + "grad_norm": 0.2451171875, + "learning_rate": 0.001997422793289825, + "loss": 0.1562, + "step": 3290 + }, + { + "epoch": 0.028567460351906666, + "grad_norm": 0.470703125, + "learning_rate": 0.0019974205420555595, + "loss": 0.1777, + "step": 3291 + }, + { + "epoch": 0.02857614083211083, + "grad_norm": 0.298828125, + "learning_rate": 0.0019974182898398886, + "loss": 0.1777, + "step": 3292 + }, + { + "epoch": 0.028584821312314997, + "grad_norm": 0.18359375, + "learning_rate": 0.0019974160366428148, + "loss": 0.2129, + "step": 3293 + }, + { + "epoch": 0.028593501792519162, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019974137824643402, + "loss": 0.2051, + "step": 3294 + }, + { + "epoch": 0.028602182272723327, + "grad_norm": 0.158203125, + "learning_rate": 0.001997411527304467, + "loss": 0.1406, + "step": 3295 + }, + { + "epoch": 0.028610862752927492, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019974092711631986, + "loss": 0.209, + "step": 3296 + }, + { + "epoch": 0.028619543233131657, + "grad_norm": 0.22265625, + "learning_rate": 0.0019974070140405366, + "loss": 0.207, + "step": 3297 + }, + { + "epoch": 0.028628223713335822, + "grad_norm": 0.08447265625, + "learning_rate": 0.001997404755936484, + "loss": 0.1797, + "step": 3298 + }, + { + "epoch": 0.028636904193539987, + "grad_norm": 0.13671875, + "learning_rate": 0.001997402496851043, + "loss": 0.1992, + "step": 3299 + }, + { + "epoch": 0.028645584673744152, + "grad_norm": 0.1904296875, + "learning_rate": 0.001997400236784216, + "loss": 0.1895, + "step": 3300 + }, + { + "epoch": 0.028654265153948318, + "grad_norm": 0.12109375, + "learning_rate": 0.0019973979757360056, + "loss": 0.1953, + "step": 3301 + }, + { + "epoch": 0.028662945634152483, + "grad_norm": 0.56640625, + "learning_rate": 0.0019973957137064137, + "loss": 0.1992, + "step": 3302 + }, + { + "epoch": 0.028671626114356648, + "grad_norm": 0.10791015625, + "learning_rate": 0.001997393450695444, + "loss": 0.2539, + "step": 3303 + }, + { + "epoch": 0.02868030659456081, + "grad_norm": 0.1015625, + "learning_rate": 0.001997391186703098, + "loss": 0.2256, + "step": 3304 + }, + { + "epoch": 0.028688987074764975, + "grad_norm": 0.11083984375, + "learning_rate": 0.001997388921729379, + "loss": 0.1426, + "step": 3305 + }, + { + "epoch": 0.02869766755496914, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019973866557742885, + "loss": 0.165, + "step": 3306 + }, + { + "epoch": 0.028706348035173305, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019973843888378296, + "loss": 0.168, + "step": 3307 + }, + { + "epoch": 0.02871502851537747, + "grad_norm": 0.0703125, + "learning_rate": 0.0019973821209200043, + "loss": 0.1992, + "step": 3308 + }, + { + "epoch": 0.028723708995581635, + "grad_norm": 0.359375, + "learning_rate": 0.001997379852020816, + "loss": 0.1328, + "step": 3309 + }, + { + "epoch": 0.0287323894757858, + "grad_norm": 0.08642578125, + "learning_rate": 0.001997377582140266, + "loss": 0.21, + "step": 3310 + }, + { + "epoch": 0.028741069955989965, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019973753112783577, + "loss": 0.1689, + "step": 3311 + }, + { + "epoch": 0.02874975043619413, + "grad_norm": 0.31640625, + "learning_rate": 0.0019973730394350934, + "loss": 0.2129, + "step": 3312 + }, + { + "epoch": 0.028758430916398296, + "grad_norm": 0.66796875, + "learning_rate": 0.001997370766610475, + "loss": 0.4082, + "step": 3313 + }, + { + "epoch": 0.02876711139660246, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019973684928045053, + "loss": 0.1865, + "step": 3314 + }, + { + "epoch": 0.028775791876806626, + "grad_norm": 0.296875, + "learning_rate": 0.0019973662180171876, + "loss": 0.166, + "step": 3315 + }, + { + "epoch": 0.02878447235701079, + "grad_norm": 0.232421875, + "learning_rate": 0.001997363942248523, + "loss": 0.2363, + "step": 3316 + }, + { + "epoch": 0.028793152837214956, + "grad_norm": 0.16796875, + "learning_rate": 0.001997361665498515, + "loss": 0.1729, + "step": 3317 + }, + { + "epoch": 0.02880183331741912, + "grad_norm": 0.26171875, + "learning_rate": 0.0019973593877671654, + "loss": 0.2148, + "step": 3318 + }, + { + "epoch": 0.028810513797623283, + "grad_norm": 0.490234375, + "learning_rate": 0.0019973571090544776, + "loss": 0.2129, + "step": 3319 + }, + { + "epoch": 0.028819194277827448, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019973548293604534, + "loss": 0.1973, + "step": 3320 + }, + { + "epoch": 0.028827874758031613, + "grad_norm": 0.52734375, + "learning_rate": 0.001997352548685095, + "loss": 0.2168, + "step": 3321 + }, + { + "epoch": 0.028836555238235778, + "grad_norm": 0.140625, + "learning_rate": 0.0019973502670284056, + "loss": 0.2021, + "step": 3322 + }, + { + "epoch": 0.028845235718439943, + "grad_norm": 0.265625, + "learning_rate": 0.001997347984390388, + "loss": 0.2461, + "step": 3323 + }, + { + "epoch": 0.02885391619864411, + "grad_norm": 0.57421875, + "learning_rate": 0.0019973457007710434, + "loss": 0.2314, + "step": 3324 + }, + { + "epoch": 0.028862596678848274, + "grad_norm": 0.228515625, + "learning_rate": 0.001997343416170375, + "loss": 0.2305, + "step": 3325 + }, + { + "epoch": 0.02887127715905244, + "grad_norm": 0.07421875, + "learning_rate": 0.001997341130588386, + "loss": 0.2012, + "step": 3326 + }, + { + "epoch": 0.028879957639256604, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019973388440250777, + "loss": 0.25, + "step": 3327 + }, + { + "epoch": 0.02888863811946077, + "grad_norm": 0.080078125, + "learning_rate": 0.001997336556480453, + "loss": 0.1807, + "step": 3328 + }, + { + "epoch": 0.028897318599664934, + "grad_norm": 0.087890625, + "learning_rate": 0.001997334267954515, + "loss": 0.2002, + "step": 3329 + }, + { + "epoch": 0.0289059990798691, + "grad_norm": 0.392578125, + "learning_rate": 0.0019973319784472652, + "loss": 0.2441, + "step": 3330 + }, + { + "epoch": 0.028914679560073264, + "grad_norm": 0.224609375, + "learning_rate": 0.001997329687958707, + "loss": 0.2617, + "step": 3331 + }, + { + "epoch": 0.02892336004027743, + "grad_norm": 0.62109375, + "learning_rate": 0.0019973273964888428, + "loss": 0.2754, + "step": 3332 + }, + { + "epoch": 0.028932040520481594, + "grad_norm": 0.2109375, + "learning_rate": 0.001997325104037674, + "loss": 0.208, + "step": 3333 + }, + { + "epoch": 0.02894072100068576, + "grad_norm": 0.107421875, + "learning_rate": 0.0019973228106052046, + "loss": 0.1709, + "step": 3334 + }, + { + "epoch": 0.02894940148088992, + "grad_norm": 0.1640625, + "learning_rate": 0.0019973205161914363, + "loss": 0.1992, + "step": 3335 + }, + { + "epoch": 0.028958081961094086, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019973182207963717, + "loss": 0.1836, + "step": 3336 + }, + { + "epoch": 0.02896676244129825, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019973159244200136, + "loss": 0.1836, + "step": 3337 + }, + { + "epoch": 0.028975442921502417, + "grad_norm": 0.46484375, + "learning_rate": 0.001997313627062364, + "loss": 0.1807, + "step": 3338 + }, + { + "epoch": 0.02898412340170658, + "grad_norm": 0.07666015625, + "learning_rate": 0.001997311328723426, + "loss": 0.1924, + "step": 3339 + }, + { + "epoch": 0.028992803881910747, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019973090294032013, + "loss": 0.2119, + "step": 3340 + }, + { + "epoch": 0.029001484362114912, + "grad_norm": 0.236328125, + "learning_rate": 0.0019973067291016934, + "loss": 0.1768, + "step": 3341 + }, + { + "epoch": 0.029010164842319077, + "grad_norm": 0.58984375, + "learning_rate": 0.0019973044278189045, + "loss": 0.2188, + "step": 3342 + }, + { + "epoch": 0.029018845322523242, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019973021255548363, + "loss": 0.2344, + "step": 3343 + }, + { + "epoch": 0.029027525802727407, + "grad_norm": 0.244140625, + "learning_rate": 0.0019972998223094923, + "loss": 0.2207, + "step": 3344 + }, + { + "epoch": 0.029036206282931572, + "grad_norm": 0.12353515625, + "learning_rate": 0.001997297518082875, + "loss": 0.1855, + "step": 3345 + }, + { + "epoch": 0.029044886763135738, + "grad_norm": 0.166015625, + "learning_rate": 0.0019972952128749864, + "loss": 0.1406, + "step": 3346 + }, + { + "epoch": 0.029053567243339903, + "grad_norm": 0.41796875, + "learning_rate": 0.001997292906685829, + "loss": 0.3105, + "step": 3347 + }, + { + "epoch": 0.029062247723544068, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019972905995154057, + "loss": 0.207, + "step": 3348 + }, + { + "epoch": 0.029070928203748233, + "grad_norm": 0.2392578125, + "learning_rate": 0.001997288291363719, + "loss": 0.1797, + "step": 3349 + }, + { + "epoch": 0.029079608683952395, + "grad_norm": 0.30859375, + "learning_rate": 0.0019972859822307712, + "loss": 0.2246, + "step": 3350 + }, + { + "epoch": 0.02908828916415656, + "grad_norm": 0.263671875, + "learning_rate": 0.001997283672116565, + "loss": 0.2344, + "step": 3351 + }, + { + "epoch": 0.029096969644360725, + "grad_norm": 0.50390625, + "learning_rate": 0.001997281361021103, + "loss": 0.2266, + "step": 3352 + }, + { + "epoch": 0.02910565012456489, + "grad_norm": 0.2314453125, + "learning_rate": 0.001997279048944387, + "loss": 0.2021, + "step": 3353 + }, + { + "epoch": 0.029114330604769055, + "grad_norm": 0.197265625, + "learning_rate": 0.001997276735886421, + "loss": 0.1895, + "step": 3354 + }, + { + "epoch": 0.02912301108497322, + "grad_norm": 0.1748046875, + "learning_rate": 0.001997274421847206, + "loss": 0.1729, + "step": 3355 + }, + { + "epoch": 0.029131691565177385, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019972721068267454, + "loss": 0.207, + "step": 3356 + }, + { + "epoch": 0.02914037204538155, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019972697908250416, + "loss": 0.1797, + "step": 3357 + }, + { + "epoch": 0.029149052525585716, + "grad_norm": 0.103515625, + "learning_rate": 0.0019972674738420967, + "loss": 0.1758, + "step": 3358 + }, + { + "epoch": 0.02915773300578988, + "grad_norm": 0.53125, + "learning_rate": 0.0019972651558779137, + "loss": 0.2266, + "step": 3359 + }, + { + "epoch": 0.029166413485994046, + "grad_norm": 0.31640625, + "learning_rate": 0.001997262836932495, + "loss": 0.2363, + "step": 3360 + }, + { + "epoch": 0.02917509396619821, + "grad_norm": 0.076171875, + "learning_rate": 0.0019972605170058434, + "loss": 0.1934, + "step": 3361 + }, + { + "epoch": 0.029183774446402376, + "grad_norm": 0.216796875, + "learning_rate": 0.001997258196097961, + "loss": 0.2734, + "step": 3362 + }, + { + "epoch": 0.02919245492660654, + "grad_norm": 0.14453125, + "learning_rate": 0.00199725587420885, + "loss": 0.2051, + "step": 3363 + }, + { + "epoch": 0.029201135406810706, + "grad_norm": 0.3359375, + "learning_rate": 0.001997253551338514, + "loss": 0.1816, + "step": 3364 + }, + { + "epoch": 0.02920981588701487, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019972512274869553, + "loss": 0.2051, + "step": 3365 + }, + { + "epoch": 0.029218496367219033, + "grad_norm": 0.10009765625, + "learning_rate": 0.001997248902654176, + "loss": 0.2021, + "step": 3366 + }, + { + "epoch": 0.029227176847423198, + "grad_norm": 0.494140625, + "learning_rate": 0.0019972465768401783, + "loss": 0.1865, + "step": 3367 + }, + { + "epoch": 0.029235857327627363, + "grad_norm": 2.046875, + "learning_rate": 0.0019972442500449657, + "loss": 0.4219, + "step": 3368 + }, + { + "epoch": 0.02924453780783153, + "grad_norm": 0.314453125, + "learning_rate": 0.00199724192226854, + "loss": 0.2539, + "step": 3369 + }, + { + "epoch": 0.029253218288035693, + "grad_norm": 0.0859375, + "learning_rate": 0.001997239593510904, + "loss": 0.1973, + "step": 3370 + }, + { + "epoch": 0.02926189876823986, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019972372637720604, + "loss": 0.1738, + "step": 3371 + }, + { + "epoch": 0.029270579248444024, + "grad_norm": 0.13671875, + "learning_rate": 0.0019972349330520116, + "loss": 0.2578, + "step": 3372 + }, + { + "epoch": 0.02927925972864819, + "grad_norm": 0.1318359375, + "learning_rate": 0.00199723260135076, + "loss": 0.2188, + "step": 3373 + }, + { + "epoch": 0.029287940208852354, + "grad_norm": 0.134765625, + "learning_rate": 0.0019972302686683085, + "loss": 0.2295, + "step": 3374 + }, + { + "epoch": 0.02929662068905652, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019972279350046595, + "loss": 0.167, + "step": 3375 + }, + { + "epoch": 0.029305301169260684, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019972256003598153, + "loss": 0.1836, + "step": 3376 + }, + { + "epoch": 0.02931398164946485, + "grad_norm": 0.44140625, + "learning_rate": 0.0019972232647337785, + "loss": 0.2246, + "step": 3377 + }, + { + "epoch": 0.029322662129669014, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019972209281265522, + "loss": 0.1699, + "step": 3378 + }, + { + "epoch": 0.02933134260987318, + "grad_norm": 0.353515625, + "learning_rate": 0.0019972185905381386, + "loss": 0.1992, + "step": 3379 + }, + { + "epoch": 0.029340023090077345, + "grad_norm": 0.310546875, + "learning_rate": 0.00199721625196854, + "loss": 0.2041, + "step": 3380 + }, + { + "epoch": 0.029348703570281506, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019972139124177593, + "loss": 0.168, + "step": 3381 + }, + { + "epoch": 0.02935738405048567, + "grad_norm": 0.455078125, + "learning_rate": 0.0019972115718857987, + "loss": 0.2197, + "step": 3382 + }, + { + "epoch": 0.029366064530689837, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019972092303726613, + "loss": 0.208, + "step": 3383 + }, + { + "epoch": 0.029374745010894, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019972068878783495, + "loss": 0.1689, + "step": 3384 + }, + { + "epoch": 0.029383425491098167, + "grad_norm": 0.09716796875, + "learning_rate": 0.001997204544402865, + "loss": 0.165, + "step": 3385 + }, + { + "epoch": 0.029392105971302332, + "grad_norm": 0.11767578125, + "learning_rate": 0.001997202199946212, + "loss": 0.2012, + "step": 3386 + }, + { + "epoch": 0.029400786451506497, + "grad_norm": 0.1337890625, + "learning_rate": 0.001997199854508392, + "loss": 0.1953, + "step": 3387 + }, + { + "epoch": 0.029409466931710662, + "grad_norm": 0.083984375, + "learning_rate": 0.001997197508089407, + "loss": 0.1641, + "step": 3388 + }, + { + "epoch": 0.029418147411914827, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019971951606892607, + "loss": 0.2031, + "step": 3389 + }, + { + "epoch": 0.029426827892118992, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019971928123079558, + "loss": 0.1689, + "step": 3390 + }, + { + "epoch": 0.029435508372323158, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019971904629454934, + "loss": 0.1855, + "step": 3391 + }, + { + "epoch": 0.029444188852527323, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019971881126018775, + "loss": 0.1504, + "step": 3392 + }, + { + "epoch": 0.029452869332731488, + "grad_norm": 0.48828125, + "learning_rate": 0.00199718576127711, + "loss": 0.1611, + "step": 3393 + }, + { + "epoch": 0.029461549812935653, + "grad_norm": 0.353515625, + "learning_rate": 0.001997183408971194, + "loss": 0.2012, + "step": 3394 + }, + { + "epoch": 0.029470230293139818, + "grad_norm": 0.12451171875, + "learning_rate": 0.001997181055684131, + "loss": 0.1934, + "step": 3395 + }, + { + "epoch": 0.02947891077334398, + "grad_norm": 0.2578125, + "learning_rate": 0.001997178701415925, + "loss": 0.2188, + "step": 3396 + }, + { + "epoch": 0.029487591253548145, + "grad_norm": 0.19140625, + "learning_rate": 0.001997176346166578, + "loss": 0.1992, + "step": 3397 + }, + { + "epoch": 0.02949627173375231, + "grad_norm": 0.333984375, + "learning_rate": 0.0019971739899360915, + "loss": 0.1592, + "step": 3398 + }, + { + "epoch": 0.029504952213956475, + "grad_norm": 0.279296875, + "learning_rate": 0.0019971716327244694, + "loss": 0.168, + "step": 3399 + }, + { + "epoch": 0.02951363269416064, + "grad_norm": 0.30859375, + "learning_rate": 0.0019971692745317142, + "loss": 0.2266, + "step": 3400 + }, + { + "epoch": 0.029522313174364805, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019971669153578276, + "loss": 0.1807, + "step": 3401 + }, + { + "epoch": 0.02953099365456897, + "grad_norm": 0.134765625, + "learning_rate": 0.0019971645552028135, + "loss": 0.1768, + "step": 3402 + }, + { + "epoch": 0.029539674134773135, + "grad_norm": 0.17578125, + "learning_rate": 0.001997162194066673, + "loss": 0.2227, + "step": 3403 + }, + { + "epoch": 0.0295483546149773, + "grad_norm": 0.22265625, + "learning_rate": 0.00199715983194941, + "loss": 0.1982, + "step": 3404 + }, + { + "epoch": 0.029557035095181466, + "grad_norm": 0.16796875, + "learning_rate": 0.001997157468851026, + "loss": 0.1699, + "step": 3405 + }, + { + "epoch": 0.02956571557538563, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019971551047715244, + "loss": 0.1895, + "step": 3406 + }, + { + "epoch": 0.029574396055589796, + "grad_norm": 0.1767578125, + "learning_rate": 0.001997152739710907, + "loss": 0.2324, + "step": 3407 + }, + { + "epoch": 0.02958307653579396, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019971503736691773, + "loss": 0.2207, + "step": 3408 + }, + { + "epoch": 0.029591757015998126, + "grad_norm": 0.259765625, + "learning_rate": 0.0019971480066463374, + "loss": 0.1768, + "step": 3409 + }, + { + "epoch": 0.02960043749620229, + "grad_norm": 0.310546875, + "learning_rate": 0.0019971456386423895, + "loss": 0.2051, + "step": 3410 + }, + { + "epoch": 0.029609117976406456, + "grad_norm": 0.2021484375, + "learning_rate": 0.001997143269657337, + "loss": 0.1719, + "step": 3411 + }, + { + "epoch": 0.029617798456610618, + "grad_norm": 0.25, + "learning_rate": 0.001997140899691182, + "loss": 0.1973, + "step": 3412 + }, + { + "epoch": 0.029626478936814783, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019971385287439274, + "loss": 0.207, + "step": 3413 + }, + { + "epoch": 0.02963515941701895, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019971361568155753, + "loss": 0.1973, + "step": 3414 + }, + { + "epoch": 0.029643839897223113, + "grad_norm": 0.4765625, + "learning_rate": 0.0019971337839061287, + "loss": 0.207, + "step": 3415 + }, + { + "epoch": 0.02965252037742728, + "grad_norm": 0.400390625, + "learning_rate": 0.00199713141001559, + "loss": 0.2236, + "step": 3416 + }, + { + "epoch": 0.029661200857631444, + "grad_norm": 0.20703125, + "learning_rate": 0.001997129035143962, + "loss": 0.1689, + "step": 3417 + }, + { + "epoch": 0.02966988133783561, + "grad_norm": 0.35546875, + "learning_rate": 0.001997126659291247, + "loss": 0.1836, + "step": 3418 + }, + { + "epoch": 0.029678561818039774, + "grad_norm": 0.080078125, + "learning_rate": 0.001997124282457448, + "loss": 0.1562, + "step": 3419 + }, + { + "epoch": 0.02968724229824394, + "grad_norm": 0.1484375, + "learning_rate": 0.001997121904642567, + "loss": 0.1777, + "step": 3420 + }, + { + "epoch": 0.029695922778448104, + "grad_norm": 0.15234375, + "learning_rate": 0.0019971195258466075, + "loss": 0.1611, + "step": 3421 + }, + { + "epoch": 0.02970460325865227, + "grad_norm": 0.4609375, + "learning_rate": 0.001997117146069571, + "loss": 0.168, + "step": 3422 + }, + { + "epoch": 0.029713283738856434, + "grad_norm": 0.10693359375, + "learning_rate": 0.001997114765311461, + "loss": 0.2031, + "step": 3423 + }, + { + "epoch": 0.0297219642190606, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019971123835722795, + "loss": 0.2275, + "step": 3424 + }, + { + "epoch": 0.029730644699264765, + "grad_norm": 0.18359375, + "learning_rate": 0.0019971100008520297, + "loss": 0.1895, + "step": 3425 + }, + { + "epoch": 0.02973932517946893, + "grad_norm": 0.306640625, + "learning_rate": 0.0019971076171507135, + "loss": 0.1895, + "step": 3426 + }, + { + "epoch": 0.02974800565967309, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019971052324683344, + "loss": 0.1777, + "step": 3427 + }, + { + "epoch": 0.029756686139877257, + "grad_norm": 0.12109375, + "learning_rate": 0.001997102846804894, + "loss": 0.1748, + "step": 3428 + }, + { + "epoch": 0.02976536662008142, + "grad_norm": 0.41015625, + "learning_rate": 0.001997100460160396, + "loss": 0.1562, + "step": 3429 + }, + { + "epoch": 0.029774047100285587, + "grad_norm": 0.140625, + "learning_rate": 0.001997098072534842, + "loss": 0.207, + "step": 3430 + }, + { + "epoch": 0.029782727580489752, + "grad_norm": 0.33203125, + "learning_rate": 0.0019970956839282347, + "loss": 0.1699, + "step": 3431 + }, + { + "epoch": 0.029791408060693917, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019970932943405777, + "loss": 0.124, + "step": 3432 + }, + { + "epoch": 0.029800088540898082, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019970909037718724, + "loss": 0.1934, + "step": 3433 + }, + { + "epoch": 0.029808769021102247, + "grad_norm": 0.15234375, + "learning_rate": 0.0019970885122221225, + "loss": 0.1826, + "step": 3434 + }, + { + "epoch": 0.029817449501306412, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019970861196913297, + "loss": 0.1494, + "step": 3435 + }, + { + "epoch": 0.029826129981510578, + "grad_norm": 0.1015625, + "learning_rate": 0.001997083726179497, + "loss": 0.168, + "step": 3436 + }, + { + "epoch": 0.029834810461714743, + "grad_norm": 0.1298828125, + "learning_rate": 0.001997081331686627, + "loss": 0.1631, + "step": 3437 + }, + { + "epoch": 0.029843490941918908, + "grad_norm": 0.125, + "learning_rate": 0.0019970789362127226, + "loss": 0.1895, + "step": 3438 + }, + { + "epoch": 0.029852171422123073, + "grad_norm": 0.396484375, + "learning_rate": 0.001997076539757786, + "loss": 0.1777, + "step": 3439 + }, + { + "epoch": 0.029860851902327238, + "grad_norm": 0.19921875, + "learning_rate": 0.00199707414232182, + "loss": 0.2119, + "step": 3440 + }, + { + "epoch": 0.029869532382531403, + "grad_norm": 0.3984375, + "learning_rate": 0.001997071743904827, + "loss": 0.2021, + "step": 3441 + }, + { + "epoch": 0.029878212862735568, + "grad_norm": 0.2734375, + "learning_rate": 0.0019970693445068104, + "loss": 0.1816, + "step": 3442 + }, + { + "epoch": 0.02988689334293973, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019970669441277717, + "loss": 0.2217, + "step": 3443 + }, + { + "epoch": 0.029895573823143895, + "grad_norm": 0.14453125, + "learning_rate": 0.0019970645427677142, + "loss": 0.1885, + "step": 3444 + }, + { + "epoch": 0.02990425430334806, + "grad_norm": 0.173828125, + "learning_rate": 0.0019970621404266403, + "loss": 0.2178, + "step": 3445 + }, + { + "epoch": 0.029912934783552225, + "grad_norm": 0.1484375, + "learning_rate": 0.001997059737104553, + "loss": 0.1826, + "step": 3446 + }, + { + "epoch": 0.02992161526375639, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019970573328014544, + "loss": 0.1709, + "step": 3447 + }, + { + "epoch": 0.029930295743960555, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019970549275173475, + "loss": 0.209, + "step": 3448 + }, + { + "epoch": 0.02993897622416472, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019970525212522345, + "loss": 0.1787, + "step": 3449 + }, + { + "epoch": 0.029947656704368886, + "grad_norm": 0.0830078125, + "learning_rate": 0.001997050114006119, + "loss": 0.1699, + "step": 3450 + }, + { + "epoch": 0.02995633718457305, + "grad_norm": 0.6796875, + "learning_rate": 0.0019970477057790026, + "loss": 0.2461, + "step": 3451 + }, + { + "epoch": 0.029965017664777216, + "grad_norm": 0.083984375, + "learning_rate": 0.001997045296570888, + "loss": 0.1787, + "step": 3452 + }, + { + "epoch": 0.02997369814498138, + "grad_norm": 0.1015625, + "learning_rate": 0.0019970428863817784, + "loss": 0.1836, + "step": 3453 + }, + { + "epoch": 0.029982378625185546, + "grad_norm": 0.279296875, + "learning_rate": 0.0019970404752116763, + "loss": 0.1855, + "step": 3454 + }, + { + "epoch": 0.02999105910538971, + "grad_norm": 0.12060546875, + "learning_rate": 0.001997038063060584, + "loss": 0.2334, + "step": 3455 + }, + { + "epoch": 0.029999739585593876, + "grad_norm": 0.73828125, + "learning_rate": 0.0019970356499285045, + "loss": 0.2207, + "step": 3456 + }, + { + "epoch": 0.03000842006579804, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019970332358154406, + "loss": 0.1885, + "step": 3457 + }, + { + "epoch": 0.030017100546002203, + "grad_norm": 0.07958984375, + "learning_rate": 0.001997030820721394, + "loss": 0.168, + "step": 3458 + }, + { + "epoch": 0.03002578102620637, + "grad_norm": 0.173828125, + "learning_rate": 0.001997028404646368, + "loss": 0.1777, + "step": 3459 + }, + { + "epoch": 0.030034461506410533, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019970259875903658, + "loss": 0.2402, + "step": 3460 + }, + { + "epoch": 0.0300431419866147, + "grad_norm": 0.12890625, + "learning_rate": 0.001997023569553389, + "loss": 0.2598, + "step": 3461 + }, + { + "epoch": 0.030051822466818864, + "grad_norm": 0.0830078125, + "learning_rate": 0.001997021150535441, + "loss": 0.1855, + "step": 3462 + }, + { + "epoch": 0.03006050294702303, + "grad_norm": 0.3984375, + "learning_rate": 0.0019970187305365238, + "loss": 0.1934, + "step": 3463 + }, + { + "epoch": 0.030069183427227194, + "grad_norm": 0.349609375, + "learning_rate": 0.0019970163095566406, + "loss": 0.25, + "step": 3464 + }, + { + "epoch": 0.03007786390743136, + "grad_norm": 0.205078125, + "learning_rate": 0.001997013887595794, + "loss": 0.1855, + "step": 3465 + }, + { + "epoch": 0.030086544387635524, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019970114646539862, + "loss": 0.2324, + "step": 3466 + }, + { + "epoch": 0.03009522486783969, + "grad_norm": 0.384765625, + "learning_rate": 0.0019970090407312206, + "loss": 0.1943, + "step": 3467 + }, + { + "epoch": 0.030103905348043854, + "grad_norm": 0.380859375, + "learning_rate": 0.0019970066158274988, + "loss": 0.1699, + "step": 3468 + }, + { + "epoch": 0.03011258582824802, + "grad_norm": 0.1669921875, + "learning_rate": 0.001997004189942824, + "loss": 0.2207, + "step": 3469 + }, + { + "epoch": 0.030121266308452185, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019970017630771995, + "loss": 0.2422, + "step": 3470 + }, + { + "epoch": 0.03012994678865635, + "grad_norm": 0.12890625, + "learning_rate": 0.001996999335230627, + "loss": 0.207, + "step": 3471 + }, + { + "epoch": 0.030138627268860515, + "grad_norm": 0.71875, + "learning_rate": 0.0019969969064031097, + "loss": 0.1875, + "step": 3472 + }, + { + "epoch": 0.030147307749064677, + "grad_norm": 0.24609375, + "learning_rate": 0.00199699447659465, + "loss": 0.1758, + "step": 3473 + }, + { + "epoch": 0.03015598822926884, + "grad_norm": 0.388671875, + "learning_rate": 0.0019969920458052506, + "loss": 0.1807, + "step": 3474 + }, + { + "epoch": 0.030164668709473007, + "grad_norm": 0.279296875, + "learning_rate": 0.0019969896140349134, + "loss": 0.2051, + "step": 3475 + }, + { + "epoch": 0.030173349189677172, + "grad_norm": 0.212890625, + "learning_rate": 0.001996987181283643, + "loss": 0.1553, + "step": 3476 + }, + { + "epoch": 0.030182029669881337, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019969847475514403, + "loss": 0.166, + "step": 3477 + }, + { + "epoch": 0.030190710150085502, + "grad_norm": 0.322265625, + "learning_rate": 0.0019969823128383087, + "loss": 0.1494, + "step": 3478 + }, + { + "epoch": 0.030199390630289667, + "grad_norm": 0.31640625, + "learning_rate": 0.0019969798771442508, + "loss": 0.1631, + "step": 3479 + }, + { + "epoch": 0.030208071110493832, + "grad_norm": 0.24609375, + "learning_rate": 0.0019969774404692687, + "loss": 0.1455, + "step": 3480 + }, + { + "epoch": 0.030216751590697997, + "grad_norm": 0.2333984375, + "learning_rate": 0.001996975002813366, + "loss": 0.1963, + "step": 3481 + }, + { + "epoch": 0.030225432070902163, + "grad_norm": 0.10400390625, + "learning_rate": 0.001996972564176545, + "loss": 0.1855, + "step": 3482 + }, + { + "epoch": 0.030234112551106328, + "grad_norm": 0.171875, + "learning_rate": 0.001996970124558808, + "loss": 0.1631, + "step": 3483 + }, + { + "epoch": 0.030242793031310493, + "grad_norm": 0.09619140625, + "learning_rate": 0.001996967683960158, + "loss": 0.2031, + "step": 3484 + }, + { + "epoch": 0.030251473511514658, + "grad_norm": 0.21875, + "learning_rate": 0.0019969652423805976, + "loss": 0.1914, + "step": 3485 + }, + { + "epoch": 0.030260153991718823, + "grad_norm": 0.330078125, + "learning_rate": 0.00199696279982013, + "loss": 0.2051, + "step": 3486 + }, + { + "epoch": 0.030268834471922988, + "grad_norm": 0.107421875, + "learning_rate": 0.001996960356278757, + "loss": 0.1826, + "step": 3487 + }, + { + "epoch": 0.030277514952127153, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019969579117564812, + "loss": 0.1846, + "step": 3488 + }, + { + "epoch": 0.030286195432331315, + "grad_norm": 0.4140625, + "learning_rate": 0.001996955466253306, + "loss": 0.2422, + "step": 3489 + }, + { + "epoch": 0.03029487591253548, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019969530197692337, + "loss": 0.1562, + "step": 3490 + }, + { + "epoch": 0.030303556392739645, + "grad_norm": 0.185546875, + "learning_rate": 0.0019969505723042677, + "loss": 0.1689, + "step": 3491 + }, + { + "epoch": 0.03031223687294381, + "grad_norm": 0.33984375, + "learning_rate": 0.0019969481238584093, + "loss": 0.1826, + "step": 3492 + }, + { + "epoch": 0.030320917353147975, + "grad_norm": 0.0859375, + "learning_rate": 0.001996945674431662, + "loss": 0.1338, + "step": 3493 + }, + { + "epoch": 0.03032959783335214, + "grad_norm": 0.353515625, + "learning_rate": 0.0019969432240240286, + "loss": 0.1885, + "step": 3494 + }, + { + "epoch": 0.030338278313556306, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019969407726355116, + "loss": 0.207, + "step": 3495 + }, + { + "epoch": 0.03034695879376047, + "grad_norm": 0.26953125, + "learning_rate": 0.0019969383202661137, + "loss": 0.1826, + "step": 3496 + }, + { + "epoch": 0.030355639273964636, + "grad_norm": 0.86328125, + "learning_rate": 0.0019969358669158373, + "loss": 0.2031, + "step": 3497 + }, + { + "epoch": 0.0303643197541688, + "grad_norm": 0.80078125, + "learning_rate": 0.0019969334125846854, + "loss": 0.2344, + "step": 3498 + }, + { + "epoch": 0.030373000234372966, + "grad_norm": 0.2890625, + "learning_rate": 0.0019969309572726605, + "loss": 0.1699, + "step": 3499 + }, + { + "epoch": 0.03038168071457713, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019969285009797657, + "loss": 0.1934, + "step": 3500 + }, + { + "epoch": 0.030390361194781296, + "grad_norm": 0.1435546875, + "learning_rate": 0.001996926043706003, + "loss": 0.1855, + "step": 3501 + }, + { + "epoch": 0.03039904167498546, + "grad_norm": 0.375, + "learning_rate": 0.0019969235854513756, + "loss": 0.1592, + "step": 3502 + }, + { + "epoch": 0.030407722155189627, + "grad_norm": 0.1494140625, + "learning_rate": 0.001996921126215886, + "loss": 0.2051, + "step": 3503 + }, + { + "epoch": 0.03041640263539379, + "grad_norm": 0.59765625, + "learning_rate": 0.001996918665999537, + "loss": 0.1855, + "step": 3504 + }, + { + "epoch": 0.030425083115597953, + "grad_norm": 0.29296875, + "learning_rate": 0.001996916204802331, + "loss": 0.1992, + "step": 3505 + }, + { + "epoch": 0.03043376359580212, + "grad_norm": 0.1884765625, + "learning_rate": 0.001996913742624271, + "loss": 0.1787, + "step": 3506 + }, + { + "epoch": 0.030442444076006284, + "grad_norm": 0.0859375, + "learning_rate": 0.00199691127946536, + "loss": 0.1973, + "step": 3507 + }, + { + "epoch": 0.03045112455621045, + "grad_norm": 0.419921875, + "learning_rate": 0.0019969088153256, + "loss": 0.1865, + "step": 3508 + }, + { + "epoch": 0.030459805036414614, + "grad_norm": 0.337890625, + "learning_rate": 0.001996906350204994, + "loss": 0.1582, + "step": 3509 + }, + { + "epoch": 0.03046848551661878, + "grad_norm": 0.375, + "learning_rate": 0.0019969038841035447, + "loss": 0.1602, + "step": 3510 + }, + { + "epoch": 0.030477165996822944, + "grad_norm": 0.1806640625, + "learning_rate": 0.001996901417021255, + "loss": 0.1543, + "step": 3511 + }, + { + "epoch": 0.03048584647702711, + "grad_norm": 0.1259765625, + "learning_rate": 0.001996898948958127, + "loss": 0.1992, + "step": 3512 + }, + { + "epoch": 0.030494526957231274, + "grad_norm": 0.640625, + "learning_rate": 0.0019968964799141637, + "loss": 0.3164, + "step": 3513 + }, + { + "epoch": 0.03050320743743544, + "grad_norm": 0.2109375, + "learning_rate": 0.0019968940098893683, + "loss": 0.1504, + "step": 3514 + }, + { + "epoch": 0.030511887917639605, + "grad_norm": 0.09375, + "learning_rate": 0.001996891538883743, + "loss": 0.1455, + "step": 3515 + }, + { + "epoch": 0.03052056839784377, + "grad_norm": 0.26171875, + "learning_rate": 0.00199688906689729, + "loss": 0.1846, + "step": 3516 + }, + { + "epoch": 0.030529248878047935, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019968865939300135, + "loss": 0.168, + "step": 3517 + }, + { + "epoch": 0.0305379293582521, + "grad_norm": 0.86328125, + "learning_rate": 0.0019968841199819146, + "loss": 0.1885, + "step": 3518 + }, + { + "epoch": 0.030546609838456265, + "grad_norm": 0.173828125, + "learning_rate": 0.0019968816450529974, + "loss": 0.1484, + "step": 3519 + }, + { + "epoch": 0.030555290318660427, + "grad_norm": 0.2431640625, + "learning_rate": 0.001996879169143263, + "loss": 0.1621, + "step": 3520 + }, + { + "epoch": 0.030563970798864592, + "grad_norm": 0.1005859375, + "learning_rate": 0.001996876692252716, + "loss": 0.1973, + "step": 3521 + }, + { + "epoch": 0.030572651279068757, + "grad_norm": 0.45703125, + "learning_rate": 0.0019968742143813578, + "loss": 0.2539, + "step": 3522 + }, + { + "epoch": 0.030581331759272922, + "grad_norm": 0.158203125, + "learning_rate": 0.001996871735529191, + "loss": 0.1641, + "step": 3523 + }, + { + "epoch": 0.030590012239477087, + "grad_norm": 0.07275390625, + "learning_rate": 0.001996869255696219, + "loss": 0.1953, + "step": 3524 + }, + { + "epoch": 0.030598692719681252, + "grad_norm": 0.125, + "learning_rate": 0.0019968667748824446, + "loss": 0.1484, + "step": 3525 + }, + { + "epoch": 0.030607373199885417, + "grad_norm": 0.130859375, + "learning_rate": 0.0019968642930878696, + "loss": 0.1914, + "step": 3526 + }, + { + "epoch": 0.030616053680089583, + "grad_norm": 0.3359375, + "learning_rate": 0.0019968618103124976, + "loss": 0.1719, + "step": 3527 + }, + { + "epoch": 0.030624734160293748, + "grad_norm": 0.0927734375, + "learning_rate": 0.001996859326556331, + "loss": 0.1973, + "step": 3528 + }, + { + "epoch": 0.030633414640497913, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019968568418193724, + "loss": 0.1738, + "step": 3529 + }, + { + "epoch": 0.030642095120702078, + "grad_norm": 0.30078125, + "learning_rate": 0.001996854356101625, + "loss": 0.2949, + "step": 3530 + }, + { + "epoch": 0.030650775600906243, + "grad_norm": 0.84765625, + "learning_rate": 0.0019968518694030908, + "loss": 0.2598, + "step": 3531 + }, + { + "epoch": 0.030659456081110408, + "grad_norm": 0.30859375, + "learning_rate": 0.0019968493817237726, + "loss": 0.1826, + "step": 3532 + }, + { + "epoch": 0.030668136561314573, + "grad_norm": 0.27734375, + "learning_rate": 0.001996846893063674, + "loss": 0.1279, + "step": 3533 + }, + { + "epoch": 0.03067681704151874, + "grad_norm": 0.474609375, + "learning_rate": 0.0019968444034227967, + "loss": 0.208, + "step": 3534 + }, + { + "epoch": 0.0306854975217229, + "grad_norm": 0.208984375, + "learning_rate": 0.0019968419128011438, + "loss": 0.1816, + "step": 3535 + }, + { + "epoch": 0.030694178001927065, + "grad_norm": 0.205078125, + "learning_rate": 0.0019968394211987185, + "loss": 0.1748, + "step": 3536 + }, + { + "epoch": 0.03070285848213123, + "grad_norm": 0.63671875, + "learning_rate": 0.0019968369286155227, + "loss": 0.2031, + "step": 3537 + }, + { + "epoch": 0.030711538962335395, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019968344350515593, + "loss": 0.1787, + "step": 3538 + }, + { + "epoch": 0.03072021944253956, + "grad_norm": 0.072265625, + "learning_rate": 0.0019968319405068314, + "loss": 0.1729, + "step": 3539 + }, + { + "epoch": 0.030728899922743726, + "grad_norm": 0.1845703125, + "learning_rate": 0.001996829444981342, + "loss": 0.1553, + "step": 3540 + }, + { + "epoch": 0.03073758040294789, + "grad_norm": 0.58203125, + "learning_rate": 0.0019968269484750933, + "loss": 0.2041, + "step": 3541 + }, + { + "epoch": 0.030746260883152056, + "grad_norm": 0.1728515625, + "learning_rate": 0.001996824450988088, + "loss": 0.1426, + "step": 3542 + }, + { + "epoch": 0.03075494136335622, + "grad_norm": 0.283203125, + "learning_rate": 0.0019968219525203284, + "loss": 0.2051, + "step": 3543 + }, + { + "epoch": 0.030763621843560386, + "grad_norm": 0.158203125, + "learning_rate": 0.0019968194530718183, + "loss": 0.248, + "step": 3544 + }, + { + "epoch": 0.03077230232376455, + "grad_norm": 0.10009765625, + "learning_rate": 0.00199681695264256, + "loss": 0.1748, + "step": 3545 + }, + { + "epoch": 0.030780982803968716, + "grad_norm": 0.353515625, + "learning_rate": 0.001996814451232556, + "loss": 0.1689, + "step": 3546 + }, + { + "epoch": 0.03078966328417288, + "grad_norm": 0.28125, + "learning_rate": 0.0019968119488418096, + "loss": 0.2305, + "step": 3547 + }, + { + "epoch": 0.030798343764377047, + "grad_norm": 0.28125, + "learning_rate": 0.0019968094454703225, + "loss": 0.2812, + "step": 3548 + }, + { + "epoch": 0.030807024244581212, + "grad_norm": 0.310546875, + "learning_rate": 0.0019968069411180982, + "loss": 0.2266, + "step": 3549 + }, + { + "epoch": 0.030815704724785373, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019968044357851398, + "loss": 0.1562, + "step": 3550 + }, + { + "epoch": 0.03082438520498954, + "grad_norm": 1.21875, + "learning_rate": 0.001996801929471449, + "loss": 0.2539, + "step": 3551 + }, + { + "epoch": 0.030833065685193704, + "grad_norm": 0.059326171875, + "learning_rate": 0.001996799422177029, + "loss": 0.1523, + "step": 3552 + }, + { + "epoch": 0.03084174616539787, + "grad_norm": 0.70703125, + "learning_rate": 0.0019967969139018833, + "loss": 0.2695, + "step": 3553 + }, + { + "epoch": 0.030850426645602034, + "grad_norm": 0.31640625, + "learning_rate": 0.0019967944046460134, + "loss": 0.1602, + "step": 3554 + }, + { + "epoch": 0.0308591071258062, + "grad_norm": 0.55078125, + "learning_rate": 0.0019967918944094228, + "loss": 0.1963, + "step": 3555 + }, + { + "epoch": 0.030867787606010364, + "grad_norm": 0.35546875, + "learning_rate": 0.001996789383192114, + "loss": 0.1777, + "step": 3556 + }, + { + "epoch": 0.03087646808621453, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019967868709940897, + "loss": 0.1543, + "step": 3557 + }, + { + "epoch": 0.030885148566418694, + "grad_norm": 0.77734375, + "learning_rate": 0.001996784357815353, + "loss": 0.2324, + "step": 3558 + }, + { + "epoch": 0.03089382904662286, + "grad_norm": 0.189453125, + "learning_rate": 0.0019967818436559064, + "loss": 0.168, + "step": 3559 + }, + { + "epoch": 0.030902509526827025, + "grad_norm": 0.2353515625, + "learning_rate": 0.001996779328515752, + "loss": 0.1719, + "step": 3560 + }, + { + "epoch": 0.03091119000703119, + "grad_norm": 0.216796875, + "learning_rate": 0.001996776812394894, + "loss": 0.1777, + "step": 3561 + }, + { + "epoch": 0.030919870487235355, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019967742952933343, + "loss": 0.2041, + "step": 3562 + }, + { + "epoch": 0.03092855096743952, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019967717772110755, + "loss": 0.1787, + "step": 3563 + }, + { + "epoch": 0.030937231447643685, + "grad_norm": 0.423828125, + "learning_rate": 0.0019967692581481203, + "loss": 0.2461, + "step": 3564 + }, + { + "epoch": 0.03094591192784785, + "grad_norm": 0.06982421875, + "learning_rate": 0.001996766738104472, + "loss": 0.2129, + "step": 3565 + }, + { + "epoch": 0.030954592408052012, + "grad_norm": 0.265625, + "learning_rate": 0.001996764217080133, + "loss": 0.1729, + "step": 3566 + }, + { + "epoch": 0.030963272888256177, + "grad_norm": 0.1494140625, + "learning_rate": 0.001996761695075106, + "loss": 0.1797, + "step": 3567 + }, + { + "epoch": 0.030971953368460342, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019967591720893942, + "loss": 0.1738, + "step": 3568 + }, + { + "epoch": 0.030980633848664507, + "grad_norm": 0.49609375, + "learning_rate": 0.001996756648123, + "loss": 0.1982, + "step": 3569 + }, + { + "epoch": 0.030989314328868672, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019967541231759264, + "loss": 0.2227, + "step": 3570 + }, + { + "epoch": 0.030997994809072837, + "grad_norm": 0.28515625, + "learning_rate": 0.001996751597248175, + "loss": 0.166, + "step": 3571 + }, + { + "epoch": 0.031006675289277003, + "grad_norm": 1.796875, + "learning_rate": 0.001996749070339751, + "loss": 0.3945, + "step": 3572 + }, + { + "epoch": 0.031015355769481168, + "grad_norm": 0.30859375, + "learning_rate": 0.0019967465424506545, + "loss": 0.1719, + "step": 3573 + }, + { + "epoch": 0.031024036249685333, + "grad_norm": 0.2578125, + "learning_rate": 0.00199674401358089, + "loss": 0.207, + "step": 3574 + }, + { + "epoch": 0.031032716729889498, + "grad_norm": 0.328125, + "learning_rate": 0.0019967414837304596, + "loss": 0.2188, + "step": 3575 + }, + { + "epoch": 0.031041397210093663, + "grad_norm": 0.359375, + "learning_rate": 0.001996738952899366, + "loss": 0.2002, + "step": 3576 + }, + { + "epoch": 0.031050077690297828, + "grad_norm": 1.1484375, + "learning_rate": 0.001996736421087612, + "loss": 0.2207, + "step": 3577 + }, + { + "epoch": 0.031058758170501993, + "grad_norm": 0.08935546875, + "learning_rate": 0.001996733888295201, + "loss": 0.1924, + "step": 3578 + }, + { + "epoch": 0.03106743865070616, + "grad_norm": 0.27734375, + "learning_rate": 0.001996731354522135, + "loss": 0.2363, + "step": 3579 + }, + { + "epoch": 0.031076119130910324, + "grad_norm": 0.578125, + "learning_rate": 0.0019967288197684173, + "loss": 0.1953, + "step": 3580 + }, + { + "epoch": 0.031084799611114485, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019967262840340505, + "loss": 0.2422, + "step": 3581 + }, + { + "epoch": 0.03109348009131865, + "grad_norm": 0.16015625, + "learning_rate": 0.0019967237473190367, + "loss": 0.1914, + "step": 3582 + }, + { + "epoch": 0.031102160571522815, + "grad_norm": 0.1083984375, + "learning_rate": 0.00199672120962338, + "loss": 0.2402, + "step": 3583 + }, + { + "epoch": 0.03111084105172698, + "grad_norm": 0.359375, + "learning_rate": 0.001996718670947082, + "loss": 0.3438, + "step": 3584 + }, + { + "epoch": 0.031119521531931146, + "grad_norm": 0.119140625, + "learning_rate": 0.0019967161312901462, + "loss": 0.1533, + "step": 3585 + }, + { + "epoch": 0.03112820201213531, + "grad_norm": 0.400390625, + "learning_rate": 0.001996713590652575, + "loss": 0.2617, + "step": 3586 + }, + { + "epoch": 0.031136882492339476, + "grad_norm": 0.328125, + "learning_rate": 0.001996711049034371, + "loss": 0.248, + "step": 3587 + }, + { + "epoch": 0.03114556297254364, + "grad_norm": 0.30078125, + "learning_rate": 0.001996708506435538, + "loss": 0.1787, + "step": 3588 + }, + { + "epoch": 0.031154243452747806, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019967059628560775, + "loss": 0.1416, + "step": 3589 + }, + { + "epoch": 0.03116292393295197, + "grad_norm": 0.31640625, + "learning_rate": 0.0019967034182959927, + "loss": 0.167, + "step": 3590 + }, + { + "epoch": 0.031171604413156136, + "grad_norm": 0.29296875, + "learning_rate": 0.001996700872755287, + "loss": 0.167, + "step": 3591 + }, + { + "epoch": 0.0311802848933603, + "grad_norm": 0.162109375, + "learning_rate": 0.001996698326233962, + "loss": 0.1758, + "step": 3592 + }, + { + "epoch": 0.031188965373564467, + "grad_norm": 0.1728515625, + "learning_rate": 0.001996695778732022, + "loss": 0.1631, + "step": 3593 + }, + { + "epoch": 0.03119764585376863, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019966932302494687, + "loss": 0.1914, + "step": 3594 + }, + { + "epoch": 0.031206326333972797, + "grad_norm": 0.59375, + "learning_rate": 0.0019966906807863047, + "loss": 0.1729, + "step": 3595 + }, + { + "epoch": 0.031215006814176962, + "grad_norm": 0.09130859375, + "learning_rate": 0.001996688130342534, + "loss": 0.1963, + "step": 3596 + }, + { + "epoch": 0.031223687294381124, + "grad_norm": 0.166015625, + "learning_rate": 0.001996685578918158, + "loss": 0.1699, + "step": 3597 + }, + { + "epoch": 0.03123236777458529, + "grad_norm": 0.296875, + "learning_rate": 0.0019966830265131805, + "loss": 0.209, + "step": 3598 + }, + { + "epoch": 0.031241048254789454, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019966804731276037, + "loss": 0.1836, + "step": 3599 + }, + { + "epoch": 0.03124972873499362, + "grad_norm": 0.337890625, + "learning_rate": 0.0019966779187614307, + "loss": 0.2793, + "step": 3600 + }, + { + "epoch": 0.031258409215197784, + "grad_norm": 0.142578125, + "learning_rate": 0.001996675363414664, + "loss": 0.1797, + "step": 3601 + }, + { + "epoch": 0.03126708969540195, + "grad_norm": 0.267578125, + "learning_rate": 0.001996672807087307, + "loss": 0.1768, + "step": 3602 + }, + { + "epoch": 0.031275770175606114, + "grad_norm": 0.09716796875, + "learning_rate": 0.001996670249779362, + "loss": 0.1963, + "step": 3603 + }, + { + "epoch": 0.03128445065581028, + "grad_norm": 0.283203125, + "learning_rate": 0.0019966676914908313, + "loss": 0.1758, + "step": 3604 + }, + { + "epoch": 0.031293131136014445, + "grad_norm": 0.365234375, + "learning_rate": 0.0019966651322217187, + "loss": 0.2559, + "step": 3605 + }, + { + "epoch": 0.03130181161621861, + "grad_norm": 0.326171875, + "learning_rate": 0.001996662571972027, + "loss": 0.2207, + "step": 3606 + }, + { + "epoch": 0.031310492096422775, + "grad_norm": 0.2265625, + "learning_rate": 0.001996660010741758, + "loss": 0.1504, + "step": 3607 + }, + { + "epoch": 0.03131917257662694, + "grad_norm": 0.39453125, + "learning_rate": 0.0019966574485309153, + "loss": 0.1641, + "step": 3608 + }, + { + "epoch": 0.031327853056831105, + "grad_norm": 0.1279296875, + "learning_rate": 0.001996654885339502, + "loss": 0.1914, + "step": 3609 + }, + { + "epoch": 0.03133653353703527, + "grad_norm": 0.1796875, + "learning_rate": 0.0019966523211675195, + "loss": 0.1523, + "step": 3610 + }, + { + "epoch": 0.031345214017239435, + "grad_norm": 0.486328125, + "learning_rate": 0.001996649756014972, + "loss": 0.2031, + "step": 3611 + }, + { + "epoch": 0.0313538944974436, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019966471898818614, + "loss": 0.1377, + "step": 3612 + }, + { + "epoch": 0.031362574977647766, + "grad_norm": 0.3359375, + "learning_rate": 0.0019966446227681913, + "loss": 0.2031, + "step": 3613 + }, + { + "epoch": 0.03137125545785193, + "grad_norm": 0.4453125, + "learning_rate": 0.001996642054673964, + "loss": 0.1777, + "step": 3614 + }, + { + "epoch": 0.031379935938056096, + "grad_norm": 0.380859375, + "learning_rate": 0.0019966394855991825, + "loss": 0.1885, + "step": 3615 + }, + { + "epoch": 0.03138861641826026, + "grad_norm": 0.427734375, + "learning_rate": 0.0019966369155438495, + "loss": 0.248, + "step": 3616 + }, + { + "epoch": 0.031397296898464426, + "grad_norm": 0.08203125, + "learning_rate": 0.001996634344507968, + "loss": 0.1543, + "step": 3617 + }, + { + "epoch": 0.03140597737866859, + "grad_norm": 0.4609375, + "learning_rate": 0.0019966317724915404, + "loss": 0.1797, + "step": 3618 + }, + { + "epoch": 0.031414657858872756, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019966291994945695, + "loss": 0.1729, + "step": 3619 + }, + { + "epoch": 0.031423338339076914, + "grad_norm": 0.126953125, + "learning_rate": 0.0019966266255170588, + "loss": 0.1709, + "step": 3620 + }, + { + "epoch": 0.03143201881928108, + "grad_norm": 0.119140625, + "learning_rate": 0.0019966240505590103, + "loss": 0.1738, + "step": 3621 + }, + { + "epoch": 0.031440699299485245, + "grad_norm": 0.111328125, + "learning_rate": 0.0019966214746204277, + "loss": 0.1855, + "step": 3622 + }, + { + "epoch": 0.03144937977968941, + "grad_norm": 0.08837890625, + "learning_rate": 0.001996618897701313, + "loss": 0.2324, + "step": 3623 + }, + { + "epoch": 0.031458060259893575, + "grad_norm": 0.07958984375, + "learning_rate": 0.001996616319801669, + "loss": 0.1641, + "step": 3624 + }, + { + "epoch": 0.03146674074009774, + "grad_norm": 0.5078125, + "learning_rate": 0.0019966137409215, + "loss": 0.1797, + "step": 3625 + }, + { + "epoch": 0.031475421220301905, + "grad_norm": 0.546875, + "learning_rate": 0.0019966111610608068, + "loss": 0.2168, + "step": 3626 + }, + { + "epoch": 0.03148410170050607, + "grad_norm": 0.4609375, + "learning_rate": 0.0019966085802195933, + "loss": 0.2773, + "step": 3627 + }, + { + "epoch": 0.031492782180710235, + "grad_norm": 0.46875, + "learning_rate": 0.001996605998397862, + "loss": 0.2148, + "step": 3628 + }, + { + "epoch": 0.0315014626609144, + "grad_norm": 0.2431640625, + "learning_rate": 0.001996603415595616, + "loss": 0.1816, + "step": 3629 + }, + { + "epoch": 0.031510143141118566, + "grad_norm": 0.232421875, + "learning_rate": 0.001996600831812858, + "loss": 0.1826, + "step": 3630 + }, + { + "epoch": 0.03151882362132273, + "grad_norm": 0.259765625, + "learning_rate": 0.001996598247049591, + "loss": 0.1465, + "step": 3631 + }, + { + "epoch": 0.031527504101526896, + "grad_norm": 0.25, + "learning_rate": 0.0019965956613058173, + "loss": 0.1719, + "step": 3632 + }, + { + "epoch": 0.03153618458173106, + "grad_norm": 0.404296875, + "learning_rate": 0.00199659307458154, + "loss": 0.1953, + "step": 3633 + }, + { + "epoch": 0.031544865061935226, + "grad_norm": 0.640625, + "learning_rate": 0.0019965904868767623, + "loss": 0.2324, + "step": 3634 + }, + { + "epoch": 0.03155354554213939, + "grad_norm": 0.068359375, + "learning_rate": 0.0019965878981914867, + "loss": 0.1484, + "step": 3635 + }, + { + "epoch": 0.031562226022343556, + "grad_norm": 0.16796875, + "learning_rate": 0.0019965853085257153, + "loss": 0.2002, + "step": 3636 + }, + { + "epoch": 0.03157090650254772, + "grad_norm": 0.5234375, + "learning_rate": 0.0019965827178794527, + "loss": 0.21, + "step": 3637 + }, + { + "epoch": 0.03157958698275189, + "grad_norm": 0.314453125, + "learning_rate": 0.0019965801262527005, + "loss": 0.2295, + "step": 3638 + }, + { + "epoch": 0.03158826746295605, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019965775336454614, + "loss": 0.166, + "step": 3639 + }, + { + "epoch": 0.03159694794316022, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019965749400577383, + "loss": 0.1445, + "step": 3640 + }, + { + "epoch": 0.03160562842336438, + "grad_norm": 0.197265625, + "learning_rate": 0.001996572345489535, + "loss": 0.1699, + "step": 3641 + }, + { + "epoch": 0.03161430890356855, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019965697499408535, + "loss": 0.1846, + "step": 3642 + }, + { + "epoch": 0.03162298938377271, + "grad_norm": 0.44140625, + "learning_rate": 0.0019965671534116964, + "loss": 0.418, + "step": 3643 + }, + { + "epoch": 0.03163166986397688, + "grad_norm": 0.1669921875, + "learning_rate": 0.001996564555902067, + "loss": 0.1787, + "step": 3644 + }, + { + "epoch": 0.03164035034418104, + "grad_norm": 0.296875, + "learning_rate": 0.0019965619574119687, + "loss": 0.1924, + "step": 3645 + }, + { + "epoch": 0.03164903082438521, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019965593579414033, + "loss": 0.1826, + "step": 3646 + }, + { + "epoch": 0.03165771130458937, + "grad_norm": 0.09814453125, + "learning_rate": 0.001996556757490374, + "loss": 0.1855, + "step": 3647 + }, + { + "epoch": 0.03166639178479354, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019965541560588835, + "loss": 0.1729, + "step": 3648 + }, + { + "epoch": 0.0316750722649977, + "grad_norm": 0.13671875, + "learning_rate": 0.001996551553646935, + "loss": 0.1387, + "step": 3649 + }, + { + "epoch": 0.03168375274520187, + "grad_norm": 0.251953125, + "learning_rate": 0.0019965489502545316, + "loss": 0.1758, + "step": 3650 + }, + { + "epoch": 0.031692433225406026, + "grad_norm": 0.26953125, + "learning_rate": 0.0019965463458816754, + "loss": 0.1885, + "step": 3651 + }, + { + "epoch": 0.03170111370561019, + "grad_norm": 0.330078125, + "learning_rate": 0.0019965437405283695, + "loss": 0.1885, + "step": 3652 + }, + { + "epoch": 0.031709794185814356, + "grad_norm": 0.443359375, + "learning_rate": 0.001996541134194617, + "loss": 0.2031, + "step": 3653 + }, + { + "epoch": 0.03171847466601852, + "grad_norm": 0.162109375, + "learning_rate": 0.0019965385268804208, + "loss": 0.166, + "step": 3654 + }, + { + "epoch": 0.03172715514622269, + "grad_norm": 0.408203125, + "learning_rate": 0.001996535918585783, + "loss": 0.2207, + "step": 3655 + }, + { + "epoch": 0.03173583562642685, + "grad_norm": 0.376953125, + "learning_rate": 0.0019965333093107072, + "loss": 0.2461, + "step": 3656 + }, + { + "epoch": 0.03174451610663102, + "grad_norm": 0.1875, + "learning_rate": 0.001996530699055196, + "loss": 0.1719, + "step": 3657 + }, + { + "epoch": 0.03175319658683518, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019965280878192523, + "loss": 0.1621, + "step": 3658 + }, + { + "epoch": 0.03176187706703935, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019965254756028794, + "loss": 0.1689, + "step": 3659 + }, + { + "epoch": 0.03177055754724351, + "grad_norm": 0.1826171875, + "learning_rate": 0.001996522862406079, + "loss": 0.1924, + "step": 3660 + }, + { + "epoch": 0.03177923802744768, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019965202482288553, + "loss": 0.1826, + "step": 3661 + }, + { + "epoch": 0.03178791850765184, + "grad_norm": 0.6015625, + "learning_rate": 0.00199651763307121, + "loss": 0.2217, + "step": 3662 + }, + { + "epoch": 0.03179659898785601, + "grad_norm": 0.1357421875, + "learning_rate": 0.001996515016933147, + "loss": 0.1689, + "step": 3663 + }, + { + "epoch": 0.03180527946806017, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019965123998146686, + "loss": 0.1621, + "step": 3664 + }, + { + "epoch": 0.03181395994826434, + "grad_norm": 0.171875, + "learning_rate": 0.0019965097817157772, + "loss": 0.2051, + "step": 3665 + }, + { + "epoch": 0.0318226404284685, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019965071626364766, + "loss": 0.1777, + "step": 3666 + }, + { + "epoch": 0.03183132090867267, + "grad_norm": 0.5546875, + "learning_rate": 0.0019965045425767687, + "loss": 0.1777, + "step": 3667 + }, + { + "epoch": 0.03184000138887683, + "grad_norm": 0.291015625, + "learning_rate": 0.0019965019215366577, + "loss": 0.2051, + "step": 3668 + }, + { + "epoch": 0.031848681869081, + "grad_norm": 0.2294921875, + "learning_rate": 0.001996499299516145, + "loss": 0.1895, + "step": 3669 + }, + { + "epoch": 0.031857362349285163, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019964966765152344, + "loss": 0.1689, + "step": 3670 + }, + { + "epoch": 0.03186604282948933, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019964940525339287, + "loss": 0.1582, + "step": 3671 + }, + { + "epoch": 0.031874723309693494, + "grad_norm": 0.34375, + "learning_rate": 0.00199649142757223, + "loss": 0.1689, + "step": 3672 + }, + { + "epoch": 0.03188340378989766, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019964888016301427, + "loss": 0.1934, + "step": 3673 + }, + { + "epoch": 0.031892084270101824, + "grad_norm": 0.48046875, + "learning_rate": 0.001996486174707668, + "loss": 0.1855, + "step": 3674 + }, + { + "epoch": 0.03190076475030599, + "grad_norm": 0.361328125, + "learning_rate": 0.0019964835468048096, + "loss": 0.3066, + "step": 3675 + }, + { + "epoch": 0.031909445230510154, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019964809179215705, + "loss": 0.2217, + "step": 3676 + }, + { + "epoch": 0.03191812571071432, + "grad_norm": 1.703125, + "learning_rate": 0.0019964782880579532, + "loss": 0.2539, + "step": 3677 + }, + { + "epoch": 0.031926806190918484, + "grad_norm": 0.13671875, + "learning_rate": 0.0019964756572139605, + "loss": 0.2383, + "step": 3678 + }, + { + "epoch": 0.03193548667112265, + "grad_norm": 0.50390625, + "learning_rate": 0.0019964730253895957, + "loss": 0.1895, + "step": 3679 + }, + { + "epoch": 0.031944167151326815, + "grad_norm": 0.65234375, + "learning_rate": 0.0019964703925848615, + "loss": 0.2227, + "step": 3680 + }, + { + "epoch": 0.03195284763153097, + "grad_norm": 0.41015625, + "learning_rate": 0.0019964677587997605, + "loss": 0.168, + "step": 3681 + }, + { + "epoch": 0.03196152811173514, + "grad_norm": 0.296875, + "learning_rate": 0.001996465124034296, + "loss": 0.1514, + "step": 3682 + }, + { + "epoch": 0.0319702085919393, + "grad_norm": 0.1806640625, + "learning_rate": 0.001996462488288471, + "loss": 0.1914, + "step": 3683 + }, + { + "epoch": 0.03197888907214347, + "grad_norm": 0.3203125, + "learning_rate": 0.0019964598515622876, + "loss": 0.1543, + "step": 3684 + }, + { + "epoch": 0.03198756955234763, + "grad_norm": 0.134765625, + "learning_rate": 0.001996457213855749, + "loss": 0.2041, + "step": 3685 + }, + { + "epoch": 0.0319962500325518, + "grad_norm": 0.154296875, + "learning_rate": 0.001996454575168859, + "loss": 0.209, + "step": 3686 + }, + { + "epoch": 0.032004930512755964, + "grad_norm": 0.095703125, + "learning_rate": 0.001996451935501619, + "loss": 0.1387, + "step": 3687 + }, + { + "epoch": 0.03201361099296013, + "grad_norm": 0.1572265625, + "learning_rate": 0.001996449294854033, + "loss": 0.1904, + "step": 3688 + }, + { + "epoch": 0.032022291473164294, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019964466532261037, + "loss": 0.2012, + "step": 3689 + }, + { + "epoch": 0.03203097195336846, + "grad_norm": 0.2294921875, + "learning_rate": 0.001996444010617834, + "loss": 0.1885, + "step": 3690 + }, + { + "epoch": 0.032039652433572624, + "grad_norm": 0.64453125, + "learning_rate": 0.001996441367029226, + "loss": 0.1963, + "step": 3691 + }, + { + "epoch": 0.03204833291377679, + "grad_norm": 0.337890625, + "learning_rate": 0.0019964387224602836, + "loss": 0.1602, + "step": 3692 + }, + { + "epoch": 0.032057013393980954, + "grad_norm": 0.2236328125, + "learning_rate": 0.001996436076911009, + "loss": 0.2383, + "step": 3693 + }, + { + "epoch": 0.03206569387418512, + "grad_norm": 0.310546875, + "learning_rate": 0.001996433430381405, + "loss": 0.1797, + "step": 3694 + }, + { + "epoch": 0.032074374354389285, + "grad_norm": 0.0908203125, + "learning_rate": 0.001996430782871476, + "loss": 0.1758, + "step": 3695 + }, + { + "epoch": 0.03208305483459345, + "grad_norm": 0.2001953125, + "learning_rate": 0.001996428134381223, + "loss": 0.1807, + "step": 3696 + }, + { + "epoch": 0.032091735314797615, + "grad_norm": 0.16796875, + "learning_rate": 0.0019964254849106494, + "loss": 0.1572, + "step": 3697 + }, + { + "epoch": 0.03210041579500178, + "grad_norm": 0.341796875, + "learning_rate": 0.001996422834459759, + "loss": 0.1855, + "step": 3698 + }, + { + "epoch": 0.032109096275205945, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019964201830285537, + "loss": 0.1758, + "step": 3699 + }, + { + "epoch": 0.03211777675541011, + "grad_norm": 0.126953125, + "learning_rate": 0.0019964175306170368, + "loss": 0.2002, + "step": 3700 + }, + { + "epoch": 0.032126457235614275, + "grad_norm": 0.32421875, + "learning_rate": 0.0019964148772252115, + "loss": 0.1738, + "step": 3701 + }, + { + "epoch": 0.03213513771581844, + "grad_norm": 0.5, + "learning_rate": 0.00199641222285308, + "loss": 0.1807, + "step": 3702 + }, + { + "epoch": 0.032143818196022605, + "grad_norm": 0.66015625, + "learning_rate": 0.0019964095675006456, + "loss": 0.1807, + "step": 3703 + }, + { + "epoch": 0.03215249867622677, + "grad_norm": 0.150390625, + "learning_rate": 0.001996406911167911, + "loss": 0.1963, + "step": 3704 + }, + { + "epoch": 0.032161179156430936, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019964042538548796, + "loss": 0.2031, + "step": 3705 + }, + { + "epoch": 0.0321698596366351, + "grad_norm": 0.5, + "learning_rate": 0.0019964015955615537, + "loss": 0.2119, + "step": 3706 + }, + { + "epoch": 0.032178540116839266, + "grad_norm": 0.2080078125, + "learning_rate": 0.001996398936287937, + "loss": 0.1572, + "step": 3707 + }, + { + "epoch": 0.03218722059704343, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019963962760340312, + "loss": 0.1572, + "step": 3708 + }, + { + "epoch": 0.032195901077247596, + "grad_norm": 0.095703125, + "learning_rate": 0.0019963936147998403, + "loss": 0.1699, + "step": 3709 + }, + { + "epoch": 0.03220458155745176, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019963909525853667, + "loss": 0.1455, + "step": 3710 + }, + { + "epoch": 0.032213262037655926, + "grad_norm": 0.25390625, + "learning_rate": 0.0019963882893906135, + "loss": 0.2227, + "step": 3711 + }, + { + "epoch": 0.032221942517860085, + "grad_norm": 0.3359375, + "learning_rate": 0.0019963856252155836, + "loss": 0.1611, + "step": 3712 + }, + { + "epoch": 0.03223062299806425, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019963829600602796, + "loss": 0.1797, + "step": 3713 + }, + { + "epoch": 0.032239303478268415, + "grad_norm": 0.55078125, + "learning_rate": 0.0019963802939247047, + "loss": 0.2012, + "step": 3714 + }, + { + "epoch": 0.03224798395847258, + "grad_norm": 0.34375, + "learning_rate": 0.001996377626808862, + "loss": 0.1611, + "step": 3715 + }, + { + "epoch": 0.032256664438676745, + "grad_norm": 0.1513671875, + "learning_rate": 0.001996374958712754, + "loss": 0.1816, + "step": 3716 + }, + { + "epoch": 0.03226534491888091, + "grad_norm": 0.337890625, + "learning_rate": 0.001996372289636384, + "loss": 0.209, + "step": 3717 + }, + { + "epoch": 0.032274025399085075, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019963696195797547, + "loss": 0.2246, + "step": 3718 + }, + { + "epoch": 0.03228270587928924, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019963669485428687, + "loss": 0.1533, + "step": 3719 + }, + { + "epoch": 0.032291386359493406, + "grad_norm": 0.5546875, + "learning_rate": 0.00199636427652573, + "loss": 0.2441, + "step": 3720 + }, + { + "epoch": 0.03230006683969757, + "grad_norm": 0.388671875, + "learning_rate": 0.00199636160352834, + "loss": 0.1914, + "step": 3721 + }, + { + "epoch": 0.032308747319901736, + "grad_norm": 0.203125, + "learning_rate": 0.0019963589295507026, + "loss": 0.1934, + "step": 3722 + }, + { + "epoch": 0.0323174278001059, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019963562545928207, + "loss": 0.1641, + "step": 3723 + }, + { + "epoch": 0.032326108280310066, + "grad_norm": 0.11328125, + "learning_rate": 0.0019963535786546973, + "loss": 0.1973, + "step": 3724 + }, + { + "epoch": 0.03233478876051423, + "grad_norm": 0.22265625, + "learning_rate": 0.0019963509017363345, + "loss": 0.1562, + "step": 3725 + }, + { + "epoch": 0.032343469240718396, + "grad_norm": 0.205078125, + "learning_rate": 0.0019963482238377363, + "loss": 0.2031, + "step": 3726 + }, + { + "epoch": 0.03235214972092256, + "grad_norm": 0.318359375, + "learning_rate": 0.001996345544958905, + "loss": 0.1582, + "step": 3727 + }, + { + "epoch": 0.03236083020112673, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019963428650998437, + "loss": 0.1738, + "step": 3728 + }, + { + "epoch": 0.03236951068133089, + "grad_norm": 0.22265625, + "learning_rate": 0.0019963401842605557, + "loss": 0.1719, + "step": 3729 + }, + { + "epoch": 0.03237819116153506, + "grad_norm": 0.78125, + "learning_rate": 0.0019963375024410427, + "loss": 0.2773, + "step": 3730 + }, + { + "epoch": 0.03238687164173922, + "grad_norm": 0.18359375, + "learning_rate": 0.001996334819641309, + "loss": 0.1885, + "step": 3731 + }, + { + "epoch": 0.03239555212194339, + "grad_norm": 0.42578125, + "learning_rate": 0.001996332135861357, + "loss": 0.1719, + "step": 3732 + }, + { + "epoch": 0.03240423260214755, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019963294511011896, + "loss": 0.1934, + "step": 3733 + }, + { + "epoch": 0.03241291308235172, + "grad_norm": 0.1357421875, + "learning_rate": 0.00199632676536081, + "loss": 0.2051, + "step": 3734 + }, + { + "epoch": 0.03242159356255588, + "grad_norm": 0.3359375, + "learning_rate": 0.0019963240786402205, + "loss": 0.2207, + "step": 3735 + }, + { + "epoch": 0.03243027404276005, + "grad_norm": 0.08203125, + "learning_rate": 0.0019963213909394246, + "loss": 0.1807, + "step": 3736 + }, + { + "epoch": 0.03243895452296421, + "grad_norm": 0.193359375, + "learning_rate": 0.0019963187022584254, + "loss": 0.2227, + "step": 3737 + }, + { + "epoch": 0.03244763500316838, + "grad_norm": 0.08154296875, + "learning_rate": 0.001996316012597225, + "loss": 0.1689, + "step": 3738 + }, + { + "epoch": 0.03245631548337254, + "grad_norm": 0.1591796875, + "learning_rate": 0.001996313321955827, + "loss": 0.1738, + "step": 3739 + }, + { + "epoch": 0.03246499596357671, + "grad_norm": 0.275390625, + "learning_rate": 0.0019963106303342347, + "loss": 0.2119, + "step": 3740 + }, + { + "epoch": 0.03247367644378087, + "grad_norm": 0.13671875, + "learning_rate": 0.00199630793773245, + "loss": 0.2285, + "step": 3741 + }, + { + "epoch": 0.03248235692398504, + "grad_norm": 0.65234375, + "learning_rate": 0.0019963052441504766, + "loss": 0.2363, + "step": 3742 + }, + { + "epoch": 0.032491037404189196, + "grad_norm": 0.146484375, + "learning_rate": 0.0019963025495883173, + "loss": 0.1895, + "step": 3743 + }, + { + "epoch": 0.03249971788439336, + "grad_norm": 1.0546875, + "learning_rate": 0.001996299854045975, + "loss": 0.4746, + "step": 3744 + }, + { + "epoch": 0.03250839836459753, + "grad_norm": 0.185546875, + "learning_rate": 0.0019962971575234524, + "loss": 0.2285, + "step": 3745 + }, + { + "epoch": 0.03251707884480169, + "grad_norm": 0.130859375, + "learning_rate": 0.0019962944600207533, + "loss": 0.1973, + "step": 3746 + }, + { + "epoch": 0.03252575932500586, + "grad_norm": 0.166015625, + "learning_rate": 0.00199629176153788, + "loss": 0.1836, + "step": 3747 + }, + { + "epoch": 0.03253443980521002, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019962890620748345, + "loss": 0.2002, + "step": 3748 + }, + { + "epoch": 0.03254312028541419, + "grad_norm": 0.345703125, + "learning_rate": 0.0019962863616316217, + "loss": 0.1533, + "step": 3749 + }, + { + "epoch": 0.03255180076561835, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019962836602082433, + "loss": 0.1934, + "step": 3750 + }, + { + "epoch": 0.03256048124582252, + "grad_norm": 0.359375, + "learning_rate": 0.0019962809578047023, + "loss": 0.2285, + "step": 3751 + }, + { + "epoch": 0.03256916172602668, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019962782544210023, + "loss": 0.1953, + "step": 3752 + }, + { + "epoch": 0.03257784220623085, + "grad_norm": 0.12890625, + "learning_rate": 0.0019962755500571457, + "loss": 0.1924, + "step": 3753 + }, + { + "epoch": 0.03258652268643501, + "grad_norm": 0.185546875, + "learning_rate": 0.0019962728447131357, + "loss": 0.166, + "step": 3754 + }, + { + "epoch": 0.03259520316663918, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019962701383889753, + "loss": 0.1562, + "step": 3755 + }, + { + "epoch": 0.03260388364684334, + "grad_norm": 0.1015625, + "learning_rate": 0.001996267431084667, + "loss": 0.2207, + "step": 3756 + }, + { + "epoch": 0.03261256412704751, + "grad_norm": 0.9609375, + "learning_rate": 0.0019962647228002144, + "loss": 0.2324, + "step": 3757 + }, + { + "epoch": 0.03262124460725167, + "grad_norm": 0.09765625, + "learning_rate": 0.00199626201353562, + "loss": 0.1924, + "step": 3758 + }, + { + "epoch": 0.03262992508745584, + "grad_norm": 0.55859375, + "learning_rate": 0.001996259303290887, + "loss": 0.1797, + "step": 3759 + }, + { + "epoch": 0.03263860556766, + "grad_norm": 0.306640625, + "learning_rate": 0.0019962565920660187, + "loss": 0.1348, + "step": 3760 + }, + { + "epoch": 0.03264728604786417, + "grad_norm": 0.357421875, + "learning_rate": 0.001996253879861017, + "loss": 0.2344, + "step": 3761 + }, + { + "epoch": 0.032655966528068334, + "grad_norm": 0.095703125, + "learning_rate": 0.001996251166675886, + "loss": 0.2148, + "step": 3762 + }, + { + "epoch": 0.0326646470082725, + "grad_norm": 0.11328125, + "learning_rate": 0.001996248452510628, + "loss": 0.166, + "step": 3763 + }, + { + "epoch": 0.032673327488476664, + "grad_norm": 0.103515625, + "learning_rate": 0.001996245737365246, + "loss": 0.1943, + "step": 3764 + }, + { + "epoch": 0.03268200796868083, + "grad_norm": 0.361328125, + "learning_rate": 0.001996243021239743, + "loss": 0.1592, + "step": 3765 + }, + { + "epoch": 0.032690688448884994, + "grad_norm": 0.06640625, + "learning_rate": 0.0019962403041341222, + "loss": 0.1895, + "step": 3766 + }, + { + "epoch": 0.03269936892908916, + "grad_norm": 0.470703125, + "learning_rate": 0.001996237586048387, + "loss": 0.1836, + "step": 3767 + }, + { + "epoch": 0.032708049409293324, + "grad_norm": 0.2080078125, + "learning_rate": 0.001996234866982539, + "loss": 0.2002, + "step": 3768 + }, + { + "epoch": 0.03271672988949749, + "grad_norm": 0.166015625, + "learning_rate": 0.001996232146936583, + "loss": 0.209, + "step": 3769 + }, + { + "epoch": 0.032725410369701655, + "grad_norm": 0.22265625, + "learning_rate": 0.0019962294259105204, + "loss": 0.1855, + "step": 3770 + }, + { + "epoch": 0.03273409084990582, + "grad_norm": 0.328125, + "learning_rate": 0.0019962267039043543, + "loss": 0.1729, + "step": 3771 + }, + { + "epoch": 0.032742771330109985, + "grad_norm": 0.11181640625, + "learning_rate": 0.001996223980918089, + "loss": 0.1934, + "step": 3772 + }, + { + "epoch": 0.03275145181031415, + "grad_norm": 0.146484375, + "learning_rate": 0.0019962212569517262, + "loss": 0.2129, + "step": 3773 + }, + { + "epoch": 0.03276013229051831, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019962185320052694, + "loss": 0.1602, + "step": 3774 + }, + { + "epoch": 0.03276881277072247, + "grad_norm": 0.3515625, + "learning_rate": 0.0019962158060787215, + "loss": 0.2285, + "step": 3775 + }, + { + "epoch": 0.03277749325092664, + "grad_norm": 0.111328125, + "learning_rate": 0.0019962130791720853, + "loss": 0.2227, + "step": 3776 + }, + { + "epoch": 0.032786173731130804, + "grad_norm": 0.0908203125, + "learning_rate": 0.001996210351285364, + "loss": 0.1699, + "step": 3777 + }, + { + "epoch": 0.03279485421133497, + "grad_norm": 0.08984375, + "learning_rate": 0.0019962076224185605, + "loss": 0.2119, + "step": 3778 + }, + { + "epoch": 0.032803534691539134, + "grad_norm": 0.2353515625, + "learning_rate": 0.001996204892571678, + "loss": 0.1738, + "step": 3779 + }, + { + "epoch": 0.0328122151717433, + "grad_norm": 0.126953125, + "learning_rate": 0.001996202161744719, + "loss": 0.1504, + "step": 3780 + }, + { + "epoch": 0.032820895651947464, + "grad_norm": 0.5234375, + "learning_rate": 0.001996199429937687, + "loss": 0.1445, + "step": 3781 + }, + { + "epoch": 0.03282957613215163, + "grad_norm": 0.60546875, + "learning_rate": 0.001996196697150585, + "loss": 0.207, + "step": 3782 + }, + { + "epoch": 0.032838256612355794, + "grad_norm": 0.408203125, + "learning_rate": 0.001996193963383416, + "loss": 0.209, + "step": 3783 + }, + { + "epoch": 0.03284693709255996, + "grad_norm": 0.0791015625, + "learning_rate": 0.001996191228636182, + "loss": 0.2227, + "step": 3784 + }, + { + "epoch": 0.032855617572764124, + "grad_norm": 0.396484375, + "learning_rate": 0.0019961884929088868, + "loss": 0.2324, + "step": 3785 + }, + { + "epoch": 0.03286429805296829, + "grad_norm": 0.283203125, + "learning_rate": 0.0019961857562015334, + "loss": 0.1846, + "step": 3786 + }, + { + "epoch": 0.032872978533172455, + "grad_norm": 0.33984375, + "learning_rate": 0.0019961830185141246, + "loss": 0.2246, + "step": 3787 + }, + { + "epoch": 0.03288165901337662, + "grad_norm": 0.0927734375, + "learning_rate": 0.001996180279846664, + "loss": 0.2031, + "step": 3788 + }, + { + "epoch": 0.032890339493580785, + "grad_norm": 0.375, + "learning_rate": 0.001996177540199154, + "loss": 0.1816, + "step": 3789 + }, + { + "epoch": 0.03289901997378495, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019961747995715976, + "loss": 0.1953, + "step": 3790 + }, + { + "epoch": 0.032907700453989115, + "grad_norm": 0.29296875, + "learning_rate": 0.001996172057963998, + "loss": 0.2051, + "step": 3791 + }, + { + "epoch": 0.03291638093419328, + "grad_norm": 0.5234375, + "learning_rate": 0.001996169315376358, + "loss": 0.2051, + "step": 3792 + }, + { + "epoch": 0.032925061414397445, + "grad_norm": 0.375, + "learning_rate": 0.0019961665718086804, + "loss": 0.1973, + "step": 3793 + }, + { + "epoch": 0.03293374189460161, + "grad_norm": 0.0986328125, + "learning_rate": 0.001996163827260969, + "loss": 0.1963, + "step": 3794 + }, + { + "epoch": 0.032942422374805776, + "grad_norm": 0.30078125, + "learning_rate": 0.001996161081733226, + "loss": 0.168, + "step": 3795 + }, + { + "epoch": 0.03295110285500994, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019961583352254546, + "loss": 0.1777, + "step": 3796 + }, + { + "epoch": 0.032959783335214106, + "grad_norm": 0.421875, + "learning_rate": 0.0019961555877376583, + "loss": 0.1807, + "step": 3797 + }, + { + "epoch": 0.03296846381541827, + "grad_norm": 0.111328125, + "learning_rate": 0.0019961528392698396, + "loss": 0.1533, + "step": 3798 + }, + { + "epoch": 0.032977144295622436, + "grad_norm": 0.296875, + "learning_rate": 0.0019961500898220013, + "loss": 0.2061, + "step": 3799 + }, + { + "epoch": 0.0329858247758266, + "grad_norm": 0.41015625, + "learning_rate": 0.001996147339394147, + "loss": 0.1885, + "step": 3800 + }, + { + "epoch": 0.032994505256030766, + "grad_norm": 0.62109375, + "learning_rate": 0.0019961445879862795, + "loss": 0.2285, + "step": 3801 + }, + { + "epoch": 0.03300318573623493, + "grad_norm": 0.625, + "learning_rate": 0.0019961418355984016, + "loss": 0.1973, + "step": 3802 + }, + { + "epoch": 0.0330118662164391, + "grad_norm": 0.259765625, + "learning_rate": 0.0019961390822305167, + "loss": 0.1611, + "step": 3803 + }, + { + "epoch": 0.03302054669664326, + "grad_norm": 0.10498046875, + "learning_rate": 0.001996136327882627, + "loss": 0.1777, + "step": 3804 + }, + { + "epoch": 0.03302922717684742, + "grad_norm": 0.51171875, + "learning_rate": 0.001996133572554737, + "loss": 0.1602, + "step": 3805 + }, + { + "epoch": 0.033037907657051585, + "grad_norm": 0.5078125, + "learning_rate": 0.0019961308162468475, + "loss": 0.3379, + "step": 3806 + }, + { + "epoch": 0.03304658813725575, + "grad_norm": 0.32421875, + "learning_rate": 0.001996128058958964, + "loss": 0.2559, + "step": 3807 + }, + { + "epoch": 0.033055268617459915, + "grad_norm": 0.79296875, + "learning_rate": 0.0019961253006910876, + "loss": 0.1621, + "step": 3808 + }, + { + "epoch": 0.03306394909766408, + "grad_norm": 0.205078125, + "learning_rate": 0.0019961225414432226, + "loss": 0.1621, + "step": 3809 + }, + { + "epoch": 0.033072629577868246, + "grad_norm": 0.119140625, + "learning_rate": 0.0019961197812153712, + "loss": 0.1758, + "step": 3810 + }, + { + "epoch": 0.03308131005807241, + "grad_norm": 0.48046875, + "learning_rate": 0.0019961170200075364, + "loss": 0.1729, + "step": 3811 + }, + { + "epoch": 0.033089990538276576, + "grad_norm": 0.65625, + "learning_rate": 0.0019961142578197215, + "loss": 0.2246, + "step": 3812 + }, + { + "epoch": 0.03309867101848074, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019961114946519297, + "loss": 0.1543, + "step": 3813 + }, + { + "epoch": 0.033107351498684906, + "grad_norm": 0.2294921875, + "learning_rate": 0.001996108730504164, + "loss": 0.1641, + "step": 3814 + }, + { + "epoch": 0.03311603197888907, + "grad_norm": 0.275390625, + "learning_rate": 0.001996105965376427, + "loss": 0.1953, + "step": 3815 + }, + { + "epoch": 0.033124712459093236, + "grad_norm": 0.08935546875, + "learning_rate": 0.001996103199268722, + "loss": 0.1582, + "step": 3816 + }, + { + "epoch": 0.0331333929392974, + "grad_norm": 0.77734375, + "learning_rate": 0.0019961004321810524, + "loss": 0.1631, + "step": 3817 + }, + { + "epoch": 0.033142073419501566, + "grad_norm": 0.12353515625, + "learning_rate": 0.00199609766411342, + "loss": 0.1572, + "step": 3818 + }, + { + "epoch": 0.03315075389970573, + "grad_norm": 0.267578125, + "learning_rate": 0.0019960948950658295, + "loss": 0.1777, + "step": 3819 + }, + { + "epoch": 0.0331594343799099, + "grad_norm": 0.2578125, + "learning_rate": 0.0019960921250382828, + "loss": 0.168, + "step": 3820 + }, + { + "epoch": 0.03316811486011406, + "grad_norm": 0.216796875, + "learning_rate": 0.0019960893540307834, + "loss": 0.168, + "step": 3821 + }, + { + "epoch": 0.03317679534031823, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019960865820433334, + "loss": 0.2129, + "step": 3822 + }, + { + "epoch": 0.03318547582052239, + "grad_norm": 0.25390625, + "learning_rate": 0.0019960838090759374, + "loss": 0.168, + "step": 3823 + }, + { + "epoch": 0.03319415630072656, + "grad_norm": 0.126953125, + "learning_rate": 0.0019960810351285973, + "loss": 0.2012, + "step": 3824 + }, + { + "epoch": 0.03320283678093072, + "grad_norm": 0.392578125, + "learning_rate": 0.0019960782602013163, + "loss": 0.1572, + "step": 3825 + }, + { + "epoch": 0.03321151726113489, + "grad_norm": 0.2001953125, + "learning_rate": 0.001996075484294098, + "loss": 0.1396, + "step": 3826 + }, + { + "epoch": 0.03322019774133905, + "grad_norm": 0.248046875, + "learning_rate": 0.0019960727074069444, + "loss": 0.1426, + "step": 3827 + }, + { + "epoch": 0.03322887822154322, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019960699295398596, + "loss": 0.1768, + "step": 3828 + }, + { + "epoch": 0.03323755870174738, + "grad_norm": 0.341796875, + "learning_rate": 0.001996067150692846, + "loss": 0.2148, + "step": 3829 + }, + { + "epoch": 0.03324623918195155, + "grad_norm": 0.259765625, + "learning_rate": 0.001996064370865907, + "loss": 0.2539, + "step": 3830 + }, + { + "epoch": 0.03325491966215571, + "grad_norm": 0.12109375, + "learning_rate": 0.0019960615900590454, + "loss": 0.1699, + "step": 3831 + }, + { + "epoch": 0.03326360014235988, + "grad_norm": 0.255859375, + "learning_rate": 0.001996058808272264, + "loss": 0.1328, + "step": 3832 + }, + { + "epoch": 0.03327228062256404, + "grad_norm": 0.400390625, + "learning_rate": 0.001996056025505567, + "loss": 0.1406, + "step": 3833 + }, + { + "epoch": 0.03328096110276821, + "grad_norm": 0.29296875, + "learning_rate": 0.001996053241758956, + "loss": 0.1855, + "step": 3834 + }, + { + "epoch": 0.03328964158297237, + "grad_norm": 0.56640625, + "learning_rate": 0.0019960504570324345, + "loss": 0.1719, + "step": 3835 + }, + { + "epoch": 0.03329832206317653, + "grad_norm": 0.4609375, + "learning_rate": 0.0019960476713260056, + "loss": 0.1963, + "step": 3836 + }, + { + "epoch": 0.0333070025433807, + "grad_norm": 0.314453125, + "learning_rate": 0.001996044884639673, + "loss": 0.2305, + "step": 3837 + }, + { + "epoch": 0.03331568302358486, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019960420969734383, + "loss": 0.1543, + "step": 3838 + }, + { + "epoch": 0.03332436350378903, + "grad_norm": 0.12890625, + "learning_rate": 0.001996039308327306, + "loss": 0.1816, + "step": 3839 + }, + { + "epoch": 0.03333304398399319, + "grad_norm": 0.24609375, + "learning_rate": 0.0019960365187012786, + "loss": 0.1738, + "step": 3840 + }, + { + "epoch": 0.03334172446419736, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019960337280953595, + "loss": 0.1885, + "step": 3841 + }, + { + "epoch": 0.03335040494440152, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019960309365095507, + "loss": 0.166, + "step": 3842 + }, + { + "epoch": 0.03335908542460569, + "grad_norm": 0.08203125, + "learning_rate": 0.0019960281439438564, + "loss": 0.1641, + "step": 3843 + }, + { + "epoch": 0.03336776590480985, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019960253503982788, + "loss": 0.2324, + "step": 3844 + }, + { + "epoch": 0.03337644638501402, + "grad_norm": 0.58984375, + "learning_rate": 0.001996022555872822, + "loss": 0.2344, + "step": 3845 + }, + { + "epoch": 0.03338512686521818, + "grad_norm": 0.30859375, + "learning_rate": 0.0019960197603674876, + "loss": 0.1846, + "step": 3846 + }, + { + "epoch": 0.03339380734542235, + "grad_norm": 0.48828125, + "learning_rate": 0.0019960169638822797, + "loss": 0.1816, + "step": 3847 + }, + { + "epoch": 0.03340248782562651, + "grad_norm": 0.056640625, + "learning_rate": 0.001996014166417201, + "loss": 0.127, + "step": 3848 + }, + { + "epoch": 0.03341116830583068, + "grad_norm": 0.2119140625, + "learning_rate": 0.001996011367972255, + "loss": 0.1553, + "step": 3849 + }, + { + "epoch": 0.03341984878603484, + "grad_norm": 0.453125, + "learning_rate": 0.0019960085685474444, + "loss": 0.1592, + "step": 3850 + }, + { + "epoch": 0.03342852926623901, + "grad_norm": 0.1884765625, + "learning_rate": 0.001996005768142772, + "loss": 0.1729, + "step": 3851 + }, + { + "epoch": 0.033437209746443174, + "grad_norm": 0.16015625, + "learning_rate": 0.0019960029667582414, + "loss": 0.1572, + "step": 3852 + }, + { + "epoch": 0.03344589022664734, + "grad_norm": 0.291015625, + "learning_rate": 0.0019960001643938552, + "loss": 0.3203, + "step": 3853 + }, + { + "epoch": 0.033454570706851504, + "grad_norm": 0.263671875, + "learning_rate": 0.0019959973610496166, + "loss": 0.1572, + "step": 3854 + }, + { + "epoch": 0.03346325118705567, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019959945567255295, + "loss": 0.25, + "step": 3855 + }, + { + "epoch": 0.033471931667259834, + "grad_norm": 0.28125, + "learning_rate": 0.0019959917514215954, + "loss": 0.1523, + "step": 3856 + }, + { + "epoch": 0.033480612147464, + "grad_norm": 0.234375, + "learning_rate": 0.0019959889451378185, + "loss": 0.1377, + "step": 3857 + }, + { + "epoch": 0.033489292627668164, + "grad_norm": 0.109375, + "learning_rate": 0.0019959861378742016, + "loss": 0.1836, + "step": 3858 + }, + { + "epoch": 0.03349797310787233, + "grad_norm": 0.30859375, + "learning_rate": 0.001995983329630748, + "loss": 0.1914, + "step": 3859 + }, + { + "epoch": 0.033506653588076495, + "grad_norm": 0.216796875, + "learning_rate": 0.00199598052040746, + "loss": 0.1699, + "step": 3860 + }, + { + "epoch": 0.03351533406828066, + "grad_norm": 0.11083984375, + "learning_rate": 0.001995977710204341, + "loss": 0.1904, + "step": 3861 + }, + { + "epoch": 0.033524014548484825, + "grad_norm": 0.60546875, + "learning_rate": 0.0019959748990213944, + "loss": 0.1963, + "step": 3862 + }, + { + "epoch": 0.03353269502868899, + "grad_norm": 0.138671875, + "learning_rate": 0.0019959720868586235, + "loss": 0.1562, + "step": 3863 + }, + { + "epoch": 0.033541375508893155, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019959692737160305, + "loss": 0.165, + "step": 3864 + }, + { + "epoch": 0.03355005598909732, + "grad_norm": 0.4765625, + "learning_rate": 0.0019959664595936193, + "loss": 0.1875, + "step": 3865 + }, + { + "epoch": 0.03355873646930148, + "grad_norm": 0.458984375, + "learning_rate": 0.001995963644491393, + "loss": 0.2402, + "step": 3866 + }, + { + "epoch": 0.033567416949505643, + "grad_norm": 0.1484375, + "learning_rate": 0.0019959608284093534, + "loss": 0.168, + "step": 3867 + }, + { + "epoch": 0.03357609742970981, + "grad_norm": 0.6484375, + "learning_rate": 0.001995958011347505, + "loss": 0.1797, + "step": 3868 + }, + { + "epoch": 0.033584777909913974, + "grad_norm": 0.24609375, + "learning_rate": 0.0019959551933058503, + "loss": 0.2168, + "step": 3869 + }, + { + "epoch": 0.03359345839011814, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019959523742843926, + "loss": 0.1318, + "step": 3870 + }, + { + "epoch": 0.033602138870322304, + "grad_norm": 0.07080078125, + "learning_rate": 0.001995949554283135, + "loss": 0.1504, + "step": 3871 + }, + { + "epoch": 0.03361081935052647, + "grad_norm": 0.3203125, + "learning_rate": 0.00199594673330208, + "loss": 0.1719, + "step": 3872 + }, + { + "epoch": 0.033619499830730634, + "grad_norm": 0.349609375, + "learning_rate": 0.0019959439113412312, + "loss": 0.21, + "step": 3873 + }, + { + "epoch": 0.0336281803109348, + "grad_norm": 0.29296875, + "learning_rate": 0.0019959410884005917, + "loss": 0.1973, + "step": 3874 + }, + { + "epoch": 0.033636860791138964, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019959382644801644, + "loss": 0.1562, + "step": 3875 + }, + { + "epoch": 0.03364554127134313, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019959354395799526, + "loss": 0.168, + "step": 3876 + }, + { + "epoch": 0.033654221751547295, + "grad_norm": 0.232421875, + "learning_rate": 0.001995932613699959, + "loss": 0.1855, + "step": 3877 + }, + { + "epoch": 0.03366290223175146, + "grad_norm": 0.3359375, + "learning_rate": 0.0019959297868401876, + "loss": 0.1875, + "step": 3878 + }, + { + "epoch": 0.033671582711955625, + "grad_norm": 0.35546875, + "learning_rate": 0.00199592695900064, + "loss": 0.2031, + "step": 3879 + }, + { + "epoch": 0.03368026319215979, + "grad_norm": 0.10546875, + "learning_rate": 0.001995924130181321, + "loss": 0.1836, + "step": 3880 + }, + { + "epoch": 0.033688943672363955, + "grad_norm": 0.30078125, + "learning_rate": 0.001995921300382232, + "loss": 0.167, + "step": 3881 + }, + { + "epoch": 0.03369762415256812, + "grad_norm": 0.60546875, + "learning_rate": 0.0019959184696033776, + "loss": 0.2539, + "step": 3882 + }, + { + "epoch": 0.033706304632772285, + "grad_norm": 0.37890625, + "learning_rate": 0.0019959156378447597, + "loss": 0.1719, + "step": 3883 + }, + { + "epoch": 0.03371498511297645, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019959128051063825, + "loss": 0.1807, + "step": 3884 + }, + { + "epoch": 0.033723665593180616, + "grad_norm": 0.10888671875, + "learning_rate": 0.001995909971388248, + "loss": 0.1719, + "step": 3885 + }, + { + "epoch": 0.03373234607338478, + "grad_norm": 0.58203125, + "learning_rate": 0.00199590713669036, + "loss": 0.1367, + "step": 3886 + }, + { + "epoch": 0.033741026553588946, + "grad_norm": 0.396484375, + "learning_rate": 0.0019959043010127214, + "loss": 0.1836, + "step": 3887 + }, + { + "epoch": 0.03374970703379311, + "grad_norm": 0.408203125, + "learning_rate": 0.0019959014643553354, + "loss": 0.1602, + "step": 3888 + }, + { + "epoch": 0.033758387513997276, + "grad_norm": 0.171875, + "learning_rate": 0.001995898626718205, + "loss": 0.1641, + "step": 3889 + }, + { + "epoch": 0.03376706799420144, + "grad_norm": 0.1455078125, + "learning_rate": 0.001995895788101333, + "loss": 0.1855, + "step": 3890 + }, + { + "epoch": 0.033775748474405606, + "grad_norm": 0.57421875, + "learning_rate": 0.001995892948504723, + "loss": 0.1426, + "step": 3891 + }, + { + "epoch": 0.03378442895460977, + "grad_norm": 0.6328125, + "learning_rate": 0.001995890107928378, + "loss": 0.1934, + "step": 3892 + }, + { + "epoch": 0.03379310943481394, + "grad_norm": 0.53515625, + "learning_rate": 0.0019958872663723014, + "loss": 0.2031, + "step": 3893 + }, + { + "epoch": 0.0338017899150181, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019958844238364957, + "loss": 0.1992, + "step": 3894 + }, + { + "epoch": 0.03381047039522227, + "grad_norm": 0.4296875, + "learning_rate": 0.001995881580320964, + "loss": 0.1738, + "step": 3895 + }, + { + "epoch": 0.03381915087542643, + "grad_norm": 0.09912109375, + "learning_rate": 0.00199587873582571, + "loss": 0.1904, + "step": 3896 + }, + { + "epoch": 0.03382783135563059, + "grad_norm": 0.5859375, + "learning_rate": 0.001995875890350736, + "loss": 0.1777, + "step": 3897 + }, + { + "epoch": 0.033836511835834755, + "grad_norm": 0.12451171875, + "learning_rate": 0.001995873043896046, + "loss": 0.1611, + "step": 3898 + }, + { + "epoch": 0.03384519231603892, + "grad_norm": 0.55078125, + "learning_rate": 0.001995870196461643, + "loss": 0.168, + "step": 3899 + }, + { + "epoch": 0.033853872796243085, + "grad_norm": 0.462890625, + "learning_rate": 0.0019958673480475293, + "loss": 0.1846, + "step": 3900 + }, + { + "epoch": 0.03386255327644725, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019958644986537086, + "loss": 0.1611, + "step": 3901 + }, + { + "epoch": 0.033871233756651416, + "grad_norm": 0.21875, + "learning_rate": 0.0019958616482801837, + "loss": 0.2266, + "step": 3902 + }, + { + "epoch": 0.03387991423685558, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019958587969269585, + "loss": 0.1387, + "step": 3903 + }, + { + "epoch": 0.033888594717059746, + "grad_norm": 2.484375, + "learning_rate": 0.001995855944594035, + "loss": 0.457, + "step": 3904 + }, + { + "epoch": 0.03389727519726391, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019958530912814174, + "loss": 0.1553, + "step": 3905 + }, + { + "epoch": 0.033905955677468076, + "grad_norm": 0.197265625, + "learning_rate": 0.001995850236989108, + "loss": 0.1504, + "step": 3906 + }, + { + "epoch": 0.03391463615767224, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019958473817171104, + "loss": 0.1797, + "step": 3907 + }, + { + "epoch": 0.033923316637876406, + "grad_norm": 1.1484375, + "learning_rate": 0.0019958445254654278, + "loss": 0.5625, + "step": 3908 + }, + { + "epoch": 0.03393199711808057, + "grad_norm": 0.50390625, + "learning_rate": 0.0019958416682340626, + "loss": 0.1885, + "step": 3909 + }, + { + "epoch": 0.03394067759828474, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019958388100230185, + "loss": 0.1641, + "step": 3910 + }, + { + "epoch": 0.0339493580784889, + "grad_norm": 0.25390625, + "learning_rate": 0.001995835950832299, + "loss": 0.1738, + "step": 3911 + }, + { + "epoch": 0.03395803855869307, + "grad_norm": 0.1611328125, + "learning_rate": 0.001995833090661906, + "loss": 0.1318, + "step": 3912 + }, + { + "epoch": 0.03396671903889723, + "grad_norm": 0.4375, + "learning_rate": 0.001995830229511844, + "loss": 0.2266, + "step": 3913 + }, + { + "epoch": 0.0339753995191014, + "grad_norm": 0.091796875, + "learning_rate": 0.0019958273673821155, + "loss": 0.2119, + "step": 3914 + }, + { + "epoch": 0.03398407999930556, + "grad_norm": 0.75390625, + "learning_rate": 0.0019958245042727236, + "loss": 0.2578, + "step": 3915 + }, + { + "epoch": 0.03399276047950973, + "grad_norm": 0.384765625, + "learning_rate": 0.0019958216401836713, + "loss": 0.3477, + "step": 3916 + }, + { + "epoch": 0.03400144095971389, + "grad_norm": 0.12060546875, + "learning_rate": 0.001995818775114962, + "loss": 0.166, + "step": 3917 + }, + { + "epoch": 0.03401012143991806, + "grad_norm": 0.359375, + "learning_rate": 0.0019958159090665986, + "loss": 0.2031, + "step": 3918 + }, + { + "epoch": 0.03401880192012222, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019958130420385848, + "loss": 0.2461, + "step": 3919 + }, + { + "epoch": 0.03402748240032639, + "grad_norm": 0.10107421875, + "learning_rate": 0.001995810174030923, + "loss": 0.1689, + "step": 3920 + }, + { + "epoch": 0.03403616288053055, + "grad_norm": 0.7578125, + "learning_rate": 0.0019958073050436167, + "loss": 0.2402, + "step": 3921 + }, + { + "epoch": 0.03404484336073472, + "grad_norm": 0.2431640625, + "learning_rate": 0.001995804435076669, + "loss": 0.165, + "step": 3922 + }, + { + "epoch": 0.03405352384093888, + "grad_norm": 0.1904296875, + "learning_rate": 0.001995801564130083, + "loss": 0.1797, + "step": 3923 + }, + { + "epoch": 0.03406220432114305, + "grad_norm": 0.671875, + "learning_rate": 0.001995798692203862, + "loss": 0.2305, + "step": 3924 + }, + { + "epoch": 0.034070884801347213, + "grad_norm": 0.373046875, + "learning_rate": 0.0019957958192980086, + "loss": 0.1992, + "step": 3925 + }, + { + "epoch": 0.03407956528155138, + "grad_norm": 0.3203125, + "learning_rate": 0.001995792945412527, + "loss": 0.2129, + "step": 3926 + }, + { + "epoch": 0.034088245761755544, + "grad_norm": 0.50390625, + "learning_rate": 0.0019957900705474194, + "loss": 0.21, + "step": 3927 + }, + { + "epoch": 0.0340969262419597, + "grad_norm": 0.2734375, + "learning_rate": 0.001995787194702689, + "loss": 0.1553, + "step": 3928 + }, + { + "epoch": 0.03410560672216387, + "grad_norm": 0.92578125, + "learning_rate": 0.0019957843178783395, + "loss": 0.4922, + "step": 3929 + }, + { + "epoch": 0.03411428720236803, + "grad_norm": 0.439453125, + "learning_rate": 0.0019957814400743735, + "loss": 0.1816, + "step": 3930 + }, + { + "epoch": 0.0341229676825722, + "grad_norm": 0.2412109375, + "learning_rate": 0.001995778561290795, + "loss": 0.2246, + "step": 3931 + }, + { + "epoch": 0.03413164816277636, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019957756815276056, + "loss": 0.1738, + "step": 3932 + }, + { + "epoch": 0.03414032864298053, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019957728007848094, + "loss": 0.2012, + "step": 3933 + }, + { + "epoch": 0.03414900912318469, + "grad_norm": 0.16796875, + "learning_rate": 0.00199576991906241, + "loss": 0.1631, + "step": 3934 + }, + { + "epoch": 0.03415768960338886, + "grad_norm": 0.1435546875, + "learning_rate": 0.00199576703636041, + "loss": 0.1572, + "step": 3935 + }, + { + "epoch": 0.03416637008359302, + "grad_norm": 0.1875, + "learning_rate": 0.0019957641526788124, + "loss": 0.2207, + "step": 3936 + }, + { + "epoch": 0.03417505056379719, + "grad_norm": 0.08251953125, + "learning_rate": 0.001995761268017621, + "loss": 0.1953, + "step": 3937 + }, + { + "epoch": 0.03418373104400135, + "grad_norm": 0.140625, + "learning_rate": 0.0019957583823768383, + "loss": 0.25, + "step": 3938 + }, + { + "epoch": 0.03419241152420552, + "grad_norm": 0.431640625, + "learning_rate": 0.0019957554957564677, + "loss": 0.1641, + "step": 3939 + }, + { + "epoch": 0.03420109200440968, + "grad_norm": 0.49609375, + "learning_rate": 0.0019957526081565125, + "loss": 0.2168, + "step": 3940 + }, + { + "epoch": 0.03420977248461385, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019957497195769755, + "loss": 0.1777, + "step": 3941 + }, + { + "epoch": 0.034218452964818014, + "grad_norm": 0.482421875, + "learning_rate": 0.0019957468300178605, + "loss": 0.2383, + "step": 3942 + }, + { + "epoch": 0.03422713344502218, + "grad_norm": 0.279296875, + "learning_rate": 0.0019957439394791693, + "loss": 0.1719, + "step": 3943 + }, + { + "epoch": 0.034235813925226344, + "grad_norm": 0.359375, + "learning_rate": 0.001995741047960907, + "loss": 0.1484, + "step": 3944 + }, + { + "epoch": 0.03424449440543051, + "grad_norm": 0.326171875, + "learning_rate": 0.001995738155463075, + "loss": 0.1729, + "step": 3945 + }, + { + "epoch": 0.034253174885634674, + "grad_norm": 0.232421875, + "learning_rate": 0.0019957352619856777, + "loss": 0.2422, + "step": 3946 + }, + { + "epoch": 0.03426185536583884, + "grad_norm": 0.08203125, + "learning_rate": 0.0019957323675287176, + "loss": 0.1504, + "step": 3947 + }, + { + "epoch": 0.034270535846043004, + "grad_norm": 0.150390625, + "learning_rate": 0.0019957294720921977, + "loss": 0.1621, + "step": 3948 + }, + { + "epoch": 0.03427921632624717, + "grad_norm": 0.66015625, + "learning_rate": 0.001995726575676122, + "loss": 0.167, + "step": 3949 + }, + { + "epoch": 0.034287896806451335, + "grad_norm": 0.6875, + "learning_rate": 0.001995723678280493, + "loss": 0.1992, + "step": 3950 + }, + { + "epoch": 0.0342965772866555, + "grad_norm": 0.1875, + "learning_rate": 0.001995720779905314, + "loss": 0.1963, + "step": 3951 + }, + { + "epoch": 0.034305257766859665, + "grad_norm": 1.6171875, + "learning_rate": 0.0019957178805505883, + "loss": 0.2852, + "step": 3952 + }, + { + "epoch": 0.03431393824706383, + "grad_norm": 0.5234375, + "learning_rate": 0.0019957149802163187, + "loss": 0.2402, + "step": 3953 + }, + { + "epoch": 0.034322618727267995, + "grad_norm": 0.30078125, + "learning_rate": 0.001995712078902509, + "loss": 0.1709, + "step": 3954 + }, + { + "epoch": 0.03433129920747216, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019957091766091618, + "loss": 0.1445, + "step": 3955 + }, + { + "epoch": 0.034339979687676325, + "grad_norm": 0.7421875, + "learning_rate": 0.0019957062733362806, + "loss": 0.1748, + "step": 3956 + }, + { + "epoch": 0.03434866016788049, + "grad_norm": 0.796875, + "learning_rate": 0.0019957033690838686, + "loss": 0.209, + "step": 3957 + }, + { + "epoch": 0.034357340648084655, + "grad_norm": 0.443359375, + "learning_rate": 0.001995700463851929, + "loss": 0.1914, + "step": 3958 + }, + { + "epoch": 0.034366021128288814, + "grad_norm": 0.404296875, + "learning_rate": 0.0019956975576404645, + "loss": 0.1816, + "step": 3959 + }, + { + "epoch": 0.03437470160849298, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019956946504494787, + "loss": 0.1992, + "step": 3960 + }, + { + "epoch": 0.034383382088697144, + "grad_norm": 0.240234375, + "learning_rate": 0.001995691742278975, + "loss": 0.1963, + "step": 3961 + }, + { + "epoch": 0.03439206256890131, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019956888331289554, + "loss": 0.1621, + "step": 3962 + }, + { + "epoch": 0.034400743049105474, + "grad_norm": 0.1533203125, + "learning_rate": 0.001995685922999425, + "loss": 0.165, + "step": 3963 + }, + { + "epoch": 0.03440942352930964, + "grad_norm": 0.1103515625, + "learning_rate": 0.001995683011890385, + "loss": 0.1699, + "step": 3964 + }, + { + "epoch": 0.034418104009513804, + "grad_norm": 0.38671875, + "learning_rate": 0.00199568009980184, + "loss": 0.207, + "step": 3965 + }, + { + "epoch": 0.03442678448971797, + "grad_norm": 0.197265625, + "learning_rate": 0.0019956771867337924, + "loss": 0.2012, + "step": 3966 + }, + { + "epoch": 0.034435464969922135, + "grad_norm": 0.1474609375, + "learning_rate": 0.001995674272686246, + "loss": 0.2246, + "step": 3967 + }, + { + "epoch": 0.0344441454501263, + "grad_norm": 0.5546875, + "learning_rate": 0.001995671357659204, + "loss": 0.1455, + "step": 3968 + }, + { + "epoch": 0.034452825930330465, + "grad_norm": 0.37890625, + "learning_rate": 0.001995668441652669, + "loss": 0.1797, + "step": 3969 + }, + { + "epoch": 0.03446150641053463, + "grad_norm": 0.337890625, + "learning_rate": 0.001995665524666644, + "loss": 0.1738, + "step": 3970 + }, + { + "epoch": 0.034470186890738795, + "grad_norm": 0.6171875, + "learning_rate": 0.001995662606701133, + "loss": 0.1982, + "step": 3971 + }, + { + "epoch": 0.03447886737094296, + "grad_norm": 0.64453125, + "learning_rate": 0.001995659687756139, + "loss": 0.248, + "step": 3972 + }, + { + "epoch": 0.034487547851147125, + "grad_norm": 0.267578125, + "learning_rate": 0.0019956567678316647, + "loss": 0.1992, + "step": 3973 + }, + { + "epoch": 0.03449622833135129, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019956538469277137, + "loss": 0.1475, + "step": 3974 + }, + { + "epoch": 0.034504908811555456, + "grad_norm": 0.4921875, + "learning_rate": 0.0019956509250442893, + "loss": 0.1543, + "step": 3975 + }, + { + "epoch": 0.03451358929175962, + "grad_norm": 0.16796875, + "learning_rate": 0.001995648002181394, + "loss": 0.1533, + "step": 3976 + }, + { + "epoch": 0.034522269771963786, + "grad_norm": 0.3359375, + "learning_rate": 0.0019956450783390318, + "loss": 0.1963, + "step": 3977 + }, + { + "epoch": 0.03453095025216795, + "grad_norm": 0.396484375, + "learning_rate": 0.0019956421535172056, + "loss": 0.1797, + "step": 3978 + }, + { + "epoch": 0.034539630732372116, + "grad_norm": 0.6875, + "learning_rate": 0.0019956392277159186, + "loss": 0.207, + "step": 3979 + }, + { + "epoch": 0.03454831121257628, + "grad_norm": 0.55859375, + "learning_rate": 0.001995636300935174, + "loss": 0.2109, + "step": 3980 + }, + { + "epoch": 0.034556991692780446, + "grad_norm": 0.12109375, + "learning_rate": 0.0019956333731749754, + "loss": 0.1621, + "step": 3981 + }, + { + "epoch": 0.03456567217298461, + "grad_norm": 0.09521484375, + "learning_rate": 0.001995630444435325, + "loss": 0.1484, + "step": 3982 + }, + { + "epoch": 0.03457435265318878, + "grad_norm": 0.234375, + "learning_rate": 0.001995627514716227, + "loss": 0.2227, + "step": 3983 + }, + { + "epoch": 0.03458303313339294, + "grad_norm": 0.1162109375, + "learning_rate": 0.001995624584017684, + "loss": 0.2197, + "step": 3984 + }, + { + "epoch": 0.03459171361359711, + "grad_norm": 0.287109375, + "learning_rate": 0.0019956216523396996, + "loss": 0.2012, + "step": 3985 + }, + { + "epoch": 0.03460039409380127, + "grad_norm": 0.07421875, + "learning_rate": 0.0019956187196822767, + "loss": 0.1562, + "step": 3986 + }, + { + "epoch": 0.03460907457400544, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019956157860454185, + "loss": 0.1729, + "step": 3987 + }, + { + "epoch": 0.0346177550542096, + "grad_norm": 0.408203125, + "learning_rate": 0.0019956128514291285, + "loss": 0.1543, + "step": 3988 + }, + { + "epoch": 0.03462643553441376, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019956099158334097, + "loss": 0.1387, + "step": 3989 + }, + { + "epoch": 0.034635116014617925, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019956069792582652, + "loss": 0.1621, + "step": 3990 + }, + { + "epoch": 0.03464379649482209, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019956040417036984, + "loss": 0.1455, + "step": 3991 + }, + { + "epoch": 0.034652476975026256, + "grad_norm": 0.33984375, + "learning_rate": 0.001995601103169713, + "loss": 0.2129, + "step": 3992 + }, + { + "epoch": 0.03466115745523042, + "grad_norm": 0.33984375, + "learning_rate": 0.001995598163656311, + "loss": 0.1885, + "step": 3993 + }, + { + "epoch": 0.034669837935434586, + "grad_norm": 1.7109375, + "learning_rate": 0.0019955952231634965, + "loss": 0.3242, + "step": 3994 + }, + { + "epoch": 0.03467851841563875, + "grad_norm": 0.37109375, + "learning_rate": 0.0019955922816912728, + "loss": 0.2441, + "step": 3995 + }, + { + "epoch": 0.034687198895842916, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019955893392396423, + "loss": 0.2012, + "step": 3996 + }, + { + "epoch": 0.03469587937604708, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019955863958086096, + "loss": 0.1641, + "step": 3997 + }, + { + "epoch": 0.034704559856251246, + "grad_norm": 0.13671875, + "learning_rate": 0.001995583451398176, + "loss": 0.1719, + "step": 3998 + }, + { + "epoch": 0.03471324033645541, + "grad_norm": 0.38671875, + "learning_rate": 0.0019955805060083466, + "loss": 0.1748, + "step": 3999 + }, + { + "epoch": 0.03472192081665958, + "grad_norm": 0.73828125, + "learning_rate": 0.0019955775596391234, + "loss": 0.1484, + "step": 4000 + }, + { + "epoch": 0.03473060129686374, + "grad_norm": 1.1953125, + "learning_rate": 0.0019955746122905104, + "loss": 0.4453, + "step": 4001 + }, + { + "epoch": 0.03473928177706791, + "grad_norm": 0.255859375, + "learning_rate": 0.00199557166396251, + "loss": 0.2109, + "step": 4002 + }, + { + "epoch": 0.03474796225727207, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019955687146551264, + "loss": 0.2441, + "step": 4003 + }, + { + "epoch": 0.03475664273747624, + "grad_norm": 0.62109375, + "learning_rate": 0.001995565764368362, + "loss": 0.2324, + "step": 4004 + }, + { + "epoch": 0.0347653232176804, + "grad_norm": 0.2265625, + "learning_rate": 0.0019955628131022203, + "loss": 0.1318, + "step": 4005 + }, + { + "epoch": 0.03477400369788457, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019955598608567045, + "loss": 0.1758, + "step": 4006 + }, + { + "epoch": 0.03478268417808873, + "grad_norm": 0.08203125, + "learning_rate": 0.001995556907631818, + "loss": 0.1572, + "step": 4007 + }, + { + "epoch": 0.0347913646582929, + "grad_norm": 0.357421875, + "learning_rate": 0.001995553953427564, + "loss": 0.2021, + "step": 4008 + }, + { + "epoch": 0.03480004513849706, + "grad_norm": 0.142578125, + "learning_rate": 0.001995550998243946, + "loss": 0.2031, + "step": 4009 + }, + { + "epoch": 0.03480872561870123, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019955480420809665, + "loss": 0.1729, + "step": 4010 + }, + { + "epoch": 0.03481740609890539, + "grad_norm": 0.294921875, + "learning_rate": 0.001995545084938629, + "loss": 0.1357, + "step": 4011 + }, + { + "epoch": 0.03482608657910956, + "grad_norm": 0.291015625, + "learning_rate": 0.001995542126816937, + "loss": 0.1582, + "step": 4012 + }, + { + "epoch": 0.03483476705931372, + "grad_norm": 0.4765625, + "learning_rate": 0.0019955391677158934, + "loss": 0.1553, + "step": 4013 + }, + { + "epoch": 0.03484344753951789, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019955362076355017, + "loss": 0.1523, + "step": 4014 + }, + { + "epoch": 0.03485212801972205, + "grad_norm": 0.236328125, + "learning_rate": 0.0019955332465757653, + "loss": 0.1396, + "step": 4015 + }, + { + "epoch": 0.03486080849992622, + "grad_norm": 0.3125, + "learning_rate": 0.001995530284536687, + "loss": 0.1855, + "step": 4016 + }, + { + "epoch": 0.034869488980130384, + "grad_norm": 0.259765625, + "learning_rate": 0.0019955273215182703, + "loss": 0.1543, + "step": 4017 + }, + { + "epoch": 0.03487816946033455, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019955243575205182, + "loss": 0.1436, + "step": 4018 + }, + { + "epoch": 0.034886849940538714, + "grad_norm": 0.1640625, + "learning_rate": 0.0019955213925434345, + "loss": 0.1553, + "step": 4019 + }, + { + "epoch": 0.03489553042074287, + "grad_norm": 0.58203125, + "learning_rate": 0.0019955184265870217, + "loss": 0.1895, + "step": 4020 + }, + { + "epoch": 0.03490421090094704, + "grad_norm": 0.31640625, + "learning_rate": 0.0019955154596512834, + "loss": 0.165, + "step": 4021 + }, + { + "epoch": 0.0349128913811512, + "grad_norm": 0.376953125, + "learning_rate": 0.001995512491736223, + "loss": 0.1533, + "step": 4022 + }, + { + "epoch": 0.03492157186135537, + "grad_norm": 0.330078125, + "learning_rate": 0.0019955095228418435, + "loss": 0.2246, + "step": 4023 + }, + { + "epoch": 0.03493025234155953, + "grad_norm": 0.23828125, + "learning_rate": 0.001995506552968148, + "loss": 0.1299, + "step": 4024 + }, + { + "epoch": 0.0349389328217637, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019955035821151404, + "loss": 0.1357, + "step": 4025 + }, + { + "epoch": 0.03494761330196786, + "grad_norm": 0.62890625, + "learning_rate": 0.0019955006102828237, + "loss": 0.1699, + "step": 4026 + }, + { + "epoch": 0.03495629378217203, + "grad_norm": 0.20703125, + "learning_rate": 0.001995497637471201, + "loss": 0.1846, + "step": 4027 + }, + { + "epoch": 0.03496497426237619, + "grad_norm": 0.2734375, + "learning_rate": 0.0019954946636802752, + "loss": 0.1494, + "step": 4028 + }, + { + "epoch": 0.03497365474258036, + "grad_norm": 0.4296875, + "learning_rate": 0.0019954916889100495, + "loss": 0.2129, + "step": 4029 + }, + { + "epoch": 0.03498233522278452, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019954887131605286, + "loss": 0.1719, + "step": 4030 + }, + { + "epoch": 0.03499101570298869, + "grad_norm": 0.68359375, + "learning_rate": 0.0019954857364317138, + "loss": 0.1289, + "step": 4031 + }, + { + "epoch": 0.034999696183192854, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019954827587236094, + "loss": 0.1953, + "step": 4032 + }, + { + "epoch": 0.03500837666339702, + "grad_norm": 0.59375, + "learning_rate": 0.001995479780036219, + "loss": 0.1641, + "step": 4033 + }, + { + "epoch": 0.035017057143601184, + "grad_norm": 0.458984375, + "learning_rate": 0.001995476800369545, + "loss": 0.1719, + "step": 4034 + }, + { + "epoch": 0.03502573762380535, + "grad_norm": 0.2255859375, + "learning_rate": 0.001995473819723591, + "loss": 0.1797, + "step": 4035 + }, + { + "epoch": 0.035034418104009514, + "grad_norm": 0.234375, + "learning_rate": 0.00199547083809836, + "loss": 0.1787, + "step": 4036 + }, + { + "epoch": 0.03504309858421368, + "grad_norm": 0.11083984375, + "learning_rate": 0.001995467855493856, + "loss": 0.2148, + "step": 4037 + }, + { + "epoch": 0.035051779064417844, + "grad_norm": 0.5625, + "learning_rate": 0.0019954648719100816, + "loss": 0.1582, + "step": 4038 + }, + { + "epoch": 0.03506045954462201, + "grad_norm": 0.177734375, + "learning_rate": 0.0019954618873470405, + "loss": 0.1689, + "step": 4039 + }, + { + "epoch": 0.035069140024826174, + "grad_norm": 0.365234375, + "learning_rate": 0.0019954589018047354, + "loss": 0.1602, + "step": 4040 + }, + { + "epoch": 0.03507782050503034, + "grad_norm": 0.5390625, + "learning_rate": 0.0019954559152831706, + "loss": 0.1895, + "step": 4041 + }, + { + "epoch": 0.035086500985234505, + "grad_norm": 0.5703125, + "learning_rate": 0.0019954529277823475, + "loss": 0.2148, + "step": 4042 + }, + { + "epoch": 0.03509518146543867, + "grad_norm": 0.400390625, + "learning_rate": 0.001995449939302271, + "loss": 0.2021, + "step": 4043 + }, + { + "epoch": 0.035103861945642835, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019954469498429444, + "loss": 0.1719, + "step": 4044 + }, + { + "epoch": 0.035112542425847, + "grad_norm": 0.375, + "learning_rate": 0.00199544395940437, + "loss": 0.1924, + "step": 4045 + }, + { + "epoch": 0.035121222906051165, + "grad_norm": 0.515625, + "learning_rate": 0.001995440967986552, + "loss": 0.1895, + "step": 4046 + }, + { + "epoch": 0.03512990338625533, + "grad_norm": 0.11376953125, + "learning_rate": 0.001995437975589493, + "loss": 0.1494, + "step": 4047 + }, + { + "epoch": 0.035138583866459495, + "grad_norm": 0.1513671875, + "learning_rate": 0.001995434982213196, + "loss": 0.1689, + "step": 4048 + }, + { + "epoch": 0.03514726434666366, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019954319878576652, + "loss": 0.1875, + "step": 4049 + }, + { + "epoch": 0.035155944826867826, + "grad_norm": 0.0634765625, + "learning_rate": 0.0019954289925229033, + "loss": 0.1719, + "step": 4050 + }, + { + "epoch": 0.035164625307071984, + "grad_norm": 0.072265625, + "learning_rate": 0.001995425996208914, + "loss": 0.123, + "step": 4051 + }, + { + "epoch": 0.03517330578727615, + "grad_norm": 0.416015625, + "learning_rate": 0.0019954229989157003, + "loss": 0.1348, + "step": 4052 + }, + { + "epoch": 0.035181986267480314, + "grad_norm": 0.24609375, + "learning_rate": 0.001995420000643265, + "loss": 0.2061, + "step": 4053 + }, + { + "epoch": 0.03519066674768448, + "grad_norm": 0.1796875, + "learning_rate": 0.001995417001391612, + "loss": 0.1699, + "step": 4054 + }, + { + "epoch": 0.035199347227888644, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019954140011607447, + "loss": 0.1816, + "step": 4055 + }, + { + "epoch": 0.03520802770809281, + "grad_norm": 0.275390625, + "learning_rate": 0.0019954109999506657, + "loss": 0.1934, + "step": 4056 + }, + { + "epoch": 0.035216708188296975, + "grad_norm": 0.166015625, + "learning_rate": 0.001995407997761379, + "loss": 0.1777, + "step": 4057 + }, + { + "epoch": 0.03522538866850114, + "grad_norm": 0.376953125, + "learning_rate": 0.0019954049945928878, + "loss": 0.2109, + "step": 4058 + }, + { + "epoch": 0.035234069148705305, + "grad_norm": 0.22265625, + "learning_rate": 0.0019954019904451946, + "loss": 0.2207, + "step": 4059 + }, + { + "epoch": 0.03524274962890947, + "grad_norm": 0.83984375, + "learning_rate": 0.0019953989853183033, + "loss": 0.1689, + "step": 4060 + }, + { + "epoch": 0.035251430109113635, + "grad_norm": 0.1630859375, + "learning_rate": 0.001995395979212217, + "loss": 0.208, + "step": 4061 + }, + { + "epoch": 0.0352601105893178, + "grad_norm": 0.26953125, + "learning_rate": 0.0019953929721269396, + "loss": 0.1641, + "step": 4062 + }, + { + "epoch": 0.035268791069521965, + "grad_norm": 0.58203125, + "learning_rate": 0.0019953899640624737, + "loss": 0.1602, + "step": 4063 + }, + { + "epoch": 0.03527747154972613, + "grad_norm": 0.365234375, + "learning_rate": 0.0019953869550188227, + "loss": 0.1836, + "step": 4064 + }, + { + "epoch": 0.035286152029930296, + "grad_norm": 0.333984375, + "learning_rate": 0.00199538394499599, + "loss": 0.1562, + "step": 4065 + }, + { + "epoch": 0.03529483251013446, + "grad_norm": 0.54296875, + "learning_rate": 0.001995380933993979, + "loss": 0.6406, + "step": 4066 + }, + { + "epoch": 0.035303512990338626, + "grad_norm": 0.55859375, + "learning_rate": 0.0019953779220127925, + "loss": 0.2227, + "step": 4067 + }, + { + "epoch": 0.03531219347054279, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019953749090524347, + "loss": 0.1816, + "step": 4068 + }, + { + "epoch": 0.035320873950746956, + "grad_norm": 0.427734375, + "learning_rate": 0.001995371895112908, + "loss": 0.1719, + "step": 4069 + }, + { + "epoch": 0.03532955443095112, + "grad_norm": 0.1103515625, + "learning_rate": 0.001995368880194216, + "loss": 0.2061, + "step": 4070 + }, + { + "epoch": 0.035338234911155286, + "grad_norm": 0.7578125, + "learning_rate": 0.001995365864296362, + "loss": 0.1855, + "step": 4071 + }, + { + "epoch": 0.03534691539135945, + "grad_norm": 0.10693359375, + "learning_rate": 0.00199536284741935, + "loss": 0.166, + "step": 4072 + }, + { + "epoch": 0.035355595871563616, + "grad_norm": 0.15234375, + "learning_rate": 0.001995359829563182, + "loss": 0.1719, + "step": 4073 + }, + { + "epoch": 0.03536427635176778, + "grad_norm": 0.34765625, + "learning_rate": 0.0019953568107278622, + "loss": 0.1504, + "step": 4074 + }, + { + "epoch": 0.03537295683197195, + "grad_norm": 0.26171875, + "learning_rate": 0.0019953537909133934, + "loss": 0.1426, + "step": 4075 + }, + { + "epoch": 0.03538163731217611, + "grad_norm": 1.0234375, + "learning_rate": 0.0019953507701197795, + "loss": 0.2988, + "step": 4076 + }, + { + "epoch": 0.03539031779238028, + "grad_norm": 0.154296875, + "learning_rate": 0.001995347748347023, + "loss": 0.1846, + "step": 4077 + }, + { + "epoch": 0.03539899827258444, + "grad_norm": 0.1259765625, + "learning_rate": 0.001995344725595128, + "loss": 0.1758, + "step": 4078 + }, + { + "epoch": 0.03540767875278861, + "grad_norm": 0.357421875, + "learning_rate": 0.0019953417018640975, + "loss": 0.1846, + "step": 4079 + }, + { + "epoch": 0.03541635923299277, + "grad_norm": 0.275390625, + "learning_rate": 0.001995338677153934, + "loss": 0.2314, + "step": 4080 + }, + { + "epoch": 0.03542503971319694, + "grad_norm": 0.330078125, + "learning_rate": 0.0019953356514646426, + "loss": 0.1514, + "step": 4081 + }, + { + "epoch": 0.035433720193401096, + "grad_norm": 0.64453125, + "learning_rate": 0.0019953326247962254, + "loss": 0.1465, + "step": 4082 + }, + { + "epoch": 0.03544240067360526, + "grad_norm": 0.302734375, + "learning_rate": 0.0019953295971486856, + "loss": 0.168, + "step": 4083 + }, + { + "epoch": 0.035451081153809426, + "grad_norm": 0.373046875, + "learning_rate": 0.0019953265685220267, + "loss": 0.1553, + "step": 4084 + }, + { + "epoch": 0.03545976163401359, + "grad_norm": 0.3046875, + "learning_rate": 0.001995323538916252, + "loss": 0.168, + "step": 4085 + }, + { + "epoch": 0.035468442114217756, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019953205083313654, + "loss": 0.1826, + "step": 4086 + }, + { + "epoch": 0.03547712259442192, + "grad_norm": 0.1953125, + "learning_rate": 0.0019953174767673694, + "loss": 0.1777, + "step": 4087 + }, + { + "epoch": 0.035485803074626086, + "grad_norm": 0.302734375, + "learning_rate": 0.0019953144442242682, + "loss": 0.1855, + "step": 4088 + }, + { + "epoch": 0.03549448355483025, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019953114107020644, + "loss": 0.1797, + "step": 4089 + }, + { + "epoch": 0.03550316403503442, + "grad_norm": 0.1279296875, + "learning_rate": 0.001995308376200761, + "loss": 0.1787, + "step": 4090 + }, + { + "epoch": 0.03551184451523858, + "grad_norm": 0.16015625, + "learning_rate": 0.0019953053407203622, + "loss": 0.1836, + "step": 4091 + }, + { + "epoch": 0.03552052499544275, + "grad_norm": 0.275390625, + "learning_rate": 0.001995302304260871, + "loss": 0.2676, + "step": 4092 + }, + { + "epoch": 0.03552920547564691, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019952992668222902, + "loss": 0.1797, + "step": 4093 + }, + { + "epoch": 0.03553788595585108, + "grad_norm": 0.2470703125, + "learning_rate": 0.001995296228404624, + "loss": 0.1504, + "step": 4094 + }, + { + "epoch": 0.03554656643605524, + "grad_norm": 0.45703125, + "learning_rate": 0.0019952931890078753, + "loss": 0.1582, + "step": 4095 + }, + { + "epoch": 0.03555524691625941, + "grad_norm": 0.494140625, + "learning_rate": 0.001995290148632047, + "loss": 0.2324, + "step": 4096 + }, + { + "epoch": 0.03556392739646357, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019952871072771434, + "loss": 0.1807, + "step": 4097 + }, + { + "epoch": 0.03557260787666774, + "grad_norm": 0.12353515625, + "learning_rate": 0.001995284064943167, + "loss": 0.2344, + "step": 4098 + }, + { + "epoch": 0.0355812883568719, + "grad_norm": 0.123046875, + "learning_rate": 0.0019952810216301215, + "loss": 0.25, + "step": 4099 + }, + { + "epoch": 0.03558996883707607, + "grad_norm": 0.39453125, + "learning_rate": 0.00199527797733801, + "loss": 0.249, + "step": 4100 + }, + { + "epoch": 0.03559864931728023, + "grad_norm": 0.1435546875, + "learning_rate": 0.001995274932066836, + "loss": 0.1865, + "step": 4101 + }, + { + "epoch": 0.0356073297974844, + "grad_norm": 0.13671875, + "learning_rate": 0.0019952718858166026, + "loss": 0.2012, + "step": 4102 + }, + { + "epoch": 0.03561601027768856, + "grad_norm": 0.2109375, + "learning_rate": 0.001995268838587314, + "loss": 0.1523, + "step": 4103 + }, + { + "epoch": 0.03562469075789273, + "grad_norm": 0.515625, + "learning_rate": 0.001995265790378972, + "loss": 0.1484, + "step": 4104 + }, + { + "epoch": 0.03563337123809689, + "grad_norm": 0.1279296875, + "learning_rate": 0.001995262741191581, + "loss": 0.1514, + "step": 4105 + }, + { + "epoch": 0.03564205171830106, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019952596910251443, + "loss": 0.1641, + "step": 4106 + }, + { + "epoch": 0.035650732198505224, + "grad_norm": 0.8125, + "learning_rate": 0.001995256639879665, + "loss": 0.457, + "step": 4107 + }, + { + "epoch": 0.03565941267870939, + "grad_norm": 0.26171875, + "learning_rate": 0.0019952535877551464, + "loss": 0.1543, + "step": 4108 + }, + { + "epoch": 0.035668093158913554, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019952505346515917, + "loss": 0.1953, + "step": 4109 + }, + { + "epoch": 0.03567677363911772, + "grad_norm": 0.39453125, + "learning_rate": 0.001995247480569005, + "loss": 0.2002, + "step": 4110 + }, + { + "epoch": 0.035685454119321884, + "grad_norm": 0.4140625, + "learning_rate": 0.001995244425507389, + "loss": 0.1826, + "step": 4111 + }, + { + "epoch": 0.03569413459952605, + "grad_norm": 0.28515625, + "learning_rate": 0.001995241369466747, + "loss": 0.2188, + "step": 4112 + }, + { + "epoch": 0.03570281507973021, + "grad_norm": 0.2265625, + "learning_rate": 0.0019952383124470824, + "loss": 0.1992, + "step": 4113 + }, + { + "epoch": 0.03571149555993437, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019952352544483987, + "loss": 0.1807, + "step": 4114 + }, + { + "epoch": 0.03572017604013854, + "grad_norm": 0.5078125, + "learning_rate": 0.0019952321954706995, + "loss": 0.1875, + "step": 4115 + }, + { + "epoch": 0.0357288565203427, + "grad_norm": 0.359375, + "learning_rate": 0.0019952291355139875, + "loss": 0.1836, + "step": 4116 + }, + { + "epoch": 0.03573753700054687, + "grad_norm": 0.43359375, + "learning_rate": 0.0019952260745782664, + "loss": 0.2168, + "step": 4117 + }, + { + "epoch": 0.03574621748075103, + "grad_norm": 0.169921875, + "learning_rate": 0.0019952230126635394, + "loss": 0.1602, + "step": 4118 + }, + { + "epoch": 0.0357548979609552, + "grad_norm": 0.21484375, + "learning_rate": 0.0019952199497698104, + "loss": 0.2314, + "step": 4119 + }, + { + "epoch": 0.03576357844115936, + "grad_norm": 0.2734375, + "learning_rate": 0.001995216885897082, + "loss": 0.1738, + "step": 4120 + }, + { + "epoch": 0.03577225892136353, + "grad_norm": 0.4609375, + "learning_rate": 0.0019952138210453576, + "loss": 0.2227, + "step": 4121 + }, + { + "epoch": 0.035780939401567693, + "grad_norm": 0.134765625, + "learning_rate": 0.001995210755214641, + "loss": 0.1719, + "step": 4122 + }, + { + "epoch": 0.03578961988177186, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019952076884049356, + "loss": 0.2148, + "step": 4123 + }, + { + "epoch": 0.035798300361976024, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019952046206162446, + "loss": 0.1953, + "step": 4124 + }, + { + "epoch": 0.03580698084218019, + "grad_norm": 0.1455078125, + "learning_rate": 0.001995201551848571, + "loss": 0.1699, + "step": 4125 + }, + { + "epoch": 0.035815661322384354, + "grad_norm": 0.12890625, + "learning_rate": 0.0019951984821019184, + "loss": 0.1758, + "step": 4126 + }, + { + "epoch": 0.03582434180258852, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019951954113762906, + "loss": 0.1445, + "step": 4127 + }, + { + "epoch": 0.035833022282792684, + "grad_norm": 0.62109375, + "learning_rate": 0.00199519233967169, + "loss": 0.1699, + "step": 4128 + }, + { + "epoch": 0.03584170276299685, + "grad_norm": 0.19921875, + "learning_rate": 0.0019951892669881205, + "loss": 0.1777, + "step": 4129 + }, + { + "epoch": 0.035850383243201014, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019951861933255855, + "loss": 0.1709, + "step": 4130 + }, + { + "epoch": 0.03585906372340518, + "grad_norm": 0.443359375, + "learning_rate": 0.001995183118684089, + "loss": 0.2188, + "step": 4131 + }, + { + "epoch": 0.035867744203609345, + "grad_norm": 0.2041015625, + "learning_rate": 0.001995180043063633, + "loss": 0.1582, + "step": 4132 + }, + { + "epoch": 0.03587642468381351, + "grad_norm": 0.1201171875, + "learning_rate": 0.001995176966464222, + "loss": 0.2578, + "step": 4133 + }, + { + "epoch": 0.035885105164017675, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019951738888858584, + "loss": 0.1953, + "step": 4134 + }, + { + "epoch": 0.03589378564422184, + "grad_norm": 0.376953125, + "learning_rate": 0.0019951708103285466, + "loss": 0.1973, + "step": 4135 + }, + { + "epoch": 0.035902466124426005, + "grad_norm": 0.0712890625, + "learning_rate": 0.001995167730792289, + "loss": 0.1631, + "step": 4136 + }, + { + "epoch": 0.03591114660463017, + "grad_norm": 0.37890625, + "learning_rate": 0.0019951646502770898, + "loss": 0.1953, + "step": 4137 + }, + { + "epoch": 0.035919827084834335, + "grad_norm": 0.197265625, + "learning_rate": 0.001995161568782952, + "loss": 0.1748, + "step": 4138 + }, + { + "epoch": 0.0359285075650385, + "grad_norm": 0.138671875, + "learning_rate": 0.0019951584863098786, + "loss": 0.1826, + "step": 4139 + }, + { + "epoch": 0.035937188045242666, + "grad_norm": 0.283203125, + "learning_rate": 0.001995155402857874, + "loss": 0.1338, + "step": 4140 + }, + { + "epoch": 0.03594586852544683, + "grad_norm": 0.310546875, + "learning_rate": 0.00199515231842694, + "loss": 0.1504, + "step": 4141 + }, + { + "epoch": 0.035954549005650996, + "grad_norm": 0.1171875, + "learning_rate": 0.0019951492330170816, + "loss": 0.1641, + "step": 4142 + }, + { + "epoch": 0.035963229485855154, + "grad_norm": 0.080078125, + "learning_rate": 0.001995146146628301, + "loss": 0.1699, + "step": 4143 + }, + { + "epoch": 0.03597190996605932, + "grad_norm": 0.7578125, + "learning_rate": 0.001995143059260602, + "loss": 0.1855, + "step": 4144 + }, + { + "epoch": 0.035980590446263484, + "grad_norm": 0.1416015625, + "learning_rate": 0.001995139970913988, + "loss": 0.1592, + "step": 4145 + }, + { + "epoch": 0.03598927092646765, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019951368815884626, + "loss": 0.1328, + "step": 4146 + }, + { + "epoch": 0.035997951406671815, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019951337912840292, + "loss": 0.1699, + "step": 4147 + }, + { + "epoch": 0.03600663188687598, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019951307000006906, + "loss": 0.1367, + "step": 4148 + }, + { + "epoch": 0.036015312367080145, + "grad_norm": 0.099609375, + "learning_rate": 0.0019951276077384505, + "loss": 0.1699, + "step": 4149 + }, + { + "epoch": 0.03602399284728431, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019951245144973124, + "loss": 0.1621, + "step": 4150 + }, + { + "epoch": 0.036032673327488475, + "grad_norm": 0.287109375, + "learning_rate": 0.0019951214202772794, + "loss": 0.1689, + "step": 4151 + }, + { + "epoch": 0.03604135380769264, + "grad_norm": 0.55859375, + "learning_rate": 0.0019951183250783554, + "loss": 0.1807, + "step": 4152 + }, + { + "epoch": 0.036050034287896805, + "grad_norm": 0.2080078125, + "learning_rate": 0.001995115228900543, + "loss": 0.1904, + "step": 4153 + }, + { + "epoch": 0.03605871476810097, + "grad_norm": 0.2138671875, + "learning_rate": 0.001995112131743846, + "loss": 0.1064, + "step": 4154 + }, + { + "epoch": 0.036067395248305135, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019951090336082686, + "loss": 0.2031, + "step": 4155 + }, + { + "epoch": 0.0360760757285093, + "grad_norm": 0.09912109375, + "learning_rate": 0.001995105934493813, + "loss": 0.2012, + "step": 4156 + }, + { + "epoch": 0.036084756208713466, + "grad_norm": 0.302734375, + "learning_rate": 0.0019951028344004826, + "loss": 0.2188, + "step": 4157 + }, + { + "epoch": 0.03609343668891763, + "grad_norm": 0.421875, + "learning_rate": 0.0019950997333282815, + "loss": 0.2031, + "step": 4158 + }, + { + "epoch": 0.036102117169121796, + "grad_norm": 0.173828125, + "learning_rate": 0.001995096631277213, + "loss": 0.1582, + "step": 4159 + }, + { + "epoch": 0.03611079764932596, + "grad_norm": 0.1435546875, + "learning_rate": 0.00199509352824728, + "loss": 0.2031, + "step": 4160 + }, + { + "epoch": 0.036119478129530126, + "grad_norm": 0.53125, + "learning_rate": 0.001995090424238486, + "loss": 0.5156, + "step": 4161 + }, + { + "epoch": 0.03612815860973429, + "grad_norm": 0.53125, + "learning_rate": 0.001995087319250835, + "loss": 0.1797, + "step": 4162 + }, + { + "epoch": 0.036136839089938456, + "grad_norm": 0.302734375, + "learning_rate": 0.0019950842132843297, + "loss": 0.1797, + "step": 4163 + }, + { + "epoch": 0.03614551957014262, + "grad_norm": 0.40234375, + "learning_rate": 0.0019950811063389738, + "loss": 0.1582, + "step": 4164 + }, + { + "epoch": 0.03615420005034679, + "grad_norm": 0.447265625, + "learning_rate": 0.0019950779984147706, + "loss": 0.1709, + "step": 4165 + }, + { + "epoch": 0.03616288053055095, + "grad_norm": 0.107421875, + "learning_rate": 0.0019950748895117237, + "loss": 0.1797, + "step": 4166 + }, + { + "epoch": 0.03617156101075512, + "grad_norm": 0.6796875, + "learning_rate": 0.001995071779629836, + "loss": 0.1973, + "step": 4167 + }, + { + "epoch": 0.03618024149095928, + "grad_norm": 0.53125, + "learning_rate": 0.0019950686687691117, + "loss": 0.2012, + "step": 4168 + }, + { + "epoch": 0.03618892197116345, + "grad_norm": 0.2421875, + "learning_rate": 0.0019950655569295535, + "loss": 0.1377, + "step": 4169 + }, + { + "epoch": 0.03619760245136761, + "grad_norm": 0.287109375, + "learning_rate": 0.001995062444111165, + "loss": 0.1914, + "step": 4170 + }, + { + "epoch": 0.03620628293157178, + "grad_norm": 0.373046875, + "learning_rate": 0.00199505933031395, + "loss": 0.1436, + "step": 4171 + }, + { + "epoch": 0.03621496341177594, + "grad_norm": 0.1787109375, + "learning_rate": 0.001995056215537911, + "loss": 0.1836, + "step": 4172 + }, + { + "epoch": 0.03622364389198011, + "grad_norm": 0.162109375, + "learning_rate": 0.0019950530997830526, + "loss": 0.126, + "step": 4173 + }, + { + "epoch": 0.036232324372184266, + "grad_norm": 0.1689453125, + "learning_rate": 0.001995049983049377, + "loss": 0.1582, + "step": 4174 + }, + { + "epoch": 0.03624100485238843, + "grad_norm": 0.1435546875, + "learning_rate": 0.001995046865336889, + "loss": 0.1328, + "step": 4175 + }, + { + "epoch": 0.036249685332592596, + "grad_norm": 0.2265625, + "learning_rate": 0.0019950437466455906, + "loss": 0.1445, + "step": 4176 + }, + { + "epoch": 0.03625836581279676, + "grad_norm": 0.1015625, + "learning_rate": 0.001995040626975486, + "loss": 0.21, + "step": 4177 + }, + { + "epoch": 0.036267046293000926, + "grad_norm": 0.12890625, + "learning_rate": 0.0019950375063265777, + "loss": 0.2031, + "step": 4178 + }, + { + "epoch": 0.03627572677320509, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019950343846988705, + "loss": 0.1787, + "step": 4179 + }, + { + "epoch": 0.03628440725340926, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019950312620923673, + "loss": 0.1514, + "step": 4180 + }, + { + "epoch": 0.03629308773361342, + "grad_norm": 0.2578125, + "learning_rate": 0.0019950281385070706, + "loss": 0.1484, + "step": 4181 + }, + { + "epoch": 0.03630176821381759, + "grad_norm": 0.29296875, + "learning_rate": 0.0019950250139429853, + "loss": 0.209, + "step": 4182 + }, + { + "epoch": 0.03631044869402175, + "grad_norm": 0.57421875, + "learning_rate": 0.001995021888400114, + "loss": 0.2188, + "step": 4183 + }, + { + "epoch": 0.03631912917422592, + "grad_norm": 0.0859375, + "learning_rate": 0.00199501876187846, + "loss": 0.1748, + "step": 4184 + }, + { + "epoch": 0.03632780965443008, + "grad_norm": 0.28515625, + "learning_rate": 0.001995015634378027, + "loss": 0.1934, + "step": 4185 + }, + { + "epoch": 0.03633649013463425, + "grad_norm": 0.130859375, + "learning_rate": 0.0019950125058988185, + "loss": 0.1787, + "step": 4186 + }, + { + "epoch": 0.03634517061483841, + "grad_norm": 0.322265625, + "learning_rate": 0.001995009376440838, + "loss": 0.2461, + "step": 4187 + }, + { + "epoch": 0.03635385109504258, + "grad_norm": 0.1005859375, + "learning_rate": 0.001995006246004088, + "loss": 0.1709, + "step": 4188 + }, + { + "epoch": 0.03636253157524674, + "grad_norm": 0.271484375, + "learning_rate": 0.0019950031145885734, + "loss": 0.1895, + "step": 4189 + }, + { + "epoch": 0.03637121205545091, + "grad_norm": 0.271484375, + "learning_rate": 0.001994999982194296, + "loss": 0.1865, + "step": 4190 + }, + { + "epoch": 0.03637989253565507, + "grad_norm": 0.55078125, + "learning_rate": 0.0019949968488212604, + "loss": 0.1758, + "step": 4191 + }, + { + "epoch": 0.03638857301585924, + "grad_norm": 0.32421875, + "learning_rate": 0.0019949937144694697, + "loss": 0.1777, + "step": 4192 + }, + { + "epoch": 0.0363972534960634, + "grad_norm": 0.13671875, + "learning_rate": 0.0019949905791389277, + "loss": 0.1758, + "step": 4193 + }, + { + "epoch": 0.03640593397626757, + "grad_norm": 0.2080078125, + "learning_rate": 0.001994987442829637, + "loss": 0.1338, + "step": 4194 + }, + { + "epoch": 0.03641461445647173, + "grad_norm": 0.1484375, + "learning_rate": 0.0019949843055416016, + "loss": 0.2188, + "step": 4195 + }, + { + "epoch": 0.0364232949366759, + "grad_norm": 0.2109375, + "learning_rate": 0.001994981167274825, + "loss": 0.123, + "step": 4196 + }, + { + "epoch": 0.036431975416880064, + "grad_norm": 0.54296875, + "learning_rate": 0.00199497802802931, + "loss": 0.1826, + "step": 4197 + }, + { + "epoch": 0.03644065589708423, + "grad_norm": 0.126953125, + "learning_rate": 0.001994974887805061, + "loss": 0.1572, + "step": 4198 + }, + { + "epoch": 0.036449336377288394, + "grad_norm": 0.58203125, + "learning_rate": 0.0019949717466020804, + "loss": 0.1973, + "step": 4199 + }, + { + "epoch": 0.03645801685749256, + "grad_norm": 0.3671875, + "learning_rate": 0.0019949686044203724, + "loss": 0.166, + "step": 4200 + }, + { + "epoch": 0.036466697337696724, + "grad_norm": 0.47265625, + "learning_rate": 0.00199496546125994, + "loss": 0.1934, + "step": 4201 + }, + { + "epoch": 0.03647537781790089, + "grad_norm": 0.25390625, + "learning_rate": 0.001994962317120787, + "loss": 0.1641, + "step": 4202 + }, + { + "epoch": 0.036484058298105054, + "grad_norm": 0.66796875, + "learning_rate": 0.001994959172002917, + "loss": 0.2314, + "step": 4203 + }, + { + "epoch": 0.03649273877830922, + "grad_norm": 0.17578125, + "learning_rate": 0.0019949560259063326, + "loss": 0.1895, + "step": 4204 + }, + { + "epoch": 0.03650141925851338, + "grad_norm": 0.142578125, + "learning_rate": 0.001994952878831038, + "loss": 0.1455, + "step": 4205 + }, + { + "epoch": 0.03651009973871754, + "grad_norm": 0.498046875, + "learning_rate": 0.0019949497307770363, + "loss": 0.2031, + "step": 4206 + }, + { + "epoch": 0.03651878021892171, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019949465817443307, + "loss": 0.1484, + "step": 4207 + }, + { + "epoch": 0.03652746069912587, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019949434317329253, + "loss": 0.1318, + "step": 4208 + }, + { + "epoch": 0.03653614117933004, + "grad_norm": 0.1591796875, + "learning_rate": 0.001994940280742823, + "loss": 0.1445, + "step": 4209 + }, + { + "epoch": 0.0365448216595342, + "grad_norm": 0.0927734375, + "learning_rate": 0.001994937128774028, + "loss": 0.1494, + "step": 4210 + }, + { + "epoch": 0.03655350213973837, + "grad_norm": 0.1669921875, + "learning_rate": 0.001994933975826543, + "loss": 0.1875, + "step": 4211 + }, + { + "epoch": 0.03656218261994253, + "grad_norm": 0.294921875, + "learning_rate": 0.001994930821900371, + "loss": 0.1572, + "step": 4212 + }, + { + "epoch": 0.0365708631001467, + "grad_norm": 0.462890625, + "learning_rate": 0.0019949276669955166, + "loss": 0.2061, + "step": 4213 + }, + { + "epoch": 0.036579543580350864, + "grad_norm": 0.3984375, + "learning_rate": 0.0019949245111119825, + "loss": 0.1436, + "step": 4214 + }, + { + "epoch": 0.03658822406055503, + "grad_norm": 0.392578125, + "learning_rate": 0.001994921354249773, + "loss": 0.1611, + "step": 4215 + }, + { + "epoch": 0.036596904540759194, + "grad_norm": 0.1455078125, + "learning_rate": 0.00199491819640889, + "loss": 0.1777, + "step": 4216 + }, + { + "epoch": 0.03660558502096336, + "grad_norm": 0.130859375, + "learning_rate": 0.0019949150375893386, + "loss": 0.1943, + "step": 4217 + }, + { + "epoch": 0.036614265501167524, + "grad_norm": 0.3203125, + "learning_rate": 0.0019949118777911217, + "loss": 0.1484, + "step": 4218 + }, + { + "epoch": 0.03662294598137169, + "grad_norm": 0.134765625, + "learning_rate": 0.0019949087170142423, + "loss": 0.1709, + "step": 4219 + }, + { + "epoch": 0.036631626461575854, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019949055552587044, + "loss": 0.1943, + "step": 4220 + }, + { + "epoch": 0.03664030694178002, + "grad_norm": 0.298828125, + "learning_rate": 0.001994902392524511, + "loss": 0.1318, + "step": 4221 + }, + { + "epoch": 0.036648987421984185, + "grad_norm": 0.181640625, + "learning_rate": 0.001994899228811666, + "loss": 0.125, + "step": 4222 + }, + { + "epoch": 0.03665766790218835, + "grad_norm": 0.5703125, + "learning_rate": 0.001994896064120172, + "loss": 0.1953, + "step": 4223 + }, + { + "epoch": 0.036666348382392515, + "grad_norm": 0.16015625, + "learning_rate": 0.001994892898450034, + "loss": 0.1855, + "step": 4224 + }, + { + "epoch": 0.03667502886259668, + "grad_norm": 0.5546875, + "learning_rate": 0.0019948897318012543, + "loss": 0.1816, + "step": 4225 + }, + { + "epoch": 0.036683709342800845, + "grad_norm": 0.09326171875, + "learning_rate": 0.001994886564173836, + "loss": 0.248, + "step": 4226 + }, + { + "epoch": 0.03669238982300501, + "grad_norm": 0.216796875, + "learning_rate": 0.001994883395567784, + "loss": 0.1543, + "step": 4227 + }, + { + "epoch": 0.036701070303209175, + "grad_norm": 0.404296875, + "learning_rate": 0.001994880225983101, + "loss": 0.1953, + "step": 4228 + }, + { + "epoch": 0.03670975078341334, + "grad_norm": 0.41015625, + "learning_rate": 0.00199487705541979, + "loss": 0.2285, + "step": 4229 + }, + { + "epoch": 0.036718431263617506, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019948738838778545, + "loss": 0.1206, + "step": 4230 + }, + { + "epoch": 0.03672711174382167, + "grad_norm": 0.12255859375, + "learning_rate": 0.001994870711357299, + "loss": 0.1934, + "step": 4231 + }, + { + "epoch": 0.036735792224025836, + "grad_norm": 0.66796875, + "learning_rate": 0.001994867537858126, + "loss": 0.1777, + "step": 4232 + }, + { + "epoch": 0.03674447270423, + "grad_norm": 0.70703125, + "learning_rate": 0.00199486436338034, + "loss": 0.168, + "step": 4233 + }, + { + "epoch": 0.036753153184434166, + "grad_norm": 0.1474609375, + "learning_rate": 0.001994861187923943, + "loss": 0.1699, + "step": 4234 + }, + { + "epoch": 0.03676183366463833, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019948580114889395, + "loss": 0.1309, + "step": 4235 + }, + { + "epoch": 0.03677051414484249, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019948548340753326, + "loss": 0.2109, + "step": 4236 + }, + { + "epoch": 0.036779194625046654, + "grad_norm": 0.29296875, + "learning_rate": 0.001994851655683126, + "loss": 0.1484, + "step": 4237 + }, + { + "epoch": 0.03678787510525082, + "grad_norm": 0.5, + "learning_rate": 0.0019948484763123234, + "loss": 0.1934, + "step": 4238 + }, + { + "epoch": 0.036796555585454985, + "grad_norm": 0.263671875, + "learning_rate": 0.0019948452959629276, + "loss": 0.1787, + "step": 4239 + }, + { + "epoch": 0.03680523606565915, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019948421146349423, + "loss": 0.2041, + "step": 4240 + }, + { + "epoch": 0.036813916545863315, + "grad_norm": 0.416015625, + "learning_rate": 0.0019948389323283717, + "loss": 0.2334, + "step": 4241 + }, + { + "epoch": 0.03682259702606748, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019948357490432184, + "loss": 0.252, + "step": 4242 + }, + { + "epoch": 0.036831277506271645, + "grad_norm": 0.287109375, + "learning_rate": 0.001994832564779486, + "loss": 0.1885, + "step": 4243 + }, + { + "epoch": 0.03683995798647581, + "grad_norm": 0.13671875, + "learning_rate": 0.001994829379537178, + "loss": 0.1426, + "step": 4244 + }, + { + "epoch": 0.036848638466679975, + "grad_norm": 0.2353515625, + "learning_rate": 0.001994826193316298, + "loss": 0.1328, + "step": 4245 + }, + { + "epoch": 0.03685731894688414, + "grad_norm": 0.365234375, + "learning_rate": 0.0019948230061168503, + "loss": 0.1738, + "step": 4246 + }, + { + "epoch": 0.036865999427088306, + "grad_norm": 0.12890625, + "learning_rate": 0.001994819817938837, + "loss": 0.1504, + "step": 4247 + }, + { + "epoch": 0.03687467990729247, + "grad_norm": 0.326171875, + "learning_rate": 0.0019948166287822623, + "loss": 0.1465, + "step": 4248 + }, + { + "epoch": 0.036883360387496636, + "grad_norm": 0.103515625, + "learning_rate": 0.0019948134386471297, + "loss": 0.1338, + "step": 4249 + }, + { + "epoch": 0.0368920408677008, + "grad_norm": 0.4375, + "learning_rate": 0.001994810247533442, + "loss": 0.2188, + "step": 4250 + }, + { + "epoch": 0.036900721347904966, + "grad_norm": 0.34765625, + "learning_rate": 0.001994807055441204, + "loss": 0.1738, + "step": 4251 + }, + { + "epoch": 0.03690940182810913, + "grad_norm": 0.08740234375, + "learning_rate": 0.001994803862370418, + "loss": 0.167, + "step": 4252 + }, + { + "epoch": 0.036918082308313296, + "grad_norm": 0.240234375, + "learning_rate": 0.001994800668321088, + "loss": 0.1758, + "step": 4253 + }, + { + "epoch": 0.03692676278851746, + "grad_norm": 0.251953125, + "learning_rate": 0.0019947974732932175, + "loss": 0.1494, + "step": 4254 + }, + { + "epoch": 0.03693544326872163, + "grad_norm": 0.220703125, + "learning_rate": 0.00199479427728681, + "loss": 0.2266, + "step": 4255 + }, + { + "epoch": 0.03694412374892579, + "grad_norm": 0.1328125, + "learning_rate": 0.001994791080301869, + "loss": 0.2119, + "step": 4256 + }, + { + "epoch": 0.03695280422912996, + "grad_norm": 0.5390625, + "learning_rate": 0.0019947878823383977, + "loss": 0.1826, + "step": 4257 + }, + { + "epoch": 0.03696148470933412, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019947846833963997, + "loss": 0.1602, + "step": 4258 + }, + { + "epoch": 0.03697016518953829, + "grad_norm": 0.2216796875, + "learning_rate": 0.001994781483475879, + "loss": 0.1133, + "step": 4259 + }, + { + "epoch": 0.03697884566974245, + "grad_norm": 0.4765625, + "learning_rate": 0.0019947782825768387, + "loss": 0.1426, + "step": 4260 + }, + { + "epoch": 0.03698752614994662, + "grad_norm": 0.443359375, + "learning_rate": 0.0019947750806992817, + "loss": 0.1641, + "step": 4261 + }, + { + "epoch": 0.03699620663015078, + "grad_norm": 0.279296875, + "learning_rate": 0.001994771877843213, + "loss": 0.166, + "step": 4262 + }, + { + "epoch": 0.03700488711035495, + "grad_norm": 0.283203125, + "learning_rate": 0.0019947686740086346, + "loss": 0.126, + "step": 4263 + }, + { + "epoch": 0.03701356759055911, + "grad_norm": 0.39453125, + "learning_rate": 0.001994765469195551, + "loss": 0.1611, + "step": 4264 + }, + { + "epoch": 0.03702224807076328, + "grad_norm": 0.5546875, + "learning_rate": 0.0019947622634039646, + "loss": 0.1934, + "step": 4265 + }, + { + "epoch": 0.03703092855096744, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019947590566338803, + "loss": 0.1621, + "step": 4266 + }, + { + "epoch": 0.0370396090311716, + "grad_norm": 0.10009765625, + "learning_rate": 0.001994755848885301, + "loss": 0.1914, + "step": 4267 + }, + { + "epoch": 0.037048289511375766, + "grad_norm": 0.150390625, + "learning_rate": 0.00199475264015823, + "loss": 0.1289, + "step": 4268 + }, + { + "epoch": 0.03705696999157993, + "grad_norm": 0.126953125, + "learning_rate": 0.0019947494304526706, + "loss": 0.1992, + "step": 4269 + }, + { + "epoch": 0.037065650471784096, + "grad_norm": 0.2275390625, + "learning_rate": 0.001994746219768627, + "loss": 0.1914, + "step": 4270 + }, + { + "epoch": 0.03707433095198826, + "grad_norm": 0.142578125, + "learning_rate": 0.001994743008106103, + "loss": 0.2012, + "step": 4271 + }, + { + "epoch": 0.03708301143219243, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019947397954651, + "loss": 0.1758, + "step": 4272 + }, + { + "epoch": 0.03709169191239659, + "grad_norm": 0.11083984375, + "learning_rate": 0.001994736581845624, + "loss": 0.1758, + "step": 4273 + }, + { + "epoch": 0.03710037239260076, + "grad_norm": 0.3828125, + "learning_rate": 0.0019947333672476774, + "loss": 0.1494, + "step": 4274 + }, + { + "epoch": 0.03710905287280492, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019947301516712638, + "loss": 0.1426, + "step": 4275 + }, + { + "epoch": 0.03711773335300909, + "grad_norm": 0.138671875, + "learning_rate": 0.0019947269351163867, + "loss": 0.1436, + "step": 4276 + }, + { + "epoch": 0.03712641383321325, + "grad_norm": 0.3359375, + "learning_rate": 0.0019947237175830497, + "loss": 0.1504, + "step": 4277 + }, + { + "epoch": 0.03713509431341742, + "grad_norm": 0.380859375, + "learning_rate": 0.0019947204990712563, + "loss": 0.208, + "step": 4278 + }, + { + "epoch": 0.03714377479362158, + "grad_norm": 0.3203125, + "learning_rate": 0.0019947172795810103, + "loss": 0.2412, + "step": 4279 + }, + { + "epoch": 0.03715245527382575, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019947140591123147, + "loss": 0.1582, + "step": 4280 + }, + { + "epoch": 0.03716113575402991, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019947108376651727, + "loss": 0.1865, + "step": 4281 + }, + { + "epoch": 0.03716981623423408, + "grad_norm": 0.12060546875, + "learning_rate": 0.001994707615239589, + "loss": 0.1582, + "step": 4282 + }, + { + "epoch": 0.03717849671443824, + "grad_norm": 0.2333984375, + "learning_rate": 0.001994704391835566, + "loss": 0.1318, + "step": 4283 + }, + { + "epoch": 0.03718717719464241, + "grad_norm": 0.330078125, + "learning_rate": 0.0019947011674531084, + "loss": 0.2539, + "step": 4284 + }, + { + "epoch": 0.03719585767484657, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019946979420922186, + "loss": 0.207, + "step": 4285 + }, + { + "epoch": 0.03720453815505074, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019946947157529005, + "loss": 0.1157, + "step": 4286 + }, + { + "epoch": 0.037213218635254904, + "grad_norm": 0.271484375, + "learning_rate": 0.001994691488435158, + "loss": 0.1904, + "step": 4287 + }, + { + "epoch": 0.03722189911545907, + "grad_norm": 0.56640625, + "learning_rate": 0.001994688260138994, + "loss": 0.1934, + "step": 4288 + }, + { + "epoch": 0.037230579595663234, + "grad_norm": 0.162109375, + "learning_rate": 0.001994685030864413, + "loss": 0.1855, + "step": 4289 + }, + { + "epoch": 0.0372392600758674, + "grad_norm": 0.1328125, + "learning_rate": 0.0019946818006114173, + "loss": 0.1572, + "step": 4290 + }, + { + "epoch": 0.037247940556071564, + "grad_norm": 0.181640625, + "learning_rate": 0.001994678569380011, + "loss": 0.1631, + "step": 4291 + }, + { + "epoch": 0.03725662103627573, + "grad_norm": 0.15234375, + "learning_rate": 0.001994675337170198, + "loss": 0.2012, + "step": 4292 + }, + { + "epoch": 0.037265301516479894, + "grad_norm": 0.10546875, + "learning_rate": 0.0019946721039819813, + "loss": 0.1836, + "step": 4293 + }, + { + "epoch": 0.03727398199668406, + "grad_norm": 0.248046875, + "learning_rate": 0.0019946688698153645, + "loss": 0.1523, + "step": 4294 + }, + { + "epoch": 0.037282662476888224, + "grad_norm": 0.25, + "learning_rate": 0.0019946656346703515, + "loss": 0.166, + "step": 4295 + }, + { + "epoch": 0.03729134295709239, + "grad_norm": 0.2236328125, + "learning_rate": 0.001994662398546946, + "loss": 0.2188, + "step": 4296 + }, + { + "epoch": 0.03730002343729655, + "grad_norm": 0.52734375, + "learning_rate": 0.0019946591614451504, + "loss": 0.1621, + "step": 4297 + }, + { + "epoch": 0.03730870391750071, + "grad_norm": 1.125, + "learning_rate": 0.0019946559233649692, + "loss": 0.2129, + "step": 4298 + }, + { + "epoch": 0.03731738439770488, + "grad_norm": 0.486328125, + "learning_rate": 0.0019946526843064058, + "loss": 0.2148, + "step": 4299 + }, + { + "epoch": 0.03732606487790904, + "grad_norm": 0.65625, + "learning_rate": 0.0019946494442694635, + "loss": 0.1885, + "step": 4300 + }, + { + "epoch": 0.03733474535811321, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019946462032541462, + "loss": 0.1992, + "step": 4301 + }, + { + "epoch": 0.03734342583831737, + "grad_norm": 0.1884765625, + "learning_rate": 0.001994642961260457, + "loss": 0.1572, + "step": 4302 + }, + { + "epoch": 0.03735210631852154, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019946397182884, + "loss": 0.2578, + "step": 4303 + }, + { + "epoch": 0.037360786798725704, + "grad_norm": 0.095703125, + "learning_rate": 0.0019946364743379788, + "loss": 0.1387, + "step": 4304 + }, + { + "epoch": 0.03736946727892987, + "grad_norm": 0.51953125, + "learning_rate": 0.0019946332294091956, + "loss": 0.1455, + "step": 4305 + }, + { + "epoch": 0.037378147759134034, + "grad_norm": 0.14453125, + "learning_rate": 0.0019946299835020558, + "loss": 0.1445, + "step": 4306 + }, + { + "epoch": 0.0373868282393382, + "grad_norm": 0.130859375, + "learning_rate": 0.001994626736616562, + "loss": 0.1348, + "step": 4307 + }, + { + "epoch": 0.037395508719542364, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019946234887527176, + "loss": 0.1553, + "step": 4308 + }, + { + "epoch": 0.03740418919974653, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019946202399105262, + "loss": 0.1602, + "step": 4309 + }, + { + "epoch": 0.037412869679950694, + "grad_norm": 0.220703125, + "learning_rate": 0.0019946169900899916, + "loss": 0.1494, + "step": 4310 + }, + { + "epoch": 0.03742155016015486, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019946137392911176, + "loss": 0.1338, + "step": 4311 + }, + { + "epoch": 0.037430230640359025, + "grad_norm": 0.41796875, + "learning_rate": 0.0019946104875139072, + "loss": 0.1602, + "step": 4312 + }, + { + "epoch": 0.03743891112056319, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019946072347583645, + "loss": 0.1719, + "step": 4313 + }, + { + "epoch": 0.037447591600767355, + "grad_norm": 0.490234375, + "learning_rate": 0.0019946039810244927, + "loss": 0.1475, + "step": 4314 + }, + { + "epoch": 0.03745627208097152, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019946007263122955, + "loss": 0.1533, + "step": 4315 + }, + { + "epoch": 0.037464952561175685, + "grad_norm": 0.4921875, + "learning_rate": 0.001994597470621776, + "loss": 0.2031, + "step": 4316 + }, + { + "epoch": 0.03747363304137985, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019945942139529384, + "loss": 0.2305, + "step": 4317 + }, + { + "epoch": 0.037482313521584015, + "grad_norm": 0.107421875, + "learning_rate": 0.001994590956305786, + "loss": 0.1562, + "step": 4318 + }, + { + "epoch": 0.03749099400178818, + "grad_norm": 0.21875, + "learning_rate": 0.0019945876976803223, + "loss": 0.1455, + "step": 4319 + }, + { + "epoch": 0.037499674481992346, + "grad_norm": 0.40625, + "learning_rate": 0.001994584438076551, + "loss": 0.2207, + "step": 4320 + }, + { + "epoch": 0.03750835496219651, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019945811774944753, + "loss": 0.1533, + "step": 4321 + }, + { + "epoch": 0.037517035442400676, + "grad_norm": 0.205078125, + "learning_rate": 0.0019945779159340994, + "loss": 0.1602, + "step": 4322 + }, + { + "epoch": 0.03752571592260484, + "grad_norm": 0.126953125, + "learning_rate": 0.0019945746533954266, + "loss": 0.1533, + "step": 4323 + }, + { + "epoch": 0.037534396402809006, + "grad_norm": 0.2890625, + "learning_rate": 0.0019945713898784603, + "loss": 0.1245, + "step": 4324 + }, + { + "epoch": 0.03754307688301317, + "grad_norm": 0.5, + "learning_rate": 0.001994568125383204, + "loss": 0.1758, + "step": 4325 + }, + { + "epoch": 0.037551757363217336, + "grad_norm": 0.2001953125, + "learning_rate": 0.001994564859909662, + "loss": 0.166, + "step": 4326 + }, + { + "epoch": 0.0375604378434215, + "grad_norm": 0.1533203125, + "learning_rate": 0.001994561593457837, + "loss": 0.2109, + "step": 4327 + }, + { + "epoch": 0.03756911832362566, + "grad_norm": 0.39453125, + "learning_rate": 0.0019945583260277326, + "loss": 0.1621, + "step": 4328 + }, + { + "epoch": 0.037577798803829825, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019945550576193527, + "loss": 0.123, + "step": 4329 + }, + { + "epoch": 0.03758647928403399, + "grad_norm": 0.08056640625, + "learning_rate": 0.001994551788232701, + "loss": 0.1826, + "step": 4330 + }, + { + "epoch": 0.037595159764238155, + "grad_norm": 0.29296875, + "learning_rate": 0.001994548517867781, + "loss": 0.1797, + "step": 4331 + }, + { + "epoch": 0.03760384024444232, + "grad_norm": 0.2001953125, + "learning_rate": 0.001994545246524596, + "loss": 0.1904, + "step": 4332 + }, + { + "epoch": 0.037612520724646485, + "grad_norm": 0.1875, + "learning_rate": 0.00199454197420315, + "loss": 0.168, + "step": 4333 + }, + { + "epoch": 0.03762120120485065, + "grad_norm": 0.28515625, + "learning_rate": 0.001994538700903446, + "loss": 0.1836, + "step": 4334 + }, + { + "epoch": 0.037629881685054815, + "grad_norm": 0.396484375, + "learning_rate": 0.001994535426625488, + "loss": 0.1689, + "step": 4335 + }, + { + "epoch": 0.03763856216525898, + "grad_norm": 0.189453125, + "learning_rate": 0.0019945321513692798, + "loss": 0.1465, + "step": 4336 + }, + { + "epoch": 0.037647242645463146, + "grad_norm": 0.1396484375, + "learning_rate": 0.001994528875134824, + "loss": 0.1758, + "step": 4337 + }, + { + "epoch": 0.03765592312566731, + "grad_norm": 0.97265625, + "learning_rate": 0.0019945255979221255, + "loss": 0.3945, + "step": 4338 + }, + { + "epoch": 0.037664603605871476, + "grad_norm": 0.5390625, + "learning_rate": 0.0019945223197311867, + "loss": 0.1953, + "step": 4339 + }, + { + "epoch": 0.03767328408607564, + "grad_norm": 1.03125, + "learning_rate": 0.0019945190405620122, + "loss": 0.168, + "step": 4340 + }, + { + "epoch": 0.037681964566279806, + "grad_norm": 0.3046875, + "learning_rate": 0.001994515760414605, + "loss": 0.1914, + "step": 4341 + }, + { + "epoch": 0.03769064504648397, + "grad_norm": 0.3515625, + "learning_rate": 0.001994512479288969, + "loss": 0.1865, + "step": 4342 + }, + { + "epoch": 0.037699325526688136, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019945091971851074, + "loss": 0.1602, + "step": 4343 + }, + { + "epoch": 0.0377080060068923, + "grad_norm": 0.115234375, + "learning_rate": 0.001994505914103024, + "loss": 0.1475, + "step": 4344 + }, + { + "epoch": 0.03771668648709647, + "grad_norm": 0.375, + "learning_rate": 0.0019945026300427224, + "loss": 0.124, + "step": 4345 + }, + { + "epoch": 0.03772536696730063, + "grad_norm": 0.2392578125, + "learning_rate": 0.001994499345004206, + "loss": 0.2217, + "step": 4346 + }, + { + "epoch": 0.0377340474475048, + "grad_norm": 0.337890625, + "learning_rate": 0.0019944960589874786, + "loss": 0.1865, + "step": 4347 + }, + { + "epoch": 0.03774272792770896, + "grad_norm": 0.458984375, + "learning_rate": 0.0019944927719925437, + "loss": 0.1699, + "step": 4348 + }, + { + "epoch": 0.03775140840791313, + "grad_norm": 0.1630859375, + "learning_rate": 0.001994489484019405, + "loss": 0.1582, + "step": 4349 + }, + { + "epoch": 0.03776008888811729, + "grad_norm": 0.232421875, + "learning_rate": 0.0019944861950680665, + "loss": 0.1523, + "step": 4350 + }, + { + "epoch": 0.03776876936832146, + "grad_norm": 0.80078125, + "learning_rate": 0.0019944829051385306, + "loss": 0.1758, + "step": 4351 + }, + { + "epoch": 0.03777744984852562, + "grad_norm": 0.9921875, + "learning_rate": 0.001994479614230802, + "loss": 0.2305, + "step": 4352 + }, + { + "epoch": 0.03778613032872979, + "grad_norm": 0.470703125, + "learning_rate": 0.001994476322344884, + "loss": 0.2266, + "step": 4353 + }, + { + "epoch": 0.03779481080893395, + "grad_norm": 0.1396484375, + "learning_rate": 0.00199447302948078, + "loss": 0.1201, + "step": 4354 + }, + { + "epoch": 0.03780349128913812, + "grad_norm": 0.10400390625, + "learning_rate": 0.001994469735638494, + "loss": 0.2129, + "step": 4355 + }, + { + "epoch": 0.03781217176934228, + "grad_norm": 0.0830078125, + "learning_rate": 0.001994466440818029, + "loss": 0.1807, + "step": 4356 + }, + { + "epoch": 0.03782085224954645, + "grad_norm": 0.11474609375, + "learning_rate": 0.001994463145019389, + "loss": 0.1582, + "step": 4357 + }, + { + "epoch": 0.03782953272975061, + "grad_norm": 0.66796875, + "learning_rate": 0.001994459848242578, + "loss": 0.1924, + "step": 4358 + }, + { + "epoch": 0.03783821320995477, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019944565504875986, + "loss": 0.2041, + "step": 4359 + }, + { + "epoch": 0.037846893690158936, + "grad_norm": 0.46484375, + "learning_rate": 0.0019944532517544547, + "loss": 0.1719, + "step": 4360 + }, + { + "epoch": 0.0378555741703631, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019944499520431506, + "loss": 0.1514, + "step": 4361 + }, + { + "epoch": 0.03786425465056727, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019944466513536897, + "loss": 0.1943, + "step": 4362 + }, + { + "epoch": 0.03787293513077143, + "grad_norm": 0.671875, + "learning_rate": 0.001994443349686075, + "loss": 0.1611, + "step": 4363 + }, + { + "epoch": 0.0378816156109756, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019944400470403106, + "loss": 0.166, + "step": 4364 + }, + { + "epoch": 0.03789029609117976, + "grad_norm": 0.98828125, + "learning_rate": 0.0019944367434164, + "loss": 0.1445, + "step": 4365 + }, + { + "epoch": 0.03789897657138393, + "grad_norm": 0.79296875, + "learning_rate": 0.0019944334388143467, + "loss": 0.1387, + "step": 4366 + }, + { + "epoch": 0.03790765705158809, + "grad_norm": 0.7890625, + "learning_rate": 0.0019944301332341548, + "loss": 0.1768, + "step": 4367 + }, + { + "epoch": 0.03791633753179226, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019944268266758273, + "loss": 0.1855, + "step": 4368 + }, + { + "epoch": 0.03792501801199642, + "grad_norm": 0.640625, + "learning_rate": 0.001994423519139368, + "loss": 0.1904, + "step": 4369 + }, + { + "epoch": 0.03793369849220059, + "grad_norm": 0.318359375, + "learning_rate": 0.0019944202106247804, + "loss": 0.165, + "step": 4370 + }, + { + "epoch": 0.03794237897240475, + "grad_norm": 0.12451171875, + "learning_rate": 0.001994416901132069, + "loss": 0.1582, + "step": 4371 + }, + { + "epoch": 0.03795105945260892, + "grad_norm": 1.2265625, + "learning_rate": 0.001994413590661236, + "loss": 0.1943, + "step": 4372 + }, + { + "epoch": 0.03795973993281308, + "grad_norm": 0.61328125, + "learning_rate": 0.001994410279212286, + "loss": 0.1719, + "step": 4373 + }, + { + "epoch": 0.03796842041301725, + "grad_norm": 0.8671875, + "learning_rate": 0.0019944069667852224, + "loss": 0.1602, + "step": 4374 + }, + { + "epoch": 0.03797710089322141, + "grad_norm": 0.9921875, + "learning_rate": 0.001994403653380049, + "loss": 0.1953, + "step": 4375 + }, + { + "epoch": 0.03798578137342558, + "grad_norm": 0.5078125, + "learning_rate": 0.001994400338996769, + "loss": 0.1621, + "step": 4376 + }, + { + "epoch": 0.037994461853629743, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019943970236353855, + "loss": 0.2031, + "step": 4377 + }, + { + "epoch": 0.03800314233383391, + "grad_norm": 0.96484375, + "learning_rate": 0.001994393707295904, + "loss": 0.1514, + "step": 4378 + }, + { + "epoch": 0.038011822814038074, + "grad_norm": 0.66015625, + "learning_rate": 0.001994390389978326, + "loss": 0.1631, + "step": 4379 + }, + { + "epoch": 0.03802050329424224, + "grad_norm": 0.7734375, + "learning_rate": 0.001994387071682657, + "loss": 0.1641, + "step": 4380 + }, + { + "epoch": 0.038029183774446404, + "grad_norm": 0.263671875, + "learning_rate": 0.0019943837524088993, + "loss": 0.1035, + "step": 4381 + }, + { + "epoch": 0.03803786425465057, + "grad_norm": 0.4453125, + "learning_rate": 0.0019943804321570567, + "loss": 0.1709, + "step": 4382 + }, + { + "epoch": 0.038046544734854734, + "grad_norm": 1.015625, + "learning_rate": 0.0019943771109271337, + "loss": 0.1934, + "step": 4383 + }, + { + "epoch": 0.0380552252150589, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019943737887191328, + "loss": 0.1631, + "step": 4384 + }, + { + "epoch": 0.038063905695263064, + "grad_norm": 0.478515625, + "learning_rate": 0.0019943704655330584, + "loss": 0.1709, + "step": 4385 + }, + { + "epoch": 0.03807258617546723, + "grad_norm": 0.384765625, + "learning_rate": 0.0019943671413689143, + "loss": 0.1738, + "step": 4386 + }, + { + "epoch": 0.038081266655671395, + "grad_norm": 0.2490234375, + "learning_rate": 0.001994363816226703, + "loss": 0.1836, + "step": 4387 + }, + { + "epoch": 0.03808994713587556, + "grad_norm": 0.78515625, + "learning_rate": 0.0019943604901064295, + "loss": 0.2129, + "step": 4388 + }, + { + "epoch": 0.038098627616079725, + "grad_norm": 0.69921875, + "learning_rate": 0.001994357163008096, + "loss": 0.1738, + "step": 4389 + }, + { + "epoch": 0.03810730809628388, + "grad_norm": 0.3359375, + "learning_rate": 0.0019943538349317075, + "loss": 0.1895, + "step": 4390 + }, + { + "epoch": 0.03811598857648805, + "grad_norm": 0.58203125, + "learning_rate": 0.001994350505877267, + "loss": 0.1445, + "step": 4391 + }, + { + "epoch": 0.03812466905669221, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019943471758447785, + "loss": 0.1758, + "step": 4392 + }, + { + "epoch": 0.03813334953689638, + "grad_norm": 0.267578125, + "learning_rate": 0.001994343844834245, + "loss": 0.166, + "step": 4393 + }, + { + "epoch": 0.038142030017100544, + "grad_norm": 0.142578125, + "learning_rate": 0.0019943405128456707, + "loss": 0.2129, + "step": 4394 + }, + { + "epoch": 0.03815071049730471, + "grad_norm": 0.392578125, + "learning_rate": 0.0019943371798790592, + "loss": 0.2031, + "step": 4395 + }, + { + "epoch": 0.038159390977508874, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019943338459344133, + "loss": 0.1465, + "step": 4396 + }, + { + "epoch": 0.03816807145771304, + "grad_norm": 0.87109375, + "learning_rate": 0.001994330511011738, + "loss": 0.1631, + "step": 4397 + }, + { + "epoch": 0.038176751937917204, + "grad_norm": 0.53515625, + "learning_rate": 0.001994327175111036, + "loss": 0.1504, + "step": 4398 + }, + { + "epoch": 0.03818543241812137, + "grad_norm": 0.5078125, + "learning_rate": 0.001994323838232312, + "loss": 0.1318, + "step": 4399 + }, + { + "epoch": 0.038194112898325534, + "grad_norm": 0.65234375, + "learning_rate": 0.0019943205003755677, + "loss": 0.2207, + "step": 4400 + }, + { + "epoch": 0.0382027933785297, + "grad_norm": 0.73046875, + "learning_rate": 0.0019943171615408083, + "loss": 0.1748, + "step": 4401 + }, + { + "epoch": 0.038211473858733865, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019943138217280378, + "loss": 0.1279, + "step": 4402 + }, + { + "epoch": 0.03822015433893803, + "grad_norm": 0.50390625, + "learning_rate": 0.0019943104809372583, + "loss": 0.2139, + "step": 4403 + }, + { + "epoch": 0.038228834819142195, + "grad_norm": 0.5, + "learning_rate": 0.0019943071391684748, + "loss": 0.1533, + "step": 4404 + }, + { + "epoch": 0.03823751529934636, + "grad_norm": 0.4609375, + "learning_rate": 0.00199430379642169, + "loss": 0.2236, + "step": 4405 + }, + { + "epoch": 0.038246195779550525, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019943004526969082, + "loss": 0.1709, + "step": 4406 + }, + { + "epoch": 0.03825487625975469, + "grad_norm": 1.46875, + "learning_rate": 0.0019942971079941326, + "loss": 0.2227, + "step": 4407 + }, + { + "epoch": 0.038263556739958855, + "grad_norm": 0.11669921875, + "learning_rate": 0.001994293762313368, + "loss": 0.1738, + "step": 4408 + }, + { + "epoch": 0.03827223722016302, + "grad_norm": 0.1484375, + "learning_rate": 0.0019942904156546163, + "loss": 0.1621, + "step": 4409 + }, + { + "epoch": 0.038280917700367185, + "grad_norm": 0.6015625, + "learning_rate": 0.0019942870680178825, + "loss": 0.2031, + "step": 4410 + }, + { + "epoch": 0.03828959818057135, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019942837194031698, + "loss": 0.1504, + "step": 4411 + }, + { + "epoch": 0.038298278660775516, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019942803698104815, + "loss": 0.1562, + "step": 4412 + }, + { + "epoch": 0.03830695914097968, + "grad_norm": 0.28515625, + "learning_rate": 0.001994277019239822, + "loss": 0.1826, + "step": 4413 + }, + { + "epoch": 0.038315639621183846, + "grad_norm": 0.1962890625, + "learning_rate": 0.001994273667691194, + "loss": 0.2129, + "step": 4414 + }, + { + "epoch": 0.03832432010138801, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019942703151646026, + "loss": 0.1641, + "step": 4415 + }, + { + "epoch": 0.038333000581592176, + "grad_norm": 0.17578125, + "learning_rate": 0.00199426696166005, + "loss": 0.1631, + "step": 4416 + }, + { + "epoch": 0.03834168106179634, + "grad_norm": 0.392578125, + "learning_rate": 0.0019942636071775405, + "loss": 0.1836, + "step": 4417 + }, + { + "epoch": 0.038350361542000506, + "grad_norm": 0.283203125, + "learning_rate": 0.001994260251717078, + "loss": 0.1475, + "step": 4418 + }, + { + "epoch": 0.03835904202220467, + "grad_norm": 0.294921875, + "learning_rate": 0.001994256895278666, + "loss": 0.1699, + "step": 4419 + }, + { + "epoch": 0.03836772250240884, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019942535378623077, + "loss": 0.2207, + "step": 4420 + }, + { + "epoch": 0.038376402982612995, + "grad_norm": 0.19140625, + "learning_rate": 0.0019942501794680077, + "loss": 0.1465, + "step": 4421 + }, + { + "epoch": 0.03838508346281716, + "grad_norm": 0.21484375, + "learning_rate": 0.001994246820095769, + "loss": 0.1621, + "step": 4422 + }, + { + "epoch": 0.038393763943021325, + "grad_norm": 0.126953125, + "learning_rate": 0.001994243459745595, + "loss": 0.1699, + "step": 4423 + }, + { + "epoch": 0.03840244442322549, + "grad_norm": 0.76953125, + "learning_rate": 0.00199424009841749, + "loss": 0.1719, + "step": 4424 + }, + { + "epoch": 0.038411124903429655, + "grad_norm": 0.12109375, + "learning_rate": 0.0019942367361114577, + "loss": 0.1758, + "step": 4425 + }, + { + "epoch": 0.03841980538363382, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019942333728275013, + "loss": 0.1562, + "step": 4426 + }, + { + "epoch": 0.038428485863837986, + "grad_norm": 0.0791015625, + "learning_rate": 0.001994230008565625, + "loss": 0.2012, + "step": 4427 + }, + { + "epoch": 0.03843716634404215, + "grad_norm": 0.173828125, + "learning_rate": 0.001994226643325832, + "loss": 0.1406, + "step": 4428 + }, + { + "epoch": 0.038445846824246316, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019942232771081258, + "loss": 0.1738, + "step": 4429 + }, + { + "epoch": 0.03845452730445048, + "grad_norm": 0.2333984375, + "learning_rate": 0.001994219909912511, + "loss": 0.1553, + "step": 4430 + }, + { + "epoch": 0.038463207784654646, + "grad_norm": 0.326171875, + "learning_rate": 0.0019942165417389905, + "loss": 0.2227, + "step": 4431 + }, + { + "epoch": 0.03847188826485881, + "grad_norm": 0.376953125, + "learning_rate": 0.0019942131725875683, + "loss": 0.1924, + "step": 4432 + }, + { + "epoch": 0.038480568745062976, + "grad_norm": 0.365234375, + "learning_rate": 0.001994209802458248, + "loss": 0.1807, + "step": 4433 + }, + { + "epoch": 0.03848924922526714, + "grad_norm": 0.1806640625, + "learning_rate": 0.001994206431351033, + "loss": 0.1592, + "step": 4434 + }, + { + "epoch": 0.03849792970547131, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019942030592659276, + "loss": 0.1738, + "step": 4435 + }, + { + "epoch": 0.03850661018567547, + "grad_norm": 0.185546875, + "learning_rate": 0.0019941996862029355, + "loss": 0.165, + "step": 4436 + }, + { + "epoch": 0.03851529066587964, + "grad_norm": 0.09228515625, + "learning_rate": 0.00199419631216206, + "loss": 0.1514, + "step": 4437 + }, + { + "epoch": 0.0385239711460838, + "grad_norm": 0.41796875, + "learning_rate": 0.001994192937143304, + "loss": 0.1533, + "step": 4438 + }, + { + "epoch": 0.03853265162628797, + "grad_norm": 0.296875, + "learning_rate": 0.001994189561146673, + "loss": 0.1367, + "step": 4439 + }, + { + "epoch": 0.03854133210649213, + "grad_norm": 0.318359375, + "learning_rate": 0.001994186184172169, + "loss": 0.1426, + "step": 4440 + }, + { + "epoch": 0.0385500125866963, + "grad_norm": 0.12451171875, + "learning_rate": 0.001994182806219797, + "loss": 0.1455, + "step": 4441 + }, + { + "epoch": 0.03855869306690046, + "grad_norm": 0.38671875, + "learning_rate": 0.0019941794272895596, + "loss": 0.1484, + "step": 4442 + }, + { + "epoch": 0.03856737354710463, + "grad_norm": 0.111328125, + "learning_rate": 0.0019941760473814614, + "loss": 0.1562, + "step": 4443 + }, + { + "epoch": 0.03857605402730879, + "grad_norm": 0.2265625, + "learning_rate": 0.0019941726664955057, + "loss": 0.1719, + "step": 4444 + }, + { + "epoch": 0.03858473450751296, + "grad_norm": 0.10546875, + "learning_rate": 0.0019941692846316963, + "loss": 0.168, + "step": 4445 + }, + { + "epoch": 0.03859341498771712, + "grad_norm": 0.5703125, + "learning_rate": 0.0019941659017900363, + "loss": 0.1504, + "step": 4446 + }, + { + "epoch": 0.03860209546792129, + "grad_norm": 0.267578125, + "learning_rate": 0.0019941625179705305, + "loss": 0.1562, + "step": 4447 + }, + { + "epoch": 0.03861077594812545, + "grad_norm": 0.33203125, + "learning_rate": 0.0019941591331731814, + "loss": 0.1973, + "step": 4448 + }, + { + "epoch": 0.03861945642832962, + "grad_norm": 0.11328125, + "learning_rate": 0.001994155747397994, + "loss": 0.1621, + "step": 4449 + }, + { + "epoch": 0.03862813690853378, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019941523606449704, + "loss": 0.166, + "step": 4450 + }, + { + "epoch": 0.03863681738873794, + "grad_norm": 0.146484375, + "learning_rate": 0.001994148972914116, + "loss": 0.1348, + "step": 4451 + }, + { + "epoch": 0.03864549786894211, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019941455842054333, + "loss": 0.2051, + "step": 4452 + }, + { + "epoch": 0.03865417834914627, + "grad_norm": 0.212890625, + "learning_rate": 0.0019941421945189265, + "loss": 0.1504, + "step": 4453 + }, + { + "epoch": 0.03866285882935044, + "grad_norm": 0.43359375, + "learning_rate": 0.0019941388038545995, + "loss": 0.1729, + "step": 4454 + }, + { + "epoch": 0.0386715393095546, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019941354122124553, + "loss": 0.1465, + "step": 4455 + }, + { + "epoch": 0.03868021978975877, + "grad_norm": 0.447265625, + "learning_rate": 0.0019941320195924982, + "loss": 0.1699, + "step": 4456 + }, + { + "epoch": 0.03868890026996293, + "grad_norm": 0.1962890625, + "learning_rate": 0.001994128625994732, + "loss": 0.1533, + "step": 4457 + }, + { + "epoch": 0.0386975807501671, + "grad_norm": 0.12255859375, + "learning_rate": 0.00199412523141916, + "loss": 0.1357, + "step": 4458 + }, + { + "epoch": 0.03870626123037126, + "grad_norm": 0.71484375, + "learning_rate": 0.001994121835865786, + "loss": 0.1738, + "step": 4459 + }, + { + "epoch": 0.03871494171057543, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019941184393346143, + "loss": 0.1416, + "step": 4460 + }, + { + "epoch": 0.03872362219077959, + "grad_norm": 0.455078125, + "learning_rate": 0.0019941150418256474, + "loss": 0.1543, + "step": 4461 + }, + { + "epoch": 0.03873230267098376, + "grad_norm": 0.359375, + "learning_rate": 0.00199411164333889, + "loss": 0.1797, + "step": 4462 + }, + { + "epoch": 0.03874098315118792, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019941082438743457, + "loss": 0.1748, + "step": 4463 + }, + { + "epoch": 0.03874966363139209, + "grad_norm": 0.353515625, + "learning_rate": 0.001994104843432018, + "loss": 0.1934, + "step": 4464 + }, + { + "epoch": 0.03875834411159625, + "grad_norm": 0.34765625, + "learning_rate": 0.0019941014420119104, + "loss": 0.2021, + "step": 4465 + }, + { + "epoch": 0.03876702459180042, + "grad_norm": 0.8046875, + "learning_rate": 0.0019940980396140275, + "loss": 0.1367, + "step": 4466 + }, + { + "epoch": 0.03877570507200458, + "grad_norm": 0.330078125, + "learning_rate": 0.001994094636238372, + "loss": 0.208, + "step": 4467 + }, + { + "epoch": 0.03878438555220875, + "grad_norm": 0.10009765625, + "learning_rate": 0.001994091231884948, + "loss": 0.1055, + "step": 4468 + }, + { + "epoch": 0.038793066032412914, + "grad_norm": 0.31640625, + "learning_rate": 0.0019940878265537593, + "loss": 0.1719, + "step": 4469 + }, + { + "epoch": 0.03880174651261708, + "grad_norm": 0.216796875, + "learning_rate": 0.0019940844202448096, + "loss": 0.1934, + "step": 4470 + }, + { + "epoch": 0.038810426992821244, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019940810129581025, + "loss": 0.1797, + "step": 4471 + }, + { + "epoch": 0.03881910747302541, + "grad_norm": 0.08447265625, + "learning_rate": 0.001994077604693642, + "loss": 0.1377, + "step": 4472 + }, + { + "epoch": 0.038827787953229574, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019940741954514317, + "loss": 0.1875, + "step": 4473 + }, + { + "epoch": 0.03883646843343374, + "grad_norm": 0.09033203125, + "learning_rate": 0.001994070785231475, + "loss": 0.1309, + "step": 4474 + }, + { + "epoch": 0.038845148913637904, + "grad_norm": 0.193359375, + "learning_rate": 0.001994067374033776, + "loss": 0.1396, + "step": 4475 + }, + { + "epoch": 0.03885382939384207, + "grad_norm": 0.5078125, + "learning_rate": 0.001994063961858339, + "loss": 0.1602, + "step": 4476 + }, + { + "epoch": 0.038862509874046235, + "grad_norm": 0.162109375, + "learning_rate": 0.0019940605487051666, + "loss": 0.1904, + "step": 4477 + }, + { + "epoch": 0.0388711903542504, + "grad_norm": 0.08154296875, + "learning_rate": 0.001994057134574263, + "loss": 0.1406, + "step": 4478 + }, + { + "epoch": 0.038879870834454565, + "grad_norm": 0.20703125, + "learning_rate": 0.0019940537194656316, + "loss": 0.1514, + "step": 4479 + }, + { + "epoch": 0.03888855131465873, + "grad_norm": 0.5, + "learning_rate": 0.0019940503033792772, + "loss": 0.2051, + "step": 4480 + }, + { + "epoch": 0.038897231794862895, + "grad_norm": 0.263671875, + "learning_rate": 0.001994046886315202, + "loss": 0.2266, + "step": 4481 + }, + { + "epoch": 0.03890591227506705, + "grad_norm": 0.111328125, + "learning_rate": 0.0019940434682734114, + "loss": 0.1465, + "step": 4482 + }, + { + "epoch": 0.03891459275527122, + "grad_norm": 0.25, + "learning_rate": 0.001994040049253908, + "loss": 0.1572, + "step": 4483 + }, + { + "epoch": 0.038923273235475384, + "grad_norm": 0.333984375, + "learning_rate": 0.0019940366292566956, + "loss": 0.2012, + "step": 4484 + }, + { + "epoch": 0.03893195371567955, + "grad_norm": 0.470703125, + "learning_rate": 0.001994033208281778, + "loss": 0.1953, + "step": 4485 + }, + { + "epoch": 0.038940634195883714, + "grad_norm": 0.24609375, + "learning_rate": 0.00199402978632916, + "loss": 0.165, + "step": 4486 + }, + { + "epoch": 0.03894931467608788, + "grad_norm": 0.3359375, + "learning_rate": 0.0019940263633988434, + "loss": 0.1934, + "step": 4487 + }, + { + "epoch": 0.038957995156292044, + "grad_norm": 0.66015625, + "learning_rate": 0.0019940229394908335, + "loss": 0.1787, + "step": 4488 + }, + { + "epoch": 0.03896667563649621, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019940195146051338, + "loss": 0.1553, + "step": 4489 + }, + { + "epoch": 0.038975356116700374, + "grad_norm": 0.091796875, + "learning_rate": 0.0019940160887417474, + "loss": 0.1455, + "step": 4490 + }, + { + "epoch": 0.03898403659690454, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019940126619006787, + "loss": 0.1094, + "step": 4491 + }, + { + "epoch": 0.038992717077108704, + "grad_norm": 0.515625, + "learning_rate": 0.001994009234081931, + "loss": 0.1172, + "step": 4492 + }, + { + "epoch": 0.03900139755731287, + "grad_norm": 0.376953125, + "learning_rate": 0.001994005805285508, + "loss": 0.1768, + "step": 4493 + }, + { + "epoch": 0.039010078037517035, + "grad_norm": 0.287109375, + "learning_rate": 0.0019940023755114144, + "loss": 0.1436, + "step": 4494 + }, + { + "epoch": 0.0390187585177212, + "grad_norm": 0.3125, + "learning_rate": 0.0019939989447596528, + "loss": 0.1553, + "step": 4495 + }, + { + "epoch": 0.039027438997925365, + "grad_norm": 0.6484375, + "learning_rate": 0.0019939955130302274, + "loss": 0.1484, + "step": 4496 + }, + { + "epoch": 0.03903611947812953, + "grad_norm": 0.49609375, + "learning_rate": 0.001993992080323142, + "loss": 0.1807, + "step": 4497 + }, + { + "epoch": 0.039044799958333695, + "grad_norm": 0.375, + "learning_rate": 0.0019939886466384, + "loss": 0.1221, + "step": 4498 + }, + { + "epoch": 0.03905348043853786, + "grad_norm": 0.443359375, + "learning_rate": 0.001993985211976006, + "loss": 0.1543, + "step": 4499 + }, + { + "epoch": 0.039062160918742025, + "grad_norm": 0.130859375, + "learning_rate": 0.0019939817763359622, + "loss": 0.1533, + "step": 4500 + }, + { + "epoch": 0.03907084139894619, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019939783397182743, + "loss": 0.1328, + "step": 4501 + }, + { + "epoch": 0.039079521879150356, + "grad_norm": 0.326171875, + "learning_rate": 0.001993974902122945, + "loss": 0.1855, + "step": 4502 + }, + { + "epoch": 0.03908820235935452, + "grad_norm": 0.1591796875, + "learning_rate": 0.001993971463549978, + "loss": 0.1377, + "step": 4503 + }, + { + "epoch": 0.039096882839558686, + "grad_norm": 0.10205078125, + "learning_rate": 0.001993968023999377, + "loss": 0.1514, + "step": 4504 + }, + { + "epoch": 0.03910556331976285, + "grad_norm": 0.2001953125, + "learning_rate": 0.001993964583471147, + "loss": 0.1523, + "step": 4505 + }, + { + "epoch": 0.039114243799967016, + "grad_norm": 0.09375, + "learning_rate": 0.0019939611419652896, + "loss": 0.1602, + "step": 4506 + }, + { + "epoch": 0.03912292428017118, + "grad_norm": 0.1923828125, + "learning_rate": 0.00199395769948181, + "loss": 0.1494, + "step": 4507 + }, + { + "epoch": 0.039131604760375346, + "grad_norm": 0.0771484375, + "learning_rate": 0.001993954256020712, + "loss": 0.1406, + "step": 4508 + }, + { + "epoch": 0.03914028524057951, + "grad_norm": 0.30078125, + "learning_rate": 0.001993950811581999, + "loss": 0.1562, + "step": 4509 + }, + { + "epoch": 0.03914896572078368, + "grad_norm": 0.162109375, + "learning_rate": 0.0019939473661656744, + "loss": 0.2012, + "step": 4510 + }, + { + "epoch": 0.03915764620098784, + "grad_norm": 0.357421875, + "learning_rate": 0.001993943919771743, + "loss": 0.165, + "step": 4511 + }, + { + "epoch": 0.03916632668119201, + "grad_norm": 0.3515625, + "learning_rate": 0.0019939404724002075, + "loss": 0.1836, + "step": 4512 + }, + { + "epoch": 0.039175007161396165, + "grad_norm": 0.07421875, + "learning_rate": 0.0019939370240510726, + "loss": 0.168, + "step": 4513 + }, + { + "epoch": 0.03918368764160033, + "grad_norm": 0.10546875, + "learning_rate": 0.0019939335747243413, + "loss": 0.1895, + "step": 4514 + }, + { + "epoch": 0.039192368121804495, + "grad_norm": 0.0859375, + "learning_rate": 0.001993930124420018, + "loss": 0.1758, + "step": 4515 + }, + { + "epoch": 0.03920104860200866, + "grad_norm": 0.11767578125, + "learning_rate": 0.001993926673138105, + "loss": 0.1826, + "step": 4516 + }, + { + "epoch": 0.039209729082212826, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019939232208786087, + "loss": 0.1172, + "step": 4517 + }, + { + "epoch": 0.03921840956241699, + "grad_norm": 0.26171875, + "learning_rate": 0.0019939197676415304, + "loss": 0.1562, + "step": 4518 + }, + { + "epoch": 0.039227090042621156, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019939163134268753, + "loss": 0.1592, + "step": 4519 + }, + { + "epoch": 0.03923577052282532, + "grad_norm": 0.09375, + "learning_rate": 0.001993912858234647, + "loss": 0.1758, + "step": 4520 + }, + { + "epoch": 0.039244451003029486, + "grad_norm": 0.296875, + "learning_rate": 0.0019939094020648487, + "loss": 0.1748, + "step": 4521 + }, + { + "epoch": 0.03925313148323365, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019939059449174843, + "loss": 0.1377, + "step": 4522 + }, + { + "epoch": 0.039261811963437816, + "grad_norm": 0.23828125, + "learning_rate": 0.001993902486792558, + "loss": 0.2363, + "step": 4523 + }, + { + "epoch": 0.03927049244364198, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019938990276900737, + "loss": 0.1797, + "step": 4524 + }, + { + "epoch": 0.039279172923846147, + "grad_norm": 0.640625, + "learning_rate": 0.0019938955676100344, + "loss": 0.1406, + "step": 4525 + }, + { + "epoch": 0.03928785340405031, + "grad_norm": 0.298828125, + "learning_rate": 0.0019938921065524445, + "loss": 0.2266, + "step": 4526 + }, + { + "epoch": 0.03929653388425448, + "grad_norm": 0.201171875, + "learning_rate": 0.0019938886445173077, + "loss": 0.1543, + "step": 4527 + }, + { + "epoch": 0.03930521436445864, + "grad_norm": 0.130859375, + "learning_rate": 0.0019938851815046277, + "loss": 0.1582, + "step": 4528 + }, + { + "epoch": 0.03931389484466281, + "grad_norm": 0.443359375, + "learning_rate": 0.001993881717514408, + "loss": 0.1514, + "step": 4529 + }, + { + "epoch": 0.03932257532486697, + "grad_norm": 0.49609375, + "learning_rate": 0.0019938782525466526, + "loss": 0.1904, + "step": 4530 + }, + { + "epoch": 0.03933125580507114, + "grad_norm": 0.375, + "learning_rate": 0.001993874786601366, + "loss": 0.1855, + "step": 4531 + }, + { + "epoch": 0.0393399362852753, + "grad_norm": 0.447265625, + "learning_rate": 0.0019938713196785514, + "loss": 0.1475, + "step": 4532 + }, + { + "epoch": 0.03934861676547947, + "grad_norm": 0.3359375, + "learning_rate": 0.0019938678517782116, + "loss": 0.2539, + "step": 4533 + }, + { + "epoch": 0.03935729724568363, + "grad_norm": 0.6171875, + "learning_rate": 0.001993864382900352, + "loss": 0.1318, + "step": 4534 + }, + { + "epoch": 0.0393659777258878, + "grad_norm": 0.390625, + "learning_rate": 0.0019938609130449756, + "loss": 0.1992, + "step": 4535 + }, + { + "epoch": 0.03937465820609196, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019938574422120867, + "loss": 0.2148, + "step": 4536 + }, + { + "epoch": 0.03938333868629613, + "grad_norm": 0.294921875, + "learning_rate": 0.001993853970401688, + "loss": 0.105, + "step": 4537 + }, + { + "epoch": 0.03939201916650029, + "grad_norm": 0.1416015625, + "learning_rate": 0.001993850497613785, + "loss": 0.2227, + "step": 4538 + }, + { + "epoch": 0.03940069964670446, + "grad_norm": 0.13671875, + "learning_rate": 0.00199384702384838, + "loss": 0.1533, + "step": 4539 + }, + { + "epoch": 0.03940938012690862, + "grad_norm": 0.0732421875, + "learning_rate": 0.001993843549105477, + "loss": 0.1387, + "step": 4540 + }, + { + "epoch": 0.03941806060711279, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019938400733850808, + "loss": 0.1348, + "step": 4541 + }, + { + "epoch": 0.039426741087316954, + "grad_norm": 0.404296875, + "learning_rate": 0.0019938365966871937, + "loss": 0.1738, + "step": 4542 + }, + { + "epoch": 0.03943542156752112, + "grad_norm": 0.19921875, + "learning_rate": 0.001993833119011821, + "loss": 0.1426, + "step": 4543 + }, + { + "epoch": 0.03944410204772528, + "grad_norm": 0.263671875, + "learning_rate": 0.0019938296403589654, + "loss": 0.1865, + "step": 4544 + }, + { + "epoch": 0.03945278252792944, + "grad_norm": 0.490234375, + "learning_rate": 0.0019938261607286316, + "loss": 0.2148, + "step": 4545 + }, + { + "epoch": 0.03946146300813361, + "grad_norm": 1.09375, + "learning_rate": 0.0019938226801208226, + "loss": 0.1582, + "step": 4546 + }, + { + "epoch": 0.03947014348833777, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019938191985355426, + "loss": 0.1406, + "step": 4547 + }, + { + "epoch": 0.03947882396854194, + "grad_norm": 0.169921875, + "learning_rate": 0.0019938157159727953, + "loss": 0.168, + "step": 4548 + }, + { + "epoch": 0.0394875044487461, + "grad_norm": 0.392578125, + "learning_rate": 0.001993812232432585, + "loss": 0.1484, + "step": 4549 + }, + { + "epoch": 0.03949618492895027, + "grad_norm": 0.4921875, + "learning_rate": 0.0019938087479149146, + "loss": 0.1436, + "step": 4550 + }, + { + "epoch": 0.03950486540915443, + "grad_norm": 0.578125, + "learning_rate": 0.0019938052624197886, + "loss": 0.1221, + "step": 4551 + }, + { + "epoch": 0.0395135458893586, + "grad_norm": 0.79296875, + "learning_rate": 0.0019938017759472105, + "loss": 0.1826, + "step": 4552 + }, + { + "epoch": 0.03952222636956276, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019937982884971842, + "loss": 0.1621, + "step": 4553 + }, + { + "epoch": 0.03953090684976693, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019937948000697133, + "loss": 0.1992, + "step": 4554 + }, + { + "epoch": 0.03953958732997109, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019937913106648024, + "loss": 0.1338, + "step": 4555 + }, + { + "epoch": 0.03954826781017526, + "grad_norm": 0.53125, + "learning_rate": 0.0019937878202824543, + "loss": 0.1699, + "step": 4556 + }, + { + "epoch": 0.03955694829037942, + "grad_norm": 0.5625, + "learning_rate": 0.0019937843289226736, + "loss": 0.2188, + "step": 4557 + }, + { + "epoch": 0.03956562877058359, + "grad_norm": 0.208984375, + "learning_rate": 0.0019937808365854633, + "loss": 0.167, + "step": 4558 + }, + { + "epoch": 0.039574309250787754, + "grad_norm": 0.068359375, + "learning_rate": 0.001993777343270828, + "loss": 0.1426, + "step": 4559 + }, + { + "epoch": 0.03958298973099192, + "grad_norm": 0.2734375, + "learning_rate": 0.0019937738489787713, + "loss": 0.168, + "step": 4560 + }, + { + "epoch": 0.039591670211196084, + "grad_norm": 0.251953125, + "learning_rate": 0.001993770353709297, + "loss": 0.1758, + "step": 4561 + }, + { + "epoch": 0.03960035069140025, + "grad_norm": 0.14453125, + "learning_rate": 0.0019937668574624085, + "loss": 0.1895, + "step": 4562 + }, + { + "epoch": 0.039609031171604414, + "grad_norm": 1.1953125, + "learning_rate": 0.0019937633602381106, + "loss": 0.3926, + "step": 4563 + }, + { + "epoch": 0.03961771165180858, + "grad_norm": 0.89453125, + "learning_rate": 0.001993759862036406, + "loss": 0.1543, + "step": 4564 + }, + { + "epoch": 0.039626392132012744, + "grad_norm": 0.765625, + "learning_rate": 0.001993756362857299, + "loss": 0.1494, + "step": 4565 + }, + { + "epoch": 0.03963507261221691, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019937528627007937, + "loss": 0.1943, + "step": 4566 + }, + { + "epoch": 0.039643753092421075, + "grad_norm": 0.28125, + "learning_rate": 0.001993749361566894, + "loss": 0.1777, + "step": 4567 + }, + { + "epoch": 0.03965243357262524, + "grad_norm": 0.220703125, + "learning_rate": 0.001993745859455603, + "loss": 0.1523, + "step": 4568 + }, + { + "epoch": 0.039661114052829405, + "grad_norm": 0.35546875, + "learning_rate": 0.001993742356366925, + "loss": 0.1533, + "step": 4569 + }, + { + "epoch": 0.03966979453303357, + "grad_norm": 0.65234375, + "learning_rate": 0.0019937388523008637, + "loss": 0.1895, + "step": 4570 + }, + { + "epoch": 0.039678475013237735, + "grad_norm": 0.259765625, + "learning_rate": 0.0019937353472574233, + "loss": 0.1328, + "step": 4571 + }, + { + "epoch": 0.0396871554934419, + "grad_norm": 0.3984375, + "learning_rate": 0.001993731841236607, + "loss": 0.1631, + "step": 4572 + }, + { + "epoch": 0.039695835973646065, + "grad_norm": 0.220703125, + "learning_rate": 0.0019937283342384192, + "loss": 0.1348, + "step": 4573 + }, + { + "epoch": 0.03970451645385023, + "grad_norm": 0.123046875, + "learning_rate": 0.0019937248262628634, + "loss": 0.1484, + "step": 4574 + }, + { + "epoch": 0.03971319693405439, + "grad_norm": 0.0849609375, + "learning_rate": 0.001993721317309944, + "loss": 0.1621, + "step": 4575 + }, + { + "epoch": 0.039721877414258554, + "grad_norm": 0.2890625, + "learning_rate": 0.001993717807379664, + "loss": 0.1973, + "step": 4576 + }, + { + "epoch": 0.03973055789446272, + "grad_norm": 0.23046875, + "learning_rate": 0.0019937142964720276, + "loss": 0.1357, + "step": 4577 + }, + { + "epoch": 0.039739238374666884, + "grad_norm": 0.126953125, + "learning_rate": 0.0019937107845870387, + "loss": 0.1309, + "step": 4578 + }, + { + "epoch": 0.03974791885487105, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019937072717247013, + "loss": 0.1777, + "step": 4579 + }, + { + "epoch": 0.039756599335075214, + "grad_norm": 0.462890625, + "learning_rate": 0.001993703757885019, + "loss": 0.2012, + "step": 4580 + }, + { + "epoch": 0.03976527981527938, + "grad_norm": 0.337890625, + "learning_rate": 0.0019937002430679956, + "loss": 0.1953, + "step": 4581 + }, + { + "epoch": 0.039773960295483544, + "grad_norm": 0.275390625, + "learning_rate": 0.001993696727273635, + "loss": 0.1533, + "step": 4582 + }, + { + "epoch": 0.03978264077568771, + "grad_norm": 0.099609375, + "learning_rate": 0.0019936932105019413, + "loss": 0.1758, + "step": 4583 + }, + { + "epoch": 0.039791321255891875, + "grad_norm": 0.15234375, + "learning_rate": 0.001993689692752918, + "loss": 0.1924, + "step": 4584 + }, + { + "epoch": 0.03980000173609604, + "grad_norm": 0.1943359375, + "learning_rate": 0.001993686174026569, + "loss": 0.1426, + "step": 4585 + }, + { + "epoch": 0.039808682216300205, + "grad_norm": 0.90234375, + "learning_rate": 0.0019936826543228985, + "loss": 0.1963, + "step": 4586 + }, + { + "epoch": 0.03981736269650437, + "grad_norm": 0.494140625, + "learning_rate": 0.00199367913364191, + "loss": 0.1475, + "step": 4587 + }, + { + "epoch": 0.039826043176708535, + "grad_norm": 0.361328125, + "learning_rate": 0.001993675611983607, + "loss": 0.1602, + "step": 4588 + }, + { + "epoch": 0.0398347236569127, + "grad_norm": 0.09130859375, + "learning_rate": 0.001993672089347994, + "loss": 0.1738, + "step": 4589 + }, + { + "epoch": 0.039843404137116865, + "grad_norm": 0.8359375, + "learning_rate": 0.0019936685657350748, + "loss": 0.209, + "step": 4590 + }, + { + "epoch": 0.03985208461732103, + "grad_norm": 0.51171875, + "learning_rate": 0.001993665041144853, + "loss": 0.1621, + "step": 4591 + }, + { + "epoch": 0.039860765097525196, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019936615155773324, + "loss": 0.1582, + "step": 4592 + }, + { + "epoch": 0.03986944557772936, + "grad_norm": 0.33984375, + "learning_rate": 0.001993657989032517, + "loss": 0.1641, + "step": 4593 + }, + { + "epoch": 0.039878126057933526, + "grad_norm": 0.11181640625, + "learning_rate": 0.001993654461510411, + "loss": 0.1416, + "step": 4594 + }, + { + "epoch": 0.03988680653813769, + "grad_norm": 0.1689453125, + "learning_rate": 0.001993650933011018, + "loss": 0.1562, + "step": 4595 + }, + { + "epoch": 0.039895487018341856, + "grad_norm": 0.283203125, + "learning_rate": 0.0019936474035343412, + "loss": 0.1514, + "step": 4596 + }, + { + "epoch": 0.03990416749854602, + "grad_norm": 0.095703125, + "learning_rate": 0.0019936438730803856, + "loss": 0.1826, + "step": 4597 + }, + { + "epoch": 0.039912847978750186, + "grad_norm": 0.69921875, + "learning_rate": 0.001993640341649154, + "loss": 0.1719, + "step": 4598 + }, + { + "epoch": 0.03992152845895435, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019936368092406506, + "loss": 0.1895, + "step": 4599 + }, + { + "epoch": 0.03993020893915852, + "grad_norm": 0.1845703125, + "learning_rate": 0.00199363327585488, + "loss": 0.1699, + "step": 4600 + }, + { + "epoch": 0.03993888941936268, + "grad_norm": 0.09033203125, + "learning_rate": 0.001993629741491845, + "loss": 0.1641, + "step": 4601 + }, + { + "epoch": 0.03994756989956685, + "grad_norm": 0.283203125, + "learning_rate": 0.0019936262061515503, + "loss": 0.1543, + "step": 4602 + }, + { + "epoch": 0.03995625037977101, + "grad_norm": 0.77734375, + "learning_rate": 0.001993622669833999, + "loss": 0.1709, + "step": 4603 + }, + { + "epoch": 0.03996493085997518, + "grad_norm": 0.365234375, + "learning_rate": 0.001993619132539196, + "loss": 0.1602, + "step": 4604 + }, + { + "epoch": 0.039973611340179335, + "grad_norm": 0.5, + "learning_rate": 0.001993615594267144, + "loss": 0.1592, + "step": 4605 + }, + { + "epoch": 0.0399822918203835, + "grad_norm": 0.427734375, + "learning_rate": 0.0019936120550178476, + "loss": 0.1426, + "step": 4606 + }, + { + "epoch": 0.039990972300587665, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019936085147913107, + "loss": 0.1895, + "step": 4607 + }, + { + "epoch": 0.03999965278079183, + "grad_norm": 0.08349609375, + "learning_rate": 0.001993604973587537, + "loss": 0.1387, + "step": 4608 + }, + { + "epoch": 0.040008333260995996, + "grad_norm": 0.31640625, + "learning_rate": 0.0019936014314065297, + "loss": 0.2656, + "step": 4609 + }, + { + "epoch": 0.04001701374120016, + "grad_norm": 0.197265625, + "learning_rate": 0.0019935978882482937, + "loss": 0.1562, + "step": 4610 + }, + { + "epoch": 0.040025694221404326, + "grad_norm": 0.87890625, + "learning_rate": 0.0019935943441128324, + "loss": 0.1748, + "step": 4611 + }, + { + "epoch": 0.04003437470160849, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019935907990001494, + "loss": 0.1387, + "step": 4612 + }, + { + "epoch": 0.040043055181812656, + "grad_norm": 0.26171875, + "learning_rate": 0.0019935872529102494, + "loss": 0.166, + "step": 4613 + }, + { + "epoch": 0.04005173566201682, + "grad_norm": 0.416015625, + "learning_rate": 0.001993583705843136, + "loss": 0.1406, + "step": 4614 + }, + { + "epoch": 0.040060416142220986, + "grad_norm": 0.201171875, + "learning_rate": 0.0019935801577988126, + "loss": 0.125, + "step": 4615 + }, + { + "epoch": 0.04006909662242515, + "grad_norm": 0.373046875, + "learning_rate": 0.0019935766087772833, + "loss": 0.1904, + "step": 4616 + }, + { + "epoch": 0.04007777710262932, + "grad_norm": 0.29296875, + "learning_rate": 0.001993573058778552, + "loss": 0.1797, + "step": 4617 + }, + { + "epoch": 0.04008645758283348, + "grad_norm": 0.5859375, + "learning_rate": 0.001993569507802623, + "loss": 0.1641, + "step": 4618 + }, + { + "epoch": 0.04009513806303765, + "grad_norm": 0.9609375, + "learning_rate": 0.0019935659558494995, + "loss": 0.1602, + "step": 4619 + }, + { + "epoch": 0.04010381854324181, + "grad_norm": 0.099609375, + "learning_rate": 0.001993562402919186, + "loss": 0.1875, + "step": 4620 + }, + { + "epoch": 0.04011249902344598, + "grad_norm": 0.69921875, + "learning_rate": 0.0019935588490116855, + "loss": 0.168, + "step": 4621 + }, + { + "epoch": 0.04012117950365014, + "grad_norm": 0.248046875, + "learning_rate": 0.001993555294127003, + "loss": 0.1826, + "step": 4622 + }, + { + "epoch": 0.04012985998385431, + "grad_norm": 0.380859375, + "learning_rate": 0.0019935517382651414, + "loss": 0.1729, + "step": 4623 + }, + { + "epoch": 0.04013854046405847, + "grad_norm": 0.19140625, + "learning_rate": 0.0019935481814261054, + "loss": 0.1543, + "step": 4624 + }, + { + "epoch": 0.04014722094426264, + "grad_norm": 0.15234375, + "learning_rate": 0.0019935446236098984, + "loss": 0.1934, + "step": 4625 + }, + { + "epoch": 0.0401559014244668, + "grad_norm": 0.5, + "learning_rate": 0.0019935410648165247, + "loss": 0.1523, + "step": 4626 + }, + { + "epoch": 0.04016458190467097, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019935375050459873, + "loss": 0.1572, + "step": 4627 + }, + { + "epoch": 0.04017326238487513, + "grad_norm": 0.66796875, + "learning_rate": 0.001993533944298291, + "loss": 0.1699, + "step": 4628 + }, + { + "epoch": 0.0401819428650793, + "grad_norm": 0.28125, + "learning_rate": 0.00199353038257344, + "loss": 0.1738, + "step": 4629 + }, + { + "epoch": 0.04019062334528346, + "grad_norm": 0.4453125, + "learning_rate": 0.0019935268198714366, + "loss": 0.2031, + "step": 4630 + }, + { + "epoch": 0.04019930382548763, + "grad_norm": 0.34375, + "learning_rate": 0.0019935232561922867, + "loss": 0.123, + "step": 4631 + }, + { + "epoch": 0.040207984305691793, + "grad_norm": 0.458984375, + "learning_rate": 0.001993519691535992, + "loss": 0.1406, + "step": 4632 + }, + { + "epoch": 0.04021666478589596, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019935161259025586, + "loss": 0.1797, + "step": 4633 + }, + { + "epoch": 0.040225345266100124, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019935125592919893, + "loss": 0.1367, + "step": 4634 + }, + { + "epoch": 0.04023402574630429, + "grad_norm": 0.09033203125, + "learning_rate": 0.001993508991704288, + "loss": 0.165, + "step": 4635 + }, + { + "epoch": 0.04024270622650845, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019935054231394584, + "loss": 0.1797, + "step": 4636 + }, + { + "epoch": 0.04025138670671261, + "grad_norm": 0.43359375, + "learning_rate": 0.0019935018535975047, + "loss": 0.1475, + "step": 4637 + }, + { + "epoch": 0.04026006718691678, + "grad_norm": 0.1650390625, + "learning_rate": 0.001993498283078431, + "loss": 0.1934, + "step": 4638 + }, + { + "epoch": 0.04026874766712094, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019934947115822408, + "loss": 0.1621, + "step": 4639 + }, + { + "epoch": 0.04027742814732511, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019934911391089384, + "loss": 0.168, + "step": 4640 + }, + { + "epoch": 0.04028610862752927, + "grad_norm": 0.75390625, + "learning_rate": 0.0019934875656585273, + "loss": 0.2168, + "step": 4641 + }, + { + "epoch": 0.04029478910773344, + "grad_norm": 0.30078125, + "learning_rate": 0.001993483991231012, + "loss": 0.2188, + "step": 4642 + }, + { + "epoch": 0.0403034695879376, + "grad_norm": 0.921875, + "learning_rate": 0.0019934804158263956, + "loss": 0.1855, + "step": 4643 + }, + { + "epoch": 0.04031215006814177, + "grad_norm": 0.42578125, + "learning_rate": 0.0019934768394446827, + "loss": 0.2334, + "step": 4644 + }, + { + "epoch": 0.04032083054834593, + "grad_norm": 0.83984375, + "learning_rate": 0.0019934732620858773, + "loss": 0.1328, + "step": 4645 + }, + { + "epoch": 0.0403295110285501, + "grad_norm": 0.29296875, + "learning_rate": 0.0019934696837499823, + "loss": 0.1699, + "step": 4646 + }, + { + "epoch": 0.04033819150875426, + "grad_norm": 0.283203125, + "learning_rate": 0.0019934661044370026, + "loss": 0.1982, + "step": 4647 + }, + { + "epoch": 0.04034687198895843, + "grad_norm": 0.212890625, + "learning_rate": 0.0019934625241469417, + "loss": 0.1602, + "step": 4648 + }, + { + "epoch": 0.040355552469162594, + "grad_norm": 0.265625, + "learning_rate": 0.0019934589428798038, + "loss": 0.1328, + "step": 4649 + }, + { + "epoch": 0.04036423294936676, + "grad_norm": 0.265625, + "learning_rate": 0.0019934553606355924, + "loss": 0.1523, + "step": 4650 + }, + { + "epoch": 0.040372913429570924, + "grad_norm": 0.232421875, + "learning_rate": 0.0019934517774143116, + "loss": 0.1475, + "step": 4651 + }, + { + "epoch": 0.04038159390977509, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019934481932159655, + "loss": 0.1416, + "step": 4652 + }, + { + "epoch": 0.040390274389979254, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019934446080405576, + "loss": 0.1914, + "step": 4653 + }, + { + "epoch": 0.04039895487018342, + "grad_norm": 0.353515625, + "learning_rate": 0.0019934410218880928, + "loss": 0.1602, + "step": 4654 + }, + { + "epoch": 0.040407635350387584, + "grad_norm": 0.451171875, + "learning_rate": 0.0019934374347585736, + "loss": 0.1699, + "step": 4655 + }, + { + "epoch": 0.04041631583059175, + "grad_norm": 0.5546875, + "learning_rate": 0.001993433846652005, + "loss": 0.1592, + "step": 4656 + }, + { + "epoch": 0.040424996310795915, + "grad_norm": 0.77734375, + "learning_rate": 0.0019934302575683903, + "loss": 0.1816, + "step": 4657 + }, + { + "epoch": 0.04043367679100008, + "grad_norm": 0.279296875, + "learning_rate": 0.001993426667507734, + "loss": 0.1504, + "step": 4658 + }, + { + "epoch": 0.040442357271204245, + "grad_norm": 0.359375, + "learning_rate": 0.0019934230764700397, + "loss": 0.1182, + "step": 4659 + }, + { + "epoch": 0.04045103775140841, + "grad_norm": 0.10498046875, + "learning_rate": 0.001993419484455311, + "loss": 0.1445, + "step": 4660 + }, + { + "epoch": 0.040459718231612575, + "grad_norm": 0.39453125, + "learning_rate": 0.001993415891463553, + "loss": 0.1484, + "step": 4661 + }, + { + "epoch": 0.04046839871181674, + "grad_norm": 0.236328125, + "learning_rate": 0.0019934122974947684, + "loss": 0.1396, + "step": 4662 + }, + { + "epoch": 0.040477079192020905, + "grad_norm": 0.271484375, + "learning_rate": 0.001993408702548961, + "loss": 0.1875, + "step": 4663 + }, + { + "epoch": 0.04048575967222507, + "grad_norm": 0.5703125, + "learning_rate": 0.0019934051066261356, + "loss": 0.2188, + "step": 4664 + }, + { + "epoch": 0.040494440152429236, + "grad_norm": 0.353515625, + "learning_rate": 0.001993401509726296, + "loss": 0.1592, + "step": 4665 + }, + { + "epoch": 0.0405031206326334, + "grad_norm": 0.74609375, + "learning_rate": 0.0019933979118494454, + "loss": 0.1426, + "step": 4666 + }, + { + "epoch": 0.04051180111283756, + "grad_norm": 0.314453125, + "learning_rate": 0.001993394312995589, + "loss": 0.1367, + "step": 4667 + }, + { + "epoch": 0.040520481593041724, + "grad_norm": 0.4140625, + "learning_rate": 0.0019933907131647294, + "loss": 0.1885, + "step": 4668 + }, + { + "epoch": 0.04052916207324589, + "grad_norm": 0.130859375, + "learning_rate": 0.0019933871123568713, + "loss": 0.1455, + "step": 4669 + }, + { + "epoch": 0.040537842553450054, + "grad_norm": 0.19140625, + "learning_rate": 0.0019933835105720187, + "loss": 0.1611, + "step": 4670 + }, + { + "epoch": 0.04054652303365422, + "grad_norm": 0.3671875, + "learning_rate": 0.001993379907810175, + "loss": 0.2148, + "step": 4671 + }, + { + "epoch": 0.040555203513858384, + "grad_norm": 0.271484375, + "learning_rate": 0.0019933763040713448, + "loss": 0.1768, + "step": 4672 + }, + { + "epoch": 0.04056388399406255, + "grad_norm": 0.423828125, + "learning_rate": 0.0019933726993555316, + "loss": 0.1069, + "step": 4673 + }, + { + "epoch": 0.040572564474266715, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019933690936627395, + "loss": 0.2051, + "step": 4674 + }, + { + "epoch": 0.04058124495447088, + "grad_norm": 0.478515625, + "learning_rate": 0.001993365486992972, + "loss": 0.1787, + "step": 4675 + }, + { + "epoch": 0.040589925434675045, + "grad_norm": 0.490234375, + "learning_rate": 0.0019933618793462338, + "loss": 0.1758, + "step": 4676 + }, + { + "epoch": 0.04059860591487921, + "grad_norm": 0.193359375, + "learning_rate": 0.0019933582707225284, + "loss": 0.1436, + "step": 4677 + }, + { + "epoch": 0.040607286395083375, + "grad_norm": 0.4765625, + "learning_rate": 0.0019933546611218596, + "loss": 0.1426, + "step": 4678 + }, + { + "epoch": 0.04061596687528754, + "grad_norm": 0.19140625, + "learning_rate": 0.0019933510505442315, + "loss": 0.2168, + "step": 4679 + }, + { + "epoch": 0.040624647355491705, + "grad_norm": 0.095703125, + "learning_rate": 0.001993347438989649, + "loss": 0.165, + "step": 4680 + }, + { + "epoch": 0.04063332783569587, + "grad_norm": 0.22265625, + "learning_rate": 0.001993343826458114, + "loss": 0.1104, + "step": 4681 + }, + { + "epoch": 0.040642008315900036, + "grad_norm": 0.76953125, + "learning_rate": 0.0019933402129496324, + "loss": 0.1963, + "step": 4682 + }, + { + "epoch": 0.0406506887961042, + "grad_norm": 0.404296875, + "learning_rate": 0.001993336598464207, + "loss": 0.1777, + "step": 4683 + }, + { + "epoch": 0.040659369276308366, + "grad_norm": 0.30859375, + "learning_rate": 0.0019933329830018423, + "loss": 0.1992, + "step": 4684 + }, + { + "epoch": 0.04066804975651253, + "grad_norm": 0.44921875, + "learning_rate": 0.001993329366562542, + "loss": 0.1553, + "step": 4685 + }, + { + "epoch": 0.040676730236716696, + "grad_norm": 0.09814453125, + "learning_rate": 0.00199332574914631, + "loss": 0.165, + "step": 4686 + }, + { + "epoch": 0.04068541071692086, + "grad_norm": 0.82421875, + "learning_rate": 0.00199332213075315, + "loss": 0.2129, + "step": 4687 + }, + { + "epoch": 0.040694091197125026, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019933185113830674, + "loss": 0.1543, + "step": 4688 + }, + { + "epoch": 0.04070277167732919, + "grad_norm": 0.63671875, + "learning_rate": 0.0019933148910360643, + "loss": 0.1465, + "step": 4689 + }, + { + "epoch": 0.04071145215753336, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019933112697121456, + "loss": 0.1445, + "step": 4690 + }, + { + "epoch": 0.04072013263773752, + "grad_norm": 0.6171875, + "learning_rate": 0.0019933076474113152, + "loss": 0.1973, + "step": 4691 + }, + { + "epoch": 0.04072881311794169, + "grad_norm": 1.1640625, + "learning_rate": 0.001993304024133577, + "loss": 0.1992, + "step": 4692 + }, + { + "epoch": 0.04073749359814585, + "grad_norm": 0.79296875, + "learning_rate": 0.001993300399878935, + "loss": 0.1553, + "step": 4693 + }, + { + "epoch": 0.04074617407835002, + "grad_norm": 0.59375, + "learning_rate": 0.001993296774647393, + "loss": 0.1699, + "step": 4694 + }, + { + "epoch": 0.04075485455855418, + "grad_norm": 0.578125, + "learning_rate": 0.001993293148438955, + "loss": 0.1602, + "step": 4695 + }, + { + "epoch": 0.04076353503875835, + "grad_norm": 0.0849609375, + "learning_rate": 0.001993289521253625, + "loss": 0.124, + "step": 4696 + }, + { + "epoch": 0.04077221551896251, + "grad_norm": 0.76171875, + "learning_rate": 0.0019932858930914073, + "loss": 0.1895, + "step": 4697 + }, + { + "epoch": 0.04078089599916667, + "grad_norm": 0.65234375, + "learning_rate": 0.001993282263952305, + "loss": 0.1572, + "step": 4698 + }, + { + "epoch": 0.040789576479370836, + "grad_norm": 1.109375, + "learning_rate": 0.0019932786338363235, + "loss": 0.1357, + "step": 4699 + }, + { + "epoch": 0.040798256959575, + "grad_norm": 1.1796875, + "learning_rate": 0.0019932750027434653, + "loss": 0.1758, + "step": 4700 + }, + { + "epoch": 0.040806937439779166, + "grad_norm": 0.578125, + "learning_rate": 0.001993271370673735, + "loss": 0.1445, + "step": 4701 + }, + { + "epoch": 0.04081561791998333, + "grad_norm": 0.62109375, + "learning_rate": 0.001993267737627137, + "loss": 0.1426, + "step": 4702 + }, + { + "epoch": 0.040824298400187496, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019932641036036745, + "loss": 0.1494, + "step": 4703 + }, + { + "epoch": 0.04083297888039166, + "grad_norm": 0.197265625, + "learning_rate": 0.0019932604686033516, + "loss": 0.1777, + "step": 4704 + }, + { + "epoch": 0.040841659360595826, + "grad_norm": 0.474609375, + "learning_rate": 0.0019932568326261725, + "loss": 0.1582, + "step": 4705 + }, + { + "epoch": 0.04085033984079999, + "grad_norm": 0.345703125, + "learning_rate": 0.0019932531956721416, + "loss": 0.1768, + "step": 4706 + }, + { + "epoch": 0.04085902032100416, + "grad_norm": 0.1015625, + "learning_rate": 0.001993249557741262, + "loss": 0.2207, + "step": 4707 + }, + { + "epoch": 0.04086770080120832, + "grad_norm": 0.50390625, + "learning_rate": 0.001993245918833538, + "loss": 0.1387, + "step": 4708 + }, + { + "epoch": 0.04087638128141249, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019932422789489743, + "loss": 0.1738, + "step": 4709 + }, + { + "epoch": 0.04088506176161665, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019932386380875737, + "loss": 0.1562, + "step": 4710 + }, + { + "epoch": 0.04089374224182082, + "grad_norm": 0.1455078125, + "learning_rate": 0.001993234996249341, + "loss": 0.166, + "step": 4711 + }, + { + "epoch": 0.04090242272202498, + "grad_norm": 0.318359375, + "learning_rate": 0.00199323135343428, + "loss": 0.1484, + "step": 4712 + }, + { + "epoch": 0.04091110320222915, + "grad_norm": 0.099609375, + "learning_rate": 0.0019932277096423945, + "loss": 0.1641, + "step": 4713 + }, + { + "epoch": 0.04091978368243331, + "grad_norm": 0.3046875, + "learning_rate": 0.0019932240648736885, + "loss": 0.1641, + "step": 4714 + }, + { + "epoch": 0.04092846416263748, + "grad_norm": 0.111328125, + "learning_rate": 0.001993220419128166, + "loss": 0.1543, + "step": 4715 + }, + { + "epoch": 0.04093714464284164, + "grad_norm": 0.1845703125, + "learning_rate": 0.001993216772405831, + "loss": 0.1689, + "step": 4716 + }, + { + "epoch": 0.04094582512304581, + "grad_norm": 1.703125, + "learning_rate": 0.001993213124706688, + "loss": 0.1895, + "step": 4717 + }, + { + "epoch": 0.04095450560324997, + "grad_norm": 0.71484375, + "learning_rate": 0.0019932094760307406, + "loss": 0.1748, + "step": 4718 + }, + { + "epoch": 0.04096318608345414, + "grad_norm": 0.59765625, + "learning_rate": 0.0019932058263779926, + "loss": 0.1367, + "step": 4719 + }, + { + "epoch": 0.0409718665636583, + "grad_norm": 0.12060546875, + "learning_rate": 0.001993202175748448, + "loss": 0.167, + "step": 4720 + }, + { + "epoch": 0.04098054704386247, + "grad_norm": 0.337890625, + "learning_rate": 0.0019931985241421106, + "loss": 0.1562, + "step": 4721 + }, + { + "epoch": 0.04098922752406663, + "grad_norm": 0.310546875, + "learning_rate": 0.001993194871558985, + "loss": 0.1855, + "step": 4722 + }, + { + "epoch": 0.0409979080042708, + "grad_norm": 0.9921875, + "learning_rate": 0.001993191217999075, + "loss": 0.1562, + "step": 4723 + }, + { + "epoch": 0.041006588484474964, + "grad_norm": 0.6015625, + "learning_rate": 0.0019931875634623844, + "loss": 0.1475, + "step": 4724 + }, + { + "epoch": 0.04101526896467913, + "grad_norm": 0.447265625, + "learning_rate": 0.0019931839079489174, + "loss": 0.1504, + "step": 4725 + }, + { + "epoch": 0.041023949444883294, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019931802514586774, + "loss": 0.1196, + "step": 4726 + }, + { + "epoch": 0.04103262992508746, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019931765939916694, + "loss": 0.1475, + "step": 4727 + }, + { + "epoch": 0.041041310405291624, + "grad_norm": 0.404296875, + "learning_rate": 0.001993172935547897, + "loss": 0.1777, + "step": 4728 + }, + { + "epoch": 0.04104999088549578, + "grad_norm": 0.09130859375, + "learning_rate": 0.001993169276127364, + "loss": 0.1387, + "step": 4729 + }, + { + "epoch": 0.04105867136569995, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019931656157300744, + "loss": 0.1455, + "step": 4730 + }, + { + "epoch": 0.04106735184590411, + "grad_norm": 0.3515625, + "learning_rate": 0.0019931619543560324, + "loss": 0.1475, + "step": 4731 + }, + { + "epoch": 0.04107603232610828, + "grad_norm": 0.35546875, + "learning_rate": 0.0019931582920052417, + "loss": 0.1436, + "step": 4732 + }, + { + "epoch": 0.04108471280631244, + "grad_norm": 0.3203125, + "learning_rate": 0.0019931546286777067, + "loss": 0.1475, + "step": 4733 + }, + { + "epoch": 0.04109339328651661, + "grad_norm": 0.40625, + "learning_rate": 0.0019931509643734313, + "loss": 0.1641, + "step": 4734 + }, + { + "epoch": 0.04110207376672077, + "grad_norm": 0.0830078125, + "learning_rate": 0.001993147299092419, + "loss": 0.1367, + "step": 4735 + }, + { + "epoch": 0.04111075424692494, + "grad_norm": 0.09326171875, + "learning_rate": 0.001993143632834675, + "loss": 0.1729, + "step": 4736 + }, + { + "epoch": 0.0411194347271291, + "grad_norm": 0.2265625, + "learning_rate": 0.001993139965600202, + "loss": 0.1875, + "step": 4737 + }, + { + "epoch": 0.04112811520733327, + "grad_norm": 0.34375, + "learning_rate": 0.0019931362973890044, + "loss": 0.1758, + "step": 4738 + }, + { + "epoch": 0.041136795687537434, + "grad_norm": 0.28125, + "learning_rate": 0.0019931326282010865, + "loss": 0.2266, + "step": 4739 + }, + { + "epoch": 0.0411454761677416, + "grad_norm": 0.58984375, + "learning_rate": 0.0019931289580364525, + "loss": 0.1875, + "step": 4740 + }, + { + "epoch": 0.041154156647945764, + "grad_norm": 0.197265625, + "learning_rate": 0.001993125286895106, + "loss": 0.1699, + "step": 4741 + }, + { + "epoch": 0.04116283712814993, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019931216147770505, + "loss": 0.1475, + "step": 4742 + }, + { + "epoch": 0.041171517608354094, + "grad_norm": 0.57421875, + "learning_rate": 0.0019931179416822916, + "loss": 0.1631, + "step": 4743 + }, + { + "epoch": 0.04118019808855826, + "grad_norm": 0.3515625, + "learning_rate": 0.0019931142676108318, + "loss": 0.1689, + "step": 4744 + }, + { + "epoch": 0.041188878568762424, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019931105925626753, + "loss": 0.1543, + "step": 4745 + }, + { + "epoch": 0.04119755904896659, + "grad_norm": 0.244140625, + "learning_rate": 0.0019931069165378267, + "loss": 0.1309, + "step": 4746 + }, + { + "epoch": 0.041206239529170754, + "grad_norm": 1.46875, + "learning_rate": 0.00199310323953629, + "loss": 0.3652, + "step": 4747 + }, + { + "epoch": 0.04121492000937492, + "grad_norm": 0.72265625, + "learning_rate": 0.001993099561558069, + "loss": 0.1533, + "step": 4748 + }, + { + "epoch": 0.041223600489579085, + "grad_norm": 0.46875, + "learning_rate": 0.001993095882603168, + "loss": 0.1562, + "step": 4749 + }, + { + "epoch": 0.04123228096978325, + "grad_norm": 0.0888671875, + "learning_rate": 0.00199309220267159, + "loss": 0.1572, + "step": 4750 + }, + { + "epoch": 0.041240961449987415, + "grad_norm": 0.380859375, + "learning_rate": 0.0019930885217633405, + "loss": 0.1826, + "step": 4751 + }, + { + "epoch": 0.04124964193019158, + "grad_norm": 0.515625, + "learning_rate": 0.001993084839878423, + "loss": 0.1426, + "step": 4752 + }, + { + "epoch": 0.041258322410395745, + "grad_norm": 0.828125, + "learning_rate": 0.0019930811570168403, + "loss": 0.1582, + "step": 4753 + }, + { + "epoch": 0.04126700289059991, + "grad_norm": 0.14453125, + "learning_rate": 0.0019930774731785985, + "loss": 0.1729, + "step": 4754 + }, + { + "epoch": 0.041275683370804075, + "grad_norm": 0.34765625, + "learning_rate": 0.0019930737883637, + "loss": 0.1387, + "step": 4755 + }, + { + "epoch": 0.04128436385100824, + "grad_norm": 1.015625, + "learning_rate": 0.0019930701025721496, + "loss": 0.1572, + "step": 4756 + }, + { + "epoch": 0.041293044331212406, + "grad_norm": 0.0888671875, + "learning_rate": 0.001993066415803951, + "loss": 0.1533, + "step": 4757 + }, + { + "epoch": 0.04130172481141657, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019930627280591085, + "loss": 0.1934, + "step": 4758 + }, + { + "epoch": 0.04131040529162073, + "grad_norm": 0.09423828125, + "learning_rate": 0.001993059039337626, + "loss": 0.1826, + "step": 4759 + }, + { + "epoch": 0.041319085771824894, + "grad_norm": 0.173828125, + "learning_rate": 0.001993055349639508, + "loss": 0.1602, + "step": 4760 + }, + { + "epoch": 0.04132776625202906, + "grad_norm": 0.490234375, + "learning_rate": 0.0019930516589647574, + "loss": 0.1719, + "step": 4761 + }, + { + "epoch": 0.041336446732233224, + "grad_norm": 0.162109375, + "learning_rate": 0.001993047967313379, + "loss": 0.127, + "step": 4762 + }, + { + "epoch": 0.04134512721243739, + "grad_norm": 0.1943359375, + "learning_rate": 0.001993044274685377, + "loss": 0.1396, + "step": 4763 + }, + { + "epoch": 0.041353807692641555, + "grad_norm": 0.310546875, + "learning_rate": 0.0019930405810807553, + "loss": 0.1504, + "step": 4764 + }, + { + "epoch": 0.04136248817284572, + "grad_norm": 0.58984375, + "learning_rate": 0.001993036886499518, + "loss": 0.1348, + "step": 4765 + }, + { + "epoch": 0.041371168653049885, + "grad_norm": 0.259765625, + "learning_rate": 0.0019930331909416682, + "loss": 0.1611, + "step": 4766 + }, + { + "epoch": 0.04137984913325405, + "grad_norm": 0.453125, + "learning_rate": 0.0019930294944072117, + "loss": 0.1572, + "step": 4767 + }, + { + "epoch": 0.041388529613458215, + "grad_norm": 0.26171875, + "learning_rate": 0.001993025796896151, + "loss": 0.1611, + "step": 4768 + }, + { + "epoch": 0.04139721009366238, + "grad_norm": 0.3828125, + "learning_rate": 0.0019930220984084902, + "loss": 0.1465, + "step": 4769 + }, + { + "epoch": 0.041405890573866545, + "grad_norm": 0.7421875, + "learning_rate": 0.001993018398944235, + "loss": 0.1602, + "step": 4770 + }, + { + "epoch": 0.04141457105407071, + "grad_norm": 0.6953125, + "learning_rate": 0.0019930146985033875, + "loss": 0.1484, + "step": 4771 + }, + { + "epoch": 0.041423251534274876, + "grad_norm": 0.244140625, + "learning_rate": 0.001993010997085953, + "loss": 0.1436, + "step": 4772 + }, + { + "epoch": 0.04143193201447904, + "grad_norm": 0.3359375, + "learning_rate": 0.0019930072946919347, + "loss": 0.1377, + "step": 4773 + }, + { + "epoch": 0.041440612494683206, + "grad_norm": 0.3671875, + "learning_rate": 0.001993003591321337, + "loss": 0.1553, + "step": 4774 + }, + { + "epoch": 0.04144929297488737, + "grad_norm": 0.484375, + "learning_rate": 0.0019929998869741643, + "loss": 0.1572, + "step": 4775 + }, + { + "epoch": 0.041457973455091536, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019929961816504203, + "loss": 0.1719, + "step": 4776 + }, + { + "epoch": 0.0414666539352957, + "grad_norm": 0.46875, + "learning_rate": 0.001992992475350109, + "loss": 0.1367, + "step": 4777 + }, + { + "epoch": 0.041475334415499866, + "grad_norm": 0.26953125, + "learning_rate": 0.0019929887680732346, + "loss": 0.1426, + "step": 4778 + }, + { + "epoch": 0.04148401489570403, + "grad_norm": 0.294921875, + "learning_rate": 0.001992985059819801, + "loss": 0.1338, + "step": 4779 + }, + { + "epoch": 0.041492695375908197, + "grad_norm": 0.23828125, + "learning_rate": 0.0019929813505898124, + "loss": 0.1475, + "step": 4780 + }, + { + "epoch": 0.04150137585611236, + "grad_norm": 0.376953125, + "learning_rate": 0.001992977640383273, + "loss": 0.1465, + "step": 4781 + }, + { + "epoch": 0.04151005633631653, + "grad_norm": 0.17578125, + "learning_rate": 0.0019929739292001863, + "loss": 0.1777, + "step": 4782 + }, + { + "epoch": 0.04151873681652069, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019929702170405567, + "loss": 0.1318, + "step": 4783 + }, + { + "epoch": 0.04152741729672486, + "grad_norm": 0.08837890625, + "learning_rate": 0.001992966503904389, + "loss": 0.1504, + "step": 4784 + }, + { + "epoch": 0.04153609777692902, + "grad_norm": 0.2265625, + "learning_rate": 0.0019929627897916856, + "loss": 0.1855, + "step": 4785 + }, + { + "epoch": 0.04154477825713319, + "grad_norm": 0.244140625, + "learning_rate": 0.001992959074702452, + "loss": 0.1211, + "step": 4786 + }, + { + "epoch": 0.04155345873733735, + "grad_norm": 0.435546875, + "learning_rate": 0.0019929553586366918, + "loss": 0.1807, + "step": 4787 + }, + { + "epoch": 0.04156213921754152, + "grad_norm": 0.12890625, + "learning_rate": 0.0019929516415944093, + "loss": 0.209, + "step": 4788 + }, + { + "epoch": 0.04157081969774568, + "grad_norm": 0.29296875, + "learning_rate": 0.001992947923575608, + "loss": 0.1523, + "step": 4789 + }, + { + "epoch": 0.04157950017794984, + "grad_norm": 0.36328125, + "learning_rate": 0.0019929442045802923, + "loss": 0.1504, + "step": 4790 + }, + { + "epoch": 0.041588180658154006, + "grad_norm": 0.1845703125, + "learning_rate": 0.001992940484608466, + "loss": 0.1182, + "step": 4791 + }, + { + "epoch": 0.04159686113835817, + "grad_norm": 0.435546875, + "learning_rate": 0.0019929367636601332, + "loss": 0.2051, + "step": 4792 + }, + { + "epoch": 0.041605541618562336, + "grad_norm": 0.169921875, + "learning_rate": 0.001992933041735299, + "loss": 0.1167, + "step": 4793 + }, + { + "epoch": 0.0416142220987665, + "grad_norm": 0.158203125, + "learning_rate": 0.0019929293188339662, + "loss": 0.1338, + "step": 4794 + }, + { + "epoch": 0.041622902578970666, + "grad_norm": 0.125, + "learning_rate": 0.001992925594956139, + "loss": 0.1445, + "step": 4795 + }, + { + "epoch": 0.04163158305917483, + "grad_norm": 0.1884765625, + "learning_rate": 0.001992921870101822, + "loss": 0.1221, + "step": 4796 + }, + { + "epoch": 0.041640263539379, + "grad_norm": 0.52734375, + "learning_rate": 0.0019929181442710194, + "loss": 0.1758, + "step": 4797 + }, + { + "epoch": 0.04164894401958316, + "grad_norm": 0.39453125, + "learning_rate": 0.0019929144174637347, + "loss": 0.1758, + "step": 4798 + }, + { + "epoch": 0.04165762449978733, + "grad_norm": 0.1298828125, + "learning_rate": 0.001992910689679972, + "loss": 0.1436, + "step": 4799 + }, + { + "epoch": 0.04166630497999149, + "grad_norm": 0.330078125, + "learning_rate": 0.0019929069609197357, + "loss": 0.1855, + "step": 4800 + }, + { + "epoch": 0.04167498546019566, + "grad_norm": 0.12890625, + "learning_rate": 0.00199290323118303, + "loss": 0.1196, + "step": 4801 + }, + { + "epoch": 0.04168366594039982, + "grad_norm": 0.375, + "learning_rate": 0.0019928995004698585, + "loss": 0.1826, + "step": 4802 + }, + { + "epoch": 0.04169234642060399, + "grad_norm": 0.138671875, + "learning_rate": 0.001992895768780226, + "loss": 0.1387, + "step": 4803 + }, + { + "epoch": 0.04170102690080815, + "grad_norm": 0.453125, + "learning_rate": 0.0019928920361141356, + "loss": 0.1514, + "step": 4804 + }, + { + "epoch": 0.04170970738101232, + "grad_norm": 0.1123046875, + "learning_rate": 0.001992888302471592, + "loss": 0.1055, + "step": 4805 + }, + { + "epoch": 0.04171838786121648, + "grad_norm": 0.59765625, + "learning_rate": 0.001992884567852599, + "loss": 0.1543, + "step": 4806 + }, + { + "epoch": 0.04172706834142065, + "grad_norm": 0.1875, + "learning_rate": 0.001992880832257161, + "loss": 0.1826, + "step": 4807 + }, + { + "epoch": 0.04173574882162481, + "grad_norm": 0.32421875, + "learning_rate": 0.001992877095685282, + "loss": 0.1699, + "step": 4808 + }, + { + "epoch": 0.04174442930182898, + "grad_norm": 0.1123046875, + "learning_rate": 0.001992873358136966, + "loss": 0.126, + "step": 4809 + }, + { + "epoch": 0.04175310978203314, + "grad_norm": 0.96875, + "learning_rate": 0.001992869619612217, + "loss": 0.1582, + "step": 4810 + }, + { + "epoch": 0.04176179026223731, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019928658801110395, + "loss": 0.1855, + "step": 4811 + }, + { + "epoch": 0.04177047074244147, + "grad_norm": 0.4140625, + "learning_rate": 0.001992862139633437, + "loss": 0.1465, + "step": 4812 + }, + { + "epoch": 0.04177915122264564, + "grad_norm": 0.33984375, + "learning_rate": 0.001992858398179414, + "loss": 0.1719, + "step": 4813 + }, + { + "epoch": 0.041787831702849804, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019928546557489743, + "loss": 0.1484, + "step": 4814 + }, + { + "epoch": 0.04179651218305397, + "grad_norm": 2.34375, + "learning_rate": 0.0019928509123421224, + "loss": 0.3672, + "step": 4815 + }, + { + "epoch": 0.041805192663258134, + "grad_norm": 0.09326171875, + "learning_rate": 0.001992847167958862, + "loss": 0.1436, + "step": 4816 + }, + { + "epoch": 0.0418138731434623, + "grad_norm": 0.51171875, + "learning_rate": 0.0019928434225991976, + "loss": 0.166, + "step": 4817 + }, + { + "epoch": 0.041822553623666464, + "grad_norm": 0.1708984375, + "learning_rate": 0.001992839676263133, + "loss": 0.2002, + "step": 4818 + }, + { + "epoch": 0.04183123410387063, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019928359289506717, + "loss": 0.126, + "step": 4819 + }, + { + "epoch": 0.041839914584074794, + "grad_norm": 0.404296875, + "learning_rate": 0.001992832180661819, + "loss": 0.1523, + "step": 4820 + }, + { + "epoch": 0.04184859506427895, + "grad_norm": 0.83984375, + "learning_rate": 0.0019928284313965785, + "loss": 0.1895, + "step": 4821 + }, + { + "epoch": 0.04185727554448312, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019928246811549543, + "loss": 0.1084, + "step": 4822 + }, + { + "epoch": 0.04186595602468728, + "grad_norm": 0.43359375, + "learning_rate": 0.00199282092993695, + "loss": 0.1992, + "step": 4823 + }, + { + "epoch": 0.04187463650489145, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019928171777425703, + "loss": 0.1973, + "step": 4824 + }, + { + "epoch": 0.04188331698509561, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019928134245718195, + "loss": 0.1582, + "step": 4825 + }, + { + "epoch": 0.04189199746529978, + "grad_norm": 0.251953125, + "learning_rate": 0.001992809670424701, + "loss": 0.1377, + "step": 4826 + }, + { + "epoch": 0.04190067794550394, + "grad_norm": 0.361328125, + "learning_rate": 0.001992805915301219, + "loss": 0.1367, + "step": 4827 + }, + { + "epoch": 0.04190935842570811, + "grad_norm": 0.54296875, + "learning_rate": 0.0019928021592013783, + "loss": 0.2891, + "step": 4828 + }, + { + "epoch": 0.041918038905912273, + "grad_norm": 0.40625, + "learning_rate": 0.001992798402125183, + "loss": 0.1641, + "step": 4829 + }, + { + "epoch": 0.04192671938611644, + "grad_norm": 0.45703125, + "learning_rate": 0.001992794644072636, + "loss": 0.1738, + "step": 4830 + }, + { + "epoch": 0.041935399866320604, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019927908850437426, + "loss": 0.21, + "step": 4831 + }, + { + "epoch": 0.04194408034652477, + "grad_norm": 0.318359375, + "learning_rate": 0.0019927871250385062, + "loss": 0.1602, + "step": 4832 + }, + { + "epoch": 0.041952760826728934, + "grad_norm": 0.189453125, + "learning_rate": 0.001992783364056932, + "loss": 0.2168, + "step": 4833 + }, + { + "epoch": 0.0419614413069331, + "grad_norm": 0.205078125, + "learning_rate": 0.001992779602099022, + "loss": 0.168, + "step": 4834 + }, + { + "epoch": 0.041970121787137264, + "grad_norm": 0.1484375, + "learning_rate": 0.001992775839164783, + "loss": 0.1172, + "step": 4835 + }, + { + "epoch": 0.04197880226734143, + "grad_norm": 0.201171875, + "learning_rate": 0.0019927720752542168, + "loss": 0.1289, + "step": 4836 + }, + { + "epoch": 0.041987482747545594, + "grad_norm": 0.61328125, + "learning_rate": 0.001992768310367329, + "loss": 0.168, + "step": 4837 + }, + { + "epoch": 0.04199616322774976, + "grad_norm": 0.375, + "learning_rate": 0.0019927645445041225, + "loss": 0.2285, + "step": 4838 + }, + { + "epoch": 0.042004843707953925, + "grad_norm": 0.337890625, + "learning_rate": 0.001992760777664603, + "loss": 0.2041, + "step": 4839 + }, + { + "epoch": 0.04201352418815809, + "grad_norm": 0.146484375, + "learning_rate": 0.0019927570098487732, + "loss": 0.1572, + "step": 4840 + }, + { + "epoch": 0.042022204668362255, + "grad_norm": 0.1455078125, + "learning_rate": 0.001992753241056638, + "loss": 0.1309, + "step": 4841 + }, + { + "epoch": 0.04203088514856642, + "grad_norm": 0.1318359375, + "learning_rate": 0.001992749471288201, + "loss": 0.1543, + "step": 4842 + }, + { + "epoch": 0.042039565628770585, + "grad_norm": 0.119140625, + "learning_rate": 0.0019927457005434667, + "loss": 0.1445, + "step": 4843 + }, + { + "epoch": 0.04204824610897475, + "grad_norm": 0.63671875, + "learning_rate": 0.0019927419288224392, + "loss": 0.1855, + "step": 4844 + }, + { + "epoch": 0.042056926589178915, + "grad_norm": 0.318359375, + "learning_rate": 0.0019927381561251224, + "loss": 0.2227, + "step": 4845 + }, + { + "epoch": 0.04206560706938308, + "grad_norm": 0.337890625, + "learning_rate": 0.001992734382451521, + "loss": 0.1406, + "step": 4846 + }, + { + "epoch": 0.042074287549587246, + "grad_norm": 0.56640625, + "learning_rate": 0.0019927306078016383, + "loss": 0.2188, + "step": 4847 + }, + { + "epoch": 0.04208296802979141, + "grad_norm": 0.416015625, + "learning_rate": 0.0019927268321754785, + "loss": 0.1523, + "step": 4848 + }, + { + "epoch": 0.042091648509995576, + "grad_norm": 0.341796875, + "learning_rate": 0.0019927230555730467, + "loss": 0.1338, + "step": 4849 + }, + { + "epoch": 0.04210032899019974, + "grad_norm": 0.07470703125, + "learning_rate": 0.001992719277994346, + "loss": 0.1455, + "step": 4850 + }, + { + "epoch": 0.042109009470403906, + "grad_norm": 0.45703125, + "learning_rate": 0.001992715499439381, + "loss": 0.1826, + "step": 4851 + }, + { + "epoch": 0.042117689950608064, + "grad_norm": 0.455078125, + "learning_rate": 0.0019927117199081555, + "loss": 0.2012, + "step": 4852 + }, + { + "epoch": 0.04212637043081223, + "grad_norm": 0.251953125, + "learning_rate": 0.0019927079394006742, + "loss": 0.1602, + "step": 4853 + }, + { + "epoch": 0.042135050911016395, + "grad_norm": 0.13671875, + "learning_rate": 0.001992704157916941, + "loss": 0.1943, + "step": 4854 + }, + { + "epoch": 0.04214373139122056, + "grad_norm": 0.671875, + "learning_rate": 0.00199270037545696, + "loss": 0.1875, + "step": 4855 + }, + { + "epoch": 0.042152411871424725, + "grad_norm": 0.076171875, + "learning_rate": 0.001992696592020735, + "loss": 0.1592, + "step": 4856 + }, + { + "epoch": 0.04216109235162889, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019926928076082705, + "loss": 0.1855, + "step": 4857 + }, + { + "epoch": 0.042169772831833055, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019926890222195705, + "loss": 0.1709, + "step": 4858 + }, + { + "epoch": 0.04217845331203722, + "grad_norm": 0.11962890625, + "learning_rate": 0.001992685235854639, + "loss": 0.168, + "step": 4859 + }, + { + "epoch": 0.042187133792241385, + "grad_norm": 0.25, + "learning_rate": 0.0019926814485134807, + "loss": 0.1396, + "step": 4860 + }, + { + "epoch": 0.04219581427244555, + "grad_norm": 0.53515625, + "learning_rate": 0.001992677660196099, + "loss": 0.1719, + "step": 4861 + }, + { + "epoch": 0.042204494752649716, + "grad_norm": 0.251953125, + "learning_rate": 0.001992673870902499, + "loss": 0.1592, + "step": 4862 + }, + { + "epoch": 0.04221317523285388, + "grad_norm": 0.158203125, + "learning_rate": 0.0019926700806326835, + "loss": 0.168, + "step": 4863 + }, + { + "epoch": 0.042221855713058046, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019926662893866584, + "loss": 0.1328, + "step": 4864 + }, + { + "epoch": 0.04223053619326221, + "grad_norm": 0.1982421875, + "learning_rate": 0.001992662497164426, + "loss": 0.1406, + "step": 4865 + }, + { + "epoch": 0.042239216673466376, + "grad_norm": 0.4140625, + "learning_rate": 0.001992658703965992, + "loss": 0.1914, + "step": 4866 + }, + { + "epoch": 0.04224789715367054, + "grad_norm": 0.1455078125, + "learning_rate": 0.001992654909791359, + "loss": 0.1543, + "step": 4867 + }, + { + "epoch": 0.042256577633874706, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019926511146405325, + "loss": 0.1709, + "step": 4868 + }, + { + "epoch": 0.04226525811407887, + "grad_norm": 0.498046875, + "learning_rate": 0.001992647318513516, + "loss": 0.1621, + "step": 4869 + }, + { + "epoch": 0.042273938594283036, + "grad_norm": 0.12109375, + "learning_rate": 0.0019926435214103143, + "loss": 0.127, + "step": 4870 + }, + { + "epoch": 0.0422826190744872, + "grad_norm": 0.328125, + "learning_rate": 0.001992639723330931, + "loss": 0.1299, + "step": 4871 + }, + { + "epoch": 0.04229129955469137, + "grad_norm": 0.546875, + "learning_rate": 0.0019926359242753698, + "loss": 0.1934, + "step": 4872 + }, + { + "epoch": 0.04229998003489553, + "grad_norm": 0.126953125, + "learning_rate": 0.001992632124243635, + "loss": 0.168, + "step": 4873 + }, + { + "epoch": 0.0423086605150997, + "grad_norm": 0.3984375, + "learning_rate": 0.0019926283232357318, + "loss": 0.1807, + "step": 4874 + }, + { + "epoch": 0.04231734099530386, + "grad_norm": 0.1640625, + "learning_rate": 0.001992624521251664, + "loss": 0.1504, + "step": 4875 + }, + { + "epoch": 0.04232602147550803, + "grad_norm": 0.17578125, + "learning_rate": 0.001992620718291435, + "loss": 0.1875, + "step": 4876 + }, + { + "epoch": 0.04233470195571219, + "grad_norm": 0.5546875, + "learning_rate": 0.001992616914355049, + "loss": 0.1245, + "step": 4877 + }, + { + "epoch": 0.04234338243591636, + "grad_norm": 0.212890625, + "learning_rate": 0.001992613109442511, + "loss": 0.168, + "step": 4878 + }, + { + "epoch": 0.04235206291612052, + "grad_norm": 0.46875, + "learning_rate": 0.0019926093035538247, + "loss": 0.2168, + "step": 4879 + }, + { + "epoch": 0.04236074339632469, + "grad_norm": 0.318359375, + "learning_rate": 0.0019926054966889943, + "loss": 0.166, + "step": 4880 + }, + { + "epoch": 0.04236942387652885, + "grad_norm": 0.1845703125, + "learning_rate": 0.001992601688848024, + "loss": 0.1934, + "step": 4881 + }, + { + "epoch": 0.04237810435673302, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019925978800309175, + "loss": 0.1582, + "step": 4882 + }, + { + "epoch": 0.042386784836937176, + "grad_norm": 0.52734375, + "learning_rate": 0.0019925940702376797, + "loss": 0.168, + "step": 4883 + }, + { + "epoch": 0.04239546531714134, + "grad_norm": 0.16015625, + "learning_rate": 0.0019925902594683147, + "loss": 0.166, + "step": 4884 + }, + { + "epoch": 0.042404145797345506, + "grad_norm": 0.5390625, + "learning_rate": 0.001992586447722826, + "loss": 0.1338, + "step": 4885 + }, + { + "epoch": 0.04241282627754967, + "grad_norm": 0.478515625, + "learning_rate": 0.001992582635001218, + "loss": 0.2695, + "step": 4886 + }, + { + "epoch": 0.04242150675775384, + "grad_norm": 0.32421875, + "learning_rate": 0.0019925788213034953, + "loss": 0.1543, + "step": 4887 + }, + { + "epoch": 0.042430187237958, + "grad_norm": 0.232421875, + "learning_rate": 0.001992575006629662, + "loss": 0.1348, + "step": 4888 + }, + { + "epoch": 0.04243886771816217, + "grad_norm": 0.158203125, + "learning_rate": 0.001992571190979722, + "loss": 0.1689, + "step": 4889 + }, + { + "epoch": 0.04244754819836633, + "grad_norm": 2.359375, + "learning_rate": 0.0019925673743536793, + "loss": 0.3301, + "step": 4890 + }, + { + "epoch": 0.0424562286785705, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019925635567515387, + "loss": 0.165, + "step": 4891 + }, + { + "epoch": 0.04246490915877466, + "grad_norm": 0.302734375, + "learning_rate": 0.001992559738173304, + "loss": 0.1348, + "step": 4892 + }, + { + "epoch": 0.04247358963897883, + "grad_norm": 0.474609375, + "learning_rate": 0.001992555918618979, + "loss": 0.124, + "step": 4893 + }, + { + "epoch": 0.04248227011918299, + "grad_norm": 0.53125, + "learning_rate": 0.0019925520980885684, + "loss": 0.1787, + "step": 4894 + }, + { + "epoch": 0.04249095059938716, + "grad_norm": 0.380859375, + "learning_rate": 0.001992548276582076, + "loss": 0.1387, + "step": 4895 + }, + { + "epoch": 0.04249963107959132, + "grad_norm": 0.11767578125, + "learning_rate": 0.001992544454099507, + "loss": 0.1631, + "step": 4896 + }, + { + "epoch": 0.04250831155979549, + "grad_norm": 0.291015625, + "learning_rate": 0.001992540630640864, + "loss": 0.1807, + "step": 4897 + }, + { + "epoch": 0.04251699203999965, + "grad_norm": 0.2265625, + "learning_rate": 0.0019925368062061522, + "loss": 0.1318, + "step": 4898 + }, + { + "epoch": 0.04252567252020382, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019925329807953755, + "loss": 0.1523, + "step": 4899 + }, + { + "epoch": 0.04253435300040798, + "grad_norm": 1.03125, + "learning_rate": 0.0019925291544085383, + "loss": 0.1748, + "step": 4900 + }, + { + "epoch": 0.04254303348061215, + "grad_norm": 0.189453125, + "learning_rate": 0.0019925253270456447, + "loss": 0.1465, + "step": 4901 + }, + { + "epoch": 0.04255171396081631, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019925214987066985, + "loss": 0.2324, + "step": 4902 + }, + { + "epoch": 0.04256039444102048, + "grad_norm": 0.73828125, + "learning_rate": 0.0019925176693917045, + "loss": 0.2012, + "step": 4903 + }, + { + "epoch": 0.042569074921224644, + "grad_norm": 0.205078125, + "learning_rate": 0.0019925138391006666, + "loss": 0.1914, + "step": 4904 + }, + { + "epoch": 0.04257775540142881, + "grad_norm": 0.08984375, + "learning_rate": 0.0019925100078335887, + "loss": 0.1641, + "step": 4905 + }, + { + "epoch": 0.042586435881632974, + "grad_norm": 0.205078125, + "learning_rate": 0.0019925061755904755, + "loss": 0.1309, + "step": 4906 + }, + { + "epoch": 0.04259511636183714, + "grad_norm": 0.39453125, + "learning_rate": 0.0019925023423713307, + "loss": 0.1328, + "step": 4907 + }, + { + "epoch": 0.042603796842041304, + "grad_norm": 1.140625, + "learning_rate": 0.0019924985081761592, + "loss": 0.165, + "step": 4908 + }, + { + "epoch": 0.04261247732224547, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019924946730049643, + "loss": 0.1221, + "step": 4909 + }, + { + "epoch": 0.042621157802449634, + "grad_norm": 0.201171875, + "learning_rate": 0.0019924908368577506, + "loss": 0.1768, + "step": 4910 + }, + { + "epoch": 0.0426298382826538, + "grad_norm": 0.61328125, + "learning_rate": 0.001992486999734523, + "loss": 0.1895, + "step": 4911 + }, + { + "epoch": 0.042638518762857965, + "grad_norm": 0.328125, + "learning_rate": 0.0019924831616352843, + "loss": 0.1436, + "step": 4912 + }, + { + "epoch": 0.04264719924306212, + "grad_norm": 0.130859375, + "learning_rate": 0.0019924793225600396, + "loss": 0.1475, + "step": 4913 + }, + { + "epoch": 0.04265587972326629, + "grad_norm": 0.328125, + "learning_rate": 0.001992475482508793, + "loss": 0.1641, + "step": 4914 + }, + { + "epoch": 0.04266456020347045, + "grad_norm": 0.421875, + "learning_rate": 0.001992471641481549, + "loss": 0.1699, + "step": 4915 + }, + { + "epoch": 0.04267324068367462, + "grad_norm": 0.7421875, + "learning_rate": 0.0019924677994783107, + "loss": 0.1523, + "step": 4916 + }, + { + "epoch": 0.04268192116387878, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019924639564990834, + "loss": 0.1533, + "step": 4917 + }, + { + "epoch": 0.04269060164408295, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019924601125438706, + "loss": 0.1709, + "step": 4918 + }, + { + "epoch": 0.04269928212428711, + "grad_norm": 0.138671875, + "learning_rate": 0.0019924562676126773, + "loss": 0.1348, + "step": 4919 + }, + { + "epoch": 0.04270796260449128, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019924524217055073, + "loss": 0.1719, + "step": 4920 + }, + { + "epoch": 0.042716643084695444, + "grad_norm": 0.74609375, + "learning_rate": 0.0019924485748223646, + "loss": 0.1504, + "step": 4921 + }, + { + "epoch": 0.04272532356489961, + "grad_norm": 0.451171875, + "learning_rate": 0.0019924447269632534, + "loss": 0.1089, + "step": 4922 + }, + { + "epoch": 0.042734004045103774, + "grad_norm": 0.0703125, + "learning_rate": 0.001992440878128178, + "loss": 0.1157, + "step": 4923 + }, + { + "epoch": 0.04274268452530794, + "grad_norm": 0.462890625, + "learning_rate": 0.0019924370283171426, + "loss": 0.1309, + "step": 4924 + }, + { + "epoch": 0.042751365005512104, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019924331775301517, + "loss": 0.1191, + "step": 4925 + }, + { + "epoch": 0.04276004548571627, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019924293257672096, + "loss": 0.2002, + "step": 4926 + }, + { + "epoch": 0.042768725965920434, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019924254730283196, + "loss": 0.1689, + "step": 4927 + }, + { + "epoch": 0.0427774064461246, + "grad_norm": 0.296875, + "learning_rate": 0.0019924216193134866, + "loss": 0.2051, + "step": 4928 + }, + { + "epoch": 0.042786086926328765, + "grad_norm": 0.10205078125, + "learning_rate": 0.001992417764622715, + "loss": 0.1465, + "step": 4929 + }, + { + "epoch": 0.04279476740653293, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019924139089560087, + "loss": 0.1201, + "step": 4930 + }, + { + "epoch": 0.042803447886737095, + "grad_norm": 0.205078125, + "learning_rate": 0.001992410052313372, + "loss": 0.1797, + "step": 4931 + }, + { + "epoch": 0.04281212836694126, + "grad_norm": 0.5, + "learning_rate": 0.001992406194694809, + "loss": 0.1045, + "step": 4932 + }, + { + "epoch": 0.042820808847145425, + "grad_norm": 0.388671875, + "learning_rate": 0.0019924023361003237, + "loss": 0.1206, + "step": 4933 + }, + { + "epoch": 0.04282948932734959, + "grad_norm": 0.10595703125, + "learning_rate": 0.001992398476529921, + "loss": 0.1357, + "step": 4934 + }, + { + "epoch": 0.042838169807553755, + "grad_norm": 0.35546875, + "learning_rate": 0.0019923946159836046, + "loss": 0.1416, + "step": 4935 + }, + { + "epoch": 0.04284685028775792, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019923907544613785, + "loss": 0.168, + "step": 4936 + }, + { + "epoch": 0.042855530767962086, + "grad_norm": 0.5390625, + "learning_rate": 0.0019923868919632477, + "loss": 0.1885, + "step": 4937 + }, + { + "epoch": 0.04286421124816625, + "grad_norm": 0.703125, + "learning_rate": 0.001992383028489216, + "loss": 0.1709, + "step": 4938 + }, + { + "epoch": 0.042872891728370416, + "grad_norm": 0.2216796875, + "learning_rate": 0.001992379164039287, + "loss": 0.1348, + "step": 4939 + }, + { + "epoch": 0.04288157220857458, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019923752986134666, + "loss": 0.1621, + "step": 4940 + }, + { + "epoch": 0.042890252688778746, + "grad_norm": 0.099609375, + "learning_rate": 0.001992371432211758, + "loss": 0.1787, + "step": 4941 + }, + { + "epoch": 0.04289893316898291, + "grad_norm": 0.3984375, + "learning_rate": 0.0019923675648341643, + "loss": 0.1396, + "step": 4942 + }, + { + "epoch": 0.042907613649187076, + "grad_norm": 0.4375, + "learning_rate": 0.0019923636964806913, + "loss": 0.1416, + "step": 4943 + }, + { + "epoch": 0.042916294129391235, + "grad_norm": 0.220703125, + "learning_rate": 0.001992359827151343, + "loss": 0.1797, + "step": 4944 + }, + { + "epoch": 0.0429249746095954, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019923559568461235, + "loss": 0.1621, + "step": 4945 + }, + { + "epoch": 0.042933655089799565, + "grad_norm": 0.515625, + "learning_rate": 0.0019923520855650366, + "loss": 0.1484, + "step": 4946 + }, + { + "epoch": 0.04294233557000373, + "grad_norm": 0.75, + "learning_rate": 0.001992348213308087, + "loss": 0.1816, + "step": 4947 + }, + { + "epoch": 0.042951016050207895, + "grad_norm": 0.294921875, + "learning_rate": 0.001992344340075279, + "loss": 0.1328, + "step": 4948 + }, + { + "epoch": 0.04295969653041206, + "grad_norm": 0.123046875, + "learning_rate": 0.001992340465866616, + "loss": 0.1357, + "step": 4949 + }, + { + "epoch": 0.042968377010616225, + "grad_norm": 0.095703125, + "learning_rate": 0.001992336590682104, + "loss": 0.1641, + "step": 4950 + }, + { + "epoch": 0.04297705749082039, + "grad_norm": 0.333984375, + "learning_rate": 0.001992332714521745, + "loss": 0.1348, + "step": 4951 + }, + { + "epoch": 0.042985737971024555, + "grad_norm": 0.25390625, + "learning_rate": 0.001992328837385545, + "loss": 0.1719, + "step": 4952 + }, + { + "epoch": 0.04299441845122872, + "grad_norm": 0.30859375, + "learning_rate": 0.001992324959273507, + "loss": 0.1914, + "step": 4953 + }, + { + "epoch": 0.043003098931432886, + "grad_norm": 0.19921875, + "learning_rate": 0.0019923210801856364, + "loss": 0.1553, + "step": 4954 + }, + { + "epoch": 0.04301177941163705, + "grad_norm": 0.609375, + "learning_rate": 0.0019923172001219368, + "loss": 0.1768, + "step": 4955 + }, + { + "epoch": 0.043020459891841216, + "grad_norm": 0.3828125, + "learning_rate": 0.0019923133190824123, + "loss": 0.1953, + "step": 4956 + }, + { + "epoch": 0.04302914037204538, + "grad_norm": 0.40234375, + "learning_rate": 0.0019923094370670673, + "loss": 0.1074, + "step": 4957 + }, + { + "epoch": 0.043037820852249546, + "grad_norm": 0.115234375, + "learning_rate": 0.0019923055540759065, + "loss": 0.1689, + "step": 4958 + }, + { + "epoch": 0.04304650133245371, + "grad_norm": 0.30078125, + "learning_rate": 0.0019923016701089338, + "loss": 0.1758, + "step": 4959 + }, + { + "epoch": 0.043055181812657876, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019922977851661527, + "loss": 0.1514, + "step": 4960 + }, + { + "epoch": 0.04306386229286204, + "grad_norm": 1.203125, + "learning_rate": 0.001992293899247569, + "loss": 0.1445, + "step": 4961 + }, + { + "epoch": 0.04307254277306621, + "grad_norm": 0.8359375, + "learning_rate": 0.001992290012353185, + "loss": 0.2559, + "step": 4962 + }, + { + "epoch": 0.04308122325327037, + "grad_norm": 0.2578125, + "learning_rate": 0.001992286124483007, + "loss": 0.166, + "step": 4963 + }, + { + "epoch": 0.04308990373347454, + "grad_norm": 0.236328125, + "learning_rate": 0.001992282235637038, + "loss": 0.1562, + "step": 4964 + }, + { + "epoch": 0.0430985842136787, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019922783458152828, + "loss": 0.1641, + "step": 4965 + }, + { + "epoch": 0.04310726469388287, + "grad_norm": 0.1953125, + "learning_rate": 0.001992274455017745, + "loss": 0.2217, + "step": 4966 + }, + { + "epoch": 0.04311594517408703, + "grad_norm": 0.28125, + "learning_rate": 0.0019922705632444294, + "loss": 0.1177, + "step": 4967 + }, + { + "epoch": 0.0431246256542912, + "grad_norm": 0.48828125, + "learning_rate": 0.00199226667049534, + "loss": 0.168, + "step": 4968 + }, + { + "epoch": 0.04313330613449536, + "grad_norm": 0.484375, + "learning_rate": 0.0019922627767704816, + "loss": 0.1543, + "step": 4969 + }, + { + "epoch": 0.04314198661469953, + "grad_norm": 0.228515625, + "learning_rate": 0.001992258882069858, + "loss": 0.124, + "step": 4970 + }, + { + "epoch": 0.04315066709490369, + "grad_norm": 0.703125, + "learning_rate": 0.0019922549863934727, + "loss": 0.1348, + "step": 4971 + }, + { + "epoch": 0.04315934757510786, + "grad_norm": 0.640625, + "learning_rate": 0.0019922510897413316, + "loss": 0.166, + "step": 4972 + }, + { + "epoch": 0.04316802805531202, + "grad_norm": 0.41796875, + "learning_rate": 0.0019922471921134375, + "loss": 0.1953, + "step": 4973 + }, + { + "epoch": 0.04317670853551619, + "grad_norm": 0.3671875, + "learning_rate": 0.0019922432935097958, + "loss": 0.1348, + "step": 4974 + }, + { + "epoch": 0.043185389015720346, + "grad_norm": 0.404296875, + "learning_rate": 0.00199223939393041, + "loss": 0.1611, + "step": 4975 + }, + { + "epoch": 0.04319406949592451, + "grad_norm": 1.0625, + "learning_rate": 0.001992235493375285, + "loss": 0.1572, + "step": 4976 + }, + { + "epoch": 0.043202749976128677, + "grad_norm": 0.69921875, + "learning_rate": 0.001992231591844424, + "loss": 0.1445, + "step": 4977 + }, + { + "epoch": 0.04321143045633284, + "grad_norm": 0.494140625, + "learning_rate": 0.0019922276893378325, + "loss": 0.1523, + "step": 4978 + }, + { + "epoch": 0.04322011093653701, + "grad_norm": 0.69140625, + "learning_rate": 0.0019922237858555135, + "loss": 0.1377, + "step": 4979 + }, + { + "epoch": 0.04322879141674117, + "grad_norm": 0.251953125, + "learning_rate": 0.0019922198813974723, + "loss": 0.1611, + "step": 4980 + }, + { + "epoch": 0.04323747189694534, + "grad_norm": 0.71484375, + "learning_rate": 0.0019922159759637134, + "loss": 0.1953, + "step": 4981 + }, + { + "epoch": 0.0432461523771495, + "grad_norm": 0.33203125, + "learning_rate": 0.0019922120695542397, + "loss": 0.1465, + "step": 4982 + }, + { + "epoch": 0.04325483285735367, + "grad_norm": 0.546875, + "learning_rate": 0.001992208162169057, + "loss": 0.1602, + "step": 4983 + }, + { + "epoch": 0.04326351333755783, + "grad_norm": 0.84765625, + "learning_rate": 0.0019922042538081685, + "loss": 0.1914, + "step": 4984 + }, + { + "epoch": 0.043272193817762, + "grad_norm": 0.10888671875, + "learning_rate": 0.001992200344471579, + "loss": 0.1816, + "step": 4985 + }, + { + "epoch": 0.04328087429796616, + "grad_norm": 0.2294921875, + "learning_rate": 0.001992196434159292, + "loss": 0.168, + "step": 4986 + }, + { + "epoch": 0.04328955477817033, + "grad_norm": 0.51953125, + "learning_rate": 0.0019921925228713134, + "loss": 0.168, + "step": 4987 + }, + { + "epoch": 0.04329823525837449, + "grad_norm": 0.25, + "learning_rate": 0.001992188610607646, + "loss": 0.1523, + "step": 4988 + }, + { + "epoch": 0.04330691573857866, + "grad_norm": 0.322265625, + "learning_rate": 0.001992184697368294, + "loss": 0.1299, + "step": 4989 + }, + { + "epoch": 0.04331559621878282, + "grad_norm": 0.376953125, + "learning_rate": 0.001992180783153263, + "loss": 0.1895, + "step": 4990 + }, + { + "epoch": 0.04332427669898699, + "grad_norm": 0.609375, + "learning_rate": 0.001992176867962556, + "loss": 0.1514, + "step": 4991 + }, + { + "epoch": 0.04333295717919115, + "grad_norm": 0.087890625, + "learning_rate": 0.001992172951796178, + "loss": 0.1416, + "step": 4992 + }, + { + "epoch": 0.04334163765939532, + "grad_norm": 0.40234375, + "learning_rate": 0.001992169034654133, + "loss": 0.1582, + "step": 4993 + }, + { + "epoch": 0.043350318139599484, + "grad_norm": 0.2373046875, + "learning_rate": 0.001992165116536425, + "loss": 0.2031, + "step": 4994 + }, + { + "epoch": 0.04335899861980365, + "grad_norm": 0.267578125, + "learning_rate": 0.0019921611974430594, + "loss": 0.127, + "step": 4995 + }, + { + "epoch": 0.043367679100007814, + "grad_norm": 0.53125, + "learning_rate": 0.001992157277374039, + "loss": 0.1475, + "step": 4996 + }, + { + "epoch": 0.04337635958021198, + "grad_norm": 0.8828125, + "learning_rate": 0.0019921533563293696, + "loss": 0.1484, + "step": 4997 + }, + { + "epoch": 0.043385040060416144, + "grad_norm": 0.3359375, + "learning_rate": 0.001992149434309054, + "loss": 0.1797, + "step": 4998 + }, + { + "epoch": 0.04339372054062031, + "grad_norm": 0.345703125, + "learning_rate": 0.001992145511313097, + "loss": 0.1514, + "step": 4999 + }, + { + "epoch": 0.043402401020824474, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019921415873415038, + "loss": 0.1387, + "step": 5000 + }, + { + "epoch": 0.04341108150102864, + "grad_norm": 0.17578125, + "learning_rate": 0.0019921376623942776, + "loss": 0.1396, + "step": 5001 + }, + { + "epoch": 0.043419761981232805, + "grad_norm": 0.2001953125, + "learning_rate": 0.001992133736471423, + "loss": 0.2402, + "step": 5002 + }, + { + "epoch": 0.04342844246143697, + "grad_norm": 0.27734375, + "learning_rate": 0.0019921298095729447, + "loss": 0.1162, + "step": 5003 + }, + { + "epoch": 0.043437122941641135, + "grad_norm": 0.439453125, + "learning_rate": 0.0019921258816988463, + "loss": 0.1338, + "step": 5004 + }, + { + "epoch": 0.0434458034218453, + "grad_norm": 0.34375, + "learning_rate": 0.001992121952849133, + "loss": 0.1289, + "step": 5005 + }, + { + "epoch": 0.04345448390204946, + "grad_norm": 0.68359375, + "learning_rate": 0.0019921180230238074, + "loss": 0.1465, + "step": 5006 + }, + { + "epoch": 0.04346316438225362, + "grad_norm": 0.357421875, + "learning_rate": 0.0019921140922228757, + "loss": 0.1611, + "step": 5007 + }, + { + "epoch": 0.04347184486245779, + "grad_norm": 0.41796875, + "learning_rate": 0.0019921101604463416, + "loss": 0.1162, + "step": 5008 + }, + { + "epoch": 0.04348052534266195, + "grad_norm": 0.62890625, + "learning_rate": 0.001992106227694209, + "loss": 0.1445, + "step": 5009 + }, + { + "epoch": 0.04348920582286612, + "grad_norm": 0.12060546875, + "learning_rate": 0.001992102293966482, + "loss": 0.1289, + "step": 5010 + }, + { + "epoch": 0.043497886303070284, + "grad_norm": 0.2353515625, + "learning_rate": 0.001992098359263166, + "loss": 0.1377, + "step": 5011 + }, + { + "epoch": 0.04350656678327445, + "grad_norm": 0.482421875, + "learning_rate": 0.0019920944235842643, + "loss": 0.1445, + "step": 5012 + }, + { + "epoch": 0.043515247263478614, + "grad_norm": 0.384765625, + "learning_rate": 0.001992090486929782, + "loss": 0.1699, + "step": 5013 + }, + { + "epoch": 0.04352392774368278, + "grad_norm": 0.1259765625, + "learning_rate": 0.001992086549299722, + "loss": 0.1719, + "step": 5014 + }, + { + "epoch": 0.043532608223886944, + "grad_norm": 0.93359375, + "learning_rate": 0.0019920826106940904, + "loss": 0.1816, + "step": 5015 + }, + { + "epoch": 0.04354128870409111, + "grad_norm": 0.92578125, + "learning_rate": 0.0019920786711128905, + "loss": 0.166, + "step": 5016 + }, + { + "epoch": 0.043549969184295274, + "grad_norm": 0.490234375, + "learning_rate": 0.0019920747305561264, + "loss": 0.2051, + "step": 5017 + }, + { + "epoch": 0.04355864966449944, + "grad_norm": 0.46875, + "learning_rate": 0.001992070789023803, + "loss": 0.2227, + "step": 5018 + }, + { + "epoch": 0.043567330144703605, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019920668465159244, + "loss": 0.1592, + "step": 5019 + }, + { + "epoch": 0.04357601062490777, + "grad_norm": 0.1171875, + "learning_rate": 0.0019920629030324953, + "loss": 0.1406, + "step": 5020 + }, + { + "epoch": 0.043584691105111935, + "grad_norm": 0.375, + "learning_rate": 0.0019920589585735193, + "loss": 0.1904, + "step": 5021 + }, + { + "epoch": 0.0435933715853161, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019920550131390007, + "loss": 0.1309, + "step": 5022 + }, + { + "epoch": 0.043602052065520265, + "grad_norm": 0.68359375, + "learning_rate": 0.0019920510667289443, + "loss": 0.1738, + "step": 5023 + }, + { + "epoch": 0.04361073254572443, + "grad_norm": 0.42578125, + "learning_rate": 0.0019920471193433545, + "loss": 0.1318, + "step": 5024 + }, + { + "epoch": 0.043619413025928595, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019920431709822355, + "loss": 0.1162, + "step": 5025 + }, + { + "epoch": 0.04362809350613276, + "grad_norm": 0.205078125, + "learning_rate": 0.001992039221645591, + "loss": 0.1738, + "step": 5026 + }, + { + "epoch": 0.043636773986336926, + "grad_norm": 0.5234375, + "learning_rate": 0.001992035271333426, + "loss": 0.1953, + "step": 5027 + }, + { + "epoch": 0.04364545446654109, + "grad_norm": 0.7734375, + "learning_rate": 0.0019920313200457447, + "loss": 0.1973, + "step": 5028 + }, + { + "epoch": 0.043654134946745256, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019920273677825513, + "loss": 0.1641, + "step": 5029 + }, + { + "epoch": 0.04366281542694942, + "grad_norm": 0.25390625, + "learning_rate": 0.00199202341454385, + "loss": 0.1113, + "step": 5030 + }, + { + "epoch": 0.043671495907153586, + "grad_norm": 0.40625, + "learning_rate": 0.0019920194603296458, + "loss": 0.1748, + "step": 5031 + }, + { + "epoch": 0.04368017638735775, + "grad_norm": 0.21484375, + "learning_rate": 0.0019920155051399422, + "loss": 0.1523, + "step": 5032 + }, + { + "epoch": 0.043688856867561916, + "grad_norm": 0.1376953125, + "learning_rate": 0.001992011548974744, + "loss": 0.1289, + "step": 5033 + }, + { + "epoch": 0.04369753734776608, + "grad_norm": 0.47265625, + "learning_rate": 0.001992007591834055, + "loss": 0.2656, + "step": 5034 + }, + { + "epoch": 0.043706217827970247, + "grad_norm": 1.015625, + "learning_rate": 0.00199200363371788, + "loss": 0.1631, + "step": 5035 + }, + { + "epoch": 0.04371489830817441, + "grad_norm": 0.240234375, + "learning_rate": 0.0019919996746262233, + "loss": 0.1289, + "step": 5036 + }, + { + "epoch": 0.04372357878837857, + "grad_norm": 0.80859375, + "learning_rate": 0.0019919957145590893, + "loss": 0.1357, + "step": 5037 + }, + { + "epoch": 0.043732259268582735, + "grad_norm": 0.10302734375, + "learning_rate": 0.001991991753516482, + "loss": 0.1973, + "step": 5038 + }, + { + "epoch": 0.0437409397487869, + "grad_norm": 0.66015625, + "learning_rate": 0.001991987791498406, + "loss": 0.1748, + "step": 5039 + }, + { + "epoch": 0.043749620228991065, + "grad_norm": 0.392578125, + "learning_rate": 0.0019919838285048655, + "loss": 0.166, + "step": 5040 + }, + { + "epoch": 0.04375830070919523, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019919798645358653, + "loss": 0.1699, + "step": 5041 + }, + { + "epoch": 0.043766981189399395, + "grad_norm": 0.068359375, + "learning_rate": 0.0019919758995914087, + "loss": 0.1099, + "step": 5042 + }, + { + "epoch": 0.04377566166960356, + "grad_norm": 0.1640625, + "learning_rate": 0.001991971933671501, + "loss": 0.1445, + "step": 5043 + }, + { + "epoch": 0.043784342149807726, + "grad_norm": 0.2734375, + "learning_rate": 0.001991967966776146, + "loss": 0.1387, + "step": 5044 + }, + { + "epoch": 0.04379302263001189, + "grad_norm": 0.46875, + "learning_rate": 0.0019919639989053484, + "loss": 0.1562, + "step": 5045 + }, + { + "epoch": 0.043801703110216056, + "grad_norm": 0.41796875, + "learning_rate": 0.0019919600300591124, + "loss": 0.1309, + "step": 5046 + }, + { + "epoch": 0.04381038359042022, + "grad_norm": 0.85546875, + "learning_rate": 0.0019919560602374422, + "loss": 0.1484, + "step": 5047 + }, + { + "epoch": 0.043819064070624386, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019919520894403422, + "loss": 0.1436, + "step": 5048 + }, + { + "epoch": 0.04382774455082855, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019919481176678172, + "loss": 0.1328, + "step": 5049 + }, + { + "epoch": 0.043836425031032716, + "grad_norm": 0.13671875, + "learning_rate": 0.0019919441449198706, + "loss": 0.1611, + "step": 5050 + }, + { + "epoch": 0.04384510551123688, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019919401711965077, + "loss": 0.1445, + "step": 5051 + }, + { + "epoch": 0.04385378599144105, + "grad_norm": 0.349609375, + "learning_rate": 0.001991936196497732, + "loss": 0.1816, + "step": 5052 + }, + { + "epoch": 0.04386246647164521, + "grad_norm": 0.53515625, + "learning_rate": 0.0019919322208235488, + "loss": 0.1641, + "step": 5053 + }, + { + "epoch": 0.04387114695184938, + "grad_norm": 0.63671875, + "learning_rate": 0.0019919282441739614, + "loss": 0.2432, + "step": 5054 + }, + { + "epoch": 0.04387982743205354, + "grad_norm": 0.173828125, + "learning_rate": 0.001991924266548975, + "loss": 0.1836, + "step": 5055 + }, + { + "epoch": 0.04388850791225771, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019919202879485937, + "loss": 0.2949, + "step": 5056 + }, + { + "epoch": 0.04389718839246187, + "grad_norm": 0.26171875, + "learning_rate": 0.0019919163083728215, + "loss": 0.1641, + "step": 5057 + }, + { + "epoch": 0.04390586887266604, + "grad_norm": 0.107421875, + "learning_rate": 0.0019919123278216632, + "loss": 0.1621, + "step": 5058 + }, + { + "epoch": 0.0439145493528702, + "grad_norm": 0.51171875, + "learning_rate": 0.001991908346295123, + "loss": 0.21, + "step": 5059 + }, + { + "epoch": 0.04392322983307437, + "grad_norm": 0.5078125, + "learning_rate": 0.001991904363793205, + "loss": 0.1465, + "step": 5060 + }, + { + "epoch": 0.04393191031327853, + "grad_norm": 0.81640625, + "learning_rate": 0.001991900380315914, + "loss": 0.208, + "step": 5061 + }, + { + "epoch": 0.0439405907934827, + "grad_norm": 0.08837890625, + "learning_rate": 0.001991896395863254, + "loss": 0.1484, + "step": 5062 + }, + { + "epoch": 0.04394927127368686, + "grad_norm": 0.1875, + "learning_rate": 0.0019918924104352295, + "loss": 0.1768, + "step": 5063 + }, + { + "epoch": 0.04395795175389103, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019918884240318455, + "loss": 0.1514, + "step": 5064 + }, + { + "epoch": 0.04396663223409519, + "grad_norm": 0.3671875, + "learning_rate": 0.001991884436653105, + "loss": 0.1992, + "step": 5065 + }, + { + "epoch": 0.04397531271429936, + "grad_norm": 0.52734375, + "learning_rate": 0.001991880448299013, + "loss": 0.168, + "step": 5066 + }, + { + "epoch": 0.043983993194503516, + "grad_norm": 0.8203125, + "learning_rate": 0.001991876458969574, + "loss": 0.1621, + "step": 5067 + }, + { + "epoch": 0.04399267367470768, + "grad_norm": 0.435546875, + "learning_rate": 0.0019918724686647927, + "loss": 0.1865, + "step": 5068 + }, + { + "epoch": 0.04400135415491185, + "grad_norm": 0.462890625, + "learning_rate": 0.001991868477384673, + "loss": 0.1494, + "step": 5069 + }, + { + "epoch": 0.04401003463511601, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019918644851292192, + "loss": 0.1465, + "step": 5070 + }, + { + "epoch": 0.04401871511532018, + "grad_norm": 0.30078125, + "learning_rate": 0.001991860491898436, + "loss": 0.167, + "step": 5071 + }, + { + "epoch": 0.04402739559552434, + "grad_norm": 0.07958984375, + "learning_rate": 0.001991856497692327, + "loss": 0.1416, + "step": 5072 + }, + { + "epoch": 0.04403607607572851, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019918525025108976, + "loss": 0.1426, + "step": 5073 + }, + { + "epoch": 0.04404475655593267, + "grad_norm": 0.251953125, + "learning_rate": 0.001991848506354152, + "loss": 0.1758, + "step": 5074 + }, + { + "epoch": 0.04405343703613684, + "grad_norm": 0.220703125, + "learning_rate": 0.0019918445092220933, + "loss": 0.1689, + "step": 5075 + }, + { + "epoch": 0.044062117516341, + "grad_norm": 0.3046875, + "learning_rate": 0.0019918405111147277, + "loss": 0.1582, + "step": 5076 + }, + { + "epoch": 0.04407079799654517, + "grad_norm": 0.45703125, + "learning_rate": 0.0019918365120320585, + "loss": 0.1895, + "step": 5077 + }, + { + "epoch": 0.04407947847674933, + "grad_norm": 0.181640625, + "learning_rate": 0.00199183251197409, + "loss": 0.1338, + "step": 5078 + }, + { + "epoch": 0.0440881589569535, + "grad_norm": 0.71875, + "learning_rate": 0.0019918285109408276, + "loss": 0.1436, + "step": 5079 + }, + { + "epoch": 0.04409683943715766, + "grad_norm": 0.5390625, + "learning_rate": 0.001991824508932274, + "loss": 0.1777, + "step": 5080 + }, + { + "epoch": 0.04410551991736183, + "grad_norm": 0.1123046875, + "learning_rate": 0.001991820505948435, + "loss": 0.1738, + "step": 5081 + }, + { + "epoch": 0.04411420039756599, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019918165019893143, + "loss": 0.1504, + "step": 5082 + }, + { + "epoch": 0.04412288087777016, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019918124970549167, + "loss": 0.1621, + "step": 5083 + }, + { + "epoch": 0.044131561357974324, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019918084911452463, + "loss": 0.1426, + "step": 5084 + }, + { + "epoch": 0.04414024183817849, + "grad_norm": 0.59375, + "learning_rate": 0.001991804484260308, + "loss": 0.1328, + "step": 5085 + }, + { + "epoch": 0.044148922318382654, + "grad_norm": 0.46875, + "learning_rate": 0.001991800476400105, + "loss": 0.1387, + "step": 5086 + }, + { + "epoch": 0.04415760279858682, + "grad_norm": 0.30859375, + "learning_rate": 0.0019917964675646425, + "loss": 0.126, + "step": 5087 + }, + { + "epoch": 0.044166283278790984, + "grad_norm": 0.15625, + "learning_rate": 0.001991792457753925, + "loss": 0.207, + "step": 5088 + }, + { + "epoch": 0.04417496375899515, + "grad_norm": 0.234375, + "learning_rate": 0.0019917884469679565, + "loss": 0.1797, + "step": 5089 + }, + { + "epoch": 0.044183644239199314, + "grad_norm": 0.72265625, + "learning_rate": 0.0019917844352067417, + "loss": 0.1572, + "step": 5090 + }, + { + "epoch": 0.04419232471940348, + "grad_norm": 0.625, + "learning_rate": 0.0019917804224702847, + "loss": 0.2012, + "step": 5091 + }, + { + "epoch": 0.044201005199607644, + "grad_norm": 0.30078125, + "learning_rate": 0.00199177640875859, + "loss": 0.1299, + "step": 5092 + }, + { + "epoch": 0.04420968567981181, + "grad_norm": 0.20703125, + "learning_rate": 0.001991772394071662, + "loss": 0.1738, + "step": 5093 + }, + { + "epoch": 0.044218366160015975, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019917683784095052, + "loss": 0.1445, + "step": 5094 + }, + { + "epoch": 0.04422704664022014, + "grad_norm": 0.103515625, + "learning_rate": 0.001991764361772124, + "loss": 0.1836, + "step": 5095 + }, + { + "epoch": 0.044235727120424305, + "grad_norm": 0.23046875, + "learning_rate": 0.0019917603441595225, + "loss": 0.1416, + "step": 5096 + }, + { + "epoch": 0.04424440760062847, + "grad_norm": 0.2890625, + "learning_rate": 0.0019917563255717054, + "loss": 0.1396, + "step": 5097 + }, + { + "epoch": 0.04425308808083263, + "grad_norm": 0.3984375, + "learning_rate": 0.001991752306008677, + "loss": 0.1523, + "step": 5098 + }, + { + "epoch": 0.04426176856103679, + "grad_norm": 0.1298828125, + "learning_rate": 0.001991748285470442, + "loss": 0.1816, + "step": 5099 + }, + { + "epoch": 0.04427044904124096, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019917442639570044, + "loss": 0.126, + "step": 5100 + }, + { + "epoch": 0.044279129521445124, + "grad_norm": 0.2080078125, + "learning_rate": 0.001991740241468368, + "loss": 0.1631, + "step": 5101 + }, + { + "epoch": 0.04428781000164929, + "grad_norm": 0.400390625, + "learning_rate": 0.001991736218004538, + "loss": 0.1338, + "step": 5102 + }, + { + "epoch": 0.044296490481853454, + "grad_norm": 0.408203125, + "learning_rate": 0.0019917321935655195, + "loss": 0.1514, + "step": 5103 + }, + { + "epoch": 0.04430517096205762, + "grad_norm": 0.111328125, + "learning_rate": 0.0019917281681513153, + "loss": 0.1338, + "step": 5104 + }, + { + "epoch": 0.044313851442261784, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019917241417619306, + "loss": 0.1562, + "step": 5105 + }, + { + "epoch": 0.04432253192246595, + "grad_norm": 0.35546875, + "learning_rate": 0.0019917201143973703, + "loss": 0.1562, + "step": 5106 + }, + { + "epoch": 0.044331212402670114, + "grad_norm": 0.306640625, + "learning_rate": 0.0019917160860576377, + "loss": 0.1436, + "step": 5107 + }, + { + "epoch": 0.04433989288287428, + "grad_norm": 0.4375, + "learning_rate": 0.001991712056742738, + "loss": 0.1729, + "step": 5108 + }, + { + "epoch": 0.044348573363078445, + "grad_norm": 0.484375, + "learning_rate": 0.0019917080264526756, + "loss": 0.1621, + "step": 5109 + }, + { + "epoch": 0.04435725384328261, + "grad_norm": 0.45703125, + "learning_rate": 0.0019917039951874546, + "loss": 0.1309, + "step": 5110 + }, + { + "epoch": 0.044365934323486775, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019916999629470793, + "loss": 0.1465, + "step": 5111 + }, + { + "epoch": 0.04437461480369094, + "grad_norm": 0.201171875, + "learning_rate": 0.0019916959297315543, + "loss": 0.1777, + "step": 5112 + }, + { + "epoch": 0.044383295283895105, + "grad_norm": 0.19140625, + "learning_rate": 0.0019916918955408844, + "loss": 0.1924, + "step": 5113 + }, + { + "epoch": 0.04439197576409927, + "grad_norm": 0.375, + "learning_rate": 0.001991687860375073, + "loss": 0.2148, + "step": 5114 + }, + { + "epoch": 0.044400656244303435, + "grad_norm": 1.0703125, + "learning_rate": 0.001991683824234126, + "loss": 0.1738, + "step": 5115 + }, + { + "epoch": 0.0444093367245076, + "grad_norm": 0.3984375, + "learning_rate": 0.001991679787118046, + "loss": 0.123, + "step": 5116 + }, + { + "epoch": 0.044418017204711766, + "grad_norm": 0.330078125, + "learning_rate": 0.0019916757490268393, + "loss": 0.1504, + "step": 5117 + }, + { + "epoch": 0.04442669768491593, + "grad_norm": 0.1396484375, + "learning_rate": 0.001991671709960509, + "loss": 0.1396, + "step": 5118 + }, + { + "epoch": 0.044435378165120096, + "grad_norm": 0.18359375, + "learning_rate": 0.0019916676699190602, + "loss": 0.2012, + "step": 5119 + }, + { + "epoch": 0.04444405864532426, + "grad_norm": 0.43359375, + "learning_rate": 0.0019916636289024968, + "loss": 0.1543, + "step": 5120 + }, + { + "epoch": 0.044452739125528426, + "grad_norm": 0.087890625, + "learning_rate": 0.0019916595869108236, + "loss": 0.1357, + "step": 5121 + }, + { + "epoch": 0.04446141960573259, + "grad_norm": 0.296875, + "learning_rate": 0.001991655543944044, + "loss": 0.1514, + "step": 5122 + }, + { + "epoch": 0.044470100085936756, + "grad_norm": 0.25390625, + "learning_rate": 0.0019916515000021643, + "loss": 0.1094, + "step": 5123 + }, + { + "epoch": 0.04447878056614092, + "grad_norm": 0.59765625, + "learning_rate": 0.0019916474550851875, + "loss": 0.1494, + "step": 5124 + }, + { + "epoch": 0.044487461046345086, + "grad_norm": 0.263671875, + "learning_rate": 0.0019916434091931186, + "loss": 0.1543, + "step": 5125 + }, + { + "epoch": 0.04449614152654925, + "grad_norm": 1.0078125, + "learning_rate": 0.001991639362325962, + "loss": 0.1953, + "step": 5126 + }, + { + "epoch": 0.04450482200675342, + "grad_norm": 0.076171875, + "learning_rate": 0.001991635314483722, + "loss": 0.1191, + "step": 5127 + }, + { + "epoch": 0.04451350248695758, + "grad_norm": 0.369140625, + "learning_rate": 0.0019916312656664023, + "loss": 0.1367, + "step": 5128 + }, + { + "epoch": 0.04452218296716174, + "grad_norm": 0.349609375, + "learning_rate": 0.001991627215874009, + "loss": 0.1562, + "step": 5129 + }, + { + "epoch": 0.044530863447365905, + "grad_norm": 0.51953125, + "learning_rate": 0.001991623165106545, + "loss": 0.1768, + "step": 5130 + }, + { + "epoch": 0.04453954392757007, + "grad_norm": 0.46484375, + "learning_rate": 0.0019916191133640153, + "loss": 0.1504, + "step": 5131 + }, + { + "epoch": 0.044548224407774235, + "grad_norm": 0.248046875, + "learning_rate": 0.0019916150606464245, + "loss": 0.1289, + "step": 5132 + }, + { + "epoch": 0.0445569048879784, + "grad_norm": 0.62109375, + "learning_rate": 0.001991611006953777, + "loss": 0.1465, + "step": 5133 + }, + { + "epoch": 0.044565585368182566, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019916069522860767, + "loss": 0.1475, + "step": 5134 + }, + { + "epoch": 0.04457426584838673, + "grad_norm": 0.1982421875, + "learning_rate": 0.001991602896643329, + "loss": 0.1748, + "step": 5135 + }, + { + "epoch": 0.044582946328590896, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019915988400255372, + "loss": 0.1602, + "step": 5136 + }, + { + "epoch": 0.04459162680879506, + "grad_norm": 0.318359375, + "learning_rate": 0.001991594782432707, + "loss": 0.1631, + "step": 5137 + }, + { + "epoch": 0.044600307288999226, + "grad_norm": 0.61328125, + "learning_rate": 0.0019915907238648414, + "loss": 0.1162, + "step": 5138 + }, + { + "epoch": 0.04460898776920339, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019915866643219456, + "loss": 0.1348, + "step": 5139 + }, + { + "epoch": 0.044617668249407556, + "grad_norm": 0.10986328125, + "learning_rate": 0.001991582603804024, + "loss": 0.1846, + "step": 5140 + }, + { + "epoch": 0.04462634872961172, + "grad_norm": 0.353515625, + "learning_rate": 0.001991578542311082, + "loss": 0.1426, + "step": 5141 + }, + { + "epoch": 0.04463502920981589, + "grad_norm": 1.171875, + "learning_rate": 0.001991574479843122, + "loss": 0.1807, + "step": 5142 + }, + { + "epoch": 0.04464370969002005, + "grad_norm": 0.5703125, + "learning_rate": 0.0019915704164001503, + "loss": 0.1289, + "step": 5143 + }, + { + "epoch": 0.04465239017022422, + "grad_norm": 0.1806640625, + "learning_rate": 0.00199156635198217, + "loss": 0.1328, + "step": 5144 + }, + { + "epoch": 0.04466107065042838, + "grad_norm": 0.91796875, + "learning_rate": 0.0019915622865891865, + "loss": 0.1777, + "step": 5145 + }, + { + "epoch": 0.04466975113063255, + "grad_norm": 0.126953125, + "learning_rate": 0.0019915582202212037, + "loss": 0.1138, + "step": 5146 + }, + { + "epoch": 0.04467843161083671, + "grad_norm": 0.76953125, + "learning_rate": 0.0019915541528782266, + "loss": 0.1582, + "step": 5147 + }, + { + "epoch": 0.04468711209104088, + "grad_norm": 1.03125, + "learning_rate": 0.001991550084560259, + "loss": 0.1797, + "step": 5148 + }, + { + "epoch": 0.04469579257124504, + "grad_norm": 0.2470703125, + "learning_rate": 0.001991546015267306, + "loss": 0.1543, + "step": 5149 + }, + { + "epoch": 0.04470447305144921, + "grad_norm": 0.38671875, + "learning_rate": 0.001991541944999371, + "loss": 0.168, + "step": 5150 + }, + { + "epoch": 0.04471315353165337, + "grad_norm": 0.2890625, + "learning_rate": 0.0019915378737564594, + "loss": 0.1396, + "step": 5151 + }, + { + "epoch": 0.04472183401185754, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019915338015385753, + "loss": 0.1514, + "step": 5152 + }, + { + "epoch": 0.0447305144920617, + "grad_norm": 0.78125, + "learning_rate": 0.0019915297283457233, + "loss": 0.2363, + "step": 5153 + }, + { + "epoch": 0.04473919497226587, + "grad_norm": 0.86328125, + "learning_rate": 0.001991525654177908, + "loss": 0.2012, + "step": 5154 + }, + { + "epoch": 0.04474787545247003, + "grad_norm": 1.109375, + "learning_rate": 0.001991521579035133, + "loss": 0.126, + "step": 5155 + }, + { + "epoch": 0.0447565559326742, + "grad_norm": 0.337890625, + "learning_rate": 0.0019915175029174043, + "loss": 0.1348, + "step": 5156 + }, + { + "epoch": 0.04476523641287836, + "grad_norm": 0.26953125, + "learning_rate": 0.0019915134258247248, + "loss": 0.1191, + "step": 5157 + }, + { + "epoch": 0.04477391689308253, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019915093477571, + "loss": 0.1611, + "step": 5158 + }, + { + "epoch": 0.044782597373286694, + "grad_norm": 0.337890625, + "learning_rate": 0.001991505268714534, + "loss": 0.1367, + "step": 5159 + }, + { + "epoch": 0.04479127785349085, + "grad_norm": 0.400390625, + "learning_rate": 0.0019915011886970308, + "loss": 0.1758, + "step": 5160 + }, + { + "epoch": 0.04479995833369502, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019914971077045953, + "loss": 0.1357, + "step": 5161 + }, + { + "epoch": 0.04480863881389918, + "grad_norm": 0.189453125, + "learning_rate": 0.0019914930257372322, + "loss": 0.1621, + "step": 5162 + }, + { + "epoch": 0.04481731929410335, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019914889427949454, + "loss": 0.1523, + "step": 5163 + }, + { + "epoch": 0.04482599977430751, + "grad_norm": 0.26171875, + "learning_rate": 0.00199148485887774, + "loss": 0.1387, + "step": 5164 + }, + { + "epoch": 0.04483468025451168, + "grad_norm": 0.40234375, + "learning_rate": 0.0019914807739856203, + "loss": 0.1826, + "step": 5165 + }, + { + "epoch": 0.04484336073471584, + "grad_norm": 0.859375, + "learning_rate": 0.0019914766881185906, + "loss": 0.1797, + "step": 5166 + }, + { + "epoch": 0.04485204121492001, + "grad_norm": 0.640625, + "learning_rate": 0.001991472601276655, + "loss": 0.1689, + "step": 5167 + }, + { + "epoch": 0.04486072169512417, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019914685134598186, + "loss": 0.123, + "step": 5168 + }, + { + "epoch": 0.04486940217532834, + "grad_norm": 0.1259765625, + "learning_rate": 0.001991464424668086, + "loss": 0.166, + "step": 5169 + }, + { + "epoch": 0.0448780826555325, + "grad_norm": 0.173828125, + "learning_rate": 0.0019914603349014607, + "loss": 0.1104, + "step": 5170 + }, + { + "epoch": 0.04488676313573667, + "grad_norm": 0.294921875, + "learning_rate": 0.001991456244159948, + "loss": 0.1855, + "step": 5171 + }, + { + "epoch": 0.04489544361594083, + "grad_norm": 1.9375, + "learning_rate": 0.0019914521524435525, + "loss": 0.1357, + "step": 5172 + }, + { + "epoch": 0.044904124096145, + "grad_norm": 0.23828125, + "learning_rate": 0.0019914480597522776, + "loss": 0.1602, + "step": 5173 + }, + { + "epoch": 0.04491280457634916, + "grad_norm": 0.1357421875, + "learning_rate": 0.001991443966086129, + "loss": 0.1455, + "step": 5174 + }, + { + "epoch": 0.04492148505655333, + "grad_norm": 0.306640625, + "learning_rate": 0.0019914398714451103, + "loss": 0.1738, + "step": 5175 + }, + { + "epoch": 0.044930165536757494, + "grad_norm": 0.57421875, + "learning_rate": 0.0019914357758292266, + "loss": 0.1533, + "step": 5176 + }, + { + "epoch": 0.04493884601696166, + "grad_norm": 0.3359375, + "learning_rate": 0.001991431679238482, + "loss": 0.166, + "step": 5177 + }, + { + "epoch": 0.044947526497165824, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019914275816728813, + "loss": 0.126, + "step": 5178 + }, + { + "epoch": 0.04495620697736999, + "grad_norm": 0.2451171875, + "learning_rate": 0.001991423483132429, + "loss": 0.1709, + "step": 5179 + }, + { + "epoch": 0.044964887457574154, + "grad_norm": 0.83984375, + "learning_rate": 0.001991419383617129, + "loss": 0.1504, + "step": 5180 + }, + { + "epoch": 0.04497356793777832, + "grad_norm": 0.318359375, + "learning_rate": 0.001991415283126986, + "loss": 0.1328, + "step": 5181 + }, + { + "epoch": 0.044982248417982484, + "grad_norm": 0.20703125, + "learning_rate": 0.0019914111816620047, + "loss": 0.1021, + "step": 5182 + }, + { + "epoch": 0.04499092889818665, + "grad_norm": 0.3046875, + "learning_rate": 0.00199140707922219, + "loss": 0.1514, + "step": 5183 + }, + { + "epoch": 0.044999609378390815, + "grad_norm": 0.271484375, + "learning_rate": 0.0019914029758075456, + "loss": 0.1387, + "step": 5184 + }, + { + "epoch": 0.04500828985859498, + "grad_norm": 0.28515625, + "learning_rate": 0.0019913988714180763, + "loss": 0.1211, + "step": 5185 + }, + { + "epoch": 0.045016970338799145, + "grad_norm": 1.328125, + "learning_rate": 0.0019913947660537864, + "loss": 0.1514, + "step": 5186 + }, + { + "epoch": 0.04502565081900331, + "grad_norm": 0.2158203125, + "learning_rate": 0.001991390659714681, + "loss": 0.1602, + "step": 5187 + }, + { + "epoch": 0.045034331299207475, + "grad_norm": 0.5234375, + "learning_rate": 0.001991386552400764, + "loss": 0.1855, + "step": 5188 + }, + { + "epoch": 0.04504301177941164, + "grad_norm": 0.7578125, + "learning_rate": 0.00199138244411204, + "loss": 0.1406, + "step": 5189 + }, + { + "epoch": 0.045051692259615805, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019913783348485137, + "loss": 0.1406, + "step": 5190 + }, + { + "epoch": 0.045060372739819964, + "grad_norm": 0.318359375, + "learning_rate": 0.001991374224610189, + "loss": 0.2051, + "step": 5191 + }, + { + "epoch": 0.04506905322002413, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019913701133970714, + "loss": 0.1416, + "step": 5192 + }, + { + "epoch": 0.045077733700228294, + "grad_norm": 0.51953125, + "learning_rate": 0.0019913660012091643, + "loss": 0.1445, + "step": 5193 + }, + { + "epoch": 0.04508641418043246, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019913618880464732, + "loss": 0.1973, + "step": 5194 + }, + { + "epoch": 0.045095094660636624, + "grad_norm": 0.349609375, + "learning_rate": 0.001991357773909002, + "loss": 0.1836, + "step": 5195 + }, + { + "epoch": 0.04510377514084079, + "grad_norm": 2.40625, + "learning_rate": 0.0019913536587967554, + "loss": 0.1738, + "step": 5196 + }, + { + "epoch": 0.045112455621044954, + "grad_norm": 0.46875, + "learning_rate": 0.0019913495427097373, + "loss": 0.1895, + "step": 5197 + }, + { + "epoch": 0.04512113610124912, + "grad_norm": 0.59765625, + "learning_rate": 0.0019913454256479533, + "loss": 0.1396, + "step": 5198 + }, + { + "epoch": 0.045129816581453285, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019913413076114075, + "loss": 0.1426, + "step": 5199 + }, + { + "epoch": 0.04513849706165745, + "grad_norm": 0.18359375, + "learning_rate": 0.0019913371886001036, + "loss": 0.1377, + "step": 5200 + }, + { + "epoch": 0.045147177541861615, + "grad_norm": 0.203125, + "learning_rate": 0.001991333068614047, + "loss": 0.1602, + "step": 5201 + }, + { + "epoch": 0.04515585802206578, + "grad_norm": 0.427734375, + "learning_rate": 0.001991328947653242, + "loss": 0.1748, + "step": 5202 + }, + { + "epoch": 0.045164538502269945, + "grad_norm": 0.3828125, + "learning_rate": 0.001991324825717693, + "loss": 0.1494, + "step": 5203 + }, + { + "epoch": 0.04517321898247411, + "grad_norm": 0.107421875, + "learning_rate": 0.0019913207028074047, + "loss": 0.1504, + "step": 5204 + }, + { + "epoch": 0.045181899462678275, + "grad_norm": 0.314453125, + "learning_rate": 0.0019913165789223813, + "loss": 0.1543, + "step": 5205 + }, + { + "epoch": 0.04519057994288244, + "grad_norm": 0.51953125, + "learning_rate": 0.0019913124540626276, + "loss": 0.1914, + "step": 5206 + }, + { + "epoch": 0.045199260423086605, + "grad_norm": 0.2294921875, + "learning_rate": 0.001991308328228148, + "loss": 0.1572, + "step": 5207 + }, + { + "epoch": 0.04520794090329077, + "grad_norm": 0.21875, + "learning_rate": 0.001991304201418947, + "loss": 0.2324, + "step": 5208 + }, + { + "epoch": 0.045216621383494936, + "grad_norm": 0.34765625, + "learning_rate": 0.001991300073635029, + "loss": 0.1562, + "step": 5209 + }, + { + "epoch": 0.0452253018636991, + "grad_norm": 1.109375, + "learning_rate": 0.001991295944876399, + "loss": 0.1973, + "step": 5210 + }, + { + "epoch": 0.045233982343903266, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019912918151430608, + "loss": 0.1738, + "step": 5211 + }, + { + "epoch": 0.04524266282410743, + "grad_norm": 0.578125, + "learning_rate": 0.001991287684435019, + "loss": 0.1846, + "step": 5212 + }, + { + "epoch": 0.045251343304311596, + "grad_norm": 0.392578125, + "learning_rate": 0.001991283552752279, + "loss": 0.1455, + "step": 5213 + }, + { + "epoch": 0.04526002378451576, + "grad_norm": 0.283203125, + "learning_rate": 0.001991279420094844, + "loss": 0.1025, + "step": 5214 + }, + { + "epoch": 0.045268704264719926, + "grad_norm": 0.416015625, + "learning_rate": 0.00199127528646272, + "loss": 0.1592, + "step": 5215 + }, + { + "epoch": 0.04527738474492409, + "grad_norm": 0.1328125, + "learning_rate": 0.0019912711518559104, + "loss": 0.1641, + "step": 5216 + }, + { + "epoch": 0.04528606522512826, + "grad_norm": 0.53515625, + "learning_rate": 0.00199126701627442, + "loss": 0.1406, + "step": 5217 + }, + { + "epoch": 0.04529474570533242, + "grad_norm": 0.29296875, + "learning_rate": 0.0019912628797182537, + "loss": 0.165, + "step": 5218 + }, + { + "epoch": 0.04530342618553659, + "grad_norm": 0.3203125, + "learning_rate": 0.001991258742187415, + "loss": 0.1113, + "step": 5219 + }, + { + "epoch": 0.04531210666574075, + "grad_norm": 0.98828125, + "learning_rate": 0.0019912546036819096, + "loss": 0.1807, + "step": 5220 + }, + { + "epoch": 0.04532078714594492, + "grad_norm": 0.16015625, + "learning_rate": 0.0019912504642017412, + "loss": 0.1602, + "step": 5221 + }, + { + "epoch": 0.045329467626149075, + "grad_norm": 0.1640625, + "learning_rate": 0.001991246323746915, + "loss": 0.1455, + "step": 5222 + }, + { + "epoch": 0.04533814810635324, + "grad_norm": 0.4140625, + "learning_rate": 0.0019912421823174353, + "loss": 0.207, + "step": 5223 + }, + { + "epoch": 0.045346828586557406, + "grad_norm": 0.38671875, + "learning_rate": 0.0019912380399133064, + "loss": 0.1426, + "step": 5224 + }, + { + "epoch": 0.04535550906676157, + "grad_norm": 0.953125, + "learning_rate": 0.001991233896534533, + "loss": 0.1396, + "step": 5225 + }, + { + "epoch": 0.045364189546965736, + "grad_norm": 0.08203125, + "learning_rate": 0.0019912297521811196, + "loss": 0.1348, + "step": 5226 + }, + { + "epoch": 0.0453728700271699, + "grad_norm": 0.65625, + "learning_rate": 0.001991225606853071, + "loss": 0.1475, + "step": 5227 + }, + { + "epoch": 0.045381550507374066, + "grad_norm": 0.162109375, + "learning_rate": 0.0019912214605503913, + "loss": 0.1562, + "step": 5228 + }, + { + "epoch": 0.04539023098757823, + "grad_norm": 0.61328125, + "learning_rate": 0.001991217313273085, + "loss": 0.1602, + "step": 5229 + }, + { + "epoch": 0.045398911467782396, + "grad_norm": 0.74609375, + "learning_rate": 0.001991213165021157, + "loss": 0.1094, + "step": 5230 + }, + { + "epoch": 0.04540759194798656, + "grad_norm": 0.162109375, + "learning_rate": 0.0019912090157946116, + "loss": 0.207, + "step": 5231 + }, + { + "epoch": 0.045416272428190727, + "grad_norm": 0.89453125, + "learning_rate": 0.0019912048655934536, + "loss": 0.1582, + "step": 5232 + }, + { + "epoch": 0.04542495290839489, + "grad_norm": 0.50390625, + "learning_rate": 0.0019912007144176867, + "loss": 0.1543, + "step": 5233 + }, + { + "epoch": 0.04543363338859906, + "grad_norm": 0.53125, + "learning_rate": 0.0019911965622673167, + "loss": 0.3008, + "step": 5234 + }, + { + "epoch": 0.04544231386880322, + "grad_norm": 0.4296875, + "learning_rate": 0.0019911924091423473, + "loss": 0.248, + "step": 5235 + }, + { + "epoch": 0.04545099434900739, + "grad_norm": 0.201171875, + "learning_rate": 0.0019911882550427834, + "loss": 0.168, + "step": 5236 + }, + { + "epoch": 0.04545967482921155, + "grad_norm": 0.234375, + "learning_rate": 0.0019911840999686293, + "loss": 0.165, + "step": 5237 + }, + { + "epoch": 0.04546835530941572, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019911799439198898, + "loss": 0.2012, + "step": 5238 + }, + { + "epoch": 0.04547703578961988, + "grad_norm": 0.291015625, + "learning_rate": 0.001991175786896569, + "loss": 0.1689, + "step": 5239 + }, + { + "epoch": 0.04548571626982405, + "grad_norm": 0.169921875, + "learning_rate": 0.001991171628898672, + "loss": 0.1504, + "step": 5240 + }, + { + "epoch": 0.04549439675002821, + "grad_norm": 0.4296875, + "learning_rate": 0.0019911674699262033, + "loss": 0.1367, + "step": 5241 + }, + { + "epoch": 0.04550307723023238, + "grad_norm": 0.1357421875, + "learning_rate": 0.001991163309979167, + "loss": 0.1187, + "step": 5242 + }, + { + "epoch": 0.04551175771043654, + "grad_norm": 0.2451171875, + "learning_rate": 0.001991159149057568, + "loss": 0.1309, + "step": 5243 + }, + { + "epoch": 0.04552043819064071, + "grad_norm": 0.3046875, + "learning_rate": 0.001991154987161411, + "loss": 0.1807, + "step": 5244 + }, + { + "epoch": 0.04552911867084487, + "grad_norm": 0.4375, + "learning_rate": 0.0019911508242906996, + "loss": 0.1719, + "step": 5245 + }, + { + "epoch": 0.04553779915104904, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019911466604454397, + "loss": 0.1641, + "step": 5246 + }, + { + "epoch": 0.0455464796312532, + "grad_norm": 0.40625, + "learning_rate": 0.0019911424956256347, + "loss": 0.1465, + "step": 5247 + }, + { + "epoch": 0.04555516011145737, + "grad_norm": 0.13671875, + "learning_rate": 0.00199113832983129, + "loss": 0.1445, + "step": 5248 + }, + { + "epoch": 0.045563840591661534, + "grad_norm": 0.33984375, + "learning_rate": 0.0019911341630624094, + "loss": 0.1777, + "step": 5249 + }, + { + "epoch": 0.0455725210718657, + "grad_norm": 1.1328125, + "learning_rate": 0.001991129995318998, + "loss": 0.1992, + "step": 5250 + }, + { + "epoch": 0.045581201552069864, + "grad_norm": 0.337890625, + "learning_rate": 0.0019911258266010604, + "loss": 0.1826, + "step": 5251 + }, + { + "epoch": 0.04558988203227402, + "grad_norm": 0.1748046875, + "learning_rate": 0.001991121656908601, + "loss": 0.1553, + "step": 5252 + }, + { + "epoch": 0.04559856251247819, + "grad_norm": 0.271484375, + "learning_rate": 0.0019911174862416244, + "loss": 0.1553, + "step": 5253 + }, + { + "epoch": 0.04560724299268235, + "grad_norm": 0.2890625, + "learning_rate": 0.001991113314600135, + "loss": 0.1484, + "step": 5254 + }, + { + "epoch": 0.04561592347288652, + "grad_norm": 0.353515625, + "learning_rate": 0.0019911091419841376, + "loss": 0.1445, + "step": 5255 + }, + { + "epoch": 0.04562460395309068, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019911049683936366, + "loss": 0.1387, + "step": 5256 + }, + { + "epoch": 0.04563328443329485, + "grad_norm": 0.294921875, + "learning_rate": 0.0019911007938286365, + "loss": 0.1299, + "step": 5257 + }, + { + "epoch": 0.04564196491349901, + "grad_norm": 0.255859375, + "learning_rate": 0.001991096618289142, + "loss": 0.1367, + "step": 5258 + }, + { + "epoch": 0.04565064539370318, + "grad_norm": 0.5234375, + "learning_rate": 0.0019910924417751576, + "loss": 0.1445, + "step": 5259 + }, + { + "epoch": 0.04565932587390734, + "grad_norm": 1.0703125, + "learning_rate": 0.001991088264286688, + "loss": 0.3281, + "step": 5260 + }, + { + "epoch": 0.04566800635411151, + "grad_norm": 0.16796875, + "learning_rate": 0.001991084085823737, + "loss": 0.1543, + "step": 5261 + }, + { + "epoch": 0.04567668683431567, + "grad_norm": 0.359375, + "learning_rate": 0.001991079906386311, + "loss": 0.126, + "step": 5262 + }, + { + "epoch": 0.04568536731451984, + "grad_norm": 0.427734375, + "learning_rate": 0.0019910757259744127, + "loss": 0.1641, + "step": 5263 + }, + { + "epoch": 0.045694047794724, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019910715445880475, + "loss": 0.125, + "step": 5264 + }, + { + "epoch": 0.04570272827492817, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019910673622272196, + "loss": 0.1543, + "step": 5265 + }, + { + "epoch": 0.045711408755132334, + "grad_norm": 0.2060546875, + "learning_rate": 0.001991063178891934, + "loss": 0.1377, + "step": 5266 + }, + { + "epoch": 0.0457200892353365, + "grad_norm": 0.212890625, + "learning_rate": 0.0019910589945821952, + "loss": 0.1367, + "step": 5267 + }, + { + "epoch": 0.045728769715540664, + "grad_norm": 0.5078125, + "learning_rate": 0.0019910548092980074, + "loss": 0.2217, + "step": 5268 + }, + { + "epoch": 0.04573745019574483, + "grad_norm": 0.7890625, + "learning_rate": 0.001991050623039376, + "loss": 0.1289, + "step": 5269 + }, + { + "epoch": 0.045746130675948994, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019910464358063045, + "loss": 0.2227, + "step": 5270 + }, + { + "epoch": 0.04575481115615316, + "grad_norm": 0.447265625, + "learning_rate": 0.001991042247598798, + "loss": 0.1797, + "step": 5271 + }, + { + "epoch": 0.045763491636357324, + "grad_norm": 0.12890625, + "learning_rate": 0.0019910380584168613, + "loss": 0.1348, + "step": 5272 + }, + { + "epoch": 0.04577217211656149, + "grad_norm": 0.453125, + "learning_rate": 0.0019910338682604988, + "loss": 0.1406, + "step": 5273 + }, + { + "epoch": 0.045780852596765655, + "grad_norm": 0.265625, + "learning_rate": 0.0019910296771297148, + "loss": 0.1289, + "step": 5274 + }, + { + "epoch": 0.04578953307696982, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019910254850245145, + "loss": 0.126, + "step": 5275 + }, + { + "epoch": 0.045798213557173985, + "grad_norm": 0.1884765625, + "learning_rate": 0.001991021291944902, + "loss": 0.1245, + "step": 5276 + }, + { + "epoch": 0.04580689403737815, + "grad_norm": 0.4375, + "learning_rate": 0.001991017097890882, + "loss": 0.207, + "step": 5277 + }, + { + "epoch": 0.045815574517582315, + "grad_norm": 0.154296875, + "learning_rate": 0.0019910129028624587, + "loss": 0.2041, + "step": 5278 + }, + { + "epoch": 0.04582425499778648, + "grad_norm": 0.3125, + "learning_rate": 0.0019910087068596375, + "loss": 0.1797, + "step": 5279 + }, + { + "epoch": 0.045832935477990645, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019910045098824225, + "loss": 0.1562, + "step": 5280 + }, + { + "epoch": 0.04584161595819481, + "grad_norm": 0.2216796875, + "learning_rate": 0.001991000311930818, + "loss": 0.2148, + "step": 5281 + }, + { + "epoch": 0.045850296438398976, + "grad_norm": 0.13671875, + "learning_rate": 0.0019909961130048295, + "loss": 0.166, + "step": 5282 + }, + { + "epoch": 0.045858976918603134, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019909919131044606, + "loss": 0.1426, + "step": 5283 + }, + { + "epoch": 0.0458676573988073, + "grad_norm": 0.25390625, + "learning_rate": 0.0019909877122297167, + "loss": 0.1523, + "step": 5284 + }, + { + "epoch": 0.045876337879011464, + "grad_norm": 0.0791015625, + "learning_rate": 0.001990983510380602, + "loss": 0.1504, + "step": 5285 + }, + { + "epoch": 0.04588501835921563, + "grad_norm": 0.328125, + "learning_rate": 0.0019909793075571208, + "loss": 0.1768, + "step": 5286 + }, + { + "epoch": 0.045893698839419794, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019909751037592784, + "loss": 0.1084, + "step": 5287 + }, + { + "epoch": 0.04590237931962396, + "grad_norm": 0.251953125, + "learning_rate": 0.0019909708989870787, + "loss": 0.1885, + "step": 5288 + }, + { + "epoch": 0.045911059799828124, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019909666932405264, + "loss": 0.1582, + "step": 5289 + }, + { + "epoch": 0.04591974028003229, + "grad_norm": 0.59765625, + "learning_rate": 0.001990962486519627, + "loss": 0.1875, + "step": 5290 + }, + { + "epoch": 0.045928420760236455, + "grad_norm": 0.56640625, + "learning_rate": 0.001990958278824384, + "loss": 0.1309, + "step": 5291 + }, + { + "epoch": 0.04593710124044062, + "grad_norm": 1.2265625, + "learning_rate": 0.001990954070154802, + "loss": 0.1562, + "step": 5292 + }, + { + "epoch": 0.045945781720644785, + "grad_norm": 0.453125, + "learning_rate": 0.001990949860510887, + "loss": 0.1895, + "step": 5293 + }, + { + "epoch": 0.04595446220084895, + "grad_norm": 0.1142578125, + "learning_rate": 0.001990945649892642, + "loss": 0.1797, + "step": 5294 + }, + { + "epoch": 0.045963142681053115, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019909414383000724, + "loss": 0.1396, + "step": 5295 + }, + { + "epoch": 0.04597182316125728, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019909372257331825, + "loss": 0.1729, + "step": 5296 + }, + { + "epoch": 0.045980503641461445, + "grad_norm": 0.453125, + "learning_rate": 0.001990933012191977, + "loss": 0.1777, + "step": 5297 + }, + { + "epoch": 0.04598918412166561, + "grad_norm": 0.80078125, + "learning_rate": 0.001990928797676461, + "loss": 0.1543, + "step": 5298 + }, + { + "epoch": 0.045997864601869776, + "grad_norm": 0.181640625, + "learning_rate": 0.0019909245821866377, + "loss": 0.1582, + "step": 5299 + }, + { + "epoch": 0.04600654508207394, + "grad_norm": 0.283203125, + "learning_rate": 0.0019909203657225137, + "loss": 0.1338, + "step": 5300 + }, + { + "epoch": 0.046015225562278106, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019909161482840917, + "loss": 0.1387, + "step": 5301 + }, + { + "epoch": 0.04602390604248227, + "grad_norm": 0.126953125, + "learning_rate": 0.001990911929871378, + "loss": 0.2197, + "step": 5302 + }, + { + "epoch": 0.046032586522686436, + "grad_norm": 0.09130859375, + "learning_rate": 0.001990907710484376, + "loss": 0.1572, + "step": 5303 + }, + { + "epoch": 0.0460412670028906, + "grad_norm": 0.333984375, + "learning_rate": 0.0019909034901230906, + "loss": 0.1445, + "step": 5304 + }, + { + "epoch": 0.046049947483094766, + "grad_norm": 0.84375, + "learning_rate": 0.0019908992687875265, + "loss": 0.1738, + "step": 5305 + }, + { + "epoch": 0.04605862796329893, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019908950464776884, + "loss": 0.1641, + "step": 5306 + }, + { + "epoch": 0.0460673084435031, + "grad_norm": 0.4375, + "learning_rate": 0.001990890823193581, + "loss": 0.1445, + "step": 5307 + }, + { + "epoch": 0.04607598892370726, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019908865989352086, + "loss": 0.2021, + "step": 5308 + }, + { + "epoch": 0.04608466940391143, + "grad_norm": 0.67578125, + "learning_rate": 0.0019908823737025765, + "loss": 0.2285, + "step": 5309 + }, + { + "epoch": 0.04609334988411559, + "grad_norm": 0.3203125, + "learning_rate": 0.001990878147495688, + "loss": 0.1289, + "step": 5310 + }, + { + "epoch": 0.04610203036431976, + "grad_norm": 0.8046875, + "learning_rate": 0.001990873920314549, + "loss": 0.1572, + "step": 5311 + }, + { + "epoch": 0.04611071084452392, + "grad_norm": 0.2734375, + "learning_rate": 0.001990869692159164, + "loss": 0.2012, + "step": 5312 + }, + { + "epoch": 0.04611939132472809, + "grad_norm": 0.1796875, + "learning_rate": 0.0019908654630295367, + "loss": 0.2109, + "step": 5313 + }, + { + "epoch": 0.046128071804932246, + "grad_norm": 0.263671875, + "learning_rate": 0.0019908612329256727, + "loss": 0.1973, + "step": 5314 + }, + { + "epoch": 0.04613675228513641, + "grad_norm": 0.3515625, + "learning_rate": 0.0019908570018475763, + "loss": 0.127, + "step": 5315 + }, + { + "epoch": 0.046145432765340576, + "grad_norm": 0.447265625, + "learning_rate": 0.001990852769795252, + "loss": 0.165, + "step": 5316 + }, + { + "epoch": 0.04615411324554474, + "grad_norm": 0.314453125, + "learning_rate": 0.001990848536768704, + "loss": 0.1226, + "step": 5317 + }, + { + "epoch": 0.046162793725748906, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019908443027679378, + "loss": 0.1807, + "step": 5318 + }, + { + "epoch": 0.04617147420595307, + "grad_norm": 0.130859375, + "learning_rate": 0.0019908400677929577, + "loss": 0.1504, + "step": 5319 + }, + { + "epoch": 0.046180154686157236, + "grad_norm": 0.1728515625, + "learning_rate": 0.001990835831843768, + "loss": 0.1426, + "step": 5320 + }, + { + "epoch": 0.0461888351663614, + "grad_norm": 0.251953125, + "learning_rate": 0.001990831594920374, + "loss": 0.1318, + "step": 5321 + }, + { + "epoch": 0.046197515646565566, + "grad_norm": 0.45703125, + "learning_rate": 0.0019908273570227795, + "loss": 0.1406, + "step": 5322 + }, + { + "epoch": 0.04620619612676973, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019908231181509904, + "loss": 0.1953, + "step": 5323 + }, + { + "epoch": 0.0462148766069739, + "grad_norm": 0.60546875, + "learning_rate": 0.00199081887830501, + "loss": 0.1787, + "step": 5324 + }, + { + "epoch": 0.04622355708717806, + "grad_norm": 0.52734375, + "learning_rate": 0.001990814637484843, + "loss": 0.1187, + "step": 5325 + }, + { + "epoch": 0.04623223756738223, + "grad_norm": 0.2158203125, + "learning_rate": 0.001990810395690495, + "loss": 0.168, + "step": 5326 + }, + { + "epoch": 0.04624091804758639, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019908061529219706, + "loss": 0.1729, + "step": 5327 + }, + { + "epoch": 0.04624959852779056, + "grad_norm": 0.25, + "learning_rate": 0.0019908019091792735, + "loss": 0.2041, + "step": 5328 + }, + { + "epoch": 0.04625827900799472, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019907976644624086, + "loss": 0.1187, + "step": 5329 + }, + { + "epoch": 0.04626695948819889, + "grad_norm": 0.275390625, + "learning_rate": 0.0019907934187713807, + "loss": 0.1787, + "step": 5330 + }, + { + "epoch": 0.04627563996840305, + "grad_norm": 0.2578125, + "learning_rate": 0.001990789172106195, + "loss": 0.1426, + "step": 5331 + }, + { + "epoch": 0.04628432044860722, + "grad_norm": 0.55078125, + "learning_rate": 0.0019907849244668553, + "loss": 0.1816, + "step": 5332 + }, + { + "epoch": 0.04629300092881138, + "grad_norm": 0.271484375, + "learning_rate": 0.001990780675853367, + "loss": 0.1523, + "step": 5333 + }, + { + "epoch": 0.04630168140901555, + "grad_norm": 0.458984375, + "learning_rate": 0.001990776426265734, + "loss": 0.1914, + "step": 5334 + }, + { + "epoch": 0.04631036188921971, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019907721757039614, + "loss": 0.1582, + "step": 5335 + }, + { + "epoch": 0.04631904236942388, + "grad_norm": 0.6171875, + "learning_rate": 0.0019907679241680538, + "loss": 0.1543, + "step": 5336 + }, + { + "epoch": 0.04632772284962804, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019907636716580152, + "loss": 0.1143, + "step": 5337 + }, + { + "epoch": 0.04633640332983221, + "grad_norm": 0.275390625, + "learning_rate": 0.0019907594181738514, + "loss": 0.1348, + "step": 5338 + }, + { + "epoch": 0.046345083810036374, + "grad_norm": 0.2578125, + "learning_rate": 0.001990755163715566, + "loss": 0.126, + "step": 5339 + }, + { + "epoch": 0.04635376429024054, + "grad_norm": 0.11376953125, + "learning_rate": 0.001990750908283165, + "loss": 0.1719, + "step": 5340 + }, + { + "epoch": 0.046362444770444704, + "grad_norm": 0.78125, + "learning_rate": 0.0019907466518766515, + "loss": 0.1543, + "step": 5341 + }, + { + "epoch": 0.04637112525064887, + "grad_norm": 0.2734375, + "learning_rate": 0.001990742394496031, + "loss": 0.105, + "step": 5342 + }, + { + "epoch": 0.046379805730853034, + "grad_norm": 0.30078125, + "learning_rate": 0.001990738136141308, + "loss": 0.1338, + "step": 5343 + }, + { + "epoch": 0.0463884862110572, + "grad_norm": 0.279296875, + "learning_rate": 0.0019907338768124874, + "loss": 0.1533, + "step": 5344 + }, + { + "epoch": 0.04639716669126136, + "grad_norm": 0.3203125, + "learning_rate": 0.001990729616509573, + "loss": 0.1455, + "step": 5345 + }, + { + "epoch": 0.04640584717146552, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019907253552325704, + "loss": 0.1182, + "step": 5346 + }, + { + "epoch": 0.04641452765166969, + "grad_norm": 0.7109375, + "learning_rate": 0.001990721092981484, + "loss": 0.1758, + "step": 5347 + }, + { + "epoch": 0.04642320813187385, + "grad_norm": 0.38671875, + "learning_rate": 0.0019907168297563184, + "loss": 0.1699, + "step": 5348 + }, + { + "epoch": 0.04643188861207802, + "grad_norm": 0.921875, + "learning_rate": 0.0019907125655570785, + "loss": 0.1699, + "step": 5349 + }, + { + "epoch": 0.04644056909228218, + "grad_norm": 0.396484375, + "learning_rate": 0.0019907083003837685, + "loss": 0.1396, + "step": 5350 + }, + { + "epoch": 0.04644924957248635, + "grad_norm": 0.337890625, + "learning_rate": 0.001990704034236393, + "loss": 0.1406, + "step": 5351 + }, + { + "epoch": 0.04645793005269051, + "grad_norm": 1.09375, + "learning_rate": 0.0019906997671149573, + "loss": 0.1602, + "step": 5352 + }, + { + "epoch": 0.04646661053289468, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019906954990194657, + "loss": 0.1494, + "step": 5353 + }, + { + "epoch": 0.04647529101309884, + "grad_norm": 0.1005859375, + "learning_rate": 0.001990691229949923, + "loss": 0.1816, + "step": 5354 + }, + { + "epoch": 0.04648397149330301, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019906869599063334, + "loss": 0.1357, + "step": 5355 + }, + { + "epoch": 0.046492651973507174, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019906826888887023, + "loss": 0.127, + "step": 5356 + }, + { + "epoch": 0.04650133245371134, + "grad_norm": 0.5234375, + "learning_rate": 0.0019906784168970335, + "loss": 0.1699, + "step": 5357 + }, + { + "epoch": 0.046510012933915504, + "grad_norm": 0.66796875, + "learning_rate": 0.0019906741439313326, + "loss": 0.1641, + "step": 5358 + }, + { + "epoch": 0.04651869341411967, + "grad_norm": 0.44140625, + "learning_rate": 0.0019906698699916035, + "loss": 0.1553, + "step": 5359 + }, + { + "epoch": 0.046527373894323834, + "grad_norm": 0.150390625, + "learning_rate": 0.0019906655950778515, + "loss": 0.1514, + "step": 5360 + }, + { + "epoch": 0.046536054374528, + "grad_norm": 0.1728515625, + "learning_rate": 0.001990661319190081, + "loss": 0.166, + "step": 5361 + }, + { + "epoch": 0.046544734854732164, + "grad_norm": 0.166015625, + "learning_rate": 0.0019906570423282965, + "loss": 0.1328, + "step": 5362 + }, + { + "epoch": 0.04655341533493633, + "grad_norm": 0.37890625, + "learning_rate": 0.0019906527644925026, + "loss": 0.0996, + "step": 5363 + }, + { + "epoch": 0.046562095815140495, + "grad_norm": 0.41796875, + "learning_rate": 0.001990648485682705, + "loss": 0.1592, + "step": 5364 + }, + { + "epoch": 0.04657077629534466, + "grad_norm": 0.2265625, + "learning_rate": 0.001990644205898907, + "loss": 0.1533, + "step": 5365 + }, + { + "epoch": 0.046579456775548825, + "grad_norm": 0.236328125, + "learning_rate": 0.001990639925141114, + "loss": 0.1387, + "step": 5366 + }, + { + "epoch": 0.04658813725575299, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019906356434093303, + "loss": 0.1118, + "step": 5367 + }, + { + "epoch": 0.046596817735957155, + "grad_norm": 0.123046875, + "learning_rate": 0.0019906313607035613, + "loss": 0.1875, + "step": 5368 + }, + { + "epoch": 0.04660549821616132, + "grad_norm": 0.0771484375, + "learning_rate": 0.001990627077023811, + "loss": 0.127, + "step": 5369 + }, + { + "epoch": 0.046614178696365485, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019906227923700845, + "loss": 0.1504, + "step": 5370 + }, + { + "epoch": 0.04662285917656965, + "grad_norm": 0.076171875, + "learning_rate": 0.001990618506742386, + "loss": 0.1387, + "step": 5371 + }, + { + "epoch": 0.046631539656773816, + "grad_norm": 0.27734375, + "learning_rate": 0.0019906142201407207, + "loss": 0.1719, + "step": 5372 + }, + { + "epoch": 0.04664022013697798, + "grad_norm": 0.08740234375, + "learning_rate": 0.001990609932565093, + "loss": 0.1719, + "step": 5373 + }, + { + "epoch": 0.046648900617182146, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019906056440155075, + "loss": 0.1152, + "step": 5374 + }, + { + "epoch": 0.04665758109738631, + "grad_norm": 0.1962890625, + "learning_rate": 0.001990601354491969, + "loss": 0.1201, + "step": 5375 + }, + { + "epoch": 0.04666626157759047, + "grad_norm": 0.134765625, + "learning_rate": 0.0019905970639944827, + "loss": 0.1602, + "step": 5376 + }, + { + "epoch": 0.046674942057794634, + "grad_norm": 0.3203125, + "learning_rate": 0.0019905927725230523, + "loss": 0.207, + "step": 5377 + }, + { + "epoch": 0.0466836225379988, + "grad_norm": 0.322265625, + "learning_rate": 0.001990588480077683, + "loss": 0.127, + "step": 5378 + }, + { + "epoch": 0.046692303018202964, + "grad_norm": 0.283203125, + "learning_rate": 0.0019905841866583803, + "loss": 0.1582, + "step": 5379 + }, + { + "epoch": 0.04670098349840713, + "grad_norm": 0.283203125, + "learning_rate": 0.0019905798922651473, + "loss": 0.1807, + "step": 5380 + }, + { + "epoch": 0.046709663978611295, + "grad_norm": 0.12890625, + "learning_rate": 0.0019905755968979903, + "loss": 0.1699, + "step": 5381 + }, + { + "epoch": 0.04671834445881546, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019905713005569123, + "loss": 0.1426, + "step": 5382 + }, + { + "epoch": 0.046727024939019625, + "grad_norm": 0.25, + "learning_rate": 0.0019905670032419193, + "loss": 0.2109, + "step": 5383 + }, + { + "epoch": 0.04673570541922379, + "grad_norm": 0.703125, + "learning_rate": 0.0019905627049530156, + "loss": 0.1191, + "step": 5384 + }, + { + "epoch": 0.046744385899427955, + "grad_norm": 0.10009765625, + "learning_rate": 0.001990558405690206, + "loss": 0.1621, + "step": 5385 + }, + { + "epoch": 0.04675306637963212, + "grad_norm": 0.1806640625, + "learning_rate": 0.001990554105453495, + "loss": 0.1709, + "step": 5386 + }, + { + "epoch": 0.046761746859836285, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019905498042428874, + "loss": 0.1309, + "step": 5387 + }, + { + "epoch": 0.04677042734004045, + "grad_norm": 0.125, + "learning_rate": 0.001990545502058388, + "loss": 0.126, + "step": 5388 + }, + { + "epoch": 0.046779107820244616, + "grad_norm": 0.2578125, + "learning_rate": 0.0019905411989000017, + "loss": 0.1104, + "step": 5389 + }, + { + "epoch": 0.04678778830044878, + "grad_norm": 0.1171875, + "learning_rate": 0.0019905368947677328, + "loss": 0.1953, + "step": 5390 + }, + { + "epoch": 0.046796468780652946, + "grad_norm": 0.244140625, + "learning_rate": 0.001990532589661586, + "loss": 0.1611, + "step": 5391 + }, + { + "epoch": 0.04680514926085711, + "grad_norm": 0.765625, + "learning_rate": 0.001990528283581566, + "loss": 0.1533, + "step": 5392 + }, + { + "epoch": 0.046813829741061276, + "grad_norm": 0.13671875, + "learning_rate": 0.001990523976527678, + "loss": 0.1514, + "step": 5393 + }, + { + "epoch": 0.04682251022126544, + "grad_norm": 0.142578125, + "learning_rate": 0.0019905196684999256, + "loss": 0.1094, + "step": 5394 + }, + { + "epoch": 0.046831190701469606, + "grad_norm": 0.166015625, + "learning_rate": 0.001990515359498315, + "loss": 0.1602, + "step": 5395 + }, + { + "epoch": 0.04683987118167377, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019905110495228497, + "loss": 0.1777, + "step": 5396 + }, + { + "epoch": 0.04684855166187794, + "grad_norm": 0.1298828125, + "learning_rate": 0.001990506738573535, + "loss": 0.1992, + "step": 5397 + }, + { + "epoch": 0.0468572321420821, + "grad_norm": 0.08203125, + "learning_rate": 0.001990502426650376, + "loss": 0.1484, + "step": 5398 + }, + { + "epoch": 0.04686591262228627, + "grad_norm": 0.3671875, + "learning_rate": 0.0019904981137533763, + "loss": 0.1543, + "step": 5399 + }, + { + "epoch": 0.04687459310249043, + "grad_norm": 0.2177734375, + "learning_rate": 0.001990493799882542, + "loss": 0.1621, + "step": 5400 + }, + { + "epoch": 0.0468832735826946, + "grad_norm": 1.28125, + "learning_rate": 0.0019904894850378764, + "loss": 0.2236, + "step": 5401 + }, + { + "epoch": 0.04689195406289876, + "grad_norm": 0.53515625, + "learning_rate": 0.001990485169219385, + "loss": 0.165, + "step": 5402 + }, + { + "epoch": 0.04690063454310293, + "grad_norm": 0.296875, + "learning_rate": 0.0019904808524270727, + "loss": 0.1621, + "step": 5403 + }, + { + "epoch": 0.04690931502330709, + "grad_norm": 0.609375, + "learning_rate": 0.0019904765346609433, + "loss": 0.1348, + "step": 5404 + }, + { + "epoch": 0.04691799550351126, + "grad_norm": 2.546875, + "learning_rate": 0.001990472215921003, + "loss": 0.5156, + "step": 5405 + }, + { + "epoch": 0.046926675983715416, + "grad_norm": 0.171875, + "learning_rate": 0.001990467896207255, + "loss": 0.1162, + "step": 5406 + }, + { + "epoch": 0.04693535646391958, + "grad_norm": 0.30078125, + "learning_rate": 0.001990463575519705, + "loss": 0.1777, + "step": 5407 + }, + { + "epoch": 0.046944036944123746, + "grad_norm": 0.109375, + "learning_rate": 0.001990459253858357, + "loss": 0.1641, + "step": 5408 + }, + { + "epoch": 0.04695271742432791, + "grad_norm": 0.396484375, + "learning_rate": 0.0019904549312232166, + "loss": 0.1709, + "step": 5409 + }, + { + "epoch": 0.046961397904532076, + "grad_norm": 0.62890625, + "learning_rate": 0.001990450607614288, + "loss": 0.1523, + "step": 5410 + }, + { + "epoch": 0.04697007838473624, + "grad_norm": 0.384765625, + "learning_rate": 0.001990446283031576, + "loss": 0.166, + "step": 5411 + }, + { + "epoch": 0.046978758864940406, + "grad_norm": 0.34765625, + "learning_rate": 0.001990441957475085, + "loss": 0.1133, + "step": 5412 + }, + { + "epoch": 0.04698743934514457, + "grad_norm": 1.0625, + "learning_rate": 0.00199043763094482, + "loss": 0.1533, + "step": 5413 + }, + { + "epoch": 0.04699611982534874, + "grad_norm": 0.294921875, + "learning_rate": 0.0019904333034407866, + "loss": 0.126, + "step": 5414 + }, + { + "epoch": 0.0470048003055529, + "grad_norm": 0.28125, + "learning_rate": 0.001990428974962988, + "loss": 0.1338, + "step": 5415 + }, + { + "epoch": 0.04701348078575707, + "grad_norm": 0.07470703125, + "learning_rate": 0.00199042464551143, + "loss": 0.1611, + "step": 5416 + }, + { + "epoch": 0.04702216126596123, + "grad_norm": 0.283203125, + "learning_rate": 0.0019904203150861166, + "loss": 0.1357, + "step": 5417 + }, + { + "epoch": 0.0470308417461654, + "grad_norm": 0.34375, + "learning_rate": 0.0019904159836870534, + "loss": 0.1387, + "step": 5418 + }, + { + "epoch": 0.04703952222636956, + "grad_norm": 0.091796875, + "learning_rate": 0.001990411651314244, + "loss": 0.1289, + "step": 5419 + }, + { + "epoch": 0.04704820270657373, + "grad_norm": 0.7109375, + "learning_rate": 0.0019904073179676943, + "loss": 0.1289, + "step": 5420 + }, + { + "epoch": 0.04705688318677789, + "grad_norm": 0.3984375, + "learning_rate": 0.001990402983647409, + "loss": 0.1006, + "step": 5421 + }, + { + "epoch": 0.04706556366698206, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019903986483533914, + "loss": 0.1309, + "step": 5422 + }, + { + "epoch": 0.04707424414718622, + "grad_norm": 0.107421875, + "learning_rate": 0.0019903943120856476, + "loss": 0.127, + "step": 5423 + }, + { + "epoch": 0.04708292462739039, + "grad_norm": 0.271484375, + "learning_rate": 0.001990389974844182, + "loss": 0.1406, + "step": 5424 + }, + { + "epoch": 0.04709160510759455, + "grad_norm": 0.494140625, + "learning_rate": 0.0019903856366289994, + "loss": 0.1089, + "step": 5425 + }, + { + "epoch": 0.04710028558779872, + "grad_norm": 0.380859375, + "learning_rate": 0.001990381297440104, + "loss": 0.1875, + "step": 5426 + }, + { + "epoch": 0.04710896606800288, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019903769572775015, + "loss": 0.1602, + "step": 5427 + }, + { + "epoch": 0.04711764654820705, + "grad_norm": 0.4921875, + "learning_rate": 0.0019903726161411956, + "loss": 0.166, + "step": 5428 + }, + { + "epoch": 0.04712632702841121, + "grad_norm": 0.259765625, + "learning_rate": 0.001990368274031192, + "loss": 0.1455, + "step": 5429 + }, + { + "epoch": 0.04713500750861538, + "grad_norm": 0.18359375, + "learning_rate": 0.001990363930947495, + "loss": 0.1621, + "step": 5430 + }, + { + "epoch": 0.047143687988819544, + "grad_norm": 0.259765625, + "learning_rate": 0.0019903595868901096, + "loss": 0.1699, + "step": 5431 + }, + { + "epoch": 0.04715236846902371, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019903552418590402, + "loss": 0.1328, + "step": 5432 + }, + { + "epoch": 0.047161048949227874, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019903508958542915, + "loss": 0.1299, + "step": 5433 + }, + { + "epoch": 0.04716972942943204, + "grad_norm": 0.23828125, + "learning_rate": 0.0019903465488758684, + "loss": 0.1699, + "step": 5434 + }, + { + "epoch": 0.047178409909636204, + "grad_norm": 0.09130859375, + "learning_rate": 0.001990342200923776, + "loss": 0.1758, + "step": 5435 + }, + { + "epoch": 0.04718709038984037, + "grad_norm": 0.5078125, + "learning_rate": 0.0019903378519980186, + "loss": 0.1729, + "step": 5436 + }, + { + "epoch": 0.04719577087004453, + "grad_norm": 0.1533203125, + "learning_rate": 0.001990333502098601, + "loss": 0.1738, + "step": 5437 + }, + { + "epoch": 0.04720445135024869, + "grad_norm": 0.3671875, + "learning_rate": 0.001990329151225528, + "loss": 0.2246, + "step": 5438 + }, + { + "epoch": 0.04721313183045286, + "grad_norm": 0.314453125, + "learning_rate": 0.0019903247993788044, + "loss": 0.1895, + "step": 5439 + }, + { + "epoch": 0.04722181231065702, + "grad_norm": 0.177734375, + "learning_rate": 0.001990320446558435, + "loss": 0.1426, + "step": 5440 + }, + { + "epoch": 0.04723049279086119, + "grad_norm": 0.216796875, + "learning_rate": 0.001990316092764425, + "loss": 0.2334, + "step": 5441 + }, + { + "epoch": 0.04723917327106535, + "grad_norm": 0.146484375, + "learning_rate": 0.001990311737996778, + "loss": 0.1533, + "step": 5442 + }, + { + "epoch": 0.04724785375126952, + "grad_norm": 0.33203125, + "learning_rate": 0.0019903073822555004, + "loss": 0.1216, + "step": 5443 + }, + { + "epoch": 0.04725653423147368, + "grad_norm": 0.341796875, + "learning_rate": 0.001990303025540595, + "loss": 0.1152, + "step": 5444 + }, + { + "epoch": 0.04726521471167785, + "grad_norm": 0.0908203125, + "learning_rate": 0.001990298667852068, + "loss": 0.1562, + "step": 5445 + }, + { + "epoch": 0.047273895191882014, + "grad_norm": 0.63671875, + "learning_rate": 0.0019902943091899238, + "loss": 0.1631, + "step": 5446 + }, + { + "epoch": 0.04728257567208618, + "grad_norm": 0.70703125, + "learning_rate": 0.001990289949554167, + "loss": 0.1406, + "step": 5447 + }, + { + "epoch": 0.047291256152290344, + "grad_norm": 0.0703125, + "learning_rate": 0.0019902855889448027, + "loss": 0.1367, + "step": 5448 + }, + { + "epoch": 0.04729993663249451, + "grad_norm": 0.462890625, + "learning_rate": 0.0019902812273618355, + "loss": 0.1973, + "step": 5449 + }, + { + "epoch": 0.047308617112698674, + "grad_norm": 1.9375, + "learning_rate": 0.0019902768648052695, + "loss": 0.2598, + "step": 5450 + }, + { + "epoch": 0.04731729759290284, + "grad_norm": 0.44921875, + "learning_rate": 0.001990272501275111, + "loss": 0.1592, + "step": 5451 + }, + { + "epoch": 0.047325978073107004, + "grad_norm": 0.171875, + "learning_rate": 0.0019902681367713632, + "loss": 0.2188, + "step": 5452 + }, + { + "epoch": 0.04733465855331117, + "grad_norm": 0.1669921875, + "learning_rate": 0.001990263771294032, + "loss": 0.1416, + "step": 5453 + }, + { + "epoch": 0.047343339033515335, + "grad_norm": 0.275390625, + "learning_rate": 0.001990259404843121, + "loss": 0.1582, + "step": 5454 + }, + { + "epoch": 0.0473520195137195, + "grad_norm": 0.609375, + "learning_rate": 0.001990255037418636, + "loss": 0.1885, + "step": 5455 + }, + { + "epoch": 0.047360699993923665, + "grad_norm": 0.08544921875, + "learning_rate": 0.001990250669020582, + "loss": 0.1045, + "step": 5456 + }, + { + "epoch": 0.04736938047412783, + "grad_norm": 0.77734375, + "learning_rate": 0.001990246299648963, + "loss": 0.166, + "step": 5457 + }, + { + "epoch": 0.047378060954331995, + "grad_norm": 0.58203125, + "learning_rate": 0.001990241929303784, + "loss": 0.1338, + "step": 5458 + }, + { + "epoch": 0.04738674143453616, + "grad_norm": 0.478515625, + "learning_rate": 0.0019902375579850494, + "loss": 0.1436, + "step": 5459 + }, + { + "epoch": 0.047395421914740325, + "grad_norm": 0.125, + "learning_rate": 0.0019902331856927647, + "loss": 0.1562, + "step": 5460 + }, + { + "epoch": 0.04740410239494449, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019902288124269345, + "loss": 0.2129, + "step": 5461 + }, + { + "epoch": 0.047412782875148655, + "grad_norm": 0.087890625, + "learning_rate": 0.0019902244381875636, + "loss": 0.1426, + "step": 5462 + }, + { + "epoch": 0.04742146335535282, + "grad_norm": 0.287109375, + "learning_rate": 0.001990220062974656, + "loss": 0.1543, + "step": 5463 + }, + { + "epoch": 0.047430143835556986, + "grad_norm": 0.5703125, + "learning_rate": 0.0019902156867882175, + "loss": 0.1426, + "step": 5464 + }, + { + "epoch": 0.04743882431576115, + "grad_norm": 0.44921875, + "learning_rate": 0.0019902113096282527, + "loss": 0.1426, + "step": 5465 + }, + { + "epoch": 0.047447504795965316, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019902069314947654, + "loss": 0.1582, + "step": 5466 + }, + { + "epoch": 0.04745618527616948, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019902025523877618, + "loss": 0.1406, + "step": 5467 + }, + { + "epoch": 0.04746486575637364, + "grad_norm": 0.2353515625, + "learning_rate": 0.001990198172307246, + "loss": 0.1729, + "step": 5468 + }, + { + "epoch": 0.047473546236577804, + "grad_norm": 0.1806640625, + "learning_rate": 0.001990193791253223, + "loss": 0.1279, + "step": 5469 + }, + { + "epoch": 0.04748222671678197, + "grad_norm": 0.314453125, + "learning_rate": 0.001990189409225697, + "loss": 0.1611, + "step": 5470 + }, + { + "epoch": 0.047490907196986135, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019901850262246737, + "loss": 0.124, + "step": 5471 + }, + { + "epoch": 0.0474995876771903, + "grad_norm": 0.4453125, + "learning_rate": 0.0019901806422501574, + "loss": 0.1348, + "step": 5472 + }, + { + "epoch": 0.047508268157394465, + "grad_norm": 0.08935546875, + "learning_rate": 0.001990176257302153, + "loss": 0.1318, + "step": 5473 + }, + { + "epoch": 0.04751694863759863, + "grad_norm": 0.359375, + "learning_rate": 0.001990171871380665, + "loss": 0.1816, + "step": 5474 + }, + { + "epoch": 0.047525629117802795, + "grad_norm": 0.359375, + "learning_rate": 0.001990167484485698, + "loss": 0.1562, + "step": 5475 + }, + { + "epoch": 0.04753430959800696, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019901630966172577, + "loss": 0.1416, + "step": 5476 + }, + { + "epoch": 0.047542990078211125, + "grad_norm": 0.76171875, + "learning_rate": 0.0019901587077753484, + "loss": 0.165, + "step": 5477 + }, + { + "epoch": 0.04755167055841529, + "grad_norm": 0.421875, + "learning_rate": 0.0019901543179599744, + "loss": 0.1641, + "step": 5478 + }, + { + "epoch": 0.047560351038619456, + "grad_norm": 0.703125, + "learning_rate": 0.0019901499271711416, + "loss": 0.1738, + "step": 5479 + }, + { + "epoch": 0.04756903151882362, + "grad_norm": 0.095703125, + "learning_rate": 0.0019901455354088544, + "loss": 0.1572, + "step": 5480 + }, + { + "epoch": 0.047577711999027786, + "grad_norm": 0.177734375, + "learning_rate": 0.0019901411426731168, + "loss": 0.1631, + "step": 5481 + }, + { + "epoch": 0.04758639247923195, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019901367489639347, + "loss": 0.166, + "step": 5482 + }, + { + "epoch": 0.047595072959436116, + "grad_norm": 0.1728515625, + "learning_rate": 0.001990132354281312, + "loss": 0.1855, + "step": 5483 + }, + { + "epoch": 0.04760375343964028, + "grad_norm": 0.23046875, + "learning_rate": 0.0019901279586252542, + "loss": 0.1738, + "step": 5484 + }, + { + "epoch": 0.047612433919844446, + "grad_norm": 0.416015625, + "learning_rate": 0.001990123561995766, + "loss": 0.1543, + "step": 5485 + }, + { + "epoch": 0.04762111440004861, + "grad_norm": 0.68359375, + "learning_rate": 0.0019901191643928516, + "loss": 0.1904, + "step": 5486 + }, + { + "epoch": 0.047629794880252777, + "grad_norm": 0.64453125, + "learning_rate": 0.001990114765816517, + "loss": 0.1562, + "step": 5487 + }, + { + "epoch": 0.04763847536045694, + "grad_norm": 0.34375, + "learning_rate": 0.0019901103662667654, + "loss": 0.1289, + "step": 5488 + }, + { + "epoch": 0.04764715584066111, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019901059657436025, + "loss": 0.1904, + "step": 5489 + }, + { + "epoch": 0.04765583632086527, + "grad_norm": 0.10986328125, + "learning_rate": 0.001990101564247034, + "loss": 0.1787, + "step": 5490 + }, + { + "epoch": 0.04766451680106944, + "grad_norm": 0.4140625, + "learning_rate": 0.001990097161777063, + "loss": 0.1758, + "step": 5491 + }, + { + "epoch": 0.0476731972812736, + "grad_norm": 0.8125, + "learning_rate": 0.0019900927583336955, + "loss": 0.1396, + "step": 5492 + }, + { + "epoch": 0.04768187776147777, + "grad_norm": 0.1875, + "learning_rate": 0.001990088353916936, + "loss": 0.165, + "step": 5493 + }, + { + "epoch": 0.04769055824168193, + "grad_norm": 0.244140625, + "learning_rate": 0.0019900839485267885, + "loss": 0.127, + "step": 5494 + }, + { + "epoch": 0.0476992387218861, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019900795421632592, + "loss": 0.1777, + "step": 5495 + }, + { + "epoch": 0.04770791920209026, + "grad_norm": 0.44140625, + "learning_rate": 0.0019900751348263523, + "loss": 0.1406, + "step": 5496 + }, + { + "epoch": 0.04771659968229443, + "grad_norm": 0.318359375, + "learning_rate": 0.0019900707265160726, + "loss": 0.1152, + "step": 5497 + }, + { + "epoch": 0.04772528016249859, + "grad_norm": 0.083984375, + "learning_rate": 0.0019900663172324248, + "loss": 0.1357, + "step": 5498 + }, + { + "epoch": 0.04773396064270275, + "grad_norm": 0.28515625, + "learning_rate": 0.0019900619069754136, + "loss": 0.1748, + "step": 5499 + }, + { + "epoch": 0.047742641122906916, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019900574957450443, + "loss": 0.1226, + "step": 5500 + }, + { + "epoch": 0.04775132160311108, + "grad_norm": 0.3359375, + "learning_rate": 0.0019900530835413217, + "loss": 0.1416, + "step": 5501 + }, + { + "epoch": 0.047760002083315246, + "grad_norm": 0.451171875, + "learning_rate": 0.00199004867036425, + "loss": 0.1309, + "step": 5502 + }, + { + "epoch": 0.04776868256351941, + "grad_norm": 0.15625, + "learning_rate": 0.001990044256213835, + "loss": 0.1367, + "step": 5503 + }, + { + "epoch": 0.04777736304372358, + "grad_norm": 0.60546875, + "learning_rate": 0.0019900398410900807, + "loss": 0.166, + "step": 5504 + }, + { + "epoch": 0.04778604352392774, + "grad_norm": 0.34375, + "learning_rate": 0.001990035424992992, + "loss": 0.1641, + "step": 5505 + }, + { + "epoch": 0.04779472400413191, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019900310079225742, + "loss": 0.1797, + "step": 5506 + }, + { + "epoch": 0.04780340448433607, + "grad_norm": 0.36328125, + "learning_rate": 0.0019900265898788318, + "loss": 0.127, + "step": 5507 + }, + { + "epoch": 0.04781208496454024, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019900221708617698, + "loss": 0.1206, + "step": 5508 + }, + { + "epoch": 0.0478207654447444, + "grad_norm": 0.328125, + "learning_rate": 0.0019900177508713926, + "loss": 0.1504, + "step": 5509 + }, + { + "epoch": 0.04782944592494857, + "grad_norm": 0.7734375, + "learning_rate": 0.001990013329907706, + "loss": 0.1973, + "step": 5510 + }, + { + "epoch": 0.04783812640515273, + "grad_norm": 0.56640625, + "learning_rate": 0.0019900089079707135, + "loss": 0.1465, + "step": 5511 + }, + { + "epoch": 0.0478468068853569, + "grad_norm": 0.275390625, + "learning_rate": 0.0019900044850604207, + "loss": 0.1582, + "step": 5512 + }, + { + "epoch": 0.04785548736556106, + "grad_norm": 0.609375, + "learning_rate": 0.0019900000611768327, + "loss": 0.1367, + "step": 5513 + }, + { + "epoch": 0.04786416784576523, + "grad_norm": 0.21875, + "learning_rate": 0.001989995636319954, + "loss": 0.1099, + "step": 5514 + }, + { + "epoch": 0.04787284832596939, + "grad_norm": 0.31640625, + "learning_rate": 0.001989991210489789, + "loss": 0.1426, + "step": 5515 + }, + { + "epoch": 0.04788152880617356, + "grad_norm": 0.26171875, + "learning_rate": 0.0019899867836863433, + "loss": 0.1475, + "step": 5516 + }, + { + "epoch": 0.04789020928637772, + "grad_norm": 0.138671875, + "learning_rate": 0.001989982355909621, + "loss": 0.1543, + "step": 5517 + }, + { + "epoch": 0.04789888976658189, + "grad_norm": 0.6484375, + "learning_rate": 0.001989977927159628, + "loss": 0.1572, + "step": 5518 + }, + { + "epoch": 0.04790757024678605, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019899734974363685, + "loss": 0.125, + "step": 5519 + }, + { + "epoch": 0.04791625072699022, + "grad_norm": 0.51953125, + "learning_rate": 0.001989969066739847, + "loss": 0.1367, + "step": 5520 + }, + { + "epoch": 0.047924931207194384, + "grad_norm": 0.130859375, + "learning_rate": 0.001989964635070069, + "loss": 0.1582, + "step": 5521 + }, + { + "epoch": 0.04793361168739855, + "grad_norm": 0.5859375, + "learning_rate": 0.001989960202427039, + "loss": 0.1533, + "step": 5522 + }, + { + "epoch": 0.047942292167602714, + "grad_norm": 0.1748046875, + "learning_rate": 0.001989955768810762, + "loss": 0.106, + "step": 5523 + }, + { + "epoch": 0.04795097264780688, + "grad_norm": 0.7265625, + "learning_rate": 0.001989951334221242, + "loss": 0.1211, + "step": 5524 + }, + { + "epoch": 0.047959653128011044, + "grad_norm": 0.4453125, + "learning_rate": 0.0019899468986584855, + "loss": 0.1582, + "step": 5525 + }, + { + "epoch": 0.04796833360821521, + "grad_norm": 0.0751953125, + "learning_rate": 0.001989942462122496, + "loss": 0.1387, + "step": 5526 + }, + { + "epoch": 0.047977014088419374, + "grad_norm": 0.126953125, + "learning_rate": 0.001989938024613279, + "loss": 0.1465, + "step": 5527 + }, + { + "epoch": 0.04798569456862354, + "grad_norm": 0.126953125, + "learning_rate": 0.001989933586130839, + "loss": 0.1748, + "step": 5528 + }, + { + "epoch": 0.047994375048827705, + "grad_norm": 0.212890625, + "learning_rate": 0.0019899291466751803, + "loss": 0.1318, + "step": 5529 + }, + { + "epoch": 0.04800305552903186, + "grad_norm": 0.380859375, + "learning_rate": 0.0019899247062463094, + "loss": 0.1455, + "step": 5530 + }, + { + "epoch": 0.04801173600923603, + "grad_norm": 0.416015625, + "learning_rate": 0.00198992026484423, + "loss": 0.1328, + "step": 5531 + }, + { + "epoch": 0.04802041648944019, + "grad_norm": 0.294921875, + "learning_rate": 0.001989915822468947, + "loss": 0.1514, + "step": 5532 + }, + { + "epoch": 0.04802909696964436, + "grad_norm": 0.19921875, + "learning_rate": 0.0019899113791204655, + "loss": 0.1348, + "step": 5533 + }, + { + "epoch": 0.04803777744984852, + "grad_norm": 0.376953125, + "learning_rate": 0.0019899069347987905, + "loss": 0.1533, + "step": 5534 + }, + { + "epoch": 0.04804645793005269, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019899024895039265, + "loss": 0.1279, + "step": 5535 + }, + { + "epoch": 0.048055138410256854, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019898980432358788, + "loss": 0.1602, + "step": 5536 + }, + { + "epoch": 0.04806381889046102, + "grad_norm": 0.2060546875, + "learning_rate": 0.001989893595994651, + "loss": 0.165, + "step": 5537 + }, + { + "epoch": 0.048072499370665184, + "grad_norm": 0.69921875, + "learning_rate": 0.00198988914778025, + "loss": 0.1943, + "step": 5538 + }, + { + "epoch": 0.04808117985086935, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019898846985926797, + "loss": 0.127, + "step": 5539 + }, + { + "epoch": 0.048089860331073514, + "grad_norm": 0.166015625, + "learning_rate": 0.0019898802484319437, + "loss": 0.1426, + "step": 5540 + }, + { + "epoch": 0.04809854081127768, + "grad_norm": 0.1650390625, + "learning_rate": 0.001989875797298049, + "loss": 0.1211, + "step": 5541 + }, + { + "epoch": 0.048107221291481844, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019898713451909993, + "loss": 0.1562, + "step": 5542 + }, + { + "epoch": 0.04811590177168601, + "grad_norm": 0.6484375, + "learning_rate": 0.0019898668921107996, + "loss": 0.1719, + "step": 5543 + }, + { + "epoch": 0.048124582251890174, + "grad_norm": 0.33984375, + "learning_rate": 0.0019898624380574547, + "loss": 0.1396, + "step": 5544 + }, + { + "epoch": 0.04813326273209434, + "grad_norm": 0.11181640625, + "learning_rate": 0.00198985798303097, + "loss": 0.1904, + "step": 5545 + }, + { + "epoch": 0.048141943212298505, + "grad_norm": 0.6484375, + "learning_rate": 0.0019898535270313493, + "loss": 0.1484, + "step": 5546 + }, + { + "epoch": 0.04815062369250267, + "grad_norm": 0.1748046875, + "learning_rate": 0.001989849070058599, + "loss": 0.1787, + "step": 5547 + }, + { + "epoch": 0.048159304172706835, + "grad_norm": 0.349609375, + "learning_rate": 0.001989844612112722, + "loss": 0.1709, + "step": 5548 + }, + { + "epoch": 0.048167984652911, + "grad_norm": 0.08251953125, + "learning_rate": 0.001989840153193725, + "loss": 0.1602, + "step": 5549 + }, + { + "epoch": 0.048176665133115165, + "grad_norm": 0.455078125, + "learning_rate": 0.0019898356933016124, + "loss": 0.1533, + "step": 5550 + }, + { + "epoch": 0.04818534561331933, + "grad_norm": 0.5859375, + "learning_rate": 0.0019898312324363888, + "loss": 0.1582, + "step": 5551 + }, + { + "epoch": 0.048194026093523495, + "grad_norm": 0.1572265625, + "learning_rate": 0.001989826770598059, + "loss": 0.1631, + "step": 5552 + }, + { + "epoch": 0.04820270657372766, + "grad_norm": 0.197265625, + "learning_rate": 0.001989822307786628, + "loss": 0.127, + "step": 5553 + }, + { + "epoch": 0.048211387053931826, + "grad_norm": 0.365234375, + "learning_rate": 0.0019898178440021005, + "loss": 0.1514, + "step": 5554 + }, + { + "epoch": 0.04822006753413599, + "grad_norm": 0.45703125, + "learning_rate": 0.0019898133792444817, + "loss": 0.1582, + "step": 5555 + }, + { + "epoch": 0.048228748014340156, + "grad_norm": 0.328125, + "learning_rate": 0.0019898089135137764, + "loss": 0.1426, + "step": 5556 + }, + { + "epoch": 0.04823742849454432, + "grad_norm": 0.169921875, + "learning_rate": 0.0019898044468099896, + "loss": 0.1465, + "step": 5557 + }, + { + "epoch": 0.048246108974748486, + "grad_norm": 0.125, + "learning_rate": 0.0019897999791331257, + "loss": 0.2275, + "step": 5558 + }, + { + "epoch": 0.04825478945495265, + "grad_norm": 0.130859375, + "learning_rate": 0.00198979551048319, + "loss": 0.1187, + "step": 5559 + }, + { + "epoch": 0.04826346993515681, + "grad_norm": 0.15234375, + "learning_rate": 0.0019897910408601875, + "loss": 0.1602, + "step": 5560 + }, + { + "epoch": 0.048272150415360975, + "grad_norm": 0.396484375, + "learning_rate": 0.0019897865702641227, + "loss": 0.1108, + "step": 5561 + }, + { + "epoch": 0.04828083089556514, + "grad_norm": 0.6953125, + "learning_rate": 0.001989782098695001, + "loss": 0.1201, + "step": 5562 + }, + { + "epoch": 0.048289511375769305, + "grad_norm": 0.455078125, + "learning_rate": 0.0019897776261528265, + "loss": 0.1143, + "step": 5563 + }, + { + "epoch": 0.04829819185597347, + "grad_norm": 1.03125, + "learning_rate": 0.001989773152637605, + "loss": 0.1982, + "step": 5564 + }, + { + "epoch": 0.048306872336177635, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019897686781493403, + "loss": 0.1895, + "step": 5565 + }, + { + "epoch": 0.0483155528163818, + "grad_norm": 0.14453125, + "learning_rate": 0.0019897642026880384, + "loss": 0.1904, + "step": 5566 + }, + { + "epoch": 0.048324233296585965, + "grad_norm": 0.287109375, + "learning_rate": 0.0019897597262537037, + "loss": 0.1309, + "step": 5567 + }, + { + "epoch": 0.04833291377679013, + "grad_norm": 0.33984375, + "learning_rate": 0.0019897552488463413, + "loss": 0.1543, + "step": 5568 + }, + { + "epoch": 0.048341594256994296, + "grad_norm": 0.25390625, + "learning_rate": 0.0019897507704659556, + "loss": 0.1494, + "step": 5569 + }, + { + "epoch": 0.04835027473719846, + "grad_norm": 0.310546875, + "learning_rate": 0.0019897462911125517, + "loss": 0.1914, + "step": 5570 + }, + { + "epoch": 0.048358955217402626, + "grad_norm": 0.41796875, + "learning_rate": 0.001989741810786135, + "loss": 0.1328, + "step": 5571 + }, + { + "epoch": 0.04836763569760679, + "grad_norm": 0.259765625, + "learning_rate": 0.0019897373294867097, + "loss": 0.123, + "step": 5572 + }, + { + "epoch": 0.048376316177810956, + "grad_norm": 0.259765625, + "learning_rate": 0.0019897328472142815, + "loss": 0.1621, + "step": 5573 + }, + { + "epoch": 0.04838499665801512, + "grad_norm": 0.380859375, + "learning_rate": 0.0019897283639688547, + "loss": 0.1777, + "step": 5574 + }, + { + "epoch": 0.048393677138219286, + "grad_norm": 0.1328125, + "learning_rate": 0.001989723879750434, + "loss": 0.1426, + "step": 5575 + }, + { + "epoch": 0.04840235761842345, + "grad_norm": 0.56640625, + "learning_rate": 0.001989719394559025, + "loss": 0.2031, + "step": 5576 + }, + { + "epoch": 0.048411038098627616, + "grad_norm": 0.1435546875, + "learning_rate": 0.001989714908394632, + "loss": 0.1143, + "step": 5577 + }, + { + "epoch": 0.04841971857883178, + "grad_norm": 0.98828125, + "learning_rate": 0.00198971042125726, + "loss": 0.1328, + "step": 5578 + }, + { + "epoch": 0.04842839905903595, + "grad_norm": 0.62109375, + "learning_rate": 0.0019897059331469144, + "loss": 0.1396, + "step": 5579 + }, + { + "epoch": 0.04843707953924011, + "grad_norm": 0.515625, + "learning_rate": 0.0019897014440635997, + "loss": 0.1426, + "step": 5580 + }, + { + "epoch": 0.04844576001944428, + "grad_norm": 0.08349609375, + "learning_rate": 0.001989696954007321, + "loss": 0.126, + "step": 5581 + }, + { + "epoch": 0.04845444049964844, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019896924629780825, + "loss": 0.1211, + "step": 5582 + }, + { + "epoch": 0.04846312097985261, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019896879709758904, + "loss": 0.1602, + "step": 5583 + }, + { + "epoch": 0.04847180146005677, + "grad_norm": 0.46875, + "learning_rate": 0.0019896834780007483, + "loss": 0.1621, + "step": 5584 + }, + { + "epoch": 0.04848048194026094, + "grad_norm": 0.5390625, + "learning_rate": 0.0019896789840526618, + "loss": 0.1426, + "step": 5585 + }, + { + "epoch": 0.0484891624204651, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019896744891316357, + "loss": 0.1445, + "step": 5586 + }, + { + "epoch": 0.04849784290066927, + "grad_norm": 0.37890625, + "learning_rate": 0.0019896699932376756, + "loss": 0.1641, + "step": 5587 + }, + { + "epoch": 0.04850652338087343, + "grad_norm": 0.263671875, + "learning_rate": 0.001989665496370785, + "loss": 0.1602, + "step": 5588 + }, + { + "epoch": 0.0485152038610776, + "grad_norm": 0.244140625, + "learning_rate": 0.00198966099853097, + "loss": 0.1152, + "step": 5589 + }, + { + "epoch": 0.04852388434128176, + "grad_norm": 0.494140625, + "learning_rate": 0.001989656499718235, + "loss": 0.167, + "step": 5590 + }, + { + "epoch": 0.04853256482148592, + "grad_norm": 0.310546875, + "learning_rate": 0.001989651999932585, + "loss": 0.1611, + "step": 5591 + }, + { + "epoch": 0.048541245301690086, + "grad_norm": 0.4140625, + "learning_rate": 0.001989647499174025, + "loss": 0.1543, + "step": 5592 + }, + { + "epoch": 0.04854992578189425, + "grad_norm": 0.15234375, + "learning_rate": 0.0019896429974425598, + "loss": 0.1318, + "step": 5593 + }, + { + "epoch": 0.04855860626209842, + "grad_norm": 0.1953125, + "learning_rate": 0.0019896384947381946, + "loss": 0.1289, + "step": 5594 + }, + { + "epoch": 0.04856728674230258, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019896339910609336, + "loss": 0.1426, + "step": 5595 + }, + { + "epoch": 0.04857596722250675, + "grad_norm": 0.1591796875, + "learning_rate": 0.001989629486410783, + "loss": 0.1621, + "step": 5596 + }, + { + "epoch": 0.04858464770271091, + "grad_norm": 0.5390625, + "learning_rate": 0.0019896249807877463, + "loss": 0.1396, + "step": 5597 + }, + { + "epoch": 0.04859332818291508, + "grad_norm": 0.359375, + "learning_rate": 0.001989620474191829, + "loss": 0.1436, + "step": 5598 + }, + { + "epoch": 0.04860200866311924, + "grad_norm": 0.15234375, + "learning_rate": 0.0019896159666230365, + "loss": 0.1631, + "step": 5599 + }, + { + "epoch": 0.04861068914332341, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019896114580813733, + "loss": 0.1436, + "step": 5600 + }, + { + "epoch": 0.04861936962352757, + "grad_norm": 0.08984375, + "learning_rate": 0.0019896069485668445, + "loss": 0.1602, + "step": 5601 + }, + { + "epoch": 0.04862805010373174, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019896024380794547, + "loss": 0.1426, + "step": 5602 + }, + { + "epoch": 0.0486367305839359, + "grad_norm": 0.34765625, + "learning_rate": 0.001989597926619209, + "loss": 0.1289, + "step": 5603 + }, + { + "epoch": 0.04864541106414007, + "grad_norm": 0.271484375, + "learning_rate": 0.0019895934141861125, + "loss": 0.1152, + "step": 5604 + }, + { + "epoch": 0.04865409154434423, + "grad_norm": 0.1328125, + "learning_rate": 0.0019895889007801697, + "loss": 0.1143, + "step": 5605 + }, + { + "epoch": 0.0486627720245484, + "grad_norm": 0.2421875, + "learning_rate": 0.001989584386401386, + "loss": 0.1484, + "step": 5606 + }, + { + "epoch": 0.04867145250475256, + "grad_norm": 0.302734375, + "learning_rate": 0.0019895798710497666, + "loss": 0.1885, + "step": 5607 + }, + { + "epoch": 0.04868013298495673, + "grad_norm": 0.07421875, + "learning_rate": 0.001989575354725316, + "loss": 0.1299, + "step": 5608 + }, + { + "epoch": 0.04868881346516089, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019895708374280388, + "loss": 0.1543, + "step": 5609 + }, + { + "epoch": 0.04869749394536506, + "grad_norm": 0.373046875, + "learning_rate": 0.00198956631915794, + "loss": 0.2031, + "step": 5610 + }, + { + "epoch": 0.048706174425569224, + "grad_norm": 0.150390625, + "learning_rate": 0.0019895617999150256, + "loss": 0.1128, + "step": 5611 + }, + { + "epoch": 0.04871485490577339, + "grad_norm": 0.201171875, + "learning_rate": 0.001989557279699299, + "loss": 0.1328, + "step": 5612 + }, + { + "epoch": 0.048723535385977554, + "grad_norm": 0.80078125, + "learning_rate": 0.0019895527585107662, + "loss": 0.1406, + "step": 5613 + }, + { + "epoch": 0.04873221586618172, + "grad_norm": 0.283203125, + "learning_rate": 0.001989548236349432, + "loss": 0.1592, + "step": 5614 + }, + { + "epoch": 0.048740896346385884, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019895437132153014, + "loss": 0.1387, + "step": 5615 + }, + { + "epoch": 0.04874957682659005, + "grad_norm": 0.49609375, + "learning_rate": 0.001989539189108379, + "loss": 0.1455, + "step": 5616 + }, + { + "epoch": 0.048758257306794214, + "grad_norm": 0.67578125, + "learning_rate": 0.0019895346640286696, + "loss": 0.1641, + "step": 5617 + }, + { + "epoch": 0.04876693778699838, + "grad_norm": 0.3046875, + "learning_rate": 0.001989530137976179, + "loss": 0.1328, + "step": 5618 + }, + { + "epoch": 0.048775618267202545, + "grad_norm": 0.482421875, + "learning_rate": 0.001989525610950911, + "loss": 0.1318, + "step": 5619 + }, + { + "epoch": 0.04878429874740671, + "grad_norm": 0.193359375, + "learning_rate": 0.0019895210829528717, + "loss": 0.1689, + "step": 5620 + }, + { + "epoch": 0.048792979227610875, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019895165539820653, + "loss": 0.1455, + "step": 5621 + }, + { + "epoch": 0.04880165970781503, + "grad_norm": 0.0810546875, + "learning_rate": 0.001989512024038497, + "loss": 0.1445, + "step": 5622 + }, + { + "epoch": 0.0488103401880192, + "grad_norm": 0.146484375, + "learning_rate": 0.0019895074931221713, + "loss": 0.1484, + "step": 5623 + }, + { + "epoch": 0.04881902066822336, + "grad_norm": 0.287109375, + "learning_rate": 0.001989502961233094, + "loss": 0.1533, + "step": 5624 + }, + { + "epoch": 0.04882770114842753, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019894984283712696, + "loss": 0.1777, + "step": 5625 + }, + { + "epoch": 0.04883638162863169, + "grad_norm": 0.322265625, + "learning_rate": 0.001989493894536703, + "loss": 0.1445, + "step": 5626 + }, + { + "epoch": 0.04884506210883586, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019894893597293996, + "loss": 0.1318, + "step": 5627 + }, + { + "epoch": 0.048853742589040024, + "grad_norm": 0.294921875, + "learning_rate": 0.0019894848239493633, + "loss": 0.1855, + "step": 5628 + }, + { + "epoch": 0.04886242306924419, + "grad_norm": 0.33984375, + "learning_rate": 0.0019894802871966003, + "loss": 0.1611, + "step": 5629 + }, + { + "epoch": 0.048871103549448354, + "grad_norm": 0.90234375, + "learning_rate": 0.001989475749471115, + "loss": 0.1602, + "step": 5630 + }, + { + "epoch": 0.04887978402965252, + "grad_norm": 0.52734375, + "learning_rate": 0.001989471210772912, + "loss": 0.1777, + "step": 5631 + }, + { + "epoch": 0.048888464509856684, + "grad_norm": 0.30078125, + "learning_rate": 0.001989466671101997, + "loss": 0.166, + "step": 5632 + }, + { + "epoch": 0.04889714499006085, + "grad_norm": 0.66015625, + "learning_rate": 0.0019894621304583746, + "loss": 0.1738, + "step": 5633 + }, + { + "epoch": 0.048905825470265014, + "grad_norm": 0.42578125, + "learning_rate": 0.00198945758884205, + "loss": 0.1318, + "step": 5634 + }, + { + "epoch": 0.04891450595046918, + "grad_norm": 0.1435546875, + "learning_rate": 0.001989453046253027, + "loss": 0.1113, + "step": 5635 + }, + { + "epoch": 0.048923186430673345, + "grad_norm": 0.91796875, + "learning_rate": 0.001989448502691312, + "loss": 0.3262, + "step": 5636 + }, + { + "epoch": 0.04893186691087751, + "grad_norm": 0.201171875, + "learning_rate": 0.0019894439581569097, + "loss": 0.1855, + "step": 5637 + }, + { + "epoch": 0.048940547391081675, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019894394126498253, + "loss": 0.1011, + "step": 5638 + }, + { + "epoch": 0.04894922787128584, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019894348661700623, + "loss": 0.125, + "step": 5639 + }, + { + "epoch": 0.048957908351490005, + "grad_norm": 0.67578125, + "learning_rate": 0.0019894303187176273, + "loss": 0.2129, + "step": 5640 + }, + { + "epoch": 0.04896658883169417, + "grad_norm": 0.134765625, + "learning_rate": 0.0019894257702925246, + "loss": 0.1338, + "step": 5641 + }, + { + "epoch": 0.048975269311898335, + "grad_norm": 0.45703125, + "learning_rate": 0.001989421220894759, + "loss": 0.1426, + "step": 5642 + }, + { + "epoch": 0.0489839497921025, + "grad_norm": 0.412109375, + "learning_rate": 0.0019894166705243365, + "loss": 0.1621, + "step": 5643 + }, + { + "epoch": 0.048992630272306666, + "grad_norm": 0.1953125, + "learning_rate": 0.00198941211918126, + "loss": 0.1816, + "step": 5644 + }, + { + "epoch": 0.04900131075251083, + "grad_norm": 0.1337890625, + "learning_rate": 0.001989407566865537, + "loss": 0.1621, + "step": 5645 + }, + { + "epoch": 0.049009991232714996, + "grad_norm": 0.478515625, + "learning_rate": 0.0019894030135771705, + "loss": 0.2031, + "step": 5646 + }, + { + "epoch": 0.04901867171291916, + "grad_norm": 0.107421875, + "learning_rate": 0.0019893984593161666, + "loss": 0.1201, + "step": 5647 + }, + { + "epoch": 0.049027352193123326, + "grad_norm": 0.091796875, + "learning_rate": 0.0019893939040825297, + "loss": 0.1582, + "step": 5648 + }, + { + "epoch": 0.04903603267332749, + "grad_norm": 0.337890625, + "learning_rate": 0.001989389347876265, + "loss": 0.2324, + "step": 5649 + }, + { + "epoch": 0.049044713153531656, + "grad_norm": 0.298828125, + "learning_rate": 0.0019893847906973777, + "loss": 0.1504, + "step": 5650 + }, + { + "epoch": 0.04905339363373582, + "grad_norm": 0.193359375, + "learning_rate": 0.001989380232545872, + "loss": 0.1406, + "step": 5651 + }, + { + "epoch": 0.04906207411393999, + "grad_norm": 0.287109375, + "learning_rate": 0.001989375673421754, + "loss": 0.1475, + "step": 5652 + }, + { + "epoch": 0.049070754594144145, + "grad_norm": 0.083984375, + "learning_rate": 0.001989371113325028, + "loss": 0.1582, + "step": 5653 + }, + { + "epoch": 0.04907943507434831, + "grad_norm": 0.08203125, + "learning_rate": 0.0019893665522556993, + "loss": 0.1123, + "step": 5654 + }, + { + "epoch": 0.049088115554552475, + "grad_norm": 0.203125, + "learning_rate": 0.0019893619902137723, + "loss": 0.124, + "step": 5655 + }, + { + "epoch": 0.04909679603475664, + "grad_norm": 0.08203125, + "learning_rate": 0.0019893574271992527, + "loss": 0.1377, + "step": 5656 + }, + { + "epoch": 0.049105476514960805, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019893528632121448, + "loss": 0.1143, + "step": 5657 + }, + { + "epoch": 0.04911415699516497, + "grad_norm": 0.1513671875, + "learning_rate": 0.001989348298252454, + "loss": 0.1416, + "step": 5658 + }, + { + "epoch": 0.049122837475369135, + "grad_norm": 0.197265625, + "learning_rate": 0.0019893437323201856, + "loss": 0.1436, + "step": 5659 + }, + { + "epoch": 0.0491315179555733, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019893391654153443, + "loss": 0.1533, + "step": 5660 + }, + { + "epoch": 0.049140198435777466, + "grad_norm": 0.1015625, + "learning_rate": 0.001989334597537935, + "loss": 0.1465, + "step": 5661 + }, + { + "epoch": 0.04914887891598163, + "grad_norm": 0.1796875, + "learning_rate": 0.0019893300286879627, + "loss": 0.1973, + "step": 5662 + }, + { + "epoch": 0.049157559396185796, + "grad_norm": 0.65234375, + "learning_rate": 0.0019893254588654323, + "loss": 0.1758, + "step": 5663 + }, + { + "epoch": 0.04916623987638996, + "grad_norm": 0.39453125, + "learning_rate": 0.0019893208880703487, + "loss": 0.1973, + "step": 5664 + }, + { + "epoch": 0.049174920356594126, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019893163163027176, + "loss": 0.1553, + "step": 5665 + }, + { + "epoch": 0.04918360083679829, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019893117435625433, + "loss": 0.1396, + "step": 5666 + }, + { + "epoch": 0.049192281317002456, + "grad_norm": 0.2431640625, + "learning_rate": 0.001989307169849831, + "loss": 0.1494, + "step": 5667 + }, + { + "epoch": 0.04920096179720662, + "grad_norm": 0.68359375, + "learning_rate": 0.001989302595164586, + "loss": 0.21, + "step": 5668 + }, + { + "epoch": 0.04920964227741079, + "grad_norm": 0.130859375, + "learning_rate": 0.001989298019506813, + "loss": 0.1709, + "step": 5669 + }, + { + "epoch": 0.04921832275761495, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019892934428765168, + "loss": 0.1348, + "step": 5670 + }, + { + "epoch": 0.04922700323781912, + "grad_norm": 0.0849609375, + "learning_rate": 0.001989288865273703, + "loss": 0.1279, + "step": 5671 + }, + { + "epoch": 0.04923568371802328, + "grad_norm": 0.2431640625, + "learning_rate": 0.001989284286698376, + "loss": 0.1533, + "step": 5672 + }, + { + "epoch": 0.04924436419822745, + "grad_norm": 0.7265625, + "learning_rate": 0.001989279707150541, + "loss": 0.1475, + "step": 5673 + }, + { + "epoch": 0.04925304467843161, + "grad_norm": 0.1875, + "learning_rate": 0.0019892751266302033, + "loss": 0.1475, + "step": 5674 + }, + { + "epoch": 0.04926172515863578, + "grad_norm": 0.4921875, + "learning_rate": 0.0019892705451373676, + "loss": 0.1709, + "step": 5675 + }, + { + "epoch": 0.04927040563883994, + "grad_norm": 0.146484375, + "learning_rate": 0.001989265962672039, + "loss": 0.1338, + "step": 5676 + }, + { + "epoch": 0.04927908611904411, + "grad_norm": 0.171875, + "learning_rate": 0.0019892613792342226, + "loss": 0.1895, + "step": 5677 + }, + { + "epoch": 0.04928776659924827, + "grad_norm": 0.53125, + "learning_rate": 0.0019892567948239233, + "loss": 0.1455, + "step": 5678 + }, + { + "epoch": 0.04929644707945244, + "grad_norm": 0.1083984375, + "learning_rate": 0.001989252209441146, + "loss": 0.1758, + "step": 5679 + }, + { + "epoch": 0.0493051275596566, + "grad_norm": 0.08984375, + "learning_rate": 0.001989247623085896, + "loss": 0.1914, + "step": 5680 + }, + { + "epoch": 0.04931380803986077, + "grad_norm": 0.52734375, + "learning_rate": 0.0019892430357581776, + "loss": 0.2012, + "step": 5681 + }, + { + "epoch": 0.04932248852006493, + "grad_norm": 0.255859375, + "learning_rate": 0.001989238447457997, + "loss": 0.1953, + "step": 5682 + }, + { + "epoch": 0.0493311690002691, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019892338581853583, + "loss": 0.208, + "step": 5683 + }, + { + "epoch": 0.049339849480473257, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019892292679402667, + "loss": 0.2305, + "step": 5684 + }, + { + "epoch": 0.04934852996067742, + "grad_norm": 0.265625, + "learning_rate": 0.0019892246767227277, + "loss": 0.1699, + "step": 5685 + }, + { + "epoch": 0.04935721044088159, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019892200845327457, + "loss": 0.166, + "step": 5686 + }, + { + "epoch": 0.04936589092108575, + "grad_norm": 0.19921875, + "learning_rate": 0.0019892154913703263, + "loss": 0.1172, + "step": 5687 + }, + { + "epoch": 0.04937457140128992, + "grad_norm": 0.234375, + "learning_rate": 0.001989210897235474, + "loss": 0.1416, + "step": 5688 + }, + { + "epoch": 0.04938325188149408, + "grad_norm": 0.1328125, + "learning_rate": 0.0019892063021281934, + "loss": 0.1406, + "step": 5689 + }, + { + "epoch": 0.04939193236169825, + "grad_norm": 0.0986328125, + "learning_rate": 0.001989201706048491, + "loss": 0.1367, + "step": 5690 + }, + { + "epoch": 0.04940061284190241, + "grad_norm": 0.205078125, + "learning_rate": 0.001989197108996371, + "loss": 0.1475, + "step": 5691 + }, + { + "epoch": 0.04940929332210658, + "grad_norm": 0.142578125, + "learning_rate": 0.001989192510971838, + "loss": 0.1445, + "step": 5692 + }, + { + "epoch": 0.04941797380231074, + "grad_norm": 0.26171875, + "learning_rate": 0.0019891879119748974, + "loss": 0.123, + "step": 5693 + }, + { + "epoch": 0.04942665428251491, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019891833120055545, + "loss": 0.1406, + "step": 5694 + }, + { + "epoch": 0.04943533476271907, + "grad_norm": 0.3359375, + "learning_rate": 0.001989178711063814, + "loss": 0.1611, + "step": 5695 + }, + { + "epoch": 0.04944401524292324, + "grad_norm": 0.083984375, + "learning_rate": 0.0019891741091496813, + "loss": 0.1611, + "step": 5696 + }, + { + "epoch": 0.0494526957231274, + "grad_norm": 0.10546875, + "learning_rate": 0.0019891695062631606, + "loss": 0.127, + "step": 5697 + }, + { + "epoch": 0.04946137620333157, + "grad_norm": 0.09033203125, + "learning_rate": 0.001989164902404258, + "loss": 0.1875, + "step": 5698 + }, + { + "epoch": 0.04947005668353573, + "grad_norm": 0.291015625, + "learning_rate": 0.0019891602975729778, + "loss": 0.1406, + "step": 5699 + }, + { + "epoch": 0.0494787371637399, + "grad_norm": 0.234375, + "learning_rate": 0.0019891556917693252, + "loss": 0.1562, + "step": 5700 + }, + { + "epoch": 0.049487417643944064, + "grad_norm": 0.326171875, + "learning_rate": 0.001989151084993305, + "loss": 0.1758, + "step": 5701 + }, + { + "epoch": 0.04949609812414823, + "grad_norm": 0.11474609375, + "learning_rate": 0.001989146477244923, + "loss": 0.1123, + "step": 5702 + }, + { + "epoch": 0.049504778604352394, + "grad_norm": 0.16796875, + "learning_rate": 0.001989141868524184, + "loss": 0.1523, + "step": 5703 + }, + { + "epoch": 0.04951345908455656, + "grad_norm": 0.1357421875, + "learning_rate": 0.001989137258831092, + "loss": 0.1748, + "step": 5704 + }, + { + "epoch": 0.049522139564760724, + "grad_norm": 0.4765625, + "learning_rate": 0.0019891326481656533, + "loss": 0.1816, + "step": 5705 + }, + { + "epoch": 0.04953082004496489, + "grad_norm": 0.482421875, + "learning_rate": 0.001989128036527873, + "loss": 0.1504, + "step": 5706 + }, + { + "epoch": 0.049539500525169054, + "grad_norm": 0.212890625, + "learning_rate": 0.001989123423917755, + "loss": 0.1641, + "step": 5707 + }, + { + "epoch": 0.04954818100537322, + "grad_norm": 0.275390625, + "learning_rate": 0.001989118810335305, + "loss": 0.1445, + "step": 5708 + }, + { + "epoch": 0.049556861485577385, + "grad_norm": 0.169921875, + "learning_rate": 0.0019891141957805282, + "loss": 0.1426, + "step": 5709 + }, + { + "epoch": 0.04956554196578155, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019891095802534297, + "loss": 0.1289, + "step": 5710 + }, + { + "epoch": 0.049574222445985715, + "grad_norm": 0.09619140625, + "learning_rate": 0.001989104963754014, + "loss": 0.1367, + "step": 5711 + }, + { + "epoch": 0.04958290292618988, + "grad_norm": 0.212890625, + "learning_rate": 0.0019891003462822864, + "loss": 0.1348, + "step": 5712 + }, + { + "epoch": 0.049591583406394045, + "grad_norm": 0.50390625, + "learning_rate": 0.001989095727838252, + "loss": 0.1182, + "step": 5713 + }, + { + "epoch": 0.0496002638865982, + "grad_norm": 0.07421875, + "learning_rate": 0.0019890911084219157, + "loss": 0.1504, + "step": 5714 + }, + { + "epoch": 0.04960894436680237, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019890864880332836, + "loss": 0.1504, + "step": 5715 + }, + { + "epoch": 0.04961762484700653, + "grad_norm": 0.1376953125, + "learning_rate": 0.001989081866672359, + "loss": 0.1191, + "step": 5716 + }, + { + "epoch": 0.0496263053272107, + "grad_norm": 0.23828125, + "learning_rate": 0.001989077244339148, + "loss": 0.1357, + "step": 5717 + }, + { + "epoch": 0.049634985807414864, + "grad_norm": 0.20703125, + "learning_rate": 0.0019890726210336556, + "loss": 0.1523, + "step": 5718 + }, + { + "epoch": 0.04964366628761903, + "grad_norm": 0.24609375, + "learning_rate": 0.0019890679967558868, + "loss": 0.166, + "step": 5719 + }, + { + "epoch": 0.049652346767823194, + "grad_norm": 0.150390625, + "learning_rate": 0.001989063371505847, + "loss": 0.1406, + "step": 5720 + }, + { + "epoch": 0.04966102724802736, + "grad_norm": 0.1328125, + "learning_rate": 0.0019890587452835403, + "loss": 0.1123, + "step": 5721 + }, + { + "epoch": 0.049669707728231524, + "grad_norm": 0.1953125, + "learning_rate": 0.0019890541180889727, + "loss": 0.1445, + "step": 5722 + }, + { + "epoch": 0.04967838820843569, + "grad_norm": 0.76953125, + "learning_rate": 0.001989049489922148, + "loss": 0.1108, + "step": 5723 + }, + { + "epoch": 0.049687068688639854, + "grad_norm": 0.1953125, + "learning_rate": 0.001989044860783073, + "loss": 0.1523, + "step": 5724 + }, + { + "epoch": 0.04969574916884402, + "grad_norm": 0.224609375, + "learning_rate": 0.0019890402306717513, + "loss": 0.1621, + "step": 5725 + }, + { + "epoch": 0.049704429649048185, + "grad_norm": 0.59765625, + "learning_rate": 0.001989035599588189, + "loss": 0.1318, + "step": 5726 + }, + { + "epoch": 0.04971311012925235, + "grad_norm": 0.361328125, + "learning_rate": 0.0019890309675323907, + "loss": 0.1582, + "step": 5727 + }, + { + "epoch": 0.049721790609456515, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019890263345043614, + "loss": 0.1602, + "step": 5728 + }, + { + "epoch": 0.04973047108966068, + "grad_norm": 0.4765625, + "learning_rate": 0.0019890217005041066, + "loss": 0.1816, + "step": 5729 + }, + { + "epoch": 0.049739151569864845, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019890170655316305, + "loss": 0.1475, + "step": 5730 + }, + { + "epoch": 0.04974783205006901, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019890124295869386, + "loss": 0.1465, + "step": 5731 + }, + { + "epoch": 0.049756512530273175, + "grad_norm": 0.0947265625, + "learning_rate": 0.001989007792670037, + "loss": 0.168, + "step": 5732 + }, + { + "epoch": 0.04976519301047734, + "grad_norm": 0.1279296875, + "learning_rate": 0.001989003154780929, + "loss": 0.1494, + "step": 5733 + }, + { + "epoch": 0.049773873490681506, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019889985159196205, + "loss": 0.1445, + "step": 5734 + }, + { + "epoch": 0.04978255397088567, + "grad_norm": 0.48828125, + "learning_rate": 0.001988993876086117, + "loss": 0.1709, + "step": 5735 + }, + { + "epoch": 0.049791234451089836, + "grad_norm": 0.2265625, + "learning_rate": 0.001988989235280423, + "loss": 0.1934, + "step": 5736 + }, + { + "epoch": 0.049799914931294, + "grad_norm": 0.53125, + "learning_rate": 0.0019889845935025436, + "loss": 0.1836, + "step": 5737 + }, + { + "epoch": 0.049808595411498166, + "grad_norm": 0.1171875, + "learning_rate": 0.0019889799507524838, + "loss": 0.1494, + "step": 5738 + }, + { + "epoch": 0.04981727589170233, + "grad_norm": 0.384765625, + "learning_rate": 0.0019889753070302493, + "loss": 0.1797, + "step": 5739 + }, + { + "epoch": 0.049825956371906496, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019889706623358447, + "loss": 0.1504, + "step": 5740 + }, + { + "epoch": 0.04983463685211066, + "grad_norm": 0.314453125, + "learning_rate": 0.001988966016669275, + "loss": 0.1299, + "step": 5741 + }, + { + "epoch": 0.049843317332314827, + "grad_norm": 0.341796875, + "learning_rate": 0.0019889613700305457, + "loss": 0.1475, + "step": 5742 + }, + { + "epoch": 0.04985199781251899, + "grad_norm": 0.1611328125, + "learning_rate": 0.001988956722419661, + "loss": 0.1855, + "step": 5743 + }, + { + "epoch": 0.04986067829272316, + "grad_norm": 0.08203125, + "learning_rate": 0.0019889520738366273, + "loss": 0.1182, + "step": 5744 + }, + { + "epoch": 0.049869358772927315, + "grad_norm": 0.375, + "learning_rate": 0.001988947424281448, + "loss": 0.1729, + "step": 5745 + }, + { + "epoch": 0.04987803925313148, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019889427737541303, + "loss": 0.1494, + "step": 5746 + }, + { + "epoch": 0.049886719733335645, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019889381222546773, + "loss": 0.1553, + "step": 5747 + }, + { + "epoch": 0.04989540021353981, + "grad_norm": 0.310546875, + "learning_rate": 0.0019889334697830953, + "loss": 0.1494, + "step": 5748 + }, + { + "epoch": 0.049904080693743975, + "grad_norm": 0.19140625, + "learning_rate": 0.001988928816339389, + "loss": 0.1396, + "step": 5749 + }, + { + "epoch": 0.04991276117394814, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019889241619235635, + "loss": 0.1348, + "step": 5750 + }, + { + "epoch": 0.049921441654152306, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019889195065356242, + "loss": 0.1699, + "step": 5751 + }, + { + "epoch": 0.04993012213435647, + "grad_norm": 0.306640625, + "learning_rate": 0.0019889148501755754, + "loss": 0.1533, + "step": 5752 + }, + { + "epoch": 0.049938802614560636, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019889101928434226, + "loss": 0.1387, + "step": 5753 + }, + { + "epoch": 0.0499474830947648, + "grad_norm": 0.6328125, + "learning_rate": 0.001988905534539171, + "loss": 0.1357, + "step": 5754 + }, + { + "epoch": 0.049956163574968966, + "grad_norm": 0.578125, + "learning_rate": 0.0019889008752628262, + "loss": 0.1484, + "step": 5755 + }, + { + "epoch": 0.04996484405517313, + "grad_norm": 0.1865234375, + "learning_rate": 0.001988896215014392, + "loss": 0.1357, + "step": 5756 + }, + { + "epoch": 0.049973524535377296, + "grad_norm": 0.455078125, + "learning_rate": 0.0019888915537938754, + "loss": 0.1631, + "step": 5757 + }, + { + "epoch": 0.04998220501558146, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019888868916012794, + "loss": 0.1309, + "step": 5758 + }, + { + "epoch": 0.04999088549578563, + "grad_norm": 0.1484375, + "learning_rate": 0.0019888822284366103, + "loss": 0.1816, + "step": 5759 + }, + { + "epoch": 0.04999956597598979, + "grad_norm": 0.224609375, + "learning_rate": 0.0019888775642998725, + "loss": 0.1475, + "step": 5760 + }, + { + "epoch": 0.05000824645619396, + "grad_norm": 0.490234375, + "learning_rate": 0.0019888728991910723, + "loss": 0.1387, + "step": 5761 + }, + { + "epoch": 0.05001692693639812, + "grad_norm": 0.8125, + "learning_rate": 0.001988868233110214, + "loss": 0.1494, + "step": 5762 + }, + { + "epoch": 0.05002560741660229, + "grad_norm": 0.486328125, + "learning_rate": 0.001988863566057302, + "loss": 0.1631, + "step": 5763 + }, + { + "epoch": 0.05003428789680645, + "grad_norm": 0.6796875, + "learning_rate": 0.001988858898032343, + "loss": 0.1738, + "step": 5764 + }, + { + "epoch": 0.05004296837701062, + "grad_norm": 0.326171875, + "learning_rate": 0.0019888542290353406, + "loss": 0.1738, + "step": 5765 + }, + { + "epoch": 0.05005164885721478, + "grad_norm": 0.2197265625, + "learning_rate": 0.001988849559066301, + "loss": 0.1143, + "step": 5766 + }, + { + "epoch": 0.05006032933741895, + "grad_norm": 0.52734375, + "learning_rate": 0.0019888448881252288, + "loss": 0.1406, + "step": 5767 + }, + { + "epoch": 0.05006900981762311, + "grad_norm": 0.224609375, + "learning_rate": 0.001988840216212129, + "loss": 0.1582, + "step": 5768 + }, + { + "epoch": 0.05007769029782728, + "grad_norm": 0.283203125, + "learning_rate": 0.001988835543327007, + "loss": 0.1138, + "step": 5769 + }, + { + "epoch": 0.05008637077803144, + "grad_norm": 0.123046875, + "learning_rate": 0.0019888308694698677, + "loss": 0.1553, + "step": 5770 + }, + { + "epoch": 0.05009505125823561, + "grad_norm": 0.73046875, + "learning_rate": 0.0019888261946407167, + "loss": 0.1416, + "step": 5771 + }, + { + "epoch": 0.05010373173843977, + "grad_norm": 0.177734375, + "learning_rate": 0.0019888215188395585, + "loss": 0.1641, + "step": 5772 + }, + { + "epoch": 0.05011241221864394, + "grad_norm": 0.796875, + "learning_rate": 0.0019888168420663984, + "loss": 0.1875, + "step": 5773 + }, + { + "epoch": 0.0501210926988481, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019888121643212415, + "loss": 0.166, + "step": 5774 + }, + { + "epoch": 0.05012977317905227, + "grad_norm": 0.1650390625, + "learning_rate": 0.001988807485604093, + "loss": 0.1875, + "step": 5775 + }, + { + "epoch": 0.05013845365925643, + "grad_norm": 0.37109375, + "learning_rate": 0.001988802805914958, + "loss": 0.1221, + "step": 5776 + }, + { + "epoch": 0.05014713413946059, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019887981252538415, + "loss": 0.1816, + "step": 5777 + }, + { + "epoch": 0.05015581461966476, + "grad_norm": 0.205078125, + "learning_rate": 0.001988793443620749, + "loss": 0.1328, + "step": 5778 + }, + { + "epoch": 0.05016449509986892, + "grad_norm": 0.1171875, + "learning_rate": 0.0019887887610156854, + "loss": 0.1631, + "step": 5779 + }, + { + "epoch": 0.05017317558007309, + "grad_norm": 0.412109375, + "learning_rate": 0.0019887840774386553, + "loss": 0.1367, + "step": 5780 + }, + { + "epoch": 0.05018185606027725, + "grad_norm": 0.3125, + "learning_rate": 0.001988779392889665, + "loss": 0.167, + "step": 5781 + }, + { + "epoch": 0.05019053654048142, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019887747073687183, + "loss": 0.1074, + "step": 5782 + }, + { + "epoch": 0.05019921702068558, + "grad_norm": 0.21875, + "learning_rate": 0.001988770020875821, + "loss": 0.1875, + "step": 5783 + }, + { + "epoch": 0.05020789750088975, + "grad_norm": 0.310546875, + "learning_rate": 0.001988765333410978, + "loss": 0.1895, + "step": 5784 + }, + { + "epoch": 0.05021657798109391, + "grad_norm": 0.2197265625, + "learning_rate": 0.001988760644974195, + "loss": 0.1641, + "step": 5785 + }, + { + "epoch": 0.05022525846129808, + "grad_norm": 0.7265625, + "learning_rate": 0.0019887559555654766, + "loss": 0.1689, + "step": 5786 + }, + { + "epoch": 0.05023393894150224, + "grad_norm": 0.091796875, + "learning_rate": 0.001988751265184828, + "loss": 0.1465, + "step": 5787 + }, + { + "epoch": 0.05024261942170641, + "grad_norm": 0.212890625, + "learning_rate": 0.0019887465738322543, + "loss": 0.1348, + "step": 5788 + }, + { + "epoch": 0.05025129990191057, + "grad_norm": 0.203125, + "learning_rate": 0.0019887418815077605, + "loss": 0.1123, + "step": 5789 + }, + { + "epoch": 0.05025998038211474, + "grad_norm": 0.23828125, + "learning_rate": 0.0019887371882113527, + "loss": 0.1689, + "step": 5790 + }, + { + "epoch": 0.050268660862318904, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019887324939430345, + "loss": 0.1562, + "step": 5791 + }, + { + "epoch": 0.05027734134252307, + "grad_norm": 0.283203125, + "learning_rate": 0.0019887277987028123, + "loss": 0.1572, + "step": 5792 + }, + { + "epoch": 0.050286021822727234, + "grad_norm": 0.33203125, + "learning_rate": 0.0019887231024906904, + "loss": 0.1338, + "step": 5793 + }, + { + "epoch": 0.0502947023029314, + "grad_norm": 0.7578125, + "learning_rate": 0.001988718405306674, + "loss": 0.1689, + "step": 5794 + }, + { + "epoch": 0.050303382783135564, + "grad_norm": 0.48828125, + "learning_rate": 0.0019887137071507694, + "loss": 0.1387, + "step": 5795 + }, + { + "epoch": 0.05031206326333973, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019887090080229802, + "loss": 0.1533, + "step": 5796 + }, + { + "epoch": 0.050320743743543894, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019887043079233123, + "loss": 0.1699, + "step": 5797 + }, + { + "epoch": 0.05032942422374806, + "grad_norm": 0.408203125, + "learning_rate": 0.0019886996068517707, + "loss": 0.1221, + "step": 5798 + }, + { + "epoch": 0.050338104703952224, + "grad_norm": 0.087890625, + "learning_rate": 0.0019886949048083607, + "loss": 0.1621, + "step": 5799 + }, + { + "epoch": 0.05034678518415639, + "grad_norm": 0.1650390625, + "learning_rate": 0.001988690201793087, + "loss": 0.1338, + "step": 5800 + }, + { + "epoch": 0.050355465664360555, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019886854978059553, + "loss": 0.1875, + "step": 5801 + }, + { + "epoch": 0.05036414614456472, + "grad_norm": 0.40234375, + "learning_rate": 0.0019886807928469704, + "loss": 0.1631, + "step": 5802 + }, + { + "epoch": 0.050372826624768885, + "grad_norm": 0.42578125, + "learning_rate": 0.0019886760869161375, + "loss": 0.1445, + "step": 5803 + }, + { + "epoch": 0.05038150710497305, + "grad_norm": 0.2734375, + "learning_rate": 0.0019886713800134617, + "loss": 0.1191, + "step": 5804 + }, + { + "epoch": 0.050390187585177215, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019886666721389483, + "loss": 0.1318, + "step": 5805 + }, + { + "epoch": 0.05039886806538138, + "grad_norm": 0.296875, + "learning_rate": 0.0019886619632926024, + "loss": 0.1562, + "step": 5806 + }, + { + "epoch": 0.05040754854558554, + "grad_norm": 0.365234375, + "learning_rate": 0.0019886572534744294, + "loss": 0.1611, + "step": 5807 + }, + { + "epoch": 0.050416229025789704, + "grad_norm": 0.193359375, + "learning_rate": 0.001988652542684434, + "loss": 0.1543, + "step": 5808 + }, + { + "epoch": 0.05042490950599387, + "grad_norm": 0.671875, + "learning_rate": 0.0019886478309226216, + "loss": 0.1562, + "step": 5809 + }, + { + "epoch": 0.050433589986198034, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019886431181889972, + "loss": 0.1289, + "step": 5810 + }, + { + "epoch": 0.0504422704664022, + "grad_norm": 0.478515625, + "learning_rate": 0.0019886384044835656, + "loss": 0.1738, + "step": 5811 + }, + { + "epoch": 0.050450950946606364, + "grad_norm": 0.22265625, + "learning_rate": 0.001988633689806333, + "loss": 0.1289, + "step": 5812 + }, + { + "epoch": 0.05045963142681053, + "grad_norm": 0.220703125, + "learning_rate": 0.001988628974157304, + "loss": 0.1494, + "step": 5813 + }, + { + "epoch": 0.050468311907014694, + "grad_norm": 0.1171875, + "learning_rate": 0.001988624257536483, + "loss": 0.1748, + "step": 5814 + }, + { + "epoch": 0.05047699238721886, + "grad_norm": 0.6796875, + "learning_rate": 0.0019886195399438766, + "loss": 0.1123, + "step": 5815 + }, + { + "epoch": 0.050485672867423025, + "grad_norm": 1.7265625, + "learning_rate": 0.001988614821379489, + "loss": 0.3867, + "step": 5816 + }, + { + "epoch": 0.05049435334762719, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019886101018433255, + "loss": 0.0957, + "step": 5817 + }, + { + "epoch": 0.050503033827831355, + "grad_norm": 0.1796875, + "learning_rate": 0.0019886053813353913, + "loss": 0.1953, + "step": 5818 + }, + { + "epoch": 0.05051171430803552, + "grad_norm": 0.48828125, + "learning_rate": 0.001988600659855692, + "loss": 0.1426, + "step": 5819 + }, + { + "epoch": 0.050520394788239685, + "grad_norm": 0.181640625, + "learning_rate": 0.0019885959374042316, + "loss": 0.2285, + "step": 5820 + }, + { + "epoch": 0.05052907526844385, + "grad_norm": 0.36328125, + "learning_rate": 0.0019885912139810166, + "loss": 0.1562, + "step": 5821 + }, + { + "epoch": 0.050537755748648015, + "grad_norm": 0.15234375, + "learning_rate": 0.0019885864895860515, + "loss": 0.1523, + "step": 5822 + }, + { + "epoch": 0.05054643622885218, + "grad_norm": 0.48828125, + "learning_rate": 0.0019885817642193416, + "loss": 0.1387, + "step": 5823 + }, + { + "epoch": 0.050555116709056346, + "grad_norm": 0.39453125, + "learning_rate": 0.001988577037880892, + "loss": 0.124, + "step": 5824 + }, + { + "epoch": 0.05056379718926051, + "grad_norm": 0.15234375, + "learning_rate": 0.0019885723105707077, + "loss": 0.1855, + "step": 5825 + }, + { + "epoch": 0.050572477669464676, + "grad_norm": 0.52734375, + "learning_rate": 0.001988567582288794, + "loss": 0.1123, + "step": 5826 + }, + { + "epoch": 0.05058115814966884, + "grad_norm": 0.400390625, + "learning_rate": 0.0019885628530351565, + "loss": 0.1523, + "step": 5827 + }, + { + "epoch": 0.050589838629873006, + "grad_norm": 0.6875, + "learning_rate": 0.0019885581228098, + "loss": 0.1641, + "step": 5828 + }, + { + "epoch": 0.05059851911007717, + "grad_norm": 0.173828125, + "learning_rate": 0.0019885533916127298, + "loss": 0.165, + "step": 5829 + }, + { + "epoch": 0.050607199590281336, + "grad_norm": 0.30859375, + "learning_rate": 0.001988548659443951, + "loss": 0.1396, + "step": 5830 + }, + { + "epoch": 0.0506158800704855, + "grad_norm": 0.44921875, + "learning_rate": 0.001988543926303468, + "loss": 0.1289, + "step": 5831 + }, + { + "epoch": 0.050624560550689666, + "grad_norm": 0.13671875, + "learning_rate": 0.0019885391921912873, + "loss": 0.1328, + "step": 5832 + }, + { + "epoch": 0.05063324103089383, + "grad_norm": 0.74609375, + "learning_rate": 0.0019885344571074137, + "loss": 0.1641, + "step": 5833 + }, + { + "epoch": 0.050641921511098, + "grad_norm": 0.9453125, + "learning_rate": 0.0019885297210518516, + "loss": 0.1602, + "step": 5834 + }, + { + "epoch": 0.05065060199130216, + "grad_norm": 0.369140625, + "learning_rate": 0.001988524984024607, + "loss": 0.1709, + "step": 5835 + }, + { + "epoch": 0.05065928247150633, + "grad_norm": 0.12060546875, + "learning_rate": 0.001988520246025685, + "loss": 0.1758, + "step": 5836 + }, + { + "epoch": 0.05066796295171049, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019885155070550904, + "loss": 0.1621, + "step": 5837 + }, + { + "epoch": 0.05067664343191465, + "grad_norm": 0.3515625, + "learning_rate": 0.0019885107671128287, + "loss": 0.1465, + "step": 5838 + }, + { + "epoch": 0.050685323912118815, + "grad_norm": 0.419921875, + "learning_rate": 0.001988506026198905, + "loss": 0.127, + "step": 5839 + }, + { + "epoch": 0.05069400439232298, + "grad_norm": 0.427734375, + "learning_rate": 0.0019885012843133243, + "loss": 0.1299, + "step": 5840 + }, + { + "epoch": 0.050702684872527146, + "grad_norm": 0.466796875, + "learning_rate": 0.001988496541456092, + "loss": 0.1279, + "step": 5841 + }, + { + "epoch": 0.05071136535273131, + "grad_norm": 0.09912109375, + "learning_rate": 0.001988491797627213, + "loss": 0.127, + "step": 5842 + }, + { + "epoch": 0.050720045832935476, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019884870528266933, + "loss": 0.1328, + "step": 5843 + }, + { + "epoch": 0.05072872631313964, + "grad_norm": 1.453125, + "learning_rate": 0.001988482307054537, + "loss": 0.3008, + "step": 5844 + }, + { + "epoch": 0.050737406793343806, + "grad_norm": 0.671875, + "learning_rate": 0.0019884775603107503, + "loss": 0.1836, + "step": 5845 + }, + { + "epoch": 0.05074608727354797, + "grad_norm": 0.2109375, + "learning_rate": 0.001988472812595337, + "loss": 0.1416, + "step": 5846 + }, + { + "epoch": 0.050754767753752136, + "grad_norm": 0.3046875, + "learning_rate": 0.001988468063908304, + "loss": 0.1445, + "step": 5847 + }, + { + "epoch": 0.0507634482339563, + "grad_norm": 1.1875, + "learning_rate": 0.0019884633142496556, + "loss": 0.1406, + "step": 5848 + }, + { + "epoch": 0.05077212871416047, + "grad_norm": 0.36328125, + "learning_rate": 0.0019884585636193966, + "loss": 0.1484, + "step": 5849 + }, + { + "epoch": 0.05078080919436463, + "grad_norm": 0.30078125, + "learning_rate": 0.001988453812017533, + "loss": 0.1357, + "step": 5850 + }, + { + "epoch": 0.0507894896745688, + "grad_norm": 0.2109375, + "learning_rate": 0.0019884490594440697, + "loss": 0.1406, + "step": 5851 + }, + { + "epoch": 0.05079817015477296, + "grad_norm": 0.73828125, + "learning_rate": 0.0019884443058990117, + "loss": 0.1465, + "step": 5852 + }, + { + "epoch": 0.05080685063497713, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019884395513823645, + "loss": 0.1631, + "step": 5853 + }, + { + "epoch": 0.05081553111518129, + "grad_norm": 0.2578125, + "learning_rate": 0.001988434795894133, + "loss": 0.168, + "step": 5854 + }, + { + "epoch": 0.05082421159538546, + "grad_norm": 1.8828125, + "learning_rate": 0.001988430039434323, + "loss": 0.1758, + "step": 5855 + }, + { + "epoch": 0.05083289207558962, + "grad_norm": 0.84375, + "learning_rate": 0.0019884252820029386, + "loss": 0.1836, + "step": 5856 + }, + { + "epoch": 0.05084157255579379, + "grad_norm": 0.41796875, + "learning_rate": 0.001988420523599986, + "loss": 0.2012, + "step": 5857 + }, + { + "epoch": 0.05085025303599795, + "grad_norm": 0.75390625, + "learning_rate": 0.00198841576422547, + "loss": 0.1953, + "step": 5858 + }, + { + "epoch": 0.05085893351620212, + "grad_norm": 0.1787109375, + "learning_rate": 0.001988411003879396, + "loss": 0.1201, + "step": 5859 + }, + { + "epoch": 0.05086761399640628, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019884062425617687, + "loss": 0.165, + "step": 5860 + }, + { + "epoch": 0.05087629447661045, + "grad_norm": 0.33984375, + "learning_rate": 0.001988401480272594, + "loss": 0.1113, + "step": 5861 + }, + { + "epoch": 0.05088497495681461, + "grad_norm": 0.44140625, + "learning_rate": 0.001988396717011877, + "loss": 0.1348, + "step": 5862 + }, + { + "epoch": 0.05089365543701878, + "grad_norm": 0.07421875, + "learning_rate": 0.0019883919527796225, + "loss": 0.1475, + "step": 5863 + }, + { + "epoch": 0.05090233591722294, + "grad_norm": 0.2265625, + "learning_rate": 0.0019883871875758354, + "loss": 0.2188, + "step": 5864 + }, + { + "epoch": 0.05091101639742711, + "grad_norm": 0.10595703125, + "learning_rate": 0.001988382421400522, + "loss": 0.1309, + "step": 5865 + }, + { + "epoch": 0.050919696877631274, + "grad_norm": 0.625, + "learning_rate": 0.001988377654253687, + "loss": 0.1855, + "step": 5866 + }, + { + "epoch": 0.05092837735783544, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019883728861353352, + "loss": 0.1211, + "step": 5867 + }, + { + "epoch": 0.0509370578380396, + "grad_norm": 0.10986328125, + "learning_rate": 0.001988368117045472, + "loss": 0.1855, + "step": 5868 + }, + { + "epoch": 0.05094573831824376, + "grad_norm": 1.03125, + "learning_rate": 0.0019883633469841037, + "loss": 0.1235, + "step": 5869 + }, + { + "epoch": 0.05095441879844793, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019883585759512337, + "loss": 0.1357, + "step": 5870 + }, + { + "epoch": 0.05096309927865209, + "grad_norm": 0.2734375, + "learning_rate": 0.0019883538039468683, + "loss": 0.1572, + "step": 5871 + }, + { + "epoch": 0.05097177975885626, + "grad_norm": 0.1875, + "learning_rate": 0.0019883490309710126, + "loss": 0.209, + "step": 5872 + }, + { + "epoch": 0.05098046023906042, + "grad_norm": 0.2001953125, + "learning_rate": 0.001988344257023672, + "loss": 0.1494, + "step": 5873 + }, + { + "epoch": 0.05098914071926459, + "grad_norm": 0.60546875, + "learning_rate": 0.0019883394821048513, + "loss": 0.1846, + "step": 5874 + }, + { + "epoch": 0.05099782119946875, + "grad_norm": 0.3046875, + "learning_rate": 0.0019883347062145556, + "loss": 0.1025, + "step": 5875 + }, + { + "epoch": 0.05100650167967292, + "grad_norm": 0.259765625, + "learning_rate": 0.0019883299293527905, + "loss": 0.1602, + "step": 5876 + }, + { + "epoch": 0.05101518215987708, + "grad_norm": 0.275390625, + "learning_rate": 0.001988325151519561, + "loss": 0.1592, + "step": 5877 + }, + { + "epoch": 0.05102386264008125, + "grad_norm": 0.20703125, + "learning_rate": 0.0019883203727148733, + "loss": 0.1387, + "step": 5878 + }, + { + "epoch": 0.05103254312028541, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019883155929387307, + "loss": 0.1465, + "step": 5879 + }, + { + "epoch": 0.05104122360048958, + "grad_norm": 0.478515625, + "learning_rate": 0.0019883108121911404, + "loss": 0.1436, + "step": 5880 + }, + { + "epoch": 0.05104990408069374, + "grad_norm": 0.52734375, + "learning_rate": 0.001988306030472106, + "loss": 0.1445, + "step": 5881 + }, + { + "epoch": 0.05105858456089791, + "grad_norm": 0.466796875, + "learning_rate": 0.0019883012477816342, + "loss": 0.1621, + "step": 5882 + }, + { + "epoch": 0.051067265041102074, + "grad_norm": 0.298828125, + "learning_rate": 0.001988296464119729, + "loss": 0.1406, + "step": 5883 + }, + { + "epoch": 0.05107594552130624, + "grad_norm": 0.62109375, + "learning_rate": 0.001988291679486396, + "loss": 0.1338, + "step": 5884 + }, + { + "epoch": 0.051084626001510404, + "grad_norm": 0.09326171875, + "learning_rate": 0.001988286893881641, + "loss": 0.1738, + "step": 5885 + }, + { + "epoch": 0.05109330648171457, + "grad_norm": 0.07373046875, + "learning_rate": 0.001988282107305469, + "loss": 0.1514, + "step": 5886 + }, + { + "epoch": 0.051101986961918734, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019882773197578843, + "loss": 0.1436, + "step": 5887 + }, + { + "epoch": 0.0511106674421229, + "grad_norm": 0.6484375, + "learning_rate": 0.0019882725312388933, + "loss": 0.1455, + "step": 5888 + }, + { + "epoch": 0.051119347922327064, + "grad_norm": 0.443359375, + "learning_rate": 0.0019882677417485005, + "loss": 0.1348, + "step": 5889 + }, + { + "epoch": 0.05112802840253123, + "grad_norm": 0.3359375, + "learning_rate": 0.001988262951286712, + "loss": 0.1758, + "step": 5890 + }, + { + "epoch": 0.051136708882735395, + "grad_norm": 0.216796875, + "learning_rate": 0.0019882581598535324, + "loss": 0.1553, + "step": 5891 + }, + { + "epoch": 0.05114538936293956, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019882533674489664, + "loss": 0.2832, + "step": 5892 + }, + { + "epoch": 0.051154069843143725, + "grad_norm": 0.92578125, + "learning_rate": 0.0019882485740730206, + "loss": 0.1436, + "step": 5893 + }, + { + "epoch": 0.05116275032334789, + "grad_norm": 0.2109375, + "learning_rate": 0.0019882437797256987, + "loss": 0.1338, + "step": 5894 + }, + { + "epoch": 0.051171430803552055, + "grad_norm": 0.357421875, + "learning_rate": 0.0019882389844070075, + "loss": 0.1621, + "step": 5895 + }, + { + "epoch": 0.05118011128375622, + "grad_norm": 0.0849609375, + "learning_rate": 0.001988234188116951, + "loss": 0.1289, + "step": 5896 + }, + { + "epoch": 0.051188791763960385, + "grad_norm": 0.130859375, + "learning_rate": 0.001988229390855535, + "loss": 0.1758, + "step": 5897 + }, + { + "epoch": 0.05119747224416455, + "grad_norm": 0.11767578125, + "learning_rate": 0.001988224592622765, + "loss": 0.1338, + "step": 5898 + }, + { + "epoch": 0.05120615272436871, + "grad_norm": 0.12109375, + "learning_rate": 0.0019882197934186457, + "loss": 0.125, + "step": 5899 + }, + { + "epoch": 0.051214833204572874, + "grad_norm": 0.6171875, + "learning_rate": 0.001988214993243183, + "loss": 0.167, + "step": 5900 + }, + { + "epoch": 0.05122351368477704, + "grad_norm": 0.1337890625, + "learning_rate": 0.001988210192096381, + "loss": 0.1465, + "step": 5901 + }, + { + "epoch": 0.051232194164981204, + "grad_norm": 0.365234375, + "learning_rate": 0.001988205389978246, + "loss": 0.1289, + "step": 5902 + }, + { + "epoch": 0.05124087464518537, + "grad_norm": 0.1513671875, + "learning_rate": 0.001988200586888783, + "loss": 0.1445, + "step": 5903 + }, + { + "epoch": 0.051249555125389534, + "grad_norm": 0.1474609375, + "learning_rate": 0.001988195782827997, + "loss": 0.1289, + "step": 5904 + }, + { + "epoch": 0.0512582356055937, + "grad_norm": 0.21875, + "learning_rate": 0.0019881909777958937, + "loss": 0.1064, + "step": 5905 + }, + { + "epoch": 0.051266916085797865, + "grad_norm": 0.8125, + "learning_rate": 0.001988186171792478, + "loss": 0.1709, + "step": 5906 + }, + { + "epoch": 0.05127559656600203, + "grad_norm": 0.51171875, + "learning_rate": 0.0019881813648177553, + "loss": 0.1309, + "step": 5907 + }, + { + "epoch": 0.051284277046206195, + "grad_norm": 0.71484375, + "learning_rate": 0.001988176556871731, + "loss": 0.1367, + "step": 5908 + }, + { + "epoch": 0.05129295752641036, + "grad_norm": 0.3984375, + "learning_rate": 0.0019881717479544097, + "loss": 0.1348, + "step": 5909 + }, + { + "epoch": 0.051301638006614525, + "grad_norm": 0.7265625, + "learning_rate": 0.0019881669380657973, + "loss": 0.1455, + "step": 5910 + }, + { + "epoch": 0.05131031848681869, + "grad_norm": 0.490234375, + "learning_rate": 0.001988162127205899, + "loss": 0.1426, + "step": 5911 + }, + { + "epoch": 0.051318998967022855, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019881573153747198, + "loss": 0.125, + "step": 5912 + }, + { + "epoch": 0.05132767944722702, + "grad_norm": 0.3046875, + "learning_rate": 0.001988152502572265, + "loss": 0.1436, + "step": 5913 + }, + { + "epoch": 0.051336359927431185, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019881476887985403, + "loss": 0.1465, + "step": 5914 + }, + { + "epoch": 0.05134504040763535, + "grad_norm": 0.5078125, + "learning_rate": 0.0019881428740535505, + "loss": 0.1216, + "step": 5915 + }, + { + "epoch": 0.051353720887839516, + "grad_norm": 0.2421875, + "learning_rate": 0.0019881380583373007, + "loss": 0.1011, + "step": 5916 + }, + { + "epoch": 0.05136240136804368, + "grad_norm": 0.1298828125, + "learning_rate": 0.001988133241649797, + "loss": 0.1445, + "step": 5917 + }, + { + "epoch": 0.051371081848247846, + "grad_norm": 0.5078125, + "learning_rate": 0.0019881284239910437, + "loss": 0.2754, + "step": 5918 + }, + { + "epoch": 0.05137976232845201, + "grad_norm": 0.07470703125, + "learning_rate": 0.001988123605361047, + "loss": 0.1436, + "step": 5919 + }, + { + "epoch": 0.051388442808656176, + "grad_norm": 0.3671875, + "learning_rate": 0.001988118785759811, + "loss": 0.2734, + "step": 5920 + }, + { + "epoch": 0.05139712328886034, + "grad_norm": 0.482421875, + "learning_rate": 0.0019881139651873416, + "loss": 0.1367, + "step": 5921 + }, + { + "epoch": 0.051405803769064506, + "grad_norm": 0.390625, + "learning_rate": 0.0019881091436436448, + "loss": 0.1221, + "step": 5922 + }, + { + "epoch": 0.05141448424926867, + "grad_norm": 0.384765625, + "learning_rate": 0.001988104321128725, + "loss": 0.1758, + "step": 5923 + }, + { + "epoch": 0.05142316472947284, + "grad_norm": 0.32421875, + "learning_rate": 0.0019880994976425874, + "loss": 0.2158, + "step": 5924 + }, + { + "epoch": 0.051431845209677, + "grad_norm": 2.3125, + "learning_rate": 0.0019880946731852377, + "loss": 0.2773, + "step": 5925 + }, + { + "epoch": 0.05144052568988117, + "grad_norm": 0.37109375, + "learning_rate": 0.001988089847756681, + "loss": 0.2314, + "step": 5926 + }, + { + "epoch": 0.05144920617008533, + "grad_norm": 0.1220703125, + "learning_rate": 0.001988085021356922, + "loss": 0.1777, + "step": 5927 + }, + { + "epoch": 0.0514578866502895, + "grad_norm": 0.09521484375, + "learning_rate": 0.001988080193985967, + "loss": 0.1455, + "step": 5928 + }, + { + "epoch": 0.05146656713049366, + "grad_norm": 0.26953125, + "learning_rate": 0.0019880753656438214, + "loss": 0.1523, + "step": 5929 + }, + { + "epoch": 0.05147524761069782, + "grad_norm": 0.3046875, + "learning_rate": 0.0019880705363304893, + "loss": 0.1484, + "step": 5930 + }, + { + "epoch": 0.051483928090901986, + "grad_norm": 0.20703125, + "learning_rate": 0.0019880657060459767, + "loss": 0.1475, + "step": 5931 + }, + { + "epoch": 0.05149260857110615, + "grad_norm": 0.26171875, + "learning_rate": 0.0019880608747902886, + "loss": 0.1689, + "step": 5932 + }, + { + "epoch": 0.051501289051310316, + "grad_norm": 0.1328125, + "learning_rate": 0.0019880560425634313, + "loss": 0.1289, + "step": 5933 + }, + { + "epoch": 0.05150996953151448, + "grad_norm": 0.16796875, + "learning_rate": 0.0019880512093654085, + "loss": 0.1533, + "step": 5934 + }, + { + "epoch": 0.051518650011718646, + "grad_norm": 0.146484375, + "learning_rate": 0.0019880463751962263, + "loss": 0.1367, + "step": 5935 + }, + { + "epoch": 0.05152733049192281, + "grad_norm": 0.1552734375, + "learning_rate": 0.00198804154005589, + "loss": 0.1504, + "step": 5936 + }, + { + "epoch": 0.051536010972126976, + "grad_norm": 0.1279296875, + "learning_rate": 0.001988036703944405, + "loss": 0.1602, + "step": 5937 + }, + { + "epoch": 0.05154469145233114, + "grad_norm": 0.37109375, + "learning_rate": 0.0019880318668617763, + "loss": 0.1069, + "step": 5938 + }, + { + "epoch": 0.05155337193253531, + "grad_norm": 0.36328125, + "learning_rate": 0.001988027028808009, + "loss": 0.1533, + "step": 5939 + }, + { + "epoch": 0.05156205241273947, + "grad_norm": 0.31640625, + "learning_rate": 0.0019880221897831092, + "loss": 0.1699, + "step": 5940 + }, + { + "epoch": 0.05157073289294364, + "grad_norm": 0.390625, + "learning_rate": 0.001988017349787081, + "loss": 0.1602, + "step": 5941 + }, + { + "epoch": 0.0515794133731478, + "grad_norm": 0.23046875, + "learning_rate": 0.0019880125088199314, + "loss": 0.1973, + "step": 5942 + }, + { + "epoch": 0.05158809385335197, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019880076668816636, + "loss": 0.1406, + "step": 5943 + }, + { + "epoch": 0.05159677433355613, + "grad_norm": 0.25390625, + "learning_rate": 0.0019880028239722845, + "loss": 0.1719, + "step": 5944 + }, + { + "epoch": 0.0516054548137603, + "grad_norm": 0.26953125, + "learning_rate": 0.0019879979800917985, + "loss": 0.1543, + "step": 5945 + }, + { + "epoch": 0.05161413529396446, + "grad_norm": 0.0634765625, + "learning_rate": 0.001987993135240212, + "loss": 0.127, + "step": 5946 + }, + { + "epoch": 0.05162281577416863, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019879882894175286, + "loss": 0.1201, + "step": 5947 + }, + { + "epoch": 0.05163149625437279, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019879834426237554, + "loss": 0.126, + "step": 5948 + }, + { + "epoch": 0.05164017673457696, + "grad_norm": 0.3203125, + "learning_rate": 0.001987978594858896, + "loss": 0.166, + "step": 5949 + }, + { + "epoch": 0.05164885721478112, + "grad_norm": 0.84375, + "learning_rate": 0.0019879737461229573, + "loss": 0.1875, + "step": 5950 + }, + { + "epoch": 0.05165753769498529, + "grad_norm": 0.47265625, + "learning_rate": 0.0019879688964159433, + "loss": 0.1758, + "step": 5951 + }, + { + "epoch": 0.05166621817518945, + "grad_norm": 0.140625, + "learning_rate": 0.00198796404573786, + "loss": 0.1445, + "step": 5952 + }, + { + "epoch": 0.05167489865539362, + "grad_norm": 0.423828125, + "learning_rate": 0.001987959194088713, + "loss": 0.1484, + "step": 5953 + }, + { + "epoch": 0.05168357913559778, + "grad_norm": 0.15234375, + "learning_rate": 0.001987954341468507, + "loss": 0.1787, + "step": 5954 + }, + { + "epoch": 0.05169225961580195, + "grad_norm": 0.291015625, + "learning_rate": 0.0019879494878772474, + "loss": 0.168, + "step": 5955 + }, + { + "epoch": 0.051700940096006114, + "grad_norm": 0.115234375, + "learning_rate": 0.0019879446333149393, + "loss": 0.123, + "step": 5956 + }, + { + "epoch": 0.05170962057621028, + "grad_norm": 0.193359375, + "learning_rate": 0.0019879397777815885, + "loss": 0.1182, + "step": 5957 + }, + { + "epoch": 0.051718301056414444, + "grad_norm": 0.1953125, + "learning_rate": 0.0019879349212772, + "loss": 0.1484, + "step": 5958 + }, + { + "epoch": 0.05172698153661861, + "grad_norm": 0.73046875, + "learning_rate": 0.0019879300638017796, + "loss": 0.1611, + "step": 5959 + }, + { + "epoch": 0.051735662016822774, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019879252053553315, + "loss": 0.1484, + "step": 5960 + }, + { + "epoch": 0.05174434249702693, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019879203459378623, + "loss": 0.1108, + "step": 5961 + }, + { + "epoch": 0.0517530229772311, + "grad_norm": 0.55859375, + "learning_rate": 0.0019879154855493766, + "loss": 0.1367, + "step": 5962 + }, + { + "epoch": 0.05176170345743526, + "grad_norm": 0.46875, + "learning_rate": 0.0019879106241898798, + "loss": 0.168, + "step": 5963 + }, + { + "epoch": 0.05177038393763943, + "grad_norm": 0.38671875, + "learning_rate": 0.0019879057618593775, + "loss": 0.1201, + "step": 5964 + }, + { + "epoch": 0.05177906441784359, + "grad_norm": 0.087890625, + "learning_rate": 0.0019879008985578747, + "loss": 0.1416, + "step": 5965 + }, + { + "epoch": 0.05178774489804776, + "grad_norm": 0.1171875, + "learning_rate": 0.001987896034285377, + "loss": 0.1279, + "step": 5966 + }, + { + "epoch": 0.05179642537825192, + "grad_norm": 0.29296875, + "learning_rate": 0.001987891169041889, + "loss": 0.2227, + "step": 5967 + }, + { + "epoch": 0.05180510585845609, + "grad_norm": 0.12060546875, + "learning_rate": 0.001987886302827417, + "loss": 0.1592, + "step": 5968 + }, + { + "epoch": 0.05181378633866025, + "grad_norm": 0.59375, + "learning_rate": 0.0019878814356419656, + "loss": 0.1196, + "step": 5969 + }, + { + "epoch": 0.05182246681886442, + "grad_norm": 0.87890625, + "learning_rate": 0.0019878765674855407, + "loss": 0.1138, + "step": 5970 + }, + { + "epoch": 0.05183114729906858, + "grad_norm": 0.455078125, + "learning_rate": 0.001987871698358147, + "loss": 0.1289, + "step": 5971 + }, + { + "epoch": 0.05183982777927275, + "grad_norm": 0.470703125, + "learning_rate": 0.0019878668282597906, + "loss": 0.1387, + "step": 5972 + }, + { + "epoch": 0.051848508259476914, + "grad_norm": 0.2265625, + "learning_rate": 0.001987861957190476, + "loss": 0.1514, + "step": 5973 + }, + { + "epoch": 0.05185718873968108, + "grad_norm": 0.10400390625, + "learning_rate": 0.001987857085150209, + "loss": 0.125, + "step": 5974 + }, + { + "epoch": 0.051865869219885244, + "grad_norm": 0.2578125, + "learning_rate": 0.0019878522121389948, + "loss": 0.1777, + "step": 5975 + }, + { + "epoch": 0.05187454970008941, + "grad_norm": 0.625, + "learning_rate": 0.0019878473381568385, + "loss": 0.1699, + "step": 5976 + }, + { + "epoch": 0.051883230180293574, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019878424632037463, + "loss": 0.125, + "step": 5977 + }, + { + "epoch": 0.05189191066049774, + "grad_norm": 0.302734375, + "learning_rate": 0.0019878375872797224, + "loss": 0.1494, + "step": 5978 + }, + { + "epoch": 0.051900591140701904, + "grad_norm": 0.404296875, + "learning_rate": 0.001987832710384773, + "loss": 0.1484, + "step": 5979 + }, + { + "epoch": 0.05190927162090607, + "grad_norm": 0.4296875, + "learning_rate": 0.001987827832518903, + "loss": 0.1338, + "step": 5980 + }, + { + "epoch": 0.051917952101110235, + "grad_norm": 0.87890625, + "learning_rate": 0.001987822953682118, + "loss": 0.1055, + "step": 5981 + }, + { + "epoch": 0.0519266325813144, + "grad_norm": 0.19140625, + "learning_rate": 0.0019878180738744228, + "loss": 0.1357, + "step": 5982 + }, + { + "epoch": 0.051935313061518565, + "grad_norm": 0.96484375, + "learning_rate": 0.001987813193095823, + "loss": 0.1621, + "step": 5983 + }, + { + "epoch": 0.05194399354172273, + "grad_norm": 0.3046875, + "learning_rate": 0.0019878083113463244, + "loss": 0.165, + "step": 5984 + }, + { + "epoch": 0.051952674021926895, + "grad_norm": 0.76171875, + "learning_rate": 0.001987803428625932, + "loss": 0.1484, + "step": 5985 + }, + { + "epoch": 0.05196135450213106, + "grad_norm": 0.27734375, + "learning_rate": 0.0019877985449346506, + "loss": 0.1455, + "step": 5986 + }, + { + "epoch": 0.051970034982335225, + "grad_norm": 0.462890625, + "learning_rate": 0.0019877936602724863, + "loss": 0.1084, + "step": 5987 + }, + { + "epoch": 0.05197871546253939, + "grad_norm": 0.4921875, + "learning_rate": 0.0019877887746394445, + "loss": 0.1836, + "step": 5988 + }, + { + "epoch": 0.051987395942743556, + "grad_norm": 0.267578125, + "learning_rate": 0.00198778388803553, + "loss": 0.1221, + "step": 5989 + }, + { + "epoch": 0.05199607642294772, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019877790004607485, + "loss": 0.1299, + "step": 5990 + }, + { + "epoch": 0.052004756903151886, + "grad_norm": 0.40234375, + "learning_rate": 0.001987774111915105, + "loss": 0.1426, + "step": 5991 + }, + { + "epoch": 0.052013437383356044, + "grad_norm": 0.57421875, + "learning_rate": 0.001987769222398605, + "loss": 0.1621, + "step": 5992 + }, + { + "epoch": 0.05202211786356021, + "grad_norm": 0.12890625, + "learning_rate": 0.0019877643319112545, + "loss": 0.1582, + "step": 5993 + }, + { + "epoch": 0.052030798343764374, + "grad_norm": 0.193359375, + "learning_rate": 0.001987759440453058, + "loss": 0.1187, + "step": 5994 + }, + { + "epoch": 0.05203947882396854, + "grad_norm": 0.11328125, + "learning_rate": 0.0019877545480240207, + "loss": 0.1377, + "step": 5995 + }, + { + "epoch": 0.052048159304172704, + "grad_norm": 0.31640625, + "learning_rate": 0.0019877496546241488, + "loss": 0.1406, + "step": 5996 + }, + { + "epoch": 0.05205683978437687, + "grad_norm": 0.63671875, + "learning_rate": 0.001987744760253447, + "loss": 0.1484, + "step": 5997 + }, + { + "epoch": 0.052065520264581035, + "grad_norm": 0.53125, + "learning_rate": 0.001987739864911921, + "loss": 0.1602, + "step": 5998 + }, + { + "epoch": 0.0520742007447852, + "grad_norm": 0.69921875, + "learning_rate": 0.0019877349685995757, + "loss": 0.1416, + "step": 5999 + }, + { + "epoch": 0.052082881224989365, + "grad_norm": 0.400390625, + "learning_rate": 0.0019877300713164172, + "loss": 0.1348, + "step": 6000 + }, + { + "epoch": 0.05209156170519353, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019877251730624503, + "loss": 0.1602, + "step": 6001 + }, + { + "epoch": 0.052100242185397695, + "grad_norm": 0.181640625, + "learning_rate": 0.0019877202738376805, + "loss": 0.1816, + "step": 6002 + }, + { + "epoch": 0.05210892266560186, + "grad_norm": 0.271484375, + "learning_rate": 0.001987715373642113, + "loss": 0.1201, + "step": 6003 + }, + { + "epoch": 0.052117603145806025, + "grad_norm": 0.10595703125, + "learning_rate": 0.001987710472475753, + "loss": 0.1514, + "step": 6004 + }, + { + "epoch": 0.05212628362601019, + "grad_norm": 0.267578125, + "learning_rate": 0.001987705570338607, + "loss": 0.1641, + "step": 6005 + }, + { + "epoch": 0.052134964106214356, + "grad_norm": 0.220703125, + "learning_rate": 0.001987700667230679, + "loss": 0.1543, + "step": 6006 + }, + { + "epoch": 0.05214364458641852, + "grad_norm": 0.53125, + "learning_rate": 0.001987695763151975, + "loss": 0.166, + "step": 6007 + }, + { + "epoch": 0.052152325066622686, + "grad_norm": 0.6484375, + "learning_rate": 0.0019876908581025004, + "loss": 0.1357, + "step": 6008 + }, + { + "epoch": 0.05216100554682685, + "grad_norm": 0.236328125, + "learning_rate": 0.0019876859520822597, + "loss": 0.1895, + "step": 6009 + }, + { + "epoch": 0.052169686027031016, + "grad_norm": 0.091796875, + "learning_rate": 0.00198768104509126, + "loss": 0.1338, + "step": 6010 + }, + { + "epoch": 0.05217836650723518, + "grad_norm": 0.57421875, + "learning_rate": 0.001987676137129505, + "loss": 0.2109, + "step": 6011 + }, + { + "epoch": 0.052187046987439346, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019876712281970007, + "loss": 0.1572, + "step": 6012 + }, + { + "epoch": 0.05219572746764351, + "grad_norm": 0.482421875, + "learning_rate": 0.001987666318293753, + "loss": 0.1025, + "step": 6013 + }, + { + "epoch": 0.05220440794784768, + "grad_norm": 0.62109375, + "learning_rate": 0.0019876614074197663, + "loss": 0.1914, + "step": 6014 + }, + { + "epoch": 0.05221308842805184, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019876564955750462, + "loss": 0.1533, + "step": 6015 + }, + { + "epoch": 0.05222176890825601, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019876515827595993, + "loss": 0.1641, + "step": 6016 + }, + { + "epoch": 0.05223044938846017, + "grad_norm": 0.255859375, + "learning_rate": 0.001987646668973429, + "loss": 0.165, + "step": 6017 + }, + { + "epoch": 0.05223912986866434, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019876417542165417, + "loss": 0.1611, + "step": 6018 + }, + { + "epoch": 0.0522478103488685, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019876368384889433, + "loss": 0.1768, + "step": 6019 + }, + { + "epoch": 0.05225649082907267, + "grad_norm": 0.380859375, + "learning_rate": 0.0019876319217906383, + "loss": 0.1377, + "step": 6020 + }, + { + "epoch": 0.05226517130927683, + "grad_norm": 0.345703125, + "learning_rate": 0.0019876270041216323, + "loss": 0.1338, + "step": 6021 + }, + { + "epoch": 0.05227385178948099, + "grad_norm": 0.306640625, + "learning_rate": 0.001987622085481931, + "loss": 0.1338, + "step": 6022 + }, + { + "epoch": 0.052282532269685156, + "grad_norm": 0.2001953125, + "learning_rate": 0.001987617165871539, + "loss": 0.1514, + "step": 6023 + }, + { + "epoch": 0.05229121274988932, + "grad_norm": 0.6015625, + "learning_rate": 0.0019876122452904626, + "loss": 0.1436, + "step": 6024 + }, + { + "epoch": 0.052299893230093486, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019876073237387066, + "loss": 0.1245, + "step": 6025 + }, + { + "epoch": 0.05230857371029765, + "grad_norm": 0.6015625, + "learning_rate": 0.0019876024012162766, + "loss": 0.1924, + "step": 6026 + }, + { + "epoch": 0.052317254190501816, + "grad_norm": 0.31640625, + "learning_rate": 0.001987597477723178, + "loss": 0.1543, + "step": 6027 + }, + { + "epoch": 0.05232593467070598, + "grad_norm": 0.828125, + "learning_rate": 0.001987592553259416, + "loss": 0.1279, + "step": 6028 + }, + { + "epoch": 0.052334615150910146, + "grad_norm": 0.162109375, + "learning_rate": 0.0019875876278249963, + "loss": 0.1543, + "step": 6029 + }, + { + "epoch": 0.05234329563111431, + "grad_norm": 0.408203125, + "learning_rate": 0.0019875827014199238, + "loss": 0.1387, + "step": 6030 + }, + { + "epoch": 0.05235197611131848, + "grad_norm": 0.75, + "learning_rate": 0.0019875777740442045, + "loss": 0.1953, + "step": 6031 + }, + { + "epoch": 0.05236065659152264, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019875728456978437, + "loss": 0.1406, + "step": 6032 + }, + { + "epoch": 0.05236933707172681, + "grad_norm": 0.212890625, + "learning_rate": 0.001987567916380846, + "loss": 0.1196, + "step": 6033 + }, + { + "epoch": 0.05237801755193097, + "grad_norm": 0.83984375, + "learning_rate": 0.001987562986093218, + "loss": 0.1465, + "step": 6034 + }, + { + "epoch": 0.05238669803213514, + "grad_norm": 0.083984375, + "learning_rate": 0.001987558054834964, + "loss": 0.0977, + "step": 6035 + }, + { + "epoch": 0.0523953785123393, + "grad_norm": 0.333984375, + "learning_rate": 0.00198755312260609, + "loss": 0.1426, + "step": 6036 + }, + { + "epoch": 0.05240405899254347, + "grad_norm": 0.51953125, + "learning_rate": 0.001987548189406601, + "loss": 0.1475, + "step": 6037 + }, + { + "epoch": 0.05241273947274763, + "grad_norm": 0.1337890625, + "learning_rate": 0.001987543255236503, + "loss": 0.1396, + "step": 6038 + }, + { + "epoch": 0.0524214199529518, + "grad_norm": 0.173828125, + "learning_rate": 0.001987538320095801, + "loss": 0.1299, + "step": 6039 + }, + { + "epoch": 0.05243010043315596, + "grad_norm": 0.291015625, + "learning_rate": 0.0019875333839845, + "loss": 0.1846, + "step": 6040 + }, + { + "epoch": 0.05243878091336013, + "grad_norm": 0.6640625, + "learning_rate": 0.0019875284469026066, + "loss": 0.1357, + "step": 6041 + }, + { + "epoch": 0.05244746139356429, + "grad_norm": 0.107421875, + "learning_rate": 0.0019875235088501245, + "loss": 0.1592, + "step": 6042 + }, + { + "epoch": 0.05245614187376846, + "grad_norm": 0.078125, + "learning_rate": 0.0019875185698270607, + "loss": 0.1641, + "step": 6043 + }, + { + "epoch": 0.05246482235397262, + "grad_norm": 1.4765625, + "learning_rate": 0.0019875136298334196, + "loss": 0.1387, + "step": 6044 + }, + { + "epoch": 0.05247350283417679, + "grad_norm": 0.234375, + "learning_rate": 0.001987508688869207, + "loss": 0.106, + "step": 6045 + }, + { + "epoch": 0.052482183314380954, + "grad_norm": 0.08740234375, + "learning_rate": 0.001987503746934428, + "loss": 0.1426, + "step": 6046 + }, + { + "epoch": 0.05249086379458512, + "grad_norm": 0.228515625, + "learning_rate": 0.001987498804029089, + "loss": 0.1914, + "step": 6047 + }, + { + "epoch": 0.052499544274789284, + "grad_norm": 0.11865234375, + "learning_rate": 0.001987493860153194, + "loss": 0.125, + "step": 6048 + }, + { + "epoch": 0.05250822475499345, + "grad_norm": 0.76953125, + "learning_rate": 0.001987488915306749, + "loss": 0.1328, + "step": 6049 + }, + { + "epoch": 0.052516905235197614, + "grad_norm": 0.21875, + "learning_rate": 0.0019874839694897597, + "loss": 0.1367, + "step": 6050 + }, + { + "epoch": 0.05252558571540178, + "grad_norm": 0.875, + "learning_rate": 0.001987479022702231, + "loss": 0.166, + "step": 6051 + }, + { + "epoch": 0.052534266195605944, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019874740749441693, + "loss": 0.1328, + "step": 6052 + }, + { + "epoch": 0.0525429466758101, + "grad_norm": 0.51171875, + "learning_rate": 0.0019874691262155787, + "loss": 0.167, + "step": 6053 + }, + { + "epoch": 0.05255162715601427, + "grad_norm": 0.150390625, + "learning_rate": 0.0019874641765164655, + "loss": 0.127, + "step": 6054 + }, + { + "epoch": 0.05256030763621843, + "grad_norm": 0.208984375, + "learning_rate": 0.0019874592258468343, + "loss": 0.0942, + "step": 6055 + }, + { + "epoch": 0.0525689881164226, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019874542742066913, + "loss": 0.1309, + "step": 6056 + }, + { + "epoch": 0.05257766859662676, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019874493215960415, + "loss": 0.2227, + "step": 6057 + }, + { + "epoch": 0.05258634907683093, + "grad_norm": 0.3671875, + "learning_rate": 0.001987444368014891, + "loss": 0.1396, + "step": 6058 + }, + { + "epoch": 0.05259502955703509, + "grad_norm": 0.38671875, + "learning_rate": 0.0019874394134632442, + "loss": 0.127, + "step": 6059 + }, + { + "epoch": 0.05260371003723926, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019874344579411066, + "loss": 0.2148, + "step": 6060 + }, + { + "epoch": 0.05261239051744342, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019874295014484845, + "loss": 0.124, + "step": 6061 + }, + { + "epoch": 0.05262107099764759, + "grad_norm": 0.427734375, + "learning_rate": 0.001987424543985383, + "loss": 0.1416, + "step": 6062 + }, + { + "epoch": 0.052629751477851754, + "grad_norm": 0.26953125, + "learning_rate": 0.001987419585551807, + "loss": 0.168, + "step": 6063 + }, + { + "epoch": 0.05263843195805592, + "grad_norm": 0.1484375, + "learning_rate": 0.001987414626147763, + "loss": 0.1221, + "step": 6064 + }, + { + "epoch": 0.052647112438260084, + "grad_norm": 0.31640625, + "learning_rate": 0.001987409665773255, + "loss": 0.168, + "step": 6065 + }, + { + "epoch": 0.05265579291846425, + "grad_norm": 0.2412109375, + "learning_rate": 0.001987404704428289, + "loss": 0.1338, + "step": 6066 + }, + { + "epoch": 0.052664473398668414, + "grad_norm": 0.2138671875, + "learning_rate": 0.001987399742112871, + "loss": 0.126, + "step": 6067 + }, + { + "epoch": 0.05267315387887258, + "grad_norm": 0.07421875, + "learning_rate": 0.0019873947788270055, + "loss": 0.1104, + "step": 6068 + }, + { + "epoch": 0.052681834359076744, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019873898145706986, + "loss": 0.1387, + "step": 6069 + }, + { + "epoch": 0.05269051483928091, + "grad_norm": 0.20703125, + "learning_rate": 0.001987384849343956, + "loss": 0.126, + "step": 6070 + }, + { + "epoch": 0.052699195319485075, + "grad_norm": 0.115234375, + "learning_rate": 0.0019873798831467823, + "loss": 0.125, + "step": 6071 + }, + { + "epoch": 0.05270787579968924, + "grad_norm": 0.162109375, + "learning_rate": 0.0019873749159791828, + "loss": 0.1475, + "step": 6072 + }, + { + "epoch": 0.052716556279893405, + "grad_norm": 0.3125, + "learning_rate": 0.001987369947841164, + "loss": 0.168, + "step": 6073 + }, + { + "epoch": 0.05272523676009757, + "grad_norm": 0.345703125, + "learning_rate": 0.001987364978732731, + "loss": 0.1445, + "step": 6074 + }, + { + "epoch": 0.052733917240301735, + "grad_norm": 0.31640625, + "learning_rate": 0.0019873600086538885, + "loss": 0.1387, + "step": 6075 + }, + { + "epoch": 0.0527425977205059, + "grad_norm": 0.26171875, + "learning_rate": 0.0019873550376046426, + "loss": 0.2383, + "step": 6076 + }, + { + "epoch": 0.052751278200710065, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019873500655849985, + "loss": 0.1592, + "step": 6077 + }, + { + "epoch": 0.05275995868091423, + "grad_norm": 0.234375, + "learning_rate": 0.001987345092594962, + "loss": 0.2207, + "step": 6078 + }, + { + "epoch": 0.052768639161118396, + "grad_norm": 0.103515625, + "learning_rate": 0.0019873401186345377, + "loss": 0.1348, + "step": 6079 + }, + { + "epoch": 0.05277731964132256, + "grad_norm": 0.10791015625, + "learning_rate": 0.001987335143703732, + "loss": 0.1416, + "step": 6080 + }, + { + "epoch": 0.052786000121526726, + "grad_norm": 0.2119140625, + "learning_rate": 0.00198733016780255, + "loss": 0.1719, + "step": 6081 + }, + { + "epoch": 0.05279468060173089, + "grad_norm": 0.37109375, + "learning_rate": 0.0019873251909309966, + "loss": 0.1641, + "step": 6082 + }, + { + "epoch": 0.052803361081935056, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019873202130890775, + "loss": 0.1055, + "step": 6083 + }, + { + "epoch": 0.052812041562139214, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019873152342767993, + "loss": 0.1348, + "step": 6084 + }, + { + "epoch": 0.05282072204234338, + "grad_norm": 0.095703125, + "learning_rate": 0.0019873102544941657, + "loss": 0.1953, + "step": 6085 + }, + { + "epoch": 0.052829402522547544, + "grad_norm": 0.361328125, + "learning_rate": 0.0019873052737411833, + "loss": 0.1172, + "step": 6086 + }, + { + "epoch": 0.05283808300275171, + "grad_norm": 0.78515625, + "learning_rate": 0.001987300292017857, + "loss": 0.1738, + "step": 6087 + }, + { + "epoch": 0.052846763482955875, + "grad_norm": 0.134765625, + "learning_rate": 0.0019872953093241926, + "loss": 0.1768, + "step": 6088 + }, + { + "epoch": 0.05285544396316004, + "grad_norm": 0.2734375, + "learning_rate": 0.001987290325660195, + "loss": 0.1396, + "step": 6089 + }, + { + "epoch": 0.052864124443364205, + "grad_norm": 0.09130859375, + "learning_rate": 0.001987285341025871, + "loss": 0.1143, + "step": 6090 + }, + { + "epoch": 0.05287280492356837, + "grad_norm": 0.255859375, + "learning_rate": 0.001987280355421224, + "loss": 0.1826, + "step": 6091 + }, + { + "epoch": 0.052881485403772535, + "grad_norm": 0.224609375, + "learning_rate": 0.001987275368846261, + "loss": 0.1562, + "step": 6092 + }, + { + "epoch": 0.0528901658839767, + "grad_norm": 0.796875, + "learning_rate": 0.001987270381300987, + "loss": 0.1641, + "step": 6093 + }, + { + "epoch": 0.052898846364180865, + "grad_norm": 0.26171875, + "learning_rate": 0.0019872653927854073, + "loss": 0.1699, + "step": 6094 + }, + { + "epoch": 0.05290752684438503, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019872604032995273, + "loss": 0.1377, + "step": 6095 + }, + { + "epoch": 0.052916207324589196, + "grad_norm": 0.5859375, + "learning_rate": 0.001987255412843353, + "loss": 0.127, + "step": 6096 + }, + { + "epoch": 0.05292488780479336, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019872504214168895, + "loss": 0.1172, + "step": 6097 + }, + { + "epoch": 0.052933568284997526, + "grad_norm": 0.396484375, + "learning_rate": 0.0019872454290201422, + "loss": 0.1621, + "step": 6098 + }, + { + "epoch": 0.05294224876520169, + "grad_norm": 0.1884765625, + "learning_rate": 0.001987240435653117, + "loss": 0.1152, + "step": 6099 + }, + { + "epoch": 0.052950929245405856, + "grad_norm": 0.15234375, + "learning_rate": 0.0019872354413158185, + "loss": 0.2197, + "step": 6100 + }, + { + "epoch": 0.05295960972561002, + "grad_norm": 0.515625, + "learning_rate": 0.0019872304460082525, + "loss": 0.1533, + "step": 6101 + }, + { + "epoch": 0.052968290205814186, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019872254497304252, + "loss": 0.1465, + "step": 6102 + }, + { + "epoch": 0.05297697068601835, + "grad_norm": 1.28125, + "learning_rate": 0.001987220452482341, + "loss": 0.2383, + "step": 6103 + }, + { + "epoch": 0.05298565116622252, + "grad_norm": 0.083984375, + "learning_rate": 0.0019872154542640063, + "loss": 0.1318, + "step": 6104 + }, + { + "epoch": 0.05299433164642668, + "grad_norm": 0.349609375, + "learning_rate": 0.001987210455075426, + "loss": 0.1523, + "step": 6105 + }, + { + "epoch": 0.05300301212663085, + "grad_norm": 0.431640625, + "learning_rate": 0.0019872054549166054, + "loss": 0.1465, + "step": 6106 + }, + { + "epoch": 0.05301169260683501, + "grad_norm": 0.26953125, + "learning_rate": 0.0019872004537875502, + "loss": 0.1533, + "step": 6107 + }, + { + "epoch": 0.05302037308703918, + "grad_norm": 0.58203125, + "learning_rate": 0.001987195451688266, + "loss": 0.1357, + "step": 6108 + }, + { + "epoch": 0.05302905356724334, + "grad_norm": 0.099609375, + "learning_rate": 0.0019871904486187583, + "loss": 0.1709, + "step": 6109 + }, + { + "epoch": 0.05303773404744751, + "grad_norm": 0.921875, + "learning_rate": 0.0019871854445790327, + "loss": 0.166, + "step": 6110 + }, + { + "epoch": 0.05304641452765167, + "grad_norm": 0.103515625, + "learning_rate": 0.001987180439569094, + "loss": 0.1099, + "step": 6111 + }, + { + "epoch": 0.05305509500785584, + "grad_norm": 0.73828125, + "learning_rate": 0.001987175433588948, + "loss": 0.1016, + "step": 6112 + }, + { + "epoch": 0.05306377548806, + "grad_norm": 0.373046875, + "learning_rate": 0.0019871704266386007, + "loss": 0.1582, + "step": 6113 + }, + { + "epoch": 0.05307245596826417, + "grad_norm": 0.44921875, + "learning_rate": 0.001987165418718057, + "loss": 0.2266, + "step": 6114 + }, + { + "epoch": 0.053081136448468326, + "grad_norm": 0.12890625, + "learning_rate": 0.0019871604098273222, + "loss": 0.1416, + "step": 6115 + }, + { + "epoch": 0.05308981692867249, + "grad_norm": 0.265625, + "learning_rate": 0.001987155399966403, + "loss": 0.125, + "step": 6116 + }, + { + "epoch": 0.053098497408876656, + "grad_norm": 0.640625, + "learning_rate": 0.001987150389135303, + "loss": 0.1406, + "step": 6117 + }, + { + "epoch": 0.05310717788908082, + "grad_norm": 0.2412109375, + "learning_rate": 0.001987145377334029, + "loss": 0.1738, + "step": 6118 + }, + { + "epoch": 0.053115858369284986, + "grad_norm": 3.203125, + "learning_rate": 0.0019871403645625864, + "loss": 0.6641, + "step": 6119 + }, + { + "epoch": 0.05312453884948915, + "grad_norm": 0.283203125, + "learning_rate": 0.00198713535082098, + "loss": 0.1406, + "step": 6120 + }, + { + "epoch": 0.05313321932969332, + "grad_norm": 3.703125, + "learning_rate": 0.0019871303361092157, + "loss": 0.1406, + "step": 6121 + }, + { + "epoch": 0.05314189980989748, + "grad_norm": 0.59765625, + "learning_rate": 0.0019871253204272995, + "loss": 0.2637, + "step": 6122 + }, + { + "epoch": 0.05315058029010165, + "grad_norm": 0.66015625, + "learning_rate": 0.001987120303775236, + "loss": 0.1279, + "step": 6123 + }, + { + "epoch": 0.05315926077030581, + "grad_norm": 0.1181640625, + "learning_rate": 0.001987115286153031, + "loss": 0.1455, + "step": 6124 + }, + { + "epoch": 0.05316794125050998, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019871102675606904, + "loss": 0.1582, + "step": 6125 + }, + { + "epoch": 0.05317662173071414, + "grad_norm": 0.765625, + "learning_rate": 0.001987105247998219, + "loss": 0.1289, + "step": 6126 + }, + { + "epoch": 0.05318530221091831, + "grad_norm": 0.7734375, + "learning_rate": 0.001987100227465623, + "loss": 0.1738, + "step": 6127 + }, + { + "epoch": 0.05319398269112247, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019870952059629072, + "loss": 0.126, + "step": 6128 + }, + { + "epoch": 0.05320266317132664, + "grad_norm": 0.201171875, + "learning_rate": 0.0019870901834900777, + "loss": 0.1318, + "step": 6129 + }, + { + "epoch": 0.0532113436515308, + "grad_norm": 0.1220703125, + "learning_rate": 0.001987085160047139, + "loss": 0.1191, + "step": 6130 + }, + { + "epoch": 0.05322002413173497, + "grad_norm": 0.2109375, + "learning_rate": 0.001987080135634098, + "loss": 0.2021, + "step": 6131 + }, + { + "epoch": 0.05322870461193913, + "grad_norm": 0.10107421875, + "learning_rate": 0.00198707511025096, + "loss": 0.1514, + "step": 6132 + }, + { + "epoch": 0.0532373850921433, + "grad_norm": 0.51171875, + "learning_rate": 0.001987070083897729, + "loss": 0.1318, + "step": 6133 + }, + { + "epoch": 0.05324606557234746, + "grad_norm": 0.83984375, + "learning_rate": 0.0019870650565744115, + "loss": 0.1631, + "step": 6134 + }, + { + "epoch": 0.05325474605255163, + "grad_norm": 0.578125, + "learning_rate": 0.0019870600282810135, + "loss": 0.1113, + "step": 6135 + }, + { + "epoch": 0.053263426532755793, + "grad_norm": 0.40625, + "learning_rate": 0.0019870549990175395, + "loss": 0.2275, + "step": 6136 + }, + { + "epoch": 0.05327210701295996, + "grad_norm": 0.0869140625, + "learning_rate": 0.001987049968783996, + "loss": 0.1006, + "step": 6137 + }, + { + "epoch": 0.053280787493164124, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019870449375803878, + "loss": 0.1055, + "step": 6138 + }, + { + "epoch": 0.05328946797336829, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019870399054067204, + "loss": 0.1475, + "step": 6139 + }, + { + "epoch": 0.053298148453572454, + "grad_norm": 0.4453125, + "learning_rate": 0.001987034872262999, + "loss": 0.1348, + "step": 6140 + }, + { + "epoch": 0.05330682893377662, + "grad_norm": 0.30859375, + "learning_rate": 0.0019870298381492305, + "loss": 0.125, + "step": 6141 + }, + { + "epoch": 0.053315509413980784, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019870248030654193, + "loss": 0.166, + "step": 6142 + }, + { + "epoch": 0.05332418989418495, + "grad_norm": 0.2265625, + "learning_rate": 0.001987019767011571, + "loss": 0.1533, + "step": 6143 + }, + { + "epoch": 0.053332870374389114, + "grad_norm": 0.4921875, + "learning_rate": 0.0019870147299876914, + "loss": 0.1406, + "step": 6144 + }, + { + "epoch": 0.05334155085459328, + "grad_norm": 0.197265625, + "learning_rate": 0.0019870096919937853, + "loss": 0.1226, + "step": 6145 + }, + { + "epoch": 0.05335023133479744, + "grad_norm": 0.52734375, + "learning_rate": 0.001987004653029859, + "loss": 0.1157, + "step": 6146 + }, + { + "epoch": 0.0533589118150016, + "grad_norm": 0.6328125, + "learning_rate": 0.001986999613095918, + "loss": 0.1504, + "step": 6147 + }, + { + "epoch": 0.05336759229520577, + "grad_norm": 0.41015625, + "learning_rate": 0.001986994572191967, + "loss": 0.1465, + "step": 6148 + }, + { + "epoch": 0.05337627277540993, + "grad_norm": 0.4140625, + "learning_rate": 0.0019869895303180124, + "loss": 0.1367, + "step": 6149 + }, + { + "epoch": 0.0533849532556141, + "grad_norm": 0.68359375, + "learning_rate": 0.0019869844874740596, + "loss": 0.1572, + "step": 6150 + }, + { + "epoch": 0.05339363373581826, + "grad_norm": 0.3828125, + "learning_rate": 0.0019869794436601137, + "loss": 0.1465, + "step": 6151 + }, + { + "epoch": 0.05340231421602243, + "grad_norm": 0.263671875, + "learning_rate": 0.00198697439887618, + "loss": 0.1611, + "step": 6152 + }, + { + "epoch": 0.053410994696226594, + "grad_norm": 0.07666015625, + "learning_rate": 0.001986969353122265, + "loss": 0.1001, + "step": 6153 + }, + { + "epoch": 0.05341967517643076, + "grad_norm": 0.64453125, + "learning_rate": 0.001986964306398373, + "loss": 0.1465, + "step": 6154 + }, + { + "epoch": 0.053428355656634924, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019869592587045113, + "loss": 0.1543, + "step": 6155 + }, + { + "epoch": 0.05343703613683909, + "grad_norm": 0.244140625, + "learning_rate": 0.0019869542100406836, + "loss": 0.1543, + "step": 6156 + }, + { + "epoch": 0.053445716617043254, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019869491604068957, + "loss": 0.1406, + "step": 6157 + }, + { + "epoch": 0.05345439709724742, + "grad_norm": 0.076171875, + "learning_rate": 0.0019869441098031536, + "loss": 0.127, + "step": 6158 + }, + { + "epoch": 0.053463077577451584, + "grad_norm": 0.546875, + "learning_rate": 0.0019869390582294634, + "loss": 0.1143, + "step": 6159 + }, + { + "epoch": 0.05347175805765575, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019869340056858295, + "loss": 0.1572, + "step": 6160 + }, + { + "epoch": 0.053480438537859915, + "grad_norm": 0.36328125, + "learning_rate": 0.0019869289521722584, + "loss": 0.123, + "step": 6161 + }, + { + "epoch": 0.05348911901806408, + "grad_norm": 0.212890625, + "learning_rate": 0.0019869238976887543, + "loss": 0.1377, + "step": 6162 + }, + { + "epoch": 0.053497799498268245, + "grad_norm": 1.2109375, + "learning_rate": 0.0019869188422353243, + "loss": 0.1318, + "step": 6163 + }, + { + "epoch": 0.05350647997847241, + "grad_norm": 0.73046875, + "learning_rate": 0.0019869137858119727, + "loss": 0.1182, + "step": 6164 + }, + { + "epoch": 0.053515160458676575, + "grad_norm": 0.23046875, + "learning_rate": 0.0019869087284187055, + "loss": 0.1543, + "step": 6165 + }, + { + "epoch": 0.05352384093888074, + "grad_norm": 0.609375, + "learning_rate": 0.0019869036700555284, + "loss": 0.1582, + "step": 6166 + }, + { + "epoch": 0.053532521419084905, + "grad_norm": 0.2236328125, + "learning_rate": 0.001986898610722447, + "loss": 0.1426, + "step": 6167 + }, + { + "epoch": 0.05354120189928907, + "grad_norm": 0.25, + "learning_rate": 0.001986893550419466, + "loss": 0.1064, + "step": 6168 + }, + { + "epoch": 0.053549882379493235, + "grad_norm": 0.39453125, + "learning_rate": 0.001986888489146592, + "loss": 0.1064, + "step": 6169 + }, + { + "epoch": 0.0535585628596974, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019868834269038305, + "loss": 0.1523, + "step": 6170 + }, + { + "epoch": 0.053567243339901566, + "grad_norm": 0.62890625, + "learning_rate": 0.001986878363691186, + "loss": 0.1768, + "step": 6171 + }, + { + "epoch": 0.05357592382010573, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019868732995086645, + "loss": 0.1406, + "step": 6172 + }, + { + "epoch": 0.053584604300309896, + "grad_norm": 0.58203125, + "learning_rate": 0.001986868234356272, + "loss": 0.1348, + "step": 6173 + }, + { + "epoch": 0.05359328478051406, + "grad_norm": 0.453125, + "learning_rate": 0.001986863168234014, + "loss": 0.1416, + "step": 6174 + }, + { + "epoch": 0.053601965260718226, + "grad_norm": 0.244140625, + "learning_rate": 0.001986858101141895, + "loss": 0.1235, + "step": 6175 + }, + { + "epoch": 0.053610645740922384, + "grad_norm": 0.462890625, + "learning_rate": 0.001986853033079922, + "loss": 0.1719, + "step": 6176 + }, + { + "epoch": 0.05361932622112655, + "grad_norm": 0.2578125, + "learning_rate": 0.0019868479640480995, + "loss": 0.1416, + "step": 6177 + }, + { + "epoch": 0.053628006701330715, + "grad_norm": 0.302734375, + "learning_rate": 0.001986842894046433, + "loss": 0.125, + "step": 6178 + }, + { + "epoch": 0.05363668718153488, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019868378230749294, + "loss": 0.1689, + "step": 6179 + }, + { + "epoch": 0.053645367661739045, + "grad_norm": 0.1748046875, + "learning_rate": 0.001986832751133593, + "loss": 0.1504, + "step": 6180 + }, + { + "epoch": 0.05365404814194321, + "grad_norm": 0.166015625, + "learning_rate": 0.001986827678222429, + "loss": 0.1797, + "step": 6181 + }, + { + "epoch": 0.053662728622147375, + "grad_norm": 0.08837890625, + "learning_rate": 0.001986822604341444, + "loss": 0.1494, + "step": 6182 + }, + { + "epoch": 0.05367140910235154, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019868175294906433, + "loss": 0.1172, + "step": 6183 + }, + { + "epoch": 0.053680089582555705, + "grad_norm": 0.92578125, + "learning_rate": 0.0019868124536700322, + "loss": 0.2168, + "step": 6184 + }, + { + "epoch": 0.05368877006275987, + "grad_norm": 0.14453125, + "learning_rate": 0.001986807376879616, + "loss": 0.1631, + "step": 6185 + }, + { + "epoch": 0.053697450542964036, + "grad_norm": 0.1083984375, + "learning_rate": 0.001986802299119401, + "loss": 0.1221, + "step": 6186 + }, + { + "epoch": 0.0537061310231682, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019867972203893923, + "loss": 0.1289, + "step": 6187 + }, + { + "epoch": 0.053714811503372366, + "grad_norm": 0.201171875, + "learning_rate": 0.001986792140689595, + "loss": 0.1387, + "step": 6188 + }, + { + "epoch": 0.05372349198357653, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019867870600200157, + "loss": 0.1245, + "step": 6189 + }, + { + "epoch": 0.053732172463780696, + "grad_norm": 0.515625, + "learning_rate": 0.001986781978380659, + "loss": 0.1553, + "step": 6190 + }, + { + "epoch": 0.05374085294398486, + "grad_norm": 0.52734375, + "learning_rate": 0.001986776895771531, + "loss": 0.123, + "step": 6191 + }, + { + "epoch": 0.053749533424189026, + "grad_norm": 0.63671875, + "learning_rate": 0.001986771812192637, + "loss": 0.1211, + "step": 6192 + }, + { + "epoch": 0.05375821390439319, + "grad_norm": 0.08544921875, + "learning_rate": 0.001986766727643983, + "loss": 0.166, + "step": 6193 + }, + { + "epoch": 0.05376689438459736, + "grad_norm": 0.5390625, + "learning_rate": 0.0019867616421255737, + "loss": 0.1855, + "step": 6194 + }, + { + "epoch": 0.05377557486480152, + "grad_norm": 0.390625, + "learning_rate": 0.0019867565556374155, + "loss": 0.1396, + "step": 6195 + }, + { + "epoch": 0.05378425534500569, + "grad_norm": 1.5234375, + "learning_rate": 0.001986751468179514, + "loss": 0.1582, + "step": 6196 + }, + { + "epoch": 0.05379293582520985, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019867463797518738, + "loss": 0.1143, + "step": 6197 + }, + { + "epoch": 0.05380161630541402, + "grad_norm": 0.09375, + "learning_rate": 0.0019867412903545015, + "loss": 0.1245, + "step": 6198 + }, + { + "epoch": 0.05381029678561818, + "grad_norm": 0.1591796875, + "learning_rate": 0.001986736199987402, + "loss": 0.1387, + "step": 6199 + }, + { + "epoch": 0.05381897726582235, + "grad_norm": 0.30078125, + "learning_rate": 0.0019867311086505814, + "loss": 0.1396, + "step": 6200 + }, + { + "epoch": 0.05382765774602651, + "grad_norm": 0.40234375, + "learning_rate": 0.001986726016344045, + "loss": 0.1021, + "step": 6201 + }, + { + "epoch": 0.05383633822623068, + "grad_norm": 0.08154296875, + "learning_rate": 0.001986720923067798, + "loss": 0.1221, + "step": 6202 + }, + { + "epoch": 0.05384501870643484, + "grad_norm": 0.30078125, + "learning_rate": 0.0019867158288218462, + "loss": 0.1777, + "step": 6203 + }, + { + "epoch": 0.05385369918663901, + "grad_norm": 0.1943359375, + "learning_rate": 0.001986710733606196, + "loss": 0.1436, + "step": 6204 + }, + { + "epoch": 0.05386237966684317, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019867056374208514, + "loss": 0.1445, + "step": 6205 + }, + { + "epoch": 0.05387106014704734, + "grad_norm": 0.33984375, + "learning_rate": 0.001986700540265819, + "loss": 0.1348, + "step": 6206 + }, + { + "epoch": 0.053879740627251496, + "grad_norm": 0.376953125, + "learning_rate": 0.001986695442141105, + "loss": 0.1602, + "step": 6207 + }, + { + "epoch": 0.05388842110745566, + "grad_norm": 0.5078125, + "learning_rate": 0.001986690343046713, + "loss": 0.1553, + "step": 6208 + }, + { + "epoch": 0.053897101587659826, + "grad_norm": 0.2578125, + "learning_rate": 0.001986685242982651, + "loss": 0.1924, + "step": 6209 + }, + { + "epoch": 0.05390578206786399, + "grad_norm": 0.52734375, + "learning_rate": 0.0019866801419489225, + "loss": 0.127, + "step": 6210 + }, + { + "epoch": 0.05391446254806816, + "grad_norm": 0.083984375, + "learning_rate": 0.0019866750399455345, + "loss": 0.1328, + "step": 6211 + }, + { + "epoch": 0.05392314302827232, + "grad_norm": 0.2578125, + "learning_rate": 0.0019866699369724913, + "loss": 0.1562, + "step": 6212 + }, + { + "epoch": 0.05393182350847649, + "grad_norm": 0.181640625, + "learning_rate": 0.0019866648330297996, + "loss": 0.1787, + "step": 6213 + }, + { + "epoch": 0.05394050398868065, + "grad_norm": 0.400390625, + "learning_rate": 0.001986659728117465, + "loss": 0.1836, + "step": 6214 + }, + { + "epoch": 0.05394918446888482, + "grad_norm": 0.1396484375, + "learning_rate": 0.001986654622235492, + "loss": 0.1533, + "step": 6215 + }, + { + "epoch": 0.05395786494908898, + "grad_norm": 0.451171875, + "learning_rate": 0.0019866495153838867, + "loss": 0.127, + "step": 6216 + }, + { + "epoch": 0.05396654542929315, + "grad_norm": 0.255859375, + "learning_rate": 0.001986644407562655, + "loss": 0.1533, + "step": 6217 + }, + { + "epoch": 0.05397522590949731, + "grad_norm": 0.30859375, + "learning_rate": 0.0019866392987718026, + "loss": 0.165, + "step": 6218 + }, + { + "epoch": 0.05398390638970148, + "grad_norm": 0.201171875, + "learning_rate": 0.0019866341890113345, + "loss": 0.1123, + "step": 6219 + }, + { + "epoch": 0.05399258686990564, + "grad_norm": 0.64453125, + "learning_rate": 0.001986629078281257, + "loss": 0.1641, + "step": 6220 + }, + { + "epoch": 0.05400126735010981, + "grad_norm": 0.3828125, + "learning_rate": 0.0019866239665815746, + "loss": 0.1719, + "step": 6221 + }, + { + "epoch": 0.05400994783031397, + "grad_norm": 0.2431640625, + "learning_rate": 0.001986618853912294, + "loss": 0.1367, + "step": 6222 + }, + { + "epoch": 0.05401862831051814, + "grad_norm": 0.48828125, + "learning_rate": 0.00198661374027342, + "loss": 0.1108, + "step": 6223 + }, + { + "epoch": 0.0540273087907223, + "grad_norm": 0.1201171875, + "learning_rate": 0.001986608625664959, + "loss": 0.1289, + "step": 6224 + }, + { + "epoch": 0.05403598927092647, + "grad_norm": 0.14453125, + "learning_rate": 0.0019866035100869163, + "loss": 0.1279, + "step": 6225 + }, + { + "epoch": 0.05404466975113063, + "grad_norm": 0.412109375, + "learning_rate": 0.0019865983935392966, + "loss": 0.208, + "step": 6226 + }, + { + "epoch": 0.0540533502313348, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019865932760221067, + "loss": 0.1289, + "step": 6227 + }, + { + "epoch": 0.054062030711538964, + "grad_norm": 0.275390625, + "learning_rate": 0.001986588157535351, + "loss": 0.1279, + "step": 6228 + }, + { + "epoch": 0.05407071119174313, + "grad_norm": 0.107421875, + "learning_rate": 0.001986583038079037, + "loss": 0.127, + "step": 6229 + }, + { + "epoch": 0.054079391671947294, + "grad_norm": 0.2333984375, + "learning_rate": 0.001986577917653168, + "loss": 0.1299, + "step": 6230 + }, + { + "epoch": 0.05408807215215146, + "grad_norm": 0.333984375, + "learning_rate": 0.0019865727962577514, + "loss": 0.1543, + "step": 6231 + }, + { + "epoch": 0.054096752632355624, + "grad_norm": 0.296875, + "learning_rate": 0.001986567673892792, + "loss": 0.1484, + "step": 6232 + }, + { + "epoch": 0.05410543311255979, + "grad_norm": 0.4921875, + "learning_rate": 0.0019865625505582954, + "loss": 0.1416, + "step": 6233 + }, + { + "epoch": 0.054114113592763954, + "grad_norm": 1.0234375, + "learning_rate": 0.001986557426254267, + "loss": 0.1406, + "step": 6234 + }, + { + "epoch": 0.05412279407296812, + "grad_norm": 0.69921875, + "learning_rate": 0.001986552300980713, + "loss": 0.1787, + "step": 6235 + }, + { + "epoch": 0.054131474553172285, + "grad_norm": 0.1904296875, + "learning_rate": 0.001986547174737639, + "loss": 0.1777, + "step": 6236 + }, + { + "epoch": 0.05414015503337645, + "grad_norm": 0.6328125, + "learning_rate": 0.0019865420475250503, + "loss": 0.1768, + "step": 6237 + }, + { + "epoch": 0.05414883551358061, + "grad_norm": 0.13671875, + "learning_rate": 0.0019865369193429517, + "loss": 0.1602, + "step": 6238 + }, + { + "epoch": 0.05415751599378477, + "grad_norm": 0.20703125, + "learning_rate": 0.0019865317901913504, + "loss": 0.123, + "step": 6239 + }, + { + "epoch": 0.05416619647398894, + "grad_norm": 0.322265625, + "learning_rate": 0.0019865266600702512, + "loss": 0.1602, + "step": 6240 + }, + { + "epoch": 0.0541748769541931, + "grad_norm": 0.333984375, + "learning_rate": 0.0019865215289796598, + "loss": 0.1445, + "step": 6241 + }, + { + "epoch": 0.05418355743439727, + "grad_norm": 0.50390625, + "learning_rate": 0.0019865163969195816, + "loss": 0.127, + "step": 6242 + }, + { + "epoch": 0.054192237914601434, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019865112638900225, + "loss": 0.1445, + "step": 6243 + }, + { + "epoch": 0.0542009183948056, + "grad_norm": 0.162109375, + "learning_rate": 0.001986506129890988, + "loss": 0.1279, + "step": 6244 + }, + { + "epoch": 0.054209598875009764, + "grad_norm": 0.94140625, + "learning_rate": 0.0019865009949224834, + "loss": 0.1094, + "step": 6245 + }, + { + "epoch": 0.05421827935521393, + "grad_norm": 0.11865234375, + "learning_rate": 0.001986495858984515, + "loss": 0.1455, + "step": 6246 + }, + { + "epoch": 0.054226959835418094, + "grad_norm": 0.6171875, + "learning_rate": 0.001986490722077088, + "loss": 0.1465, + "step": 6247 + }, + { + "epoch": 0.05423564031562226, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019864855842002082, + "loss": 0.1387, + "step": 6248 + }, + { + "epoch": 0.054244320795826424, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019864804453538806, + "loss": 0.1436, + "step": 6249 + }, + { + "epoch": 0.05425300127603059, + "grad_norm": 0.11962890625, + "learning_rate": 0.001986475305538112, + "loss": 0.1631, + "step": 6250 + }, + { + "epoch": 0.054261681756234754, + "grad_norm": 0.404296875, + "learning_rate": 0.001986470164752907, + "loss": 0.1279, + "step": 6251 + }, + { + "epoch": 0.05427036223643892, + "grad_norm": 0.451171875, + "learning_rate": 0.001986465022998271, + "loss": 0.1484, + "step": 6252 + }, + { + "epoch": 0.054279042716643085, + "grad_norm": 0.373046875, + "learning_rate": 0.001986459880274211, + "loss": 0.1865, + "step": 6253 + }, + { + "epoch": 0.05428772319684725, + "grad_norm": 0.31640625, + "learning_rate": 0.001986454736580731, + "loss": 0.1543, + "step": 6254 + }, + { + "epoch": 0.054296403677051415, + "grad_norm": 0.11376953125, + "learning_rate": 0.001986449591917838, + "loss": 0.1611, + "step": 6255 + }, + { + "epoch": 0.05430508415725558, + "grad_norm": 0.095703125, + "learning_rate": 0.0019864444462855367, + "loss": 0.1387, + "step": 6256 + }, + { + "epoch": 0.054313764637459745, + "grad_norm": 0.427734375, + "learning_rate": 0.0019864392996838333, + "loss": 0.1494, + "step": 6257 + }, + { + "epoch": 0.05432244511766391, + "grad_norm": 0.396484375, + "learning_rate": 0.0019864341521127335, + "loss": 0.1572, + "step": 6258 + }, + { + "epoch": 0.054331125597868075, + "grad_norm": 0.58984375, + "learning_rate": 0.001986429003572242, + "loss": 0.1768, + "step": 6259 + }, + { + "epoch": 0.05433980607807224, + "grad_norm": 0.140625, + "learning_rate": 0.0019864238540623653, + "loss": 0.1797, + "step": 6260 + }, + { + "epoch": 0.054348486558276406, + "grad_norm": 0.11279296875, + "learning_rate": 0.001986418703583109, + "loss": 0.1172, + "step": 6261 + }, + { + "epoch": 0.05435716703848057, + "grad_norm": 0.55859375, + "learning_rate": 0.0019864135521344786, + "loss": 0.1074, + "step": 6262 + }, + { + "epoch": 0.054365847518684736, + "grad_norm": 0.298828125, + "learning_rate": 0.0019864083997164794, + "loss": 0.209, + "step": 6263 + }, + { + "epoch": 0.0543745279988889, + "grad_norm": 0.294921875, + "learning_rate": 0.001986403246329117, + "loss": 0.1455, + "step": 6264 + }, + { + "epoch": 0.054383208479093066, + "grad_norm": 0.10205078125, + "learning_rate": 0.001986398091972398, + "loss": 0.1924, + "step": 6265 + }, + { + "epoch": 0.05439188895929723, + "grad_norm": 0.1240234375, + "learning_rate": 0.001986392936646327, + "loss": 0.0977, + "step": 6266 + }, + { + "epoch": 0.054400569439501396, + "grad_norm": 0.09765625, + "learning_rate": 0.00198638778035091, + "loss": 0.1719, + "step": 6267 + }, + { + "epoch": 0.05440924991970556, + "grad_norm": 1.1953125, + "learning_rate": 0.0019863826230861526, + "loss": 0.1562, + "step": 6268 + }, + { + "epoch": 0.05441793039990972, + "grad_norm": 0.056884765625, + "learning_rate": 0.00198637746485206, + "loss": 0.123, + "step": 6269 + }, + { + "epoch": 0.054426610880113885, + "grad_norm": 0.41796875, + "learning_rate": 0.0019863723056486398, + "loss": 0.1436, + "step": 6270 + }, + { + "epoch": 0.05443529136031805, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019863671454758945, + "loss": 0.1582, + "step": 6271 + }, + { + "epoch": 0.054443971840522215, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019863619843338326, + "loss": 0.1108, + "step": 6272 + }, + { + "epoch": 0.05445265232072638, + "grad_norm": 0.2109375, + "learning_rate": 0.001986356822222458, + "loss": 0.104, + "step": 6273 + }, + { + "epoch": 0.054461332800930545, + "grad_norm": 0.3671875, + "learning_rate": 0.001986351659141777, + "loss": 0.1533, + "step": 6274 + }, + { + "epoch": 0.05447001328113471, + "grad_norm": 0.1298828125, + "learning_rate": 0.001986346495091795, + "loss": 0.0986, + "step": 6275 + }, + { + "epoch": 0.054478693761338876, + "grad_norm": 0.2373046875, + "learning_rate": 0.001986341330072518, + "loss": 0.1089, + "step": 6276 + }, + { + "epoch": 0.05448737424154304, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019863361640839513, + "loss": 0.1426, + "step": 6277 + }, + { + "epoch": 0.054496054721747206, + "grad_norm": 0.2578125, + "learning_rate": 0.001986330997126101, + "loss": 0.1025, + "step": 6278 + }, + { + "epoch": 0.05450473520195137, + "grad_norm": 0.1337890625, + "learning_rate": 0.001986325829198972, + "loss": 0.1445, + "step": 6279 + }, + { + "epoch": 0.054513415682155536, + "grad_norm": 0.1015625, + "learning_rate": 0.0019863206603025706, + "loss": 0.1367, + "step": 6280 + }, + { + "epoch": 0.0545220961623597, + "grad_norm": 0.45703125, + "learning_rate": 0.0019863154904369022, + "loss": 0.1787, + "step": 6281 + }, + { + "epoch": 0.054530776642563866, + "grad_norm": 0.26953125, + "learning_rate": 0.0019863103196019723, + "loss": 0.1406, + "step": 6282 + }, + { + "epoch": 0.05453945712276803, + "grad_norm": 0.33984375, + "learning_rate": 0.0019863051477977873, + "loss": 0.126, + "step": 6283 + }, + { + "epoch": 0.054548137602972196, + "grad_norm": 0.466796875, + "learning_rate": 0.0019862999750243516, + "loss": 0.124, + "step": 6284 + }, + { + "epoch": 0.05455681808317636, + "grad_norm": 0.5625, + "learning_rate": 0.001986294801281672, + "loss": 0.1484, + "step": 6285 + }, + { + "epoch": 0.05456549856338053, + "grad_norm": 0.359375, + "learning_rate": 0.001986289626569754, + "loss": 0.1279, + "step": 6286 + }, + { + "epoch": 0.05457417904358469, + "grad_norm": 0.52734375, + "learning_rate": 0.0019862844508886025, + "loss": 0.1592, + "step": 6287 + }, + { + "epoch": 0.05458285952378886, + "grad_norm": 0.1875, + "learning_rate": 0.001986279274238224, + "loss": 0.1143, + "step": 6288 + }, + { + "epoch": 0.05459154000399302, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019862740966186234, + "loss": 0.124, + "step": 6289 + }, + { + "epoch": 0.05460022048419719, + "grad_norm": 0.1572265625, + "learning_rate": 0.001986268918029807, + "loss": 0.1338, + "step": 6290 + }, + { + "epoch": 0.05460890096440135, + "grad_norm": 0.2578125, + "learning_rate": 0.00198626373847178, + "loss": 0.1221, + "step": 6291 + }, + { + "epoch": 0.05461758144460552, + "grad_norm": 0.3359375, + "learning_rate": 0.0019862585579445485, + "loss": 0.1592, + "step": 6292 + }, + { + "epoch": 0.05462626192480968, + "grad_norm": 0.1796875, + "learning_rate": 0.001986253376448118, + "loss": 0.1118, + "step": 6293 + }, + { + "epoch": 0.05463494240501385, + "grad_norm": 0.365234375, + "learning_rate": 0.001986248193982494, + "loss": 0.1523, + "step": 6294 + }, + { + "epoch": 0.05464362288521801, + "grad_norm": 0.1220703125, + "learning_rate": 0.001986243010547682, + "loss": 0.1748, + "step": 6295 + }, + { + "epoch": 0.05465230336542218, + "grad_norm": 0.53125, + "learning_rate": 0.0019862378261436883, + "loss": 0.1318, + "step": 6296 + }, + { + "epoch": 0.05466098384562634, + "grad_norm": 0.51953125, + "learning_rate": 0.0019862326407705183, + "loss": 0.1289, + "step": 6297 + }, + { + "epoch": 0.05466966432583051, + "grad_norm": 0.59765625, + "learning_rate": 0.0019862274544281775, + "loss": 0.1089, + "step": 6298 + }, + { + "epoch": 0.05467834480603467, + "grad_norm": 0.1650390625, + "learning_rate": 0.001986222267116672, + "loss": 0.1582, + "step": 6299 + }, + { + "epoch": 0.05468702528623883, + "grad_norm": 0.08251953125, + "learning_rate": 0.001986217078836007, + "loss": 0.1738, + "step": 6300 + }, + { + "epoch": 0.054695705766443, + "grad_norm": 0.19921875, + "learning_rate": 0.001986211889586188, + "loss": 0.1465, + "step": 6301 + }, + { + "epoch": 0.05470438624664716, + "grad_norm": 0.08154296875, + "learning_rate": 0.001986206699367221, + "loss": 0.1357, + "step": 6302 + }, + { + "epoch": 0.05471306672685133, + "grad_norm": 0.50390625, + "learning_rate": 0.001986201508179112, + "loss": 0.1016, + "step": 6303 + }, + { + "epoch": 0.05472174720705549, + "grad_norm": 0.130859375, + "learning_rate": 0.0019861963160218658, + "loss": 0.126, + "step": 6304 + }, + { + "epoch": 0.05473042768725966, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019861911228954894, + "loss": 0.1221, + "step": 6305 + }, + { + "epoch": 0.05473910816746382, + "grad_norm": 0.1455078125, + "learning_rate": 0.001986185928799987, + "loss": 0.1602, + "step": 6306 + }, + { + "epoch": 0.05474778864766799, + "grad_norm": 0.62109375, + "learning_rate": 0.0019861807337353656, + "loss": 0.1396, + "step": 6307 + }, + { + "epoch": 0.05475646912787215, + "grad_norm": 0.349609375, + "learning_rate": 0.00198617553770163, + "loss": 0.166, + "step": 6308 + }, + { + "epoch": 0.05476514960807632, + "grad_norm": 0.3515625, + "learning_rate": 0.001986170340698786, + "loss": 0.1768, + "step": 6309 + }, + { + "epoch": 0.05477383008828048, + "grad_norm": 0.306640625, + "learning_rate": 0.0019861651427268393, + "loss": 0.1855, + "step": 6310 + }, + { + "epoch": 0.05478251056848465, + "grad_norm": 0.486328125, + "learning_rate": 0.001986159943785796, + "loss": 0.1162, + "step": 6311 + }, + { + "epoch": 0.05479119104868881, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019861547438756615, + "loss": 0.127, + "step": 6312 + }, + { + "epoch": 0.05479987152889298, + "grad_norm": 0.2001953125, + "learning_rate": 0.001986149542996441, + "loss": 0.1885, + "step": 6313 + }, + { + "epoch": 0.05480855200909714, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019861443411481414, + "loss": 0.1787, + "step": 6314 + }, + { + "epoch": 0.05481723248930131, + "grad_norm": 0.103515625, + "learning_rate": 0.001986139138330767, + "loss": 0.1318, + "step": 6315 + }, + { + "epoch": 0.05482591296950547, + "grad_norm": 0.234375, + "learning_rate": 0.001986133934544325, + "loss": 0.166, + "step": 6316 + }, + { + "epoch": 0.05483459344970964, + "grad_norm": 0.357421875, + "learning_rate": 0.00198612872978882, + "loss": 0.1357, + "step": 6317 + }, + { + "epoch": 0.054843273929913804, + "grad_norm": 0.0888671875, + "learning_rate": 0.001986123524064257, + "loss": 0.1299, + "step": 6318 + }, + { + "epoch": 0.05485195441011797, + "grad_norm": 0.19140625, + "learning_rate": 0.0019861183173706435, + "loss": 0.1226, + "step": 6319 + }, + { + "epoch": 0.054860634890322134, + "grad_norm": 0.318359375, + "learning_rate": 0.0019861131097079843, + "loss": 0.1621, + "step": 6320 + }, + { + "epoch": 0.0548693153705263, + "grad_norm": 0.2490234375, + "learning_rate": 0.001986107901076285, + "loss": 0.1475, + "step": 6321 + }, + { + "epoch": 0.054877995850730464, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019861026914755513, + "loss": 0.1816, + "step": 6322 + }, + { + "epoch": 0.05488667633093463, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019860974809057888, + "loss": 0.1465, + "step": 6323 + }, + { + "epoch": 0.054895356811138794, + "grad_norm": 0.193359375, + "learning_rate": 0.001986092269367004, + "loss": 0.1475, + "step": 6324 + }, + { + "epoch": 0.05490403729134296, + "grad_norm": 0.1875, + "learning_rate": 0.0019860870568592015, + "loss": 0.1533, + "step": 6325 + }, + { + "epoch": 0.054912717771547125, + "grad_norm": 0.150390625, + "learning_rate": 0.001986081843382388, + "loss": 0.1533, + "step": 6326 + }, + { + "epoch": 0.05492139825175129, + "grad_norm": 0.1962890625, + "learning_rate": 0.001986076628936568, + "loss": 0.1338, + "step": 6327 + }, + { + "epoch": 0.054930078731955455, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019860714135217484, + "loss": 0.1396, + "step": 6328 + }, + { + "epoch": 0.05493875921215962, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019860661971379343, + "loss": 0.1147, + "step": 6329 + }, + { + "epoch": 0.05494743969236378, + "grad_norm": 0.189453125, + "learning_rate": 0.0019860609797851315, + "loss": 0.126, + "step": 6330 + }, + { + "epoch": 0.05495612017256794, + "grad_norm": 0.080078125, + "learning_rate": 0.001986055761463346, + "loss": 0.1436, + "step": 6331 + }, + { + "epoch": 0.05496480065277211, + "grad_norm": 0.458984375, + "learning_rate": 0.001986050542172583, + "loss": 0.1309, + "step": 6332 + }, + { + "epoch": 0.054973481132976273, + "grad_norm": 0.271484375, + "learning_rate": 0.0019860453219128485, + "loss": 0.1494, + "step": 6333 + }, + { + "epoch": 0.05498216161318044, + "grad_norm": 0.384765625, + "learning_rate": 0.0019860401006841478, + "loss": 0.1797, + "step": 6334 + }, + { + "epoch": 0.054990842093384604, + "grad_norm": 0.2421875, + "learning_rate": 0.001986034878486488, + "loss": 0.1416, + "step": 6335 + }, + { + "epoch": 0.05499952257358877, + "grad_norm": 0.1328125, + "learning_rate": 0.0019860296553198724, + "loss": 0.1157, + "step": 6336 + }, + { + "epoch": 0.055008203053792934, + "grad_norm": 0.265625, + "learning_rate": 0.0019860244311843086, + "loss": 0.1289, + "step": 6337 + }, + { + "epoch": 0.0550168835339971, + "grad_norm": 0.2099609375, + "learning_rate": 0.001986019206079802, + "loss": 0.1406, + "step": 6338 + }, + { + "epoch": 0.055025564014201264, + "grad_norm": 0.1044921875, + "learning_rate": 0.001986013980006358, + "loss": 0.126, + "step": 6339 + }, + { + "epoch": 0.05503424449440543, + "grad_norm": 0.62109375, + "learning_rate": 0.001986008752963983, + "loss": 0.124, + "step": 6340 + }, + { + "epoch": 0.055042924974609594, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019860035249526812, + "loss": 0.1167, + "step": 6341 + }, + { + "epoch": 0.05505160545481376, + "grad_norm": 0.58203125, + "learning_rate": 0.0019859982959724598, + "loss": 0.1025, + "step": 6342 + }, + { + "epoch": 0.055060285935017925, + "grad_norm": 0.220703125, + "learning_rate": 0.0019859930660233237, + "loss": 0.127, + "step": 6343 + }, + { + "epoch": 0.05506896641522209, + "grad_norm": 0.70703125, + "learning_rate": 0.001985987835105279, + "loss": 0.1172, + "step": 6344 + }, + { + "epoch": 0.055077646895426255, + "grad_norm": 0.1591796875, + "learning_rate": 0.001985982603218331, + "loss": 0.1289, + "step": 6345 + }, + { + "epoch": 0.05508632737563042, + "grad_norm": 0.388671875, + "learning_rate": 0.0019859773703624862, + "loss": 0.1377, + "step": 6346 + }, + { + "epoch": 0.055095007855834585, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019859721365377498, + "loss": 0.1572, + "step": 6347 + }, + { + "epoch": 0.05510368833603875, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019859669017441273, + "loss": 0.126, + "step": 6348 + }, + { + "epoch": 0.055112368816242915, + "grad_norm": 0.173828125, + "learning_rate": 0.001985961665981625, + "loss": 0.1758, + "step": 6349 + }, + { + "epoch": 0.05512104929644708, + "grad_norm": 0.279296875, + "learning_rate": 0.0019859564292502487, + "loss": 0.1514, + "step": 6350 + }, + { + "epoch": 0.055129729776651246, + "grad_norm": 0.5, + "learning_rate": 0.001985951191550003, + "loss": 0.1367, + "step": 6351 + }, + { + "epoch": 0.05513841025685541, + "grad_norm": 0.345703125, + "learning_rate": 0.0019859459528808947, + "loss": 0.1426, + "step": 6352 + }, + { + "epoch": 0.055147090737059576, + "grad_norm": 0.41015625, + "learning_rate": 0.0019859407132429294, + "loss": 0.1396, + "step": 6353 + }, + { + "epoch": 0.05515577121726374, + "grad_norm": 0.52734375, + "learning_rate": 0.0019859354726361124, + "loss": 0.1885, + "step": 6354 + }, + { + "epoch": 0.055164451697467906, + "grad_norm": 0.33203125, + "learning_rate": 0.0019859302310604493, + "loss": 0.1348, + "step": 6355 + }, + { + "epoch": 0.05517313217767207, + "grad_norm": 0.5859375, + "learning_rate": 0.001985924988515947, + "loss": 0.2031, + "step": 6356 + }, + { + "epoch": 0.055181812657876236, + "grad_norm": 0.333984375, + "learning_rate": 0.00198591974500261, + "loss": 0.1016, + "step": 6357 + }, + { + "epoch": 0.0551904931380804, + "grad_norm": 0.208984375, + "learning_rate": 0.0019859145005204446, + "loss": 0.1221, + "step": 6358 + }, + { + "epoch": 0.05519917361828457, + "grad_norm": 0.1640625, + "learning_rate": 0.0019859092550694566, + "loss": 0.1021, + "step": 6359 + }, + { + "epoch": 0.05520785409848873, + "grad_norm": 0.8203125, + "learning_rate": 0.001985904008649651, + "loss": 0.1328, + "step": 6360 + }, + { + "epoch": 0.05521653457869289, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019858987612610345, + "loss": 0.1738, + "step": 6361 + }, + { + "epoch": 0.055225215058897055, + "grad_norm": 0.9296875, + "learning_rate": 0.0019858935129036122, + "loss": 0.1172, + "step": 6362 + }, + { + "epoch": 0.05523389553910122, + "grad_norm": 0.54296875, + "learning_rate": 0.0019858882635773903, + "loss": 0.1387, + "step": 6363 + }, + { + "epoch": 0.055242576019305385, + "grad_norm": 0.16796875, + "learning_rate": 0.0019858830132823743, + "loss": 0.2051, + "step": 6364 + }, + { + "epoch": 0.05525125649950955, + "grad_norm": 0.1806640625, + "learning_rate": 0.00198587776201857, + "loss": 0.1152, + "step": 6365 + }, + { + "epoch": 0.055259936979713715, + "grad_norm": 0.09716796875, + "learning_rate": 0.001985872509785983, + "loss": 0.1602, + "step": 6366 + }, + { + "epoch": 0.05526861745991788, + "grad_norm": 0.39453125, + "learning_rate": 0.001985867256584619, + "loss": 0.1621, + "step": 6367 + }, + { + "epoch": 0.055277297940122046, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019858620024144837, + "loss": 0.1328, + "step": 6368 + }, + { + "epoch": 0.05528597842032621, + "grad_norm": 0.220703125, + "learning_rate": 0.0019858567472755832, + "loss": 0.123, + "step": 6369 + }, + { + "epoch": 0.055294658900530376, + "grad_norm": 0.181640625, + "learning_rate": 0.0019858514911679235, + "loss": 0.168, + "step": 6370 + }, + { + "epoch": 0.05530333938073454, + "grad_norm": 0.15625, + "learning_rate": 0.0019858462340915095, + "loss": 0.1416, + "step": 6371 + }, + { + "epoch": 0.055312019860938706, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019858409760463475, + "loss": 0.1235, + "step": 6372 + }, + { + "epoch": 0.05532070034114287, + "grad_norm": 0.1259765625, + "learning_rate": 0.001985835717032443, + "loss": 0.1221, + "step": 6373 + }, + { + "epoch": 0.055329380821347036, + "grad_norm": 0.322265625, + "learning_rate": 0.001985830457049802, + "loss": 0.124, + "step": 6374 + }, + { + "epoch": 0.0553380613015512, + "grad_norm": 0.474609375, + "learning_rate": 0.0019858251960984297, + "loss": 0.1484, + "step": 6375 + }, + { + "epoch": 0.05534674178175537, + "grad_norm": 0.5078125, + "learning_rate": 0.0019858199341783326, + "loss": 0.1816, + "step": 6376 + }, + { + "epoch": 0.05535542226195953, + "grad_norm": 0.28515625, + "learning_rate": 0.0019858146712895164, + "loss": 0.126, + "step": 6377 + }, + { + "epoch": 0.0553641027421637, + "grad_norm": 0.181640625, + "learning_rate": 0.0019858094074319864, + "loss": 0.1602, + "step": 6378 + }, + { + "epoch": 0.05537278322236786, + "grad_norm": 0.490234375, + "learning_rate": 0.0019858041426057483, + "loss": 0.1387, + "step": 6379 + }, + { + "epoch": 0.05538146370257203, + "grad_norm": 0.166015625, + "learning_rate": 0.001985798876810808, + "loss": 0.1201, + "step": 6380 + }, + { + "epoch": 0.05539014418277619, + "grad_norm": 0.2451171875, + "learning_rate": 0.001985793610047172, + "loss": 0.1367, + "step": 6381 + }, + { + "epoch": 0.05539882466298036, + "grad_norm": 0.22265625, + "learning_rate": 0.001985788342314845, + "loss": 0.1309, + "step": 6382 + }, + { + "epoch": 0.05540750514318452, + "grad_norm": 0.201171875, + "learning_rate": 0.001985783073613833, + "loss": 0.1621, + "step": 6383 + }, + { + "epoch": 0.05541618562338869, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019857778039441424, + "loss": 0.1104, + "step": 6384 + }, + { + "epoch": 0.05542486610359285, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019857725333057782, + "loss": 0.1562, + "step": 6385 + }, + { + "epoch": 0.05543354658379702, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019857672616987467, + "loss": 0.1689, + "step": 6386 + }, + { + "epoch": 0.05544222706400118, + "grad_norm": 0.255859375, + "learning_rate": 0.0019857619891230534, + "loss": 0.1729, + "step": 6387 + }, + { + "epoch": 0.05545090754420535, + "grad_norm": 0.453125, + "learning_rate": 0.001985756715578704, + "loss": 0.1641, + "step": 6388 + }, + { + "epoch": 0.05545958802440951, + "grad_norm": 0.171875, + "learning_rate": 0.0019857514410657044, + "loss": 0.1543, + "step": 6389 + }, + { + "epoch": 0.05546826850461368, + "grad_norm": 0.25, + "learning_rate": 0.00198574616558406, + "loss": 0.1338, + "step": 6390 + }, + { + "epoch": 0.055476948984817843, + "grad_norm": 0.68359375, + "learning_rate": 0.001985740889133777, + "loss": 0.1611, + "step": 6391 + }, + { + "epoch": 0.055485629465022, + "grad_norm": 0.5390625, + "learning_rate": 0.0019857356117148612, + "loss": 0.1416, + "step": 6392 + }, + { + "epoch": 0.05549430994522617, + "grad_norm": 0.302734375, + "learning_rate": 0.0019857303333273187, + "loss": 0.1387, + "step": 6393 + }, + { + "epoch": 0.05550299042543033, + "grad_norm": 0.61328125, + "learning_rate": 0.001985725053971154, + "loss": 0.1484, + "step": 6394 + }, + { + "epoch": 0.0555116709056345, + "grad_norm": 0.703125, + "learning_rate": 0.0019857197736463743, + "loss": 0.1484, + "step": 6395 + }, + { + "epoch": 0.05552035138583866, + "grad_norm": 0.310546875, + "learning_rate": 0.0019857144923529843, + "loss": 0.1406, + "step": 6396 + }, + { + "epoch": 0.05552903186604283, + "grad_norm": 0.412109375, + "learning_rate": 0.0019857092100909906, + "loss": 0.1191, + "step": 6397 + }, + { + "epoch": 0.05553771234624699, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019857039268603984, + "loss": 0.1504, + "step": 6398 + }, + { + "epoch": 0.05554639282645116, + "grad_norm": 0.2890625, + "learning_rate": 0.001985698642661214, + "loss": 0.1738, + "step": 6399 + }, + { + "epoch": 0.05555507330665532, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019856933574934426, + "loss": 0.1357, + "step": 6400 + }, + { + "epoch": 0.05556375378685949, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019856880713570906, + "loss": 0.1631, + "step": 6401 + }, + { + "epoch": 0.05557243426706365, + "grad_norm": 0.13671875, + "learning_rate": 0.001985682784252163, + "loss": 0.1543, + "step": 6402 + }, + { + "epoch": 0.05558111474726782, + "grad_norm": 0.482421875, + "learning_rate": 0.0019856774961786662, + "loss": 0.1582, + "step": 6403 + }, + { + "epoch": 0.05558979522747198, + "grad_norm": 0.83984375, + "learning_rate": 0.0019856722071366056, + "loss": 0.1504, + "step": 6404 + }, + { + "epoch": 0.05559847570767615, + "grad_norm": 0.17578125, + "learning_rate": 0.001985666917125987, + "loss": 0.1309, + "step": 6405 + }, + { + "epoch": 0.05560715618788031, + "grad_norm": 0.205078125, + "learning_rate": 0.001985661626146817, + "loss": 0.1113, + "step": 6406 + }, + { + "epoch": 0.05561583666808448, + "grad_norm": 0.359375, + "learning_rate": 0.0019856563341991004, + "loss": 0.1289, + "step": 6407 + }, + { + "epoch": 0.055624517148288644, + "grad_norm": 0.7109375, + "learning_rate": 0.0019856510412828436, + "loss": 0.1797, + "step": 6408 + }, + { + "epoch": 0.05563319762849281, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019856457473980517, + "loss": 0.1621, + "step": 6409 + }, + { + "epoch": 0.055641878108696974, + "grad_norm": 0.33203125, + "learning_rate": 0.001985640452544731, + "loss": 0.1504, + "step": 6410 + }, + { + "epoch": 0.05565055858890114, + "grad_norm": 0.107421875, + "learning_rate": 0.0019856351567228876, + "loss": 0.1367, + "step": 6411 + }, + { + "epoch": 0.055659239069105304, + "grad_norm": 0.25, + "learning_rate": 0.0019856298599325267, + "loss": 0.1143, + "step": 6412 + }, + { + "epoch": 0.05566791954930947, + "grad_norm": 0.4375, + "learning_rate": 0.001985624562173654, + "loss": 0.127, + "step": 6413 + }, + { + "epoch": 0.055676600029513634, + "grad_norm": 1.0703125, + "learning_rate": 0.001985619263446276, + "loss": 0.3711, + "step": 6414 + }, + { + "epoch": 0.0556852805097178, + "grad_norm": 0.40234375, + "learning_rate": 0.001985613963750398, + "loss": 0.1289, + "step": 6415 + }, + { + "epoch": 0.055693960989921965, + "grad_norm": 0.185546875, + "learning_rate": 0.0019856086630860256, + "loss": 0.126, + "step": 6416 + }, + { + "epoch": 0.05570264147012613, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019856033614531654, + "loss": 0.1211, + "step": 6417 + }, + { + "epoch": 0.055711321950330295, + "grad_norm": 0.1357421875, + "learning_rate": 0.001985598058851822, + "loss": 0.1621, + "step": 6418 + }, + { + "epoch": 0.05572000243053446, + "grad_norm": 0.28125, + "learning_rate": 0.0019855927552820027, + "loss": 0.1738, + "step": 6419 + }, + { + "epoch": 0.055728682910738625, + "grad_norm": 0.291015625, + "learning_rate": 0.001985587450743712, + "loss": 0.1885, + "step": 6420 + }, + { + "epoch": 0.05573736339094279, + "grad_norm": 0.380859375, + "learning_rate": 0.0019855821452369558, + "loss": 0.1338, + "step": 6421 + }, + { + "epoch": 0.055746043871146955, + "grad_norm": 0.419921875, + "learning_rate": 0.001985576838761741, + "loss": 0.1816, + "step": 6422 + }, + { + "epoch": 0.05575472435135111, + "grad_norm": 0.201171875, + "learning_rate": 0.0019855715313180718, + "loss": 0.1709, + "step": 6423 + }, + { + "epoch": 0.05576340483155528, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019855662229059557, + "loss": 0.1289, + "step": 6424 + }, + { + "epoch": 0.055772085311759444, + "grad_norm": 0.08837890625, + "learning_rate": 0.001985560913525397, + "loss": 0.1387, + "step": 6425 + }, + { + "epoch": 0.05578076579196361, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019855556031764024, + "loss": 0.2002, + "step": 6426 + }, + { + "epoch": 0.055789446272167774, + "grad_norm": 0.0849609375, + "learning_rate": 0.001985550291858978, + "loss": 0.125, + "step": 6427 + }, + { + "epoch": 0.05579812675237194, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019855449795731285, + "loss": 0.1719, + "step": 6428 + }, + { + "epoch": 0.055806807232576104, + "grad_norm": 0.2275390625, + "learning_rate": 0.001985539666318861, + "loss": 0.1777, + "step": 6429 + }, + { + "epoch": 0.05581548771278027, + "grad_norm": 0.51953125, + "learning_rate": 0.0019855343520961796, + "loss": 0.1367, + "step": 6430 + }, + { + "epoch": 0.055824168192984434, + "grad_norm": 0.21484375, + "learning_rate": 0.001985529036905092, + "loss": 0.2051, + "step": 6431 + }, + { + "epoch": 0.0558328486731886, + "grad_norm": 0.2314453125, + "learning_rate": 0.001985523720745603, + "loss": 0.1338, + "step": 6432 + }, + { + "epoch": 0.055841529153392765, + "grad_norm": 0.11767578125, + "learning_rate": 0.001985518403617718, + "loss": 0.1797, + "step": 6433 + }, + { + "epoch": 0.05585020963359693, + "grad_norm": 0.34375, + "learning_rate": 0.0019855130855214445, + "loss": 0.1396, + "step": 6434 + }, + { + "epoch": 0.055858890113801095, + "grad_norm": 0.1962890625, + "learning_rate": 0.001985507766456786, + "loss": 0.1348, + "step": 6435 + }, + { + "epoch": 0.05586757059400526, + "grad_norm": 1.1875, + "learning_rate": 0.00198550244642375, + "loss": 0.1279, + "step": 6436 + }, + { + "epoch": 0.055876251074209425, + "grad_norm": 0.08447265625, + "learning_rate": 0.001985497125422342, + "loss": 0.1289, + "step": 6437 + }, + { + "epoch": 0.05588493155441359, + "grad_norm": 0.59375, + "learning_rate": 0.0019854918034525673, + "loss": 0.1465, + "step": 6438 + }, + { + "epoch": 0.055893612034617755, + "grad_norm": 0.2734375, + "learning_rate": 0.0019854864805144323, + "loss": 0.1289, + "step": 6439 + }, + { + "epoch": 0.05590229251482192, + "grad_norm": 0.275390625, + "learning_rate": 0.0019854811566079426, + "loss": 0.1582, + "step": 6440 + }, + { + "epoch": 0.055910972995026086, + "grad_norm": 0.291015625, + "learning_rate": 0.0019854758317331043, + "loss": 0.1465, + "step": 6441 + }, + { + "epoch": 0.05591965347523025, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019854705058899224, + "loss": 0.1475, + "step": 6442 + }, + { + "epoch": 0.055928333955434416, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019854651790784036, + "loss": 0.1523, + "step": 6443 + }, + { + "epoch": 0.05593701443563858, + "grad_norm": 0.34765625, + "learning_rate": 0.001985459851298553, + "loss": 0.1426, + "step": 6444 + }, + { + "epoch": 0.055945694915842746, + "grad_norm": 0.1953125, + "learning_rate": 0.0019854545225503773, + "loss": 0.1279, + "step": 6445 + }, + { + "epoch": 0.05595437539604691, + "grad_norm": 0.1875, + "learning_rate": 0.0019854491928338815, + "loss": 0.1162, + "step": 6446 + }, + { + "epoch": 0.055963055876251076, + "grad_norm": 0.51953125, + "learning_rate": 0.001985443862149072, + "loss": 0.1709, + "step": 6447 + }, + { + "epoch": 0.05597173635645524, + "grad_norm": 0.1689453125, + "learning_rate": 0.001985438530495954, + "loss": 0.1279, + "step": 6448 + }, + { + "epoch": 0.05598041683665941, + "grad_norm": 0.5703125, + "learning_rate": 0.001985433197874534, + "loss": 0.1787, + "step": 6449 + }, + { + "epoch": 0.05598909731686357, + "grad_norm": 0.083984375, + "learning_rate": 0.0019854278642848174, + "loss": 0.1245, + "step": 6450 + }, + { + "epoch": 0.05599777779706774, + "grad_norm": 0.41796875, + "learning_rate": 0.0019854225297268103, + "loss": 0.1182, + "step": 6451 + }, + { + "epoch": 0.0560064582772719, + "grad_norm": 0.29296875, + "learning_rate": 0.0019854171942005187, + "loss": 0.1816, + "step": 6452 + }, + { + "epoch": 0.05601513875747607, + "grad_norm": 0.3046875, + "learning_rate": 0.001985411857705948, + "loss": 0.1621, + "step": 6453 + }, + { + "epoch": 0.056023819237680225, + "grad_norm": 0.10498046875, + "learning_rate": 0.001985406520243104, + "loss": 0.1787, + "step": 6454 + }, + { + "epoch": 0.05603249971788439, + "grad_norm": 1.03125, + "learning_rate": 0.001985401181811993, + "loss": 0.1348, + "step": 6455 + }, + { + "epoch": 0.056041180198088555, + "grad_norm": 0.310546875, + "learning_rate": 0.0019853958424126206, + "loss": 0.1201, + "step": 6456 + }, + { + "epoch": 0.05604986067829272, + "grad_norm": 0.87890625, + "learning_rate": 0.001985390502044992, + "loss": 0.1562, + "step": 6457 + }, + { + "epoch": 0.056058541158496886, + "grad_norm": 0.1875, + "learning_rate": 0.001985385160709114, + "loss": 0.1455, + "step": 6458 + }, + { + "epoch": 0.05606722163870105, + "grad_norm": 0.5390625, + "learning_rate": 0.0019853798184049924, + "loss": 0.1094, + "step": 6459 + }, + { + "epoch": 0.056075902118905216, + "grad_norm": 0.189453125, + "learning_rate": 0.0019853744751326323, + "loss": 0.1465, + "step": 6460 + }, + { + "epoch": 0.05608458259910938, + "grad_norm": 0.1162109375, + "learning_rate": 0.00198536913089204, + "loss": 0.1191, + "step": 6461 + }, + { + "epoch": 0.056093263079313546, + "grad_norm": 0.314453125, + "learning_rate": 0.001985363785683222, + "loss": 0.1553, + "step": 6462 + }, + { + "epoch": 0.05610194355951771, + "grad_norm": 0.3828125, + "learning_rate": 0.0019853584395061828, + "loss": 0.165, + "step": 6463 + }, + { + "epoch": 0.056110624039721876, + "grad_norm": 0.1650390625, + "learning_rate": 0.001985353092360929, + "loss": 0.1758, + "step": 6464 + }, + { + "epoch": 0.05611930451992604, + "grad_norm": 0.37109375, + "learning_rate": 0.0019853477442474663, + "loss": 0.1748, + "step": 6465 + }, + { + "epoch": 0.05612798500013021, + "grad_norm": 0.11669921875, + "learning_rate": 0.001985342395165801, + "loss": 0.1523, + "step": 6466 + }, + { + "epoch": 0.05613666548033437, + "grad_norm": 0.267578125, + "learning_rate": 0.001985337045115938, + "loss": 0.1777, + "step": 6467 + }, + { + "epoch": 0.05614534596053854, + "grad_norm": 0.474609375, + "learning_rate": 0.001985331694097884, + "loss": 0.1167, + "step": 6468 + }, + { + "epoch": 0.0561540264407427, + "grad_norm": 0.08447265625, + "learning_rate": 0.001985326342111645, + "loss": 0.1152, + "step": 6469 + }, + { + "epoch": 0.05616270692094687, + "grad_norm": 0.50390625, + "learning_rate": 0.001985320989157226, + "loss": 0.1377, + "step": 6470 + }, + { + "epoch": 0.05617138740115103, + "grad_norm": 0.515625, + "learning_rate": 0.001985315635234633, + "loss": 0.106, + "step": 6471 + }, + { + "epoch": 0.0561800678813552, + "grad_norm": 0.13671875, + "learning_rate": 0.0019853102803438724, + "loss": 0.1582, + "step": 6472 + }, + { + "epoch": 0.05618874836155936, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019853049244849498, + "loss": 0.1357, + "step": 6473 + }, + { + "epoch": 0.05619742884176353, + "grad_norm": 0.236328125, + "learning_rate": 0.001985299567657871, + "loss": 0.1445, + "step": 6474 + }, + { + "epoch": 0.05620610932196769, + "grad_norm": 0.181640625, + "learning_rate": 0.001985294209862642, + "loss": 0.1006, + "step": 6475 + }, + { + "epoch": 0.05621478980217186, + "grad_norm": 0.14453125, + "learning_rate": 0.0019852888510992683, + "loss": 0.1475, + "step": 6476 + }, + { + "epoch": 0.05622347028237602, + "grad_norm": 0.4921875, + "learning_rate": 0.001985283491367756, + "loss": 0.166, + "step": 6477 + }, + { + "epoch": 0.05623215076258019, + "grad_norm": 0.56640625, + "learning_rate": 0.0019852781306681114, + "loss": 0.1055, + "step": 6478 + }, + { + "epoch": 0.05624083124278435, + "grad_norm": 1.015625, + "learning_rate": 0.0019852727690003397, + "loss": 0.1641, + "step": 6479 + }, + { + "epoch": 0.05624951172298852, + "grad_norm": 0.28515625, + "learning_rate": 0.0019852674063644466, + "loss": 0.1406, + "step": 6480 + }, + { + "epoch": 0.05625819220319268, + "grad_norm": 0.201171875, + "learning_rate": 0.001985262042760439, + "loss": 0.127, + "step": 6481 + }, + { + "epoch": 0.05626687268339685, + "grad_norm": 0.27734375, + "learning_rate": 0.001985256678188322, + "loss": 0.1167, + "step": 6482 + }, + { + "epoch": 0.056275553163601014, + "grad_norm": 0.16796875, + "learning_rate": 0.0019852513126481014, + "loss": 0.1367, + "step": 6483 + }, + { + "epoch": 0.05628423364380517, + "grad_norm": 0.58984375, + "learning_rate": 0.0019852459461397833, + "loss": 0.1357, + "step": 6484 + }, + { + "epoch": 0.05629291412400934, + "grad_norm": 0.419921875, + "learning_rate": 0.0019852405786633734, + "loss": 0.1797, + "step": 6485 + }, + { + "epoch": 0.0563015946042135, + "grad_norm": 0.353515625, + "learning_rate": 0.001985235210218878, + "loss": 0.1377, + "step": 6486 + }, + { + "epoch": 0.05631027508441767, + "grad_norm": 0.13671875, + "learning_rate": 0.0019852298408063024, + "loss": 0.1338, + "step": 6487 + }, + { + "epoch": 0.05631895556462183, + "grad_norm": 0.3046875, + "learning_rate": 0.001985224470425653, + "loss": 0.1582, + "step": 6488 + }, + { + "epoch": 0.056327636044826, + "grad_norm": 0.279296875, + "learning_rate": 0.0019852190990769353, + "loss": 0.1309, + "step": 6489 + }, + { + "epoch": 0.05633631652503016, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019852137267601555, + "loss": 0.0898, + "step": 6490 + }, + { + "epoch": 0.05634499700523433, + "grad_norm": 0.06103515625, + "learning_rate": 0.001985208353475319, + "loss": 0.1055, + "step": 6491 + }, + { + "epoch": 0.05635367748543849, + "grad_norm": 0.41015625, + "learning_rate": 0.001985202979222432, + "loss": 0.1162, + "step": 6492 + }, + { + "epoch": 0.05636235796564266, + "grad_norm": 1.0859375, + "learning_rate": 0.0019851976040015005, + "loss": 0.1777, + "step": 6493 + }, + { + "epoch": 0.05637103844584682, + "grad_norm": 0.1025390625, + "learning_rate": 0.00198519222781253, + "loss": 0.1406, + "step": 6494 + }, + { + "epoch": 0.05637971892605099, + "grad_norm": 0.263671875, + "learning_rate": 0.001985186850655527, + "loss": 0.1289, + "step": 6495 + }, + { + "epoch": 0.05638839940625515, + "grad_norm": 0.412109375, + "learning_rate": 0.001985181472530496, + "loss": 0.1553, + "step": 6496 + }, + { + "epoch": 0.05639707988645932, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019851760934374444, + "loss": 0.1318, + "step": 6497 + }, + { + "epoch": 0.056405760366663484, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019851707133763776, + "loss": 0.1904, + "step": 6498 + }, + { + "epoch": 0.05641444084686765, + "grad_norm": 0.216796875, + "learning_rate": 0.0019851653323473013, + "loss": 0.1523, + "step": 6499 + }, + { + "epoch": 0.056423121327071814, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019851599503502215, + "loss": 0.1465, + "step": 6500 + }, + { + "epoch": 0.05643180180727598, + "grad_norm": 0.1650390625, + "learning_rate": 0.001985154567385144, + "loss": 0.1152, + "step": 6501 + }, + { + "epoch": 0.056440482287480144, + "grad_norm": 0.40625, + "learning_rate": 0.001985149183452075, + "loss": 0.1377, + "step": 6502 + }, + { + "epoch": 0.05644916276768431, + "grad_norm": 0.283203125, + "learning_rate": 0.00198514379855102, + "loss": 0.1484, + "step": 6503 + }, + { + "epoch": 0.056457843247888474, + "grad_norm": 0.09033203125, + "learning_rate": 0.001985138412681985, + "loss": 0.1445, + "step": 6504 + }, + { + "epoch": 0.05646652372809264, + "grad_norm": 0.18359375, + "learning_rate": 0.0019851330258449764, + "loss": 0.1377, + "step": 6505 + }, + { + "epoch": 0.056475204208296804, + "grad_norm": 0.1640625, + "learning_rate": 0.001985127638039999, + "loss": 0.1562, + "step": 6506 + }, + { + "epoch": 0.05648388468850097, + "grad_norm": 0.263671875, + "learning_rate": 0.0019851222492670594, + "loss": 0.1406, + "step": 6507 + }, + { + "epoch": 0.056492565168705135, + "grad_norm": 0.197265625, + "learning_rate": 0.0019851168595261637, + "loss": 0.1289, + "step": 6508 + }, + { + "epoch": 0.0565012456489093, + "grad_norm": 0.353515625, + "learning_rate": 0.0019851114688173173, + "loss": 0.1143, + "step": 6509 + }, + { + "epoch": 0.056509926129113465, + "grad_norm": 0.07666015625, + "learning_rate": 0.001985106077140526, + "loss": 0.1143, + "step": 6510 + }, + { + "epoch": 0.05651860660931763, + "grad_norm": 0.470703125, + "learning_rate": 0.001985100684495797, + "loss": 0.1016, + "step": 6511 + }, + { + "epoch": 0.056527287089521795, + "grad_norm": 0.380859375, + "learning_rate": 0.0019850952908831345, + "loss": 0.125, + "step": 6512 + }, + { + "epoch": 0.05653596756972596, + "grad_norm": 0.45703125, + "learning_rate": 0.001985089896302545, + "loss": 0.1279, + "step": 6513 + }, + { + "epoch": 0.056544648049930125, + "grad_norm": 0.416015625, + "learning_rate": 0.0019850845007540344, + "loss": 0.1045, + "step": 6514 + }, + { + "epoch": 0.056553328530134284, + "grad_norm": 0.076171875, + "learning_rate": 0.0019850791042376093, + "loss": 0.1221, + "step": 6515 + }, + { + "epoch": 0.05656200901033845, + "grad_norm": 0.52734375, + "learning_rate": 0.0019850737067532744, + "loss": 0.1465, + "step": 6516 + }, + { + "epoch": 0.056570689490542614, + "grad_norm": 0.0654296875, + "learning_rate": 0.001985068308301036, + "loss": 0.1064, + "step": 6517 + }, + { + "epoch": 0.05657936997074678, + "grad_norm": 0.1640625, + "learning_rate": 0.0019850629088809008, + "loss": 0.1196, + "step": 6518 + }, + { + "epoch": 0.056588050450950944, + "grad_norm": 0.89453125, + "learning_rate": 0.001985057508492874, + "loss": 0.1338, + "step": 6519 + }, + { + "epoch": 0.05659673093115511, + "grad_norm": 0.39453125, + "learning_rate": 0.0019850521071369614, + "loss": 0.1113, + "step": 6520 + }, + { + "epoch": 0.056605411411359274, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019850467048131695, + "loss": 0.1289, + "step": 6521 + }, + { + "epoch": 0.05661409189156344, + "grad_norm": 0.53125, + "learning_rate": 0.0019850413015215034, + "loss": 0.1631, + "step": 6522 + }, + { + "epoch": 0.056622772371767605, + "grad_norm": 0.146484375, + "learning_rate": 0.0019850358972619696, + "loss": 0.1089, + "step": 6523 + }, + { + "epoch": 0.05663145285197177, + "grad_norm": 0.86328125, + "learning_rate": 0.0019850304920345737, + "loss": 0.1504, + "step": 6524 + }, + { + "epoch": 0.056640133332175935, + "grad_norm": 0.6171875, + "learning_rate": 0.001985025085839322, + "loss": 0.1309, + "step": 6525 + }, + { + "epoch": 0.0566488138123801, + "grad_norm": 0.423828125, + "learning_rate": 0.0019850196786762196, + "loss": 0.1719, + "step": 6526 + }, + { + "epoch": 0.056657494292584265, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019850142705452735, + "loss": 0.165, + "step": 6527 + }, + { + "epoch": 0.05666617477278843, + "grad_norm": 0.0869140625, + "learning_rate": 0.001985008861446489, + "loss": 0.1436, + "step": 6528 + }, + { + "epoch": 0.056674855252992595, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019850034513798722, + "loss": 0.1504, + "step": 6529 + }, + { + "epoch": 0.05668353573319676, + "grad_norm": 0.228515625, + "learning_rate": 0.001984998040345429, + "loss": 0.1406, + "step": 6530 + }, + { + "epoch": 0.056692216213400926, + "grad_norm": 0.1474609375, + "learning_rate": 0.001984992628343165, + "loss": 0.1523, + "step": 6531 + }, + { + "epoch": 0.05670089669360509, + "grad_norm": 0.166015625, + "learning_rate": 0.0019849872153730862, + "loss": 0.0991, + "step": 6532 + }, + { + "epoch": 0.056709577173809256, + "grad_norm": 0.3203125, + "learning_rate": 0.001984981801435199, + "loss": 0.1377, + "step": 6533 + }, + { + "epoch": 0.05671825765401342, + "grad_norm": 0.42578125, + "learning_rate": 0.0019849763865295088, + "loss": 0.1602, + "step": 6534 + }, + { + "epoch": 0.056726938134217586, + "grad_norm": 0.298828125, + "learning_rate": 0.0019849709706560217, + "loss": 0.1455, + "step": 6535 + }, + { + "epoch": 0.05673561861442175, + "grad_norm": 0.15625, + "learning_rate": 0.0019849655538147436, + "loss": 0.1357, + "step": 6536 + }, + { + "epoch": 0.056744299094625916, + "grad_norm": 0.08984375, + "learning_rate": 0.0019849601360056807, + "loss": 0.1641, + "step": 6537 + }, + { + "epoch": 0.05675297957483008, + "grad_norm": 0.30859375, + "learning_rate": 0.0019849547172288385, + "loss": 0.123, + "step": 6538 + }, + { + "epoch": 0.056761660055034246, + "grad_norm": 0.7265625, + "learning_rate": 0.001984949297484223, + "loss": 0.1543, + "step": 6539 + }, + { + "epoch": 0.05677034053523841, + "grad_norm": 0.609375, + "learning_rate": 0.0019849438767718408, + "loss": 0.127, + "step": 6540 + }, + { + "epoch": 0.05677902101544258, + "grad_norm": 0.373046875, + "learning_rate": 0.001984938455091697, + "loss": 0.1709, + "step": 6541 + }, + { + "epoch": 0.05678770149564674, + "grad_norm": 0.12890625, + "learning_rate": 0.0019849330324437976, + "loss": 0.1348, + "step": 6542 + }, + { + "epoch": 0.05679638197585091, + "grad_norm": 0.1396484375, + "learning_rate": 0.001984927608828149, + "loss": 0.1436, + "step": 6543 + }, + { + "epoch": 0.05680506245605507, + "grad_norm": 0.27734375, + "learning_rate": 0.0019849221842447567, + "loss": 0.1592, + "step": 6544 + }, + { + "epoch": 0.05681374293625924, + "grad_norm": 0.478515625, + "learning_rate": 0.001984916758693627, + "loss": 0.1328, + "step": 6545 + }, + { + "epoch": 0.056822423416463395, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019849113321747654, + "loss": 0.0942, + "step": 6546 + }, + { + "epoch": 0.05683110389666756, + "grad_norm": 0.458984375, + "learning_rate": 0.001984905904688178, + "loss": 0.1133, + "step": 6547 + }, + { + "epoch": 0.056839784376871726, + "grad_norm": 0.18359375, + "learning_rate": 0.001984900476233871, + "loss": 0.1543, + "step": 6548 + }, + { + "epoch": 0.05684846485707589, + "grad_norm": 0.1875, + "learning_rate": 0.00198489504681185, + "loss": 0.1279, + "step": 6549 + }, + { + "epoch": 0.056857145337280056, + "grad_norm": 0.189453125, + "learning_rate": 0.001984889616422121, + "loss": 0.0879, + "step": 6550 + }, + { + "epoch": 0.05686582581748422, + "grad_norm": 0.6640625, + "learning_rate": 0.0019848841850646903, + "loss": 0.1719, + "step": 6551 + }, + { + "epoch": 0.056874506297688386, + "grad_norm": 0.314453125, + "learning_rate": 0.001984878752739563, + "loss": 0.1895, + "step": 6552 + }, + { + "epoch": 0.05688318677789255, + "grad_norm": 0.404296875, + "learning_rate": 0.0019848733194467462, + "loss": 0.1299, + "step": 6553 + }, + { + "epoch": 0.056891867258096716, + "grad_norm": 0.45703125, + "learning_rate": 0.001984867885186245, + "loss": 0.1543, + "step": 6554 + }, + { + "epoch": 0.05690054773830088, + "grad_norm": 0.18359375, + "learning_rate": 0.0019848624499580656, + "loss": 0.1504, + "step": 6555 + }, + { + "epoch": 0.05690922821850505, + "grad_norm": 0.49609375, + "learning_rate": 0.001984857013762214, + "loss": 0.127, + "step": 6556 + }, + { + "epoch": 0.05691790869870921, + "grad_norm": 0.28125, + "learning_rate": 0.001984851576598696, + "loss": 0.125, + "step": 6557 + }, + { + "epoch": 0.05692658917891338, + "grad_norm": 0.5703125, + "learning_rate": 0.0019848461384675176, + "loss": 0.1299, + "step": 6558 + }, + { + "epoch": 0.05693526965911754, + "grad_norm": 0.21875, + "learning_rate": 0.0019848406993686844, + "loss": 0.1035, + "step": 6559 + }, + { + "epoch": 0.05694395013932171, + "grad_norm": 0.7734375, + "learning_rate": 0.0019848352593022033, + "loss": 0.1758, + "step": 6560 + }, + { + "epoch": 0.05695263061952587, + "grad_norm": 0.1259765625, + "learning_rate": 0.001984829818268079, + "loss": 0.1338, + "step": 6561 + }, + { + "epoch": 0.05696131109973004, + "grad_norm": 0.224609375, + "learning_rate": 0.001984824376266319, + "loss": 0.105, + "step": 6562 + }, + { + "epoch": 0.0569699915799342, + "grad_norm": 0.1796875, + "learning_rate": 0.0019848189332969273, + "loss": 0.1836, + "step": 6563 + }, + { + "epoch": 0.05697867206013837, + "grad_norm": 0.3515625, + "learning_rate": 0.0019848134893599117, + "loss": 0.1348, + "step": 6564 + }, + { + "epoch": 0.05698735254034253, + "grad_norm": 0.099609375, + "learning_rate": 0.0019848080444552774, + "loss": 0.1309, + "step": 6565 + }, + { + "epoch": 0.0569960330205467, + "grad_norm": 0.322265625, + "learning_rate": 0.00198480259858303, + "loss": 0.1553, + "step": 6566 + }, + { + "epoch": 0.05700471350075086, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019847971517431757, + "loss": 0.0957, + "step": 6567 + }, + { + "epoch": 0.05701339398095503, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019847917039357206, + "loss": 0.2021, + "step": 6568 + }, + { + "epoch": 0.05702207446115919, + "grad_norm": 0.72265625, + "learning_rate": 0.0019847862551606705, + "loss": 0.1445, + "step": 6569 + }, + { + "epoch": 0.05703075494136336, + "grad_norm": 0.146484375, + "learning_rate": 0.0019847808054180315, + "loss": 0.1406, + "step": 6570 + }, + { + "epoch": 0.05703943542156752, + "grad_norm": 0.3046875, + "learning_rate": 0.00198477535470781, + "loss": 0.126, + "step": 6571 + }, + { + "epoch": 0.05704811590177169, + "grad_norm": 0.2275390625, + "learning_rate": 0.001984769903030011, + "loss": 0.1777, + "step": 6572 + }, + { + "epoch": 0.057056796381975854, + "grad_norm": 0.1708984375, + "learning_rate": 0.001984764450384641, + "loss": 0.127, + "step": 6573 + }, + { + "epoch": 0.05706547686218002, + "grad_norm": 0.23046875, + "learning_rate": 0.0019847589967717056, + "loss": 0.1436, + "step": 6574 + }, + { + "epoch": 0.057074157342384184, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019847535421912113, + "loss": 0.127, + "step": 6575 + }, + { + "epoch": 0.05708283782258835, + "grad_norm": 0.1904296875, + "learning_rate": 0.001984748086643164, + "loss": 0.1689, + "step": 6576 + }, + { + "epoch": 0.05709151830279251, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019847426301275693, + "loss": 0.105, + "step": 6577 + }, + { + "epoch": 0.05710019878299667, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019847371726444333, + "loss": 0.1777, + "step": 6578 + }, + { + "epoch": 0.05710887926320084, + "grad_norm": 0.64453125, + "learning_rate": 0.0019847317141937623, + "loss": 0.1245, + "step": 6579 + }, + { + "epoch": 0.057117559743405, + "grad_norm": 0.357421875, + "learning_rate": 0.001984726254775562, + "loss": 0.1738, + "step": 6580 + }, + { + "epoch": 0.05712624022360917, + "grad_norm": 0.53515625, + "learning_rate": 0.001984720794389838, + "loss": 0.1416, + "step": 6581 + }, + { + "epoch": 0.05713492070381333, + "grad_norm": 0.142578125, + "learning_rate": 0.0019847153330365967, + "loss": 0.1338, + "step": 6582 + }, + { + "epoch": 0.0571436011840175, + "grad_norm": 0.15234375, + "learning_rate": 0.001984709870715844, + "loss": 0.166, + "step": 6583 + }, + { + "epoch": 0.05715228166422166, + "grad_norm": 0.177734375, + "learning_rate": 0.001984704407427586, + "loss": 0.1201, + "step": 6584 + }, + { + "epoch": 0.05716096214442583, + "grad_norm": 0.7109375, + "learning_rate": 0.0019846989431718284, + "loss": 0.1621, + "step": 6585 + }, + { + "epoch": 0.05716964262462999, + "grad_norm": 0.37109375, + "learning_rate": 0.001984693477948578, + "loss": 0.1523, + "step": 6586 + }, + { + "epoch": 0.05717832310483416, + "grad_norm": 0.51953125, + "learning_rate": 0.001984688011757839, + "loss": 0.1592, + "step": 6587 + }, + { + "epoch": 0.057187003585038323, + "grad_norm": 0.6015625, + "learning_rate": 0.001984682544599619, + "loss": 0.1787, + "step": 6588 + }, + { + "epoch": 0.05719568406524249, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019846770764739233, + "loss": 0.1182, + "step": 6589 + }, + { + "epoch": 0.057204364545446654, + "grad_norm": 0.58203125, + "learning_rate": 0.001984671607380759, + "loss": 0.1699, + "step": 6590 + }, + { + "epoch": 0.05721304502565082, + "grad_norm": 0.2451171875, + "learning_rate": 0.00198466613732013, + "loss": 0.1504, + "step": 6591 + }, + { + "epoch": 0.057221725505854984, + "grad_norm": 0.083984375, + "learning_rate": 0.0019846606662920437, + "loss": 0.1475, + "step": 6592 + }, + { + "epoch": 0.05723040598605915, + "grad_norm": 0.10302734375, + "learning_rate": 0.001984655194296506, + "loss": 0.1465, + "step": 6593 + }, + { + "epoch": 0.057239086466263314, + "grad_norm": 0.3359375, + "learning_rate": 0.001984649721333522, + "loss": 0.1484, + "step": 6594 + }, + { + "epoch": 0.05724776694646748, + "grad_norm": 0.8203125, + "learning_rate": 0.0019846442474030996, + "loss": 0.1396, + "step": 6595 + }, + { + "epoch": 0.057256447426671644, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019846387725052424, + "loss": 0.1533, + "step": 6596 + }, + { + "epoch": 0.05726512790687581, + "grad_norm": 0.25390625, + "learning_rate": 0.0019846332966399582, + "loss": 0.127, + "step": 6597 + }, + { + "epoch": 0.057273808387079975, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019846278198072517, + "loss": 0.1318, + "step": 6598 + }, + { + "epoch": 0.05728248886728414, + "grad_norm": 0.384765625, + "learning_rate": 0.00198462234200713, + "loss": 0.1309, + "step": 6599 + }, + { + "epoch": 0.057291169347488305, + "grad_norm": 0.326171875, + "learning_rate": 0.0019846168632395985, + "loss": 0.1572, + "step": 6600 + }, + { + "epoch": 0.05729984982769247, + "grad_norm": 0.1064453125, + "learning_rate": 0.001984611383504663, + "loss": 0.123, + "step": 6601 + }, + { + "epoch": 0.057308530307896635, + "grad_norm": 1.390625, + "learning_rate": 0.0019846059028023303, + "loss": 0.1582, + "step": 6602 + }, + { + "epoch": 0.0573172107881008, + "grad_norm": 0.287109375, + "learning_rate": 0.0019846004211326058, + "loss": 0.1182, + "step": 6603 + }, + { + "epoch": 0.057325891268304965, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019845949384954954, + "loss": 0.1289, + "step": 6604 + }, + { + "epoch": 0.05733457174850913, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019845894548910052, + "loss": 0.1699, + "step": 6605 + }, + { + "epoch": 0.057343252228713296, + "grad_norm": 0.189453125, + "learning_rate": 0.0019845839703191413, + "loss": 0.1377, + "step": 6606 + }, + { + "epoch": 0.05735193270891746, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019845784847799097, + "loss": 0.1465, + "step": 6607 + }, + { + "epoch": 0.05736061318912162, + "grad_norm": 0.291015625, + "learning_rate": 0.0019845729982733164, + "loss": 0.124, + "step": 6608 + }, + { + "epoch": 0.057369293669325784, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019845675107993672, + "loss": 0.1348, + "step": 6609 + }, + { + "epoch": 0.05737797414952995, + "grad_norm": 0.255859375, + "learning_rate": 0.0019845620223580686, + "loss": 0.207, + "step": 6610 + }, + { + "epoch": 0.057386654629734114, + "grad_norm": 0.0703125, + "learning_rate": 0.0019845565329494256, + "loss": 0.1523, + "step": 6611 + }, + { + "epoch": 0.05739533510993828, + "grad_norm": 0.24609375, + "learning_rate": 0.0019845510425734454, + "loss": 0.1416, + "step": 6612 + }, + { + "epoch": 0.057404015590142445, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019845455512301335, + "loss": 0.1484, + "step": 6613 + }, + { + "epoch": 0.05741269607034661, + "grad_norm": 0.431640625, + "learning_rate": 0.001984540058919496, + "loss": 0.124, + "step": 6614 + }, + { + "epoch": 0.057421376550550775, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019845345656415384, + "loss": 0.1338, + "step": 6615 + }, + { + "epoch": 0.05743005703075494, + "grad_norm": 0.3125, + "learning_rate": 0.001984529071396267, + "loss": 0.1699, + "step": 6616 + }, + { + "epoch": 0.057438737510959105, + "grad_norm": 0.2392578125, + "learning_rate": 0.001984523576183688, + "loss": 0.1123, + "step": 6617 + }, + { + "epoch": 0.05744741799116327, + "grad_norm": 0.150390625, + "learning_rate": 0.0019845180800038076, + "loss": 0.1279, + "step": 6618 + }, + { + "epoch": 0.057456098471367435, + "grad_norm": 0.474609375, + "learning_rate": 0.0019845125828566314, + "loss": 0.1079, + "step": 6619 + }, + { + "epoch": 0.0574647789515716, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019845070847421655, + "loss": 0.1221, + "step": 6620 + }, + { + "epoch": 0.057473459431775765, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019845015856604156, + "loss": 0.1748, + "step": 6621 + }, + { + "epoch": 0.05748213991197993, + "grad_norm": 0.130859375, + "learning_rate": 0.0019844960856113883, + "loss": 0.1504, + "step": 6622 + }, + { + "epoch": 0.057490820392184096, + "grad_norm": 0.56640625, + "learning_rate": 0.0019844905845950896, + "loss": 0.1328, + "step": 6623 + }, + { + "epoch": 0.05749950087238826, + "grad_norm": 0.6640625, + "learning_rate": 0.001984485082611525, + "loss": 0.1572, + "step": 6624 + }, + { + "epoch": 0.057508181352592426, + "grad_norm": 0.326171875, + "learning_rate": 0.0019844795796607013, + "loss": 0.1328, + "step": 6625 + }, + { + "epoch": 0.05751686183279659, + "grad_norm": 0.12890625, + "learning_rate": 0.0019844740757426235, + "loss": 0.1533, + "step": 6626 + }, + { + "epoch": 0.057525542313000756, + "grad_norm": 0.2021484375, + "learning_rate": 0.001984468570857298, + "loss": 0.126, + "step": 6627 + }, + { + "epoch": 0.05753422279320492, + "grad_norm": 0.1484375, + "learning_rate": 0.0019844630650047313, + "loss": 0.1235, + "step": 6628 + }, + { + "epoch": 0.057542903273409086, + "grad_norm": 0.31640625, + "learning_rate": 0.001984457558184929, + "loss": 0.1309, + "step": 6629 + }, + { + "epoch": 0.05755158375361325, + "grad_norm": 0.8359375, + "learning_rate": 0.0019844520503978975, + "loss": 0.209, + "step": 6630 + }, + { + "epoch": 0.05756026423381742, + "grad_norm": 0.46875, + "learning_rate": 0.0019844465416436422, + "loss": 0.1328, + "step": 6631 + }, + { + "epoch": 0.05756894471402158, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019844410319221693, + "loss": 0.124, + "step": 6632 + }, + { + "epoch": 0.05757762519422575, + "grad_norm": 0.765625, + "learning_rate": 0.0019844355212334853, + "loss": 0.2031, + "step": 6633 + }, + { + "epoch": 0.05758630567442991, + "grad_norm": 0.5859375, + "learning_rate": 0.001984430009577596, + "loss": 0.1221, + "step": 6634 + }, + { + "epoch": 0.05759498615463408, + "grad_norm": 0.244140625, + "learning_rate": 0.0019844244969545073, + "loss": 0.1494, + "step": 6635 + }, + { + "epoch": 0.05760366663483824, + "grad_norm": 0.0859375, + "learning_rate": 0.0019844189833642246, + "loss": 0.1494, + "step": 6636 + }, + { + "epoch": 0.05761234711504241, + "grad_norm": 0.2490234375, + "learning_rate": 0.001984413468806755, + "loss": 0.1465, + "step": 6637 + }, + { + "epoch": 0.057621027595246566, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019844079532821044, + "loss": 0.1494, + "step": 6638 + }, + { + "epoch": 0.05762970807545073, + "grad_norm": 0.08154296875, + "learning_rate": 0.001984402436790278, + "loss": 0.1143, + "step": 6639 + }, + { + "epoch": 0.057638388555654896, + "grad_norm": 0.5859375, + "learning_rate": 0.001984396919331283, + "loss": 0.1152, + "step": 6640 + }, + { + "epoch": 0.05764706903585906, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019843914009051245, + "loss": 0.1387, + "step": 6641 + }, + { + "epoch": 0.057655749516063226, + "grad_norm": 0.1005859375, + "learning_rate": 0.001984385881511809, + "loss": 0.1152, + "step": 6642 + }, + { + "epoch": 0.05766442999626739, + "grad_norm": 0.26953125, + "learning_rate": 0.001984380361151342, + "loss": 0.1572, + "step": 6643 + }, + { + "epoch": 0.057673110476471556, + "grad_norm": 0.392578125, + "learning_rate": 0.00198437483982373, + "loss": 0.1689, + "step": 6644 + }, + { + "epoch": 0.05768179095667572, + "grad_norm": 0.212890625, + "learning_rate": 0.0019843693175289794, + "loss": 0.1211, + "step": 6645 + }, + { + "epoch": 0.05769047143687989, + "grad_norm": 0.65234375, + "learning_rate": 0.0019843637942670955, + "loss": 0.2051, + "step": 6646 + }, + { + "epoch": 0.05769915191708405, + "grad_norm": 0.37109375, + "learning_rate": 0.001984358270038085, + "loss": 0.1602, + "step": 6647 + }, + { + "epoch": 0.05770783239728822, + "grad_norm": 0.291015625, + "learning_rate": 0.0019843527448419533, + "loss": 0.1455, + "step": 6648 + }, + { + "epoch": 0.05771651287749238, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019843472186787067, + "loss": 0.1465, + "step": 6649 + }, + { + "epoch": 0.05772519335769655, + "grad_norm": 0.294921875, + "learning_rate": 0.0019843416915483513, + "loss": 0.1318, + "step": 6650 + }, + { + "epoch": 0.05773387383790071, + "grad_norm": 0.408203125, + "learning_rate": 0.0019843361634508928, + "loss": 0.1157, + "step": 6651 + }, + { + "epoch": 0.05774255431810488, + "grad_norm": 0.56640625, + "learning_rate": 0.001984330634386338, + "loss": 0.1416, + "step": 6652 + }, + { + "epoch": 0.05775123479830904, + "grad_norm": 0.265625, + "learning_rate": 0.0019843251043546923, + "loss": 0.1445, + "step": 6653 + }, + { + "epoch": 0.05775991527851321, + "grad_norm": 0.296875, + "learning_rate": 0.001984319573355962, + "loss": 0.1504, + "step": 6654 + }, + { + "epoch": 0.05776859575871737, + "grad_norm": 0.58984375, + "learning_rate": 0.001984314041390153, + "loss": 0.1504, + "step": 6655 + }, + { + "epoch": 0.05777727623892154, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019843085084572717, + "loss": 0.1592, + "step": 6656 + }, + { + "epoch": 0.0577859567191257, + "grad_norm": 0.67578125, + "learning_rate": 0.0019843029745573236, + "loss": 0.1846, + "step": 6657 + }, + { + "epoch": 0.05779463719932987, + "grad_norm": 0.7265625, + "learning_rate": 0.0019842974396903153, + "loss": 0.2246, + "step": 6658 + }, + { + "epoch": 0.05780331767953403, + "grad_norm": 0.296875, + "learning_rate": 0.0019842919038562524, + "loss": 0.124, + "step": 6659 + }, + { + "epoch": 0.0578119981597382, + "grad_norm": 0.216796875, + "learning_rate": 0.0019842863670551415, + "loss": 0.1162, + "step": 6660 + }, + { + "epoch": 0.05782067863994236, + "grad_norm": 0.875, + "learning_rate": 0.001984280829286988, + "loss": 0.1465, + "step": 6661 + }, + { + "epoch": 0.05782935912014653, + "grad_norm": 0.3515625, + "learning_rate": 0.0019842752905517985, + "loss": 0.1104, + "step": 6662 + }, + { + "epoch": 0.057838039600350694, + "grad_norm": 0.345703125, + "learning_rate": 0.0019842697508495786, + "loss": 0.125, + "step": 6663 + }, + { + "epoch": 0.05784672008055486, + "grad_norm": 0.61328125, + "learning_rate": 0.001984264210180335, + "loss": 0.1436, + "step": 6664 + }, + { + "epoch": 0.057855400560759024, + "grad_norm": 0.6796875, + "learning_rate": 0.0019842586685440726, + "loss": 0.2334, + "step": 6665 + }, + { + "epoch": 0.05786408104096319, + "grad_norm": 0.138671875, + "learning_rate": 0.0019842531259407986, + "loss": 0.1138, + "step": 6666 + }, + { + "epoch": 0.057872761521167354, + "grad_norm": 0.388671875, + "learning_rate": 0.001984247582370519, + "loss": 0.1211, + "step": 6667 + }, + { + "epoch": 0.05788144200137152, + "grad_norm": 0.1630859375, + "learning_rate": 0.001984242037833239, + "loss": 0.1279, + "step": 6668 + }, + { + "epoch": 0.05789012248157568, + "grad_norm": 0.7578125, + "learning_rate": 0.0019842364923289657, + "loss": 0.165, + "step": 6669 + }, + { + "epoch": 0.05789880296177984, + "grad_norm": 0.328125, + "learning_rate": 0.0019842309458577042, + "loss": 0.1396, + "step": 6670 + }, + { + "epoch": 0.05790748344198401, + "grad_norm": 0.0966796875, + "learning_rate": 0.001984225398419461, + "loss": 0.1221, + "step": 6671 + }, + { + "epoch": 0.05791616392218817, + "grad_norm": 0.3671875, + "learning_rate": 0.0019842198500142428, + "loss": 0.1768, + "step": 6672 + }, + { + "epoch": 0.05792484440239234, + "grad_norm": 0.11328125, + "learning_rate": 0.0019842143006420544, + "loss": 0.1318, + "step": 6673 + }, + { + "epoch": 0.0579335248825965, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019842087503029026, + "loss": 0.166, + "step": 6674 + }, + { + "epoch": 0.05794220536280067, + "grad_norm": 0.2158203125, + "learning_rate": 0.001984203198996794, + "loss": 0.126, + "step": 6675 + }, + { + "epoch": 0.05795088584300483, + "grad_norm": 0.51953125, + "learning_rate": 0.0019841976467237336, + "loss": 0.1396, + "step": 6676 + }, + { + "epoch": 0.057959566323209, + "grad_norm": 0.103515625, + "learning_rate": 0.0019841920934837278, + "loss": 0.1533, + "step": 6677 + }, + { + "epoch": 0.05796824680341316, + "grad_norm": 0.2890625, + "learning_rate": 0.001984186539276783, + "loss": 0.1475, + "step": 6678 + }, + { + "epoch": 0.05797692728361733, + "grad_norm": 0.51171875, + "learning_rate": 0.001984180984102905, + "loss": 0.1699, + "step": 6679 + }, + { + "epoch": 0.057985607763821494, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019841754279621, + "loss": 0.125, + "step": 6680 + }, + { + "epoch": 0.05799428824402566, + "grad_norm": 0.0986328125, + "learning_rate": 0.001984169870854374, + "loss": 0.1357, + "step": 6681 + }, + { + "epoch": 0.058002968724229824, + "grad_norm": 0.60546875, + "learning_rate": 0.001984164312779733, + "loss": 0.1367, + "step": 6682 + }, + { + "epoch": 0.05801164920443399, + "grad_norm": 0.490234375, + "learning_rate": 0.0019841587537381836, + "loss": 0.1855, + "step": 6683 + }, + { + "epoch": 0.058020329684638154, + "grad_norm": 0.306640625, + "learning_rate": 0.0019841531937297313, + "loss": 0.1289, + "step": 6684 + }, + { + "epoch": 0.05802901016484232, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019841476327543822, + "loss": 0.1621, + "step": 6685 + }, + { + "epoch": 0.058037690645046484, + "grad_norm": 0.75, + "learning_rate": 0.0019841420708121425, + "loss": 0.1309, + "step": 6686 + }, + { + "epoch": 0.05804637112525065, + "grad_norm": 0.158203125, + "learning_rate": 0.0019841365079030187, + "loss": 0.124, + "step": 6687 + }, + { + "epoch": 0.058055051605454815, + "grad_norm": 0.2255859375, + "learning_rate": 0.001984130944027016, + "loss": 0.1494, + "step": 6688 + }, + { + "epoch": 0.05806373208565898, + "grad_norm": 0.5859375, + "learning_rate": 0.0019841253791841418, + "loss": 0.1387, + "step": 6689 + }, + { + "epoch": 0.058072412565863145, + "grad_norm": 0.384765625, + "learning_rate": 0.0019841198133744007, + "loss": 0.126, + "step": 6690 + }, + { + "epoch": 0.05808109304606731, + "grad_norm": 0.4140625, + "learning_rate": 0.0019841142465977994, + "loss": 0.1719, + "step": 6691 + }, + { + "epoch": 0.058089773526271475, + "grad_norm": 0.38671875, + "learning_rate": 0.001984108678854344, + "loss": 0.1055, + "step": 6692 + }, + { + "epoch": 0.05809845400647564, + "grad_norm": 0.11279296875, + "learning_rate": 0.001984103110144041, + "loss": 0.123, + "step": 6693 + }, + { + "epoch": 0.058107134486679805, + "grad_norm": 0.390625, + "learning_rate": 0.001984097540466896, + "loss": 0.1377, + "step": 6694 + }, + { + "epoch": 0.05811581496688397, + "grad_norm": 0.06640625, + "learning_rate": 0.001984091969822915, + "loss": 0.1191, + "step": 6695 + }, + { + "epoch": 0.058124495447088136, + "grad_norm": 0.181640625, + "learning_rate": 0.0019840863982121047, + "loss": 0.1357, + "step": 6696 + }, + { + "epoch": 0.0581331759272923, + "grad_norm": 0.439453125, + "learning_rate": 0.0019840808256344706, + "loss": 0.1172, + "step": 6697 + }, + { + "epoch": 0.058141856407496466, + "grad_norm": 0.44140625, + "learning_rate": 0.001984075252090019, + "loss": 0.1172, + "step": 6698 + }, + { + "epoch": 0.05815053688770063, + "grad_norm": 0.671875, + "learning_rate": 0.001984069677578756, + "loss": 0.1357, + "step": 6699 + }, + { + "epoch": 0.05815921736790479, + "grad_norm": 0.484375, + "learning_rate": 0.0019840641021006875, + "loss": 0.1465, + "step": 6700 + }, + { + "epoch": 0.058167897848108954, + "grad_norm": 0.55078125, + "learning_rate": 0.00198405852565582, + "loss": 0.0942, + "step": 6701 + }, + { + "epoch": 0.05817657832831312, + "grad_norm": 0.50390625, + "learning_rate": 0.001984052948244159, + "loss": 0.2178, + "step": 6702 + }, + { + "epoch": 0.058185258808517284, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019840473698657115, + "loss": 0.1709, + "step": 6703 + }, + { + "epoch": 0.05819393928872145, + "grad_norm": 0.859375, + "learning_rate": 0.001984041790520483, + "loss": 0.1416, + "step": 6704 + }, + { + "epoch": 0.058202619768925615, + "grad_norm": 0.671875, + "learning_rate": 0.0019840362102084793, + "loss": 0.1484, + "step": 6705 + }, + { + "epoch": 0.05821130024912978, + "grad_norm": 0.79296875, + "learning_rate": 0.0019840306289297073, + "loss": 0.124, + "step": 6706 + }, + { + "epoch": 0.058219980729333945, + "grad_norm": 0.275390625, + "learning_rate": 0.001984025046684172, + "loss": 0.1357, + "step": 6707 + }, + { + "epoch": 0.05822866120953811, + "grad_norm": 0.365234375, + "learning_rate": 0.0019840194634718807, + "loss": 0.1982, + "step": 6708 + }, + { + "epoch": 0.058237341689742275, + "grad_norm": 0.08203125, + "learning_rate": 0.0019840138792928392, + "loss": 0.1445, + "step": 6709 + }, + { + "epoch": 0.05824602216994644, + "grad_norm": 0.404296875, + "learning_rate": 0.0019840082941470533, + "loss": 0.1592, + "step": 6710 + }, + { + "epoch": 0.058254702650150605, + "grad_norm": 0.1513671875, + "learning_rate": 0.001984002708034529, + "loss": 0.1348, + "step": 6711 + }, + { + "epoch": 0.05826338313035477, + "grad_norm": 0.796875, + "learning_rate": 0.001983997120955273, + "loss": 0.1348, + "step": 6712 + }, + { + "epoch": 0.058272063610558936, + "grad_norm": 0.59375, + "learning_rate": 0.0019839915329092905, + "loss": 0.1172, + "step": 6713 + }, + { + "epoch": 0.0582807440907631, + "grad_norm": 0.859375, + "learning_rate": 0.0019839859438965887, + "loss": 0.1465, + "step": 6714 + }, + { + "epoch": 0.058289424570967266, + "grad_norm": 0.396484375, + "learning_rate": 0.0019839803539171727, + "loss": 0.1299, + "step": 6715 + }, + { + "epoch": 0.05829810505117143, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019839747629710495, + "loss": 0.124, + "step": 6716 + }, + { + "epoch": 0.058306785531375596, + "grad_norm": 1.0, + "learning_rate": 0.0019839691710582244, + "loss": 0.1592, + "step": 6717 + }, + { + "epoch": 0.05831546601157976, + "grad_norm": 0.466796875, + "learning_rate": 0.0019839635781787037, + "loss": 0.1523, + "step": 6718 + }, + { + "epoch": 0.058324146491783926, + "grad_norm": 0.984375, + "learning_rate": 0.001983957984332494, + "loss": 0.1328, + "step": 6719 + }, + { + "epoch": 0.05833282697198809, + "grad_norm": 0.19921875, + "learning_rate": 0.0019839523895196012, + "loss": 0.1162, + "step": 6720 + }, + { + "epoch": 0.05834150745219226, + "grad_norm": 0.82421875, + "learning_rate": 0.0019839467937400315, + "loss": 0.1719, + "step": 6721 + }, + { + "epoch": 0.05835018793239642, + "grad_norm": 0.4609375, + "learning_rate": 0.0019839411969937906, + "loss": 0.1348, + "step": 6722 + }, + { + "epoch": 0.05835886841260059, + "grad_norm": 0.486328125, + "learning_rate": 0.001983935599280885, + "loss": 0.1602, + "step": 6723 + }, + { + "epoch": 0.05836754889280475, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019839300006013207, + "loss": 0.1816, + "step": 6724 + }, + { + "epoch": 0.05837622937300892, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019839244009551035, + "loss": 0.1807, + "step": 6725 + }, + { + "epoch": 0.05838490985321308, + "grad_norm": 0.15234375, + "learning_rate": 0.0019839188003422403, + "loss": 0.1152, + "step": 6726 + }, + { + "epoch": 0.05839359033341725, + "grad_norm": 1.484375, + "learning_rate": 0.001983913198762736, + "loss": 0.1934, + "step": 6727 + }, + { + "epoch": 0.05840227081362141, + "grad_norm": 0.33203125, + "learning_rate": 0.0019839075962165987, + "loss": 0.1328, + "step": 6728 + }, + { + "epoch": 0.05841095129382558, + "grad_norm": 0.45703125, + "learning_rate": 0.001983901992703833, + "loss": 0.124, + "step": 6729 + }, + { + "epoch": 0.05841963177402974, + "grad_norm": 0.78515625, + "learning_rate": 0.0019838963882244445, + "loss": 0.1279, + "step": 6730 + }, + { + "epoch": 0.0584283122542339, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019838907827784413, + "loss": 0.126, + "step": 6731 + }, + { + "epoch": 0.058436992734438066, + "grad_norm": 0.5625, + "learning_rate": 0.0019838851763658277, + "loss": 0.1748, + "step": 6732 + }, + { + "epoch": 0.05844567321464223, + "grad_norm": 0.369140625, + "learning_rate": 0.001983879568986611, + "loss": 0.1699, + "step": 6733 + }, + { + "epoch": 0.058454353694846396, + "grad_norm": 0.302734375, + "learning_rate": 0.0019838739606407967, + "loss": 0.1104, + "step": 6734 + }, + { + "epoch": 0.05846303417505056, + "grad_norm": 0.2119140625, + "learning_rate": 0.001983868351328391, + "loss": 0.1523, + "step": 6735 + }, + { + "epoch": 0.058471714655254727, + "grad_norm": 0.337890625, + "learning_rate": 0.0019838627410494, + "loss": 0.123, + "step": 6736 + }, + { + "epoch": 0.05848039513545889, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019838571298038303, + "loss": 0.1318, + "step": 6737 + }, + { + "epoch": 0.05848907561566306, + "grad_norm": 0.134765625, + "learning_rate": 0.0019838515175916874, + "loss": 0.1245, + "step": 6738 + }, + { + "epoch": 0.05849775609586722, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019838459044129782, + "loss": 0.1152, + "step": 6739 + }, + { + "epoch": 0.05850643657607139, + "grad_norm": 0.6328125, + "learning_rate": 0.0019838402902677085, + "loss": 0.1377, + "step": 6740 + }, + { + "epoch": 0.05851511705627555, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019838346751558837, + "loss": 0.1895, + "step": 6741 + }, + { + "epoch": 0.05852379753647972, + "grad_norm": 0.087890625, + "learning_rate": 0.001983829059077511, + "loss": 0.1099, + "step": 6742 + }, + { + "epoch": 0.05853247801668388, + "grad_norm": 0.11572265625, + "learning_rate": 0.001983823442032596, + "loss": 0.1211, + "step": 6743 + }, + { + "epoch": 0.05854115849688805, + "grad_norm": 0.373046875, + "learning_rate": 0.0019838178240211453, + "loss": 0.1611, + "step": 6744 + }, + { + "epoch": 0.05854983897709221, + "grad_norm": 0.26953125, + "learning_rate": 0.0019838122050431645, + "loss": 0.1699, + "step": 6745 + }, + { + "epoch": 0.05855851945729638, + "grad_norm": 0.36328125, + "learning_rate": 0.0019838065850986595, + "loss": 0.127, + "step": 6746 + }, + { + "epoch": 0.05856719993750054, + "grad_norm": 0.365234375, + "learning_rate": 0.0019838009641876373, + "loss": 0.1406, + "step": 6747 + }, + { + "epoch": 0.05857588041770471, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019837953423101034, + "loss": 0.127, + "step": 6748 + }, + { + "epoch": 0.05858456089790887, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019837897194660645, + "loss": 0.1504, + "step": 6749 + }, + { + "epoch": 0.05859324137811304, + "grad_norm": 1.1328125, + "learning_rate": 0.0019837840956555266, + "loss": 0.1377, + "step": 6750 + }, + { + "epoch": 0.0586019218583172, + "grad_norm": 0.22265625, + "learning_rate": 0.0019837784708784953, + "loss": 0.1377, + "step": 6751 + }, + { + "epoch": 0.05861060233852137, + "grad_norm": 0.76171875, + "learning_rate": 0.001983772845134977, + "loss": 0.1123, + "step": 6752 + }, + { + "epoch": 0.058619282818725534, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019837672184249786, + "loss": 0.0991, + "step": 6753 + }, + { + "epoch": 0.0586279632989297, + "grad_norm": 0.734375, + "learning_rate": 0.0019837615907485054, + "loss": 0.1445, + "step": 6754 + }, + { + "epoch": 0.058636643779133864, + "grad_norm": 0.30078125, + "learning_rate": 0.0019837559621055634, + "loss": 0.1562, + "step": 6755 + }, + { + "epoch": 0.05864532425933803, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019837503324961598, + "loss": 0.1699, + "step": 6756 + }, + { + "epoch": 0.058654004739542194, + "grad_norm": 0.25, + "learning_rate": 0.0019837447019202996, + "loss": 0.1201, + "step": 6757 + }, + { + "epoch": 0.05866268521974636, + "grad_norm": 0.69140625, + "learning_rate": 0.0019837390703779894, + "loss": 0.1758, + "step": 6758 + }, + { + "epoch": 0.058671365699950524, + "grad_norm": 0.220703125, + "learning_rate": 0.001983733437869236, + "loss": 0.1201, + "step": 6759 + }, + { + "epoch": 0.05868004618015469, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019837278043940447, + "loss": 0.1357, + "step": 6760 + }, + { + "epoch": 0.058688726660358854, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019837221699524218, + "loss": 0.1572, + "step": 6761 + }, + { + "epoch": 0.05869740714056301, + "grad_norm": 0.4921875, + "learning_rate": 0.001983716534544374, + "loss": 0.1562, + "step": 6762 + }, + { + "epoch": 0.05870608762076718, + "grad_norm": 0.33984375, + "learning_rate": 0.001983710898169907, + "loss": 0.1172, + "step": 6763 + }, + { + "epoch": 0.05871476810097134, + "grad_norm": 0.8125, + "learning_rate": 0.001983705260829027, + "loss": 0.1602, + "step": 6764 + }, + { + "epoch": 0.05872344858117551, + "grad_norm": 0.158203125, + "learning_rate": 0.0019836996225217397, + "loss": 0.1147, + "step": 6765 + }, + { + "epoch": 0.05873212906137967, + "grad_norm": 0.11376953125, + "learning_rate": 0.001983693983248052, + "loss": 0.1641, + "step": 6766 + }, + { + "epoch": 0.05874080954158384, + "grad_norm": 0.84765625, + "learning_rate": 0.00198368834300797, + "loss": 0.1504, + "step": 6767 + }, + { + "epoch": 0.058749490021788, + "grad_norm": 0.55859375, + "learning_rate": 0.0019836827018015, + "loss": 0.1553, + "step": 6768 + }, + { + "epoch": 0.05875817050199217, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019836770596286476, + "loss": 0.1133, + "step": 6769 + }, + { + "epoch": 0.058766850982196334, + "grad_norm": 0.21484375, + "learning_rate": 0.001983671416489419, + "loss": 0.1465, + "step": 6770 + }, + { + "epoch": 0.0587755314624005, + "grad_norm": 0.384765625, + "learning_rate": 0.001983665772383821, + "loss": 0.1094, + "step": 6771 + }, + { + "epoch": 0.058784211942604664, + "grad_norm": 0.0693359375, + "learning_rate": 0.001983660127311859, + "loss": 0.1157, + "step": 6772 + }, + { + "epoch": 0.05879289242280883, + "grad_norm": 3.609375, + "learning_rate": 0.00198365448127354, + "loss": 0.2578, + "step": 6773 + }, + { + "epoch": 0.058801572903012994, + "grad_norm": 0.09326171875, + "learning_rate": 0.001983648834268869, + "loss": 0.1289, + "step": 6774 + }, + { + "epoch": 0.05881025338321716, + "grad_norm": 0.078125, + "learning_rate": 0.0019836431862978537, + "loss": 0.126, + "step": 6775 + }, + { + "epoch": 0.058818933863421324, + "grad_norm": 0.58203125, + "learning_rate": 0.0019836375373604993, + "loss": 0.1309, + "step": 6776 + }, + { + "epoch": 0.05882761434362549, + "grad_norm": 0.9453125, + "learning_rate": 0.001983631887456812, + "loss": 0.167, + "step": 6777 + }, + { + "epoch": 0.058836294823829655, + "grad_norm": 0.09326171875, + "learning_rate": 0.001983626236586798, + "loss": 0.1523, + "step": 6778 + }, + { + "epoch": 0.05884497530403382, + "grad_norm": 0.091796875, + "learning_rate": 0.001983620584750464, + "loss": 0.127, + "step": 6779 + }, + { + "epoch": 0.058853655784237985, + "grad_norm": 0.275390625, + "learning_rate": 0.0019836149319478157, + "loss": 0.1406, + "step": 6780 + }, + { + "epoch": 0.05886233626444215, + "grad_norm": 0.322265625, + "learning_rate": 0.001983609278178859, + "loss": 0.125, + "step": 6781 + }, + { + "epoch": 0.058871016744646315, + "grad_norm": 0.12353515625, + "learning_rate": 0.001983603623443601, + "loss": 0.1475, + "step": 6782 + }, + { + "epoch": 0.05887969722485048, + "grad_norm": 0.66015625, + "learning_rate": 0.0019835979677420477, + "loss": 0.1426, + "step": 6783 + }, + { + "epoch": 0.058888377705054645, + "grad_norm": 0.16015625, + "learning_rate": 0.001983592311074204, + "loss": 0.1377, + "step": 6784 + }, + { + "epoch": 0.05889705818525881, + "grad_norm": 0.19140625, + "learning_rate": 0.001983586653440077, + "loss": 0.1875, + "step": 6785 + }, + { + "epoch": 0.058905738665462976, + "grad_norm": 0.19921875, + "learning_rate": 0.0019835809948396737, + "loss": 0.1289, + "step": 6786 + }, + { + "epoch": 0.05891441914566714, + "grad_norm": 0.08740234375, + "learning_rate": 0.001983575335272999, + "loss": 0.1045, + "step": 6787 + }, + { + "epoch": 0.058923099625871306, + "grad_norm": 0.28515625, + "learning_rate": 0.00198356967474006, + "loss": 0.1426, + "step": 6788 + }, + { + "epoch": 0.05893178010607547, + "grad_norm": 0.142578125, + "learning_rate": 0.0019835640132408617, + "loss": 0.1572, + "step": 6789 + }, + { + "epoch": 0.058940460586279636, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019835583507754116, + "loss": 0.1338, + "step": 6790 + }, + { + "epoch": 0.0589491410664838, + "grad_norm": 0.462890625, + "learning_rate": 0.0019835526873437158, + "loss": 0.1865, + "step": 6791 + }, + { + "epoch": 0.05895782154668796, + "grad_norm": 0.3046875, + "learning_rate": 0.0019835470229457794, + "loss": 0.1758, + "step": 6792 + }, + { + "epoch": 0.058966502026892124, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019835413575816094, + "loss": 0.1191, + "step": 6793 + }, + { + "epoch": 0.05897518250709629, + "grad_norm": 0.123046875, + "learning_rate": 0.001983535691251212, + "loss": 0.1299, + "step": 6794 + }, + { + "epoch": 0.058983862987300455, + "grad_norm": 0.81640625, + "learning_rate": 0.0019835300239545934, + "loss": 0.1338, + "step": 6795 + }, + { + "epoch": 0.05899254346750462, + "grad_norm": 0.1162109375, + "learning_rate": 0.001983524355691759, + "loss": 0.126, + "step": 6796 + }, + { + "epoch": 0.059001223947708785, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019835186864627158, + "loss": 0.2129, + "step": 6797 + }, + { + "epoch": 0.05900990442791295, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019835130162674706, + "loss": 0.1377, + "step": 6798 + }, + { + "epoch": 0.059018584908117115, + "grad_norm": 0.94921875, + "learning_rate": 0.001983507345106028, + "loss": 0.1201, + "step": 6799 + }, + { + "epoch": 0.05902726538832128, + "grad_norm": 3.203125, + "learning_rate": 0.001983501672978395, + "loss": 0.1309, + "step": 6800 + }, + { + "epoch": 0.059035945868525445, + "grad_norm": 0.70703125, + "learning_rate": 0.0019834959998845784, + "loss": 0.1445, + "step": 6801 + }, + { + "epoch": 0.05904462634872961, + "grad_norm": 0.08837890625, + "learning_rate": 0.001983490325824584, + "loss": 0.125, + "step": 6802 + }, + { + "epoch": 0.059053306828933776, + "grad_norm": 0.515625, + "learning_rate": 0.001983484650798417, + "loss": 0.1611, + "step": 6803 + }, + { + "epoch": 0.05906198730913794, + "grad_norm": 1.421875, + "learning_rate": 0.001983478974806085, + "loss": 0.2285, + "step": 6804 + }, + { + "epoch": 0.059070667789342106, + "grad_norm": 0.318359375, + "learning_rate": 0.0019834732978475937, + "loss": 0.1455, + "step": 6805 + }, + { + "epoch": 0.05907934826954627, + "grad_norm": 0.408203125, + "learning_rate": 0.001983467619922949, + "loss": 0.1465, + "step": 6806 + }, + { + "epoch": 0.059088028749750436, + "grad_norm": 0.5, + "learning_rate": 0.0019834619410321577, + "loss": 0.2422, + "step": 6807 + }, + { + "epoch": 0.0590967092299546, + "grad_norm": 0.333984375, + "learning_rate": 0.0019834562611752257, + "loss": 0.1182, + "step": 6808 + }, + { + "epoch": 0.059105389710158766, + "grad_norm": 0.5390625, + "learning_rate": 0.001983450580352159, + "loss": 0.1396, + "step": 6809 + }, + { + "epoch": 0.05911407019036293, + "grad_norm": 0.55078125, + "learning_rate": 0.0019834448985629636, + "loss": 0.1406, + "step": 6810 + }, + { + "epoch": 0.0591227506705671, + "grad_norm": 0.87109375, + "learning_rate": 0.0019834392158076474, + "loss": 0.2041, + "step": 6811 + }, + { + "epoch": 0.05913143115077126, + "grad_norm": 0.38671875, + "learning_rate": 0.0019834335320862142, + "loss": 0.1367, + "step": 6812 + }, + { + "epoch": 0.05914011163097543, + "grad_norm": 0.365234375, + "learning_rate": 0.001983427847398672, + "loss": 0.1226, + "step": 6813 + }, + { + "epoch": 0.05914879211117959, + "grad_norm": 0.1953125, + "learning_rate": 0.0019834221617450257, + "loss": 0.1465, + "step": 6814 + }, + { + "epoch": 0.05915747259138376, + "grad_norm": 0.55859375, + "learning_rate": 0.001983416475125283, + "loss": 0.1465, + "step": 6815 + }, + { + "epoch": 0.05916615307158792, + "grad_norm": 0.345703125, + "learning_rate": 0.0019834107875394487, + "loss": 0.1162, + "step": 6816 + }, + { + "epoch": 0.05917483355179209, + "grad_norm": 0.62890625, + "learning_rate": 0.0019834050989875297, + "loss": 0.1602, + "step": 6817 + }, + { + "epoch": 0.05918351403199625, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019833994094695323, + "loss": 0.127, + "step": 6818 + }, + { + "epoch": 0.05919219451220042, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019833937189854627, + "loss": 0.1504, + "step": 6819 + }, + { + "epoch": 0.05920087499240458, + "grad_norm": 0.423828125, + "learning_rate": 0.001983388027535327, + "loss": 0.1641, + "step": 6820 + }, + { + "epoch": 0.05920955547260875, + "grad_norm": 0.41015625, + "learning_rate": 0.0019833823351191313, + "loss": 0.1533, + "step": 6821 + }, + { + "epoch": 0.05921823595281291, + "grad_norm": 0.58984375, + "learning_rate": 0.0019833766417368818, + "loss": 0.166, + "step": 6822 + }, + { + "epoch": 0.05922691643301707, + "grad_norm": 0.267578125, + "learning_rate": 0.001983370947388585, + "loss": 0.168, + "step": 6823 + }, + { + "epoch": 0.059235596913221236, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019833652520742473, + "loss": 0.1182, + "step": 6824 + }, + { + "epoch": 0.0592442773934254, + "grad_norm": 0.193359375, + "learning_rate": 0.001983359555793874, + "loss": 0.1465, + "step": 6825 + }, + { + "epoch": 0.059252957873629566, + "grad_norm": 0.98046875, + "learning_rate": 0.001983353858547473, + "loss": 0.1133, + "step": 6826 + }, + { + "epoch": 0.05926163835383373, + "grad_norm": 0.400390625, + "learning_rate": 0.0019833481603350484, + "loss": 0.1094, + "step": 6827 + }, + { + "epoch": 0.0592703188340379, + "grad_norm": 0.640625, + "learning_rate": 0.001983342461156608, + "loss": 0.1348, + "step": 6828 + }, + { + "epoch": 0.05927899931424206, + "grad_norm": 0.353515625, + "learning_rate": 0.001983336761012158, + "loss": 0.1738, + "step": 6829 + }, + { + "epoch": 0.05928767979444623, + "grad_norm": 0.20703125, + "learning_rate": 0.0019833310599017035, + "loss": 0.1279, + "step": 6830 + }, + { + "epoch": 0.05929636027465039, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019833253578252514, + "loss": 0.1299, + "step": 6831 + }, + { + "epoch": 0.05930504075485456, + "grad_norm": 0.1044921875, + "learning_rate": 0.001983319654782808, + "loss": 0.125, + "step": 6832 + }, + { + "epoch": 0.05931372123505872, + "grad_norm": 0.341796875, + "learning_rate": 0.00198331395077438, + "loss": 0.1123, + "step": 6833 + }, + { + "epoch": 0.05932240171526289, + "grad_norm": 0.52734375, + "learning_rate": 0.001983308245799973, + "loss": 0.1328, + "step": 6834 + }, + { + "epoch": 0.05933108219546705, + "grad_norm": 0.447265625, + "learning_rate": 0.001983302539859593, + "loss": 0.2012, + "step": 6835 + }, + { + "epoch": 0.05933976267567122, + "grad_norm": 0.6796875, + "learning_rate": 0.001983296832953247, + "loss": 0.1387, + "step": 6836 + }, + { + "epoch": 0.05934844315587538, + "grad_norm": 1.0078125, + "learning_rate": 0.0019832911250809403, + "loss": 0.1406, + "step": 6837 + }, + { + "epoch": 0.05935712363607955, + "grad_norm": 0.373046875, + "learning_rate": 0.0019832854162426805, + "loss": 0.1426, + "step": 6838 + }, + { + "epoch": 0.05936580411628371, + "grad_norm": 0.341796875, + "learning_rate": 0.001983279706438472, + "loss": 0.106, + "step": 6839 + }, + { + "epoch": 0.05937448459648788, + "grad_norm": 0.330078125, + "learning_rate": 0.001983273995668323, + "loss": 0.1113, + "step": 6840 + }, + { + "epoch": 0.05938316507669204, + "grad_norm": 1.7421875, + "learning_rate": 0.0019832682839322387, + "loss": 0.1406, + "step": 6841 + }, + { + "epoch": 0.05939184555689621, + "grad_norm": 0.1455078125, + "learning_rate": 0.001983262571230225, + "loss": 0.1309, + "step": 6842 + }, + { + "epoch": 0.059400526037100373, + "grad_norm": 0.53515625, + "learning_rate": 0.001983256857562289, + "loss": 0.1895, + "step": 6843 + }, + { + "epoch": 0.05940920651730454, + "grad_norm": 0.578125, + "learning_rate": 0.0019832511429284366, + "loss": 0.1084, + "step": 6844 + }, + { + "epoch": 0.059417886997508704, + "grad_norm": 0.212890625, + "learning_rate": 0.0019832454273286735, + "loss": 0.1504, + "step": 6845 + }, + { + "epoch": 0.05942656747771287, + "grad_norm": 0.119140625, + "learning_rate": 0.0019832397107630074, + "loss": 0.1445, + "step": 6846 + }, + { + "epoch": 0.059435247957917034, + "grad_norm": 0.146484375, + "learning_rate": 0.001983233993231443, + "loss": 0.1699, + "step": 6847 + }, + { + "epoch": 0.0594439284381212, + "grad_norm": 0.203125, + "learning_rate": 0.001983228274733987, + "loss": 0.1289, + "step": 6848 + }, + { + "epoch": 0.059452608918325364, + "grad_norm": 0.84375, + "learning_rate": 0.001983222555270646, + "loss": 0.1855, + "step": 6849 + }, + { + "epoch": 0.05946128939852953, + "grad_norm": 0.20703125, + "learning_rate": 0.0019832168348414268, + "loss": 0.1982, + "step": 6850 + }, + { + "epoch": 0.059469969878733694, + "grad_norm": 1.0859375, + "learning_rate": 0.001983211113446334, + "loss": 0.1797, + "step": 6851 + }, + { + "epoch": 0.05947865035893786, + "grad_norm": 0.890625, + "learning_rate": 0.001983205391085375, + "loss": 0.1494, + "step": 6852 + }, + { + "epoch": 0.059487330839142025, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019831996677585565, + "loss": 0.1582, + "step": 6853 + }, + { + "epoch": 0.05949601131934618, + "grad_norm": 0.228515625, + "learning_rate": 0.0019831939434658834, + "loss": 0.1104, + "step": 6854 + }, + { + "epoch": 0.05950469179955035, + "grad_norm": 0.6484375, + "learning_rate": 0.0019831882182073626, + "loss": 0.1289, + "step": 6855 + }, + { + "epoch": 0.05951337227975451, + "grad_norm": 0.361328125, + "learning_rate": 0.0019831824919830008, + "loss": 0.1699, + "step": 6856 + }, + { + "epoch": 0.05952205275995868, + "grad_norm": 0.443359375, + "learning_rate": 0.001983176764792804, + "loss": 0.1689, + "step": 6857 + }, + { + "epoch": 0.05953073324016284, + "grad_norm": 0.5703125, + "learning_rate": 0.001983171036636778, + "loss": 0.1089, + "step": 6858 + }, + { + "epoch": 0.05953941372036701, + "grad_norm": 0.279296875, + "learning_rate": 0.0019831653075149296, + "loss": 0.1436, + "step": 6859 + }, + { + "epoch": 0.059548094200571174, + "grad_norm": 0.408203125, + "learning_rate": 0.001983159577427265, + "loss": 0.1211, + "step": 6860 + }, + { + "epoch": 0.05955677468077534, + "grad_norm": 0.3359375, + "learning_rate": 0.00198315384637379, + "loss": 0.1641, + "step": 6861 + }, + { + "epoch": 0.059565455160979504, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019831481143545117, + "loss": 0.1289, + "step": 6862 + }, + { + "epoch": 0.05957413564118367, + "grad_norm": 0.380859375, + "learning_rate": 0.0019831423813694354, + "loss": 0.166, + "step": 6863 + }, + { + "epoch": 0.059582816121387834, + "grad_norm": 0.3359375, + "learning_rate": 0.0019831366474185683, + "loss": 0.166, + "step": 6864 + }, + { + "epoch": 0.059591496601592, + "grad_norm": 0.3828125, + "learning_rate": 0.001983130912501916, + "loss": 0.166, + "step": 6865 + }, + { + "epoch": 0.059600177081796164, + "grad_norm": 0.349609375, + "learning_rate": 0.001983125176619485, + "loss": 0.1484, + "step": 6866 + }, + { + "epoch": 0.05960885756200033, + "grad_norm": 0.376953125, + "learning_rate": 0.0019831194397712817, + "loss": 0.1455, + "step": 6867 + }, + { + "epoch": 0.059617538042204495, + "grad_norm": 0.69140625, + "learning_rate": 0.001983113701957312, + "loss": 0.1807, + "step": 6868 + }, + { + "epoch": 0.05962621852240866, + "grad_norm": 0.91015625, + "learning_rate": 0.001983107963177583, + "loss": 0.1426, + "step": 6869 + }, + { + "epoch": 0.059634899002612825, + "grad_norm": 0.271484375, + "learning_rate": 0.0019831022234320997, + "loss": 0.1289, + "step": 6870 + }, + { + "epoch": 0.05964357948281699, + "grad_norm": 0.64453125, + "learning_rate": 0.001983096482720869, + "loss": 0.1221, + "step": 6871 + }, + { + "epoch": 0.059652259963021155, + "grad_norm": 0.28515625, + "learning_rate": 0.001983090741043898, + "loss": 0.1455, + "step": 6872 + }, + { + "epoch": 0.05966094044322532, + "grad_norm": 0.296875, + "learning_rate": 0.001983084998401192, + "loss": 0.0996, + "step": 6873 + }, + { + "epoch": 0.059669620923429485, + "grad_norm": 0.443359375, + "learning_rate": 0.001983079254792757, + "loss": 0.0903, + "step": 6874 + }, + { + "epoch": 0.05967830140363365, + "grad_norm": 0.263671875, + "learning_rate": 0.0019830735102186006, + "loss": 0.126, + "step": 6875 + }, + { + "epoch": 0.059686981883837815, + "grad_norm": 0.12109375, + "learning_rate": 0.0019830677646787277, + "loss": 0.0962, + "step": 6876 + }, + { + "epoch": 0.05969566236404198, + "grad_norm": 0.44140625, + "learning_rate": 0.001983062018173145, + "loss": 0.1699, + "step": 6877 + }, + { + "epoch": 0.059704342844246146, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019830562707018598, + "loss": 0.1455, + "step": 6878 + }, + { + "epoch": 0.05971302332445031, + "grad_norm": 0.26953125, + "learning_rate": 0.001983050522264877, + "loss": 0.1465, + "step": 6879 + }, + { + "epoch": 0.059721703804654476, + "grad_norm": 0.58203125, + "learning_rate": 0.0019830447728622033, + "loss": 0.1245, + "step": 6880 + }, + { + "epoch": 0.05973038428485864, + "grad_norm": 0.125, + "learning_rate": 0.001983039022493845, + "loss": 0.1128, + "step": 6881 + }, + { + "epoch": 0.059739064765062806, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019830332711598085, + "loss": 0.0996, + "step": 6882 + }, + { + "epoch": 0.05974774524526697, + "grad_norm": 1.0625, + "learning_rate": 0.0019830275188601003, + "loss": 0.1123, + "step": 6883 + }, + { + "epoch": 0.059756425725471136, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019830217655947267, + "loss": 0.1133, + "step": 6884 + }, + { + "epoch": 0.059765106205675295, + "grad_norm": 0.3828125, + "learning_rate": 0.0019830160113636936, + "loss": 0.1309, + "step": 6885 + }, + { + "epoch": 0.05977378668587946, + "grad_norm": 1.4375, + "learning_rate": 0.001983010256167007, + "loss": 0.1523, + "step": 6886 + }, + { + "epoch": 0.059782467166083625, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019830045000046744, + "loss": 0.1426, + "step": 6887 + }, + { + "epoch": 0.05979114764628779, + "grad_norm": 0.470703125, + "learning_rate": 0.001982998742876701, + "loss": 0.1104, + "step": 6888 + }, + { + "epoch": 0.059799828126491955, + "grad_norm": 0.09423828125, + "learning_rate": 0.001982992984783093, + "loss": 0.1533, + "step": 6889 + }, + { + "epoch": 0.05980850860669612, + "grad_norm": 0.05908203125, + "learning_rate": 0.001982987225723858, + "loss": 0.1113, + "step": 6890 + }, + { + "epoch": 0.059817189086900285, + "grad_norm": 0.12890625, + "learning_rate": 0.0019829814656990006, + "loss": 0.1592, + "step": 6891 + }, + { + "epoch": 0.05982586956710445, + "grad_norm": 0.4375, + "learning_rate": 0.001982975704708528, + "loss": 0.1777, + "step": 6892 + }, + { + "epoch": 0.059834550047308616, + "grad_norm": 0.08349609375, + "learning_rate": 0.001982969942752447, + "loss": 0.1406, + "step": 6893 + }, + { + "epoch": 0.05984323052751278, + "grad_norm": 1.6875, + "learning_rate": 0.0019829641798307634, + "loss": 0.1895, + "step": 6894 + }, + { + "epoch": 0.059851911007716946, + "grad_norm": 0.58203125, + "learning_rate": 0.001982958415943483, + "loss": 0.1328, + "step": 6895 + }, + { + "epoch": 0.05986059148792111, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019829526510906125, + "loss": 0.1416, + "step": 6896 + }, + { + "epoch": 0.059869271968125276, + "grad_norm": 0.4609375, + "learning_rate": 0.0019829468852721587, + "loss": 0.1641, + "step": 6897 + }, + { + "epoch": 0.05987795244832944, + "grad_norm": 0.60546875, + "learning_rate": 0.001982941118488127, + "loss": 0.126, + "step": 6898 + }, + { + "epoch": 0.059886632928533606, + "grad_norm": 0.26171875, + "learning_rate": 0.0019829353507385245, + "loss": 0.1816, + "step": 6899 + }, + { + "epoch": 0.05989531340873777, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019829295820233566, + "loss": 0.1377, + "step": 6900 + }, + { + "epoch": 0.05990399388894194, + "grad_norm": 0.84765625, + "learning_rate": 0.001982923812342631, + "loss": 0.1758, + "step": 6901 + }, + { + "epoch": 0.0599126743691461, + "grad_norm": 0.68359375, + "learning_rate": 0.0019829180416963527, + "loss": 0.2656, + "step": 6902 + }, + { + "epoch": 0.05992135484935027, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019829122700845284, + "loss": 0.1416, + "step": 6903 + }, + { + "epoch": 0.05993003532955443, + "grad_norm": 0.283203125, + "learning_rate": 0.001982906497507165, + "loss": 0.1162, + "step": 6904 + }, + { + "epoch": 0.0599387158097586, + "grad_norm": 0.5, + "learning_rate": 0.0019829007239642676, + "loss": 0.1865, + "step": 6905 + }, + { + "epoch": 0.05994739628996276, + "grad_norm": 0.27734375, + "learning_rate": 0.001982894949455844, + "loss": 0.1045, + "step": 6906 + }, + { + "epoch": 0.05995607677016693, + "grad_norm": 0.306640625, + "learning_rate": 0.001982889173981899, + "loss": 0.1104, + "step": 6907 + }, + { + "epoch": 0.05996475725037109, + "grad_norm": 0.1787109375, + "learning_rate": 0.00198288339754244, + "loss": 0.1455, + "step": 6908 + }, + { + "epoch": 0.05997343773057526, + "grad_norm": 0.7890625, + "learning_rate": 0.001982877620137473, + "loss": 0.1309, + "step": 6909 + }, + { + "epoch": 0.05998211821077942, + "grad_norm": 0.08837890625, + "learning_rate": 0.001982871841767005, + "loss": 0.1309, + "step": 6910 + }, + { + "epoch": 0.05999079869098359, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019828660624310406, + "loss": 0.1406, + "step": 6911 + }, + { + "epoch": 0.05999947917118775, + "grad_norm": 0.322265625, + "learning_rate": 0.0019828602821295874, + "loss": 0.1465, + "step": 6912 + }, + { + "epoch": 0.06000815965139192, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019828545008626513, + "loss": 0.1299, + "step": 6913 + }, + { + "epoch": 0.06001684013159608, + "grad_norm": 0.29296875, + "learning_rate": 0.0019828487186302392, + "loss": 0.1475, + "step": 6914 + }, + { + "epoch": 0.06002552061180025, + "grad_norm": 0.44921875, + "learning_rate": 0.001982842935432357, + "loss": 0.1182, + "step": 6915 + }, + { + "epoch": 0.060034201092004406, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019828371512690107, + "loss": 0.1631, + "step": 6916 + }, + { + "epoch": 0.06004288157220857, + "grad_norm": 0.703125, + "learning_rate": 0.001982831366140207, + "loss": 0.127, + "step": 6917 + }, + { + "epoch": 0.06005156205241274, + "grad_norm": 0.49609375, + "learning_rate": 0.0019828255800459524, + "loss": 0.5977, + "step": 6918 + }, + { + "epoch": 0.0600602425326169, + "grad_norm": 0.298828125, + "learning_rate": 0.001982819792986253, + "loss": 0.1348, + "step": 6919 + }, + { + "epoch": 0.06006892301282107, + "grad_norm": 0.076171875, + "learning_rate": 0.001982814004961115, + "loss": 0.124, + "step": 6920 + }, + { + "epoch": 0.06007760349302523, + "grad_norm": 0.38671875, + "learning_rate": 0.001982808215970545, + "loss": 0.1611, + "step": 6921 + }, + { + "epoch": 0.0600862839732294, + "grad_norm": 0.154296875, + "learning_rate": 0.001982802426014549, + "loss": 0.1758, + "step": 6922 + }, + { + "epoch": 0.06009496445343356, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019827966350931332, + "loss": 0.124, + "step": 6923 + }, + { + "epoch": 0.06010364493363773, + "grad_norm": 0.59375, + "learning_rate": 0.001982790843206305, + "loss": 0.1602, + "step": 6924 + }, + { + "epoch": 0.06011232541384189, + "grad_norm": 0.69921875, + "learning_rate": 0.00198278505035407, + "loss": 0.1602, + "step": 6925 + }, + { + "epoch": 0.06012100589404606, + "grad_norm": 0.76171875, + "learning_rate": 0.0019827792565364336, + "loss": 0.1738, + "step": 6926 + }, + { + "epoch": 0.06012968637425022, + "grad_norm": 0.474609375, + "learning_rate": 0.001982773461753404, + "loss": 0.1865, + "step": 6927 + }, + { + "epoch": 0.06013836685445439, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019827676660049863, + "loss": 0.1592, + "step": 6928 + }, + { + "epoch": 0.06014704733465855, + "grad_norm": 1.125, + "learning_rate": 0.001982761869291187, + "loss": 0.1133, + "step": 6929 + }, + { + "epoch": 0.06015572781486272, + "grad_norm": 0.65625, + "learning_rate": 0.0019827560716120128, + "loss": 0.1465, + "step": 6930 + }, + { + "epoch": 0.06016440829506688, + "grad_norm": 0.0888671875, + "learning_rate": 0.00198275027296747, + "loss": 0.127, + "step": 6931 + }, + { + "epoch": 0.06017308877527105, + "grad_norm": 0.2890625, + "learning_rate": 0.0019827444733575643, + "loss": 0.1543, + "step": 6932 + }, + { + "epoch": 0.06018176925547521, + "grad_norm": 0.50390625, + "learning_rate": 0.0019827386727823025, + "loss": 0.1172, + "step": 6933 + }, + { + "epoch": 0.06019044973567938, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019827328712416913, + "loss": 0.1289, + "step": 6934 + }, + { + "epoch": 0.060199130215883544, + "grad_norm": 0.1611328125, + "learning_rate": 0.001982727068735737, + "loss": 0.1094, + "step": 6935 + }, + { + "epoch": 0.06020781069608771, + "grad_norm": 0.322265625, + "learning_rate": 0.0019827212652644447, + "loss": 0.1279, + "step": 6936 + }, + { + "epoch": 0.060216491176291874, + "grad_norm": 0.50390625, + "learning_rate": 0.0019827154608278224, + "loss": 0.1221, + "step": 6937 + }, + { + "epoch": 0.06022517165649604, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019827096554258755, + "loss": 0.1367, + "step": 6938 + }, + { + "epoch": 0.060233852136700204, + "grad_norm": 0.515625, + "learning_rate": 0.0019827038490586105, + "loss": 0.1147, + "step": 6939 + }, + { + "epoch": 0.06024253261690437, + "grad_norm": 0.140625, + "learning_rate": 0.0019826980417260343, + "loss": 0.1406, + "step": 6940 + }, + { + "epoch": 0.060251213097108534, + "grad_norm": 0.6015625, + "learning_rate": 0.001982692233428152, + "loss": 0.1357, + "step": 6941 + }, + { + "epoch": 0.0602598935773127, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019826864241649714, + "loss": 0.1211, + "step": 6942 + }, + { + "epoch": 0.060268574057516865, + "grad_norm": 0.345703125, + "learning_rate": 0.001982680613936498, + "loss": 0.1143, + "step": 6943 + }, + { + "epoch": 0.06027725453772103, + "grad_norm": 0.216796875, + "learning_rate": 0.0019826748027427384, + "loss": 0.1738, + "step": 6944 + }, + { + "epoch": 0.060285935017925195, + "grad_norm": 0.462890625, + "learning_rate": 0.0019826689905836988, + "loss": 0.1055, + "step": 6945 + }, + { + "epoch": 0.06029461549812935, + "grad_norm": 0.166015625, + "learning_rate": 0.0019826631774593857, + "loss": 0.1582, + "step": 6946 + }, + { + "epoch": 0.06030329597833352, + "grad_norm": 0.8515625, + "learning_rate": 0.0019826573633698052, + "loss": 0.1084, + "step": 6947 + }, + { + "epoch": 0.06031197645853768, + "grad_norm": 0.205078125, + "learning_rate": 0.001982651548314964, + "loss": 0.1572, + "step": 6948 + }, + { + "epoch": 0.06032065693874185, + "grad_norm": 0.240234375, + "learning_rate": 0.0019826457322948683, + "loss": 0.2051, + "step": 6949 + }, + { + "epoch": 0.060329337418946014, + "grad_norm": 0.80078125, + "learning_rate": 0.0019826399153095244, + "loss": 0.1143, + "step": 6950 + }, + { + "epoch": 0.06033801789915018, + "grad_norm": 0.58984375, + "learning_rate": 0.001982634097358939, + "loss": 0.1758, + "step": 6951 + }, + { + "epoch": 0.060346698379354344, + "grad_norm": 1.2109375, + "learning_rate": 0.001982628278443118, + "loss": 0.1045, + "step": 6952 + }, + { + "epoch": 0.06035537885955851, + "grad_norm": 0.291015625, + "learning_rate": 0.0019826224585620684, + "loss": 0.1387, + "step": 6953 + }, + { + "epoch": 0.060364059339762674, + "grad_norm": 0.142578125, + "learning_rate": 0.001982616637715796, + "loss": 0.1191, + "step": 6954 + }, + { + "epoch": 0.06037273981996684, + "grad_norm": 0.07177734375, + "learning_rate": 0.001982610815904307, + "loss": 0.1162, + "step": 6955 + }, + { + "epoch": 0.060381420300171004, + "grad_norm": 0.150390625, + "learning_rate": 0.001982604993127608, + "loss": 0.1289, + "step": 6956 + }, + { + "epoch": 0.06039010078037517, + "grad_norm": 0.3203125, + "learning_rate": 0.001982599169385706, + "loss": 0.165, + "step": 6957 + }, + { + "epoch": 0.060398781260579334, + "grad_norm": 0.150390625, + "learning_rate": 0.0019825933446786063, + "loss": 0.1699, + "step": 6958 + }, + { + "epoch": 0.0604074617407835, + "grad_norm": 0.07958984375, + "learning_rate": 0.001982587519006316, + "loss": 0.1523, + "step": 6959 + }, + { + "epoch": 0.060416142220987665, + "grad_norm": 0.1357421875, + "learning_rate": 0.001982581692368841, + "loss": 0.1543, + "step": 6960 + }, + { + "epoch": 0.06042482270119183, + "grad_norm": 0.279296875, + "learning_rate": 0.0019825758647661882, + "loss": 0.1689, + "step": 6961 + }, + { + "epoch": 0.060433503181395995, + "grad_norm": 0.9921875, + "learning_rate": 0.0019825700361983637, + "loss": 0.1465, + "step": 6962 + }, + { + "epoch": 0.06044218366160016, + "grad_norm": 0.921875, + "learning_rate": 0.0019825642066653736, + "loss": 0.1152, + "step": 6963 + }, + { + "epoch": 0.060450864141804325, + "grad_norm": 0.302734375, + "learning_rate": 0.0019825583761672247, + "loss": 0.1504, + "step": 6964 + }, + { + "epoch": 0.06045954462200849, + "grad_norm": 0.234375, + "learning_rate": 0.0019825525447039233, + "loss": 0.0996, + "step": 6965 + }, + { + "epoch": 0.060468225102212655, + "grad_norm": 0.16796875, + "learning_rate": 0.0019825467122754755, + "loss": 0.1069, + "step": 6966 + }, + { + "epoch": 0.06047690558241682, + "grad_norm": 0.353515625, + "learning_rate": 0.001982540878881888, + "loss": 0.1094, + "step": 6967 + }, + { + "epoch": 0.060485586062620986, + "grad_norm": 0.2578125, + "learning_rate": 0.001982535044523167, + "loss": 0.1328, + "step": 6968 + }, + { + "epoch": 0.06049426654282515, + "grad_norm": 0.78125, + "learning_rate": 0.001982529209199319, + "loss": 0.1934, + "step": 6969 + }, + { + "epoch": 0.060502947023029316, + "grad_norm": 0.8046875, + "learning_rate": 0.0019825233729103505, + "loss": 0.1206, + "step": 6970 + }, + { + "epoch": 0.06051162750323348, + "grad_norm": 0.53515625, + "learning_rate": 0.0019825175356562677, + "loss": 0.1309, + "step": 6971 + }, + { + "epoch": 0.060520307983437646, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019825116974370766, + "loss": 0.168, + "step": 6972 + }, + { + "epoch": 0.06052898846364181, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019825058582527842, + "loss": 0.1523, + "step": 6973 + }, + { + "epoch": 0.060537668943845976, + "grad_norm": 0.283203125, + "learning_rate": 0.0019825000181033967, + "loss": 0.1445, + "step": 6974 + }, + { + "epoch": 0.06054634942405014, + "grad_norm": 0.296875, + "learning_rate": 0.0019824941769889205, + "loss": 0.1416, + "step": 6975 + }, + { + "epoch": 0.06055502990425431, + "grad_norm": 0.79296875, + "learning_rate": 0.0019824883349093612, + "loss": 0.1201, + "step": 6976 + }, + { + "epoch": 0.060563710384458465, + "grad_norm": 0.62890625, + "learning_rate": 0.0019824824918647267, + "loss": 0.1504, + "step": 6977 + }, + { + "epoch": 0.06057239086466263, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019824766478550226, + "loss": 0.1172, + "step": 6978 + }, + { + "epoch": 0.060581071344866795, + "grad_norm": 0.1416015625, + "learning_rate": 0.001982470802880255, + "loss": 0.1426, + "step": 6979 + }, + { + "epoch": 0.06058975182507096, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019824649569404307, + "loss": 0.1177, + "step": 6980 + }, + { + "epoch": 0.060598432305275125, + "grad_norm": 0.1181640625, + "learning_rate": 0.001982459110035556, + "loss": 0.1406, + "step": 6981 + }, + { + "epoch": 0.06060711278547929, + "grad_norm": 0.98828125, + "learning_rate": 0.0019824532621656376, + "loss": 0.1455, + "step": 6982 + }, + { + "epoch": 0.060615793265683456, + "grad_norm": 0.115234375, + "learning_rate": 0.0019824474133306813, + "loss": 0.1377, + "step": 6983 + }, + { + "epoch": 0.06062447374588762, + "grad_norm": 0.482421875, + "learning_rate": 0.001982441563530693, + "loss": 0.1514, + "step": 6984 + }, + { + "epoch": 0.060633154226091786, + "grad_norm": 0.1572265625, + "learning_rate": 0.001982435712765681, + "loss": 0.1147, + "step": 6985 + }, + { + "epoch": 0.06064183470629595, + "grad_norm": 0.37109375, + "learning_rate": 0.00198242986103565, + "loss": 0.1562, + "step": 6986 + }, + { + "epoch": 0.060650515186500116, + "grad_norm": 0.302734375, + "learning_rate": 0.0019824240083406073, + "loss": 0.1699, + "step": 6987 + }, + { + "epoch": 0.06065919566670428, + "grad_norm": 0.146484375, + "learning_rate": 0.0019824181546805584, + "loss": 0.123, + "step": 6988 + }, + { + "epoch": 0.060667876146908446, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019824123000555106, + "loss": 0.1387, + "step": 6989 + }, + { + "epoch": 0.06067655662711261, + "grad_norm": 0.80859375, + "learning_rate": 0.00198240644446547, + "loss": 0.1553, + "step": 6990 + }, + { + "epoch": 0.060685237107316777, + "grad_norm": 0.74609375, + "learning_rate": 0.001982400587910443, + "loss": 0.1191, + "step": 6991 + }, + { + "epoch": 0.06069391758752094, + "grad_norm": 0.578125, + "learning_rate": 0.001982394730390436, + "loss": 0.1445, + "step": 6992 + }, + { + "epoch": 0.06070259806772511, + "grad_norm": 0.50390625, + "learning_rate": 0.0019823888719054552, + "loss": 0.1387, + "step": 6993 + }, + { + "epoch": 0.06071127854792927, + "grad_norm": 0.408203125, + "learning_rate": 0.0019823830124555074, + "loss": 0.1357, + "step": 6994 + }, + { + "epoch": 0.06071995902813344, + "grad_norm": 0.248046875, + "learning_rate": 0.0019823771520405984, + "loss": 0.1289, + "step": 6995 + }, + { + "epoch": 0.0607286395083376, + "grad_norm": 0.50390625, + "learning_rate": 0.0019823712906607356, + "loss": 0.1582, + "step": 6996 + }, + { + "epoch": 0.06073731998854177, + "grad_norm": 0.09375, + "learning_rate": 0.001982365428315924, + "loss": 0.1128, + "step": 6997 + }, + { + "epoch": 0.06074600046874593, + "grad_norm": 0.0703125, + "learning_rate": 0.001982359565006172, + "loss": 0.1445, + "step": 6998 + }, + { + "epoch": 0.0607546809489501, + "grad_norm": 0.1484375, + "learning_rate": 0.001982353700731484, + "loss": 0.1377, + "step": 6999 + }, + { + "epoch": 0.06076336142915426, + "grad_norm": 0.1328125, + "learning_rate": 0.0019823478354918673, + "loss": 0.1504, + "step": 7000 + }, + { + "epoch": 0.06077204190935843, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019823419692873282, + "loss": 0.1611, + "step": 7001 + }, + { + "epoch": 0.06078072238956259, + "grad_norm": 0.5859375, + "learning_rate": 0.0019823361021178735, + "loss": 0.1426, + "step": 7002 + }, + { + "epoch": 0.06078940286976676, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019823302339835092, + "loss": 0.1084, + "step": 7003 + }, + { + "epoch": 0.06079808334997092, + "grad_norm": 0.75, + "learning_rate": 0.001982324364884242, + "loss": 0.1338, + "step": 7004 + }, + { + "epoch": 0.06080676383017509, + "grad_norm": 0.1357421875, + "learning_rate": 0.001982318494820078, + "loss": 0.1221, + "step": 7005 + }, + { + "epoch": 0.06081544431037925, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019823126237910236, + "loss": 0.166, + "step": 7006 + }, + { + "epoch": 0.06082412479058342, + "grad_norm": 0.083984375, + "learning_rate": 0.001982306751797085, + "loss": 0.1836, + "step": 7007 + }, + { + "epoch": 0.06083280527078758, + "grad_norm": 0.341796875, + "learning_rate": 0.0019823008788382697, + "loss": 0.127, + "step": 7008 + }, + { + "epoch": 0.06084148575099174, + "grad_norm": 0.232421875, + "learning_rate": 0.0019822950049145833, + "loss": 0.1504, + "step": 7009 + }, + { + "epoch": 0.06085016623119591, + "grad_norm": 0.2275390625, + "learning_rate": 0.001982289130026032, + "loss": 0.1221, + "step": 7010 + }, + { + "epoch": 0.06085884671140007, + "grad_norm": 1.4921875, + "learning_rate": 0.0019822832541726227, + "loss": 0.168, + "step": 7011 + }, + { + "epoch": 0.06086752719160424, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019822773773543615, + "loss": 0.1104, + "step": 7012 + }, + { + "epoch": 0.0608762076718084, + "grad_norm": 0.82421875, + "learning_rate": 0.0019822714995712553, + "loss": 0.127, + "step": 7013 + }, + { + "epoch": 0.06088488815201257, + "grad_norm": 0.48046875, + "learning_rate": 0.0019822656208233103, + "loss": 0.1514, + "step": 7014 + }, + { + "epoch": 0.06089356863221673, + "grad_norm": 0.1826171875, + "learning_rate": 0.001982259741110533, + "loss": 0.1387, + "step": 7015 + }, + { + "epoch": 0.0609022491124209, + "grad_norm": 0.291015625, + "learning_rate": 0.001982253860432929, + "loss": 0.1689, + "step": 7016 + }, + { + "epoch": 0.06091092959262506, + "grad_norm": 0.177734375, + "learning_rate": 0.001982247978790506, + "loss": 0.1738, + "step": 7017 + }, + { + "epoch": 0.06091961007282923, + "grad_norm": 0.232421875, + "learning_rate": 0.0019822420961832696, + "loss": 0.1689, + "step": 7018 + }, + { + "epoch": 0.06092829055303339, + "grad_norm": 0.11572265625, + "learning_rate": 0.001982236212611227, + "loss": 0.1523, + "step": 7019 + }, + { + "epoch": 0.06093697103323756, + "grad_norm": 0.119140625, + "learning_rate": 0.001982230328074384, + "loss": 0.1836, + "step": 7020 + }, + { + "epoch": 0.06094565151344172, + "grad_norm": 0.3828125, + "learning_rate": 0.0019822244425727466, + "loss": 0.1416, + "step": 7021 + }, + { + "epoch": 0.06095433199364589, + "grad_norm": 0.1611328125, + "learning_rate": 0.001982218556106322, + "loss": 0.1553, + "step": 7022 + }, + { + "epoch": 0.06096301247385005, + "grad_norm": 0.345703125, + "learning_rate": 0.001982212668675117, + "loss": 0.1377, + "step": 7023 + }, + { + "epoch": 0.06097169295405422, + "grad_norm": 0.484375, + "learning_rate": 0.001982206780279137, + "loss": 0.1455, + "step": 7024 + }, + { + "epoch": 0.060980373434258384, + "grad_norm": 0.64453125, + "learning_rate": 0.001982200890918389, + "loss": 0.1348, + "step": 7025 + }, + { + "epoch": 0.06098905391446255, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019821950005928794, + "loss": 0.166, + "step": 7026 + }, + { + "epoch": 0.060997734394666714, + "grad_norm": 0.18359375, + "learning_rate": 0.0019821891093026144, + "loss": 0.0874, + "step": 7027 + }, + { + "epoch": 0.06100641487487088, + "grad_norm": 0.255859375, + "learning_rate": 0.001982183217047601, + "loss": 0.1426, + "step": 7028 + }, + { + "epoch": 0.061015095355075044, + "grad_norm": 0.6484375, + "learning_rate": 0.001982177323827845, + "loss": 0.1133, + "step": 7029 + }, + { + "epoch": 0.06102377583527921, + "grad_norm": 0.154296875, + "learning_rate": 0.0019821714296433535, + "loss": 0.1699, + "step": 7030 + }, + { + "epoch": 0.061032456315483374, + "grad_norm": 0.31640625, + "learning_rate": 0.0019821655344941323, + "loss": 0.1523, + "step": 7031 + }, + { + "epoch": 0.06104113679568754, + "grad_norm": 0.1640625, + "learning_rate": 0.001982159638380188, + "loss": 0.1367, + "step": 7032 + }, + { + "epoch": 0.061049817275891705, + "grad_norm": 0.197265625, + "learning_rate": 0.001982153741301527, + "loss": 0.1582, + "step": 7033 + }, + { + "epoch": 0.06105849775609587, + "grad_norm": 0.091796875, + "learning_rate": 0.0019821478432581565, + "loss": 0.1777, + "step": 7034 + }, + { + "epoch": 0.061067178236300035, + "grad_norm": 0.57421875, + "learning_rate": 0.001982141944250082, + "loss": 0.1465, + "step": 7035 + }, + { + "epoch": 0.0610758587165042, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019821360442773103, + "loss": 0.1318, + "step": 7036 + }, + { + "epoch": 0.061084539196708365, + "grad_norm": 0.09765625, + "learning_rate": 0.001982130143339848, + "loss": 0.1182, + "step": 7037 + }, + { + "epoch": 0.06109321967691253, + "grad_norm": 0.255859375, + "learning_rate": 0.0019821242414377015, + "loss": 0.1191, + "step": 7038 + }, + { + "epoch": 0.06110190015711669, + "grad_norm": 0.392578125, + "learning_rate": 0.0019821183385708772, + "loss": 0.1592, + "step": 7039 + }, + { + "epoch": 0.061110580637320853, + "grad_norm": 0.103515625, + "learning_rate": 0.001982112434739382, + "loss": 0.1318, + "step": 7040 + }, + { + "epoch": 0.06111926111752502, + "grad_norm": 0.439453125, + "learning_rate": 0.001982106529943221, + "loss": 0.208, + "step": 7041 + }, + { + "epoch": 0.061127941597729184, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019821006241824016, + "loss": 0.1011, + "step": 7042 + }, + { + "epoch": 0.06113662207793335, + "grad_norm": 0.1943359375, + "learning_rate": 0.001982094717456931, + "loss": 0.1348, + "step": 7043 + }, + { + "epoch": 0.061145302558137514, + "grad_norm": 0.302734375, + "learning_rate": 0.0019820888097668143, + "loss": 0.168, + "step": 7044 + }, + { + "epoch": 0.06115398303834168, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019820829011120583, + "loss": 0.1108, + "step": 7045 + }, + { + "epoch": 0.061162663518545844, + "grad_norm": 0.43359375, + "learning_rate": 0.0019820769914926702, + "loss": 0.126, + "step": 7046 + }, + { + "epoch": 0.06117134399875001, + "grad_norm": 0.28515625, + "learning_rate": 0.001982071080908656, + "loss": 0.1299, + "step": 7047 + }, + { + "epoch": 0.061180024478954174, + "grad_norm": 0.27734375, + "learning_rate": 0.0019820651693600215, + "loss": 0.1523, + "step": 7048 + }, + { + "epoch": 0.06118870495915834, + "grad_norm": 0.275390625, + "learning_rate": 0.0019820592568467743, + "loss": 0.1719, + "step": 7049 + }, + { + "epoch": 0.061197385439362505, + "grad_norm": 0.171875, + "learning_rate": 0.00198205334336892, + "loss": 0.1777, + "step": 7050 + }, + { + "epoch": 0.06120606591956667, + "grad_norm": 0.671875, + "learning_rate": 0.0019820474289264658, + "loss": 0.2012, + "step": 7051 + }, + { + "epoch": 0.061214746399770835, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019820415135194175, + "loss": 0.1279, + "step": 7052 + }, + { + "epoch": 0.061223426879975, + "grad_norm": 0.318359375, + "learning_rate": 0.0019820355971477822, + "loss": 0.1191, + "step": 7053 + }, + { + "epoch": 0.061232107360179165, + "grad_norm": 0.3984375, + "learning_rate": 0.0019820296798115657, + "loss": 0.1074, + "step": 7054 + }, + { + "epoch": 0.06124078784038333, + "grad_norm": 0.5625, + "learning_rate": 0.001982023761510775, + "loss": 0.2227, + "step": 7055 + }, + { + "epoch": 0.061249468320587495, + "grad_norm": 0.0859375, + "learning_rate": 0.001982017842245416, + "loss": 0.126, + "step": 7056 + }, + { + "epoch": 0.06125814880079166, + "grad_norm": 0.1337890625, + "learning_rate": 0.001982011922015496, + "loss": 0.1074, + "step": 7057 + }, + { + "epoch": 0.061266829280995826, + "grad_norm": 0.26953125, + "learning_rate": 0.001982006000821021, + "loss": 0.1279, + "step": 7058 + }, + { + "epoch": 0.06127550976119999, + "grad_norm": 0.328125, + "learning_rate": 0.0019820000786619973, + "loss": 0.1494, + "step": 7059 + }, + { + "epoch": 0.061284190241404156, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019819941555384315, + "loss": 0.1602, + "step": 7060 + }, + { + "epoch": 0.06129287072160832, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019819882314503303, + "loss": 0.1211, + "step": 7061 + }, + { + "epoch": 0.061301551201812486, + "grad_norm": 0.2734375, + "learning_rate": 0.0019819823063977, + "loss": 0.1641, + "step": 7062 + }, + { + "epoch": 0.06131023168201665, + "grad_norm": 0.1630859375, + "learning_rate": 0.001981976380380547, + "loss": 0.1045, + "step": 7063 + }, + { + "epoch": 0.061318912162220816, + "grad_norm": 0.30859375, + "learning_rate": 0.001981970453398878, + "loss": 0.1494, + "step": 7064 + }, + { + "epoch": 0.06132759264242498, + "grad_norm": 0.5390625, + "learning_rate": 0.0019819645254526993, + "loss": 0.1484, + "step": 7065 + }, + { + "epoch": 0.06133627312262915, + "grad_norm": 0.5, + "learning_rate": 0.001981958596542017, + "loss": 0.1445, + "step": 7066 + }, + { + "epoch": 0.06134495360283331, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019819526666668384, + "loss": 0.127, + "step": 7067 + }, + { + "epoch": 0.06135363408303748, + "grad_norm": 0.23828125, + "learning_rate": 0.00198194673582717, + "loss": 0.1553, + "step": 7068 + }, + { + "epoch": 0.06136231456324164, + "grad_norm": 0.171875, + "learning_rate": 0.001981940804023017, + "loss": 0.1387, + "step": 7069 + }, + { + "epoch": 0.0613709950434458, + "grad_norm": 0.396484375, + "learning_rate": 0.001981934871254388, + "loss": 0.125, + "step": 7070 + }, + { + "epoch": 0.061379675523649965, + "grad_norm": 0.1875, + "learning_rate": 0.0019819289375212876, + "loss": 0.0996, + "step": 7071 + }, + { + "epoch": 0.06138835600385413, + "grad_norm": 0.12255859375, + "learning_rate": 0.001981923002823723, + "loss": 0.1445, + "step": 7072 + }, + { + "epoch": 0.061397036484058296, + "grad_norm": 1.03125, + "learning_rate": 0.0019819170671617004, + "loss": 0.1484, + "step": 7073 + }, + { + "epoch": 0.06140571696426246, + "grad_norm": 0.27734375, + "learning_rate": 0.001981911130535227, + "loss": 0.1689, + "step": 7074 + }, + { + "epoch": 0.061414397444466626, + "grad_norm": 0.146484375, + "learning_rate": 0.001981905192944309, + "loss": 0.1289, + "step": 7075 + }, + { + "epoch": 0.06142307792467079, + "grad_norm": 0.30078125, + "learning_rate": 0.001981899254388952, + "loss": 0.1396, + "step": 7076 + }, + { + "epoch": 0.061431758404874956, + "grad_norm": 0.0712890625, + "learning_rate": 0.0019818933148691635, + "loss": 0.1289, + "step": 7077 + }, + { + "epoch": 0.06144043888507912, + "grad_norm": 0.1513671875, + "learning_rate": 0.00198188737438495, + "loss": 0.1348, + "step": 7078 + }, + { + "epoch": 0.061449119365283286, + "grad_norm": 0.2421875, + "learning_rate": 0.001981881432936318, + "loss": 0.1602, + "step": 7079 + }, + { + "epoch": 0.06145779984548745, + "grad_norm": 0.341796875, + "learning_rate": 0.0019818754905232727, + "loss": 0.1084, + "step": 7080 + }, + { + "epoch": 0.061466480325691616, + "grad_norm": 0.2275390625, + "learning_rate": 0.001981869547145822, + "loss": 0.1133, + "step": 7081 + }, + { + "epoch": 0.06147516080589578, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019818636028039725, + "loss": 0.1582, + "step": 7082 + }, + { + "epoch": 0.06148384128609995, + "grad_norm": 0.244140625, + "learning_rate": 0.00198185765749773, + "loss": 0.1191, + "step": 7083 + }, + { + "epoch": 0.06149252176630411, + "grad_norm": 0.73046875, + "learning_rate": 0.0019818517112271013, + "loss": 0.1206, + "step": 7084 + }, + { + "epoch": 0.06150120224650828, + "grad_norm": 0.1884765625, + "learning_rate": 0.001981845763992092, + "loss": 0.1309, + "step": 7085 + }, + { + "epoch": 0.06150988272671244, + "grad_norm": 0.412109375, + "learning_rate": 0.0019818398157927103, + "loss": 0.1514, + "step": 7086 + }, + { + "epoch": 0.06151856320691661, + "grad_norm": 0.287109375, + "learning_rate": 0.0019818338666289614, + "loss": 0.1211, + "step": 7087 + }, + { + "epoch": 0.06152724368712077, + "grad_norm": 0.6796875, + "learning_rate": 0.0019818279165008524, + "loss": 0.1514, + "step": 7088 + }, + { + "epoch": 0.06153592416732494, + "grad_norm": 0.33984375, + "learning_rate": 0.00198182196540839, + "loss": 0.125, + "step": 7089 + }, + { + "epoch": 0.0615446046475291, + "grad_norm": 0.2734375, + "learning_rate": 0.00198181601335158, + "loss": 0.0913, + "step": 7090 + }, + { + "epoch": 0.06155328512773327, + "grad_norm": 0.1201171875, + "learning_rate": 0.001981810060330429, + "loss": 0.1318, + "step": 7091 + }, + { + "epoch": 0.06156196560793743, + "grad_norm": 0.09130859375, + "learning_rate": 0.001981804106344944, + "loss": 0.127, + "step": 7092 + }, + { + "epoch": 0.0615706460881416, + "grad_norm": 0.375, + "learning_rate": 0.0019817981513951316, + "loss": 0.1172, + "step": 7093 + }, + { + "epoch": 0.06157932656834576, + "grad_norm": 0.13671875, + "learning_rate": 0.0019817921954809977, + "loss": 0.1416, + "step": 7094 + }, + { + "epoch": 0.06158800704854993, + "grad_norm": 0.60546875, + "learning_rate": 0.001981786238602549, + "loss": 0.1562, + "step": 7095 + }, + { + "epoch": 0.06159668752875409, + "grad_norm": 0.50390625, + "learning_rate": 0.001981780280759792, + "loss": 0.1147, + "step": 7096 + }, + { + "epoch": 0.06160536800895826, + "grad_norm": 0.126953125, + "learning_rate": 0.001981774321952734, + "loss": 0.1406, + "step": 7097 + }, + { + "epoch": 0.061614048489162423, + "grad_norm": 0.76953125, + "learning_rate": 0.0019817683621813804, + "loss": 0.1543, + "step": 7098 + }, + { + "epoch": 0.06162272896936659, + "grad_norm": 0.2041015625, + "learning_rate": 0.001981762401445738, + "loss": 0.1504, + "step": 7099 + }, + { + "epoch": 0.06163140944957075, + "grad_norm": 0.6328125, + "learning_rate": 0.0019817564397458134, + "loss": 0.1289, + "step": 7100 + }, + { + "epoch": 0.06164008992977491, + "grad_norm": 0.357421875, + "learning_rate": 0.0019817504770816134, + "loss": 0.1162, + "step": 7101 + }, + { + "epoch": 0.06164877040997908, + "grad_norm": 0.294921875, + "learning_rate": 0.0019817445134531443, + "loss": 0.123, + "step": 7102 + }, + { + "epoch": 0.06165745089018324, + "grad_norm": 0.67578125, + "learning_rate": 0.001981738548860413, + "loss": 0.1377, + "step": 7103 + }, + { + "epoch": 0.06166613137038741, + "grad_norm": 0.421875, + "learning_rate": 0.0019817325833034255, + "loss": 0.1367, + "step": 7104 + }, + { + "epoch": 0.06167481185059157, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019817266167821883, + "loss": 0.1426, + "step": 7105 + }, + { + "epoch": 0.06168349233079574, + "grad_norm": 0.259765625, + "learning_rate": 0.001981720649296708, + "loss": 0.1582, + "step": 7106 + }, + { + "epoch": 0.0616921728109999, + "grad_norm": 0.474609375, + "learning_rate": 0.0019817146808469915, + "loss": 0.1279, + "step": 7107 + }, + { + "epoch": 0.06170085329120407, + "grad_norm": 0.181640625, + "learning_rate": 0.0019817087114330453, + "loss": 0.1797, + "step": 7108 + }, + { + "epoch": 0.06170953377140823, + "grad_norm": 0.8125, + "learning_rate": 0.0019817027410548756, + "loss": 0.2168, + "step": 7109 + }, + { + "epoch": 0.0617182142516124, + "grad_norm": 0.119140625, + "learning_rate": 0.0019816967697124884, + "loss": 0.0991, + "step": 7110 + }, + { + "epoch": 0.06172689473181656, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019816907974058917, + "loss": 0.1094, + "step": 7111 + }, + { + "epoch": 0.06173557521202073, + "grad_norm": 1.0, + "learning_rate": 0.0019816848241350905, + "loss": 0.1758, + "step": 7112 + }, + { + "epoch": 0.06174425569222489, + "grad_norm": 1.4921875, + "learning_rate": 0.0019816788499000927, + "loss": 0.1621, + "step": 7113 + }, + { + "epoch": 0.06175293617242906, + "grad_norm": 0.298828125, + "learning_rate": 0.0019816728747009036, + "loss": 0.1187, + "step": 7114 + }, + { + "epoch": 0.061761616652633224, + "grad_norm": 0.33203125, + "learning_rate": 0.0019816668985375304, + "loss": 0.1514, + "step": 7115 + }, + { + "epoch": 0.06177029713283739, + "grad_norm": 0.279296875, + "learning_rate": 0.0019816609214099797, + "loss": 0.1089, + "step": 7116 + }, + { + "epoch": 0.061778977613041554, + "grad_norm": 0.64453125, + "learning_rate": 0.001981654943318258, + "loss": 0.1426, + "step": 7117 + }, + { + "epoch": 0.06178765809324572, + "grad_norm": 0.421875, + "learning_rate": 0.0019816489642623713, + "loss": 0.1348, + "step": 7118 + }, + { + "epoch": 0.061796338573449884, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019816429842423266, + "loss": 0.167, + "step": 7119 + }, + { + "epoch": 0.06180501905365405, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019816370032581304, + "loss": 0.1084, + "step": 7120 + }, + { + "epoch": 0.061813699533858214, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019816310213097896, + "loss": 0.1064, + "step": 7121 + }, + { + "epoch": 0.06182238001406238, + "grad_norm": 0.087890625, + "learning_rate": 0.00198162503839731, + "loss": 0.1426, + "step": 7122 + }, + { + "epoch": 0.061831060494266545, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019816190545206987, + "loss": 0.1299, + "step": 7123 + }, + { + "epoch": 0.06183974097447071, + "grad_norm": 0.12353515625, + "learning_rate": 0.001981613069679962, + "loss": 0.1206, + "step": 7124 + }, + { + "epoch": 0.061848421454674875, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019816070838751063, + "loss": 0.1777, + "step": 7125 + }, + { + "epoch": 0.06185710193487904, + "grad_norm": 0.421875, + "learning_rate": 0.0019816010971061386, + "loss": 0.1089, + "step": 7126 + }, + { + "epoch": 0.061865782415083205, + "grad_norm": 0.74609375, + "learning_rate": 0.001981595109373065, + "loss": 0.1348, + "step": 7127 + }, + { + "epoch": 0.06187446289528737, + "grad_norm": 0.67578125, + "learning_rate": 0.001981589120675892, + "loss": 0.1011, + "step": 7128 + }, + { + "epoch": 0.061883143375491535, + "grad_norm": 0.224609375, + "learning_rate": 0.001981583131014627, + "loss": 0.1309, + "step": 7129 + }, + { + "epoch": 0.0618918238556957, + "grad_norm": 0.515625, + "learning_rate": 0.0019815771403892757, + "loss": 0.1445, + "step": 7130 + }, + { + "epoch": 0.06190050433589986, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019815711487998445, + "loss": 0.1357, + "step": 7131 + }, + { + "epoch": 0.061909184816104024, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019815651562463408, + "loss": 0.168, + "step": 7132 + }, + { + "epoch": 0.06191786529630819, + "grad_norm": 0.310546875, + "learning_rate": 0.0019815591627287705, + "loss": 0.1572, + "step": 7133 + }, + { + "epoch": 0.061926545776512354, + "grad_norm": 0.63671875, + "learning_rate": 0.0019815531682471402, + "loss": 0.1758, + "step": 7134 + }, + { + "epoch": 0.06193522625671652, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019815471728014565, + "loss": 0.1221, + "step": 7135 + }, + { + "epoch": 0.061943906736920684, + "grad_norm": 0.248046875, + "learning_rate": 0.0019815411763917266, + "loss": 0.1787, + "step": 7136 + }, + { + "epoch": 0.06195258721712485, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019815351790179563, + "loss": 0.125, + "step": 7137 + }, + { + "epoch": 0.061961267697329014, + "grad_norm": 0.416015625, + "learning_rate": 0.001981529180680152, + "loss": 0.1475, + "step": 7138 + }, + { + "epoch": 0.06196994817753318, + "grad_norm": 0.08544921875, + "learning_rate": 0.001981523181378321, + "loss": 0.1338, + "step": 7139 + }, + { + "epoch": 0.061978628657737345, + "grad_norm": 0.62890625, + "learning_rate": 0.0019815171811124696, + "loss": 0.1553, + "step": 7140 + }, + { + "epoch": 0.06198730913794151, + "grad_norm": 0.1494140625, + "learning_rate": 0.001981511179882604, + "loss": 0.1465, + "step": 7141 + }, + { + "epoch": 0.061995989618145675, + "grad_norm": 0.453125, + "learning_rate": 0.0019815051776887315, + "loss": 0.1396, + "step": 7142 + }, + { + "epoch": 0.06200467009834984, + "grad_norm": 0.0859375, + "learning_rate": 0.0019814991745308575, + "loss": 0.1045, + "step": 7143 + }, + { + "epoch": 0.062013350578554005, + "grad_norm": 0.3203125, + "learning_rate": 0.0019814931704089894, + "loss": 0.1689, + "step": 7144 + }, + { + "epoch": 0.06202203105875817, + "grad_norm": 0.287109375, + "learning_rate": 0.0019814871653231337, + "loss": 0.1226, + "step": 7145 + }, + { + "epoch": 0.062030711538962335, + "grad_norm": 0.90234375, + "learning_rate": 0.001981481159273297, + "loss": 0.1484, + "step": 7146 + }, + { + "epoch": 0.0620393920191665, + "grad_norm": 0.341796875, + "learning_rate": 0.001981475152259486, + "loss": 0.1016, + "step": 7147 + }, + { + "epoch": 0.062048072499370666, + "grad_norm": 0.82421875, + "learning_rate": 0.0019814691442817064, + "loss": 0.1484, + "step": 7148 + }, + { + "epoch": 0.06205675297957483, + "grad_norm": 0.255859375, + "learning_rate": 0.0019814631353399655, + "loss": 0.1562, + "step": 7149 + }, + { + "epoch": 0.062065433459778996, + "grad_norm": 0.09228515625, + "learning_rate": 0.00198145712543427, + "loss": 0.1357, + "step": 7150 + }, + { + "epoch": 0.06207411393998316, + "grad_norm": 0.26171875, + "learning_rate": 0.0019814511145646264, + "loss": 0.1514, + "step": 7151 + }, + { + "epoch": 0.062082794420187326, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019814451027310408, + "loss": 0.1992, + "step": 7152 + }, + { + "epoch": 0.06209147490039149, + "grad_norm": 0.1533203125, + "learning_rate": 0.00198143908993352, + "loss": 0.1309, + "step": 7153 + }, + { + "epoch": 0.062100155380595656, + "grad_norm": 0.10693359375, + "learning_rate": 0.001981433076172071, + "loss": 0.1123, + "step": 7154 + }, + { + "epoch": 0.06210883586079982, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019814270614467, + "loss": 0.1855, + "step": 7155 + }, + { + "epoch": 0.06211751634100399, + "grad_norm": 0.7734375, + "learning_rate": 0.001981421045757413, + "loss": 0.1367, + "step": 7156 + }, + { + "epoch": 0.06212619682120815, + "grad_norm": 0.341796875, + "learning_rate": 0.001981415029104218, + "loss": 0.1406, + "step": 7157 + }, + { + "epoch": 0.06213487730141232, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019814090114871207, + "loss": 0.1396, + "step": 7158 + }, + { + "epoch": 0.06214355778161648, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019814029929061273, + "loss": 0.125, + "step": 7159 + }, + { + "epoch": 0.06215223826182065, + "grad_norm": 0.3203125, + "learning_rate": 0.0019813969733612448, + "loss": 0.1484, + "step": 7160 + }, + { + "epoch": 0.06216091874202481, + "grad_norm": 0.890625, + "learning_rate": 0.00198139095285248, + "loss": 0.1426, + "step": 7161 + }, + { + "epoch": 0.06216959922222897, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019813849313798392, + "loss": 0.1582, + "step": 7162 + }, + { + "epoch": 0.062178279702433135, + "grad_norm": 0.361328125, + "learning_rate": 0.0019813789089433292, + "loss": 0.1719, + "step": 7163 + }, + { + "epoch": 0.0621869601826373, + "grad_norm": 0.236328125, + "learning_rate": 0.0019813728855429566, + "loss": 0.1309, + "step": 7164 + }, + { + "epoch": 0.062195640662841466, + "grad_norm": 0.341796875, + "learning_rate": 0.0019813668611787278, + "loss": 0.1226, + "step": 7165 + }, + { + "epoch": 0.06220432114304563, + "grad_norm": 0.095703125, + "learning_rate": 0.0019813608358506493, + "loss": 0.1348, + "step": 7166 + }, + { + "epoch": 0.062213001623249796, + "grad_norm": 0.078125, + "learning_rate": 0.001981354809558728, + "loss": 0.1553, + "step": 7167 + }, + { + "epoch": 0.06222168210345396, + "grad_norm": 0.25390625, + "learning_rate": 0.00198134878230297, + "loss": 0.1621, + "step": 7168 + }, + { + "epoch": 0.062230362583658126, + "grad_norm": 0.26953125, + "learning_rate": 0.001981342754083382, + "loss": 0.1055, + "step": 7169 + }, + { + "epoch": 0.06223904306386229, + "grad_norm": 0.189453125, + "learning_rate": 0.0019813367248999714, + "loss": 0.1875, + "step": 7170 + }, + { + "epoch": 0.062247723544066456, + "grad_norm": 0.125, + "learning_rate": 0.001981330694752744, + "loss": 0.1445, + "step": 7171 + }, + { + "epoch": 0.06225640402427062, + "grad_norm": 0.068359375, + "learning_rate": 0.0019813246636417067, + "loss": 0.127, + "step": 7172 + }, + { + "epoch": 0.06226508450447479, + "grad_norm": 0.458984375, + "learning_rate": 0.0019813186315668662, + "loss": 0.1484, + "step": 7173 + }, + { + "epoch": 0.06227376498467895, + "grad_norm": 0.173828125, + "learning_rate": 0.0019813125985282285, + "loss": 0.1514, + "step": 7174 + }, + { + "epoch": 0.06228244546488312, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019813065645258006, + "loss": 0.1279, + "step": 7175 + }, + { + "epoch": 0.06229112594508728, + "grad_norm": 0.11474609375, + "learning_rate": 0.001981300529559589, + "loss": 0.1196, + "step": 7176 + }, + { + "epoch": 0.06229980642529145, + "grad_norm": 0.197265625, + "learning_rate": 0.0019812944936296005, + "loss": 0.1494, + "step": 7177 + }, + { + "epoch": 0.06230848690549561, + "grad_norm": 0.265625, + "learning_rate": 0.0019812884567358417, + "loss": 0.1475, + "step": 7178 + }, + { + "epoch": 0.06231716738569978, + "grad_norm": 0.44921875, + "learning_rate": 0.001981282418878319, + "loss": 0.1055, + "step": 7179 + }, + { + "epoch": 0.06232584786590394, + "grad_norm": 1.0546875, + "learning_rate": 0.0019812763800570394, + "loss": 0.1094, + "step": 7180 + }, + { + "epoch": 0.06233452834610811, + "grad_norm": 0.2734375, + "learning_rate": 0.0019812703402720087, + "loss": 0.1143, + "step": 7181 + }, + { + "epoch": 0.06234320882631227, + "grad_norm": 0.361328125, + "learning_rate": 0.0019812642995232343, + "loss": 0.1182, + "step": 7182 + }, + { + "epoch": 0.06235188930651644, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019812582578107225, + "loss": 0.166, + "step": 7183 + }, + { + "epoch": 0.0623605697867206, + "grad_norm": 0.34375, + "learning_rate": 0.00198125221513448, + "loss": 0.1211, + "step": 7184 + }, + { + "epoch": 0.06236925026692477, + "grad_norm": 0.2265625, + "learning_rate": 0.001981246171494513, + "loss": 0.1191, + "step": 7185 + }, + { + "epoch": 0.06237793074712893, + "grad_norm": 0.140625, + "learning_rate": 0.001981240126890828, + "loss": 0.1445, + "step": 7186 + }, + { + "epoch": 0.0623866112273331, + "grad_norm": 0.84375, + "learning_rate": 0.001981234081323433, + "loss": 0.1484, + "step": 7187 + }, + { + "epoch": 0.06239529170753726, + "grad_norm": 0.4375, + "learning_rate": 0.0019812280347923333, + "loss": 0.1113, + "step": 7188 + }, + { + "epoch": 0.06240397218774143, + "grad_norm": 0.25390625, + "learning_rate": 0.001981221987297536, + "loss": 0.1367, + "step": 7189 + }, + { + "epoch": 0.062412652667945594, + "grad_norm": 0.283203125, + "learning_rate": 0.001981215938839047, + "loss": 0.127, + "step": 7190 + }, + { + "epoch": 0.06242133314814976, + "grad_norm": 0.20703125, + "learning_rate": 0.0019812098894168738, + "loss": 0.1523, + "step": 7191 + }, + { + "epoch": 0.062430013628353924, + "grad_norm": 0.2265625, + "learning_rate": 0.0019812038390310226, + "loss": 0.1562, + "step": 7192 + }, + { + "epoch": 0.06243869410855808, + "grad_norm": 0.82421875, + "learning_rate": 0.0019811977876815004, + "loss": 0.1758, + "step": 7193 + }, + { + "epoch": 0.06244737458876225, + "grad_norm": 0.30078125, + "learning_rate": 0.0019811917353683133, + "loss": 0.1641, + "step": 7194 + }, + { + "epoch": 0.06245605506896641, + "grad_norm": 0.5, + "learning_rate": 0.0019811856820914683, + "loss": 0.1777, + "step": 7195 + }, + { + "epoch": 0.06246473554917058, + "grad_norm": 0.55078125, + "learning_rate": 0.0019811796278509717, + "loss": 0.1299, + "step": 7196 + }, + { + "epoch": 0.06247341602937474, + "grad_norm": 0.25390625, + "learning_rate": 0.0019811735726468307, + "loss": 0.1289, + "step": 7197 + }, + { + "epoch": 0.06248209650957891, + "grad_norm": 0.484375, + "learning_rate": 0.001981167516479051, + "loss": 0.1475, + "step": 7198 + }, + { + "epoch": 0.06249077698978307, + "grad_norm": 0.462890625, + "learning_rate": 0.00198116145934764, + "loss": 0.1348, + "step": 7199 + }, + { + "epoch": 0.06249945746998724, + "grad_norm": 0.80078125, + "learning_rate": 0.0019811554012526046, + "loss": 0.1113, + "step": 7200 + }, + { + "epoch": 0.0625081379501914, + "grad_norm": 0.46875, + "learning_rate": 0.00198114934219395, + "loss": 0.1465, + "step": 7201 + }, + { + "epoch": 0.06251681843039557, + "grad_norm": 0.234375, + "learning_rate": 0.0019811432821716843, + "loss": 0.1387, + "step": 7202 + }, + { + "epoch": 0.06252549891059973, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019811372211858136, + "loss": 0.1377, + "step": 7203 + }, + { + "epoch": 0.0625341793908039, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019811311592363443, + "loss": 0.1387, + "step": 7204 + }, + { + "epoch": 0.06254285987100806, + "grad_norm": 0.3828125, + "learning_rate": 0.001981125096323283, + "loss": 0.1318, + "step": 7205 + }, + { + "epoch": 0.06255154035121223, + "grad_norm": 0.30859375, + "learning_rate": 0.0019811190324466365, + "loss": 0.1157, + "step": 7206 + }, + { + "epoch": 0.0625602208314164, + "grad_norm": 0.73046875, + "learning_rate": 0.001981112967606412, + "loss": 0.1045, + "step": 7207 + }, + { + "epoch": 0.06256890131162056, + "grad_norm": 0.37890625, + "learning_rate": 0.0019811069018026146, + "loss": 0.1582, + "step": 7208 + }, + { + "epoch": 0.06257758179182472, + "grad_norm": 0.26953125, + "learning_rate": 0.0019811008350352527, + "loss": 0.1328, + "step": 7209 + }, + { + "epoch": 0.06258626227202889, + "grad_norm": 0.1806640625, + "learning_rate": 0.001981094767304332, + "loss": 0.1416, + "step": 7210 + }, + { + "epoch": 0.06259494275223305, + "grad_norm": 0.328125, + "learning_rate": 0.0019810886986098594, + "loss": 0.1289, + "step": 7211 + }, + { + "epoch": 0.06260362323243722, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019810826289518415, + "loss": 0.1592, + "step": 7212 + }, + { + "epoch": 0.06261230371264138, + "grad_norm": 0.95703125, + "learning_rate": 0.0019810765583302844, + "loss": 0.1621, + "step": 7213 + }, + { + "epoch": 0.06262098419284555, + "grad_norm": 0.65625, + "learning_rate": 0.0019810704867451955, + "loss": 0.1221, + "step": 7214 + }, + { + "epoch": 0.06262966467304971, + "grad_norm": 0.5390625, + "learning_rate": 0.0019810644141965814, + "loss": 0.1465, + "step": 7215 + }, + { + "epoch": 0.06263834515325388, + "grad_norm": 0.08984375, + "learning_rate": 0.001981058340684448, + "loss": 0.1455, + "step": 7216 + }, + { + "epoch": 0.06264702563345804, + "grad_norm": 0.34375, + "learning_rate": 0.001981052266208803, + "loss": 0.1465, + "step": 7217 + }, + { + "epoch": 0.06265570611366221, + "grad_norm": 0.345703125, + "learning_rate": 0.001981046190769652, + "loss": 0.1406, + "step": 7218 + }, + { + "epoch": 0.06266438659386638, + "grad_norm": 3.609375, + "learning_rate": 0.0019810401143670023, + "loss": 0.3086, + "step": 7219 + }, + { + "epoch": 0.06267306707407054, + "grad_norm": 0.81640625, + "learning_rate": 0.0019810340370008607, + "loss": 0.1455, + "step": 7220 + }, + { + "epoch": 0.0626817475542747, + "grad_norm": 0.4375, + "learning_rate": 0.0019810279586712333, + "loss": 0.123, + "step": 7221 + }, + { + "epoch": 0.06269042803447887, + "grad_norm": 0.408203125, + "learning_rate": 0.0019810218793781267, + "loss": 0.1367, + "step": 7222 + }, + { + "epoch": 0.06269910851468304, + "grad_norm": 0.7578125, + "learning_rate": 0.001981015799121548, + "loss": 0.1309, + "step": 7223 + }, + { + "epoch": 0.0627077889948872, + "grad_norm": 0.232421875, + "learning_rate": 0.001981009717901504, + "loss": 0.1406, + "step": 7224 + }, + { + "epoch": 0.06271646947509137, + "grad_norm": 0.28515625, + "learning_rate": 0.0019810036357180002, + "loss": 0.1514, + "step": 7225 + }, + { + "epoch": 0.06272514995529553, + "grad_norm": 0.314453125, + "learning_rate": 0.0019809975525710444, + "loss": 0.1592, + "step": 7226 + }, + { + "epoch": 0.0627338304354997, + "grad_norm": 0.15625, + "learning_rate": 0.001980991468460643, + "loss": 0.1118, + "step": 7227 + }, + { + "epoch": 0.06274251091570386, + "grad_norm": 0.3046875, + "learning_rate": 0.001980985383386803, + "loss": 0.1113, + "step": 7228 + }, + { + "epoch": 0.06275119139590803, + "grad_norm": 0.443359375, + "learning_rate": 0.00198097929734953, + "loss": 0.1387, + "step": 7229 + }, + { + "epoch": 0.06275987187611219, + "grad_norm": 0.37109375, + "learning_rate": 0.0019809732103488317, + "loss": 0.1455, + "step": 7230 + }, + { + "epoch": 0.06276855235631636, + "grad_norm": 0.5546875, + "learning_rate": 0.001980967122384714, + "loss": 0.1299, + "step": 7231 + }, + { + "epoch": 0.06277723283652052, + "grad_norm": 0.1396484375, + "learning_rate": 0.001980961033457184, + "loss": 0.0967, + "step": 7232 + }, + { + "epoch": 0.06278591331672469, + "grad_norm": 0.08447265625, + "learning_rate": 0.001980954943566248, + "loss": 0.1201, + "step": 7233 + }, + { + "epoch": 0.06279459379692885, + "grad_norm": 0.2734375, + "learning_rate": 0.0019809488527119134, + "loss": 0.1406, + "step": 7234 + }, + { + "epoch": 0.06280327427713302, + "grad_norm": 0.3125, + "learning_rate": 0.001980942760894186, + "loss": 0.1953, + "step": 7235 + }, + { + "epoch": 0.06281195475733718, + "grad_norm": 0.23828125, + "learning_rate": 0.001980936668113073, + "loss": 0.105, + "step": 7236 + }, + { + "epoch": 0.06282063523754135, + "grad_norm": 0.263671875, + "learning_rate": 0.001980930574368581, + "loss": 0.1387, + "step": 7237 + }, + { + "epoch": 0.06282931571774551, + "grad_norm": 0.337890625, + "learning_rate": 0.0019809244796607163, + "loss": 0.1602, + "step": 7238 + }, + { + "epoch": 0.06283799619794966, + "grad_norm": 0.353515625, + "learning_rate": 0.0019809183839894864, + "loss": 0.2109, + "step": 7239 + }, + { + "epoch": 0.06284667667815383, + "grad_norm": 0.349609375, + "learning_rate": 0.001980912287354897, + "loss": 0.1309, + "step": 7240 + }, + { + "epoch": 0.062855357158358, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019809061897569547, + "loss": 0.1055, + "step": 7241 + }, + { + "epoch": 0.06286403763856216, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019809000911956668, + "loss": 0.1768, + "step": 7242 + }, + { + "epoch": 0.06287271811876632, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019808939916710405, + "loss": 0.1162, + "step": 7243 + }, + { + "epoch": 0.06288139859897049, + "grad_norm": 0.431640625, + "learning_rate": 0.001980887891183081, + "loss": 0.1182, + "step": 7244 + }, + { + "epoch": 0.06289007907917465, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019808817897317963, + "loss": 0.1445, + "step": 7245 + }, + { + "epoch": 0.06289875955937882, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019808756873171923, + "loss": 0.1221, + "step": 7246 + }, + { + "epoch": 0.06290744003958298, + "grad_norm": 0.205078125, + "learning_rate": 0.0019808695839392755, + "loss": 0.1152, + "step": 7247 + }, + { + "epoch": 0.06291612051978715, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019808634795980534, + "loss": 0.1562, + "step": 7248 + }, + { + "epoch": 0.06292480099999131, + "grad_norm": 0.2255859375, + "learning_rate": 0.001980857374293532, + "loss": 0.1387, + "step": 7249 + }, + { + "epoch": 0.06293348148019548, + "grad_norm": 0.345703125, + "learning_rate": 0.0019808512680257184, + "loss": 0.1055, + "step": 7250 + }, + { + "epoch": 0.06294216196039965, + "grad_norm": 0.56640625, + "learning_rate": 0.001980845160794619, + "loss": 0.1914, + "step": 7251 + }, + { + "epoch": 0.06295084244060381, + "grad_norm": 0.1123046875, + "learning_rate": 0.001980839052600241, + "loss": 0.1216, + "step": 7252 + }, + { + "epoch": 0.06295952292080798, + "grad_norm": 0.2138671875, + "learning_rate": 0.00198083294344259, + "loss": 0.1328, + "step": 7253 + }, + { + "epoch": 0.06296820340101214, + "grad_norm": 0.451171875, + "learning_rate": 0.0019808268333216736, + "loss": 0.1113, + "step": 7254 + }, + { + "epoch": 0.0629768838812163, + "grad_norm": 0.52734375, + "learning_rate": 0.001980820722237498, + "loss": 0.0933, + "step": 7255 + }, + { + "epoch": 0.06298556436142047, + "grad_norm": 0.875, + "learning_rate": 0.0019808146101900704, + "loss": 0.1367, + "step": 7256 + }, + { + "epoch": 0.06299424484162464, + "grad_norm": 0.35546875, + "learning_rate": 0.001980808497179397, + "loss": 0.1465, + "step": 7257 + }, + { + "epoch": 0.0630029253218288, + "grad_norm": 0.310546875, + "learning_rate": 0.0019808023832054845, + "loss": 0.1279, + "step": 7258 + }, + { + "epoch": 0.06301160580203297, + "grad_norm": 0.443359375, + "learning_rate": 0.00198079626826834, + "loss": 0.1465, + "step": 7259 + }, + { + "epoch": 0.06302028628223713, + "grad_norm": 0.13671875, + "learning_rate": 0.00198079015236797, + "loss": 0.1016, + "step": 7260 + }, + { + "epoch": 0.0630289667624413, + "grad_norm": 0.490234375, + "learning_rate": 0.0019807840355043813, + "loss": 0.1299, + "step": 7261 + }, + { + "epoch": 0.06303764724264546, + "grad_norm": 0.400390625, + "learning_rate": 0.0019807779176775797, + "loss": 0.1641, + "step": 7262 + }, + { + "epoch": 0.06304632772284963, + "grad_norm": 0.51953125, + "learning_rate": 0.001980771798887573, + "loss": 0.1523, + "step": 7263 + }, + { + "epoch": 0.06305500820305379, + "grad_norm": 0.34375, + "learning_rate": 0.0019807656791343674, + "loss": 0.1406, + "step": 7264 + }, + { + "epoch": 0.06306368868325796, + "grad_norm": 0.408203125, + "learning_rate": 0.0019807595584179698, + "loss": 0.1416, + "step": 7265 + }, + { + "epoch": 0.06307236916346212, + "grad_norm": 0.75390625, + "learning_rate": 0.001980753436738387, + "loss": 0.1895, + "step": 7266 + }, + { + "epoch": 0.06308104964366629, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019807473140956255, + "loss": 0.1377, + "step": 7267 + }, + { + "epoch": 0.06308973012387045, + "grad_norm": 0.27734375, + "learning_rate": 0.0019807411904896916, + "loss": 0.1416, + "step": 7268 + }, + { + "epoch": 0.06309841060407462, + "grad_norm": 0.41015625, + "learning_rate": 0.0019807350659205923, + "loss": 0.165, + "step": 7269 + }, + { + "epoch": 0.06310709108427878, + "grad_norm": 0.150390625, + "learning_rate": 0.0019807289403883342, + "loss": 0.1895, + "step": 7270 + }, + { + "epoch": 0.06311577156448295, + "grad_norm": 0.69140625, + "learning_rate": 0.001980722813892925, + "loss": 0.1582, + "step": 7271 + }, + { + "epoch": 0.06312445204468711, + "grad_norm": 0.34765625, + "learning_rate": 0.0019807166864343703, + "loss": 0.1338, + "step": 7272 + }, + { + "epoch": 0.06313313252489128, + "grad_norm": 0.255859375, + "learning_rate": 0.0019807105580126765, + "loss": 0.1445, + "step": 7273 + }, + { + "epoch": 0.06314181300509544, + "grad_norm": 0.08056640625, + "learning_rate": 0.001980704428627851, + "loss": 0.1318, + "step": 7274 + }, + { + "epoch": 0.06315049348529961, + "grad_norm": 0.380859375, + "learning_rate": 0.0019806982982799007, + "loss": 0.1621, + "step": 7275 + }, + { + "epoch": 0.06315917396550377, + "grad_norm": 0.1416015625, + "learning_rate": 0.001980692166968832, + "loss": 0.1426, + "step": 7276 + }, + { + "epoch": 0.06316785444570794, + "grad_norm": 0.0703125, + "learning_rate": 0.0019806860346946514, + "loss": 0.1074, + "step": 7277 + }, + { + "epoch": 0.0631765349259121, + "grad_norm": 0.34765625, + "learning_rate": 0.001980679901457366, + "loss": 0.1426, + "step": 7278 + }, + { + "epoch": 0.06318521540611627, + "grad_norm": 0.11767578125, + "learning_rate": 0.001980673767256982, + "loss": 0.123, + "step": 7279 + }, + { + "epoch": 0.06319389588632043, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019806676320935066, + "loss": 0.1211, + "step": 7280 + }, + { + "epoch": 0.0632025763665246, + "grad_norm": 0.11767578125, + "learning_rate": 0.001980661495966946, + "loss": 0.1875, + "step": 7281 + }, + { + "epoch": 0.06321125684672876, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019806553588773076, + "loss": 0.1235, + "step": 7282 + }, + { + "epoch": 0.06321993732693293, + "grad_norm": 0.3515625, + "learning_rate": 0.0019806492208245974, + "loss": 0.1348, + "step": 7283 + }, + { + "epoch": 0.0632286178071371, + "grad_norm": 0.28125, + "learning_rate": 0.001980643081808823, + "loss": 0.1738, + "step": 7284 + }, + { + "epoch": 0.06323729828734126, + "grad_norm": 0.3671875, + "learning_rate": 0.00198063694182999, + "loss": 0.1455, + "step": 7285 + }, + { + "epoch": 0.06324597876754542, + "grad_norm": 0.478515625, + "learning_rate": 0.0019806308008881056, + "loss": 0.166, + "step": 7286 + }, + { + "epoch": 0.06325465924774959, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019806246589831773, + "loss": 0.1216, + "step": 7287 + }, + { + "epoch": 0.06326333972795375, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019806185161152104, + "loss": 0.1172, + "step": 7288 + }, + { + "epoch": 0.06327202020815792, + "grad_norm": 0.255859375, + "learning_rate": 0.0019806123722842127, + "loss": 0.2266, + "step": 7289 + }, + { + "epoch": 0.06328070068836208, + "grad_norm": 0.3515625, + "learning_rate": 0.0019806062274901905, + "loss": 0.1143, + "step": 7290 + }, + { + "epoch": 0.06328938116856625, + "grad_norm": 0.1025390625, + "learning_rate": 0.00198060008173315, + "loss": 0.1465, + "step": 7291 + }, + { + "epoch": 0.06329806164877042, + "grad_norm": 0.29296875, + "learning_rate": 0.001980593935013099, + "loss": 0.1299, + "step": 7292 + }, + { + "epoch": 0.06330674212897458, + "grad_norm": 0.09423828125, + "learning_rate": 0.001980587787330044, + "loss": 0.1465, + "step": 7293 + }, + { + "epoch": 0.06331542260917875, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019805816386839906, + "loss": 0.1377, + "step": 7294 + }, + { + "epoch": 0.06332410308938291, + "grad_norm": 0.84765625, + "learning_rate": 0.0019805754890749473, + "loss": 0.1436, + "step": 7295 + }, + { + "epoch": 0.06333278356958708, + "grad_norm": 0.455078125, + "learning_rate": 0.001980569338502919, + "loss": 0.166, + "step": 7296 + }, + { + "epoch": 0.06334146404979124, + "grad_norm": 0.48046875, + "learning_rate": 0.0019805631869679137, + "loss": 0.1426, + "step": 7297 + }, + { + "epoch": 0.0633501445299954, + "grad_norm": 0.0947265625, + "learning_rate": 0.001980557034469938, + "loss": 0.1318, + "step": 7298 + }, + { + "epoch": 0.06335882501019957, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019805508810089977, + "loss": 0.1523, + "step": 7299 + }, + { + "epoch": 0.06336750549040374, + "grad_norm": 0.181640625, + "learning_rate": 0.001980544726585101, + "loss": 0.1113, + "step": 7300 + }, + { + "epoch": 0.06337618597060789, + "grad_norm": 0.439453125, + "learning_rate": 0.001980538571198253, + "loss": 0.1484, + "step": 7301 + }, + { + "epoch": 0.06338486645081205, + "grad_norm": 0.220703125, + "learning_rate": 0.001980532414848462, + "loss": 0.1504, + "step": 7302 + }, + { + "epoch": 0.06339354693101622, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019805262575357334, + "loss": 0.125, + "step": 7303 + }, + { + "epoch": 0.06340222741122038, + "grad_norm": 0.158203125, + "learning_rate": 0.001980520099260075, + "loss": 0.1074, + "step": 7304 + }, + { + "epoch": 0.06341090789142455, + "grad_norm": 0.26953125, + "learning_rate": 0.0019805139400214927, + "loss": 0.1621, + "step": 7305 + }, + { + "epoch": 0.06341958837162871, + "grad_norm": 0.11328125, + "learning_rate": 0.0019805077798199936, + "loss": 0.1445, + "step": 7306 + }, + { + "epoch": 0.06342826885183288, + "grad_norm": 0.546875, + "learning_rate": 0.0019805016186555846, + "loss": 0.1182, + "step": 7307 + }, + { + "epoch": 0.06343694933203704, + "grad_norm": 0.357421875, + "learning_rate": 0.0019804954565282723, + "loss": 0.1553, + "step": 7308 + }, + { + "epoch": 0.06344562981224121, + "grad_norm": 0.875, + "learning_rate": 0.001980489293438063, + "loss": 0.127, + "step": 7309 + }, + { + "epoch": 0.06345431029244537, + "grad_norm": 0.37890625, + "learning_rate": 0.001980483129384964, + "loss": 0.1592, + "step": 7310 + }, + { + "epoch": 0.06346299077264954, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019804769643689824, + "loss": 0.1562, + "step": 7311 + }, + { + "epoch": 0.0634716712528537, + "grad_norm": 0.27734375, + "learning_rate": 0.0019804707983901238, + "loss": 0.1299, + "step": 7312 + }, + { + "epoch": 0.06348035173305787, + "grad_norm": 1.0703125, + "learning_rate": 0.0019804646314483957, + "loss": 0.1426, + "step": 7313 + }, + { + "epoch": 0.06348903221326203, + "grad_norm": 0.890625, + "learning_rate": 0.0019804584635438045, + "loss": 0.125, + "step": 7314 + }, + { + "epoch": 0.0634977126934662, + "grad_norm": 0.61328125, + "learning_rate": 0.0019804522946763577, + "loss": 0.1436, + "step": 7315 + }, + { + "epoch": 0.06350639317367036, + "grad_norm": 1.0, + "learning_rate": 0.001980446124846061, + "loss": 0.1279, + "step": 7316 + }, + { + "epoch": 0.06351507365387453, + "grad_norm": 0.52734375, + "learning_rate": 0.0019804399540529222, + "loss": 0.1631, + "step": 7317 + }, + { + "epoch": 0.0635237541340787, + "grad_norm": 0.244140625, + "learning_rate": 0.001980433782296947, + "loss": 0.1377, + "step": 7318 + }, + { + "epoch": 0.06353243461428286, + "grad_norm": 0.42578125, + "learning_rate": 0.0019804276095781426, + "loss": 0.1465, + "step": 7319 + }, + { + "epoch": 0.06354111509448702, + "grad_norm": 0.6875, + "learning_rate": 0.0019804214358965164, + "loss": 0.1279, + "step": 7320 + }, + { + "epoch": 0.06354979557469119, + "grad_norm": 0.51171875, + "learning_rate": 0.001980415261252074, + "loss": 0.1289, + "step": 7321 + }, + { + "epoch": 0.06355847605489535, + "grad_norm": 0.66796875, + "learning_rate": 0.001980409085644823, + "loss": 0.1064, + "step": 7322 + }, + { + "epoch": 0.06356715653509952, + "grad_norm": 0.578125, + "learning_rate": 0.0019804029090747695, + "loss": 0.1299, + "step": 7323 + }, + { + "epoch": 0.06357583701530369, + "grad_norm": 0.34375, + "learning_rate": 0.0019803967315419205, + "loss": 0.2012, + "step": 7324 + }, + { + "epoch": 0.06358451749550785, + "grad_norm": 0.29296875, + "learning_rate": 0.0019803905530462835, + "loss": 0.1523, + "step": 7325 + }, + { + "epoch": 0.06359319797571202, + "grad_norm": 0.2265625, + "learning_rate": 0.001980384373587864, + "loss": 0.1069, + "step": 7326 + }, + { + "epoch": 0.06360187845591618, + "grad_norm": 0.412109375, + "learning_rate": 0.0019803781931666694, + "loss": 0.1182, + "step": 7327 + }, + { + "epoch": 0.06361055893612035, + "grad_norm": 0.390625, + "learning_rate": 0.001980372011782707, + "loss": 0.2002, + "step": 7328 + }, + { + "epoch": 0.06361923941632451, + "grad_norm": 0.380859375, + "learning_rate": 0.0019803658294359826, + "loss": 0.1348, + "step": 7329 + }, + { + "epoch": 0.06362791989652868, + "grad_norm": 0.197265625, + "learning_rate": 0.0019803596461265035, + "loss": 0.1807, + "step": 7330 + }, + { + "epoch": 0.06363660037673284, + "grad_norm": 0.66796875, + "learning_rate": 0.0019803534618542764, + "loss": 0.1543, + "step": 7331 + }, + { + "epoch": 0.063645280856937, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019803472766193077, + "loss": 0.126, + "step": 7332 + }, + { + "epoch": 0.06365396133714117, + "grad_norm": 0.51171875, + "learning_rate": 0.0019803410904216047, + "loss": 0.1328, + "step": 7333 + }, + { + "epoch": 0.06366264181734534, + "grad_norm": 0.4296875, + "learning_rate": 0.0019803349032611736, + "loss": 0.1484, + "step": 7334 + }, + { + "epoch": 0.0636713222975495, + "grad_norm": 0.51953125, + "learning_rate": 0.0019803287151380217, + "loss": 0.1367, + "step": 7335 + }, + { + "epoch": 0.06368000277775367, + "grad_norm": 0.79296875, + "learning_rate": 0.0019803225260521555, + "loss": 0.2344, + "step": 7336 + }, + { + "epoch": 0.06368868325795783, + "grad_norm": 0.79296875, + "learning_rate": 0.001980316336003582, + "loss": 0.1299, + "step": 7337 + }, + { + "epoch": 0.063697363738162, + "grad_norm": 0.3359375, + "learning_rate": 0.0019803101449923075, + "loss": 0.1465, + "step": 7338 + }, + { + "epoch": 0.06370604421836616, + "grad_norm": 0.1875, + "learning_rate": 0.0019803039530183393, + "loss": 0.125, + "step": 7339 + }, + { + "epoch": 0.06371472469857033, + "grad_norm": 0.21484375, + "learning_rate": 0.0019802977600816836, + "loss": 0.1484, + "step": 7340 + }, + { + "epoch": 0.06372340517877449, + "grad_norm": 0.1259765625, + "learning_rate": 0.001980291566182348, + "loss": 0.124, + "step": 7341 + }, + { + "epoch": 0.06373208565897866, + "grad_norm": 0.48828125, + "learning_rate": 0.0019802853713203382, + "loss": 0.166, + "step": 7342 + }, + { + "epoch": 0.06374076613918282, + "grad_norm": 0.390625, + "learning_rate": 0.001980279175495662, + "loss": 0.1562, + "step": 7343 + }, + { + "epoch": 0.06374944661938699, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019802729787083257, + "loss": 0.1172, + "step": 7344 + }, + { + "epoch": 0.06375812709959115, + "grad_norm": 0.1484375, + "learning_rate": 0.0019802667809583354, + "loss": 0.1221, + "step": 7345 + }, + { + "epoch": 0.06376680757979532, + "grad_norm": 0.671875, + "learning_rate": 0.0019802605822456993, + "loss": 0.1328, + "step": 7346 + }, + { + "epoch": 0.06377548805999948, + "grad_norm": 1.3671875, + "learning_rate": 0.001980254382570423, + "loss": 0.1309, + "step": 7347 + }, + { + "epoch": 0.06378416854020365, + "grad_norm": 0.322265625, + "learning_rate": 0.001980248181932514, + "loss": 0.1367, + "step": 7348 + }, + { + "epoch": 0.06379284902040781, + "grad_norm": 0.2001953125, + "learning_rate": 0.001980241980331979, + "loss": 0.1118, + "step": 7349 + }, + { + "epoch": 0.06380152950061198, + "grad_norm": 0.63671875, + "learning_rate": 0.0019802357777688245, + "loss": 0.1328, + "step": 7350 + }, + { + "epoch": 0.06381020998081614, + "grad_norm": 0.07421875, + "learning_rate": 0.001980229574243057, + "loss": 0.1016, + "step": 7351 + }, + { + "epoch": 0.06381889046102031, + "grad_norm": 0.2275390625, + "learning_rate": 0.001980223369754684, + "loss": 0.1182, + "step": 7352 + }, + { + "epoch": 0.06382757094122447, + "grad_norm": 0.453125, + "learning_rate": 0.0019802171643037115, + "loss": 0.1118, + "step": 7353 + }, + { + "epoch": 0.06383625142142864, + "grad_norm": 0.42578125, + "learning_rate": 0.0019802109578901475, + "loss": 0.123, + "step": 7354 + }, + { + "epoch": 0.0638449319016328, + "grad_norm": 0.478515625, + "learning_rate": 0.0019802047505139975, + "loss": 0.1113, + "step": 7355 + }, + { + "epoch": 0.06385361238183697, + "grad_norm": 0.251953125, + "learning_rate": 0.001980198542175269, + "loss": 0.1562, + "step": 7356 + }, + { + "epoch": 0.06386229286204113, + "grad_norm": 1.0703125, + "learning_rate": 0.001980192332873969, + "loss": 0.1348, + "step": 7357 + }, + { + "epoch": 0.0638709733422453, + "grad_norm": 0.205078125, + "learning_rate": 0.001980186122610103, + "loss": 0.1699, + "step": 7358 + }, + { + "epoch": 0.06387965382244946, + "grad_norm": 0.27734375, + "learning_rate": 0.001980179911383679, + "loss": 0.1289, + "step": 7359 + }, + { + "epoch": 0.06388833430265363, + "grad_norm": 0.5625, + "learning_rate": 0.001980173699194704, + "loss": 0.1162, + "step": 7360 + }, + { + "epoch": 0.0638970147828578, + "grad_norm": 0.271484375, + "learning_rate": 0.0019801674860431837, + "loss": 0.1592, + "step": 7361 + }, + { + "epoch": 0.06390569526306195, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019801612719291257, + "loss": 0.1504, + "step": 7362 + }, + { + "epoch": 0.06391437574326611, + "grad_norm": 0.353515625, + "learning_rate": 0.001980155056852536, + "loss": 0.1504, + "step": 7363 + }, + { + "epoch": 0.06392305622347028, + "grad_norm": 1.0703125, + "learning_rate": 0.0019801488408134225, + "loss": 0.1719, + "step": 7364 + }, + { + "epoch": 0.06393173670367444, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019801426238117916, + "loss": 0.1279, + "step": 7365 + }, + { + "epoch": 0.0639404171838786, + "grad_norm": 0.55078125, + "learning_rate": 0.0019801364058476497, + "loss": 0.1328, + "step": 7366 + }, + { + "epoch": 0.06394909766408277, + "grad_norm": 0.09228515625, + "learning_rate": 0.001980130186921004, + "loss": 0.1641, + "step": 7367 + }, + { + "epoch": 0.06395777814428694, + "grad_norm": 0.333984375, + "learning_rate": 0.001980123967031861, + "loss": 0.0938, + "step": 7368 + }, + { + "epoch": 0.0639664586244911, + "grad_norm": 0.330078125, + "learning_rate": 0.001980117746180228, + "loss": 0.1133, + "step": 7369 + }, + { + "epoch": 0.06397513910469527, + "grad_norm": 0.3359375, + "learning_rate": 0.0019801115243661105, + "loss": 0.168, + "step": 7370 + }, + { + "epoch": 0.06398381958489943, + "grad_norm": 0.875, + "learning_rate": 0.001980105301589517, + "loss": 0.1367, + "step": 7371 + }, + { + "epoch": 0.0639925000651036, + "grad_norm": 0.82421875, + "learning_rate": 0.001980099077850454, + "loss": 0.1436, + "step": 7372 + }, + { + "epoch": 0.06400118054530776, + "grad_norm": 0.26953125, + "learning_rate": 0.001980092853148927, + "loss": 0.1279, + "step": 7373 + }, + { + "epoch": 0.06400986102551193, + "grad_norm": 0.373046875, + "learning_rate": 0.001980086627484944, + "loss": 0.1631, + "step": 7374 + }, + { + "epoch": 0.06401854150571609, + "grad_norm": 0.2890625, + "learning_rate": 0.0019800804008585113, + "loss": 0.1357, + "step": 7375 + }, + { + "epoch": 0.06402722198592026, + "grad_norm": 0.55078125, + "learning_rate": 0.001980074173269636, + "loss": 0.1504, + "step": 7376 + }, + { + "epoch": 0.06403590246612442, + "grad_norm": 0.5234375, + "learning_rate": 0.001980067944718325, + "loss": 0.0928, + "step": 7377 + }, + { + "epoch": 0.06404458294632859, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019800617152045848, + "loss": 0.1309, + "step": 7378 + }, + { + "epoch": 0.06405326342653275, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019800554847284224, + "loss": 0.1157, + "step": 7379 + }, + { + "epoch": 0.06406194390673692, + "grad_norm": 0.197265625, + "learning_rate": 0.0019800492532898446, + "loss": 0.1211, + "step": 7380 + }, + { + "epoch": 0.06407062438694108, + "grad_norm": 0.328125, + "learning_rate": 0.001980043020888858, + "loss": 0.127, + "step": 7381 + }, + { + "epoch": 0.06407930486714525, + "grad_norm": 0.30078125, + "learning_rate": 0.00198003678752547, + "loss": 0.1523, + "step": 7382 + }, + { + "epoch": 0.06408798534734941, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019800305531996864, + "loss": 0.1504, + "step": 7383 + }, + { + "epoch": 0.06409666582755358, + "grad_norm": 0.177734375, + "learning_rate": 0.0019800243179115143, + "loss": 0.1143, + "step": 7384 + }, + { + "epoch": 0.06410534630775774, + "grad_norm": 0.3046875, + "learning_rate": 0.001980018081660962, + "loss": 0.1465, + "step": 7385 + }, + { + "epoch": 0.06411402678796191, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019800118444480343, + "loss": 0.1064, + "step": 7386 + }, + { + "epoch": 0.06412270726816607, + "grad_norm": 0.41015625, + "learning_rate": 0.001980005606272739, + "loss": 0.1318, + "step": 7387 + }, + { + "epoch": 0.06413138774837024, + "grad_norm": 0.74609375, + "learning_rate": 0.001979999367135083, + "loss": 0.1465, + "step": 7388 + }, + { + "epoch": 0.0641400682285744, + "grad_norm": 0.36328125, + "learning_rate": 0.0019799931270350728, + "loss": 0.124, + "step": 7389 + }, + { + "epoch": 0.06414874870877857, + "grad_norm": 0.384765625, + "learning_rate": 0.0019799868859727153, + "loss": 0.1328, + "step": 7390 + }, + { + "epoch": 0.06415742918898273, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019799806439480174, + "loss": 0.1201, + "step": 7391 + }, + { + "epoch": 0.0641661096691869, + "grad_norm": 0.1005859375, + "learning_rate": 0.001979974400960986, + "loss": 0.1855, + "step": 7392 + }, + { + "epoch": 0.06417479014939106, + "grad_norm": 0.58203125, + "learning_rate": 0.0019799681570116276, + "loss": 0.1133, + "step": 7393 + }, + { + "epoch": 0.06418347062959523, + "grad_norm": 0.56640625, + "learning_rate": 0.001979961912099949, + "loss": 0.0977, + "step": 7394 + }, + { + "epoch": 0.0641921511097994, + "grad_norm": 0.158203125, + "learning_rate": 0.001979955666225958, + "loss": 0.2617, + "step": 7395 + }, + { + "epoch": 0.06420083159000356, + "grad_norm": 0.23046875, + "learning_rate": 0.00197994941938966, + "loss": 0.168, + "step": 7396 + }, + { + "epoch": 0.06420951207020772, + "grad_norm": 0.2392578125, + "learning_rate": 0.001979943171591063, + "loss": 0.1133, + "step": 7397 + }, + { + "epoch": 0.06421819255041189, + "grad_norm": 0.6171875, + "learning_rate": 0.0019799369228301732, + "loss": 0.1211, + "step": 7398 + }, + { + "epoch": 0.06422687303061606, + "grad_norm": 0.439453125, + "learning_rate": 0.0019799306731069976, + "loss": 0.1104, + "step": 7399 + }, + { + "epoch": 0.06423555351082022, + "grad_norm": 0.18359375, + "learning_rate": 0.001979924422421543, + "loss": 0.1533, + "step": 7400 + }, + { + "epoch": 0.06424423399102439, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019799181707738167, + "loss": 0.1367, + "step": 7401 + }, + { + "epoch": 0.06425291447122855, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019799119181638245, + "loss": 0.1416, + "step": 7402 + }, + { + "epoch": 0.06426159495143272, + "grad_norm": 0.1162109375, + "learning_rate": 0.001979905664591574, + "loss": 0.1436, + "step": 7403 + }, + { + "epoch": 0.06427027543163688, + "grad_norm": 0.73828125, + "learning_rate": 0.0019798994100570723, + "loss": 0.166, + "step": 7404 + }, + { + "epoch": 0.06427895591184105, + "grad_norm": 0.24609375, + "learning_rate": 0.001979893154560325, + "loss": 0.1562, + "step": 7405 + }, + { + "epoch": 0.06428763639204521, + "grad_norm": 0.287109375, + "learning_rate": 0.001979886898101341, + "loss": 0.1426, + "step": 7406 + }, + { + "epoch": 0.06429631687224938, + "grad_norm": 0.275390625, + "learning_rate": 0.0019798806406801247, + "loss": 0.1289, + "step": 7407 + }, + { + "epoch": 0.06430499735245354, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019798743822966843, + "loss": 0.127, + "step": 7408 + }, + { + "epoch": 0.0643136778326577, + "grad_norm": 0.73046875, + "learning_rate": 0.001979868122951027, + "loss": 0.1396, + "step": 7409 + }, + { + "epoch": 0.06432235831286187, + "grad_norm": 0.1884765625, + "learning_rate": 0.001979861862643159, + "loss": 0.1484, + "step": 7410 + }, + { + "epoch": 0.06433103879306604, + "grad_norm": 0.78515625, + "learning_rate": 0.001979855601373087, + "loss": 0.1611, + "step": 7411 + }, + { + "epoch": 0.0643397192732702, + "grad_norm": 0.2421875, + "learning_rate": 0.001979849339140819, + "loss": 0.1147, + "step": 7412 + }, + { + "epoch": 0.06434839975347437, + "grad_norm": 0.1630859375, + "learning_rate": 0.00197984307594636, + "loss": 0.1543, + "step": 7413 + }, + { + "epoch": 0.06435708023367853, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019798368117897184, + "loss": 0.165, + "step": 7414 + }, + { + "epoch": 0.0643657607138827, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019798305466709, + "loss": 0.1426, + "step": 7415 + }, + { + "epoch": 0.06437444119408686, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019798242805899127, + "loss": 0.1465, + "step": 7416 + }, + { + "epoch": 0.06438312167429103, + "grad_norm": 0.73828125, + "learning_rate": 0.001979818013546762, + "loss": 0.124, + "step": 7417 + }, + { + "epoch": 0.06439180215449519, + "grad_norm": 1.109375, + "learning_rate": 0.0019798117455414564, + "loss": 0.106, + "step": 7418 + }, + { + "epoch": 0.06440048263469936, + "grad_norm": 0.271484375, + "learning_rate": 0.0019798054765740015, + "loss": 0.1758, + "step": 7419 + }, + { + "epoch": 0.06440916311490352, + "grad_norm": 0.099609375, + "learning_rate": 0.0019797992066444047, + "loss": 0.1182, + "step": 7420 + }, + { + "epoch": 0.06441784359510769, + "grad_norm": 0.5234375, + "learning_rate": 0.0019797929357526726, + "loss": 0.1025, + "step": 7421 + }, + { + "epoch": 0.06442652407531185, + "grad_norm": 0.89453125, + "learning_rate": 0.001979786663898812, + "loss": 0.1885, + "step": 7422 + }, + { + "epoch": 0.06443520455551602, + "grad_norm": 0.255859375, + "learning_rate": 0.00197978039108283, + "loss": 0.1787, + "step": 7423 + }, + { + "epoch": 0.06444388503572017, + "grad_norm": 0.34375, + "learning_rate": 0.0019797741173047338, + "loss": 0.1128, + "step": 7424 + }, + { + "epoch": 0.06445256551592433, + "grad_norm": 0.65234375, + "learning_rate": 0.0019797678425645293, + "loss": 0.127, + "step": 7425 + }, + { + "epoch": 0.0644612459961285, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019797615668622243, + "loss": 0.0952, + "step": 7426 + }, + { + "epoch": 0.06446992647633266, + "grad_norm": 0.265625, + "learning_rate": 0.0019797552901978247, + "loss": 0.1367, + "step": 7427 + }, + { + "epoch": 0.06447860695653683, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019797490125713383, + "loss": 0.1514, + "step": 7428 + }, + { + "epoch": 0.064487287436741, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019797427339827717, + "loss": 0.1426, + "step": 7429 + }, + { + "epoch": 0.06449596791694516, + "grad_norm": 0.2080078125, + "learning_rate": 0.001979736454432132, + "loss": 0.1465, + "step": 7430 + }, + { + "epoch": 0.06450464839714933, + "grad_norm": 0.059814453125, + "learning_rate": 0.0019797301739194247, + "loss": 0.1206, + "step": 7431 + }, + { + "epoch": 0.06451332887735349, + "grad_norm": 0.357421875, + "learning_rate": 0.0019797238924446586, + "loss": 0.1328, + "step": 7432 + }, + { + "epoch": 0.06452200935755766, + "grad_norm": 0.625, + "learning_rate": 0.001979717610007839, + "loss": 0.1543, + "step": 7433 + }, + { + "epoch": 0.06453068983776182, + "grad_norm": 0.06787109375, + "learning_rate": 0.001979711326608974, + "loss": 0.0938, + "step": 7434 + }, + { + "epoch": 0.06453937031796599, + "grad_norm": 0.5859375, + "learning_rate": 0.00197970504224807, + "loss": 0.0874, + "step": 7435 + }, + { + "epoch": 0.06454805079817015, + "grad_norm": 0.64453125, + "learning_rate": 0.0019796987569251335, + "loss": 0.2012, + "step": 7436 + }, + { + "epoch": 0.06455673127837432, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019796924706401717, + "loss": 0.125, + "step": 7437 + }, + { + "epoch": 0.06456541175857848, + "grad_norm": 0.345703125, + "learning_rate": 0.0019796861833931913, + "loss": 0.1245, + "step": 7438 + }, + { + "epoch": 0.06457409223878265, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019796798951841996, + "loss": 0.1602, + "step": 7439 + }, + { + "epoch": 0.06458277271898681, + "grad_norm": 0.447265625, + "learning_rate": 0.0019796736060132027, + "loss": 0.1299, + "step": 7440 + }, + { + "epoch": 0.06459145319919098, + "grad_norm": 0.46875, + "learning_rate": 0.0019796673158802083, + "loss": 0.1133, + "step": 7441 + }, + { + "epoch": 0.06460013367939514, + "grad_norm": 0.578125, + "learning_rate": 0.001979661024785223, + "loss": 0.1641, + "step": 7442 + }, + { + "epoch": 0.0646088141595993, + "grad_norm": 0.6484375, + "learning_rate": 0.0019796547327282535, + "loss": 0.1201, + "step": 7443 + }, + { + "epoch": 0.06461749463980347, + "grad_norm": 0.78515625, + "learning_rate": 0.001979648439709307, + "loss": 0.1602, + "step": 7444 + }, + { + "epoch": 0.06462617512000764, + "grad_norm": 0.98828125, + "learning_rate": 0.0019796421457283896, + "loss": 0.125, + "step": 7445 + }, + { + "epoch": 0.0646348556002118, + "grad_norm": 0.216796875, + "learning_rate": 0.0019796358507855093, + "loss": 0.1289, + "step": 7446 + }, + { + "epoch": 0.06464353608041597, + "grad_norm": 0.478515625, + "learning_rate": 0.0019796295548806723, + "loss": 0.1465, + "step": 7447 + }, + { + "epoch": 0.06465221656062013, + "grad_norm": 0.58203125, + "learning_rate": 0.0019796232580138857, + "loss": 0.1113, + "step": 7448 + }, + { + "epoch": 0.0646608970408243, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019796169601851562, + "loss": 0.104, + "step": 7449 + }, + { + "epoch": 0.06466957752102846, + "grad_norm": 0.185546875, + "learning_rate": 0.0019796106613944905, + "loss": 0.1289, + "step": 7450 + }, + { + "epoch": 0.06467825800123263, + "grad_norm": 0.32421875, + "learning_rate": 0.0019796043616418964, + "loss": 0.1299, + "step": 7451 + }, + { + "epoch": 0.06468693848143679, + "grad_norm": 0.2373046875, + "learning_rate": 0.00197959806092738, + "loss": 0.1602, + "step": 7452 + }, + { + "epoch": 0.06469561896164096, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019795917592509483, + "loss": 0.1094, + "step": 7453 + }, + { + "epoch": 0.06470429944184512, + "grad_norm": 0.5625, + "learning_rate": 0.0019795854566126083, + "loss": 0.123, + "step": 7454 + }, + { + "epoch": 0.06471297992204929, + "grad_norm": 0.7109375, + "learning_rate": 0.001979579153012367, + "loss": 0.1846, + "step": 7455 + }, + { + "epoch": 0.06472166040225345, + "grad_norm": 0.66015625, + "learning_rate": 0.001979572848450231, + "loss": 0.1621, + "step": 7456 + }, + { + "epoch": 0.06473034088245762, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019795665429262073, + "loss": 0.166, + "step": 7457 + }, + { + "epoch": 0.06473902136266178, + "grad_norm": 0.09619140625, + "learning_rate": 0.001979560236440303, + "loss": 0.1416, + "step": 7458 + }, + { + "epoch": 0.06474770184286595, + "grad_norm": 0.22265625, + "learning_rate": 0.0019795539289925246, + "loss": 0.1309, + "step": 7459 + }, + { + "epoch": 0.06475638232307011, + "grad_norm": 0.3046875, + "learning_rate": 0.0019795476205828797, + "loss": 0.167, + "step": 7460 + }, + { + "epoch": 0.06476506280327428, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019795413112113744, + "loss": 0.1289, + "step": 7461 + }, + { + "epoch": 0.06477374328347844, + "grad_norm": 0.36328125, + "learning_rate": 0.0019795350008780157, + "loss": 0.127, + "step": 7462 + }, + { + "epoch": 0.06478242376368261, + "grad_norm": 0.1357421875, + "learning_rate": 0.001979528689582811, + "loss": 0.1309, + "step": 7463 + }, + { + "epoch": 0.06479110424388677, + "grad_norm": 0.48046875, + "learning_rate": 0.001979522377325767, + "loss": 0.1562, + "step": 7464 + }, + { + "epoch": 0.06479978472409094, + "grad_norm": 0.6484375, + "learning_rate": 0.0019795160641068905, + "loss": 0.1211, + "step": 7465 + }, + { + "epoch": 0.0648084652042951, + "grad_norm": 0.171875, + "learning_rate": 0.0019795097499261884, + "loss": 0.1289, + "step": 7466 + }, + { + "epoch": 0.06481714568449927, + "grad_norm": 0.1640625, + "learning_rate": 0.0019795034347836675, + "loss": 0.127, + "step": 7467 + }, + { + "epoch": 0.06482582616470343, + "grad_norm": 0.08984375, + "learning_rate": 0.0019794971186793352, + "loss": 0.1348, + "step": 7468 + }, + { + "epoch": 0.0648345066449076, + "grad_norm": 0.205078125, + "learning_rate": 0.0019794908016131977, + "loss": 0.1582, + "step": 7469 + }, + { + "epoch": 0.06484318712511176, + "grad_norm": 0.1171875, + "learning_rate": 0.0019794844835852627, + "loss": 0.1738, + "step": 7470 + }, + { + "epoch": 0.06485186760531593, + "grad_norm": 0.10791015625, + "learning_rate": 0.001979478164595536, + "loss": 0.1318, + "step": 7471 + }, + { + "epoch": 0.0648605480855201, + "grad_norm": 0.142578125, + "learning_rate": 0.001979471844644026, + "loss": 0.1572, + "step": 7472 + }, + { + "epoch": 0.06486922856572426, + "grad_norm": 0.275390625, + "learning_rate": 0.0019794655237307386, + "loss": 0.1523, + "step": 7473 + }, + { + "epoch": 0.06487790904592843, + "grad_norm": 0.349609375, + "learning_rate": 0.001979459201855681, + "loss": 0.1328, + "step": 7474 + }, + { + "epoch": 0.06488658952613259, + "grad_norm": 0.26171875, + "learning_rate": 0.0019794528790188595, + "loss": 0.1445, + "step": 7475 + }, + { + "epoch": 0.06489527000633676, + "grad_norm": 0.111328125, + "learning_rate": 0.001979446555220282, + "loss": 0.1309, + "step": 7476 + }, + { + "epoch": 0.06490395048654092, + "grad_norm": 0.1845703125, + "learning_rate": 0.001979440230459955, + "loss": 0.1016, + "step": 7477 + }, + { + "epoch": 0.06491263096674509, + "grad_norm": 0.130859375, + "learning_rate": 0.001979433904737885, + "loss": 0.1426, + "step": 7478 + }, + { + "epoch": 0.06492131144694925, + "grad_norm": 0.6484375, + "learning_rate": 0.001979427578054079, + "loss": 0.1357, + "step": 7479 + }, + { + "epoch": 0.06492999192715342, + "grad_norm": 0.69140625, + "learning_rate": 0.001979421250408545, + "loss": 0.1504, + "step": 7480 + }, + { + "epoch": 0.06493867240735758, + "grad_norm": 0.357421875, + "learning_rate": 0.0019794149218012886, + "loss": 0.1118, + "step": 7481 + }, + { + "epoch": 0.06494735288756175, + "grad_norm": 0.296875, + "learning_rate": 0.0019794085922323176, + "loss": 0.1279, + "step": 7482 + }, + { + "epoch": 0.06495603336776591, + "grad_norm": 0.5390625, + "learning_rate": 0.0019794022617016386, + "loss": 0.1719, + "step": 7483 + }, + { + "epoch": 0.06496471384797008, + "grad_norm": 0.7265625, + "learning_rate": 0.0019793959302092583, + "loss": 0.1309, + "step": 7484 + }, + { + "epoch": 0.06497339432817423, + "grad_norm": 0.322265625, + "learning_rate": 0.0019793895977551837, + "loss": 0.1289, + "step": 7485 + }, + { + "epoch": 0.06498207480837839, + "grad_norm": 2.40625, + "learning_rate": 0.0019793832643394227, + "loss": 0.2021, + "step": 7486 + }, + { + "epoch": 0.06499075528858256, + "grad_norm": 0.09423828125, + "learning_rate": 0.00197937692996198, + "loss": 0.1484, + "step": 7487 + }, + { + "epoch": 0.06499943576878672, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019793705946228647, + "loss": 0.1396, + "step": 7488 + }, + { + "epoch": 0.06500811624899089, + "grad_norm": 0.462890625, + "learning_rate": 0.001979364258322083, + "loss": 0.1309, + "step": 7489 + }, + { + "epoch": 0.06501679672919505, + "grad_norm": 0.1103515625, + "learning_rate": 0.001979357921059642, + "loss": 0.1641, + "step": 7490 + }, + { + "epoch": 0.06502547720939922, + "grad_norm": 0.255859375, + "learning_rate": 0.001979351582835548, + "loss": 0.0869, + "step": 7491 + }, + { + "epoch": 0.06503415768960338, + "grad_norm": 0.4140625, + "learning_rate": 0.0019793452436498078, + "loss": 0.1465, + "step": 7492 + }, + { + "epoch": 0.06504283816980755, + "grad_norm": 0.40234375, + "learning_rate": 0.0019793389035024293, + "loss": 0.1299, + "step": 7493 + }, + { + "epoch": 0.06505151865001171, + "grad_norm": 0.51953125, + "learning_rate": 0.001979332562393419, + "loss": 0.1758, + "step": 7494 + }, + { + "epoch": 0.06506019913021588, + "grad_norm": 0.474609375, + "learning_rate": 0.0019793262203227843, + "loss": 0.127, + "step": 7495 + }, + { + "epoch": 0.06506887961042004, + "grad_norm": 0.58984375, + "learning_rate": 0.001979319877290531, + "loss": 0.0996, + "step": 7496 + }, + { + "epoch": 0.06507756009062421, + "grad_norm": 0.48046875, + "learning_rate": 0.001979313533296667, + "loss": 0.1426, + "step": 7497 + }, + { + "epoch": 0.06508624057082837, + "grad_norm": 0.2392578125, + "learning_rate": 0.001979307188341199, + "loss": 0.1543, + "step": 7498 + }, + { + "epoch": 0.06509492105103254, + "grad_norm": 0.068359375, + "learning_rate": 0.0019793008424241337, + "loss": 0.0962, + "step": 7499 + }, + { + "epoch": 0.0651036015312367, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019792944955454783, + "loss": 0.1211, + "step": 7500 + }, + { + "epoch": 0.06511228201144087, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019792881477052396, + "loss": 0.1465, + "step": 7501 + }, + { + "epoch": 0.06512096249164503, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019792817989034246, + "loss": 0.1748, + "step": 7502 + }, + { + "epoch": 0.0651296429718492, + "grad_norm": 1.0078125, + "learning_rate": 0.0019792754491400402, + "loss": 0.1318, + "step": 7503 + }, + { + "epoch": 0.06513832345205336, + "grad_norm": 0.33984375, + "learning_rate": 0.001979269098415093, + "loss": 0.1045, + "step": 7504 + }, + { + "epoch": 0.06514700393225753, + "grad_norm": 0.66015625, + "learning_rate": 0.001979262746728591, + "loss": 0.1416, + "step": 7505 + }, + { + "epoch": 0.0651556844124617, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019792563940805403, + "loss": 0.1348, + "step": 7506 + }, + { + "epoch": 0.06516436489266586, + "grad_norm": 0.07177734375, + "learning_rate": 0.001979250040470948, + "loss": 0.1621, + "step": 7507 + }, + { + "epoch": 0.06517304537287003, + "grad_norm": 0.5390625, + "learning_rate": 0.0019792436858998208, + "loss": 0.1406, + "step": 7508 + }, + { + "epoch": 0.06518172585307419, + "grad_norm": 0.201171875, + "learning_rate": 0.001979237330367166, + "loss": 0.1152, + "step": 7509 + }, + { + "epoch": 0.06519040633327836, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019792309738729907, + "loss": 0.1191, + "step": 7510 + }, + { + "epoch": 0.06519908681348252, + "grad_norm": 0.380859375, + "learning_rate": 0.0019792246164173012, + "loss": 0.1377, + "step": 7511 + }, + { + "epoch": 0.06520776729368669, + "grad_norm": 0.3828125, + "learning_rate": 0.0019792182580001053, + "loss": 0.1494, + "step": 7512 + }, + { + "epoch": 0.06521644777389085, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019792118986214093, + "loss": 0.165, + "step": 7513 + }, + { + "epoch": 0.06522512825409502, + "grad_norm": 0.2265625, + "learning_rate": 0.0019792055382812203, + "loss": 0.1377, + "step": 7514 + }, + { + "epoch": 0.06523380873429918, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019791991769795457, + "loss": 0.1455, + "step": 7515 + }, + { + "epoch": 0.06524248921450335, + "grad_norm": 0.73828125, + "learning_rate": 0.001979192814716392, + "loss": 0.1465, + "step": 7516 + }, + { + "epoch": 0.06525116969470751, + "grad_norm": 0.1455078125, + "learning_rate": 0.001979186451491766, + "loss": 0.1357, + "step": 7517 + }, + { + "epoch": 0.06525985017491168, + "grad_norm": 0.09765625, + "learning_rate": 0.001979180087305675, + "loss": 0.1602, + "step": 7518 + }, + { + "epoch": 0.06526853065511584, + "grad_norm": 0.12451171875, + "learning_rate": 0.001979173722158126, + "loss": 0.1201, + "step": 7519 + }, + { + "epoch": 0.06527721113532, + "grad_norm": 0.80078125, + "learning_rate": 0.0019791673560491255, + "loss": 0.1367, + "step": 7520 + }, + { + "epoch": 0.06528589161552417, + "grad_norm": 0.640625, + "learning_rate": 0.0019791609889786813, + "loss": 0.126, + "step": 7521 + }, + { + "epoch": 0.06529457209572834, + "grad_norm": 0.283203125, + "learning_rate": 0.0019791546209467995, + "loss": 0.1079, + "step": 7522 + }, + { + "epoch": 0.0653032525759325, + "grad_norm": 0.283203125, + "learning_rate": 0.001979148251953487, + "loss": 0.1089, + "step": 7523 + }, + { + "epoch": 0.06531193305613667, + "grad_norm": 0.6796875, + "learning_rate": 0.001979141881998752, + "loss": 0.1602, + "step": 7524 + }, + { + "epoch": 0.06532061353634083, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019791355110826, + "loss": 0.1445, + "step": 7525 + }, + { + "epoch": 0.065329294016545, + "grad_norm": 0.37890625, + "learning_rate": 0.001979129139205039, + "loss": 0.2383, + "step": 7526 + }, + { + "epoch": 0.06533797449674916, + "grad_norm": 0.12890625, + "learning_rate": 0.001979122766366076, + "loss": 0.1602, + "step": 7527 + }, + { + "epoch": 0.06534665497695333, + "grad_norm": 1.015625, + "learning_rate": 0.001979116392565717, + "loss": 0.1201, + "step": 7528 + }, + { + "epoch": 0.06535533545715749, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019791100178039693, + "loss": 0.1387, + "step": 7529 + }, + { + "epoch": 0.06536401593736166, + "grad_norm": 0.65625, + "learning_rate": 0.00197910364208084, + "loss": 0.1494, + "step": 7530 + }, + { + "epoch": 0.06537269641756582, + "grad_norm": 0.083984375, + "learning_rate": 0.001979097265396337, + "loss": 0.1006, + "step": 7531 + }, + { + "epoch": 0.06538137689776999, + "grad_norm": 0.19921875, + "learning_rate": 0.0019790908877504658, + "loss": 0.1367, + "step": 7532 + }, + { + "epoch": 0.06539005737797415, + "grad_norm": 0.232421875, + "learning_rate": 0.0019790845091432344, + "loss": 0.1279, + "step": 7533 + }, + { + "epoch": 0.06539873785817832, + "grad_norm": 0.255859375, + "learning_rate": 0.001979078129574649, + "loss": 0.0898, + "step": 7534 + }, + { + "epoch": 0.06540741833838248, + "grad_norm": 0.1953125, + "learning_rate": 0.001979071749044717, + "loss": 0.1445, + "step": 7535 + }, + { + "epoch": 0.06541609881858665, + "grad_norm": 0.25390625, + "learning_rate": 0.0019790653675534455, + "loss": 0.2109, + "step": 7536 + }, + { + "epoch": 0.06542477929879081, + "grad_norm": 0.55078125, + "learning_rate": 0.0019790589851008413, + "loss": 0.1426, + "step": 7537 + }, + { + "epoch": 0.06543345977899498, + "grad_norm": 0.291015625, + "learning_rate": 0.0019790526016869115, + "loss": 0.1406, + "step": 7538 + }, + { + "epoch": 0.06544214025919914, + "grad_norm": 0.400390625, + "learning_rate": 0.001979046217311663, + "loss": 0.1064, + "step": 7539 + }, + { + "epoch": 0.06545082073940331, + "grad_norm": 0.15234375, + "learning_rate": 0.0019790398319751027, + "loss": 0.1484, + "step": 7540 + }, + { + "epoch": 0.06545950121960747, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019790334456772376, + "loss": 0.1562, + "step": 7541 + }, + { + "epoch": 0.06546818169981164, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019790270584180746, + "loss": 0.1611, + "step": 7542 + }, + { + "epoch": 0.0654768621800158, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019790206701976207, + "loss": 0.1299, + "step": 7543 + }, + { + "epoch": 0.06548554266021997, + "grad_norm": 0.275390625, + "learning_rate": 0.001979014281015883, + "loss": 0.1152, + "step": 7544 + }, + { + "epoch": 0.06549422314042413, + "grad_norm": 0.765625, + "learning_rate": 0.001979007890872869, + "loss": 0.125, + "step": 7545 + }, + { + "epoch": 0.0655029036206283, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019790014997685852, + "loss": 0.1318, + "step": 7546 + }, + { + "epoch": 0.06551158410083245, + "grad_norm": 0.1396484375, + "learning_rate": 0.001978995107703038, + "loss": 0.1533, + "step": 7547 + }, + { + "epoch": 0.06552026458103662, + "grad_norm": 0.7890625, + "learning_rate": 0.0019789887146762354, + "loss": 0.1182, + "step": 7548 + }, + { + "epoch": 0.06552894506124078, + "grad_norm": 0.28515625, + "learning_rate": 0.0019789823206881837, + "loss": 0.1553, + "step": 7549 + }, + { + "epoch": 0.06553762554144495, + "grad_norm": 0.083984375, + "learning_rate": 0.0019789759257388905, + "loss": 0.1299, + "step": 7550 + }, + { + "epoch": 0.06554630602164911, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019789695298283623, + "loss": 0.1416, + "step": 7551 + }, + { + "epoch": 0.06555498650185328, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019789631329566056, + "loss": 0.1465, + "step": 7552 + }, + { + "epoch": 0.06556366698205744, + "grad_norm": 0.494140625, + "learning_rate": 0.0019789567351236287, + "loss": 0.1064, + "step": 7553 + }, + { + "epoch": 0.06557234746226161, + "grad_norm": 0.21484375, + "learning_rate": 0.0019789503363294375, + "loss": 0.123, + "step": 7554 + }, + { + "epoch": 0.06558102794246577, + "grad_norm": 0.451171875, + "learning_rate": 0.0019789439365740396, + "loss": 0.1182, + "step": 7555 + }, + { + "epoch": 0.06558970842266994, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019789375358574422, + "loss": 0.1729, + "step": 7556 + }, + { + "epoch": 0.0655983889028741, + "grad_norm": 0.4140625, + "learning_rate": 0.0019789311341796515, + "loss": 0.1162, + "step": 7557 + }, + { + "epoch": 0.06560706938307827, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019789247315406752, + "loss": 0.1289, + "step": 7558 + }, + { + "epoch": 0.06561574986328243, + "grad_norm": 0.60546875, + "learning_rate": 0.00197891832794052, + "loss": 0.1167, + "step": 7559 + }, + { + "epoch": 0.0656244303434866, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019789119233791924, + "loss": 0.1084, + "step": 7560 + }, + { + "epoch": 0.06563311082369076, + "grad_norm": 0.275390625, + "learning_rate": 0.0019789055178567002, + "loss": 0.1504, + "step": 7561 + }, + { + "epoch": 0.06564179130389493, + "grad_norm": 0.2578125, + "learning_rate": 0.00197889911137305, + "loss": 0.1133, + "step": 7562 + }, + { + "epoch": 0.0656504717840991, + "grad_norm": 0.380859375, + "learning_rate": 0.0019788927039282493, + "loss": 0.1143, + "step": 7563 + }, + { + "epoch": 0.06565915226430326, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019788862955223046, + "loss": 0.1035, + "step": 7564 + }, + { + "epoch": 0.06566783274450742, + "grad_norm": 0.1806640625, + "learning_rate": 0.001978879886155223, + "loss": 0.1348, + "step": 7565 + }, + { + "epoch": 0.06567651322471159, + "grad_norm": 0.11181640625, + "learning_rate": 0.001978873475827012, + "loss": 0.1738, + "step": 7566 + }, + { + "epoch": 0.06568519370491575, + "grad_norm": 0.201171875, + "learning_rate": 0.0019788670645376773, + "loss": 0.1377, + "step": 7567 + }, + { + "epoch": 0.06569387418511992, + "grad_norm": 0.306640625, + "learning_rate": 0.0019788606522872275, + "loss": 0.125, + "step": 7568 + }, + { + "epoch": 0.06570255466532408, + "grad_norm": 0.0869140625, + "learning_rate": 0.001978854239075669, + "loss": 0.1133, + "step": 7569 + }, + { + "epoch": 0.06571123514552825, + "grad_norm": 0.3046875, + "learning_rate": 0.001978847824903008, + "loss": 0.1445, + "step": 7570 + }, + { + "epoch": 0.06571991562573241, + "grad_norm": 0.396484375, + "learning_rate": 0.0019788414097692527, + "loss": 0.126, + "step": 7571 + }, + { + "epoch": 0.06572859610593658, + "grad_norm": 0.1484375, + "learning_rate": 0.0019788349936744093, + "loss": 0.1221, + "step": 7572 + }, + { + "epoch": 0.06573727658614074, + "grad_norm": 0.09765625, + "learning_rate": 0.0019788285766184853, + "loss": 0.1367, + "step": 7573 + }, + { + "epoch": 0.06574595706634491, + "grad_norm": 0.142578125, + "learning_rate": 0.0019788221586014877, + "loss": 0.1729, + "step": 7574 + }, + { + "epoch": 0.06575463754654907, + "grad_norm": 0.255859375, + "learning_rate": 0.0019788157396234234, + "loss": 0.1592, + "step": 7575 + }, + { + "epoch": 0.06576331802675324, + "grad_norm": 0.41015625, + "learning_rate": 0.001978809319684299, + "loss": 0.2061, + "step": 7576 + }, + { + "epoch": 0.0657719985069574, + "grad_norm": 0.283203125, + "learning_rate": 0.0019788028987841226, + "loss": 0.1699, + "step": 7577 + }, + { + "epoch": 0.06578067898716157, + "grad_norm": 0.083984375, + "learning_rate": 0.0019787964769229, + "loss": 0.1533, + "step": 7578 + }, + { + "epoch": 0.06578935946736574, + "grad_norm": 0.068359375, + "learning_rate": 0.0019787900541006396, + "loss": 0.1064, + "step": 7579 + }, + { + "epoch": 0.0657980399475699, + "grad_norm": 0.3671875, + "learning_rate": 0.001978783630317347, + "loss": 0.1123, + "step": 7580 + }, + { + "epoch": 0.06580672042777407, + "grad_norm": 0.21484375, + "learning_rate": 0.0019787772055730297, + "loss": 0.1172, + "step": 7581 + }, + { + "epoch": 0.06581540090797823, + "grad_norm": 0.30078125, + "learning_rate": 0.0019787707798676946, + "loss": 0.1279, + "step": 7582 + }, + { + "epoch": 0.0658240813881824, + "grad_norm": 0.154296875, + "learning_rate": 0.0019787643532013497, + "loss": 0.1582, + "step": 7583 + }, + { + "epoch": 0.06583276186838656, + "grad_norm": 0.06298828125, + "learning_rate": 0.001978757925574001, + "loss": 0.1172, + "step": 7584 + }, + { + "epoch": 0.06584144234859073, + "grad_norm": 0.6875, + "learning_rate": 0.0019787514969856557, + "loss": 0.1328, + "step": 7585 + }, + { + "epoch": 0.06585012282879489, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019787450674363214, + "loss": 0.1309, + "step": 7586 + }, + { + "epoch": 0.06585880330899906, + "grad_norm": 0.1982421875, + "learning_rate": 0.001978738636926004, + "loss": 0.1133, + "step": 7587 + }, + { + "epoch": 0.06586748378920322, + "grad_norm": 0.275390625, + "learning_rate": 0.001978732205454712, + "loss": 0.1162, + "step": 7588 + }, + { + "epoch": 0.06587616426940739, + "grad_norm": 0.2236328125, + "learning_rate": 0.001978725773022451, + "loss": 0.126, + "step": 7589 + }, + { + "epoch": 0.06588484474961155, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019787193396292287, + "loss": 0.207, + "step": 7590 + }, + { + "epoch": 0.06589352522981572, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019787129052750525, + "loss": 0.1445, + "step": 7591 + }, + { + "epoch": 0.06590220571001988, + "grad_norm": 0.40625, + "learning_rate": 0.0019787064699599293, + "loss": 0.1465, + "step": 7592 + }, + { + "epoch": 0.06591088619022405, + "grad_norm": 0.1318359375, + "learning_rate": 0.001978700033683865, + "loss": 0.1357, + "step": 7593 + }, + { + "epoch": 0.06591956667042821, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019786935964468686, + "loss": 0.1211, + "step": 7594 + }, + { + "epoch": 0.06592824715063238, + "grad_norm": 0.126953125, + "learning_rate": 0.0019786871582489454, + "loss": 0.1318, + "step": 7595 + }, + { + "epoch": 0.06593692763083654, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019786807190901035, + "loss": 0.1602, + "step": 7596 + }, + { + "epoch": 0.06594560811104071, + "grad_norm": 1.328125, + "learning_rate": 0.001978674278970349, + "loss": 0.1523, + "step": 7597 + }, + { + "epoch": 0.06595428859124487, + "grad_norm": 0.31640625, + "learning_rate": 0.0019786678378896903, + "loss": 0.1064, + "step": 7598 + }, + { + "epoch": 0.06596296907144904, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019786613958481334, + "loss": 0.1328, + "step": 7599 + }, + { + "epoch": 0.0659716495516532, + "grad_norm": 0.11376953125, + "learning_rate": 0.001978654952845685, + "loss": 0.1514, + "step": 7600 + }, + { + "epoch": 0.06598033003185737, + "grad_norm": 0.103515625, + "learning_rate": 0.0019786485088823537, + "loss": 0.1069, + "step": 7601 + }, + { + "epoch": 0.06598901051206153, + "grad_norm": 0.10791015625, + "learning_rate": 0.001978642063958145, + "loss": 0.1113, + "step": 7602 + }, + { + "epoch": 0.0659976909922657, + "grad_norm": 0.4375, + "learning_rate": 0.0019786356180730665, + "loss": 0.0977, + "step": 7603 + }, + { + "epoch": 0.06600637147246986, + "grad_norm": 0.56640625, + "learning_rate": 0.0019786291712271253, + "loss": 0.127, + "step": 7604 + }, + { + "epoch": 0.06601505195267403, + "grad_norm": 1.8046875, + "learning_rate": 0.0019786227234203286, + "loss": 0.1465, + "step": 7605 + }, + { + "epoch": 0.0660237324328782, + "grad_norm": 0.2734375, + "learning_rate": 0.0019786162746526833, + "loss": 0.1348, + "step": 7606 + }, + { + "epoch": 0.06603241291308236, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019786098249241963, + "loss": 0.1396, + "step": 7607 + }, + { + "epoch": 0.06604109339328652, + "grad_norm": 0.5234375, + "learning_rate": 0.001978603374234875, + "loss": 0.1006, + "step": 7608 + }, + { + "epoch": 0.06604977387349067, + "grad_norm": 0.28515625, + "learning_rate": 0.001978596922584726, + "loss": 0.1631, + "step": 7609 + }, + { + "epoch": 0.06605845435369484, + "grad_norm": 0.2197265625, + "learning_rate": 0.001978590469973757, + "loss": 0.1191, + "step": 7610 + }, + { + "epoch": 0.066067134833899, + "grad_norm": 0.41015625, + "learning_rate": 0.001978584016401974, + "loss": 0.166, + "step": 7611 + }, + { + "epoch": 0.06607581531410317, + "grad_norm": 1.34375, + "learning_rate": 0.0019785775618693852, + "loss": 0.1523, + "step": 7612 + }, + { + "epoch": 0.06608449579430734, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019785711063759976, + "loss": 0.1543, + "step": 7613 + }, + { + "epoch": 0.0660931762745115, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019785646499218172, + "loss": 0.123, + "step": 7614 + }, + { + "epoch": 0.06610185675471567, + "grad_norm": 0.5078125, + "learning_rate": 0.0019785581925068517, + "loss": 0.1523, + "step": 7615 + }, + { + "epoch": 0.06611053723491983, + "grad_norm": 0.23046875, + "learning_rate": 0.0019785517341311086, + "loss": 0.1143, + "step": 7616 + }, + { + "epoch": 0.066119217715124, + "grad_norm": 0.0830078125, + "learning_rate": 0.001978545274794594, + "loss": 0.1328, + "step": 7617 + }, + { + "epoch": 0.06612789819532816, + "grad_norm": 0.58984375, + "learning_rate": 0.0019785388144973157, + "loss": 0.1211, + "step": 7618 + }, + { + "epoch": 0.06613657867553233, + "grad_norm": 0.380859375, + "learning_rate": 0.0019785323532392806, + "loss": 0.1128, + "step": 7619 + }, + { + "epoch": 0.06614525915573649, + "grad_norm": 0.2578125, + "learning_rate": 0.0019785258910204957, + "loss": 0.105, + "step": 7620 + }, + { + "epoch": 0.06615393963594066, + "grad_norm": 0.419921875, + "learning_rate": 0.001978519427840968, + "loss": 0.125, + "step": 7621 + }, + { + "epoch": 0.06616262011614482, + "grad_norm": 0.43359375, + "learning_rate": 0.0019785129637007047, + "loss": 0.1309, + "step": 7622 + }, + { + "epoch": 0.06617130059634899, + "grad_norm": 0.275390625, + "learning_rate": 0.001978506498599713, + "loss": 0.1514, + "step": 7623 + }, + { + "epoch": 0.06617998107655315, + "grad_norm": 0.255859375, + "learning_rate": 0.001978500032538, + "loss": 0.0938, + "step": 7624 + }, + { + "epoch": 0.06618866155675732, + "grad_norm": 0.55859375, + "learning_rate": 0.001978493565515572, + "loss": 0.1426, + "step": 7625 + }, + { + "epoch": 0.06619734203696148, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019784870975324367, + "loss": 0.1172, + "step": 7626 + }, + { + "epoch": 0.06620602251716565, + "grad_norm": 0.2734375, + "learning_rate": 0.0019784806285886012, + "loss": 0.125, + "step": 7627 + }, + { + "epoch": 0.06621470299736981, + "grad_norm": 0.2333984375, + "learning_rate": 0.001978474158684073, + "loss": 0.1436, + "step": 7628 + }, + { + "epoch": 0.06622338347757398, + "grad_norm": 0.3828125, + "learning_rate": 0.0019784676878188577, + "loss": 0.1123, + "step": 7629 + }, + { + "epoch": 0.06623206395777814, + "grad_norm": 0.60546875, + "learning_rate": 0.001978461215992964, + "loss": 0.1445, + "step": 7630 + }, + { + "epoch": 0.06624074443798231, + "grad_norm": 0.2216796875, + "learning_rate": 0.001978454743206398, + "loss": 0.1406, + "step": 7631 + }, + { + "epoch": 0.06624942491818647, + "grad_norm": 0.345703125, + "learning_rate": 0.0019784482694591675, + "loss": 0.1523, + "step": 7632 + }, + { + "epoch": 0.06625810539839064, + "grad_norm": 0.388671875, + "learning_rate": 0.0019784417947512787, + "loss": 0.0986, + "step": 7633 + }, + { + "epoch": 0.0662667858785948, + "grad_norm": 0.27734375, + "learning_rate": 0.0019784353190827396, + "loss": 0.1895, + "step": 7634 + }, + { + "epoch": 0.06627546635879897, + "grad_norm": 0.369140625, + "learning_rate": 0.001978428842453557, + "loss": 0.1289, + "step": 7635 + }, + { + "epoch": 0.06628414683900313, + "grad_norm": 0.1171875, + "learning_rate": 0.001978422364863737, + "loss": 0.1445, + "step": 7636 + }, + { + "epoch": 0.0662928273192073, + "grad_norm": 0.078125, + "learning_rate": 0.001978415886313288, + "loss": 0.1099, + "step": 7637 + }, + { + "epoch": 0.06630150779941146, + "grad_norm": 0.255859375, + "learning_rate": 0.0019784094068022165, + "loss": 0.1006, + "step": 7638 + }, + { + "epoch": 0.06631018827961563, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019784029263305294, + "loss": 0.1221, + "step": 7639 + }, + { + "epoch": 0.0663188687598198, + "grad_norm": 0.09375, + "learning_rate": 0.0019783964448982345, + "loss": 0.1143, + "step": 7640 + }, + { + "epoch": 0.06632754924002396, + "grad_norm": 0.33984375, + "learning_rate": 0.001978389962505338, + "loss": 0.1631, + "step": 7641 + }, + { + "epoch": 0.06633622972022812, + "grad_norm": 0.08349609375, + "learning_rate": 0.001978383479151848, + "loss": 0.1318, + "step": 7642 + }, + { + "epoch": 0.06634491020043229, + "grad_norm": 0.32421875, + "learning_rate": 0.0019783769948377708, + "loss": 0.1299, + "step": 7643 + }, + { + "epoch": 0.06635359068063645, + "grad_norm": 0.4921875, + "learning_rate": 0.0019783705095631136, + "loss": 0.1621, + "step": 7644 + }, + { + "epoch": 0.06636227116084062, + "grad_norm": 0.345703125, + "learning_rate": 0.0019783640233278837, + "loss": 0.1416, + "step": 7645 + }, + { + "epoch": 0.06637095164104478, + "grad_norm": 0.1591796875, + "learning_rate": 0.001978357536132088, + "loss": 0.0996, + "step": 7646 + }, + { + "epoch": 0.06637963212124895, + "grad_norm": 1.1484375, + "learning_rate": 0.001978351047975734, + "loss": 0.1836, + "step": 7647 + }, + { + "epoch": 0.06638831260145311, + "grad_norm": 0.10791015625, + "learning_rate": 0.001978344558858828, + "loss": 0.1084, + "step": 7648 + }, + { + "epoch": 0.06639699308165728, + "grad_norm": 0.11767578125, + "learning_rate": 0.001978338068781378, + "loss": 0.1445, + "step": 7649 + }, + { + "epoch": 0.06640567356186144, + "grad_norm": 0.72265625, + "learning_rate": 0.0019783315777433903, + "loss": 0.0952, + "step": 7650 + }, + { + "epoch": 0.06641435404206561, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019783250857448725, + "loss": 0.125, + "step": 7651 + }, + { + "epoch": 0.06642303452226977, + "grad_norm": 0.42578125, + "learning_rate": 0.001978318592785832, + "loss": 0.104, + "step": 7652 + }, + { + "epoch": 0.06643171500247394, + "grad_norm": 0.412109375, + "learning_rate": 0.001978312098866275, + "loss": 0.2061, + "step": 7653 + }, + { + "epoch": 0.0664403954826781, + "grad_norm": 0.28515625, + "learning_rate": 0.0019783056039862092, + "loss": 0.1514, + "step": 7654 + }, + { + "epoch": 0.06644907596288227, + "grad_norm": 0.14453125, + "learning_rate": 0.001978299108145642, + "loss": 0.1445, + "step": 7655 + }, + { + "epoch": 0.06645775644308644, + "grad_norm": 0.34375, + "learning_rate": 0.0019782926113445795, + "loss": 0.124, + "step": 7656 + }, + { + "epoch": 0.0664664369232906, + "grad_norm": 0.1318359375, + "learning_rate": 0.00197828611358303, + "loss": 0.1543, + "step": 7657 + }, + { + "epoch": 0.06647511740349477, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019782796148609992, + "loss": 0.1592, + "step": 7658 + }, + { + "epoch": 0.06648379788369893, + "grad_norm": 0.83984375, + "learning_rate": 0.0019782731151784957, + "loss": 0.1309, + "step": 7659 + }, + { + "epoch": 0.0664924783639031, + "grad_norm": 0.251953125, + "learning_rate": 0.0019782666145355256, + "loss": 0.1592, + "step": 7660 + }, + { + "epoch": 0.06650115884410726, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019782601129320963, + "loss": 0.1729, + "step": 7661 + }, + { + "epoch": 0.06650983932431143, + "grad_norm": 0.66015625, + "learning_rate": 0.001978253610368215, + "loss": 0.1475, + "step": 7662 + }, + { + "epoch": 0.06651851980451559, + "grad_norm": 0.6328125, + "learning_rate": 0.0019782471068438886, + "loss": 0.123, + "step": 7663 + }, + { + "epoch": 0.06652720028471976, + "grad_norm": 0.283203125, + "learning_rate": 0.001978240602359125, + "loss": 0.1025, + "step": 7664 + }, + { + "epoch": 0.06653588076492392, + "grad_norm": 0.93359375, + "learning_rate": 0.0019782340969139297, + "loss": 0.1914, + "step": 7665 + }, + { + "epoch": 0.06654456124512809, + "grad_norm": 0.46875, + "learning_rate": 0.0019782275905083115, + "loss": 0.1338, + "step": 7666 + }, + { + "epoch": 0.06655324172533225, + "grad_norm": 0.212890625, + "learning_rate": 0.0019782210831422765, + "loss": 0.127, + "step": 7667 + }, + { + "epoch": 0.06656192220553642, + "grad_norm": 1.09375, + "learning_rate": 0.0019782145748158324, + "loss": 0.209, + "step": 7668 + }, + { + "epoch": 0.06657060268574058, + "grad_norm": 1.1328125, + "learning_rate": 0.001978208065528986, + "loss": 0.1494, + "step": 7669 + }, + { + "epoch": 0.06657928316594473, + "grad_norm": 0.8828125, + "learning_rate": 0.001978201555281744, + "loss": 0.1758, + "step": 7670 + }, + { + "epoch": 0.0665879636461489, + "grad_norm": 1.4140625, + "learning_rate": 0.0019781950440741145, + "loss": 0.1494, + "step": 7671 + }, + { + "epoch": 0.06659664412635306, + "grad_norm": 0.50390625, + "learning_rate": 0.001978188531906104, + "loss": 0.1416, + "step": 7672 + }, + { + "epoch": 0.06660532460655723, + "grad_norm": 2.859375, + "learning_rate": 0.001978182018777719, + "loss": 0.1719, + "step": 7673 + }, + { + "epoch": 0.0666140050867614, + "grad_norm": 0.06640625, + "learning_rate": 0.001978175504688968, + "loss": 0.127, + "step": 7674 + }, + { + "epoch": 0.06662268556696556, + "grad_norm": 0.58984375, + "learning_rate": 0.0019781689896398577, + "loss": 0.1982, + "step": 7675 + }, + { + "epoch": 0.06663136604716972, + "grad_norm": 0.20703125, + "learning_rate": 0.0019781624736303943, + "loss": 0.1123, + "step": 7676 + }, + { + "epoch": 0.06664004652737389, + "grad_norm": 0.87890625, + "learning_rate": 0.0019781559566605864, + "loss": 0.1328, + "step": 7677 + }, + { + "epoch": 0.06664872700757805, + "grad_norm": 0.48046875, + "learning_rate": 0.00197814943873044, + "loss": 0.1562, + "step": 7678 + }, + { + "epoch": 0.06665740748778222, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019781429198399625, + "loss": 0.1387, + "step": 7679 + }, + { + "epoch": 0.06666608796798638, + "grad_norm": 0.76171875, + "learning_rate": 0.001978136399989161, + "loss": 0.1504, + "step": 7680 + }, + { + "epoch": 0.06667476844819055, + "grad_norm": 0.171875, + "learning_rate": 0.0019781298791780428, + "loss": 0.1318, + "step": 7681 + }, + { + "epoch": 0.06668344892839471, + "grad_norm": 0.09375, + "learning_rate": 0.0019781233574066148, + "loss": 0.168, + "step": 7682 + }, + { + "epoch": 0.06669212940859888, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019781168346748846, + "loss": 0.0918, + "step": 7683 + }, + { + "epoch": 0.06670080988880304, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019781103109828588, + "loss": 0.126, + "step": 7684 + }, + { + "epoch": 0.06670949036900721, + "grad_norm": 0.251953125, + "learning_rate": 0.0019781037863305447, + "loss": 0.1074, + "step": 7685 + }, + { + "epoch": 0.06671817084921138, + "grad_norm": 0.09423828125, + "learning_rate": 0.00197809726071795, + "loss": 0.1143, + "step": 7686 + }, + { + "epoch": 0.06672685132941554, + "grad_norm": 0.2578125, + "learning_rate": 0.0019780907341450807, + "loss": 0.1045, + "step": 7687 + }, + { + "epoch": 0.0667355318096197, + "grad_norm": 0.3125, + "learning_rate": 0.0019780842066119447, + "loss": 0.1123, + "step": 7688 + }, + { + "epoch": 0.06674421228982387, + "grad_norm": 0.8671875, + "learning_rate": 0.0019780776781185496, + "loss": 0.1924, + "step": 7689 + }, + { + "epoch": 0.06675289277002804, + "grad_norm": 0.71875, + "learning_rate": 0.0019780711486649012, + "loss": 0.252, + "step": 7690 + }, + { + "epoch": 0.0667615732502322, + "grad_norm": 0.208984375, + "learning_rate": 0.001978064618251008, + "loss": 0.1426, + "step": 7691 + }, + { + "epoch": 0.06677025373043637, + "grad_norm": 0.236328125, + "learning_rate": 0.001978058086876876, + "loss": 0.1396, + "step": 7692 + }, + { + "epoch": 0.06677893421064053, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019780515545425132, + "loss": 0.125, + "step": 7693 + }, + { + "epoch": 0.0667876146908447, + "grad_norm": 0.0986328125, + "learning_rate": 0.001978045021247926, + "loss": 0.1621, + "step": 7694 + }, + { + "epoch": 0.06679629517104886, + "grad_norm": 0.07568359375, + "learning_rate": 0.001978038486993122, + "loss": 0.1504, + "step": 7695 + }, + { + "epoch": 0.06680497565125303, + "grad_norm": 0.6484375, + "learning_rate": 0.0019780319517781085, + "loss": 0.1377, + "step": 7696 + }, + { + "epoch": 0.06681365613145719, + "grad_norm": 0.28515625, + "learning_rate": 0.001978025415602893, + "loss": 0.1465, + "step": 7697 + }, + { + "epoch": 0.06682233661166136, + "grad_norm": 0.08984375, + "learning_rate": 0.0019780188784674817, + "loss": 0.1621, + "step": 7698 + }, + { + "epoch": 0.06683101709186552, + "grad_norm": 0.40625, + "learning_rate": 0.0019780123403718816, + "loss": 0.1387, + "step": 7699 + }, + { + "epoch": 0.06683969757206969, + "grad_norm": 0.69921875, + "learning_rate": 0.001978005801316101, + "loss": 0.1484, + "step": 7700 + }, + { + "epoch": 0.06684837805227385, + "grad_norm": 0.6328125, + "learning_rate": 0.0019779992613001463, + "loss": 0.1279, + "step": 7701 + }, + { + "epoch": 0.06685705853247802, + "grad_norm": 2.65625, + "learning_rate": 0.001977992720324025, + "loss": 0.3203, + "step": 7702 + }, + { + "epoch": 0.06686573901268218, + "grad_norm": 0.35546875, + "learning_rate": 0.0019779861783877435, + "loss": 0.1611, + "step": 7703 + }, + { + "epoch": 0.06687441949288635, + "grad_norm": 0.11083984375, + "learning_rate": 0.00197797963549131, + "loss": 0.0977, + "step": 7704 + }, + { + "epoch": 0.06688309997309051, + "grad_norm": 0.66015625, + "learning_rate": 0.001977973091634731, + "loss": 0.125, + "step": 7705 + }, + { + "epoch": 0.06689178045329468, + "grad_norm": 0.52734375, + "learning_rate": 0.001977966546818014, + "loss": 0.1377, + "step": 7706 + }, + { + "epoch": 0.06690046093349884, + "grad_norm": 0.8671875, + "learning_rate": 0.0019779600010411654, + "loss": 0.2148, + "step": 7707 + }, + { + "epoch": 0.06690914141370301, + "grad_norm": 0.12890625, + "learning_rate": 0.0019779534543041934, + "loss": 0.1699, + "step": 7708 + }, + { + "epoch": 0.06691782189390717, + "grad_norm": 1.28125, + "learning_rate": 0.001977946906607105, + "loss": 0.1299, + "step": 7709 + }, + { + "epoch": 0.06692650237411134, + "grad_norm": 1.140625, + "learning_rate": 0.0019779403579499063, + "loss": 0.168, + "step": 7710 + }, + { + "epoch": 0.0669351828543155, + "grad_norm": 0.267578125, + "learning_rate": 0.0019779338083326056, + "loss": 0.1816, + "step": 7711 + }, + { + "epoch": 0.06694386333451967, + "grad_norm": 0.87890625, + "learning_rate": 0.0019779272577552095, + "loss": 0.1289, + "step": 7712 + }, + { + "epoch": 0.06695254381472383, + "grad_norm": 0.1826171875, + "learning_rate": 0.001977920706217726, + "loss": 0.1279, + "step": 7713 + }, + { + "epoch": 0.066961224294928, + "grad_norm": 0.439453125, + "learning_rate": 0.0019779141537201606, + "loss": 0.1709, + "step": 7714 + }, + { + "epoch": 0.06696990477513216, + "grad_norm": 0.58984375, + "learning_rate": 0.001977907600262522, + "loss": 0.1982, + "step": 7715 + }, + { + "epoch": 0.06697858525533633, + "grad_norm": 0.33984375, + "learning_rate": 0.001977901045844817, + "loss": 0.1445, + "step": 7716 + }, + { + "epoch": 0.0669872657355405, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019778944904670524, + "loss": 0.1143, + "step": 7717 + }, + { + "epoch": 0.06699594621574466, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019778879341292356, + "loss": 0.1235, + "step": 7718 + }, + { + "epoch": 0.06700462669594882, + "grad_norm": 0.201171875, + "learning_rate": 0.0019778813768313737, + "loss": 0.1475, + "step": 7719 + }, + { + "epoch": 0.06701330717615299, + "grad_norm": 0.439453125, + "learning_rate": 0.001977874818573474, + "loss": 0.1328, + "step": 7720 + }, + { + "epoch": 0.06702198765635715, + "grad_norm": 0.111328125, + "learning_rate": 0.001977868259355544, + "loss": 0.1104, + "step": 7721 + }, + { + "epoch": 0.06703066813656132, + "grad_norm": 0.486328125, + "learning_rate": 0.0019778616991775896, + "loss": 0.1162, + "step": 7722 + }, + { + "epoch": 0.06703934861676548, + "grad_norm": 0.2890625, + "learning_rate": 0.0019778551380396194, + "loss": 0.1182, + "step": 7723 + }, + { + "epoch": 0.06704802909696965, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019778485759416402, + "loss": 0.2051, + "step": 7724 + }, + { + "epoch": 0.06705670957717381, + "grad_norm": 0.099609375, + "learning_rate": 0.001977842012883658, + "loss": 0.1299, + "step": 7725 + }, + { + "epoch": 0.06706539005737798, + "grad_norm": 1.015625, + "learning_rate": 0.001977835448865682, + "loss": 0.2285, + "step": 7726 + }, + { + "epoch": 0.06707407053758214, + "grad_norm": 0.197265625, + "learning_rate": 0.001977828883887718, + "loss": 0.126, + "step": 7727 + }, + { + "epoch": 0.06708275101778631, + "grad_norm": 0.24609375, + "learning_rate": 0.0019778223179497735, + "loss": 0.1455, + "step": 7728 + }, + { + "epoch": 0.06709143149799048, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019778157510518556, + "loss": 0.125, + "step": 7729 + }, + { + "epoch": 0.06710011197819464, + "grad_norm": 0.142578125, + "learning_rate": 0.001977809183193972, + "loss": 0.1235, + "step": 7730 + }, + { + "epoch": 0.0671087924583988, + "grad_norm": 0.2470703125, + "learning_rate": 0.001977802614376129, + "loss": 0.1699, + "step": 7731 + }, + { + "epoch": 0.06711747293860296, + "grad_norm": 0.35546875, + "learning_rate": 0.001977796044598334, + "loss": 0.1348, + "step": 7732 + }, + { + "epoch": 0.06712615341880712, + "grad_norm": 0.21875, + "learning_rate": 0.001977789473860595, + "loss": 0.1025, + "step": 7733 + }, + { + "epoch": 0.06713483389901129, + "grad_norm": 0.3203125, + "learning_rate": 0.001977782902162918, + "loss": 0.1387, + "step": 7734 + }, + { + "epoch": 0.06714351437921545, + "grad_norm": 0.40625, + "learning_rate": 0.0019777763295053116, + "loss": 0.1001, + "step": 7735 + }, + { + "epoch": 0.06715219485941962, + "grad_norm": 0.205078125, + "learning_rate": 0.001977769755887782, + "loss": 0.1104, + "step": 7736 + }, + { + "epoch": 0.06716087533962378, + "grad_norm": 0.42578125, + "learning_rate": 0.001977763181310336, + "loss": 0.1582, + "step": 7737 + }, + { + "epoch": 0.06716955581982795, + "grad_norm": 0.21484375, + "learning_rate": 0.001977756605772982, + "loss": 0.103, + "step": 7738 + }, + { + "epoch": 0.06717823630003211, + "grad_norm": 0.134765625, + "learning_rate": 0.0019777500292757263, + "loss": 0.1289, + "step": 7739 + }, + { + "epoch": 0.06718691678023628, + "grad_norm": 0.47265625, + "learning_rate": 0.0019777434518185766, + "loss": 0.1504, + "step": 7740 + }, + { + "epoch": 0.06719559726044044, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019777368734015394, + "loss": 0.1387, + "step": 7741 + }, + { + "epoch": 0.06720427774064461, + "grad_norm": 0.12158203125, + "learning_rate": 0.001977730294024623, + "loss": 0.1582, + "step": 7742 + }, + { + "epoch": 0.06721295822084877, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019777237136878335, + "loss": 0.1504, + "step": 7743 + }, + { + "epoch": 0.06722163870105294, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019777171323911782, + "loss": 0.1069, + "step": 7744 + }, + { + "epoch": 0.0672303191812571, + "grad_norm": 0.431640625, + "learning_rate": 0.0019777105501346654, + "loss": 0.1191, + "step": 7745 + }, + { + "epoch": 0.06723899966146127, + "grad_norm": 0.412109375, + "learning_rate": 0.001977703966918301, + "loss": 0.1108, + "step": 7746 + }, + { + "epoch": 0.06724768014166543, + "grad_norm": 0.123046875, + "learning_rate": 0.001977697382742093, + "loss": 0.1738, + "step": 7747 + }, + { + "epoch": 0.0672563606218696, + "grad_norm": 0.486328125, + "learning_rate": 0.001977690797606048, + "loss": 0.1514, + "step": 7748 + }, + { + "epoch": 0.06726504110207376, + "grad_norm": 0.08203125, + "learning_rate": 0.001977684211510174, + "loss": 0.1504, + "step": 7749 + }, + { + "epoch": 0.06727372158227793, + "grad_norm": 0.23046875, + "learning_rate": 0.0019776776244544774, + "loss": 0.168, + "step": 7750 + }, + { + "epoch": 0.0672824020624821, + "grad_norm": 0.435546875, + "learning_rate": 0.001977671036438966, + "loss": 0.1445, + "step": 7751 + }, + { + "epoch": 0.06729108254268626, + "grad_norm": 0.56640625, + "learning_rate": 0.001977664447463646, + "loss": 0.1177, + "step": 7752 + }, + { + "epoch": 0.06729976302289042, + "grad_norm": 0.3671875, + "learning_rate": 0.0019776578575285263, + "loss": 0.1729, + "step": 7753 + }, + { + "epoch": 0.06730844350309459, + "grad_norm": 0.4765625, + "learning_rate": 0.0019776512666336127, + "loss": 0.1426, + "step": 7754 + }, + { + "epoch": 0.06731712398329875, + "grad_norm": 0.0966796875, + "learning_rate": 0.001977644674778913, + "loss": 0.1895, + "step": 7755 + }, + { + "epoch": 0.06732580446350292, + "grad_norm": 0.322265625, + "learning_rate": 0.001977638081964434, + "loss": 0.1641, + "step": 7756 + }, + { + "epoch": 0.06733448494370708, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019776314881901833, + "loss": 0.1553, + "step": 7757 + }, + { + "epoch": 0.06734316542391125, + "grad_norm": 0.087890625, + "learning_rate": 0.0019776248934561683, + "loss": 0.1396, + "step": 7758 + }, + { + "epoch": 0.06735184590411541, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019776182977623955, + "loss": 0.1758, + "step": 7759 + }, + { + "epoch": 0.06736052638431958, + "grad_norm": 0.73046875, + "learning_rate": 0.0019776117011088726, + "loss": 0.1602, + "step": 7760 + }, + { + "epoch": 0.06736920686452375, + "grad_norm": 0.52734375, + "learning_rate": 0.001977605103495607, + "loss": 0.0996, + "step": 7761 + }, + { + "epoch": 0.06737788734472791, + "grad_norm": 0.294921875, + "learning_rate": 0.001977598504922605, + "loss": 0.1123, + "step": 7762 + }, + { + "epoch": 0.06738656782493208, + "grad_norm": 0.193359375, + "learning_rate": 0.0019775919053898754, + "loss": 0.1162, + "step": 7763 + }, + { + "epoch": 0.06739524830513624, + "grad_norm": 0.29296875, + "learning_rate": 0.0019775853048974237, + "loss": 0.1123, + "step": 7764 + }, + { + "epoch": 0.0674039287853404, + "grad_norm": 0.380859375, + "learning_rate": 0.0019775787034452584, + "loss": 0.127, + "step": 7765 + }, + { + "epoch": 0.06741260926554457, + "grad_norm": 0.07666015625, + "learning_rate": 0.001977572101033386, + "loss": 0.1348, + "step": 7766 + }, + { + "epoch": 0.06742128974574874, + "grad_norm": 0.36328125, + "learning_rate": 0.001977565497661814, + "loss": 0.1035, + "step": 7767 + }, + { + "epoch": 0.0674299702259529, + "grad_norm": 0.0751953125, + "learning_rate": 0.001977558893330549, + "loss": 0.1201, + "step": 7768 + }, + { + "epoch": 0.06743865070615707, + "grad_norm": 0.490234375, + "learning_rate": 0.0019775522880395996, + "loss": 0.1104, + "step": 7769 + }, + { + "epoch": 0.06744733118636123, + "grad_norm": 0.345703125, + "learning_rate": 0.001977545681788972, + "loss": 0.1543, + "step": 7770 + }, + { + "epoch": 0.0674560116665654, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019775390745786735, + "loss": 0.1777, + "step": 7771 + }, + { + "epoch": 0.06746469214676956, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019775324664087113, + "loss": 0.1201, + "step": 7772 + }, + { + "epoch": 0.06747337262697373, + "grad_norm": 0.474609375, + "learning_rate": 0.001977525857279093, + "loss": 0.1187, + "step": 7773 + }, + { + "epoch": 0.06748205310717789, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019775192471898255, + "loss": 0.1738, + "step": 7774 + }, + { + "epoch": 0.06749073358738206, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019775126361409163, + "loss": 0.1113, + "step": 7775 + }, + { + "epoch": 0.06749941406758622, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019775060241323723, + "loss": 0.1729, + "step": 7776 + }, + { + "epoch": 0.06750809454779039, + "grad_norm": 0.08447265625, + "learning_rate": 0.001977499411164201, + "loss": 0.1475, + "step": 7777 + }, + { + "epoch": 0.06751677502799455, + "grad_norm": 0.6484375, + "learning_rate": 0.0019774927972364092, + "loss": 0.1396, + "step": 7778 + }, + { + "epoch": 0.06752545550819872, + "grad_norm": 0.15625, + "learning_rate": 0.0019774861823490045, + "loss": 0.123, + "step": 7779 + }, + { + "epoch": 0.06753413598840288, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019774795665019945, + "loss": 0.1367, + "step": 7780 + }, + { + "epoch": 0.06754281646860705, + "grad_norm": 0.232421875, + "learning_rate": 0.001977472949695386, + "loss": 0.1387, + "step": 7781 + }, + { + "epoch": 0.06755149694881121, + "grad_norm": 0.25390625, + "learning_rate": 0.001977466331929186, + "loss": 0.1387, + "step": 7782 + }, + { + "epoch": 0.06756017742901538, + "grad_norm": 0.2490234375, + "learning_rate": 0.001977459713203402, + "loss": 0.1738, + "step": 7783 + }, + { + "epoch": 0.06756885790921954, + "grad_norm": 0.75, + "learning_rate": 0.0019774530935180415, + "loss": 0.1885, + "step": 7784 + }, + { + "epoch": 0.06757753838942371, + "grad_norm": 0.60546875, + "learning_rate": 0.0019774464728731112, + "loss": 0.1455, + "step": 7785 + }, + { + "epoch": 0.06758621886962787, + "grad_norm": 0.33984375, + "learning_rate": 0.001977439851268619, + "loss": 0.1406, + "step": 7786 + }, + { + "epoch": 0.06759489934983204, + "grad_norm": 0.353515625, + "learning_rate": 0.0019774332287045716, + "loss": 0.1387, + "step": 7787 + }, + { + "epoch": 0.0676035798300362, + "grad_norm": 0.13671875, + "learning_rate": 0.001977426605180976, + "loss": 0.1152, + "step": 7788 + }, + { + "epoch": 0.06761226031024037, + "grad_norm": 0.287109375, + "learning_rate": 0.0019774199806978404, + "loss": 0.1045, + "step": 7789 + }, + { + "epoch": 0.06762094079044453, + "grad_norm": 0.2333984375, + "learning_rate": 0.001977413355255171, + "loss": 0.1299, + "step": 7790 + }, + { + "epoch": 0.0676296212706487, + "grad_norm": 0.10302734375, + "learning_rate": 0.001977406728852976, + "loss": 0.1348, + "step": 7791 + }, + { + "epoch": 0.06763830175085286, + "grad_norm": 0.34765625, + "learning_rate": 0.0019774001014912623, + "loss": 0.1621, + "step": 7792 + }, + { + "epoch": 0.06764698223105703, + "grad_norm": 0.98046875, + "learning_rate": 0.0019773934731700366, + "loss": 0.125, + "step": 7793 + }, + { + "epoch": 0.06765566271126118, + "grad_norm": 0.248046875, + "learning_rate": 0.0019773868438893068, + "loss": 0.1475, + "step": 7794 + }, + { + "epoch": 0.06766434319146535, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019773802136490797, + "loss": 0.1021, + "step": 7795 + }, + { + "epoch": 0.06767302367166951, + "grad_norm": 0.330078125, + "learning_rate": 0.0019773735824493628, + "loss": 0.1211, + "step": 7796 + }, + { + "epoch": 0.06768170415187368, + "grad_norm": 0.42578125, + "learning_rate": 0.0019773669502901633, + "loss": 0.1406, + "step": 7797 + }, + { + "epoch": 0.06769038463207784, + "grad_norm": 0.6640625, + "learning_rate": 0.0019773603171714888, + "loss": 0.1299, + "step": 7798 + }, + { + "epoch": 0.067699065112282, + "grad_norm": 0.392578125, + "learning_rate": 0.0019773536830933465, + "loss": 0.1895, + "step": 7799 + }, + { + "epoch": 0.06770774559248617, + "grad_norm": 0.25390625, + "learning_rate": 0.0019773470480557425, + "loss": 0.1426, + "step": 7800 + }, + { + "epoch": 0.06771642607269034, + "grad_norm": 0.69921875, + "learning_rate": 0.0019773404120586855, + "loss": 0.1504, + "step": 7801 + }, + { + "epoch": 0.0677251065528945, + "grad_norm": 0.337890625, + "learning_rate": 0.001977333775102182, + "loss": 0.125, + "step": 7802 + }, + { + "epoch": 0.06773378703309867, + "grad_norm": 0.48046875, + "learning_rate": 0.00197732713718624, + "loss": 0.1348, + "step": 7803 + }, + { + "epoch": 0.06774246751330283, + "grad_norm": 0.365234375, + "learning_rate": 0.0019773204983108655, + "loss": 0.1279, + "step": 7804 + }, + { + "epoch": 0.067751147993507, + "grad_norm": 0.11279296875, + "learning_rate": 0.001977313858476067, + "loss": 0.1289, + "step": 7805 + }, + { + "epoch": 0.06775982847371116, + "grad_norm": 0.130859375, + "learning_rate": 0.0019773072176818506, + "loss": 0.124, + "step": 7806 + }, + { + "epoch": 0.06776850895391533, + "grad_norm": 0.875, + "learning_rate": 0.001977300575928225, + "loss": 0.3457, + "step": 7807 + }, + { + "epoch": 0.06777718943411949, + "grad_norm": 0.306640625, + "learning_rate": 0.0019772939332151963, + "loss": 0.1279, + "step": 7808 + }, + { + "epoch": 0.06778586991432366, + "grad_norm": 0.435546875, + "learning_rate": 0.001977287289542772, + "loss": 0.1348, + "step": 7809 + }, + { + "epoch": 0.06779455039452782, + "grad_norm": 0.126953125, + "learning_rate": 0.00197728064491096, + "loss": 0.124, + "step": 7810 + }, + { + "epoch": 0.06780323087473199, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019772739993197666, + "loss": 0.1455, + "step": 7811 + }, + { + "epoch": 0.06781191135493615, + "grad_norm": 0.484375, + "learning_rate": 0.0019772673527691997, + "loss": 0.1426, + "step": 7812 + }, + { + "epoch": 0.06782059183514032, + "grad_norm": 0.83984375, + "learning_rate": 0.0019772607052592666, + "loss": 0.1504, + "step": 7813 + }, + { + "epoch": 0.06782927231534448, + "grad_norm": 0.08984375, + "learning_rate": 0.0019772540567899737, + "loss": 0.1094, + "step": 7814 + }, + { + "epoch": 0.06783795279554865, + "grad_norm": 0.94140625, + "learning_rate": 0.0019772474073613297, + "loss": 0.1816, + "step": 7815 + }, + { + "epoch": 0.06784663327575281, + "grad_norm": 1.328125, + "learning_rate": 0.0019772407569733402, + "loss": 0.1738, + "step": 7816 + }, + { + "epoch": 0.06785531375595698, + "grad_norm": 0.326171875, + "learning_rate": 0.0019772341056260145, + "loss": 0.1348, + "step": 7817 + }, + { + "epoch": 0.06786399423616114, + "grad_norm": 0.71484375, + "learning_rate": 0.001977227453319358, + "loss": 0.5586, + "step": 7818 + }, + { + "epoch": 0.06787267471636531, + "grad_norm": 0.1728515625, + "learning_rate": 0.001977220800053379, + "loss": 0.1504, + "step": 7819 + }, + { + "epoch": 0.06788135519656947, + "grad_norm": 1.1484375, + "learning_rate": 0.0019772141458280846, + "loss": 0.1318, + "step": 7820 + }, + { + "epoch": 0.06789003567677364, + "grad_norm": 0.265625, + "learning_rate": 0.001977207490643482, + "loss": 0.1533, + "step": 7821 + }, + { + "epoch": 0.0678987161569778, + "grad_norm": 0.1181640625, + "learning_rate": 0.001977200834499578, + "loss": 0.1084, + "step": 7822 + }, + { + "epoch": 0.06790739663718197, + "grad_norm": 0.4921875, + "learning_rate": 0.001977194177396381, + "loss": 0.1318, + "step": 7823 + }, + { + "epoch": 0.06791607711738613, + "grad_norm": 0.2158203125, + "learning_rate": 0.001977187519333897, + "loss": 0.1406, + "step": 7824 + }, + { + "epoch": 0.0679247575975903, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019771808603121347, + "loss": 0.1416, + "step": 7825 + }, + { + "epoch": 0.06793343807779446, + "grad_norm": 0.142578125, + "learning_rate": 0.0019771742003311, + "loss": 0.1465, + "step": 7826 + }, + { + "epoch": 0.06794211855799863, + "grad_norm": 0.13671875, + "learning_rate": 0.001977167539390801, + "loss": 0.1396, + "step": 7827 + }, + { + "epoch": 0.0679507990382028, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019771608774912445, + "loss": 0.1167, + "step": 7828 + }, + { + "epoch": 0.06795947951840696, + "grad_norm": 0.408203125, + "learning_rate": 0.001977154214632438, + "loss": 0.166, + "step": 7829 + }, + { + "epoch": 0.06796815999861112, + "grad_norm": 0.18359375, + "learning_rate": 0.001977147550814389, + "loss": 0.1094, + "step": 7830 + }, + { + "epoch": 0.06797684047881529, + "grad_norm": 0.443359375, + "learning_rate": 0.001977140886037105, + "loss": 0.1602, + "step": 7831 + }, + { + "epoch": 0.06798552095901945, + "grad_norm": 0.279296875, + "learning_rate": 0.0019771342203005926, + "loss": 0.1289, + "step": 7832 + }, + { + "epoch": 0.06799420143922362, + "grad_norm": 0.306640625, + "learning_rate": 0.0019771275536048594, + "loss": 0.168, + "step": 7833 + }, + { + "epoch": 0.06800288191942779, + "grad_norm": 0.10888671875, + "learning_rate": 0.001977120885949912, + "loss": 0.168, + "step": 7834 + }, + { + "epoch": 0.06801156239963195, + "grad_norm": 0.59375, + "learning_rate": 0.0019771142173357595, + "loss": 0.1445, + "step": 7835 + }, + { + "epoch": 0.06802024287983612, + "grad_norm": 0.5234375, + "learning_rate": 0.0019771075477624076, + "loss": 0.1299, + "step": 7836 + }, + { + "epoch": 0.06802892336004028, + "grad_norm": 0.13671875, + "learning_rate": 0.001977100877229864, + "loss": 0.1299, + "step": 7837 + }, + { + "epoch": 0.06803760384024445, + "grad_norm": 0.08349609375, + "learning_rate": 0.001977094205738136, + "loss": 0.1123, + "step": 7838 + }, + { + "epoch": 0.06804628432044861, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019770875332872314, + "loss": 0.104, + "step": 7839 + }, + { + "epoch": 0.06805496480065278, + "grad_norm": 0.5390625, + "learning_rate": 0.0019770808598771564, + "loss": 0.1592, + "step": 7840 + }, + { + "epoch": 0.06806364528085694, + "grad_norm": 0.86328125, + "learning_rate": 0.0019770741855079195, + "loss": 0.1113, + "step": 7841 + }, + { + "epoch": 0.0680723257610611, + "grad_norm": 0.18359375, + "learning_rate": 0.001977067510179527, + "loss": 0.1445, + "step": 7842 + }, + { + "epoch": 0.06808100624126527, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019770608338919872, + "loss": 0.1475, + "step": 7843 + }, + { + "epoch": 0.06808968672146944, + "grad_norm": 0.158203125, + "learning_rate": 0.0019770541566453063, + "loss": 0.1514, + "step": 7844 + }, + { + "epoch": 0.0680983672016736, + "grad_norm": 0.6796875, + "learning_rate": 0.0019770474784394925, + "loss": 0.126, + "step": 7845 + }, + { + "epoch": 0.06810704768187777, + "grad_norm": 0.140625, + "learning_rate": 0.0019770407992745526, + "loss": 0.168, + "step": 7846 + }, + { + "epoch": 0.06811572816208193, + "grad_norm": 0.08544921875, + "learning_rate": 0.001977034119150494, + "loss": 0.1187, + "step": 7847 + }, + { + "epoch": 0.0681244086422861, + "grad_norm": 0.5546875, + "learning_rate": 0.0019770274380673243, + "loss": 0.1011, + "step": 7848 + }, + { + "epoch": 0.06813308912249026, + "grad_norm": 0.73046875, + "learning_rate": 0.00197702075602505, + "loss": 0.1602, + "step": 7849 + }, + { + "epoch": 0.06814176960269443, + "grad_norm": 0.45703125, + "learning_rate": 0.0019770140730236796, + "loss": 0.123, + "step": 7850 + }, + { + "epoch": 0.06815045008289859, + "grad_norm": 0.07861328125, + "learning_rate": 0.00197700738906322, + "loss": 0.1206, + "step": 7851 + }, + { + "epoch": 0.06815913056310276, + "grad_norm": 0.203125, + "learning_rate": 0.0019770007041436774, + "loss": 0.1406, + "step": 7852 + }, + { + "epoch": 0.06816781104330692, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019769940182650606, + "loss": 0.1006, + "step": 7853 + }, + { + "epoch": 0.06817649152351109, + "grad_norm": 0.0859375, + "learning_rate": 0.0019769873314273764, + "loss": 0.1201, + "step": 7854 + }, + { + "epoch": 0.06818517200371524, + "grad_norm": 0.384765625, + "learning_rate": 0.001976980643630632, + "loss": 0.3047, + "step": 7855 + }, + { + "epoch": 0.0681938524839194, + "grad_norm": 0.5390625, + "learning_rate": 0.001976973954874834, + "loss": 0.124, + "step": 7856 + }, + { + "epoch": 0.06820253296412357, + "grad_norm": 0.375, + "learning_rate": 0.0019769672651599917, + "loss": 0.1123, + "step": 7857 + }, + { + "epoch": 0.06821121344432773, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019769605744861104, + "loss": 0.1367, + "step": 7858 + }, + { + "epoch": 0.0682198939245319, + "grad_norm": 0.1533203125, + "learning_rate": 0.001976953882853198, + "loss": 0.1035, + "step": 7859 + }, + { + "epoch": 0.06822857440473606, + "grad_norm": 0.66015625, + "learning_rate": 0.001976947190261262, + "loss": 0.1182, + "step": 7860 + }, + { + "epoch": 0.06823725488494023, + "grad_norm": 0.21875, + "learning_rate": 0.0019769404967103104, + "loss": 0.1621, + "step": 7861 + }, + { + "epoch": 0.0682459353651444, + "grad_norm": 0.640625, + "learning_rate": 0.0019769338022003497, + "loss": 0.1885, + "step": 7862 + }, + { + "epoch": 0.06825461584534856, + "grad_norm": 0.640625, + "learning_rate": 0.001976927106731387, + "loss": 0.1592, + "step": 7863 + }, + { + "epoch": 0.06826329632555272, + "grad_norm": 0.1669921875, + "learning_rate": 0.00197692041030343, + "loss": 0.1797, + "step": 7864 + }, + { + "epoch": 0.06827197680575689, + "grad_norm": 0.314453125, + "learning_rate": 0.0019769137129164865, + "loss": 0.1279, + "step": 7865 + }, + { + "epoch": 0.06828065728596105, + "grad_norm": 0.53515625, + "learning_rate": 0.001976907014570563, + "loss": 0.1436, + "step": 7866 + }, + { + "epoch": 0.06828933776616522, + "grad_norm": 0.3125, + "learning_rate": 0.001976900315265667, + "loss": 0.1338, + "step": 7867 + }, + { + "epoch": 0.06829801824636939, + "grad_norm": 0.435546875, + "learning_rate": 0.0019768936150018064, + "loss": 0.1523, + "step": 7868 + }, + { + "epoch": 0.06830669872657355, + "grad_norm": 0.6015625, + "learning_rate": 0.001976886913778988, + "loss": 0.1318, + "step": 7869 + }, + { + "epoch": 0.06831537920677772, + "grad_norm": 0.478515625, + "learning_rate": 0.001976880211597219, + "loss": 0.1157, + "step": 7870 + }, + { + "epoch": 0.06832405968698188, + "grad_norm": 0.40625, + "learning_rate": 0.001976873508456507, + "loss": 0.1396, + "step": 7871 + }, + { + "epoch": 0.06833274016718605, + "grad_norm": 0.390625, + "learning_rate": 0.001976866804356859, + "loss": 0.1455, + "step": 7872 + }, + { + "epoch": 0.06834142064739021, + "grad_norm": 0.64453125, + "learning_rate": 0.0019768600992982836, + "loss": 0.1035, + "step": 7873 + }, + { + "epoch": 0.06835010112759438, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019768533932807866, + "loss": 0.1689, + "step": 7874 + }, + { + "epoch": 0.06835878160779854, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019768466863043757, + "loss": 0.1318, + "step": 7875 + }, + { + "epoch": 0.0683674620880027, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019768399783690587, + "loss": 0.1357, + "step": 7876 + }, + { + "epoch": 0.06837614256820687, + "grad_norm": 0.197265625, + "learning_rate": 0.001976833269474843, + "loss": 0.0991, + "step": 7877 + }, + { + "epoch": 0.06838482304841104, + "grad_norm": 0.244140625, + "learning_rate": 0.001976826559621735, + "loss": 0.1562, + "step": 7878 + }, + { + "epoch": 0.0683935035286152, + "grad_norm": 0.087890625, + "learning_rate": 0.001976819848809743, + "loss": 0.1504, + "step": 7879 + }, + { + "epoch": 0.06840218400881937, + "grad_norm": 0.6328125, + "learning_rate": 0.0019768131370388737, + "loss": 0.1104, + "step": 7880 + }, + { + "epoch": 0.06841086448902353, + "grad_norm": 0.2294921875, + "learning_rate": 0.001976806424309135, + "loss": 0.1182, + "step": 7881 + }, + { + "epoch": 0.0684195449692277, + "grad_norm": 0.7578125, + "learning_rate": 0.0019767997106205338, + "loss": 0.1504, + "step": 7882 + }, + { + "epoch": 0.06842822544943186, + "grad_norm": 0.34765625, + "learning_rate": 0.001976792995973078, + "loss": 0.1484, + "step": 7883 + }, + { + "epoch": 0.06843690592963603, + "grad_norm": 0.16015625, + "learning_rate": 0.001976786280366774, + "loss": 0.1328, + "step": 7884 + }, + { + "epoch": 0.06844558640984019, + "grad_norm": 0.11328125, + "learning_rate": 0.00197677956380163, + "loss": 0.1533, + "step": 7885 + }, + { + "epoch": 0.06845426689004436, + "grad_norm": 0.287109375, + "learning_rate": 0.001976772846277653, + "loss": 0.1182, + "step": 7886 + }, + { + "epoch": 0.06846294737024852, + "grad_norm": 0.19921875, + "learning_rate": 0.0019767661277948502, + "loss": 0.1514, + "step": 7887 + }, + { + "epoch": 0.06847162785045269, + "grad_norm": 1.0390625, + "learning_rate": 0.001976759408353229, + "loss": 0.166, + "step": 7888 + }, + { + "epoch": 0.06848030833065685, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019767526879527976, + "loss": 0.1406, + "step": 7889 + }, + { + "epoch": 0.06848898881086102, + "grad_norm": 0.375, + "learning_rate": 0.0019767459665935624, + "loss": 0.1377, + "step": 7890 + }, + { + "epoch": 0.06849766929106518, + "grad_norm": 0.076171875, + "learning_rate": 0.0019767392442755307, + "loss": 0.1123, + "step": 7891 + }, + { + "epoch": 0.06850634977126935, + "grad_norm": 0.203125, + "learning_rate": 0.00197673252099871, + "loss": 0.1426, + "step": 7892 + }, + { + "epoch": 0.06851503025147351, + "grad_norm": 0.3203125, + "learning_rate": 0.001976725796763108, + "loss": 0.124, + "step": 7893 + }, + { + "epoch": 0.06852371073167768, + "grad_norm": 0.349609375, + "learning_rate": 0.0019767190715687323, + "loss": 0.1416, + "step": 7894 + }, + { + "epoch": 0.06853239121188184, + "grad_norm": 0.267578125, + "learning_rate": 0.001976712345415589, + "loss": 0.127, + "step": 7895 + }, + { + "epoch": 0.06854107169208601, + "grad_norm": 0.314453125, + "learning_rate": 0.0019767056183036865, + "loss": 0.0918, + "step": 7896 + }, + { + "epoch": 0.06854975217229017, + "grad_norm": 0.134765625, + "learning_rate": 0.001976698890233032, + "loss": 0.105, + "step": 7897 + }, + { + "epoch": 0.06855843265249434, + "grad_norm": 0.16796875, + "learning_rate": 0.0019766921612036333, + "loss": 0.166, + "step": 7898 + }, + { + "epoch": 0.0685671131326985, + "grad_norm": 0.18359375, + "learning_rate": 0.0019766854312154966, + "loss": 0.1011, + "step": 7899 + }, + { + "epoch": 0.06857579361290267, + "grad_norm": 0.416015625, + "learning_rate": 0.00197667870026863, + "loss": 0.1016, + "step": 7900 + }, + { + "epoch": 0.06858447409310683, + "grad_norm": 0.1884765625, + "learning_rate": 0.001976671968363041, + "loss": 0.1758, + "step": 7901 + }, + { + "epoch": 0.068593154573311, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019766652354987362, + "loss": 0.1523, + "step": 7902 + }, + { + "epoch": 0.06860183505351516, + "grad_norm": 0.470703125, + "learning_rate": 0.0019766585016757236, + "loss": 0.1187, + "step": 7903 + }, + { + "epoch": 0.06861051553371933, + "grad_norm": 0.28515625, + "learning_rate": 0.001976651766894011, + "loss": 0.1377, + "step": 7904 + }, + { + "epoch": 0.0686191960139235, + "grad_norm": 0.5625, + "learning_rate": 0.0019766450311536044, + "loss": 0.1543, + "step": 7905 + }, + { + "epoch": 0.06862787649412766, + "grad_norm": 0.13671875, + "learning_rate": 0.0019766382944545126, + "loss": 0.1318, + "step": 7906 + }, + { + "epoch": 0.06863655697433182, + "grad_norm": 0.1650390625, + "learning_rate": 0.001976631556796742, + "loss": 0.1523, + "step": 7907 + }, + { + "epoch": 0.06864523745453599, + "grad_norm": 0.142578125, + "learning_rate": 0.0019766248181803, + "loss": 0.1484, + "step": 7908 + }, + { + "epoch": 0.06865391793474016, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019766180786051947, + "loss": 0.1113, + "step": 7909 + }, + { + "epoch": 0.06866259841494432, + "grad_norm": 0.328125, + "learning_rate": 0.0019766113380714333, + "loss": 0.1514, + "step": 7910 + }, + { + "epoch": 0.06867127889514849, + "grad_norm": 0.3046875, + "learning_rate": 0.0019766045965790226, + "loss": 0.1123, + "step": 7911 + }, + { + "epoch": 0.06867995937535265, + "grad_norm": 0.326171875, + "learning_rate": 0.0019765978541279703, + "loss": 0.1143, + "step": 7912 + }, + { + "epoch": 0.06868863985555682, + "grad_norm": 0.6875, + "learning_rate": 0.0019765911107182833, + "loss": 0.1475, + "step": 7913 + }, + { + "epoch": 0.06869732033576098, + "grad_norm": 0.55859375, + "learning_rate": 0.00197658436634997, + "loss": 0.1484, + "step": 7914 + }, + { + "epoch": 0.06870600081596515, + "grad_norm": 0.57421875, + "learning_rate": 0.001976577621023037, + "loss": 0.1338, + "step": 7915 + }, + { + "epoch": 0.06871468129616931, + "grad_norm": 0.251953125, + "learning_rate": 0.001976570874737492, + "loss": 0.1279, + "step": 7916 + }, + { + "epoch": 0.06872336177637346, + "grad_norm": 0.4375, + "learning_rate": 0.0019765641274933424, + "loss": 0.1514, + "step": 7917 + }, + { + "epoch": 0.06873204225657763, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019765573792905953, + "loss": 0.1611, + "step": 7918 + }, + { + "epoch": 0.06874072273678179, + "grad_norm": 0.212890625, + "learning_rate": 0.001976550630129258, + "loss": 0.1689, + "step": 7919 + }, + { + "epoch": 0.06874940321698596, + "grad_norm": 0.427734375, + "learning_rate": 0.0019765438800093383, + "loss": 0.1299, + "step": 7920 + }, + { + "epoch": 0.06875808369719012, + "grad_norm": 0.416015625, + "learning_rate": 0.0019765371289308434, + "loss": 0.1338, + "step": 7921 + }, + { + "epoch": 0.06876676417739429, + "grad_norm": 0.10546875, + "learning_rate": 0.0019765303768937807, + "loss": 0.1191, + "step": 7922 + }, + { + "epoch": 0.06877544465759845, + "grad_norm": 0.16015625, + "learning_rate": 0.001976523623898157, + "loss": 0.1191, + "step": 7923 + }, + { + "epoch": 0.06878412513780262, + "grad_norm": 0.515625, + "learning_rate": 0.001976516869943981, + "loss": 0.1484, + "step": 7924 + }, + { + "epoch": 0.06879280561800678, + "grad_norm": 0.765625, + "learning_rate": 0.001976510115031259, + "loss": 0.1035, + "step": 7925 + }, + { + "epoch": 0.06880148609821095, + "grad_norm": 0.2734375, + "learning_rate": 0.001976503359159999, + "loss": 0.127, + "step": 7926 + }, + { + "epoch": 0.06881016657841511, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019764966023302077, + "loss": 0.1309, + "step": 7927 + }, + { + "epoch": 0.06881884705861928, + "grad_norm": 0.103515625, + "learning_rate": 0.001976489844541893, + "loss": 0.1396, + "step": 7928 + }, + { + "epoch": 0.06882752753882344, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019764830857950623, + "loss": 0.1328, + "step": 7929 + }, + { + "epoch": 0.06883620801902761, + "grad_norm": 0.373046875, + "learning_rate": 0.0019764763260897225, + "loss": 0.1504, + "step": 7930 + }, + { + "epoch": 0.06884488849923177, + "grad_norm": 0.271484375, + "learning_rate": 0.001976469565425882, + "loss": 0.1494, + "step": 7931 + }, + { + "epoch": 0.06885356897943594, + "grad_norm": 0.3359375, + "learning_rate": 0.001976462803803547, + "loss": 0.1235, + "step": 7932 + }, + { + "epoch": 0.0688622494596401, + "grad_norm": 0.1171875, + "learning_rate": 0.0019764560412227255, + "loss": 0.1211, + "step": 7933 + }, + { + "epoch": 0.06887092993984427, + "grad_norm": 0.185546875, + "learning_rate": 0.001976449277683425, + "loss": 0.084, + "step": 7934 + }, + { + "epoch": 0.06887961042004843, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019764425131856525, + "loss": 0.1719, + "step": 7935 + }, + { + "epoch": 0.0688882909002526, + "grad_norm": 0.56640625, + "learning_rate": 0.001976435747729416, + "loss": 0.1191, + "step": 7936 + }, + { + "epoch": 0.06889697138045676, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019764289813147223, + "loss": 0.1426, + "step": 7937 + }, + { + "epoch": 0.06890565186066093, + "grad_norm": 0.08837890625, + "learning_rate": 0.001976422213941579, + "loss": 0.1523, + "step": 7938 + }, + { + "epoch": 0.0689143323408651, + "grad_norm": 0.11376953125, + "learning_rate": 0.001976415445609994, + "loss": 0.1348, + "step": 7939 + }, + { + "epoch": 0.06892301282106926, + "grad_norm": 0.439453125, + "learning_rate": 0.001976408676319974, + "loss": 0.1055, + "step": 7940 + }, + { + "epoch": 0.06893169330127343, + "grad_norm": 0.27734375, + "learning_rate": 0.0019764019060715264, + "loss": 0.1162, + "step": 7941 + }, + { + "epoch": 0.06894037378147759, + "grad_norm": 0.3984375, + "learning_rate": 0.0019763951348646587, + "loss": 0.1006, + "step": 7942 + }, + { + "epoch": 0.06894905426168176, + "grad_norm": 0.62109375, + "learning_rate": 0.0019763883626993787, + "loss": 0.1084, + "step": 7943 + }, + { + "epoch": 0.06895773474188592, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019763815895756935, + "loss": 0.1079, + "step": 7944 + }, + { + "epoch": 0.06896641522209009, + "grad_norm": 0.30859375, + "learning_rate": 0.001976374815493611, + "loss": 0.1094, + "step": 7945 + }, + { + "epoch": 0.06897509570229425, + "grad_norm": 0.111328125, + "learning_rate": 0.0019763680404531374, + "loss": 0.1182, + "step": 7946 + }, + { + "epoch": 0.06898377618249842, + "grad_norm": 0.427734375, + "learning_rate": 0.0019763612644542817, + "loss": 0.0957, + "step": 7947 + }, + { + "epoch": 0.06899245666270258, + "grad_norm": 0.20703125, + "learning_rate": 0.0019763544874970497, + "loss": 0.1138, + "step": 7948 + }, + { + "epoch": 0.06900113714290675, + "grad_norm": 0.46875, + "learning_rate": 0.0019763477095814503, + "loss": 0.1147, + "step": 7949 + }, + { + "epoch": 0.06900981762311091, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019763409307074894, + "loss": 0.1387, + "step": 7950 + }, + { + "epoch": 0.06901849810331508, + "grad_norm": 0.09619140625, + "learning_rate": 0.001976334150875176, + "loss": 0.1377, + "step": 7951 + }, + { + "epoch": 0.06902717858351924, + "grad_norm": 0.255859375, + "learning_rate": 0.0019763273700845164, + "loss": 0.1162, + "step": 7952 + }, + { + "epoch": 0.0690358590637234, + "grad_norm": 0.0751953125, + "learning_rate": 0.001976320588335518, + "loss": 0.1543, + "step": 7953 + }, + { + "epoch": 0.06904453954392757, + "grad_norm": 0.07861328125, + "learning_rate": 0.001976313805628189, + "loss": 0.1167, + "step": 7954 + }, + { + "epoch": 0.06905322002413174, + "grad_norm": 0.294921875, + "learning_rate": 0.0019763070219625365, + "loss": 0.124, + "step": 7955 + }, + { + "epoch": 0.0690619005043359, + "grad_norm": 0.380859375, + "learning_rate": 0.001976300237338568, + "loss": 0.0947, + "step": 7956 + }, + { + "epoch": 0.06907058098454007, + "grad_norm": 0.359375, + "learning_rate": 0.0019762934517562904, + "loss": 0.2656, + "step": 7957 + }, + { + "epoch": 0.06907926146474423, + "grad_norm": 0.47265625, + "learning_rate": 0.001976286665215711, + "loss": 0.125, + "step": 7958 + }, + { + "epoch": 0.0690879419449484, + "grad_norm": 0.10107421875, + "learning_rate": 0.001976279877716838, + "loss": 0.1699, + "step": 7959 + }, + { + "epoch": 0.06909662242515256, + "grad_norm": 0.265625, + "learning_rate": 0.001976273089259679, + "loss": 0.1289, + "step": 7960 + }, + { + "epoch": 0.06910530290535673, + "grad_norm": 0.10107421875, + "learning_rate": 0.00197626629984424, + "loss": 0.1157, + "step": 7961 + }, + { + "epoch": 0.06911398338556089, + "grad_norm": 0.11767578125, + "learning_rate": 0.00197625950947053, + "loss": 0.1182, + "step": 7962 + }, + { + "epoch": 0.06912266386576506, + "grad_norm": 0.0859375, + "learning_rate": 0.0019762527181385555, + "loss": 0.1348, + "step": 7963 + }, + { + "epoch": 0.06913134434596922, + "grad_norm": 0.291015625, + "learning_rate": 0.001976245925848325, + "loss": 0.1206, + "step": 7964 + }, + { + "epoch": 0.06914002482617339, + "grad_norm": 0.80078125, + "learning_rate": 0.001976239132599844, + "loss": 0.1494, + "step": 7965 + }, + { + "epoch": 0.06914870530637755, + "grad_norm": 0.34375, + "learning_rate": 0.0019762323383931218, + "loss": 0.1113, + "step": 7966 + }, + { + "epoch": 0.06915738578658172, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019762255432281646, + "loss": 0.1504, + "step": 7967 + }, + { + "epoch": 0.06916606626678588, + "grad_norm": 0.380859375, + "learning_rate": 0.0019762187471049804, + "loss": 0.1099, + "step": 7968 + }, + { + "epoch": 0.06917474674699005, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019762119500235766, + "loss": 0.1084, + "step": 7969 + }, + { + "epoch": 0.06918342722719421, + "grad_norm": 0.62890625, + "learning_rate": 0.0019762051519839605, + "loss": 0.2109, + "step": 7970 + }, + { + "epoch": 0.06919210770739838, + "grad_norm": 0.7265625, + "learning_rate": 0.0019761983529861396, + "loss": 0.1455, + "step": 7971 + }, + { + "epoch": 0.06920078818760254, + "grad_norm": 0.3515625, + "learning_rate": 0.0019761915530301217, + "loss": 0.1055, + "step": 7972 + }, + { + "epoch": 0.06920946866780671, + "grad_norm": 0.318359375, + "learning_rate": 0.0019761847521159136, + "loss": 0.1328, + "step": 7973 + }, + { + "epoch": 0.06921814914801087, + "grad_norm": 0.267578125, + "learning_rate": 0.0019761779502435227, + "loss": 0.1445, + "step": 7974 + }, + { + "epoch": 0.06922682962821504, + "grad_norm": 0.376953125, + "learning_rate": 0.0019761711474129574, + "loss": 0.1777, + "step": 7975 + }, + { + "epoch": 0.0692355101084192, + "grad_norm": 0.138671875, + "learning_rate": 0.001976164343624224, + "loss": 0.1523, + "step": 7976 + }, + { + "epoch": 0.06924419058862337, + "grad_norm": 0.419921875, + "learning_rate": 0.001976157538877331, + "loss": 0.1309, + "step": 7977 + }, + { + "epoch": 0.06925287106882752, + "grad_norm": 0.34765625, + "learning_rate": 0.001976150733172285, + "loss": 0.1406, + "step": 7978 + }, + { + "epoch": 0.06926155154903169, + "grad_norm": 0.2421875, + "learning_rate": 0.0019761439265090935, + "loss": 0.1816, + "step": 7979 + }, + { + "epoch": 0.06927023202923585, + "grad_norm": 0.1494140625, + "learning_rate": 0.001976137118887764, + "loss": 0.0903, + "step": 7980 + }, + { + "epoch": 0.06927891250944002, + "grad_norm": 0.263671875, + "learning_rate": 0.0019761303103083046, + "loss": 0.0903, + "step": 7981 + }, + { + "epoch": 0.06928759298964418, + "grad_norm": 0.546875, + "learning_rate": 0.001976123500770722, + "loss": 0.1396, + "step": 7982 + }, + { + "epoch": 0.06929627346984835, + "grad_norm": 0.283203125, + "learning_rate": 0.001976116690275024, + "loss": 0.123, + "step": 7983 + }, + { + "epoch": 0.06930495395005251, + "grad_norm": 0.2490234375, + "learning_rate": 0.001976109878821218, + "loss": 0.1406, + "step": 7984 + }, + { + "epoch": 0.06931363443025668, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019761030664093115, + "loss": 0.1084, + "step": 7985 + }, + { + "epoch": 0.06932231491046084, + "grad_norm": 0.48828125, + "learning_rate": 0.0019760962530393114, + "loss": 0.1445, + "step": 7986 + }, + { + "epoch": 0.069330995390665, + "grad_norm": 0.5078125, + "learning_rate": 0.0019760894387112257, + "loss": 0.1504, + "step": 7987 + }, + { + "epoch": 0.06933967587086917, + "grad_norm": 0.18359375, + "learning_rate": 0.001976082623425062, + "loss": 0.1216, + "step": 7988 + }, + { + "epoch": 0.06934835635107334, + "grad_norm": 0.72265625, + "learning_rate": 0.0019760758071808277, + "loss": 0.1191, + "step": 7989 + }, + { + "epoch": 0.0693570368312775, + "grad_norm": 0.326171875, + "learning_rate": 0.0019760689899785294, + "loss": 0.126, + "step": 7990 + }, + { + "epoch": 0.06936571731148167, + "grad_norm": 0.095703125, + "learning_rate": 0.001976062171818176, + "loss": 0.1299, + "step": 7991 + }, + { + "epoch": 0.06937439779168583, + "grad_norm": 0.28125, + "learning_rate": 0.0019760553526997735, + "loss": 0.1099, + "step": 7992 + }, + { + "epoch": 0.06938307827189, + "grad_norm": 0.091796875, + "learning_rate": 0.00197604853262333, + "loss": 0.123, + "step": 7993 + }, + { + "epoch": 0.06939175875209416, + "grad_norm": 0.361328125, + "learning_rate": 0.001976041711588854, + "loss": 0.1182, + "step": 7994 + }, + { + "epoch": 0.06940043923229833, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019760348895963512, + "loss": 0.1621, + "step": 7995 + }, + { + "epoch": 0.06940911971250249, + "grad_norm": 0.64453125, + "learning_rate": 0.00197602806664583, + "loss": 0.1543, + "step": 7996 + }, + { + "epoch": 0.06941780019270666, + "grad_norm": 0.3828125, + "learning_rate": 0.0019760212427372974, + "loss": 0.1445, + "step": 7997 + }, + { + "epoch": 0.06942648067291082, + "grad_norm": 0.7578125, + "learning_rate": 0.001976014417870761, + "loss": 0.0977, + "step": 7998 + }, + { + "epoch": 0.06943516115311499, + "grad_norm": 0.12158203125, + "learning_rate": 0.001976007592046229, + "loss": 0.124, + "step": 7999 + }, + { + "epoch": 0.06944384163331915, + "grad_norm": 0.94921875, + "learning_rate": 0.0019760007652637078, + "loss": 0.2266, + "step": 8000 + }, + { + "epoch": 0.06945252211352332, + "grad_norm": 0.107421875, + "learning_rate": 0.001975993937523205, + "loss": 0.1445, + "step": 8001 + }, + { + "epoch": 0.06946120259372748, + "grad_norm": 0.1572265625, + "learning_rate": 0.001975987108824729, + "loss": 0.1357, + "step": 8002 + }, + { + "epoch": 0.06946988307393165, + "grad_norm": 0.21875, + "learning_rate": 0.0019759802791682866, + "loss": 0.1602, + "step": 8003 + }, + { + "epoch": 0.06947856355413581, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019759734485538852, + "loss": 0.1021, + "step": 8004 + }, + { + "epoch": 0.06948724403433998, + "grad_norm": 0.33984375, + "learning_rate": 0.0019759666169815323, + "loss": 0.1182, + "step": 8005 + }, + { + "epoch": 0.06949592451454414, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019759597844512355, + "loss": 0.1016, + "step": 8006 + }, + { + "epoch": 0.06950460499474831, + "grad_norm": 0.3359375, + "learning_rate": 0.0019759529509630024, + "loss": 0.1387, + "step": 8007 + }, + { + "epoch": 0.06951328547495247, + "grad_norm": 1.0078125, + "learning_rate": 0.00197594611651684, + "loss": 0.2246, + "step": 8008 + }, + { + "epoch": 0.06952196595515664, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019759392811127567, + "loss": 0.1328, + "step": 8009 + }, + { + "epoch": 0.0695306464353608, + "grad_norm": 0.328125, + "learning_rate": 0.001975932444750759, + "loss": 0.1602, + "step": 8010 + }, + { + "epoch": 0.06953932691556497, + "grad_norm": 0.87890625, + "learning_rate": 0.0019759256074308547, + "loss": 0.1104, + "step": 8011 + }, + { + "epoch": 0.06954800739576913, + "grad_norm": 0.50390625, + "learning_rate": 0.0019759187691530513, + "loss": 0.1235, + "step": 8012 + }, + { + "epoch": 0.0695566878759733, + "grad_norm": 0.416015625, + "learning_rate": 0.0019759119299173566, + "loss": 0.124, + "step": 8013 + }, + { + "epoch": 0.06956536835617746, + "grad_norm": 0.41015625, + "learning_rate": 0.0019759050897237775, + "loss": 0.1689, + "step": 8014 + }, + { + "epoch": 0.06957404883638163, + "grad_norm": 0.734375, + "learning_rate": 0.0019758982485723214, + "loss": 0.1406, + "step": 8015 + }, + { + "epoch": 0.0695827293165858, + "grad_norm": 0.111328125, + "learning_rate": 0.0019758914064629965, + "loss": 0.125, + "step": 8016 + }, + { + "epoch": 0.06959140979678996, + "grad_norm": 0.1611328125, + "learning_rate": 0.00197588456339581, + "loss": 0.123, + "step": 8017 + }, + { + "epoch": 0.06960009027699413, + "grad_norm": 0.15625, + "learning_rate": 0.001975877719370769, + "loss": 0.1367, + "step": 8018 + }, + { + "epoch": 0.06960877075719829, + "grad_norm": 0.21875, + "learning_rate": 0.0019758708743878818, + "loss": 0.1279, + "step": 8019 + }, + { + "epoch": 0.06961745123740246, + "grad_norm": 1.203125, + "learning_rate": 0.001975864028447155, + "loss": 0.1602, + "step": 8020 + }, + { + "epoch": 0.06962613171760662, + "grad_norm": 0.13671875, + "learning_rate": 0.0019758571815485966, + "loss": 0.1426, + "step": 8021 + }, + { + "epoch": 0.06963481219781079, + "grad_norm": 0.60546875, + "learning_rate": 0.001975850333692214, + "loss": 0.1162, + "step": 8022 + }, + { + "epoch": 0.06964349267801495, + "grad_norm": 0.96875, + "learning_rate": 0.0019758434848780146, + "loss": 0.1055, + "step": 8023 + }, + { + "epoch": 0.06965217315821912, + "grad_norm": 0.3359375, + "learning_rate": 0.0019758366351060055, + "loss": 0.1328, + "step": 8024 + }, + { + "epoch": 0.06966085363842328, + "grad_norm": 0.1142578125, + "learning_rate": 0.001975829784376195, + "loss": 0.1279, + "step": 8025 + }, + { + "epoch": 0.06966953411862745, + "grad_norm": 0.173828125, + "learning_rate": 0.0019758229326885903, + "loss": 0.1426, + "step": 8026 + }, + { + "epoch": 0.06967821459883161, + "grad_norm": 0.5859375, + "learning_rate": 0.0019758160800431987, + "loss": 0.1426, + "step": 8027 + }, + { + "epoch": 0.06968689507903578, + "grad_norm": 0.142578125, + "learning_rate": 0.001975809226440028, + "loss": 0.1309, + "step": 8028 + }, + { + "epoch": 0.06969557555923994, + "grad_norm": 0.275390625, + "learning_rate": 0.001975802371879085, + "loss": 0.1562, + "step": 8029 + }, + { + "epoch": 0.0697042560394441, + "grad_norm": 0.1376953125, + "learning_rate": 0.001975795516360378, + "loss": 0.1523, + "step": 8030 + }, + { + "epoch": 0.06971293651964827, + "grad_norm": 0.328125, + "learning_rate": 0.001975788659883914, + "loss": 0.1069, + "step": 8031 + }, + { + "epoch": 0.06972161699985244, + "grad_norm": 0.244140625, + "learning_rate": 0.0019757818024497007, + "loss": 0.1367, + "step": 8032 + }, + { + "epoch": 0.0697302974800566, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019757749440577456, + "loss": 0.1592, + "step": 8033 + }, + { + "epoch": 0.06973897796026077, + "grad_norm": 0.193359375, + "learning_rate": 0.0019757680847080562, + "loss": 0.1699, + "step": 8034 + }, + { + "epoch": 0.06974765844046493, + "grad_norm": 0.208984375, + "learning_rate": 0.0019757612244006404, + "loss": 0.1445, + "step": 8035 + }, + { + "epoch": 0.0697563389206691, + "grad_norm": 0.1845703125, + "learning_rate": 0.001975754363135505, + "loss": 0.0977, + "step": 8036 + }, + { + "epoch": 0.06976501940087326, + "grad_norm": 0.30078125, + "learning_rate": 0.001975747500912657, + "loss": 0.1816, + "step": 8037 + }, + { + "epoch": 0.06977369988107743, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019757406377321055, + "loss": 0.127, + "step": 8038 + }, + { + "epoch": 0.06978238036128159, + "grad_norm": 0.369140625, + "learning_rate": 0.0019757337735938573, + "loss": 0.0996, + "step": 8039 + }, + { + "epoch": 0.06979106084148574, + "grad_norm": 0.59375, + "learning_rate": 0.00197572690849792, + "loss": 0.1807, + "step": 8040 + }, + { + "epoch": 0.06979974132168991, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019757200424443, + "loss": 0.1309, + "step": 8041 + }, + { + "epoch": 0.06980842180189407, + "grad_norm": 0.5078125, + "learning_rate": 0.0019757131754330063, + "loss": 0.1836, + "step": 8042 + }, + { + "epoch": 0.06981710228209824, + "grad_norm": 0.3359375, + "learning_rate": 0.001975706307464046, + "loss": 0.1396, + "step": 8043 + }, + { + "epoch": 0.0698257827623024, + "grad_norm": 0.150390625, + "learning_rate": 0.001975699438537426, + "loss": 0.1748, + "step": 8044 + }, + { + "epoch": 0.06983446324250657, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019756925686531543, + "loss": 0.1367, + "step": 8045 + }, + { + "epoch": 0.06984314372271073, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019756856978112385, + "loss": 0.1406, + "step": 8046 + }, + { + "epoch": 0.0698518242029149, + "grad_norm": 0.1484375, + "learning_rate": 0.0019756788260116864, + "loss": 0.1465, + "step": 8047 + }, + { + "epoch": 0.06986050468311907, + "grad_norm": 0.40234375, + "learning_rate": 0.0019756719532545044, + "loss": 0.1211, + "step": 8048 + }, + { + "epoch": 0.06986918516332323, + "grad_norm": 0.166015625, + "learning_rate": 0.001975665079539701, + "loss": 0.1494, + "step": 8049 + }, + { + "epoch": 0.0698778656435274, + "grad_norm": 0.21484375, + "learning_rate": 0.0019756582048672837, + "loss": 0.1426, + "step": 8050 + }, + { + "epoch": 0.06988654612373156, + "grad_norm": 1.09375, + "learning_rate": 0.001975651329237259, + "loss": 0.125, + "step": 8051 + }, + { + "epoch": 0.06989522660393573, + "grad_norm": 0.072265625, + "learning_rate": 0.0019756444526496356, + "loss": 0.0835, + "step": 8052 + }, + { + "epoch": 0.06990390708413989, + "grad_norm": 0.578125, + "learning_rate": 0.001975637575104421, + "loss": 0.1367, + "step": 8053 + }, + { + "epoch": 0.06991258756434406, + "grad_norm": 0.36328125, + "learning_rate": 0.001975630696601622, + "loss": 0.1543, + "step": 8054 + }, + { + "epoch": 0.06992126804454822, + "grad_norm": 0.44921875, + "learning_rate": 0.0019756238171412463, + "loss": 0.125, + "step": 8055 + }, + { + "epoch": 0.06992994852475239, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019756169367233013, + "loss": 0.1592, + "step": 8056 + }, + { + "epoch": 0.06993862900495655, + "grad_norm": 0.2333984375, + "learning_rate": 0.001975610055347795, + "loss": 0.1143, + "step": 8057 + }, + { + "epoch": 0.06994730948516072, + "grad_norm": 0.1337890625, + "learning_rate": 0.001975603173014735, + "loss": 0.1758, + "step": 8058 + }, + { + "epoch": 0.06995598996536488, + "grad_norm": 0.7265625, + "learning_rate": 0.0019755962897241284, + "loss": 0.123, + "step": 8059 + }, + { + "epoch": 0.06996467044556905, + "grad_norm": 0.318359375, + "learning_rate": 0.0019755894054759825, + "loss": 0.1406, + "step": 8060 + }, + { + "epoch": 0.06997335092577321, + "grad_norm": 0.6328125, + "learning_rate": 0.001975582520270305, + "loss": 0.1426, + "step": 8061 + }, + { + "epoch": 0.06998203140597738, + "grad_norm": 0.5625, + "learning_rate": 0.0019755756341071043, + "loss": 0.125, + "step": 8062 + }, + { + "epoch": 0.06999071188618154, + "grad_norm": 0.451171875, + "learning_rate": 0.001975568746986387, + "loss": 0.127, + "step": 8063 + }, + { + "epoch": 0.06999939236638571, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019755618589081606, + "loss": 0.1221, + "step": 8064 + }, + { + "epoch": 0.07000807284658987, + "grad_norm": 0.671875, + "learning_rate": 0.0019755549698724337, + "loss": 0.167, + "step": 8065 + }, + { + "epoch": 0.07001675332679404, + "grad_norm": 0.337890625, + "learning_rate": 0.0019755480798792123, + "loss": 0.1924, + "step": 8066 + }, + { + "epoch": 0.0700254338069982, + "grad_norm": 0.83203125, + "learning_rate": 0.0019755411889285047, + "loss": 0.1484, + "step": 8067 + }, + { + "epoch": 0.07003411428720237, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019755342970203186, + "loss": 0.1484, + "step": 8068 + }, + { + "epoch": 0.07004279476740653, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019755274041546615, + "loss": 0.1338, + "step": 8069 + }, + { + "epoch": 0.0700514752476107, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019755205103315406, + "loss": 0.1523, + "step": 8070 + }, + { + "epoch": 0.07006015572781486, + "grad_norm": 0.80859375, + "learning_rate": 0.001975513615550964, + "loss": 0.2021, + "step": 8071 + }, + { + "epoch": 0.07006883620801903, + "grad_norm": 0.306640625, + "learning_rate": 0.001975506719812938, + "loss": 0.1797, + "step": 8072 + }, + { + "epoch": 0.0700775166882232, + "grad_norm": 0.8203125, + "learning_rate": 0.0019754998231174717, + "loss": 0.1191, + "step": 8073 + }, + { + "epoch": 0.07008619716842736, + "grad_norm": 0.3046875, + "learning_rate": 0.0019754929254645714, + "loss": 0.0986, + "step": 8074 + }, + { + "epoch": 0.07009487764863152, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019754860268542456, + "loss": 0.1226, + "step": 8075 + }, + { + "epoch": 0.07010355812883569, + "grad_norm": 0.361328125, + "learning_rate": 0.0019754791272865017, + "loss": 0.1279, + "step": 8076 + }, + { + "epoch": 0.07011223860903985, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019754722267613465, + "loss": 0.1025, + "step": 8077 + }, + { + "epoch": 0.07012091908924402, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019754653252787883, + "loss": 0.0977, + "step": 8078 + }, + { + "epoch": 0.07012959956944818, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019754584228388345, + "loss": 0.1279, + "step": 8079 + }, + { + "epoch": 0.07013828004965235, + "grad_norm": 0.2470703125, + "learning_rate": 0.001975451519441492, + "loss": 0.1348, + "step": 8080 + }, + { + "epoch": 0.07014696052985651, + "grad_norm": 0.51171875, + "learning_rate": 0.0019754446150867687, + "loss": 0.1064, + "step": 8081 + }, + { + "epoch": 0.07015564101006068, + "grad_norm": 0.1796875, + "learning_rate": 0.001975437709774673, + "loss": 0.1289, + "step": 8082 + }, + { + "epoch": 0.07016432149026484, + "grad_norm": 0.56640625, + "learning_rate": 0.001975430803505212, + "loss": 0.1484, + "step": 8083 + }, + { + "epoch": 0.07017300197046901, + "grad_norm": 0.5546875, + "learning_rate": 0.0019754238962783922, + "loss": 0.0957, + "step": 8084 + }, + { + "epoch": 0.07018168245067317, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019754169880942226, + "loss": 0.1338, + "step": 8085 + }, + { + "epoch": 0.07019036293087734, + "grad_norm": 0.486328125, + "learning_rate": 0.0019754100789527097, + "loss": 0.1377, + "step": 8086 + }, + { + "epoch": 0.0701990434110815, + "grad_norm": 0.130859375, + "learning_rate": 0.0019754031688538616, + "loss": 0.1035, + "step": 8087 + }, + { + "epoch": 0.07020772389128567, + "grad_norm": 1.3203125, + "learning_rate": 0.001975396257797686, + "loss": 0.1611, + "step": 8088 + }, + { + "epoch": 0.07021640437148984, + "grad_norm": 0.490234375, + "learning_rate": 0.00197538934578419, + "loss": 0.1367, + "step": 8089 + }, + { + "epoch": 0.070225084851694, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019753824328133814, + "loss": 0.1318, + "step": 8090 + }, + { + "epoch": 0.07023376533189817, + "grad_norm": 0.10400390625, + "learning_rate": 0.001975375518885268, + "loss": 0.1621, + "step": 8091 + }, + { + "epoch": 0.07024244581210233, + "grad_norm": 0.6953125, + "learning_rate": 0.0019753686039998566, + "loss": 0.1641, + "step": 8092 + }, + { + "epoch": 0.0702511262923065, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019753616881571555, + "loss": 0.0928, + "step": 8093 + }, + { + "epoch": 0.07025980677251066, + "grad_norm": 0.48828125, + "learning_rate": 0.0019753547713571716, + "loss": 0.125, + "step": 8094 + }, + { + "epoch": 0.07026848725271483, + "grad_norm": 0.29296875, + "learning_rate": 0.0019753478535999135, + "loss": 0.1426, + "step": 8095 + }, + { + "epoch": 0.07027716773291899, + "grad_norm": 0.91796875, + "learning_rate": 0.0019753409348853878, + "loss": 0.1279, + "step": 8096 + }, + { + "epoch": 0.07028584821312316, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019753340152136025, + "loss": 0.1338, + "step": 8097 + }, + { + "epoch": 0.07029452869332732, + "grad_norm": 0.4453125, + "learning_rate": 0.001975327094584565, + "loss": 0.1631, + "step": 8098 + }, + { + "epoch": 0.07030320917353149, + "grad_norm": 0.296875, + "learning_rate": 0.001975320172998283, + "loss": 0.1963, + "step": 8099 + }, + { + "epoch": 0.07031188965373565, + "grad_norm": 0.27734375, + "learning_rate": 0.0019753132504547644, + "loss": 0.1719, + "step": 8100 + }, + { + "epoch": 0.07032057013393982, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019753063269540155, + "loss": 0.0923, + "step": 8101 + }, + { + "epoch": 0.07032925061414397, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019752994024960454, + "loss": 0.1143, + "step": 8102 + }, + { + "epoch": 0.07033793109434813, + "grad_norm": 0.58984375, + "learning_rate": 0.001975292477080861, + "loss": 0.1196, + "step": 8103 + }, + { + "epoch": 0.0703466115745523, + "grad_norm": 0.28515625, + "learning_rate": 0.0019752855507084695, + "loss": 0.1348, + "step": 8104 + }, + { + "epoch": 0.07035529205475646, + "grad_norm": 0.439453125, + "learning_rate": 0.0019752786233788793, + "loss": 0.1299, + "step": 8105 + }, + { + "epoch": 0.07036397253496063, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019752716950920974, + "loss": 0.1592, + "step": 8106 + }, + { + "epoch": 0.0703726530151648, + "grad_norm": 0.515625, + "learning_rate": 0.0019752647658481315, + "loss": 0.127, + "step": 8107 + }, + { + "epoch": 0.07038133349536896, + "grad_norm": 0.25390625, + "learning_rate": 0.0019752578356469893, + "loss": 0.1152, + "step": 8108 + }, + { + "epoch": 0.07039001397557312, + "grad_norm": 0.203125, + "learning_rate": 0.001975250904488678, + "loss": 0.1025, + "step": 8109 + }, + { + "epoch": 0.07039869445577729, + "grad_norm": 0.26953125, + "learning_rate": 0.0019752439723732056, + "loss": 0.125, + "step": 8110 + }, + { + "epoch": 0.07040737493598145, + "grad_norm": 1.65625, + "learning_rate": 0.0019752370393005796, + "loss": 0.1641, + "step": 8111 + }, + { + "epoch": 0.07041605541618562, + "grad_norm": 0.16796875, + "learning_rate": 0.001975230105270808, + "loss": 0.1387, + "step": 8112 + }, + { + "epoch": 0.07042473589638978, + "grad_norm": 0.142578125, + "learning_rate": 0.001975223170283897, + "loss": 0.1289, + "step": 8113 + }, + { + "epoch": 0.07043341637659395, + "grad_norm": 0.0673828125, + "learning_rate": 0.001975216234339856, + "loss": 0.0996, + "step": 8114 + }, + { + "epoch": 0.07044209685679811, + "grad_norm": 0.353515625, + "learning_rate": 0.0019752092974386908, + "loss": 0.0908, + "step": 8115 + }, + { + "epoch": 0.07045077733700228, + "grad_norm": 0.53515625, + "learning_rate": 0.0019752023595804103, + "loss": 0.1133, + "step": 8116 + }, + { + "epoch": 0.07045945781720644, + "grad_norm": 0.62890625, + "learning_rate": 0.0019751954207650217, + "loss": 0.1797, + "step": 8117 + }, + { + "epoch": 0.07046813829741061, + "grad_norm": 0.92578125, + "learning_rate": 0.0019751884809925324, + "loss": 0.1445, + "step": 8118 + }, + { + "epoch": 0.07047681877761477, + "grad_norm": 0.080078125, + "learning_rate": 0.00197518154026295, + "loss": 0.125, + "step": 8119 + }, + { + "epoch": 0.07048549925781894, + "grad_norm": 0.6015625, + "learning_rate": 0.0019751745985762824, + "loss": 0.1357, + "step": 8120 + }, + { + "epoch": 0.0704941797380231, + "grad_norm": 0.2119140625, + "learning_rate": 0.001975167655932537, + "loss": 0.1338, + "step": 8121 + }, + { + "epoch": 0.07050286021822727, + "grad_norm": 0.265625, + "learning_rate": 0.0019751607123317215, + "loss": 0.1377, + "step": 8122 + }, + { + "epoch": 0.07051154069843144, + "grad_norm": 0.1923828125, + "learning_rate": 0.001975153767773843, + "loss": 0.1787, + "step": 8123 + }, + { + "epoch": 0.0705202211786356, + "grad_norm": 0.2470703125, + "learning_rate": 0.00197514682225891, + "loss": 0.124, + "step": 8124 + }, + { + "epoch": 0.07052890165883977, + "grad_norm": 0.080078125, + "learning_rate": 0.0019751398757869294, + "loss": 0.1455, + "step": 8125 + }, + { + "epoch": 0.07053758213904393, + "grad_norm": 0.16015625, + "learning_rate": 0.001975132928357909, + "loss": 0.1035, + "step": 8126 + }, + { + "epoch": 0.0705462626192481, + "grad_norm": 0.60546875, + "learning_rate": 0.001975125979971856, + "loss": 0.1152, + "step": 8127 + }, + { + "epoch": 0.07055494309945226, + "grad_norm": 1.3671875, + "learning_rate": 0.001975119030628779, + "loss": 0.1533, + "step": 8128 + }, + { + "epoch": 0.07056362357965643, + "grad_norm": 0.353515625, + "learning_rate": 0.0019751120803286845, + "loss": 0.1523, + "step": 8129 + }, + { + "epoch": 0.07057230405986059, + "grad_norm": 0.458984375, + "learning_rate": 0.001975105129071581, + "loss": 0.1465, + "step": 8130 + }, + { + "epoch": 0.07058098454006476, + "grad_norm": 0.13671875, + "learning_rate": 0.0019750981768574755, + "loss": 0.1475, + "step": 8131 + }, + { + "epoch": 0.07058966502026892, + "grad_norm": 0.1982421875, + "learning_rate": 0.001975091223686376, + "loss": 0.1162, + "step": 8132 + }, + { + "epoch": 0.07059834550047309, + "grad_norm": 0.44921875, + "learning_rate": 0.001975084269558289, + "loss": 0.1494, + "step": 8133 + }, + { + "epoch": 0.07060702598067725, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019750773144732235, + "loss": 0.1279, + "step": 8134 + }, + { + "epoch": 0.07061570646088142, + "grad_norm": 0.48046875, + "learning_rate": 0.001975070358431187, + "loss": 0.126, + "step": 8135 + }, + { + "epoch": 0.07062438694108558, + "grad_norm": 0.52734375, + "learning_rate": 0.0019750634014321865, + "loss": 0.126, + "step": 8136 + }, + { + "epoch": 0.07063306742128975, + "grad_norm": 0.37890625, + "learning_rate": 0.00197505644347623, + "loss": 0.0933, + "step": 8137 + }, + { + "epoch": 0.07064174790149391, + "grad_norm": 0.90234375, + "learning_rate": 0.0019750494845633244, + "loss": 0.1211, + "step": 8138 + }, + { + "epoch": 0.07065042838169808, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019750425246934777, + "loss": 0.1357, + "step": 8139 + }, + { + "epoch": 0.07065910886190224, + "grad_norm": 0.09765625, + "learning_rate": 0.001975035563866698, + "loss": 0.1201, + "step": 8140 + }, + { + "epoch": 0.07066778934210641, + "grad_norm": 0.375, + "learning_rate": 0.001975028602082993, + "loss": 0.1514, + "step": 8141 + }, + { + "epoch": 0.07067646982231057, + "grad_norm": 0.2734375, + "learning_rate": 0.001975021639342369, + "loss": 0.1079, + "step": 8142 + }, + { + "epoch": 0.07068515030251474, + "grad_norm": 0.162109375, + "learning_rate": 0.001975014675644835, + "loss": 0.1172, + "step": 8143 + }, + { + "epoch": 0.0706938307827189, + "grad_norm": 0.33203125, + "learning_rate": 0.001975007710990398, + "loss": 0.1602, + "step": 8144 + }, + { + "epoch": 0.07070251126292307, + "grad_norm": 0.314453125, + "learning_rate": 0.0019750007453790655, + "loss": 0.1572, + "step": 8145 + }, + { + "epoch": 0.07071119174312723, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019749937788108456, + "loss": 0.1182, + "step": 8146 + }, + { + "epoch": 0.0707198722233314, + "grad_norm": 0.330078125, + "learning_rate": 0.001974986811285746, + "loss": 0.1504, + "step": 8147 + }, + { + "epoch": 0.07072855270353556, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019749798428037734, + "loss": 0.1123, + "step": 8148 + }, + { + "epoch": 0.07073723318373973, + "grad_norm": 0.455078125, + "learning_rate": 0.0019749728733649365, + "loss": 0.1299, + "step": 8149 + }, + { + "epoch": 0.0707459136639439, + "grad_norm": 0.1201171875, + "learning_rate": 0.001974965902969242, + "loss": 0.0986, + "step": 8150 + }, + { + "epoch": 0.07075459414414806, + "grad_norm": 0.177734375, + "learning_rate": 0.0019749589316166976, + "loss": 0.1328, + "step": 8151 + }, + { + "epoch": 0.07076327462435222, + "grad_norm": 0.34375, + "learning_rate": 0.001974951959307312, + "loss": 0.0913, + "step": 8152 + }, + { + "epoch": 0.07077195510455639, + "grad_norm": 0.494140625, + "learning_rate": 0.0019749449860410914, + "loss": 0.1318, + "step": 8153 + }, + { + "epoch": 0.07078063558476055, + "grad_norm": 0.58984375, + "learning_rate": 0.0019749380118180447, + "loss": 0.1182, + "step": 8154 + }, + { + "epoch": 0.07078931606496472, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019749310366381787, + "loss": 0.1084, + "step": 8155 + }, + { + "epoch": 0.07079799654516888, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019749240605015012, + "loss": 0.125, + "step": 8156 + }, + { + "epoch": 0.07080667702537305, + "grad_norm": 0.52734375, + "learning_rate": 0.00197491708340802, + "loss": 0.1543, + "step": 8157 + }, + { + "epoch": 0.07081535750557721, + "grad_norm": 0.15234375, + "learning_rate": 0.0019749101053577424, + "loss": 0.1465, + "step": 8158 + }, + { + "epoch": 0.07082403798578138, + "grad_norm": 0.091796875, + "learning_rate": 0.0019749031263506765, + "loss": 0.1387, + "step": 8159 + }, + { + "epoch": 0.07083271846598554, + "grad_norm": 0.150390625, + "learning_rate": 0.0019748961463868296, + "loss": 0.1406, + "step": 8160 + }, + { + "epoch": 0.07084139894618971, + "grad_norm": 0.3671875, + "learning_rate": 0.0019748891654662094, + "loss": 0.0967, + "step": 8161 + }, + { + "epoch": 0.07085007942639387, + "grad_norm": 0.625, + "learning_rate": 0.0019748821835888237, + "loss": 0.1318, + "step": 8162 + }, + { + "epoch": 0.07085875990659803, + "grad_norm": 0.37890625, + "learning_rate": 0.00197487520075468, + "loss": 0.1406, + "step": 8163 + }, + { + "epoch": 0.07086744038680219, + "grad_norm": 0.640625, + "learning_rate": 0.0019748682169637858, + "loss": 0.1914, + "step": 8164 + }, + { + "epoch": 0.07087612086700636, + "grad_norm": 0.13671875, + "learning_rate": 0.0019748612322161487, + "loss": 0.0835, + "step": 8165 + }, + { + "epoch": 0.07088480134721052, + "grad_norm": 0.43359375, + "learning_rate": 0.0019748542465117765, + "loss": 0.1123, + "step": 8166 + }, + { + "epoch": 0.07089348182741469, + "grad_norm": 0.09912109375, + "learning_rate": 0.001974847259850677, + "loss": 0.1084, + "step": 8167 + }, + { + "epoch": 0.07090216230761885, + "grad_norm": 0.072265625, + "learning_rate": 0.001974840272232858, + "loss": 0.1094, + "step": 8168 + }, + { + "epoch": 0.07091084278782302, + "grad_norm": 0.1455078125, + "learning_rate": 0.001974833283658326, + "loss": 0.1709, + "step": 8169 + }, + { + "epoch": 0.07091952326802718, + "grad_norm": 0.625, + "learning_rate": 0.0019748262941270906, + "loss": 0.1523, + "step": 8170 + }, + { + "epoch": 0.07092820374823135, + "grad_norm": 0.138671875, + "learning_rate": 0.0019748193036391577, + "loss": 0.126, + "step": 8171 + }, + { + "epoch": 0.07093688422843551, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019748123121945352, + "loss": 0.1133, + "step": 8172 + }, + { + "epoch": 0.07094556470863968, + "grad_norm": 0.30859375, + "learning_rate": 0.0019748053197932314, + "loss": 0.1113, + "step": 8173 + }, + { + "epoch": 0.07095424518884384, + "grad_norm": 0.69921875, + "learning_rate": 0.0019747983264352537, + "loss": 0.1006, + "step": 8174 + }, + { + "epoch": 0.07096292566904801, + "grad_norm": 0.32421875, + "learning_rate": 0.0019747913321206097, + "loss": 0.2168, + "step": 8175 + }, + { + "epoch": 0.07097160614925217, + "grad_norm": 0.25, + "learning_rate": 0.0019747843368493074, + "loss": 0.1201, + "step": 8176 + }, + { + "epoch": 0.07098028662945634, + "grad_norm": 0.302734375, + "learning_rate": 0.001974777340621353, + "loss": 0.1348, + "step": 8177 + }, + { + "epoch": 0.0709889671096605, + "grad_norm": 0.353515625, + "learning_rate": 0.0019747703434367567, + "loss": 0.123, + "step": 8178 + }, + { + "epoch": 0.07099764758986467, + "grad_norm": 0.474609375, + "learning_rate": 0.0019747633452955235, + "loss": 0.1396, + "step": 8179 + }, + { + "epoch": 0.07100632807006883, + "grad_norm": 0.10986328125, + "learning_rate": 0.001974756346197663, + "loss": 0.1191, + "step": 8180 + }, + { + "epoch": 0.071015008550273, + "grad_norm": 0.90625, + "learning_rate": 0.001974749346143182, + "loss": 0.1729, + "step": 8181 + }, + { + "epoch": 0.07102368903047716, + "grad_norm": 0.12890625, + "learning_rate": 0.0019747423451320877, + "loss": 0.1338, + "step": 8182 + }, + { + "epoch": 0.07103236951068133, + "grad_norm": 0.8984375, + "learning_rate": 0.0019747353431643884, + "loss": 0.1416, + "step": 8183 + }, + { + "epoch": 0.0710410499908855, + "grad_norm": 0.27734375, + "learning_rate": 0.001974728340240092, + "loss": 0.0986, + "step": 8184 + }, + { + "epoch": 0.07104973047108966, + "grad_norm": 0.66796875, + "learning_rate": 0.001974721336359206, + "loss": 0.1523, + "step": 8185 + }, + { + "epoch": 0.07105841095129382, + "grad_norm": 0.125, + "learning_rate": 0.001974714331521738, + "loss": 0.2012, + "step": 8186 + }, + { + "epoch": 0.07106709143149799, + "grad_norm": 0.388671875, + "learning_rate": 0.0019747073257276953, + "loss": 0.125, + "step": 8187 + }, + { + "epoch": 0.07107577191170215, + "grad_norm": 0.640625, + "learning_rate": 0.0019747003189770854, + "loss": 0.1162, + "step": 8188 + }, + { + "epoch": 0.07108445239190632, + "grad_norm": 0.1396484375, + "learning_rate": 0.001974693311269917, + "loss": 0.1387, + "step": 8189 + }, + { + "epoch": 0.07109313287211048, + "grad_norm": 0.7578125, + "learning_rate": 0.0019746863026061964, + "loss": 0.1123, + "step": 8190 + }, + { + "epoch": 0.07110181335231465, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019746792929859324, + "loss": 0.1279, + "step": 8191 + }, + { + "epoch": 0.07111049383251881, + "grad_norm": 0.400390625, + "learning_rate": 0.0019746722824091323, + "loss": 0.1426, + "step": 8192 + }, + { + "epoch": 0.07111917431272298, + "grad_norm": 0.1572265625, + "learning_rate": 0.001974665270875804, + "loss": 0.1621, + "step": 8193 + }, + { + "epoch": 0.07112785479292714, + "grad_norm": 0.734375, + "learning_rate": 0.0019746582583859544, + "loss": 0.1084, + "step": 8194 + }, + { + "epoch": 0.07113653527313131, + "grad_norm": 0.10107421875, + "learning_rate": 0.001974651244939592, + "loss": 0.168, + "step": 8195 + }, + { + "epoch": 0.07114521575333548, + "grad_norm": 1.03125, + "learning_rate": 0.001974644230536724, + "loss": 0.1436, + "step": 8196 + }, + { + "epoch": 0.07115389623353964, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019746372151773586, + "loss": 0.1157, + "step": 8197 + }, + { + "epoch": 0.0711625767137438, + "grad_norm": 0.314453125, + "learning_rate": 0.0019746301988615026, + "loss": 0.1221, + "step": 8198 + }, + { + "epoch": 0.07117125719394797, + "grad_norm": 0.85546875, + "learning_rate": 0.0019746231815891643, + "loss": 0.1025, + "step": 8199 + }, + { + "epoch": 0.07117993767415214, + "grad_norm": 0.2353515625, + "learning_rate": 0.001974616163360351, + "loss": 0.1826, + "step": 8200 + }, + { + "epoch": 0.0711886181543563, + "grad_norm": 0.09521484375, + "learning_rate": 0.001974609144175071, + "loss": 0.1201, + "step": 8201 + }, + { + "epoch": 0.07119729863456047, + "grad_norm": 0.97265625, + "learning_rate": 0.0019746021240333316, + "loss": 0.1348, + "step": 8202 + }, + { + "epoch": 0.07120597911476463, + "grad_norm": 0.1630859375, + "learning_rate": 0.00197459510293514, + "loss": 0.1133, + "step": 8203 + }, + { + "epoch": 0.0712146595949688, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019745880808805048, + "loss": 0.125, + "step": 8204 + }, + { + "epoch": 0.07122334007517296, + "grad_norm": 0.1630859375, + "learning_rate": 0.001974581057869433, + "loss": 0.1416, + "step": 8205 + }, + { + "epoch": 0.07123202055537713, + "grad_norm": 0.314453125, + "learning_rate": 0.0019745740339019327, + "loss": 0.1113, + "step": 8206 + }, + { + "epoch": 0.07124070103558129, + "grad_norm": 0.6640625, + "learning_rate": 0.0019745670089780106, + "loss": 0.1523, + "step": 8207 + }, + { + "epoch": 0.07124938151578546, + "grad_norm": 0.171875, + "learning_rate": 0.001974559983097676, + "loss": 0.1436, + "step": 8208 + }, + { + "epoch": 0.07125806199598962, + "grad_norm": 0.578125, + "learning_rate": 0.0019745529562609355, + "loss": 0.0967, + "step": 8209 + }, + { + "epoch": 0.07126674247619379, + "grad_norm": 0.6171875, + "learning_rate": 0.0019745459284677973, + "loss": 0.1113, + "step": 8210 + }, + { + "epoch": 0.07127542295639795, + "grad_norm": 0.7421875, + "learning_rate": 0.0019745388997182683, + "loss": 0.1562, + "step": 8211 + }, + { + "epoch": 0.07128410343660212, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019745318700123567, + "loss": 0.0933, + "step": 8212 + }, + { + "epoch": 0.07129278391680628, + "grad_norm": 0.333984375, + "learning_rate": 0.001974524839350071, + "loss": 0.1118, + "step": 8213 + }, + { + "epoch": 0.07130146439701045, + "grad_norm": 0.251953125, + "learning_rate": 0.001974517807731417, + "loss": 0.125, + "step": 8214 + }, + { + "epoch": 0.07131014487721461, + "grad_norm": 0.431640625, + "learning_rate": 0.0019745107751564044, + "loss": 0.1396, + "step": 8215 + }, + { + "epoch": 0.07131882535741878, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019745037416250394, + "loss": 0.1406, + "step": 8216 + }, + { + "epoch": 0.07132750583762294, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019744967071373305, + "loss": 0.1572, + "step": 8217 + }, + { + "epoch": 0.07133618631782711, + "grad_norm": 0.390625, + "learning_rate": 0.001974489671693285, + "loss": 0.1504, + "step": 8218 + }, + { + "epoch": 0.07134486679803127, + "grad_norm": 0.37109375, + "learning_rate": 0.001974482635292911, + "loss": 0.1138, + "step": 8219 + }, + { + "epoch": 0.07135354727823544, + "grad_norm": 0.2119140625, + "learning_rate": 0.001974475597936215, + "loss": 0.123, + "step": 8220 + }, + { + "epoch": 0.0713622277584396, + "grad_norm": 0.70703125, + "learning_rate": 0.0019744685596232068, + "loss": 0.1523, + "step": 8221 + }, + { + "epoch": 0.07137090823864377, + "grad_norm": 0.3515625, + "learning_rate": 0.001974461520353892, + "loss": 0.1934, + "step": 8222 + }, + { + "epoch": 0.07137958871884793, + "grad_norm": 0.796875, + "learning_rate": 0.0019744544801282796, + "loss": 0.1201, + "step": 8223 + }, + { + "epoch": 0.0713882691990521, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019744474389463773, + "loss": 0.1553, + "step": 8224 + }, + { + "epoch": 0.07139694967925625, + "grad_norm": 0.14453125, + "learning_rate": 0.001974440396808192, + "loss": 0.1611, + "step": 8225 + }, + { + "epoch": 0.07140563015946041, + "grad_norm": 1.046875, + "learning_rate": 0.001974433353713732, + "loss": 0.1543, + "step": 8226 + }, + { + "epoch": 0.07141431063966458, + "grad_norm": 0.38671875, + "learning_rate": 0.0019744263096630045, + "loss": 0.1133, + "step": 8227 + }, + { + "epoch": 0.07142299111986875, + "grad_norm": 0.341796875, + "learning_rate": 0.001974419264656018, + "loss": 0.1816, + "step": 8228 + }, + { + "epoch": 0.07143167160007291, + "grad_norm": 0.50390625, + "learning_rate": 0.001974412218692779, + "loss": 0.1055, + "step": 8229 + }, + { + "epoch": 0.07144035208027708, + "grad_norm": 0.216796875, + "learning_rate": 0.001974405171773297, + "loss": 0.1016, + "step": 8230 + }, + { + "epoch": 0.07144903256048124, + "grad_norm": 0.267578125, + "learning_rate": 0.001974398123897578, + "loss": 0.1318, + "step": 8231 + }, + { + "epoch": 0.0714577130406854, + "grad_norm": 0.640625, + "learning_rate": 0.00197439107506563, + "loss": 0.1396, + "step": 8232 + }, + { + "epoch": 0.07146639352088957, + "grad_norm": 0.20703125, + "learning_rate": 0.001974384025277462, + "loss": 0.1064, + "step": 8233 + }, + { + "epoch": 0.07147507400109374, + "grad_norm": 0.4609375, + "learning_rate": 0.00197437697453308, + "loss": 0.1455, + "step": 8234 + }, + { + "epoch": 0.0714837544812979, + "grad_norm": 1.1015625, + "learning_rate": 0.0019743699228324924, + "loss": 0.1523, + "step": 8235 + }, + { + "epoch": 0.07149243496150207, + "grad_norm": 0.74609375, + "learning_rate": 0.0019743628701757074, + "loss": 0.1089, + "step": 8236 + }, + { + "epoch": 0.07150111544170623, + "grad_norm": 0.72265625, + "learning_rate": 0.001974355816562732, + "loss": 0.1533, + "step": 8237 + }, + { + "epoch": 0.0715097959219104, + "grad_norm": 0.498046875, + "learning_rate": 0.001974348761993575, + "loss": 0.1377, + "step": 8238 + }, + { + "epoch": 0.07151847640211456, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019743417064682425, + "loss": 0.1357, + "step": 8239 + }, + { + "epoch": 0.07152715688231873, + "grad_norm": 0.1328125, + "learning_rate": 0.0019743346499867436, + "loss": 0.1465, + "step": 8240 + }, + { + "epoch": 0.07153583736252289, + "grad_norm": 0.2578125, + "learning_rate": 0.001974327592549085, + "loss": 0.1455, + "step": 8241 + }, + { + "epoch": 0.07154451784272706, + "grad_norm": 0.62109375, + "learning_rate": 0.0019743205341552746, + "loss": 0.1211, + "step": 8242 + }, + { + "epoch": 0.07155319832293122, + "grad_norm": 0.37109375, + "learning_rate": 0.001974313474805321, + "loss": 0.1074, + "step": 8243 + }, + { + "epoch": 0.07156187880313539, + "grad_norm": 0.462890625, + "learning_rate": 0.0019743064144992313, + "loss": 0.1299, + "step": 8244 + }, + { + "epoch": 0.07157055928333955, + "grad_norm": 0.14453125, + "learning_rate": 0.001974299353237013, + "loss": 0.1318, + "step": 8245 + }, + { + "epoch": 0.07157923976354372, + "grad_norm": 0.337890625, + "learning_rate": 0.0019742922910186743, + "loss": 0.1216, + "step": 8246 + }, + { + "epoch": 0.07158792024374788, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019742852278442223, + "loss": 0.1045, + "step": 8247 + }, + { + "epoch": 0.07159660072395205, + "grad_norm": 0.076171875, + "learning_rate": 0.0019742781637136653, + "loss": 0.1211, + "step": 8248 + }, + { + "epoch": 0.07160528120415621, + "grad_norm": 0.421875, + "learning_rate": 0.001974271098627011, + "loss": 0.1582, + "step": 8249 + }, + { + "epoch": 0.07161396168436038, + "grad_norm": 0.25390625, + "learning_rate": 0.0019742640325842668, + "loss": 0.167, + "step": 8250 + }, + { + "epoch": 0.07162264216456454, + "grad_norm": 0.333984375, + "learning_rate": 0.001974256965585441, + "loss": 0.1689, + "step": 8251 + }, + { + "epoch": 0.07163132264476871, + "grad_norm": 0.302734375, + "learning_rate": 0.0019742498976305406, + "loss": 0.1133, + "step": 8252 + }, + { + "epoch": 0.07164000312497287, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019742428287195736, + "loss": 0.1143, + "step": 8253 + }, + { + "epoch": 0.07164868360517704, + "grad_norm": 0.38671875, + "learning_rate": 0.0019742357588525476, + "loss": 0.104, + "step": 8254 + }, + { + "epoch": 0.0716573640853812, + "grad_norm": 0.1455078125, + "learning_rate": 0.001974228688029471, + "loss": 0.127, + "step": 8255 + }, + { + "epoch": 0.07166604456558537, + "grad_norm": 0.173828125, + "learning_rate": 0.0019742216162503506, + "loss": 0.1758, + "step": 8256 + }, + { + "epoch": 0.07167472504578953, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019742145435151944, + "loss": 0.1299, + "step": 8257 + }, + { + "epoch": 0.0716834055259937, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019742074698240113, + "loss": 0.1309, + "step": 8258 + }, + { + "epoch": 0.07169208600619786, + "grad_norm": 0.314453125, + "learning_rate": 0.0019742003951768072, + "loss": 0.0986, + "step": 8259 + }, + { + "epoch": 0.07170076648640203, + "grad_norm": 0.283203125, + "learning_rate": 0.0019741933195735906, + "loss": 0.1348, + "step": 8260 + }, + { + "epoch": 0.0717094469666062, + "grad_norm": 0.75390625, + "learning_rate": 0.0019741862430143696, + "loss": 0.1191, + "step": 8261 + }, + { + "epoch": 0.07171812744681036, + "grad_norm": 0.130859375, + "learning_rate": 0.0019741791654991517, + "loss": 0.1172, + "step": 8262 + }, + { + "epoch": 0.07172680792701452, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019741720870279445, + "loss": 0.0942, + "step": 8263 + }, + { + "epoch": 0.07173548840721869, + "grad_norm": 0.40234375, + "learning_rate": 0.001974165007600756, + "loss": 0.1348, + "step": 8264 + }, + { + "epoch": 0.07174416888742285, + "grad_norm": 0.236328125, + "learning_rate": 0.0019741579272175938, + "loss": 0.1553, + "step": 8265 + }, + { + "epoch": 0.07175284936762702, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019741508458784654, + "loss": 0.1104, + "step": 8266 + }, + { + "epoch": 0.07176152984783118, + "grad_norm": 0.66015625, + "learning_rate": 0.0019741437635833786, + "loss": 0.1406, + "step": 8267 + }, + { + "epoch": 0.07177021032803535, + "grad_norm": 0.36328125, + "learning_rate": 0.0019741366803323417, + "loss": 0.1348, + "step": 8268 + }, + { + "epoch": 0.07177889080823951, + "grad_norm": 0.1669921875, + "learning_rate": 0.001974129596125362, + "loss": 0.1406, + "step": 8269 + }, + { + "epoch": 0.07178757128844368, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019741225109624473, + "loss": 0.1074, + "step": 8270 + }, + { + "epoch": 0.07179625176864785, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019741154248436054, + "loss": 0.105, + "step": 8271 + }, + { + "epoch": 0.07180493224885201, + "grad_norm": 0.1953125, + "learning_rate": 0.0019741083377688437, + "loss": 0.1523, + "step": 8272 + }, + { + "epoch": 0.07181361272905618, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019741012497381705, + "loss": 0.1504, + "step": 8273 + }, + { + "epoch": 0.07182229320926034, + "grad_norm": 0.294921875, + "learning_rate": 0.001974094160751593, + "loss": 0.1289, + "step": 8274 + }, + { + "epoch": 0.0718309736894645, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019740870708091197, + "loss": 0.1235, + "step": 8275 + }, + { + "epoch": 0.07183965416966867, + "grad_norm": 0.91796875, + "learning_rate": 0.0019740799799107577, + "loss": 0.1309, + "step": 8276 + }, + { + "epoch": 0.07184833464987284, + "grad_norm": 0.134765625, + "learning_rate": 0.001974072888056515, + "loss": 0.1079, + "step": 8277 + }, + { + "epoch": 0.071857015130077, + "grad_norm": 0.66015625, + "learning_rate": 0.0019740657952463992, + "loss": 0.1338, + "step": 8278 + }, + { + "epoch": 0.07186569561028117, + "grad_norm": 0.578125, + "learning_rate": 0.0019740587014804183, + "loss": 0.125, + "step": 8279 + }, + { + "epoch": 0.07187437609048533, + "grad_norm": 0.12255859375, + "learning_rate": 0.00197405160675858, + "loss": 0.1172, + "step": 8280 + }, + { + "epoch": 0.0718830565706895, + "grad_norm": 0.265625, + "learning_rate": 0.001974044511080892, + "loss": 0.1309, + "step": 8281 + }, + { + "epoch": 0.07189173705089366, + "grad_norm": 0.291015625, + "learning_rate": 0.001974037414447362, + "loss": 0.1562, + "step": 8282 + }, + { + "epoch": 0.07190041753109783, + "grad_norm": 0.158203125, + "learning_rate": 0.0019740303168579976, + "loss": 0.1318, + "step": 8283 + }, + { + "epoch": 0.07190909801130199, + "grad_norm": 0.10888671875, + "learning_rate": 0.001974023218312807, + "loss": 0.1084, + "step": 8284 + }, + { + "epoch": 0.07191777849150616, + "grad_norm": 0.1865234375, + "learning_rate": 0.001974016118811798, + "loss": 0.1226, + "step": 8285 + }, + { + "epoch": 0.07192645897171031, + "grad_norm": 0.28515625, + "learning_rate": 0.001974009018354978, + "loss": 0.1592, + "step": 8286 + }, + { + "epoch": 0.07193513945191447, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019740019169423545, + "loss": 0.1445, + "step": 8287 + }, + { + "epoch": 0.07194381993211864, + "grad_norm": 0.310546875, + "learning_rate": 0.001973994814573936, + "loss": 0.127, + "step": 8288 + }, + { + "epoch": 0.0719525004123228, + "grad_norm": 1.1796875, + "learning_rate": 0.001973987711249729, + "loss": 0.1104, + "step": 8289 + }, + { + "epoch": 0.07196118089252697, + "grad_norm": 0.6328125, + "learning_rate": 0.0019739806069697433, + "loss": 0.1357, + "step": 8290 + }, + { + "epoch": 0.07196986137273113, + "grad_norm": 0.5390625, + "learning_rate": 0.0019739735017339855, + "loss": 0.1133, + "step": 8291 + }, + { + "epoch": 0.0719785418529353, + "grad_norm": 0.2412109375, + "learning_rate": 0.001973966395542463, + "loss": 0.0801, + "step": 8292 + }, + { + "epoch": 0.07198722233313946, + "grad_norm": 0.189453125, + "learning_rate": 0.001973959288395184, + "loss": 0.0986, + "step": 8293 + }, + { + "epoch": 0.07199590281334363, + "grad_norm": 0.2158203125, + "learning_rate": 0.001973952180292156, + "loss": 0.1055, + "step": 8294 + }, + { + "epoch": 0.0720045832935478, + "grad_norm": 0.40625, + "learning_rate": 0.0019739450712333876, + "loss": 0.1011, + "step": 8295 + }, + { + "epoch": 0.07201326377375196, + "grad_norm": 0.474609375, + "learning_rate": 0.0019739379612188856, + "loss": 0.1484, + "step": 8296 + }, + { + "epoch": 0.07202194425395612, + "grad_norm": 0.453125, + "learning_rate": 0.001973930850248658, + "loss": 0.082, + "step": 8297 + }, + { + "epoch": 0.07203062473416029, + "grad_norm": 0.78125, + "learning_rate": 0.001973923738322713, + "loss": 0.1045, + "step": 8298 + }, + { + "epoch": 0.07203930521436445, + "grad_norm": 0.37109375, + "learning_rate": 0.0019739166254410584, + "loss": 0.1406, + "step": 8299 + }, + { + "epoch": 0.07204798569456862, + "grad_norm": 0.54296875, + "learning_rate": 0.0019739095116037015, + "loss": 0.1299, + "step": 8300 + }, + { + "epoch": 0.07205666617477278, + "grad_norm": 0.640625, + "learning_rate": 0.0019739023968106503, + "loss": 0.0957, + "step": 8301 + }, + { + "epoch": 0.07206534665497695, + "grad_norm": 0.9921875, + "learning_rate": 0.0019738952810619127, + "loss": 0.1484, + "step": 8302 + }, + { + "epoch": 0.07207402713518112, + "grad_norm": 0.07470703125, + "learning_rate": 0.001973888164357496, + "loss": 0.125, + "step": 8303 + }, + { + "epoch": 0.07208270761538528, + "grad_norm": 0.451171875, + "learning_rate": 0.0019738810466974088, + "loss": 0.1152, + "step": 8304 + }, + { + "epoch": 0.07209138809558945, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019738739280816577, + "loss": 0.1182, + "step": 8305 + }, + { + "epoch": 0.07210006857579361, + "grad_norm": 0.134765625, + "learning_rate": 0.0019738668085102517, + "loss": 0.126, + "step": 8306 + }, + { + "epoch": 0.07210874905599778, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019738596879831983, + "loss": 0.1377, + "step": 8307 + }, + { + "epoch": 0.07211742953620194, + "grad_norm": 0.189453125, + "learning_rate": 0.001973852566500505, + "loss": 0.1289, + "step": 8308 + }, + { + "epoch": 0.0721261100164061, + "grad_norm": 0.109375, + "learning_rate": 0.001973845444062179, + "loss": 0.1016, + "step": 8309 + }, + { + "epoch": 0.07213479049661027, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019738383206682297, + "loss": 0.1523, + "step": 8310 + }, + { + "epoch": 0.07214347097681444, + "grad_norm": 0.255859375, + "learning_rate": 0.0019738311963186637, + "loss": 0.1289, + "step": 8311 + }, + { + "epoch": 0.0721521514570186, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019738240710134884, + "loss": 0.1514, + "step": 8312 + }, + { + "epoch": 0.07216083193722277, + "grad_norm": 0.058837890625, + "learning_rate": 0.0019738169447527125, + "loss": 0.1001, + "step": 8313 + }, + { + "epoch": 0.07216951241742693, + "grad_norm": 0.31640625, + "learning_rate": 0.0019738098175363437, + "loss": 0.1484, + "step": 8314 + }, + { + "epoch": 0.0721781928976311, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019738026893643896, + "loss": 0.1001, + "step": 8315 + }, + { + "epoch": 0.07218687337783526, + "grad_norm": 0.490234375, + "learning_rate": 0.001973795560236858, + "loss": 0.1484, + "step": 8316 + }, + { + "epoch": 0.07219555385803943, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019737884301537566, + "loss": 0.1191, + "step": 8317 + }, + { + "epoch": 0.07220423433824359, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019737812991150933, + "loss": 0.1357, + "step": 8318 + }, + { + "epoch": 0.07221291481844776, + "grad_norm": 0.275390625, + "learning_rate": 0.001973774167120876, + "loss": 0.1143, + "step": 8319 + }, + { + "epoch": 0.07222159529865192, + "grad_norm": 0.216796875, + "learning_rate": 0.0019737670341711125, + "loss": 0.1187, + "step": 8320 + }, + { + "epoch": 0.07223027577885609, + "grad_norm": 0.55078125, + "learning_rate": 0.00197375990026581, + "loss": 0.127, + "step": 8321 + }, + { + "epoch": 0.07223895625906025, + "grad_norm": 0.255859375, + "learning_rate": 0.0019737527654049773, + "loss": 0.127, + "step": 8322 + }, + { + "epoch": 0.07224763673926442, + "grad_norm": 0.640625, + "learning_rate": 0.001973745629588622, + "loss": 0.1123, + "step": 8323 + }, + { + "epoch": 0.07225631721946858, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019737384928167506, + "loss": 0.1172, + "step": 8324 + }, + { + "epoch": 0.07226499769967275, + "grad_norm": 0.2431640625, + "learning_rate": 0.001973731355089373, + "loss": 0.1484, + "step": 8325 + }, + { + "epoch": 0.07227367817987691, + "grad_norm": 0.259765625, + "learning_rate": 0.001973724216406495, + "loss": 0.1133, + "step": 8326 + }, + { + "epoch": 0.07228235866008108, + "grad_norm": 0.1328125, + "learning_rate": 0.001973717076768126, + "loss": 0.1416, + "step": 8327 + }, + { + "epoch": 0.07229103914028524, + "grad_norm": 0.283203125, + "learning_rate": 0.001973709936174273, + "loss": 0.1504, + "step": 8328 + }, + { + "epoch": 0.07229971962048941, + "grad_norm": 0.470703125, + "learning_rate": 0.0019737027946249435, + "loss": 0.1045, + "step": 8329 + }, + { + "epoch": 0.07230840010069357, + "grad_norm": 0.1611328125, + "learning_rate": 0.001973695652120146, + "loss": 0.127, + "step": 8330 + }, + { + "epoch": 0.07231708058089774, + "grad_norm": 0.130859375, + "learning_rate": 0.0019736885086598884, + "loss": 0.1289, + "step": 8331 + }, + { + "epoch": 0.0723257610611019, + "grad_norm": 1.078125, + "learning_rate": 0.001973681364244178, + "loss": 0.1816, + "step": 8332 + }, + { + "epoch": 0.07233444154130607, + "grad_norm": 0.8203125, + "learning_rate": 0.0019736742188730225, + "loss": 0.1221, + "step": 8333 + }, + { + "epoch": 0.07234312202151023, + "grad_norm": 0.193359375, + "learning_rate": 0.0019736670725464304, + "loss": 0.1211, + "step": 8334 + }, + { + "epoch": 0.0723518025017144, + "grad_norm": 0.7265625, + "learning_rate": 0.0019736599252644085, + "loss": 0.1162, + "step": 8335 + }, + { + "epoch": 0.07236048298191856, + "grad_norm": 0.60546875, + "learning_rate": 0.0019736527770269656, + "loss": 0.1553, + "step": 8336 + }, + { + "epoch": 0.07236916346212273, + "grad_norm": 0.53125, + "learning_rate": 0.0019736456278341094, + "loss": 0.1523, + "step": 8337 + }, + { + "epoch": 0.0723778439423269, + "grad_norm": 1.2421875, + "learning_rate": 0.0019736384776858473, + "loss": 0.1367, + "step": 8338 + }, + { + "epoch": 0.07238652442253106, + "grad_norm": 0.1630859375, + "learning_rate": 0.001973631326582187, + "loss": 0.1235, + "step": 8339 + }, + { + "epoch": 0.07239520490273522, + "grad_norm": 0.5625, + "learning_rate": 0.001973624174523137, + "loss": 0.1406, + "step": 8340 + }, + { + "epoch": 0.07240388538293939, + "grad_norm": 0.7109375, + "learning_rate": 0.0019736170215087045, + "loss": 0.1387, + "step": 8341 + }, + { + "epoch": 0.07241256586314355, + "grad_norm": 0.4140625, + "learning_rate": 0.001973609867538898, + "loss": 0.1289, + "step": 8342 + }, + { + "epoch": 0.07242124634334772, + "grad_norm": 0.28515625, + "learning_rate": 0.001973602712613724, + "loss": 0.1289, + "step": 8343 + }, + { + "epoch": 0.07242992682355189, + "grad_norm": 0.283203125, + "learning_rate": 0.001973595556733192, + "loss": 0.1309, + "step": 8344 + }, + { + "epoch": 0.07243860730375605, + "grad_norm": 0.609375, + "learning_rate": 0.0019735883998973085, + "loss": 0.1377, + "step": 8345 + }, + { + "epoch": 0.07244728778396022, + "grad_norm": 0.134765625, + "learning_rate": 0.001973581242106082, + "loss": 0.1445, + "step": 8346 + }, + { + "epoch": 0.07245596826416438, + "grad_norm": 0.5625, + "learning_rate": 0.0019735740833595203, + "loss": 0.1436, + "step": 8347 + }, + { + "epoch": 0.07246464874436853, + "grad_norm": 0.390625, + "learning_rate": 0.0019735669236576312, + "loss": 0.1289, + "step": 8348 + }, + { + "epoch": 0.0724733292245727, + "grad_norm": 0.6328125, + "learning_rate": 0.001973559763000422, + "loss": 0.1514, + "step": 8349 + }, + { + "epoch": 0.07248200970477686, + "grad_norm": 0.330078125, + "learning_rate": 0.0019735526013879017, + "loss": 0.1689, + "step": 8350 + }, + { + "epoch": 0.07249069018498103, + "grad_norm": 0.232421875, + "learning_rate": 0.001973545438820077, + "loss": 0.0981, + "step": 8351 + }, + { + "epoch": 0.07249937066518519, + "grad_norm": 0.279296875, + "learning_rate": 0.0019735382752969557, + "loss": 0.1465, + "step": 8352 + }, + { + "epoch": 0.07250805114538936, + "grad_norm": 0.68359375, + "learning_rate": 0.0019735311108185468, + "loss": 0.1348, + "step": 8353 + }, + { + "epoch": 0.07251673162559352, + "grad_norm": 0.890625, + "learning_rate": 0.001973523945384857, + "loss": 0.1396, + "step": 8354 + }, + { + "epoch": 0.07252541210579769, + "grad_norm": 0.373046875, + "learning_rate": 0.0019735167789958945, + "loss": 0.1406, + "step": 8355 + }, + { + "epoch": 0.07253409258600185, + "grad_norm": 0.3671875, + "learning_rate": 0.0019735096116516673, + "loss": 0.1553, + "step": 8356 + }, + { + "epoch": 0.07254277306620602, + "grad_norm": 0.37109375, + "learning_rate": 0.0019735024433521833, + "loss": 0.1162, + "step": 8357 + }, + { + "epoch": 0.07255145354641018, + "grad_norm": 0.138671875, + "learning_rate": 0.00197349527409745, + "loss": 0.1201, + "step": 8358 + }, + { + "epoch": 0.07256013402661435, + "grad_norm": 0.625, + "learning_rate": 0.0019734881038874752, + "loss": 0.1113, + "step": 8359 + }, + { + "epoch": 0.07256881450681851, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019734809327222667, + "loss": 0.1089, + "step": 8360 + }, + { + "epoch": 0.07257749498702268, + "grad_norm": 0.4453125, + "learning_rate": 0.001973473760601833, + "loss": 0.1318, + "step": 8361 + }, + { + "epoch": 0.07258617546722684, + "grad_norm": 0.359375, + "learning_rate": 0.001973466587526182, + "loss": 0.1289, + "step": 8362 + }, + { + "epoch": 0.07259485594743101, + "grad_norm": 0.255859375, + "learning_rate": 0.0019734594134953203, + "loss": 0.0884, + "step": 8363 + }, + { + "epoch": 0.07260353642763517, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019734522385092567, + "loss": 0.1338, + "step": 8364 + }, + { + "epoch": 0.07261221690783934, + "grad_norm": 0.1689453125, + "learning_rate": 0.001973445062567999, + "loss": 0.0898, + "step": 8365 + }, + { + "epoch": 0.0726208973880435, + "grad_norm": 0.337890625, + "learning_rate": 0.0019734378856715545, + "loss": 0.0957, + "step": 8366 + }, + { + "epoch": 0.07262957786824767, + "grad_norm": 0.10498046875, + "learning_rate": 0.001973430707819932, + "loss": 0.1914, + "step": 8367 + }, + { + "epoch": 0.07263825834845183, + "grad_norm": 0.419921875, + "learning_rate": 0.001973423529013138, + "loss": 0.1079, + "step": 8368 + }, + { + "epoch": 0.072646938828656, + "grad_norm": 0.265625, + "learning_rate": 0.001973416349251182, + "loss": 0.1172, + "step": 8369 + }, + { + "epoch": 0.07265561930886016, + "grad_norm": 0.216796875, + "learning_rate": 0.0019734091685340706, + "loss": 0.1523, + "step": 8370 + }, + { + "epoch": 0.07266429978906433, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019734019868618124, + "loss": 0.0977, + "step": 8371 + }, + { + "epoch": 0.0726729802692685, + "grad_norm": 0.3046875, + "learning_rate": 0.0019733948042344146, + "loss": 0.1289, + "step": 8372 + }, + { + "epoch": 0.07268166074947266, + "grad_norm": 0.07958984375, + "learning_rate": 0.001973387620651886, + "loss": 0.1338, + "step": 8373 + }, + { + "epoch": 0.07269034122967682, + "grad_norm": 0.19921875, + "learning_rate": 0.001973380436114233, + "loss": 0.1162, + "step": 8374 + }, + { + "epoch": 0.07269902170988099, + "grad_norm": 0.10107421875, + "learning_rate": 0.001973373250621464, + "loss": 0.1377, + "step": 8375 + }, + { + "epoch": 0.07270770219008516, + "grad_norm": 0.21484375, + "learning_rate": 0.001973366064173588, + "loss": 0.1138, + "step": 8376 + }, + { + "epoch": 0.07271638267028932, + "grad_norm": 0.63671875, + "learning_rate": 0.001973358876770612, + "loss": 0.0889, + "step": 8377 + }, + { + "epoch": 0.07272506315049349, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019733516884125435, + "loss": 0.0771, + "step": 8378 + }, + { + "epoch": 0.07273374363069765, + "grad_norm": 0.4453125, + "learning_rate": 0.0019733444990993904, + "loss": 0.1309, + "step": 8379 + }, + { + "epoch": 0.07274242411090182, + "grad_norm": 0.38671875, + "learning_rate": 0.001973337308831161, + "loss": 0.1016, + "step": 8380 + }, + { + "epoch": 0.07275110459110598, + "grad_norm": 0.26171875, + "learning_rate": 0.0019733301176078637, + "loss": 0.1001, + "step": 8381 + }, + { + "epoch": 0.07275978507131015, + "grad_norm": 0.84765625, + "learning_rate": 0.001973322925429505, + "loss": 0.1543, + "step": 8382 + }, + { + "epoch": 0.07276846555151431, + "grad_norm": 0.205078125, + "learning_rate": 0.0019733157322960938, + "loss": 0.1523, + "step": 8383 + }, + { + "epoch": 0.07277714603171848, + "grad_norm": 0.34765625, + "learning_rate": 0.0019733085382076373, + "loss": 0.1377, + "step": 8384 + }, + { + "epoch": 0.07278582651192264, + "grad_norm": 0.6796875, + "learning_rate": 0.001973301343164144, + "loss": 0.1611, + "step": 8385 + }, + { + "epoch": 0.0727945069921268, + "grad_norm": 0.671875, + "learning_rate": 0.0019732941471656217, + "loss": 0.1201, + "step": 8386 + }, + { + "epoch": 0.07280318747233097, + "grad_norm": 1.9609375, + "learning_rate": 0.0019732869502120777, + "loss": 0.127, + "step": 8387 + }, + { + "epoch": 0.07281186795253514, + "grad_norm": 1.0703125, + "learning_rate": 0.0019732797523035204, + "loss": 0.1338, + "step": 8388 + }, + { + "epoch": 0.0728205484327393, + "grad_norm": 0.97265625, + "learning_rate": 0.0019732725534399574, + "loss": 0.4219, + "step": 8389 + }, + { + "epoch": 0.07282922891294347, + "grad_norm": 0.63671875, + "learning_rate": 0.0019732653536213966, + "loss": 0.123, + "step": 8390 + }, + { + "epoch": 0.07283790939314763, + "grad_norm": 0.35546875, + "learning_rate": 0.001973258152847846, + "loss": 0.1309, + "step": 8391 + }, + { + "epoch": 0.0728465898733518, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019732509511193133, + "loss": 0.1631, + "step": 8392 + }, + { + "epoch": 0.07285527035355596, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019732437484358068, + "loss": 0.1387, + "step": 8393 + }, + { + "epoch": 0.07286395083376013, + "grad_norm": 0.30078125, + "learning_rate": 0.0019732365447973333, + "loss": 0.1152, + "step": 8394 + }, + { + "epoch": 0.07287263131396429, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019732293402039023, + "loss": 0.1973, + "step": 8395 + }, + { + "epoch": 0.07288131179416846, + "grad_norm": 0.30859375, + "learning_rate": 0.0019732221346555204, + "loss": 0.1592, + "step": 8396 + }, + { + "epoch": 0.07288999227437262, + "grad_norm": 0.390625, + "learning_rate": 0.0019732149281521958, + "loss": 0.1162, + "step": 8397 + }, + { + "epoch": 0.07289867275457679, + "grad_norm": 0.5546875, + "learning_rate": 0.001973207720693937, + "loss": 0.1221, + "step": 8398 + }, + { + "epoch": 0.07290735323478095, + "grad_norm": 0.1982421875, + "learning_rate": 0.001973200512280751, + "loss": 0.166, + "step": 8399 + }, + { + "epoch": 0.07291603371498512, + "grad_norm": 0.37890625, + "learning_rate": 0.0019731933029126457, + "loss": 0.1396, + "step": 8400 + }, + { + "epoch": 0.07292471419518928, + "grad_norm": 0.392578125, + "learning_rate": 0.00197318609258963, + "loss": 0.0835, + "step": 8401 + }, + { + "epoch": 0.07293339467539345, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019731788813117106, + "loss": 0.166, + "step": 8402 + }, + { + "epoch": 0.07294207515559761, + "grad_norm": 0.14453125, + "learning_rate": 0.001973171669078896, + "loss": 0.1299, + "step": 8403 + }, + { + "epoch": 0.07295075563580178, + "grad_norm": 0.51171875, + "learning_rate": 0.001973164455891194, + "loss": 0.1338, + "step": 8404 + }, + { + "epoch": 0.07295943611600594, + "grad_norm": 0.091796875, + "learning_rate": 0.0019731572417486126, + "loss": 0.1816, + "step": 8405 + }, + { + "epoch": 0.07296811659621011, + "grad_norm": 0.73828125, + "learning_rate": 0.0019731500266511595, + "loss": 0.1641, + "step": 8406 + }, + { + "epoch": 0.07297679707641427, + "grad_norm": 0.28125, + "learning_rate": 0.0019731428105988426, + "loss": 0.1221, + "step": 8407 + }, + { + "epoch": 0.07298547755661844, + "grad_norm": 0.4609375, + "learning_rate": 0.0019731355935916697, + "loss": 0.0947, + "step": 8408 + }, + { + "epoch": 0.0729941580368226, + "grad_norm": 0.1083984375, + "learning_rate": 0.001973128375629649, + "loss": 0.1338, + "step": 8409 + }, + { + "epoch": 0.07300283851702676, + "grad_norm": 0.10107421875, + "learning_rate": 0.001973121156712788, + "loss": 0.1641, + "step": 8410 + }, + { + "epoch": 0.07301151899723092, + "grad_norm": 0.12890625, + "learning_rate": 0.0019731139368410954, + "loss": 0.1123, + "step": 8411 + }, + { + "epoch": 0.07302019947743509, + "grad_norm": 0.08544921875, + "learning_rate": 0.001973106716014578, + "loss": 0.1484, + "step": 8412 + }, + { + "epoch": 0.07302887995763925, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019730994942332445, + "loss": 0.1289, + "step": 8413 + }, + { + "epoch": 0.07303756043784342, + "grad_norm": 0.359375, + "learning_rate": 0.0019730922714971027, + "loss": 0.1426, + "step": 8414 + }, + { + "epoch": 0.07304624091804758, + "grad_norm": 0.73828125, + "learning_rate": 0.0019730850478061595, + "loss": 0.1768, + "step": 8415 + }, + { + "epoch": 0.07305492139825175, + "grad_norm": 0.1865234375, + "learning_rate": 0.001973077823160424, + "loss": 0.166, + "step": 8416 + }, + { + "epoch": 0.07306360187845591, + "grad_norm": 0.734375, + "learning_rate": 0.001973070597559904, + "loss": 0.1533, + "step": 8417 + }, + { + "epoch": 0.07307228235866008, + "grad_norm": 0.8125, + "learning_rate": 0.001973063371004607, + "loss": 0.1089, + "step": 8418 + }, + { + "epoch": 0.07308096283886424, + "grad_norm": 0.1904296875, + "learning_rate": 0.001973056143494541, + "loss": 0.1074, + "step": 8419 + }, + { + "epoch": 0.0730896433190684, + "grad_norm": 0.384765625, + "learning_rate": 0.0019730489150297137, + "loss": 0.1357, + "step": 8420 + }, + { + "epoch": 0.07309832379927257, + "grad_norm": 0.341796875, + "learning_rate": 0.0019730416856101334, + "loss": 0.1357, + "step": 8421 + }, + { + "epoch": 0.07310700427947674, + "grad_norm": 0.1162109375, + "learning_rate": 0.001973034455235808, + "loss": 0.1387, + "step": 8422 + }, + { + "epoch": 0.0731156847596809, + "grad_norm": 0.259765625, + "learning_rate": 0.0019730272239067447, + "loss": 0.1699, + "step": 8423 + }, + { + "epoch": 0.07312436523988507, + "grad_norm": 0.326171875, + "learning_rate": 0.0019730199916229525, + "loss": 0.207, + "step": 8424 + }, + { + "epoch": 0.07313304572008923, + "grad_norm": 0.42578125, + "learning_rate": 0.0019730127583844383, + "loss": 0.1094, + "step": 8425 + }, + { + "epoch": 0.0731417262002934, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019730055241912107, + "loss": 0.1445, + "step": 8426 + }, + { + "epoch": 0.07315040668049756, + "grad_norm": 0.6640625, + "learning_rate": 0.0019729982890432777, + "loss": 0.1504, + "step": 8427 + }, + { + "epoch": 0.07315908716070173, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019729910529406466, + "loss": 0.1387, + "step": 8428 + }, + { + "epoch": 0.07316776764090589, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019729838158833255, + "loss": 0.1855, + "step": 8429 + }, + { + "epoch": 0.07317644812111006, + "grad_norm": 0.228515625, + "learning_rate": 0.0019729765778713224, + "loss": 0.1221, + "step": 8430 + }, + { + "epoch": 0.07318512860131422, + "grad_norm": 0.7890625, + "learning_rate": 0.0019729693389046454, + "loss": 0.1758, + "step": 8431 + }, + { + "epoch": 0.07319380908151839, + "grad_norm": 0.12353515625, + "learning_rate": 0.001972962098983302, + "loss": 0.1289, + "step": 8432 + }, + { + "epoch": 0.07320248956172255, + "grad_norm": 0.26953125, + "learning_rate": 0.0019729548581073008, + "loss": 0.1152, + "step": 8433 + }, + { + "epoch": 0.07321117004192672, + "grad_norm": 0.111328125, + "learning_rate": 0.0019729476162766487, + "loss": 0.1484, + "step": 8434 + }, + { + "epoch": 0.07321985052213088, + "grad_norm": 0.1982421875, + "learning_rate": 0.001972940373491355, + "loss": 0.1465, + "step": 8435 + }, + { + "epoch": 0.07322853100233505, + "grad_norm": 0.640625, + "learning_rate": 0.001972933129751426, + "loss": 0.1465, + "step": 8436 + }, + { + "epoch": 0.07323721148253921, + "grad_norm": 0.6953125, + "learning_rate": 0.0019729258850568706, + "loss": 0.1309, + "step": 8437 + }, + { + "epoch": 0.07324589196274338, + "grad_norm": 0.228515625, + "learning_rate": 0.0019729186394076966, + "loss": 0.1523, + "step": 8438 + }, + { + "epoch": 0.07325457244294754, + "grad_norm": 0.3203125, + "learning_rate": 0.0019729113928039126, + "loss": 0.1309, + "step": 8439 + }, + { + "epoch": 0.07326325292315171, + "grad_norm": 0.4140625, + "learning_rate": 0.0019729041452455254, + "loss": 0.1641, + "step": 8440 + }, + { + "epoch": 0.07327193340335587, + "grad_norm": 0.203125, + "learning_rate": 0.001972896896732543, + "loss": 0.1953, + "step": 8441 + }, + { + "epoch": 0.07328061388356004, + "grad_norm": 1.0390625, + "learning_rate": 0.0019728896472649736, + "loss": 0.2227, + "step": 8442 + }, + { + "epoch": 0.0732892943637642, + "grad_norm": 0.0859375, + "learning_rate": 0.0019728823968428255, + "loss": 0.1201, + "step": 8443 + }, + { + "epoch": 0.07329797484396837, + "grad_norm": 1.46875, + "learning_rate": 0.0019728751454661064, + "loss": 0.1113, + "step": 8444 + }, + { + "epoch": 0.07330665532417253, + "grad_norm": 0.154296875, + "learning_rate": 0.001972867893134824, + "loss": 0.1445, + "step": 8445 + }, + { + "epoch": 0.0733153358043767, + "grad_norm": 0.83984375, + "learning_rate": 0.0019728606398489866, + "loss": 0.1279, + "step": 8446 + }, + { + "epoch": 0.07332401628458086, + "grad_norm": 0.24609375, + "learning_rate": 0.0019728533856086016, + "loss": 0.1035, + "step": 8447 + }, + { + "epoch": 0.07333269676478503, + "grad_norm": 0.85546875, + "learning_rate": 0.0019728461304136776, + "loss": 0.1484, + "step": 8448 + }, + { + "epoch": 0.0733413772449892, + "grad_norm": 0.1796875, + "learning_rate": 0.001972838874264222, + "loss": 0.1152, + "step": 8449 + }, + { + "epoch": 0.07335005772519336, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019728316171602426, + "loss": 0.1328, + "step": 8450 + }, + { + "epoch": 0.07335873820539753, + "grad_norm": 0.314453125, + "learning_rate": 0.001972824359101748, + "loss": 0.1807, + "step": 8451 + }, + { + "epoch": 0.07336741868560169, + "grad_norm": 0.224609375, + "learning_rate": 0.001972817100088746, + "loss": 0.1533, + "step": 8452 + }, + { + "epoch": 0.07337609916580586, + "grad_norm": 0.1708984375, + "learning_rate": 0.001972809840121244, + "loss": 0.1289, + "step": 8453 + }, + { + "epoch": 0.07338477964601002, + "grad_norm": 0.16015625, + "learning_rate": 0.0019728025791992503, + "loss": 0.1582, + "step": 8454 + }, + { + "epoch": 0.07339346012621419, + "grad_norm": 0.2734375, + "learning_rate": 0.001972795317322773, + "loss": 0.1182, + "step": 8455 + }, + { + "epoch": 0.07340214060641835, + "grad_norm": 0.63671875, + "learning_rate": 0.0019727880544918195, + "loss": 0.1475, + "step": 8456 + }, + { + "epoch": 0.07341082108662252, + "grad_norm": 0.30078125, + "learning_rate": 0.0019727807907063987, + "loss": 0.2051, + "step": 8457 + }, + { + "epoch": 0.07341950156682668, + "grad_norm": 0.189453125, + "learning_rate": 0.0019727735259665174, + "loss": 0.1357, + "step": 8458 + }, + { + "epoch": 0.07342818204703085, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019727662602721844, + "loss": 0.1328, + "step": 8459 + }, + { + "epoch": 0.07343686252723501, + "grad_norm": 0.35546875, + "learning_rate": 0.001972758993623407, + "loss": 0.1523, + "step": 8460 + }, + { + "epoch": 0.07344554300743918, + "grad_norm": 0.255859375, + "learning_rate": 0.0019727517260201936, + "loss": 0.0938, + "step": 8461 + }, + { + "epoch": 0.07345422348764334, + "grad_norm": 0.16796875, + "learning_rate": 0.0019727444574625524, + "loss": 0.1426, + "step": 8462 + }, + { + "epoch": 0.0734629039678475, + "grad_norm": 0.546875, + "learning_rate": 0.0019727371879504907, + "loss": 0.1631, + "step": 8463 + }, + { + "epoch": 0.07347158444805167, + "grad_norm": 0.11328125, + "learning_rate": 0.0019727299174840166, + "loss": 0.1064, + "step": 8464 + }, + { + "epoch": 0.07348026492825584, + "grad_norm": 0.328125, + "learning_rate": 0.0019727226460631384, + "loss": 0.0869, + "step": 8465 + }, + { + "epoch": 0.07348894540846, + "grad_norm": 0.15625, + "learning_rate": 0.0019727153736878635, + "loss": 0.1426, + "step": 8466 + }, + { + "epoch": 0.07349762588866417, + "grad_norm": 0.279296875, + "learning_rate": 0.0019727081003582, + "loss": 0.1201, + "step": 8467 + }, + { + "epoch": 0.07350630636886833, + "grad_norm": 0.228515625, + "learning_rate": 0.001972700826074157, + "loss": 0.1562, + "step": 8468 + }, + { + "epoch": 0.0735149868490725, + "grad_norm": 0.244140625, + "learning_rate": 0.001972693550835741, + "loss": 0.1504, + "step": 8469 + }, + { + "epoch": 0.07352366732927666, + "grad_norm": 0.193359375, + "learning_rate": 0.0019726862746429603, + "loss": 0.125, + "step": 8470 + }, + { + "epoch": 0.07353234780948081, + "grad_norm": 1.0546875, + "learning_rate": 0.0019726789974958228, + "loss": 0.1621, + "step": 8471 + }, + { + "epoch": 0.07354102828968498, + "grad_norm": 0.06494140625, + "learning_rate": 0.0019726717193943367, + "loss": 0.1084, + "step": 8472 + }, + { + "epoch": 0.07354970876988914, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019726644403385107, + "loss": 0.1221, + "step": 8473 + }, + { + "epoch": 0.07355838925009331, + "grad_norm": 0.1611328125, + "learning_rate": 0.001972657160328351, + "loss": 0.1201, + "step": 8474 + }, + { + "epoch": 0.07356706973029747, + "grad_norm": 0.314453125, + "learning_rate": 0.001972649879363867, + "loss": 0.2129, + "step": 8475 + }, + { + "epoch": 0.07357575021050164, + "grad_norm": 0.0859375, + "learning_rate": 0.0019726425974450662, + "loss": 0.0977, + "step": 8476 + }, + { + "epoch": 0.0735844306907058, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019726353145719565, + "loss": 0.1387, + "step": 8477 + }, + { + "epoch": 0.07359311117090997, + "grad_norm": 0.37890625, + "learning_rate": 0.001972628030744546, + "loss": 0.1104, + "step": 8478 + }, + { + "epoch": 0.07360179165111413, + "grad_norm": 0.515625, + "learning_rate": 0.0019726207459628423, + "loss": 0.127, + "step": 8479 + }, + { + "epoch": 0.0736104721313183, + "grad_norm": 3.515625, + "learning_rate": 0.0019726134602268543, + "loss": 0.3281, + "step": 8480 + }, + { + "epoch": 0.07361915261152246, + "grad_norm": 0.73828125, + "learning_rate": 0.0019726061735365888, + "loss": 0.1396, + "step": 8481 + }, + { + "epoch": 0.07362783309172663, + "grad_norm": 0.921875, + "learning_rate": 0.0019725988858920545, + "loss": 0.1172, + "step": 8482 + }, + { + "epoch": 0.0736365135719308, + "grad_norm": 0.162109375, + "learning_rate": 0.001972591597293259, + "loss": 0.1289, + "step": 8483 + }, + { + "epoch": 0.07364519405213496, + "grad_norm": 0.42578125, + "learning_rate": 0.0019725843077402104, + "loss": 0.1104, + "step": 8484 + }, + { + "epoch": 0.07365387453233913, + "grad_norm": 0.1494140625, + "learning_rate": 0.001972577017232917, + "loss": 0.1143, + "step": 8485 + }, + { + "epoch": 0.07366255501254329, + "grad_norm": 0.087890625, + "learning_rate": 0.0019725697257713865, + "loss": 0.1191, + "step": 8486 + }, + { + "epoch": 0.07367123549274746, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019725624333556267, + "loss": 0.1279, + "step": 8487 + }, + { + "epoch": 0.07367991597295162, + "grad_norm": 0.5078125, + "learning_rate": 0.0019725551399856454, + "loss": 0.1006, + "step": 8488 + }, + { + "epoch": 0.07368859645315579, + "grad_norm": 0.27734375, + "learning_rate": 0.0019725478456614513, + "loss": 0.1211, + "step": 8489 + }, + { + "epoch": 0.07369727693335995, + "grad_norm": 0.2197265625, + "learning_rate": 0.001972540550383052, + "loss": 0.1084, + "step": 8490 + }, + { + "epoch": 0.07370595741356412, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019725332541504553, + "loss": 0.1436, + "step": 8491 + }, + { + "epoch": 0.07371463789376828, + "grad_norm": 0.291015625, + "learning_rate": 0.0019725259569636695, + "loss": 0.1553, + "step": 8492 + }, + { + "epoch": 0.07372331837397245, + "grad_norm": 0.248046875, + "learning_rate": 0.0019725186588227025, + "loss": 0.1426, + "step": 8493 + }, + { + "epoch": 0.07373199885417661, + "grad_norm": 0.3203125, + "learning_rate": 0.001972511359727562, + "loss": 0.1162, + "step": 8494 + }, + { + "epoch": 0.07374067933438078, + "grad_norm": 0.318359375, + "learning_rate": 0.001972504059678256, + "loss": 0.103, + "step": 8495 + }, + { + "epoch": 0.07374935981458494, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019724967586747934, + "loss": 0.1201, + "step": 8496 + }, + { + "epoch": 0.0737580402947891, + "grad_norm": 0.5703125, + "learning_rate": 0.001972489456717181, + "loss": 0.1279, + "step": 8497 + }, + { + "epoch": 0.07376672077499327, + "grad_norm": 0.16015625, + "learning_rate": 0.001972482153805427, + "loss": 0.1631, + "step": 8498 + }, + { + "epoch": 0.07377540125519744, + "grad_norm": 0.53125, + "learning_rate": 0.00197247484993954, + "loss": 0.1416, + "step": 8499 + }, + { + "epoch": 0.0737840817354016, + "grad_norm": 1.0390625, + "learning_rate": 0.0019724675451195275, + "loss": 0.1123, + "step": 8500 + }, + { + "epoch": 0.07379276221560577, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019724602393453973, + "loss": 0.0903, + "step": 8501 + }, + { + "epoch": 0.07380144269580993, + "grad_norm": 0.53125, + "learning_rate": 0.001972452932617158, + "loss": 0.1855, + "step": 8502 + }, + { + "epoch": 0.0738101231760141, + "grad_norm": 0.3125, + "learning_rate": 0.001972445624934817, + "loss": 0.1455, + "step": 8503 + }, + { + "epoch": 0.07381880365621826, + "grad_norm": 0.279296875, + "learning_rate": 0.0019724383162983833, + "loss": 0.127, + "step": 8504 + }, + { + "epoch": 0.07382748413642243, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019724310067078636, + "loss": 0.1436, + "step": 8505 + }, + { + "epoch": 0.07383616461662659, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019724236961632664, + "loss": 0.1426, + "step": 8506 + }, + { + "epoch": 0.07384484509683076, + "grad_norm": 0.39453125, + "learning_rate": 0.0019724163846646, + "loss": 0.207, + "step": 8507 + }, + { + "epoch": 0.07385352557703492, + "grad_norm": 0.126953125, + "learning_rate": 0.001972409072211872, + "loss": 0.1123, + "step": 8508 + }, + { + "epoch": 0.07386220605723909, + "grad_norm": 0.29296875, + "learning_rate": 0.0019724017588050903, + "loss": 0.1328, + "step": 8509 + }, + { + "epoch": 0.07387088653744325, + "grad_norm": 0.4609375, + "learning_rate": 0.0019723944444442637, + "loss": 0.1309, + "step": 8510 + }, + { + "epoch": 0.07387956701764742, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019723871291293994, + "loss": 0.1543, + "step": 8511 + }, + { + "epoch": 0.07388824749785158, + "grad_norm": 0.12109375, + "learning_rate": 0.0019723798128605057, + "loss": 0.1357, + "step": 8512 + }, + { + "epoch": 0.07389692797805575, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019723724956375904, + "loss": 0.1396, + "step": 8513 + }, + { + "epoch": 0.07390560845825991, + "grad_norm": 0.76171875, + "learning_rate": 0.0019723651774606614, + "loss": 0.1455, + "step": 8514 + }, + { + "epoch": 0.07391428893846408, + "grad_norm": 0.203125, + "learning_rate": 0.0019723578583297273, + "loss": 0.1113, + "step": 8515 + }, + { + "epoch": 0.07392296941866824, + "grad_norm": 0.361328125, + "learning_rate": 0.001972350538244796, + "loss": 0.1309, + "step": 8516 + }, + { + "epoch": 0.07393164989887241, + "grad_norm": 0.162109375, + "learning_rate": 0.001972343217205875, + "loss": 0.1143, + "step": 8517 + }, + { + "epoch": 0.07394033037907657, + "grad_norm": 0.33984375, + "learning_rate": 0.0019723358952129723, + "loss": 0.123, + "step": 8518 + }, + { + "epoch": 0.07394901085928074, + "grad_norm": 0.443359375, + "learning_rate": 0.0019723285722660962, + "loss": 0.1602, + "step": 8519 + }, + { + "epoch": 0.0739576913394849, + "grad_norm": 0.412109375, + "learning_rate": 0.001972321248365255, + "loss": 0.1128, + "step": 8520 + }, + { + "epoch": 0.07396637181968907, + "grad_norm": 0.78515625, + "learning_rate": 0.0019723139235104563, + "loss": 0.1201, + "step": 8521 + }, + { + "epoch": 0.07397505229989323, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019723065977017085, + "loss": 0.1621, + "step": 8522 + }, + { + "epoch": 0.0739837327800974, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019722992709390185, + "loss": 0.1211, + "step": 8523 + }, + { + "epoch": 0.07399241326030156, + "grad_norm": 0.46875, + "learning_rate": 0.0019722919432223957, + "loss": 0.1357, + "step": 8524 + }, + { + "epoch": 0.07400109374050573, + "grad_norm": 1.3046875, + "learning_rate": 0.0019722846145518473, + "loss": 0.1348, + "step": 8525 + }, + { + "epoch": 0.0740097742207099, + "grad_norm": 0.76171875, + "learning_rate": 0.001972277284927382, + "loss": 0.1123, + "step": 8526 + }, + { + "epoch": 0.07401845470091406, + "grad_norm": 0.123046875, + "learning_rate": 0.001972269954349007, + "loss": 0.1621, + "step": 8527 + }, + { + "epoch": 0.07402713518111823, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019722626228167307, + "loss": 0.1152, + "step": 8528 + }, + { + "epoch": 0.07403581566132239, + "grad_norm": 0.46875, + "learning_rate": 0.0019722552903305613, + "loss": 0.1152, + "step": 8529 + }, + { + "epoch": 0.07404449614152656, + "grad_norm": 0.1533203125, + "learning_rate": 0.001972247956890506, + "loss": 0.1523, + "step": 8530 + }, + { + "epoch": 0.07405317662173072, + "grad_norm": 0.2119140625, + "learning_rate": 0.001972240622496574, + "loss": 0.0986, + "step": 8531 + }, + { + "epoch": 0.07406185710193489, + "grad_norm": 0.275390625, + "learning_rate": 0.0019722332871487725, + "loss": 0.1289, + "step": 8532 + }, + { + "epoch": 0.07407053758213904, + "grad_norm": 0.08349609375, + "learning_rate": 0.00197222595084711, + "loss": 0.1089, + "step": 8533 + }, + { + "epoch": 0.0740792180623432, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019722186135915946, + "loss": 0.124, + "step": 8534 + }, + { + "epoch": 0.07408789854254737, + "grad_norm": 0.78515625, + "learning_rate": 0.0019722112753822337, + "loss": 0.1396, + "step": 8535 + }, + { + "epoch": 0.07409657902275153, + "grad_norm": 0.51171875, + "learning_rate": 0.0019722039362190355, + "loss": 0.126, + "step": 8536 + }, + { + "epoch": 0.0741052595029557, + "grad_norm": 1.1796875, + "learning_rate": 0.0019721965961020083, + "loss": 0.1221, + "step": 8537 + }, + { + "epoch": 0.07411393998315986, + "grad_norm": 0.7265625, + "learning_rate": 0.00197218925503116, + "loss": 0.1436, + "step": 8538 + }, + { + "epoch": 0.07412262046336403, + "grad_norm": 0.99609375, + "learning_rate": 0.0019721819130064985, + "loss": 0.1396, + "step": 8539 + }, + { + "epoch": 0.07413130094356819, + "grad_norm": 0.197265625, + "learning_rate": 0.0019721745700280323, + "loss": 0.1055, + "step": 8540 + }, + { + "epoch": 0.07413998142377236, + "grad_norm": 0.451171875, + "learning_rate": 0.0019721672260957688, + "loss": 0.1875, + "step": 8541 + }, + { + "epoch": 0.07414866190397652, + "grad_norm": 0.1630859375, + "learning_rate": 0.001972159881209717, + "loss": 0.1074, + "step": 8542 + }, + { + "epoch": 0.07415734238418069, + "grad_norm": 0.423828125, + "learning_rate": 0.0019721525353698834, + "loss": 0.1436, + "step": 8543 + }, + { + "epoch": 0.07416602286438485, + "grad_norm": 0.92578125, + "learning_rate": 0.001972145188576277, + "loss": 0.1172, + "step": 8544 + }, + { + "epoch": 0.07417470334458902, + "grad_norm": 1.03125, + "learning_rate": 0.0019721378408289057, + "loss": 0.1758, + "step": 8545 + }, + { + "epoch": 0.07418338382479318, + "grad_norm": 0.72265625, + "learning_rate": 0.001972130492127778, + "loss": 0.1387, + "step": 8546 + }, + { + "epoch": 0.07419206430499735, + "grad_norm": 0.5859375, + "learning_rate": 0.001972123142472901, + "loss": 0.1201, + "step": 8547 + }, + { + "epoch": 0.07420074478520151, + "grad_norm": 0.45703125, + "learning_rate": 0.0019721157918642838, + "loss": 0.127, + "step": 8548 + }, + { + "epoch": 0.07420942526540568, + "grad_norm": 0.341796875, + "learning_rate": 0.0019721084403019336, + "loss": 0.1357, + "step": 8549 + }, + { + "epoch": 0.07421810574560984, + "grad_norm": 0.40625, + "learning_rate": 0.001972101087785859, + "loss": 0.1748, + "step": 8550 + }, + { + "epoch": 0.07422678622581401, + "grad_norm": 0.11767578125, + "learning_rate": 0.001972093734316067, + "loss": 0.1885, + "step": 8551 + }, + { + "epoch": 0.07423546670601817, + "grad_norm": 0.5390625, + "learning_rate": 0.0019720863798925667, + "loss": 0.082, + "step": 8552 + }, + { + "epoch": 0.07424414718622234, + "grad_norm": 0.52734375, + "learning_rate": 0.001972079024515366, + "loss": 0.1299, + "step": 8553 + }, + { + "epoch": 0.0742528276664265, + "grad_norm": 0.2421875, + "learning_rate": 0.001972071668184473, + "loss": 0.1279, + "step": 8554 + }, + { + "epoch": 0.07426150814663067, + "grad_norm": 0.361328125, + "learning_rate": 0.001972064310899895, + "loss": 0.0942, + "step": 8555 + }, + { + "epoch": 0.07427018862683483, + "grad_norm": 0.50390625, + "learning_rate": 0.0019720569526616412, + "loss": 0.1309, + "step": 8556 + }, + { + "epoch": 0.074278869107039, + "grad_norm": 0.130859375, + "learning_rate": 0.0019720495934697183, + "loss": 0.127, + "step": 8557 + }, + { + "epoch": 0.07428754958724317, + "grad_norm": 0.2265625, + "learning_rate": 0.001972042233324136, + "loss": 0.127, + "step": 8558 + }, + { + "epoch": 0.07429623006744733, + "grad_norm": 0.61328125, + "learning_rate": 0.0019720348722249005, + "loss": 0.1758, + "step": 8559 + }, + { + "epoch": 0.0743049105476515, + "grad_norm": 0.38671875, + "learning_rate": 0.0019720275101720213, + "loss": 0.1113, + "step": 8560 + }, + { + "epoch": 0.07431359102785566, + "grad_norm": 0.9140625, + "learning_rate": 0.0019720201471655056, + "loss": 0.1553, + "step": 8561 + }, + { + "epoch": 0.07432227150805983, + "grad_norm": 0.27734375, + "learning_rate": 0.001972012783205362, + "loss": 0.1484, + "step": 8562 + }, + { + "epoch": 0.07433095198826399, + "grad_norm": 0.19140625, + "learning_rate": 0.001972005418291598, + "loss": 0.1104, + "step": 8563 + }, + { + "epoch": 0.07433963246846816, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019719980524242225, + "loss": 0.126, + "step": 8564 + }, + { + "epoch": 0.07434831294867232, + "grad_norm": 0.1416015625, + "learning_rate": 0.001971990685603243, + "loss": 0.1543, + "step": 8565 + }, + { + "epoch": 0.07435699342887649, + "grad_norm": 0.275390625, + "learning_rate": 0.001971983317828667, + "loss": 0.1357, + "step": 8566 + }, + { + "epoch": 0.07436567390908065, + "grad_norm": 0.63671875, + "learning_rate": 0.0019719759491005034, + "loss": 0.1045, + "step": 8567 + }, + { + "epoch": 0.07437435438928482, + "grad_norm": 0.228515625, + "learning_rate": 0.00197196857941876, + "loss": 0.1191, + "step": 8568 + }, + { + "epoch": 0.07438303486948898, + "grad_norm": 0.3359375, + "learning_rate": 0.001971961208783445, + "loss": 0.1348, + "step": 8569 + }, + { + "epoch": 0.07439171534969315, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019719538371945665, + "loss": 0.1094, + "step": 8570 + }, + { + "epoch": 0.07440039582989731, + "grad_norm": 2.203125, + "learning_rate": 0.0019719464646521323, + "loss": 0.25, + "step": 8571 + }, + { + "epoch": 0.07440907631010148, + "grad_norm": 0.373046875, + "learning_rate": 0.00197193909115615, + "loss": 0.1533, + "step": 8572 + }, + { + "epoch": 0.07441775679030564, + "grad_norm": 0.095703125, + "learning_rate": 0.001971931716706629, + "loss": 0.0972, + "step": 8573 + }, + { + "epoch": 0.07442643727050981, + "grad_norm": 0.328125, + "learning_rate": 0.001971924341303576, + "loss": 0.1611, + "step": 8574 + }, + { + "epoch": 0.07443511775071397, + "grad_norm": 1.109375, + "learning_rate": 0.001971916964947, + "loss": 0.1133, + "step": 8575 + }, + { + "epoch": 0.07444379823091814, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019719095876369083, + "loss": 0.1543, + "step": 8576 + }, + { + "epoch": 0.0744524787111223, + "grad_norm": 0.111328125, + "learning_rate": 0.0019719022093733097, + "loss": 0.1396, + "step": 8577 + }, + { + "epoch": 0.07446115919132647, + "grad_norm": 0.82421875, + "learning_rate": 0.0019718948301562116, + "loss": 0.1123, + "step": 8578 + }, + { + "epoch": 0.07446983967153063, + "grad_norm": 0.09326171875, + "learning_rate": 0.001971887449985623, + "loss": 0.1475, + "step": 8579 + }, + { + "epoch": 0.0744785201517348, + "grad_norm": 0.2373046875, + "learning_rate": 0.001971880068861551, + "loss": 0.1128, + "step": 8580 + }, + { + "epoch": 0.07448720063193896, + "grad_norm": 0.08837890625, + "learning_rate": 0.001971872686784004, + "loss": 0.168, + "step": 8581 + }, + { + "epoch": 0.07449588111214313, + "grad_norm": 0.40625, + "learning_rate": 0.0019718653037529904, + "loss": 0.1475, + "step": 8582 + }, + { + "epoch": 0.0745045615923473, + "grad_norm": 0.40625, + "learning_rate": 0.0019718579197685175, + "loss": 0.1279, + "step": 8583 + }, + { + "epoch": 0.07451324207255146, + "grad_norm": 0.181640625, + "learning_rate": 0.0019718505348305946, + "loss": 0.1436, + "step": 8584 + }, + { + "epoch": 0.07452192255275562, + "grad_norm": 0.416015625, + "learning_rate": 0.001971843148939228, + "loss": 0.0977, + "step": 8585 + }, + { + "epoch": 0.07453060303295979, + "grad_norm": 0.91796875, + "learning_rate": 0.0019718357620944283, + "loss": 0.127, + "step": 8586 + }, + { + "epoch": 0.07453928351316395, + "grad_norm": 0.279296875, + "learning_rate": 0.001971828374296201, + "loss": 0.124, + "step": 8587 + }, + { + "epoch": 0.07454796399336812, + "grad_norm": 0.62109375, + "learning_rate": 0.0019718209855445557, + "loss": 0.126, + "step": 8588 + }, + { + "epoch": 0.07455664447357228, + "grad_norm": 0.359375, + "learning_rate": 0.0019718135958395, + "loss": 0.1338, + "step": 8589 + }, + { + "epoch": 0.07456532495377645, + "grad_norm": 0.322265625, + "learning_rate": 0.001971806205181042, + "loss": 0.1089, + "step": 8590 + }, + { + "epoch": 0.07457400543398061, + "grad_norm": 0.12451171875, + "learning_rate": 0.00197179881356919, + "loss": 0.127, + "step": 8591 + }, + { + "epoch": 0.07458268591418478, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019717914210039514, + "loss": 0.1221, + "step": 8592 + }, + { + "epoch": 0.07459136639438894, + "grad_norm": 0.34765625, + "learning_rate": 0.0019717840274853355, + "loss": 0.1445, + "step": 8593 + }, + { + "epoch": 0.0746000468745931, + "grad_norm": 0.099609375, + "learning_rate": 0.001971776633013349, + "loss": 0.1152, + "step": 8594 + }, + { + "epoch": 0.07460872735479726, + "grad_norm": 0.458984375, + "learning_rate": 0.0019717692375880013, + "loss": 0.2344, + "step": 8595 + }, + { + "epoch": 0.07461740783500143, + "grad_norm": 0.279296875, + "learning_rate": 0.0019717618412092995, + "loss": 0.0967, + "step": 8596 + }, + { + "epoch": 0.07462608831520559, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019717544438772523, + "loss": 0.1143, + "step": 8597 + }, + { + "epoch": 0.07463476879540976, + "grad_norm": 0.447265625, + "learning_rate": 0.0019717470455918673, + "loss": 0.1318, + "step": 8598 + }, + { + "epoch": 0.07464344927561392, + "grad_norm": 0.2265625, + "learning_rate": 0.001971739646353153, + "loss": 0.1025, + "step": 8599 + }, + { + "epoch": 0.07465212975581809, + "grad_norm": 0.16796875, + "learning_rate": 0.001971732246161117, + "loss": 0.126, + "step": 8600 + }, + { + "epoch": 0.07466081023602225, + "grad_norm": 0.130859375, + "learning_rate": 0.001971724845015768, + "loss": 0.1387, + "step": 8601 + }, + { + "epoch": 0.07466949071622642, + "grad_norm": 0.15234375, + "learning_rate": 0.001971717442917114, + "loss": 0.127, + "step": 8602 + }, + { + "epoch": 0.07467817119643058, + "grad_norm": 0.21875, + "learning_rate": 0.0019717100398651626, + "loss": 0.2207, + "step": 8603 + }, + { + "epoch": 0.07468685167663475, + "grad_norm": 0.07763671875, + "learning_rate": 0.001971702635859922, + "loss": 0.105, + "step": 8604 + }, + { + "epoch": 0.07469553215683891, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019716952309014007, + "loss": 0.1621, + "step": 8605 + }, + { + "epoch": 0.07470421263704308, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019716878249896068, + "loss": 0.1104, + "step": 8606 + }, + { + "epoch": 0.07471289311724724, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019716804181245484, + "loss": 0.1123, + "step": 8607 + }, + { + "epoch": 0.07472157359745141, + "grad_norm": 0.373046875, + "learning_rate": 0.0019716730103062326, + "loss": 0.1025, + "step": 8608 + }, + { + "epoch": 0.07473025407765557, + "grad_norm": 0.06787109375, + "learning_rate": 0.001971665601534669, + "loss": 0.1123, + "step": 8609 + }, + { + "epoch": 0.07473893455785974, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019716581918098643, + "loss": 0.1289, + "step": 8610 + }, + { + "epoch": 0.0747476150380639, + "grad_norm": 0.30078125, + "learning_rate": 0.0019716507811318278, + "loss": 0.1094, + "step": 8611 + }, + { + "epoch": 0.07475629551826807, + "grad_norm": 1.453125, + "learning_rate": 0.001971643369500567, + "loss": 0.3652, + "step": 8612 + }, + { + "epoch": 0.07476497599847223, + "grad_norm": 0.10595703125, + "learning_rate": 0.00197163595691609, + "loss": 0.0991, + "step": 8613 + }, + { + "epoch": 0.0747736564786764, + "grad_norm": 0.10498046875, + "learning_rate": 0.001971628543378405, + "loss": 0.1738, + "step": 8614 + }, + { + "epoch": 0.07478233695888056, + "grad_norm": 0.2578125, + "learning_rate": 0.0019716211288875207, + "loss": 0.1289, + "step": 8615 + }, + { + "epoch": 0.07479101743908473, + "grad_norm": 0.14453125, + "learning_rate": 0.001971613713443444, + "loss": 0.1465, + "step": 8616 + }, + { + "epoch": 0.0747996979192889, + "grad_norm": 0.306640625, + "learning_rate": 0.001971606297046184, + "loss": 0.1006, + "step": 8617 + }, + { + "epoch": 0.07480837839949306, + "grad_norm": 0.455078125, + "learning_rate": 0.001971598879695748, + "loss": 0.1299, + "step": 8618 + }, + { + "epoch": 0.07481705887969722, + "grad_norm": 0.57421875, + "learning_rate": 0.001971591461392145, + "loss": 0.1396, + "step": 8619 + }, + { + "epoch": 0.07482573935990139, + "grad_norm": 0.46875, + "learning_rate": 0.0019715840421353826, + "loss": 0.1157, + "step": 8620 + }, + { + "epoch": 0.07483441984010555, + "grad_norm": 0.42578125, + "learning_rate": 0.0019715766219254686, + "loss": 0.168, + "step": 8621 + }, + { + "epoch": 0.07484310032030972, + "grad_norm": 0.53125, + "learning_rate": 0.001971569200762412, + "loss": 0.1177, + "step": 8622 + }, + { + "epoch": 0.07485178080051388, + "grad_norm": 0.1767578125, + "learning_rate": 0.00197156177864622, + "loss": 0.124, + "step": 8623 + }, + { + "epoch": 0.07486046128071805, + "grad_norm": 0.140625, + "learning_rate": 0.0019715543555769016, + "loss": 0.1406, + "step": 8624 + }, + { + "epoch": 0.07486914176092221, + "grad_norm": 0.6328125, + "learning_rate": 0.001971546931554464, + "loss": 0.1084, + "step": 8625 + }, + { + "epoch": 0.07487782224112638, + "grad_norm": 0.2412109375, + "learning_rate": 0.001971539506578916, + "loss": 0.1426, + "step": 8626 + }, + { + "epoch": 0.07488650272133054, + "grad_norm": 0.078125, + "learning_rate": 0.0019715320806502653, + "loss": 0.125, + "step": 8627 + }, + { + "epoch": 0.07489518320153471, + "grad_norm": 0.30859375, + "learning_rate": 0.0019715246537685208, + "loss": 0.1348, + "step": 8628 + }, + { + "epoch": 0.07490386368173887, + "grad_norm": 0.9375, + "learning_rate": 0.0019715172259336896, + "loss": 0.1309, + "step": 8629 + }, + { + "epoch": 0.07491254416194304, + "grad_norm": 0.1650390625, + "learning_rate": 0.00197150979714578, + "loss": 0.0771, + "step": 8630 + }, + { + "epoch": 0.0749212246421472, + "grad_norm": 0.1640625, + "learning_rate": 0.0019715023674048006, + "loss": 0.125, + "step": 8631 + }, + { + "epoch": 0.07492990512235137, + "grad_norm": 0.236328125, + "learning_rate": 0.001971494936710759, + "loss": 0.124, + "step": 8632 + }, + { + "epoch": 0.07493858560255554, + "grad_norm": 0.181640625, + "learning_rate": 0.0019714875050636646, + "loss": 0.1367, + "step": 8633 + }, + { + "epoch": 0.0749472660827597, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019714800724635237, + "loss": 0.1699, + "step": 8634 + }, + { + "epoch": 0.07495594656296387, + "grad_norm": 0.703125, + "learning_rate": 0.001971472638910346, + "loss": 0.1021, + "step": 8635 + }, + { + "epoch": 0.07496462704316803, + "grad_norm": 0.283203125, + "learning_rate": 0.001971465204404138, + "loss": 0.1836, + "step": 8636 + }, + { + "epoch": 0.0749733075233722, + "grad_norm": 0.71484375, + "learning_rate": 0.0019714577689449093, + "loss": 0.1006, + "step": 8637 + }, + { + "epoch": 0.07498198800357636, + "grad_norm": 1.25, + "learning_rate": 0.0019714503325326672, + "loss": 0.1426, + "step": 8638 + }, + { + "epoch": 0.07499066848378053, + "grad_norm": 1.03125, + "learning_rate": 0.00197144289516742, + "loss": 0.1133, + "step": 8639 + }, + { + "epoch": 0.07499934896398469, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019714354568491762, + "loss": 0.1328, + "step": 8640 + }, + { + "epoch": 0.07500802944418886, + "grad_norm": 0.130859375, + "learning_rate": 0.0019714280175779438, + "loss": 0.0991, + "step": 8641 + }, + { + "epoch": 0.07501670992439302, + "grad_norm": 0.37890625, + "learning_rate": 0.001971420577353731, + "loss": 0.0889, + "step": 8642 + }, + { + "epoch": 0.07502539040459719, + "grad_norm": 0.251953125, + "learning_rate": 0.0019714131361765453, + "loss": 0.1543, + "step": 8643 + }, + { + "epoch": 0.07503407088480135, + "grad_norm": 0.486328125, + "learning_rate": 0.0019714056940463953, + "loss": 0.1514, + "step": 8644 + }, + { + "epoch": 0.07504275136500552, + "grad_norm": 0.6953125, + "learning_rate": 0.0019713982509632897, + "loss": 0.167, + "step": 8645 + }, + { + "epoch": 0.07505143184520968, + "grad_norm": 0.828125, + "learning_rate": 0.0019713908069272355, + "loss": 0.106, + "step": 8646 + }, + { + "epoch": 0.07506011232541385, + "grad_norm": 0.796875, + "learning_rate": 0.0019713833619382413, + "loss": 0.1543, + "step": 8647 + }, + { + "epoch": 0.07506879280561801, + "grad_norm": 0.53515625, + "learning_rate": 0.0019713759159963157, + "loss": 0.1191, + "step": 8648 + }, + { + "epoch": 0.07507747328582218, + "grad_norm": 0.52734375, + "learning_rate": 0.0019713684691014663, + "loss": 0.127, + "step": 8649 + }, + { + "epoch": 0.07508615376602634, + "grad_norm": 0.326171875, + "learning_rate": 0.0019713610212537015, + "loss": 0.1592, + "step": 8650 + }, + { + "epoch": 0.07509483424623051, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019713535724530293, + "loss": 0.0952, + "step": 8651 + }, + { + "epoch": 0.07510351472643467, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019713461226994584, + "loss": 0.127, + "step": 8652 + }, + { + "epoch": 0.07511219520663884, + "grad_norm": 0.087890625, + "learning_rate": 0.001971338671992996, + "loss": 0.1387, + "step": 8653 + }, + { + "epoch": 0.075120875686843, + "grad_norm": 0.53515625, + "learning_rate": 0.001971331220333651, + "loss": 0.1211, + "step": 8654 + }, + { + "epoch": 0.07512955616704717, + "grad_norm": 0.18359375, + "learning_rate": 0.001971323767721431, + "loss": 0.1172, + "step": 8655 + }, + { + "epoch": 0.07513823664725132, + "grad_norm": 0.2734375, + "learning_rate": 0.001971316314156344, + "loss": 0.1602, + "step": 8656 + }, + { + "epoch": 0.07514691712745548, + "grad_norm": 0.1328125, + "learning_rate": 0.0019713088596383993, + "loss": 0.1289, + "step": 8657 + }, + { + "epoch": 0.07515559760765965, + "grad_norm": 0.12109375, + "learning_rate": 0.001971301404167604, + "loss": 0.1094, + "step": 8658 + }, + { + "epoch": 0.07516427808786381, + "grad_norm": 0.130859375, + "learning_rate": 0.0019712939477439665, + "loss": 0.0874, + "step": 8659 + }, + { + "epoch": 0.07517295856806798, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019712864903674954, + "loss": 0.1143, + "step": 8660 + }, + { + "epoch": 0.07518163904827214, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019712790320381983, + "loss": 0.1172, + "step": 8661 + }, + { + "epoch": 0.07519031952847631, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019712715727560836, + "loss": 0.1523, + "step": 8662 + }, + { + "epoch": 0.07519900000868047, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019712641125211594, + "loss": 0.0962, + "step": 8663 + }, + { + "epoch": 0.07520768048888464, + "grad_norm": 0.73046875, + "learning_rate": 0.0019712566513334336, + "loss": 0.1523, + "step": 8664 + }, + { + "epoch": 0.0752163609690888, + "grad_norm": 0.77734375, + "learning_rate": 0.001971249189192915, + "loss": 0.1289, + "step": 8665 + }, + { + "epoch": 0.07522504144929297, + "grad_norm": 0.318359375, + "learning_rate": 0.001971241726099611, + "loss": 0.1865, + "step": 8666 + }, + { + "epoch": 0.07523372192949714, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019712342620535304, + "loss": 0.1226, + "step": 8667 + }, + { + "epoch": 0.0752424024097013, + "grad_norm": 0.61328125, + "learning_rate": 0.001971226797054681, + "loss": 0.1084, + "step": 8668 + }, + { + "epoch": 0.07525108288990547, + "grad_norm": 0.06396484375, + "learning_rate": 0.001971219331103071, + "loss": 0.1021, + "step": 8669 + }, + { + "epoch": 0.07525976337010963, + "grad_norm": 0.859375, + "learning_rate": 0.0019712118641987087, + "loss": 0.1348, + "step": 8670 + }, + { + "epoch": 0.0752684438503138, + "grad_norm": 0.275390625, + "learning_rate": 0.001971204396341602, + "loss": 0.124, + "step": 8671 + }, + { + "epoch": 0.07527712433051796, + "grad_norm": 0.13671875, + "learning_rate": 0.0019711969275317594, + "loss": 0.1396, + "step": 8672 + }, + { + "epoch": 0.07528580481072213, + "grad_norm": 0.26953125, + "learning_rate": 0.0019711894577691892, + "loss": 0.1289, + "step": 8673 + }, + { + "epoch": 0.07529448529092629, + "grad_norm": 0.1669921875, + "learning_rate": 0.001971181987053899, + "loss": 0.127, + "step": 8674 + }, + { + "epoch": 0.07530316577113046, + "grad_norm": 0.2734375, + "learning_rate": 0.001971174515385897, + "loss": 0.1455, + "step": 8675 + }, + { + "epoch": 0.07531184625133462, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019711670427651922, + "loss": 0.0845, + "step": 8676 + }, + { + "epoch": 0.07532052673153879, + "grad_norm": 0.099609375, + "learning_rate": 0.0019711595691917923, + "loss": 0.1279, + "step": 8677 + }, + { + "epoch": 0.07532920721174295, + "grad_norm": 0.50390625, + "learning_rate": 0.001971152094665705, + "loss": 0.166, + "step": 8678 + }, + { + "epoch": 0.07533788769194712, + "grad_norm": 0.52734375, + "learning_rate": 0.001971144619186939, + "loss": 0.1387, + "step": 8679 + }, + { + "epoch": 0.07534656817215128, + "grad_norm": 0.85546875, + "learning_rate": 0.0019711371427555023, + "loss": 0.1064, + "step": 8680 + }, + { + "epoch": 0.07535524865235545, + "grad_norm": 0.349609375, + "learning_rate": 0.0019711296653714032, + "loss": 0.1611, + "step": 8681 + }, + { + "epoch": 0.07536392913255961, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019711221870346496, + "loss": 0.1152, + "step": 8682 + }, + { + "epoch": 0.07537260961276378, + "grad_norm": 0.26953125, + "learning_rate": 0.00197111470774525, + "loss": 0.1289, + "step": 8683 + }, + { + "epoch": 0.07538129009296794, + "grad_norm": 0.326171875, + "learning_rate": 0.001971107227503212, + "loss": 0.1201, + "step": 8684 + }, + { + "epoch": 0.07538997057317211, + "grad_norm": 1.3203125, + "learning_rate": 0.0019710997463085445, + "loss": 0.123, + "step": 8685 + }, + { + "epoch": 0.07539865105337627, + "grad_norm": 0.515625, + "learning_rate": 0.001971092264161256, + "loss": 0.1006, + "step": 8686 + }, + { + "epoch": 0.07540733153358044, + "grad_norm": 0.37890625, + "learning_rate": 0.0019710847810613533, + "loss": 0.1187, + "step": 8687 + }, + { + "epoch": 0.0754160120137846, + "grad_norm": 0.51953125, + "learning_rate": 0.001971077297008846, + "loss": 0.1079, + "step": 8688 + }, + { + "epoch": 0.07542469249398877, + "grad_norm": 0.458984375, + "learning_rate": 0.001971069812003741, + "loss": 0.1406, + "step": 8689 + }, + { + "epoch": 0.07543337297419293, + "grad_norm": 0.2041015625, + "learning_rate": 0.001971062326046048, + "loss": 0.1289, + "step": 8690 + }, + { + "epoch": 0.0754420534543971, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019710548391357733, + "loss": 0.1191, + "step": 8691 + }, + { + "epoch": 0.07545073393460126, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019710473512729266, + "loss": 0.1016, + "step": 8692 + }, + { + "epoch": 0.07545941441480543, + "grad_norm": 1.21875, + "learning_rate": 0.0019710398624575155, + "loss": 0.1543, + "step": 8693 + }, + { + "epoch": 0.0754680948950096, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019710323726895487, + "loss": 0.1084, + "step": 8694 + }, + { + "epoch": 0.07547677537521376, + "grad_norm": 0.287109375, + "learning_rate": 0.001971024881969034, + "loss": 0.1455, + "step": 8695 + }, + { + "epoch": 0.07548545585541792, + "grad_norm": 0.4609375, + "learning_rate": 0.001971017390295979, + "loss": 0.1523, + "step": 8696 + }, + { + "epoch": 0.07549413633562209, + "grad_norm": 0.78515625, + "learning_rate": 0.0019710098976703926, + "loss": 0.1699, + "step": 8697 + }, + { + "epoch": 0.07550281681582625, + "grad_norm": 0.1318359375, + "learning_rate": 0.001971002404092283, + "loss": 0.1523, + "step": 8698 + }, + { + "epoch": 0.07551149729603042, + "grad_norm": 0.96484375, + "learning_rate": 0.001970994909561658, + "loss": 0.1367, + "step": 8699 + }, + { + "epoch": 0.07552017777623458, + "grad_norm": 0.6640625, + "learning_rate": 0.0019709874140785267, + "loss": 0.1289, + "step": 8700 + }, + { + "epoch": 0.07552885825643875, + "grad_norm": 0.75390625, + "learning_rate": 0.001970979917642896, + "loss": 0.168, + "step": 8701 + }, + { + "epoch": 0.07553753873664291, + "grad_norm": 0.0810546875, + "learning_rate": 0.001970972420254775, + "loss": 0.1211, + "step": 8702 + }, + { + "epoch": 0.07554621921684708, + "grad_norm": 0.57421875, + "learning_rate": 0.0019709649219141717, + "loss": 0.1133, + "step": 8703 + }, + { + "epoch": 0.07555489969705124, + "grad_norm": 0.439453125, + "learning_rate": 0.0019709574226210945, + "loss": 0.1582, + "step": 8704 + }, + { + "epoch": 0.07556358017725541, + "grad_norm": 1.1875, + "learning_rate": 0.001970949922375551, + "loss": 0.1328, + "step": 8705 + }, + { + "epoch": 0.07557226065745958, + "grad_norm": 0.1298828125, + "learning_rate": 0.00197094242117755, + "loss": 0.1211, + "step": 8706 + }, + { + "epoch": 0.07558094113766374, + "grad_norm": 0.30859375, + "learning_rate": 0.001970934919027099, + "loss": 0.0913, + "step": 8707 + }, + { + "epoch": 0.0755896216178679, + "grad_norm": 0.330078125, + "learning_rate": 0.001970927415924207, + "loss": 0.1143, + "step": 8708 + }, + { + "epoch": 0.07559830209807207, + "grad_norm": 0.71484375, + "learning_rate": 0.001970919911868882, + "loss": 0.123, + "step": 8709 + }, + { + "epoch": 0.07560698257827624, + "grad_norm": 0.8671875, + "learning_rate": 0.0019709124068611316, + "loss": 0.1338, + "step": 8710 + }, + { + "epoch": 0.0756156630584804, + "grad_norm": 0.146484375, + "learning_rate": 0.001970904900900965, + "loss": 0.1147, + "step": 8711 + }, + { + "epoch": 0.07562434353868457, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019708973939883898, + "loss": 0.1182, + "step": 8712 + }, + { + "epoch": 0.07563302401888873, + "grad_norm": 0.330078125, + "learning_rate": 0.0019708898861234138, + "loss": 0.1289, + "step": 8713 + }, + { + "epoch": 0.0756417044990929, + "grad_norm": 0.23046875, + "learning_rate": 0.001970882377306046, + "loss": 0.168, + "step": 8714 + }, + { + "epoch": 0.07565038497929706, + "grad_norm": 0.451171875, + "learning_rate": 0.001970874867536294, + "loss": 0.1797, + "step": 8715 + }, + { + "epoch": 0.07565906545950123, + "grad_norm": 0.61328125, + "learning_rate": 0.001970867356814167, + "loss": 0.1221, + "step": 8716 + }, + { + "epoch": 0.07566774593970539, + "grad_norm": 0.98046875, + "learning_rate": 0.001970859845139672, + "loss": 0.1533, + "step": 8717 + }, + { + "epoch": 0.07567642641990954, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019708523325128184, + "loss": 0.1309, + "step": 8718 + }, + { + "epoch": 0.07568510690011371, + "grad_norm": 0.0712890625, + "learning_rate": 0.001970844818933613, + "loss": 0.1387, + "step": 8719 + }, + { + "epoch": 0.07569378738031787, + "grad_norm": 0.84765625, + "learning_rate": 0.001970837304402065, + "loss": 0.1465, + "step": 8720 + }, + { + "epoch": 0.07570246786052204, + "grad_norm": 0.75, + "learning_rate": 0.001970829788918183, + "loss": 0.127, + "step": 8721 + }, + { + "epoch": 0.0757111483407262, + "grad_norm": 0.1826171875, + "learning_rate": 0.001970822272481974, + "loss": 0.1025, + "step": 8722 + }, + { + "epoch": 0.07571982882093037, + "grad_norm": 0.09765625, + "learning_rate": 0.001970814755093447, + "loss": 0.1299, + "step": 8723 + }, + { + "epoch": 0.07572850930113453, + "grad_norm": 0.345703125, + "learning_rate": 0.00197080723675261, + "loss": 0.1152, + "step": 8724 + }, + { + "epoch": 0.0757371897813387, + "grad_norm": 0.46875, + "learning_rate": 0.0019707997174594712, + "loss": 0.127, + "step": 8725 + }, + { + "epoch": 0.07574587026154286, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019707921972140392, + "loss": 0.1416, + "step": 8726 + }, + { + "epoch": 0.07575455074174703, + "grad_norm": 0.19921875, + "learning_rate": 0.0019707846760163216, + "loss": 0.1348, + "step": 8727 + }, + { + "epoch": 0.0757632312219512, + "grad_norm": 0.6015625, + "learning_rate": 0.001970777153866327, + "loss": 0.1465, + "step": 8728 + }, + { + "epoch": 0.07577191170215536, + "grad_norm": 0.419921875, + "learning_rate": 0.001970769630764064, + "loss": 0.1338, + "step": 8729 + }, + { + "epoch": 0.07578059218235952, + "grad_norm": 0.384765625, + "learning_rate": 0.00197076210670954, + "loss": 0.1006, + "step": 8730 + }, + { + "epoch": 0.07578927266256369, + "grad_norm": 1.140625, + "learning_rate": 0.0019707545817027644, + "loss": 0.1162, + "step": 8731 + }, + { + "epoch": 0.07579795314276785, + "grad_norm": 0.33984375, + "learning_rate": 0.001970747055743744, + "loss": 0.0762, + "step": 8732 + }, + { + "epoch": 0.07580663362297202, + "grad_norm": 0.1796875, + "learning_rate": 0.0019707395288324878, + "loss": 0.1309, + "step": 8733 + }, + { + "epoch": 0.07581531410317618, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019707320009690037, + "loss": 0.1299, + "step": 8734 + }, + { + "epoch": 0.07582399458338035, + "grad_norm": 0.251953125, + "learning_rate": 0.0019707244721533005, + "loss": 0.1221, + "step": 8735 + }, + { + "epoch": 0.07583267506358451, + "grad_norm": 0.16015625, + "learning_rate": 0.0019707169423853863, + "loss": 0.0977, + "step": 8736 + }, + { + "epoch": 0.07584135554378868, + "grad_norm": 0.4921875, + "learning_rate": 0.0019707094116652685, + "loss": 0.1367, + "step": 8737 + }, + { + "epoch": 0.07585003602399285, + "grad_norm": 0.173828125, + "learning_rate": 0.001970701879992956, + "loss": 0.1455, + "step": 8738 + }, + { + "epoch": 0.07585871650419701, + "grad_norm": 0.09326171875, + "learning_rate": 0.001970694347368458, + "loss": 0.1504, + "step": 8739 + }, + { + "epoch": 0.07586739698440118, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019706868137917806, + "loss": 0.1113, + "step": 8740 + }, + { + "epoch": 0.07587607746460534, + "grad_norm": 0.3359375, + "learning_rate": 0.001970679279262934, + "loss": 0.1377, + "step": 8741 + }, + { + "epoch": 0.0758847579448095, + "grad_norm": 0.47265625, + "learning_rate": 0.001970671743781925, + "loss": 0.1201, + "step": 8742 + }, + { + "epoch": 0.07589343842501367, + "grad_norm": 0.72265625, + "learning_rate": 0.0019706642073487627, + "loss": 0.124, + "step": 8743 + }, + { + "epoch": 0.07590211890521784, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019706566699634553, + "loss": 0.1152, + "step": 8744 + }, + { + "epoch": 0.075910799385422, + "grad_norm": 0.33203125, + "learning_rate": 0.0019706491316260105, + "loss": 0.0908, + "step": 8745 + }, + { + "epoch": 0.07591947986562617, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019706415923364373, + "loss": 0.1113, + "step": 8746 + }, + { + "epoch": 0.07592816034583033, + "grad_norm": 0.8125, + "learning_rate": 0.001970634052094743, + "loss": 0.1211, + "step": 8747 + }, + { + "epoch": 0.0759368408260345, + "grad_norm": 0.0888671875, + "learning_rate": 0.001970626510900937, + "loss": 0.1035, + "step": 8748 + }, + { + "epoch": 0.07594552130623866, + "grad_norm": 0.185546875, + "learning_rate": 0.0019706189687550267, + "loss": 0.1035, + "step": 8749 + }, + { + "epoch": 0.07595420178644283, + "grad_norm": 0.1943359375, + "learning_rate": 0.00197061142565702, + "loss": 0.1562, + "step": 8750 + }, + { + "epoch": 0.07596288226664699, + "grad_norm": 0.17578125, + "learning_rate": 0.0019706038816069264, + "loss": 0.167, + "step": 8751 + }, + { + "epoch": 0.07597156274685116, + "grad_norm": 0.310546875, + "learning_rate": 0.0019705963366047534, + "loss": 0.1191, + "step": 8752 + }, + { + "epoch": 0.07598024322705532, + "grad_norm": 0.107421875, + "learning_rate": 0.001970588790650509, + "loss": 0.1445, + "step": 8753 + }, + { + "epoch": 0.07598892370725949, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019705812437442022, + "loss": 0.0957, + "step": 8754 + }, + { + "epoch": 0.07599760418746365, + "grad_norm": 0.361328125, + "learning_rate": 0.0019705736958858405, + "loss": 0.1055, + "step": 8755 + }, + { + "epoch": 0.07600628466766782, + "grad_norm": 0.212890625, + "learning_rate": 0.001970566147075433, + "loss": 0.1572, + "step": 8756 + }, + { + "epoch": 0.07601496514787198, + "grad_norm": 0.796875, + "learning_rate": 0.001970558597312987, + "loss": 0.1025, + "step": 8757 + }, + { + "epoch": 0.07602364562807615, + "grad_norm": 0.357421875, + "learning_rate": 0.001970551046598511, + "loss": 0.1494, + "step": 8758 + }, + { + "epoch": 0.07603232610828031, + "grad_norm": 0.7890625, + "learning_rate": 0.0019705434949320137, + "loss": 0.1387, + "step": 8759 + }, + { + "epoch": 0.07604100658848448, + "grad_norm": 0.388671875, + "learning_rate": 0.001970535942313503, + "loss": 0.1162, + "step": 8760 + }, + { + "epoch": 0.07604968706868864, + "grad_norm": 0.345703125, + "learning_rate": 0.001970528388742988, + "loss": 0.1445, + "step": 8761 + }, + { + "epoch": 0.07605836754889281, + "grad_norm": 0.412109375, + "learning_rate": 0.001970520834220475, + "loss": 0.168, + "step": 8762 + }, + { + "epoch": 0.07606704802909697, + "grad_norm": 0.384765625, + "learning_rate": 0.0019705132787459743, + "loss": 0.1055, + "step": 8763 + }, + { + "epoch": 0.07607572850930114, + "grad_norm": 0.38671875, + "learning_rate": 0.0019705057223194933, + "loss": 0.1357, + "step": 8764 + }, + { + "epoch": 0.0760844089895053, + "grad_norm": 0.328125, + "learning_rate": 0.00197049816494104, + "loss": 0.1182, + "step": 8765 + }, + { + "epoch": 0.07609308946970947, + "grad_norm": 0.5546875, + "learning_rate": 0.0019704906066106234, + "loss": 0.1094, + "step": 8766 + }, + { + "epoch": 0.07610176994991363, + "grad_norm": 0.29296875, + "learning_rate": 0.001970483047328251, + "loss": 0.1514, + "step": 8767 + }, + { + "epoch": 0.0761104504301178, + "grad_norm": 0.185546875, + "learning_rate": 0.0019704754870939318, + "loss": 0.1172, + "step": 8768 + }, + { + "epoch": 0.07611913091032196, + "grad_norm": 0.482421875, + "learning_rate": 0.0019704679259076735, + "loss": 0.1113, + "step": 8769 + }, + { + "epoch": 0.07612781139052613, + "grad_norm": 0.166015625, + "learning_rate": 0.0019704603637694844, + "loss": 0.1445, + "step": 8770 + }, + { + "epoch": 0.0761364918707303, + "grad_norm": 0.462890625, + "learning_rate": 0.001970452800679373, + "loss": 0.1797, + "step": 8771 + }, + { + "epoch": 0.07614517235093446, + "grad_norm": 0.6484375, + "learning_rate": 0.0019704452366373477, + "loss": 0.1094, + "step": 8772 + }, + { + "epoch": 0.07615385283113862, + "grad_norm": 0.40625, + "learning_rate": 0.0019704376716434165, + "loss": 0.1406, + "step": 8773 + }, + { + "epoch": 0.07616253331134279, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019704301056975874, + "loss": 0.126, + "step": 8774 + }, + { + "epoch": 0.07617121379154695, + "grad_norm": 0.474609375, + "learning_rate": 0.001970422538799869, + "loss": 0.1201, + "step": 8775 + }, + { + "epoch": 0.07617989427175112, + "grad_norm": 0.48828125, + "learning_rate": 0.00197041497095027, + "loss": 0.1289, + "step": 8776 + }, + { + "epoch": 0.07618857475195528, + "grad_norm": 0.302734375, + "learning_rate": 0.001970407402148798, + "loss": 0.1201, + "step": 8777 + }, + { + "epoch": 0.07619725523215945, + "grad_norm": 0.3671875, + "learning_rate": 0.001970399832395462, + "loss": 0.1211, + "step": 8778 + }, + { + "epoch": 0.0762059357123636, + "grad_norm": 0.50390625, + "learning_rate": 0.0019703922616902693, + "loss": 0.127, + "step": 8779 + }, + { + "epoch": 0.07621461619256777, + "grad_norm": 1.109375, + "learning_rate": 0.001970384690033229, + "loss": 0.1592, + "step": 8780 + }, + { + "epoch": 0.07622329667277193, + "grad_norm": 0.1484375, + "learning_rate": 0.0019703771174243486, + "loss": 0.1367, + "step": 8781 + }, + { + "epoch": 0.0762319771529761, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019703695438636377, + "loss": 0.125, + "step": 8782 + }, + { + "epoch": 0.07624065763318026, + "grad_norm": 0.322265625, + "learning_rate": 0.0019703619693511028, + "loss": 0.1021, + "step": 8783 + }, + { + "epoch": 0.07624933811338443, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019703543938867536, + "loss": 0.1758, + "step": 8784 + }, + { + "epoch": 0.07625801859358859, + "grad_norm": 0.421875, + "learning_rate": 0.001970346817470598, + "loss": 0.1133, + "step": 8785 + }, + { + "epoch": 0.07626669907379276, + "grad_norm": 0.2119140625, + "learning_rate": 0.001970339240102644, + "loss": 0.0947, + "step": 8786 + }, + { + "epoch": 0.07627537955399692, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019703316617829003, + "loss": 0.1318, + "step": 8787 + }, + { + "epoch": 0.07628406003420109, + "grad_norm": 0.1328125, + "learning_rate": 0.0019703240825113748, + "loss": 0.1089, + "step": 8788 + }, + { + "epoch": 0.07629274051440525, + "grad_norm": 0.349609375, + "learning_rate": 0.0019703165022880763, + "loss": 0.1289, + "step": 8789 + }, + { + "epoch": 0.07630142099460942, + "grad_norm": 1.890625, + "learning_rate": 0.0019703089211130123, + "loss": 0.1641, + "step": 8790 + }, + { + "epoch": 0.07631010147481358, + "grad_norm": 0.328125, + "learning_rate": 0.001970301338986192, + "loss": 0.1328, + "step": 8791 + }, + { + "epoch": 0.07631878195501775, + "grad_norm": 0.61328125, + "learning_rate": 0.0019702937559076227, + "loss": 0.1426, + "step": 8792 + }, + { + "epoch": 0.07632746243522191, + "grad_norm": 1.125, + "learning_rate": 0.0019702861718773135, + "loss": 0.1182, + "step": 8793 + }, + { + "epoch": 0.07633614291542608, + "grad_norm": 0.53125, + "learning_rate": 0.001970278586895272, + "loss": 0.1641, + "step": 8794 + }, + { + "epoch": 0.07634482339563024, + "grad_norm": 0.53515625, + "learning_rate": 0.0019702710009615074, + "loss": 0.1162, + "step": 8795 + }, + { + "epoch": 0.07635350387583441, + "grad_norm": 0.453125, + "learning_rate": 0.0019702634140760272, + "loss": 0.1426, + "step": 8796 + }, + { + "epoch": 0.07636218435603857, + "grad_norm": 0.287109375, + "learning_rate": 0.0019702558262388405, + "loss": 0.1484, + "step": 8797 + }, + { + "epoch": 0.07637086483624274, + "grad_norm": 0.42578125, + "learning_rate": 0.0019702482374499546, + "loss": 0.127, + "step": 8798 + }, + { + "epoch": 0.0763795453164469, + "grad_norm": 0.390625, + "learning_rate": 0.0019702406477093786, + "loss": 0.1426, + "step": 8799 + }, + { + "epoch": 0.07638822579665107, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019702330570171202, + "loss": 0.1289, + "step": 8800 + }, + { + "epoch": 0.07639690627685523, + "grad_norm": 0.1640625, + "learning_rate": 0.0019702254653731883, + "loss": 0.104, + "step": 8801 + }, + { + "epoch": 0.0764055867570594, + "grad_norm": 1.7265625, + "learning_rate": 0.001970217872777591, + "loss": 0.1582, + "step": 8802 + }, + { + "epoch": 0.07641426723726356, + "grad_norm": 1.125, + "learning_rate": 0.0019702102792303364, + "loss": 0.168, + "step": 8803 + }, + { + "epoch": 0.07642294771746773, + "grad_norm": 0.33203125, + "learning_rate": 0.001970202684731433, + "loss": 0.1152, + "step": 8804 + }, + { + "epoch": 0.0764316281976719, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019701950892808883, + "loss": 0.123, + "step": 8805 + }, + { + "epoch": 0.07644030867787606, + "grad_norm": 0.392578125, + "learning_rate": 0.001970187492878712, + "loss": 0.127, + "step": 8806 + }, + { + "epoch": 0.07644898915808022, + "grad_norm": 0.365234375, + "learning_rate": 0.001970179895524912, + "loss": 0.1836, + "step": 8807 + }, + { + "epoch": 0.07645766963828439, + "grad_norm": 0.66796875, + "learning_rate": 0.001970172297219496, + "loss": 0.168, + "step": 8808 + }, + { + "epoch": 0.07646635011848855, + "grad_norm": 0.6328125, + "learning_rate": 0.0019701646979624725, + "loss": 0.0996, + "step": 8809 + }, + { + "epoch": 0.07647503059869272, + "grad_norm": 0.2021484375, + "learning_rate": 0.00197015709775385, + "loss": 0.0894, + "step": 8810 + }, + { + "epoch": 0.07648371107889688, + "grad_norm": 0.18359375, + "learning_rate": 0.0019701494965936365, + "loss": 0.1484, + "step": 8811 + }, + { + "epoch": 0.07649239155910105, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019701418944818414, + "loss": 0.1113, + "step": 8812 + }, + { + "epoch": 0.07650107203930522, + "grad_norm": 0.28125, + "learning_rate": 0.0019701342914184715, + "loss": 0.1309, + "step": 8813 + }, + { + "epoch": 0.07650975251950938, + "grad_norm": 0.162109375, + "learning_rate": 0.001970126687403536, + "loss": 0.1484, + "step": 8814 + }, + { + "epoch": 0.07651843299971355, + "grad_norm": 0.80078125, + "learning_rate": 0.001970119082437043, + "loss": 0.1621, + "step": 8815 + }, + { + "epoch": 0.07652711347991771, + "grad_norm": 0.283203125, + "learning_rate": 0.001970111476519001, + "loss": 0.085, + "step": 8816 + }, + { + "epoch": 0.07653579396012188, + "grad_norm": 0.07421875, + "learning_rate": 0.001970103869649418, + "loss": 0.1094, + "step": 8817 + }, + { + "epoch": 0.07654447444032604, + "grad_norm": 0.419921875, + "learning_rate": 0.0019700962618283027, + "loss": 0.0869, + "step": 8818 + }, + { + "epoch": 0.0765531549205302, + "grad_norm": 0.30078125, + "learning_rate": 0.001970088653055663, + "loss": 0.127, + "step": 8819 + }, + { + "epoch": 0.07656183540073437, + "grad_norm": 0.478515625, + "learning_rate": 0.0019700810433315075, + "loss": 0.1475, + "step": 8820 + }, + { + "epoch": 0.07657051588093854, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019700734326558444, + "loss": 0.1338, + "step": 8821 + }, + { + "epoch": 0.0765791963611427, + "grad_norm": 0.10888671875, + "learning_rate": 0.001970065821028682, + "loss": 0.1064, + "step": 8822 + }, + { + "epoch": 0.07658787684134687, + "grad_norm": 0.189453125, + "learning_rate": 0.0019700582084500287, + "loss": 0.1157, + "step": 8823 + }, + { + "epoch": 0.07659655732155103, + "grad_norm": 0.2890625, + "learning_rate": 0.001970050594919893, + "loss": 0.0928, + "step": 8824 + }, + { + "epoch": 0.0766052378017552, + "grad_norm": 0.158203125, + "learning_rate": 0.001970042980438283, + "loss": 0.1416, + "step": 8825 + }, + { + "epoch": 0.07661391828195936, + "grad_norm": 0.11328125, + "learning_rate": 0.001970035365005207, + "loss": 0.166, + "step": 8826 + }, + { + "epoch": 0.07662259876216353, + "grad_norm": 0.318359375, + "learning_rate": 0.0019700277486206733, + "loss": 0.1553, + "step": 8827 + }, + { + "epoch": 0.07663127924236769, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019700201312846904, + "loss": 0.1562, + "step": 8828 + }, + { + "epoch": 0.07663995972257186, + "grad_norm": 0.302734375, + "learning_rate": 0.0019700125129972668, + "loss": 0.1465, + "step": 8829 + }, + { + "epoch": 0.07664864020277602, + "grad_norm": 0.97265625, + "learning_rate": 0.00197000489375841, + "loss": 0.1182, + "step": 8830 + }, + { + "epoch": 0.07665732068298019, + "grad_norm": 0.435546875, + "learning_rate": 0.0019699972735681293, + "loss": 0.1396, + "step": 8831 + }, + { + "epoch": 0.07666600116318435, + "grad_norm": 0.50390625, + "learning_rate": 0.001969989652426433, + "loss": 0.123, + "step": 8832 + }, + { + "epoch": 0.07667468164338852, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019699820303333286, + "loss": 0.1436, + "step": 8833 + }, + { + "epoch": 0.07668336212359268, + "grad_norm": 0.2353515625, + "learning_rate": 0.001969974407288825, + "loss": 0.1445, + "step": 8834 + }, + { + "epoch": 0.07669204260379685, + "grad_norm": 0.384765625, + "learning_rate": 0.0019699667832929306, + "loss": 0.1089, + "step": 8835 + }, + { + "epoch": 0.07670072308400101, + "grad_norm": 0.20703125, + "learning_rate": 0.0019699591583456537, + "loss": 0.1582, + "step": 8836 + }, + { + "epoch": 0.07670940356420518, + "grad_norm": 0.287109375, + "learning_rate": 0.001969951532447002, + "loss": 0.1113, + "step": 8837 + }, + { + "epoch": 0.07671808404440934, + "grad_norm": 0.578125, + "learning_rate": 0.0019699439055969846, + "loss": 0.1113, + "step": 8838 + }, + { + "epoch": 0.07672676452461351, + "grad_norm": 0.294921875, + "learning_rate": 0.00196993627779561, + "loss": 0.1318, + "step": 8839 + }, + { + "epoch": 0.07673544500481767, + "grad_norm": 0.130859375, + "learning_rate": 0.001969928649042886, + "loss": 0.1445, + "step": 8840 + }, + { + "epoch": 0.07674412548502182, + "grad_norm": 0.22265625, + "learning_rate": 0.0019699210193388213, + "loss": 0.1445, + "step": 8841 + }, + { + "epoch": 0.07675280596522599, + "grad_norm": 0.181640625, + "learning_rate": 0.0019699133886834233, + "loss": 0.0879, + "step": 8842 + }, + { + "epoch": 0.07676148644543015, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019699057570767018, + "loss": 0.1543, + "step": 8843 + }, + { + "epoch": 0.07677016692563432, + "grad_norm": 0.55859375, + "learning_rate": 0.001969898124518664, + "loss": 0.1099, + "step": 8844 + }, + { + "epoch": 0.07677884740583849, + "grad_norm": 0.76171875, + "learning_rate": 0.001969890491009319, + "loss": 0.0957, + "step": 8845 + }, + { + "epoch": 0.07678752788604265, + "grad_norm": 0.578125, + "learning_rate": 0.0019698828565486745, + "loss": 0.1309, + "step": 8846 + }, + { + "epoch": 0.07679620836624682, + "grad_norm": 0.50390625, + "learning_rate": 0.0019698752211367394, + "loss": 0.1504, + "step": 8847 + }, + { + "epoch": 0.07680488884645098, + "grad_norm": 0.087890625, + "learning_rate": 0.001969867584773522, + "loss": 0.1182, + "step": 8848 + }, + { + "epoch": 0.07681356932665515, + "grad_norm": 0.134765625, + "learning_rate": 0.00196985994745903, + "loss": 0.123, + "step": 8849 + }, + { + "epoch": 0.07682224980685931, + "grad_norm": 0.251953125, + "learning_rate": 0.0019698523091932723, + "loss": 0.1416, + "step": 8850 + }, + { + "epoch": 0.07683093028706348, + "grad_norm": 0.65234375, + "learning_rate": 0.0019698446699762577, + "loss": 0.1602, + "step": 8851 + }, + { + "epoch": 0.07683961076726764, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019698370298079936, + "loss": 0.1016, + "step": 8852 + }, + { + "epoch": 0.0768482912474718, + "grad_norm": 0.130859375, + "learning_rate": 0.0019698293886884887, + "loss": 0.1328, + "step": 8853 + }, + { + "epoch": 0.07685697172767597, + "grad_norm": 0.23046875, + "learning_rate": 0.0019698217466177516, + "loss": 0.1113, + "step": 8854 + }, + { + "epoch": 0.07686565220788014, + "grad_norm": 0.18359375, + "learning_rate": 0.0019698141035957907, + "loss": 0.1299, + "step": 8855 + }, + { + "epoch": 0.0768743326880843, + "grad_norm": 0.25, + "learning_rate": 0.001969806459622614, + "loss": 0.1133, + "step": 8856 + }, + { + "epoch": 0.07688301316828847, + "grad_norm": 0.26171875, + "learning_rate": 0.0019697988146982304, + "loss": 0.0977, + "step": 8857 + }, + { + "epoch": 0.07689169364849263, + "grad_norm": 0.349609375, + "learning_rate": 0.0019697911688226475, + "loss": 0.1582, + "step": 8858 + }, + { + "epoch": 0.0769003741286968, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019697835219958737, + "loss": 0.1338, + "step": 8859 + }, + { + "epoch": 0.07690905460890096, + "grad_norm": 0.255859375, + "learning_rate": 0.0019697758742179185, + "loss": 0.1113, + "step": 8860 + }, + { + "epoch": 0.07691773508910513, + "grad_norm": 0.431640625, + "learning_rate": 0.001969768225488789, + "loss": 0.0928, + "step": 8861 + }, + { + "epoch": 0.07692641556930929, + "grad_norm": 0.50390625, + "learning_rate": 0.001969760575808494, + "loss": 0.1348, + "step": 8862 + }, + { + "epoch": 0.07693509604951346, + "grad_norm": 0.392578125, + "learning_rate": 0.0019697529251770417, + "loss": 0.124, + "step": 8863 + }, + { + "epoch": 0.07694377652971762, + "grad_norm": 0.80859375, + "learning_rate": 0.001969745273594441, + "loss": 0.1455, + "step": 8864 + }, + { + "epoch": 0.07695245700992179, + "grad_norm": 0.423828125, + "learning_rate": 0.0019697376210607, + "loss": 0.1226, + "step": 8865 + }, + { + "epoch": 0.07696113749012595, + "grad_norm": 1.5234375, + "learning_rate": 0.001969729967575827, + "loss": 0.1416, + "step": 8866 + }, + { + "epoch": 0.07696981797033012, + "grad_norm": 0.1435546875, + "learning_rate": 0.00196972231313983, + "loss": 0.1182, + "step": 8867 + }, + { + "epoch": 0.07697849845053428, + "grad_norm": 0.3046875, + "learning_rate": 0.001969714657752718, + "loss": 0.0986, + "step": 8868 + }, + { + "epoch": 0.07698717893073845, + "grad_norm": 0.322265625, + "learning_rate": 0.0019697070014144994, + "loss": 0.1377, + "step": 8869 + }, + { + "epoch": 0.07699585941094261, + "grad_norm": 0.08984375, + "learning_rate": 0.001969699344125182, + "loss": 0.1069, + "step": 8870 + }, + { + "epoch": 0.07700453989114678, + "grad_norm": 0.369140625, + "learning_rate": 0.0019696916858847747, + "loss": 0.1328, + "step": 8871 + }, + { + "epoch": 0.07701322037135094, + "grad_norm": 0.3203125, + "learning_rate": 0.0019696840266932854, + "loss": 0.125, + "step": 8872 + }, + { + "epoch": 0.07702190085155511, + "grad_norm": 1.1484375, + "learning_rate": 0.0019696763665507226, + "loss": 0.1758, + "step": 8873 + }, + { + "epoch": 0.07703058133175927, + "grad_norm": 0.27734375, + "learning_rate": 0.001969668705457095, + "loss": 0.1084, + "step": 8874 + }, + { + "epoch": 0.07703926181196344, + "grad_norm": 0.25390625, + "learning_rate": 0.0019696610434124105, + "loss": 0.1152, + "step": 8875 + }, + { + "epoch": 0.0770479422921676, + "grad_norm": 0.416015625, + "learning_rate": 0.0019696533804166782, + "loss": 0.1084, + "step": 8876 + }, + { + "epoch": 0.07705662277237177, + "grad_norm": 0.189453125, + "learning_rate": 0.001969645716469906, + "loss": 0.166, + "step": 8877 + }, + { + "epoch": 0.07706530325257593, + "grad_norm": 0.4296875, + "learning_rate": 0.001969638051572102, + "loss": 0.1191, + "step": 8878 + }, + { + "epoch": 0.0770739837327801, + "grad_norm": 0.1494140625, + "learning_rate": 0.001969630385723275, + "loss": 0.1992, + "step": 8879 + }, + { + "epoch": 0.07708266421298426, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019696227189234332, + "loss": 0.1133, + "step": 8880 + }, + { + "epoch": 0.07709134469318843, + "grad_norm": 0.431640625, + "learning_rate": 0.0019696150511725853, + "loss": 0.1572, + "step": 8881 + }, + { + "epoch": 0.0771000251733926, + "grad_norm": 0.134765625, + "learning_rate": 0.001969607382470739, + "loss": 0.1089, + "step": 8882 + }, + { + "epoch": 0.07710870565359676, + "grad_norm": 0.44140625, + "learning_rate": 0.0019695997128179035, + "loss": 0.1328, + "step": 8883 + }, + { + "epoch": 0.07711738613380092, + "grad_norm": 0.26953125, + "learning_rate": 0.001969592042214087, + "loss": 0.1816, + "step": 8884 + }, + { + "epoch": 0.07712606661400509, + "grad_norm": 0.275390625, + "learning_rate": 0.0019695843706592974, + "loss": 0.124, + "step": 8885 + }, + { + "epoch": 0.07713474709420926, + "grad_norm": 0.146484375, + "learning_rate": 0.0019695766981535434, + "loss": 0.1621, + "step": 8886 + }, + { + "epoch": 0.07714342757441342, + "grad_norm": 0.34375, + "learning_rate": 0.001969569024696833, + "loss": 0.1201, + "step": 8887 + }, + { + "epoch": 0.07715210805461759, + "grad_norm": 0.4296875, + "learning_rate": 0.001969561350289176, + "loss": 0.1216, + "step": 8888 + }, + { + "epoch": 0.07716078853482175, + "grad_norm": 0.0888671875, + "learning_rate": 0.001969553674930579, + "loss": 0.1504, + "step": 8889 + }, + { + "epoch": 0.07716946901502592, + "grad_norm": 0.13671875, + "learning_rate": 0.0019695459986210514, + "loss": 0.1631, + "step": 8890 + }, + { + "epoch": 0.07717814949523008, + "grad_norm": 0.07421875, + "learning_rate": 0.001969538321360601, + "loss": 0.127, + "step": 8891 + }, + { + "epoch": 0.07718682997543425, + "grad_norm": 0.12353515625, + "learning_rate": 0.001969530643149237, + "loss": 0.1553, + "step": 8892 + }, + { + "epoch": 0.07719551045563841, + "grad_norm": 0.36328125, + "learning_rate": 0.001969522963986967, + "loss": 0.1094, + "step": 8893 + }, + { + "epoch": 0.07720419093584258, + "grad_norm": 1.0, + "learning_rate": 0.0019695152838738, + "loss": 0.1377, + "step": 8894 + }, + { + "epoch": 0.07721287141604674, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019695076028097445, + "loss": 0.1318, + "step": 8895 + }, + { + "epoch": 0.0772215518962509, + "grad_norm": 0.66796875, + "learning_rate": 0.001969499920794808, + "loss": 0.0796, + "step": 8896 + }, + { + "epoch": 0.07723023237645507, + "grad_norm": 0.18359375, + "learning_rate": 0.0019694922378289998, + "loss": 0.0933, + "step": 8897 + }, + { + "epoch": 0.07723891285665924, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019694845539123275, + "loss": 0.1494, + "step": 8898 + }, + { + "epoch": 0.0772475933368634, + "grad_norm": 0.07421875, + "learning_rate": 0.0019694768690448005, + "loss": 0.1089, + "step": 8899 + }, + { + "epoch": 0.07725627381706757, + "grad_norm": 0.609375, + "learning_rate": 0.0019694691832264264, + "loss": 0.168, + "step": 8900 + }, + { + "epoch": 0.07726495429727173, + "grad_norm": 0.1845703125, + "learning_rate": 0.001969461496457214, + "loss": 0.1416, + "step": 8901 + }, + { + "epoch": 0.07727363477747588, + "grad_norm": 0.36328125, + "learning_rate": 0.0019694538087371713, + "loss": 0.1074, + "step": 8902 + }, + { + "epoch": 0.07728231525768005, + "grad_norm": 0.44921875, + "learning_rate": 0.001969446120066307, + "loss": 0.1221, + "step": 8903 + }, + { + "epoch": 0.07729099573788421, + "grad_norm": 0.89453125, + "learning_rate": 0.0019694384304446296, + "loss": 0.1924, + "step": 8904 + }, + { + "epoch": 0.07729967621808838, + "grad_norm": 0.640625, + "learning_rate": 0.0019694307398721474, + "loss": 0.1143, + "step": 8905 + }, + { + "epoch": 0.07730835669829254, + "grad_norm": 0.32421875, + "learning_rate": 0.001969423048348869, + "loss": 0.1406, + "step": 8906 + }, + { + "epoch": 0.07731703717849671, + "grad_norm": 0.3125, + "learning_rate": 0.001969415355874802, + "loss": 0.1426, + "step": 8907 + }, + { + "epoch": 0.07732571765870087, + "grad_norm": 0.51171875, + "learning_rate": 0.001969407662449956, + "loss": 0.1348, + "step": 8908 + }, + { + "epoch": 0.07733439813890504, + "grad_norm": 0.203125, + "learning_rate": 0.0019693999680743387, + "loss": 0.0972, + "step": 8909 + }, + { + "epoch": 0.0773430786191092, + "grad_norm": 0.76171875, + "learning_rate": 0.0019693922727479587, + "loss": 0.1069, + "step": 8910 + }, + { + "epoch": 0.07735175909931337, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019693845764708244, + "loss": 0.1201, + "step": 8911 + }, + { + "epoch": 0.07736043957951753, + "grad_norm": 0.09912109375, + "learning_rate": 0.001969376879242944, + "loss": 0.1543, + "step": 8912 + }, + { + "epoch": 0.0773691200597217, + "grad_norm": 0.345703125, + "learning_rate": 0.001969369181064326, + "loss": 0.0981, + "step": 8913 + }, + { + "epoch": 0.07737780053992586, + "grad_norm": 0.72265625, + "learning_rate": 0.001969361481934979, + "loss": 0.127, + "step": 8914 + }, + { + "epoch": 0.07738648102013003, + "grad_norm": 0.0849609375, + "learning_rate": 0.001969353781854912, + "loss": 0.1055, + "step": 8915 + }, + { + "epoch": 0.0773951615003342, + "grad_norm": 0.33984375, + "learning_rate": 0.001969346080824132, + "loss": 0.125, + "step": 8916 + }, + { + "epoch": 0.07740384198053836, + "grad_norm": 0.41015625, + "learning_rate": 0.001969338378842649, + "loss": 0.1523, + "step": 8917 + }, + { + "epoch": 0.07741252246074252, + "grad_norm": 0.3203125, + "learning_rate": 0.0019693306759104696, + "loss": 0.1104, + "step": 8918 + }, + { + "epoch": 0.07742120294094669, + "grad_norm": 0.458984375, + "learning_rate": 0.0019693229720276037, + "loss": 0.1133, + "step": 8919 + }, + { + "epoch": 0.07742988342115086, + "grad_norm": 0.353515625, + "learning_rate": 0.001969315267194059, + "loss": 0.1084, + "step": 8920 + }, + { + "epoch": 0.07743856390135502, + "grad_norm": 0.455078125, + "learning_rate": 0.0019693075614098444, + "loss": 0.1582, + "step": 8921 + }, + { + "epoch": 0.07744724438155919, + "grad_norm": 0.49609375, + "learning_rate": 0.0019692998546749683, + "loss": 0.1719, + "step": 8922 + }, + { + "epoch": 0.07745592486176335, + "grad_norm": 0.373046875, + "learning_rate": 0.001969292146989439, + "loss": 0.1309, + "step": 8923 + }, + { + "epoch": 0.07746460534196752, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019692844383532642, + "loss": 0.1289, + "step": 8924 + }, + { + "epoch": 0.07747328582217168, + "grad_norm": 0.69140625, + "learning_rate": 0.0019692767287664537, + "loss": 0.1182, + "step": 8925 + }, + { + "epoch": 0.07748196630237585, + "grad_norm": 0.62890625, + "learning_rate": 0.001969269018229015, + "loss": 0.127, + "step": 8926 + }, + { + "epoch": 0.07749064678258001, + "grad_norm": 0.31640625, + "learning_rate": 0.0019692613067409568, + "loss": 0.1182, + "step": 8927 + }, + { + "epoch": 0.07749932726278418, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019692535943022873, + "loss": 0.166, + "step": 8928 + }, + { + "epoch": 0.07750800774298834, + "grad_norm": 0.3828125, + "learning_rate": 0.0019692458809130154, + "loss": 0.1367, + "step": 8929 + }, + { + "epoch": 0.0775166882231925, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019692381665731493, + "loss": 0.1328, + "step": 8930 + }, + { + "epoch": 0.07752536870339667, + "grad_norm": 0.26953125, + "learning_rate": 0.001969230451282697, + "loss": 0.1035, + "step": 8931 + }, + { + "epoch": 0.07753404918360084, + "grad_norm": 0.38671875, + "learning_rate": 0.001969222735041668, + "loss": 0.1494, + "step": 8932 + }, + { + "epoch": 0.077542729663805, + "grad_norm": 0.078125, + "learning_rate": 0.0019692150178500694, + "loss": 0.1021, + "step": 8933 + }, + { + "epoch": 0.07755141014400917, + "grad_norm": 0.34765625, + "learning_rate": 0.0019692072997079106, + "loss": 0.165, + "step": 8934 + }, + { + "epoch": 0.07756009062421333, + "grad_norm": 0.123046875, + "learning_rate": 0.0019691995806151996, + "loss": 0.1084, + "step": 8935 + }, + { + "epoch": 0.0775687711044175, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019691918605719448, + "loss": 0.1475, + "step": 8936 + }, + { + "epoch": 0.07757745158462166, + "grad_norm": 0.47265625, + "learning_rate": 0.001969184139578155, + "loss": 0.1143, + "step": 8937 + }, + { + "epoch": 0.07758613206482583, + "grad_norm": 0.287109375, + "learning_rate": 0.001969176417633839, + "loss": 0.1338, + "step": 8938 + }, + { + "epoch": 0.07759481254502999, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019691686947390045, + "loss": 0.1992, + "step": 8939 + }, + { + "epoch": 0.07760349302523416, + "grad_norm": 0.255859375, + "learning_rate": 0.00196916097089366, + "loss": 0.1221, + "step": 8940 + }, + { + "epoch": 0.07761217350543832, + "grad_norm": 0.15234375, + "learning_rate": 0.001969153246097814, + "loss": 0.1455, + "step": 8941 + }, + { + "epoch": 0.07762085398564249, + "grad_norm": 0.435546875, + "learning_rate": 0.0019691455203514753, + "loss": 0.1387, + "step": 8942 + }, + { + "epoch": 0.07762953446584665, + "grad_norm": 0.6015625, + "learning_rate": 0.001969137793654652, + "loss": 0.1309, + "step": 8943 + }, + { + "epoch": 0.07763821494605082, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019691300660073527, + "loss": 0.1895, + "step": 8944 + }, + { + "epoch": 0.07764689542625498, + "grad_norm": 0.69140625, + "learning_rate": 0.0019691223374095857, + "loss": 0.1035, + "step": 8945 + }, + { + "epoch": 0.07765557590645915, + "grad_norm": 0.1640625, + "learning_rate": 0.00196911460786136, + "loss": 0.1309, + "step": 8946 + }, + { + "epoch": 0.07766425638666331, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019691068773626834, + "loss": 0.0981, + "step": 8947 + }, + { + "epoch": 0.07767293686686748, + "grad_norm": 0.142578125, + "learning_rate": 0.0019690991459135644, + "loss": 0.0874, + "step": 8948 + }, + { + "epoch": 0.07768161734707164, + "grad_norm": 0.19921875, + "learning_rate": 0.0019690914135140117, + "loss": 0.1973, + "step": 8949 + }, + { + "epoch": 0.07769029782727581, + "grad_norm": 0.083984375, + "learning_rate": 0.0019690836801640335, + "loss": 0.1104, + "step": 8950 + }, + { + "epoch": 0.07769897830747997, + "grad_norm": 0.51953125, + "learning_rate": 0.001969075945863639, + "loss": 0.127, + "step": 8951 + }, + { + "epoch": 0.07770765878768414, + "grad_norm": 0.15625, + "learning_rate": 0.001969068210612836, + "loss": 0.1113, + "step": 8952 + }, + { + "epoch": 0.0777163392678883, + "grad_norm": 0.83984375, + "learning_rate": 0.0019690604744116327, + "loss": 0.2695, + "step": 8953 + }, + { + "epoch": 0.07772501974809247, + "grad_norm": 0.494140625, + "learning_rate": 0.001969052737260038, + "loss": 0.1309, + "step": 8954 + }, + { + "epoch": 0.07773370022829663, + "grad_norm": 0.279296875, + "learning_rate": 0.00196904499915806, + "loss": 0.166, + "step": 8955 + }, + { + "epoch": 0.0777423807085008, + "grad_norm": 0.1005859375, + "learning_rate": 0.001969037260105708, + "loss": 0.165, + "step": 8956 + }, + { + "epoch": 0.07775106118870496, + "grad_norm": 0.4375, + "learning_rate": 0.00196902952010299, + "loss": 0.127, + "step": 8957 + }, + { + "epoch": 0.07775974166890913, + "grad_norm": 0.1220703125, + "learning_rate": 0.001969021779149914, + "loss": 0.1426, + "step": 8958 + }, + { + "epoch": 0.0777684221491133, + "grad_norm": 0.48828125, + "learning_rate": 0.0019690140372464887, + "loss": 0.1387, + "step": 8959 + }, + { + "epoch": 0.07777710262931746, + "grad_norm": 0.5234375, + "learning_rate": 0.001969006294392723, + "loss": 0.1104, + "step": 8960 + }, + { + "epoch": 0.07778578310952163, + "grad_norm": 0.81640625, + "learning_rate": 0.001968998550588625, + "loss": 0.123, + "step": 8961 + }, + { + "epoch": 0.07779446358972579, + "grad_norm": 0.67578125, + "learning_rate": 0.001968990805834203, + "loss": 0.124, + "step": 8962 + }, + { + "epoch": 0.07780314406992996, + "grad_norm": 0.5546875, + "learning_rate": 0.001968983060129466, + "loss": 0.1104, + "step": 8963 + }, + { + "epoch": 0.0778118245501341, + "grad_norm": 0.3515625, + "learning_rate": 0.001968975313474422, + "loss": 0.1572, + "step": 8964 + }, + { + "epoch": 0.07782050503033827, + "grad_norm": 0.279296875, + "learning_rate": 0.00196896756586908, + "loss": 0.1953, + "step": 8965 + }, + { + "epoch": 0.07782918551054244, + "grad_norm": 0.115234375, + "learning_rate": 0.001968959817313448, + "loss": 0.1592, + "step": 8966 + }, + { + "epoch": 0.0778378659907466, + "grad_norm": 2.046875, + "learning_rate": 0.0019689520678075347, + "loss": 0.123, + "step": 8967 + }, + { + "epoch": 0.07784654647095077, + "grad_norm": 0.2177734375, + "learning_rate": 0.001968944317351348, + "loss": 0.1211, + "step": 8968 + }, + { + "epoch": 0.07785522695115493, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019689365659448974, + "loss": 0.1426, + "step": 8969 + }, + { + "epoch": 0.0778639074313591, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019689288135881903, + "loss": 0.1123, + "step": 8970 + }, + { + "epoch": 0.07787258791156326, + "grad_norm": 0.66796875, + "learning_rate": 0.0019689210602812364, + "loss": 0.1152, + "step": 8971 + }, + { + "epoch": 0.07788126839176743, + "grad_norm": 0.5859375, + "learning_rate": 0.0019689133060240426, + "loss": 0.105, + "step": 8972 + }, + { + "epoch": 0.07788994887197159, + "grad_norm": 0.12255859375, + "learning_rate": 0.001968905550816619, + "loss": 0.123, + "step": 8973 + }, + { + "epoch": 0.07789862935217576, + "grad_norm": 0.83984375, + "learning_rate": 0.0019688977946589734, + "loss": 0.1484, + "step": 8974 + }, + { + "epoch": 0.07790730983237992, + "grad_norm": 0.65234375, + "learning_rate": 0.0019688900375511133, + "loss": 0.1299, + "step": 8975 + }, + { + "epoch": 0.07791599031258409, + "grad_norm": 0.138671875, + "learning_rate": 0.001968882279493049, + "loss": 0.1045, + "step": 8976 + }, + { + "epoch": 0.07792467079278825, + "grad_norm": 0.4921875, + "learning_rate": 0.0019688745204847874, + "loss": 0.1226, + "step": 8977 + }, + { + "epoch": 0.07793335127299242, + "grad_norm": 0.1357421875, + "learning_rate": 0.001968866760526338, + "loss": 0.1504, + "step": 8978 + }, + { + "epoch": 0.07794203175319658, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019688589996177093, + "loss": 0.1021, + "step": 8979 + }, + { + "epoch": 0.07795071223340075, + "grad_norm": 0.1328125, + "learning_rate": 0.001968851237758909, + "loss": 0.1758, + "step": 8980 + }, + { + "epoch": 0.07795939271360491, + "grad_norm": 0.341796875, + "learning_rate": 0.0019688434749499466, + "loss": 0.1611, + "step": 8981 + }, + { + "epoch": 0.07796807319380908, + "grad_norm": 0.310546875, + "learning_rate": 0.0019688357111908297, + "loss": 0.1172, + "step": 8982 + }, + { + "epoch": 0.07797675367401324, + "grad_norm": 0.2158203125, + "learning_rate": 0.001968827946481567, + "loss": 0.1523, + "step": 8983 + }, + { + "epoch": 0.07798543415421741, + "grad_norm": 0.578125, + "learning_rate": 0.001968820180822167, + "loss": 0.1133, + "step": 8984 + }, + { + "epoch": 0.07799411463442157, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019688124142126385, + "loss": 0.0977, + "step": 8985 + }, + { + "epoch": 0.07800279511462574, + "grad_norm": 0.078125, + "learning_rate": 0.00196880464665299, + "loss": 0.0977, + "step": 8986 + }, + { + "epoch": 0.0780114755948299, + "grad_norm": 0.35546875, + "learning_rate": 0.0019687968781432297, + "loss": 0.1367, + "step": 8987 + }, + { + "epoch": 0.07802015607503407, + "grad_norm": 0.123046875, + "learning_rate": 0.001968789108683366, + "loss": 0.1045, + "step": 8988 + }, + { + "epoch": 0.07802883655523823, + "grad_norm": 0.66015625, + "learning_rate": 0.0019687813382734077, + "loss": 0.1064, + "step": 8989 + }, + { + "epoch": 0.0780375170354424, + "grad_norm": 0.283203125, + "learning_rate": 0.001968773566913363, + "loss": 0.1133, + "step": 8990 + }, + { + "epoch": 0.07804619751564656, + "grad_norm": 0.359375, + "learning_rate": 0.0019687657946032408, + "loss": 0.1533, + "step": 8991 + }, + { + "epoch": 0.07805487799585073, + "grad_norm": 0.72265625, + "learning_rate": 0.0019687580213430494, + "loss": 0.1709, + "step": 8992 + }, + { + "epoch": 0.0780635584760549, + "grad_norm": 0.2333984375, + "learning_rate": 0.001968750247132797, + "loss": 0.1289, + "step": 8993 + }, + { + "epoch": 0.07807223895625906, + "grad_norm": 0.16796875, + "learning_rate": 0.0019687424719724927, + "loss": 0.1387, + "step": 8994 + }, + { + "epoch": 0.07808091943646323, + "grad_norm": 0.427734375, + "learning_rate": 0.0019687346958621446, + "loss": 0.167, + "step": 8995 + }, + { + "epoch": 0.07808959991666739, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019687269188017613, + "loss": 0.1318, + "step": 8996 + }, + { + "epoch": 0.07809828039687156, + "grad_norm": 0.2578125, + "learning_rate": 0.001968719140791351, + "loss": 0.1216, + "step": 8997 + }, + { + "epoch": 0.07810696087707572, + "grad_norm": 0.22265625, + "learning_rate": 0.001968711361830923, + "loss": 0.1074, + "step": 8998 + }, + { + "epoch": 0.07811564135727989, + "grad_norm": 0.56640625, + "learning_rate": 0.001968703581920485, + "loss": 0.1201, + "step": 8999 + }, + { + "epoch": 0.07812432183748405, + "grad_norm": 0.07666015625, + "learning_rate": 0.001968695801060046, + "loss": 0.1094, + "step": 9000 + }, + { + "epoch": 0.07813300231768822, + "grad_norm": 0.375, + "learning_rate": 0.0019686880192496137, + "loss": 0.1113, + "step": 9001 + }, + { + "epoch": 0.07814168279789238, + "grad_norm": 0.09130859375, + "learning_rate": 0.001968680236489198, + "loss": 0.1104, + "step": 9002 + }, + { + "epoch": 0.07815036327809655, + "grad_norm": 0.1064453125, + "learning_rate": 0.001968672452778806, + "loss": 0.1348, + "step": 9003 + }, + { + "epoch": 0.07815904375830071, + "grad_norm": 0.41015625, + "learning_rate": 0.0019686646681184475, + "loss": 0.1543, + "step": 9004 + }, + { + "epoch": 0.07816772423850488, + "grad_norm": 0.17578125, + "learning_rate": 0.00196865688250813, + "loss": 0.1367, + "step": 9005 + }, + { + "epoch": 0.07817640471870904, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019686490959478623, + "loss": 0.1211, + "step": 9006 + }, + { + "epoch": 0.0781850851989132, + "grad_norm": 0.91796875, + "learning_rate": 0.0019686413084376536, + "loss": 0.1289, + "step": 9007 + }, + { + "epoch": 0.07819376567911737, + "grad_norm": 0.32421875, + "learning_rate": 0.001968633519977511, + "loss": 0.1279, + "step": 9008 + }, + { + "epoch": 0.07820244615932154, + "grad_norm": 0.95703125, + "learning_rate": 0.0019686257305674445, + "loss": 0.2031, + "step": 9009 + }, + { + "epoch": 0.0782111266395257, + "grad_norm": 0.33203125, + "learning_rate": 0.0019686179402074616, + "loss": 0.166, + "step": 9010 + }, + { + "epoch": 0.07821980711972987, + "grad_norm": 0.6484375, + "learning_rate": 0.0019686101488975713, + "loss": 0.1621, + "step": 9011 + }, + { + "epoch": 0.07822848759993403, + "grad_norm": 0.546875, + "learning_rate": 0.001968602356637782, + "loss": 0.1338, + "step": 9012 + }, + { + "epoch": 0.0782371680801382, + "grad_norm": 0.5234375, + "learning_rate": 0.001968594563428102, + "loss": 0.1211, + "step": 9013 + }, + { + "epoch": 0.07824584856034236, + "grad_norm": 0.73828125, + "learning_rate": 0.0019685867692685404, + "loss": 0.127, + "step": 9014 + }, + { + "epoch": 0.07825452904054653, + "grad_norm": 1.078125, + "learning_rate": 0.001968578974159105, + "loss": 0.1465, + "step": 9015 + }, + { + "epoch": 0.07826320952075069, + "grad_norm": 0.1015625, + "learning_rate": 0.001968571178099805, + "loss": 0.1221, + "step": 9016 + }, + { + "epoch": 0.07827189000095486, + "grad_norm": 0.5, + "learning_rate": 0.001968563381090648, + "loss": 0.165, + "step": 9017 + }, + { + "epoch": 0.07828057048115902, + "grad_norm": 1.234375, + "learning_rate": 0.0019685555831316435, + "loss": 0.127, + "step": 9018 + }, + { + "epoch": 0.07828925096136319, + "grad_norm": 0.34375, + "learning_rate": 0.0019685477842228, + "loss": 0.1504, + "step": 9019 + }, + { + "epoch": 0.07829793144156735, + "grad_norm": 0.8046875, + "learning_rate": 0.0019685399843641254, + "loss": 0.1396, + "step": 9020 + }, + { + "epoch": 0.07830661192177152, + "grad_norm": 0.94921875, + "learning_rate": 0.001968532183555629, + "loss": 0.1348, + "step": 9021 + }, + { + "epoch": 0.07831529240197568, + "grad_norm": 1.796875, + "learning_rate": 0.001968524381797318, + "loss": 0.1641, + "step": 9022 + }, + { + "epoch": 0.07832397288217985, + "grad_norm": 0.1259765625, + "learning_rate": 0.001968516579089202, + "loss": 0.166, + "step": 9023 + }, + { + "epoch": 0.07833265336238401, + "grad_norm": 0.0791015625, + "learning_rate": 0.00196850877543129, + "loss": 0.1211, + "step": 9024 + }, + { + "epoch": 0.07834133384258818, + "grad_norm": 0.6484375, + "learning_rate": 0.001968500970823589, + "loss": 0.1729, + "step": 9025 + }, + { + "epoch": 0.07835001432279233, + "grad_norm": 0.45703125, + "learning_rate": 0.001968493165266109, + "loss": 0.1338, + "step": 9026 + }, + { + "epoch": 0.0783586948029965, + "grad_norm": 0.17578125, + "learning_rate": 0.0019684853587588575, + "loss": 0.125, + "step": 9027 + }, + { + "epoch": 0.07836737528320066, + "grad_norm": 0.68359375, + "learning_rate": 0.0019684775513018437, + "loss": 0.1426, + "step": 9028 + }, + { + "epoch": 0.07837605576340483, + "grad_norm": 0.146484375, + "learning_rate": 0.001968469742895076, + "loss": 0.1221, + "step": 9029 + }, + { + "epoch": 0.07838473624360899, + "grad_norm": 0.29296875, + "learning_rate": 0.0019684619335385623, + "loss": 0.1211, + "step": 9030 + }, + { + "epoch": 0.07839341672381316, + "grad_norm": 0.296875, + "learning_rate": 0.0019684541232323125, + "loss": 0.1064, + "step": 9031 + }, + { + "epoch": 0.07840209720401732, + "grad_norm": 0.404296875, + "learning_rate": 0.0019684463119763337, + "loss": 0.1367, + "step": 9032 + }, + { + "epoch": 0.07841077768422149, + "grad_norm": 0.439453125, + "learning_rate": 0.0019684384997706353, + "loss": 0.1455, + "step": 9033 + }, + { + "epoch": 0.07841945816442565, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019684306866152254, + "loss": 0.1582, + "step": 9034 + }, + { + "epoch": 0.07842813864462982, + "grad_norm": 0.1650390625, + "learning_rate": 0.001968422872510113, + "loss": 0.1162, + "step": 9035 + }, + { + "epoch": 0.07843681912483398, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019684150574553064, + "loss": 0.1055, + "step": 9036 + }, + { + "epoch": 0.07844549960503815, + "grad_norm": 0.404296875, + "learning_rate": 0.001968407241450814, + "loss": 0.127, + "step": 9037 + }, + { + "epoch": 0.07845418008524231, + "grad_norm": 0.515625, + "learning_rate": 0.0019683994244966444, + "loss": 0.2139, + "step": 9038 + }, + { + "epoch": 0.07846286056544648, + "grad_norm": 0.291015625, + "learning_rate": 0.0019683916065928063, + "loss": 0.1885, + "step": 9039 + }, + { + "epoch": 0.07847154104565064, + "grad_norm": 0.125, + "learning_rate": 0.001968383787739308, + "loss": 0.1191, + "step": 9040 + }, + { + "epoch": 0.0784802215258548, + "grad_norm": 0.255859375, + "learning_rate": 0.001968375967936159, + "loss": 0.1162, + "step": 9041 + }, + { + "epoch": 0.07848890200605897, + "grad_norm": 0.21484375, + "learning_rate": 0.001968368147183366, + "loss": 0.1289, + "step": 9042 + }, + { + "epoch": 0.07849758248626314, + "grad_norm": 0.197265625, + "learning_rate": 0.0019683603254809397, + "loss": 0.1055, + "step": 9043 + }, + { + "epoch": 0.0785062629664673, + "grad_norm": 0.2119140625, + "learning_rate": 0.001968352502828887, + "loss": 0.1172, + "step": 9044 + }, + { + "epoch": 0.07851494344667147, + "grad_norm": 0.1435546875, + "learning_rate": 0.001968344679227217, + "loss": 0.1387, + "step": 9045 + }, + { + "epoch": 0.07852362392687563, + "grad_norm": 0.0859375, + "learning_rate": 0.0019683368546759383, + "loss": 0.1123, + "step": 9046 + }, + { + "epoch": 0.0785323044070798, + "grad_norm": 0.34375, + "learning_rate": 0.00196832902917506, + "loss": 0.1182, + "step": 9047 + }, + { + "epoch": 0.07854098488728396, + "grad_norm": 0.76953125, + "learning_rate": 0.001968321202724589, + "loss": 0.1143, + "step": 9048 + }, + { + "epoch": 0.07854966536748813, + "grad_norm": 0.10986328125, + "learning_rate": 0.001968313375324536, + "loss": 0.1357, + "step": 9049 + }, + { + "epoch": 0.07855834584769229, + "grad_norm": 0.7265625, + "learning_rate": 0.0019683055469749086, + "loss": 0.1143, + "step": 9050 + }, + { + "epoch": 0.07856702632789646, + "grad_norm": 0.08935546875, + "learning_rate": 0.001968297717675715, + "loss": 0.0986, + "step": 9051 + }, + { + "epoch": 0.07857570680810062, + "grad_norm": 0.3046875, + "learning_rate": 0.0019682898874269637, + "loss": 0.1445, + "step": 9052 + }, + { + "epoch": 0.07858438728830479, + "grad_norm": 0.107421875, + "learning_rate": 0.001968282056228664, + "loss": 0.1514, + "step": 9053 + }, + { + "epoch": 0.07859306776850895, + "grad_norm": 0.515625, + "learning_rate": 0.001968274224080824, + "loss": 0.123, + "step": 9054 + }, + { + "epoch": 0.07860174824871312, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019682663909834524, + "loss": 0.1201, + "step": 9055 + }, + { + "epoch": 0.07861042872891728, + "grad_norm": 0.373046875, + "learning_rate": 0.0019682585569365577, + "loss": 0.0957, + "step": 9056 + }, + { + "epoch": 0.07861910920912145, + "grad_norm": 0.2265625, + "learning_rate": 0.0019682507219401487, + "loss": 0.1602, + "step": 9057 + }, + { + "epoch": 0.07862778968932561, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019682428859942335, + "loss": 0.1113, + "step": 9058 + }, + { + "epoch": 0.07863647016952978, + "grad_norm": 0.28125, + "learning_rate": 0.001968235049098821, + "loss": 0.1279, + "step": 9059 + }, + { + "epoch": 0.07864515064973394, + "grad_norm": 0.236328125, + "learning_rate": 0.0019682272112539196, + "loss": 0.0908, + "step": 9060 + }, + { + "epoch": 0.07865383112993811, + "grad_norm": 0.31640625, + "learning_rate": 0.0019682193724595383, + "loss": 0.1289, + "step": 9061 + }, + { + "epoch": 0.07866251161014227, + "grad_norm": 0.07666015625, + "learning_rate": 0.001968211532715685, + "loss": 0.1328, + "step": 9062 + }, + { + "epoch": 0.07867119209034644, + "grad_norm": 1.3828125, + "learning_rate": 0.0019682036920223686, + "loss": 0.1104, + "step": 9063 + }, + { + "epoch": 0.0786798725705506, + "grad_norm": 0.228515625, + "learning_rate": 0.0019681958503795977, + "loss": 0.1338, + "step": 9064 + }, + { + "epoch": 0.07868855305075477, + "grad_norm": 0.33203125, + "learning_rate": 0.001968188007787381, + "loss": 0.1201, + "step": 9065 + }, + { + "epoch": 0.07869723353095893, + "grad_norm": 0.337890625, + "learning_rate": 0.0019681801642457268, + "loss": 0.1118, + "step": 9066 + }, + { + "epoch": 0.0787059140111631, + "grad_norm": 0.103515625, + "learning_rate": 0.001968172319754644, + "loss": 0.1187, + "step": 9067 + }, + { + "epoch": 0.07871459449136727, + "grad_norm": 0.12158203125, + "learning_rate": 0.001968164474314141, + "loss": 0.1104, + "step": 9068 + }, + { + "epoch": 0.07872327497157143, + "grad_norm": 0.58984375, + "learning_rate": 0.001968156627924226, + "loss": 0.1621, + "step": 9069 + }, + { + "epoch": 0.0787319554517756, + "grad_norm": 0.1220703125, + "learning_rate": 0.001968148780584908, + "loss": 0.1289, + "step": 9070 + }, + { + "epoch": 0.07874063593197976, + "grad_norm": 0.166015625, + "learning_rate": 0.001968140932296196, + "loss": 0.1738, + "step": 9071 + }, + { + "epoch": 0.07874931641218393, + "grad_norm": 0.421875, + "learning_rate": 0.001968133083058098, + "loss": 0.1074, + "step": 9072 + }, + { + "epoch": 0.07875799689238809, + "grad_norm": 0.640625, + "learning_rate": 0.0019681252328706228, + "loss": 0.1235, + "step": 9073 + }, + { + "epoch": 0.07876667737259226, + "grad_norm": 0.44921875, + "learning_rate": 0.0019681173817337784, + "loss": 0.1191, + "step": 9074 + }, + { + "epoch": 0.07877535785279642, + "grad_norm": 0.73046875, + "learning_rate": 0.001968109529647574, + "loss": 0.123, + "step": 9075 + }, + { + "epoch": 0.07878403833300059, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019681016766120185, + "loss": 0.1328, + "step": 9076 + }, + { + "epoch": 0.07879271881320475, + "grad_norm": 0.1875, + "learning_rate": 0.00196809382262712, + "loss": 0.1572, + "step": 9077 + }, + { + "epoch": 0.07880139929340892, + "grad_norm": 0.0927734375, + "learning_rate": 0.001968085967692887, + "loss": 0.1318, + "step": 9078 + }, + { + "epoch": 0.07881007977361308, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019680781118093277, + "loss": 0.166, + "step": 9079 + }, + { + "epoch": 0.07881876025381725, + "grad_norm": 0.470703125, + "learning_rate": 0.0019680702549764516, + "loss": 0.207, + "step": 9080 + }, + { + "epoch": 0.07882744073402141, + "grad_norm": 0.4453125, + "learning_rate": 0.0019680623971942666, + "loss": 0.1118, + "step": 9081 + }, + { + "epoch": 0.07883612121422558, + "grad_norm": 0.26953125, + "learning_rate": 0.001968054538462782, + "loss": 0.1357, + "step": 9082 + }, + { + "epoch": 0.07884480169442974, + "grad_norm": 0.67578125, + "learning_rate": 0.0019680466787820065, + "loss": 0.1357, + "step": 9083 + }, + { + "epoch": 0.07885348217463391, + "grad_norm": 0.6328125, + "learning_rate": 0.0019680388181519473, + "loss": 0.1895, + "step": 9084 + }, + { + "epoch": 0.07886216265483807, + "grad_norm": 0.65234375, + "learning_rate": 0.0019680309565726145, + "loss": 0.1079, + "step": 9085 + }, + { + "epoch": 0.07887084313504224, + "grad_norm": 0.1279296875, + "learning_rate": 0.001968023094044016, + "loss": 0.104, + "step": 9086 + }, + { + "epoch": 0.07887952361524639, + "grad_norm": 0.251953125, + "learning_rate": 0.00196801523056616, + "loss": 0.0859, + "step": 9087 + }, + { + "epoch": 0.07888820409545055, + "grad_norm": 0.416015625, + "learning_rate": 0.001968007366139056, + "loss": 0.1846, + "step": 9088 + }, + { + "epoch": 0.07889688457565472, + "grad_norm": 0.1396484375, + "learning_rate": 0.001967999500762712, + "loss": 0.1602, + "step": 9089 + }, + { + "epoch": 0.07890556505585888, + "grad_norm": 0.169921875, + "learning_rate": 0.001967991634437137, + "loss": 0.1426, + "step": 9090 + }, + { + "epoch": 0.07891424553606305, + "grad_norm": 1.3515625, + "learning_rate": 0.0019679837671623393, + "loss": 0.1885, + "step": 9091 + }, + { + "epoch": 0.07892292601626721, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019679758989383273, + "loss": 0.085, + "step": 9092 + }, + { + "epoch": 0.07893160649647138, + "grad_norm": 0.5546875, + "learning_rate": 0.00196796802976511, + "loss": 0.1328, + "step": 9093 + }, + { + "epoch": 0.07894028697667554, + "grad_norm": 0.09033203125, + "learning_rate": 0.001967960159642696, + "loss": 0.1348, + "step": 9094 + }, + { + "epoch": 0.07894896745687971, + "grad_norm": 0.12109375, + "learning_rate": 0.001967952288571094, + "loss": 0.1201, + "step": 9095 + }, + { + "epoch": 0.07895764793708387, + "grad_norm": 0.58984375, + "learning_rate": 0.0019679444165503126, + "loss": 0.1406, + "step": 9096 + }, + { + "epoch": 0.07896632841728804, + "grad_norm": 0.65234375, + "learning_rate": 0.00196793654358036, + "loss": 0.1348, + "step": 9097 + }, + { + "epoch": 0.0789750088974922, + "grad_norm": 0.9765625, + "learning_rate": 0.0019679286696612447, + "loss": 0.1328, + "step": 9098 + }, + { + "epoch": 0.07898368937769637, + "grad_norm": 0.197265625, + "learning_rate": 0.001967920794792976, + "loss": 0.1143, + "step": 9099 + }, + { + "epoch": 0.07899236985790054, + "grad_norm": 0.7265625, + "learning_rate": 0.001967912918975562, + "loss": 0.1357, + "step": 9100 + }, + { + "epoch": 0.0790010503381047, + "grad_norm": 0.427734375, + "learning_rate": 0.0019679050422090113, + "loss": 0.1016, + "step": 9101 + }, + { + "epoch": 0.07900973081830887, + "grad_norm": 0.5625, + "learning_rate": 0.001967897164493333, + "loss": 0.1455, + "step": 9102 + }, + { + "epoch": 0.07901841129851303, + "grad_norm": 0.158203125, + "learning_rate": 0.0019678892858285358, + "loss": 0.1016, + "step": 9103 + }, + { + "epoch": 0.0790270917787172, + "grad_norm": 0.55078125, + "learning_rate": 0.001967881406214627, + "loss": 0.1143, + "step": 9104 + }, + { + "epoch": 0.07903577225892136, + "grad_norm": 0.10400390625, + "learning_rate": 0.001967873525651617, + "loss": 0.1377, + "step": 9105 + }, + { + "epoch": 0.07904445273912553, + "grad_norm": 0.197265625, + "learning_rate": 0.001967865644139513, + "loss": 0.0825, + "step": 9106 + }, + { + "epoch": 0.07905313321932969, + "grad_norm": 0.62109375, + "learning_rate": 0.001967857761678324, + "loss": 0.2158, + "step": 9107 + }, + { + "epoch": 0.07906181369953386, + "grad_norm": 0.3046875, + "learning_rate": 0.0019678498782680592, + "loss": 0.1123, + "step": 9108 + }, + { + "epoch": 0.07907049417973802, + "grad_norm": 0.349609375, + "learning_rate": 0.0019678419939087267, + "loss": 0.1113, + "step": 9109 + }, + { + "epoch": 0.07907917465994219, + "grad_norm": 0.181640625, + "learning_rate": 0.0019678341086003352, + "loss": 0.1484, + "step": 9110 + }, + { + "epoch": 0.07908785514014635, + "grad_norm": 0.26171875, + "learning_rate": 0.0019678262223428934, + "loss": 0.1406, + "step": 9111 + }, + { + "epoch": 0.07909653562035052, + "grad_norm": 1.796875, + "learning_rate": 0.00196781833513641, + "loss": 0.1357, + "step": 9112 + }, + { + "epoch": 0.07910521610055468, + "grad_norm": 0.4453125, + "learning_rate": 0.0019678104469808932, + "loss": 0.1143, + "step": 9113 + }, + { + "epoch": 0.07911389658075885, + "grad_norm": 0.609375, + "learning_rate": 0.001967802557876352, + "loss": 0.1191, + "step": 9114 + }, + { + "epoch": 0.07912257706096301, + "grad_norm": 0.369140625, + "learning_rate": 0.0019677946678227955, + "loss": 0.1079, + "step": 9115 + }, + { + "epoch": 0.07913125754116718, + "grad_norm": 0.76171875, + "learning_rate": 0.001967786776820231, + "loss": 0.1484, + "step": 9116 + }, + { + "epoch": 0.07913993802137134, + "grad_norm": 0.486328125, + "learning_rate": 0.0019677788848686682, + "loss": 0.123, + "step": 9117 + }, + { + "epoch": 0.07914861850157551, + "grad_norm": 0.1220703125, + "learning_rate": 0.001967770991968116, + "loss": 0.1235, + "step": 9118 + }, + { + "epoch": 0.07915729898177967, + "grad_norm": 0.1591796875, + "learning_rate": 0.001967763098118582, + "loss": 0.0972, + "step": 9119 + }, + { + "epoch": 0.07916597946198384, + "grad_norm": 0.1533203125, + "learning_rate": 0.001967755203320075, + "loss": 0.123, + "step": 9120 + }, + { + "epoch": 0.079174659942188, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019677473075726043, + "loss": 0.123, + "step": 9121 + }, + { + "epoch": 0.07918334042239217, + "grad_norm": 0.33984375, + "learning_rate": 0.001967739410876178, + "loss": 0.1768, + "step": 9122 + }, + { + "epoch": 0.07919202090259633, + "grad_norm": 0.328125, + "learning_rate": 0.0019677315132308054, + "loss": 0.1318, + "step": 9123 + }, + { + "epoch": 0.0792007013828005, + "grad_norm": 0.15625, + "learning_rate": 0.0019677236146364943, + "loss": 0.1533, + "step": 9124 + }, + { + "epoch": 0.07920938186300466, + "grad_norm": 0.203125, + "learning_rate": 0.001967715715093254, + "loss": 0.1758, + "step": 9125 + }, + { + "epoch": 0.07921806234320883, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019677078146010923, + "loss": 0.1074, + "step": 9126 + }, + { + "epoch": 0.079226742823413, + "grad_norm": 0.2177734375, + "learning_rate": 0.001967699913160019, + "loss": 0.1396, + "step": 9127 + }, + { + "epoch": 0.07923542330361716, + "grad_norm": 0.26171875, + "learning_rate": 0.0019676920107700416, + "loss": 0.0894, + "step": 9128 + }, + { + "epoch": 0.07924410378382132, + "grad_norm": 0.57421875, + "learning_rate": 0.001967684107431169, + "loss": 0.1045, + "step": 9129 + }, + { + "epoch": 0.07925278426402549, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019676762031434107, + "loss": 0.1367, + "step": 9130 + }, + { + "epoch": 0.07926146474422965, + "grad_norm": 0.41796875, + "learning_rate": 0.0019676682979067744, + "loss": 0.1357, + "step": 9131 + }, + { + "epoch": 0.07927014522443382, + "grad_norm": 0.46875, + "learning_rate": 0.0019676603917212696, + "loss": 0.1094, + "step": 9132 + }, + { + "epoch": 0.07927882570463798, + "grad_norm": 0.0888671875, + "learning_rate": 0.001967652484586904, + "loss": 0.1045, + "step": 9133 + }, + { + "epoch": 0.07928750618484215, + "grad_norm": 0.091796875, + "learning_rate": 0.0019676445765036865, + "loss": 0.1152, + "step": 9134 + }, + { + "epoch": 0.07929618666504631, + "grad_norm": 0.546875, + "learning_rate": 0.001967636667471626, + "loss": 0.1338, + "step": 9135 + }, + { + "epoch": 0.07930486714525048, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019676287574907315, + "loss": 0.1211, + "step": 9136 + }, + { + "epoch": 0.07931354762545464, + "grad_norm": 0.60546875, + "learning_rate": 0.0019676208465610107, + "loss": 0.1621, + "step": 9137 + }, + { + "epoch": 0.07932222810565881, + "grad_norm": 0.384765625, + "learning_rate": 0.001967612934682473, + "loss": 0.1523, + "step": 9138 + }, + { + "epoch": 0.07933090858586297, + "grad_norm": 0.427734375, + "learning_rate": 0.001967605021855127, + "loss": 0.1484, + "step": 9139 + }, + { + "epoch": 0.07933958906606714, + "grad_norm": 0.359375, + "learning_rate": 0.0019675971080789813, + "loss": 0.1279, + "step": 9140 + }, + { + "epoch": 0.0793482695462713, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019675891933540438, + "loss": 0.124, + "step": 9141 + }, + { + "epoch": 0.07935695002647547, + "grad_norm": 0.134765625, + "learning_rate": 0.0019675812776803244, + "loss": 0.1104, + "step": 9142 + }, + { + "epoch": 0.07936563050667964, + "grad_norm": 0.3203125, + "learning_rate": 0.001967573361057831, + "loss": 0.1108, + "step": 9143 + }, + { + "epoch": 0.0793743109868838, + "grad_norm": 0.59375, + "learning_rate": 0.0019675654434865722, + "loss": 0.1602, + "step": 9144 + }, + { + "epoch": 0.07938299146708797, + "grad_norm": 0.1416015625, + "learning_rate": 0.001967557524966557, + "loss": 0.1221, + "step": 9145 + }, + { + "epoch": 0.07939167194729213, + "grad_norm": 0.203125, + "learning_rate": 0.001967549605497794, + "loss": 0.1279, + "step": 9146 + }, + { + "epoch": 0.0794003524274963, + "grad_norm": 0.58203125, + "learning_rate": 0.0019675416850802917, + "loss": 0.1221, + "step": 9147 + }, + { + "epoch": 0.07940903290770046, + "grad_norm": 0.248046875, + "learning_rate": 0.001967533763714059, + "loss": 0.1348, + "step": 9148 + }, + { + "epoch": 0.07941771338790461, + "grad_norm": 0.279296875, + "learning_rate": 0.0019675258413991044, + "loss": 0.1162, + "step": 9149 + }, + { + "epoch": 0.07942639386810878, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019675179181354364, + "loss": 0.1494, + "step": 9150 + }, + { + "epoch": 0.07943507434831294, + "grad_norm": 0.08984375, + "learning_rate": 0.0019675099939230637, + "loss": 0.1289, + "step": 9151 + }, + { + "epoch": 0.07944375482851711, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019675020687619955, + "loss": 0.1025, + "step": 9152 + }, + { + "epoch": 0.07945243530872127, + "grad_norm": 1.0703125, + "learning_rate": 0.0019674941426522396, + "loss": 0.103, + "step": 9153 + }, + { + "epoch": 0.07946111578892544, + "grad_norm": 0.65625, + "learning_rate": 0.0019674862155938055, + "loss": 0.1069, + "step": 9154 + }, + { + "epoch": 0.0794697962691296, + "grad_norm": 0.32421875, + "learning_rate": 0.0019674782875867013, + "loss": 0.1221, + "step": 9155 + }, + { + "epoch": 0.07947847674933377, + "grad_norm": 0.38671875, + "learning_rate": 0.001967470358630936, + "loss": 0.127, + "step": 9156 + }, + { + "epoch": 0.07948715722953793, + "grad_norm": 0.2197265625, + "learning_rate": 0.001967462428726518, + "loss": 0.1602, + "step": 9157 + }, + { + "epoch": 0.0794958377097421, + "grad_norm": 0.0869140625, + "learning_rate": 0.001967454497873457, + "loss": 0.123, + "step": 9158 + }, + { + "epoch": 0.07950451818994626, + "grad_norm": 0.181640625, + "learning_rate": 0.0019674465660717596, + "loss": 0.1187, + "step": 9159 + }, + { + "epoch": 0.07951319867015043, + "grad_norm": 0.057861328125, + "learning_rate": 0.0019674386333214363, + "loss": 0.105, + "step": 9160 + }, + { + "epoch": 0.0795218791503546, + "grad_norm": 0.578125, + "learning_rate": 0.001967430699622495, + "loss": 0.1187, + "step": 9161 + }, + { + "epoch": 0.07953055963055876, + "grad_norm": 0.326171875, + "learning_rate": 0.0019674227649749444, + "loss": 0.1216, + "step": 9162 + }, + { + "epoch": 0.07953924011076292, + "grad_norm": 0.359375, + "learning_rate": 0.0019674148293787933, + "loss": 0.1465, + "step": 9163 + }, + { + "epoch": 0.07954792059096709, + "grad_norm": 0.4296875, + "learning_rate": 0.0019674068928340507, + "loss": 0.1348, + "step": 9164 + }, + { + "epoch": 0.07955660107117125, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019673989553407245, + "loss": 0.1377, + "step": 9165 + }, + { + "epoch": 0.07956528155137542, + "grad_norm": 0.32421875, + "learning_rate": 0.001967391016898824, + "loss": 0.1045, + "step": 9166 + }, + { + "epoch": 0.07957396203157958, + "grad_norm": 0.11376953125, + "learning_rate": 0.001967383077508358, + "loss": 0.1377, + "step": 9167 + }, + { + "epoch": 0.07958264251178375, + "grad_norm": 0.298828125, + "learning_rate": 0.0019673751371693344, + "loss": 0.0889, + "step": 9168 + }, + { + "epoch": 0.07959132299198791, + "grad_norm": 0.404296875, + "learning_rate": 0.001967367195881763, + "loss": 0.1152, + "step": 9169 + }, + { + "epoch": 0.07960000347219208, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019673592536456515, + "loss": 0.1328, + "step": 9170 + }, + { + "epoch": 0.07960868395239624, + "grad_norm": 0.0791015625, + "learning_rate": 0.001967351310461009, + "loss": 0.0747, + "step": 9171 + }, + { + "epoch": 0.07961736443260041, + "grad_norm": 0.2392578125, + "learning_rate": 0.001967343366327844, + "loss": 0.1221, + "step": 9172 + }, + { + "epoch": 0.07962604491280457, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019673354212461653, + "loss": 0.0918, + "step": 9173 + }, + { + "epoch": 0.07963472539300874, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019673274752159816, + "loss": 0.1416, + "step": 9174 + }, + { + "epoch": 0.0796434058732129, + "grad_norm": 1.3046875, + "learning_rate": 0.001967319528237302, + "loss": 0.5742, + "step": 9175 + }, + { + "epoch": 0.07965208635341707, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019673115803101344, + "loss": 0.1523, + "step": 9176 + }, + { + "epoch": 0.07966076683362124, + "grad_norm": 0.4453125, + "learning_rate": 0.001967303631434488, + "loss": 0.1226, + "step": 9177 + }, + { + "epoch": 0.0796694473138254, + "grad_norm": 0.6015625, + "learning_rate": 0.0019672956816103714, + "loss": 0.1143, + "step": 9178 + }, + { + "epoch": 0.07967812779402957, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019672877308377934, + "loss": 0.1221, + "step": 9179 + }, + { + "epoch": 0.07968680827423373, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019672797791167626, + "loss": 0.1025, + "step": 9180 + }, + { + "epoch": 0.0796954887544379, + "grad_norm": 0.44921875, + "learning_rate": 0.0019672718264472874, + "loss": 0.2031, + "step": 9181 + }, + { + "epoch": 0.07970416923464206, + "grad_norm": 0.0966796875, + "learning_rate": 0.001967263872829377, + "loss": 0.126, + "step": 9182 + }, + { + "epoch": 0.07971284971484623, + "grad_norm": 0.1953125, + "learning_rate": 0.00196725591826304, + "loss": 0.1074, + "step": 9183 + }, + { + "epoch": 0.07972153019505039, + "grad_norm": 0.6875, + "learning_rate": 0.0019672479627482844, + "loss": 0.1133, + "step": 9184 + }, + { + "epoch": 0.07973021067525456, + "grad_norm": 0.55859375, + "learning_rate": 0.00196724000628512, + "loss": 0.1299, + "step": 9185 + }, + { + "epoch": 0.07973889115545872, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019672320488735543, + "loss": 0.1367, + "step": 9186 + }, + { + "epoch": 0.07974757163566289, + "grad_norm": 0.42578125, + "learning_rate": 0.0019672240905135975, + "loss": 0.1338, + "step": 9187 + }, + { + "epoch": 0.07975625211586705, + "grad_norm": 0.2158203125, + "learning_rate": 0.001967216131205257, + "loss": 0.1113, + "step": 9188 + }, + { + "epoch": 0.07976493259607122, + "grad_norm": 0.49609375, + "learning_rate": 0.001967208170948542, + "loss": 0.1289, + "step": 9189 + }, + { + "epoch": 0.07977361307627538, + "grad_norm": 0.376953125, + "learning_rate": 0.001967200209743461, + "loss": 0.1553, + "step": 9190 + }, + { + "epoch": 0.07978229355647955, + "grad_norm": 0.306640625, + "learning_rate": 0.001967192247590023, + "loss": 0.1309, + "step": 9191 + }, + { + "epoch": 0.07979097403668371, + "grad_norm": 0.41796875, + "learning_rate": 0.0019671842844882366, + "loss": 0.1855, + "step": 9192 + }, + { + "epoch": 0.07979965451688788, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019671763204381107, + "loss": 0.1221, + "step": 9193 + }, + { + "epoch": 0.07980833499709204, + "grad_norm": 0.2119140625, + "learning_rate": 0.001967168355439654, + "loss": 0.123, + "step": 9194 + }, + { + "epoch": 0.07981701547729621, + "grad_norm": 0.11181640625, + "learning_rate": 0.001967160389492874, + "loss": 0.1426, + "step": 9195 + }, + { + "epoch": 0.07982569595750037, + "grad_norm": 0.57421875, + "learning_rate": 0.0019671524225977816, + "loss": 0.1309, + "step": 9196 + }, + { + "epoch": 0.07983437643770454, + "grad_norm": 0.16015625, + "learning_rate": 0.0019671444547543836, + "loss": 0.1104, + "step": 9197 + }, + { + "epoch": 0.0798430569179087, + "grad_norm": 0.67578125, + "learning_rate": 0.0019671364859626898, + "loss": 0.2012, + "step": 9198 + }, + { + "epoch": 0.07985173739811287, + "grad_norm": 0.5703125, + "learning_rate": 0.0019671285162227085, + "loss": 0.1167, + "step": 9199 + }, + { + "epoch": 0.07986041787831703, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019671205455344485, + "loss": 0.1226, + "step": 9200 + }, + { + "epoch": 0.0798690983585212, + "grad_norm": 0.091796875, + "learning_rate": 0.0019671125738979184, + "loss": 0.1177, + "step": 9201 + }, + { + "epoch": 0.07987777883872536, + "grad_norm": 0.6328125, + "learning_rate": 0.0019671046013131272, + "loss": 0.168, + "step": 9202 + }, + { + "epoch": 0.07988645931892953, + "grad_norm": 0.2177734375, + "learning_rate": 0.001967096627780083, + "loss": 0.1309, + "step": 9203 + }, + { + "epoch": 0.0798951397991337, + "grad_norm": 0.20703125, + "learning_rate": 0.0019670886532987954, + "loss": 0.1289, + "step": 9204 + }, + { + "epoch": 0.07990382027933786, + "grad_norm": 0.197265625, + "learning_rate": 0.0019670806778692724, + "loss": 0.1416, + "step": 9205 + }, + { + "epoch": 0.07991250075954202, + "grad_norm": 0.59375, + "learning_rate": 0.001967072701491523, + "loss": 0.1738, + "step": 9206 + }, + { + "epoch": 0.07992118123974619, + "grad_norm": 0.359375, + "learning_rate": 0.0019670647241655562, + "loss": 0.1328, + "step": 9207 + }, + { + "epoch": 0.07992986171995035, + "grad_norm": 0.203125, + "learning_rate": 0.0019670567458913804, + "loss": 0.1475, + "step": 9208 + }, + { + "epoch": 0.07993854220015452, + "grad_norm": 0.107421875, + "learning_rate": 0.001967048766669004, + "loss": 0.1113, + "step": 9209 + }, + { + "epoch": 0.07994722268035867, + "grad_norm": 0.63671875, + "learning_rate": 0.001967040786498437, + "loss": 0.124, + "step": 9210 + }, + { + "epoch": 0.07995590316056284, + "grad_norm": 0.48046875, + "learning_rate": 0.001967032805379686, + "loss": 0.1357, + "step": 9211 + }, + { + "epoch": 0.079964583640767, + "grad_norm": 0.6015625, + "learning_rate": 0.001967024823312762, + "loss": 0.1055, + "step": 9212 + }, + { + "epoch": 0.07997326412097117, + "grad_norm": 0.2392578125, + "learning_rate": 0.001967016840297672, + "loss": 0.0977, + "step": 9213 + }, + { + "epoch": 0.07998194460117533, + "grad_norm": 0.205078125, + "learning_rate": 0.0019670088563344254, + "loss": 0.1045, + "step": 9214 + }, + { + "epoch": 0.0799906250813795, + "grad_norm": 0.146484375, + "learning_rate": 0.0019670008714230314, + "loss": 0.1572, + "step": 9215 + }, + { + "epoch": 0.07999930556158366, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019669928855634983, + "loss": 0.0889, + "step": 9216 + }, + { + "epoch": 0.08000798604178783, + "grad_norm": 0.73046875, + "learning_rate": 0.0019669848987558343, + "loss": 0.1367, + "step": 9217 + }, + { + "epoch": 0.08001666652199199, + "grad_norm": 0.44140625, + "learning_rate": 0.001966976911000049, + "loss": 0.1699, + "step": 9218 + }, + { + "epoch": 0.08002534700219616, + "grad_norm": 0.46484375, + "learning_rate": 0.0019669689222961504, + "loss": 0.1221, + "step": 9219 + }, + { + "epoch": 0.08003402748240032, + "grad_norm": 0.26953125, + "learning_rate": 0.0019669609326441483, + "loss": 0.1592, + "step": 9220 + }, + { + "epoch": 0.08004270796260449, + "grad_norm": 0.5546875, + "learning_rate": 0.00196695294204405, + "loss": 0.124, + "step": 9221 + }, + { + "epoch": 0.08005138844280865, + "grad_norm": 0.138671875, + "learning_rate": 0.0019669449504958656, + "loss": 0.127, + "step": 9222 + }, + { + "epoch": 0.08006006892301282, + "grad_norm": 0.357421875, + "learning_rate": 0.0019669369579996027, + "loss": 0.1523, + "step": 9223 + }, + { + "epoch": 0.08006874940321698, + "grad_norm": 0.2451171875, + "learning_rate": 0.001966928964555271, + "loss": 0.1514, + "step": 9224 + }, + { + "epoch": 0.08007742988342115, + "grad_norm": 0.376953125, + "learning_rate": 0.0019669209701628785, + "loss": 0.1143, + "step": 9225 + }, + { + "epoch": 0.08008611036362531, + "grad_norm": 0.58203125, + "learning_rate": 0.0019669129748224345, + "loss": 0.1758, + "step": 9226 + }, + { + "epoch": 0.08009479084382948, + "grad_norm": 0.18359375, + "learning_rate": 0.0019669049785339472, + "loss": 0.0938, + "step": 9227 + }, + { + "epoch": 0.08010347132403364, + "grad_norm": 0.201171875, + "learning_rate": 0.001966896981297426, + "loss": 0.1699, + "step": 9228 + }, + { + "epoch": 0.08011215180423781, + "grad_norm": 0.146484375, + "learning_rate": 0.001966888983112879, + "loss": 0.1309, + "step": 9229 + }, + { + "epoch": 0.08012083228444197, + "grad_norm": 0.10107421875, + "learning_rate": 0.001966880983980315, + "loss": 0.1406, + "step": 9230 + }, + { + "epoch": 0.08012951276464614, + "grad_norm": 0.59375, + "learning_rate": 0.0019668729838997435, + "loss": 0.126, + "step": 9231 + }, + { + "epoch": 0.0801381932448503, + "grad_norm": 0.2265625, + "learning_rate": 0.0019668649828711725, + "loss": 0.0986, + "step": 9232 + }, + { + "epoch": 0.08014687372505447, + "grad_norm": 0.271484375, + "learning_rate": 0.001966856980894611, + "loss": 0.0967, + "step": 9233 + }, + { + "epoch": 0.08015555420525863, + "grad_norm": 0.1884765625, + "learning_rate": 0.001966848977970068, + "loss": 0.0898, + "step": 9234 + }, + { + "epoch": 0.0801642346854628, + "grad_norm": 0.15234375, + "learning_rate": 0.001966840974097552, + "loss": 0.1543, + "step": 9235 + }, + { + "epoch": 0.08017291516566696, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019668329692770713, + "loss": 0.1099, + "step": 9236 + }, + { + "epoch": 0.08018159564587113, + "grad_norm": 0.56640625, + "learning_rate": 0.001966824963508635, + "loss": 0.1494, + "step": 9237 + }, + { + "epoch": 0.0801902761260753, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019668169567922523, + "loss": 0.1641, + "step": 9238 + }, + { + "epoch": 0.08019895660627946, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019668089491279316, + "loss": 0.1367, + "step": 9239 + }, + { + "epoch": 0.08020763708648362, + "grad_norm": 0.244140625, + "learning_rate": 0.0019668009405156813, + "loss": 0.1211, + "step": 9240 + }, + { + "epoch": 0.08021631756668779, + "grad_norm": 0.330078125, + "learning_rate": 0.0019667929309555112, + "loss": 0.1338, + "step": 9241 + }, + { + "epoch": 0.08022499804689195, + "grad_norm": 0.33203125, + "learning_rate": 0.0019667849204474284, + "loss": 0.1436, + "step": 9242 + }, + { + "epoch": 0.08023367852709612, + "grad_norm": 0.244140625, + "learning_rate": 0.0019667769089914437, + "loss": 0.1475, + "step": 9243 + }, + { + "epoch": 0.08024235900730028, + "grad_norm": 0.470703125, + "learning_rate": 0.001966768896587564, + "loss": 0.1172, + "step": 9244 + }, + { + "epoch": 0.08025103948750445, + "grad_norm": 0.9140625, + "learning_rate": 0.0019667608832357993, + "loss": 0.1226, + "step": 9245 + }, + { + "epoch": 0.08025971996770861, + "grad_norm": 0.498046875, + "learning_rate": 0.0019667528689361574, + "loss": 0.1221, + "step": 9246 + }, + { + "epoch": 0.08026840044791278, + "grad_norm": 0.7890625, + "learning_rate": 0.0019667448536886483, + "loss": 0.1025, + "step": 9247 + }, + { + "epoch": 0.08027708092811695, + "grad_norm": 0.376953125, + "learning_rate": 0.0019667368374932793, + "loss": 0.1436, + "step": 9248 + }, + { + "epoch": 0.08028576140832111, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019667288203500605, + "loss": 0.1279, + "step": 9249 + }, + { + "epoch": 0.08029444188852528, + "grad_norm": 0.353515625, + "learning_rate": 0.001966720802259, + "loss": 0.1553, + "step": 9250 + }, + { + "epoch": 0.08030312236872944, + "grad_norm": 0.078125, + "learning_rate": 0.0019667127832201066, + "loss": 0.0781, + "step": 9251 + }, + { + "epoch": 0.0803118028489336, + "grad_norm": 0.244140625, + "learning_rate": 0.001966704763233389, + "loss": 0.1328, + "step": 9252 + }, + { + "epoch": 0.08032048332913777, + "grad_norm": 0.3984375, + "learning_rate": 0.001966696742298856, + "loss": 0.1377, + "step": 9253 + }, + { + "epoch": 0.08032916380934194, + "grad_norm": 0.361328125, + "learning_rate": 0.001966688720416517, + "loss": 0.1621, + "step": 9254 + }, + { + "epoch": 0.0803378442895461, + "grad_norm": 0.26953125, + "learning_rate": 0.0019666806975863795, + "loss": 0.1465, + "step": 9255 + }, + { + "epoch": 0.08034652476975027, + "grad_norm": 0.2890625, + "learning_rate": 0.0019666726738084535, + "loss": 0.1279, + "step": 9256 + }, + { + "epoch": 0.08035520524995443, + "grad_norm": 0.24609375, + "learning_rate": 0.001966664649082747, + "loss": 0.1348, + "step": 9257 + }, + { + "epoch": 0.0803638857301586, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019666566234092693, + "loss": 0.1128, + "step": 9258 + }, + { + "epoch": 0.08037256621036276, + "grad_norm": 0.72265625, + "learning_rate": 0.001966648596788029, + "loss": 0.1055, + "step": 9259 + }, + { + "epoch": 0.08038124669056693, + "grad_norm": 1.1171875, + "learning_rate": 0.0019666405692190347, + "loss": 0.1523, + "step": 9260 + }, + { + "epoch": 0.08038992717077109, + "grad_norm": 0.458984375, + "learning_rate": 0.0019666325407022953, + "loss": 0.1416, + "step": 9261 + }, + { + "epoch": 0.08039860765097526, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019666245112378196, + "loss": 0.1279, + "step": 9262 + }, + { + "epoch": 0.08040728813117942, + "grad_norm": 0.251953125, + "learning_rate": 0.0019666164808256163, + "loss": 0.1328, + "step": 9263 + }, + { + "epoch": 0.08041596861138359, + "grad_norm": 1.03125, + "learning_rate": 0.001966608449465694, + "loss": 0.1523, + "step": 9264 + }, + { + "epoch": 0.08042464909158775, + "grad_norm": 0.640625, + "learning_rate": 0.001966600417158062, + "loss": 0.1196, + "step": 9265 + }, + { + "epoch": 0.08043332957179192, + "grad_norm": 0.07568359375, + "learning_rate": 0.001966592383902729, + "loss": 0.123, + "step": 9266 + }, + { + "epoch": 0.08044201005199608, + "grad_norm": 0.330078125, + "learning_rate": 0.001966584349699703, + "loss": 0.1201, + "step": 9267 + }, + { + "epoch": 0.08045069053220025, + "grad_norm": 0.1640625, + "learning_rate": 0.001966576314548994, + "loss": 0.1016, + "step": 9268 + }, + { + "epoch": 0.08045937101240441, + "grad_norm": 0.39453125, + "learning_rate": 0.00196656827845061, + "loss": 0.1367, + "step": 9269 + }, + { + "epoch": 0.08046805149260858, + "grad_norm": 0.64453125, + "learning_rate": 0.00196656024140456, + "loss": 0.1348, + "step": 9270 + }, + { + "epoch": 0.08047673197281274, + "grad_norm": 0.1171875, + "learning_rate": 0.0019665522034108523, + "loss": 0.1318, + "step": 9271 + }, + { + "epoch": 0.0804854124530169, + "grad_norm": 0.353515625, + "learning_rate": 0.0019665441644694964, + "loss": 0.124, + "step": 9272 + }, + { + "epoch": 0.08049409293322106, + "grad_norm": 1.140625, + "learning_rate": 0.001966536124580501, + "loss": 0.1738, + "step": 9273 + }, + { + "epoch": 0.08050277341342522, + "grad_norm": 0.90234375, + "learning_rate": 0.001966528083743874, + "loss": 0.1055, + "step": 9274 + }, + { + "epoch": 0.08051145389362939, + "grad_norm": 0.19921875, + "learning_rate": 0.0019665200419596257, + "loss": 0.1113, + "step": 9275 + }, + { + "epoch": 0.08052013437383355, + "grad_norm": 1.109375, + "learning_rate": 0.001966511999227764, + "loss": 0.124, + "step": 9276 + }, + { + "epoch": 0.08052881485403772, + "grad_norm": 0.5234375, + "learning_rate": 0.0019665039555482977, + "loss": 0.0991, + "step": 9277 + }, + { + "epoch": 0.08053749533424188, + "grad_norm": 0.42578125, + "learning_rate": 0.0019664959109212355, + "loss": 0.1836, + "step": 9278 + }, + { + "epoch": 0.08054617581444605, + "grad_norm": 0.68359375, + "learning_rate": 0.0019664878653465868, + "loss": 0.1172, + "step": 9279 + }, + { + "epoch": 0.08055485629465022, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019664798188243596, + "loss": 0.1011, + "step": 9280 + }, + { + "epoch": 0.08056353677485438, + "grad_norm": 0.0732421875, + "learning_rate": 0.001966471771354563, + "loss": 0.1055, + "step": 9281 + }, + { + "epoch": 0.08057221725505855, + "grad_norm": 0.330078125, + "learning_rate": 0.001966463722937206, + "loss": 0.1162, + "step": 9282 + }, + { + "epoch": 0.08058089773526271, + "grad_norm": 0.119140625, + "learning_rate": 0.0019664556735722975, + "loss": 0.1445, + "step": 9283 + }, + { + "epoch": 0.08058957821546688, + "grad_norm": 0.4453125, + "learning_rate": 0.001966447623259846, + "loss": 0.1094, + "step": 9284 + }, + { + "epoch": 0.08059825869567104, + "grad_norm": 0.54296875, + "learning_rate": 0.0019664395719998606, + "loss": 0.1367, + "step": 9285 + }, + { + "epoch": 0.0806069391758752, + "grad_norm": 0.50390625, + "learning_rate": 0.0019664315197923496, + "loss": 0.0728, + "step": 9286 + }, + { + "epoch": 0.08061561965607937, + "grad_norm": 0.236328125, + "learning_rate": 0.0019664234666373222, + "loss": 0.0918, + "step": 9287 + }, + { + "epoch": 0.08062430013628354, + "grad_norm": 0.07080078125, + "learning_rate": 0.001966415412534787, + "loss": 0.103, + "step": 9288 + }, + { + "epoch": 0.0806329806164877, + "grad_norm": 0.310546875, + "learning_rate": 0.001966407357484753, + "loss": 0.1641, + "step": 9289 + }, + { + "epoch": 0.08064166109669187, + "grad_norm": 0.60546875, + "learning_rate": 0.001966399301487229, + "loss": 0.1133, + "step": 9290 + }, + { + "epoch": 0.08065034157689603, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019663912445422238, + "loss": 0.1211, + "step": 9291 + }, + { + "epoch": 0.0806590220571002, + "grad_norm": 0.482421875, + "learning_rate": 0.001966383186649746, + "loss": 0.125, + "step": 9292 + }, + { + "epoch": 0.08066770253730436, + "grad_norm": 1.15625, + "learning_rate": 0.0019663751278098046, + "loss": 0.1533, + "step": 9293 + }, + { + "epoch": 0.08067638301750853, + "grad_norm": 0.111328125, + "learning_rate": 0.0019663670680224086, + "loss": 0.0996, + "step": 9294 + }, + { + "epoch": 0.08068506349771269, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019663590072875664, + "loss": 0.1445, + "step": 9295 + }, + { + "epoch": 0.08069374397791686, + "grad_norm": 0.796875, + "learning_rate": 0.001966350945605287, + "loss": 0.1348, + "step": 9296 + }, + { + "epoch": 0.08070242445812102, + "grad_norm": 0.2470703125, + "learning_rate": 0.001966342882975579, + "loss": 0.127, + "step": 9297 + }, + { + "epoch": 0.08071110493832519, + "grad_norm": 0.5, + "learning_rate": 0.001966334819398452, + "loss": 0.1074, + "step": 9298 + }, + { + "epoch": 0.08071978541852935, + "grad_norm": 0.64453125, + "learning_rate": 0.001966326754873914, + "loss": 0.1206, + "step": 9299 + }, + { + "epoch": 0.08072846589873352, + "grad_norm": 0.447265625, + "learning_rate": 0.001966318689401974, + "loss": 0.1074, + "step": 9300 + }, + { + "epoch": 0.08073714637893768, + "grad_norm": 0.55078125, + "learning_rate": 0.0019663106229826405, + "loss": 0.1348, + "step": 9301 + }, + { + "epoch": 0.08074582685914185, + "grad_norm": 0.1318359375, + "learning_rate": 0.001966302555615923, + "loss": 0.1074, + "step": 9302 + }, + { + "epoch": 0.08075450733934601, + "grad_norm": 0.30078125, + "learning_rate": 0.0019662944873018303, + "loss": 0.125, + "step": 9303 + }, + { + "epoch": 0.08076318781955018, + "grad_norm": 0.3046875, + "learning_rate": 0.001966286418040371, + "loss": 0.1104, + "step": 9304 + }, + { + "epoch": 0.08077186829975434, + "grad_norm": 0.12109375, + "learning_rate": 0.0019662783478315536, + "loss": 0.1299, + "step": 9305 + }, + { + "epoch": 0.08078054877995851, + "grad_norm": 0.19921875, + "learning_rate": 0.0019662702766753875, + "loss": 0.1465, + "step": 9306 + }, + { + "epoch": 0.08078922926016267, + "grad_norm": 0.234375, + "learning_rate": 0.001966262204571881, + "loss": 0.0957, + "step": 9307 + }, + { + "epoch": 0.08079790974036684, + "grad_norm": 0.4375, + "learning_rate": 0.0019662541315210434, + "loss": 0.125, + "step": 9308 + }, + { + "epoch": 0.080806590220571, + "grad_norm": 0.26953125, + "learning_rate": 0.0019662460575228828, + "loss": 0.1152, + "step": 9309 + }, + { + "epoch": 0.08081527070077517, + "grad_norm": 0.13671875, + "learning_rate": 0.001966237982577409, + "loss": 0.1826, + "step": 9310 + }, + { + "epoch": 0.08082395118097933, + "grad_norm": 0.55078125, + "learning_rate": 0.0019662299066846305, + "loss": 0.1016, + "step": 9311 + }, + { + "epoch": 0.0808326316611835, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019662218298445553, + "loss": 0.1484, + "step": 9312 + }, + { + "epoch": 0.08084131214138766, + "grad_norm": 0.4140625, + "learning_rate": 0.0019662137520571935, + "loss": 0.1177, + "step": 9313 + }, + { + "epoch": 0.08084999262159183, + "grad_norm": 0.69921875, + "learning_rate": 0.001966205673322553, + "loss": 0.123, + "step": 9314 + }, + { + "epoch": 0.080858673101796, + "grad_norm": 0.388671875, + "learning_rate": 0.0019661975936406434, + "loss": 0.1006, + "step": 9315 + }, + { + "epoch": 0.08086735358200016, + "grad_norm": 0.63671875, + "learning_rate": 0.001966189513011473, + "loss": 0.1748, + "step": 9316 + }, + { + "epoch": 0.08087603406220432, + "grad_norm": 0.87109375, + "learning_rate": 0.0019661814314350505, + "loss": 0.1201, + "step": 9317 + }, + { + "epoch": 0.08088471454240849, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019661733489113853, + "loss": 0.1108, + "step": 9318 + }, + { + "epoch": 0.08089339502261265, + "grad_norm": 0.33984375, + "learning_rate": 0.0019661652654404858, + "loss": 0.168, + "step": 9319 + }, + { + "epoch": 0.08090207550281682, + "grad_norm": 0.99609375, + "learning_rate": 0.001966157181022361, + "loss": 0.1123, + "step": 9320 + }, + { + "epoch": 0.08091075598302098, + "grad_norm": 0.11328125, + "learning_rate": 0.0019661490956570196, + "loss": 0.0869, + "step": 9321 + }, + { + "epoch": 0.08091943646322515, + "grad_norm": 0.3671875, + "learning_rate": 0.0019661410093444708, + "loss": 0.1147, + "step": 9322 + }, + { + "epoch": 0.08092811694342932, + "grad_norm": 0.345703125, + "learning_rate": 0.001966132922084723, + "loss": 0.1328, + "step": 9323 + }, + { + "epoch": 0.08093679742363348, + "grad_norm": 0.06494140625, + "learning_rate": 0.001966124833877785, + "loss": 0.1172, + "step": 9324 + }, + { + "epoch": 0.08094547790383765, + "grad_norm": 0.578125, + "learning_rate": 0.0019661167447236664, + "loss": 0.1099, + "step": 9325 + }, + { + "epoch": 0.08095415838404181, + "grad_norm": 0.466796875, + "learning_rate": 0.0019661086546223752, + "loss": 0.1006, + "step": 9326 + }, + { + "epoch": 0.08096283886424598, + "grad_norm": 0.23046875, + "learning_rate": 0.0019661005635739207, + "loss": 0.1758, + "step": 9327 + }, + { + "epoch": 0.08097151934445014, + "grad_norm": 0.69921875, + "learning_rate": 0.001966092471578312, + "loss": 0.1738, + "step": 9328 + }, + { + "epoch": 0.0809801998246543, + "grad_norm": 0.134765625, + "learning_rate": 0.0019660843786355573, + "loss": 0.1426, + "step": 9329 + }, + { + "epoch": 0.08098888030485847, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019660762847456656, + "loss": 0.1152, + "step": 9330 + }, + { + "epoch": 0.08099756078506264, + "grad_norm": 0.158203125, + "learning_rate": 0.0019660681899086457, + "loss": 0.1514, + "step": 9331 + }, + { + "epoch": 0.0810062412652668, + "grad_norm": 0.09326171875, + "learning_rate": 0.001966060094124507, + "loss": 0.0928, + "step": 9332 + }, + { + "epoch": 0.08101492174547097, + "grad_norm": 0.1708984375, + "learning_rate": 0.001966051997393258, + "loss": 0.126, + "step": 9333 + }, + { + "epoch": 0.08102360222567512, + "grad_norm": 0.5625, + "learning_rate": 0.0019660438997149074, + "loss": 0.1631, + "step": 9334 + }, + { + "epoch": 0.08103228270587928, + "grad_norm": 0.451171875, + "learning_rate": 0.0019660358010894644, + "loss": 0.1387, + "step": 9335 + }, + { + "epoch": 0.08104096318608345, + "grad_norm": 0.0927734375, + "learning_rate": 0.001966027701516938, + "loss": 0.1367, + "step": 9336 + }, + { + "epoch": 0.08104964366628761, + "grad_norm": 0.8515625, + "learning_rate": 0.001966019600997336, + "loss": 0.1172, + "step": 9337 + }, + { + "epoch": 0.08105832414649178, + "grad_norm": 0.359375, + "learning_rate": 0.0019660114995306683, + "loss": 0.1338, + "step": 9338 + }, + { + "epoch": 0.08106700462669594, + "grad_norm": 0.37109375, + "learning_rate": 0.0019660033971169435, + "loss": 0.1196, + "step": 9339 + }, + { + "epoch": 0.08107568510690011, + "grad_norm": 0.671875, + "learning_rate": 0.00196599529375617, + "loss": 0.1318, + "step": 9340 + }, + { + "epoch": 0.08108436558710427, + "grad_norm": 0.5390625, + "learning_rate": 0.0019659871894483574, + "loss": 0.1177, + "step": 9341 + }, + { + "epoch": 0.08109304606730844, + "grad_norm": 0.5625, + "learning_rate": 0.001965979084193514, + "loss": 0.1309, + "step": 9342 + }, + { + "epoch": 0.0811017265475126, + "grad_norm": 0.4453125, + "learning_rate": 0.001965970977991649, + "loss": 0.1572, + "step": 9343 + }, + { + "epoch": 0.08111040702771677, + "grad_norm": 0.94921875, + "learning_rate": 0.001965962870842771, + "loss": 0.1387, + "step": 9344 + }, + { + "epoch": 0.08111908750792093, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019659547627468897, + "loss": 0.1445, + "step": 9345 + }, + { + "epoch": 0.0811277679881251, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019659466537040125, + "loss": 0.1416, + "step": 9346 + }, + { + "epoch": 0.08113644846832926, + "grad_norm": 0.349609375, + "learning_rate": 0.0019659385437141496, + "loss": 0.1099, + "step": 9347 + }, + { + "epoch": 0.08114512894853343, + "grad_norm": 0.115234375, + "learning_rate": 0.001965930432777309, + "loss": 0.1289, + "step": 9348 + }, + { + "epoch": 0.0811538094287376, + "grad_norm": 0.69921875, + "learning_rate": 0.0019659223208935, + "loss": 0.1055, + "step": 9349 + }, + { + "epoch": 0.08116248990894176, + "grad_norm": 0.265625, + "learning_rate": 0.001965914208062731, + "loss": 0.1406, + "step": 9350 + }, + { + "epoch": 0.08117117038914592, + "grad_norm": 0.3828125, + "learning_rate": 0.0019659060942850116, + "loss": 0.1045, + "step": 9351 + }, + { + "epoch": 0.08117985086935009, + "grad_norm": 0.71875, + "learning_rate": 0.0019658979795603502, + "loss": 0.1021, + "step": 9352 + }, + { + "epoch": 0.08118853134955425, + "grad_norm": 0.310546875, + "learning_rate": 0.001965889863888756, + "loss": 0.0908, + "step": 9353 + }, + { + "epoch": 0.08119721182975842, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019658817472702375, + "loss": 0.168, + "step": 9354 + }, + { + "epoch": 0.08120589230996259, + "grad_norm": 0.263671875, + "learning_rate": 0.001965873629704804, + "loss": 0.1089, + "step": 9355 + }, + { + "epoch": 0.08121457279016675, + "grad_norm": 0.388671875, + "learning_rate": 0.0019658655111924636, + "loss": 0.1436, + "step": 9356 + }, + { + "epoch": 0.08122325327037092, + "grad_norm": 0.298828125, + "learning_rate": 0.001965857391733226, + "loss": 0.1006, + "step": 9357 + }, + { + "epoch": 0.08123193375057508, + "grad_norm": 0.470703125, + "learning_rate": 0.0019658492713270992, + "loss": 0.1079, + "step": 9358 + }, + { + "epoch": 0.08124061423077925, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019658411499740933, + "loss": 0.1865, + "step": 9359 + }, + { + "epoch": 0.08124929471098341, + "grad_norm": 0.203125, + "learning_rate": 0.0019658330276742164, + "loss": 0.1006, + "step": 9360 + }, + { + "epoch": 0.08125797519118758, + "grad_norm": 0.361328125, + "learning_rate": 0.001965824904427477, + "loss": 0.126, + "step": 9361 + }, + { + "epoch": 0.08126665567139174, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019658167802338853, + "loss": 0.127, + "step": 9362 + }, + { + "epoch": 0.0812753361515959, + "grad_norm": 0.875, + "learning_rate": 0.001965808655093449, + "loss": 0.1094, + "step": 9363 + }, + { + "epoch": 0.08128401663180007, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019658005290061776, + "loss": 0.1426, + "step": 9364 + }, + { + "epoch": 0.08129269711200424, + "grad_norm": 0.162109375, + "learning_rate": 0.0019657924019720794, + "loss": 0.0986, + "step": 9365 + }, + { + "epoch": 0.0813013775922084, + "grad_norm": 0.302734375, + "learning_rate": 0.0019657842739911635, + "loss": 0.1079, + "step": 9366 + }, + { + "epoch": 0.08131005807241257, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019657761450634394, + "loss": 0.1069, + "step": 9367 + }, + { + "epoch": 0.08131873855261673, + "grad_norm": 0.7109375, + "learning_rate": 0.0019657680151889153, + "loss": 0.1387, + "step": 9368 + }, + { + "epoch": 0.0813274190328209, + "grad_norm": 0.310546875, + "learning_rate": 0.0019657598843676005, + "loss": 0.1025, + "step": 9369 + }, + { + "epoch": 0.08133609951302506, + "grad_norm": 0.416015625, + "learning_rate": 0.0019657517525995035, + "loss": 0.1289, + "step": 9370 + }, + { + "epoch": 0.08134477999322923, + "grad_norm": 0.2373046875, + "learning_rate": 0.001965743619884633, + "loss": 0.2393, + "step": 9371 + }, + { + "epoch": 0.08135346047343339, + "grad_norm": 0.07421875, + "learning_rate": 0.0019657354862229986, + "loss": 0.0981, + "step": 9372 + }, + { + "epoch": 0.08136214095363756, + "grad_norm": 0.7109375, + "learning_rate": 0.0019657273516146094, + "loss": 0.1367, + "step": 9373 + }, + { + "epoch": 0.08137082143384172, + "grad_norm": 0.1474609375, + "learning_rate": 0.001965719216059473, + "loss": 0.0889, + "step": 9374 + }, + { + "epoch": 0.08137950191404589, + "grad_norm": 0.353515625, + "learning_rate": 0.0019657110795576, + "loss": 0.1074, + "step": 9375 + }, + { + "epoch": 0.08138818239425005, + "grad_norm": 0.79296875, + "learning_rate": 0.0019657029421089977, + "loss": 0.1426, + "step": 9376 + }, + { + "epoch": 0.08139686287445422, + "grad_norm": 0.5234375, + "learning_rate": 0.0019656948037136755, + "loss": 0.1699, + "step": 9377 + }, + { + "epoch": 0.08140554335465838, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019656866643716427, + "loss": 0.127, + "step": 9378 + }, + { + "epoch": 0.08141422383486255, + "grad_norm": 0.412109375, + "learning_rate": 0.001965678524082908, + "loss": 0.1025, + "step": 9379 + }, + { + "epoch": 0.08142290431506671, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019656703828474804, + "loss": 0.1226, + "step": 9380 + }, + { + "epoch": 0.08143158479527088, + "grad_norm": 0.26171875, + "learning_rate": 0.0019656622406653687, + "loss": 0.1064, + "step": 9381 + }, + { + "epoch": 0.08144026527547504, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019656540975365815, + "loss": 0.1318, + "step": 9382 + }, + { + "epoch": 0.08144894575567921, + "grad_norm": 0.20703125, + "learning_rate": 0.001965645953461128, + "loss": 0.1504, + "step": 9383 + }, + { + "epoch": 0.08145762623588337, + "grad_norm": 0.8046875, + "learning_rate": 0.0019656378084390172, + "loss": 0.1934, + "step": 9384 + }, + { + "epoch": 0.08146630671608754, + "grad_norm": 0.478515625, + "learning_rate": 0.001965629662470258, + "loss": 0.1084, + "step": 9385 + }, + { + "epoch": 0.0814749871962917, + "grad_norm": 0.29296875, + "learning_rate": 0.001965621515554859, + "loss": 0.0938, + "step": 9386 + }, + { + "epoch": 0.08148366767649587, + "grad_norm": 0.201171875, + "learning_rate": 0.0019656133676928295, + "loss": 0.1167, + "step": 9387 + }, + { + "epoch": 0.08149234815670003, + "grad_norm": 0.142578125, + "learning_rate": 0.001965605218884178, + "loss": 0.1104, + "step": 9388 + }, + { + "epoch": 0.0815010286369042, + "grad_norm": 0.337890625, + "learning_rate": 0.001965597069128914, + "loss": 0.1182, + "step": 9389 + }, + { + "epoch": 0.08150970911710836, + "grad_norm": 1.234375, + "learning_rate": 0.0019655889184270457, + "loss": 0.1562, + "step": 9390 + }, + { + "epoch": 0.08151838959731253, + "grad_norm": 0.349609375, + "learning_rate": 0.0019655807667785827, + "loss": 0.1089, + "step": 9391 + }, + { + "epoch": 0.0815270700775167, + "grad_norm": 0.193359375, + "learning_rate": 0.001965572614183533, + "loss": 0.1367, + "step": 9392 + }, + { + "epoch": 0.08153575055772086, + "grad_norm": 0.4765625, + "learning_rate": 0.001965564460641907, + "loss": 0.1953, + "step": 9393 + }, + { + "epoch": 0.08154443103792502, + "grad_norm": 0.3984375, + "learning_rate": 0.001965556306153712, + "loss": 0.1123, + "step": 9394 + }, + { + "epoch": 0.08155311151812918, + "grad_norm": 0.3828125, + "learning_rate": 0.001965548150718958, + "loss": 0.1055, + "step": 9395 + }, + { + "epoch": 0.08156179199833334, + "grad_norm": 0.263671875, + "learning_rate": 0.001965539994337653, + "loss": 0.1299, + "step": 9396 + }, + { + "epoch": 0.0815704724785375, + "grad_norm": 0.30078125, + "learning_rate": 0.001965531837009807, + "loss": 0.1196, + "step": 9397 + }, + { + "epoch": 0.08157915295874167, + "grad_norm": 0.443359375, + "learning_rate": 0.0019655236787354284, + "loss": 0.1152, + "step": 9398 + }, + { + "epoch": 0.08158783343894584, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019655155195145257, + "loss": 0.1191, + "step": 9399 + }, + { + "epoch": 0.08159651391915, + "grad_norm": 0.53515625, + "learning_rate": 0.0019655073593471085, + "loss": 0.1533, + "step": 9400 + }, + { + "epoch": 0.08160519439935417, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019654991982331854, + "loss": 0.1133, + "step": 9401 + }, + { + "epoch": 0.08161387487955833, + "grad_norm": 0.546875, + "learning_rate": 0.0019654910361727655, + "loss": 0.127, + "step": 9402 + }, + { + "epoch": 0.0816225553597625, + "grad_norm": 0.4140625, + "learning_rate": 0.0019654828731658575, + "loss": 0.1973, + "step": 9403 + }, + { + "epoch": 0.08163123583996666, + "grad_norm": 0.53125, + "learning_rate": 0.0019654747092124705, + "loss": 0.1162, + "step": 9404 + }, + { + "epoch": 0.08163991632017083, + "grad_norm": 0.17578125, + "learning_rate": 0.001965466544312613, + "loss": 0.1177, + "step": 9405 + }, + { + "epoch": 0.08164859680037499, + "grad_norm": 0.74609375, + "learning_rate": 0.0019654583784662946, + "loss": 0.1182, + "step": 9406 + }, + { + "epoch": 0.08165727728057916, + "grad_norm": 1.6640625, + "learning_rate": 0.001965450211673524, + "loss": 0.1582, + "step": 9407 + }, + { + "epoch": 0.08166595776078332, + "grad_norm": 0.396484375, + "learning_rate": 0.00196544204393431, + "loss": 0.1133, + "step": 9408 + }, + { + "epoch": 0.08167463824098749, + "grad_norm": 0.314453125, + "learning_rate": 0.0019654338752486617, + "loss": 0.1113, + "step": 9409 + }, + { + "epoch": 0.08168331872119165, + "grad_norm": 0.66796875, + "learning_rate": 0.0019654257056165875, + "loss": 0.1143, + "step": 9410 + }, + { + "epoch": 0.08169199920139582, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019654175350380973, + "loss": 0.126, + "step": 9411 + }, + { + "epoch": 0.08170067968159998, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019654093635131987, + "loss": 0.1543, + "step": 9412 + }, + { + "epoch": 0.08170936016180415, + "grad_norm": 0.515625, + "learning_rate": 0.001965401191041902, + "loss": 0.1689, + "step": 9413 + }, + { + "epoch": 0.08171804064200831, + "grad_norm": 0.76171875, + "learning_rate": 0.0019653930176242156, + "loss": 0.1562, + "step": 9414 + }, + { + "epoch": 0.08172672112221248, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019653848432601485, + "loss": 0.0996, + "step": 9415 + }, + { + "epoch": 0.08173540160241664, + "grad_norm": 0.35546875, + "learning_rate": 0.0019653766679497094, + "loss": 0.1543, + "step": 9416 + }, + { + "epoch": 0.08174408208262081, + "grad_norm": 0.271484375, + "learning_rate": 0.001965368491692907, + "loss": 0.1406, + "step": 9417 + }, + { + "epoch": 0.08175276256282497, + "grad_norm": 0.62109375, + "learning_rate": 0.0019653603144897513, + "loss": 0.167, + "step": 9418 + }, + { + "epoch": 0.08176144304302914, + "grad_norm": 0.2080078125, + "learning_rate": 0.00196535213634025, + "loss": 0.1426, + "step": 9419 + }, + { + "epoch": 0.0817701235232333, + "grad_norm": 0.0634765625, + "learning_rate": 0.0019653439572444127, + "loss": 0.0962, + "step": 9420 + }, + { + "epoch": 0.08177880400343747, + "grad_norm": 0.71484375, + "learning_rate": 0.0019653357772022482, + "loss": 0.1641, + "step": 9421 + }, + { + "epoch": 0.08178748448364163, + "grad_norm": 0.158203125, + "learning_rate": 0.0019653275962137656, + "loss": 0.1035, + "step": 9422 + }, + { + "epoch": 0.0817961649638458, + "grad_norm": 0.2197265625, + "learning_rate": 0.001965319414278974, + "loss": 0.0889, + "step": 9423 + }, + { + "epoch": 0.08180484544404996, + "grad_norm": 0.5546875, + "learning_rate": 0.0019653112313978817, + "loss": 0.1582, + "step": 9424 + }, + { + "epoch": 0.08181352592425413, + "grad_norm": 0.43359375, + "learning_rate": 0.0019653030475704986, + "loss": 0.1162, + "step": 9425 + }, + { + "epoch": 0.0818222064044583, + "grad_norm": 0.28125, + "learning_rate": 0.0019652948627968325, + "loss": 0.0933, + "step": 9426 + }, + { + "epoch": 0.08183088688466246, + "grad_norm": 0.150390625, + "learning_rate": 0.001965286677076893, + "loss": 0.166, + "step": 9427 + }, + { + "epoch": 0.08183956736486662, + "grad_norm": 0.294921875, + "learning_rate": 0.0019652784904106895, + "loss": 0.1157, + "step": 9428 + }, + { + "epoch": 0.08184824784507079, + "grad_norm": 0.12890625, + "learning_rate": 0.00196527030279823, + "loss": 0.167, + "step": 9429 + }, + { + "epoch": 0.08185692832527496, + "grad_norm": 0.3984375, + "learning_rate": 0.001965262114239524, + "loss": 0.1279, + "step": 9430 + }, + { + "epoch": 0.08186560880547912, + "grad_norm": 0.25, + "learning_rate": 0.0019652539247345803, + "loss": 0.1216, + "step": 9431 + }, + { + "epoch": 0.08187428928568329, + "grad_norm": 0.40625, + "learning_rate": 0.001965245734283408, + "loss": 0.1016, + "step": 9432 + }, + { + "epoch": 0.08188296976588745, + "grad_norm": 0.4765625, + "learning_rate": 0.0019652375428860155, + "loss": 0.0957, + "step": 9433 + }, + { + "epoch": 0.08189165024609162, + "grad_norm": 0.2255859375, + "learning_rate": 0.001965229350542413, + "loss": 0.1128, + "step": 9434 + }, + { + "epoch": 0.08190033072629578, + "grad_norm": 0.25390625, + "learning_rate": 0.001965221157252608, + "loss": 0.1631, + "step": 9435 + }, + { + "epoch": 0.08190901120649995, + "grad_norm": 0.51171875, + "learning_rate": 0.0019652129630166105, + "loss": 0.1309, + "step": 9436 + }, + { + "epoch": 0.08191769168670411, + "grad_norm": 0.25390625, + "learning_rate": 0.001965204767834429, + "loss": 0.1514, + "step": 9437 + }, + { + "epoch": 0.08192637216690828, + "grad_norm": 0.2197265625, + "learning_rate": 0.001965196571706073, + "loss": 0.1895, + "step": 9438 + }, + { + "epoch": 0.08193505264711244, + "grad_norm": 0.259765625, + "learning_rate": 0.0019651883746315503, + "loss": 0.1367, + "step": 9439 + }, + { + "epoch": 0.0819437331273166, + "grad_norm": 0.46875, + "learning_rate": 0.001965180176610871, + "loss": 0.1152, + "step": 9440 + }, + { + "epoch": 0.08195241360752077, + "grad_norm": 0.60546875, + "learning_rate": 0.0019651719776440433, + "loss": 0.1553, + "step": 9441 + }, + { + "epoch": 0.08196109408772494, + "grad_norm": 0.373046875, + "learning_rate": 0.001965163777731077, + "loss": 0.1367, + "step": 9442 + }, + { + "epoch": 0.0819697745679291, + "grad_norm": 0.85546875, + "learning_rate": 0.0019651555768719804, + "loss": 0.1143, + "step": 9443 + }, + { + "epoch": 0.08197845504813327, + "grad_norm": 0.68359375, + "learning_rate": 0.001965147375066763, + "loss": 0.1367, + "step": 9444 + }, + { + "epoch": 0.08198713552833743, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019651391723154326, + "loss": 0.084, + "step": 9445 + }, + { + "epoch": 0.0819958160085416, + "grad_norm": 0.6640625, + "learning_rate": 0.001965130968618, + "loss": 0.1357, + "step": 9446 + }, + { + "epoch": 0.08200449648874576, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019651227639744724, + "loss": 0.1221, + "step": 9447 + }, + { + "epoch": 0.08201317696894993, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019651145583848596, + "loss": 0.1138, + "step": 9448 + }, + { + "epoch": 0.08202185744915409, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019651063518491708, + "loss": 0.0938, + "step": 9449 + }, + { + "epoch": 0.08203053792935826, + "grad_norm": 0.431640625, + "learning_rate": 0.0019650981443674145, + "loss": 0.085, + "step": 9450 + }, + { + "epoch": 0.08203921840956242, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019650899359396003, + "loss": 0.1973, + "step": 9451 + }, + { + "epoch": 0.08204789888976659, + "grad_norm": 0.6171875, + "learning_rate": 0.0019650817265657364, + "loss": 0.124, + "step": 9452 + }, + { + "epoch": 0.08205657936997075, + "grad_norm": 0.09130859375, + "learning_rate": 0.001965073516245832, + "loss": 0.1055, + "step": 9453 + }, + { + "epoch": 0.08206525985017492, + "grad_norm": 0.212890625, + "learning_rate": 0.001965065304979896, + "loss": 0.1299, + "step": 9454 + }, + { + "epoch": 0.08207394033037908, + "grad_norm": 0.6328125, + "learning_rate": 0.0019650570927679384, + "loss": 0.1357, + "step": 9455 + }, + { + "epoch": 0.08208262081058325, + "grad_norm": 0.2470703125, + "learning_rate": 0.001965048879609967, + "loss": 0.1367, + "step": 9456 + }, + { + "epoch": 0.0820913012907874, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019650406655059905, + "loss": 0.1167, + "step": 9457 + }, + { + "epoch": 0.08209998177099156, + "grad_norm": 0.56640625, + "learning_rate": 0.0019650324504560195, + "loss": 0.1396, + "step": 9458 + }, + { + "epoch": 0.08210866225119573, + "grad_norm": 0.369140625, + "learning_rate": 0.0019650242344600613, + "loss": 0.104, + "step": 9459 + }, + { + "epoch": 0.0821173427313999, + "grad_norm": 0.142578125, + "learning_rate": 0.001965016017518126, + "loss": 0.125, + "step": 9460 + }, + { + "epoch": 0.08212602321160406, + "grad_norm": 0.18359375, + "learning_rate": 0.0019650077996302222, + "loss": 0.1016, + "step": 9461 + }, + { + "epoch": 0.08213470369180823, + "grad_norm": 0.236328125, + "learning_rate": 0.0019649995807963587, + "loss": 0.1338, + "step": 9462 + }, + { + "epoch": 0.08214338417201239, + "grad_norm": 0.9296875, + "learning_rate": 0.0019649913610165445, + "loss": 0.1445, + "step": 9463 + }, + { + "epoch": 0.08215206465221656, + "grad_norm": 0.84765625, + "learning_rate": 0.001964983140290789, + "loss": 0.1484, + "step": 9464 + }, + { + "epoch": 0.08216074513242072, + "grad_norm": 0.427734375, + "learning_rate": 0.001964974918619101, + "loss": 0.125, + "step": 9465 + }, + { + "epoch": 0.08216942561262489, + "grad_norm": 0.138671875, + "learning_rate": 0.001964966696001489, + "loss": 0.1455, + "step": 9466 + }, + { + "epoch": 0.08217810609282905, + "grad_norm": 0.3828125, + "learning_rate": 0.0019649584724379626, + "loss": 0.1216, + "step": 9467 + }, + { + "epoch": 0.08218678657303322, + "grad_norm": 0.146484375, + "learning_rate": 0.001964950247928531, + "loss": 0.1826, + "step": 9468 + }, + { + "epoch": 0.08219546705323738, + "grad_norm": 0.140625, + "learning_rate": 0.001964942022473202, + "loss": 0.0791, + "step": 9469 + }, + { + "epoch": 0.08220414753344155, + "grad_norm": 0.25390625, + "learning_rate": 0.001964933796071986, + "loss": 0.1338, + "step": 9470 + }, + { + "epoch": 0.08221282801364571, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019649255687248916, + "loss": 0.1001, + "step": 9471 + }, + { + "epoch": 0.08222150849384988, + "grad_norm": 0.373046875, + "learning_rate": 0.001964917340431927, + "loss": 0.1035, + "step": 9472 + }, + { + "epoch": 0.08223018897405404, + "grad_norm": 0.88671875, + "learning_rate": 0.0019649091111931024, + "loss": 0.1426, + "step": 9473 + }, + { + "epoch": 0.0822388694542582, + "grad_norm": 0.470703125, + "learning_rate": 0.001964900881008426, + "loss": 0.1035, + "step": 9474 + }, + { + "epoch": 0.08224754993446237, + "grad_norm": 0.12890625, + "learning_rate": 0.0019648926498779065, + "loss": 0.1318, + "step": 9475 + }, + { + "epoch": 0.08225623041466654, + "grad_norm": 0.171875, + "learning_rate": 0.001964884417801554, + "loss": 0.1162, + "step": 9476 + }, + { + "epoch": 0.0822649108948707, + "grad_norm": 0.228515625, + "learning_rate": 0.001964876184779376, + "loss": 0.1104, + "step": 9477 + }, + { + "epoch": 0.08227359137507487, + "grad_norm": 0.3203125, + "learning_rate": 0.0019648679508113835, + "loss": 0.084, + "step": 9478 + }, + { + "epoch": 0.08228227185527903, + "grad_norm": 0.96875, + "learning_rate": 0.001964859715897584, + "loss": 0.1348, + "step": 9479 + }, + { + "epoch": 0.0822909523354832, + "grad_norm": 0.5078125, + "learning_rate": 0.001964851480037987, + "loss": 0.1182, + "step": 9480 + }, + { + "epoch": 0.08229963281568736, + "grad_norm": 0.12255859375, + "learning_rate": 0.001964843243232601, + "loss": 0.1494, + "step": 9481 + }, + { + "epoch": 0.08230831329589153, + "grad_norm": 1.328125, + "learning_rate": 0.001964835005481436, + "loss": 0.1484, + "step": 9482 + }, + { + "epoch": 0.08231699377609569, + "grad_norm": 0.388671875, + "learning_rate": 0.0019648267667845003, + "loss": 0.1094, + "step": 9483 + }, + { + "epoch": 0.08232567425629986, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019648185271418026, + "loss": 0.0967, + "step": 9484 + }, + { + "epoch": 0.08233435473650402, + "grad_norm": 0.376953125, + "learning_rate": 0.0019648102865533527, + "loss": 0.1113, + "step": 9485 + }, + { + "epoch": 0.08234303521670819, + "grad_norm": 0.42578125, + "learning_rate": 0.001964802045019159, + "loss": 0.1245, + "step": 9486 + }, + { + "epoch": 0.08235171569691235, + "grad_norm": 0.11572265625, + "learning_rate": 0.001964793802539231, + "loss": 0.0977, + "step": 9487 + }, + { + "epoch": 0.08236039617711652, + "grad_norm": 0.400390625, + "learning_rate": 0.0019647855591135776, + "loss": 0.1357, + "step": 9488 + }, + { + "epoch": 0.08236907665732068, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019647773147422076, + "loss": 0.1201, + "step": 9489 + }, + { + "epoch": 0.08237775713752485, + "grad_norm": 0.2734375, + "learning_rate": 0.00196476906942513, + "loss": 0.1064, + "step": 9490 + }, + { + "epoch": 0.08238643761772901, + "grad_norm": 0.92578125, + "learning_rate": 0.001964760823162354, + "loss": 0.1035, + "step": 9491 + }, + { + "epoch": 0.08239511809793318, + "grad_norm": 0.125, + "learning_rate": 0.0019647525759538885, + "loss": 0.1445, + "step": 9492 + }, + { + "epoch": 0.08240379857813734, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019647443277997427, + "loss": 0.0986, + "step": 9493 + }, + { + "epoch": 0.08241247905834151, + "grad_norm": 0.337890625, + "learning_rate": 0.0019647360786999254, + "loss": 0.1045, + "step": 9494 + }, + { + "epoch": 0.08242115953854567, + "grad_norm": 0.171875, + "learning_rate": 0.0019647278286544457, + "loss": 0.1465, + "step": 9495 + }, + { + "epoch": 0.08242984001874984, + "grad_norm": 0.330078125, + "learning_rate": 0.0019647195776633128, + "loss": 0.1279, + "step": 9496 + }, + { + "epoch": 0.082438520498954, + "grad_norm": 0.09765625, + "learning_rate": 0.0019647113257265356, + "loss": 0.1074, + "step": 9497 + }, + { + "epoch": 0.08244720097915817, + "grad_norm": 0.1572265625, + "learning_rate": 0.001964703072844123, + "loss": 0.1123, + "step": 9498 + }, + { + "epoch": 0.08245588145936233, + "grad_norm": 0.126953125, + "learning_rate": 0.001964694819016084, + "loss": 0.1348, + "step": 9499 + }, + { + "epoch": 0.0824645619395665, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019646865642424277, + "loss": 0.1211, + "step": 9500 + }, + { + "epoch": 0.08247324241977066, + "grad_norm": 0.7265625, + "learning_rate": 0.0019646783085231633, + "loss": 0.1045, + "step": 9501 + }, + { + "epoch": 0.08248192289997483, + "grad_norm": 0.28125, + "learning_rate": 0.0019646700518582997, + "loss": 0.1191, + "step": 9502 + }, + { + "epoch": 0.082490603380179, + "grad_norm": 0.306640625, + "learning_rate": 0.001964661794247846, + "loss": 0.1191, + "step": 9503 + }, + { + "epoch": 0.08249928386038316, + "grad_norm": 0.345703125, + "learning_rate": 0.001964653535691811, + "loss": 0.1807, + "step": 9504 + }, + { + "epoch": 0.08250796434058733, + "grad_norm": 0.380859375, + "learning_rate": 0.0019646452761902036, + "loss": 0.1143, + "step": 9505 + }, + { + "epoch": 0.08251664482079149, + "grad_norm": 0.5234375, + "learning_rate": 0.0019646370157430336, + "loss": 0.1055, + "step": 9506 + }, + { + "epoch": 0.08252532530099566, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019646287543503096, + "loss": 0.1416, + "step": 9507 + }, + { + "epoch": 0.08253400578119982, + "grad_norm": 0.546875, + "learning_rate": 0.0019646204920120403, + "loss": 0.1445, + "step": 9508 + }, + { + "epoch": 0.08254268626140399, + "grad_norm": 0.65234375, + "learning_rate": 0.0019646122287282353, + "loss": 0.1914, + "step": 9509 + }, + { + "epoch": 0.08255136674160815, + "grad_norm": 0.109375, + "learning_rate": 0.001964603964498903, + "loss": 0.1406, + "step": 9510 + }, + { + "epoch": 0.08256004722181232, + "grad_norm": 0.55078125, + "learning_rate": 0.001964595699324053, + "loss": 0.1572, + "step": 9511 + }, + { + "epoch": 0.08256872770201648, + "grad_norm": 0.51171875, + "learning_rate": 0.001964587433203694, + "loss": 0.1367, + "step": 9512 + }, + { + "epoch": 0.08257740818222065, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019645791661378354, + "loss": 0.0815, + "step": 9513 + }, + { + "epoch": 0.08258608866242481, + "grad_norm": 0.482421875, + "learning_rate": 0.0019645708981264856, + "loss": 0.1001, + "step": 9514 + }, + { + "epoch": 0.08259476914262898, + "grad_norm": 0.85546875, + "learning_rate": 0.0019645626291696547, + "loss": 0.0938, + "step": 9515 + }, + { + "epoch": 0.08260344962283314, + "grad_norm": 2.859375, + "learning_rate": 0.0019645543592673505, + "loss": 0.293, + "step": 9516 + }, + { + "epoch": 0.0826121301030373, + "grad_norm": 0.103515625, + "learning_rate": 0.001964546088419583, + "loss": 0.1162, + "step": 9517 + }, + { + "epoch": 0.08262081058324146, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019645378166263604, + "loss": 0.0928, + "step": 9518 + }, + { + "epoch": 0.08262949106344562, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019645295438876927, + "loss": 0.124, + "step": 9519 + }, + { + "epoch": 0.08263817154364979, + "grad_norm": 0.251953125, + "learning_rate": 0.0019645212702035886, + "loss": 0.1406, + "step": 9520 + }, + { + "epoch": 0.08264685202385395, + "grad_norm": 0.490234375, + "learning_rate": 0.0019645129955740566, + "loss": 0.1289, + "step": 9521 + }, + { + "epoch": 0.08265553250405812, + "grad_norm": 0.490234375, + "learning_rate": 0.0019645047199991065, + "loss": 0.1348, + "step": 9522 + }, + { + "epoch": 0.08266421298426228, + "grad_norm": 0.2080078125, + "learning_rate": 0.001964496443478747, + "loss": 0.0972, + "step": 9523 + }, + { + "epoch": 0.08267289346446645, + "grad_norm": 0.255859375, + "learning_rate": 0.001964488166012987, + "loss": 0.1147, + "step": 9524 + }, + { + "epoch": 0.08268157394467061, + "grad_norm": 0.12255859375, + "learning_rate": 0.001964479887601836, + "loss": 0.1465, + "step": 9525 + }, + { + "epoch": 0.08269025442487478, + "grad_norm": 0.208984375, + "learning_rate": 0.0019644716082453024, + "loss": 0.125, + "step": 9526 + }, + { + "epoch": 0.08269893490507894, + "grad_norm": 0.1181640625, + "learning_rate": 0.001964463327943396, + "loss": 0.1621, + "step": 9527 + }, + { + "epoch": 0.08270761538528311, + "grad_norm": 0.474609375, + "learning_rate": 0.001964455046696125, + "loss": 0.1562, + "step": 9528 + }, + { + "epoch": 0.08271629586548727, + "grad_norm": 0.408203125, + "learning_rate": 0.0019644467645034995, + "loss": 0.1182, + "step": 9529 + }, + { + "epoch": 0.08272497634569144, + "grad_norm": 1.375, + "learning_rate": 0.001964438481365528, + "loss": 0.0962, + "step": 9530 + }, + { + "epoch": 0.0827336568258956, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019644301972822193, + "loss": 0.1289, + "step": 9531 + }, + { + "epoch": 0.08274233730609977, + "grad_norm": 0.279296875, + "learning_rate": 0.0019644219122535826, + "loss": 0.1504, + "step": 9532 + }, + { + "epoch": 0.08275101778630393, + "grad_norm": 0.5546875, + "learning_rate": 0.0019644136262796275, + "loss": 0.1089, + "step": 9533 + }, + { + "epoch": 0.0827596982665081, + "grad_norm": 0.609375, + "learning_rate": 0.001964405339360362, + "loss": 0.1084, + "step": 9534 + }, + { + "epoch": 0.08276837874671227, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019643970514957964, + "loss": 0.1074, + "step": 9535 + }, + { + "epoch": 0.08277705922691643, + "grad_norm": 0.416015625, + "learning_rate": 0.0019643887626859388, + "loss": 0.1328, + "step": 9536 + }, + { + "epoch": 0.0827857397071206, + "grad_norm": 0.546875, + "learning_rate": 0.001964380472930799, + "loss": 0.1455, + "step": 9537 + }, + { + "epoch": 0.08279442018732476, + "grad_norm": 0.39453125, + "learning_rate": 0.0019643721822303855, + "loss": 0.1021, + "step": 9538 + }, + { + "epoch": 0.08280310066752893, + "grad_norm": 0.50390625, + "learning_rate": 0.0019643638905847078, + "loss": 0.1592, + "step": 9539 + }, + { + "epoch": 0.08281178114773309, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019643555979937744, + "loss": 0.1021, + "step": 9540 + }, + { + "epoch": 0.08282046162793726, + "grad_norm": 0.29296875, + "learning_rate": 0.001964347304457595, + "loss": 0.127, + "step": 9541 + }, + { + "epoch": 0.08282914210814142, + "grad_norm": 0.09033203125, + "learning_rate": 0.001964339009976178, + "loss": 0.1289, + "step": 9542 + }, + { + "epoch": 0.08283782258834559, + "grad_norm": 0.291015625, + "learning_rate": 0.001964330714549533, + "loss": 0.127, + "step": 9543 + }, + { + "epoch": 0.08284650306854975, + "grad_norm": 0.40625, + "learning_rate": 0.0019643224181776693, + "loss": 0.1455, + "step": 9544 + }, + { + "epoch": 0.08285518354875392, + "grad_norm": 0.921875, + "learning_rate": 0.001964314120860595, + "loss": 0.0845, + "step": 9545 + }, + { + "epoch": 0.08286386402895808, + "grad_norm": 0.30859375, + "learning_rate": 0.00196430582259832, + "loss": 0.1201, + "step": 9546 + }, + { + "epoch": 0.08287254450916225, + "grad_norm": 0.11767578125, + "learning_rate": 0.001964297523390853, + "loss": 0.1289, + "step": 9547 + }, + { + "epoch": 0.08288122498936641, + "grad_norm": 1.03125, + "learning_rate": 0.0019642892232382037, + "loss": 0.1475, + "step": 9548 + }, + { + "epoch": 0.08288990546957058, + "grad_norm": 0.357421875, + "learning_rate": 0.0019642809221403802, + "loss": 0.1084, + "step": 9549 + }, + { + "epoch": 0.08289858594977474, + "grad_norm": 0.255859375, + "learning_rate": 0.0019642726200973923, + "loss": 0.127, + "step": 9550 + }, + { + "epoch": 0.08290726642997891, + "grad_norm": 0.51171875, + "learning_rate": 0.001964264317109249, + "loss": 0.1211, + "step": 9551 + }, + { + "epoch": 0.08291594691018307, + "grad_norm": 0.72265625, + "learning_rate": 0.0019642560131759587, + "loss": 0.1523, + "step": 9552 + }, + { + "epoch": 0.08292462739038724, + "grad_norm": 0.4140625, + "learning_rate": 0.0019642477082975314, + "loss": 0.1641, + "step": 9553 + }, + { + "epoch": 0.0829333078705914, + "grad_norm": 0.30859375, + "learning_rate": 0.001964239402473976, + "loss": 0.1426, + "step": 9554 + }, + { + "epoch": 0.08294198835079557, + "grad_norm": 0.275390625, + "learning_rate": 0.0019642310957053008, + "loss": 0.1309, + "step": 9555 + }, + { + "epoch": 0.08295066883099973, + "grad_norm": 1.1796875, + "learning_rate": 0.001964222787991516, + "loss": 0.1123, + "step": 9556 + }, + { + "epoch": 0.0829593493112039, + "grad_norm": 0.271484375, + "learning_rate": 0.0019642144793326297, + "loss": 0.1201, + "step": 9557 + }, + { + "epoch": 0.08296802979140806, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019642061697286517, + "loss": 0.1289, + "step": 9558 + }, + { + "epoch": 0.08297671027161223, + "grad_norm": 0.80859375, + "learning_rate": 0.0019641978591795907, + "loss": 0.1143, + "step": 9559 + }, + { + "epoch": 0.08298539075181639, + "grad_norm": 0.2734375, + "learning_rate": 0.001964189547685456, + "loss": 0.106, + "step": 9560 + }, + { + "epoch": 0.08299407123202056, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019641812352462567, + "loss": 0.1201, + "step": 9561 + }, + { + "epoch": 0.08300275171222472, + "grad_norm": 0.494140625, + "learning_rate": 0.0019641729218620014, + "loss": 0.123, + "step": 9562 + }, + { + "epoch": 0.08301143219242889, + "grad_norm": 0.3828125, + "learning_rate": 0.0019641646075327, + "loss": 0.1143, + "step": 9563 + }, + { + "epoch": 0.08302011267263305, + "grad_norm": 0.62890625, + "learning_rate": 0.0019641562922583606, + "loss": 0.1104, + "step": 9564 + }, + { + "epoch": 0.08302879315283722, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019641479760389934, + "loss": 0.1201, + "step": 9565 + }, + { + "epoch": 0.08303747363304138, + "grad_norm": 0.212890625, + "learning_rate": 0.001964139658874607, + "loss": 0.1172, + "step": 9566 + }, + { + "epoch": 0.08304615411324555, + "grad_norm": 0.283203125, + "learning_rate": 0.0019641313407652104, + "loss": 0.1299, + "step": 9567 + }, + { + "epoch": 0.08305483459344971, + "grad_norm": 0.220703125, + "learning_rate": 0.0019641230217108123, + "loss": 0.1221, + "step": 9568 + }, + { + "epoch": 0.08306351507365388, + "grad_norm": 0.2578125, + "learning_rate": 0.0019641147017114224, + "loss": 0.0947, + "step": 9569 + }, + { + "epoch": 0.08307219555385804, + "grad_norm": 0.326171875, + "learning_rate": 0.00196410638076705, + "loss": 0.0991, + "step": 9570 + }, + { + "epoch": 0.08308087603406221, + "grad_norm": 0.1572265625, + "learning_rate": 0.001964098058877703, + "loss": 0.1016, + "step": 9571 + }, + { + "epoch": 0.08308955651426637, + "grad_norm": 0.314453125, + "learning_rate": 0.0019640897360433925, + "loss": 0.1162, + "step": 9572 + }, + { + "epoch": 0.08309823699447054, + "grad_norm": 0.46484375, + "learning_rate": 0.001964081412264126, + "loss": 0.1162, + "step": 9573 + }, + { + "epoch": 0.0831069174746747, + "grad_norm": 0.279296875, + "learning_rate": 0.001964073087539913, + "loss": 0.1523, + "step": 9574 + }, + { + "epoch": 0.08311559795487887, + "grad_norm": 0.48828125, + "learning_rate": 0.0019640647618707624, + "loss": 0.1309, + "step": 9575 + }, + { + "epoch": 0.08312427843508303, + "grad_norm": 0.2275390625, + "learning_rate": 0.001964056435256684, + "loss": 0.1387, + "step": 9576 + }, + { + "epoch": 0.0831329589152872, + "grad_norm": 0.326171875, + "learning_rate": 0.001964048107697686, + "loss": 0.1475, + "step": 9577 + }, + { + "epoch": 0.08314163939549137, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019640397791937784, + "loss": 0.1699, + "step": 9578 + }, + { + "epoch": 0.08315031987569553, + "grad_norm": 0.5546875, + "learning_rate": 0.0019640314497449698, + "loss": 0.1245, + "step": 9579 + }, + { + "epoch": 0.08315900035589968, + "grad_norm": 0.259765625, + "learning_rate": 0.0019640231193512694, + "loss": 0.1318, + "step": 9580 + }, + { + "epoch": 0.08316768083610385, + "grad_norm": 0.0849609375, + "learning_rate": 0.001964014788012686, + "loss": 0.0996, + "step": 9581 + }, + { + "epoch": 0.08317636131630801, + "grad_norm": 0.451171875, + "learning_rate": 0.001964006455729229, + "loss": 0.1216, + "step": 9582 + }, + { + "epoch": 0.08318504179651218, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019639981225009076, + "loss": 0.1299, + "step": 9583 + }, + { + "epoch": 0.08319372227671634, + "grad_norm": 0.5078125, + "learning_rate": 0.0019639897883277314, + "loss": 0.1289, + "step": 9584 + }, + { + "epoch": 0.08320240275692051, + "grad_norm": 0.337890625, + "learning_rate": 0.0019639814532097085, + "loss": 0.1094, + "step": 9585 + }, + { + "epoch": 0.08321108323712467, + "grad_norm": 1.0390625, + "learning_rate": 0.0019639731171468486, + "loss": 0.1045, + "step": 9586 + }, + { + "epoch": 0.08321976371732884, + "grad_norm": 0.44140625, + "learning_rate": 0.00196396478013916, + "loss": 0.1011, + "step": 9587 + }, + { + "epoch": 0.083228444197533, + "grad_norm": 0.38671875, + "learning_rate": 0.0019639564421866534, + "loss": 0.1484, + "step": 9588 + }, + { + "epoch": 0.08323712467773717, + "grad_norm": 0.671875, + "learning_rate": 0.001963948103289337, + "loss": 0.1064, + "step": 9589 + }, + { + "epoch": 0.08324580515794133, + "grad_norm": 0.345703125, + "learning_rate": 0.001963939763447219, + "loss": 0.1089, + "step": 9590 + }, + { + "epoch": 0.0832544856381455, + "grad_norm": 0.208984375, + "learning_rate": 0.0019639314226603103, + "loss": 0.1055, + "step": 9591 + }, + { + "epoch": 0.08326316611834966, + "grad_norm": 0.703125, + "learning_rate": 0.0019639230809286186, + "loss": 0.1416, + "step": 9592 + }, + { + "epoch": 0.08327184659855383, + "grad_norm": 0.267578125, + "learning_rate": 0.0019639147382521545, + "loss": 0.1113, + "step": 9593 + }, + { + "epoch": 0.083280527078758, + "grad_norm": 0.384765625, + "learning_rate": 0.0019639063946309252, + "loss": 0.1455, + "step": 9594 + }, + { + "epoch": 0.08328920755896216, + "grad_norm": 0.30078125, + "learning_rate": 0.0019638980500649413, + "loss": 0.0903, + "step": 9595 + }, + { + "epoch": 0.08329788803916632, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019638897045542118, + "loss": 0.104, + "step": 9596 + }, + { + "epoch": 0.08330656851937049, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019638813580987453, + "loss": 0.123, + "step": 9597 + }, + { + "epoch": 0.08331524899957465, + "grad_norm": 0.40625, + "learning_rate": 0.0019638730106985506, + "loss": 0.1147, + "step": 9598 + }, + { + "epoch": 0.08332392947977882, + "grad_norm": 0.251953125, + "learning_rate": 0.0019638646623536377, + "loss": 0.1396, + "step": 9599 + }, + { + "epoch": 0.08333260995998298, + "grad_norm": 0.421875, + "learning_rate": 0.0019638563130640156, + "loss": 0.1396, + "step": 9600 + }, + { + "epoch": 0.08334129044018715, + "grad_norm": 0.357421875, + "learning_rate": 0.001963847962829693, + "loss": 0.1104, + "step": 9601 + }, + { + "epoch": 0.08334997092039131, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019638396116506795, + "loss": 0.1416, + "step": 9602 + }, + { + "epoch": 0.08335865140059548, + "grad_norm": 0.640625, + "learning_rate": 0.0019638312595269838, + "loss": 0.1611, + "step": 9603 + }, + { + "epoch": 0.08336733188079964, + "grad_norm": 0.373046875, + "learning_rate": 0.0019638229064586153, + "loss": 0.1465, + "step": 9604 + }, + { + "epoch": 0.08337601236100381, + "grad_norm": 0.64453125, + "learning_rate": 0.0019638145524455827, + "loss": 0.1211, + "step": 9605 + }, + { + "epoch": 0.08338469284120797, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019638061974878956, + "loss": 0.1064, + "step": 9606 + }, + { + "epoch": 0.08339337332141214, + "grad_norm": 0.177734375, + "learning_rate": 0.001963797841585563, + "loss": 0.166, + "step": 9607 + }, + { + "epoch": 0.0834020538016163, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019637894847385944, + "loss": 0.1162, + "step": 9608 + }, + { + "epoch": 0.08341073428182047, + "grad_norm": 0.271484375, + "learning_rate": 0.0019637811269469984, + "loss": 0.1621, + "step": 9609 + }, + { + "epoch": 0.08341941476202464, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019637727682107844, + "loss": 0.1221, + "step": 9610 + }, + { + "epoch": 0.0834280952422288, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019637644085299614, + "loss": 0.1074, + "step": 9611 + }, + { + "epoch": 0.08343677572243297, + "grad_norm": 2.09375, + "learning_rate": 0.0019637560479045385, + "loss": 0.293, + "step": 9612 + }, + { + "epoch": 0.08344545620263713, + "grad_norm": 0.16796875, + "learning_rate": 0.0019637476863345253, + "loss": 0.1318, + "step": 9613 + }, + { + "epoch": 0.0834541366828413, + "grad_norm": 0.54296875, + "learning_rate": 0.0019637393238199304, + "loss": 0.1182, + "step": 9614 + }, + { + "epoch": 0.08346281716304546, + "grad_norm": 0.546875, + "learning_rate": 0.001963730960360763, + "loss": 0.1396, + "step": 9615 + }, + { + "epoch": 0.08347149764324963, + "grad_norm": 0.376953125, + "learning_rate": 0.0019637225959570326, + "loss": 0.1118, + "step": 9616 + }, + { + "epoch": 0.08348017812345379, + "grad_norm": 0.2734375, + "learning_rate": 0.0019637142306087483, + "loss": 0.1387, + "step": 9617 + }, + { + "epoch": 0.08348885860365796, + "grad_norm": 0.44140625, + "learning_rate": 0.001963705864315919, + "loss": 0.104, + "step": 9618 + }, + { + "epoch": 0.08349753908386212, + "grad_norm": 0.37109375, + "learning_rate": 0.0019636974970785536, + "loss": 0.1592, + "step": 9619 + }, + { + "epoch": 0.08350621956406629, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019636891288966622, + "loss": 0.1016, + "step": 9620 + }, + { + "epoch": 0.08351490004427045, + "grad_norm": 0.1962890625, + "learning_rate": 0.001963680759770253, + "loss": 0.1069, + "step": 9621 + }, + { + "epoch": 0.08352358052447462, + "grad_norm": 0.80078125, + "learning_rate": 0.001963672389699335, + "loss": 0.1367, + "step": 9622 + }, + { + "epoch": 0.08353226100467878, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019636640186839185, + "loss": 0.1562, + "step": 9623 + }, + { + "epoch": 0.08354094148488295, + "grad_norm": 0.474609375, + "learning_rate": 0.001963655646724012, + "loss": 0.0952, + "step": 9624 + }, + { + "epoch": 0.08354962196508711, + "grad_norm": 0.25, + "learning_rate": 0.0019636472738196243, + "loss": 0.0938, + "step": 9625 + }, + { + "epoch": 0.08355830244529128, + "grad_norm": 0.412109375, + "learning_rate": 0.001963638899970765, + "loss": 0.1206, + "step": 9626 + }, + { + "epoch": 0.08356698292549544, + "grad_norm": 0.119140625, + "learning_rate": 0.0019636305251774437, + "loss": 0.123, + "step": 9627 + }, + { + "epoch": 0.08357566340569961, + "grad_norm": 0.61328125, + "learning_rate": 0.0019636221494396685, + "loss": 0.126, + "step": 9628 + }, + { + "epoch": 0.08358434388590377, + "grad_norm": 0.349609375, + "learning_rate": 0.001963613772757449, + "loss": 0.1387, + "step": 9629 + }, + { + "epoch": 0.08359302436610794, + "grad_norm": 0.47265625, + "learning_rate": 0.001963605395130795, + "loss": 0.0991, + "step": 9630 + }, + { + "epoch": 0.0836017048463121, + "grad_norm": 0.2578125, + "learning_rate": 0.0019635970165597146, + "loss": 0.1025, + "step": 9631 + }, + { + "epoch": 0.08361038532651627, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019635886370442178, + "loss": 0.127, + "step": 9632 + }, + { + "epoch": 0.08361906580672043, + "grad_norm": 0.275390625, + "learning_rate": 0.0019635802565843136, + "loss": 0.1162, + "step": 9633 + }, + { + "epoch": 0.0836277462869246, + "grad_norm": 0.1875, + "learning_rate": 0.0019635718751800104, + "loss": 0.1406, + "step": 9634 + }, + { + "epoch": 0.08363642676712876, + "grad_norm": 0.1611328125, + "learning_rate": 0.001963563492831318, + "loss": 0.1055, + "step": 9635 + }, + { + "epoch": 0.08364510724733293, + "grad_norm": 0.140625, + "learning_rate": 0.001963555109538246, + "loss": 0.1426, + "step": 9636 + }, + { + "epoch": 0.0836537877275371, + "grad_norm": 0.75, + "learning_rate": 0.0019635467253008034, + "loss": 0.1348, + "step": 9637 + }, + { + "epoch": 0.08366246820774126, + "grad_norm": 0.46484375, + "learning_rate": 0.0019635383401189984, + "loss": 0.1064, + "step": 9638 + }, + { + "epoch": 0.08367114868794542, + "grad_norm": 0.287109375, + "learning_rate": 0.0019635299539928408, + "loss": 0.1309, + "step": 9639 + }, + { + "epoch": 0.08367982916814959, + "grad_norm": 0.28125, + "learning_rate": 0.0019635215669223404, + "loss": 0.1299, + "step": 9640 + }, + { + "epoch": 0.08368850964835375, + "grad_norm": 0.11767578125, + "learning_rate": 0.001963513178907505, + "loss": 0.1582, + "step": 9641 + }, + { + "epoch": 0.0836971901285579, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019635047899483455, + "loss": 0.1309, + "step": 9642 + }, + { + "epoch": 0.08370587060876207, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019634964000448695, + "loss": 0.125, + "step": 9643 + }, + { + "epoch": 0.08371455108896624, + "grad_norm": 1.34375, + "learning_rate": 0.001963488009197087, + "loss": 0.1562, + "step": 9644 + }, + { + "epoch": 0.0837232315691704, + "grad_norm": 0.142578125, + "learning_rate": 0.001963479617405007, + "loss": 0.1006, + "step": 9645 + }, + { + "epoch": 0.08373191204937457, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019634712246686388, + "loss": 0.1094, + "step": 9646 + }, + { + "epoch": 0.08374059252957873, + "grad_norm": 1.1328125, + "learning_rate": 0.001963462830987991, + "loss": 0.1992, + "step": 9647 + }, + { + "epoch": 0.0837492730097829, + "grad_norm": 0.22265625, + "learning_rate": 0.0019634544363630736, + "loss": 0.1299, + "step": 9648 + }, + { + "epoch": 0.08375795348998706, + "grad_norm": 0.43359375, + "learning_rate": 0.0019634460407938954, + "loss": 0.1357, + "step": 9649 + }, + { + "epoch": 0.08376663397019123, + "grad_norm": 0.30859375, + "learning_rate": 0.0019634376442804655, + "loss": 0.1055, + "step": 9650 + }, + { + "epoch": 0.08377531445039539, + "grad_norm": 0.5078125, + "learning_rate": 0.0019634292468227927, + "loss": 0.1338, + "step": 9651 + }, + { + "epoch": 0.08378399493059956, + "grad_norm": 0.400390625, + "learning_rate": 0.0019634208484208874, + "loss": 0.1035, + "step": 9652 + }, + { + "epoch": 0.08379267541080372, + "grad_norm": 0.48828125, + "learning_rate": 0.0019634124490747577, + "loss": 0.1182, + "step": 9653 + }, + { + "epoch": 0.08380135589100789, + "grad_norm": 0.396484375, + "learning_rate": 0.001963404048784413, + "loss": 0.1152, + "step": 9654 + }, + { + "epoch": 0.08381003637121205, + "grad_norm": 0.369140625, + "learning_rate": 0.001963395647549863, + "loss": 0.1289, + "step": 9655 + }, + { + "epoch": 0.08381871685141622, + "grad_norm": 0.65625, + "learning_rate": 0.0019633872453711163, + "loss": 0.1406, + "step": 9656 + }, + { + "epoch": 0.08382739733162038, + "grad_norm": 0.333984375, + "learning_rate": 0.001963378842248182, + "loss": 0.1475, + "step": 9657 + }, + { + "epoch": 0.08383607781182455, + "grad_norm": 0.3828125, + "learning_rate": 0.00196337043818107, + "loss": 0.123, + "step": 9658 + }, + { + "epoch": 0.08384475829202871, + "grad_norm": 0.53515625, + "learning_rate": 0.0019633620331697887, + "loss": 0.1133, + "step": 9659 + }, + { + "epoch": 0.08385343877223288, + "grad_norm": 0.287109375, + "learning_rate": 0.001963353627214348, + "loss": 0.123, + "step": 9660 + }, + { + "epoch": 0.08386211925243704, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019633452203147565, + "loss": 0.1328, + "step": 9661 + }, + { + "epoch": 0.08387079973264121, + "grad_norm": 0.19921875, + "learning_rate": 0.0019633368124710236, + "loss": 0.1289, + "step": 9662 + }, + { + "epoch": 0.08387948021284537, + "grad_norm": 0.125, + "learning_rate": 0.0019633284036831584, + "loss": 0.1138, + "step": 9663 + }, + { + "epoch": 0.08388816069304954, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019633199939511708, + "loss": 0.1611, + "step": 9664 + }, + { + "epoch": 0.0838968411732537, + "grad_norm": 0.55859375, + "learning_rate": 0.001963311583275069, + "loss": 0.1484, + "step": 9665 + }, + { + "epoch": 0.08390552165345787, + "grad_norm": 0.294921875, + "learning_rate": 0.0019633031716548626, + "loss": 0.1055, + "step": 9666 + }, + { + "epoch": 0.08391420213366203, + "grad_norm": 0.87109375, + "learning_rate": 0.001963294759090561, + "loss": 0.1309, + "step": 9667 + }, + { + "epoch": 0.0839228826138662, + "grad_norm": 0.23828125, + "learning_rate": 0.001963286345582173, + "loss": 0.1465, + "step": 9668 + }, + { + "epoch": 0.08393156309407036, + "grad_norm": 0.97265625, + "learning_rate": 0.0019632779311297083, + "loss": 0.1309, + "step": 9669 + }, + { + "epoch": 0.08394024357427453, + "grad_norm": 0.515625, + "learning_rate": 0.0019632695157331755, + "loss": 0.124, + "step": 9670 + }, + { + "epoch": 0.0839489240544787, + "grad_norm": 0.1015625, + "learning_rate": 0.0019632610993925844, + "loss": 0.1328, + "step": 9671 + }, + { + "epoch": 0.08395760453468286, + "grad_norm": 1.1640625, + "learning_rate": 0.001963252682107944, + "loss": 0.2383, + "step": 9672 + }, + { + "epoch": 0.08396628501488702, + "grad_norm": 0.62109375, + "learning_rate": 0.0019632442638792633, + "loss": 0.0972, + "step": 9673 + }, + { + "epoch": 0.08397496549509119, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019632358447065515, + "loss": 0.1289, + "step": 9674 + }, + { + "epoch": 0.08398364597529535, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019632274245898183, + "loss": 0.1113, + "step": 9675 + }, + { + "epoch": 0.08399232645549952, + "grad_norm": 0.1689453125, + "learning_rate": 0.001963219003529072, + "loss": 0.1084, + "step": 9676 + }, + { + "epoch": 0.08400100693570368, + "grad_norm": 0.076171875, + "learning_rate": 0.0019632105815243226, + "loss": 0.1035, + "step": 9677 + }, + { + "epoch": 0.08400968741590785, + "grad_norm": 0.435546875, + "learning_rate": 0.0019632021585755796, + "loss": 0.1074, + "step": 9678 + }, + { + "epoch": 0.08401836789611201, + "grad_norm": 0.16796875, + "learning_rate": 0.001963193734682851, + "loss": 0.1289, + "step": 9679 + }, + { + "epoch": 0.08402704837631618, + "grad_norm": 0.6953125, + "learning_rate": 0.001963185309846147, + "loss": 0.2363, + "step": 9680 + }, + { + "epoch": 0.08403572885652034, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019631768840654767, + "loss": 0.1475, + "step": 9681 + }, + { + "epoch": 0.08404440933672451, + "grad_norm": 0.228515625, + "learning_rate": 0.0019631684573408487, + "loss": 0.1436, + "step": 9682 + }, + { + "epoch": 0.08405308981692867, + "grad_norm": 0.55078125, + "learning_rate": 0.0019631600296722727, + "loss": 0.1387, + "step": 9683 + }, + { + "epoch": 0.08406177029713284, + "grad_norm": 0.201171875, + "learning_rate": 0.001963151601059758, + "loss": 0.1221, + "step": 9684 + }, + { + "epoch": 0.084070450777337, + "grad_norm": 0.453125, + "learning_rate": 0.001963143171503314, + "loss": 0.2051, + "step": 9685 + }, + { + "epoch": 0.08407913125754117, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019631347410029496, + "loss": 0.1543, + "step": 9686 + }, + { + "epoch": 0.08408781173774534, + "grad_norm": 0.318359375, + "learning_rate": 0.0019631263095586733, + "loss": 0.1709, + "step": 9687 + }, + { + "epoch": 0.0840964922179495, + "grad_norm": 0.396484375, + "learning_rate": 0.0019631178771704954, + "loss": 0.127, + "step": 9688 + }, + { + "epoch": 0.08410517269815367, + "grad_norm": 0.81640625, + "learning_rate": 0.001963109443838425, + "loss": 0.1455, + "step": 9689 + }, + { + "epoch": 0.08411385317835783, + "grad_norm": 0.263671875, + "learning_rate": 0.0019631010095624707, + "loss": 0.1191, + "step": 9690 + }, + { + "epoch": 0.084122533658562, + "grad_norm": 0.98828125, + "learning_rate": 0.001963092574342642, + "loss": 0.1992, + "step": 9691 + }, + { + "epoch": 0.08413121413876616, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019630841381789487, + "loss": 0.1123, + "step": 9692 + }, + { + "epoch": 0.08413989461897033, + "grad_norm": 0.5859375, + "learning_rate": 0.0019630757010713996, + "loss": 0.1084, + "step": 9693 + }, + { + "epoch": 0.08414857509917449, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019630672630200035, + "loss": 0.1318, + "step": 9694 + }, + { + "epoch": 0.08415725557937866, + "grad_norm": 0.41015625, + "learning_rate": 0.00196305882402477, + "loss": 0.1895, + "step": 9695 + }, + { + "epoch": 0.08416593605958282, + "grad_norm": 0.244140625, + "learning_rate": 0.0019630503840857086, + "loss": 0.1104, + "step": 9696 + }, + { + "epoch": 0.08417461653978699, + "grad_norm": 0.35546875, + "learning_rate": 0.0019630419432028283, + "loss": 0.1709, + "step": 9697 + }, + { + "epoch": 0.08418329701999115, + "grad_norm": 0.2890625, + "learning_rate": 0.0019630335013761384, + "loss": 0.127, + "step": 9698 + }, + { + "epoch": 0.08419197750019532, + "grad_norm": 0.26953125, + "learning_rate": 0.0019630250586056474, + "loss": 0.1348, + "step": 9699 + }, + { + "epoch": 0.08420065798039948, + "grad_norm": 0.171875, + "learning_rate": 0.001963016614891366, + "loss": 0.1143, + "step": 9700 + }, + { + "epoch": 0.08420933846060365, + "grad_norm": 0.2275390625, + "learning_rate": 0.001963008170233302, + "loss": 0.1836, + "step": 9701 + }, + { + "epoch": 0.08421801894080781, + "grad_norm": 0.5078125, + "learning_rate": 0.0019629997246314655, + "loss": 0.1162, + "step": 9702 + }, + { + "epoch": 0.08422669942101196, + "grad_norm": 0.75390625, + "learning_rate": 0.0019629912780858654, + "loss": 0.0957, + "step": 9703 + }, + { + "epoch": 0.08423537990121613, + "grad_norm": 0.4609375, + "learning_rate": 0.001962982830596511, + "loss": 0.1406, + "step": 9704 + }, + { + "epoch": 0.0842440603814203, + "grad_norm": 0.32421875, + "learning_rate": 0.0019629743821634113, + "loss": 0.1553, + "step": 9705 + }, + { + "epoch": 0.08425274086162446, + "grad_norm": 1.5234375, + "learning_rate": 0.001962965932786576, + "loss": 0.2617, + "step": 9706 + }, + { + "epoch": 0.08426142134182862, + "grad_norm": 0.330078125, + "learning_rate": 0.001962957482466014, + "loss": 0.1377, + "step": 9707 + }, + { + "epoch": 0.08427010182203279, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019629490312017346, + "loss": 0.1191, + "step": 9708 + }, + { + "epoch": 0.08427878230223695, + "grad_norm": 0.546875, + "learning_rate": 0.0019629405789937477, + "loss": 0.1475, + "step": 9709 + }, + { + "epoch": 0.08428746278244112, + "grad_norm": 0.23046875, + "learning_rate": 0.0019629321258420613, + "loss": 0.125, + "step": 9710 + }, + { + "epoch": 0.08429614326264528, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019629236717466857, + "loss": 0.1108, + "step": 9711 + }, + { + "epoch": 0.08430482374284945, + "grad_norm": 0.953125, + "learning_rate": 0.0019629152167076297, + "loss": 0.1631, + "step": 9712 + }, + { + "epoch": 0.08431350422305361, + "grad_norm": 0.498046875, + "learning_rate": 0.0019629067607249023, + "loss": 0.125, + "step": 9713 + }, + { + "epoch": 0.08432218470325778, + "grad_norm": 0.220703125, + "learning_rate": 0.001962898303798513, + "loss": 0.1172, + "step": 9714 + }, + { + "epoch": 0.08433086518346194, + "grad_norm": 0.23828125, + "learning_rate": 0.001962889845928471, + "loss": 0.127, + "step": 9715 + }, + { + "epoch": 0.08433954566366611, + "grad_norm": 0.1748046875, + "learning_rate": 0.001962881387114786, + "loss": 0.1172, + "step": 9716 + }, + { + "epoch": 0.08434822614387028, + "grad_norm": 0.73046875, + "learning_rate": 0.001962872927357467, + "loss": 0.1592, + "step": 9717 + }, + { + "epoch": 0.08435690662407444, + "grad_norm": 1.171875, + "learning_rate": 0.001962864466656523, + "loss": 0.2695, + "step": 9718 + }, + { + "epoch": 0.0843655871042786, + "grad_norm": 0.146484375, + "learning_rate": 0.001962856005011963, + "loss": 0.1113, + "step": 9719 + }, + { + "epoch": 0.08437426758448277, + "grad_norm": 0.369140625, + "learning_rate": 0.001962847542423797, + "loss": 0.0962, + "step": 9720 + }, + { + "epoch": 0.08438294806468694, + "grad_norm": 0.75, + "learning_rate": 0.0019628390788920336, + "loss": 0.1748, + "step": 9721 + }, + { + "epoch": 0.0843916285448911, + "grad_norm": 0.1123046875, + "learning_rate": 0.001962830614416683, + "loss": 0.1113, + "step": 9722 + }, + { + "epoch": 0.08440030902509527, + "grad_norm": 1.3125, + "learning_rate": 0.0019628221489977533, + "loss": 0.168, + "step": 9723 + }, + { + "epoch": 0.08440898950529943, + "grad_norm": 0.1748046875, + "learning_rate": 0.001962813682635254, + "loss": 0.165, + "step": 9724 + }, + { + "epoch": 0.0844176699855036, + "grad_norm": 0.1298828125, + "learning_rate": 0.001962805215329195, + "loss": 0.1426, + "step": 9725 + }, + { + "epoch": 0.08442635046570776, + "grad_norm": 0.671875, + "learning_rate": 0.001962796747079585, + "loss": 0.0854, + "step": 9726 + }, + { + "epoch": 0.08443503094591193, + "grad_norm": 0.36328125, + "learning_rate": 0.0019627882778864335, + "loss": 0.1348, + "step": 9727 + }, + { + "epoch": 0.08444371142611609, + "grad_norm": 0.1806640625, + "learning_rate": 0.00196277980774975, + "loss": 0.1221, + "step": 9728 + }, + { + "epoch": 0.08445239190632026, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019627713366695432, + "loss": 0.1128, + "step": 9729 + }, + { + "epoch": 0.08446107238652442, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019627628646458227, + "loss": 0.1128, + "step": 9730 + }, + { + "epoch": 0.08446975286672859, + "grad_norm": 0.271484375, + "learning_rate": 0.0019627543916785977, + "loss": 0.1367, + "step": 9731 + }, + { + "epoch": 0.08447843334693275, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019627459177678774, + "loss": 0.1118, + "step": 9732 + }, + { + "epoch": 0.08448711382713692, + "grad_norm": 0.26171875, + "learning_rate": 0.001962737442913671, + "loss": 0.1309, + "step": 9733 + }, + { + "epoch": 0.08449579430734108, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019627289671159884, + "loss": 0.1484, + "step": 9734 + }, + { + "epoch": 0.08450447478754525, + "grad_norm": 0.1162109375, + "learning_rate": 0.001962720490374838, + "loss": 0.127, + "step": 9735 + }, + { + "epoch": 0.08451315526774941, + "grad_norm": 1.1328125, + "learning_rate": 0.001962712012690229, + "loss": 0.1523, + "step": 9736 + }, + { + "epoch": 0.08452183574795358, + "grad_norm": 0.5625, + "learning_rate": 0.0019627035340621713, + "loss": 0.1582, + "step": 9737 + }, + { + "epoch": 0.08453051622815774, + "grad_norm": 0.306640625, + "learning_rate": 0.0019626950544906745, + "loss": 0.1338, + "step": 9738 + }, + { + "epoch": 0.08453919670836191, + "grad_norm": 0.6015625, + "learning_rate": 0.001962686573975747, + "loss": 0.1182, + "step": 9739 + }, + { + "epoch": 0.08454787718856607, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019626780925173985, + "loss": 0.1338, + "step": 9740 + }, + { + "epoch": 0.08455655766877024, + "grad_norm": 0.173828125, + "learning_rate": 0.0019626696101156384, + "loss": 0.1001, + "step": 9741 + }, + { + "epoch": 0.0845652381489744, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019626611267704753, + "loss": 0.0933, + "step": 9742 + }, + { + "epoch": 0.08457391862917857, + "grad_norm": 0.078125, + "learning_rate": 0.0019626526424819193, + "loss": 0.1187, + "step": 9743 + }, + { + "epoch": 0.08458259910938273, + "grad_norm": 0.421875, + "learning_rate": 0.0019626441572499795, + "loss": 0.1875, + "step": 9744 + }, + { + "epoch": 0.0845912795895869, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019626356710746645, + "loss": 0.127, + "step": 9745 + }, + { + "epoch": 0.08459996006979106, + "grad_norm": 0.130859375, + "learning_rate": 0.0019626271839559843, + "loss": 0.1055, + "step": 9746 + }, + { + "epoch": 0.08460864054999523, + "grad_norm": 0.115234375, + "learning_rate": 0.001962618695893948, + "loss": 0.168, + "step": 9747 + }, + { + "epoch": 0.0846173210301994, + "grad_norm": 0.263671875, + "learning_rate": 0.0019626102068885654, + "loss": 0.1152, + "step": 9748 + }, + { + "epoch": 0.08462600151040356, + "grad_norm": 0.37890625, + "learning_rate": 0.0019626017169398443, + "loss": 0.125, + "step": 9749 + }, + { + "epoch": 0.08463468199060772, + "grad_norm": 0.2734375, + "learning_rate": 0.0019625932260477954, + "loss": 0.167, + "step": 9750 + }, + { + "epoch": 0.08464336247081189, + "grad_norm": 0.15625, + "learning_rate": 0.0019625847342124277, + "loss": 0.1396, + "step": 9751 + }, + { + "epoch": 0.08465204295101605, + "grad_norm": 0.09619140625, + "learning_rate": 0.00196257624143375, + "loss": 0.0933, + "step": 9752 + }, + { + "epoch": 0.08466072343122022, + "grad_norm": 0.6171875, + "learning_rate": 0.0019625677477117715, + "loss": 0.1299, + "step": 9753 + }, + { + "epoch": 0.08466940391142438, + "grad_norm": 0.25390625, + "learning_rate": 0.0019625592530465026, + "loss": 0.1133, + "step": 9754 + }, + { + "epoch": 0.08467808439162855, + "grad_norm": 0.466796875, + "learning_rate": 0.0019625507574379517, + "loss": 0.1738, + "step": 9755 + }, + { + "epoch": 0.08468676487183271, + "grad_norm": 0.56640625, + "learning_rate": 0.0019625422608861277, + "loss": 0.1689, + "step": 9756 + }, + { + "epoch": 0.08469544535203688, + "grad_norm": 0.09033203125, + "learning_rate": 0.001962533763391041, + "loss": 0.1582, + "step": 9757 + }, + { + "epoch": 0.08470412583224105, + "grad_norm": 0.37109375, + "learning_rate": 0.0019625252649527, + "loss": 0.1172, + "step": 9758 + }, + { + "epoch": 0.08471280631244521, + "grad_norm": 1.25, + "learning_rate": 0.001962516765571115, + "loss": 0.1484, + "step": 9759 + }, + { + "epoch": 0.08472148679264938, + "grad_norm": 0.150390625, + "learning_rate": 0.001962508265246294, + "loss": 0.1377, + "step": 9760 + }, + { + "epoch": 0.08473016727285354, + "grad_norm": 0.48828125, + "learning_rate": 0.001962499763978247, + "loss": 0.1367, + "step": 9761 + }, + { + "epoch": 0.0847388477530577, + "grad_norm": 0.95703125, + "learning_rate": 0.0019624912617669833, + "loss": 0.1377, + "step": 9762 + }, + { + "epoch": 0.08474752823326187, + "grad_norm": 0.0859375, + "learning_rate": 0.001962482758612512, + "loss": 0.1523, + "step": 9763 + }, + { + "epoch": 0.08475620871346604, + "grad_norm": 0.67578125, + "learning_rate": 0.0019624742545148424, + "loss": 0.1514, + "step": 9764 + }, + { + "epoch": 0.08476488919367019, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019624657494739842, + "loss": 0.1357, + "step": 9765 + }, + { + "epoch": 0.08477356967387435, + "grad_norm": 0.169921875, + "learning_rate": 0.0019624572434899464, + "loss": 0.1143, + "step": 9766 + }, + { + "epoch": 0.08478225015407852, + "grad_norm": 0.07666015625, + "learning_rate": 0.001962448736562738, + "loss": 0.1582, + "step": 9767 + }, + { + "epoch": 0.08479093063428268, + "grad_norm": 0.09033203125, + "learning_rate": 0.001962440228692369, + "loss": 0.1138, + "step": 9768 + }, + { + "epoch": 0.08479961111448685, + "grad_norm": 0.416015625, + "learning_rate": 0.0019624317198788485, + "loss": 0.1084, + "step": 9769 + }, + { + "epoch": 0.08480829159469101, + "grad_norm": 0.142578125, + "learning_rate": 0.0019624232101221853, + "loss": 0.0938, + "step": 9770 + }, + { + "epoch": 0.08481697207489518, + "grad_norm": 0.130859375, + "learning_rate": 0.001962414699422389, + "loss": 0.1211, + "step": 9771 + }, + { + "epoch": 0.08482565255509934, + "grad_norm": 0.341796875, + "learning_rate": 0.001962406187779469, + "loss": 0.1162, + "step": 9772 + }, + { + "epoch": 0.08483433303530351, + "grad_norm": 0.2060546875, + "learning_rate": 0.001962397675193435, + "loss": 0.124, + "step": 9773 + }, + { + "epoch": 0.08484301351550767, + "grad_norm": 0.484375, + "learning_rate": 0.001962389161664295, + "loss": 0.1191, + "step": 9774 + }, + { + "epoch": 0.08485169399571184, + "grad_norm": 0.59375, + "learning_rate": 0.0019623806471920596, + "loss": 0.1328, + "step": 9775 + }, + { + "epoch": 0.084860374475916, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019623721317767375, + "loss": 0.1045, + "step": 9776 + }, + { + "epoch": 0.08486905495612017, + "grad_norm": 0.26171875, + "learning_rate": 0.001962363615418338, + "loss": 0.1191, + "step": 9777 + }, + { + "epoch": 0.08487773543632433, + "grad_norm": 0.115234375, + "learning_rate": 0.0019623550981168712, + "loss": 0.1289, + "step": 9778 + }, + { + "epoch": 0.0848864159165285, + "grad_norm": 0.28125, + "learning_rate": 0.0019623465798723457, + "loss": 0.0962, + "step": 9779 + }, + { + "epoch": 0.08489509639673266, + "grad_norm": 0.255859375, + "learning_rate": 0.001962338060684771, + "loss": 0.1143, + "step": 9780 + }, + { + "epoch": 0.08490377687693683, + "grad_norm": 0.1171875, + "learning_rate": 0.001962329540554156, + "loss": 0.127, + "step": 9781 + }, + { + "epoch": 0.084912457357141, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019623210194805105, + "loss": 0.1201, + "step": 9782 + }, + { + "epoch": 0.08492113783734516, + "grad_norm": 0.5546875, + "learning_rate": 0.001962312497463844, + "loss": 0.1621, + "step": 9783 + }, + { + "epoch": 0.08492981831754932, + "grad_norm": 0.1630859375, + "learning_rate": 0.001962303974504165, + "loss": 0.1367, + "step": 9784 + }, + { + "epoch": 0.08493849879775349, + "grad_norm": 0.453125, + "learning_rate": 0.001962295450601484, + "loss": 0.123, + "step": 9785 + }, + { + "epoch": 0.08494717927795765, + "grad_norm": 0.73828125, + "learning_rate": 0.001962286925755809, + "loss": 0.1416, + "step": 9786 + }, + { + "epoch": 0.08495585975816182, + "grad_norm": 0.3828125, + "learning_rate": 0.00196227839996715, + "loss": 0.1484, + "step": 9787 + }, + { + "epoch": 0.08496454023836598, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019622698732355164, + "loss": 0.1104, + "step": 9788 + }, + { + "epoch": 0.08497322071857015, + "grad_norm": 0.15625, + "learning_rate": 0.001962261345560918, + "loss": 0.1729, + "step": 9789 + }, + { + "epoch": 0.08498190119877432, + "grad_norm": 0.384765625, + "learning_rate": 0.0019622528169433626, + "loss": 0.1738, + "step": 9790 + }, + { + "epoch": 0.08499058167897848, + "grad_norm": 0.224609375, + "learning_rate": 0.001962244287382861, + "loss": 0.1201, + "step": 9791 + }, + { + "epoch": 0.08499926215918265, + "grad_norm": 0.328125, + "learning_rate": 0.001962235756879422, + "loss": 0.1187, + "step": 9792 + }, + { + "epoch": 0.08500794263938681, + "grad_norm": 0.1005859375, + "learning_rate": 0.001962227225433055, + "loss": 0.1157, + "step": 9793 + }, + { + "epoch": 0.08501662311959098, + "grad_norm": 0.33984375, + "learning_rate": 0.001962218693043769, + "loss": 0.1562, + "step": 9794 + }, + { + "epoch": 0.08502530359979514, + "grad_norm": 0.357421875, + "learning_rate": 0.0019622101597115736, + "loss": 0.0928, + "step": 9795 + }, + { + "epoch": 0.0850339840799993, + "grad_norm": 1.0234375, + "learning_rate": 0.0019622016254364784, + "loss": 0.1543, + "step": 9796 + }, + { + "epoch": 0.08504266456020347, + "grad_norm": 0.123046875, + "learning_rate": 0.0019621930902184925, + "loss": 0.127, + "step": 9797 + }, + { + "epoch": 0.08505134504040764, + "grad_norm": 0.396484375, + "learning_rate": 0.001962184554057625, + "loss": 0.1543, + "step": 9798 + }, + { + "epoch": 0.0850600255206118, + "grad_norm": 0.484375, + "learning_rate": 0.0019621760169538853, + "loss": 0.1426, + "step": 9799 + }, + { + "epoch": 0.08506870600081597, + "grad_norm": 0.1240234375, + "learning_rate": 0.001962167478907283, + "loss": 0.1221, + "step": 9800 + }, + { + "epoch": 0.08507738648102013, + "grad_norm": 0.314453125, + "learning_rate": 0.001962158939917827, + "loss": 0.1021, + "step": 9801 + }, + { + "epoch": 0.0850860669612243, + "grad_norm": 0.419921875, + "learning_rate": 0.0019621503999855275, + "loss": 0.1514, + "step": 9802 + }, + { + "epoch": 0.08509474744142846, + "grad_norm": 0.396484375, + "learning_rate": 0.001962141859110393, + "loss": 0.1182, + "step": 9803 + }, + { + "epoch": 0.08510342792163263, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019621333172924332, + "loss": 0.0933, + "step": 9804 + }, + { + "epoch": 0.08511210840183679, + "grad_norm": 0.69140625, + "learning_rate": 0.001962124774531657, + "loss": 0.127, + "step": 9805 + }, + { + "epoch": 0.08512078888204096, + "grad_norm": 0.419921875, + "learning_rate": 0.001962116230828075, + "loss": 0.1187, + "step": 9806 + }, + { + "epoch": 0.08512946936224512, + "grad_norm": 0.263671875, + "learning_rate": 0.0019621076861816946, + "loss": 0.2168, + "step": 9807 + }, + { + "epoch": 0.08513814984244929, + "grad_norm": 0.2177734375, + "learning_rate": 0.001962099140592527, + "loss": 0.1338, + "step": 9808 + }, + { + "epoch": 0.08514683032265345, + "grad_norm": 0.50390625, + "learning_rate": 0.00196209059406058, + "loss": 0.125, + "step": 9809 + }, + { + "epoch": 0.08515551080285762, + "grad_norm": 0.494140625, + "learning_rate": 0.001962082046585864, + "loss": 0.1543, + "step": 9810 + }, + { + "epoch": 0.08516419128306178, + "grad_norm": 0.28515625, + "learning_rate": 0.001962073498168388, + "loss": 0.1299, + "step": 9811 + }, + { + "epoch": 0.08517287176326595, + "grad_norm": 0.26953125, + "learning_rate": 0.0019620649488081613, + "loss": 0.1455, + "step": 9812 + }, + { + "epoch": 0.08518155224347011, + "grad_norm": 0.1513671875, + "learning_rate": 0.001962056398505194, + "loss": 0.1201, + "step": 9813 + }, + { + "epoch": 0.08519023272367428, + "grad_norm": 0.6640625, + "learning_rate": 0.0019620478472594936, + "loss": 0.1602, + "step": 9814 + }, + { + "epoch": 0.08519891320387844, + "grad_norm": 0.91015625, + "learning_rate": 0.0019620392950710714, + "loss": 0.1836, + "step": 9815 + }, + { + "epoch": 0.08520759368408261, + "grad_norm": 0.32421875, + "learning_rate": 0.0019620307419399357, + "loss": 0.1079, + "step": 9816 + }, + { + "epoch": 0.08521627416428677, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019620221878660963, + "loss": 0.1299, + "step": 9817 + }, + { + "epoch": 0.08522495464449094, + "grad_norm": 0.255859375, + "learning_rate": 0.0019620136328495625, + "loss": 0.1025, + "step": 9818 + }, + { + "epoch": 0.0852336351246951, + "grad_norm": 0.138671875, + "learning_rate": 0.0019620050768903437, + "loss": 0.1475, + "step": 9819 + }, + { + "epoch": 0.08524231560489927, + "grad_norm": 0.494140625, + "learning_rate": 0.0019619965199884483, + "loss": 0.0942, + "step": 9820 + }, + { + "epoch": 0.08525099608510343, + "grad_norm": 0.408203125, + "learning_rate": 0.0019619879621438872, + "loss": 0.1406, + "step": 9821 + }, + { + "epoch": 0.0852596765653076, + "grad_norm": 0.2275390625, + "learning_rate": 0.001961979403356669, + "loss": 0.106, + "step": 9822 + }, + { + "epoch": 0.08526835704551176, + "grad_norm": 0.435546875, + "learning_rate": 0.0019619708436268025, + "loss": 0.1855, + "step": 9823 + }, + { + "epoch": 0.08527703752571593, + "grad_norm": 0.1435546875, + "learning_rate": 0.001961962282954298, + "loss": 0.1191, + "step": 9824 + }, + { + "epoch": 0.0852857180059201, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019619537213391644, + "loss": 0.124, + "step": 9825 + }, + { + "epoch": 0.08529439848612425, + "grad_norm": 0.470703125, + "learning_rate": 0.0019619451587814113, + "loss": 0.1309, + "step": 9826 + }, + { + "epoch": 0.08530307896632841, + "grad_norm": 0.490234375, + "learning_rate": 0.0019619365952810476, + "loss": 0.1143, + "step": 9827 + }, + { + "epoch": 0.08531175944653258, + "grad_norm": 0.201171875, + "learning_rate": 0.0019619280308380838, + "loss": 0.0967, + "step": 9828 + }, + { + "epoch": 0.08532043992673674, + "grad_norm": 0.318359375, + "learning_rate": 0.0019619194654525276, + "loss": 0.1562, + "step": 9829 + }, + { + "epoch": 0.0853291204069409, + "grad_norm": 0.87109375, + "learning_rate": 0.00196191089912439, + "loss": 0.1211, + "step": 9830 + }, + { + "epoch": 0.08533780088714507, + "grad_norm": 0.2890625, + "learning_rate": 0.001961902331853679, + "loss": 0.0898, + "step": 9831 + }, + { + "epoch": 0.08534648136734924, + "grad_norm": 0.41796875, + "learning_rate": 0.0019618937636404045, + "loss": 0.1113, + "step": 9832 + }, + { + "epoch": 0.0853551618475534, + "grad_norm": 0.2255859375, + "learning_rate": 0.001961885194484576, + "loss": 0.1035, + "step": 9833 + }, + { + "epoch": 0.08536384232775757, + "grad_norm": 0.123046875, + "learning_rate": 0.001961876624386203, + "loss": 0.106, + "step": 9834 + }, + { + "epoch": 0.08537252280796173, + "grad_norm": 0.11767578125, + "learning_rate": 0.001961868053345294, + "loss": 0.1162, + "step": 9835 + }, + { + "epoch": 0.0853812032881659, + "grad_norm": 0.1298828125, + "learning_rate": 0.00196185948136186, + "loss": 0.1118, + "step": 9836 + }, + { + "epoch": 0.08538988376837006, + "grad_norm": 0.2021484375, + "learning_rate": 0.001961850908435909, + "loss": 0.1387, + "step": 9837 + }, + { + "epoch": 0.08539856424857423, + "grad_norm": 0.36328125, + "learning_rate": 0.0019618423345674507, + "loss": 0.2539, + "step": 9838 + }, + { + "epoch": 0.08540724472877839, + "grad_norm": 0.27734375, + "learning_rate": 0.0019618337597564948, + "loss": 0.1172, + "step": 9839 + }, + { + "epoch": 0.08541592520898256, + "grad_norm": 1.296875, + "learning_rate": 0.0019618251840030506, + "loss": 0.1387, + "step": 9840 + }, + { + "epoch": 0.08542460568918672, + "grad_norm": 0.921875, + "learning_rate": 0.0019618166073071272, + "loss": 0.1309, + "step": 9841 + }, + { + "epoch": 0.08543328616939089, + "grad_norm": 0.37109375, + "learning_rate": 0.001961808029668734, + "loss": 0.1201, + "step": 9842 + }, + { + "epoch": 0.08544196664959505, + "grad_norm": 0.271484375, + "learning_rate": 0.0019617994510878803, + "loss": 0.1348, + "step": 9843 + }, + { + "epoch": 0.08545064712979922, + "grad_norm": 1.0078125, + "learning_rate": 0.001961790871564576, + "loss": 0.1055, + "step": 9844 + }, + { + "epoch": 0.08545932761000338, + "grad_norm": 0.05908203125, + "learning_rate": 0.00196178229109883, + "loss": 0.0942, + "step": 9845 + }, + { + "epoch": 0.08546800809020755, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019617737096906517, + "loss": 0.0938, + "step": 9846 + }, + { + "epoch": 0.08547668857041171, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019617651273400507, + "loss": 0.1201, + "step": 9847 + }, + { + "epoch": 0.08548536905061588, + "grad_norm": 0.6875, + "learning_rate": 0.0019617565440470364, + "loss": 0.1523, + "step": 9848 + }, + { + "epoch": 0.08549404953082004, + "grad_norm": 0.26953125, + "learning_rate": 0.001961747959811618, + "loss": 0.1201, + "step": 9849 + }, + { + "epoch": 0.08550273001102421, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019617393746338054, + "loss": 0.1152, + "step": 9850 + }, + { + "epoch": 0.08551141049122837, + "grad_norm": 0.765625, + "learning_rate": 0.001961730788513607, + "loss": 0.1348, + "step": 9851 + }, + { + "epoch": 0.08552009097143254, + "grad_norm": 0.1396484375, + "learning_rate": 0.001961722201451033, + "loss": 0.1504, + "step": 9852 + }, + { + "epoch": 0.0855287714516367, + "grad_norm": 0.86328125, + "learning_rate": 0.0019617136134460924, + "loss": 0.1069, + "step": 9853 + }, + { + "epoch": 0.08553745193184087, + "grad_norm": 0.314453125, + "learning_rate": 0.0019617050244987947, + "loss": 0.0874, + "step": 9854 + }, + { + "epoch": 0.08554613241204503, + "grad_norm": 1.3515625, + "learning_rate": 0.0019616964346091496, + "loss": 0.1738, + "step": 9855 + }, + { + "epoch": 0.0855548128922492, + "grad_norm": 0.26171875, + "learning_rate": 0.0019616878437771663, + "loss": 0.1035, + "step": 9856 + }, + { + "epoch": 0.08556349337245336, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019616792520028535, + "loss": 0.1318, + "step": 9857 + }, + { + "epoch": 0.08557217385265753, + "grad_norm": 0.11181640625, + "learning_rate": 0.001961670659286222, + "loss": 0.1138, + "step": 9858 + }, + { + "epoch": 0.0855808543328617, + "grad_norm": 0.2275390625, + "learning_rate": 0.00196166206562728, + "loss": 0.2227, + "step": 9859 + }, + { + "epoch": 0.08558953481306586, + "grad_norm": 0.416015625, + "learning_rate": 0.001961653471026037, + "loss": 0.1113, + "step": 9860 + }, + { + "epoch": 0.08559821529327002, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019616448754825036, + "loss": 0.126, + "step": 9861 + }, + { + "epoch": 0.08560689577347419, + "grad_norm": 0.94921875, + "learning_rate": 0.0019616362789966874, + "loss": 0.1104, + "step": 9862 + }, + { + "epoch": 0.08561557625367835, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019616276815685994, + "loss": 0.1338, + "step": 9863 + }, + { + "epoch": 0.08562425673388252, + "grad_norm": 0.431640625, + "learning_rate": 0.0019616190831982477, + "loss": 0.1138, + "step": 9864 + }, + { + "epoch": 0.08563293721408669, + "grad_norm": 0.44140625, + "learning_rate": 0.001961610483885643, + "loss": 0.1196, + "step": 9865 + }, + { + "epoch": 0.08564161769429085, + "grad_norm": 0.447265625, + "learning_rate": 0.0019616018836307934, + "loss": 0.1074, + "step": 9866 + }, + { + "epoch": 0.08565029817449502, + "grad_norm": 0.125, + "learning_rate": 0.0019615932824337094, + "loss": 0.1069, + "step": 9867 + }, + { + "epoch": 0.08565897865469918, + "grad_norm": 0.326171875, + "learning_rate": 0.0019615846802943995, + "loss": 0.1055, + "step": 9868 + }, + { + "epoch": 0.08566765913490335, + "grad_norm": 0.5390625, + "learning_rate": 0.0019615760772128737, + "loss": 0.1035, + "step": 9869 + }, + { + "epoch": 0.08567633961510751, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019615674731891415, + "loss": 0.1719, + "step": 9870 + }, + { + "epoch": 0.08568502009531168, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019615588682232116, + "loss": 0.1299, + "step": 9871 + }, + { + "epoch": 0.08569370057551584, + "grad_norm": 0.90234375, + "learning_rate": 0.0019615502623150944, + "loss": 0.1562, + "step": 9872 + }, + { + "epoch": 0.08570238105572, + "grad_norm": 0.298828125, + "learning_rate": 0.001961541655464798, + "loss": 0.1201, + "step": 9873 + }, + { + "epoch": 0.08571106153592417, + "grad_norm": 0.8125, + "learning_rate": 0.0019615330476723336, + "loss": 0.1182, + "step": 9874 + }, + { + "epoch": 0.08571974201612834, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019615244389377087, + "loss": 0.1318, + "step": 9875 + }, + { + "epoch": 0.0857284224963325, + "grad_norm": 0.30078125, + "learning_rate": 0.0019615158292609343, + "loss": 0.0923, + "step": 9876 + }, + { + "epoch": 0.08573710297653667, + "grad_norm": 0.326171875, + "learning_rate": 0.0019615072186420185, + "loss": 0.1138, + "step": 9877 + }, + { + "epoch": 0.08574578345674083, + "grad_norm": 0.5390625, + "learning_rate": 0.0019614986070809718, + "loss": 0.165, + "step": 9878 + }, + { + "epoch": 0.085754463936945, + "grad_norm": 0.470703125, + "learning_rate": 0.001961489994577803, + "loss": 0.125, + "step": 9879 + }, + { + "epoch": 0.08576314441714916, + "grad_norm": 0.61328125, + "learning_rate": 0.001961481381132522, + "loss": 0.1367, + "step": 9880 + }, + { + "epoch": 0.08577182489735333, + "grad_norm": 0.099609375, + "learning_rate": 0.0019614727667451376, + "loss": 0.1211, + "step": 9881 + }, + { + "epoch": 0.08578050537755749, + "grad_norm": 0.408203125, + "learning_rate": 0.0019614641514156595, + "loss": 0.1279, + "step": 9882 + }, + { + "epoch": 0.08578918585776166, + "grad_norm": 0.28515625, + "learning_rate": 0.0019614555351440974, + "loss": 0.1074, + "step": 9883 + }, + { + "epoch": 0.08579786633796582, + "grad_norm": 0.30078125, + "learning_rate": 0.0019614469179304603, + "loss": 0.1328, + "step": 9884 + }, + { + "epoch": 0.08580654681816999, + "grad_norm": 0.75390625, + "learning_rate": 0.0019614382997747573, + "loss": 0.1118, + "step": 9885 + }, + { + "epoch": 0.08581522729837415, + "grad_norm": 0.65625, + "learning_rate": 0.001961429680676999, + "loss": 0.1187, + "step": 9886 + }, + { + "epoch": 0.08582390777857832, + "grad_norm": 0.6484375, + "learning_rate": 0.001961421060637194, + "loss": 0.1182, + "step": 9887 + }, + { + "epoch": 0.08583258825878247, + "grad_norm": 0.337890625, + "learning_rate": 0.0019614124396553517, + "loss": 0.1177, + "step": 9888 + }, + { + "epoch": 0.08584126873898663, + "grad_norm": 0.56640625, + "learning_rate": 0.001961403817731482, + "loss": 0.126, + "step": 9889 + }, + { + "epoch": 0.0858499492191908, + "grad_norm": 0.365234375, + "learning_rate": 0.0019613951948655936, + "loss": 0.1602, + "step": 9890 + }, + { + "epoch": 0.08585862969939496, + "grad_norm": 1.109375, + "learning_rate": 0.0019613865710576966, + "loss": 0.1484, + "step": 9891 + }, + { + "epoch": 0.08586731017959913, + "grad_norm": 0.287109375, + "learning_rate": 0.0019613779463078006, + "loss": 0.124, + "step": 9892 + }, + { + "epoch": 0.0858759906598033, + "grad_norm": 0.1015625, + "learning_rate": 0.001961369320615914, + "loss": 0.1172, + "step": 9893 + }, + { + "epoch": 0.08588467114000746, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019613606939820473, + "loss": 0.1162, + "step": 9894 + }, + { + "epoch": 0.08589335162021162, + "grad_norm": 0.279296875, + "learning_rate": 0.001961352066406209, + "loss": 0.123, + "step": 9895 + }, + { + "epoch": 0.08590203210041579, + "grad_norm": 0.78125, + "learning_rate": 0.0019613434378884095, + "loss": 0.1074, + "step": 9896 + }, + { + "epoch": 0.08591071258061996, + "grad_norm": 1.2265625, + "learning_rate": 0.0019613348084286573, + "loss": 0.1699, + "step": 9897 + }, + { + "epoch": 0.08591939306082412, + "grad_norm": 0.40234375, + "learning_rate": 0.0019613261780269627, + "loss": 0.1152, + "step": 9898 + }, + { + "epoch": 0.08592807354102829, + "grad_norm": 0.1083984375, + "learning_rate": 0.001961317546683334, + "loss": 0.1152, + "step": 9899 + }, + { + "epoch": 0.08593675402123245, + "grad_norm": 0.5390625, + "learning_rate": 0.0019613089143977825, + "loss": 0.166, + "step": 9900 + }, + { + "epoch": 0.08594543450143662, + "grad_norm": 0.36328125, + "learning_rate": 0.001961300281170316, + "loss": 0.125, + "step": 9901 + }, + { + "epoch": 0.08595411498164078, + "grad_norm": 0.08642578125, + "learning_rate": 0.001961291647000944, + "loss": 0.1318, + "step": 9902 + }, + { + "epoch": 0.08596279546184495, + "grad_norm": 0.158203125, + "learning_rate": 0.001961283011889677, + "loss": 0.1055, + "step": 9903 + }, + { + "epoch": 0.08597147594204911, + "grad_norm": 0.115234375, + "learning_rate": 0.0019612743758365236, + "loss": 0.126, + "step": 9904 + }, + { + "epoch": 0.08598015642225328, + "grad_norm": 0.263671875, + "learning_rate": 0.0019612657388414934, + "loss": 0.1348, + "step": 9905 + }, + { + "epoch": 0.08598883690245744, + "grad_norm": 0.48046875, + "learning_rate": 0.0019612571009045964, + "loss": 0.1406, + "step": 9906 + }, + { + "epoch": 0.0859975173826616, + "grad_norm": 0.07666015625, + "learning_rate": 0.001961248462025841, + "loss": 0.083, + "step": 9907 + }, + { + "epoch": 0.08600619786286577, + "grad_norm": 0.08203125, + "learning_rate": 0.0019612398222052375, + "loss": 0.1396, + "step": 9908 + }, + { + "epoch": 0.08601487834306994, + "grad_norm": 0.71484375, + "learning_rate": 0.001961231181442795, + "loss": 0.1406, + "step": 9909 + }, + { + "epoch": 0.0860235588232741, + "grad_norm": 0.451171875, + "learning_rate": 0.001961222539738523, + "loss": 0.1836, + "step": 9910 + }, + { + "epoch": 0.08603223930347827, + "grad_norm": 0.34765625, + "learning_rate": 0.001961213897092431, + "loss": 0.1123, + "step": 9911 + }, + { + "epoch": 0.08604091978368243, + "grad_norm": 0.4765625, + "learning_rate": 0.0019612052535045283, + "loss": 0.1055, + "step": 9912 + }, + { + "epoch": 0.0860496002638866, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019611966089748247, + "loss": 0.1318, + "step": 9913 + }, + { + "epoch": 0.08605828074409076, + "grad_norm": 0.333984375, + "learning_rate": 0.0019611879635033293, + "loss": 0.0981, + "step": 9914 + }, + { + "epoch": 0.08606696122429493, + "grad_norm": 0.291015625, + "learning_rate": 0.0019611793170900517, + "loss": 0.0918, + "step": 9915 + }, + { + "epoch": 0.08607564170449909, + "grad_norm": 0.49609375, + "learning_rate": 0.0019611706697350013, + "loss": 0.0928, + "step": 9916 + }, + { + "epoch": 0.08608432218470326, + "grad_norm": 1.1328125, + "learning_rate": 0.001961162021438188, + "loss": 0.1738, + "step": 9917 + }, + { + "epoch": 0.08609300266490742, + "grad_norm": 0.361328125, + "learning_rate": 0.00196115337219962, + "loss": 0.0938, + "step": 9918 + }, + { + "epoch": 0.08610168314511159, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019611447220193082, + "loss": 0.1318, + "step": 9919 + }, + { + "epoch": 0.08611036362531575, + "grad_norm": 0.12451171875, + "learning_rate": 0.001961136070897261, + "loss": 0.1191, + "step": 9920 + }, + { + "epoch": 0.08611904410551992, + "grad_norm": 0.240234375, + "learning_rate": 0.001961127418833489, + "loss": 0.1309, + "step": 9921 + }, + { + "epoch": 0.08612772458572408, + "grad_norm": 0.330078125, + "learning_rate": 0.001961118765828001, + "loss": 0.1367, + "step": 9922 + }, + { + "epoch": 0.08613640506592825, + "grad_norm": 1.625, + "learning_rate": 0.001961110111880806, + "loss": 0.1357, + "step": 9923 + }, + { + "epoch": 0.08614508554613241, + "grad_norm": 0.2109375, + "learning_rate": 0.001961101456991914, + "loss": 0.1201, + "step": 9924 + }, + { + "epoch": 0.08615376602633658, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019610928011613345, + "loss": 0.1387, + "step": 9925 + }, + { + "epoch": 0.08616244650654074, + "grad_norm": 0.294921875, + "learning_rate": 0.0019610841443890767, + "loss": 0.1328, + "step": 9926 + }, + { + "epoch": 0.08617112698674491, + "grad_norm": 0.3828125, + "learning_rate": 0.0019610754866751503, + "loss": 0.1348, + "step": 9927 + }, + { + "epoch": 0.08617980746694907, + "grad_norm": 0.3515625, + "learning_rate": 0.001961066828019565, + "loss": 0.0918, + "step": 9928 + }, + { + "epoch": 0.08618848794715324, + "grad_norm": 0.244140625, + "learning_rate": 0.00196105816842233, + "loss": 0.1245, + "step": 9929 + }, + { + "epoch": 0.0861971684273574, + "grad_norm": 1.3515625, + "learning_rate": 0.001961049507883454, + "loss": 0.1328, + "step": 9930 + }, + { + "epoch": 0.08620584890756157, + "grad_norm": 0.2578125, + "learning_rate": 0.0019610408464029476, + "loss": 0.1553, + "step": 9931 + }, + { + "epoch": 0.08621452938776573, + "grad_norm": 0.224609375, + "learning_rate": 0.0019610321839808194, + "loss": 0.1357, + "step": 9932 + }, + { + "epoch": 0.0862232098679699, + "grad_norm": 0.298828125, + "learning_rate": 0.00196102352061708, + "loss": 0.1035, + "step": 9933 + }, + { + "epoch": 0.08623189034817406, + "grad_norm": 0.2373046875, + "learning_rate": 0.001961014856311738, + "loss": 0.1318, + "step": 9934 + }, + { + "epoch": 0.08624057082837823, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019610061910648032, + "loss": 0.0986, + "step": 9935 + }, + { + "epoch": 0.0862492513085824, + "grad_norm": 0.392578125, + "learning_rate": 0.0019609975248762847, + "loss": 0.1455, + "step": 9936 + }, + { + "epoch": 0.08625793178878656, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019609888577461925, + "loss": 0.1113, + "step": 9937 + }, + { + "epoch": 0.08626661226899072, + "grad_norm": 0.08203125, + "learning_rate": 0.001960980189674536, + "loss": 0.0894, + "step": 9938 + }, + { + "epoch": 0.08627529274919489, + "grad_norm": 0.12890625, + "learning_rate": 0.001960971520661324, + "loss": 0.125, + "step": 9939 + }, + { + "epoch": 0.08628397322939906, + "grad_norm": 0.490234375, + "learning_rate": 0.001960962850706567, + "loss": 0.1436, + "step": 9940 + }, + { + "epoch": 0.08629265370960322, + "grad_norm": 0.33984375, + "learning_rate": 0.0019609541798102734, + "loss": 0.1543, + "step": 9941 + }, + { + "epoch": 0.08630133418980739, + "grad_norm": 0.2578125, + "learning_rate": 0.0019609455079724536, + "loss": 0.1177, + "step": 9942 + }, + { + "epoch": 0.08631001467001155, + "grad_norm": 0.3984375, + "learning_rate": 0.0019609368351931164, + "loss": 0.1104, + "step": 9943 + }, + { + "epoch": 0.08631869515021572, + "grad_norm": 0.158203125, + "learning_rate": 0.001960928161472272, + "loss": 0.1201, + "step": 9944 + }, + { + "epoch": 0.08632737563041988, + "grad_norm": 0.37890625, + "learning_rate": 0.0019609194868099294, + "loss": 0.1416, + "step": 9945 + }, + { + "epoch": 0.08633605611062405, + "grad_norm": 0.314453125, + "learning_rate": 0.001960910811206098, + "loss": 0.126, + "step": 9946 + }, + { + "epoch": 0.08634473659082821, + "grad_norm": 0.298828125, + "learning_rate": 0.0019609021346607878, + "loss": 0.1147, + "step": 9947 + }, + { + "epoch": 0.08635341707103238, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019608934571740074, + "loss": 0.1377, + "step": 9948 + }, + { + "epoch": 0.08636209755123654, + "grad_norm": 0.328125, + "learning_rate": 0.0019608847787457675, + "loss": 0.1143, + "step": 9949 + }, + { + "epoch": 0.08637077803144069, + "grad_norm": 0.15625, + "learning_rate": 0.0019608760993760767, + "loss": 0.1016, + "step": 9950 + }, + { + "epoch": 0.08637945851164486, + "grad_norm": 0.8125, + "learning_rate": 0.001960867419064945, + "loss": 0.1123, + "step": 9951 + }, + { + "epoch": 0.08638813899184902, + "grad_norm": 0.46484375, + "learning_rate": 0.001960858737812381, + "loss": 0.1162, + "step": 9952 + }, + { + "epoch": 0.08639681947205319, + "grad_norm": 0.5234375, + "learning_rate": 0.0019608500556183954, + "loss": 0.124, + "step": 9953 + }, + { + "epoch": 0.08640549995225735, + "grad_norm": 0.94140625, + "learning_rate": 0.0019608413724829963, + "loss": 0.0986, + "step": 9954 + }, + { + "epoch": 0.08641418043246152, + "grad_norm": 0.1630859375, + "learning_rate": 0.001960832688406195, + "loss": 0.0957, + "step": 9955 + }, + { + "epoch": 0.08642286091266568, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019608240033879994, + "loss": 0.124, + "step": 9956 + }, + { + "epoch": 0.08643154139286985, + "grad_norm": 0.4765625, + "learning_rate": 0.00196081531742842, + "loss": 0.1289, + "step": 9957 + }, + { + "epoch": 0.08644022187307401, + "grad_norm": 0.208984375, + "learning_rate": 0.001960806630527465, + "loss": 0.1191, + "step": 9958 + }, + { + "epoch": 0.08644890235327818, + "grad_norm": 0.267578125, + "learning_rate": 0.0019607979426851455, + "loss": 0.127, + "step": 9959 + }, + { + "epoch": 0.08645758283348234, + "grad_norm": 0.478515625, + "learning_rate": 0.0019607892539014707, + "loss": 0.1152, + "step": 9960 + }, + { + "epoch": 0.08646626331368651, + "grad_norm": 0.392578125, + "learning_rate": 0.0019607805641764496, + "loss": 0.1064, + "step": 9961 + }, + { + "epoch": 0.08647494379389067, + "grad_norm": 0.2001953125, + "learning_rate": 0.001960771873510091, + "loss": 0.0996, + "step": 9962 + }, + { + "epoch": 0.08648362427409484, + "grad_norm": 0.50390625, + "learning_rate": 0.001960763181902406, + "loss": 0.125, + "step": 9963 + }, + { + "epoch": 0.086492304754299, + "grad_norm": 1.0234375, + "learning_rate": 0.0019607544893534027, + "loss": 0.1934, + "step": 9964 + }, + { + "epoch": 0.08650098523450317, + "grad_norm": 0.578125, + "learning_rate": 0.001960745795863092, + "loss": 0.1094, + "step": 9965 + }, + { + "epoch": 0.08650966571470733, + "grad_norm": 1.234375, + "learning_rate": 0.001960737101431482, + "loss": 0.1602, + "step": 9966 + }, + { + "epoch": 0.0865183461949115, + "grad_norm": 0.51953125, + "learning_rate": 0.0019607284060585827, + "loss": 0.085, + "step": 9967 + }, + { + "epoch": 0.08652702667511566, + "grad_norm": 0.67578125, + "learning_rate": 0.001960719709744404, + "loss": 0.1621, + "step": 9968 + }, + { + "epoch": 0.08653570715531983, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019607110124889556, + "loss": 0.1133, + "step": 9969 + }, + { + "epoch": 0.086544387635524, + "grad_norm": 0.28515625, + "learning_rate": 0.0019607023142922462, + "loss": 0.1084, + "step": 9970 + }, + { + "epoch": 0.08655306811572816, + "grad_norm": 0.85546875, + "learning_rate": 0.0019606936151542855, + "loss": 0.1035, + "step": 9971 + }, + { + "epoch": 0.08656174859593233, + "grad_norm": 0.6015625, + "learning_rate": 0.0019606849150750833, + "loss": 0.1445, + "step": 9972 + }, + { + "epoch": 0.08657042907613649, + "grad_norm": 0.45703125, + "learning_rate": 0.001960676214054649, + "loss": 0.1289, + "step": 9973 + }, + { + "epoch": 0.08657910955634066, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019606675120929923, + "loss": 0.127, + "step": 9974 + }, + { + "epoch": 0.08658779003654482, + "grad_norm": 0.78515625, + "learning_rate": 0.0019606588091901226, + "loss": 0.1504, + "step": 9975 + }, + { + "epoch": 0.08659647051674899, + "grad_norm": 0.2578125, + "learning_rate": 0.001960650105346049, + "loss": 0.1846, + "step": 9976 + }, + { + "epoch": 0.08660515099695315, + "grad_norm": 0.396484375, + "learning_rate": 0.001960641400560782, + "loss": 0.0884, + "step": 9977 + }, + { + "epoch": 0.08661383147715732, + "grad_norm": 0.357421875, + "learning_rate": 0.0019606326948343297, + "loss": 0.0986, + "step": 9978 + }, + { + "epoch": 0.08662251195736148, + "grad_norm": 0.5703125, + "learning_rate": 0.001960623988166703, + "loss": 0.1279, + "step": 9979 + }, + { + "epoch": 0.08663119243756565, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019606152805579104, + "loss": 0.126, + "step": 9980 + }, + { + "epoch": 0.08663987291776981, + "grad_norm": 0.30078125, + "learning_rate": 0.001960606572007962, + "loss": 0.1357, + "step": 9981 + }, + { + "epoch": 0.08664855339797398, + "grad_norm": 0.234375, + "learning_rate": 0.001960597862516867, + "loss": 0.125, + "step": 9982 + }, + { + "epoch": 0.08665723387817814, + "grad_norm": 0.74609375, + "learning_rate": 0.0019605891520846357, + "loss": 0.1309, + "step": 9983 + }, + { + "epoch": 0.0866659143583823, + "grad_norm": 0.326171875, + "learning_rate": 0.0019605804407112765, + "loss": 0.1201, + "step": 9984 + }, + { + "epoch": 0.08667459483858647, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019605717283968003, + "loss": 0.0752, + "step": 9985 + }, + { + "epoch": 0.08668327531879064, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019605630151412145, + "loss": 0.1309, + "step": 9986 + }, + { + "epoch": 0.0866919557989948, + "grad_norm": 0.09375, + "learning_rate": 0.001960554300944531, + "loss": 0.1328, + "step": 9987 + }, + { + "epoch": 0.08670063627919897, + "grad_norm": 0.1484375, + "learning_rate": 0.0019605455858067574, + "loss": 0.1318, + "step": 9988 + }, + { + "epoch": 0.08670931675940313, + "grad_norm": 0.21484375, + "learning_rate": 0.001960536869727905, + "loss": 0.1074, + "step": 9989 + }, + { + "epoch": 0.0867179972396073, + "grad_norm": 0.61328125, + "learning_rate": 0.0019605281527079813, + "loss": 0.123, + "step": 9990 + }, + { + "epoch": 0.08672667771981146, + "grad_norm": 0.63671875, + "learning_rate": 0.0019605194347469975, + "loss": 0.1387, + "step": 9991 + }, + { + "epoch": 0.08673535820001563, + "grad_norm": 0.125, + "learning_rate": 0.0019605107158449627, + "loss": 0.1348, + "step": 9992 + }, + { + "epoch": 0.08674403868021979, + "grad_norm": 0.4765625, + "learning_rate": 0.0019605019960018863, + "loss": 0.1719, + "step": 9993 + }, + { + "epoch": 0.08675271916042396, + "grad_norm": 0.318359375, + "learning_rate": 0.001960493275217778, + "loss": 0.123, + "step": 9994 + }, + { + "epoch": 0.08676139964062812, + "grad_norm": 0.08447265625, + "learning_rate": 0.001960484553492647, + "loss": 0.1328, + "step": 9995 + }, + { + "epoch": 0.08677008012083229, + "grad_norm": 0.07177734375, + "learning_rate": 0.001960475830826503, + "loss": 0.1157, + "step": 9996 + }, + { + "epoch": 0.08677876060103645, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019604671072193554, + "loss": 0.1123, + "step": 9997 + }, + { + "epoch": 0.08678744108124062, + "grad_norm": 0.77734375, + "learning_rate": 0.0019604583826712137, + "loss": 0.1118, + "step": 9998 + }, + { + "epoch": 0.08679612156144478, + "grad_norm": 0.1015625, + "learning_rate": 0.0019604496571820882, + "loss": 0.1357, + "step": 9999 + }, + { + "epoch": 0.08680480204164895, + "grad_norm": 0.333984375, + "learning_rate": 0.0019604409307519876, + "loss": 0.1104, + "step": 10000 + }, + { + "epoch": 0.08681348252185311, + "grad_norm": 0.625, + "learning_rate": 0.0019604322033809217, + "loss": 0.103, + "step": 10001 + }, + { + "epoch": 0.08682216300205728, + "grad_norm": 0.2734375, + "learning_rate": 0.0019604234750689002, + "loss": 0.1182, + "step": 10002 + }, + { + "epoch": 0.08683084348226144, + "grad_norm": 0.37109375, + "learning_rate": 0.0019604147458159326, + "loss": 0.1797, + "step": 10003 + }, + { + "epoch": 0.08683952396246561, + "grad_norm": 0.205078125, + "learning_rate": 0.0019604060156220285, + "loss": 0.1191, + "step": 10004 + }, + { + "epoch": 0.08684820444266977, + "grad_norm": 0.48828125, + "learning_rate": 0.001960397284487197, + "loss": 0.1187, + "step": 10005 + }, + { + "epoch": 0.08685688492287394, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019603885524114477, + "loss": 0.1084, + "step": 10006 + }, + { + "epoch": 0.0868655654030781, + "grad_norm": 0.25, + "learning_rate": 0.0019603798193947907, + "loss": 0.1177, + "step": 10007 + }, + { + "epoch": 0.08687424588328227, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019603710854372352, + "loss": 0.1689, + "step": 10008 + }, + { + "epoch": 0.08688292636348643, + "grad_norm": 0.921875, + "learning_rate": 0.001960362350538791, + "loss": 0.1426, + "step": 10009 + }, + { + "epoch": 0.0868916068436906, + "grad_norm": 0.3046875, + "learning_rate": 0.0019603536146994673, + "loss": 0.1211, + "step": 10010 + }, + { + "epoch": 0.08690028732389475, + "grad_norm": 2.4375, + "learning_rate": 0.001960344877919274, + "loss": 0.1406, + "step": 10011 + }, + { + "epoch": 0.08690896780409892, + "grad_norm": 0.23046875, + "learning_rate": 0.0019603361401982202, + "loss": 0.0923, + "step": 10012 + }, + { + "epoch": 0.08691764828430308, + "grad_norm": 0.1435546875, + "learning_rate": 0.001960327401536316, + "loss": 0.1309, + "step": 10013 + }, + { + "epoch": 0.08692632876450725, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019603186619335704, + "loss": 0.1523, + "step": 10014 + }, + { + "epoch": 0.08693500924471141, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019603099213899933, + "loss": 0.127, + "step": 10015 + }, + { + "epoch": 0.08694368972491558, + "grad_norm": 0.267578125, + "learning_rate": 0.001960301179905594, + "loss": 0.1235, + "step": 10016 + }, + { + "epoch": 0.08695237020511974, + "grad_norm": 0.52734375, + "learning_rate": 0.001960292437480383, + "loss": 0.1113, + "step": 10017 + }, + { + "epoch": 0.0869610506853239, + "grad_norm": 0.306640625, + "learning_rate": 0.0019602836941143686, + "loss": 0.123, + "step": 10018 + }, + { + "epoch": 0.08696973116552807, + "grad_norm": 0.66015625, + "learning_rate": 0.0019602749498075604, + "loss": 0.1963, + "step": 10019 + }, + { + "epoch": 0.08697841164573224, + "grad_norm": 0.17578125, + "learning_rate": 0.0019602662045599692, + "loss": 0.1074, + "step": 10020 + }, + { + "epoch": 0.0869870921259364, + "grad_norm": 0.392578125, + "learning_rate": 0.0019602574583716037, + "loss": 0.1465, + "step": 10021 + }, + { + "epoch": 0.08699577260614057, + "grad_norm": 0.265625, + "learning_rate": 0.0019602487112424733, + "loss": 0.1338, + "step": 10022 + }, + { + "epoch": 0.08700445308634473, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019602399631725876, + "loss": 0.1123, + "step": 10023 + }, + { + "epoch": 0.0870131335665489, + "grad_norm": 0.349609375, + "learning_rate": 0.001960231214161957, + "loss": 0.1084, + "step": 10024 + }, + { + "epoch": 0.08702181404675306, + "grad_norm": 0.2578125, + "learning_rate": 0.0019602224642105903, + "loss": 0.1069, + "step": 10025 + }, + { + "epoch": 0.08703049452695723, + "grad_norm": 0.451171875, + "learning_rate": 0.001960213713318497, + "loss": 0.1279, + "step": 10026 + }, + { + "epoch": 0.08703917500716139, + "grad_norm": 0.3828125, + "learning_rate": 0.001960204961485687, + "loss": 0.1152, + "step": 10027 + }, + { + "epoch": 0.08704785548736556, + "grad_norm": 0.291015625, + "learning_rate": 0.00196019620871217, + "loss": 0.1455, + "step": 10028 + }, + { + "epoch": 0.08705653596756972, + "grad_norm": 0.29296875, + "learning_rate": 0.001960187454997955, + "loss": 0.1221, + "step": 10029 + }, + { + "epoch": 0.08706521644777389, + "grad_norm": 0.298828125, + "learning_rate": 0.0019601787003430526, + "loss": 0.1504, + "step": 10030 + }, + { + "epoch": 0.08707389692797805, + "grad_norm": 0.216796875, + "learning_rate": 0.001960169944747471, + "loss": 0.1143, + "step": 10031 + }, + { + "epoch": 0.08708257740818222, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019601611882112207, + "loss": 0.1309, + "step": 10032 + }, + { + "epoch": 0.08709125788838638, + "grad_norm": 0.275390625, + "learning_rate": 0.0019601524307343112, + "loss": 0.0859, + "step": 10033 + }, + { + "epoch": 0.08709993836859055, + "grad_norm": 0.357421875, + "learning_rate": 0.0019601436723167514, + "loss": 0.165, + "step": 10034 + }, + { + "epoch": 0.08710861884879471, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019601349129585517, + "loss": 0.083, + "step": 10035 + }, + { + "epoch": 0.08711729932899888, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019601261526597216, + "loss": 0.1035, + "step": 10036 + }, + { + "epoch": 0.08712597980920304, + "grad_norm": 0.314453125, + "learning_rate": 0.0019601173914202703, + "loss": 0.1328, + "step": 10037 + }, + { + "epoch": 0.08713466028940721, + "grad_norm": 0.30859375, + "learning_rate": 0.0019601086292402076, + "loss": 0.1182, + "step": 10038 + }, + { + "epoch": 0.08714334076961137, + "grad_norm": 0.55859375, + "learning_rate": 0.0019600998661195427, + "loss": 0.1167, + "step": 10039 + }, + { + "epoch": 0.08715202124981554, + "grad_norm": 0.1328125, + "learning_rate": 0.001960091102058286, + "loss": 0.1426, + "step": 10040 + }, + { + "epoch": 0.0871607017300197, + "grad_norm": 0.55859375, + "learning_rate": 0.0019600823370564463, + "loss": 0.126, + "step": 10041 + }, + { + "epoch": 0.08716938221022387, + "grad_norm": 0.5625, + "learning_rate": 0.0019600735711140338, + "loss": 0.0938, + "step": 10042 + }, + { + "epoch": 0.08717806269042803, + "grad_norm": 0.263671875, + "learning_rate": 0.0019600648042310577, + "loss": 0.1187, + "step": 10043 + }, + { + "epoch": 0.0871867431706322, + "grad_norm": 0.0888671875, + "learning_rate": 0.001960056036407527, + "loss": 0.1094, + "step": 10044 + }, + { + "epoch": 0.08719542365083637, + "grad_norm": 0.5, + "learning_rate": 0.0019600472676434525, + "loss": 0.1177, + "step": 10045 + }, + { + "epoch": 0.08720410413104053, + "grad_norm": 0.41015625, + "learning_rate": 0.0019600384979388437, + "loss": 0.1426, + "step": 10046 + }, + { + "epoch": 0.0872127846112447, + "grad_norm": 0.1328125, + "learning_rate": 0.001960029727293709, + "loss": 0.0869, + "step": 10047 + }, + { + "epoch": 0.08722146509144886, + "grad_norm": 0.236328125, + "learning_rate": 0.0019600209557080585, + "loss": 0.1387, + "step": 10048 + }, + { + "epoch": 0.08723014557165303, + "grad_norm": 0.51953125, + "learning_rate": 0.0019600121831819025, + "loss": 0.1377, + "step": 10049 + }, + { + "epoch": 0.08723882605185719, + "grad_norm": 0.2236328125, + "learning_rate": 0.00196000340971525, + "loss": 0.1055, + "step": 10050 + }, + { + "epoch": 0.08724750653206136, + "grad_norm": 0.12109375, + "learning_rate": 0.0019599946353081104, + "loss": 0.1475, + "step": 10051 + }, + { + "epoch": 0.08725618701226552, + "grad_norm": 0.32421875, + "learning_rate": 0.001959985859960494, + "loss": 0.1035, + "step": 10052 + }, + { + "epoch": 0.08726486749246969, + "grad_norm": 0.59765625, + "learning_rate": 0.0019599770836724096, + "loss": 0.1133, + "step": 10053 + }, + { + "epoch": 0.08727354797267385, + "grad_norm": 0.345703125, + "learning_rate": 0.001959968306443868, + "loss": 0.1348, + "step": 10054 + }, + { + "epoch": 0.08728222845287802, + "grad_norm": 0.09375, + "learning_rate": 0.0019599595282748772, + "loss": 0.1963, + "step": 10055 + }, + { + "epoch": 0.08729090893308218, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019599507491654482, + "loss": 0.1289, + "step": 10056 + }, + { + "epoch": 0.08729958941328635, + "grad_norm": 0.57421875, + "learning_rate": 0.001959941969115589, + "loss": 0.1689, + "step": 10057 + }, + { + "epoch": 0.08730826989349051, + "grad_norm": 0.6171875, + "learning_rate": 0.001959933188125311, + "loss": 0.2754, + "step": 10058 + }, + { + "epoch": 0.08731695037369468, + "grad_norm": 0.10546875, + "learning_rate": 0.0019599244061946227, + "loss": 0.1328, + "step": 10059 + }, + { + "epoch": 0.08732563085389884, + "grad_norm": 0.07763671875, + "learning_rate": 0.001959915623323534, + "loss": 0.1226, + "step": 10060 + }, + { + "epoch": 0.08733431133410301, + "grad_norm": 0.26171875, + "learning_rate": 0.001959906839512055, + "loss": 0.1016, + "step": 10061 + }, + { + "epoch": 0.08734299181430717, + "grad_norm": 0.1513671875, + "learning_rate": 0.001959898054760194, + "loss": 0.1328, + "step": 10062 + }, + { + "epoch": 0.08735167229451134, + "grad_norm": 1.3828125, + "learning_rate": 0.0019598892690679613, + "loss": 0.1582, + "step": 10063 + }, + { + "epoch": 0.0873603527747155, + "grad_norm": 0.359375, + "learning_rate": 0.0019598804824353675, + "loss": 0.1079, + "step": 10064 + }, + { + "epoch": 0.08736903325491967, + "grad_norm": 0.267578125, + "learning_rate": 0.001959871694862421, + "loss": 0.1309, + "step": 10065 + }, + { + "epoch": 0.08737771373512383, + "grad_norm": 0.2734375, + "learning_rate": 0.0019598629063491313, + "loss": 0.1387, + "step": 10066 + }, + { + "epoch": 0.087386394215328, + "grad_norm": 0.56640625, + "learning_rate": 0.001959854116895509, + "loss": 0.1211, + "step": 10067 + }, + { + "epoch": 0.08739507469553216, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019598453265015628, + "loss": 0.1064, + "step": 10068 + }, + { + "epoch": 0.08740375517573633, + "grad_norm": 0.470703125, + "learning_rate": 0.001959836535167303, + "loss": 0.1279, + "step": 10069 + }, + { + "epoch": 0.08741243565594049, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019598277428927386, + "loss": 0.1138, + "step": 10070 + }, + { + "epoch": 0.08742111613614466, + "grad_norm": 0.1630859375, + "learning_rate": 0.001959818949677879, + "loss": 0.1357, + "step": 10071 + }, + { + "epoch": 0.08742979661634882, + "grad_norm": 0.228515625, + "learning_rate": 0.001959810155522735, + "loss": 0.0889, + "step": 10072 + }, + { + "epoch": 0.08743847709655297, + "grad_norm": 0.57421875, + "learning_rate": 0.0019598013604273158, + "loss": 0.1543, + "step": 10073 + }, + { + "epoch": 0.08744715757675714, + "grad_norm": 0.146484375, + "learning_rate": 0.0019597925643916304, + "loss": 0.1504, + "step": 10074 + }, + { + "epoch": 0.0874558380569613, + "grad_norm": 0.419921875, + "learning_rate": 0.001959783767415689, + "loss": 0.1084, + "step": 10075 + }, + { + "epoch": 0.08746451853716547, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019597749694995003, + "loss": 0.1709, + "step": 10076 + }, + { + "epoch": 0.08747319901736963, + "grad_norm": 0.6875, + "learning_rate": 0.001959766170643075, + "loss": 0.1406, + "step": 10077 + }, + { + "epoch": 0.0874818794975738, + "grad_norm": 0.35546875, + "learning_rate": 0.001959757370846422, + "loss": 0.0933, + "step": 10078 + }, + { + "epoch": 0.08749055997777797, + "grad_norm": 0.10693359375, + "learning_rate": 0.001959748570109552, + "loss": 0.1113, + "step": 10079 + }, + { + "epoch": 0.08749924045798213, + "grad_norm": 0.11328125, + "learning_rate": 0.0019597397684324733, + "loss": 0.1191, + "step": 10080 + }, + { + "epoch": 0.0875079209381863, + "grad_norm": 0.44921875, + "learning_rate": 0.0019597309658151963, + "loss": 0.1094, + "step": 10081 + }, + { + "epoch": 0.08751660141839046, + "grad_norm": 0.2412109375, + "learning_rate": 0.001959722162257731, + "loss": 0.1328, + "step": 10082 + }, + { + "epoch": 0.08752528189859463, + "grad_norm": 0.4375, + "learning_rate": 0.0019597133577600855, + "loss": 0.082, + "step": 10083 + }, + { + "epoch": 0.08753396237879879, + "grad_norm": 0.255859375, + "learning_rate": 0.001959704552322271, + "loss": 0.1396, + "step": 10084 + }, + { + "epoch": 0.08754264285900296, + "grad_norm": 0.2734375, + "learning_rate": 0.001959695745944296, + "loss": 0.125, + "step": 10085 + }, + { + "epoch": 0.08755132333920712, + "grad_norm": 0.73828125, + "learning_rate": 0.001959686938626171, + "loss": 0.1738, + "step": 10086 + }, + { + "epoch": 0.08756000381941129, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019596781303679056, + "loss": 0.1113, + "step": 10087 + }, + { + "epoch": 0.08756868429961545, + "grad_norm": 0.232421875, + "learning_rate": 0.0019596693211695088, + "loss": 0.0859, + "step": 10088 + }, + { + "epoch": 0.08757736477981962, + "grad_norm": 0.92578125, + "learning_rate": 0.0019596605110309903, + "loss": 0.1494, + "step": 10089 + }, + { + "epoch": 0.08758604526002378, + "grad_norm": 0.45703125, + "learning_rate": 0.0019596516999523597, + "loss": 0.127, + "step": 10090 + }, + { + "epoch": 0.08759472574022795, + "grad_norm": 0.423828125, + "learning_rate": 0.001959642887933628, + "loss": 0.1201, + "step": 10091 + }, + { + "epoch": 0.08760340622043211, + "grad_norm": 0.099609375, + "learning_rate": 0.001959634074974803, + "loss": 0.1426, + "step": 10092 + }, + { + "epoch": 0.08761208670063628, + "grad_norm": 0.37109375, + "learning_rate": 0.0019596252610758953, + "loss": 0.1172, + "step": 10093 + }, + { + "epoch": 0.08762076718084044, + "grad_norm": 0.09521484375, + "learning_rate": 0.001959616446236914, + "loss": 0.1309, + "step": 10094 + }, + { + "epoch": 0.08762944766104461, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019596076304578694, + "loss": 0.1699, + "step": 10095 + }, + { + "epoch": 0.08763812814124877, + "grad_norm": 0.375, + "learning_rate": 0.0019595988137387707, + "loss": 0.1289, + "step": 10096 + }, + { + "epoch": 0.08764680862145294, + "grad_norm": 0.39453125, + "learning_rate": 0.001959589996079628, + "loss": 0.1377, + "step": 10097 + }, + { + "epoch": 0.0876554891016571, + "grad_norm": 0.2431640625, + "learning_rate": 0.00195958117748045, + "loss": 0.0957, + "step": 10098 + }, + { + "epoch": 0.08766416958186127, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019595723579412475, + "loss": 0.1123, + "step": 10099 + }, + { + "epoch": 0.08767285006206543, + "grad_norm": 0.11328125, + "learning_rate": 0.001959563537462029, + "loss": 0.126, + "step": 10100 + }, + { + "epoch": 0.0876815305422696, + "grad_norm": 0.396484375, + "learning_rate": 0.0019595547160428054, + "loss": 0.1445, + "step": 10101 + }, + { + "epoch": 0.08769021102247376, + "grad_norm": 0.3984375, + "learning_rate": 0.001959545893683585, + "loss": 0.1406, + "step": 10102 + }, + { + "epoch": 0.08769889150267793, + "grad_norm": 0.0888671875, + "learning_rate": 0.001959537070384379, + "loss": 0.126, + "step": 10103 + }, + { + "epoch": 0.0877075719828821, + "grad_norm": 0.466796875, + "learning_rate": 0.0019595282461451952, + "loss": 0.1133, + "step": 10104 + }, + { + "epoch": 0.08771625246308626, + "grad_norm": 0.30859375, + "learning_rate": 0.001959519420966045, + "loss": 0.1562, + "step": 10105 + }, + { + "epoch": 0.08772493294329042, + "grad_norm": 0.138671875, + "learning_rate": 0.0019595105948469367, + "loss": 0.1309, + "step": 10106 + }, + { + "epoch": 0.08773361342349459, + "grad_norm": 0.0712890625, + "learning_rate": 0.001959501767787881, + "loss": 0.0898, + "step": 10107 + }, + { + "epoch": 0.08774229390369875, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019594929397888866, + "loss": 0.1484, + "step": 10108 + }, + { + "epoch": 0.08775097438390292, + "grad_norm": 0.1435546875, + "learning_rate": 0.001959484110849964, + "loss": 0.1387, + "step": 10109 + }, + { + "epoch": 0.08775965486410708, + "grad_norm": 0.306640625, + "learning_rate": 0.0019594752809711223, + "loss": 0.1289, + "step": 10110 + }, + { + "epoch": 0.08776833534431125, + "grad_norm": 0.44921875, + "learning_rate": 0.0019594664501523715, + "loss": 0.2012, + "step": 10111 + }, + { + "epoch": 0.08777701582451541, + "grad_norm": 0.1171875, + "learning_rate": 0.0019594576183937207, + "loss": 0.0972, + "step": 10112 + }, + { + "epoch": 0.08778569630471958, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019594487856951803, + "loss": 0.0967, + "step": 10113 + }, + { + "epoch": 0.08779437678492374, + "grad_norm": 0.2109375, + "learning_rate": 0.0019594399520567597, + "loss": 0.123, + "step": 10114 + }, + { + "epoch": 0.08780305726512791, + "grad_norm": 1.5078125, + "learning_rate": 0.0019594311174784686, + "loss": 0.1621, + "step": 10115 + }, + { + "epoch": 0.08781173774533207, + "grad_norm": 0.166015625, + "learning_rate": 0.0019594222819603164, + "loss": 0.0815, + "step": 10116 + }, + { + "epoch": 0.08782041822553624, + "grad_norm": 0.38671875, + "learning_rate": 0.0019594134455023127, + "loss": 0.1582, + "step": 10117 + }, + { + "epoch": 0.0878290987057404, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019594046081044675, + "loss": 0.0952, + "step": 10118 + }, + { + "epoch": 0.08783777918594457, + "grad_norm": 0.36328125, + "learning_rate": 0.0019593957697667903, + "loss": 0.1201, + "step": 10119 + }, + { + "epoch": 0.08784645966614874, + "grad_norm": 0.2578125, + "learning_rate": 0.0019593869304892907, + "loss": 0.1309, + "step": 10120 + }, + { + "epoch": 0.0878551401463529, + "grad_norm": 0.4609375, + "learning_rate": 0.001959378090271979, + "loss": 0.1426, + "step": 10121 + }, + { + "epoch": 0.08786382062655707, + "grad_norm": 0.359375, + "learning_rate": 0.0019593692491148636, + "loss": 0.0942, + "step": 10122 + }, + { + "epoch": 0.08787250110676123, + "grad_norm": 0.07958984375, + "learning_rate": 0.001959360407017955, + "loss": 0.1025, + "step": 10123 + }, + { + "epoch": 0.0878811815869654, + "grad_norm": 0.208984375, + "learning_rate": 0.001959351563981263, + "loss": 0.0938, + "step": 10124 + }, + { + "epoch": 0.08788986206716956, + "grad_norm": 0.32421875, + "learning_rate": 0.0019593427200047973, + "loss": 0.123, + "step": 10125 + }, + { + "epoch": 0.08789854254737373, + "grad_norm": 0.34765625, + "learning_rate": 0.001959333875088567, + "loss": 0.1602, + "step": 10126 + }, + { + "epoch": 0.08790722302757789, + "grad_norm": 0.08203125, + "learning_rate": 0.001959325029232582, + "loss": 0.1416, + "step": 10127 + }, + { + "epoch": 0.08791590350778206, + "grad_norm": 0.388671875, + "learning_rate": 0.0019593161824368525, + "loss": 0.1553, + "step": 10128 + }, + { + "epoch": 0.08792458398798622, + "grad_norm": 0.322265625, + "learning_rate": 0.0019593073347013872, + "loss": 0.1436, + "step": 10129 + }, + { + "epoch": 0.08793326446819039, + "grad_norm": 0.32421875, + "learning_rate": 0.0019592984860261967, + "loss": 0.0923, + "step": 10130 + }, + { + "epoch": 0.08794194494839455, + "grad_norm": 0.65625, + "learning_rate": 0.00195928963641129, + "loss": 0.1689, + "step": 10131 + }, + { + "epoch": 0.08795062542859872, + "grad_norm": 0.609375, + "learning_rate": 0.0019592807858566774, + "loss": 0.1104, + "step": 10132 + }, + { + "epoch": 0.08795930590880288, + "grad_norm": 0.1953125, + "learning_rate": 0.001959271934362368, + "loss": 0.1133, + "step": 10133 + }, + { + "epoch": 0.08796798638900703, + "grad_norm": 0.123046875, + "learning_rate": 0.001959263081928372, + "loss": 0.1455, + "step": 10134 + }, + { + "epoch": 0.0879766668692112, + "grad_norm": 0.6015625, + "learning_rate": 0.0019592542285546983, + "loss": 0.1138, + "step": 10135 + }, + { + "epoch": 0.08798534734941536, + "grad_norm": 0.16796875, + "learning_rate": 0.0019592453742413576, + "loss": 0.127, + "step": 10136 + }, + { + "epoch": 0.08799402782961953, + "grad_norm": 0.1181640625, + "learning_rate": 0.001959236518988359, + "loss": 0.1357, + "step": 10137 + }, + { + "epoch": 0.0880027083098237, + "grad_norm": 0.115234375, + "learning_rate": 0.0019592276627957116, + "loss": 0.1133, + "step": 10138 + }, + { + "epoch": 0.08801138879002786, + "grad_norm": 0.10400390625, + "learning_rate": 0.001959218805663426, + "loss": 0.123, + "step": 10139 + }, + { + "epoch": 0.08802006927023202, + "grad_norm": 0.06982421875, + "learning_rate": 0.001959209947591512, + "loss": 0.0889, + "step": 10140 + }, + { + "epoch": 0.08802874975043619, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019592010885799787, + "loss": 0.0947, + "step": 10141 + }, + { + "epoch": 0.08803743023064035, + "grad_norm": 0.2490234375, + "learning_rate": 0.001959192228628836, + "loss": 0.1357, + "step": 10142 + }, + { + "epoch": 0.08804611071084452, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019591833677380936, + "loss": 0.1445, + "step": 10143 + }, + { + "epoch": 0.08805479119104868, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019591745059077616, + "loss": 0.0967, + "step": 10144 + }, + { + "epoch": 0.08806347167125285, + "grad_norm": 0.498046875, + "learning_rate": 0.0019591656431378487, + "loss": 0.1396, + "step": 10145 + }, + { + "epoch": 0.08807215215145701, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019591567794283653, + "loss": 0.1118, + "step": 10146 + }, + { + "epoch": 0.08808083263166118, + "grad_norm": 0.2890625, + "learning_rate": 0.001959147914779321, + "loss": 0.1641, + "step": 10147 + }, + { + "epoch": 0.08808951311186534, + "grad_norm": 0.08984375, + "learning_rate": 0.0019591390491907256, + "loss": 0.1172, + "step": 10148 + }, + { + "epoch": 0.08809819359206951, + "grad_norm": 0.578125, + "learning_rate": 0.001959130182662588, + "loss": 0.1104, + "step": 10149 + }, + { + "epoch": 0.08810687407227367, + "grad_norm": 0.236328125, + "learning_rate": 0.0019591213151949195, + "loss": 0.1172, + "step": 10150 + }, + { + "epoch": 0.08811555455247784, + "grad_norm": 0.466796875, + "learning_rate": 0.001959112446787728, + "loss": 0.1309, + "step": 10151 + }, + { + "epoch": 0.088124235032682, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019591035774410245, + "loss": 0.1064, + "step": 10152 + }, + { + "epoch": 0.08813291551288617, + "grad_norm": 0.140625, + "learning_rate": 0.0019590947071548184, + "loss": 0.1143, + "step": 10153 + }, + { + "epoch": 0.08814159599309034, + "grad_norm": 0.478515625, + "learning_rate": 0.0019590858359291186, + "loss": 0.1357, + "step": 10154 + }, + { + "epoch": 0.0881502764732945, + "grad_norm": 0.0810546875, + "learning_rate": 0.001959076963763936, + "loss": 0.126, + "step": 10155 + }, + { + "epoch": 0.08815895695349867, + "grad_norm": 0.8359375, + "learning_rate": 0.0019590680906592796, + "loss": 0.1123, + "step": 10156 + }, + { + "epoch": 0.08816763743370283, + "grad_norm": 0.5625, + "learning_rate": 0.0019590592166151594, + "loss": 0.1348, + "step": 10157 + }, + { + "epoch": 0.088176317913907, + "grad_norm": 0.4375, + "learning_rate": 0.0019590503416315845, + "loss": 0.1396, + "step": 10158 + }, + { + "epoch": 0.08818499839411116, + "grad_norm": 0.458984375, + "learning_rate": 0.0019590414657085654, + "loss": 0.1113, + "step": 10159 + }, + { + "epoch": 0.08819367887431533, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019590325888461115, + "loss": 0.0957, + "step": 10160 + }, + { + "epoch": 0.08820235935451949, + "grad_norm": 0.796875, + "learning_rate": 0.001959023711044232, + "loss": 0.1006, + "step": 10161 + }, + { + "epoch": 0.08821103983472366, + "grad_norm": 0.6015625, + "learning_rate": 0.0019590148323029377, + "loss": 0.0918, + "step": 10162 + }, + { + "epoch": 0.08821972031492782, + "grad_norm": 0.20703125, + "learning_rate": 0.0019590059526222374, + "loss": 0.1309, + "step": 10163 + }, + { + "epoch": 0.08822840079513199, + "grad_norm": 0.21875, + "learning_rate": 0.001958997072002141, + "loss": 0.0771, + "step": 10164 + }, + { + "epoch": 0.08823708127533615, + "grad_norm": 0.365234375, + "learning_rate": 0.0019589881904426582, + "loss": 0.0928, + "step": 10165 + }, + { + "epoch": 0.08824576175554032, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019589793079437994, + "loss": 0.0918, + "step": 10166 + }, + { + "epoch": 0.08825444223574448, + "grad_norm": 0.330078125, + "learning_rate": 0.001958970424505573, + "loss": 0.1152, + "step": 10167 + }, + { + "epoch": 0.08826312271594865, + "grad_norm": 1.09375, + "learning_rate": 0.00195896154012799, + "loss": 0.1465, + "step": 10168 + }, + { + "epoch": 0.08827180319615281, + "grad_norm": 0.09765625, + "learning_rate": 0.001958952654811059, + "loss": 0.1211, + "step": 10169 + }, + { + "epoch": 0.08828048367635698, + "grad_norm": 0.1787109375, + "learning_rate": 0.001958943768554791, + "loss": 0.1416, + "step": 10170 + }, + { + "epoch": 0.08828916415656114, + "grad_norm": 0.46484375, + "learning_rate": 0.0019589348813591947, + "loss": 0.1318, + "step": 10171 + }, + { + "epoch": 0.08829784463676531, + "grad_norm": 0.373046875, + "learning_rate": 0.00195892599322428, + "loss": 0.1201, + "step": 10172 + }, + { + "epoch": 0.08830652511696947, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019589171041500568, + "loss": 0.1064, + "step": 10173 + }, + { + "epoch": 0.08831520559717364, + "grad_norm": 1.0546875, + "learning_rate": 0.0019589082141365345, + "loss": 0.1445, + "step": 10174 + }, + { + "epoch": 0.0883238860773778, + "grad_norm": 0.142578125, + "learning_rate": 0.0019588993231837238, + "loss": 0.1436, + "step": 10175 + }, + { + "epoch": 0.08833256655758197, + "grad_norm": 0.427734375, + "learning_rate": 0.001958890431291633, + "loss": 0.1123, + "step": 10176 + }, + { + "epoch": 0.08834124703778613, + "grad_norm": 0.49609375, + "learning_rate": 0.001958881538460273, + "loss": 0.1016, + "step": 10177 + }, + { + "epoch": 0.0883499275179903, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019588726446896527, + "loss": 0.1553, + "step": 10178 + }, + { + "epoch": 0.08835860799819446, + "grad_norm": 0.09375, + "learning_rate": 0.001958863749979782, + "loss": 0.1328, + "step": 10179 + }, + { + "epoch": 0.08836728847839863, + "grad_norm": 0.298828125, + "learning_rate": 0.001958854854330671, + "loss": 0.1221, + "step": 10180 + }, + { + "epoch": 0.0883759689586028, + "grad_norm": 0.51171875, + "learning_rate": 0.0019588459577423297, + "loss": 0.1875, + "step": 10181 + }, + { + "epoch": 0.08838464943880696, + "grad_norm": 0.095703125, + "learning_rate": 0.001958837060214767, + "loss": 0.1089, + "step": 10182 + }, + { + "epoch": 0.08839332991901112, + "grad_norm": 0.56640625, + "learning_rate": 0.001958828161747993, + "loss": 0.1436, + "step": 10183 + }, + { + "epoch": 0.08840201039921529, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019588192623420176, + "loss": 0.125, + "step": 10184 + }, + { + "epoch": 0.08841069087941945, + "grad_norm": 0.154296875, + "learning_rate": 0.00195881036199685, + "loss": 0.1494, + "step": 10185 + }, + { + "epoch": 0.08841937135962362, + "grad_norm": 0.5390625, + "learning_rate": 0.0019588014607125007, + "loss": 0.1465, + "step": 10186 + }, + { + "epoch": 0.08842805183982778, + "grad_norm": 0.29296875, + "learning_rate": 0.0019587925584889786, + "loss": 0.1992, + "step": 10187 + }, + { + "epoch": 0.08843673232003195, + "grad_norm": 0.46875, + "learning_rate": 0.0019587836553262945, + "loss": 0.1553, + "step": 10188 + }, + { + "epoch": 0.08844541280023611, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019587747512244567, + "loss": 0.1309, + "step": 10189 + }, + { + "epoch": 0.08845409328044028, + "grad_norm": 0.1953125, + "learning_rate": 0.001958765846183476, + "loss": 0.0986, + "step": 10190 + }, + { + "epoch": 0.08846277376064444, + "grad_norm": 0.09130859375, + "learning_rate": 0.001958756940203362, + "loss": 0.1094, + "step": 10191 + }, + { + "epoch": 0.08847145424084861, + "grad_norm": 1.5078125, + "learning_rate": 0.0019587480332841243, + "loss": 0.1855, + "step": 10192 + }, + { + "epoch": 0.08848013472105277, + "grad_norm": 0.115234375, + "learning_rate": 0.0019587391254257728, + "loss": 0.1289, + "step": 10193 + }, + { + "epoch": 0.08848881520125694, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019587302166283174, + "loss": 0.1094, + "step": 10194 + }, + { + "epoch": 0.0884974956814611, + "grad_norm": 0.216796875, + "learning_rate": 0.001958721306891767, + "loss": 0.103, + "step": 10195 + }, + { + "epoch": 0.08850617616166526, + "grad_norm": 0.66015625, + "learning_rate": 0.0019587123962161316, + "loss": 0.2305, + "step": 10196 + }, + { + "epoch": 0.08851485664186942, + "grad_norm": 0.23046875, + "learning_rate": 0.0019587034846014215, + "loss": 0.1338, + "step": 10197 + }, + { + "epoch": 0.08852353712207359, + "grad_norm": 1.34375, + "learning_rate": 0.0019586945720476466, + "loss": 0.1138, + "step": 10198 + }, + { + "epoch": 0.08853221760227775, + "grad_norm": 0.28515625, + "learning_rate": 0.0019586856585548157, + "loss": 0.1172, + "step": 10199 + }, + { + "epoch": 0.08854089808248192, + "grad_norm": 0.59765625, + "learning_rate": 0.001958676744122939, + "loss": 0.1123, + "step": 10200 + }, + { + "epoch": 0.08854957856268608, + "grad_norm": 0.392578125, + "learning_rate": 0.001958667828752027, + "loss": 0.1523, + "step": 10201 + }, + { + "epoch": 0.08855825904289025, + "grad_norm": 0.2578125, + "learning_rate": 0.0019586589124420883, + "loss": 0.1377, + "step": 10202 + }, + { + "epoch": 0.08856693952309441, + "grad_norm": 0.41796875, + "learning_rate": 0.001958649995193133, + "loss": 0.1064, + "step": 10203 + }, + { + "epoch": 0.08857562000329858, + "grad_norm": 0.158203125, + "learning_rate": 0.001958641077005171, + "loss": 0.1406, + "step": 10204 + }, + { + "epoch": 0.08858430048350274, + "grad_norm": 0.8046875, + "learning_rate": 0.001958632157878212, + "loss": 0.1016, + "step": 10205 + }, + { + "epoch": 0.08859298096370691, + "grad_norm": 0.6015625, + "learning_rate": 0.001958623237812266, + "loss": 0.1245, + "step": 10206 + }, + { + "epoch": 0.08860166144391107, + "grad_norm": 0.671875, + "learning_rate": 0.0019586143168073425, + "loss": 0.1055, + "step": 10207 + }, + { + "epoch": 0.08861034192411524, + "grad_norm": 0.33203125, + "learning_rate": 0.0019586053948634514, + "loss": 0.1592, + "step": 10208 + }, + { + "epoch": 0.0886190224043194, + "grad_norm": 0.07666015625, + "learning_rate": 0.001958596471980602, + "loss": 0.1289, + "step": 10209 + }, + { + "epoch": 0.08862770288452357, + "grad_norm": 0.2890625, + "learning_rate": 0.0019585875481588043, + "loss": 0.1602, + "step": 10210 + }, + { + "epoch": 0.08863638336472773, + "grad_norm": 0.17578125, + "learning_rate": 0.0019585786233980687, + "loss": 0.125, + "step": 10211 + }, + { + "epoch": 0.0886450638449319, + "grad_norm": 0.2021484375, + "learning_rate": 0.001958569697698404, + "loss": 0.0903, + "step": 10212 + }, + { + "epoch": 0.08865374432513606, + "grad_norm": 0.08203125, + "learning_rate": 0.001958560771059821, + "loss": 0.1025, + "step": 10213 + }, + { + "epoch": 0.08866242480534023, + "grad_norm": 0.2080078125, + "learning_rate": 0.001958551843482328, + "loss": 0.1582, + "step": 10214 + }, + { + "epoch": 0.0886711052855444, + "grad_norm": 0.48828125, + "learning_rate": 0.001958542914965936, + "loss": 0.1143, + "step": 10215 + }, + { + "epoch": 0.08867978576574856, + "grad_norm": 0.50390625, + "learning_rate": 0.001958533985510654, + "loss": 0.105, + "step": 10216 + }, + { + "epoch": 0.08868846624595272, + "grad_norm": 0.431640625, + "learning_rate": 0.0019585250551164926, + "loss": 0.1328, + "step": 10217 + }, + { + "epoch": 0.08869714672615689, + "grad_norm": 0.294921875, + "learning_rate": 0.0019585161237834613, + "loss": 0.1719, + "step": 10218 + }, + { + "epoch": 0.08870582720636105, + "grad_norm": 0.2021484375, + "learning_rate": 0.001958507191511569, + "loss": 0.1191, + "step": 10219 + }, + { + "epoch": 0.08871450768656522, + "grad_norm": 0.10498046875, + "learning_rate": 0.001958498258300827, + "loss": 0.1406, + "step": 10220 + }, + { + "epoch": 0.08872318816676938, + "grad_norm": 0.185546875, + "learning_rate": 0.001958489324151243, + "loss": 0.127, + "step": 10221 + }, + { + "epoch": 0.08873186864697355, + "grad_norm": 0.37890625, + "learning_rate": 0.0019584803890628287, + "loss": 0.1406, + "step": 10222 + }, + { + "epoch": 0.08874054912717771, + "grad_norm": 0.30859375, + "learning_rate": 0.0019584714530355932, + "loss": 0.1475, + "step": 10223 + }, + { + "epoch": 0.08874922960738188, + "grad_norm": 0.099609375, + "learning_rate": 0.001958462516069546, + "loss": 0.1191, + "step": 10224 + }, + { + "epoch": 0.08875791008758604, + "grad_norm": 0.8203125, + "learning_rate": 0.0019584535781646973, + "loss": 0.125, + "step": 10225 + }, + { + "epoch": 0.08876659056779021, + "grad_norm": 0.1171875, + "learning_rate": 0.0019584446393210563, + "loss": 0.106, + "step": 10226 + }, + { + "epoch": 0.08877527104799438, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019584356995386333, + "loss": 0.166, + "step": 10227 + }, + { + "epoch": 0.08878395152819854, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019584267588174382, + "loss": 0.0879, + "step": 10228 + }, + { + "epoch": 0.0887926320084027, + "grad_norm": 0.279296875, + "learning_rate": 0.00195841781715748, + "loss": 0.1211, + "step": 10229 + }, + { + "epoch": 0.08880131248860687, + "grad_norm": 0.4140625, + "learning_rate": 0.0019584088745587696, + "loss": 0.1157, + "step": 10230 + }, + { + "epoch": 0.08880999296881104, + "grad_norm": 0.87890625, + "learning_rate": 0.0019583999310213156, + "loss": 0.1553, + "step": 10231 + }, + { + "epoch": 0.0888186734490152, + "grad_norm": 0.1923828125, + "learning_rate": 0.001958390986545129, + "loss": 0.1016, + "step": 10232 + }, + { + "epoch": 0.08882735392921937, + "grad_norm": 1.5859375, + "learning_rate": 0.0019583820411302184, + "loss": 0.1543, + "step": 10233 + }, + { + "epoch": 0.08883603440942353, + "grad_norm": 0.09912109375, + "learning_rate": 0.001958373094776594, + "loss": 0.1201, + "step": 10234 + }, + { + "epoch": 0.0888447148896277, + "grad_norm": 0.345703125, + "learning_rate": 0.001958364147484266, + "loss": 0.1406, + "step": 10235 + }, + { + "epoch": 0.08885339536983186, + "grad_norm": 0.2265625, + "learning_rate": 0.001958355199253244, + "loss": 0.1689, + "step": 10236 + }, + { + "epoch": 0.08886207585003603, + "grad_norm": 0.1572265625, + "learning_rate": 0.001958346250083537, + "loss": 0.1582, + "step": 10237 + }, + { + "epoch": 0.08887075633024019, + "grad_norm": 0.279296875, + "learning_rate": 0.001958337299975156, + "loss": 0.0908, + "step": 10238 + }, + { + "epoch": 0.08887943681044436, + "grad_norm": 0.287109375, + "learning_rate": 0.00195832834892811, + "loss": 0.0889, + "step": 10239 + }, + { + "epoch": 0.08888811729064852, + "grad_norm": 0.07568359375, + "learning_rate": 0.001958319396942409, + "loss": 0.1089, + "step": 10240 + }, + { + "epoch": 0.08889679777085269, + "grad_norm": 0.267578125, + "learning_rate": 0.0019583104440180628, + "loss": 0.0889, + "step": 10241 + }, + { + "epoch": 0.08890547825105685, + "grad_norm": 0.65234375, + "learning_rate": 0.001958301490155081, + "loss": 0.1816, + "step": 10242 + }, + { + "epoch": 0.08891415873126102, + "grad_norm": 0.60546875, + "learning_rate": 0.001958292535353474, + "loss": 0.1172, + "step": 10243 + }, + { + "epoch": 0.08892283921146518, + "grad_norm": 0.11279296875, + "learning_rate": 0.001958283579613251, + "loss": 0.124, + "step": 10244 + }, + { + "epoch": 0.08893151969166935, + "grad_norm": 0.1259765625, + "learning_rate": 0.001958274622934422, + "loss": 0.124, + "step": 10245 + }, + { + "epoch": 0.08894020017187351, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019582656653169966, + "loss": 0.1152, + "step": 10246 + }, + { + "epoch": 0.08894888065207768, + "grad_norm": 0.376953125, + "learning_rate": 0.001958256706760985, + "loss": 0.1094, + "step": 10247 + }, + { + "epoch": 0.08895756113228184, + "grad_norm": 1.1171875, + "learning_rate": 0.0019582477472663965, + "loss": 0.125, + "step": 10248 + }, + { + "epoch": 0.08896624161248601, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019582387868332415, + "loss": 0.0869, + "step": 10249 + }, + { + "epoch": 0.08897492209269017, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019582298254615296, + "loss": 0.1426, + "step": 10250 + }, + { + "epoch": 0.08898360257289434, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019582208631512697, + "loss": 0.1396, + "step": 10251 + }, + { + "epoch": 0.0889922830530985, + "grad_norm": 0.2060546875, + "learning_rate": 0.001958211899902473, + "loss": 0.1123, + "step": 10252 + }, + { + "epoch": 0.08900096353330267, + "grad_norm": 0.314453125, + "learning_rate": 0.001958202935715148, + "loss": 0.1914, + "step": 10253 + }, + { + "epoch": 0.08900964401350683, + "grad_norm": 0.9609375, + "learning_rate": 0.001958193970589306, + "loss": 0.106, + "step": 10254 + }, + { + "epoch": 0.089018324493711, + "grad_norm": 0.15625, + "learning_rate": 0.0019581850045249554, + "loss": 0.124, + "step": 10255 + }, + { + "epoch": 0.08902700497391516, + "grad_norm": 0.248046875, + "learning_rate": 0.001958176037522107, + "loss": 0.1211, + "step": 10256 + }, + { + "epoch": 0.08903568545411933, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019581670695807698, + "loss": 0.0889, + "step": 10257 + }, + { + "epoch": 0.08904436593432348, + "grad_norm": 0.140625, + "learning_rate": 0.0019581581007009543, + "loss": 0.0996, + "step": 10258 + }, + { + "epoch": 0.08905304641452765, + "grad_norm": 0.109375, + "learning_rate": 0.00195814913088267, + "loss": 0.0947, + "step": 10259 + }, + { + "epoch": 0.08906172689473181, + "grad_norm": 0.14453125, + "learning_rate": 0.001958140160125926, + "loss": 0.1084, + "step": 10260 + }, + { + "epoch": 0.08907040737493598, + "grad_norm": 0.671875, + "learning_rate": 0.001958131188430734, + "loss": 0.127, + "step": 10261 + }, + { + "epoch": 0.08907908785514014, + "grad_norm": 0.40625, + "learning_rate": 0.0019581222157971016, + "loss": 0.1201, + "step": 10262 + }, + { + "epoch": 0.0890877683353443, + "grad_norm": 0.6640625, + "learning_rate": 0.00195811324222504, + "loss": 0.1426, + "step": 10263 + }, + { + "epoch": 0.08909644881554847, + "grad_norm": 0.6875, + "learning_rate": 0.0019581042677145584, + "loss": 0.1016, + "step": 10264 + }, + { + "epoch": 0.08910512929575264, + "grad_norm": 0.263671875, + "learning_rate": 0.0019580952922656674, + "loss": 0.1299, + "step": 10265 + }, + { + "epoch": 0.0891138097759568, + "grad_norm": 0.416015625, + "learning_rate": 0.001958086315878376, + "loss": 0.0957, + "step": 10266 + }, + { + "epoch": 0.08912249025616097, + "grad_norm": 0.130859375, + "learning_rate": 0.001958077338552694, + "loss": 0.1201, + "step": 10267 + }, + { + "epoch": 0.08913117073636513, + "grad_norm": 0.703125, + "learning_rate": 0.001958068360288632, + "loss": 0.1445, + "step": 10268 + }, + { + "epoch": 0.0891398512165693, + "grad_norm": 1.09375, + "learning_rate": 0.0019580593810861993, + "loss": 0.1436, + "step": 10269 + }, + { + "epoch": 0.08914853169677346, + "grad_norm": 0.55859375, + "learning_rate": 0.0019580504009454056, + "loss": 0.1064, + "step": 10270 + }, + { + "epoch": 0.08915721217697763, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019580414198662605, + "loss": 0.0947, + "step": 10271 + }, + { + "epoch": 0.08916589265718179, + "grad_norm": 0.29296875, + "learning_rate": 0.0019580324378487748, + "loss": 0.1348, + "step": 10272 + }, + { + "epoch": 0.08917457313738596, + "grad_norm": 0.34765625, + "learning_rate": 0.0019580234548929576, + "loss": 0.0947, + "step": 10273 + }, + { + "epoch": 0.08918325361759012, + "grad_norm": 0.80078125, + "learning_rate": 0.0019580144709988186, + "loss": 0.1367, + "step": 10274 + }, + { + "epoch": 0.08919193409779429, + "grad_norm": 1.5859375, + "learning_rate": 0.0019580054861663685, + "loss": 0.1484, + "step": 10275 + }, + { + "epoch": 0.08920061457799845, + "grad_norm": 0.248046875, + "learning_rate": 0.0019579965003956156, + "loss": 0.1592, + "step": 10276 + }, + { + "epoch": 0.08920929505820262, + "grad_norm": 0.43359375, + "learning_rate": 0.001957987513686571, + "loss": 0.1367, + "step": 10277 + }, + { + "epoch": 0.08921797553840678, + "grad_norm": 0.431640625, + "learning_rate": 0.0019579785260392443, + "loss": 0.1416, + "step": 10278 + }, + { + "epoch": 0.08922665601861095, + "grad_norm": 0.4140625, + "learning_rate": 0.001957969537453645, + "loss": 0.125, + "step": 10279 + }, + { + "epoch": 0.08923533649881511, + "grad_norm": 0.111328125, + "learning_rate": 0.001957960547929783, + "loss": 0.1064, + "step": 10280 + }, + { + "epoch": 0.08924401697901928, + "grad_norm": 0.3671875, + "learning_rate": 0.0019579515574676683, + "loss": 0.1416, + "step": 10281 + }, + { + "epoch": 0.08925269745922344, + "grad_norm": 0.09619140625, + "learning_rate": 0.001957942566067311, + "loss": 0.1543, + "step": 10282 + }, + { + "epoch": 0.08926137793942761, + "grad_norm": 0.54296875, + "learning_rate": 0.00195793357372872, + "loss": 0.1729, + "step": 10283 + }, + { + "epoch": 0.08927005841963177, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019579245804519064, + "loss": 0.1064, + "step": 10284 + }, + { + "epoch": 0.08927873889983594, + "grad_norm": 1.3515625, + "learning_rate": 0.001957915586236879, + "loss": 0.1475, + "step": 10285 + }, + { + "epoch": 0.0892874193800401, + "grad_norm": 0.09814453125, + "learning_rate": 0.001957906591083648, + "loss": 0.168, + "step": 10286 + }, + { + "epoch": 0.08929609986024427, + "grad_norm": 0.3671875, + "learning_rate": 0.001957897594992223, + "loss": 0.1001, + "step": 10287 + }, + { + "epoch": 0.08930478034044843, + "grad_norm": 0.08203125, + "learning_rate": 0.001957888597962615, + "loss": 0.1147, + "step": 10288 + }, + { + "epoch": 0.0893134608206526, + "grad_norm": 0.1298828125, + "learning_rate": 0.001957879599994832, + "loss": 0.125, + "step": 10289 + }, + { + "epoch": 0.08932214130085676, + "grad_norm": 0.10986328125, + "learning_rate": 0.001957870601088885, + "loss": 0.1143, + "step": 10290 + }, + { + "epoch": 0.08933082178106093, + "grad_norm": 0.25390625, + "learning_rate": 0.0019578616012447835, + "loss": 0.1494, + "step": 10291 + }, + { + "epoch": 0.0893395022612651, + "grad_norm": 0.578125, + "learning_rate": 0.0019578526004625377, + "loss": 0.126, + "step": 10292 + }, + { + "epoch": 0.08934818274146926, + "grad_norm": 0.22265625, + "learning_rate": 0.001957843598742157, + "loss": 0.0947, + "step": 10293 + }, + { + "epoch": 0.08935686322167342, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019578345960836515, + "loss": 0.1206, + "step": 10294 + }, + { + "epoch": 0.08936554370187759, + "grad_norm": 0.18359375, + "learning_rate": 0.001957825592487031, + "loss": 0.0903, + "step": 10295 + }, + { + "epoch": 0.08937422418208175, + "grad_norm": 0.44140625, + "learning_rate": 0.0019578165879523056, + "loss": 0.1006, + "step": 10296 + }, + { + "epoch": 0.08938290466228592, + "grad_norm": 1.0078125, + "learning_rate": 0.0019578075824794846, + "loss": 0.126, + "step": 10297 + }, + { + "epoch": 0.08939158514249008, + "grad_norm": 1.6015625, + "learning_rate": 0.0019577985760685776, + "loss": 0.2266, + "step": 10298 + }, + { + "epoch": 0.08940026562269425, + "grad_norm": 0.423828125, + "learning_rate": 0.001957789568719596, + "loss": 0.1387, + "step": 10299 + }, + { + "epoch": 0.08940894610289842, + "grad_norm": 0.57421875, + "learning_rate": 0.0019577805604325477, + "loss": 0.1035, + "step": 10300 + }, + { + "epoch": 0.08941762658310258, + "grad_norm": 0.1279296875, + "learning_rate": 0.001957771551207444, + "loss": 0.1895, + "step": 10301 + }, + { + "epoch": 0.08942630706330675, + "grad_norm": 0.87890625, + "learning_rate": 0.001957762541044294, + "loss": 0.1143, + "step": 10302 + }, + { + "epoch": 0.08943498754351091, + "grad_norm": 0.453125, + "learning_rate": 0.0019577535299431076, + "loss": 0.1016, + "step": 10303 + }, + { + "epoch": 0.08944366802371508, + "grad_norm": 0.66015625, + "learning_rate": 0.001957744517903895, + "loss": 0.127, + "step": 10304 + }, + { + "epoch": 0.08945234850391924, + "grad_norm": 0.78515625, + "learning_rate": 0.0019577355049266664, + "loss": 0.1191, + "step": 10305 + }, + { + "epoch": 0.0894610289841234, + "grad_norm": 0.166015625, + "learning_rate": 0.00195772649101143, + "loss": 0.1406, + "step": 10306 + }, + { + "epoch": 0.08946970946432757, + "grad_norm": 0.310546875, + "learning_rate": 0.001957717476158198, + "loss": 0.1074, + "step": 10307 + }, + { + "epoch": 0.08947838994453174, + "grad_norm": 0.474609375, + "learning_rate": 0.0019577084603669788, + "loss": 0.1318, + "step": 10308 + }, + { + "epoch": 0.0894870704247359, + "grad_norm": 0.373046875, + "learning_rate": 0.001957699443637782, + "loss": 0.1719, + "step": 10309 + }, + { + "epoch": 0.08949575090494007, + "grad_norm": 0.373046875, + "learning_rate": 0.0019576904259706184, + "loss": 0.1367, + "step": 10310 + }, + { + "epoch": 0.08950443138514423, + "grad_norm": 0.578125, + "learning_rate": 0.001957681407365497, + "loss": 0.1465, + "step": 10311 + }, + { + "epoch": 0.0895131118653484, + "grad_norm": 0.5234375, + "learning_rate": 0.0019576723878224284, + "loss": 0.1074, + "step": 10312 + }, + { + "epoch": 0.08952179234555256, + "grad_norm": 0.291015625, + "learning_rate": 0.001957663367341422, + "loss": 0.1318, + "step": 10313 + }, + { + "epoch": 0.08953047282575673, + "grad_norm": 0.3125, + "learning_rate": 0.0019576543459224878, + "loss": 0.1445, + "step": 10314 + }, + { + "epoch": 0.08953915330596089, + "grad_norm": 0.2890625, + "learning_rate": 0.001957645323565636, + "loss": 0.125, + "step": 10315 + }, + { + "epoch": 0.08954783378616506, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019576363002708763, + "loss": 0.0972, + "step": 10316 + }, + { + "epoch": 0.08955651426636922, + "grad_norm": 0.1533203125, + "learning_rate": 0.001957627276038218, + "loss": 0.0879, + "step": 10317 + }, + { + "epoch": 0.08956519474657339, + "grad_norm": 0.63671875, + "learning_rate": 0.0019576182508676716, + "loss": 0.1582, + "step": 10318 + }, + { + "epoch": 0.08957387522677754, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019576092247592467, + "loss": 0.1465, + "step": 10319 + }, + { + "epoch": 0.0895825557069817, + "grad_norm": 0.220703125, + "learning_rate": 0.001957600197712953, + "loss": 0.1562, + "step": 10320 + }, + { + "epoch": 0.08959123618718587, + "grad_norm": 0.369140625, + "learning_rate": 0.0019575911697288007, + "loss": 0.1143, + "step": 10321 + }, + { + "epoch": 0.08959991666739003, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019575821408067996, + "loss": 0.1172, + "step": 10322 + }, + { + "epoch": 0.0896085971475942, + "grad_norm": 0.275390625, + "learning_rate": 0.0019575731109469597, + "loss": 0.1484, + "step": 10323 + }, + { + "epoch": 0.08961727762779836, + "grad_norm": 0.70703125, + "learning_rate": 0.001957564080149291, + "loss": 0.1182, + "step": 10324 + }, + { + "epoch": 0.08962595810800253, + "grad_norm": 0.5078125, + "learning_rate": 0.0019575550484138026, + "loss": 0.1328, + "step": 10325 + }, + { + "epoch": 0.0896346385882067, + "grad_norm": 0.59375, + "learning_rate": 0.0019575460157405054, + "loss": 0.1416, + "step": 10326 + }, + { + "epoch": 0.08964331906841086, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019575369821294084, + "loss": 0.1069, + "step": 10327 + }, + { + "epoch": 0.08965199954861502, + "grad_norm": 0.451171875, + "learning_rate": 0.001957527947580522, + "loss": 0.124, + "step": 10328 + }, + { + "epoch": 0.08966068002881919, + "grad_norm": 0.326171875, + "learning_rate": 0.0019575189120938554, + "loss": 0.1035, + "step": 10329 + }, + { + "epoch": 0.08966936050902335, + "grad_norm": 0.76953125, + "learning_rate": 0.0019575098756694194, + "loss": 0.1406, + "step": 10330 + }, + { + "epoch": 0.08967804098922752, + "grad_norm": 0.11474609375, + "learning_rate": 0.001957500838307224, + "loss": 0.1836, + "step": 10331 + }, + { + "epoch": 0.08968672146943168, + "grad_norm": 0.2177734375, + "learning_rate": 0.001957491800007278, + "loss": 0.1328, + "step": 10332 + }, + { + "epoch": 0.08969540194963585, + "grad_norm": 0.734375, + "learning_rate": 0.001957482760769592, + "loss": 0.0908, + "step": 10333 + }, + { + "epoch": 0.08970408242984002, + "grad_norm": 0.25390625, + "learning_rate": 0.0019574737205941754, + "loss": 0.165, + "step": 10334 + }, + { + "epoch": 0.08971276291004418, + "grad_norm": 0.57421875, + "learning_rate": 0.0019574646794810386, + "loss": 0.1445, + "step": 10335 + }, + { + "epoch": 0.08972144339024835, + "grad_norm": 0.091796875, + "learning_rate": 0.0019574556374301914, + "loss": 0.1533, + "step": 10336 + }, + { + "epoch": 0.08973012387045251, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019574465944416438, + "loss": 0.1221, + "step": 10337 + }, + { + "epoch": 0.08973880435065668, + "grad_norm": 0.65625, + "learning_rate": 0.0019574375505154048, + "loss": 0.1094, + "step": 10338 + }, + { + "epoch": 0.08974748483086084, + "grad_norm": 0.81640625, + "learning_rate": 0.0019574285056514857, + "loss": 0.1328, + "step": 10339 + }, + { + "epoch": 0.089756165311065, + "grad_norm": 0.2109375, + "learning_rate": 0.0019574194598498953, + "loss": 0.1455, + "step": 10340 + }, + { + "epoch": 0.08976484579126917, + "grad_norm": 0.08056640625, + "learning_rate": 0.001957410413110644, + "loss": 0.0933, + "step": 10341 + }, + { + "epoch": 0.08977352627147334, + "grad_norm": 0.1796875, + "learning_rate": 0.0019574013654337415, + "loss": 0.1348, + "step": 10342 + }, + { + "epoch": 0.0897822067516775, + "grad_norm": 0.15234375, + "learning_rate": 0.001957392316819198, + "loss": 0.1084, + "step": 10343 + }, + { + "epoch": 0.08979088723188167, + "grad_norm": 0.1875, + "learning_rate": 0.001957383267267023, + "loss": 0.0869, + "step": 10344 + }, + { + "epoch": 0.08979956771208583, + "grad_norm": 0.16796875, + "learning_rate": 0.0019573742167772264, + "loss": 0.1338, + "step": 10345 + }, + { + "epoch": 0.08980824819229, + "grad_norm": 0.345703125, + "learning_rate": 0.0019573651653498184, + "loss": 0.1963, + "step": 10346 + }, + { + "epoch": 0.08981692867249416, + "grad_norm": 0.57421875, + "learning_rate": 0.001957356112984809, + "loss": 0.1147, + "step": 10347 + }, + { + "epoch": 0.08982560915269833, + "grad_norm": 0.3828125, + "learning_rate": 0.001957347059682207, + "loss": 0.1533, + "step": 10348 + }, + { + "epoch": 0.08983428963290249, + "grad_norm": 0.3828125, + "learning_rate": 0.001957338005442024, + "loss": 0.1113, + "step": 10349 + }, + { + "epoch": 0.08984297011310666, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019573289502642682, + "loss": 0.0962, + "step": 10350 + }, + { + "epoch": 0.08985165059331082, + "grad_norm": 0.10888671875, + "learning_rate": 0.001957319894148951, + "loss": 0.1572, + "step": 10351 + }, + { + "epoch": 0.08986033107351499, + "grad_norm": 0.353515625, + "learning_rate": 0.0019573108370960816, + "loss": 0.1445, + "step": 10352 + }, + { + "epoch": 0.08986901155371915, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019573017791056696, + "loss": 0.0615, + "step": 10353 + }, + { + "epoch": 0.08987769203392332, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019572927201777254, + "loss": 0.123, + "step": 10354 + }, + { + "epoch": 0.08988637251412748, + "grad_norm": 0.4453125, + "learning_rate": 0.0019572836603122586, + "loss": 0.0972, + "step": 10355 + }, + { + "epoch": 0.08989505299433165, + "grad_norm": 0.40625, + "learning_rate": 0.0019572745995092796, + "loss": 0.1021, + "step": 10356 + }, + { + "epoch": 0.08990373347453581, + "grad_norm": 0.55859375, + "learning_rate": 0.0019572655377687984, + "loss": 0.1484, + "step": 10357 + }, + { + "epoch": 0.08991241395473998, + "grad_norm": 0.21875, + "learning_rate": 0.0019572564750908236, + "loss": 0.1123, + "step": 10358 + }, + { + "epoch": 0.08992109443494414, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019572474114753664, + "loss": 0.1182, + "step": 10359 + }, + { + "epoch": 0.08992977491514831, + "grad_norm": 0.318359375, + "learning_rate": 0.001957238346922436, + "loss": 0.0952, + "step": 10360 + }, + { + "epoch": 0.08993845539535247, + "grad_norm": 0.2041015625, + "learning_rate": 0.001957229281432043, + "loss": 0.1543, + "step": 10361 + }, + { + "epoch": 0.08994713587555664, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019572202150041967, + "loss": 0.1582, + "step": 10362 + }, + { + "epoch": 0.0899558163557608, + "grad_norm": 0.1484375, + "learning_rate": 0.0019572111476389075, + "loss": 0.1206, + "step": 10363 + }, + { + "epoch": 0.08996449683596497, + "grad_norm": 0.224609375, + "learning_rate": 0.001957202079336185, + "loss": 0.1235, + "step": 10364 + }, + { + "epoch": 0.08997317731616913, + "grad_norm": 0.3671875, + "learning_rate": 0.0019571930100960393, + "loss": 0.1025, + "step": 10365 + }, + { + "epoch": 0.0899818577963733, + "grad_norm": 0.1171875, + "learning_rate": 0.0019571839399184798, + "loss": 0.1514, + "step": 10366 + }, + { + "epoch": 0.08999053827657746, + "grad_norm": 0.4609375, + "learning_rate": 0.0019571748688035173, + "loss": 0.1387, + "step": 10367 + }, + { + "epoch": 0.08999921875678163, + "grad_norm": 0.6875, + "learning_rate": 0.001957165796751161, + "loss": 0.1377, + "step": 10368 + }, + { + "epoch": 0.0900078992369858, + "grad_norm": 0.3203125, + "learning_rate": 0.001957156723761421, + "loss": 0.1147, + "step": 10369 + }, + { + "epoch": 0.09001657971718996, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019571476498343073, + "loss": 0.1006, + "step": 10370 + }, + { + "epoch": 0.09002526019739412, + "grad_norm": 0.1259765625, + "learning_rate": 0.00195713857496983, + "loss": 0.1855, + "step": 10371 + }, + { + "epoch": 0.09003394067759829, + "grad_norm": 0.357421875, + "learning_rate": 0.0019571294991679983, + "loss": 0.1484, + "step": 10372 + }, + { + "epoch": 0.09004262115780245, + "grad_norm": 0.376953125, + "learning_rate": 0.0019571204224288234, + "loss": 0.0977, + "step": 10373 + }, + { + "epoch": 0.09005130163800662, + "grad_norm": 0.98828125, + "learning_rate": 0.001957111344752314, + "loss": 0.1055, + "step": 10374 + }, + { + "epoch": 0.09005998211821079, + "grad_norm": 0.42578125, + "learning_rate": 0.0019571022661384807, + "loss": 0.1235, + "step": 10375 + }, + { + "epoch": 0.09006866259841495, + "grad_norm": 0.20703125, + "learning_rate": 0.0019570931865873333, + "loss": 0.1289, + "step": 10376 + }, + { + "epoch": 0.09007734307861912, + "grad_norm": 0.66796875, + "learning_rate": 0.0019570841060988814, + "loss": 0.1094, + "step": 10377 + }, + { + "epoch": 0.09008602355882328, + "grad_norm": 0.28125, + "learning_rate": 0.0019570750246731355, + "loss": 0.1475, + "step": 10378 + }, + { + "epoch": 0.09009470403902745, + "grad_norm": 0.734375, + "learning_rate": 0.001957065942310105, + "loss": 0.1377, + "step": 10379 + }, + { + "epoch": 0.09010338451923161, + "grad_norm": 0.66796875, + "learning_rate": 0.0019570568590098003, + "loss": 0.1279, + "step": 10380 + }, + { + "epoch": 0.09011206499943576, + "grad_norm": 0.451171875, + "learning_rate": 0.0019570477747722307, + "loss": 0.1084, + "step": 10381 + }, + { + "epoch": 0.09012074547963993, + "grad_norm": 0.19921875, + "learning_rate": 0.001957038689597407, + "loss": 0.1104, + "step": 10382 + }, + { + "epoch": 0.09012942595984409, + "grad_norm": 0.80078125, + "learning_rate": 0.001957029603485338, + "loss": 0.1055, + "step": 10383 + }, + { + "epoch": 0.09013810644004826, + "grad_norm": 0.359375, + "learning_rate": 0.0019570205164360353, + "loss": 0.1011, + "step": 10384 + }, + { + "epoch": 0.09014678692025242, + "grad_norm": 0.341796875, + "learning_rate": 0.001957011428449507, + "loss": 0.207, + "step": 10385 + }, + { + "epoch": 0.09015546740045659, + "grad_norm": 0.10302734375, + "learning_rate": 0.001957002339525764, + "loss": 0.1045, + "step": 10386 + }, + { + "epoch": 0.09016414788066075, + "grad_norm": 0.88671875, + "learning_rate": 0.0019569932496648164, + "loss": 0.1387, + "step": 10387 + }, + { + "epoch": 0.09017282836086492, + "grad_norm": 1.0703125, + "learning_rate": 0.001956984158866674, + "loss": 0.1035, + "step": 10388 + }, + { + "epoch": 0.09018150884106908, + "grad_norm": 0.7578125, + "learning_rate": 0.001956975067131346, + "loss": 0.1143, + "step": 10389 + }, + { + "epoch": 0.09019018932127325, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019569659744588432, + "loss": 0.1289, + "step": 10390 + }, + { + "epoch": 0.09019886980147741, + "grad_norm": 0.474609375, + "learning_rate": 0.0019569568808491754, + "loss": 0.1143, + "step": 10391 + }, + { + "epoch": 0.09020755028168158, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019569477863023523, + "loss": 0.124, + "step": 10392 + }, + { + "epoch": 0.09021623076188574, + "grad_norm": 0.458984375, + "learning_rate": 0.0019569386908183843, + "loss": 0.1172, + "step": 10393 + }, + { + "epoch": 0.09022491124208991, + "grad_norm": 0.5078125, + "learning_rate": 0.0019569295943972806, + "loss": 0.1523, + "step": 10394 + }, + { + "epoch": 0.09023359172229407, + "grad_norm": 0.10302734375, + "learning_rate": 0.001956920497039052, + "loss": 0.1021, + "step": 10395 + }, + { + "epoch": 0.09024227220249824, + "grad_norm": 0.33984375, + "learning_rate": 0.0019569113987437073, + "loss": 0.1006, + "step": 10396 + }, + { + "epoch": 0.0902509526827024, + "grad_norm": 0.11962890625, + "learning_rate": 0.001956902299511258, + "loss": 0.1123, + "step": 10397 + }, + { + "epoch": 0.09025963316290657, + "grad_norm": 0.126953125, + "learning_rate": 0.0019568931993417128, + "loss": 0.2012, + "step": 10398 + }, + { + "epoch": 0.09026831364311073, + "grad_norm": 0.75, + "learning_rate": 0.0019568840982350823, + "loss": 0.1016, + "step": 10399 + }, + { + "epoch": 0.0902769941233149, + "grad_norm": 0.23046875, + "learning_rate": 0.001956874996191376, + "loss": 0.1357, + "step": 10400 + }, + { + "epoch": 0.09028567460351906, + "grad_norm": 0.6953125, + "learning_rate": 0.001956865893210604, + "loss": 0.127, + "step": 10401 + }, + { + "epoch": 0.09029435508372323, + "grad_norm": 0.73828125, + "learning_rate": 0.001956856789292777, + "loss": 0.1338, + "step": 10402 + }, + { + "epoch": 0.0903030355639274, + "grad_norm": 0.384765625, + "learning_rate": 0.0019568476844379033, + "loss": 0.126, + "step": 10403 + }, + { + "epoch": 0.09031171604413156, + "grad_norm": 0.4609375, + "learning_rate": 0.001956838578645994, + "loss": 0.1348, + "step": 10404 + }, + { + "epoch": 0.09032039652433572, + "grad_norm": 0.189453125, + "learning_rate": 0.0019568294719170595, + "loss": 0.1357, + "step": 10405 + }, + { + "epoch": 0.09032907700453989, + "grad_norm": 0.71875, + "learning_rate": 0.001956820364251109, + "loss": 0.1445, + "step": 10406 + }, + { + "epoch": 0.09033775748474406, + "grad_norm": 0.67578125, + "learning_rate": 0.0019568112556481525, + "loss": 0.1201, + "step": 10407 + }, + { + "epoch": 0.09034643796494822, + "grad_norm": 0.546875, + "learning_rate": 0.0019568021461082, + "loss": 0.0991, + "step": 10408 + }, + { + "epoch": 0.09035511844515239, + "grad_norm": 0.4921875, + "learning_rate": 0.001956793035631262, + "loss": 0.0923, + "step": 10409 + }, + { + "epoch": 0.09036379892535655, + "grad_norm": 0.53515625, + "learning_rate": 0.0019567839242173473, + "loss": 0.127, + "step": 10410 + }, + { + "epoch": 0.09037247940556072, + "grad_norm": 1.015625, + "learning_rate": 0.001956774811866467, + "loss": 0.1211, + "step": 10411 + }, + { + "epoch": 0.09038115988576488, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019567656985786308, + "loss": 0.1201, + "step": 10412 + }, + { + "epoch": 0.09038984036596905, + "grad_norm": 0.70703125, + "learning_rate": 0.001956756584353848, + "loss": 0.1069, + "step": 10413 + }, + { + "epoch": 0.09039852084617321, + "grad_norm": 0.4140625, + "learning_rate": 0.0019567474691921296, + "loss": 0.0996, + "step": 10414 + }, + { + "epoch": 0.09040720132637738, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019567383530934845, + "loss": 0.1328, + "step": 10415 + }, + { + "epoch": 0.09041588180658154, + "grad_norm": 0.54296875, + "learning_rate": 0.0019567292360579236, + "loss": 0.1367, + "step": 10416 + }, + { + "epoch": 0.0904245622867857, + "grad_norm": 0.2109375, + "learning_rate": 0.0019567201180854562, + "loss": 0.1211, + "step": 10417 + }, + { + "epoch": 0.09043324276698987, + "grad_norm": 0.07861328125, + "learning_rate": 0.001956710999176093, + "loss": 0.1172, + "step": 10418 + }, + { + "epoch": 0.09044192324719404, + "grad_norm": 0.234375, + "learning_rate": 0.001956701879329843, + "loss": 0.1299, + "step": 10419 + }, + { + "epoch": 0.0904506037273982, + "grad_norm": 0.294921875, + "learning_rate": 0.001956692758546717, + "loss": 0.1641, + "step": 10420 + }, + { + "epoch": 0.09045928420760237, + "grad_norm": 0.1611328125, + "learning_rate": 0.001956683636826725, + "loss": 0.1562, + "step": 10421 + }, + { + "epoch": 0.09046796468780653, + "grad_norm": 0.890625, + "learning_rate": 0.0019566745141698758, + "loss": 0.1582, + "step": 10422 + }, + { + "epoch": 0.0904766451680107, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019566653905761806, + "loss": 0.0928, + "step": 10423 + }, + { + "epoch": 0.09048532564821486, + "grad_norm": 0.54296875, + "learning_rate": 0.0019566562660456492, + "loss": 0.1367, + "step": 10424 + }, + { + "epoch": 0.09049400612841903, + "grad_norm": 0.828125, + "learning_rate": 0.0019566471405782914, + "loss": 0.1377, + "step": 10425 + }, + { + "epoch": 0.09050268660862319, + "grad_norm": 0.30078125, + "learning_rate": 0.001956638014174117, + "loss": 0.1777, + "step": 10426 + }, + { + "epoch": 0.09051136708882736, + "grad_norm": 0.15234375, + "learning_rate": 0.0019566288868331358, + "loss": 0.1138, + "step": 10427 + }, + { + "epoch": 0.09052004756903152, + "grad_norm": 0.50390625, + "learning_rate": 0.001956619758555359, + "loss": 0.127, + "step": 10428 + }, + { + "epoch": 0.09052872804923569, + "grad_norm": 0.25390625, + "learning_rate": 0.001956610629340795, + "loss": 0.2334, + "step": 10429 + }, + { + "epoch": 0.09053740852943985, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019566014991894546, + "loss": 0.1143, + "step": 10430 + }, + { + "epoch": 0.09054608900964402, + "grad_norm": 0.2734375, + "learning_rate": 0.0019565923681013476, + "loss": 0.125, + "step": 10431 + }, + { + "epoch": 0.09055476948984818, + "grad_norm": 0.400390625, + "learning_rate": 0.0019565832360764843, + "loss": 0.1089, + "step": 10432 + }, + { + "epoch": 0.09056344997005235, + "grad_norm": 0.166015625, + "learning_rate": 0.0019565741031148743, + "loss": 0.1494, + "step": 10433 + }, + { + "epoch": 0.09057213045025651, + "grad_norm": 0.11474609375, + "learning_rate": 0.001956564969216528, + "loss": 0.1504, + "step": 10434 + }, + { + "epoch": 0.09058081093046068, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019565558343814546, + "loss": 0.1191, + "step": 10435 + }, + { + "epoch": 0.09058949141066484, + "grad_norm": 0.375, + "learning_rate": 0.001956546698609665, + "loss": 0.1162, + "step": 10436 + }, + { + "epoch": 0.09059817189086901, + "grad_norm": 0.255859375, + "learning_rate": 0.0019565375619011688, + "loss": 0.1226, + "step": 10437 + }, + { + "epoch": 0.09060685237107317, + "grad_norm": 0.373046875, + "learning_rate": 0.0019565284242559757, + "loss": 0.1562, + "step": 10438 + }, + { + "epoch": 0.09061553285127734, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019565192856740962, + "loss": 0.1465, + "step": 10439 + }, + { + "epoch": 0.0906242133314815, + "grad_norm": 0.416015625, + "learning_rate": 0.00195651014615554, + "loss": 0.1187, + "step": 10440 + }, + { + "epoch": 0.09063289381168567, + "grad_norm": 0.7265625, + "learning_rate": 0.001956501005700317, + "loss": 0.1357, + "step": 10441 + }, + { + "epoch": 0.09064157429188983, + "grad_norm": 0.359375, + "learning_rate": 0.001956491864308438, + "loss": 0.1182, + "step": 10442 + }, + { + "epoch": 0.09065025477209399, + "grad_norm": 0.423828125, + "learning_rate": 0.001956482721979912, + "loss": 0.1143, + "step": 10443 + }, + { + "epoch": 0.09065893525229815, + "grad_norm": 1.078125, + "learning_rate": 0.001956473578714749, + "loss": 0.2637, + "step": 10444 + }, + { + "epoch": 0.09066761573250232, + "grad_norm": 0.306640625, + "learning_rate": 0.0019564644345129597, + "loss": 0.166, + "step": 10445 + }, + { + "epoch": 0.09067629621270648, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019564552893745536, + "loss": 0.0874, + "step": 10446 + }, + { + "epoch": 0.09068497669291065, + "grad_norm": 0.169921875, + "learning_rate": 0.001956446143299541, + "loss": 0.0918, + "step": 10447 + }, + { + "epoch": 0.09069365717311481, + "grad_norm": 0.0869140625, + "learning_rate": 0.001956436996287932, + "loss": 0.1079, + "step": 10448 + }, + { + "epoch": 0.09070233765331898, + "grad_norm": 1.421875, + "learning_rate": 0.0019564278483397363, + "loss": 0.1445, + "step": 10449 + }, + { + "epoch": 0.09071101813352314, + "grad_norm": 1.515625, + "learning_rate": 0.0019564186994549637, + "loss": 0.209, + "step": 10450 + }, + { + "epoch": 0.0907196986137273, + "grad_norm": 0.2734375, + "learning_rate": 0.0019564095496336243, + "loss": 0.1084, + "step": 10451 + }, + { + "epoch": 0.09072837909393147, + "grad_norm": 0.1845703125, + "learning_rate": 0.001956400398875729, + "loss": 0.1572, + "step": 10452 + }, + { + "epoch": 0.09073705957413564, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019563912471812868, + "loss": 0.1328, + "step": 10453 + }, + { + "epoch": 0.0907457400543398, + "grad_norm": 0.189453125, + "learning_rate": 0.0019563820945503077, + "loss": 0.125, + "step": 10454 + }, + { + "epoch": 0.09075442053454397, + "grad_norm": 0.099609375, + "learning_rate": 0.001956372940982802, + "loss": 0.1533, + "step": 10455 + }, + { + "epoch": 0.09076310101474813, + "grad_norm": 0.2373046875, + "learning_rate": 0.00195636378647878, + "loss": 0.082, + "step": 10456 + }, + { + "epoch": 0.0907717814949523, + "grad_norm": 0.15234375, + "learning_rate": 0.0019563546310382516, + "loss": 0.1406, + "step": 10457 + }, + { + "epoch": 0.09078046197515646, + "grad_norm": 0.326171875, + "learning_rate": 0.001956345474661226, + "loss": 0.1177, + "step": 10458 + }, + { + "epoch": 0.09078914245536063, + "grad_norm": 0.53515625, + "learning_rate": 0.0019563363173477144, + "loss": 0.0977, + "step": 10459 + }, + { + "epoch": 0.09079782293556479, + "grad_norm": 0.08984375, + "learning_rate": 0.0019563271590977258, + "loss": 0.1484, + "step": 10460 + }, + { + "epoch": 0.09080650341576896, + "grad_norm": 0.10546875, + "learning_rate": 0.001956317999911271, + "loss": 0.1484, + "step": 10461 + }, + { + "epoch": 0.09081518389597312, + "grad_norm": 0.423828125, + "learning_rate": 0.00195630883978836, + "loss": 0.166, + "step": 10462 + }, + { + "epoch": 0.09082386437617729, + "grad_norm": 0.482421875, + "learning_rate": 0.001956299678729002, + "loss": 0.0942, + "step": 10463 + }, + { + "epoch": 0.09083254485638145, + "grad_norm": 0.46484375, + "learning_rate": 0.0019562905167332077, + "loss": 0.1064, + "step": 10464 + }, + { + "epoch": 0.09084122533658562, + "grad_norm": 0.1533203125, + "learning_rate": 0.001956281353800987, + "loss": 0.1387, + "step": 10465 + }, + { + "epoch": 0.09084990581678978, + "grad_norm": 0.71875, + "learning_rate": 0.00195627218993235, + "loss": 0.1504, + "step": 10466 + }, + { + "epoch": 0.09085858629699395, + "grad_norm": 0.2109375, + "learning_rate": 0.0019562630251273062, + "loss": 0.1455, + "step": 10467 + }, + { + "epoch": 0.09086726677719811, + "grad_norm": 0.267578125, + "learning_rate": 0.0019562538593858665, + "loss": 0.1689, + "step": 10468 + }, + { + "epoch": 0.09087594725740228, + "grad_norm": 0.11328125, + "learning_rate": 0.00195624469270804, + "loss": 0.1562, + "step": 10469 + }, + { + "epoch": 0.09088462773760644, + "grad_norm": 0.349609375, + "learning_rate": 0.0019562355250938375, + "loss": 0.123, + "step": 10470 + }, + { + "epoch": 0.09089330821781061, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019562263565432686, + "loss": 0.1079, + "step": 10471 + }, + { + "epoch": 0.09090198869801477, + "grad_norm": 0.640625, + "learning_rate": 0.0019562171870563436, + "loss": 0.1289, + "step": 10472 + }, + { + "epoch": 0.09091066917821894, + "grad_norm": 0.6484375, + "learning_rate": 0.001956208016633072, + "loss": 0.1152, + "step": 10473 + }, + { + "epoch": 0.0909193496584231, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019561988452734646, + "loss": 0.1084, + "step": 10474 + }, + { + "epoch": 0.09092803013862727, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019561896729775306, + "loss": 0.0986, + "step": 10475 + }, + { + "epoch": 0.09093671061883143, + "grad_norm": 0.69921875, + "learning_rate": 0.001956180499745281, + "loss": 0.1172, + "step": 10476 + }, + { + "epoch": 0.0909453910990356, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019561713255767245, + "loss": 0.1904, + "step": 10477 + }, + { + "epoch": 0.09095407157923976, + "grad_norm": 0.22265625, + "learning_rate": 0.001956162150471872, + "loss": 0.1436, + "step": 10478 + }, + { + "epoch": 0.09096275205944393, + "grad_norm": 0.5234375, + "learning_rate": 0.0019561529744307336, + "loss": 0.1221, + "step": 10479 + }, + { + "epoch": 0.0909714325396481, + "grad_norm": 0.26953125, + "learning_rate": 0.0019561437974533196, + "loss": 0.1475, + "step": 10480 + }, + { + "epoch": 0.09098011301985226, + "grad_norm": 0.21875, + "learning_rate": 0.001956134619539639, + "loss": 0.0962, + "step": 10481 + }, + { + "epoch": 0.09098879350005643, + "grad_norm": 0.447265625, + "learning_rate": 0.001956125440689703, + "loss": 0.1162, + "step": 10482 + }, + { + "epoch": 0.09099747398026059, + "grad_norm": 0.2080078125, + "learning_rate": 0.001956116260903521, + "loss": 0.1445, + "step": 10483 + }, + { + "epoch": 0.09100615446046476, + "grad_norm": 0.83984375, + "learning_rate": 0.0019561070801811025, + "loss": 0.1387, + "step": 10484 + }, + { + "epoch": 0.09101483494066892, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019560978985224587, + "loss": 0.1572, + "step": 10485 + }, + { + "epoch": 0.09102351542087309, + "grad_norm": 0.333984375, + "learning_rate": 0.001956088715927599, + "loss": 0.0952, + "step": 10486 + }, + { + "epoch": 0.09103219590107725, + "grad_norm": 0.474609375, + "learning_rate": 0.0019560795323965334, + "loss": 0.1475, + "step": 10487 + }, + { + "epoch": 0.09104087638128142, + "grad_norm": 0.61328125, + "learning_rate": 0.0019560703479292725, + "loss": 0.1221, + "step": 10488 + }, + { + "epoch": 0.09104955686148558, + "grad_norm": 0.357421875, + "learning_rate": 0.0019560611625258255, + "loss": 0.1191, + "step": 10489 + }, + { + "epoch": 0.09105823734168975, + "grad_norm": 0.35546875, + "learning_rate": 0.001956051976186203, + "loss": 0.1235, + "step": 10490 + }, + { + "epoch": 0.09106691782189391, + "grad_norm": 0.67578125, + "learning_rate": 0.0019560427889104147, + "loss": 0.1582, + "step": 10491 + }, + { + "epoch": 0.09107559830209808, + "grad_norm": 0.38671875, + "learning_rate": 0.0019560336006984713, + "loss": 0.0947, + "step": 10492 + }, + { + "epoch": 0.09108427878230224, + "grad_norm": 0.30859375, + "learning_rate": 0.001956024411550382, + "loss": 0.127, + "step": 10493 + }, + { + "epoch": 0.0910929592625064, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019560152214661577, + "loss": 0.125, + "step": 10494 + }, + { + "epoch": 0.09110163974271057, + "grad_norm": 0.10595703125, + "learning_rate": 0.001956006030445808, + "loss": 0.1055, + "step": 10495 + }, + { + "epoch": 0.09111032022291474, + "grad_norm": 0.76953125, + "learning_rate": 0.001955996838489343, + "loss": 0.1377, + "step": 10496 + }, + { + "epoch": 0.0911190007031189, + "grad_norm": 0.322265625, + "learning_rate": 0.001955987645596772, + "loss": 0.1089, + "step": 10497 + }, + { + "epoch": 0.09112768118332307, + "grad_norm": 0.166015625, + "learning_rate": 0.001955978451768106, + "loss": 0.1133, + "step": 10498 + }, + { + "epoch": 0.09113636166352723, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019559692570033553, + "loss": 0.1006, + "step": 10499 + }, + { + "epoch": 0.0911450421437314, + "grad_norm": 0.296875, + "learning_rate": 0.001955960061302529, + "loss": 0.1641, + "step": 10500 + }, + { + "epoch": 0.09115372262393556, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019559508646656383, + "loss": 0.1025, + "step": 10501 + }, + { + "epoch": 0.09116240310413973, + "grad_norm": 0.7421875, + "learning_rate": 0.001955941667092692, + "loss": 0.1465, + "step": 10502 + }, + { + "epoch": 0.09117108358434389, + "grad_norm": 0.0654296875, + "learning_rate": 0.0019559324685837014, + "loss": 0.1006, + "step": 10503 + }, + { + "epoch": 0.09117976406454804, + "grad_norm": 0.80078125, + "learning_rate": 0.0019559232691386754, + "loss": 0.1216, + "step": 10504 + }, + { + "epoch": 0.09118844454475221, + "grad_norm": 0.412109375, + "learning_rate": 0.0019559140687576246, + "loss": 0.1011, + "step": 10505 + }, + { + "epoch": 0.09119712502495637, + "grad_norm": 0.482421875, + "learning_rate": 0.0019559048674405587, + "loss": 0.0869, + "step": 10506 + }, + { + "epoch": 0.09120580550516054, + "grad_norm": 0.373046875, + "learning_rate": 0.001955895665187489, + "loss": 0.1025, + "step": 10507 + }, + { + "epoch": 0.0912144859853647, + "grad_norm": 0.1435546875, + "learning_rate": 0.001955886461998424, + "loss": 0.1035, + "step": 10508 + }, + { + "epoch": 0.09122316646556887, + "grad_norm": 0.318359375, + "learning_rate": 0.001955877257873374, + "loss": 0.1011, + "step": 10509 + }, + { + "epoch": 0.09123184694577303, + "grad_norm": 0.4921875, + "learning_rate": 0.0019558680528123503, + "loss": 0.167, + "step": 10510 + }, + { + "epoch": 0.0912405274259772, + "grad_norm": 0.14453125, + "learning_rate": 0.001955858846815362, + "loss": 0.1289, + "step": 10511 + }, + { + "epoch": 0.09124920790618136, + "grad_norm": 0.3671875, + "learning_rate": 0.001955849639882419, + "loss": 0.1426, + "step": 10512 + }, + { + "epoch": 0.09125788838638553, + "grad_norm": 0.412109375, + "learning_rate": 0.001955840432013532, + "loss": 0.0796, + "step": 10513 + }, + { + "epoch": 0.0912665688665897, + "grad_norm": 0.5625, + "learning_rate": 0.001955831223208711, + "loss": 0.1152, + "step": 10514 + }, + { + "epoch": 0.09127524934679386, + "grad_norm": 0.138671875, + "learning_rate": 0.0019558220134679653, + "loss": 0.1245, + "step": 10515 + }, + { + "epoch": 0.09128392982699803, + "grad_norm": 0.52734375, + "learning_rate": 0.001955812802791306, + "loss": 0.1162, + "step": 10516 + }, + { + "epoch": 0.09129261030720219, + "grad_norm": 0.62109375, + "learning_rate": 0.001955803591178742, + "loss": 0.1318, + "step": 10517 + }, + { + "epoch": 0.09130129078740636, + "grad_norm": 1.2265625, + "learning_rate": 0.0019557943786302845, + "loss": 0.1328, + "step": 10518 + }, + { + "epoch": 0.09130997126761052, + "grad_norm": 0.1552734375, + "learning_rate": 0.001955785165145943, + "loss": 0.0918, + "step": 10519 + }, + { + "epoch": 0.09131865174781469, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019557759507257276, + "loss": 0.1357, + "step": 10520 + }, + { + "epoch": 0.09132733222801885, + "grad_norm": 0.1875, + "learning_rate": 0.0019557667353696487, + "loss": 0.1152, + "step": 10521 + }, + { + "epoch": 0.09133601270822302, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019557575190777164, + "loss": 0.1328, + "step": 10522 + }, + { + "epoch": 0.09134469318842718, + "grad_norm": 0.6015625, + "learning_rate": 0.0019557483018499402, + "loss": 0.0972, + "step": 10523 + }, + { + "epoch": 0.09135337366863135, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019557390836863306, + "loss": 0.1279, + "step": 10524 + }, + { + "epoch": 0.09136205414883551, + "grad_norm": 0.58203125, + "learning_rate": 0.0019557298645868975, + "loss": 0.1123, + "step": 10525 + }, + { + "epoch": 0.09137073462903968, + "grad_norm": 1.2578125, + "learning_rate": 0.001955720644551651, + "loss": 0.2393, + "step": 10526 + }, + { + "epoch": 0.09137941510924384, + "grad_norm": 0.33984375, + "learning_rate": 0.0019557114235806013, + "loss": 0.1016, + "step": 10527 + }, + { + "epoch": 0.091388095589448, + "grad_norm": 0.474609375, + "learning_rate": 0.0019557022016737586, + "loss": 0.1357, + "step": 10528 + }, + { + "epoch": 0.09139677606965217, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019556929788311327, + "loss": 0.1504, + "step": 10529 + }, + { + "epoch": 0.09140545654985634, + "grad_norm": 0.54296875, + "learning_rate": 0.001955683755052734, + "loss": 0.1514, + "step": 10530 + }, + { + "epoch": 0.0914141370300605, + "grad_norm": 0.458984375, + "learning_rate": 0.0019556745303385727, + "loss": 0.0962, + "step": 10531 + }, + { + "epoch": 0.09142281751026467, + "grad_norm": 0.69140625, + "learning_rate": 0.0019556653046886576, + "loss": 0.1211, + "step": 10532 + }, + { + "epoch": 0.09143149799046883, + "grad_norm": 0.703125, + "learning_rate": 0.0019556560781030007, + "loss": 0.1406, + "step": 10533 + }, + { + "epoch": 0.091440178470673, + "grad_norm": 0.427734375, + "learning_rate": 0.001955646850581611, + "loss": 0.1201, + "step": 10534 + }, + { + "epoch": 0.09144885895087716, + "grad_norm": 0.62890625, + "learning_rate": 0.0019556376221244983, + "loss": 0.1084, + "step": 10535 + }, + { + "epoch": 0.09145753943108133, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019556283927316736, + "loss": 0.106, + "step": 10536 + }, + { + "epoch": 0.09146621991128549, + "grad_norm": 0.0849609375, + "learning_rate": 0.001955619162403146, + "loss": 0.0898, + "step": 10537 + }, + { + "epoch": 0.09147490039148966, + "grad_norm": 0.3046875, + "learning_rate": 0.0019556099311389265, + "loss": 0.1201, + "step": 10538 + }, + { + "epoch": 0.09148358087169382, + "grad_norm": 0.1640625, + "learning_rate": 0.0019556006989390248, + "loss": 0.1064, + "step": 10539 + }, + { + "epoch": 0.09149226135189799, + "grad_norm": 0.216796875, + "learning_rate": 0.0019555914658034514, + "loss": 0.1074, + "step": 10540 + }, + { + "epoch": 0.09150094183210215, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019555822317322154, + "loss": 0.1445, + "step": 10541 + }, + { + "epoch": 0.09150962231230632, + "grad_norm": 0.25390625, + "learning_rate": 0.0019555729967253277, + "loss": 0.1182, + "step": 10542 + }, + { + "epoch": 0.09151830279251048, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019555637607827987, + "loss": 0.1309, + "step": 10543 + }, + { + "epoch": 0.09152698327271465, + "grad_norm": 0.482421875, + "learning_rate": 0.0019555545239046375, + "loss": 0.1416, + "step": 10544 + }, + { + "epoch": 0.09153566375291881, + "grad_norm": 0.625, + "learning_rate": 0.001955545286090855, + "loss": 0.1484, + "step": 10545 + }, + { + "epoch": 0.09154434423312298, + "grad_norm": 0.20703125, + "learning_rate": 0.001955536047341461, + "loss": 0.1001, + "step": 10546 + }, + { + "epoch": 0.09155302471332714, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019555268076564655, + "loss": 0.1118, + "step": 10547 + }, + { + "epoch": 0.09156170519353131, + "grad_norm": 0.162109375, + "learning_rate": 0.001955517567035879, + "loss": 0.1572, + "step": 10548 + }, + { + "epoch": 0.09157038567373547, + "grad_norm": 0.17578125, + "learning_rate": 0.0019555083254797112, + "loss": 0.1348, + "step": 10549 + }, + { + "epoch": 0.09157906615393964, + "grad_norm": 0.462890625, + "learning_rate": 0.0019554990829879723, + "loss": 0.1104, + "step": 10550 + }, + { + "epoch": 0.0915877466341438, + "grad_norm": 0.375, + "learning_rate": 0.001955489839560672, + "loss": 0.1074, + "step": 10551 + }, + { + "epoch": 0.09159642711434797, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019554805951978215, + "loss": 0.1104, + "step": 10552 + }, + { + "epoch": 0.09160510759455213, + "grad_norm": 0.150390625, + "learning_rate": 0.0019554713498994305, + "loss": 0.1309, + "step": 10553 + }, + { + "epoch": 0.0916137880747563, + "grad_norm": 0.126953125, + "learning_rate": 0.0019554621036655083, + "loss": 0.0996, + "step": 10554 + }, + { + "epoch": 0.09162246855496047, + "grad_norm": 0.06689453125, + "learning_rate": 0.001955452856496066, + "loss": 0.123, + "step": 10555 + }, + { + "epoch": 0.09163114903516463, + "grad_norm": 0.349609375, + "learning_rate": 0.0019554436083911125, + "loss": 0.1191, + "step": 10556 + }, + { + "epoch": 0.0916398295153688, + "grad_norm": 0.150390625, + "learning_rate": 0.0019554343593506597, + "loss": 0.1338, + "step": 10557 + }, + { + "epoch": 0.09164850999557296, + "grad_norm": 0.60546875, + "learning_rate": 0.001955425109374716, + "loss": 0.0977, + "step": 10558 + }, + { + "epoch": 0.09165719047577713, + "grad_norm": 0.5703125, + "learning_rate": 0.001955415858463293, + "loss": 0.1182, + "step": 10559 + }, + { + "epoch": 0.09166587095598129, + "grad_norm": 0.265625, + "learning_rate": 0.0019554066066163997, + "loss": 0.1387, + "step": 10560 + }, + { + "epoch": 0.09167455143618546, + "grad_norm": 0.171875, + "learning_rate": 0.0019553973538340465, + "loss": 0.1167, + "step": 10561 + }, + { + "epoch": 0.09168323191638962, + "grad_norm": 0.103515625, + "learning_rate": 0.0019553881001162436, + "loss": 0.0859, + "step": 10562 + }, + { + "epoch": 0.09169191239659379, + "grad_norm": 0.1669921875, + "learning_rate": 0.001955378845463001, + "loss": 0.1152, + "step": 10563 + }, + { + "epoch": 0.09170059287679795, + "grad_norm": 0.326171875, + "learning_rate": 0.0019553695898743293, + "loss": 0.1201, + "step": 10564 + }, + { + "epoch": 0.09170927335700212, + "grad_norm": 0.51953125, + "learning_rate": 0.001955360333350238, + "loss": 0.1211, + "step": 10565 + }, + { + "epoch": 0.09171795383720627, + "grad_norm": 0.490234375, + "learning_rate": 0.001955351075890738, + "loss": 0.1318, + "step": 10566 + }, + { + "epoch": 0.09172663431741043, + "grad_norm": 1.171875, + "learning_rate": 0.001955341817495838, + "loss": 0.1797, + "step": 10567 + }, + { + "epoch": 0.0917353147976146, + "grad_norm": 0.103515625, + "learning_rate": 0.0019553325581655495, + "loss": 0.0918, + "step": 10568 + }, + { + "epoch": 0.09174399527781876, + "grad_norm": 0.169921875, + "learning_rate": 0.001955323297899883, + "loss": 0.1016, + "step": 10569 + }, + { + "epoch": 0.09175267575802293, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019553140366988467, + "loss": 0.1011, + "step": 10570 + }, + { + "epoch": 0.0917613562382271, + "grad_norm": 0.69921875, + "learning_rate": 0.001955304774562452, + "loss": 0.0889, + "step": 10571 + }, + { + "epoch": 0.09177003671843126, + "grad_norm": 0.150390625, + "learning_rate": 0.0019552955114907093, + "loss": 0.1416, + "step": 10572 + }, + { + "epoch": 0.09177871719863542, + "grad_norm": 0.095703125, + "learning_rate": 0.001955286247483628, + "loss": 0.1406, + "step": 10573 + }, + { + "epoch": 0.09178739767883959, + "grad_norm": 0.2236328125, + "learning_rate": 0.001955276982541218, + "loss": 0.1147, + "step": 10574 + }, + { + "epoch": 0.09179607815904375, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019552677166634904, + "loss": 0.1436, + "step": 10575 + }, + { + "epoch": 0.09180475863924792, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019552584498504547, + "loss": 0.1934, + "step": 10576 + }, + { + "epoch": 0.09181343911945208, + "grad_norm": 0.486328125, + "learning_rate": 0.0019552491821021216, + "loss": 0.1631, + "step": 10577 + }, + { + "epoch": 0.09182211959965625, + "grad_norm": 0.171875, + "learning_rate": 0.0019552399134185007, + "loss": 0.1406, + "step": 10578 + }, + { + "epoch": 0.09183080007986041, + "grad_norm": 0.3046875, + "learning_rate": 0.0019552306437996022, + "loss": 0.1426, + "step": 10579 + }, + { + "epoch": 0.09183948056006458, + "grad_norm": 0.205078125, + "learning_rate": 0.0019552213732454363, + "loss": 0.0933, + "step": 10580 + }, + { + "epoch": 0.09184816104026874, + "grad_norm": 0.7578125, + "learning_rate": 0.0019552121017560133, + "loss": 0.1182, + "step": 10581 + }, + { + "epoch": 0.09185684152047291, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019552028293313433, + "loss": 0.123, + "step": 10582 + }, + { + "epoch": 0.09186552200067707, + "grad_norm": 1.2734375, + "learning_rate": 0.001955193555971436, + "loss": 0.1562, + "step": 10583 + }, + { + "epoch": 0.09187420248088124, + "grad_norm": 0.3359375, + "learning_rate": 0.001955184281676302, + "loss": 0.2246, + "step": 10584 + }, + { + "epoch": 0.0918828829610854, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019551750064459514, + "loss": 0.0952, + "step": 10585 + }, + { + "epoch": 0.09189156344128957, + "grad_norm": 0.302734375, + "learning_rate": 0.001955165730280394, + "loss": 0.1289, + "step": 10586 + }, + { + "epoch": 0.09190024392149373, + "grad_norm": 0.97265625, + "learning_rate": 0.0019551564531796405, + "loss": 0.1699, + "step": 10587 + }, + { + "epoch": 0.0919089244016979, + "grad_norm": 0.166015625, + "learning_rate": 0.0019551471751437007, + "loss": 0.1738, + "step": 10588 + }, + { + "epoch": 0.09191760488190207, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019551378961725845, + "loss": 0.1055, + "step": 10589 + }, + { + "epoch": 0.09192628536210623, + "grad_norm": 0.408203125, + "learning_rate": 0.0019551286162663027, + "loss": 0.1299, + "step": 10590 + }, + { + "epoch": 0.0919349658423104, + "grad_norm": 0.12353515625, + "learning_rate": 0.001955119335424865, + "loss": 0.1069, + "step": 10591 + }, + { + "epoch": 0.09194364632251456, + "grad_norm": 0.212890625, + "learning_rate": 0.0019551100536482815, + "loss": 0.1016, + "step": 10592 + }, + { + "epoch": 0.09195232680271873, + "grad_norm": 0.546875, + "learning_rate": 0.001955100770936563, + "loss": 0.1138, + "step": 10593 + }, + { + "epoch": 0.09196100728292289, + "grad_norm": 2.453125, + "learning_rate": 0.0019550914872897185, + "loss": 0.1943, + "step": 10594 + }, + { + "epoch": 0.09196968776312706, + "grad_norm": 0.83203125, + "learning_rate": 0.001955082202707759, + "loss": 0.1602, + "step": 10595 + }, + { + "epoch": 0.09197836824333122, + "grad_norm": 0.44140625, + "learning_rate": 0.0019550729171906945, + "loss": 0.1138, + "step": 10596 + }, + { + "epoch": 0.09198704872353539, + "grad_norm": 0.61328125, + "learning_rate": 0.001955063630738535, + "loss": 0.1582, + "step": 10597 + }, + { + "epoch": 0.09199572920373955, + "grad_norm": 0.408203125, + "learning_rate": 0.001955054343351291, + "loss": 0.1211, + "step": 10598 + }, + { + "epoch": 0.09200440968394372, + "grad_norm": 0.55078125, + "learning_rate": 0.001955045055028972, + "loss": 0.1465, + "step": 10599 + }, + { + "epoch": 0.09201309016414788, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019550357657715893, + "loss": 0.1011, + "step": 10600 + }, + { + "epoch": 0.09202177064435205, + "grad_norm": 0.333984375, + "learning_rate": 0.001955026475579152, + "loss": 0.1182, + "step": 10601 + }, + { + "epoch": 0.09203045112455621, + "grad_norm": 0.0703125, + "learning_rate": 0.0019550171844516704, + "loss": 0.1191, + "step": 10602 + }, + { + "epoch": 0.09203913160476038, + "grad_norm": 0.3984375, + "learning_rate": 0.0019550078923891547, + "loss": 0.123, + "step": 10603 + }, + { + "epoch": 0.09204781208496454, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019549985993916157, + "loss": 0.1162, + "step": 10604 + }, + { + "epoch": 0.09205649256516871, + "grad_norm": 0.765625, + "learning_rate": 0.001954989305459063, + "loss": 0.1328, + "step": 10605 + }, + { + "epoch": 0.09206517304537287, + "grad_norm": 0.357421875, + "learning_rate": 0.0019549800105915067, + "loss": 0.125, + "step": 10606 + }, + { + "epoch": 0.09207385352557704, + "grad_norm": 0.18359375, + "learning_rate": 0.0019549707147889566, + "loss": 0.1621, + "step": 10607 + }, + { + "epoch": 0.0920825340057812, + "grad_norm": 0.078125, + "learning_rate": 0.001954961418051424, + "loss": 0.0767, + "step": 10608 + }, + { + "epoch": 0.09209121448598537, + "grad_norm": 0.85546875, + "learning_rate": 0.001954952120378918, + "loss": 0.1074, + "step": 10609 + }, + { + "epoch": 0.09209989496618953, + "grad_norm": 0.203125, + "learning_rate": 0.00195494282177145, + "loss": 0.1084, + "step": 10610 + }, + { + "epoch": 0.0921085754463937, + "grad_norm": 0.224609375, + "learning_rate": 0.001954933522229029, + "loss": 0.1064, + "step": 10611 + }, + { + "epoch": 0.09211725592659786, + "grad_norm": 1.375, + "learning_rate": 0.001954924221751665, + "loss": 0.1387, + "step": 10612 + }, + { + "epoch": 0.09212593640680203, + "grad_norm": 0.376953125, + "learning_rate": 0.0019549149203393693, + "loss": 0.1758, + "step": 10613 + }, + { + "epoch": 0.0921346168870062, + "grad_norm": 0.263671875, + "learning_rate": 0.001954905617992151, + "loss": 0.1206, + "step": 10614 + }, + { + "epoch": 0.09214329736721036, + "grad_norm": 0.146484375, + "learning_rate": 0.0019548963147100217, + "loss": 0.106, + "step": 10615 + }, + { + "epoch": 0.09215197784741452, + "grad_norm": 0.2158203125, + "learning_rate": 0.00195488701049299, + "loss": 0.1221, + "step": 10616 + }, + { + "epoch": 0.09216065832761869, + "grad_norm": 0.375, + "learning_rate": 0.0019548777053410666, + "loss": 0.1094, + "step": 10617 + }, + { + "epoch": 0.09216933880782285, + "grad_norm": 0.23828125, + "learning_rate": 0.001954868399254262, + "loss": 0.127, + "step": 10618 + }, + { + "epoch": 0.09217801928802702, + "grad_norm": 0.5546875, + "learning_rate": 0.001954859092232586, + "loss": 0.1104, + "step": 10619 + }, + { + "epoch": 0.09218669976823118, + "grad_norm": 0.34375, + "learning_rate": 0.001954849784276049, + "loss": 0.0869, + "step": 10620 + }, + { + "epoch": 0.09219538024843535, + "grad_norm": 0.625, + "learning_rate": 0.001954840475384661, + "loss": 0.1387, + "step": 10621 + }, + { + "epoch": 0.09220406072863951, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019548311655584326, + "loss": 0.1533, + "step": 10622 + }, + { + "epoch": 0.09221274120884368, + "grad_norm": 0.296875, + "learning_rate": 0.0019548218547973733, + "loss": 0.1289, + "step": 10623 + }, + { + "epoch": 0.09222142168904784, + "grad_norm": 0.06591796875, + "learning_rate": 0.001954812543101494, + "loss": 0.1045, + "step": 10624 + }, + { + "epoch": 0.09223010216925201, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019548032304708045, + "loss": 0.1064, + "step": 10625 + }, + { + "epoch": 0.09223878264945617, + "grad_norm": 0.150390625, + "learning_rate": 0.001954793916905315, + "loss": 0.1357, + "step": 10626 + }, + { + "epoch": 0.09224746312966033, + "grad_norm": 0.7109375, + "learning_rate": 0.001954784602405036, + "loss": 0.1406, + "step": 10627 + }, + { + "epoch": 0.09225614360986449, + "grad_norm": 0.09130859375, + "learning_rate": 0.001954775286969977, + "loss": 0.1069, + "step": 10628 + }, + { + "epoch": 0.09226482409006866, + "grad_norm": 0.640625, + "learning_rate": 0.0019547659706001487, + "loss": 0.1543, + "step": 10629 + }, + { + "epoch": 0.09227350457027282, + "grad_norm": 0.5859375, + "learning_rate": 0.001954756653295561, + "loss": 0.1045, + "step": 10630 + }, + { + "epoch": 0.09228218505047699, + "grad_norm": 0.734375, + "learning_rate": 0.0019547473350562243, + "loss": 0.1611, + "step": 10631 + }, + { + "epoch": 0.09229086553068115, + "grad_norm": 0.1640625, + "learning_rate": 0.001954738015882149, + "loss": 0.1055, + "step": 10632 + }, + { + "epoch": 0.09229954601088532, + "grad_norm": 0.59765625, + "learning_rate": 0.001954728695773345, + "loss": 0.1406, + "step": 10633 + }, + { + "epoch": 0.09230822649108948, + "grad_norm": 0.390625, + "learning_rate": 0.0019547193747298224, + "loss": 0.1001, + "step": 10634 + }, + { + "epoch": 0.09231690697129365, + "grad_norm": 0.6796875, + "learning_rate": 0.0019547100527515914, + "loss": 0.1055, + "step": 10635 + }, + { + "epoch": 0.09232558745149781, + "grad_norm": 0.6328125, + "learning_rate": 0.0019547007298386623, + "loss": 0.1191, + "step": 10636 + }, + { + "epoch": 0.09233426793170198, + "grad_norm": 0.57421875, + "learning_rate": 0.0019546914059910456, + "loss": 0.1084, + "step": 10637 + }, + { + "epoch": 0.09234294841190614, + "grad_norm": 0.490234375, + "learning_rate": 0.001954682081208751, + "loss": 0.1182, + "step": 10638 + }, + { + "epoch": 0.09235162889211031, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019546727554917895, + "loss": 0.1099, + "step": 10639 + }, + { + "epoch": 0.09236030937231447, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019546634288401706, + "loss": 0.1152, + "step": 10640 + }, + { + "epoch": 0.09236898985251864, + "grad_norm": 0.365234375, + "learning_rate": 0.0019546541012539042, + "loss": 0.1445, + "step": 10641 + }, + { + "epoch": 0.0923776703327228, + "grad_norm": 0.25, + "learning_rate": 0.001954644772733001, + "loss": 0.0977, + "step": 10642 + }, + { + "epoch": 0.09238635081292697, + "grad_norm": 0.392578125, + "learning_rate": 0.001954635443277471, + "loss": 0.1001, + "step": 10643 + }, + { + "epoch": 0.09239503129313113, + "grad_norm": 1.8828125, + "learning_rate": 0.001954626112887325, + "loss": 0.2021, + "step": 10644 + }, + { + "epoch": 0.0924037117733353, + "grad_norm": 0.1015625, + "learning_rate": 0.001954616781562572, + "loss": 0.1289, + "step": 10645 + }, + { + "epoch": 0.09241239225353946, + "grad_norm": 0.154296875, + "learning_rate": 0.0019546074493032237, + "loss": 0.1309, + "step": 10646 + }, + { + "epoch": 0.09242107273374363, + "grad_norm": 0.90234375, + "learning_rate": 0.001954598116109289, + "loss": 0.124, + "step": 10647 + }, + { + "epoch": 0.0924297532139478, + "grad_norm": 0.296875, + "learning_rate": 0.001954588781980779, + "loss": 0.1016, + "step": 10648 + }, + { + "epoch": 0.09243843369415196, + "grad_norm": 0.1923828125, + "learning_rate": 0.001954579446917703, + "loss": 0.1152, + "step": 10649 + }, + { + "epoch": 0.09244711417435612, + "grad_norm": 0.330078125, + "learning_rate": 0.0019545701109200723, + "loss": 0.1377, + "step": 10650 + }, + { + "epoch": 0.09245579465456029, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019545607739878965, + "loss": 0.1191, + "step": 10651 + }, + { + "epoch": 0.09246447513476445, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019545514361211857, + "loss": 0.1426, + "step": 10652 + }, + { + "epoch": 0.09247315561496862, + "grad_norm": 0.5390625, + "learning_rate": 0.00195454209731995, + "loss": 0.1157, + "step": 10653 + }, + { + "epoch": 0.09248183609517278, + "grad_norm": 0.1875, + "learning_rate": 0.0019545327575842004, + "loss": 0.103, + "step": 10654 + }, + { + "epoch": 0.09249051657537695, + "grad_norm": 0.3046875, + "learning_rate": 0.0019545234169139466, + "loss": 0.1641, + "step": 10655 + }, + { + "epoch": 0.09249919705558111, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019545140753091986, + "loss": 0.103, + "step": 10656 + }, + { + "epoch": 0.09250787753578528, + "grad_norm": 0.6015625, + "learning_rate": 0.001954504732769967, + "loss": 0.127, + "step": 10657 + }, + { + "epoch": 0.09251655801598944, + "grad_norm": 0.08203125, + "learning_rate": 0.001954495389296262, + "loss": 0.1191, + "step": 10658 + }, + { + "epoch": 0.09252523849619361, + "grad_norm": 0.404296875, + "learning_rate": 0.001954486044888093, + "loss": 0.082, + "step": 10659 + }, + { + "epoch": 0.09253391897639777, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019544766995454716, + "loss": 0.1094, + "step": 10660 + }, + { + "epoch": 0.09254259945660194, + "grad_norm": 0.150390625, + "learning_rate": 0.001954467353268407, + "loss": 0.1201, + "step": 10661 + }, + { + "epoch": 0.0925512799368061, + "grad_norm": 0.1923828125, + "learning_rate": 0.00195445800605691, + "loss": 0.1504, + "step": 10662 + }, + { + "epoch": 0.09255996041701027, + "grad_norm": 1.8125, + "learning_rate": 0.0019544486579109906, + "loss": 0.127, + "step": 10663 + }, + { + "epoch": 0.09256864089721444, + "grad_norm": 0.283203125, + "learning_rate": 0.0019544393088306586, + "loss": 0.1523, + "step": 10664 + }, + { + "epoch": 0.0925773213774186, + "grad_norm": 0.2578125, + "learning_rate": 0.001954429958815925, + "loss": 0.1494, + "step": 10665 + }, + { + "epoch": 0.09258600185762277, + "grad_norm": 0.061767578125, + "learning_rate": 0.001954420607866799, + "loss": 0.1104, + "step": 10666 + }, + { + "epoch": 0.09259468233782693, + "grad_norm": 0.10400390625, + "learning_rate": 0.001954411255983292, + "loss": 0.0996, + "step": 10667 + }, + { + "epoch": 0.0926033628180311, + "grad_norm": 0.25390625, + "learning_rate": 0.0019544019031654133, + "loss": 0.1309, + "step": 10668 + }, + { + "epoch": 0.09261204329823526, + "grad_norm": 0.5390625, + "learning_rate": 0.001954392549413174, + "loss": 0.1201, + "step": 10669 + }, + { + "epoch": 0.09262072377843943, + "grad_norm": 0.73828125, + "learning_rate": 0.0019543831947265835, + "loss": 0.1104, + "step": 10670 + }, + { + "epoch": 0.09262940425864359, + "grad_norm": 0.81640625, + "learning_rate": 0.0019543738391056527, + "loss": 0.1104, + "step": 10671 + }, + { + "epoch": 0.09263808473884776, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019543644825503912, + "loss": 0.1104, + "step": 10672 + }, + { + "epoch": 0.09264676521905192, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019543551250608094, + "loss": 0.125, + "step": 10673 + }, + { + "epoch": 0.09265544569925609, + "grad_norm": 0.203125, + "learning_rate": 0.001954345766636918, + "loss": 0.1445, + "step": 10674 + }, + { + "epoch": 0.09266412617946025, + "grad_norm": 0.53515625, + "learning_rate": 0.001954336407278727, + "loss": 0.1318, + "step": 10675 + }, + { + "epoch": 0.09267280665966442, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019543270469862466, + "loss": 0.1348, + "step": 10676 + }, + { + "epoch": 0.09268148713986858, + "grad_norm": 0.37109375, + "learning_rate": 0.0019543176857594866, + "loss": 0.1543, + "step": 10677 + }, + { + "epoch": 0.09269016762007275, + "grad_norm": 0.400390625, + "learning_rate": 0.0019543083235984576, + "loss": 0.1514, + "step": 10678 + }, + { + "epoch": 0.09269884810027691, + "grad_norm": 0.07958984375, + "learning_rate": 0.00195429896050317, + "loss": 0.1436, + "step": 10679 + }, + { + "epoch": 0.09270752858048108, + "grad_norm": 0.2431640625, + "learning_rate": 0.001954289596473634, + "loss": 0.0967, + "step": 10680 + }, + { + "epoch": 0.09271620906068524, + "grad_norm": 0.189453125, + "learning_rate": 0.00195428023150986, + "loss": 0.1182, + "step": 10681 + }, + { + "epoch": 0.09272488954088941, + "grad_norm": 1.359375, + "learning_rate": 0.0019542708656118575, + "loss": 0.1523, + "step": 10682 + }, + { + "epoch": 0.09273357002109357, + "grad_norm": 0.130859375, + "learning_rate": 0.0019542614987796375, + "loss": 0.1201, + "step": 10683 + }, + { + "epoch": 0.09274225050129774, + "grad_norm": 0.08203125, + "learning_rate": 0.0019542521310132095, + "loss": 0.0972, + "step": 10684 + }, + { + "epoch": 0.0927509309815019, + "grad_norm": 0.08935546875, + "learning_rate": 0.0019542427623125843, + "loss": 0.0889, + "step": 10685 + }, + { + "epoch": 0.09275961146170607, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019542333926777723, + "loss": 0.1289, + "step": 10686 + }, + { + "epoch": 0.09276829194191023, + "grad_norm": 0.55078125, + "learning_rate": 0.0019542240221087836, + "loss": 0.0869, + "step": 10687 + }, + { + "epoch": 0.0927769724221144, + "grad_norm": 0.05078125, + "learning_rate": 0.001954214650605628, + "loss": 0.1001, + "step": 10688 + }, + { + "epoch": 0.09278565290231855, + "grad_norm": 0.42578125, + "learning_rate": 0.0019542052781683166, + "loss": 0.1416, + "step": 10689 + }, + { + "epoch": 0.09279433338252271, + "grad_norm": 0.7890625, + "learning_rate": 0.0019541959047968587, + "loss": 0.1182, + "step": 10690 + }, + { + "epoch": 0.09280301386272688, + "grad_norm": 0.3515625, + "learning_rate": 0.001954186530491265, + "loss": 0.1045, + "step": 10691 + }, + { + "epoch": 0.09281169434293104, + "grad_norm": 0.1787109375, + "learning_rate": 0.001954177155251546, + "loss": 0.1426, + "step": 10692 + }, + { + "epoch": 0.09282037482313521, + "grad_norm": 0.11376953125, + "learning_rate": 0.0019541677790777113, + "loss": 0.1475, + "step": 10693 + }, + { + "epoch": 0.09282905530333938, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019541584019697722, + "loss": 0.125, + "step": 10694 + }, + { + "epoch": 0.09283773578354354, + "grad_norm": 0.1279296875, + "learning_rate": 0.001954149023927738, + "loss": 0.1025, + "step": 10695 + }, + { + "epoch": 0.0928464162637477, + "grad_norm": 0.177734375, + "learning_rate": 0.001954139644951619, + "loss": 0.1289, + "step": 10696 + }, + { + "epoch": 0.09285509674395187, + "grad_norm": 0.2890625, + "learning_rate": 0.001954130265041426, + "loss": 0.1328, + "step": 10697 + }, + { + "epoch": 0.09286377722415604, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019541208841971688, + "loss": 0.1455, + "step": 10698 + }, + { + "epoch": 0.0928724577043602, + "grad_norm": 0.150390625, + "learning_rate": 0.001954111502418858, + "loss": 0.1406, + "step": 10699 + }, + { + "epoch": 0.09288113818456437, + "grad_norm": 0.86328125, + "learning_rate": 0.0019541021197065035, + "loss": 0.1387, + "step": 10700 + }, + { + "epoch": 0.09288981866476853, + "grad_norm": 0.390625, + "learning_rate": 0.001954092736060116, + "loss": 0.1416, + "step": 10701 + }, + { + "epoch": 0.0928984991449727, + "grad_norm": 0.64453125, + "learning_rate": 0.0019540833514797053, + "loss": 0.0942, + "step": 10702 + }, + { + "epoch": 0.09290717962517686, + "grad_norm": 0.373046875, + "learning_rate": 0.001954073965965282, + "loss": 0.0859, + "step": 10703 + }, + { + "epoch": 0.09291586010538103, + "grad_norm": 0.232421875, + "learning_rate": 0.001954064579516856, + "loss": 0.1011, + "step": 10704 + }, + { + "epoch": 0.09292454058558519, + "grad_norm": 0.13671875, + "learning_rate": 0.0019540551921344382, + "loss": 0.1338, + "step": 10705 + }, + { + "epoch": 0.09293322106578936, + "grad_norm": 0.12890625, + "learning_rate": 0.001954045803818038, + "loss": 0.1279, + "step": 10706 + }, + { + "epoch": 0.09294190154599352, + "grad_norm": 0.27734375, + "learning_rate": 0.0019540364145676664, + "loss": 0.082, + "step": 10707 + }, + { + "epoch": 0.09295058202619769, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019540270243833336, + "loss": 0.127, + "step": 10708 + }, + { + "epoch": 0.09295926250640185, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019540176332650495, + "loss": 0.1162, + "step": 10709 + }, + { + "epoch": 0.09296794298660602, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019540082412128246, + "loss": 0.1484, + "step": 10710 + }, + { + "epoch": 0.09297662346681018, + "grad_norm": 1.21875, + "learning_rate": 0.001953998848226669, + "loss": 0.1152, + "step": 10711 + }, + { + "epoch": 0.09298530394701435, + "grad_norm": 0.1455078125, + "learning_rate": 0.001953989454306593, + "loss": 0.1758, + "step": 10712 + }, + { + "epoch": 0.09299398442721851, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019539800594526075, + "loss": 0.1123, + "step": 10713 + }, + { + "epoch": 0.09300266490742268, + "grad_norm": 0.2470703125, + "learning_rate": 0.001953970663664722, + "loss": 0.1377, + "step": 10714 + }, + { + "epoch": 0.09301134538762684, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019539612669429466, + "loss": 0.1758, + "step": 10715 + }, + { + "epoch": 0.09302002586783101, + "grad_norm": 0.37109375, + "learning_rate": 0.001953951869287292, + "loss": 0.1172, + "step": 10716 + }, + { + "epoch": 0.09302870634803517, + "grad_norm": 0.189453125, + "learning_rate": 0.0019539424706977687, + "loss": 0.1206, + "step": 10717 + }, + { + "epoch": 0.09303738682823934, + "grad_norm": 0.28125, + "learning_rate": 0.0019539330711743873, + "loss": 0.1118, + "step": 10718 + }, + { + "epoch": 0.0930460673084435, + "grad_norm": 0.1396484375, + "learning_rate": 0.001953923670717157, + "loss": 0.127, + "step": 10719 + }, + { + "epoch": 0.09305474778864767, + "grad_norm": 0.2255859375, + "learning_rate": 0.001953914269326088, + "loss": 0.1172, + "step": 10720 + }, + { + "epoch": 0.09306342826885183, + "grad_norm": 0.4140625, + "learning_rate": 0.0019539048670011924, + "loss": 0.1211, + "step": 10721 + }, + { + "epoch": 0.093072108749056, + "grad_norm": 0.3359375, + "learning_rate": 0.001953895463742478, + "loss": 0.1768, + "step": 10722 + }, + { + "epoch": 0.09308078922926016, + "grad_norm": 0.1953125, + "learning_rate": 0.0019538860595499576, + "loss": 0.125, + "step": 10723 + }, + { + "epoch": 0.09308946970946433, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019538766544236396, + "loss": 0.1602, + "step": 10724 + }, + { + "epoch": 0.0930981501896685, + "grad_norm": 0.443359375, + "learning_rate": 0.001953867248363535, + "loss": 0.0693, + "step": 10725 + }, + { + "epoch": 0.09310683066987266, + "grad_norm": 0.07666015625, + "learning_rate": 0.001953857841369654, + "loss": 0.1172, + "step": 10726 + }, + { + "epoch": 0.09311551115007682, + "grad_norm": 0.33203125, + "learning_rate": 0.0019538484334420065, + "loss": 0.103, + "step": 10727 + }, + { + "epoch": 0.09312419163028099, + "grad_norm": 0.10498046875, + "learning_rate": 0.001953839024580604, + "loss": 0.1348, + "step": 10728 + }, + { + "epoch": 0.09313287211048515, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019538296147854554, + "loss": 0.1426, + "step": 10729 + }, + { + "epoch": 0.09314155259068932, + "grad_norm": 0.388671875, + "learning_rate": 0.0019538202040565718, + "loss": 0.0996, + "step": 10730 + }, + { + "epoch": 0.09315023307089348, + "grad_norm": 2.015625, + "learning_rate": 0.0019538107923939628, + "loss": 0.1338, + "step": 10731 + }, + { + "epoch": 0.09315891355109765, + "grad_norm": 0.37890625, + "learning_rate": 0.0019538013797976393, + "loss": 0.0879, + "step": 10732 + }, + { + "epoch": 0.09316759403130181, + "grad_norm": 0.2265625, + "learning_rate": 0.001953791966267612, + "loss": 0.1172, + "step": 10733 + }, + { + "epoch": 0.09317627451150598, + "grad_norm": 0.251953125, + "learning_rate": 0.00195378255180389, + "loss": 0.1069, + "step": 10734 + }, + { + "epoch": 0.09318495499171014, + "grad_norm": 0.388671875, + "learning_rate": 0.0019537731364064844, + "loss": 0.1187, + "step": 10735 + }, + { + "epoch": 0.09319363547191431, + "grad_norm": 0.330078125, + "learning_rate": 0.0019537637200754054, + "loss": 0.0947, + "step": 10736 + }, + { + "epoch": 0.09320231595211848, + "grad_norm": 0.275390625, + "learning_rate": 0.0019537543028106635, + "loss": 0.1289, + "step": 10737 + }, + { + "epoch": 0.09321099643232264, + "grad_norm": 0.2578125, + "learning_rate": 0.0019537448846122683, + "loss": 0.1309, + "step": 10738 + }, + { + "epoch": 0.0932196769125268, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019537354654802306, + "loss": 0.0928, + "step": 10739 + }, + { + "epoch": 0.09322835739273097, + "grad_norm": 0.7265625, + "learning_rate": 0.001953726045414561, + "loss": 0.1021, + "step": 10740 + }, + { + "epoch": 0.09323703787293514, + "grad_norm": 0.25, + "learning_rate": 0.001953716624415269, + "loss": 0.1465, + "step": 10741 + }, + { + "epoch": 0.0932457183531393, + "grad_norm": 0.302734375, + "learning_rate": 0.0019537072024823653, + "loss": 0.1128, + "step": 10742 + }, + { + "epoch": 0.09325439883334347, + "grad_norm": 0.5234375, + "learning_rate": 0.0019536977796158605, + "loss": 0.1543, + "step": 10743 + }, + { + "epoch": 0.09326307931354763, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019536883558157647, + "loss": 0.1113, + "step": 10744 + }, + { + "epoch": 0.0932717597937518, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019536789310820877, + "loss": 0.1699, + "step": 10745 + }, + { + "epoch": 0.09328044027395596, + "grad_norm": 0.369140625, + "learning_rate": 0.0019536695054148405, + "loss": 0.1099, + "step": 10746 + }, + { + "epoch": 0.09328912075416013, + "grad_norm": 0.3046875, + "learning_rate": 0.0019536600788140333, + "loss": 0.1162, + "step": 10747 + }, + { + "epoch": 0.09329780123436429, + "grad_norm": 0.1962890625, + "learning_rate": 0.001953650651279676, + "loss": 0.1299, + "step": 10748 + }, + { + "epoch": 0.09330648171456846, + "grad_norm": 0.29296875, + "learning_rate": 0.0019536412228117793, + "loss": 0.1357, + "step": 10749 + }, + { + "epoch": 0.09331516219477262, + "grad_norm": 0.162109375, + "learning_rate": 0.0019536317934103534, + "loss": 0.1069, + "step": 10750 + }, + { + "epoch": 0.09332384267497677, + "grad_norm": 0.390625, + "learning_rate": 0.0019536223630754086, + "loss": 0.1289, + "step": 10751 + }, + { + "epoch": 0.09333252315518094, + "grad_norm": 0.68359375, + "learning_rate": 0.001953612931806955, + "loss": 0.1504, + "step": 10752 + }, + { + "epoch": 0.0933412036353851, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019536034996050036, + "loss": 0.1699, + "step": 10753 + }, + { + "epoch": 0.09334988411558927, + "grad_norm": 0.404296875, + "learning_rate": 0.0019535940664695637, + "loss": 0.1855, + "step": 10754 + }, + { + "epoch": 0.09335856459579343, + "grad_norm": 0.404296875, + "learning_rate": 0.001953584632400646, + "loss": 0.2051, + "step": 10755 + }, + { + "epoch": 0.0933672450759976, + "grad_norm": 0.96484375, + "learning_rate": 0.0019535751973982617, + "loss": 0.1377, + "step": 10756 + }, + { + "epoch": 0.09337592555620176, + "grad_norm": 0.71484375, + "learning_rate": 0.0019535657614624205, + "loss": 0.1484, + "step": 10757 + }, + { + "epoch": 0.09338460603640593, + "grad_norm": 0.2421875, + "learning_rate": 0.001953556324593132, + "loss": 0.1357, + "step": 10758 + }, + { + "epoch": 0.0933932865166101, + "grad_norm": 0.15234375, + "learning_rate": 0.001953546886790407, + "loss": 0.105, + "step": 10759 + }, + { + "epoch": 0.09340196699681426, + "grad_norm": 0.287109375, + "learning_rate": 0.0019535374480542564, + "loss": 0.1416, + "step": 10760 + }, + { + "epoch": 0.09341064747701842, + "grad_norm": 0.068359375, + "learning_rate": 0.0019535280083846896, + "loss": 0.124, + "step": 10761 + }, + { + "epoch": 0.09341932795722259, + "grad_norm": 0.953125, + "learning_rate": 0.0019535185677817184, + "loss": 0.1641, + "step": 10762 + }, + { + "epoch": 0.09342800843742675, + "grad_norm": 0.220703125, + "learning_rate": 0.001953509126245351, + "loss": 0.1602, + "step": 10763 + }, + { + "epoch": 0.09343668891763092, + "grad_norm": 0.88671875, + "learning_rate": 0.0019534996837755995, + "loss": 0.1416, + "step": 10764 + }, + { + "epoch": 0.09344536939783508, + "grad_norm": 0.3828125, + "learning_rate": 0.0019534902403724732, + "loss": 0.1367, + "step": 10765 + }, + { + "epoch": 0.09345404987803925, + "grad_norm": 0.52734375, + "learning_rate": 0.001953480796035983, + "loss": 0.1016, + "step": 10766 + }, + { + "epoch": 0.09346273035824341, + "grad_norm": 0.423828125, + "learning_rate": 0.001953471350766139, + "loss": 0.1201, + "step": 10767 + }, + { + "epoch": 0.09347141083844758, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019534619045629515, + "loss": 0.126, + "step": 10768 + }, + { + "epoch": 0.09348009131865175, + "grad_norm": 0.2890625, + "learning_rate": 0.0019534524574264306, + "loss": 0.1377, + "step": 10769 + }, + { + "epoch": 0.09348877179885591, + "grad_norm": 0.62109375, + "learning_rate": 0.0019534430093565876, + "loss": 0.084, + "step": 10770 + }, + { + "epoch": 0.09349745227906008, + "grad_norm": 0.48046875, + "learning_rate": 0.0019534335603534317, + "loss": 0.1465, + "step": 10771 + }, + { + "epoch": 0.09350613275926424, + "grad_norm": 0.66015625, + "learning_rate": 0.001953424110416974, + "loss": 0.1182, + "step": 10772 + }, + { + "epoch": 0.0935148132394684, + "grad_norm": 0.15625, + "learning_rate": 0.001953414659547224, + "loss": 0.1885, + "step": 10773 + }, + { + "epoch": 0.09352349371967257, + "grad_norm": 0.44140625, + "learning_rate": 0.001953405207744193, + "loss": 0.127, + "step": 10774 + }, + { + "epoch": 0.09353217419987674, + "grad_norm": 0.306640625, + "learning_rate": 0.00195339575500789, + "loss": 0.1465, + "step": 10775 + }, + { + "epoch": 0.0935408546800809, + "grad_norm": 0.43359375, + "learning_rate": 0.0019533863013383272, + "loss": 0.1279, + "step": 10776 + }, + { + "epoch": 0.09354953516028507, + "grad_norm": 0.322265625, + "learning_rate": 0.001953376846735514, + "loss": 0.1113, + "step": 10777 + }, + { + "epoch": 0.09355821564048923, + "grad_norm": 0.478515625, + "learning_rate": 0.00195336739119946, + "loss": 0.105, + "step": 10778 + }, + { + "epoch": 0.0935668961206934, + "grad_norm": 0.28515625, + "learning_rate": 0.0019533579347301767, + "loss": 0.0898, + "step": 10779 + }, + { + "epoch": 0.09357557660089756, + "grad_norm": 0.38671875, + "learning_rate": 0.001953348477327674, + "loss": 0.1543, + "step": 10780 + }, + { + "epoch": 0.09358425708110173, + "grad_norm": 0.1650390625, + "learning_rate": 0.001953339018991962, + "loss": 0.125, + "step": 10781 + }, + { + "epoch": 0.09359293756130589, + "grad_norm": 0.39453125, + "learning_rate": 0.0019533295597230515, + "loss": 0.1436, + "step": 10782 + }, + { + "epoch": 0.09360161804151006, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019533200995209524, + "loss": 0.1113, + "step": 10783 + }, + { + "epoch": 0.09361029852171422, + "grad_norm": 0.3125, + "learning_rate": 0.0019533106383856757, + "loss": 0.1084, + "step": 10784 + }, + { + "epoch": 0.09361897900191839, + "grad_norm": 0.1416015625, + "learning_rate": 0.001953301176317231, + "loss": 0.1611, + "step": 10785 + }, + { + "epoch": 0.09362765948212255, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019532917133156286, + "loss": 0.1074, + "step": 10786 + }, + { + "epoch": 0.09363633996232672, + "grad_norm": 0.44140625, + "learning_rate": 0.00195328224938088, + "loss": 0.1162, + "step": 10787 + }, + { + "epoch": 0.09364502044253088, + "grad_norm": 0.6328125, + "learning_rate": 0.0019532727845129943, + "loss": 0.1211, + "step": 10788 + }, + { + "epoch": 0.09365370092273505, + "grad_norm": 0.2294921875, + "learning_rate": 0.001953263318711982, + "loss": 0.1064, + "step": 10789 + }, + { + "epoch": 0.09366238140293921, + "grad_norm": 0.267578125, + "learning_rate": 0.001953253851977854, + "loss": 0.1416, + "step": 10790 + }, + { + "epoch": 0.09367106188314338, + "grad_norm": 0.416015625, + "learning_rate": 0.0019532443843106205, + "loss": 0.1367, + "step": 10791 + }, + { + "epoch": 0.09367974236334754, + "grad_norm": 0.142578125, + "learning_rate": 0.0019532349157102918, + "loss": 0.1377, + "step": 10792 + }, + { + "epoch": 0.09368842284355171, + "grad_norm": 0.4921875, + "learning_rate": 0.0019532254461768783, + "loss": 0.1328, + "step": 10793 + }, + { + "epoch": 0.09369710332375587, + "grad_norm": 0.82421875, + "learning_rate": 0.0019532159757103904, + "loss": 0.1279, + "step": 10794 + }, + { + "epoch": 0.09370578380396004, + "grad_norm": 0.08203125, + "learning_rate": 0.0019532065043108377, + "loss": 0.0898, + "step": 10795 + }, + { + "epoch": 0.0937144642841642, + "grad_norm": 0.408203125, + "learning_rate": 0.0019531970319782315, + "loss": 0.1465, + "step": 10796 + }, + { + "epoch": 0.09372314476436837, + "grad_norm": 0.203125, + "learning_rate": 0.0019531875587125817, + "loss": 0.1221, + "step": 10797 + }, + { + "epoch": 0.09373182524457253, + "grad_norm": 0.1787109375, + "learning_rate": 0.001953178084513899, + "loss": 0.1299, + "step": 10798 + }, + { + "epoch": 0.0937405057247767, + "grad_norm": 0.212890625, + "learning_rate": 0.0019531686093821935, + "loss": 0.1377, + "step": 10799 + }, + { + "epoch": 0.09374918620498086, + "grad_norm": 0.81640625, + "learning_rate": 0.001953159133317476, + "loss": 0.1172, + "step": 10800 + }, + { + "epoch": 0.09375786668518503, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019531496563197563, + "loss": 0.1201, + "step": 10801 + }, + { + "epoch": 0.0937665471653892, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019531401783890448, + "loss": 0.1016, + "step": 10802 + }, + { + "epoch": 0.09377522764559336, + "grad_norm": 0.21875, + "learning_rate": 0.001953130699525352, + "loss": 0.1182, + "step": 10803 + }, + { + "epoch": 0.09378390812579752, + "grad_norm": 0.11083984375, + "learning_rate": 0.001953121219728688, + "loss": 0.1475, + "step": 10804 + }, + { + "epoch": 0.09379258860600169, + "grad_norm": 0.77734375, + "learning_rate": 0.001953111738999064, + "loss": 0.1416, + "step": 10805 + }, + { + "epoch": 0.09380126908620585, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019531022573364893, + "loss": 0.1172, + "step": 10806 + }, + { + "epoch": 0.09380994956641002, + "grad_norm": 0.291015625, + "learning_rate": 0.0019530927747409754, + "loss": 0.1084, + "step": 10807 + }, + { + "epoch": 0.09381863004661418, + "grad_norm": 0.095703125, + "learning_rate": 0.0019530832912125317, + "loss": 0.1055, + "step": 10808 + }, + { + "epoch": 0.09382731052681835, + "grad_norm": 0.283203125, + "learning_rate": 0.0019530738067511692, + "loss": 0.1016, + "step": 10809 + }, + { + "epoch": 0.09383599100702252, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019530643213568975, + "loss": 0.1172, + "step": 10810 + }, + { + "epoch": 0.09384467148722668, + "grad_norm": 0.306640625, + "learning_rate": 0.0019530548350297277, + "loss": 0.1094, + "step": 10811 + }, + { + "epoch": 0.09385335196743083, + "grad_norm": 0.126953125, + "learning_rate": 0.00195304534776967, + "loss": 0.1758, + "step": 10812 + }, + { + "epoch": 0.093862032447635, + "grad_norm": 0.40625, + "learning_rate": 0.0019530358595767348, + "loss": 0.1367, + "step": 10813 + }, + { + "epoch": 0.09387071292783916, + "grad_norm": 0.087890625, + "learning_rate": 0.0019530263704509324, + "loss": 0.0801, + "step": 10814 + }, + { + "epoch": 0.09387939340804333, + "grad_norm": 0.30859375, + "learning_rate": 0.0019530168803922729, + "loss": 0.1328, + "step": 10815 + }, + { + "epoch": 0.09388807388824749, + "grad_norm": 0.474609375, + "learning_rate": 0.0019530073894007672, + "loss": 0.0957, + "step": 10816 + }, + { + "epoch": 0.09389675436845166, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019529978974764256, + "loss": 0.124, + "step": 10817 + }, + { + "epoch": 0.09390543484865582, + "grad_norm": 0.2490234375, + "learning_rate": 0.001952988404619258, + "loss": 0.0869, + "step": 10818 + }, + { + "epoch": 0.09391411532885999, + "grad_norm": 0.455078125, + "learning_rate": 0.0019529789108292752, + "loss": 0.1465, + "step": 10819 + }, + { + "epoch": 0.09392279580906415, + "grad_norm": 0.474609375, + "learning_rate": 0.0019529694161064875, + "loss": 0.1074, + "step": 10820 + }, + { + "epoch": 0.09393147628926832, + "grad_norm": 0.404296875, + "learning_rate": 0.0019529599204509052, + "loss": 0.1465, + "step": 10821 + }, + { + "epoch": 0.09394015676947248, + "grad_norm": 0.29296875, + "learning_rate": 0.0019529504238625388, + "loss": 0.0938, + "step": 10822 + }, + { + "epoch": 0.09394883724967665, + "grad_norm": 0.26171875, + "learning_rate": 0.0019529409263413986, + "loss": 0.1299, + "step": 10823 + }, + { + "epoch": 0.09395751772988081, + "grad_norm": 0.255859375, + "learning_rate": 0.0019529314278874952, + "loss": 0.1201, + "step": 10824 + }, + { + "epoch": 0.09396619821008498, + "grad_norm": 0.267578125, + "learning_rate": 0.0019529219285008385, + "loss": 0.1416, + "step": 10825 + }, + { + "epoch": 0.09397487869028914, + "grad_norm": 0.48828125, + "learning_rate": 0.0019529124281814395, + "loss": 0.0938, + "step": 10826 + }, + { + "epoch": 0.09398355917049331, + "grad_norm": 0.2578125, + "learning_rate": 0.0019529029269293081, + "loss": 0.1123, + "step": 10827 + }, + { + "epoch": 0.09399223965069747, + "grad_norm": 0.5078125, + "learning_rate": 0.0019528934247444547, + "loss": 0.1318, + "step": 10828 + }, + { + "epoch": 0.09400092013090164, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019528839216268903, + "loss": 0.1406, + "step": 10829 + }, + { + "epoch": 0.0940096006111058, + "grad_norm": 0.890625, + "learning_rate": 0.0019528744175766244, + "loss": 0.1128, + "step": 10830 + }, + { + "epoch": 0.09401828109130997, + "grad_norm": 0.08642578125, + "learning_rate": 0.001952864912593668, + "loss": 0.1128, + "step": 10831 + }, + { + "epoch": 0.09402696157151413, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019528554066780318, + "loss": 0.105, + "step": 10832 + }, + { + "epoch": 0.0940356420517183, + "grad_norm": 0.5078125, + "learning_rate": 0.0019528458998297252, + "loss": 0.1406, + "step": 10833 + }, + { + "epoch": 0.09404432253192246, + "grad_norm": 0.251953125, + "learning_rate": 0.0019528363920487595, + "loss": 0.1367, + "step": 10834 + }, + { + "epoch": 0.09405300301212663, + "grad_norm": 0.267578125, + "learning_rate": 0.0019528268833351444, + "loss": 0.1436, + "step": 10835 + }, + { + "epoch": 0.0940616834923308, + "grad_norm": 0.4140625, + "learning_rate": 0.001952817373688891, + "loss": 0.1377, + "step": 10836 + }, + { + "epoch": 0.09407036397253496, + "grad_norm": 0.4296875, + "learning_rate": 0.0019528078631100088, + "loss": 0.1416, + "step": 10837 + }, + { + "epoch": 0.09407904445273912, + "grad_norm": 0.9921875, + "learning_rate": 0.0019527983515985094, + "loss": 0.1543, + "step": 10838 + }, + { + "epoch": 0.09408772493294329, + "grad_norm": 0.20703125, + "learning_rate": 0.001952788839154402, + "loss": 0.1338, + "step": 10839 + }, + { + "epoch": 0.09409640541314745, + "grad_norm": 0.255859375, + "learning_rate": 0.0019527793257776975, + "loss": 0.1201, + "step": 10840 + }, + { + "epoch": 0.09410508589335162, + "grad_norm": 2.265625, + "learning_rate": 0.0019527698114684068, + "loss": 0.1406, + "step": 10841 + }, + { + "epoch": 0.09411376637355579, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019527602962265396, + "loss": 0.1143, + "step": 10842 + }, + { + "epoch": 0.09412244685375995, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019527507800521063, + "loss": 0.1055, + "step": 10843 + }, + { + "epoch": 0.09413112733396412, + "grad_norm": 0.314453125, + "learning_rate": 0.0019527412629451177, + "loss": 0.1514, + "step": 10844 + }, + { + "epoch": 0.09413980781416828, + "grad_norm": 0.107421875, + "learning_rate": 0.0019527317449055846, + "loss": 0.1084, + "step": 10845 + }, + { + "epoch": 0.09414848829437245, + "grad_norm": 0.291015625, + "learning_rate": 0.0019527222259335165, + "loss": 0.0918, + "step": 10846 + }, + { + "epoch": 0.09415716877457661, + "grad_norm": 0.359375, + "learning_rate": 0.0019527127060289238, + "loss": 0.1045, + "step": 10847 + }, + { + "epoch": 0.09416584925478078, + "grad_norm": 0.16796875, + "learning_rate": 0.0019527031851918177, + "loss": 0.1021, + "step": 10848 + }, + { + "epoch": 0.09417452973498494, + "grad_norm": 0.16015625, + "learning_rate": 0.0019526936634222084, + "loss": 0.1904, + "step": 10849 + }, + { + "epoch": 0.0941832102151891, + "grad_norm": 0.259765625, + "learning_rate": 0.0019526841407201056, + "loss": 0.1055, + "step": 10850 + }, + { + "epoch": 0.09419189069539327, + "grad_norm": 0.287109375, + "learning_rate": 0.0019526746170855204, + "loss": 0.1143, + "step": 10851 + }, + { + "epoch": 0.09420057117559744, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019526650925184635, + "loss": 0.1216, + "step": 10852 + }, + { + "epoch": 0.0942092516558016, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019526555670189444, + "loss": 0.1079, + "step": 10853 + }, + { + "epoch": 0.09421793213600577, + "grad_norm": 0.7734375, + "learning_rate": 0.001952646040586974, + "loss": 0.1377, + "step": 10854 + }, + { + "epoch": 0.09422661261620993, + "grad_norm": 0.09033203125, + "learning_rate": 0.001952636513222563, + "loss": 0.1348, + "step": 10855 + }, + { + "epoch": 0.0942352930964141, + "grad_norm": 0.443359375, + "learning_rate": 0.0019526269849257211, + "loss": 0.1523, + "step": 10856 + }, + { + "epoch": 0.09424397357661826, + "grad_norm": 0.146484375, + "learning_rate": 0.0019526174556964595, + "loss": 0.1328, + "step": 10857 + }, + { + "epoch": 0.09425265405682243, + "grad_norm": 0.294921875, + "learning_rate": 0.0019526079255347882, + "loss": 0.1133, + "step": 10858 + }, + { + "epoch": 0.09426133453702659, + "grad_norm": 0.365234375, + "learning_rate": 0.0019525983944407178, + "loss": 0.124, + "step": 10859 + }, + { + "epoch": 0.09427001501723076, + "grad_norm": 0.427734375, + "learning_rate": 0.0019525888624142582, + "loss": 0.1514, + "step": 10860 + }, + { + "epoch": 0.09427869549743492, + "grad_norm": 0.59765625, + "learning_rate": 0.0019525793294554205, + "loss": 0.1572, + "step": 10861 + }, + { + "epoch": 0.09428737597763909, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019525697955642147, + "loss": 0.1484, + "step": 10862 + }, + { + "epoch": 0.09429605645784325, + "grad_norm": 0.224609375, + "learning_rate": 0.0019525602607406516, + "loss": 0.0815, + "step": 10863 + }, + { + "epoch": 0.09430473693804742, + "grad_norm": 0.1328125, + "learning_rate": 0.0019525507249847412, + "loss": 0.1523, + "step": 10864 + }, + { + "epoch": 0.09431341741825158, + "grad_norm": 0.251953125, + "learning_rate": 0.0019525411882964942, + "loss": 0.0947, + "step": 10865 + }, + { + "epoch": 0.09432209789845575, + "grad_norm": 0.10400390625, + "learning_rate": 0.001952531650675921, + "loss": 0.1201, + "step": 10866 + }, + { + "epoch": 0.09433077837865991, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019525221121230322, + "loss": 0.1074, + "step": 10867 + }, + { + "epoch": 0.09433945885886408, + "grad_norm": 0.76953125, + "learning_rate": 0.001952512572637838, + "loss": 0.123, + "step": 10868 + }, + { + "epoch": 0.09434813933906824, + "grad_norm": 0.3828125, + "learning_rate": 0.0019525030322203483, + "loss": 0.1045, + "step": 10869 + }, + { + "epoch": 0.09435681981927241, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019524934908705745, + "loss": 0.1445, + "step": 10870 + }, + { + "epoch": 0.09436550029947657, + "grad_norm": 0.388671875, + "learning_rate": 0.0019524839485885266, + "loss": 0.1309, + "step": 10871 + }, + { + "epoch": 0.09437418077968074, + "grad_norm": 0.5703125, + "learning_rate": 0.001952474405374215, + "loss": 0.1465, + "step": 10872 + }, + { + "epoch": 0.0943828612598849, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019524648612276503, + "loss": 0.1279, + "step": 10873 + }, + { + "epoch": 0.09439154174008905, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019524553161488429, + "loss": 0.1299, + "step": 10874 + }, + { + "epoch": 0.09440022222029322, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019524457701378028, + "loss": 0.166, + "step": 10875 + }, + { + "epoch": 0.09440890270049739, + "grad_norm": 0.23046875, + "learning_rate": 0.0019524362231945413, + "loss": 0.0928, + "step": 10876 + }, + { + "epoch": 0.09441758318070155, + "grad_norm": 0.07421875, + "learning_rate": 0.0019524266753190682, + "loss": 0.0933, + "step": 10877 + }, + { + "epoch": 0.09442626366090572, + "grad_norm": 0.373046875, + "learning_rate": 0.001952417126511394, + "loss": 0.1338, + "step": 10878 + }, + { + "epoch": 0.09443494414110988, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019524075767715292, + "loss": 0.1108, + "step": 10879 + }, + { + "epoch": 0.09444362462131405, + "grad_norm": 0.353515625, + "learning_rate": 0.0019523980260994845, + "loss": 0.1543, + "step": 10880 + }, + { + "epoch": 0.09445230510151821, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019523884744952701, + "loss": 0.1094, + "step": 10881 + }, + { + "epoch": 0.09446098558172238, + "grad_norm": 0.349609375, + "learning_rate": 0.0019523789219588962, + "loss": 0.126, + "step": 10882 + }, + { + "epoch": 0.09446966606192654, + "grad_norm": 0.345703125, + "learning_rate": 0.0019523693684903737, + "loss": 0.1191, + "step": 10883 + }, + { + "epoch": 0.0944783465421307, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019523598140897128, + "loss": 0.1582, + "step": 10884 + }, + { + "epoch": 0.09448702702233487, + "grad_norm": 0.326171875, + "learning_rate": 0.001952350258756924, + "loss": 0.1011, + "step": 10885 + }, + { + "epoch": 0.09449570750253904, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019523407024920176, + "loss": 0.0879, + "step": 10886 + }, + { + "epoch": 0.0945043879827432, + "grad_norm": 0.408203125, + "learning_rate": 0.0019523311452950048, + "loss": 0.1348, + "step": 10887 + }, + { + "epoch": 0.09451306846294737, + "grad_norm": 0.099609375, + "learning_rate": 0.001952321587165895, + "loss": 0.1279, + "step": 10888 + }, + { + "epoch": 0.09452174894315153, + "grad_norm": 0.46875, + "learning_rate": 0.001952312028104699, + "loss": 0.127, + "step": 10889 + }, + { + "epoch": 0.0945304294233557, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019523024681114276, + "loss": 0.1328, + "step": 10890 + }, + { + "epoch": 0.09453910990355986, + "grad_norm": 0.2734375, + "learning_rate": 0.001952292907186091, + "loss": 0.0942, + "step": 10891 + }, + { + "epoch": 0.09454779038376403, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019522833453286999, + "loss": 0.1279, + "step": 10892 + }, + { + "epoch": 0.09455647086396819, + "grad_norm": 0.1904296875, + "learning_rate": 0.001952273782539264, + "loss": 0.1064, + "step": 10893 + }, + { + "epoch": 0.09456515134417236, + "grad_norm": 0.126953125, + "learning_rate": 0.001952264218817795, + "loss": 0.1191, + "step": 10894 + }, + { + "epoch": 0.09457383182437652, + "grad_norm": 0.1806640625, + "learning_rate": 0.001952254654164302, + "loss": 0.1328, + "step": 10895 + }, + { + "epoch": 0.09458251230458069, + "grad_norm": 0.76171875, + "learning_rate": 0.0019522450885787964, + "loss": 0.1357, + "step": 10896 + }, + { + "epoch": 0.09459119278478485, + "grad_norm": 0.10546875, + "learning_rate": 0.0019522355220612886, + "loss": 0.1191, + "step": 10897 + }, + { + "epoch": 0.09459987326498902, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019522259546117884, + "loss": 0.1133, + "step": 10898 + }, + { + "epoch": 0.09460855374519318, + "grad_norm": 0.0947265625, + "learning_rate": 0.001952216386230307, + "loss": 0.1367, + "step": 10899 + }, + { + "epoch": 0.09461723422539735, + "grad_norm": 0.095703125, + "learning_rate": 0.0019522068169168544, + "loss": 0.0957, + "step": 10900 + }, + { + "epoch": 0.09462591470560151, + "grad_norm": 0.181640625, + "learning_rate": 0.0019521972466714414, + "loss": 0.1211, + "step": 10901 + }, + { + "epoch": 0.09463459518580568, + "grad_norm": 0.279296875, + "learning_rate": 0.0019521876754940782, + "loss": 0.0938, + "step": 10902 + }, + { + "epoch": 0.09464327566600984, + "grad_norm": 0.345703125, + "learning_rate": 0.0019521781033847753, + "loss": 0.1055, + "step": 10903 + }, + { + "epoch": 0.09465195614621401, + "grad_norm": 0.4765625, + "learning_rate": 0.0019521685303435434, + "loss": 0.1191, + "step": 10904 + }, + { + "epoch": 0.09466063662641817, + "grad_norm": 0.08984375, + "learning_rate": 0.0019521589563703924, + "loss": 0.1128, + "step": 10905 + }, + { + "epoch": 0.09466931710662234, + "grad_norm": 0.4375, + "learning_rate": 0.0019521493814653337, + "loss": 0.1299, + "step": 10906 + }, + { + "epoch": 0.0946779975868265, + "grad_norm": 0.51171875, + "learning_rate": 0.001952139805628377, + "loss": 0.1016, + "step": 10907 + }, + { + "epoch": 0.09468667806703067, + "grad_norm": 0.16796875, + "learning_rate": 0.0019521302288595332, + "loss": 0.103, + "step": 10908 + }, + { + "epoch": 0.09469535854723483, + "grad_norm": 0.0849609375, + "learning_rate": 0.001952120651158812, + "loss": 0.1104, + "step": 10909 + }, + { + "epoch": 0.094704039027439, + "grad_norm": 0.10205078125, + "learning_rate": 0.001952111072526225, + "loss": 0.1021, + "step": 10910 + }, + { + "epoch": 0.09471271950764316, + "grad_norm": 0.212890625, + "learning_rate": 0.0019521014929617822, + "loss": 0.1367, + "step": 10911 + }, + { + "epoch": 0.09472139998784733, + "grad_norm": 0.30078125, + "learning_rate": 0.0019520919124654938, + "loss": 0.1309, + "step": 10912 + }, + { + "epoch": 0.0947300804680515, + "grad_norm": 0.3515625, + "learning_rate": 0.0019520823310373704, + "loss": 0.1455, + "step": 10913 + }, + { + "epoch": 0.09473876094825566, + "grad_norm": 0.4140625, + "learning_rate": 0.0019520727486774226, + "loss": 0.1455, + "step": 10914 + }, + { + "epoch": 0.09474744142845982, + "grad_norm": 0.85546875, + "learning_rate": 0.0019520631653856609, + "loss": 0.1582, + "step": 10915 + }, + { + "epoch": 0.09475612190866399, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019520535811620957, + "loss": 0.1484, + "step": 10916 + }, + { + "epoch": 0.09476480238886816, + "grad_norm": 0.2578125, + "learning_rate": 0.0019520439960067375, + "loss": 0.1523, + "step": 10917 + }, + { + "epoch": 0.09477348286907232, + "grad_norm": 0.0927734375, + "learning_rate": 0.001952034409919597, + "loss": 0.1011, + "step": 10918 + }, + { + "epoch": 0.09478216334927649, + "grad_norm": 0.228515625, + "learning_rate": 0.0019520248229006843, + "loss": 0.1641, + "step": 10919 + }, + { + "epoch": 0.09479084382948065, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019520152349500098, + "loss": 0.1465, + "step": 10920 + }, + { + "epoch": 0.09479952430968482, + "grad_norm": 0.53125, + "learning_rate": 0.0019520056460675846, + "loss": 0.1211, + "step": 10921 + }, + { + "epoch": 0.09480820478988898, + "grad_norm": 0.15625, + "learning_rate": 0.001951996056253419, + "loss": 0.0796, + "step": 10922 + }, + { + "epoch": 0.09481688527009315, + "grad_norm": 0.4140625, + "learning_rate": 0.001951986465507523, + "loss": 0.0732, + "step": 10923 + }, + { + "epoch": 0.09482556575029731, + "grad_norm": 0.3671875, + "learning_rate": 0.0019519768738299073, + "loss": 0.1182, + "step": 10924 + }, + { + "epoch": 0.09483424623050148, + "grad_norm": 0.22265625, + "learning_rate": 0.0019519672812205824, + "loss": 0.1445, + "step": 10925 + }, + { + "epoch": 0.09484292671070564, + "grad_norm": 0.13671875, + "learning_rate": 0.001951957687679559, + "loss": 0.0977, + "step": 10926 + }, + { + "epoch": 0.0948516071909098, + "grad_norm": 0.5078125, + "learning_rate": 0.0019519480932068476, + "loss": 0.0957, + "step": 10927 + }, + { + "epoch": 0.09486028767111397, + "grad_norm": 0.345703125, + "learning_rate": 0.0019519384978024586, + "loss": 0.1113, + "step": 10928 + }, + { + "epoch": 0.09486896815131814, + "grad_norm": 0.142578125, + "learning_rate": 0.0019519289014664023, + "loss": 0.1611, + "step": 10929 + }, + { + "epoch": 0.0948776486315223, + "grad_norm": 0.376953125, + "learning_rate": 0.0019519193041986896, + "loss": 0.0908, + "step": 10930 + }, + { + "epoch": 0.09488632911172647, + "grad_norm": 0.39453125, + "learning_rate": 0.0019519097059993305, + "loss": 0.0967, + "step": 10931 + }, + { + "epoch": 0.09489500959193063, + "grad_norm": 0.80859375, + "learning_rate": 0.001951900106868336, + "loss": 0.1406, + "step": 10932 + }, + { + "epoch": 0.0949036900721348, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019518905068057163, + "loss": 0.1143, + "step": 10933 + }, + { + "epoch": 0.09491237055233896, + "grad_norm": 0.275390625, + "learning_rate": 0.0019518809058114814, + "loss": 0.1016, + "step": 10934 + }, + { + "epoch": 0.09492105103254311, + "grad_norm": 0.0673828125, + "learning_rate": 0.001951871303885643, + "loss": 0.0967, + "step": 10935 + }, + { + "epoch": 0.09492973151274728, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019518617010282108, + "loss": 0.1318, + "step": 10936 + }, + { + "epoch": 0.09493841199295144, + "grad_norm": 0.427734375, + "learning_rate": 0.0019518520972391954, + "loss": 0.1533, + "step": 10937 + }, + { + "epoch": 0.09494709247315561, + "grad_norm": 0.08203125, + "learning_rate": 0.0019518424925186071, + "loss": 0.1094, + "step": 10938 + }, + { + "epoch": 0.09495577295335977, + "grad_norm": 0.75, + "learning_rate": 0.001951832886866457, + "loss": 0.1367, + "step": 10939 + }, + { + "epoch": 0.09496445343356394, + "grad_norm": 0.279296875, + "learning_rate": 0.0019518232802827552, + "loss": 0.127, + "step": 10940 + }, + { + "epoch": 0.0949731339137681, + "grad_norm": 0.318359375, + "learning_rate": 0.001951813672767512, + "loss": 0.1426, + "step": 10941 + }, + { + "epoch": 0.09498181439397227, + "grad_norm": 0.59765625, + "learning_rate": 0.0019518040643207385, + "loss": 0.2715, + "step": 10942 + }, + { + "epoch": 0.09499049487417643, + "grad_norm": 0.416015625, + "learning_rate": 0.0019517944549424444, + "loss": 0.1318, + "step": 10943 + }, + { + "epoch": 0.0949991753543806, + "grad_norm": 0.31640625, + "learning_rate": 0.0019517848446326412, + "loss": 0.1133, + "step": 10944 + }, + { + "epoch": 0.09500785583458476, + "grad_norm": 0.37109375, + "learning_rate": 0.0019517752333913386, + "loss": 0.1172, + "step": 10945 + }, + { + "epoch": 0.09501653631478893, + "grad_norm": 0.474609375, + "learning_rate": 0.0019517656212185476, + "loss": 0.1104, + "step": 10946 + }, + { + "epoch": 0.0950252167949931, + "grad_norm": 0.107421875, + "learning_rate": 0.0019517560081142782, + "loss": 0.1289, + "step": 10947 + }, + { + "epoch": 0.09503389727519726, + "grad_norm": 0.4921875, + "learning_rate": 0.0019517463940785414, + "loss": 0.0996, + "step": 10948 + }, + { + "epoch": 0.09504257775540143, + "grad_norm": 0.53125, + "learning_rate": 0.0019517367791113476, + "loss": 0.1104, + "step": 10949 + }, + { + "epoch": 0.09505125823560559, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019517271632127073, + "loss": 0.1167, + "step": 10950 + }, + { + "epoch": 0.09505993871580976, + "grad_norm": 0.47265625, + "learning_rate": 0.0019517175463826311, + "loss": 0.1123, + "step": 10951 + }, + { + "epoch": 0.09506861919601392, + "grad_norm": 0.10107421875, + "learning_rate": 0.001951707928621129, + "loss": 0.1289, + "step": 10952 + }, + { + "epoch": 0.09507729967621809, + "grad_norm": 0.177734375, + "learning_rate": 0.001951698309928212, + "loss": 0.1279, + "step": 10953 + }, + { + "epoch": 0.09508598015642225, + "grad_norm": 0.21875, + "learning_rate": 0.0019516886903038907, + "loss": 0.1201, + "step": 10954 + }, + { + "epoch": 0.09509466063662642, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019516790697481751, + "loss": 0.1035, + "step": 10955 + }, + { + "epoch": 0.09510334111683058, + "grad_norm": 0.515625, + "learning_rate": 0.0019516694482610765, + "loss": 0.0977, + "step": 10956 + }, + { + "epoch": 0.09511202159703475, + "grad_norm": 0.69921875, + "learning_rate": 0.001951659825842605, + "loss": 0.125, + "step": 10957 + }, + { + "epoch": 0.09512070207723891, + "grad_norm": 0.240234375, + "learning_rate": 0.0019516502024927707, + "loss": 0.1143, + "step": 10958 + }, + { + "epoch": 0.09512938255744308, + "grad_norm": 0.314453125, + "learning_rate": 0.0019516405782115847, + "loss": 0.1279, + "step": 10959 + }, + { + "epoch": 0.09513806303764724, + "grad_norm": 0.484375, + "learning_rate": 0.0019516309529990575, + "loss": 0.127, + "step": 10960 + }, + { + "epoch": 0.0951467435178514, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019516213268551995, + "loss": 0.1602, + "step": 10961 + }, + { + "epoch": 0.09515542399805557, + "grad_norm": 0.263671875, + "learning_rate": 0.001951611699780021, + "loss": 0.1309, + "step": 10962 + }, + { + "epoch": 0.09516410447825974, + "grad_norm": 0.81640625, + "learning_rate": 0.0019516020717735328, + "loss": 0.166, + "step": 10963 + }, + { + "epoch": 0.0951727849584639, + "grad_norm": 0.396484375, + "learning_rate": 0.0019515924428357455, + "loss": 0.127, + "step": 10964 + }, + { + "epoch": 0.09518146543866807, + "grad_norm": 0.25390625, + "learning_rate": 0.0019515828129666693, + "loss": 0.1748, + "step": 10965 + }, + { + "epoch": 0.09519014591887223, + "grad_norm": 0.126953125, + "learning_rate": 0.0019515731821663149, + "loss": 0.1758, + "step": 10966 + }, + { + "epoch": 0.0951988263990764, + "grad_norm": 0.2578125, + "learning_rate": 0.0019515635504346932, + "loss": 0.1602, + "step": 10967 + }, + { + "epoch": 0.09520750687928056, + "grad_norm": 0.130859375, + "learning_rate": 0.0019515539177718142, + "loss": 0.1514, + "step": 10968 + }, + { + "epoch": 0.09521618735948473, + "grad_norm": 0.3984375, + "learning_rate": 0.0019515442841776885, + "loss": 0.1035, + "step": 10969 + }, + { + "epoch": 0.09522486783968889, + "grad_norm": 0.12890625, + "learning_rate": 0.001951534649652327, + "loss": 0.1572, + "step": 10970 + }, + { + "epoch": 0.09523354831989306, + "grad_norm": 0.37109375, + "learning_rate": 0.0019515250141957398, + "loss": 0.0952, + "step": 10971 + }, + { + "epoch": 0.09524222880009722, + "grad_norm": 0.35546875, + "learning_rate": 0.001951515377807938, + "loss": 0.167, + "step": 10972 + }, + { + "epoch": 0.09525090928030139, + "grad_norm": 0.37890625, + "learning_rate": 0.0019515057404889313, + "loss": 0.1367, + "step": 10973 + }, + { + "epoch": 0.09525958976050555, + "grad_norm": 0.7578125, + "learning_rate": 0.0019514961022387312, + "loss": 0.1445, + "step": 10974 + }, + { + "epoch": 0.09526827024070972, + "grad_norm": 0.78515625, + "learning_rate": 0.0019514864630573471, + "loss": 0.1045, + "step": 10975 + }, + { + "epoch": 0.09527695072091388, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019514768229447908, + "loss": 0.104, + "step": 10976 + }, + { + "epoch": 0.09528563120111805, + "grad_norm": 0.1962890625, + "learning_rate": 0.001951467181901072, + "loss": 0.125, + "step": 10977 + }, + { + "epoch": 0.09529431168132221, + "grad_norm": 0.474609375, + "learning_rate": 0.0019514575399262016, + "loss": 0.1045, + "step": 10978 + }, + { + "epoch": 0.09530299216152638, + "grad_norm": 0.33203125, + "learning_rate": 0.00195144789702019, + "loss": 0.1182, + "step": 10979 + }, + { + "epoch": 0.09531167264173054, + "grad_norm": 0.34375, + "learning_rate": 0.0019514382531830474, + "loss": 0.1118, + "step": 10980 + }, + { + "epoch": 0.09532035312193471, + "grad_norm": 0.58984375, + "learning_rate": 0.0019514286084147854, + "loss": 0.1025, + "step": 10981 + }, + { + "epoch": 0.09532903360213887, + "grad_norm": 0.51953125, + "learning_rate": 0.0019514189627154134, + "loss": 0.1426, + "step": 10982 + }, + { + "epoch": 0.09533771408234304, + "grad_norm": 0.83984375, + "learning_rate": 0.0019514093160849424, + "loss": 0.1064, + "step": 10983 + }, + { + "epoch": 0.0953463945625472, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019513996685233833, + "loss": 0.1396, + "step": 10984 + }, + { + "epoch": 0.09535507504275137, + "grad_norm": 0.171875, + "learning_rate": 0.001951390020030746, + "loss": 0.1172, + "step": 10985 + }, + { + "epoch": 0.09536375552295553, + "grad_norm": 0.85546875, + "learning_rate": 0.0019513803706070415, + "loss": 0.1367, + "step": 10986 + }, + { + "epoch": 0.0953724360031597, + "grad_norm": 0.1015625, + "learning_rate": 0.00195137072025228, + "loss": 0.1211, + "step": 10987 + }, + { + "epoch": 0.09538111648336386, + "grad_norm": 0.30859375, + "learning_rate": 0.0019513610689664725, + "loss": 0.0947, + "step": 10988 + }, + { + "epoch": 0.09538979696356803, + "grad_norm": 1.171875, + "learning_rate": 0.0019513514167496294, + "loss": 0.1079, + "step": 10989 + }, + { + "epoch": 0.0953984774437722, + "grad_norm": 0.451171875, + "learning_rate": 0.0019513417636017607, + "loss": 0.1045, + "step": 10990 + }, + { + "epoch": 0.09540715792397636, + "grad_norm": 0.326171875, + "learning_rate": 0.0019513321095228782, + "loss": 0.1455, + "step": 10991 + }, + { + "epoch": 0.09541583840418053, + "grad_norm": 0.357421875, + "learning_rate": 0.001951322454512991, + "loss": 0.1025, + "step": 10992 + }, + { + "epoch": 0.09542451888438469, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019513127985721108, + "loss": 0.1465, + "step": 10993 + }, + { + "epoch": 0.09543319936458886, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019513031417002476, + "loss": 0.124, + "step": 10994 + }, + { + "epoch": 0.09544187984479302, + "grad_norm": 0.061767578125, + "learning_rate": 0.0019512934838974122, + "loss": 0.0874, + "step": 10995 + }, + { + "epoch": 0.09545056032499719, + "grad_norm": 0.14453125, + "learning_rate": 0.0019512838251636146, + "loss": 0.1084, + "step": 10996 + }, + { + "epoch": 0.09545924080520134, + "grad_norm": 0.173828125, + "learning_rate": 0.001951274165498866, + "loss": 0.1206, + "step": 10997 + }, + { + "epoch": 0.0954679212854055, + "grad_norm": 1.0078125, + "learning_rate": 0.001951264504903177, + "loss": 0.1182, + "step": 10998 + }, + { + "epoch": 0.09547660176560967, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019512548433765575, + "loss": 0.1201, + "step": 10999 + }, + { + "epoch": 0.09548528224581383, + "grad_norm": 0.0927734375, + "learning_rate": 0.001951245180919019, + "loss": 0.123, + "step": 11000 + }, + { + "epoch": 0.095493962726018, + "grad_norm": 0.703125, + "learning_rate": 0.001951235517530571, + "loss": 0.127, + "step": 11001 + }, + { + "epoch": 0.09550264320622216, + "grad_norm": 0.1005859375, + "learning_rate": 0.001951225853211225, + "loss": 0.1191, + "step": 11002 + }, + { + "epoch": 0.09551132368642633, + "grad_norm": 0.412109375, + "learning_rate": 0.0019512161879609914, + "loss": 0.1387, + "step": 11003 + }, + { + "epoch": 0.09552000416663049, + "grad_norm": 0.74609375, + "learning_rate": 0.0019512065217798802, + "loss": 0.126, + "step": 11004 + }, + { + "epoch": 0.09552868464683466, + "grad_norm": 0.310546875, + "learning_rate": 0.0019511968546679025, + "loss": 0.1035, + "step": 11005 + }, + { + "epoch": 0.09553736512703882, + "grad_norm": 0.56640625, + "learning_rate": 0.0019511871866250685, + "loss": 0.1338, + "step": 11006 + }, + { + "epoch": 0.09554604560724299, + "grad_norm": 0.119140625, + "learning_rate": 0.0019511775176513893, + "loss": 0.1113, + "step": 11007 + }, + { + "epoch": 0.09555472608744715, + "grad_norm": 0.1533203125, + "learning_rate": 0.001951167847746875, + "loss": 0.1147, + "step": 11008 + }, + { + "epoch": 0.09556340656765132, + "grad_norm": 0.18359375, + "learning_rate": 0.0019511581769115363, + "loss": 0.1113, + "step": 11009 + }, + { + "epoch": 0.09557208704785548, + "grad_norm": 0.490234375, + "learning_rate": 0.0019511485051453836, + "loss": 0.1094, + "step": 11010 + }, + { + "epoch": 0.09558076752805965, + "grad_norm": 0.337890625, + "learning_rate": 0.001951138832448428, + "loss": 0.1138, + "step": 11011 + }, + { + "epoch": 0.09558944800826381, + "grad_norm": 0.2353515625, + "learning_rate": 0.00195112915882068, + "loss": 0.1055, + "step": 11012 + }, + { + "epoch": 0.09559812848846798, + "grad_norm": 0.6875, + "learning_rate": 0.0019511194842621493, + "loss": 0.1562, + "step": 11013 + }, + { + "epoch": 0.09560680896867214, + "grad_norm": 0.2890625, + "learning_rate": 0.0019511098087728474, + "loss": 0.0928, + "step": 11014 + }, + { + "epoch": 0.09561548944887631, + "grad_norm": 0.125, + "learning_rate": 0.0019511001323527849, + "loss": 0.0977, + "step": 11015 + }, + { + "epoch": 0.09562416992908047, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019510904550019716, + "loss": 0.1602, + "step": 11016 + }, + { + "epoch": 0.09563285040928464, + "grad_norm": 0.640625, + "learning_rate": 0.0019510807767204189, + "loss": 0.1289, + "step": 11017 + }, + { + "epoch": 0.0956415308894888, + "grad_norm": 0.111328125, + "learning_rate": 0.0019510710975081367, + "loss": 0.0967, + "step": 11018 + }, + { + "epoch": 0.09565021136969297, + "grad_norm": 0.359375, + "learning_rate": 0.0019510614173651364, + "loss": 0.1279, + "step": 11019 + }, + { + "epoch": 0.09565889184989713, + "grad_norm": 1.0, + "learning_rate": 0.0019510517362914277, + "loss": 0.1035, + "step": 11020 + }, + { + "epoch": 0.0956675723301013, + "grad_norm": 0.25390625, + "learning_rate": 0.001951042054287022, + "loss": 0.0928, + "step": 11021 + }, + { + "epoch": 0.09567625281030546, + "grad_norm": 0.1328125, + "learning_rate": 0.0019510323713519291, + "loss": 0.1406, + "step": 11022 + }, + { + "epoch": 0.09568493329050963, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019510226874861604, + "loss": 0.1484, + "step": 11023 + }, + { + "epoch": 0.0956936137707138, + "grad_norm": 0.203125, + "learning_rate": 0.0019510130026897258, + "loss": 0.125, + "step": 11024 + }, + { + "epoch": 0.09570229425091796, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019510033169626362, + "loss": 0.1211, + "step": 11025 + }, + { + "epoch": 0.09571097473112213, + "grad_norm": 0.30078125, + "learning_rate": 0.0019509936303049023, + "loss": 0.1191, + "step": 11026 + }, + { + "epoch": 0.09571965521132629, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019509839427165343, + "loss": 0.1562, + "step": 11027 + }, + { + "epoch": 0.09572833569153046, + "grad_norm": 0.169921875, + "learning_rate": 0.0019509742541975433, + "loss": 0.1328, + "step": 11028 + }, + { + "epoch": 0.09573701617173462, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019509645647479393, + "loss": 0.1221, + "step": 11029 + }, + { + "epoch": 0.09574569665193879, + "grad_norm": 0.240234375, + "learning_rate": 0.0019509548743677334, + "loss": 0.1162, + "step": 11030 + }, + { + "epoch": 0.09575437713214295, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019509451830569363, + "loss": 0.1504, + "step": 11031 + }, + { + "epoch": 0.09576305761234712, + "grad_norm": 0.080078125, + "learning_rate": 0.0019509354908155581, + "loss": 0.1084, + "step": 11032 + }, + { + "epoch": 0.09577173809255128, + "grad_norm": 0.326171875, + "learning_rate": 0.0019509257976436094, + "loss": 0.0962, + "step": 11033 + }, + { + "epoch": 0.09578041857275545, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019509161035411017, + "loss": 0.1094, + "step": 11034 + }, + { + "epoch": 0.09578909905295961, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019509064085080443, + "loss": 0.0938, + "step": 11035 + }, + { + "epoch": 0.09579777953316378, + "grad_norm": 0.181640625, + "learning_rate": 0.0019508967125444487, + "loss": 0.1543, + "step": 11036 + }, + { + "epoch": 0.09580646001336794, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019508870156503251, + "loss": 0.1172, + "step": 11037 + }, + { + "epoch": 0.0958151404935721, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019508773178256841, + "loss": 0.0894, + "step": 11038 + }, + { + "epoch": 0.09582382097377627, + "grad_norm": 0.1533203125, + "learning_rate": 0.001950867619070537, + "loss": 0.1025, + "step": 11039 + }, + { + "epoch": 0.09583250145398044, + "grad_norm": 0.099609375, + "learning_rate": 0.0019508579193848934, + "loss": 0.1367, + "step": 11040 + }, + { + "epoch": 0.0958411819341846, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019508482187687644, + "loss": 0.1357, + "step": 11041 + }, + { + "epoch": 0.09584986241438877, + "grad_norm": 0.482421875, + "learning_rate": 0.0019508385172221608, + "loss": 0.1196, + "step": 11042 + }, + { + "epoch": 0.09585854289459293, + "grad_norm": 0.453125, + "learning_rate": 0.001950828814745093, + "loss": 0.0962, + "step": 11043 + }, + { + "epoch": 0.0958672233747971, + "grad_norm": 0.330078125, + "learning_rate": 0.0019508191113375714, + "loss": 0.1025, + "step": 11044 + }, + { + "epoch": 0.09587590385500126, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019508094069996067, + "loss": 0.1191, + "step": 11045 + }, + { + "epoch": 0.09588458433520543, + "grad_norm": 0.232421875, + "learning_rate": 0.0019507997017312098, + "loss": 0.1191, + "step": 11046 + }, + { + "epoch": 0.09589326481540959, + "grad_norm": 0.369140625, + "learning_rate": 0.0019507899955323912, + "loss": 0.1016, + "step": 11047 + }, + { + "epoch": 0.09590194529561376, + "grad_norm": 0.24609375, + "learning_rate": 0.0019507802884031616, + "loss": 0.1289, + "step": 11048 + }, + { + "epoch": 0.09591062577581792, + "grad_norm": 0.1875, + "learning_rate": 0.001950770580343531, + "loss": 0.1177, + "step": 11049 + }, + { + "epoch": 0.09591930625602209, + "grad_norm": 0.384765625, + "learning_rate": 0.0019507608713535106, + "loss": 0.1436, + "step": 11050 + }, + { + "epoch": 0.09592798673622625, + "grad_norm": 0.18359375, + "learning_rate": 0.001950751161433111, + "loss": 0.0967, + "step": 11051 + }, + { + "epoch": 0.09593666721643042, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019507414505823427, + "loss": 0.1016, + "step": 11052 + }, + { + "epoch": 0.09594534769663458, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019507317388012164, + "loss": 0.124, + "step": 11053 + }, + { + "epoch": 0.09595402817683875, + "grad_norm": 0.158203125, + "learning_rate": 0.0019507220260897426, + "loss": 0.1504, + "step": 11054 + }, + { + "epoch": 0.09596270865704291, + "grad_norm": 0.1552734375, + "learning_rate": 0.001950712312447932, + "loss": 0.0967, + "step": 11055 + }, + { + "epoch": 0.09597138913724708, + "grad_norm": 0.3671875, + "learning_rate": 0.0019507025978757948, + "loss": 0.1641, + "step": 11056 + }, + { + "epoch": 0.09598006961745124, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019506928823733428, + "loss": 0.1152, + "step": 11057 + }, + { + "epoch": 0.09598875009765541, + "grad_norm": 0.453125, + "learning_rate": 0.0019506831659405852, + "loss": 0.1162, + "step": 11058 + }, + { + "epoch": 0.09599743057785956, + "grad_norm": 0.2734375, + "learning_rate": 0.0019506734485775332, + "loss": 0.1138, + "step": 11059 + }, + { + "epoch": 0.09600611105806373, + "grad_norm": 0.470703125, + "learning_rate": 0.0019506637302841982, + "loss": 0.1348, + "step": 11060 + }, + { + "epoch": 0.09601479153826789, + "grad_norm": 0.150390625, + "learning_rate": 0.0019506540110605895, + "loss": 0.1367, + "step": 11061 + }, + { + "epoch": 0.09602347201847206, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019506442909067187, + "loss": 0.1348, + "step": 11062 + }, + { + "epoch": 0.09603215249867622, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019506345698225957, + "loss": 0.1064, + "step": 11063 + }, + { + "epoch": 0.09604083297888039, + "grad_norm": 0.59765625, + "learning_rate": 0.001950624847808232, + "loss": 0.1123, + "step": 11064 + }, + { + "epoch": 0.09604951345908455, + "grad_norm": 0.384765625, + "learning_rate": 0.0019506151248636374, + "loss": 0.123, + "step": 11065 + }, + { + "epoch": 0.09605819393928872, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019506054009888233, + "loss": 0.0957, + "step": 11066 + }, + { + "epoch": 0.09606687441949288, + "grad_norm": 0.26953125, + "learning_rate": 0.0019505956761837996, + "loss": 0.0869, + "step": 11067 + }, + { + "epoch": 0.09607555489969705, + "grad_norm": 0.5625, + "learning_rate": 0.001950585950448577, + "loss": 0.1279, + "step": 11068 + }, + { + "epoch": 0.09608423537990121, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019505762237831666, + "loss": 0.1543, + "step": 11069 + }, + { + "epoch": 0.09609291586010538, + "grad_norm": 0.306640625, + "learning_rate": 0.0019505664961875788, + "loss": 0.1445, + "step": 11070 + }, + { + "epoch": 0.09610159634030954, + "grad_norm": 0.68359375, + "learning_rate": 0.0019505567676618243, + "loss": 0.1543, + "step": 11071 + }, + { + "epoch": 0.09611027682051371, + "grad_norm": 0.28515625, + "learning_rate": 0.0019505470382059139, + "loss": 0.126, + "step": 11072 + }, + { + "epoch": 0.09611895730071787, + "grad_norm": 0.326171875, + "learning_rate": 0.0019505373078198577, + "loss": 0.1357, + "step": 11073 + }, + { + "epoch": 0.09612763778092204, + "grad_norm": 0.11328125, + "learning_rate": 0.001950527576503667, + "loss": 0.1396, + "step": 11074 + }, + { + "epoch": 0.0961363182611262, + "grad_norm": 0.16796875, + "learning_rate": 0.0019505178442573519, + "loss": 0.1289, + "step": 11075 + }, + { + "epoch": 0.09614499874133037, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019505081110809232, + "loss": 0.1006, + "step": 11076 + }, + { + "epoch": 0.09615367922153453, + "grad_norm": 0.47265625, + "learning_rate": 0.0019504983769743918, + "loss": 0.1299, + "step": 11077 + }, + { + "epoch": 0.0961623597017387, + "grad_norm": 0.10546875, + "learning_rate": 0.001950488641937768, + "loss": 0.1348, + "step": 11078 + }, + { + "epoch": 0.09617104018194286, + "grad_norm": 0.6015625, + "learning_rate": 0.0019504789059710628, + "loss": 0.1367, + "step": 11079 + }, + { + "epoch": 0.09617972066214703, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019504691690742865, + "loss": 0.1592, + "step": 11080 + }, + { + "epoch": 0.0961884011423512, + "grad_norm": 0.1796875, + "learning_rate": 0.0019504594312474497, + "loss": 0.1523, + "step": 11081 + }, + { + "epoch": 0.09619708162255536, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019504496924905637, + "loss": 0.1172, + "step": 11082 + }, + { + "epoch": 0.09620576210275952, + "grad_norm": 0.296875, + "learning_rate": 0.0019504399528036383, + "loss": 0.1357, + "step": 11083 + }, + { + "epoch": 0.09621444258296369, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019504302121866848, + "loss": 0.0928, + "step": 11084 + }, + { + "epoch": 0.09622312306316785, + "grad_norm": 0.51953125, + "learning_rate": 0.0019504204706397132, + "loss": 0.0967, + "step": 11085 + }, + { + "epoch": 0.09623180354337202, + "grad_norm": 0.134765625, + "learning_rate": 0.0019504107281627353, + "loss": 0.126, + "step": 11086 + }, + { + "epoch": 0.09624048402357618, + "grad_norm": 0.62890625, + "learning_rate": 0.0019504009847557604, + "loss": 0.1309, + "step": 11087 + }, + { + "epoch": 0.09624916450378035, + "grad_norm": 0.271484375, + "learning_rate": 0.0019503912404188, + "loss": 0.1533, + "step": 11088 + }, + { + "epoch": 0.09625784498398451, + "grad_norm": 0.30078125, + "learning_rate": 0.0019503814951518643, + "loss": 0.1318, + "step": 11089 + }, + { + "epoch": 0.09626652546418868, + "grad_norm": 0.2265625, + "learning_rate": 0.0019503717489549645, + "loss": 0.0986, + "step": 11090 + }, + { + "epoch": 0.09627520594439284, + "grad_norm": 0.3125, + "learning_rate": 0.0019503620018281108, + "loss": 0.1138, + "step": 11091 + }, + { + "epoch": 0.09628388642459701, + "grad_norm": 0.298828125, + "learning_rate": 0.0019503522537713137, + "loss": 0.1162, + "step": 11092 + }, + { + "epoch": 0.09629256690480117, + "grad_norm": 0.33984375, + "learning_rate": 0.0019503425047845847, + "loss": 0.1191, + "step": 11093 + }, + { + "epoch": 0.09630124738500534, + "grad_norm": 0.41796875, + "learning_rate": 0.0019503327548679333, + "loss": 0.0962, + "step": 11094 + }, + { + "epoch": 0.0963099278652095, + "grad_norm": 0.3828125, + "learning_rate": 0.001950323004021371, + "loss": 0.1416, + "step": 11095 + }, + { + "epoch": 0.09631860834541367, + "grad_norm": 0.1640625, + "learning_rate": 0.0019503132522449085, + "loss": 0.1143, + "step": 11096 + }, + { + "epoch": 0.09632728882561784, + "grad_norm": 0.23828125, + "learning_rate": 0.001950303499538556, + "loss": 0.1543, + "step": 11097 + }, + { + "epoch": 0.096335969305822, + "grad_norm": 0.224609375, + "learning_rate": 0.0019502937459023244, + "loss": 0.1084, + "step": 11098 + }, + { + "epoch": 0.09634464978602617, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019502839913362244, + "loss": 0.1348, + "step": 11099 + }, + { + "epoch": 0.09635333026623033, + "grad_norm": 0.314453125, + "learning_rate": 0.0019502742358402663, + "loss": 0.1436, + "step": 11100 + }, + { + "epoch": 0.0963620107464345, + "grad_norm": 0.40234375, + "learning_rate": 0.0019502644794144612, + "loss": 0.1113, + "step": 11101 + }, + { + "epoch": 0.09637069122663866, + "grad_norm": 1.1015625, + "learning_rate": 0.0019502547220588199, + "loss": 0.1187, + "step": 11102 + }, + { + "epoch": 0.09637937170684283, + "grad_norm": 0.1015625, + "learning_rate": 0.0019502449637733526, + "loss": 0.1182, + "step": 11103 + }, + { + "epoch": 0.09638805218704699, + "grad_norm": 0.083984375, + "learning_rate": 0.0019502352045580702, + "loss": 0.127, + "step": 11104 + }, + { + "epoch": 0.09639673266725116, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019502254444129833, + "loss": 0.124, + "step": 11105 + }, + { + "epoch": 0.09640541314745532, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019502156833381028, + "loss": 0.1465, + "step": 11106 + }, + { + "epoch": 0.09641409362765949, + "grad_norm": 0.130859375, + "learning_rate": 0.001950205921333439, + "loss": 0.124, + "step": 11107 + }, + { + "epoch": 0.09642277410786365, + "grad_norm": 0.353515625, + "learning_rate": 0.001950196158399003, + "loss": 0.1475, + "step": 11108 + }, + { + "epoch": 0.09643145458806782, + "grad_norm": 0.453125, + "learning_rate": 0.001950186394534805, + "loss": 0.0972, + "step": 11109 + }, + { + "epoch": 0.09644013506827198, + "grad_norm": 0.283203125, + "learning_rate": 0.0019501766297408563, + "loss": 0.1182, + "step": 11110 + }, + { + "epoch": 0.09644881554847615, + "grad_norm": 0.50390625, + "learning_rate": 0.0019501668640171668, + "loss": 0.0894, + "step": 11111 + }, + { + "epoch": 0.09645749602868031, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019501570973637478, + "loss": 0.1455, + "step": 11112 + }, + { + "epoch": 0.09646617650888448, + "grad_norm": 0.53515625, + "learning_rate": 0.00195014732978061, + "loss": 0.1016, + "step": 11113 + }, + { + "epoch": 0.09647485698908864, + "grad_norm": 0.53515625, + "learning_rate": 0.0019501375612677634, + "loss": 0.0986, + "step": 11114 + }, + { + "epoch": 0.09648353746929281, + "grad_norm": 1.7265625, + "learning_rate": 0.0019501277918252194, + "loss": 0.1592, + "step": 11115 + }, + { + "epoch": 0.09649221794949697, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019501180214529883, + "loss": 0.126, + "step": 11116 + }, + { + "epoch": 0.09650089842970114, + "grad_norm": 0.412109375, + "learning_rate": 0.0019501082501510811, + "loss": 0.1309, + "step": 11117 + }, + { + "epoch": 0.0965095789099053, + "grad_norm": 0.5703125, + "learning_rate": 0.0019500984779195082, + "loss": 0.105, + "step": 11118 + }, + { + "epoch": 0.09651825939010947, + "grad_norm": 0.20703125, + "learning_rate": 0.00195008870475828, + "loss": 0.1084, + "step": 11119 + }, + { + "epoch": 0.09652693987031362, + "grad_norm": 0.26953125, + "learning_rate": 0.0019500789306674082, + "loss": 0.1309, + "step": 11120 + }, + { + "epoch": 0.09653562035051778, + "grad_norm": 0.275390625, + "learning_rate": 0.0019500691556469023, + "loss": 0.125, + "step": 11121 + }, + { + "epoch": 0.09654430083072195, + "grad_norm": 0.1748046875, + "learning_rate": 0.001950059379696774, + "loss": 0.0957, + "step": 11122 + }, + { + "epoch": 0.09655298131092611, + "grad_norm": 0.45703125, + "learning_rate": 0.0019500496028170332, + "loss": 0.1172, + "step": 11123 + }, + { + "epoch": 0.09656166179113028, + "grad_norm": 0.46875, + "learning_rate": 0.0019500398250076913, + "loss": 0.1211, + "step": 11124 + }, + { + "epoch": 0.09657034227133444, + "grad_norm": 0.75390625, + "learning_rate": 0.0019500300462687583, + "loss": 0.1426, + "step": 11125 + }, + { + "epoch": 0.09657902275153861, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019500202666002454, + "loss": 0.1025, + "step": 11126 + }, + { + "epoch": 0.09658770323174277, + "grad_norm": 0.4140625, + "learning_rate": 0.0019500104860021628, + "loss": 0.0991, + "step": 11127 + }, + { + "epoch": 0.09659638371194694, + "grad_norm": 0.48046875, + "learning_rate": 0.0019500007044745218, + "loss": 0.1279, + "step": 11128 + }, + { + "epoch": 0.0966050641921511, + "grad_norm": 0.283203125, + "learning_rate": 0.0019499909220173327, + "loss": 0.0708, + "step": 11129 + }, + { + "epoch": 0.09661374467235527, + "grad_norm": 0.77734375, + "learning_rate": 0.0019499811386306063, + "loss": 0.1157, + "step": 11130 + }, + { + "epoch": 0.09662242515255944, + "grad_norm": 0.185546875, + "learning_rate": 0.0019499713543143532, + "loss": 0.1011, + "step": 11131 + }, + { + "epoch": 0.0966311056327636, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019499615690685844, + "loss": 0.1523, + "step": 11132 + }, + { + "epoch": 0.09663978611296777, + "grad_norm": 0.2890625, + "learning_rate": 0.0019499517828933102, + "loss": 0.125, + "step": 11133 + }, + { + "epoch": 0.09664846659317193, + "grad_norm": 0.3359375, + "learning_rate": 0.0019499419957885416, + "loss": 0.1416, + "step": 11134 + }, + { + "epoch": 0.0966571470733761, + "grad_norm": 1.2265625, + "learning_rate": 0.0019499322077542892, + "loss": 0.1562, + "step": 11135 + }, + { + "epoch": 0.09666582755358026, + "grad_norm": 0.349609375, + "learning_rate": 0.001949922418790564, + "loss": 0.1006, + "step": 11136 + }, + { + "epoch": 0.09667450803378443, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019499126288973758, + "loss": 0.0859, + "step": 11137 + }, + { + "epoch": 0.09668318851398859, + "grad_norm": 0.376953125, + "learning_rate": 0.0019499028380747364, + "loss": 0.0962, + "step": 11138 + }, + { + "epoch": 0.09669186899419276, + "grad_norm": 0.23828125, + "learning_rate": 0.0019498930463226558, + "loss": 0.1006, + "step": 11139 + }, + { + "epoch": 0.09670054947439692, + "grad_norm": 0.1083984375, + "learning_rate": 0.001949883253641145, + "loss": 0.083, + "step": 11140 + }, + { + "epoch": 0.09670922995460109, + "grad_norm": 1.6015625, + "learning_rate": 0.0019498734600302146, + "loss": 0.1035, + "step": 11141 + }, + { + "epoch": 0.09671791043480525, + "grad_norm": 0.078125, + "learning_rate": 0.0019498636654898755, + "loss": 0.1191, + "step": 11142 + }, + { + "epoch": 0.09672659091500942, + "grad_norm": 0.58203125, + "learning_rate": 0.001949853870020138, + "loss": 0.0747, + "step": 11143 + }, + { + "epoch": 0.09673527139521358, + "grad_norm": 0.482421875, + "learning_rate": 0.0019498440736210132, + "loss": 0.1074, + "step": 11144 + }, + { + "epoch": 0.09674395187541775, + "grad_norm": 0.255859375, + "learning_rate": 0.001949834276292512, + "loss": 0.1016, + "step": 11145 + }, + { + "epoch": 0.09675263235562191, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019498244780346446, + "loss": 0.0957, + "step": 11146 + }, + { + "epoch": 0.09676131283582608, + "grad_norm": 1.09375, + "learning_rate": 0.0019498146788474217, + "loss": 0.1406, + "step": 11147 + }, + { + "epoch": 0.09676999331603024, + "grad_norm": 0.5078125, + "learning_rate": 0.0019498048787308542, + "loss": 0.1611, + "step": 11148 + }, + { + "epoch": 0.09677867379623441, + "grad_norm": 0.33984375, + "learning_rate": 0.0019497950776849533, + "loss": 0.1406, + "step": 11149 + }, + { + "epoch": 0.09678735427643857, + "grad_norm": 0.447265625, + "learning_rate": 0.0019497852757097288, + "loss": 0.1328, + "step": 11150 + }, + { + "epoch": 0.09679603475664274, + "grad_norm": 0.255859375, + "learning_rate": 0.001949775472805192, + "loss": 0.126, + "step": 11151 + }, + { + "epoch": 0.0968047152368469, + "grad_norm": 0.44140625, + "learning_rate": 0.0019497656689713538, + "loss": 0.1235, + "step": 11152 + }, + { + "epoch": 0.09681339571705107, + "grad_norm": 0.89453125, + "learning_rate": 0.0019497558642082246, + "loss": 0.1348, + "step": 11153 + }, + { + "epoch": 0.09682207619725523, + "grad_norm": 0.341796875, + "learning_rate": 0.0019497460585158146, + "loss": 0.1357, + "step": 11154 + }, + { + "epoch": 0.0968307566774594, + "grad_norm": 0.28125, + "learning_rate": 0.0019497362518941356, + "loss": 0.1167, + "step": 11155 + }, + { + "epoch": 0.09683943715766356, + "grad_norm": 0.185546875, + "learning_rate": 0.0019497264443431978, + "loss": 0.0977, + "step": 11156 + }, + { + "epoch": 0.09684811763786773, + "grad_norm": 0.248046875, + "learning_rate": 0.0019497166358630118, + "loss": 0.125, + "step": 11157 + }, + { + "epoch": 0.0968567981180719, + "grad_norm": 0.34375, + "learning_rate": 0.0019497068264535883, + "loss": 0.1074, + "step": 11158 + }, + { + "epoch": 0.09686547859827606, + "grad_norm": 0.96875, + "learning_rate": 0.0019496970161149384, + "loss": 0.1553, + "step": 11159 + }, + { + "epoch": 0.09687415907848022, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019496872048470725, + "loss": 0.1167, + "step": 11160 + }, + { + "epoch": 0.09688283955868439, + "grad_norm": 0.158203125, + "learning_rate": 0.0019496773926500016, + "loss": 0.1406, + "step": 11161 + }, + { + "epoch": 0.09689152003888855, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019496675795237363, + "loss": 0.1206, + "step": 11162 + }, + { + "epoch": 0.09690020051909272, + "grad_norm": 0.7421875, + "learning_rate": 0.0019496577654682873, + "loss": 0.1387, + "step": 11163 + }, + { + "epoch": 0.09690888099929688, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019496479504836647, + "loss": 0.0942, + "step": 11164 + }, + { + "epoch": 0.09691756147950105, + "grad_norm": 0.35546875, + "learning_rate": 0.0019496381345698807, + "loss": 0.1025, + "step": 11165 + }, + { + "epoch": 0.09692624195970521, + "grad_norm": 0.0986328125, + "learning_rate": 0.001949628317726945, + "loss": 0.125, + "step": 11166 + }, + { + "epoch": 0.09693492243990938, + "grad_norm": 0.7265625, + "learning_rate": 0.0019496184999548683, + "loss": 0.1187, + "step": 11167 + }, + { + "epoch": 0.09694360292011354, + "grad_norm": 0.77734375, + "learning_rate": 0.0019496086812536616, + "loss": 0.125, + "step": 11168 + }, + { + "epoch": 0.09695228340031771, + "grad_norm": 0.7578125, + "learning_rate": 0.0019495988616233355, + "loss": 0.1484, + "step": 11169 + }, + { + "epoch": 0.09696096388052187, + "grad_norm": 0.158203125, + "learning_rate": 0.0019495890410639012, + "loss": 0.1001, + "step": 11170 + }, + { + "epoch": 0.09696964436072604, + "grad_norm": 0.85546875, + "learning_rate": 0.001949579219575369, + "loss": 0.1348, + "step": 11171 + }, + { + "epoch": 0.0969783248409302, + "grad_norm": 0.546875, + "learning_rate": 0.0019495693971577498, + "loss": 0.1162, + "step": 11172 + }, + { + "epoch": 0.09698700532113437, + "grad_norm": 0.13671875, + "learning_rate": 0.001949559573811054, + "loss": 0.0942, + "step": 11173 + }, + { + "epoch": 0.09699568580133854, + "grad_norm": 0.310546875, + "learning_rate": 0.0019495497495352926, + "loss": 0.1416, + "step": 11174 + }, + { + "epoch": 0.0970043662815427, + "grad_norm": 0.205078125, + "learning_rate": 0.0019495399243304767, + "loss": 0.1445, + "step": 11175 + }, + { + "epoch": 0.09701304676174687, + "grad_norm": 0.55078125, + "learning_rate": 0.0019495300981966165, + "loss": 0.1094, + "step": 11176 + }, + { + "epoch": 0.09702172724195103, + "grad_norm": 0.84765625, + "learning_rate": 0.001949520271133723, + "loss": 0.1006, + "step": 11177 + }, + { + "epoch": 0.0970304077221552, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019495104431418067, + "loss": 0.1328, + "step": 11178 + }, + { + "epoch": 0.09703908820235936, + "grad_norm": 0.84375, + "learning_rate": 0.0019495006142208786, + "loss": 0.1777, + "step": 11179 + }, + { + "epoch": 0.09704776868256353, + "grad_norm": 0.26953125, + "learning_rate": 0.0019494907843709496, + "loss": 0.1172, + "step": 11180 + }, + { + "epoch": 0.09705644916276769, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019494809535920302, + "loss": 0.1094, + "step": 11181 + }, + { + "epoch": 0.09706512964297184, + "grad_norm": 0.0859375, + "learning_rate": 0.001949471121884131, + "loss": 0.1143, + "step": 11182 + }, + { + "epoch": 0.09707381012317601, + "grad_norm": 1.0546875, + "learning_rate": 0.001949461289247263, + "loss": 0.1533, + "step": 11183 + }, + { + "epoch": 0.09708249060338017, + "grad_norm": 1.21875, + "learning_rate": 0.0019494514556814369, + "loss": 0.1133, + "step": 11184 + }, + { + "epoch": 0.09709117108358434, + "grad_norm": 0.16015625, + "learning_rate": 0.0019494416211866636, + "loss": 0.1396, + "step": 11185 + }, + { + "epoch": 0.0970998515637885, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019494317857629536, + "loss": 0.1689, + "step": 11186 + }, + { + "epoch": 0.09710853204399267, + "grad_norm": 0.29296875, + "learning_rate": 0.0019494219494103177, + "loss": 0.1855, + "step": 11187 + }, + { + "epoch": 0.09711721252419683, + "grad_norm": 0.3671875, + "learning_rate": 0.0019494121121287672, + "loss": 0.126, + "step": 11188 + }, + { + "epoch": 0.097125893004401, + "grad_norm": 0.2265625, + "learning_rate": 0.0019494022739183117, + "loss": 0.0801, + "step": 11189 + }, + { + "epoch": 0.09713457348460516, + "grad_norm": 0.36328125, + "learning_rate": 0.001949392434778963, + "loss": 0.1055, + "step": 11190 + }, + { + "epoch": 0.09714325396480933, + "grad_norm": 0.177734375, + "learning_rate": 0.0019493825947107313, + "loss": 0.1069, + "step": 11191 + }, + { + "epoch": 0.0971519344450135, + "grad_norm": 0.361328125, + "learning_rate": 0.0019493727537136277, + "loss": 0.0781, + "step": 11192 + }, + { + "epoch": 0.09716061492521766, + "grad_norm": 0.16796875, + "learning_rate": 0.0019493629117876626, + "loss": 0.0825, + "step": 11193 + }, + { + "epoch": 0.09716929540542182, + "grad_norm": 0.82421875, + "learning_rate": 0.0019493530689328473, + "loss": 0.1338, + "step": 11194 + }, + { + "epoch": 0.09717797588562599, + "grad_norm": 0.349609375, + "learning_rate": 0.0019493432251491921, + "loss": 0.1113, + "step": 11195 + }, + { + "epoch": 0.09718665636583015, + "grad_norm": 0.150390625, + "learning_rate": 0.0019493333804367079, + "loss": 0.1104, + "step": 11196 + }, + { + "epoch": 0.09719533684603432, + "grad_norm": 0.337890625, + "learning_rate": 0.0019493235347954054, + "loss": 0.127, + "step": 11197 + }, + { + "epoch": 0.09720401732623848, + "grad_norm": 0.392578125, + "learning_rate": 0.0019493136882252958, + "loss": 0.1426, + "step": 11198 + }, + { + "epoch": 0.09721269780644265, + "grad_norm": 0.41796875, + "learning_rate": 0.001949303840726389, + "loss": 0.1172, + "step": 11199 + }, + { + "epoch": 0.09722137828664681, + "grad_norm": 0.388671875, + "learning_rate": 0.0019492939922986965, + "loss": 0.0986, + "step": 11200 + }, + { + "epoch": 0.09723005876685098, + "grad_norm": 0.0869140625, + "learning_rate": 0.001949284142942229, + "loss": 0.125, + "step": 11201 + }, + { + "epoch": 0.09723873924705514, + "grad_norm": 0.103515625, + "learning_rate": 0.001949274292656997, + "loss": 0.1738, + "step": 11202 + }, + { + "epoch": 0.09724741972725931, + "grad_norm": 0.365234375, + "learning_rate": 0.0019492644414430117, + "loss": 0.1211, + "step": 11203 + }, + { + "epoch": 0.09725610020746348, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019492545893002831, + "loss": 0.1387, + "step": 11204 + }, + { + "epoch": 0.09726478068766764, + "grad_norm": 0.30078125, + "learning_rate": 0.0019492447362288224, + "loss": 0.1455, + "step": 11205 + }, + { + "epoch": 0.0972734611678718, + "grad_norm": 0.55859375, + "learning_rate": 0.0019492348822286406, + "loss": 0.1045, + "step": 11206 + }, + { + "epoch": 0.09728214164807597, + "grad_norm": 0.66015625, + "learning_rate": 0.0019492250272997486, + "loss": 0.1465, + "step": 11207 + }, + { + "epoch": 0.09729082212828014, + "grad_norm": 0.380859375, + "learning_rate": 0.0019492151714421565, + "loss": 0.1367, + "step": 11208 + }, + { + "epoch": 0.0972995026084843, + "grad_norm": 0.9609375, + "learning_rate": 0.0019492053146558752, + "loss": 0.0884, + "step": 11209 + }, + { + "epoch": 0.09730818308868847, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019491954569409162, + "loss": 0.1406, + "step": 11210 + }, + { + "epoch": 0.09731686356889263, + "grad_norm": 0.43359375, + "learning_rate": 0.0019491855982972897, + "loss": 0.168, + "step": 11211 + }, + { + "epoch": 0.0973255440490968, + "grad_norm": 0.26953125, + "learning_rate": 0.0019491757387250066, + "loss": 0.1318, + "step": 11212 + }, + { + "epoch": 0.09733422452930096, + "grad_norm": 0.421875, + "learning_rate": 0.0019491658782240775, + "loss": 0.1289, + "step": 11213 + }, + { + "epoch": 0.09734290500950513, + "grad_norm": 0.1328125, + "learning_rate": 0.0019491560167945138, + "loss": 0.0869, + "step": 11214 + }, + { + "epoch": 0.09735158548970929, + "grad_norm": 0.63671875, + "learning_rate": 0.0019491461544363252, + "loss": 0.123, + "step": 11215 + }, + { + "epoch": 0.09736026596991346, + "grad_norm": 0.54296875, + "learning_rate": 0.0019491362911495233, + "loss": 0.1572, + "step": 11216 + }, + { + "epoch": 0.09736894645011762, + "grad_norm": 0.34375, + "learning_rate": 0.001949126426934119, + "loss": 0.0967, + "step": 11217 + }, + { + "epoch": 0.09737762693032179, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019491165617901228, + "loss": 0.0889, + "step": 11218 + }, + { + "epoch": 0.09738630741052595, + "grad_norm": 0.55859375, + "learning_rate": 0.0019491066957175451, + "loss": 0.1211, + "step": 11219 + }, + { + "epoch": 0.09739498789073012, + "grad_norm": 0.474609375, + "learning_rate": 0.0019490968287163976, + "loss": 0.1143, + "step": 11220 + }, + { + "epoch": 0.09740366837093428, + "grad_norm": 0.54296875, + "learning_rate": 0.00194908696078669, + "loss": 0.1074, + "step": 11221 + }, + { + "epoch": 0.09741234885113845, + "grad_norm": 0.34765625, + "learning_rate": 0.001949077091928434, + "loss": 0.1777, + "step": 11222 + }, + { + "epoch": 0.09742102933134261, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019490672221416398, + "loss": 0.1445, + "step": 11223 + }, + { + "epoch": 0.09742970981154678, + "grad_norm": 0.13671875, + "learning_rate": 0.0019490573514263187, + "loss": 0.1357, + "step": 11224 + }, + { + "epoch": 0.09743839029175094, + "grad_norm": 0.63671875, + "learning_rate": 0.0019490474797824813, + "loss": 0.1299, + "step": 11225 + }, + { + "epoch": 0.09744707077195511, + "grad_norm": 0.244140625, + "learning_rate": 0.001949037607210138, + "loss": 0.1025, + "step": 11226 + }, + { + "epoch": 0.09745575125215927, + "grad_norm": 0.3125, + "learning_rate": 0.0019490277337093002, + "loss": 0.1016, + "step": 11227 + }, + { + "epoch": 0.09746443173236344, + "grad_norm": 0.6484375, + "learning_rate": 0.0019490178592799785, + "loss": 0.1367, + "step": 11228 + }, + { + "epoch": 0.0974731122125676, + "grad_norm": 0.546875, + "learning_rate": 0.0019490079839221835, + "loss": 0.1816, + "step": 11229 + }, + { + "epoch": 0.09748179269277177, + "grad_norm": 0.10498046875, + "learning_rate": 0.001948998107635926, + "loss": 0.1089, + "step": 11230 + }, + { + "epoch": 0.09749047317297593, + "grad_norm": 1.6171875, + "learning_rate": 0.001948988230421217, + "loss": 0.1133, + "step": 11231 + }, + { + "epoch": 0.0974991536531801, + "grad_norm": 0.19140625, + "learning_rate": 0.0019489783522780675, + "loss": 0.0967, + "step": 11232 + }, + { + "epoch": 0.09750783413338426, + "grad_norm": 0.82421875, + "learning_rate": 0.0019489684732064876, + "loss": 0.1377, + "step": 11233 + }, + { + "epoch": 0.09751651461358843, + "grad_norm": 0.302734375, + "learning_rate": 0.0019489585932064887, + "loss": 0.1602, + "step": 11234 + }, + { + "epoch": 0.0975251950937926, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019489487122780814, + "loss": 0.1064, + "step": 11235 + }, + { + "epoch": 0.09753387557399676, + "grad_norm": 0.400390625, + "learning_rate": 0.0019489388304212766, + "loss": 0.1187, + "step": 11236 + }, + { + "epoch": 0.09754255605420092, + "grad_norm": 0.408203125, + "learning_rate": 0.0019489289476360852, + "loss": 0.1191, + "step": 11237 + }, + { + "epoch": 0.09755123653440509, + "grad_norm": 0.66015625, + "learning_rate": 0.0019489190639225176, + "loss": 0.1426, + "step": 11238 + }, + { + "epoch": 0.09755991701460925, + "grad_norm": 0.451171875, + "learning_rate": 0.0019489091792805851, + "loss": 0.1377, + "step": 11239 + }, + { + "epoch": 0.09756859749481342, + "grad_norm": 0.462890625, + "learning_rate": 0.001948899293710298, + "loss": 0.1387, + "step": 11240 + }, + { + "epoch": 0.09757727797501758, + "grad_norm": 1.265625, + "learning_rate": 0.0019488894072116677, + "loss": 0.1582, + "step": 11241 + }, + { + "epoch": 0.09758595845522175, + "grad_norm": 0.8515625, + "learning_rate": 0.0019488795197847044, + "loss": 0.1484, + "step": 11242 + }, + { + "epoch": 0.0975946389354259, + "grad_norm": 0.197265625, + "learning_rate": 0.0019488696314294194, + "loss": 0.1084, + "step": 11243 + }, + { + "epoch": 0.09760331941563007, + "grad_norm": 0.49609375, + "learning_rate": 0.0019488597421458231, + "loss": 0.125, + "step": 11244 + }, + { + "epoch": 0.09761199989583423, + "grad_norm": 0.263671875, + "learning_rate": 0.0019488498519339267, + "loss": 0.1387, + "step": 11245 + }, + { + "epoch": 0.0976206803760384, + "grad_norm": 0.107421875, + "learning_rate": 0.001948839960793741, + "loss": 0.1592, + "step": 11246 + }, + { + "epoch": 0.09762936085624256, + "grad_norm": 0.25390625, + "learning_rate": 0.0019488300687252765, + "loss": 0.1611, + "step": 11247 + }, + { + "epoch": 0.09763804133644673, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019488201757285441, + "loss": 0.1191, + "step": 11248 + }, + { + "epoch": 0.09764672181665089, + "grad_norm": 0.52734375, + "learning_rate": 0.001948810281803555, + "loss": 0.1641, + "step": 11249 + }, + { + "epoch": 0.09765540229685506, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019488003869503198, + "loss": 0.105, + "step": 11250 + }, + { + "epoch": 0.09766408277705922, + "grad_norm": 0.50390625, + "learning_rate": 0.0019487904911688488, + "loss": 0.0972, + "step": 11251 + }, + { + "epoch": 0.09767276325726339, + "grad_norm": 0.486328125, + "learning_rate": 0.0019487805944591536, + "loss": 0.0903, + "step": 11252 + }, + { + "epoch": 0.09768144373746755, + "grad_norm": 0.236328125, + "learning_rate": 0.0019487706968212444, + "loss": 0.1348, + "step": 11253 + }, + { + "epoch": 0.09769012421767172, + "grad_norm": 0.19140625, + "learning_rate": 0.0019487607982551325, + "loss": 0.0996, + "step": 11254 + }, + { + "epoch": 0.09769880469787588, + "grad_norm": 0.34765625, + "learning_rate": 0.0019487508987608287, + "loss": 0.1504, + "step": 11255 + }, + { + "epoch": 0.09770748517808005, + "grad_norm": 0.275390625, + "learning_rate": 0.0019487409983383434, + "loss": 0.1445, + "step": 11256 + }, + { + "epoch": 0.09771616565828421, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019487310969876876, + "loss": 0.1211, + "step": 11257 + }, + { + "epoch": 0.09772484613848838, + "grad_norm": 0.875, + "learning_rate": 0.0019487211947088728, + "loss": 0.127, + "step": 11258 + }, + { + "epoch": 0.09773352661869254, + "grad_norm": 0.484375, + "learning_rate": 0.0019487112915019087, + "loss": 0.1074, + "step": 11259 + }, + { + "epoch": 0.09774220709889671, + "grad_norm": 0.4375, + "learning_rate": 0.001948701387366807, + "loss": 0.1445, + "step": 11260 + }, + { + "epoch": 0.09775088757910087, + "grad_norm": 0.083984375, + "learning_rate": 0.0019486914823035782, + "loss": 0.1455, + "step": 11261 + }, + { + "epoch": 0.09775956805930504, + "grad_norm": 0.10205078125, + "learning_rate": 0.001948681576312233, + "loss": 0.1064, + "step": 11262 + }, + { + "epoch": 0.0977682485395092, + "grad_norm": 0.671875, + "learning_rate": 0.0019486716693927824, + "loss": 0.0859, + "step": 11263 + }, + { + "epoch": 0.09777692901971337, + "grad_norm": 0.365234375, + "learning_rate": 0.0019486617615452372, + "loss": 0.1196, + "step": 11264 + }, + { + "epoch": 0.09778560949991753, + "grad_norm": 0.380859375, + "learning_rate": 0.0019486518527696082, + "loss": 0.1045, + "step": 11265 + }, + { + "epoch": 0.0977942899801217, + "grad_norm": 0.73046875, + "learning_rate": 0.0019486419430659063, + "loss": 0.1475, + "step": 11266 + }, + { + "epoch": 0.09780297046032586, + "grad_norm": 0.91015625, + "learning_rate": 0.0019486320324341427, + "loss": 0.1147, + "step": 11267 + }, + { + "epoch": 0.09781165094053003, + "grad_norm": 0.5703125, + "learning_rate": 0.0019486221208743274, + "loss": 0.1641, + "step": 11268 + }, + { + "epoch": 0.0978203314207342, + "grad_norm": 0.126953125, + "learning_rate": 0.0019486122083864717, + "loss": 0.1475, + "step": 11269 + }, + { + "epoch": 0.09782901190093836, + "grad_norm": 0.103515625, + "learning_rate": 0.0019486022949705866, + "loss": 0.1504, + "step": 11270 + }, + { + "epoch": 0.09783769238114252, + "grad_norm": 0.25, + "learning_rate": 0.0019485923806266828, + "loss": 0.126, + "step": 11271 + }, + { + "epoch": 0.09784637286134669, + "grad_norm": 0.11181640625, + "learning_rate": 0.001948582465354771, + "loss": 0.1289, + "step": 11272 + }, + { + "epoch": 0.09785505334155085, + "grad_norm": 0.478515625, + "learning_rate": 0.0019485725491548624, + "loss": 0.1162, + "step": 11273 + }, + { + "epoch": 0.09786373382175502, + "grad_norm": 0.2890625, + "learning_rate": 0.0019485626320269675, + "loss": 0.1084, + "step": 11274 + }, + { + "epoch": 0.09787241430195918, + "grad_norm": 0.17578125, + "learning_rate": 0.001948552713971097, + "loss": 0.1289, + "step": 11275 + }, + { + "epoch": 0.09788109478216335, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019485427949872624, + "loss": 0.1328, + "step": 11276 + }, + { + "epoch": 0.09788977526236751, + "grad_norm": 0.412109375, + "learning_rate": 0.0019485328750754739, + "loss": 0.1133, + "step": 11277 + }, + { + "epoch": 0.09789845574257168, + "grad_norm": 0.466796875, + "learning_rate": 0.0019485229542357426, + "loss": 0.1123, + "step": 11278 + }, + { + "epoch": 0.09790713622277585, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019485130324680795, + "loss": 0.1162, + "step": 11279 + }, + { + "epoch": 0.09791581670298001, + "grad_norm": 0.09033203125, + "learning_rate": 0.001948503109772495, + "loss": 0.1045, + "step": 11280 + }, + { + "epoch": 0.09792449718318418, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019484931861490004, + "loss": 0.0967, + "step": 11281 + }, + { + "epoch": 0.09793317766338834, + "grad_norm": 0.5859375, + "learning_rate": 0.0019484832615976066, + "loss": 0.125, + "step": 11282 + }, + { + "epoch": 0.0979418581435925, + "grad_norm": 0.263671875, + "learning_rate": 0.0019484733361183242, + "loss": 0.1069, + "step": 11283 + }, + { + "epoch": 0.09795053862379667, + "grad_norm": 0.2099609375, + "learning_rate": 0.001948463409711164, + "loss": 0.1299, + "step": 11284 + }, + { + "epoch": 0.09795921910400084, + "grad_norm": 0.8828125, + "learning_rate": 0.001948453482376137, + "loss": 0.125, + "step": 11285 + }, + { + "epoch": 0.097967899584205, + "grad_norm": 0.197265625, + "learning_rate": 0.0019484435541132542, + "loss": 0.1113, + "step": 11286 + }, + { + "epoch": 0.09797658006440917, + "grad_norm": 0.11328125, + "learning_rate": 0.001948433624922526, + "loss": 0.1055, + "step": 11287 + }, + { + "epoch": 0.09798526054461333, + "grad_norm": 0.318359375, + "learning_rate": 0.0019484236948039636, + "loss": 0.125, + "step": 11288 + }, + { + "epoch": 0.0979939410248175, + "grad_norm": 0.61328125, + "learning_rate": 0.0019484137637575778, + "loss": 0.1221, + "step": 11289 + }, + { + "epoch": 0.09800262150502166, + "grad_norm": 0.5078125, + "learning_rate": 0.0019484038317833792, + "loss": 0.1152, + "step": 11290 + }, + { + "epoch": 0.09801130198522583, + "grad_norm": 0.126953125, + "learning_rate": 0.0019483938988813794, + "loss": 0.1147, + "step": 11291 + }, + { + "epoch": 0.09801998246542999, + "grad_norm": 0.41015625, + "learning_rate": 0.0019483839650515885, + "loss": 0.0947, + "step": 11292 + }, + { + "epoch": 0.09802866294563416, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019483740302940176, + "loss": 0.1011, + "step": 11293 + }, + { + "epoch": 0.09803734342583832, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019483640946086778, + "loss": 0.1377, + "step": 11294 + }, + { + "epoch": 0.09804602390604249, + "grad_norm": 0.390625, + "learning_rate": 0.0019483541579955797, + "loss": 0.1201, + "step": 11295 + }, + { + "epoch": 0.09805470438624665, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019483442204547342, + "loss": 0.168, + "step": 11296 + }, + { + "epoch": 0.09806338486645082, + "grad_norm": 0.6484375, + "learning_rate": 0.001948334281986152, + "loss": 0.1191, + "step": 11297 + }, + { + "epoch": 0.09807206534665498, + "grad_norm": 0.189453125, + "learning_rate": 0.0019483243425898445, + "loss": 0.1152, + "step": 11298 + }, + { + "epoch": 0.09808074582685915, + "grad_norm": 0.08984375, + "learning_rate": 0.001948314402265822, + "loss": 0.1221, + "step": 11299 + }, + { + "epoch": 0.09808942630706331, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019483044610140961, + "loss": 0.1289, + "step": 11300 + }, + { + "epoch": 0.09809810678726748, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019482945188346765, + "loss": 0.1406, + "step": 11301 + }, + { + "epoch": 0.09810678726747164, + "grad_norm": 0.546875, + "learning_rate": 0.0019482845757275754, + "loss": 0.1338, + "step": 11302 + }, + { + "epoch": 0.09811546774767581, + "grad_norm": 0.26953125, + "learning_rate": 0.0019482746316928025, + "loss": 0.0747, + "step": 11303 + }, + { + "epoch": 0.09812414822787997, + "grad_norm": 0.64453125, + "learning_rate": 0.0019482646867303697, + "loss": 0.1162, + "step": 11304 + }, + { + "epoch": 0.09813282870808412, + "grad_norm": 0.81640625, + "learning_rate": 0.001948254740840287, + "loss": 0.1128, + "step": 11305 + }, + { + "epoch": 0.09814150918828829, + "grad_norm": 0.455078125, + "learning_rate": 0.001948244794022566, + "loss": 0.0854, + "step": 11306 + }, + { + "epoch": 0.09815018966849245, + "grad_norm": 0.435546875, + "learning_rate": 0.001948234846277217, + "loss": 0.1289, + "step": 11307 + }, + { + "epoch": 0.09815887014869662, + "grad_norm": 0.11474609375, + "learning_rate": 0.001948224897604251, + "loss": 0.126, + "step": 11308 + }, + { + "epoch": 0.09816755062890078, + "grad_norm": 0.62109375, + "learning_rate": 0.0019482149480036792, + "loss": 0.0962, + "step": 11309 + }, + { + "epoch": 0.09817623110910495, + "grad_norm": 0.78515625, + "learning_rate": 0.0019482049974755124, + "loss": 0.1553, + "step": 11310 + }, + { + "epoch": 0.09818491158930912, + "grad_norm": 0.45703125, + "learning_rate": 0.0019481950460197611, + "loss": 0.123, + "step": 11311 + }, + { + "epoch": 0.09819359206951328, + "grad_norm": 0.125, + "learning_rate": 0.0019481850936364366, + "loss": 0.1123, + "step": 11312 + }, + { + "epoch": 0.09820227254971745, + "grad_norm": 0.390625, + "learning_rate": 0.0019481751403255495, + "loss": 0.125, + "step": 11313 + }, + { + "epoch": 0.09821095302992161, + "grad_norm": 0.75, + "learning_rate": 0.0019481651860871107, + "loss": 0.1196, + "step": 11314 + }, + { + "epoch": 0.09821963351012578, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019481552309211314, + "loss": 0.1426, + "step": 11315 + }, + { + "epoch": 0.09822831399032994, + "grad_norm": 1.0, + "learning_rate": 0.0019481452748276223, + "loss": 0.1377, + "step": 11316 + }, + { + "epoch": 0.0982369944705341, + "grad_norm": 0.087890625, + "learning_rate": 0.0019481353178065941, + "loss": 0.1162, + "step": 11317 + }, + { + "epoch": 0.09824567495073827, + "grad_norm": 0.86328125, + "learning_rate": 0.0019481253598580583, + "loss": 0.1064, + "step": 11318 + }, + { + "epoch": 0.09825435543094244, + "grad_norm": 1.1796875, + "learning_rate": 0.0019481154009820245, + "loss": 0.125, + "step": 11319 + }, + { + "epoch": 0.0982630359111466, + "grad_norm": 0.361328125, + "learning_rate": 0.001948105441178505, + "loss": 0.1445, + "step": 11320 + }, + { + "epoch": 0.09827171639135077, + "grad_norm": 0.181640625, + "learning_rate": 0.0019480954804475105, + "loss": 0.1104, + "step": 11321 + }, + { + "epoch": 0.09828039687155493, + "grad_norm": 0.76953125, + "learning_rate": 0.001948085518789051, + "loss": 0.1582, + "step": 11322 + }, + { + "epoch": 0.0982890773517591, + "grad_norm": 0.51953125, + "learning_rate": 0.001948075556203138, + "loss": 0.1123, + "step": 11323 + }, + { + "epoch": 0.09829775783196326, + "grad_norm": 1.640625, + "learning_rate": 0.0019480655926897824, + "loss": 0.1172, + "step": 11324 + }, + { + "epoch": 0.09830643831216743, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019480556282489947, + "loss": 0.1016, + "step": 11325 + }, + { + "epoch": 0.09831511879237159, + "grad_norm": 0.21484375, + "learning_rate": 0.0019480456628807865, + "loss": 0.0684, + "step": 11326 + }, + { + "epoch": 0.09832379927257576, + "grad_norm": 0.78515625, + "learning_rate": 0.0019480356965851679, + "loss": 0.1123, + "step": 11327 + }, + { + "epoch": 0.09833247975277992, + "grad_norm": 0.5546875, + "learning_rate": 0.001948025729362151, + "loss": 0.1445, + "step": 11328 + }, + { + "epoch": 0.09834116023298409, + "grad_norm": 0.44921875, + "learning_rate": 0.0019480157612117453, + "loss": 0.1221, + "step": 11329 + }, + { + "epoch": 0.09834984071318825, + "grad_norm": 0.78125, + "learning_rate": 0.0019480057921339622, + "loss": 0.0962, + "step": 11330 + }, + { + "epoch": 0.09835852119339242, + "grad_norm": 1.140625, + "learning_rate": 0.001947995822128813, + "loss": 0.1172, + "step": 11331 + }, + { + "epoch": 0.09836720167359658, + "grad_norm": 0.330078125, + "learning_rate": 0.001947985851196308, + "loss": 0.1357, + "step": 11332 + }, + { + "epoch": 0.09837588215380075, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019479758793364587, + "loss": 0.0962, + "step": 11333 + }, + { + "epoch": 0.09838456263400491, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019479659065492756, + "loss": 0.1699, + "step": 11334 + }, + { + "epoch": 0.09839324311420908, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019479559328347698, + "loss": 0.0864, + "step": 11335 + }, + { + "epoch": 0.09840192359441324, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019479459581929523, + "loss": 0.1084, + "step": 11336 + }, + { + "epoch": 0.09841060407461741, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019479359826238335, + "loss": 0.0996, + "step": 11337 + }, + { + "epoch": 0.09841928455482157, + "grad_norm": 0.06982421875, + "learning_rate": 0.001947926006127425, + "loss": 0.0781, + "step": 11338 + }, + { + "epoch": 0.09842796503502574, + "grad_norm": 0.345703125, + "learning_rate": 0.0019479160287037367, + "loss": 0.1543, + "step": 11339 + }, + { + "epoch": 0.0984366455152299, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019479060503527809, + "loss": 0.1123, + "step": 11340 + }, + { + "epoch": 0.09844532599543407, + "grad_norm": 0.27734375, + "learning_rate": 0.0019478960710745674, + "loss": 0.0752, + "step": 11341 + }, + { + "epoch": 0.09845400647563823, + "grad_norm": 0.271484375, + "learning_rate": 0.0019478860908691074, + "loss": 0.1118, + "step": 11342 + }, + { + "epoch": 0.0984626869558424, + "grad_norm": 0.08642578125, + "learning_rate": 0.001947876109736412, + "loss": 0.0938, + "step": 11343 + }, + { + "epoch": 0.09847136743604656, + "grad_norm": 0.3671875, + "learning_rate": 0.0019478661276764923, + "loss": 0.1094, + "step": 11344 + }, + { + "epoch": 0.09848004791625073, + "grad_norm": 0.298828125, + "learning_rate": 0.0019478561446893588, + "loss": 0.0938, + "step": 11345 + }, + { + "epoch": 0.0984887283964549, + "grad_norm": 0.216796875, + "learning_rate": 0.0019478461607750227, + "loss": 0.1367, + "step": 11346 + }, + { + "epoch": 0.09849740887665906, + "grad_norm": 0.78125, + "learning_rate": 0.0019478361759334947, + "loss": 0.1206, + "step": 11347 + }, + { + "epoch": 0.09850608935686322, + "grad_norm": 0.87890625, + "learning_rate": 0.0019478261901647856, + "loss": 0.1455, + "step": 11348 + }, + { + "epoch": 0.09851476983706739, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019478162034689064, + "loss": 0.123, + "step": 11349 + }, + { + "epoch": 0.09852345031727155, + "grad_norm": 0.314453125, + "learning_rate": 0.0019478062158458686, + "loss": 0.1377, + "step": 11350 + }, + { + "epoch": 0.09853213079747572, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019477962272956822, + "loss": 0.1436, + "step": 11351 + }, + { + "epoch": 0.09854081127767989, + "grad_norm": 0.59375, + "learning_rate": 0.001947786237818359, + "loss": 0.0977, + "step": 11352 + }, + { + "epoch": 0.09854949175788405, + "grad_norm": 0.2060546875, + "learning_rate": 0.001947776247413909, + "loss": 0.0898, + "step": 11353 + }, + { + "epoch": 0.09855817223808822, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019477662560823442, + "loss": 0.124, + "step": 11354 + }, + { + "epoch": 0.09856685271829238, + "grad_norm": 0.6484375, + "learning_rate": 0.0019477562638236746, + "loss": 0.0986, + "step": 11355 + }, + { + "epoch": 0.09857553319849655, + "grad_norm": 0.6171875, + "learning_rate": 0.0019477462706379114, + "loss": 0.0918, + "step": 11356 + }, + { + "epoch": 0.09858421367870071, + "grad_norm": 0.224609375, + "learning_rate": 0.001947736276525066, + "loss": 0.1367, + "step": 11357 + }, + { + "epoch": 0.09859289415890488, + "grad_norm": 0.212890625, + "learning_rate": 0.0019477262814851489, + "loss": 0.1396, + "step": 11358 + }, + { + "epoch": 0.09860157463910904, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019477162855181707, + "loss": 0.1221, + "step": 11359 + }, + { + "epoch": 0.0986102551193132, + "grad_norm": 0.8359375, + "learning_rate": 0.001947706288624143, + "loss": 0.1064, + "step": 11360 + }, + { + "epoch": 0.09861893559951737, + "grad_norm": 0.470703125, + "learning_rate": 0.0019476962908030763, + "loss": 0.0981, + "step": 11361 + }, + { + "epoch": 0.09862761607972154, + "grad_norm": 0.546875, + "learning_rate": 0.0019476862920549819, + "loss": 0.1377, + "step": 11362 + }, + { + "epoch": 0.0986362965599257, + "grad_norm": 0.416015625, + "learning_rate": 0.0019476762923798703, + "loss": 0.0889, + "step": 11363 + }, + { + "epoch": 0.09864497704012987, + "grad_norm": 0.91796875, + "learning_rate": 0.0019476662917777526, + "loss": 0.1094, + "step": 11364 + }, + { + "epoch": 0.09865365752033403, + "grad_norm": 0.5234375, + "learning_rate": 0.0019476562902486399, + "loss": 0.1602, + "step": 11365 + }, + { + "epoch": 0.0986623380005382, + "grad_norm": 0.0830078125, + "learning_rate": 0.001947646287792543, + "loss": 0.1172, + "step": 11366 + }, + { + "epoch": 0.09867101848074235, + "grad_norm": 2.21875, + "learning_rate": 0.0019476362844094727, + "loss": 0.2695, + "step": 11367 + }, + { + "epoch": 0.09867969896094651, + "grad_norm": 0.248046875, + "learning_rate": 0.0019476262800994404, + "loss": 0.1367, + "step": 11368 + }, + { + "epoch": 0.09868837944115068, + "grad_norm": 0.451171875, + "learning_rate": 0.0019476162748624567, + "loss": 0.1367, + "step": 11369 + }, + { + "epoch": 0.09869705992135484, + "grad_norm": 0.10546875, + "learning_rate": 0.0019476062686985323, + "loss": 0.1348, + "step": 11370 + }, + { + "epoch": 0.09870574040155901, + "grad_norm": 0.453125, + "learning_rate": 0.0019475962616076787, + "loss": 0.1221, + "step": 11371 + }, + { + "epoch": 0.09871442088176317, + "grad_norm": 0.302734375, + "learning_rate": 0.0019475862535899063, + "loss": 0.1406, + "step": 11372 + }, + { + "epoch": 0.09872310136196734, + "grad_norm": 0.279296875, + "learning_rate": 0.0019475762446452263, + "loss": 0.1533, + "step": 11373 + }, + { + "epoch": 0.0987317818421715, + "grad_norm": 0.169921875, + "learning_rate": 0.00194756623477365, + "loss": 0.1084, + "step": 11374 + }, + { + "epoch": 0.09874046232237567, + "grad_norm": 0.421875, + "learning_rate": 0.0019475562239751878, + "loss": 0.1523, + "step": 11375 + }, + { + "epoch": 0.09874914280257983, + "grad_norm": 0.78515625, + "learning_rate": 0.001947546212249851, + "loss": 0.1416, + "step": 11376 + }, + { + "epoch": 0.098757823282784, + "grad_norm": 0.6171875, + "learning_rate": 0.0019475361995976502, + "loss": 0.1406, + "step": 11377 + }, + { + "epoch": 0.09876650376298816, + "grad_norm": 0.291015625, + "learning_rate": 0.0019475261860185963, + "loss": 0.1138, + "step": 11378 + }, + { + "epoch": 0.09877518424319233, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019475161715127012, + "loss": 0.1084, + "step": 11379 + }, + { + "epoch": 0.0987838647233965, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019475061560799746, + "loss": 0.1035, + "step": 11380 + }, + { + "epoch": 0.09879254520360066, + "grad_norm": 0.59765625, + "learning_rate": 0.0019474961397204283, + "loss": 0.1309, + "step": 11381 + }, + { + "epoch": 0.09880122568380482, + "grad_norm": 0.6796875, + "learning_rate": 0.0019474861224340724, + "loss": 0.1104, + "step": 11382 + }, + { + "epoch": 0.09880990616400899, + "grad_norm": 0.61328125, + "learning_rate": 0.001947476104220919, + "loss": 0.2246, + "step": 11383 + }, + { + "epoch": 0.09881858664421315, + "grad_norm": 0.328125, + "learning_rate": 0.0019474660850809782, + "loss": 0.0972, + "step": 11384 + }, + { + "epoch": 0.09882726712441732, + "grad_norm": 0.80859375, + "learning_rate": 0.0019474560650142613, + "loss": 0.1172, + "step": 11385 + }, + { + "epoch": 0.09883594760462149, + "grad_norm": 0.203125, + "learning_rate": 0.0019474460440207792, + "loss": 0.126, + "step": 11386 + }, + { + "epoch": 0.09884462808482565, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019474360221005426, + "loss": 0.1328, + "step": 11387 + }, + { + "epoch": 0.09885330856502982, + "grad_norm": 0.150390625, + "learning_rate": 0.0019474259992535632, + "loss": 0.125, + "step": 11388 + }, + { + "epoch": 0.09886198904523398, + "grad_norm": 0.107421875, + "learning_rate": 0.001947415975479851, + "loss": 0.1123, + "step": 11389 + }, + { + "epoch": 0.09887066952543815, + "grad_norm": 0.30859375, + "learning_rate": 0.0019474059507794176, + "loss": 0.2207, + "step": 11390 + }, + { + "epoch": 0.09887935000564231, + "grad_norm": 0.546875, + "learning_rate": 0.001947395925152274, + "loss": 0.1128, + "step": 11391 + }, + { + "epoch": 0.09888803048584648, + "grad_norm": 0.123046875, + "learning_rate": 0.0019473858985984307, + "loss": 0.1045, + "step": 11392 + }, + { + "epoch": 0.09889671096605064, + "grad_norm": 0.166015625, + "learning_rate": 0.0019473758711178988, + "loss": 0.1064, + "step": 11393 + }, + { + "epoch": 0.0989053914462548, + "grad_norm": 0.31640625, + "learning_rate": 0.0019473658427106896, + "loss": 0.123, + "step": 11394 + }, + { + "epoch": 0.09891407192645897, + "grad_norm": 0.251953125, + "learning_rate": 0.0019473558133768136, + "loss": 0.1172, + "step": 11395 + }, + { + "epoch": 0.09892275240666314, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019473457831162822, + "loss": 0.0996, + "step": 11396 + }, + { + "epoch": 0.0989314328868673, + "grad_norm": 0.6953125, + "learning_rate": 0.0019473357519291061, + "loss": 0.0996, + "step": 11397 + }, + { + "epoch": 0.09894011336707147, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019473257198152965, + "loss": 0.1416, + "step": 11398 + }, + { + "epoch": 0.09894879384727563, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019473156867748642, + "loss": 0.1152, + "step": 11399 + }, + { + "epoch": 0.0989574743274798, + "grad_norm": 0.150390625, + "learning_rate": 0.00194730565280782, + "loss": 0.123, + "step": 11400 + }, + { + "epoch": 0.09896615480768396, + "grad_norm": 0.298828125, + "learning_rate": 0.0019472956179141753, + "loss": 0.1045, + "step": 11401 + }, + { + "epoch": 0.09897483528788813, + "grad_norm": 0.515625, + "learning_rate": 0.0019472855820939405, + "loss": 0.1182, + "step": 11402 + }, + { + "epoch": 0.09898351576809229, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019472755453471273, + "loss": 0.1006, + "step": 11403 + }, + { + "epoch": 0.09899219624829646, + "grad_norm": 0.16796875, + "learning_rate": 0.0019472655076737462, + "loss": 0.0894, + "step": 11404 + }, + { + "epoch": 0.09900087672850062, + "grad_norm": 0.43359375, + "learning_rate": 0.0019472554690738084, + "loss": 0.1016, + "step": 11405 + }, + { + "epoch": 0.09900955720870479, + "grad_norm": 0.158203125, + "learning_rate": 0.0019472454295473246, + "loss": 0.1406, + "step": 11406 + }, + { + "epoch": 0.09901823768890895, + "grad_norm": 0.408203125, + "learning_rate": 0.0019472353890943057, + "loss": 0.1182, + "step": 11407 + }, + { + "epoch": 0.09902691816911312, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019472253477147631, + "loss": 0.1279, + "step": 11408 + }, + { + "epoch": 0.09903559864931728, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019472153054087076, + "loss": 0.1338, + "step": 11409 + }, + { + "epoch": 0.09904427912952145, + "grad_norm": 0.294921875, + "learning_rate": 0.0019472052621761502, + "loss": 0.1553, + "step": 11410 + }, + { + "epoch": 0.09905295960972561, + "grad_norm": 0.25390625, + "learning_rate": 0.0019471952180171019, + "loss": 0.1553, + "step": 11411 + }, + { + "epoch": 0.09906164008992978, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019471851729315737, + "loss": 0.1055, + "step": 11412 + }, + { + "epoch": 0.09907032057013394, + "grad_norm": 0.369140625, + "learning_rate": 0.0019471751269195762, + "loss": 0.0986, + "step": 11413 + }, + { + "epoch": 0.09907900105033811, + "grad_norm": 0.3203125, + "learning_rate": 0.001947165079981121, + "loss": 0.1299, + "step": 11414 + }, + { + "epoch": 0.09908768153054227, + "grad_norm": 0.400390625, + "learning_rate": 0.0019471550321162186, + "loss": 0.1118, + "step": 11415 + }, + { + "epoch": 0.09909636201074644, + "grad_norm": 0.09765625, + "learning_rate": 0.0019471449833248801, + "loss": 0.1201, + "step": 11416 + }, + { + "epoch": 0.0991050424909506, + "grad_norm": 0.26171875, + "learning_rate": 0.001947134933607117, + "loss": 0.1201, + "step": 11417 + }, + { + "epoch": 0.09911372297115477, + "grad_norm": 0.515625, + "learning_rate": 0.0019471248829629397, + "loss": 0.1406, + "step": 11418 + }, + { + "epoch": 0.09912240345135893, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019471148313923593, + "loss": 0.0933, + "step": 11419 + }, + { + "epoch": 0.0991310839315631, + "grad_norm": 0.421875, + "learning_rate": 0.0019471047788953866, + "loss": 0.1045, + "step": 11420 + }, + { + "epoch": 0.09913976441176726, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019470947254720333, + "loss": 0.1123, + "step": 11421 + }, + { + "epoch": 0.09914844489197143, + "grad_norm": 0.142578125, + "learning_rate": 0.0019470846711223096, + "loss": 0.1377, + "step": 11422 + }, + { + "epoch": 0.0991571253721756, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019470746158462271, + "loss": 0.1211, + "step": 11423 + }, + { + "epoch": 0.09916580585237976, + "grad_norm": 0.44921875, + "learning_rate": 0.0019470645596437965, + "loss": 0.0938, + "step": 11424 + }, + { + "epoch": 0.09917448633258392, + "grad_norm": 0.2109375, + "learning_rate": 0.0019470545025150286, + "loss": 0.0879, + "step": 11425 + }, + { + "epoch": 0.09918316681278809, + "grad_norm": 0.115234375, + "learning_rate": 0.0019470444444599346, + "loss": 0.1416, + "step": 11426 + }, + { + "epoch": 0.09919184729299226, + "grad_norm": 0.45703125, + "learning_rate": 0.0019470343854785254, + "loss": 0.0776, + "step": 11427 + }, + { + "epoch": 0.0992005277731964, + "grad_norm": 0.46875, + "learning_rate": 0.0019470243255708124, + "loss": 0.0977, + "step": 11428 + }, + { + "epoch": 0.09920920825340057, + "grad_norm": 0.154296875, + "learning_rate": 0.0019470142647368062, + "loss": 0.1211, + "step": 11429 + }, + { + "epoch": 0.09921788873360474, + "grad_norm": 0.93359375, + "learning_rate": 0.0019470042029765182, + "loss": 0.1455, + "step": 11430 + }, + { + "epoch": 0.0992265692138089, + "grad_norm": 0.76171875, + "learning_rate": 0.001946994140289959, + "loss": 0.1387, + "step": 11431 + }, + { + "epoch": 0.09923524969401307, + "grad_norm": 0.625, + "learning_rate": 0.0019469840766771394, + "loss": 0.1406, + "step": 11432 + }, + { + "epoch": 0.09924393017421723, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019469740121380711, + "loss": 0.1143, + "step": 11433 + }, + { + "epoch": 0.0992526106544214, + "grad_norm": 0.2734375, + "learning_rate": 0.0019469639466727646, + "loss": 0.124, + "step": 11434 + }, + { + "epoch": 0.09926129113462556, + "grad_norm": 0.494140625, + "learning_rate": 0.0019469538802812313, + "loss": 0.1299, + "step": 11435 + }, + { + "epoch": 0.09926997161482973, + "grad_norm": 0.271484375, + "learning_rate": 0.0019469438129634817, + "loss": 0.1318, + "step": 11436 + }, + { + "epoch": 0.09927865209503389, + "grad_norm": 0.625, + "learning_rate": 0.0019469337447195272, + "loss": 0.0898, + "step": 11437 + }, + { + "epoch": 0.09928733257523806, + "grad_norm": 0.5546875, + "learning_rate": 0.0019469236755493785, + "loss": 0.1484, + "step": 11438 + }, + { + "epoch": 0.09929601305544222, + "grad_norm": 0.2099609375, + "learning_rate": 0.001946913605453047, + "loss": 0.1299, + "step": 11439 + }, + { + "epoch": 0.09930469353564639, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019469035344305433, + "loss": 0.127, + "step": 11440 + }, + { + "epoch": 0.09931337401585055, + "grad_norm": 0.796875, + "learning_rate": 0.0019468934624818789, + "loss": 0.1201, + "step": 11441 + }, + { + "epoch": 0.09932205449605472, + "grad_norm": 0.412109375, + "learning_rate": 0.0019468833896070644, + "loss": 0.1318, + "step": 11442 + }, + { + "epoch": 0.09933073497625888, + "grad_norm": 0.3984375, + "learning_rate": 0.0019468733158061107, + "loss": 0.1543, + "step": 11443 + }, + { + "epoch": 0.09933941545646305, + "grad_norm": 0.68359375, + "learning_rate": 0.0019468632410790294, + "loss": 0.0991, + "step": 11444 + }, + { + "epoch": 0.09934809593666721, + "grad_norm": 0.1142578125, + "learning_rate": 0.001946853165425831, + "loss": 0.1426, + "step": 11445 + }, + { + "epoch": 0.09935677641687138, + "grad_norm": 0.298828125, + "learning_rate": 0.0019468430888465269, + "loss": 0.1143, + "step": 11446 + }, + { + "epoch": 0.09936545689707554, + "grad_norm": 0.625, + "learning_rate": 0.0019468330113411277, + "loss": 0.1455, + "step": 11447 + }, + { + "epoch": 0.09937413737727971, + "grad_norm": 1.265625, + "learning_rate": 0.001946822932909645, + "loss": 0.1143, + "step": 11448 + }, + { + "epoch": 0.09938281785748387, + "grad_norm": 0.12353515625, + "learning_rate": 0.001946812853552089, + "loss": 0.1162, + "step": 11449 + }, + { + "epoch": 0.09939149833768804, + "grad_norm": 0.35546875, + "learning_rate": 0.0019468027732684713, + "loss": 0.1582, + "step": 11450 + }, + { + "epoch": 0.0994001788178922, + "grad_norm": 0.51171875, + "learning_rate": 0.001946792692058803, + "loss": 0.0991, + "step": 11451 + }, + { + "epoch": 0.09940885929809637, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019467826099230947, + "loss": 0.125, + "step": 11452 + }, + { + "epoch": 0.09941753977830053, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019467725268613582, + "loss": 0.103, + "step": 11453 + }, + { + "epoch": 0.0994262202585047, + "grad_norm": 0.19921875, + "learning_rate": 0.0019467624428736033, + "loss": 0.1328, + "step": 11454 + }, + { + "epoch": 0.09943490073870886, + "grad_norm": 0.30859375, + "learning_rate": 0.0019467523579598418, + "loss": 0.1445, + "step": 11455 + }, + { + "epoch": 0.09944358121891303, + "grad_norm": 0.259765625, + "learning_rate": 0.001946742272120085, + "loss": 0.166, + "step": 11456 + }, + { + "epoch": 0.0994522616991172, + "grad_norm": 0.75, + "learning_rate": 0.0019467321853543435, + "loss": 0.1582, + "step": 11457 + }, + { + "epoch": 0.09946094217932136, + "grad_norm": 0.197265625, + "learning_rate": 0.0019467220976626283, + "loss": 0.106, + "step": 11458 + }, + { + "epoch": 0.09946962265952553, + "grad_norm": 1.3828125, + "learning_rate": 0.0019467120090449505, + "loss": 0.1055, + "step": 11459 + }, + { + "epoch": 0.09947830313972969, + "grad_norm": 0.216796875, + "learning_rate": 0.0019467019195013213, + "loss": 0.1191, + "step": 11460 + }, + { + "epoch": 0.09948698361993386, + "grad_norm": 0.35546875, + "learning_rate": 0.0019466918290317514, + "loss": 0.1162, + "step": 11461 + }, + { + "epoch": 0.09949566410013802, + "grad_norm": 1.1796875, + "learning_rate": 0.0019466817376362522, + "loss": 0.1494, + "step": 11462 + }, + { + "epoch": 0.09950434458034219, + "grad_norm": 0.95703125, + "learning_rate": 0.0019466716453148346, + "loss": 0.1426, + "step": 11463 + }, + { + "epoch": 0.09951302506054635, + "grad_norm": 0.267578125, + "learning_rate": 0.0019466615520675097, + "loss": 0.1025, + "step": 11464 + }, + { + "epoch": 0.09952170554075052, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019466514578942882, + "loss": 0.1016, + "step": 11465 + }, + { + "epoch": 0.09953038602095468, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019466413627951813, + "loss": 0.0967, + "step": 11466 + }, + { + "epoch": 0.09953906650115885, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019466312667702004, + "loss": 0.1245, + "step": 11467 + }, + { + "epoch": 0.09954774698136301, + "grad_norm": 0.58203125, + "learning_rate": 0.0019466211698193564, + "loss": 0.1289, + "step": 11468 + }, + { + "epoch": 0.09955642746156718, + "grad_norm": 0.84765625, + "learning_rate": 0.0019466110719426597, + "loss": 0.1221, + "step": 11469 + }, + { + "epoch": 0.09956510794177134, + "grad_norm": 0.5078125, + "learning_rate": 0.0019466009731401222, + "loss": 0.1104, + "step": 11470 + }, + { + "epoch": 0.0995737884219755, + "grad_norm": 0.328125, + "learning_rate": 0.0019465908734117544, + "loss": 0.1611, + "step": 11471 + }, + { + "epoch": 0.09958246890217967, + "grad_norm": 0.306640625, + "learning_rate": 0.0019465807727575676, + "loss": 0.1455, + "step": 11472 + }, + { + "epoch": 0.09959114938238384, + "grad_norm": 0.71875, + "learning_rate": 0.001946570671177573, + "loss": 0.1738, + "step": 11473 + }, + { + "epoch": 0.099599829862588, + "grad_norm": 0.4921875, + "learning_rate": 0.0019465605686717813, + "loss": 0.1377, + "step": 11474 + }, + { + "epoch": 0.09960851034279217, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019465504652402038, + "loss": 0.1162, + "step": 11475 + }, + { + "epoch": 0.09961719082299633, + "grad_norm": 0.171875, + "learning_rate": 0.0019465403608828512, + "loss": 0.0898, + "step": 11476 + }, + { + "epoch": 0.0996258713032005, + "grad_norm": 1.1953125, + "learning_rate": 0.0019465302555997349, + "loss": 0.127, + "step": 11477 + }, + { + "epoch": 0.09963455178340466, + "grad_norm": 0.62890625, + "learning_rate": 0.001946520149390866, + "loss": 0.0996, + "step": 11478 + }, + { + "epoch": 0.09964323226360883, + "grad_norm": 0.796875, + "learning_rate": 0.001946510042256255, + "loss": 0.126, + "step": 11479 + }, + { + "epoch": 0.09965191274381299, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019464999341959135, + "loss": 0.1426, + "step": 11480 + }, + { + "epoch": 0.09966059322401716, + "grad_norm": 0.51171875, + "learning_rate": 0.0019464898252098523, + "loss": 0.1387, + "step": 11481 + }, + { + "epoch": 0.09966927370422132, + "grad_norm": 0.064453125, + "learning_rate": 0.0019464797152980828, + "loss": 0.1113, + "step": 11482 + }, + { + "epoch": 0.09967795418442549, + "grad_norm": 0.50390625, + "learning_rate": 0.0019464696044606158, + "loss": 0.1055, + "step": 11483 + }, + { + "epoch": 0.09968663466462965, + "grad_norm": 0.474609375, + "learning_rate": 0.0019464594926974622, + "loss": 0.1172, + "step": 11484 + }, + { + "epoch": 0.09969531514483382, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019464493800086331, + "loss": 0.127, + "step": 11485 + }, + { + "epoch": 0.09970399562503798, + "grad_norm": 0.306640625, + "learning_rate": 0.00194643926639414, + "loss": 0.1309, + "step": 11486 + }, + { + "epoch": 0.09971267610524215, + "grad_norm": 0.45703125, + "learning_rate": 0.0019464291518539933, + "loss": 0.0762, + "step": 11487 + }, + { + "epoch": 0.09972135658544631, + "grad_norm": 0.318359375, + "learning_rate": 0.0019464190363882047, + "loss": 0.1123, + "step": 11488 + }, + { + "epoch": 0.09973003706565048, + "grad_norm": 0.55078125, + "learning_rate": 0.0019464089199967846, + "loss": 0.0981, + "step": 11489 + }, + { + "epoch": 0.09973871754585463, + "grad_norm": 0.416015625, + "learning_rate": 0.0019463988026797447, + "loss": 0.1143, + "step": 11490 + }, + { + "epoch": 0.0997473980260588, + "grad_norm": 0.734375, + "learning_rate": 0.001946388684437096, + "loss": 0.1553, + "step": 11491 + }, + { + "epoch": 0.09975607850626296, + "grad_norm": 0.267578125, + "learning_rate": 0.001946378565268849, + "loss": 0.1211, + "step": 11492 + }, + { + "epoch": 0.09976475898646713, + "grad_norm": 0.255859375, + "learning_rate": 0.001946368445175015, + "loss": 0.1602, + "step": 11493 + }, + { + "epoch": 0.09977343946667129, + "grad_norm": 0.59375, + "learning_rate": 0.0019463583241556056, + "loss": 0.1396, + "step": 11494 + }, + { + "epoch": 0.09978211994687546, + "grad_norm": 0.5859375, + "learning_rate": 0.0019463482022106311, + "loss": 0.1289, + "step": 11495 + }, + { + "epoch": 0.09979080042707962, + "grad_norm": 0.287109375, + "learning_rate": 0.001946338079340103, + "loss": 0.106, + "step": 11496 + }, + { + "epoch": 0.09979948090728379, + "grad_norm": 0.66796875, + "learning_rate": 0.0019463279555440324, + "loss": 0.1074, + "step": 11497 + }, + { + "epoch": 0.09980816138748795, + "grad_norm": 0.2421875, + "learning_rate": 0.00194631783082243, + "loss": 0.1436, + "step": 11498 + }, + { + "epoch": 0.09981684186769212, + "grad_norm": 0.53125, + "learning_rate": 0.0019463077051753077, + "loss": 0.1631, + "step": 11499 + }, + { + "epoch": 0.09982552234789628, + "grad_norm": 0.166015625, + "learning_rate": 0.0019462975786026756, + "loss": 0.1719, + "step": 11500 + }, + { + "epoch": 0.09983420282810045, + "grad_norm": 0.11962890625, + "learning_rate": 0.001946287451104545, + "loss": 0.0859, + "step": 11501 + }, + { + "epoch": 0.09984288330830461, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019462773226809277, + "loss": 0.0981, + "step": 11502 + }, + { + "epoch": 0.09985156378850878, + "grad_norm": 0.6875, + "learning_rate": 0.0019462671933318338, + "loss": 0.1123, + "step": 11503 + }, + { + "epoch": 0.09986024426871294, + "grad_norm": 0.484375, + "learning_rate": 0.0019462570630572752, + "loss": 0.0952, + "step": 11504 + }, + { + "epoch": 0.0998689247489171, + "grad_norm": 0.208984375, + "learning_rate": 0.0019462469318572621, + "loss": 0.0835, + "step": 11505 + }, + { + "epoch": 0.09987760522912127, + "grad_norm": 0.216796875, + "learning_rate": 0.0019462367997318062, + "loss": 0.1162, + "step": 11506 + }, + { + "epoch": 0.09988628570932544, + "grad_norm": 0.126953125, + "learning_rate": 0.0019462266666809187, + "loss": 0.1289, + "step": 11507 + }, + { + "epoch": 0.0998949661895296, + "grad_norm": 0.76171875, + "learning_rate": 0.0019462165327046102, + "loss": 0.1113, + "step": 11508 + }, + { + "epoch": 0.09990364666973377, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019462063978028922, + "loss": 0.1064, + "step": 11509 + }, + { + "epoch": 0.09991232714993793, + "grad_norm": 0.36328125, + "learning_rate": 0.0019461962619757756, + "loss": 0.1191, + "step": 11510 + }, + { + "epoch": 0.0999210076301421, + "grad_norm": 0.18359375, + "learning_rate": 0.0019461861252232713, + "loss": 0.248, + "step": 11511 + }, + { + "epoch": 0.09992968811034626, + "grad_norm": 0.1875, + "learning_rate": 0.0019461759875453906, + "loss": 0.0977, + "step": 11512 + }, + { + "epoch": 0.09993836859055043, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019461658489421446, + "loss": 0.084, + "step": 11513 + }, + { + "epoch": 0.09994704907075459, + "grad_norm": 0.35546875, + "learning_rate": 0.0019461557094135444, + "loss": 0.166, + "step": 11514 + }, + { + "epoch": 0.09995572955095876, + "grad_norm": 0.58203125, + "learning_rate": 0.001946145568959601, + "loss": 0.1338, + "step": 11515 + }, + { + "epoch": 0.09996441003116292, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019461354275803254, + "loss": 0.1064, + "step": 11516 + }, + { + "epoch": 0.09997309051136709, + "grad_norm": 0.859375, + "learning_rate": 0.001946125285275729, + "loss": 0.1455, + "step": 11517 + }, + { + "epoch": 0.09998177099157125, + "grad_norm": 1.078125, + "learning_rate": 0.0019461151420458228, + "loss": 0.1006, + "step": 11518 + }, + { + "epoch": 0.09999045147177542, + "grad_norm": 0.14453125, + "learning_rate": 0.001946104997890617, + "loss": 0.1089, + "step": 11519 + }, + { + "epoch": 0.09999913195197958, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019460948528101244, + "loss": 0.1187, + "step": 11520 + }, + { + "epoch": 0.10000781243218375, + "grad_norm": 0.1943359375, + "learning_rate": 0.0019460847068043547, + "loss": 0.1309, + "step": 11521 + }, + { + "epoch": 0.10001649291238791, + "grad_norm": 0.482421875, + "learning_rate": 0.00194607455987332, + "loss": 0.1328, + "step": 11522 + }, + { + "epoch": 0.10002517339259208, + "grad_norm": 0.18359375, + "learning_rate": 0.0019460644120170305, + "loss": 0.1475, + "step": 11523 + }, + { + "epoch": 0.10003385387279624, + "grad_norm": 0.400390625, + "learning_rate": 0.0019460542632354976, + "loss": 0.0957, + "step": 11524 + }, + { + "epoch": 0.10004253435300041, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019460441135287325, + "loss": 0.1328, + "step": 11525 + }, + { + "epoch": 0.10005121483320457, + "grad_norm": 0.58203125, + "learning_rate": 0.0019460339628967463, + "loss": 0.1064, + "step": 11526 + }, + { + "epoch": 0.10005989531340874, + "grad_norm": 0.10693359375, + "learning_rate": 0.00194602381133955, + "loss": 0.1299, + "step": 11527 + }, + { + "epoch": 0.1000685757936129, + "grad_norm": 0.7734375, + "learning_rate": 0.001946013658857155, + "loss": 0.0967, + "step": 11528 + }, + { + "epoch": 0.10007725627381707, + "grad_norm": 0.296875, + "learning_rate": 0.0019460035054495721, + "loss": 0.084, + "step": 11529 + }, + { + "epoch": 0.10008593675402123, + "grad_norm": 0.6015625, + "learning_rate": 0.0019459933511168125, + "loss": 0.124, + "step": 11530 + }, + { + "epoch": 0.1000946172342254, + "grad_norm": 1.1171875, + "learning_rate": 0.001945983195858887, + "loss": 0.2207, + "step": 11531 + }, + { + "epoch": 0.10010329771442956, + "grad_norm": 0.83984375, + "learning_rate": 0.0019459730396758073, + "loss": 0.1191, + "step": 11532 + }, + { + "epoch": 0.10011197819463373, + "grad_norm": 0.39453125, + "learning_rate": 0.001945962882567584, + "loss": 0.0986, + "step": 11533 + }, + { + "epoch": 0.1001206586748379, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019459527245342288, + "loss": 0.1064, + "step": 11534 + }, + { + "epoch": 0.10012933915504206, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019459425655757518, + "loss": 0.1562, + "step": 11535 + }, + { + "epoch": 0.10013801963524623, + "grad_norm": 0.7578125, + "learning_rate": 0.001945932405692165, + "loss": 0.1367, + "step": 11536 + }, + { + "epoch": 0.10014670011545039, + "grad_norm": 0.14453125, + "learning_rate": 0.0019459222448834795, + "loss": 0.1074, + "step": 11537 + }, + { + "epoch": 0.10015538059565456, + "grad_norm": 0.40625, + "learning_rate": 0.0019459120831497056, + "loss": 0.1172, + "step": 11538 + }, + { + "epoch": 0.10016406107585872, + "grad_norm": 0.408203125, + "learning_rate": 0.0019459019204908553, + "loss": 0.1113, + "step": 11539 + }, + { + "epoch": 0.10017274155606289, + "grad_norm": 5.46875, + "learning_rate": 0.0019458917569069394, + "loss": 0.2012, + "step": 11540 + }, + { + "epoch": 0.10018142203626705, + "grad_norm": 0.1005859375, + "learning_rate": 0.001945881592397969, + "loss": 0.1787, + "step": 11541 + }, + { + "epoch": 0.10019010251647122, + "grad_norm": 0.703125, + "learning_rate": 0.001945871426963955, + "loss": 0.0967, + "step": 11542 + }, + { + "epoch": 0.10019878299667538, + "grad_norm": 0.287109375, + "learning_rate": 0.0019458612606049086, + "loss": 0.1406, + "step": 11543 + }, + { + "epoch": 0.10020746347687955, + "grad_norm": 0.703125, + "learning_rate": 0.0019458510933208415, + "loss": 0.1348, + "step": 11544 + }, + { + "epoch": 0.10021614395708371, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019458409251117642, + "loss": 0.123, + "step": 11545 + }, + { + "epoch": 0.10022482443728788, + "grad_norm": 0.345703125, + "learning_rate": 0.0019458307559776877, + "loss": 0.1143, + "step": 11546 + }, + { + "epoch": 0.10023350491749204, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019458205859186235, + "loss": 0.0947, + "step": 11547 + }, + { + "epoch": 0.1002421853976962, + "grad_norm": 0.474609375, + "learning_rate": 0.0019458104149345829, + "loss": 0.1602, + "step": 11548 + }, + { + "epoch": 0.10025086587790037, + "grad_norm": 0.54296875, + "learning_rate": 0.0019458002430255764, + "loss": 0.165, + "step": 11549 + }, + { + "epoch": 0.10025954635810454, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019457900701916155, + "loss": 0.1719, + "step": 11550 + }, + { + "epoch": 0.10026822683830869, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019457798964327114, + "loss": 0.1396, + "step": 11551 + }, + { + "epoch": 0.10027690731851285, + "grad_norm": 0.66796875, + "learning_rate": 0.0019457697217488753, + "loss": 0.0918, + "step": 11552 + }, + { + "epoch": 0.10028558779871702, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019457595461401178, + "loss": 0.1094, + "step": 11553 + }, + { + "epoch": 0.10029426827892118, + "grad_norm": 1.15625, + "learning_rate": 0.0019457493696064506, + "loss": 0.1187, + "step": 11554 + }, + { + "epoch": 0.10030294875912535, + "grad_norm": 0.38671875, + "learning_rate": 0.0019457391921478841, + "loss": 0.0918, + "step": 11555 + }, + { + "epoch": 0.10031162923932951, + "grad_norm": 0.3046875, + "learning_rate": 0.0019457290137644306, + "loss": 0.1455, + "step": 11556 + }, + { + "epoch": 0.10032030971953368, + "grad_norm": 0.388671875, + "learning_rate": 0.0019457188344561003, + "loss": 0.1143, + "step": 11557 + }, + { + "epoch": 0.10032899019973784, + "grad_norm": 0.7734375, + "learning_rate": 0.0019457086542229048, + "loss": 0.1245, + "step": 11558 + }, + { + "epoch": 0.10033767067994201, + "grad_norm": 0.44140625, + "learning_rate": 0.0019456984730648547, + "loss": 0.1328, + "step": 11559 + }, + { + "epoch": 0.10034635116014617, + "grad_norm": 0.24609375, + "learning_rate": 0.0019456882909819613, + "loss": 0.1133, + "step": 11560 + }, + { + "epoch": 0.10035503164035034, + "grad_norm": 0.37890625, + "learning_rate": 0.0019456781079742363, + "loss": 0.0898, + "step": 11561 + }, + { + "epoch": 0.1003637121205545, + "grad_norm": 0.5234375, + "learning_rate": 0.00194566792404169, + "loss": 0.1289, + "step": 11562 + }, + { + "epoch": 0.10037239260075867, + "grad_norm": 0.72265625, + "learning_rate": 0.0019456577391843343, + "loss": 0.1348, + "step": 11563 + }, + { + "epoch": 0.10038107308096283, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019456475534021802, + "loss": 0.1719, + "step": 11564 + }, + { + "epoch": 0.100389753561167, + "grad_norm": 0.66015625, + "learning_rate": 0.0019456373666952383, + "loss": 0.0859, + "step": 11565 + }, + { + "epoch": 0.10039843404137117, + "grad_norm": 0.10595703125, + "learning_rate": 0.00194562717906352, + "loss": 0.125, + "step": 11566 + }, + { + "epoch": 0.10040711452157533, + "grad_norm": 0.9296875, + "learning_rate": 0.0019456169905070367, + "loss": 0.1143, + "step": 11567 + }, + { + "epoch": 0.1004157950017795, + "grad_norm": 0.1328125, + "learning_rate": 0.0019456068010257994, + "loss": 0.1279, + "step": 11568 + }, + { + "epoch": 0.10042447548198366, + "grad_norm": 0.451171875, + "learning_rate": 0.0019455966106198192, + "loss": 0.0957, + "step": 11569 + }, + { + "epoch": 0.10043315596218783, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019455864192891071, + "loss": 0.125, + "step": 11570 + }, + { + "epoch": 0.10044183644239199, + "grad_norm": 0.703125, + "learning_rate": 0.0019455762270336747, + "loss": 0.1465, + "step": 11571 + }, + { + "epoch": 0.10045051692259616, + "grad_norm": 0.984375, + "learning_rate": 0.0019455660338535325, + "loss": 0.1094, + "step": 11572 + }, + { + "epoch": 0.10045919740280032, + "grad_norm": 0.10791015625, + "learning_rate": 0.001945555839748692, + "loss": 0.1006, + "step": 11573 + }, + { + "epoch": 0.10046787788300449, + "grad_norm": 0.65625, + "learning_rate": 0.0019455456447191647, + "loss": 0.1572, + "step": 11574 + }, + { + "epoch": 0.10047655836320865, + "grad_norm": 0.10595703125, + "learning_rate": 0.001945535448764961, + "loss": 0.0986, + "step": 11575 + }, + { + "epoch": 0.10048523884341282, + "grad_norm": 0.080078125, + "learning_rate": 0.0019455252518860927, + "loss": 0.084, + "step": 11576 + }, + { + "epoch": 0.10049391932361698, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019455150540825703, + "loss": 0.1611, + "step": 11577 + }, + { + "epoch": 0.10050259980382115, + "grad_norm": 0.28125, + "learning_rate": 0.0019455048553544058, + "loss": 0.1201, + "step": 11578 + }, + { + "epoch": 0.10051128028402531, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019454946557016098, + "loss": 0.1045, + "step": 11579 + }, + { + "epoch": 0.10051996076422948, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019454844551241933, + "loss": 0.0942, + "step": 11580 + }, + { + "epoch": 0.10052864124443364, + "grad_norm": 0.201171875, + "learning_rate": 0.001945474253622168, + "loss": 0.1465, + "step": 11581 + }, + { + "epoch": 0.10053732172463781, + "grad_norm": 0.0859375, + "learning_rate": 0.0019454640511955446, + "loss": 0.1123, + "step": 11582 + }, + { + "epoch": 0.10054600220484197, + "grad_norm": 0.7421875, + "learning_rate": 0.0019454538478443344, + "loss": 0.1182, + "step": 11583 + }, + { + "epoch": 0.10055468268504614, + "grad_norm": 0.390625, + "learning_rate": 0.0019454436435685488, + "loss": 0.1709, + "step": 11584 + }, + { + "epoch": 0.1005633631652503, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019454334383681984, + "loss": 0.124, + "step": 11585 + }, + { + "epoch": 0.10057204364545447, + "grad_norm": 0.46875, + "learning_rate": 0.0019454232322432949, + "loss": 0.0986, + "step": 11586 + }, + { + "epoch": 0.10058072412565863, + "grad_norm": 0.216796875, + "learning_rate": 0.0019454130251938492, + "loss": 0.1147, + "step": 11587 + }, + { + "epoch": 0.1005894046058628, + "grad_norm": 0.26953125, + "learning_rate": 0.0019454028172198727, + "loss": 0.1104, + "step": 11588 + }, + { + "epoch": 0.10059808508606696, + "grad_norm": 0.82421875, + "learning_rate": 0.001945392608321376, + "loss": 0.2051, + "step": 11589 + }, + { + "epoch": 0.10060676556627113, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019453823984983706, + "loss": 0.1289, + "step": 11590 + }, + { + "epoch": 0.1006154460464753, + "grad_norm": 0.275390625, + "learning_rate": 0.0019453721877508684, + "loss": 0.1143, + "step": 11591 + }, + { + "epoch": 0.10062412652667946, + "grad_norm": 0.1484375, + "learning_rate": 0.0019453619760788792, + "loss": 0.1143, + "step": 11592 + }, + { + "epoch": 0.10063280700688362, + "grad_norm": 0.0927734375, + "learning_rate": 0.001945351763482415, + "loss": 0.1162, + "step": 11593 + }, + { + "epoch": 0.10064148748708779, + "grad_norm": 0.796875, + "learning_rate": 0.0019453415499614871, + "loss": 0.1118, + "step": 11594 + }, + { + "epoch": 0.10065016796729195, + "grad_norm": 0.376953125, + "learning_rate": 0.0019453313355161062, + "loss": 0.1211, + "step": 11595 + }, + { + "epoch": 0.10065884844749612, + "grad_norm": 0.314453125, + "learning_rate": 0.0019453211201462836, + "loss": 0.1318, + "step": 11596 + }, + { + "epoch": 0.10066752892770028, + "grad_norm": 0.36328125, + "learning_rate": 0.0019453109038520306, + "loss": 0.1118, + "step": 11597 + }, + { + "epoch": 0.10067620940790445, + "grad_norm": 0.1640625, + "learning_rate": 0.0019453006866333583, + "loss": 0.0908, + "step": 11598 + }, + { + "epoch": 0.10068488988810861, + "grad_norm": 0.54296875, + "learning_rate": 0.0019452904684902775, + "loss": 0.1455, + "step": 11599 + }, + { + "epoch": 0.10069357036831278, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019452802494228, + "loss": 0.1172, + "step": 11600 + }, + { + "epoch": 0.10070225084851694, + "grad_norm": 0.6875, + "learning_rate": 0.0019452700294309369, + "loss": 0.1621, + "step": 11601 + }, + { + "epoch": 0.10071093132872111, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019452598085146992, + "loss": 0.1216, + "step": 11602 + }, + { + "epoch": 0.10071961180892527, + "grad_norm": 0.36328125, + "learning_rate": 0.001945249586674098, + "loss": 0.1123, + "step": 11603 + }, + { + "epoch": 0.10072829228912944, + "grad_norm": 0.279296875, + "learning_rate": 0.0019452393639091444, + "loss": 0.1221, + "step": 11604 + }, + { + "epoch": 0.1007369727693336, + "grad_norm": 0.271484375, + "learning_rate": 0.0019452291402198498, + "loss": 0.1221, + "step": 11605 + }, + { + "epoch": 0.10074565324953777, + "grad_norm": 0.08203125, + "learning_rate": 0.0019452189156062252, + "loss": 0.1191, + "step": 11606 + }, + { + "epoch": 0.10075433372974194, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019452086900682822, + "loss": 0.1094, + "step": 11607 + }, + { + "epoch": 0.1007630142099461, + "grad_norm": 0.384765625, + "learning_rate": 0.0019451984636060316, + "loss": 0.082, + "step": 11608 + }, + { + "epoch": 0.10077169469015027, + "grad_norm": 0.73828125, + "learning_rate": 0.0019451882362194845, + "loss": 0.2031, + "step": 11609 + }, + { + "epoch": 0.10078037517035443, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019451780079086523, + "loss": 0.1289, + "step": 11610 + }, + { + "epoch": 0.1007890556505586, + "grad_norm": 0.408203125, + "learning_rate": 0.0019451677786735462, + "loss": 0.125, + "step": 11611 + }, + { + "epoch": 0.10079773613076276, + "grad_norm": 0.734375, + "learning_rate": 0.0019451575485141773, + "loss": 0.1338, + "step": 11612 + }, + { + "epoch": 0.10080641661096691, + "grad_norm": 0.51171875, + "learning_rate": 0.0019451473174305568, + "loss": 0.0869, + "step": 11613 + }, + { + "epoch": 0.10081509709117108, + "grad_norm": 0.271484375, + "learning_rate": 0.0019451370854226958, + "loss": 0.1035, + "step": 11614 + }, + { + "epoch": 0.10082377757137524, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019451268524906056, + "loss": 0.1143, + "step": 11615 + }, + { + "epoch": 0.10083245805157941, + "grad_norm": 0.609375, + "learning_rate": 0.0019451166186342974, + "loss": 0.1445, + "step": 11616 + }, + { + "epoch": 0.10084113853178357, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019451063838537826, + "loss": 0.1221, + "step": 11617 + }, + { + "epoch": 0.10084981901198774, + "grad_norm": 0.56640625, + "learning_rate": 0.0019450961481490717, + "loss": 0.1143, + "step": 11618 + }, + { + "epoch": 0.1008584994921919, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019450859115201764, + "loss": 0.1484, + "step": 11619 + }, + { + "epoch": 0.10086717997239607, + "grad_norm": 0.146484375, + "learning_rate": 0.0019450756739671084, + "loss": 0.0913, + "step": 11620 + }, + { + "epoch": 0.10087586045260023, + "grad_norm": 0.4765625, + "learning_rate": 0.0019450654354898778, + "loss": 0.0835, + "step": 11621 + }, + { + "epoch": 0.1008845409328044, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019450551960884964, + "loss": 0.1309, + "step": 11622 + }, + { + "epoch": 0.10089322141300856, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019450449557629754, + "loss": 0.1436, + "step": 11623 + }, + { + "epoch": 0.10090190189321273, + "grad_norm": 0.35546875, + "learning_rate": 0.0019450347145133263, + "loss": 0.0825, + "step": 11624 + }, + { + "epoch": 0.1009105823734169, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019450244723395595, + "loss": 0.1367, + "step": 11625 + }, + { + "epoch": 0.10091926285362106, + "grad_norm": 0.4453125, + "learning_rate": 0.0019450142292416867, + "loss": 0.0957, + "step": 11626 + }, + { + "epoch": 0.10092794333382522, + "grad_norm": 0.3671875, + "learning_rate": 0.0019450039852197188, + "loss": 0.1309, + "step": 11627 + }, + { + "epoch": 0.10093662381402939, + "grad_norm": 0.158203125, + "learning_rate": 0.0019449937402736678, + "loss": 0.1377, + "step": 11628 + }, + { + "epoch": 0.10094530429423355, + "grad_norm": 0.27734375, + "learning_rate": 0.0019449834944035439, + "loss": 0.123, + "step": 11629 + }, + { + "epoch": 0.10095398477443772, + "grad_norm": 0.283203125, + "learning_rate": 0.001944973247609359, + "loss": 0.1445, + "step": 11630 + }, + { + "epoch": 0.10096266525464188, + "grad_norm": 0.6171875, + "learning_rate": 0.0019449629998911238, + "loss": 0.2344, + "step": 11631 + }, + { + "epoch": 0.10097134573484605, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019449527512488498, + "loss": 0.1113, + "step": 11632 + }, + { + "epoch": 0.10098002621505021, + "grad_norm": 0.31640625, + "learning_rate": 0.0019449425016825487, + "loss": 0.1631, + "step": 11633 + }, + { + "epoch": 0.10098870669525438, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019449322511922306, + "loss": 0.1348, + "step": 11634 + }, + { + "epoch": 0.10099738717545854, + "grad_norm": 0.26953125, + "learning_rate": 0.0019449219997779072, + "loss": 0.1245, + "step": 11635 + }, + { + "epoch": 0.10100606765566271, + "grad_norm": 0.5, + "learning_rate": 0.0019449117474395904, + "loss": 0.0952, + "step": 11636 + }, + { + "epoch": 0.10101474813586687, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019449014941772902, + "loss": 0.1094, + "step": 11637 + }, + { + "epoch": 0.10102342861607104, + "grad_norm": 0.7578125, + "learning_rate": 0.0019448912399910186, + "loss": 0.0981, + "step": 11638 + }, + { + "epoch": 0.1010321090962752, + "grad_norm": 0.5078125, + "learning_rate": 0.0019448809848807867, + "loss": 0.123, + "step": 11639 + }, + { + "epoch": 0.10104078957647937, + "grad_norm": 0.294921875, + "learning_rate": 0.0019448707288466057, + "loss": 0.1074, + "step": 11640 + }, + { + "epoch": 0.10104947005668354, + "grad_norm": 0.65234375, + "learning_rate": 0.0019448604718884867, + "loss": 0.1328, + "step": 11641 + }, + { + "epoch": 0.1010581505368877, + "grad_norm": 0.34375, + "learning_rate": 0.001944850214006441, + "loss": 0.1533, + "step": 11642 + }, + { + "epoch": 0.10106683101709187, + "grad_norm": 0.42578125, + "learning_rate": 0.0019448399552004798, + "loss": 0.1152, + "step": 11643 + }, + { + "epoch": 0.10107551149729603, + "grad_norm": 0.251953125, + "learning_rate": 0.0019448296954706142, + "loss": 0.1348, + "step": 11644 + }, + { + "epoch": 0.1010841919775002, + "grad_norm": 0.29296875, + "learning_rate": 0.0019448194348168557, + "loss": 0.0933, + "step": 11645 + }, + { + "epoch": 0.10109287245770436, + "grad_norm": 0.4765625, + "learning_rate": 0.0019448091732392153, + "loss": 0.123, + "step": 11646 + }, + { + "epoch": 0.10110155293790853, + "grad_norm": 0.25390625, + "learning_rate": 0.0019447989107377044, + "loss": 0.1069, + "step": 11647 + }, + { + "epoch": 0.10111023341811269, + "grad_norm": 0.1337890625, + "learning_rate": 0.001944788647312334, + "loss": 0.127, + "step": 11648 + }, + { + "epoch": 0.10111891389831686, + "grad_norm": 0.09912109375, + "learning_rate": 0.001944778382963115, + "loss": 0.0942, + "step": 11649 + }, + { + "epoch": 0.10112759437852102, + "grad_norm": 0.70703125, + "learning_rate": 0.0019447681176900599, + "loss": 0.1201, + "step": 11650 + }, + { + "epoch": 0.10113627485872519, + "grad_norm": 0.3828125, + "learning_rate": 0.0019447578514931786, + "loss": 0.1133, + "step": 11651 + }, + { + "epoch": 0.10114495533892935, + "grad_norm": 0.50390625, + "learning_rate": 0.0019447475843724826, + "loss": 0.1475, + "step": 11652 + }, + { + "epoch": 0.10115363581913352, + "grad_norm": 0.15234375, + "learning_rate": 0.001944737316327984, + "loss": 0.1113, + "step": 11653 + }, + { + "epoch": 0.10116231629933768, + "grad_norm": 0.55859375, + "learning_rate": 0.0019447270473596928, + "loss": 0.1484, + "step": 11654 + }, + { + "epoch": 0.10117099677954185, + "grad_norm": 0.12109375, + "learning_rate": 0.001944716777467621, + "loss": 0.1133, + "step": 11655 + }, + { + "epoch": 0.10117967725974601, + "grad_norm": 0.080078125, + "learning_rate": 0.0019447065066517796, + "loss": 0.1152, + "step": 11656 + }, + { + "epoch": 0.10118835773995018, + "grad_norm": 0.13671875, + "learning_rate": 0.0019446962349121797, + "loss": 0.1289, + "step": 11657 + }, + { + "epoch": 0.10119703822015434, + "grad_norm": 0.111328125, + "learning_rate": 0.001944685962248833, + "loss": 0.1167, + "step": 11658 + }, + { + "epoch": 0.10120571870035851, + "grad_norm": 0.123046875, + "learning_rate": 0.0019446756886617501, + "loss": 0.1245, + "step": 11659 + }, + { + "epoch": 0.10121439918056267, + "grad_norm": 0.48046875, + "learning_rate": 0.0019446654141509429, + "loss": 0.1084, + "step": 11660 + }, + { + "epoch": 0.10122307966076684, + "grad_norm": 0.140625, + "learning_rate": 0.001944655138716422, + "loss": 0.1055, + "step": 11661 + }, + { + "epoch": 0.101231760140971, + "grad_norm": 0.23046875, + "learning_rate": 0.0019446448623581987, + "loss": 0.1436, + "step": 11662 + }, + { + "epoch": 0.10124044062117517, + "grad_norm": 0.65234375, + "learning_rate": 0.0019446345850762847, + "loss": 0.1484, + "step": 11663 + }, + { + "epoch": 0.10124912110137933, + "grad_norm": 0.30078125, + "learning_rate": 0.0019446243068706914, + "loss": 0.1113, + "step": 11664 + }, + { + "epoch": 0.1012578015815835, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019446140277414293, + "loss": 0.1045, + "step": 11665 + }, + { + "epoch": 0.10126648206178766, + "grad_norm": 0.359375, + "learning_rate": 0.0019446037476885099, + "loss": 0.1816, + "step": 11666 + }, + { + "epoch": 0.10127516254199183, + "grad_norm": 0.11181640625, + "learning_rate": 0.001944593466711945, + "loss": 0.1201, + "step": 11667 + }, + { + "epoch": 0.101283843022196, + "grad_norm": 0.8046875, + "learning_rate": 0.0019445831848117448, + "loss": 0.1445, + "step": 11668 + }, + { + "epoch": 0.10129252350240016, + "grad_norm": 0.212890625, + "learning_rate": 0.0019445729019879215, + "loss": 0.1104, + "step": 11669 + }, + { + "epoch": 0.10130120398260432, + "grad_norm": 0.166015625, + "learning_rate": 0.0019445626182404859, + "loss": 0.2012, + "step": 11670 + }, + { + "epoch": 0.10130988446280849, + "grad_norm": 0.40625, + "learning_rate": 0.001944552333569449, + "loss": 0.1045, + "step": 11671 + }, + { + "epoch": 0.10131856494301265, + "grad_norm": 0.41796875, + "learning_rate": 0.0019445420479748225, + "loss": 0.1387, + "step": 11672 + }, + { + "epoch": 0.10132724542321682, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019445317614566178, + "loss": 0.1084, + "step": 11673 + }, + { + "epoch": 0.10133592590342098, + "grad_norm": 0.083984375, + "learning_rate": 0.0019445214740148457, + "loss": 0.1123, + "step": 11674 + }, + { + "epoch": 0.10134460638362514, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019445111856495176, + "loss": 0.0942, + "step": 11675 + }, + { + "epoch": 0.1013532868638293, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019445008963606448, + "loss": 0.082, + "step": 11676 + }, + { + "epoch": 0.10136196734403347, + "grad_norm": 0.412109375, + "learning_rate": 0.0019444906061482383, + "loss": 0.1006, + "step": 11677 + }, + { + "epoch": 0.10137064782423763, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019444803150123096, + "loss": 0.1387, + "step": 11678 + }, + { + "epoch": 0.1013793283044418, + "grad_norm": 0.083984375, + "learning_rate": 0.00194447002295287, + "loss": 0.123, + "step": 11679 + }, + { + "epoch": 0.10138800878464596, + "grad_norm": 0.248046875, + "learning_rate": 0.0019444597299699308, + "loss": 0.0928, + "step": 11680 + }, + { + "epoch": 0.10139668926485013, + "grad_norm": 0.6015625, + "learning_rate": 0.001944449436063503, + "loss": 0.127, + "step": 11681 + }, + { + "epoch": 0.10140536974505429, + "grad_norm": 0.087890625, + "learning_rate": 0.0019444391412335978, + "loss": 0.1465, + "step": 11682 + }, + { + "epoch": 0.10141405022525846, + "grad_norm": 0.06298828125, + "learning_rate": 0.0019444288454802268, + "loss": 0.0903, + "step": 11683 + }, + { + "epoch": 0.10142273070546262, + "grad_norm": 0.224609375, + "learning_rate": 0.0019444185488034012, + "loss": 0.1406, + "step": 11684 + }, + { + "epoch": 0.10143141118566679, + "grad_norm": 0.103515625, + "learning_rate": 0.0019444082512031322, + "loss": 0.1084, + "step": 11685 + }, + { + "epoch": 0.10144009166587095, + "grad_norm": 0.46875, + "learning_rate": 0.0019443979526794308, + "loss": 0.1582, + "step": 11686 + }, + { + "epoch": 0.10144877214607512, + "grad_norm": 0.12109375, + "learning_rate": 0.0019443876532323086, + "loss": 0.123, + "step": 11687 + }, + { + "epoch": 0.10145745262627928, + "grad_norm": 0.51953125, + "learning_rate": 0.0019443773528617766, + "loss": 0.1846, + "step": 11688 + }, + { + "epoch": 0.10146613310648345, + "grad_norm": 0.12451171875, + "learning_rate": 0.0019443670515678464, + "loss": 0.1113, + "step": 11689 + }, + { + "epoch": 0.10147481358668761, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019443567493505293, + "loss": 0.1201, + "step": 11690 + }, + { + "epoch": 0.10148349406689178, + "grad_norm": 0.171875, + "learning_rate": 0.001944346446209836, + "loss": 0.1006, + "step": 11691 + }, + { + "epoch": 0.10149217454709594, + "grad_norm": 0.09716796875, + "learning_rate": 0.001944336142145778, + "loss": 0.123, + "step": 11692 + }, + { + "epoch": 0.10150085502730011, + "grad_norm": 0.7890625, + "learning_rate": 0.001944325837158367, + "loss": 0.126, + "step": 11693 + }, + { + "epoch": 0.10150953550750427, + "grad_norm": 0.296875, + "learning_rate": 0.0019443155312476139, + "loss": 0.1201, + "step": 11694 + }, + { + "epoch": 0.10151821598770844, + "grad_norm": 0.474609375, + "learning_rate": 0.0019443052244135298, + "loss": 0.1001, + "step": 11695 + }, + { + "epoch": 0.1015268964679126, + "grad_norm": 0.7109375, + "learning_rate": 0.0019442949166561265, + "loss": 0.1196, + "step": 11696 + }, + { + "epoch": 0.10153557694811677, + "grad_norm": 0.66796875, + "learning_rate": 0.001944284607975415, + "loss": 0.1143, + "step": 11697 + }, + { + "epoch": 0.10154425742832093, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019442742983714063, + "loss": 0.1494, + "step": 11698 + }, + { + "epoch": 0.1015529379085251, + "grad_norm": 0.57421875, + "learning_rate": 0.001944263987844112, + "loss": 0.1191, + "step": 11699 + }, + { + "epoch": 0.10156161838872926, + "grad_norm": 0.26953125, + "learning_rate": 0.001944253676393543, + "loss": 0.123, + "step": 11700 + }, + { + "epoch": 0.10157029886893343, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019442433640197113, + "loss": 0.1113, + "step": 11701 + }, + { + "epoch": 0.1015789793491376, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019442330507226273, + "loss": 0.124, + "step": 11702 + }, + { + "epoch": 0.10158765982934176, + "grad_norm": 0.140625, + "learning_rate": 0.0019442227365023032, + "loss": 0.1377, + "step": 11703 + }, + { + "epoch": 0.10159634030954592, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019442124213587494, + "loss": 0.0889, + "step": 11704 + }, + { + "epoch": 0.10160502078975009, + "grad_norm": 0.455078125, + "learning_rate": 0.0019442021052919777, + "loss": 0.124, + "step": 11705 + }, + { + "epoch": 0.10161370126995425, + "grad_norm": 0.494140625, + "learning_rate": 0.0019441917883019995, + "loss": 0.1523, + "step": 11706 + }, + { + "epoch": 0.10162238175015842, + "grad_norm": 0.486328125, + "learning_rate": 0.0019441814703888257, + "loss": 0.0967, + "step": 11707 + }, + { + "epoch": 0.10163106223036258, + "grad_norm": 0.404296875, + "learning_rate": 0.0019441711515524676, + "loss": 0.1299, + "step": 11708 + }, + { + "epoch": 0.10163974271056675, + "grad_norm": 0.6875, + "learning_rate": 0.0019441608317929367, + "loss": 0.1797, + "step": 11709 + }, + { + "epoch": 0.10164842319077091, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019441505111102441, + "loss": 0.1123, + "step": 11710 + }, + { + "epoch": 0.10165710367097508, + "grad_norm": 0.28125, + "learning_rate": 0.0019441401895044013, + "loss": 0.0674, + "step": 11711 + }, + { + "epoch": 0.10166578415117924, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019441298669754196, + "loss": 0.1006, + "step": 11712 + }, + { + "epoch": 0.10167446463138341, + "grad_norm": 0.126953125, + "learning_rate": 0.0019441195435233101, + "loss": 0.1348, + "step": 11713 + }, + { + "epoch": 0.10168314511158758, + "grad_norm": 0.46484375, + "learning_rate": 0.0019441092191480839, + "loss": 0.104, + "step": 11714 + }, + { + "epoch": 0.10169182559179174, + "grad_norm": 0.51171875, + "learning_rate": 0.001944098893849753, + "loss": 0.1299, + "step": 11715 + }, + { + "epoch": 0.1017005060719959, + "grad_norm": 0.150390625, + "learning_rate": 0.0019440885676283274, + "loss": 0.1006, + "step": 11716 + }, + { + "epoch": 0.10170918655220007, + "grad_norm": 0.3671875, + "learning_rate": 0.00194407824048382, + "loss": 0.1074, + "step": 11717 + }, + { + "epoch": 0.10171786703240424, + "grad_norm": 0.51171875, + "learning_rate": 0.001944067912416241, + "loss": 0.1299, + "step": 11718 + }, + { + "epoch": 0.1017265475126084, + "grad_norm": 0.115234375, + "learning_rate": 0.0019440575834256022, + "loss": 0.1006, + "step": 11719 + }, + { + "epoch": 0.10173522799281257, + "grad_norm": 0.65234375, + "learning_rate": 0.0019440472535119148, + "loss": 0.1016, + "step": 11720 + }, + { + "epoch": 0.10174390847301673, + "grad_norm": 0.625, + "learning_rate": 0.0019440369226751894, + "loss": 0.1021, + "step": 11721 + }, + { + "epoch": 0.1017525889532209, + "grad_norm": 0.302734375, + "learning_rate": 0.0019440265909154385, + "loss": 0.1162, + "step": 11722 + }, + { + "epoch": 0.10176126943342506, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019440162582326727, + "loss": 0.123, + "step": 11723 + }, + { + "epoch": 0.10176994991362923, + "grad_norm": 0.2197265625, + "learning_rate": 0.001944005924626903, + "loss": 0.1118, + "step": 11724 + }, + { + "epoch": 0.10177863039383339, + "grad_norm": 0.259765625, + "learning_rate": 0.0019439955900981416, + "loss": 0.1582, + "step": 11725 + }, + { + "epoch": 0.10178731087403756, + "grad_norm": 0.08544921875, + "learning_rate": 0.001943985254646399, + "loss": 0.1445, + "step": 11726 + }, + { + "epoch": 0.10179599135424172, + "grad_norm": 0.53515625, + "learning_rate": 0.001943974918271687, + "loss": 0.1191, + "step": 11727 + }, + { + "epoch": 0.10180467183444589, + "grad_norm": 0.953125, + "learning_rate": 0.0019439645809740168, + "loss": 0.127, + "step": 11728 + }, + { + "epoch": 0.10181335231465005, + "grad_norm": 1.3125, + "learning_rate": 0.0019439542427533993, + "loss": 0.1201, + "step": 11729 + }, + { + "epoch": 0.10182203279485422, + "grad_norm": 0.318359375, + "learning_rate": 0.0019439439036098463, + "loss": 0.1006, + "step": 11730 + }, + { + "epoch": 0.10183071327505838, + "grad_norm": 0.23046875, + "learning_rate": 0.0019439335635433688, + "loss": 0.1128, + "step": 11731 + }, + { + "epoch": 0.10183939375526255, + "grad_norm": 0.251953125, + "learning_rate": 0.0019439232225539785, + "loss": 0.1035, + "step": 11732 + }, + { + "epoch": 0.10184807423546671, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019439128806416864, + "loss": 0.1143, + "step": 11733 + }, + { + "epoch": 0.10185675471567088, + "grad_norm": 0.5234375, + "learning_rate": 0.0019439025378065037, + "loss": 0.1221, + "step": 11734 + }, + { + "epoch": 0.10186543519587504, + "grad_norm": 0.08056640625, + "learning_rate": 0.001943892194048442, + "loss": 0.1011, + "step": 11735 + }, + { + "epoch": 0.1018741156760792, + "grad_norm": 0.189453125, + "learning_rate": 0.001943881849367512, + "loss": 0.0889, + "step": 11736 + }, + { + "epoch": 0.10188279615628336, + "grad_norm": 0.41796875, + "learning_rate": 0.001943871503763726, + "loss": 0.1367, + "step": 11737 + }, + { + "epoch": 0.10189147663648752, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019438611572370945, + "loss": 0.1152, + "step": 11738 + }, + { + "epoch": 0.10190015711669169, + "grad_norm": 0.80078125, + "learning_rate": 0.0019438508097876295, + "loss": 0.1416, + "step": 11739 + }, + { + "epoch": 0.10190883759689585, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019438404614153418, + "loss": 0.1523, + "step": 11740 + }, + { + "epoch": 0.10191751807710002, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019438301121202425, + "loss": 0.124, + "step": 11741 + }, + { + "epoch": 0.10192619855730418, + "grad_norm": 0.82421875, + "learning_rate": 0.0019438197619023436, + "loss": 0.1533, + "step": 11742 + }, + { + "epoch": 0.10193487903750835, + "grad_norm": 0.162109375, + "learning_rate": 0.001943809410761656, + "loss": 0.084, + "step": 11743 + }, + { + "epoch": 0.10194355951771251, + "grad_norm": 0.1376953125, + "learning_rate": 0.001943799058698191, + "loss": 0.0938, + "step": 11744 + }, + { + "epoch": 0.10195223999791668, + "grad_norm": 0.439453125, + "learning_rate": 0.0019437887057119603, + "loss": 0.1196, + "step": 11745 + }, + { + "epoch": 0.10196092047812085, + "grad_norm": 0.103515625, + "learning_rate": 0.0019437783518029747, + "loss": 0.1396, + "step": 11746 + }, + { + "epoch": 0.10196960095832501, + "grad_norm": 0.427734375, + "learning_rate": 0.0019437679969712461, + "loss": 0.1289, + "step": 11747 + }, + { + "epoch": 0.10197828143852918, + "grad_norm": 0.65234375, + "learning_rate": 0.0019437576412167851, + "loss": 0.0879, + "step": 11748 + }, + { + "epoch": 0.10198696191873334, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019437472845396037, + "loss": 0.1582, + "step": 11749 + }, + { + "epoch": 0.1019956423989375, + "grad_norm": 0.310546875, + "learning_rate": 0.0019437369269397126, + "loss": 0.1094, + "step": 11750 + }, + { + "epoch": 0.10200432287914167, + "grad_norm": 0.546875, + "learning_rate": 0.001943726568417124, + "loss": 0.166, + "step": 11751 + }, + { + "epoch": 0.10201300335934584, + "grad_norm": 0.2109375, + "learning_rate": 0.0019437162089718484, + "loss": 0.0991, + "step": 11752 + }, + { + "epoch": 0.10202168383955, + "grad_norm": 0.2734375, + "learning_rate": 0.0019437058486038972, + "loss": 0.1455, + "step": 11753 + }, + { + "epoch": 0.10203036431975417, + "grad_norm": 0.251953125, + "learning_rate": 0.0019436954873132822, + "loss": 0.0908, + "step": 11754 + }, + { + "epoch": 0.10203904479995833, + "grad_norm": 0.130859375, + "learning_rate": 0.0019436851251000147, + "loss": 0.1289, + "step": 11755 + }, + { + "epoch": 0.1020477252801625, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019436747619641056, + "loss": 0.1289, + "step": 11756 + }, + { + "epoch": 0.10205640576036666, + "grad_norm": 0.357421875, + "learning_rate": 0.0019436643979055665, + "loss": 0.1309, + "step": 11757 + }, + { + "epoch": 0.10206508624057083, + "grad_norm": 0.5078125, + "learning_rate": 0.0019436540329244086, + "loss": 0.1016, + "step": 11758 + }, + { + "epoch": 0.10207376672077499, + "grad_norm": 0.193359375, + "learning_rate": 0.0019436436670206433, + "loss": 0.1406, + "step": 11759 + }, + { + "epoch": 0.10208244720097916, + "grad_norm": 0.384765625, + "learning_rate": 0.0019436333001942821, + "loss": 0.1455, + "step": 11760 + }, + { + "epoch": 0.10209112768118332, + "grad_norm": 0.09228515625, + "learning_rate": 0.001943622932445336, + "loss": 0.1064, + "step": 11761 + }, + { + "epoch": 0.10209980816138749, + "grad_norm": 0.208984375, + "learning_rate": 0.0019436125637738166, + "loss": 0.1094, + "step": 11762 + }, + { + "epoch": 0.10210848864159165, + "grad_norm": 0.49609375, + "learning_rate": 0.0019436021941797354, + "loss": 0.1191, + "step": 11763 + }, + { + "epoch": 0.10211716912179582, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019435918236631036, + "loss": 0.0996, + "step": 11764 + }, + { + "epoch": 0.10212584960199998, + "grad_norm": 0.341796875, + "learning_rate": 0.0019435814522239321, + "loss": 0.1494, + "step": 11765 + }, + { + "epoch": 0.10213453008220415, + "grad_norm": 0.49609375, + "learning_rate": 0.0019435710798622328, + "loss": 0.0947, + "step": 11766 + }, + { + "epoch": 0.10214321056240831, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019435607065780168, + "loss": 0.0942, + "step": 11767 + }, + { + "epoch": 0.10215189104261248, + "grad_norm": 1.125, + "learning_rate": 0.0019435503323712953, + "loss": 0.1177, + "step": 11768 + }, + { + "epoch": 0.10216057152281664, + "grad_norm": 0.80078125, + "learning_rate": 0.00194353995724208, + "loss": 0.166, + "step": 11769 + }, + { + "epoch": 0.10216925200302081, + "grad_norm": 0.166015625, + "learning_rate": 0.001943529581190382, + "loss": 0.0957, + "step": 11770 + }, + { + "epoch": 0.10217793248322497, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019435192042162128, + "loss": 0.1348, + "step": 11771 + }, + { + "epoch": 0.10218661296342914, + "grad_norm": 0.5625, + "learning_rate": 0.0019435088263195838, + "loss": 0.0967, + "step": 11772 + }, + { + "epoch": 0.1021952934436333, + "grad_norm": 0.5234375, + "learning_rate": 0.0019434984475005062, + "loss": 0.0977, + "step": 11773 + }, + { + "epoch": 0.10220397392383747, + "grad_norm": 0.337890625, + "learning_rate": 0.0019434880677589913, + "loss": 0.0996, + "step": 11774 + }, + { + "epoch": 0.10221265440404163, + "grad_norm": 0.45703125, + "learning_rate": 0.0019434776870950504, + "loss": 0.1133, + "step": 11775 + }, + { + "epoch": 0.1022213348842458, + "grad_norm": 0.7109375, + "learning_rate": 0.0019434673055086949, + "loss": 0.1416, + "step": 11776 + }, + { + "epoch": 0.10223001536444996, + "grad_norm": 0.4296875, + "learning_rate": 0.0019434569229999366, + "loss": 0.1079, + "step": 11777 + }, + { + "epoch": 0.10223869584465413, + "grad_norm": 0.134765625, + "learning_rate": 0.001943446539568786, + "loss": 0.1533, + "step": 11778 + }, + { + "epoch": 0.1022473763248583, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019434361552152556, + "loss": 0.0913, + "step": 11779 + }, + { + "epoch": 0.10225605680506246, + "grad_norm": 0.44921875, + "learning_rate": 0.0019434257699393555, + "loss": 0.106, + "step": 11780 + }, + { + "epoch": 0.10226473728526662, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019434153837410977, + "loss": 0.1055, + "step": 11781 + }, + { + "epoch": 0.10227341776547079, + "grad_norm": 0.341796875, + "learning_rate": 0.0019434049966204937, + "loss": 0.0986, + "step": 11782 + }, + { + "epoch": 0.10228209824567495, + "grad_norm": 0.41015625, + "learning_rate": 0.0019433946085775547, + "loss": 0.125, + "step": 11783 + }, + { + "epoch": 0.10229077872587912, + "grad_norm": 0.31640625, + "learning_rate": 0.001943384219612292, + "loss": 0.124, + "step": 11784 + }, + { + "epoch": 0.10229945920608328, + "grad_norm": 0.26953125, + "learning_rate": 0.001943373829724717, + "loss": 0.1348, + "step": 11785 + }, + { + "epoch": 0.10230813968628745, + "grad_norm": 0.220703125, + "learning_rate": 0.0019433634389148407, + "loss": 0.0986, + "step": 11786 + }, + { + "epoch": 0.10231682016649161, + "grad_norm": 0.216796875, + "learning_rate": 0.0019433530471826752, + "loss": 0.0947, + "step": 11787 + }, + { + "epoch": 0.10232550064669578, + "grad_norm": 0.373046875, + "learning_rate": 0.0019433426545282313, + "loss": 0.125, + "step": 11788 + }, + { + "epoch": 0.10233418112689995, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019433322609515207, + "loss": 0.1143, + "step": 11789 + }, + { + "epoch": 0.10234286160710411, + "grad_norm": 0.5, + "learning_rate": 0.0019433218664525545, + "loss": 0.1279, + "step": 11790 + }, + { + "epoch": 0.10235154208730828, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019433114710313442, + "loss": 0.0723, + "step": 11791 + }, + { + "epoch": 0.10236022256751244, + "grad_norm": 0.365234375, + "learning_rate": 0.0019433010746879012, + "loss": 0.1484, + "step": 11792 + }, + { + "epoch": 0.1023689030477166, + "grad_norm": 0.49609375, + "learning_rate": 0.0019432906774222367, + "loss": 0.1221, + "step": 11793 + }, + { + "epoch": 0.10237758352792077, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019432802792343625, + "loss": 0.0918, + "step": 11794 + }, + { + "epoch": 0.10238626400812494, + "grad_norm": 0.6953125, + "learning_rate": 0.0019432698801242893, + "loss": 0.1162, + "step": 11795 + }, + { + "epoch": 0.1023949444883291, + "grad_norm": 0.115234375, + "learning_rate": 0.001943259480092029, + "loss": 0.1221, + "step": 11796 + }, + { + "epoch": 0.10240362496853327, + "grad_norm": 0.859375, + "learning_rate": 0.0019432490791375925, + "loss": 0.1445, + "step": 11797 + }, + { + "epoch": 0.10241230544873742, + "grad_norm": 0.1240234375, + "learning_rate": 0.001943238677260992, + "loss": 0.1162, + "step": 11798 + }, + { + "epoch": 0.10242098592894158, + "grad_norm": 0.41015625, + "learning_rate": 0.001943228274462238, + "loss": 0.1084, + "step": 11799 + }, + { + "epoch": 0.10242966640914575, + "grad_norm": 1.953125, + "learning_rate": 0.0019432178707413424, + "loss": 0.1973, + "step": 11800 + }, + { + "epoch": 0.10243834688934991, + "grad_norm": 0.310546875, + "learning_rate": 0.0019432074660983162, + "loss": 0.1074, + "step": 11801 + }, + { + "epoch": 0.10244702736955408, + "grad_norm": 0.2314453125, + "learning_rate": 0.001943197060533171, + "loss": 0.1064, + "step": 11802 + }, + { + "epoch": 0.10245570784975824, + "grad_norm": 0.22265625, + "learning_rate": 0.0019431866540459183, + "loss": 0.1113, + "step": 11803 + }, + { + "epoch": 0.10246438832996241, + "grad_norm": 1.109375, + "learning_rate": 0.0019431762466365695, + "loss": 0.125, + "step": 11804 + }, + { + "epoch": 0.10247306881016657, + "grad_norm": 0.51953125, + "learning_rate": 0.0019431658383051355, + "loss": 0.1045, + "step": 11805 + }, + { + "epoch": 0.10248174929037074, + "grad_norm": 1.0546875, + "learning_rate": 0.001943155429051628, + "loss": 0.0928, + "step": 11806 + }, + { + "epoch": 0.1024904297705749, + "grad_norm": 0.291015625, + "learning_rate": 0.0019431450188760586, + "loss": 0.0845, + "step": 11807 + }, + { + "epoch": 0.10249911025077907, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019431346077784385, + "loss": 0.103, + "step": 11808 + }, + { + "epoch": 0.10250779073098323, + "grad_norm": 0.251953125, + "learning_rate": 0.0019431241957587788, + "loss": 0.1338, + "step": 11809 + }, + { + "epoch": 0.1025164712111874, + "grad_norm": 0.68359375, + "learning_rate": 0.0019431137828170915, + "loss": 0.1406, + "step": 11810 + }, + { + "epoch": 0.10252515169139156, + "grad_norm": 0.5703125, + "learning_rate": 0.0019431033689533872, + "loss": 0.1328, + "step": 11811 + }, + { + "epoch": 0.10253383217159573, + "grad_norm": 0.416015625, + "learning_rate": 0.0019430929541676782, + "loss": 0.0845, + "step": 11812 + }, + { + "epoch": 0.1025425126517999, + "grad_norm": 0.625, + "learning_rate": 0.001943082538459975, + "loss": 0.1357, + "step": 11813 + }, + { + "epoch": 0.10255119313200406, + "grad_norm": 0.58984375, + "learning_rate": 0.0019430721218302897, + "loss": 0.1289, + "step": 11814 + }, + { + "epoch": 0.10255987361220822, + "grad_norm": 0.81640625, + "learning_rate": 0.0019430617042786333, + "loss": 0.0991, + "step": 11815 + }, + { + "epoch": 0.10256855409241239, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019430512858050171, + "loss": 0.1172, + "step": 11816 + }, + { + "epoch": 0.10257723457261655, + "grad_norm": 0.107421875, + "learning_rate": 0.001943040866409453, + "loss": 0.1021, + "step": 11817 + }, + { + "epoch": 0.10258591505282072, + "grad_norm": 0.52734375, + "learning_rate": 0.001943030446091952, + "loss": 0.0913, + "step": 11818 + }, + { + "epoch": 0.10259459553302488, + "grad_norm": 0.671875, + "learning_rate": 0.0019430200248525254, + "loss": 0.0996, + "step": 11819 + }, + { + "epoch": 0.10260327601322905, + "grad_norm": 0.251953125, + "learning_rate": 0.001943009602691185, + "loss": 0.1621, + "step": 11820 + }, + { + "epoch": 0.10261195649343322, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019429991796079415, + "loss": 0.1211, + "step": 11821 + }, + { + "epoch": 0.10262063697363738, + "grad_norm": 0.055419921875, + "learning_rate": 0.0019429887556028074, + "loss": 0.1094, + "step": 11822 + }, + { + "epoch": 0.10262931745384155, + "grad_norm": 0.671875, + "learning_rate": 0.0019429783306757929, + "loss": 0.1758, + "step": 11823 + }, + { + "epoch": 0.10263799793404571, + "grad_norm": 0.3671875, + "learning_rate": 0.0019429679048269103, + "loss": 0.1387, + "step": 11824 + }, + { + "epoch": 0.10264667841424988, + "grad_norm": 0.77734375, + "learning_rate": 0.0019429574780561708, + "loss": 0.207, + "step": 11825 + }, + { + "epoch": 0.10265535889445404, + "grad_norm": 0.328125, + "learning_rate": 0.0019429470503635853, + "loss": 0.0732, + "step": 11826 + }, + { + "epoch": 0.1026640393746582, + "grad_norm": 0.134765625, + "learning_rate": 0.0019429366217491657, + "loss": 0.1143, + "step": 11827 + }, + { + "epoch": 0.10267271985486237, + "grad_norm": 0.25, + "learning_rate": 0.0019429261922129233, + "loss": 0.1216, + "step": 11828 + }, + { + "epoch": 0.10268140033506654, + "grad_norm": 0.423828125, + "learning_rate": 0.0019429157617548694, + "loss": 0.1104, + "step": 11829 + }, + { + "epoch": 0.1026900808152707, + "grad_norm": 0.345703125, + "learning_rate": 0.0019429053303750157, + "loss": 0.0996, + "step": 11830 + }, + { + "epoch": 0.10269876129547487, + "grad_norm": 0.294921875, + "learning_rate": 0.0019428948980733732, + "loss": 0.1064, + "step": 11831 + }, + { + "epoch": 0.10270744177567903, + "grad_norm": 0.306640625, + "learning_rate": 0.0019428844648499534, + "loss": 0.1157, + "step": 11832 + }, + { + "epoch": 0.1027161222558832, + "grad_norm": 0.66796875, + "learning_rate": 0.0019428740307047684, + "loss": 0.1025, + "step": 11833 + }, + { + "epoch": 0.10272480273608736, + "grad_norm": 0.9296875, + "learning_rate": 0.0019428635956378283, + "loss": 0.127, + "step": 11834 + }, + { + "epoch": 0.10273348321629153, + "grad_norm": 0.2109375, + "learning_rate": 0.0019428531596491457, + "loss": 0.1455, + "step": 11835 + }, + { + "epoch": 0.10274216369649569, + "grad_norm": 0.13671875, + "learning_rate": 0.0019428427227387312, + "loss": 0.1455, + "step": 11836 + }, + { + "epoch": 0.10275084417669986, + "grad_norm": 0.50390625, + "learning_rate": 0.0019428322849065966, + "loss": 0.0996, + "step": 11837 + }, + { + "epoch": 0.10275952465690402, + "grad_norm": 0.208984375, + "learning_rate": 0.0019428218461527535, + "loss": 0.1055, + "step": 11838 + }, + { + "epoch": 0.10276820513710819, + "grad_norm": 0.87890625, + "learning_rate": 0.0019428114064772128, + "loss": 0.1133, + "step": 11839 + }, + { + "epoch": 0.10277688561731235, + "grad_norm": 0.07177734375, + "learning_rate": 0.0019428009658799866, + "loss": 0.0864, + "step": 11840 + }, + { + "epoch": 0.10278556609751652, + "grad_norm": 0.08544921875, + "learning_rate": 0.001942790524361086, + "loss": 0.1133, + "step": 11841 + }, + { + "epoch": 0.10279424657772068, + "grad_norm": 0.2421875, + "learning_rate": 0.001942780081920522, + "loss": 0.0957, + "step": 11842 + }, + { + "epoch": 0.10280292705792485, + "grad_norm": 0.1640625, + "learning_rate": 0.0019427696385583065, + "loss": 0.1045, + "step": 11843 + }, + { + "epoch": 0.10281160753812901, + "grad_norm": 0.58203125, + "learning_rate": 0.0019427591942744508, + "loss": 0.1191, + "step": 11844 + }, + { + "epoch": 0.10282028801833318, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019427487490689661, + "loss": 0.1016, + "step": 11845 + }, + { + "epoch": 0.10282896849853734, + "grad_norm": 0.73046875, + "learning_rate": 0.0019427383029418641, + "loss": 0.2158, + "step": 11846 + }, + { + "epoch": 0.10283764897874151, + "grad_norm": 0.6640625, + "learning_rate": 0.0019427278558931563, + "loss": 0.124, + "step": 11847 + }, + { + "epoch": 0.10284632945894567, + "grad_norm": 0.3359375, + "learning_rate": 0.001942717407922854, + "loss": 0.085, + "step": 11848 + }, + { + "epoch": 0.10285500993914984, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019427069590309686, + "loss": 0.126, + "step": 11849 + }, + { + "epoch": 0.102863690419354, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019426965092175113, + "loss": 0.1191, + "step": 11850 + }, + { + "epoch": 0.10287237089955817, + "grad_norm": 0.1474609375, + "learning_rate": 0.001942686058482494, + "loss": 0.1172, + "step": 11851 + }, + { + "epoch": 0.10288105137976233, + "grad_norm": 0.515625, + "learning_rate": 0.0019426756068259277, + "loss": 0.1064, + "step": 11852 + }, + { + "epoch": 0.1028897318599665, + "grad_norm": 0.361328125, + "learning_rate": 0.0019426651542478243, + "loss": 0.1338, + "step": 11853 + }, + { + "epoch": 0.10289841234017066, + "grad_norm": 0.458984375, + "learning_rate": 0.0019426547007481948, + "loss": 0.1216, + "step": 11854 + }, + { + "epoch": 0.10290709282037483, + "grad_norm": 0.3671875, + "learning_rate": 0.0019426442463270507, + "loss": 0.1182, + "step": 11855 + }, + { + "epoch": 0.102915773300579, + "grad_norm": 0.703125, + "learning_rate": 0.0019426337909844034, + "loss": 0.1836, + "step": 11856 + }, + { + "epoch": 0.10292445378078316, + "grad_norm": 0.66015625, + "learning_rate": 0.0019426233347202646, + "loss": 0.0957, + "step": 11857 + }, + { + "epoch": 0.10293313426098732, + "grad_norm": 0.150390625, + "learning_rate": 0.0019426128775346455, + "loss": 0.1201, + "step": 11858 + }, + { + "epoch": 0.10294181474119148, + "grad_norm": 0.33203125, + "learning_rate": 0.0019426024194275579, + "loss": 0.1167, + "step": 11859 + }, + { + "epoch": 0.10295049522139564, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019425919603990127, + "loss": 0.1216, + "step": 11860 + }, + { + "epoch": 0.1029591757015998, + "grad_norm": 0.2109375, + "learning_rate": 0.0019425815004490215, + "loss": 0.0996, + "step": 11861 + }, + { + "epoch": 0.10296785618180397, + "grad_norm": 0.7421875, + "learning_rate": 0.0019425710395775961, + "loss": 0.1045, + "step": 11862 + }, + { + "epoch": 0.10297653666200814, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019425605777847474, + "loss": 0.1719, + "step": 11863 + }, + { + "epoch": 0.1029852171422123, + "grad_norm": 0.58203125, + "learning_rate": 0.0019425501150704873, + "loss": 0.0903, + "step": 11864 + }, + { + "epoch": 0.10299389762241647, + "grad_norm": 0.259765625, + "learning_rate": 0.001942539651434827, + "loss": 0.1475, + "step": 11865 + }, + { + "epoch": 0.10300257810262063, + "grad_norm": 1.4296875, + "learning_rate": 0.0019425291868777781, + "loss": 0.1299, + "step": 11866 + }, + { + "epoch": 0.1030112585828248, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019425187213993518, + "loss": 0.127, + "step": 11867 + }, + { + "epoch": 0.10301993906302896, + "grad_norm": 0.259765625, + "learning_rate": 0.0019425082549995598, + "loss": 0.1035, + "step": 11868 + }, + { + "epoch": 0.10302861954323313, + "grad_norm": 0.2734375, + "learning_rate": 0.0019424977876784137, + "loss": 0.1143, + "step": 11869 + }, + { + "epoch": 0.10303730002343729, + "grad_norm": 0.6875, + "learning_rate": 0.001942487319435924, + "loss": 0.1104, + "step": 11870 + }, + { + "epoch": 0.10304598050364146, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019424768502721032, + "loss": 0.1289, + "step": 11871 + }, + { + "epoch": 0.10305466098384562, + "grad_norm": 0.169921875, + "learning_rate": 0.0019424663801869623, + "loss": 0.1406, + "step": 11872 + }, + { + "epoch": 0.10306334146404979, + "grad_norm": 0.40234375, + "learning_rate": 0.0019424559091805128, + "loss": 0.1172, + "step": 11873 + }, + { + "epoch": 0.10307202194425395, + "grad_norm": 0.486328125, + "learning_rate": 0.0019424454372527662, + "loss": 0.1104, + "step": 11874 + }, + { + "epoch": 0.10308070242445812, + "grad_norm": 0.21875, + "learning_rate": 0.0019424349644037342, + "loss": 0.1133, + "step": 11875 + }, + { + "epoch": 0.10308938290466228, + "grad_norm": 0.53515625, + "learning_rate": 0.0019424244906334277, + "loss": 0.1445, + "step": 11876 + }, + { + "epoch": 0.10309806338486645, + "grad_norm": 0.470703125, + "learning_rate": 0.0019424140159418585, + "loss": 0.1885, + "step": 11877 + }, + { + "epoch": 0.10310674386507061, + "grad_norm": 0.1337890625, + "learning_rate": 0.001942403540329038, + "loss": 0.1426, + "step": 11878 + }, + { + "epoch": 0.10311542434527478, + "grad_norm": 0.40625, + "learning_rate": 0.0019423930637949777, + "loss": 0.1484, + "step": 11879 + }, + { + "epoch": 0.10312410482547894, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019423825863396887, + "loss": 0.1143, + "step": 11880 + }, + { + "epoch": 0.10313278530568311, + "grad_norm": 0.6953125, + "learning_rate": 0.001942372107963183, + "loss": 0.1006, + "step": 11881 + }, + { + "epoch": 0.10314146578588727, + "grad_norm": 0.50390625, + "learning_rate": 0.0019423616286654718, + "loss": 0.1221, + "step": 11882 + }, + { + "epoch": 0.10315014626609144, + "grad_norm": 0.38671875, + "learning_rate": 0.0019423511484465668, + "loss": 0.1064, + "step": 11883 + }, + { + "epoch": 0.1031588267462956, + "grad_norm": 0.1328125, + "learning_rate": 0.001942340667306479, + "loss": 0.0933, + "step": 11884 + }, + { + "epoch": 0.10316750722649977, + "grad_norm": 0.314453125, + "learning_rate": 0.00194233018524522, + "loss": 0.1011, + "step": 11885 + }, + { + "epoch": 0.10317618770670393, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019423197022628016, + "loss": 0.1035, + "step": 11886 + }, + { + "epoch": 0.1031848681869081, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019423092183592347, + "loss": 0.0928, + "step": 11887 + }, + { + "epoch": 0.10319354866711226, + "grad_norm": 0.89453125, + "learning_rate": 0.0019422987335345314, + "loss": 0.1289, + "step": 11888 + }, + { + "epoch": 0.10320222914731643, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019422882477887027, + "loss": 0.1035, + "step": 11889 + }, + { + "epoch": 0.1032109096275206, + "grad_norm": 0.1328125, + "learning_rate": 0.0019422777611217603, + "loss": 0.1455, + "step": 11890 + }, + { + "epoch": 0.10321959010772476, + "grad_norm": 0.2109375, + "learning_rate": 0.0019422672735337158, + "loss": 0.1758, + "step": 11891 + }, + { + "epoch": 0.10322827058792892, + "grad_norm": 0.244140625, + "learning_rate": 0.0019422567850245803, + "loss": 0.1211, + "step": 11892 + }, + { + "epoch": 0.10323695106813309, + "grad_norm": 0.33984375, + "learning_rate": 0.0019422462955943653, + "loss": 0.1377, + "step": 11893 + }, + { + "epoch": 0.10324563154833725, + "grad_norm": 0.3671875, + "learning_rate": 0.0019422358052430823, + "loss": 0.1045, + "step": 11894 + }, + { + "epoch": 0.10325431202854142, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019422253139707436, + "loss": 0.1406, + "step": 11895 + }, + { + "epoch": 0.10326299250874559, + "grad_norm": 0.134765625, + "learning_rate": 0.001942214821777359, + "loss": 0.1143, + "step": 11896 + }, + { + "epoch": 0.10327167298894975, + "grad_norm": 0.82421875, + "learning_rate": 0.0019422043286629416, + "loss": 0.1123, + "step": 11897 + }, + { + "epoch": 0.10328035346915392, + "grad_norm": 0.236328125, + "learning_rate": 0.0019421938346275017, + "loss": 0.1699, + "step": 11898 + }, + { + "epoch": 0.10328903394935808, + "grad_norm": 0.734375, + "learning_rate": 0.0019421833396710516, + "loss": 0.1377, + "step": 11899 + }, + { + "epoch": 0.10329771442956225, + "grad_norm": 0.146484375, + "learning_rate": 0.0019421728437936023, + "loss": 0.1152, + "step": 11900 + }, + { + "epoch": 0.10330639490976641, + "grad_norm": 0.125, + "learning_rate": 0.001942162346995166, + "loss": 0.0957, + "step": 11901 + }, + { + "epoch": 0.10331507538997058, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019421518492757528, + "loss": 0.1426, + "step": 11902 + }, + { + "epoch": 0.10332375587017474, + "grad_norm": 0.365234375, + "learning_rate": 0.0019421413506353754, + "loss": 0.1113, + "step": 11903 + }, + { + "epoch": 0.1033324363503789, + "grad_norm": 0.216796875, + "learning_rate": 0.0019421308510740448, + "loss": 0.168, + "step": 11904 + }, + { + "epoch": 0.10334111683058307, + "grad_norm": 0.490234375, + "learning_rate": 0.0019421203505917727, + "loss": 0.1611, + "step": 11905 + }, + { + "epoch": 0.10334979731078724, + "grad_norm": 1.453125, + "learning_rate": 0.00194210984918857, + "loss": 0.1162, + "step": 11906 + }, + { + "epoch": 0.1033584777909914, + "grad_norm": 0.2451171875, + "learning_rate": 0.001942099346864449, + "loss": 0.1289, + "step": 11907 + }, + { + "epoch": 0.10336715827119557, + "grad_norm": 0.20703125, + "learning_rate": 0.0019420888436194206, + "loss": 0.1699, + "step": 11908 + }, + { + "epoch": 0.10337583875139973, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019420783394534969, + "loss": 0.1079, + "step": 11909 + }, + { + "epoch": 0.1033845192316039, + "grad_norm": 0.306640625, + "learning_rate": 0.0019420678343666884, + "loss": 0.1367, + "step": 11910 + }, + { + "epoch": 0.10339319971180806, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019420573283590076, + "loss": 0.1299, + "step": 11911 + }, + { + "epoch": 0.10340188019201223, + "grad_norm": 0.57421875, + "learning_rate": 0.0019420468214304653, + "loss": 0.1406, + "step": 11912 + }, + { + "epoch": 0.10341056067221639, + "grad_norm": 0.35546875, + "learning_rate": 0.0019420363135810735, + "loss": 0.1523, + "step": 11913 + }, + { + "epoch": 0.10341924115242056, + "grad_norm": 0.6953125, + "learning_rate": 0.001942025804810843, + "loss": 0.1187, + "step": 11914 + }, + { + "epoch": 0.10342792163262472, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019420152951197863, + "loss": 0.1035, + "step": 11915 + }, + { + "epoch": 0.10343660211282889, + "grad_norm": 0.28125, + "learning_rate": 0.0019420047845079141, + "loss": 0.0933, + "step": 11916 + }, + { + "epoch": 0.10344528259303305, + "grad_norm": 0.376953125, + "learning_rate": 0.0019419942729752382, + "loss": 0.1562, + "step": 11917 + }, + { + "epoch": 0.10345396307323722, + "grad_norm": 0.71875, + "learning_rate": 0.0019419837605217697, + "loss": 0.1055, + "step": 11918 + }, + { + "epoch": 0.10346264355344138, + "grad_norm": 0.142578125, + "learning_rate": 0.0019419732471475204, + "loss": 0.1221, + "step": 11919 + }, + { + "epoch": 0.10347132403364555, + "grad_norm": 0.86328125, + "learning_rate": 0.0019419627328525022, + "loss": 0.127, + "step": 11920 + }, + { + "epoch": 0.1034800045138497, + "grad_norm": 0.55078125, + "learning_rate": 0.0019419522176367262, + "loss": 0.1074, + "step": 11921 + }, + { + "epoch": 0.10348868499405386, + "grad_norm": 0.271484375, + "learning_rate": 0.0019419417015002034, + "loss": 0.1504, + "step": 11922 + }, + { + "epoch": 0.10349736547425803, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019419311844429462, + "loss": 0.1318, + "step": 11923 + }, + { + "epoch": 0.1035060459544622, + "grad_norm": 0.494140625, + "learning_rate": 0.0019419206664649657, + "loss": 0.1045, + "step": 11924 + }, + { + "epoch": 0.10351472643466636, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019419101475662733, + "loss": 0.1553, + "step": 11925 + }, + { + "epoch": 0.10352340691487052, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019418996277468807, + "loss": 0.1484, + "step": 11926 + }, + { + "epoch": 0.10353208739507469, + "grad_norm": 1.0625, + "learning_rate": 0.0019418891070067993, + "loss": 0.1099, + "step": 11927 + }, + { + "epoch": 0.10354076787527886, + "grad_norm": 1.3515625, + "learning_rate": 0.0019418785853460405, + "loss": 0.1836, + "step": 11928 + }, + { + "epoch": 0.10354944835548302, + "grad_norm": 0.28515625, + "learning_rate": 0.001941868062764616, + "loss": 0.1445, + "step": 11929 + }, + { + "epoch": 0.10355812883568719, + "grad_norm": 0.16015625, + "learning_rate": 0.0019418575392625374, + "loss": 0.1357, + "step": 11930 + }, + { + "epoch": 0.10356680931589135, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019418470148398158, + "loss": 0.1797, + "step": 11931 + }, + { + "epoch": 0.10357548979609552, + "grad_norm": 0.3359375, + "learning_rate": 0.0019418364894964628, + "loss": 0.0972, + "step": 11932 + }, + { + "epoch": 0.10358417027629968, + "grad_norm": 0.625, + "learning_rate": 0.0019418259632324904, + "loss": 0.1514, + "step": 11933 + }, + { + "epoch": 0.10359285075650385, + "grad_norm": 0.28125, + "learning_rate": 0.00194181543604791, + "loss": 0.1543, + "step": 11934 + }, + { + "epoch": 0.10360153123670801, + "grad_norm": 0.4921875, + "learning_rate": 0.0019418049079427322, + "loss": 0.1699, + "step": 11935 + }, + { + "epoch": 0.10361021171691218, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019417943789169698, + "loss": 0.1216, + "step": 11936 + }, + { + "epoch": 0.10361889219711634, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019417838489706335, + "loss": 0.126, + "step": 11937 + }, + { + "epoch": 0.1036275726773205, + "grad_norm": 0.11669921875, + "learning_rate": 0.001941773318103735, + "loss": 0.1289, + "step": 11938 + }, + { + "epoch": 0.10363625315752467, + "grad_norm": 0.4453125, + "learning_rate": 0.0019417627863162858, + "loss": 0.0986, + "step": 11939 + }, + { + "epoch": 0.10364493363772884, + "grad_norm": 0.25, + "learning_rate": 0.0019417522536082977, + "loss": 0.1367, + "step": 11940 + }, + { + "epoch": 0.103653614117933, + "grad_norm": 0.5078125, + "learning_rate": 0.0019417417199797816, + "loss": 0.1328, + "step": 11941 + }, + { + "epoch": 0.10366229459813717, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019417311854307495, + "loss": 0.125, + "step": 11942 + }, + { + "epoch": 0.10367097507834133, + "grad_norm": 0.318359375, + "learning_rate": 0.0019417206499612132, + "loss": 0.1162, + "step": 11943 + }, + { + "epoch": 0.1036796555585455, + "grad_norm": 0.306640625, + "learning_rate": 0.0019417101135711834, + "loss": 0.1226, + "step": 11944 + }, + { + "epoch": 0.10368833603874966, + "grad_norm": 0.341796875, + "learning_rate": 0.0019416995762606725, + "loss": 0.1006, + "step": 11945 + }, + { + "epoch": 0.10369701651895383, + "grad_norm": 1.4375, + "learning_rate": 0.0019416890380296913, + "loss": 0.4785, + "step": 11946 + }, + { + "epoch": 0.10370569699915799, + "grad_norm": 1.1484375, + "learning_rate": 0.0019416784988782517, + "loss": 0.1289, + "step": 11947 + }, + { + "epoch": 0.10371437747936216, + "grad_norm": 0.19921875, + "learning_rate": 0.001941667958806365, + "loss": 0.104, + "step": 11948 + }, + { + "epoch": 0.10372305795956632, + "grad_norm": 0.2890625, + "learning_rate": 0.001941657417814043, + "loss": 0.0938, + "step": 11949 + }, + { + "epoch": 0.10373173843977049, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019416468759012968, + "loss": 0.0698, + "step": 11950 + }, + { + "epoch": 0.10374041891997465, + "grad_norm": 0.59375, + "learning_rate": 0.0019416363330681387, + "loss": 0.1455, + "step": 11951 + }, + { + "epoch": 0.10374909940017882, + "grad_norm": 0.32421875, + "learning_rate": 0.0019416257893145794, + "loss": 0.1523, + "step": 11952 + }, + { + "epoch": 0.10375777988038298, + "grad_norm": 0.8671875, + "learning_rate": 0.001941615244640631, + "loss": 0.1221, + "step": 11953 + }, + { + "epoch": 0.10376646036058715, + "grad_norm": 0.421875, + "learning_rate": 0.0019416046990463048, + "loss": 0.1006, + "step": 11954 + }, + { + "epoch": 0.10377514084079131, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019415941525316124, + "loss": 0.1182, + "step": 11955 + }, + { + "epoch": 0.10378382132099548, + "grad_norm": 1.171875, + "learning_rate": 0.0019415836050965653, + "loss": 0.1475, + "step": 11956 + }, + { + "epoch": 0.10379250180119964, + "grad_norm": 0.255859375, + "learning_rate": 0.0019415730567411745, + "loss": 0.1338, + "step": 11957 + }, + { + "epoch": 0.10380118228140381, + "grad_norm": 0.70703125, + "learning_rate": 0.0019415625074654527, + "loss": 0.1162, + "step": 11958 + }, + { + "epoch": 0.10380986276160797, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019415519572694104, + "loss": 0.1152, + "step": 11959 + }, + { + "epoch": 0.10381854324181214, + "grad_norm": 0.52734375, + "learning_rate": 0.0019415414061530597, + "loss": 0.1406, + "step": 11960 + }, + { + "epoch": 0.1038272237220163, + "grad_norm": 0.275390625, + "learning_rate": 0.0019415308541164114, + "loss": 0.1299, + "step": 11961 + }, + { + "epoch": 0.10383590420222047, + "grad_norm": 0.396484375, + "learning_rate": 0.0019415203011594784, + "loss": 0.1289, + "step": 11962 + }, + { + "epoch": 0.10384458468242463, + "grad_norm": 0.1708984375, + "learning_rate": 0.001941509747282271, + "loss": 0.1357, + "step": 11963 + }, + { + "epoch": 0.1038532651626288, + "grad_norm": 0.56640625, + "learning_rate": 0.0019414991924848013, + "loss": 0.126, + "step": 11964 + }, + { + "epoch": 0.10386194564283296, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019414886367670808, + "loss": 0.1055, + "step": 11965 + }, + { + "epoch": 0.10387062612303713, + "grad_norm": 0.47265625, + "learning_rate": 0.001941478080129121, + "loss": 0.1504, + "step": 11966 + }, + { + "epoch": 0.1038793066032413, + "grad_norm": 0.107421875, + "learning_rate": 0.0019414675225709332, + "loss": 0.0977, + "step": 11967 + }, + { + "epoch": 0.10388798708344546, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019414569640925292, + "loss": 0.0688, + "step": 11968 + }, + { + "epoch": 0.10389666756364963, + "grad_norm": 0.185546875, + "learning_rate": 0.0019414464046939206, + "loss": 0.1187, + "step": 11969 + }, + { + "epoch": 0.10390534804385379, + "grad_norm": 0.51171875, + "learning_rate": 0.001941435844375119, + "loss": 0.1357, + "step": 11970 + }, + { + "epoch": 0.10391402852405796, + "grad_norm": 0.3125, + "learning_rate": 0.0019414252831361357, + "loss": 0.126, + "step": 11971 + }, + { + "epoch": 0.10392270900426212, + "grad_norm": 0.482421875, + "learning_rate": 0.0019414147209769821, + "loss": 0.1465, + "step": 11972 + }, + { + "epoch": 0.10393138948446629, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019414041578976705, + "loss": 0.1011, + "step": 11973 + }, + { + "epoch": 0.10394006996467045, + "grad_norm": 0.291015625, + "learning_rate": 0.0019413935938982117, + "loss": 0.1426, + "step": 11974 + }, + { + "epoch": 0.10394875044487462, + "grad_norm": 0.1328125, + "learning_rate": 0.0019413830289786171, + "loss": 0.1079, + "step": 11975 + }, + { + "epoch": 0.10395743092507878, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019413724631388994, + "loss": 0.123, + "step": 11976 + }, + { + "epoch": 0.10396611140528295, + "grad_norm": 0.38671875, + "learning_rate": 0.0019413618963790688, + "loss": 0.1182, + "step": 11977 + }, + { + "epoch": 0.10397479188548711, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019413513286991378, + "loss": 0.1001, + "step": 11978 + }, + { + "epoch": 0.10398347236569128, + "grad_norm": 0.369140625, + "learning_rate": 0.0019413407600991175, + "loss": 0.1387, + "step": 11979 + }, + { + "epoch": 0.10399215284589544, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019413301905790196, + "loss": 0.1475, + "step": 11980 + }, + { + "epoch": 0.1040008333260996, + "grad_norm": 2.765625, + "learning_rate": 0.0019413196201388556, + "loss": 0.3945, + "step": 11981 + }, + { + "epoch": 0.10400951380630377, + "grad_norm": 0.14453125, + "learning_rate": 0.0019413090487786372, + "loss": 0.0908, + "step": 11982 + }, + { + "epoch": 0.10401819428650792, + "grad_norm": 0.64453125, + "learning_rate": 0.0019412984764983759, + "loss": 0.1406, + "step": 11983 + }, + { + "epoch": 0.10402687476671209, + "grad_norm": 0.166015625, + "learning_rate": 0.0019412879032980833, + "loss": 0.1055, + "step": 11984 + }, + { + "epoch": 0.10403555524691625, + "grad_norm": 0.09765625, + "learning_rate": 0.0019412773291777705, + "loss": 0.1147, + "step": 11985 + }, + { + "epoch": 0.10404423572712042, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019412667541374498, + "loss": 0.1099, + "step": 11986 + }, + { + "epoch": 0.10405291620732458, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019412561781771323, + "loss": 0.1416, + "step": 11987 + }, + { + "epoch": 0.10406159668752875, + "grad_norm": 0.06884765625, + "learning_rate": 0.0019412456012968301, + "loss": 0.0879, + "step": 11988 + }, + { + "epoch": 0.10407027716773291, + "grad_norm": 0.2734375, + "learning_rate": 0.0019412350234965536, + "loss": 0.168, + "step": 11989 + }, + { + "epoch": 0.10407895764793708, + "grad_norm": 0.21875, + "learning_rate": 0.0019412244447763157, + "loss": 0.0942, + "step": 11990 + }, + { + "epoch": 0.10408763812814124, + "grad_norm": 0.498046875, + "learning_rate": 0.0019412138651361268, + "loss": 0.1357, + "step": 11991 + }, + { + "epoch": 0.10409631860834541, + "grad_norm": 0.427734375, + "learning_rate": 0.0019412032845759998, + "loss": 0.1406, + "step": 11992 + }, + { + "epoch": 0.10410499908854957, + "grad_norm": 0.5234375, + "learning_rate": 0.0019411927030959452, + "loss": 0.1211, + "step": 11993 + }, + { + "epoch": 0.10411367956875374, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019411821206959745, + "loss": 0.1533, + "step": 11994 + }, + { + "epoch": 0.1041223600489579, + "grad_norm": 0.265625, + "learning_rate": 0.0019411715373761, + "loss": 0.1133, + "step": 11995 + }, + { + "epoch": 0.10413104052916207, + "grad_norm": 0.173828125, + "learning_rate": 0.001941160953136333, + "loss": 0.0928, + "step": 11996 + }, + { + "epoch": 0.10413972100936623, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019411503679766851, + "loss": 0.125, + "step": 11997 + }, + { + "epoch": 0.1041484014895704, + "grad_norm": 0.068359375, + "learning_rate": 0.0019411397818971676, + "loss": 0.0796, + "step": 11998 + }, + { + "epoch": 0.10415708196977456, + "grad_norm": 0.466796875, + "learning_rate": 0.0019411291948977923, + "loss": 0.1143, + "step": 11999 + }, + { + "epoch": 0.10416576244997873, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019411186069785708, + "loss": 0.1123, + "step": 12000 + }, + { + "epoch": 0.1041744429301829, + "grad_norm": 0.337890625, + "learning_rate": 0.0019411080181395148, + "loss": 0.1006, + "step": 12001 + }, + { + "epoch": 0.10418312341038706, + "grad_norm": 0.419921875, + "learning_rate": 0.0019410974283806356, + "loss": 0.1108, + "step": 12002 + }, + { + "epoch": 0.10419180389059123, + "grad_norm": 1.0859375, + "learning_rate": 0.0019410868377019448, + "loss": 0.1875, + "step": 12003 + }, + { + "epoch": 0.10420048437079539, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019410762461034542, + "loss": 0.1396, + "step": 12004 + }, + { + "epoch": 0.10420916485099956, + "grad_norm": 1.1171875, + "learning_rate": 0.001941065653585175, + "loss": 0.1113, + "step": 12005 + }, + { + "epoch": 0.10421784533120372, + "grad_norm": 0.76953125, + "learning_rate": 0.0019410550601471194, + "loss": 0.1426, + "step": 12006 + }, + { + "epoch": 0.10422652581140789, + "grad_norm": 0.2890625, + "learning_rate": 0.0019410444657892983, + "loss": 0.1289, + "step": 12007 + }, + { + "epoch": 0.10423520629161205, + "grad_norm": 0.53125, + "learning_rate": 0.0019410338705117237, + "loss": 0.1191, + "step": 12008 + }, + { + "epoch": 0.10424388677181622, + "grad_norm": 0.32421875, + "learning_rate": 0.0019410232743144072, + "loss": 0.1309, + "step": 12009 + }, + { + "epoch": 0.10425256725202038, + "grad_norm": 0.4375, + "learning_rate": 0.0019410126771973602, + "loss": 0.1641, + "step": 12010 + }, + { + "epoch": 0.10426124773222455, + "grad_norm": 0.31640625, + "learning_rate": 0.0019410020791605941, + "loss": 0.1104, + "step": 12011 + }, + { + "epoch": 0.10426992821242871, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019409914802041212, + "loss": 0.1699, + "step": 12012 + }, + { + "epoch": 0.10427860869263288, + "grad_norm": 0.30078125, + "learning_rate": 0.0019409808803279526, + "loss": 0.1016, + "step": 12013 + }, + { + "epoch": 0.10428728917283704, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019409702795320995, + "loss": 0.125, + "step": 12014 + }, + { + "epoch": 0.1042959696530412, + "grad_norm": 0.162109375, + "learning_rate": 0.0019409596778165745, + "loss": 0.125, + "step": 12015 + }, + { + "epoch": 0.10430465013324537, + "grad_norm": 0.51953125, + "learning_rate": 0.0019409490751813882, + "loss": 0.1455, + "step": 12016 + }, + { + "epoch": 0.10431333061344954, + "grad_norm": 0.2119140625, + "learning_rate": 0.001940938471626553, + "loss": 0.0991, + "step": 12017 + }, + { + "epoch": 0.1043220110936537, + "grad_norm": 0.310546875, + "learning_rate": 0.00194092786715208, + "loss": 0.1299, + "step": 12018 + }, + { + "epoch": 0.10433069157385787, + "grad_norm": 0.48828125, + "learning_rate": 0.0019409172617579809, + "loss": 0.1904, + "step": 12019 + }, + { + "epoch": 0.10433937205406203, + "grad_norm": 0.63671875, + "learning_rate": 0.0019409066554442673, + "loss": 0.1221, + "step": 12020 + }, + { + "epoch": 0.1043480525342662, + "grad_norm": 0.0830078125, + "learning_rate": 0.001940896048210951, + "loss": 0.1045, + "step": 12021 + }, + { + "epoch": 0.10435673301447036, + "grad_norm": 0.73828125, + "learning_rate": 0.0019408854400580434, + "loss": 0.1201, + "step": 12022 + }, + { + "epoch": 0.10436541349467453, + "grad_norm": 0.57421875, + "learning_rate": 0.0019408748309855555, + "loss": 0.1445, + "step": 12023 + }, + { + "epoch": 0.10437409397487869, + "grad_norm": 0.0859375, + "learning_rate": 0.0019408642209935002, + "loss": 0.1416, + "step": 12024 + }, + { + "epoch": 0.10438277445508286, + "grad_norm": 0.14453125, + "learning_rate": 0.0019408536100818882, + "loss": 0.1445, + "step": 12025 + }, + { + "epoch": 0.10439145493528702, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019408429982507311, + "loss": 0.127, + "step": 12026 + }, + { + "epoch": 0.10440013541549119, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019408323855000412, + "loss": 0.126, + "step": 12027 + }, + { + "epoch": 0.10440881589569535, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019408217718298293, + "loss": 0.0967, + "step": 12028 + }, + { + "epoch": 0.10441749637589952, + "grad_norm": 0.69140625, + "learning_rate": 0.0019408111572401072, + "loss": 0.1113, + "step": 12029 + }, + { + "epoch": 0.10442617685610368, + "grad_norm": 0.19921875, + "learning_rate": 0.0019408005417308868, + "loss": 0.1094, + "step": 12030 + }, + { + "epoch": 0.10443485733630785, + "grad_norm": 0.72265625, + "learning_rate": 0.00194078992530218, + "loss": 0.1709, + "step": 12031 + }, + { + "epoch": 0.10444353781651201, + "grad_norm": 0.458984375, + "learning_rate": 0.0019407793079539975, + "loss": 0.1182, + "step": 12032 + }, + { + "epoch": 0.10445221829671618, + "grad_norm": 0.3515625, + "learning_rate": 0.0019407686896863515, + "loss": 0.1494, + "step": 12033 + }, + { + "epoch": 0.10446089877692034, + "grad_norm": 0.408203125, + "learning_rate": 0.0019407580704992532, + "loss": 0.1035, + "step": 12034 + }, + { + "epoch": 0.10446957925712451, + "grad_norm": 0.423828125, + "learning_rate": 0.0019407474503927151, + "loss": 0.1318, + "step": 12035 + }, + { + "epoch": 0.10447825973732867, + "grad_norm": 0.5234375, + "learning_rate": 0.001940736829366748, + "loss": 0.1191, + "step": 12036 + }, + { + "epoch": 0.10448694021753284, + "grad_norm": 0.134765625, + "learning_rate": 0.0019407262074213633, + "loss": 0.1396, + "step": 12037 + }, + { + "epoch": 0.104495620697737, + "grad_norm": 0.14453125, + "learning_rate": 0.0019407155845565737, + "loss": 0.126, + "step": 12038 + }, + { + "epoch": 0.10450430117794117, + "grad_norm": 0.291015625, + "learning_rate": 0.0019407049607723897, + "loss": 0.1582, + "step": 12039 + }, + { + "epoch": 0.10451298165814533, + "grad_norm": 0.2421875, + "learning_rate": 0.0019406943360688236, + "loss": 0.127, + "step": 12040 + }, + { + "epoch": 0.1045216621383495, + "grad_norm": 0.44921875, + "learning_rate": 0.0019406837104458867, + "loss": 0.1152, + "step": 12041 + }, + { + "epoch": 0.10453034261855366, + "grad_norm": 0.5546875, + "learning_rate": 0.0019406730839035906, + "loss": 0.1436, + "step": 12042 + }, + { + "epoch": 0.10453902309875783, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019406624564419475, + "loss": 0.1035, + "step": 12043 + }, + { + "epoch": 0.10454770357896198, + "grad_norm": 0.609375, + "learning_rate": 0.0019406518280609684, + "loss": 0.1138, + "step": 12044 + }, + { + "epoch": 0.10455638405916615, + "grad_norm": 0.0595703125, + "learning_rate": 0.0019406411987606647, + "loss": 0.106, + "step": 12045 + }, + { + "epoch": 0.10456506453937031, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019406305685410487, + "loss": 0.0996, + "step": 12046 + }, + { + "epoch": 0.10457374501957448, + "grad_norm": 0.384765625, + "learning_rate": 0.0019406199374021318, + "loss": 0.1299, + "step": 12047 + }, + { + "epoch": 0.10458242549977864, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019406093053439255, + "loss": 0.106, + "step": 12048 + }, + { + "epoch": 0.1045911059799828, + "grad_norm": 0.34765625, + "learning_rate": 0.0019405986723664416, + "loss": 0.1338, + "step": 12049 + }, + { + "epoch": 0.10459978646018697, + "grad_norm": 0.50390625, + "learning_rate": 0.0019405880384696914, + "loss": 0.0942, + "step": 12050 + }, + { + "epoch": 0.10460846694039114, + "grad_norm": 0.416015625, + "learning_rate": 0.001940577403653687, + "loss": 0.1318, + "step": 12051 + }, + { + "epoch": 0.1046171474205953, + "grad_norm": 0.6328125, + "learning_rate": 0.0019405667679184395, + "loss": 0.1045, + "step": 12052 + }, + { + "epoch": 0.10462582790079947, + "grad_norm": 0.14453125, + "learning_rate": 0.0019405561312639613, + "loss": 0.0728, + "step": 12053 + }, + { + "epoch": 0.10463450838100363, + "grad_norm": 0.5234375, + "learning_rate": 0.0019405454936902628, + "loss": 0.1162, + "step": 12054 + }, + { + "epoch": 0.1046431888612078, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019405348551973572, + "loss": 0.1084, + "step": 12055 + }, + { + "epoch": 0.10465186934141196, + "grad_norm": 0.96875, + "learning_rate": 0.0019405242157852548, + "loss": 0.1152, + "step": 12056 + }, + { + "epoch": 0.10466054982161613, + "grad_norm": 1.0, + "learning_rate": 0.0019405135754539677, + "loss": 0.1143, + "step": 12057 + }, + { + "epoch": 0.10466923030182029, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019405029342035077, + "loss": 0.0815, + "step": 12058 + }, + { + "epoch": 0.10467791078202446, + "grad_norm": 0.2421875, + "learning_rate": 0.0019404922920338865, + "loss": 0.125, + "step": 12059 + }, + { + "epoch": 0.10468659126222862, + "grad_norm": 0.462890625, + "learning_rate": 0.0019404816489451153, + "loss": 0.1104, + "step": 12060 + }, + { + "epoch": 0.10469527174243279, + "grad_norm": 0.302734375, + "learning_rate": 0.001940471004937206, + "loss": 0.1592, + "step": 12061 + }, + { + "epoch": 0.10470395222263695, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019404603600101705, + "loss": 0.1504, + "step": 12062 + }, + { + "epoch": 0.10471263270284112, + "grad_norm": 0.1669921875, + "learning_rate": 0.00194044971416402, + "loss": 0.166, + "step": 12063 + }, + { + "epoch": 0.10472131318304528, + "grad_norm": 1.2109375, + "learning_rate": 0.0019404390673987664, + "loss": 0.1357, + "step": 12064 + }, + { + "epoch": 0.10472999366324945, + "grad_norm": 0.24609375, + "learning_rate": 0.001940428419714421, + "loss": 0.1143, + "step": 12065 + }, + { + "epoch": 0.10473867414345361, + "grad_norm": 4.3125, + "learning_rate": 0.001940417771110996, + "loss": 0.4766, + "step": 12066 + }, + { + "epoch": 0.10474735462365778, + "grad_norm": 0.3515625, + "learning_rate": 0.0019404071215885028, + "loss": 0.1245, + "step": 12067 + }, + { + "epoch": 0.10475603510386194, + "grad_norm": 0.36328125, + "learning_rate": 0.0019403964711469527, + "loss": 0.125, + "step": 12068 + }, + { + "epoch": 0.10476471558406611, + "grad_norm": 0.265625, + "learning_rate": 0.001940385819786358, + "loss": 0.1016, + "step": 12069 + }, + { + "epoch": 0.10477339606427027, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019403751675067297, + "loss": 0.0981, + "step": 12070 + }, + { + "epoch": 0.10478207654447444, + "grad_norm": 0.171875, + "learning_rate": 0.0019403645143080798, + "loss": 0.2598, + "step": 12071 + }, + { + "epoch": 0.1047907570246786, + "grad_norm": 0.0966796875, + "learning_rate": 0.00194035386019042, + "loss": 0.1318, + "step": 12072 + }, + { + "epoch": 0.10479943750488277, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019403432051537618, + "loss": 0.0923, + "step": 12073 + }, + { + "epoch": 0.10480811798508693, + "grad_norm": 0.34765625, + "learning_rate": 0.001940332549198117, + "loss": 0.1035, + "step": 12074 + }, + { + "epoch": 0.1048167984652911, + "grad_norm": 0.7578125, + "learning_rate": 0.001940321892323497, + "loss": 0.126, + "step": 12075 + }, + { + "epoch": 0.10482547894549527, + "grad_norm": 0.59375, + "learning_rate": 0.0019403112345299136, + "loss": 0.1177, + "step": 12076 + }, + { + "epoch": 0.10483415942569943, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019403005758173784, + "loss": 0.1074, + "step": 12077 + }, + { + "epoch": 0.1048428399059036, + "grad_norm": 0.8203125, + "learning_rate": 0.0019402899161859034, + "loss": 0.124, + "step": 12078 + }, + { + "epoch": 0.10485152038610776, + "grad_norm": 0.216796875, + "learning_rate": 0.0019402792556354996, + "loss": 0.1094, + "step": 12079 + }, + { + "epoch": 0.10486020086631193, + "grad_norm": 0.23828125, + "learning_rate": 0.0019402685941661793, + "loss": 0.1172, + "step": 12080 + }, + { + "epoch": 0.10486888134651609, + "grad_norm": 0.482421875, + "learning_rate": 0.0019402579317779535, + "loss": 0.0986, + "step": 12081 + }, + { + "epoch": 0.10487756182672026, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019402472684708347, + "loss": 0.1553, + "step": 12082 + }, + { + "epoch": 0.10488624230692442, + "grad_norm": 0.103515625, + "learning_rate": 0.001940236604244834, + "loss": 0.0986, + "step": 12083 + }, + { + "epoch": 0.10489492278712859, + "grad_norm": 0.2158203125, + "learning_rate": 0.001940225939099963, + "loss": 0.1621, + "step": 12084 + }, + { + "epoch": 0.10490360326733275, + "grad_norm": 0.08154296875, + "learning_rate": 0.001940215273036234, + "loss": 0.1118, + "step": 12085 + }, + { + "epoch": 0.10491228374753692, + "grad_norm": 0.494140625, + "learning_rate": 0.0019402046060536578, + "loss": 0.1025, + "step": 12086 + }, + { + "epoch": 0.10492096422774108, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019401939381522466, + "loss": 0.1045, + "step": 12087 + }, + { + "epoch": 0.10492964470794525, + "grad_norm": 0.2890625, + "learning_rate": 0.0019401832693320118, + "loss": 0.1494, + "step": 12088 + }, + { + "epoch": 0.10493832518814941, + "grad_norm": 0.423828125, + "learning_rate": 0.0019401725995929655, + "loss": 0.1138, + "step": 12089 + }, + { + "epoch": 0.10494700566835358, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019401619289351184, + "loss": 0.1133, + "step": 12090 + }, + { + "epoch": 0.10495568614855774, + "grad_norm": 0.45703125, + "learning_rate": 0.0019401512573584837, + "loss": 0.166, + "step": 12091 + }, + { + "epoch": 0.10496436662876191, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019401405848630715, + "loss": 0.1104, + "step": 12092 + }, + { + "epoch": 0.10497304710896607, + "grad_norm": 0.359375, + "learning_rate": 0.0019401299114488945, + "loss": 0.1484, + "step": 12093 + }, + { + "epoch": 0.10498172758917024, + "grad_norm": 0.470703125, + "learning_rate": 0.0019401192371159644, + "loss": 0.1113, + "step": 12094 + }, + { + "epoch": 0.1049904080693744, + "grad_norm": 0.09814453125, + "learning_rate": 0.001940108561864292, + "loss": 0.1699, + "step": 12095 + }, + { + "epoch": 0.10499908854957857, + "grad_norm": 0.34765625, + "learning_rate": 0.0019400978856938895, + "loss": 0.1396, + "step": 12096 + }, + { + "epoch": 0.10500776902978273, + "grad_norm": 0.119140625, + "learning_rate": 0.0019400872086047687, + "loss": 0.1719, + "step": 12097 + }, + { + "epoch": 0.1050164495099869, + "grad_norm": 0.140625, + "learning_rate": 0.0019400765305969412, + "loss": 0.1162, + "step": 12098 + }, + { + "epoch": 0.10502512999019106, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019400658516704186, + "loss": 0.1445, + "step": 12099 + }, + { + "epoch": 0.10503381047039523, + "grad_norm": 0.26953125, + "learning_rate": 0.0019400551718252127, + "loss": 0.1445, + "step": 12100 + }, + { + "epoch": 0.1050424909505994, + "grad_norm": 0.140625, + "learning_rate": 0.0019400444910613348, + "loss": 0.1113, + "step": 12101 + }, + { + "epoch": 0.10505117143080356, + "grad_norm": 0.365234375, + "learning_rate": 0.0019400338093787974, + "loss": 0.1826, + "step": 12102 + }, + { + "epoch": 0.10505985191100772, + "grad_norm": 0.7109375, + "learning_rate": 0.0019400231267776111, + "loss": 0.1279, + "step": 12103 + }, + { + "epoch": 0.10506853239121189, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019400124432577884, + "loss": 0.1182, + "step": 12104 + }, + { + "epoch": 0.10507721287141605, + "grad_norm": 0.25, + "learning_rate": 0.0019400017588193407, + "loss": 0.1387, + "step": 12105 + }, + { + "epoch": 0.1050858933516202, + "grad_norm": 0.251953125, + "learning_rate": 0.0019399910734622797, + "loss": 0.1143, + "step": 12106 + }, + { + "epoch": 0.10509457383182437, + "grad_norm": 0.55078125, + "learning_rate": 0.0019399803871866168, + "loss": 0.1553, + "step": 12107 + }, + { + "epoch": 0.10510325431202854, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019399696999923644, + "loss": 0.1025, + "step": 12108 + }, + { + "epoch": 0.1051119347922327, + "grad_norm": 0.298828125, + "learning_rate": 0.0019399590118795336, + "loss": 0.1089, + "step": 12109 + }, + { + "epoch": 0.10512061527243687, + "grad_norm": 0.4609375, + "learning_rate": 0.001939948322848136, + "loss": 0.1289, + "step": 12110 + }, + { + "epoch": 0.10512929575264103, + "grad_norm": 1.0078125, + "learning_rate": 0.0019399376328981838, + "loss": 0.1787, + "step": 12111 + }, + { + "epoch": 0.1051379762328452, + "grad_norm": 0.36328125, + "learning_rate": 0.0019399269420296884, + "loss": 0.1074, + "step": 12112 + }, + { + "epoch": 0.10514665671304936, + "grad_norm": 0.26953125, + "learning_rate": 0.0019399162502426615, + "loss": 0.1172, + "step": 12113 + }, + { + "epoch": 0.10515533719325353, + "grad_norm": 0.13671875, + "learning_rate": 0.001939905557537115, + "loss": 0.1211, + "step": 12114 + }, + { + "epoch": 0.10516401767345769, + "grad_norm": 0.63671875, + "learning_rate": 0.0019398948639130597, + "loss": 0.125, + "step": 12115 + }, + { + "epoch": 0.10517269815366186, + "grad_norm": 0.322265625, + "learning_rate": 0.0019398841693705087, + "loss": 0.1279, + "step": 12116 + }, + { + "epoch": 0.10518137863386602, + "grad_norm": 0.3203125, + "learning_rate": 0.0019398734739094728, + "loss": 0.1289, + "step": 12117 + }, + { + "epoch": 0.10519005911407019, + "grad_norm": 0.2080078125, + "learning_rate": 0.001939862777529964, + "loss": 0.127, + "step": 12118 + }, + { + "epoch": 0.10519873959427435, + "grad_norm": 0.23828125, + "learning_rate": 0.0019398520802319935, + "loss": 0.1777, + "step": 12119 + }, + { + "epoch": 0.10520742007447852, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019398413820155738, + "loss": 0.1484, + "step": 12120 + }, + { + "epoch": 0.10521610055468268, + "grad_norm": 0.11083984375, + "learning_rate": 0.001939830682880716, + "loss": 0.1318, + "step": 12121 + }, + { + "epoch": 0.10522478103488685, + "grad_norm": 0.11865234375, + "learning_rate": 0.001939819982827432, + "loss": 0.1289, + "step": 12122 + }, + { + "epoch": 0.10523346151509101, + "grad_norm": 0.177734375, + "learning_rate": 0.0019398092818557337, + "loss": 0.1182, + "step": 12123 + }, + { + "epoch": 0.10524214199529518, + "grad_norm": 0.396484375, + "learning_rate": 0.0019397985799656321, + "loss": 0.1445, + "step": 12124 + }, + { + "epoch": 0.10525082247549934, + "grad_norm": 0.46484375, + "learning_rate": 0.0019397878771571398, + "loss": 0.1406, + "step": 12125 + }, + { + "epoch": 0.10525950295570351, + "grad_norm": 0.498046875, + "learning_rate": 0.001939777173430268, + "loss": 0.1611, + "step": 12126 + }, + { + "epoch": 0.10526818343590767, + "grad_norm": 0.427734375, + "learning_rate": 0.0019397664687850283, + "loss": 0.1118, + "step": 12127 + }, + { + "epoch": 0.10527686391611184, + "grad_norm": 0.26171875, + "learning_rate": 0.0019397557632214327, + "loss": 0.0889, + "step": 12128 + }, + { + "epoch": 0.105285544396316, + "grad_norm": 0.1025390625, + "learning_rate": 0.001939745056739493, + "loss": 0.1582, + "step": 12129 + }, + { + "epoch": 0.10529422487652017, + "grad_norm": 0.5703125, + "learning_rate": 0.0019397343493392206, + "loss": 0.125, + "step": 12130 + }, + { + "epoch": 0.10530290535672433, + "grad_norm": 0.64453125, + "learning_rate": 0.0019397236410206274, + "loss": 0.1406, + "step": 12131 + }, + { + "epoch": 0.1053115858369285, + "grad_norm": 0.30859375, + "learning_rate": 0.001939712931783725, + "loss": 0.1553, + "step": 12132 + }, + { + "epoch": 0.10532026631713266, + "grad_norm": 0.095703125, + "learning_rate": 0.001939702221628525, + "loss": 0.0762, + "step": 12133 + }, + { + "epoch": 0.10532894679733683, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019396915105550395, + "loss": 0.1309, + "step": 12134 + }, + { + "epoch": 0.105337627277541, + "grad_norm": 0.26953125, + "learning_rate": 0.0019396807985632798, + "loss": 0.1128, + "step": 12135 + }, + { + "epoch": 0.10534630775774516, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019396700856532578, + "loss": 0.123, + "step": 12136 + }, + { + "epoch": 0.10535498823794932, + "grad_norm": 0.70703125, + "learning_rate": 0.0019396593718249853, + "loss": 0.0977, + "step": 12137 + }, + { + "epoch": 0.10536366871815349, + "grad_norm": 0.201171875, + "learning_rate": 0.0019396486570784739, + "loss": 0.0674, + "step": 12138 + }, + { + "epoch": 0.10537234919835765, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019396379414137357, + "loss": 0.0991, + "step": 12139 + }, + { + "epoch": 0.10538102967856182, + "grad_norm": 0.107421875, + "learning_rate": 0.0019396272248307814, + "loss": 0.1084, + "step": 12140 + }, + { + "epoch": 0.10538971015876598, + "grad_norm": 0.201171875, + "learning_rate": 0.001939616507329624, + "loss": 0.1348, + "step": 12141 + }, + { + "epoch": 0.10539839063897015, + "grad_norm": 0.345703125, + "learning_rate": 0.001939605788910274, + "loss": 0.1221, + "step": 12142 + }, + { + "epoch": 0.10540707111917431, + "grad_norm": 0.384765625, + "learning_rate": 0.0019395950695727442, + "loss": 0.1104, + "step": 12143 + }, + { + "epoch": 0.10541575159937848, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019395843493170456, + "loss": 0.207, + "step": 12144 + }, + { + "epoch": 0.10542443207958264, + "grad_norm": 0.5625, + "learning_rate": 0.0019395736281431904, + "loss": 0.1201, + "step": 12145 + }, + { + "epoch": 0.10543311255978681, + "grad_norm": 0.21875, + "learning_rate": 0.00193956290605119, + "loss": 0.1777, + "step": 12146 + }, + { + "epoch": 0.10544179303999097, + "grad_norm": 0.92578125, + "learning_rate": 0.001939552183041056, + "loss": 0.1016, + "step": 12147 + }, + { + "epoch": 0.10545047352019514, + "grad_norm": 0.15234375, + "learning_rate": 0.0019395414591128006, + "loss": 0.1123, + "step": 12148 + }, + { + "epoch": 0.1054591540003993, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019395307342664353, + "loss": 0.1143, + "step": 12149 + }, + { + "epoch": 0.10546783448060347, + "grad_norm": 0.5546875, + "learning_rate": 0.0019395200085019716, + "loss": 0.0986, + "step": 12150 + }, + { + "epoch": 0.10547651496080764, + "grad_norm": 0.412109375, + "learning_rate": 0.0019395092818194214, + "loss": 0.1089, + "step": 12151 + }, + { + "epoch": 0.1054851954410118, + "grad_norm": 0.8828125, + "learning_rate": 0.0019394985542187965, + "loss": 0.1157, + "step": 12152 + }, + { + "epoch": 0.10549387592121597, + "grad_norm": 0.158203125, + "learning_rate": 0.0019394878257001087, + "loss": 0.0977, + "step": 12153 + }, + { + "epoch": 0.10550255640142013, + "grad_norm": 0.326171875, + "learning_rate": 0.0019394770962633697, + "loss": 0.084, + "step": 12154 + }, + { + "epoch": 0.1055112368816243, + "grad_norm": 0.52734375, + "learning_rate": 0.001939466365908591, + "loss": 0.0884, + "step": 12155 + }, + { + "epoch": 0.10551991736182846, + "grad_norm": 0.171875, + "learning_rate": 0.0019394556346357845, + "loss": 0.1514, + "step": 12156 + }, + { + "epoch": 0.10552859784203263, + "grad_norm": 0.404296875, + "learning_rate": 0.001939444902444962, + "loss": 0.127, + "step": 12157 + }, + { + "epoch": 0.10553727832223679, + "grad_norm": 0.302734375, + "learning_rate": 0.0019394341693361353, + "loss": 0.0962, + "step": 12158 + }, + { + "epoch": 0.10554595880244096, + "grad_norm": 0.6953125, + "learning_rate": 0.001939423435309316, + "loss": 0.1465, + "step": 12159 + }, + { + "epoch": 0.10555463928264512, + "grad_norm": 0.384765625, + "learning_rate": 0.0019394127003645157, + "loss": 0.1172, + "step": 12160 + }, + { + "epoch": 0.10556331976284929, + "grad_norm": 0.07958984375, + "learning_rate": 0.001939401964501746, + "loss": 0.0991, + "step": 12161 + }, + { + "epoch": 0.10557200024305345, + "grad_norm": 0.1796875, + "learning_rate": 0.0019393912277210193, + "loss": 0.1133, + "step": 12162 + }, + { + "epoch": 0.10558068072325762, + "grad_norm": 0.333984375, + "learning_rate": 0.0019393804900223468, + "loss": 0.1797, + "step": 12163 + }, + { + "epoch": 0.10558936120346178, + "grad_norm": 0.48828125, + "learning_rate": 0.0019393697514057404, + "loss": 0.1201, + "step": 12164 + }, + { + "epoch": 0.10559804168366595, + "grad_norm": 0.064453125, + "learning_rate": 0.001939359011871212, + "loss": 0.0859, + "step": 12165 + }, + { + "epoch": 0.10560672216387011, + "grad_norm": 0.150390625, + "learning_rate": 0.0019393482714187731, + "loss": 0.1318, + "step": 12166 + }, + { + "epoch": 0.10561540264407426, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019393375300484359, + "loss": 0.0996, + "step": 12167 + }, + { + "epoch": 0.10562408312427843, + "grad_norm": 0.15625, + "learning_rate": 0.0019393267877602111, + "loss": 0.1191, + "step": 12168 + }, + { + "epoch": 0.1056327636044826, + "grad_norm": 0.345703125, + "learning_rate": 0.0019393160445541119, + "loss": 0.1426, + "step": 12169 + }, + { + "epoch": 0.10564144408468676, + "grad_norm": 0.103515625, + "learning_rate": 0.0019393053004301485, + "loss": 0.1328, + "step": 12170 + }, + { + "epoch": 0.10565012456489092, + "grad_norm": 0.134765625, + "learning_rate": 0.001939294555388334, + "loss": 0.1045, + "step": 12171 + }, + { + "epoch": 0.10565880504509509, + "grad_norm": 0.240234375, + "learning_rate": 0.0019392838094286792, + "loss": 0.1064, + "step": 12172 + }, + { + "epoch": 0.10566748552529925, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019392730625511965, + "loss": 0.1123, + "step": 12173 + }, + { + "epoch": 0.10567616600550342, + "grad_norm": 0.228515625, + "learning_rate": 0.0019392623147558978, + "loss": 0.083, + "step": 12174 + }, + { + "epoch": 0.10568484648570758, + "grad_norm": 0.15234375, + "learning_rate": 0.0019392515660427937, + "loss": 0.1377, + "step": 12175 + }, + { + "epoch": 0.10569352696591175, + "grad_norm": 0.1357421875, + "learning_rate": 0.001939240816411897, + "loss": 0.1006, + "step": 12176 + }, + { + "epoch": 0.10570220744611591, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019392300658632189, + "loss": 0.1338, + "step": 12177 + }, + { + "epoch": 0.10571088792632008, + "grad_norm": 0.59375, + "learning_rate": 0.0019392193143967717, + "loss": 0.1016, + "step": 12178 + }, + { + "epoch": 0.10571956840652424, + "grad_norm": 0.263671875, + "learning_rate": 0.0019392085620125667, + "loss": 0.1396, + "step": 12179 + }, + { + "epoch": 0.10572824888672841, + "grad_norm": 0.412109375, + "learning_rate": 0.0019391978087106162, + "loss": 0.1318, + "step": 12180 + }, + { + "epoch": 0.10573692936693257, + "grad_norm": 0.57421875, + "learning_rate": 0.001939187054490931, + "loss": 0.1104, + "step": 12181 + }, + { + "epoch": 0.10574560984713674, + "grad_norm": 0.318359375, + "learning_rate": 0.001939176299353524, + "loss": 0.1699, + "step": 12182 + }, + { + "epoch": 0.1057542903273409, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019391655432984061, + "loss": 0.1826, + "step": 12183 + }, + { + "epoch": 0.10576297080754507, + "grad_norm": 0.146484375, + "learning_rate": 0.0019391547863255893, + "loss": 0.1494, + "step": 12184 + }, + { + "epoch": 0.10577165128774924, + "grad_norm": 0.8046875, + "learning_rate": 0.0019391440284350857, + "loss": 0.1582, + "step": 12185 + }, + { + "epoch": 0.1057803317679534, + "grad_norm": 0.408203125, + "learning_rate": 0.0019391332696269065, + "loss": 0.1221, + "step": 12186 + }, + { + "epoch": 0.10578901224815757, + "grad_norm": 0.1962890625, + "learning_rate": 0.001939122509901064, + "loss": 0.1055, + "step": 12187 + }, + { + "epoch": 0.10579769272836173, + "grad_norm": 1.296875, + "learning_rate": 0.0019391117492575697, + "loss": 0.1699, + "step": 12188 + }, + { + "epoch": 0.1058063732085659, + "grad_norm": 0.59375, + "learning_rate": 0.0019391009876964353, + "loss": 0.1099, + "step": 12189 + }, + { + "epoch": 0.10581505368877006, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019390902252176727, + "loss": 0.1084, + "step": 12190 + }, + { + "epoch": 0.10582373416897423, + "grad_norm": 0.177734375, + "learning_rate": 0.001939079461821294, + "loss": 0.123, + "step": 12191 + }, + { + "epoch": 0.10583241464917839, + "grad_norm": 0.2080078125, + "learning_rate": 0.00193906869750731, + "loss": 0.1377, + "step": 12192 + }, + { + "epoch": 0.10584109512938256, + "grad_norm": 0.515625, + "learning_rate": 0.0019390579322757335, + "loss": 0.1289, + "step": 12193 + }, + { + "epoch": 0.10584977560958672, + "grad_norm": 0.13671875, + "learning_rate": 0.0019390471661265756, + "loss": 0.125, + "step": 12194 + }, + { + "epoch": 0.10585845608979089, + "grad_norm": 0.302734375, + "learning_rate": 0.0019390363990598482, + "loss": 0.1602, + "step": 12195 + }, + { + "epoch": 0.10586713656999505, + "grad_norm": 1.046875, + "learning_rate": 0.0019390256310755636, + "loss": 0.1543, + "step": 12196 + }, + { + "epoch": 0.10587581705019922, + "grad_norm": 0.0830078125, + "learning_rate": 0.001939014862173733, + "loss": 0.1562, + "step": 12197 + }, + { + "epoch": 0.10588449753040338, + "grad_norm": 0.314453125, + "learning_rate": 0.0019390040923543683, + "loss": 0.1592, + "step": 12198 + }, + { + "epoch": 0.10589317801060755, + "grad_norm": 0.3203125, + "learning_rate": 0.0019389933216174815, + "loss": 0.127, + "step": 12199 + }, + { + "epoch": 0.10590185849081171, + "grad_norm": 1.7265625, + "learning_rate": 0.001938982549963084, + "loss": 0.1348, + "step": 12200 + }, + { + "epoch": 0.10591053897101588, + "grad_norm": 0.294921875, + "learning_rate": 0.001938971777391188, + "loss": 0.1357, + "step": 12201 + }, + { + "epoch": 0.10591921945122004, + "grad_norm": 0.5, + "learning_rate": 0.0019389610039018048, + "loss": 0.104, + "step": 12202 + }, + { + "epoch": 0.10592789993142421, + "grad_norm": 0.345703125, + "learning_rate": 0.0019389502294949467, + "loss": 0.1133, + "step": 12203 + }, + { + "epoch": 0.10593658041162837, + "grad_norm": 0.130859375, + "learning_rate": 0.0019389394541706251, + "loss": 0.0889, + "step": 12204 + }, + { + "epoch": 0.10594526089183254, + "grad_norm": 0.1875, + "learning_rate": 0.0019389286779288519, + "loss": 0.084, + "step": 12205 + }, + { + "epoch": 0.1059539413720367, + "grad_norm": 0.373046875, + "learning_rate": 0.0019389179007696389, + "loss": 0.123, + "step": 12206 + }, + { + "epoch": 0.10596262185224087, + "grad_norm": 0.486328125, + "learning_rate": 0.001938907122692998, + "loss": 0.123, + "step": 12207 + }, + { + "epoch": 0.10597130233244503, + "grad_norm": 0.1181640625, + "learning_rate": 0.001938896343698941, + "loss": 0.1055, + "step": 12208 + }, + { + "epoch": 0.1059799828126492, + "grad_norm": 0.5625, + "learning_rate": 0.0019388855637874792, + "loss": 0.0947, + "step": 12209 + }, + { + "epoch": 0.10598866329285336, + "grad_norm": 0.267578125, + "learning_rate": 0.0019388747829586252, + "loss": 0.1211, + "step": 12210 + }, + { + "epoch": 0.10599734377305753, + "grad_norm": 0.189453125, + "learning_rate": 0.0019388640012123898, + "loss": 0.0957, + "step": 12211 + }, + { + "epoch": 0.1060060242532617, + "grad_norm": 1.1328125, + "learning_rate": 0.0019388532185487858, + "loss": 0.1602, + "step": 12212 + }, + { + "epoch": 0.10601470473346586, + "grad_norm": 0.060302734375, + "learning_rate": 0.0019388424349678242, + "loss": 0.1074, + "step": 12213 + }, + { + "epoch": 0.10602338521367002, + "grad_norm": 0.765625, + "learning_rate": 0.0019388316504695172, + "loss": 0.1133, + "step": 12214 + }, + { + "epoch": 0.10603206569387419, + "grad_norm": 0.30078125, + "learning_rate": 0.0019388208650538766, + "loss": 0.1504, + "step": 12215 + }, + { + "epoch": 0.10604074617407835, + "grad_norm": 0.4375, + "learning_rate": 0.0019388100787209141, + "loss": 0.1602, + "step": 12216 + }, + { + "epoch": 0.10604942665428252, + "grad_norm": 0.25, + "learning_rate": 0.0019387992914706416, + "loss": 0.1167, + "step": 12217 + }, + { + "epoch": 0.10605810713448668, + "grad_norm": 0.3828125, + "learning_rate": 0.0019387885033030705, + "loss": 0.0991, + "step": 12218 + }, + { + "epoch": 0.10606678761469085, + "grad_norm": 0.5625, + "learning_rate": 0.0019387777142182133, + "loss": 0.1196, + "step": 12219 + }, + { + "epoch": 0.10607546809489501, + "grad_norm": 0.40625, + "learning_rate": 0.001938766924216081, + "loss": 0.1064, + "step": 12220 + }, + { + "epoch": 0.10608414857509918, + "grad_norm": 0.47265625, + "learning_rate": 0.0019387561332966858, + "loss": 0.1445, + "step": 12221 + }, + { + "epoch": 0.10609282905530334, + "grad_norm": 1.359375, + "learning_rate": 0.0019387453414600397, + "loss": 0.1348, + "step": 12222 + }, + { + "epoch": 0.10610150953550751, + "grad_norm": 0.21875, + "learning_rate": 0.0019387345487061542, + "loss": 0.1084, + "step": 12223 + }, + { + "epoch": 0.10611019001571168, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019387237550350415, + "loss": 0.1035, + "step": 12224 + }, + { + "epoch": 0.10611887049591584, + "grad_norm": 0.17578125, + "learning_rate": 0.0019387129604467128, + "loss": 0.1094, + "step": 12225 + }, + { + "epoch": 0.10612755097612, + "grad_norm": 0.146484375, + "learning_rate": 0.0019387021649411801, + "loss": 0.1592, + "step": 12226 + }, + { + "epoch": 0.10613623145632417, + "grad_norm": 0.130859375, + "learning_rate": 0.0019386913685184555, + "loss": 0.1045, + "step": 12227 + }, + { + "epoch": 0.10614491193652834, + "grad_norm": 0.81640625, + "learning_rate": 0.0019386805711785506, + "loss": 0.127, + "step": 12228 + }, + { + "epoch": 0.10615359241673249, + "grad_norm": 0.2158203125, + "learning_rate": 0.001938669772921477, + "loss": 0.0952, + "step": 12229 + }, + { + "epoch": 0.10616227289693665, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019386589737472471, + "loss": 0.1069, + "step": 12230 + }, + { + "epoch": 0.10617095337714082, + "grad_norm": 0.7421875, + "learning_rate": 0.0019386481736558723, + "loss": 0.1064, + "step": 12231 + }, + { + "epoch": 0.10617963385734498, + "grad_norm": 0.361328125, + "learning_rate": 0.0019386373726473644, + "loss": 0.1338, + "step": 12232 + }, + { + "epoch": 0.10618831433754915, + "grad_norm": 0.279296875, + "learning_rate": 0.0019386265707217352, + "loss": 0.1367, + "step": 12233 + }, + { + "epoch": 0.10619699481775331, + "grad_norm": 0.4296875, + "learning_rate": 0.0019386157678789966, + "loss": 0.1226, + "step": 12234 + }, + { + "epoch": 0.10620567529795748, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019386049641191604, + "loss": 0.1465, + "step": 12235 + }, + { + "epoch": 0.10621435577816164, + "grad_norm": 0.39453125, + "learning_rate": 0.0019385941594422386, + "loss": 0.1504, + "step": 12236 + }, + { + "epoch": 0.10622303625836581, + "grad_norm": 0.48828125, + "learning_rate": 0.0019385833538482424, + "loss": 0.123, + "step": 12237 + }, + { + "epoch": 0.10623171673856997, + "grad_norm": 0.6484375, + "learning_rate": 0.0019385725473371844, + "loss": 0.083, + "step": 12238 + }, + { + "epoch": 0.10624039721877414, + "grad_norm": 0.66015625, + "learning_rate": 0.0019385617399090757, + "loss": 0.125, + "step": 12239 + }, + { + "epoch": 0.1062490776989783, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019385509315639287, + "loss": 0.1006, + "step": 12240 + }, + { + "epoch": 0.10625775817918247, + "grad_norm": 0.86328125, + "learning_rate": 0.0019385401223017552, + "loss": 0.1235, + "step": 12241 + }, + { + "epoch": 0.10626643865938663, + "grad_norm": 0.279296875, + "learning_rate": 0.001938529312122567, + "loss": 0.1201, + "step": 12242 + }, + { + "epoch": 0.1062751191395908, + "grad_norm": 0.275390625, + "learning_rate": 0.0019385185010263749, + "loss": 0.1191, + "step": 12243 + }, + { + "epoch": 0.10628379961979496, + "grad_norm": 0.2353515625, + "learning_rate": 0.001938507689013192, + "loss": 0.1216, + "step": 12244 + }, + { + "epoch": 0.10629248009999913, + "grad_norm": 0.095703125, + "learning_rate": 0.0019384968760830302, + "loss": 0.1445, + "step": 12245 + }, + { + "epoch": 0.1063011605802033, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019384860622359001, + "loss": 0.1807, + "step": 12246 + }, + { + "epoch": 0.10630984106040746, + "grad_norm": 0.8046875, + "learning_rate": 0.0019384752474718146, + "loss": 0.1562, + "step": 12247 + }, + { + "epoch": 0.10631852154061162, + "grad_norm": 0.53125, + "learning_rate": 0.001938464431790785, + "loss": 0.1289, + "step": 12248 + }, + { + "epoch": 0.10632720202081579, + "grad_norm": 0.37109375, + "learning_rate": 0.0019384536151928235, + "loss": 0.1328, + "step": 12249 + }, + { + "epoch": 0.10633588250101995, + "grad_norm": 0.337890625, + "learning_rate": 0.0019384427976779414, + "loss": 0.0864, + "step": 12250 + }, + { + "epoch": 0.10634456298122412, + "grad_norm": 0.11328125, + "learning_rate": 0.0019384319792461512, + "loss": 0.1084, + "step": 12251 + }, + { + "epoch": 0.10635324346142828, + "grad_norm": 0.384765625, + "learning_rate": 0.001938421159897464, + "loss": 0.1211, + "step": 12252 + }, + { + "epoch": 0.10636192394163245, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019384103396318925, + "loss": 0.1113, + "step": 12253 + }, + { + "epoch": 0.10637060442183661, + "grad_norm": 0.255859375, + "learning_rate": 0.0019383995184494476, + "loss": 0.1187, + "step": 12254 + }, + { + "epoch": 0.10637928490204078, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019383886963501421, + "loss": 0.0938, + "step": 12255 + }, + { + "epoch": 0.10638796538224495, + "grad_norm": 0.390625, + "learning_rate": 0.0019383778733339868, + "loss": 0.1279, + "step": 12256 + }, + { + "epoch": 0.10639664586244911, + "grad_norm": 0.103515625, + "learning_rate": 0.0019383670494009942, + "loss": 0.1299, + "step": 12257 + }, + { + "epoch": 0.10640532634265328, + "grad_norm": 0.1748046875, + "learning_rate": 0.001938356224551176, + "loss": 0.1138, + "step": 12258 + }, + { + "epoch": 0.10641400682285744, + "grad_norm": 0.5625, + "learning_rate": 0.001938345398784544, + "loss": 0.1221, + "step": 12259 + }, + { + "epoch": 0.1064226873030616, + "grad_norm": 0.08203125, + "learning_rate": 0.0019383345721011103, + "loss": 0.1069, + "step": 12260 + }, + { + "epoch": 0.10643136778326577, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019383237445008861, + "loss": 0.1113, + "step": 12261 + }, + { + "epoch": 0.10644004826346994, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019383129159838842, + "loss": 0.1211, + "step": 12262 + }, + { + "epoch": 0.1064487287436741, + "grad_norm": 0.287109375, + "learning_rate": 0.0019383020865501152, + "loss": 0.103, + "step": 12263 + }, + { + "epoch": 0.10645740922387827, + "grad_norm": 0.08203125, + "learning_rate": 0.0019382912561995923, + "loss": 0.0894, + "step": 12264 + }, + { + "epoch": 0.10646608970408243, + "grad_norm": 0.59765625, + "learning_rate": 0.0019382804249323262, + "loss": 0.2109, + "step": 12265 + }, + { + "epoch": 0.1064747701842866, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019382695927483299, + "loss": 0.0977, + "step": 12266 + }, + { + "epoch": 0.10648345066449076, + "grad_norm": 0.302734375, + "learning_rate": 0.0019382587596476137, + "loss": 0.1289, + "step": 12267 + }, + { + "epoch": 0.10649213114469493, + "grad_norm": 0.1103515625, + "learning_rate": 0.001938247925630191, + "loss": 0.1445, + "step": 12268 + }, + { + "epoch": 0.10650081162489909, + "grad_norm": 0.1640625, + "learning_rate": 0.0019382370906960724, + "loss": 0.1089, + "step": 12269 + }, + { + "epoch": 0.10650949210510326, + "grad_norm": 0.7421875, + "learning_rate": 0.0019382262548452707, + "loss": 0.1279, + "step": 12270 + }, + { + "epoch": 0.10651817258530742, + "grad_norm": 0.1455078125, + "learning_rate": 0.0019382154180777973, + "loss": 0.1309, + "step": 12271 + }, + { + "epoch": 0.10652685306551159, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019382045803936641, + "loss": 0.0991, + "step": 12272 + }, + { + "epoch": 0.10653553354571575, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019381937417928827, + "loss": 0.1465, + "step": 12273 + }, + { + "epoch": 0.10654421402591992, + "grad_norm": 0.376953125, + "learning_rate": 0.0019381829022754656, + "loss": 0.1172, + "step": 12274 + }, + { + "epoch": 0.10655289450612408, + "grad_norm": 0.75390625, + "learning_rate": 0.001938172061841424, + "loss": 0.1289, + "step": 12275 + }, + { + "epoch": 0.10656157498632825, + "grad_norm": 0.91015625, + "learning_rate": 0.00193816122049077, + "loss": 0.1152, + "step": 12276 + }, + { + "epoch": 0.10657025546653241, + "grad_norm": 0.2294921875, + "learning_rate": 0.001938150378223516, + "loss": 0.1309, + "step": 12277 + }, + { + "epoch": 0.10657893594673658, + "grad_norm": 2.09375, + "learning_rate": 0.001938139535039673, + "loss": 0.1836, + "step": 12278 + }, + { + "epoch": 0.10658761642694074, + "grad_norm": 0.1533203125, + "learning_rate": 0.001938128690939253, + "loss": 0.1611, + "step": 12279 + }, + { + "epoch": 0.10659629690714491, + "grad_norm": 0.14453125, + "learning_rate": 0.001938117845922268, + "loss": 0.1445, + "step": 12280 + }, + { + "epoch": 0.10660497738734907, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019381069999887303, + "loss": 0.1328, + "step": 12281 + }, + { + "epoch": 0.10661365786755324, + "grad_norm": 0.328125, + "learning_rate": 0.0019380961531386511, + "loss": 0.1104, + "step": 12282 + }, + { + "epoch": 0.1066223383477574, + "grad_norm": 0.474609375, + "learning_rate": 0.0019380853053720427, + "loss": 0.1035, + "step": 12283 + }, + { + "epoch": 0.10663101882796157, + "grad_norm": 0.302734375, + "learning_rate": 0.0019380744566889169, + "loss": 0.105, + "step": 12284 + }, + { + "epoch": 0.10663969930816573, + "grad_norm": 0.388671875, + "learning_rate": 0.0019380636070892852, + "loss": 0.123, + "step": 12285 + }, + { + "epoch": 0.1066483797883699, + "grad_norm": 0.2109375, + "learning_rate": 0.0019380527565731598, + "loss": 0.1484, + "step": 12286 + }, + { + "epoch": 0.10665706026857406, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019380419051405527, + "loss": 0.1436, + "step": 12287 + }, + { + "epoch": 0.10666574074877823, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019380310527914752, + "loss": 0.1416, + "step": 12288 + }, + { + "epoch": 0.1066744212289824, + "grad_norm": 0.84765625, + "learning_rate": 0.00193802019952594, + "loss": 0.1387, + "step": 12289 + }, + { + "epoch": 0.10668310170918656, + "grad_norm": 0.2451171875, + "learning_rate": 0.001938009345343958, + "loss": 0.1104, + "step": 12290 + }, + { + "epoch": 0.10669178218939071, + "grad_norm": 0.357421875, + "learning_rate": 0.0019379984902455419, + "loss": 0.1064, + "step": 12291 + }, + { + "epoch": 0.10670046266959488, + "grad_norm": 0.400390625, + "learning_rate": 0.001937987634230703, + "loss": 0.0977, + "step": 12292 + }, + { + "epoch": 0.10670914314979904, + "grad_norm": 0.296875, + "learning_rate": 0.0019379767772994539, + "loss": 0.124, + "step": 12293 + }, + { + "epoch": 0.1067178236300032, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019379659194518053, + "loss": 0.0923, + "step": 12294 + }, + { + "epoch": 0.10672650411020737, + "grad_norm": 0.3125, + "learning_rate": 0.0019379550606877702, + "loss": 0.1084, + "step": 12295 + }, + { + "epoch": 0.10673518459041154, + "grad_norm": 0.1552734375, + "learning_rate": 0.00193794420100736, + "loss": 0.1152, + "step": 12296 + }, + { + "epoch": 0.1067438650706157, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019379333404105866, + "loss": 0.1475, + "step": 12297 + }, + { + "epoch": 0.10675254555081987, + "grad_norm": 0.369140625, + "learning_rate": 0.001937922478897462, + "loss": 0.2305, + "step": 12298 + }, + { + "epoch": 0.10676122603102403, + "grad_norm": 0.248046875, + "learning_rate": 0.0019379116164679976, + "loss": 0.1494, + "step": 12299 + }, + { + "epoch": 0.1067699065112282, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019379007531222059, + "loss": 0.1562, + "step": 12300 + }, + { + "epoch": 0.10677858699143236, + "grad_norm": 1.578125, + "learning_rate": 0.0019378898888600985, + "loss": 0.1514, + "step": 12301 + }, + { + "epoch": 0.10678726747163653, + "grad_norm": 0.150390625, + "learning_rate": 0.0019378790236816875, + "loss": 0.1211, + "step": 12302 + }, + { + "epoch": 0.10679594795184069, + "grad_norm": 0.5859375, + "learning_rate": 0.0019378681575869842, + "loss": 0.0996, + "step": 12303 + }, + { + "epoch": 0.10680462843204486, + "grad_norm": 0.62890625, + "learning_rate": 0.001937857290576001, + "loss": 0.1309, + "step": 12304 + }, + { + "epoch": 0.10681330891224902, + "grad_norm": 0.55078125, + "learning_rate": 0.0019378464226487498, + "loss": 0.0894, + "step": 12305 + }, + { + "epoch": 0.10682198939245319, + "grad_norm": 0.279296875, + "learning_rate": 0.001937835553805242, + "loss": 0.1162, + "step": 12306 + }, + { + "epoch": 0.10683066987265735, + "grad_norm": 0.33203125, + "learning_rate": 0.0019378246840454903, + "loss": 0.1201, + "step": 12307 + }, + { + "epoch": 0.10683935035286152, + "grad_norm": 0.345703125, + "learning_rate": 0.001937813813369506, + "loss": 0.1221, + "step": 12308 + }, + { + "epoch": 0.10684803083306568, + "grad_norm": 1.1953125, + "learning_rate": 0.0019378029417773009, + "loss": 0.1406, + "step": 12309 + }, + { + "epoch": 0.10685671131326985, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019377920692688873, + "loss": 0.1484, + "step": 12310 + }, + { + "epoch": 0.10686539179347401, + "grad_norm": 0.087890625, + "learning_rate": 0.0019377811958442765, + "loss": 0.1211, + "step": 12311 + }, + { + "epoch": 0.10687407227367818, + "grad_norm": 0.34765625, + "learning_rate": 0.001937770321503481, + "loss": 0.1064, + "step": 12312 + }, + { + "epoch": 0.10688275275388234, + "grad_norm": 0.412109375, + "learning_rate": 0.0019377594462465125, + "loss": 0.1641, + "step": 12313 + }, + { + "epoch": 0.10689143323408651, + "grad_norm": 0.1416015625, + "learning_rate": 0.001937748570073383, + "loss": 0.1475, + "step": 12314 + }, + { + "epoch": 0.10690011371429067, + "grad_norm": 0.095703125, + "learning_rate": 0.0019377376929841043, + "loss": 0.1445, + "step": 12315 + }, + { + "epoch": 0.10690879419449484, + "grad_norm": 0.4609375, + "learning_rate": 0.001937726814978688, + "loss": 0.1191, + "step": 12316 + }, + { + "epoch": 0.106917474674699, + "grad_norm": 0.392578125, + "learning_rate": 0.0019377159360571463, + "loss": 0.0962, + "step": 12317 + }, + { + "epoch": 0.10692615515490317, + "grad_norm": 0.365234375, + "learning_rate": 0.001937705056219491, + "loss": 0.0962, + "step": 12318 + }, + { + "epoch": 0.10693483563510733, + "grad_norm": 0.486328125, + "learning_rate": 0.0019376941754657337, + "loss": 0.123, + "step": 12319 + }, + { + "epoch": 0.1069435161153115, + "grad_norm": 0.515625, + "learning_rate": 0.0019376832937958873, + "loss": 0.1318, + "step": 12320 + }, + { + "epoch": 0.10695219659551566, + "grad_norm": 0.373046875, + "learning_rate": 0.0019376724112099627, + "loss": 0.0977, + "step": 12321 + }, + { + "epoch": 0.10696087707571983, + "grad_norm": 0.421875, + "learning_rate": 0.0019376615277079724, + "loss": 0.1562, + "step": 12322 + }, + { + "epoch": 0.106969557555924, + "grad_norm": 0.41015625, + "learning_rate": 0.0019376506432899276, + "loss": 0.1289, + "step": 12323 + }, + { + "epoch": 0.10697823803612816, + "grad_norm": 0.1103515625, + "learning_rate": 0.001937639757955841, + "loss": 0.1201, + "step": 12324 + }, + { + "epoch": 0.10698691851633232, + "grad_norm": 0.10791015625, + "learning_rate": 0.001937628871705724, + "loss": 0.126, + "step": 12325 + }, + { + "epoch": 0.10699559899653649, + "grad_norm": 0.125, + "learning_rate": 0.0019376179845395885, + "loss": 0.1011, + "step": 12326 + }, + { + "epoch": 0.10700427947674065, + "grad_norm": 0.193359375, + "learning_rate": 0.001937607096457447, + "loss": 0.1094, + "step": 12327 + }, + { + "epoch": 0.10701295995694482, + "grad_norm": 0.16015625, + "learning_rate": 0.0019375962074593103, + "loss": 0.0957, + "step": 12328 + }, + { + "epoch": 0.10702164043714898, + "grad_norm": 0.19140625, + "learning_rate": 0.0019375853175451917, + "loss": 0.126, + "step": 12329 + }, + { + "epoch": 0.10703032091735315, + "grad_norm": 0.56640625, + "learning_rate": 0.0019375744267151021, + "loss": 0.1318, + "step": 12330 + }, + { + "epoch": 0.10703900139755732, + "grad_norm": 0.361328125, + "learning_rate": 0.0019375635349690532, + "loss": 0.1348, + "step": 12331 + }, + { + "epoch": 0.10704768187776148, + "grad_norm": 0.126953125, + "learning_rate": 0.0019375526423070582, + "loss": 0.1289, + "step": 12332 + }, + { + "epoch": 0.10705636235796565, + "grad_norm": 0.3828125, + "learning_rate": 0.0019375417487291276, + "loss": 0.127, + "step": 12333 + }, + { + "epoch": 0.10706504283816981, + "grad_norm": 0.16015625, + "learning_rate": 0.0019375308542352741, + "loss": 0.1348, + "step": 12334 + }, + { + "epoch": 0.10707372331837398, + "grad_norm": 0.443359375, + "learning_rate": 0.0019375199588255096, + "loss": 0.1123, + "step": 12335 + }, + { + "epoch": 0.10708240379857814, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019375090624998458, + "loss": 0.168, + "step": 12336 + }, + { + "epoch": 0.1070910842787823, + "grad_norm": 0.52734375, + "learning_rate": 0.0019374981652582946, + "loss": 0.1973, + "step": 12337 + }, + { + "epoch": 0.10709976475898647, + "grad_norm": 0.421875, + "learning_rate": 0.001937487267100868, + "loss": 0.1104, + "step": 12338 + }, + { + "epoch": 0.10710844523919064, + "grad_norm": 0.53515625, + "learning_rate": 0.0019374763680275783, + "loss": 0.1211, + "step": 12339 + }, + { + "epoch": 0.1071171257193948, + "grad_norm": 0.349609375, + "learning_rate": 0.0019374654680384365, + "loss": 0.123, + "step": 12340 + }, + { + "epoch": 0.10712580619959897, + "grad_norm": 0.31640625, + "learning_rate": 0.001937454567133455, + "loss": 0.1167, + "step": 12341 + }, + { + "epoch": 0.10713448667980313, + "grad_norm": 1.2421875, + "learning_rate": 0.0019374436653126462, + "loss": 0.0996, + "step": 12342 + }, + { + "epoch": 0.1071431671600073, + "grad_norm": 0.5078125, + "learning_rate": 0.0019374327625760212, + "loss": 0.0957, + "step": 12343 + }, + { + "epoch": 0.10715184764021146, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019374218589235926, + "loss": 0.0942, + "step": 12344 + }, + { + "epoch": 0.10716052812041563, + "grad_norm": 0.1259765625, + "learning_rate": 0.001937410954355372, + "loss": 0.0962, + "step": 12345 + }, + { + "epoch": 0.10716920860061979, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019374000488713713, + "loss": 0.0845, + "step": 12346 + }, + { + "epoch": 0.10717788908082396, + "grad_norm": 0.443359375, + "learning_rate": 0.0019373891424716023, + "loss": 0.127, + "step": 12347 + }, + { + "epoch": 0.10718656956102812, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019373782351560774, + "loss": 0.0952, + "step": 12348 + }, + { + "epoch": 0.10719525004123229, + "grad_norm": 0.37890625, + "learning_rate": 0.0019373673269248081, + "loss": 0.0918, + "step": 12349 + }, + { + "epoch": 0.10720393052143645, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019373564177778064, + "loss": 0.0962, + "step": 12350 + }, + { + "epoch": 0.10721261100164062, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019373455077150844, + "loss": 0.1348, + "step": 12351 + }, + { + "epoch": 0.10722129148184477, + "grad_norm": 0.294921875, + "learning_rate": 0.001937334596736654, + "loss": 0.1016, + "step": 12352 + }, + { + "epoch": 0.10722997196204893, + "grad_norm": 0.193359375, + "learning_rate": 0.001937323684842527, + "loss": 0.1279, + "step": 12353 + }, + { + "epoch": 0.1072386524422531, + "grad_norm": 0.5703125, + "learning_rate": 0.0019373127720327154, + "loss": 0.126, + "step": 12354 + }, + { + "epoch": 0.10724733292245726, + "grad_norm": 0.09423828125, + "learning_rate": 0.001937301858307231, + "loss": 0.0947, + "step": 12355 + }, + { + "epoch": 0.10725601340266143, + "grad_norm": 0.55078125, + "learning_rate": 0.0019372909436660862, + "loss": 0.1484, + "step": 12356 + }, + { + "epoch": 0.1072646938828656, + "grad_norm": 0.142578125, + "learning_rate": 0.0019372800281092927, + "loss": 0.0957, + "step": 12357 + }, + { + "epoch": 0.10727337436306976, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019372691116368618, + "loss": 0.124, + "step": 12358 + }, + { + "epoch": 0.10728205484327392, + "grad_norm": 0.28515625, + "learning_rate": 0.0019372581942488061, + "loss": 0.1079, + "step": 12359 + }, + { + "epoch": 0.10729073532347809, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019372472759451378, + "loss": 0.123, + "step": 12360 + }, + { + "epoch": 0.10729941580368225, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019372363567258682, + "loss": 0.1641, + "step": 12361 + }, + { + "epoch": 0.10730809628388642, + "grad_norm": 0.427734375, + "learning_rate": 0.0019372254365910093, + "loss": 0.1104, + "step": 12362 + }, + { + "epoch": 0.10731677676409059, + "grad_norm": 0.5390625, + "learning_rate": 0.0019372145155405734, + "loss": 0.166, + "step": 12363 + }, + { + "epoch": 0.10732545724429475, + "grad_norm": 0.796875, + "learning_rate": 0.0019372035935745725, + "loss": 0.2969, + "step": 12364 + }, + { + "epoch": 0.10733413772449892, + "grad_norm": 0.15234375, + "learning_rate": 0.0019371926706930182, + "loss": 0.0933, + "step": 12365 + }, + { + "epoch": 0.10734281820470308, + "grad_norm": 1.1640625, + "learning_rate": 0.0019371817468959226, + "loss": 0.1035, + "step": 12366 + }, + { + "epoch": 0.10735149868490725, + "grad_norm": 0.171875, + "learning_rate": 0.0019371708221832974, + "loss": 0.1094, + "step": 12367 + }, + { + "epoch": 0.10736017916511141, + "grad_norm": 0.8671875, + "learning_rate": 0.001937159896555155, + "loss": 0.1328, + "step": 12368 + }, + { + "epoch": 0.10736885964531558, + "grad_norm": 0.248046875, + "learning_rate": 0.001937148970011507, + "loss": 0.1328, + "step": 12369 + }, + { + "epoch": 0.10737754012551974, + "grad_norm": 0.275390625, + "learning_rate": 0.0019371380425523654, + "loss": 0.1143, + "step": 12370 + }, + { + "epoch": 0.1073862206057239, + "grad_norm": 0.205078125, + "learning_rate": 0.0019371271141777422, + "loss": 0.1064, + "step": 12371 + }, + { + "epoch": 0.10739490108592807, + "grad_norm": 0.05810546875, + "learning_rate": 0.0019371161848876497, + "loss": 0.0889, + "step": 12372 + }, + { + "epoch": 0.10740358156613224, + "grad_norm": 0.59765625, + "learning_rate": 0.0019371052546820994, + "loss": 0.2188, + "step": 12373 + }, + { + "epoch": 0.1074122620463364, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019370943235611031, + "loss": 0.1602, + "step": 12374 + }, + { + "epoch": 0.10742094252654057, + "grad_norm": 0.48828125, + "learning_rate": 0.0019370833915246734, + "loss": 0.1387, + "step": 12375 + }, + { + "epoch": 0.10742962300674473, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019370724585728216, + "loss": 0.123, + "step": 12376 + }, + { + "epoch": 0.1074383034869489, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019370615247055602, + "loss": 0.1055, + "step": 12377 + }, + { + "epoch": 0.10744698396715306, + "grad_norm": 0.421875, + "learning_rate": 0.0019370505899229005, + "loss": 0.1533, + "step": 12378 + }, + { + "epoch": 0.10745566444735723, + "grad_norm": 0.51953125, + "learning_rate": 0.0019370396542248553, + "loss": 0.1777, + "step": 12379 + }, + { + "epoch": 0.10746434492756139, + "grad_norm": 0.53125, + "learning_rate": 0.0019370287176114357, + "loss": 0.1099, + "step": 12380 + }, + { + "epoch": 0.10747302540776556, + "grad_norm": 0.3046875, + "learning_rate": 0.0019370177800826542, + "loss": 0.0996, + "step": 12381 + }, + { + "epoch": 0.10748170588796972, + "grad_norm": 0.072265625, + "learning_rate": 0.0019370068416385228, + "loss": 0.1201, + "step": 12382 + }, + { + "epoch": 0.10749038636817389, + "grad_norm": 0.2353515625, + "learning_rate": 0.001936995902279053, + "loss": 0.1309, + "step": 12383 + }, + { + "epoch": 0.10749906684837805, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019369849620042572, + "loss": 0.126, + "step": 12384 + }, + { + "epoch": 0.10750774732858222, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019369740208141472, + "loss": 0.1279, + "step": 12385 + }, + { + "epoch": 0.10751642780878638, + "grad_norm": 0.427734375, + "learning_rate": 0.0019369630787087352, + "loss": 0.1035, + "step": 12386 + }, + { + "epoch": 0.10752510828899055, + "grad_norm": 0.109375, + "learning_rate": 0.0019369521356880327, + "loss": 0.1045, + "step": 12387 + }, + { + "epoch": 0.10753378876919471, + "grad_norm": 0.388671875, + "learning_rate": 0.0019369411917520521, + "loss": 0.1152, + "step": 12388 + }, + { + "epoch": 0.10754246924939888, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019369302469008053, + "loss": 0.1436, + "step": 12389 + }, + { + "epoch": 0.10755114972960304, + "grad_norm": 0.294921875, + "learning_rate": 0.0019369193011343035, + "loss": 0.1152, + "step": 12390 + }, + { + "epoch": 0.10755983020980721, + "grad_norm": 0.59765625, + "learning_rate": 0.00193690835445256, + "loss": 0.083, + "step": 12391 + }, + { + "epoch": 0.10756851069001137, + "grad_norm": 0.421875, + "learning_rate": 0.001936897406855586, + "loss": 0.1035, + "step": 12392 + }, + { + "epoch": 0.10757719117021554, + "grad_norm": 0.72265625, + "learning_rate": 0.0019368864583433933, + "loss": 0.1621, + "step": 12393 + }, + { + "epoch": 0.1075858716504197, + "grad_norm": 0.38671875, + "learning_rate": 0.0019368755089159944, + "loss": 0.1162, + "step": 12394 + }, + { + "epoch": 0.10759455213062387, + "grad_norm": 0.58203125, + "learning_rate": 0.0019368645585734008, + "loss": 0.2031, + "step": 12395 + }, + { + "epoch": 0.10760323261082803, + "grad_norm": 0.546875, + "learning_rate": 0.0019368536073156248, + "loss": 0.1006, + "step": 12396 + }, + { + "epoch": 0.1076119130910322, + "grad_norm": 0.921875, + "learning_rate": 0.0019368426551426782, + "loss": 0.1152, + "step": 12397 + }, + { + "epoch": 0.10762059357123636, + "grad_norm": 0.138671875, + "learning_rate": 0.001936831702054573, + "loss": 0.0903, + "step": 12398 + }, + { + "epoch": 0.10762927405144053, + "grad_norm": 0.4140625, + "learning_rate": 0.0019368207480513215, + "loss": 0.1157, + "step": 12399 + }, + { + "epoch": 0.1076379545316447, + "grad_norm": 0.259765625, + "learning_rate": 0.0019368097931329353, + "loss": 0.209, + "step": 12400 + }, + { + "epoch": 0.10764663501184886, + "grad_norm": 0.5703125, + "learning_rate": 0.0019367988372994264, + "loss": 0.1533, + "step": 12401 + }, + { + "epoch": 0.10765531549205302, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019367878805508067, + "loss": 0.1167, + "step": 12402 + }, + { + "epoch": 0.10766399597225719, + "grad_norm": 0.265625, + "learning_rate": 0.0019367769228870887, + "loss": 0.1152, + "step": 12403 + }, + { + "epoch": 0.10767267645246135, + "grad_norm": 0.9140625, + "learning_rate": 0.0019367659643082838, + "loss": 0.1221, + "step": 12404 + }, + { + "epoch": 0.10768135693266552, + "grad_norm": 0.34375, + "learning_rate": 0.0019367550048144042, + "loss": 0.1396, + "step": 12405 + }, + { + "epoch": 0.10769003741286969, + "grad_norm": 0.2138671875, + "learning_rate": 0.001936744044405462, + "loss": 0.168, + "step": 12406 + }, + { + "epoch": 0.10769871789307385, + "grad_norm": 0.3203125, + "learning_rate": 0.0019367330830814692, + "loss": 0.1162, + "step": 12407 + }, + { + "epoch": 0.10770739837327802, + "grad_norm": 0.373046875, + "learning_rate": 0.0019367221208424375, + "loss": 0.1133, + "step": 12408 + }, + { + "epoch": 0.10771607885348218, + "grad_norm": 0.296875, + "learning_rate": 0.001936711157688379, + "loss": 0.1221, + "step": 12409 + }, + { + "epoch": 0.10772475933368635, + "grad_norm": 0.48828125, + "learning_rate": 0.001936700193619306, + "loss": 0.1602, + "step": 12410 + }, + { + "epoch": 0.10773343981389051, + "grad_norm": 0.265625, + "learning_rate": 0.00193668922863523, + "loss": 0.1113, + "step": 12411 + }, + { + "epoch": 0.10774212029409468, + "grad_norm": 0.33203125, + "learning_rate": 0.0019366782627361632, + "loss": 0.1289, + "step": 12412 + }, + { + "epoch": 0.10775080077429884, + "grad_norm": 0.248046875, + "learning_rate": 0.001936667295922118, + "loss": 0.1182, + "step": 12413 + }, + { + "epoch": 0.10775948125450299, + "grad_norm": 0.123046875, + "learning_rate": 0.0019366563281931054, + "loss": 0.0996, + "step": 12414 + }, + { + "epoch": 0.10776816173470716, + "grad_norm": 0.4609375, + "learning_rate": 0.0019366453595491384, + "loss": 0.1128, + "step": 12415 + }, + { + "epoch": 0.10777684221491132, + "grad_norm": 0.64453125, + "learning_rate": 0.0019366343899902287, + "loss": 0.1035, + "step": 12416 + }, + { + "epoch": 0.10778552269511549, + "grad_norm": 0.380859375, + "learning_rate": 0.0019366234195163882, + "loss": 0.125, + "step": 12417 + }, + { + "epoch": 0.10779420317531965, + "grad_norm": 0.294921875, + "learning_rate": 0.0019366124481276286, + "loss": 0.0854, + "step": 12418 + }, + { + "epoch": 0.10780288365552382, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019366014758239623, + "loss": 0.1445, + "step": 12419 + }, + { + "epoch": 0.10781156413572798, + "grad_norm": 0.57421875, + "learning_rate": 0.0019365905026054013, + "loss": 0.1562, + "step": 12420 + }, + { + "epoch": 0.10782024461593215, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019365795284719577, + "loss": 0.1367, + "step": 12421 + }, + { + "epoch": 0.10782892509613631, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019365685534236427, + "loss": 0.1523, + "step": 12422 + }, + { + "epoch": 0.10783760557634048, + "grad_norm": 0.177734375, + "learning_rate": 0.0019365575774604696, + "loss": 0.123, + "step": 12423 + }, + { + "epoch": 0.10784628605654464, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019365466005824492, + "loss": 0.1162, + "step": 12424 + }, + { + "epoch": 0.10785496653674881, + "grad_norm": 1.3125, + "learning_rate": 0.0019365356227895943, + "loss": 0.1465, + "step": 12425 + }, + { + "epoch": 0.10786364701695297, + "grad_norm": 0.25390625, + "learning_rate": 0.0019365246440819165, + "loss": 0.1387, + "step": 12426 + }, + { + "epoch": 0.10787232749715714, + "grad_norm": 0.765625, + "learning_rate": 0.0019365136644594278, + "loss": 0.2812, + "step": 12427 + }, + { + "epoch": 0.1078810079773613, + "grad_norm": 0.322265625, + "learning_rate": 0.0019365026839221402, + "loss": 0.1162, + "step": 12428 + }, + { + "epoch": 0.10788968845756547, + "grad_norm": 0.07763671875, + "learning_rate": 0.001936491702470066, + "loss": 0.1182, + "step": 12429 + }, + { + "epoch": 0.10789836893776963, + "grad_norm": 0.126953125, + "learning_rate": 0.0019364807201032173, + "loss": 0.1348, + "step": 12430 + }, + { + "epoch": 0.1079070494179738, + "grad_norm": 0.65234375, + "learning_rate": 0.0019364697368216056, + "loss": 0.0947, + "step": 12431 + }, + { + "epoch": 0.10791572989817796, + "grad_norm": 0.50390625, + "learning_rate": 0.001936458752625243, + "loss": 0.1455, + "step": 12432 + }, + { + "epoch": 0.10792441037838213, + "grad_norm": 0.6015625, + "learning_rate": 0.001936447767514142, + "loss": 0.1699, + "step": 12433 + }, + { + "epoch": 0.1079330908585863, + "grad_norm": 0.384765625, + "learning_rate": 0.001936436781488314, + "loss": 0.1357, + "step": 12434 + }, + { + "epoch": 0.10794177133879046, + "grad_norm": 0.138671875, + "learning_rate": 0.0019364257945477716, + "loss": 0.1113, + "step": 12435 + }, + { + "epoch": 0.10795045181899462, + "grad_norm": 0.458984375, + "learning_rate": 0.0019364148066925264, + "loss": 0.1445, + "step": 12436 + }, + { + "epoch": 0.10795913229919879, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019364038179225907, + "loss": 0.1309, + "step": 12437 + }, + { + "epoch": 0.10796781277940296, + "grad_norm": 0.94140625, + "learning_rate": 0.0019363928282379761, + "loss": 0.1494, + "step": 12438 + }, + { + "epoch": 0.10797649325960712, + "grad_norm": 0.119140625, + "learning_rate": 0.001936381837638695, + "loss": 0.124, + "step": 12439 + }, + { + "epoch": 0.10798517373981129, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019363708461247594, + "loss": 0.1025, + "step": 12440 + }, + { + "epoch": 0.10799385422001545, + "grad_norm": 0.154296875, + "learning_rate": 0.001936359853696181, + "loss": 0.1289, + "step": 12441 + }, + { + "epoch": 0.10800253470021962, + "grad_norm": 0.12109375, + "learning_rate": 0.0019363488603529719, + "loss": 0.1123, + "step": 12442 + }, + { + "epoch": 0.10801121518042378, + "grad_norm": 0.10546875, + "learning_rate": 0.0019363378660951448, + "loss": 0.127, + "step": 12443 + }, + { + "epoch": 0.10801989566062795, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019363268709227104, + "loss": 0.1133, + "step": 12444 + }, + { + "epoch": 0.10802857614083211, + "grad_norm": 0.80859375, + "learning_rate": 0.0019363158748356818, + "loss": 0.1162, + "step": 12445 + }, + { + "epoch": 0.10803725662103628, + "grad_norm": 0.171875, + "learning_rate": 0.001936304877834071, + "loss": 0.1367, + "step": 12446 + }, + { + "epoch": 0.10804593710124044, + "grad_norm": 0.419921875, + "learning_rate": 0.0019362938799178898, + "loss": 0.1318, + "step": 12447 + }, + { + "epoch": 0.1080546175814446, + "grad_norm": 0.431640625, + "learning_rate": 0.00193628288108715, + "loss": 0.1162, + "step": 12448 + }, + { + "epoch": 0.10806329806164877, + "grad_norm": 0.0869140625, + "learning_rate": 0.0019362718813418637, + "loss": 0.1045, + "step": 12449 + }, + { + "epoch": 0.10807197854185294, + "grad_norm": 0.1611328125, + "learning_rate": 0.001936260880682043, + "loss": 0.1279, + "step": 12450 + }, + { + "epoch": 0.1080806590220571, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019362498791077, + "loss": 0.125, + "step": 12451 + }, + { + "epoch": 0.10808933950226127, + "grad_norm": 0.357421875, + "learning_rate": 0.0019362388766188468, + "loss": 0.0977, + "step": 12452 + }, + { + "epoch": 0.10809801998246543, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019362278732154954, + "loss": 0.0898, + "step": 12453 + }, + { + "epoch": 0.1081067004626696, + "grad_norm": 0.423828125, + "learning_rate": 0.0019362168688976575, + "loss": 0.1172, + "step": 12454 + }, + { + "epoch": 0.10811538094287376, + "grad_norm": 0.56640625, + "learning_rate": 0.0019362058636653457, + "loss": 0.1377, + "step": 12455 + }, + { + "epoch": 0.10812406142307793, + "grad_norm": 0.22265625, + "learning_rate": 0.0019361948575185716, + "loss": 0.1816, + "step": 12456 + }, + { + "epoch": 0.10813274190328209, + "grad_norm": 0.322265625, + "learning_rate": 0.0019361838504573475, + "loss": 0.0928, + "step": 12457 + }, + { + "epoch": 0.10814142238348626, + "grad_norm": 1.5078125, + "learning_rate": 0.0019361728424816848, + "loss": 0.4199, + "step": 12458 + }, + { + "epoch": 0.10815010286369042, + "grad_norm": 0.171875, + "learning_rate": 0.0019361618335915964, + "loss": 0.1191, + "step": 12459 + }, + { + "epoch": 0.10815878334389459, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019361508237870939, + "loss": 0.0918, + "step": 12460 + }, + { + "epoch": 0.10816746382409875, + "grad_norm": 0.482421875, + "learning_rate": 0.0019361398130681897, + "loss": 0.1113, + "step": 12461 + }, + { + "epoch": 0.10817614430430292, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019361288014348953, + "loss": 0.1426, + "step": 12462 + }, + { + "epoch": 0.10818482478450708, + "grad_norm": 0.115234375, + "learning_rate": 0.0019361177888872231, + "loss": 0.1123, + "step": 12463 + }, + { + "epoch": 0.10819350526471125, + "grad_norm": 0.6171875, + "learning_rate": 0.0019361067754251852, + "loss": 0.1123, + "step": 12464 + }, + { + "epoch": 0.10820218574491541, + "grad_norm": 0.1640625, + "learning_rate": 0.001936095761048793, + "loss": 0.1641, + "step": 12465 + }, + { + "epoch": 0.10821086622511958, + "grad_norm": 0.30078125, + "learning_rate": 0.0019360847457580595, + "loss": 0.1074, + "step": 12466 + }, + { + "epoch": 0.10821954670532374, + "grad_norm": 0.39453125, + "learning_rate": 0.0019360737295529963, + "loss": 0.1162, + "step": 12467 + }, + { + "epoch": 0.10822822718552791, + "grad_norm": 0.330078125, + "learning_rate": 0.0019360627124336153, + "loss": 0.1318, + "step": 12468 + }, + { + "epoch": 0.10823690766573207, + "grad_norm": 0.52734375, + "learning_rate": 0.0019360516943999284, + "loss": 0.1699, + "step": 12469 + }, + { + "epoch": 0.10824558814593624, + "grad_norm": 0.42578125, + "learning_rate": 0.0019360406754519483, + "loss": 0.1504, + "step": 12470 + }, + { + "epoch": 0.1082542686261404, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019360296555896866, + "loss": 0.0986, + "step": 12471 + }, + { + "epoch": 0.10826294910634457, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019360186348131555, + "loss": 0.0835, + "step": 12472 + }, + { + "epoch": 0.10827162958654873, + "grad_norm": 0.205078125, + "learning_rate": 0.0019360076131223665, + "loss": 0.124, + "step": 12473 + }, + { + "epoch": 0.1082803100667529, + "grad_norm": 0.53515625, + "learning_rate": 0.0019359965905173327, + "loss": 0.1191, + "step": 12474 + }, + { + "epoch": 0.10828899054695705, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019359855669980653, + "loss": 0.1436, + "step": 12475 + }, + { + "epoch": 0.10829767102716122, + "grad_norm": 0.177734375, + "learning_rate": 0.0019359745425645766, + "loss": 0.1104, + "step": 12476 + }, + { + "epoch": 0.10830635150736538, + "grad_norm": 0.126953125, + "learning_rate": 0.001935963517216879, + "loss": 0.1367, + "step": 12477 + }, + { + "epoch": 0.10831503198756955, + "grad_norm": 0.1982421875, + "learning_rate": 0.001935952490954984, + "loss": 0.1357, + "step": 12478 + }, + { + "epoch": 0.10832371246777371, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019359414637789042, + "loss": 0.1426, + "step": 12479 + }, + { + "epoch": 0.10833239294797788, + "grad_norm": 0.28125, + "learning_rate": 0.0019359304356886512, + "loss": 0.1582, + "step": 12480 + }, + { + "epoch": 0.10834107342818204, + "grad_norm": 0.150390625, + "learning_rate": 0.0019359194066842372, + "loss": 0.1152, + "step": 12481 + }, + { + "epoch": 0.1083497539083862, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019359083767656743, + "loss": 0.1172, + "step": 12482 + }, + { + "epoch": 0.10835843438859037, + "grad_norm": 0.31640625, + "learning_rate": 0.0019358973459329745, + "loss": 0.1167, + "step": 12483 + }, + { + "epoch": 0.10836711486879454, + "grad_norm": 0.232421875, + "learning_rate": 0.00193588631418615, + "loss": 0.1582, + "step": 12484 + }, + { + "epoch": 0.1083757953489987, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019358752815252127, + "loss": 0.1826, + "step": 12485 + }, + { + "epoch": 0.10838447582920287, + "grad_norm": 0.1708984375, + "learning_rate": 0.001935864247950175, + "loss": 0.0986, + "step": 12486 + }, + { + "epoch": 0.10839315630940703, + "grad_norm": 0.166015625, + "learning_rate": 0.0019358532134610485, + "loss": 0.124, + "step": 12487 + }, + { + "epoch": 0.1084018367896112, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019358421780578455, + "loss": 0.1094, + "step": 12488 + }, + { + "epoch": 0.10841051726981536, + "grad_norm": 0.796875, + "learning_rate": 0.0019358311417405783, + "loss": 0.1504, + "step": 12489 + }, + { + "epoch": 0.10841919775001953, + "grad_norm": 0.3515625, + "learning_rate": 0.0019358201045092582, + "loss": 0.0884, + "step": 12490 + }, + { + "epoch": 0.10842787823022369, + "grad_norm": 0.6171875, + "learning_rate": 0.0019358090663638982, + "loss": 0.1089, + "step": 12491 + }, + { + "epoch": 0.10843655871042786, + "grad_norm": 0.232421875, + "learning_rate": 0.0019357980273045096, + "loss": 0.1182, + "step": 12492 + }, + { + "epoch": 0.10844523919063202, + "grad_norm": 0.330078125, + "learning_rate": 0.0019357869873311056, + "loss": 0.1396, + "step": 12493 + }, + { + "epoch": 0.10845391967083619, + "grad_norm": 0.0791015625, + "learning_rate": 0.0019357759464436969, + "loss": 0.0977, + "step": 12494 + }, + { + "epoch": 0.10846260015104035, + "grad_norm": 0.1669921875, + "learning_rate": 0.001935764904642296, + "loss": 0.1079, + "step": 12495 + }, + { + "epoch": 0.10847128063124452, + "grad_norm": 0.296875, + "learning_rate": 0.0019357538619269158, + "loss": 0.1504, + "step": 12496 + }, + { + "epoch": 0.10847996111144868, + "grad_norm": 0.291015625, + "learning_rate": 0.0019357428182975671, + "loss": 0.084, + "step": 12497 + }, + { + "epoch": 0.10848864159165285, + "grad_norm": 0.0810546875, + "learning_rate": 0.001935731773754263, + "loss": 0.1152, + "step": 12498 + }, + { + "epoch": 0.10849732207185701, + "grad_norm": 0.1416015625, + "learning_rate": 0.001935720728297015, + "loss": 0.1143, + "step": 12499 + }, + { + "epoch": 0.10850600255206118, + "grad_norm": 0.404296875, + "learning_rate": 0.0019357096819258355, + "loss": 0.1797, + "step": 12500 + }, + { + "epoch": 0.10851468303226534, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019356986346407365, + "loss": 0.0977, + "step": 12501 + }, + { + "epoch": 0.10852336351246951, + "grad_norm": 0.248046875, + "learning_rate": 0.00193568758644173, + "loss": 0.1123, + "step": 12502 + }, + { + "epoch": 0.10853204399267367, + "grad_norm": 0.671875, + "learning_rate": 0.0019356765373288278, + "loss": 0.1299, + "step": 12503 + }, + { + "epoch": 0.10854072447287784, + "grad_norm": 0.35546875, + "learning_rate": 0.0019356654873020425, + "loss": 0.0762, + "step": 12504 + }, + { + "epoch": 0.108549404953082, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019356544363613862, + "loss": 0.1699, + "step": 12505 + }, + { + "epoch": 0.10855808543328617, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019356433845068705, + "loss": 0.2031, + "step": 12506 + }, + { + "epoch": 0.10856676591349033, + "grad_norm": 0.39453125, + "learning_rate": 0.0019356323317385078, + "loss": 0.1152, + "step": 12507 + }, + { + "epoch": 0.1085754463936945, + "grad_norm": 0.353515625, + "learning_rate": 0.0019356212780563102, + "loss": 0.1562, + "step": 12508 + }, + { + "epoch": 0.10858412687389866, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019356102234602897, + "loss": 0.1113, + "step": 12509 + }, + { + "epoch": 0.10859280735410283, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019355991679504585, + "loss": 0.1147, + "step": 12510 + }, + { + "epoch": 0.108601487834307, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019355881115268284, + "loss": 0.1582, + "step": 12511 + }, + { + "epoch": 0.10861016831451116, + "grad_norm": 0.16015625, + "learning_rate": 0.0019355770541894118, + "loss": 0.1104, + "step": 12512 + }, + { + "epoch": 0.10861884879471533, + "grad_norm": 0.68359375, + "learning_rate": 0.0019355659959382206, + "loss": 0.1289, + "step": 12513 + }, + { + "epoch": 0.10862752927491949, + "grad_norm": 0.125, + "learning_rate": 0.0019355549367732672, + "loss": 0.1143, + "step": 12514 + }, + { + "epoch": 0.10863620975512366, + "grad_norm": 0.32421875, + "learning_rate": 0.0019355438766945634, + "loss": 0.1064, + "step": 12515 + }, + { + "epoch": 0.10864489023532782, + "grad_norm": 1.125, + "learning_rate": 0.0019355328157021213, + "loss": 0.1152, + "step": 12516 + }, + { + "epoch": 0.10865357071553199, + "grad_norm": 0.640625, + "learning_rate": 0.001935521753795953, + "loss": 0.123, + "step": 12517 + }, + { + "epoch": 0.10866225119573615, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019355106909760707, + "loss": 0.0977, + "step": 12518 + }, + { + "epoch": 0.10867093167594032, + "grad_norm": 0.078125, + "learning_rate": 0.0019354996272424867, + "loss": 0.1011, + "step": 12519 + }, + { + "epoch": 0.10867961215614448, + "grad_norm": 0.94921875, + "learning_rate": 0.0019354885625952128, + "loss": 0.168, + "step": 12520 + }, + { + "epoch": 0.10868829263634865, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019354774970342609, + "loss": 0.1582, + "step": 12521 + }, + { + "epoch": 0.10869697311655281, + "grad_norm": 0.271484375, + "learning_rate": 0.0019354664305596435, + "loss": 0.0957, + "step": 12522 + }, + { + "epoch": 0.10870565359675698, + "grad_norm": 0.06396484375, + "learning_rate": 0.0019354553631713727, + "loss": 0.1045, + "step": 12523 + }, + { + "epoch": 0.10871433407696114, + "grad_norm": 0.10205078125, + "learning_rate": 0.00193544429486946, + "loss": 0.1172, + "step": 12524 + }, + { + "epoch": 0.1087230145571653, + "grad_norm": 0.51171875, + "learning_rate": 0.0019354332256539182, + "loss": 0.165, + "step": 12525 + }, + { + "epoch": 0.10873169503736947, + "grad_norm": 0.609375, + "learning_rate": 0.0019354221555247595, + "loss": 0.1338, + "step": 12526 + }, + { + "epoch": 0.10874037551757364, + "grad_norm": 0.34375, + "learning_rate": 0.0019354110844819956, + "loss": 0.1807, + "step": 12527 + }, + { + "epoch": 0.1087490559977778, + "grad_norm": 0.12890625, + "learning_rate": 0.0019354000125256384, + "loss": 0.1172, + "step": 12528 + }, + { + "epoch": 0.10875773647798197, + "grad_norm": 0.2109375, + "learning_rate": 0.0019353889396557006, + "loss": 0.1045, + "step": 12529 + }, + { + "epoch": 0.10876641695818613, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019353778658721937, + "loss": 0.1123, + "step": 12530 + }, + { + "epoch": 0.1087750974383903, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019353667911751302, + "loss": 0.1445, + "step": 12531 + }, + { + "epoch": 0.10878377791859446, + "grad_norm": 0.21875, + "learning_rate": 0.0019353557155645223, + "loss": 0.1279, + "step": 12532 + }, + { + "epoch": 0.10879245839879863, + "grad_norm": 0.1728515625, + "learning_rate": 0.001935344639040382, + "loss": 0.0869, + "step": 12533 + }, + { + "epoch": 0.10880113887900279, + "grad_norm": 0.482421875, + "learning_rate": 0.001935333561602721, + "loss": 0.1553, + "step": 12534 + }, + { + "epoch": 0.10880981935920696, + "grad_norm": 0.427734375, + "learning_rate": 0.001935322483251552, + "loss": 0.1357, + "step": 12535 + }, + { + "epoch": 0.10881849983941112, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019353114039868868, + "loss": 0.125, + "step": 12536 + }, + { + "epoch": 0.10882718031961527, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019353003238087377, + "loss": 0.1182, + "step": 12537 + }, + { + "epoch": 0.10883586079981944, + "grad_norm": 0.20703125, + "learning_rate": 0.0019352892427171166, + "loss": 0.1245, + "step": 12538 + }, + { + "epoch": 0.1088445412800236, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019352781607120358, + "loss": 0.1445, + "step": 12539 + }, + { + "epoch": 0.10885322176022777, + "grad_norm": 0.09375, + "learning_rate": 0.0019352670777935074, + "loss": 0.1211, + "step": 12540 + }, + { + "epoch": 0.10886190224043193, + "grad_norm": 0.2890625, + "learning_rate": 0.0019352559939615436, + "loss": 0.1152, + "step": 12541 + }, + { + "epoch": 0.1088705827206361, + "grad_norm": 0.1865234375, + "learning_rate": 0.001935244909216156, + "loss": 0.1094, + "step": 12542 + }, + { + "epoch": 0.10887926320084026, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019352338235573573, + "loss": 0.1084, + "step": 12543 + }, + { + "epoch": 0.10888794368104443, + "grad_norm": 0.5234375, + "learning_rate": 0.0019352227369851594, + "loss": 0.1074, + "step": 12544 + }, + { + "epoch": 0.1088966241612486, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019352116494995746, + "loss": 0.1187, + "step": 12545 + }, + { + "epoch": 0.10890530464145276, + "grad_norm": 0.5859375, + "learning_rate": 0.0019352005611006147, + "loss": 0.3398, + "step": 12546 + }, + { + "epoch": 0.10891398512165693, + "grad_norm": 0.154296875, + "learning_rate": 0.0019351894717882924, + "loss": 0.1182, + "step": 12547 + }, + { + "epoch": 0.10892266560186109, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019351783815626192, + "loss": 0.0996, + "step": 12548 + }, + { + "epoch": 0.10893134608206526, + "grad_norm": 0.93359375, + "learning_rate": 0.0019351672904236073, + "loss": 0.1328, + "step": 12549 + }, + { + "epoch": 0.10894002656226942, + "grad_norm": 0.1875, + "learning_rate": 0.001935156198371269, + "loss": 0.1084, + "step": 12550 + }, + { + "epoch": 0.10894870704247359, + "grad_norm": 1.0078125, + "learning_rate": 0.0019351451054056166, + "loss": 0.0996, + "step": 12551 + }, + { + "epoch": 0.10895738752267775, + "grad_norm": 0.4375, + "learning_rate": 0.0019351340115266625, + "loss": 0.0786, + "step": 12552 + }, + { + "epoch": 0.10896606800288192, + "grad_norm": 0.255859375, + "learning_rate": 0.0019351229167344176, + "loss": 0.0996, + "step": 12553 + }, + { + "epoch": 0.10897474848308608, + "grad_norm": 0.625, + "learning_rate": 0.0019351118210288953, + "loss": 0.1201, + "step": 12554 + }, + { + "epoch": 0.10898342896329025, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019351007244101071, + "loss": 0.1211, + "step": 12555 + }, + { + "epoch": 0.10899210944349441, + "grad_norm": 0.24609375, + "learning_rate": 0.0019350896268780654, + "loss": 0.1191, + "step": 12556 + }, + { + "epoch": 0.10900078992369858, + "grad_norm": 0.5546875, + "learning_rate": 0.001935078528432782, + "loss": 0.1035, + "step": 12557 + }, + { + "epoch": 0.10900947040390274, + "grad_norm": 0.8203125, + "learning_rate": 0.0019350674290742693, + "loss": 0.1235, + "step": 12558 + }, + { + "epoch": 0.1090181508841069, + "grad_norm": 0.625, + "learning_rate": 0.0019350563288025396, + "loss": 0.1001, + "step": 12559 + }, + { + "epoch": 0.10902683136431107, + "grad_norm": 0.65234375, + "learning_rate": 0.001935045227617605, + "loss": 0.1123, + "step": 12560 + }, + { + "epoch": 0.10903551184451524, + "grad_norm": 0.1533203125, + "learning_rate": 0.001935034125519477, + "loss": 0.1729, + "step": 12561 + }, + { + "epoch": 0.1090441923247194, + "grad_norm": 0.14453125, + "learning_rate": 0.0019350230225081686, + "loss": 0.1133, + "step": 12562 + }, + { + "epoch": 0.10905287280492357, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019350119185836915, + "loss": 0.0933, + "step": 12563 + }, + { + "epoch": 0.10906155328512773, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019350008137460578, + "loss": 0.1436, + "step": 12564 + }, + { + "epoch": 0.1090702337653319, + "grad_norm": 0.337890625, + "learning_rate": 0.0019349897079952798, + "loss": 0.1826, + "step": 12565 + }, + { + "epoch": 0.10907891424553606, + "grad_norm": 0.7109375, + "learning_rate": 0.0019349786013313696, + "loss": 0.1484, + "step": 12566 + }, + { + "epoch": 0.10908759472574023, + "grad_norm": 0.314453125, + "learning_rate": 0.0019349674937543396, + "loss": 0.1191, + "step": 12567 + }, + { + "epoch": 0.10909627520594439, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019349563852642014, + "loss": 0.1108, + "step": 12568 + }, + { + "epoch": 0.10910495568614856, + "grad_norm": 0.134765625, + "learning_rate": 0.0019349452758609673, + "loss": 0.0801, + "step": 12569 + }, + { + "epoch": 0.10911363616635272, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019349341655446497, + "loss": 0.0903, + "step": 12570 + }, + { + "epoch": 0.10912231664655689, + "grad_norm": 0.150390625, + "learning_rate": 0.001934923054315261, + "loss": 0.1455, + "step": 12571 + }, + { + "epoch": 0.10913099712676105, + "grad_norm": 0.392578125, + "learning_rate": 0.0019349119421728127, + "loss": 0.1113, + "step": 12572 + }, + { + "epoch": 0.10913967760696522, + "grad_norm": 0.43359375, + "learning_rate": 0.0019349008291173172, + "loss": 0.1021, + "step": 12573 + }, + { + "epoch": 0.10914835808716938, + "grad_norm": 0.37109375, + "learning_rate": 0.0019348897151487867, + "loss": 0.1426, + "step": 12574 + }, + { + "epoch": 0.10915703856737355, + "grad_norm": 0.154296875, + "learning_rate": 0.0019348786002672334, + "loss": 0.1328, + "step": 12575 + }, + { + "epoch": 0.10916571904757771, + "grad_norm": 0.30078125, + "learning_rate": 0.0019348674844726695, + "loss": 0.1191, + "step": 12576 + }, + { + "epoch": 0.10917439952778188, + "grad_norm": 0.41015625, + "learning_rate": 0.001934856367765107, + "loss": 0.1387, + "step": 12577 + }, + { + "epoch": 0.10918308000798604, + "grad_norm": 0.365234375, + "learning_rate": 0.0019348452501445582, + "loss": 0.1123, + "step": 12578 + }, + { + "epoch": 0.10919176048819021, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019348341316110349, + "loss": 0.1162, + "step": 12579 + }, + { + "epoch": 0.10920044096839437, + "grad_norm": 0.283203125, + "learning_rate": 0.0019348230121645495, + "loss": 0.1118, + "step": 12580 + }, + { + "epoch": 0.10920912144859854, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019348118918051148, + "loss": 0.1006, + "step": 12581 + }, + { + "epoch": 0.1092178019288027, + "grad_norm": 0.07861328125, + "learning_rate": 0.001934800770532742, + "loss": 0.0908, + "step": 12582 + }, + { + "epoch": 0.10922648240900687, + "grad_norm": 0.28125, + "learning_rate": 0.0019347896483474434, + "loss": 0.1084, + "step": 12583 + }, + { + "epoch": 0.10923516288921103, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019347785252492318, + "loss": 0.0806, + "step": 12584 + }, + { + "epoch": 0.1092438433694152, + "grad_norm": 0.341796875, + "learning_rate": 0.0019347674012381187, + "loss": 0.0972, + "step": 12585 + }, + { + "epoch": 0.10925252384961937, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019347562763141165, + "loss": 0.1191, + "step": 12586 + }, + { + "epoch": 0.10926120432982353, + "grad_norm": 0.251953125, + "learning_rate": 0.0019347451504772373, + "loss": 0.1025, + "step": 12587 + }, + { + "epoch": 0.1092698848100277, + "grad_norm": 0.3984375, + "learning_rate": 0.0019347340237274937, + "loss": 0.1602, + "step": 12588 + }, + { + "epoch": 0.10927856529023186, + "grad_norm": 0.71875, + "learning_rate": 0.001934722896064897, + "loss": 0.1309, + "step": 12589 + }, + { + "epoch": 0.10928724577043603, + "grad_norm": 0.353515625, + "learning_rate": 0.0019347117674894604, + "loss": 0.1426, + "step": 12590 + }, + { + "epoch": 0.10929592625064019, + "grad_norm": 0.333984375, + "learning_rate": 0.0019347006380011954, + "loss": 0.1201, + "step": 12591 + }, + { + "epoch": 0.10930460673084436, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019346895076001141, + "loss": 0.1484, + "step": 12592 + }, + { + "epoch": 0.10931328721104852, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019346783762862292, + "loss": 0.0962, + "step": 12593 + }, + { + "epoch": 0.10932196769125269, + "grad_norm": 0.458984375, + "learning_rate": 0.0019346672440595522, + "loss": 0.1201, + "step": 12594 + }, + { + "epoch": 0.10933064817145685, + "grad_norm": 0.388671875, + "learning_rate": 0.001934656110920096, + "loss": 0.1045, + "step": 12595 + }, + { + "epoch": 0.10933932865166102, + "grad_norm": 0.3203125, + "learning_rate": 0.0019346449768678723, + "loss": 0.106, + "step": 12596 + }, + { + "epoch": 0.10934800913186518, + "grad_norm": 0.62109375, + "learning_rate": 0.0019346338419028932, + "loss": 0.1992, + "step": 12597 + }, + { + "epoch": 0.10935668961206935, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019346227060251713, + "loss": 0.1367, + "step": 12598 + }, + { + "epoch": 0.1093653700922735, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019346115692347183, + "loss": 0.1201, + "step": 12599 + }, + { + "epoch": 0.10937405057247766, + "grad_norm": 0.07763671875, + "learning_rate": 0.001934600431531547, + "loss": 0.1191, + "step": 12600 + }, + { + "epoch": 0.10938273105268183, + "grad_norm": 0.59375, + "learning_rate": 0.001934589292915669, + "loss": 0.1328, + "step": 12601 + }, + { + "epoch": 0.109391411532886, + "grad_norm": 0.70703125, + "learning_rate": 0.0019345781533870966, + "loss": 0.1562, + "step": 12602 + }, + { + "epoch": 0.10940009201309016, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019345670129458422, + "loss": 0.1104, + "step": 12603 + }, + { + "epoch": 0.10940877249329432, + "grad_norm": 0.15234375, + "learning_rate": 0.001934555871591918, + "loss": 0.1084, + "step": 12604 + }, + { + "epoch": 0.10941745297349849, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019345447293253359, + "loss": 0.124, + "step": 12605 + }, + { + "epoch": 0.10942613345370265, + "grad_norm": 0.1875, + "learning_rate": 0.001934533586146108, + "loss": 0.104, + "step": 12606 + }, + { + "epoch": 0.10943481393390682, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019345224420542467, + "loss": 0.0962, + "step": 12607 + }, + { + "epoch": 0.10944349441411098, + "grad_norm": 0.345703125, + "learning_rate": 0.0019345112970497646, + "loss": 0.1299, + "step": 12608 + }, + { + "epoch": 0.10945217489431515, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019345001511326735, + "loss": 0.1445, + "step": 12609 + }, + { + "epoch": 0.10946085537451931, + "grad_norm": 0.61328125, + "learning_rate": 0.001934489004302985, + "loss": 0.125, + "step": 12610 + }, + { + "epoch": 0.10946953585472348, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019344778565607123, + "loss": 0.1064, + "step": 12611 + }, + { + "epoch": 0.10947821633492764, + "grad_norm": 0.9609375, + "learning_rate": 0.001934466707905867, + "loss": 0.1484, + "step": 12612 + }, + { + "epoch": 0.10948689681513181, + "grad_norm": 0.341796875, + "learning_rate": 0.0019344555583384615, + "loss": 0.0815, + "step": 12613 + }, + { + "epoch": 0.10949557729533597, + "grad_norm": 0.11767578125, + "learning_rate": 0.001934444407858508, + "loss": 0.1455, + "step": 12614 + }, + { + "epoch": 0.10950425777554014, + "grad_norm": 0.48046875, + "learning_rate": 0.0019344332564660185, + "loss": 0.1045, + "step": 12615 + }, + { + "epoch": 0.1095129382557443, + "grad_norm": 0.466796875, + "learning_rate": 0.0019344221041610054, + "loss": 0.1025, + "step": 12616 + }, + { + "epoch": 0.10952161873594847, + "grad_norm": 0.466796875, + "learning_rate": 0.0019344109509434808, + "loss": 0.1348, + "step": 12617 + }, + { + "epoch": 0.10953029921615264, + "grad_norm": 0.267578125, + "learning_rate": 0.0019343997968134573, + "loss": 0.127, + "step": 12618 + }, + { + "epoch": 0.1095389796963568, + "grad_norm": 0.65625, + "learning_rate": 0.0019343886417709463, + "loss": 0.0996, + "step": 12619 + }, + { + "epoch": 0.10954766017656097, + "grad_norm": 0.42578125, + "learning_rate": 0.0019343774858159603, + "loss": 0.1504, + "step": 12620 + }, + { + "epoch": 0.10955634065676513, + "grad_norm": 1.28125, + "learning_rate": 0.0019343663289485117, + "loss": 0.165, + "step": 12621 + }, + { + "epoch": 0.1095650211369693, + "grad_norm": 0.142578125, + "learning_rate": 0.001934355171168613, + "loss": 0.1836, + "step": 12622 + }, + { + "epoch": 0.10957370161717346, + "grad_norm": 0.455078125, + "learning_rate": 0.0019343440124762756, + "loss": 0.1055, + "step": 12623 + }, + { + "epoch": 0.10958238209737763, + "grad_norm": 0.80078125, + "learning_rate": 0.0019343328528715125, + "loss": 0.1699, + "step": 12624 + }, + { + "epoch": 0.10959106257758179, + "grad_norm": 0.30859375, + "learning_rate": 0.0019343216923543352, + "loss": 0.1309, + "step": 12625 + }, + { + "epoch": 0.10959974305778596, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019343105309247565, + "loss": 0.1182, + "step": 12626 + }, + { + "epoch": 0.10960842353799012, + "grad_norm": 0.52734375, + "learning_rate": 0.0019342993685827878, + "loss": 0.1138, + "step": 12627 + }, + { + "epoch": 0.10961710401819429, + "grad_norm": 0.60546875, + "learning_rate": 0.0019342882053284426, + "loss": 0.1348, + "step": 12628 + }, + { + "epoch": 0.10962578449839845, + "grad_norm": 0.400390625, + "learning_rate": 0.0019342770411617317, + "loss": 0.1084, + "step": 12629 + }, + { + "epoch": 0.10963446497860262, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019342658760826683, + "loss": 0.1182, + "step": 12630 + }, + { + "epoch": 0.10964314545880678, + "grad_norm": 0.29296875, + "learning_rate": 0.0019342547100912645, + "loss": 0.1016, + "step": 12631 + }, + { + "epoch": 0.10965182593901095, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019342435431875318, + "loss": 0.1309, + "step": 12632 + }, + { + "epoch": 0.10966050641921511, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019342323753714832, + "loss": 0.1143, + "step": 12633 + }, + { + "epoch": 0.10966918689941928, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019342212066431302, + "loss": 0.1138, + "step": 12634 + }, + { + "epoch": 0.10967786737962344, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019342100370024857, + "loss": 0.1104, + "step": 12635 + }, + { + "epoch": 0.10968654785982761, + "grad_norm": 0.3671875, + "learning_rate": 0.001934198866449562, + "loss": 0.1406, + "step": 12636 + }, + { + "epoch": 0.10969522834003177, + "grad_norm": 0.349609375, + "learning_rate": 0.0019341876949843704, + "loss": 0.0942, + "step": 12637 + }, + { + "epoch": 0.10970390882023594, + "grad_norm": 0.2109375, + "learning_rate": 0.0019341765226069238, + "loss": 0.1689, + "step": 12638 + }, + { + "epoch": 0.1097125893004401, + "grad_norm": 0.412109375, + "learning_rate": 0.0019341653493172345, + "loss": 0.1035, + "step": 12639 + }, + { + "epoch": 0.10972126978064427, + "grad_norm": 0.3046875, + "learning_rate": 0.0019341541751153142, + "loss": 0.1079, + "step": 12640 + }, + { + "epoch": 0.10972995026084843, + "grad_norm": 0.5390625, + "learning_rate": 0.0019341430000011756, + "loss": 0.1533, + "step": 12641 + }, + { + "epoch": 0.1097386307410526, + "grad_norm": 0.26953125, + "learning_rate": 0.0019341318239748308, + "loss": 0.1309, + "step": 12642 + }, + { + "epoch": 0.10974731122125676, + "grad_norm": 0.1005859375, + "learning_rate": 0.001934120647036292, + "loss": 0.1328, + "step": 12643 + }, + { + "epoch": 0.10975599170146093, + "grad_norm": 0.134765625, + "learning_rate": 0.001934109469185571, + "loss": 0.1089, + "step": 12644 + }, + { + "epoch": 0.1097646721816651, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019340982904226811, + "loss": 0.1016, + "step": 12645 + }, + { + "epoch": 0.10977335266186926, + "grad_norm": 0.5703125, + "learning_rate": 0.0019340871107476334, + "loss": 0.1084, + "step": 12646 + }, + { + "epoch": 0.10978203314207342, + "grad_norm": 0.58984375, + "learning_rate": 0.0019340759301604408, + "loss": 0.1094, + "step": 12647 + }, + { + "epoch": 0.10979071362227759, + "grad_norm": 0.1708984375, + "learning_rate": 0.001934064748661115, + "loss": 0.1523, + "step": 12648 + }, + { + "epoch": 0.10979939410248175, + "grad_norm": 0.2421875, + "learning_rate": 0.0019340535662496687, + "loss": 0.104, + "step": 12649 + }, + { + "epoch": 0.10980807458268592, + "grad_norm": 0.33984375, + "learning_rate": 0.001934042382926114, + "loss": 0.1113, + "step": 12650 + }, + { + "epoch": 0.10981675506289008, + "grad_norm": 0.375, + "learning_rate": 0.0019340311986904629, + "loss": 0.1211, + "step": 12651 + }, + { + "epoch": 0.10982543554309425, + "grad_norm": 0.1259765625, + "learning_rate": 0.001934020013542728, + "loss": 0.1289, + "step": 12652 + }, + { + "epoch": 0.10983411602329841, + "grad_norm": 0.2734375, + "learning_rate": 0.001934008827482921, + "loss": 0.1299, + "step": 12653 + }, + { + "epoch": 0.10984279650350258, + "grad_norm": 0.09375, + "learning_rate": 0.0019339976405110548, + "loss": 0.0815, + "step": 12654 + }, + { + "epoch": 0.10985147698370674, + "grad_norm": 0.216796875, + "learning_rate": 0.0019339864526271414, + "loss": 0.1191, + "step": 12655 + }, + { + "epoch": 0.10986015746391091, + "grad_norm": 0.24609375, + "learning_rate": 0.0019339752638311929, + "loss": 0.0928, + "step": 12656 + }, + { + "epoch": 0.10986883794411507, + "grad_norm": 0.23828125, + "learning_rate": 0.0019339640741232213, + "loss": 0.1387, + "step": 12657 + }, + { + "epoch": 0.10987751842431924, + "grad_norm": 1.1171875, + "learning_rate": 0.0019339528835032392, + "loss": 0.127, + "step": 12658 + }, + { + "epoch": 0.1098861989045234, + "grad_norm": 0.1748046875, + "learning_rate": 0.001933941691971259, + "loss": 0.1338, + "step": 12659 + }, + { + "epoch": 0.10989487938472756, + "grad_norm": 0.140625, + "learning_rate": 0.0019339304995272923, + "loss": 0.1104, + "step": 12660 + }, + { + "epoch": 0.10990355986493172, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019339193061713522, + "loss": 0.1226, + "step": 12661 + }, + { + "epoch": 0.10991224034513589, + "grad_norm": 0.1962890625, + "learning_rate": 0.00193390811190345, + "loss": 0.0933, + "step": 12662 + }, + { + "epoch": 0.10992092082534005, + "grad_norm": 0.11572265625, + "learning_rate": 0.001933896916723599, + "loss": 0.1299, + "step": 12663 + }, + { + "epoch": 0.10992960130554422, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019338857206318103, + "loss": 0.1016, + "step": 12664 + }, + { + "epoch": 0.10993828178574838, + "grad_norm": 0.158203125, + "learning_rate": 0.001933874523628097, + "loss": 0.1367, + "step": 12665 + }, + { + "epoch": 0.10994696226595255, + "grad_norm": 0.54296875, + "learning_rate": 0.001933863325712471, + "loss": 0.084, + "step": 12666 + }, + { + "epoch": 0.10995564274615671, + "grad_norm": 1.2890625, + "learning_rate": 0.0019338521268849444, + "loss": 0.1445, + "step": 12667 + }, + { + "epoch": 0.10996432322636088, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019338409271455299, + "loss": 0.1162, + "step": 12668 + }, + { + "epoch": 0.10997300370656504, + "grad_norm": 0.166015625, + "learning_rate": 0.0019338297264942393, + "loss": 0.1182, + "step": 12669 + }, + { + "epoch": 0.10998168418676921, + "grad_norm": 0.474609375, + "learning_rate": 0.0019338185249310853, + "loss": 0.1797, + "step": 12670 + }, + { + "epoch": 0.10999036466697337, + "grad_norm": 0.35546875, + "learning_rate": 0.00193380732245608, + "loss": 0.1055, + "step": 12671 + }, + { + "epoch": 0.10999904514717754, + "grad_norm": 0.8359375, + "learning_rate": 0.001933796119069235, + "loss": 0.1289, + "step": 12672 + }, + { + "epoch": 0.1100077256273817, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019337849147705634, + "loss": 0.0869, + "step": 12673 + }, + { + "epoch": 0.11001640610758587, + "grad_norm": 0.21484375, + "learning_rate": 0.001933773709560077, + "loss": 0.1074, + "step": 12674 + }, + { + "epoch": 0.11002508658779003, + "grad_norm": 0.328125, + "learning_rate": 0.0019337625034377883, + "loss": 0.1074, + "step": 12675 + }, + { + "epoch": 0.1100337670679942, + "grad_norm": 0.427734375, + "learning_rate": 0.0019337512964037096, + "loss": 0.1016, + "step": 12676 + }, + { + "epoch": 0.11004244754819836, + "grad_norm": 0.361328125, + "learning_rate": 0.0019337400884578528, + "loss": 0.1328, + "step": 12677 + }, + { + "epoch": 0.11005112802840253, + "grad_norm": 0.43359375, + "learning_rate": 0.0019337288796002306, + "loss": 0.1484, + "step": 12678 + }, + { + "epoch": 0.1100598085086067, + "grad_norm": 0.1484375, + "learning_rate": 0.0019337176698308548, + "loss": 0.0928, + "step": 12679 + }, + { + "epoch": 0.11006848898881086, + "grad_norm": 0.375, + "learning_rate": 0.001933706459149738, + "loss": 0.125, + "step": 12680 + }, + { + "epoch": 0.11007716946901502, + "grad_norm": 0.51171875, + "learning_rate": 0.0019336952475568923, + "loss": 0.1328, + "step": 12681 + }, + { + "epoch": 0.11008584994921919, + "grad_norm": 0.1875, + "learning_rate": 0.00193368403505233, + "loss": 0.0928, + "step": 12682 + }, + { + "epoch": 0.11009453042942335, + "grad_norm": 0.427734375, + "learning_rate": 0.0019336728216360632, + "loss": 0.1104, + "step": 12683 + }, + { + "epoch": 0.11010321090962752, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019336616073081046, + "loss": 0.1016, + "step": 12684 + }, + { + "epoch": 0.11011189138983168, + "grad_norm": 0.1728515625, + "learning_rate": 0.001933650392068466, + "loss": 0.1084, + "step": 12685 + }, + { + "epoch": 0.11012057187003585, + "grad_norm": 0.1103515625, + "learning_rate": 0.00193363917591716, + "loss": 0.1523, + "step": 12686 + }, + { + "epoch": 0.11012925235024001, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019336279588541988, + "loss": 0.0903, + "step": 12687 + }, + { + "epoch": 0.11013793283044418, + "grad_norm": 0.3203125, + "learning_rate": 0.0019336167408795944, + "loss": 0.1069, + "step": 12688 + }, + { + "epoch": 0.11014661331064834, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019336055219933593, + "loss": 0.1445, + "step": 12689 + }, + { + "epoch": 0.11015529379085251, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019335943021955057, + "loss": 0.0991, + "step": 12690 + }, + { + "epoch": 0.11016397427105667, + "grad_norm": 0.154296875, + "learning_rate": 0.0019335830814860458, + "loss": 0.1436, + "step": 12691 + }, + { + "epoch": 0.11017265475126084, + "grad_norm": 0.21484375, + "learning_rate": 0.0019335718598649922, + "loss": 0.125, + "step": 12692 + }, + { + "epoch": 0.110181335231465, + "grad_norm": 0.58984375, + "learning_rate": 0.0019335606373323568, + "loss": 0.1416, + "step": 12693 + }, + { + "epoch": 0.11019001571166917, + "grad_norm": 0.65625, + "learning_rate": 0.0019335494138881522, + "loss": 0.1172, + "step": 12694 + }, + { + "epoch": 0.11019869619187334, + "grad_norm": 0.28125, + "learning_rate": 0.0019335381895323903, + "loss": 0.1006, + "step": 12695 + }, + { + "epoch": 0.1102073766720775, + "grad_norm": 0.353515625, + "learning_rate": 0.0019335269642650837, + "loss": 0.1621, + "step": 12696 + }, + { + "epoch": 0.11021605715228167, + "grad_norm": 0.353515625, + "learning_rate": 0.0019335157380862443, + "loss": 0.1387, + "step": 12697 + }, + { + "epoch": 0.11022473763248583, + "grad_norm": 0.154296875, + "learning_rate": 0.0019335045109958847, + "loss": 0.0771, + "step": 12698 + }, + { + "epoch": 0.11023341811269, + "grad_norm": 0.50390625, + "learning_rate": 0.001933493282994017, + "loss": 0.125, + "step": 12699 + }, + { + "epoch": 0.11024209859289416, + "grad_norm": 0.234375, + "learning_rate": 0.0019334820540806538, + "loss": 0.0957, + "step": 12700 + }, + { + "epoch": 0.11025077907309833, + "grad_norm": 0.341796875, + "learning_rate": 0.001933470824255807, + "loss": 0.0962, + "step": 12701 + }, + { + "epoch": 0.11025945955330249, + "grad_norm": 0.177734375, + "learning_rate": 0.0019334595935194892, + "loss": 0.1221, + "step": 12702 + }, + { + "epoch": 0.11026814003350666, + "grad_norm": 0.1171875, + "learning_rate": 0.0019334483618717122, + "loss": 0.0952, + "step": 12703 + }, + { + "epoch": 0.11027682051371082, + "grad_norm": 0.30859375, + "learning_rate": 0.0019334371293124888, + "loss": 0.1523, + "step": 12704 + }, + { + "epoch": 0.11028550099391499, + "grad_norm": 0.380859375, + "learning_rate": 0.0019334258958418312, + "loss": 0.084, + "step": 12705 + }, + { + "epoch": 0.11029418147411915, + "grad_norm": 0.353515625, + "learning_rate": 0.0019334146614597512, + "loss": 0.1123, + "step": 12706 + }, + { + "epoch": 0.11030286195432332, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019334034261662618, + "loss": 0.1133, + "step": 12707 + }, + { + "epoch": 0.11031154243452748, + "grad_norm": 1.09375, + "learning_rate": 0.0019333921899613747, + "loss": 0.0977, + "step": 12708 + }, + { + "epoch": 0.11032022291473165, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019333809528451027, + "loss": 0.0898, + "step": 12709 + }, + { + "epoch": 0.11032890339493581, + "grad_norm": 0.68359375, + "learning_rate": 0.0019333697148174575, + "loss": 0.1138, + "step": 12710 + }, + { + "epoch": 0.11033758387513998, + "grad_norm": 0.21484375, + "learning_rate": 0.001933358475878452, + "loss": 0.0781, + "step": 12711 + }, + { + "epoch": 0.11034626435534414, + "grad_norm": 0.318359375, + "learning_rate": 0.001933347236028098, + "loss": 0.1006, + "step": 12712 + }, + { + "epoch": 0.11035494483554831, + "grad_norm": 0.369140625, + "learning_rate": 0.001933335995266408, + "loss": 0.0815, + "step": 12713 + }, + { + "epoch": 0.11036362531575247, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019333247535933942, + "loss": 0.1001, + "step": 12714 + }, + { + "epoch": 0.11037230579595664, + "grad_norm": 0.50390625, + "learning_rate": 0.0019333135110090691, + "loss": 0.0938, + "step": 12715 + }, + { + "epoch": 0.1103809862761608, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019333022675134448, + "loss": 0.126, + "step": 12716 + }, + { + "epoch": 0.11038966675636497, + "grad_norm": 0.1484375, + "learning_rate": 0.0019332910231065338, + "loss": 0.1328, + "step": 12717 + }, + { + "epoch": 0.11039834723656913, + "grad_norm": 0.90234375, + "learning_rate": 0.0019332797777883483, + "loss": 0.1035, + "step": 12718 + }, + { + "epoch": 0.1104070277167733, + "grad_norm": 0.318359375, + "learning_rate": 0.0019332685315589003, + "loss": 0.1348, + "step": 12719 + }, + { + "epoch": 0.11041570819697746, + "grad_norm": 0.95703125, + "learning_rate": 0.0019332572844182022, + "loss": 0.1914, + "step": 12720 + }, + { + "epoch": 0.11042438867718163, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019332460363662668, + "loss": 0.125, + "step": 12721 + }, + { + "epoch": 0.11043306915738578, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019332347874031063, + "loss": 0.0996, + "step": 12722 + }, + { + "epoch": 0.11044174963758994, + "grad_norm": 0.6171875, + "learning_rate": 0.0019332235375287322, + "loss": 0.0957, + "step": 12723 + }, + { + "epoch": 0.11045043011779411, + "grad_norm": 0.34765625, + "learning_rate": 0.0019332122867431576, + "loss": 0.1445, + "step": 12724 + }, + { + "epoch": 0.11045911059799828, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019332010350463944, + "loss": 0.1094, + "step": 12725 + }, + { + "epoch": 0.11046779107820244, + "grad_norm": 0.271484375, + "learning_rate": 0.0019331897824384553, + "loss": 0.1055, + "step": 12726 + }, + { + "epoch": 0.1104764715584066, + "grad_norm": 0.296875, + "learning_rate": 0.0019331785289193521, + "loss": 0.1074, + "step": 12727 + }, + { + "epoch": 0.11048515203861077, + "grad_norm": 0.734375, + "learning_rate": 0.0019331672744890979, + "loss": 0.1182, + "step": 12728 + }, + { + "epoch": 0.11049383251881494, + "grad_norm": 0.380859375, + "learning_rate": 0.001933156019147704, + "loss": 0.1758, + "step": 12729 + }, + { + "epoch": 0.1105025129990191, + "grad_norm": 0.326171875, + "learning_rate": 0.0019331447628951834, + "loss": 0.1211, + "step": 12730 + }, + { + "epoch": 0.11051119347922327, + "grad_norm": 0.333984375, + "learning_rate": 0.001933133505731548, + "loss": 0.1396, + "step": 12731 + }, + { + "epoch": 0.11051987395942743, + "grad_norm": 0.37109375, + "learning_rate": 0.0019331222476568106, + "loss": 0.1006, + "step": 12732 + }, + { + "epoch": 0.1105285544396316, + "grad_norm": 0.640625, + "learning_rate": 0.001933110988670983, + "loss": 0.1592, + "step": 12733 + }, + { + "epoch": 0.11053723491983576, + "grad_norm": 0.078125, + "learning_rate": 0.0019330997287740777, + "loss": 0.1426, + "step": 12734 + }, + { + "epoch": 0.11054591540003993, + "grad_norm": 0.89453125, + "learning_rate": 0.0019330884679661072, + "loss": 0.0898, + "step": 12735 + }, + { + "epoch": 0.11055459588024409, + "grad_norm": 0.412109375, + "learning_rate": 0.001933077206247084, + "loss": 0.0815, + "step": 12736 + }, + { + "epoch": 0.11056327636044826, + "grad_norm": 0.61328125, + "learning_rate": 0.0019330659436170195, + "loss": 0.0889, + "step": 12737 + }, + { + "epoch": 0.11057195684065242, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019330546800759268, + "loss": 0.1006, + "step": 12738 + }, + { + "epoch": 0.11058063732085659, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019330434156238179, + "loss": 0.1367, + "step": 12739 + }, + { + "epoch": 0.11058931780106075, + "grad_norm": 0.318359375, + "learning_rate": 0.0019330321502607055, + "loss": 0.1318, + "step": 12740 + }, + { + "epoch": 0.11059799828126492, + "grad_norm": 0.5546875, + "learning_rate": 0.0019330208839866017, + "loss": 0.1172, + "step": 12741 + }, + { + "epoch": 0.11060667876146908, + "grad_norm": 0.18359375, + "learning_rate": 0.0019330096168015188, + "loss": 0.1094, + "step": 12742 + }, + { + "epoch": 0.11061535924167325, + "grad_norm": 0.353515625, + "learning_rate": 0.0019329983487054686, + "loss": 0.1348, + "step": 12743 + }, + { + "epoch": 0.11062403972187741, + "grad_norm": 0.51171875, + "learning_rate": 0.0019329870796984644, + "loss": 0.1436, + "step": 12744 + }, + { + "epoch": 0.11063272020208158, + "grad_norm": 0.66015625, + "learning_rate": 0.001932975809780518, + "loss": 0.1504, + "step": 12745 + }, + { + "epoch": 0.11064140068228574, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019329645389516415, + "loss": 0.1021, + "step": 12746 + }, + { + "epoch": 0.11065008116248991, + "grad_norm": 0.32421875, + "learning_rate": 0.0019329532672118475, + "loss": 0.0967, + "step": 12747 + }, + { + "epoch": 0.11065876164269407, + "grad_norm": 1.2734375, + "learning_rate": 0.0019329419945611486, + "loss": 0.106, + "step": 12748 + }, + { + "epoch": 0.11066744212289824, + "grad_norm": 0.65625, + "learning_rate": 0.0019329307209995567, + "loss": 0.2051, + "step": 12749 + }, + { + "epoch": 0.1106761226031024, + "grad_norm": 0.439453125, + "learning_rate": 0.0019329194465270844, + "loss": 0.1016, + "step": 12750 + }, + { + "epoch": 0.11068480308330657, + "grad_norm": 0.208984375, + "learning_rate": 0.0019329081711437438, + "loss": 0.1113, + "step": 12751 + }, + { + "epoch": 0.11069348356351073, + "grad_norm": 1.0546875, + "learning_rate": 0.0019328968948495472, + "loss": 0.1309, + "step": 12752 + }, + { + "epoch": 0.1107021640437149, + "grad_norm": 0.330078125, + "learning_rate": 0.0019328856176445072, + "loss": 0.1074, + "step": 12753 + }, + { + "epoch": 0.11071084452391906, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019328743395286361, + "loss": 0.1182, + "step": 12754 + }, + { + "epoch": 0.11071952500412323, + "grad_norm": 0.15625, + "learning_rate": 0.001932863060501946, + "loss": 0.1074, + "step": 12755 + }, + { + "epoch": 0.1107282054843274, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019328517805644496, + "loss": 0.1436, + "step": 12756 + }, + { + "epoch": 0.11073688596453156, + "grad_norm": 0.126953125, + "learning_rate": 0.0019328404997161588, + "loss": 0.1611, + "step": 12757 + }, + { + "epoch": 0.11074556644473572, + "grad_norm": 0.10302734375, + "learning_rate": 0.001932829217957086, + "loss": 0.123, + "step": 12758 + }, + { + "epoch": 0.11075424692493989, + "grad_norm": 0.11669921875, + "learning_rate": 0.001932817935287244, + "loss": 0.1289, + "step": 12759 + }, + { + "epoch": 0.11076292740514405, + "grad_norm": 0.7265625, + "learning_rate": 0.0019328066517066444, + "loss": 0.1055, + "step": 12760 + }, + { + "epoch": 0.11077160788534822, + "grad_norm": 0.58203125, + "learning_rate": 0.0019327953672153004, + "loss": 0.1309, + "step": 12761 + }, + { + "epoch": 0.11078028836555238, + "grad_norm": 0.412109375, + "learning_rate": 0.0019327840818132238, + "loss": 0.125, + "step": 12762 + }, + { + "epoch": 0.11078896884575655, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019327727955004272, + "loss": 0.1396, + "step": 12763 + }, + { + "epoch": 0.11079764932596071, + "grad_norm": 0.6484375, + "learning_rate": 0.0019327615082769226, + "loss": 0.1016, + "step": 12764 + }, + { + "epoch": 0.11080632980616488, + "grad_norm": 0.29296875, + "learning_rate": 0.0019327502201427222, + "loss": 0.1562, + "step": 12765 + }, + { + "epoch": 0.11081501028636905, + "grad_norm": 0.65234375, + "learning_rate": 0.001932738931097839, + "loss": 0.0967, + "step": 12766 + }, + { + "epoch": 0.11082369076657321, + "grad_norm": 0.59765625, + "learning_rate": 0.0019327276411422848, + "loss": 0.126, + "step": 12767 + }, + { + "epoch": 0.11083237124677738, + "grad_norm": 0.33203125, + "learning_rate": 0.0019327163502760728, + "loss": 0.1289, + "step": 12768 + }, + { + "epoch": 0.11084105172698154, + "grad_norm": 0.14453125, + "learning_rate": 0.0019327050584992138, + "loss": 0.1689, + "step": 12769 + }, + { + "epoch": 0.1108497322071857, + "grad_norm": 0.279296875, + "learning_rate": 0.0019326937658117216, + "loss": 0.1328, + "step": 12770 + }, + { + "epoch": 0.11085841268738987, + "grad_norm": 0.53125, + "learning_rate": 0.001932682472213608, + "loss": 0.1357, + "step": 12771 + }, + { + "epoch": 0.11086709316759404, + "grad_norm": 0.341796875, + "learning_rate": 0.0019326711777048853, + "loss": 0.123, + "step": 12772 + }, + { + "epoch": 0.1108757736477982, + "grad_norm": 0.380859375, + "learning_rate": 0.001932659882285566, + "loss": 0.1035, + "step": 12773 + }, + { + "epoch": 0.11088445412800237, + "grad_norm": 0.0849609375, + "learning_rate": 0.001932648585955662, + "loss": 0.0933, + "step": 12774 + }, + { + "epoch": 0.11089313460820653, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019326372887151866, + "loss": 0.1504, + "step": 12775 + }, + { + "epoch": 0.1109018150884107, + "grad_norm": 0.48046875, + "learning_rate": 0.001932625990564151, + "loss": 0.1011, + "step": 12776 + }, + { + "epoch": 0.11091049556861486, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019326146915025683, + "loss": 0.0928, + "step": 12777 + }, + { + "epoch": 0.11091917604881903, + "grad_norm": 0.435546875, + "learning_rate": 0.0019326033915304508, + "loss": 0.126, + "step": 12778 + }, + { + "epoch": 0.11092785652902319, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019325920906478108, + "loss": 0.1084, + "step": 12779 + }, + { + "epoch": 0.11093653700922736, + "grad_norm": 0.216796875, + "learning_rate": 0.0019325807888546606, + "loss": 0.0869, + "step": 12780 + }, + { + "epoch": 0.11094521748943152, + "grad_norm": 0.671875, + "learning_rate": 0.0019325694861510121, + "loss": 0.0991, + "step": 12781 + }, + { + "epoch": 0.11095389796963569, + "grad_norm": 0.58203125, + "learning_rate": 0.0019325581825368789, + "loss": 0.1279, + "step": 12782 + }, + { + "epoch": 0.11096257844983984, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019325468780122719, + "loss": 0.1025, + "step": 12783 + }, + { + "epoch": 0.110971258930044, + "grad_norm": 0.46484375, + "learning_rate": 0.0019325355725772042, + "loss": 0.1035, + "step": 12784 + }, + { + "epoch": 0.11097993941024817, + "grad_norm": 0.091796875, + "learning_rate": 0.0019325242662316885, + "loss": 0.0981, + "step": 12785 + }, + { + "epoch": 0.11098861989045233, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019325129589757367, + "loss": 0.1318, + "step": 12786 + }, + { + "epoch": 0.1109973003706565, + "grad_norm": 0.310546875, + "learning_rate": 0.0019325016508093611, + "loss": 0.126, + "step": 12787 + }, + { + "epoch": 0.11100598085086066, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019324903417325741, + "loss": 0.1152, + "step": 12788 + }, + { + "epoch": 0.11101466133106483, + "grad_norm": 0.2734375, + "learning_rate": 0.0019324790317453882, + "loss": 0.0981, + "step": 12789 + }, + { + "epoch": 0.111023341811269, + "grad_norm": 0.734375, + "learning_rate": 0.001932467720847816, + "loss": 0.0957, + "step": 12790 + }, + { + "epoch": 0.11103202229147316, + "grad_norm": 0.142578125, + "learning_rate": 0.0019324564090398692, + "loss": 0.1299, + "step": 12791 + }, + { + "epoch": 0.11104070277167732, + "grad_norm": 1.046875, + "learning_rate": 0.001932445096321561, + "loss": 0.123, + "step": 12792 + }, + { + "epoch": 0.11104938325188149, + "grad_norm": 0.125, + "learning_rate": 0.001932433782692903, + "loss": 0.0698, + "step": 12793 + }, + { + "epoch": 0.11105806373208565, + "grad_norm": 0.1484375, + "learning_rate": 0.0019324224681539083, + "loss": 0.1191, + "step": 12794 + }, + { + "epoch": 0.11106674421228982, + "grad_norm": 0.068359375, + "learning_rate": 0.0019324111527045885, + "loss": 0.084, + "step": 12795 + }, + { + "epoch": 0.11107542469249398, + "grad_norm": 0.5078125, + "learning_rate": 0.0019323998363449564, + "loss": 0.1094, + "step": 12796 + }, + { + "epoch": 0.11108410517269815, + "grad_norm": 0.41796875, + "learning_rate": 0.0019323885190750247, + "loss": 0.1396, + "step": 12797 + }, + { + "epoch": 0.11109278565290231, + "grad_norm": 0.087890625, + "learning_rate": 0.001932377200894805, + "loss": 0.1133, + "step": 12798 + }, + { + "epoch": 0.11110146613310648, + "grad_norm": 0.65625, + "learning_rate": 0.0019323658818043107, + "loss": 0.1426, + "step": 12799 + }, + { + "epoch": 0.11111014661331065, + "grad_norm": 0.0947265625, + "learning_rate": 0.001932354561803553, + "loss": 0.0811, + "step": 12800 + }, + { + "epoch": 0.11111882709351481, + "grad_norm": 0.60546875, + "learning_rate": 0.001932343240892545, + "loss": 0.1348, + "step": 12801 + }, + { + "epoch": 0.11112750757371898, + "grad_norm": 0.625, + "learning_rate": 0.0019323319190712991, + "loss": 0.0996, + "step": 12802 + }, + { + "epoch": 0.11113618805392314, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019323205963398273, + "loss": 0.1484, + "step": 12803 + }, + { + "epoch": 0.1111448685341273, + "grad_norm": 0.353515625, + "learning_rate": 0.0019323092726981424, + "loss": 0.1494, + "step": 12804 + }, + { + "epoch": 0.11115354901433147, + "grad_norm": 0.287109375, + "learning_rate": 0.0019322979481462563, + "loss": 0.1475, + "step": 12805 + }, + { + "epoch": 0.11116222949453564, + "grad_norm": 0.1640625, + "learning_rate": 0.001932286622684182, + "loss": 0.125, + "step": 12806 + }, + { + "epoch": 0.1111709099747398, + "grad_norm": 0.130859375, + "learning_rate": 0.0019322752963119314, + "loss": 0.0874, + "step": 12807 + }, + { + "epoch": 0.11117959045494397, + "grad_norm": 0.5546875, + "learning_rate": 0.001932263969029517, + "loss": 0.1099, + "step": 12808 + }, + { + "epoch": 0.11118827093514813, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019322526408369514, + "loss": 0.126, + "step": 12809 + }, + { + "epoch": 0.1111969514153523, + "grad_norm": 0.453125, + "learning_rate": 0.0019322413117342466, + "loss": 0.1289, + "step": 12810 + }, + { + "epoch": 0.11120563189555646, + "grad_norm": 0.5546875, + "learning_rate": 0.0019322299817214152, + "loss": 0.1191, + "step": 12811 + }, + { + "epoch": 0.11121431237576063, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019322186507984699, + "loss": 0.1152, + "step": 12812 + }, + { + "epoch": 0.11122299285596479, + "grad_norm": 0.7421875, + "learning_rate": 0.0019322073189654222, + "loss": 0.1484, + "step": 12813 + }, + { + "epoch": 0.11123167333616896, + "grad_norm": 0.126953125, + "learning_rate": 0.0019321959862222857, + "loss": 0.123, + "step": 12814 + }, + { + "epoch": 0.11124035381637312, + "grad_norm": 0.93359375, + "learning_rate": 0.0019321846525690721, + "loss": 0.1602, + "step": 12815 + }, + { + "epoch": 0.11124903429657729, + "grad_norm": 0.392578125, + "learning_rate": 0.0019321733180057937, + "loss": 0.1699, + "step": 12816 + }, + { + "epoch": 0.11125771477678145, + "grad_norm": 0.71875, + "learning_rate": 0.001932161982532463, + "loss": 0.1699, + "step": 12817 + }, + { + "epoch": 0.11126639525698562, + "grad_norm": 0.09765625, + "learning_rate": 0.0019321506461490924, + "loss": 0.0801, + "step": 12818 + }, + { + "epoch": 0.11127507573718978, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019321393088556945, + "loss": 0.0879, + "step": 12819 + }, + { + "epoch": 0.11128375621739395, + "grad_norm": 0.41015625, + "learning_rate": 0.0019321279706522817, + "loss": 0.1235, + "step": 12820 + }, + { + "epoch": 0.11129243669759811, + "grad_norm": 0.08447265625, + "learning_rate": 0.001932116631538866, + "loss": 0.1094, + "step": 12821 + }, + { + "epoch": 0.11130111717780228, + "grad_norm": 0.203125, + "learning_rate": 0.0019321052915154601, + "loss": 0.1309, + "step": 12822 + }, + { + "epoch": 0.11130979765800644, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019320939505820766, + "loss": 0.1348, + "step": 12823 + }, + { + "epoch": 0.11131847813821061, + "grad_norm": 0.60546875, + "learning_rate": 0.0019320826087387273, + "loss": 0.123, + "step": 12824 + }, + { + "epoch": 0.11132715861841477, + "grad_norm": 0.25, + "learning_rate": 0.0019320712659854251, + "loss": 0.1162, + "step": 12825 + }, + { + "epoch": 0.11133583909861894, + "grad_norm": 0.328125, + "learning_rate": 0.0019320599223221825, + "loss": 0.1328, + "step": 12826 + }, + { + "epoch": 0.1113445195788231, + "grad_norm": 0.62890625, + "learning_rate": 0.0019320485777490112, + "loss": 0.1582, + "step": 12827 + }, + { + "epoch": 0.11135320005902727, + "grad_norm": 0.35546875, + "learning_rate": 0.0019320372322659245, + "loss": 0.0933, + "step": 12828 + }, + { + "epoch": 0.11136188053923143, + "grad_norm": 0.15625, + "learning_rate": 0.0019320258858729342, + "loss": 0.1387, + "step": 12829 + }, + { + "epoch": 0.1113705610194356, + "grad_norm": 0.369140625, + "learning_rate": 0.0019320145385700528, + "loss": 0.1172, + "step": 12830 + }, + { + "epoch": 0.11137924149963976, + "grad_norm": 0.5703125, + "learning_rate": 0.001932003190357293, + "loss": 0.1318, + "step": 12831 + }, + { + "epoch": 0.11138792197984393, + "grad_norm": 0.259765625, + "learning_rate": 0.0019319918412346669, + "loss": 0.1162, + "step": 12832 + }, + { + "epoch": 0.1113966024600481, + "grad_norm": 0.74609375, + "learning_rate": 0.001931980491202187, + "loss": 0.1416, + "step": 12833 + }, + { + "epoch": 0.11140528294025226, + "grad_norm": 0.111328125, + "learning_rate": 0.0019319691402598657, + "loss": 0.1123, + "step": 12834 + }, + { + "epoch": 0.11141396342045642, + "grad_norm": 0.28125, + "learning_rate": 0.0019319577884077156, + "loss": 0.1016, + "step": 12835 + }, + { + "epoch": 0.11142264390066059, + "grad_norm": 0.140625, + "learning_rate": 0.001931946435645749, + "loss": 0.1035, + "step": 12836 + }, + { + "epoch": 0.11143132438086475, + "grad_norm": 0.5078125, + "learning_rate": 0.0019319350819739782, + "loss": 0.1289, + "step": 12837 + }, + { + "epoch": 0.11144000486106892, + "grad_norm": 0.76171875, + "learning_rate": 0.0019319237273924155, + "loss": 0.1768, + "step": 12838 + }, + { + "epoch": 0.11144868534127308, + "grad_norm": 0.53125, + "learning_rate": 0.0019319123719010738, + "loss": 0.1279, + "step": 12839 + }, + { + "epoch": 0.11145736582147725, + "grad_norm": 0.353515625, + "learning_rate": 0.001931901015499965, + "loss": 0.1084, + "step": 12840 + }, + { + "epoch": 0.11146604630168142, + "grad_norm": 0.1845703125, + "learning_rate": 0.001931889658189102, + "loss": 0.1177, + "step": 12841 + }, + { + "epoch": 0.11147472678188558, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019318782999684968, + "loss": 0.0928, + "step": 12842 + }, + { + "epoch": 0.11148340726208975, + "grad_norm": 0.2255859375, + "learning_rate": 0.001931866940838162, + "loss": 0.124, + "step": 12843 + }, + { + "epoch": 0.11149208774229391, + "grad_norm": 0.12109375, + "learning_rate": 0.00193185558079811, + "loss": 0.1016, + "step": 12844 + }, + { + "epoch": 0.11150076822249806, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019318442198483533, + "loss": 0.1074, + "step": 12845 + }, + { + "epoch": 0.11150944870270223, + "grad_norm": 1.046875, + "learning_rate": 0.0019318328579889042, + "loss": 0.1309, + "step": 12846 + }, + { + "epoch": 0.11151812918290639, + "grad_norm": 0.2265625, + "learning_rate": 0.0019318214952197751, + "loss": 0.2461, + "step": 12847 + }, + { + "epoch": 0.11152680966311056, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019318101315409786, + "loss": 0.082, + "step": 12848 + }, + { + "epoch": 0.11153549014331472, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019317987669525273, + "loss": 0.1318, + "step": 12849 + }, + { + "epoch": 0.11154417062351889, + "grad_norm": 0.2734375, + "learning_rate": 0.001931787401454433, + "loss": 0.1025, + "step": 12850 + }, + { + "epoch": 0.11155285110372305, + "grad_norm": 0.248046875, + "learning_rate": 0.0019317760350467087, + "loss": 0.0898, + "step": 12851 + }, + { + "epoch": 0.11156153158392722, + "grad_norm": 0.62109375, + "learning_rate": 0.0019317646677293665, + "loss": 0.1201, + "step": 12852 + }, + { + "epoch": 0.11157021206413138, + "grad_norm": 0.53125, + "learning_rate": 0.001931753299502419, + "loss": 0.1416, + "step": 12853 + }, + { + "epoch": 0.11157889254433555, + "grad_norm": 0.328125, + "learning_rate": 0.0019317419303658786, + "loss": 0.0977, + "step": 12854 + }, + { + "epoch": 0.11158757302453971, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019317305603197576, + "loss": 0.0903, + "step": 12855 + }, + { + "epoch": 0.11159625350474388, + "grad_norm": 0.37890625, + "learning_rate": 0.0019317191893640688, + "loss": 0.0918, + "step": 12856 + }, + { + "epoch": 0.11160493398494804, + "grad_norm": 0.2412109375, + "learning_rate": 0.001931707817498824, + "loss": 0.1045, + "step": 12857 + }, + { + "epoch": 0.11161361446515221, + "grad_norm": 5.75, + "learning_rate": 0.0019316964447240366, + "loss": 0.4863, + "step": 12858 + }, + { + "epoch": 0.11162229494535637, + "grad_norm": 0.10693359375, + "learning_rate": 0.001931685071039718, + "loss": 0.1543, + "step": 12859 + }, + { + "epoch": 0.11163097542556054, + "grad_norm": 0.26953125, + "learning_rate": 0.0019316736964458815, + "loss": 0.1113, + "step": 12860 + }, + { + "epoch": 0.1116396559057647, + "grad_norm": 0.27734375, + "learning_rate": 0.0019316623209425386, + "loss": 0.0903, + "step": 12861 + }, + { + "epoch": 0.11164833638596887, + "grad_norm": 0.82421875, + "learning_rate": 0.0019316509445297027, + "loss": 0.124, + "step": 12862 + }, + { + "epoch": 0.11165701686617303, + "grad_norm": 0.201171875, + "learning_rate": 0.0019316395672073856, + "loss": 0.1309, + "step": 12863 + }, + { + "epoch": 0.1116656973463772, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019316281889756003, + "loss": 0.1064, + "step": 12864 + }, + { + "epoch": 0.11167437782658136, + "grad_norm": 0.373046875, + "learning_rate": 0.0019316168098343586, + "loss": 0.1162, + "step": 12865 + }, + { + "epoch": 0.11168305830678553, + "grad_norm": 0.30078125, + "learning_rate": 0.0019316054297836731, + "loss": 0.127, + "step": 12866 + }, + { + "epoch": 0.1116917387869897, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019315940488235568, + "loss": 0.1143, + "step": 12867 + }, + { + "epoch": 0.11170041926719386, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019315826669540216, + "loss": 0.1152, + "step": 12868 + }, + { + "epoch": 0.11170909974739802, + "grad_norm": 0.181640625, + "learning_rate": 0.00193157128417508, + "loss": 0.126, + "step": 12869 + }, + { + "epoch": 0.11171778022760219, + "grad_norm": 0.4609375, + "learning_rate": 0.0019315599004867446, + "loss": 0.1299, + "step": 12870 + }, + { + "epoch": 0.11172646070780635, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019315485158890276, + "loss": 0.0952, + "step": 12871 + }, + { + "epoch": 0.11173514118801052, + "grad_norm": 0.396484375, + "learning_rate": 0.001931537130381942, + "loss": 0.123, + "step": 12872 + }, + { + "epoch": 0.11174382166821469, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019315257439654994, + "loss": 0.0952, + "step": 12873 + }, + { + "epoch": 0.11175250214841885, + "grad_norm": 0.53515625, + "learning_rate": 0.001931514356639713, + "loss": 0.1777, + "step": 12874 + }, + { + "epoch": 0.11176118262862302, + "grad_norm": 0.267578125, + "learning_rate": 0.0019315029684045953, + "loss": 0.1055, + "step": 12875 + }, + { + "epoch": 0.11176986310882718, + "grad_norm": 0.609375, + "learning_rate": 0.001931491579260158, + "loss": 0.1367, + "step": 12876 + }, + { + "epoch": 0.11177854358903135, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019314801892064142, + "loss": 0.0864, + "step": 12877 + }, + { + "epoch": 0.11178722406923551, + "grad_norm": 0.26953125, + "learning_rate": 0.0019314687982433761, + "loss": 0.106, + "step": 12878 + }, + { + "epoch": 0.11179590454943968, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019314574063710563, + "loss": 0.1079, + "step": 12879 + }, + { + "epoch": 0.11180458502964384, + "grad_norm": 0.859375, + "learning_rate": 0.0019314460135894673, + "loss": 0.1416, + "step": 12880 + }, + { + "epoch": 0.111813265509848, + "grad_norm": 0.4609375, + "learning_rate": 0.0019314346198986213, + "loss": 0.0996, + "step": 12881 + }, + { + "epoch": 0.11182194599005217, + "grad_norm": 0.3515625, + "learning_rate": 0.0019314232252985307, + "loss": 0.124, + "step": 12882 + }, + { + "epoch": 0.11183062647025634, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019314118297892085, + "loss": 0.1094, + "step": 12883 + }, + { + "epoch": 0.1118393069504605, + "grad_norm": 0.58984375, + "learning_rate": 0.0019314004333706666, + "loss": 0.1299, + "step": 12884 + }, + { + "epoch": 0.11184798743066467, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019313890360429175, + "loss": 0.0957, + "step": 12885 + }, + { + "epoch": 0.11185666791086883, + "grad_norm": 0.19140625, + "learning_rate": 0.0019313776378059742, + "loss": 0.1426, + "step": 12886 + }, + { + "epoch": 0.111865348391073, + "grad_norm": 0.306640625, + "learning_rate": 0.0019313662386598485, + "loss": 0.1074, + "step": 12887 + }, + { + "epoch": 0.11187402887127716, + "grad_norm": 0.4296875, + "learning_rate": 0.0019313548386045533, + "loss": 0.123, + "step": 12888 + }, + { + "epoch": 0.11188270935148133, + "grad_norm": 0.47265625, + "learning_rate": 0.001931343437640101, + "loss": 0.125, + "step": 12889 + }, + { + "epoch": 0.11189138983168549, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019313320357665041, + "loss": 0.1221, + "step": 12890 + }, + { + "epoch": 0.11190007031188966, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019313206329837746, + "loss": 0.1377, + "step": 12891 + }, + { + "epoch": 0.11190875079209382, + "grad_norm": 0.431640625, + "learning_rate": 0.0019313092292919257, + "loss": 0.1143, + "step": 12892 + }, + { + "epoch": 0.11191743127229799, + "grad_norm": 0.7734375, + "learning_rate": 0.0019312978246909696, + "loss": 0.1543, + "step": 12893 + }, + { + "epoch": 0.11192611175250215, + "grad_norm": 0.06787109375, + "learning_rate": 0.0019312864191809182, + "loss": 0.0981, + "step": 12894 + }, + { + "epoch": 0.11193479223270632, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019312750127617848, + "loss": 0.1504, + "step": 12895 + }, + { + "epoch": 0.11194347271291048, + "grad_norm": 0.21875, + "learning_rate": 0.0019312636054335814, + "loss": 0.0996, + "step": 12896 + }, + { + "epoch": 0.11195215319311465, + "grad_norm": 0.1416015625, + "learning_rate": 0.001931252197196321, + "loss": 0.1406, + "step": 12897 + }, + { + "epoch": 0.11196083367331881, + "grad_norm": 0.44140625, + "learning_rate": 0.0019312407880500153, + "loss": 0.1689, + "step": 12898 + }, + { + "epoch": 0.11196951415352298, + "grad_norm": 0.3515625, + "learning_rate": 0.001931229377994677, + "loss": 0.127, + "step": 12899 + }, + { + "epoch": 0.11197819463372714, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019312179670303194, + "loss": 0.1104, + "step": 12900 + }, + { + "epoch": 0.11198687511393131, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019312065551569538, + "loss": 0.0874, + "step": 12901 + }, + { + "epoch": 0.11199555559413547, + "grad_norm": 0.3671875, + "learning_rate": 0.0019311951423745934, + "loss": 0.1162, + "step": 12902 + }, + { + "epoch": 0.11200423607433964, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019311837286832502, + "loss": 0.0977, + "step": 12903 + }, + { + "epoch": 0.1120129165545438, + "grad_norm": 0.3515625, + "learning_rate": 0.001931172314082937, + "loss": 0.1064, + "step": 12904 + }, + { + "epoch": 0.11202159703474797, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019311608985736666, + "loss": 0.1367, + "step": 12905 + }, + { + "epoch": 0.11203027751495213, + "grad_norm": 0.228515625, + "learning_rate": 0.001931149482155451, + "loss": 0.1025, + "step": 12906 + }, + { + "epoch": 0.11203895799515629, + "grad_norm": 0.08203125, + "learning_rate": 0.0019311380648283027, + "loss": 0.1245, + "step": 12907 + }, + { + "epoch": 0.11204763847536045, + "grad_norm": 0.337890625, + "learning_rate": 0.0019311266465922342, + "loss": 0.1011, + "step": 12908 + }, + { + "epoch": 0.11205631895556462, + "grad_norm": 0.94140625, + "learning_rate": 0.0019311152274472583, + "loss": 0.1318, + "step": 12909 + }, + { + "epoch": 0.11206499943576878, + "grad_norm": 0.7109375, + "learning_rate": 0.0019311038073933874, + "loss": 0.0957, + "step": 12910 + }, + { + "epoch": 0.11207367991597295, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019310923864306335, + "loss": 0.1162, + "step": 12911 + }, + { + "epoch": 0.11208236039617711, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019310809645590097, + "loss": 0.1465, + "step": 12912 + }, + { + "epoch": 0.11209104087638128, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019310695417785282, + "loss": 0.1465, + "step": 12913 + }, + { + "epoch": 0.11209972135658544, + "grad_norm": 0.451171875, + "learning_rate": 0.0019310581180892014, + "loss": 0.1201, + "step": 12914 + }, + { + "epoch": 0.1121084018367896, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019310466934910422, + "loss": 0.0913, + "step": 12915 + }, + { + "epoch": 0.11211708231699377, + "grad_norm": 0.265625, + "learning_rate": 0.0019310352679840625, + "loss": 0.1201, + "step": 12916 + }, + { + "epoch": 0.11212576279719794, + "grad_norm": 0.345703125, + "learning_rate": 0.0019310238415682752, + "loss": 0.126, + "step": 12917 + }, + { + "epoch": 0.1121344432774021, + "grad_norm": 0.271484375, + "learning_rate": 0.001931012414243693, + "loss": 0.1299, + "step": 12918 + }, + { + "epoch": 0.11214312375760627, + "grad_norm": 0.24609375, + "learning_rate": 0.0019310009860103278, + "loss": 0.1465, + "step": 12919 + }, + { + "epoch": 0.11215180423781043, + "grad_norm": 0.98828125, + "learning_rate": 0.0019309895568681925, + "loss": 0.1875, + "step": 12920 + }, + { + "epoch": 0.1121604847180146, + "grad_norm": 0.234375, + "learning_rate": 0.0019309781268172994, + "loss": 0.1699, + "step": 12921 + }, + { + "epoch": 0.11216916519821876, + "grad_norm": 0.8984375, + "learning_rate": 0.0019309666958576613, + "loss": 0.0933, + "step": 12922 + }, + { + "epoch": 0.11217784567842293, + "grad_norm": 0.54296875, + "learning_rate": 0.0019309552639892905, + "loss": 0.1436, + "step": 12923 + }, + { + "epoch": 0.11218652615862709, + "grad_norm": 0.27734375, + "learning_rate": 0.0019309438312121994, + "loss": 0.1084, + "step": 12924 + }, + { + "epoch": 0.11219520663883126, + "grad_norm": 0.50390625, + "learning_rate": 0.0019309323975264007, + "loss": 0.1069, + "step": 12925 + }, + { + "epoch": 0.11220388711903542, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019309209629319063, + "loss": 0.1055, + "step": 12926 + }, + { + "epoch": 0.11221256759923959, + "grad_norm": 0.1201171875, + "learning_rate": 0.00193090952742873, + "loss": 0.1074, + "step": 12927 + }, + { + "epoch": 0.11222124807944375, + "grad_norm": 0.275390625, + "learning_rate": 0.0019308980910168832, + "loss": 0.103, + "step": 12928 + }, + { + "epoch": 0.11222992855964792, + "grad_norm": 0.1640625, + "learning_rate": 0.0019308866536963786, + "loss": 0.1064, + "step": 12929 + }, + { + "epoch": 0.11223860903985208, + "grad_norm": 0.59375, + "learning_rate": 0.0019308752154672294, + "loss": 0.1289, + "step": 12930 + }, + { + "epoch": 0.11224728952005625, + "grad_norm": 0.6875, + "learning_rate": 0.0019308637763294468, + "loss": 0.1328, + "step": 12931 + }, + { + "epoch": 0.11225597000026041, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019308523362830446, + "loss": 0.0786, + "step": 12932 + }, + { + "epoch": 0.11226465048046458, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019308408953280342, + "loss": 0.1768, + "step": 12933 + }, + { + "epoch": 0.11227333096066874, + "grad_norm": 0.34375, + "learning_rate": 0.0019308294534644292, + "loss": 0.167, + "step": 12934 + }, + { + "epoch": 0.11228201144087291, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019308180106922415, + "loss": 0.1143, + "step": 12935 + }, + { + "epoch": 0.11229069192107707, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019308065670114835, + "loss": 0.0938, + "step": 12936 + }, + { + "epoch": 0.11229937240128124, + "grad_norm": 0.578125, + "learning_rate": 0.001930795122422168, + "loss": 0.0986, + "step": 12937 + }, + { + "epoch": 0.1123080528814854, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019307836769243078, + "loss": 0.1084, + "step": 12938 + }, + { + "epoch": 0.11231673336168957, + "grad_norm": 0.427734375, + "learning_rate": 0.0019307722305179145, + "loss": 0.1001, + "step": 12939 + }, + { + "epoch": 0.11232541384189373, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019307607832030015, + "loss": 0.0938, + "step": 12940 + }, + { + "epoch": 0.1123340943220979, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019307493349795808, + "loss": 0.1177, + "step": 12941 + }, + { + "epoch": 0.11234277480230206, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019307378858476653, + "loss": 0.168, + "step": 12942 + }, + { + "epoch": 0.11235145528250623, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019307264358072672, + "loss": 0.1172, + "step": 12943 + }, + { + "epoch": 0.1123601357627104, + "grad_norm": 0.79296875, + "learning_rate": 0.001930714984858399, + "loss": 0.1221, + "step": 12944 + }, + { + "epoch": 0.11236881624291456, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019307035330010738, + "loss": 0.1099, + "step": 12945 + }, + { + "epoch": 0.11237749672311872, + "grad_norm": 0.27734375, + "learning_rate": 0.0019306920802353033, + "loss": 0.124, + "step": 12946 + }, + { + "epoch": 0.11238617720332289, + "grad_norm": 0.298828125, + "learning_rate": 0.0019306806265611005, + "loss": 0.0889, + "step": 12947 + }, + { + "epoch": 0.11239485768352706, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019306691719784782, + "loss": 0.127, + "step": 12948 + }, + { + "epoch": 0.11240353816373122, + "grad_norm": 0.32421875, + "learning_rate": 0.001930657716487448, + "loss": 0.1045, + "step": 12949 + }, + { + "epoch": 0.11241221864393539, + "grad_norm": 1.15625, + "learning_rate": 0.0019306462600880237, + "loss": 0.1133, + "step": 12950 + }, + { + "epoch": 0.11242089912413955, + "grad_norm": 0.86328125, + "learning_rate": 0.0019306348027802165, + "loss": 0.1895, + "step": 12951 + }, + { + "epoch": 0.11242957960434372, + "grad_norm": 0.38671875, + "learning_rate": 0.0019306233445640399, + "loss": 0.0903, + "step": 12952 + }, + { + "epoch": 0.11243826008454788, + "grad_norm": 1.3515625, + "learning_rate": 0.001930611885439506, + "loss": 0.1152, + "step": 12953 + }, + { + "epoch": 0.11244694056475205, + "grad_norm": 0.412109375, + "learning_rate": 0.0019306004254066274, + "loss": 0.1641, + "step": 12954 + }, + { + "epoch": 0.11245562104495621, + "grad_norm": 0.18359375, + "learning_rate": 0.0019305889644654164, + "loss": 0.123, + "step": 12955 + }, + { + "epoch": 0.11246430152516038, + "grad_norm": 0.24609375, + "learning_rate": 0.0019305775026158864, + "loss": 0.1187, + "step": 12956 + }, + { + "epoch": 0.11247298200536454, + "grad_norm": 0.5546875, + "learning_rate": 0.0019305660398580488, + "loss": 0.0972, + "step": 12957 + }, + { + "epoch": 0.1124816624855687, + "grad_norm": 0.10205078125, + "learning_rate": 0.001930554576191917, + "loss": 0.1001, + "step": 12958 + }, + { + "epoch": 0.11249034296577287, + "grad_norm": 0.57421875, + "learning_rate": 0.001930543111617503, + "loss": 0.1309, + "step": 12959 + }, + { + "epoch": 0.11249902344597704, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019305316461348196, + "loss": 0.1191, + "step": 12960 + }, + { + "epoch": 0.1125077039261812, + "grad_norm": 0.60546875, + "learning_rate": 0.001930520179743879, + "loss": 0.1035, + "step": 12961 + }, + { + "epoch": 0.11251638440638537, + "grad_norm": 0.404296875, + "learning_rate": 0.0019305087124446942, + "loss": 0.127, + "step": 12962 + }, + { + "epoch": 0.11252506488658953, + "grad_norm": 0.125, + "learning_rate": 0.0019304972442372777, + "loss": 0.124, + "step": 12963 + }, + { + "epoch": 0.1125337453667937, + "grad_norm": 0.490234375, + "learning_rate": 0.0019304857751216418, + "loss": 0.0986, + "step": 12964 + }, + { + "epoch": 0.11254242584699786, + "grad_norm": 0.53125, + "learning_rate": 0.0019304743050977994, + "loss": 0.0957, + "step": 12965 + }, + { + "epoch": 0.11255110632720203, + "grad_norm": 0.734375, + "learning_rate": 0.0019304628341657623, + "loss": 0.1338, + "step": 12966 + }, + { + "epoch": 0.11255978680740619, + "grad_norm": 0.23046875, + "learning_rate": 0.0019304513623255436, + "loss": 0.127, + "step": 12967 + }, + { + "epoch": 0.11256846728761034, + "grad_norm": 1.265625, + "learning_rate": 0.001930439889577156, + "loss": 0.1182, + "step": 12968 + }, + { + "epoch": 0.11257714776781451, + "grad_norm": 0.68359375, + "learning_rate": 0.0019304284159206114, + "loss": 0.083, + "step": 12969 + }, + { + "epoch": 0.11258582824801867, + "grad_norm": 0.3671875, + "learning_rate": 0.0019304169413559232, + "loss": 0.1006, + "step": 12970 + }, + { + "epoch": 0.11259450872822284, + "grad_norm": 0.2734375, + "learning_rate": 0.0019304054658831034, + "loss": 0.1074, + "step": 12971 + }, + { + "epoch": 0.112603189208427, + "grad_norm": 0.8515625, + "learning_rate": 0.0019303939895021648, + "loss": 0.1172, + "step": 12972 + }, + { + "epoch": 0.11261186968863117, + "grad_norm": 0.109375, + "learning_rate": 0.0019303825122131196, + "loss": 0.1074, + "step": 12973 + }, + { + "epoch": 0.11262055016883533, + "grad_norm": 0.462890625, + "learning_rate": 0.0019303710340159803, + "loss": 0.1133, + "step": 12974 + }, + { + "epoch": 0.1126292306490395, + "grad_norm": 0.283203125, + "learning_rate": 0.0019303595549107601, + "loss": 0.1426, + "step": 12975 + }, + { + "epoch": 0.11263791112924366, + "grad_norm": 0.48828125, + "learning_rate": 0.0019303480748974712, + "loss": 0.1172, + "step": 12976 + }, + { + "epoch": 0.11264659160944783, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019303365939761258, + "loss": 0.1416, + "step": 12977 + }, + { + "epoch": 0.112655272089652, + "grad_norm": 0.94140625, + "learning_rate": 0.001930325112146737, + "loss": 0.1299, + "step": 12978 + }, + { + "epoch": 0.11266395256985616, + "grad_norm": 0.12353515625, + "learning_rate": 0.001930313629409317, + "loss": 0.1138, + "step": 12979 + }, + { + "epoch": 0.11267263305006033, + "grad_norm": 0.37890625, + "learning_rate": 0.0019303021457638787, + "loss": 0.1191, + "step": 12980 + }, + { + "epoch": 0.11268131353026449, + "grad_norm": 0.291015625, + "learning_rate": 0.0019302906612104346, + "loss": 0.1387, + "step": 12981 + }, + { + "epoch": 0.11268999401046866, + "grad_norm": 1.03125, + "learning_rate": 0.0019302791757489965, + "loss": 0.1836, + "step": 12982 + }, + { + "epoch": 0.11269867449067282, + "grad_norm": 0.361328125, + "learning_rate": 0.0019302676893795782, + "loss": 0.1562, + "step": 12983 + }, + { + "epoch": 0.11270735497087699, + "grad_norm": 0.365234375, + "learning_rate": 0.0019302562021021915, + "loss": 0.1387, + "step": 12984 + }, + { + "epoch": 0.11271603545108115, + "grad_norm": 0.171875, + "learning_rate": 0.0019302447139168486, + "loss": 0.1562, + "step": 12985 + }, + { + "epoch": 0.11272471593128532, + "grad_norm": 0.294921875, + "learning_rate": 0.0019302332248235631, + "loss": 0.1201, + "step": 12986 + }, + { + "epoch": 0.11273339641148948, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019302217348223468, + "loss": 0.0977, + "step": 12987 + }, + { + "epoch": 0.11274207689169365, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019302102439132125, + "loss": 0.123, + "step": 12988 + }, + { + "epoch": 0.11275075737189781, + "grad_norm": 0.322265625, + "learning_rate": 0.0019301987520961728, + "loss": 0.1445, + "step": 12989 + }, + { + "epoch": 0.11275943785210198, + "grad_norm": 0.3359375, + "learning_rate": 0.0019301872593712399, + "loss": 0.1328, + "step": 12990 + }, + { + "epoch": 0.11276811833230614, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019301757657384274, + "loss": 0.1089, + "step": 12991 + }, + { + "epoch": 0.1127767988125103, + "grad_norm": 0.30859375, + "learning_rate": 0.0019301642711977466, + "loss": 0.105, + "step": 12992 + }, + { + "epoch": 0.11278547929271447, + "grad_norm": 0.23828125, + "learning_rate": 0.0019301527757492106, + "loss": 0.0752, + "step": 12993 + }, + { + "epoch": 0.11279415977291864, + "grad_norm": 0.34375, + "learning_rate": 0.001930141279392832, + "loss": 0.166, + "step": 12994 + }, + { + "epoch": 0.1128028402531228, + "grad_norm": 0.55078125, + "learning_rate": 0.0019301297821286238, + "loss": 0.1309, + "step": 12995 + }, + { + "epoch": 0.11281152073332697, + "grad_norm": 0.23046875, + "learning_rate": 0.0019301182839565976, + "loss": 0.1084, + "step": 12996 + }, + { + "epoch": 0.11282020121353113, + "grad_norm": 0.212890625, + "learning_rate": 0.001930106784876767, + "loss": 0.1426, + "step": 12997 + }, + { + "epoch": 0.1128288816937353, + "grad_norm": 0.333984375, + "learning_rate": 0.0019300952848891437, + "loss": 0.1729, + "step": 12998 + }, + { + "epoch": 0.11283756217393946, + "grad_norm": 0.435546875, + "learning_rate": 0.0019300837839937408, + "loss": 0.1211, + "step": 12999 + }, + { + "epoch": 0.11284624265414363, + "grad_norm": 0.2421875, + "learning_rate": 0.001930072282190571, + "loss": 0.0894, + "step": 13000 + }, + { + "epoch": 0.11285492313434779, + "grad_norm": 0.404296875, + "learning_rate": 0.0019300607794796463, + "loss": 0.1089, + "step": 13001 + }, + { + "epoch": 0.11286360361455196, + "grad_norm": 0.54296875, + "learning_rate": 0.0019300492758609795, + "loss": 0.1299, + "step": 13002 + }, + { + "epoch": 0.11287228409475612, + "grad_norm": 0.40234375, + "learning_rate": 0.0019300377713345836, + "loss": 0.123, + "step": 13003 + }, + { + "epoch": 0.11288096457496029, + "grad_norm": 0.115234375, + "learning_rate": 0.0019300262659004707, + "loss": 0.1504, + "step": 13004 + }, + { + "epoch": 0.11288964505516445, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019300147595586537, + "loss": 0.166, + "step": 13005 + }, + { + "epoch": 0.11289832553536862, + "grad_norm": 0.8515625, + "learning_rate": 0.0019300032523091452, + "loss": 0.127, + "step": 13006 + }, + { + "epoch": 0.11290700601557278, + "grad_norm": 0.158203125, + "learning_rate": 0.001929991744151957, + "loss": 0.123, + "step": 13007 + }, + { + "epoch": 0.11291568649577695, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019299802350871028, + "loss": 0.1338, + "step": 13008 + }, + { + "epoch": 0.11292436697598111, + "grad_norm": 0.61328125, + "learning_rate": 0.0019299687251145945, + "loss": 0.0801, + "step": 13009 + }, + { + "epoch": 0.11293304745618528, + "grad_norm": 0.1328125, + "learning_rate": 0.001929957214234445, + "loss": 0.1045, + "step": 13010 + }, + { + "epoch": 0.11294172793638944, + "grad_norm": 0.3359375, + "learning_rate": 0.0019299457024466666, + "loss": 0.1621, + "step": 13011 + }, + { + "epoch": 0.11295040841659361, + "grad_norm": 0.291015625, + "learning_rate": 0.001929934189751272, + "loss": 0.1182, + "step": 13012 + }, + { + "epoch": 0.11295908889679777, + "grad_norm": 0.10986328125, + "learning_rate": 0.001929922676148274, + "loss": 0.1221, + "step": 13013 + }, + { + "epoch": 0.11296776937700194, + "grad_norm": 0.84765625, + "learning_rate": 0.001929911161637685, + "loss": 0.1006, + "step": 13014 + }, + { + "epoch": 0.1129764498572061, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019298996462195175, + "loss": 0.0801, + "step": 13015 + }, + { + "epoch": 0.11298513033741027, + "grad_norm": 0.546875, + "learning_rate": 0.0019298881298937846, + "loss": 0.1221, + "step": 13016 + }, + { + "epoch": 0.11299381081761443, + "grad_norm": 0.203125, + "learning_rate": 0.0019298766126604981, + "loss": 0.1191, + "step": 13017 + }, + { + "epoch": 0.1130024912978186, + "grad_norm": 0.29296875, + "learning_rate": 0.0019298650945196712, + "loss": 0.0894, + "step": 13018 + }, + { + "epoch": 0.11301117177802276, + "grad_norm": 0.09326171875, + "learning_rate": 0.001929853575471316, + "loss": 0.0693, + "step": 13019 + }, + { + "epoch": 0.11301985225822693, + "grad_norm": 0.353515625, + "learning_rate": 0.0019298420555154457, + "loss": 0.1143, + "step": 13020 + }, + { + "epoch": 0.1130285327384311, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019298305346520726, + "loss": 0.125, + "step": 13021 + }, + { + "epoch": 0.11303721321863526, + "grad_norm": 0.26171875, + "learning_rate": 0.0019298190128812093, + "loss": 0.0859, + "step": 13022 + }, + { + "epoch": 0.11304589369883943, + "grad_norm": 0.1201171875, + "learning_rate": 0.001929807490202868, + "loss": 0.1006, + "step": 13023 + }, + { + "epoch": 0.11305457417904359, + "grad_norm": 0.578125, + "learning_rate": 0.0019297959666170622, + "loss": 0.1465, + "step": 13024 + }, + { + "epoch": 0.11306325465924776, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019297844421238039, + "loss": 0.0908, + "step": 13025 + }, + { + "epoch": 0.11307193513945192, + "grad_norm": 0.1953125, + "learning_rate": 0.0019297729167231056, + "loss": 0.1318, + "step": 13026 + }, + { + "epoch": 0.11308061561965609, + "grad_norm": 0.5625, + "learning_rate": 0.0019297613904149802, + "loss": 0.1328, + "step": 13027 + }, + { + "epoch": 0.11308929609986025, + "grad_norm": 0.3828125, + "learning_rate": 0.0019297498631994404, + "loss": 0.0884, + "step": 13028 + }, + { + "epoch": 0.11309797658006442, + "grad_norm": 0.376953125, + "learning_rate": 0.0019297383350764983, + "loss": 0.1631, + "step": 13029 + }, + { + "epoch": 0.11310665706026857, + "grad_norm": 0.46875, + "learning_rate": 0.001929726806046167, + "loss": 0.1162, + "step": 13030 + }, + { + "epoch": 0.11311533754047273, + "grad_norm": 0.54296875, + "learning_rate": 0.0019297152761084588, + "loss": 0.1387, + "step": 13031 + }, + { + "epoch": 0.1131240180206769, + "grad_norm": 0.64453125, + "learning_rate": 0.0019297037452633864, + "loss": 0.1069, + "step": 13032 + }, + { + "epoch": 0.11313269850088106, + "grad_norm": 0.69921875, + "learning_rate": 0.0019296922135109626, + "loss": 0.1074, + "step": 13033 + }, + { + "epoch": 0.11314137898108523, + "grad_norm": 1.1953125, + "learning_rate": 0.0019296806808511999, + "loss": 0.1602, + "step": 13034 + }, + { + "epoch": 0.11315005946128939, + "grad_norm": 0.443359375, + "learning_rate": 0.001929669147284111, + "loss": 0.0894, + "step": 13035 + }, + { + "epoch": 0.11315873994149356, + "grad_norm": 0.205078125, + "learning_rate": 0.0019296576128097083, + "loss": 0.1084, + "step": 13036 + }, + { + "epoch": 0.11316742042169772, + "grad_norm": 0.50390625, + "learning_rate": 0.0019296460774280045, + "loss": 0.1094, + "step": 13037 + }, + { + "epoch": 0.11317610090190189, + "grad_norm": 1.5390625, + "learning_rate": 0.001929634541139012, + "loss": 0.1338, + "step": 13038 + }, + { + "epoch": 0.11318478138210605, + "grad_norm": 0.1728515625, + "learning_rate": 0.001929623003942744, + "loss": 0.1562, + "step": 13039 + }, + { + "epoch": 0.11319346186231022, + "grad_norm": 0.447265625, + "learning_rate": 0.0019296114658392124, + "loss": 0.1211, + "step": 13040 + }, + { + "epoch": 0.11320214234251438, + "grad_norm": 0.5, + "learning_rate": 0.0019295999268284301, + "loss": 0.1206, + "step": 13041 + }, + { + "epoch": 0.11321082282271855, + "grad_norm": 0.1044921875, + "learning_rate": 0.00192958838691041, + "loss": 0.0889, + "step": 13042 + }, + { + "epoch": 0.11321950330292271, + "grad_norm": 0.609375, + "learning_rate": 0.0019295768460851645, + "loss": 0.1406, + "step": 13043 + }, + { + "epoch": 0.11322818378312688, + "grad_norm": 0.50390625, + "learning_rate": 0.001929565304352706, + "loss": 0.1504, + "step": 13044 + }, + { + "epoch": 0.11323686426333104, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019295537617130475, + "loss": 0.1074, + "step": 13045 + }, + { + "epoch": 0.11324554474353521, + "grad_norm": 0.173828125, + "learning_rate": 0.0019295422181662017, + "loss": 0.1113, + "step": 13046 + }, + { + "epoch": 0.11325422522373937, + "grad_norm": 0.1591796875, + "learning_rate": 0.001929530673712181, + "loss": 0.1426, + "step": 13047 + }, + { + "epoch": 0.11326290570394354, + "grad_norm": 0.283203125, + "learning_rate": 0.0019295191283509978, + "loss": 0.0986, + "step": 13048 + }, + { + "epoch": 0.1132715861841477, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019295075820826646, + "loss": 0.1318, + "step": 13049 + }, + { + "epoch": 0.11328026666435187, + "grad_norm": 0.5, + "learning_rate": 0.001929496034907195, + "loss": 0.1309, + "step": 13050 + }, + { + "epoch": 0.11328894714455603, + "grad_norm": 0.25, + "learning_rate": 0.0019294844868246006, + "loss": 0.1328, + "step": 13051 + }, + { + "epoch": 0.1132976276247602, + "grad_norm": 0.421875, + "learning_rate": 0.0019294729378348946, + "loss": 0.1006, + "step": 13052 + }, + { + "epoch": 0.11330630810496437, + "grad_norm": 0.0771484375, + "learning_rate": 0.0019294613879380896, + "loss": 0.1152, + "step": 13053 + }, + { + "epoch": 0.11331498858516853, + "grad_norm": 0.34375, + "learning_rate": 0.0019294498371341976, + "loss": 0.1172, + "step": 13054 + }, + { + "epoch": 0.1133236690653727, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019294382854232323, + "loss": 0.1074, + "step": 13055 + }, + { + "epoch": 0.11333234954557686, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019294267328052056, + "loss": 0.106, + "step": 13056 + }, + { + "epoch": 0.11334103002578103, + "grad_norm": 0.2080078125, + "learning_rate": 0.00192941517928013, + "loss": 0.0942, + "step": 13057 + }, + { + "epoch": 0.11334971050598519, + "grad_norm": 0.26171875, + "learning_rate": 0.0019294036248480185, + "loss": 0.1367, + "step": 13058 + }, + { + "epoch": 0.11335839098618936, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019293920695088837, + "loss": 0.1338, + "step": 13059 + }, + { + "epoch": 0.11336707146639352, + "grad_norm": 0.5859375, + "learning_rate": 0.0019293805132627384, + "loss": 0.1001, + "step": 13060 + }, + { + "epoch": 0.11337575194659769, + "grad_norm": 0.59765625, + "learning_rate": 0.0019293689561095948, + "loss": 0.1147, + "step": 13061 + }, + { + "epoch": 0.11338443242680185, + "grad_norm": 0.78125, + "learning_rate": 0.0019293573980494658, + "loss": 0.125, + "step": 13062 + }, + { + "epoch": 0.11339311290700602, + "grad_norm": 0.35546875, + "learning_rate": 0.0019293458390823642, + "loss": 0.2207, + "step": 13063 + }, + { + "epoch": 0.11340179338721018, + "grad_norm": 0.21875, + "learning_rate": 0.0019293342792083023, + "loss": 0.1055, + "step": 13064 + }, + { + "epoch": 0.11341047386741435, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019293227184272927, + "loss": 0.1064, + "step": 13065 + }, + { + "epoch": 0.11341915434761851, + "grad_norm": 0.9296875, + "learning_rate": 0.0019293111567393487, + "loss": 0.1309, + "step": 13066 + }, + { + "epoch": 0.11342783482782268, + "grad_norm": 0.296875, + "learning_rate": 0.0019292995941444823, + "loss": 0.1113, + "step": 13067 + }, + { + "epoch": 0.11343651530802684, + "grad_norm": 0.1640625, + "learning_rate": 0.0019292880306427062, + "loss": 0.0596, + "step": 13068 + }, + { + "epoch": 0.113445195788231, + "grad_norm": 0.1904296875, + "learning_rate": 0.001929276466234033, + "loss": 0.0845, + "step": 13069 + }, + { + "epoch": 0.11345387626843517, + "grad_norm": 0.34765625, + "learning_rate": 0.0019292649009184759, + "loss": 0.1387, + "step": 13070 + }, + { + "epoch": 0.11346255674863934, + "grad_norm": 0.15234375, + "learning_rate": 0.001929253334696047, + "loss": 0.1045, + "step": 13071 + }, + { + "epoch": 0.1134712372288435, + "grad_norm": 0.83984375, + "learning_rate": 0.001929241767566759, + "loss": 0.1348, + "step": 13072 + }, + { + "epoch": 0.11347991770904767, + "grad_norm": 0.1318359375, + "learning_rate": 0.001929230199530625, + "loss": 0.1357, + "step": 13073 + }, + { + "epoch": 0.11348859818925183, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019292186305876569, + "loss": 0.0923, + "step": 13074 + }, + { + "epoch": 0.113497278669456, + "grad_norm": 0.65234375, + "learning_rate": 0.0019292070607378677, + "loss": 0.1328, + "step": 13075 + }, + { + "epoch": 0.11350595914966016, + "grad_norm": 0.169921875, + "learning_rate": 0.0019291954899812705, + "loss": 0.064, + "step": 13076 + }, + { + "epoch": 0.11351463962986433, + "grad_norm": 0.388671875, + "learning_rate": 0.0019291839183178774, + "loss": 0.0938, + "step": 13077 + }, + { + "epoch": 0.11352332011006849, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019291723457477014, + "loss": 0.125, + "step": 13078 + }, + { + "epoch": 0.11353200059027266, + "grad_norm": 0.75390625, + "learning_rate": 0.001929160772270755, + "loss": 0.0977, + "step": 13079 + }, + { + "epoch": 0.11354068107047682, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019291491978870505, + "loss": 0.1055, + "step": 13080 + }, + { + "epoch": 0.11354936155068099, + "grad_norm": 0.185546875, + "learning_rate": 0.001929137622596601, + "loss": 0.127, + "step": 13081 + }, + { + "epoch": 0.11355804203088515, + "grad_norm": 0.30859375, + "learning_rate": 0.0019291260463994194, + "loss": 0.1074, + "step": 13082 + }, + { + "epoch": 0.11356672251108932, + "grad_norm": 0.259765625, + "learning_rate": 0.0019291144692955176, + "loss": 0.1016, + "step": 13083 + }, + { + "epoch": 0.11357540299129348, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019291028912849088, + "loss": 0.0703, + "step": 13084 + }, + { + "epoch": 0.11358408347149765, + "grad_norm": 0.197265625, + "learning_rate": 0.0019290913123676055, + "loss": 0.0767, + "step": 13085 + }, + { + "epoch": 0.11359276395170181, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019290797325436204, + "loss": 0.0908, + "step": 13086 + }, + { + "epoch": 0.11360144443190598, + "grad_norm": 0.6171875, + "learning_rate": 0.0019290681518129664, + "loss": 0.1289, + "step": 13087 + }, + { + "epoch": 0.11361012491211014, + "grad_norm": 0.478515625, + "learning_rate": 0.0019290565701756556, + "loss": 0.1543, + "step": 13088 + }, + { + "epoch": 0.11361880539231431, + "grad_norm": 0.337890625, + "learning_rate": 0.0019290449876317012, + "loss": 0.124, + "step": 13089 + }, + { + "epoch": 0.11362748587251847, + "grad_norm": 0.72265625, + "learning_rate": 0.0019290334041811156, + "loss": 0.1074, + "step": 13090 + }, + { + "epoch": 0.11363616635272264, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019290218198239114, + "loss": 0.1182, + "step": 13091 + }, + { + "epoch": 0.11364484683292679, + "grad_norm": 0.306640625, + "learning_rate": 0.0019290102345601015, + "loss": 0.125, + "step": 13092 + }, + { + "epoch": 0.11365352731313096, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019289986483896984, + "loss": 0.168, + "step": 13093 + }, + { + "epoch": 0.11366220779333512, + "grad_norm": 0.69921875, + "learning_rate": 0.001928987061312715, + "loss": 0.1582, + "step": 13094 + }, + { + "epoch": 0.11367088827353929, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019289754733291637, + "loss": 0.1099, + "step": 13095 + }, + { + "epoch": 0.11367956875374345, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019289638844390573, + "loss": 0.125, + "step": 13096 + }, + { + "epoch": 0.11368824923394762, + "grad_norm": 0.31640625, + "learning_rate": 0.0019289522946424082, + "loss": 0.123, + "step": 13097 + }, + { + "epoch": 0.11369692971415178, + "grad_norm": 0.2412109375, + "learning_rate": 0.00192894070393923, + "loss": 0.0786, + "step": 13098 + }, + { + "epoch": 0.11370561019435595, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019289291123295341, + "loss": 0.1045, + "step": 13099 + }, + { + "epoch": 0.11371429067456011, + "grad_norm": 0.546875, + "learning_rate": 0.001928917519813334, + "loss": 0.1201, + "step": 13100 + }, + { + "epoch": 0.11372297115476428, + "grad_norm": 0.365234375, + "learning_rate": 0.001928905926390642, + "loss": 0.127, + "step": 13101 + }, + { + "epoch": 0.11373165163496844, + "grad_norm": 0.1318359375, + "learning_rate": 0.001928894332061471, + "loss": 0.1279, + "step": 13102 + }, + { + "epoch": 0.11374033211517261, + "grad_norm": 1.03125, + "learning_rate": 0.0019288827368258337, + "loss": 0.124, + "step": 13103 + }, + { + "epoch": 0.11374901259537677, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019288711406837424, + "loss": 0.1021, + "step": 13104 + }, + { + "epoch": 0.11375769307558094, + "grad_norm": 0.111328125, + "learning_rate": 0.0019288595436352106, + "loss": 0.0977, + "step": 13105 + }, + { + "epoch": 0.1137663735557851, + "grad_norm": 0.5390625, + "learning_rate": 0.00192884794568025, + "loss": 0.1543, + "step": 13106 + }, + { + "epoch": 0.11377505403598927, + "grad_norm": 0.1943359375, + "learning_rate": 0.001928836346818874, + "loss": 0.0947, + "step": 13107 + }, + { + "epoch": 0.11378373451619343, + "grad_norm": 0.275390625, + "learning_rate": 0.0019288247470510946, + "loss": 0.0962, + "step": 13108 + }, + { + "epoch": 0.1137924149963976, + "grad_norm": 0.123046875, + "learning_rate": 0.0019288131463769255, + "loss": 0.0977, + "step": 13109 + }, + { + "epoch": 0.11380109547660176, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019288015447963785, + "loss": 0.0996, + "step": 13110 + }, + { + "epoch": 0.11380977595680593, + "grad_norm": 0.51953125, + "learning_rate": 0.0019287899423094665, + "loss": 0.1396, + "step": 13111 + }, + { + "epoch": 0.1138184564370101, + "grad_norm": 0.2421875, + "learning_rate": 0.0019287783389162025, + "loss": 0.0967, + "step": 13112 + }, + { + "epoch": 0.11382713691721426, + "grad_norm": 0.142578125, + "learning_rate": 0.0019287667346165988, + "loss": 0.1357, + "step": 13113 + }, + { + "epoch": 0.11383581739741842, + "grad_norm": 0.58203125, + "learning_rate": 0.0019287551294106683, + "loss": 0.125, + "step": 13114 + }, + { + "epoch": 0.11384449787762259, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019287435232984236, + "loss": 0.1309, + "step": 13115 + }, + { + "epoch": 0.11385317835782675, + "grad_norm": 0.466796875, + "learning_rate": 0.0019287319162798776, + "loss": 0.1465, + "step": 13116 + }, + { + "epoch": 0.11386185883803092, + "grad_norm": 0.263671875, + "learning_rate": 0.0019287203083550425, + "loss": 0.1328, + "step": 13117 + }, + { + "epoch": 0.11387053931823508, + "grad_norm": 1.0625, + "learning_rate": 0.0019287086995239317, + "loss": 0.1377, + "step": 13118 + }, + { + "epoch": 0.11387921979843925, + "grad_norm": 1.1484375, + "learning_rate": 0.0019286970897865571, + "loss": 0.1348, + "step": 13119 + }, + { + "epoch": 0.11388790027864341, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019286854791429322, + "loss": 0.1187, + "step": 13120 + }, + { + "epoch": 0.11389658075884758, + "grad_norm": 0.10302734375, + "learning_rate": 0.0019286738675930693, + "loss": 0.0942, + "step": 13121 + }, + { + "epoch": 0.11390526123905174, + "grad_norm": 0.470703125, + "learning_rate": 0.001928662255136981, + "loss": 0.1191, + "step": 13122 + }, + { + "epoch": 0.11391394171925591, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019286506417746801, + "loss": 0.1123, + "step": 13123 + }, + { + "epoch": 0.11392262219946007, + "grad_norm": 0.138671875, + "learning_rate": 0.0019286390275061792, + "loss": 0.1309, + "step": 13124 + }, + { + "epoch": 0.11393130267966424, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019286274123314912, + "loss": 0.1348, + "step": 13125 + }, + { + "epoch": 0.1139399831598684, + "grad_norm": 0.2265625, + "learning_rate": 0.0019286157962506286, + "loss": 0.1445, + "step": 13126 + }, + { + "epoch": 0.11394866364007257, + "grad_norm": 0.0810546875, + "learning_rate": 0.0019286041792636044, + "loss": 0.1201, + "step": 13127 + }, + { + "epoch": 0.11395734412027674, + "grad_norm": 0.1376953125, + "learning_rate": 0.001928592561370431, + "loss": 0.1162, + "step": 13128 + }, + { + "epoch": 0.1139660246004809, + "grad_norm": 0.345703125, + "learning_rate": 0.0019285809425711215, + "loss": 0.1318, + "step": 13129 + }, + { + "epoch": 0.11397470508068507, + "grad_norm": 0.5625, + "learning_rate": 0.001928569322865688, + "loss": 0.1387, + "step": 13130 + }, + { + "epoch": 0.11398338556088923, + "grad_norm": 0.498046875, + "learning_rate": 0.0019285577022541435, + "loss": 0.1309, + "step": 13131 + }, + { + "epoch": 0.1139920660410934, + "grad_norm": 2.09375, + "learning_rate": 0.001928546080736501, + "loss": 0.1162, + "step": 13132 + }, + { + "epoch": 0.11400074652129756, + "grad_norm": 0.1728515625, + "learning_rate": 0.001928534458312773, + "loss": 0.1099, + "step": 13133 + }, + { + "epoch": 0.11400942700150173, + "grad_norm": 0.10546875, + "learning_rate": 0.0019285228349829717, + "loss": 0.1426, + "step": 13134 + }, + { + "epoch": 0.11401810748170589, + "grad_norm": 0.62890625, + "learning_rate": 0.0019285112107471107, + "loss": 0.0859, + "step": 13135 + }, + { + "epoch": 0.11402678796191006, + "grad_norm": 1.046875, + "learning_rate": 0.0019284995856052023, + "loss": 0.1465, + "step": 13136 + }, + { + "epoch": 0.11403546844211422, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019284879595572592, + "loss": 0.1021, + "step": 13137 + }, + { + "epoch": 0.11404414892231839, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019284763326032938, + "loss": 0.0972, + "step": 13138 + }, + { + "epoch": 0.11405282940252255, + "grad_norm": 0.484375, + "learning_rate": 0.0019284647047433192, + "loss": 0.127, + "step": 13139 + }, + { + "epoch": 0.11406150988272672, + "grad_norm": 1.2890625, + "learning_rate": 0.0019284530759773482, + "loss": 0.125, + "step": 13140 + }, + { + "epoch": 0.11407019036293088, + "grad_norm": 0.7578125, + "learning_rate": 0.0019284414463053934, + "loss": 0.1201, + "step": 13141 + }, + { + "epoch": 0.11407887084313505, + "grad_norm": 0.310546875, + "learning_rate": 0.0019284298157274675, + "loss": 0.1914, + "step": 13142 + }, + { + "epoch": 0.11408755132333921, + "grad_norm": 0.4375, + "learning_rate": 0.0019284181842435832, + "loss": 0.1289, + "step": 13143 + }, + { + "epoch": 0.11409623180354338, + "grad_norm": 0.2138671875, + "learning_rate": 0.001928406551853753, + "loss": 0.1172, + "step": 13144 + }, + { + "epoch": 0.11410491228374754, + "grad_norm": 0.310546875, + "learning_rate": 0.0019283949185579898, + "loss": 0.1084, + "step": 13145 + }, + { + "epoch": 0.11411359276395171, + "grad_norm": 0.251953125, + "learning_rate": 0.0019283832843563066, + "loss": 0.0913, + "step": 13146 + }, + { + "epoch": 0.11412227324415587, + "grad_norm": 0.40625, + "learning_rate": 0.001928371649248716, + "loss": 0.0928, + "step": 13147 + }, + { + "epoch": 0.11413095372436004, + "grad_norm": 0.109375, + "learning_rate": 0.0019283600132352305, + "loss": 0.1201, + "step": 13148 + }, + { + "epoch": 0.1141396342045642, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019283483763158627, + "loss": 0.1162, + "step": 13149 + }, + { + "epoch": 0.11414831468476837, + "grad_norm": 0.34765625, + "learning_rate": 0.0019283367384906255, + "loss": 0.1064, + "step": 13150 + }, + { + "epoch": 0.11415699516497253, + "grad_norm": 0.3828125, + "learning_rate": 0.001928325099759532, + "loss": 0.1436, + "step": 13151 + }, + { + "epoch": 0.1141656756451767, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019283134601225946, + "loss": 0.1348, + "step": 13152 + }, + { + "epoch": 0.11417435612538085, + "grad_norm": 0.99609375, + "learning_rate": 0.0019283018195798258, + "loss": 0.1455, + "step": 13153 + }, + { + "epoch": 0.11418303660558501, + "grad_norm": 0.322265625, + "learning_rate": 0.001928290178131239, + "loss": 0.1143, + "step": 13154 + }, + { + "epoch": 0.11419171708578918, + "grad_norm": 0.67578125, + "learning_rate": 0.0019282785357768461, + "loss": 0.1602, + "step": 13155 + }, + { + "epoch": 0.11420039756599334, + "grad_norm": 0.208984375, + "learning_rate": 0.0019282668925166607, + "loss": 0.1055, + "step": 13156 + }, + { + "epoch": 0.11420907804619751, + "grad_norm": 0.58984375, + "learning_rate": 0.0019282552483506946, + "loss": 0.125, + "step": 13157 + }, + { + "epoch": 0.11421775852640167, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019282436032789611, + "loss": 0.1543, + "step": 13158 + }, + { + "epoch": 0.11422643900660584, + "grad_norm": 1.109375, + "learning_rate": 0.0019282319573014732, + "loss": 0.1426, + "step": 13159 + }, + { + "epoch": 0.11423511948681, + "grad_norm": 0.296875, + "learning_rate": 0.001928220310418243, + "loss": 0.1543, + "step": 13160 + }, + { + "epoch": 0.11424379996701417, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019282086626292833, + "loss": 0.125, + "step": 13161 + }, + { + "epoch": 0.11425248044721834, + "grad_norm": 0.44921875, + "learning_rate": 0.0019281970139346071, + "loss": 0.1094, + "step": 13162 + }, + { + "epoch": 0.1142611609274225, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019281853643342275, + "loss": 0.1289, + "step": 13163 + }, + { + "epoch": 0.11426984140762667, + "grad_norm": 0.12890625, + "learning_rate": 0.0019281737138281566, + "loss": 0.0952, + "step": 13164 + }, + { + "epoch": 0.11427852188783083, + "grad_norm": 1.2890625, + "learning_rate": 0.0019281620624164073, + "loss": 0.083, + "step": 13165 + }, + { + "epoch": 0.114287202368035, + "grad_norm": 0.44140625, + "learning_rate": 0.0019281504100989925, + "loss": 0.0859, + "step": 13166 + }, + { + "epoch": 0.11429588284823916, + "grad_norm": 0.60546875, + "learning_rate": 0.0019281387568759248, + "loss": 0.1582, + "step": 13167 + }, + { + "epoch": 0.11430456332844333, + "grad_norm": 0.400390625, + "learning_rate": 0.001928127102747217, + "loss": 0.0962, + "step": 13168 + }, + { + "epoch": 0.11431324380864749, + "grad_norm": 0.55859375, + "learning_rate": 0.001928115447712882, + "loss": 0.1201, + "step": 13169 + }, + { + "epoch": 0.11432192428885166, + "grad_norm": 0.484375, + "learning_rate": 0.001928103791772932, + "loss": 0.126, + "step": 13170 + }, + { + "epoch": 0.11433060476905582, + "grad_norm": 0.392578125, + "learning_rate": 0.0019280921349273804, + "loss": 0.0972, + "step": 13171 + }, + { + "epoch": 0.11433928524925999, + "grad_norm": 0.390625, + "learning_rate": 0.0019280804771762397, + "loss": 0.123, + "step": 13172 + }, + { + "epoch": 0.11434796572946415, + "grad_norm": 0.453125, + "learning_rate": 0.0019280688185195226, + "loss": 0.105, + "step": 13173 + }, + { + "epoch": 0.11435664620966832, + "grad_norm": 0.291015625, + "learning_rate": 0.0019280571589572418, + "loss": 0.082, + "step": 13174 + }, + { + "epoch": 0.11436532668987248, + "grad_norm": 0.091796875, + "learning_rate": 0.0019280454984894103, + "loss": 0.1108, + "step": 13175 + }, + { + "epoch": 0.11437400717007665, + "grad_norm": 0.306640625, + "learning_rate": 0.0019280338371160403, + "loss": 0.1143, + "step": 13176 + }, + { + "epoch": 0.11438268765028081, + "grad_norm": 0.255859375, + "learning_rate": 0.0019280221748371454, + "loss": 0.1191, + "step": 13177 + }, + { + "epoch": 0.11439136813048498, + "grad_norm": 0.310546875, + "learning_rate": 0.0019280105116527377, + "loss": 0.1182, + "step": 13178 + }, + { + "epoch": 0.11440004861068914, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019279988475628302, + "loss": 0.0986, + "step": 13179 + }, + { + "epoch": 0.11440872909089331, + "grad_norm": 0.8671875, + "learning_rate": 0.0019279871825674355, + "loss": 0.1699, + "step": 13180 + }, + { + "epoch": 0.11441740957109747, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019279755166665665, + "loss": 0.1211, + "step": 13181 + }, + { + "epoch": 0.11442609005130164, + "grad_norm": 0.138671875, + "learning_rate": 0.0019279638498602358, + "loss": 0.0869, + "step": 13182 + }, + { + "epoch": 0.1144347705315058, + "grad_norm": 0.4296875, + "learning_rate": 0.0019279521821484566, + "loss": 0.0977, + "step": 13183 + }, + { + "epoch": 0.11444345101170997, + "grad_norm": 0.34375, + "learning_rate": 0.0019279405135312411, + "loss": 0.125, + "step": 13184 + }, + { + "epoch": 0.11445213149191413, + "grad_norm": 1.0703125, + "learning_rate": 0.0019279288440086026, + "loss": 0.1094, + "step": 13185 + }, + { + "epoch": 0.1144608119721183, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019279171735805533, + "loss": 0.0889, + "step": 13186 + }, + { + "epoch": 0.11446949245232246, + "grad_norm": 0.236328125, + "learning_rate": 0.001927905502247106, + "loss": 0.1123, + "step": 13187 + }, + { + "epoch": 0.11447817293252663, + "grad_norm": 0.5078125, + "learning_rate": 0.0019278938300082739, + "loss": 0.1152, + "step": 13188 + }, + { + "epoch": 0.1144868534127308, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019278821568640694, + "loss": 0.1138, + "step": 13189 + }, + { + "epoch": 0.11449553389293496, + "grad_norm": 0.408203125, + "learning_rate": 0.0019278704828145058, + "loss": 0.1445, + "step": 13190 + }, + { + "epoch": 0.11450421437313912, + "grad_norm": 0.6796875, + "learning_rate": 0.0019278588078595952, + "loss": 0.1279, + "step": 13191 + }, + { + "epoch": 0.11451289485334329, + "grad_norm": 0.271484375, + "learning_rate": 0.0019278471319993508, + "loss": 0.1133, + "step": 13192 + }, + { + "epoch": 0.11452157533354745, + "grad_norm": 0.94140625, + "learning_rate": 0.0019278354552337846, + "loss": 0.1523, + "step": 13193 + }, + { + "epoch": 0.11453025581375162, + "grad_norm": 0.2255859375, + "learning_rate": 0.001927823777562911, + "loss": 0.0952, + "step": 13194 + }, + { + "epoch": 0.11453893629395578, + "grad_norm": 0.10498046875, + "learning_rate": 0.001927812098986741, + "loss": 0.1221, + "step": 13195 + }, + { + "epoch": 0.11454761677415995, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019278004195052883, + "loss": 0.1123, + "step": 13196 + }, + { + "epoch": 0.11455629725436411, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019277887391185658, + "loss": 0.0645, + "step": 13197 + }, + { + "epoch": 0.11456497773456828, + "grad_norm": 0.103515625, + "learning_rate": 0.0019277770578265856, + "loss": 0.1152, + "step": 13198 + }, + { + "epoch": 0.11457365821477244, + "grad_norm": 0.138671875, + "learning_rate": 0.0019277653756293609, + "loss": 0.127, + "step": 13199 + }, + { + "epoch": 0.11458233869497661, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019277536925269047, + "loss": 0.1011, + "step": 13200 + }, + { + "epoch": 0.11459101917518077, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019277420085192292, + "loss": 0.104, + "step": 13201 + }, + { + "epoch": 0.11459969965538494, + "grad_norm": 0.419921875, + "learning_rate": 0.0019277303236063476, + "loss": 0.127, + "step": 13202 + }, + { + "epoch": 0.1146083801355891, + "grad_norm": 0.94921875, + "learning_rate": 0.0019277186377882722, + "loss": 0.1152, + "step": 13203 + }, + { + "epoch": 0.11461706061579327, + "grad_norm": 0.486328125, + "learning_rate": 0.0019277069510650166, + "loss": 0.0977, + "step": 13204 + }, + { + "epoch": 0.11462574109599744, + "grad_norm": 1.015625, + "learning_rate": 0.001927695263436593, + "loss": 0.1318, + "step": 13205 + }, + { + "epoch": 0.1146344215762016, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019276835749030144, + "loss": 0.1104, + "step": 13206 + }, + { + "epoch": 0.11464310205640577, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019276718854642934, + "loss": 0.1143, + "step": 13207 + }, + { + "epoch": 0.11465178253660993, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019276601951204427, + "loss": 0.1416, + "step": 13208 + }, + { + "epoch": 0.1146604630168141, + "grad_norm": 0.267578125, + "learning_rate": 0.0019276485038714755, + "loss": 0.084, + "step": 13209 + }, + { + "epoch": 0.11466914349701826, + "grad_norm": 0.546875, + "learning_rate": 0.0019276368117174042, + "loss": 0.1201, + "step": 13210 + }, + { + "epoch": 0.11467782397722243, + "grad_norm": 0.41015625, + "learning_rate": 0.0019276251186582417, + "loss": 0.127, + "step": 13211 + }, + { + "epoch": 0.11468650445742659, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019276134246940008, + "loss": 0.1553, + "step": 13212 + }, + { + "epoch": 0.11469518493763076, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019276017298246939, + "loss": 0.1055, + "step": 13213 + }, + { + "epoch": 0.11470386541783492, + "grad_norm": 0.59765625, + "learning_rate": 0.0019275900340503347, + "loss": 0.1064, + "step": 13214 + }, + { + "epoch": 0.11471254589803907, + "grad_norm": 0.166015625, + "learning_rate": 0.0019275783373709358, + "loss": 0.1484, + "step": 13215 + }, + { + "epoch": 0.11472122637824324, + "grad_norm": 0.26953125, + "learning_rate": 0.0019275666397865088, + "loss": 0.1211, + "step": 13216 + }, + { + "epoch": 0.1147299068584474, + "grad_norm": 0.7109375, + "learning_rate": 0.0019275549412970676, + "loss": 0.1201, + "step": 13217 + }, + { + "epoch": 0.11473858733865157, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019275432419026254, + "loss": 0.1191, + "step": 13218 + }, + { + "epoch": 0.11474726781885573, + "grad_norm": 0.169921875, + "learning_rate": 0.0019275315416031936, + "loss": 0.1079, + "step": 13219 + }, + { + "epoch": 0.1147559482990599, + "grad_norm": 0.29296875, + "learning_rate": 0.001927519840398786, + "loss": 0.1123, + "step": 13220 + }, + { + "epoch": 0.11476462877926406, + "grad_norm": 0.60546875, + "learning_rate": 0.0019275081382894154, + "loss": 0.1357, + "step": 13221 + }, + { + "epoch": 0.11477330925946823, + "grad_norm": 0.74609375, + "learning_rate": 0.0019274964352750939, + "loss": 0.168, + "step": 13222 + }, + { + "epoch": 0.1147819897396724, + "grad_norm": 0.48828125, + "learning_rate": 0.0019274847313558348, + "loss": 0.1064, + "step": 13223 + }, + { + "epoch": 0.11479067021987656, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019274730265316509, + "loss": 0.1357, + "step": 13224 + }, + { + "epoch": 0.11479935070008072, + "grad_norm": 0.62109375, + "learning_rate": 0.0019274613208025552, + "loss": 0.1396, + "step": 13225 + }, + { + "epoch": 0.11480803118028489, + "grad_norm": 0.390625, + "learning_rate": 0.0019274496141685601, + "loss": 0.1289, + "step": 13226 + }, + { + "epoch": 0.11481671166048905, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019274379066296784, + "loss": 0.082, + "step": 13227 + }, + { + "epoch": 0.11482539214069322, + "grad_norm": 0.099609375, + "learning_rate": 0.001927426198185923, + "loss": 0.1289, + "step": 13228 + }, + { + "epoch": 0.11483407262089738, + "grad_norm": 0.2216796875, + "learning_rate": 0.001927414488837307, + "loss": 0.123, + "step": 13229 + }, + { + "epoch": 0.11484275310110155, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019274027785838425, + "loss": 0.0952, + "step": 13230 + }, + { + "epoch": 0.11485143358130571, + "grad_norm": 0.076171875, + "learning_rate": 0.001927391067425543, + "loss": 0.0952, + "step": 13231 + }, + { + "epoch": 0.11486011406150988, + "grad_norm": 0.36328125, + "learning_rate": 0.0019273793553624212, + "loss": 0.1055, + "step": 13232 + }, + { + "epoch": 0.11486879454171404, + "grad_norm": 0.34375, + "learning_rate": 0.00192736764239449, + "loss": 0.1162, + "step": 13233 + }, + { + "epoch": 0.11487747502191821, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019273559285217612, + "loss": 0.0996, + "step": 13234 + }, + { + "epoch": 0.11488615550212238, + "grad_norm": 0.30078125, + "learning_rate": 0.001927344213744249, + "loss": 0.0967, + "step": 13235 + }, + { + "epoch": 0.11489483598232654, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019273324980619653, + "loss": 0.1445, + "step": 13236 + }, + { + "epoch": 0.1149035164625307, + "grad_norm": 0.4140625, + "learning_rate": 0.0019273207814749233, + "loss": 0.0972, + "step": 13237 + }, + { + "epoch": 0.11491219694273487, + "grad_norm": 0.60546875, + "learning_rate": 0.001927309063983136, + "loss": 0.1523, + "step": 13238 + }, + { + "epoch": 0.11492087742293904, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019272973455866154, + "loss": 0.1104, + "step": 13239 + }, + { + "epoch": 0.1149295579031432, + "grad_norm": 0.52734375, + "learning_rate": 0.0019272856262853753, + "loss": 0.1279, + "step": 13240 + }, + { + "epoch": 0.11493823838334737, + "grad_norm": 0.232421875, + "learning_rate": 0.001927273906079428, + "loss": 0.1055, + "step": 13241 + }, + { + "epoch": 0.11494691886355153, + "grad_norm": 0.482421875, + "learning_rate": 0.001927262184968786, + "loss": 0.1484, + "step": 13242 + }, + { + "epoch": 0.1149555993437557, + "grad_norm": 0.310546875, + "learning_rate": 0.001927250462953463, + "loss": 0.0996, + "step": 13243 + }, + { + "epoch": 0.11496427982395986, + "grad_norm": 0.2265625, + "learning_rate": 0.001927238740033471, + "loss": 0.1387, + "step": 13244 + }, + { + "epoch": 0.11497296030416403, + "grad_norm": 0.4765625, + "learning_rate": 0.0019272270162088235, + "loss": 0.1172, + "step": 13245 + }, + { + "epoch": 0.11498164078436819, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019272152914795325, + "loss": 0.0859, + "step": 13246 + }, + { + "epoch": 0.11499032126457236, + "grad_norm": 0.25, + "learning_rate": 0.0019272035658456114, + "loss": 0.1113, + "step": 13247 + }, + { + "epoch": 0.11499900174477652, + "grad_norm": 0.287109375, + "learning_rate": 0.0019271918393070733, + "loss": 0.1338, + "step": 13248 + }, + { + "epoch": 0.11500768222498069, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019271801118639304, + "loss": 0.1152, + "step": 13249 + }, + { + "epoch": 0.11501636270518485, + "grad_norm": 0.27734375, + "learning_rate": 0.0019271683835161957, + "loss": 0.1211, + "step": 13250 + }, + { + "epoch": 0.11502504318538902, + "grad_norm": 0.2236328125, + "learning_rate": 0.001927156654263882, + "loss": 0.1504, + "step": 13251 + }, + { + "epoch": 0.11503372366559318, + "grad_norm": 0.63671875, + "learning_rate": 0.0019271449241070023, + "loss": 0.0928, + "step": 13252 + }, + { + "epoch": 0.11504240414579735, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019271331930455693, + "loss": 0.1123, + "step": 13253 + }, + { + "epoch": 0.11505108462600151, + "grad_norm": 0.404296875, + "learning_rate": 0.001927121461079596, + "loss": 0.1445, + "step": 13254 + }, + { + "epoch": 0.11505976510620568, + "grad_norm": 0.6015625, + "learning_rate": 0.0019271097282090952, + "loss": 0.1069, + "step": 13255 + }, + { + "epoch": 0.11506844558640984, + "grad_norm": 0.208984375, + "learning_rate": 0.0019270979944340791, + "loss": 0.125, + "step": 13256 + }, + { + "epoch": 0.11507712606661401, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019270862597545616, + "loss": 0.1016, + "step": 13257 + }, + { + "epoch": 0.11508580654681817, + "grad_norm": 0.7421875, + "learning_rate": 0.001927074524170555, + "loss": 0.1924, + "step": 13258 + }, + { + "epoch": 0.11509448702702234, + "grad_norm": 0.166015625, + "learning_rate": 0.001927062787682072, + "loss": 0.1133, + "step": 13259 + }, + { + "epoch": 0.1151031675072265, + "grad_norm": 0.18359375, + "learning_rate": 0.001927051050289125, + "loss": 0.0957, + "step": 13260 + }, + { + "epoch": 0.11511184798743067, + "grad_norm": 1.8125, + "learning_rate": 0.0019270393119917282, + "loss": 0.1445, + "step": 13261 + }, + { + "epoch": 0.11512052846763483, + "grad_norm": 0.263671875, + "learning_rate": 0.0019270275727898932, + "loss": 0.0864, + "step": 13262 + }, + { + "epoch": 0.115129208947839, + "grad_norm": 0.72265625, + "learning_rate": 0.0019270158326836334, + "loss": 0.1523, + "step": 13263 + }, + { + "epoch": 0.11513788942804316, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019270040916729615, + "loss": 0.0801, + "step": 13264 + }, + { + "epoch": 0.11514656990824733, + "grad_norm": 0.267578125, + "learning_rate": 0.0019269923497578906, + "loss": 0.1191, + "step": 13265 + }, + { + "epoch": 0.1151552503884515, + "grad_norm": 2.15625, + "learning_rate": 0.001926980606938433, + "loss": 0.1895, + "step": 13266 + }, + { + "epoch": 0.11516393086865566, + "grad_norm": 0.41015625, + "learning_rate": 0.0019269688632146016, + "loss": 0.1133, + "step": 13267 + }, + { + "epoch": 0.11517261134885982, + "grad_norm": 0.2431640625, + "learning_rate": 0.00192695711858641, + "loss": 0.1426, + "step": 13268 + }, + { + "epoch": 0.11518129182906399, + "grad_norm": 0.38671875, + "learning_rate": 0.00192694537305387, + "loss": 0.1562, + "step": 13269 + }, + { + "epoch": 0.11518997230926815, + "grad_norm": 0.193359375, + "learning_rate": 0.0019269336266169953, + "loss": 0.1299, + "step": 13270 + }, + { + "epoch": 0.11519865278947232, + "grad_norm": 0.5234375, + "learning_rate": 0.0019269218792757983, + "loss": 0.0957, + "step": 13271 + }, + { + "epoch": 0.11520733326967648, + "grad_norm": 0.12158203125, + "learning_rate": 0.001926910131030292, + "loss": 0.0938, + "step": 13272 + }, + { + "epoch": 0.11521601374988065, + "grad_norm": 0.765625, + "learning_rate": 0.0019268983818804893, + "loss": 0.1328, + "step": 13273 + }, + { + "epoch": 0.11522469423008481, + "grad_norm": 0.5703125, + "learning_rate": 0.0019268866318264029, + "loss": 0.0918, + "step": 13274 + }, + { + "epoch": 0.11523337471028898, + "grad_norm": 0.369140625, + "learning_rate": 0.0019268748808680455, + "loss": 0.1318, + "step": 13275 + }, + { + "epoch": 0.11524205519049313, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019268631290054303, + "loss": 0.1279, + "step": 13276 + }, + { + "epoch": 0.1152507356706973, + "grad_norm": 1.140625, + "learning_rate": 0.0019268513762385703, + "loss": 0.1064, + "step": 13277 + }, + { + "epoch": 0.11525941615090146, + "grad_norm": 0.232421875, + "learning_rate": 0.0019268396225674777, + "loss": 0.1367, + "step": 13278 + }, + { + "epoch": 0.11526809663110563, + "grad_norm": 0.921875, + "learning_rate": 0.0019268278679921657, + "loss": 0.1064, + "step": 13279 + }, + { + "epoch": 0.11527677711130979, + "grad_norm": 0.84765625, + "learning_rate": 0.0019268161125126472, + "loss": 0.2969, + "step": 13280 + }, + { + "epoch": 0.11528545759151396, + "grad_norm": 0.16796875, + "learning_rate": 0.0019268043561289352, + "loss": 0.0938, + "step": 13281 + }, + { + "epoch": 0.11529413807171812, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019267925988410424, + "loss": 0.1211, + "step": 13282 + }, + { + "epoch": 0.11530281855192229, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019267808406489813, + "loss": 0.0894, + "step": 13283 + }, + { + "epoch": 0.11531149903212645, + "grad_norm": 0.2109375, + "learning_rate": 0.0019267690815527652, + "loss": 0.0996, + "step": 13284 + }, + { + "epoch": 0.11532017951233062, + "grad_norm": 0.5546875, + "learning_rate": 0.001926757321552407, + "loss": 0.1104, + "step": 13285 + }, + { + "epoch": 0.11532885999253478, + "grad_norm": 0.166015625, + "learning_rate": 0.0019267455606479195, + "loss": 0.1084, + "step": 13286 + }, + { + "epoch": 0.11533754047273895, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019267337988393154, + "loss": 0.1172, + "step": 13287 + }, + { + "epoch": 0.11534622095294311, + "grad_norm": 0.671875, + "learning_rate": 0.0019267220361266076, + "loss": 0.1396, + "step": 13288 + }, + { + "epoch": 0.11535490143314728, + "grad_norm": 0.10400390625, + "learning_rate": 0.001926710272509809, + "loss": 0.084, + "step": 13289 + }, + { + "epoch": 0.11536358191335144, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019266985079889323, + "loss": 0.1504, + "step": 13290 + }, + { + "epoch": 0.11537226239355561, + "grad_norm": 0.1982421875, + "learning_rate": 0.001926686742563991, + "loss": 0.1045, + "step": 13291 + }, + { + "epoch": 0.11538094287375977, + "grad_norm": 0.169921875, + "learning_rate": 0.0019266749762349969, + "loss": 0.1069, + "step": 13292 + }, + { + "epoch": 0.11538962335396394, + "grad_norm": 0.19921875, + "learning_rate": 0.0019266632090019639, + "loss": 0.1045, + "step": 13293 + }, + { + "epoch": 0.1153983038341681, + "grad_norm": 0.39453125, + "learning_rate": 0.001926651440864904, + "loss": 0.1182, + "step": 13294 + }, + { + "epoch": 0.11540698431437227, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019266396718238311, + "loss": 0.1069, + "step": 13295 + }, + { + "epoch": 0.11541566479457643, + "grad_norm": 0.291015625, + "learning_rate": 0.0019266279018787572, + "loss": 0.104, + "step": 13296 + }, + { + "epoch": 0.1154243452747806, + "grad_norm": 0.18359375, + "learning_rate": 0.0019266161310296953, + "loss": 0.125, + "step": 13297 + }, + { + "epoch": 0.11543302575498476, + "grad_norm": 0.490234375, + "learning_rate": 0.0019266043592766585, + "loss": 0.1758, + "step": 13298 + }, + { + "epoch": 0.11544170623518893, + "grad_norm": 0.28515625, + "learning_rate": 0.0019265925866196597, + "loss": 0.1143, + "step": 13299 + }, + { + "epoch": 0.1154503867153931, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019265808130587115, + "loss": 0.1309, + "step": 13300 + }, + { + "epoch": 0.11545906719559726, + "grad_norm": 0.171875, + "learning_rate": 0.001926569038593827, + "loss": 0.1328, + "step": 13301 + }, + { + "epoch": 0.11546774767580142, + "grad_norm": 0.19140625, + "learning_rate": 0.001926557263225019, + "loss": 0.0776, + "step": 13302 + }, + { + "epoch": 0.11547642815600559, + "grad_norm": 0.439453125, + "learning_rate": 0.0019265454869523005, + "loss": 0.0967, + "step": 13303 + }, + { + "epoch": 0.11548510863620975, + "grad_norm": 0.271484375, + "learning_rate": 0.001926533709775684, + "loss": 0.0947, + "step": 13304 + }, + { + "epoch": 0.11549378911641392, + "grad_norm": 0.07080078125, + "learning_rate": 0.0019265219316951833, + "loss": 0.082, + "step": 13305 + }, + { + "epoch": 0.11550246959661808, + "grad_norm": 0.46875, + "learning_rate": 0.0019265101527108101, + "loss": 0.0859, + "step": 13306 + }, + { + "epoch": 0.11551115007682225, + "grad_norm": 0.2255859375, + "learning_rate": 0.001926498372822578, + "loss": 0.0864, + "step": 13307 + }, + { + "epoch": 0.11551983055702642, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019264865920304995, + "loss": 0.1133, + "step": 13308 + }, + { + "epoch": 0.11552851103723058, + "grad_norm": 0.13671875, + "learning_rate": 0.0019264748103345877, + "loss": 0.0972, + "step": 13309 + }, + { + "epoch": 0.11553719151743475, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019264630277348556, + "loss": 0.1777, + "step": 13310 + }, + { + "epoch": 0.11554587199763891, + "grad_norm": 0.259765625, + "learning_rate": 0.0019264512442313162, + "loss": 0.1113, + "step": 13311 + }, + { + "epoch": 0.11555455247784308, + "grad_norm": 0.18359375, + "learning_rate": 0.0019264394598239817, + "loss": 0.0918, + "step": 13312 + }, + { + "epoch": 0.11556323295804724, + "grad_norm": 0.29296875, + "learning_rate": 0.0019264276745128658, + "loss": 0.1128, + "step": 13313 + }, + { + "epoch": 0.1155719134382514, + "grad_norm": 0.86328125, + "learning_rate": 0.0019264158882979804, + "loss": 0.1094, + "step": 13314 + }, + { + "epoch": 0.11558059391845557, + "grad_norm": 0.32421875, + "learning_rate": 0.0019264041011793397, + "loss": 0.1201, + "step": 13315 + }, + { + "epoch": 0.11558927439865974, + "grad_norm": 0.10546875, + "learning_rate": 0.0019263923131569553, + "loss": 0.1128, + "step": 13316 + }, + { + "epoch": 0.1155979548788639, + "grad_norm": 0.203125, + "learning_rate": 0.001926380524230841, + "loss": 0.0864, + "step": 13317 + }, + { + "epoch": 0.11560663535906807, + "grad_norm": 0.69921875, + "learning_rate": 0.0019263687344010095, + "loss": 0.1289, + "step": 13318 + }, + { + "epoch": 0.11561531583927223, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019263569436674733, + "loss": 0.1338, + "step": 13319 + }, + { + "epoch": 0.1156239963194764, + "grad_norm": 0.443359375, + "learning_rate": 0.0019263451520302457, + "loss": 0.1182, + "step": 13320 + }, + { + "epoch": 0.11563267679968056, + "grad_norm": 0.1669921875, + "learning_rate": 0.0019263333594893392, + "loss": 0.1377, + "step": 13321 + }, + { + "epoch": 0.11564135727988473, + "grad_norm": 0.466796875, + "learning_rate": 0.0019263215660447674, + "loss": 0.1309, + "step": 13322 + }, + { + "epoch": 0.11565003776008889, + "grad_norm": 0.4921875, + "learning_rate": 0.0019263097716965426, + "loss": 0.0801, + "step": 13323 + }, + { + "epoch": 0.11565871824029306, + "grad_norm": 0.66796875, + "learning_rate": 0.001926297976444678, + "loss": 0.1162, + "step": 13324 + }, + { + "epoch": 0.11566739872049722, + "grad_norm": 0.62109375, + "learning_rate": 0.001926286180289186, + "loss": 0.1216, + "step": 13325 + }, + { + "epoch": 0.11567607920070139, + "grad_norm": 0.19921875, + "learning_rate": 0.00192627438323008, + "loss": 0.0928, + "step": 13326 + }, + { + "epoch": 0.11568475968090555, + "grad_norm": 0.94921875, + "learning_rate": 0.0019262625852673727, + "loss": 0.1895, + "step": 13327 + }, + { + "epoch": 0.11569344016110972, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019262507864010773, + "loss": 0.1328, + "step": 13328 + }, + { + "epoch": 0.11570212064131388, + "grad_norm": 0.142578125, + "learning_rate": 0.001926238986631206, + "loss": 0.126, + "step": 13329 + }, + { + "epoch": 0.11571080112151805, + "grad_norm": 0.361328125, + "learning_rate": 0.0019262271859577722, + "loss": 0.1055, + "step": 13330 + }, + { + "epoch": 0.11571948160172221, + "grad_norm": 0.267578125, + "learning_rate": 0.001926215384380789, + "loss": 0.1318, + "step": 13331 + }, + { + "epoch": 0.11572816208192638, + "grad_norm": 0.73828125, + "learning_rate": 0.0019262035819002693, + "loss": 0.1084, + "step": 13332 + }, + { + "epoch": 0.11573684256213054, + "grad_norm": 0.294921875, + "learning_rate": 0.0019261917785162253, + "loss": 0.1377, + "step": 13333 + }, + { + "epoch": 0.11574552304233471, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019261799742286705, + "loss": 0.0757, + "step": 13334 + }, + { + "epoch": 0.11575420352253887, + "grad_norm": 0.33984375, + "learning_rate": 0.001926168169037618, + "loss": 0.1182, + "step": 13335 + }, + { + "epoch": 0.11576288400274304, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019261563629430802, + "loss": 0.1074, + "step": 13336 + }, + { + "epoch": 0.1157715644829472, + "grad_norm": 0.31640625, + "learning_rate": 0.00192614455594507, + "loss": 0.1118, + "step": 13337 + }, + { + "epoch": 0.11578024496315135, + "grad_norm": 0.38671875, + "learning_rate": 0.001926132748043601, + "loss": 0.1279, + "step": 13338 + }, + { + "epoch": 0.11578892544335552, + "grad_norm": 0.53515625, + "learning_rate": 0.0019261209392386854, + "loss": 0.0908, + "step": 13339 + }, + { + "epoch": 0.11579760592355968, + "grad_norm": 0.408203125, + "learning_rate": 0.0019261091295303362, + "loss": 0.1367, + "step": 13340 + }, + { + "epoch": 0.11580628640376385, + "grad_norm": 0.365234375, + "learning_rate": 0.0019260973189185663, + "loss": 0.1055, + "step": 13341 + }, + { + "epoch": 0.11581496688396802, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019260855074033892, + "loss": 0.0967, + "step": 13342 + }, + { + "epoch": 0.11582364736417218, + "grad_norm": 0.1328125, + "learning_rate": 0.0019260736949848175, + "loss": 0.1162, + "step": 13343 + }, + { + "epoch": 0.11583232784437635, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019260618816628638, + "loss": 0.0928, + "step": 13344 + }, + { + "epoch": 0.11584100832458051, + "grad_norm": 0.130859375, + "learning_rate": 0.0019260500674375411, + "loss": 0.126, + "step": 13345 + }, + { + "epoch": 0.11584968880478468, + "grad_norm": 0.45703125, + "learning_rate": 0.0019260382523088625, + "loss": 0.0923, + "step": 13346 + }, + { + "epoch": 0.11585836928498884, + "grad_norm": 0.58203125, + "learning_rate": 0.001926026436276841, + "loss": 0.1157, + "step": 13347 + }, + { + "epoch": 0.115867049765193, + "grad_norm": 0.275390625, + "learning_rate": 0.0019260146193414893, + "loss": 0.125, + "step": 13348 + }, + { + "epoch": 0.11587573024539717, + "grad_norm": 0.55078125, + "learning_rate": 0.0019260028015028205, + "loss": 0.1504, + "step": 13349 + }, + { + "epoch": 0.11588441072560134, + "grad_norm": 0.40625, + "learning_rate": 0.0019259909827608476, + "loss": 0.1299, + "step": 13350 + }, + { + "epoch": 0.1158930912058055, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019259791631155829, + "loss": 0.0972, + "step": 13351 + }, + { + "epoch": 0.11590177168600967, + "grad_norm": 0.40625, + "learning_rate": 0.00192596734256704, + "loss": 0.0957, + "step": 13352 + }, + { + "epoch": 0.11591045216621383, + "grad_norm": 0.150390625, + "learning_rate": 0.001925955521115232, + "loss": 0.106, + "step": 13353 + }, + { + "epoch": 0.115919132646418, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019259436987601713, + "loss": 0.106, + "step": 13354 + }, + { + "epoch": 0.11592781312662216, + "grad_norm": 0.4453125, + "learning_rate": 0.0019259318755018707, + "loss": 0.1357, + "step": 13355 + }, + { + "epoch": 0.11593649360682633, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019259200513403435, + "loss": 0.1191, + "step": 13356 + }, + { + "epoch": 0.11594517408703049, + "grad_norm": 0.38671875, + "learning_rate": 0.0019259082262756027, + "loss": 0.1123, + "step": 13357 + }, + { + "epoch": 0.11595385456723466, + "grad_norm": 0.181640625, + "learning_rate": 0.0019258964003076608, + "loss": 0.1074, + "step": 13358 + }, + { + "epoch": 0.11596253504743882, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019258845734365314, + "loss": 0.1367, + "step": 13359 + }, + { + "epoch": 0.11597121552764299, + "grad_norm": 0.10546875, + "learning_rate": 0.0019258727456622268, + "loss": 0.0947, + "step": 13360 + }, + { + "epoch": 0.11597989600784715, + "grad_norm": 0.37109375, + "learning_rate": 0.0019258609169847602, + "loss": 0.1216, + "step": 13361 + }, + { + "epoch": 0.11598857648805132, + "grad_norm": 0.244140625, + "learning_rate": 0.0019258490874041446, + "loss": 0.1157, + "step": 13362 + }, + { + "epoch": 0.11599725696825548, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019258372569203929, + "loss": 0.1338, + "step": 13363 + }, + { + "epoch": 0.11600593744845965, + "grad_norm": 0.166015625, + "learning_rate": 0.0019258254255335176, + "loss": 0.1201, + "step": 13364 + }, + { + "epoch": 0.11601461792866381, + "grad_norm": 0.765625, + "learning_rate": 0.0019258135932435324, + "loss": 0.1553, + "step": 13365 + }, + { + "epoch": 0.11602329840886798, + "grad_norm": 0.609375, + "learning_rate": 0.0019258017600504499, + "loss": 0.1045, + "step": 13366 + }, + { + "epoch": 0.11603197888907214, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019257899259542829, + "loss": 0.0996, + "step": 13367 + }, + { + "epoch": 0.11604065936927631, + "grad_norm": 0.14453125, + "learning_rate": 0.0019257780909550442, + "loss": 0.1592, + "step": 13368 + }, + { + "epoch": 0.11604933984948047, + "grad_norm": 0.388671875, + "learning_rate": 0.0019257662550527471, + "loss": 0.1084, + "step": 13369 + }, + { + "epoch": 0.11605802032968464, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019257544182474042, + "loss": 0.0898, + "step": 13370 + }, + { + "epoch": 0.1160667008098888, + "grad_norm": 0.138671875, + "learning_rate": 0.0019257425805390292, + "loss": 0.1094, + "step": 13371 + }, + { + "epoch": 0.11607538129009297, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019257307419276342, + "loss": 0.1357, + "step": 13372 + }, + { + "epoch": 0.11608406177029713, + "grad_norm": 0.421875, + "learning_rate": 0.0019257189024132328, + "loss": 0.1172, + "step": 13373 + }, + { + "epoch": 0.1160927422505013, + "grad_norm": 0.376953125, + "learning_rate": 0.0019257070619958373, + "loss": 0.1055, + "step": 13374 + }, + { + "epoch": 0.11610142273070546, + "grad_norm": 0.26171875, + "learning_rate": 0.001925695220675461, + "loss": 0.0967, + "step": 13375 + }, + { + "epoch": 0.11611010321090963, + "grad_norm": 0.19140625, + "learning_rate": 0.0019256833784521167, + "loss": 0.1592, + "step": 13376 + }, + { + "epoch": 0.1161187836911138, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019256715353258176, + "loss": 0.0947, + "step": 13377 + }, + { + "epoch": 0.11612746417131796, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019256596912965768, + "loss": 0.0952, + "step": 13378 + }, + { + "epoch": 0.11613614465152212, + "grad_norm": 0.357421875, + "learning_rate": 0.0019256478463644065, + "loss": 0.105, + "step": 13379 + }, + { + "epoch": 0.11614482513172629, + "grad_norm": 0.8046875, + "learning_rate": 0.0019256360005293204, + "loss": 0.1289, + "step": 13380 + }, + { + "epoch": 0.11615350561193045, + "grad_norm": 0.5078125, + "learning_rate": 0.001925624153791331, + "loss": 0.1338, + "step": 13381 + }, + { + "epoch": 0.11616218609213462, + "grad_norm": 0.259765625, + "learning_rate": 0.0019256123061504515, + "loss": 0.0923, + "step": 13382 + }, + { + "epoch": 0.11617086657233879, + "grad_norm": 0.81640625, + "learning_rate": 0.0019256004576066947, + "loss": 0.1025, + "step": 13383 + }, + { + "epoch": 0.11617954705254295, + "grad_norm": 0.12890625, + "learning_rate": 0.0019255886081600737, + "loss": 0.1797, + "step": 13384 + }, + { + "epoch": 0.11618822753274712, + "grad_norm": 0.103515625, + "learning_rate": 0.0019255767578106016, + "loss": 0.0947, + "step": 13385 + }, + { + "epoch": 0.11619690801295128, + "grad_norm": 0.255859375, + "learning_rate": 0.001925564906558291, + "loss": 0.0869, + "step": 13386 + }, + { + "epoch": 0.11620558849315545, + "grad_norm": 0.328125, + "learning_rate": 0.0019255530544031551, + "loss": 0.1182, + "step": 13387 + }, + { + "epoch": 0.11621426897335961, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019255412013452068, + "loss": 0.1396, + "step": 13388 + }, + { + "epoch": 0.11622294945356378, + "grad_norm": 0.11328125, + "learning_rate": 0.0019255293473844591, + "loss": 0.0688, + "step": 13389 + }, + { + "epoch": 0.11623162993376794, + "grad_norm": 0.392578125, + "learning_rate": 0.0019255174925209245, + "loss": 0.0801, + "step": 13390 + }, + { + "epoch": 0.1162403104139721, + "grad_norm": 0.390625, + "learning_rate": 0.0019255056367546166, + "loss": 0.1357, + "step": 13391 + }, + { + "epoch": 0.11624899089417627, + "grad_norm": 0.83203125, + "learning_rate": 0.0019254937800855484, + "loss": 0.0957, + "step": 13392 + }, + { + "epoch": 0.11625767137438044, + "grad_norm": 0.13671875, + "learning_rate": 0.0019254819225137322, + "loss": 0.1035, + "step": 13393 + }, + { + "epoch": 0.1162663518545846, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019254700640391818, + "loss": 0.1484, + "step": 13394 + }, + { + "epoch": 0.11627503233478877, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019254582046619093, + "loss": 0.1182, + "step": 13395 + }, + { + "epoch": 0.11628371281499293, + "grad_norm": 0.2412109375, + "learning_rate": 0.001925446344381928, + "loss": 0.1006, + "step": 13396 + }, + { + "epoch": 0.1162923932951971, + "grad_norm": 0.30859375, + "learning_rate": 0.0019254344831992518, + "loss": 0.1621, + "step": 13397 + }, + { + "epoch": 0.11630107377540126, + "grad_norm": 0.333984375, + "learning_rate": 0.001925422621113892, + "loss": 0.0918, + "step": 13398 + }, + { + "epoch": 0.11630975425560543, + "grad_norm": 0.158203125, + "learning_rate": 0.001925410758125863, + "loss": 0.1318, + "step": 13399 + }, + { + "epoch": 0.11631843473580958, + "grad_norm": 0.376953125, + "learning_rate": 0.0019253988942351768, + "loss": 0.1172, + "step": 13400 + }, + { + "epoch": 0.11632711521601374, + "grad_norm": 0.69140625, + "learning_rate": 0.001925387029441847, + "loss": 0.1152, + "step": 13401 + }, + { + "epoch": 0.11633579569621791, + "grad_norm": 0.9140625, + "learning_rate": 0.0019253751637458864, + "loss": 0.1055, + "step": 13402 + }, + { + "epoch": 0.11634447617642207, + "grad_norm": 0.41015625, + "learning_rate": 0.0019253632971473079, + "loss": 0.126, + "step": 13403 + }, + { + "epoch": 0.11635315665662624, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019253514296461243, + "loss": 0.1279, + "step": 13404 + }, + { + "epoch": 0.1163618371368304, + "grad_norm": 0.53125, + "learning_rate": 0.0019253395612423487, + "loss": 0.1377, + "step": 13405 + }, + { + "epoch": 0.11637051761703457, + "grad_norm": 0.251953125, + "learning_rate": 0.0019253276919359945, + "loss": 0.1289, + "step": 13406 + }, + { + "epoch": 0.11637919809723873, + "grad_norm": 0.26171875, + "learning_rate": 0.0019253158217270741, + "loss": 0.1191, + "step": 13407 + }, + { + "epoch": 0.1163878785774429, + "grad_norm": 0.11669921875, + "learning_rate": 0.001925303950615601, + "loss": 0.0874, + "step": 13408 + }, + { + "epoch": 0.11639655905764706, + "grad_norm": 0.16796875, + "learning_rate": 0.0019252920786015876, + "loss": 0.1138, + "step": 13409 + }, + { + "epoch": 0.11640523953785123, + "grad_norm": 0.5390625, + "learning_rate": 0.001925280205685047, + "loss": 0.1123, + "step": 13410 + }, + { + "epoch": 0.1164139200180554, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019252683318659927, + "loss": 0.0947, + "step": 13411 + }, + { + "epoch": 0.11642260049825956, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019252564571444375, + "loss": 0.103, + "step": 13412 + }, + { + "epoch": 0.11643128097846372, + "grad_norm": 0.1474609375, + "learning_rate": 0.001925244581520394, + "loss": 0.1201, + "step": 13413 + }, + { + "epoch": 0.11643996145866789, + "grad_norm": 0.37109375, + "learning_rate": 0.0019252327049938756, + "loss": 0.1201, + "step": 13414 + }, + { + "epoch": 0.11644864193887206, + "grad_norm": 0.1796875, + "learning_rate": 0.0019252208275648947, + "loss": 0.106, + "step": 13415 + }, + { + "epoch": 0.11645732241907622, + "grad_norm": 0.2216796875, + "learning_rate": 0.001925208949233465, + "loss": 0.1074, + "step": 13416 + }, + { + "epoch": 0.11646600289928039, + "grad_norm": 0.291015625, + "learning_rate": 0.0019251970699995992, + "loss": 0.1396, + "step": 13417 + }, + { + "epoch": 0.11647468337948455, + "grad_norm": 0.220703125, + "learning_rate": 0.0019251851898633102, + "loss": 0.1064, + "step": 13418 + }, + { + "epoch": 0.11648336385968872, + "grad_norm": 0.263671875, + "learning_rate": 0.0019251733088246114, + "loss": 0.0889, + "step": 13419 + }, + { + "epoch": 0.11649204433989288, + "grad_norm": 0.11181640625, + "learning_rate": 0.001925161426883515, + "loss": 0.1172, + "step": 13420 + }, + { + "epoch": 0.11650072482009705, + "grad_norm": 0.1484375, + "learning_rate": 0.0019251495440400347, + "loss": 0.1055, + "step": 13421 + }, + { + "epoch": 0.11650940530030121, + "grad_norm": 0.625, + "learning_rate": 0.001925137660294183, + "loss": 0.124, + "step": 13422 + }, + { + "epoch": 0.11651808578050538, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019251257756459734, + "loss": 0.1191, + "step": 13423 + }, + { + "epoch": 0.11652676626070954, + "grad_norm": 0.42578125, + "learning_rate": 0.0019251138900954186, + "loss": 0.0757, + "step": 13424 + }, + { + "epoch": 0.1165354467409137, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019251020036425317, + "loss": 0.103, + "step": 13425 + }, + { + "epoch": 0.11654412722111787, + "grad_norm": 0.71484375, + "learning_rate": 0.0019250901162873256, + "loss": 0.1309, + "step": 13426 + }, + { + "epoch": 0.11655280770132204, + "grad_norm": 0.38671875, + "learning_rate": 0.0019250782280298133, + "loss": 0.126, + "step": 13427 + }, + { + "epoch": 0.1165614881815262, + "grad_norm": 0.337890625, + "learning_rate": 0.001925066338870008, + "loss": 0.1011, + "step": 13428 + }, + { + "epoch": 0.11657016866173037, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019250544488079223, + "loss": 0.1143, + "step": 13429 + }, + { + "epoch": 0.11657884914193453, + "grad_norm": 0.326171875, + "learning_rate": 0.0019250425578435698, + "loss": 0.0972, + "step": 13430 + }, + { + "epoch": 0.1165875296221387, + "grad_norm": 0.08837890625, + "learning_rate": 0.001925030665976963, + "loss": 0.0957, + "step": 13431 + }, + { + "epoch": 0.11659621010234286, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019250187732081149, + "loss": 0.127, + "step": 13432 + }, + { + "epoch": 0.11660489058254703, + "grad_norm": 0.294921875, + "learning_rate": 0.0019250068795370388, + "loss": 0.1279, + "step": 13433 + }, + { + "epoch": 0.11661357106275119, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019249949849637475, + "loss": 0.1201, + "step": 13434 + }, + { + "epoch": 0.11662225154295536, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019249830894882544, + "loss": 0.123, + "step": 13435 + }, + { + "epoch": 0.11663093202315952, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019249711931105715, + "loss": 0.1504, + "step": 13436 + }, + { + "epoch": 0.11663961250336369, + "grad_norm": 0.2431640625, + "learning_rate": 0.001924959295830713, + "loss": 0.0947, + "step": 13437 + }, + { + "epoch": 0.11664829298356785, + "grad_norm": 0.41015625, + "learning_rate": 0.0019249473976486913, + "loss": 0.126, + "step": 13438 + }, + { + "epoch": 0.11665697346377202, + "grad_norm": 0.5, + "learning_rate": 0.0019249354985645195, + "loss": 0.1396, + "step": 13439 + }, + { + "epoch": 0.11666565394397618, + "grad_norm": 0.087890625, + "learning_rate": 0.0019249235985782107, + "loss": 0.0781, + "step": 13440 + }, + { + "epoch": 0.11667433442418035, + "grad_norm": 0.1796875, + "learning_rate": 0.0019249116976897778, + "loss": 0.083, + "step": 13441 + }, + { + "epoch": 0.11668301490438451, + "grad_norm": 0.0869140625, + "learning_rate": 0.001924899795899234, + "loss": 0.1006, + "step": 13442 + }, + { + "epoch": 0.11669169538458868, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019248878932065921, + "loss": 0.1221, + "step": 13443 + }, + { + "epoch": 0.11670037586479284, + "grad_norm": 0.1796875, + "learning_rate": 0.001924875989611865, + "loss": 0.1465, + "step": 13444 + }, + { + "epoch": 0.11670905634499701, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019248640851150661, + "loss": 0.1465, + "step": 13445 + }, + { + "epoch": 0.11671773682520117, + "grad_norm": 0.439453125, + "learning_rate": 0.0019248521797162082, + "loss": 0.1299, + "step": 13446 + }, + { + "epoch": 0.11672641730540534, + "grad_norm": 0.75390625, + "learning_rate": 0.0019248402734153046, + "loss": 0.1074, + "step": 13447 + }, + { + "epoch": 0.1167350977856095, + "grad_norm": 1.0234375, + "learning_rate": 0.0019248283662123675, + "loss": 0.1533, + "step": 13448 + }, + { + "epoch": 0.11674377826581367, + "grad_norm": 0.39453125, + "learning_rate": 0.0019248164581074107, + "loss": 0.1738, + "step": 13449 + }, + { + "epoch": 0.11675245874601783, + "grad_norm": 0.203125, + "learning_rate": 0.0019248045491004474, + "loss": 0.1602, + "step": 13450 + }, + { + "epoch": 0.116761139226222, + "grad_norm": 0.32421875, + "learning_rate": 0.00192479263919149, + "loss": 0.1152, + "step": 13451 + }, + { + "epoch": 0.11676981970642616, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019247807283805517, + "loss": 0.1387, + "step": 13452 + }, + { + "epoch": 0.11677850018663033, + "grad_norm": 0.1328125, + "learning_rate": 0.0019247688166676455, + "loss": 0.1094, + "step": 13453 + }, + { + "epoch": 0.1167871806668345, + "grad_norm": 0.2578125, + "learning_rate": 0.0019247569040527846, + "loss": 0.1377, + "step": 13454 + }, + { + "epoch": 0.11679586114703866, + "grad_norm": 0.875, + "learning_rate": 0.001924744990535982, + "loss": 0.125, + "step": 13455 + }, + { + "epoch": 0.11680454162724282, + "grad_norm": 0.55859375, + "learning_rate": 0.0019247330761172506, + "loss": 0.1084, + "step": 13456 + }, + { + "epoch": 0.11681322210744699, + "grad_norm": 0.58203125, + "learning_rate": 0.0019247211607966036, + "loss": 0.1143, + "step": 13457 + }, + { + "epoch": 0.11682190258765116, + "grad_norm": 0.154296875, + "learning_rate": 0.0019247092445740536, + "loss": 0.0879, + "step": 13458 + }, + { + "epoch": 0.11683058306785532, + "grad_norm": 0.59765625, + "learning_rate": 0.0019246973274496145, + "loss": 0.1934, + "step": 13459 + }, + { + "epoch": 0.11683926354805949, + "grad_norm": 0.5625, + "learning_rate": 0.0019246854094232984, + "loss": 0.166, + "step": 13460 + }, + { + "epoch": 0.11684794402826364, + "grad_norm": 0.208984375, + "learning_rate": 0.001924673490495119, + "loss": 0.1113, + "step": 13461 + }, + { + "epoch": 0.1168566245084678, + "grad_norm": 0.26171875, + "learning_rate": 0.001924661570665089, + "loss": 0.1064, + "step": 13462 + }, + { + "epoch": 0.11686530498867197, + "grad_norm": 0.181640625, + "learning_rate": 0.0019246496499332212, + "loss": 0.1152, + "step": 13463 + }, + { + "epoch": 0.11687398546887613, + "grad_norm": 0.1240234375, + "learning_rate": 0.001924637728299529, + "loss": 0.1143, + "step": 13464 + }, + { + "epoch": 0.1168826659490803, + "grad_norm": 0.29296875, + "learning_rate": 0.0019246258057640257, + "loss": 0.1367, + "step": 13465 + }, + { + "epoch": 0.11689134642928446, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019246138823267238, + "loss": 0.0791, + "step": 13466 + }, + { + "epoch": 0.11690002690948863, + "grad_norm": 0.24609375, + "learning_rate": 0.0019246019579876367, + "loss": 0.166, + "step": 13467 + }, + { + "epoch": 0.11690870738969279, + "grad_norm": 0.08251953125, + "learning_rate": 0.001924590032746777, + "loss": 0.1182, + "step": 13468 + }, + { + "epoch": 0.11691738786989696, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019245781066041584, + "loss": 0.1221, + "step": 13469 + }, + { + "epoch": 0.11692606835010112, + "grad_norm": 0.10791015625, + "learning_rate": 0.001924566179559793, + "loss": 0.1104, + "step": 13470 + }, + { + "epoch": 0.11693474883030529, + "grad_norm": 0.48828125, + "learning_rate": 0.001924554251613695, + "loss": 0.1001, + "step": 13471 + }, + { + "epoch": 0.11694342931050945, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019245423227658765, + "loss": 0.124, + "step": 13472 + }, + { + "epoch": 0.11695210979071362, + "grad_norm": 0.369140625, + "learning_rate": 0.001924530393016351, + "loss": 0.124, + "step": 13473 + }, + { + "epoch": 0.11696079027091778, + "grad_norm": 0.337890625, + "learning_rate": 0.0019245184623651315, + "loss": 0.1045, + "step": 13474 + }, + { + "epoch": 0.11696947075112195, + "grad_norm": 0.29296875, + "learning_rate": 0.0019245065308122308, + "loss": 0.1064, + "step": 13475 + }, + { + "epoch": 0.11697815123132611, + "grad_norm": 0.50390625, + "learning_rate": 0.0019244945983576626, + "loss": 0.0884, + "step": 13476 + }, + { + "epoch": 0.11698683171153028, + "grad_norm": 0.400390625, + "learning_rate": 0.0019244826650014393, + "loss": 0.0908, + "step": 13477 + }, + { + "epoch": 0.11699551219173444, + "grad_norm": 0.390625, + "learning_rate": 0.001924470730743574, + "loss": 0.1348, + "step": 13478 + }, + { + "epoch": 0.11700419267193861, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019244587955840797, + "loss": 0.1074, + "step": 13479 + }, + { + "epoch": 0.11701287315214277, + "grad_norm": 0.396484375, + "learning_rate": 0.00192444685952297, + "loss": 0.0908, + "step": 13480 + }, + { + "epoch": 0.11702155363234694, + "grad_norm": 0.462890625, + "learning_rate": 0.0019244349225602576, + "loss": 0.1064, + "step": 13481 + }, + { + "epoch": 0.1170302341125511, + "grad_norm": 0.111328125, + "learning_rate": 0.0019244229846959555, + "loss": 0.1064, + "step": 13482 + }, + { + "epoch": 0.11703891459275527, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019244110459300768, + "loss": 0.1084, + "step": 13483 + }, + { + "epoch": 0.11704759507295943, + "grad_norm": 0.146484375, + "learning_rate": 0.0019243991062626348, + "loss": 0.1699, + "step": 13484 + }, + { + "epoch": 0.1170562755531636, + "grad_norm": 0.1953125, + "learning_rate": 0.0019243871656936418, + "loss": 0.1426, + "step": 13485 + }, + { + "epoch": 0.11706495603336776, + "grad_norm": 0.08056640625, + "learning_rate": 0.0019243752242231115, + "loss": 0.1143, + "step": 13486 + }, + { + "epoch": 0.11707363651357193, + "grad_norm": 0.25390625, + "learning_rate": 0.001924363281851057, + "loss": 0.0986, + "step": 13487 + }, + { + "epoch": 0.1170823169937761, + "grad_norm": 0.080078125, + "learning_rate": 0.0019243513385774913, + "loss": 0.1152, + "step": 13488 + }, + { + "epoch": 0.11709099747398026, + "grad_norm": 0.9140625, + "learning_rate": 0.0019243393944024271, + "loss": 0.1914, + "step": 13489 + }, + { + "epoch": 0.11709967795418443, + "grad_norm": 1.421875, + "learning_rate": 0.001924327449325878, + "loss": 0.1641, + "step": 13490 + }, + { + "epoch": 0.11710835843438859, + "grad_norm": 0.455078125, + "learning_rate": 0.0019243155033478567, + "loss": 0.0977, + "step": 13491 + }, + { + "epoch": 0.11711703891459276, + "grad_norm": 0.296875, + "learning_rate": 0.0019243035564683764, + "loss": 0.1475, + "step": 13492 + }, + { + "epoch": 0.11712571939479692, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019242916086874501, + "loss": 0.1138, + "step": 13493 + }, + { + "epoch": 0.11713439987500109, + "grad_norm": 1.0546875, + "learning_rate": 0.0019242796600050904, + "loss": 0.1396, + "step": 13494 + }, + { + "epoch": 0.11714308035520525, + "grad_norm": 0.419921875, + "learning_rate": 0.0019242677104213117, + "loss": 0.2129, + "step": 13495 + }, + { + "epoch": 0.11715176083540942, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019242557599361257, + "loss": 0.105, + "step": 13496 + }, + { + "epoch": 0.11716044131561358, + "grad_norm": 0.474609375, + "learning_rate": 0.0019242438085495459, + "loss": 0.1816, + "step": 13497 + }, + { + "epoch": 0.11716912179581775, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019242318562615857, + "loss": 0.1191, + "step": 13498 + }, + { + "epoch": 0.11717780227602191, + "grad_norm": 0.126953125, + "learning_rate": 0.0019242199030722578, + "loss": 0.1416, + "step": 13499 + }, + { + "epoch": 0.11718648275622608, + "grad_norm": 0.2001953125, + "learning_rate": 0.0019242079489815757, + "loss": 0.1309, + "step": 13500 + }, + { + "epoch": 0.11719516323643024, + "grad_norm": 0.41015625, + "learning_rate": 0.0019241959939895518, + "loss": 0.0996, + "step": 13501 + }, + { + "epoch": 0.1172038437166344, + "grad_norm": 0.314453125, + "learning_rate": 0.0019241840380961994, + "loss": 0.1289, + "step": 13502 + }, + { + "epoch": 0.11721252419683857, + "grad_norm": 0.49609375, + "learning_rate": 0.0019241720813015322, + "loss": 0.1465, + "step": 13503 + }, + { + "epoch": 0.11722120467704274, + "grad_norm": 0.98046875, + "learning_rate": 0.0019241601236055625, + "loss": 0.1113, + "step": 13504 + }, + { + "epoch": 0.1172298851572469, + "grad_norm": 0.62890625, + "learning_rate": 0.0019241481650083038, + "loss": 0.1348, + "step": 13505 + }, + { + "epoch": 0.11723856563745107, + "grad_norm": 0.86328125, + "learning_rate": 0.001924136205509769, + "loss": 0.1079, + "step": 13506 + }, + { + "epoch": 0.11724724611765523, + "grad_norm": 0.5625, + "learning_rate": 0.001924124245109971, + "loss": 0.1533, + "step": 13507 + }, + { + "epoch": 0.1172559265978594, + "grad_norm": 0.15234375, + "learning_rate": 0.0019241122838089236, + "loss": 0.1328, + "step": 13508 + }, + { + "epoch": 0.11726460707806356, + "grad_norm": 0.314453125, + "learning_rate": 0.0019241003216066389, + "loss": 0.127, + "step": 13509 + }, + { + "epoch": 0.11727328755826773, + "grad_norm": 0.3984375, + "learning_rate": 0.0019240883585031308, + "loss": 0.1758, + "step": 13510 + }, + { + "epoch": 0.11728196803847189, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019240763944984118, + "loss": 0.1143, + "step": 13511 + }, + { + "epoch": 0.11729064851867606, + "grad_norm": 0.37890625, + "learning_rate": 0.0019240644295924958, + "loss": 0.1338, + "step": 13512 + }, + { + "epoch": 0.11729932899888022, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019240524637853947, + "loss": 0.1377, + "step": 13513 + }, + { + "epoch": 0.11730800947908439, + "grad_norm": 0.91796875, + "learning_rate": 0.0019240404970771224, + "loss": 0.1357, + "step": 13514 + }, + { + "epoch": 0.11731668995928855, + "grad_norm": 0.3046875, + "learning_rate": 0.0019240285294676918, + "loss": 0.1465, + "step": 13515 + }, + { + "epoch": 0.11732537043949272, + "grad_norm": 0.43359375, + "learning_rate": 0.001924016560957116, + "loss": 0.1396, + "step": 13516 + }, + { + "epoch": 0.11733405091969688, + "grad_norm": 0.275390625, + "learning_rate": 0.0019240045915454081, + "loss": 0.1177, + "step": 13517 + }, + { + "epoch": 0.11734273139990105, + "grad_norm": 0.177734375, + "learning_rate": 0.001923992621232581, + "loss": 0.083, + "step": 13518 + }, + { + "epoch": 0.11735141188010521, + "grad_norm": 0.08984375, + "learning_rate": 0.0019239806500186481, + "loss": 0.1147, + "step": 13519 + }, + { + "epoch": 0.11736009236030938, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019239686779036222, + "loss": 0.1133, + "step": 13520 + }, + { + "epoch": 0.11736877284051354, + "grad_norm": 0.337890625, + "learning_rate": 0.0019239567048875165, + "loss": 0.1025, + "step": 13521 + }, + { + "epoch": 0.11737745332071771, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019239447309703445, + "loss": 0.1445, + "step": 13522 + }, + { + "epoch": 0.11738613380092186, + "grad_norm": 1.0546875, + "learning_rate": 0.0019239327561521187, + "loss": 0.1592, + "step": 13523 + }, + { + "epoch": 0.11739481428112603, + "grad_norm": 0.75390625, + "learning_rate": 0.0019239207804328524, + "loss": 0.123, + "step": 13524 + }, + { + "epoch": 0.11740349476133019, + "grad_norm": 0.1884765625, + "learning_rate": 0.001923908803812559, + "loss": 0.1094, + "step": 13525 + }, + { + "epoch": 0.11741217524153436, + "grad_norm": 0.271484375, + "learning_rate": 0.001923896826291251, + "loss": 0.123, + "step": 13526 + }, + { + "epoch": 0.11742085572173852, + "grad_norm": 0.369140625, + "learning_rate": 0.0019238848478689418, + "loss": 0.1045, + "step": 13527 + }, + { + "epoch": 0.11742953620194269, + "grad_norm": 0.337890625, + "learning_rate": 0.0019238728685456443, + "loss": 0.0811, + "step": 13528 + }, + { + "epoch": 0.11743821668214685, + "grad_norm": 0.1181640625, + "learning_rate": 0.001923860888321372, + "loss": 0.0869, + "step": 13529 + }, + { + "epoch": 0.11744689716235102, + "grad_norm": 0.1953125, + "learning_rate": 0.001923848907196138, + "loss": 0.0874, + "step": 13530 + }, + { + "epoch": 0.11745557764255518, + "grad_norm": 0.25, + "learning_rate": 0.001923836925169955, + "loss": 0.0977, + "step": 13531 + }, + { + "epoch": 0.11746425812275935, + "grad_norm": 0.107421875, + "learning_rate": 0.0019238249422428365, + "loss": 0.1113, + "step": 13532 + }, + { + "epoch": 0.11747293860296351, + "grad_norm": 0.404296875, + "learning_rate": 0.0019238129584147952, + "loss": 0.0962, + "step": 13533 + }, + { + "epoch": 0.11748161908316768, + "grad_norm": 0.6171875, + "learning_rate": 0.0019238009736858447, + "loss": 0.1074, + "step": 13534 + }, + { + "epoch": 0.11749029956337184, + "grad_norm": 1.0, + "learning_rate": 0.0019237889880559975, + "loss": 0.0825, + "step": 13535 + }, + { + "epoch": 0.117498980043576, + "grad_norm": 0.423828125, + "learning_rate": 0.0019237770015252673, + "loss": 0.1045, + "step": 13536 + }, + { + "epoch": 0.11750766052378017, + "grad_norm": 0.56640625, + "learning_rate": 0.0019237650140936668, + "loss": 0.1621, + "step": 13537 + }, + { + "epoch": 0.11751634100398434, + "grad_norm": 0.224609375, + "learning_rate": 0.0019237530257612092, + "loss": 0.0996, + "step": 13538 + }, + { + "epoch": 0.1175250214841885, + "grad_norm": 0.32421875, + "learning_rate": 0.001923741036527908, + "loss": 0.1289, + "step": 13539 + }, + { + "epoch": 0.11753370196439267, + "grad_norm": 0.4296875, + "learning_rate": 0.0019237290463937759, + "loss": 0.1523, + "step": 13540 + }, + { + "epoch": 0.11754238244459683, + "grad_norm": 0.32421875, + "learning_rate": 0.001923717055358826, + "loss": 0.1045, + "step": 13541 + }, + { + "epoch": 0.117551062924801, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019237050634230714, + "loss": 0.1162, + "step": 13542 + }, + { + "epoch": 0.11755974340500516, + "grad_norm": 0.2578125, + "learning_rate": 0.0019236930705865253, + "loss": 0.1299, + "step": 13543 + }, + { + "epoch": 0.11756842388520933, + "grad_norm": 0.07958984375, + "learning_rate": 0.001923681076849201, + "loss": 0.0996, + "step": 13544 + }, + { + "epoch": 0.11757710436541349, + "grad_norm": 0.337890625, + "learning_rate": 0.0019236690822111114, + "loss": 0.1338, + "step": 13545 + }, + { + "epoch": 0.11758578484561766, + "grad_norm": 0.357421875, + "learning_rate": 0.0019236570866722697, + "loss": 0.0859, + "step": 13546 + }, + { + "epoch": 0.11759446532582182, + "grad_norm": 0.2890625, + "learning_rate": 0.0019236450902326888, + "loss": 0.1104, + "step": 13547 + }, + { + "epoch": 0.11760314580602599, + "grad_norm": 0.419921875, + "learning_rate": 0.0019236330928923823, + "loss": 0.1221, + "step": 13548 + }, + { + "epoch": 0.11761182628623015, + "grad_norm": 0.34765625, + "learning_rate": 0.0019236210946513626, + "loss": 0.1289, + "step": 13549 + }, + { + "epoch": 0.11762050676643432, + "grad_norm": 0.53515625, + "learning_rate": 0.0019236090955096435, + "loss": 0.1484, + "step": 13550 + }, + { + "epoch": 0.11762918724663848, + "grad_norm": 0.15234375, + "learning_rate": 0.0019235970954672378, + "loss": 0.1309, + "step": 13551 + }, + { + "epoch": 0.11763786772684265, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019235850945241587, + "loss": 0.0923, + "step": 13552 + }, + { + "epoch": 0.11764654820704681, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019235730926804196, + "loss": 0.1211, + "step": 13553 + }, + { + "epoch": 0.11765522868725098, + "grad_norm": 0.33984375, + "learning_rate": 0.001923561089936033, + "loss": 0.0996, + "step": 13554 + }, + { + "epoch": 0.11766390916745514, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019235490862910126, + "loss": 0.1367, + "step": 13555 + }, + { + "epoch": 0.11767258964765931, + "grad_norm": 0.40625, + "learning_rate": 0.0019235370817453712, + "loss": 0.1074, + "step": 13556 + }, + { + "epoch": 0.11768127012786347, + "grad_norm": 0.65625, + "learning_rate": 0.001923525076299122, + "loss": 0.1777, + "step": 13557 + }, + { + "epoch": 0.11768995060806764, + "grad_norm": 0.291015625, + "learning_rate": 0.001923513069952278, + "loss": 0.1289, + "step": 13558 + }, + { + "epoch": 0.1176986310882718, + "grad_norm": 0.287109375, + "learning_rate": 0.0019235010627048526, + "loss": 0.123, + "step": 13559 + }, + { + "epoch": 0.11770731156847597, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019234890545568588, + "loss": 0.0967, + "step": 13560 + }, + { + "epoch": 0.11771599204868013, + "grad_norm": 0.400390625, + "learning_rate": 0.0019234770455083098, + "loss": 0.1562, + "step": 13561 + }, + { + "epoch": 0.1177246725288843, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019234650355592186, + "loss": 0.0698, + "step": 13562 + }, + { + "epoch": 0.11773335300908847, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019234530247095984, + "loss": 0.1084, + "step": 13563 + }, + { + "epoch": 0.11774203348929263, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019234410129594626, + "loss": 0.0898, + "step": 13564 + }, + { + "epoch": 0.1177507139694968, + "grad_norm": 0.185546875, + "learning_rate": 0.0019234290003088237, + "loss": 0.1104, + "step": 13565 + }, + { + "epoch": 0.11775939444970096, + "grad_norm": 1.65625, + "learning_rate": 0.0019234169867576954, + "loss": 0.1216, + "step": 13566 + }, + { + "epoch": 0.11776807492990513, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019234049723060908, + "loss": 0.1367, + "step": 13567 + }, + { + "epoch": 0.11777675541010929, + "grad_norm": 0.796875, + "learning_rate": 0.0019233929569540226, + "loss": 0.1562, + "step": 13568 + }, + { + "epoch": 0.11778543589031346, + "grad_norm": 0.1630859375, + "learning_rate": 0.0019233809407015047, + "loss": 0.1055, + "step": 13569 + }, + { + "epoch": 0.11779411637051762, + "grad_norm": 0.8046875, + "learning_rate": 0.0019233689235485492, + "loss": 0.0801, + "step": 13570 + }, + { + "epoch": 0.11780279685072179, + "grad_norm": 0.32421875, + "learning_rate": 0.0019233569054951701, + "loss": 0.1167, + "step": 13571 + }, + { + "epoch": 0.11781147733092595, + "grad_norm": 0.77734375, + "learning_rate": 0.0019233448865413802, + "loss": 0.0869, + "step": 13572 + }, + { + "epoch": 0.11782015781113012, + "grad_norm": 0.328125, + "learning_rate": 0.001923332866687193, + "loss": 0.1094, + "step": 13573 + }, + { + "epoch": 0.11782883829133428, + "grad_norm": 0.1640625, + "learning_rate": 0.0019233208459326209, + "loss": 0.1064, + "step": 13574 + }, + { + "epoch": 0.11783751877153845, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019233088242776777, + "loss": 0.0781, + "step": 13575 + }, + { + "epoch": 0.11784619925174261, + "grad_norm": 0.150390625, + "learning_rate": 0.0019232968017223763, + "loss": 0.1416, + "step": 13576 + }, + { + "epoch": 0.11785487973194678, + "grad_norm": 0.061279296875, + "learning_rate": 0.0019232847782667303, + "loss": 0.0854, + "step": 13577 + }, + { + "epoch": 0.11786356021215094, + "grad_norm": 0.55859375, + "learning_rate": 0.0019232727539107515, + "loss": 0.0996, + "step": 13578 + }, + { + "epoch": 0.1178722406923551, + "grad_norm": 1.2734375, + "learning_rate": 0.0019232607286544547, + "loss": 0.1416, + "step": 13579 + }, + { + "epoch": 0.11788092117255927, + "grad_norm": 0.74609375, + "learning_rate": 0.0019232487024978522, + "loss": 0.1113, + "step": 13580 + }, + { + "epoch": 0.11788960165276344, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019232366754409571, + "loss": 0.1201, + "step": 13581 + }, + { + "epoch": 0.1178982821329676, + "grad_norm": 0.396484375, + "learning_rate": 0.001923224647483783, + "loss": 0.1182, + "step": 13582 + }, + { + "epoch": 0.11790696261317177, + "grad_norm": 0.490234375, + "learning_rate": 0.0019232126186263427, + "loss": 0.1025, + "step": 13583 + }, + { + "epoch": 0.11791564309337592, + "grad_norm": 0.392578125, + "learning_rate": 0.0019232005888686495, + "loss": 0.1157, + "step": 13584 + }, + { + "epoch": 0.11792432357358008, + "grad_norm": 0.33203125, + "learning_rate": 0.0019231885582107165, + "loss": 0.0996, + "step": 13585 + }, + { + "epoch": 0.11793300405378425, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019231765266525568, + "loss": 0.1064, + "step": 13586 + }, + { + "epoch": 0.11794168453398841, + "grad_norm": 1.3828125, + "learning_rate": 0.0019231644941941834, + "loss": 0.126, + "step": 13587 + }, + { + "epoch": 0.11795036501419258, + "grad_norm": 0.078125, + "learning_rate": 0.0019231524608356101, + "loss": 0.126, + "step": 13588 + }, + { + "epoch": 0.11795904549439674, + "grad_norm": 0.06201171875, + "learning_rate": 0.0019231404265768492, + "loss": 0.0908, + "step": 13589 + }, + { + "epoch": 0.11796772597460091, + "grad_norm": 0.85546875, + "learning_rate": 0.0019231283914179147, + "loss": 0.1484, + "step": 13590 + }, + { + "epoch": 0.11797640645480507, + "grad_norm": 1.109375, + "learning_rate": 0.001923116355358819, + "loss": 0.1289, + "step": 13591 + }, + { + "epoch": 0.11798508693500924, + "grad_norm": 0.361328125, + "learning_rate": 0.001923104318399576, + "loss": 0.1309, + "step": 13592 + }, + { + "epoch": 0.1179937674152134, + "grad_norm": 0.65625, + "learning_rate": 0.0019230922805401982, + "loss": 0.1104, + "step": 13593 + }, + { + "epoch": 0.11800244789541757, + "grad_norm": 0.20703125, + "learning_rate": 0.0019230802417806992, + "loss": 0.1357, + "step": 13594 + }, + { + "epoch": 0.11801112837562173, + "grad_norm": 0.58984375, + "learning_rate": 0.001923068202121092, + "loss": 0.0996, + "step": 13595 + }, + { + "epoch": 0.1180198088558259, + "grad_norm": 0.9296875, + "learning_rate": 0.0019230561615613897, + "loss": 0.1191, + "step": 13596 + }, + { + "epoch": 0.11802848933603007, + "grad_norm": 0.4140625, + "learning_rate": 0.0019230441201016058, + "loss": 0.0771, + "step": 13597 + }, + { + "epoch": 0.11803716981623423, + "grad_norm": 0.0966796875, + "learning_rate": 0.001923032077741753, + "loss": 0.1211, + "step": 13598 + }, + { + "epoch": 0.1180458502964384, + "grad_norm": 0.494140625, + "learning_rate": 0.0019230200344818446, + "loss": 0.1152, + "step": 13599 + }, + { + "epoch": 0.11805453077664256, + "grad_norm": 0.80078125, + "learning_rate": 0.0019230079903218937, + "loss": 0.1133, + "step": 13600 + }, + { + "epoch": 0.11806321125684673, + "grad_norm": 0.11279296875, + "learning_rate": 0.001922995945261914, + "loss": 0.0967, + "step": 13601 + }, + { + "epoch": 0.11807189173705089, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019229838993019182, + "loss": 0.0737, + "step": 13602 + }, + { + "epoch": 0.11808057221725506, + "grad_norm": 0.283203125, + "learning_rate": 0.0019229718524419197, + "loss": 0.1562, + "step": 13603 + }, + { + "epoch": 0.11808925269745922, + "grad_norm": 0.7265625, + "learning_rate": 0.0019229598046819312, + "loss": 0.1191, + "step": 13604 + }, + { + "epoch": 0.11809793317766339, + "grad_norm": 1.484375, + "learning_rate": 0.0019229477560219663, + "loss": 0.1133, + "step": 13605 + }, + { + "epoch": 0.11810661365786755, + "grad_norm": 0.267578125, + "learning_rate": 0.0019229357064620386, + "loss": 0.0977, + "step": 13606 + }, + { + "epoch": 0.11811529413807172, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019229236560021601, + "loss": 0.1084, + "step": 13607 + }, + { + "epoch": 0.11812397461827588, + "grad_norm": 0.08251953125, + "learning_rate": 0.001922911604642345, + "loss": 0.0728, + "step": 13608 + }, + { + "epoch": 0.11813265509848005, + "grad_norm": 0.263671875, + "learning_rate": 0.0019228995523826063, + "loss": 0.1123, + "step": 13609 + }, + { + "epoch": 0.11814133557868421, + "grad_norm": 0.10498046875, + "learning_rate": 0.0019228874992229567, + "loss": 0.1021, + "step": 13610 + }, + { + "epoch": 0.11815001605888838, + "grad_norm": 0.2421875, + "learning_rate": 0.0019228754451634097, + "loss": 0.1348, + "step": 13611 + }, + { + "epoch": 0.11815869653909254, + "grad_norm": 0.291015625, + "learning_rate": 0.0019228633902039786, + "loss": 0.0845, + "step": 13612 + }, + { + "epoch": 0.11816737701929671, + "grad_norm": 0.2373046875, + "learning_rate": 0.0019228513343446762, + "loss": 0.1631, + "step": 13613 + }, + { + "epoch": 0.11817605749950087, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019228392775855164, + "loss": 0.127, + "step": 13614 + }, + { + "epoch": 0.11818473797970504, + "grad_norm": 0.349609375, + "learning_rate": 0.0019228272199265115, + "loss": 0.0947, + "step": 13615 + }, + { + "epoch": 0.1181934184599092, + "grad_norm": 0.51953125, + "learning_rate": 0.0019228151613676753, + "loss": 0.0654, + "step": 13616 + }, + { + "epoch": 0.11820209894011337, + "grad_norm": 0.193359375, + "learning_rate": 0.0019228031019090208, + "loss": 0.1133, + "step": 13617 + }, + { + "epoch": 0.11821077942031753, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019227910415505615, + "loss": 0.1006, + "step": 13618 + }, + { + "epoch": 0.1182194599005217, + "grad_norm": 0.09814453125, + "learning_rate": 0.00192277898029231, + "loss": 0.1289, + "step": 13619 + }, + { + "epoch": 0.11822814038072586, + "grad_norm": 1.2890625, + "learning_rate": 0.0019227669181342796, + "loss": 0.1982, + "step": 13620 + }, + { + "epoch": 0.11823682086093003, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019227548550764838, + "loss": 0.1226, + "step": 13621 + }, + { + "epoch": 0.1182455013411342, + "grad_norm": 1.0546875, + "learning_rate": 0.0019227427911189356, + "loss": 0.0996, + "step": 13622 + }, + { + "epoch": 0.11825418182133836, + "grad_norm": 0.78125, + "learning_rate": 0.0019227307262616481, + "loss": 0.124, + "step": 13623 + }, + { + "epoch": 0.11826286230154252, + "grad_norm": 1.546875, + "learning_rate": 0.0019227186605046353, + "loss": 0.1133, + "step": 13624 + }, + { + "epoch": 0.11827154278174669, + "grad_norm": 0.478515625, + "learning_rate": 0.0019227065938479092, + "loss": 0.0996, + "step": 13625 + }, + { + "epoch": 0.11828022326195085, + "grad_norm": 0.703125, + "learning_rate": 0.0019226945262914837, + "loss": 0.1177, + "step": 13626 + }, + { + "epoch": 0.11828890374215502, + "grad_norm": 0.29296875, + "learning_rate": 0.0019226824578353714, + "loss": 0.0913, + "step": 13627 + }, + { + "epoch": 0.11829758422235918, + "grad_norm": 0.40234375, + "learning_rate": 0.0019226703884795864, + "loss": 0.1016, + "step": 13628 + }, + { + "epoch": 0.11830626470256335, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019226583182241413, + "loss": 0.125, + "step": 13629 + }, + { + "epoch": 0.11831494518276751, + "grad_norm": 0.37890625, + "learning_rate": 0.0019226462470690493, + "loss": 0.209, + "step": 13630 + }, + { + "epoch": 0.11832362566297168, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019226341750143239, + "loss": 0.1006, + "step": 13631 + }, + { + "epoch": 0.11833230614317584, + "grad_norm": 0.416015625, + "learning_rate": 0.0019226221020599782, + "loss": 0.127, + "step": 13632 + }, + { + "epoch": 0.11834098662338001, + "grad_norm": 0.5234375, + "learning_rate": 0.0019226100282060251, + "loss": 0.0986, + "step": 13633 + }, + { + "epoch": 0.11834966710358417, + "grad_norm": 0.421875, + "learning_rate": 0.001922597953452478, + "loss": 0.0996, + "step": 13634 + }, + { + "epoch": 0.11835834758378834, + "grad_norm": 0.63671875, + "learning_rate": 0.0019225858777993507, + "loss": 0.0957, + "step": 13635 + }, + { + "epoch": 0.1183670280639925, + "grad_norm": 0.5546875, + "learning_rate": 0.0019225738012466549, + "loss": 0.1064, + "step": 13636 + }, + { + "epoch": 0.11837570854419667, + "grad_norm": 0.4140625, + "learning_rate": 0.0019225617237944054, + "loss": 0.0986, + "step": 13637 + }, + { + "epoch": 0.11838438902440084, + "grad_norm": 0.33203125, + "learning_rate": 0.001922549645442615, + "loss": 0.1367, + "step": 13638 + }, + { + "epoch": 0.118393069504605, + "grad_norm": 0.80078125, + "learning_rate": 0.001922537566191296, + "loss": 0.1221, + "step": 13639 + }, + { + "epoch": 0.11840174998480917, + "grad_norm": 0.1171875, + "learning_rate": 0.0019225254860404627, + "loss": 0.1533, + "step": 13640 + }, + { + "epoch": 0.11841043046501333, + "grad_norm": 0.365234375, + "learning_rate": 0.0019225134049901274, + "loss": 0.1016, + "step": 13641 + }, + { + "epoch": 0.1184191109452175, + "grad_norm": 0.18359375, + "learning_rate": 0.0019225013230403044, + "loss": 0.1035, + "step": 13642 + }, + { + "epoch": 0.11842779142542166, + "grad_norm": 0.404296875, + "learning_rate": 0.0019224892401910059, + "loss": 0.1729, + "step": 13643 + }, + { + "epoch": 0.11843647190562583, + "grad_norm": 0.296875, + "learning_rate": 0.0019224771564422456, + "loss": 0.1348, + "step": 13644 + }, + { + "epoch": 0.11844515238582999, + "grad_norm": 0.1396484375, + "learning_rate": 0.0019224650717940368, + "loss": 0.1021, + "step": 13645 + }, + { + "epoch": 0.11845383286603414, + "grad_norm": 0.287109375, + "learning_rate": 0.0019224529862463926, + "loss": 0.1699, + "step": 13646 + }, + { + "epoch": 0.11846251334623831, + "grad_norm": 0.279296875, + "learning_rate": 0.0019224408997993258, + "loss": 0.1177, + "step": 13647 + }, + { + "epoch": 0.11847119382644247, + "grad_norm": 0.390625, + "learning_rate": 0.0019224288124528503, + "loss": 0.1348, + "step": 13648 + }, + { + "epoch": 0.11847987430664664, + "grad_norm": 0.330078125, + "learning_rate": 0.001922416724206979, + "loss": 0.1348, + "step": 13649 + }, + { + "epoch": 0.1184885547868508, + "grad_norm": 0.8203125, + "learning_rate": 0.001922404635061725, + "loss": 0.1152, + "step": 13650 + }, + { + "epoch": 0.11849723526705497, + "grad_norm": 0.46875, + "learning_rate": 0.0019223925450171017, + "loss": 0.0898, + "step": 13651 + }, + { + "epoch": 0.11850591574725913, + "grad_norm": 0.65234375, + "learning_rate": 0.0019223804540731224, + "loss": 0.1064, + "step": 13652 + }, + { + "epoch": 0.1185145962274633, + "grad_norm": 0.41796875, + "learning_rate": 0.0019223683622298, + "loss": 0.125, + "step": 13653 + }, + { + "epoch": 0.11852327670766746, + "grad_norm": 0.287109375, + "learning_rate": 0.001922356269487148, + "loss": 0.083, + "step": 13654 + }, + { + "epoch": 0.11853195718787163, + "grad_norm": 0.7265625, + "learning_rate": 0.0019223441758451797, + "loss": 0.0947, + "step": 13655 + }, + { + "epoch": 0.1185406376680758, + "grad_norm": 0.65234375, + "learning_rate": 0.001922332081303908, + "loss": 0.1299, + "step": 13656 + }, + { + "epoch": 0.11854931814827996, + "grad_norm": 0.126953125, + "learning_rate": 0.001922319985863346, + "loss": 0.0806, + "step": 13657 + }, + { + "epoch": 0.11855799862848412, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019223078895235079, + "loss": 0.0918, + "step": 13658 + }, + { + "epoch": 0.11856667910868829, + "grad_norm": 0.30859375, + "learning_rate": 0.0019222957922844055, + "loss": 0.0752, + "step": 13659 + }, + { + "epoch": 0.11857535958889245, + "grad_norm": 0.4453125, + "learning_rate": 0.0019222836941460532, + "loss": 0.124, + "step": 13660 + }, + { + "epoch": 0.11858404006909662, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019222715951084637, + "loss": 0.1045, + "step": 13661 + }, + { + "epoch": 0.11859272054930078, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019222594951716504, + "loss": 0.1001, + "step": 13662 + }, + { + "epoch": 0.11860140102950495, + "grad_norm": 0.4296875, + "learning_rate": 0.0019222473943356266, + "loss": 0.1104, + "step": 13663 + }, + { + "epoch": 0.11861008150970911, + "grad_norm": 0.98046875, + "learning_rate": 0.0019222352926004053, + "loss": 0.1309, + "step": 13664 + }, + { + "epoch": 0.11861876198991328, + "grad_norm": 0.259765625, + "learning_rate": 0.0019222231899659998, + "loss": 0.1201, + "step": 13665 + }, + { + "epoch": 0.11862744247011744, + "grad_norm": 0.97265625, + "learning_rate": 0.0019222110864324231, + "loss": 0.1406, + "step": 13666 + }, + { + "epoch": 0.11863612295032161, + "grad_norm": 0.341796875, + "learning_rate": 0.001922198981999689, + "loss": 0.1279, + "step": 13667 + }, + { + "epoch": 0.11864480343052577, + "grad_norm": 0.73828125, + "learning_rate": 0.0019221868766678106, + "loss": 0.1387, + "step": 13668 + }, + { + "epoch": 0.11865348391072994, + "grad_norm": 0.19140625, + "learning_rate": 0.0019221747704368007, + "loss": 0.1074, + "step": 13669 + }, + { + "epoch": 0.1186621643909341, + "grad_norm": 0.22265625, + "learning_rate": 0.0019221626633066729, + "loss": 0.1123, + "step": 13670 + }, + { + "epoch": 0.11867084487113827, + "grad_norm": 0.59375, + "learning_rate": 0.0019221505552774406, + "loss": 0.0889, + "step": 13671 + }, + { + "epoch": 0.11867952535134244, + "grad_norm": 0.255859375, + "learning_rate": 0.0019221384463491164, + "loss": 0.0767, + "step": 13672 + }, + { + "epoch": 0.1186882058315466, + "grad_norm": 0.78515625, + "learning_rate": 0.001922126336521714, + "loss": 0.1187, + "step": 13673 + }, + { + "epoch": 0.11869688631175077, + "grad_norm": 0.337890625, + "learning_rate": 0.0019221142257952468, + "loss": 0.1318, + "step": 13674 + }, + { + "epoch": 0.11870556679195493, + "grad_norm": 0.625, + "learning_rate": 0.0019221021141697277, + "loss": 0.1196, + "step": 13675 + }, + { + "epoch": 0.1187142472721591, + "grad_norm": 0.5703125, + "learning_rate": 0.0019220900016451703, + "loss": 0.1309, + "step": 13676 + }, + { + "epoch": 0.11872292775236326, + "grad_norm": 0.392578125, + "learning_rate": 0.0019220778882215875, + "loss": 0.0977, + "step": 13677 + }, + { + "epoch": 0.11873160823256743, + "grad_norm": 0.578125, + "learning_rate": 0.0019220657738989928, + "loss": 0.1118, + "step": 13678 + }, + { + "epoch": 0.11874028871277159, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019220536586773988, + "loss": 0.1328, + "step": 13679 + }, + { + "epoch": 0.11874896919297576, + "grad_norm": 0.1416015625, + "learning_rate": 0.00192204154255682, + "loss": 0.1201, + "step": 13680 + }, + { + "epoch": 0.11875764967317992, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019220294255372684, + "loss": 0.0952, + "step": 13681 + }, + { + "epoch": 0.11876633015338409, + "grad_norm": 0.447265625, + "learning_rate": 0.001922017307618758, + "loss": 0.1162, + "step": 13682 + }, + { + "epoch": 0.11877501063358825, + "grad_norm": 0.310546875, + "learning_rate": 0.0019220051888013017, + "loss": 0.1484, + "step": 13683 + }, + { + "epoch": 0.11878369111379242, + "grad_norm": 0.5234375, + "learning_rate": 0.001921993069084913, + "loss": 0.1318, + "step": 13684 + }, + { + "epoch": 0.11879237159399658, + "grad_norm": 0.349609375, + "learning_rate": 0.001921980948469605, + "loss": 0.0938, + "step": 13685 + }, + { + "epoch": 0.11880105207420075, + "grad_norm": 0.1845703125, + "learning_rate": 0.001921968826955391, + "loss": 0.0952, + "step": 13686 + }, + { + "epoch": 0.11880973255440491, + "grad_norm": 0.0751953125, + "learning_rate": 0.0019219567045422842, + "loss": 0.0967, + "step": 13687 + }, + { + "epoch": 0.11881841303460908, + "grad_norm": 0.140625, + "learning_rate": 0.0019219445812302977, + "loss": 0.123, + "step": 13688 + }, + { + "epoch": 0.11882709351481324, + "grad_norm": 0.318359375, + "learning_rate": 0.0019219324570194452, + "loss": 0.106, + "step": 13689 + }, + { + "epoch": 0.11883577399501741, + "grad_norm": 0.333984375, + "learning_rate": 0.0019219203319097396, + "loss": 0.1152, + "step": 13690 + }, + { + "epoch": 0.11884445447522157, + "grad_norm": 0.16796875, + "learning_rate": 0.0019219082059011943, + "loss": 0.0869, + "step": 13691 + }, + { + "epoch": 0.11885313495542574, + "grad_norm": 0.51171875, + "learning_rate": 0.0019218960789938227, + "loss": 0.0923, + "step": 13692 + }, + { + "epoch": 0.1188618154356299, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019218839511876379, + "loss": 0.1436, + "step": 13693 + }, + { + "epoch": 0.11887049591583407, + "grad_norm": 0.7890625, + "learning_rate": 0.0019218718224826529, + "loss": 0.0957, + "step": 13694 + }, + { + "epoch": 0.11887917639603823, + "grad_norm": 0.07373046875, + "learning_rate": 0.0019218596928788812, + "loss": 0.0996, + "step": 13695 + }, + { + "epoch": 0.1188878568762424, + "grad_norm": 0.322265625, + "learning_rate": 0.0019218475623763362, + "loss": 0.1064, + "step": 13696 + }, + { + "epoch": 0.11889653735644656, + "grad_norm": 0.197265625, + "learning_rate": 0.0019218354309750313, + "loss": 0.1641, + "step": 13697 + }, + { + "epoch": 0.11890521783665073, + "grad_norm": 0.30078125, + "learning_rate": 0.001921823298674979, + "loss": 0.126, + "step": 13698 + }, + { + "epoch": 0.1189138983168549, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019218111654761936, + "loss": 0.0742, + "step": 13699 + }, + { + "epoch": 0.11892257879705906, + "grad_norm": 0.443359375, + "learning_rate": 0.0019217990313786875, + "loss": 0.1182, + "step": 13700 + }, + { + "epoch": 0.11893125927726322, + "grad_norm": 0.443359375, + "learning_rate": 0.0019217868963824747, + "loss": 0.1543, + "step": 13701 + }, + { + "epoch": 0.11893993975746739, + "grad_norm": 0.984375, + "learning_rate": 0.0019217747604875677, + "loss": 0.1211, + "step": 13702 + }, + { + "epoch": 0.11894862023767155, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019217626236939802, + "loss": 0.1279, + "step": 13703 + }, + { + "epoch": 0.11895730071787572, + "grad_norm": 0.08056640625, + "learning_rate": 0.001921750486001726, + "loss": 0.0957, + "step": 13704 + }, + { + "epoch": 0.11896598119807988, + "grad_norm": 0.58984375, + "learning_rate": 0.001921738347410817, + "loss": 0.1621, + "step": 13705 + }, + { + "epoch": 0.11897466167828405, + "grad_norm": 0.423828125, + "learning_rate": 0.0019217262079212678, + "loss": 0.1338, + "step": 13706 + }, + { + "epoch": 0.11898334215848821, + "grad_norm": 0.1201171875, + "learning_rate": 0.001921714067533091, + "loss": 0.1191, + "step": 13707 + }, + { + "epoch": 0.11899202263869237, + "grad_norm": 0.416015625, + "learning_rate": 0.0019217019262463002, + "loss": 0.1523, + "step": 13708 + }, + { + "epoch": 0.11900070311889653, + "grad_norm": 0.74609375, + "learning_rate": 0.0019216897840609085, + "loss": 0.1455, + "step": 13709 + }, + { + "epoch": 0.1190093835991007, + "grad_norm": 0.65625, + "learning_rate": 0.001921677640976929, + "loss": 0.1387, + "step": 13710 + }, + { + "epoch": 0.11901806407930486, + "grad_norm": 0.337890625, + "learning_rate": 0.0019216654969943753, + "loss": 0.1016, + "step": 13711 + }, + { + "epoch": 0.11902674455950903, + "grad_norm": 0.51953125, + "learning_rate": 0.0019216533521132604, + "loss": 0.0908, + "step": 13712 + }, + { + "epoch": 0.11903542503971319, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019216412063335982, + "loss": 0.0977, + "step": 13713 + }, + { + "epoch": 0.11904410551991736, + "grad_norm": 0.283203125, + "learning_rate": 0.001921629059655401, + "loss": 0.1196, + "step": 13714 + }, + { + "epoch": 0.11905278600012152, + "grad_norm": 0.17578125, + "learning_rate": 0.0019216169120786829, + "loss": 0.0947, + "step": 13715 + }, + { + "epoch": 0.11906146648032569, + "grad_norm": 0.5859375, + "learning_rate": 0.0019216047636034569, + "loss": 0.1621, + "step": 13716 + }, + { + "epoch": 0.11907014696052985, + "grad_norm": 0.94140625, + "learning_rate": 0.0019215926142297363, + "loss": 0.1113, + "step": 13717 + }, + { + "epoch": 0.11907882744073402, + "grad_norm": 0.4296875, + "learning_rate": 0.0019215804639575338, + "loss": 0.1318, + "step": 13718 + }, + { + "epoch": 0.11908750792093818, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019215683127868637, + "loss": 0.1172, + "step": 13719 + }, + { + "epoch": 0.11909618840114235, + "grad_norm": 0.408203125, + "learning_rate": 0.001921556160717739, + "loss": 0.1523, + "step": 13720 + }, + { + "epoch": 0.11910486888134651, + "grad_norm": 0.6484375, + "learning_rate": 0.0019215440077501728, + "loss": 0.1211, + "step": 13721 + }, + { + "epoch": 0.11911354936155068, + "grad_norm": 0.306640625, + "learning_rate": 0.0019215318538841782, + "loss": 0.1143, + "step": 13722 + }, + { + "epoch": 0.11912222984175484, + "grad_norm": 0.099609375, + "learning_rate": 0.0019215196991197688, + "loss": 0.0962, + "step": 13723 + }, + { + "epoch": 0.11913091032195901, + "grad_norm": 1.1796875, + "learning_rate": 0.0019215075434569576, + "loss": 0.1367, + "step": 13724 + }, + { + "epoch": 0.11913959080216317, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019214953868957583, + "loss": 0.1133, + "step": 13725 + }, + { + "epoch": 0.11914827128236734, + "grad_norm": 0.431640625, + "learning_rate": 0.0019214832294361841, + "loss": 0.0962, + "step": 13726 + }, + { + "epoch": 0.1191569517625715, + "grad_norm": 0.58203125, + "learning_rate": 0.0019214710710782482, + "loss": 0.1113, + "step": 13727 + }, + { + "epoch": 0.11916563224277567, + "grad_norm": 0.0693359375, + "learning_rate": 0.0019214589118219634, + "loss": 0.0732, + "step": 13728 + }, + { + "epoch": 0.11917431272297983, + "grad_norm": 0.404296875, + "learning_rate": 0.0019214467516673439, + "loss": 0.1621, + "step": 13729 + }, + { + "epoch": 0.119182993203184, + "grad_norm": 0.67578125, + "learning_rate": 0.0019214345906144023, + "loss": 0.1074, + "step": 13730 + }, + { + "epoch": 0.11919167368338816, + "grad_norm": 0.625, + "learning_rate": 0.0019214224286631526, + "loss": 0.0864, + "step": 13731 + }, + { + "epoch": 0.11920035416359233, + "grad_norm": 0.328125, + "learning_rate": 0.0019214102658136074, + "loss": 0.1602, + "step": 13732 + }, + { + "epoch": 0.1192090346437965, + "grad_norm": 0.439453125, + "learning_rate": 0.0019213981020657802, + "loss": 0.1377, + "step": 13733 + }, + { + "epoch": 0.11921771512400066, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019213859374196845, + "loss": 0.1719, + "step": 13734 + }, + { + "epoch": 0.11922639560420482, + "grad_norm": 0.154296875, + "learning_rate": 0.0019213737718753335, + "loss": 0.0811, + "step": 13735 + }, + { + "epoch": 0.11923507608440899, + "grad_norm": 0.3125, + "learning_rate": 0.0019213616054327403, + "loss": 0.1377, + "step": 13736 + }, + { + "epoch": 0.11924375656461315, + "grad_norm": 0.33984375, + "learning_rate": 0.0019213494380919188, + "loss": 0.1807, + "step": 13737 + }, + { + "epoch": 0.11925243704481732, + "grad_norm": 0.3203125, + "learning_rate": 0.0019213372698528815, + "loss": 0.0986, + "step": 13738 + }, + { + "epoch": 0.11926111752502148, + "grad_norm": 0.73828125, + "learning_rate": 0.0019213251007156422, + "loss": 0.1309, + "step": 13739 + }, + { + "epoch": 0.11926979800522565, + "grad_norm": 0.326171875, + "learning_rate": 0.0019213129306802142, + "loss": 0.0874, + "step": 13740 + }, + { + "epoch": 0.11927847848542981, + "grad_norm": 0.390625, + "learning_rate": 0.0019213007597466106, + "loss": 0.0996, + "step": 13741 + }, + { + "epoch": 0.11928715896563398, + "grad_norm": 0.087890625, + "learning_rate": 0.0019212885879148448, + "loss": 0.0942, + "step": 13742 + }, + { + "epoch": 0.11929583944583814, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019212764151849302, + "loss": 0.1289, + "step": 13743 + }, + { + "epoch": 0.11930451992604231, + "grad_norm": 0.068359375, + "learning_rate": 0.0019212642415568804, + "loss": 0.0796, + "step": 13744 + }, + { + "epoch": 0.11931320040624648, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019212520670307078, + "loss": 0.1016, + "step": 13745 + }, + { + "epoch": 0.11932188088645064, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019212398916064264, + "loss": 0.1377, + "step": 13746 + }, + { + "epoch": 0.1193305613666548, + "grad_norm": 0.107421875, + "learning_rate": 0.0019212277152840496, + "loss": 0.1162, + "step": 13747 + }, + { + "epoch": 0.11933924184685897, + "grad_norm": 0.953125, + "learning_rate": 0.0019212155380635902, + "loss": 0.1138, + "step": 13748 + }, + { + "epoch": 0.11934792232706314, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019212033599450625, + "loss": 0.1484, + "step": 13749 + }, + { + "epoch": 0.1193566028072673, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019211911809284786, + "loss": 0.1562, + "step": 13750 + }, + { + "epoch": 0.11936528328747147, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019211790010138523, + "loss": 0.104, + "step": 13751 + }, + { + "epoch": 0.11937396376767563, + "grad_norm": 0.306640625, + "learning_rate": 0.0019211668202011973, + "loss": 0.0796, + "step": 13752 + }, + { + "epoch": 0.1193826442478798, + "grad_norm": 0.1728515625, + "learning_rate": 0.0019211546384905263, + "loss": 0.1152, + "step": 13753 + }, + { + "epoch": 0.11939132472808396, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019211424558818533, + "loss": 0.1133, + "step": 13754 + }, + { + "epoch": 0.11940000520828813, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019211302723751908, + "loss": 0.1377, + "step": 13755 + }, + { + "epoch": 0.11940868568849229, + "grad_norm": 0.2099609375, + "learning_rate": 0.001921118087970553, + "loss": 0.1143, + "step": 13756 + }, + { + "epoch": 0.11941736616869646, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019211059026679523, + "loss": 0.0752, + "step": 13757 + }, + { + "epoch": 0.11942604664890062, + "grad_norm": 0.12109375, + "learning_rate": 0.001921093716467403, + "loss": 0.0972, + "step": 13758 + }, + { + "epoch": 0.11943472712910479, + "grad_norm": 0.09375, + "learning_rate": 0.0019210815293689177, + "loss": 0.1006, + "step": 13759 + }, + { + "epoch": 0.11944340760930895, + "grad_norm": 0.294921875, + "learning_rate": 0.00192106934137251, + "loss": 0.106, + "step": 13760 + }, + { + "epoch": 0.11945208808951312, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019210571524781932, + "loss": 0.127, + "step": 13761 + }, + { + "epoch": 0.11946076856971728, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019210449626859806, + "loss": 0.1523, + "step": 13762 + }, + { + "epoch": 0.11946944904992145, + "grad_norm": 0.515625, + "learning_rate": 0.0019210327719958859, + "loss": 0.0986, + "step": 13763 + }, + { + "epoch": 0.11947812953012561, + "grad_norm": 0.5390625, + "learning_rate": 0.0019210205804079217, + "loss": 0.0928, + "step": 13764 + }, + { + "epoch": 0.11948681001032978, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019210083879221019, + "loss": 0.1187, + "step": 13765 + }, + { + "epoch": 0.11949549049053394, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019209961945384395, + "loss": 0.1099, + "step": 13766 + }, + { + "epoch": 0.11950417097073811, + "grad_norm": 0.271484375, + "learning_rate": 0.0019209840002569483, + "loss": 0.1963, + "step": 13767 + }, + { + "epoch": 0.11951285145094227, + "grad_norm": 0.25390625, + "learning_rate": 0.001920971805077641, + "loss": 0.0986, + "step": 13768 + }, + { + "epoch": 0.11952153193114642, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019209596090005315, + "loss": 0.0933, + "step": 13769 + }, + { + "epoch": 0.11953021241135059, + "grad_norm": 0.3125, + "learning_rate": 0.0019209474120256327, + "loss": 0.1445, + "step": 13770 + }, + { + "epoch": 0.11953889289155475, + "grad_norm": 0.41015625, + "learning_rate": 0.001920935214152958, + "loss": 0.125, + "step": 13771 + }, + { + "epoch": 0.11954757337175892, + "grad_norm": 0.4375, + "learning_rate": 0.0019209230153825216, + "loss": 0.0986, + "step": 13772 + }, + { + "epoch": 0.11955625385196308, + "grad_norm": 0.34765625, + "learning_rate": 0.0019209108157143356, + "loss": 0.0933, + "step": 13773 + }, + { + "epoch": 0.11956493433216725, + "grad_norm": 0.11279296875, + "learning_rate": 0.001920898615148414, + "loss": 0.1172, + "step": 13774 + }, + { + "epoch": 0.11957361481237141, + "grad_norm": 0.162109375, + "learning_rate": 0.0019208864136847697, + "loss": 0.1084, + "step": 13775 + }, + { + "epoch": 0.11958229529257558, + "grad_norm": 0.11328125, + "learning_rate": 0.0019208742113234168, + "loss": 0.1367, + "step": 13776 + }, + { + "epoch": 0.11959097577277975, + "grad_norm": 0.07275390625, + "learning_rate": 0.001920862008064368, + "loss": 0.1182, + "step": 13777 + }, + { + "epoch": 0.11959965625298391, + "grad_norm": 1.6015625, + "learning_rate": 0.0019208498039076367, + "loss": 0.1719, + "step": 13778 + }, + { + "epoch": 0.11960833673318808, + "grad_norm": 0.2421875, + "learning_rate": 0.0019208375988532366, + "loss": 0.1006, + "step": 13779 + }, + { + "epoch": 0.11961701721339224, + "grad_norm": 0.283203125, + "learning_rate": 0.001920825392901181, + "loss": 0.1113, + "step": 13780 + }, + { + "epoch": 0.1196256976935964, + "grad_norm": 0.64453125, + "learning_rate": 0.0019208131860514825, + "loss": 0.1006, + "step": 13781 + }, + { + "epoch": 0.11963437817380057, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019208009783041556, + "loss": 0.1592, + "step": 13782 + }, + { + "epoch": 0.11964305865400474, + "grad_norm": 0.373046875, + "learning_rate": 0.0019207887696592128, + "loss": 0.1621, + "step": 13783 + }, + { + "epoch": 0.1196517391342089, + "grad_norm": 0.74609375, + "learning_rate": 0.0019207765601166678, + "loss": 0.126, + "step": 13784 + }, + { + "epoch": 0.11966041961441307, + "grad_norm": 0.2109375, + "learning_rate": 0.001920764349676534, + "loss": 0.1504, + "step": 13785 + }, + { + "epoch": 0.11966910009461723, + "grad_norm": 0.63671875, + "learning_rate": 0.0019207521383388245, + "loss": 0.1094, + "step": 13786 + }, + { + "epoch": 0.1196777805748214, + "grad_norm": 0.3046875, + "learning_rate": 0.0019207399261035524, + "loss": 0.1201, + "step": 13787 + }, + { + "epoch": 0.11968646105502556, + "grad_norm": 0.3984375, + "learning_rate": 0.0019207277129707323, + "loss": 0.1562, + "step": 13788 + }, + { + "epoch": 0.11969514153522973, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019207154989403761, + "loss": 0.1133, + "step": 13789 + }, + { + "epoch": 0.11970382201543389, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019207032840124981, + "loss": 0.1138, + "step": 13790 + }, + { + "epoch": 0.11971250249563806, + "grad_norm": 0.1640625, + "learning_rate": 0.0019206910681871113, + "loss": 0.1025, + "step": 13791 + }, + { + "epoch": 0.11972118297584222, + "grad_norm": 0.09375, + "learning_rate": 0.0019206788514642286, + "loss": 0.1016, + "step": 13792 + }, + { + "epoch": 0.11972986345604639, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019206666338438642, + "loss": 0.1162, + "step": 13793 + }, + { + "epoch": 0.11973854393625055, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019206544153260313, + "loss": 0.1367, + "step": 13794 + }, + { + "epoch": 0.11974722441645472, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019206421959107425, + "loss": 0.1138, + "step": 13795 + }, + { + "epoch": 0.11975590489665888, + "grad_norm": 0.115234375, + "learning_rate": 0.0019206299755980122, + "loss": 0.1108, + "step": 13796 + }, + { + "epoch": 0.11976458537686305, + "grad_norm": 0.140625, + "learning_rate": 0.001920617754387853, + "loss": 0.125, + "step": 13797 + }, + { + "epoch": 0.11977326585706721, + "grad_norm": 0.234375, + "learning_rate": 0.0019206055322802788, + "loss": 0.0977, + "step": 13798 + }, + { + "epoch": 0.11978194633727138, + "grad_norm": 0.134765625, + "learning_rate": 0.0019205933092753028, + "loss": 0.1221, + "step": 13799 + }, + { + "epoch": 0.11979062681747554, + "grad_norm": 0.59375, + "learning_rate": 0.0019205810853729379, + "loss": 0.1006, + "step": 13800 + }, + { + "epoch": 0.11979930729767971, + "grad_norm": 0.26171875, + "learning_rate": 0.001920568860573198, + "loss": 0.1025, + "step": 13801 + }, + { + "epoch": 0.11980798777788387, + "grad_norm": 0.380859375, + "learning_rate": 0.0019205566348760965, + "loss": 0.1025, + "step": 13802 + }, + { + "epoch": 0.11981666825808804, + "grad_norm": 0.177734375, + "learning_rate": 0.0019205444082816463, + "loss": 0.1172, + "step": 13803 + }, + { + "epoch": 0.1198253487382922, + "grad_norm": 0.66015625, + "learning_rate": 0.0019205321807898613, + "loss": 0.1309, + "step": 13804 + }, + { + "epoch": 0.11983402921849637, + "grad_norm": 0.79296875, + "learning_rate": 0.001920519952400755, + "loss": 0.0977, + "step": 13805 + }, + { + "epoch": 0.11984270969870053, + "grad_norm": 0.380859375, + "learning_rate": 0.0019205077231143398, + "loss": 0.1523, + "step": 13806 + }, + { + "epoch": 0.1198513901789047, + "grad_norm": 0.291015625, + "learning_rate": 0.0019204954929306298, + "loss": 0.0977, + "step": 13807 + }, + { + "epoch": 0.11986007065910886, + "grad_norm": 0.3515625, + "learning_rate": 0.0019204832618496383, + "loss": 0.1143, + "step": 13808 + }, + { + "epoch": 0.11986875113931303, + "grad_norm": 0.50390625, + "learning_rate": 0.0019204710298713786, + "loss": 0.1133, + "step": 13809 + }, + { + "epoch": 0.1198774316195172, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019204587969958642, + "loss": 0.125, + "step": 13810 + }, + { + "epoch": 0.11988611209972136, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019204465632231083, + "loss": 0.127, + "step": 13811 + }, + { + "epoch": 0.11989479257992552, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019204343285531245, + "loss": 0.127, + "step": 13812 + }, + { + "epoch": 0.11990347306012969, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019204220929859258, + "loss": 0.1045, + "step": 13813 + }, + { + "epoch": 0.11991215354033385, + "grad_norm": 0.53515625, + "learning_rate": 0.001920409856521526, + "loss": 0.1177, + "step": 13814 + }, + { + "epoch": 0.11992083402053802, + "grad_norm": 0.1357421875, + "learning_rate": 0.001920397619159938, + "loss": 0.1206, + "step": 13815 + }, + { + "epoch": 0.11992951450074218, + "grad_norm": 0.4375, + "learning_rate": 0.0019203853809011762, + "loss": 0.1426, + "step": 13816 + }, + { + "epoch": 0.11993819498094635, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019203731417452529, + "loss": 0.1338, + "step": 13817 + }, + { + "epoch": 0.11994687546115052, + "grad_norm": 0.14453125, + "learning_rate": 0.0019203609016921817, + "loss": 0.0933, + "step": 13818 + }, + { + "epoch": 0.11995555594135468, + "grad_norm": 0.107421875, + "learning_rate": 0.0019203486607419762, + "loss": 0.0723, + "step": 13819 + }, + { + "epoch": 0.11996423642155885, + "grad_norm": 0.51953125, + "learning_rate": 0.00192033641889465, + "loss": 0.0967, + "step": 13820 + }, + { + "epoch": 0.11997291690176301, + "grad_norm": 0.84765625, + "learning_rate": 0.0019203241761502159, + "loss": 0.127, + "step": 13821 + }, + { + "epoch": 0.11998159738196718, + "grad_norm": 0.1005859375, + "learning_rate": 0.0019203119325086875, + "loss": 0.0879, + "step": 13822 + }, + { + "epoch": 0.11999027786217134, + "grad_norm": 0.392578125, + "learning_rate": 0.0019202996879700788, + "loss": 0.1152, + "step": 13823 + }, + { + "epoch": 0.1199989583423755, + "grad_norm": 0.45703125, + "learning_rate": 0.001920287442534402, + "loss": 0.123, + "step": 13824 + }, + { + "epoch": 0.12000763882257967, + "grad_norm": 0.66796875, + "learning_rate": 0.0019202751962016718, + "loss": 0.1426, + "step": 13825 + }, + { + "epoch": 0.12001631930278384, + "grad_norm": 0.0947265625, + "learning_rate": 0.001920262948971901, + "loss": 0.1206, + "step": 13826 + }, + { + "epoch": 0.120024999782988, + "grad_norm": 0.375, + "learning_rate": 0.0019202507008451025, + "loss": 0.1074, + "step": 13827 + }, + { + "epoch": 0.12003368026319217, + "grad_norm": 0.50390625, + "learning_rate": 0.0019202384518212903, + "loss": 0.0913, + "step": 13828 + }, + { + "epoch": 0.12004236074339633, + "grad_norm": 0.337890625, + "learning_rate": 0.0019202262019004776, + "loss": 0.1152, + "step": 13829 + }, + { + "epoch": 0.1200510412236005, + "grad_norm": 0.77734375, + "learning_rate": 0.0019202139510826781, + "loss": 0.127, + "step": 13830 + }, + { + "epoch": 0.12005972170380465, + "grad_norm": 0.443359375, + "learning_rate": 0.0019202016993679048, + "loss": 0.123, + "step": 13831 + }, + { + "epoch": 0.12006840218400881, + "grad_norm": 0.71484375, + "learning_rate": 0.0019201894467561712, + "loss": 0.1094, + "step": 13832 + }, + { + "epoch": 0.12007708266421298, + "grad_norm": 0.14453125, + "learning_rate": 0.001920177193247491, + "loss": 0.1338, + "step": 13833 + }, + { + "epoch": 0.12008576314441714, + "grad_norm": 0.09765625, + "learning_rate": 0.0019201649388418769, + "loss": 0.1699, + "step": 13834 + }, + { + "epoch": 0.12009444362462131, + "grad_norm": 0.3359375, + "learning_rate": 0.0019201526835393432, + "loss": 0.1055, + "step": 13835 + }, + { + "epoch": 0.12010312410482547, + "grad_norm": 0.388671875, + "learning_rate": 0.0019201404273399027, + "loss": 0.0947, + "step": 13836 + }, + { + "epoch": 0.12011180458502964, + "grad_norm": 0.255859375, + "learning_rate": 0.001920128170243569, + "loss": 0.1128, + "step": 13837 + }, + { + "epoch": 0.1201204850652338, + "grad_norm": 0.546875, + "learning_rate": 0.0019201159122503551, + "loss": 0.1084, + "step": 13838 + }, + { + "epoch": 0.12012916554543797, + "grad_norm": 0.232421875, + "learning_rate": 0.001920103653360275, + "loss": 0.1055, + "step": 13839 + }, + { + "epoch": 0.12013784602564213, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019200913935733417, + "loss": 0.1113, + "step": 13840 + }, + { + "epoch": 0.1201465265058463, + "grad_norm": 0.404296875, + "learning_rate": 0.001920079132889569, + "loss": 0.1157, + "step": 13841 + }, + { + "epoch": 0.12015520698605046, + "grad_norm": 0.41015625, + "learning_rate": 0.00192006687130897, + "loss": 0.1357, + "step": 13842 + }, + { + "epoch": 0.12016388746625463, + "grad_norm": 0.494140625, + "learning_rate": 0.0019200546088315583, + "loss": 0.1318, + "step": 13843 + }, + { + "epoch": 0.1201725679464588, + "grad_norm": 0.62890625, + "learning_rate": 0.0019200423454573472, + "loss": 0.167, + "step": 13844 + }, + { + "epoch": 0.12018124842666296, + "grad_norm": 0.54296875, + "learning_rate": 0.0019200300811863503, + "loss": 0.123, + "step": 13845 + }, + { + "epoch": 0.12018992890686712, + "grad_norm": 0.1640625, + "learning_rate": 0.0019200178160185804, + "loss": 0.124, + "step": 13846 + }, + { + "epoch": 0.12019860938707129, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019200055499540515, + "loss": 0.0708, + "step": 13847 + }, + { + "epoch": 0.12020728986727545, + "grad_norm": 0.3125, + "learning_rate": 0.0019199932829927766, + "loss": 0.1367, + "step": 13848 + }, + { + "epoch": 0.12021597034747962, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019199810151347696, + "loss": 0.1055, + "step": 13849 + }, + { + "epoch": 0.12022465082768378, + "grad_norm": 0.212890625, + "learning_rate": 0.001919968746380044, + "loss": 0.0898, + "step": 13850 + }, + { + "epoch": 0.12023333130788795, + "grad_norm": 0.5078125, + "learning_rate": 0.0019199564767286123, + "loss": 0.1152, + "step": 13851 + }, + { + "epoch": 0.12024201178809212, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019199442061804887, + "loss": 0.0947, + "step": 13852 + }, + { + "epoch": 0.12025069226829628, + "grad_norm": 0.33203125, + "learning_rate": 0.0019199319347356868, + "loss": 0.1143, + "step": 13853 + }, + { + "epoch": 0.12025937274850045, + "grad_norm": 0.3671875, + "learning_rate": 0.0019199196623942193, + "loss": 0.1309, + "step": 13854 + }, + { + "epoch": 0.12026805322870461, + "grad_norm": 0.94140625, + "learning_rate": 0.0019199073891561004, + "loss": 0.1182, + "step": 13855 + }, + { + "epoch": 0.12027673370890878, + "grad_norm": 1.015625, + "learning_rate": 0.0019198951150213423, + "loss": 0.1816, + "step": 13856 + }, + { + "epoch": 0.12028541418911294, + "grad_norm": 0.1064453125, + "learning_rate": 0.00191988283998996, + "loss": 0.0967, + "step": 13857 + }, + { + "epoch": 0.1202940946693171, + "grad_norm": 0.166015625, + "learning_rate": 0.0019198705640619657, + "loss": 0.0957, + "step": 13858 + }, + { + "epoch": 0.12030277514952127, + "grad_norm": 0.71875, + "learning_rate": 0.0019198582872373736, + "loss": 0.1289, + "step": 13859 + }, + { + "epoch": 0.12031145562972544, + "grad_norm": 0.71484375, + "learning_rate": 0.0019198460095161962, + "loss": 0.1309, + "step": 13860 + }, + { + "epoch": 0.1203201361099296, + "grad_norm": 0.90234375, + "learning_rate": 0.0019198337308984482, + "loss": 0.168, + "step": 13861 + }, + { + "epoch": 0.12032881659013377, + "grad_norm": 0.119140625, + "learning_rate": 0.001919821451384142, + "loss": 0.1162, + "step": 13862 + }, + { + "epoch": 0.12033749707033793, + "grad_norm": 1.1015625, + "learning_rate": 0.0019198091709732915, + "loss": 0.0962, + "step": 13863 + }, + { + "epoch": 0.1203461775505421, + "grad_norm": 0.390625, + "learning_rate": 0.0019197968896659098, + "loss": 0.1553, + "step": 13864 + }, + { + "epoch": 0.12035485803074626, + "grad_norm": 0.9140625, + "learning_rate": 0.001919784607462011, + "loss": 0.1592, + "step": 13865 + }, + { + "epoch": 0.12036353851095043, + "grad_norm": 0.326171875, + "learning_rate": 0.0019197723243616076, + "loss": 0.1162, + "step": 13866 + }, + { + "epoch": 0.12037221899115459, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019197600403647137, + "loss": 0.1758, + "step": 13867 + }, + { + "epoch": 0.12038089947135876, + "grad_norm": 0.07763671875, + "learning_rate": 0.0019197477554713425, + "loss": 0.105, + "step": 13868 + }, + { + "epoch": 0.12038957995156292, + "grad_norm": 0.43359375, + "learning_rate": 0.0019197354696815075, + "loss": 0.1396, + "step": 13869 + }, + { + "epoch": 0.12039826043176709, + "grad_norm": 0.1240234375, + "learning_rate": 0.001919723182995222, + "loss": 0.1299, + "step": 13870 + }, + { + "epoch": 0.12040694091197125, + "grad_norm": 0.291015625, + "learning_rate": 0.0019197108954124998, + "loss": 0.1025, + "step": 13871 + }, + { + "epoch": 0.12041562139217542, + "grad_norm": 0.40625, + "learning_rate": 0.001919698606933354, + "loss": 0.1914, + "step": 13872 + }, + { + "epoch": 0.12042430187237958, + "grad_norm": 0.6796875, + "learning_rate": 0.0019196863175577979, + "loss": 0.1035, + "step": 13873 + }, + { + "epoch": 0.12043298235258375, + "grad_norm": 1.3203125, + "learning_rate": 0.0019196740272858454, + "loss": 0.6211, + "step": 13874 + }, + { + "epoch": 0.12044166283278791, + "grad_norm": 0.27734375, + "learning_rate": 0.0019196617361175094, + "loss": 0.1445, + "step": 13875 + }, + { + "epoch": 0.12045034331299208, + "grad_norm": 0.50390625, + "learning_rate": 0.001919649444052804, + "loss": 0.1445, + "step": 13876 + }, + { + "epoch": 0.12045902379319624, + "grad_norm": 0.0859375, + "learning_rate": 0.001919637151091742, + "loss": 0.0913, + "step": 13877 + }, + { + "epoch": 0.12046770427340041, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019196248572343373, + "loss": 0.1279, + "step": 13878 + }, + { + "epoch": 0.12047638475360457, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019196125624806031, + "loss": 0.1074, + "step": 13879 + }, + { + "epoch": 0.12048506523380874, + "grad_norm": 0.1103515625, + "learning_rate": 0.001919600266830553, + "loss": 0.1533, + "step": 13880 + }, + { + "epoch": 0.1204937457140129, + "grad_norm": 0.353515625, + "learning_rate": 0.0019195879702842004, + "loss": 0.0908, + "step": 13881 + }, + { + "epoch": 0.12050242619421707, + "grad_norm": 0.71875, + "learning_rate": 0.0019195756728415583, + "loss": 0.1069, + "step": 13882 + }, + { + "epoch": 0.12051110667442123, + "grad_norm": 0.140625, + "learning_rate": 0.001919563374502641, + "loss": 0.0947, + "step": 13883 + }, + { + "epoch": 0.1205197871546254, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019195510752674611, + "loss": 0.1182, + "step": 13884 + }, + { + "epoch": 0.12052846763482956, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019195387751360332, + "loss": 0.1602, + "step": 13885 + }, + { + "epoch": 0.12053714811503373, + "grad_norm": 1.0546875, + "learning_rate": 0.0019195264741083693, + "loss": 0.1289, + "step": 13886 + }, + { + "epoch": 0.1205458285952379, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019195141721844837, + "loss": 0.1445, + "step": 13887 + }, + { + "epoch": 0.12055450907544206, + "grad_norm": 0.453125, + "learning_rate": 0.00191950186936439, + "loss": 0.1045, + "step": 13888 + }, + { + "epoch": 0.12056318955564622, + "grad_norm": 0.259765625, + "learning_rate": 0.0019194895656481008, + "loss": 0.1328, + "step": 13889 + }, + { + "epoch": 0.12057187003585039, + "grad_norm": 0.46484375, + "learning_rate": 0.0019194772610356308, + "loss": 0.1621, + "step": 13890 + }, + { + "epoch": 0.12058055051605455, + "grad_norm": 0.267578125, + "learning_rate": 0.0019194649555269922, + "loss": 0.0986, + "step": 13891 + }, + { + "epoch": 0.1205892309962587, + "grad_norm": 0.365234375, + "learning_rate": 0.0019194526491221994, + "loss": 0.1211, + "step": 13892 + }, + { + "epoch": 0.12059791147646287, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019194403418212656, + "loss": 0.1084, + "step": 13893 + }, + { + "epoch": 0.12060659195666704, + "grad_norm": 0.50390625, + "learning_rate": 0.001919428033624204, + "loss": 0.127, + "step": 13894 + }, + { + "epoch": 0.1206152724368712, + "grad_norm": 0.65625, + "learning_rate": 0.001919415724531028, + "loss": 0.1211, + "step": 13895 + }, + { + "epoch": 0.12062395291707537, + "grad_norm": 0.75, + "learning_rate": 0.001919403414541752, + "loss": 0.0874, + "step": 13896 + }, + { + "epoch": 0.12063263339727953, + "grad_norm": 0.71875, + "learning_rate": 0.0019193911036563878, + "loss": 0.1211, + "step": 13897 + }, + { + "epoch": 0.1206413138774837, + "grad_norm": 0.1796875, + "learning_rate": 0.0019193787918749504, + "loss": 0.1016, + "step": 13898 + }, + { + "epoch": 0.12064999435768786, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019193664791974524, + "loss": 0.0957, + "step": 13899 + }, + { + "epoch": 0.12065867483789203, + "grad_norm": 1.015625, + "learning_rate": 0.0019193541656239075, + "loss": 0.1348, + "step": 13900 + }, + { + "epoch": 0.12066735531809619, + "grad_norm": 0.16015625, + "learning_rate": 0.0019193418511543293, + "loss": 0.0835, + "step": 13901 + }, + { + "epoch": 0.12067603579830036, + "grad_norm": 0.251953125, + "learning_rate": 0.001919329535788731, + "loss": 0.1201, + "step": 13902 + }, + { + "epoch": 0.12068471627850452, + "grad_norm": 0.404296875, + "learning_rate": 0.0019193172195271265, + "loss": 0.3574, + "step": 13903 + }, + { + "epoch": 0.12069339675870869, + "grad_norm": 0.314453125, + "learning_rate": 0.0019193049023695289, + "loss": 0.0928, + "step": 13904 + }, + { + "epoch": 0.12070207723891285, + "grad_norm": 1.0625, + "learning_rate": 0.0019192925843159517, + "loss": 0.1543, + "step": 13905 + }, + { + "epoch": 0.12071075771911702, + "grad_norm": 0.61328125, + "learning_rate": 0.0019192802653664085, + "loss": 0.0928, + "step": 13906 + }, + { + "epoch": 0.12071943819932118, + "grad_norm": 0.291015625, + "learning_rate": 0.0019192679455209128, + "loss": 0.1738, + "step": 13907 + }, + { + "epoch": 0.12072811867952535, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019192556247794776, + "loss": 0.1074, + "step": 13908 + }, + { + "epoch": 0.12073679915972951, + "grad_norm": 0.08154296875, + "learning_rate": 0.0019192433031421173, + "loss": 0.1011, + "step": 13909 + }, + { + "epoch": 0.12074547963993368, + "grad_norm": 0.64453125, + "learning_rate": 0.0019192309806088444, + "loss": 0.124, + "step": 13910 + }, + { + "epoch": 0.12075416012013784, + "grad_norm": 0.4140625, + "learning_rate": 0.001919218657179673, + "loss": 0.0903, + "step": 13911 + }, + { + "epoch": 0.12076284060034201, + "grad_norm": 0.625, + "learning_rate": 0.0019192063328546164, + "loss": 0.1387, + "step": 13912 + }, + { + "epoch": 0.12077152108054617, + "grad_norm": 0.1982421875, + "learning_rate": 0.001919194007633688, + "loss": 0.126, + "step": 13913 + }, + { + "epoch": 0.12078020156075034, + "grad_norm": 0.5078125, + "learning_rate": 0.0019191816815169013, + "loss": 0.1797, + "step": 13914 + }, + { + "epoch": 0.1207888820409545, + "grad_norm": 0.26171875, + "learning_rate": 0.0019191693545042702, + "loss": 0.1147, + "step": 13915 + }, + { + "epoch": 0.12079756252115867, + "grad_norm": 0.291015625, + "learning_rate": 0.001919157026595807, + "loss": 0.0918, + "step": 13916 + }, + { + "epoch": 0.12080624300136283, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019191446977915268, + "loss": 0.0938, + "step": 13917 + }, + { + "epoch": 0.120814923481567, + "grad_norm": 0.359375, + "learning_rate": 0.001919132368091442, + "loss": 0.1177, + "step": 13918 + }, + { + "epoch": 0.12082360396177116, + "grad_norm": 1.2265625, + "learning_rate": 0.0019191200374955662, + "loss": 0.1221, + "step": 13919 + }, + { + "epoch": 0.12083228444197533, + "grad_norm": 0.2734375, + "learning_rate": 0.001919107706003913, + "loss": 0.1396, + "step": 13920 + }, + { + "epoch": 0.1208409649221795, + "grad_norm": 0.6796875, + "learning_rate": 0.0019190953736164963, + "loss": 0.1514, + "step": 13921 + }, + { + "epoch": 0.12084964540238366, + "grad_norm": 0.55859375, + "learning_rate": 0.0019190830403333287, + "loss": 0.1094, + "step": 13922 + }, + { + "epoch": 0.12085832588258782, + "grad_norm": 0.79296875, + "learning_rate": 0.0019190707061544246, + "loss": 0.1348, + "step": 13923 + }, + { + "epoch": 0.12086700636279199, + "grad_norm": 0.55078125, + "learning_rate": 0.0019190583710797968, + "loss": 0.085, + "step": 13924 + }, + { + "epoch": 0.12087568684299616, + "grad_norm": 0.146484375, + "learning_rate": 0.0019190460351094594, + "loss": 0.0913, + "step": 13925 + }, + { + "epoch": 0.12088436732320032, + "grad_norm": 0.11083984375, + "learning_rate": 0.0019190336982434254, + "loss": 0.1328, + "step": 13926 + }, + { + "epoch": 0.12089304780340449, + "grad_norm": 0.271484375, + "learning_rate": 0.0019190213604817084, + "loss": 0.1396, + "step": 13927 + }, + { + "epoch": 0.12090172828360865, + "grad_norm": 0.166015625, + "learning_rate": 0.0019190090218243216, + "loss": 0.1133, + "step": 13928 + }, + { + "epoch": 0.12091040876381282, + "grad_norm": 0.388671875, + "learning_rate": 0.0019189966822712795, + "loss": 0.1055, + "step": 13929 + }, + { + "epoch": 0.12091908924401698, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019189843418225946, + "loss": 0.1738, + "step": 13930 + }, + { + "epoch": 0.12092776972422115, + "grad_norm": 0.20703125, + "learning_rate": 0.0019189720004782807, + "loss": 0.1064, + "step": 13931 + }, + { + "epoch": 0.12093645020442531, + "grad_norm": 0.22265625, + "learning_rate": 0.0019189596582383512, + "loss": 0.1084, + "step": 13932 + }, + { + "epoch": 0.12094513068462948, + "grad_norm": 0.734375, + "learning_rate": 0.00191894731510282, + "loss": 0.1152, + "step": 13933 + }, + { + "epoch": 0.12095381116483364, + "grad_norm": 1.0625, + "learning_rate": 0.0019189349710717004, + "loss": 0.1182, + "step": 13934 + }, + { + "epoch": 0.1209624916450378, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019189226261450054, + "loss": 0.084, + "step": 13935 + }, + { + "epoch": 0.12097117212524197, + "grad_norm": 0.64453125, + "learning_rate": 0.0019189102803227494, + "loss": 0.1006, + "step": 13936 + }, + { + "epoch": 0.12097985260544614, + "grad_norm": 0.15625, + "learning_rate": 0.0019188979336049452, + "loss": 0.1191, + "step": 13937 + }, + { + "epoch": 0.1209885330856503, + "grad_norm": 0.62109375, + "learning_rate": 0.0019188855859916063, + "loss": 0.0752, + "step": 13938 + }, + { + "epoch": 0.12099721356585447, + "grad_norm": 0.34375, + "learning_rate": 0.001918873237482747, + "loss": 0.1514, + "step": 13939 + }, + { + "epoch": 0.12100589404605863, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019188608880783798, + "loss": 0.0972, + "step": 13940 + }, + { + "epoch": 0.1210145745262628, + "grad_norm": 0.40625, + "learning_rate": 0.0019188485377785185, + "loss": 0.0801, + "step": 13941 + }, + { + "epoch": 0.12102325500646696, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019188361865831772, + "loss": 0.1182, + "step": 13942 + }, + { + "epoch": 0.12103193548667113, + "grad_norm": 0.474609375, + "learning_rate": 0.0019188238344923685, + "loss": 0.1309, + "step": 13943 + }, + { + "epoch": 0.12104061596687529, + "grad_norm": 1.3984375, + "learning_rate": 0.0019188114815061066, + "loss": 0.1855, + "step": 13944 + }, + { + "epoch": 0.12104929644707946, + "grad_norm": 0.546875, + "learning_rate": 0.0019187991276244049, + "loss": 0.1523, + "step": 13945 + }, + { + "epoch": 0.12105797692728362, + "grad_norm": 0.234375, + "learning_rate": 0.0019187867728472766, + "loss": 0.1348, + "step": 13946 + }, + { + "epoch": 0.12106665740748779, + "grad_norm": 0.203125, + "learning_rate": 0.0019187744171747351, + "loss": 0.1133, + "step": 13947 + }, + { + "epoch": 0.12107533788769195, + "grad_norm": 0.193359375, + "learning_rate": 0.0019187620606067946, + "loss": 0.0737, + "step": 13948 + }, + { + "epoch": 0.12108401836789612, + "grad_norm": 0.232421875, + "learning_rate": 0.001918749703143468, + "loss": 0.1133, + "step": 13949 + }, + { + "epoch": 0.12109269884810028, + "grad_norm": 1.4765625, + "learning_rate": 0.001918737344784769, + "loss": 0.2305, + "step": 13950 + }, + { + "epoch": 0.12110137932830445, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019187249855307114, + "loss": 0.1172, + "step": 13951 + }, + { + "epoch": 0.12111005980850861, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019187126253813083, + "loss": 0.1445, + "step": 13952 + }, + { + "epoch": 0.12111874028871278, + "grad_norm": 0.435546875, + "learning_rate": 0.0019187002643365736, + "loss": 0.1309, + "step": 13953 + }, + { + "epoch": 0.12112742076891693, + "grad_norm": 0.076171875, + "learning_rate": 0.0019186879023965204, + "loss": 0.1367, + "step": 13954 + }, + { + "epoch": 0.1211361012491211, + "grad_norm": 0.228515625, + "learning_rate": 0.0019186755395611624, + "loss": 0.0942, + "step": 13955 + }, + { + "epoch": 0.12114478172932526, + "grad_norm": 0.421875, + "learning_rate": 0.001918663175830513, + "loss": 0.0938, + "step": 13956 + }, + { + "epoch": 0.12115346220952943, + "grad_norm": 0.19921875, + "learning_rate": 0.0019186508112045862, + "loss": 0.1309, + "step": 13957 + }, + { + "epoch": 0.12116214268973359, + "grad_norm": 0.9765625, + "learning_rate": 0.0019186384456833948, + "loss": 0.1152, + "step": 13958 + }, + { + "epoch": 0.12117082316993776, + "grad_norm": 0.1259765625, + "learning_rate": 0.001918626079266953, + "loss": 0.0801, + "step": 13959 + }, + { + "epoch": 0.12117950365014192, + "grad_norm": 0.051513671875, + "learning_rate": 0.0019186137119552738, + "loss": 0.1006, + "step": 13960 + }, + { + "epoch": 0.12118818413034609, + "grad_norm": 0.12890625, + "learning_rate": 0.001918601343748371, + "loss": 0.1338, + "step": 13961 + }, + { + "epoch": 0.12119686461055025, + "grad_norm": 0.28515625, + "learning_rate": 0.001918588974646258, + "loss": 0.1357, + "step": 13962 + }, + { + "epoch": 0.12120554509075442, + "grad_norm": 0.68359375, + "learning_rate": 0.0019185766046489485, + "loss": 0.0791, + "step": 13963 + }, + { + "epoch": 0.12121422557095858, + "grad_norm": 0.2109375, + "learning_rate": 0.001918564233756456, + "loss": 0.1064, + "step": 13964 + }, + { + "epoch": 0.12122290605116275, + "grad_norm": 0.359375, + "learning_rate": 0.0019185518619687938, + "loss": 0.1309, + "step": 13965 + }, + { + "epoch": 0.12123158653136691, + "grad_norm": 0.265625, + "learning_rate": 0.0019185394892859758, + "loss": 0.123, + "step": 13966 + }, + { + "epoch": 0.12124026701157108, + "grad_norm": 0.81640625, + "learning_rate": 0.0019185271157080152, + "loss": 0.0991, + "step": 13967 + }, + { + "epoch": 0.12124894749177524, + "grad_norm": 0.373046875, + "learning_rate": 0.0019185147412349253, + "loss": 0.1504, + "step": 13968 + }, + { + "epoch": 0.1212576279719794, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019185023658667203, + "loss": 0.0938, + "step": 13969 + }, + { + "epoch": 0.12126630845218357, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019184899896034133, + "loss": 0.1279, + "step": 13970 + }, + { + "epoch": 0.12127498893238774, + "grad_norm": 0.109375, + "learning_rate": 0.0019184776124450183, + "loss": 0.1367, + "step": 13971 + }, + { + "epoch": 0.1212836694125919, + "grad_norm": 0.77734375, + "learning_rate": 0.0019184652343915481, + "loss": 0.1143, + "step": 13972 + }, + { + "epoch": 0.12129234989279607, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019184528554430165, + "loss": 0.1055, + "step": 13973 + }, + { + "epoch": 0.12130103037300023, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019184404755994376, + "loss": 0.1348, + "step": 13974 + }, + { + "epoch": 0.1213097108532044, + "grad_norm": 0.400390625, + "learning_rate": 0.0019184280948608243, + "loss": 0.1162, + "step": 13975 + }, + { + "epoch": 0.12131839133340856, + "grad_norm": 0.267578125, + "learning_rate": 0.0019184157132271903, + "loss": 0.1104, + "step": 13976 + }, + { + "epoch": 0.12132707181361273, + "grad_norm": 0.2001953125, + "learning_rate": 0.001918403330698549, + "loss": 0.1436, + "step": 13977 + }, + { + "epoch": 0.12133575229381689, + "grad_norm": 0.91015625, + "learning_rate": 0.0019183909472749144, + "loss": 0.0806, + "step": 13978 + }, + { + "epoch": 0.12134443277402106, + "grad_norm": 0.3046875, + "learning_rate": 0.0019183785629562997, + "loss": 0.1123, + "step": 13979 + }, + { + "epoch": 0.12135311325422522, + "grad_norm": 0.1875, + "learning_rate": 0.0019183661777427187, + "loss": 0.124, + "step": 13980 + }, + { + "epoch": 0.12136179373442939, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019183537916341842, + "loss": 0.0913, + "step": 13981 + }, + { + "epoch": 0.12137047421463355, + "grad_norm": 0.123046875, + "learning_rate": 0.0019183414046307107, + "loss": 0.1133, + "step": 13982 + }, + { + "epoch": 0.12137915469483772, + "grad_norm": 0.205078125, + "learning_rate": 0.0019183290167323114, + "loss": 0.1201, + "step": 13983 + }, + { + "epoch": 0.12138783517504188, + "grad_norm": 0.177734375, + "learning_rate": 0.0019183166279389992, + "loss": 0.1177, + "step": 13984 + }, + { + "epoch": 0.12139651565524605, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019183042382507887, + "loss": 0.1455, + "step": 13985 + }, + { + "epoch": 0.12140519613545021, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019182918476676931, + "loss": 0.1025, + "step": 13986 + }, + { + "epoch": 0.12141387661565438, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019182794561897257, + "loss": 0.0957, + "step": 13987 + }, + { + "epoch": 0.12142255709585854, + "grad_norm": 0.28125, + "learning_rate": 0.0019182670638168999, + "loss": 0.1602, + "step": 13988 + }, + { + "epoch": 0.12143123757606271, + "grad_norm": 0.322265625, + "learning_rate": 0.00191825467054923, + "loss": 0.1045, + "step": 13989 + }, + { + "epoch": 0.12143991805626687, + "grad_norm": 0.287109375, + "learning_rate": 0.0019182422763867287, + "loss": 0.1143, + "step": 13990 + }, + { + "epoch": 0.12144859853647104, + "grad_norm": 0.515625, + "learning_rate": 0.00191822988132941, + "loss": 0.1104, + "step": 13991 + }, + { + "epoch": 0.1214572790166752, + "grad_norm": 0.2265625, + "learning_rate": 0.0019182174853772877, + "loss": 0.1025, + "step": 13992 + }, + { + "epoch": 0.12146595949687937, + "grad_norm": 0.15625, + "learning_rate": 0.0019182050885303747, + "loss": 0.1162, + "step": 13993 + }, + { + "epoch": 0.12147463997708353, + "grad_norm": 0.314453125, + "learning_rate": 0.0019181926907886851, + "loss": 0.1211, + "step": 13994 + }, + { + "epoch": 0.1214833204572877, + "grad_norm": 0.7265625, + "learning_rate": 0.0019181802921522323, + "loss": 0.1309, + "step": 13995 + }, + { + "epoch": 0.12149200093749186, + "grad_norm": 0.126953125, + "learning_rate": 0.0019181678926210295, + "loss": 0.1016, + "step": 13996 + }, + { + "epoch": 0.12150068141769603, + "grad_norm": 0.984375, + "learning_rate": 0.001918155492195091, + "loss": 0.1348, + "step": 13997 + }, + { + "epoch": 0.1215093618979002, + "grad_norm": 0.47265625, + "learning_rate": 0.0019181430908744297, + "loss": 0.1328, + "step": 13998 + }, + { + "epoch": 0.12151804237810436, + "grad_norm": 0.193359375, + "learning_rate": 0.0019181306886590596, + "loss": 0.1113, + "step": 13999 + }, + { + "epoch": 0.12152672285830853, + "grad_norm": 0.328125, + "learning_rate": 0.0019181182855489939, + "loss": 0.1152, + "step": 14000 + }, + { + "epoch": 0.12153540333851269, + "grad_norm": 0.302734375, + "learning_rate": 0.0019181058815442463, + "loss": 0.1143, + "step": 14001 + }, + { + "epoch": 0.12154408381871686, + "grad_norm": 0.11865234375, + "learning_rate": 0.0019180934766448306, + "loss": 0.1162, + "step": 14002 + }, + { + "epoch": 0.12155276429892102, + "grad_norm": 0.150390625, + "learning_rate": 0.0019180810708507598, + "loss": 0.1123, + "step": 14003 + }, + { + "epoch": 0.12156144477912519, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019180686641620483, + "loss": 0.0972, + "step": 14004 + }, + { + "epoch": 0.12157012525932935, + "grad_norm": 0.10986328125, + "learning_rate": 0.0019180562565787088, + "loss": 0.1406, + "step": 14005 + }, + { + "epoch": 0.12157880573953352, + "grad_norm": 0.3203125, + "learning_rate": 0.0019180438481007556, + "loss": 0.1182, + "step": 14006 + }, + { + "epoch": 0.12158748621973768, + "grad_norm": 0.9453125, + "learning_rate": 0.001918031438728202, + "loss": 0.1309, + "step": 14007 + }, + { + "epoch": 0.12159616669994185, + "grad_norm": 0.1630859375, + "learning_rate": 0.001918019028461061, + "loss": 0.1245, + "step": 14008 + }, + { + "epoch": 0.12160484718014601, + "grad_norm": 0.29296875, + "learning_rate": 0.001918006617299347, + "loss": 0.0996, + "step": 14009 + }, + { + "epoch": 0.12161352766035018, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019179942052430733, + "loss": 0.125, + "step": 14010 + }, + { + "epoch": 0.12162220814055434, + "grad_norm": 0.5234375, + "learning_rate": 0.0019179817922922533, + "loss": 0.1187, + "step": 14011 + }, + { + "epoch": 0.1216308886207585, + "grad_norm": 0.61328125, + "learning_rate": 0.0019179693784469008, + "loss": 0.124, + "step": 14012 + }, + { + "epoch": 0.12163956910096267, + "grad_norm": 0.154296875, + "learning_rate": 0.001917956963707029, + "loss": 0.1816, + "step": 14013 + }, + { + "epoch": 0.12164824958116684, + "grad_norm": 0.3515625, + "learning_rate": 0.0019179445480726521, + "loss": 0.1016, + "step": 14014 + }, + { + "epoch": 0.121656930061371, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019179321315437835, + "loss": 0.1221, + "step": 14015 + }, + { + "epoch": 0.12166561054157515, + "grad_norm": 0.201171875, + "learning_rate": 0.0019179197141204363, + "loss": 0.1357, + "step": 14016 + }, + { + "epoch": 0.12167429102177932, + "grad_norm": 0.373046875, + "learning_rate": 0.0019179072958026243, + "loss": 0.1387, + "step": 14017 + }, + { + "epoch": 0.12168297150198348, + "grad_norm": 0.76953125, + "learning_rate": 0.0019178948765903611, + "loss": 0.1104, + "step": 14018 + }, + { + "epoch": 0.12169165198218765, + "grad_norm": 0.23828125, + "learning_rate": 0.0019178824564836609, + "loss": 0.1074, + "step": 14019 + }, + { + "epoch": 0.12170033246239181, + "grad_norm": 0.26953125, + "learning_rate": 0.001917870035482536, + "loss": 0.123, + "step": 14020 + }, + { + "epoch": 0.12170901294259598, + "grad_norm": 0.90625, + "learning_rate": 0.0019178576135870011, + "loss": 0.123, + "step": 14021 + }, + { + "epoch": 0.12171769342280014, + "grad_norm": 0.32421875, + "learning_rate": 0.0019178451907970696, + "loss": 0.1182, + "step": 14022 + }, + { + "epoch": 0.12172637390300431, + "grad_norm": 0.412109375, + "learning_rate": 0.0019178327671127546, + "loss": 0.1602, + "step": 14023 + }, + { + "epoch": 0.12173505438320847, + "grad_norm": 0.443359375, + "learning_rate": 0.00191782034253407, + "loss": 0.1143, + "step": 14024 + }, + { + "epoch": 0.12174373486341264, + "grad_norm": 0.44140625, + "learning_rate": 0.0019178079170610294, + "loss": 0.1465, + "step": 14025 + }, + { + "epoch": 0.1217524153436168, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019177954906936467, + "loss": 0.1143, + "step": 14026 + }, + { + "epoch": 0.12176109582382097, + "grad_norm": 0.86328125, + "learning_rate": 0.0019177830634319346, + "loss": 0.1104, + "step": 14027 + }, + { + "epoch": 0.12176977630402513, + "grad_norm": 0.25390625, + "learning_rate": 0.0019177706352759077, + "loss": 0.1094, + "step": 14028 + }, + { + "epoch": 0.1217784567842293, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019177582062255788, + "loss": 0.0991, + "step": 14029 + }, + { + "epoch": 0.12178713726443346, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019177457762809618, + "loss": 0.0952, + "step": 14030 + }, + { + "epoch": 0.12179581774463763, + "grad_norm": 0.14453125, + "learning_rate": 0.0019177333454420704, + "loss": 0.165, + "step": 14031 + }, + { + "epoch": 0.1218044982248418, + "grad_norm": 0.15625, + "learning_rate": 0.001917720913708918, + "loss": 0.1123, + "step": 14032 + }, + { + "epoch": 0.12181317870504596, + "grad_norm": 0.2578125, + "learning_rate": 0.0019177084810815183, + "loss": 0.0947, + "step": 14033 + }, + { + "epoch": 0.12182185918525013, + "grad_norm": 0.1875, + "learning_rate": 0.001917696047559885, + "loss": 0.1045, + "step": 14034 + }, + { + "epoch": 0.12183053966545429, + "grad_norm": 0.55859375, + "learning_rate": 0.0019176836131440313, + "loss": 0.0752, + "step": 14035 + }, + { + "epoch": 0.12183922014565846, + "grad_norm": 0.19921875, + "learning_rate": 0.0019176711778339714, + "loss": 0.0908, + "step": 14036 + }, + { + "epoch": 0.12184790062586262, + "grad_norm": 0.6171875, + "learning_rate": 0.0019176587416297186, + "loss": 0.0967, + "step": 14037 + }, + { + "epoch": 0.12185658110606679, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019176463045312863, + "loss": 0.126, + "step": 14038 + }, + { + "epoch": 0.12186526158627095, + "grad_norm": 0.404296875, + "learning_rate": 0.0019176338665386886, + "loss": 0.0742, + "step": 14039 + }, + { + "epoch": 0.12187394206647512, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019176214276519386, + "loss": 0.1172, + "step": 14040 + }, + { + "epoch": 0.12188262254667928, + "grad_norm": 0.1484375, + "learning_rate": 0.00191760898787105, + "loss": 0.1348, + "step": 14041 + }, + { + "epoch": 0.12189130302688345, + "grad_norm": 0.57421875, + "learning_rate": 0.0019175965471960367, + "loss": 0.1133, + "step": 14042 + }, + { + "epoch": 0.12189998350708761, + "grad_norm": 0.2314453125, + "learning_rate": 0.001917584105626912, + "loss": 0.123, + "step": 14043 + }, + { + "epoch": 0.12190866398729178, + "grad_norm": 0.400390625, + "learning_rate": 0.0019175716631636894, + "loss": 0.1357, + "step": 14044 + }, + { + "epoch": 0.12191734446749594, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019175592198063828, + "loss": 0.1113, + "step": 14045 + }, + { + "epoch": 0.1219260249477001, + "grad_norm": 0.29296875, + "learning_rate": 0.001917546775555006, + "loss": 0.1191, + "step": 14046 + }, + { + "epoch": 0.12193470542790427, + "grad_norm": 0.515625, + "learning_rate": 0.0019175343304095724, + "loss": 0.1094, + "step": 14047 + }, + { + "epoch": 0.12194338590810844, + "grad_norm": 0.70703125, + "learning_rate": 0.001917521884370095, + "loss": 0.0991, + "step": 14048 + }, + { + "epoch": 0.1219520663883126, + "grad_norm": 0.470703125, + "learning_rate": 0.0019175094374365882, + "loss": 0.1094, + "step": 14049 + }, + { + "epoch": 0.12196074686851677, + "grad_norm": 0.134765625, + "learning_rate": 0.0019174969896090656, + "loss": 0.1426, + "step": 14050 + }, + { + "epoch": 0.12196942734872093, + "grad_norm": 0.29296875, + "learning_rate": 0.0019174845408875402, + "loss": 0.0869, + "step": 14051 + }, + { + "epoch": 0.1219781078289251, + "grad_norm": 0.1953125, + "learning_rate": 0.001917472091272026, + "loss": 0.1367, + "step": 14052 + }, + { + "epoch": 0.12198678830912926, + "grad_norm": 0.248046875, + "learning_rate": 0.0019174596407625368, + "loss": 0.1104, + "step": 14053 + }, + { + "epoch": 0.12199546878933343, + "grad_norm": 1.4921875, + "learning_rate": 0.001917447189359086, + "loss": 0.1533, + "step": 14054 + }, + { + "epoch": 0.12200414926953759, + "grad_norm": 0.2451171875, + "learning_rate": 0.001917434737061687, + "loss": 0.126, + "step": 14055 + }, + { + "epoch": 0.12201282974974176, + "grad_norm": 0.34375, + "learning_rate": 0.0019174222838703543, + "loss": 0.1074, + "step": 14056 + }, + { + "epoch": 0.12202151022994592, + "grad_norm": 0.52734375, + "learning_rate": 0.0019174098297851004, + "loss": 0.1191, + "step": 14057 + }, + { + "epoch": 0.12203019071015009, + "grad_norm": 0.1484375, + "learning_rate": 0.0019173973748059393, + "loss": 0.0977, + "step": 14058 + }, + { + "epoch": 0.12203887119035425, + "grad_norm": 0.53515625, + "learning_rate": 0.001917384918932885, + "loss": 0.0845, + "step": 14059 + }, + { + "epoch": 0.12204755167055842, + "grad_norm": 0.62109375, + "learning_rate": 0.0019173724621659506, + "loss": 0.0981, + "step": 14060 + }, + { + "epoch": 0.12205623215076258, + "grad_norm": 0.236328125, + "learning_rate": 0.00191736000450515, + "loss": 0.127, + "step": 14061 + }, + { + "epoch": 0.12206491263096675, + "grad_norm": 0.328125, + "learning_rate": 0.001917347545950497, + "loss": 0.0986, + "step": 14062 + }, + { + "epoch": 0.12207359311117091, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019173350865020048, + "loss": 0.127, + "step": 14063 + }, + { + "epoch": 0.12208227359137508, + "grad_norm": 0.259765625, + "learning_rate": 0.001917322626159687, + "loss": 0.1504, + "step": 14064 + }, + { + "epoch": 0.12209095407157924, + "grad_norm": 0.404296875, + "learning_rate": 0.0019173101649235579, + "loss": 0.1484, + "step": 14065 + }, + { + "epoch": 0.12209963455178341, + "grad_norm": 0.341796875, + "learning_rate": 0.0019172977027936302, + "loss": 0.1045, + "step": 14066 + }, + { + "epoch": 0.12210831503198757, + "grad_norm": 0.28125, + "learning_rate": 0.0019172852397699185, + "loss": 0.1045, + "step": 14067 + }, + { + "epoch": 0.12211699551219174, + "grad_norm": 0.29296875, + "learning_rate": 0.0019172727758524357, + "loss": 0.1245, + "step": 14068 + }, + { + "epoch": 0.1221256759923959, + "grad_norm": 0.5, + "learning_rate": 0.0019172603110411957, + "loss": 0.1055, + "step": 14069 + }, + { + "epoch": 0.12213435647260007, + "grad_norm": 0.609375, + "learning_rate": 0.001917247845336212, + "loss": 0.104, + "step": 14070 + }, + { + "epoch": 0.12214303695280423, + "grad_norm": 0.69140625, + "learning_rate": 0.0019172353787374988, + "loss": 0.166, + "step": 14071 + }, + { + "epoch": 0.1221517174330084, + "grad_norm": 0.16796875, + "learning_rate": 0.0019172229112450687, + "loss": 0.1143, + "step": 14072 + }, + { + "epoch": 0.12216039791321257, + "grad_norm": 0.11083984375, + "learning_rate": 0.001917210442858936, + "loss": 0.1074, + "step": 14073 + }, + { + "epoch": 0.12216907839341673, + "grad_norm": 0.16796875, + "learning_rate": 0.0019171979735791145, + "loss": 0.0835, + "step": 14074 + }, + { + "epoch": 0.1221777588736209, + "grad_norm": 0.640625, + "learning_rate": 0.0019171855034056174, + "loss": 0.1377, + "step": 14075 + }, + { + "epoch": 0.12218643935382506, + "grad_norm": 0.435546875, + "learning_rate": 0.0019171730323384586, + "loss": 0.1221, + "step": 14076 + }, + { + "epoch": 0.12219511983402921, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019171605603776516, + "loss": 0.0903, + "step": 14077 + }, + { + "epoch": 0.12220380031423338, + "grad_norm": 0.1708984375, + "learning_rate": 0.00191714808752321, + "loss": 0.126, + "step": 14078 + }, + { + "epoch": 0.12221248079443754, + "grad_norm": 1.078125, + "learning_rate": 0.0019171356137751478, + "loss": 0.207, + "step": 14079 + }, + { + "epoch": 0.12222116127464171, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019171231391334781, + "loss": 0.083, + "step": 14080 + }, + { + "epoch": 0.12222984175484587, + "grad_norm": 0.3984375, + "learning_rate": 0.0019171106635982149, + "loss": 0.125, + "step": 14081 + }, + { + "epoch": 0.12223852223505004, + "grad_norm": 1.2734375, + "learning_rate": 0.0019170981871693717, + "loss": 0.1562, + "step": 14082 + }, + { + "epoch": 0.1222472027152542, + "grad_norm": 0.12255859375, + "learning_rate": 0.0019170857098469625, + "loss": 0.1338, + "step": 14083 + }, + { + "epoch": 0.12225588319545837, + "grad_norm": 0.45703125, + "learning_rate": 0.0019170732316310002, + "loss": 0.085, + "step": 14084 + }, + { + "epoch": 0.12226456367566253, + "grad_norm": 0.1875, + "learning_rate": 0.0019170607525214988, + "loss": 0.127, + "step": 14085 + }, + { + "epoch": 0.1222732441558667, + "grad_norm": 0.126953125, + "learning_rate": 0.0019170482725184728, + "loss": 0.1108, + "step": 14086 + }, + { + "epoch": 0.12228192463607086, + "grad_norm": 0.46875, + "learning_rate": 0.0019170357916219345, + "loss": 0.1602, + "step": 14087 + }, + { + "epoch": 0.12229060511627503, + "grad_norm": 0.5859375, + "learning_rate": 0.001917023309831898, + "loss": 0.1289, + "step": 14088 + }, + { + "epoch": 0.1222992855964792, + "grad_norm": 0.490234375, + "learning_rate": 0.0019170108271483776, + "loss": 0.1582, + "step": 14089 + }, + { + "epoch": 0.12230796607668336, + "grad_norm": 0.1474609375, + "learning_rate": 0.001916998343571386, + "loss": 0.0825, + "step": 14090 + }, + { + "epoch": 0.12231664655688752, + "grad_norm": 0.388671875, + "learning_rate": 0.0019169858591009375, + "loss": 0.1001, + "step": 14091 + }, + { + "epoch": 0.12232532703709169, + "grad_norm": 0.294921875, + "learning_rate": 0.0019169733737370454, + "loss": 0.1357, + "step": 14092 + }, + { + "epoch": 0.12233400751729585, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019169608874797237, + "loss": 0.1172, + "step": 14093 + }, + { + "epoch": 0.12234268799750002, + "grad_norm": 0.25390625, + "learning_rate": 0.0019169484003289858, + "loss": 0.1216, + "step": 14094 + }, + { + "epoch": 0.12235136847770418, + "grad_norm": 0.56640625, + "learning_rate": 0.0019169359122848454, + "loss": 0.0815, + "step": 14095 + }, + { + "epoch": 0.12236004895790835, + "grad_norm": 0.37109375, + "learning_rate": 0.0019169234233473158, + "loss": 0.1416, + "step": 14096 + }, + { + "epoch": 0.12236872943811251, + "grad_norm": 0.095703125, + "learning_rate": 0.0019169109335164115, + "loss": 0.0947, + "step": 14097 + }, + { + "epoch": 0.12237740991831668, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019168984427921454, + "loss": 0.1426, + "step": 14098 + }, + { + "epoch": 0.12238609039852084, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019168859511745316, + "loss": 0.124, + "step": 14099 + }, + { + "epoch": 0.12239477087872501, + "grad_norm": 0.134765625, + "learning_rate": 0.0019168734586635834, + "loss": 0.1084, + "step": 14100 + }, + { + "epoch": 0.12240345135892917, + "grad_norm": 0.51953125, + "learning_rate": 0.001916860965259315, + "loss": 0.0913, + "step": 14101 + }, + { + "epoch": 0.12241213183913334, + "grad_norm": 0.8203125, + "learning_rate": 0.0019168484709617395, + "loss": 0.1064, + "step": 14102 + }, + { + "epoch": 0.1224208123193375, + "grad_norm": 0.10205078125, + "learning_rate": 0.001916835975770871, + "loss": 0.1377, + "step": 14103 + }, + { + "epoch": 0.12242949279954167, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019168234796867224, + "loss": 0.0938, + "step": 14104 + }, + { + "epoch": 0.12243817327974583, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019168109827093084, + "loss": 0.1079, + "step": 14105 + }, + { + "epoch": 0.12244685375995, + "grad_norm": 0.1064453125, + "learning_rate": 0.001916798484838642, + "loss": 0.127, + "step": 14106 + }, + { + "epoch": 0.12245553424015417, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019167859860747371, + "loss": 0.0991, + "step": 14107 + }, + { + "epoch": 0.12246421472035833, + "grad_norm": 0.494140625, + "learning_rate": 0.0019167734864176075, + "loss": 0.0933, + "step": 14108 + }, + { + "epoch": 0.1224728952005625, + "grad_norm": 0.2275390625, + "learning_rate": 0.001916760985867266, + "loss": 0.1187, + "step": 14109 + }, + { + "epoch": 0.12248157568076666, + "grad_norm": 0.185546875, + "learning_rate": 0.0019167484844237277, + "loss": 0.1182, + "step": 14110 + }, + { + "epoch": 0.12249025616097083, + "grad_norm": 0.255859375, + "learning_rate": 0.0019167359820870052, + "loss": 0.1191, + "step": 14111 + }, + { + "epoch": 0.12249893664117499, + "grad_norm": 0.55859375, + "learning_rate": 0.0019167234788571128, + "loss": 0.1143, + "step": 14112 + }, + { + "epoch": 0.12250761712137916, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019167109747340635, + "loss": 0.1211, + "step": 14113 + }, + { + "epoch": 0.12251629760158332, + "grad_norm": 0.447265625, + "learning_rate": 0.0019166984697178717, + "loss": 0.1279, + "step": 14114 + }, + { + "epoch": 0.12252497808178749, + "grad_norm": 0.51171875, + "learning_rate": 0.0019166859638085503, + "loss": 0.1211, + "step": 14115 + }, + { + "epoch": 0.12253365856199165, + "grad_norm": 0.71484375, + "learning_rate": 0.0019166734570061138, + "loss": 0.1035, + "step": 14116 + }, + { + "epoch": 0.12254233904219582, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019166609493105752, + "loss": 0.1279, + "step": 14117 + }, + { + "epoch": 0.12255101952239998, + "grad_norm": 0.57421875, + "learning_rate": 0.0019166484407219485, + "loss": 0.104, + "step": 14118 + }, + { + "epoch": 0.12255970000260415, + "grad_norm": 0.1796875, + "learning_rate": 0.0019166359312402475, + "loss": 0.1191, + "step": 14119 + }, + { + "epoch": 0.12256838048280831, + "grad_norm": 0.65234375, + "learning_rate": 0.0019166234208654854, + "loss": 0.1387, + "step": 14120 + }, + { + "epoch": 0.12257706096301248, + "grad_norm": 0.39453125, + "learning_rate": 0.0019166109095976765, + "loss": 0.1562, + "step": 14121 + }, + { + "epoch": 0.12258574144321664, + "grad_norm": 0.251953125, + "learning_rate": 0.0019165983974368341, + "loss": 0.0996, + "step": 14122 + }, + { + "epoch": 0.12259442192342081, + "grad_norm": 0.154296875, + "learning_rate": 0.001916585884382972, + "loss": 0.1377, + "step": 14123 + }, + { + "epoch": 0.12260310240362497, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019165733704361039, + "loss": 0.0767, + "step": 14124 + }, + { + "epoch": 0.12261178288382914, + "grad_norm": 0.201171875, + "learning_rate": 0.0019165608555962433, + "loss": 0.1357, + "step": 14125 + }, + { + "epoch": 0.1226204633640333, + "grad_norm": 0.74609375, + "learning_rate": 0.0019165483398634038, + "loss": 0.1152, + "step": 14126 + }, + { + "epoch": 0.12262914384423747, + "grad_norm": 0.126953125, + "learning_rate": 0.0019165358232375999, + "loss": 0.0967, + "step": 14127 + }, + { + "epoch": 0.12263782432444163, + "grad_norm": 0.16015625, + "learning_rate": 0.0019165233057188442, + "loss": 0.1074, + "step": 14128 + }, + { + "epoch": 0.1226465048046458, + "grad_norm": 0.28515625, + "learning_rate": 0.001916510787307151, + "loss": 0.1216, + "step": 14129 + }, + { + "epoch": 0.12265518528484996, + "grad_norm": 0.1142578125, + "learning_rate": 0.001916498268002534, + "loss": 0.1221, + "step": 14130 + }, + { + "epoch": 0.12266386576505413, + "grad_norm": 0.72265625, + "learning_rate": 0.0019164857478050066, + "loss": 0.168, + "step": 14131 + }, + { + "epoch": 0.1226725462452583, + "grad_norm": 0.134765625, + "learning_rate": 0.0019164732267145828, + "loss": 0.1504, + "step": 14132 + }, + { + "epoch": 0.12268122672546246, + "grad_norm": 0.11669921875, + "learning_rate": 0.0019164607047312762, + "loss": 0.1045, + "step": 14133 + }, + { + "epoch": 0.12268990720566662, + "grad_norm": 0.73046875, + "learning_rate": 0.0019164481818551, + "loss": 0.1104, + "step": 14134 + }, + { + "epoch": 0.12269858768587079, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019164356580860687, + "loss": 0.1074, + "step": 14135 + }, + { + "epoch": 0.12270726816607495, + "grad_norm": 0.17578125, + "learning_rate": 0.0019164231334241955, + "loss": 0.1021, + "step": 14136 + }, + { + "epoch": 0.12271594864627912, + "grad_norm": 0.44140625, + "learning_rate": 0.0019164106078694942, + "loss": 0.085, + "step": 14137 + }, + { + "epoch": 0.12272462912648328, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019163980814219787, + "loss": 0.0942, + "step": 14138 + }, + { + "epoch": 0.12273330960668744, + "grad_norm": 0.83203125, + "learning_rate": 0.0019163855540816623, + "loss": 0.1089, + "step": 14139 + }, + { + "epoch": 0.1227419900868916, + "grad_norm": 0.1328125, + "learning_rate": 0.0019163730258485593, + "loss": 0.1133, + "step": 14140 + }, + { + "epoch": 0.12275067056709577, + "grad_norm": 0.287109375, + "learning_rate": 0.0019163604967226827, + "loss": 0.1074, + "step": 14141 + }, + { + "epoch": 0.12275935104729993, + "grad_norm": 0.470703125, + "learning_rate": 0.0019163479667040468, + "loss": 0.1416, + "step": 14142 + }, + { + "epoch": 0.1227680315275041, + "grad_norm": 0.283203125, + "learning_rate": 0.0019163354357926645, + "loss": 0.1191, + "step": 14143 + }, + { + "epoch": 0.12277671200770826, + "grad_norm": 0.275390625, + "learning_rate": 0.0019163229039885505, + "loss": 0.1299, + "step": 14144 + }, + { + "epoch": 0.12278539248791243, + "grad_norm": 0.134765625, + "learning_rate": 0.001916310371291718, + "loss": 0.126, + "step": 14145 + }, + { + "epoch": 0.12279407296811659, + "grad_norm": 0.486328125, + "learning_rate": 0.0019162978377021806, + "loss": 0.127, + "step": 14146 + }, + { + "epoch": 0.12280275344832076, + "grad_norm": 2.75, + "learning_rate": 0.0019162853032199524, + "loss": 0.3652, + "step": 14147 + }, + { + "epoch": 0.12281143392852492, + "grad_norm": 0.14453125, + "learning_rate": 0.0019162727678450468, + "loss": 0.0986, + "step": 14148 + }, + { + "epoch": 0.12282011440872909, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019162602315774772, + "loss": 0.1143, + "step": 14149 + }, + { + "epoch": 0.12282879488893325, + "grad_norm": 0.314453125, + "learning_rate": 0.0019162476944172577, + "loss": 0.085, + "step": 14150 + }, + { + "epoch": 0.12283747536913742, + "grad_norm": 1.125, + "learning_rate": 0.0019162351563644022, + "loss": 0.1357, + "step": 14151 + }, + { + "epoch": 0.12284615584934158, + "grad_norm": 0.625, + "learning_rate": 0.0019162226174189243, + "loss": 0.1201, + "step": 14152 + }, + { + "epoch": 0.12285483632954575, + "grad_norm": 0.109375, + "learning_rate": 0.0019162100775808376, + "loss": 0.1309, + "step": 14153 + }, + { + "epoch": 0.12286351680974991, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019161975368501556, + "loss": 0.1465, + "step": 14154 + }, + { + "epoch": 0.12287219728995408, + "grad_norm": 0.091796875, + "learning_rate": 0.0019161849952268925, + "loss": 0.1582, + "step": 14155 + }, + { + "epoch": 0.12288087777015824, + "grad_norm": 0.09326171875, + "learning_rate": 0.0019161724527110616, + "loss": 0.1641, + "step": 14156 + }, + { + "epoch": 0.12288955825036241, + "grad_norm": 0.064453125, + "learning_rate": 0.0019161599093026767, + "loss": 0.0811, + "step": 14157 + }, + { + "epoch": 0.12289823873056657, + "grad_norm": 0.3359375, + "learning_rate": 0.0019161473650017515, + "loss": 0.0918, + "step": 14158 + }, + { + "epoch": 0.12290691921077074, + "grad_norm": 0.484375, + "learning_rate": 0.0019161348198083001, + "loss": 0.0815, + "step": 14159 + }, + { + "epoch": 0.1229155996909749, + "grad_norm": 0.48046875, + "learning_rate": 0.0019161222737223357, + "loss": 0.083, + "step": 14160 + }, + { + "epoch": 0.12292428017117907, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019161097267438726, + "loss": 0.0962, + "step": 14161 + }, + { + "epoch": 0.12293296065138323, + "grad_norm": 0.330078125, + "learning_rate": 0.0019160971788729237, + "loss": 0.124, + "step": 14162 + }, + { + "epoch": 0.1229416411315874, + "grad_norm": 0.78125, + "learning_rate": 0.0019160846301095036, + "loss": 0.1069, + "step": 14163 + }, + { + "epoch": 0.12295032161179156, + "grad_norm": 0.41015625, + "learning_rate": 0.0019160720804536253, + "loss": 0.1406, + "step": 14164 + }, + { + "epoch": 0.12295900209199573, + "grad_norm": 0.10595703125, + "learning_rate": 0.001916059529905303, + "loss": 0.1147, + "step": 14165 + }, + { + "epoch": 0.1229676825721999, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019160469784645503, + "loss": 0.1021, + "step": 14166 + }, + { + "epoch": 0.12297636305240406, + "grad_norm": 0.42578125, + "learning_rate": 0.0019160344261313805, + "loss": 0.1152, + "step": 14167 + }, + { + "epoch": 0.12298504353260822, + "grad_norm": 0.5703125, + "learning_rate": 0.0019160218729058081, + "loss": 0.1543, + "step": 14168 + }, + { + "epoch": 0.12299372401281239, + "grad_norm": 0.3671875, + "learning_rate": 0.0019160093187878461, + "loss": 0.1143, + "step": 14169 + }, + { + "epoch": 0.12300240449301655, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019159967637775088, + "loss": 0.1074, + "step": 14170 + }, + { + "epoch": 0.12301108497322072, + "grad_norm": 0.11376953125, + "learning_rate": 0.00191598420787481, + "loss": 0.1436, + "step": 14171 + }, + { + "epoch": 0.12301976545342488, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019159716510797628, + "loss": 0.1162, + "step": 14172 + }, + { + "epoch": 0.12302844593362905, + "grad_norm": 0.69140625, + "learning_rate": 0.0019159590933923811, + "loss": 0.1045, + "step": 14173 + }, + { + "epoch": 0.12303712641383321, + "grad_norm": 0.92578125, + "learning_rate": 0.001915946534812679, + "loss": 0.1465, + "step": 14174 + }, + { + "epoch": 0.12304580689403738, + "grad_norm": 0.06982421875, + "learning_rate": 0.0019159339753406703, + "loss": 0.1006, + "step": 14175 + }, + { + "epoch": 0.12305448737424154, + "grad_norm": 0.30078125, + "learning_rate": 0.001915921414976368, + "loss": 0.1001, + "step": 14176 + }, + { + "epoch": 0.12306316785444571, + "grad_norm": 0.91796875, + "learning_rate": 0.0019159088537197865, + "loss": 0.0933, + "step": 14177 + }, + { + "epoch": 0.12307184833464987, + "grad_norm": 0.8125, + "learning_rate": 0.0019158962915709393, + "loss": 0.127, + "step": 14178 + }, + { + "epoch": 0.12308052881485404, + "grad_norm": 0.5859375, + "learning_rate": 0.0019158837285298404, + "loss": 0.0942, + "step": 14179 + }, + { + "epoch": 0.1230892092950582, + "grad_norm": 0.314453125, + "learning_rate": 0.0019158711645965029, + "loss": 0.0928, + "step": 14180 + }, + { + "epoch": 0.12309788977526237, + "grad_norm": 0.333984375, + "learning_rate": 0.0019158585997709411, + "loss": 0.1201, + "step": 14181 + }, + { + "epoch": 0.12310657025546654, + "grad_norm": 0.2890625, + "learning_rate": 0.0019158460340531686, + "loss": 0.0889, + "step": 14182 + }, + { + "epoch": 0.1231152507356707, + "grad_norm": 0.66796875, + "learning_rate": 0.0019158334674431992, + "loss": 0.0742, + "step": 14183 + }, + { + "epoch": 0.12312393121587487, + "grad_norm": 0.55078125, + "learning_rate": 0.0019158208999410467, + "loss": 0.085, + "step": 14184 + }, + { + "epoch": 0.12313261169607903, + "grad_norm": 0.34765625, + "learning_rate": 0.0019158083315467245, + "loss": 0.1572, + "step": 14185 + }, + { + "epoch": 0.1231412921762832, + "grad_norm": 0.71484375, + "learning_rate": 0.0019157957622602468, + "loss": 0.1602, + "step": 14186 + }, + { + "epoch": 0.12314997265648736, + "grad_norm": 0.671875, + "learning_rate": 0.001915783192081627, + "loss": 0.1572, + "step": 14187 + }, + { + "epoch": 0.12315865313669153, + "grad_norm": 0.671875, + "learning_rate": 0.001915770621010879, + "loss": 0.1094, + "step": 14188 + }, + { + "epoch": 0.12316733361689569, + "grad_norm": 0.38671875, + "learning_rate": 0.0019157580490480161, + "loss": 0.1006, + "step": 14189 + }, + { + "epoch": 0.12317601409709986, + "grad_norm": 0.28125, + "learning_rate": 0.001915745476193053, + "loss": 0.1143, + "step": 14190 + }, + { + "epoch": 0.12318469457730402, + "grad_norm": 0.13671875, + "learning_rate": 0.0019157329024460025, + "loss": 0.0938, + "step": 14191 + }, + { + "epoch": 0.12319337505750819, + "grad_norm": 0.388671875, + "learning_rate": 0.0019157203278068792, + "loss": 0.1768, + "step": 14192 + }, + { + "epoch": 0.12320205553771235, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019157077522756957, + "loss": 0.1533, + "step": 14193 + }, + { + "epoch": 0.12321073601791652, + "grad_norm": 0.32421875, + "learning_rate": 0.001915695175852467, + "loss": 0.0845, + "step": 14194 + }, + { + "epoch": 0.12321941649812068, + "grad_norm": 0.181640625, + "learning_rate": 0.0019156825985372065, + "loss": 0.1143, + "step": 14195 + }, + { + "epoch": 0.12322809697832485, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019156700203299274, + "loss": 0.127, + "step": 14196 + }, + { + "epoch": 0.12323677745852901, + "grad_norm": 0.5234375, + "learning_rate": 0.0019156574412306437, + "loss": 0.1084, + "step": 14197 + }, + { + "epoch": 0.12324545793873318, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019156448612393695, + "loss": 0.0986, + "step": 14198 + }, + { + "epoch": 0.12325413841893734, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019156322803561183, + "loss": 0.1299, + "step": 14199 + }, + { + "epoch": 0.1232628188991415, + "grad_norm": 0.205078125, + "learning_rate": 0.001915619698580904, + "loss": 0.124, + "step": 14200 + }, + { + "epoch": 0.12327149937934566, + "grad_norm": 0.173828125, + "learning_rate": 0.0019156071159137401, + "loss": 0.1133, + "step": 14201 + }, + { + "epoch": 0.12328017985954982, + "grad_norm": 0.0673828125, + "learning_rate": 0.0019155945323546407, + "loss": 0.0986, + "step": 14202 + }, + { + "epoch": 0.12328886033975399, + "grad_norm": 0.09912109375, + "learning_rate": 0.0019155819479036195, + "loss": 0.1064, + "step": 14203 + }, + { + "epoch": 0.12329754081995815, + "grad_norm": 0.58984375, + "learning_rate": 0.0019155693625606896, + "loss": 0.1582, + "step": 14204 + }, + { + "epoch": 0.12330622130016232, + "grad_norm": 0.85546875, + "learning_rate": 0.0019155567763258657, + "loss": 0.123, + "step": 14205 + }, + { + "epoch": 0.12331490178036648, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019155441891991608, + "loss": 0.0825, + "step": 14206 + }, + { + "epoch": 0.12332358226057065, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019155316011805893, + "loss": 0.125, + "step": 14207 + }, + { + "epoch": 0.12333226274077481, + "grad_norm": 0.07958984375, + "learning_rate": 0.0019155190122701648, + "loss": 0.104, + "step": 14208 + }, + { + "epoch": 0.12334094322097898, + "grad_norm": 0.6640625, + "learning_rate": 0.0019155064224679007, + "loss": 0.0918, + "step": 14209 + }, + { + "epoch": 0.12334962370118314, + "grad_norm": 0.13671875, + "learning_rate": 0.0019154938317738112, + "loss": 0.1338, + "step": 14210 + }, + { + "epoch": 0.12335830418138731, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019154812401879098, + "loss": 0.1201, + "step": 14211 + }, + { + "epoch": 0.12336698466159148, + "grad_norm": 0.60546875, + "learning_rate": 0.0019154686477102106, + "loss": 0.1211, + "step": 14212 + }, + { + "epoch": 0.12337566514179564, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019154560543407269, + "loss": 0.1836, + "step": 14213 + }, + { + "epoch": 0.1233843456219998, + "grad_norm": 0.06640625, + "learning_rate": 0.0019154434600794727, + "loss": 0.0977, + "step": 14214 + }, + { + "epoch": 0.12339302610220397, + "grad_norm": 0.33203125, + "learning_rate": 0.0019154308649264617, + "loss": 0.1338, + "step": 14215 + }, + { + "epoch": 0.12340170658240814, + "grad_norm": 0.2734375, + "learning_rate": 0.0019154182688817078, + "loss": 0.0791, + "step": 14216 + }, + { + "epoch": 0.1234103870626123, + "grad_norm": 0.474609375, + "learning_rate": 0.0019154056719452254, + "loss": 0.0938, + "step": 14217 + }, + { + "epoch": 0.12341906754281647, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019153930741170268, + "loss": 0.0938, + "step": 14218 + }, + { + "epoch": 0.12342774802302063, + "grad_norm": 0.333984375, + "learning_rate": 0.001915380475397127, + "loss": 0.2539, + "step": 14219 + }, + { + "epoch": 0.1234364285032248, + "grad_norm": 0.2890625, + "learning_rate": 0.001915367875785539, + "loss": 0.0786, + "step": 14220 + }, + { + "epoch": 0.12344510898342896, + "grad_norm": 0.71875, + "learning_rate": 0.001915355275282277, + "loss": 0.1084, + "step": 14221 + }, + { + "epoch": 0.12345378946363313, + "grad_norm": 0.34765625, + "learning_rate": 0.001915342673887355, + "loss": 0.0874, + "step": 14222 + }, + { + "epoch": 0.12346246994383729, + "grad_norm": 0.474609375, + "learning_rate": 0.0019153300716007865, + "loss": 0.1191, + "step": 14223 + }, + { + "epoch": 0.12347115042404146, + "grad_norm": 0.37109375, + "learning_rate": 0.0019153174684225852, + "loss": 0.0972, + "step": 14224 + }, + { + "epoch": 0.12347983090424562, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019153048643527648, + "loss": 0.0825, + "step": 14225 + }, + { + "epoch": 0.12348851138444979, + "grad_norm": 0.765625, + "learning_rate": 0.0019152922593913397, + "loss": 0.0933, + "step": 14226 + }, + { + "epoch": 0.12349719186465395, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019152796535383224, + "loss": 0.1484, + "step": 14227 + }, + { + "epoch": 0.12350587234485812, + "grad_norm": 0.6796875, + "learning_rate": 0.0019152670467937282, + "loss": 0.1543, + "step": 14228 + }, + { + "epoch": 0.12351455282506228, + "grad_norm": 0.103515625, + "learning_rate": 0.0019152544391575698, + "loss": 0.1123, + "step": 14229 + }, + { + "epoch": 0.12352323330526645, + "grad_norm": 0.39453125, + "learning_rate": 0.0019152418306298618, + "loss": 0.1436, + "step": 14230 + }, + { + "epoch": 0.12353191378547061, + "grad_norm": 0.13671875, + "learning_rate": 0.0019152292212106176, + "loss": 0.1006, + "step": 14231 + }, + { + "epoch": 0.12354059426567478, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019152166108998506, + "loss": 0.1152, + "step": 14232 + }, + { + "epoch": 0.12354927474587894, + "grad_norm": 0.228515625, + "learning_rate": 0.0019152039996975752, + "loss": 0.1328, + "step": 14233 + }, + { + "epoch": 0.12355795522608311, + "grad_norm": 0.181640625, + "learning_rate": 0.0019151913876038046, + "loss": 0.125, + "step": 14234 + }, + { + "epoch": 0.12356663570628727, + "grad_norm": 0.515625, + "learning_rate": 0.0019151787746185536, + "loss": 0.1094, + "step": 14235 + }, + { + "epoch": 0.12357531618649144, + "grad_norm": 0.796875, + "learning_rate": 0.0019151661607418348, + "loss": 0.1357, + "step": 14236 + }, + { + "epoch": 0.1235839966666956, + "grad_norm": 0.4375, + "learning_rate": 0.001915153545973663, + "loss": 0.1069, + "step": 14237 + }, + { + "epoch": 0.12359267714689977, + "grad_norm": 0.26171875, + "learning_rate": 0.001915140930314051, + "loss": 0.1006, + "step": 14238 + }, + { + "epoch": 0.12360135762710393, + "grad_norm": 0.2578125, + "learning_rate": 0.0019151283137630135, + "loss": 0.1113, + "step": 14239 + }, + { + "epoch": 0.1236100381073081, + "grad_norm": 0.5078125, + "learning_rate": 0.0019151156963205637, + "loss": 0.0869, + "step": 14240 + }, + { + "epoch": 0.12361871858751226, + "grad_norm": 0.4296875, + "learning_rate": 0.0019151030779867159, + "loss": 0.1094, + "step": 14241 + }, + { + "epoch": 0.12362739906771643, + "grad_norm": 0.0576171875, + "learning_rate": 0.0019150904587614832, + "loss": 0.0928, + "step": 14242 + }, + { + "epoch": 0.1236360795479206, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019150778386448802, + "loss": 0.1079, + "step": 14243 + }, + { + "epoch": 0.12364476002812476, + "grad_norm": 0.2236328125, + "learning_rate": 0.00191506521763692, + "loss": 0.1387, + "step": 14244 + }, + { + "epoch": 0.12365344050832892, + "grad_norm": 0.333984375, + "learning_rate": 0.001915052595737617, + "loss": 0.1719, + "step": 14245 + }, + { + "epoch": 0.12366212098853309, + "grad_norm": 0.51953125, + "learning_rate": 0.0019150399729469848, + "loss": 0.1152, + "step": 14246 + }, + { + "epoch": 0.12367080146873725, + "grad_norm": 0.17578125, + "learning_rate": 0.0019150273492650367, + "loss": 0.0776, + "step": 14247 + }, + { + "epoch": 0.12367948194894142, + "grad_norm": 0.375, + "learning_rate": 0.0019150147246917875, + "loss": 0.1152, + "step": 14248 + }, + { + "epoch": 0.12368816242914558, + "grad_norm": 0.1806640625, + "learning_rate": 0.00191500209922725, + "loss": 0.0952, + "step": 14249 + }, + { + "epoch": 0.12369684290934975, + "grad_norm": 0.5234375, + "learning_rate": 0.0019149894728714388, + "loss": 0.1069, + "step": 14250 + }, + { + "epoch": 0.12370552338955391, + "grad_norm": 0.3203125, + "learning_rate": 0.001914976845624367, + "loss": 0.1128, + "step": 14251 + }, + { + "epoch": 0.12371420386975808, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019149642174860488, + "loss": 0.1309, + "step": 14252 + }, + { + "epoch": 0.12372288434996224, + "grad_norm": 0.3984375, + "learning_rate": 0.0019149515884564982, + "loss": 0.1006, + "step": 14253 + }, + { + "epoch": 0.12373156483016641, + "grad_norm": 0.37890625, + "learning_rate": 0.0019149389585357286, + "loss": 0.1504, + "step": 14254 + }, + { + "epoch": 0.12374024531037058, + "grad_norm": 0.26171875, + "learning_rate": 0.0019149263277237541, + "loss": 0.1035, + "step": 14255 + }, + { + "epoch": 0.12374892579057474, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019149136960205887, + "loss": 0.1118, + "step": 14256 + }, + { + "epoch": 0.1237576062707789, + "grad_norm": 0.380859375, + "learning_rate": 0.0019149010634262455, + "loss": 0.1289, + "step": 14257 + }, + { + "epoch": 0.12376628675098307, + "grad_norm": 0.107421875, + "learning_rate": 0.0019148884299407386, + "loss": 0.1445, + "step": 14258 + }, + { + "epoch": 0.12377496723118724, + "grad_norm": 0.1806640625, + "learning_rate": 0.0019148757955640824, + "loss": 0.1084, + "step": 14259 + }, + { + "epoch": 0.1237836477113914, + "grad_norm": 0.80859375, + "learning_rate": 0.00191486316029629, + "loss": 0.1201, + "step": 14260 + }, + { + "epoch": 0.12379232819159557, + "grad_norm": 0.107421875, + "learning_rate": 0.0019148505241373757, + "loss": 0.1143, + "step": 14261 + }, + { + "epoch": 0.12380100867179972, + "grad_norm": 0.115234375, + "learning_rate": 0.0019148378870873528, + "loss": 0.1025, + "step": 14262 + }, + { + "epoch": 0.12380968915200388, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019148252491462358, + "loss": 0.1152, + "step": 14263 + }, + { + "epoch": 0.12381836963220805, + "grad_norm": 0.419921875, + "learning_rate": 0.001914812610314038, + "loss": 0.1328, + "step": 14264 + }, + { + "epoch": 0.12382705011241221, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019147999705907733, + "loss": 0.0742, + "step": 14265 + }, + { + "epoch": 0.12383573059261638, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019147873299764555, + "loss": 0.1328, + "step": 14266 + }, + { + "epoch": 0.12384441107282054, + "grad_norm": 0.85546875, + "learning_rate": 0.0019147746884710989, + "loss": 0.1533, + "step": 14267 + }, + { + "epoch": 0.12385309155302471, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019147620460747163, + "loss": 0.1475, + "step": 14268 + }, + { + "epoch": 0.12386177203322887, + "grad_norm": 0.44921875, + "learning_rate": 0.0019147494027873227, + "loss": 0.103, + "step": 14269 + }, + { + "epoch": 0.12387045251343304, + "grad_norm": 0.380859375, + "learning_rate": 0.0019147367586089315, + "loss": 0.0894, + "step": 14270 + }, + { + "epoch": 0.1238791329936372, + "grad_norm": 0.330078125, + "learning_rate": 0.001914724113539556, + "loss": 0.1035, + "step": 14271 + }, + { + "epoch": 0.12388781347384137, + "grad_norm": 0.64453125, + "learning_rate": 0.0019147114675792105, + "loss": 0.1602, + "step": 14272 + }, + { + "epoch": 0.12389649395404553, + "grad_norm": 0.318359375, + "learning_rate": 0.0019146988207279093, + "loss": 0.0918, + "step": 14273 + }, + { + "epoch": 0.1239051744342497, + "grad_norm": 0.240234375, + "learning_rate": 0.0019146861729856654, + "loss": 0.1172, + "step": 14274 + }, + { + "epoch": 0.12391385491445386, + "grad_norm": 0.46875, + "learning_rate": 0.0019146735243524926, + "loss": 0.0835, + "step": 14275 + }, + { + "epoch": 0.12392253539465803, + "grad_norm": 0.54296875, + "learning_rate": 0.0019146608748284054, + "loss": 0.1504, + "step": 14276 + }, + { + "epoch": 0.1239312158748622, + "grad_norm": 0.22265625, + "learning_rate": 0.0019146482244134172, + "loss": 0.1025, + "step": 14277 + }, + { + "epoch": 0.12393989635506636, + "grad_norm": 0.46875, + "learning_rate": 0.001914635573107542, + "loss": 0.0767, + "step": 14278 + }, + { + "epoch": 0.12394857683527052, + "grad_norm": 0.3515625, + "learning_rate": 0.0019146229209107935, + "loss": 0.125, + "step": 14279 + }, + { + "epoch": 0.12395725731547469, + "grad_norm": 0.1953125, + "learning_rate": 0.001914610267823186, + "loss": 0.0947, + "step": 14280 + }, + { + "epoch": 0.12396593779567885, + "grad_norm": 0.24609375, + "learning_rate": 0.0019145976138447323, + "loss": 0.1162, + "step": 14281 + }, + { + "epoch": 0.12397461827588302, + "grad_norm": 0.46875, + "learning_rate": 0.0019145849589754472, + "loss": 0.1416, + "step": 14282 + }, + { + "epoch": 0.12398329875608718, + "grad_norm": 0.53125, + "learning_rate": 0.0019145723032153443, + "loss": 0.1504, + "step": 14283 + }, + { + "epoch": 0.12399197923629135, + "grad_norm": 0.427734375, + "learning_rate": 0.0019145596465644375, + "loss": 0.1309, + "step": 14284 + }, + { + "epoch": 0.12400065971649551, + "grad_norm": 0.48046875, + "learning_rate": 0.0019145469890227403, + "loss": 0.0996, + "step": 14285 + }, + { + "epoch": 0.12400934019669968, + "grad_norm": 0.126953125, + "learning_rate": 0.0019145343305902668, + "loss": 0.1001, + "step": 14286 + }, + { + "epoch": 0.12401802067690385, + "grad_norm": 1.15625, + "learning_rate": 0.0019145216712670309, + "loss": 0.1328, + "step": 14287 + }, + { + "epoch": 0.12402670115710801, + "grad_norm": 0.1572265625, + "learning_rate": 0.001914509011053046, + "loss": 0.0928, + "step": 14288 + }, + { + "epoch": 0.12403538163731218, + "grad_norm": 0.314453125, + "learning_rate": 0.0019144963499483265, + "loss": 0.0942, + "step": 14289 + }, + { + "epoch": 0.12404406211751634, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019144836879528859, + "loss": 0.1396, + "step": 14290 + }, + { + "epoch": 0.1240527425977205, + "grad_norm": 0.06103515625, + "learning_rate": 0.0019144710250667382, + "loss": 0.0957, + "step": 14291 + }, + { + "epoch": 0.12406142307792467, + "grad_norm": 0.37109375, + "learning_rate": 0.0019144583612898975, + "loss": 0.1113, + "step": 14292 + }, + { + "epoch": 0.12407010355812884, + "grad_norm": 0.330078125, + "learning_rate": 0.0019144456966223773, + "loss": 0.1123, + "step": 14293 + }, + { + "epoch": 0.124078784038333, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019144330310641915, + "loss": 0.1133, + "step": 14294 + }, + { + "epoch": 0.12408746451853717, + "grad_norm": 0.796875, + "learning_rate": 0.0019144203646153538, + "loss": 0.0898, + "step": 14295 + }, + { + "epoch": 0.12409614499874133, + "grad_norm": 1.2734375, + "learning_rate": 0.0019144076972758784, + "loss": 0.1289, + "step": 14296 + }, + { + "epoch": 0.1241048254789455, + "grad_norm": 0.33203125, + "learning_rate": 0.0019143950290457787, + "loss": 0.1338, + "step": 14297 + }, + { + "epoch": 0.12411350595914966, + "grad_norm": 0.37890625, + "learning_rate": 0.001914382359925069, + "loss": 0.0845, + "step": 14298 + }, + { + "epoch": 0.12412218643935383, + "grad_norm": 0.365234375, + "learning_rate": 0.001914369689913763, + "loss": 0.1099, + "step": 14299 + }, + { + "epoch": 0.12413086691955799, + "grad_norm": 0.376953125, + "learning_rate": 0.0019143570190118748, + "loss": 0.0752, + "step": 14300 + }, + { + "epoch": 0.12413954739976216, + "grad_norm": 0.3359375, + "learning_rate": 0.0019143443472194178, + "loss": 0.1064, + "step": 14301 + }, + { + "epoch": 0.12414822787996632, + "grad_norm": 0.16015625, + "learning_rate": 0.001914331674536406, + "loss": 0.1108, + "step": 14302 + }, + { + "epoch": 0.12415690836017049, + "grad_norm": 0.287109375, + "learning_rate": 0.0019143190009628533, + "loss": 0.1748, + "step": 14303 + }, + { + "epoch": 0.12416558884037465, + "grad_norm": 0.61328125, + "learning_rate": 0.0019143063264987735, + "loss": 0.124, + "step": 14304 + }, + { + "epoch": 0.12417426932057882, + "grad_norm": 0.17578125, + "learning_rate": 0.0019142936511441805, + "loss": 0.1318, + "step": 14305 + }, + { + "epoch": 0.12418294980078298, + "grad_norm": 0.37890625, + "learning_rate": 0.0019142809748990883, + "loss": 0.1006, + "step": 14306 + }, + { + "epoch": 0.12419163028098715, + "grad_norm": 0.828125, + "learning_rate": 0.0019142682977635111, + "loss": 0.1719, + "step": 14307 + }, + { + "epoch": 0.12420031076119131, + "grad_norm": 0.2890625, + "learning_rate": 0.0019142556197374616, + "loss": 0.1279, + "step": 14308 + }, + { + "epoch": 0.12420899124139548, + "grad_norm": 0.74609375, + "learning_rate": 0.001914242940820955, + "loss": 0.1426, + "step": 14309 + }, + { + "epoch": 0.12421767172159964, + "grad_norm": 0.404296875, + "learning_rate": 0.001914230261014004, + "loss": 0.1172, + "step": 14310 + }, + { + "epoch": 0.12422635220180381, + "grad_norm": 0.28125, + "learning_rate": 0.0019142175803166233, + "loss": 0.1182, + "step": 14311 + }, + { + "epoch": 0.12423503268200797, + "grad_norm": 0.51953125, + "learning_rate": 0.0019142048987288265, + "loss": 0.1279, + "step": 14312 + }, + { + "epoch": 0.12424371316221214, + "grad_norm": 0.45703125, + "learning_rate": 0.0019141922162506275, + "loss": 0.1455, + "step": 14313 + }, + { + "epoch": 0.1242523936424163, + "grad_norm": 0.30859375, + "learning_rate": 0.00191417953288204, + "loss": 0.1416, + "step": 14314 + }, + { + "epoch": 0.12426107412262047, + "grad_norm": 0.25390625, + "learning_rate": 0.0019141668486230782, + "loss": 0.1426, + "step": 14315 + }, + { + "epoch": 0.12426975460282463, + "grad_norm": 0.478515625, + "learning_rate": 0.0019141541634737555, + "loss": 0.1143, + "step": 14316 + }, + { + "epoch": 0.1242784350830288, + "grad_norm": 0.07275390625, + "learning_rate": 0.0019141414774340863, + "loss": 0.0884, + "step": 14317 + }, + { + "epoch": 0.12428711556323296, + "grad_norm": 0.06640625, + "learning_rate": 0.001914128790504084, + "loss": 0.0864, + "step": 14318 + }, + { + "epoch": 0.12429579604343713, + "grad_norm": 0.8125, + "learning_rate": 0.0019141161026837627, + "loss": 0.1523, + "step": 14319 + }, + { + "epoch": 0.1243044765236413, + "grad_norm": 0.66796875, + "learning_rate": 0.001914103413973136, + "loss": 0.168, + "step": 14320 + }, + { + "epoch": 0.12431315700384546, + "grad_norm": 0.703125, + "learning_rate": 0.0019140907243722185, + "loss": 0.1387, + "step": 14321 + }, + { + "epoch": 0.12432183748404962, + "grad_norm": 0.7109375, + "learning_rate": 0.0019140780338810237, + "loss": 0.1182, + "step": 14322 + }, + { + "epoch": 0.12433051796425379, + "grad_norm": 0.451171875, + "learning_rate": 0.0019140653424995651, + "loss": 0.167, + "step": 14323 + }, + { + "epoch": 0.12433919844445794, + "grad_norm": 0.1318359375, + "learning_rate": 0.001914052650227857, + "loss": 0.1152, + "step": 14324 + }, + { + "epoch": 0.1243478789246621, + "grad_norm": 0.984375, + "learning_rate": 0.001914039957065913, + "loss": 0.1025, + "step": 14325 + }, + { + "epoch": 0.12435655940486627, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019140272630137475, + "loss": 0.0928, + "step": 14326 + }, + { + "epoch": 0.12436523988507044, + "grad_norm": 0.28515625, + "learning_rate": 0.0019140145680713737, + "loss": 0.0879, + "step": 14327 + }, + { + "epoch": 0.1243739203652746, + "grad_norm": 0.58203125, + "learning_rate": 0.0019140018722388057, + "loss": 0.1211, + "step": 14328 + }, + { + "epoch": 0.12438260084547877, + "grad_norm": 0.126953125, + "learning_rate": 0.001913989175516058, + "loss": 0.0996, + "step": 14329 + }, + { + "epoch": 0.12439128132568293, + "grad_norm": 0.201171875, + "learning_rate": 0.0019139764779031432, + "loss": 0.1309, + "step": 14330 + }, + { + "epoch": 0.1243999618058871, + "grad_norm": 2.546875, + "learning_rate": 0.0019139637794000765, + "loss": 0.1572, + "step": 14331 + }, + { + "epoch": 0.12440864228609126, + "grad_norm": 0.42578125, + "learning_rate": 0.0019139510800068708, + "loss": 0.1143, + "step": 14332 + }, + { + "epoch": 0.12441732276629543, + "grad_norm": 0.2001953125, + "learning_rate": 0.001913938379723541, + "loss": 0.168, + "step": 14333 + }, + { + "epoch": 0.12442600324649959, + "grad_norm": 0.703125, + "learning_rate": 0.0019139256785501004, + "loss": 0.1299, + "step": 14334 + }, + { + "epoch": 0.12443468372670376, + "grad_norm": 0.25, + "learning_rate": 0.0019139129764865624, + "loss": 0.1143, + "step": 14335 + }, + { + "epoch": 0.12444336420690792, + "grad_norm": 0.3671875, + "learning_rate": 0.0019139002735329414, + "loss": 0.1641, + "step": 14336 + }, + { + "epoch": 0.12445204468711209, + "grad_norm": 0.84765625, + "learning_rate": 0.0019138875696892517, + "loss": 0.1162, + "step": 14337 + }, + { + "epoch": 0.12446072516731625, + "grad_norm": 0.4609375, + "learning_rate": 0.0019138748649555066, + "loss": 0.1211, + "step": 14338 + }, + { + "epoch": 0.12446940564752042, + "grad_norm": 0.341796875, + "learning_rate": 0.0019138621593317202, + "loss": 0.1099, + "step": 14339 + }, + { + "epoch": 0.12447808612772458, + "grad_norm": 0.44921875, + "learning_rate": 0.0019138494528179065, + "loss": 0.1328, + "step": 14340 + }, + { + "epoch": 0.12448676660792875, + "grad_norm": 0.41796875, + "learning_rate": 0.0019138367454140791, + "loss": 0.21, + "step": 14341 + }, + { + "epoch": 0.12449544708813291, + "grad_norm": 0.2265625, + "learning_rate": 0.001913824037120252, + "loss": 0.1211, + "step": 14342 + }, + { + "epoch": 0.12450412756833708, + "grad_norm": 0.83984375, + "learning_rate": 0.0019138113279364394, + "loss": 0.1182, + "step": 14343 + }, + { + "epoch": 0.12451280804854124, + "grad_norm": 0.27734375, + "learning_rate": 0.0019137986178626547, + "loss": 0.1543, + "step": 14344 + }, + { + "epoch": 0.12452148852874541, + "grad_norm": 0.12060546875, + "learning_rate": 0.0019137859068989122, + "loss": 0.0952, + "step": 14345 + }, + { + "epoch": 0.12453016900894957, + "grad_norm": 0.8046875, + "learning_rate": 0.0019137731950452255, + "loss": 0.1348, + "step": 14346 + }, + { + "epoch": 0.12453884948915374, + "grad_norm": 0.52734375, + "learning_rate": 0.0019137604823016088, + "loss": 0.125, + "step": 14347 + }, + { + "epoch": 0.1245475299693579, + "grad_norm": 0.7890625, + "learning_rate": 0.001913747768668076, + "loss": 0.1177, + "step": 14348 + }, + { + "epoch": 0.12455621044956207, + "grad_norm": 0.53515625, + "learning_rate": 0.0019137350541446403, + "loss": 0.083, + "step": 14349 + }, + { + "epoch": 0.12456489092976623, + "grad_norm": 0.91796875, + "learning_rate": 0.001913722338731317, + "loss": 0.1328, + "step": 14350 + }, + { + "epoch": 0.1245735714099704, + "grad_norm": 0.6015625, + "learning_rate": 0.0019137096224281186, + "loss": 0.1094, + "step": 14351 + }, + { + "epoch": 0.12458225189017456, + "grad_norm": 0.298828125, + "learning_rate": 0.00191369690523506, + "loss": 0.1177, + "step": 14352 + }, + { + "epoch": 0.12459093237037873, + "grad_norm": 0.451171875, + "learning_rate": 0.0019136841871521541, + "loss": 0.0898, + "step": 14353 + }, + { + "epoch": 0.1245996128505829, + "grad_norm": 0.1083984375, + "learning_rate": 0.0019136714681794158, + "loss": 0.084, + "step": 14354 + }, + { + "epoch": 0.12460829333078706, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019136587483168585, + "loss": 0.1006, + "step": 14355 + }, + { + "epoch": 0.12461697381099122, + "grad_norm": 0.36328125, + "learning_rate": 0.0019136460275644962, + "loss": 0.0981, + "step": 14356 + }, + { + "epoch": 0.12462565429119539, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019136333059223429, + "loss": 0.0825, + "step": 14357 + }, + { + "epoch": 0.12463433477139955, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019136205833904126, + "loss": 0.103, + "step": 14358 + }, + { + "epoch": 0.12464301525160372, + "grad_norm": 0.76171875, + "learning_rate": 0.001913607859968719, + "loss": 0.1172, + "step": 14359 + }, + { + "epoch": 0.12465169573180788, + "grad_norm": 0.408203125, + "learning_rate": 0.001913595135657276, + "loss": 0.1016, + "step": 14360 + }, + { + "epoch": 0.12466037621201205, + "grad_norm": 0.318359375, + "learning_rate": 0.0019135824104560973, + "loss": 0.0879, + "step": 14361 + }, + { + "epoch": 0.12466905669221622, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019135696843651975, + "loss": 0.124, + "step": 14362 + }, + { + "epoch": 0.12467773717242038, + "grad_norm": 0.25390625, + "learning_rate": 0.0019135569573845899, + "loss": 0.083, + "step": 14363 + }, + { + "epoch": 0.12468641765262455, + "grad_norm": 0.44921875, + "learning_rate": 0.001913544229514289, + "loss": 0.1562, + "step": 14364 + }, + { + "epoch": 0.12469509813282871, + "grad_norm": 0.54296875, + "learning_rate": 0.0019135315007543078, + "loss": 0.1104, + "step": 14365 + }, + { + "epoch": 0.12470377861303288, + "grad_norm": 0.7109375, + "learning_rate": 0.001913518771104661, + "loss": 0.0811, + "step": 14366 + }, + { + "epoch": 0.12471245909323704, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019135060405653624, + "loss": 0.0771, + "step": 14367 + }, + { + "epoch": 0.1247211395734412, + "grad_norm": 0.384765625, + "learning_rate": 0.0019134933091364257, + "loss": 0.1201, + "step": 14368 + }, + { + "epoch": 0.12472982005364537, + "grad_norm": 1.09375, + "learning_rate": 0.001913480576817865, + "loss": 0.125, + "step": 14369 + }, + { + "epoch": 0.12473850053384954, + "grad_norm": 0.1337890625, + "learning_rate": 0.001913467843609694, + "loss": 0.1094, + "step": 14370 + }, + { + "epoch": 0.1247471810140537, + "grad_norm": 0.2255859375, + "learning_rate": 0.001913455109511927, + "loss": 0.1133, + "step": 14371 + }, + { + "epoch": 0.12475586149425787, + "grad_norm": 0.3359375, + "learning_rate": 0.0019134423745245774, + "loss": 0.1104, + "step": 14372 + }, + { + "epoch": 0.12476454197446203, + "grad_norm": 0.365234375, + "learning_rate": 0.0019134296386476596, + "loss": 0.0884, + "step": 14373 + }, + { + "epoch": 0.1247732224546662, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019134169018811878, + "loss": 0.126, + "step": 14374 + }, + { + "epoch": 0.12478190293487036, + "grad_norm": 0.3515625, + "learning_rate": 0.001913404164225175, + "loss": 0.1221, + "step": 14375 + }, + { + "epoch": 0.12479058341507453, + "grad_norm": 0.1240234375, + "learning_rate": 0.0019133914256796356, + "loss": 0.1016, + "step": 14376 + }, + { + "epoch": 0.12479926389527869, + "grad_norm": 0.498046875, + "learning_rate": 0.001913378686244584, + "loss": 0.0957, + "step": 14377 + }, + { + "epoch": 0.12480794437548286, + "grad_norm": 0.4375, + "learning_rate": 0.0019133659459200335, + "loss": 0.1416, + "step": 14378 + }, + { + "epoch": 0.12481662485568702, + "grad_norm": 0.1845703125, + "learning_rate": 0.001913353204705998, + "loss": 0.1094, + "step": 14379 + }, + { + "epoch": 0.12482530533589119, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019133404626024918, + "loss": 0.0854, + "step": 14380 + }, + { + "epoch": 0.12483398581609535, + "grad_norm": 0.36328125, + "learning_rate": 0.0019133277196095285, + "loss": 0.1069, + "step": 14381 + }, + { + "epoch": 0.12484266629629952, + "grad_norm": 0.466796875, + "learning_rate": 0.0019133149757271227, + "loss": 0.1123, + "step": 14382 + }, + { + "epoch": 0.12485134677650368, + "grad_norm": 0.30859375, + "learning_rate": 0.0019133022309552874, + "loss": 0.1094, + "step": 14383 + }, + { + "epoch": 0.12486002725670785, + "grad_norm": 0.45703125, + "learning_rate": 0.0019132894852940373, + "loss": 0.1328, + "step": 14384 + }, + { + "epoch": 0.124868707736912, + "grad_norm": 0.1875, + "learning_rate": 0.001913276738743386, + "loss": 0.1113, + "step": 14385 + }, + { + "epoch": 0.12487738821711616, + "grad_norm": 0.373046875, + "learning_rate": 0.0019132639913033473, + "loss": 0.1455, + "step": 14386 + }, + { + "epoch": 0.12488606869732033, + "grad_norm": 0.39453125, + "learning_rate": 0.0019132512429739354, + "loss": 0.0889, + "step": 14387 + }, + { + "epoch": 0.1248947491775245, + "grad_norm": 0.404296875, + "learning_rate": 0.0019132384937551642, + "loss": 0.1094, + "step": 14388 + }, + { + "epoch": 0.12490342965772866, + "grad_norm": 0.4140625, + "learning_rate": 0.0019132257436470475, + "loss": 0.1128, + "step": 14389 + }, + { + "epoch": 0.12491211013793282, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019132129926495995, + "loss": 0.1128, + "step": 14390 + }, + { + "epoch": 0.12492079061813699, + "grad_norm": 1.0234375, + "learning_rate": 0.0019132002407628338, + "loss": 0.1396, + "step": 14391 + }, + { + "epoch": 0.12492947109834115, + "grad_norm": 0.65234375, + "learning_rate": 0.0019131874879867645, + "loss": 0.1562, + "step": 14392 + }, + { + "epoch": 0.12493815157854532, + "grad_norm": 0.63671875, + "learning_rate": 0.001913174734321406, + "loss": 0.1885, + "step": 14393 + }, + { + "epoch": 0.12494683205874949, + "grad_norm": 0.111328125, + "learning_rate": 0.0019131619797667716, + "loss": 0.1348, + "step": 14394 + }, + { + "epoch": 0.12495551253895365, + "grad_norm": 0.291015625, + "learning_rate": 0.0019131492243228752, + "loss": 0.0977, + "step": 14395 + }, + { + "epoch": 0.12496419301915782, + "grad_norm": 1.0625, + "learning_rate": 0.0019131364679897313, + "loss": 0.083, + "step": 14396 + }, + { + "epoch": 0.12497287349936198, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019131237107673537, + "loss": 0.106, + "step": 14397 + }, + { + "epoch": 0.12498155397956615, + "grad_norm": 0.3203125, + "learning_rate": 0.001913110952655756, + "loss": 0.1455, + "step": 14398 + }, + { + "epoch": 0.12499023445977031, + "grad_norm": 0.1220703125, + "learning_rate": 0.0019130981936549523, + "loss": 0.1235, + "step": 14399 + }, + { + "epoch": 0.12499891493997448, + "grad_norm": 0.162109375, + "learning_rate": 0.001913085433764957, + "loss": 0.1279, + "step": 14400 + }, + { + "epoch": 0.12500759542017864, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019130726729857835, + "loss": 0.1143, + "step": 14401 + }, + { + "epoch": 0.1250162759003828, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019130599113174457, + "loss": 0.1162, + "step": 14402 + }, + { + "epoch": 0.12502495638058697, + "grad_norm": 0.51953125, + "learning_rate": 0.001913047148759958, + "loss": 0.1064, + "step": 14403 + }, + { + "epoch": 0.12503363686079114, + "grad_norm": 0.458984375, + "learning_rate": 0.0019130343853133343, + "loss": 0.1201, + "step": 14404 + }, + { + "epoch": 0.1250423173409953, + "grad_norm": 0.384765625, + "learning_rate": 0.0019130216209775881, + "loss": 0.1113, + "step": 14405 + }, + { + "epoch": 0.12505099782119947, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019130088557527338, + "loss": 0.125, + "step": 14406 + }, + { + "epoch": 0.12505967830140363, + "grad_norm": 0.1884765625, + "learning_rate": 0.0019129960896387855, + "loss": 0.0947, + "step": 14407 + }, + { + "epoch": 0.1250683587816078, + "grad_norm": 0.1796875, + "learning_rate": 0.0019129833226357565, + "loss": 0.1191, + "step": 14408 + }, + { + "epoch": 0.12507703926181196, + "grad_norm": 0.94921875, + "learning_rate": 0.0019129705547436613, + "loss": 0.0967, + "step": 14409 + }, + { + "epoch": 0.12508571974201613, + "grad_norm": 0.63671875, + "learning_rate": 0.0019129577859625137, + "loss": 0.1113, + "step": 14410 + }, + { + "epoch": 0.1250944002222203, + "grad_norm": 0.447265625, + "learning_rate": 0.0019129450162923276, + "loss": 0.0776, + "step": 14411 + }, + { + "epoch": 0.12510308070242446, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019129322457331174, + "loss": 0.0898, + "step": 14412 + }, + { + "epoch": 0.12511176118262862, + "grad_norm": 0.267578125, + "learning_rate": 0.0019129194742848966, + "loss": 0.0957, + "step": 14413 + }, + { + "epoch": 0.1251204416628328, + "grad_norm": 0.58203125, + "learning_rate": 0.0019129067019476788, + "loss": 0.1436, + "step": 14414 + }, + { + "epoch": 0.12512912214303695, + "grad_norm": 0.1357421875, + "learning_rate": 0.001912893928721479, + "loss": 0.1445, + "step": 14415 + }, + { + "epoch": 0.12513780262324112, + "grad_norm": 0.61328125, + "learning_rate": 0.0019128811546063103, + "loss": 0.085, + "step": 14416 + }, + { + "epoch": 0.12514648310344528, + "grad_norm": 0.72265625, + "learning_rate": 0.001912868379602187, + "loss": 0.1602, + "step": 14417 + }, + { + "epoch": 0.12515516358364945, + "grad_norm": 0.310546875, + "learning_rate": 0.0019128556037091233, + "loss": 0.1123, + "step": 14418 + }, + { + "epoch": 0.1251638440638536, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019128428269271326, + "loss": 0.1211, + "step": 14419 + }, + { + "epoch": 0.12517252454405778, + "grad_norm": 0.96484375, + "learning_rate": 0.0019128300492562294, + "loss": 0.1221, + "step": 14420 + }, + { + "epoch": 0.12518120502426194, + "grad_norm": 0.423828125, + "learning_rate": 0.0019128172706964273, + "loss": 0.1191, + "step": 14421 + }, + { + "epoch": 0.1251898855044661, + "grad_norm": 0.921875, + "learning_rate": 0.0019128044912477408, + "loss": 0.0928, + "step": 14422 + }, + { + "epoch": 0.12519856598467027, + "grad_norm": 0.265625, + "learning_rate": 0.0019127917109101832, + "loss": 0.1021, + "step": 14423 + }, + { + "epoch": 0.12520724646487444, + "grad_norm": 0.1484375, + "learning_rate": 0.001912778929683769, + "loss": 0.0977, + "step": 14424 + }, + { + "epoch": 0.1252159269450786, + "grad_norm": 0.384765625, + "learning_rate": 0.0019127661475685117, + "loss": 0.127, + "step": 14425 + }, + { + "epoch": 0.12522460742528277, + "grad_norm": 0.302734375, + "learning_rate": 0.0019127533645644258, + "loss": 0.1299, + "step": 14426 + }, + { + "epoch": 0.12523328790548693, + "grad_norm": 0.298828125, + "learning_rate": 0.0019127405806715248, + "loss": 0.1191, + "step": 14427 + }, + { + "epoch": 0.1252419683856911, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019127277958898231, + "loss": 0.1406, + "step": 14428 + }, + { + "epoch": 0.12525064886589526, + "grad_norm": 0.8828125, + "learning_rate": 0.0019127150102193344, + "loss": 0.2129, + "step": 14429 + }, + { + "epoch": 0.12525932934609943, + "grad_norm": 0.11279296875, + "learning_rate": 0.001912702223660073, + "loss": 0.1133, + "step": 14430 + }, + { + "epoch": 0.1252680098263036, + "grad_norm": 0.421875, + "learning_rate": 0.0019126894362120523, + "loss": 0.1006, + "step": 14431 + }, + { + "epoch": 0.12527669030650776, + "grad_norm": 0.7890625, + "learning_rate": 0.001912676647875287, + "loss": 0.1963, + "step": 14432 + }, + { + "epoch": 0.12528537078671192, + "grad_norm": 0.142578125, + "learning_rate": 0.0019126638586497904, + "loss": 0.1182, + "step": 14433 + }, + { + "epoch": 0.1252940512669161, + "grad_norm": 0.10009765625, + "learning_rate": 0.001912651068535577, + "loss": 0.0923, + "step": 14434 + }, + { + "epoch": 0.12530273174712026, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019126382775326604, + "loss": 0.1182, + "step": 14435 + }, + { + "epoch": 0.12531141222732442, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019126254856410554, + "loss": 0.1089, + "step": 14436 + }, + { + "epoch": 0.12532009270752859, + "grad_norm": 0.453125, + "learning_rate": 0.0019126126928607746, + "loss": 0.1055, + "step": 14437 + }, + { + "epoch": 0.12532877318773275, + "grad_norm": 0.43359375, + "learning_rate": 0.0019125998991918333, + "loss": 0.0825, + "step": 14438 + }, + { + "epoch": 0.12533745366793692, + "grad_norm": 0.130859375, + "learning_rate": 0.0019125871046342447, + "loss": 0.1235, + "step": 14439 + }, + { + "epoch": 0.12534613414814108, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019125743091880233, + "loss": 0.082, + "step": 14440 + }, + { + "epoch": 0.12535481462834525, + "grad_norm": 0.279296875, + "learning_rate": 0.0019125615128531826, + "loss": 0.1035, + "step": 14441 + }, + { + "epoch": 0.1253634951085494, + "grad_norm": 0.1044921875, + "learning_rate": 0.001912548715629737, + "loss": 0.1167, + "step": 14442 + }, + { + "epoch": 0.12537217558875358, + "grad_norm": 0.09765625, + "learning_rate": 0.0019125359175177, + "loss": 0.1328, + "step": 14443 + }, + { + "epoch": 0.12538085606895774, + "grad_norm": 0.27734375, + "learning_rate": 0.001912523118517086, + "loss": 0.123, + "step": 14444 + }, + { + "epoch": 0.1253895365491619, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019125103186279093, + "loss": 0.0928, + "step": 14445 + }, + { + "epoch": 0.12539821702936607, + "grad_norm": 0.451171875, + "learning_rate": 0.0019124975178501835, + "loss": 0.1738, + "step": 14446 + }, + { + "epoch": 0.12540689750957024, + "grad_norm": 0.0947265625, + "learning_rate": 0.0019124847161839226, + "loss": 0.1064, + "step": 14447 + }, + { + "epoch": 0.1254155779897744, + "grad_norm": 0.58203125, + "learning_rate": 0.0019124719136291402, + "loss": 0.1426, + "step": 14448 + }, + { + "epoch": 0.12542425846997857, + "grad_norm": 0.3046875, + "learning_rate": 0.001912459110185851, + "loss": 0.0952, + "step": 14449 + }, + { + "epoch": 0.12543293895018273, + "grad_norm": 0.4375, + "learning_rate": 0.0019124463058540686, + "loss": 0.1108, + "step": 14450 + }, + { + "epoch": 0.1254416194303869, + "grad_norm": 0.51953125, + "learning_rate": 0.0019124335006338073, + "loss": 0.1523, + "step": 14451 + }, + { + "epoch": 0.12545029991059106, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019124206945250809, + "loss": 0.0859, + "step": 14452 + }, + { + "epoch": 0.12545898039079523, + "grad_norm": 0.162109375, + "learning_rate": 0.0019124078875279035, + "loss": 0.1318, + "step": 14453 + }, + { + "epoch": 0.1254676608709994, + "grad_norm": 0.078125, + "learning_rate": 0.001912395079642289, + "loss": 0.0928, + "step": 14454 + }, + { + "epoch": 0.12547634135120356, + "grad_norm": 0.4609375, + "learning_rate": 0.001912382270868251, + "loss": 0.1309, + "step": 14455 + }, + { + "epoch": 0.12548502183140772, + "grad_norm": 0.265625, + "learning_rate": 0.0019123694612058046, + "loss": 0.0938, + "step": 14456 + }, + { + "epoch": 0.1254937023116119, + "grad_norm": 0.162109375, + "learning_rate": 0.0019123566506549629, + "loss": 0.1016, + "step": 14457 + }, + { + "epoch": 0.12550238279181605, + "grad_norm": 0.0908203125, + "learning_rate": 0.00191234383921574, + "loss": 0.0967, + "step": 14458 + }, + { + "epoch": 0.12551106327202022, + "grad_norm": 0.240234375, + "learning_rate": 0.0019123310268881503, + "loss": 0.1187, + "step": 14459 + }, + { + "epoch": 0.12551974375222438, + "grad_norm": 0.30859375, + "learning_rate": 0.0019123182136722074, + "loss": 0.1387, + "step": 14460 + }, + { + "epoch": 0.12552842423242855, + "grad_norm": 0.126953125, + "learning_rate": 0.0019123053995679258, + "loss": 0.103, + "step": 14461 + }, + { + "epoch": 0.1255371047126327, + "grad_norm": 0.283203125, + "learning_rate": 0.0019122925845753193, + "loss": 0.1504, + "step": 14462 + }, + { + "epoch": 0.12554578519283688, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019122797686944013, + "loss": 0.1221, + "step": 14463 + }, + { + "epoch": 0.12555446567304104, + "grad_norm": 0.361328125, + "learning_rate": 0.0019122669519251868, + "loss": 0.103, + "step": 14464 + }, + { + "epoch": 0.1255631461532452, + "grad_norm": 0.166015625, + "learning_rate": 0.0019122541342676894, + "loss": 0.127, + "step": 14465 + }, + { + "epoch": 0.12557182663344937, + "grad_norm": 0.431640625, + "learning_rate": 0.0019122413157219229, + "loss": 0.1357, + "step": 14466 + }, + { + "epoch": 0.12558050711365354, + "grad_norm": 0.291015625, + "learning_rate": 0.0019122284962879014, + "loss": 0.1514, + "step": 14467 + }, + { + "epoch": 0.1255891875938577, + "grad_norm": 0.58203125, + "learning_rate": 0.001912215675965639, + "loss": 0.1211, + "step": 14468 + }, + { + "epoch": 0.12559786807406187, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019122028547551497, + "loss": 0.1748, + "step": 14469 + }, + { + "epoch": 0.12560654855426603, + "grad_norm": 0.392578125, + "learning_rate": 0.0019121900326564479, + "loss": 0.1221, + "step": 14470 + }, + { + "epoch": 0.1256152290344702, + "grad_norm": 0.4375, + "learning_rate": 0.001912177209669547, + "loss": 0.165, + "step": 14471 + }, + { + "epoch": 0.12562390951467436, + "grad_norm": 0.53515625, + "learning_rate": 0.0019121643857944617, + "loss": 0.1299, + "step": 14472 + }, + { + "epoch": 0.12563258999487853, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019121515610312052, + "loss": 0.1484, + "step": 14473 + }, + { + "epoch": 0.1256412704750827, + "grad_norm": 1.296875, + "learning_rate": 0.0019121387353797923, + "loss": 0.1348, + "step": 14474 + }, + { + "epoch": 0.12564995095528686, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019121259088402363, + "loss": 0.1157, + "step": 14475 + }, + { + "epoch": 0.12565863143549102, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019121130814125518, + "loss": 0.1011, + "step": 14476 + }, + { + "epoch": 0.1256673119156952, + "grad_norm": 0.08984375, + "learning_rate": 0.0019121002530967527, + "loss": 0.127, + "step": 14477 + }, + { + "epoch": 0.12567599239589933, + "grad_norm": 0.8203125, + "learning_rate": 0.001912087423892853, + "loss": 0.1025, + "step": 14478 + }, + { + "epoch": 0.1256846728761035, + "grad_norm": 0.11572265625, + "learning_rate": 0.0019120745938008663, + "loss": 0.0698, + "step": 14479 + }, + { + "epoch": 0.12569335335630766, + "grad_norm": 0.099609375, + "learning_rate": 0.0019120617628208077, + "loss": 0.1387, + "step": 14480 + }, + { + "epoch": 0.12570203383651182, + "grad_norm": 0.48828125, + "learning_rate": 0.00191204893095269, + "loss": 0.127, + "step": 14481 + }, + { + "epoch": 0.125710714316716, + "grad_norm": 0.177734375, + "learning_rate": 0.001912036098196528, + "loss": 0.1748, + "step": 14482 + }, + { + "epoch": 0.12571939479692015, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019120232645523354, + "loss": 0.1309, + "step": 14483 + }, + { + "epoch": 0.12572807527712432, + "grad_norm": 0.125, + "learning_rate": 0.0019120104300201267, + "loss": 0.1348, + "step": 14484 + }, + { + "epoch": 0.12573675575732848, + "grad_norm": 0.53515625, + "learning_rate": 0.001911997594599915, + "loss": 0.1299, + "step": 14485 + }, + { + "epoch": 0.12574543623753265, + "grad_norm": 0.224609375, + "learning_rate": 0.0019119847582917153, + "loss": 0.0913, + "step": 14486 + }, + { + "epoch": 0.1257541167177368, + "grad_norm": 0.453125, + "learning_rate": 0.0019119719210955414, + "loss": 0.1348, + "step": 14487 + }, + { + "epoch": 0.12576279719794098, + "grad_norm": 0.1123046875, + "learning_rate": 0.001911959083011407, + "loss": 0.1025, + "step": 14488 + }, + { + "epoch": 0.12577147767814514, + "grad_norm": 0.3984375, + "learning_rate": 0.0019119462440393262, + "loss": 0.1641, + "step": 14489 + }, + { + "epoch": 0.1257801581583493, + "grad_norm": 0.1484375, + "learning_rate": 0.0019119334041793135, + "loss": 0.0635, + "step": 14490 + }, + { + "epoch": 0.12578883863855347, + "grad_norm": 0.498046875, + "learning_rate": 0.0019119205634313826, + "loss": 0.1064, + "step": 14491 + }, + { + "epoch": 0.12579751911875764, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019119077217955474, + "loss": 0.1416, + "step": 14492 + }, + { + "epoch": 0.1258061995989618, + "grad_norm": 0.27734375, + "learning_rate": 0.0019118948792718222, + "loss": 0.1182, + "step": 14493 + }, + { + "epoch": 0.12581488007916597, + "grad_norm": 0.423828125, + "learning_rate": 0.0019118820358602212, + "loss": 0.0854, + "step": 14494 + }, + { + "epoch": 0.12582356055937013, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019118691915607578, + "loss": 0.1084, + "step": 14495 + }, + { + "epoch": 0.1258322410395743, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019118563463734464, + "loss": 0.1191, + "step": 14496 + }, + { + "epoch": 0.12584092151977846, + "grad_norm": 0.259765625, + "learning_rate": 0.0019118435002983016, + "loss": 0.1206, + "step": 14497 + }, + { + "epoch": 0.12584960199998263, + "grad_norm": 0.2578125, + "learning_rate": 0.0019118306533353368, + "loss": 0.1348, + "step": 14498 + }, + { + "epoch": 0.1258582824801868, + "grad_norm": 0.2109375, + "learning_rate": 0.0019118178054845657, + "loss": 0.0957, + "step": 14499 + }, + { + "epoch": 0.12586696296039096, + "grad_norm": 0.2119140625, + "learning_rate": 0.0019118049567460034, + "loss": 0.1299, + "step": 14500 + }, + { + "epoch": 0.12587564344059513, + "grad_norm": 0.259765625, + "learning_rate": 0.0019117921071196632, + "loss": 0.1621, + "step": 14501 + }, + { + "epoch": 0.1258843239207993, + "grad_norm": 0.294921875, + "learning_rate": 0.0019117792566055592, + "loss": 0.1069, + "step": 14502 + }, + { + "epoch": 0.12589300440100346, + "grad_norm": 0.140625, + "learning_rate": 0.0019117664052037057, + "loss": 0.1162, + "step": 14503 + }, + { + "epoch": 0.12590168488120762, + "grad_norm": 0.154296875, + "learning_rate": 0.0019117535529141168, + "loss": 0.1328, + "step": 14504 + }, + { + "epoch": 0.12591036536141179, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019117406997368064, + "loss": 0.127, + "step": 14505 + }, + { + "epoch": 0.12591904584161595, + "grad_norm": 0.1533203125, + "learning_rate": 0.0019117278456717885, + "loss": 0.1416, + "step": 14506 + }, + { + "epoch": 0.12592772632182012, + "grad_norm": 0.310546875, + "learning_rate": 0.0019117149907190772, + "loss": 0.126, + "step": 14507 + }, + { + "epoch": 0.12593640680202428, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019117021348786865, + "loss": 0.1592, + "step": 14508 + }, + { + "epoch": 0.12594508728222845, + "grad_norm": 0.57421875, + "learning_rate": 0.0019116892781506306, + "loss": 0.1621, + "step": 14509 + }, + { + "epoch": 0.1259537677624326, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019116764205349235, + "loss": 0.1289, + "step": 14510 + }, + { + "epoch": 0.12596244824263678, + "grad_norm": 0.57421875, + "learning_rate": 0.0019116635620315793, + "loss": 0.1367, + "step": 14511 + }, + { + "epoch": 0.12597112872284094, + "grad_norm": 0.5, + "learning_rate": 0.0019116507026406121, + "loss": 0.1011, + "step": 14512 + }, + { + "epoch": 0.1259798092030451, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019116378423620355, + "loss": 0.1152, + "step": 14513 + }, + { + "epoch": 0.12598848968324927, + "grad_norm": 0.1748046875, + "learning_rate": 0.0019116249811958644, + "loss": 0.1035, + "step": 14514 + }, + { + "epoch": 0.12599717016345344, + "grad_norm": 0.294921875, + "learning_rate": 0.0019116121191421122, + "loss": 0.1084, + "step": 14515 + }, + { + "epoch": 0.1260058506436576, + "grad_norm": 0.56640625, + "learning_rate": 0.0019115992562007934, + "loss": 0.1523, + "step": 14516 + }, + { + "epoch": 0.12601453112386177, + "grad_norm": 0.220703125, + "learning_rate": 0.0019115863923719217, + "loss": 0.1328, + "step": 14517 + }, + { + "epoch": 0.12602321160406593, + "grad_norm": 0.255859375, + "learning_rate": 0.001911573527655511, + "loss": 0.123, + "step": 14518 + }, + { + "epoch": 0.1260318920842701, + "grad_norm": 0.482421875, + "learning_rate": 0.0019115606620515762, + "loss": 0.1055, + "step": 14519 + }, + { + "epoch": 0.12604057256447426, + "grad_norm": 0.373046875, + "learning_rate": 0.0019115477955601306, + "loss": 0.1338, + "step": 14520 + }, + { + "epoch": 0.12604925304467843, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019115349281811887, + "loss": 0.105, + "step": 14521 + }, + { + "epoch": 0.1260579335248826, + "grad_norm": 0.27734375, + "learning_rate": 0.0019115220599147643, + "loss": 0.1621, + "step": 14522 + }, + { + "epoch": 0.12606661400508676, + "grad_norm": 0.310546875, + "learning_rate": 0.0019115091907608713, + "loss": 0.1123, + "step": 14523 + }, + { + "epoch": 0.12607529448529092, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019114963207195244, + "loss": 0.1069, + "step": 14524 + }, + { + "epoch": 0.1260839749654951, + "grad_norm": 0.400390625, + "learning_rate": 0.0019114834497907373, + "loss": 0.1094, + "step": 14525 + }, + { + "epoch": 0.12609265544569925, + "grad_norm": 0.09619140625, + "learning_rate": 0.001911470577974524, + "loss": 0.1187, + "step": 14526 + }, + { + "epoch": 0.12610133592590342, + "grad_norm": 0.9140625, + "learning_rate": 0.0019114577052708984, + "loss": 0.1465, + "step": 14527 + }, + { + "epoch": 0.12611001640610758, + "grad_norm": 0.3046875, + "learning_rate": 0.0019114448316798751, + "loss": 0.0874, + "step": 14528 + }, + { + "epoch": 0.12611869688631175, + "grad_norm": 0.1767578125, + "learning_rate": 0.001911431957201468, + "loss": 0.083, + "step": 14529 + }, + { + "epoch": 0.1261273773665159, + "grad_norm": 0.421875, + "learning_rate": 0.0019114190818356911, + "loss": 0.1523, + "step": 14530 + }, + { + "epoch": 0.12613605784672008, + "grad_norm": 0.279296875, + "learning_rate": 0.0019114062055825582, + "loss": 0.126, + "step": 14531 + }, + { + "epoch": 0.12614473832692424, + "grad_norm": 0.62890625, + "learning_rate": 0.0019113933284420838, + "loss": 0.1172, + "step": 14532 + }, + { + "epoch": 0.1261534188071284, + "grad_norm": 1.296875, + "learning_rate": 0.001911380450414282, + "loss": 0.1289, + "step": 14533 + }, + { + "epoch": 0.12616209928733257, + "grad_norm": 0.7109375, + "learning_rate": 0.001911367571499167, + "loss": 0.1191, + "step": 14534 + }, + { + "epoch": 0.12617077976753674, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019113546916967518, + "loss": 0.1406, + "step": 14535 + }, + { + "epoch": 0.1261794602477409, + "grad_norm": 0.244140625, + "learning_rate": 0.0019113418110070519, + "loss": 0.1162, + "step": 14536 + }, + { + "epoch": 0.12618814072794507, + "grad_norm": 0.15234375, + "learning_rate": 0.0019113289294300808, + "loss": 0.1699, + "step": 14537 + }, + { + "epoch": 0.12619682120814923, + "grad_norm": 0.349609375, + "learning_rate": 0.0019113160469658524, + "loss": 0.1143, + "step": 14538 + }, + { + "epoch": 0.1262055016883534, + "grad_norm": 0.29296875, + "learning_rate": 0.001911303163614381, + "loss": 0.1289, + "step": 14539 + }, + { + "epoch": 0.12621418216855756, + "grad_norm": 0.70703125, + "learning_rate": 0.0019112902793756805, + "loss": 0.1328, + "step": 14540 + }, + { + "epoch": 0.12622286264876173, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019112773942497655, + "loss": 0.1055, + "step": 14541 + }, + { + "epoch": 0.1262315431289659, + "grad_norm": 0.51171875, + "learning_rate": 0.0019112645082366494, + "loss": 0.1299, + "step": 14542 + }, + { + "epoch": 0.12624022360917006, + "grad_norm": 0.1845703125, + "learning_rate": 0.0019112516213363464, + "loss": 0.1064, + "step": 14543 + }, + { + "epoch": 0.12624890408937423, + "grad_norm": 0.421875, + "learning_rate": 0.0019112387335488712, + "loss": 0.1113, + "step": 14544 + }, + { + "epoch": 0.1262575845695784, + "grad_norm": 0.21875, + "learning_rate": 0.0019112258448742377, + "loss": 0.127, + "step": 14545 + }, + { + "epoch": 0.12626626504978256, + "grad_norm": 0.171875, + "learning_rate": 0.0019112129553124596, + "loss": 0.0957, + "step": 14546 + }, + { + "epoch": 0.12627494552998672, + "grad_norm": 0.34375, + "learning_rate": 0.0019112000648635512, + "loss": 0.1084, + "step": 14547 + }, + { + "epoch": 0.12628362601019089, + "grad_norm": 0.78125, + "learning_rate": 0.0019111871735275265, + "loss": 0.1475, + "step": 14548 + }, + { + "epoch": 0.12629230649039505, + "grad_norm": 0.349609375, + "learning_rate": 0.0019111742813043998, + "loss": 0.085, + "step": 14549 + }, + { + "epoch": 0.12630098697059922, + "grad_norm": 0.11328125, + "learning_rate": 0.001911161388194185, + "loss": 0.1074, + "step": 14550 + }, + { + "epoch": 0.12630966745080338, + "grad_norm": 0.150390625, + "learning_rate": 0.0019111484941968962, + "loss": 0.1025, + "step": 14551 + }, + { + "epoch": 0.12631834793100755, + "grad_norm": 0.375, + "learning_rate": 0.001911135599312548, + "loss": 0.0879, + "step": 14552 + }, + { + "epoch": 0.1263270284112117, + "grad_norm": 0.796875, + "learning_rate": 0.0019111227035411536, + "loss": 0.0996, + "step": 14553 + }, + { + "epoch": 0.12633570889141588, + "grad_norm": 0.73828125, + "learning_rate": 0.001911109806882728, + "loss": 0.1172, + "step": 14554 + }, + { + "epoch": 0.12634438937162004, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019110969093372844, + "loss": 0.0962, + "step": 14555 + }, + { + "epoch": 0.1263530698518242, + "grad_norm": 0.7109375, + "learning_rate": 0.001911084010904838, + "loss": 0.1152, + "step": 14556 + }, + { + "epoch": 0.12636175033202837, + "grad_norm": 0.412109375, + "learning_rate": 0.0019110711115854018, + "loss": 0.0835, + "step": 14557 + }, + { + "epoch": 0.12637043081223254, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019110582113789907, + "loss": 0.1172, + "step": 14558 + }, + { + "epoch": 0.1263791112924367, + "grad_norm": 0.1201171875, + "learning_rate": 0.0019110453102856187, + "loss": 0.1406, + "step": 14559 + }, + { + "epoch": 0.12638779177264087, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019110324083052995, + "loss": 0.1016, + "step": 14560 + }, + { + "epoch": 0.12639647225284503, + "grad_norm": 0.48828125, + "learning_rate": 0.0019110195054380474, + "loss": 0.1357, + "step": 14561 + }, + { + "epoch": 0.1264051527330492, + "grad_norm": 0.29296875, + "learning_rate": 0.0019110066016838768, + "loss": 0.1211, + "step": 14562 + }, + { + "epoch": 0.12641383321325336, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019109936970428012, + "loss": 0.0879, + "step": 14563 + }, + { + "epoch": 0.12642251369345753, + "grad_norm": 0.2099609375, + "learning_rate": 0.001910980791514835, + "loss": 0.0923, + "step": 14564 + }, + { + "epoch": 0.1264311941736617, + "grad_norm": 0.08740234375, + "learning_rate": 0.0019109678850999928, + "loss": 0.125, + "step": 14565 + }, + { + "epoch": 0.12643987465386586, + "grad_norm": 0.201171875, + "learning_rate": 0.0019109549777982884, + "loss": 0.1113, + "step": 14566 + }, + { + "epoch": 0.12644855513407002, + "grad_norm": 0.60546875, + "learning_rate": 0.0019109420696097354, + "loss": 0.0786, + "step": 14567 + }, + { + "epoch": 0.1264572356142742, + "grad_norm": 0.515625, + "learning_rate": 0.0019109291605343486, + "loss": 0.1167, + "step": 14568 + }, + { + "epoch": 0.12646591609447835, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019109162505721417, + "loss": 0.0957, + "step": 14569 + }, + { + "epoch": 0.12647459657468252, + "grad_norm": 0.73046875, + "learning_rate": 0.001910903339723129, + "loss": 0.085, + "step": 14570 + }, + { + "epoch": 0.12648327705488668, + "grad_norm": 0.7734375, + "learning_rate": 0.0019108904279873248, + "loss": 0.0967, + "step": 14571 + }, + { + "epoch": 0.12649195753509085, + "grad_norm": 0.6484375, + "learning_rate": 0.001910877515364743, + "loss": 0.1279, + "step": 14572 + }, + { + "epoch": 0.126500638015295, + "grad_norm": 0.470703125, + "learning_rate": 0.0019108646018553975, + "loss": 0.1006, + "step": 14573 + }, + { + "epoch": 0.12650931849549918, + "grad_norm": 0.3125, + "learning_rate": 0.0019108516874593027, + "loss": 0.0938, + "step": 14574 + }, + { + "epoch": 0.12651799897570334, + "grad_norm": 0.81640625, + "learning_rate": 0.0019108387721764727, + "loss": 0.124, + "step": 14575 + }, + { + "epoch": 0.1265266794559075, + "grad_norm": 0.69921875, + "learning_rate": 0.0019108258560069217, + "loss": 0.123, + "step": 14576 + }, + { + "epoch": 0.12653535993611167, + "grad_norm": 0.1689453125, + "learning_rate": 0.0019108129389506635, + "loss": 0.1367, + "step": 14577 + }, + { + "epoch": 0.12654404041631584, + "grad_norm": 0.44921875, + "learning_rate": 0.0019108000210077125, + "loss": 0.1152, + "step": 14578 + }, + { + "epoch": 0.12655272089652, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019107871021780832, + "loss": 0.127, + "step": 14579 + }, + { + "epoch": 0.12656140137672417, + "grad_norm": 0.58203125, + "learning_rate": 0.0019107741824617888, + "loss": 0.2031, + "step": 14580 + }, + { + "epoch": 0.12657008185692833, + "grad_norm": 0.091796875, + "learning_rate": 0.001910761261858844, + "loss": 0.0952, + "step": 14581 + }, + { + "epoch": 0.1265787623371325, + "grad_norm": 0.09765625, + "learning_rate": 0.001910748340369263, + "loss": 0.0938, + "step": 14582 + }, + { + "epoch": 0.12658744281733667, + "grad_norm": 0.43359375, + "learning_rate": 0.0019107354179930599, + "loss": 0.1162, + "step": 14583 + }, + { + "epoch": 0.12659612329754083, + "grad_norm": 0.25390625, + "learning_rate": 0.0019107224947302486, + "loss": 0.1465, + "step": 14584 + }, + { + "epoch": 0.126604803777745, + "grad_norm": 0.158203125, + "learning_rate": 0.0019107095705808432, + "loss": 0.1182, + "step": 14585 + }, + { + "epoch": 0.12661348425794916, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019106966455448582, + "loss": 0.082, + "step": 14586 + }, + { + "epoch": 0.12662216473815333, + "grad_norm": 0.54296875, + "learning_rate": 0.0019106837196223075, + "loss": 0.1201, + "step": 14587 + }, + { + "epoch": 0.1266308452183575, + "grad_norm": 0.5390625, + "learning_rate": 0.0019106707928132054, + "loss": 0.165, + "step": 14588 + }, + { + "epoch": 0.12663952569856166, + "grad_norm": 0.1767578125, + "learning_rate": 0.001910657865117566, + "loss": 0.125, + "step": 14589 + }, + { + "epoch": 0.12664820617876582, + "grad_norm": 0.10205078125, + "learning_rate": 0.0019106449365354028, + "loss": 0.127, + "step": 14590 + }, + { + "epoch": 0.12665688665896999, + "grad_norm": 1.0390625, + "learning_rate": 0.0019106320070667309, + "loss": 0.1406, + "step": 14591 + }, + { + "epoch": 0.12666556713917415, + "grad_norm": 0.283203125, + "learning_rate": 0.0019106190767115639, + "loss": 0.1426, + "step": 14592 + }, + { + "epoch": 0.12667424761937832, + "grad_norm": 0.107421875, + "learning_rate": 0.0019106061454699159, + "loss": 0.1089, + "step": 14593 + }, + { + "epoch": 0.12668292809958248, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019105932133418014, + "loss": 0.0796, + "step": 14594 + }, + { + "epoch": 0.12669160857978665, + "grad_norm": 1.421875, + "learning_rate": 0.0019105802803272343, + "loss": 0.1816, + "step": 14595 + }, + { + "epoch": 0.1267002890599908, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019105673464262288, + "loss": 0.084, + "step": 14596 + }, + { + "epoch": 0.12670896954019498, + "grad_norm": 0.1796875, + "learning_rate": 0.0019105544116387989, + "loss": 0.1289, + "step": 14597 + }, + { + "epoch": 0.12671765002039914, + "grad_norm": 1.9140625, + "learning_rate": 0.0019105414759649588, + "loss": 0.2041, + "step": 14598 + }, + { + "epoch": 0.1267263305006033, + "grad_norm": 0.76171875, + "learning_rate": 0.0019105285394047232, + "loss": 0.1191, + "step": 14599 + }, + { + "epoch": 0.12673501098080747, + "grad_norm": 0.115234375, + "learning_rate": 0.0019105156019581055, + "loss": 0.124, + "step": 14600 + }, + { + "epoch": 0.1267436914610116, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019105026636251203, + "loss": 0.1699, + "step": 14601 + }, + { + "epoch": 0.12675237194121577, + "grad_norm": 0.5703125, + "learning_rate": 0.0019104897244057808, + "loss": 0.1299, + "step": 14602 + }, + { + "epoch": 0.12676105242141994, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019104767843001027, + "loss": 0.1357, + "step": 14603 + }, + { + "epoch": 0.1267697329016241, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019104638433080992, + "loss": 0.1016, + "step": 14604 + }, + { + "epoch": 0.12677841338182827, + "grad_norm": 0.4140625, + "learning_rate": 0.0019104509014297847, + "loss": 0.1143, + "step": 14605 + }, + { + "epoch": 0.12678709386203244, + "grad_norm": 0.17578125, + "learning_rate": 0.0019104379586651727, + "loss": 0.127, + "step": 14606 + }, + { + "epoch": 0.1267957743422366, + "grad_norm": 0.859375, + "learning_rate": 0.0019104250150142785, + "loss": 0.1084, + "step": 14607 + }, + { + "epoch": 0.12680445482244077, + "grad_norm": 0.1875, + "learning_rate": 0.0019104120704771155, + "loss": 0.1152, + "step": 14608 + }, + { + "epoch": 0.12681313530264493, + "grad_norm": 0.486328125, + "learning_rate": 0.001910399125053698, + "loss": 0.083, + "step": 14609 + }, + { + "epoch": 0.1268218157828491, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019103861787440402, + "loss": 0.1035, + "step": 14610 + }, + { + "epoch": 0.12683049626305326, + "grad_norm": 0.4609375, + "learning_rate": 0.0019103732315481564, + "loss": 0.125, + "step": 14611 + }, + { + "epoch": 0.12683917674325743, + "grad_norm": 2.40625, + "learning_rate": 0.0019103602834660606, + "loss": 0.1494, + "step": 14612 + }, + { + "epoch": 0.1268478572234616, + "grad_norm": 0.353515625, + "learning_rate": 0.001910347334497767, + "loss": 0.0996, + "step": 14613 + }, + { + "epoch": 0.12685653770366576, + "grad_norm": 0.796875, + "learning_rate": 0.0019103343846432895, + "loss": 0.1025, + "step": 14614 + }, + { + "epoch": 0.12686521818386992, + "grad_norm": 0.443359375, + "learning_rate": 0.0019103214339026425, + "loss": 0.1177, + "step": 14615 + }, + { + "epoch": 0.1268738986640741, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019103084822758402, + "loss": 0.1387, + "step": 14616 + }, + { + "epoch": 0.12688257914427825, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019102955297628968, + "loss": 0.1543, + "step": 14617 + }, + { + "epoch": 0.12689125962448242, + "grad_norm": 0.1875, + "learning_rate": 0.0019102825763638266, + "loss": 0.1426, + "step": 14618 + }, + { + "epoch": 0.12689994010468658, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019102696220786432, + "loss": 0.1465, + "step": 14619 + }, + { + "epoch": 0.12690862058489075, + "grad_norm": 0.16796875, + "learning_rate": 0.0019102566669073609, + "loss": 0.1475, + "step": 14620 + }, + { + "epoch": 0.1269173010650949, + "grad_norm": 0.29296875, + "learning_rate": 0.0019102437108499946, + "loss": 0.1279, + "step": 14621 + }, + { + "epoch": 0.12692598154529908, + "grad_norm": 0.53515625, + "learning_rate": 0.0019102307539065578, + "loss": 0.1416, + "step": 14622 + }, + { + "epoch": 0.12693466202550324, + "grad_norm": 0.423828125, + "learning_rate": 0.0019102177960770648, + "loss": 0.084, + "step": 14623 + }, + { + "epoch": 0.1269433425057074, + "grad_norm": 0.98046875, + "learning_rate": 0.0019102048373615297, + "loss": 0.0947, + "step": 14624 + }, + { + "epoch": 0.12695202298591157, + "grad_norm": 0.7265625, + "learning_rate": 0.0019101918777599668, + "loss": 0.1025, + "step": 14625 + }, + { + "epoch": 0.12696070346611574, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019101789172723904, + "loss": 0.0996, + "step": 14626 + }, + { + "epoch": 0.1269693839463199, + "grad_norm": 0.12109375, + "learning_rate": 0.0019101659558988142, + "loss": 0.1064, + "step": 14627 + }, + { + "epoch": 0.12697806442652407, + "grad_norm": 0.283203125, + "learning_rate": 0.001910152993639253, + "loss": 0.1113, + "step": 14628 + }, + { + "epoch": 0.12698674490672823, + "grad_norm": 0.6484375, + "learning_rate": 0.0019101400304937205, + "loss": 0.1221, + "step": 14629 + }, + { + "epoch": 0.1269954253869324, + "grad_norm": 0.490234375, + "learning_rate": 0.001910127066462231, + "loss": 0.1104, + "step": 14630 + }, + { + "epoch": 0.12700410586713656, + "grad_norm": 0.31640625, + "learning_rate": 0.001910114101544799, + "loss": 0.1758, + "step": 14631 + }, + { + "epoch": 0.12701278634734073, + "grad_norm": 0.431640625, + "learning_rate": 0.001910101135741438, + "loss": 0.1162, + "step": 14632 + }, + { + "epoch": 0.1270214668275449, + "grad_norm": 0.369140625, + "learning_rate": 0.0019100881690521628, + "loss": 0.1143, + "step": 14633 + }, + { + "epoch": 0.12703014730774906, + "grad_norm": 0.2119140625, + "learning_rate": 0.001910075201476987, + "loss": 0.1021, + "step": 14634 + }, + { + "epoch": 0.12703882778795322, + "grad_norm": 0.267578125, + "learning_rate": 0.0019100622330159255, + "loss": 0.1055, + "step": 14635 + }, + { + "epoch": 0.1270475082681574, + "grad_norm": 0.298828125, + "learning_rate": 0.001910049263668992, + "loss": 0.0811, + "step": 14636 + }, + { + "epoch": 0.12705618874836155, + "grad_norm": 0.8046875, + "learning_rate": 0.001910036293436201, + "loss": 0.1279, + "step": 14637 + }, + { + "epoch": 0.12706486922856572, + "grad_norm": 0.92578125, + "learning_rate": 0.0019100233223175664, + "loss": 0.1172, + "step": 14638 + }, + { + "epoch": 0.12707354970876988, + "grad_norm": 0.4375, + "learning_rate": 0.001910010350313102, + "loss": 0.0928, + "step": 14639 + }, + { + "epoch": 0.12708223018897405, + "grad_norm": 0.1357421875, + "learning_rate": 0.001909997377422823, + "loss": 0.1172, + "step": 14640 + }, + { + "epoch": 0.12709091066917821, + "grad_norm": 0.25390625, + "learning_rate": 0.0019099844036467425, + "loss": 0.1201, + "step": 14641 + }, + { + "epoch": 0.12709959114938238, + "grad_norm": 0.365234375, + "learning_rate": 0.0019099714289848757, + "loss": 0.1367, + "step": 14642 + }, + { + "epoch": 0.12710827162958654, + "grad_norm": 0.32421875, + "learning_rate": 0.0019099584534372363, + "loss": 0.1426, + "step": 14643 + }, + { + "epoch": 0.1271169521097907, + "grad_norm": 0.1494140625, + "learning_rate": 0.0019099454770038385, + "loss": 0.1089, + "step": 14644 + }, + { + "epoch": 0.12712563258999487, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019099324996846962, + "loss": 0.1025, + "step": 14645 + }, + { + "epoch": 0.12713431307019904, + "grad_norm": 0.275390625, + "learning_rate": 0.0019099195214798243, + "loss": 0.0986, + "step": 14646 + }, + { + "epoch": 0.1271429935504032, + "grad_norm": 0.84375, + "learning_rate": 0.0019099065423892362, + "loss": 0.1465, + "step": 14647 + }, + { + "epoch": 0.12715167403060737, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019098935624129467, + "loss": 0.1328, + "step": 14648 + }, + { + "epoch": 0.12716035451081154, + "grad_norm": 0.49609375, + "learning_rate": 0.0019098805815509696, + "loss": 0.1338, + "step": 14649 + }, + { + "epoch": 0.1271690349910157, + "grad_norm": 0.185546875, + "learning_rate": 0.0019098675998033194, + "loss": 0.1406, + "step": 14650 + }, + { + "epoch": 0.12717771547121987, + "grad_norm": 0.1572265625, + "learning_rate": 0.00190985461717001, + "loss": 0.1221, + "step": 14651 + }, + { + "epoch": 0.12718639595142403, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019098416336510563, + "loss": 0.1035, + "step": 14652 + }, + { + "epoch": 0.1271950764316282, + "grad_norm": 0.4453125, + "learning_rate": 0.0019098286492464714, + "loss": 0.1523, + "step": 14653 + }, + { + "epoch": 0.12720375691183236, + "grad_norm": 0.56640625, + "learning_rate": 0.0019098156639562702, + "loss": 0.1465, + "step": 14654 + }, + { + "epoch": 0.12721243739203653, + "grad_norm": 0.404296875, + "learning_rate": 0.0019098026777804669, + "loss": 0.0835, + "step": 14655 + }, + { + "epoch": 0.1272211178722407, + "grad_norm": 0.1796875, + "learning_rate": 0.0019097896907190753, + "loss": 0.168, + "step": 14656 + }, + { + "epoch": 0.12722979835244486, + "grad_norm": 0.373046875, + "learning_rate": 0.00190977670277211, + "loss": 0.1187, + "step": 14657 + }, + { + "epoch": 0.12723847883264902, + "grad_norm": 0.427734375, + "learning_rate": 0.001909763713939585, + "loss": 0.1025, + "step": 14658 + }, + { + "epoch": 0.1272471593128532, + "grad_norm": 0.44921875, + "learning_rate": 0.0019097507242215146, + "loss": 0.0874, + "step": 14659 + }, + { + "epoch": 0.12725583979305735, + "grad_norm": 0.1328125, + "learning_rate": 0.001909737733617913, + "loss": 0.1289, + "step": 14660 + }, + { + "epoch": 0.12726452027326152, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019097247421287944, + "loss": 0.1133, + "step": 14661 + }, + { + "epoch": 0.12727320075346568, + "grad_norm": 0.345703125, + "learning_rate": 0.0019097117497541727, + "loss": 0.0996, + "step": 14662 + }, + { + "epoch": 0.12728188123366985, + "grad_norm": 0.07958984375, + "learning_rate": 0.001909698756494063, + "loss": 0.1069, + "step": 14663 + }, + { + "epoch": 0.127290561713874, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019096857623484786, + "loss": 0.1289, + "step": 14664 + }, + { + "epoch": 0.12729924219407818, + "grad_norm": 0.1767578125, + "learning_rate": 0.0019096727673174338, + "loss": 0.0825, + "step": 14665 + }, + { + "epoch": 0.12730792267428234, + "grad_norm": 0.1875, + "learning_rate": 0.001909659771400943, + "loss": 0.1157, + "step": 14666 + }, + { + "epoch": 0.1273166031544865, + "grad_norm": 0.2177734375, + "learning_rate": 0.0019096467745990209, + "loss": 0.1211, + "step": 14667 + }, + { + "epoch": 0.12732528363469067, + "grad_norm": 0.5234375, + "learning_rate": 0.001909633776911681, + "loss": 0.1021, + "step": 14668 + }, + { + "epoch": 0.12733396411489484, + "grad_norm": 0.37890625, + "learning_rate": 0.0019096207783389377, + "loss": 0.1172, + "step": 14669 + }, + { + "epoch": 0.127342644595099, + "grad_norm": 0.5, + "learning_rate": 0.0019096077788808054, + "loss": 0.1406, + "step": 14670 + }, + { + "epoch": 0.12735132507530317, + "grad_norm": 0.26953125, + "learning_rate": 0.001909594778537298, + "loss": 0.1084, + "step": 14671 + }, + { + "epoch": 0.12736000555550733, + "grad_norm": 0.130859375, + "learning_rate": 0.0019095817773084301, + "loss": 0.1484, + "step": 14672 + }, + { + "epoch": 0.1273686860357115, + "grad_norm": 0.380859375, + "learning_rate": 0.0019095687751942154, + "loss": 0.1152, + "step": 14673 + }, + { + "epoch": 0.12737736651591566, + "grad_norm": 0.953125, + "learning_rate": 0.001909555772194669, + "loss": 0.1084, + "step": 14674 + }, + { + "epoch": 0.12738604699611983, + "grad_norm": 1.015625, + "learning_rate": 0.0019095427683098044, + "loss": 0.0791, + "step": 14675 + }, + { + "epoch": 0.127394727476324, + "grad_norm": 0.17578125, + "learning_rate": 0.001909529763539636, + "loss": 0.1006, + "step": 14676 + }, + { + "epoch": 0.12740340795652816, + "grad_norm": 0.07177734375, + "learning_rate": 0.001909516757884178, + "loss": 0.1152, + "step": 14677 + }, + { + "epoch": 0.12741208843673232, + "grad_norm": 0.5234375, + "learning_rate": 0.0019095037513434442, + "loss": 0.0933, + "step": 14678 + }, + { + "epoch": 0.1274207689169365, + "grad_norm": 0.23828125, + "learning_rate": 0.0019094907439174498, + "loss": 0.1377, + "step": 14679 + }, + { + "epoch": 0.12742944939714065, + "grad_norm": 0.0947265625, + "learning_rate": 0.001909477735606208, + "loss": 0.1289, + "step": 14680 + }, + { + "epoch": 0.12743812987734482, + "grad_norm": 0.08203125, + "learning_rate": 0.001909464726409734, + "loss": 0.0918, + "step": 14681 + }, + { + "epoch": 0.12744681035754898, + "grad_norm": 0.2158203125, + "learning_rate": 0.001909451716328041, + "loss": 0.1348, + "step": 14682 + }, + { + "epoch": 0.12745549083775315, + "grad_norm": 0.3671875, + "learning_rate": 0.0019094387053611446, + "loss": 0.1719, + "step": 14683 + }, + { + "epoch": 0.12746417131795731, + "grad_norm": 0.9375, + "learning_rate": 0.0019094256935090574, + "loss": 0.3652, + "step": 14684 + }, + { + "epoch": 0.12747285179816148, + "grad_norm": 0.169921875, + "learning_rate": 0.001909412680771795, + "loss": 0.1045, + "step": 14685 + }, + { + "epoch": 0.12748153227836564, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019093996671493705, + "loss": 0.0967, + "step": 14686 + }, + { + "epoch": 0.1274902127585698, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019093866526417988, + "loss": 0.123, + "step": 14687 + }, + { + "epoch": 0.12749889323877397, + "grad_norm": 0.21875, + "learning_rate": 0.0019093736372490947, + "loss": 0.0801, + "step": 14688 + }, + { + "epoch": 0.12750757371897814, + "grad_norm": 0.130859375, + "learning_rate": 0.001909360620971271, + "loss": 0.1152, + "step": 14689 + }, + { + "epoch": 0.1275162541991823, + "grad_norm": 0.12060546875, + "learning_rate": 0.001909347603808343, + "loss": 0.1074, + "step": 14690 + }, + { + "epoch": 0.12752493467938647, + "grad_norm": 0.3125, + "learning_rate": 0.0019093345857603243, + "loss": 0.1309, + "step": 14691 + }, + { + "epoch": 0.12753361515959064, + "grad_norm": 0.130859375, + "learning_rate": 0.00190932156682723, + "loss": 0.104, + "step": 14692 + }, + { + "epoch": 0.1275422956397948, + "grad_norm": 0.423828125, + "learning_rate": 0.0019093085470090734, + "loss": 0.1553, + "step": 14693 + }, + { + "epoch": 0.12755097611999897, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019092955263058692, + "loss": 0.0957, + "step": 14694 + }, + { + "epoch": 0.12755965660020313, + "grad_norm": 0.12890625, + "learning_rate": 0.0019092825047176314, + "loss": 0.1108, + "step": 14695 + }, + { + "epoch": 0.1275683370804073, + "grad_norm": 0.353515625, + "learning_rate": 0.0019092694822443748, + "loss": 0.0806, + "step": 14696 + }, + { + "epoch": 0.12757701756061146, + "grad_norm": 0.1611328125, + "learning_rate": 0.001909256458886113, + "loss": 0.1104, + "step": 14697 + }, + { + "epoch": 0.12758569804081563, + "grad_norm": 0.33203125, + "learning_rate": 0.0019092434346428607, + "loss": 0.1064, + "step": 14698 + }, + { + "epoch": 0.1275943785210198, + "grad_norm": 0.154296875, + "learning_rate": 0.0019092304095146317, + "loss": 0.1318, + "step": 14699 + }, + { + "epoch": 0.12760305900122396, + "grad_norm": 0.58203125, + "learning_rate": 0.0019092173835014406, + "loss": 0.2109, + "step": 14700 + }, + { + "epoch": 0.12761173948142812, + "grad_norm": 0.1435546875, + "learning_rate": 0.0019092043566033014, + "loss": 0.0879, + "step": 14701 + }, + { + "epoch": 0.1276204199616323, + "grad_norm": 0.640625, + "learning_rate": 0.0019091913288202284, + "loss": 0.166, + "step": 14702 + }, + { + "epoch": 0.12762910044183645, + "grad_norm": 0.388671875, + "learning_rate": 0.0019091783001522363, + "loss": 0.0967, + "step": 14703 + }, + { + "epoch": 0.12763778092204062, + "grad_norm": 0.4375, + "learning_rate": 0.0019091652705993385, + "loss": 0.1377, + "step": 14704 + }, + { + "epoch": 0.12764646140224478, + "grad_norm": 0.158203125, + "learning_rate": 0.0019091522401615503, + "loss": 0.1455, + "step": 14705 + }, + { + "epoch": 0.12765514188244895, + "grad_norm": 0.12890625, + "learning_rate": 0.0019091392088388848, + "loss": 0.1162, + "step": 14706 + }, + { + "epoch": 0.1276638223626531, + "grad_norm": 0.0888671875, + "learning_rate": 0.0019091261766313575, + "loss": 0.1079, + "step": 14707 + }, + { + "epoch": 0.12767250284285728, + "grad_norm": 0.2236328125, + "learning_rate": 0.0019091131435389813, + "loss": 0.1206, + "step": 14708 + }, + { + "epoch": 0.12768118332306144, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019091001095617713, + "loss": 0.0923, + "step": 14709 + }, + { + "epoch": 0.1276898638032656, + "grad_norm": 0.29296875, + "learning_rate": 0.0019090870746997415, + "loss": 0.103, + "step": 14710 + }, + { + "epoch": 0.12769854428346977, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019090740389529066, + "loss": 0.1045, + "step": 14711 + }, + { + "epoch": 0.12770722476367394, + "grad_norm": 0.421875, + "learning_rate": 0.0019090610023212802, + "loss": 0.1416, + "step": 14712 + }, + { + "epoch": 0.1277159052438781, + "grad_norm": 0.0751953125, + "learning_rate": 0.001909047964804877, + "loss": 0.124, + "step": 14713 + }, + { + "epoch": 0.12772458572408227, + "grad_norm": 0.294921875, + "learning_rate": 0.0019090349264037111, + "loss": 0.1045, + "step": 14714 + }, + { + "epoch": 0.12773326620428643, + "grad_norm": 0.1484375, + "learning_rate": 0.0019090218871177964, + "loss": 0.0952, + "step": 14715 + }, + { + "epoch": 0.1277419466844906, + "grad_norm": 0.193359375, + "learning_rate": 0.0019090088469471478, + "loss": 0.1406, + "step": 14716 + }, + { + "epoch": 0.12775062716469476, + "grad_norm": 0.4453125, + "learning_rate": 0.0019089958058917797, + "loss": 0.1055, + "step": 14717 + }, + { + "epoch": 0.12775930764489893, + "grad_norm": 0.220703125, + "learning_rate": 0.0019089827639517053, + "loss": 0.1406, + "step": 14718 + }, + { + "epoch": 0.1277679881251031, + "grad_norm": 1.1875, + "learning_rate": 0.0019089697211269394, + "loss": 0.2324, + "step": 14719 + }, + { + "epoch": 0.12777666860530726, + "grad_norm": 0.39453125, + "learning_rate": 0.001908956677417497, + "loss": 0.1426, + "step": 14720 + }, + { + "epoch": 0.12778534908551142, + "grad_norm": 0.37890625, + "learning_rate": 0.001908943632823391, + "loss": 0.0845, + "step": 14721 + }, + { + "epoch": 0.1277940295657156, + "grad_norm": 0.36328125, + "learning_rate": 0.001908930587344637, + "loss": 0.1221, + "step": 14722 + }, + { + "epoch": 0.12780271004591975, + "grad_norm": 0.53125, + "learning_rate": 0.0019089175409812487, + "loss": 0.103, + "step": 14723 + }, + { + "epoch": 0.1278113905261239, + "grad_norm": 0.1279296875, + "learning_rate": 0.0019089044937332397, + "loss": 0.1211, + "step": 14724 + }, + { + "epoch": 0.12782007100632806, + "grad_norm": 0.1640625, + "learning_rate": 0.0019088914456006254, + "loss": 0.0947, + "step": 14725 + }, + { + "epoch": 0.12782875148653222, + "grad_norm": 0.2109375, + "learning_rate": 0.0019088783965834195, + "loss": 0.0991, + "step": 14726 + }, + { + "epoch": 0.1278374319667364, + "grad_norm": 0.72265625, + "learning_rate": 0.0019088653466816363, + "loss": 0.1699, + "step": 14727 + }, + { + "epoch": 0.12784611244694055, + "grad_norm": 0.189453125, + "learning_rate": 0.0019088522958952902, + "loss": 0.1377, + "step": 14728 + }, + { + "epoch": 0.12785479292714472, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019088392442243953, + "loss": 0.1182, + "step": 14729 + }, + { + "epoch": 0.12786347340734888, + "grad_norm": 0.3359375, + "learning_rate": 0.001908826191668966, + "loss": 0.1465, + "step": 14730 + }, + { + "epoch": 0.12787215388755305, + "grad_norm": 0.08203125, + "learning_rate": 0.0019088131382290164, + "loss": 0.1006, + "step": 14731 + }, + { + "epoch": 0.1278808343677572, + "grad_norm": 0.07861328125, + "learning_rate": 0.001908800083904561, + "loss": 0.1035, + "step": 14732 + }, + { + "epoch": 0.12788951484796138, + "grad_norm": 0.24609375, + "learning_rate": 0.0019087870286956141, + "loss": 0.0933, + "step": 14733 + }, + { + "epoch": 0.12789819532816554, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019087739726021899, + "loss": 0.1279, + "step": 14734 + }, + { + "epoch": 0.1279068758083697, + "grad_norm": 0.20703125, + "learning_rate": 0.0019087609156243023, + "loss": 0.0674, + "step": 14735 + }, + { + "epoch": 0.12791555628857387, + "grad_norm": 0.474609375, + "learning_rate": 0.001908747857761966, + "loss": 0.1602, + "step": 14736 + }, + { + "epoch": 0.12792423676877804, + "grad_norm": 0.1796875, + "learning_rate": 0.0019087347990151954, + "loss": 0.1084, + "step": 14737 + }, + { + "epoch": 0.1279329172489822, + "grad_norm": 0.451171875, + "learning_rate": 0.0019087217393840045, + "loss": 0.1201, + "step": 14738 + }, + { + "epoch": 0.12794159772918637, + "grad_norm": 0.54296875, + "learning_rate": 0.0019087086788684073, + "loss": 0.0908, + "step": 14739 + }, + { + "epoch": 0.12795027820939053, + "grad_norm": 0.27734375, + "learning_rate": 0.0019086956174684188, + "loss": 0.0947, + "step": 14740 + }, + { + "epoch": 0.1279589586895947, + "grad_norm": 0.333984375, + "learning_rate": 0.0019086825551840527, + "loss": 0.1396, + "step": 14741 + }, + { + "epoch": 0.12796763916979886, + "grad_norm": 0.59375, + "learning_rate": 0.0019086694920153239, + "loss": 0.0996, + "step": 14742 + }, + { + "epoch": 0.12797631965000303, + "grad_norm": 0.875, + "learning_rate": 0.001908656427962246, + "loss": 0.0996, + "step": 14743 + }, + { + "epoch": 0.1279850001302072, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019086433630248337, + "loss": 0.126, + "step": 14744 + }, + { + "epoch": 0.12799368061041136, + "grad_norm": 0.07958984375, + "learning_rate": 0.001908630297203101, + "loss": 0.1123, + "step": 14745 + }, + { + "epoch": 0.12800236109061552, + "grad_norm": 0.671875, + "learning_rate": 0.0019086172304970629, + "loss": 0.0967, + "step": 14746 + }, + { + "epoch": 0.1280110415708197, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019086041629067325, + "loss": 0.1113, + "step": 14747 + }, + { + "epoch": 0.12801972205102385, + "grad_norm": 0.80078125, + "learning_rate": 0.0019085910944321247, + "loss": 0.125, + "step": 14748 + }, + { + "epoch": 0.12802840253122802, + "grad_norm": 0.5078125, + "learning_rate": 0.0019085780250732542, + "loss": 0.1328, + "step": 14749 + }, + { + "epoch": 0.12803708301143218, + "grad_norm": 0.984375, + "learning_rate": 0.0019085649548301348, + "loss": 0.1318, + "step": 14750 + }, + { + "epoch": 0.12804576349163635, + "grad_norm": 0.50390625, + "learning_rate": 0.0019085518837027812, + "loss": 0.1191, + "step": 14751 + }, + { + "epoch": 0.12805444397184051, + "grad_norm": 0.345703125, + "learning_rate": 0.001908538811691207, + "loss": 0.1387, + "step": 14752 + }, + { + "epoch": 0.12806312445204468, + "grad_norm": 0.08984375, + "learning_rate": 0.001908525738795427, + "loss": 0.0942, + "step": 14753 + }, + { + "epoch": 0.12807180493224884, + "grad_norm": 0.119140625, + "learning_rate": 0.0019085126650154555, + "loss": 0.1094, + "step": 14754 + }, + { + "epoch": 0.128080485412453, + "grad_norm": 0.341796875, + "learning_rate": 0.0019084995903513064, + "loss": 0.1108, + "step": 14755 + }, + { + "epoch": 0.12808916589265718, + "grad_norm": 0.201171875, + "learning_rate": 0.0019084865148029945, + "loss": 0.1172, + "step": 14756 + }, + { + "epoch": 0.12809784637286134, + "grad_norm": 0.486328125, + "learning_rate": 0.001908473438370534, + "loss": 0.1289, + "step": 14757 + }, + { + "epoch": 0.1281065268530655, + "grad_norm": 0.1259765625, + "learning_rate": 0.001908460361053939, + "loss": 0.1348, + "step": 14758 + }, + { + "epoch": 0.12811520733326967, + "grad_norm": 0.07763671875, + "learning_rate": 0.001908447282853224, + "loss": 0.1191, + "step": 14759 + }, + { + "epoch": 0.12812388781347384, + "grad_norm": 0.09130859375, + "learning_rate": 0.001908434203768403, + "loss": 0.0991, + "step": 14760 + }, + { + "epoch": 0.128132568293678, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019084211237994906, + "loss": 0.1182, + "step": 14761 + }, + { + "epoch": 0.12814124877388217, + "grad_norm": 0.67578125, + "learning_rate": 0.001908408042946501, + "loss": 0.1396, + "step": 14762 + }, + { + "epoch": 0.12814992925408633, + "grad_norm": 0.87109375, + "learning_rate": 0.0019083949612094487, + "loss": 0.1094, + "step": 14763 + }, + { + "epoch": 0.1281586097342905, + "grad_norm": 0.69921875, + "learning_rate": 0.0019083818785883475, + "loss": 0.2793, + "step": 14764 + }, + { + "epoch": 0.12816729021449466, + "grad_norm": 0.4296875, + "learning_rate": 0.0019083687950832123, + "loss": 0.0996, + "step": 14765 + }, + { + "epoch": 0.12817597069469883, + "grad_norm": 0.310546875, + "learning_rate": 0.001908355710694057, + "loss": 0.1602, + "step": 14766 + }, + { + "epoch": 0.128184651174903, + "grad_norm": 0.67578125, + "learning_rate": 0.0019083426254208962, + "loss": 0.1406, + "step": 14767 + }, + { + "epoch": 0.12819333165510716, + "grad_norm": 0.412109375, + "learning_rate": 0.0019083295392637437, + "loss": 0.1611, + "step": 14768 + }, + { + "epoch": 0.12820201213531132, + "grad_norm": 0.66015625, + "learning_rate": 0.0019083164522226141, + "loss": 0.1079, + "step": 14769 + }, + { + "epoch": 0.1282106926155155, + "grad_norm": 0.287109375, + "learning_rate": 0.0019083033642975224, + "loss": 0.1572, + "step": 14770 + }, + { + "epoch": 0.12821937309571965, + "grad_norm": 0.4296875, + "learning_rate": 0.0019082902754884816, + "loss": 0.1221, + "step": 14771 + }, + { + "epoch": 0.12822805357592382, + "grad_norm": 0.498046875, + "learning_rate": 0.0019082771857955073, + "loss": 0.1104, + "step": 14772 + }, + { + "epoch": 0.12823673405612798, + "grad_norm": 0.2890625, + "learning_rate": 0.0019082640952186127, + "loss": 0.1226, + "step": 14773 + }, + { + "epoch": 0.12824541453633215, + "grad_norm": 0.08544921875, + "learning_rate": 0.001908251003757813, + "loss": 0.0942, + "step": 14774 + }, + { + "epoch": 0.1282540950165363, + "grad_norm": 0.1572265625, + "learning_rate": 0.001908237911413122, + "loss": 0.1377, + "step": 14775 + }, + { + "epoch": 0.12826277549674048, + "grad_norm": 0.458984375, + "learning_rate": 0.001908224818184554, + "loss": 0.1553, + "step": 14776 + }, + { + "epoch": 0.12827145597694464, + "grad_norm": 0.400390625, + "learning_rate": 0.0019082117240721238, + "loss": 0.1011, + "step": 14777 + }, + { + "epoch": 0.1282801364571488, + "grad_norm": 0.17578125, + "learning_rate": 0.001908198629075845, + "loss": 0.1543, + "step": 14778 + }, + { + "epoch": 0.12828881693735297, + "grad_norm": 0.54296875, + "learning_rate": 0.0019081855331957328, + "loss": 0.0894, + "step": 14779 + }, + { + "epoch": 0.12829749741755714, + "grad_norm": 0.357421875, + "learning_rate": 0.0019081724364318007, + "loss": 0.1216, + "step": 14780 + }, + { + "epoch": 0.1283061778977613, + "grad_norm": 0.453125, + "learning_rate": 0.0019081593387840635, + "loss": 0.1016, + "step": 14781 + }, + { + "epoch": 0.12831485837796547, + "grad_norm": 0.392578125, + "learning_rate": 0.0019081462402525354, + "loss": 0.1221, + "step": 14782 + }, + { + "epoch": 0.12832353885816963, + "grad_norm": 0.8828125, + "learning_rate": 0.001908133140837231, + "loss": 0.1094, + "step": 14783 + }, + { + "epoch": 0.1283322193383738, + "grad_norm": 0.1171875, + "learning_rate": 0.0019081200405381637, + "loss": 0.1206, + "step": 14784 + }, + { + "epoch": 0.12834089981857796, + "grad_norm": 1.1171875, + "learning_rate": 0.0019081069393553488, + "loss": 0.1523, + "step": 14785 + }, + { + "epoch": 0.12834958029878213, + "grad_norm": 0.30078125, + "learning_rate": 0.0019080938372888002, + "loss": 0.1035, + "step": 14786 + }, + { + "epoch": 0.1283582607789863, + "grad_norm": 0.193359375, + "learning_rate": 0.0019080807343385324, + "loss": 0.0996, + "step": 14787 + }, + { + "epoch": 0.12836694125919046, + "grad_norm": 0.55078125, + "learning_rate": 0.0019080676305045599, + "loss": 0.1016, + "step": 14788 + }, + { + "epoch": 0.12837562173939462, + "grad_norm": 0.44140625, + "learning_rate": 0.0019080545257868964, + "loss": 0.0918, + "step": 14789 + }, + { + "epoch": 0.1283843022195988, + "grad_norm": 0.2041015625, + "learning_rate": 0.0019080414201855568, + "loss": 0.1226, + "step": 14790 + }, + { + "epoch": 0.12839298269980295, + "grad_norm": 0.5078125, + "learning_rate": 0.0019080283137005549, + "loss": 0.126, + "step": 14791 + }, + { + "epoch": 0.12840166318000712, + "grad_norm": 0.64453125, + "learning_rate": 0.0019080152063319057, + "loss": 0.104, + "step": 14792 + }, + { + "epoch": 0.12841034366021128, + "grad_norm": 0.59765625, + "learning_rate": 0.0019080020980796234, + "loss": 0.1084, + "step": 14793 + }, + { + "epoch": 0.12841902414041545, + "grad_norm": 0.54296875, + "learning_rate": 0.0019079889889437216, + "loss": 0.0977, + "step": 14794 + }, + { + "epoch": 0.12842770462061961, + "grad_norm": 0.1953125, + "learning_rate": 0.0019079758789242156, + "loss": 0.0957, + "step": 14795 + }, + { + "epoch": 0.12843638510082378, + "grad_norm": 0.361328125, + "learning_rate": 0.0019079627680211194, + "loss": 0.1299, + "step": 14796 + }, + { + "epoch": 0.12844506558102795, + "grad_norm": 0.1201171875, + "learning_rate": 0.001907949656234447, + "loss": 0.1021, + "step": 14797 + }, + { + "epoch": 0.1284537460612321, + "grad_norm": 0.10595703125, + "learning_rate": 0.001907936543564213, + "loss": 0.0952, + "step": 14798 + }, + { + "epoch": 0.12846242654143628, + "grad_norm": 0.3046875, + "learning_rate": 0.001907923430010432, + "loss": 0.0981, + "step": 14799 + }, + { + "epoch": 0.12847110702164044, + "grad_norm": 0.236328125, + "learning_rate": 0.0019079103155731178, + "loss": 0.0874, + "step": 14800 + }, + { + "epoch": 0.1284797875018446, + "grad_norm": 0.111328125, + "learning_rate": 0.001907897200252285, + "loss": 0.1035, + "step": 14801 + }, + { + "epoch": 0.12848846798204877, + "grad_norm": 0.271484375, + "learning_rate": 0.0019078840840479484, + "loss": 0.1167, + "step": 14802 + }, + { + "epoch": 0.12849714846225294, + "grad_norm": 0.25, + "learning_rate": 0.001907870966960121, + "loss": 0.1299, + "step": 14803 + }, + { + "epoch": 0.1285058289424571, + "grad_norm": 0.11083984375, + "learning_rate": 0.001907857848988819, + "loss": 0.0933, + "step": 14804 + }, + { + "epoch": 0.12851450942266127, + "grad_norm": 0.3984375, + "learning_rate": 0.0019078447301340553, + "loss": 0.1064, + "step": 14805 + }, + { + "epoch": 0.12852318990286543, + "grad_norm": 0.421875, + "learning_rate": 0.0019078316103958448, + "loss": 0.0898, + "step": 14806 + }, + { + "epoch": 0.1285318703830696, + "grad_norm": 1.6875, + "learning_rate": 0.0019078184897742018, + "loss": 0.1738, + "step": 14807 + }, + { + "epoch": 0.12854055086327376, + "grad_norm": 0.244140625, + "learning_rate": 0.0019078053682691407, + "loss": 0.1196, + "step": 14808 + }, + { + "epoch": 0.12854923134347793, + "grad_norm": 0.35546875, + "learning_rate": 0.0019077922458806758, + "loss": 0.127, + "step": 14809 + }, + { + "epoch": 0.1285579118236821, + "grad_norm": 0.3984375, + "learning_rate": 0.0019077791226088212, + "loss": 0.1211, + "step": 14810 + }, + { + "epoch": 0.12856659230388626, + "grad_norm": 0.46484375, + "learning_rate": 0.001907765998453592, + "loss": 0.0884, + "step": 14811 + }, + { + "epoch": 0.12857527278409042, + "grad_norm": 0.373046875, + "learning_rate": 0.0019077528734150015, + "loss": 0.1191, + "step": 14812 + }, + { + "epoch": 0.1285839532642946, + "grad_norm": 0.546875, + "learning_rate": 0.001907739747493065, + "loss": 0.1162, + "step": 14813 + }, + { + "epoch": 0.12859263374449875, + "grad_norm": 0.59375, + "learning_rate": 0.001907726620687796, + "loss": 0.1748, + "step": 14814 + }, + { + "epoch": 0.12860131422470292, + "grad_norm": 0.099609375, + "learning_rate": 0.0019077134929992097, + "loss": 0.1025, + "step": 14815 + }, + { + "epoch": 0.12860999470490708, + "grad_norm": 0.271484375, + "learning_rate": 0.0019077003644273202, + "loss": 0.1064, + "step": 14816 + }, + { + "epoch": 0.12861867518511125, + "grad_norm": 0.640625, + "learning_rate": 0.0019076872349721413, + "loss": 0.1426, + "step": 14817 + }, + { + "epoch": 0.1286273556653154, + "grad_norm": 0.52734375, + "learning_rate": 0.001907674104633688, + "loss": 0.1328, + "step": 14818 + }, + { + "epoch": 0.12863603614551958, + "grad_norm": 0.625, + "learning_rate": 0.0019076609734119743, + "loss": 0.123, + "step": 14819 + }, + { + "epoch": 0.12864471662572374, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019076478413070147, + "loss": 0.1055, + "step": 14820 + }, + { + "epoch": 0.1286533971059279, + "grad_norm": 0.97265625, + "learning_rate": 0.0019076347083188236, + "loss": 0.0942, + "step": 14821 + }, + { + "epoch": 0.12866207758613207, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019076215744474152, + "loss": 0.1133, + "step": 14822 + }, + { + "epoch": 0.12867075806633624, + "grad_norm": 0.4375, + "learning_rate": 0.0019076084396928042, + "loss": 0.1074, + "step": 14823 + }, + { + "epoch": 0.1286794385465404, + "grad_norm": 0.345703125, + "learning_rate": 0.0019075953040550045, + "loss": 0.1104, + "step": 14824 + }, + { + "epoch": 0.12868811902674457, + "grad_norm": 0.197265625, + "learning_rate": 0.0019075821675340313, + "loss": 0.124, + "step": 14825 + }, + { + "epoch": 0.12869679950694873, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019075690301298977, + "loss": 0.0957, + "step": 14826 + }, + { + "epoch": 0.1287054799871529, + "grad_norm": 1.109375, + "learning_rate": 0.0019075558918426189, + "loss": 0.1426, + "step": 14827 + }, + { + "epoch": 0.12871416046735706, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019075427526722092, + "loss": 0.1328, + "step": 14828 + }, + { + "epoch": 0.12872284094756123, + "grad_norm": 0.193359375, + "learning_rate": 0.001907529612618683, + "loss": 0.1641, + "step": 14829 + }, + { + "epoch": 0.1287315214277654, + "grad_norm": 0.494140625, + "learning_rate": 0.0019075164716820543, + "loss": 0.1787, + "step": 14830 + }, + { + "epoch": 0.12874020190796956, + "grad_norm": 0.16796875, + "learning_rate": 0.0019075033298623378, + "loss": 0.0938, + "step": 14831 + }, + { + "epoch": 0.12874888238817372, + "grad_norm": 0.2451171875, + "learning_rate": 0.0019074901871595478, + "loss": 0.1162, + "step": 14832 + }, + { + "epoch": 0.1287575628683779, + "grad_norm": 0.275390625, + "learning_rate": 0.0019074770435736988, + "loss": 0.1309, + "step": 14833 + }, + { + "epoch": 0.12876624334858205, + "grad_norm": 0.73046875, + "learning_rate": 0.001907463899104805, + "loss": 0.1416, + "step": 14834 + }, + { + "epoch": 0.12877492382878622, + "grad_norm": 0.59375, + "learning_rate": 0.0019074507537528804, + "loss": 0.124, + "step": 14835 + }, + { + "epoch": 0.12878360430899038, + "grad_norm": 0.67578125, + "learning_rate": 0.0019074376075179404, + "loss": 0.1182, + "step": 14836 + }, + { + "epoch": 0.12879228478919455, + "grad_norm": 0.51171875, + "learning_rate": 0.0019074244603999983, + "loss": 0.1016, + "step": 14837 + }, + { + "epoch": 0.12880096526939872, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019074113123990692, + "loss": 0.1895, + "step": 14838 + }, + { + "epoch": 0.12880964574960288, + "grad_norm": 0.828125, + "learning_rate": 0.0019073981635151672, + "loss": 0.1143, + "step": 14839 + }, + { + "epoch": 0.12881832622980705, + "grad_norm": 0.111328125, + "learning_rate": 0.0019073850137483067, + "loss": 0.1299, + "step": 14840 + }, + { + "epoch": 0.1288270067100112, + "grad_norm": 0.466796875, + "learning_rate": 0.001907371863098502, + "loss": 0.1094, + "step": 14841 + }, + { + "epoch": 0.12883568719021538, + "grad_norm": 0.3828125, + "learning_rate": 0.0019073587115657677, + "loss": 0.1055, + "step": 14842 + }, + { + "epoch": 0.12884436767041954, + "grad_norm": 0.1953125, + "learning_rate": 0.0019073455591501179, + "loss": 0.1377, + "step": 14843 + }, + { + "epoch": 0.1288530481506237, + "grad_norm": 0.3515625, + "learning_rate": 0.0019073324058515673, + "loss": 0.1523, + "step": 14844 + }, + { + "epoch": 0.12886172863082787, + "grad_norm": 0.328125, + "learning_rate": 0.0019073192516701298, + "loss": 0.1191, + "step": 14845 + }, + { + "epoch": 0.12887040911103204, + "grad_norm": 1.984375, + "learning_rate": 0.0019073060966058202, + "loss": 0.6523, + "step": 14846 + }, + { + "epoch": 0.12887908959123617, + "grad_norm": 0.341796875, + "learning_rate": 0.001907292940658653, + "loss": 0.0938, + "step": 14847 + }, + { + "epoch": 0.12888777007144034, + "grad_norm": 0.171875, + "learning_rate": 0.0019072797838286422, + "loss": 0.1465, + "step": 14848 + }, + { + "epoch": 0.1288964505516445, + "grad_norm": 0.1376953125, + "learning_rate": 0.0019072666261158026, + "loss": 0.0718, + "step": 14849 + }, + { + "epoch": 0.12890513103184867, + "grad_norm": 0.337890625, + "learning_rate": 0.0019072534675201478, + "loss": 0.1201, + "step": 14850 + }, + { + "epoch": 0.12891381151205283, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019072403080416932, + "loss": 0.1074, + "step": 14851 + }, + { + "epoch": 0.128922491992257, + "grad_norm": 0.40625, + "learning_rate": 0.0019072271476804523, + "loss": 0.1157, + "step": 14852 + }, + { + "epoch": 0.12893117247246116, + "grad_norm": 0.337890625, + "learning_rate": 0.0019072139864364404, + "loss": 0.1045, + "step": 14853 + }, + { + "epoch": 0.12893985295266533, + "grad_norm": 0.2470703125, + "learning_rate": 0.0019072008243096713, + "loss": 0.1113, + "step": 14854 + }, + { + "epoch": 0.1289485334328695, + "grad_norm": 0.13671875, + "learning_rate": 0.0019071876613001592, + "loss": 0.1143, + "step": 14855 + }, + { + "epoch": 0.12895721391307366, + "grad_norm": 0.388671875, + "learning_rate": 0.001907174497407919, + "loss": 0.0947, + "step": 14856 + }, + { + "epoch": 0.12896589439327782, + "grad_norm": 0.0791015625, + "learning_rate": 0.001907161332632965, + "loss": 0.084, + "step": 14857 + }, + { + "epoch": 0.128974574873482, + "grad_norm": 0.248046875, + "learning_rate": 0.0019071481669753116, + "loss": 0.0864, + "step": 14858 + }, + { + "epoch": 0.12898325535368615, + "grad_norm": 0.3125, + "learning_rate": 0.0019071350004349727, + "loss": 0.1172, + "step": 14859 + }, + { + "epoch": 0.12899193583389032, + "grad_norm": 0.29296875, + "learning_rate": 0.0019071218330119636, + "loss": 0.0859, + "step": 14860 + }, + { + "epoch": 0.12900061631409449, + "grad_norm": 0.15234375, + "learning_rate": 0.0019071086647062978, + "loss": 0.106, + "step": 14861 + }, + { + "epoch": 0.12900929679429865, + "grad_norm": 0.080078125, + "learning_rate": 0.0019070954955179903, + "loss": 0.1084, + "step": 14862 + }, + { + "epoch": 0.12901797727450282, + "grad_norm": 0.12158203125, + "learning_rate": 0.0019070823254470548, + "loss": 0.1079, + "step": 14863 + }, + { + "epoch": 0.12902665775470698, + "grad_norm": 0.10546875, + "learning_rate": 0.001907069154493507, + "loss": 0.1016, + "step": 14864 + }, + { + "epoch": 0.12903533823491115, + "grad_norm": 0.279296875, + "learning_rate": 0.0019070559826573597, + "loss": 0.1045, + "step": 14865 + }, + { + "epoch": 0.1290440187151153, + "grad_norm": 0.171875, + "learning_rate": 0.0019070428099386288, + "loss": 0.1533, + "step": 14866 + }, + { + "epoch": 0.12905269919531948, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019070296363373274, + "loss": 0.1387, + "step": 14867 + }, + { + "epoch": 0.12906137967552364, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019070164618534711, + "loss": 0.1016, + "step": 14868 + }, + { + "epoch": 0.1290700601557278, + "grad_norm": 0.345703125, + "learning_rate": 0.0019070032864870732, + "loss": 0.0898, + "step": 14869 + }, + { + "epoch": 0.12907874063593197, + "grad_norm": 0.58203125, + "learning_rate": 0.0019069901102381488, + "loss": 0.1191, + "step": 14870 + }, + { + "epoch": 0.12908742111613614, + "grad_norm": 0.498046875, + "learning_rate": 0.0019069769331067125, + "loss": 0.1318, + "step": 14871 + }, + { + "epoch": 0.1290961015963403, + "grad_norm": 0.201171875, + "learning_rate": 0.0019069637550927778, + "loss": 0.1377, + "step": 14872 + }, + { + "epoch": 0.12910478207654447, + "grad_norm": 0.6015625, + "learning_rate": 0.0019069505761963599, + "loss": 0.126, + "step": 14873 + }, + { + "epoch": 0.12911346255674863, + "grad_norm": 0.578125, + "learning_rate": 0.001906937396417473, + "loss": 0.1025, + "step": 14874 + }, + { + "epoch": 0.1291221430369528, + "grad_norm": 0.48828125, + "learning_rate": 0.0019069242157561314, + "loss": 0.0957, + "step": 14875 + }, + { + "epoch": 0.12913082351715696, + "grad_norm": 0.2021484375, + "learning_rate": 0.0019069110342123498, + "loss": 0.1123, + "step": 14876 + }, + { + "epoch": 0.12913950399736113, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019068978517861423, + "loss": 0.1016, + "step": 14877 + }, + { + "epoch": 0.1291481844775653, + "grad_norm": 0.361328125, + "learning_rate": 0.001906884668477523, + "loss": 0.1455, + "step": 14878 + }, + { + "epoch": 0.12915686495776946, + "grad_norm": 0.09423828125, + "learning_rate": 0.0019068714842865077, + "loss": 0.1582, + "step": 14879 + }, + { + "epoch": 0.12916554543797362, + "grad_norm": 0.490234375, + "learning_rate": 0.0019068582992131091, + "loss": 0.0869, + "step": 14880 + }, + { + "epoch": 0.1291742259181778, + "grad_norm": 0.75390625, + "learning_rate": 0.0019068451132573429, + "loss": 0.1289, + "step": 14881 + }, + { + "epoch": 0.12918290639838195, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019068319264192225, + "loss": 0.1631, + "step": 14882 + }, + { + "epoch": 0.12919158687858612, + "grad_norm": 0.12353515625, + "learning_rate": 0.0019068187386987632, + "loss": 0.1074, + "step": 14883 + }, + { + "epoch": 0.12920026735879028, + "grad_norm": 0.3125, + "learning_rate": 0.0019068055500959789, + "loss": 0.0713, + "step": 14884 + }, + { + "epoch": 0.12920894783899445, + "grad_norm": 0.126953125, + "learning_rate": 0.001906792360610884, + "loss": 0.1216, + "step": 14885 + }, + { + "epoch": 0.1292176283191986, + "grad_norm": 0.22265625, + "learning_rate": 0.0019067791702434933, + "loss": 0.1436, + "step": 14886 + }, + { + "epoch": 0.12922630879940278, + "grad_norm": 0.27734375, + "learning_rate": 0.0019067659789938213, + "loss": 0.124, + "step": 14887 + }, + { + "epoch": 0.12923498927960694, + "grad_norm": 0.193359375, + "learning_rate": 0.0019067527868618817, + "loss": 0.1279, + "step": 14888 + }, + { + "epoch": 0.1292436697598111, + "grad_norm": 0.1904296875, + "learning_rate": 0.0019067395938476897, + "loss": 0.0928, + "step": 14889 + }, + { + "epoch": 0.12925235024001527, + "grad_norm": 0.20703125, + "learning_rate": 0.001906726399951259, + "loss": 0.0967, + "step": 14890 + }, + { + "epoch": 0.12926103072021944, + "grad_norm": 0.11767578125, + "learning_rate": 0.0019067132051726046, + "loss": 0.0981, + "step": 14891 + }, + { + "epoch": 0.1292697112004236, + "grad_norm": 0.271484375, + "learning_rate": 0.001906700009511741, + "loss": 0.1025, + "step": 14892 + }, + { + "epoch": 0.12927839168062777, + "grad_norm": 0.7890625, + "learning_rate": 0.0019066868129686823, + "loss": 0.1357, + "step": 14893 + }, + { + "epoch": 0.12928707216083193, + "grad_norm": 0.54296875, + "learning_rate": 0.0019066736155434428, + "loss": 0.3633, + "step": 14894 + }, + { + "epoch": 0.1292957526410361, + "grad_norm": 0.25390625, + "learning_rate": 0.001906660417236037, + "loss": 0.1138, + "step": 14895 + }, + { + "epoch": 0.12930443312124026, + "grad_norm": 0.08349609375, + "learning_rate": 0.00190664721804648, + "loss": 0.124, + "step": 14896 + }, + { + "epoch": 0.12931311360144443, + "grad_norm": 0.71484375, + "learning_rate": 0.0019066340179747857, + "loss": 0.124, + "step": 14897 + }, + { + "epoch": 0.1293217940816486, + "grad_norm": 0.318359375, + "learning_rate": 0.001906620817020968, + "loss": 0.0889, + "step": 14898 + }, + { + "epoch": 0.12933047456185276, + "grad_norm": 0.359375, + "learning_rate": 0.0019066076151850423, + "loss": 0.1001, + "step": 14899 + }, + { + "epoch": 0.12933915504205692, + "grad_norm": 1.9140625, + "learning_rate": 0.0019065944124670225, + "loss": 0.1328, + "step": 14900 + }, + { + "epoch": 0.1293478355222611, + "grad_norm": 0.06591796875, + "learning_rate": 0.0019065812088669232, + "loss": 0.1055, + "step": 14901 + }, + { + "epoch": 0.12935651600246525, + "grad_norm": 0.50390625, + "learning_rate": 0.001906568004384759, + "loss": 0.1104, + "step": 14902 + }, + { + "epoch": 0.12936519648266942, + "grad_norm": 0.65625, + "learning_rate": 0.001906554799020544, + "loss": 0.0879, + "step": 14903 + }, + { + "epoch": 0.12937387696287359, + "grad_norm": 0.92578125, + "learning_rate": 0.0019065415927742926, + "loss": 0.1611, + "step": 14904 + }, + { + "epoch": 0.12938255744307775, + "grad_norm": 0.306640625, + "learning_rate": 0.0019065283856460199, + "loss": 0.1094, + "step": 14905 + }, + { + "epoch": 0.12939123792328192, + "grad_norm": 0.5703125, + "learning_rate": 0.0019065151776357394, + "loss": 0.1084, + "step": 14906 + }, + { + "epoch": 0.12939991840348608, + "grad_norm": 0.7265625, + "learning_rate": 0.0019065019687434664, + "loss": 0.1426, + "step": 14907 + }, + { + "epoch": 0.12940859888369025, + "grad_norm": 0.220703125, + "learning_rate": 0.0019064887589692148, + "loss": 0.1064, + "step": 14908 + }, + { + "epoch": 0.1294172793638944, + "grad_norm": 0.248046875, + "learning_rate": 0.0019064755483129989, + "loss": 0.104, + "step": 14909 + }, + { + "epoch": 0.12942595984409858, + "grad_norm": 0.19921875, + "learning_rate": 0.001906462336774834, + "loss": 0.1191, + "step": 14910 + }, + { + "epoch": 0.12943464032430274, + "grad_norm": 0.56640625, + "learning_rate": 0.0019064491243547335, + "loss": 0.1299, + "step": 14911 + }, + { + "epoch": 0.1294433208045069, + "grad_norm": 0.44140625, + "learning_rate": 0.0019064359110527128, + "loss": 0.1016, + "step": 14912 + }, + { + "epoch": 0.12945200128471107, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019064226968687856, + "loss": 0.123, + "step": 14913 + }, + { + "epoch": 0.12946068176491524, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019064094818029668, + "loss": 0.1084, + "step": 14914 + }, + { + "epoch": 0.1294693622451194, + "grad_norm": 0.333984375, + "learning_rate": 0.0019063962658552706, + "loss": 0.1113, + "step": 14915 + }, + { + "epoch": 0.12947804272532357, + "grad_norm": 0.25390625, + "learning_rate": 0.0019063830490257117, + "loss": 0.1201, + "step": 14916 + }, + { + "epoch": 0.12948672320552773, + "grad_norm": 0.57421875, + "learning_rate": 0.0019063698313143043, + "loss": 0.0859, + "step": 14917 + }, + { + "epoch": 0.1294954036857319, + "grad_norm": 0.392578125, + "learning_rate": 0.0019063566127210632, + "loss": 0.125, + "step": 14918 + }, + { + "epoch": 0.12950408416593606, + "grad_norm": 1.921875, + "learning_rate": 0.0019063433932460024, + "loss": 0.1172, + "step": 14919 + }, + { + "epoch": 0.12951276464614023, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019063301728891365, + "loss": 0.1289, + "step": 14920 + }, + { + "epoch": 0.1295214451263444, + "grad_norm": 0.7578125, + "learning_rate": 0.0019063169516504801, + "loss": 0.1152, + "step": 14921 + }, + { + "epoch": 0.12953012560654856, + "grad_norm": 0.375, + "learning_rate": 0.0019063037295300475, + "loss": 0.0791, + "step": 14922 + }, + { + "epoch": 0.12953880608675272, + "grad_norm": 0.1044921875, + "learning_rate": 0.0019062905065278537, + "loss": 0.0825, + "step": 14923 + }, + { + "epoch": 0.1295474865669569, + "grad_norm": 0.18359375, + "learning_rate": 0.0019062772826439122, + "loss": 0.0908, + "step": 14924 + }, + { + "epoch": 0.12955616704716105, + "grad_norm": 0.6015625, + "learning_rate": 0.0019062640578782384, + "loss": 0.1279, + "step": 14925 + }, + { + "epoch": 0.12956484752736522, + "grad_norm": 0.3046875, + "learning_rate": 0.0019062508322308463, + "loss": 0.1533, + "step": 14926 + }, + { + "epoch": 0.12957352800756938, + "grad_norm": 0.70703125, + "learning_rate": 0.0019062376057017501, + "loss": 0.1006, + "step": 14927 + }, + { + "epoch": 0.12958220848777355, + "grad_norm": 0.423828125, + "learning_rate": 0.0019062243782909647, + "loss": 0.0894, + "step": 14928 + }, + { + "epoch": 0.1295908889679777, + "grad_norm": 0.130859375, + "learning_rate": 0.0019062111499985044, + "loss": 0.1133, + "step": 14929 + }, + { + "epoch": 0.12959956944818188, + "grad_norm": 0.3984375, + "learning_rate": 0.0019061979208243839, + "loss": 0.1133, + "step": 14930 + }, + { + "epoch": 0.12960824992838604, + "grad_norm": 0.486328125, + "learning_rate": 0.0019061846907686174, + "loss": 0.209, + "step": 14931 + }, + { + "epoch": 0.1296169304085902, + "grad_norm": 0.330078125, + "learning_rate": 0.0019061714598312192, + "loss": 0.0996, + "step": 14932 + }, + { + "epoch": 0.12962561088879437, + "grad_norm": 0.1611328125, + "learning_rate": 0.0019061582280122042, + "loss": 0.1079, + "step": 14933 + }, + { + "epoch": 0.12963429136899854, + "grad_norm": 0.26953125, + "learning_rate": 0.0019061449953115866, + "loss": 0.1055, + "step": 14934 + }, + { + "epoch": 0.1296429718492027, + "grad_norm": 0.07421875, + "learning_rate": 0.0019061317617293809, + "loss": 0.1094, + "step": 14935 + }, + { + "epoch": 0.12965165232940687, + "grad_norm": 0.197265625, + "learning_rate": 0.0019061185272656017, + "loss": 0.1465, + "step": 14936 + }, + { + "epoch": 0.12966033280961103, + "grad_norm": 0.5546875, + "learning_rate": 0.0019061052919202635, + "loss": 0.1484, + "step": 14937 + }, + { + "epoch": 0.1296690132898152, + "grad_norm": 0.447265625, + "learning_rate": 0.0019060920556933808, + "loss": 0.1465, + "step": 14938 + }, + { + "epoch": 0.12967769377001936, + "grad_norm": 0.09814453125, + "learning_rate": 0.0019060788185849672, + "loss": 0.0898, + "step": 14939 + }, + { + "epoch": 0.12968637425022353, + "grad_norm": 0.30859375, + "learning_rate": 0.0019060655805950386, + "loss": 0.1216, + "step": 14940 + }, + { + "epoch": 0.1296950547304277, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019060523417236087, + "loss": 0.085, + "step": 14941 + }, + { + "epoch": 0.12970373521063186, + "grad_norm": 0.4375, + "learning_rate": 0.0019060391019706919, + "loss": 0.0967, + "step": 14942 + }, + { + "epoch": 0.12971241569083602, + "grad_norm": 0.275390625, + "learning_rate": 0.001906025861336303, + "loss": 0.1172, + "step": 14943 + }, + { + "epoch": 0.1297210961710402, + "grad_norm": 0.05810546875, + "learning_rate": 0.0019060126198204561, + "loss": 0.0908, + "step": 14944 + }, + { + "epoch": 0.12972977665124436, + "grad_norm": 0.35546875, + "learning_rate": 0.001905999377423166, + "loss": 0.126, + "step": 14945 + }, + { + "epoch": 0.12973845713144852, + "grad_norm": 0.3828125, + "learning_rate": 0.0019059861341444472, + "loss": 0.1445, + "step": 14946 + }, + { + "epoch": 0.12974713761165269, + "grad_norm": 0.11474609375, + "learning_rate": 0.0019059728899843142, + "loss": 0.1162, + "step": 14947 + }, + { + "epoch": 0.12975581809185685, + "grad_norm": 0.234375, + "learning_rate": 0.0019059596449427809, + "loss": 0.1396, + "step": 14948 + }, + { + "epoch": 0.12976449857206102, + "grad_norm": 0.10107421875, + "learning_rate": 0.0019059463990198624, + "loss": 0.1348, + "step": 14949 + }, + { + "epoch": 0.12977317905226518, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019059331522155733, + "loss": 0.1016, + "step": 14950 + }, + { + "epoch": 0.12978185953246935, + "grad_norm": 0.1923828125, + "learning_rate": 0.0019059199045299278, + "loss": 0.1162, + "step": 14951 + }, + { + "epoch": 0.1297905400126735, + "grad_norm": 0.59765625, + "learning_rate": 0.0019059066559629404, + "loss": 0.0977, + "step": 14952 + }, + { + "epoch": 0.12979922049287768, + "grad_norm": 0.251953125, + "learning_rate": 0.0019058934065146254, + "loss": 0.1064, + "step": 14953 + }, + { + "epoch": 0.12980790097308184, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019058801561849974, + "loss": 0.127, + "step": 14954 + }, + { + "epoch": 0.129816581453286, + "grad_norm": 0.474609375, + "learning_rate": 0.0019058669049740711, + "loss": 0.1289, + "step": 14955 + }, + { + "epoch": 0.12982526193349017, + "grad_norm": 0.73046875, + "learning_rate": 0.0019058536528818609, + "loss": 0.1396, + "step": 14956 + }, + { + "epoch": 0.12983394241369434, + "grad_norm": 0.25, + "learning_rate": 0.0019058403999083812, + "loss": 0.1123, + "step": 14957 + }, + { + "epoch": 0.1298426228938985, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019058271460536465, + "loss": 0.0908, + "step": 14958 + }, + { + "epoch": 0.12985130337410267, + "grad_norm": 1.515625, + "learning_rate": 0.0019058138913176715, + "loss": 0.3477, + "step": 14959 + }, + { + "epoch": 0.12985998385430683, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019058006357004702, + "loss": 0.1035, + "step": 14960 + }, + { + "epoch": 0.129868664334511, + "grad_norm": 0.11328125, + "learning_rate": 0.0019057873792020577, + "loss": 0.0771, + "step": 14961 + }, + { + "epoch": 0.12987734481471516, + "grad_norm": 0.1103515625, + "learning_rate": 0.0019057741218224484, + "loss": 0.1143, + "step": 14962 + }, + { + "epoch": 0.12988602529491933, + "grad_norm": 0.1875, + "learning_rate": 0.0019057608635616563, + "loss": 0.1064, + "step": 14963 + }, + { + "epoch": 0.1298947057751235, + "grad_norm": 0.41015625, + "learning_rate": 0.0019057476044196963, + "loss": 0.1289, + "step": 14964 + }, + { + "epoch": 0.12990338625532766, + "grad_norm": 0.287109375, + "learning_rate": 0.001905734344396583, + "loss": 0.1406, + "step": 14965 + }, + { + "epoch": 0.12991206673553182, + "grad_norm": 0.103515625, + "learning_rate": 0.0019057210834923306, + "loss": 0.1289, + "step": 14966 + }, + { + "epoch": 0.129920747215736, + "grad_norm": 0.65234375, + "learning_rate": 0.0019057078217069538, + "loss": 0.1279, + "step": 14967 + }, + { + "epoch": 0.12992942769594015, + "grad_norm": 0.1484375, + "learning_rate": 0.0019056945590404668, + "loss": 0.1133, + "step": 14968 + }, + { + "epoch": 0.12993810817614432, + "grad_norm": 0.38671875, + "learning_rate": 0.0019056812954928845, + "loss": 0.1504, + "step": 14969 + }, + { + "epoch": 0.12994678865634846, + "grad_norm": 0.20703125, + "learning_rate": 0.0019056680310642212, + "loss": 0.1318, + "step": 14970 + }, + { + "epoch": 0.12995546913655262, + "grad_norm": 0.154296875, + "learning_rate": 0.0019056547657544914, + "loss": 0.1152, + "step": 14971 + }, + { + "epoch": 0.12996414961675679, + "grad_norm": 0.11279296875, + "learning_rate": 0.0019056414995637096, + "loss": 0.1553, + "step": 14972 + }, + { + "epoch": 0.12997283009696095, + "grad_norm": 0.109375, + "learning_rate": 0.0019056282324918905, + "loss": 0.0938, + "step": 14973 + }, + { + "epoch": 0.12998151057716512, + "grad_norm": 0.1875, + "learning_rate": 0.0019056149645390486, + "loss": 0.0923, + "step": 14974 + }, + { + "epoch": 0.12999019105736928, + "grad_norm": 0.087890625, + "learning_rate": 0.0019056016957051984, + "loss": 0.1021, + "step": 14975 + }, + { + "epoch": 0.12999887153757345, + "grad_norm": 0.6796875, + "learning_rate": 0.001905588425990354, + "loss": 0.1045, + "step": 14976 + }, + { + "epoch": 0.1300075520177776, + "grad_norm": 0.61328125, + "learning_rate": 0.0019055751553945302, + "loss": 0.0942, + "step": 14977 + }, + { + "epoch": 0.13001623249798178, + "grad_norm": 0.349609375, + "learning_rate": 0.0019055618839177416, + "loss": 0.1777, + "step": 14978 + }, + { + "epoch": 0.13002491297818594, + "grad_norm": 0.09375, + "learning_rate": 0.0019055486115600027, + "loss": 0.1416, + "step": 14979 + }, + { + "epoch": 0.1300335934583901, + "grad_norm": 0.11181640625, + "learning_rate": 0.0019055353383213276, + "loss": 0.0869, + "step": 14980 + }, + { + "epoch": 0.13004227393859427, + "grad_norm": 0.70703125, + "learning_rate": 0.0019055220642017312, + "loss": 0.1055, + "step": 14981 + }, + { + "epoch": 0.13005095441879844, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019055087892012284, + "loss": 0.1504, + "step": 14982 + }, + { + "epoch": 0.1300596348990026, + "grad_norm": 0.08837890625, + "learning_rate": 0.0019054955133198331, + "loss": 0.1104, + "step": 14983 + }, + { + "epoch": 0.13006831537920677, + "grad_norm": 0.2236328125, + "learning_rate": 0.00190548223655756, + "loss": 0.1035, + "step": 14984 + }, + { + "epoch": 0.13007699585941093, + "grad_norm": 0.08642578125, + "learning_rate": 0.0019054689589144233, + "loss": 0.0938, + "step": 14985 + }, + { + "epoch": 0.1300856763396151, + "grad_norm": 0.16015625, + "learning_rate": 0.0019054556803904381, + "loss": 0.0957, + "step": 14986 + }, + { + "epoch": 0.13009435681981926, + "grad_norm": 0.09765625, + "learning_rate": 0.0019054424009856189, + "loss": 0.1133, + "step": 14987 + }, + { + "epoch": 0.13010303730002343, + "grad_norm": 1.0078125, + "learning_rate": 0.0019054291206999797, + "loss": 0.1279, + "step": 14988 + }, + { + "epoch": 0.1301117177802276, + "grad_norm": 0.310546875, + "learning_rate": 0.0019054158395335355, + "loss": 0.0771, + "step": 14989 + }, + { + "epoch": 0.13012039826043176, + "grad_norm": 0.146484375, + "learning_rate": 0.0019054025574863004, + "loss": 0.0957, + "step": 14990 + }, + { + "epoch": 0.13012907874063592, + "grad_norm": 0.197265625, + "learning_rate": 0.0019053892745582895, + "loss": 0.0879, + "step": 14991 + }, + { + "epoch": 0.1301377592208401, + "grad_norm": 0.130859375, + "learning_rate": 0.0019053759907495167, + "loss": 0.0923, + "step": 14992 + }, + { + "epoch": 0.13014643970104425, + "grad_norm": 0.7421875, + "learning_rate": 0.0019053627060599968, + "loss": 0.3535, + "step": 14993 + }, + { + "epoch": 0.13015512018124842, + "grad_norm": 0.55078125, + "learning_rate": 0.0019053494204897446, + "loss": 0.084, + "step": 14994 + }, + { + "epoch": 0.13016380066145258, + "grad_norm": 0.255859375, + "learning_rate": 0.0019053361340387743, + "loss": 0.0942, + "step": 14995 + }, + { + "epoch": 0.13017248114165675, + "grad_norm": 0.2314453125, + "learning_rate": 0.0019053228467071001, + "loss": 0.1143, + "step": 14996 + }, + { + "epoch": 0.1301811616218609, + "grad_norm": 0.29296875, + "learning_rate": 0.0019053095584947374, + "loss": 0.1221, + "step": 14997 + }, + { + "epoch": 0.13018984210206508, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019052962694017003, + "loss": 0.1064, + "step": 14998 + }, + { + "epoch": 0.13019852258226924, + "grad_norm": 0.125, + "learning_rate": 0.0019052829794280028, + "loss": 0.1221, + "step": 14999 + }, + { + "epoch": 0.1302072030624734, + "grad_norm": 0.263671875, + "learning_rate": 0.0019052696885736608, + "loss": 0.1206, + "step": 15000 + }, + { + "epoch": 0.13021588354267757, + "grad_norm": 0.109375, + "learning_rate": 0.001905256396838687, + "loss": 0.0762, + "step": 15001 + }, + { + "epoch": 0.13022456402288174, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019052431042230976, + "loss": 0.1445, + "step": 15002 + }, + { + "epoch": 0.1302332445030859, + "grad_norm": 0.10009765625, + "learning_rate": 0.0019052298107269062, + "loss": 0.1128, + "step": 15003 + }, + { + "epoch": 0.13024192498329007, + "grad_norm": 0.19140625, + "learning_rate": 0.0019052165163501273, + "loss": 0.0947, + "step": 15004 + }, + { + "epoch": 0.13025060546349423, + "grad_norm": 0.2578125, + "learning_rate": 0.001905203221092776, + "loss": 0.1445, + "step": 15005 + }, + { + "epoch": 0.1302592859436984, + "grad_norm": 0.41796875, + "learning_rate": 0.0019051899249548667, + "loss": 0.0854, + "step": 15006 + }, + { + "epoch": 0.13026796642390256, + "grad_norm": 0.73046875, + "learning_rate": 0.0019051766279364135, + "loss": 0.1562, + "step": 15007 + }, + { + "epoch": 0.13027664690410673, + "grad_norm": 0.400390625, + "learning_rate": 0.0019051633300374313, + "loss": 0.1211, + "step": 15008 + }, + { + "epoch": 0.1302853273843109, + "grad_norm": 0.703125, + "learning_rate": 0.0019051500312579348, + "loss": 0.1523, + "step": 15009 + }, + { + "epoch": 0.13029400786451506, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019051367315979382, + "loss": 0.1719, + "step": 15010 + }, + { + "epoch": 0.13030268834471923, + "grad_norm": 0.470703125, + "learning_rate": 0.0019051234310574561, + "loss": 0.1133, + "step": 15011 + }, + { + "epoch": 0.1303113688249234, + "grad_norm": 0.15234375, + "learning_rate": 0.001905110129636503, + "loss": 0.1084, + "step": 15012 + }, + { + "epoch": 0.13032004930512756, + "grad_norm": 0.18359375, + "learning_rate": 0.0019050968273350939, + "loss": 0.1045, + "step": 15013 + }, + { + "epoch": 0.13032872978533172, + "grad_norm": 0.140625, + "learning_rate": 0.0019050835241532426, + "loss": 0.1172, + "step": 15014 + }, + { + "epoch": 0.13033741026553589, + "grad_norm": 0.87109375, + "learning_rate": 0.0019050702200909644, + "loss": 0.2773, + "step": 15015 + }, + { + "epoch": 0.13034609074574005, + "grad_norm": 0.10595703125, + "learning_rate": 0.0019050569151482736, + "loss": 0.1338, + "step": 15016 + }, + { + "epoch": 0.13035477122594422, + "grad_norm": 0.51953125, + "learning_rate": 0.0019050436093251842, + "loss": 0.1543, + "step": 15017 + }, + { + "epoch": 0.13036345170614838, + "grad_norm": 0.5546875, + "learning_rate": 0.0019050303026217116, + "loss": 0.104, + "step": 15018 + }, + { + "epoch": 0.13037213218635255, + "grad_norm": 1.59375, + "learning_rate": 0.0019050169950378696, + "loss": 0.1094, + "step": 15019 + }, + { + "epoch": 0.1303808126665567, + "grad_norm": 0.294921875, + "learning_rate": 0.0019050036865736733, + "loss": 0.1484, + "step": 15020 + }, + { + "epoch": 0.13038949314676088, + "grad_norm": 0.1044921875, + "learning_rate": 0.001904990377229137, + "loss": 0.1133, + "step": 15021 + }, + { + "epoch": 0.13039817362696504, + "grad_norm": 0.169921875, + "learning_rate": 0.0019049770670042756, + "loss": 0.127, + "step": 15022 + }, + { + "epoch": 0.1304068541071692, + "grad_norm": 0.12451171875, + "learning_rate": 0.001904963755899103, + "loss": 0.1592, + "step": 15023 + }, + { + "epoch": 0.13041553458737337, + "grad_norm": 0.67578125, + "learning_rate": 0.0019049504439136342, + "loss": 0.1318, + "step": 15024 + }, + { + "epoch": 0.13042421506757754, + "grad_norm": 0.2060546875, + "learning_rate": 0.0019049371310478837, + "loss": 0.1104, + "step": 15025 + }, + { + "epoch": 0.1304328955477817, + "grad_norm": 0.2109375, + "learning_rate": 0.0019049238173018661, + "loss": 0.1523, + "step": 15026 + }, + { + "epoch": 0.13044157602798587, + "grad_norm": 0.2294921875, + "learning_rate": 0.0019049105026755961, + "loss": 0.1719, + "step": 15027 + }, + { + "epoch": 0.13045025650819003, + "grad_norm": 0.3359375, + "learning_rate": 0.001904897187169088, + "loss": 0.0986, + "step": 15028 + }, + { + "epoch": 0.1304589369883942, + "grad_norm": 0.287109375, + "learning_rate": 0.0019048838707823562, + "loss": 0.0913, + "step": 15029 + }, + { + "epoch": 0.13046761746859836, + "grad_norm": 0.1416015625, + "learning_rate": 0.0019048705535154157, + "loss": 0.1631, + "step": 15030 + }, + { + "epoch": 0.13047629794880253, + "grad_norm": 0.29296875, + "learning_rate": 0.0019048572353682807, + "loss": 0.1279, + "step": 15031 + }, + { + "epoch": 0.1304849784290067, + "grad_norm": 0.15625, + "learning_rate": 0.001904843916340966, + "loss": 0.1426, + "step": 15032 + }, + { + "epoch": 0.13049365890921086, + "grad_norm": 0.2314453125, + "learning_rate": 0.001904830596433486, + "loss": 0.1445, + "step": 15033 + }, + { + "epoch": 0.13050233938941502, + "grad_norm": 0.10791015625, + "learning_rate": 0.0019048172756458556, + "loss": 0.1182, + "step": 15034 + }, + { + "epoch": 0.1305110198696192, + "grad_norm": 0.2392578125, + "learning_rate": 0.0019048039539780889, + "loss": 0.1128, + "step": 15035 + }, + { + "epoch": 0.13051970034982335, + "grad_norm": 0.11962890625, + "learning_rate": 0.0019047906314302008, + "loss": 0.1299, + "step": 15036 + }, + { + "epoch": 0.13052838083002752, + "grad_norm": 0.251953125, + "learning_rate": 0.0019047773080022056, + "loss": 0.1084, + "step": 15037 + }, + { + "epoch": 0.13053706131023168, + "grad_norm": 0.271484375, + "learning_rate": 0.0019047639836941182, + "loss": 0.1035, + "step": 15038 + }, + { + "epoch": 0.13054574179043585, + "grad_norm": 0.4609375, + "learning_rate": 0.0019047506585059529, + "loss": 0.1045, + "step": 15039 + }, + { + "epoch": 0.13055442227064, + "grad_norm": 0.48828125, + "learning_rate": 0.0019047373324377244, + "loss": 0.1318, + "step": 15040 + }, + { + "epoch": 0.13056310275084418, + "grad_norm": 0.1318359375, + "learning_rate": 0.001904724005489447, + "loss": 0.1113, + "step": 15041 + }, + { + "epoch": 0.13057178323104834, + "grad_norm": 0.45703125, + "learning_rate": 0.001904710677661136, + "loss": 0.1128, + "step": 15042 + }, + { + "epoch": 0.1305804637112525, + "grad_norm": 0.53515625, + "learning_rate": 0.001904697348952805, + "loss": 0.0977, + "step": 15043 + }, + { + "epoch": 0.13058914419145667, + "grad_norm": 0.9765625, + "learning_rate": 0.0019046840193644695, + "loss": 0.2363, + "step": 15044 + }, + { + "epoch": 0.13059782467166084, + "grad_norm": 0.2734375, + "learning_rate": 0.0019046706888961435, + "loss": 0.1191, + "step": 15045 + }, + { + "epoch": 0.130606505151865, + "grad_norm": 0.298828125, + "learning_rate": 0.0019046573575478417, + "loss": 0.1143, + "step": 15046 + }, + { + "epoch": 0.13061518563206917, + "grad_norm": 0.515625, + "learning_rate": 0.0019046440253195785, + "loss": 0.1025, + "step": 15047 + }, + { + "epoch": 0.13062386611227333, + "grad_norm": 0.224609375, + "learning_rate": 0.001904630692211369, + "loss": 0.1182, + "step": 15048 + }, + { + "epoch": 0.1306325465924775, + "grad_norm": 0.49609375, + "learning_rate": 0.0019046173582232273, + "loss": 0.1172, + "step": 15049 + }, + { + "epoch": 0.13064122707268166, + "grad_norm": 0.267578125, + "learning_rate": 0.0019046040233551686, + "loss": 0.1299, + "step": 15050 + }, + { + "epoch": 0.13064990755288583, + "grad_norm": 0.416015625, + "learning_rate": 0.0019045906876072063, + "loss": 0.1191, + "step": 15051 + }, + { + "epoch": 0.13065858803309, + "grad_norm": 0.3515625, + "learning_rate": 0.001904577350979356, + "loss": 0.1064, + "step": 15052 + }, + { + "epoch": 0.13066726851329416, + "grad_norm": 0.7578125, + "learning_rate": 0.0019045640134716323, + "loss": 0.1758, + "step": 15053 + }, + { + "epoch": 0.13067594899349833, + "grad_norm": 0.123046875, + "learning_rate": 0.0019045506750840489, + "loss": 0.126, + "step": 15054 + }, + { + "epoch": 0.1306846294737025, + "grad_norm": 0.25, + "learning_rate": 0.0019045373358166215, + "loss": 0.1016, + "step": 15055 + }, + { + "epoch": 0.13069330995390666, + "grad_norm": 0.1328125, + "learning_rate": 0.0019045239956693638, + "loss": 0.1113, + "step": 15056 + }, + { + "epoch": 0.13070199043411082, + "grad_norm": 0.333984375, + "learning_rate": 0.0019045106546422908, + "loss": 0.1133, + "step": 15057 + }, + { + "epoch": 0.13071067091431499, + "grad_norm": 0.474609375, + "learning_rate": 0.001904497312735417, + "loss": 0.1416, + "step": 15058 + }, + { + "epoch": 0.13071935139451915, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019044839699487574, + "loss": 0.1279, + "step": 15059 + }, + { + "epoch": 0.13072803187472332, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019044706262823258, + "loss": 0.1133, + "step": 15060 + }, + { + "epoch": 0.13073671235492748, + "grad_norm": 0.162109375, + "learning_rate": 0.0019044572817361375, + "loss": 0.0747, + "step": 15061 + }, + { + "epoch": 0.13074539283513165, + "grad_norm": 0.48828125, + "learning_rate": 0.0019044439363102065, + "loss": 0.084, + "step": 15062 + }, + { + "epoch": 0.1307540733153358, + "grad_norm": 0.9765625, + "learning_rate": 0.001904430590004548, + "loss": 0.1328, + "step": 15063 + }, + { + "epoch": 0.13076275379553998, + "grad_norm": 0.154296875, + "learning_rate": 0.0019044172428191763, + "loss": 0.1069, + "step": 15064 + }, + { + "epoch": 0.13077143427574414, + "grad_norm": 0.1982421875, + "learning_rate": 0.0019044038947541055, + "loss": 0.1001, + "step": 15065 + }, + { + "epoch": 0.1307801147559483, + "grad_norm": 0.134765625, + "learning_rate": 0.0019043905458093514, + "loss": 0.1309, + "step": 15066 + }, + { + "epoch": 0.13078879523615247, + "grad_norm": 0.083984375, + "learning_rate": 0.0019043771959849274, + "loss": 0.1016, + "step": 15067 + }, + { + "epoch": 0.13079747571635664, + "grad_norm": 0.328125, + "learning_rate": 0.0019043638452808486, + "loss": 0.0986, + "step": 15068 + }, + { + "epoch": 0.1308061561965608, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019043504936971296, + "loss": 0.0981, + "step": 15069 + }, + { + "epoch": 0.13081483667676497, + "grad_norm": 0.08251953125, + "learning_rate": 0.0019043371412337854, + "loss": 0.0947, + "step": 15070 + }, + { + "epoch": 0.13082351715696913, + "grad_norm": 0.451171875, + "learning_rate": 0.0019043237878908297, + "loss": 0.1182, + "step": 15071 + }, + { + "epoch": 0.1308321976371733, + "grad_norm": 0.4765625, + "learning_rate": 0.001904310433668278, + "loss": 0.1279, + "step": 15072 + }, + { + "epoch": 0.13084087811737746, + "grad_norm": 0.25390625, + "learning_rate": 0.0019042970785661441, + "loss": 0.1416, + "step": 15073 + }, + { + "epoch": 0.13084955859758163, + "grad_norm": 0.578125, + "learning_rate": 0.0019042837225844431, + "loss": 0.1123, + "step": 15074 + }, + { + "epoch": 0.1308582390777858, + "grad_norm": 0.30859375, + "learning_rate": 0.0019042703657231896, + "loss": 0.1494, + "step": 15075 + }, + { + "epoch": 0.13086691955798996, + "grad_norm": 0.1103515625, + "learning_rate": 0.001904257007982398, + "loss": 0.1201, + "step": 15076 + }, + { + "epoch": 0.13087560003819412, + "grad_norm": 0.21484375, + "learning_rate": 0.0019042436493620831, + "loss": 0.0986, + "step": 15077 + }, + { + "epoch": 0.1308842805183983, + "grad_norm": 0.3515625, + "learning_rate": 0.0019042302898622594, + "loss": 0.084, + "step": 15078 + }, + { + "epoch": 0.13089296099860245, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019042169294829416, + "loss": 0.0903, + "step": 15079 + }, + { + "epoch": 0.13090164147880662, + "grad_norm": 0.34375, + "learning_rate": 0.0019042035682241443, + "loss": 0.1113, + "step": 15080 + }, + { + "epoch": 0.13091032195901078, + "grad_norm": 0.26171875, + "learning_rate": 0.0019041902060858817, + "loss": 0.0762, + "step": 15081 + }, + { + "epoch": 0.13091900243921495, + "grad_norm": 0.125, + "learning_rate": 0.0019041768430681694, + "loss": 0.1084, + "step": 15082 + }, + { + "epoch": 0.1309276829194191, + "grad_norm": 0.83203125, + "learning_rate": 0.001904163479171021, + "loss": 0.1011, + "step": 15083 + }, + { + "epoch": 0.13093636339962328, + "grad_norm": 0.4375, + "learning_rate": 0.0019041501143944514, + "loss": 0.1514, + "step": 15084 + }, + { + "epoch": 0.13094504387982744, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019041367487384756, + "loss": 0.0625, + "step": 15085 + }, + { + "epoch": 0.1309537243600316, + "grad_norm": 0.302734375, + "learning_rate": 0.0019041233822031079, + "loss": 0.1367, + "step": 15086 + }, + { + "epoch": 0.13096240484023577, + "grad_norm": 0.8203125, + "learning_rate": 0.0019041100147883627, + "loss": 0.1416, + "step": 15087 + }, + { + "epoch": 0.13097108532043994, + "grad_norm": 0.328125, + "learning_rate": 0.0019040966464942551, + "loss": 0.1328, + "step": 15088 + }, + { + "epoch": 0.1309797658006441, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019040832773207991, + "loss": 0.1045, + "step": 15089 + }, + { + "epoch": 0.13098844628084827, + "grad_norm": 0.26171875, + "learning_rate": 0.00190406990726801, + "loss": 0.123, + "step": 15090 + }, + { + "epoch": 0.13099712676105243, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019040565363359024, + "loss": 0.127, + "step": 15091 + }, + { + "epoch": 0.1310058072412566, + "grad_norm": 0.59375, + "learning_rate": 0.0019040431645244902, + "loss": 0.1035, + "step": 15092 + }, + { + "epoch": 0.13101448772146077, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019040297918337889, + "loss": 0.1025, + "step": 15093 + }, + { + "epoch": 0.1310231682016649, + "grad_norm": 0.09033203125, + "learning_rate": 0.0019040164182638122, + "loss": 0.1172, + "step": 15094 + }, + { + "epoch": 0.13103184868186907, + "grad_norm": 0.337890625, + "learning_rate": 0.0019040030438145757, + "loss": 0.1094, + "step": 15095 + }, + { + "epoch": 0.13104052916207323, + "grad_norm": 0.2333984375, + "learning_rate": 0.0019039896684860931, + "loss": 0.1221, + "step": 15096 + }, + { + "epoch": 0.1310492096422774, + "grad_norm": 0.2255859375, + "learning_rate": 0.0019039762922783798, + "loss": 0.1123, + "step": 15097 + }, + { + "epoch": 0.13105789012248156, + "grad_norm": 0.22265625, + "learning_rate": 0.0019039629151914501, + "loss": 0.1201, + "step": 15098 + }, + { + "epoch": 0.13106657060268573, + "grad_norm": 0.2099609375, + "learning_rate": 0.0019039495372253185, + "loss": 0.1211, + "step": 15099 + }, + { + "epoch": 0.1310752510828899, + "grad_norm": 0.9765625, + "learning_rate": 0.0019039361583799996, + "loss": 0.0933, + "step": 15100 + }, + { + "epoch": 0.13108393156309406, + "grad_norm": 0.08447265625, + "learning_rate": 0.0019039227786555087, + "loss": 0.1064, + "step": 15101 + }, + { + "epoch": 0.13109261204329822, + "grad_norm": 0.34375, + "learning_rate": 0.0019039093980518596, + "loss": 0.1035, + "step": 15102 + }, + { + "epoch": 0.1311012925235024, + "grad_norm": 0.416015625, + "learning_rate": 0.0019038960165690673, + "loss": 0.1035, + "step": 15103 + }, + { + "epoch": 0.13110997300370655, + "grad_norm": 0.18359375, + "learning_rate": 0.0019038826342071464, + "loss": 0.1099, + "step": 15104 + }, + { + "epoch": 0.13111865348391072, + "grad_norm": 0.1025390625, + "learning_rate": 0.0019038692509661117, + "loss": 0.1377, + "step": 15105 + }, + { + "epoch": 0.13112733396411488, + "grad_norm": 0.423828125, + "learning_rate": 0.0019038558668459775, + "loss": 0.1055, + "step": 15106 + }, + { + "epoch": 0.13113601444431905, + "grad_norm": 0.142578125, + "learning_rate": 0.0019038424818467587, + "loss": 0.0791, + "step": 15107 + }, + { + "epoch": 0.13114469492452321, + "grad_norm": 0.21484375, + "learning_rate": 0.0019038290959684694, + "loss": 0.1289, + "step": 15108 + }, + { + "epoch": 0.13115337540472738, + "grad_norm": 0.3125, + "learning_rate": 0.0019038157092111253, + "loss": 0.0967, + "step": 15109 + }, + { + "epoch": 0.13116205588493154, + "grad_norm": 0.478515625, + "learning_rate": 0.00190380232157474, + "loss": 0.1406, + "step": 15110 + }, + { + "epoch": 0.1311707363651357, + "grad_norm": 0.0859375, + "learning_rate": 0.0019037889330593286, + "loss": 0.1377, + "step": 15111 + }, + { + "epoch": 0.13117941684533987, + "grad_norm": 0.2734375, + "learning_rate": 0.0019037755436649058, + "loss": 0.1299, + "step": 15112 + }, + { + "epoch": 0.13118809732554404, + "grad_norm": 0.3515625, + "learning_rate": 0.0019037621533914862, + "loss": 0.1133, + "step": 15113 + }, + { + "epoch": 0.1311967778057482, + "grad_norm": 0.6640625, + "learning_rate": 0.0019037487622390844, + "loss": 0.0967, + "step": 15114 + }, + { + "epoch": 0.13120545828595237, + "grad_norm": 0.330078125, + "learning_rate": 0.001903735370207715, + "loss": 0.0771, + "step": 15115 + }, + { + "epoch": 0.13121413876615654, + "grad_norm": 1.046875, + "learning_rate": 0.0019037219772973927, + "loss": 0.1377, + "step": 15116 + }, + { + "epoch": 0.1312228192463607, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019037085835081318, + "loss": 0.124, + "step": 15117 + }, + { + "epoch": 0.13123149972656487, + "grad_norm": 0.41796875, + "learning_rate": 0.0019036951888399473, + "loss": 0.1484, + "step": 15118 + }, + { + "epoch": 0.13124018020676903, + "grad_norm": 0.408203125, + "learning_rate": 0.0019036817932928543, + "loss": 0.1455, + "step": 15119 + }, + { + "epoch": 0.1312488606869732, + "grad_norm": 0.296875, + "learning_rate": 0.0019036683968668667, + "loss": 0.1143, + "step": 15120 + }, + { + "epoch": 0.13125754116717736, + "grad_norm": 0.060302734375, + "learning_rate": 0.0019036549995619996, + "loss": 0.1016, + "step": 15121 + }, + { + "epoch": 0.13126622164738153, + "grad_norm": 0.11669921875, + "learning_rate": 0.001903641601378267, + "loss": 0.0952, + "step": 15122 + }, + { + "epoch": 0.1312749021275857, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019036282023156843, + "loss": 0.0762, + "step": 15123 + }, + { + "epoch": 0.13128358260778986, + "grad_norm": 0.37109375, + "learning_rate": 0.0019036148023742659, + "loss": 0.1602, + "step": 15124 + }, + { + "epoch": 0.13129226308799402, + "grad_norm": 0.5078125, + "learning_rate": 0.0019036014015540263, + "loss": 0.125, + "step": 15125 + }, + { + "epoch": 0.1313009435681982, + "grad_norm": 0.3046875, + "learning_rate": 0.0019035879998549805, + "loss": 0.1191, + "step": 15126 + }, + { + "epoch": 0.13130962404840235, + "grad_norm": 0.33984375, + "learning_rate": 0.0019035745972771428, + "loss": 0.127, + "step": 15127 + }, + { + "epoch": 0.13131830452860652, + "grad_norm": 0.64453125, + "learning_rate": 0.0019035611938205282, + "loss": 0.3496, + "step": 15128 + }, + { + "epoch": 0.13132698500881068, + "grad_norm": 0.0927734375, + "learning_rate": 0.0019035477894851509, + "loss": 0.1309, + "step": 15129 + }, + { + "epoch": 0.13133566548901485, + "grad_norm": 1.015625, + "learning_rate": 0.0019035343842710257, + "loss": 0.1758, + "step": 15130 + }, + { + "epoch": 0.131344345969219, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019035209781781675, + "loss": 0.126, + "step": 15131 + }, + { + "epoch": 0.13135302644942318, + "grad_norm": 0.2109375, + "learning_rate": 0.001903507571206591, + "loss": 0.1914, + "step": 15132 + }, + { + "epoch": 0.13136170692962734, + "grad_norm": 0.9140625, + "learning_rate": 0.0019034941633563106, + "loss": 0.1055, + "step": 15133 + }, + { + "epoch": 0.1313703874098315, + "grad_norm": 0.8125, + "learning_rate": 0.001903480754627341, + "loss": 0.1562, + "step": 15134 + }, + { + "epoch": 0.13137906789003567, + "grad_norm": 0.189453125, + "learning_rate": 0.0019034673450196969, + "loss": 0.1328, + "step": 15135 + }, + { + "epoch": 0.13138774837023984, + "grad_norm": 0.89453125, + "learning_rate": 0.001903453934533393, + "loss": 0.0962, + "step": 15136 + }, + { + "epoch": 0.131396428850444, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019034405231684443, + "loss": 0.1318, + "step": 15137 + }, + { + "epoch": 0.13140510933064817, + "grad_norm": 0.7734375, + "learning_rate": 0.0019034271109248646, + "loss": 0.1445, + "step": 15138 + }, + { + "epoch": 0.13141378981085233, + "grad_norm": 0.1259765625, + "learning_rate": 0.0019034136978026692, + "loss": 0.1211, + "step": 15139 + }, + { + "epoch": 0.1314224702910565, + "grad_norm": 0.2080078125, + "learning_rate": 0.0019034002838018726, + "loss": 0.1128, + "step": 15140 + }, + { + "epoch": 0.13143115077126066, + "grad_norm": 0.318359375, + "learning_rate": 0.0019033868689224897, + "loss": 0.1855, + "step": 15141 + }, + { + "epoch": 0.13143983125146483, + "grad_norm": 0.345703125, + "learning_rate": 0.0019033734531645353, + "loss": 0.166, + "step": 15142 + }, + { + "epoch": 0.131448511731669, + "grad_norm": 0.3671875, + "learning_rate": 0.0019033600365280232, + "loss": 0.1211, + "step": 15143 + }, + { + "epoch": 0.13145719221187316, + "grad_norm": 0.5859375, + "learning_rate": 0.001903346619012969, + "loss": 0.1104, + "step": 15144 + }, + { + "epoch": 0.13146587269207732, + "grad_norm": 1.2421875, + "learning_rate": 0.0019033332006193867, + "loss": 0.1182, + "step": 15145 + }, + { + "epoch": 0.1314745531722815, + "grad_norm": 1.1484375, + "learning_rate": 0.0019033197813472914, + "loss": 0.1641, + "step": 15146 + }, + { + "epoch": 0.13148323365248565, + "grad_norm": 0.37109375, + "learning_rate": 0.0019033063611966978, + "loss": 0.1592, + "step": 15147 + }, + { + "epoch": 0.13149191413268982, + "grad_norm": 0.1591796875, + "learning_rate": 0.0019032929401676202, + "loss": 0.1113, + "step": 15148 + }, + { + "epoch": 0.13150059461289398, + "grad_norm": 0.34375, + "learning_rate": 0.0019032795182600734, + "loss": 0.0991, + "step": 15149 + }, + { + "epoch": 0.13150927509309815, + "grad_norm": 0.26171875, + "learning_rate": 0.0019032660954740727, + "loss": 0.1045, + "step": 15150 + }, + { + "epoch": 0.13151795557330231, + "grad_norm": 0.3125, + "learning_rate": 0.001903252671809632, + "loss": 0.1279, + "step": 15151 + }, + { + "epoch": 0.13152663605350648, + "grad_norm": 0.21875, + "learning_rate": 0.0019032392472667662, + "loss": 0.1309, + "step": 15152 + }, + { + "epoch": 0.13153531653371064, + "grad_norm": 0.78515625, + "learning_rate": 0.00190322582184549, + "loss": 0.1045, + "step": 15153 + }, + { + "epoch": 0.1315439970139148, + "grad_norm": 0.57421875, + "learning_rate": 0.0019032123955458182, + "loss": 0.127, + "step": 15154 + }, + { + "epoch": 0.13155267749411897, + "grad_norm": 0.439453125, + "learning_rate": 0.0019031989683677654, + "loss": 0.0996, + "step": 15155 + }, + { + "epoch": 0.13156135797432314, + "grad_norm": 0.390625, + "learning_rate": 0.001903185540311346, + "loss": 0.1221, + "step": 15156 + }, + { + "epoch": 0.1315700384545273, + "grad_norm": 0.470703125, + "learning_rate": 0.0019031721113765753, + "loss": 0.1328, + "step": 15157 + }, + { + "epoch": 0.13157871893473147, + "grad_norm": 0.49609375, + "learning_rate": 0.0019031586815634676, + "loss": 0.1221, + "step": 15158 + }, + { + "epoch": 0.13158739941493564, + "grad_norm": 0.2890625, + "learning_rate": 0.0019031452508720375, + "loss": 0.1289, + "step": 15159 + }, + { + "epoch": 0.1315960798951398, + "grad_norm": 0.140625, + "learning_rate": 0.0019031318193023, + "loss": 0.124, + "step": 15160 + }, + { + "epoch": 0.13160476037534397, + "grad_norm": 0.1787109375, + "learning_rate": 0.0019031183868542694, + "loss": 0.1094, + "step": 15161 + }, + { + "epoch": 0.13161344085554813, + "grad_norm": 0.76171875, + "learning_rate": 0.0019031049535279607, + "loss": 0.1348, + "step": 15162 + }, + { + "epoch": 0.1316221213357523, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019030915193233887, + "loss": 0.1084, + "step": 15163 + }, + { + "epoch": 0.13163080181595646, + "grad_norm": 0.55078125, + "learning_rate": 0.0019030780842405675, + "loss": 0.1758, + "step": 15164 + }, + { + "epoch": 0.13163948229616063, + "grad_norm": 0.734375, + "learning_rate": 0.0019030646482795124, + "loss": 0.1201, + "step": 15165 + }, + { + "epoch": 0.1316481627763648, + "grad_norm": 0.419921875, + "learning_rate": 0.0019030512114402377, + "loss": 0.1523, + "step": 15166 + }, + { + "epoch": 0.13165684325656896, + "grad_norm": 0.1953125, + "learning_rate": 0.0019030377737227586, + "loss": 0.124, + "step": 15167 + }, + { + "epoch": 0.13166552373677312, + "grad_norm": 0.1416015625, + "learning_rate": 0.001903024335127089, + "loss": 0.1147, + "step": 15168 + }, + { + "epoch": 0.1316742042169773, + "grad_norm": 0.150390625, + "learning_rate": 0.0019030108956532446, + "loss": 0.1582, + "step": 15169 + }, + { + "epoch": 0.13168288469718145, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019029974553012392, + "loss": 0.1152, + "step": 15170 + }, + { + "epoch": 0.13169156517738562, + "grad_norm": 0.5625, + "learning_rate": 0.001902984014071088, + "loss": 0.1484, + "step": 15171 + }, + { + "epoch": 0.13170024565758978, + "grad_norm": 0.1220703125, + "learning_rate": 0.001902970571962805, + "loss": 0.1084, + "step": 15172 + }, + { + "epoch": 0.13170892613779395, + "grad_norm": 0.1640625, + "learning_rate": 0.001902957128976406, + "loss": 0.1279, + "step": 15173 + }, + { + "epoch": 0.1317176066179981, + "grad_norm": 0.1162109375, + "learning_rate": 0.0019029436851119047, + "loss": 0.1104, + "step": 15174 + }, + { + "epoch": 0.13172628709820228, + "grad_norm": 0.490234375, + "learning_rate": 0.0019029302403693168, + "loss": 0.0918, + "step": 15175 + }, + { + "epoch": 0.13173496757840644, + "grad_norm": 0.28125, + "learning_rate": 0.0019029167947486564, + "loss": 0.1387, + "step": 15176 + }, + { + "epoch": 0.1317436480586106, + "grad_norm": 0.75, + "learning_rate": 0.0019029033482499378, + "loss": 0.0977, + "step": 15177 + }, + { + "epoch": 0.13175232853881477, + "grad_norm": 0.43359375, + "learning_rate": 0.0019028899008731767, + "loss": 0.0977, + "step": 15178 + }, + { + "epoch": 0.13176100901901894, + "grad_norm": 0.671875, + "learning_rate": 0.0019028764526183867, + "loss": 0.084, + "step": 15179 + }, + { + "epoch": 0.1317696894992231, + "grad_norm": 0.248046875, + "learning_rate": 0.0019028630034855839, + "loss": 0.0898, + "step": 15180 + }, + { + "epoch": 0.13177836997942727, + "grad_norm": 0.365234375, + "learning_rate": 0.0019028495534747815, + "loss": 0.1191, + "step": 15181 + }, + { + "epoch": 0.13178705045963143, + "grad_norm": 0.43359375, + "learning_rate": 0.001902836102585995, + "loss": 0.1191, + "step": 15182 + }, + { + "epoch": 0.1317957309398356, + "grad_norm": 0.5546875, + "learning_rate": 0.001902822650819239, + "loss": 0.125, + "step": 15183 + }, + { + "epoch": 0.13180441142003976, + "grad_norm": 0.275390625, + "learning_rate": 0.0019028091981745283, + "loss": 0.1221, + "step": 15184 + }, + { + "epoch": 0.13181309190024393, + "grad_norm": 0.2412109375, + "learning_rate": 0.0019027957446518776, + "loss": 0.0908, + "step": 15185 + }, + { + "epoch": 0.1318217723804481, + "grad_norm": 0.373046875, + "learning_rate": 0.0019027822902513013, + "loss": 0.1562, + "step": 15186 + }, + { + "epoch": 0.13183045286065226, + "grad_norm": 0.421875, + "learning_rate": 0.0019027688349728146, + "loss": 0.1001, + "step": 15187 + }, + { + "epoch": 0.13183913334085642, + "grad_norm": 0.1455078125, + "learning_rate": 0.001902755378816432, + "loss": 0.0957, + "step": 15188 + }, + { + "epoch": 0.1318478138210606, + "grad_norm": 0.55859375, + "learning_rate": 0.0019027419217821679, + "loss": 0.1318, + "step": 15189 + }, + { + "epoch": 0.13185649430126475, + "grad_norm": 0.66796875, + "learning_rate": 0.0019027284638700374, + "loss": 0.1289, + "step": 15190 + }, + { + "epoch": 0.13186517478146892, + "grad_norm": 0.96484375, + "learning_rate": 0.001902715005080055, + "loss": 0.1133, + "step": 15191 + }, + { + "epoch": 0.13187385526167308, + "grad_norm": 0.154296875, + "learning_rate": 0.0019027015454122357, + "loss": 0.1211, + "step": 15192 + }, + { + "epoch": 0.13188253574187725, + "grad_norm": 0.08349609375, + "learning_rate": 0.0019026880848665943, + "loss": 0.1162, + "step": 15193 + }, + { + "epoch": 0.13189121622208141, + "grad_norm": 0.1572265625, + "learning_rate": 0.0019026746234431446, + "loss": 0.1279, + "step": 15194 + }, + { + "epoch": 0.13189989670228558, + "grad_norm": 0.185546875, + "learning_rate": 0.0019026611611419025, + "loss": 0.1104, + "step": 15195 + }, + { + "epoch": 0.13190857718248974, + "grad_norm": 0.107421875, + "learning_rate": 0.0019026476979628819, + "loss": 0.0942, + "step": 15196 + }, + { + "epoch": 0.1319172576626939, + "grad_norm": 0.0849609375, + "learning_rate": 0.0019026342339060983, + "loss": 0.1211, + "step": 15197 + }, + { + "epoch": 0.13192593814289807, + "grad_norm": 0.279296875, + "learning_rate": 0.0019026207689715655, + "loss": 0.1104, + "step": 15198 + }, + { + "epoch": 0.13193461862310224, + "grad_norm": 0.60546875, + "learning_rate": 0.001902607303159299, + "loss": 0.1494, + "step": 15199 + }, + { + "epoch": 0.1319432991033064, + "grad_norm": 0.2021484375, + "learning_rate": 0.001902593836469313, + "loss": 0.0845, + "step": 15200 + }, + { + "epoch": 0.13195197958351057, + "grad_norm": 0.091796875, + "learning_rate": 0.0019025803689016226, + "loss": 0.1328, + "step": 15201 + }, + { + "epoch": 0.13196066006371474, + "grad_norm": 0.279296875, + "learning_rate": 0.0019025669004562424, + "loss": 0.1025, + "step": 15202 + }, + { + "epoch": 0.1319693405439189, + "grad_norm": 0.07568359375, + "learning_rate": 0.0019025534311331868, + "loss": 0.0869, + "step": 15203 + }, + { + "epoch": 0.13197802102412307, + "grad_norm": 0.06201171875, + "learning_rate": 0.001902539960932471, + "loss": 0.0962, + "step": 15204 + }, + { + "epoch": 0.13198670150432723, + "grad_norm": 0.42578125, + "learning_rate": 0.0019025264898541096, + "loss": 0.125, + "step": 15205 + }, + { + "epoch": 0.1319953819845314, + "grad_norm": 0.08544921875, + "learning_rate": 0.0019025130178981174, + "loss": 0.1133, + "step": 15206 + }, + { + "epoch": 0.13200406246473556, + "grad_norm": 0.0830078125, + "learning_rate": 0.0019024995450645087, + "loss": 0.1226, + "step": 15207 + }, + { + "epoch": 0.13201274294493973, + "grad_norm": 0.19140625, + "learning_rate": 0.0019024860713532988, + "loss": 0.0859, + "step": 15208 + }, + { + "epoch": 0.1320214234251439, + "grad_norm": 0.296875, + "learning_rate": 0.001902472596764502, + "loss": 0.126, + "step": 15209 + }, + { + "epoch": 0.13203010390534806, + "grad_norm": 1.0859375, + "learning_rate": 0.0019024591212981337, + "loss": 0.127, + "step": 15210 + }, + { + "epoch": 0.13203878438555222, + "grad_norm": 0.2578125, + "learning_rate": 0.0019024456449542077, + "loss": 0.1289, + "step": 15211 + }, + { + "epoch": 0.1320474648657564, + "grad_norm": 0.1298828125, + "learning_rate": 0.0019024321677327393, + "loss": 0.1221, + "step": 15212 + }, + { + "epoch": 0.13205614534596055, + "grad_norm": 0.1650390625, + "learning_rate": 0.0019024186896337432, + "loss": 0.0928, + "step": 15213 + }, + { + "epoch": 0.13206482582616472, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019024052106572339, + "loss": 0.0718, + "step": 15214 + }, + { + "epoch": 0.13207350630636888, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019023917308032263, + "loss": 0.0938, + "step": 15215 + }, + { + "epoch": 0.13208218678657305, + "grad_norm": 0.41015625, + "learning_rate": 0.0019023782500717354, + "loss": 0.1279, + "step": 15216 + }, + { + "epoch": 0.13209086726677718, + "grad_norm": 0.146484375, + "learning_rate": 0.0019023647684627755, + "loss": 0.0962, + "step": 15217 + }, + { + "epoch": 0.13209954774698135, + "grad_norm": 0.1513671875, + "learning_rate": 0.0019023512859763618, + "loss": 0.0996, + "step": 15218 + }, + { + "epoch": 0.13210822822718551, + "grad_norm": 0.228515625, + "learning_rate": 0.0019023378026125086, + "loss": 0.1611, + "step": 15219 + }, + { + "epoch": 0.13211690870738968, + "grad_norm": 0.251953125, + "learning_rate": 0.001902324318371231, + "loss": 0.1309, + "step": 15220 + }, + { + "epoch": 0.13212558918759384, + "grad_norm": 0.50390625, + "learning_rate": 0.0019023108332525433, + "loss": 0.1152, + "step": 15221 + }, + { + "epoch": 0.132134269667798, + "grad_norm": 0.17578125, + "learning_rate": 0.0019022973472564608, + "loss": 0.1445, + "step": 15222 + }, + { + "epoch": 0.13214295014800218, + "grad_norm": 0.0908203125, + "learning_rate": 0.0019022838603829978, + "loss": 0.1035, + "step": 15223 + }, + { + "epoch": 0.13215163062820634, + "grad_norm": 0.10400390625, + "learning_rate": 0.0019022703726321692, + "loss": 0.1406, + "step": 15224 + }, + { + "epoch": 0.1321603111084105, + "grad_norm": 0.5, + "learning_rate": 0.00190225688400399, + "loss": 0.1367, + "step": 15225 + }, + { + "epoch": 0.13216899158861467, + "grad_norm": 0.7578125, + "learning_rate": 0.0019022433944984745, + "loss": 0.1035, + "step": 15226 + }, + { + "epoch": 0.13217767206881884, + "grad_norm": 0.271484375, + "learning_rate": 0.0019022299041156377, + "loss": 0.1719, + "step": 15227 + }, + { + "epoch": 0.132186352549023, + "grad_norm": 0.1357421875, + "learning_rate": 0.0019022164128554943, + "loss": 0.127, + "step": 15228 + }, + { + "epoch": 0.13219503302922717, + "grad_norm": 0.1318359375, + "learning_rate": 0.0019022029207180593, + "loss": 0.1143, + "step": 15229 + }, + { + "epoch": 0.13220371350943133, + "grad_norm": 0.484375, + "learning_rate": 0.0019021894277033472, + "loss": 0.1309, + "step": 15230 + }, + { + "epoch": 0.1322123939896355, + "grad_norm": 0.2431640625, + "learning_rate": 0.0019021759338113727, + "loss": 0.1045, + "step": 15231 + }, + { + "epoch": 0.13222107446983966, + "grad_norm": 0.240234375, + "learning_rate": 0.0019021624390421507, + "loss": 0.125, + "step": 15232 + }, + { + "epoch": 0.13222975495004383, + "grad_norm": 0.44140625, + "learning_rate": 0.0019021489433956958, + "loss": 0.166, + "step": 15233 + }, + { + "epoch": 0.132238435430248, + "grad_norm": 0.5625, + "learning_rate": 0.0019021354468720228, + "loss": 0.1201, + "step": 15234 + }, + { + "epoch": 0.13224711591045216, + "grad_norm": 0.0732421875, + "learning_rate": 0.0019021219494711468, + "loss": 0.1104, + "step": 15235 + }, + { + "epoch": 0.13225579639065632, + "grad_norm": 0.1865234375, + "learning_rate": 0.0019021084511930825, + "loss": 0.124, + "step": 15236 + }, + { + "epoch": 0.1322644768708605, + "grad_norm": 0.126953125, + "learning_rate": 0.001902094952037844, + "loss": 0.1108, + "step": 15237 + }, + { + "epoch": 0.13227315735106465, + "grad_norm": 0.1826171875, + "learning_rate": 0.0019020814520054464, + "loss": 0.1094, + "step": 15238 + }, + { + "epoch": 0.13228183783126882, + "grad_norm": 0.478515625, + "learning_rate": 0.0019020679510959046, + "loss": 0.0845, + "step": 15239 + }, + { + "epoch": 0.13229051831147298, + "grad_norm": 0.19140625, + "learning_rate": 0.0019020544493092337, + "loss": 0.0898, + "step": 15240 + }, + { + "epoch": 0.13229919879167715, + "grad_norm": 0.5390625, + "learning_rate": 0.0019020409466454478, + "loss": 0.0986, + "step": 15241 + }, + { + "epoch": 0.1323078792718813, + "grad_norm": 0.365234375, + "learning_rate": 0.0019020274431045623, + "loss": 0.1543, + "step": 15242 + }, + { + "epoch": 0.13231655975208548, + "grad_norm": 0.193359375, + "learning_rate": 0.0019020139386865913, + "loss": 0.1128, + "step": 15243 + }, + { + "epoch": 0.13232524023228964, + "grad_norm": 0.64453125, + "learning_rate": 0.0019020004333915501, + "loss": 0.0952, + "step": 15244 + }, + { + "epoch": 0.1323339207124938, + "grad_norm": 0.162109375, + "learning_rate": 0.0019019869272194534, + "loss": 0.1211, + "step": 15245 + }, + { + "epoch": 0.13234260119269797, + "grad_norm": 0.2158203125, + "learning_rate": 0.0019019734201703154, + "loss": 0.1416, + "step": 15246 + }, + { + "epoch": 0.13235128167290214, + "grad_norm": 0.8203125, + "learning_rate": 0.0019019599122441517, + "loss": 0.1074, + "step": 15247 + }, + { + "epoch": 0.1323599621531063, + "grad_norm": 0.33984375, + "learning_rate": 0.0019019464034409767, + "loss": 0.1377, + "step": 15248 + }, + { + "epoch": 0.13236864263331047, + "grad_norm": 0.36328125, + "learning_rate": 0.001901932893760805, + "loss": 0.1162, + "step": 15249 + }, + { + "epoch": 0.13237732311351463, + "grad_norm": 0.283203125, + "learning_rate": 0.0019019193832036515, + "loss": 0.0957, + "step": 15250 + }, + { + "epoch": 0.1323860035937188, + "grad_norm": 0.1943359375, + "learning_rate": 0.001901905871769531, + "loss": 0.1816, + "step": 15251 + }, + { + "epoch": 0.13239468407392296, + "grad_norm": 0.48046875, + "learning_rate": 0.0019018923594584585, + "loss": 0.1875, + "step": 15252 + }, + { + "epoch": 0.13240336455412713, + "grad_norm": 0.1142578125, + "learning_rate": 0.0019018788462704483, + "loss": 0.1445, + "step": 15253 + }, + { + "epoch": 0.1324120450343313, + "grad_norm": 0.189453125, + "learning_rate": 0.0019018653322055157, + "loss": 0.1182, + "step": 15254 + }, + { + "epoch": 0.13242072551453546, + "grad_norm": 0.12890625, + "learning_rate": 0.0019018518172636751, + "loss": 0.1045, + "step": 15255 + }, + { + "epoch": 0.13242940599473962, + "grad_norm": 0.115234375, + "learning_rate": 0.0019018383014449414, + "loss": 0.1025, + "step": 15256 + }, + { + "epoch": 0.1324380864749438, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019018247847493294, + "loss": 0.0918, + "step": 15257 + }, + { + "epoch": 0.13244676695514795, + "grad_norm": 0.19140625, + "learning_rate": 0.001901811267176854, + "loss": 0.1211, + "step": 15258 + }, + { + "epoch": 0.13245544743535212, + "grad_norm": 1.09375, + "learning_rate": 0.0019017977487275296, + "loss": 0.1084, + "step": 15259 + }, + { + "epoch": 0.13246412791555628, + "grad_norm": 0.1962890625, + "learning_rate": 0.0019017842294013713, + "loss": 0.1143, + "step": 15260 + }, + { + "epoch": 0.13247280839576045, + "grad_norm": 0.09619140625, + "learning_rate": 0.0019017707091983937, + "loss": 0.1328, + "step": 15261 + }, + { + "epoch": 0.13248148887596461, + "grad_norm": 0.546875, + "learning_rate": 0.0019017571881186118, + "loss": 0.1436, + "step": 15262 + }, + { + "epoch": 0.13249016935616878, + "grad_norm": 0.6171875, + "learning_rate": 0.0019017436661620407, + "loss": 0.1211, + "step": 15263 + }, + { + "epoch": 0.13249884983637294, + "grad_norm": 0.240234375, + "learning_rate": 0.0019017301433286943, + "loss": 0.085, + "step": 15264 + }, + { + "epoch": 0.1325075303165771, + "grad_norm": 0.26171875, + "learning_rate": 0.0019017166196185877, + "loss": 0.1348, + "step": 15265 + }, + { + "epoch": 0.13251621079678128, + "grad_norm": 0.388671875, + "learning_rate": 0.0019017030950317362, + "loss": 0.1064, + "step": 15266 + }, + { + "epoch": 0.13252489127698544, + "grad_norm": 0.28125, + "learning_rate": 0.0019016895695681542, + "loss": 0.1426, + "step": 15267 + }, + { + "epoch": 0.1325335717571896, + "grad_norm": 0.09228515625, + "learning_rate": 0.0019016760432278563, + "loss": 0.0874, + "step": 15268 + }, + { + "epoch": 0.13254225223739377, + "grad_norm": 0.271484375, + "learning_rate": 0.0019016625160108579, + "loss": 0.1465, + "step": 15269 + }, + { + "epoch": 0.13255093271759794, + "grad_norm": 0.1064453125, + "learning_rate": 0.0019016489879171731, + "loss": 0.1406, + "step": 15270 + }, + { + "epoch": 0.1325596131978021, + "grad_norm": 0.2138671875, + "learning_rate": 0.0019016354589468172, + "loss": 0.1309, + "step": 15271 + }, + { + "epoch": 0.13256829367800627, + "grad_norm": 0.26171875, + "learning_rate": 0.0019016219290998047, + "loss": 0.1602, + "step": 15272 + }, + { + "epoch": 0.13257697415821043, + "grad_norm": 0.421875, + "learning_rate": 0.0019016083983761506, + "loss": 0.1016, + "step": 15273 + }, + { + "epoch": 0.1325856546384146, + "grad_norm": 0.1337890625, + "learning_rate": 0.0019015948667758696, + "loss": 0.1289, + "step": 15274 + }, + { + "epoch": 0.13259433511861876, + "grad_norm": 0.2216796875, + "learning_rate": 0.0019015813342989766, + "loss": 0.1182, + "step": 15275 + }, + { + "epoch": 0.13260301559882293, + "grad_norm": 0.255859375, + "learning_rate": 0.0019015678009454858, + "loss": 0.0889, + "step": 15276 + }, + { + "epoch": 0.1326116960790271, + "grad_norm": 0.53515625, + "learning_rate": 0.0019015542667154127, + "loss": 0.0869, + "step": 15277 + }, + { + "epoch": 0.13262037655923126, + "grad_norm": 0.4296875, + "learning_rate": 0.001901540731608772, + "loss": 0.127, + "step": 15278 + }, + { + "epoch": 0.13262905703943542, + "grad_norm": 0.298828125, + "learning_rate": 0.0019015271956255785, + "loss": 0.0781, + "step": 15279 + }, + { + "epoch": 0.1326377375196396, + "grad_norm": 0.12060546875, + "learning_rate": 0.001901513658765847, + "loss": 0.1084, + "step": 15280 + }, + { + "epoch": 0.13264641799984375, + "grad_norm": 0.1318359375, + "learning_rate": 0.001901500121029592, + "loss": 0.1074, + "step": 15281 + }, + { + "epoch": 0.13265509848004792, + "grad_norm": 0.14453125, + "learning_rate": 0.0019014865824168285, + "loss": 0.1123, + "step": 15282 + }, + { + "epoch": 0.13266377896025208, + "grad_norm": 0.515625, + "learning_rate": 0.0019014730429275713, + "loss": 0.0933, + "step": 15283 + }, + { + "epoch": 0.13267245944045625, + "grad_norm": 0.1123046875, + "learning_rate": 0.0019014595025618353, + "loss": 0.1094, + "step": 15284 + }, + { + "epoch": 0.1326811399206604, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019014459613196353, + "loss": 0.0786, + "step": 15285 + }, + { + "epoch": 0.13268982040086458, + "grad_norm": 0.478515625, + "learning_rate": 0.001901432419200986, + "loss": 0.1099, + "step": 15286 + }, + { + "epoch": 0.13269850088106874, + "grad_norm": 0.5859375, + "learning_rate": 0.001901418876205902, + "loss": 0.1387, + "step": 15287 + }, + { + "epoch": 0.1327071813612729, + "grad_norm": 0.9375, + "learning_rate": 0.0019014053323343988, + "loss": 0.1191, + "step": 15288 + }, + { + "epoch": 0.13271586184147707, + "grad_norm": 0.1181640625, + "learning_rate": 0.0019013917875864903, + "loss": 0.1348, + "step": 15289 + }, + { + "epoch": 0.13272454232168124, + "grad_norm": 0.2431640625, + "learning_rate": 0.001901378241962192, + "loss": 0.1045, + "step": 15290 + }, + { + "epoch": 0.1327332228018854, + "grad_norm": 0.263671875, + "learning_rate": 0.0019013646954615185, + "loss": 0.1465, + "step": 15291 + }, + { + "epoch": 0.13274190328208957, + "grad_norm": 0.240234375, + "learning_rate": 0.0019013511480844847, + "loss": 0.1021, + "step": 15292 + }, + { + "epoch": 0.13275058376229373, + "grad_norm": 0.1005859375, + "learning_rate": 0.001901337599831105, + "loss": 0.1289, + "step": 15293 + }, + { + "epoch": 0.1327592642424979, + "grad_norm": 0.423828125, + "learning_rate": 0.001901324050701395, + "loss": 0.103, + "step": 15294 + }, + { + "epoch": 0.13276794472270206, + "grad_norm": 0.1181640625, + "learning_rate": 0.001901310500695369, + "loss": 0.1289, + "step": 15295 + }, + { + "epoch": 0.13277662520290623, + "grad_norm": 0.65625, + "learning_rate": 0.0019012969498130414, + "loss": 0.1289, + "step": 15296 + }, + { + "epoch": 0.1327853056831104, + "grad_norm": 0.078125, + "learning_rate": 0.0019012833980544279, + "loss": 0.1021, + "step": 15297 + }, + { + "epoch": 0.13279398616331456, + "grad_norm": 0.2275390625, + "learning_rate": 0.0019012698454195429, + "loss": 0.1167, + "step": 15298 + }, + { + "epoch": 0.13280266664351872, + "grad_norm": 1.3125, + "learning_rate": 0.0019012562919084014, + "loss": 0.1309, + "step": 15299 + }, + { + "epoch": 0.1328113471237229, + "grad_norm": 0.169921875, + "learning_rate": 0.0019012427375210176, + "loss": 0.126, + "step": 15300 + }, + { + "epoch": 0.13282002760392705, + "grad_norm": 0.296875, + "learning_rate": 0.0019012291822574068, + "loss": 0.1123, + "step": 15301 + }, + { + "epoch": 0.13282870808413122, + "grad_norm": 0.349609375, + "learning_rate": 0.0019012156261175841, + "loss": 0.1289, + "step": 15302 + }, + { + "epoch": 0.13283738856433538, + "grad_norm": 0.17578125, + "learning_rate": 0.001901202069101564, + "loss": 0.0879, + "step": 15303 + }, + { + "epoch": 0.13284606904453955, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019011885112093611, + "loss": 0.0801, + "step": 15304 + }, + { + "epoch": 0.13285474952474371, + "grad_norm": 0.57421875, + "learning_rate": 0.001901174952440991, + "loss": 0.1465, + "step": 15305 + }, + { + "epoch": 0.13286343000494788, + "grad_norm": 0.10693359375, + "learning_rate": 0.0019011613927964672, + "loss": 0.1221, + "step": 15306 + }, + { + "epoch": 0.13287211048515205, + "grad_norm": 0.322265625, + "learning_rate": 0.0019011478322758062, + "loss": 0.105, + "step": 15307 + }, + { + "epoch": 0.1328807909653562, + "grad_norm": 0.0986328125, + "learning_rate": 0.0019011342708790213, + "loss": 0.1143, + "step": 15308 + }, + { + "epoch": 0.13288947144556038, + "grad_norm": 0.19921875, + "learning_rate": 0.0019011207086061285, + "loss": 0.1367, + "step": 15309 + }, + { + "epoch": 0.13289815192576454, + "grad_norm": 0.138671875, + "learning_rate": 0.001901107145457142, + "loss": 0.0918, + "step": 15310 + }, + { + "epoch": 0.1329068324059687, + "grad_norm": 0.330078125, + "learning_rate": 0.0019010935814320765, + "loss": 0.1475, + "step": 15311 + }, + { + "epoch": 0.13291551288617287, + "grad_norm": 0.16796875, + "learning_rate": 0.0019010800165309473, + "loss": 0.0981, + "step": 15312 + }, + { + "epoch": 0.13292419336637704, + "grad_norm": 0.201171875, + "learning_rate": 0.001901066450753769, + "loss": 0.125, + "step": 15313 + }, + { + "epoch": 0.1329328738465812, + "grad_norm": 0.1708984375, + "learning_rate": 0.0019010528841005565, + "loss": 0.1191, + "step": 15314 + }, + { + "epoch": 0.13294155432678537, + "grad_norm": 0.455078125, + "learning_rate": 0.0019010393165713245, + "loss": 0.166, + "step": 15315 + }, + { + "epoch": 0.13295023480698953, + "grad_norm": 0.2197265625, + "learning_rate": 0.001901025748166088, + "loss": 0.0874, + "step": 15316 + }, + { + "epoch": 0.1329589152871937, + "grad_norm": 0.419921875, + "learning_rate": 0.001901012178884862, + "loss": 0.1094, + "step": 15317 + }, + { + "epoch": 0.13296759576739786, + "grad_norm": 0.439453125, + "learning_rate": 0.0019009986087276608, + "loss": 0.1045, + "step": 15318 + }, + { + "epoch": 0.13297627624760203, + "grad_norm": 0.5703125, + "learning_rate": 0.0019009850376944994, + "loss": 0.1196, + "step": 15319 + }, + { + "epoch": 0.1329849567278062, + "grad_norm": 0.55859375, + "learning_rate": 0.0019009714657853931, + "loss": 0.1406, + "step": 15320 + }, + { + "epoch": 0.13299363720801036, + "grad_norm": 0.0966796875, + "learning_rate": 0.0019009578930003566, + "loss": 0.1016, + "step": 15321 + }, + { + "epoch": 0.13300231768821452, + "grad_norm": 0.3046875, + "learning_rate": 0.0019009443193394042, + "loss": 0.1187, + "step": 15322 + }, + { + "epoch": 0.1330109981684187, + "grad_norm": 0.09130859375, + "learning_rate": 0.0019009307448025513, + "loss": 0.0723, + "step": 15323 + }, + { + "epoch": 0.13301967864862285, + "grad_norm": 0.453125, + "learning_rate": 0.0019009171693898125, + "loss": 0.1514, + "step": 15324 + }, + { + "epoch": 0.13302835912882702, + "grad_norm": 0.1689453125, + "learning_rate": 0.001900903593101203, + "loss": 0.1289, + "step": 15325 + }, + { + "epoch": 0.13303703960903118, + "grad_norm": 0.10888671875, + "learning_rate": 0.001900890015936737, + "loss": 0.0923, + "step": 15326 + }, + { + "epoch": 0.13304572008923535, + "grad_norm": 0.1650390625, + "learning_rate": 0.00190087643789643, + "loss": 0.1055, + "step": 15327 + }, + { + "epoch": 0.1330544005694395, + "grad_norm": 0.31640625, + "learning_rate": 0.0019008628589802963, + "loss": 0.1289, + "step": 15328 + }, + { + "epoch": 0.13306308104964368, + "grad_norm": 0.3046875, + "learning_rate": 0.001900849279188351, + "loss": 0.1348, + "step": 15329 + }, + { + "epoch": 0.13307176152984784, + "grad_norm": 0.09716796875, + "learning_rate": 0.0019008356985206092, + "loss": 0.1143, + "step": 15330 + }, + { + "epoch": 0.133080442010052, + "grad_norm": 0.2890625, + "learning_rate": 0.0019008221169770857, + "loss": 0.1348, + "step": 15331 + }, + { + "epoch": 0.13308912249025617, + "grad_norm": 0.326171875, + "learning_rate": 0.0019008085345577949, + "loss": 0.0923, + "step": 15332 + }, + { + "epoch": 0.13309780297046034, + "grad_norm": 0.41796875, + "learning_rate": 0.0019007949512627519, + "loss": 0.1201, + "step": 15333 + }, + { + "epoch": 0.1331064834506645, + "grad_norm": 0.212890625, + "learning_rate": 0.0019007813670919714, + "loss": 0.124, + "step": 15334 + }, + { + "epoch": 0.13311516393086867, + "grad_norm": 0.10888671875, + "learning_rate": 0.0019007677820454686, + "loss": 0.1113, + "step": 15335 + }, + { + "epoch": 0.13312384441107283, + "grad_norm": 0.62109375, + "learning_rate": 0.0019007541961232582, + "loss": 0.1631, + "step": 15336 + }, + { + "epoch": 0.133132524891277, + "grad_norm": 0.50390625, + "learning_rate": 0.0019007406093253549, + "loss": 0.0898, + "step": 15337 + }, + { + "epoch": 0.13314120537148116, + "grad_norm": 0.1455078125, + "learning_rate": 0.001900727021651774, + "loss": 0.1396, + "step": 15338 + }, + { + "epoch": 0.13314988585168533, + "grad_norm": 0.890625, + "learning_rate": 0.0019007134331025298, + "loss": 0.1719, + "step": 15339 + }, + { + "epoch": 0.13315856633188947, + "grad_norm": 0.3125, + "learning_rate": 0.0019006998436776376, + "loss": 0.125, + "step": 15340 + }, + { + "epoch": 0.13316724681209363, + "grad_norm": 0.2197265625, + "learning_rate": 0.0019006862533771118, + "loss": 0.1025, + "step": 15341 + }, + { + "epoch": 0.1331759272922978, + "grad_norm": 0.361328125, + "learning_rate": 0.0019006726622009676, + "loss": 0.1289, + "step": 15342 + }, + { + "epoch": 0.13318460777250196, + "grad_norm": 0.1962890625, + "learning_rate": 0.00190065907014922, + "loss": 0.0952, + "step": 15343 + }, + { + "epoch": 0.13319328825270613, + "grad_norm": 0.1552734375, + "learning_rate": 0.0019006454772218838, + "loss": 0.1357, + "step": 15344 + }, + { + "epoch": 0.1332019687329103, + "grad_norm": 0.91796875, + "learning_rate": 0.0019006318834189732, + "loss": 0.1133, + "step": 15345 + }, + { + "epoch": 0.13321064921311446, + "grad_norm": 0.09521484375, + "learning_rate": 0.0019006182887405043, + "loss": 0.1084, + "step": 15346 + }, + { + "epoch": 0.13321932969331862, + "grad_norm": 0.32421875, + "learning_rate": 0.0019006046931864908, + "loss": 0.1206, + "step": 15347 + }, + { + "epoch": 0.1332280101735228, + "grad_norm": 0.69921875, + "learning_rate": 0.0019005910967569485, + "loss": 0.1328, + "step": 15348 + }, + { + "epoch": 0.13323669065372695, + "grad_norm": 0.244140625, + "learning_rate": 0.0019005774994518912, + "loss": 0.1143, + "step": 15349 + }, + { + "epoch": 0.13324537113393112, + "grad_norm": 0.5625, + "learning_rate": 0.0019005639012713346, + "loss": 0.1191, + "step": 15350 + }, + { + "epoch": 0.13325405161413528, + "grad_norm": 0.07861328125, + "learning_rate": 0.0019005503022152934, + "loss": 0.0874, + "step": 15351 + }, + { + "epoch": 0.13326273209433945, + "grad_norm": 0.51953125, + "learning_rate": 0.0019005367022837823, + "loss": 0.1582, + "step": 15352 + }, + { + "epoch": 0.1332714125745436, + "grad_norm": 4.21875, + "learning_rate": 0.0019005231014768162, + "loss": 0.3047, + "step": 15353 + }, + { + "epoch": 0.13328009305474778, + "grad_norm": 0.234375, + "learning_rate": 0.0019005094997944104, + "loss": 0.1328, + "step": 15354 + }, + { + "epoch": 0.13328877353495194, + "grad_norm": 0.07666015625, + "learning_rate": 0.0019004958972365796, + "loss": 0.0894, + "step": 15355 + }, + { + "epoch": 0.1332974540151561, + "grad_norm": 0.162109375, + "learning_rate": 0.001900482293803338, + "loss": 0.1104, + "step": 15356 + }, + { + "epoch": 0.13330613449536027, + "grad_norm": 0.2490234375, + "learning_rate": 0.0019004686894947011, + "loss": 0.0918, + "step": 15357 + }, + { + "epoch": 0.13331481497556444, + "grad_norm": 0.1328125, + "learning_rate": 0.0019004550843106838, + "loss": 0.1152, + "step": 15358 + }, + { + "epoch": 0.1333234954557686, + "grad_norm": 0.416015625, + "learning_rate": 0.0019004414782513007, + "loss": 0.1147, + "step": 15359 + }, + { + "epoch": 0.13333217593597277, + "grad_norm": 0.3125, + "learning_rate": 0.0019004278713165667, + "loss": 0.0996, + "step": 15360 + }, + { + "epoch": 0.13334085641617693, + "grad_norm": 0.482421875, + "learning_rate": 0.0019004142635064974, + "loss": 0.1191, + "step": 15361 + }, + { + "epoch": 0.1333495368963811, + "grad_norm": 0.208984375, + "learning_rate": 0.0019004006548211068, + "loss": 0.085, + "step": 15362 + }, + { + "epoch": 0.13335821737658526, + "grad_norm": 0.6796875, + "learning_rate": 0.00190038704526041, + "loss": 0.0889, + "step": 15363 + }, + { + "epoch": 0.13336689785678943, + "grad_norm": 0.271484375, + "learning_rate": 0.0019003734348244218, + "loss": 0.125, + "step": 15364 + }, + { + "epoch": 0.1333755783369936, + "grad_norm": 0.55859375, + "learning_rate": 0.0019003598235131573, + "loss": 0.1318, + "step": 15365 + }, + { + "epoch": 0.13338425881719776, + "grad_norm": 0.07470703125, + "learning_rate": 0.0019003462113266316, + "loss": 0.1064, + "step": 15366 + }, + { + "epoch": 0.13339293929740192, + "grad_norm": 0.08984375, + "learning_rate": 0.001900332598264859, + "loss": 0.0903, + "step": 15367 + }, + { + "epoch": 0.1334016197776061, + "grad_norm": 0.294921875, + "learning_rate": 0.001900318984327855, + "loss": 0.1309, + "step": 15368 + }, + { + "epoch": 0.13341030025781025, + "grad_norm": 0.21875, + "learning_rate": 0.0019003053695156335, + "loss": 0.126, + "step": 15369 + }, + { + "epoch": 0.13341898073801442, + "grad_norm": 0.51171875, + "learning_rate": 0.001900291753828211, + "loss": 0.125, + "step": 15370 + }, + { + "epoch": 0.13342766121821859, + "grad_norm": 0.2353515625, + "learning_rate": 0.0019002781372656008, + "loss": 0.1289, + "step": 15371 + }, + { + "epoch": 0.13343634169842275, + "grad_norm": 0.416015625, + "learning_rate": 0.001900264519827819, + "loss": 0.1147, + "step": 15372 + }, + { + "epoch": 0.13344502217862692, + "grad_norm": 0.6796875, + "learning_rate": 0.0019002509015148797, + "loss": 0.1523, + "step": 15373 + }, + { + "epoch": 0.13345370265883108, + "grad_norm": 0.3125, + "learning_rate": 0.0019002372823267977, + "loss": 0.0986, + "step": 15374 + }, + { + "epoch": 0.13346238313903525, + "grad_norm": 0.28125, + "learning_rate": 0.0019002236622635887, + "loss": 0.1162, + "step": 15375 + }, + { + "epoch": 0.1334710636192394, + "grad_norm": 0.341796875, + "learning_rate": 0.0019002100413252666, + "loss": 0.0898, + "step": 15376 + }, + { + "epoch": 0.13347974409944358, + "grad_norm": 0.50390625, + "learning_rate": 0.0019001964195118474, + "loss": 0.125, + "step": 15377 + }, + { + "epoch": 0.13348842457964774, + "grad_norm": 0.078125, + "learning_rate": 0.001900182796823345, + "loss": 0.1084, + "step": 15378 + }, + { + "epoch": 0.1334971050598519, + "grad_norm": 0.8125, + "learning_rate": 0.0019001691732597747, + "loss": 0.1797, + "step": 15379 + }, + { + "epoch": 0.13350578554005607, + "grad_norm": 0.18359375, + "learning_rate": 0.0019001555488211519, + "loss": 0.106, + "step": 15380 + }, + { + "epoch": 0.13351446602026024, + "grad_norm": 0.3671875, + "learning_rate": 0.0019001419235074907, + "loss": 0.1035, + "step": 15381 + }, + { + "epoch": 0.1335231465004644, + "grad_norm": 0.1474609375, + "learning_rate": 0.0019001282973188061, + "loss": 0.1055, + "step": 15382 + }, + { + "epoch": 0.13353182698066857, + "grad_norm": 0.65234375, + "learning_rate": 0.0019001146702551136, + "loss": 0.0977, + "step": 15383 + }, + { + "epoch": 0.13354050746087273, + "grad_norm": 0.33203125, + "learning_rate": 0.0019001010423164272, + "loss": 0.083, + "step": 15384 + }, + { + "epoch": 0.1335491879410769, + "grad_norm": 0.375, + "learning_rate": 0.001900087413502763, + "loss": 0.1172, + "step": 15385 + }, + { + "epoch": 0.13355786842128106, + "grad_norm": 0.265625, + "learning_rate": 0.0019000737838141349, + "loss": 0.1436, + "step": 15386 + }, + { + "epoch": 0.13356654890148523, + "grad_norm": 0.140625, + "learning_rate": 0.0019000601532505582, + "loss": 0.1084, + "step": 15387 + }, + { + "epoch": 0.1335752293816894, + "grad_norm": 0.60546875, + "learning_rate": 0.0019000465218120474, + "loss": 0.1182, + "step": 15388 + }, + { + "epoch": 0.13358390986189356, + "grad_norm": 0.0673828125, + "learning_rate": 0.001900032889498618, + "loss": 0.1011, + "step": 15389 + }, + { + "epoch": 0.13359259034209772, + "grad_norm": 0.2353515625, + "learning_rate": 0.001900019256310285, + "loss": 0.1221, + "step": 15390 + }, + { + "epoch": 0.1336012708223019, + "grad_norm": 0.490234375, + "learning_rate": 0.0019000056222470624, + "loss": 0.1089, + "step": 15391 + }, + { + "epoch": 0.13360995130250605, + "grad_norm": 0.07763671875, + "learning_rate": 0.001899991987308966, + "loss": 0.125, + "step": 15392 + }, + { + "epoch": 0.13361863178271022, + "grad_norm": 0.169921875, + "learning_rate": 0.0018999783514960103, + "loss": 0.0967, + "step": 15393 + }, + { + "epoch": 0.13362731226291438, + "grad_norm": 0.134765625, + "learning_rate": 0.0018999647148082105, + "loss": 0.1133, + "step": 15394 + }, + { + "epoch": 0.13363599274311855, + "grad_norm": 0.07958984375, + "learning_rate": 0.0018999510772455808, + "loss": 0.0942, + "step": 15395 + }, + { + "epoch": 0.1336446732233227, + "grad_norm": 1.09375, + "learning_rate": 0.001899937438808137, + "loss": 0.1162, + "step": 15396 + }, + { + "epoch": 0.13365335370352688, + "grad_norm": 0.298828125, + "learning_rate": 0.0018999237994958935, + "loss": 0.1152, + "step": 15397 + }, + { + "epoch": 0.13366203418373104, + "grad_norm": 0.294921875, + "learning_rate": 0.0018999101593088655, + "loss": 0.1328, + "step": 15398 + }, + { + "epoch": 0.1336707146639352, + "grad_norm": 0.4609375, + "learning_rate": 0.0018998965182470675, + "loss": 0.1152, + "step": 15399 + }, + { + "epoch": 0.13367939514413937, + "grad_norm": 0.06591796875, + "learning_rate": 0.001899882876310515, + "loss": 0.1006, + "step": 15400 + }, + { + "epoch": 0.13368807562434354, + "grad_norm": 0.23828125, + "learning_rate": 0.0018998692334992223, + "loss": 0.1523, + "step": 15401 + }, + { + "epoch": 0.1336967561045477, + "grad_norm": 0.10302734375, + "learning_rate": 0.0018998555898132048, + "loss": 0.1045, + "step": 15402 + }, + { + "epoch": 0.13370543658475187, + "grad_norm": 0.09130859375, + "learning_rate": 0.0018998419452524774, + "loss": 0.1143, + "step": 15403 + }, + { + "epoch": 0.13371411706495603, + "grad_norm": 0.484375, + "learning_rate": 0.0018998282998170543, + "loss": 0.127, + "step": 15404 + }, + { + "epoch": 0.1337227975451602, + "grad_norm": 0.275390625, + "learning_rate": 0.0018998146535069518, + "loss": 0.0898, + "step": 15405 + }, + { + "epoch": 0.13373147802536436, + "grad_norm": 0.4296875, + "learning_rate": 0.0018998010063221835, + "loss": 0.1055, + "step": 15406 + }, + { + "epoch": 0.13374015850556853, + "grad_norm": 0.466796875, + "learning_rate": 0.001899787358262765, + "loss": 0.0923, + "step": 15407 + }, + { + "epoch": 0.1337488389857727, + "grad_norm": 0.427734375, + "learning_rate": 0.0018997737093287108, + "loss": 0.126, + "step": 15408 + }, + { + "epoch": 0.13375751946597686, + "grad_norm": 0.07763671875, + "learning_rate": 0.0018997600595200364, + "loss": 0.1084, + "step": 15409 + }, + { + "epoch": 0.13376619994618102, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018997464088367562, + "loss": 0.1016, + "step": 15410 + }, + { + "epoch": 0.1337748804263852, + "grad_norm": 0.287109375, + "learning_rate": 0.0018997327572788853, + "loss": 0.0996, + "step": 15411 + }, + { + "epoch": 0.13378356090658935, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018997191048464388, + "loss": 0.1079, + "step": 15412 + }, + { + "epoch": 0.13379224138679352, + "grad_norm": 0.162109375, + "learning_rate": 0.0018997054515394317, + "loss": 0.1318, + "step": 15413 + }, + { + "epoch": 0.13380092186699769, + "grad_norm": 0.8359375, + "learning_rate": 0.0018996917973578785, + "loss": 0.1025, + "step": 15414 + }, + { + "epoch": 0.13380960234720185, + "grad_norm": 0.37109375, + "learning_rate": 0.0018996781423017944, + "loss": 0.1245, + "step": 15415 + }, + { + "epoch": 0.13381828282740602, + "grad_norm": 0.06298828125, + "learning_rate": 0.0018996644863711943, + "loss": 0.0903, + "step": 15416 + }, + { + "epoch": 0.13382696330761018, + "grad_norm": 0.38671875, + "learning_rate": 0.001899650829566093, + "loss": 0.1074, + "step": 15417 + }, + { + "epoch": 0.13383564378781435, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018996371718865057, + "loss": 0.1226, + "step": 15418 + }, + { + "epoch": 0.1338443242680185, + "grad_norm": 0.32421875, + "learning_rate": 0.0018996235133324473, + "loss": 0.1328, + "step": 15419 + }, + { + "epoch": 0.13385300474822268, + "grad_norm": 0.359375, + "learning_rate": 0.0018996098539039326, + "loss": 0.1396, + "step": 15420 + }, + { + "epoch": 0.13386168522842684, + "grad_norm": 0.08251953125, + "learning_rate": 0.0018995961936009768, + "loss": 0.0908, + "step": 15421 + }, + { + "epoch": 0.133870365708631, + "grad_norm": 0.1630859375, + "learning_rate": 0.001899582532423594, + "loss": 0.1069, + "step": 15422 + }, + { + "epoch": 0.13387904618883517, + "grad_norm": 0.365234375, + "learning_rate": 0.0018995688703718002, + "loss": 0.1172, + "step": 15423 + }, + { + "epoch": 0.13388772666903934, + "grad_norm": 0.546875, + "learning_rate": 0.0018995552074456098, + "loss": 0.1348, + "step": 15424 + }, + { + "epoch": 0.1338964071492435, + "grad_norm": 0.064453125, + "learning_rate": 0.001899541543645038, + "loss": 0.0928, + "step": 15425 + }, + { + "epoch": 0.13390508762944767, + "grad_norm": 0.06201171875, + "learning_rate": 0.0018995278789700992, + "loss": 0.0957, + "step": 15426 + }, + { + "epoch": 0.13391376810965183, + "grad_norm": 0.44140625, + "learning_rate": 0.001899514213420809, + "loss": 0.1172, + "step": 15427 + }, + { + "epoch": 0.133922448589856, + "grad_norm": 0.427734375, + "learning_rate": 0.001899500546997182, + "loss": 0.1152, + "step": 15428 + }, + { + "epoch": 0.13393112907006016, + "grad_norm": 0.11181640625, + "learning_rate": 0.0018994868796992333, + "loss": 0.1289, + "step": 15429 + }, + { + "epoch": 0.13393980955026433, + "grad_norm": 0.10107421875, + "learning_rate": 0.0018994732115269776, + "loss": 0.0898, + "step": 15430 + }, + { + "epoch": 0.1339484900304685, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018994595424804302, + "loss": 0.1069, + "step": 15431 + }, + { + "epoch": 0.13395717051067266, + "grad_norm": 0.5625, + "learning_rate": 0.0018994458725596058, + "loss": 0.1123, + "step": 15432 + }, + { + "epoch": 0.13396585099087682, + "grad_norm": 0.08984375, + "learning_rate": 0.0018994322017645193, + "loss": 0.124, + "step": 15433 + }, + { + "epoch": 0.133974531471081, + "grad_norm": 0.255859375, + "learning_rate": 0.0018994185300951856, + "loss": 0.127, + "step": 15434 + }, + { + "epoch": 0.13398321195128515, + "grad_norm": 0.59375, + "learning_rate": 0.0018994048575516204, + "loss": 0.1309, + "step": 15435 + }, + { + "epoch": 0.13399189243148932, + "grad_norm": 0.22265625, + "learning_rate": 0.0018993911841338374, + "loss": 0.1133, + "step": 15436 + }, + { + "epoch": 0.13400057291169348, + "grad_norm": 0.375, + "learning_rate": 0.0018993775098418526, + "loss": 0.1084, + "step": 15437 + }, + { + "epoch": 0.13400925339189765, + "grad_norm": 0.373046875, + "learning_rate": 0.0018993638346756804, + "loss": 0.1064, + "step": 15438 + }, + { + "epoch": 0.1340179338721018, + "grad_norm": 0.181640625, + "learning_rate": 0.001899350158635336, + "loss": 0.0942, + "step": 15439 + }, + { + "epoch": 0.13402661435230598, + "grad_norm": 0.412109375, + "learning_rate": 0.0018993364817208342, + "loss": 0.2285, + "step": 15440 + }, + { + "epoch": 0.13403529483251014, + "grad_norm": 0.0732421875, + "learning_rate": 0.00189932280393219, + "loss": 0.0811, + "step": 15441 + }, + { + "epoch": 0.1340439753127143, + "grad_norm": 0.6171875, + "learning_rate": 0.0018993091252694188, + "loss": 0.1406, + "step": 15442 + }, + { + "epoch": 0.13405265579291847, + "grad_norm": 0.77734375, + "learning_rate": 0.0018992954457325347, + "loss": 0.0977, + "step": 15443 + }, + { + "epoch": 0.13406133627312264, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018992817653215532, + "loss": 0.1021, + "step": 15444 + }, + { + "epoch": 0.1340700167533268, + "grad_norm": 0.5390625, + "learning_rate": 0.001899268084036489, + "loss": 0.0854, + "step": 15445 + }, + { + "epoch": 0.13407869723353097, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018992544018773577, + "loss": 0.0928, + "step": 15446 + }, + { + "epoch": 0.13408737771373513, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018992407188441734, + "loss": 0.085, + "step": 15447 + }, + { + "epoch": 0.1340960581939393, + "grad_norm": 0.21484375, + "learning_rate": 0.0018992270349369516, + "loss": 0.0972, + "step": 15448 + }, + { + "epoch": 0.13410473867414346, + "grad_norm": 0.134765625, + "learning_rate": 0.0018992133501557072, + "loss": 0.0864, + "step": 15449 + }, + { + "epoch": 0.13411341915434763, + "grad_norm": 0.390625, + "learning_rate": 0.001899199664500455, + "loss": 0.0913, + "step": 15450 + }, + { + "epoch": 0.1341220996345518, + "grad_norm": 0.296875, + "learning_rate": 0.00189918597797121, + "loss": 0.1289, + "step": 15451 + }, + { + "epoch": 0.13413078011475596, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018991722905679874, + "loss": 0.1221, + "step": 15452 + }, + { + "epoch": 0.13413946059496012, + "grad_norm": 0.1328125, + "learning_rate": 0.0018991586022908016, + "loss": 0.127, + "step": 15453 + }, + { + "epoch": 0.1341481410751643, + "grad_norm": 0.1328125, + "learning_rate": 0.0018991449131396682, + "loss": 0.0767, + "step": 15454 + }, + { + "epoch": 0.13415682155536846, + "grad_norm": 0.259765625, + "learning_rate": 0.0018991312231146024, + "loss": 0.1128, + "step": 15455 + }, + { + "epoch": 0.13416550203557262, + "grad_norm": 0.302734375, + "learning_rate": 0.001899117532215618, + "loss": 0.1094, + "step": 15456 + }, + { + "epoch": 0.13417418251577679, + "grad_norm": 0.1357421875, + "learning_rate": 0.001899103840442731, + "loss": 0.1143, + "step": 15457 + }, + { + "epoch": 0.13418286299598095, + "grad_norm": 0.3203125, + "learning_rate": 0.001899090147795956, + "loss": 0.1074, + "step": 15458 + }, + { + "epoch": 0.13419154347618512, + "grad_norm": 0.388671875, + "learning_rate": 0.001899076454275308, + "loss": 0.1211, + "step": 15459 + }, + { + "epoch": 0.13420022395638928, + "grad_norm": 0.14453125, + "learning_rate": 0.001899062759880802, + "loss": 0.0913, + "step": 15460 + }, + { + "epoch": 0.13420890443659345, + "grad_norm": 0.4765625, + "learning_rate": 0.001899049064612453, + "loss": 0.1094, + "step": 15461 + }, + { + "epoch": 0.1342175849167976, + "grad_norm": 0.173828125, + "learning_rate": 0.0018990353684702759, + "loss": 0.0967, + "step": 15462 + }, + { + "epoch": 0.13422626539700175, + "grad_norm": 0.7109375, + "learning_rate": 0.001899021671454286, + "loss": 0.0957, + "step": 15463 + }, + { + "epoch": 0.1342349458772059, + "grad_norm": 1.2421875, + "learning_rate": 0.0018990079735644973, + "loss": 0.1377, + "step": 15464 + }, + { + "epoch": 0.13424362635741008, + "grad_norm": 0.359375, + "learning_rate": 0.0018989942748009261, + "loss": 0.0981, + "step": 15465 + }, + { + "epoch": 0.13425230683761424, + "grad_norm": 0.0771484375, + "learning_rate": 0.0018989805751635864, + "loss": 0.085, + "step": 15466 + }, + { + "epoch": 0.1342609873178184, + "grad_norm": 0.48046875, + "learning_rate": 0.0018989668746524939, + "loss": 0.1152, + "step": 15467 + }, + { + "epoch": 0.13426966779802257, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018989531732676631, + "loss": 0.1143, + "step": 15468 + }, + { + "epoch": 0.13427834827822674, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018989394710091092, + "loss": 0.126, + "step": 15469 + }, + { + "epoch": 0.1342870287584309, + "grad_norm": 0.33984375, + "learning_rate": 0.0018989257678768468, + "loss": 0.0996, + "step": 15470 + }, + { + "epoch": 0.13429570923863507, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018989120638708914, + "loss": 0.1138, + "step": 15471 + }, + { + "epoch": 0.13430438971883923, + "grad_norm": 0.30859375, + "learning_rate": 0.001898898358991258, + "loss": 0.124, + "step": 15472 + }, + { + "epoch": 0.1343130701990434, + "grad_norm": 0.435546875, + "learning_rate": 0.0018988846532379609, + "loss": 0.1299, + "step": 15473 + }, + { + "epoch": 0.13432175067924756, + "grad_norm": 0.36328125, + "learning_rate": 0.0018988709466110157, + "loss": 0.1699, + "step": 15474 + }, + { + "epoch": 0.13433043115945173, + "grad_norm": 0.228515625, + "learning_rate": 0.0018988572391104373, + "loss": 0.1221, + "step": 15475 + }, + { + "epoch": 0.1343391116396559, + "grad_norm": 0.150390625, + "learning_rate": 0.0018988435307362406, + "loss": 0.168, + "step": 15476 + }, + { + "epoch": 0.13434779211986006, + "grad_norm": 0.7578125, + "learning_rate": 0.0018988298214884407, + "loss": 0.127, + "step": 15477 + }, + { + "epoch": 0.13435647260006423, + "grad_norm": 0.0703125, + "learning_rate": 0.0018988161113670523, + "loss": 0.0938, + "step": 15478 + }, + { + "epoch": 0.1343651530802684, + "grad_norm": 0.90234375, + "learning_rate": 0.0018988024003720907, + "loss": 0.0825, + "step": 15479 + }, + { + "epoch": 0.13437383356047256, + "grad_norm": 0.6640625, + "learning_rate": 0.001898788688503571, + "loss": 0.1445, + "step": 15480 + }, + { + "epoch": 0.13438251404067672, + "grad_norm": 0.197265625, + "learning_rate": 0.001898774975761508, + "loss": 0.1436, + "step": 15481 + }, + { + "epoch": 0.13439119452088089, + "grad_norm": 0.52734375, + "learning_rate": 0.0018987612621459164, + "loss": 0.1113, + "step": 15482 + }, + { + "epoch": 0.13439987500108505, + "grad_norm": 0.328125, + "learning_rate": 0.0018987475476568117, + "loss": 0.1133, + "step": 15483 + }, + { + "epoch": 0.13440855548128922, + "grad_norm": 0.267578125, + "learning_rate": 0.0018987338322942085, + "loss": 0.0913, + "step": 15484 + }, + { + "epoch": 0.13441723596149338, + "grad_norm": 0.2392578125, + "learning_rate": 0.001898720116058122, + "loss": 0.1572, + "step": 15485 + }, + { + "epoch": 0.13442591644169755, + "grad_norm": 0.490234375, + "learning_rate": 0.0018987063989485673, + "loss": 0.1177, + "step": 15486 + }, + { + "epoch": 0.1344345969219017, + "grad_norm": 0.390625, + "learning_rate": 0.0018986926809655593, + "loss": 0.125, + "step": 15487 + }, + { + "epoch": 0.13444327740210588, + "grad_norm": 0.55859375, + "learning_rate": 0.0018986789621091129, + "loss": 0.1113, + "step": 15488 + }, + { + "epoch": 0.13445195788231004, + "grad_norm": 0.12060546875, + "learning_rate": 0.0018986652423792433, + "loss": 0.1147, + "step": 15489 + }, + { + "epoch": 0.1344606383625142, + "grad_norm": 0.484375, + "learning_rate": 0.0018986515217759652, + "loss": 0.1104, + "step": 15490 + }, + { + "epoch": 0.13446931884271837, + "grad_norm": 0.09423828125, + "learning_rate": 0.001898637800299294, + "loss": 0.103, + "step": 15491 + }, + { + "epoch": 0.13447799932292254, + "grad_norm": 0.11328125, + "learning_rate": 0.0018986240779492443, + "loss": 0.1143, + "step": 15492 + }, + { + "epoch": 0.1344866798031267, + "grad_norm": 0.93359375, + "learning_rate": 0.0018986103547258312, + "loss": 0.0942, + "step": 15493 + }, + { + "epoch": 0.13449536028333087, + "grad_norm": 0.30078125, + "learning_rate": 0.00189859663062907, + "loss": 0.0732, + "step": 15494 + }, + { + "epoch": 0.13450404076353503, + "grad_norm": 0.134765625, + "learning_rate": 0.0018985829056589756, + "loss": 0.125, + "step": 15495 + }, + { + "epoch": 0.1345127212437392, + "grad_norm": 0.0673828125, + "learning_rate": 0.0018985691798155632, + "loss": 0.1167, + "step": 15496 + }, + { + "epoch": 0.13452140172394336, + "grad_norm": 0.154296875, + "learning_rate": 0.001898555453098847, + "loss": 0.1133, + "step": 15497 + }, + { + "epoch": 0.13453008220414753, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018985417255088426, + "loss": 0.0962, + "step": 15498 + }, + { + "epoch": 0.1345387626843517, + "grad_norm": 0.279296875, + "learning_rate": 0.0018985279970455654, + "loss": 0.1133, + "step": 15499 + }, + { + "epoch": 0.13454744316455586, + "grad_norm": 0.244140625, + "learning_rate": 0.0018985142677090297, + "loss": 0.0996, + "step": 15500 + }, + { + "epoch": 0.13455612364476002, + "grad_norm": 0.158203125, + "learning_rate": 0.0018985005374992506, + "loss": 0.1055, + "step": 15501 + }, + { + "epoch": 0.1345648041249642, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018984868064162437, + "loss": 0.0957, + "step": 15502 + }, + { + "epoch": 0.13457348460516835, + "grad_norm": 0.369140625, + "learning_rate": 0.0018984730744600233, + "loss": 0.168, + "step": 15503 + }, + { + "epoch": 0.13458216508537252, + "grad_norm": 0.5390625, + "learning_rate": 0.0018984593416306048, + "loss": 0.1172, + "step": 15504 + }, + { + "epoch": 0.13459084556557668, + "grad_norm": 0.23046875, + "learning_rate": 0.0018984456079280032, + "loss": 0.1602, + "step": 15505 + }, + { + "epoch": 0.13459952604578085, + "grad_norm": 0.318359375, + "learning_rate": 0.001898431873352234, + "loss": 0.1221, + "step": 15506 + }, + { + "epoch": 0.134608206525985, + "grad_norm": 0.265625, + "learning_rate": 0.001898418137903311, + "loss": 0.0898, + "step": 15507 + }, + { + "epoch": 0.13461688700618918, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018984044015812503, + "loss": 0.1562, + "step": 15508 + }, + { + "epoch": 0.13462556748639334, + "grad_norm": 0.4375, + "learning_rate": 0.0018983906643860661, + "loss": 0.1123, + "step": 15509 + }, + { + "epoch": 0.1346342479665975, + "grad_norm": 0.291015625, + "learning_rate": 0.0018983769263177742, + "loss": 0.1484, + "step": 15510 + }, + { + "epoch": 0.13464292844680167, + "grad_norm": 0.462890625, + "learning_rate": 0.0018983631873763893, + "loss": 0.0952, + "step": 15511 + }, + { + "epoch": 0.13465160892700584, + "grad_norm": 0.318359375, + "learning_rate": 0.0018983494475619262, + "loss": 0.1143, + "step": 15512 + }, + { + "epoch": 0.13466028940721, + "grad_norm": 0.6328125, + "learning_rate": 0.0018983357068744005, + "loss": 0.1016, + "step": 15513 + }, + { + "epoch": 0.13466896988741417, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018983219653138266, + "loss": 0.1177, + "step": 15514 + }, + { + "epoch": 0.13467765036761833, + "grad_norm": 0.314453125, + "learning_rate": 0.00189830822288022, + "loss": 0.1187, + "step": 15515 + }, + { + "epoch": 0.1346863308478225, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018982944795735953, + "loss": 0.0977, + "step": 15516 + }, + { + "epoch": 0.13469501132802666, + "grad_norm": 0.109375, + "learning_rate": 0.0018982807353939677, + "loss": 0.1118, + "step": 15517 + }, + { + "epoch": 0.13470369180823083, + "grad_norm": 0.103515625, + "learning_rate": 0.0018982669903413526, + "loss": 0.1016, + "step": 15518 + }, + { + "epoch": 0.134712372288435, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018982532444157645, + "loss": 0.0938, + "step": 15519 + }, + { + "epoch": 0.13472105276863916, + "grad_norm": 0.2265625, + "learning_rate": 0.0018982394976172186, + "loss": 0.0864, + "step": 15520 + }, + { + "epoch": 0.13472973324884333, + "grad_norm": 0.39453125, + "learning_rate": 0.00189822574994573, + "loss": 0.1328, + "step": 15521 + }, + { + "epoch": 0.1347384137290475, + "grad_norm": 0.1796875, + "learning_rate": 0.0018982120014013138, + "loss": 0.1377, + "step": 15522 + }, + { + "epoch": 0.13474709420925166, + "grad_norm": 0.74609375, + "learning_rate": 0.0018981982519839851, + "loss": 0.125, + "step": 15523 + }, + { + "epoch": 0.13475577468945582, + "grad_norm": 0.625, + "learning_rate": 0.0018981845016937584, + "loss": 0.1797, + "step": 15524 + }, + { + "epoch": 0.13476445516965999, + "grad_norm": 0.26171875, + "learning_rate": 0.0018981707505306492, + "loss": 0.0957, + "step": 15525 + }, + { + "epoch": 0.13477313564986415, + "grad_norm": 0.4375, + "learning_rate": 0.001898156998494673, + "loss": 0.1133, + "step": 15526 + }, + { + "epoch": 0.13478181613006832, + "grad_norm": 1.2265625, + "learning_rate": 0.0018981432455858435, + "loss": 0.1836, + "step": 15527 + }, + { + "epoch": 0.13479049661027248, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018981294918041772, + "loss": 0.1045, + "step": 15528 + }, + { + "epoch": 0.13479917709047665, + "grad_norm": 0.412109375, + "learning_rate": 0.0018981157371496882, + "loss": 0.082, + "step": 15529 + }, + { + "epoch": 0.1348078575706808, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018981019816223916, + "loss": 0.0679, + "step": 15530 + }, + { + "epoch": 0.13481653805088498, + "grad_norm": 0.107421875, + "learning_rate": 0.001898088225222303, + "loss": 0.1206, + "step": 15531 + }, + { + "epoch": 0.13482521853108914, + "grad_norm": 0.4765625, + "learning_rate": 0.001898074467949437, + "loss": 0.1621, + "step": 15532 + }, + { + "epoch": 0.1348338990112933, + "grad_norm": 0.205078125, + "learning_rate": 0.0018980607098038084, + "loss": 0.1021, + "step": 15533 + }, + { + "epoch": 0.13484257949149747, + "grad_norm": 0.66796875, + "learning_rate": 0.0018980469507854333, + "loss": 0.127, + "step": 15534 + }, + { + "epoch": 0.13485125997170164, + "grad_norm": 0.3984375, + "learning_rate": 0.0018980331908943254, + "loss": 0.0986, + "step": 15535 + }, + { + "epoch": 0.1348599404519058, + "grad_norm": 0.26953125, + "learning_rate": 0.001898019430130501, + "loss": 0.1309, + "step": 15536 + }, + { + "epoch": 0.13486862093210997, + "grad_norm": 0.169921875, + "learning_rate": 0.001898005668493974, + "loss": 0.1123, + "step": 15537 + }, + { + "epoch": 0.13487730141231413, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018979919059847604, + "loss": 0.1104, + "step": 15538 + }, + { + "epoch": 0.1348859818925183, + "grad_norm": 0.296875, + "learning_rate": 0.0018979781426028746, + "loss": 0.166, + "step": 15539 + }, + { + "epoch": 0.13489466237272246, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018979643783483319, + "loss": 0.1504, + "step": 15540 + }, + { + "epoch": 0.13490334285292663, + "grad_norm": 0.267578125, + "learning_rate": 0.0018979506132211475, + "loss": 0.1133, + "step": 15541 + }, + { + "epoch": 0.1349120233331308, + "grad_norm": 0.447265625, + "learning_rate": 0.0018979368472213359, + "loss": 0.1143, + "step": 15542 + }, + { + "epoch": 0.13492070381333496, + "grad_norm": 0.279296875, + "learning_rate": 0.0018979230803489132, + "loss": 0.1147, + "step": 15543 + }, + { + "epoch": 0.13492938429353912, + "grad_norm": 0.212890625, + "learning_rate": 0.0018979093126038932, + "loss": 0.1523, + "step": 15544 + }, + { + "epoch": 0.1349380647737433, + "grad_norm": 0.0849609375, + "learning_rate": 0.0018978955439862917, + "loss": 0.1021, + "step": 15545 + }, + { + "epoch": 0.13494674525394745, + "grad_norm": 0.6640625, + "learning_rate": 0.0018978817744961238, + "loss": 0.1455, + "step": 15546 + }, + { + "epoch": 0.13495542573415162, + "grad_norm": 0.64453125, + "learning_rate": 0.0018978680041334044, + "loss": 0.1533, + "step": 15547 + }, + { + "epoch": 0.13496410621435578, + "grad_norm": 0.228515625, + "learning_rate": 0.0018978542328981485, + "loss": 0.1162, + "step": 15548 + }, + { + "epoch": 0.13497278669455995, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018978404607903712, + "loss": 0.0894, + "step": 15549 + }, + { + "epoch": 0.1349814671747641, + "grad_norm": 0.25, + "learning_rate": 0.0018978266878100873, + "loss": 0.1074, + "step": 15550 + }, + { + "epoch": 0.13499014765496828, + "grad_norm": 0.76953125, + "learning_rate": 0.0018978129139573124, + "loss": 0.1318, + "step": 15551 + }, + { + "epoch": 0.13499882813517244, + "grad_norm": 0.7890625, + "learning_rate": 0.0018977991392320616, + "loss": 0.127, + "step": 15552 + }, + { + "epoch": 0.1350075086153766, + "grad_norm": 0.203125, + "learning_rate": 0.001897785363634349, + "loss": 0.1201, + "step": 15553 + }, + { + "epoch": 0.13501618909558077, + "grad_norm": 0.208984375, + "learning_rate": 0.0018977715871641907, + "loss": 0.1196, + "step": 15554 + }, + { + "epoch": 0.13502486957578494, + "grad_norm": 0.294921875, + "learning_rate": 0.001897757809821601, + "loss": 0.127, + "step": 15555 + }, + { + "epoch": 0.1350335500559891, + "grad_norm": 0.1953125, + "learning_rate": 0.0018977440316065956, + "loss": 0.1318, + "step": 15556 + }, + { + "epoch": 0.13504223053619327, + "grad_norm": 0.109375, + "learning_rate": 0.0018977302525191896, + "loss": 0.1406, + "step": 15557 + }, + { + "epoch": 0.13505091101639743, + "grad_norm": 0.234375, + "learning_rate": 0.0018977164725593975, + "loss": 0.1182, + "step": 15558 + }, + { + "epoch": 0.1350595914966016, + "grad_norm": 0.345703125, + "learning_rate": 0.0018977026917272343, + "loss": 0.0903, + "step": 15559 + }, + { + "epoch": 0.13506827197680576, + "grad_norm": 0.11181640625, + "learning_rate": 0.0018976889100227158, + "loss": 0.1289, + "step": 15560 + }, + { + "epoch": 0.13507695245700993, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018976751274458567, + "loss": 0.1035, + "step": 15561 + }, + { + "epoch": 0.1350856329372141, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018976613439966722, + "loss": 0.1641, + "step": 15562 + }, + { + "epoch": 0.13509431341741826, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018976475596751772, + "loss": 0.0889, + "step": 15563 + }, + { + "epoch": 0.13510299389762243, + "grad_norm": 0.3125, + "learning_rate": 0.0018976337744813865, + "loss": 0.1318, + "step": 15564 + }, + { + "epoch": 0.1351116743778266, + "grad_norm": 0.7734375, + "learning_rate": 0.0018976199884153157, + "loss": 0.1143, + "step": 15565 + }, + { + "epoch": 0.13512035485803076, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018976062014769794, + "loss": 0.0908, + "step": 15566 + }, + { + "epoch": 0.13512903533823492, + "grad_norm": 0.46484375, + "learning_rate": 0.0018975924136663933, + "loss": 0.1592, + "step": 15567 + }, + { + "epoch": 0.13513771581843909, + "grad_norm": 0.322265625, + "learning_rate": 0.0018975786249835723, + "loss": 0.0928, + "step": 15568 + }, + { + "epoch": 0.13514639629864325, + "grad_norm": 0.083984375, + "learning_rate": 0.0018975648354285308, + "loss": 0.1348, + "step": 15569 + }, + { + "epoch": 0.13515507677884742, + "grad_norm": 0.431640625, + "learning_rate": 0.0018975510450012847, + "loss": 0.0977, + "step": 15570 + }, + { + "epoch": 0.13516375725905158, + "grad_norm": 0.2578125, + "learning_rate": 0.0018975372537018485, + "loss": 0.1445, + "step": 15571 + }, + { + "epoch": 0.13517243773925575, + "grad_norm": 0.400390625, + "learning_rate": 0.0018975234615302378, + "loss": 0.1318, + "step": 15572 + }, + { + "epoch": 0.1351811182194599, + "grad_norm": 0.16015625, + "learning_rate": 0.0018975096684864673, + "loss": 0.1367, + "step": 15573 + }, + { + "epoch": 0.13518979869966408, + "grad_norm": 0.1015625, + "learning_rate": 0.0018974958745705522, + "loss": 0.1123, + "step": 15574 + }, + { + "epoch": 0.13519847917986824, + "grad_norm": 0.361328125, + "learning_rate": 0.0018974820797825075, + "loss": 0.0981, + "step": 15575 + }, + { + "epoch": 0.1352071596600724, + "grad_norm": 0.126953125, + "learning_rate": 0.0018974682841223485, + "loss": 0.1396, + "step": 15576 + }, + { + "epoch": 0.13521584014027657, + "grad_norm": 0.88671875, + "learning_rate": 0.00189745448759009, + "loss": 0.1299, + "step": 15577 + }, + { + "epoch": 0.13522452062048074, + "grad_norm": 0.244140625, + "learning_rate": 0.0018974406901857476, + "loss": 0.1069, + "step": 15578 + }, + { + "epoch": 0.1352332011006849, + "grad_norm": 0.1630859375, + "learning_rate": 0.001897426891909336, + "loss": 0.1387, + "step": 15579 + }, + { + "epoch": 0.13524188158088907, + "grad_norm": 0.19140625, + "learning_rate": 0.0018974130927608699, + "loss": 0.0918, + "step": 15580 + }, + { + "epoch": 0.13525056206109323, + "grad_norm": 0.2158203125, + "learning_rate": 0.0018973992927403653, + "loss": 0.0967, + "step": 15581 + }, + { + "epoch": 0.1352592425412974, + "grad_norm": 0.1875, + "learning_rate": 0.0018973854918478367, + "loss": 0.1338, + "step": 15582 + }, + { + "epoch": 0.13526792302150156, + "grad_norm": 0.43359375, + "learning_rate": 0.0018973716900832992, + "loss": 0.1074, + "step": 15583 + }, + { + "epoch": 0.13527660350170573, + "grad_norm": 0.80078125, + "learning_rate": 0.0018973578874467679, + "loss": 0.1152, + "step": 15584 + }, + { + "epoch": 0.1352852839819099, + "grad_norm": 0.140625, + "learning_rate": 0.001897344083938258, + "loss": 0.0791, + "step": 15585 + }, + { + "epoch": 0.13529396446211406, + "grad_norm": 0.97265625, + "learning_rate": 0.0018973302795577847, + "loss": 0.1104, + "step": 15586 + }, + { + "epoch": 0.1353026449423182, + "grad_norm": 0.5390625, + "learning_rate": 0.001897316474305363, + "loss": 0.1475, + "step": 15587 + }, + { + "epoch": 0.13531132542252236, + "grad_norm": 0.330078125, + "learning_rate": 0.001897302668181008, + "loss": 0.0996, + "step": 15588 + }, + { + "epoch": 0.13532000590272653, + "grad_norm": 0.177734375, + "learning_rate": 0.001897288861184735, + "loss": 0.1138, + "step": 15589 + }, + { + "epoch": 0.1353286863829307, + "grad_norm": 0.177734375, + "learning_rate": 0.0018972750533165585, + "loss": 0.1133, + "step": 15590 + }, + { + "epoch": 0.13533736686313486, + "grad_norm": 0.33203125, + "learning_rate": 0.0018972612445764942, + "loss": 0.1035, + "step": 15591 + }, + { + "epoch": 0.13534604734333902, + "grad_norm": 0.328125, + "learning_rate": 0.0018972474349645567, + "loss": 0.1641, + "step": 15592 + }, + { + "epoch": 0.13535472782354319, + "grad_norm": 0.3515625, + "learning_rate": 0.0018972336244807617, + "loss": 0.1455, + "step": 15593 + }, + { + "epoch": 0.13536340830374735, + "grad_norm": 0.2890625, + "learning_rate": 0.0018972198131251238, + "loss": 0.126, + "step": 15594 + }, + { + "epoch": 0.13537208878395152, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018972060008976583, + "loss": 0.1045, + "step": 15595 + }, + { + "epoch": 0.13538076926415568, + "grad_norm": 0.07177734375, + "learning_rate": 0.0018971921877983804, + "loss": 0.1211, + "step": 15596 + }, + { + "epoch": 0.13538944974435985, + "grad_norm": 0.3046875, + "learning_rate": 0.0018971783738273053, + "loss": 0.1523, + "step": 15597 + }, + { + "epoch": 0.135398130224564, + "grad_norm": 0.09423828125, + "learning_rate": 0.0018971645589844475, + "loss": 0.0879, + "step": 15598 + }, + { + "epoch": 0.13540681070476818, + "grad_norm": 0.41796875, + "learning_rate": 0.0018971507432698224, + "loss": 0.1299, + "step": 15599 + }, + { + "epoch": 0.13541549118497234, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018971369266834457, + "loss": 0.1104, + "step": 15600 + }, + { + "epoch": 0.1354241716651765, + "grad_norm": 0.0771484375, + "learning_rate": 0.0018971231092253318, + "loss": 0.1006, + "step": 15601 + }, + { + "epoch": 0.13543285214538067, + "grad_norm": 0.2373046875, + "learning_rate": 0.0018971092908954964, + "loss": 0.1021, + "step": 15602 + }, + { + "epoch": 0.13544153262558484, + "grad_norm": 0.1318359375, + "learning_rate": 0.001897095471693954, + "loss": 0.1191, + "step": 15603 + }, + { + "epoch": 0.135450213105789, + "grad_norm": 0.421875, + "learning_rate": 0.00189708165162072, + "loss": 0.1279, + "step": 15604 + }, + { + "epoch": 0.13545889358599317, + "grad_norm": 0.306640625, + "learning_rate": 0.001897067830675809, + "loss": 0.1172, + "step": 15605 + }, + { + "epoch": 0.13546757406619733, + "grad_norm": 0.10302734375, + "learning_rate": 0.0018970540088592371, + "loss": 0.1221, + "step": 15606 + }, + { + "epoch": 0.1354762545464015, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018970401861710187, + "loss": 0.0967, + "step": 15607 + }, + { + "epoch": 0.13548493502660566, + "grad_norm": 0.234375, + "learning_rate": 0.0018970263626111694, + "loss": 0.1006, + "step": 15608 + }, + { + "epoch": 0.13549361550680983, + "grad_norm": 0.2138671875, + "learning_rate": 0.001897012538179704, + "loss": 0.1123, + "step": 15609 + }, + { + "epoch": 0.135502295987014, + "grad_norm": 0.193359375, + "learning_rate": 0.0018969987128766375, + "loss": 0.1108, + "step": 15610 + }, + { + "epoch": 0.13551097646721816, + "grad_norm": 0.51953125, + "learning_rate": 0.001896984886701985, + "loss": 0.1543, + "step": 15611 + }, + { + "epoch": 0.13551965694742232, + "grad_norm": 0.1796875, + "learning_rate": 0.0018969710596557622, + "loss": 0.0942, + "step": 15612 + }, + { + "epoch": 0.1355283374276265, + "grad_norm": 1.1328125, + "learning_rate": 0.0018969572317379837, + "loss": 0.2158, + "step": 15613 + }, + { + "epoch": 0.13553701790783065, + "grad_norm": 0.609375, + "learning_rate": 0.0018969434029486646, + "loss": 0.1279, + "step": 15614 + }, + { + "epoch": 0.13554569838803482, + "grad_norm": 0.498046875, + "learning_rate": 0.0018969295732878203, + "loss": 0.1133, + "step": 15615 + }, + { + "epoch": 0.13555437886823898, + "grad_norm": 0.349609375, + "learning_rate": 0.0018969157427554657, + "loss": 0.1533, + "step": 15616 + }, + { + "epoch": 0.13556305934844315, + "grad_norm": 3.421875, + "learning_rate": 0.001896901911351616, + "loss": 0.7305, + "step": 15617 + }, + { + "epoch": 0.13557173982864731, + "grad_norm": 0.31640625, + "learning_rate": 0.0018968880790762863, + "loss": 0.1016, + "step": 15618 + }, + { + "epoch": 0.13558042030885148, + "grad_norm": 0.38671875, + "learning_rate": 0.0018968742459294918, + "loss": 0.1191, + "step": 15619 + }, + { + "epoch": 0.13558910078905564, + "grad_norm": 0.189453125, + "learning_rate": 0.0018968604119112477, + "loss": 0.0874, + "step": 15620 + }, + { + "epoch": 0.1355977812692598, + "grad_norm": 1.0859375, + "learning_rate": 0.0018968465770215692, + "loss": 0.1445, + "step": 15621 + }, + { + "epoch": 0.13560646174946397, + "grad_norm": 0.90234375, + "learning_rate": 0.0018968327412604712, + "loss": 0.1465, + "step": 15622 + }, + { + "epoch": 0.13561514222966814, + "grad_norm": 1.1015625, + "learning_rate": 0.0018968189046279687, + "loss": 0.1118, + "step": 15623 + }, + { + "epoch": 0.1356238227098723, + "grad_norm": 0.21484375, + "learning_rate": 0.0018968050671240768, + "loss": 0.1201, + "step": 15624 + }, + { + "epoch": 0.13563250319007647, + "grad_norm": 0.87890625, + "learning_rate": 0.0018967912287488112, + "loss": 0.2363, + "step": 15625 + }, + { + "epoch": 0.13564118367028064, + "grad_norm": 0.17578125, + "learning_rate": 0.0018967773895021865, + "loss": 0.1094, + "step": 15626 + }, + { + "epoch": 0.1356498641504848, + "grad_norm": 0.154296875, + "learning_rate": 0.0018967635493842186, + "loss": 0.123, + "step": 15627 + }, + { + "epoch": 0.13565854463068897, + "grad_norm": 0.2470703125, + "learning_rate": 0.0018967497083949214, + "loss": 0.1357, + "step": 15628 + }, + { + "epoch": 0.13566722511089313, + "grad_norm": 0.3046875, + "learning_rate": 0.0018967358665343108, + "loss": 0.1191, + "step": 15629 + }, + { + "epoch": 0.1356759055910973, + "grad_norm": 0.65234375, + "learning_rate": 0.0018967220238024017, + "loss": 0.1367, + "step": 15630 + }, + { + "epoch": 0.13568458607130146, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018967081801992097, + "loss": 0.1201, + "step": 15631 + }, + { + "epoch": 0.13569326655150563, + "grad_norm": 0.21875, + "learning_rate": 0.0018966943357247494, + "loss": 0.1367, + "step": 15632 + }, + { + "epoch": 0.1357019470317098, + "grad_norm": 0.5390625, + "learning_rate": 0.0018966804903790364, + "loss": 0.126, + "step": 15633 + }, + { + "epoch": 0.13571062751191396, + "grad_norm": 0.490234375, + "learning_rate": 0.0018966666441620854, + "loss": 0.1172, + "step": 15634 + }, + { + "epoch": 0.13571930799211812, + "grad_norm": 0.4296875, + "learning_rate": 0.0018966527970739118, + "loss": 0.1465, + "step": 15635 + }, + { + "epoch": 0.1357279884723223, + "grad_norm": 0.326171875, + "learning_rate": 0.0018966389491145306, + "loss": 0.1602, + "step": 15636 + }, + { + "epoch": 0.13573666895252645, + "grad_norm": 0.22265625, + "learning_rate": 0.0018966251002839568, + "loss": 0.1118, + "step": 15637 + }, + { + "epoch": 0.13574534943273062, + "grad_norm": 0.0947265625, + "learning_rate": 0.001896611250582206, + "loss": 0.1006, + "step": 15638 + }, + { + "epoch": 0.13575402991293478, + "grad_norm": 0.1025390625, + "learning_rate": 0.001896597400009293, + "loss": 0.0991, + "step": 15639 + }, + { + "epoch": 0.13576271039313895, + "grad_norm": 0.109375, + "learning_rate": 0.0018965835485652332, + "loss": 0.1309, + "step": 15640 + }, + { + "epoch": 0.1357713908733431, + "grad_norm": 0.404296875, + "learning_rate": 0.0018965696962500416, + "loss": 0.1396, + "step": 15641 + }, + { + "epoch": 0.13578007135354728, + "grad_norm": 0.099609375, + "learning_rate": 0.001896555843063733, + "loss": 0.0879, + "step": 15642 + }, + { + "epoch": 0.13578875183375144, + "grad_norm": 0.6796875, + "learning_rate": 0.0018965419890063233, + "loss": 0.1191, + "step": 15643 + }, + { + "epoch": 0.1357974323139556, + "grad_norm": 0.08837890625, + "learning_rate": 0.0018965281340778274, + "loss": 0.1484, + "step": 15644 + }, + { + "epoch": 0.13580611279415977, + "grad_norm": 0.20703125, + "learning_rate": 0.0018965142782782598, + "loss": 0.1523, + "step": 15645 + }, + { + "epoch": 0.13581479327436394, + "grad_norm": 0.248046875, + "learning_rate": 0.0018965004216076364, + "loss": 0.1055, + "step": 15646 + }, + { + "epoch": 0.1358234737545681, + "grad_norm": 0.07861328125, + "learning_rate": 0.001896486564065972, + "loss": 0.1484, + "step": 15647 + }, + { + "epoch": 0.13583215423477227, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018964727056532819, + "loss": 0.1436, + "step": 15648 + }, + { + "epoch": 0.13584083471497643, + "grad_norm": 0.318359375, + "learning_rate": 0.0018964588463695811, + "loss": 0.1123, + "step": 15649 + }, + { + "epoch": 0.1358495151951806, + "grad_norm": 0.734375, + "learning_rate": 0.001896444986214885, + "loss": 0.1484, + "step": 15650 + }, + { + "epoch": 0.13585819567538476, + "grad_norm": 0.62109375, + "learning_rate": 0.001896431125189209, + "loss": 0.1147, + "step": 15651 + }, + { + "epoch": 0.13586687615558893, + "grad_norm": 0.14453125, + "learning_rate": 0.0018964172632925673, + "loss": 0.0928, + "step": 15652 + }, + { + "epoch": 0.1358755566357931, + "grad_norm": 0.91015625, + "learning_rate": 0.0018964034005249759, + "loss": 0.1484, + "step": 15653 + }, + { + "epoch": 0.13588423711599726, + "grad_norm": 0.3828125, + "learning_rate": 0.0018963895368864499, + "loss": 0.1094, + "step": 15654 + }, + { + "epoch": 0.13589291759620142, + "grad_norm": 0.25390625, + "learning_rate": 0.0018963756723770037, + "loss": 0.1328, + "step": 15655 + }, + { + "epoch": 0.1359015980764056, + "grad_norm": 0.220703125, + "learning_rate": 0.0018963618069966538, + "loss": 0.0986, + "step": 15656 + }, + { + "epoch": 0.13591027855660975, + "grad_norm": 0.50390625, + "learning_rate": 0.0018963479407454141, + "loss": 0.1318, + "step": 15657 + }, + { + "epoch": 0.13591895903681392, + "grad_norm": 0.78125, + "learning_rate": 0.0018963340736233001, + "loss": 0.1113, + "step": 15658 + }, + { + "epoch": 0.13592763951701808, + "grad_norm": 0.330078125, + "learning_rate": 0.0018963202056303276, + "loss": 0.1396, + "step": 15659 + }, + { + "epoch": 0.13593631999722225, + "grad_norm": 0.232421875, + "learning_rate": 0.001896306336766511, + "loss": 0.0928, + "step": 15660 + }, + { + "epoch": 0.13594500047742641, + "grad_norm": 0.20703125, + "learning_rate": 0.0018962924670318658, + "loss": 0.0918, + "step": 15661 + }, + { + "epoch": 0.13595368095763058, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018962785964264073, + "loss": 0.1001, + "step": 15662 + }, + { + "epoch": 0.13596236143783474, + "grad_norm": 0.234375, + "learning_rate": 0.0018962647249501502, + "loss": 0.1475, + "step": 15663 + }, + { + "epoch": 0.1359710419180389, + "grad_norm": 0.126953125, + "learning_rate": 0.0018962508526031102, + "loss": 0.0981, + "step": 15664 + }, + { + "epoch": 0.13597972239824307, + "grad_norm": 0.2080078125, + "learning_rate": 0.0018962369793853022, + "loss": 0.1426, + "step": 15665 + }, + { + "epoch": 0.13598840287844724, + "grad_norm": 0.298828125, + "learning_rate": 0.0018962231052967412, + "loss": 0.1162, + "step": 15666 + }, + { + "epoch": 0.1359970833586514, + "grad_norm": 0.359375, + "learning_rate": 0.0018962092303374427, + "loss": 0.0918, + "step": 15667 + }, + { + "epoch": 0.13600576383885557, + "grad_norm": 0.396484375, + "learning_rate": 0.0018961953545074217, + "loss": 0.0986, + "step": 15668 + }, + { + "epoch": 0.13601444431905974, + "grad_norm": 0.27734375, + "learning_rate": 0.0018961814778066936, + "loss": 0.106, + "step": 15669 + }, + { + "epoch": 0.1360231247992639, + "grad_norm": 0.20703125, + "learning_rate": 0.0018961676002352732, + "loss": 0.105, + "step": 15670 + }, + { + "epoch": 0.13603180527946807, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018961537217931758, + "loss": 0.0889, + "step": 15671 + }, + { + "epoch": 0.13604048575967223, + "grad_norm": 0.390625, + "learning_rate": 0.0018961398424804168, + "loss": 0.1138, + "step": 15672 + }, + { + "epoch": 0.1360491662398764, + "grad_norm": 0.2578125, + "learning_rate": 0.0018961259622970112, + "loss": 0.1152, + "step": 15673 + }, + { + "epoch": 0.13605784672008056, + "grad_norm": 0.734375, + "learning_rate": 0.001896112081242974, + "loss": 0.1299, + "step": 15674 + }, + { + "epoch": 0.13606652720028473, + "grad_norm": 0.1943359375, + "learning_rate": 0.001896098199318321, + "loss": 0.1025, + "step": 15675 + }, + { + "epoch": 0.1360752076804889, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018960843165230668, + "loss": 0.1299, + "step": 15676 + }, + { + "epoch": 0.13608388816069306, + "grad_norm": 1.0234375, + "learning_rate": 0.001896070432857227, + "loss": 0.0908, + "step": 15677 + }, + { + "epoch": 0.13609256864089722, + "grad_norm": 0.201171875, + "learning_rate": 0.0018960565483208162, + "loss": 0.125, + "step": 15678 + }, + { + "epoch": 0.1361012491211014, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018960426629138498, + "loss": 0.0991, + "step": 15679 + }, + { + "epoch": 0.13610992960130555, + "grad_norm": 0.734375, + "learning_rate": 0.0018960287766363432, + "loss": 0.168, + "step": 15680 + }, + { + "epoch": 0.13611861008150972, + "grad_norm": 0.365234375, + "learning_rate": 0.001896014889488312, + "loss": 0.1377, + "step": 15681 + }, + { + "epoch": 0.13612729056171388, + "grad_norm": 0.08837890625, + "learning_rate": 0.00189600100146977, + "loss": 0.1289, + "step": 15682 + }, + { + "epoch": 0.13613597104191805, + "grad_norm": 0.478515625, + "learning_rate": 0.001895987112580734, + "loss": 0.1133, + "step": 15683 + }, + { + "epoch": 0.1361446515221222, + "grad_norm": 0.38671875, + "learning_rate": 0.0018959732228212183, + "loss": 0.1338, + "step": 15684 + }, + { + "epoch": 0.13615333200232638, + "grad_norm": 0.09521484375, + "learning_rate": 0.001895959332191238, + "loss": 0.1201, + "step": 15685 + }, + { + "epoch": 0.13616201248253054, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018959454406908085, + "loss": 0.1445, + "step": 15686 + }, + { + "epoch": 0.1361706929627347, + "grad_norm": 0.1787109375, + "learning_rate": 0.0018959315483199452, + "loss": 0.1699, + "step": 15687 + }, + { + "epoch": 0.13617937344293887, + "grad_norm": 0.1015625, + "learning_rate": 0.0018959176550786634, + "loss": 0.1436, + "step": 15688 + }, + { + "epoch": 0.13618805392314304, + "grad_norm": 0.91015625, + "learning_rate": 0.0018959037609669778, + "loss": 0.084, + "step": 15689 + }, + { + "epoch": 0.1361967344033472, + "grad_norm": 0.59375, + "learning_rate": 0.0018958898659849036, + "loss": 0.0947, + "step": 15690 + }, + { + "epoch": 0.13620541488355137, + "grad_norm": 0.53515625, + "learning_rate": 0.0018958759701324561, + "loss": 0.0913, + "step": 15691 + }, + { + "epoch": 0.13621409536375553, + "grad_norm": 0.91796875, + "learning_rate": 0.001895862073409651, + "loss": 0.1089, + "step": 15692 + }, + { + "epoch": 0.1362227758439597, + "grad_norm": 0.11865234375, + "learning_rate": 0.001895848175816503, + "loss": 0.1196, + "step": 15693 + }, + { + "epoch": 0.13623145632416386, + "grad_norm": 0.494140625, + "learning_rate": 0.0018958342773530274, + "loss": 0.1006, + "step": 15694 + }, + { + "epoch": 0.13624013680436803, + "grad_norm": 0.220703125, + "learning_rate": 0.001895820378019239, + "loss": 0.1074, + "step": 15695 + }, + { + "epoch": 0.1362488172845722, + "grad_norm": 0.47265625, + "learning_rate": 0.001895806477815154, + "loss": 0.123, + "step": 15696 + }, + { + "epoch": 0.13625749776477636, + "grad_norm": 0.36328125, + "learning_rate": 0.0018957925767407865, + "loss": 0.1484, + "step": 15697 + }, + { + "epoch": 0.13626617824498052, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018957786747961524, + "loss": 0.0835, + "step": 15698 + }, + { + "epoch": 0.1362748587251847, + "grad_norm": 0.287109375, + "learning_rate": 0.0018957647719812668, + "loss": 0.0967, + "step": 15699 + }, + { + "epoch": 0.13628353920538885, + "grad_norm": 0.181640625, + "learning_rate": 0.0018957508682961445, + "loss": 0.1221, + "step": 15700 + }, + { + "epoch": 0.13629221968559302, + "grad_norm": 0.75390625, + "learning_rate": 0.0018957369637408012, + "loss": 0.1152, + "step": 15701 + }, + { + "epoch": 0.13630090016579718, + "grad_norm": 0.6875, + "learning_rate": 0.001895723058315252, + "loss": 0.1416, + "step": 15702 + }, + { + "epoch": 0.13630958064600135, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018957091520195118, + "loss": 0.1074, + "step": 15703 + }, + { + "epoch": 0.13631826112620551, + "grad_norm": 0.484375, + "learning_rate": 0.0018956952448535961, + "loss": 0.2109, + "step": 15704 + }, + { + "epoch": 0.13632694160640968, + "grad_norm": 0.09619140625, + "learning_rate": 0.00189568133681752, + "loss": 0.1348, + "step": 15705 + }, + { + "epoch": 0.13633562208661384, + "grad_norm": 0.458984375, + "learning_rate": 0.0018956674279112988, + "loss": 0.1201, + "step": 15706 + }, + { + "epoch": 0.136344302566818, + "grad_norm": 0.1640625, + "learning_rate": 0.0018956535181349479, + "loss": 0.1133, + "step": 15707 + }, + { + "epoch": 0.13635298304702217, + "grad_norm": 0.640625, + "learning_rate": 0.001895639607488482, + "loss": 0.1289, + "step": 15708 + }, + { + "epoch": 0.13636166352722634, + "grad_norm": 0.443359375, + "learning_rate": 0.0018956256959719168, + "loss": 0.1235, + "step": 15709 + }, + { + "epoch": 0.13637034400743048, + "grad_norm": 0.259765625, + "learning_rate": 0.001895611783585267, + "loss": 0.167, + "step": 15710 + }, + { + "epoch": 0.13637902448763464, + "grad_norm": 0.310546875, + "learning_rate": 0.0018955978703285481, + "loss": 0.1162, + "step": 15711 + }, + { + "epoch": 0.1363877049678388, + "grad_norm": 0.19140625, + "learning_rate": 0.0018955839562017757, + "loss": 0.085, + "step": 15712 + }, + { + "epoch": 0.13639638544804297, + "grad_norm": 0.46484375, + "learning_rate": 0.0018955700412049642, + "loss": 0.0796, + "step": 15713 + }, + { + "epoch": 0.13640506592824714, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018955561253381295, + "loss": 0.1289, + "step": 15714 + }, + { + "epoch": 0.1364137464084513, + "grad_norm": 0.2265625, + "learning_rate": 0.0018955422086012863, + "loss": 0.1201, + "step": 15715 + }, + { + "epoch": 0.13642242688865547, + "grad_norm": 0.73828125, + "learning_rate": 0.0018955282909944504, + "loss": 0.0806, + "step": 15716 + }, + { + "epoch": 0.13643110736885963, + "grad_norm": 0.5859375, + "learning_rate": 0.0018955143725176368, + "loss": 0.1016, + "step": 15717 + }, + { + "epoch": 0.1364397878490638, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018955004531708605, + "loss": 0.1699, + "step": 15718 + }, + { + "epoch": 0.13644846832926796, + "grad_norm": 0.482421875, + "learning_rate": 0.0018954865329541366, + "loss": 0.1025, + "step": 15719 + }, + { + "epoch": 0.13645714880947213, + "grad_norm": 0.373046875, + "learning_rate": 0.001895472611867481, + "loss": 0.1226, + "step": 15720 + }, + { + "epoch": 0.1364658292896763, + "grad_norm": 0.3671875, + "learning_rate": 0.0018954586899109084, + "loss": 0.1211, + "step": 15721 + }, + { + "epoch": 0.13647450976988046, + "grad_norm": 0.125, + "learning_rate": 0.0018954447670844338, + "loss": 0.1416, + "step": 15722 + }, + { + "epoch": 0.13648319025008462, + "grad_norm": 0.177734375, + "learning_rate": 0.001895430843388073, + "loss": 0.1221, + "step": 15723 + }, + { + "epoch": 0.1364918707302888, + "grad_norm": 0.10498046875, + "learning_rate": 0.001895416918821841, + "loss": 0.0913, + "step": 15724 + }, + { + "epoch": 0.13650055121049295, + "grad_norm": 0.58984375, + "learning_rate": 0.0018954029933857528, + "loss": 0.1123, + "step": 15725 + }, + { + "epoch": 0.13650923169069712, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018953890670798243, + "loss": 0.1069, + "step": 15726 + }, + { + "epoch": 0.13651791217090128, + "grad_norm": 0.296875, + "learning_rate": 0.0018953751399040698, + "loss": 0.083, + "step": 15727 + }, + { + "epoch": 0.13652659265110545, + "grad_norm": 0.369140625, + "learning_rate": 0.001895361211858505, + "loss": 0.1162, + "step": 15728 + }, + { + "epoch": 0.13653527313130961, + "grad_norm": 0.208984375, + "learning_rate": 0.0018953472829431456, + "loss": 0.1084, + "step": 15729 + }, + { + "epoch": 0.13654395361151378, + "grad_norm": 0.546875, + "learning_rate": 0.001895333353158006, + "loss": 0.0918, + "step": 15730 + }, + { + "epoch": 0.13655263409171794, + "grad_norm": 0.091796875, + "learning_rate": 0.0018953194225031019, + "loss": 0.1064, + "step": 15731 + }, + { + "epoch": 0.1365613145719221, + "grad_norm": 0.220703125, + "learning_rate": 0.0018953054909784483, + "loss": 0.1357, + "step": 15732 + }, + { + "epoch": 0.13656999505212628, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018952915585840604, + "loss": 0.1104, + "step": 15733 + }, + { + "epoch": 0.13657867553233044, + "grad_norm": 0.271484375, + "learning_rate": 0.001895277625319954, + "loss": 0.1108, + "step": 15734 + }, + { + "epoch": 0.1365873560125346, + "grad_norm": 1.546875, + "learning_rate": 0.001895263691186144, + "loss": 0.0952, + "step": 15735 + }, + { + "epoch": 0.13659603649273877, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018952497561826455, + "loss": 0.0947, + "step": 15736 + }, + { + "epoch": 0.13660471697294294, + "grad_norm": 0.4453125, + "learning_rate": 0.0018952358203094734, + "loss": 0.1279, + "step": 15737 + }, + { + "epoch": 0.1366133974531471, + "grad_norm": 0.28515625, + "learning_rate": 0.0018952218835666438, + "loss": 0.1118, + "step": 15738 + }, + { + "epoch": 0.13662207793335127, + "grad_norm": 0.67578125, + "learning_rate": 0.0018952079459541712, + "loss": 0.1279, + "step": 15739 + }, + { + "epoch": 0.13663075841355543, + "grad_norm": 0.240234375, + "learning_rate": 0.0018951940074720714, + "loss": 0.125, + "step": 15740 + }, + { + "epoch": 0.1366394388937596, + "grad_norm": 0.2177734375, + "learning_rate": 0.001895180068120359, + "loss": 0.1299, + "step": 15741 + }, + { + "epoch": 0.13664811937396376, + "grad_norm": 0.326171875, + "learning_rate": 0.00189516612789905, + "loss": 0.1064, + "step": 15742 + }, + { + "epoch": 0.13665679985416793, + "grad_norm": 0.9375, + "learning_rate": 0.0018951521868081592, + "loss": 0.126, + "step": 15743 + }, + { + "epoch": 0.1366654803343721, + "grad_norm": 0.1865234375, + "learning_rate": 0.001895138244847702, + "loss": 0.1592, + "step": 15744 + }, + { + "epoch": 0.13667416081457626, + "grad_norm": 0.15625, + "learning_rate": 0.0018951243020176934, + "loss": 0.1504, + "step": 15745 + }, + { + "epoch": 0.13668284129478042, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018951103583181486, + "loss": 0.1406, + "step": 15746 + }, + { + "epoch": 0.1366915217749846, + "grad_norm": 0.57421875, + "learning_rate": 0.0018950964137490834, + "loss": 0.1299, + "step": 15747 + }, + { + "epoch": 0.13670020225518875, + "grad_norm": 0.08984375, + "learning_rate": 0.0018950824683105125, + "loss": 0.0933, + "step": 15748 + }, + { + "epoch": 0.13670888273539292, + "grad_norm": 0.33203125, + "learning_rate": 0.0018950685220024514, + "loss": 0.1621, + "step": 15749 + }, + { + "epoch": 0.13671756321559708, + "grad_norm": 0.107421875, + "learning_rate": 0.0018950545748249157, + "loss": 0.1426, + "step": 15750 + }, + { + "epoch": 0.13672624369580125, + "grad_norm": 0.1337890625, + "learning_rate": 0.00189504062677792, + "loss": 0.0918, + "step": 15751 + }, + { + "epoch": 0.1367349241760054, + "grad_norm": 0.322265625, + "learning_rate": 0.0018950266778614796, + "loss": 0.1011, + "step": 15752 + }, + { + "epoch": 0.13674360465620958, + "grad_norm": 0.125, + "learning_rate": 0.00189501272807561, + "loss": 0.1338, + "step": 15753 + }, + { + "epoch": 0.13675228513641374, + "grad_norm": 0.29296875, + "learning_rate": 0.0018949987774203266, + "loss": 0.1201, + "step": 15754 + }, + { + "epoch": 0.1367609656166179, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018949848258956446, + "loss": 0.1309, + "step": 15755 + }, + { + "epoch": 0.13676964609682207, + "grad_norm": 0.10302734375, + "learning_rate": 0.001894970873501579, + "loss": 0.1191, + "step": 15756 + }, + { + "epoch": 0.13677832657702624, + "grad_norm": 0.091796875, + "learning_rate": 0.001894956920238145, + "loss": 0.0947, + "step": 15757 + }, + { + "epoch": 0.1367870070572304, + "grad_norm": 1.1484375, + "learning_rate": 0.0018949429661053582, + "loss": 0.1533, + "step": 15758 + }, + { + "epoch": 0.13679568753743457, + "grad_norm": 0.53125, + "learning_rate": 0.0018949290111032338, + "loss": 0.1143, + "step": 15759 + }, + { + "epoch": 0.13680436801763873, + "grad_norm": 0.08544921875, + "learning_rate": 0.0018949150552317869, + "loss": 0.126, + "step": 15760 + }, + { + "epoch": 0.1368130484978429, + "grad_norm": 0.84765625, + "learning_rate": 0.0018949010984910327, + "loss": 0.1074, + "step": 15761 + }, + { + "epoch": 0.13682172897804706, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018948871408809868, + "loss": 0.0957, + "step": 15762 + }, + { + "epoch": 0.13683040945825123, + "grad_norm": 0.30859375, + "learning_rate": 0.0018948731824016641, + "loss": 0.1138, + "step": 15763 + }, + { + "epoch": 0.1368390899384554, + "grad_norm": 0.365234375, + "learning_rate": 0.0018948592230530804, + "loss": 0.1138, + "step": 15764 + }, + { + "epoch": 0.13684777041865956, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018948452628352501, + "loss": 0.1992, + "step": 15765 + }, + { + "epoch": 0.13685645089886372, + "grad_norm": 0.34375, + "learning_rate": 0.0018948313017481894, + "loss": 0.0991, + "step": 15766 + }, + { + "epoch": 0.1368651313790679, + "grad_norm": 0.50390625, + "learning_rate": 0.0018948173397919132, + "loss": 0.1699, + "step": 15767 + }, + { + "epoch": 0.13687381185927205, + "grad_norm": 0.375, + "learning_rate": 0.0018948033769664362, + "loss": 0.127, + "step": 15768 + }, + { + "epoch": 0.13688249233947622, + "grad_norm": 0.298828125, + "learning_rate": 0.0018947894132717742, + "loss": 0.1182, + "step": 15769 + }, + { + "epoch": 0.13689117281968038, + "grad_norm": 0.99609375, + "learning_rate": 0.001894775448707943, + "loss": 0.1816, + "step": 15770 + }, + { + "epoch": 0.13689985329988455, + "grad_norm": 0.296875, + "learning_rate": 0.001894761483274957, + "loss": 0.1021, + "step": 15771 + }, + { + "epoch": 0.13690853378008871, + "grad_norm": 0.52734375, + "learning_rate": 0.0018947475169728315, + "loss": 0.1064, + "step": 15772 + }, + { + "epoch": 0.13691721426029288, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018947335498015823, + "loss": 0.1201, + "step": 15773 + }, + { + "epoch": 0.13692589474049705, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018947195817612247, + "loss": 0.0996, + "step": 15774 + }, + { + "epoch": 0.1369345752207012, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018947056128517732, + "loss": 0.1406, + "step": 15775 + }, + { + "epoch": 0.13694325570090538, + "grad_norm": 0.318359375, + "learning_rate": 0.001894691643073244, + "loss": 0.0957, + "step": 15776 + }, + { + "epoch": 0.13695193618110954, + "grad_norm": 0.35546875, + "learning_rate": 0.0018946776724256514, + "loss": 0.1104, + "step": 15777 + }, + { + "epoch": 0.1369606166613137, + "grad_norm": 0.291015625, + "learning_rate": 0.0018946637009090118, + "loss": 0.085, + "step": 15778 + }, + { + "epoch": 0.13696929714151787, + "grad_norm": 0.6015625, + "learning_rate": 0.0018946497285233397, + "loss": 0.1162, + "step": 15779 + }, + { + "epoch": 0.13697797762172204, + "grad_norm": 0.3046875, + "learning_rate": 0.0018946357552686506, + "loss": 0.124, + "step": 15780 + }, + { + "epoch": 0.1369866581019262, + "grad_norm": 1.015625, + "learning_rate": 0.0018946217811449597, + "loss": 0.0713, + "step": 15781 + }, + { + "epoch": 0.13699533858213037, + "grad_norm": 0.5625, + "learning_rate": 0.0018946078061522827, + "loss": 0.1172, + "step": 15782 + }, + { + "epoch": 0.13700401906233453, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018945938302906344, + "loss": 0.1367, + "step": 15783 + }, + { + "epoch": 0.1370126995425387, + "grad_norm": 0.3984375, + "learning_rate": 0.00189457985356003, + "loss": 0.1196, + "step": 15784 + }, + { + "epoch": 0.13702138002274286, + "grad_norm": 0.07861328125, + "learning_rate": 0.001894565875960485, + "loss": 0.0923, + "step": 15785 + }, + { + "epoch": 0.13703006050294703, + "grad_norm": 0.33203125, + "learning_rate": 0.0018945518974920151, + "loss": 0.1279, + "step": 15786 + }, + { + "epoch": 0.1370387409831512, + "grad_norm": 0.10009765625, + "learning_rate": 0.001894537918154635, + "loss": 0.0967, + "step": 15787 + }, + { + "epoch": 0.13704742146335536, + "grad_norm": 0.267578125, + "learning_rate": 0.0018945239379483598, + "loss": 0.1289, + "step": 15788 + }, + { + "epoch": 0.13705610194355952, + "grad_norm": 0.080078125, + "learning_rate": 0.0018945099568732054, + "loss": 0.083, + "step": 15789 + }, + { + "epoch": 0.1370647824237637, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018944959749291868, + "loss": 0.0894, + "step": 15790 + }, + { + "epoch": 0.13707346290396785, + "grad_norm": 0.5859375, + "learning_rate": 0.0018944819921163196, + "loss": 0.1064, + "step": 15791 + }, + { + "epoch": 0.13708214338417202, + "grad_norm": 0.3359375, + "learning_rate": 0.0018944680084346187, + "loss": 0.1445, + "step": 15792 + }, + { + "epoch": 0.13709082386437618, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018944540238840994, + "loss": 0.1152, + "step": 15793 + }, + { + "epoch": 0.13709950434458035, + "grad_norm": 0.390625, + "learning_rate": 0.0018944400384647772, + "loss": 0.1279, + "step": 15794 + }, + { + "epoch": 0.1371081848247845, + "grad_norm": 0.06640625, + "learning_rate": 0.0018944260521766673, + "loss": 0.1104, + "step": 15795 + }, + { + "epoch": 0.13711686530498868, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018944120650197854, + "loss": 0.1177, + "step": 15796 + }, + { + "epoch": 0.13712554578519284, + "grad_norm": 0.466796875, + "learning_rate": 0.001894398076994146, + "loss": 0.1465, + "step": 15797 + }, + { + "epoch": 0.137134226265397, + "grad_norm": 0.271484375, + "learning_rate": 0.0018943840880997647, + "loss": 0.0991, + "step": 15798 + }, + { + "epoch": 0.13714290674560117, + "grad_norm": 1.265625, + "learning_rate": 0.001894370098336657, + "loss": 0.1338, + "step": 15799 + }, + { + "epoch": 0.13715158722580534, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018943561077048381, + "loss": 0.1367, + "step": 15800 + }, + { + "epoch": 0.1371602677060095, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018943421162043233, + "loss": 0.1157, + "step": 15801 + }, + { + "epoch": 0.13716894818621367, + "grad_norm": 0.484375, + "learning_rate": 0.001894328123835128, + "loss": 0.1592, + "step": 15802 + }, + { + "epoch": 0.13717762866641783, + "grad_norm": 0.404296875, + "learning_rate": 0.0018943141305972674, + "loss": 0.1426, + "step": 15803 + }, + { + "epoch": 0.137186309146622, + "grad_norm": 0.35546875, + "learning_rate": 0.0018943001364907567, + "loss": 0.1182, + "step": 15804 + }, + { + "epoch": 0.13719498962682616, + "grad_norm": 0.224609375, + "learning_rate": 0.001894286141515611, + "loss": 0.0889, + "step": 15805 + }, + { + "epoch": 0.13720367010703033, + "grad_norm": 0.5546875, + "learning_rate": 0.0018942721456718461, + "loss": 0.125, + "step": 15806 + }, + { + "epoch": 0.1372123505872345, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018942581489594774, + "loss": 0.1758, + "step": 15807 + }, + { + "epoch": 0.13722103106743866, + "grad_norm": 1.125, + "learning_rate": 0.00189424415137852, + "loss": 0.1094, + "step": 15808 + }, + { + "epoch": 0.13722971154764282, + "grad_norm": 0.83203125, + "learning_rate": 0.001894230152928989, + "loss": 0.1475, + "step": 15809 + }, + { + "epoch": 0.137238392027847, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018942161536108994, + "loss": 0.1309, + "step": 15810 + }, + { + "epoch": 0.13724707250805115, + "grad_norm": 0.111328125, + "learning_rate": 0.0018942021534242672, + "loss": 0.0957, + "step": 15811 + }, + { + "epoch": 0.13725575298825532, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018941881523691076, + "loss": 0.1094, + "step": 15812 + }, + { + "epoch": 0.13726443346845948, + "grad_norm": 0.5859375, + "learning_rate": 0.001894174150445436, + "loss": 0.1143, + "step": 15813 + }, + { + "epoch": 0.13727311394866365, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018941601476532668, + "loss": 0.1099, + "step": 15814 + }, + { + "epoch": 0.13728179442886781, + "grad_norm": 0.392578125, + "learning_rate": 0.0018941461439926166, + "loss": 0.1094, + "step": 15815 + }, + { + "epoch": 0.13729047490907198, + "grad_norm": 0.546875, + "learning_rate": 0.0018941321394634998, + "loss": 0.0928, + "step": 15816 + }, + { + "epoch": 0.13729915538927615, + "grad_norm": 0.12060546875, + "learning_rate": 0.0018941181340659322, + "loss": 0.1113, + "step": 15817 + }, + { + "epoch": 0.1373078358694803, + "grad_norm": 0.244140625, + "learning_rate": 0.0018941041277999286, + "loss": 0.1533, + "step": 15818 + }, + { + "epoch": 0.13731651634968448, + "grad_norm": 0.3046875, + "learning_rate": 0.0018940901206655053, + "loss": 0.1172, + "step": 15819 + }, + { + "epoch": 0.13732519682988864, + "grad_norm": 0.083984375, + "learning_rate": 0.0018940761126626765, + "loss": 0.1089, + "step": 15820 + }, + { + "epoch": 0.1373338773100928, + "grad_norm": 0.09033203125, + "learning_rate": 0.001894062103791458, + "loss": 0.1152, + "step": 15821 + }, + { + "epoch": 0.13734255779029697, + "grad_norm": 0.11376953125, + "learning_rate": 0.0018940480940518652, + "loss": 0.1055, + "step": 15822 + }, + { + "epoch": 0.13735123827050114, + "grad_norm": 0.42578125, + "learning_rate": 0.0018940340834439134, + "loss": 0.1582, + "step": 15823 + }, + { + "epoch": 0.1373599187507053, + "grad_norm": 0.76953125, + "learning_rate": 0.001894020071967618, + "loss": 0.0967, + "step": 15824 + }, + { + "epoch": 0.13736859923090947, + "grad_norm": 0.69921875, + "learning_rate": 0.0018940060596229936, + "loss": 0.1147, + "step": 15825 + }, + { + "epoch": 0.13737727971111363, + "grad_norm": 0.09912109375, + "learning_rate": 0.0018939920464100567, + "loss": 0.1147, + "step": 15826 + }, + { + "epoch": 0.1373859601913178, + "grad_norm": 0.625, + "learning_rate": 0.0018939780323288218, + "loss": 0.1182, + "step": 15827 + }, + { + "epoch": 0.13739464067152196, + "grad_norm": 0.279296875, + "learning_rate": 0.0018939640173793045, + "loss": 0.104, + "step": 15828 + }, + { + "epoch": 0.13740332115172613, + "grad_norm": 0.392578125, + "learning_rate": 0.00189395000156152, + "loss": 0.1191, + "step": 15829 + }, + { + "epoch": 0.1374120016319303, + "grad_norm": 0.062255859375, + "learning_rate": 0.0018939359848754835, + "loss": 0.1074, + "step": 15830 + }, + { + "epoch": 0.13742068211213446, + "grad_norm": 0.11962890625, + "learning_rate": 0.001893921967321211, + "loss": 0.1104, + "step": 15831 + }, + { + "epoch": 0.13742936259233862, + "grad_norm": 0.0771484375, + "learning_rate": 0.0018939079488987168, + "loss": 0.1016, + "step": 15832 + }, + { + "epoch": 0.13743804307254276, + "grad_norm": 0.71484375, + "learning_rate": 0.0018938939296080173, + "loss": 0.125, + "step": 15833 + }, + { + "epoch": 0.13744672355274692, + "grad_norm": 0.1953125, + "learning_rate": 0.0018938799094491271, + "loss": 0.125, + "step": 15834 + }, + { + "epoch": 0.1374554040329511, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018938658884220618, + "loss": 0.085, + "step": 15835 + }, + { + "epoch": 0.13746408451315525, + "grad_norm": 0.10302734375, + "learning_rate": 0.0018938518665268368, + "loss": 0.1543, + "step": 15836 + }, + { + "epoch": 0.13747276499335942, + "grad_norm": 0.1064453125, + "learning_rate": 0.001893837843763467, + "loss": 0.1357, + "step": 15837 + }, + { + "epoch": 0.13748144547356358, + "grad_norm": 0.38671875, + "learning_rate": 0.0018938238201319686, + "loss": 0.1006, + "step": 15838 + }, + { + "epoch": 0.13749012595376775, + "grad_norm": 0.388671875, + "learning_rate": 0.001893809795632356, + "loss": 0.1045, + "step": 15839 + }, + { + "epoch": 0.13749880643397192, + "grad_norm": 0.07373046875, + "learning_rate": 0.0018937957702646448, + "loss": 0.0967, + "step": 15840 + }, + { + "epoch": 0.13750748691417608, + "grad_norm": 0.271484375, + "learning_rate": 0.0018937817440288509, + "loss": 0.1621, + "step": 15841 + }, + { + "epoch": 0.13751616739438025, + "grad_norm": 0.1845703125, + "learning_rate": 0.001893767716924989, + "loss": 0.1177, + "step": 15842 + }, + { + "epoch": 0.1375248478745844, + "grad_norm": 0.251953125, + "learning_rate": 0.001893753688953075, + "loss": 0.1035, + "step": 15843 + }, + { + "epoch": 0.13753352835478858, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018937396601131232, + "loss": 0.1748, + "step": 15844 + }, + { + "epoch": 0.13754220883499274, + "grad_norm": 0.25390625, + "learning_rate": 0.0018937256304051501, + "loss": 0.0928, + "step": 15845 + }, + { + "epoch": 0.1375508893151969, + "grad_norm": 0.5625, + "learning_rate": 0.0018937115998291702, + "loss": 0.0928, + "step": 15846 + }, + { + "epoch": 0.13755956979540107, + "grad_norm": 0.263671875, + "learning_rate": 0.0018936975683851995, + "loss": 0.1211, + "step": 15847 + }, + { + "epoch": 0.13756825027560524, + "grad_norm": 0.302734375, + "learning_rate": 0.0018936835360732532, + "loss": 0.1172, + "step": 15848 + }, + { + "epoch": 0.1375769307558094, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018936695028933465, + "loss": 0.124, + "step": 15849 + }, + { + "epoch": 0.13758561123601357, + "grad_norm": 1.03125, + "learning_rate": 0.0018936554688454945, + "loss": 0.0942, + "step": 15850 + }, + { + "epoch": 0.13759429171621773, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018936414339297129, + "loss": 0.1123, + "step": 15851 + }, + { + "epoch": 0.1376029721964219, + "grad_norm": 1.28125, + "learning_rate": 0.0018936273981460172, + "loss": 0.1719, + "step": 15852 + }, + { + "epoch": 0.13761165267662606, + "grad_norm": 0.375, + "learning_rate": 0.0018936133614944222, + "loss": 0.1582, + "step": 15853 + }, + { + "epoch": 0.13762033315683023, + "grad_norm": 0.59375, + "learning_rate": 0.0018935993239749436, + "loss": 0.1172, + "step": 15854 + }, + { + "epoch": 0.1376290136370344, + "grad_norm": 0.197265625, + "learning_rate": 0.001893585285587597, + "loss": 0.1523, + "step": 15855 + }, + { + "epoch": 0.13763769411723856, + "grad_norm": 0.58203125, + "learning_rate": 0.0018935712463323967, + "loss": 0.1973, + "step": 15856 + }, + { + "epoch": 0.13764637459744272, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018935572062093594, + "loss": 0.1064, + "step": 15857 + }, + { + "epoch": 0.1376550550776469, + "grad_norm": 0.130859375, + "learning_rate": 0.0018935431652184998, + "loss": 0.1167, + "step": 15858 + }, + { + "epoch": 0.13766373555785105, + "grad_norm": 0.419921875, + "learning_rate": 0.0018935291233598331, + "loss": 0.1055, + "step": 15859 + }, + { + "epoch": 0.13767241603805522, + "grad_norm": 0.5, + "learning_rate": 0.001893515080633375, + "loss": 0.1504, + "step": 15860 + }, + { + "epoch": 0.13768109651825938, + "grad_norm": 0.322265625, + "learning_rate": 0.001893501037039141, + "loss": 0.1406, + "step": 15861 + }, + { + "epoch": 0.13768977699846355, + "grad_norm": 0.6484375, + "learning_rate": 0.001893486992577146, + "loss": 0.1143, + "step": 15862 + }, + { + "epoch": 0.1376984574786677, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018934729472474055, + "loss": 0.1309, + "step": 15863 + }, + { + "epoch": 0.13770713795887188, + "grad_norm": 0.30859375, + "learning_rate": 0.001893458901049935, + "loss": 0.1553, + "step": 15864 + }, + { + "epoch": 0.13771581843907604, + "grad_norm": 0.083984375, + "learning_rate": 0.0018934448539847494, + "loss": 0.0957, + "step": 15865 + }, + { + "epoch": 0.1377244989192802, + "grad_norm": 0.55859375, + "learning_rate": 0.001893430806051865, + "loss": 0.1406, + "step": 15866 + }, + { + "epoch": 0.13773317939948437, + "grad_norm": 0.6484375, + "learning_rate": 0.0018934167572512962, + "loss": 0.1016, + "step": 15867 + }, + { + "epoch": 0.13774185987968854, + "grad_norm": 0.89453125, + "learning_rate": 0.001893402707583059, + "loss": 0.1357, + "step": 15868 + }, + { + "epoch": 0.1377505403598927, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018933886570471684, + "loss": 0.1309, + "step": 15869 + }, + { + "epoch": 0.13775922084009687, + "grad_norm": 0.080078125, + "learning_rate": 0.0018933746056436398, + "loss": 0.1133, + "step": 15870 + }, + { + "epoch": 0.13776790132030103, + "grad_norm": 0.6640625, + "learning_rate": 0.0018933605533724886, + "loss": 0.105, + "step": 15871 + }, + { + "epoch": 0.1377765818005052, + "grad_norm": 0.248046875, + "learning_rate": 0.0018933465002337305, + "loss": 0.0879, + "step": 15872 + }, + { + "epoch": 0.13778526228070936, + "grad_norm": 0.396484375, + "learning_rate": 0.0018933324462273803, + "loss": 0.0918, + "step": 15873 + }, + { + "epoch": 0.13779394276091353, + "grad_norm": 0.294921875, + "learning_rate": 0.0018933183913534537, + "loss": 0.1035, + "step": 15874 + }, + { + "epoch": 0.1378026232411177, + "grad_norm": 0.578125, + "learning_rate": 0.0018933043356119663, + "loss": 0.1406, + "step": 15875 + }, + { + "epoch": 0.13781130372132186, + "grad_norm": 0.1826171875, + "learning_rate": 0.001893290279002933, + "loss": 0.0991, + "step": 15876 + }, + { + "epoch": 0.13781998420152602, + "grad_norm": 0.1162109375, + "learning_rate": 0.0018932762215263692, + "loss": 0.1084, + "step": 15877 + }, + { + "epoch": 0.1378286646817302, + "grad_norm": 0.27734375, + "learning_rate": 0.0018932621631822905, + "loss": 0.1279, + "step": 15878 + }, + { + "epoch": 0.13783734516193435, + "grad_norm": 1.0234375, + "learning_rate": 0.0018932481039707126, + "loss": 0.1055, + "step": 15879 + }, + { + "epoch": 0.13784602564213852, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018932340438916502, + "loss": 0.1699, + "step": 15880 + }, + { + "epoch": 0.13785470612234269, + "grad_norm": 0.5078125, + "learning_rate": 0.001893219982945119, + "loss": 0.123, + "step": 15881 + }, + { + "epoch": 0.13786338660254685, + "grad_norm": 0.5390625, + "learning_rate": 0.0018932059211311343, + "loss": 0.1445, + "step": 15882 + }, + { + "epoch": 0.13787206708275102, + "grad_norm": 0.07861328125, + "learning_rate": 0.0018931918584497115, + "loss": 0.0977, + "step": 15883 + }, + { + "epoch": 0.13788074756295518, + "grad_norm": 0.390625, + "learning_rate": 0.001893177794900866, + "loss": 0.124, + "step": 15884 + }, + { + "epoch": 0.13788942804315935, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018931637304846133, + "loss": 0.0913, + "step": 15885 + }, + { + "epoch": 0.1378981085233635, + "grad_norm": 0.220703125, + "learning_rate": 0.0018931496652009685, + "loss": 0.084, + "step": 15886 + }, + { + "epoch": 0.13790678900356768, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018931355990499475, + "loss": 0.1543, + "step": 15887 + }, + { + "epoch": 0.13791546948377184, + "grad_norm": 0.875, + "learning_rate": 0.0018931215320315649, + "loss": 0.1113, + "step": 15888 + }, + { + "epoch": 0.137924149963976, + "grad_norm": 0.3203125, + "learning_rate": 0.0018931074641458368, + "loss": 0.1118, + "step": 15889 + }, + { + "epoch": 0.13793283044418017, + "grad_norm": 0.85546875, + "learning_rate": 0.001893093395392778, + "loss": 0.1064, + "step": 15890 + }, + { + "epoch": 0.13794151092438434, + "grad_norm": 0.16796875, + "learning_rate": 0.0018930793257724045, + "loss": 0.1216, + "step": 15891 + }, + { + "epoch": 0.1379501914045885, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018930652552847313, + "loss": 0.127, + "step": 15892 + }, + { + "epoch": 0.13795887188479267, + "grad_norm": 0.1435546875, + "learning_rate": 0.001893051183929774, + "loss": 0.1147, + "step": 15893 + }, + { + "epoch": 0.13796755236499683, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018930371117075473, + "loss": 0.0913, + "step": 15894 + }, + { + "epoch": 0.137976232845201, + "grad_norm": 0.498046875, + "learning_rate": 0.0018930230386180673, + "loss": 0.0884, + "step": 15895 + }, + { + "epoch": 0.13798491332540516, + "grad_norm": 0.69140625, + "learning_rate": 0.0018930089646613496, + "loss": 0.1289, + "step": 15896 + }, + { + "epoch": 0.13799359380560933, + "grad_norm": 0.330078125, + "learning_rate": 0.001892994889837409, + "loss": 0.0938, + "step": 15897 + }, + { + "epoch": 0.1380022742858135, + "grad_norm": 0.97265625, + "learning_rate": 0.001892980814146261, + "loss": 0.1396, + "step": 15898 + }, + { + "epoch": 0.13801095476601766, + "grad_norm": 0.2373046875, + "learning_rate": 0.0018929667375879211, + "loss": 0.1001, + "step": 15899 + }, + { + "epoch": 0.13801963524622182, + "grad_norm": 0.1376953125, + "learning_rate": 0.001892952660162405, + "loss": 0.1055, + "step": 15900 + }, + { + "epoch": 0.138028315726426, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018929385818697274, + "loss": 0.1152, + "step": 15901 + }, + { + "epoch": 0.13803699620663015, + "grad_norm": 0.15625, + "learning_rate": 0.0018929245027099043, + "loss": 0.1118, + "step": 15902 + }, + { + "epoch": 0.13804567668683432, + "grad_norm": 0.24609375, + "learning_rate": 0.0018929104226829507, + "loss": 0.0986, + "step": 15903 + }, + { + "epoch": 0.13805435716703848, + "grad_norm": 0.6953125, + "learning_rate": 0.0018928963417888822, + "loss": 0.1187, + "step": 15904 + }, + { + "epoch": 0.13806303764724265, + "grad_norm": 0.0849609375, + "learning_rate": 0.0018928822600277142, + "loss": 0.1055, + "step": 15905 + }, + { + "epoch": 0.1380717181274468, + "grad_norm": 0.2734375, + "learning_rate": 0.0018928681773994623, + "loss": 0.1162, + "step": 15906 + }, + { + "epoch": 0.13808039860765098, + "grad_norm": 0.36328125, + "learning_rate": 0.0018928540939041416, + "loss": 0.1113, + "step": 15907 + }, + { + "epoch": 0.13808907908785514, + "grad_norm": 0.3828125, + "learning_rate": 0.0018928400095417676, + "loss": 0.124, + "step": 15908 + }, + { + "epoch": 0.1380977595680593, + "grad_norm": 0.291015625, + "learning_rate": 0.0018928259243123553, + "loss": 0.1094, + "step": 15909 + }, + { + "epoch": 0.13810644004826347, + "grad_norm": 0.42578125, + "learning_rate": 0.0018928118382159208, + "loss": 0.0942, + "step": 15910 + }, + { + "epoch": 0.13811512052846764, + "grad_norm": 0.19140625, + "learning_rate": 0.0018927977512524792, + "loss": 0.127, + "step": 15911 + }, + { + "epoch": 0.1381238010086718, + "grad_norm": 0.69921875, + "learning_rate": 0.0018927836634220457, + "loss": 0.1177, + "step": 15912 + }, + { + "epoch": 0.13813248148887597, + "grad_norm": 0.09521484375, + "learning_rate": 0.001892769574724636, + "loss": 0.1152, + "step": 15913 + }, + { + "epoch": 0.13814116196908013, + "grad_norm": 1.1328125, + "learning_rate": 0.0018927554851602656, + "loss": 0.1094, + "step": 15914 + }, + { + "epoch": 0.1381498424492843, + "grad_norm": 0.490234375, + "learning_rate": 0.0018927413947289495, + "loss": 0.1387, + "step": 15915 + }, + { + "epoch": 0.13815852292948846, + "grad_norm": 0.5390625, + "learning_rate": 0.0018927273034307034, + "loss": 0.1143, + "step": 15916 + }, + { + "epoch": 0.13816720340969263, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018927132112655425, + "loss": 0.1094, + "step": 15917 + }, + { + "epoch": 0.1381758838898968, + "grad_norm": 0.251953125, + "learning_rate": 0.0018926991182334823, + "loss": 0.1074, + "step": 15918 + }, + { + "epoch": 0.13818456437010096, + "grad_norm": 0.1640625, + "learning_rate": 0.0018926850243345384, + "loss": 0.1001, + "step": 15919 + }, + { + "epoch": 0.13819324485030512, + "grad_norm": 0.640625, + "learning_rate": 0.0018926709295687262, + "loss": 0.1758, + "step": 15920 + }, + { + "epoch": 0.1382019253305093, + "grad_norm": 0.337890625, + "learning_rate": 0.0018926568339360608, + "loss": 0.123, + "step": 15921 + }, + { + "epoch": 0.13821060581071345, + "grad_norm": 0.50390625, + "learning_rate": 0.0018926427374365577, + "loss": 0.1094, + "step": 15922 + }, + { + "epoch": 0.13821928629091762, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018926286400702326, + "loss": 0.1147, + "step": 15923 + }, + { + "epoch": 0.13822796677112179, + "grad_norm": 0.494140625, + "learning_rate": 0.0018926145418371007, + "loss": 0.1094, + "step": 15924 + }, + { + "epoch": 0.13823664725132595, + "grad_norm": 0.166015625, + "learning_rate": 0.0018926004427371774, + "loss": 0.103, + "step": 15925 + }, + { + "epoch": 0.13824532773153012, + "grad_norm": 0.392578125, + "learning_rate": 0.0018925863427704782, + "loss": 0.1289, + "step": 15926 + }, + { + "epoch": 0.13825400821173428, + "grad_norm": 1.1328125, + "learning_rate": 0.0018925722419370185, + "loss": 0.1445, + "step": 15927 + }, + { + "epoch": 0.13826268869193845, + "grad_norm": 0.84765625, + "learning_rate": 0.0018925581402368134, + "loss": 0.0815, + "step": 15928 + }, + { + "epoch": 0.1382713691721426, + "grad_norm": 0.609375, + "learning_rate": 0.001892544037669879, + "loss": 0.1123, + "step": 15929 + }, + { + "epoch": 0.13828004965234678, + "grad_norm": 0.86328125, + "learning_rate": 0.0018925299342362303, + "loss": 0.1309, + "step": 15930 + }, + { + "epoch": 0.13828873013255094, + "grad_norm": 0.138671875, + "learning_rate": 0.0018925158299358826, + "loss": 0.0889, + "step": 15931 + }, + { + "epoch": 0.1382974106127551, + "grad_norm": 0.33203125, + "learning_rate": 0.001892501724768852, + "loss": 0.1426, + "step": 15932 + }, + { + "epoch": 0.13830609109295927, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018924876187351527, + "loss": 0.1177, + "step": 15933 + }, + { + "epoch": 0.13831477157316344, + "grad_norm": 0.41015625, + "learning_rate": 0.001892473511834801, + "loss": 0.1226, + "step": 15934 + }, + { + "epoch": 0.1383234520533676, + "grad_norm": 0.26953125, + "learning_rate": 0.0018924594040678124, + "loss": 0.0786, + "step": 15935 + }, + { + "epoch": 0.13833213253357177, + "grad_norm": 0.703125, + "learning_rate": 0.0018924452954342025, + "loss": 0.1045, + "step": 15936 + }, + { + "epoch": 0.13834081301377593, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018924311859339858, + "loss": 0.1074, + "step": 15937 + }, + { + "epoch": 0.1383494934939801, + "grad_norm": 0.375, + "learning_rate": 0.0018924170755671781, + "loss": 0.1172, + "step": 15938 + }, + { + "epoch": 0.13835817397418426, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018924029643337954, + "loss": 0.124, + "step": 15939 + }, + { + "epoch": 0.13836685445438843, + "grad_norm": 0.27734375, + "learning_rate": 0.0018923888522338523, + "loss": 0.1201, + "step": 15940 + }, + { + "epoch": 0.1383755349345926, + "grad_norm": 0.15625, + "learning_rate": 0.001892374739267365, + "loss": 0.0659, + "step": 15941 + }, + { + "epoch": 0.13838421541479676, + "grad_norm": 0.287109375, + "learning_rate": 0.0018923606254343487, + "loss": 0.1162, + "step": 15942 + }, + { + "epoch": 0.13839289589500092, + "grad_norm": 0.2080078125, + "learning_rate": 0.0018923465107348184, + "loss": 0.0762, + "step": 15943 + }, + { + "epoch": 0.1384015763752051, + "grad_norm": 0.150390625, + "learning_rate": 0.0018923323951687903, + "loss": 0.1157, + "step": 15944 + }, + { + "epoch": 0.13841025685540925, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018923182787362789, + "loss": 0.1069, + "step": 15945 + }, + { + "epoch": 0.13841893733561342, + "grad_norm": 0.1484375, + "learning_rate": 0.0018923041614373002, + "loss": 0.1396, + "step": 15946 + }, + { + "epoch": 0.13842761781581758, + "grad_norm": 0.40234375, + "learning_rate": 0.0018922900432718696, + "loss": 0.1064, + "step": 15947 + }, + { + "epoch": 0.13843629829602175, + "grad_norm": 0.38671875, + "learning_rate": 0.0018922759242400028, + "loss": 0.085, + "step": 15948 + }, + { + "epoch": 0.1384449787762259, + "grad_norm": 0.97265625, + "learning_rate": 0.0018922618043417149, + "loss": 0.124, + "step": 15949 + }, + { + "epoch": 0.13845365925643008, + "grad_norm": 0.361328125, + "learning_rate": 0.001892247683577021, + "loss": 0.1172, + "step": 15950 + }, + { + "epoch": 0.13846233973663424, + "grad_norm": 0.1044921875, + "learning_rate": 0.0018922335619459372, + "loss": 0.1025, + "step": 15951 + }, + { + "epoch": 0.1384710202168384, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018922194394484788, + "loss": 0.0977, + "step": 15952 + }, + { + "epoch": 0.13847970069704257, + "grad_norm": 0.3125, + "learning_rate": 0.001892205316084661, + "loss": 0.1289, + "step": 15953 + }, + { + "epoch": 0.13848838117724674, + "grad_norm": 0.33984375, + "learning_rate": 0.0018921911918544994, + "loss": 0.1221, + "step": 15954 + }, + { + "epoch": 0.1384970616574509, + "grad_norm": 0.275390625, + "learning_rate": 0.0018921770667580094, + "loss": 0.0776, + "step": 15955 + }, + { + "epoch": 0.13850574213765504, + "grad_norm": 0.38671875, + "learning_rate": 0.0018921629407952062, + "loss": 0.0977, + "step": 15956 + }, + { + "epoch": 0.1385144226178592, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018921488139661059, + "loss": 0.1934, + "step": 15957 + }, + { + "epoch": 0.13852310309806337, + "grad_norm": 0.216796875, + "learning_rate": 0.0018921346862707234, + "loss": 0.1123, + "step": 15958 + }, + { + "epoch": 0.13853178357826754, + "grad_norm": 0.08837890625, + "learning_rate": 0.0018921205577090746, + "loss": 0.0791, + "step": 15959 + }, + { + "epoch": 0.1385404640584717, + "grad_norm": 0.228515625, + "learning_rate": 0.0018921064282811741, + "loss": 0.1016, + "step": 15960 + }, + { + "epoch": 0.13854914453867587, + "grad_norm": 0.71484375, + "learning_rate": 0.0018920922979870382, + "loss": 0.1055, + "step": 15961 + }, + { + "epoch": 0.13855782501888003, + "grad_norm": 0.5859375, + "learning_rate": 0.0018920781668266822, + "loss": 0.0967, + "step": 15962 + }, + { + "epoch": 0.1385665054990842, + "grad_norm": 0.83203125, + "learning_rate": 0.0018920640348001213, + "loss": 0.1289, + "step": 15963 + }, + { + "epoch": 0.13857518597928836, + "grad_norm": 0.427734375, + "learning_rate": 0.001892049901907371, + "loss": 0.0884, + "step": 15964 + }, + { + "epoch": 0.13858386645949253, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018920357681484468, + "loss": 0.1445, + "step": 15965 + }, + { + "epoch": 0.1385925469396967, + "grad_norm": 0.5, + "learning_rate": 0.0018920216335233642, + "loss": 0.165, + "step": 15966 + }, + { + "epoch": 0.13860122741990086, + "grad_norm": 0.251953125, + "learning_rate": 0.0018920074980321388, + "loss": 0.1216, + "step": 15967 + }, + { + "epoch": 0.13860990790010502, + "grad_norm": 0.5625, + "learning_rate": 0.0018919933616747858, + "loss": 0.0869, + "step": 15968 + }, + { + "epoch": 0.1386185883803092, + "grad_norm": 0.140625, + "learning_rate": 0.0018919792244513208, + "loss": 0.1006, + "step": 15969 + }, + { + "epoch": 0.13862726886051335, + "grad_norm": 0.39453125, + "learning_rate": 0.0018919650863617591, + "loss": 0.0962, + "step": 15970 + }, + { + "epoch": 0.13863594934071752, + "grad_norm": 0.23828125, + "learning_rate": 0.0018919509474061163, + "loss": 0.085, + "step": 15971 + }, + { + "epoch": 0.13864462982092168, + "grad_norm": 0.3828125, + "learning_rate": 0.001891936807584408, + "loss": 0.1016, + "step": 15972 + }, + { + "epoch": 0.13865331030112585, + "grad_norm": 1.0, + "learning_rate": 0.0018919226668966493, + "loss": 0.104, + "step": 15973 + }, + { + "epoch": 0.13866199078133, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018919085253428561, + "loss": 0.1348, + "step": 15974 + }, + { + "epoch": 0.13867067126153418, + "grad_norm": 0.115234375, + "learning_rate": 0.0018918943829230434, + "loss": 0.1055, + "step": 15975 + }, + { + "epoch": 0.13867935174173834, + "grad_norm": 0.6484375, + "learning_rate": 0.0018918802396372272, + "loss": 0.127, + "step": 15976 + }, + { + "epoch": 0.1386880322219425, + "grad_norm": 0.1328125, + "learning_rate": 0.0018918660954854224, + "loss": 0.1099, + "step": 15977 + }, + { + "epoch": 0.13869671270214667, + "grad_norm": 0.361328125, + "learning_rate": 0.0018918519504676449, + "loss": 0.1611, + "step": 15978 + }, + { + "epoch": 0.13870539318235084, + "grad_norm": 1.0546875, + "learning_rate": 0.00189183780458391, + "loss": 0.1235, + "step": 15979 + }, + { + "epoch": 0.138714073662555, + "grad_norm": 0.416015625, + "learning_rate": 0.0018918236578342332, + "loss": 0.1147, + "step": 15980 + }, + { + "epoch": 0.13872275414275917, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018918095102186297, + "loss": 0.0947, + "step": 15981 + }, + { + "epoch": 0.13873143462296333, + "grad_norm": 0.63671875, + "learning_rate": 0.0018917953617371152, + "loss": 0.1289, + "step": 15982 + }, + { + "epoch": 0.1387401151031675, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018917812123897056, + "loss": 0.1162, + "step": 15983 + }, + { + "epoch": 0.13874879558337166, + "grad_norm": 0.091796875, + "learning_rate": 0.0018917670621764156, + "loss": 0.1025, + "step": 15984 + }, + { + "epoch": 0.13875747606357583, + "grad_norm": 0.2236328125, + "learning_rate": 0.001891752911097261, + "loss": 0.1279, + "step": 15985 + }, + { + "epoch": 0.13876615654378, + "grad_norm": 0.486328125, + "learning_rate": 0.0018917387591522578, + "loss": 0.1445, + "step": 15986 + }, + { + "epoch": 0.13877483702398416, + "grad_norm": 0.53125, + "learning_rate": 0.0018917246063414204, + "loss": 0.1348, + "step": 15987 + }, + { + "epoch": 0.13878351750418833, + "grad_norm": 0.373046875, + "learning_rate": 0.0018917104526647653, + "loss": 0.0889, + "step": 15988 + }, + { + "epoch": 0.1387921979843925, + "grad_norm": 0.65234375, + "learning_rate": 0.0018916962981223072, + "loss": 0.1289, + "step": 15989 + }, + { + "epoch": 0.13880087846459666, + "grad_norm": 0.14453125, + "learning_rate": 0.001891682142714062, + "loss": 0.1318, + "step": 15990 + }, + { + "epoch": 0.13880955894480082, + "grad_norm": 0.59765625, + "learning_rate": 0.0018916679864400454, + "loss": 0.1143, + "step": 15991 + }, + { + "epoch": 0.13881823942500499, + "grad_norm": 0.279296875, + "learning_rate": 0.0018916538293002722, + "loss": 0.1348, + "step": 15992 + }, + { + "epoch": 0.13882691990520915, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018916396712947586, + "loss": 0.0938, + "step": 15993 + }, + { + "epoch": 0.13883560038541332, + "grad_norm": 1.0625, + "learning_rate": 0.0018916255124235195, + "loss": 0.1367, + "step": 15994 + }, + { + "epoch": 0.13884428086561748, + "grad_norm": 0.54296875, + "learning_rate": 0.0018916113526865708, + "loss": 0.1094, + "step": 15995 + }, + { + "epoch": 0.13885296134582165, + "grad_norm": 1.0703125, + "learning_rate": 0.0018915971920839276, + "loss": 0.127, + "step": 15996 + }, + { + "epoch": 0.1388616418260258, + "grad_norm": 0.57421875, + "learning_rate": 0.0018915830306156058, + "loss": 0.1211, + "step": 15997 + }, + { + "epoch": 0.13887032230622998, + "grad_norm": 0.72265625, + "learning_rate": 0.0018915688682816206, + "loss": 0.0967, + "step": 15998 + }, + { + "epoch": 0.13887900278643414, + "grad_norm": 0.54296875, + "learning_rate": 0.001891554705081988, + "loss": 0.0938, + "step": 15999 + }, + { + "epoch": 0.1388876832666383, + "grad_norm": 0.1328125, + "learning_rate": 0.0018915405410167225, + "loss": 0.1094, + "step": 16000 + }, + { + "epoch": 0.13889636374684247, + "grad_norm": 0.458984375, + "learning_rate": 0.0018915263760858401, + "loss": 0.1582, + "step": 16001 + }, + { + "epoch": 0.13890504422704664, + "grad_norm": 0.212890625, + "learning_rate": 0.0018915122102893568, + "loss": 0.125, + "step": 16002 + }, + { + "epoch": 0.1389137247072508, + "grad_norm": 0.58203125, + "learning_rate": 0.0018914980436272873, + "loss": 0.1406, + "step": 16003 + }, + { + "epoch": 0.13892240518745497, + "grad_norm": 0.11279296875, + "learning_rate": 0.0018914838760996475, + "loss": 0.0864, + "step": 16004 + }, + { + "epoch": 0.13893108566765913, + "grad_norm": 0.271484375, + "learning_rate": 0.0018914697077064531, + "loss": 0.0869, + "step": 16005 + }, + { + "epoch": 0.1389397661478633, + "grad_norm": 0.251953125, + "learning_rate": 0.001891455538447719, + "loss": 0.125, + "step": 16006 + }, + { + "epoch": 0.13894844662806746, + "grad_norm": 0.169921875, + "learning_rate": 0.0018914413683234614, + "loss": 0.0801, + "step": 16007 + }, + { + "epoch": 0.13895712710827163, + "grad_norm": 0.0771484375, + "learning_rate": 0.0018914271973336947, + "loss": 0.1074, + "step": 16008 + }, + { + "epoch": 0.1389658075884758, + "grad_norm": 0.5625, + "learning_rate": 0.0018914130254784355, + "loss": 0.1699, + "step": 16009 + }, + { + "epoch": 0.13897448806867996, + "grad_norm": 0.451171875, + "learning_rate": 0.0018913988527576994, + "loss": 0.123, + "step": 16010 + }, + { + "epoch": 0.13898316854888412, + "grad_norm": 0.09423828125, + "learning_rate": 0.0018913846791715006, + "loss": 0.0894, + "step": 16011 + }, + { + "epoch": 0.1389918490290883, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018913705047198558, + "loss": 0.1191, + "step": 16012 + }, + { + "epoch": 0.13900052950929245, + "grad_norm": 0.3828125, + "learning_rate": 0.00189135632940278, + "loss": 0.1514, + "step": 16013 + }, + { + "epoch": 0.13900920998949662, + "grad_norm": 0.4140625, + "learning_rate": 0.0018913421532202887, + "loss": 0.0972, + "step": 16014 + }, + { + "epoch": 0.13901789046970078, + "grad_norm": 0.62890625, + "learning_rate": 0.001891327976172398, + "loss": 0.1826, + "step": 16015 + }, + { + "epoch": 0.13902657094990495, + "grad_norm": 0.279296875, + "learning_rate": 0.0018913137982591226, + "loss": 0.0967, + "step": 16016 + }, + { + "epoch": 0.1390352514301091, + "grad_norm": 0.08984375, + "learning_rate": 0.0018912996194804783, + "loss": 0.1084, + "step": 16017 + }, + { + "epoch": 0.13904393191031328, + "grad_norm": 0.1328125, + "learning_rate": 0.0018912854398364805, + "loss": 0.0889, + "step": 16018 + }, + { + "epoch": 0.13905261239051744, + "grad_norm": 0.07373046875, + "learning_rate": 0.001891271259327145, + "loss": 0.103, + "step": 16019 + }, + { + "epoch": 0.1390612928707216, + "grad_norm": 0.2421875, + "learning_rate": 0.001891257077952487, + "loss": 0.1348, + "step": 16020 + }, + { + "epoch": 0.13906997335092577, + "grad_norm": 0.275390625, + "learning_rate": 0.0018912428957125222, + "loss": 0.1094, + "step": 16021 + }, + { + "epoch": 0.13907865383112994, + "grad_norm": 0.3671875, + "learning_rate": 0.0018912287126072664, + "loss": 0.0762, + "step": 16022 + }, + { + "epoch": 0.1390873343113341, + "grad_norm": 0.32421875, + "learning_rate": 0.0018912145286367344, + "loss": 0.1182, + "step": 16023 + }, + { + "epoch": 0.13909601479153827, + "grad_norm": 0.333984375, + "learning_rate": 0.0018912003438009421, + "loss": 0.1045, + "step": 16024 + }, + { + "epoch": 0.13910469527174243, + "grad_norm": 0.578125, + "learning_rate": 0.0018911861580999053, + "loss": 0.0791, + "step": 16025 + }, + { + "epoch": 0.1391133757519466, + "grad_norm": 0.181640625, + "learning_rate": 0.0018911719715336388, + "loss": 0.1289, + "step": 16026 + }, + { + "epoch": 0.13912205623215076, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018911577841021587, + "loss": 0.1162, + "step": 16027 + }, + { + "epoch": 0.13913073671235493, + "grad_norm": 0.40234375, + "learning_rate": 0.0018911435958054806, + "loss": 0.1484, + "step": 16028 + }, + { + "epoch": 0.1391394171925591, + "grad_norm": 0.384765625, + "learning_rate": 0.001891129406643619, + "loss": 0.0898, + "step": 16029 + }, + { + "epoch": 0.13914809767276326, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018911152166165908, + "loss": 0.1167, + "step": 16030 + }, + { + "epoch": 0.13915677815296743, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018911010257244105, + "loss": 0.1162, + "step": 16031 + }, + { + "epoch": 0.1391654586331716, + "grad_norm": 0.078125, + "learning_rate": 0.0018910868339670945, + "loss": 0.0889, + "step": 16032 + }, + { + "epoch": 0.13917413911337576, + "grad_norm": 0.55078125, + "learning_rate": 0.0018910726413446577, + "loss": 0.1377, + "step": 16033 + }, + { + "epoch": 0.13918281959357992, + "grad_norm": 0.41796875, + "learning_rate": 0.0018910584478571155, + "loss": 0.0898, + "step": 16034 + }, + { + "epoch": 0.13919150007378409, + "grad_norm": 0.490234375, + "learning_rate": 0.0018910442535044836, + "loss": 0.1182, + "step": 16035 + }, + { + "epoch": 0.13920018055398825, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018910300582867779, + "loss": 0.125, + "step": 16036 + }, + { + "epoch": 0.13920886103419242, + "grad_norm": 0.25390625, + "learning_rate": 0.0018910158622040135, + "loss": 0.0698, + "step": 16037 + }, + { + "epoch": 0.13921754151439658, + "grad_norm": 0.5390625, + "learning_rate": 0.0018910016652562058, + "loss": 0.123, + "step": 16038 + }, + { + "epoch": 0.13922622199460075, + "grad_norm": 0.279296875, + "learning_rate": 0.001890987467443371, + "loss": 0.1016, + "step": 16039 + }, + { + "epoch": 0.1392349024748049, + "grad_norm": 0.294921875, + "learning_rate": 0.001890973268765524, + "loss": 0.104, + "step": 16040 + }, + { + "epoch": 0.13924358295500908, + "grad_norm": 0.189453125, + "learning_rate": 0.0018909590692226803, + "loss": 0.0938, + "step": 16041 + }, + { + "epoch": 0.13925226343521324, + "grad_norm": 0.65625, + "learning_rate": 0.0018909448688148559, + "loss": 0.1045, + "step": 16042 + }, + { + "epoch": 0.1392609439154174, + "grad_norm": 0.111328125, + "learning_rate": 0.001890930667542066, + "loss": 0.1289, + "step": 16043 + }, + { + "epoch": 0.13926962439562157, + "grad_norm": 0.1796875, + "learning_rate": 0.001890916465404326, + "loss": 0.1582, + "step": 16044 + }, + { + "epoch": 0.13927830487582574, + "grad_norm": 0.74609375, + "learning_rate": 0.001890902262401652, + "loss": 0.166, + "step": 16045 + }, + { + "epoch": 0.1392869853560299, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018908880585340592, + "loss": 0.0737, + "step": 16046 + }, + { + "epoch": 0.13929566583623407, + "grad_norm": 0.3515625, + "learning_rate": 0.0018908738538015628, + "loss": 0.1396, + "step": 16047 + }, + { + "epoch": 0.13930434631643823, + "grad_norm": 0.37109375, + "learning_rate": 0.0018908596482041786, + "loss": 0.1152, + "step": 16048 + }, + { + "epoch": 0.1393130267966424, + "grad_norm": 0.546875, + "learning_rate": 0.0018908454417419223, + "loss": 0.123, + "step": 16049 + }, + { + "epoch": 0.13932170727684656, + "grad_norm": 0.08837890625, + "learning_rate": 0.0018908312344148095, + "loss": 0.0996, + "step": 16050 + }, + { + "epoch": 0.13933038775705073, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018908170262228552, + "loss": 0.103, + "step": 16051 + }, + { + "epoch": 0.1393390682372549, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018908028171660753, + "loss": 0.0991, + "step": 16052 + }, + { + "epoch": 0.13934774871745906, + "grad_norm": 0.5390625, + "learning_rate": 0.0018907886072444858, + "loss": 0.1187, + "step": 16053 + }, + { + "epoch": 0.13935642919766322, + "grad_norm": 0.140625, + "learning_rate": 0.0018907743964581012, + "loss": 0.1279, + "step": 16054 + }, + { + "epoch": 0.1393651096778674, + "grad_norm": 0.14453125, + "learning_rate": 0.001890760184806938, + "loss": 0.1592, + "step": 16055 + }, + { + "epoch": 0.13937379015807155, + "grad_norm": 0.7421875, + "learning_rate": 0.001890745972291011, + "loss": 0.1338, + "step": 16056 + }, + { + "epoch": 0.13938247063827572, + "grad_norm": 0.72265625, + "learning_rate": 0.0018907317589103364, + "loss": 0.0874, + "step": 16057 + }, + { + "epoch": 0.13939115111847988, + "grad_norm": 0.390625, + "learning_rate": 0.0018907175446649291, + "loss": 0.0942, + "step": 16058 + }, + { + "epoch": 0.13939983159868405, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018907033295548052, + "loss": 0.0913, + "step": 16059 + }, + { + "epoch": 0.1394085120788882, + "grad_norm": 0.353515625, + "learning_rate": 0.00189068911357998, + "loss": 0.1348, + "step": 16060 + }, + { + "epoch": 0.13941719255909238, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018906748967404692, + "loss": 0.0938, + "step": 16061 + }, + { + "epoch": 0.13942587303929654, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018906606790362879, + "loss": 0.1006, + "step": 16062 + }, + { + "epoch": 0.1394345535195007, + "grad_norm": 0.294921875, + "learning_rate": 0.0018906464604674523, + "loss": 0.1475, + "step": 16063 + }, + { + "epoch": 0.13944323399970487, + "grad_norm": 0.50390625, + "learning_rate": 0.0018906322410339774, + "loss": 0.1094, + "step": 16064 + }, + { + "epoch": 0.13945191447990904, + "grad_norm": 0.267578125, + "learning_rate": 0.001890618020735879, + "loss": 0.1338, + "step": 16065 + }, + { + "epoch": 0.1394605949601132, + "grad_norm": 0.40234375, + "learning_rate": 0.0018906037995731727, + "loss": 0.1299, + "step": 16066 + }, + { + "epoch": 0.13946927544031737, + "grad_norm": 0.67578125, + "learning_rate": 0.001890589577545874, + "loss": 0.1162, + "step": 16067 + }, + { + "epoch": 0.13947795592052153, + "grad_norm": 0.4453125, + "learning_rate": 0.0018905753546539978, + "loss": 0.0991, + "step": 16068 + }, + { + "epoch": 0.1394866364007257, + "grad_norm": 0.1767578125, + "learning_rate": 0.001890561130897561, + "loss": 0.0986, + "step": 16069 + }, + { + "epoch": 0.13949531688092986, + "grad_norm": 0.18359375, + "learning_rate": 0.0018905469062765782, + "loss": 0.1504, + "step": 16070 + }, + { + "epoch": 0.13950399736113403, + "grad_norm": 0.1337890625, + "learning_rate": 0.001890532680791065, + "loss": 0.1118, + "step": 16071 + }, + { + "epoch": 0.1395126778413382, + "grad_norm": 0.72265625, + "learning_rate": 0.0018905184544410374, + "loss": 0.1465, + "step": 16072 + }, + { + "epoch": 0.13952135832154236, + "grad_norm": 0.224609375, + "learning_rate": 0.0018905042272265105, + "loss": 0.1279, + "step": 16073 + }, + { + "epoch": 0.13953003880174653, + "grad_norm": 0.380859375, + "learning_rate": 0.0018904899991475004, + "loss": 0.0776, + "step": 16074 + }, + { + "epoch": 0.1395387192819507, + "grad_norm": 0.09375, + "learning_rate": 0.0018904757702040218, + "loss": 0.1094, + "step": 16075 + }, + { + "epoch": 0.13954739976215486, + "grad_norm": 0.140625, + "learning_rate": 0.0018904615403960913, + "loss": 0.1055, + "step": 16076 + }, + { + "epoch": 0.13955608024235902, + "grad_norm": 0.89453125, + "learning_rate": 0.0018904473097237236, + "loss": 0.1729, + "step": 16077 + }, + { + "epoch": 0.13956476072256319, + "grad_norm": 0.09423828125, + "learning_rate": 0.0018904330781869345, + "loss": 0.0962, + "step": 16078 + }, + { + "epoch": 0.13957344120276732, + "grad_norm": 0.263671875, + "learning_rate": 0.00189041884578574, + "loss": 0.1299, + "step": 16079 + }, + { + "epoch": 0.1395821216829715, + "grad_norm": 0.546875, + "learning_rate": 0.0018904046125201553, + "loss": 0.1113, + "step": 16080 + }, + { + "epoch": 0.13959080216317565, + "grad_norm": 0.7109375, + "learning_rate": 0.0018903903783901957, + "loss": 0.0947, + "step": 16081 + }, + { + "epoch": 0.13959948264337982, + "grad_norm": 0.279296875, + "learning_rate": 0.0018903761433958772, + "loss": 0.1289, + "step": 16082 + }, + { + "epoch": 0.13960816312358398, + "grad_norm": 0.3828125, + "learning_rate": 0.0018903619075372154, + "loss": 0.0859, + "step": 16083 + }, + { + "epoch": 0.13961684360378815, + "grad_norm": 0.62890625, + "learning_rate": 0.0018903476708142253, + "loss": 0.1201, + "step": 16084 + }, + { + "epoch": 0.13962552408399231, + "grad_norm": 0.1494140625, + "learning_rate": 0.001890333433226923, + "loss": 0.1211, + "step": 16085 + }, + { + "epoch": 0.13963420456419648, + "grad_norm": 0.197265625, + "learning_rate": 0.0018903191947753242, + "loss": 0.1279, + "step": 16086 + }, + { + "epoch": 0.13964288504440064, + "grad_norm": 0.462890625, + "learning_rate": 0.001890304955459444, + "loss": 0.0981, + "step": 16087 + }, + { + "epoch": 0.1396515655246048, + "grad_norm": 0.1572265625, + "learning_rate": 0.001890290715279298, + "loss": 0.1299, + "step": 16088 + }, + { + "epoch": 0.13966024600480897, + "grad_norm": 0.150390625, + "learning_rate": 0.0018902764742349024, + "loss": 0.1196, + "step": 16089 + }, + { + "epoch": 0.13966892648501314, + "grad_norm": 0.259765625, + "learning_rate": 0.001890262232326272, + "loss": 0.0967, + "step": 16090 + }, + { + "epoch": 0.1396776069652173, + "grad_norm": 0.388671875, + "learning_rate": 0.0018902479895534228, + "loss": 0.0908, + "step": 16091 + }, + { + "epoch": 0.13968628744542147, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018902337459163704, + "loss": 0.0732, + "step": 16092 + }, + { + "epoch": 0.13969496792562563, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018902195014151297, + "loss": 0.1172, + "step": 16093 + }, + { + "epoch": 0.1397036484058298, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018902052560497173, + "loss": 0.1074, + "step": 16094 + }, + { + "epoch": 0.13971232888603397, + "grad_norm": 0.828125, + "learning_rate": 0.0018901910098201484, + "loss": 0.1709, + "step": 16095 + }, + { + "epoch": 0.13972100936623813, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018901767627264384, + "loss": 0.0977, + "step": 16096 + }, + { + "epoch": 0.1397296898464423, + "grad_norm": 0.2109375, + "learning_rate": 0.0018901625147686029, + "loss": 0.0835, + "step": 16097 + }, + { + "epoch": 0.13973837032664646, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018901482659466574, + "loss": 0.0928, + "step": 16098 + }, + { + "epoch": 0.13974705080685063, + "grad_norm": 0.142578125, + "learning_rate": 0.0018901340162606178, + "loss": 0.1367, + "step": 16099 + }, + { + "epoch": 0.1397557312870548, + "grad_norm": 0.333984375, + "learning_rate": 0.0018901197657104993, + "loss": 0.1025, + "step": 16100 + }, + { + "epoch": 0.13976441176725896, + "grad_norm": 0.64453125, + "learning_rate": 0.001890105514296318, + "loss": 0.2041, + "step": 16101 + }, + { + "epoch": 0.13977309224746312, + "grad_norm": 0.609375, + "learning_rate": 0.001890091262018089, + "loss": 0.1216, + "step": 16102 + }, + { + "epoch": 0.13978177272766729, + "grad_norm": 0.09130859375, + "learning_rate": 0.0018900770088758278, + "loss": 0.0918, + "step": 16103 + }, + { + "epoch": 0.13979045320787145, + "grad_norm": 0.2314453125, + "learning_rate": 0.0018900627548695506, + "loss": 0.1021, + "step": 16104 + }, + { + "epoch": 0.13979913368807562, + "grad_norm": 0.7421875, + "learning_rate": 0.0018900484999992727, + "loss": 0.1279, + "step": 16105 + }, + { + "epoch": 0.13980781416827978, + "grad_norm": 0.166015625, + "learning_rate": 0.0018900342442650095, + "loss": 0.1289, + "step": 16106 + }, + { + "epoch": 0.13981649464848395, + "grad_norm": 0.68359375, + "learning_rate": 0.0018900199876667769, + "loss": 0.1094, + "step": 16107 + }, + { + "epoch": 0.1398251751286881, + "grad_norm": 0.375, + "learning_rate": 0.00189000573020459, + "loss": 0.1729, + "step": 16108 + }, + { + "epoch": 0.13983385560889228, + "grad_norm": 0.16796875, + "learning_rate": 0.001889991471878465, + "loss": 0.0972, + "step": 16109 + }, + { + "epoch": 0.13984253608909644, + "grad_norm": 0.640625, + "learning_rate": 0.001889977212688417, + "loss": 0.1216, + "step": 16110 + }, + { + "epoch": 0.1398512165693006, + "grad_norm": 0.189453125, + "learning_rate": 0.001889962952634462, + "loss": 0.1104, + "step": 16111 + }, + { + "epoch": 0.13985989704950477, + "grad_norm": 0.51953125, + "learning_rate": 0.001889948691716615, + "loss": 0.0771, + "step": 16112 + }, + { + "epoch": 0.13986857752970894, + "grad_norm": 0.62890625, + "learning_rate": 0.0018899344299348922, + "loss": 0.1138, + "step": 16113 + }, + { + "epoch": 0.1398772580099131, + "grad_norm": 0.08203125, + "learning_rate": 0.0018899201672893092, + "loss": 0.1157, + "step": 16114 + }, + { + "epoch": 0.13988593849011727, + "grad_norm": 0.107421875, + "learning_rate": 0.001889905903779881, + "loss": 0.1177, + "step": 16115 + }, + { + "epoch": 0.13989461897032143, + "grad_norm": 0.451171875, + "learning_rate": 0.0018898916394066237, + "loss": 0.1074, + "step": 16116 + }, + { + "epoch": 0.1399032994505256, + "grad_norm": 0.39453125, + "learning_rate": 0.001889877374169553, + "loss": 0.1318, + "step": 16117 + }, + { + "epoch": 0.13991197993072976, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018898631080686839, + "loss": 0.0908, + "step": 16118 + }, + { + "epoch": 0.13992066041093393, + "grad_norm": 0.263671875, + "learning_rate": 0.0018898488411040327, + "loss": 0.1074, + "step": 16119 + }, + { + "epoch": 0.1399293408911381, + "grad_norm": 0.30078125, + "learning_rate": 0.0018898345732756144, + "loss": 0.1934, + "step": 16120 + }, + { + "epoch": 0.13993802137134226, + "grad_norm": 0.470703125, + "learning_rate": 0.001889820304583445, + "loss": 0.1045, + "step": 16121 + }, + { + "epoch": 0.13994670185154642, + "grad_norm": 0.28515625, + "learning_rate": 0.00188980603502754, + "loss": 0.1309, + "step": 16122 + }, + { + "epoch": 0.1399553823317506, + "grad_norm": 0.205078125, + "learning_rate": 0.0018897917646079148, + "loss": 0.1006, + "step": 16123 + }, + { + "epoch": 0.13996406281195475, + "grad_norm": 0.2734375, + "learning_rate": 0.0018897774933245852, + "loss": 0.0986, + "step": 16124 + }, + { + "epoch": 0.13997274329215892, + "grad_norm": 0.8203125, + "learning_rate": 0.001889763221177567, + "loss": 0.1113, + "step": 16125 + }, + { + "epoch": 0.13998142377236308, + "grad_norm": 0.76171875, + "learning_rate": 0.0018897489481668756, + "loss": 0.1396, + "step": 16126 + }, + { + "epoch": 0.13999010425256725, + "grad_norm": 0.28125, + "learning_rate": 0.001889734674292527, + "loss": 0.0918, + "step": 16127 + }, + { + "epoch": 0.13999878473277141, + "grad_norm": 0.259765625, + "learning_rate": 0.0018897203995545356, + "loss": 0.085, + "step": 16128 + }, + { + "epoch": 0.14000746521297558, + "grad_norm": 0.19140625, + "learning_rate": 0.0018897061239529186, + "loss": 0.1387, + "step": 16129 + }, + { + "epoch": 0.14001614569317974, + "grad_norm": 0.35546875, + "learning_rate": 0.0018896918474876905, + "loss": 0.0908, + "step": 16130 + }, + { + "epoch": 0.1400248261733839, + "grad_norm": 0.12109375, + "learning_rate": 0.0018896775701588674, + "loss": 0.1221, + "step": 16131 + }, + { + "epoch": 0.14003350665358807, + "grad_norm": 0.322265625, + "learning_rate": 0.0018896632919664644, + "loss": 0.1016, + "step": 16132 + }, + { + "epoch": 0.14004218713379224, + "grad_norm": 0.205078125, + "learning_rate": 0.0018896490129104975, + "loss": 0.1123, + "step": 16133 + }, + { + "epoch": 0.1400508676139964, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018896347329909826, + "loss": 0.1162, + "step": 16134 + }, + { + "epoch": 0.14005954809420057, + "grad_norm": 0.25390625, + "learning_rate": 0.001889620452207935, + "loss": 0.1279, + "step": 16135 + }, + { + "epoch": 0.14006822857440474, + "grad_norm": 0.26953125, + "learning_rate": 0.0018896061705613702, + "loss": 0.1094, + "step": 16136 + }, + { + "epoch": 0.1400769090546089, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018895918880513043, + "loss": 0.1416, + "step": 16137 + }, + { + "epoch": 0.14008558953481307, + "grad_norm": 0.333984375, + "learning_rate": 0.001889577604677752, + "loss": 0.1089, + "step": 16138 + }, + { + "epoch": 0.14009427001501723, + "grad_norm": 0.53125, + "learning_rate": 0.0018895633204407299, + "loss": 0.1191, + "step": 16139 + }, + { + "epoch": 0.1401029504952214, + "grad_norm": 0.1123046875, + "learning_rate": 0.001889549035340253, + "loss": 0.1514, + "step": 16140 + }, + { + "epoch": 0.14011163097542556, + "grad_norm": 0.44921875, + "learning_rate": 0.0018895347493763372, + "loss": 0.0918, + "step": 16141 + }, + { + "epoch": 0.14012031145562973, + "grad_norm": 0.126953125, + "learning_rate": 0.0018895204625489982, + "loss": 0.1152, + "step": 16142 + }, + { + "epoch": 0.1401289919358339, + "grad_norm": 0.330078125, + "learning_rate": 0.0018895061748582514, + "loss": 0.0767, + "step": 16143 + }, + { + "epoch": 0.14013767241603806, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018894918863041127, + "loss": 0.1133, + "step": 16144 + }, + { + "epoch": 0.14014635289624222, + "grad_norm": 0.353515625, + "learning_rate": 0.001889477596886597, + "loss": 0.1455, + "step": 16145 + }, + { + "epoch": 0.1401550333764464, + "grad_norm": 0.15625, + "learning_rate": 0.001889463306605721, + "loss": 0.1108, + "step": 16146 + }, + { + "epoch": 0.14016371385665055, + "grad_norm": 0.25, + "learning_rate": 0.0018894490154614996, + "loss": 0.1221, + "step": 16147 + }, + { + "epoch": 0.14017239433685472, + "grad_norm": 0.99609375, + "learning_rate": 0.0018894347234539486, + "loss": 0.1348, + "step": 16148 + }, + { + "epoch": 0.14018107481705888, + "grad_norm": 0.322265625, + "learning_rate": 0.0018894204305830837, + "loss": 0.1128, + "step": 16149 + }, + { + "epoch": 0.14018975529726305, + "grad_norm": 0.482421875, + "learning_rate": 0.0018894061368489204, + "loss": 0.1436, + "step": 16150 + }, + { + "epoch": 0.1401984357774672, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018893918422514745, + "loss": 0.106, + "step": 16151 + }, + { + "epoch": 0.14020711625767138, + "grad_norm": 0.224609375, + "learning_rate": 0.0018893775467907616, + "loss": 0.1445, + "step": 16152 + }, + { + "epoch": 0.14021579673787554, + "grad_norm": 0.40625, + "learning_rate": 0.0018893632504667973, + "loss": 0.1885, + "step": 16153 + }, + { + "epoch": 0.1402244772180797, + "grad_norm": 0.28125, + "learning_rate": 0.0018893489532795969, + "loss": 0.1182, + "step": 16154 + }, + { + "epoch": 0.14023315769828387, + "grad_norm": 0.498046875, + "learning_rate": 0.0018893346552291766, + "loss": 0.1094, + "step": 16155 + }, + { + "epoch": 0.14024183817848804, + "grad_norm": 0.3984375, + "learning_rate": 0.0018893203563155514, + "loss": 0.0972, + "step": 16156 + }, + { + "epoch": 0.1402505186586922, + "grad_norm": 0.69921875, + "learning_rate": 0.0018893060565387378, + "loss": 0.0957, + "step": 16157 + }, + { + "epoch": 0.14025919913889637, + "grad_norm": 0.275390625, + "learning_rate": 0.001889291755898751, + "loss": 0.1211, + "step": 16158 + }, + { + "epoch": 0.14026787961910053, + "grad_norm": 0.8984375, + "learning_rate": 0.001889277454395606, + "loss": 0.1494, + "step": 16159 + }, + { + "epoch": 0.1402765600993047, + "grad_norm": 0.2109375, + "learning_rate": 0.0018892631520293194, + "loss": 0.1113, + "step": 16160 + }, + { + "epoch": 0.14028524057950886, + "grad_norm": 0.4765625, + "learning_rate": 0.0018892488487999067, + "loss": 0.1787, + "step": 16161 + }, + { + "epoch": 0.14029392105971303, + "grad_norm": 0.423828125, + "learning_rate": 0.0018892345447073832, + "loss": 0.1367, + "step": 16162 + }, + { + "epoch": 0.1403026015399172, + "grad_norm": 0.67578125, + "learning_rate": 0.0018892202397517643, + "loss": 0.1064, + "step": 16163 + }, + { + "epoch": 0.14031128202012136, + "grad_norm": 0.3984375, + "learning_rate": 0.0018892059339330662, + "loss": 0.0986, + "step": 16164 + }, + { + "epoch": 0.14031996250032552, + "grad_norm": 0.375, + "learning_rate": 0.0018891916272513044, + "loss": 0.1035, + "step": 16165 + }, + { + "epoch": 0.1403286429805297, + "grad_norm": 0.25390625, + "learning_rate": 0.0018891773197064947, + "loss": 0.0977, + "step": 16166 + }, + { + "epoch": 0.14033732346073385, + "grad_norm": 0.53125, + "learning_rate": 0.0018891630112986522, + "loss": 0.0811, + "step": 16167 + }, + { + "epoch": 0.14034600394093802, + "grad_norm": 0.10302734375, + "learning_rate": 0.001889148702027793, + "loss": 0.1084, + "step": 16168 + }, + { + "epoch": 0.14035468442114218, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018891343918939327, + "loss": 0.1113, + "step": 16169 + }, + { + "epoch": 0.14036336490134635, + "grad_norm": 0.30859375, + "learning_rate": 0.001889120080897087, + "loss": 0.1191, + "step": 16170 + }, + { + "epoch": 0.14037204538155051, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018891057690372712, + "loss": 0.1216, + "step": 16171 + }, + { + "epoch": 0.14038072586175468, + "grad_norm": 0.140625, + "learning_rate": 0.0018890914563145015, + "loss": 0.0977, + "step": 16172 + }, + { + "epoch": 0.14038940634195884, + "grad_norm": 0.462890625, + "learning_rate": 0.0018890771427287926, + "loss": 0.1699, + "step": 16173 + }, + { + "epoch": 0.140398086822163, + "grad_norm": 1.2109375, + "learning_rate": 0.0018890628282801616, + "loss": 0.1465, + "step": 16174 + }, + { + "epoch": 0.14040676730236717, + "grad_norm": 0.126953125, + "learning_rate": 0.001889048512968623, + "loss": 0.1621, + "step": 16175 + }, + { + "epoch": 0.14041544778257134, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018890341967941926, + "loss": 0.0869, + "step": 16176 + }, + { + "epoch": 0.1404241282627755, + "grad_norm": 0.62109375, + "learning_rate": 0.0018890198797568863, + "loss": 0.1338, + "step": 16177 + }, + { + "epoch": 0.14043280874297967, + "grad_norm": 0.314453125, + "learning_rate": 0.0018890055618567198, + "loss": 0.1045, + "step": 16178 + }, + { + "epoch": 0.14044148922318384, + "grad_norm": 0.390625, + "learning_rate": 0.0018889912430937088, + "loss": 0.085, + "step": 16179 + }, + { + "epoch": 0.140450169703388, + "grad_norm": 0.474609375, + "learning_rate": 0.0018889769234678687, + "loss": 0.0693, + "step": 16180 + }, + { + "epoch": 0.14045885018359217, + "grad_norm": 0.466796875, + "learning_rate": 0.0018889626029792153, + "loss": 0.375, + "step": 16181 + }, + { + "epoch": 0.14046753066379633, + "grad_norm": 0.4453125, + "learning_rate": 0.0018889482816277644, + "loss": 0.1221, + "step": 16182 + }, + { + "epoch": 0.1404762111440005, + "grad_norm": 4.65625, + "learning_rate": 0.0018889339594135312, + "loss": 0.1387, + "step": 16183 + }, + { + "epoch": 0.14048489162420466, + "grad_norm": 0.68359375, + "learning_rate": 0.001888919636336532, + "loss": 0.1543, + "step": 16184 + }, + { + "epoch": 0.14049357210440883, + "grad_norm": 0.35546875, + "learning_rate": 0.001888905312396782, + "loss": 0.1133, + "step": 16185 + }, + { + "epoch": 0.140502252584613, + "grad_norm": 0.1240234375, + "learning_rate": 0.001888890987594297, + "loss": 0.0806, + "step": 16186 + }, + { + "epoch": 0.14051093306481716, + "grad_norm": 0.357421875, + "learning_rate": 0.0018888766619290928, + "loss": 0.1943, + "step": 16187 + }, + { + "epoch": 0.14051961354502132, + "grad_norm": 1.0703125, + "learning_rate": 0.0018888623354011847, + "loss": 0.1465, + "step": 16188 + }, + { + "epoch": 0.1405282940252255, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018888480080105888, + "loss": 0.0713, + "step": 16189 + }, + { + "epoch": 0.14053697450542965, + "grad_norm": 0.302734375, + "learning_rate": 0.0018888336797573204, + "loss": 0.0884, + "step": 16190 + }, + { + "epoch": 0.14054565498563382, + "grad_norm": 0.166015625, + "learning_rate": 0.0018888193506413954, + "loss": 0.1113, + "step": 16191 + }, + { + "epoch": 0.14055433546583798, + "grad_norm": 0.63671875, + "learning_rate": 0.0018888050206628294, + "loss": 0.0986, + "step": 16192 + }, + { + "epoch": 0.14056301594604215, + "grad_norm": 0.25390625, + "learning_rate": 0.0018887906898216382, + "loss": 0.1406, + "step": 16193 + }, + { + "epoch": 0.1405716964262463, + "grad_norm": 0.2431640625, + "learning_rate": 0.0018887763581178375, + "loss": 0.1182, + "step": 16194 + }, + { + "epoch": 0.14058037690645048, + "grad_norm": 0.2578125, + "learning_rate": 0.0018887620255514425, + "loss": 0.1133, + "step": 16195 + }, + { + "epoch": 0.14058905738665464, + "grad_norm": 0.11572265625, + "learning_rate": 0.001888747692122469, + "loss": 0.0972, + "step": 16196 + }, + { + "epoch": 0.1405977378668588, + "grad_norm": 0.52734375, + "learning_rate": 0.0018887333578309334, + "loss": 0.103, + "step": 16197 + }, + { + "epoch": 0.14060641834706297, + "grad_norm": 0.37109375, + "learning_rate": 0.0018887190226768505, + "loss": 0.1191, + "step": 16198 + }, + { + "epoch": 0.14061509882726714, + "grad_norm": 0.06982421875, + "learning_rate": 0.0018887046866602366, + "loss": 0.0806, + "step": 16199 + }, + { + "epoch": 0.1406237793074713, + "grad_norm": 0.66015625, + "learning_rate": 0.0018886903497811068, + "loss": 0.1357, + "step": 16200 + }, + { + "epoch": 0.14063245978767547, + "grad_norm": 0.31640625, + "learning_rate": 0.0018886760120394772, + "loss": 0.1572, + "step": 16201 + }, + { + "epoch": 0.14064114026787963, + "grad_norm": 0.2392578125, + "learning_rate": 0.0018886616734353635, + "loss": 0.1152, + "step": 16202 + }, + { + "epoch": 0.14064982074808377, + "grad_norm": 0.279296875, + "learning_rate": 0.001888647333968781, + "loss": 0.1602, + "step": 16203 + }, + { + "epoch": 0.14065850122828794, + "grad_norm": 0.41015625, + "learning_rate": 0.001888632993639746, + "loss": 0.1147, + "step": 16204 + }, + { + "epoch": 0.1406671817084921, + "grad_norm": 0.142578125, + "learning_rate": 0.0018886186524482735, + "loss": 0.1523, + "step": 16205 + }, + { + "epoch": 0.14067586218869627, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018886043103943796, + "loss": 0.0918, + "step": 16206 + }, + { + "epoch": 0.14068454266890043, + "grad_norm": 0.314453125, + "learning_rate": 0.0018885899674780796, + "loss": 0.0859, + "step": 16207 + }, + { + "epoch": 0.1406932231491046, + "grad_norm": 0.59765625, + "learning_rate": 0.00188857562369939, + "loss": 0.0859, + "step": 16208 + }, + { + "epoch": 0.14070190362930876, + "grad_norm": 0.30859375, + "learning_rate": 0.0018885612790583255, + "loss": 0.1348, + "step": 16209 + }, + { + "epoch": 0.14071058410951293, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018885469335549023, + "loss": 0.0967, + "step": 16210 + }, + { + "epoch": 0.1407192645897171, + "grad_norm": 0.130859375, + "learning_rate": 0.0018885325871891361, + "loss": 0.1182, + "step": 16211 + }, + { + "epoch": 0.14072794506992126, + "grad_norm": 0.09765625, + "learning_rate": 0.0018885182399610426, + "loss": 0.1445, + "step": 16212 + }, + { + "epoch": 0.14073662555012542, + "grad_norm": 0.146484375, + "learning_rate": 0.0018885038918706374, + "loss": 0.1221, + "step": 16213 + }, + { + "epoch": 0.1407453060303296, + "grad_norm": 0.462890625, + "learning_rate": 0.0018884895429179358, + "loss": 0.126, + "step": 16214 + }, + { + "epoch": 0.14075398651053375, + "grad_norm": 0.703125, + "learning_rate": 0.0018884751931029546, + "loss": 0.1162, + "step": 16215 + }, + { + "epoch": 0.14076266699073792, + "grad_norm": 0.6171875, + "learning_rate": 0.0018884608424257082, + "loss": 0.1172, + "step": 16216 + }, + { + "epoch": 0.14077134747094208, + "grad_norm": 0.2734375, + "learning_rate": 0.0018884464908862132, + "loss": 0.0991, + "step": 16217 + }, + { + "epoch": 0.14078002795114625, + "grad_norm": 0.62109375, + "learning_rate": 0.0018884321384844847, + "loss": 0.1162, + "step": 16218 + }, + { + "epoch": 0.1407887084313504, + "grad_norm": 0.185546875, + "learning_rate": 0.0018884177852205386, + "loss": 0.1201, + "step": 16219 + }, + { + "epoch": 0.14079738891155458, + "grad_norm": 0.5703125, + "learning_rate": 0.001888403431094391, + "loss": 0.124, + "step": 16220 + }, + { + "epoch": 0.14080606939175874, + "grad_norm": 1.453125, + "learning_rate": 0.0018883890761060569, + "loss": 0.1152, + "step": 16221 + }, + { + "epoch": 0.1408147498719629, + "grad_norm": 0.50390625, + "learning_rate": 0.0018883747202555526, + "loss": 0.1177, + "step": 16222 + }, + { + "epoch": 0.14082343035216707, + "grad_norm": 0.3125, + "learning_rate": 0.0018883603635428934, + "loss": 0.1099, + "step": 16223 + }, + { + "epoch": 0.14083211083237124, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018883460059680953, + "loss": 0.1348, + "step": 16224 + }, + { + "epoch": 0.1408407913125754, + "grad_norm": 0.11083984375, + "learning_rate": 0.001888331647531174, + "loss": 0.1245, + "step": 16225 + }, + { + "epoch": 0.14084947179277957, + "grad_norm": 0.333984375, + "learning_rate": 0.0018883172882321443, + "loss": 0.0918, + "step": 16226 + }, + { + "epoch": 0.14085815227298373, + "grad_norm": 0.359375, + "learning_rate": 0.0018883029280710234, + "loss": 0.1006, + "step": 16227 + }, + { + "epoch": 0.1408668327531879, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018882885670478261, + "loss": 0.123, + "step": 16228 + }, + { + "epoch": 0.14087551323339206, + "grad_norm": 0.1328125, + "learning_rate": 0.001888274205162568, + "loss": 0.1104, + "step": 16229 + }, + { + "epoch": 0.14088419371359623, + "grad_norm": 0.203125, + "learning_rate": 0.0018882598424152654, + "loss": 0.1777, + "step": 16230 + }, + { + "epoch": 0.1408928741938004, + "grad_norm": 0.48046875, + "learning_rate": 0.0018882454788059335, + "loss": 0.1621, + "step": 16231 + }, + { + "epoch": 0.14090155467400456, + "grad_norm": 0.45703125, + "learning_rate": 0.0018882311143345884, + "loss": 0.1299, + "step": 16232 + }, + { + "epoch": 0.14091023515420872, + "grad_norm": 0.5078125, + "learning_rate": 0.0018882167490012454, + "loss": 0.0913, + "step": 16233 + }, + { + "epoch": 0.1409189156344129, + "grad_norm": 0.490234375, + "learning_rate": 0.0018882023828059205, + "loss": 0.1426, + "step": 16234 + }, + { + "epoch": 0.14092759611461705, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018881880157486292, + "loss": 0.2129, + "step": 16235 + }, + { + "epoch": 0.14093627659482122, + "grad_norm": 0.435546875, + "learning_rate": 0.0018881736478293873, + "loss": 0.0952, + "step": 16236 + }, + { + "epoch": 0.14094495707502538, + "grad_norm": 0.1025390625, + "learning_rate": 0.001888159279048211, + "loss": 0.1133, + "step": 16237 + }, + { + "epoch": 0.14095363755522955, + "grad_norm": 2.328125, + "learning_rate": 0.001888144909405115, + "loss": 0.1992, + "step": 16238 + }, + { + "epoch": 0.14096231803543371, + "grad_norm": 0.08984375, + "learning_rate": 0.001888130538900116, + "loss": 0.1172, + "step": 16239 + }, + { + "epoch": 0.14097099851563788, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018881161675332288, + "loss": 0.1064, + "step": 16240 + }, + { + "epoch": 0.14097967899584204, + "grad_norm": 0.107421875, + "learning_rate": 0.00188810179530447, + "loss": 0.1289, + "step": 16241 + }, + { + "epoch": 0.1409883594760462, + "grad_norm": 0.27734375, + "learning_rate": 0.0018880874222138547, + "loss": 0.1396, + "step": 16242 + }, + { + "epoch": 0.14099703995625038, + "grad_norm": 1.0390625, + "learning_rate": 0.0018880730482613987, + "loss": 0.0718, + "step": 16243 + }, + { + "epoch": 0.14100572043645454, + "grad_norm": 0.37890625, + "learning_rate": 0.0018880586734471181, + "loss": 0.1221, + "step": 16244 + }, + { + "epoch": 0.1410144009166587, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018880442977710283, + "loss": 0.1055, + "step": 16245 + }, + { + "epoch": 0.14102308139686287, + "grad_norm": 0.66015625, + "learning_rate": 0.0018880299212331454, + "loss": 0.125, + "step": 16246 + }, + { + "epoch": 0.14103176187706704, + "grad_norm": 0.212890625, + "learning_rate": 0.0018880155438334846, + "loss": 0.1152, + "step": 16247 + }, + { + "epoch": 0.1410404423572712, + "grad_norm": 0.56640625, + "learning_rate": 0.0018880011655720615, + "loss": 0.1172, + "step": 16248 + }, + { + "epoch": 0.14104912283747537, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018879867864488925, + "loss": 0.0874, + "step": 16249 + }, + { + "epoch": 0.14105780331767953, + "grad_norm": 0.298828125, + "learning_rate": 0.001887972406463993, + "loss": 0.1162, + "step": 16250 + }, + { + "epoch": 0.1410664837978837, + "grad_norm": 0.77734375, + "learning_rate": 0.0018879580256173788, + "loss": 0.0972, + "step": 16251 + }, + { + "epoch": 0.14107516427808786, + "grad_norm": 0.6875, + "learning_rate": 0.0018879436439090654, + "loss": 0.1035, + "step": 16252 + }, + { + "epoch": 0.14108384475829203, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018879292613390684, + "loss": 0.0864, + "step": 16253 + }, + { + "epoch": 0.1410925252384962, + "grad_norm": 0.25390625, + "learning_rate": 0.0018879148779074042, + "loss": 0.0879, + "step": 16254 + }, + { + "epoch": 0.14110120571870036, + "grad_norm": 0.5078125, + "learning_rate": 0.0018879004936140878, + "loss": 0.1104, + "step": 16255 + }, + { + "epoch": 0.14110988619890452, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018878861084591356, + "loss": 0.1543, + "step": 16256 + }, + { + "epoch": 0.1411185666791087, + "grad_norm": 0.47265625, + "learning_rate": 0.001887871722442563, + "loss": 0.1143, + "step": 16257 + }, + { + "epoch": 0.14112724715931285, + "grad_norm": 0.609375, + "learning_rate": 0.0018878573355643855, + "loss": 0.1162, + "step": 16258 + }, + { + "epoch": 0.14113592763951702, + "grad_norm": 0.1005859375, + "learning_rate": 0.0018878429478246192, + "loss": 0.1128, + "step": 16259 + }, + { + "epoch": 0.14114460811972118, + "grad_norm": 0.09912109375, + "learning_rate": 0.0018878285592232795, + "loss": 0.1504, + "step": 16260 + }, + { + "epoch": 0.14115328859992535, + "grad_norm": 0.1171875, + "learning_rate": 0.0018878141697603824, + "loss": 0.1123, + "step": 16261 + }, + { + "epoch": 0.1411619690801295, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018877997794359439, + "loss": 0.0942, + "step": 16262 + }, + { + "epoch": 0.14117064956033368, + "grad_norm": 1.1640625, + "learning_rate": 0.001887785388249979, + "loss": 0.1621, + "step": 16263 + }, + { + "epoch": 0.14117933004053784, + "grad_norm": 0.51171875, + "learning_rate": 0.0018877709962025037, + "loss": 0.1445, + "step": 16264 + }, + { + "epoch": 0.141188010520742, + "grad_norm": 0.56640625, + "learning_rate": 0.001887756603293534, + "loss": 0.1211, + "step": 16265 + }, + { + "epoch": 0.14119669100094617, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018877422095230857, + "loss": 0.1318, + "step": 16266 + }, + { + "epoch": 0.14120537148115034, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018877278148911746, + "loss": 0.1079, + "step": 16267 + }, + { + "epoch": 0.1412140519613545, + "grad_norm": 1.2734375, + "learning_rate": 0.0018877134193978156, + "loss": 0.0918, + "step": 16268 + }, + { + "epoch": 0.14122273244155867, + "grad_norm": 0.173828125, + "learning_rate": 0.0018876990230430252, + "loss": 0.1006, + "step": 16269 + }, + { + "epoch": 0.14123141292176283, + "grad_norm": 0.201171875, + "learning_rate": 0.0018876846258268191, + "loss": 0.1035, + "step": 16270 + }, + { + "epoch": 0.141240093401967, + "grad_norm": 0.69140625, + "learning_rate": 0.001887670227749213, + "loss": 0.1143, + "step": 16271 + }, + { + "epoch": 0.14124877388217116, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018876558288102227, + "loss": 0.0977, + "step": 16272 + }, + { + "epoch": 0.14125745436237533, + "grad_norm": 0.314453125, + "learning_rate": 0.0018876414290098635, + "loss": 0.1543, + "step": 16273 + }, + { + "epoch": 0.1412661348425795, + "grad_norm": 0.333984375, + "learning_rate": 0.0018876270283481515, + "loss": 0.1533, + "step": 16274 + }, + { + "epoch": 0.14127481532278366, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018876126268251026, + "loss": 0.123, + "step": 16275 + }, + { + "epoch": 0.14128349580298782, + "grad_norm": 0.17578125, + "learning_rate": 0.0018875982244407324, + "loss": 0.1191, + "step": 16276 + }, + { + "epoch": 0.141292176283192, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018875838211950562, + "loss": 0.1436, + "step": 16277 + }, + { + "epoch": 0.14130085676339615, + "grad_norm": 0.322265625, + "learning_rate": 0.0018875694170880906, + "loss": 0.0859, + "step": 16278 + }, + { + "epoch": 0.14130953724360032, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018875550121198509, + "loss": 0.0879, + "step": 16279 + }, + { + "epoch": 0.14131821772380448, + "grad_norm": 0.25390625, + "learning_rate": 0.0018875406062903526, + "loss": 0.1055, + "step": 16280 + }, + { + "epoch": 0.14132689820400865, + "grad_norm": 0.248046875, + "learning_rate": 0.0018875261995996118, + "loss": 0.1309, + "step": 16281 + }, + { + "epoch": 0.14133557868421281, + "grad_norm": 0.291015625, + "learning_rate": 0.0018875117920476444, + "loss": 0.1357, + "step": 16282 + }, + { + "epoch": 0.14134425916441698, + "grad_norm": 0.984375, + "learning_rate": 0.0018874973836344657, + "loss": 0.1465, + "step": 16283 + }, + { + "epoch": 0.14135293964462115, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018874829743600917, + "loss": 0.1279, + "step": 16284 + }, + { + "epoch": 0.1413616201248253, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018874685642245384, + "loss": 0.0947, + "step": 16285 + }, + { + "epoch": 0.14137030060502948, + "grad_norm": 0.765625, + "learning_rate": 0.0018874541532278208, + "loss": 0.0986, + "step": 16286 + }, + { + "epoch": 0.14137898108523364, + "grad_norm": 0.48828125, + "learning_rate": 0.0018874397413699557, + "loss": 0.1182, + "step": 16287 + }, + { + "epoch": 0.1413876615654378, + "grad_norm": 0.265625, + "learning_rate": 0.001887425328650958, + "loss": 0.1318, + "step": 16288 + }, + { + "epoch": 0.14139634204564197, + "grad_norm": 0.60546875, + "learning_rate": 0.0018874109150708441, + "loss": 0.1094, + "step": 16289 + }, + { + "epoch": 0.14140502252584614, + "grad_norm": 0.5078125, + "learning_rate": 0.0018873965006296293, + "loss": 0.1045, + "step": 16290 + }, + { + "epoch": 0.1414137030060503, + "grad_norm": 0.462890625, + "learning_rate": 0.0018873820853273294, + "loss": 0.1001, + "step": 16291 + }, + { + "epoch": 0.14142238348625447, + "grad_norm": 3.84375, + "learning_rate": 0.0018873676691639606, + "loss": 0.2891, + "step": 16292 + }, + { + "epoch": 0.14143106396645863, + "grad_norm": 0.2060546875, + "learning_rate": 0.001887353252139538, + "loss": 0.1118, + "step": 16293 + }, + { + "epoch": 0.1414397444466628, + "grad_norm": 0.35546875, + "learning_rate": 0.001887338834254078, + "loss": 0.123, + "step": 16294 + }, + { + "epoch": 0.14144842492686696, + "grad_norm": 0.291015625, + "learning_rate": 0.0018873244155075957, + "loss": 0.1055, + "step": 16295 + }, + { + "epoch": 0.14145710540707113, + "grad_norm": 0.30078125, + "learning_rate": 0.0018873099959001076, + "loss": 0.1484, + "step": 16296 + }, + { + "epoch": 0.1414657858872753, + "grad_norm": 0.1259765625, + "learning_rate": 0.001887295575431629, + "loss": 0.1426, + "step": 16297 + }, + { + "epoch": 0.14147446636747946, + "grad_norm": 0.5703125, + "learning_rate": 0.0018872811541021757, + "loss": 0.127, + "step": 16298 + }, + { + "epoch": 0.14148314684768362, + "grad_norm": 0.67578125, + "learning_rate": 0.0018872667319117637, + "loss": 0.1089, + "step": 16299 + }, + { + "epoch": 0.1414918273278878, + "grad_norm": 0.73046875, + "learning_rate": 0.0018872523088604085, + "loss": 0.0967, + "step": 16300 + }, + { + "epoch": 0.14150050780809195, + "grad_norm": 0.55078125, + "learning_rate": 0.001887237884948126, + "loss": 0.1123, + "step": 16301 + }, + { + "epoch": 0.14150918828829612, + "grad_norm": 0.11376953125, + "learning_rate": 0.0018872234601749322, + "loss": 0.1191, + "step": 16302 + }, + { + "epoch": 0.14151786876850028, + "grad_norm": 0.40625, + "learning_rate": 0.0018872090345408423, + "loss": 0.1104, + "step": 16303 + }, + { + "epoch": 0.14152654924870445, + "grad_norm": 0.63671875, + "learning_rate": 0.0018871946080458728, + "loss": 0.1172, + "step": 16304 + }, + { + "epoch": 0.1415352297289086, + "grad_norm": 0.1201171875, + "learning_rate": 0.001887180180690039, + "loss": 0.1289, + "step": 16305 + }, + { + "epoch": 0.14154391020911278, + "grad_norm": 0.0810546875, + "learning_rate": 0.0018871657524733567, + "loss": 0.0967, + "step": 16306 + }, + { + "epoch": 0.14155259068931694, + "grad_norm": 0.05810546875, + "learning_rate": 0.0018871513233958417, + "loss": 0.0713, + "step": 16307 + }, + { + "epoch": 0.1415612711695211, + "grad_norm": 0.2021484375, + "learning_rate": 0.00188713689345751, + "loss": 0.1543, + "step": 16308 + }, + { + "epoch": 0.14156995164972527, + "grad_norm": 0.396484375, + "learning_rate": 0.001887122462658377, + "loss": 0.1621, + "step": 16309 + }, + { + "epoch": 0.14157863212992944, + "grad_norm": 0.57421875, + "learning_rate": 0.001887108030998459, + "loss": 0.1709, + "step": 16310 + }, + { + "epoch": 0.1415873126101336, + "grad_norm": 0.14453125, + "learning_rate": 0.001887093598477771, + "loss": 0.1182, + "step": 16311 + }, + { + "epoch": 0.14159599309033777, + "grad_norm": 0.197265625, + "learning_rate": 0.00188707916509633, + "loss": 0.1406, + "step": 16312 + }, + { + "epoch": 0.14160467357054193, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018870647308541503, + "loss": 0.0889, + "step": 16313 + }, + { + "epoch": 0.1416133540507461, + "grad_norm": 0.6875, + "learning_rate": 0.001887050295751249, + "loss": 0.1152, + "step": 16314 + }, + { + "epoch": 0.14162203453095026, + "grad_norm": 0.21484375, + "learning_rate": 0.0018870358597876408, + "loss": 0.1416, + "step": 16315 + }, + { + "epoch": 0.14163071501115443, + "grad_norm": 0.30859375, + "learning_rate": 0.0018870214229633425, + "loss": 0.1406, + "step": 16316 + }, + { + "epoch": 0.1416393954913586, + "grad_norm": 0.412109375, + "learning_rate": 0.0018870069852783692, + "loss": 0.1206, + "step": 16317 + }, + { + "epoch": 0.14164807597156276, + "grad_norm": 0.06396484375, + "learning_rate": 0.0018869925467327367, + "loss": 0.0957, + "step": 16318 + }, + { + "epoch": 0.14165675645176692, + "grad_norm": 0.515625, + "learning_rate": 0.0018869781073264614, + "loss": 0.1387, + "step": 16319 + }, + { + "epoch": 0.1416654369319711, + "grad_norm": 0.25390625, + "learning_rate": 0.0018869636670595582, + "loss": 0.1289, + "step": 16320 + }, + { + "epoch": 0.14167411741217525, + "grad_norm": 0.333984375, + "learning_rate": 0.0018869492259320439, + "loss": 0.0908, + "step": 16321 + }, + { + "epoch": 0.14168279789237942, + "grad_norm": 0.1328125, + "learning_rate": 0.0018869347839439333, + "loss": 0.125, + "step": 16322 + }, + { + "epoch": 0.14169147837258358, + "grad_norm": 0.443359375, + "learning_rate": 0.0018869203410952426, + "loss": 0.1406, + "step": 16323 + }, + { + "epoch": 0.14170015885278775, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018869058973859881, + "loss": 0.1191, + "step": 16324 + }, + { + "epoch": 0.14170883933299191, + "grad_norm": 0.220703125, + "learning_rate": 0.0018868914528161846, + "loss": 0.1162, + "step": 16325 + }, + { + "epoch": 0.14171751981319605, + "grad_norm": 0.84375, + "learning_rate": 0.0018868770073858488, + "loss": 0.1016, + "step": 16326 + }, + { + "epoch": 0.14172620029340022, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018868625610949958, + "loss": 0.1348, + "step": 16327 + }, + { + "epoch": 0.14173488077360438, + "grad_norm": 0.11865234375, + "learning_rate": 0.001886848113943642, + "loss": 0.123, + "step": 16328 + }, + { + "epoch": 0.14174356125380855, + "grad_norm": 0.58203125, + "learning_rate": 0.001886833665931803, + "loss": 0.1367, + "step": 16329 + }, + { + "epoch": 0.1417522417340127, + "grad_norm": 0.08984375, + "learning_rate": 0.0018868192170594944, + "loss": 0.1113, + "step": 16330 + }, + { + "epoch": 0.14176092221421688, + "grad_norm": 0.498046875, + "learning_rate": 0.001886804767326732, + "loss": 0.1289, + "step": 16331 + }, + { + "epoch": 0.14176960269442104, + "grad_norm": 0.703125, + "learning_rate": 0.0018867903167335321, + "loss": 0.0962, + "step": 16332 + }, + { + "epoch": 0.1417782831746252, + "grad_norm": 0.09375, + "learning_rate": 0.0018867758652799096, + "loss": 0.1221, + "step": 16333 + }, + { + "epoch": 0.14178696365482937, + "grad_norm": 0.8125, + "learning_rate": 0.001886761412965881, + "loss": 0.1289, + "step": 16334 + }, + { + "epoch": 0.14179564413503354, + "grad_norm": 0.298828125, + "learning_rate": 0.0018867469597914621, + "loss": 0.1387, + "step": 16335 + }, + { + "epoch": 0.1418043246152377, + "grad_norm": 0.31640625, + "learning_rate": 0.0018867325057566684, + "loss": 0.1328, + "step": 16336 + }, + { + "epoch": 0.14181300509544187, + "grad_norm": 0.169921875, + "learning_rate": 0.001886718050861516, + "loss": 0.1079, + "step": 16337 + }, + { + "epoch": 0.14182168557564603, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018867035951060207, + "loss": 0.1133, + "step": 16338 + }, + { + "epoch": 0.1418303660558502, + "grad_norm": 0.5078125, + "learning_rate": 0.001886689138490198, + "loss": 0.1309, + "step": 16339 + }, + { + "epoch": 0.14183904653605436, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018866746810140638, + "loss": 0.1074, + "step": 16340 + }, + { + "epoch": 0.14184772701625853, + "grad_norm": 0.08203125, + "learning_rate": 0.0018866602226776342, + "loss": 0.1035, + "step": 16341 + }, + { + "epoch": 0.1418564074964627, + "grad_norm": 0.0791015625, + "learning_rate": 0.0018866457634809245, + "loss": 0.126, + "step": 16342 + }, + { + "epoch": 0.14186508797666686, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018866313034239506, + "loss": 0.1562, + "step": 16343 + }, + { + "epoch": 0.14187376845687102, + "grad_norm": 0.87109375, + "learning_rate": 0.001886616842506729, + "loss": 0.0991, + "step": 16344 + }, + { + "epoch": 0.1418824489370752, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018866023807292752, + "loss": 0.1035, + "step": 16345 + }, + { + "epoch": 0.14189112941727935, + "grad_norm": 0.33203125, + "learning_rate": 0.0018865879180916044, + "loss": 0.1094, + "step": 16346 + }, + { + "epoch": 0.14189980989748352, + "grad_norm": 0.734375, + "learning_rate": 0.001886573454593733, + "loss": 0.1426, + "step": 16347 + }, + { + "epoch": 0.14190849037768768, + "grad_norm": 0.435546875, + "learning_rate": 0.001886558990235677, + "loss": 0.1064, + "step": 16348 + }, + { + "epoch": 0.14191717085789185, + "grad_norm": 0.12109375, + "learning_rate": 0.0018865445250174514, + "loss": 0.126, + "step": 16349 + }, + { + "epoch": 0.14192585133809602, + "grad_norm": 0.2177734375, + "learning_rate": 0.0018865300589390728, + "loss": 0.1309, + "step": 16350 + }, + { + "epoch": 0.14193453181830018, + "grad_norm": 0.2158203125, + "learning_rate": 0.0018865155920005565, + "loss": 0.1367, + "step": 16351 + }, + { + "epoch": 0.14194321229850435, + "grad_norm": 0.08203125, + "learning_rate": 0.001886501124201919, + "loss": 0.0981, + "step": 16352 + }, + { + "epoch": 0.1419518927787085, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018864866555431755, + "loss": 0.1279, + "step": 16353 + }, + { + "epoch": 0.14196057325891268, + "grad_norm": 0.08935546875, + "learning_rate": 0.001886472186024342, + "loss": 0.0986, + "step": 16354 + }, + { + "epoch": 0.14196925373911684, + "grad_norm": 0.875, + "learning_rate": 0.0018864577156454342, + "loss": 0.1426, + "step": 16355 + }, + { + "epoch": 0.141977934219321, + "grad_norm": 0.427734375, + "learning_rate": 0.0018864432444064682, + "loss": 0.1699, + "step": 16356 + }, + { + "epoch": 0.14198661469952517, + "grad_norm": 0.25390625, + "learning_rate": 0.0018864287723074596, + "loss": 0.1562, + "step": 16357 + }, + { + "epoch": 0.14199529517972934, + "grad_norm": 0.53125, + "learning_rate": 0.0018864142993484244, + "loss": 0.1299, + "step": 16358 + }, + { + "epoch": 0.1420039756599335, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018863998255293781, + "loss": 0.1494, + "step": 16359 + }, + { + "epoch": 0.14201265614013767, + "grad_norm": 0.189453125, + "learning_rate": 0.001886385350850337, + "loss": 0.1055, + "step": 16360 + }, + { + "epoch": 0.14202133662034183, + "grad_norm": 0.8046875, + "learning_rate": 0.001886370875311317, + "loss": 0.1016, + "step": 16361 + }, + { + "epoch": 0.142030017100546, + "grad_norm": 0.765625, + "learning_rate": 0.0018863563989123331, + "loss": 0.1006, + "step": 16362 + }, + { + "epoch": 0.14203869758075016, + "grad_norm": 0.73828125, + "learning_rate": 0.001886341921653402, + "loss": 0.0801, + "step": 16363 + }, + { + "epoch": 0.14204737806095433, + "grad_norm": 0.2275390625, + "learning_rate": 0.0018863274435345389, + "loss": 0.0996, + "step": 16364 + }, + { + "epoch": 0.1420560585411585, + "grad_norm": 0.58984375, + "learning_rate": 0.00188631296455576, + "loss": 0.1162, + "step": 16365 + }, + { + "epoch": 0.14206473902136266, + "grad_norm": 0.181640625, + "learning_rate": 0.001886298484717081, + "loss": 0.1182, + "step": 16366 + }, + { + "epoch": 0.14207341950156682, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018862840040185182, + "loss": 0.0889, + "step": 16367 + }, + { + "epoch": 0.142082099981771, + "grad_norm": 0.486328125, + "learning_rate": 0.0018862695224600864, + "loss": 0.1328, + "step": 16368 + }, + { + "epoch": 0.14209078046197515, + "grad_norm": 0.828125, + "learning_rate": 0.0018862550400418026, + "loss": 0.1641, + "step": 16369 + }, + { + "epoch": 0.14209946094217932, + "grad_norm": 0.18359375, + "learning_rate": 0.0018862405567636818, + "loss": 0.1152, + "step": 16370 + }, + { + "epoch": 0.14210814142238348, + "grad_norm": 0.076171875, + "learning_rate": 0.0018862260726257402, + "loss": 0.0933, + "step": 16371 + }, + { + "epoch": 0.14211682190258765, + "grad_norm": 0.052978515625, + "learning_rate": 0.0018862115876279936, + "loss": 0.0957, + "step": 16372 + }, + { + "epoch": 0.1421255023827918, + "grad_norm": 0.11865234375, + "learning_rate": 0.001886197101770458, + "loss": 0.0908, + "step": 16373 + }, + { + "epoch": 0.14213418286299598, + "grad_norm": 0.73046875, + "learning_rate": 0.0018861826150531487, + "loss": 0.0752, + "step": 16374 + }, + { + "epoch": 0.14214286334320014, + "grad_norm": 1.0078125, + "learning_rate": 0.0018861681274760823, + "loss": 0.0835, + "step": 16375 + }, + { + "epoch": 0.1421515438234043, + "grad_norm": 0.369140625, + "learning_rate": 0.001886153639039274, + "loss": 0.1172, + "step": 16376 + }, + { + "epoch": 0.14216022430360847, + "grad_norm": 0.62890625, + "learning_rate": 0.00188613914974274, + "loss": 0.1475, + "step": 16377 + }, + { + "epoch": 0.14216890478381264, + "grad_norm": 0.34765625, + "learning_rate": 0.0018861246595864955, + "loss": 0.166, + "step": 16378 + }, + { + "epoch": 0.1421775852640168, + "grad_norm": 0.44140625, + "learning_rate": 0.0018861101685705578, + "loss": 0.1582, + "step": 16379 + }, + { + "epoch": 0.14218626574422097, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018860956766949414, + "loss": 0.127, + "step": 16380 + }, + { + "epoch": 0.14219494622442513, + "grad_norm": 0.259765625, + "learning_rate": 0.0018860811839596623, + "loss": 0.103, + "step": 16381 + }, + { + "epoch": 0.1422036267046293, + "grad_norm": 0.357421875, + "learning_rate": 0.001886066690364737, + "loss": 0.1133, + "step": 16382 + }, + { + "epoch": 0.14221230718483346, + "grad_norm": 0.2451171875, + "learning_rate": 0.001886052195910181, + "loss": 0.1182, + "step": 16383 + }, + { + "epoch": 0.14222098766503763, + "grad_norm": 0.2255859375, + "learning_rate": 0.00188603770059601, + "loss": 0.1172, + "step": 16384 + }, + { + "epoch": 0.1422296681452418, + "grad_norm": 0.326171875, + "learning_rate": 0.00188602320442224, + "loss": 0.1387, + "step": 16385 + }, + { + "epoch": 0.14223834862544596, + "grad_norm": 0.234375, + "learning_rate": 0.0018860087073888866, + "loss": 0.0864, + "step": 16386 + }, + { + "epoch": 0.14224702910565012, + "grad_norm": 0.75, + "learning_rate": 0.0018859942094959664, + "loss": 0.0942, + "step": 16387 + }, + { + "epoch": 0.1422557095858543, + "grad_norm": 0.546875, + "learning_rate": 0.001885979710743494, + "loss": 0.0957, + "step": 16388 + }, + { + "epoch": 0.14226439006605845, + "grad_norm": 0.4609375, + "learning_rate": 0.0018859652111314867, + "loss": 0.1533, + "step": 16389 + }, + { + "epoch": 0.14227307054626262, + "grad_norm": 0.31640625, + "learning_rate": 0.0018859507106599594, + "loss": 0.1367, + "step": 16390 + }, + { + "epoch": 0.14228175102646679, + "grad_norm": 0.12451171875, + "learning_rate": 0.001885936209328928, + "loss": 0.103, + "step": 16391 + }, + { + "epoch": 0.14229043150667095, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018859217071384088, + "loss": 0.0669, + "step": 16392 + }, + { + "epoch": 0.14229911198687512, + "grad_norm": 0.455078125, + "learning_rate": 0.0018859072040884176, + "loss": 0.0991, + "step": 16393 + }, + { + "epoch": 0.14230779246707928, + "grad_norm": 0.15625, + "learning_rate": 0.0018858927001789698, + "loss": 0.1719, + "step": 16394 + }, + { + "epoch": 0.14231647294728345, + "grad_norm": 0.11328125, + "learning_rate": 0.0018858781954100817, + "loss": 0.0889, + "step": 16395 + }, + { + "epoch": 0.1423251534274876, + "grad_norm": 0.06640625, + "learning_rate": 0.001885863689781769, + "loss": 0.0801, + "step": 16396 + }, + { + "epoch": 0.14233383390769178, + "grad_norm": 0.376953125, + "learning_rate": 0.0018858491832940477, + "loss": 0.1143, + "step": 16397 + }, + { + "epoch": 0.14234251438789594, + "grad_norm": 0.48046875, + "learning_rate": 0.001885834675946933, + "loss": 0.1445, + "step": 16398 + }, + { + "epoch": 0.1423511948681001, + "grad_norm": 0.4140625, + "learning_rate": 0.0018858201677404418, + "loss": 0.1211, + "step": 16399 + }, + { + "epoch": 0.14235987534830427, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018858056586745892, + "loss": 0.083, + "step": 16400 + }, + { + "epoch": 0.14236855582850844, + "grad_norm": 0.91796875, + "learning_rate": 0.0018857911487493917, + "loss": 0.1211, + "step": 16401 + }, + { + "epoch": 0.1423772363087126, + "grad_norm": 0.36328125, + "learning_rate": 0.0018857766379648646, + "loss": 0.1123, + "step": 16402 + }, + { + "epoch": 0.14238591678891677, + "grad_norm": 0.46875, + "learning_rate": 0.001885762126321024, + "loss": 0.1099, + "step": 16403 + }, + { + "epoch": 0.14239459726912093, + "grad_norm": 0.94921875, + "learning_rate": 0.001885747613817886, + "loss": 0.1396, + "step": 16404 + }, + { + "epoch": 0.1424032777493251, + "grad_norm": 0.353515625, + "learning_rate": 0.0018857331004554657, + "loss": 0.1167, + "step": 16405 + }, + { + "epoch": 0.14241195822952926, + "grad_norm": 0.365234375, + "learning_rate": 0.0018857185862337798, + "loss": 0.1191, + "step": 16406 + }, + { + "epoch": 0.14242063870973343, + "grad_norm": 0.310546875, + "learning_rate": 0.0018857040711528438, + "loss": 0.125, + "step": 16407 + }, + { + "epoch": 0.1424293191899376, + "grad_norm": 0.2890625, + "learning_rate": 0.0018856895552126737, + "loss": 0.1289, + "step": 16408 + }, + { + "epoch": 0.14243799967014176, + "grad_norm": 0.0810546875, + "learning_rate": 0.001885675038413285, + "loss": 0.1289, + "step": 16409 + }, + { + "epoch": 0.14244668015034592, + "grad_norm": 0.412109375, + "learning_rate": 0.001885660520754694, + "loss": 0.1064, + "step": 16410 + }, + { + "epoch": 0.1424553606305501, + "grad_norm": 0.111328125, + "learning_rate": 0.0018856460022369165, + "loss": 0.1064, + "step": 16411 + }, + { + "epoch": 0.14246404111075425, + "grad_norm": 0.455078125, + "learning_rate": 0.0018856314828599686, + "loss": 0.1348, + "step": 16412 + }, + { + "epoch": 0.14247272159095842, + "grad_norm": 0.3515625, + "learning_rate": 0.0018856169626238657, + "loss": 0.0659, + "step": 16413 + }, + { + "epoch": 0.14248140207116258, + "grad_norm": 0.41015625, + "learning_rate": 0.001885602441528624, + "loss": 0.127, + "step": 16414 + }, + { + "epoch": 0.14249008255136675, + "grad_norm": 0.470703125, + "learning_rate": 0.0018855879195742592, + "loss": 0.1108, + "step": 16415 + }, + { + "epoch": 0.1424987630315709, + "grad_norm": 0.240234375, + "learning_rate": 0.0018855733967607874, + "loss": 0.0752, + "step": 16416 + }, + { + "epoch": 0.14250744351177508, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018855588730882237, + "loss": 0.0967, + "step": 16417 + }, + { + "epoch": 0.14251612399197924, + "grad_norm": 0.177734375, + "learning_rate": 0.0018855443485565854, + "loss": 0.1074, + "step": 16418 + }, + { + "epoch": 0.1425248044721834, + "grad_norm": 0.11279296875, + "learning_rate": 0.0018855298231658874, + "loss": 0.1025, + "step": 16419 + }, + { + "epoch": 0.14253348495238757, + "grad_norm": 0.77734375, + "learning_rate": 0.0018855152969161455, + "loss": 0.1416, + "step": 16420 + }, + { + "epoch": 0.14254216543259174, + "grad_norm": 0.2158203125, + "learning_rate": 0.001885500769807376, + "loss": 0.1191, + "step": 16421 + }, + { + "epoch": 0.1425508459127959, + "grad_norm": 0.1328125, + "learning_rate": 0.0018854862418395948, + "loss": 0.1289, + "step": 16422 + }, + { + "epoch": 0.14255952639300007, + "grad_norm": 0.08984375, + "learning_rate": 0.0018854717130128175, + "loss": 0.1162, + "step": 16423 + }, + { + "epoch": 0.14256820687320423, + "grad_norm": 0.4140625, + "learning_rate": 0.0018854571833270602, + "loss": 0.1191, + "step": 16424 + }, + { + "epoch": 0.1425768873534084, + "grad_norm": 0.220703125, + "learning_rate": 0.0018854426527823388, + "loss": 0.1504, + "step": 16425 + }, + { + "epoch": 0.14258556783361256, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018854281213786692, + "loss": 0.127, + "step": 16426 + }, + { + "epoch": 0.14259424831381673, + "grad_norm": 0.30859375, + "learning_rate": 0.0018854135891160668, + "loss": 0.1235, + "step": 16427 + }, + { + "epoch": 0.1426029287940209, + "grad_norm": 0.66796875, + "learning_rate": 0.0018853990559945483, + "loss": 0.1182, + "step": 16428 + }, + { + "epoch": 0.14261160927422506, + "grad_norm": 0.1357421875, + "learning_rate": 0.001885384522014129, + "loss": 0.1113, + "step": 16429 + }, + { + "epoch": 0.14262028975442922, + "grad_norm": 0.08447265625, + "learning_rate": 0.0018853699871748252, + "loss": 0.103, + "step": 16430 + }, + { + "epoch": 0.1426289702346334, + "grad_norm": 0.134765625, + "learning_rate": 0.0018853554514766523, + "loss": 0.127, + "step": 16431 + }, + { + "epoch": 0.14263765071483755, + "grad_norm": 0.80859375, + "learning_rate": 0.0018853409149196265, + "loss": 0.1387, + "step": 16432 + }, + { + "epoch": 0.14264633119504172, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018853263775037638, + "loss": 0.0864, + "step": 16433 + }, + { + "epoch": 0.14265501167524589, + "grad_norm": 0.0791015625, + "learning_rate": 0.00188531183922908, + "loss": 0.1138, + "step": 16434 + }, + { + "epoch": 0.14266369215545005, + "grad_norm": 0.0859375, + "learning_rate": 0.001885297300095591, + "loss": 0.0996, + "step": 16435 + }, + { + "epoch": 0.14267237263565422, + "grad_norm": 0.474609375, + "learning_rate": 0.0018852827601033124, + "loss": 0.1221, + "step": 16436 + }, + { + "epoch": 0.14268105311585838, + "grad_norm": 0.373046875, + "learning_rate": 0.0018852682192522605, + "loss": 0.0752, + "step": 16437 + }, + { + "epoch": 0.14268973359606255, + "grad_norm": 0.359375, + "learning_rate": 0.0018852536775424511, + "loss": 0.1484, + "step": 16438 + }, + { + "epoch": 0.1426984140762667, + "grad_norm": 0.63671875, + "learning_rate": 0.0018852391349739003, + "loss": 0.127, + "step": 16439 + }, + { + "epoch": 0.14270709455647088, + "grad_norm": 1.8828125, + "learning_rate": 0.0018852245915466235, + "loss": 0.1162, + "step": 16440 + }, + { + "epoch": 0.14271577503667504, + "grad_norm": 0.85546875, + "learning_rate": 0.001885210047260637, + "loss": 0.1162, + "step": 16441 + }, + { + "epoch": 0.1427244555168792, + "grad_norm": 0.34765625, + "learning_rate": 0.0018851955021159566, + "loss": 0.1133, + "step": 16442 + }, + { + "epoch": 0.14273313599708337, + "grad_norm": 0.20703125, + "learning_rate": 0.0018851809561125981, + "loss": 0.1533, + "step": 16443 + }, + { + "epoch": 0.14274181647728754, + "grad_norm": 0.578125, + "learning_rate": 0.0018851664092505778, + "loss": 0.1416, + "step": 16444 + }, + { + "epoch": 0.1427504969574917, + "grad_norm": 0.12060546875, + "learning_rate": 0.0018851518615299108, + "loss": 0.0962, + "step": 16445 + }, + { + "epoch": 0.14275917743769587, + "grad_norm": 0.54296875, + "learning_rate": 0.001885137312950614, + "loss": 0.1113, + "step": 16446 + }, + { + "epoch": 0.14276785791790003, + "grad_norm": 0.8046875, + "learning_rate": 0.0018851227635127024, + "loss": 0.1108, + "step": 16447 + }, + { + "epoch": 0.1427765383981042, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018851082132161927, + "loss": 0.0947, + "step": 16448 + }, + { + "epoch": 0.14278521887830833, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018850936620611002, + "loss": 0.0918, + "step": 16449 + }, + { + "epoch": 0.1427938993585125, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018850791100474414, + "loss": 0.1699, + "step": 16450 + }, + { + "epoch": 0.14280257983871666, + "grad_norm": 0.255859375, + "learning_rate": 0.0018850645571752316, + "loss": 0.0806, + "step": 16451 + }, + { + "epoch": 0.14281126031892083, + "grad_norm": 0.12890625, + "learning_rate": 0.001885050003444487, + "loss": 0.1338, + "step": 16452 + }, + { + "epoch": 0.142819940799125, + "grad_norm": 0.205078125, + "learning_rate": 0.0018850354488552238, + "loss": 0.0967, + "step": 16453 + }, + { + "epoch": 0.14282862127932916, + "grad_norm": 0.3125, + "learning_rate": 0.001885020893407457, + "loss": 0.0996, + "step": 16454 + }, + { + "epoch": 0.14283730175953332, + "grad_norm": 1.03125, + "learning_rate": 0.0018850063371012038, + "loss": 0.1084, + "step": 16455 + }, + { + "epoch": 0.1428459822397375, + "grad_norm": 0.1640625, + "learning_rate": 0.001884991779936479, + "loss": 0.0977, + "step": 16456 + }, + { + "epoch": 0.14285466271994166, + "grad_norm": 0.55859375, + "learning_rate": 0.0018849772219132993, + "loss": 0.126, + "step": 16457 + }, + { + "epoch": 0.14286334320014582, + "grad_norm": 0.255859375, + "learning_rate": 0.0018849626630316801, + "loss": 0.1445, + "step": 16458 + }, + { + "epoch": 0.14287202368034999, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018849481032916378, + "loss": 0.0869, + "step": 16459 + }, + { + "epoch": 0.14288070416055415, + "grad_norm": 0.173828125, + "learning_rate": 0.0018849335426931878, + "loss": 0.1035, + "step": 16460 + }, + { + "epoch": 0.14288938464075832, + "grad_norm": 0.369140625, + "learning_rate": 0.0018849189812363465, + "loss": 0.1328, + "step": 16461 + }, + { + "epoch": 0.14289806512096248, + "grad_norm": 0.6328125, + "learning_rate": 0.0018849044189211293, + "loss": 0.0952, + "step": 16462 + }, + { + "epoch": 0.14290674560116665, + "grad_norm": 0.404296875, + "learning_rate": 0.0018848898557475525, + "loss": 0.0918, + "step": 16463 + }, + { + "epoch": 0.1429154260813708, + "grad_norm": 0.447265625, + "learning_rate": 0.001884875291715632, + "loss": 0.2246, + "step": 16464 + }, + { + "epoch": 0.14292410656157498, + "grad_norm": 0.4140625, + "learning_rate": 0.0018848607268253839, + "loss": 0.1016, + "step": 16465 + }, + { + "epoch": 0.14293278704177914, + "grad_norm": 0.2265625, + "learning_rate": 0.0018848461610768237, + "loss": 0.1367, + "step": 16466 + }, + { + "epoch": 0.1429414675219833, + "grad_norm": 0.302734375, + "learning_rate": 0.0018848315944699674, + "loss": 0.1484, + "step": 16467 + }, + { + "epoch": 0.14295014800218747, + "grad_norm": 0.208984375, + "learning_rate": 0.0018848170270048313, + "loss": 0.123, + "step": 16468 + }, + { + "epoch": 0.14295882848239164, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018848024586814308, + "loss": 0.1592, + "step": 16469 + }, + { + "epoch": 0.1429675089625958, + "grad_norm": 0.63671875, + "learning_rate": 0.0018847878894997826, + "loss": 0.3047, + "step": 16470 + }, + { + "epoch": 0.14297618944279997, + "grad_norm": 0.41015625, + "learning_rate": 0.0018847733194599014, + "loss": 0.1006, + "step": 16471 + }, + { + "epoch": 0.14298486992300413, + "grad_norm": 0.62109375, + "learning_rate": 0.0018847587485618046, + "loss": 0.1201, + "step": 16472 + }, + { + "epoch": 0.1429935504032083, + "grad_norm": 0.2255859375, + "learning_rate": 0.001884744176805507, + "loss": 0.126, + "step": 16473 + }, + { + "epoch": 0.14300223088341246, + "grad_norm": 0.2392578125, + "learning_rate": 0.001884729604191025, + "loss": 0.1025, + "step": 16474 + }, + { + "epoch": 0.14301091136361663, + "grad_norm": 0.90234375, + "learning_rate": 0.0018847150307183748, + "loss": 0.165, + "step": 16475 + }, + { + "epoch": 0.1430195918438208, + "grad_norm": 0.88671875, + "learning_rate": 0.0018847004563875718, + "loss": 0.1016, + "step": 16476 + }, + { + "epoch": 0.14302827232402496, + "grad_norm": 0.10205078125, + "learning_rate": 0.0018846858811986324, + "loss": 0.1113, + "step": 16477 + }, + { + "epoch": 0.14303695280422912, + "grad_norm": 1.671875, + "learning_rate": 0.001884671305151572, + "loss": 0.3145, + "step": 16478 + }, + { + "epoch": 0.1430456332844333, + "grad_norm": 0.42578125, + "learning_rate": 0.001884656728246407, + "loss": 0.083, + "step": 16479 + }, + { + "epoch": 0.14305431376463745, + "grad_norm": 0.09912109375, + "learning_rate": 0.0018846421504831532, + "loss": 0.0854, + "step": 16480 + }, + { + "epoch": 0.14306299424484162, + "grad_norm": 0.4375, + "learning_rate": 0.0018846275718618264, + "loss": 0.1108, + "step": 16481 + }, + { + "epoch": 0.14307167472504578, + "grad_norm": 0.5703125, + "learning_rate": 0.0018846129923824429, + "loss": 0.1084, + "step": 16482 + }, + { + "epoch": 0.14308035520524995, + "grad_norm": 0.421875, + "learning_rate": 0.0018845984120450185, + "loss": 0.1582, + "step": 16483 + }, + { + "epoch": 0.1430890356854541, + "grad_norm": 0.71484375, + "learning_rate": 0.0018845838308495686, + "loss": 0.125, + "step": 16484 + }, + { + "epoch": 0.14309771616565828, + "grad_norm": 0.1904296875, + "learning_rate": 0.00188456924879611, + "loss": 0.0967, + "step": 16485 + }, + { + "epoch": 0.14310639664586244, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018845546658846583, + "loss": 0.083, + "step": 16486 + }, + { + "epoch": 0.1431150771260666, + "grad_norm": 0.498046875, + "learning_rate": 0.001884540082115229, + "loss": 0.1641, + "step": 16487 + }, + { + "epoch": 0.14312375760627077, + "grad_norm": 0.9921875, + "learning_rate": 0.0018845254974878386, + "loss": 0.1426, + "step": 16488 + }, + { + "epoch": 0.14313243808647494, + "grad_norm": 0.4296875, + "learning_rate": 0.0018845109120025033, + "loss": 0.1406, + "step": 16489 + }, + { + "epoch": 0.1431411185666791, + "grad_norm": 0.5546875, + "learning_rate": 0.001884496325659238, + "loss": 0.1182, + "step": 16490 + }, + { + "epoch": 0.14314979904688327, + "grad_norm": 0.5390625, + "learning_rate": 0.0018844817384580598, + "loss": 0.1025, + "step": 16491 + }, + { + "epoch": 0.14315847952708743, + "grad_norm": 0.384765625, + "learning_rate": 0.0018844671503989842, + "loss": 0.1055, + "step": 16492 + }, + { + "epoch": 0.1431671600072916, + "grad_norm": 0.1640625, + "learning_rate": 0.0018844525614820267, + "loss": 0.1196, + "step": 16493 + }, + { + "epoch": 0.14317584048749576, + "grad_norm": 0.546875, + "learning_rate": 0.0018844379717072037, + "loss": 0.0845, + "step": 16494 + }, + { + "epoch": 0.14318452096769993, + "grad_norm": 0.455078125, + "learning_rate": 0.0018844233810745314, + "loss": 0.0913, + "step": 16495 + }, + { + "epoch": 0.1431932014479041, + "grad_norm": 0.62109375, + "learning_rate": 0.0018844087895840257, + "loss": 0.1562, + "step": 16496 + }, + { + "epoch": 0.14320188192810826, + "grad_norm": 0.26171875, + "learning_rate": 0.001884394197235702, + "loss": 0.1279, + "step": 16497 + }, + { + "epoch": 0.14321056240831243, + "grad_norm": 0.162109375, + "learning_rate": 0.0018843796040295764, + "loss": 0.1191, + "step": 16498 + }, + { + "epoch": 0.1432192428885166, + "grad_norm": 0.53125, + "learning_rate": 0.0018843650099656653, + "loss": 0.123, + "step": 16499 + }, + { + "epoch": 0.14322792336872076, + "grad_norm": 0.66796875, + "learning_rate": 0.0018843504150439843, + "loss": 0.1108, + "step": 16500 + }, + { + "epoch": 0.14323660384892492, + "grad_norm": 0.08056640625, + "learning_rate": 0.0018843358192645494, + "loss": 0.103, + "step": 16501 + }, + { + "epoch": 0.14324528432912909, + "grad_norm": 0.140625, + "learning_rate": 0.001884321222627377, + "loss": 0.0752, + "step": 16502 + }, + { + "epoch": 0.14325396480933325, + "grad_norm": 0.5703125, + "learning_rate": 0.0018843066251324825, + "loss": 0.1562, + "step": 16503 + }, + { + "epoch": 0.14326264528953742, + "grad_norm": 0.287109375, + "learning_rate": 0.001884292026779882, + "loss": 0.1279, + "step": 16504 + }, + { + "epoch": 0.14327132576974158, + "grad_norm": 0.19921875, + "learning_rate": 0.0018842774275695915, + "loss": 0.1328, + "step": 16505 + }, + { + "epoch": 0.14328000624994575, + "grad_norm": 0.19921875, + "learning_rate": 0.001884262827501627, + "loss": 0.1211, + "step": 16506 + }, + { + "epoch": 0.1432886867301499, + "grad_norm": 0.265625, + "learning_rate": 0.0018842482265760045, + "loss": 0.1318, + "step": 16507 + }, + { + "epoch": 0.14329736721035408, + "grad_norm": 0.32421875, + "learning_rate": 0.0018842336247927398, + "loss": 0.1465, + "step": 16508 + }, + { + "epoch": 0.14330604769055824, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018842190221518493, + "loss": 0.126, + "step": 16509 + }, + { + "epoch": 0.1433147281707624, + "grad_norm": 0.453125, + "learning_rate": 0.0018842044186533485, + "loss": 0.1079, + "step": 16510 + }, + { + "epoch": 0.14332340865096657, + "grad_norm": 0.216796875, + "learning_rate": 0.0018841898142972536, + "loss": 0.1133, + "step": 16511 + }, + { + "epoch": 0.14333208913117074, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018841752090835802, + "loss": 0.1279, + "step": 16512 + }, + { + "epoch": 0.1433407696113749, + "grad_norm": 0.111328125, + "learning_rate": 0.0018841606030123449, + "loss": 0.1074, + "step": 16513 + }, + { + "epoch": 0.14334945009157907, + "grad_norm": 0.0859375, + "learning_rate": 0.0018841459960835631, + "loss": 0.1089, + "step": 16514 + }, + { + "epoch": 0.14335813057178323, + "grad_norm": 0.35546875, + "learning_rate": 0.0018841313882972512, + "loss": 0.127, + "step": 16515 + }, + { + "epoch": 0.1433668110519874, + "grad_norm": 0.373046875, + "learning_rate": 0.0018841167796534249, + "loss": 0.1543, + "step": 16516 + }, + { + "epoch": 0.14337549153219156, + "grad_norm": 0.130859375, + "learning_rate": 0.0018841021701521002, + "loss": 0.083, + "step": 16517 + }, + { + "epoch": 0.14338417201239573, + "grad_norm": 0.61328125, + "learning_rate": 0.0018840875597932933, + "loss": 0.1226, + "step": 16518 + }, + { + "epoch": 0.1433928524925999, + "grad_norm": 0.337890625, + "learning_rate": 0.00188407294857702, + "loss": 0.0889, + "step": 16519 + }, + { + "epoch": 0.14340153297280406, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018840583365032965, + "loss": 0.0977, + "step": 16520 + }, + { + "epoch": 0.14341021345300822, + "grad_norm": 0.119140625, + "learning_rate": 0.0018840437235721384, + "loss": 0.1426, + "step": 16521 + }, + { + "epoch": 0.1434188939332124, + "grad_norm": 0.263671875, + "learning_rate": 0.0018840291097835615, + "loss": 0.124, + "step": 16522 + }, + { + "epoch": 0.14342757441341655, + "grad_norm": 0.3203125, + "learning_rate": 0.0018840144951375827, + "loss": 0.126, + "step": 16523 + }, + { + "epoch": 0.14343625489362072, + "grad_norm": 0.1796875, + "learning_rate": 0.001883999879634217, + "loss": 0.1025, + "step": 16524 + }, + { + "epoch": 0.14344493537382488, + "grad_norm": 0.87890625, + "learning_rate": 0.0018839852632734813, + "loss": 0.1436, + "step": 16525 + }, + { + "epoch": 0.14345361585402905, + "grad_norm": 0.498046875, + "learning_rate": 0.0018839706460553906, + "loss": 0.1289, + "step": 16526 + }, + { + "epoch": 0.1434622963342332, + "grad_norm": 0.375, + "learning_rate": 0.0018839560279799614, + "loss": 0.1196, + "step": 16527 + }, + { + "epoch": 0.14347097681443738, + "grad_norm": 0.384765625, + "learning_rate": 0.00188394140904721, + "loss": 0.1289, + "step": 16528 + }, + { + "epoch": 0.14347965729464154, + "grad_norm": 0.328125, + "learning_rate": 0.0018839267892571522, + "loss": 0.1426, + "step": 16529 + }, + { + "epoch": 0.1434883377748457, + "grad_norm": 0.353515625, + "learning_rate": 0.0018839121686098036, + "loss": 0.1201, + "step": 16530 + }, + { + "epoch": 0.14349701825504987, + "grad_norm": 0.13671875, + "learning_rate": 0.0018838975471051802, + "loss": 0.0957, + "step": 16531 + }, + { + "epoch": 0.14350569873525404, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018838829247432983, + "loss": 0.0947, + "step": 16532 + }, + { + "epoch": 0.1435143792154582, + "grad_norm": 0.85546875, + "learning_rate": 0.0018838683015241738, + "loss": 0.1523, + "step": 16533 + }, + { + "epoch": 0.14352305969566237, + "grad_norm": 0.57421875, + "learning_rate": 0.0018838536774478233, + "loss": 0.0981, + "step": 16534 + }, + { + "epoch": 0.14353174017586653, + "grad_norm": 0.65234375, + "learning_rate": 0.0018838390525142614, + "loss": 0.1406, + "step": 16535 + }, + { + "epoch": 0.1435404206560707, + "grad_norm": 0.609375, + "learning_rate": 0.0018838244267235051, + "loss": 0.1465, + "step": 16536 + }, + { + "epoch": 0.14354910113627486, + "grad_norm": 0.2177734375, + "learning_rate": 0.0018838098000755704, + "loss": 0.0903, + "step": 16537 + }, + { + "epoch": 0.14355778161647903, + "grad_norm": 0.349609375, + "learning_rate": 0.0018837951725704728, + "loss": 0.1279, + "step": 16538 + }, + { + "epoch": 0.1435664620966832, + "grad_norm": 1.3515625, + "learning_rate": 0.0018837805442082289, + "loss": 0.0981, + "step": 16539 + }, + { + "epoch": 0.14357514257688736, + "grad_norm": 0.1201171875, + "learning_rate": 0.0018837659149888541, + "loss": 0.1245, + "step": 16540 + }, + { + "epoch": 0.14358382305709153, + "grad_norm": 0.67578125, + "learning_rate": 0.001883751284912365, + "loss": 0.1191, + "step": 16541 + }, + { + "epoch": 0.1435925035372957, + "grad_norm": 0.486328125, + "learning_rate": 0.0018837366539787767, + "loss": 0.1162, + "step": 16542 + }, + { + "epoch": 0.14360118401749986, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018837220221881064, + "loss": 0.0967, + "step": 16543 + }, + { + "epoch": 0.14360986449770402, + "grad_norm": 0.392578125, + "learning_rate": 0.0018837073895403692, + "loss": 0.1162, + "step": 16544 + }, + { + "epoch": 0.14361854497790819, + "grad_norm": 1.5078125, + "learning_rate": 0.0018836927560355815, + "loss": 0.1504, + "step": 16545 + }, + { + "epoch": 0.14362722545811235, + "grad_norm": 0.75, + "learning_rate": 0.001883678121673759, + "loss": 0.1562, + "step": 16546 + }, + { + "epoch": 0.14363590593831652, + "grad_norm": 0.765625, + "learning_rate": 0.001883663486454918, + "loss": 0.1338, + "step": 16547 + }, + { + "epoch": 0.14364458641852068, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018836488503790744, + "loss": 0.1074, + "step": 16548 + }, + { + "epoch": 0.14365326689872485, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018836342134462439, + "loss": 0.1143, + "step": 16549 + }, + { + "epoch": 0.143661947378929, + "grad_norm": 0.375, + "learning_rate": 0.001883619575656443, + "loss": 0.0771, + "step": 16550 + }, + { + "epoch": 0.14367062785913318, + "grad_norm": 0.193359375, + "learning_rate": 0.0018836049370096876, + "loss": 0.1172, + "step": 16551 + }, + { + "epoch": 0.14367930833933734, + "grad_norm": 0.486328125, + "learning_rate": 0.0018835902975059935, + "loss": 0.0991, + "step": 16552 + }, + { + "epoch": 0.1436879888195415, + "grad_norm": 1.140625, + "learning_rate": 0.001883575657145377, + "loss": 0.127, + "step": 16553 + }, + { + "epoch": 0.14369666929974567, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018835610159278536, + "loss": 0.1035, + "step": 16554 + }, + { + "epoch": 0.14370534977994984, + "grad_norm": 0.52734375, + "learning_rate": 0.00188354637385344, + "loss": 0.1152, + "step": 16555 + }, + { + "epoch": 0.143714030260154, + "grad_norm": 0.78125, + "learning_rate": 0.0018835317309221515, + "loss": 0.0977, + "step": 16556 + }, + { + "epoch": 0.14372271074035817, + "grad_norm": 0.703125, + "learning_rate": 0.0018835170871340048, + "loss": 0.084, + "step": 16557 + }, + { + "epoch": 0.14373139122056233, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018835024424890153, + "loss": 0.1055, + "step": 16558 + }, + { + "epoch": 0.1437400717007665, + "grad_norm": 0.5, + "learning_rate": 0.0018834877969871994, + "loss": 0.1152, + "step": 16559 + }, + { + "epoch": 0.14374875218097066, + "grad_norm": 0.1904296875, + "learning_rate": 0.001883473150628573, + "loss": 0.0991, + "step": 16560 + }, + { + "epoch": 0.14375743266117483, + "grad_norm": 0.298828125, + "learning_rate": 0.0018834585034131525, + "loss": 0.1074, + "step": 16561 + }, + { + "epoch": 0.143766113141379, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018834438553409533, + "loss": 0.1377, + "step": 16562 + }, + { + "epoch": 0.14377479362158316, + "grad_norm": 0.15234375, + "learning_rate": 0.0018834292064119915, + "loss": 0.1719, + "step": 16563 + }, + { + "epoch": 0.14378347410178732, + "grad_norm": 0.19921875, + "learning_rate": 0.0018834145566262834, + "loss": 0.1006, + "step": 16564 + }, + { + "epoch": 0.1437921545819915, + "grad_norm": 0.33203125, + "learning_rate": 0.0018833999059838448, + "loss": 0.1299, + "step": 16565 + }, + { + "epoch": 0.14380083506219565, + "grad_norm": 1.265625, + "learning_rate": 0.0018833852544846918, + "loss": 0.1436, + "step": 16566 + }, + { + "epoch": 0.14380951554239982, + "grad_norm": 0.2392578125, + "learning_rate": 0.001883370602128841, + "loss": 0.0938, + "step": 16567 + }, + { + "epoch": 0.14381819602260398, + "grad_norm": 1.359375, + "learning_rate": 0.0018833559489163075, + "loss": 0.1104, + "step": 16568 + }, + { + "epoch": 0.14382687650280815, + "grad_norm": 0.640625, + "learning_rate": 0.0018833412948471076, + "loss": 0.1011, + "step": 16569 + }, + { + "epoch": 0.1438355569830123, + "grad_norm": 0.283203125, + "learning_rate": 0.0018833266399212573, + "loss": 0.0986, + "step": 16570 + }, + { + "epoch": 0.14384423746321648, + "grad_norm": 0.458984375, + "learning_rate": 0.0018833119841387732, + "loss": 0.1309, + "step": 16571 + }, + { + "epoch": 0.14385291794342062, + "grad_norm": 0.109375, + "learning_rate": 0.0018832973274996708, + "loss": 0.1162, + "step": 16572 + }, + { + "epoch": 0.14386159842362478, + "grad_norm": 0.71484375, + "learning_rate": 0.0018832826700039657, + "loss": 0.1064, + "step": 16573 + }, + { + "epoch": 0.14387027890382895, + "grad_norm": 0.27734375, + "learning_rate": 0.001883268011651675, + "loss": 0.123, + "step": 16574 + }, + { + "epoch": 0.1438789593840331, + "grad_norm": 0.80859375, + "learning_rate": 0.001883253352442814, + "loss": 0.1602, + "step": 16575 + }, + { + "epoch": 0.14388763986423728, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018832386923773993, + "loss": 0.0957, + "step": 16576 + }, + { + "epoch": 0.14389632034444144, + "grad_norm": 1.5234375, + "learning_rate": 0.0018832240314554459, + "loss": 0.1318, + "step": 16577 + }, + { + "epoch": 0.1439050008246456, + "grad_norm": 0.70703125, + "learning_rate": 0.0018832093696769705, + "loss": 0.1104, + "step": 16578 + }, + { + "epoch": 0.14391368130484977, + "grad_norm": 0.73828125, + "learning_rate": 0.0018831947070419896, + "loss": 0.082, + "step": 16579 + }, + { + "epoch": 0.14392236178505394, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018831800435505181, + "loss": 0.127, + "step": 16580 + }, + { + "epoch": 0.1439310422652581, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018831653792025732, + "loss": 0.1035, + "step": 16581 + }, + { + "epoch": 0.14393972274546227, + "grad_norm": 0.361328125, + "learning_rate": 0.0018831507139981702, + "loss": 0.1309, + "step": 16582 + }, + { + "epoch": 0.14394840322566643, + "grad_norm": 0.44921875, + "learning_rate": 0.0018831360479373254, + "loss": 0.1738, + "step": 16583 + }, + { + "epoch": 0.1439570837058706, + "grad_norm": 0.439453125, + "learning_rate": 0.0018831213810200547, + "loss": 0.1416, + "step": 16584 + }, + { + "epoch": 0.14396576418607476, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018831067132463741, + "loss": 0.127, + "step": 16585 + }, + { + "epoch": 0.14397444466627893, + "grad_norm": 1.296875, + "learning_rate": 0.0018830920446163003, + "loss": 0.1777, + "step": 16586 + }, + { + "epoch": 0.1439831251464831, + "grad_norm": 0.26953125, + "learning_rate": 0.0018830773751298481, + "loss": 0.1045, + "step": 16587 + }, + { + "epoch": 0.14399180562668726, + "grad_norm": 0.40234375, + "learning_rate": 0.0018830627047870345, + "loss": 0.1123, + "step": 16588 + }, + { + "epoch": 0.14400048610689142, + "grad_norm": 0.37109375, + "learning_rate": 0.0018830480335878755, + "loss": 0.0889, + "step": 16589 + }, + { + "epoch": 0.1440091665870956, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018830333615323866, + "loss": 0.0977, + "step": 16590 + }, + { + "epoch": 0.14401784706729975, + "grad_norm": 0.140625, + "learning_rate": 0.0018830186886205848, + "loss": 0.1367, + "step": 16591 + }, + { + "epoch": 0.14402652754750392, + "grad_norm": 0.126953125, + "learning_rate": 0.0018830040148524849, + "loss": 0.1543, + "step": 16592 + }, + { + "epoch": 0.14403520802770808, + "grad_norm": 0.10986328125, + "learning_rate": 0.001882989340228104, + "loss": 0.0864, + "step": 16593 + }, + { + "epoch": 0.14404388850791225, + "grad_norm": 0.376953125, + "learning_rate": 0.0018829746647474575, + "loss": 0.083, + "step": 16594 + }, + { + "epoch": 0.14405256898811641, + "grad_norm": 0.76171875, + "learning_rate": 0.0018829599884105618, + "loss": 0.1069, + "step": 16595 + }, + { + "epoch": 0.14406124946832058, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018829453112174325, + "loss": 0.0996, + "step": 16596 + }, + { + "epoch": 0.14406992994852474, + "grad_norm": 0.412109375, + "learning_rate": 0.0018829306331680862, + "loss": 0.1396, + "step": 16597 + }, + { + "epoch": 0.1440786104287289, + "grad_norm": 1.0625, + "learning_rate": 0.0018829159542625385, + "loss": 0.1416, + "step": 16598 + }, + { + "epoch": 0.14408729090893307, + "grad_norm": 0.37109375, + "learning_rate": 0.001882901274500806, + "loss": 0.1152, + "step": 16599 + }, + { + "epoch": 0.14409597138913724, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018828865938829043, + "loss": 0.1348, + "step": 16600 + }, + { + "epoch": 0.1441046518693414, + "grad_norm": 0.2578125, + "learning_rate": 0.0018828719124088498, + "loss": 0.1045, + "step": 16601 + }, + { + "epoch": 0.14411333234954557, + "grad_norm": 0.7109375, + "learning_rate": 0.001882857230078658, + "loss": 0.062, + "step": 16602 + }, + { + "epoch": 0.14412201282974973, + "grad_norm": 0.1025390625, + "learning_rate": 0.0018828425468923455, + "loss": 0.1191, + "step": 16603 + }, + { + "epoch": 0.1441306933099539, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018828278628499281, + "loss": 0.0952, + "step": 16604 + }, + { + "epoch": 0.14413937379015807, + "grad_norm": 0.765625, + "learning_rate": 0.0018828131779514217, + "loss": 0.1172, + "step": 16605 + }, + { + "epoch": 0.14414805427036223, + "grad_norm": 0.84375, + "learning_rate": 0.001882798492196843, + "loss": 0.1504, + "step": 16606 + }, + { + "epoch": 0.1441567347505664, + "grad_norm": 0.361328125, + "learning_rate": 0.0018827838055862072, + "loss": 0.125, + "step": 16607 + }, + { + "epoch": 0.14416541523077056, + "grad_norm": 0.1357421875, + "learning_rate": 0.001882769118119531, + "loss": 0.1025, + "step": 16608 + }, + { + "epoch": 0.14417409571097473, + "grad_norm": 0.08056640625, + "learning_rate": 0.0018827544297968302, + "loss": 0.1074, + "step": 16609 + }, + { + "epoch": 0.1441827761911789, + "grad_norm": 0.384765625, + "learning_rate": 0.001882739740618121, + "loss": 0.1543, + "step": 16610 + }, + { + "epoch": 0.14419145667138306, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018827250505834194, + "loss": 0.1055, + "step": 16611 + }, + { + "epoch": 0.14420013715158722, + "grad_norm": 0.62109375, + "learning_rate": 0.0018827103596927414, + "loss": 0.0781, + "step": 16612 + }, + { + "epoch": 0.1442088176317914, + "grad_norm": 0.11669921875, + "learning_rate": 0.001882695667946103, + "loss": 0.0854, + "step": 16613 + }, + { + "epoch": 0.14421749811199555, + "grad_norm": 0.275390625, + "learning_rate": 0.0018826809753435204, + "loss": 0.1133, + "step": 16614 + }, + { + "epoch": 0.14422617859219972, + "grad_norm": 0.59765625, + "learning_rate": 0.0018826662818850098, + "loss": 0.1152, + "step": 16615 + }, + { + "epoch": 0.14423485907240388, + "grad_norm": 0.349609375, + "learning_rate": 0.001882651587570587, + "loss": 0.1416, + "step": 16616 + }, + { + "epoch": 0.14424353955260805, + "grad_norm": 0.390625, + "learning_rate": 0.001882636892400268, + "loss": 0.0957, + "step": 16617 + }, + { + "epoch": 0.1442522200328122, + "grad_norm": 0.22265625, + "learning_rate": 0.0018826221963740696, + "loss": 0.0947, + "step": 16618 + }, + { + "epoch": 0.14426090051301638, + "grad_norm": 0.255859375, + "learning_rate": 0.001882607499492007, + "loss": 0.1484, + "step": 16619 + }, + { + "epoch": 0.14426958099322054, + "grad_norm": 0.365234375, + "learning_rate": 0.0018825928017540963, + "loss": 0.0981, + "step": 16620 + }, + { + "epoch": 0.1442782614734247, + "grad_norm": 0.671875, + "learning_rate": 0.0018825781031603544, + "loss": 0.1123, + "step": 16621 + }, + { + "epoch": 0.14428694195362887, + "grad_norm": 0.06884765625, + "learning_rate": 0.0018825634037107964, + "loss": 0.0791, + "step": 16622 + }, + { + "epoch": 0.14429562243383304, + "grad_norm": 1.21875, + "learning_rate": 0.001882548703405439, + "loss": 0.1484, + "step": 16623 + }, + { + "epoch": 0.1443043029140372, + "grad_norm": 0.65234375, + "learning_rate": 0.0018825340022442982, + "loss": 0.1162, + "step": 16624 + }, + { + "epoch": 0.14431298339424137, + "grad_norm": 0.310546875, + "learning_rate": 0.0018825193002273899, + "loss": 0.0879, + "step": 16625 + }, + { + "epoch": 0.14432166387444553, + "grad_norm": 0.08154296875, + "learning_rate": 0.00188250459735473, + "loss": 0.1001, + "step": 16626 + }, + { + "epoch": 0.1443303443546497, + "grad_norm": 0.58984375, + "learning_rate": 0.0018824898936263353, + "loss": 0.1191, + "step": 16627 + }, + { + "epoch": 0.14433902483485386, + "grad_norm": 0.2099609375, + "learning_rate": 0.001882475189042221, + "loss": 0.1035, + "step": 16628 + }, + { + "epoch": 0.14434770531505803, + "grad_norm": 0.291015625, + "learning_rate": 0.0018824604836024039, + "loss": 0.1299, + "step": 16629 + }, + { + "epoch": 0.1443563857952622, + "grad_norm": 0.267578125, + "learning_rate": 0.0018824457773068998, + "loss": 0.1094, + "step": 16630 + }, + { + "epoch": 0.14436506627546636, + "grad_norm": 0.24609375, + "learning_rate": 0.0018824310701557243, + "loss": 0.1299, + "step": 16631 + }, + { + "epoch": 0.14437374675567052, + "grad_norm": 0.384765625, + "learning_rate": 0.0018824163621488942, + "loss": 0.1035, + "step": 16632 + }, + { + "epoch": 0.1443824272358747, + "grad_norm": 0.22265625, + "learning_rate": 0.0018824016532864257, + "loss": 0.125, + "step": 16633 + }, + { + "epoch": 0.14439110771607885, + "grad_norm": 0.2373046875, + "learning_rate": 0.0018823869435683341, + "loss": 0.1064, + "step": 16634 + }, + { + "epoch": 0.14439978819628302, + "grad_norm": 0.15625, + "learning_rate": 0.0018823722329946357, + "loss": 0.0796, + "step": 16635 + }, + { + "epoch": 0.14440846867648718, + "grad_norm": 0.240234375, + "learning_rate": 0.0018823575215653472, + "loss": 0.1494, + "step": 16636 + }, + { + "epoch": 0.14441714915669135, + "grad_norm": 0.31640625, + "learning_rate": 0.001882342809280484, + "loss": 0.1406, + "step": 16637 + }, + { + "epoch": 0.14442582963689551, + "grad_norm": 0.72265625, + "learning_rate": 0.0018823280961400627, + "loss": 0.207, + "step": 16638 + }, + { + "epoch": 0.14443451011709968, + "grad_norm": 0.09130859375, + "learning_rate": 0.0018823133821440987, + "loss": 0.1079, + "step": 16639 + }, + { + "epoch": 0.14444319059730384, + "grad_norm": 0.07958984375, + "learning_rate": 0.001882298667292609, + "loss": 0.0796, + "step": 16640 + }, + { + "epoch": 0.144451871077508, + "grad_norm": 0.423828125, + "learning_rate": 0.0018822839515856091, + "loss": 0.1191, + "step": 16641 + }, + { + "epoch": 0.14446055155771217, + "grad_norm": 0.228515625, + "learning_rate": 0.0018822692350231151, + "loss": 0.0879, + "step": 16642 + }, + { + "epoch": 0.14446923203791634, + "grad_norm": 0.07373046875, + "learning_rate": 0.0018822545176051435, + "loss": 0.1035, + "step": 16643 + }, + { + "epoch": 0.1444779125181205, + "grad_norm": 0.095703125, + "learning_rate": 0.00188223979933171, + "loss": 0.1206, + "step": 16644 + }, + { + "epoch": 0.14448659299832467, + "grad_norm": 0.498046875, + "learning_rate": 0.0018822250802028307, + "loss": 0.1143, + "step": 16645 + }, + { + "epoch": 0.14449527347852884, + "grad_norm": 0.35546875, + "learning_rate": 0.001882210360218522, + "loss": 0.1289, + "step": 16646 + }, + { + "epoch": 0.144503953958733, + "grad_norm": 0.453125, + "learning_rate": 0.0018821956393787996, + "loss": 0.0933, + "step": 16647 + }, + { + "epoch": 0.14451263443893717, + "grad_norm": 1.125, + "learning_rate": 0.0018821809176836797, + "loss": 0.1406, + "step": 16648 + }, + { + "epoch": 0.14452131491914133, + "grad_norm": 0.2734375, + "learning_rate": 0.0018821661951331787, + "loss": 0.1152, + "step": 16649 + }, + { + "epoch": 0.1445299953993455, + "grad_norm": 0.375, + "learning_rate": 0.0018821514717273125, + "loss": 0.0938, + "step": 16650 + }, + { + "epoch": 0.14453867587954966, + "grad_norm": 0.50390625, + "learning_rate": 0.001882136747466097, + "loss": 0.1162, + "step": 16651 + }, + { + "epoch": 0.14454735635975383, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018821220223495486, + "loss": 0.1133, + "step": 16652 + }, + { + "epoch": 0.144556036839958, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018821072963776835, + "loss": 0.1108, + "step": 16653 + }, + { + "epoch": 0.14456471732016216, + "grad_norm": 0.0859375, + "learning_rate": 0.0018820925695505176, + "loss": 0.0967, + "step": 16654 + }, + { + "epoch": 0.14457339780036632, + "grad_norm": 0.2451171875, + "learning_rate": 0.0018820778418680668, + "loss": 0.0913, + "step": 16655 + }, + { + "epoch": 0.1445820782805705, + "grad_norm": 3.609375, + "learning_rate": 0.0018820631133303475, + "loss": 0.3047, + "step": 16656 + }, + { + "epoch": 0.14459075876077465, + "grad_norm": 0.76953125, + "learning_rate": 0.001882048383937376, + "loss": 0.1289, + "step": 16657 + }, + { + "epoch": 0.14459943924097882, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018820336536891679, + "loss": 0.1738, + "step": 16658 + }, + { + "epoch": 0.14460811972118298, + "grad_norm": 0.51171875, + "learning_rate": 0.0018820189225857394, + "loss": 0.1045, + "step": 16659 + }, + { + "epoch": 0.14461680020138715, + "grad_norm": 0.140625, + "learning_rate": 0.0018820041906271068, + "loss": 0.1348, + "step": 16660 + }, + { + "epoch": 0.1446254806815913, + "grad_norm": 0.36328125, + "learning_rate": 0.0018819894578132861, + "loss": 0.1113, + "step": 16661 + }, + { + "epoch": 0.14463416116179548, + "grad_norm": 0.283203125, + "learning_rate": 0.0018819747241442938, + "loss": 0.1211, + "step": 16662 + }, + { + "epoch": 0.14464284164199964, + "grad_norm": 0.416015625, + "learning_rate": 0.0018819599896201455, + "loss": 0.1484, + "step": 16663 + }, + { + "epoch": 0.1446515221222038, + "grad_norm": 0.33984375, + "learning_rate": 0.0018819452542408574, + "loss": 0.0957, + "step": 16664 + }, + { + "epoch": 0.14466020260240797, + "grad_norm": 0.333984375, + "learning_rate": 0.001881930518006446, + "loss": 0.1035, + "step": 16665 + }, + { + "epoch": 0.14466888308261214, + "grad_norm": 0.74609375, + "learning_rate": 0.0018819157809169269, + "loss": 0.1235, + "step": 16666 + }, + { + "epoch": 0.1446775635628163, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018819010429723167, + "loss": 0.1309, + "step": 16667 + }, + { + "epoch": 0.14468624404302047, + "grad_norm": 0.142578125, + "learning_rate": 0.001881886304172631, + "loss": 0.0796, + "step": 16668 + }, + { + "epoch": 0.14469492452322463, + "grad_norm": 0.84375, + "learning_rate": 0.0018818715645178862, + "loss": 0.126, + "step": 16669 + }, + { + "epoch": 0.1447036050034288, + "grad_norm": 0.138671875, + "learning_rate": 0.0018818568240080986, + "loss": 0.1299, + "step": 16670 + }, + { + "epoch": 0.14471228548363296, + "grad_norm": 0.205078125, + "learning_rate": 0.001881842082643284, + "loss": 0.1089, + "step": 16671 + }, + { + "epoch": 0.14472096596383713, + "grad_norm": 0.5546875, + "learning_rate": 0.0018818273404234587, + "loss": 0.1094, + "step": 16672 + }, + { + "epoch": 0.1447296464440413, + "grad_norm": 0.54296875, + "learning_rate": 0.0018818125973486387, + "loss": 0.0957, + "step": 16673 + }, + { + "epoch": 0.14473832692424546, + "grad_norm": 0.279296875, + "learning_rate": 0.0018817978534188403, + "loss": 0.1167, + "step": 16674 + }, + { + "epoch": 0.14474700740444962, + "grad_norm": 1.0, + "learning_rate": 0.0018817831086340793, + "loss": 0.1182, + "step": 16675 + }, + { + "epoch": 0.1447556878846538, + "grad_norm": 0.384765625, + "learning_rate": 0.0018817683629943722, + "loss": 0.1309, + "step": 16676 + }, + { + "epoch": 0.14476436836485795, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018817536164997348, + "loss": 0.1309, + "step": 16677 + }, + { + "epoch": 0.14477304884506212, + "grad_norm": 0.365234375, + "learning_rate": 0.0018817388691501837, + "loss": 0.1221, + "step": 16678 + }, + { + "epoch": 0.14478172932526628, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018817241209457343, + "loss": 0.1172, + "step": 16679 + }, + { + "epoch": 0.14479040980547045, + "grad_norm": 0.53125, + "learning_rate": 0.0018817093718864035, + "loss": 0.1221, + "step": 16680 + }, + { + "epoch": 0.14479909028567461, + "grad_norm": 0.322265625, + "learning_rate": 0.001881694621972207, + "loss": 0.1162, + "step": 16681 + }, + { + "epoch": 0.14480777076587878, + "grad_norm": 0.1533203125, + "learning_rate": 0.001881679871203161, + "loss": 0.1445, + "step": 16682 + }, + { + "epoch": 0.14481645124608294, + "grad_norm": 0.8828125, + "learning_rate": 0.0018816651195792816, + "loss": 0.0957, + "step": 16683 + }, + { + "epoch": 0.1448251317262871, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018816503671005849, + "loss": 0.1348, + "step": 16684 + }, + { + "epoch": 0.14483381220649127, + "grad_norm": 0.26171875, + "learning_rate": 0.001881635613767087, + "loss": 0.1123, + "step": 16685 + }, + { + "epoch": 0.14484249268669544, + "grad_norm": 0.8046875, + "learning_rate": 0.0018816208595788047, + "loss": 0.0942, + "step": 16686 + }, + { + "epoch": 0.1448511731668996, + "grad_norm": 0.275390625, + "learning_rate": 0.0018816061045357532, + "loss": 0.1172, + "step": 16687 + }, + { + "epoch": 0.14485985364710377, + "grad_norm": 0.6953125, + "learning_rate": 0.001881591348637949, + "loss": 0.1309, + "step": 16688 + }, + { + "epoch": 0.14486853412730794, + "grad_norm": 0.75, + "learning_rate": 0.001881576591885408, + "loss": 0.1172, + "step": 16689 + }, + { + "epoch": 0.1448772146075121, + "grad_norm": 0.275390625, + "learning_rate": 0.001881561834278147, + "loss": 0.1191, + "step": 16690 + }, + { + "epoch": 0.14488589508771627, + "grad_norm": 0.5234375, + "learning_rate": 0.0018815470758161814, + "loss": 0.1172, + "step": 16691 + }, + { + "epoch": 0.14489457556792043, + "grad_norm": 0.458984375, + "learning_rate": 0.001881532316499528, + "loss": 0.1387, + "step": 16692 + }, + { + "epoch": 0.1449032560481246, + "grad_norm": 0.287109375, + "learning_rate": 0.0018815175563282023, + "loss": 0.1338, + "step": 16693 + }, + { + "epoch": 0.14491193652832876, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018815027953022207, + "loss": 0.1196, + "step": 16694 + }, + { + "epoch": 0.1449206170085329, + "grad_norm": 0.2734375, + "learning_rate": 0.0018814880334215999, + "loss": 0.1055, + "step": 16695 + }, + { + "epoch": 0.14492929748873706, + "grad_norm": 0.59765625, + "learning_rate": 0.001881473270686355, + "loss": 0.1348, + "step": 16696 + }, + { + "epoch": 0.14493797796894123, + "grad_norm": 0.2578125, + "learning_rate": 0.001881458507096503, + "loss": 0.1328, + "step": 16697 + }, + { + "epoch": 0.1449466584491454, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018814437426520594, + "loss": 0.124, + "step": 16698 + }, + { + "epoch": 0.14495533892934956, + "grad_norm": 0.16015625, + "learning_rate": 0.0018814289773530406, + "loss": 0.1211, + "step": 16699 + }, + { + "epoch": 0.14496401940955372, + "grad_norm": 0.265625, + "learning_rate": 0.0018814142111994632, + "loss": 0.1069, + "step": 16700 + }, + { + "epoch": 0.1449726998897579, + "grad_norm": 0.29296875, + "learning_rate": 0.0018813994441913427, + "loss": 0.0942, + "step": 16701 + }, + { + "epoch": 0.14498138036996205, + "grad_norm": 0.267578125, + "learning_rate": 0.0018813846763286958, + "loss": 0.1123, + "step": 16702 + }, + { + "epoch": 0.14499006085016622, + "grad_norm": 0.1640625, + "learning_rate": 0.0018813699076115381, + "loss": 0.1221, + "step": 16703 + }, + { + "epoch": 0.14499874133037038, + "grad_norm": 0.275390625, + "learning_rate": 0.001881355138039886, + "loss": 0.1221, + "step": 16704 + }, + { + "epoch": 0.14500742181057455, + "grad_norm": 0.0771484375, + "learning_rate": 0.0018813403676137556, + "loss": 0.0752, + "step": 16705 + }, + { + "epoch": 0.14501610229077871, + "grad_norm": 0.423828125, + "learning_rate": 0.0018813255963331633, + "loss": 0.1182, + "step": 16706 + }, + { + "epoch": 0.14502478277098288, + "grad_norm": 0.2255859375, + "learning_rate": 0.001881310824198125, + "loss": 0.1055, + "step": 16707 + }, + { + "epoch": 0.14503346325118704, + "grad_norm": 0.22265625, + "learning_rate": 0.001881296051208657, + "loss": 0.1045, + "step": 16708 + }, + { + "epoch": 0.1450421437313912, + "grad_norm": 0.56640625, + "learning_rate": 0.001881281277364775, + "loss": 0.0806, + "step": 16709 + }, + { + "epoch": 0.14505082421159537, + "grad_norm": 0.263671875, + "learning_rate": 0.001881266502666496, + "loss": 0.0898, + "step": 16710 + }, + { + "epoch": 0.14505950469179954, + "grad_norm": 0.24609375, + "learning_rate": 0.001881251727113836, + "loss": 0.1348, + "step": 16711 + }, + { + "epoch": 0.1450681851720037, + "grad_norm": 0.306640625, + "learning_rate": 0.00188123695070681, + "loss": 0.1152, + "step": 16712 + }, + { + "epoch": 0.14507686565220787, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018812221734454356, + "loss": 0.0996, + "step": 16713 + }, + { + "epoch": 0.14508554613241204, + "grad_norm": 0.91796875, + "learning_rate": 0.0018812073953297281, + "loss": 0.2363, + "step": 16714 + }, + { + "epoch": 0.1450942266126162, + "grad_norm": 0.2216796875, + "learning_rate": 0.001881192616359704, + "loss": 0.1113, + "step": 16715 + }, + { + "epoch": 0.14510290709282037, + "grad_norm": 0.61328125, + "learning_rate": 0.001881177836535379, + "loss": 0.0996, + "step": 16716 + }, + { + "epoch": 0.14511158757302453, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018811630558567703, + "loss": 0.1123, + "step": 16717 + }, + { + "epoch": 0.1451202680532287, + "grad_norm": 0.095703125, + "learning_rate": 0.0018811482743238933, + "loss": 0.0815, + "step": 16718 + }, + { + "epoch": 0.14512894853343286, + "grad_norm": 0.1572265625, + "learning_rate": 0.001881133491936764, + "loss": 0.1348, + "step": 16719 + }, + { + "epoch": 0.14513762901363703, + "grad_norm": 1.0234375, + "learning_rate": 0.001881118708695399, + "loss": 0.1357, + "step": 16720 + }, + { + "epoch": 0.1451463094938412, + "grad_norm": 0.875, + "learning_rate": 0.0018811039245998143, + "loss": 0.103, + "step": 16721 + }, + { + "epoch": 0.14515498997404536, + "grad_norm": 0.1533203125, + "learning_rate": 0.001881089139650026, + "loss": 0.1348, + "step": 16722 + }, + { + "epoch": 0.14516367045424952, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018810743538460505, + "loss": 0.0996, + "step": 16723 + }, + { + "epoch": 0.1451723509344537, + "grad_norm": 0.6875, + "learning_rate": 0.0018810595671879037, + "loss": 0.0796, + "step": 16724 + }, + { + "epoch": 0.14518103141465785, + "grad_norm": 0.103515625, + "learning_rate": 0.001881044779675602, + "loss": 0.1104, + "step": 16725 + }, + { + "epoch": 0.14518971189486202, + "grad_norm": 0.400390625, + "learning_rate": 0.0018810299913091615, + "loss": 0.1221, + "step": 16726 + }, + { + "epoch": 0.14519839237506618, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018810152020885982, + "loss": 0.1426, + "step": 16727 + }, + { + "epoch": 0.14520707285527035, + "grad_norm": 0.29296875, + "learning_rate": 0.0018810004120139283, + "loss": 0.1143, + "step": 16728 + }, + { + "epoch": 0.1452157533354745, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018809856210851683, + "loss": 0.085, + "step": 16729 + }, + { + "epoch": 0.14522443381567868, + "grad_norm": 0.515625, + "learning_rate": 0.0018809708293023343, + "loss": 0.1299, + "step": 16730 + }, + { + "epoch": 0.14523311429588284, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018809560366654422, + "loss": 0.1133, + "step": 16731 + }, + { + "epoch": 0.145241794776087, + "grad_norm": 0.423828125, + "learning_rate": 0.0018809412431745078, + "loss": 0.0825, + "step": 16732 + }, + { + "epoch": 0.14525047525629117, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018809264488295486, + "loss": 0.1445, + "step": 16733 + }, + { + "epoch": 0.14525915573649534, + "grad_norm": 0.091796875, + "learning_rate": 0.0018809116536305794, + "loss": 0.1094, + "step": 16734 + }, + { + "epoch": 0.1452678362166995, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018808968575776174, + "loss": 0.0947, + "step": 16735 + }, + { + "epoch": 0.14527651669690367, + "grad_norm": 0.11865234375, + "learning_rate": 0.001880882060670678, + "loss": 0.1104, + "step": 16736 + }, + { + "epoch": 0.14528519717710783, + "grad_norm": 0.369140625, + "learning_rate": 0.0018808672629097775, + "loss": 0.0996, + "step": 16737 + }, + { + "epoch": 0.145293877657312, + "grad_norm": 0.73828125, + "learning_rate": 0.001880852464294933, + "loss": 0.1406, + "step": 16738 + }, + { + "epoch": 0.14530255813751616, + "grad_norm": 0.322265625, + "learning_rate": 0.0018808376648261594, + "loss": 0.1123, + "step": 16739 + }, + { + "epoch": 0.14531123861772033, + "grad_norm": 1.75, + "learning_rate": 0.0018808228645034739, + "loss": 0.3457, + "step": 16740 + }, + { + "epoch": 0.1453199190979245, + "grad_norm": 0.296875, + "learning_rate": 0.0018808080633268918, + "loss": 0.1221, + "step": 16741 + }, + { + "epoch": 0.14532859957812866, + "grad_norm": 0.482421875, + "learning_rate": 0.00188079326129643, + "loss": 0.1133, + "step": 16742 + }, + { + "epoch": 0.14533728005833282, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018807784584121041, + "loss": 0.0854, + "step": 16743 + }, + { + "epoch": 0.145345960538537, + "grad_norm": 0.13671875, + "learning_rate": 0.0018807636546739309, + "loss": 0.1172, + "step": 16744 + }, + { + "epoch": 0.14535464101874115, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018807488500819263, + "loss": 0.1216, + "step": 16745 + }, + { + "epoch": 0.14536332149894532, + "grad_norm": 0.53515625, + "learning_rate": 0.0018807340446361065, + "loss": 0.1309, + "step": 16746 + }, + { + "epoch": 0.14537200197914948, + "grad_norm": 0.22265625, + "learning_rate": 0.0018807192383364878, + "loss": 0.1016, + "step": 16747 + }, + { + "epoch": 0.14538068245935365, + "grad_norm": 0.78125, + "learning_rate": 0.001880704431183086, + "loss": 0.1138, + "step": 16748 + }, + { + "epoch": 0.14538936293955781, + "grad_norm": 0.171875, + "learning_rate": 0.0018806896231759174, + "loss": 0.1011, + "step": 16749 + }, + { + "epoch": 0.14539804341976198, + "grad_norm": 0.158203125, + "learning_rate": 0.0018806748143149988, + "loss": 0.127, + "step": 16750 + }, + { + "epoch": 0.14540672389996614, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018806600046003456, + "loss": 0.0986, + "step": 16751 + }, + { + "epoch": 0.1454154043801703, + "grad_norm": 0.0751953125, + "learning_rate": 0.001880645194031975, + "loss": 0.1045, + "step": 16752 + }, + { + "epoch": 0.14542408486037448, + "grad_norm": 0.158203125, + "learning_rate": 0.0018806303826099015, + "loss": 0.1084, + "step": 16753 + }, + { + "epoch": 0.14543276534057864, + "grad_norm": 0.16796875, + "learning_rate": 0.001880615570334143, + "loss": 0.1206, + "step": 16754 + }, + { + "epoch": 0.1454414458207828, + "grad_norm": 0.1005859375, + "learning_rate": 0.001880600757204715, + "loss": 0.1436, + "step": 16755 + }, + { + "epoch": 0.14545012630098697, + "grad_norm": 0.328125, + "learning_rate": 0.0018805859432216338, + "loss": 0.1318, + "step": 16756 + }, + { + "epoch": 0.14545880678119114, + "grad_norm": 0.5390625, + "learning_rate": 0.0018805711283849155, + "loss": 0.1406, + "step": 16757 + }, + { + "epoch": 0.1454674872613953, + "grad_norm": 0.6484375, + "learning_rate": 0.001880556312694576, + "loss": 0.1543, + "step": 16758 + }, + { + "epoch": 0.14547616774159947, + "grad_norm": 0.1787109375, + "learning_rate": 0.0018805414961506322, + "loss": 0.1191, + "step": 16759 + }, + { + "epoch": 0.14548484822180363, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018805266787530997, + "loss": 0.126, + "step": 16760 + }, + { + "epoch": 0.1454935287020078, + "grad_norm": 0.41015625, + "learning_rate": 0.001880511860501995, + "loss": 0.126, + "step": 16761 + }, + { + "epoch": 0.14550220918221196, + "grad_norm": 0.208984375, + "learning_rate": 0.0018804970413973346, + "loss": 0.1279, + "step": 16762 + }, + { + "epoch": 0.14551088966241613, + "grad_norm": 0.1552734375, + "learning_rate": 0.001880482221439134, + "loss": 0.1021, + "step": 16763 + }, + { + "epoch": 0.1455195701426203, + "grad_norm": 0.4609375, + "learning_rate": 0.00188046740062741, + "loss": 0.1436, + "step": 16764 + }, + { + "epoch": 0.14552825062282446, + "grad_norm": 0.59375, + "learning_rate": 0.0018804525789621785, + "loss": 0.0908, + "step": 16765 + }, + { + "epoch": 0.14553693110302862, + "grad_norm": 0.126953125, + "learning_rate": 0.0018804377564434556, + "loss": 0.1455, + "step": 16766 + }, + { + "epoch": 0.1455456115832328, + "grad_norm": 0.216796875, + "learning_rate": 0.001880422933071258, + "loss": 0.1348, + "step": 16767 + }, + { + "epoch": 0.14555429206343695, + "grad_norm": 0.08349609375, + "learning_rate": 0.0018804081088456012, + "loss": 0.125, + "step": 16768 + }, + { + "epoch": 0.14556297254364112, + "grad_norm": 0.115234375, + "learning_rate": 0.001880393283766502, + "loss": 0.0811, + "step": 16769 + }, + { + "epoch": 0.14557165302384528, + "grad_norm": 0.37109375, + "learning_rate": 0.001880378457833977, + "loss": 0.1084, + "step": 16770 + }, + { + "epoch": 0.14558033350404945, + "grad_norm": 0.5234375, + "learning_rate": 0.001880363631048041, + "loss": 0.1309, + "step": 16771 + }, + { + "epoch": 0.1455890139842536, + "grad_norm": 0.267578125, + "learning_rate": 0.0018803488034087115, + "loss": 0.1484, + "step": 16772 + }, + { + "epoch": 0.14559769446445778, + "grad_norm": 0.240234375, + "learning_rate": 0.0018803339749160041, + "loss": 0.0815, + "step": 16773 + }, + { + "epoch": 0.14560637494466194, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018803191455699355, + "loss": 0.1172, + "step": 16774 + }, + { + "epoch": 0.1456150554248661, + "grad_norm": 0.66796875, + "learning_rate": 0.0018803043153705213, + "loss": 0.1758, + "step": 16775 + }, + { + "epoch": 0.14562373590507027, + "grad_norm": 0.5, + "learning_rate": 0.0018802894843177778, + "loss": 0.1875, + "step": 16776 + }, + { + "epoch": 0.14563241638527444, + "grad_norm": 0.1533203125, + "learning_rate": 0.001880274652411722, + "loss": 0.0869, + "step": 16777 + }, + { + "epoch": 0.1456410968654786, + "grad_norm": 0.48828125, + "learning_rate": 0.0018802598196523692, + "loss": 0.0996, + "step": 16778 + }, + { + "epoch": 0.14564977734568277, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018802449860397363, + "loss": 0.1108, + "step": 16779 + }, + { + "epoch": 0.14565845782588693, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018802301515738389, + "loss": 0.1113, + "step": 16780 + }, + { + "epoch": 0.1456671383060911, + "grad_norm": 2.140625, + "learning_rate": 0.0018802153162546936, + "loss": 0.1865, + "step": 16781 + }, + { + "epoch": 0.14567581878629526, + "grad_norm": 0.3515625, + "learning_rate": 0.0018802004800823164, + "loss": 0.1123, + "step": 16782 + }, + { + "epoch": 0.14568449926649943, + "grad_norm": 0.125, + "learning_rate": 0.001880185643056724, + "loss": 0.1162, + "step": 16783 + }, + { + "epoch": 0.1456931797467036, + "grad_norm": 0.1533203125, + "learning_rate": 0.001880170805177932, + "loss": 0.1299, + "step": 16784 + }, + { + "epoch": 0.14570186022690776, + "grad_norm": 0.31640625, + "learning_rate": 0.001880155966445957, + "loss": 0.1025, + "step": 16785 + }, + { + "epoch": 0.14571054070711192, + "grad_norm": 0.65234375, + "learning_rate": 0.0018801411268608156, + "loss": 0.1631, + "step": 16786 + }, + { + "epoch": 0.1457192211873161, + "grad_norm": 0.2265625, + "learning_rate": 0.001880126286422523, + "loss": 0.1084, + "step": 16787 + }, + { + "epoch": 0.14572790166752025, + "grad_norm": 0.373046875, + "learning_rate": 0.0018801114451310963, + "loss": 0.1309, + "step": 16788 + }, + { + "epoch": 0.14573658214772442, + "grad_norm": 0.345703125, + "learning_rate": 0.0018800966029865515, + "loss": 0.1914, + "step": 16789 + }, + { + "epoch": 0.14574526262792858, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018800817599889047, + "loss": 0.0908, + "step": 16790 + }, + { + "epoch": 0.14575394310813275, + "grad_norm": 0.380859375, + "learning_rate": 0.0018800669161381722, + "loss": 0.0811, + "step": 16791 + }, + { + "epoch": 0.14576262358833691, + "grad_norm": 0.484375, + "learning_rate": 0.00188005207143437, + "loss": 0.1328, + "step": 16792 + }, + { + "epoch": 0.14577130406854108, + "grad_norm": 0.58203125, + "learning_rate": 0.0018800372258775148, + "loss": 0.165, + "step": 16793 + }, + { + "epoch": 0.14577998454874525, + "grad_norm": 0.439453125, + "learning_rate": 0.0018800223794676228, + "loss": 0.1108, + "step": 16794 + }, + { + "epoch": 0.1457886650289494, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018800075322047097, + "loss": 0.0967, + "step": 16795 + }, + { + "epoch": 0.14579734550915358, + "grad_norm": 0.248046875, + "learning_rate": 0.0018799926840887922, + "loss": 0.1133, + "step": 16796 + }, + { + "epoch": 0.14580602598935774, + "grad_norm": 0.484375, + "learning_rate": 0.0018799778351198863, + "loss": 0.1826, + "step": 16797 + }, + { + "epoch": 0.1458147064695619, + "grad_norm": 0.296875, + "learning_rate": 0.0018799629852980088, + "loss": 0.1338, + "step": 16798 + }, + { + "epoch": 0.14582338694976607, + "grad_norm": 0.1552734375, + "learning_rate": 0.001879948134623175, + "loss": 0.0874, + "step": 16799 + }, + { + "epoch": 0.14583206742997024, + "grad_norm": 0.6171875, + "learning_rate": 0.0018799332830954018, + "loss": 0.1406, + "step": 16800 + }, + { + "epoch": 0.1458407479101744, + "grad_norm": 0.3828125, + "learning_rate": 0.0018799184307147054, + "loss": 0.1357, + "step": 16801 + }, + { + "epoch": 0.14584942839037857, + "grad_norm": 0.21484375, + "learning_rate": 0.001879903577481102, + "loss": 0.0981, + "step": 16802 + }, + { + "epoch": 0.14585810887058273, + "grad_norm": 0.11328125, + "learning_rate": 0.0018798887233946076, + "loss": 0.1182, + "step": 16803 + }, + { + "epoch": 0.1458667893507869, + "grad_norm": 0.298828125, + "learning_rate": 0.0018798738684552385, + "loss": 0.0947, + "step": 16804 + }, + { + "epoch": 0.14587546983099106, + "grad_norm": 0.259765625, + "learning_rate": 0.0018798590126630113, + "loss": 0.0981, + "step": 16805 + }, + { + "epoch": 0.14588415031119523, + "grad_norm": 0.6171875, + "learning_rate": 0.0018798441560179417, + "loss": 0.1221, + "step": 16806 + }, + { + "epoch": 0.1458928307913994, + "grad_norm": 0.31640625, + "learning_rate": 0.0018798292985200465, + "loss": 0.0991, + "step": 16807 + }, + { + "epoch": 0.14590151127160356, + "grad_norm": 0.54296875, + "learning_rate": 0.0018798144401693416, + "loss": 0.1196, + "step": 16808 + }, + { + "epoch": 0.14591019175180772, + "grad_norm": 0.1328125, + "learning_rate": 0.0018797995809658435, + "loss": 0.105, + "step": 16809 + }, + { + "epoch": 0.1459188722320119, + "grad_norm": 0.1748046875, + "learning_rate": 0.001879784720909568, + "loss": 0.1172, + "step": 16810 + }, + { + "epoch": 0.14592755271221605, + "grad_norm": 0.578125, + "learning_rate": 0.0018797698600005318, + "loss": 0.1143, + "step": 16811 + }, + { + "epoch": 0.14593623319242022, + "grad_norm": 0.27734375, + "learning_rate": 0.0018797549982387512, + "loss": 0.1758, + "step": 16812 + }, + { + "epoch": 0.14594491367262438, + "grad_norm": 0.36328125, + "learning_rate": 0.001879740135624242, + "loss": 0.1445, + "step": 16813 + }, + { + "epoch": 0.14595359415282855, + "grad_norm": 0.6015625, + "learning_rate": 0.0018797252721570207, + "loss": 0.0981, + "step": 16814 + }, + { + "epoch": 0.1459622746330327, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018797104078371034, + "loss": 0.124, + "step": 16815 + }, + { + "epoch": 0.14597095511323688, + "grad_norm": 0.326171875, + "learning_rate": 0.001879695542664507, + "loss": 0.1045, + "step": 16816 + }, + { + "epoch": 0.14597963559344104, + "grad_norm": 0.474609375, + "learning_rate": 0.001879680676639247, + "loss": 0.1367, + "step": 16817 + }, + { + "epoch": 0.1459883160736452, + "grad_norm": 0.201171875, + "learning_rate": 0.00187966580976134, + "loss": 0.1089, + "step": 16818 + }, + { + "epoch": 0.14599699655384935, + "grad_norm": 0.1435546875, + "learning_rate": 0.001879650942030802, + "loss": 0.1201, + "step": 16819 + }, + { + "epoch": 0.1460056770340535, + "grad_norm": 0.4296875, + "learning_rate": 0.0018796360734476497, + "loss": 0.1016, + "step": 16820 + }, + { + "epoch": 0.14601435751425768, + "grad_norm": 0.455078125, + "learning_rate": 0.0018796212040118987, + "loss": 0.1143, + "step": 16821 + }, + { + "epoch": 0.14602303799446184, + "grad_norm": 0.34765625, + "learning_rate": 0.0018796063337235657, + "loss": 0.126, + "step": 16822 + }, + { + "epoch": 0.146031718474666, + "grad_norm": 0.6171875, + "learning_rate": 0.0018795914625826676, + "loss": 0.1084, + "step": 16823 + }, + { + "epoch": 0.14604039895487017, + "grad_norm": 0.39453125, + "learning_rate": 0.0018795765905892192, + "loss": 0.126, + "step": 16824 + }, + { + "epoch": 0.14604907943507434, + "grad_norm": 0.2099609375, + "learning_rate": 0.001879561717743238, + "loss": 0.1348, + "step": 16825 + }, + { + "epoch": 0.1460577599152785, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018795468440447399, + "loss": 0.1494, + "step": 16826 + }, + { + "epoch": 0.14606644039548267, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018795319694937408, + "loss": 0.123, + "step": 16827 + }, + { + "epoch": 0.14607512087568683, + "grad_norm": 0.384765625, + "learning_rate": 0.0018795170940902573, + "loss": 0.126, + "step": 16828 + }, + { + "epoch": 0.146083801355891, + "grad_norm": 0.703125, + "learning_rate": 0.0018795022178343058, + "loss": 0.0947, + "step": 16829 + }, + { + "epoch": 0.14609248183609516, + "grad_norm": 0.5390625, + "learning_rate": 0.0018794873407259021, + "loss": 0.123, + "step": 16830 + }, + { + "epoch": 0.14610116231629933, + "grad_norm": 0.058837890625, + "learning_rate": 0.0018794724627650627, + "loss": 0.085, + "step": 16831 + }, + { + "epoch": 0.1461098427965035, + "grad_norm": 0.271484375, + "learning_rate": 0.0018794575839518042, + "loss": 0.1123, + "step": 16832 + }, + { + "epoch": 0.14611852327670766, + "grad_norm": 0.3984375, + "learning_rate": 0.0018794427042861427, + "loss": 0.1338, + "step": 16833 + }, + { + "epoch": 0.14612720375691182, + "grad_norm": 0.333984375, + "learning_rate": 0.001879427823768094, + "loss": 0.0703, + "step": 16834 + }, + { + "epoch": 0.146135884237116, + "grad_norm": 0.23828125, + "learning_rate": 0.001879412942397675, + "loss": 0.1064, + "step": 16835 + }, + { + "epoch": 0.14614456471732015, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018793980601749017, + "loss": 0.1055, + "step": 16836 + }, + { + "epoch": 0.14615324519752432, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018793831770997904, + "loss": 0.0791, + "step": 16837 + }, + { + "epoch": 0.14616192567772848, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018793682931723572, + "loss": 0.1387, + "step": 16838 + }, + { + "epoch": 0.14617060615793265, + "grad_norm": 0.220703125, + "learning_rate": 0.0018793534083926186, + "loss": 0.0967, + "step": 16839 + }, + { + "epoch": 0.1461792866381368, + "grad_norm": 0.74609375, + "learning_rate": 0.001879338522760591, + "loss": 0.1226, + "step": 16840 + }, + { + "epoch": 0.14618796711834098, + "grad_norm": 0.15625, + "learning_rate": 0.0018793236362762904, + "loss": 0.1167, + "step": 16841 + }, + { + "epoch": 0.14619664759854514, + "grad_norm": 0.47265625, + "learning_rate": 0.001879308748939733, + "loss": 0.1396, + "step": 16842 + }, + { + "epoch": 0.1462053280787493, + "grad_norm": 0.6796875, + "learning_rate": 0.0018792938607509356, + "loss": 0.0903, + "step": 16843 + }, + { + "epoch": 0.14621400855895347, + "grad_norm": 0.94921875, + "learning_rate": 0.001879278971709914, + "loss": 0.1279, + "step": 16844 + }, + { + "epoch": 0.14622268903915764, + "grad_norm": 0.55078125, + "learning_rate": 0.0018792640818166846, + "loss": 0.1348, + "step": 16845 + }, + { + "epoch": 0.1462313695193618, + "grad_norm": 0.58984375, + "learning_rate": 0.0018792491910712638, + "loss": 0.0859, + "step": 16846 + }, + { + "epoch": 0.14624004999956597, + "grad_norm": 0.62890625, + "learning_rate": 0.0018792342994736677, + "loss": 0.1064, + "step": 16847 + }, + { + "epoch": 0.14624873047977013, + "grad_norm": 0.34375, + "learning_rate": 0.0018792194070239125, + "loss": 0.1226, + "step": 16848 + }, + { + "epoch": 0.1462574109599743, + "grad_norm": 0.10693359375, + "learning_rate": 0.001879204513722015, + "loss": 0.085, + "step": 16849 + }, + { + "epoch": 0.14626609144017846, + "grad_norm": 0.1484375, + "learning_rate": 0.001879189619567991, + "loss": 0.1196, + "step": 16850 + }, + { + "epoch": 0.14627477192038263, + "grad_norm": 0.53125, + "learning_rate": 0.0018791747245618566, + "loss": 0.1494, + "step": 16851 + }, + { + "epoch": 0.1462834524005868, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018791598287036292, + "loss": 0.0972, + "step": 16852 + }, + { + "epoch": 0.14629213288079096, + "grad_norm": 0.09375, + "learning_rate": 0.0018791449319933238, + "loss": 0.1064, + "step": 16853 + }, + { + "epoch": 0.14630081336099512, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018791300344309571, + "loss": 0.0923, + "step": 16854 + }, + { + "epoch": 0.1463094938411993, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018791151360165457, + "loss": 0.1396, + "step": 16855 + }, + { + "epoch": 0.14631817432140345, + "grad_norm": 0.6328125, + "learning_rate": 0.0018791002367501057, + "loss": 0.1201, + "step": 16856 + }, + { + "epoch": 0.14632685480160762, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018790853366316533, + "loss": 0.1011, + "step": 16857 + }, + { + "epoch": 0.14633553528181178, + "grad_norm": 0.349609375, + "learning_rate": 0.001879070435661205, + "loss": 0.1816, + "step": 16858 + }, + { + "epoch": 0.14634421576201595, + "grad_norm": 0.390625, + "learning_rate": 0.0018790555338387771, + "loss": 0.1445, + "step": 16859 + }, + { + "epoch": 0.14635289624222012, + "grad_norm": 0.52734375, + "learning_rate": 0.0018790406311643853, + "loss": 0.1143, + "step": 16860 + }, + { + "epoch": 0.14636157672242428, + "grad_norm": 0.9296875, + "learning_rate": 0.0018790257276380467, + "loss": 0.1377, + "step": 16861 + }, + { + "epoch": 0.14637025720262845, + "grad_norm": 0.388671875, + "learning_rate": 0.0018790108232597774, + "loss": 0.0967, + "step": 16862 + }, + { + "epoch": 0.1463789376828326, + "grad_norm": 1.15625, + "learning_rate": 0.0018789959180295934, + "loss": 0.1152, + "step": 16863 + }, + { + "epoch": 0.14638761816303678, + "grad_norm": 0.6796875, + "learning_rate": 0.0018789810119475112, + "loss": 0.1035, + "step": 16864 + }, + { + "epoch": 0.14639629864324094, + "grad_norm": 0.46484375, + "learning_rate": 0.0018789661050135468, + "loss": 0.1299, + "step": 16865 + }, + { + "epoch": 0.1464049791234451, + "grad_norm": 0.189453125, + "learning_rate": 0.001878951197227717, + "loss": 0.1348, + "step": 16866 + }, + { + "epoch": 0.14641365960364927, + "grad_norm": 0.26953125, + "learning_rate": 0.001878936288590038, + "loss": 0.1206, + "step": 16867 + }, + { + "epoch": 0.14642234008385344, + "grad_norm": 0.1025390625, + "learning_rate": 0.001878921379100526, + "loss": 0.1016, + "step": 16868 + }, + { + "epoch": 0.1464310205640576, + "grad_norm": 0.361328125, + "learning_rate": 0.0018789064687591973, + "loss": 0.0801, + "step": 16869 + }, + { + "epoch": 0.14643970104426177, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018788915575660677, + "loss": 0.125, + "step": 16870 + }, + { + "epoch": 0.14644838152446593, + "grad_norm": 0.56640625, + "learning_rate": 0.0018788766455211544, + "loss": 0.1396, + "step": 16871 + }, + { + "epoch": 0.1464570620046701, + "grad_norm": 0.84375, + "learning_rate": 0.0018788617326244731, + "loss": 0.0859, + "step": 16872 + }, + { + "epoch": 0.14646574248487426, + "grad_norm": 0.53125, + "learning_rate": 0.0018788468188760405, + "loss": 0.1084, + "step": 16873 + }, + { + "epoch": 0.14647442296507843, + "grad_norm": 0.34765625, + "learning_rate": 0.0018788319042758728, + "loss": 0.0854, + "step": 16874 + }, + { + "epoch": 0.1464831034452826, + "grad_norm": 3.109375, + "learning_rate": 0.001878816988823986, + "loss": 0.5625, + "step": 16875 + }, + { + "epoch": 0.14649178392548676, + "grad_norm": 0.1796875, + "learning_rate": 0.0018788020725203968, + "loss": 0.0879, + "step": 16876 + }, + { + "epoch": 0.14650046440569092, + "grad_norm": 0.296875, + "learning_rate": 0.0018787871553651212, + "loss": 0.0889, + "step": 16877 + }, + { + "epoch": 0.1465091448858951, + "grad_norm": 0.197265625, + "learning_rate": 0.001878772237358176, + "loss": 0.166, + "step": 16878 + }, + { + "epoch": 0.14651782536609925, + "grad_norm": 0.201171875, + "learning_rate": 0.0018787573184995768, + "loss": 0.1006, + "step": 16879 + }, + { + "epoch": 0.14652650584630342, + "grad_norm": 1.3671875, + "learning_rate": 0.0018787423987893406, + "loss": 0.1177, + "step": 16880 + }, + { + "epoch": 0.14653518632650758, + "grad_norm": 0.26953125, + "learning_rate": 0.001878727478227483, + "loss": 0.0972, + "step": 16881 + }, + { + "epoch": 0.14654386680671175, + "grad_norm": 0.09375, + "learning_rate": 0.0018787125568140212, + "loss": 0.1016, + "step": 16882 + }, + { + "epoch": 0.1465525472869159, + "grad_norm": 0.240234375, + "learning_rate": 0.0018786976345489706, + "loss": 0.1299, + "step": 16883 + }, + { + "epoch": 0.14656122776712008, + "grad_norm": 0.71484375, + "learning_rate": 0.0018786827114323484, + "loss": 0.0967, + "step": 16884 + }, + { + "epoch": 0.14656990824732424, + "grad_norm": 0.6875, + "learning_rate": 0.0018786677874641704, + "loss": 0.1562, + "step": 16885 + }, + { + "epoch": 0.1465785887275284, + "grad_norm": 0.341796875, + "learning_rate": 0.0018786528626444528, + "loss": 0.1011, + "step": 16886 + }, + { + "epoch": 0.14658726920773257, + "grad_norm": 0.54296875, + "learning_rate": 0.001878637936973212, + "loss": 0.1201, + "step": 16887 + }, + { + "epoch": 0.14659594968793674, + "grad_norm": 0.609375, + "learning_rate": 0.0018786230104504648, + "loss": 0.0801, + "step": 16888 + }, + { + "epoch": 0.1466046301681409, + "grad_norm": 0.0888671875, + "learning_rate": 0.001878608083076227, + "loss": 0.1143, + "step": 16889 + }, + { + "epoch": 0.14661331064834507, + "grad_norm": 0.51953125, + "learning_rate": 0.0018785931548505152, + "loss": 0.1074, + "step": 16890 + }, + { + "epoch": 0.14662199112854923, + "grad_norm": 0.103515625, + "learning_rate": 0.0018785782257733454, + "loss": 0.1387, + "step": 16891 + }, + { + "epoch": 0.1466306716087534, + "grad_norm": 0.9140625, + "learning_rate": 0.0018785632958447345, + "loss": 0.1133, + "step": 16892 + }, + { + "epoch": 0.14663935208895756, + "grad_norm": 0.9453125, + "learning_rate": 0.0018785483650646982, + "loss": 0.1504, + "step": 16893 + }, + { + "epoch": 0.14664803256916173, + "grad_norm": 0.158203125, + "learning_rate": 0.0018785334334332534, + "loss": 0.1836, + "step": 16894 + }, + { + "epoch": 0.1466567130493659, + "grad_norm": 0.302734375, + "learning_rate": 0.001878518500950416, + "loss": 0.1157, + "step": 16895 + }, + { + "epoch": 0.14666539352957006, + "grad_norm": 0.2734375, + "learning_rate": 0.0018785035676162024, + "loss": 0.1055, + "step": 16896 + }, + { + "epoch": 0.14667407400977422, + "grad_norm": 0.060302734375, + "learning_rate": 0.001878488633430629, + "loss": 0.105, + "step": 16897 + }, + { + "epoch": 0.1466827544899784, + "grad_norm": 0.53515625, + "learning_rate": 0.001878473698393712, + "loss": 0.1387, + "step": 16898 + }, + { + "epoch": 0.14669143497018255, + "grad_norm": 0.490234375, + "learning_rate": 0.0018784587625054678, + "loss": 0.1196, + "step": 16899 + }, + { + "epoch": 0.14670011545038672, + "grad_norm": 0.609375, + "learning_rate": 0.0018784438257659132, + "loss": 0.1318, + "step": 16900 + }, + { + "epoch": 0.14670879593059089, + "grad_norm": 0.3828125, + "learning_rate": 0.0018784288881750638, + "loss": 0.1221, + "step": 16901 + }, + { + "epoch": 0.14671747641079505, + "grad_norm": 0.3984375, + "learning_rate": 0.0018784139497329364, + "loss": 0.0913, + "step": 16902 + }, + { + "epoch": 0.14672615689099922, + "grad_norm": 0.361328125, + "learning_rate": 0.0018783990104395472, + "loss": 0.1592, + "step": 16903 + }, + { + "epoch": 0.14673483737120338, + "grad_norm": 1.25, + "learning_rate": 0.0018783840702949126, + "loss": 0.1582, + "step": 16904 + }, + { + "epoch": 0.14674351785140755, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018783691292990489, + "loss": 0.1187, + "step": 16905 + }, + { + "epoch": 0.1467521983316117, + "grad_norm": 0.2314453125, + "learning_rate": 0.001878354187451972, + "loss": 0.1299, + "step": 16906 + }, + { + "epoch": 0.14676087881181588, + "grad_norm": 1.265625, + "learning_rate": 0.001878339244753699, + "loss": 0.1152, + "step": 16907 + }, + { + "epoch": 0.14676955929202004, + "grad_norm": 0.8046875, + "learning_rate": 0.001878324301204246, + "loss": 0.0859, + "step": 16908 + }, + { + "epoch": 0.1467782397722242, + "grad_norm": 0.369140625, + "learning_rate": 0.0018783093568036288, + "loss": 0.1328, + "step": 16909 + }, + { + "epoch": 0.14678692025242837, + "grad_norm": 0.20703125, + "learning_rate": 0.0018782944115518648, + "loss": 0.0996, + "step": 16910 + }, + { + "epoch": 0.14679560073263254, + "grad_norm": 0.2177734375, + "learning_rate": 0.0018782794654489694, + "loss": 0.1357, + "step": 16911 + }, + { + "epoch": 0.1468042812128367, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018782645184949591, + "loss": 0.1079, + "step": 16912 + }, + { + "epoch": 0.14681296169304087, + "grad_norm": 0.828125, + "learning_rate": 0.0018782495706898509, + "loss": 0.125, + "step": 16913 + }, + { + "epoch": 0.14682164217324503, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018782346220336602, + "loss": 0.127, + "step": 16914 + }, + { + "epoch": 0.1468303226534492, + "grad_norm": 0.58984375, + "learning_rate": 0.001878219672526404, + "loss": 0.1465, + "step": 16915 + }, + { + "epoch": 0.14683900313365336, + "grad_norm": 0.86328125, + "learning_rate": 0.0018782047221680982, + "loss": 0.3555, + "step": 16916 + }, + { + "epoch": 0.14684768361385753, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018781897709587599, + "loss": 0.1055, + "step": 16917 + }, + { + "epoch": 0.1468563640940617, + "grad_norm": 0.47265625, + "learning_rate": 0.0018781748188984044, + "loss": 0.0771, + "step": 16918 + }, + { + "epoch": 0.14686504457426586, + "grad_norm": 0.5859375, + "learning_rate": 0.001878159865987049, + "loss": 0.1152, + "step": 16919 + }, + { + "epoch": 0.14687372505447002, + "grad_norm": 0.181640625, + "learning_rate": 0.0018781449122247096, + "loss": 0.1377, + "step": 16920 + }, + { + "epoch": 0.1468824055346742, + "grad_norm": 0.07958984375, + "learning_rate": 0.0018781299576114024, + "loss": 0.1006, + "step": 16921 + }, + { + "epoch": 0.14689108601487835, + "grad_norm": 0.1689453125, + "learning_rate": 0.0018781150021471443, + "loss": 0.123, + "step": 16922 + }, + { + "epoch": 0.14689976649508252, + "grad_norm": 0.330078125, + "learning_rate": 0.001878100045831951, + "loss": 0.1309, + "step": 16923 + }, + { + "epoch": 0.14690844697528668, + "grad_norm": 0.279296875, + "learning_rate": 0.0018780850886658395, + "loss": 0.1221, + "step": 16924 + }, + { + "epoch": 0.14691712745549085, + "grad_norm": 0.451171875, + "learning_rate": 0.0018780701306488254, + "loss": 0.1406, + "step": 16925 + }, + { + "epoch": 0.146925807935695, + "grad_norm": 0.427734375, + "learning_rate": 0.001878055171780926, + "loss": 0.1157, + "step": 16926 + }, + { + "epoch": 0.14693448841589918, + "grad_norm": 0.69921875, + "learning_rate": 0.0018780402120621572, + "loss": 0.1621, + "step": 16927 + }, + { + "epoch": 0.14694316889610334, + "grad_norm": 0.59765625, + "learning_rate": 0.0018780252514925347, + "loss": 0.1162, + "step": 16928 + }, + { + "epoch": 0.1469518493763075, + "grad_norm": 0.71484375, + "learning_rate": 0.001878010290072076, + "loss": 0.1143, + "step": 16929 + }, + { + "epoch": 0.14696052985651167, + "grad_norm": 0.26171875, + "learning_rate": 0.0018779953278007966, + "loss": 0.1123, + "step": 16930 + }, + { + "epoch": 0.14696921033671584, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018779803646787135, + "loss": 0.1299, + "step": 16931 + }, + { + "epoch": 0.14697789081692, + "grad_norm": 0.396484375, + "learning_rate": 0.0018779654007058427, + "loss": 0.0835, + "step": 16932 + }, + { + "epoch": 0.14698657129712417, + "grad_norm": 0.333984375, + "learning_rate": 0.0018779504358822005, + "loss": 0.1216, + "step": 16933 + }, + { + "epoch": 0.14699525177732833, + "grad_norm": 1.0859375, + "learning_rate": 0.0018779354702078033, + "loss": 0.1143, + "step": 16934 + }, + { + "epoch": 0.1470039322575325, + "grad_norm": 0.18359375, + "learning_rate": 0.0018779205036826676, + "loss": 0.1426, + "step": 16935 + }, + { + "epoch": 0.14701261273773666, + "grad_norm": 0.51171875, + "learning_rate": 0.00187790553630681, + "loss": 0.1152, + "step": 16936 + }, + { + "epoch": 0.14702129321794083, + "grad_norm": 0.384765625, + "learning_rate": 0.0018778905680802464, + "loss": 0.0991, + "step": 16937 + }, + { + "epoch": 0.147029973698145, + "grad_norm": 0.298828125, + "learning_rate": 0.0018778755990029935, + "loss": 0.1543, + "step": 16938 + }, + { + "epoch": 0.14703865417834916, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018778606290750673, + "loss": 0.1934, + "step": 16939 + }, + { + "epoch": 0.14704733465855332, + "grad_norm": 0.19140625, + "learning_rate": 0.0018778456582964845, + "loss": 0.1035, + "step": 16940 + }, + { + "epoch": 0.1470560151387575, + "grad_norm": 0.51171875, + "learning_rate": 0.0018778306866672613, + "loss": 0.1992, + "step": 16941 + }, + { + "epoch": 0.14706469561896163, + "grad_norm": 0.671875, + "learning_rate": 0.0018778157141874144, + "loss": 0.2188, + "step": 16942 + }, + { + "epoch": 0.1470733760991658, + "grad_norm": 0.23046875, + "learning_rate": 0.0018778007408569598, + "loss": 0.1191, + "step": 16943 + }, + { + "epoch": 0.14708205657936996, + "grad_norm": 1.25, + "learning_rate": 0.001877785766675914, + "loss": 0.1436, + "step": 16944 + }, + { + "epoch": 0.14709073705957412, + "grad_norm": 0.26953125, + "learning_rate": 0.0018777707916442933, + "loss": 0.1182, + "step": 16945 + }, + { + "epoch": 0.1470994175397783, + "grad_norm": 0.2451171875, + "learning_rate": 0.001877755815762114, + "loss": 0.1143, + "step": 16946 + }, + { + "epoch": 0.14710809801998245, + "grad_norm": 0.1845703125, + "learning_rate": 0.0018777408390293928, + "loss": 0.1309, + "step": 16947 + }, + { + "epoch": 0.14711677850018662, + "grad_norm": 0.291015625, + "learning_rate": 0.001877725861446146, + "loss": 0.1338, + "step": 16948 + }, + { + "epoch": 0.14712545898039078, + "grad_norm": 0.224609375, + "learning_rate": 0.0018777108830123899, + "loss": 0.0991, + "step": 16949 + }, + { + "epoch": 0.14713413946059495, + "grad_norm": 0.333984375, + "learning_rate": 0.0018776959037281405, + "loss": 0.083, + "step": 16950 + }, + { + "epoch": 0.1471428199407991, + "grad_norm": 0.51953125, + "learning_rate": 0.001877680923593415, + "loss": 0.1543, + "step": 16951 + }, + { + "epoch": 0.14715150042100328, + "grad_norm": 0.205078125, + "learning_rate": 0.001877665942608229, + "loss": 0.1338, + "step": 16952 + }, + { + "epoch": 0.14716018090120744, + "grad_norm": 0.271484375, + "learning_rate": 0.0018776509607725993, + "loss": 0.1357, + "step": 16953 + }, + { + "epoch": 0.1471688613814116, + "grad_norm": 0.19140625, + "learning_rate": 0.0018776359780865424, + "loss": 0.1123, + "step": 16954 + }, + { + "epoch": 0.14717754186161577, + "grad_norm": 0.4296875, + "learning_rate": 0.0018776209945500742, + "loss": 0.1289, + "step": 16955 + }, + { + "epoch": 0.14718622234181994, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018776060101632116, + "loss": 0.1011, + "step": 16956 + }, + { + "epoch": 0.1471949028220241, + "grad_norm": 0.92578125, + "learning_rate": 0.0018775910249259704, + "loss": 0.0991, + "step": 16957 + }, + { + "epoch": 0.14720358330222827, + "grad_norm": 0.318359375, + "learning_rate": 0.0018775760388383676, + "loss": 0.1084, + "step": 16958 + }, + { + "epoch": 0.14721226378243243, + "grad_norm": 0.07958984375, + "learning_rate": 0.001877561051900419, + "loss": 0.1426, + "step": 16959 + }, + { + "epoch": 0.1472209442626366, + "grad_norm": 0.27734375, + "learning_rate": 0.001877546064112142, + "loss": 0.1123, + "step": 16960 + }, + { + "epoch": 0.14722962474284076, + "grad_norm": 0.166015625, + "learning_rate": 0.0018775310754735517, + "loss": 0.1416, + "step": 16961 + }, + { + "epoch": 0.14723830522304493, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018775160859846656, + "loss": 0.1484, + "step": 16962 + }, + { + "epoch": 0.1472469857032491, + "grad_norm": 0.275390625, + "learning_rate": 0.0018775010956454993, + "loss": 0.0854, + "step": 16963 + }, + { + "epoch": 0.14725566618345326, + "grad_norm": 0.384765625, + "learning_rate": 0.0018774861044560693, + "loss": 0.1191, + "step": 16964 + }, + { + "epoch": 0.14726434666365742, + "grad_norm": 0.267578125, + "learning_rate": 0.0018774711124163925, + "loss": 0.1016, + "step": 16965 + }, + { + "epoch": 0.1472730271438616, + "grad_norm": 0.326171875, + "learning_rate": 0.0018774561195264849, + "loss": 0.1123, + "step": 16966 + }, + { + "epoch": 0.14728170762406576, + "grad_norm": 0.404296875, + "learning_rate": 0.0018774411257863625, + "loss": 0.1367, + "step": 16967 + }, + { + "epoch": 0.14729038810426992, + "grad_norm": 0.263671875, + "learning_rate": 0.0018774261311960427, + "loss": 0.124, + "step": 16968 + }, + { + "epoch": 0.14729906858447409, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018774111357555412, + "loss": 0.1299, + "step": 16969 + }, + { + "epoch": 0.14730774906467825, + "grad_norm": 0.2275390625, + "learning_rate": 0.0018773961394648747, + "loss": 0.1455, + "step": 16970 + }, + { + "epoch": 0.14731642954488242, + "grad_norm": 0.66015625, + "learning_rate": 0.0018773811423240595, + "loss": 0.1279, + "step": 16971 + }, + { + "epoch": 0.14732511002508658, + "grad_norm": 0.2578125, + "learning_rate": 0.0018773661443331115, + "loss": 0.0947, + "step": 16972 + }, + { + "epoch": 0.14733379050529075, + "grad_norm": 0.12158203125, + "learning_rate": 0.001877351145492048, + "loss": 0.1206, + "step": 16973 + }, + { + "epoch": 0.1473424709854949, + "grad_norm": 0.21875, + "learning_rate": 0.0018773361458008849, + "loss": 0.104, + "step": 16974 + }, + { + "epoch": 0.14735115146569908, + "grad_norm": 0.40625, + "learning_rate": 0.0018773211452596387, + "loss": 0.1045, + "step": 16975 + }, + { + "epoch": 0.14735983194590324, + "grad_norm": 0.16015625, + "learning_rate": 0.0018773061438683257, + "loss": 0.1465, + "step": 16976 + }, + { + "epoch": 0.1473685124261074, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018772911416269622, + "loss": 0.1006, + "step": 16977 + }, + { + "epoch": 0.14737719290631157, + "grad_norm": 0.380859375, + "learning_rate": 0.001877276138535565, + "loss": 0.1426, + "step": 16978 + }, + { + "epoch": 0.14738587338651574, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018772611345941504, + "loss": 0.1094, + "step": 16979 + }, + { + "epoch": 0.1473945538667199, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018772461298027345, + "loss": 0.1357, + "step": 16980 + }, + { + "epoch": 0.14740323434692407, + "grad_norm": 0.66796875, + "learning_rate": 0.001877231124161334, + "loss": 0.0977, + "step": 16981 + }, + { + "epoch": 0.14741191482712823, + "grad_norm": 0.2421875, + "learning_rate": 0.001877216117669965, + "loss": 0.1348, + "step": 16982 + }, + { + "epoch": 0.1474205953073324, + "grad_norm": 0.85546875, + "learning_rate": 0.0018772011103286444, + "loss": 0.1221, + "step": 16983 + }, + { + "epoch": 0.14742927578753656, + "grad_norm": 0.94140625, + "learning_rate": 0.0018771861021373883, + "loss": 0.1162, + "step": 16984 + }, + { + "epoch": 0.14743795626774073, + "grad_norm": 0.46875, + "learning_rate": 0.0018771710930962132, + "loss": 0.0859, + "step": 16985 + }, + { + "epoch": 0.1474466367479449, + "grad_norm": 1.1328125, + "learning_rate": 0.0018771560832051353, + "loss": 0.1895, + "step": 16986 + }, + { + "epoch": 0.14745531722814906, + "grad_norm": 0.140625, + "learning_rate": 0.001877141072464171, + "loss": 0.1104, + "step": 16987 + }, + { + "epoch": 0.14746399770835322, + "grad_norm": 0.232421875, + "learning_rate": 0.0018771260608733374, + "loss": 0.083, + "step": 16988 + }, + { + "epoch": 0.1474726781885574, + "grad_norm": 0.58203125, + "learning_rate": 0.0018771110484326501, + "loss": 0.0952, + "step": 16989 + }, + { + "epoch": 0.14748135866876155, + "grad_norm": 0.1259765625, + "learning_rate": 0.001877096035142126, + "loss": 0.1147, + "step": 16990 + }, + { + "epoch": 0.14749003914896572, + "grad_norm": 0.234375, + "learning_rate": 0.0018770810210017813, + "loss": 0.0908, + "step": 16991 + }, + { + "epoch": 0.14749871962916988, + "grad_norm": 0.53515625, + "learning_rate": 0.0018770660060116325, + "loss": 0.1611, + "step": 16992 + }, + { + "epoch": 0.14750740010937405, + "grad_norm": 0.62890625, + "learning_rate": 0.0018770509901716956, + "loss": 0.1074, + "step": 16993 + }, + { + "epoch": 0.1475160805895782, + "grad_norm": 0.30078125, + "learning_rate": 0.001877035973481988, + "loss": 0.1143, + "step": 16994 + }, + { + "epoch": 0.14752476106978238, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018770209559425253, + "loss": 0.1357, + "step": 16995 + }, + { + "epoch": 0.14753344154998654, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018770059375533242, + "loss": 0.1035, + "step": 16996 + }, + { + "epoch": 0.1475421220301907, + "grad_norm": 0.6171875, + "learning_rate": 0.001876990918314401, + "loss": 0.1201, + "step": 16997 + }, + { + "epoch": 0.14755080251039487, + "grad_norm": 0.345703125, + "learning_rate": 0.001876975898225772, + "loss": 0.1367, + "step": 16998 + }, + { + "epoch": 0.14755948299059904, + "grad_norm": 0.1484375, + "learning_rate": 0.0018769608772874543, + "loss": 0.1182, + "step": 16999 + }, + { + "epoch": 0.1475681634708032, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018769458554994635, + "loss": 0.1348, + "step": 17000 + }, + { + "epoch": 0.14757684395100737, + "grad_norm": 0.40234375, + "learning_rate": 0.0018769308328618166, + "loss": 0.125, + "step": 17001 + }, + { + "epoch": 0.14758552443121153, + "grad_norm": 0.181640625, + "learning_rate": 0.0018769158093745295, + "loss": 0.0874, + "step": 17002 + }, + { + "epoch": 0.1475942049114157, + "grad_norm": 0.427734375, + "learning_rate": 0.0018769007850376193, + "loss": 0.0781, + "step": 17003 + }, + { + "epoch": 0.14760288539161986, + "grad_norm": 0.484375, + "learning_rate": 0.0018768857598511018, + "loss": 0.1406, + "step": 17004 + }, + { + "epoch": 0.14761156587182403, + "grad_norm": 0.7265625, + "learning_rate": 0.001876870733814994, + "loss": 0.1016, + "step": 17005 + }, + { + "epoch": 0.1476202463520282, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018768557069293117, + "loss": 0.1113, + "step": 17006 + }, + { + "epoch": 0.14762892683223236, + "grad_norm": 0.08349609375, + "learning_rate": 0.001876840679194072, + "loss": 0.1177, + "step": 17007 + }, + { + "epoch": 0.14763760731243653, + "grad_norm": 0.14453125, + "learning_rate": 0.001876825650609291, + "loss": 0.1426, + "step": 17008 + }, + { + "epoch": 0.1476462877926407, + "grad_norm": 0.23828125, + "learning_rate": 0.0018768106211749847, + "loss": 0.0977, + "step": 17009 + }, + { + "epoch": 0.14765496827284486, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018767955908911705, + "loss": 0.1025, + "step": 17010 + }, + { + "epoch": 0.14766364875304902, + "grad_norm": 0.4609375, + "learning_rate": 0.001876780559757864, + "loss": 0.1309, + "step": 17011 + }, + { + "epoch": 0.14767232923325319, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018767655277750822, + "loss": 0.1484, + "step": 17012 + }, + { + "epoch": 0.14768100971345735, + "grad_norm": 0.220703125, + "learning_rate": 0.001876750494942841, + "loss": 0.1299, + "step": 17013 + }, + { + "epoch": 0.14768969019366152, + "grad_norm": 0.61328125, + "learning_rate": 0.0018767354612611571, + "loss": 0.126, + "step": 17014 + }, + { + "epoch": 0.14769837067386568, + "grad_norm": 0.5546875, + "learning_rate": 0.001876720426730047, + "loss": 0.1416, + "step": 17015 + }, + { + "epoch": 0.14770705115406985, + "grad_norm": 0.224609375, + "learning_rate": 0.0018767053913495274, + "loss": 0.1226, + "step": 17016 + }, + { + "epoch": 0.147715731634274, + "grad_norm": 0.5390625, + "learning_rate": 0.0018766903551196141, + "loss": 0.0898, + "step": 17017 + }, + { + "epoch": 0.14772441211447818, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018766753180403245, + "loss": 0.1426, + "step": 17018 + }, + { + "epoch": 0.14773309259468234, + "grad_norm": 0.275390625, + "learning_rate": 0.001876660280111674, + "loss": 0.1514, + "step": 17019 + }, + { + "epoch": 0.1477417730748865, + "grad_norm": 0.5625, + "learning_rate": 0.0018766452413336793, + "loss": 0.168, + "step": 17020 + }, + { + "epoch": 0.14775045355509067, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018766302017063571, + "loss": 0.1016, + "step": 17021 + }, + { + "epoch": 0.14775913403529484, + "grad_norm": 0.52734375, + "learning_rate": 0.001876615161229724, + "loss": 0.106, + "step": 17022 + }, + { + "epoch": 0.147767814515499, + "grad_norm": 0.80078125, + "learning_rate": 0.0018766001199037963, + "loss": 0.166, + "step": 17023 + }, + { + "epoch": 0.14777649499570317, + "grad_norm": 0.404296875, + "learning_rate": 0.0018765850777285901, + "loss": 0.1855, + "step": 17024 + }, + { + "epoch": 0.14778517547590733, + "grad_norm": 0.1162109375, + "learning_rate": 0.0018765700347041223, + "loss": 0.0986, + "step": 17025 + }, + { + "epoch": 0.1477938559561115, + "grad_norm": 0.28125, + "learning_rate": 0.0018765549908304092, + "loss": 0.1436, + "step": 17026 + }, + { + "epoch": 0.14780253643631566, + "grad_norm": 0.318359375, + "learning_rate": 0.001876539946107467, + "loss": 0.166, + "step": 17027 + }, + { + "epoch": 0.14781121691651983, + "grad_norm": 0.427734375, + "learning_rate": 0.0018765249005353128, + "loss": 0.1445, + "step": 17028 + }, + { + "epoch": 0.147819897396724, + "grad_norm": 0.16796875, + "learning_rate": 0.0018765098541139622, + "loss": 0.1396, + "step": 17029 + }, + { + "epoch": 0.14782857787692816, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018764948068434324, + "loss": 0.082, + "step": 17030 + }, + { + "epoch": 0.14783725835713232, + "grad_norm": 0.080078125, + "learning_rate": 0.0018764797587237393, + "loss": 0.1182, + "step": 17031 + }, + { + "epoch": 0.1478459388373365, + "grad_norm": 0.46484375, + "learning_rate": 0.0018764647097548996, + "loss": 0.0977, + "step": 17032 + }, + { + "epoch": 0.14785461931754065, + "grad_norm": 0.24609375, + "learning_rate": 0.00187644965993693, + "loss": 0.1348, + "step": 17033 + }, + { + "epoch": 0.14786329979774482, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018764346092698468, + "loss": 0.1069, + "step": 17034 + }, + { + "epoch": 0.14787198027794898, + "grad_norm": 0.423828125, + "learning_rate": 0.001876419557753666, + "loss": 0.1387, + "step": 17035 + }, + { + "epoch": 0.14788066075815315, + "grad_norm": 0.140625, + "learning_rate": 0.0018764045053884049, + "loss": 0.1113, + "step": 17036 + }, + { + "epoch": 0.1478893412383573, + "grad_norm": 0.2021484375, + "learning_rate": 0.001876389452174079, + "loss": 0.1309, + "step": 17037 + }, + { + "epoch": 0.14789802171856148, + "grad_norm": 0.3125, + "learning_rate": 0.0018763743981107055, + "loss": 0.1011, + "step": 17038 + }, + { + "epoch": 0.14790670219876564, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018763593431983007, + "loss": 0.0781, + "step": 17039 + }, + { + "epoch": 0.1479153826789698, + "grad_norm": 0.48828125, + "learning_rate": 0.0018763442874368808, + "loss": 0.2041, + "step": 17040 + }, + { + "epoch": 0.14792406315917397, + "grad_norm": 0.107421875, + "learning_rate": 0.0018763292308264623, + "loss": 0.1211, + "step": 17041 + }, + { + "epoch": 0.14793274363937814, + "grad_norm": 0.1767578125, + "learning_rate": 0.001876314173367062, + "loss": 0.124, + "step": 17042 + }, + { + "epoch": 0.1479414241195823, + "grad_norm": 0.115234375, + "learning_rate": 0.0018762991150586964, + "loss": 0.1147, + "step": 17043 + }, + { + "epoch": 0.14795010459978647, + "grad_norm": 0.31640625, + "learning_rate": 0.0018762840559013816, + "loss": 0.1172, + "step": 17044 + }, + { + "epoch": 0.14795878507999063, + "grad_norm": 0.3125, + "learning_rate": 0.0018762689958951345, + "loss": 0.1289, + "step": 17045 + }, + { + "epoch": 0.1479674655601948, + "grad_norm": 0.51953125, + "learning_rate": 0.0018762539350399708, + "loss": 0.1211, + "step": 17046 + }, + { + "epoch": 0.14797614604039896, + "grad_norm": 0.431640625, + "learning_rate": 0.0018762388733359076, + "loss": 0.1035, + "step": 17047 + }, + { + "epoch": 0.14798482652060313, + "grad_norm": 0.40234375, + "learning_rate": 0.0018762238107829612, + "loss": 0.1621, + "step": 17048 + }, + { + "epoch": 0.1479935070008073, + "grad_norm": 0.57421875, + "learning_rate": 0.0018762087473811484, + "loss": 0.0947, + "step": 17049 + }, + { + "epoch": 0.14800218748101146, + "grad_norm": 0.70703125, + "learning_rate": 0.001876193683130485, + "loss": 0.1631, + "step": 17050 + }, + { + "epoch": 0.14801086796121563, + "grad_norm": 0.263671875, + "learning_rate": 0.0018761786180309882, + "loss": 0.1162, + "step": 17051 + }, + { + "epoch": 0.1480195484414198, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018761635520826738, + "loss": 0.0962, + "step": 17052 + }, + { + "epoch": 0.14802822892162396, + "grad_norm": 0.21875, + "learning_rate": 0.0018761484852855591, + "loss": 0.1152, + "step": 17053 + }, + { + "epoch": 0.14803690940182812, + "grad_norm": 0.5859375, + "learning_rate": 0.0018761334176396597, + "loss": 0.0874, + "step": 17054 + }, + { + "epoch": 0.14804558988203229, + "grad_norm": 0.9609375, + "learning_rate": 0.0018761183491449922, + "loss": 0.1025, + "step": 17055 + }, + { + "epoch": 0.14805427036223645, + "grad_norm": 0.17578125, + "learning_rate": 0.0018761032798015737, + "loss": 0.1143, + "step": 17056 + }, + { + "epoch": 0.14806295084244062, + "grad_norm": 0.26171875, + "learning_rate": 0.0018760882096094202, + "loss": 0.1123, + "step": 17057 + }, + { + "epoch": 0.14807163132264478, + "grad_norm": 0.353515625, + "learning_rate": 0.0018760731385685481, + "loss": 0.1377, + "step": 17058 + }, + { + "epoch": 0.14808031180284895, + "grad_norm": 0.30078125, + "learning_rate": 0.0018760580666789745, + "loss": 0.1562, + "step": 17059 + }, + { + "epoch": 0.1480889922830531, + "grad_norm": 0.404296875, + "learning_rate": 0.0018760429939407154, + "loss": 0.1143, + "step": 17060 + }, + { + "epoch": 0.14809767276325728, + "grad_norm": 0.412109375, + "learning_rate": 0.0018760279203537868, + "loss": 0.1104, + "step": 17061 + }, + { + "epoch": 0.14810635324346144, + "grad_norm": 0.130859375, + "learning_rate": 0.0018760128459182064, + "loss": 0.1221, + "step": 17062 + }, + { + "epoch": 0.1481150337236656, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018759977706339897, + "loss": 0.1553, + "step": 17063 + }, + { + "epoch": 0.14812371420386977, + "grad_norm": 0.81640625, + "learning_rate": 0.0018759826945011536, + "loss": 0.1445, + "step": 17064 + }, + { + "epoch": 0.1481323946840739, + "grad_norm": 0.373046875, + "learning_rate": 0.0018759676175197143, + "loss": 0.1406, + "step": 17065 + }, + { + "epoch": 0.14814107516427807, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018759525396896885, + "loss": 0.1328, + "step": 17066 + }, + { + "epoch": 0.14814975564448224, + "grad_norm": 0.345703125, + "learning_rate": 0.0018759374610110926, + "loss": 0.1279, + "step": 17067 + }, + { + "epoch": 0.1481584361246864, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018759223814839435, + "loss": 0.1094, + "step": 17068 + }, + { + "epoch": 0.14816711660489057, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018759073011082573, + "loss": 0.1445, + "step": 17069 + }, + { + "epoch": 0.14817579708509473, + "grad_norm": 0.44140625, + "learning_rate": 0.0018758922198840503, + "loss": 0.1465, + "step": 17070 + }, + { + "epoch": 0.1481844775652989, + "grad_norm": 0.294921875, + "learning_rate": 0.0018758771378113393, + "loss": 0.1895, + "step": 17071 + }, + { + "epoch": 0.14819315804550307, + "grad_norm": 0.60546875, + "learning_rate": 0.0018758620548901409, + "loss": 0.1016, + "step": 17072 + }, + { + "epoch": 0.14820183852570723, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018758469711204712, + "loss": 0.1201, + "step": 17073 + }, + { + "epoch": 0.1482105190059114, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018758318865023467, + "loss": 0.1484, + "step": 17074 + }, + { + "epoch": 0.14821919948611556, + "grad_norm": 0.28515625, + "learning_rate": 0.0018758168010357843, + "loss": 0.1152, + "step": 17075 + }, + { + "epoch": 0.14822787996631973, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018758017147208004, + "loss": 0.1221, + "step": 17076 + }, + { + "epoch": 0.1482365604465239, + "grad_norm": 0.263671875, + "learning_rate": 0.0018757866275574114, + "loss": 0.1104, + "step": 17077 + }, + { + "epoch": 0.14824524092672806, + "grad_norm": 0.291015625, + "learning_rate": 0.0018757715395456337, + "loss": 0.1309, + "step": 17078 + }, + { + "epoch": 0.14825392140693222, + "grad_norm": 0.78125, + "learning_rate": 0.001875756450685484, + "loss": 0.1201, + "step": 17079 + }, + { + "epoch": 0.14826260188713639, + "grad_norm": 0.1640625, + "learning_rate": 0.001875741360976979, + "loss": 0.1113, + "step": 17080 + }, + { + "epoch": 0.14827128236734055, + "grad_norm": 0.388671875, + "learning_rate": 0.0018757262704201343, + "loss": 0.1611, + "step": 17081 + }, + { + "epoch": 0.14827996284754472, + "grad_norm": 0.453125, + "learning_rate": 0.0018757111790149673, + "loss": 0.0903, + "step": 17082 + }, + { + "epoch": 0.14828864332774888, + "grad_norm": 0.158203125, + "learning_rate": 0.0018756960867614943, + "loss": 0.1025, + "step": 17083 + }, + { + "epoch": 0.14829732380795305, + "grad_norm": 0.142578125, + "learning_rate": 0.0018756809936597313, + "loss": 0.0957, + "step": 17084 + }, + { + "epoch": 0.1483060042881572, + "grad_norm": 0.12353515625, + "learning_rate": 0.0018756658997096955, + "loss": 0.124, + "step": 17085 + }, + { + "epoch": 0.14831468476836138, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018756508049114033, + "loss": 0.1348, + "step": 17086 + }, + { + "epoch": 0.14832336524856554, + "grad_norm": 0.2890625, + "learning_rate": 0.0018756357092648709, + "loss": 0.0986, + "step": 17087 + }, + { + "epoch": 0.1483320457287697, + "grad_norm": 0.359375, + "learning_rate": 0.0018756206127701147, + "loss": 0.1309, + "step": 17088 + }, + { + "epoch": 0.14834072620897387, + "grad_norm": 0.373046875, + "learning_rate": 0.001875605515427152, + "loss": 0.0938, + "step": 17089 + }, + { + "epoch": 0.14834940668917804, + "grad_norm": 0.29296875, + "learning_rate": 0.0018755904172359983, + "loss": 0.127, + "step": 17090 + }, + { + "epoch": 0.1483580871693822, + "grad_norm": 0.58203125, + "learning_rate": 0.0018755753181966706, + "loss": 0.1318, + "step": 17091 + }, + { + "epoch": 0.14836676764958637, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018755602183091857, + "loss": 0.0981, + "step": 17092 + }, + { + "epoch": 0.14837544812979053, + "grad_norm": 0.1640625, + "learning_rate": 0.0018755451175735593, + "loss": 0.124, + "step": 17093 + }, + { + "epoch": 0.1483841286099947, + "grad_norm": 0.59765625, + "learning_rate": 0.0018755300159898087, + "loss": 0.1177, + "step": 17094 + }, + { + "epoch": 0.14839280909019886, + "grad_norm": 0.17578125, + "learning_rate": 0.0018755149135579504, + "loss": 0.1426, + "step": 17095 + }, + { + "epoch": 0.14840148957040303, + "grad_norm": 0.55078125, + "learning_rate": 0.0018754998102780004, + "loss": 0.1104, + "step": 17096 + }, + { + "epoch": 0.1484101700506072, + "grad_norm": 0.421875, + "learning_rate": 0.0018754847061499753, + "loss": 0.1094, + "step": 17097 + }, + { + "epoch": 0.14841885053081136, + "grad_norm": 0.083984375, + "learning_rate": 0.001875469601173892, + "loss": 0.0977, + "step": 17098 + }, + { + "epoch": 0.14842753101101552, + "grad_norm": 0.62890625, + "learning_rate": 0.0018754544953497668, + "loss": 0.1309, + "step": 17099 + }, + { + "epoch": 0.1484362114912197, + "grad_norm": 0.4375, + "learning_rate": 0.0018754393886776157, + "loss": 0.1982, + "step": 17100 + }, + { + "epoch": 0.14844489197142385, + "grad_norm": 0.484375, + "learning_rate": 0.0018754242811574564, + "loss": 0.0811, + "step": 17101 + }, + { + "epoch": 0.14845357245162802, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018754091727893043, + "loss": 0.1016, + "step": 17102 + }, + { + "epoch": 0.14846225293183218, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018753940635731765, + "loss": 0.1387, + "step": 17103 + }, + { + "epoch": 0.14847093341203635, + "grad_norm": 0.38671875, + "learning_rate": 0.0018753789535090896, + "loss": 0.0874, + "step": 17104 + }, + { + "epoch": 0.14847961389224051, + "grad_norm": 0.23046875, + "learning_rate": 0.0018753638425970597, + "loss": 0.0815, + "step": 17105 + }, + { + "epoch": 0.14848829437244468, + "grad_norm": 0.96875, + "learning_rate": 0.0018753487308371037, + "loss": 0.1074, + "step": 17106 + }, + { + "epoch": 0.14849697485264884, + "grad_norm": 0.609375, + "learning_rate": 0.001875333618229238, + "loss": 0.0933, + "step": 17107 + }, + { + "epoch": 0.148505655332853, + "grad_norm": 0.130859375, + "learning_rate": 0.001875318504773479, + "loss": 0.1367, + "step": 17108 + }, + { + "epoch": 0.14851433581305717, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018753033904698436, + "loss": 0.1387, + "step": 17109 + }, + { + "epoch": 0.14852301629326134, + "grad_norm": 0.0849609375, + "learning_rate": 0.0018752882753183477, + "loss": 0.105, + "step": 17110 + }, + { + "epoch": 0.1485316967734655, + "grad_norm": 0.1240234375, + "learning_rate": 0.0018752731593190083, + "loss": 0.1455, + "step": 17111 + }, + { + "epoch": 0.14854037725366967, + "grad_norm": 0.353515625, + "learning_rate": 0.0018752580424718417, + "loss": 0.1055, + "step": 17112 + }, + { + "epoch": 0.14854905773387383, + "grad_norm": 0.56640625, + "learning_rate": 0.001875242924776865, + "loss": 0.1592, + "step": 17113 + }, + { + "epoch": 0.148557738214078, + "grad_norm": 0.322265625, + "learning_rate": 0.001875227806234094, + "loss": 0.1143, + "step": 17114 + }, + { + "epoch": 0.14856641869428217, + "grad_norm": 0.62890625, + "learning_rate": 0.0018752126868435456, + "loss": 0.124, + "step": 17115 + }, + { + "epoch": 0.14857509917448633, + "grad_norm": 0.54296875, + "learning_rate": 0.0018751975666052361, + "loss": 0.0869, + "step": 17116 + }, + { + "epoch": 0.1485837796546905, + "grad_norm": 1.2421875, + "learning_rate": 0.0018751824455191819, + "loss": 0.1064, + "step": 17117 + }, + { + "epoch": 0.14859246013489466, + "grad_norm": 0.47265625, + "learning_rate": 0.0018751673235854005, + "loss": 0.125, + "step": 17118 + }, + { + "epoch": 0.14860114061509883, + "grad_norm": 0.333984375, + "learning_rate": 0.0018751522008039075, + "loss": 0.1289, + "step": 17119 + }, + { + "epoch": 0.148609821095303, + "grad_norm": 0.490234375, + "learning_rate": 0.0018751370771747196, + "loss": 0.1426, + "step": 17120 + }, + { + "epoch": 0.14861850157550716, + "grad_norm": 0.109375, + "learning_rate": 0.0018751219526978535, + "loss": 0.1113, + "step": 17121 + }, + { + "epoch": 0.14862718205571132, + "grad_norm": 0.181640625, + "learning_rate": 0.0018751068273733253, + "loss": 0.0957, + "step": 17122 + }, + { + "epoch": 0.1486358625359155, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018750917012011524, + "loss": 0.1021, + "step": 17123 + }, + { + "epoch": 0.14864454301611965, + "grad_norm": 0.287109375, + "learning_rate": 0.001875076574181351, + "loss": 0.1367, + "step": 17124 + }, + { + "epoch": 0.14865322349632382, + "grad_norm": 0.115234375, + "learning_rate": 0.0018750614463139371, + "loss": 0.1201, + "step": 17125 + }, + { + "epoch": 0.14866190397652798, + "grad_norm": 0.54296875, + "learning_rate": 0.0018750463175989276, + "loss": 0.1211, + "step": 17126 + }, + { + "epoch": 0.14867058445673215, + "grad_norm": 0.26171875, + "learning_rate": 0.0018750311880363394, + "loss": 0.0894, + "step": 17127 + }, + { + "epoch": 0.1486792649369363, + "grad_norm": 0.130859375, + "learning_rate": 0.0018750160576261885, + "loss": 0.1855, + "step": 17128 + }, + { + "epoch": 0.14868794541714048, + "grad_norm": 0.09521484375, + "learning_rate": 0.001875000926368492, + "loss": 0.1113, + "step": 17129 + }, + { + "epoch": 0.14869662589734464, + "grad_norm": 0.48828125, + "learning_rate": 0.001874985794263266, + "loss": 0.1211, + "step": 17130 + }, + { + "epoch": 0.1487053063775488, + "grad_norm": 0.0888671875, + "learning_rate": 0.001874970661310527, + "loss": 0.0918, + "step": 17131 + }, + { + "epoch": 0.14871398685775297, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018749555275102917, + "loss": 0.1191, + "step": 17132 + }, + { + "epoch": 0.14872266733795714, + "grad_norm": 0.1318359375, + "learning_rate": 0.001874940392862577, + "loss": 0.124, + "step": 17133 + }, + { + "epoch": 0.1487313478181613, + "grad_norm": 0.482421875, + "learning_rate": 0.001874925257367399, + "loss": 0.1377, + "step": 17134 + }, + { + "epoch": 0.14874002829836547, + "grad_norm": 0.2431640625, + "learning_rate": 0.001874910121024774, + "loss": 0.1177, + "step": 17135 + }, + { + "epoch": 0.14874870877856963, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018748949838347196, + "loss": 0.1133, + "step": 17136 + }, + { + "epoch": 0.1487573892587738, + "grad_norm": 0.7890625, + "learning_rate": 0.0018748798457972511, + "loss": 0.0986, + "step": 17137 + }, + { + "epoch": 0.14876606973897796, + "grad_norm": 0.369140625, + "learning_rate": 0.0018748647069123858, + "loss": 0.1211, + "step": 17138 + }, + { + "epoch": 0.14877475021918213, + "grad_norm": 0.09375, + "learning_rate": 0.0018748495671801404, + "loss": 0.0977, + "step": 17139 + }, + { + "epoch": 0.1487834306993863, + "grad_norm": 0.138671875, + "learning_rate": 0.001874834426600531, + "loss": 0.0942, + "step": 17140 + }, + { + "epoch": 0.14879211117959046, + "grad_norm": 0.38671875, + "learning_rate": 0.0018748192851735742, + "loss": 0.1084, + "step": 17141 + }, + { + "epoch": 0.14880079165979462, + "grad_norm": 0.166015625, + "learning_rate": 0.001874804142899287, + "loss": 0.125, + "step": 17142 + }, + { + "epoch": 0.1488094721399988, + "grad_norm": 0.0908203125, + "learning_rate": 0.001874788999777685, + "loss": 0.1123, + "step": 17143 + }, + { + "epoch": 0.14881815262020295, + "grad_norm": 0.1396484375, + "learning_rate": 0.001874773855808786, + "loss": 0.1064, + "step": 17144 + }, + { + "epoch": 0.14882683310040712, + "grad_norm": 0.5234375, + "learning_rate": 0.0018747587109926057, + "loss": 0.1309, + "step": 17145 + }, + { + "epoch": 0.14883551358061128, + "grad_norm": 0.296875, + "learning_rate": 0.0018747435653291608, + "loss": 0.1162, + "step": 17146 + }, + { + "epoch": 0.14884419406081545, + "grad_norm": 0.07177734375, + "learning_rate": 0.001874728418818468, + "loss": 0.1089, + "step": 17147 + }, + { + "epoch": 0.14885287454101961, + "grad_norm": 0.46875, + "learning_rate": 0.001874713271460544, + "loss": 0.1846, + "step": 17148 + }, + { + "epoch": 0.14886155502122378, + "grad_norm": 0.2890625, + "learning_rate": 0.0018746981232554052, + "loss": 0.1182, + "step": 17149 + }, + { + "epoch": 0.14887023550142794, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018746829742030682, + "loss": 0.1582, + "step": 17150 + }, + { + "epoch": 0.1488789159816321, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018746678243035492, + "loss": 0.1118, + "step": 17151 + }, + { + "epoch": 0.14888759646183627, + "grad_norm": 0.29296875, + "learning_rate": 0.0018746526735568653, + "loss": 0.0737, + "step": 17152 + }, + { + "epoch": 0.14889627694204044, + "grad_norm": 0.87109375, + "learning_rate": 0.0018746375219630333, + "loss": 0.127, + "step": 17153 + }, + { + "epoch": 0.1489049574222446, + "grad_norm": 0.1337890625, + "learning_rate": 0.001874622369522069, + "loss": 0.168, + "step": 17154 + }, + { + "epoch": 0.14891363790244877, + "grad_norm": 0.291015625, + "learning_rate": 0.0018746072162339892, + "loss": 0.1289, + "step": 17155 + }, + { + "epoch": 0.14892231838265294, + "grad_norm": 0.08251953125, + "learning_rate": 0.0018745920620988107, + "loss": 0.1016, + "step": 17156 + }, + { + "epoch": 0.1489309988628571, + "grad_norm": 0.2109375, + "learning_rate": 0.00187457690711655, + "loss": 0.1641, + "step": 17157 + }, + { + "epoch": 0.14893967934306127, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018745617512872235, + "loss": 0.1025, + "step": 17158 + }, + { + "epoch": 0.14894835982326543, + "grad_norm": 0.255859375, + "learning_rate": 0.001874546594610848, + "loss": 0.0938, + "step": 17159 + }, + { + "epoch": 0.1489570403034696, + "grad_norm": 0.15625, + "learning_rate": 0.0018745314370874398, + "loss": 0.166, + "step": 17160 + }, + { + "epoch": 0.14896572078367376, + "grad_norm": 0.65234375, + "learning_rate": 0.001874516278717016, + "loss": 0.1777, + "step": 17161 + }, + { + "epoch": 0.14897440126387793, + "grad_norm": 0.83203125, + "learning_rate": 0.0018745011194995927, + "loss": 0.1348, + "step": 17162 + }, + { + "epoch": 0.1489830817440821, + "grad_norm": 0.18359375, + "learning_rate": 0.0018744859594351864, + "loss": 0.123, + "step": 17163 + }, + { + "epoch": 0.14899176222428626, + "grad_norm": 0.171875, + "learning_rate": 0.0018744707985238142, + "loss": 0.1045, + "step": 17164 + }, + { + "epoch": 0.14900044270449042, + "grad_norm": 0.2119140625, + "learning_rate": 0.001874455636765492, + "loss": 0.1318, + "step": 17165 + }, + { + "epoch": 0.1490091231846946, + "grad_norm": 0.33203125, + "learning_rate": 0.001874440474160237, + "loss": 0.1396, + "step": 17166 + }, + { + "epoch": 0.14901780366489875, + "grad_norm": 0.37109375, + "learning_rate": 0.0018744253107080655, + "loss": 0.1084, + "step": 17167 + }, + { + "epoch": 0.14902648414510292, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018744101464089945, + "loss": 0.1123, + "step": 17168 + }, + { + "epoch": 0.14903516462530708, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018743949812630397, + "loss": 0.1108, + "step": 17169 + }, + { + "epoch": 0.14904384510551125, + "grad_norm": 0.2734375, + "learning_rate": 0.001874379815270218, + "loss": 0.1016, + "step": 17170 + }, + { + "epoch": 0.1490525255857154, + "grad_norm": 0.71875, + "learning_rate": 0.0018743646484305468, + "loss": 0.0928, + "step": 17171 + }, + { + "epoch": 0.14906120606591958, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018743494807440416, + "loss": 0.0889, + "step": 17172 + }, + { + "epoch": 0.14906988654612374, + "grad_norm": 0.337890625, + "learning_rate": 0.0018743343122107196, + "loss": 0.1689, + "step": 17173 + }, + { + "epoch": 0.1490785670263279, + "grad_norm": 0.59765625, + "learning_rate": 0.0018743191428305976, + "loss": 0.1064, + "step": 17174 + }, + { + "epoch": 0.14908724750653207, + "grad_norm": 1.359375, + "learning_rate": 0.001874303972603691, + "loss": 0.1143, + "step": 17175 + }, + { + "epoch": 0.14909592798673624, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018742888015300177, + "loss": 0.127, + "step": 17176 + }, + { + "epoch": 0.1491046084669404, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018742736296095937, + "loss": 0.1416, + "step": 17177 + }, + { + "epoch": 0.14911328894714457, + "grad_norm": 0.059814453125, + "learning_rate": 0.0018742584568424358, + "loss": 0.0728, + "step": 17178 + }, + { + "epoch": 0.14912196942734873, + "grad_norm": 0.16796875, + "learning_rate": 0.0018742432832285604, + "loss": 0.1133, + "step": 17179 + }, + { + "epoch": 0.1491306499075529, + "grad_norm": 0.42578125, + "learning_rate": 0.001874228108767984, + "loss": 0.1396, + "step": 17180 + }, + { + "epoch": 0.14913933038775706, + "grad_norm": 0.2421875, + "learning_rate": 0.0018742129334607235, + "loss": 0.1191, + "step": 17181 + }, + { + "epoch": 0.14914801086796123, + "grad_norm": 0.39453125, + "learning_rate": 0.0018741977573067956, + "loss": 0.1543, + "step": 17182 + }, + { + "epoch": 0.1491566913481654, + "grad_norm": 0.21875, + "learning_rate": 0.0018741825803062168, + "loss": 0.1143, + "step": 17183 + }, + { + "epoch": 0.14916537182836956, + "grad_norm": 0.296875, + "learning_rate": 0.0018741674024590031, + "loss": 0.0967, + "step": 17184 + }, + { + "epoch": 0.14917405230857372, + "grad_norm": 0.251953125, + "learning_rate": 0.0018741522237651718, + "loss": 0.1099, + "step": 17185 + }, + { + "epoch": 0.1491827327887779, + "grad_norm": 0.3125, + "learning_rate": 0.0018741370442247392, + "loss": 0.083, + "step": 17186 + }, + { + "epoch": 0.14919141326898205, + "grad_norm": 0.2421875, + "learning_rate": 0.0018741218638377217, + "loss": 0.1182, + "step": 17187 + }, + { + "epoch": 0.1492000937491862, + "grad_norm": 0.416015625, + "learning_rate": 0.0018741066826041366, + "loss": 0.0898, + "step": 17188 + }, + { + "epoch": 0.14920877422939036, + "grad_norm": 0.138671875, + "learning_rate": 0.0018740915005239999, + "loss": 0.1689, + "step": 17189 + }, + { + "epoch": 0.14921745470959452, + "grad_norm": 0.5, + "learning_rate": 0.0018740763175973281, + "loss": 0.1099, + "step": 17190 + }, + { + "epoch": 0.1492261351897987, + "grad_norm": 0.0703125, + "learning_rate": 0.0018740611338241384, + "loss": 0.0996, + "step": 17191 + }, + { + "epoch": 0.14923481567000285, + "grad_norm": 0.2890625, + "learning_rate": 0.0018740459492044468, + "loss": 0.1035, + "step": 17192 + }, + { + "epoch": 0.14924349615020702, + "grad_norm": 0.6171875, + "learning_rate": 0.0018740307637382706, + "loss": 0.1143, + "step": 17193 + }, + { + "epoch": 0.14925217663041118, + "grad_norm": 0.486328125, + "learning_rate": 0.0018740155774256257, + "loss": 0.0815, + "step": 17194 + }, + { + "epoch": 0.14926085711061535, + "grad_norm": 0.1376953125, + "learning_rate": 0.001874000390266529, + "loss": 0.123, + "step": 17195 + }, + { + "epoch": 0.1492695375908195, + "grad_norm": 0.55859375, + "learning_rate": 0.0018739852022609971, + "loss": 0.1133, + "step": 17196 + }, + { + "epoch": 0.14927821807102368, + "grad_norm": 0.10546875, + "learning_rate": 0.0018739700134090466, + "loss": 0.1021, + "step": 17197 + }, + { + "epoch": 0.14928689855122784, + "grad_norm": 0.640625, + "learning_rate": 0.0018739548237106944, + "loss": 0.1084, + "step": 17198 + }, + { + "epoch": 0.149295579031432, + "grad_norm": 0.306640625, + "learning_rate": 0.0018739396331659563, + "loss": 0.1396, + "step": 17199 + }, + { + "epoch": 0.14930425951163617, + "grad_norm": 1.234375, + "learning_rate": 0.0018739244417748501, + "loss": 0.1309, + "step": 17200 + }, + { + "epoch": 0.14931293999184034, + "grad_norm": 0.07861328125, + "learning_rate": 0.0018739092495373912, + "loss": 0.0811, + "step": 17201 + }, + { + "epoch": 0.1493216204720445, + "grad_norm": 0.16015625, + "learning_rate": 0.001873894056453597, + "loss": 0.1543, + "step": 17202 + }, + { + "epoch": 0.14933030095224867, + "grad_norm": 0.59375, + "learning_rate": 0.0018738788625234839, + "loss": 0.1055, + "step": 17203 + }, + { + "epoch": 0.14933898143245283, + "grad_norm": 0.228515625, + "learning_rate": 0.0018738636677470687, + "loss": 0.0986, + "step": 17204 + }, + { + "epoch": 0.149347661912657, + "grad_norm": 0.474609375, + "learning_rate": 0.0018738484721243674, + "loss": 0.1133, + "step": 17205 + }, + { + "epoch": 0.14935634239286116, + "grad_norm": 0.091796875, + "learning_rate": 0.0018738332756553973, + "loss": 0.123, + "step": 17206 + }, + { + "epoch": 0.14936502287306533, + "grad_norm": 0.3046875, + "learning_rate": 0.0018738180783401746, + "loss": 0.1357, + "step": 17207 + }, + { + "epoch": 0.1493737033532695, + "grad_norm": 0.3984375, + "learning_rate": 0.001873802880178716, + "loss": 0.0981, + "step": 17208 + }, + { + "epoch": 0.14938238383347366, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018737876811710387, + "loss": 0.0918, + "step": 17209 + }, + { + "epoch": 0.14939106431367782, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018737724813171583, + "loss": 0.127, + "step": 17210 + }, + { + "epoch": 0.149399744793882, + "grad_norm": 0.11376953125, + "learning_rate": 0.0018737572806170925, + "loss": 0.1064, + "step": 17211 + }, + { + "epoch": 0.14940842527408615, + "grad_norm": 0.189453125, + "learning_rate": 0.0018737420790708565, + "loss": 0.1279, + "step": 17212 + }, + { + "epoch": 0.14941710575429032, + "grad_norm": 0.5859375, + "learning_rate": 0.0018737268766784684, + "loss": 0.1079, + "step": 17213 + }, + { + "epoch": 0.14942578623449448, + "grad_norm": 0.62109375, + "learning_rate": 0.0018737116734399441, + "loss": 0.1367, + "step": 17214 + }, + { + "epoch": 0.14943446671469865, + "grad_norm": 0.302734375, + "learning_rate": 0.0018736964693553003, + "loss": 0.1162, + "step": 17215 + }, + { + "epoch": 0.14944314719490281, + "grad_norm": 0.057861328125, + "learning_rate": 0.0018736812644245537, + "loss": 0.0815, + "step": 17216 + }, + { + "epoch": 0.14945182767510698, + "grad_norm": 0.91015625, + "learning_rate": 0.001873666058647721, + "loss": 0.0801, + "step": 17217 + }, + { + "epoch": 0.14946050815531114, + "grad_norm": 0.462890625, + "learning_rate": 0.0018736508520248185, + "loss": 0.127, + "step": 17218 + }, + { + "epoch": 0.1494691886355153, + "grad_norm": 0.470703125, + "learning_rate": 0.0018736356445558631, + "loss": 0.1079, + "step": 17219 + }, + { + "epoch": 0.14947786911571947, + "grad_norm": 0.49609375, + "learning_rate": 0.0018736204362408717, + "loss": 0.1069, + "step": 17220 + }, + { + "epoch": 0.14948654959592364, + "grad_norm": 0.283203125, + "learning_rate": 0.0018736052270798603, + "loss": 0.1055, + "step": 17221 + }, + { + "epoch": 0.1494952300761278, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018735900170728458, + "loss": 0.1084, + "step": 17222 + }, + { + "epoch": 0.14950391055633197, + "grad_norm": 0.1015625, + "learning_rate": 0.001873574806219845, + "loss": 0.1143, + "step": 17223 + }, + { + "epoch": 0.14951259103653614, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018735595945208743, + "loss": 0.0942, + "step": 17224 + }, + { + "epoch": 0.1495212715167403, + "grad_norm": 0.498046875, + "learning_rate": 0.0018735443819759506, + "loss": 0.0859, + "step": 17225 + }, + { + "epoch": 0.14952995199694447, + "grad_norm": 0.64453125, + "learning_rate": 0.0018735291685850904, + "loss": 0.0854, + "step": 17226 + }, + { + "epoch": 0.14953863247714863, + "grad_norm": 0.16015625, + "learning_rate": 0.0018735139543483105, + "loss": 0.124, + "step": 17227 + }, + { + "epoch": 0.1495473129573528, + "grad_norm": 0.56640625, + "learning_rate": 0.001873498739265627, + "loss": 0.106, + "step": 17228 + }, + { + "epoch": 0.14955599343755696, + "grad_norm": 0.34765625, + "learning_rate": 0.001873483523337057, + "loss": 0.1328, + "step": 17229 + }, + { + "epoch": 0.14956467391776113, + "grad_norm": 0.498046875, + "learning_rate": 0.001873468306562617, + "loss": 0.0835, + "step": 17230 + }, + { + "epoch": 0.1495733543979653, + "grad_norm": 0.30078125, + "learning_rate": 0.0018734530889423237, + "loss": 0.1084, + "step": 17231 + }, + { + "epoch": 0.14958203487816946, + "grad_norm": 0.373046875, + "learning_rate": 0.0018734378704761939, + "loss": 0.1133, + "step": 17232 + }, + { + "epoch": 0.14959071535837362, + "grad_norm": 0.373046875, + "learning_rate": 0.0018734226511642439, + "loss": 0.1064, + "step": 17233 + }, + { + "epoch": 0.1495993958385778, + "grad_norm": 0.35546875, + "learning_rate": 0.0018734074310064906, + "loss": 0.0947, + "step": 17234 + }, + { + "epoch": 0.14960807631878195, + "grad_norm": 0.08984375, + "learning_rate": 0.0018733922100029503, + "loss": 0.0942, + "step": 17235 + }, + { + "epoch": 0.14961675679898612, + "grad_norm": 0.1845703125, + "learning_rate": 0.0018733769881536402, + "loss": 0.1387, + "step": 17236 + }, + { + "epoch": 0.14962543727919028, + "grad_norm": 0.58203125, + "learning_rate": 0.0018733617654585763, + "loss": 0.1289, + "step": 17237 + }, + { + "epoch": 0.14963411775939445, + "grad_norm": 0.7109375, + "learning_rate": 0.001873346541917776, + "loss": 0.1328, + "step": 17238 + }, + { + "epoch": 0.1496427982395986, + "grad_norm": 0.275390625, + "learning_rate": 0.0018733313175312554, + "loss": 0.1309, + "step": 17239 + }, + { + "epoch": 0.14965147871980278, + "grad_norm": 0.08447265625, + "learning_rate": 0.0018733160922990314, + "loss": 0.1021, + "step": 17240 + }, + { + "epoch": 0.14966015920000694, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018733008662211203, + "loss": 0.0952, + "step": 17241 + }, + { + "epoch": 0.1496688396802111, + "grad_norm": 0.451171875, + "learning_rate": 0.001873285639297539, + "loss": 0.0908, + "step": 17242 + }, + { + "epoch": 0.14967752016041527, + "grad_norm": 0.099609375, + "learning_rate": 0.001873270411528304, + "loss": 0.0962, + "step": 17243 + }, + { + "epoch": 0.14968620064061944, + "grad_norm": 3.6875, + "learning_rate": 0.0018732551829134325, + "loss": 0.252, + "step": 17244 + }, + { + "epoch": 0.1496948811208236, + "grad_norm": 0.8359375, + "learning_rate": 0.0018732399534529408, + "loss": 0.1201, + "step": 17245 + }, + { + "epoch": 0.14970356160102777, + "grad_norm": 0.10107421875, + "learning_rate": 0.0018732247231468453, + "loss": 0.0952, + "step": 17246 + }, + { + "epoch": 0.14971224208123193, + "grad_norm": 0.130859375, + "learning_rate": 0.0018732094919951627, + "loss": 0.126, + "step": 17247 + }, + { + "epoch": 0.1497209225614361, + "grad_norm": 0.8359375, + "learning_rate": 0.00187319425999791, + "loss": 0.1191, + "step": 17248 + }, + { + "epoch": 0.14972960304164026, + "grad_norm": 0.341796875, + "learning_rate": 0.0018731790271551037, + "loss": 0.0986, + "step": 17249 + }, + { + "epoch": 0.14973828352184443, + "grad_norm": 0.5078125, + "learning_rate": 0.0018731637934667603, + "loss": 0.1523, + "step": 17250 + }, + { + "epoch": 0.1497469640020486, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018731485589328968, + "loss": 0.1182, + "step": 17251 + }, + { + "epoch": 0.14975564448225276, + "grad_norm": 0.259765625, + "learning_rate": 0.0018731333235535296, + "loss": 0.1455, + "step": 17252 + }, + { + "epoch": 0.14976432496245692, + "grad_norm": 0.58203125, + "learning_rate": 0.0018731180873286753, + "loss": 0.126, + "step": 17253 + }, + { + "epoch": 0.1497730054426611, + "grad_norm": 0.412109375, + "learning_rate": 0.0018731028502583505, + "loss": 0.1196, + "step": 17254 + }, + { + "epoch": 0.14978168592286525, + "grad_norm": 0.1396484375, + "learning_rate": 0.001873087612342572, + "loss": 0.1064, + "step": 17255 + }, + { + "epoch": 0.14979036640306942, + "grad_norm": 0.1728515625, + "learning_rate": 0.001873072373581357, + "loss": 0.1118, + "step": 17256 + }, + { + "epoch": 0.14979904688327358, + "grad_norm": 0.2890625, + "learning_rate": 0.0018730571339747215, + "loss": 0.1348, + "step": 17257 + }, + { + "epoch": 0.14980772736347775, + "grad_norm": 0.384765625, + "learning_rate": 0.0018730418935226823, + "loss": 0.0806, + "step": 17258 + }, + { + "epoch": 0.14981640784368191, + "grad_norm": 0.150390625, + "learning_rate": 0.0018730266522252559, + "loss": 0.1006, + "step": 17259 + }, + { + "epoch": 0.14982508832388608, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018730114100824593, + "loss": 0.1025, + "step": 17260 + }, + { + "epoch": 0.14983376880409024, + "grad_norm": 0.2451171875, + "learning_rate": 0.0018729961670943091, + "loss": 0.125, + "step": 17261 + }, + { + "epoch": 0.1498424492842944, + "grad_norm": 0.1435546875, + "learning_rate": 0.001872980923260822, + "loss": 0.1245, + "step": 17262 + }, + { + "epoch": 0.14985112976449858, + "grad_norm": 0.1875, + "learning_rate": 0.0018729656785820147, + "loss": 0.1699, + "step": 17263 + }, + { + "epoch": 0.14985981024470274, + "grad_norm": 0.5390625, + "learning_rate": 0.0018729504330579033, + "loss": 0.1309, + "step": 17264 + }, + { + "epoch": 0.1498684907249069, + "grad_norm": 0.640625, + "learning_rate": 0.001872935186688505, + "loss": 0.1064, + "step": 17265 + }, + { + "epoch": 0.14987717120511107, + "grad_norm": 0.154296875, + "learning_rate": 0.0018729199394738364, + "loss": 0.1055, + "step": 17266 + }, + { + "epoch": 0.14988585168531524, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018729046914139145, + "loss": 0.1201, + "step": 17267 + }, + { + "epoch": 0.1498945321655194, + "grad_norm": 0.466796875, + "learning_rate": 0.001872889442508755, + "loss": 0.1641, + "step": 17268 + }, + { + "epoch": 0.14990321264572357, + "grad_norm": 0.33203125, + "learning_rate": 0.001872874192758376, + "loss": 0.1162, + "step": 17269 + }, + { + "epoch": 0.14991189312592773, + "grad_norm": 0.130859375, + "learning_rate": 0.0018728589421627931, + "loss": 0.1445, + "step": 17270 + }, + { + "epoch": 0.1499205736061319, + "grad_norm": 0.271484375, + "learning_rate": 0.001872843690722023, + "loss": 0.1123, + "step": 17271 + }, + { + "epoch": 0.14992925408633606, + "grad_norm": 0.1201171875, + "learning_rate": 0.001872828438436083, + "loss": 0.0879, + "step": 17272 + }, + { + "epoch": 0.14993793456654023, + "grad_norm": 0.201171875, + "learning_rate": 0.001872813185304989, + "loss": 0.125, + "step": 17273 + }, + { + "epoch": 0.1499466150467444, + "grad_norm": 0.82421875, + "learning_rate": 0.0018727979313287587, + "loss": 0.1094, + "step": 17274 + }, + { + "epoch": 0.14995529552694856, + "grad_norm": 0.109375, + "learning_rate": 0.0018727826765074077, + "loss": 0.1279, + "step": 17275 + }, + { + "epoch": 0.14996397600715272, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018727674208409532, + "loss": 0.1348, + "step": 17276 + }, + { + "epoch": 0.1499726564873569, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018727521643294124, + "loss": 0.1177, + "step": 17277 + }, + { + "epoch": 0.14998133696756105, + "grad_norm": 0.328125, + "learning_rate": 0.001872736906972801, + "loss": 0.1484, + "step": 17278 + }, + { + "epoch": 0.14999001744776522, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018727216487711361, + "loss": 0.1035, + "step": 17279 + }, + { + "epoch": 0.14999869792796938, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018727063897244344, + "loss": 0.1235, + "step": 17280 + }, + { + "epoch": 0.15000737840817355, + "grad_norm": 0.470703125, + "learning_rate": 0.0018726911298327126, + "loss": 0.123, + "step": 17281 + }, + { + "epoch": 0.1500160588883777, + "grad_norm": 0.07666015625, + "learning_rate": 0.0018726758690959875, + "loss": 0.0825, + "step": 17282 + }, + { + "epoch": 0.15002473936858188, + "grad_norm": 0.337890625, + "learning_rate": 0.0018726606075142758, + "loss": 0.0942, + "step": 17283 + }, + { + "epoch": 0.15003341984878604, + "grad_norm": 0.146484375, + "learning_rate": 0.001872645345087594, + "loss": 0.1016, + "step": 17284 + }, + { + "epoch": 0.1500421003289902, + "grad_norm": 0.29296875, + "learning_rate": 0.0018726300818159586, + "loss": 0.1113, + "step": 17285 + }, + { + "epoch": 0.15005078080919437, + "grad_norm": 0.486328125, + "learning_rate": 0.001872614817699387, + "loss": 0.0991, + "step": 17286 + }, + { + "epoch": 0.15005946128939854, + "grad_norm": 0.291015625, + "learning_rate": 0.001872599552737895, + "loss": 0.1104, + "step": 17287 + }, + { + "epoch": 0.1500681417696027, + "grad_norm": 1.265625, + "learning_rate": 0.0018725842869314999, + "loss": 0.1426, + "step": 17288 + }, + { + "epoch": 0.15007682224980687, + "grad_norm": 0.341796875, + "learning_rate": 0.0018725690202802184, + "loss": 0.1123, + "step": 17289 + }, + { + "epoch": 0.15008550273001103, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018725537527840667, + "loss": 0.1348, + "step": 17290 + }, + { + "epoch": 0.1500941832102152, + "grad_norm": 0.8359375, + "learning_rate": 0.0018725384844430623, + "loss": 0.1045, + "step": 17291 + }, + { + "epoch": 0.15010286369041936, + "grad_norm": 0.50390625, + "learning_rate": 0.0018725232152572212, + "loss": 0.1025, + "step": 17292 + }, + { + "epoch": 0.15011154417062353, + "grad_norm": 0.1630859375, + "learning_rate": 0.00187250794522656, + "loss": 0.1167, + "step": 17293 + }, + { + "epoch": 0.1501202246508277, + "grad_norm": 0.169921875, + "learning_rate": 0.0018724926743510961, + "loss": 0.0796, + "step": 17294 + }, + { + "epoch": 0.15012890513103186, + "grad_norm": 0.7109375, + "learning_rate": 0.0018724774026308458, + "loss": 0.1367, + "step": 17295 + }, + { + "epoch": 0.15013758561123602, + "grad_norm": 0.2470703125, + "learning_rate": 0.0018724621300658256, + "loss": 0.1719, + "step": 17296 + }, + { + "epoch": 0.1501462660914402, + "grad_norm": 0.412109375, + "learning_rate": 0.0018724468566560526, + "loss": 0.1309, + "step": 17297 + }, + { + "epoch": 0.15015494657164435, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018724315824015431, + "loss": 0.1104, + "step": 17298 + }, + { + "epoch": 0.15016362705184852, + "grad_norm": 0.2470703125, + "learning_rate": 0.0018724163073023145, + "loss": 0.127, + "step": 17299 + }, + { + "epoch": 0.15017230753205268, + "grad_norm": 0.14453125, + "learning_rate": 0.0018724010313583826, + "loss": 0.1133, + "step": 17300 + }, + { + "epoch": 0.15018098801225685, + "grad_norm": 0.14453125, + "learning_rate": 0.0018723857545697646, + "loss": 0.1475, + "step": 17301 + }, + { + "epoch": 0.15018966849246101, + "grad_norm": 0.09033203125, + "learning_rate": 0.001872370476936477, + "loss": 0.0923, + "step": 17302 + }, + { + "epoch": 0.15019834897266518, + "grad_norm": 0.337890625, + "learning_rate": 0.0018723551984585372, + "loss": 0.1152, + "step": 17303 + }, + { + "epoch": 0.15020702945286935, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018723399191359607, + "loss": 0.1201, + "step": 17304 + }, + { + "epoch": 0.1502157099330735, + "grad_norm": 0.62109375, + "learning_rate": 0.0018723246389687651, + "loss": 0.0869, + "step": 17305 + }, + { + "epoch": 0.15022439041327768, + "grad_norm": 0.42578125, + "learning_rate": 0.0018723093579569672, + "loss": 0.0898, + "step": 17306 + }, + { + "epoch": 0.15023307089348184, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018722940761005831, + "loss": 0.0996, + "step": 17307 + }, + { + "epoch": 0.150241751373686, + "grad_norm": 0.8828125, + "learning_rate": 0.0018722787933996299, + "loss": 0.3867, + "step": 17308 + }, + { + "epoch": 0.15025043185389017, + "grad_norm": 0.181640625, + "learning_rate": 0.0018722635098541243, + "loss": 0.0933, + "step": 17309 + }, + { + "epoch": 0.15025911233409434, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018722482254640827, + "loss": 0.1328, + "step": 17310 + }, + { + "epoch": 0.15026779281429847, + "grad_norm": 0.51171875, + "learning_rate": 0.0018722329402295222, + "loss": 0.0996, + "step": 17311 + }, + { + "epoch": 0.15027647329450264, + "grad_norm": 0.65625, + "learning_rate": 0.0018722176541504593, + "loss": 0.1572, + "step": 17312 + }, + { + "epoch": 0.1502851537747068, + "grad_norm": 1.234375, + "learning_rate": 0.0018722023672269107, + "loss": 0.1523, + "step": 17313 + }, + { + "epoch": 0.15029383425491097, + "grad_norm": 0.65234375, + "learning_rate": 0.001872187079458893, + "loss": 0.1055, + "step": 17314 + }, + { + "epoch": 0.15030251473511513, + "grad_norm": 0.1015625, + "learning_rate": 0.0018721717908464235, + "loss": 0.1113, + "step": 17315 + }, + { + "epoch": 0.1503111952153193, + "grad_norm": 0.455078125, + "learning_rate": 0.0018721565013895185, + "loss": 0.1338, + "step": 17316 + }, + { + "epoch": 0.15031987569552346, + "grad_norm": 0.91796875, + "learning_rate": 0.0018721412110881946, + "loss": 0.1338, + "step": 17317 + }, + { + "epoch": 0.15032855617572763, + "grad_norm": 0.337890625, + "learning_rate": 0.0018721259199424686, + "loss": 0.1162, + "step": 17318 + }, + { + "epoch": 0.1503372366559318, + "grad_norm": 0.68359375, + "learning_rate": 0.0018721106279523576, + "loss": 0.1445, + "step": 17319 + }, + { + "epoch": 0.15034591713613596, + "grad_norm": 0.447265625, + "learning_rate": 0.0018720953351178777, + "loss": 0.0903, + "step": 17320 + }, + { + "epoch": 0.15035459761634012, + "grad_norm": 1.0234375, + "learning_rate": 0.0018720800414390463, + "loss": 0.1084, + "step": 17321 + }, + { + "epoch": 0.1503632780965443, + "grad_norm": 1.0390625, + "learning_rate": 0.0018720647469158795, + "loss": 0.1348, + "step": 17322 + }, + { + "epoch": 0.15037195857674845, + "grad_norm": 0.47265625, + "learning_rate": 0.0018720494515483944, + "loss": 0.0928, + "step": 17323 + }, + { + "epoch": 0.15038063905695262, + "grad_norm": 0.185546875, + "learning_rate": 0.0018720341553366074, + "loss": 0.1318, + "step": 17324 + }, + { + "epoch": 0.15038931953715678, + "grad_norm": 0.671875, + "learning_rate": 0.0018720188582805357, + "loss": 0.0874, + "step": 17325 + }, + { + "epoch": 0.15039800001736095, + "grad_norm": 0.08056640625, + "learning_rate": 0.001872003560380196, + "loss": 0.1064, + "step": 17326 + }, + { + "epoch": 0.15040668049756512, + "grad_norm": 0.61328125, + "learning_rate": 0.0018719882616356042, + "loss": 0.1133, + "step": 17327 + }, + { + "epoch": 0.15041536097776928, + "grad_norm": 0.51171875, + "learning_rate": 0.001871972962046778, + "loss": 0.1309, + "step": 17328 + }, + { + "epoch": 0.15042404145797345, + "grad_norm": 0.26171875, + "learning_rate": 0.0018719576616137335, + "loss": 0.1221, + "step": 17329 + }, + { + "epoch": 0.1504327219381776, + "grad_norm": 0.640625, + "learning_rate": 0.001871942360336488, + "loss": 0.1021, + "step": 17330 + }, + { + "epoch": 0.15044140241838178, + "grad_norm": 0.322265625, + "learning_rate": 0.0018719270582150578, + "loss": 0.0977, + "step": 17331 + }, + { + "epoch": 0.15045008289858594, + "grad_norm": 0.69140625, + "learning_rate": 0.00187191175524946, + "loss": 0.1064, + "step": 17332 + }, + { + "epoch": 0.1504587633787901, + "grad_norm": 0.15625, + "learning_rate": 0.001871896451439711, + "loss": 0.1191, + "step": 17333 + }, + { + "epoch": 0.15046744385899427, + "grad_norm": 0.21484375, + "learning_rate": 0.0018718811467858274, + "loss": 0.0947, + "step": 17334 + }, + { + "epoch": 0.15047612433919844, + "grad_norm": 0.271484375, + "learning_rate": 0.0018718658412878262, + "loss": 0.1045, + "step": 17335 + }, + { + "epoch": 0.1504848048194026, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018718505349457244, + "loss": 0.1152, + "step": 17336 + }, + { + "epoch": 0.15049348529960677, + "grad_norm": 0.451171875, + "learning_rate": 0.001871835227759538, + "loss": 0.1367, + "step": 17337 + }, + { + "epoch": 0.15050216577981093, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018718199197292847, + "loss": 0.1367, + "step": 17338 + }, + { + "epoch": 0.1505108462600151, + "grad_norm": 0.388671875, + "learning_rate": 0.0018718046108549804, + "loss": 0.0986, + "step": 17339 + }, + { + "epoch": 0.15051952674021926, + "grad_norm": 0.5234375, + "learning_rate": 0.0018717893011366423, + "loss": 0.1162, + "step": 17340 + }, + { + "epoch": 0.15052820722042343, + "grad_norm": 0.119140625, + "learning_rate": 0.0018717739905742873, + "loss": 0.104, + "step": 17341 + }, + { + "epoch": 0.1505368877006276, + "grad_norm": 0.087890625, + "learning_rate": 0.0018717586791679314, + "loss": 0.1289, + "step": 17342 + }, + { + "epoch": 0.15054556818083176, + "grad_norm": 0.474609375, + "learning_rate": 0.001871743366917592, + "loss": 0.1729, + "step": 17343 + }, + { + "epoch": 0.15055424866103592, + "grad_norm": 0.1025390625, + "learning_rate": 0.0018717280538232855, + "loss": 0.103, + "step": 17344 + }, + { + "epoch": 0.1505629291412401, + "grad_norm": 0.16015625, + "learning_rate": 0.0018717127398850289, + "loss": 0.127, + "step": 17345 + }, + { + "epoch": 0.15057160962144425, + "grad_norm": 0.146484375, + "learning_rate": 0.0018716974251028389, + "loss": 0.0977, + "step": 17346 + }, + { + "epoch": 0.15058029010164842, + "grad_norm": 1.28125, + "learning_rate": 0.001871682109476732, + "loss": 0.1426, + "step": 17347 + }, + { + "epoch": 0.15058897058185258, + "grad_norm": 0.49609375, + "learning_rate": 0.0018716667930067253, + "loss": 0.0903, + "step": 17348 + }, + { + "epoch": 0.15059765106205675, + "grad_norm": 0.427734375, + "learning_rate": 0.0018716514756928354, + "loss": 0.123, + "step": 17349 + }, + { + "epoch": 0.1506063315422609, + "grad_norm": 0.447265625, + "learning_rate": 0.001871636157535079, + "loss": 0.1113, + "step": 17350 + }, + { + "epoch": 0.15061501202246508, + "grad_norm": 0.62109375, + "learning_rate": 0.001871620838533473, + "loss": 0.0898, + "step": 17351 + }, + { + "epoch": 0.15062369250266924, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018716055186880339, + "loss": 0.127, + "step": 17352 + }, + { + "epoch": 0.1506323729828734, + "grad_norm": 0.302734375, + "learning_rate": 0.0018715901979987786, + "loss": 0.123, + "step": 17353 + }, + { + "epoch": 0.15064105346307757, + "grad_norm": 0.119140625, + "learning_rate": 0.0018715748764657242, + "loss": 0.125, + "step": 17354 + }, + { + "epoch": 0.15064973394328174, + "grad_norm": 0.76171875, + "learning_rate": 0.0018715595540888868, + "loss": 0.1523, + "step": 17355 + }, + { + "epoch": 0.1506584144234859, + "grad_norm": 0.56640625, + "learning_rate": 0.001871544230868283, + "loss": 0.1367, + "step": 17356 + }, + { + "epoch": 0.15066709490369007, + "grad_norm": 1.2421875, + "learning_rate": 0.0018715289068039307, + "loss": 0.1514, + "step": 17357 + }, + { + "epoch": 0.15067577538389423, + "grad_norm": 0.416015625, + "learning_rate": 0.001871513581895846, + "loss": 0.0967, + "step": 17358 + }, + { + "epoch": 0.1506844558640984, + "grad_norm": 0.271484375, + "learning_rate": 0.001871498256144045, + "loss": 0.083, + "step": 17359 + }, + { + "epoch": 0.15069313634430256, + "grad_norm": 0.16015625, + "learning_rate": 0.0018714829295485458, + "loss": 0.1182, + "step": 17360 + }, + { + "epoch": 0.15070181682450673, + "grad_norm": 0.37109375, + "learning_rate": 0.0018714676021093642, + "loss": 0.209, + "step": 17361 + }, + { + "epoch": 0.1507104973047109, + "grad_norm": 0.48046875, + "learning_rate": 0.0018714522738265172, + "loss": 0.1582, + "step": 17362 + }, + { + "epoch": 0.15071917778491506, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018714369447000213, + "loss": 0.1064, + "step": 17363 + }, + { + "epoch": 0.15072785826511922, + "grad_norm": 0.11279296875, + "learning_rate": 0.001871421614729894, + "loss": 0.1211, + "step": 17364 + }, + { + "epoch": 0.1507365387453234, + "grad_norm": 0.24609375, + "learning_rate": 0.0018714062839161514, + "loss": 0.1221, + "step": 17365 + }, + { + "epoch": 0.15074521922552755, + "grad_norm": 0.30078125, + "learning_rate": 0.0018713909522588106, + "loss": 0.1084, + "step": 17366 + }, + { + "epoch": 0.15075389970573172, + "grad_norm": 1.328125, + "learning_rate": 0.0018713756197578882, + "loss": 0.1177, + "step": 17367 + }, + { + "epoch": 0.15076258018593588, + "grad_norm": 0.271484375, + "learning_rate": 0.0018713602864134007, + "loss": 0.1069, + "step": 17368 + }, + { + "epoch": 0.15077126066614005, + "grad_norm": 0.515625, + "learning_rate": 0.0018713449522253654, + "loss": 0.0981, + "step": 17369 + }, + { + "epoch": 0.15077994114634422, + "grad_norm": 0.140625, + "learning_rate": 0.001871329617193799, + "loss": 0.1318, + "step": 17370 + }, + { + "epoch": 0.15078862162654838, + "grad_norm": 0.201171875, + "learning_rate": 0.0018713142813187181, + "loss": 0.0986, + "step": 17371 + }, + { + "epoch": 0.15079730210675255, + "grad_norm": 0.30859375, + "learning_rate": 0.0018712989446001394, + "loss": 0.126, + "step": 17372 + }, + { + "epoch": 0.1508059825869567, + "grad_norm": 0.384765625, + "learning_rate": 0.0018712836070380796, + "loss": 0.1045, + "step": 17373 + }, + { + "epoch": 0.15081466306716088, + "grad_norm": 0.21875, + "learning_rate": 0.0018712682686325558, + "loss": 0.0918, + "step": 17374 + }, + { + "epoch": 0.15082334354736504, + "grad_norm": 0.515625, + "learning_rate": 0.0018712529293835847, + "loss": 0.1387, + "step": 17375 + }, + { + "epoch": 0.1508320240275692, + "grad_norm": 0.306640625, + "learning_rate": 0.001871237589291183, + "loss": 0.168, + "step": 17376 + }, + { + "epoch": 0.15084070450777337, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018712222483553672, + "loss": 0.1138, + "step": 17377 + }, + { + "epoch": 0.15084938498797754, + "grad_norm": 0.419921875, + "learning_rate": 0.0018712069065761545, + "loss": 0.1406, + "step": 17378 + }, + { + "epoch": 0.1508580654681817, + "grad_norm": 0.150390625, + "learning_rate": 0.0018711915639535617, + "loss": 0.0933, + "step": 17379 + }, + { + "epoch": 0.15086674594838587, + "grad_norm": 0.291015625, + "learning_rate": 0.0018711762204876052, + "loss": 0.0806, + "step": 17380 + }, + { + "epoch": 0.15087542642859003, + "grad_norm": 0.3125, + "learning_rate": 0.0018711608761783022, + "loss": 0.1138, + "step": 17381 + }, + { + "epoch": 0.1508841069087942, + "grad_norm": 1.484375, + "learning_rate": 0.0018711455310256686, + "loss": 0.1475, + "step": 17382 + }, + { + "epoch": 0.15089278738899836, + "grad_norm": 0.353515625, + "learning_rate": 0.0018711301850297224, + "loss": 0.0977, + "step": 17383 + }, + { + "epoch": 0.15090146786920253, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018711148381904796, + "loss": 0.0967, + "step": 17384 + }, + { + "epoch": 0.1509101483494067, + "grad_norm": 0.255859375, + "learning_rate": 0.0018710994905079572, + "loss": 0.1484, + "step": 17385 + }, + { + "epoch": 0.15091882882961086, + "grad_norm": 0.578125, + "learning_rate": 0.0018710841419821722, + "loss": 0.124, + "step": 17386 + }, + { + "epoch": 0.15092750930981502, + "grad_norm": 1.2890625, + "learning_rate": 0.0018710687926131412, + "loss": 0.1953, + "step": 17387 + }, + { + "epoch": 0.1509361897900192, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018710534424008806, + "loss": 0.1172, + "step": 17388 + }, + { + "epoch": 0.15094487027022335, + "grad_norm": 0.68359375, + "learning_rate": 0.0018710380913454078, + "loss": 0.1475, + "step": 17389 + }, + { + "epoch": 0.15095355075042752, + "grad_norm": 0.5234375, + "learning_rate": 0.0018710227394467393, + "loss": 0.1309, + "step": 17390 + }, + { + "epoch": 0.15096223123063168, + "grad_norm": 0.345703125, + "learning_rate": 0.0018710073867048915, + "loss": 0.0811, + "step": 17391 + }, + { + "epoch": 0.15097091171083585, + "grad_norm": 0.3828125, + "learning_rate": 0.0018709920331198822, + "loss": 0.1104, + "step": 17392 + }, + { + "epoch": 0.15097959219104, + "grad_norm": 0.1240234375, + "learning_rate": 0.0018709766786917273, + "loss": 0.1006, + "step": 17393 + }, + { + "epoch": 0.15098827267124418, + "grad_norm": 0.29296875, + "learning_rate": 0.001870961323420444, + "loss": 0.1279, + "step": 17394 + }, + { + "epoch": 0.15099695315144834, + "grad_norm": 0.46484375, + "learning_rate": 0.0018709459673060492, + "loss": 0.1025, + "step": 17395 + }, + { + "epoch": 0.1510056336316525, + "grad_norm": 0.4296875, + "learning_rate": 0.001870930610348559, + "loss": 0.1211, + "step": 17396 + }, + { + "epoch": 0.15101431411185667, + "grad_norm": 0.314453125, + "learning_rate": 0.001870915252547991, + "loss": 0.1504, + "step": 17397 + }, + { + "epoch": 0.15102299459206084, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018708998939043614, + "loss": 0.1074, + "step": 17398 + }, + { + "epoch": 0.151031675072265, + "grad_norm": 0.6015625, + "learning_rate": 0.0018708845344176878, + "loss": 0.1123, + "step": 17399 + }, + { + "epoch": 0.15104035555246917, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018708691740879858, + "loss": 0.1445, + "step": 17400 + }, + { + "epoch": 0.15104903603267333, + "grad_norm": 0.29296875, + "learning_rate": 0.0018708538129152733, + "loss": 0.1006, + "step": 17401 + }, + { + "epoch": 0.1510577165128775, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018708384508995663, + "loss": 0.1123, + "step": 17402 + }, + { + "epoch": 0.15106639699308166, + "grad_norm": 0.1240234375, + "learning_rate": 0.001870823088040882, + "loss": 0.0864, + "step": 17403 + }, + { + "epoch": 0.15107507747328583, + "grad_norm": 0.154296875, + "learning_rate": 0.0018708077243392372, + "loss": 0.1011, + "step": 17404 + }, + { + "epoch": 0.15108375795349, + "grad_norm": 0.41796875, + "learning_rate": 0.0018707923597946487, + "loss": 0.126, + "step": 17405 + }, + { + "epoch": 0.15109243843369416, + "grad_norm": 0.7265625, + "learning_rate": 0.0018707769944071332, + "loss": 0.1133, + "step": 17406 + }, + { + "epoch": 0.15110111891389832, + "grad_norm": 0.07666015625, + "learning_rate": 0.0018707616281767078, + "loss": 0.1011, + "step": 17407 + }, + { + "epoch": 0.1511097993941025, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018707462611033889, + "loss": 0.1328, + "step": 17408 + }, + { + "epoch": 0.15111847987430665, + "grad_norm": 0.248046875, + "learning_rate": 0.0018707308931871936, + "loss": 0.0889, + "step": 17409 + }, + { + "epoch": 0.15112716035451082, + "grad_norm": 0.19140625, + "learning_rate": 0.0018707155244281383, + "loss": 0.1582, + "step": 17410 + }, + { + "epoch": 0.15113584083471499, + "grad_norm": 0.42578125, + "learning_rate": 0.00187070015482624, + "loss": 0.1094, + "step": 17411 + }, + { + "epoch": 0.15114452131491915, + "grad_norm": 0.1240234375, + "learning_rate": 0.001870684784381516, + "loss": 0.166, + "step": 17412 + }, + { + "epoch": 0.15115320179512332, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018706694130939828, + "loss": 0.1016, + "step": 17413 + }, + { + "epoch": 0.15116188227532748, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018706540409636565, + "loss": 0.1016, + "step": 17414 + }, + { + "epoch": 0.15117056275553165, + "grad_norm": 0.31640625, + "learning_rate": 0.001870638667990555, + "loss": 0.0879, + "step": 17415 + }, + { + "epoch": 0.1511792432357358, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018706232941746944, + "loss": 0.0933, + "step": 17416 + }, + { + "epoch": 0.15118792371593998, + "grad_norm": 0.95703125, + "learning_rate": 0.0018706079195160916, + "loss": 0.1445, + "step": 17417 + }, + { + "epoch": 0.15119660419614414, + "grad_norm": 0.0830078125, + "learning_rate": 0.001870592544014764, + "loss": 0.1543, + "step": 17418 + }, + { + "epoch": 0.1512052846763483, + "grad_norm": 0.61328125, + "learning_rate": 0.0018705771676707278, + "loss": 0.0962, + "step": 17419 + }, + { + "epoch": 0.15121396515655247, + "grad_norm": 0.287109375, + "learning_rate": 0.001870561790484, + "loss": 0.1836, + "step": 17420 + }, + { + "epoch": 0.15122264563675664, + "grad_norm": 0.365234375, + "learning_rate": 0.0018705464124545974, + "loss": 0.125, + "step": 17421 + }, + { + "epoch": 0.1512313261169608, + "grad_norm": 0.255859375, + "learning_rate": 0.0018705310335825366, + "loss": 0.0791, + "step": 17422 + }, + { + "epoch": 0.15124000659716497, + "grad_norm": 0.1689453125, + "learning_rate": 0.001870515653867835, + "loss": 0.1357, + "step": 17423 + }, + { + "epoch": 0.15124868707736913, + "grad_norm": 1.296875, + "learning_rate": 0.0018705002733105088, + "loss": 0.1934, + "step": 17424 + }, + { + "epoch": 0.1512573675575733, + "grad_norm": 0.216796875, + "learning_rate": 0.001870484891910575, + "loss": 0.0801, + "step": 17425 + }, + { + "epoch": 0.15126604803777746, + "grad_norm": 0.37109375, + "learning_rate": 0.0018704695096680506, + "loss": 0.103, + "step": 17426 + }, + { + "epoch": 0.15127472851798163, + "grad_norm": 0.8125, + "learning_rate": 0.0018704541265829524, + "loss": 0.1309, + "step": 17427 + }, + { + "epoch": 0.1512834089981858, + "grad_norm": 0.28125, + "learning_rate": 0.0018704387426552972, + "loss": 0.0874, + "step": 17428 + }, + { + "epoch": 0.15129208947838996, + "grad_norm": 0.265625, + "learning_rate": 0.0018704233578851015, + "loss": 0.0986, + "step": 17429 + }, + { + "epoch": 0.15130076995859412, + "grad_norm": 1.1953125, + "learning_rate": 0.0018704079722723825, + "loss": 0.1299, + "step": 17430 + }, + { + "epoch": 0.1513094504387983, + "grad_norm": 0.298828125, + "learning_rate": 0.0018703925858171572, + "loss": 0.1309, + "step": 17431 + }, + { + "epoch": 0.15131813091900245, + "grad_norm": 0.146484375, + "learning_rate": 0.0018703771985194417, + "loss": 0.0962, + "step": 17432 + }, + { + "epoch": 0.15132681139920662, + "grad_norm": 0.076171875, + "learning_rate": 0.0018703618103792536, + "loss": 0.105, + "step": 17433 + }, + { + "epoch": 0.15133549187941078, + "grad_norm": 0.4453125, + "learning_rate": 0.0018703464213966092, + "loss": 0.1543, + "step": 17434 + }, + { + "epoch": 0.15134417235961492, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018703310315715256, + "loss": 0.1523, + "step": 17435 + }, + { + "epoch": 0.15135285283981909, + "grad_norm": 0.1689453125, + "learning_rate": 0.0018703156409040197, + "loss": 0.1416, + "step": 17436 + }, + { + "epoch": 0.15136153332002325, + "grad_norm": 0.408203125, + "learning_rate": 0.001870300249394108, + "loss": 0.1094, + "step": 17437 + }, + { + "epoch": 0.15137021380022742, + "grad_norm": 0.72265625, + "learning_rate": 0.0018702848570418076, + "loss": 0.1582, + "step": 17438 + }, + { + "epoch": 0.15137889428043158, + "grad_norm": 1.3359375, + "learning_rate": 0.0018702694638471353, + "loss": 0.1719, + "step": 17439 + }, + { + "epoch": 0.15138757476063575, + "grad_norm": 0.3828125, + "learning_rate": 0.0018702540698101078, + "loss": 0.1406, + "step": 17440 + }, + { + "epoch": 0.1513962552408399, + "grad_norm": 0.484375, + "learning_rate": 0.0018702386749307421, + "loss": 0.0986, + "step": 17441 + }, + { + "epoch": 0.15140493572104408, + "grad_norm": 1.0546875, + "learning_rate": 0.001870223279209055, + "loss": 0.168, + "step": 17442 + }, + { + "epoch": 0.15141361620124824, + "grad_norm": 0.2236328125, + "learning_rate": 0.001870207882645063, + "loss": 0.1357, + "step": 17443 + }, + { + "epoch": 0.1514222966814524, + "grad_norm": 0.5546875, + "learning_rate": 0.0018701924852387835, + "loss": 0.1177, + "step": 17444 + }, + { + "epoch": 0.15143097716165657, + "grad_norm": 0.25390625, + "learning_rate": 0.001870177086990233, + "loss": 0.1289, + "step": 17445 + }, + { + "epoch": 0.15143965764186074, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018701616878994286, + "loss": 0.1025, + "step": 17446 + }, + { + "epoch": 0.1514483381220649, + "grad_norm": 0.265625, + "learning_rate": 0.0018701462879663867, + "loss": 0.083, + "step": 17447 + }, + { + "epoch": 0.15145701860226907, + "grad_norm": 0.220703125, + "learning_rate": 0.0018701308871911242, + "loss": 0.1348, + "step": 17448 + }, + { + "epoch": 0.15146569908247323, + "grad_norm": 0.32421875, + "learning_rate": 0.0018701154855736586, + "loss": 0.1338, + "step": 17449 + }, + { + "epoch": 0.1514743795626774, + "grad_norm": 0.138671875, + "learning_rate": 0.001870100083114006, + "loss": 0.1396, + "step": 17450 + }, + { + "epoch": 0.15148306004288156, + "grad_norm": 0.177734375, + "learning_rate": 0.0018700846798121835, + "loss": 0.0903, + "step": 17451 + }, + { + "epoch": 0.15149174052308573, + "grad_norm": 0.35546875, + "learning_rate": 0.0018700692756682081, + "loss": 0.1191, + "step": 17452 + }, + { + "epoch": 0.1515004210032899, + "grad_norm": 0.0859375, + "learning_rate": 0.0018700538706820965, + "loss": 0.1006, + "step": 17453 + }, + { + "epoch": 0.15150910148349406, + "grad_norm": 1.4296875, + "learning_rate": 0.0018700384648538653, + "loss": 0.1484, + "step": 17454 + }, + { + "epoch": 0.15151778196369822, + "grad_norm": 0.1845703125, + "learning_rate": 0.001870023058183532, + "loss": 0.0869, + "step": 17455 + }, + { + "epoch": 0.1515264624439024, + "grad_norm": 0.310546875, + "learning_rate": 0.001870007650671113, + "loss": 0.1113, + "step": 17456 + }, + { + "epoch": 0.15153514292410655, + "grad_norm": 0.515625, + "learning_rate": 0.0018699922423166249, + "loss": 0.084, + "step": 17457 + }, + { + "epoch": 0.15154382340431072, + "grad_norm": 0.140625, + "learning_rate": 0.001869976833120085, + "loss": 0.1514, + "step": 17458 + }, + { + "epoch": 0.15155250388451488, + "grad_norm": 0.48046875, + "learning_rate": 0.00186996142308151, + "loss": 0.127, + "step": 17459 + }, + { + "epoch": 0.15156118436471905, + "grad_norm": 0.47265625, + "learning_rate": 0.0018699460122009169, + "loss": 0.1074, + "step": 17460 + }, + { + "epoch": 0.1515698648449232, + "grad_norm": 0.447265625, + "learning_rate": 0.0018699306004783223, + "loss": 0.1279, + "step": 17461 + }, + { + "epoch": 0.15157854532512738, + "grad_norm": 0.244140625, + "learning_rate": 0.0018699151879137434, + "loss": 0.1113, + "step": 17462 + }, + { + "epoch": 0.15158722580533154, + "grad_norm": 0.3046875, + "learning_rate": 0.0018698997745071962, + "loss": 0.1191, + "step": 17463 + }, + { + "epoch": 0.1515959062855357, + "grad_norm": 0.271484375, + "learning_rate": 0.0018698843602586986, + "loss": 0.1235, + "step": 17464 + }, + { + "epoch": 0.15160458676573987, + "grad_norm": 0.150390625, + "learning_rate": 0.001869868945168267, + "loss": 0.1406, + "step": 17465 + }, + { + "epoch": 0.15161326724594404, + "grad_norm": 0.546875, + "learning_rate": 0.0018698535292359184, + "loss": 0.1523, + "step": 17466 + }, + { + "epoch": 0.1516219477261482, + "grad_norm": 0.16796875, + "learning_rate": 0.0018698381124616695, + "loss": 0.0928, + "step": 17467 + }, + { + "epoch": 0.15163062820635237, + "grad_norm": 0.232421875, + "learning_rate": 0.001869822694845537, + "loss": 0.0728, + "step": 17468 + }, + { + "epoch": 0.15163930868655653, + "grad_norm": 0.640625, + "learning_rate": 0.001869807276387538, + "loss": 0.0991, + "step": 17469 + }, + { + "epoch": 0.1516479891667607, + "grad_norm": 0.4609375, + "learning_rate": 0.0018697918570876894, + "loss": 0.1016, + "step": 17470 + }, + { + "epoch": 0.15165666964696486, + "grad_norm": 0.24609375, + "learning_rate": 0.0018697764369460079, + "loss": 0.1338, + "step": 17471 + }, + { + "epoch": 0.15166535012716903, + "grad_norm": 0.09521484375, + "learning_rate": 0.0018697610159625108, + "loss": 0.0791, + "step": 17472 + }, + { + "epoch": 0.1516740306073732, + "grad_norm": 0.306640625, + "learning_rate": 0.0018697455941372144, + "loss": 0.1143, + "step": 17473 + }, + { + "epoch": 0.15168271108757736, + "grad_norm": 0.91015625, + "learning_rate": 0.0018697301714701356, + "loss": 0.1064, + "step": 17474 + }, + { + "epoch": 0.15169139156778152, + "grad_norm": 0.21875, + "learning_rate": 0.0018697147479612916, + "loss": 0.0986, + "step": 17475 + }, + { + "epoch": 0.1517000720479857, + "grad_norm": 0.357421875, + "learning_rate": 0.0018696993236106995, + "loss": 0.127, + "step": 17476 + }, + { + "epoch": 0.15170875252818986, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018696838984183753, + "loss": 0.0762, + "step": 17477 + }, + { + "epoch": 0.15171743300839402, + "grad_norm": 0.06005859375, + "learning_rate": 0.0018696684723843364, + "loss": 0.0874, + "step": 17478 + }, + { + "epoch": 0.15172611348859819, + "grad_norm": 0.314453125, + "learning_rate": 0.0018696530455085997, + "loss": 0.1299, + "step": 17479 + }, + { + "epoch": 0.15173479396880235, + "grad_norm": 0.25, + "learning_rate": 0.001869637617791182, + "loss": 0.0791, + "step": 17480 + }, + { + "epoch": 0.15174347444900652, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018696221892321, + "loss": 0.1377, + "step": 17481 + }, + { + "epoch": 0.15175215492921068, + "grad_norm": 0.13671875, + "learning_rate": 0.0018696067598313708, + "loss": 0.0986, + "step": 17482 + }, + { + "epoch": 0.15176083540941485, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018695913295890116, + "loss": 0.0889, + "step": 17483 + }, + { + "epoch": 0.151769515889619, + "grad_norm": 0.68359375, + "learning_rate": 0.0018695758985050384, + "loss": 0.1211, + "step": 17484 + }, + { + "epoch": 0.15177819636982318, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018695604665794688, + "loss": 0.0762, + "step": 17485 + }, + { + "epoch": 0.15178687685002734, + "grad_norm": 0.365234375, + "learning_rate": 0.0018695450338123195, + "loss": 0.1201, + "step": 17486 + }, + { + "epoch": 0.1517955573302315, + "grad_norm": 0.26171875, + "learning_rate": 0.001869529600203607, + "loss": 0.1475, + "step": 17487 + }, + { + "epoch": 0.15180423781043567, + "grad_norm": 0.609375, + "learning_rate": 0.0018695141657533488, + "loss": 0.1426, + "step": 17488 + }, + { + "epoch": 0.15181291829063984, + "grad_norm": 0.310546875, + "learning_rate": 0.0018694987304615613, + "loss": 0.0952, + "step": 17489 + }, + { + "epoch": 0.151821598770844, + "grad_norm": 0.703125, + "learning_rate": 0.0018694832943282615, + "loss": 0.1123, + "step": 17490 + }, + { + "epoch": 0.15183027925104817, + "grad_norm": 1.46875, + "learning_rate": 0.0018694678573534667, + "loss": 0.1455, + "step": 17491 + }, + { + "epoch": 0.15183895973125233, + "grad_norm": 0.08203125, + "learning_rate": 0.0018694524195371931, + "loss": 0.0898, + "step": 17492 + }, + { + "epoch": 0.1518476402114565, + "grad_norm": 0.177734375, + "learning_rate": 0.001869436980879458, + "loss": 0.1162, + "step": 17493 + }, + { + "epoch": 0.15185632069166066, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018694215413802783, + "loss": 0.1035, + "step": 17494 + }, + { + "epoch": 0.15186500117186483, + "grad_norm": 0.13671875, + "learning_rate": 0.0018694061010396704, + "loss": 0.0928, + "step": 17495 + }, + { + "epoch": 0.151873681652069, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018693906598576519, + "loss": 0.0996, + "step": 17496 + }, + { + "epoch": 0.15188236213227316, + "grad_norm": 0.453125, + "learning_rate": 0.001869375217834239, + "loss": 0.0967, + "step": 17497 + }, + { + "epoch": 0.15189104261247732, + "grad_norm": 0.416015625, + "learning_rate": 0.001869359774969449, + "loss": 0.0869, + "step": 17498 + }, + { + "epoch": 0.1518997230926815, + "grad_norm": 0.44140625, + "learning_rate": 0.001869344331263299, + "loss": 0.2012, + "step": 17499 + }, + { + "epoch": 0.15190840357288565, + "grad_norm": 0.29296875, + "learning_rate": 0.0018693288867158053, + "loss": 0.1104, + "step": 17500 + }, + { + "epoch": 0.15191708405308982, + "grad_norm": 0.30859375, + "learning_rate": 0.0018693134413269852, + "loss": 0.1094, + "step": 17501 + }, + { + "epoch": 0.15192576453329398, + "grad_norm": 0.26953125, + "learning_rate": 0.0018692979950968554, + "loss": 0.1108, + "step": 17502 + }, + { + "epoch": 0.15193444501349815, + "grad_norm": 0.6484375, + "learning_rate": 0.0018692825480254329, + "loss": 0.1104, + "step": 17503 + }, + { + "epoch": 0.1519431254937023, + "grad_norm": 0.6953125, + "learning_rate": 0.0018692671001127347, + "loss": 0.1191, + "step": 17504 + }, + { + "epoch": 0.15195180597390648, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018692516513587774, + "loss": 0.1113, + "step": 17505 + }, + { + "epoch": 0.15196048645411064, + "grad_norm": 0.236328125, + "learning_rate": 0.0018692362017635783, + "loss": 0.0972, + "step": 17506 + }, + { + "epoch": 0.1519691669343148, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018692207513271537, + "loss": 0.127, + "step": 17507 + }, + { + "epoch": 0.15197784741451897, + "grad_norm": 0.51171875, + "learning_rate": 0.001869205300049521, + "loss": 0.1211, + "step": 17508 + }, + { + "epoch": 0.15198652789472314, + "grad_norm": 0.24609375, + "learning_rate": 0.0018691898479306968, + "loss": 0.0967, + "step": 17509 + }, + { + "epoch": 0.1519952083749273, + "grad_norm": 0.609375, + "learning_rate": 0.0018691743949706984, + "loss": 0.1289, + "step": 17510 + }, + { + "epoch": 0.15200388885513147, + "grad_norm": 0.251953125, + "learning_rate": 0.0018691589411695424, + "loss": 0.1128, + "step": 17511 + }, + { + "epoch": 0.15201256933533563, + "grad_norm": 1.171875, + "learning_rate": 0.0018691434865272456, + "loss": 0.1396, + "step": 17512 + }, + { + "epoch": 0.1520212498155398, + "grad_norm": 0.87890625, + "learning_rate": 0.001869128031043825, + "loss": 0.1289, + "step": 17513 + }, + { + "epoch": 0.15202993029574396, + "grad_norm": 0.19921875, + "learning_rate": 0.0018691125747192976, + "loss": 0.1318, + "step": 17514 + }, + { + "epoch": 0.15203861077594813, + "grad_norm": 0.62890625, + "learning_rate": 0.0018690971175536807, + "loss": 0.1504, + "step": 17515 + }, + { + "epoch": 0.1520472912561523, + "grad_norm": 0.6328125, + "learning_rate": 0.0018690816595469902, + "loss": 0.1816, + "step": 17516 + }, + { + "epoch": 0.15205597173635646, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018690662006992436, + "loss": 0.1201, + "step": 17517 + }, + { + "epoch": 0.15206465221656063, + "grad_norm": 0.30078125, + "learning_rate": 0.0018690507410104578, + "loss": 0.1113, + "step": 17518 + }, + { + "epoch": 0.1520733326967648, + "grad_norm": 0.396484375, + "learning_rate": 0.0018690352804806496, + "loss": 0.0908, + "step": 17519 + }, + { + "epoch": 0.15208201317696896, + "grad_norm": 0.126953125, + "learning_rate": 0.0018690198191098361, + "loss": 0.1162, + "step": 17520 + }, + { + "epoch": 0.15209069365717312, + "grad_norm": 0.353515625, + "learning_rate": 0.0018690043568980344, + "loss": 0.1104, + "step": 17521 + }, + { + "epoch": 0.15209937413737729, + "grad_norm": 0.412109375, + "learning_rate": 0.0018689888938452604, + "loss": 0.0918, + "step": 17522 + }, + { + "epoch": 0.15210805461758145, + "grad_norm": 0.396484375, + "learning_rate": 0.0018689734299515322, + "loss": 0.1016, + "step": 17523 + }, + { + "epoch": 0.15211673509778562, + "grad_norm": 0.65625, + "learning_rate": 0.001868957965216866, + "loss": 0.1025, + "step": 17524 + }, + { + "epoch": 0.15212541557798978, + "grad_norm": 0.36328125, + "learning_rate": 0.0018689424996412793, + "loss": 0.1177, + "step": 17525 + }, + { + "epoch": 0.15213409605819395, + "grad_norm": 0.1962890625, + "learning_rate": 0.001868927033224788, + "loss": 0.1162, + "step": 17526 + }, + { + "epoch": 0.1521427765383981, + "grad_norm": 0.66796875, + "learning_rate": 0.0018689115659674099, + "loss": 0.1104, + "step": 17527 + }, + { + "epoch": 0.15215145701860228, + "grad_norm": 0.63671875, + "learning_rate": 0.0018688960978691619, + "loss": 0.1338, + "step": 17528 + }, + { + "epoch": 0.15216013749880644, + "grad_norm": 0.546875, + "learning_rate": 0.0018688806289300604, + "loss": 0.1123, + "step": 17529 + }, + { + "epoch": 0.1521688179790106, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018688651591501227, + "loss": 0.1426, + "step": 17530 + }, + { + "epoch": 0.15217749845921477, + "grad_norm": 0.1123046875, + "learning_rate": 0.0018688496885293658, + "loss": 0.0737, + "step": 17531 + }, + { + "epoch": 0.15218617893941894, + "grad_norm": 0.62109375, + "learning_rate": 0.001868834217067806, + "loss": 0.0825, + "step": 17532 + }, + { + "epoch": 0.1521948594196231, + "grad_norm": 0.373046875, + "learning_rate": 0.001868818744765461, + "loss": 0.0947, + "step": 17533 + }, + { + "epoch": 0.15220353989982727, + "grad_norm": 0.376953125, + "learning_rate": 0.0018688032716223473, + "loss": 0.0879, + "step": 17534 + }, + { + "epoch": 0.15221222038003143, + "grad_norm": 0.54296875, + "learning_rate": 0.0018687877976384817, + "loss": 0.1055, + "step": 17535 + }, + { + "epoch": 0.1522209008602356, + "grad_norm": 0.462890625, + "learning_rate": 0.0018687723228138818, + "loss": 0.1328, + "step": 17536 + }, + { + "epoch": 0.15222958134043976, + "grad_norm": 0.55078125, + "learning_rate": 0.0018687568471485636, + "loss": 0.1211, + "step": 17537 + }, + { + "epoch": 0.15223826182064393, + "grad_norm": 0.419921875, + "learning_rate": 0.0018687413706425444, + "loss": 0.1055, + "step": 17538 + }, + { + "epoch": 0.1522469423008481, + "grad_norm": 0.234375, + "learning_rate": 0.0018687258932958415, + "loss": 0.1543, + "step": 17539 + }, + { + "epoch": 0.15225562278105226, + "grad_norm": 0.95703125, + "learning_rate": 0.0018687104151084715, + "loss": 0.1387, + "step": 17540 + }, + { + "epoch": 0.15226430326125642, + "grad_norm": 0.609375, + "learning_rate": 0.001868694936080451, + "loss": 0.0957, + "step": 17541 + }, + { + "epoch": 0.1522729837414606, + "grad_norm": 0.365234375, + "learning_rate": 0.0018686794562117978, + "loss": 0.1357, + "step": 17542 + }, + { + "epoch": 0.15228166422166475, + "grad_norm": 0.37890625, + "learning_rate": 0.0018686639755025282, + "loss": 0.127, + "step": 17543 + }, + { + "epoch": 0.15229034470186892, + "grad_norm": 0.3828125, + "learning_rate": 0.0018686484939526588, + "loss": 0.0952, + "step": 17544 + }, + { + "epoch": 0.15229902518207308, + "grad_norm": 0.58984375, + "learning_rate": 0.0018686330115622073, + "loss": 0.1206, + "step": 17545 + }, + { + "epoch": 0.15230770566227725, + "grad_norm": 1.0703125, + "learning_rate": 0.00186861752833119, + "loss": 0.1836, + "step": 17546 + }, + { + "epoch": 0.1523163861424814, + "grad_norm": 0.263671875, + "learning_rate": 0.0018686020442596244, + "loss": 0.1064, + "step": 17547 + }, + { + "epoch": 0.15232506662268558, + "grad_norm": 0.3828125, + "learning_rate": 0.0018685865593475268, + "loss": 0.0928, + "step": 17548 + }, + { + "epoch": 0.15233374710288974, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018685710735949147, + "loss": 0.1396, + "step": 17549 + }, + { + "epoch": 0.1523424275830939, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018685555870018048, + "loss": 0.1055, + "step": 17550 + }, + { + "epoch": 0.15235110806329807, + "grad_norm": 0.25390625, + "learning_rate": 0.0018685400995682143, + "loss": 0.0986, + "step": 17551 + }, + { + "epoch": 0.15235978854350224, + "grad_norm": 0.24609375, + "learning_rate": 0.0018685246112941597, + "loss": 0.0977, + "step": 17552 + }, + { + "epoch": 0.1523684690237064, + "grad_norm": 0.126953125, + "learning_rate": 0.0018685091221796583, + "loss": 0.0894, + "step": 17553 + }, + { + "epoch": 0.15237714950391057, + "grad_norm": 0.330078125, + "learning_rate": 0.0018684936322247263, + "loss": 0.1196, + "step": 17554 + }, + { + "epoch": 0.15238582998411473, + "grad_norm": 0.1357421875, + "learning_rate": 0.001868478141429382, + "loss": 0.1016, + "step": 17555 + }, + { + "epoch": 0.1523945104643189, + "grad_norm": 0.302734375, + "learning_rate": 0.001868462649793641, + "loss": 0.1089, + "step": 17556 + }, + { + "epoch": 0.15240319094452306, + "grad_norm": 0.314453125, + "learning_rate": 0.001868447157317521, + "loss": 0.126, + "step": 17557 + }, + { + "epoch": 0.1524118714247272, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018684316640010389, + "loss": 0.1553, + "step": 17558 + }, + { + "epoch": 0.15242055190493137, + "grad_norm": 0.279296875, + "learning_rate": 0.001868416169844211, + "loss": 0.1357, + "step": 17559 + }, + { + "epoch": 0.15242923238513553, + "grad_norm": 0.169921875, + "learning_rate": 0.001868400674847055, + "loss": 0.1138, + "step": 17560 + }, + { + "epoch": 0.1524379128653397, + "grad_norm": 0.455078125, + "learning_rate": 0.0018683851790095873, + "loss": 0.0811, + "step": 17561 + }, + { + "epoch": 0.15244659334554386, + "grad_norm": 0.1796875, + "learning_rate": 0.0018683696823318257, + "loss": 0.1045, + "step": 17562 + }, + { + "epoch": 0.15245527382574803, + "grad_norm": 0.248046875, + "learning_rate": 0.001868354184813786, + "loss": 0.1445, + "step": 17563 + }, + { + "epoch": 0.1524639543059522, + "grad_norm": 0.07568359375, + "learning_rate": 0.0018683386864554862, + "loss": 0.124, + "step": 17564 + }, + { + "epoch": 0.15247263478615636, + "grad_norm": 0.46484375, + "learning_rate": 0.0018683231872569422, + "loss": 0.0928, + "step": 17565 + }, + { + "epoch": 0.15248131526636052, + "grad_norm": 0.138671875, + "learning_rate": 0.001868307687218172, + "loss": 0.166, + "step": 17566 + }, + { + "epoch": 0.1524899957465647, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018682921863391918, + "loss": 0.1162, + "step": 17567 + }, + { + "epoch": 0.15249867622676885, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018682766846200188, + "loss": 0.1553, + "step": 17568 + }, + { + "epoch": 0.15250735670697302, + "grad_norm": 0.73828125, + "learning_rate": 0.0018682611820606701, + "loss": 0.1914, + "step": 17569 + }, + { + "epoch": 0.15251603718717718, + "grad_norm": 0.35546875, + "learning_rate": 0.0018682456786611624, + "loss": 0.1299, + "step": 17570 + }, + { + "epoch": 0.15252471766738135, + "grad_norm": 0.154296875, + "learning_rate": 0.0018682301744215128, + "loss": 0.1289, + "step": 17571 + }, + { + "epoch": 0.1525333981475855, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018682146693417383, + "loss": 0.1104, + "step": 17572 + }, + { + "epoch": 0.15254207862778968, + "grad_norm": 0.36328125, + "learning_rate": 0.0018681991634218555, + "loss": 0.0962, + "step": 17573 + }, + { + "epoch": 0.15255075910799384, + "grad_norm": 0.1474609375, + "learning_rate": 0.001868183656661882, + "loss": 0.1211, + "step": 17574 + }, + { + "epoch": 0.152559439588198, + "grad_norm": 0.6640625, + "learning_rate": 0.001868168149061834, + "loss": 0.1562, + "step": 17575 + }, + { + "epoch": 0.15256812006840217, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018681526406217294, + "loss": 0.1084, + "step": 17576 + }, + { + "epoch": 0.15257680054860634, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018681371313415843, + "loss": 0.1279, + "step": 17577 + }, + { + "epoch": 0.1525854810288105, + "grad_norm": 0.12451171875, + "learning_rate": 0.001868121621221416, + "loss": 0.1318, + "step": 17578 + }, + { + "epoch": 0.15259416150901467, + "grad_norm": 0.1875, + "learning_rate": 0.0018681061102612413, + "loss": 0.123, + "step": 17579 + }, + { + "epoch": 0.15260284198921883, + "grad_norm": 0.3046875, + "learning_rate": 0.0018680905984610773, + "loss": 0.0938, + "step": 17580 + }, + { + "epoch": 0.152611522469423, + "grad_norm": 0.169921875, + "learning_rate": 0.0018680750858209411, + "loss": 0.1445, + "step": 17581 + }, + { + "epoch": 0.15262020294962717, + "grad_norm": 0.578125, + "learning_rate": 0.0018680595723408499, + "loss": 0.1079, + "step": 17582 + }, + { + "epoch": 0.15262888342983133, + "grad_norm": 0.3125, + "learning_rate": 0.0018680440580208198, + "loss": 0.1475, + "step": 17583 + }, + { + "epoch": 0.1526375639100355, + "grad_norm": 0.3671875, + "learning_rate": 0.0018680285428608683, + "loss": 0.0991, + "step": 17584 + }, + { + "epoch": 0.15264624439023966, + "grad_norm": 0.58984375, + "learning_rate": 0.0018680130268610125, + "loss": 0.0894, + "step": 17585 + }, + { + "epoch": 0.15265492487044383, + "grad_norm": 0.29296875, + "learning_rate": 0.0018679975100212689, + "loss": 0.1992, + "step": 17586 + }, + { + "epoch": 0.152663605350648, + "grad_norm": 0.1796875, + "learning_rate": 0.001867981992341655, + "loss": 0.1387, + "step": 17587 + }, + { + "epoch": 0.15267228583085216, + "grad_norm": 0.373046875, + "learning_rate": 0.0018679664738221876, + "loss": 0.1387, + "step": 17588 + }, + { + "epoch": 0.15268096631105632, + "grad_norm": 0.357421875, + "learning_rate": 0.0018679509544628835, + "loss": 0.1191, + "step": 17589 + }, + { + "epoch": 0.15268964679126049, + "grad_norm": 0.14453125, + "learning_rate": 0.0018679354342637595, + "loss": 0.0996, + "step": 17590 + }, + { + "epoch": 0.15269832727146465, + "grad_norm": 0.421875, + "learning_rate": 0.001867919913224833, + "loss": 0.1152, + "step": 17591 + }, + { + "epoch": 0.15270700775166882, + "grad_norm": 0.1015625, + "learning_rate": 0.0018679043913461212, + "loss": 0.1055, + "step": 17592 + }, + { + "epoch": 0.15271568823187298, + "grad_norm": 0.275390625, + "learning_rate": 0.0018678888686276404, + "loss": 0.0762, + "step": 17593 + }, + { + "epoch": 0.15272436871207715, + "grad_norm": 0.12060546875, + "learning_rate": 0.0018678733450694077, + "loss": 0.1064, + "step": 17594 + }, + { + "epoch": 0.1527330491922813, + "grad_norm": 0.08154296875, + "learning_rate": 0.0018678578206714404, + "loss": 0.0938, + "step": 17595 + }, + { + "epoch": 0.15274172967248548, + "grad_norm": 0.5546875, + "learning_rate": 0.0018678422954337556, + "loss": 0.127, + "step": 17596 + }, + { + "epoch": 0.15275041015268964, + "grad_norm": 0.486328125, + "learning_rate": 0.0018678267693563698, + "loss": 0.1045, + "step": 17597 + }, + { + "epoch": 0.1527590906328938, + "grad_norm": 0.54296875, + "learning_rate": 0.0018678112424392998, + "loss": 0.1436, + "step": 17598 + }, + { + "epoch": 0.15276777111309797, + "grad_norm": 0.61328125, + "learning_rate": 0.0018677957146825633, + "loss": 0.1025, + "step": 17599 + }, + { + "epoch": 0.15277645159330214, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018677801860861772, + "loss": 0.1523, + "step": 17600 + }, + { + "epoch": 0.1527851320735063, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018677646566501577, + "loss": 0.1113, + "step": 17601 + }, + { + "epoch": 0.15279381255371047, + "grad_norm": 0.55859375, + "learning_rate": 0.0018677491263745225, + "loss": 0.1074, + "step": 17602 + }, + { + "epoch": 0.15280249303391463, + "grad_norm": 1.03125, + "learning_rate": 0.0018677335952592886, + "loss": 0.1064, + "step": 17603 + }, + { + "epoch": 0.1528111735141188, + "grad_norm": 0.1171875, + "learning_rate": 0.0018677180633044728, + "loss": 0.1426, + "step": 17604 + }, + { + "epoch": 0.15281985399432296, + "grad_norm": 0.431640625, + "learning_rate": 0.0018677025305100917, + "loss": 0.1309, + "step": 17605 + }, + { + "epoch": 0.15282853447452713, + "grad_norm": 0.1123046875, + "learning_rate": 0.001867686996876163, + "loss": 0.1035, + "step": 17606 + }, + { + "epoch": 0.1528372149547313, + "grad_norm": 0.1943359375, + "learning_rate": 0.001867671462402703, + "loss": 0.0933, + "step": 17607 + }, + { + "epoch": 0.15284589543493546, + "grad_norm": 1.0, + "learning_rate": 0.0018676559270897294, + "loss": 0.1206, + "step": 17608 + }, + { + "epoch": 0.15285457591513962, + "grad_norm": 0.375, + "learning_rate": 0.0018676403909372583, + "loss": 0.1113, + "step": 17609 + }, + { + "epoch": 0.1528632563953438, + "grad_norm": 0.470703125, + "learning_rate": 0.0018676248539453077, + "loss": 0.1504, + "step": 17610 + }, + { + "epoch": 0.15287193687554795, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018676093161138939, + "loss": 0.1074, + "step": 17611 + }, + { + "epoch": 0.15288061735575212, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018675937774430343, + "loss": 0.1035, + "step": 17612 + }, + { + "epoch": 0.15288929783595628, + "grad_norm": 0.25390625, + "learning_rate": 0.0018675782379327453, + "loss": 0.1074, + "step": 17613 + }, + { + "epoch": 0.15289797831616045, + "grad_norm": 0.54296875, + "learning_rate": 0.0018675626975830444, + "loss": 0.1475, + "step": 17614 + }, + { + "epoch": 0.15290665879636461, + "grad_norm": 0.435546875, + "learning_rate": 0.0018675471563939486, + "loss": 0.1055, + "step": 17615 + }, + { + "epoch": 0.15291533927656878, + "grad_norm": 0.59375, + "learning_rate": 0.001867531614365475, + "loss": 0.1875, + "step": 17616 + }, + { + "epoch": 0.15292401975677294, + "grad_norm": 0.171875, + "learning_rate": 0.0018675160714976397, + "loss": 0.1299, + "step": 17617 + }, + { + "epoch": 0.1529327002369771, + "grad_norm": 0.27734375, + "learning_rate": 0.0018675005277904607, + "loss": 0.1084, + "step": 17618 + }, + { + "epoch": 0.15294138071718127, + "grad_norm": 0.1005859375, + "learning_rate": 0.0018674849832439546, + "loss": 0.083, + "step": 17619 + }, + { + "epoch": 0.15295006119738544, + "grad_norm": 0.138671875, + "learning_rate": 0.0018674694378581385, + "loss": 0.1289, + "step": 17620 + }, + { + "epoch": 0.1529587416775896, + "grad_norm": 0.359375, + "learning_rate": 0.0018674538916330294, + "loss": 0.085, + "step": 17621 + }, + { + "epoch": 0.15296742215779377, + "grad_norm": 0.30078125, + "learning_rate": 0.001867438344568644, + "loss": 0.1025, + "step": 17622 + }, + { + "epoch": 0.15297610263799793, + "grad_norm": 0.45703125, + "learning_rate": 0.001867422796665, + "loss": 0.1064, + "step": 17623 + }, + { + "epoch": 0.1529847831182021, + "grad_norm": 0.5234375, + "learning_rate": 0.0018674072479221138, + "loss": 0.1562, + "step": 17624 + }, + { + "epoch": 0.15299346359840627, + "grad_norm": 0.388671875, + "learning_rate": 0.0018673916983400023, + "loss": 0.1113, + "step": 17625 + }, + { + "epoch": 0.15300214407861043, + "grad_norm": 0.279296875, + "learning_rate": 0.001867376147918683, + "loss": 0.1191, + "step": 17626 + }, + { + "epoch": 0.1530108245588146, + "grad_norm": 0.224609375, + "learning_rate": 0.0018673605966581725, + "loss": 0.125, + "step": 17627 + }, + { + "epoch": 0.15301950503901876, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018673450445584883, + "loss": 0.1289, + "step": 17628 + }, + { + "epoch": 0.15302818551922293, + "grad_norm": 0.265625, + "learning_rate": 0.0018673294916196467, + "loss": 0.0928, + "step": 17629 + }, + { + "epoch": 0.1530368659994271, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018673139378416657, + "loss": 0.1045, + "step": 17630 + }, + { + "epoch": 0.15304554647963126, + "grad_norm": 0.384765625, + "learning_rate": 0.0018672983832245611, + "loss": 0.1162, + "step": 17631 + }, + { + "epoch": 0.15305422695983542, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018672828277683507, + "loss": 0.0986, + "step": 17632 + }, + { + "epoch": 0.1530629074400396, + "grad_norm": 0.61328125, + "learning_rate": 0.0018672672714730514, + "loss": 0.1177, + "step": 17633 + }, + { + "epoch": 0.15307158792024375, + "grad_norm": 0.46875, + "learning_rate": 0.0018672517143386801, + "loss": 0.1719, + "step": 17634 + }, + { + "epoch": 0.15308026840044792, + "grad_norm": 0.3203125, + "learning_rate": 0.001867236156365254, + "loss": 0.1191, + "step": 17635 + }, + { + "epoch": 0.15308894888065208, + "grad_norm": 0.25390625, + "learning_rate": 0.0018672205975527895, + "loss": 0.1128, + "step": 17636 + }, + { + "epoch": 0.15309762936085625, + "grad_norm": 0.1796875, + "learning_rate": 0.0018672050379013043, + "loss": 0.1191, + "step": 17637 + }, + { + "epoch": 0.1531063098410604, + "grad_norm": 0.263671875, + "learning_rate": 0.0018671894774108153, + "loss": 0.124, + "step": 17638 + }, + { + "epoch": 0.15311499032126458, + "grad_norm": 0.8125, + "learning_rate": 0.0018671739160813394, + "loss": 0.0815, + "step": 17639 + }, + { + "epoch": 0.15312367080146874, + "grad_norm": 0.33203125, + "learning_rate": 0.0018671583539128938, + "loss": 0.1484, + "step": 17640 + }, + { + "epoch": 0.1531323512816729, + "grad_norm": 0.380859375, + "learning_rate": 0.0018671427909054952, + "loss": 0.0942, + "step": 17641 + }, + { + "epoch": 0.15314103176187707, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018671272270591609, + "loss": 0.1006, + "step": 17642 + }, + { + "epoch": 0.15314971224208124, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018671116623739078, + "loss": 0.1543, + "step": 17643 + }, + { + "epoch": 0.1531583927222854, + "grad_norm": 0.35546875, + "learning_rate": 0.0018670960968497526, + "loss": 0.127, + "step": 17644 + }, + { + "epoch": 0.15316707320248957, + "grad_norm": 0.58203125, + "learning_rate": 0.0018670805304867126, + "loss": 0.1621, + "step": 17645 + }, + { + "epoch": 0.15317575368269373, + "grad_norm": 0.27734375, + "learning_rate": 0.001867064963284805, + "loss": 0.1416, + "step": 17646 + }, + { + "epoch": 0.1531844341628979, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018670493952440469, + "loss": 0.104, + "step": 17647 + }, + { + "epoch": 0.15319311464310206, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018670338263644549, + "loss": 0.1201, + "step": 17648 + }, + { + "epoch": 0.15320179512330623, + "grad_norm": 0.470703125, + "learning_rate": 0.001867018256646046, + "loss": 0.1211, + "step": 17649 + }, + { + "epoch": 0.1532104756035104, + "grad_norm": 0.80078125, + "learning_rate": 0.001867002686088838, + "loss": 0.0918, + "step": 17650 + }, + { + "epoch": 0.15321915608371456, + "grad_norm": 0.78125, + "learning_rate": 0.0018669871146928473, + "loss": 0.0947, + "step": 17651 + }, + { + "epoch": 0.15322783656391872, + "grad_norm": 0.1123046875, + "learning_rate": 0.0018669715424580906, + "loss": 0.0967, + "step": 17652 + }, + { + "epoch": 0.1532365170441229, + "grad_norm": 0.3515625, + "learning_rate": 0.0018669559693845854, + "loss": 0.0762, + "step": 17653 + }, + { + "epoch": 0.15324519752432705, + "grad_norm": 0.58203125, + "learning_rate": 0.001866940395472349, + "loss": 0.1523, + "step": 17654 + }, + { + "epoch": 0.15325387800453122, + "grad_norm": 0.150390625, + "learning_rate": 0.001866924820721398, + "loss": 0.0991, + "step": 17655 + }, + { + "epoch": 0.15326255848473538, + "grad_norm": 0.14453125, + "learning_rate": 0.0018669092451317495, + "loss": 0.1074, + "step": 17656 + }, + { + "epoch": 0.15327123896493955, + "grad_norm": 0.349609375, + "learning_rate": 0.0018668936687034207, + "loss": 0.0811, + "step": 17657 + }, + { + "epoch": 0.15327991944514371, + "grad_norm": 0.4453125, + "learning_rate": 0.0018668780914364283, + "loss": 0.1104, + "step": 17658 + }, + { + "epoch": 0.15328859992534788, + "grad_norm": 0.5859375, + "learning_rate": 0.0018668625133307895, + "loss": 0.0928, + "step": 17659 + }, + { + "epoch": 0.15329728040555204, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018668469343865216, + "loss": 0.165, + "step": 17660 + }, + { + "epoch": 0.1533059608857562, + "grad_norm": 0.208984375, + "learning_rate": 0.0018668313546036414, + "loss": 0.0933, + "step": 17661 + }, + { + "epoch": 0.15331464136596037, + "grad_norm": 1.328125, + "learning_rate": 0.0018668157739821659, + "loss": 0.1514, + "step": 17662 + }, + { + "epoch": 0.15332332184616454, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018668001925221118, + "loss": 0.104, + "step": 17663 + }, + { + "epoch": 0.1533320023263687, + "grad_norm": 0.72265625, + "learning_rate": 0.0018667846102234972, + "loss": 0.123, + "step": 17664 + }, + { + "epoch": 0.15334068280657287, + "grad_norm": 0.68359375, + "learning_rate": 0.001866769027086338, + "loss": 0.1689, + "step": 17665 + }, + { + "epoch": 0.15334936328677704, + "grad_norm": 1.1484375, + "learning_rate": 0.001866753443110652, + "loss": 0.1436, + "step": 17666 + }, + { + "epoch": 0.1533580437669812, + "grad_norm": 0.10546875, + "learning_rate": 0.0018667378582964559, + "loss": 0.1279, + "step": 17667 + }, + { + "epoch": 0.15336672424718537, + "grad_norm": 0.1015625, + "learning_rate": 0.0018667222726437664, + "loss": 0.1045, + "step": 17668 + }, + { + "epoch": 0.15337540472738953, + "grad_norm": 0.65234375, + "learning_rate": 0.0018667066861526012, + "loss": 0.1748, + "step": 17669 + }, + { + "epoch": 0.1533840852075937, + "grad_norm": 0.08984375, + "learning_rate": 0.0018666910988229775, + "loss": 0.1016, + "step": 17670 + }, + { + "epoch": 0.15339276568779786, + "grad_norm": 0.341796875, + "learning_rate": 0.0018666755106549113, + "loss": 0.1318, + "step": 17671 + }, + { + "epoch": 0.15340144616800203, + "grad_norm": 0.384765625, + "learning_rate": 0.0018666599216484207, + "loss": 0.1035, + "step": 17672 + }, + { + "epoch": 0.1534101266482062, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018666443318035221, + "loss": 0.0894, + "step": 17673 + }, + { + "epoch": 0.15341880712841036, + "grad_norm": 0.447265625, + "learning_rate": 0.0018666287411202326, + "loss": 0.168, + "step": 17674 + }, + { + "epoch": 0.15342748760861452, + "grad_norm": 0.99609375, + "learning_rate": 0.0018666131495985701, + "loss": 0.1357, + "step": 17675 + }, + { + "epoch": 0.1534361680888187, + "grad_norm": 0.7578125, + "learning_rate": 0.0018665975572385504, + "loss": 0.1367, + "step": 17676 + }, + { + "epoch": 0.15344484856902285, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018665819640401914, + "loss": 0.1328, + "step": 17677 + }, + { + "epoch": 0.15345352904922702, + "grad_norm": 0.66015625, + "learning_rate": 0.0018665663700035099, + "loss": 0.1021, + "step": 17678 + }, + { + "epoch": 0.15346220952943118, + "grad_norm": 0.34375, + "learning_rate": 0.0018665507751285228, + "loss": 0.1357, + "step": 17679 + }, + { + "epoch": 0.15347089000963535, + "grad_norm": 0.95703125, + "learning_rate": 0.001866535179415247, + "loss": 0.1289, + "step": 17680 + }, + { + "epoch": 0.15347957048983948, + "grad_norm": 0.625, + "learning_rate": 0.0018665195828637003, + "loss": 0.1074, + "step": 17681 + }, + { + "epoch": 0.15348825097004365, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018665039854738991, + "loss": 0.0835, + "step": 17682 + }, + { + "epoch": 0.15349693145024781, + "grad_norm": 0.28125, + "learning_rate": 0.001866488387245861, + "loss": 0.1621, + "step": 17683 + }, + { + "epoch": 0.15350561193045198, + "grad_norm": 0.49609375, + "learning_rate": 0.0018664727881796022, + "loss": 0.0967, + "step": 17684 + }, + { + "epoch": 0.15351429241065614, + "grad_norm": 0.70703125, + "learning_rate": 0.0018664571882751405, + "loss": 0.104, + "step": 17685 + }, + { + "epoch": 0.1535229728908603, + "grad_norm": 0.111328125, + "learning_rate": 0.0018664415875324926, + "loss": 0.1094, + "step": 17686 + }, + { + "epoch": 0.15353165337106447, + "grad_norm": 0.248046875, + "learning_rate": 0.001866425985951676, + "loss": 0.1299, + "step": 17687 + }, + { + "epoch": 0.15354033385126864, + "grad_norm": 0.0703125, + "learning_rate": 0.0018664103835327073, + "loss": 0.0908, + "step": 17688 + }, + { + "epoch": 0.1535490143314728, + "grad_norm": 0.142578125, + "learning_rate": 0.0018663947802756038, + "loss": 0.125, + "step": 17689 + }, + { + "epoch": 0.15355769481167697, + "grad_norm": 0.14453125, + "learning_rate": 0.0018663791761803824, + "loss": 0.0991, + "step": 17690 + }, + { + "epoch": 0.15356637529188114, + "grad_norm": 0.64453125, + "learning_rate": 0.0018663635712470603, + "loss": 0.1445, + "step": 17691 + }, + { + "epoch": 0.1535750557720853, + "grad_norm": 0.3359375, + "learning_rate": 0.0018663479654756543, + "loss": 0.1338, + "step": 17692 + }, + { + "epoch": 0.15358373625228947, + "grad_norm": 0.203125, + "learning_rate": 0.001866332358866182, + "loss": 0.1104, + "step": 17693 + }, + { + "epoch": 0.15359241673249363, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018663167514186602, + "loss": 0.1162, + "step": 17694 + }, + { + "epoch": 0.1536010972126978, + "grad_norm": 0.72265625, + "learning_rate": 0.0018663011431331053, + "loss": 0.1318, + "step": 17695 + }, + { + "epoch": 0.15360977769290196, + "grad_norm": 0.1240234375, + "learning_rate": 0.0018662855340095355, + "loss": 0.124, + "step": 17696 + }, + { + "epoch": 0.15361845817310613, + "grad_norm": 0.345703125, + "learning_rate": 0.0018662699240479675, + "loss": 0.0879, + "step": 17697 + }, + { + "epoch": 0.1536271386533103, + "grad_norm": 0.3515625, + "learning_rate": 0.0018662543132484176, + "loss": 0.1001, + "step": 17698 + }, + { + "epoch": 0.15363581913351446, + "grad_norm": 0.2138671875, + "learning_rate": 0.001866238701610904, + "loss": 0.1006, + "step": 17699 + }, + { + "epoch": 0.15364449961371862, + "grad_norm": 0.248046875, + "learning_rate": 0.0018662230891354434, + "loss": 0.0854, + "step": 17700 + }, + { + "epoch": 0.1536531800939228, + "grad_norm": 0.20703125, + "learning_rate": 0.0018662074758220523, + "loss": 0.1069, + "step": 17701 + }, + { + "epoch": 0.15366186057412695, + "grad_norm": 0.1689453125, + "learning_rate": 0.0018661918616707483, + "loss": 0.0918, + "step": 17702 + }, + { + "epoch": 0.15367054105433112, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018661762466815484, + "loss": 0.1484, + "step": 17703 + }, + { + "epoch": 0.15367922153453528, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018661606308544695, + "loss": 0.1309, + "step": 17704 + }, + { + "epoch": 0.15368790201473945, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018661450141895292, + "loss": 0.0952, + "step": 17705 + }, + { + "epoch": 0.1536965824949436, + "grad_norm": 0.095703125, + "learning_rate": 0.001866129396686744, + "loss": 0.0962, + "step": 17706 + }, + { + "epoch": 0.15370526297514778, + "grad_norm": 0.298828125, + "learning_rate": 0.0018661137783461314, + "loss": 0.1182, + "step": 17707 + }, + { + "epoch": 0.15371394345535194, + "grad_norm": 0.484375, + "learning_rate": 0.0018660981591677084, + "loss": 0.1006, + "step": 17708 + }, + { + "epoch": 0.1537226239355561, + "grad_norm": 0.146484375, + "learning_rate": 0.0018660825391514917, + "loss": 0.0771, + "step": 17709 + }, + { + "epoch": 0.15373130441576027, + "grad_norm": 0.65234375, + "learning_rate": 0.0018660669182974988, + "loss": 0.1299, + "step": 17710 + }, + { + "epoch": 0.15373998489596444, + "grad_norm": 0.41015625, + "learning_rate": 0.0018660512966057466, + "loss": 0.1143, + "step": 17711 + }, + { + "epoch": 0.1537486653761686, + "grad_norm": 0.212890625, + "learning_rate": 0.0018660356740762521, + "loss": 0.1338, + "step": 17712 + }, + { + "epoch": 0.15375734585637277, + "grad_norm": 0.37890625, + "learning_rate": 0.0018660200507090324, + "loss": 0.1006, + "step": 17713 + }, + { + "epoch": 0.15376602633657693, + "grad_norm": 0.455078125, + "learning_rate": 0.0018660044265041047, + "loss": 0.1006, + "step": 17714 + }, + { + "epoch": 0.1537747068167811, + "grad_norm": 0.2080078125, + "learning_rate": 0.0018659888014614862, + "loss": 0.1104, + "step": 17715 + }, + { + "epoch": 0.15378338729698526, + "grad_norm": 0.36328125, + "learning_rate": 0.0018659731755811943, + "loss": 0.1328, + "step": 17716 + }, + { + "epoch": 0.15379206777718943, + "grad_norm": 0.224609375, + "learning_rate": 0.0018659575488632447, + "loss": 0.1133, + "step": 17717 + }, + { + "epoch": 0.1538007482573936, + "grad_norm": 0.109375, + "learning_rate": 0.0018659419213076561, + "loss": 0.0664, + "step": 17718 + }, + { + "epoch": 0.15380942873759776, + "grad_norm": 0.69140625, + "learning_rate": 0.0018659262929144448, + "loss": 0.0908, + "step": 17719 + }, + { + "epoch": 0.15381810921780192, + "grad_norm": 0.61328125, + "learning_rate": 0.0018659106636836277, + "loss": 0.0806, + "step": 17720 + }, + { + "epoch": 0.1538267896980061, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018658950336152228, + "loss": 0.1206, + "step": 17721 + }, + { + "epoch": 0.15383547017821025, + "grad_norm": 0.232421875, + "learning_rate": 0.0018658794027092464, + "loss": 0.082, + "step": 17722 + }, + { + "epoch": 0.15384415065841442, + "grad_norm": 0.31640625, + "learning_rate": 0.0018658637709657153, + "loss": 0.1172, + "step": 17723 + }, + { + "epoch": 0.15385283113861858, + "grad_norm": 0.42578125, + "learning_rate": 0.0018658481383846475, + "loss": 0.1011, + "step": 17724 + }, + { + "epoch": 0.15386151161882275, + "grad_norm": 1.21875, + "learning_rate": 0.0018658325049660598, + "loss": 0.1328, + "step": 17725 + }, + { + "epoch": 0.15387019209902691, + "grad_norm": 0.45703125, + "learning_rate": 0.0018658168707099687, + "loss": 0.1133, + "step": 17726 + }, + { + "epoch": 0.15387887257923108, + "grad_norm": 1.21875, + "learning_rate": 0.0018658012356163923, + "loss": 0.1387, + "step": 17727 + }, + { + "epoch": 0.15388755305943524, + "grad_norm": 0.10205078125, + "learning_rate": 0.001865785599685347, + "loss": 0.0918, + "step": 17728 + }, + { + "epoch": 0.1538962335396394, + "grad_norm": 0.18359375, + "learning_rate": 0.0018657699629168502, + "loss": 0.1099, + "step": 17729 + }, + { + "epoch": 0.15390491401984358, + "grad_norm": 0.07177734375, + "learning_rate": 0.0018657543253109185, + "loss": 0.0811, + "step": 17730 + }, + { + "epoch": 0.15391359450004774, + "grad_norm": 0.35546875, + "learning_rate": 0.0018657386868675698, + "loss": 0.1377, + "step": 17731 + }, + { + "epoch": 0.1539222749802519, + "grad_norm": 0.2109375, + "learning_rate": 0.0018657230475868202, + "loss": 0.1191, + "step": 17732 + }, + { + "epoch": 0.15393095546045607, + "grad_norm": 0.244140625, + "learning_rate": 0.001865707407468688, + "loss": 0.1123, + "step": 17733 + }, + { + "epoch": 0.15393963594066024, + "grad_norm": 0.130859375, + "learning_rate": 0.0018656917665131893, + "loss": 0.126, + "step": 17734 + }, + { + "epoch": 0.1539483164208644, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018656761247203418, + "loss": 0.0972, + "step": 17735 + }, + { + "epoch": 0.15395699690106857, + "grad_norm": 0.296875, + "learning_rate": 0.0018656604820901622, + "loss": 0.0977, + "step": 17736 + }, + { + "epoch": 0.15396567738127273, + "grad_norm": 0.51953125, + "learning_rate": 0.0018656448386226681, + "loss": 0.1934, + "step": 17737 + }, + { + "epoch": 0.1539743578614769, + "grad_norm": 0.875, + "learning_rate": 0.001865629194317876, + "loss": 0.1216, + "step": 17738 + }, + { + "epoch": 0.15398303834168106, + "grad_norm": 0.130859375, + "learning_rate": 0.0018656135491758033, + "loss": 0.1016, + "step": 17739 + }, + { + "epoch": 0.15399171882188523, + "grad_norm": 0.35546875, + "learning_rate": 0.0018655979031964674, + "loss": 0.1123, + "step": 17740 + }, + { + "epoch": 0.1540003993020894, + "grad_norm": 0.6171875, + "learning_rate": 0.0018655822563798847, + "loss": 0.1025, + "step": 17741 + }, + { + "epoch": 0.15400907978229356, + "grad_norm": 1.1640625, + "learning_rate": 0.0018655666087260734, + "loss": 0.1162, + "step": 17742 + }, + { + "epoch": 0.15401776026249772, + "grad_norm": 0.361328125, + "learning_rate": 0.0018655509602350496, + "loss": 0.1338, + "step": 17743 + }, + { + "epoch": 0.1540264407427019, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018655353109068306, + "loss": 0.125, + "step": 17744 + }, + { + "epoch": 0.15403512122290605, + "grad_norm": 0.359375, + "learning_rate": 0.001865519660741434, + "loss": 0.1348, + "step": 17745 + }, + { + "epoch": 0.15404380170311022, + "grad_norm": 0.125, + "learning_rate": 0.0018655040097388762, + "loss": 0.1523, + "step": 17746 + }, + { + "epoch": 0.15405248218331438, + "grad_norm": 0.27734375, + "learning_rate": 0.0018654883578991754, + "loss": 0.1045, + "step": 17747 + }, + { + "epoch": 0.15406116266351855, + "grad_norm": 0.326171875, + "learning_rate": 0.0018654727052223473, + "loss": 0.1143, + "step": 17748 + }, + { + "epoch": 0.1540698431437227, + "grad_norm": 0.2080078125, + "learning_rate": 0.0018654570517084102, + "loss": 0.1162, + "step": 17749 + }, + { + "epoch": 0.15407852362392688, + "grad_norm": 0.15234375, + "learning_rate": 0.0018654413973573805, + "loss": 0.1143, + "step": 17750 + }, + { + "epoch": 0.15408720410413104, + "grad_norm": 0.70703125, + "learning_rate": 0.0018654257421692757, + "loss": 0.1074, + "step": 17751 + }, + { + "epoch": 0.1540958845843352, + "grad_norm": 0.306640625, + "learning_rate": 0.001865410086144113, + "loss": 0.1377, + "step": 17752 + }, + { + "epoch": 0.15410456506453937, + "grad_norm": 0.1904296875, + "learning_rate": 0.001865394429281909, + "loss": 0.1211, + "step": 17753 + }, + { + "epoch": 0.15411324554474354, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018653787715826816, + "loss": 0.0957, + "step": 17754 + }, + { + "epoch": 0.1541219260249477, + "grad_norm": 0.369140625, + "learning_rate": 0.001865363113046447, + "loss": 0.1001, + "step": 17755 + }, + { + "epoch": 0.15413060650515187, + "grad_norm": 0.6484375, + "learning_rate": 0.001865347453673223, + "loss": 0.1465, + "step": 17756 + }, + { + "epoch": 0.15413928698535603, + "grad_norm": 0.380859375, + "learning_rate": 0.0018653317934630262, + "loss": 0.1094, + "step": 17757 + }, + { + "epoch": 0.1541479674655602, + "grad_norm": 0.353515625, + "learning_rate": 0.0018653161324158748, + "loss": 0.1118, + "step": 17758 + }, + { + "epoch": 0.15415664794576436, + "grad_norm": 0.203125, + "learning_rate": 0.0018653004705317846, + "loss": 0.1514, + "step": 17759 + }, + { + "epoch": 0.15416532842596853, + "grad_norm": 0.138671875, + "learning_rate": 0.0018652848078107734, + "loss": 0.1011, + "step": 17760 + }, + { + "epoch": 0.1541740089061727, + "grad_norm": 0.12109375, + "learning_rate": 0.0018652691442528585, + "loss": 0.1147, + "step": 17761 + }, + { + "epoch": 0.15418268938637686, + "grad_norm": 0.177734375, + "learning_rate": 0.0018652534798580562, + "loss": 0.085, + "step": 17762 + }, + { + "epoch": 0.15419136986658102, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018652378146263849, + "loss": 0.1172, + "step": 17763 + }, + { + "epoch": 0.1542000503467852, + "grad_norm": 0.22265625, + "learning_rate": 0.0018652221485578604, + "loss": 0.1118, + "step": 17764 + }, + { + "epoch": 0.15420873082698935, + "grad_norm": 0.08251953125, + "learning_rate": 0.0018652064816525005, + "loss": 0.1045, + "step": 17765 + }, + { + "epoch": 0.15421741130719352, + "grad_norm": 0.74609375, + "learning_rate": 0.0018651908139103227, + "loss": 0.1123, + "step": 17766 + }, + { + "epoch": 0.15422609178739768, + "grad_norm": 0.169921875, + "learning_rate": 0.0018651751453313435, + "loss": 0.103, + "step": 17767 + }, + { + "epoch": 0.15423477226760185, + "grad_norm": 0.63671875, + "learning_rate": 0.0018651594759155803, + "loss": 0.1025, + "step": 17768 + }, + { + "epoch": 0.15424345274780601, + "grad_norm": 0.11669921875, + "learning_rate": 0.00186514380566305, + "loss": 0.1133, + "step": 17769 + }, + { + "epoch": 0.15425213322801018, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018651281345737703, + "loss": 0.1099, + "step": 17770 + }, + { + "epoch": 0.15426081370821434, + "grad_norm": 0.18359375, + "learning_rate": 0.0018651124626477577, + "loss": 0.0972, + "step": 17771 + }, + { + "epoch": 0.1542694941884185, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018650967898850297, + "loss": 0.1094, + "step": 17772 + }, + { + "epoch": 0.15427817466862268, + "grad_norm": 0.30078125, + "learning_rate": 0.0018650811162856033, + "loss": 0.1309, + "step": 17773 + }, + { + "epoch": 0.15428685514882684, + "grad_norm": 0.60546875, + "learning_rate": 0.0018650654418494957, + "loss": 0.209, + "step": 17774 + }, + { + "epoch": 0.154295535629031, + "grad_norm": 0.19921875, + "learning_rate": 0.001865049766576724, + "loss": 0.1699, + "step": 17775 + }, + { + "epoch": 0.15430421610923517, + "grad_norm": 0.8671875, + "learning_rate": 0.0018650340904673055, + "loss": 0.1011, + "step": 17776 + }, + { + "epoch": 0.15431289658943934, + "grad_norm": 0.06640625, + "learning_rate": 0.0018650184135212571, + "loss": 0.0737, + "step": 17777 + }, + { + "epoch": 0.1543215770696435, + "grad_norm": 0.6796875, + "learning_rate": 0.001865002735738596, + "loss": 0.103, + "step": 17778 + }, + { + "epoch": 0.15433025754984767, + "grad_norm": 0.25390625, + "learning_rate": 0.0018649870571193394, + "loss": 0.1201, + "step": 17779 + }, + { + "epoch": 0.15433893803005183, + "grad_norm": 0.06005859375, + "learning_rate": 0.0018649713776635046, + "loss": 0.0977, + "step": 17780 + }, + { + "epoch": 0.154347618510256, + "grad_norm": 0.3203125, + "learning_rate": 0.0018649556973711085, + "loss": 0.0874, + "step": 17781 + }, + { + "epoch": 0.15435629899046016, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018649400162421687, + "loss": 0.1562, + "step": 17782 + }, + { + "epoch": 0.15436497947066433, + "grad_norm": 0.44921875, + "learning_rate": 0.0018649243342767016, + "loss": 0.1338, + "step": 17783 + }, + { + "epoch": 0.1543736599508685, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018649086514747244, + "loss": 0.1064, + "step": 17784 + }, + { + "epoch": 0.15438234043107266, + "grad_norm": 0.0986328125, + "learning_rate": 0.001864892967836255, + "loss": 0.1035, + "step": 17785 + }, + { + "epoch": 0.15439102091127682, + "grad_norm": 0.0859375, + "learning_rate": 0.0018648772833613104, + "loss": 0.1318, + "step": 17786 + }, + { + "epoch": 0.154399701391481, + "grad_norm": 0.51953125, + "learning_rate": 0.0018648615980499073, + "loss": 0.1387, + "step": 17787 + }, + { + "epoch": 0.15440838187168515, + "grad_norm": 0.55859375, + "learning_rate": 0.001864845911902063, + "loss": 0.0972, + "step": 17788 + }, + { + "epoch": 0.15441706235188932, + "grad_norm": 0.6328125, + "learning_rate": 0.0018648302249177947, + "loss": 0.1514, + "step": 17789 + }, + { + "epoch": 0.15442574283209348, + "grad_norm": 0.57421875, + "learning_rate": 0.0018648145370971192, + "loss": 0.1182, + "step": 17790 + }, + { + "epoch": 0.15443442331229765, + "grad_norm": 0.9140625, + "learning_rate": 0.0018647988484400547, + "loss": 0.1494, + "step": 17791 + }, + { + "epoch": 0.1544431037925018, + "grad_norm": 0.2490234375, + "learning_rate": 0.001864783158946617, + "loss": 0.1445, + "step": 17792 + }, + { + "epoch": 0.15445178427270598, + "grad_norm": 0.31640625, + "learning_rate": 0.0018647674686168244, + "loss": 0.1211, + "step": 17793 + }, + { + "epoch": 0.15446046475291014, + "grad_norm": 0.67578125, + "learning_rate": 0.0018647517774506932, + "loss": 0.1064, + "step": 17794 + }, + { + "epoch": 0.1544691452331143, + "grad_norm": 0.484375, + "learning_rate": 0.0018647360854482414, + "loss": 0.1123, + "step": 17795 + }, + { + "epoch": 0.15447782571331847, + "grad_norm": 0.201171875, + "learning_rate": 0.0018647203926094853, + "loss": 0.126, + "step": 17796 + }, + { + "epoch": 0.15448650619352264, + "grad_norm": 0.42578125, + "learning_rate": 0.0018647046989344429, + "loss": 0.0957, + "step": 17797 + }, + { + "epoch": 0.1544951866737268, + "grad_norm": 1.140625, + "learning_rate": 0.0018646890044231308, + "loss": 0.0996, + "step": 17798 + }, + { + "epoch": 0.15450386715393097, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018646733090755658, + "loss": 0.0942, + "step": 17799 + }, + { + "epoch": 0.15451254763413513, + "grad_norm": 0.12060546875, + "learning_rate": 0.001864657612891766, + "loss": 0.1104, + "step": 17800 + }, + { + "epoch": 0.1545212281143393, + "grad_norm": 0.671875, + "learning_rate": 0.0018646419158717482, + "loss": 0.1426, + "step": 17801 + }, + { + "epoch": 0.15452990859454346, + "grad_norm": 0.072265625, + "learning_rate": 0.0018646262180155293, + "loss": 0.0977, + "step": 17802 + }, + { + "epoch": 0.15453858907474763, + "grad_norm": 0.0791015625, + "learning_rate": 0.0018646105193231267, + "loss": 0.1055, + "step": 17803 + }, + { + "epoch": 0.15454726955495177, + "grad_norm": 0.2158203125, + "learning_rate": 0.0018645948197945576, + "loss": 0.1245, + "step": 17804 + }, + { + "epoch": 0.15455595003515593, + "grad_norm": 0.302734375, + "learning_rate": 0.001864579119429839, + "loss": 0.0869, + "step": 17805 + }, + { + "epoch": 0.1545646305153601, + "grad_norm": 0.15625, + "learning_rate": 0.0018645634182289882, + "loss": 0.0981, + "step": 17806 + }, + { + "epoch": 0.15457331099556426, + "grad_norm": 0.337890625, + "learning_rate": 0.0018645477161920224, + "loss": 0.0791, + "step": 17807 + }, + { + "epoch": 0.15458199147576843, + "grad_norm": 0.25, + "learning_rate": 0.0018645320133189585, + "loss": 0.0996, + "step": 17808 + }, + { + "epoch": 0.1545906719559726, + "grad_norm": 0.77734375, + "learning_rate": 0.0018645163096098138, + "loss": 0.1396, + "step": 17809 + }, + { + "epoch": 0.15459935243617676, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018645006050646061, + "loss": 0.1309, + "step": 17810 + }, + { + "epoch": 0.15460803291638092, + "grad_norm": 1.109375, + "learning_rate": 0.0018644848996833515, + "loss": 0.1055, + "step": 17811 + }, + { + "epoch": 0.1546167133965851, + "grad_norm": 0.19140625, + "learning_rate": 0.0018644691934660677, + "loss": 0.1147, + "step": 17812 + }, + { + "epoch": 0.15462539387678925, + "grad_norm": 0.259765625, + "learning_rate": 0.0018644534864127723, + "loss": 0.1562, + "step": 17813 + }, + { + "epoch": 0.15463407435699342, + "grad_norm": 0.52734375, + "learning_rate": 0.0018644377785234816, + "loss": 0.1118, + "step": 17814 + }, + { + "epoch": 0.15464275483719758, + "grad_norm": 0.326171875, + "learning_rate": 0.0018644220697982136, + "loss": 0.1182, + "step": 17815 + }, + { + "epoch": 0.15465143531740175, + "grad_norm": 0.53125, + "learning_rate": 0.0018644063602369849, + "loss": 0.1001, + "step": 17816 + }, + { + "epoch": 0.1546601157976059, + "grad_norm": 0.244140625, + "learning_rate": 0.0018643906498398129, + "loss": 0.0972, + "step": 17817 + }, + { + "epoch": 0.15466879627781008, + "grad_norm": 1.53125, + "learning_rate": 0.0018643749386067147, + "loss": 0.2148, + "step": 17818 + }, + { + "epoch": 0.15467747675801424, + "grad_norm": 0.193359375, + "learning_rate": 0.0018643592265377075, + "loss": 0.1006, + "step": 17819 + }, + { + "epoch": 0.1546861572382184, + "grad_norm": 0.51953125, + "learning_rate": 0.001864343513632809, + "loss": 0.1182, + "step": 17820 + }, + { + "epoch": 0.15469483771842257, + "grad_norm": 0.115234375, + "learning_rate": 0.0018643277998920357, + "loss": 0.1001, + "step": 17821 + }, + { + "epoch": 0.15470351819862674, + "grad_norm": 0.087890625, + "learning_rate": 0.0018643120853154048, + "loss": 0.127, + "step": 17822 + }, + { + "epoch": 0.1547121986788309, + "grad_norm": 0.6328125, + "learning_rate": 0.001864296369902934, + "loss": 0.1719, + "step": 17823 + }, + { + "epoch": 0.15472087915903507, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018642806536546398, + "loss": 0.0908, + "step": 17824 + }, + { + "epoch": 0.15472955963923923, + "grad_norm": 0.185546875, + "learning_rate": 0.00186426493657054, + "loss": 0.1367, + "step": 17825 + }, + { + "epoch": 0.1547382401194434, + "grad_norm": 0.130859375, + "learning_rate": 0.0018642492186506517, + "loss": 0.1357, + "step": 17826 + }, + { + "epoch": 0.15474692059964756, + "grad_norm": 0.431640625, + "learning_rate": 0.0018642334998949914, + "loss": 0.0615, + "step": 17827 + }, + { + "epoch": 0.15475560107985173, + "grad_norm": 0.51953125, + "learning_rate": 0.0018642177803035772, + "loss": 0.0928, + "step": 17828 + }, + { + "epoch": 0.1547642815600559, + "grad_norm": 0.7421875, + "learning_rate": 0.001864202059876426, + "loss": 0.1719, + "step": 17829 + }, + { + "epoch": 0.15477296204026006, + "grad_norm": 0.5546875, + "learning_rate": 0.0018641863386135546, + "loss": 0.1001, + "step": 17830 + }, + { + "epoch": 0.15478164252046422, + "grad_norm": 0.251953125, + "learning_rate": 0.0018641706165149808, + "loss": 0.1211, + "step": 17831 + }, + { + "epoch": 0.1547903230006684, + "grad_norm": 0.3515625, + "learning_rate": 0.0018641548935807212, + "loss": 0.1006, + "step": 17832 + }, + { + "epoch": 0.15479900348087255, + "grad_norm": 0.2265625, + "learning_rate": 0.0018641391698107934, + "loss": 0.0977, + "step": 17833 + }, + { + "epoch": 0.15480768396107672, + "grad_norm": 0.95703125, + "learning_rate": 0.0018641234452052148, + "loss": 0.1162, + "step": 17834 + }, + { + "epoch": 0.15481636444128088, + "grad_norm": 0.8828125, + "learning_rate": 0.001864107719764002, + "loss": 0.1445, + "step": 17835 + }, + { + "epoch": 0.15482504492148505, + "grad_norm": 0.84375, + "learning_rate": 0.0018640919934871725, + "loss": 0.1025, + "step": 17836 + }, + { + "epoch": 0.15483372540168922, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018640762663747435, + "loss": 0.0835, + "step": 17837 + }, + { + "epoch": 0.15484240588189338, + "grad_norm": 0.490234375, + "learning_rate": 0.0018640605384267319, + "loss": 0.166, + "step": 17838 + }, + { + "epoch": 0.15485108636209755, + "grad_norm": 0.4140625, + "learning_rate": 0.0018640448096431555, + "loss": 0.0874, + "step": 17839 + }, + { + "epoch": 0.1548597668423017, + "grad_norm": 0.134765625, + "learning_rate": 0.0018640290800240311, + "loss": 0.1387, + "step": 17840 + }, + { + "epoch": 0.15486844732250588, + "grad_norm": 0.09521484375, + "learning_rate": 0.0018640133495693761, + "loss": 0.125, + "step": 17841 + }, + { + "epoch": 0.15487712780271004, + "grad_norm": 0.279296875, + "learning_rate": 0.0018639976182792074, + "loss": 0.106, + "step": 17842 + }, + { + "epoch": 0.1548858082829142, + "grad_norm": 0.32421875, + "learning_rate": 0.0018639818861535425, + "loss": 0.0913, + "step": 17843 + }, + { + "epoch": 0.15489448876311837, + "grad_norm": 0.4296875, + "learning_rate": 0.0018639661531923985, + "loss": 0.1475, + "step": 17844 + }, + { + "epoch": 0.15490316924332254, + "grad_norm": 0.23828125, + "learning_rate": 0.0018639504193957925, + "loss": 0.1162, + "step": 17845 + }, + { + "epoch": 0.1549118497235267, + "grad_norm": 0.84375, + "learning_rate": 0.0018639346847637416, + "loss": 0.1069, + "step": 17846 + }, + { + "epoch": 0.15492053020373087, + "grad_norm": 0.2080078125, + "learning_rate": 0.0018639189492962636, + "loss": 0.1309, + "step": 17847 + }, + { + "epoch": 0.15492921068393503, + "grad_norm": 0.94921875, + "learning_rate": 0.001863903212993375, + "loss": 0.1289, + "step": 17848 + }, + { + "epoch": 0.1549378911641392, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018638874758550937, + "loss": 0.0996, + "step": 17849 + }, + { + "epoch": 0.15494657164434336, + "grad_norm": 0.5390625, + "learning_rate": 0.0018638717378814363, + "loss": 0.0869, + "step": 17850 + }, + { + "epoch": 0.15495525212454753, + "grad_norm": 0.33203125, + "learning_rate": 0.0018638559990724203, + "loss": 0.1084, + "step": 17851 + }, + { + "epoch": 0.1549639326047517, + "grad_norm": 0.51953125, + "learning_rate": 0.0018638402594280629, + "loss": 0.082, + "step": 17852 + }, + { + "epoch": 0.15497261308495586, + "grad_norm": 0.5234375, + "learning_rate": 0.001863824518948381, + "loss": 0.1182, + "step": 17853 + }, + { + "epoch": 0.15498129356516002, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018638087776333924, + "loss": 0.1523, + "step": 17854 + }, + { + "epoch": 0.1549899740453642, + "grad_norm": 0.478515625, + "learning_rate": 0.001863793035483114, + "loss": 0.0933, + "step": 17855 + }, + { + "epoch": 0.15499865452556835, + "grad_norm": 1.484375, + "learning_rate": 0.0018637772924975627, + "loss": 0.1289, + "step": 17856 + }, + { + "epoch": 0.15500733500577252, + "grad_norm": 0.494140625, + "learning_rate": 0.0018637615486767562, + "loss": 0.123, + "step": 17857 + }, + { + "epoch": 0.15501601548597668, + "grad_norm": 1.25, + "learning_rate": 0.0018637458040207115, + "loss": 0.1396, + "step": 17858 + }, + { + "epoch": 0.15502469596618085, + "grad_norm": 0.1328125, + "learning_rate": 0.001863730058529446, + "loss": 0.1133, + "step": 17859 + }, + { + "epoch": 0.155033376446385, + "grad_norm": 0.486328125, + "learning_rate": 0.0018637143122029764, + "loss": 0.0947, + "step": 17860 + }, + { + "epoch": 0.15504205692658918, + "grad_norm": 0.287109375, + "learning_rate": 0.0018636985650413208, + "loss": 0.1021, + "step": 17861 + }, + { + "epoch": 0.15505073740679334, + "grad_norm": 0.26171875, + "learning_rate": 0.0018636828170444954, + "loss": 0.1221, + "step": 17862 + }, + { + "epoch": 0.1550594178869975, + "grad_norm": 0.251953125, + "learning_rate": 0.0018636670682125186, + "loss": 0.1201, + "step": 17863 + }, + { + "epoch": 0.15506809836720167, + "grad_norm": 0.263671875, + "learning_rate": 0.0018636513185454064, + "loss": 0.0938, + "step": 17864 + }, + { + "epoch": 0.15507677884740584, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018636355680431772, + "loss": 0.1211, + "step": 17865 + }, + { + "epoch": 0.15508545932761, + "grad_norm": 0.1279296875, + "learning_rate": 0.001863619816705847, + "loss": 0.123, + "step": 17866 + }, + { + "epoch": 0.15509413980781417, + "grad_norm": 0.2021484375, + "learning_rate": 0.001863604064533434, + "loss": 0.0991, + "step": 17867 + }, + { + "epoch": 0.15510282028801833, + "grad_norm": 0.0634765625, + "learning_rate": 0.0018635883115259548, + "loss": 0.0991, + "step": 17868 + }, + { + "epoch": 0.1551115007682225, + "grad_norm": 0.458984375, + "learning_rate": 0.001863572557683427, + "loss": 0.0938, + "step": 17869 + }, + { + "epoch": 0.15512018124842666, + "grad_norm": 0.4140625, + "learning_rate": 0.001863556803005868, + "loss": 0.123, + "step": 17870 + }, + { + "epoch": 0.15512886172863083, + "grad_norm": 0.13671875, + "learning_rate": 0.0018635410474932943, + "loss": 0.1289, + "step": 17871 + }, + { + "epoch": 0.155137542208835, + "grad_norm": 0.482421875, + "learning_rate": 0.0018635252911457238, + "loss": 0.0913, + "step": 17872 + }, + { + "epoch": 0.15514622268903916, + "grad_norm": 0.15625, + "learning_rate": 0.0018635095339631735, + "loss": 0.1484, + "step": 17873 + }, + { + "epoch": 0.15515490316924332, + "grad_norm": 0.287109375, + "learning_rate": 0.0018634937759456607, + "loss": 0.1328, + "step": 17874 + }, + { + "epoch": 0.1551635836494475, + "grad_norm": 0.357421875, + "learning_rate": 0.0018634780170932027, + "loss": 0.1089, + "step": 17875 + }, + { + "epoch": 0.15517226412965165, + "grad_norm": 0.13671875, + "learning_rate": 0.0018634622574058163, + "loss": 0.0942, + "step": 17876 + }, + { + "epoch": 0.15518094460985582, + "grad_norm": 0.51171875, + "learning_rate": 0.0018634464968835193, + "loss": 0.1099, + "step": 17877 + }, + { + "epoch": 0.15518962509005998, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018634307355263285, + "loss": 0.1152, + "step": 17878 + }, + { + "epoch": 0.15519830557026415, + "grad_norm": 0.345703125, + "learning_rate": 0.0018634149733342616, + "loss": 0.125, + "step": 17879 + }, + { + "epoch": 0.15520698605046832, + "grad_norm": 0.275390625, + "learning_rate": 0.001863399210307335, + "loss": 0.1191, + "step": 17880 + }, + { + "epoch": 0.15521566653067248, + "grad_norm": 0.234375, + "learning_rate": 0.0018633834464455673, + "loss": 0.1133, + "step": 17881 + }, + { + "epoch": 0.15522434701087665, + "grad_norm": 0.2734375, + "learning_rate": 0.0018633676817489742, + "loss": 0.1182, + "step": 17882 + }, + { + "epoch": 0.1552330274910808, + "grad_norm": 0.337890625, + "learning_rate": 0.0018633519162175739, + "loss": 0.1816, + "step": 17883 + }, + { + "epoch": 0.15524170797128498, + "grad_norm": 0.228515625, + "learning_rate": 0.0018633361498513836, + "loss": 0.1123, + "step": 17884 + }, + { + "epoch": 0.15525038845148914, + "grad_norm": 0.15234375, + "learning_rate": 0.0018633203826504202, + "loss": 0.1143, + "step": 17885 + }, + { + "epoch": 0.1552590689316933, + "grad_norm": 0.09033203125, + "learning_rate": 0.001863304614614701, + "loss": 0.0967, + "step": 17886 + }, + { + "epoch": 0.15526774941189747, + "grad_norm": 0.400390625, + "learning_rate": 0.0018632888457442436, + "loss": 0.168, + "step": 17887 + }, + { + "epoch": 0.15527642989210164, + "grad_norm": 0.0810546875, + "learning_rate": 0.0018632730760390647, + "loss": 0.0942, + "step": 17888 + }, + { + "epoch": 0.1552851103723058, + "grad_norm": 0.99609375, + "learning_rate": 0.001863257305499182, + "loss": 0.104, + "step": 17889 + }, + { + "epoch": 0.15529379085250997, + "grad_norm": 0.08837890625, + "learning_rate": 0.0018632415341246127, + "loss": 0.0889, + "step": 17890 + }, + { + "epoch": 0.15530247133271413, + "grad_norm": 0.71484375, + "learning_rate": 0.0018632257619153737, + "loss": 0.0908, + "step": 17891 + }, + { + "epoch": 0.1553111518129183, + "grad_norm": 0.203125, + "learning_rate": 0.0018632099888714825, + "loss": 0.124, + "step": 17892 + }, + { + "epoch": 0.15531983229312246, + "grad_norm": 0.84375, + "learning_rate": 0.0018631942149929567, + "loss": 0.1167, + "step": 17893 + }, + { + "epoch": 0.15532851277332663, + "grad_norm": 0.150390625, + "learning_rate": 0.0018631784402798129, + "loss": 0.1455, + "step": 17894 + }, + { + "epoch": 0.1553371932535308, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018631626647320682, + "loss": 0.1006, + "step": 17895 + }, + { + "epoch": 0.15534587373373496, + "grad_norm": 1.1328125, + "learning_rate": 0.001863146888349741, + "loss": 0.1182, + "step": 17896 + }, + { + "epoch": 0.15535455421393912, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018631311111328474, + "loss": 0.1299, + "step": 17897 + }, + { + "epoch": 0.1553632346941433, + "grad_norm": 0.107421875, + "learning_rate": 0.001863115333081405, + "loss": 0.0864, + "step": 17898 + }, + { + "epoch": 0.15537191517434745, + "grad_norm": 0.140625, + "learning_rate": 0.0018630995541954315, + "loss": 0.0957, + "step": 17899 + }, + { + "epoch": 0.15538059565455162, + "grad_norm": 0.6015625, + "learning_rate": 0.0018630837744749436, + "loss": 0.1133, + "step": 17900 + }, + { + "epoch": 0.15538927613475578, + "grad_norm": 0.158203125, + "learning_rate": 0.0018630679939199586, + "loss": 0.1357, + "step": 17901 + }, + { + "epoch": 0.15539795661495995, + "grad_norm": 0.162109375, + "learning_rate": 0.001863052212530494, + "loss": 0.1118, + "step": 17902 + }, + { + "epoch": 0.1554066370951641, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018630364303065673, + "loss": 0.1572, + "step": 17903 + }, + { + "epoch": 0.15541531757536828, + "grad_norm": 0.1083984375, + "learning_rate": 0.0018630206472481949, + "loss": 0.0938, + "step": 17904 + }, + { + "epoch": 0.15542399805557244, + "grad_norm": 0.302734375, + "learning_rate": 0.0018630048633553951, + "loss": 0.124, + "step": 17905 + }, + { + "epoch": 0.1554326785357766, + "grad_norm": 0.87109375, + "learning_rate": 0.001862989078628184, + "loss": 0.168, + "step": 17906 + }, + { + "epoch": 0.15544135901598077, + "grad_norm": 0.8671875, + "learning_rate": 0.00186297329306658, + "loss": 0.1094, + "step": 17907 + }, + { + "epoch": 0.15545003949618494, + "grad_norm": 0.27734375, + "learning_rate": 0.0018629575066705997, + "loss": 0.0811, + "step": 17908 + }, + { + "epoch": 0.1554587199763891, + "grad_norm": 0.302734375, + "learning_rate": 0.0018629417194402607, + "loss": 0.1182, + "step": 17909 + }, + { + "epoch": 0.15546740045659327, + "grad_norm": 0.1318359375, + "learning_rate": 0.00186292593137558, + "loss": 0.1055, + "step": 17910 + }, + { + "epoch": 0.15547608093679743, + "grad_norm": 0.75, + "learning_rate": 0.0018629101424765748, + "loss": 0.1113, + "step": 17911 + }, + { + "epoch": 0.1554847614170016, + "grad_norm": 0.42578125, + "learning_rate": 0.001862894352743263, + "loss": 0.1758, + "step": 17912 + }, + { + "epoch": 0.15549344189720576, + "grad_norm": 0.2421875, + "learning_rate": 0.001862878562175661, + "loss": 0.1123, + "step": 17913 + }, + { + "epoch": 0.15550212237740993, + "grad_norm": 0.1005859375, + "learning_rate": 0.0018628627707737864, + "loss": 0.0869, + "step": 17914 + }, + { + "epoch": 0.1555108028576141, + "grad_norm": 0.2373046875, + "learning_rate": 0.001862846978537657, + "loss": 0.0913, + "step": 17915 + }, + { + "epoch": 0.15551948333781826, + "grad_norm": 0.53125, + "learning_rate": 0.001862831185467289, + "loss": 0.0977, + "step": 17916 + }, + { + "epoch": 0.15552816381802242, + "grad_norm": 0.125, + "learning_rate": 0.0018628153915627005, + "loss": 0.1094, + "step": 17917 + }, + { + "epoch": 0.1555368442982266, + "grad_norm": 0.85546875, + "learning_rate": 0.001862799596823909, + "loss": 0.0879, + "step": 17918 + }, + { + "epoch": 0.15554552477843075, + "grad_norm": 0.60546875, + "learning_rate": 0.0018627838012509309, + "loss": 0.1426, + "step": 17919 + }, + { + "epoch": 0.15555420525863492, + "grad_norm": 0.265625, + "learning_rate": 0.0018627680048437837, + "loss": 0.1465, + "step": 17920 + }, + { + "epoch": 0.15556288573883909, + "grad_norm": 0.7734375, + "learning_rate": 0.001862752207602485, + "loss": 0.1035, + "step": 17921 + }, + { + "epoch": 0.15557156621904325, + "grad_norm": 0.4765625, + "learning_rate": 0.001862736409527052, + "loss": 0.1055, + "step": 17922 + }, + { + "epoch": 0.15558024669924742, + "grad_norm": 0.236328125, + "learning_rate": 0.001862720610617502, + "loss": 0.1328, + "step": 17923 + }, + { + "epoch": 0.15558892717945158, + "grad_norm": 0.07421875, + "learning_rate": 0.0018627048108738522, + "loss": 0.0815, + "step": 17924 + }, + { + "epoch": 0.15559760765965575, + "grad_norm": 0.318359375, + "learning_rate": 0.00186268901029612, + "loss": 0.0854, + "step": 17925 + }, + { + "epoch": 0.1556062881398599, + "grad_norm": 0.52734375, + "learning_rate": 0.0018626732088843224, + "loss": 0.1084, + "step": 17926 + }, + { + "epoch": 0.15561496862006408, + "grad_norm": 0.24609375, + "learning_rate": 0.001862657406638477, + "loss": 0.1377, + "step": 17927 + }, + { + "epoch": 0.1556236491002682, + "grad_norm": 0.91015625, + "learning_rate": 0.0018626416035586004, + "loss": 0.1143, + "step": 17928 + }, + { + "epoch": 0.15563232958047238, + "grad_norm": 0.54296875, + "learning_rate": 0.0018626257996447108, + "loss": 0.1523, + "step": 17929 + }, + { + "epoch": 0.15564101006067654, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018626099948968252, + "loss": 0.1533, + "step": 17930 + }, + { + "epoch": 0.1556496905408807, + "grad_norm": 0.365234375, + "learning_rate": 0.0018625941893149606, + "loss": 0.1602, + "step": 17931 + }, + { + "epoch": 0.15565837102108487, + "grad_norm": 0.62109375, + "learning_rate": 0.0018625783828991346, + "loss": 0.1533, + "step": 17932 + }, + { + "epoch": 0.15566705150128904, + "grad_norm": 0.412109375, + "learning_rate": 0.001862562575649364, + "loss": 0.1445, + "step": 17933 + }, + { + "epoch": 0.1556757319814932, + "grad_norm": 0.5078125, + "learning_rate": 0.0018625467675656667, + "loss": 0.0967, + "step": 17934 + }, + { + "epoch": 0.15568441246169737, + "grad_norm": 0.125, + "learning_rate": 0.0018625309586480596, + "loss": 0.1074, + "step": 17935 + }, + { + "epoch": 0.15569309294190153, + "grad_norm": 0.244140625, + "learning_rate": 0.0018625151488965601, + "loss": 0.1094, + "step": 17936 + }, + { + "epoch": 0.1557017734221057, + "grad_norm": 0.384765625, + "learning_rate": 0.0018624993383111856, + "loss": 0.0957, + "step": 17937 + }, + { + "epoch": 0.15571045390230986, + "grad_norm": 0.23046875, + "learning_rate": 0.0018624835268919533, + "loss": 0.1172, + "step": 17938 + }, + { + "epoch": 0.15571913438251403, + "grad_norm": 0.69140625, + "learning_rate": 0.0018624677146388804, + "loss": 0.085, + "step": 17939 + }, + { + "epoch": 0.1557278148627182, + "grad_norm": 0.2431640625, + "learning_rate": 0.001862451901551984, + "loss": 0.1309, + "step": 17940 + }, + { + "epoch": 0.15573649534292236, + "grad_norm": 0.150390625, + "learning_rate": 0.001862436087631282, + "loss": 0.1187, + "step": 17941 + }, + { + "epoch": 0.15574517582312652, + "grad_norm": 0.2265625, + "learning_rate": 0.0018624202728767913, + "loss": 0.0957, + "step": 17942 + }, + { + "epoch": 0.1557538563033307, + "grad_norm": 0.5078125, + "learning_rate": 0.0018624044572885293, + "loss": 0.125, + "step": 17943 + }, + { + "epoch": 0.15576253678353486, + "grad_norm": 0.75, + "learning_rate": 0.001862388640866513, + "loss": 0.1504, + "step": 17944 + }, + { + "epoch": 0.15577121726373902, + "grad_norm": 0.421875, + "learning_rate": 0.00186237282361076, + "loss": 0.1064, + "step": 17945 + }, + { + "epoch": 0.15577989774394319, + "grad_norm": 0.294921875, + "learning_rate": 0.0018623570055212877, + "loss": 0.1504, + "step": 17946 + }, + { + "epoch": 0.15578857822414735, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018623411865981133, + "loss": 0.0898, + "step": 17947 + }, + { + "epoch": 0.15579725870435152, + "grad_norm": 0.09912109375, + "learning_rate": 0.001862325366841254, + "loss": 0.0947, + "step": 17948 + }, + { + "epoch": 0.15580593918455568, + "grad_norm": 0.365234375, + "learning_rate": 0.0018623095462507267, + "loss": 0.1182, + "step": 17949 + }, + { + "epoch": 0.15581461966475985, + "grad_norm": 0.126953125, + "learning_rate": 0.0018622937248265494, + "loss": 0.123, + "step": 17950 + }, + { + "epoch": 0.155823300144964, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018622779025687392, + "loss": 0.1133, + "step": 17951 + }, + { + "epoch": 0.15583198062516818, + "grad_norm": 0.51953125, + "learning_rate": 0.0018622620794773136, + "loss": 0.125, + "step": 17952 + }, + { + "epoch": 0.15584066110537234, + "grad_norm": 1.21875, + "learning_rate": 0.0018622462555522893, + "loss": 0.1855, + "step": 17953 + }, + { + "epoch": 0.1558493415855765, + "grad_norm": 0.234375, + "learning_rate": 0.001862230430793684, + "loss": 0.1045, + "step": 17954 + }, + { + "epoch": 0.15585802206578067, + "grad_norm": 0.10107421875, + "learning_rate": 0.001862214605201515, + "loss": 0.0952, + "step": 17955 + }, + { + "epoch": 0.15586670254598484, + "grad_norm": 0.369140625, + "learning_rate": 0.0018621987787757997, + "loss": 0.1562, + "step": 17956 + }, + { + "epoch": 0.155875383026189, + "grad_norm": 0.59765625, + "learning_rate": 0.0018621829515165548, + "loss": 0.1104, + "step": 17957 + }, + { + "epoch": 0.15588406350639317, + "grad_norm": 0.09033203125, + "learning_rate": 0.0018621671234237982, + "loss": 0.0938, + "step": 17958 + }, + { + "epoch": 0.15589274398659733, + "grad_norm": 0.265625, + "learning_rate": 0.0018621512944975475, + "loss": 0.1377, + "step": 17959 + }, + { + "epoch": 0.1559014244668015, + "grad_norm": 0.162109375, + "learning_rate": 0.0018621354647378192, + "loss": 0.1104, + "step": 17960 + }, + { + "epoch": 0.15591010494700566, + "grad_norm": 1.0625, + "learning_rate": 0.001862119634144631, + "loss": 0.1895, + "step": 17961 + }, + { + "epoch": 0.15591878542720983, + "grad_norm": 0.337890625, + "learning_rate": 0.0018621038027180005, + "loss": 0.1396, + "step": 17962 + }, + { + "epoch": 0.155927465907414, + "grad_norm": 0.177734375, + "learning_rate": 0.0018620879704579446, + "loss": 0.1211, + "step": 17963 + }, + { + "epoch": 0.15593614638761816, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018620721373644809, + "loss": 0.0942, + "step": 17964 + }, + { + "epoch": 0.15594482686782232, + "grad_norm": 0.203125, + "learning_rate": 0.0018620563034376263, + "loss": 0.125, + "step": 17965 + }, + { + "epoch": 0.1559535073480265, + "grad_norm": 0.2158203125, + "learning_rate": 0.0018620404686773987, + "loss": 0.1113, + "step": 17966 + }, + { + "epoch": 0.15596218782823065, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018620246330838149, + "loss": 0.106, + "step": 17967 + }, + { + "epoch": 0.15597086830843482, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018620087966568922, + "loss": 0.0942, + "step": 17968 + }, + { + "epoch": 0.15597954878863898, + "grad_norm": 0.279296875, + "learning_rate": 0.0018619929593966485, + "loss": 0.123, + "step": 17969 + }, + { + "epoch": 0.15598822926884315, + "grad_norm": 0.189453125, + "learning_rate": 0.0018619771213031005, + "loss": 0.1504, + "step": 17970 + }, + { + "epoch": 0.1559969097490473, + "grad_norm": 0.8046875, + "learning_rate": 0.0018619612823762657, + "loss": 0.1289, + "step": 17971 + }, + { + "epoch": 0.15600559022925148, + "grad_norm": 0.306640625, + "learning_rate": 0.0018619454426161617, + "loss": 0.1328, + "step": 17972 + }, + { + "epoch": 0.15601427070945564, + "grad_norm": 0.11962890625, + "learning_rate": 0.0018619296020228055, + "loss": 0.1338, + "step": 17973 + }, + { + "epoch": 0.1560229511896598, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018619137605962147, + "loss": 0.1689, + "step": 17974 + }, + { + "epoch": 0.15603163166986397, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018618979183364062, + "loss": 0.0767, + "step": 17975 + }, + { + "epoch": 0.15604031215006814, + "grad_norm": 0.474609375, + "learning_rate": 0.001861882075243398, + "loss": 0.1196, + "step": 17976 + }, + { + "epoch": 0.1560489926302723, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018618662313172065, + "loss": 0.1201, + "step": 17977 + }, + { + "epoch": 0.15605767311047647, + "grad_norm": 0.416015625, + "learning_rate": 0.0018618503865578497, + "loss": 0.0957, + "step": 17978 + }, + { + "epoch": 0.15606635359068063, + "grad_norm": 0.0888671875, + "learning_rate": 0.001861834540965345, + "loss": 0.0874, + "step": 17979 + }, + { + "epoch": 0.1560750340708848, + "grad_norm": 0.5625, + "learning_rate": 0.0018618186945397093, + "loss": 0.063, + "step": 17980 + }, + { + "epoch": 0.15608371455108896, + "grad_norm": 0.734375, + "learning_rate": 0.00186180284728096, + "loss": 0.1426, + "step": 17981 + }, + { + "epoch": 0.15609239503129313, + "grad_norm": 0.1640625, + "learning_rate": 0.001861786999189115, + "loss": 0.1758, + "step": 17982 + }, + { + "epoch": 0.1561010755114973, + "grad_norm": 0.65234375, + "learning_rate": 0.0018617711502641905, + "loss": 0.1553, + "step": 17983 + }, + { + "epoch": 0.15610975599170146, + "grad_norm": 0.345703125, + "learning_rate": 0.001861755300506205, + "loss": 0.1572, + "step": 17984 + }, + { + "epoch": 0.15611843647190563, + "grad_norm": 0.90234375, + "learning_rate": 0.0018617394499151755, + "loss": 0.1138, + "step": 17985 + }, + { + "epoch": 0.1561271169521098, + "grad_norm": 0.58203125, + "learning_rate": 0.0018617235984911187, + "loss": 0.085, + "step": 17986 + }, + { + "epoch": 0.15613579743231396, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018617077462340526, + "loss": 0.1523, + "step": 17987 + }, + { + "epoch": 0.15614447791251812, + "grad_norm": 0.462890625, + "learning_rate": 0.0018616918931439943, + "loss": 0.1069, + "step": 17988 + }, + { + "epoch": 0.15615315839272229, + "grad_norm": 0.08349609375, + "learning_rate": 0.0018616760392209616, + "loss": 0.1011, + "step": 17989 + }, + { + "epoch": 0.15616183887292645, + "grad_norm": 0.93359375, + "learning_rate": 0.0018616601844649713, + "loss": 0.125, + "step": 17990 + }, + { + "epoch": 0.15617051935313062, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018616443288760405, + "loss": 0.1143, + "step": 17991 + }, + { + "epoch": 0.15617919983333478, + "grad_norm": 0.19140625, + "learning_rate": 0.001861628472454187, + "loss": 0.1035, + "step": 17992 + }, + { + "epoch": 0.15618788031353895, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018616126151994282, + "loss": 0.1426, + "step": 17993 + }, + { + "epoch": 0.1561965607937431, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018615967571117811, + "loss": 0.127, + "step": 17994 + }, + { + "epoch": 0.15620524127394728, + "grad_norm": 0.302734375, + "learning_rate": 0.0018615808981912634, + "loss": 0.1172, + "step": 17995 + }, + { + "epoch": 0.15621392175415144, + "grad_norm": 0.8828125, + "learning_rate": 0.0018615650384378921, + "loss": 0.1357, + "step": 17996 + }, + { + "epoch": 0.1562226022343556, + "grad_norm": 0.94140625, + "learning_rate": 0.0018615491778516849, + "loss": 0.1406, + "step": 17997 + }, + { + "epoch": 0.15623128271455977, + "grad_norm": 0.337890625, + "learning_rate": 0.0018615333164326588, + "loss": 0.1299, + "step": 17998 + }, + { + "epoch": 0.15623996319476394, + "grad_norm": 0.6640625, + "learning_rate": 0.0018615174541808314, + "loss": 0.0977, + "step": 17999 + }, + { + "epoch": 0.1562486436749681, + "grad_norm": 0.1337890625, + "learning_rate": 0.00186150159109622, + "loss": 0.0928, + "step": 18000 + }, + { + "epoch": 0.15625732415517227, + "grad_norm": 1.1484375, + "learning_rate": 0.0018614857271788418, + "loss": 0.1426, + "step": 18001 + }, + { + "epoch": 0.15626600463537643, + "grad_norm": 0.330078125, + "learning_rate": 0.0018614698624287145, + "loss": 0.123, + "step": 18002 + }, + { + "epoch": 0.1562746851155806, + "grad_norm": 0.10009765625, + "learning_rate": 0.001861453996845855, + "loss": 0.1338, + "step": 18003 + }, + { + "epoch": 0.15628336559578476, + "grad_norm": 0.267578125, + "learning_rate": 0.001861438130430281, + "loss": 0.1187, + "step": 18004 + }, + { + "epoch": 0.15629204607598893, + "grad_norm": 0.19921875, + "learning_rate": 0.0018614222631820093, + "loss": 0.1279, + "step": 18005 + }, + { + "epoch": 0.1563007265561931, + "grad_norm": 0.984375, + "learning_rate": 0.0018614063951010581, + "loss": 0.1211, + "step": 18006 + }, + { + "epoch": 0.15630940703639726, + "grad_norm": 0.62890625, + "learning_rate": 0.001861390526187444, + "loss": 0.1201, + "step": 18007 + }, + { + "epoch": 0.15631808751660142, + "grad_norm": 0.328125, + "learning_rate": 0.0018613746564411849, + "loss": 0.105, + "step": 18008 + }, + { + "epoch": 0.1563267679968056, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018613587858622976, + "loss": 0.1328, + "step": 18009 + }, + { + "epoch": 0.15633544847700975, + "grad_norm": 0.546875, + "learning_rate": 0.0018613429144508, + "loss": 0.1108, + "step": 18010 + }, + { + "epoch": 0.15634412895721392, + "grad_norm": 0.71875, + "learning_rate": 0.0018613270422067096, + "loss": 0.0938, + "step": 18011 + }, + { + "epoch": 0.15635280943741808, + "grad_norm": 1.640625, + "learning_rate": 0.0018613111691300427, + "loss": 0.1289, + "step": 18012 + }, + { + "epoch": 0.15636148991762225, + "grad_norm": 0.97265625, + "learning_rate": 0.0018612952952208176, + "loss": 0.0947, + "step": 18013 + }, + { + "epoch": 0.1563701703978264, + "grad_norm": 0.5546875, + "learning_rate": 0.0018612794204790513, + "loss": 0.1172, + "step": 18014 + }, + { + "epoch": 0.15637885087803058, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018612635449047614, + "loss": 0.1152, + "step": 18015 + }, + { + "epoch": 0.15638753135823474, + "grad_norm": 0.29296875, + "learning_rate": 0.0018612476684979654, + "loss": 0.0938, + "step": 18016 + }, + { + "epoch": 0.1563962118384389, + "grad_norm": 0.212890625, + "learning_rate": 0.00186123179125868, + "loss": 0.0918, + "step": 18017 + }, + { + "epoch": 0.15640489231864307, + "grad_norm": 0.12255859375, + "learning_rate": 0.001861215913186923, + "loss": 0.1177, + "step": 18018 + }, + { + "epoch": 0.15641357279884724, + "grad_norm": 0.734375, + "learning_rate": 0.0018612000342827118, + "loss": 0.1221, + "step": 18019 + }, + { + "epoch": 0.1564222532790514, + "grad_norm": 0.369140625, + "learning_rate": 0.0018611841545460635, + "loss": 0.1084, + "step": 18020 + }, + { + "epoch": 0.15643093375925557, + "grad_norm": 0.189453125, + "learning_rate": 0.0018611682739769959, + "loss": 0.1133, + "step": 18021 + }, + { + "epoch": 0.15643961423945973, + "grad_norm": 0.828125, + "learning_rate": 0.001861152392575526, + "loss": 0.1309, + "step": 18022 + }, + { + "epoch": 0.1564482947196639, + "grad_norm": 0.56640625, + "learning_rate": 0.0018611365103416713, + "loss": 0.1348, + "step": 18023 + }, + { + "epoch": 0.15645697519986806, + "grad_norm": 0.1357421875, + "learning_rate": 0.001861120627275449, + "loss": 0.0898, + "step": 18024 + }, + { + "epoch": 0.15646565568007223, + "grad_norm": 0.15625, + "learning_rate": 0.0018611047433768771, + "loss": 0.1172, + "step": 18025 + }, + { + "epoch": 0.1564743361602764, + "grad_norm": 0.11376953125, + "learning_rate": 0.001861088858645972, + "loss": 0.1094, + "step": 18026 + }, + { + "epoch": 0.15648301664048056, + "grad_norm": 0.6875, + "learning_rate": 0.0018610729730827516, + "loss": 0.0859, + "step": 18027 + }, + { + "epoch": 0.15649169712068473, + "grad_norm": 0.234375, + "learning_rate": 0.0018610570866872333, + "loss": 0.1016, + "step": 18028 + }, + { + "epoch": 0.1565003776008889, + "grad_norm": 0.248046875, + "learning_rate": 0.0018610411994594346, + "loss": 0.1016, + "step": 18029 + }, + { + "epoch": 0.15650905808109306, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018610253113993723, + "loss": 0.0654, + "step": 18030 + }, + { + "epoch": 0.15651773856129722, + "grad_norm": 0.150390625, + "learning_rate": 0.0018610094225070643, + "loss": 0.1621, + "step": 18031 + }, + { + "epoch": 0.15652641904150139, + "grad_norm": 0.330078125, + "learning_rate": 0.001860993532782528, + "loss": 0.1064, + "step": 18032 + }, + { + "epoch": 0.15653509952170555, + "grad_norm": 0.62890625, + "learning_rate": 0.0018609776422257808, + "loss": 0.1465, + "step": 18033 + }, + { + "epoch": 0.15654378000190972, + "grad_norm": 0.365234375, + "learning_rate": 0.0018609617508368395, + "loss": 0.1035, + "step": 18034 + }, + { + "epoch": 0.15655246048211388, + "grad_norm": 0.9296875, + "learning_rate": 0.0018609458586157218, + "loss": 0.1084, + "step": 18035 + }, + { + "epoch": 0.15656114096231805, + "grad_norm": 0.2421875, + "learning_rate": 0.0018609299655624454, + "loss": 0.1465, + "step": 18036 + }, + { + "epoch": 0.1565698214425222, + "grad_norm": 1.703125, + "learning_rate": 0.0018609140716770271, + "loss": 0.1299, + "step": 18037 + }, + { + "epoch": 0.15657850192272638, + "grad_norm": 0.46484375, + "learning_rate": 0.001860898176959485, + "loss": 0.1543, + "step": 18038 + }, + { + "epoch": 0.15658718240293054, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018608822814098359, + "loss": 0.0977, + "step": 18039 + }, + { + "epoch": 0.1565958628831347, + "grad_norm": 0.1064453125, + "learning_rate": 0.0018608663850280974, + "loss": 0.1113, + "step": 18040 + }, + { + "epoch": 0.15660454336333887, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018608504878142868, + "loss": 0.0986, + "step": 18041 + }, + { + "epoch": 0.15661322384354304, + "grad_norm": 0.201171875, + "learning_rate": 0.0018608345897684218, + "loss": 0.1143, + "step": 18042 + }, + { + "epoch": 0.1566219043237472, + "grad_norm": 0.08056640625, + "learning_rate": 0.001860818690890519, + "loss": 0.0938, + "step": 18043 + }, + { + "epoch": 0.15663058480395137, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018608027911805967, + "loss": 0.083, + "step": 18044 + }, + { + "epoch": 0.15663926528415553, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018607868906386719, + "loss": 0.1001, + "step": 18045 + }, + { + "epoch": 0.1566479457643597, + "grad_norm": 0.5703125, + "learning_rate": 0.0018607709892647618, + "loss": 0.0977, + "step": 18046 + }, + { + "epoch": 0.15665662624456386, + "grad_norm": 0.220703125, + "learning_rate": 0.0018607550870588842, + "loss": 0.1289, + "step": 18047 + }, + { + "epoch": 0.15666530672476803, + "grad_norm": 0.30859375, + "learning_rate": 0.001860739184021056, + "loss": 0.0806, + "step": 18048 + }, + { + "epoch": 0.1566739872049722, + "grad_norm": 0.0927734375, + "learning_rate": 0.001860723280151295, + "loss": 0.1094, + "step": 18049 + }, + { + "epoch": 0.15668266768517636, + "grad_norm": 0.171875, + "learning_rate": 0.0018607073754496185, + "loss": 0.1025, + "step": 18050 + }, + { + "epoch": 0.1566913481653805, + "grad_norm": 0.328125, + "learning_rate": 0.0018606914699160438, + "loss": 0.1377, + "step": 18051 + }, + { + "epoch": 0.15670002864558466, + "grad_norm": 0.140625, + "learning_rate": 0.0018606755635505884, + "loss": 0.1006, + "step": 18052 + }, + { + "epoch": 0.15670870912578883, + "grad_norm": 0.61328125, + "learning_rate": 0.0018606596563532694, + "loss": 0.1152, + "step": 18053 + }, + { + "epoch": 0.156717389605993, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018606437483241044, + "loss": 0.0947, + "step": 18054 + }, + { + "epoch": 0.15672607008619716, + "grad_norm": 0.6015625, + "learning_rate": 0.0018606278394631113, + "loss": 0.2012, + "step": 18055 + }, + { + "epoch": 0.15673475056640132, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018606119297703066, + "loss": 0.0933, + "step": 18056 + }, + { + "epoch": 0.15674343104660549, + "grad_norm": 0.470703125, + "learning_rate": 0.0018605960192457083, + "loss": 0.1309, + "step": 18057 + }, + { + "epoch": 0.15675211152680965, + "grad_norm": 0.26171875, + "learning_rate": 0.0018605801078893335, + "loss": 0.1914, + "step": 18058 + }, + { + "epoch": 0.15676079200701382, + "grad_norm": 0.486328125, + "learning_rate": 0.0018605641957012, + "loss": 0.1504, + "step": 18059 + }, + { + "epoch": 0.15676947248721798, + "grad_norm": 0.671875, + "learning_rate": 0.0018605482826813244, + "loss": 0.125, + "step": 18060 + }, + { + "epoch": 0.15677815296742215, + "grad_norm": 0.28125, + "learning_rate": 0.001860532368829725, + "loss": 0.1064, + "step": 18061 + }, + { + "epoch": 0.1567868334476263, + "grad_norm": 0.1953125, + "learning_rate": 0.0018605164541464188, + "loss": 0.0811, + "step": 18062 + }, + { + "epoch": 0.15679551392783048, + "grad_norm": 0.578125, + "learning_rate": 0.0018605005386314233, + "loss": 0.126, + "step": 18063 + }, + { + "epoch": 0.15680419440803464, + "grad_norm": 0.1064453125, + "learning_rate": 0.0018604846222847558, + "loss": 0.084, + "step": 18064 + }, + { + "epoch": 0.1568128748882388, + "grad_norm": 0.451171875, + "learning_rate": 0.0018604687051064336, + "loss": 0.1035, + "step": 18065 + }, + { + "epoch": 0.15682155536844297, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018604527870964742, + "loss": 0.0933, + "step": 18066 + }, + { + "epoch": 0.15683023584864714, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018604368682548953, + "loss": 0.0854, + "step": 18067 + }, + { + "epoch": 0.1568389163288513, + "grad_norm": 0.103515625, + "learning_rate": 0.001860420948581714, + "loss": 0.1152, + "step": 18068 + }, + { + "epoch": 0.15684759680905547, + "grad_norm": 0.1240234375, + "learning_rate": 0.0018604050280769478, + "loss": 0.0952, + "step": 18069 + }, + { + "epoch": 0.15685627728925963, + "grad_norm": 0.609375, + "learning_rate": 0.001860389106740614, + "loss": 0.1289, + "step": 18070 + }, + { + "epoch": 0.1568649577694638, + "grad_norm": 0.130859375, + "learning_rate": 0.0018603731845727302, + "loss": 0.1279, + "step": 18071 + }, + { + "epoch": 0.15687363824966796, + "grad_norm": 0.1123046875, + "learning_rate": 0.0018603572615733136, + "loss": 0.1162, + "step": 18072 + }, + { + "epoch": 0.15688231872987213, + "grad_norm": 0.4921875, + "learning_rate": 0.0018603413377423818, + "loss": 0.1201, + "step": 18073 + }, + { + "epoch": 0.1568909992100763, + "grad_norm": 0.4609375, + "learning_rate": 0.001860325413079952, + "loss": 0.1211, + "step": 18074 + }, + { + "epoch": 0.15689967969028046, + "grad_norm": 0.8125, + "learning_rate": 0.0018603094875860417, + "loss": 0.1533, + "step": 18075 + }, + { + "epoch": 0.15690836017048462, + "grad_norm": 0.23046875, + "learning_rate": 0.0018602935612606689, + "loss": 0.1123, + "step": 18076 + }, + { + "epoch": 0.1569170406506888, + "grad_norm": 0.24609375, + "learning_rate": 0.00186027763410385, + "loss": 0.103, + "step": 18077 + }, + { + "epoch": 0.15692572113089295, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018602617061156028, + "loss": 0.1973, + "step": 18078 + }, + { + "epoch": 0.15693440161109712, + "grad_norm": 0.1953125, + "learning_rate": 0.001860245777295945, + "loss": 0.1338, + "step": 18079 + }, + { + "epoch": 0.15694308209130128, + "grad_norm": 0.06689453125, + "learning_rate": 0.001860229847644894, + "loss": 0.1025, + "step": 18080 + }, + { + "epoch": 0.15695176257150545, + "grad_norm": 0.36328125, + "learning_rate": 0.0018602139171624667, + "loss": 0.1543, + "step": 18081 + }, + { + "epoch": 0.1569604430517096, + "grad_norm": 0.62109375, + "learning_rate": 0.0018601979858486812, + "loss": 0.1211, + "step": 18082 + }, + { + "epoch": 0.15696912353191378, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018601820537035546, + "loss": 0.1069, + "step": 18083 + }, + { + "epoch": 0.15697780401211794, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018601661207271044, + "loss": 0.1309, + "step": 18084 + }, + { + "epoch": 0.1569864844923221, + "grad_norm": 0.90625, + "learning_rate": 0.0018601501869193475, + "loss": 0.1001, + "step": 18085 + }, + { + "epoch": 0.15699516497252627, + "grad_norm": 0.388671875, + "learning_rate": 0.001860134252280302, + "loss": 0.1299, + "step": 18086 + }, + { + "epoch": 0.15700384545273044, + "grad_norm": 0.345703125, + "learning_rate": 0.0018601183168099852, + "loss": 0.1211, + "step": 18087 + }, + { + "epoch": 0.1570125259329346, + "grad_norm": 0.64453125, + "learning_rate": 0.0018601023805084143, + "loss": 0.1211, + "step": 18088 + }, + { + "epoch": 0.15702120641313877, + "grad_norm": 0.259765625, + "learning_rate": 0.001860086443375607, + "loss": 0.1455, + "step": 18089 + }, + { + "epoch": 0.15702988689334293, + "grad_norm": 0.1484375, + "learning_rate": 0.0018600705054115808, + "loss": 0.1152, + "step": 18090 + }, + { + "epoch": 0.1570385673735471, + "grad_norm": 0.68359375, + "learning_rate": 0.0018600545666163527, + "loss": 0.1328, + "step": 18091 + }, + { + "epoch": 0.15704724785375127, + "grad_norm": 0.275390625, + "learning_rate": 0.00186003862698994, + "loss": 0.1045, + "step": 18092 + }, + { + "epoch": 0.15705592833395543, + "grad_norm": 0.228515625, + "learning_rate": 0.001860022686532361, + "loss": 0.0781, + "step": 18093 + }, + { + "epoch": 0.1570646088141596, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018600067452436324, + "loss": 0.1104, + "step": 18094 + }, + { + "epoch": 0.15707328929436376, + "grad_norm": 0.12890625, + "learning_rate": 0.001859990803123772, + "loss": 0.0664, + "step": 18095 + }, + { + "epoch": 0.15708196977456793, + "grad_norm": 0.126953125, + "learning_rate": 0.0018599748601727966, + "loss": 0.1143, + "step": 18096 + }, + { + "epoch": 0.1570906502547721, + "grad_norm": 0.267578125, + "learning_rate": 0.0018599589163907245, + "loss": 0.1162, + "step": 18097 + }, + { + "epoch": 0.15709933073497626, + "grad_norm": 0.435546875, + "learning_rate": 0.0018599429717775728, + "loss": 0.1309, + "step": 18098 + }, + { + "epoch": 0.15710801121518042, + "grad_norm": 0.1162109375, + "learning_rate": 0.0018599270263333588, + "loss": 0.0889, + "step": 18099 + }, + { + "epoch": 0.15711669169538459, + "grad_norm": 1.328125, + "learning_rate": 0.0018599110800580996, + "loss": 0.126, + "step": 18100 + }, + { + "epoch": 0.15712537217558875, + "grad_norm": 0.43359375, + "learning_rate": 0.0018598951329518137, + "loss": 0.1377, + "step": 18101 + }, + { + "epoch": 0.15713405265579292, + "grad_norm": 2.625, + "learning_rate": 0.0018598791850145174, + "loss": 0.1211, + "step": 18102 + }, + { + "epoch": 0.15714273313599708, + "grad_norm": 0.55078125, + "learning_rate": 0.0018598632362462292, + "loss": 0.127, + "step": 18103 + }, + { + "epoch": 0.15715141361620125, + "grad_norm": 0.498046875, + "learning_rate": 0.0018598472866469655, + "loss": 0.0918, + "step": 18104 + }, + { + "epoch": 0.1571600940964054, + "grad_norm": 0.318359375, + "learning_rate": 0.0018598313362167443, + "loss": 0.1055, + "step": 18105 + }, + { + "epoch": 0.15716877457660958, + "grad_norm": 0.146484375, + "learning_rate": 0.0018598153849555832, + "loss": 0.082, + "step": 18106 + }, + { + "epoch": 0.15717745505681374, + "grad_norm": 0.21484375, + "learning_rate": 0.001859799432863499, + "loss": 0.1025, + "step": 18107 + }, + { + "epoch": 0.1571861355370179, + "grad_norm": 0.2041015625, + "learning_rate": 0.00185978347994051, + "loss": 0.1074, + "step": 18108 + }, + { + "epoch": 0.15719481601722207, + "grad_norm": 0.390625, + "learning_rate": 0.001859767526186633, + "loss": 0.1504, + "step": 18109 + }, + { + "epoch": 0.15720349649742624, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018597515716018856, + "loss": 0.1338, + "step": 18110 + }, + { + "epoch": 0.1572121769776304, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018597356161862853, + "loss": 0.1338, + "step": 18111 + }, + { + "epoch": 0.15722085745783457, + "grad_norm": 0.275390625, + "learning_rate": 0.0018597196599398499, + "loss": 0.0898, + "step": 18112 + }, + { + "epoch": 0.15722953793803873, + "grad_norm": 0.546875, + "learning_rate": 0.001859703702862596, + "loss": 0.1367, + "step": 18113 + }, + { + "epoch": 0.1572382184182429, + "grad_norm": 0.50390625, + "learning_rate": 0.0018596877449545417, + "loss": 0.1045, + "step": 18114 + }, + { + "epoch": 0.15724689889844706, + "grad_norm": 0.07666015625, + "learning_rate": 0.0018596717862157043, + "loss": 0.1104, + "step": 18115 + }, + { + "epoch": 0.15725557937865123, + "grad_norm": 0.287109375, + "learning_rate": 0.0018596558266461014, + "loss": 0.0923, + "step": 18116 + }, + { + "epoch": 0.1572642598588554, + "grad_norm": 0.296875, + "learning_rate": 0.00185963986624575, + "loss": 0.1465, + "step": 18117 + }, + { + "epoch": 0.15727294033905956, + "grad_norm": 1.046875, + "learning_rate": 0.0018596239050146682, + "loss": 0.1709, + "step": 18118 + }, + { + "epoch": 0.15728162081926372, + "grad_norm": 0.478515625, + "learning_rate": 0.001859607942952873, + "loss": 0.1172, + "step": 18119 + }, + { + "epoch": 0.1572903012994679, + "grad_norm": 0.578125, + "learning_rate": 0.001859591980060382, + "loss": 0.1182, + "step": 18120 + }, + { + "epoch": 0.15729898177967205, + "grad_norm": 0.498046875, + "learning_rate": 0.0018595760163372126, + "loss": 0.0869, + "step": 18121 + }, + { + "epoch": 0.15730766225987622, + "grad_norm": 0.25, + "learning_rate": 0.001859560051783382, + "loss": 0.1113, + "step": 18122 + }, + { + "epoch": 0.15731634274008038, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018595440863989083, + "loss": 0.0728, + "step": 18123 + }, + { + "epoch": 0.15732502322028455, + "grad_norm": 0.83203125, + "learning_rate": 0.0018595281201838083, + "loss": 0.1191, + "step": 18124 + }, + { + "epoch": 0.15733370370048871, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018595121531380999, + "loss": 0.1147, + "step": 18125 + }, + { + "epoch": 0.15734238418069288, + "grad_norm": 0.203125, + "learning_rate": 0.0018594961852618005, + "loss": 0.0923, + "step": 18126 + }, + { + "epoch": 0.15735106466089704, + "grad_norm": 0.08154296875, + "learning_rate": 0.0018594802165549276, + "loss": 0.1089, + "step": 18127 + }, + { + "epoch": 0.1573597451411012, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018594642470174984, + "loss": 0.1167, + "step": 18128 + }, + { + "epoch": 0.15736842562130537, + "grad_norm": 0.087890625, + "learning_rate": 0.0018594482766495304, + "loss": 0.1069, + "step": 18129 + }, + { + "epoch": 0.15737710610150954, + "grad_norm": 0.07080078125, + "learning_rate": 0.0018594323054510415, + "loss": 0.0981, + "step": 18130 + }, + { + "epoch": 0.1573857865817137, + "grad_norm": 0.451171875, + "learning_rate": 0.0018594163334220482, + "loss": 0.1074, + "step": 18131 + }, + { + "epoch": 0.15739446706191787, + "grad_norm": 0.07275390625, + "learning_rate": 0.0018594003605625692, + "loss": 0.1016, + "step": 18132 + }, + { + "epoch": 0.15740314754212203, + "grad_norm": 0.55078125, + "learning_rate": 0.0018593843868726211, + "loss": 0.1123, + "step": 18133 + }, + { + "epoch": 0.1574118280223262, + "grad_norm": 0.11279296875, + "learning_rate": 0.0018593684123522217, + "loss": 0.1279, + "step": 18134 + }, + { + "epoch": 0.15742050850253037, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018593524370013885, + "loss": 0.0991, + "step": 18135 + }, + { + "epoch": 0.15742918898273453, + "grad_norm": 0.4609375, + "learning_rate": 0.001859336460820139, + "loss": 0.1396, + "step": 18136 + }, + { + "epoch": 0.1574378694629387, + "grad_norm": 0.453125, + "learning_rate": 0.00185932048380849, + "loss": 0.1357, + "step": 18137 + }, + { + "epoch": 0.15744654994314286, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018593045059664603, + "loss": 0.1206, + "step": 18138 + }, + { + "epoch": 0.15745523042334703, + "grad_norm": 0.388671875, + "learning_rate": 0.001859288527294066, + "loss": 0.1123, + "step": 18139 + }, + { + "epoch": 0.1574639109035512, + "grad_norm": 0.06103515625, + "learning_rate": 0.0018592725477913255, + "loss": 0.0967, + "step": 18140 + }, + { + "epoch": 0.15747259138375536, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018592565674582558, + "loss": 0.104, + "step": 18141 + }, + { + "epoch": 0.15748127186395952, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018592405862948747, + "loss": 0.0684, + "step": 18142 + }, + { + "epoch": 0.1574899523441637, + "grad_norm": 0.490234375, + "learning_rate": 0.0018592246043011994, + "loss": 0.1216, + "step": 18143 + }, + { + "epoch": 0.15749863282436785, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018592086214772475, + "loss": 0.123, + "step": 18144 + }, + { + "epoch": 0.15750731330457202, + "grad_norm": 0.384765625, + "learning_rate": 0.0018591926378230359, + "loss": 0.083, + "step": 18145 + }, + { + "epoch": 0.15751599378477618, + "grad_norm": 0.12109375, + "learning_rate": 0.0018591766533385837, + "loss": 0.0933, + "step": 18146 + }, + { + "epoch": 0.15752467426498035, + "grad_norm": 0.6875, + "learning_rate": 0.0018591606680239065, + "loss": 0.123, + "step": 18147 + }, + { + "epoch": 0.1575333547451845, + "grad_norm": 0.1455078125, + "learning_rate": 0.001859144681879023, + "loss": 0.0991, + "step": 18148 + }, + { + "epoch": 0.15754203522538868, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018591286949039502, + "loss": 0.1021, + "step": 18149 + }, + { + "epoch": 0.15755071570559284, + "grad_norm": 0.65625, + "learning_rate": 0.0018591127070987055, + "loss": 0.1182, + "step": 18150 + }, + { + "epoch": 0.157559396185797, + "grad_norm": 0.33203125, + "learning_rate": 0.0018590967184633068, + "loss": 0.1777, + "step": 18151 + }, + { + "epoch": 0.15756807666600117, + "grad_norm": 0.28515625, + "learning_rate": 0.0018590807289977712, + "loss": 0.1133, + "step": 18152 + }, + { + "epoch": 0.15757675714620534, + "grad_norm": 0.72265625, + "learning_rate": 0.001859064738702116, + "loss": 0.1309, + "step": 18153 + }, + { + "epoch": 0.1575854376264095, + "grad_norm": 0.1640625, + "learning_rate": 0.0018590487475763596, + "loss": 0.1289, + "step": 18154 + }, + { + "epoch": 0.15759411810661367, + "grad_norm": 0.177734375, + "learning_rate": 0.0018590327556205185, + "loss": 0.1025, + "step": 18155 + }, + { + "epoch": 0.15760279858681783, + "grad_norm": 0.169921875, + "learning_rate": 0.0018590167628346106, + "loss": 0.1025, + "step": 18156 + }, + { + "epoch": 0.157611479067022, + "grad_norm": 0.1787109375, + "learning_rate": 0.0018590007692186534, + "loss": 0.125, + "step": 18157 + }, + { + "epoch": 0.15762015954722616, + "grad_norm": 0.3046875, + "learning_rate": 0.0018589847747726646, + "loss": 0.1133, + "step": 18158 + }, + { + "epoch": 0.15762884002743033, + "grad_norm": 0.12890625, + "learning_rate": 0.0018589687794966611, + "loss": 0.1025, + "step": 18159 + }, + { + "epoch": 0.1576375205076345, + "grad_norm": 0.224609375, + "learning_rate": 0.001858952783390661, + "loss": 0.1562, + "step": 18160 + }, + { + "epoch": 0.15764620098783866, + "grad_norm": 0.15234375, + "learning_rate": 0.0018589367864546816, + "loss": 0.1396, + "step": 18161 + }, + { + "epoch": 0.15765488146804282, + "grad_norm": 0.162109375, + "learning_rate": 0.0018589207886887402, + "loss": 0.0757, + "step": 18162 + }, + { + "epoch": 0.157663561948247, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018589047900928544, + "loss": 0.1279, + "step": 18163 + }, + { + "epoch": 0.15767224242845115, + "grad_norm": 0.58203125, + "learning_rate": 0.0018588887906670418, + "loss": 0.1074, + "step": 18164 + }, + { + "epoch": 0.15768092290865532, + "grad_norm": 0.42578125, + "learning_rate": 0.0018588727904113196, + "loss": 0.1133, + "step": 18165 + }, + { + "epoch": 0.15768960338885948, + "grad_norm": 0.109375, + "learning_rate": 0.0018588567893257058, + "loss": 0.1143, + "step": 18166 + }, + { + "epoch": 0.15769828386906365, + "grad_norm": 0.287109375, + "learning_rate": 0.0018588407874102176, + "loss": 0.1328, + "step": 18167 + }, + { + "epoch": 0.15770696434926781, + "grad_norm": 0.275390625, + "learning_rate": 0.0018588247846648722, + "loss": 0.0977, + "step": 18168 + }, + { + "epoch": 0.15771564482947198, + "grad_norm": 0.423828125, + "learning_rate": 0.0018588087810896877, + "loss": 0.105, + "step": 18169 + }, + { + "epoch": 0.15772432530967614, + "grad_norm": 0.435546875, + "learning_rate": 0.0018587927766846814, + "loss": 0.1104, + "step": 18170 + }, + { + "epoch": 0.1577330057898803, + "grad_norm": 0.31640625, + "learning_rate": 0.0018587767714498705, + "loss": 0.1426, + "step": 18171 + }, + { + "epoch": 0.15774168627008447, + "grad_norm": 0.87890625, + "learning_rate": 0.0018587607653852726, + "loss": 0.1133, + "step": 18172 + }, + { + "epoch": 0.15775036675028864, + "grad_norm": 0.484375, + "learning_rate": 0.0018587447584909056, + "loss": 0.1064, + "step": 18173 + }, + { + "epoch": 0.15775904723049278, + "grad_norm": 0.44921875, + "learning_rate": 0.0018587287507667867, + "loss": 0.1191, + "step": 18174 + }, + { + "epoch": 0.15776772771069694, + "grad_norm": 0.39453125, + "learning_rate": 0.0018587127422129335, + "loss": 0.0747, + "step": 18175 + }, + { + "epoch": 0.1577764081909011, + "grad_norm": 0.41796875, + "learning_rate": 0.0018586967328293633, + "loss": 0.0942, + "step": 18176 + }, + { + "epoch": 0.15778508867110527, + "grad_norm": 0.27734375, + "learning_rate": 0.0018586807226160938, + "loss": 0.0952, + "step": 18177 + }, + { + "epoch": 0.15779376915130944, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018586647115731425, + "loss": 0.0957, + "step": 18178 + }, + { + "epoch": 0.1578024496315136, + "grad_norm": 0.4296875, + "learning_rate": 0.0018586486997005268, + "loss": 0.1973, + "step": 18179 + }, + { + "epoch": 0.15781113011171777, + "grad_norm": 0.546875, + "learning_rate": 0.001858632686998264, + "loss": 0.0977, + "step": 18180 + }, + { + "epoch": 0.15781981059192193, + "grad_norm": 0.390625, + "learning_rate": 0.0018586166734663725, + "loss": 0.1118, + "step": 18181 + }, + { + "epoch": 0.1578284910721261, + "grad_norm": 0.75, + "learning_rate": 0.0018586006591048687, + "loss": 0.1445, + "step": 18182 + }, + { + "epoch": 0.15783717155233026, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018585846439137712, + "loss": 0.1406, + "step": 18183 + }, + { + "epoch": 0.15784585203253443, + "grad_norm": 0.11279296875, + "learning_rate": 0.0018585686278930961, + "loss": 0.1152, + "step": 18184 + }, + { + "epoch": 0.1578545325127386, + "grad_norm": 0.5234375, + "learning_rate": 0.0018585526110428627, + "loss": 0.1309, + "step": 18185 + }, + { + "epoch": 0.15786321299294276, + "grad_norm": 0.162109375, + "learning_rate": 0.0018585365933630868, + "loss": 0.1777, + "step": 18186 + }, + { + "epoch": 0.15787189347314692, + "grad_norm": 0.375, + "learning_rate": 0.001858520574853787, + "loss": 0.0996, + "step": 18187 + }, + { + "epoch": 0.1578805739533511, + "grad_norm": 0.87890625, + "learning_rate": 0.0018585045555149803, + "loss": 0.0859, + "step": 18188 + }, + { + "epoch": 0.15788925443355525, + "grad_norm": 0.388671875, + "learning_rate": 0.0018584885353466847, + "loss": 0.0986, + "step": 18189 + }, + { + "epoch": 0.15789793491375942, + "grad_norm": 0.134765625, + "learning_rate": 0.0018584725143489173, + "loss": 0.1396, + "step": 18190 + }, + { + "epoch": 0.15790661539396358, + "grad_norm": 0.123046875, + "learning_rate": 0.0018584564925216958, + "loss": 0.1406, + "step": 18191 + }, + { + "epoch": 0.15791529587416775, + "grad_norm": 0.384765625, + "learning_rate": 0.0018584404698650378, + "loss": 0.1436, + "step": 18192 + }, + { + "epoch": 0.15792397635437191, + "grad_norm": 0.11328125, + "learning_rate": 0.0018584244463789603, + "loss": 0.1348, + "step": 18193 + }, + { + "epoch": 0.15793265683457608, + "grad_norm": 0.150390625, + "learning_rate": 0.0018584084220634814, + "loss": 0.1504, + "step": 18194 + }, + { + "epoch": 0.15794133731478024, + "grad_norm": 0.51171875, + "learning_rate": 0.0018583923969186184, + "loss": 0.1133, + "step": 18195 + }, + { + "epoch": 0.1579500177949844, + "grad_norm": 0.34765625, + "learning_rate": 0.001858376370944389, + "loss": 0.1123, + "step": 18196 + }, + { + "epoch": 0.15795869827518857, + "grad_norm": 0.36328125, + "learning_rate": 0.0018583603441408106, + "loss": 0.0942, + "step": 18197 + }, + { + "epoch": 0.15796737875539274, + "grad_norm": 0.65625, + "learning_rate": 0.0018583443165079009, + "loss": 0.0913, + "step": 18198 + }, + { + "epoch": 0.1579760592355969, + "grad_norm": 1.2578125, + "learning_rate": 0.001858328288045677, + "loss": 0.1172, + "step": 18199 + }, + { + "epoch": 0.15798473971580107, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018583122587541563, + "loss": 0.1328, + "step": 18200 + }, + { + "epoch": 0.15799342019600524, + "grad_norm": 0.146484375, + "learning_rate": 0.0018582962286333572, + "loss": 0.1177, + "step": 18201 + }, + { + "epoch": 0.1580021006762094, + "grad_norm": 0.1044921875, + "learning_rate": 0.0018582801976832965, + "loss": 0.1016, + "step": 18202 + }, + { + "epoch": 0.15801078115641357, + "grad_norm": 0.13671875, + "learning_rate": 0.0018582641659039925, + "loss": 0.1309, + "step": 18203 + }, + { + "epoch": 0.15801946163661773, + "grad_norm": 0.1953125, + "learning_rate": 0.0018582481332954618, + "loss": 0.1108, + "step": 18204 + }, + { + "epoch": 0.1580281421168219, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018582320998577222, + "loss": 0.0854, + "step": 18205 + }, + { + "epoch": 0.15803682259702606, + "grad_norm": 0.6015625, + "learning_rate": 0.0018582160655907917, + "loss": 0.1299, + "step": 18206 + }, + { + "epoch": 0.15804550307723023, + "grad_norm": 0.087890625, + "learning_rate": 0.001858200030494687, + "loss": 0.1289, + "step": 18207 + }, + { + "epoch": 0.1580541835574344, + "grad_norm": 0.310546875, + "learning_rate": 0.0018581839945694266, + "loss": 0.1074, + "step": 18208 + }, + { + "epoch": 0.15806286403763856, + "grad_norm": 0.298828125, + "learning_rate": 0.0018581679578150274, + "loss": 0.1475, + "step": 18209 + }, + { + "epoch": 0.15807154451784272, + "grad_norm": 0.63671875, + "learning_rate": 0.0018581519202315068, + "loss": 0.1035, + "step": 18210 + }, + { + "epoch": 0.1580802249980469, + "grad_norm": 0.2109375, + "learning_rate": 0.0018581358818188834, + "loss": 0.1494, + "step": 18211 + }, + { + "epoch": 0.15808890547825105, + "grad_norm": 0.60546875, + "learning_rate": 0.0018581198425771735, + "loss": 0.1172, + "step": 18212 + }, + { + "epoch": 0.15809758595845522, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018581038025063952, + "loss": 0.1387, + "step": 18213 + }, + { + "epoch": 0.15810626643865938, + "grad_norm": 0.27734375, + "learning_rate": 0.0018580877616065658, + "loss": 0.0845, + "step": 18214 + }, + { + "epoch": 0.15811494691886355, + "grad_norm": 1.2109375, + "learning_rate": 0.0018580717198777032, + "loss": 0.1523, + "step": 18215 + }, + { + "epoch": 0.1581236273990677, + "grad_norm": 0.53515625, + "learning_rate": 0.001858055677319825, + "loss": 0.0723, + "step": 18216 + }, + { + "epoch": 0.15813230787927188, + "grad_norm": 0.212890625, + "learning_rate": 0.0018580396339329482, + "loss": 0.1128, + "step": 18217 + }, + { + "epoch": 0.15814098835947604, + "grad_norm": 0.66796875, + "learning_rate": 0.0018580235897170908, + "loss": 0.085, + "step": 18218 + }, + { + "epoch": 0.1581496688396802, + "grad_norm": 0.7265625, + "learning_rate": 0.00185800754467227, + "loss": 0.1504, + "step": 18219 + }, + { + "epoch": 0.15815834931988437, + "grad_norm": 0.3359375, + "learning_rate": 0.0018579914987985036, + "loss": 0.1235, + "step": 18220 + }, + { + "epoch": 0.15816702980008854, + "grad_norm": 0.2890625, + "learning_rate": 0.0018579754520958091, + "loss": 0.1045, + "step": 18221 + }, + { + "epoch": 0.1581757102802927, + "grad_norm": 0.263671875, + "learning_rate": 0.0018579594045642041, + "loss": 0.1289, + "step": 18222 + }, + { + "epoch": 0.15818439076049687, + "grad_norm": 0.1298828125, + "learning_rate": 0.001857943356203706, + "loss": 0.1309, + "step": 18223 + }, + { + "epoch": 0.15819307124070103, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018579273070143325, + "loss": 0.0845, + "step": 18224 + }, + { + "epoch": 0.1582017517209052, + "grad_norm": 0.30078125, + "learning_rate": 0.0018579112569961011, + "loss": 0.1084, + "step": 18225 + }, + { + "epoch": 0.15821043220110936, + "grad_norm": 0.8125, + "learning_rate": 0.0018578952061490293, + "loss": 0.1211, + "step": 18226 + }, + { + "epoch": 0.15821911268131353, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018578791544731346, + "loss": 0.0957, + "step": 18227 + }, + { + "epoch": 0.1582277931615177, + "grad_norm": 0.78515625, + "learning_rate": 0.001857863101968435, + "loss": 0.1094, + "step": 18228 + }, + { + "epoch": 0.15823647364172186, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018578470486349473, + "loss": 0.0796, + "step": 18229 + }, + { + "epoch": 0.15824515412192602, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018578309944726895, + "loss": 0.104, + "step": 18230 + }, + { + "epoch": 0.1582538346021302, + "grad_norm": 0.275390625, + "learning_rate": 0.001857814939481679, + "loss": 0.084, + "step": 18231 + }, + { + "epoch": 0.15826251508233435, + "grad_norm": 0.447265625, + "learning_rate": 0.0018577988836619337, + "loss": 0.1328, + "step": 18232 + }, + { + "epoch": 0.15827119556253852, + "grad_norm": 0.5703125, + "learning_rate": 0.0018577828270134708, + "loss": 0.0801, + "step": 18233 + }, + { + "epoch": 0.15827987604274268, + "grad_norm": 0.56640625, + "learning_rate": 0.001857766769536308, + "loss": 0.1172, + "step": 18234 + }, + { + "epoch": 0.15828855652294685, + "grad_norm": 0.1640625, + "learning_rate": 0.001857750711230463, + "loss": 0.1621, + "step": 18235 + }, + { + "epoch": 0.15829723700315101, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018577346520959532, + "loss": 0.1094, + "step": 18236 + }, + { + "epoch": 0.15830591748335518, + "grad_norm": 0.134765625, + "learning_rate": 0.0018577185921327963, + "loss": 0.105, + "step": 18237 + }, + { + "epoch": 0.15831459796355934, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018577025313410095, + "loss": 0.1172, + "step": 18238 + }, + { + "epoch": 0.1583232784437635, + "grad_norm": 0.265625, + "learning_rate": 0.0018576864697206106, + "loss": 0.1182, + "step": 18239 + }, + { + "epoch": 0.15833195892396768, + "grad_norm": 0.453125, + "learning_rate": 0.0018576704072716173, + "loss": 0.1426, + "step": 18240 + }, + { + "epoch": 0.15834063940417184, + "grad_norm": 0.490234375, + "learning_rate": 0.0018576543439940468, + "loss": 0.127, + "step": 18241 + }, + { + "epoch": 0.158349319884376, + "grad_norm": 0.240234375, + "learning_rate": 0.001857638279887917, + "loss": 0.0933, + "step": 18242 + }, + { + "epoch": 0.15835800036458017, + "grad_norm": 1.4296875, + "learning_rate": 0.0018576222149532453, + "loss": 0.1289, + "step": 18243 + }, + { + "epoch": 0.15836668084478434, + "grad_norm": 0.69921875, + "learning_rate": 0.0018576061491900496, + "loss": 0.1025, + "step": 18244 + }, + { + "epoch": 0.1583753613249885, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018575900825983469, + "loss": 0.1211, + "step": 18245 + }, + { + "epoch": 0.15838404180519267, + "grad_norm": 0.11328125, + "learning_rate": 0.001857574015178155, + "loss": 0.0874, + "step": 18246 + }, + { + "epoch": 0.15839272228539683, + "grad_norm": 0.287109375, + "learning_rate": 0.0018575579469294915, + "loss": 0.1152, + "step": 18247 + }, + { + "epoch": 0.158401402765601, + "grad_norm": 0.30859375, + "learning_rate": 0.0018575418778523744, + "loss": 0.1221, + "step": 18248 + }, + { + "epoch": 0.15841008324580516, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018575258079468206, + "loss": 0.1309, + "step": 18249 + }, + { + "epoch": 0.15841876372600933, + "grad_norm": 0.66796875, + "learning_rate": 0.0018575097372128483, + "loss": 0.1582, + "step": 18250 + }, + { + "epoch": 0.1584274442062135, + "grad_norm": 0.78515625, + "learning_rate": 0.0018574936656504744, + "loss": 0.1592, + "step": 18251 + }, + { + "epoch": 0.15843612468641766, + "grad_norm": 0.439453125, + "learning_rate": 0.001857477593259717, + "loss": 0.1221, + "step": 18252 + }, + { + "epoch": 0.15844480516662182, + "grad_norm": 0.232421875, + "learning_rate": 0.0018574615200405934, + "loss": 0.083, + "step": 18253 + }, + { + "epoch": 0.158453485646826, + "grad_norm": 0.390625, + "learning_rate": 0.001857445445993121, + "loss": 0.1328, + "step": 18254 + }, + { + "epoch": 0.15846216612703015, + "grad_norm": 0.359375, + "learning_rate": 0.001857429371117318, + "loss": 0.1084, + "step": 18255 + }, + { + "epoch": 0.15847084660723432, + "grad_norm": 0.267578125, + "learning_rate": 0.0018574132954132015, + "loss": 0.1211, + "step": 18256 + }, + { + "epoch": 0.15847952708743848, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018573972188807891, + "loss": 0.1016, + "step": 18257 + }, + { + "epoch": 0.15848820756764265, + "grad_norm": 0.353515625, + "learning_rate": 0.0018573811415200986, + "loss": 0.1074, + "step": 18258 + }, + { + "epoch": 0.1584968880478468, + "grad_norm": 0.341796875, + "learning_rate": 0.0018573650633311475, + "loss": 0.1177, + "step": 18259 + }, + { + "epoch": 0.15850556852805098, + "grad_norm": 0.185546875, + "learning_rate": 0.0018573489843139532, + "loss": 0.0889, + "step": 18260 + }, + { + "epoch": 0.15851424900825514, + "grad_norm": 0.20703125, + "learning_rate": 0.0018573329044685334, + "loss": 0.127, + "step": 18261 + }, + { + "epoch": 0.1585229294884593, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018573168237949061, + "loss": 0.1025, + "step": 18262 + }, + { + "epoch": 0.15853160996866347, + "grad_norm": 0.27734375, + "learning_rate": 0.0018573007422930884, + "loss": 0.1001, + "step": 18263 + }, + { + "epoch": 0.15854029044886764, + "grad_norm": 0.28125, + "learning_rate": 0.0018572846599630977, + "loss": 0.0796, + "step": 18264 + }, + { + "epoch": 0.1585489709290718, + "grad_norm": 0.10693359375, + "learning_rate": 0.001857268576804952, + "loss": 0.1152, + "step": 18265 + }, + { + "epoch": 0.15855765140927597, + "grad_norm": 0.4140625, + "learning_rate": 0.0018572524928186687, + "loss": 0.1074, + "step": 18266 + }, + { + "epoch": 0.15856633188948013, + "grad_norm": 0.2451171875, + "learning_rate": 0.0018572364080042655, + "loss": 0.1118, + "step": 18267 + }, + { + "epoch": 0.1585750123696843, + "grad_norm": 0.45703125, + "learning_rate": 0.00185722032236176, + "loss": 0.1123, + "step": 18268 + }, + { + "epoch": 0.15858369284988846, + "grad_norm": 0.10546875, + "learning_rate": 0.0018572042358911695, + "loss": 0.1113, + "step": 18269 + }, + { + "epoch": 0.15859237333009263, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018571881485925123, + "loss": 0.0669, + "step": 18270 + }, + { + "epoch": 0.1586010538102968, + "grad_norm": 0.73046875, + "learning_rate": 0.0018571720604658051, + "loss": 0.1084, + "step": 18271 + }, + { + "epoch": 0.15860973429050096, + "grad_norm": 0.390625, + "learning_rate": 0.001857155971511066, + "loss": 0.084, + "step": 18272 + }, + { + "epoch": 0.15861841477070512, + "grad_norm": 0.7890625, + "learning_rate": 0.0018571398817283126, + "loss": 0.1582, + "step": 18273 + }, + { + "epoch": 0.1586270952509093, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018571237911175623, + "loss": 0.1084, + "step": 18274 + }, + { + "epoch": 0.15863577573111345, + "grad_norm": 0.443359375, + "learning_rate": 0.0018571076996788327, + "loss": 0.1123, + "step": 18275 + }, + { + "epoch": 0.15864445621131762, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018570916074121416, + "loss": 0.1211, + "step": 18276 + }, + { + "epoch": 0.15865313669152178, + "grad_norm": 1.0546875, + "learning_rate": 0.0018570755143175067, + "loss": 0.1133, + "step": 18277 + }, + { + "epoch": 0.15866181717172595, + "grad_norm": 0.40625, + "learning_rate": 0.0018570594203949453, + "loss": 0.1621, + "step": 18278 + }, + { + "epoch": 0.15867049765193011, + "grad_norm": 0.216796875, + "learning_rate": 0.001857043325644475, + "loss": 0.0918, + "step": 18279 + }, + { + "epoch": 0.15867917813213428, + "grad_norm": 0.37890625, + "learning_rate": 0.0018570272300661133, + "loss": 0.1855, + "step": 18280 + }, + { + "epoch": 0.15868785861233844, + "grad_norm": 0.53515625, + "learning_rate": 0.0018570111336598783, + "loss": 0.1289, + "step": 18281 + }, + { + "epoch": 0.1586965390925426, + "grad_norm": 0.12890625, + "learning_rate": 0.0018569950364257873, + "loss": 0.1201, + "step": 18282 + }, + { + "epoch": 0.15870521957274678, + "grad_norm": 0.4765625, + "learning_rate": 0.0018569789383638578, + "loss": 0.1377, + "step": 18283 + }, + { + "epoch": 0.15871390005295094, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018569628394741074, + "loss": 0.0879, + "step": 18284 + }, + { + "epoch": 0.1587225805331551, + "grad_norm": 0.60546875, + "learning_rate": 0.0018569467397565541, + "loss": 0.1025, + "step": 18285 + }, + { + "epoch": 0.15873126101335927, + "grad_norm": 0.8828125, + "learning_rate": 0.001856930639211215, + "loss": 0.1523, + "step": 18286 + }, + { + "epoch": 0.15873994149356344, + "grad_norm": 0.1201171875, + "learning_rate": 0.0018569145378381081, + "loss": 0.0977, + "step": 18287 + }, + { + "epoch": 0.1587486219737676, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018568984356372505, + "loss": 0.1211, + "step": 18288 + }, + { + "epoch": 0.15875730245397177, + "grad_norm": 0.349609375, + "learning_rate": 0.0018568823326086603, + "loss": 0.1104, + "step": 18289 + }, + { + "epoch": 0.15876598293417593, + "grad_norm": 0.205078125, + "learning_rate": 0.001856866228752355, + "loss": 0.1514, + "step": 18290 + }, + { + "epoch": 0.1587746634143801, + "grad_norm": 0.193359375, + "learning_rate": 0.0018568501240683521, + "loss": 0.0869, + "step": 18291 + }, + { + "epoch": 0.15878334389458426, + "grad_norm": 0.08642578125, + "learning_rate": 0.0018568340185566696, + "loss": 0.1016, + "step": 18292 + }, + { + "epoch": 0.15879202437478843, + "grad_norm": 0.162109375, + "learning_rate": 0.0018568179122173242, + "loss": 0.0957, + "step": 18293 + }, + { + "epoch": 0.1588007048549926, + "grad_norm": 0.421875, + "learning_rate": 0.0018568018050503345, + "loss": 0.127, + "step": 18294 + }, + { + "epoch": 0.15880938533519676, + "grad_norm": 0.859375, + "learning_rate": 0.0018567856970557177, + "loss": 0.1367, + "step": 18295 + }, + { + "epoch": 0.15881806581540092, + "grad_norm": 0.310546875, + "learning_rate": 0.0018567695882334913, + "loss": 0.1069, + "step": 18296 + }, + { + "epoch": 0.15882674629560506, + "grad_norm": 0.1015625, + "learning_rate": 0.001856753478583673, + "loss": 0.1162, + "step": 18297 + }, + { + "epoch": 0.15883542677580922, + "grad_norm": 0.294921875, + "learning_rate": 0.001856737368106281, + "loss": 0.0967, + "step": 18298 + }, + { + "epoch": 0.1588441072560134, + "grad_norm": 0.28125, + "learning_rate": 0.001856721256801332, + "loss": 0.1309, + "step": 18299 + }, + { + "epoch": 0.15885278773621755, + "grad_norm": 0.40625, + "learning_rate": 0.001856705144668844, + "loss": 0.1245, + "step": 18300 + }, + { + "epoch": 0.15886146821642172, + "grad_norm": 0.3046875, + "learning_rate": 0.0018566890317088343, + "loss": 0.1328, + "step": 18301 + }, + { + "epoch": 0.15887014869662588, + "grad_norm": 0.1484375, + "learning_rate": 0.0018566729179213214, + "loss": 0.0928, + "step": 18302 + }, + { + "epoch": 0.15887882917683005, + "grad_norm": 0.796875, + "learning_rate": 0.001856656803306322, + "loss": 0.1021, + "step": 18303 + }, + { + "epoch": 0.15888750965703421, + "grad_norm": 0.57421875, + "learning_rate": 0.001856640687863854, + "loss": 0.0957, + "step": 18304 + }, + { + "epoch": 0.15889619013723838, + "grad_norm": 0.1796875, + "learning_rate": 0.0018566245715939355, + "loss": 0.0825, + "step": 18305 + }, + { + "epoch": 0.15890487061744255, + "grad_norm": 0.06103515625, + "learning_rate": 0.0018566084544965836, + "loss": 0.0918, + "step": 18306 + }, + { + "epoch": 0.1589135510976467, + "grad_norm": 0.1318359375, + "learning_rate": 0.001856592336571816, + "loss": 0.1074, + "step": 18307 + }, + { + "epoch": 0.15892223157785088, + "grad_norm": 0.09912109375, + "learning_rate": 0.0018565762178196503, + "loss": 0.1167, + "step": 18308 + }, + { + "epoch": 0.15893091205805504, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018565600982401046, + "loss": 0.1475, + "step": 18309 + }, + { + "epoch": 0.1589395925382592, + "grad_norm": 0.146484375, + "learning_rate": 0.0018565439778331957, + "loss": 0.1191, + "step": 18310 + }, + { + "epoch": 0.15894827301846337, + "grad_norm": 0.5625, + "learning_rate": 0.001856527856598942, + "loss": 0.1367, + "step": 18311 + }, + { + "epoch": 0.15895695349866754, + "grad_norm": 1.2734375, + "learning_rate": 0.0018565117345373605, + "loss": 0.1182, + "step": 18312 + }, + { + "epoch": 0.1589656339788717, + "grad_norm": 0.53125, + "learning_rate": 0.0018564956116484692, + "loss": 0.1021, + "step": 18313 + }, + { + "epoch": 0.15897431445907587, + "grad_norm": 0.203125, + "learning_rate": 0.0018564794879322858, + "loss": 0.1357, + "step": 18314 + }, + { + "epoch": 0.15898299493928003, + "grad_norm": 0.58984375, + "learning_rate": 0.0018564633633888275, + "loss": 0.0986, + "step": 18315 + }, + { + "epoch": 0.1589916754194842, + "grad_norm": 0.89453125, + "learning_rate": 0.0018564472380181126, + "loss": 0.1118, + "step": 18316 + }, + { + "epoch": 0.15900035589968836, + "grad_norm": 0.171875, + "learning_rate": 0.0018564311118201581, + "loss": 0.0957, + "step": 18317 + }, + { + "epoch": 0.15900903637989253, + "grad_norm": 0.31640625, + "learning_rate": 0.0018564149847949823, + "loss": 0.1133, + "step": 18318 + }, + { + "epoch": 0.1590177168600967, + "grad_norm": 0.57421875, + "learning_rate": 0.0018563988569426022, + "loss": 0.0952, + "step": 18319 + }, + { + "epoch": 0.15902639734030086, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018563827282630356, + "loss": 0.1201, + "step": 18320 + }, + { + "epoch": 0.15903507782050502, + "grad_norm": 0.453125, + "learning_rate": 0.0018563665987563, + "loss": 0.1108, + "step": 18321 + }, + { + "epoch": 0.1590437583007092, + "grad_norm": 0.369140625, + "learning_rate": 0.0018563504684224138, + "loss": 0.1094, + "step": 18322 + }, + { + "epoch": 0.15905243878091335, + "grad_norm": 0.216796875, + "learning_rate": 0.0018563343372613936, + "loss": 0.1191, + "step": 18323 + }, + { + "epoch": 0.15906111926111752, + "grad_norm": 0.2890625, + "learning_rate": 0.0018563182052732578, + "loss": 0.1289, + "step": 18324 + }, + { + "epoch": 0.15906979974132168, + "grad_norm": 0.341796875, + "learning_rate": 0.001856302072458024, + "loss": 0.123, + "step": 18325 + }, + { + "epoch": 0.15907848022152585, + "grad_norm": 0.3359375, + "learning_rate": 0.001856285938815709, + "loss": 0.1406, + "step": 18326 + }, + { + "epoch": 0.15908716070173, + "grad_norm": 0.357421875, + "learning_rate": 0.0018562698043463313, + "loss": 0.1089, + "step": 18327 + }, + { + "epoch": 0.15909584118193418, + "grad_norm": 0.455078125, + "learning_rate": 0.0018562536690499086, + "loss": 0.1357, + "step": 18328 + }, + { + "epoch": 0.15910452166213834, + "grad_norm": 0.515625, + "learning_rate": 0.0018562375329264581, + "loss": 0.085, + "step": 18329 + }, + { + "epoch": 0.1591132021423425, + "grad_norm": 0.1015625, + "learning_rate": 0.0018562213959759973, + "loss": 0.1123, + "step": 18330 + }, + { + "epoch": 0.15912188262254667, + "grad_norm": 0.8125, + "learning_rate": 0.0018562052581985447, + "loss": 0.103, + "step": 18331 + }, + { + "epoch": 0.15913056310275084, + "grad_norm": 0.34375, + "learning_rate": 0.001856189119594117, + "loss": 0.1475, + "step": 18332 + }, + { + "epoch": 0.159139243582955, + "grad_norm": 0.265625, + "learning_rate": 0.0018561729801627321, + "loss": 0.0977, + "step": 18333 + }, + { + "epoch": 0.15914792406315917, + "grad_norm": 0.8046875, + "learning_rate": 0.0018561568399044083, + "loss": 0.1387, + "step": 18334 + }, + { + "epoch": 0.15915660454336333, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018561406988191623, + "loss": 0.1035, + "step": 18335 + }, + { + "epoch": 0.1591652850235675, + "grad_norm": 1.328125, + "learning_rate": 0.0018561245569070123, + "loss": 0.3164, + "step": 18336 + }, + { + "epoch": 0.15917396550377166, + "grad_norm": 0.21484375, + "learning_rate": 0.0018561084141679757, + "loss": 0.0913, + "step": 18337 + }, + { + "epoch": 0.15918264598397583, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018560922706020706, + "loss": 0.1504, + "step": 18338 + }, + { + "epoch": 0.15919132646418, + "grad_norm": 0.17578125, + "learning_rate": 0.0018560761262093142, + "loss": 0.1045, + "step": 18339 + }, + { + "epoch": 0.15920000694438416, + "grad_norm": 2.3125, + "learning_rate": 0.0018560599809897245, + "loss": 0.1543, + "step": 18340 + }, + { + "epoch": 0.15920868742458832, + "grad_norm": 0.154296875, + "learning_rate": 0.0018560438349433186, + "loss": 0.1445, + "step": 18341 + }, + { + "epoch": 0.1592173679047925, + "grad_norm": 0.134765625, + "learning_rate": 0.001856027688070115, + "loss": 0.1504, + "step": 18342 + }, + { + "epoch": 0.15922604838499665, + "grad_norm": 0.353515625, + "learning_rate": 0.0018560115403701304, + "loss": 0.0977, + "step": 18343 + }, + { + "epoch": 0.15923472886520082, + "grad_norm": 0.6328125, + "learning_rate": 0.001855995391843383, + "loss": 0.1514, + "step": 18344 + }, + { + "epoch": 0.15924340934540498, + "grad_norm": 0.404296875, + "learning_rate": 0.0018559792424898908, + "loss": 0.1113, + "step": 18345 + }, + { + "epoch": 0.15925208982560915, + "grad_norm": 0.07421875, + "learning_rate": 0.001855963092309671, + "loss": 0.1089, + "step": 18346 + }, + { + "epoch": 0.15926077030581332, + "grad_norm": 0.98046875, + "learning_rate": 0.001855946941302741, + "loss": 0.0957, + "step": 18347 + }, + { + "epoch": 0.15926945078601748, + "grad_norm": 0.84765625, + "learning_rate": 0.0018559307894691185, + "loss": 0.1445, + "step": 18348 + }, + { + "epoch": 0.15927813126622165, + "grad_norm": 0.0791015625, + "learning_rate": 0.0018559146368088222, + "loss": 0.124, + "step": 18349 + }, + { + "epoch": 0.1592868117464258, + "grad_norm": 0.2392578125, + "learning_rate": 0.0018558984833218686, + "loss": 0.1094, + "step": 18350 + }, + { + "epoch": 0.15929549222662998, + "grad_norm": 0.1025390625, + "learning_rate": 0.0018558823290082758, + "loss": 0.1172, + "step": 18351 + }, + { + "epoch": 0.15930417270683414, + "grad_norm": 0.392578125, + "learning_rate": 0.0018558661738680615, + "loss": 0.0894, + "step": 18352 + }, + { + "epoch": 0.1593128531870383, + "grad_norm": 0.51953125, + "learning_rate": 0.0018558500179012435, + "loss": 0.1172, + "step": 18353 + }, + { + "epoch": 0.15932153366724247, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018558338611078392, + "loss": 0.1514, + "step": 18354 + }, + { + "epoch": 0.15933021414744664, + "grad_norm": 0.263671875, + "learning_rate": 0.0018558177034878659, + "loss": 0.1543, + "step": 18355 + }, + { + "epoch": 0.1593388946276508, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018558015450413422, + "loss": 0.1357, + "step": 18356 + }, + { + "epoch": 0.15934757510785497, + "grad_norm": 0.60546875, + "learning_rate": 0.001855785385768285, + "loss": 0.1357, + "step": 18357 + }, + { + "epoch": 0.15935625558805913, + "grad_norm": 0.07958984375, + "learning_rate": 0.0018557692256687124, + "loss": 0.123, + "step": 18358 + }, + { + "epoch": 0.1593649360682633, + "grad_norm": 0.11083984375, + "learning_rate": 0.001855753064742642, + "loss": 0.1094, + "step": 18359 + }, + { + "epoch": 0.15937361654846746, + "grad_norm": 0.35546875, + "learning_rate": 0.0018557369029900913, + "loss": 0.127, + "step": 18360 + }, + { + "epoch": 0.15938229702867163, + "grad_norm": 0.291015625, + "learning_rate": 0.0018557207404110779, + "loss": 0.1162, + "step": 18361 + }, + { + "epoch": 0.1593909775088758, + "grad_norm": 0.390625, + "learning_rate": 0.0018557045770056198, + "loss": 0.1123, + "step": 18362 + }, + { + "epoch": 0.15939965798907996, + "grad_norm": 0.453125, + "learning_rate": 0.0018556884127737345, + "loss": 0.0967, + "step": 18363 + }, + { + "epoch": 0.15940833846928412, + "grad_norm": 0.2060546875, + "learning_rate": 0.00185567224771544, + "loss": 0.1582, + "step": 18364 + }, + { + "epoch": 0.1594170189494883, + "grad_norm": 0.296875, + "learning_rate": 0.0018556560818307536, + "loss": 0.1147, + "step": 18365 + }, + { + "epoch": 0.15942569942969245, + "grad_norm": 0.173828125, + "learning_rate": 0.0018556399151196928, + "loss": 0.1143, + "step": 18366 + }, + { + "epoch": 0.15943437990989662, + "grad_norm": 0.423828125, + "learning_rate": 0.0018556237475822754, + "loss": 0.1074, + "step": 18367 + }, + { + "epoch": 0.15944306039010078, + "grad_norm": 0.2734375, + "learning_rate": 0.0018556075792185197, + "loss": 0.1113, + "step": 18368 + }, + { + "epoch": 0.15945174087030495, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018555914100284428, + "loss": 0.0996, + "step": 18369 + }, + { + "epoch": 0.1594604213505091, + "grad_norm": 0.5390625, + "learning_rate": 0.0018555752400120622, + "loss": 0.1011, + "step": 18370 + }, + { + "epoch": 0.15946910183071328, + "grad_norm": 0.232421875, + "learning_rate": 0.001855559069169396, + "loss": 0.1094, + "step": 18371 + }, + { + "epoch": 0.15947778231091744, + "grad_norm": 0.10302734375, + "learning_rate": 0.0018555428975004617, + "loss": 0.1055, + "step": 18372 + }, + { + "epoch": 0.1594864627911216, + "grad_norm": 0.154296875, + "learning_rate": 0.0018555267250052775, + "loss": 0.126, + "step": 18373 + }, + { + "epoch": 0.15949514327132577, + "grad_norm": 0.671875, + "learning_rate": 0.0018555105516838602, + "loss": 0.1631, + "step": 18374 + }, + { + "epoch": 0.15950382375152994, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018554943775362278, + "loss": 0.125, + "step": 18375 + }, + { + "epoch": 0.1595125042317341, + "grad_norm": 0.51171875, + "learning_rate": 0.0018554782025623983, + "loss": 0.124, + "step": 18376 + }, + { + "epoch": 0.15952118471193827, + "grad_norm": 0.83203125, + "learning_rate": 0.0018554620267623892, + "loss": 0.1001, + "step": 18377 + }, + { + "epoch": 0.15952986519214243, + "grad_norm": 0.458984375, + "learning_rate": 0.0018554458501362182, + "loss": 0.168, + "step": 18378 + }, + { + "epoch": 0.1595385456723466, + "grad_norm": 0.6171875, + "learning_rate": 0.0018554296726839027, + "loss": 0.1455, + "step": 18379 + }, + { + "epoch": 0.15954722615255076, + "grad_norm": 0.494140625, + "learning_rate": 0.001855413494405461, + "loss": 0.0977, + "step": 18380 + }, + { + "epoch": 0.15955590663275493, + "grad_norm": 0.5625, + "learning_rate": 0.0018553973153009103, + "loss": 0.0903, + "step": 18381 + }, + { + "epoch": 0.1595645871129591, + "grad_norm": 0.2890625, + "learning_rate": 0.0018553811353702685, + "loss": 0.1289, + "step": 18382 + }, + { + "epoch": 0.15957326759316326, + "grad_norm": 0.193359375, + "learning_rate": 0.001855364954613553, + "loss": 0.1025, + "step": 18383 + }, + { + "epoch": 0.15958194807336742, + "grad_norm": 0.234375, + "learning_rate": 0.0018553487730307823, + "loss": 0.1504, + "step": 18384 + }, + { + "epoch": 0.1595906285535716, + "grad_norm": 0.291015625, + "learning_rate": 0.001855332590621973, + "loss": 0.1016, + "step": 18385 + }, + { + "epoch": 0.15959930903377575, + "grad_norm": 0.236328125, + "learning_rate": 0.0018553164073871434, + "loss": 0.1348, + "step": 18386 + }, + { + "epoch": 0.15960798951397992, + "grad_norm": 0.447265625, + "learning_rate": 0.0018553002233263115, + "loss": 0.1914, + "step": 18387 + }, + { + "epoch": 0.15961666999418408, + "grad_norm": 1.078125, + "learning_rate": 0.0018552840384394944, + "loss": 0.1177, + "step": 18388 + }, + { + "epoch": 0.15962535047438825, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018552678527267098, + "loss": 0.104, + "step": 18389 + }, + { + "epoch": 0.15963403095459242, + "grad_norm": 0.283203125, + "learning_rate": 0.0018552516661879757, + "loss": 0.1387, + "step": 18390 + }, + { + "epoch": 0.15964271143479658, + "grad_norm": 0.796875, + "learning_rate": 0.00185523547882331, + "loss": 0.0942, + "step": 18391 + }, + { + "epoch": 0.15965139191500075, + "grad_norm": 0.125, + "learning_rate": 0.0018552192906327299, + "loss": 0.0972, + "step": 18392 + }, + { + "epoch": 0.1596600723952049, + "grad_norm": 0.255859375, + "learning_rate": 0.0018552031016162533, + "loss": 0.0776, + "step": 18393 + }, + { + "epoch": 0.15966875287540908, + "grad_norm": 0.140625, + "learning_rate": 0.0018551869117738981, + "loss": 0.127, + "step": 18394 + }, + { + "epoch": 0.15967743335561324, + "grad_norm": 0.15625, + "learning_rate": 0.0018551707211056818, + "loss": 0.1104, + "step": 18395 + }, + { + "epoch": 0.1596861138358174, + "grad_norm": 0.310546875, + "learning_rate": 0.001855154529611622, + "loss": 0.125, + "step": 18396 + }, + { + "epoch": 0.15969479431602157, + "grad_norm": 0.162109375, + "learning_rate": 0.0018551383372917367, + "loss": 0.1143, + "step": 18397 + }, + { + "epoch": 0.15970347479622574, + "grad_norm": 0.3359375, + "learning_rate": 0.0018551221441460435, + "loss": 0.1123, + "step": 18398 + }, + { + "epoch": 0.1597121552764299, + "grad_norm": 0.0732421875, + "learning_rate": 0.00185510595017456, + "loss": 0.0894, + "step": 18399 + }, + { + "epoch": 0.15972083575663407, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018550897553773038, + "loss": 0.0918, + "step": 18400 + }, + { + "epoch": 0.15972951623683823, + "grad_norm": 0.111328125, + "learning_rate": 0.001855073559754293, + "loss": 0.0947, + "step": 18401 + }, + { + "epoch": 0.1597381967170424, + "grad_norm": 0.06298828125, + "learning_rate": 0.0018550573633055452, + "loss": 0.0864, + "step": 18402 + }, + { + "epoch": 0.15974687719724656, + "grad_norm": 0.138671875, + "learning_rate": 0.0018550411660310778, + "loss": 0.1172, + "step": 18403 + }, + { + "epoch": 0.15975555767745073, + "grad_norm": 0.10595703125, + "learning_rate": 0.001855024967930909, + "loss": 0.1416, + "step": 18404 + }, + { + "epoch": 0.1597642381576549, + "grad_norm": 0.7890625, + "learning_rate": 0.0018550087690050558, + "loss": 0.0986, + "step": 18405 + }, + { + "epoch": 0.15977291863785906, + "grad_norm": 0.171875, + "learning_rate": 0.0018549925692535366, + "loss": 0.1387, + "step": 18406 + }, + { + "epoch": 0.15978159911806322, + "grad_norm": 0.265625, + "learning_rate": 0.0018549763686763689, + "loss": 0.0967, + "step": 18407 + }, + { + "epoch": 0.1597902795982674, + "grad_norm": 0.46484375, + "learning_rate": 0.0018549601672735703, + "loss": 0.0801, + "step": 18408 + }, + { + "epoch": 0.15979896007847155, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018549439650451588, + "loss": 0.1035, + "step": 18409 + }, + { + "epoch": 0.15980764055867572, + "grad_norm": 0.15234375, + "learning_rate": 0.0018549277619911515, + "loss": 0.0796, + "step": 18410 + }, + { + "epoch": 0.15981632103887988, + "grad_norm": 0.64453125, + "learning_rate": 0.001854911558111567, + "loss": 0.124, + "step": 18411 + }, + { + "epoch": 0.15982500151908405, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018548953534064223, + "loss": 0.1035, + "step": 18412 + }, + { + "epoch": 0.1598336819992882, + "grad_norm": 0.158203125, + "learning_rate": 0.0018548791478757353, + "loss": 0.1025, + "step": 18413 + }, + { + "epoch": 0.15984236247949238, + "grad_norm": 0.283203125, + "learning_rate": 0.0018548629415195241, + "loss": 0.1289, + "step": 18414 + }, + { + "epoch": 0.15985104295969654, + "grad_norm": 0.43359375, + "learning_rate": 0.0018548467343378061, + "loss": 0.1187, + "step": 18415 + }, + { + "epoch": 0.1598597234399007, + "grad_norm": 0.126953125, + "learning_rate": 0.001854830526330599, + "loss": 0.1406, + "step": 18416 + }, + { + "epoch": 0.15986840392010487, + "grad_norm": 1.0546875, + "learning_rate": 0.0018548143174979203, + "loss": 0.1045, + "step": 18417 + }, + { + "epoch": 0.15987708440030904, + "grad_norm": 0.1162109375, + "learning_rate": 0.0018547981078397885, + "loss": 0.1426, + "step": 18418 + }, + { + "epoch": 0.1598857648805132, + "grad_norm": 0.279296875, + "learning_rate": 0.0018547818973562206, + "loss": 0.1113, + "step": 18419 + }, + { + "epoch": 0.15989444536071734, + "grad_norm": 0.189453125, + "learning_rate": 0.0018547656860472344, + "loss": 0.1289, + "step": 18420 + }, + { + "epoch": 0.1599031258409215, + "grad_norm": 0.1923828125, + "learning_rate": 0.001854749473912848, + "loss": 0.1416, + "step": 18421 + }, + { + "epoch": 0.15991180632112567, + "grad_norm": 0.107421875, + "learning_rate": 0.0018547332609530787, + "loss": 0.0752, + "step": 18422 + }, + { + "epoch": 0.15992048680132984, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018547170471679443, + "loss": 0.123, + "step": 18423 + }, + { + "epoch": 0.159929167281534, + "grad_norm": 0.341796875, + "learning_rate": 0.0018547008325574632, + "loss": 0.1104, + "step": 18424 + }, + { + "epoch": 0.15993784776173817, + "grad_norm": 0.1953125, + "learning_rate": 0.0018546846171216523, + "loss": 0.1113, + "step": 18425 + }, + { + "epoch": 0.15994652824194233, + "grad_norm": 0.11962890625, + "learning_rate": 0.0018546684008605295, + "loss": 0.1104, + "step": 18426 + }, + { + "epoch": 0.1599552087221465, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018546521837741132, + "loss": 0.0884, + "step": 18427 + }, + { + "epoch": 0.15996388920235066, + "grad_norm": 0.197265625, + "learning_rate": 0.00185463596586242, + "loss": 0.1631, + "step": 18428 + }, + { + "epoch": 0.15997256968255483, + "grad_norm": 0.32421875, + "learning_rate": 0.0018546197471254688, + "loss": 0.127, + "step": 18429 + }, + { + "epoch": 0.159981250162759, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018546035275632763, + "loss": 0.1045, + "step": 18430 + }, + { + "epoch": 0.15998993064296316, + "grad_norm": 0.232421875, + "learning_rate": 0.001854587307175861, + "loss": 0.0991, + "step": 18431 + }, + { + "epoch": 0.15999861112316732, + "grad_norm": 0.337890625, + "learning_rate": 0.0018545710859632404, + "loss": 0.1289, + "step": 18432 + }, + { + "epoch": 0.1600072916033715, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018545548639254319, + "loss": 0.1484, + "step": 18433 + }, + { + "epoch": 0.16001597208357565, + "grad_norm": 0.3828125, + "learning_rate": 0.0018545386410624537, + "loss": 0.1128, + "step": 18434 + }, + { + "epoch": 0.16002465256377982, + "grad_norm": 0.400390625, + "learning_rate": 0.0018545224173743234, + "loss": 0.1162, + "step": 18435 + }, + { + "epoch": 0.16003333304398398, + "grad_norm": 0.443359375, + "learning_rate": 0.0018545061928610588, + "loss": 0.1494, + "step": 18436 + }, + { + "epoch": 0.16004201352418815, + "grad_norm": 0.1640625, + "learning_rate": 0.0018544899675226774, + "loss": 0.1045, + "step": 18437 + }, + { + "epoch": 0.1600506940043923, + "grad_norm": 0.365234375, + "learning_rate": 0.0018544737413591973, + "loss": 0.127, + "step": 18438 + }, + { + "epoch": 0.16005937448459648, + "grad_norm": 0.216796875, + "learning_rate": 0.0018544575143706358, + "loss": 0.1064, + "step": 18439 + }, + { + "epoch": 0.16006805496480064, + "grad_norm": 0.34375, + "learning_rate": 0.001854441286557011, + "loss": 0.1533, + "step": 18440 + }, + { + "epoch": 0.1600767354450048, + "grad_norm": 0.353515625, + "learning_rate": 0.0018544250579183405, + "loss": 0.0864, + "step": 18441 + }, + { + "epoch": 0.16008541592520897, + "grad_norm": 0.4375, + "learning_rate": 0.0018544088284546422, + "loss": 0.0947, + "step": 18442 + }, + { + "epoch": 0.16009409640541314, + "grad_norm": 0.1484375, + "learning_rate": 0.0018543925981659337, + "loss": 0.0928, + "step": 18443 + }, + { + "epoch": 0.1601027768856173, + "grad_norm": 0.357421875, + "learning_rate": 0.001854376367052233, + "loss": 0.1152, + "step": 18444 + }, + { + "epoch": 0.16011145736582147, + "grad_norm": 0.61328125, + "learning_rate": 0.0018543601351135575, + "loss": 0.1113, + "step": 18445 + }, + { + "epoch": 0.16012013784602563, + "grad_norm": 0.2421875, + "learning_rate": 0.001854343902349925, + "loss": 0.1055, + "step": 18446 + }, + { + "epoch": 0.1601288183262298, + "grad_norm": 0.53125, + "learning_rate": 0.0018543276687613534, + "loss": 0.1025, + "step": 18447 + }, + { + "epoch": 0.16013749880643396, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018543114343478602, + "loss": 0.1123, + "step": 18448 + }, + { + "epoch": 0.16014617928663813, + "grad_norm": 0.1689453125, + "learning_rate": 0.0018542951991094637, + "loss": 0.1133, + "step": 18449 + }, + { + "epoch": 0.1601548597668423, + "grad_norm": 0.328125, + "learning_rate": 0.001854278963046181, + "loss": 0.0986, + "step": 18450 + }, + { + "epoch": 0.16016354024704646, + "grad_norm": 0.08349609375, + "learning_rate": 0.0018542627261580302, + "loss": 0.103, + "step": 18451 + }, + { + "epoch": 0.16017222072725062, + "grad_norm": 0.296875, + "learning_rate": 0.0018542464884450291, + "loss": 0.1226, + "step": 18452 + }, + { + "epoch": 0.1601809012074548, + "grad_norm": 0.45703125, + "learning_rate": 0.0018542302499071957, + "loss": 0.1025, + "step": 18453 + }, + { + "epoch": 0.16018958168765896, + "grad_norm": 0.54296875, + "learning_rate": 0.001854214010544547, + "loss": 0.1094, + "step": 18454 + }, + { + "epoch": 0.16019826216786312, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018541977703571014, + "loss": 0.0947, + "step": 18455 + }, + { + "epoch": 0.16020694264806729, + "grad_norm": 0.087890625, + "learning_rate": 0.0018541815293448762, + "loss": 0.0991, + "step": 18456 + }, + { + "epoch": 0.16021562312827145, + "grad_norm": 0.271484375, + "learning_rate": 0.0018541652875078893, + "loss": 0.1113, + "step": 18457 + }, + { + "epoch": 0.16022430360847562, + "grad_norm": 0.09130859375, + "learning_rate": 0.001854149044846159, + "loss": 0.0898, + "step": 18458 + }, + { + "epoch": 0.16023298408867978, + "grad_norm": 0.5078125, + "learning_rate": 0.0018541328013597026, + "loss": 0.1206, + "step": 18459 + }, + { + "epoch": 0.16024166456888395, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018541165570485376, + "loss": 0.1289, + "step": 18460 + }, + { + "epoch": 0.1602503450490881, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018541003119126823, + "loss": 0.1289, + "step": 18461 + }, + { + "epoch": 0.16025902552929228, + "grad_norm": 0.283203125, + "learning_rate": 0.0018540840659521543, + "loss": 0.1475, + "step": 18462 + }, + { + "epoch": 0.16026770600949644, + "grad_norm": 0.42578125, + "learning_rate": 0.001854067819166971, + "loss": 0.1152, + "step": 18463 + }, + { + "epoch": 0.1602763864897006, + "grad_norm": 0.474609375, + "learning_rate": 0.0018540515715571506, + "loss": 0.1045, + "step": 18464 + }, + { + "epoch": 0.16028506696990477, + "grad_norm": 0.0810546875, + "learning_rate": 0.0018540353231227106, + "loss": 0.1211, + "step": 18465 + }, + { + "epoch": 0.16029374745010894, + "grad_norm": 0.34375, + "learning_rate": 0.0018540190738636692, + "loss": 0.0874, + "step": 18466 + }, + { + "epoch": 0.1603024279303131, + "grad_norm": 0.59375, + "learning_rate": 0.0018540028237800437, + "loss": 0.125, + "step": 18467 + }, + { + "epoch": 0.16031110841051727, + "grad_norm": 0.171875, + "learning_rate": 0.0018539865728718521, + "loss": 0.1299, + "step": 18468 + }, + { + "epoch": 0.16031978889072143, + "grad_norm": 0.10986328125, + "learning_rate": 0.001853970321139112, + "loss": 0.0957, + "step": 18469 + }, + { + "epoch": 0.1603284693709256, + "grad_norm": 0.126953125, + "learning_rate": 0.0018539540685818415, + "loss": 0.1162, + "step": 18470 + }, + { + "epoch": 0.16033714985112976, + "grad_norm": 0.3515625, + "learning_rate": 0.0018539378152000578, + "loss": 0.0986, + "step": 18471 + }, + { + "epoch": 0.16034583033133393, + "grad_norm": 0.70703125, + "learning_rate": 0.0018539215609937793, + "loss": 0.166, + "step": 18472 + }, + { + "epoch": 0.1603545108115381, + "grad_norm": 1.1171875, + "learning_rate": 0.0018539053059630235, + "loss": 0.103, + "step": 18473 + }, + { + "epoch": 0.16036319129174226, + "grad_norm": 0.5390625, + "learning_rate": 0.0018538890501078082, + "loss": 0.168, + "step": 18474 + }, + { + "epoch": 0.16037187177194642, + "grad_norm": 0.08642578125, + "learning_rate": 0.0018538727934281512, + "loss": 0.1045, + "step": 18475 + }, + { + "epoch": 0.1603805522521506, + "grad_norm": 0.1279296875, + "learning_rate": 0.00185385653592407, + "loss": 0.0996, + "step": 18476 + }, + { + "epoch": 0.16038923273235475, + "grad_norm": 0.2158203125, + "learning_rate": 0.0018538402775955826, + "loss": 0.1699, + "step": 18477 + }, + { + "epoch": 0.16039791321255892, + "grad_norm": 0.0751953125, + "learning_rate": 0.001853824018442707, + "loss": 0.0879, + "step": 18478 + }, + { + "epoch": 0.16040659369276308, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018538077584654606, + "loss": 0.1533, + "step": 18479 + }, + { + "epoch": 0.16041527417296725, + "grad_norm": 0.40234375, + "learning_rate": 0.0018537914976638618, + "loss": 0.1348, + "step": 18480 + }, + { + "epoch": 0.1604239546531714, + "grad_norm": 0.28125, + "learning_rate": 0.0018537752360379277, + "loss": 0.1338, + "step": 18481 + }, + { + "epoch": 0.16043263513337558, + "grad_norm": 0.361328125, + "learning_rate": 0.001853758973587676, + "loss": 0.1191, + "step": 18482 + }, + { + "epoch": 0.16044131561357974, + "grad_norm": 0.291015625, + "learning_rate": 0.0018537427103131252, + "loss": 0.123, + "step": 18483 + }, + { + "epoch": 0.1604499960937839, + "grad_norm": 0.546875, + "learning_rate": 0.0018537264462142925, + "loss": 0.1099, + "step": 18484 + }, + { + "epoch": 0.16045867657398807, + "grad_norm": 0.2431640625, + "learning_rate": 0.001853710181291196, + "loss": 0.1475, + "step": 18485 + }, + { + "epoch": 0.16046735705419224, + "grad_norm": 0.283203125, + "learning_rate": 0.0018536939155438534, + "loss": 0.1113, + "step": 18486 + }, + { + "epoch": 0.1604760375343964, + "grad_norm": 0.79296875, + "learning_rate": 0.0018536776489722824, + "loss": 0.1006, + "step": 18487 + }, + { + "epoch": 0.16048471801460057, + "grad_norm": 0.609375, + "learning_rate": 0.0018536613815765008, + "loss": 0.1196, + "step": 18488 + }, + { + "epoch": 0.16049339849480473, + "grad_norm": 0.2734375, + "learning_rate": 0.0018536451133565265, + "loss": 0.1367, + "step": 18489 + }, + { + "epoch": 0.1605020789750089, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018536288443123771, + "loss": 0.1245, + "step": 18490 + }, + { + "epoch": 0.16051075945521306, + "grad_norm": 0.4453125, + "learning_rate": 0.0018536125744440706, + "loss": 0.1172, + "step": 18491 + }, + { + "epoch": 0.16051943993541723, + "grad_norm": 0.1689453125, + "learning_rate": 0.0018535963037516249, + "loss": 0.0723, + "step": 18492 + }, + { + "epoch": 0.1605281204156214, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018535800322350576, + "loss": 0.0967, + "step": 18493 + }, + { + "epoch": 0.16053680089582556, + "grad_norm": 0.0859375, + "learning_rate": 0.0018535637598943863, + "loss": 0.0649, + "step": 18494 + }, + { + "epoch": 0.16054548137602973, + "grad_norm": 0.333984375, + "learning_rate": 0.001853547486729629, + "loss": 0.1367, + "step": 18495 + }, + { + "epoch": 0.1605541618562339, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018535312127408035, + "loss": 0.0967, + "step": 18496 + }, + { + "epoch": 0.16056284233643806, + "grad_norm": 0.5625, + "learning_rate": 0.0018535149379279273, + "loss": 0.1846, + "step": 18497 + }, + { + "epoch": 0.16057152281664222, + "grad_norm": 0.291015625, + "learning_rate": 0.001853498662291019, + "loss": 0.1211, + "step": 18498 + }, + { + "epoch": 0.16058020329684639, + "grad_norm": 0.267578125, + "learning_rate": 0.0018534823858300958, + "loss": 0.1338, + "step": 18499 + }, + { + "epoch": 0.16058888377705055, + "grad_norm": 0.609375, + "learning_rate": 0.0018534661085451755, + "loss": 0.1221, + "step": 18500 + }, + { + "epoch": 0.16059756425725472, + "grad_norm": 0.48046875, + "learning_rate": 0.0018534498304362756, + "loss": 0.2109, + "step": 18501 + }, + { + "epoch": 0.16060624473745888, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018534335515034148, + "loss": 0.1182, + "step": 18502 + }, + { + "epoch": 0.16061492521766305, + "grad_norm": 0.2734375, + "learning_rate": 0.0018534172717466102, + "loss": 0.1113, + "step": 18503 + }, + { + "epoch": 0.1606236056978672, + "grad_norm": 0.2216796875, + "learning_rate": 0.00185340099116588, + "loss": 0.1011, + "step": 18504 + }, + { + "epoch": 0.16063228617807138, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018533847097612413, + "loss": 0.1172, + "step": 18505 + }, + { + "epoch": 0.16064096665827554, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018533684275327128, + "loss": 0.1221, + "step": 18506 + }, + { + "epoch": 0.1606496471384797, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018533521444803118, + "loss": 0.1104, + "step": 18507 + }, + { + "epoch": 0.16065832761868387, + "grad_norm": 0.7578125, + "learning_rate": 0.0018533358606040561, + "loss": 0.1328, + "step": 18508 + }, + { + "epoch": 0.16066700809888804, + "grad_norm": 0.12255859375, + "learning_rate": 0.001853319575903964, + "loss": 0.0942, + "step": 18509 + }, + { + "epoch": 0.1606756885790922, + "grad_norm": 0.376953125, + "learning_rate": 0.0018533032903800526, + "loss": 0.1143, + "step": 18510 + }, + { + "epoch": 0.16068436905929637, + "grad_norm": 0.1015625, + "learning_rate": 0.0018532870040323401, + "loss": 0.0928, + "step": 18511 + }, + { + "epoch": 0.16069304953950053, + "grad_norm": 0.158203125, + "learning_rate": 0.0018532707168608443, + "loss": 0.1348, + "step": 18512 + }, + { + "epoch": 0.1607017300197047, + "grad_norm": 0.6484375, + "learning_rate": 0.001853254428865583, + "loss": 0.1182, + "step": 18513 + }, + { + "epoch": 0.16071041049990886, + "grad_norm": 0.17578125, + "learning_rate": 0.0018532381400465736, + "loss": 0.0928, + "step": 18514 + }, + { + "epoch": 0.16071909098011303, + "grad_norm": 0.80078125, + "learning_rate": 0.0018532218504038348, + "loss": 0.126, + "step": 18515 + }, + { + "epoch": 0.1607277714603172, + "grad_norm": 0.234375, + "learning_rate": 0.0018532055599373837, + "loss": 0.0947, + "step": 18516 + }, + { + "epoch": 0.16073645194052136, + "grad_norm": 0.490234375, + "learning_rate": 0.0018531892686472382, + "loss": 0.1167, + "step": 18517 + }, + { + "epoch": 0.16074513242072552, + "grad_norm": 0.2373046875, + "learning_rate": 0.0018531729765334163, + "loss": 0.0898, + "step": 18518 + }, + { + "epoch": 0.1607538129009297, + "grad_norm": 0.109375, + "learning_rate": 0.0018531566835959358, + "loss": 0.1025, + "step": 18519 + }, + { + "epoch": 0.16076249338113385, + "grad_norm": 0.373046875, + "learning_rate": 0.0018531403898348147, + "loss": 0.1006, + "step": 18520 + }, + { + "epoch": 0.16077117386133802, + "grad_norm": 0.107421875, + "learning_rate": 0.0018531240952500702, + "loss": 0.1035, + "step": 18521 + }, + { + "epoch": 0.16077985434154218, + "grad_norm": 0.0810546875, + "learning_rate": 0.0018531077998417207, + "loss": 0.1055, + "step": 18522 + }, + { + "epoch": 0.16078853482174635, + "grad_norm": 0.58203125, + "learning_rate": 0.001853091503609784, + "loss": 0.1309, + "step": 18523 + }, + { + "epoch": 0.1607972153019505, + "grad_norm": 1.0, + "learning_rate": 0.0018530752065542774, + "loss": 0.1162, + "step": 18524 + }, + { + "epoch": 0.16080589578215468, + "grad_norm": 0.1171875, + "learning_rate": 0.0018530589086752191, + "loss": 0.0981, + "step": 18525 + }, + { + "epoch": 0.16081457626235884, + "grad_norm": 0.125, + "learning_rate": 0.001853042609972627, + "loss": 0.1235, + "step": 18526 + }, + { + "epoch": 0.160823256742563, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018530263104465188, + "loss": 0.125, + "step": 18527 + }, + { + "epoch": 0.16083193722276717, + "grad_norm": 0.154296875, + "learning_rate": 0.0018530100100969124, + "loss": 0.1123, + "step": 18528 + }, + { + "epoch": 0.16084061770297134, + "grad_norm": 0.3515625, + "learning_rate": 0.0018529937089238253, + "loss": 0.106, + "step": 18529 + }, + { + "epoch": 0.1608492981831755, + "grad_norm": 0.2294921875, + "learning_rate": 0.001852977406927276, + "loss": 0.0781, + "step": 18530 + }, + { + "epoch": 0.16085797866337967, + "grad_norm": 0.3828125, + "learning_rate": 0.0018529611041072818, + "loss": 0.1138, + "step": 18531 + }, + { + "epoch": 0.16086665914358383, + "grad_norm": 0.19140625, + "learning_rate": 0.0018529448004638606, + "loss": 0.084, + "step": 18532 + }, + { + "epoch": 0.160875339623788, + "grad_norm": 0.73046875, + "learning_rate": 0.0018529284959970301, + "loss": 0.1079, + "step": 18533 + }, + { + "epoch": 0.16088402010399216, + "grad_norm": 0.34765625, + "learning_rate": 0.0018529121907068086, + "loss": 0.0586, + "step": 18534 + }, + { + "epoch": 0.16089270058419633, + "grad_norm": 0.091796875, + "learning_rate": 0.0018528958845932134, + "loss": 0.1104, + "step": 18535 + }, + { + "epoch": 0.1609013810644005, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018528795776562627, + "loss": 0.124, + "step": 18536 + }, + { + "epoch": 0.16091006154460466, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018528632698959742, + "loss": 0.1221, + "step": 18537 + }, + { + "epoch": 0.16091874202480883, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018528469613123657, + "loss": 0.0815, + "step": 18538 + }, + { + "epoch": 0.160927422505013, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018528306519054552, + "loss": 0.0991, + "step": 18539 + }, + { + "epoch": 0.16093610298521716, + "grad_norm": 0.52734375, + "learning_rate": 0.0018528143416752603, + "loss": 0.1006, + "step": 18540 + }, + { + "epoch": 0.16094478346542132, + "grad_norm": 0.52734375, + "learning_rate": 0.0018527980306217988, + "loss": 0.0845, + "step": 18541 + }, + { + "epoch": 0.16095346394562549, + "grad_norm": 0.134765625, + "learning_rate": 0.0018527817187450888, + "loss": 0.0962, + "step": 18542 + }, + { + "epoch": 0.16096214442582965, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018527654060451482, + "loss": 0.1504, + "step": 18543 + }, + { + "epoch": 0.1609708249060338, + "grad_norm": 0.349609375, + "learning_rate": 0.0018527490925219945, + "loss": 0.1836, + "step": 18544 + }, + { + "epoch": 0.16097950538623795, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018527327781756456, + "loss": 0.1006, + "step": 18545 + }, + { + "epoch": 0.16098818586644212, + "grad_norm": 0.77734375, + "learning_rate": 0.0018527164630061197, + "loss": 0.1514, + "step": 18546 + }, + { + "epoch": 0.16099686634664628, + "grad_norm": 0.1572265625, + "learning_rate": 0.001852700147013434, + "loss": 0.082, + "step": 18547 + }, + { + "epoch": 0.16100554682685045, + "grad_norm": 0.185546875, + "learning_rate": 0.0018526838301976072, + "loss": 0.0986, + "step": 18548 + }, + { + "epoch": 0.1610142273070546, + "grad_norm": 0.15234375, + "learning_rate": 0.0018526675125586562, + "loss": 0.1123, + "step": 18549 + }, + { + "epoch": 0.16102290778725878, + "grad_norm": 0.47265625, + "learning_rate": 0.0018526511940966, + "loss": 0.0913, + "step": 18550 + }, + { + "epoch": 0.16103158826746294, + "grad_norm": 0.07568359375, + "learning_rate": 0.0018526348748114552, + "loss": 0.1045, + "step": 18551 + }, + { + "epoch": 0.1610402687476671, + "grad_norm": 0.314453125, + "learning_rate": 0.0018526185547032403, + "loss": 0.0918, + "step": 18552 + }, + { + "epoch": 0.16104894922787127, + "grad_norm": 0.0791015625, + "learning_rate": 0.001852602233771973, + "loss": 0.0723, + "step": 18553 + }, + { + "epoch": 0.16105762970807544, + "grad_norm": 0.392578125, + "learning_rate": 0.0018525859120176714, + "loss": 0.0977, + "step": 18554 + }, + { + "epoch": 0.1610663101882796, + "grad_norm": 0.412109375, + "learning_rate": 0.0018525695894403532, + "loss": 0.1367, + "step": 18555 + }, + { + "epoch": 0.16107499066848377, + "grad_norm": 0.8984375, + "learning_rate": 0.001852553266040036, + "loss": 0.1621, + "step": 18556 + }, + { + "epoch": 0.16108367114868793, + "grad_norm": 0.16015625, + "learning_rate": 0.0018525369418167378, + "loss": 0.0977, + "step": 18557 + }, + { + "epoch": 0.1610923516288921, + "grad_norm": 0.359375, + "learning_rate": 0.0018525206167704766, + "loss": 0.1367, + "step": 18558 + }, + { + "epoch": 0.16110103210909626, + "grad_norm": 0.333984375, + "learning_rate": 0.0018525042909012701, + "loss": 0.1289, + "step": 18559 + }, + { + "epoch": 0.16110971258930043, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018524879642091365, + "loss": 0.2305, + "step": 18560 + }, + { + "epoch": 0.1611183930695046, + "grad_norm": 0.451171875, + "learning_rate": 0.0018524716366940933, + "loss": 0.1094, + "step": 18561 + }, + { + "epoch": 0.16112707354970876, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018524553083561581, + "loss": 0.1035, + "step": 18562 + }, + { + "epoch": 0.16113575402991293, + "grad_norm": 0.265625, + "learning_rate": 0.0018524389791953492, + "loss": 0.1123, + "step": 18563 + }, + { + "epoch": 0.1611444345101171, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018524226492116845, + "loss": 0.0947, + "step": 18564 + }, + { + "epoch": 0.16115311499032126, + "grad_norm": 0.1640625, + "learning_rate": 0.0018524063184051818, + "loss": 0.0977, + "step": 18565 + }, + { + "epoch": 0.16116179547052542, + "grad_norm": 0.62109375, + "learning_rate": 0.0018523899867758584, + "loss": 0.1143, + "step": 18566 + }, + { + "epoch": 0.16117047595072959, + "grad_norm": 0.1484375, + "learning_rate": 0.0018523736543237328, + "loss": 0.1118, + "step": 18567 + }, + { + "epoch": 0.16117915643093375, + "grad_norm": 0.1630859375, + "learning_rate": 0.001852357321048823, + "loss": 0.0898, + "step": 18568 + }, + { + "epoch": 0.16118783691113792, + "grad_norm": 0.28515625, + "learning_rate": 0.0018523409869511463, + "loss": 0.0991, + "step": 18569 + }, + { + "epoch": 0.16119651739134208, + "grad_norm": 0.2109375, + "learning_rate": 0.001852324652030721, + "loss": 0.0957, + "step": 18570 + }, + { + "epoch": 0.16120519787154625, + "grad_norm": 0.7109375, + "learning_rate": 0.0018523083162875643, + "loss": 0.1367, + "step": 18571 + }, + { + "epoch": 0.1612138783517504, + "grad_norm": 0.333984375, + "learning_rate": 0.0018522919797216949, + "loss": 0.1201, + "step": 18572 + }, + { + "epoch": 0.16122255883195458, + "grad_norm": 0.65234375, + "learning_rate": 0.0018522756423331298, + "loss": 0.0947, + "step": 18573 + }, + { + "epoch": 0.16123123931215874, + "grad_norm": 0.3984375, + "learning_rate": 0.0018522593041218878, + "loss": 0.0967, + "step": 18574 + }, + { + "epoch": 0.1612399197923629, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018522429650879863, + "loss": 0.1245, + "step": 18575 + }, + { + "epoch": 0.16124860027256707, + "grad_norm": 0.2734375, + "learning_rate": 0.0018522266252314434, + "loss": 0.0977, + "step": 18576 + }, + { + "epoch": 0.16125728075277124, + "grad_norm": 0.193359375, + "learning_rate": 0.0018522102845522765, + "loss": 0.0947, + "step": 18577 + }, + { + "epoch": 0.1612659612329754, + "grad_norm": 0.328125, + "learning_rate": 0.0018521939430505036, + "loss": 0.082, + "step": 18578 + }, + { + "epoch": 0.16127464171317957, + "grad_norm": 0.49609375, + "learning_rate": 0.0018521776007261429, + "loss": 0.0771, + "step": 18579 + }, + { + "epoch": 0.16128332219338373, + "grad_norm": 0.40234375, + "learning_rate": 0.0018521612575792123, + "loss": 0.1533, + "step": 18580 + }, + { + "epoch": 0.1612920026735879, + "grad_norm": 0.6171875, + "learning_rate": 0.001852144913609729, + "loss": 0.1172, + "step": 18581 + }, + { + "epoch": 0.16130068315379206, + "grad_norm": 0.51171875, + "learning_rate": 0.0018521285688177114, + "loss": 0.0957, + "step": 18582 + }, + { + "epoch": 0.16130936363399623, + "grad_norm": 0.8671875, + "learning_rate": 0.0018521122232031775, + "loss": 0.1006, + "step": 18583 + }, + { + "epoch": 0.1613180441142004, + "grad_norm": 0.10400390625, + "learning_rate": 0.001852095876766145, + "loss": 0.1318, + "step": 18584 + }, + { + "epoch": 0.16132672459440456, + "grad_norm": 0.345703125, + "learning_rate": 0.0018520795295066314, + "loss": 0.0972, + "step": 18585 + }, + { + "epoch": 0.16133540507460872, + "grad_norm": 0.498046875, + "learning_rate": 0.0018520631814246554, + "loss": 0.1289, + "step": 18586 + }, + { + "epoch": 0.1613440855548129, + "grad_norm": 0.1220703125, + "learning_rate": 0.001852046832520234, + "loss": 0.1006, + "step": 18587 + }, + { + "epoch": 0.16135276603501705, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018520304827933855, + "loss": 0.1123, + "step": 18588 + }, + { + "epoch": 0.16136144651522122, + "grad_norm": 1.0234375, + "learning_rate": 0.0018520141322441281, + "loss": 0.1504, + "step": 18589 + }, + { + "epoch": 0.16137012699542538, + "grad_norm": 0.2353515625, + "learning_rate": 0.001851997780872479, + "loss": 0.1035, + "step": 18590 + }, + { + "epoch": 0.16137880747562955, + "grad_norm": 0.19921875, + "learning_rate": 0.0018519814286784566, + "loss": 0.1191, + "step": 18591 + }, + { + "epoch": 0.1613874879558337, + "grad_norm": 0.69140625, + "learning_rate": 0.001851965075662079, + "loss": 0.1152, + "step": 18592 + }, + { + "epoch": 0.16139616843603788, + "grad_norm": 0.392578125, + "learning_rate": 0.001851948721823363, + "loss": 0.1089, + "step": 18593 + }, + { + "epoch": 0.16140484891624204, + "grad_norm": 0.279296875, + "learning_rate": 0.0018519323671623276, + "loss": 0.1191, + "step": 18594 + }, + { + "epoch": 0.1614135293964462, + "grad_norm": 0.3046875, + "learning_rate": 0.0018519160116789901, + "loss": 0.1221, + "step": 18595 + }, + { + "epoch": 0.16142220987665037, + "grad_norm": 0.70703125, + "learning_rate": 0.0018518996553733686, + "loss": 0.0918, + "step": 18596 + }, + { + "epoch": 0.16143089035685454, + "grad_norm": 0.1025390625, + "learning_rate": 0.0018518832982454807, + "loss": 0.1914, + "step": 18597 + }, + { + "epoch": 0.1614395708370587, + "grad_norm": 0.61328125, + "learning_rate": 0.0018518669402953448, + "loss": 0.1064, + "step": 18598 + }, + { + "epoch": 0.16144825131726287, + "grad_norm": 0.91796875, + "learning_rate": 0.0018518505815229786, + "loss": 0.1152, + "step": 18599 + }, + { + "epoch": 0.16145693179746703, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018518342219283994, + "loss": 0.1006, + "step": 18600 + }, + { + "epoch": 0.1614656122776712, + "grad_norm": 0.8671875, + "learning_rate": 0.001851817861511626, + "loss": 0.0981, + "step": 18601 + }, + { + "epoch": 0.16147429275787537, + "grad_norm": 0.51953125, + "learning_rate": 0.001851801500272676, + "loss": 0.126, + "step": 18602 + }, + { + "epoch": 0.16148297323807953, + "grad_norm": 0.80078125, + "learning_rate": 0.0018517851382115669, + "loss": 0.1357, + "step": 18603 + }, + { + "epoch": 0.1614916537182837, + "grad_norm": 0.1640625, + "learning_rate": 0.001851768775328317, + "loss": 0.1523, + "step": 18604 + }, + { + "epoch": 0.16150033419848786, + "grad_norm": 0.2138671875, + "learning_rate": 0.001851752411622944, + "loss": 0.0933, + "step": 18605 + }, + { + "epoch": 0.16150901467869203, + "grad_norm": 0.67578125, + "learning_rate": 0.0018517360470954659, + "loss": 0.0967, + "step": 18606 + }, + { + "epoch": 0.1615176951588962, + "grad_norm": 0.240234375, + "learning_rate": 0.0018517196817459003, + "loss": 0.1016, + "step": 18607 + }, + { + "epoch": 0.16152637563910036, + "grad_norm": 0.54296875, + "learning_rate": 0.0018517033155742658, + "loss": 0.1182, + "step": 18608 + }, + { + "epoch": 0.16153505611930452, + "grad_norm": 0.11767578125, + "learning_rate": 0.0018516869485805797, + "loss": 0.0923, + "step": 18609 + }, + { + "epoch": 0.16154373659950869, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018516705807648597, + "loss": 0.1426, + "step": 18610 + }, + { + "epoch": 0.16155241707971285, + "grad_norm": 0.765625, + "learning_rate": 0.0018516542121271246, + "loss": 0.1797, + "step": 18611 + }, + { + "epoch": 0.16156109755991702, + "grad_norm": 0.396484375, + "learning_rate": 0.0018516378426673915, + "loss": 0.1074, + "step": 18612 + }, + { + "epoch": 0.16156977804012118, + "grad_norm": 0.78125, + "learning_rate": 0.0018516214723856784, + "loss": 0.0986, + "step": 18613 + }, + { + "epoch": 0.16157845852032535, + "grad_norm": 0.16796875, + "learning_rate": 0.0018516051012820035, + "loss": 0.0952, + "step": 18614 + }, + { + "epoch": 0.1615871390005295, + "grad_norm": 0.08056640625, + "learning_rate": 0.0018515887293563845, + "loss": 0.0796, + "step": 18615 + }, + { + "epoch": 0.16159581948073368, + "grad_norm": 0.56640625, + "learning_rate": 0.0018515723566088394, + "loss": 0.0981, + "step": 18616 + }, + { + "epoch": 0.16160449996093784, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018515559830393857, + "loss": 0.1016, + "step": 18617 + }, + { + "epoch": 0.161613180441142, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018515396086480423, + "loss": 0.1128, + "step": 18618 + }, + { + "epoch": 0.16162186092134617, + "grad_norm": 0.126953125, + "learning_rate": 0.0018515232334348263, + "loss": 0.1045, + "step": 18619 + }, + { + "epoch": 0.16163054140155034, + "grad_norm": 0.146484375, + "learning_rate": 0.0018515068573997555, + "loss": 0.1543, + "step": 18620 + }, + { + "epoch": 0.1616392218817545, + "grad_norm": 0.263671875, + "learning_rate": 0.0018514904805428483, + "loss": 0.1318, + "step": 18621 + }, + { + "epoch": 0.16164790236195867, + "grad_norm": 0.41796875, + "learning_rate": 0.0018514741028641223, + "loss": 0.0815, + "step": 18622 + }, + { + "epoch": 0.16165658284216283, + "grad_norm": 0.203125, + "learning_rate": 0.0018514577243635957, + "loss": 0.0957, + "step": 18623 + }, + { + "epoch": 0.161665263322367, + "grad_norm": 0.291015625, + "learning_rate": 0.001851441345041286, + "loss": 0.1138, + "step": 18624 + }, + { + "epoch": 0.16167394380257116, + "grad_norm": 0.099609375, + "learning_rate": 0.0018514249648972116, + "loss": 0.0933, + "step": 18625 + }, + { + "epoch": 0.16168262428277533, + "grad_norm": 0.53125, + "learning_rate": 0.00185140858393139, + "loss": 0.1206, + "step": 18626 + }, + { + "epoch": 0.1616913047629795, + "grad_norm": 0.1162109375, + "learning_rate": 0.0018513922021438389, + "loss": 0.0933, + "step": 18627 + }, + { + "epoch": 0.16169998524318366, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018513758195345769, + "loss": 0.1182, + "step": 18628 + }, + { + "epoch": 0.16170866572338782, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018513594361036219, + "loss": 0.1094, + "step": 18629 + }, + { + "epoch": 0.161717346203592, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018513430518509913, + "loss": 0.1245, + "step": 18630 + }, + { + "epoch": 0.16172602668379615, + "grad_norm": 0.09130859375, + "learning_rate": 0.001851326666776703, + "loss": 0.1162, + "step": 18631 + }, + { + "epoch": 0.16173470716400032, + "grad_norm": 0.16796875, + "learning_rate": 0.0018513102808807753, + "loss": 0.1387, + "step": 18632 + }, + { + "epoch": 0.16174338764420448, + "grad_norm": 0.3125, + "learning_rate": 0.001851293894163226, + "loss": 0.1021, + "step": 18633 + }, + { + "epoch": 0.16175206812440865, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018512775066240727, + "loss": 0.1182, + "step": 18634 + }, + { + "epoch": 0.16176074860461281, + "grad_norm": 0.072265625, + "learning_rate": 0.001851261118263334, + "loss": 0.0903, + "step": 18635 + }, + { + "epoch": 0.16176942908481698, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018512447290810275, + "loss": 0.1187, + "step": 18636 + }, + { + "epoch": 0.16177810956502114, + "grad_norm": 0.56640625, + "learning_rate": 0.0018512283390771705, + "loss": 0.127, + "step": 18637 + }, + { + "epoch": 0.1617867900452253, + "grad_norm": 0.53125, + "learning_rate": 0.001851211948251782, + "loss": 0.0903, + "step": 18638 + }, + { + "epoch": 0.16179547052542947, + "grad_norm": 0.1328125, + "learning_rate": 0.0018511955566048793, + "loss": 0.1152, + "step": 18639 + }, + { + "epoch": 0.16180415100563364, + "grad_norm": 0.3125, + "learning_rate": 0.0018511791641364803, + "loss": 0.1221, + "step": 18640 + }, + { + "epoch": 0.1618128314858378, + "grad_norm": 0.74609375, + "learning_rate": 0.0018511627708466033, + "loss": 0.084, + "step": 18641 + }, + { + "epoch": 0.16182151196604197, + "grad_norm": 0.2451171875, + "learning_rate": 0.0018511463767352655, + "loss": 0.0938, + "step": 18642 + }, + { + "epoch": 0.16183019244624613, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018511299818024856, + "loss": 0.0776, + "step": 18643 + }, + { + "epoch": 0.1618388729264503, + "grad_norm": 0.09814453125, + "learning_rate": 0.001851113586048281, + "loss": 0.103, + "step": 18644 + }, + { + "epoch": 0.16184755340665447, + "grad_norm": 0.083984375, + "learning_rate": 0.0018510971894726703, + "loss": 0.124, + "step": 18645 + }, + { + "epoch": 0.16185623388685863, + "grad_norm": 0.30078125, + "learning_rate": 0.0018510807920756706, + "loss": 0.1592, + "step": 18646 + }, + { + "epoch": 0.1618649143670628, + "grad_norm": 0.41796875, + "learning_rate": 0.0018510643938573006, + "loss": 0.1167, + "step": 18647 + }, + { + "epoch": 0.16187359484726696, + "grad_norm": 0.9609375, + "learning_rate": 0.001851047994817578, + "loss": 0.1289, + "step": 18648 + }, + { + "epoch": 0.16188227532747113, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018510315949565198, + "loss": 0.1289, + "step": 18649 + }, + { + "epoch": 0.1618909558076753, + "grad_norm": 0.0908203125, + "learning_rate": 0.001851015194274145, + "loss": 0.1533, + "step": 18650 + }, + { + "epoch": 0.16189963628787946, + "grad_norm": 0.609375, + "learning_rate": 0.0018509987927704716, + "loss": 0.1206, + "step": 18651 + }, + { + "epoch": 0.16190831676808362, + "grad_norm": 0.10205078125, + "learning_rate": 0.0018509823904455174, + "loss": 0.0796, + "step": 18652 + }, + { + "epoch": 0.1619169972482878, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018509659872992996, + "loss": 0.1152, + "step": 18653 + }, + { + "epoch": 0.16192567772849195, + "grad_norm": 0.11181640625, + "learning_rate": 0.001850949583331837, + "loss": 0.124, + "step": 18654 + }, + { + "epoch": 0.16193435820869612, + "grad_norm": 0.26953125, + "learning_rate": 0.001850933178543147, + "loss": 0.0854, + "step": 18655 + }, + { + "epoch": 0.16194303868890028, + "grad_norm": 0.275390625, + "learning_rate": 0.001850916772933248, + "loss": 0.1562, + "step": 18656 + }, + { + "epoch": 0.16195171916910445, + "grad_norm": 0.29296875, + "learning_rate": 0.0018509003665021576, + "loss": 0.1172, + "step": 18657 + }, + { + "epoch": 0.1619603996493086, + "grad_norm": 0.392578125, + "learning_rate": 0.001850883959249894, + "loss": 0.1201, + "step": 18658 + }, + { + "epoch": 0.16196908012951278, + "grad_norm": 0.296875, + "learning_rate": 0.0018508675511764749, + "loss": 0.1064, + "step": 18659 + }, + { + "epoch": 0.16197776060971694, + "grad_norm": 0.5, + "learning_rate": 0.0018508511422819182, + "loss": 0.127, + "step": 18660 + }, + { + "epoch": 0.1619864410899211, + "grad_norm": 0.0791015625, + "learning_rate": 0.0018508347325662418, + "loss": 0.0757, + "step": 18661 + }, + { + "epoch": 0.16199512157012527, + "grad_norm": 0.09375, + "learning_rate": 0.0018508183220294638, + "loss": 0.0796, + "step": 18662 + }, + { + "epoch": 0.16200380205032944, + "grad_norm": 0.373046875, + "learning_rate": 0.0018508019106716026, + "loss": 0.123, + "step": 18663 + }, + { + "epoch": 0.1620124825305336, + "grad_norm": 0.51171875, + "learning_rate": 0.0018507854984926753, + "loss": 0.1104, + "step": 18664 + }, + { + "epoch": 0.16202116301073777, + "grad_norm": 0.6328125, + "learning_rate": 0.0018507690854927005, + "loss": 0.0825, + "step": 18665 + }, + { + "epoch": 0.16202984349094193, + "grad_norm": 0.1826171875, + "learning_rate": 0.001850752671671696, + "loss": 0.0967, + "step": 18666 + }, + { + "epoch": 0.16203852397114607, + "grad_norm": 0.388671875, + "learning_rate": 0.0018507362570296792, + "loss": 0.0757, + "step": 18667 + }, + { + "epoch": 0.16204720445135024, + "grad_norm": 0.49609375, + "learning_rate": 0.001850719841566669, + "loss": 0.1064, + "step": 18668 + }, + { + "epoch": 0.1620558849315544, + "grad_norm": 0.326171875, + "learning_rate": 0.0018507034252826828, + "loss": 0.1123, + "step": 18669 + }, + { + "epoch": 0.16206456541175857, + "grad_norm": 0.318359375, + "learning_rate": 0.0018506870081777382, + "loss": 0.0869, + "step": 18670 + }, + { + "epoch": 0.16207324589196273, + "grad_norm": 0.255859375, + "learning_rate": 0.0018506705902518541, + "loss": 0.1562, + "step": 18671 + }, + { + "epoch": 0.1620819263721669, + "grad_norm": 0.375, + "learning_rate": 0.0018506541715050478, + "loss": 0.1367, + "step": 18672 + }, + { + "epoch": 0.16209060685237106, + "grad_norm": 0.388671875, + "learning_rate": 0.0018506377519373373, + "loss": 0.125, + "step": 18673 + }, + { + "epoch": 0.16209928733257523, + "grad_norm": 0.53515625, + "learning_rate": 0.0018506213315487405, + "loss": 0.1016, + "step": 18674 + }, + { + "epoch": 0.1621079678127794, + "grad_norm": 0.45703125, + "learning_rate": 0.0018506049103392758, + "loss": 0.105, + "step": 18675 + }, + { + "epoch": 0.16211664829298356, + "grad_norm": 0.09375, + "learning_rate": 0.0018505884883089606, + "loss": 0.127, + "step": 18676 + }, + { + "epoch": 0.16212532877318772, + "grad_norm": 0.42578125, + "learning_rate": 0.0018505720654578128, + "loss": 0.0747, + "step": 18677 + }, + { + "epoch": 0.1621340092533919, + "grad_norm": 0.251953125, + "learning_rate": 0.001850555641785851, + "loss": 0.0942, + "step": 18678 + }, + { + "epoch": 0.16214268973359605, + "grad_norm": 0.333984375, + "learning_rate": 0.0018505392172930928, + "loss": 0.125, + "step": 18679 + }, + { + "epoch": 0.16215137021380022, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018505227919795564, + "loss": 0.104, + "step": 18680 + }, + { + "epoch": 0.16216005069400438, + "grad_norm": 0.65625, + "learning_rate": 0.0018505063658452595, + "loss": 0.1221, + "step": 18681 + }, + { + "epoch": 0.16216873117420855, + "grad_norm": 0.447265625, + "learning_rate": 0.0018504899388902196, + "loss": 0.1094, + "step": 18682 + }, + { + "epoch": 0.1621774116544127, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018504735111144557, + "loss": 0.0918, + "step": 18683 + }, + { + "epoch": 0.16218609213461688, + "grad_norm": 0.34375, + "learning_rate": 0.001850457082517985, + "loss": 0.0928, + "step": 18684 + }, + { + "epoch": 0.16219477261482104, + "grad_norm": 0.18359375, + "learning_rate": 0.001850440653100826, + "loss": 0.1318, + "step": 18685 + }, + { + "epoch": 0.1622034530950252, + "grad_norm": 0.10498046875, + "learning_rate": 0.001850424222862996, + "loss": 0.1338, + "step": 18686 + }, + { + "epoch": 0.16221213357522937, + "grad_norm": 1.359375, + "learning_rate": 0.001850407791804514, + "loss": 0.1523, + "step": 18687 + }, + { + "epoch": 0.16222081405543354, + "grad_norm": 0.53515625, + "learning_rate": 0.0018503913599253966, + "loss": 0.1099, + "step": 18688 + }, + { + "epoch": 0.1622294945356377, + "grad_norm": 0.126953125, + "learning_rate": 0.001850374927225663, + "loss": 0.0815, + "step": 18689 + }, + { + "epoch": 0.16223817501584187, + "grad_norm": 0.447265625, + "learning_rate": 0.0018503584937053304, + "loss": 0.1621, + "step": 18690 + }, + { + "epoch": 0.16224685549604603, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018503420593644169, + "loss": 0.085, + "step": 18691 + }, + { + "epoch": 0.1622555359762502, + "grad_norm": 0.19140625, + "learning_rate": 0.0018503256242029407, + "loss": 0.0728, + "step": 18692 + }, + { + "epoch": 0.16226421645645436, + "grad_norm": 0.98046875, + "learning_rate": 0.0018503091882209196, + "loss": 0.1123, + "step": 18693 + }, + { + "epoch": 0.16227289693665853, + "grad_norm": 0.38671875, + "learning_rate": 0.0018502927514183719, + "loss": 0.1445, + "step": 18694 + }, + { + "epoch": 0.1622815774168627, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018502763137953151, + "loss": 0.1045, + "step": 18695 + }, + { + "epoch": 0.16229025789706686, + "grad_norm": 0.37109375, + "learning_rate": 0.0018502598753517673, + "loss": 0.1641, + "step": 18696 + }, + { + "epoch": 0.16229893837727102, + "grad_norm": 0.15625, + "learning_rate": 0.001850243436087747, + "loss": 0.0986, + "step": 18697 + }, + { + "epoch": 0.1623076188574752, + "grad_norm": 0.275390625, + "learning_rate": 0.0018502269960032713, + "loss": 0.1055, + "step": 18698 + }, + { + "epoch": 0.16231629933767935, + "grad_norm": 0.263671875, + "learning_rate": 0.0018502105550983587, + "loss": 0.0923, + "step": 18699 + }, + { + "epoch": 0.16232497981788352, + "grad_norm": 0.0859375, + "learning_rate": 0.0018501941133730274, + "loss": 0.0874, + "step": 18700 + }, + { + "epoch": 0.16233366029808768, + "grad_norm": 0.259765625, + "learning_rate": 0.001850177670827295, + "loss": 0.0835, + "step": 18701 + }, + { + "epoch": 0.16234234077829185, + "grad_norm": 0.640625, + "learning_rate": 0.0018501612274611791, + "loss": 0.0869, + "step": 18702 + }, + { + "epoch": 0.16235102125849601, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018501447832746988, + "loss": 0.123, + "step": 18703 + }, + { + "epoch": 0.16235970173870018, + "grad_norm": 0.267578125, + "learning_rate": 0.001850128338267871, + "loss": 0.0869, + "step": 18704 + }, + { + "epoch": 0.16236838221890434, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018501118924407144, + "loss": 0.085, + "step": 18705 + }, + { + "epoch": 0.1623770626991085, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018500954457932467, + "loss": 0.0859, + "step": 18706 + }, + { + "epoch": 0.16238574317931267, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018500789983254857, + "loss": 0.1133, + "step": 18707 + }, + { + "epoch": 0.16239442365951684, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018500625500374498, + "loss": 0.1426, + "step": 18708 + }, + { + "epoch": 0.162403104139721, + "grad_norm": 0.27734375, + "learning_rate": 0.0018500461009291562, + "loss": 0.0938, + "step": 18709 + }, + { + "epoch": 0.16241178461992517, + "grad_norm": 0.50390625, + "learning_rate": 0.0018500296510006243, + "loss": 0.1104, + "step": 18710 + }, + { + "epoch": 0.16242046510012934, + "grad_norm": 0.2431640625, + "learning_rate": 0.0018500132002518706, + "loss": 0.0952, + "step": 18711 + }, + { + "epoch": 0.1624291455803335, + "grad_norm": 0.3515625, + "learning_rate": 0.0018499967486829139, + "loss": 0.1289, + "step": 18712 + }, + { + "epoch": 0.16243782606053767, + "grad_norm": 0.41796875, + "learning_rate": 0.0018499802962937723, + "loss": 0.0884, + "step": 18713 + }, + { + "epoch": 0.16244650654074183, + "grad_norm": 0.224609375, + "learning_rate": 0.0018499638430844632, + "loss": 0.085, + "step": 18714 + }, + { + "epoch": 0.162455187020946, + "grad_norm": 0.0625, + "learning_rate": 0.001849947389055005, + "loss": 0.084, + "step": 18715 + }, + { + "epoch": 0.16246386750115016, + "grad_norm": 0.97265625, + "learning_rate": 0.0018499309342054156, + "loss": 0.1309, + "step": 18716 + }, + { + "epoch": 0.16247254798135433, + "grad_norm": 0.8515625, + "learning_rate": 0.0018499144785357133, + "loss": 0.0908, + "step": 18717 + }, + { + "epoch": 0.1624812284615585, + "grad_norm": 0.166015625, + "learning_rate": 0.0018498980220459152, + "loss": 0.0972, + "step": 18718 + }, + { + "epoch": 0.16248990894176266, + "grad_norm": 0.375, + "learning_rate": 0.00184988156473604, + "loss": 0.1494, + "step": 18719 + }, + { + "epoch": 0.16249858942196682, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018498651066061061, + "loss": 0.1147, + "step": 18720 + }, + { + "epoch": 0.162507269902171, + "grad_norm": 0.11767578125, + "learning_rate": 0.0018498486476561304, + "loss": 0.1465, + "step": 18721 + }, + { + "epoch": 0.16251595038237515, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018498321878861318, + "loss": 0.1123, + "step": 18722 + }, + { + "epoch": 0.16252463086257932, + "grad_norm": 0.84375, + "learning_rate": 0.0018498157272961277, + "loss": 0.0908, + "step": 18723 + }, + { + "epoch": 0.16253331134278348, + "grad_norm": 0.091796875, + "learning_rate": 0.0018497992658861367, + "loss": 0.1387, + "step": 18724 + }, + { + "epoch": 0.16254199182298765, + "grad_norm": 0.734375, + "learning_rate": 0.0018497828036561767, + "loss": 0.1055, + "step": 18725 + }, + { + "epoch": 0.1625506723031918, + "grad_norm": 0.1083984375, + "learning_rate": 0.001849766340606265, + "loss": 0.0781, + "step": 18726 + }, + { + "epoch": 0.16255935278339598, + "grad_norm": 0.11279296875, + "learning_rate": 0.0018497498767364206, + "loss": 0.0962, + "step": 18727 + }, + { + "epoch": 0.16256803326360014, + "grad_norm": 0.5390625, + "learning_rate": 0.0018497334120466604, + "loss": 0.0742, + "step": 18728 + }, + { + "epoch": 0.1625767137438043, + "grad_norm": 0.97265625, + "learning_rate": 0.0018497169465370035, + "loss": 0.1367, + "step": 18729 + }, + { + "epoch": 0.16258539422400847, + "grad_norm": 0.9453125, + "learning_rate": 0.0018497004802074672, + "loss": 0.1191, + "step": 18730 + }, + { + "epoch": 0.16259407470421264, + "grad_norm": 0.78125, + "learning_rate": 0.0018496840130580695, + "loss": 0.0903, + "step": 18731 + }, + { + "epoch": 0.1626027551844168, + "grad_norm": 0.8828125, + "learning_rate": 0.001849667545088829, + "loss": 0.0947, + "step": 18732 + }, + { + "epoch": 0.16261143566462097, + "grad_norm": 0.7734375, + "learning_rate": 0.001849651076299763, + "loss": 0.1055, + "step": 18733 + }, + { + "epoch": 0.16262011614482513, + "grad_norm": 0.53515625, + "learning_rate": 0.00184963460669089, + "loss": 0.0869, + "step": 18734 + }, + { + "epoch": 0.1626287966250293, + "grad_norm": 0.8359375, + "learning_rate": 0.001849618136262228, + "loss": 0.0977, + "step": 18735 + }, + { + "epoch": 0.16263747710523346, + "grad_norm": 0.306640625, + "learning_rate": 0.0018496016650137949, + "loss": 0.1357, + "step": 18736 + }, + { + "epoch": 0.16264615758543763, + "grad_norm": 0.423828125, + "learning_rate": 0.0018495851929456084, + "loss": 0.1128, + "step": 18737 + }, + { + "epoch": 0.1626548380656418, + "grad_norm": 0.212890625, + "learning_rate": 0.0018495687200576869, + "loss": 0.123, + "step": 18738 + }, + { + "epoch": 0.16266351854584596, + "grad_norm": 0.244140625, + "learning_rate": 0.0018495522463500484, + "loss": 0.1416, + "step": 18739 + }, + { + "epoch": 0.16267219902605012, + "grad_norm": 0.77734375, + "learning_rate": 0.0018495357718227108, + "loss": 0.0879, + "step": 18740 + }, + { + "epoch": 0.1626808795062543, + "grad_norm": 1.0859375, + "learning_rate": 0.001849519296475692, + "loss": 0.1064, + "step": 18741 + }, + { + "epoch": 0.16268955998645845, + "grad_norm": 0.59765625, + "learning_rate": 0.0018495028203090104, + "loss": 0.1025, + "step": 18742 + }, + { + "epoch": 0.16269824046666262, + "grad_norm": 0.3359375, + "learning_rate": 0.0018494863433226837, + "loss": 0.1426, + "step": 18743 + }, + { + "epoch": 0.16270692094686678, + "grad_norm": 0.61328125, + "learning_rate": 0.00184946986551673, + "loss": 0.1162, + "step": 18744 + }, + { + "epoch": 0.16271560142707095, + "grad_norm": 0.57421875, + "learning_rate": 0.0018494533868911674, + "loss": 0.0928, + "step": 18745 + }, + { + "epoch": 0.16272428190727511, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018494369074460136, + "loss": 0.1221, + "step": 18746 + }, + { + "epoch": 0.16273296238747928, + "grad_norm": 0.2412109375, + "learning_rate": 0.001849420427181287, + "loss": 0.125, + "step": 18747 + }, + { + "epoch": 0.16274164286768344, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018494039460970053, + "loss": 0.1113, + "step": 18748 + }, + { + "epoch": 0.1627503233478876, + "grad_norm": 0.1142578125, + "learning_rate": 0.001849387464193187, + "loss": 0.1235, + "step": 18749 + }, + { + "epoch": 0.16275900382809178, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018493709814698493, + "loss": 0.0928, + "step": 18750 + }, + { + "epoch": 0.16276768430829594, + "grad_norm": 0.1025390625, + "learning_rate": 0.0018493544979270114, + "loss": 0.1143, + "step": 18751 + }, + { + "epoch": 0.1627763647885001, + "grad_norm": 1.28125, + "learning_rate": 0.0018493380135646905, + "loss": 0.1816, + "step": 18752 + }, + { + "epoch": 0.16278504526870427, + "grad_norm": 0.42578125, + "learning_rate": 0.0018493215283829046, + "loss": 0.0801, + "step": 18753 + }, + { + "epoch": 0.16279372574890844, + "grad_norm": 0.8203125, + "learning_rate": 0.001849305042381672, + "loss": 0.124, + "step": 18754 + }, + { + "epoch": 0.1628024062291126, + "grad_norm": 0.330078125, + "learning_rate": 0.0018492885555610105, + "loss": 0.1064, + "step": 18755 + }, + { + "epoch": 0.16281108670931677, + "grad_norm": 0.60546875, + "learning_rate": 0.0018492720679209387, + "loss": 0.1426, + "step": 18756 + }, + { + "epoch": 0.16281976718952093, + "grad_norm": 0.3984375, + "learning_rate": 0.001849255579461474, + "loss": 0.1445, + "step": 18757 + }, + { + "epoch": 0.1628284476697251, + "grad_norm": 0.16796875, + "learning_rate": 0.0018492390901826347, + "loss": 0.1475, + "step": 18758 + }, + { + "epoch": 0.16283712814992926, + "grad_norm": 0.158203125, + "learning_rate": 0.0018492226000844387, + "loss": 0.1201, + "step": 18759 + }, + { + "epoch": 0.16284580863013343, + "grad_norm": 0.1328125, + "learning_rate": 0.0018492061091669044, + "loss": 0.0898, + "step": 18760 + }, + { + "epoch": 0.1628544891103376, + "grad_norm": 0.09423828125, + "learning_rate": 0.0018491896174300491, + "loss": 0.085, + "step": 18761 + }, + { + "epoch": 0.16286316959054176, + "grad_norm": 0.61328125, + "learning_rate": 0.0018491731248738916, + "loss": 0.1191, + "step": 18762 + }, + { + "epoch": 0.16287185007074592, + "grad_norm": 0.5078125, + "learning_rate": 0.00184915663149845, + "loss": 0.1318, + "step": 18763 + }, + { + "epoch": 0.1628805305509501, + "grad_norm": 0.224609375, + "learning_rate": 0.0018491401373037413, + "loss": 0.1147, + "step": 18764 + }, + { + "epoch": 0.16288921103115425, + "grad_norm": 0.291015625, + "learning_rate": 0.0018491236422897843, + "loss": 0.1318, + "step": 18765 + }, + { + "epoch": 0.16289789151135842, + "grad_norm": 0.12255859375, + "learning_rate": 0.001849107146456597, + "loss": 0.1484, + "step": 18766 + }, + { + "epoch": 0.16290657199156258, + "grad_norm": 0.265625, + "learning_rate": 0.0018490906498041977, + "loss": 0.1113, + "step": 18767 + }, + { + "epoch": 0.16291525247176675, + "grad_norm": 0.28125, + "learning_rate": 0.0018490741523326038, + "loss": 0.124, + "step": 18768 + }, + { + "epoch": 0.1629239329519709, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018490576540418336, + "loss": 0.1084, + "step": 18769 + }, + { + "epoch": 0.16293261343217508, + "grad_norm": 0.126953125, + "learning_rate": 0.0018490411549319055, + "loss": 0.1299, + "step": 18770 + }, + { + "epoch": 0.16294129391237924, + "grad_norm": 0.5078125, + "learning_rate": 0.0018490246550028368, + "loss": 0.1299, + "step": 18771 + }, + { + "epoch": 0.1629499743925834, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018490081542546465, + "loss": 0.1206, + "step": 18772 + }, + { + "epoch": 0.16295865487278757, + "grad_norm": 0.51171875, + "learning_rate": 0.0018489916526873518, + "loss": 0.1235, + "step": 18773 + }, + { + "epoch": 0.16296733535299174, + "grad_norm": 0.20703125, + "learning_rate": 0.0018489751503009712, + "loss": 0.1235, + "step": 18774 + }, + { + "epoch": 0.1629760158331959, + "grad_norm": 0.69140625, + "learning_rate": 0.001848958647095523, + "loss": 0.1035, + "step": 18775 + }, + { + "epoch": 0.16298469631340007, + "grad_norm": 0.263671875, + "learning_rate": 0.0018489421430710243, + "loss": 0.1113, + "step": 18776 + }, + { + "epoch": 0.16299337679360423, + "grad_norm": 0.0908203125, + "learning_rate": 0.001848925638227494, + "loss": 0.1338, + "step": 18777 + }, + { + "epoch": 0.1630020572738084, + "grad_norm": 0.1044921875, + "learning_rate": 0.0018489091325649496, + "loss": 0.1348, + "step": 18778 + }, + { + "epoch": 0.16301073775401256, + "grad_norm": 0.474609375, + "learning_rate": 0.0018488926260834095, + "loss": 0.1592, + "step": 18779 + }, + { + "epoch": 0.16301941823421673, + "grad_norm": 0.73828125, + "learning_rate": 0.0018488761187828919, + "loss": 0.1133, + "step": 18780 + }, + { + "epoch": 0.1630280987144209, + "grad_norm": 0.3671875, + "learning_rate": 0.0018488596106634148, + "loss": 0.0771, + "step": 18781 + }, + { + "epoch": 0.16303677919462506, + "grad_norm": 0.12890625, + "learning_rate": 0.0018488431017249958, + "loss": 0.1123, + "step": 18782 + }, + { + "epoch": 0.16304545967482922, + "grad_norm": 0.17578125, + "learning_rate": 0.001848826591967653, + "loss": 0.0981, + "step": 18783 + }, + { + "epoch": 0.1630541401550334, + "grad_norm": 0.0849609375, + "learning_rate": 0.0018488100813914053, + "loss": 0.0854, + "step": 18784 + }, + { + "epoch": 0.16306282063523755, + "grad_norm": 0.3203125, + "learning_rate": 0.0018487935699962696, + "loss": 0.1201, + "step": 18785 + }, + { + "epoch": 0.16307150111544172, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018487770577822647, + "loss": 0.1562, + "step": 18786 + }, + { + "epoch": 0.16308018159564588, + "grad_norm": 0.337890625, + "learning_rate": 0.0018487605447494084, + "loss": 0.1572, + "step": 18787 + }, + { + "epoch": 0.16308886207585005, + "grad_norm": 0.62109375, + "learning_rate": 0.0018487440308977189, + "loss": 0.1738, + "step": 18788 + }, + { + "epoch": 0.16309754255605421, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018487275162272138, + "loss": 0.1245, + "step": 18789 + }, + { + "epoch": 0.16310622303625835, + "grad_norm": 0.671875, + "learning_rate": 0.0018487110007379122, + "loss": 0.124, + "step": 18790 + }, + { + "epoch": 0.16311490351646252, + "grad_norm": 0.30078125, + "learning_rate": 0.001848694484429831, + "loss": 0.0986, + "step": 18791 + }, + { + "epoch": 0.16312358399666668, + "grad_norm": 0.33984375, + "learning_rate": 0.0018486779673029892, + "loss": 0.1162, + "step": 18792 + }, + { + "epoch": 0.16313226447687085, + "grad_norm": 0.08837890625, + "learning_rate": 0.001848661449357404, + "loss": 0.1094, + "step": 18793 + }, + { + "epoch": 0.163140944957075, + "grad_norm": 0.25, + "learning_rate": 0.0018486449305930943, + "loss": 0.1152, + "step": 18794 + }, + { + "epoch": 0.16314962543727918, + "grad_norm": 0.35546875, + "learning_rate": 0.0018486284110100773, + "loss": 0.1357, + "step": 18795 + }, + { + "epoch": 0.16315830591748334, + "grad_norm": 0.400390625, + "learning_rate": 0.0018486118906083718, + "loss": 0.1182, + "step": 18796 + }, + { + "epoch": 0.1631669863976875, + "grad_norm": 0.08349609375, + "learning_rate": 0.0018485953693879956, + "loss": 0.0688, + "step": 18797 + }, + { + "epoch": 0.16317566687789167, + "grad_norm": 0.25, + "learning_rate": 0.0018485788473489666, + "loss": 0.0918, + "step": 18798 + }, + { + "epoch": 0.16318434735809584, + "grad_norm": 0.11669921875, + "learning_rate": 0.001848562324491303, + "loss": 0.0972, + "step": 18799 + }, + { + "epoch": 0.1631930278383, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018485458008150229, + "loss": 0.1006, + "step": 18800 + }, + { + "epoch": 0.16320170831850417, + "grad_norm": 0.314453125, + "learning_rate": 0.0018485292763201445, + "loss": 0.1289, + "step": 18801 + }, + { + "epoch": 0.16321038879870833, + "grad_norm": 0.58984375, + "learning_rate": 0.0018485127510066857, + "loss": 0.1094, + "step": 18802 + }, + { + "epoch": 0.1632190692789125, + "grad_norm": 0.486328125, + "learning_rate": 0.0018484962248746645, + "loss": 0.1377, + "step": 18803 + }, + { + "epoch": 0.16322774975911666, + "grad_norm": 0.06396484375, + "learning_rate": 0.0018484796979240992, + "loss": 0.0703, + "step": 18804 + }, + { + "epoch": 0.16323643023932083, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018484631701550078, + "loss": 0.1357, + "step": 18805 + }, + { + "epoch": 0.163245110719525, + "grad_norm": 0.45703125, + "learning_rate": 0.001848446641567408, + "loss": 0.0923, + "step": 18806 + }, + { + "epoch": 0.16325379119972916, + "grad_norm": 0.65625, + "learning_rate": 0.0018484301121613185, + "loss": 0.168, + "step": 18807 + }, + { + "epoch": 0.16326247167993332, + "grad_norm": 0.1015625, + "learning_rate": 0.001848413581936757, + "loss": 0.1074, + "step": 18808 + }, + { + "epoch": 0.1632711521601375, + "grad_norm": 0.82421875, + "learning_rate": 0.0018483970508937418, + "loss": 0.1367, + "step": 18809 + }, + { + "epoch": 0.16327983264034165, + "grad_norm": 0.162109375, + "learning_rate": 0.0018483805190322904, + "loss": 0.1152, + "step": 18810 + }, + { + "epoch": 0.16328851312054582, + "grad_norm": 0.057373046875, + "learning_rate": 0.0018483639863524216, + "loss": 0.0659, + "step": 18811 + }, + { + "epoch": 0.16329719360074998, + "grad_norm": 0.2109375, + "learning_rate": 0.001848347452854153, + "loss": 0.1191, + "step": 18812 + }, + { + "epoch": 0.16330587408095415, + "grad_norm": 0.1240234375, + "learning_rate": 0.0018483309185375032, + "loss": 0.1553, + "step": 18813 + }, + { + "epoch": 0.16331455456115831, + "grad_norm": 0.080078125, + "learning_rate": 0.0018483143834024897, + "loss": 0.0933, + "step": 18814 + }, + { + "epoch": 0.16332323504136248, + "grad_norm": 0.07568359375, + "learning_rate": 0.0018482978474491309, + "loss": 0.0967, + "step": 18815 + }, + { + "epoch": 0.16333191552156665, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018482813106774447, + "loss": 0.1064, + "step": 18816 + }, + { + "epoch": 0.1633405960017708, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018482647730874496, + "loss": 0.1113, + "step": 18817 + }, + { + "epoch": 0.16334927648197498, + "grad_norm": 0.828125, + "learning_rate": 0.001848248234679163, + "loss": 0.1162, + "step": 18818 + }, + { + "epoch": 0.16335795696217914, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018482316954526038, + "loss": 0.0898, + "step": 18819 + }, + { + "epoch": 0.1633666374423833, + "grad_norm": 0.4375, + "learning_rate": 0.0018482151554077896, + "loss": 0.1445, + "step": 18820 + }, + { + "epoch": 0.16337531792258747, + "grad_norm": 0.7265625, + "learning_rate": 0.0018481986145447383, + "loss": 0.0825, + "step": 18821 + }, + { + "epoch": 0.16338399840279164, + "grad_norm": 1.203125, + "learning_rate": 0.0018481820728634684, + "loss": 0.1064, + "step": 18822 + }, + { + "epoch": 0.1633926788829958, + "grad_norm": 1.46875, + "learning_rate": 0.0018481655303639976, + "loss": 0.1104, + "step": 18823 + }, + { + "epoch": 0.16340135936319997, + "grad_norm": 0.2431640625, + "learning_rate": 0.0018481489870463444, + "loss": 0.126, + "step": 18824 + }, + { + "epoch": 0.16341003984340413, + "grad_norm": 0.13671875, + "learning_rate": 0.0018481324429105268, + "loss": 0.0947, + "step": 18825 + }, + { + "epoch": 0.1634187203236083, + "grad_norm": 0.79296875, + "learning_rate": 0.0018481158979565624, + "loss": 0.1602, + "step": 18826 + }, + { + "epoch": 0.16342740080381246, + "grad_norm": 0.1796875, + "learning_rate": 0.00184809935218447, + "loss": 0.0879, + "step": 18827 + }, + { + "epoch": 0.16343608128401663, + "grad_norm": 0.154296875, + "learning_rate": 0.0018480828055942677, + "loss": 0.0913, + "step": 18828 + }, + { + "epoch": 0.1634447617642208, + "grad_norm": 0.4296875, + "learning_rate": 0.0018480662581859728, + "loss": 0.1279, + "step": 18829 + }, + { + "epoch": 0.16345344224442496, + "grad_norm": 0.609375, + "learning_rate": 0.0018480497099596042, + "loss": 0.0923, + "step": 18830 + }, + { + "epoch": 0.16346212272462912, + "grad_norm": 0.56640625, + "learning_rate": 0.0018480331609151792, + "loss": 0.1426, + "step": 18831 + }, + { + "epoch": 0.1634708032048333, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018480166110527167, + "loss": 0.0913, + "step": 18832 + }, + { + "epoch": 0.16347948368503745, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018480000603722346, + "loss": 0.0791, + "step": 18833 + }, + { + "epoch": 0.16348816416524162, + "grad_norm": 0.1689453125, + "learning_rate": 0.001847983508873751, + "loss": 0.125, + "step": 18834 + }, + { + "epoch": 0.16349684464544578, + "grad_norm": 0.26171875, + "learning_rate": 0.0018479669565572834, + "loss": 0.1196, + "step": 18835 + }, + { + "epoch": 0.16350552512564995, + "grad_norm": 0.36328125, + "learning_rate": 0.0018479504034228508, + "loss": 0.0898, + "step": 18836 + }, + { + "epoch": 0.1635142056058541, + "grad_norm": 0.061767578125, + "learning_rate": 0.0018479338494704703, + "loss": 0.0781, + "step": 18837 + }, + { + "epoch": 0.16352288608605828, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018479172947001611, + "loss": 0.0845, + "step": 18838 + }, + { + "epoch": 0.16353156656626244, + "grad_norm": 0.4140625, + "learning_rate": 0.0018479007391119405, + "loss": 0.0898, + "step": 18839 + }, + { + "epoch": 0.1635402470464666, + "grad_norm": 0.32421875, + "learning_rate": 0.001847884182705827, + "loss": 0.1182, + "step": 18840 + }, + { + "epoch": 0.16354892752667077, + "grad_norm": 0.61328125, + "learning_rate": 0.0018478676254818384, + "loss": 0.1162, + "step": 18841 + }, + { + "epoch": 0.16355760800687494, + "grad_norm": 0.166015625, + "learning_rate": 0.0018478510674399934, + "loss": 0.1357, + "step": 18842 + }, + { + "epoch": 0.1635662884870791, + "grad_norm": 0.09521484375, + "learning_rate": 0.0018478345085803094, + "loss": 0.123, + "step": 18843 + }, + { + "epoch": 0.16357496896728327, + "grad_norm": 0.103515625, + "learning_rate": 0.0018478179489028052, + "loss": 0.1016, + "step": 18844 + }, + { + "epoch": 0.16358364944748743, + "grad_norm": 0.1103515625, + "learning_rate": 0.001847801388407498, + "loss": 0.1162, + "step": 18845 + }, + { + "epoch": 0.1635923299276916, + "grad_norm": 0.10400390625, + "learning_rate": 0.001847784827094407, + "loss": 0.105, + "step": 18846 + }, + { + "epoch": 0.16360101040789576, + "grad_norm": 0.2265625, + "learning_rate": 0.0018477682649635492, + "loss": 0.1113, + "step": 18847 + }, + { + "epoch": 0.16360969088809993, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018477517020149437, + "loss": 0.0996, + "step": 18848 + }, + { + "epoch": 0.1636183713683041, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018477351382486077, + "loss": 0.1045, + "step": 18849 + }, + { + "epoch": 0.16362705184850826, + "grad_norm": 0.60546875, + "learning_rate": 0.0018477185736645602, + "loss": 0.1201, + "step": 18850 + }, + { + "epoch": 0.16363573232871242, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018477020082628187, + "loss": 0.1104, + "step": 18851 + }, + { + "epoch": 0.1636444128089166, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018476854420434016, + "loss": 0.127, + "step": 18852 + }, + { + "epoch": 0.16365309328912075, + "grad_norm": 0.55859375, + "learning_rate": 0.0018476688750063272, + "loss": 0.1074, + "step": 18853 + }, + { + "epoch": 0.16366177376932492, + "grad_norm": 0.2412109375, + "learning_rate": 0.001847652307151613, + "loss": 0.1406, + "step": 18854 + }, + { + "epoch": 0.16367045424952908, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018476357384792775, + "loss": 0.1069, + "step": 18855 + }, + { + "epoch": 0.16367913472973325, + "grad_norm": 0.193359375, + "learning_rate": 0.0018476191689893386, + "loss": 0.1064, + "step": 18856 + }, + { + "epoch": 0.16368781520993742, + "grad_norm": 0.1845703125, + "learning_rate": 0.0018476025986818151, + "loss": 0.1025, + "step": 18857 + }, + { + "epoch": 0.16369649569014158, + "grad_norm": 0.177734375, + "learning_rate": 0.0018475860275567243, + "loss": 0.1172, + "step": 18858 + }, + { + "epoch": 0.16370517617034575, + "grad_norm": 0.46875, + "learning_rate": 0.0018475694556140847, + "loss": 0.106, + "step": 18859 + }, + { + "epoch": 0.1637138566505499, + "grad_norm": 0.1328125, + "learning_rate": 0.0018475528828539142, + "loss": 0.1006, + "step": 18860 + }, + { + "epoch": 0.16372253713075408, + "grad_norm": 0.5234375, + "learning_rate": 0.0018475363092762315, + "loss": 0.1328, + "step": 18861 + }, + { + "epoch": 0.16373121761095824, + "grad_norm": 0.2109375, + "learning_rate": 0.0018475197348810542, + "loss": 0.0742, + "step": 18862 + }, + { + "epoch": 0.1637398980911624, + "grad_norm": 0.318359375, + "learning_rate": 0.0018475031596684002, + "loss": 0.1318, + "step": 18863 + }, + { + "epoch": 0.16374857857136657, + "grad_norm": 0.125, + "learning_rate": 0.0018474865836382883, + "loss": 0.0996, + "step": 18864 + }, + { + "epoch": 0.16375725905157074, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018474700067907363, + "loss": 0.0942, + "step": 18865 + }, + { + "epoch": 0.1637659395317749, + "grad_norm": 0.181640625, + "learning_rate": 0.0018474534291257623, + "loss": 0.1621, + "step": 18866 + }, + { + "epoch": 0.16377462001197907, + "grad_norm": 0.3203125, + "learning_rate": 0.0018474368506433843, + "loss": 0.0713, + "step": 18867 + }, + { + "epoch": 0.16378330049218323, + "grad_norm": 0.15234375, + "learning_rate": 0.0018474202713436208, + "loss": 0.0996, + "step": 18868 + }, + { + "epoch": 0.1637919809723874, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018474036912264895, + "loss": 0.1699, + "step": 18869 + }, + { + "epoch": 0.16380066145259156, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018473871102920093, + "loss": 0.1113, + "step": 18870 + }, + { + "epoch": 0.16380934193279573, + "grad_norm": 0.326171875, + "learning_rate": 0.0018473705285401974, + "loss": 0.0898, + "step": 18871 + }, + { + "epoch": 0.1638180224129999, + "grad_norm": 0.251953125, + "learning_rate": 0.0018473539459710722, + "loss": 0.1318, + "step": 18872 + }, + { + "epoch": 0.16382670289320406, + "grad_norm": 0.265625, + "learning_rate": 0.0018473373625846519, + "loss": 0.1387, + "step": 18873 + }, + { + "epoch": 0.16383538337340822, + "grad_norm": 0.3828125, + "learning_rate": 0.001847320778380955, + "loss": 0.0986, + "step": 18874 + }, + { + "epoch": 0.1638440638536124, + "grad_norm": 0.1015625, + "learning_rate": 0.001847304193359999, + "loss": 0.105, + "step": 18875 + }, + { + "epoch": 0.16385274433381655, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018472876075218026, + "loss": 0.1318, + "step": 18876 + }, + { + "epoch": 0.16386142481402072, + "grad_norm": 0.419921875, + "learning_rate": 0.0018472710208663836, + "loss": 0.0986, + "step": 18877 + }, + { + "epoch": 0.16387010529422488, + "grad_norm": 0.73828125, + "learning_rate": 0.0018472544333937603, + "loss": 0.103, + "step": 18878 + }, + { + "epoch": 0.16387878577442905, + "grad_norm": 0.3828125, + "learning_rate": 0.0018472378451039508, + "loss": 0.0884, + "step": 18879 + }, + { + "epoch": 0.1638874662546332, + "grad_norm": 0.099609375, + "learning_rate": 0.0018472212559969732, + "loss": 0.1094, + "step": 18880 + }, + { + "epoch": 0.16389614673483738, + "grad_norm": 0.16796875, + "learning_rate": 0.0018472046660728456, + "loss": 0.1143, + "step": 18881 + }, + { + "epoch": 0.16390482721504154, + "grad_norm": 0.3515625, + "learning_rate": 0.0018471880753315865, + "loss": 0.1118, + "step": 18882 + }, + { + "epoch": 0.1639135076952457, + "grad_norm": 0.142578125, + "learning_rate": 0.0018471714837732132, + "loss": 0.1719, + "step": 18883 + }, + { + "epoch": 0.16392218817544987, + "grad_norm": 0.19921875, + "learning_rate": 0.0018471548913977448, + "loss": 0.1357, + "step": 18884 + }, + { + "epoch": 0.16393086865565404, + "grad_norm": 0.3125, + "learning_rate": 0.001847138298205199, + "loss": 0.1064, + "step": 18885 + }, + { + "epoch": 0.1639395491358582, + "grad_norm": 0.326171875, + "learning_rate": 0.001847121704195594, + "loss": 0.1113, + "step": 18886 + }, + { + "epoch": 0.16394822961606237, + "grad_norm": 0.36328125, + "learning_rate": 0.0018471051093689476, + "loss": 0.1128, + "step": 18887 + }, + { + "epoch": 0.16395691009626653, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018470885137252787, + "loss": 0.1104, + "step": 18888 + }, + { + "epoch": 0.1639655905764707, + "grad_norm": 0.21875, + "learning_rate": 0.0018470719172646046, + "loss": 0.0791, + "step": 18889 + }, + { + "epoch": 0.16397427105667486, + "grad_norm": 0.130859375, + "learning_rate": 0.0018470553199869443, + "loss": 0.1016, + "step": 18890 + }, + { + "epoch": 0.16398295153687903, + "grad_norm": 0.15234375, + "learning_rate": 0.0018470387218923153, + "loss": 0.0996, + "step": 18891 + }, + { + "epoch": 0.1639916320170832, + "grad_norm": 0.44140625, + "learning_rate": 0.0018470221229807359, + "loss": 0.0986, + "step": 18892 + }, + { + "epoch": 0.16400031249728736, + "grad_norm": 0.220703125, + "learning_rate": 0.0018470055232522244, + "loss": 0.0957, + "step": 18893 + }, + { + "epoch": 0.16400899297749152, + "grad_norm": 0.55859375, + "learning_rate": 0.0018469889227067991, + "loss": 0.1094, + "step": 18894 + }, + { + "epoch": 0.1640176734576957, + "grad_norm": 0.21875, + "learning_rate": 0.0018469723213444778, + "loss": 0.1289, + "step": 18895 + }, + { + "epoch": 0.16402635393789985, + "grad_norm": 0.150390625, + "learning_rate": 0.001846955719165279, + "loss": 0.1758, + "step": 18896 + }, + { + "epoch": 0.16403503441810402, + "grad_norm": 0.1513671875, + "learning_rate": 0.00184693911616922, + "loss": 0.1113, + "step": 18897 + }, + { + "epoch": 0.16404371489830818, + "grad_norm": 0.359375, + "learning_rate": 0.0018469225123563203, + "loss": 0.0869, + "step": 18898 + }, + { + "epoch": 0.16405239537851235, + "grad_norm": 0.57421875, + "learning_rate": 0.001846905907726597, + "loss": 0.0879, + "step": 18899 + }, + { + "epoch": 0.16406107585871652, + "grad_norm": 0.28125, + "learning_rate": 0.0018468893022800686, + "loss": 0.0981, + "step": 18900 + }, + { + "epoch": 0.16406975633892068, + "grad_norm": 0.625, + "learning_rate": 0.0018468726960167534, + "loss": 0.1279, + "step": 18901 + }, + { + "epoch": 0.16407843681912485, + "grad_norm": 0.3203125, + "learning_rate": 0.0018468560889366695, + "loss": 0.1504, + "step": 18902 + }, + { + "epoch": 0.164087117299329, + "grad_norm": 0.421875, + "learning_rate": 0.0018468394810398348, + "loss": 0.0898, + "step": 18903 + }, + { + "epoch": 0.16409579777953318, + "grad_norm": 0.30859375, + "learning_rate": 0.0018468228723262676, + "loss": 0.0835, + "step": 18904 + }, + { + "epoch": 0.16410447825973734, + "grad_norm": 0.107421875, + "learning_rate": 0.0018468062627959865, + "loss": 0.0664, + "step": 18905 + }, + { + "epoch": 0.1641131587399415, + "grad_norm": 0.359375, + "learning_rate": 0.001846789652449009, + "loss": 0.1094, + "step": 18906 + }, + { + "epoch": 0.16412183922014567, + "grad_norm": 0.54296875, + "learning_rate": 0.001846773041285354, + "loss": 0.1064, + "step": 18907 + }, + { + "epoch": 0.16413051970034984, + "grad_norm": 0.291015625, + "learning_rate": 0.0018467564293050385, + "loss": 0.1289, + "step": 18908 + }, + { + "epoch": 0.164139200180554, + "grad_norm": 0.2041015625, + "learning_rate": 0.001846739816508082, + "loss": 0.1216, + "step": 18909 + }, + { + "epoch": 0.16414788066075817, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018467232028945015, + "loss": 0.125, + "step": 18910 + }, + { + "epoch": 0.16415656114096233, + "grad_norm": 0.1796875, + "learning_rate": 0.0018467065884643162, + "loss": 0.1123, + "step": 18911 + }, + { + "epoch": 0.1641652416211665, + "grad_norm": 0.283203125, + "learning_rate": 0.0018466899732175438, + "loss": 0.0996, + "step": 18912 + }, + { + "epoch": 0.16417392210137063, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018466733571542023, + "loss": 0.1006, + "step": 18913 + }, + { + "epoch": 0.1641826025815748, + "grad_norm": 0.3671875, + "learning_rate": 0.0018466567402743103, + "loss": 0.1309, + "step": 18914 + }, + { + "epoch": 0.16419128306177896, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018466401225778852, + "loss": 0.1377, + "step": 18915 + }, + { + "epoch": 0.16419996354198313, + "grad_norm": 0.33984375, + "learning_rate": 0.0018466235040649457, + "loss": 0.1055, + "step": 18916 + }, + { + "epoch": 0.1642086440221873, + "grad_norm": 0.388671875, + "learning_rate": 0.0018466068847355103, + "loss": 0.1016, + "step": 18917 + }, + { + "epoch": 0.16421732450239146, + "grad_norm": 0.1923828125, + "learning_rate": 0.001846590264589597, + "loss": 0.1172, + "step": 18918 + }, + { + "epoch": 0.16422600498259562, + "grad_norm": 0.96484375, + "learning_rate": 0.0018465736436272231, + "loss": 0.1162, + "step": 18919 + }, + { + "epoch": 0.1642346854627998, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018465570218484083, + "loss": 0.1138, + "step": 18920 + }, + { + "epoch": 0.16424336594300395, + "grad_norm": 0.875, + "learning_rate": 0.0018465403992531694, + "loss": 0.1416, + "step": 18921 + }, + { + "epoch": 0.16425204642320812, + "grad_norm": 0.318359375, + "learning_rate": 0.0018465237758415252, + "loss": 0.1211, + "step": 18922 + }, + { + "epoch": 0.16426072690341229, + "grad_norm": 0.07421875, + "learning_rate": 0.001846507151613494, + "loss": 0.0835, + "step": 18923 + }, + { + "epoch": 0.16426940738361645, + "grad_norm": 0.1572265625, + "learning_rate": 0.001846490526569094, + "loss": 0.0869, + "step": 18924 + }, + { + "epoch": 0.16427808786382062, + "grad_norm": 0.322265625, + "learning_rate": 0.0018464739007083427, + "loss": 0.1084, + "step": 18925 + }, + { + "epoch": 0.16428676834402478, + "grad_norm": 0.953125, + "learning_rate": 0.001846457274031259, + "loss": 0.1348, + "step": 18926 + }, + { + "epoch": 0.16429544882422895, + "grad_norm": 0.671875, + "learning_rate": 0.0018464406465378605, + "loss": 0.1152, + "step": 18927 + }, + { + "epoch": 0.1643041293044331, + "grad_norm": 0.35546875, + "learning_rate": 0.001846424018228166, + "loss": 0.1182, + "step": 18928 + }, + { + "epoch": 0.16431280978463728, + "grad_norm": 0.365234375, + "learning_rate": 0.0018464073891021936, + "loss": 0.0796, + "step": 18929 + }, + { + "epoch": 0.16432149026484144, + "grad_norm": 0.294921875, + "learning_rate": 0.0018463907591599609, + "loss": 0.0923, + "step": 18930 + }, + { + "epoch": 0.1643301707450456, + "grad_norm": 0.1005859375, + "learning_rate": 0.0018463741284014867, + "loss": 0.1143, + "step": 18931 + }, + { + "epoch": 0.16433885122524977, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018463574968267892, + "loss": 0.1406, + "step": 18932 + }, + { + "epoch": 0.16434753170545394, + "grad_norm": 0.228515625, + "learning_rate": 0.001846340864435886, + "loss": 0.1191, + "step": 18933 + }, + { + "epoch": 0.1643562121856581, + "grad_norm": 0.43359375, + "learning_rate": 0.0018463242312287958, + "loss": 0.1875, + "step": 18934 + }, + { + "epoch": 0.16436489266586227, + "grad_norm": 0.3203125, + "learning_rate": 0.0018463075972055365, + "loss": 0.1582, + "step": 18935 + }, + { + "epoch": 0.16437357314606643, + "grad_norm": 0.54296875, + "learning_rate": 0.0018462909623661267, + "loss": 0.1167, + "step": 18936 + }, + { + "epoch": 0.1643822536262706, + "grad_norm": 0.1630859375, + "learning_rate": 0.001846274326710584, + "loss": 0.1084, + "step": 18937 + }, + { + "epoch": 0.16439093410647476, + "grad_norm": 0.07373046875, + "learning_rate": 0.001846257690238927, + "loss": 0.0806, + "step": 18938 + }, + { + "epoch": 0.16439961458667893, + "grad_norm": 0.11083984375, + "learning_rate": 0.001846241052951174, + "loss": 0.0781, + "step": 18939 + }, + { + "epoch": 0.1644082950668831, + "grad_norm": 0.212890625, + "learning_rate": 0.0018462244148473425, + "loss": 0.1191, + "step": 18940 + }, + { + "epoch": 0.16441697554708726, + "grad_norm": 0.337890625, + "learning_rate": 0.0018462077759274517, + "loss": 0.1309, + "step": 18941 + }, + { + "epoch": 0.16442565602729142, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018461911361915193, + "loss": 0.1211, + "step": 18942 + }, + { + "epoch": 0.1644343365074956, + "grad_norm": 0.7421875, + "learning_rate": 0.0018461744956395632, + "loss": 0.125, + "step": 18943 + }, + { + "epoch": 0.16444301698769975, + "grad_norm": 0.2275390625, + "learning_rate": 0.001846157854271602, + "loss": 0.084, + "step": 18944 + }, + { + "epoch": 0.16445169746790392, + "grad_norm": 0.388671875, + "learning_rate": 0.001846141212087654, + "loss": 0.1406, + "step": 18945 + }, + { + "epoch": 0.16446037794810808, + "grad_norm": 0.78125, + "learning_rate": 0.0018461245690877366, + "loss": 0.1064, + "step": 18946 + }, + { + "epoch": 0.16446905842831225, + "grad_norm": 0.70703125, + "learning_rate": 0.001846107925271869, + "loss": 0.1504, + "step": 18947 + }, + { + "epoch": 0.1644777389085164, + "grad_norm": 0.2373046875, + "learning_rate": 0.001846091280640069, + "loss": 0.0986, + "step": 18948 + }, + { + "epoch": 0.16448641938872058, + "grad_norm": 0.1201171875, + "learning_rate": 0.0018460746351923547, + "loss": 0.1128, + "step": 18949 + }, + { + "epoch": 0.16449509986892474, + "grad_norm": 0.365234375, + "learning_rate": 0.0018460579889287444, + "loss": 0.1494, + "step": 18950 + }, + { + "epoch": 0.1645037803491289, + "grad_norm": 0.220703125, + "learning_rate": 0.0018460413418492562, + "loss": 0.1211, + "step": 18951 + }, + { + "epoch": 0.16451246082933307, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018460246939539086, + "loss": 0.0752, + "step": 18952 + }, + { + "epoch": 0.16452114130953724, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018460080452427192, + "loss": 0.1748, + "step": 18953 + }, + { + "epoch": 0.1645298217897414, + "grad_norm": 0.50390625, + "learning_rate": 0.0018459913957157071, + "loss": 0.1001, + "step": 18954 + }, + { + "epoch": 0.16453850226994557, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018459747453728897, + "loss": 0.0933, + "step": 18955 + }, + { + "epoch": 0.16454718275014973, + "grad_norm": 0.43359375, + "learning_rate": 0.0018459580942142857, + "loss": 0.1328, + "step": 18956 + }, + { + "epoch": 0.1645558632303539, + "grad_norm": 0.244140625, + "learning_rate": 0.001845941442239913, + "loss": 0.0918, + "step": 18957 + }, + { + "epoch": 0.16456454371055806, + "grad_norm": 0.322265625, + "learning_rate": 0.0018459247894497901, + "loss": 0.0874, + "step": 18958 + }, + { + "epoch": 0.16457322419076223, + "grad_norm": 0.310546875, + "learning_rate": 0.0018459081358439351, + "loss": 0.0928, + "step": 18959 + }, + { + "epoch": 0.1645819046709664, + "grad_norm": 1.703125, + "learning_rate": 0.001845891481422366, + "loss": 0.293, + "step": 18960 + }, + { + "epoch": 0.16459058515117056, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018458748261851012, + "loss": 0.1211, + "step": 18961 + }, + { + "epoch": 0.16459926563137472, + "grad_norm": 0.421875, + "learning_rate": 0.0018458581701321591, + "loss": 0.1221, + "step": 18962 + }, + { + "epoch": 0.1646079461115789, + "grad_norm": 0.52734375, + "learning_rate": 0.0018458415132635573, + "loss": 0.1191, + "step": 18963 + }, + { + "epoch": 0.16461662659178306, + "grad_norm": 0.5, + "learning_rate": 0.0018458248555793147, + "loss": 0.0918, + "step": 18964 + }, + { + "epoch": 0.16462530707198722, + "grad_norm": 0.13671875, + "learning_rate": 0.0018458081970794491, + "loss": 0.1699, + "step": 18965 + }, + { + "epoch": 0.16463398755219139, + "grad_norm": 0.1328125, + "learning_rate": 0.001845791537763979, + "loss": 0.0918, + "step": 18966 + }, + { + "epoch": 0.16464266803239555, + "grad_norm": 0.203125, + "learning_rate": 0.0018457748776329226, + "loss": 0.0938, + "step": 18967 + }, + { + "epoch": 0.16465134851259972, + "grad_norm": 0.482421875, + "learning_rate": 0.0018457582166862979, + "loss": 0.125, + "step": 18968 + }, + { + "epoch": 0.16466002899280388, + "grad_norm": 0.384765625, + "learning_rate": 0.0018457415549241233, + "loss": 0.1455, + "step": 18969 + }, + { + "epoch": 0.16466870947300805, + "grad_norm": 0.81640625, + "learning_rate": 0.0018457248923464168, + "loss": 0.1211, + "step": 18970 + }, + { + "epoch": 0.1646773899532122, + "grad_norm": 0.65625, + "learning_rate": 0.0018457082289531966, + "loss": 0.1035, + "step": 18971 + }, + { + "epoch": 0.16468607043341638, + "grad_norm": 0.55078125, + "learning_rate": 0.0018456915647444812, + "loss": 0.3848, + "step": 18972 + }, + { + "epoch": 0.16469475091362054, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018456748997202888, + "loss": 0.1104, + "step": 18973 + }, + { + "epoch": 0.1647034313938247, + "grad_norm": 0.251953125, + "learning_rate": 0.0018456582338806375, + "loss": 0.0928, + "step": 18974 + }, + { + "epoch": 0.16471211187402887, + "grad_norm": 0.384765625, + "learning_rate": 0.0018456415672255457, + "loss": 0.0903, + "step": 18975 + }, + { + "epoch": 0.16472079235423304, + "grad_norm": 0.34375, + "learning_rate": 0.001845624899755031, + "loss": 0.1104, + "step": 18976 + }, + { + "epoch": 0.1647294728344372, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018456082314691126, + "loss": 0.1221, + "step": 18977 + }, + { + "epoch": 0.16473815331464137, + "grad_norm": 0.2373046875, + "learning_rate": 0.001845591562367808, + "loss": 0.1201, + "step": 18978 + }, + { + "epoch": 0.16474683379484553, + "grad_norm": 0.1455078125, + "learning_rate": 0.001845574892451136, + "loss": 0.1123, + "step": 18979 + }, + { + "epoch": 0.1647555142750497, + "grad_norm": 0.13671875, + "learning_rate": 0.001845558221719114, + "loss": 0.1396, + "step": 18980 + }, + { + "epoch": 0.16476419475525386, + "grad_norm": 0.65625, + "learning_rate": 0.0018455415501717613, + "loss": 0.0796, + "step": 18981 + }, + { + "epoch": 0.16477287523545803, + "grad_norm": 0.51171875, + "learning_rate": 0.0018455248778090952, + "loss": 0.1299, + "step": 18982 + }, + { + "epoch": 0.1647815557156622, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018455082046311343, + "loss": 0.1172, + "step": 18983 + }, + { + "epoch": 0.16479023619586636, + "grad_norm": 0.29296875, + "learning_rate": 0.001845491530637897, + "loss": 0.1562, + "step": 18984 + }, + { + "epoch": 0.16479891667607052, + "grad_norm": 0.79296875, + "learning_rate": 0.0018454748558294012, + "loss": 0.1348, + "step": 18985 + }, + { + "epoch": 0.1648075971562747, + "grad_norm": 1.0078125, + "learning_rate": 0.0018454581802056654, + "loss": 0.1533, + "step": 18986 + }, + { + "epoch": 0.16481627763647885, + "grad_norm": 0.0751953125, + "learning_rate": 0.0018454415037667076, + "loss": 0.0771, + "step": 18987 + }, + { + "epoch": 0.16482495811668302, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018454248265125461, + "loss": 0.0908, + "step": 18988 + }, + { + "epoch": 0.16483363859688718, + "grad_norm": 0.4375, + "learning_rate": 0.0018454081484431996, + "loss": 0.123, + "step": 18989 + }, + { + "epoch": 0.16484231907709135, + "grad_norm": 0.53125, + "learning_rate": 0.0018453914695586856, + "loss": 0.1338, + "step": 18990 + }, + { + "epoch": 0.1648509995572955, + "grad_norm": 1.5390625, + "learning_rate": 0.0018453747898590227, + "loss": 0.0908, + "step": 18991 + }, + { + "epoch": 0.16485968003749968, + "grad_norm": 0.57421875, + "learning_rate": 0.0018453581093442293, + "loss": 0.1406, + "step": 18992 + }, + { + "epoch": 0.16486836051770384, + "grad_norm": 0.38671875, + "learning_rate": 0.0018453414280143238, + "loss": 0.1099, + "step": 18993 + }, + { + "epoch": 0.164877040997908, + "grad_norm": 0.65625, + "learning_rate": 0.0018453247458693235, + "loss": 0.124, + "step": 18994 + }, + { + "epoch": 0.16488572147811217, + "grad_norm": 0.09765625, + "learning_rate": 0.0018453080629092472, + "loss": 0.1221, + "step": 18995 + }, + { + "epoch": 0.16489440195831634, + "grad_norm": 1.109375, + "learning_rate": 0.0018452913791341136, + "loss": 0.1309, + "step": 18996 + }, + { + "epoch": 0.1649030824385205, + "grad_norm": 0.2275390625, + "learning_rate": 0.0018452746945439402, + "loss": 0.1123, + "step": 18997 + }, + { + "epoch": 0.16491176291872467, + "grad_norm": 0.109375, + "learning_rate": 0.001845258009138746, + "loss": 0.1357, + "step": 18998 + }, + { + "epoch": 0.16492044339892883, + "grad_norm": 0.146484375, + "learning_rate": 0.0018452413229185487, + "loss": 0.1133, + "step": 18999 + }, + { + "epoch": 0.164929123879133, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018452246358833667, + "loss": 0.1582, + "step": 19000 + }, + { + "epoch": 0.16493780435933716, + "grad_norm": 0.7734375, + "learning_rate": 0.001845207948033218, + "loss": 0.3594, + "step": 19001 + }, + { + "epoch": 0.16494648483954133, + "grad_norm": 0.2275390625, + "learning_rate": 0.0018451912593681213, + "loss": 0.1367, + "step": 19002 + }, + { + "epoch": 0.1649551653197455, + "grad_norm": 0.51171875, + "learning_rate": 0.0018451745698880946, + "loss": 0.0771, + "step": 19003 + }, + { + "epoch": 0.16496384579994966, + "grad_norm": 0.09423828125, + "learning_rate": 0.0018451578795931564, + "loss": 0.1191, + "step": 19004 + }, + { + "epoch": 0.16497252628015383, + "grad_norm": 0.75, + "learning_rate": 0.0018451411884833242, + "loss": 0.1152, + "step": 19005 + }, + { + "epoch": 0.164981206760358, + "grad_norm": 0.208984375, + "learning_rate": 0.0018451244965586172, + "loss": 0.1074, + "step": 19006 + }, + { + "epoch": 0.16498988724056216, + "grad_norm": 0.447265625, + "learning_rate": 0.001845107803819053, + "loss": 0.126, + "step": 19007 + }, + { + "epoch": 0.16499856772076632, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018450911102646504, + "loss": 0.1396, + "step": 19008 + }, + { + "epoch": 0.16500724820097049, + "grad_norm": 0.69140625, + "learning_rate": 0.0018450744158954272, + "loss": 0.1006, + "step": 19009 + }, + { + "epoch": 0.16501592868117465, + "grad_norm": 0.30859375, + "learning_rate": 0.0018450577207114018, + "loss": 0.1069, + "step": 19010 + }, + { + "epoch": 0.16502460916137882, + "grad_norm": 0.6484375, + "learning_rate": 0.0018450410247125924, + "loss": 0.1426, + "step": 19011 + }, + { + "epoch": 0.16503328964158298, + "grad_norm": 0.263671875, + "learning_rate": 0.0018450243278990176, + "loss": 0.1533, + "step": 19012 + }, + { + "epoch": 0.16504197012178715, + "grad_norm": 0.119140625, + "learning_rate": 0.001845007630270695, + "loss": 0.1074, + "step": 19013 + }, + { + "epoch": 0.1650506506019913, + "grad_norm": 0.66796875, + "learning_rate": 0.0018449909318276436, + "loss": 0.1416, + "step": 19014 + }, + { + "epoch": 0.16505933108219548, + "grad_norm": 0.12451171875, + "learning_rate": 0.001844974232569881, + "loss": 0.1133, + "step": 19015 + }, + { + "epoch": 0.16506801156239964, + "grad_norm": 0.765625, + "learning_rate": 0.001844957532497426, + "loss": 0.1426, + "step": 19016 + }, + { + "epoch": 0.1650766920426038, + "grad_norm": 0.2197265625, + "learning_rate": 0.001844940831610297, + "loss": 0.1533, + "step": 19017 + }, + { + "epoch": 0.16508537252280797, + "grad_norm": 0.06884765625, + "learning_rate": 0.0018449241299085114, + "loss": 0.0771, + "step": 19018 + }, + { + "epoch": 0.16509405300301214, + "grad_norm": 0.09765625, + "learning_rate": 0.0018449074273920877, + "loss": 0.1094, + "step": 19019 + }, + { + "epoch": 0.1651027334832163, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018448907240610451, + "loss": 0.1123, + "step": 19020 + }, + { + "epoch": 0.16511141396342047, + "grad_norm": 0.494140625, + "learning_rate": 0.0018448740199154005, + "loss": 0.126, + "step": 19021 + }, + { + "epoch": 0.16512009444362463, + "grad_norm": 0.205078125, + "learning_rate": 0.0018448573149551734, + "loss": 0.1016, + "step": 19022 + }, + { + "epoch": 0.1651287749238288, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018448406091803814, + "loss": 0.1279, + "step": 19023 + }, + { + "epoch": 0.16513745540403296, + "grad_norm": 0.1689453125, + "learning_rate": 0.001844823902591043, + "loss": 0.1465, + "step": 19024 + }, + { + "epoch": 0.16514613588423713, + "grad_norm": 0.283203125, + "learning_rate": 0.001844807195187176, + "loss": 0.1182, + "step": 19025 + }, + { + "epoch": 0.1651548163644413, + "grad_norm": 0.126953125, + "learning_rate": 0.0018447904869687993, + "loss": 0.0952, + "step": 19026 + }, + { + "epoch": 0.16516349684464546, + "grad_norm": 0.1025390625, + "learning_rate": 0.001844773777935931, + "loss": 0.1016, + "step": 19027 + }, + { + "epoch": 0.16517217732484962, + "grad_norm": 0.1708984375, + "learning_rate": 0.001844757068088589, + "loss": 0.1201, + "step": 19028 + }, + { + "epoch": 0.1651808578050538, + "grad_norm": 0.353515625, + "learning_rate": 0.0018447403574267923, + "loss": 0.1064, + "step": 19029 + }, + { + "epoch": 0.16518953828525795, + "grad_norm": 0.240234375, + "learning_rate": 0.0018447236459505584, + "loss": 0.1289, + "step": 19030 + }, + { + "epoch": 0.16519821876546212, + "grad_norm": 0.1025390625, + "learning_rate": 0.001844706933659906, + "loss": 0.085, + "step": 19031 + }, + { + "epoch": 0.16520689924566628, + "grad_norm": 0.3125, + "learning_rate": 0.001844690220554853, + "loss": 0.1006, + "step": 19032 + }, + { + "epoch": 0.16521557972587045, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018446735066354185, + "loss": 0.0889, + "step": 19033 + }, + { + "epoch": 0.1652242602060746, + "grad_norm": 0.08642578125, + "learning_rate": 0.0018446567919016197, + "loss": 0.0977, + "step": 19034 + }, + { + "epoch": 0.16523294068627878, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018446400763534759, + "loss": 0.1074, + "step": 19035 + }, + { + "epoch": 0.16524162116648292, + "grad_norm": 0.1484375, + "learning_rate": 0.0018446233599910048, + "loss": 0.1133, + "step": 19036 + }, + { + "epoch": 0.16525030164668708, + "grad_norm": 0.203125, + "learning_rate": 0.0018446066428142247, + "loss": 0.126, + "step": 19037 + }, + { + "epoch": 0.16525898212689125, + "grad_norm": 0.328125, + "learning_rate": 0.0018445899248231536, + "loss": 0.1113, + "step": 19038 + }, + { + "epoch": 0.1652676626070954, + "grad_norm": 0.625, + "learning_rate": 0.0018445732060178106, + "loss": 0.1523, + "step": 19039 + }, + { + "epoch": 0.16527634308729958, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018445564863982134, + "loss": 0.0879, + "step": 19040 + }, + { + "epoch": 0.16528502356750374, + "grad_norm": 0.25390625, + "learning_rate": 0.0018445397659643805, + "loss": 0.1279, + "step": 19041 + }, + { + "epoch": 0.1652937040477079, + "grad_norm": 0.25390625, + "learning_rate": 0.00184452304471633, + "loss": 0.123, + "step": 19042 + }, + { + "epoch": 0.16530238452791207, + "grad_norm": 0.291015625, + "learning_rate": 0.0018445063226540804, + "loss": 0.0981, + "step": 19043 + }, + { + "epoch": 0.16531106500811624, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018444895997776496, + "loss": 0.1279, + "step": 19044 + }, + { + "epoch": 0.1653197454883204, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018444728760870564, + "loss": 0.127, + "step": 19045 + }, + { + "epoch": 0.16532842596852457, + "grad_norm": 0.162109375, + "learning_rate": 0.001844456151582319, + "loss": 0.1182, + "step": 19046 + }, + { + "epoch": 0.16533710644872873, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018444394262634549, + "loss": 0.0811, + "step": 19047 + }, + { + "epoch": 0.1653457869289329, + "grad_norm": 0.10302734375, + "learning_rate": 0.0018444227001304836, + "loss": 0.0957, + "step": 19048 + }, + { + "epoch": 0.16535446740913706, + "grad_norm": 0.130859375, + "learning_rate": 0.0018444059731834229, + "loss": 0.1006, + "step": 19049 + }, + { + "epoch": 0.16536314788934123, + "grad_norm": 0.267578125, + "learning_rate": 0.0018443892454222907, + "loss": 0.0947, + "step": 19050 + }, + { + "epoch": 0.1653718283695454, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018443725168471054, + "loss": 0.1553, + "step": 19051 + }, + { + "epoch": 0.16538050884974956, + "grad_norm": 0.5, + "learning_rate": 0.001844355787457886, + "loss": 0.1133, + "step": 19052 + }, + { + "epoch": 0.16538918932995372, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018443390572546502, + "loss": 0.1191, + "step": 19053 + }, + { + "epoch": 0.1653978698101579, + "grad_norm": 0.08349609375, + "learning_rate": 0.0018443223262374164, + "loss": 0.1128, + "step": 19054 + }, + { + "epoch": 0.16540655029036205, + "grad_norm": 0.33203125, + "learning_rate": 0.0018443055944062025, + "loss": 0.1201, + "step": 19055 + }, + { + "epoch": 0.16541523077056622, + "grad_norm": 1.078125, + "learning_rate": 0.0018442888617610277, + "loss": 0.0986, + "step": 19056 + }, + { + "epoch": 0.16542391125077038, + "grad_norm": 0.345703125, + "learning_rate": 0.0018442721283019093, + "loss": 0.1299, + "step": 19057 + }, + { + "epoch": 0.16543259173097455, + "grad_norm": 0.29296875, + "learning_rate": 0.0018442553940288663, + "loss": 0.0996, + "step": 19058 + }, + { + "epoch": 0.1654412722111787, + "grad_norm": 0.322265625, + "learning_rate": 0.001844238658941917, + "loss": 0.0986, + "step": 19059 + }, + { + "epoch": 0.16544995269138288, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018442219230410792, + "loss": 0.0869, + "step": 19060 + }, + { + "epoch": 0.16545863317158704, + "grad_norm": 0.58203125, + "learning_rate": 0.0018442051863263719, + "loss": 0.1221, + "step": 19061 + }, + { + "epoch": 0.1654673136517912, + "grad_norm": 0.296875, + "learning_rate": 0.0018441884487978124, + "loss": 0.1118, + "step": 19062 + }, + { + "epoch": 0.16547599413199537, + "grad_norm": 0.095703125, + "learning_rate": 0.0018441717104554199, + "loss": 0.0991, + "step": 19063 + }, + { + "epoch": 0.16548467461219954, + "grad_norm": 0.345703125, + "learning_rate": 0.0018441549712992123, + "loss": 0.127, + "step": 19064 + }, + { + "epoch": 0.1654933550924037, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018441382313292081, + "loss": 0.1108, + "step": 19065 + }, + { + "epoch": 0.16550203557260787, + "grad_norm": 0.095703125, + "learning_rate": 0.0018441214905454255, + "loss": 0.1074, + "step": 19066 + }, + { + "epoch": 0.16551071605281203, + "grad_norm": 0.6171875, + "learning_rate": 0.001844104748947883, + "loss": 0.1191, + "step": 19067 + }, + { + "epoch": 0.1655193965330162, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018440880065365983, + "loss": 0.1748, + "step": 19068 + }, + { + "epoch": 0.16552807701322036, + "grad_norm": 0.1728515625, + "learning_rate": 0.00184407126331159, + "loss": 0.1377, + "step": 19069 + }, + { + "epoch": 0.16553675749342453, + "grad_norm": 0.177734375, + "learning_rate": 0.001844054519272877, + "loss": 0.1016, + "step": 19070 + }, + { + "epoch": 0.1655454379736287, + "grad_norm": 0.193359375, + "learning_rate": 0.0018440377744204772, + "loss": 0.0996, + "step": 19071 + }, + { + "epoch": 0.16555411845383286, + "grad_norm": 0.32421875, + "learning_rate": 0.0018440210287544085, + "loss": 0.1689, + "step": 19072 + }, + { + "epoch": 0.16556279893403703, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018440042822746902, + "loss": 0.124, + "step": 19073 + }, + { + "epoch": 0.1655714794142412, + "grad_norm": 0.66015625, + "learning_rate": 0.0018439875349813394, + "loss": 0.1328, + "step": 19074 + }, + { + "epoch": 0.16558015989444536, + "grad_norm": 0.16015625, + "learning_rate": 0.0018439707868743755, + "loss": 0.123, + "step": 19075 + }, + { + "epoch": 0.16558884037464952, + "grad_norm": 0.326171875, + "learning_rate": 0.0018439540379538157, + "loss": 0.104, + "step": 19076 + }, + { + "epoch": 0.16559752085485369, + "grad_norm": 0.57421875, + "learning_rate": 0.0018439372882196794, + "loss": 0.1084, + "step": 19077 + }, + { + "epoch": 0.16560620133505785, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018439205376719846, + "loss": 0.1309, + "step": 19078 + }, + { + "epoch": 0.16561488181526202, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018439037863107488, + "loss": 0.1211, + "step": 19079 + }, + { + "epoch": 0.16562356229546618, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018438870341359916, + "loss": 0.1162, + "step": 19080 + }, + { + "epoch": 0.16563224277567035, + "grad_norm": 0.3203125, + "learning_rate": 0.0018438702811477304, + "loss": 0.1377, + "step": 19081 + }, + { + "epoch": 0.1656409232558745, + "grad_norm": 0.0869140625, + "learning_rate": 0.001843853527345984, + "loss": 0.1216, + "step": 19082 + }, + { + "epoch": 0.16564960373607868, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018438367727307706, + "loss": 0.1318, + "step": 19083 + }, + { + "epoch": 0.16565828421628284, + "grad_norm": 0.15625, + "learning_rate": 0.0018438200173021085, + "loss": 0.0986, + "step": 19084 + }, + { + "epoch": 0.165666964696487, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018438032610600158, + "loss": 0.0889, + "step": 19085 + }, + { + "epoch": 0.16567564517669117, + "grad_norm": 0.265625, + "learning_rate": 0.0018437865040045111, + "loss": 0.1992, + "step": 19086 + }, + { + "epoch": 0.16568432565689534, + "grad_norm": 1.8125, + "learning_rate": 0.0018437697461356126, + "loss": 0.0972, + "step": 19087 + }, + { + "epoch": 0.1656930061370995, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018437529874533389, + "loss": 0.1465, + "step": 19088 + }, + { + "epoch": 0.16570168661730367, + "grad_norm": 0.2080078125, + "learning_rate": 0.001843736227957708, + "loss": 0.1035, + "step": 19089 + }, + { + "epoch": 0.16571036709750783, + "grad_norm": 0.08349609375, + "learning_rate": 0.0018437194676487383, + "loss": 0.0986, + "step": 19090 + }, + { + "epoch": 0.165719047577712, + "grad_norm": 0.265625, + "learning_rate": 0.0018437027065264483, + "loss": 0.1191, + "step": 19091 + }, + { + "epoch": 0.16572772805791616, + "grad_norm": 0.3125, + "learning_rate": 0.001843685944590856, + "loss": 0.1309, + "step": 19092 + }, + { + "epoch": 0.16573640853812033, + "grad_norm": 0.33984375, + "learning_rate": 0.0018436691818419799, + "loss": 0.103, + "step": 19093 + }, + { + "epoch": 0.1657450890183245, + "grad_norm": 0.33984375, + "learning_rate": 0.0018436524182798384, + "loss": 0.1064, + "step": 19094 + }, + { + "epoch": 0.16575376949852866, + "grad_norm": 0.171875, + "learning_rate": 0.0018436356539044498, + "loss": 0.1143, + "step": 19095 + }, + { + "epoch": 0.16576244997873282, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018436188887158325, + "loss": 0.0952, + "step": 19096 + }, + { + "epoch": 0.165771130458937, + "grad_norm": 0.173828125, + "learning_rate": 0.0018436021227140047, + "loss": 0.0859, + "step": 19097 + }, + { + "epoch": 0.16577981093914115, + "grad_norm": 0.263671875, + "learning_rate": 0.001843585355898985, + "loss": 0.1157, + "step": 19098 + }, + { + "epoch": 0.16578849141934532, + "grad_norm": 0.37890625, + "learning_rate": 0.0018435685882707913, + "loss": 0.1123, + "step": 19099 + }, + { + "epoch": 0.16579717189954948, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018435518198294423, + "loss": 0.1079, + "step": 19100 + }, + { + "epoch": 0.16580585237975365, + "grad_norm": 0.244140625, + "learning_rate": 0.001843535050574956, + "loss": 0.0879, + "step": 19101 + }, + { + "epoch": 0.16581453285995781, + "grad_norm": 0.34375, + "learning_rate": 0.001843518280507351, + "loss": 0.0815, + "step": 19102 + }, + { + "epoch": 0.16582321334016198, + "grad_norm": 0.080078125, + "learning_rate": 0.0018435015096266456, + "loss": 0.0889, + "step": 19103 + }, + { + "epoch": 0.16583189382036614, + "grad_norm": 0.462890625, + "learning_rate": 0.0018434847379328583, + "loss": 0.1055, + "step": 19104 + }, + { + "epoch": 0.1658405743005703, + "grad_norm": 0.1513671875, + "learning_rate": 0.001843467965426007, + "loss": 0.1094, + "step": 19105 + }, + { + "epoch": 0.16584925478077447, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018434511921061104, + "loss": 0.0752, + "step": 19106 + }, + { + "epoch": 0.16585793526097864, + "grad_norm": 0.11962890625, + "learning_rate": 0.0018434344179731868, + "loss": 0.1104, + "step": 19107 + }, + { + "epoch": 0.1658666157411828, + "grad_norm": 0.59765625, + "learning_rate": 0.0018434176430272545, + "loss": 0.0854, + "step": 19108 + }, + { + "epoch": 0.16587529622138697, + "grad_norm": 0.7109375, + "learning_rate": 0.0018434008672683316, + "loss": 0.1514, + "step": 19109 + }, + { + "epoch": 0.16588397670159113, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018433840906964371, + "loss": 0.1338, + "step": 19110 + }, + { + "epoch": 0.1658926571817953, + "grad_norm": 0.369140625, + "learning_rate": 0.0018433673133115885, + "loss": 0.1211, + "step": 19111 + }, + { + "epoch": 0.16590133766199947, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018433505351138052, + "loss": 0.1045, + "step": 19112 + }, + { + "epoch": 0.16591001814220363, + "grad_norm": 0.15625, + "learning_rate": 0.0018433337561031043, + "loss": 0.1367, + "step": 19113 + }, + { + "epoch": 0.1659186986224078, + "grad_norm": 0.57421875, + "learning_rate": 0.0018433169762795047, + "loss": 0.1152, + "step": 19114 + }, + { + "epoch": 0.16592737910261196, + "grad_norm": 0.28515625, + "learning_rate": 0.0018433001956430254, + "loss": 0.1001, + "step": 19115 + }, + { + "epoch": 0.16593605958281613, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018432834141936837, + "loss": 0.1201, + "step": 19116 + }, + { + "epoch": 0.1659447400630203, + "grad_norm": 0.181640625, + "learning_rate": 0.0018432666319314988, + "loss": 0.1279, + "step": 19117 + }, + { + "epoch": 0.16595342054322446, + "grad_norm": 0.11767578125, + "learning_rate": 0.001843249848856488, + "loss": 0.1035, + "step": 19118 + }, + { + "epoch": 0.16596210102342862, + "grad_norm": 0.5703125, + "learning_rate": 0.001843233064968671, + "loss": 0.1299, + "step": 19119 + }, + { + "epoch": 0.16597078150363279, + "grad_norm": 0.32421875, + "learning_rate": 0.0018432162802680652, + "loss": 0.1211, + "step": 19120 + }, + { + "epoch": 0.16597946198383695, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018431994947546895, + "loss": 0.1201, + "step": 19121 + }, + { + "epoch": 0.16598814246404112, + "grad_norm": 0.6171875, + "learning_rate": 0.001843182708428562, + "loss": 0.1055, + "step": 19122 + }, + { + "epoch": 0.16599682294424528, + "grad_norm": 0.515625, + "learning_rate": 0.0018431659212897007, + "loss": 0.1367, + "step": 19123 + }, + { + "epoch": 0.16600550342444945, + "grad_norm": 0.431640625, + "learning_rate": 0.0018431491333381245, + "loss": 0.1035, + "step": 19124 + }, + { + "epoch": 0.1660141839046536, + "grad_norm": 0.4296875, + "learning_rate": 0.0018431323445738516, + "loss": 0.1035, + "step": 19125 + }, + { + "epoch": 0.16602286438485778, + "grad_norm": 0.55078125, + "learning_rate": 0.0018431155549969002, + "loss": 0.0996, + "step": 19126 + }, + { + "epoch": 0.16603154486506194, + "grad_norm": 0.1279296875, + "learning_rate": 0.001843098764607289, + "loss": 0.123, + "step": 19127 + }, + { + "epoch": 0.1660402253452661, + "grad_norm": 0.19140625, + "learning_rate": 0.0018430819734050361, + "loss": 0.1377, + "step": 19128 + }, + { + "epoch": 0.16604890582547027, + "grad_norm": 0.453125, + "learning_rate": 0.0018430651813901597, + "loss": 0.1201, + "step": 19129 + }, + { + "epoch": 0.16605758630567444, + "grad_norm": 0.68359375, + "learning_rate": 0.0018430483885626785, + "loss": 0.1069, + "step": 19130 + }, + { + "epoch": 0.1660662667858786, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018430315949226111, + "loss": 0.1211, + "step": 19131 + }, + { + "epoch": 0.16607494726608277, + "grad_norm": 0.482421875, + "learning_rate": 0.0018430148004699753, + "loss": 0.1138, + "step": 19132 + }, + { + "epoch": 0.16608362774628693, + "grad_norm": 0.65234375, + "learning_rate": 0.0018429980052047893, + "loss": 0.1523, + "step": 19133 + }, + { + "epoch": 0.1660923082264911, + "grad_norm": 0.32421875, + "learning_rate": 0.001842981209127072, + "loss": 0.1123, + "step": 19134 + }, + { + "epoch": 0.16610098870669526, + "grad_norm": 0.412109375, + "learning_rate": 0.0018429644122368418, + "loss": 0.0986, + "step": 19135 + }, + { + "epoch": 0.16610966918689943, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018429476145341171, + "loss": 0.1167, + "step": 19136 + }, + { + "epoch": 0.1661183496671036, + "grad_norm": 0.396484375, + "learning_rate": 0.0018429308160189159, + "loss": 0.1016, + "step": 19137 + }, + { + "epoch": 0.16612703014730776, + "grad_norm": 0.302734375, + "learning_rate": 0.0018429140166912566, + "loss": 0.1445, + "step": 19138 + }, + { + "epoch": 0.16613571062751192, + "grad_norm": 0.185546875, + "learning_rate": 0.0018428972165511575, + "loss": 0.1895, + "step": 19139 + }, + { + "epoch": 0.1661443911077161, + "grad_norm": 0.154296875, + "learning_rate": 0.0018428804155986374, + "loss": 0.126, + "step": 19140 + }, + { + "epoch": 0.16615307158792025, + "grad_norm": 0.0693359375, + "learning_rate": 0.0018428636138337145, + "loss": 0.0747, + "step": 19141 + }, + { + "epoch": 0.16616175206812442, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018428468112564069, + "loss": 0.1201, + "step": 19142 + }, + { + "epoch": 0.16617043254832858, + "grad_norm": 0.609375, + "learning_rate": 0.0018428300078667333, + "loss": 0.123, + "step": 19143 + }, + { + "epoch": 0.16617911302853275, + "grad_norm": 0.234375, + "learning_rate": 0.001842813203664712, + "loss": 0.1133, + "step": 19144 + }, + { + "epoch": 0.16618779350873691, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018427963986503613, + "loss": 0.1123, + "step": 19145 + }, + { + "epoch": 0.16619647398894108, + "grad_norm": 0.3046875, + "learning_rate": 0.0018427795928236998, + "loss": 0.1045, + "step": 19146 + }, + { + "epoch": 0.16620515446914524, + "grad_norm": 0.400390625, + "learning_rate": 0.0018427627861847453, + "loss": 0.0981, + "step": 19147 + }, + { + "epoch": 0.1662138349493494, + "grad_norm": 0.1171875, + "learning_rate": 0.0018427459787335168, + "loss": 0.1328, + "step": 19148 + }, + { + "epoch": 0.16622251542955357, + "grad_norm": 0.349609375, + "learning_rate": 0.0018427291704700326, + "loss": 0.1094, + "step": 19149 + }, + { + "epoch": 0.16623119590975774, + "grad_norm": 0.7578125, + "learning_rate": 0.0018427123613943108, + "loss": 0.1855, + "step": 19150 + }, + { + "epoch": 0.1662398763899619, + "grad_norm": 0.08544921875, + "learning_rate": 0.0018426955515063702, + "loss": 0.1279, + "step": 19151 + }, + { + "epoch": 0.16624855687016607, + "grad_norm": 0.73046875, + "learning_rate": 0.0018426787408062281, + "loss": 0.0967, + "step": 19152 + }, + { + "epoch": 0.16625723735037023, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018426619292939044, + "loss": 0.0957, + "step": 19153 + }, + { + "epoch": 0.1662659178305744, + "grad_norm": 0.228515625, + "learning_rate": 0.0018426451169694165, + "loss": 0.0854, + "step": 19154 + }, + { + "epoch": 0.16627459831077857, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018426283038327832, + "loss": 0.0791, + "step": 19155 + }, + { + "epoch": 0.16628327879098273, + "grad_norm": 0.1240234375, + "learning_rate": 0.0018426114898840227, + "loss": 0.0947, + "step": 19156 + }, + { + "epoch": 0.1662919592711869, + "grad_norm": 0.56640625, + "learning_rate": 0.0018425946751231535, + "loss": 0.1318, + "step": 19157 + }, + { + "epoch": 0.16630063975139106, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018425778595501936, + "loss": 0.1543, + "step": 19158 + }, + { + "epoch": 0.16630932023159523, + "grad_norm": 0.640625, + "learning_rate": 0.0018425610431651624, + "loss": 0.0986, + "step": 19159 + }, + { + "epoch": 0.16631800071179936, + "grad_norm": 0.8046875, + "learning_rate": 0.0018425442259680771, + "loss": 0.0996, + "step": 19160 + }, + { + "epoch": 0.16632668119200353, + "grad_norm": 0.435546875, + "learning_rate": 0.0018425274079589566, + "loss": 0.166, + "step": 19161 + }, + { + "epoch": 0.1663353616722077, + "grad_norm": 0.462890625, + "learning_rate": 0.0018425105891378192, + "loss": 0.1079, + "step": 19162 + }, + { + "epoch": 0.16634404215241186, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018424937695046834, + "loss": 0.1357, + "step": 19163 + }, + { + "epoch": 0.16635272263261602, + "grad_norm": 0.1455078125, + "learning_rate": 0.001842476949059568, + "loss": 0.1484, + "step": 19164 + }, + { + "epoch": 0.1663614031128202, + "grad_norm": 0.46484375, + "learning_rate": 0.0018424601278024906, + "loss": 0.1387, + "step": 19165 + }, + { + "epoch": 0.16637008359302435, + "grad_norm": 0.51171875, + "learning_rate": 0.00184244330573347, + "loss": 0.0972, + "step": 19166 + }, + { + "epoch": 0.16637876407322852, + "grad_norm": 0.9453125, + "learning_rate": 0.0018424264828525245, + "loss": 0.1436, + "step": 19167 + }, + { + "epoch": 0.16638744455343268, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018424096591596723, + "loss": 0.0977, + "step": 19168 + }, + { + "epoch": 0.16639612503363685, + "grad_norm": 0.51953125, + "learning_rate": 0.0018423928346549326, + "loss": 0.1191, + "step": 19169 + }, + { + "epoch": 0.16640480551384101, + "grad_norm": 0.95703125, + "learning_rate": 0.001842376009338323, + "loss": 0.1582, + "step": 19170 + }, + { + "epoch": 0.16641348599404518, + "grad_norm": 0.6796875, + "learning_rate": 0.001842359183209862, + "loss": 0.0977, + "step": 19171 + }, + { + "epoch": 0.16642216647424934, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018423423562695684, + "loss": 0.1309, + "step": 19172 + }, + { + "epoch": 0.1664308469544535, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018423255285174604, + "loss": 0.1348, + "step": 19173 + }, + { + "epoch": 0.16643952743465767, + "grad_norm": 0.234375, + "learning_rate": 0.001842308699953556, + "loss": 0.0981, + "step": 19174 + }, + { + "epoch": 0.16644820791486184, + "grad_norm": 0.76171875, + "learning_rate": 0.0018422918705778741, + "loss": 0.0669, + "step": 19175 + }, + { + "epoch": 0.166456888395066, + "grad_norm": 0.66015625, + "learning_rate": 0.0018422750403904334, + "loss": 0.1387, + "step": 19176 + }, + { + "epoch": 0.16646556887527017, + "grad_norm": 0.80859375, + "learning_rate": 0.0018422582093912513, + "loss": 0.3906, + "step": 19177 + }, + { + "epoch": 0.16647424935547434, + "grad_norm": 0.6015625, + "learning_rate": 0.001842241377580347, + "loss": 0.1172, + "step": 19178 + }, + { + "epoch": 0.1664829298356785, + "grad_norm": 0.259765625, + "learning_rate": 0.0018422245449577386, + "loss": 0.0933, + "step": 19179 + }, + { + "epoch": 0.16649161031588267, + "grad_norm": 0.87890625, + "learning_rate": 0.0018422077115234448, + "loss": 0.0918, + "step": 19180 + }, + { + "epoch": 0.16650029079608683, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018421908772774836, + "loss": 0.1494, + "step": 19181 + }, + { + "epoch": 0.166508971276291, + "grad_norm": 0.427734375, + "learning_rate": 0.0018421740422198737, + "loss": 0.1328, + "step": 19182 + }, + { + "epoch": 0.16651765175649516, + "grad_norm": 0.8671875, + "learning_rate": 0.0018421572063506333, + "loss": 0.1504, + "step": 19183 + }, + { + "epoch": 0.16652633223669933, + "grad_norm": 0.0888671875, + "learning_rate": 0.001842140369669781, + "loss": 0.1235, + "step": 19184 + }, + { + "epoch": 0.1665350127169035, + "grad_norm": 0.255859375, + "learning_rate": 0.0018421235321773352, + "loss": 0.1348, + "step": 19185 + }, + { + "epoch": 0.16654369319710766, + "grad_norm": 0.33203125, + "learning_rate": 0.001842106693873314, + "loss": 0.1445, + "step": 19186 + }, + { + "epoch": 0.16655237367731182, + "grad_norm": 0.67578125, + "learning_rate": 0.0018420898547577365, + "loss": 0.1309, + "step": 19187 + }, + { + "epoch": 0.166561054157516, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018420730148306205, + "loss": 0.124, + "step": 19188 + }, + { + "epoch": 0.16656973463772015, + "grad_norm": 0.248046875, + "learning_rate": 0.0018420561740919847, + "loss": 0.1377, + "step": 19189 + }, + { + "epoch": 0.16657841511792432, + "grad_norm": 0.13671875, + "learning_rate": 0.001842039332541847, + "loss": 0.1084, + "step": 19190 + }, + { + "epoch": 0.16658709559812848, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018420224901802267, + "loss": 0.1445, + "step": 19191 + }, + { + "epoch": 0.16659577607833265, + "grad_norm": 0.28125, + "learning_rate": 0.0018420056470071415, + "loss": 0.1055, + "step": 19192 + }, + { + "epoch": 0.1666044565585368, + "grad_norm": 0.466796875, + "learning_rate": 0.00184198880302261, + "loss": 0.1279, + "step": 19193 + }, + { + "epoch": 0.16661313703874098, + "grad_norm": 0.2734375, + "learning_rate": 0.001841971958226651, + "loss": 0.0908, + "step": 19194 + }, + { + "epoch": 0.16662181751894514, + "grad_norm": 0.6953125, + "learning_rate": 0.0018419551126192825, + "loss": 0.0771, + "step": 19195 + }, + { + "epoch": 0.1666304979991493, + "grad_norm": 0.47265625, + "learning_rate": 0.0018419382662005228, + "loss": 0.0947, + "step": 19196 + }, + { + "epoch": 0.16663917847935347, + "grad_norm": 0.095703125, + "learning_rate": 0.0018419214189703908, + "loss": 0.1074, + "step": 19197 + }, + { + "epoch": 0.16664785895955764, + "grad_norm": 0.3515625, + "learning_rate": 0.0018419045709289045, + "loss": 0.1455, + "step": 19198 + }, + { + "epoch": 0.1666565394397618, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018418877220760827, + "loss": 0.0957, + "step": 19199 + }, + { + "epoch": 0.16666521991996597, + "grad_norm": 0.3046875, + "learning_rate": 0.0018418708724119437, + "loss": 0.1016, + "step": 19200 + }, + { + "epoch": 0.16667390040017013, + "grad_norm": 0.3828125, + "learning_rate": 0.001841854021936506, + "loss": 0.1621, + "step": 19201 + }, + { + "epoch": 0.1666825808803743, + "grad_norm": 0.490234375, + "learning_rate": 0.0018418371706497873, + "loss": 0.1338, + "step": 19202 + }, + { + "epoch": 0.16669126136057846, + "grad_norm": 0.154296875, + "learning_rate": 0.001841820318551807, + "loss": 0.1084, + "step": 19203 + }, + { + "epoch": 0.16669994184078263, + "grad_norm": 0.60546875, + "learning_rate": 0.0018418034656425828, + "loss": 0.0889, + "step": 19204 + }, + { + "epoch": 0.1667086223209868, + "grad_norm": 0.87890625, + "learning_rate": 0.001841786611922134, + "loss": 0.123, + "step": 19205 + }, + { + "epoch": 0.16671730280119096, + "grad_norm": 0.2490234375, + "learning_rate": 0.001841769757390478, + "loss": 0.0986, + "step": 19206 + }, + { + "epoch": 0.16672598328139512, + "grad_norm": 0.291015625, + "learning_rate": 0.0018417529020476342, + "loss": 0.1055, + "step": 19207 + }, + { + "epoch": 0.1667346637615993, + "grad_norm": 0.373046875, + "learning_rate": 0.0018417360458936206, + "loss": 0.1641, + "step": 19208 + }, + { + "epoch": 0.16674334424180345, + "grad_norm": 0.341796875, + "learning_rate": 0.0018417191889284553, + "loss": 0.1123, + "step": 19209 + }, + { + "epoch": 0.16675202472200762, + "grad_norm": 0.197265625, + "learning_rate": 0.001841702331152157, + "loss": 0.1699, + "step": 19210 + }, + { + "epoch": 0.16676070520221178, + "grad_norm": 0.29296875, + "learning_rate": 0.0018416854725647443, + "loss": 0.0845, + "step": 19211 + }, + { + "epoch": 0.16676938568241595, + "grad_norm": 0.103515625, + "learning_rate": 0.0018416686131662355, + "loss": 0.1494, + "step": 19212 + }, + { + "epoch": 0.16677806616262011, + "grad_norm": 0.4296875, + "learning_rate": 0.001841651752956649, + "loss": 0.1582, + "step": 19213 + }, + { + "epoch": 0.16678674664282428, + "grad_norm": 0.302734375, + "learning_rate": 0.0018416348919360033, + "loss": 0.1572, + "step": 19214 + }, + { + "epoch": 0.16679542712302844, + "grad_norm": 0.1904296875, + "learning_rate": 0.001841618030104317, + "loss": 0.1484, + "step": 19215 + }, + { + "epoch": 0.1668041076032326, + "grad_norm": 0.30859375, + "learning_rate": 0.001841601167461608, + "loss": 0.0903, + "step": 19216 + }, + { + "epoch": 0.16681278808343677, + "grad_norm": 0.39453125, + "learning_rate": 0.0018415843040078955, + "loss": 0.0898, + "step": 19217 + }, + { + "epoch": 0.16682146856364094, + "grad_norm": 0.318359375, + "learning_rate": 0.0018415674397431977, + "loss": 0.1094, + "step": 19218 + }, + { + "epoch": 0.1668301490438451, + "grad_norm": 0.19140625, + "learning_rate": 0.0018415505746675323, + "loss": 0.0923, + "step": 19219 + }, + { + "epoch": 0.16683882952404927, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018415337087809185, + "loss": 0.1191, + "step": 19220 + }, + { + "epoch": 0.16684751000425344, + "grad_norm": 0.259765625, + "learning_rate": 0.0018415168420833747, + "loss": 0.124, + "step": 19221 + }, + { + "epoch": 0.1668561904844576, + "grad_norm": 0.3359375, + "learning_rate": 0.0018414999745749192, + "loss": 0.1084, + "step": 19222 + }, + { + "epoch": 0.16686487096466177, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018414831062555706, + "loss": 0.1221, + "step": 19223 + }, + { + "epoch": 0.16687355144486593, + "grad_norm": 0.134765625, + "learning_rate": 0.001841466237125347, + "loss": 0.1045, + "step": 19224 + }, + { + "epoch": 0.1668822319250701, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018414493671842671, + "loss": 0.1045, + "step": 19225 + }, + { + "epoch": 0.16689091240527426, + "grad_norm": 0.11328125, + "learning_rate": 0.0018414324964323495, + "loss": 0.0854, + "step": 19226 + }, + { + "epoch": 0.16689959288547843, + "grad_norm": 0.26171875, + "learning_rate": 0.0018414156248696123, + "loss": 0.1094, + "step": 19227 + }, + { + "epoch": 0.1669082733656826, + "grad_norm": 0.1591796875, + "learning_rate": 0.001841398752496074, + "loss": 0.0796, + "step": 19228 + }, + { + "epoch": 0.16691695384588676, + "grad_norm": 0.71484375, + "learning_rate": 0.0018413818793117534, + "loss": 0.1426, + "step": 19229 + }, + { + "epoch": 0.16692563432609092, + "grad_norm": 0.80859375, + "learning_rate": 0.0018413650053166685, + "loss": 0.1172, + "step": 19230 + }, + { + "epoch": 0.1669343148062951, + "grad_norm": 0.30078125, + "learning_rate": 0.0018413481305108383, + "loss": 0.0908, + "step": 19231 + }, + { + "epoch": 0.16694299528649925, + "grad_norm": 0.30078125, + "learning_rate": 0.0018413312548942807, + "loss": 0.1138, + "step": 19232 + }, + { + "epoch": 0.16695167576670342, + "grad_norm": 0.828125, + "learning_rate": 0.0018413143784670144, + "loss": 0.1006, + "step": 19233 + }, + { + "epoch": 0.16696035624690758, + "grad_norm": 0.11181640625, + "learning_rate": 0.0018412975012290578, + "loss": 0.1143, + "step": 19234 + }, + { + "epoch": 0.16696903672711175, + "grad_norm": 0.154296875, + "learning_rate": 0.0018412806231804296, + "loss": 0.125, + "step": 19235 + }, + { + "epoch": 0.1669777172073159, + "grad_norm": 0.130859375, + "learning_rate": 0.0018412637443211476, + "loss": 0.1064, + "step": 19236 + }, + { + "epoch": 0.16698639768752008, + "grad_norm": 0.09912109375, + "learning_rate": 0.001841246864651231, + "loss": 0.126, + "step": 19237 + }, + { + "epoch": 0.16699507816772424, + "grad_norm": 0.080078125, + "learning_rate": 0.001841229984170698, + "loss": 0.0713, + "step": 19238 + }, + { + "epoch": 0.1670037586479284, + "grad_norm": 0.431640625, + "learning_rate": 0.0018412131028795668, + "loss": 0.1074, + "step": 19239 + }, + { + "epoch": 0.16701243912813257, + "grad_norm": 0.185546875, + "learning_rate": 0.0018411962207778565, + "loss": 0.0908, + "step": 19240 + }, + { + "epoch": 0.16702111960833674, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018411793378655845, + "loss": 0.1064, + "step": 19241 + }, + { + "epoch": 0.1670298000885409, + "grad_norm": 0.10107421875, + "learning_rate": 0.0018411624541427704, + "loss": 0.1172, + "step": 19242 + }, + { + "epoch": 0.16703848056874507, + "grad_norm": 0.427734375, + "learning_rate": 0.001841145569609432, + "loss": 0.1191, + "step": 19243 + }, + { + "epoch": 0.16704716104894923, + "grad_norm": 0.125, + "learning_rate": 0.001841128684265588, + "loss": 0.1133, + "step": 19244 + }, + { + "epoch": 0.1670558415291534, + "grad_norm": 0.087890625, + "learning_rate": 0.001841111798111257, + "loss": 0.1104, + "step": 19245 + }, + { + "epoch": 0.16706452200935756, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018410949111464573, + "loss": 0.0957, + "step": 19246 + }, + { + "epoch": 0.16707320248956173, + "grad_norm": 0.12158203125, + "learning_rate": 0.001841078023371207, + "loss": 0.105, + "step": 19247 + }, + { + "epoch": 0.1670818829697659, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018410611347855251, + "loss": 0.1123, + "step": 19248 + }, + { + "epoch": 0.16709056344997006, + "grad_norm": 0.2470703125, + "learning_rate": 0.00184104424538943, + "loss": 0.0957, + "step": 19249 + }, + { + "epoch": 0.16709924393017422, + "grad_norm": 0.40625, + "learning_rate": 0.0018410273551829396, + "loss": 0.1523, + "step": 19250 + }, + { + "epoch": 0.1671079244103784, + "grad_norm": 0.5546875, + "learning_rate": 0.0018410104641660731, + "loss": 0.1338, + "step": 19251 + }, + { + "epoch": 0.16711660489058255, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018409935723388488, + "loss": 0.1089, + "step": 19252 + }, + { + "epoch": 0.16712528537078672, + "grad_norm": 0.259765625, + "learning_rate": 0.001840976679701285, + "loss": 0.1177, + "step": 19253 + }, + { + "epoch": 0.16713396585099088, + "grad_norm": 0.2421875, + "learning_rate": 0.0018409597862534, + "loss": 0.0996, + "step": 19254 + }, + { + "epoch": 0.16714264633119505, + "grad_norm": 0.32421875, + "learning_rate": 0.0018409428919952129, + "loss": 0.1099, + "step": 19255 + }, + { + "epoch": 0.16715132681139921, + "grad_norm": 0.1953125, + "learning_rate": 0.0018409259969267414, + "loss": 0.127, + "step": 19256 + }, + { + "epoch": 0.16716000729160338, + "grad_norm": 0.181640625, + "learning_rate": 0.0018409091010480045, + "loss": 0.1582, + "step": 19257 + }, + { + "epoch": 0.16716868777180754, + "grad_norm": 0.404296875, + "learning_rate": 0.0018408922043590206, + "loss": 0.1357, + "step": 19258 + }, + { + "epoch": 0.1671773682520117, + "grad_norm": 0.42578125, + "learning_rate": 0.0018408753068598082, + "loss": 0.1367, + "step": 19259 + }, + { + "epoch": 0.16718604873221588, + "grad_norm": 0.251953125, + "learning_rate": 0.0018408584085503855, + "loss": 0.1191, + "step": 19260 + }, + { + "epoch": 0.16719472921242004, + "grad_norm": 0.462890625, + "learning_rate": 0.0018408415094307714, + "loss": 0.123, + "step": 19261 + }, + { + "epoch": 0.1672034096926242, + "grad_norm": 0.259765625, + "learning_rate": 0.0018408246095009842, + "loss": 0.1187, + "step": 19262 + }, + { + "epoch": 0.16721209017282837, + "grad_norm": 0.3203125, + "learning_rate": 0.001840807708761042, + "loss": 0.1592, + "step": 19263 + }, + { + "epoch": 0.16722077065303254, + "grad_norm": 0.306640625, + "learning_rate": 0.0018407908072109638, + "loss": 0.0942, + "step": 19264 + }, + { + "epoch": 0.1672294511332367, + "grad_norm": 0.1201171875, + "learning_rate": 0.0018407739048507679, + "loss": 0.0977, + "step": 19265 + }, + { + "epoch": 0.16723813161344087, + "grad_norm": 0.275390625, + "learning_rate": 0.0018407570016804728, + "loss": 0.1152, + "step": 19266 + }, + { + "epoch": 0.16724681209364503, + "grad_norm": 0.60546875, + "learning_rate": 0.0018407400977000967, + "loss": 0.1074, + "step": 19267 + }, + { + "epoch": 0.1672554925738492, + "grad_norm": 0.47265625, + "learning_rate": 0.0018407231929096588, + "loss": 0.1621, + "step": 19268 + }, + { + "epoch": 0.16726417305405336, + "grad_norm": 0.9140625, + "learning_rate": 0.0018407062873091768, + "loss": 0.1177, + "step": 19269 + }, + { + "epoch": 0.16727285353425753, + "grad_norm": 0.26953125, + "learning_rate": 0.0018406893808986698, + "loss": 0.1191, + "step": 19270 + }, + { + "epoch": 0.1672815340144617, + "grad_norm": 0.396484375, + "learning_rate": 0.001840672473678156, + "loss": 0.126, + "step": 19271 + }, + { + "epoch": 0.16729021449466586, + "grad_norm": 0.419921875, + "learning_rate": 0.0018406555656476536, + "loss": 0.084, + "step": 19272 + }, + { + "epoch": 0.16729889497487002, + "grad_norm": 1.2890625, + "learning_rate": 0.0018406386568071817, + "loss": 0.0938, + "step": 19273 + }, + { + "epoch": 0.1673075754550742, + "grad_norm": 0.11181640625, + "learning_rate": 0.0018406217471567585, + "loss": 0.1064, + "step": 19274 + }, + { + "epoch": 0.16731625593527835, + "grad_norm": 0.55859375, + "learning_rate": 0.0018406048366964023, + "loss": 0.1133, + "step": 19275 + }, + { + "epoch": 0.16732493641548252, + "grad_norm": 0.2265625, + "learning_rate": 0.0018405879254261319, + "loss": 0.0771, + "step": 19276 + }, + { + "epoch": 0.16733361689568668, + "grad_norm": 0.76953125, + "learning_rate": 0.0018405710133459657, + "loss": 0.1338, + "step": 19277 + }, + { + "epoch": 0.16734229737589085, + "grad_norm": 0.25, + "learning_rate": 0.0018405541004559223, + "loss": 0.1426, + "step": 19278 + }, + { + "epoch": 0.167350977856095, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018405371867560197, + "loss": 0.0942, + "step": 19279 + }, + { + "epoch": 0.16735965833629918, + "grad_norm": 0.58984375, + "learning_rate": 0.0018405202722462767, + "loss": 0.1631, + "step": 19280 + }, + { + "epoch": 0.16736833881650334, + "grad_norm": 0.357421875, + "learning_rate": 0.0018405033569267124, + "loss": 0.0986, + "step": 19281 + }, + { + "epoch": 0.1673770192967075, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018404864407973445, + "loss": 0.1025, + "step": 19282 + }, + { + "epoch": 0.16738569977691165, + "grad_norm": 0.17578125, + "learning_rate": 0.0018404695238581916, + "loss": 0.1777, + "step": 19283 + }, + { + "epoch": 0.1673943802571158, + "grad_norm": 0.412109375, + "learning_rate": 0.0018404526061092724, + "loss": 0.1533, + "step": 19284 + }, + { + "epoch": 0.16740306073731998, + "grad_norm": 0.62109375, + "learning_rate": 0.0018404356875506055, + "loss": 0.1709, + "step": 19285 + }, + { + "epoch": 0.16741174121752414, + "grad_norm": 0.79296875, + "learning_rate": 0.0018404187681822092, + "loss": 0.1099, + "step": 19286 + }, + { + "epoch": 0.1674204216977283, + "grad_norm": 0.73046875, + "learning_rate": 0.001840401848004102, + "loss": 0.1309, + "step": 19287 + }, + { + "epoch": 0.16742910217793247, + "grad_norm": 0.490234375, + "learning_rate": 0.0018403849270163025, + "loss": 0.1196, + "step": 19288 + }, + { + "epoch": 0.16743778265813664, + "grad_norm": 0.171875, + "learning_rate": 0.0018403680052188293, + "loss": 0.1328, + "step": 19289 + }, + { + "epoch": 0.1674464631383408, + "grad_norm": 0.3984375, + "learning_rate": 0.0018403510826117005, + "loss": 0.0879, + "step": 19290 + }, + { + "epoch": 0.16745514361854497, + "grad_norm": 0.12451171875, + "learning_rate": 0.001840334159194935, + "loss": 0.1006, + "step": 19291 + }, + { + "epoch": 0.16746382409874913, + "grad_norm": 0.0771484375, + "learning_rate": 0.0018403172349685513, + "loss": 0.1113, + "step": 19292 + }, + { + "epoch": 0.1674725045789533, + "grad_norm": 0.98828125, + "learning_rate": 0.0018403003099325675, + "loss": 0.1162, + "step": 19293 + }, + { + "epoch": 0.16748118505915746, + "grad_norm": 0.26171875, + "learning_rate": 0.0018402833840870025, + "loss": 0.1069, + "step": 19294 + }, + { + "epoch": 0.16748986553936163, + "grad_norm": 0.2421875, + "learning_rate": 0.001840266457431875, + "loss": 0.1045, + "step": 19295 + }, + { + "epoch": 0.1674985460195658, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018402495299672031, + "loss": 0.0908, + "step": 19296 + }, + { + "epoch": 0.16750722649976996, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018402326016930051, + "loss": 0.0947, + "step": 19297 + }, + { + "epoch": 0.16751590697997412, + "grad_norm": 0.384765625, + "learning_rate": 0.0018402156726093002, + "loss": 0.0825, + "step": 19298 + }, + { + "epoch": 0.1675245874601783, + "grad_norm": 0.46484375, + "learning_rate": 0.0018401987427161064, + "loss": 0.1504, + "step": 19299 + }, + { + "epoch": 0.16753326794038245, + "grad_norm": 0.447265625, + "learning_rate": 0.0018401818120134426, + "loss": 0.0942, + "step": 19300 + }, + { + "epoch": 0.16754194842058662, + "grad_norm": 0.74609375, + "learning_rate": 0.001840164880501327, + "loss": 0.1167, + "step": 19301 + }, + { + "epoch": 0.16755062890079078, + "grad_norm": 0.095703125, + "learning_rate": 0.0018401479481797783, + "loss": 0.0771, + "step": 19302 + }, + { + "epoch": 0.16755930938099495, + "grad_norm": 0.2890625, + "learning_rate": 0.0018401310150488147, + "loss": 0.1348, + "step": 19303 + }, + { + "epoch": 0.1675679898611991, + "grad_norm": 0.57421875, + "learning_rate": 0.001840114081108455, + "loss": 0.1143, + "step": 19304 + }, + { + "epoch": 0.16757667034140328, + "grad_norm": 0.11376953125, + "learning_rate": 0.0018400971463587176, + "loss": 0.1279, + "step": 19305 + }, + { + "epoch": 0.16758535082160744, + "grad_norm": 0.09912109375, + "learning_rate": 0.0018400802107996213, + "loss": 0.1177, + "step": 19306 + }, + { + "epoch": 0.1675940313018116, + "grad_norm": 0.59765625, + "learning_rate": 0.0018400632744311844, + "loss": 0.125, + "step": 19307 + }, + { + "epoch": 0.16760271178201577, + "grad_norm": 0.283203125, + "learning_rate": 0.001840046337253425, + "loss": 0.1094, + "step": 19308 + }, + { + "epoch": 0.16761139226221994, + "grad_norm": 0.337890625, + "learning_rate": 0.0018400293992663623, + "loss": 0.1289, + "step": 19309 + }, + { + "epoch": 0.1676200727424241, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018400124604700148, + "loss": 0.166, + "step": 19310 + }, + { + "epoch": 0.16762875322262827, + "grad_norm": 0.119140625, + "learning_rate": 0.0018399955208644005, + "loss": 0.1348, + "step": 19311 + }, + { + "epoch": 0.16763743370283243, + "grad_norm": 0.3828125, + "learning_rate": 0.0018399785804495386, + "loss": 0.1318, + "step": 19312 + }, + { + "epoch": 0.1676461141830366, + "grad_norm": 0.09814453125, + "learning_rate": 0.001839961639225447, + "loss": 0.1191, + "step": 19313 + }, + { + "epoch": 0.16765479466324076, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018399446971921443, + "loss": 0.0967, + "step": 19314 + }, + { + "epoch": 0.16766347514344493, + "grad_norm": 0.357421875, + "learning_rate": 0.0018399277543496493, + "loss": 0.1328, + "step": 19315 + }, + { + "epoch": 0.1676721556236491, + "grad_norm": 0.4375, + "learning_rate": 0.0018399108106979803, + "loss": 0.1147, + "step": 19316 + }, + { + "epoch": 0.16768083610385326, + "grad_norm": 0.07421875, + "learning_rate": 0.0018398938662371562, + "loss": 0.0957, + "step": 19317 + }, + { + "epoch": 0.16768951658405742, + "grad_norm": 0.11376953125, + "learning_rate": 0.0018398769209671951, + "loss": 0.0962, + "step": 19318 + }, + { + "epoch": 0.1676981970642616, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018398599748881159, + "loss": 0.0986, + "step": 19319 + }, + { + "epoch": 0.16770687754446575, + "grad_norm": 0.302734375, + "learning_rate": 0.0018398430279999366, + "loss": 0.0898, + "step": 19320 + }, + { + "epoch": 0.16771555802466992, + "grad_norm": 0.1640625, + "learning_rate": 0.0018398260803026765, + "loss": 0.0996, + "step": 19321 + }, + { + "epoch": 0.16772423850487408, + "grad_norm": 0.255859375, + "learning_rate": 0.0018398091317963538, + "loss": 0.1196, + "step": 19322 + }, + { + "epoch": 0.16773291898507825, + "grad_norm": 0.318359375, + "learning_rate": 0.0018397921824809865, + "loss": 0.0962, + "step": 19323 + }, + { + "epoch": 0.16774159946528241, + "grad_norm": 0.1279296875, + "learning_rate": 0.001839775232356594, + "loss": 0.1309, + "step": 19324 + }, + { + "epoch": 0.16775027994548658, + "grad_norm": 0.1357421875, + "learning_rate": 0.001839758281423194, + "loss": 0.1089, + "step": 19325 + }, + { + "epoch": 0.16775896042569075, + "grad_norm": 0.107421875, + "learning_rate": 0.0018397413296808056, + "loss": 0.0947, + "step": 19326 + }, + { + "epoch": 0.1677676409058949, + "grad_norm": 0.48828125, + "learning_rate": 0.001839724377129447, + "loss": 0.0942, + "step": 19327 + }, + { + "epoch": 0.16777632138609908, + "grad_norm": 0.234375, + "learning_rate": 0.0018397074237691373, + "loss": 0.1152, + "step": 19328 + }, + { + "epoch": 0.16778500186630324, + "grad_norm": 0.255859375, + "learning_rate": 0.001839690469599895, + "loss": 0.0615, + "step": 19329 + }, + { + "epoch": 0.1677936823465074, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018396735146217376, + "loss": 0.1426, + "step": 19330 + }, + { + "epoch": 0.16780236282671157, + "grad_norm": 0.478515625, + "learning_rate": 0.0018396565588346848, + "loss": 0.0947, + "step": 19331 + }, + { + "epoch": 0.16781104330691574, + "grad_norm": 0.27734375, + "learning_rate": 0.0018396396022387542, + "loss": 0.1104, + "step": 19332 + }, + { + "epoch": 0.1678197237871199, + "grad_norm": 0.234375, + "learning_rate": 0.0018396226448339655, + "loss": 0.0835, + "step": 19333 + }, + { + "epoch": 0.16782840426732407, + "grad_norm": 0.111328125, + "learning_rate": 0.0018396056866203361, + "loss": 0.1162, + "step": 19334 + }, + { + "epoch": 0.16783708474752823, + "grad_norm": 0.296875, + "learning_rate": 0.0018395887275978852, + "loss": 0.0996, + "step": 19335 + }, + { + "epoch": 0.1678457652277324, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018395717677666312, + "loss": 0.0957, + "step": 19336 + }, + { + "epoch": 0.16785444570793656, + "grad_norm": 0.625, + "learning_rate": 0.0018395548071265927, + "loss": 0.1006, + "step": 19337 + }, + { + "epoch": 0.16786312618814073, + "grad_norm": 0.296875, + "learning_rate": 0.0018395378456777882, + "loss": 0.1279, + "step": 19338 + }, + { + "epoch": 0.1678718066683449, + "grad_norm": 0.091796875, + "learning_rate": 0.001839520883420236, + "loss": 0.1172, + "step": 19339 + }, + { + "epoch": 0.16788048714854906, + "grad_norm": 0.28515625, + "learning_rate": 0.0018395039203539547, + "loss": 0.0791, + "step": 19340 + }, + { + "epoch": 0.16788916762875322, + "grad_norm": 1.0, + "learning_rate": 0.0018394869564789635, + "loss": 0.1621, + "step": 19341 + }, + { + "epoch": 0.1678978481089574, + "grad_norm": 0.48828125, + "learning_rate": 0.0018394699917952803, + "loss": 0.1699, + "step": 19342 + }, + { + "epoch": 0.16790652858916155, + "grad_norm": 0.091796875, + "learning_rate": 0.0018394530263029238, + "loss": 0.0898, + "step": 19343 + }, + { + "epoch": 0.16791520906936572, + "grad_norm": 0.1064453125, + "learning_rate": 0.0018394360600019126, + "loss": 0.0879, + "step": 19344 + }, + { + "epoch": 0.16792388954956988, + "grad_norm": 0.4921875, + "learning_rate": 0.0018394190928922652, + "loss": 0.1182, + "step": 19345 + }, + { + "epoch": 0.16793257002977405, + "grad_norm": 0.15625, + "learning_rate": 0.0018394021249740003, + "loss": 0.0972, + "step": 19346 + }, + { + "epoch": 0.1679412505099782, + "grad_norm": 0.921875, + "learning_rate": 0.0018393851562471361, + "loss": 0.0996, + "step": 19347 + }, + { + "epoch": 0.16794993099018238, + "grad_norm": 0.337890625, + "learning_rate": 0.0018393681867116918, + "loss": 0.209, + "step": 19348 + }, + { + "epoch": 0.16795861147038654, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018393512163676854, + "loss": 0.1787, + "step": 19349 + }, + { + "epoch": 0.1679672919505907, + "grad_norm": 0.4609375, + "learning_rate": 0.0018393342452151353, + "loss": 0.0869, + "step": 19350 + }, + { + "epoch": 0.16797597243079487, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018393172732540607, + "loss": 0.1133, + "step": 19351 + }, + { + "epoch": 0.16798465291099904, + "grad_norm": 0.3671875, + "learning_rate": 0.0018393003004844796, + "loss": 0.1504, + "step": 19352 + }, + { + "epoch": 0.1679933333912032, + "grad_norm": 0.439453125, + "learning_rate": 0.0018392833269064112, + "loss": 0.126, + "step": 19353 + }, + { + "epoch": 0.16800201387140737, + "grad_norm": 0.3515625, + "learning_rate": 0.0018392663525198733, + "loss": 0.1494, + "step": 19354 + }, + { + "epoch": 0.16801069435161153, + "grad_norm": 0.4453125, + "learning_rate": 0.0018392493773248851, + "loss": 0.1084, + "step": 19355 + }, + { + "epoch": 0.1680193748318157, + "grad_norm": 0.201171875, + "learning_rate": 0.0018392324013214644, + "loss": 0.125, + "step": 19356 + }, + { + "epoch": 0.16802805531201986, + "grad_norm": 0.07275390625, + "learning_rate": 0.001839215424509631, + "loss": 0.084, + "step": 19357 + }, + { + "epoch": 0.16803673579222403, + "grad_norm": 0.5546875, + "learning_rate": 0.001839198446889402, + "loss": 0.1436, + "step": 19358 + }, + { + "epoch": 0.1680454162724282, + "grad_norm": 0.65234375, + "learning_rate": 0.0018391814684607971, + "loss": 0.1157, + "step": 19359 + }, + { + "epoch": 0.16805409675263236, + "grad_norm": 0.37890625, + "learning_rate": 0.0018391644892238343, + "loss": 0.1387, + "step": 19360 + }, + { + "epoch": 0.16806277723283652, + "grad_norm": 0.259765625, + "learning_rate": 0.0018391475091785324, + "loss": 0.0967, + "step": 19361 + }, + { + "epoch": 0.1680714577130407, + "grad_norm": 0.287109375, + "learning_rate": 0.00183913052832491, + "loss": 0.1279, + "step": 19362 + }, + { + "epoch": 0.16808013819324485, + "grad_norm": 0.25, + "learning_rate": 0.0018391135466629854, + "loss": 0.1035, + "step": 19363 + }, + { + "epoch": 0.16808881867344902, + "grad_norm": 0.138671875, + "learning_rate": 0.0018390965641927773, + "loss": 0.1602, + "step": 19364 + }, + { + "epoch": 0.16809749915365318, + "grad_norm": 0.95703125, + "learning_rate": 0.0018390795809143043, + "loss": 0.1367, + "step": 19365 + }, + { + "epoch": 0.16810617963385735, + "grad_norm": 0.5078125, + "learning_rate": 0.001839062596827585, + "loss": 0.1182, + "step": 19366 + }, + { + "epoch": 0.16811486011406152, + "grad_norm": 0.337890625, + "learning_rate": 0.0018390456119326381, + "loss": 0.1055, + "step": 19367 + }, + { + "epoch": 0.16812354059426568, + "grad_norm": 0.4375, + "learning_rate": 0.0018390286262294818, + "loss": 0.1895, + "step": 19368 + }, + { + "epoch": 0.16813222107446985, + "grad_norm": 0.216796875, + "learning_rate": 0.001839011639718135, + "loss": 0.0879, + "step": 19369 + }, + { + "epoch": 0.168140901554674, + "grad_norm": 0.16796875, + "learning_rate": 0.0018389946523986159, + "loss": 0.1201, + "step": 19370 + }, + { + "epoch": 0.16814958203487818, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018389776642709437, + "loss": 0.1328, + "step": 19371 + }, + { + "epoch": 0.16815826251508234, + "grad_norm": 0.345703125, + "learning_rate": 0.001838960675335137, + "loss": 0.1152, + "step": 19372 + }, + { + "epoch": 0.1681669429952865, + "grad_norm": 0.419921875, + "learning_rate": 0.0018389436855912131, + "loss": 0.1045, + "step": 19373 + }, + { + "epoch": 0.16817562347549067, + "grad_norm": 1.1484375, + "learning_rate": 0.001838926695039192, + "loss": 0.1572, + "step": 19374 + }, + { + "epoch": 0.16818430395569484, + "grad_norm": 0.2080078125, + "learning_rate": 0.0018389097036790917, + "loss": 0.0928, + "step": 19375 + }, + { + "epoch": 0.168192984435899, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018388927115109309, + "loss": 0.1118, + "step": 19376 + }, + { + "epoch": 0.16820166491610317, + "grad_norm": 0.1845703125, + "learning_rate": 0.0018388757185347283, + "loss": 0.1396, + "step": 19377 + }, + { + "epoch": 0.16821034539630733, + "grad_norm": 0.1953125, + "learning_rate": 0.0018388587247505019, + "loss": 0.127, + "step": 19378 + }, + { + "epoch": 0.1682190258765115, + "grad_norm": 0.18359375, + "learning_rate": 0.001838841730158271, + "loss": 0.1387, + "step": 19379 + }, + { + "epoch": 0.16822770635671566, + "grad_norm": 0.44921875, + "learning_rate": 0.0018388247347580538, + "loss": 0.1191, + "step": 19380 + }, + { + "epoch": 0.16823638683691983, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018388077385498689, + "loss": 0.1289, + "step": 19381 + }, + { + "epoch": 0.168245067317124, + "grad_norm": 0.306640625, + "learning_rate": 0.0018387907415337351, + "loss": 0.123, + "step": 19382 + }, + { + "epoch": 0.16825374779732816, + "grad_norm": 0.423828125, + "learning_rate": 0.0018387737437096706, + "loss": 0.0923, + "step": 19383 + }, + { + "epoch": 0.16826242827753232, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018387567450776944, + "loss": 0.0923, + "step": 19384 + }, + { + "epoch": 0.1682711087577365, + "grad_norm": 0.51953125, + "learning_rate": 0.001838739745637825, + "loss": 0.1543, + "step": 19385 + }, + { + "epoch": 0.16827978923794065, + "grad_norm": 0.734375, + "learning_rate": 0.0018387227453900807, + "loss": 0.1641, + "step": 19386 + }, + { + "epoch": 0.16828846971814482, + "grad_norm": 0.458984375, + "learning_rate": 0.0018387057443344805, + "loss": 0.0908, + "step": 19387 + }, + { + "epoch": 0.16829715019834898, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018386887424710425, + "loss": 0.1328, + "step": 19388 + }, + { + "epoch": 0.16830583067855315, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018386717397997862, + "loss": 0.1123, + "step": 19389 + }, + { + "epoch": 0.1683145111587573, + "grad_norm": 0.287109375, + "learning_rate": 0.001838654736320729, + "loss": 0.1318, + "step": 19390 + }, + { + "epoch": 0.16832319163896148, + "grad_norm": 0.66796875, + "learning_rate": 0.0018386377320338903, + "loss": 0.1221, + "step": 19391 + }, + { + "epoch": 0.16833187211916564, + "grad_norm": 0.5859375, + "learning_rate": 0.0018386207269392887, + "loss": 0.1191, + "step": 19392 + }, + { + "epoch": 0.1683405525993698, + "grad_norm": 0.083984375, + "learning_rate": 0.0018386037210369422, + "loss": 0.0928, + "step": 19393 + }, + { + "epoch": 0.16834923307957397, + "grad_norm": 0.52734375, + "learning_rate": 0.0018385867143268697, + "loss": 0.1289, + "step": 19394 + }, + { + "epoch": 0.16835791355977814, + "grad_norm": 0.86328125, + "learning_rate": 0.0018385697068090904, + "loss": 0.1562, + "step": 19395 + }, + { + "epoch": 0.1683665940399823, + "grad_norm": 0.09130859375, + "learning_rate": 0.0018385526984836219, + "loss": 0.105, + "step": 19396 + }, + { + "epoch": 0.16837527452018647, + "grad_norm": 0.1005859375, + "learning_rate": 0.0018385356893504832, + "loss": 0.0781, + "step": 19397 + }, + { + "epoch": 0.16838395500039063, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018385186794096934, + "loss": 0.1816, + "step": 19398 + }, + { + "epoch": 0.1683926354805948, + "grad_norm": 0.2578125, + "learning_rate": 0.0018385016686612704, + "loss": 0.1021, + "step": 19399 + }, + { + "epoch": 0.16840131596079896, + "grad_norm": 0.265625, + "learning_rate": 0.0018384846571052333, + "loss": 0.1309, + "step": 19400 + }, + { + "epoch": 0.16840999644100313, + "grad_norm": 0.609375, + "learning_rate": 0.0018384676447416, + "loss": 0.0967, + "step": 19401 + }, + { + "epoch": 0.1684186769212073, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018384506315703899, + "loss": 0.0693, + "step": 19402 + }, + { + "epoch": 0.16842735740141146, + "grad_norm": 0.37890625, + "learning_rate": 0.0018384336175916215, + "loss": 0.1123, + "step": 19403 + }, + { + "epoch": 0.16843603788161562, + "grad_norm": 0.45703125, + "learning_rate": 0.0018384166028053127, + "loss": 0.1143, + "step": 19404 + }, + { + "epoch": 0.1684447183618198, + "grad_norm": 0.267578125, + "learning_rate": 0.0018383995872114827, + "loss": 0.1025, + "step": 19405 + }, + { + "epoch": 0.16845339884202393, + "grad_norm": 0.408203125, + "learning_rate": 0.0018383825708101502, + "loss": 0.0938, + "step": 19406 + }, + { + "epoch": 0.1684620793222281, + "grad_norm": 0.625, + "learning_rate": 0.0018383655536013338, + "loss": 0.0957, + "step": 19407 + }, + { + "epoch": 0.16847075980243226, + "grad_norm": 0.44140625, + "learning_rate": 0.0018383485355850519, + "loss": 0.0874, + "step": 19408 + }, + { + "epoch": 0.16847944028263642, + "grad_norm": 0.8203125, + "learning_rate": 0.0018383315167613227, + "loss": 0.1406, + "step": 19409 + }, + { + "epoch": 0.1684881207628406, + "grad_norm": 0.16796875, + "learning_rate": 0.0018383144971301653, + "loss": 0.1094, + "step": 19410 + }, + { + "epoch": 0.16849680124304475, + "grad_norm": 0.083984375, + "learning_rate": 0.0018382974766915985, + "loss": 0.1328, + "step": 19411 + }, + { + "epoch": 0.16850548172324892, + "grad_norm": 0.126953125, + "learning_rate": 0.0018382804554456405, + "loss": 0.1787, + "step": 19412 + }, + { + "epoch": 0.16851416220345308, + "grad_norm": 0.21875, + "learning_rate": 0.0018382634333923102, + "loss": 0.1406, + "step": 19413 + }, + { + "epoch": 0.16852284268365725, + "grad_norm": 0.75, + "learning_rate": 0.001838246410531626, + "loss": 0.1309, + "step": 19414 + }, + { + "epoch": 0.1685315231638614, + "grad_norm": 0.333984375, + "learning_rate": 0.001838229386863607, + "loss": 0.0908, + "step": 19415 + }, + { + "epoch": 0.16854020364406558, + "grad_norm": 0.5546875, + "learning_rate": 0.0018382123623882712, + "loss": 0.1182, + "step": 19416 + }, + { + "epoch": 0.16854888412426974, + "grad_norm": 0.234375, + "learning_rate": 0.001838195337105637, + "loss": 0.0874, + "step": 19417 + }, + { + "epoch": 0.1685575646044739, + "grad_norm": 0.193359375, + "learning_rate": 0.0018381783110157238, + "loss": 0.1348, + "step": 19418 + }, + { + "epoch": 0.16856624508467807, + "grad_norm": 0.1103515625, + "learning_rate": 0.00183816128411855, + "loss": 0.0859, + "step": 19419 + }, + { + "epoch": 0.16857492556488224, + "grad_norm": 0.1796875, + "learning_rate": 0.0018381442564141342, + "loss": 0.1523, + "step": 19420 + }, + { + "epoch": 0.1685836060450864, + "grad_norm": 0.1953125, + "learning_rate": 0.0018381272279024948, + "loss": 0.0972, + "step": 19421 + }, + { + "epoch": 0.16859228652529057, + "grad_norm": 0.765625, + "learning_rate": 0.0018381101985836506, + "loss": 0.1113, + "step": 19422 + }, + { + "epoch": 0.16860096700549473, + "grad_norm": 0.28125, + "learning_rate": 0.0018380931684576199, + "loss": 0.123, + "step": 19423 + }, + { + "epoch": 0.1686096474856989, + "grad_norm": 0.28125, + "learning_rate": 0.0018380761375244219, + "loss": 0.0752, + "step": 19424 + }, + { + "epoch": 0.16861832796590306, + "grad_norm": 0.08447265625, + "learning_rate": 0.001838059105784075, + "loss": 0.1123, + "step": 19425 + }, + { + "epoch": 0.16862700844610723, + "grad_norm": 0.33203125, + "learning_rate": 0.0018380420732365974, + "loss": 0.1182, + "step": 19426 + }, + { + "epoch": 0.1686356889263114, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018380250398820084, + "loss": 0.1001, + "step": 19427 + }, + { + "epoch": 0.16864436940651556, + "grad_norm": 0.10693359375, + "learning_rate": 0.001838008005720326, + "loss": 0.1157, + "step": 19428 + }, + { + "epoch": 0.16865304988671972, + "grad_norm": 0.32421875, + "learning_rate": 0.0018379909707515695, + "loss": 0.1279, + "step": 19429 + }, + { + "epoch": 0.1686617303669239, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018379739349757569, + "loss": 0.0967, + "step": 19430 + }, + { + "epoch": 0.16867041084712805, + "grad_norm": 0.671875, + "learning_rate": 0.0018379568983929071, + "loss": 0.1221, + "step": 19431 + }, + { + "epoch": 0.16867909132733222, + "grad_norm": 0.474609375, + "learning_rate": 0.0018379398610030386, + "loss": 0.0801, + "step": 19432 + }, + { + "epoch": 0.16868777180753639, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018379228228061707, + "loss": 0.1099, + "step": 19433 + }, + { + "epoch": 0.16869645228774055, + "grad_norm": 0.06982421875, + "learning_rate": 0.0018379057838023211, + "loss": 0.0977, + "step": 19434 + }, + { + "epoch": 0.16870513276794472, + "grad_norm": 0.0712890625, + "learning_rate": 0.0018378887439915088, + "loss": 0.1104, + "step": 19435 + }, + { + "epoch": 0.16871381324814888, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018378717033737525, + "loss": 0.1279, + "step": 19436 + }, + { + "epoch": 0.16872249372835305, + "grad_norm": 0.328125, + "learning_rate": 0.0018378546619490707, + "loss": 0.1426, + "step": 19437 + }, + { + "epoch": 0.1687311742085572, + "grad_norm": 0.56640625, + "learning_rate": 0.0018378376197174823, + "loss": 0.0962, + "step": 19438 + }, + { + "epoch": 0.16873985468876138, + "grad_norm": 1.5546875, + "learning_rate": 0.0018378205766790058, + "loss": 0.127, + "step": 19439 + }, + { + "epoch": 0.16874853516896554, + "grad_norm": 0.32421875, + "learning_rate": 0.0018378035328336596, + "loss": 0.125, + "step": 19440 + }, + { + "epoch": 0.1687572156491697, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018377864881814626, + "loss": 0.1089, + "step": 19441 + }, + { + "epoch": 0.16876589612937387, + "grad_norm": 0.392578125, + "learning_rate": 0.0018377694427224333, + "loss": 0.1172, + "step": 19442 + }, + { + "epoch": 0.16877457660957804, + "grad_norm": 0.279296875, + "learning_rate": 0.0018377523964565904, + "loss": 0.1162, + "step": 19443 + }, + { + "epoch": 0.1687832570897822, + "grad_norm": 0.1953125, + "learning_rate": 0.001837735349383953, + "loss": 0.1211, + "step": 19444 + }, + { + "epoch": 0.16879193756998637, + "grad_norm": 0.37109375, + "learning_rate": 0.001837718301504539, + "loss": 0.0801, + "step": 19445 + }, + { + "epoch": 0.16880061805019053, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018377012528183673, + "loss": 0.1035, + "step": 19446 + }, + { + "epoch": 0.1688092985303947, + "grad_norm": 0.125, + "learning_rate": 0.0018376842033254565, + "loss": 0.1113, + "step": 19447 + }, + { + "epoch": 0.16881797901059886, + "grad_norm": 0.69921875, + "learning_rate": 0.0018376671530258258, + "loss": 0.1387, + "step": 19448 + }, + { + "epoch": 0.16882665949080303, + "grad_norm": 0.28515625, + "learning_rate": 0.0018376501019194932, + "loss": 0.127, + "step": 19449 + }, + { + "epoch": 0.1688353399710072, + "grad_norm": 0.078125, + "learning_rate": 0.0018376330500064772, + "loss": 0.085, + "step": 19450 + }, + { + "epoch": 0.16884402045121136, + "grad_norm": 0.390625, + "learning_rate": 0.0018376159972867969, + "loss": 0.1621, + "step": 19451 + }, + { + "epoch": 0.16885270093141552, + "grad_norm": 0.5078125, + "learning_rate": 0.001837598943760471, + "loss": 0.0972, + "step": 19452 + }, + { + "epoch": 0.1688613814116197, + "grad_norm": 0.443359375, + "learning_rate": 0.001837581889427518, + "loss": 0.1216, + "step": 19453 + }, + { + "epoch": 0.16887006189182385, + "grad_norm": 0.7109375, + "learning_rate": 0.0018375648342879563, + "loss": 0.1045, + "step": 19454 + }, + { + "epoch": 0.16887874237202802, + "grad_norm": 0.283203125, + "learning_rate": 0.001837547778341805, + "loss": 0.1226, + "step": 19455 + }, + { + "epoch": 0.16888742285223218, + "grad_norm": 0.267578125, + "learning_rate": 0.0018375307215890823, + "loss": 0.0864, + "step": 19456 + }, + { + "epoch": 0.16889610333243635, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018375136640298073, + "loss": 0.1211, + "step": 19457 + }, + { + "epoch": 0.1689047838126405, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018374966056639984, + "loss": 0.1074, + "step": 19458 + }, + { + "epoch": 0.16891346429284468, + "grad_norm": 0.43359375, + "learning_rate": 0.0018374795464916745, + "loss": 0.123, + "step": 19459 + }, + { + "epoch": 0.16892214477304884, + "grad_norm": 0.07958984375, + "learning_rate": 0.0018374624865128534, + "loss": 0.106, + "step": 19460 + }, + { + "epoch": 0.168930825253253, + "grad_norm": 0.71484375, + "learning_rate": 0.001837445425727555, + "loss": 0.1504, + "step": 19461 + }, + { + "epoch": 0.16893950573345717, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018374283641357971, + "loss": 0.0913, + "step": 19462 + }, + { + "epoch": 0.16894818621366134, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018374113017375986, + "loss": 0.1426, + "step": 19463 + }, + { + "epoch": 0.1689568666938655, + "grad_norm": 1.2265625, + "learning_rate": 0.0018373942385329785, + "loss": 0.1133, + "step": 19464 + }, + { + "epoch": 0.16896554717406967, + "grad_norm": 0.5546875, + "learning_rate": 0.0018373771745219547, + "loss": 0.0957, + "step": 19465 + }, + { + "epoch": 0.16897422765427383, + "grad_norm": 0.66796875, + "learning_rate": 0.0018373601097045468, + "loss": 0.123, + "step": 19466 + }, + { + "epoch": 0.168982908134478, + "grad_norm": 0.16015625, + "learning_rate": 0.0018373430440807726, + "loss": 0.0947, + "step": 19467 + }, + { + "epoch": 0.16899158861468216, + "grad_norm": 0.37890625, + "learning_rate": 0.001837325977650651, + "loss": 0.126, + "step": 19468 + }, + { + "epoch": 0.16900026909488633, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018373089104142008, + "loss": 0.1328, + "step": 19469 + }, + { + "epoch": 0.1690089495750905, + "grad_norm": 0.50390625, + "learning_rate": 0.001837291842371441, + "loss": 0.1133, + "step": 19470 + }, + { + "epoch": 0.16901763005529466, + "grad_norm": 0.50390625, + "learning_rate": 0.0018372747735223899, + "loss": 0.1816, + "step": 19471 + }, + { + "epoch": 0.16902631053549882, + "grad_norm": 1.7734375, + "learning_rate": 0.001837257703867066, + "loss": 0.166, + "step": 19472 + }, + { + "epoch": 0.169034991015703, + "grad_norm": 0.474609375, + "learning_rate": 0.0018372406334054882, + "loss": 0.1543, + "step": 19473 + }, + { + "epoch": 0.16904367149590716, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018372235621376752, + "loss": 0.1318, + "step": 19474 + }, + { + "epoch": 0.16905235197611132, + "grad_norm": 0.7421875, + "learning_rate": 0.0018372064900636456, + "loss": 0.1089, + "step": 19475 + }, + { + "epoch": 0.16906103245631549, + "grad_norm": 0.06591796875, + "learning_rate": 0.0018371894171834177, + "loss": 0.0732, + "step": 19476 + }, + { + "epoch": 0.16906971293651965, + "grad_norm": 0.115234375, + "learning_rate": 0.001837172343497011, + "loss": 0.1436, + "step": 19477 + }, + { + "epoch": 0.16907839341672382, + "grad_norm": 0.1787109375, + "learning_rate": 0.0018371552690044435, + "loss": 0.1133, + "step": 19478 + }, + { + "epoch": 0.16908707389692798, + "grad_norm": 0.255859375, + "learning_rate": 0.001837138193705734, + "loss": 0.1016, + "step": 19479 + }, + { + "epoch": 0.16909575437713215, + "grad_norm": 0.150390625, + "learning_rate": 0.0018371211176009012, + "loss": 0.0938, + "step": 19480 + }, + { + "epoch": 0.1691044348573363, + "grad_norm": 0.103515625, + "learning_rate": 0.001837104040689964, + "loss": 0.1016, + "step": 19481 + }, + { + "epoch": 0.16911311533754048, + "grad_norm": 0.60546875, + "learning_rate": 0.001837086962972941, + "loss": 0.103, + "step": 19482 + }, + { + "epoch": 0.16912179581774464, + "grad_norm": 0.7578125, + "learning_rate": 0.0018370698844498506, + "loss": 0.1377, + "step": 19483 + }, + { + "epoch": 0.1691304762979488, + "grad_norm": 0.6640625, + "learning_rate": 0.0018370528051207117, + "loss": 0.125, + "step": 19484 + }, + { + "epoch": 0.16913915677815297, + "grad_norm": 0.26171875, + "learning_rate": 0.0018370357249855426, + "loss": 0.0981, + "step": 19485 + }, + { + "epoch": 0.16914783725835714, + "grad_norm": 0.3984375, + "learning_rate": 0.0018370186440443625, + "loss": 0.123, + "step": 19486 + }, + { + "epoch": 0.1691565177385613, + "grad_norm": 0.1875, + "learning_rate": 0.0018370015622971901, + "loss": 0.1006, + "step": 19487 + }, + { + "epoch": 0.16916519821876547, + "grad_norm": 0.419921875, + "learning_rate": 0.0018369844797440438, + "loss": 0.1582, + "step": 19488 + }, + { + "epoch": 0.16917387869896963, + "grad_norm": 0.5859375, + "learning_rate": 0.0018369673963849426, + "loss": 0.126, + "step": 19489 + }, + { + "epoch": 0.1691825591791738, + "grad_norm": 0.5, + "learning_rate": 0.0018369503122199042, + "loss": 0.1123, + "step": 19490 + }, + { + "epoch": 0.16919123965937796, + "grad_norm": 0.423828125, + "learning_rate": 0.0018369332272489486, + "loss": 0.1514, + "step": 19491 + }, + { + "epoch": 0.16919992013958213, + "grad_norm": 0.7578125, + "learning_rate": 0.0018369161414720937, + "loss": 0.1787, + "step": 19492 + }, + { + "epoch": 0.1692086006197863, + "grad_norm": 0.275390625, + "learning_rate": 0.0018368990548893582, + "loss": 0.0913, + "step": 19493 + }, + { + "epoch": 0.16921728109999046, + "grad_norm": 0.142578125, + "learning_rate": 0.0018368819675007612, + "loss": 0.1055, + "step": 19494 + }, + { + "epoch": 0.16922596158019462, + "grad_norm": 0.0751953125, + "learning_rate": 0.0018368648793063212, + "loss": 0.0894, + "step": 19495 + }, + { + "epoch": 0.1692346420603988, + "grad_norm": 0.30078125, + "learning_rate": 0.0018368477903060567, + "loss": 0.1045, + "step": 19496 + }, + { + "epoch": 0.16924332254060295, + "grad_norm": 0.12109375, + "learning_rate": 0.0018368307004999866, + "loss": 0.0869, + "step": 19497 + }, + { + "epoch": 0.16925200302080712, + "grad_norm": 0.08447265625, + "learning_rate": 0.0018368136098881294, + "loss": 0.1504, + "step": 19498 + }, + { + "epoch": 0.16926068350101128, + "grad_norm": 0.119140625, + "learning_rate": 0.0018367965184705043, + "loss": 0.0747, + "step": 19499 + }, + { + "epoch": 0.16926936398121545, + "grad_norm": 0.205078125, + "learning_rate": 0.0018367794262471291, + "loss": 0.0918, + "step": 19500 + }, + { + "epoch": 0.1692780444614196, + "grad_norm": 0.1484375, + "learning_rate": 0.0018367623332180235, + "loss": 0.1689, + "step": 19501 + }, + { + "epoch": 0.16928672494162378, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018367452393832052, + "loss": 0.1094, + "step": 19502 + }, + { + "epoch": 0.16929540542182794, + "grad_norm": 0.2578125, + "learning_rate": 0.0018367281447426932, + "loss": 0.1406, + "step": 19503 + }, + { + "epoch": 0.1693040859020321, + "grad_norm": 0.53125, + "learning_rate": 0.001836711049296507, + "loss": 0.0918, + "step": 19504 + }, + { + "epoch": 0.16931276638223627, + "grad_norm": 0.333984375, + "learning_rate": 0.0018366939530446644, + "loss": 0.1104, + "step": 19505 + }, + { + "epoch": 0.16932144686244044, + "grad_norm": 0.27734375, + "learning_rate": 0.0018366768559871843, + "loss": 0.1118, + "step": 19506 + }, + { + "epoch": 0.1693301273426446, + "grad_norm": 0.33203125, + "learning_rate": 0.0018366597581240855, + "loss": 0.124, + "step": 19507 + }, + { + "epoch": 0.16933880782284877, + "grad_norm": 0.2578125, + "learning_rate": 0.0018366426594553867, + "loss": 0.1172, + "step": 19508 + }, + { + "epoch": 0.16934748830305293, + "grad_norm": 0.1064453125, + "learning_rate": 0.0018366255599811063, + "loss": 0.127, + "step": 19509 + }, + { + "epoch": 0.1693561687832571, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018366084597012638, + "loss": 0.1123, + "step": 19510 + }, + { + "epoch": 0.16936484926346126, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018365913586158768, + "loss": 0.1001, + "step": 19511 + }, + { + "epoch": 0.16937352974366543, + "grad_norm": 0.296875, + "learning_rate": 0.0018365742567249647, + "loss": 0.0771, + "step": 19512 + }, + { + "epoch": 0.1693822102238696, + "grad_norm": 0.8515625, + "learning_rate": 0.0018365571540285464, + "loss": 0.1221, + "step": 19513 + }, + { + "epoch": 0.16939089070407376, + "grad_norm": 0.380859375, + "learning_rate": 0.0018365400505266402, + "loss": 0.1445, + "step": 19514 + }, + { + "epoch": 0.16939957118427793, + "grad_norm": 0.314453125, + "learning_rate": 0.0018365229462192643, + "loss": 0.1084, + "step": 19515 + }, + { + "epoch": 0.1694082516644821, + "grad_norm": 0.482421875, + "learning_rate": 0.0018365058411064386, + "loss": 0.0967, + "step": 19516 + }, + { + "epoch": 0.16941693214468626, + "grad_norm": 0.212890625, + "learning_rate": 0.0018364887351881808, + "loss": 0.0771, + "step": 19517 + }, + { + "epoch": 0.16942561262489042, + "grad_norm": 0.1279296875, + "learning_rate": 0.00183647162846451, + "loss": 0.1436, + "step": 19518 + }, + { + "epoch": 0.16943429310509459, + "grad_norm": 0.2578125, + "learning_rate": 0.001836454520935445, + "loss": 0.0737, + "step": 19519 + }, + { + "epoch": 0.16944297358529875, + "grad_norm": 0.318359375, + "learning_rate": 0.0018364374126010046, + "loss": 0.126, + "step": 19520 + }, + { + "epoch": 0.16945165406550292, + "grad_norm": 0.11669921875, + "learning_rate": 0.001836420303461207, + "loss": 0.1074, + "step": 19521 + }, + { + "epoch": 0.16946033454570708, + "grad_norm": 0.263671875, + "learning_rate": 0.0018364031935160713, + "loss": 0.2031, + "step": 19522 + }, + { + "epoch": 0.16946901502591125, + "grad_norm": 0.302734375, + "learning_rate": 0.0018363860827656162, + "loss": 0.0938, + "step": 19523 + }, + { + "epoch": 0.1694776955061154, + "grad_norm": 0.65234375, + "learning_rate": 0.0018363689712098603, + "loss": 0.1055, + "step": 19524 + }, + { + "epoch": 0.16948637598631958, + "grad_norm": 0.58984375, + "learning_rate": 0.0018363518588488223, + "loss": 0.125, + "step": 19525 + }, + { + "epoch": 0.16949505646652374, + "grad_norm": 0.6171875, + "learning_rate": 0.001836334745682521, + "loss": 0.0942, + "step": 19526 + }, + { + "epoch": 0.1695037369467279, + "grad_norm": 0.34375, + "learning_rate": 0.001836317631710975, + "loss": 0.1025, + "step": 19527 + }, + { + "epoch": 0.16951241742693207, + "grad_norm": 0.953125, + "learning_rate": 0.0018363005169342033, + "loss": 0.0698, + "step": 19528 + }, + { + "epoch": 0.1695210979071362, + "grad_norm": 0.423828125, + "learning_rate": 0.0018362834013522244, + "loss": 0.1396, + "step": 19529 + }, + { + "epoch": 0.16952977838734037, + "grad_norm": 0.287109375, + "learning_rate": 0.001836266284965057, + "loss": 0.0811, + "step": 19530 + }, + { + "epoch": 0.16953845886754454, + "grad_norm": 0.62890625, + "learning_rate": 0.00183624916777272, + "loss": 0.1426, + "step": 19531 + }, + { + "epoch": 0.1695471393477487, + "grad_norm": 0.16796875, + "learning_rate": 0.0018362320497752318, + "loss": 0.1465, + "step": 19532 + }, + { + "epoch": 0.16955581982795287, + "grad_norm": 0.119140625, + "learning_rate": 0.0018362149309726114, + "loss": 0.1211, + "step": 19533 + }, + { + "epoch": 0.16956450030815703, + "grad_norm": 0.298828125, + "learning_rate": 0.0018361978113648771, + "loss": 0.1196, + "step": 19534 + }, + { + "epoch": 0.1695731807883612, + "grad_norm": 1.015625, + "learning_rate": 0.0018361806909520484, + "loss": 0.1045, + "step": 19535 + }, + { + "epoch": 0.16958186126856536, + "grad_norm": 0.490234375, + "learning_rate": 0.001836163569734143, + "loss": 0.1309, + "step": 19536 + }, + { + "epoch": 0.16959054174876953, + "grad_norm": 0.431640625, + "learning_rate": 0.0018361464477111807, + "loss": 0.1182, + "step": 19537 + }, + { + "epoch": 0.1695992222289737, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018361293248831793, + "loss": 0.0811, + "step": 19538 + }, + { + "epoch": 0.16960790270917786, + "grad_norm": 0.890625, + "learning_rate": 0.001836112201250158, + "loss": 0.1377, + "step": 19539 + }, + { + "epoch": 0.16961658318938203, + "grad_norm": 0.421875, + "learning_rate": 0.0018360950768121357, + "loss": 0.125, + "step": 19540 + }, + { + "epoch": 0.1696252636695862, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018360779515691308, + "loss": 0.0845, + "step": 19541 + }, + { + "epoch": 0.16963394414979036, + "grad_norm": 0.40625, + "learning_rate": 0.001836060825521162, + "loss": 0.1416, + "step": 19542 + }, + { + "epoch": 0.16964262462999452, + "grad_norm": 0.50390625, + "learning_rate": 0.0018360436986682483, + "loss": 0.1094, + "step": 19543 + }, + { + "epoch": 0.16965130511019869, + "grad_norm": 0.302734375, + "learning_rate": 0.0018360265710104082, + "loss": 0.1079, + "step": 19544 + }, + { + "epoch": 0.16965998559040285, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018360094425476606, + "loss": 0.1426, + "step": 19545 + }, + { + "epoch": 0.16966866607060702, + "grad_norm": 0.40234375, + "learning_rate": 0.0018359923132800237, + "loss": 0.1079, + "step": 19546 + }, + { + "epoch": 0.16967734655081118, + "grad_norm": 0.29296875, + "learning_rate": 0.0018359751832075172, + "loss": 0.1367, + "step": 19547 + }, + { + "epoch": 0.16968602703101535, + "grad_norm": 0.56640625, + "learning_rate": 0.001835958052330159, + "loss": 0.0996, + "step": 19548 + }, + { + "epoch": 0.1696947075112195, + "grad_norm": 0.27734375, + "learning_rate": 0.0018359409206479683, + "loss": 0.083, + "step": 19549 + }, + { + "epoch": 0.16970338799142368, + "grad_norm": 0.201171875, + "learning_rate": 0.0018359237881609635, + "loss": 0.0967, + "step": 19550 + }, + { + "epoch": 0.16971206847162784, + "grad_norm": 0.65234375, + "learning_rate": 0.0018359066548691635, + "loss": 0.1084, + "step": 19551 + }, + { + "epoch": 0.169720748951832, + "grad_norm": 0.130859375, + "learning_rate": 0.0018358895207725872, + "loss": 0.0806, + "step": 19552 + }, + { + "epoch": 0.16972942943203617, + "grad_norm": 0.365234375, + "learning_rate": 0.0018358723858712532, + "loss": 0.1084, + "step": 19553 + }, + { + "epoch": 0.16973810991224034, + "grad_norm": 0.6015625, + "learning_rate": 0.00183585525016518, + "loss": 0.0996, + "step": 19554 + }, + { + "epoch": 0.1697467903924445, + "grad_norm": 0.1064453125, + "learning_rate": 0.0018358381136543867, + "loss": 0.1221, + "step": 19555 + }, + { + "epoch": 0.16975547087264867, + "grad_norm": 0.5, + "learning_rate": 0.0018358209763388918, + "loss": 0.2852, + "step": 19556 + }, + { + "epoch": 0.16976415135285283, + "grad_norm": 0.15234375, + "learning_rate": 0.0018358038382187143, + "loss": 0.2285, + "step": 19557 + }, + { + "epoch": 0.169772831833057, + "grad_norm": 0.58984375, + "learning_rate": 0.0018357866992938726, + "loss": 0.125, + "step": 19558 + }, + { + "epoch": 0.16978151231326116, + "grad_norm": 0.30859375, + "learning_rate": 0.0018357695595643858, + "loss": 0.0742, + "step": 19559 + }, + { + "epoch": 0.16979019279346533, + "grad_norm": 0.1640625, + "learning_rate": 0.001835752419030272, + "loss": 0.0854, + "step": 19560 + }, + { + "epoch": 0.1697988732736695, + "grad_norm": 0.73828125, + "learning_rate": 0.001835735277691551, + "loss": 0.1455, + "step": 19561 + }, + { + "epoch": 0.16980755375387366, + "grad_norm": 0.0849609375, + "learning_rate": 0.0018357181355482409, + "loss": 0.1138, + "step": 19562 + }, + { + "epoch": 0.16981623423407782, + "grad_norm": 0.5078125, + "learning_rate": 0.00183570099260036, + "loss": 0.1094, + "step": 19563 + }, + { + "epoch": 0.169824914714282, + "grad_norm": 0.2353515625, + "learning_rate": 0.001835683848847928, + "loss": 0.1338, + "step": 19564 + }, + { + "epoch": 0.16983359519448615, + "grad_norm": 0.09033203125, + "learning_rate": 0.001835666704290963, + "loss": 0.0854, + "step": 19565 + }, + { + "epoch": 0.16984227567469032, + "grad_norm": 0.1416015625, + "learning_rate": 0.001835649558929484, + "loss": 0.105, + "step": 19566 + }, + { + "epoch": 0.16985095615489448, + "grad_norm": 0.640625, + "learning_rate": 0.0018356324127635098, + "loss": 0.1064, + "step": 19567 + }, + { + "epoch": 0.16985963663509865, + "grad_norm": 0.087890625, + "learning_rate": 0.0018356152657930589, + "loss": 0.0825, + "step": 19568 + }, + { + "epoch": 0.1698683171153028, + "grad_norm": 0.26953125, + "learning_rate": 0.0018355981180181503, + "loss": 0.0986, + "step": 19569 + }, + { + "epoch": 0.16987699759550698, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018355809694388023, + "loss": 0.1045, + "step": 19570 + }, + { + "epoch": 0.16988567807571114, + "grad_norm": 0.984375, + "learning_rate": 0.0018355638200550346, + "loss": 0.1406, + "step": 19571 + }, + { + "epoch": 0.1698943585559153, + "grad_norm": 0.357421875, + "learning_rate": 0.001835546669866865, + "loss": 0.127, + "step": 19572 + }, + { + "epoch": 0.16990303903611947, + "grad_norm": 0.5, + "learning_rate": 0.001835529518874313, + "loss": 0.1328, + "step": 19573 + }, + { + "epoch": 0.16991171951632364, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018355123670773966, + "loss": 0.1924, + "step": 19574 + }, + { + "epoch": 0.1699203999965278, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018354952144761347, + "loss": 0.106, + "step": 19575 + }, + { + "epoch": 0.16992908047673197, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018354780610705466, + "loss": 0.1069, + "step": 19576 + }, + { + "epoch": 0.16993776095693613, + "grad_norm": 0.125, + "learning_rate": 0.0018354609068606507, + "loss": 0.1108, + "step": 19577 + }, + { + "epoch": 0.1699464414371403, + "grad_norm": 0.3203125, + "learning_rate": 0.0018354437518464657, + "loss": 0.1074, + "step": 19578 + }, + { + "epoch": 0.16995512191734446, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018354265960280108, + "loss": 0.1045, + "step": 19579 + }, + { + "epoch": 0.16996380239754863, + "grad_norm": 0.291015625, + "learning_rate": 0.0018354094394053041, + "loss": 0.1074, + "step": 19580 + }, + { + "epoch": 0.1699724828777528, + "grad_norm": 0.12353515625, + "learning_rate": 0.0018353922819783646, + "loss": 0.0659, + "step": 19581 + }, + { + "epoch": 0.16998116335795696, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018353751237472117, + "loss": 0.1108, + "step": 19582 + }, + { + "epoch": 0.16998984383816113, + "grad_norm": 0.6640625, + "learning_rate": 0.001835357964711863, + "loss": 0.1494, + "step": 19583 + }, + { + "epoch": 0.1699985243183653, + "grad_norm": 0.158203125, + "learning_rate": 0.001835340804872338, + "loss": 0.1108, + "step": 19584 + }, + { + "epoch": 0.17000720479856946, + "grad_norm": 0.369140625, + "learning_rate": 0.0018353236442286557, + "loss": 0.1406, + "step": 19585 + }, + { + "epoch": 0.17001588527877362, + "grad_norm": 0.4296875, + "learning_rate": 0.001835306482780834, + "loss": 0.1328, + "step": 19586 + }, + { + "epoch": 0.17002456575897779, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018352893205288929, + "loss": 0.0894, + "step": 19587 + }, + { + "epoch": 0.17003324623918195, + "grad_norm": 0.8046875, + "learning_rate": 0.0018352721574728496, + "loss": 0.1157, + "step": 19588 + }, + { + "epoch": 0.17004192671938612, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018352549936127241, + "loss": 0.1279, + "step": 19589 + }, + { + "epoch": 0.17005060719959028, + "grad_norm": 0.2392578125, + "learning_rate": 0.001835237828948535, + "loss": 0.1309, + "step": 19590 + }, + { + "epoch": 0.17005928767979445, + "grad_norm": 0.361328125, + "learning_rate": 0.0018352206634803007, + "loss": 0.1089, + "step": 19591 + }, + { + "epoch": 0.1700679681599986, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018352034972080398, + "loss": 0.1118, + "step": 19592 + }, + { + "epoch": 0.17007664864020278, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018351863301317717, + "loss": 0.0786, + "step": 19593 + }, + { + "epoch": 0.17008532912040694, + "grad_norm": 0.275390625, + "learning_rate": 0.0018351691622515152, + "loss": 0.0845, + "step": 19594 + }, + { + "epoch": 0.1700940096006111, + "grad_norm": 0.1064453125, + "learning_rate": 0.0018351519935672883, + "loss": 0.1367, + "step": 19595 + }, + { + "epoch": 0.17010269008081527, + "grad_norm": 0.16015625, + "learning_rate": 0.00183513482407911, + "loss": 0.1128, + "step": 19596 + }, + { + "epoch": 0.17011137056101944, + "grad_norm": 0.3203125, + "learning_rate": 0.001835117653787, + "loss": 0.1143, + "step": 19597 + }, + { + "epoch": 0.1701200510412236, + "grad_norm": 0.337890625, + "learning_rate": 0.001835100482690976, + "loss": 0.0762, + "step": 19598 + }, + { + "epoch": 0.17012873152142777, + "grad_norm": 0.6640625, + "learning_rate": 0.001835083310791057, + "loss": 0.2793, + "step": 19599 + }, + { + "epoch": 0.17013741200163193, + "grad_norm": 0.265625, + "learning_rate": 0.0018350661380872621, + "loss": 0.1074, + "step": 19600 + }, + { + "epoch": 0.1701460924818361, + "grad_norm": 0.515625, + "learning_rate": 0.00183504896457961, + "loss": 0.1426, + "step": 19601 + }, + { + "epoch": 0.17015477296204026, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018350317902681194, + "loss": 0.124, + "step": 19602 + }, + { + "epoch": 0.17016345344224443, + "grad_norm": 0.58984375, + "learning_rate": 0.0018350146151528092, + "loss": 0.1309, + "step": 19603 + }, + { + "epoch": 0.1701721339224486, + "grad_norm": 0.546875, + "learning_rate": 0.0018349974392336977, + "loss": 0.0967, + "step": 19604 + }, + { + "epoch": 0.17018081440265276, + "grad_norm": 0.51953125, + "learning_rate": 0.001834980262510804, + "loss": 0.1064, + "step": 19605 + }, + { + "epoch": 0.17018949488285692, + "grad_norm": 0.255859375, + "learning_rate": 0.0018349630849841471, + "loss": 0.1074, + "step": 19606 + }, + { + "epoch": 0.1701981753630611, + "grad_norm": 0.65234375, + "learning_rate": 0.001834945906653746, + "loss": 0.123, + "step": 19607 + }, + { + "epoch": 0.17020685584326525, + "grad_norm": 1.1015625, + "learning_rate": 0.0018349287275196188, + "loss": 0.1621, + "step": 19608 + }, + { + "epoch": 0.17021553632346942, + "grad_norm": 0.34765625, + "learning_rate": 0.0018349115475817844, + "loss": 0.1309, + "step": 19609 + }, + { + "epoch": 0.17022421680367358, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018348943668402618, + "loss": 0.1099, + "step": 19610 + }, + { + "epoch": 0.17023289728387775, + "grad_norm": 0.419921875, + "learning_rate": 0.0018348771852950698, + "loss": 0.0991, + "step": 19611 + }, + { + "epoch": 0.17024157776408191, + "grad_norm": 0.8046875, + "learning_rate": 0.0018348600029462275, + "loss": 0.6641, + "step": 19612 + }, + { + "epoch": 0.17025025824428608, + "grad_norm": 0.380859375, + "learning_rate": 0.0018348428197937528, + "loss": 0.1211, + "step": 19613 + }, + { + "epoch": 0.17025893872449024, + "grad_norm": 0.265625, + "learning_rate": 0.0018348256358376653, + "loss": 0.1128, + "step": 19614 + }, + { + "epoch": 0.1702676192046944, + "grad_norm": 0.130859375, + "learning_rate": 0.0018348084510779834, + "loss": 0.083, + "step": 19615 + }, + { + "epoch": 0.17027629968489857, + "grad_norm": 0.9296875, + "learning_rate": 0.0018347912655147262, + "loss": 0.1406, + "step": 19616 + }, + { + "epoch": 0.17028498016510274, + "grad_norm": 0.275390625, + "learning_rate": 0.0018347740791479124, + "loss": 0.1152, + "step": 19617 + }, + { + "epoch": 0.1702936606453069, + "grad_norm": 1.65625, + "learning_rate": 0.0018347568919775603, + "loss": 0.0918, + "step": 19618 + }, + { + "epoch": 0.17030234112551107, + "grad_norm": 0.21484375, + "learning_rate": 0.0018347397040036895, + "loss": 0.1396, + "step": 19619 + }, + { + "epoch": 0.17031102160571523, + "grad_norm": 0.1875, + "learning_rate": 0.0018347225152263184, + "loss": 0.1104, + "step": 19620 + }, + { + "epoch": 0.1703197020859194, + "grad_norm": 0.271484375, + "learning_rate": 0.0018347053256454657, + "loss": 0.1914, + "step": 19621 + }, + { + "epoch": 0.17032838256612357, + "grad_norm": 0.51953125, + "learning_rate": 0.00183468813526115, + "loss": 0.082, + "step": 19622 + }, + { + "epoch": 0.17033706304632773, + "grad_norm": 0.1796875, + "learning_rate": 0.0018346709440733907, + "loss": 0.0781, + "step": 19623 + }, + { + "epoch": 0.1703457435265319, + "grad_norm": 0.2431640625, + "learning_rate": 0.0018346537520822065, + "loss": 0.1113, + "step": 19624 + }, + { + "epoch": 0.17035442400673606, + "grad_norm": 0.26171875, + "learning_rate": 0.0018346365592876157, + "loss": 0.0947, + "step": 19625 + }, + { + "epoch": 0.17036310448694023, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018346193656896372, + "loss": 0.0898, + "step": 19626 + }, + { + "epoch": 0.1703717849671444, + "grad_norm": 0.255859375, + "learning_rate": 0.0018346021712882903, + "loss": 0.1279, + "step": 19627 + }, + { + "epoch": 0.17038046544734856, + "grad_norm": 0.67578125, + "learning_rate": 0.0018345849760835935, + "loss": 0.125, + "step": 19628 + }, + { + "epoch": 0.17038914592755272, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018345677800755656, + "loss": 0.0806, + "step": 19629 + }, + { + "epoch": 0.17039782640775689, + "grad_norm": 0.53125, + "learning_rate": 0.0018345505832642252, + "loss": 0.1035, + "step": 19630 + }, + { + "epoch": 0.17040650688796105, + "grad_norm": 0.08251953125, + "learning_rate": 0.0018345333856495914, + "loss": 0.0845, + "step": 19631 + }, + { + "epoch": 0.17041518736816522, + "grad_norm": 0.12060546875, + "learning_rate": 0.001834516187231683, + "loss": 0.1191, + "step": 19632 + }, + { + "epoch": 0.17042386784836938, + "grad_norm": 0.275390625, + "learning_rate": 0.0018344989880105188, + "loss": 0.0859, + "step": 19633 + }, + { + "epoch": 0.17043254832857355, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018344817879861175, + "loss": 0.125, + "step": 19634 + }, + { + "epoch": 0.1704412288087777, + "grad_norm": 0.142578125, + "learning_rate": 0.0018344645871584979, + "loss": 0.1064, + "step": 19635 + }, + { + "epoch": 0.17044990928898188, + "grad_norm": 0.158203125, + "learning_rate": 0.001834447385527679, + "loss": 0.1045, + "step": 19636 + }, + { + "epoch": 0.17045858976918604, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018344301830936793, + "loss": 0.103, + "step": 19637 + }, + { + "epoch": 0.1704672702493902, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018344129798565179, + "loss": 0.1406, + "step": 19638 + }, + { + "epoch": 0.17047595072959437, + "grad_norm": 0.12353515625, + "learning_rate": 0.0018343957758162135, + "loss": 0.082, + "step": 19639 + }, + { + "epoch": 0.17048463120979854, + "grad_norm": 0.671875, + "learning_rate": 0.0018343785709727847, + "loss": 0.1289, + "step": 19640 + }, + { + "epoch": 0.1704933116900027, + "grad_norm": 0.2451171875, + "learning_rate": 0.0018343613653262509, + "loss": 0.127, + "step": 19641 + }, + { + "epoch": 0.17050199217020687, + "grad_norm": 0.1611328125, + "learning_rate": 0.00183434415887663, + "loss": 0.1387, + "step": 19642 + }, + { + "epoch": 0.17051067265041103, + "grad_norm": 0.115234375, + "learning_rate": 0.0018343269516239418, + "loss": 0.1328, + "step": 19643 + }, + { + "epoch": 0.1705193531306152, + "grad_norm": 0.1904296875, + "learning_rate": 0.0018343097435682047, + "loss": 0.1162, + "step": 19644 + }, + { + "epoch": 0.17052803361081936, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018342925347094373, + "loss": 0.1201, + "step": 19645 + }, + { + "epoch": 0.17053671409102353, + "grad_norm": 0.345703125, + "learning_rate": 0.0018342753250476587, + "loss": 0.1162, + "step": 19646 + }, + { + "epoch": 0.1705453945712277, + "grad_norm": 0.353515625, + "learning_rate": 0.0018342581145828877, + "loss": 0.123, + "step": 19647 + }, + { + "epoch": 0.17055407505143186, + "grad_norm": 0.123046875, + "learning_rate": 0.0018342409033151427, + "loss": 0.0908, + "step": 19648 + }, + { + "epoch": 0.17056275553163602, + "grad_norm": 0.1640625, + "learning_rate": 0.0018342236912444435, + "loss": 0.1387, + "step": 19649 + }, + { + "epoch": 0.1705714360118402, + "grad_norm": 0.6484375, + "learning_rate": 0.001834206478370808, + "loss": 0.207, + "step": 19650 + }, + { + "epoch": 0.17058011649204435, + "grad_norm": 0.20703125, + "learning_rate": 0.001834189264694255, + "loss": 0.083, + "step": 19651 + }, + { + "epoch": 0.1705887969722485, + "grad_norm": 0.09130859375, + "learning_rate": 0.0018341720502148041, + "loss": 0.1396, + "step": 19652 + }, + { + "epoch": 0.17059747745245266, + "grad_norm": 0.1689453125, + "learning_rate": 0.0018341548349324736, + "loss": 0.082, + "step": 19653 + }, + { + "epoch": 0.17060615793265682, + "grad_norm": 0.216796875, + "learning_rate": 0.0018341376188472825, + "loss": 0.1074, + "step": 19654 + }, + { + "epoch": 0.170614838412861, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018341204019592492, + "loss": 0.1396, + "step": 19655 + }, + { + "epoch": 0.17062351889306515, + "grad_norm": 0.16796875, + "learning_rate": 0.0018341031842683928, + "loss": 0.0996, + "step": 19656 + }, + { + "epoch": 0.17063219937326932, + "grad_norm": 0.4140625, + "learning_rate": 0.0018340859657747324, + "loss": 0.1064, + "step": 19657 + }, + { + "epoch": 0.17064087985347348, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018340687464782866, + "loss": 0.1377, + "step": 19658 + }, + { + "epoch": 0.17064956033367765, + "grad_norm": 0.390625, + "learning_rate": 0.0018340515263790742, + "loss": 0.0938, + "step": 19659 + }, + { + "epoch": 0.1706582408138818, + "grad_norm": 0.5, + "learning_rate": 0.0018340343054771143, + "loss": 0.1172, + "step": 19660 + }, + { + "epoch": 0.17066692129408598, + "grad_norm": 0.578125, + "learning_rate": 0.0018340170837724253, + "loss": 0.127, + "step": 19661 + }, + { + "epoch": 0.17067560177429014, + "grad_norm": 0.333984375, + "learning_rate": 0.0018339998612650263, + "loss": 0.1074, + "step": 19662 + }, + { + "epoch": 0.1706842822544943, + "grad_norm": 0.515625, + "learning_rate": 0.0018339826379549363, + "loss": 0.1484, + "step": 19663 + }, + { + "epoch": 0.17069296273469847, + "grad_norm": 0.251953125, + "learning_rate": 0.0018339654138421735, + "loss": 0.1084, + "step": 19664 + }, + { + "epoch": 0.17070164321490264, + "grad_norm": 0.435546875, + "learning_rate": 0.0018339481889267574, + "loss": 0.1299, + "step": 19665 + }, + { + "epoch": 0.1707103236951068, + "grad_norm": 0.208984375, + "learning_rate": 0.0018339309632087065, + "loss": 0.103, + "step": 19666 + }, + { + "epoch": 0.17071900417531097, + "grad_norm": 0.0654296875, + "learning_rate": 0.00183391373668804, + "loss": 0.0713, + "step": 19667 + }, + { + "epoch": 0.17072768465551513, + "grad_norm": 0.09765625, + "learning_rate": 0.001833896509364776, + "loss": 0.0884, + "step": 19668 + }, + { + "epoch": 0.1707363651357193, + "grad_norm": 0.59765625, + "learning_rate": 0.0018338792812389341, + "loss": 0.1143, + "step": 19669 + }, + { + "epoch": 0.17074504561592346, + "grad_norm": 0.640625, + "learning_rate": 0.0018338620523105328, + "loss": 0.1797, + "step": 19670 + }, + { + "epoch": 0.17075372609612763, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018338448225795909, + "loss": 0.1514, + "step": 19671 + }, + { + "epoch": 0.1707624065763318, + "grad_norm": 0.515625, + "learning_rate": 0.0018338275920461273, + "loss": 0.1357, + "step": 19672 + }, + { + "epoch": 0.17077108705653596, + "grad_norm": 0.439453125, + "learning_rate": 0.0018338103607101611, + "loss": 0.085, + "step": 19673 + }, + { + "epoch": 0.17077976753674012, + "grad_norm": 0.443359375, + "learning_rate": 0.0018337931285717108, + "loss": 0.1152, + "step": 19674 + }, + { + "epoch": 0.1707884480169443, + "grad_norm": 0.26953125, + "learning_rate": 0.0018337758956307954, + "loss": 0.1826, + "step": 19675 + }, + { + "epoch": 0.17079712849714845, + "grad_norm": 0.28125, + "learning_rate": 0.0018337586618874335, + "loss": 0.1445, + "step": 19676 + }, + { + "epoch": 0.17080580897735262, + "grad_norm": 0.306640625, + "learning_rate": 0.0018337414273416443, + "loss": 0.127, + "step": 19677 + }, + { + "epoch": 0.17081448945755678, + "grad_norm": 0.21875, + "learning_rate": 0.0018337241919934466, + "loss": 0.1133, + "step": 19678 + }, + { + "epoch": 0.17082316993776095, + "grad_norm": 0.287109375, + "learning_rate": 0.001833706955842859, + "loss": 0.1289, + "step": 19679 + }, + { + "epoch": 0.17083185041796511, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018336897188899004, + "loss": 0.0913, + "step": 19680 + }, + { + "epoch": 0.17084053089816928, + "grad_norm": 0.125, + "learning_rate": 0.0018336724811345902, + "loss": 0.0889, + "step": 19681 + }, + { + "epoch": 0.17084921137837344, + "grad_norm": 0.150390625, + "learning_rate": 0.0018336552425769464, + "loss": 0.127, + "step": 19682 + }, + { + "epoch": 0.1708578918585776, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018336380032169884, + "loss": 0.0835, + "step": 19683 + }, + { + "epoch": 0.17086657233878177, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018336207630547346, + "loss": 0.1504, + "step": 19684 + }, + { + "epoch": 0.17087525281898594, + "grad_norm": 0.99609375, + "learning_rate": 0.0018336035220902045, + "loss": 0.1191, + "step": 19685 + }, + { + "epoch": 0.1708839332991901, + "grad_norm": 0.310546875, + "learning_rate": 0.0018335862803234165, + "loss": 0.0723, + "step": 19686 + }, + { + "epoch": 0.17089261377939427, + "grad_norm": 0.318359375, + "learning_rate": 0.0018335690377543896, + "loss": 0.1045, + "step": 19687 + }, + { + "epoch": 0.17090129425959844, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018335517943831423, + "loss": 0.1045, + "step": 19688 + }, + { + "epoch": 0.1709099747398026, + "grad_norm": 0.404296875, + "learning_rate": 0.0018335345502096946, + "loss": 0.1094, + "step": 19689 + }, + { + "epoch": 0.17091865522000677, + "grad_norm": 0.162109375, + "learning_rate": 0.0018335173052340637, + "loss": 0.0977, + "step": 19690 + }, + { + "epoch": 0.17092733570021093, + "grad_norm": 0.134765625, + "learning_rate": 0.0018335000594562694, + "loss": 0.1143, + "step": 19691 + }, + { + "epoch": 0.1709360161804151, + "grad_norm": 0.09423828125, + "learning_rate": 0.001833482812876331, + "loss": 0.0928, + "step": 19692 + }, + { + "epoch": 0.17094469666061926, + "grad_norm": 0.09912109375, + "learning_rate": 0.0018334655654942665, + "loss": 0.0991, + "step": 19693 + }, + { + "epoch": 0.17095337714082343, + "grad_norm": 0.1787109375, + "learning_rate": 0.001833448317310095, + "loss": 0.1094, + "step": 19694 + }, + { + "epoch": 0.1709620576210276, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018334310683238353, + "loss": 0.1079, + "step": 19695 + }, + { + "epoch": 0.17097073810123176, + "grad_norm": 0.74609375, + "learning_rate": 0.0018334138185355066, + "loss": 0.1035, + "step": 19696 + }, + { + "epoch": 0.17097941858143592, + "grad_norm": 0.11376953125, + "learning_rate": 0.0018333965679451277, + "loss": 0.1152, + "step": 19697 + }, + { + "epoch": 0.1709880990616401, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018333793165527172, + "loss": 0.1367, + "step": 19698 + }, + { + "epoch": 0.17099677954184425, + "grad_norm": 0.68359375, + "learning_rate": 0.0018333620643582941, + "loss": 0.0952, + "step": 19699 + }, + { + "epoch": 0.17100546002204842, + "grad_norm": 0.37890625, + "learning_rate": 0.0018333448113618772, + "loss": 0.0947, + "step": 19700 + }, + { + "epoch": 0.17101414050225258, + "grad_norm": 0.072265625, + "learning_rate": 0.0018333275575634854, + "loss": 0.1162, + "step": 19701 + }, + { + "epoch": 0.17102282098245675, + "grad_norm": 0.51953125, + "learning_rate": 0.0018333103029631378, + "loss": 0.1455, + "step": 19702 + }, + { + "epoch": 0.1710315014626609, + "grad_norm": 0.173828125, + "learning_rate": 0.0018332930475608532, + "loss": 0.1289, + "step": 19703 + }, + { + "epoch": 0.17104018194286508, + "grad_norm": 0.263671875, + "learning_rate": 0.00183327579135665, + "loss": 0.1582, + "step": 19704 + }, + { + "epoch": 0.17104886242306924, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018332585343505475, + "loss": 0.1357, + "step": 19705 + }, + { + "epoch": 0.1710575429032734, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018332412765425642, + "loss": 0.0771, + "step": 19706 + }, + { + "epoch": 0.17106622338347757, + "grad_norm": 0.859375, + "learning_rate": 0.0018332240179327198, + "loss": 0.1064, + "step": 19707 + }, + { + "epoch": 0.17107490386368174, + "grad_norm": 0.123046875, + "learning_rate": 0.0018332067585210324, + "loss": 0.1152, + "step": 19708 + }, + { + "epoch": 0.1710835843438859, + "grad_norm": 1.0234375, + "learning_rate": 0.001833189498307521, + "loss": 0.0977, + "step": 19709 + }, + { + "epoch": 0.17109226482409007, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018331722372922046, + "loss": 0.1504, + "step": 19710 + }, + { + "epoch": 0.17110094530429423, + "grad_norm": 0.10546875, + "learning_rate": 0.0018331549754751022, + "loss": 0.1113, + "step": 19711 + }, + { + "epoch": 0.1711096257844984, + "grad_norm": 0.29296875, + "learning_rate": 0.0018331377128562325, + "loss": 0.1348, + "step": 19712 + }, + { + "epoch": 0.17111830626470256, + "grad_norm": 0.330078125, + "learning_rate": 0.0018331204494356143, + "loss": 0.1387, + "step": 19713 + }, + { + "epoch": 0.17112698674490673, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018331031852132666, + "loss": 0.0708, + "step": 19714 + }, + { + "epoch": 0.1711356672251109, + "grad_norm": 0.111328125, + "learning_rate": 0.0018330859201892086, + "loss": 0.0977, + "step": 19715 + }, + { + "epoch": 0.17114434770531506, + "grad_norm": 0.1845703125, + "learning_rate": 0.0018330686543634586, + "loss": 0.0879, + "step": 19716 + }, + { + "epoch": 0.17115302818551922, + "grad_norm": 0.2265625, + "learning_rate": 0.0018330513877360356, + "loss": 0.0781, + "step": 19717 + }, + { + "epoch": 0.1711617086657234, + "grad_norm": 0.271484375, + "learning_rate": 0.0018330341203069587, + "loss": 0.1069, + "step": 19718 + }, + { + "epoch": 0.17117038914592755, + "grad_norm": 0.140625, + "learning_rate": 0.0018330168520762466, + "loss": 0.1484, + "step": 19719 + }, + { + "epoch": 0.17117906962613172, + "grad_norm": 0.85546875, + "learning_rate": 0.0018329995830439186, + "loss": 0.0996, + "step": 19720 + }, + { + "epoch": 0.17118775010633588, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018329823132099932, + "loss": 0.1045, + "step": 19721 + }, + { + "epoch": 0.17119643058654005, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018329650425744893, + "loss": 0.1475, + "step": 19722 + }, + { + "epoch": 0.17120511106674421, + "grad_norm": 0.22265625, + "learning_rate": 0.0018329477711374255, + "loss": 0.1699, + "step": 19723 + }, + { + "epoch": 0.17121379154694838, + "grad_norm": 0.1044921875, + "learning_rate": 0.0018329304988988215, + "loss": 0.1279, + "step": 19724 + }, + { + "epoch": 0.17122247202715254, + "grad_norm": 0.11962890625, + "learning_rate": 0.0018329132258586955, + "loss": 0.1289, + "step": 19725 + }, + { + "epoch": 0.1712311525073567, + "grad_norm": 0.267578125, + "learning_rate": 0.0018328959520170667, + "loss": 0.124, + "step": 19726 + }, + { + "epoch": 0.17123983298756087, + "grad_norm": 0.67578125, + "learning_rate": 0.0018328786773739538, + "loss": 0.1611, + "step": 19727 + }, + { + "epoch": 0.17124851346776504, + "grad_norm": 0.296875, + "learning_rate": 0.001832861401929376, + "loss": 0.1387, + "step": 19728 + }, + { + "epoch": 0.1712571939479692, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018328441256833519, + "loss": 0.0972, + "step": 19729 + }, + { + "epoch": 0.17126587442817337, + "grad_norm": 0.30859375, + "learning_rate": 0.0018328268486359003, + "loss": 0.1104, + "step": 19730 + }, + { + "epoch": 0.17127455490837754, + "grad_norm": 0.49609375, + "learning_rate": 0.0018328095707870402, + "loss": 0.127, + "step": 19731 + }, + { + "epoch": 0.1712832353885817, + "grad_norm": 0.70703125, + "learning_rate": 0.001832792292136791, + "loss": 0.1572, + "step": 19732 + }, + { + "epoch": 0.17129191586878587, + "grad_norm": 0.30859375, + "learning_rate": 0.0018327750126851709, + "loss": 0.1025, + "step": 19733 + }, + { + "epoch": 0.17130059634899003, + "grad_norm": 0.41796875, + "learning_rate": 0.0018327577324321988, + "loss": 0.1211, + "step": 19734 + }, + { + "epoch": 0.1713092768291942, + "grad_norm": 0.318359375, + "learning_rate": 0.001832740451377894, + "loss": 0.0933, + "step": 19735 + }, + { + "epoch": 0.17131795730939836, + "grad_norm": 0.359375, + "learning_rate": 0.0018327231695222754, + "loss": 0.1055, + "step": 19736 + }, + { + "epoch": 0.17132663778960253, + "grad_norm": 0.48828125, + "learning_rate": 0.001832705886865362, + "loss": 0.0938, + "step": 19737 + }, + { + "epoch": 0.1713353182698067, + "grad_norm": 0.2333984375, + "learning_rate": 0.001832688603407172, + "loss": 0.0762, + "step": 19738 + }, + { + "epoch": 0.17134399875001086, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018326713191477249, + "loss": 0.1006, + "step": 19739 + }, + { + "epoch": 0.17135267923021502, + "grad_norm": 0.1103515625, + "learning_rate": 0.0018326540340870394, + "loss": 0.123, + "step": 19740 + }, + { + "epoch": 0.1713613597104192, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018326367482251345, + "loss": 0.1299, + "step": 19741 + }, + { + "epoch": 0.17137004019062335, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018326194615620292, + "loss": 0.1582, + "step": 19742 + }, + { + "epoch": 0.17137872067082752, + "grad_norm": 0.08056640625, + "learning_rate": 0.0018326021740977419, + "loss": 0.1157, + "step": 19743 + }, + { + "epoch": 0.17138740115103168, + "grad_norm": 0.357421875, + "learning_rate": 0.0018325848858322922, + "loss": 0.1201, + "step": 19744 + }, + { + "epoch": 0.17139608163123585, + "grad_norm": 0.69140625, + "learning_rate": 0.0018325675967656987, + "loss": 0.0967, + "step": 19745 + }, + { + "epoch": 0.17140476211144, + "grad_norm": 0.44921875, + "learning_rate": 0.0018325503068979802, + "loss": 0.1025, + "step": 19746 + }, + { + "epoch": 0.17141344259164418, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018325330162291555, + "loss": 0.1299, + "step": 19747 + }, + { + "epoch": 0.17142212307184834, + "grad_norm": 0.66015625, + "learning_rate": 0.0018325157247592436, + "loss": 0.1099, + "step": 19748 + }, + { + "epoch": 0.1714308035520525, + "grad_norm": 0.546875, + "learning_rate": 0.0018324984324882643, + "loss": 0.1475, + "step": 19749 + }, + { + "epoch": 0.17143948403225667, + "grad_norm": 0.55078125, + "learning_rate": 0.001832481139416235, + "loss": 0.1035, + "step": 19750 + }, + { + "epoch": 0.17144816451246084, + "grad_norm": 0.10546875, + "learning_rate": 0.0018324638455431755, + "loss": 0.1094, + "step": 19751 + }, + { + "epoch": 0.171456844992665, + "grad_norm": 0.353515625, + "learning_rate": 0.0018324465508691046, + "loss": 0.1494, + "step": 19752 + }, + { + "epoch": 0.17146552547286917, + "grad_norm": 0.474609375, + "learning_rate": 0.0018324292553940412, + "loss": 0.085, + "step": 19753 + }, + { + "epoch": 0.17147420595307333, + "grad_norm": 0.34375, + "learning_rate": 0.001832411959118004, + "loss": 0.123, + "step": 19754 + }, + { + "epoch": 0.1714828864332775, + "grad_norm": 0.416015625, + "learning_rate": 0.001832394662041012, + "loss": 0.1221, + "step": 19755 + }, + { + "epoch": 0.17149156691348166, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018323773641630847, + "loss": 0.124, + "step": 19756 + }, + { + "epoch": 0.17150024739368583, + "grad_norm": 0.06591796875, + "learning_rate": 0.00183236006548424, + "loss": 0.0708, + "step": 19757 + }, + { + "epoch": 0.17150892787389, + "grad_norm": 0.89453125, + "learning_rate": 0.0018323427660044973, + "loss": 0.0952, + "step": 19758 + }, + { + "epoch": 0.17151760835409416, + "grad_norm": 1.1484375, + "learning_rate": 0.0018323254657238762, + "loss": 0.0981, + "step": 19759 + }, + { + "epoch": 0.17152628883429832, + "grad_norm": 0.2392578125, + "learning_rate": 0.0018323081646423945, + "loss": 0.0977, + "step": 19760 + }, + { + "epoch": 0.1715349693145025, + "grad_norm": 0.25390625, + "learning_rate": 0.0018322908627600714, + "loss": 0.1299, + "step": 19761 + }, + { + "epoch": 0.17154364979470665, + "grad_norm": 0.26953125, + "learning_rate": 0.0018322735600769264, + "loss": 0.1641, + "step": 19762 + }, + { + "epoch": 0.17155233027491082, + "grad_norm": 0.33203125, + "learning_rate": 0.001832256256592978, + "loss": 0.085, + "step": 19763 + }, + { + "epoch": 0.17156101075511498, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018322389523082452, + "loss": 0.1182, + "step": 19764 + }, + { + "epoch": 0.17156969123531915, + "grad_norm": 0.5859375, + "learning_rate": 0.0018322216472227467, + "loss": 0.1309, + "step": 19765 + }, + { + "epoch": 0.17157837171552331, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018322043413365014, + "loss": 0.125, + "step": 19766 + }, + { + "epoch": 0.17158705219572748, + "grad_norm": 0.27734375, + "learning_rate": 0.0018321870346495288, + "loss": 0.123, + "step": 19767 + }, + { + "epoch": 0.17159573267593164, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018321697271618475, + "loss": 0.1055, + "step": 19768 + }, + { + "epoch": 0.1716044131561358, + "grad_norm": 0.275390625, + "learning_rate": 0.0018321524188734761, + "loss": 0.1138, + "step": 19769 + }, + { + "epoch": 0.17161309363633998, + "grad_norm": 0.12353515625, + "learning_rate": 0.0018321351097844339, + "loss": 0.1602, + "step": 19770 + }, + { + "epoch": 0.17162177411654414, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018321177998947397, + "loss": 0.1045, + "step": 19771 + }, + { + "epoch": 0.1716304545967483, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018321004892044126, + "loss": 0.1416, + "step": 19772 + }, + { + "epoch": 0.17163913507695247, + "grad_norm": 0.259765625, + "learning_rate": 0.0018320831777134715, + "loss": 0.1162, + "step": 19773 + }, + { + "epoch": 0.17164781555715664, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018320658654219352, + "loss": 0.0972, + "step": 19774 + }, + { + "epoch": 0.1716564960373608, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018320485523298225, + "loss": 0.1631, + "step": 19775 + }, + { + "epoch": 0.17166517651756494, + "grad_norm": 0.375, + "learning_rate": 0.0018320312384371524, + "loss": 0.0977, + "step": 19776 + }, + { + "epoch": 0.1716738569977691, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018320139237439444, + "loss": 0.1328, + "step": 19777 + }, + { + "epoch": 0.17168253747797327, + "grad_norm": 0.33984375, + "learning_rate": 0.0018319966082502167, + "loss": 0.0811, + "step": 19778 + }, + { + "epoch": 0.17169121795817743, + "grad_norm": 0.91796875, + "learning_rate": 0.0018319792919559883, + "loss": 0.1113, + "step": 19779 + }, + { + "epoch": 0.1716998984383816, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018319619748612787, + "loss": 0.1201, + "step": 19780 + }, + { + "epoch": 0.17170857891858576, + "grad_norm": 0.310546875, + "learning_rate": 0.0018319446569661061, + "loss": 0.0889, + "step": 19781 + }, + { + "epoch": 0.17171725939878993, + "grad_norm": 0.130859375, + "learning_rate": 0.00183192733827049, + "loss": 0.0898, + "step": 19782 + }, + { + "epoch": 0.1717259398789941, + "grad_norm": 0.373046875, + "learning_rate": 0.0018319100187744496, + "loss": 0.1191, + "step": 19783 + }, + { + "epoch": 0.17173462035919826, + "grad_norm": 0.396484375, + "learning_rate": 0.001831892698478003, + "loss": 0.1001, + "step": 19784 + }, + { + "epoch": 0.17174330083940242, + "grad_norm": 0.3046875, + "learning_rate": 0.0018318753773811695, + "loss": 0.1143, + "step": 19785 + }, + { + "epoch": 0.1717519813196066, + "grad_norm": 0.345703125, + "learning_rate": 0.0018318580554839681, + "loss": 0.1143, + "step": 19786 + }, + { + "epoch": 0.17176066179981075, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018318407327864176, + "loss": 0.0996, + "step": 19787 + }, + { + "epoch": 0.17176934228001492, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018318234092885373, + "loss": 0.1104, + "step": 19788 + }, + { + "epoch": 0.17177802276021908, + "grad_norm": 0.5546875, + "learning_rate": 0.001831806084990346, + "loss": 0.1309, + "step": 19789 + }, + { + "epoch": 0.17178670324042325, + "grad_norm": 0.251953125, + "learning_rate": 0.001831788759891862, + "loss": 0.0908, + "step": 19790 + }, + { + "epoch": 0.17179538372062741, + "grad_norm": 0.67578125, + "learning_rate": 0.0018317714339931056, + "loss": 0.1172, + "step": 19791 + }, + { + "epoch": 0.17180406420083158, + "grad_norm": 1.7109375, + "learning_rate": 0.0018317541072940947, + "loss": 0.1826, + "step": 19792 + }, + { + "epoch": 0.17181274468103575, + "grad_norm": 0.77734375, + "learning_rate": 0.0018317367797948483, + "loss": 0.0996, + "step": 19793 + }, + { + "epoch": 0.1718214251612399, + "grad_norm": 0.314453125, + "learning_rate": 0.0018317194514953854, + "loss": 0.1758, + "step": 19794 + }, + { + "epoch": 0.17183010564144408, + "grad_norm": 0.546875, + "learning_rate": 0.0018317021223957252, + "loss": 0.0981, + "step": 19795 + }, + { + "epoch": 0.17183878612164824, + "grad_norm": 0.57421875, + "learning_rate": 0.0018316847924958867, + "loss": 0.1104, + "step": 19796 + }, + { + "epoch": 0.1718474666018524, + "grad_norm": 0.27734375, + "learning_rate": 0.0018316674617958888, + "loss": 0.1318, + "step": 19797 + }, + { + "epoch": 0.17185614708205657, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018316501302957503, + "loss": 0.0933, + "step": 19798 + }, + { + "epoch": 0.17186482756226074, + "grad_norm": 0.451171875, + "learning_rate": 0.0018316327979954904, + "loss": 0.0874, + "step": 19799 + }, + { + "epoch": 0.1718735080424649, + "grad_norm": 0.15625, + "learning_rate": 0.0018316154648951274, + "loss": 0.0796, + "step": 19800 + }, + { + "epoch": 0.17188218852266907, + "grad_norm": 0.3359375, + "learning_rate": 0.001831598130994681, + "loss": 0.1416, + "step": 19801 + }, + { + "epoch": 0.17189086900287323, + "grad_norm": 0.216796875, + "learning_rate": 0.0018315807962941695, + "loss": 0.0986, + "step": 19802 + }, + { + "epoch": 0.1718995494830774, + "grad_norm": 0.2578125, + "learning_rate": 0.0018315634607936127, + "loss": 0.0781, + "step": 19803 + }, + { + "epoch": 0.17190822996328156, + "grad_norm": 0.09765625, + "learning_rate": 0.0018315461244930293, + "loss": 0.0928, + "step": 19804 + }, + { + "epoch": 0.17191691044348573, + "grad_norm": 0.8515625, + "learning_rate": 0.0018315287873924373, + "loss": 0.1211, + "step": 19805 + }, + { + "epoch": 0.1719255909236899, + "grad_norm": 0.60546875, + "learning_rate": 0.0018315114494918568, + "loss": 0.1074, + "step": 19806 + }, + { + "epoch": 0.17193427140389406, + "grad_norm": 0.181640625, + "learning_rate": 0.0018314941107913064, + "loss": 0.1094, + "step": 19807 + }, + { + "epoch": 0.17194295188409822, + "grad_norm": 0.267578125, + "learning_rate": 0.0018314767712908052, + "loss": 0.0859, + "step": 19808 + }, + { + "epoch": 0.1719516323643024, + "grad_norm": 0.07958984375, + "learning_rate": 0.0018314594309903716, + "loss": 0.1133, + "step": 19809 + }, + { + "epoch": 0.17196031284450655, + "grad_norm": 0.2734375, + "learning_rate": 0.001831442089890025, + "loss": 0.1211, + "step": 19810 + }, + { + "epoch": 0.17196899332471072, + "grad_norm": 0.146484375, + "learning_rate": 0.0018314247479897846, + "loss": 0.126, + "step": 19811 + }, + { + "epoch": 0.17197767380491488, + "grad_norm": 0.365234375, + "learning_rate": 0.001831407405289669, + "loss": 0.0811, + "step": 19812 + }, + { + "epoch": 0.17198635428511905, + "grad_norm": 0.13671875, + "learning_rate": 0.001831390061789697, + "loss": 0.1089, + "step": 19813 + }, + { + "epoch": 0.1719950347653232, + "grad_norm": 0.365234375, + "learning_rate": 0.0018313727174898884, + "loss": 0.0889, + "step": 19814 + }, + { + "epoch": 0.17200371524552738, + "grad_norm": 0.271484375, + "learning_rate": 0.001831355372390261, + "loss": 0.0923, + "step": 19815 + }, + { + "epoch": 0.17201239572573154, + "grad_norm": 0.1484375, + "learning_rate": 0.0018313380264908345, + "loss": 0.1465, + "step": 19816 + }, + { + "epoch": 0.1720210762059357, + "grad_norm": 0.140625, + "learning_rate": 0.001831320679791628, + "loss": 0.1035, + "step": 19817 + }, + { + "epoch": 0.17202975668613987, + "grad_norm": 0.2421875, + "learning_rate": 0.0018313033322926598, + "loss": 0.0889, + "step": 19818 + }, + { + "epoch": 0.17203843716634404, + "grad_norm": 0.47265625, + "learning_rate": 0.0018312859839939492, + "loss": 0.1079, + "step": 19819 + }, + { + "epoch": 0.1720471176465482, + "grad_norm": 0.380859375, + "learning_rate": 0.0018312686348955153, + "loss": 0.1021, + "step": 19820 + }, + { + "epoch": 0.17205579812675237, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018312512849973776, + "loss": 0.1436, + "step": 19821 + }, + { + "epoch": 0.17206447860695653, + "grad_norm": 0.6953125, + "learning_rate": 0.001831233934299554, + "loss": 0.0737, + "step": 19822 + }, + { + "epoch": 0.1720731590871607, + "grad_norm": 0.6015625, + "learning_rate": 0.0018312165828020639, + "loss": 0.1221, + "step": 19823 + }, + { + "epoch": 0.17208183956736486, + "grad_norm": 0.169921875, + "learning_rate": 0.0018311992305049265, + "loss": 0.1016, + "step": 19824 + }, + { + "epoch": 0.17209052004756903, + "grad_norm": 0.3359375, + "learning_rate": 0.0018311818774081605, + "loss": 0.1475, + "step": 19825 + }, + { + "epoch": 0.1720992005277732, + "grad_norm": 0.171875, + "learning_rate": 0.001831164523511785, + "loss": 0.1147, + "step": 19826 + }, + { + "epoch": 0.17210788100797736, + "grad_norm": 0.3984375, + "learning_rate": 0.001831147168815819, + "loss": 0.1426, + "step": 19827 + }, + { + "epoch": 0.17211656148818152, + "grad_norm": 0.267578125, + "learning_rate": 0.0018311298133202817, + "loss": 0.1133, + "step": 19828 + }, + { + "epoch": 0.1721252419683857, + "grad_norm": 0.263671875, + "learning_rate": 0.0018311124570251913, + "loss": 0.1357, + "step": 19829 + }, + { + "epoch": 0.17213392244858985, + "grad_norm": 0.2734375, + "learning_rate": 0.0018310950999305675, + "loss": 0.1011, + "step": 19830 + }, + { + "epoch": 0.17214260292879402, + "grad_norm": 0.341796875, + "learning_rate": 0.0018310777420364292, + "loss": 0.0835, + "step": 19831 + }, + { + "epoch": 0.17215128340899818, + "grad_norm": 0.474609375, + "learning_rate": 0.0018310603833427955, + "loss": 0.1445, + "step": 19832 + }, + { + "epoch": 0.17215996388920235, + "grad_norm": 0.21875, + "learning_rate": 0.001831043023849685, + "loss": 0.1006, + "step": 19833 + }, + { + "epoch": 0.17216864436940651, + "grad_norm": 0.130859375, + "learning_rate": 0.0018310256635571166, + "loss": 0.1387, + "step": 19834 + }, + { + "epoch": 0.17217732484961068, + "grad_norm": 0.1640625, + "learning_rate": 0.0018310083024651095, + "loss": 0.1162, + "step": 19835 + }, + { + "epoch": 0.17218600532981485, + "grad_norm": 0.314453125, + "learning_rate": 0.001830990940573683, + "loss": 0.1562, + "step": 19836 + }, + { + "epoch": 0.172194685810019, + "grad_norm": 0.3671875, + "learning_rate": 0.0018309735778828553, + "loss": 0.124, + "step": 19837 + }, + { + "epoch": 0.17220336629022318, + "grad_norm": 0.5078125, + "learning_rate": 0.0018309562143926463, + "loss": 0.1187, + "step": 19838 + }, + { + "epoch": 0.17221204677042734, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018309388501030744, + "loss": 0.168, + "step": 19839 + }, + { + "epoch": 0.1722207272506315, + "grad_norm": 0.1337890625, + "learning_rate": 0.001830921485014159, + "loss": 0.1172, + "step": 19840 + }, + { + "epoch": 0.17222940773083567, + "grad_norm": 0.142578125, + "learning_rate": 0.0018309041191259185, + "loss": 0.0903, + "step": 19841 + }, + { + "epoch": 0.17223808821103984, + "grad_norm": 0.53515625, + "learning_rate": 0.0018308867524383726, + "loss": 0.1143, + "step": 19842 + }, + { + "epoch": 0.172246768691244, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018308693849515396, + "loss": 0.1172, + "step": 19843 + }, + { + "epoch": 0.17225544917144817, + "grad_norm": 0.294921875, + "learning_rate": 0.0018308520166654388, + "loss": 0.1768, + "step": 19844 + }, + { + "epoch": 0.17226412965165233, + "grad_norm": 0.07373046875, + "learning_rate": 0.0018308346475800895, + "loss": 0.0967, + "step": 19845 + }, + { + "epoch": 0.1722728101318565, + "grad_norm": 0.10009765625, + "learning_rate": 0.00183081727769551, + "loss": 0.1113, + "step": 19846 + }, + { + "epoch": 0.17228149061206066, + "grad_norm": 0.427734375, + "learning_rate": 0.0018307999070117202, + "loss": 0.1177, + "step": 19847 + }, + { + "epoch": 0.17229017109226483, + "grad_norm": 0.400390625, + "learning_rate": 0.0018307825355287383, + "loss": 0.1035, + "step": 19848 + }, + { + "epoch": 0.172298851572469, + "grad_norm": 0.298828125, + "learning_rate": 0.0018307651632465838, + "loss": 0.1064, + "step": 19849 + }, + { + "epoch": 0.17230753205267316, + "grad_norm": 0.72265625, + "learning_rate": 0.0018307477901652754, + "loss": 0.1143, + "step": 19850 + }, + { + "epoch": 0.17231621253287732, + "grad_norm": 0.625, + "learning_rate": 0.001830730416284832, + "loss": 0.1094, + "step": 19851 + }, + { + "epoch": 0.1723248930130815, + "grad_norm": 0.486328125, + "learning_rate": 0.001830713041605273, + "loss": 0.0928, + "step": 19852 + }, + { + "epoch": 0.17233357349328565, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018306956661266172, + "loss": 0.0786, + "step": 19853 + }, + { + "epoch": 0.17234225397348982, + "grad_norm": 0.9296875, + "learning_rate": 0.0018306782898488837, + "loss": 0.0957, + "step": 19854 + }, + { + "epoch": 0.17235093445369398, + "grad_norm": 0.63671875, + "learning_rate": 0.0018306609127720912, + "loss": 0.1045, + "step": 19855 + }, + { + "epoch": 0.17235961493389815, + "grad_norm": 0.435546875, + "learning_rate": 0.0018306435348962589, + "loss": 0.1128, + "step": 19856 + }, + { + "epoch": 0.1723682954141023, + "grad_norm": 0.095703125, + "learning_rate": 0.001830626156221406, + "loss": 0.103, + "step": 19857 + }, + { + "epoch": 0.17237697589430648, + "grad_norm": 0.146484375, + "learning_rate": 0.0018306087767475509, + "loss": 0.1367, + "step": 19858 + }, + { + "epoch": 0.17238565637451064, + "grad_norm": 0.244140625, + "learning_rate": 0.0018305913964747136, + "loss": 0.0884, + "step": 19859 + }, + { + "epoch": 0.1723943368547148, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018305740154029122, + "loss": 0.0859, + "step": 19860 + }, + { + "epoch": 0.17240301733491897, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018305566335321659, + "loss": 0.126, + "step": 19861 + }, + { + "epoch": 0.17241169781512314, + "grad_norm": 0.33984375, + "learning_rate": 0.0018305392508624941, + "loss": 0.1143, + "step": 19862 + }, + { + "epoch": 0.1724203782953273, + "grad_norm": 0.236328125, + "learning_rate": 0.0018305218673939155, + "loss": 0.1602, + "step": 19863 + }, + { + "epoch": 0.17242905877553147, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018305044831264492, + "loss": 0.1035, + "step": 19864 + }, + { + "epoch": 0.17243773925573563, + "grad_norm": 0.45703125, + "learning_rate": 0.0018304870980601138, + "loss": 0.1211, + "step": 19865 + }, + { + "epoch": 0.1724464197359398, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018304697121949293, + "loss": 0.1045, + "step": 19866 + }, + { + "epoch": 0.17245510021614396, + "grad_norm": 0.271484375, + "learning_rate": 0.0018304523255309136, + "loss": 0.1001, + "step": 19867 + }, + { + "epoch": 0.17246378069634813, + "grad_norm": 0.2431640625, + "learning_rate": 0.0018304349380680862, + "loss": 0.0835, + "step": 19868 + }, + { + "epoch": 0.1724724611765523, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018304175498064666, + "loss": 0.127, + "step": 19869 + }, + { + "epoch": 0.17248114165675646, + "grad_norm": 0.62109375, + "learning_rate": 0.001830400160746073, + "loss": 0.1143, + "step": 19870 + }, + { + "epoch": 0.17248982213696062, + "grad_norm": 0.103515625, + "learning_rate": 0.0018303827708869248, + "loss": 0.1387, + "step": 19871 + }, + { + "epoch": 0.1724985026171648, + "grad_norm": 0.447265625, + "learning_rate": 0.001830365380229041, + "loss": 0.0952, + "step": 19872 + }, + { + "epoch": 0.17250718309736895, + "grad_norm": 0.1123046875, + "learning_rate": 0.0018303479887724406, + "loss": 0.1196, + "step": 19873 + }, + { + "epoch": 0.17251586357757312, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018303305965171423, + "loss": 0.1035, + "step": 19874 + }, + { + "epoch": 0.17252454405777728, + "grad_norm": 0.75, + "learning_rate": 0.0018303132034631659, + "loss": 0.1006, + "step": 19875 + }, + { + "epoch": 0.17253322453798145, + "grad_norm": 0.3515625, + "learning_rate": 0.0018302958096105299, + "loss": 0.0811, + "step": 19876 + }, + { + "epoch": 0.17254190501818562, + "grad_norm": 0.76953125, + "learning_rate": 0.0018302784149592532, + "loss": 0.0869, + "step": 19877 + }, + { + "epoch": 0.17255058549838978, + "grad_norm": 0.423828125, + "learning_rate": 0.001830261019509355, + "loss": 0.1289, + "step": 19878 + }, + { + "epoch": 0.17255926597859395, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018302436232608544, + "loss": 0.1465, + "step": 19879 + }, + { + "epoch": 0.1725679464587981, + "grad_norm": 0.275390625, + "learning_rate": 0.0018302262262137703, + "loss": 0.1216, + "step": 19880 + }, + { + "epoch": 0.17257662693900228, + "grad_norm": 0.193359375, + "learning_rate": 0.001830208828368122, + "loss": 0.1426, + "step": 19881 + }, + { + "epoch": 0.17258530741920644, + "grad_norm": 0.310546875, + "learning_rate": 0.001830191429723928, + "loss": 0.124, + "step": 19882 + }, + { + "epoch": 0.1725939878994106, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018301740302812078, + "loss": 0.1196, + "step": 19883 + }, + { + "epoch": 0.17260266837961477, + "grad_norm": 0.5078125, + "learning_rate": 0.0018301566300399804, + "loss": 0.1035, + "step": 19884 + }, + { + "epoch": 0.17261134885981894, + "grad_norm": 0.6484375, + "learning_rate": 0.001830139229000264, + "loss": 0.3398, + "step": 19885 + }, + { + "epoch": 0.1726200293400231, + "grad_norm": 0.609375, + "learning_rate": 0.0018301218271620787, + "loss": 0.127, + "step": 19886 + }, + { + "epoch": 0.17262870982022727, + "grad_norm": 0.62890625, + "learning_rate": 0.0018301044245254432, + "loss": 0.1465, + "step": 19887 + }, + { + "epoch": 0.17263739030043143, + "grad_norm": 1.359375, + "learning_rate": 0.0018300870210903766, + "loss": 0.1377, + "step": 19888 + }, + { + "epoch": 0.1726460707806356, + "grad_norm": 0.443359375, + "learning_rate": 0.0018300696168568977, + "loss": 0.1348, + "step": 19889 + }, + { + "epoch": 0.17265475126083976, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018300522118250258, + "loss": 0.2197, + "step": 19890 + }, + { + "epoch": 0.17266343174104393, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018300348059947793, + "loss": 0.0825, + "step": 19891 + }, + { + "epoch": 0.1726721122212481, + "grad_norm": 0.1826171875, + "learning_rate": 0.0018300173993661781, + "loss": 0.1738, + "step": 19892 + }, + { + "epoch": 0.17268079270145226, + "grad_norm": 0.18359375, + "learning_rate": 0.0018299999919392405, + "loss": 0.1138, + "step": 19893 + }, + { + "epoch": 0.17268947318165642, + "grad_norm": 0.4140625, + "learning_rate": 0.0018299825837139864, + "loss": 0.1133, + "step": 19894 + }, + { + "epoch": 0.1726981536618606, + "grad_norm": 0.064453125, + "learning_rate": 0.001829965174690434, + "loss": 0.1118, + "step": 19895 + }, + { + "epoch": 0.17270683414206475, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018299477648686024, + "loss": 0.1602, + "step": 19896 + }, + { + "epoch": 0.17271551462226892, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018299303542485112, + "loss": 0.123, + "step": 19897 + }, + { + "epoch": 0.17272419510247308, + "grad_norm": 0.8671875, + "learning_rate": 0.0018299129428301793, + "loss": 0.1191, + "step": 19898 + }, + { + "epoch": 0.17273287558267722, + "grad_norm": 0.546875, + "learning_rate": 0.0018298955306136253, + "loss": 0.1318, + "step": 19899 + }, + { + "epoch": 0.17274155606288139, + "grad_norm": 0.3515625, + "learning_rate": 0.0018298781175988686, + "loss": 0.1113, + "step": 19900 + }, + { + "epoch": 0.17275023654308555, + "grad_norm": 0.310546875, + "learning_rate": 0.001829860703785928, + "loss": 0.0791, + "step": 19901 + }, + { + "epoch": 0.17275891702328972, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018298432891748228, + "loss": 0.1025, + "step": 19902 + }, + { + "epoch": 0.17276759750349388, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018298258737655723, + "loss": 0.0996, + "step": 19903 + }, + { + "epoch": 0.17277627798369805, + "grad_norm": 0.27734375, + "learning_rate": 0.0018298084575581947, + "loss": 0.1367, + "step": 19904 + }, + { + "epoch": 0.1727849584639022, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018297910405527095, + "loss": 0.0903, + "step": 19905 + }, + { + "epoch": 0.17279363894410638, + "grad_norm": 0.166015625, + "learning_rate": 0.0018297736227491361, + "loss": 0.127, + "step": 19906 + }, + { + "epoch": 0.17280231942431054, + "grad_norm": 0.59375, + "learning_rate": 0.001829756204147493, + "loss": 0.1279, + "step": 19907 + }, + { + "epoch": 0.1728109999045147, + "grad_norm": 1.0859375, + "learning_rate": 0.0018297387847477998, + "loss": 0.168, + "step": 19908 + }, + { + "epoch": 0.17281968038471887, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018297213645500749, + "loss": 0.1113, + "step": 19909 + }, + { + "epoch": 0.17282836086492304, + "grad_norm": 0.25390625, + "learning_rate": 0.001829703943554338, + "loss": 0.103, + "step": 19910 + }, + { + "epoch": 0.1728370413451272, + "grad_norm": 0.296875, + "learning_rate": 0.0018296865217606077, + "loss": 0.1074, + "step": 19911 + }, + { + "epoch": 0.17284572182533137, + "grad_norm": 0.115234375, + "learning_rate": 0.0018296690991689028, + "loss": 0.1309, + "step": 19912 + }, + { + "epoch": 0.17285440230553553, + "grad_norm": 1.4609375, + "learning_rate": 0.0018296516757792434, + "loss": 0.1094, + "step": 19913 + }, + { + "epoch": 0.1728630827857397, + "grad_norm": 0.16015625, + "learning_rate": 0.0018296342515916474, + "loss": 0.1035, + "step": 19914 + }, + { + "epoch": 0.17287176326594386, + "grad_norm": 0.255859375, + "learning_rate": 0.0018296168266061343, + "loss": 0.1191, + "step": 19915 + }, + { + "epoch": 0.17288044374614803, + "grad_norm": 0.06298828125, + "learning_rate": 0.0018295994008227234, + "loss": 0.0884, + "step": 19916 + }, + { + "epoch": 0.1728891242263522, + "grad_norm": 0.3203125, + "learning_rate": 0.0018295819742414339, + "loss": 0.1162, + "step": 19917 + }, + { + "epoch": 0.17289780470655636, + "grad_norm": 0.330078125, + "learning_rate": 0.001829564546862284, + "loss": 0.1406, + "step": 19918 + }, + { + "epoch": 0.17290648518676052, + "grad_norm": 0.162109375, + "learning_rate": 0.0018295471186852935, + "loss": 0.0928, + "step": 19919 + }, + { + "epoch": 0.1729151656669647, + "grad_norm": 0.65234375, + "learning_rate": 0.001829529689710481, + "loss": 0.1099, + "step": 19920 + }, + { + "epoch": 0.17292384614716885, + "grad_norm": 0.173828125, + "learning_rate": 0.0018295122599378659, + "loss": 0.1113, + "step": 19921 + }, + { + "epoch": 0.17293252662737302, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018294948293674674, + "loss": 0.1465, + "step": 19922 + }, + { + "epoch": 0.17294120710757718, + "grad_norm": 0.5859375, + "learning_rate": 0.001829477397999304, + "loss": 0.1641, + "step": 19923 + }, + { + "epoch": 0.17294988758778135, + "grad_norm": 0.5078125, + "learning_rate": 0.0018294599658333953, + "loss": 0.0986, + "step": 19924 + }, + { + "epoch": 0.1729585680679855, + "grad_norm": 0.07861328125, + "learning_rate": 0.0018294425328697602, + "loss": 0.1079, + "step": 19925 + }, + { + "epoch": 0.17296724854818968, + "grad_norm": 0.28515625, + "learning_rate": 0.0018294250991084176, + "loss": 0.1221, + "step": 19926 + }, + { + "epoch": 0.17297592902839384, + "grad_norm": 0.90625, + "learning_rate": 0.0018294076645493863, + "loss": 0.1465, + "step": 19927 + }, + { + "epoch": 0.172984609508598, + "grad_norm": 0.287109375, + "learning_rate": 0.0018293902291926863, + "loss": 0.126, + "step": 19928 + }, + { + "epoch": 0.17299328998880217, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018293727930383359, + "loss": 0.1079, + "step": 19929 + }, + { + "epoch": 0.17300197046900634, + "grad_norm": 0.12890625, + "learning_rate": 0.0018293553560863544, + "loss": 0.1045, + "step": 19930 + }, + { + "epoch": 0.1730106509492105, + "grad_norm": 0.318359375, + "learning_rate": 0.0018293379183367605, + "loss": 0.1001, + "step": 19931 + }, + { + "epoch": 0.17301933142941467, + "grad_norm": 0.51171875, + "learning_rate": 0.001829320479789574, + "loss": 0.1084, + "step": 19932 + }, + { + "epoch": 0.17302801190961883, + "grad_norm": 0.431640625, + "learning_rate": 0.0018293030404448137, + "loss": 0.1299, + "step": 19933 + }, + { + "epoch": 0.173036692389823, + "grad_norm": 0.5625, + "learning_rate": 0.001829285600302498, + "loss": 0.1201, + "step": 19934 + }, + { + "epoch": 0.17304537287002716, + "grad_norm": 0.2734375, + "learning_rate": 0.001829268159362647, + "loss": 0.1289, + "step": 19935 + }, + { + "epoch": 0.17305405335023133, + "grad_norm": 0.11962890625, + "learning_rate": 0.001829250717625279, + "loss": 0.1475, + "step": 19936 + }, + { + "epoch": 0.1730627338304355, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018292332750904138, + "loss": 0.1108, + "step": 19937 + }, + { + "epoch": 0.17307141431063966, + "grad_norm": 0.58984375, + "learning_rate": 0.0018292158317580697, + "loss": 0.127, + "step": 19938 + }, + { + "epoch": 0.17308009479084382, + "grad_norm": 0.224609375, + "learning_rate": 0.001829198387628266, + "loss": 0.1055, + "step": 19939 + }, + { + "epoch": 0.173088775271048, + "grad_norm": 1.125, + "learning_rate": 0.0018291809427010222, + "loss": 0.1143, + "step": 19940 + }, + { + "epoch": 0.17309745575125215, + "grad_norm": 0.10205078125, + "learning_rate": 0.001829163496976357, + "loss": 0.0869, + "step": 19941 + }, + { + "epoch": 0.17310613623145632, + "grad_norm": 0.28125, + "learning_rate": 0.0018291460504542894, + "loss": 0.0757, + "step": 19942 + }, + { + "epoch": 0.17311481671166049, + "grad_norm": 0.6328125, + "learning_rate": 0.0018291286031348391, + "loss": 0.126, + "step": 19943 + }, + { + "epoch": 0.17312349719186465, + "grad_norm": 0.1376953125, + "learning_rate": 0.001829111155018024, + "loss": 0.1011, + "step": 19944 + }, + { + "epoch": 0.17313217767206882, + "grad_norm": 0.310546875, + "learning_rate": 0.0018290937061038643, + "loss": 0.0952, + "step": 19945 + }, + { + "epoch": 0.17314085815227298, + "grad_norm": 0.208984375, + "learning_rate": 0.0018290762563923787, + "loss": 0.1045, + "step": 19946 + }, + { + "epoch": 0.17314953863247715, + "grad_norm": 0.375, + "learning_rate": 0.0018290588058835864, + "loss": 0.1094, + "step": 19947 + }, + { + "epoch": 0.1731582191126813, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018290413545775062, + "loss": 0.1045, + "step": 19948 + }, + { + "epoch": 0.17316689959288548, + "grad_norm": 0.171875, + "learning_rate": 0.0018290239024741572, + "loss": 0.1035, + "step": 19949 + }, + { + "epoch": 0.17317558007308964, + "grad_norm": 0.11279296875, + "learning_rate": 0.0018290064495735585, + "loss": 0.1289, + "step": 19950 + }, + { + "epoch": 0.1731842605532938, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018289889958757295, + "loss": 0.1465, + "step": 19951 + }, + { + "epoch": 0.17319294103349797, + "grad_norm": 0.345703125, + "learning_rate": 0.0018289715413806892, + "loss": 0.106, + "step": 19952 + }, + { + "epoch": 0.17320162151370214, + "grad_norm": 0.4375, + "learning_rate": 0.0018289540860884565, + "loss": 0.1211, + "step": 19953 + }, + { + "epoch": 0.1732103019939063, + "grad_norm": 0.7265625, + "learning_rate": 0.0018289366299990505, + "loss": 0.1475, + "step": 19954 + }, + { + "epoch": 0.17321898247411047, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018289191731124903, + "loss": 0.1738, + "step": 19955 + }, + { + "epoch": 0.17322766295431463, + "grad_norm": 0.48046875, + "learning_rate": 0.0018289017154287952, + "loss": 0.1221, + "step": 19956 + }, + { + "epoch": 0.1732363434345188, + "grad_norm": 0.1689453125, + "learning_rate": 0.001828884256947984, + "loss": 0.1045, + "step": 19957 + }, + { + "epoch": 0.17324502391472296, + "grad_norm": 0.287109375, + "learning_rate": 0.001828866797670076, + "loss": 0.1592, + "step": 19958 + }, + { + "epoch": 0.17325370439492713, + "grad_norm": 0.875, + "learning_rate": 0.0018288493375950901, + "loss": 0.1475, + "step": 19959 + }, + { + "epoch": 0.1732623848751313, + "grad_norm": 0.08984375, + "learning_rate": 0.0018288318767230457, + "loss": 0.1187, + "step": 19960 + }, + { + "epoch": 0.17327106535533546, + "grad_norm": 0.16796875, + "learning_rate": 0.0018288144150539617, + "loss": 0.0874, + "step": 19961 + }, + { + "epoch": 0.17327974583553962, + "grad_norm": 0.56640625, + "learning_rate": 0.0018287969525878568, + "loss": 0.1099, + "step": 19962 + }, + { + "epoch": 0.1732884263157438, + "grad_norm": 0.50390625, + "learning_rate": 0.001828779489324751, + "loss": 0.0977, + "step": 19963 + }, + { + "epoch": 0.17329710679594795, + "grad_norm": 0.62890625, + "learning_rate": 0.0018287620252646628, + "loss": 0.1211, + "step": 19964 + }, + { + "epoch": 0.17330578727615212, + "grad_norm": 0.1083984375, + "learning_rate": 0.0018287445604076114, + "loss": 0.0894, + "step": 19965 + }, + { + "epoch": 0.17331446775635628, + "grad_norm": 0.16796875, + "learning_rate": 0.0018287270947536158, + "loss": 0.1055, + "step": 19966 + }, + { + "epoch": 0.17332314823656045, + "grad_norm": 0.1181640625, + "learning_rate": 0.0018287096283026952, + "loss": 0.0938, + "step": 19967 + }, + { + "epoch": 0.1733318287167646, + "grad_norm": 0.216796875, + "learning_rate": 0.0018286921610548687, + "loss": 0.1543, + "step": 19968 + }, + { + "epoch": 0.17334050919696878, + "grad_norm": 0.40625, + "learning_rate": 0.0018286746930101557, + "loss": 0.1025, + "step": 19969 + }, + { + "epoch": 0.17334918967717294, + "grad_norm": 0.23828125, + "learning_rate": 0.0018286572241685746, + "loss": 0.1162, + "step": 19970 + }, + { + "epoch": 0.1733578701573771, + "grad_norm": 0.26171875, + "learning_rate": 0.001828639754530145, + "loss": 0.1055, + "step": 19971 + }, + { + "epoch": 0.17336655063758127, + "grad_norm": 0.3515625, + "learning_rate": 0.001828622284094886, + "loss": 0.125, + "step": 19972 + }, + { + "epoch": 0.17337523111778544, + "grad_norm": 0.212890625, + "learning_rate": 0.0018286048128628165, + "loss": 0.1104, + "step": 19973 + }, + { + "epoch": 0.1733839115979896, + "grad_norm": 0.169921875, + "learning_rate": 0.0018285873408339557, + "loss": 0.1162, + "step": 19974 + }, + { + "epoch": 0.17339259207819377, + "grad_norm": 0.82421875, + "learning_rate": 0.001828569868008323, + "loss": 0.1104, + "step": 19975 + }, + { + "epoch": 0.17340127255839793, + "grad_norm": 0.259765625, + "learning_rate": 0.001828552394385937, + "loss": 0.0908, + "step": 19976 + }, + { + "epoch": 0.1734099530386021, + "grad_norm": 1.2578125, + "learning_rate": 0.001828534919966817, + "loss": 0.2461, + "step": 19977 + }, + { + "epoch": 0.17341863351880626, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018285174447509823, + "loss": 0.1172, + "step": 19978 + }, + { + "epoch": 0.17342731399901043, + "grad_norm": 0.0703125, + "learning_rate": 0.0018284999687384518, + "loss": 0.0898, + "step": 19979 + }, + { + "epoch": 0.1734359944792146, + "grad_norm": 0.435546875, + "learning_rate": 0.001828482491929245, + "loss": 0.1162, + "step": 19980 + }, + { + "epoch": 0.17344467495941876, + "grad_norm": 0.271484375, + "learning_rate": 0.0018284650143233804, + "loss": 0.1289, + "step": 19981 + }, + { + "epoch": 0.17345335543962292, + "grad_norm": 0.4609375, + "learning_rate": 0.0018284475359208775, + "loss": 0.127, + "step": 19982 + }, + { + "epoch": 0.1734620359198271, + "grad_norm": 0.341796875, + "learning_rate": 0.0018284300567217551, + "loss": 0.1318, + "step": 19983 + }, + { + "epoch": 0.17347071640003126, + "grad_norm": 0.47265625, + "learning_rate": 0.0018284125767260323, + "loss": 0.1025, + "step": 19984 + }, + { + "epoch": 0.17347939688023542, + "grad_norm": 0.423828125, + "learning_rate": 0.001828395095933729, + "loss": 0.126, + "step": 19985 + }, + { + "epoch": 0.17348807736043959, + "grad_norm": 0.25, + "learning_rate": 0.0018283776143448638, + "loss": 0.0771, + "step": 19986 + }, + { + "epoch": 0.17349675784064375, + "grad_norm": 0.546875, + "learning_rate": 0.0018283601319594553, + "loss": 0.1152, + "step": 19987 + }, + { + "epoch": 0.17350543832084792, + "grad_norm": 0.232421875, + "learning_rate": 0.0018283426487775234, + "loss": 0.166, + "step": 19988 + }, + { + "epoch": 0.17351411880105208, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018283251647990874, + "loss": 0.1133, + "step": 19989 + }, + { + "epoch": 0.17352279928125625, + "grad_norm": 0.5546875, + "learning_rate": 0.0018283076800241652, + "loss": 0.1133, + "step": 19990 + }, + { + "epoch": 0.1735314797614604, + "grad_norm": 0.59765625, + "learning_rate": 0.001828290194452777, + "loss": 0.1406, + "step": 19991 + }, + { + "epoch": 0.17354016024166458, + "grad_norm": 0.244140625, + "learning_rate": 0.0018282727080849413, + "loss": 0.1172, + "step": 19992 + }, + { + "epoch": 0.17354884072186874, + "grad_norm": 0.72265625, + "learning_rate": 0.0018282552209206778, + "loss": 0.1309, + "step": 19993 + }, + { + "epoch": 0.1735575212020729, + "grad_norm": 0.78125, + "learning_rate": 0.001828237732960005, + "loss": 0.1621, + "step": 19994 + }, + { + "epoch": 0.17356620168227707, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018282202442029428, + "loss": 0.103, + "step": 19995 + }, + { + "epoch": 0.17357488216248124, + "grad_norm": 0.455078125, + "learning_rate": 0.0018282027546495097, + "loss": 0.126, + "step": 19996 + }, + { + "epoch": 0.1735835626426854, + "grad_norm": 0.482421875, + "learning_rate": 0.0018281852642997249, + "loss": 0.0889, + "step": 19997 + }, + { + "epoch": 0.17359224312288957, + "grad_norm": 0.4375, + "learning_rate": 0.0018281677731536077, + "loss": 0.1035, + "step": 19998 + }, + { + "epoch": 0.17360092360309373, + "grad_norm": 0.1015625, + "learning_rate": 0.0018281502812111771, + "loss": 0.0776, + "step": 19999 + }, + { + "epoch": 0.1736096040832979, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018281327884724526, + "loss": 0.0977, + "step": 20000 + }, + { + "epoch": 0.17361828456350206, + "grad_norm": 0.208984375, + "learning_rate": 0.0018281152949374527, + "loss": 0.1182, + "step": 20001 + }, + { + "epoch": 0.17362696504370623, + "grad_norm": 0.35546875, + "learning_rate": 0.001828097800606197, + "loss": 0.1104, + "step": 20002 + }, + { + "epoch": 0.1736356455239104, + "grad_norm": 0.365234375, + "learning_rate": 0.0018280803054787043, + "loss": 0.1475, + "step": 20003 + }, + { + "epoch": 0.17364432600411456, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018280628095549943, + "loss": 0.083, + "step": 20004 + }, + { + "epoch": 0.17365300648431872, + "grad_norm": 1.1953125, + "learning_rate": 0.0018280453128350855, + "loss": 0.125, + "step": 20005 + }, + { + "epoch": 0.1736616869645229, + "grad_norm": 0.29296875, + "learning_rate": 0.0018280278153189973, + "loss": 0.1084, + "step": 20006 + }, + { + "epoch": 0.17367036744472705, + "grad_norm": 0.3828125, + "learning_rate": 0.0018280103170067487, + "loss": 0.1055, + "step": 20007 + }, + { + "epoch": 0.17367904792493122, + "grad_norm": 0.5234375, + "learning_rate": 0.001827992817898359, + "loss": 0.123, + "step": 20008 + }, + { + "epoch": 0.17368772840513538, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018279753179938473, + "loss": 0.1089, + "step": 20009 + }, + { + "epoch": 0.17369640888533955, + "grad_norm": 0.40625, + "learning_rate": 0.001827957817293233, + "loss": 0.1221, + "step": 20010 + }, + { + "epoch": 0.1737050893655437, + "grad_norm": 0.427734375, + "learning_rate": 0.0018279403157965347, + "loss": 0.1328, + "step": 20011 + }, + { + "epoch": 0.17371376984574788, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018279228135037718, + "loss": 0.0859, + "step": 20012 + }, + { + "epoch": 0.17372245032595204, + "grad_norm": 0.306640625, + "learning_rate": 0.0018279053104149634, + "loss": 0.1118, + "step": 20013 + }, + { + "epoch": 0.1737311308061562, + "grad_norm": 0.435546875, + "learning_rate": 0.0018278878065301288, + "loss": 0.0938, + "step": 20014 + }, + { + "epoch": 0.17373981128636037, + "grad_norm": 0.5, + "learning_rate": 0.001827870301849287, + "loss": 0.1367, + "step": 20015 + }, + { + "epoch": 0.17374849176656454, + "grad_norm": 0.58203125, + "learning_rate": 0.0018278527963724574, + "loss": 0.1504, + "step": 20016 + }, + { + "epoch": 0.1737571722467687, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018278352900996586, + "loss": 0.1113, + "step": 20017 + }, + { + "epoch": 0.17376585272697287, + "grad_norm": 0.61328125, + "learning_rate": 0.0018278177830309103, + "loss": 0.1484, + "step": 20018 + }, + { + "epoch": 0.17377453320717703, + "grad_norm": 1.015625, + "learning_rate": 0.0018278002751662312, + "loss": 0.1357, + "step": 20019 + }, + { + "epoch": 0.1737832136873812, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018277827665056409, + "loss": 0.0962, + "step": 20020 + }, + { + "epoch": 0.17379189416758536, + "grad_norm": 0.81640625, + "learning_rate": 0.0018277652570491579, + "loss": 0.1309, + "step": 20021 + }, + { + "epoch": 0.1738005746477895, + "grad_norm": 0.166015625, + "learning_rate": 0.0018277477467968021, + "loss": 0.1172, + "step": 20022 + }, + { + "epoch": 0.17380925512799367, + "grad_norm": 0.60546875, + "learning_rate": 0.0018277302357485924, + "loss": 0.1187, + "step": 20023 + }, + { + "epoch": 0.17381793560819783, + "grad_norm": 0.51953125, + "learning_rate": 0.0018277127239045474, + "loss": 0.1338, + "step": 20024 + }, + { + "epoch": 0.173826616088402, + "grad_norm": 0.197265625, + "learning_rate": 0.001827695211264687, + "loss": 0.1084, + "step": 20025 + }, + { + "epoch": 0.17383529656860616, + "grad_norm": 0.1005859375, + "learning_rate": 0.00182767769782903, + "loss": 0.0957, + "step": 20026 + }, + { + "epoch": 0.17384397704881033, + "grad_norm": 0.22265625, + "learning_rate": 0.0018276601835975957, + "loss": 0.1133, + "step": 20027 + }, + { + "epoch": 0.1738526575290145, + "grad_norm": 0.294921875, + "learning_rate": 0.001827642668570403, + "loss": 0.0991, + "step": 20028 + }, + { + "epoch": 0.17386133800921866, + "grad_norm": 0.26171875, + "learning_rate": 0.0018276251527474714, + "loss": 0.0981, + "step": 20029 + }, + { + "epoch": 0.17387001848942282, + "grad_norm": 0.2197265625, + "learning_rate": 0.00182760763612882, + "loss": 0.1138, + "step": 20030 + }, + { + "epoch": 0.173878698969627, + "grad_norm": 0.1357421875, + "learning_rate": 0.001827590118714467, + "loss": 0.1152, + "step": 20031 + }, + { + "epoch": 0.17388737944983115, + "grad_norm": 0.058349609375, + "learning_rate": 0.001827572600504433, + "loss": 0.082, + "step": 20032 + }, + { + "epoch": 0.17389605993003532, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018275550814987366, + "loss": 0.1147, + "step": 20033 + }, + { + "epoch": 0.17390474041023948, + "grad_norm": 0.55078125, + "learning_rate": 0.0018275375616973967, + "loss": 0.1309, + "step": 20034 + }, + { + "epoch": 0.17391342089044365, + "grad_norm": 0.498046875, + "learning_rate": 0.0018275200411004327, + "loss": 0.1045, + "step": 20035 + }, + { + "epoch": 0.1739221013706478, + "grad_norm": 0.1171875, + "learning_rate": 0.001827502519707864, + "loss": 0.1025, + "step": 20036 + }, + { + "epoch": 0.17393078185085198, + "grad_norm": 0.29296875, + "learning_rate": 0.001827484997519709, + "loss": 0.126, + "step": 20037 + }, + { + "epoch": 0.17393946233105614, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018274674745359877, + "loss": 0.0908, + "step": 20038 + }, + { + "epoch": 0.1739481428112603, + "grad_norm": 0.625, + "learning_rate": 0.0018274499507567185, + "loss": 0.0879, + "step": 20039 + }, + { + "epoch": 0.17395682329146447, + "grad_norm": 0.8984375, + "learning_rate": 0.001827432426181921, + "loss": 0.1572, + "step": 20040 + }, + { + "epoch": 0.17396550377166864, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018274149008116148, + "loss": 0.0967, + "step": 20041 + }, + { + "epoch": 0.1739741842518728, + "grad_norm": 0.6015625, + "learning_rate": 0.0018273973746458184, + "loss": 0.1367, + "step": 20042 + }, + { + "epoch": 0.17398286473207697, + "grad_norm": 0.404296875, + "learning_rate": 0.0018273798476845509, + "loss": 0.0908, + "step": 20043 + }, + { + "epoch": 0.17399154521228113, + "grad_norm": 0.1767578125, + "learning_rate": 0.001827362319927832, + "loss": 0.1157, + "step": 20044 + }, + { + "epoch": 0.1740002256924853, + "grad_norm": 0.2177734375, + "learning_rate": 0.0018273447913756808, + "loss": 0.0947, + "step": 20045 + }, + { + "epoch": 0.17400890617268946, + "grad_norm": 0.482421875, + "learning_rate": 0.0018273272620281157, + "loss": 0.1543, + "step": 20046 + }, + { + "epoch": 0.17401758665289363, + "grad_norm": 0.25, + "learning_rate": 0.0018273097318851567, + "loss": 0.1123, + "step": 20047 + }, + { + "epoch": 0.1740262671330978, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018272922009468229, + "loss": 0.1147, + "step": 20048 + }, + { + "epoch": 0.17403494761330196, + "grad_norm": 0.1484375, + "learning_rate": 0.0018272746692131329, + "loss": 0.0996, + "step": 20049 + }, + { + "epoch": 0.17404362809350613, + "grad_norm": 0.146484375, + "learning_rate": 0.0018272571366841064, + "loss": 0.0938, + "step": 20050 + }, + { + "epoch": 0.1740523085737103, + "grad_norm": 0.333984375, + "learning_rate": 0.0018272396033597626, + "loss": 0.0957, + "step": 20051 + }, + { + "epoch": 0.17406098905391446, + "grad_norm": 0.9375, + "learning_rate": 0.0018272220692401202, + "loss": 0.1553, + "step": 20052 + }, + { + "epoch": 0.17406966953411862, + "grad_norm": 0.59765625, + "learning_rate": 0.001827204534325199, + "loss": 0.1201, + "step": 20053 + }, + { + "epoch": 0.17407835001432279, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018271869986150177, + "loss": 0.126, + "step": 20054 + }, + { + "epoch": 0.17408703049452695, + "grad_norm": 0.28515625, + "learning_rate": 0.0018271694621095956, + "loss": 0.0977, + "step": 20055 + }, + { + "epoch": 0.17409571097473112, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018271519248089518, + "loss": 0.1108, + "step": 20056 + }, + { + "epoch": 0.17410439145493528, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018271343867131058, + "loss": 0.1377, + "step": 20057 + }, + { + "epoch": 0.17411307193513945, + "grad_norm": 0.52734375, + "learning_rate": 0.0018271168478220764, + "loss": 0.0879, + "step": 20058 + }, + { + "epoch": 0.1741217524153436, + "grad_norm": 0.5625, + "learning_rate": 0.001827099308135883, + "loss": 0.1318, + "step": 20059 + }, + { + "epoch": 0.17413043289554778, + "grad_norm": 0.50390625, + "learning_rate": 0.0018270817676545447, + "loss": 0.1055, + "step": 20060 + }, + { + "epoch": 0.17413911337575194, + "grad_norm": 0.11962890625, + "learning_rate": 0.0018270642263780805, + "loss": 0.1348, + "step": 20061 + }, + { + "epoch": 0.1741477938559561, + "grad_norm": 0.482421875, + "learning_rate": 0.00182704668430651, + "loss": 0.1162, + "step": 20062 + }, + { + "epoch": 0.17415647433616027, + "grad_norm": 0.392578125, + "learning_rate": 0.001827029141439852, + "loss": 0.0898, + "step": 20063 + }, + { + "epoch": 0.17416515481636444, + "grad_norm": 0.2294921875, + "learning_rate": 0.0018270115977781266, + "loss": 0.1143, + "step": 20064 + }, + { + "epoch": 0.1741738352965686, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018269940533213514, + "loss": 0.1445, + "step": 20065 + }, + { + "epoch": 0.17418251577677277, + "grad_norm": 0.7265625, + "learning_rate": 0.0018269765080695466, + "loss": 0.168, + "step": 20066 + }, + { + "epoch": 0.17419119625697693, + "grad_norm": 0.08544921875, + "learning_rate": 0.0018269589620227313, + "loss": 0.1045, + "step": 20067 + }, + { + "epoch": 0.1741998767371811, + "grad_norm": 0.1953125, + "learning_rate": 0.0018269414151809247, + "loss": 0.1279, + "step": 20068 + }, + { + "epoch": 0.17420855721738526, + "grad_norm": 0.2734375, + "learning_rate": 0.0018269238675441457, + "loss": 0.1641, + "step": 20069 + }, + { + "epoch": 0.17421723769758943, + "grad_norm": 0.71484375, + "learning_rate": 0.0018269063191124135, + "loss": 0.0962, + "step": 20070 + }, + { + "epoch": 0.1742259181777936, + "grad_norm": 0.302734375, + "learning_rate": 0.0018268887698857479, + "loss": 0.1055, + "step": 20071 + }, + { + "epoch": 0.17423459865799776, + "grad_norm": 0.5703125, + "learning_rate": 0.0018268712198641678, + "loss": 0.124, + "step": 20072 + }, + { + "epoch": 0.17424327913820192, + "grad_norm": 0.97265625, + "learning_rate": 0.0018268536690476918, + "loss": 0.1387, + "step": 20073 + }, + { + "epoch": 0.1742519596184061, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018268361174363396, + "loss": 0.125, + "step": 20074 + }, + { + "epoch": 0.17426064009861025, + "grad_norm": 0.64453125, + "learning_rate": 0.0018268185650301306, + "loss": 0.1123, + "step": 20075 + }, + { + "epoch": 0.17426932057881442, + "grad_norm": 0.158203125, + "learning_rate": 0.0018268010118290836, + "loss": 0.1172, + "step": 20076 + }, + { + "epoch": 0.17427800105901858, + "grad_norm": 0.1748046875, + "learning_rate": 0.001826783457833218, + "loss": 0.0708, + "step": 20077 + }, + { + "epoch": 0.17428668153922275, + "grad_norm": 1.2578125, + "learning_rate": 0.0018267659030425528, + "loss": 0.1309, + "step": 20078 + }, + { + "epoch": 0.1742953620194269, + "grad_norm": 0.625, + "learning_rate": 0.0018267483474571071, + "loss": 0.1465, + "step": 20079 + }, + { + "epoch": 0.17430404249963108, + "grad_norm": 0.423828125, + "learning_rate": 0.001826730791076901, + "loss": 0.1475, + "step": 20080 + }, + { + "epoch": 0.17431272297983524, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018267132339019525, + "loss": 0.1079, + "step": 20081 + }, + { + "epoch": 0.1743214034600394, + "grad_norm": 0.3203125, + "learning_rate": 0.0018266956759322814, + "loss": 0.1035, + "step": 20082 + }, + { + "epoch": 0.17433008394024357, + "grad_norm": 0.138671875, + "learning_rate": 0.001826678117167907, + "loss": 0.0986, + "step": 20083 + }, + { + "epoch": 0.17433876442044774, + "grad_norm": 0.6484375, + "learning_rate": 0.0018266605576088481, + "loss": 0.105, + "step": 20084 + }, + { + "epoch": 0.1743474449006519, + "grad_norm": 0.08251953125, + "learning_rate": 0.0018266429972551245, + "loss": 0.0933, + "step": 20085 + }, + { + "epoch": 0.17435612538085607, + "grad_norm": 0.328125, + "learning_rate": 0.0018266254361067546, + "loss": 0.1611, + "step": 20086 + }, + { + "epoch": 0.17436480586106023, + "grad_norm": 0.341796875, + "learning_rate": 0.0018266078741637581, + "loss": 0.106, + "step": 20087 + }, + { + "epoch": 0.1743734863412644, + "grad_norm": 0.54296875, + "learning_rate": 0.0018265903114261543, + "loss": 0.1191, + "step": 20088 + }, + { + "epoch": 0.17438216682146856, + "grad_norm": 0.71484375, + "learning_rate": 0.0018265727478939623, + "loss": 0.1318, + "step": 20089 + }, + { + "epoch": 0.17439084730167273, + "grad_norm": 0.4609375, + "learning_rate": 0.0018265551835672013, + "loss": 0.1582, + "step": 20090 + }, + { + "epoch": 0.1743995277818769, + "grad_norm": 0.154296875, + "learning_rate": 0.0018265376184458902, + "loss": 0.105, + "step": 20091 + }, + { + "epoch": 0.17440820826208106, + "grad_norm": 0.5078125, + "learning_rate": 0.0018265200525300486, + "loss": 0.1279, + "step": 20092 + }, + { + "epoch": 0.17441688874228523, + "grad_norm": 0.376953125, + "learning_rate": 0.0018265024858196955, + "loss": 0.1196, + "step": 20093 + }, + { + "epoch": 0.1744255692224894, + "grad_norm": 0.302734375, + "learning_rate": 0.0018264849183148506, + "loss": 0.126, + "step": 20094 + }, + { + "epoch": 0.17443424970269356, + "grad_norm": 0.25, + "learning_rate": 0.0018264673500155322, + "loss": 0.1348, + "step": 20095 + }, + { + "epoch": 0.17444293018289772, + "grad_norm": 0.4765625, + "learning_rate": 0.0018264497809217602, + "loss": 0.1201, + "step": 20096 + }, + { + "epoch": 0.17445161066310189, + "grad_norm": 0.53515625, + "learning_rate": 0.001826432211033554, + "loss": 0.1465, + "step": 20097 + }, + { + "epoch": 0.17446029114330605, + "grad_norm": 0.431640625, + "learning_rate": 0.001826414640350932, + "loss": 0.1045, + "step": 20098 + }, + { + "epoch": 0.17446897162351022, + "grad_norm": 0.625, + "learning_rate": 0.001826397068873914, + "loss": 0.1201, + "step": 20099 + }, + { + "epoch": 0.17447765210371438, + "grad_norm": 0.3828125, + "learning_rate": 0.0018263794966025188, + "loss": 0.0986, + "step": 20100 + }, + { + "epoch": 0.17448633258391855, + "grad_norm": 0.53515625, + "learning_rate": 0.0018263619235367662, + "loss": 0.0918, + "step": 20101 + }, + { + "epoch": 0.1744950130641227, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018263443496766751, + "loss": 0.1025, + "step": 20102 + }, + { + "epoch": 0.17450369354432688, + "grad_norm": 0.46484375, + "learning_rate": 0.0018263267750222648, + "loss": 0.0967, + "step": 20103 + }, + { + "epoch": 0.17451237402453104, + "grad_norm": 0.79296875, + "learning_rate": 0.0018263091995735544, + "loss": 0.1875, + "step": 20104 + }, + { + "epoch": 0.1745210545047352, + "grad_norm": 0.2333984375, + "learning_rate": 0.001826291623330563, + "loss": 0.1016, + "step": 20105 + }, + { + "epoch": 0.17452973498493937, + "grad_norm": 0.43359375, + "learning_rate": 0.0018262740462933102, + "loss": 0.1221, + "step": 20106 + }, + { + "epoch": 0.17453841546514354, + "grad_norm": 0.173828125, + "learning_rate": 0.001826256468461815, + "loss": 0.0981, + "step": 20107 + }, + { + "epoch": 0.1745470959453477, + "grad_norm": 0.390625, + "learning_rate": 0.0018262388898360965, + "loss": 0.123, + "step": 20108 + }, + { + "epoch": 0.17455577642555187, + "grad_norm": 1.078125, + "learning_rate": 0.0018262213104161742, + "loss": 0.1123, + "step": 20109 + }, + { + "epoch": 0.17456445690575603, + "grad_norm": 0.3046875, + "learning_rate": 0.0018262037302020672, + "loss": 0.0947, + "step": 20110 + }, + { + "epoch": 0.1745731373859602, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018261861491937945, + "loss": 0.0874, + "step": 20111 + }, + { + "epoch": 0.17458181786616436, + "grad_norm": 0.123046875, + "learning_rate": 0.0018261685673913756, + "loss": 0.1318, + "step": 20112 + }, + { + "epoch": 0.17459049834636853, + "grad_norm": 0.498046875, + "learning_rate": 0.0018261509847948298, + "loss": 0.125, + "step": 20113 + }, + { + "epoch": 0.1745991788265727, + "grad_norm": 0.3125, + "learning_rate": 0.0018261334014041761, + "loss": 0.1006, + "step": 20114 + }, + { + "epoch": 0.17460785930677686, + "grad_norm": 0.5, + "learning_rate": 0.0018261158172194342, + "loss": 0.1074, + "step": 20115 + }, + { + "epoch": 0.17461653978698102, + "grad_norm": 0.75, + "learning_rate": 0.0018260982322406223, + "loss": 0.1118, + "step": 20116 + }, + { + "epoch": 0.1746252202671852, + "grad_norm": 0.2451171875, + "learning_rate": 0.0018260806464677607, + "loss": 0.0864, + "step": 20117 + }, + { + "epoch": 0.17463390074738935, + "grad_norm": 0.07763671875, + "learning_rate": 0.0018260630599008679, + "loss": 0.0801, + "step": 20118 + }, + { + "epoch": 0.17464258122759352, + "grad_norm": 1.3359375, + "learning_rate": 0.0018260454725399639, + "loss": 0.0967, + "step": 20119 + }, + { + "epoch": 0.17465126170779768, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018260278843850672, + "loss": 0.1035, + "step": 20120 + }, + { + "epoch": 0.17465994218800185, + "grad_norm": 0.59375, + "learning_rate": 0.0018260102954361975, + "loss": 0.1289, + "step": 20121 + }, + { + "epoch": 0.17466862266820601, + "grad_norm": 0.85546875, + "learning_rate": 0.0018259927056933733, + "loss": 0.1055, + "step": 20122 + }, + { + "epoch": 0.17467730314841018, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018259751151566148, + "loss": 0.0708, + "step": 20123 + }, + { + "epoch": 0.17468598362861434, + "grad_norm": 0.07080078125, + "learning_rate": 0.001825957523825941, + "loss": 0.1338, + "step": 20124 + }, + { + "epoch": 0.1746946641088185, + "grad_norm": 1.6015625, + "learning_rate": 0.001825939931701371, + "loss": 0.1309, + "step": 20125 + }, + { + "epoch": 0.17470334458902267, + "grad_norm": 0.154296875, + "learning_rate": 0.0018259223387829236, + "loss": 0.1133, + "step": 20126 + }, + { + "epoch": 0.17471202506922684, + "grad_norm": 0.3125, + "learning_rate": 0.0018259047450706186, + "loss": 0.1201, + "step": 20127 + }, + { + "epoch": 0.174720705549431, + "grad_norm": 0.1337890625, + "learning_rate": 0.001825887150564475, + "loss": 0.1123, + "step": 20128 + }, + { + "epoch": 0.17472938602963517, + "grad_norm": 0.37890625, + "learning_rate": 0.0018258695552645125, + "loss": 0.1064, + "step": 20129 + }, + { + "epoch": 0.17473806650983933, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018258519591707494, + "loss": 0.1055, + "step": 20130 + }, + { + "epoch": 0.1747467469900435, + "grad_norm": 0.275390625, + "learning_rate": 0.0018258343622832057, + "loss": 0.1289, + "step": 20131 + }, + { + "epoch": 0.17475542747024767, + "grad_norm": 0.109375, + "learning_rate": 0.0018258167646019005, + "loss": 0.1357, + "step": 20132 + }, + { + "epoch": 0.17476410795045183, + "grad_norm": 0.1328125, + "learning_rate": 0.0018257991661268529, + "loss": 0.1279, + "step": 20133 + }, + { + "epoch": 0.174772788430656, + "grad_norm": 0.3984375, + "learning_rate": 0.0018257815668580825, + "loss": 0.1221, + "step": 20134 + }, + { + "epoch": 0.17478146891086016, + "grad_norm": 0.2265625, + "learning_rate": 0.001825763966795608, + "loss": 0.165, + "step": 20135 + }, + { + "epoch": 0.17479014939106433, + "grad_norm": 0.37109375, + "learning_rate": 0.0018257463659394492, + "loss": 0.1011, + "step": 20136 + }, + { + "epoch": 0.1747988298712685, + "grad_norm": 0.123046875, + "learning_rate": 0.0018257287642896248, + "loss": 0.1016, + "step": 20137 + }, + { + "epoch": 0.17480751035147266, + "grad_norm": 0.2890625, + "learning_rate": 0.0018257111618461546, + "loss": 0.1387, + "step": 20138 + }, + { + "epoch": 0.17481619083167682, + "grad_norm": 0.07763671875, + "learning_rate": 0.0018256935586090574, + "loss": 0.1069, + "step": 20139 + }, + { + "epoch": 0.17482487131188099, + "grad_norm": 0.263671875, + "learning_rate": 0.0018256759545783525, + "loss": 0.1143, + "step": 20140 + }, + { + "epoch": 0.17483355179208515, + "grad_norm": 0.232421875, + "learning_rate": 0.0018256583497540595, + "loss": 0.1206, + "step": 20141 + }, + { + "epoch": 0.17484223227228932, + "grad_norm": 0.15234375, + "learning_rate": 0.0018256407441361972, + "loss": 0.1162, + "step": 20142 + }, + { + "epoch": 0.17485091275249348, + "grad_norm": 0.2265625, + "learning_rate": 0.0018256231377247855, + "loss": 0.0947, + "step": 20143 + }, + { + "epoch": 0.17485959323269765, + "grad_norm": 0.30078125, + "learning_rate": 0.001825605530519843, + "loss": 0.1021, + "step": 20144 + }, + { + "epoch": 0.17486827371290178, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018255879225213892, + "loss": 0.1094, + "step": 20145 + }, + { + "epoch": 0.17487695419310595, + "grad_norm": 0.181640625, + "learning_rate": 0.0018255703137294433, + "loss": 0.1465, + "step": 20146 + }, + { + "epoch": 0.17488563467331011, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018255527041440246, + "loss": 0.0947, + "step": 20147 + }, + { + "epoch": 0.17489431515351428, + "grad_norm": 0.3046875, + "learning_rate": 0.0018255350937651524, + "loss": 0.1523, + "step": 20148 + }, + { + "epoch": 0.17490299563371844, + "grad_norm": 0.419921875, + "learning_rate": 0.001825517482592846, + "loss": 0.1016, + "step": 20149 + }, + { + "epoch": 0.1749116761139226, + "grad_norm": 0.130859375, + "learning_rate": 0.0018254998706271244, + "loss": 0.1035, + "step": 20150 + }, + { + "epoch": 0.17492035659412677, + "grad_norm": 0.0849609375, + "learning_rate": 0.0018254822578680073, + "loss": 0.0825, + "step": 20151 + }, + { + "epoch": 0.17492903707433094, + "grad_norm": 0.134765625, + "learning_rate": 0.0018254646443155134, + "loss": 0.1416, + "step": 20152 + }, + { + "epoch": 0.1749377175545351, + "grad_norm": 0.287109375, + "learning_rate": 0.0018254470299696624, + "loss": 0.1138, + "step": 20153 + }, + { + "epoch": 0.17494639803473927, + "grad_norm": 0.423828125, + "learning_rate": 0.0018254294148304735, + "loss": 0.1309, + "step": 20154 + }, + { + "epoch": 0.17495507851494344, + "grad_norm": 0.88671875, + "learning_rate": 0.0018254117988979659, + "loss": 0.1113, + "step": 20155 + }, + { + "epoch": 0.1749637589951476, + "grad_norm": 0.58984375, + "learning_rate": 0.0018253941821721586, + "loss": 0.1025, + "step": 20156 + }, + { + "epoch": 0.17497243947535177, + "grad_norm": 0.390625, + "learning_rate": 0.0018253765646530713, + "loss": 0.1133, + "step": 20157 + }, + { + "epoch": 0.17498111995555593, + "grad_norm": 0.359375, + "learning_rate": 0.0018253589463407232, + "loss": 0.1006, + "step": 20158 + }, + { + "epoch": 0.1749898004357601, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018253413272351336, + "loss": 0.0967, + "step": 20159 + }, + { + "epoch": 0.17499848091596426, + "grad_norm": 0.10791015625, + "learning_rate": 0.001825323707336321, + "loss": 0.0898, + "step": 20160 + }, + { + "epoch": 0.17500716139616843, + "grad_norm": 0.158203125, + "learning_rate": 0.0018253060866443062, + "loss": 0.0811, + "step": 20161 + }, + { + "epoch": 0.1750158418763726, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018252884651591068, + "loss": 0.1133, + "step": 20162 + }, + { + "epoch": 0.17502452235657676, + "grad_norm": 0.35546875, + "learning_rate": 0.001825270842880743, + "loss": 0.1445, + "step": 20163 + }, + { + "epoch": 0.17503320283678092, + "grad_norm": 0.2275390625, + "learning_rate": 0.001825253219809234, + "loss": 0.167, + "step": 20164 + }, + { + "epoch": 0.1750418833169851, + "grad_norm": 0.97265625, + "learning_rate": 0.001825235595944599, + "loss": 0.0986, + "step": 20165 + }, + { + "epoch": 0.17505056379718925, + "grad_norm": 0.1904296875, + "learning_rate": 0.001825217971286857, + "loss": 0.0889, + "step": 20166 + }, + { + "epoch": 0.17505924427739342, + "grad_norm": 0.130859375, + "learning_rate": 0.0018252003458360277, + "loss": 0.1045, + "step": 20167 + }, + { + "epoch": 0.17506792475759758, + "grad_norm": 0.2421875, + "learning_rate": 0.0018251827195921305, + "loss": 0.1758, + "step": 20168 + }, + { + "epoch": 0.17507660523780175, + "grad_norm": 0.087890625, + "learning_rate": 0.0018251650925551841, + "loss": 0.0942, + "step": 20169 + }, + { + "epoch": 0.1750852857180059, + "grad_norm": 0.451171875, + "learning_rate": 0.001825147464725208, + "loss": 0.1089, + "step": 20170 + }, + { + "epoch": 0.17509396619821008, + "grad_norm": 0.71484375, + "learning_rate": 0.0018251298361022213, + "loss": 0.1338, + "step": 20171 + }, + { + "epoch": 0.17510264667841424, + "grad_norm": 0.173828125, + "learning_rate": 0.001825112206686244, + "loss": 0.1445, + "step": 20172 + }, + { + "epoch": 0.1751113271586184, + "grad_norm": 0.62109375, + "learning_rate": 0.001825094576477295, + "loss": 0.1367, + "step": 20173 + }, + { + "epoch": 0.17512000763882257, + "grad_norm": 0.0703125, + "learning_rate": 0.0018250769454753928, + "loss": 0.0864, + "step": 20174 + }, + { + "epoch": 0.17512868811902674, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018250593136805576, + "loss": 0.0972, + "step": 20175 + }, + { + "epoch": 0.1751373685992309, + "grad_norm": 0.09375, + "learning_rate": 0.0018250416810928086, + "loss": 0.0713, + "step": 20176 + }, + { + "epoch": 0.17514604907943507, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018250240477121647, + "loss": 0.085, + "step": 20177 + }, + { + "epoch": 0.17515472955963923, + "grad_norm": 0.23046875, + "learning_rate": 0.0018250064135386457, + "loss": 0.0981, + "step": 20178 + }, + { + "epoch": 0.1751634100398434, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018249887785722703, + "loss": 0.0864, + "step": 20179 + }, + { + "epoch": 0.17517209052004756, + "grad_norm": 0.3203125, + "learning_rate": 0.0018249711428130581, + "loss": 0.0977, + "step": 20180 + }, + { + "epoch": 0.17518077100025173, + "grad_norm": 0.310546875, + "learning_rate": 0.0018249535062610284, + "loss": 0.0806, + "step": 20181 + }, + { + "epoch": 0.1751894514804559, + "grad_norm": 0.58984375, + "learning_rate": 0.0018249358689162002, + "loss": 0.1016, + "step": 20182 + }, + { + "epoch": 0.17519813196066006, + "grad_norm": 0.201171875, + "learning_rate": 0.0018249182307785937, + "loss": 0.0811, + "step": 20183 + }, + { + "epoch": 0.17520681244086422, + "grad_norm": 0.66796875, + "learning_rate": 0.0018249005918482269, + "loss": 0.1108, + "step": 20184 + }, + { + "epoch": 0.1752154929210684, + "grad_norm": 0.10107421875, + "learning_rate": 0.0018248829521251197, + "loss": 0.0933, + "step": 20185 + }, + { + "epoch": 0.17522417340127255, + "grad_norm": 0.11572265625, + "learning_rate": 0.0018248653116092917, + "loss": 0.1133, + "step": 20186 + }, + { + "epoch": 0.17523285388147672, + "grad_norm": 0.30078125, + "learning_rate": 0.0018248476703007616, + "loss": 0.1074, + "step": 20187 + }, + { + "epoch": 0.17524153436168088, + "grad_norm": 0.109375, + "learning_rate": 0.0018248300281995494, + "loss": 0.0713, + "step": 20188 + }, + { + "epoch": 0.17525021484188505, + "grad_norm": 0.3515625, + "learning_rate": 0.0018248123853056735, + "loss": 0.0967, + "step": 20189 + }, + { + "epoch": 0.17525889532208921, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018247947416191541, + "loss": 0.1104, + "step": 20190 + }, + { + "epoch": 0.17526757580229338, + "grad_norm": 0.8828125, + "learning_rate": 0.0018247770971400097, + "loss": 0.1445, + "step": 20191 + }, + { + "epoch": 0.17527625628249754, + "grad_norm": 0.482421875, + "learning_rate": 0.0018247594518682603, + "loss": 0.1729, + "step": 20192 + }, + { + "epoch": 0.1752849367627017, + "grad_norm": 1.3203125, + "learning_rate": 0.0018247418058039244, + "loss": 0.1177, + "step": 20193 + }, + { + "epoch": 0.17529361724290587, + "grad_norm": 0.330078125, + "learning_rate": 0.001824724158947022, + "loss": 0.0918, + "step": 20194 + }, + { + "epoch": 0.17530229772311004, + "grad_norm": 0.578125, + "learning_rate": 0.0018247065112975724, + "loss": 0.0874, + "step": 20195 + }, + { + "epoch": 0.1753109782033142, + "grad_norm": 0.232421875, + "learning_rate": 0.0018246888628555944, + "loss": 0.0615, + "step": 20196 + }, + { + "epoch": 0.17531965868351837, + "grad_norm": 0.494140625, + "learning_rate": 0.0018246712136211075, + "loss": 0.1582, + "step": 20197 + }, + { + "epoch": 0.17532833916372254, + "grad_norm": 0.58984375, + "learning_rate": 0.001824653563594131, + "loss": 0.1455, + "step": 20198 + }, + { + "epoch": 0.1753370196439267, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018246359127746846, + "loss": 0.0977, + "step": 20199 + }, + { + "epoch": 0.17534570012413087, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018246182611627868, + "loss": 0.1079, + "step": 20200 + }, + { + "epoch": 0.17535438060433503, + "grad_norm": 0.09423828125, + "learning_rate": 0.0018246006087584577, + "loss": 0.1328, + "step": 20201 + }, + { + "epoch": 0.1753630610845392, + "grad_norm": 0.400390625, + "learning_rate": 0.001824582955561716, + "loss": 0.1143, + "step": 20202 + }, + { + "epoch": 0.17537174156474336, + "grad_norm": 0.189453125, + "learning_rate": 0.0018245653015725814, + "loss": 0.1069, + "step": 20203 + }, + { + "epoch": 0.17538042204494753, + "grad_norm": 0.306640625, + "learning_rate": 0.0018245476467910729, + "loss": 0.1143, + "step": 20204 + }, + { + "epoch": 0.1753891025251517, + "grad_norm": 0.1044921875, + "learning_rate": 0.0018245299912172102, + "loss": 0.1152, + "step": 20205 + }, + { + "epoch": 0.17539778300535586, + "grad_norm": 0.431640625, + "learning_rate": 0.0018245123348510125, + "loss": 0.103, + "step": 20206 + }, + { + "epoch": 0.17540646348556002, + "grad_norm": 0.84375, + "learning_rate": 0.0018244946776924988, + "loss": 0.1084, + "step": 20207 + }, + { + "epoch": 0.1754151439657642, + "grad_norm": 0.08203125, + "learning_rate": 0.0018244770197416884, + "loss": 0.0972, + "step": 20208 + }, + { + "epoch": 0.17542382444596835, + "grad_norm": 0.349609375, + "learning_rate": 0.001824459360998601, + "loss": 0.1152, + "step": 20209 + }, + { + "epoch": 0.17543250492617252, + "grad_norm": 0.248046875, + "learning_rate": 0.0018244417014632562, + "loss": 0.0928, + "step": 20210 + }, + { + "epoch": 0.17544118540637668, + "grad_norm": 0.50390625, + "learning_rate": 0.001824424041135672, + "loss": 0.1084, + "step": 20211 + }, + { + "epoch": 0.17544986588658085, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018244063800158692, + "loss": 0.0913, + "step": 20212 + }, + { + "epoch": 0.175458546366785, + "grad_norm": 0.314453125, + "learning_rate": 0.0018243887181038662, + "loss": 0.1182, + "step": 20213 + }, + { + "epoch": 0.17546722684698918, + "grad_norm": 0.236328125, + "learning_rate": 0.0018243710553996826, + "loss": 0.1064, + "step": 20214 + }, + { + "epoch": 0.17547590732719334, + "grad_norm": 0.55859375, + "learning_rate": 0.0018243533919033378, + "loss": 0.1436, + "step": 20215 + }, + { + "epoch": 0.1754845878073975, + "grad_norm": 0.263671875, + "learning_rate": 0.0018243357276148508, + "loss": 0.0757, + "step": 20216 + }, + { + "epoch": 0.17549326828760167, + "grad_norm": 0.150390625, + "learning_rate": 0.0018243180625342414, + "loss": 0.1113, + "step": 20217 + }, + { + "epoch": 0.17550194876780584, + "grad_norm": 0.28515625, + "learning_rate": 0.0018243003966615285, + "loss": 0.123, + "step": 20218 + }, + { + "epoch": 0.17551062924801, + "grad_norm": 0.8359375, + "learning_rate": 0.0018242827299967315, + "loss": 0.124, + "step": 20219 + }, + { + "epoch": 0.17551930972821417, + "grad_norm": 0.2294921875, + "learning_rate": 0.00182426506253987, + "loss": 0.1172, + "step": 20220 + }, + { + "epoch": 0.17552799020841833, + "grad_norm": 0.271484375, + "learning_rate": 0.0018242473942909627, + "loss": 0.1533, + "step": 20221 + }, + { + "epoch": 0.1755366706886225, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018242297252500296, + "loss": 0.103, + "step": 20222 + }, + { + "epoch": 0.17554535116882666, + "grad_norm": 0.66015625, + "learning_rate": 0.0018242120554170897, + "loss": 0.1006, + "step": 20223 + }, + { + "epoch": 0.17555403164903083, + "grad_norm": 0.259765625, + "learning_rate": 0.0018241943847921627, + "loss": 0.1387, + "step": 20224 + }, + { + "epoch": 0.175562712129235, + "grad_norm": 0.765625, + "learning_rate": 0.0018241767133752673, + "loss": 0.1387, + "step": 20225 + }, + { + "epoch": 0.17557139260943916, + "grad_norm": 0.68359375, + "learning_rate": 0.001824159041166423, + "loss": 0.1436, + "step": 20226 + }, + { + "epoch": 0.17558007308964332, + "grad_norm": 0.140625, + "learning_rate": 0.0018241413681656493, + "loss": 0.1553, + "step": 20227 + }, + { + "epoch": 0.1755887535698475, + "grad_norm": 1.1640625, + "learning_rate": 0.0018241236943729656, + "loss": 0.1309, + "step": 20228 + }, + { + "epoch": 0.17559743405005165, + "grad_norm": 0.2294921875, + "learning_rate": 0.0018241060197883914, + "loss": 0.1328, + "step": 20229 + }, + { + "epoch": 0.17560611453025582, + "grad_norm": 0.298828125, + "learning_rate": 0.0018240883444119451, + "loss": 0.1406, + "step": 20230 + }, + { + "epoch": 0.17561479501045998, + "grad_norm": 0.134765625, + "learning_rate": 0.0018240706682436474, + "loss": 0.1484, + "step": 20231 + }, + { + "epoch": 0.17562347549066415, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018240529912835163, + "loss": 0.123, + "step": 20232 + }, + { + "epoch": 0.17563215597086831, + "grad_norm": 0.2265625, + "learning_rate": 0.001824035313531572, + "loss": 0.0903, + "step": 20233 + }, + { + "epoch": 0.17564083645107248, + "grad_norm": 0.51953125, + "learning_rate": 0.0018240176349878334, + "loss": 0.125, + "step": 20234 + }, + { + "epoch": 0.17564951693127664, + "grad_norm": 0.109375, + "learning_rate": 0.00182399995565232, + "loss": 0.1099, + "step": 20235 + }, + { + "epoch": 0.1756581974114808, + "grad_norm": 0.11083984375, + "learning_rate": 0.0018239822755250514, + "loss": 0.1338, + "step": 20236 + }, + { + "epoch": 0.17566687789168497, + "grad_norm": 0.138671875, + "learning_rate": 0.0018239645946060464, + "loss": 0.1089, + "step": 20237 + }, + { + "epoch": 0.17567555837188914, + "grad_norm": 0.5625, + "learning_rate": 0.0018239469128953248, + "loss": 0.1836, + "step": 20238 + }, + { + "epoch": 0.1756842388520933, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018239292303929057, + "loss": 0.1338, + "step": 20239 + }, + { + "epoch": 0.17569291933229747, + "grad_norm": 0.197265625, + "learning_rate": 0.0018239115470988082, + "loss": 0.0752, + "step": 20240 + }, + { + "epoch": 0.17570159981250164, + "grad_norm": 0.1552734375, + "learning_rate": 0.001823893863013052, + "loss": 0.1201, + "step": 20241 + }, + { + "epoch": 0.1757102802927058, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018238761781356565, + "loss": 0.0898, + "step": 20242 + }, + { + "epoch": 0.17571896077290997, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018238584924666406, + "loss": 0.1348, + "step": 20243 + }, + { + "epoch": 0.17572764125311413, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018238408060060243, + "loss": 0.1196, + "step": 20244 + }, + { + "epoch": 0.1757363217333183, + "grad_norm": 0.283203125, + "learning_rate": 0.0018238231187538265, + "loss": 0.1221, + "step": 20245 + }, + { + "epoch": 0.17574500221352246, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018238054307100665, + "loss": 0.1318, + "step": 20246 + }, + { + "epoch": 0.17575368269372663, + "grad_norm": 0.166015625, + "learning_rate": 0.001823787741874764, + "loss": 0.1387, + "step": 20247 + }, + { + "epoch": 0.1757623631739308, + "grad_norm": 0.08984375, + "learning_rate": 0.0018237700522479377, + "loss": 0.1367, + "step": 20248 + }, + { + "epoch": 0.17577104365413496, + "grad_norm": 0.4609375, + "learning_rate": 0.0018237523618296074, + "loss": 0.1162, + "step": 20249 + }, + { + "epoch": 0.17577972413433912, + "grad_norm": 0.322265625, + "learning_rate": 0.0018237346706197926, + "loss": 0.0679, + "step": 20250 + }, + { + "epoch": 0.1757884046145433, + "grad_norm": 1.65625, + "learning_rate": 0.0018237169786185124, + "loss": 0.1465, + "step": 20251 + }, + { + "epoch": 0.17579708509474745, + "grad_norm": 0.42578125, + "learning_rate": 0.001823699285825786, + "loss": 0.0825, + "step": 20252 + }, + { + "epoch": 0.17580576557495162, + "grad_norm": 0.3203125, + "learning_rate": 0.0018236815922416335, + "loss": 0.126, + "step": 20253 + }, + { + "epoch": 0.17581444605515578, + "grad_norm": 0.302734375, + "learning_rate": 0.0018236638978660728, + "loss": 0.1172, + "step": 20254 + }, + { + "epoch": 0.17582312653535995, + "grad_norm": 0.140625, + "learning_rate": 0.001823646202699125, + "loss": 0.126, + "step": 20255 + }, + { + "epoch": 0.1758318070155641, + "grad_norm": 0.345703125, + "learning_rate": 0.001823628506740808, + "loss": 0.0874, + "step": 20256 + }, + { + "epoch": 0.17584048749576828, + "grad_norm": 0.392578125, + "learning_rate": 0.001823610809991142, + "loss": 0.1475, + "step": 20257 + }, + { + "epoch": 0.17584916797597244, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018235931124501459, + "loss": 0.1094, + "step": 20258 + }, + { + "epoch": 0.1758578484561766, + "grad_norm": 0.7890625, + "learning_rate": 0.0018235754141178394, + "loss": 0.082, + "step": 20259 + }, + { + "epoch": 0.17586652893638077, + "grad_norm": 0.322265625, + "learning_rate": 0.0018235577149942415, + "loss": 0.1367, + "step": 20260 + }, + { + "epoch": 0.17587520941658494, + "grad_norm": 0.12890625, + "learning_rate": 0.001823540015079372, + "loss": 0.1191, + "step": 20261 + }, + { + "epoch": 0.1758838898967891, + "grad_norm": 0.259765625, + "learning_rate": 0.00182352231437325, + "loss": 0.126, + "step": 20262 + }, + { + "epoch": 0.17589257037699327, + "grad_norm": 0.33984375, + "learning_rate": 0.0018235046128758947, + "loss": 0.1143, + "step": 20263 + }, + { + "epoch": 0.17590125085719743, + "grad_norm": 0.33203125, + "learning_rate": 0.001823486910587326, + "loss": 0.126, + "step": 20264 + }, + { + "epoch": 0.1759099313374016, + "grad_norm": 1.515625, + "learning_rate": 0.0018234692075075622, + "loss": 0.7734, + "step": 20265 + }, + { + "epoch": 0.17591861181760576, + "grad_norm": 0.41796875, + "learning_rate": 0.0018234515036366237, + "loss": 0.082, + "step": 20266 + }, + { + "epoch": 0.17592729229780993, + "grad_norm": 0.138671875, + "learning_rate": 0.0018234337989745297, + "loss": 0.1191, + "step": 20267 + }, + { + "epoch": 0.17593597277801407, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018234160935212994, + "loss": 0.0938, + "step": 20268 + }, + { + "epoch": 0.17594465325821823, + "grad_norm": 0.107421875, + "learning_rate": 0.001823398387276952, + "loss": 0.0903, + "step": 20269 + }, + { + "epoch": 0.1759533337384224, + "grad_norm": 0.09228515625, + "learning_rate": 0.0018233806802415067, + "loss": 0.0947, + "step": 20270 + }, + { + "epoch": 0.17596201421862656, + "grad_norm": 0.25390625, + "learning_rate": 0.0018233629724149832, + "loss": 0.1865, + "step": 20271 + }, + { + "epoch": 0.17597069469883073, + "grad_norm": 0.2294921875, + "learning_rate": 0.0018233452637974013, + "loss": 0.1089, + "step": 20272 + }, + { + "epoch": 0.1759793751790349, + "grad_norm": 0.4296875, + "learning_rate": 0.0018233275543887795, + "loss": 0.1191, + "step": 20273 + }, + { + "epoch": 0.17598805565923906, + "grad_norm": 0.294921875, + "learning_rate": 0.0018233098441891378, + "loss": 0.1074, + "step": 20274 + }, + { + "epoch": 0.17599673613944322, + "grad_norm": 0.310546875, + "learning_rate": 0.001823292133198495, + "loss": 0.1299, + "step": 20275 + }, + { + "epoch": 0.1760054166196474, + "grad_norm": 0.60546875, + "learning_rate": 0.001823274421416871, + "loss": 0.1157, + "step": 20276 + }, + { + "epoch": 0.17601409709985155, + "grad_norm": 0.41015625, + "learning_rate": 0.0018232567088442848, + "loss": 0.1191, + "step": 20277 + }, + { + "epoch": 0.17602277758005572, + "grad_norm": 0.11279296875, + "learning_rate": 0.0018232389954807562, + "loss": 0.1113, + "step": 20278 + }, + { + "epoch": 0.17603145806025988, + "grad_norm": 0.455078125, + "learning_rate": 0.0018232212813263044, + "loss": 0.1094, + "step": 20279 + }, + { + "epoch": 0.17604013854046405, + "grad_norm": 0.51171875, + "learning_rate": 0.0018232035663809482, + "loss": 0.1016, + "step": 20280 + }, + { + "epoch": 0.1760488190206682, + "grad_norm": 0.07421875, + "learning_rate": 0.0018231858506447078, + "loss": 0.1089, + "step": 20281 + }, + { + "epoch": 0.17605749950087238, + "grad_norm": 0.07666015625, + "learning_rate": 0.001823168134117602, + "loss": 0.1152, + "step": 20282 + }, + { + "epoch": 0.17606617998107654, + "grad_norm": 0.703125, + "learning_rate": 0.0018231504167996505, + "loss": 0.1182, + "step": 20283 + }, + { + "epoch": 0.1760748604612807, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018231326986908724, + "loss": 0.1079, + "step": 20284 + }, + { + "epoch": 0.17608354094148487, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018231149797912876, + "loss": 0.1406, + "step": 20285 + }, + { + "epoch": 0.17609222142168904, + "grad_norm": 0.10302734375, + "learning_rate": 0.001823097260100915, + "loss": 0.0796, + "step": 20286 + }, + { + "epoch": 0.1761009019018932, + "grad_norm": 0.2578125, + "learning_rate": 0.001823079539619774, + "loss": 0.1396, + "step": 20287 + }, + { + "epoch": 0.17610958238209737, + "grad_norm": 0.4609375, + "learning_rate": 0.0018230618183478842, + "loss": 0.1318, + "step": 20288 + }, + { + "epoch": 0.17611826286230153, + "grad_norm": 0.6484375, + "learning_rate": 0.0018230440962852645, + "loss": 0.166, + "step": 20289 + }, + { + "epoch": 0.1761269433425057, + "grad_norm": 0.24609375, + "learning_rate": 0.001823026373431935, + "loss": 0.1279, + "step": 20290 + }, + { + "epoch": 0.17613562382270986, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018230086497879145, + "loss": 0.1226, + "step": 20291 + }, + { + "epoch": 0.17614430430291403, + "grad_norm": 0.423828125, + "learning_rate": 0.001822990925353223, + "loss": 0.1533, + "step": 20292 + }, + { + "epoch": 0.1761529847831182, + "grad_norm": 0.498046875, + "learning_rate": 0.0018229732001278792, + "loss": 0.0967, + "step": 20293 + }, + { + "epoch": 0.17616166526332236, + "grad_norm": 0.73046875, + "learning_rate": 0.0018229554741119028, + "loss": 0.1172, + "step": 20294 + }, + { + "epoch": 0.17617034574352652, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018229377473053133, + "loss": 0.0874, + "step": 20295 + }, + { + "epoch": 0.1761790262237307, + "grad_norm": 0.5390625, + "learning_rate": 0.0018229200197081295, + "loss": 0.1514, + "step": 20296 + }, + { + "epoch": 0.17618770670393485, + "grad_norm": 0.73828125, + "learning_rate": 0.0018229022913203714, + "loss": 0.1182, + "step": 20297 + }, + { + "epoch": 0.17619638718413902, + "grad_norm": 0.173828125, + "learning_rate": 0.0018228845621420585, + "loss": 0.104, + "step": 20298 + }, + { + "epoch": 0.17620506766434318, + "grad_norm": 0.1455078125, + "learning_rate": 0.0018228668321732097, + "loss": 0.085, + "step": 20299 + }, + { + "epoch": 0.17621374814454735, + "grad_norm": 0.3984375, + "learning_rate": 0.0018228491014138446, + "loss": 0.1465, + "step": 20300 + }, + { + "epoch": 0.17622242862475151, + "grad_norm": 0.56640625, + "learning_rate": 0.0018228313698639828, + "loss": 0.1592, + "step": 20301 + }, + { + "epoch": 0.17623110910495568, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018228136375236432, + "loss": 0.1338, + "step": 20302 + }, + { + "epoch": 0.17623978958515985, + "grad_norm": 0.77734375, + "learning_rate": 0.0018227959043928457, + "loss": 0.0991, + "step": 20303 + }, + { + "epoch": 0.176248470065364, + "grad_norm": 0.25390625, + "learning_rate": 0.0018227781704716094, + "loss": 0.1338, + "step": 20304 + }, + { + "epoch": 0.17625715054556818, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018227604357599534, + "loss": 0.0913, + "step": 20305 + }, + { + "epoch": 0.17626583102577234, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018227427002578977, + "loss": 0.1318, + "step": 20306 + }, + { + "epoch": 0.1762745115059765, + "grad_norm": 0.0712890625, + "learning_rate": 0.0018227249639654614, + "loss": 0.0908, + "step": 20307 + }, + { + "epoch": 0.17628319198618067, + "grad_norm": 0.427734375, + "learning_rate": 0.0018227072268826641, + "loss": 0.0967, + "step": 20308 + }, + { + "epoch": 0.17629187246638484, + "grad_norm": 0.66015625, + "learning_rate": 0.0018226894890095249, + "loss": 0.1133, + "step": 20309 + }, + { + "epoch": 0.176300552946589, + "grad_norm": 0.0791015625, + "learning_rate": 0.0018226717503460634, + "loss": 0.0938, + "step": 20310 + }, + { + "epoch": 0.17630923342679317, + "grad_norm": 0.2890625, + "learning_rate": 0.001822654010892299, + "loss": 0.0928, + "step": 20311 + }, + { + "epoch": 0.17631791390699733, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018226362706482509, + "loss": 0.1123, + "step": 20312 + }, + { + "epoch": 0.1763265943872015, + "grad_norm": 0.78515625, + "learning_rate": 0.0018226185296139387, + "loss": 0.1465, + "step": 20313 + }, + { + "epoch": 0.17633527486740566, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018226007877893816, + "loss": 0.1001, + "step": 20314 + }, + { + "epoch": 0.17634395534760983, + "grad_norm": 0.2890625, + "learning_rate": 0.0018225830451745993, + "loss": 0.127, + "step": 20315 + }, + { + "epoch": 0.176352635827814, + "grad_norm": 0.78515625, + "learning_rate": 0.001822565301769611, + "loss": 0.1113, + "step": 20316 + }, + { + "epoch": 0.17636131630801816, + "grad_norm": 0.2119140625, + "learning_rate": 0.001822547557574436, + "loss": 0.0747, + "step": 20317 + }, + { + "epoch": 0.17636999678822232, + "grad_norm": 0.09765625, + "learning_rate": 0.0018225298125890941, + "loss": 0.124, + "step": 20318 + }, + { + "epoch": 0.1763786772684265, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018225120668136042, + "loss": 0.1182, + "step": 20319 + }, + { + "epoch": 0.17638735774863065, + "grad_norm": 0.203125, + "learning_rate": 0.001822494320247986, + "loss": 0.1299, + "step": 20320 + }, + { + "epoch": 0.17639603822883482, + "grad_norm": 0.2578125, + "learning_rate": 0.001822476572892259, + "loss": 0.104, + "step": 20321 + }, + { + "epoch": 0.17640471870903898, + "grad_norm": 0.30859375, + "learning_rate": 0.0018224588247464425, + "loss": 0.1406, + "step": 20322 + }, + { + "epoch": 0.17641339918924315, + "grad_norm": 0.302734375, + "learning_rate": 0.0018224410758105557, + "loss": 0.0845, + "step": 20323 + }, + { + "epoch": 0.1764220796694473, + "grad_norm": 0.423828125, + "learning_rate": 0.0018224233260846184, + "loss": 0.0879, + "step": 20324 + }, + { + "epoch": 0.17643076014965148, + "grad_norm": 0.36328125, + "learning_rate": 0.0018224055755686494, + "loss": 0.0918, + "step": 20325 + }, + { + "epoch": 0.17643944062985564, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018223878242626687, + "loss": 0.1245, + "step": 20326 + }, + { + "epoch": 0.1764481211100598, + "grad_norm": 0.158203125, + "learning_rate": 0.001822370072166696, + "loss": 0.1494, + "step": 20327 + }, + { + "epoch": 0.17645680159026397, + "grad_norm": 0.3671875, + "learning_rate": 0.0018223523192807496, + "loss": 0.1396, + "step": 20328 + }, + { + "epoch": 0.17646548207046814, + "grad_norm": 0.921875, + "learning_rate": 0.0018223345656048502, + "loss": 0.1035, + "step": 20329 + }, + { + "epoch": 0.1764741625506723, + "grad_norm": 0.15234375, + "learning_rate": 0.001822316811139016, + "loss": 0.1338, + "step": 20330 + }, + { + "epoch": 0.17648284303087647, + "grad_norm": 0.56640625, + "learning_rate": 0.0018222990558832673, + "loss": 0.1172, + "step": 20331 + }, + { + "epoch": 0.17649152351108063, + "grad_norm": 0.275390625, + "learning_rate": 0.0018222812998376233, + "loss": 0.0962, + "step": 20332 + }, + { + "epoch": 0.1765002039912848, + "grad_norm": 0.6015625, + "learning_rate": 0.0018222635430021029, + "loss": 0.0996, + "step": 20333 + }, + { + "epoch": 0.17650888447148896, + "grad_norm": 0.337890625, + "learning_rate": 0.0018222457853767262, + "loss": 0.1133, + "step": 20334 + }, + { + "epoch": 0.17651756495169313, + "grad_norm": 0.439453125, + "learning_rate": 0.0018222280269615124, + "loss": 0.0991, + "step": 20335 + }, + { + "epoch": 0.1765262454318973, + "grad_norm": 0.34375, + "learning_rate": 0.0018222102677564805, + "loss": 0.0845, + "step": 20336 + }, + { + "epoch": 0.17653492591210146, + "grad_norm": 0.0732421875, + "learning_rate": 0.0018221925077616511, + "loss": 0.1172, + "step": 20337 + }, + { + "epoch": 0.17654360639230562, + "grad_norm": 0.66015625, + "learning_rate": 0.001822174746977042, + "loss": 0.1104, + "step": 20338 + }, + { + "epoch": 0.1765522868725098, + "grad_norm": 0.181640625, + "learning_rate": 0.0018221569854026739, + "loss": 0.0977, + "step": 20339 + }, + { + "epoch": 0.17656096735271395, + "grad_norm": 0.390625, + "learning_rate": 0.0018221392230385657, + "loss": 0.0752, + "step": 20340 + }, + { + "epoch": 0.17656964783291812, + "grad_norm": 0.330078125, + "learning_rate": 0.001822121459884737, + "loss": 0.1826, + "step": 20341 + }, + { + "epoch": 0.17657832831312228, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018221036959412068, + "loss": 0.082, + "step": 20342 + }, + { + "epoch": 0.17658700879332645, + "grad_norm": 0.1484375, + "learning_rate": 0.0018220859312079952, + "loss": 0.1162, + "step": 20343 + }, + { + "epoch": 0.17659568927353061, + "grad_norm": 0.0888671875, + "learning_rate": 0.0018220681656851208, + "loss": 0.0767, + "step": 20344 + }, + { + "epoch": 0.17660436975373478, + "grad_norm": 0.146484375, + "learning_rate": 0.001822050399372604, + "loss": 0.1001, + "step": 20345 + }, + { + "epoch": 0.17661305023393895, + "grad_norm": 0.62109375, + "learning_rate": 0.0018220326322704635, + "loss": 0.1113, + "step": 20346 + }, + { + "epoch": 0.1766217307141431, + "grad_norm": 0.17578125, + "learning_rate": 0.0018220148643787188, + "loss": 0.1011, + "step": 20347 + }, + { + "epoch": 0.17663041119434728, + "grad_norm": 0.287109375, + "learning_rate": 0.0018219970956973898, + "loss": 0.1279, + "step": 20348 + }, + { + "epoch": 0.17663909167455144, + "grad_norm": 0.447265625, + "learning_rate": 0.0018219793262264955, + "loss": 0.1143, + "step": 20349 + }, + { + "epoch": 0.1766477721547556, + "grad_norm": 0.36328125, + "learning_rate": 0.0018219615559660557, + "loss": 0.1064, + "step": 20350 + }, + { + "epoch": 0.17665645263495977, + "grad_norm": 0.24609375, + "learning_rate": 0.001821943784916089, + "loss": 0.0947, + "step": 20351 + }, + { + "epoch": 0.17666513311516394, + "grad_norm": 0.0654296875, + "learning_rate": 0.001821926013076616, + "loss": 0.0977, + "step": 20352 + }, + { + "epoch": 0.1766738135953681, + "grad_norm": 0.53125, + "learning_rate": 0.0018219082404476552, + "loss": 0.1211, + "step": 20353 + }, + { + "epoch": 0.17668249407557227, + "grad_norm": 0.212890625, + "learning_rate": 0.0018218904670292265, + "loss": 0.0977, + "step": 20354 + }, + { + "epoch": 0.17669117455577643, + "grad_norm": 0.12255859375, + "learning_rate": 0.001821872692821349, + "loss": 0.1377, + "step": 20355 + }, + { + "epoch": 0.1766998550359806, + "grad_norm": 0.162109375, + "learning_rate": 0.0018218549178240428, + "loss": 0.1055, + "step": 20356 + }, + { + "epoch": 0.17670853551618476, + "grad_norm": 0.44921875, + "learning_rate": 0.001821837142037327, + "loss": 0.1611, + "step": 20357 + }, + { + "epoch": 0.17671721599638893, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018218193654612207, + "loss": 0.1133, + "step": 20358 + }, + { + "epoch": 0.1767258964765931, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018218015880957434, + "loss": 0.1021, + "step": 20359 + }, + { + "epoch": 0.17673457695679726, + "grad_norm": 0.3515625, + "learning_rate": 0.001821783809940915, + "loss": 0.1299, + "step": 20360 + }, + { + "epoch": 0.17674325743700142, + "grad_norm": 0.19921875, + "learning_rate": 0.0018217660309967545, + "loss": 0.1484, + "step": 20361 + }, + { + "epoch": 0.1767519379172056, + "grad_norm": 0.66015625, + "learning_rate": 0.0018217482512632814, + "loss": 0.1309, + "step": 20362 + }, + { + "epoch": 0.17676061839740975, + "grad_norm": 0.66796875, + "learning_rate": 0.0018217304707405157, + "loss": 0.1719, + "step": 20363 + }, + { + "epoch": 0.17676929887761392, + "grad_norm": 3.65625, + "learning_rate": 0.0018217126894284762, + "loss": 0.375, + "step": 20364 + }, + { + "epoch": 0.17677797935781808, + "grad_norm": 1.1328125, + "learning_rate": 0.0018216949073271824, + "loss": 0.1201, + "step": 20365 + }, + { + "epoch": 0.17678665983802225, + "grad_norm": 0.44921875, + "learning_rate": 0.001821677124436654, + "loss": 0.0967, + "step": 20366 + }, + { + "epoch": 0.1767953403182264, + "grad_norm": 0.1875, + "learning_rate": 0.00182165934075691, + "loss": 0.0957, + "step": 20367 + }, + { + "epoch": 0.17680402079843058, + "grad_norm": 0.703125, + "learning_rate": 0.0018216415562879707, + "loss": 0.291, + "step": 20368 + }, + { + "epoch": 0.17681270127863474, + "grad_norm": 0.107421875, + "learning_rate": 0.0018216237710298548, + "loss": 0.1465, + "step": 20369 + }, + { + "epoch": 0.1768213817588389, + "grad_norm": 0.95703125, + "learning_rate": 0.0018216059849825823, + "loss": 0.1348, + "step": 20370 + }, + { + "epoch": 0.17683006223904307, + "grad_norm": 0.48046875, + "learning_rate": 0.0018215881981461718, + "loss": 0.126, + "step": 20371 + }, + { + "epoch": 0.17683874271924724, + "grad_norm": 0.384765625, + "learning_rate": 0.0018215704105206437, + "loss": 0.1553, + "step": 20372 + }, + { + "epoch": 0.1768474231994514, + "grad_norm": 0.6875, + "learning_rate": 0.001821552622106017, + "loss": 0.1328, + "step": 20373 + }, + { + "epoch": 0.17685610367965557, + "grad_norm": 1.0078125, + "learning_rate": 0.0018215348329023112, + "loss": 0.1016, + "step": 20374 + }, + { + "epoch": 0.17686478415985973, + "grad_norm": 0.7421875, + "learning_rate": 0.0018215170429095454, + "loss": 0.105, + "step": 20375 + }, + { + "epoch": 0.1768734646400639, + "grad_norm": 0.169921875, + "learning_rate": 0.0018214992521277396, + "loss": 0.1289, + "step": 20376 + }, + { + "epoch": 0.17688214512026806, + "grad_norm": 0.58984375, + "learning_rate": 0.0018214814605569131, + "loss": 0.1299, + "step": 20377 + }, + { + "epoch": 0.17689082560047223, + "grad_norm": 0.1240234375, + "learning_rate": 0.0018214636681970855, + "loss": 0.1025, + "step": 20378 + }, + { + "epoch": 0.1768995060806764, + "grad_norm": 0.984375, + "learning_rate": 0.001821445875048276, + "loss": 0.106, + "step": 20379 + }, + { + "epoch": 0.17690818656088056, + "grad_norm": 0.1005859375, + "learning_rate": 0.0018214280811105039, + "loss": 0.0986, + "step": 20380 + }, + { + "epoch": 0.17691686704108472, + "grad_norm": 0.099609375, + "learning_rate": 0.001821410286383789, + "loss": 0.0986, + "step": 20381 + }, + { + "epoch": 0.1769255475212889, + "grad_norm": 0.3828125, + "learning_rate": 0.0018213924908681508, + "loss": 0.1113, + "step": 20382 + }, + { + "epoch": 0.17693422800149305, + "grad_norm": 0.67578125, + "learning_rate": 0.0018213746945636082, + "loss": 0.1289, + "step": 20383 + }, + { + "epoch": 0.17694290848169722, + "grad_norm": 0.1171875, + "learning_rate": 0.0018213568974701814, + "loss": 0.0957, + "step": 20384 + }, + { + "epoch": 0.17695158896190138, + "grad_norm": 0.69921875, + "learning_rate": 0.0018213390995878895, + "loss": 0.1104, + "step": 20385 + }, + { + "epoch": 0.17696026944210555, + "grad_norm": 0.1396484375, + "learning_rate": 0.001821321300916752, + "loss": 0.0776, + "step": 20386 + }, + { + "epoch": 0.17696894992230972, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018213035014567882, + "loss": 0.0952, + "step": 20387 + }, + { + "epoch": 0.17697763040251388, + "grad_norm": 0.1728515625, + "learning_rate": 0.001821285701208018, + "loss": 0.106, + "step": 20388 + }, + { + "epoch": 0.17698631088271805, + "grad_norm": 0.083984375, + "learning_rate": 0.0018212679001704604, + "loss": 0.1035, + "step": 20389 + }, + { + "epoch": 0.1769949913629222, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018212500983441353, + "loss": 0.1162, + "step": 20390 + }, + { + "epoch": 0.17700367184312638, + "grad_norm": 0.392578125, + "learning_rate": 0.0018212322957290618, + "loss": 0.1084, + "step": 20391 + }, + { + "epoch": 0.1770123523233305, + "grad_norm": 0.259765625, + "learning_rate": 0.0018212144923252592, + "loss": 0.1133, + "step": 20392 + }, + { + "epoch": 0.17702103280353468, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018211966881327478, + "loss": 0.0859, + "step": 20393 + }, + { + "epoch": 0.17702971328373884, + "grad_norm": 0.515625, + "learning_rate": 0.0018211788831515462, + "loss": 0.127, + "step": 20394 + }, + { + "epoch": 0.177038393763943, + "grad_norm": 2.890625, + "learning_rate": 0.0018211610773816742, + "loss": 0.2129, + "step": 20395 + }, + { + "epoch": 0.17704707424414717, + "grad_norm": 0.47265625, + "learning_rate": 0.0018211432708231511, + "loss": 0.1094, + "step": 20396 + }, + { + "epoch": 0.17705575472435134, + "grad_norm": 0.337890625, + "learning_rate": 0.001821125463475997, + "loss": 0.1035, + "step": 20397 + }, + { + "epoch": 0.1770644352045555, + "grad_norm": 0.703125, + "learning_rate": 0.0018211076553402308, + "loss": 0.1387, + "step": 20398 + }, + { + "epoch": 0.17707311568475967, + "grad_norm": 0.1484375, + "learning_rate": 0.001821089846415872, + "loss": 0.1182, + "step": 20399 + }, + { + "epoch": 0.17708179616496383, + "grad_norm": 0.57421875, + "learning_rate": 0.0018210720367029401, + "loss": 0.083, + "step": 20400 + }, + { + "epoch": 0.177090476645168, + "grad_norm": 0.296875, + "learning_rate": 0.0018210542262014546, + "loss": 0.1328, + "step": 20401 + }, + { + "epoch": 0.17709915712537216, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018210364149114356, + "loss": 0.083, + "step": 20402 + }, + { + "epoch": 0.17710783760557633, + "grad_norm": 0.482421875, + "learning_rate": 0.0018210186028329014, + "loss": 0.1328, + "step": 20403 + }, + { + "epoch": 0.1771165180857805, + "grad_norm": 0.0673828125, + "learning_rate": 0.0018210007899658724, + "loss": 0.0708, + "step": 20404 + }, + { + "epoch": 0.17712519856598466, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018209829763103675, + "loss": 0.1152, + "step": 20405 + }, + { + "epoch": 0.17713387904618882, + "grad_norm": 0.5625, + "learning_rate": 0.0018209651618664067, + "loss": 0.1436, + "step": 20406 + }, + { + "epoch": 0.177142559526393, + "grad_norm": 0.154296875, + "learning_rate": 0.001820947346634009, + "loss": 0.1553, + "step": 20407 + }, + { + "epoch": 0.17715124000659715, + "grad_norm": 0.271484375, + "learning_rate": 0.0018209295306131947, + "loss": 0.1328, + "step": 20408 + }, + { + "epoch": 0.17715992048680132, + "grad_norm": 0.28515625, + "learning_rate": 0.0018209117138039822, + "loss": 0.1201, + "step": 20409 + }, + { + "epoch": 0.17716860096700549, + "grad_norm": 0.1123046875, + "learning_rate": 0.0018208938962063913, + "loss": 0.0879, + "step": 20410 + }, + { + "epoch": 0.17717728144720965, + "grad_norm": 0.53125, + "learning_rate": 0.001820876077820442, + "loss": 0.1099, + "step": 20411 + }, + { + "epoch": 0.17718596192741382, + "grad_norm": 0.13671875, + "learning_rate": 0.0018208582586461534, + "loss": 0.0986, + "step": 20412 + }, + { + "epoch": 0.17719464240761798, + "grad_norm": 0.47265625, + "learning_rate": 0.0018208404386835448, + "loss": 0.1221, + "step": 20413 + }, + { + "epoch": 0.17720332288782215, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018208226179326364, + "loss": 0.1094, + "step": 20414 + }, + { + "epoch": 0.1772120033680263, + "grad_norm": 0.380859375, + "learning_rate": 0.0018208047963934467, + "loss": 0.0933, + "step": 20415 + }, + { + "epoch": 0.17722068384823048, + "grad_norm": 0.23046875, + "learning_rate": 0.001820786974065996, + "loss": 0.1021, + "step": 20416 + }, + { + "epoch": 0.17722936432843464, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018207691509503035, + "loss": 0.0679, + "step": 20417 + }, + { + "epoch": 0.1772380448086388, + "grad_norm": 0.46875, + "learning_rate": 0.0018207513270463885, + "loss": 0.0806, + "step": 20418 + }, + { + "epoch": 0.17724672528884297, + "grad_norm": 0.291015625, + "learning_rate": 0.0018207335023542708, + "loss": 0.0879, + "step": 20419 + }, + { + "epoch": 0.17725540576904714, + "grad_norm": 0.1513671875, + "learning_rate": 0.00182071567687397, + "loss": 0.1035, + "step": 20420 + }, + { + "epoch": 0.1772640862492513, + "grad_norm": 0.09619140625, + "learning_rate": 0.001820697850605505, + "loss": 0.0898, + "step": 20421 + }, + { + "epoch": 0.17727276672945547, + "grad_norm": 0.330078125, + "learning_rate": 0.001820680023548896, + "loss": 0.1074, + "step": 20422 + }, + { + "epoch": 0.17728144720965963, + "grad_norm": 0.55078125, + "learning_rate": 0.0018206621957041619, + "loss": 0.1357, + "step": 20423 + }, + { + "epoch": 0.1772901276898638, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018206443670713223, + "loss": 0.1504, + "step": 20424 + }, + { + "epoch": 0.17729880817006796, + "grad_norm": 0.15234375, + "learning_rate": 0.001820626537650397, + "loss": 0.1201, + "step": 20425 + }, + { + "epoch": 0.17730748865027213, + "grad_norm": 0.29296875, + "learning_rate": 0.0018206087074414054, + "loss": 0.1201, + "step": 20426 + }, + { + "epoch": 0.1773161691304763, + "grad_norm": 0.22265625, + "learning_rate": 0.0018205908764443671, + "loss": 0.1523, + "step": 20427 + }, + { + "epoch": 0.17732484961068046, + "grad_norm": 0.34375, + "learning_rate": 0.0018205730446593012, + "loss": 0.1016, + "step": 20428 + }, + { + "epoch": 0.17733353009088462, + "grad_norm": 0.275390625, + "learning_rate": 0.0018205552120862274, + "loss": 0.1123, + "step": 20429 + }, + { + "epoch": 0.1773422105710888, + "grad_norm": 0.51953125, + "learning_rate": 0.0018205373787251653, + "loss": 0.1025, + "step": 20430 + }, + { + "epoch": 0.17735089105129295, + "grad_norm": 0.380859375, + "learning_rate": 0.0018205195445761347, + "loss": 0.1357, + "step": 20431 + }, + { + "epoch": 0.17735957153149712, + "grad_norm": 0.326171875, + "learning_rate": 0.0018205017096391545, + "loss": 0.082, + "step": 20432 + }, + { + "epoch": 0.17736825201170128, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018204838739142445, + "loss": 0.0957, + "step": 20433 + }, + { + "epoch": 0.17737693249190545, + "grad_norm": 0.380859375, + "learning_rate": 0.0018204660374014242, + "loss": 0.0898, + "step": 20434 + }, + { + "epoch": 0.1773856129721096, + "grad_norm": 0.3671875, + "learning_rate": 0.0018204482001007128, + "loss": 0.1191, + "step": 20435 + }, + { + "epoch": 0.17739429345231378, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018204303620121302, + "loss": 0.1025, + "step": 20436 + }, + { + "epoch": 0.17740297393251794, + "grad_norm": 0.10205078125, + "learning_rate": 0.0018204125231356958, + "loss": 0.127, + "step": 20437 + }, + { + "epoch": 0.1774116544127221, + "grad_norm": 0.431640625, + "learning_rate": 0.0018203946834714291, + "loss": 0.124, + "step": 20438 + }, + { + "epoch": 0.17742033489292627, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018203768430193495, + "loss": 0.1143, + "step": 20439 + }, + { + "epoch": 0.17742901537313044, + "grad_norm": 2.28125, + "learning_rate": 0.0018203590017794769, + "loss": 0.1562, + "step": 20440 + }, + { + "epoch": 0.1774376958533346, + "grad_norm": 1.0390625, + "learning_rate": 0.0018203411597518303, + "loss": 0.1289, + "step": 20441 + }, + { + "epoch": 0.17744637633353877, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018203233169364296, + "loss": 0.1089, + "step": 20442 + }, + { + "epoch": 0.17745505681374293, + "grad_norm": 0.53125, + "learning_rate": 0.0018203054733332935, + "loss": 0.125, + "step": 20443 + }, + { + "epoch": 0.1774637372939471, + "grad_norm": 0.447265625, + "learning_rate": 0.001820287628942443, + "loss": 0.1113, + "step": 20444 + }, + { + "epoch": 0.17747241777415126, + "grad_norm": 0.515625, + "learning_rate": 0.0018202697837638965, + "loss": 0.0918, + "step": 20445 + }, + { + "epoch": 0.17748109825435543, + "grad_norm": 0.1728515625, + "learning_rate": 0.0018202519377976734, + "loss": 0.1079, + "step": 20446 + }, + { + "epoch": 0.1774897787345596, + "grad_norm": 0.1494140625, + "learning_rate": 0.001820234091043794, + "loss": 0.0923, + "step": 20447 + }, + { + "epoch": 0.17749845921476376, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018202162435022773, + "loss": 0.0996, + "step": 20448 + }, + { + "epoch": 0.17750713969496792, + "grad_norm": 0.65625, + "learning_rate": 0.001820198395173143, + "loss": 0.126, + "step": 20449 + }, + { + "epoch": 0.1775158201751721, + "grad_norm": 0.41796875, + "learning_rate": 0.0018201805460564106, + "loss": 0.083, + "step": 20450 + }, + { + "epoch": 0.17752450065537626, + "grad_norm": 0.50390625, + "learning_rate": 0.0018201626961520997, + "loss": 0.1416, + "step": 20451 + }, + { + "epoch": 0.17753318113558042, + "grad_norm": 0.30859375, + "learning_rate": 0.0018201448454602295, + "loss": 0.1143, + "step": 20452 + }, + { + "epoch": 0.17754186161578459, + "grad_norm": 0.60546875, + "learning_rate": 0.0018201269939808196, + "loss": 0.1182, + "step": 20453 + }, + { + "epoch": 0.17755054209598875, + "grad_norm": 0.28515625, + "learning_rate": 0.0018201091417138896, + "loss": 0.125, + "step": 20454 + }, + { + "epoch": 0.17755922257619292, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018200912886594592, + "loss": 0.1455, + "step": 20455 + }, + { + "epoch": 0.17756790305639708, + "grad_norm": 0.37109375, + "learning_rate": 0.0018200734348175478, + "loss": 0.1074, + "step": 20456 + }, + { + "epoch": 0.17757658353660125, + "grad_norm": 0.5078125, + "learning_rate": 0.0018200555801881752, + "loss": 0.1201, + "step": 20457 + }, + { + "epoch": 0.1775852640168054, + "grad_norm": 0.42578125, + "learning_rate": 0.0018200377247713602, + "loss": 0.1074, + "step": 20458 + }, + { + "epoch": 0.17759394449700958, + "grad_norm": 0.1435546875, + "learning_rate": 0.001820019868567123, + "loss": 0.1328, + "step": 20459 + }, + { + "epoch": 0.17760262497721374, + "grad_norm": 0.34765625, + "learning_rate": 0.001820002011575483, + "loss": 0.1523, + "step": 20460 + }, + { + "epoch": 0.1776113054574179, + "grad_norm": 0.19921875, + "learning_rate": 0.0018199841537964595, + "loss": 0.1533, + "step": 20461 + }, + { + "epoch": 0.17761998593762207, + "grad_norm": 0.68359375, + "learning_rate": 0.0018199662952300716, + "loss": 0.1162, + "step": 20462 + }, + { + "epoch": 0.17762866641782624, + "grad_norm": 0.57421875, + "learning_rate": 0.00181994843587634, + "loss": 0.1011, + "step": 20463 + }, + { + "epoch": 0.1776373468980304, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018199305757352834, + "loss": 0.1143, + "step": 20464 + }, + { + "epoch": 0.17764602737823457, + "grad_norm": 0.482421875, + "learning_rate": 0.0018199127148069217, + "loss": 0.0947, + "step": 20465 + }, + { + "epoch": 0.17765470785843873, + "grad_norm": 1.140625, + "learning_rate": 0.0018198948530912741, + "loss": 0.1221, + "step": 20466 + }, + { + "epoch": 0.1776633883386429, + "grad_norm": 0.447265625, + "learning_rate": 0.0018198769905883603, + "loss": 0.1152, + "step": 20467 + }, + { + "epoch": 0.17767206881884706, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018198591272982, + "loss": 0.0967, + "step": 20468 + }, + { + "epoch": 0.17768074929905123, + "grad_norm": 0.09375, + "learning_rate": 0.0018198412632208125, + "loss": 0.0986, + "step": 20469 + }, + { + "epoch": 0.1776894297792554, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018198233983562175, + "loss": 0.1631, + "step": 20470 + }, + { + "epoch": 0.17769811025945956, + "grad_norm": 0.5, + "learning_rate": 0.0018198055327044342, + "loss": 0.1182, + "step": 20471 + }, + { + "epoch": 0.17770679073966372, + "grad_norm": 0.7265625, + "learning_rate": 0.0018197876662654825, + "loss": 0.1406, + "step": 20472 + }, + { + "epoch": 0.1777154712198679, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018197697990393817, + "loss": 0.1611, + "step": 20473 + }, + { + "epoch": 0.17772415170007205, + "grad_norm": 0.119140625, + "learning_rate": 0.0018197519310261519, + "loss": 0.0928, + "step": 20474 + }, + { + "epoch": 0.17773283218027622, + "grad_norm": 0.142578125, + "learning_rate": 0.001819734062225812, + "loss": 0.0947, + "step": 20475 + }, + { + "epoch": 0.17774151266048038, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018197161926383817, + "loss": 0.0903, + "step": 20476 + }, + { + "epoch": 0.17775019314068455, + "grad_norm": 0.302734375, + "learning_rate": 0.0018196983222638803, + "loss": 0.1318, + "step": 20477 + }, + { + "epoch": 0.1777588736208887, + "grad_norm": 0.40625, + "learning_rate": 0.0018196804511023282, + "loss": 0.0967, + "step": 20478 + }, + { + "epoch": 0.17776755410109288, + "grad_norm": 0.158203125, + "learning_rate": 0.0018196625791537437, + "loss": 0.1055, + "step": 20479 + }, + { + "epoch": 0.17777623458129704, + "grad_norm": 0.81640625, + "learning_rate": 0.0018196447064181475, + "loss": 0.1055, + "step": 20480 + }, + { + "epoch": 0.1777849150615012, + "grad_norm": 0.08154296875, + "learning_rate": 0.0018196268328955583, + "loss": 0.0874, + "step": 20481 + }, + { + "epoch": 0.17779359554170537, + "grad_norm": 0.490234375, + "learning_rate": 0.0018196089585859963, + "loss": 0.1387, + "step": 20482 + }, + { + "epoch": 0.17780227602190954, + "grad_norm": 0.609375, + "learning_rate": 0.0018195910834894807, + "loss": 0.1016, + "step": 20483 + }, + { + "epoch": 0.1778109565021137, + "grad_norm": 0.50390625, + "learning_rate": 0.0018195732076060312, + "loss": 0.105, + "step": 20484 + }, + { + "epoch": 0.17781963698231787, + "grad_norm": 0.3984375, + "learning_rate": 0.001819555330935667, + "loss": 0.0908, + "step": 20485 + }, + { + "epoch": 0.17782831746252203, + "grad_norm": 0.283203125, + "learning_rate": 0.0018195374534784082, + "loss": 0.1147, + "step": 20486 + }, + { + "epoch": 0.1778369979427262, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018195195752342738, + "loss": 0.1055, + "step": 20487 + }, + { + "epoch": 0.17784567842293036, + "grad_norm": 0.404296875, + "learning_rate": 0.0018195016962032835, + "loss": 0.0723, + "step": 20488 + }, + { + "epoch": 0.17785435890313453, + "grad_norm": 0.443359375, + "learning_rate": 0.0018194838163854573, + "loss": 0.1816, + "step": 20489 + }, + { + "epoch": 0.1778630393833387, + "grad_norm": 0.515625, + "learning_rate": 0.0018194659357808143, + "loss": 0.106, + "step": 20490 + }, + { + "epoch": 0.17787171986354286, + "grad_norm": 0.380859375, + "learning_rate": 0.001819448054389374, + "loss": 0.1094, + "step": 20491 + }, + { + "epoch": 0.17788040034374702, + "grad_norm": 0.484375, + "learning_rate": 0.0018194301722111563, + "loss": 0.0889, + "step": 20492 + }, + { + "epoch": 0.1778890808239512, + "grad_norm": 0.203125, + "learning_rate": 0.0018194122892461807, + "loss": 0.1074, + "step": 20493 + }, + { + "epoch": 0.17789776130415536, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018193944054944661, + "loss": 0.0962, + "step": 20494 + }, + { + "epoch": 0.17790644178435952, + "grad_norm": 0.181640625, + "learning_rate": 0.0018193765209560332, + "loss": 0.124, + "step": 20495 + }, + { + "epoch": 0.17791512226456369, + "grad_norm": 0.310546875, + "learning_rate": 0.0018193586356309009, + "loss": 0.1216, + "step": 20496 + }, + { + "epoch": 0.17792380274476785, + "grad_norm": 0.373046875, + "learning_rate": 0.0018193407495190883, + "loss": 0.1084, + "step": 20497 + }, + { + "epoch": 0.17793248322497202, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018193228626206157, + "loss": 0.0996, + "step": 20498 + }, + { + "epoch": 0.17794116370517618, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018193049749355027, + "loss": 0.0874, + "step": 20499 + }, + { + "epoch": 0.17794984418538035, + "grad_norm": 0.318359375, + "learning_rate": 0.0018192870864637682, + "loss": 0.1172, + "step": 20500 + }, + { + "epoch": 0.1779585246655845, + "grad_norm": 0.6875, + "learning_rate": 0.0018192691972054322, + "loss": 0.0889, + "step": 20501 + }, + { + "epoch": 0.17796720514578868, + "grad_norm": 0.259765625, + "learning_rate": 0.0018192513071605143, + "loss": 0.1396, + "step": 20502 + }, + { + "epoch": 0.17797588562599284, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018192334163290341, + "loss": 0.1338, + "step": 20503 + }, + { + "epoch": 0.177984566106197, + "grad_norm": 0.10546875, + "learning_rate": 0.0018192155247110109, + "loss": 0.082, + "step": 20504 + }, + { + "epoch": 0.17799324658640117, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018191976323064646, + "loss": 0.1309, + "step": 20505 + }, + { + "epoch": 0.17800192706660534, + "grad_norm": 0.15625, + "learning_rate": 0.0018191797391154142, + "loss": 0.0801, + "step": 20506 + }, + { + "epoch": 0.1780106075468095, + "grad_norm": 0.32421875, + "learning_rate": 0.00181916184513788, + "loss": 0.1377, + "step": 20507 + }, + { + "epoch": 0.17801928802701367, + "grad_norm": 0.431640625, + "learning_rate": 0.0018191439503738812, + "loss": 0.1572, + "step": 20508 + }, + { + "epoch": 0.17802796850721783, + "grad_norm": 0.640625, + "learning_rate": 0.0018191260548234371, + "loss": 0.1084, + "step": 20509 + }, + { + "epoch": 0.178036648987422, + "grad_norm": 0.30078125, + "learning_rate": 0.0018191081584865679, + "loss": 0.1021, + "step": 20510 + }, + { + "epoch": 0.17804532946762616, + "grad_norm": 0.419921875, + "learning_rate": 0.0018190902613632925, + "loss": 0.0977, + "step": 20511 + }, + { + "epoch": 0.17805400994783033, + "grad_norm": 0.13671875, + "learning_rate": 0.001819072363453631, + "loss": 0.0977, + "step": 20512 + }, + { + "epoch": 0.1780626904280345, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018190544647576025, + "loss": 0.0957, + "step": 20513 + }, + { + "epoch": 0.17807137090823866, + "grad_norm": 0.7109375, + "learning_rate": 0.0018190365652752272, + "loss": 0.1035, + "step": 20514 + }, + { + "epoch": 0.1780800513884428, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018190186650065242, + "loss": 0.1104, + "step": 20515 + }, + { + "epoch": 0.17808873186864696, + "grad_norm": 0.23828125, + "learning_rate": 0.001819000763951513, + "loss": 0.1484, + "step": 20516 + }, + { + "epoch": 0.17809741234885113, + "grad_norm": 0.45703125, + "learning_rate": 0.0018189828621102135, + "loss": 0.1406, + "step": 20517 + }, + { + "epoch": 0.1781060928290553, + "grad_norm": 0.220703125, + "learning_rate": 0.0018189649594826452, + "loss": 0.0986, + "step": 20518 + }, + { + "epoch": 0.17811477330925946, + "grad_norm": 0.953125, + "learning_rate": 0.0018189470560688278, + "loss": 0.105, + "step": 20519 + }, + { + "epoch": 0.17812345378946362, + "grad_norm": 0.8203125, + "learning_rate": 0.0018189291518687805, + "loss": 0.1162, + "step": 20520 + }, + { + "epoch": 0.17813213426966779, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018189112468825228, + "loss": 0.0918, + "step": 20521 + }, + { + "epoch": 0.17814081474987195, + "grad_norm": 0.1396484375, + "learning_rate": 0.0018188933411100747, + "loss": 0.1289, + "step": 20522 + }, + { + "epoch": 0.17814949523007612, + "grad_norm": 0.11767578125, + "learning_rate": 0.001818875434551456, + "loss": 0.1187, + "step": 20523 + }, + { + "epoch": 0.17815817571028028, + "grad_norm": 0.5546875, + "learning_rate": 0.0018188575272066853, + "loss": 0.123, + "step": 20524 + }, + { + "epoch": 0.17816685619048445, + "grad_norm": 0.455078125, + "learning_rate": 0.0018188396190757835, + "loss": 0.1113, + "step": 20525 + }, + { + "epoch": 0.1781755366706886, + "grad_norm": 0.2890625, + "learning_rate": 0.0018188217101587692, + "loss": 0.1357, + "step": 20526 + }, + { + "epoch": 0.17818421715089278, + "grad_norm": 0.119140625, + "learning_rate": 0.0018188038004556622, + "loss": 0.1289, + "step": 20527 + }, + { + "epoch": 0.17819289763109694, + "grad_norm": 0.54296875, + "learning_rate": 0.0018187858899664823, + "loss": 0.125, + "step": 20528 + }, + { + "epoch": 0.1782015781113011, + "grad_norm": 0.1484375, + "learning_rate": 0.001818767978691249, + "loss": 0.1523, + "step": 20529 + }, + { + "epoch": 0.17821025859150527, + "grad_norm": 0.333984375, + "learning_rate": 0.0018187500666299818, + "loss": 0.1191, + "step": 20530 + }, + { + "epoch": 0.17821893907170944, + "grad_norm": 0.53125, + "learning_rate": 0.0018187321537827, + "loss": 0.0908, + "step": 20531 + }, + { + "epoch": 0.1782276195519136, + "grad_norm": 0.2294921875, + "learning_rate": 0.001818714240149424, + "loss": 0.0928, + "step": 20532 + }, + { + "epoch": 0.17823630003211777, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018186963257301725, + "loss": 0.1172, + "step": 20533 + }, + { + "epoch": 0.17824498051232193, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018186784105249656, + "loss": 0.1445, + "step": 20534 + }, + { + "epoch": 0.1782536609925261, + "grad_norm": 0.51171875, + "learning_rate": 0.001818660494533823, + "loss": 0.1045, + "step": 20535 + }, + { + "epoch": 0.17826234147273026, + "grad_norm": 0.578125, + "learning_rate": 0.001818642577756764, + "loss": 0.0957, + "step": 20536 + }, + { + "epoch": 0.17827102195293443, + "grad_norm": 0.232421875, + "learning_rate": 0.001818624660193808, + "loss": 0.1138, + "step": 20537 + }, + { + "epoch": 0.1782797024331386, + "grad_norm": 0.251953125, + "learning_rate": 0.001818606741844975, + "loss": 0.0889, + "step": 20538 + }, + { + "epoch": 0.17828838291334276, + "grad_norm": 0.71875, + "learning_rate": 0.0018185888227102846, + "loss": 0.0996, + "step": 20539 + }, + { + "epoch": 0.17829706339354692, + "grad_norm": 0.69140625, + "learning_rate": 0.0018185709027897564, + "loss": 0.1367, + "step": 20540 + }, + { + "epoch": 0.1783057438737511, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018185529820834096, + "loss": 0.0752, + "step": 20541 + }, + { + "epoch": 0.17831442435395525, + "grad_norm": 0.2890625, + "learning_rate": 0.001818535060591264, + "loss": 0.1152, + "step": 20542 + }, + { + "epoch": 0.17832310483415942, + "grad_norm": 0.302734375, + "learning_rate": 0.0018185171383133395, + "loss": 0.1455, + "step": 20543 + }, + { + "epoch": 0.17833178531436358, + "grad_norm": 0.578125, + "learning_rate": 0.001818499215249655, + "loss": 0.0977, + "step": 20544 + }, + { + "epoch": 0.17834046579456775, + "grad_norm": 0.126953125, + "learning_rate": 0.001818481291400231, + "loss": 0.1553, + "step": 20545 + }, + { + "epoch": 0.1783491462747719, + "grad_norm": 0.578125, + "learning_rate": 0.0018184633667650866, + "loss": 0.1436, + "step": 20546 + }, + { + "epoch": 0.17835782675497608, + "grad_norm": 0.201171875, + "learning_rate": 0.0018184454413442414, + "loss": 0.0898, + "step": 20547 + }, + { + "epoch": 0.17836650723518024, + "grad_norm": 0.7734375, + "learning_rate": 0.0018184275151377149, + "loss": 0.1377, + "step": 20548 + }, + { + "epoch": 0.1783751877153844, + "grad_norm": 0.099609375, + "learning_rate": 0.001818409588145527, + "loss": 0.0972, + "step": 20549 + }, + { + "epoch": 0.17838386819558857, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018183916603676975, + "loss": 0.1211, + "step": 20550 + }, + { + "epoch": 0.17839254867579274, + "grad_norm": 0.39453125, + "learning_rate": 0.0018183737318042453, + "loss": 0.1177, + "step": 20551 + }, + { + "epoch": 0.1784012291559969, + "grad_norm": 0.220703125, + "learning_rate": 0.0018183558024551902, + "loss": 0.106, + "step": 20552 + }, + { + "epoch": 0.17840990963620107, + "grad_norm": 0.58984375, + "learning_rate": 0.0018183378723205523, + "loss": 0.0996, + "step": 20553 + }, + { + "epoch": 0.17841859011640523, + "grad_norm": 0.2294921875, + "learning_rate": 0.0018183199414003508, + "loss": 0.085, + "step": 20554 + }, + { + "epoch": 0.1784272705966094, + "grad_norm": 0.52734375, + "learning_rate": 0.0018183020096946055, + "loss": 0.0684, + "step": 20555 + }, + { + "epoch": 0.17843595107681356, + "grad_norm": 0.4921875, + "learning_rate": 0.001818284077203336, + "loss": 0.1875, + "step": 20556 + }, + { + "epoch": 0.17844463155701773, + "grad_norm": 0.203125, + "learning_rate": 0.0018182661439265614, + "loss": 0.1289, + "step": 20557 + }, + { + "epoch": 0.1784533120372219, + "grad_norm": 0.2001953125, + "learning_rate": 0.001818248209864302, + "loss": 0.0981, + "step": 20558 + }, + { + "epoch": 0.17846199251742606, + "grad_norm": 0.3984375, + "learning_rate": 0.0018182302750165771, + "loss": 0.1099, + "step": 20559 + }, + { + "epoch": 0.17847067299763023, + "grad_norm": 0.37109375, + "learning_rate": 0.0018182123393834064, + "loss": 0.1025, + "step": 20560 + }, + { + "epoch": 0.1784793534778344, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018181944029648097, + "loss": 0.1118, + "step": 20561 + }, + { + "epoch": 0.17848803395803856, + "grad_norm": 0.326171875, + "learning_rate": 0.0018181764657608061, + "loss": 0.1035, + "step": 20562 + }, + { + "epoch": 0.17849671443824272, + "grad_norm": 0.416015625, + "learning_rate": 0.0018181585277714156, + "loss": 0.0986, + "step": 20563 + }, + { + "epoch": 0.17850539491844689, + "grad_norm": 0.353515625, + "learning_rate": 0.0018181405889966579, + "loss": 0.1113, + "step": 20564 + }, + { + "epoch": 0.17851407539865105, + "grad_norm": 0.1572265625, + "learning_rate": 0.0018181226494365522, + "loss": 0.0996, + "step": 20565 + }, + { + "epoch": 0.17852275587885522, + "grad_norm": 0.267578125, + "learning_rate": 0.0018181047090911181, + "loss": 0.0996, + "step": 20566 + }, + { + "epoch": 0.17853143635905938, + "grad_norm": 0.1767578125, + "learning_rate": 0.0018180867679603759, + "loss": 0.0972, + "step": 20567 + }, + { + "epoch": 0.17854011683926355, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018180688260443448, + "loss": 0.1309, + "step": 20568 + }, + { + "epoch": 0.1785487973194677, + "grad_norm": 0.314453125, + "learning_rate": 0.0018180508833430442, + "loss": 0.1631, + "step": 20569 + }, + { + "epoch": 0.17855747779967188, + "grad_norm": 0.47265625, + "learning_rate": 0.001818032939856494, + "loss": 0.1309, + "step": 20570 + }, + { + "epoch": 0.17856615827987604, + "grad_norm": 0.18359375, + "learning_rate": 0.0018180149955847138, + "loss": 0.0894, + "step": 20571 + }, + { + "epoch": 0.1785748387600802, + "grad_norm": 0.39453125, + "learning_rate": 0.0018179970505277233, + "loss": 0.0933, + "step": 20572 + }, + { + "epoch": 0.17858351924028437, + "grad_norm": 0.400390625, + "learning_rate": 0.001817979104685542, + "loss": 0.0942, + "step": 20573 + }, + { + "epoch": 0.17859219972048854, + "grad_norm": 0.236328125, + "learning_rate": 0.0018179611580581895, + "loss": 0.1035, + "step": 20574 + }, + { + "epoch": 0.1786008802006927, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018179432106456852, + "loss": 0.0879, + "step": 20575 + }, + { + "epoch": 0.17860956068089687, + "grad_norm": 0.36328125, + "learning_rate": 0.0018179252624480494, + "loss": 0.1025, + "step": 20576 + }, + { + "epoch": 0.17861824116110103, + "grad_norm": 0.388671875, + "learning_rate": 0.001817907313465301, + "loss": 0.166, + "step": 20577 + }, + { + "epoch": 0.1786269216413052, + "grad_norm": 0.44140625, + "learning_rate": 0.0018178893636974602, + "loss": 0.1211, + "step": 20578 + }, + { + "epoch": 0.17863560212150936, + "grad_norm": 0.11328125, + "learning_rate": 0.0018178714131445461, + "loss": 0.0894, + "step": 20579 + }, + { + "epoch": 0.17864428260171353, + "grad_norm": 0.1279296875, + "learning_rate": 0.001817853461806579, + "loss": 0.0928, + "step": 20580 + }, + { + "epoch": 0.1786529630819177, + "grad_norm": 0.1123046875, + "learning_rate": 0.001817835509683578, + "loss": 0.1367, + "step": 20581 + }, + { + "epoch": 0.17866164356212186, + "grad_norm": 0.09814453125, + "learning_rate": 0.0018178175567755625, + "loss": 0.1123, + "step": 20582 + }, + { + "epoch": 0.17867032404232602, + "grad_norm": 0.0986328125, + "learning_rate": 0.001817799603082553, + "loss": 0.105, + "step": 20583 + }, + { + "epoch": 0.1786790045225302, + "grad_norm": 0.3828125, + "learning_rate": 0.0018177816486045684, + "loss": 0.1367, + "step": 20584 + }, + { + "epoch": 0.17868768500273435, + "grad_norm": 0.62109375, + "learning_rate": 0.0018177636933416287, + "loss": 0.1113, + "step": 20585 + }, + { + "epoch": 0.17869636548293852, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018177457372937531, + "loss": 0.1108, + "step": 20586 + }, + { + "epoch": 0.17870504596314268, + "grad_norm": 0.482421875, + "learning_rate": 0.0018177277804609618, + "loss": 0.1191, + "step": 20587 + }, + { + "epoch": 0.17871372644334685, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018177098228432742, + "loss": 0.1001, + "step": 20588 + }, + { + "epoch": 0.178722406923551, + "grad_norm": 0.26171875, + "learning_rate": 0.0018176918644407097, + "loss": 0.1289, + "step": 20589 + }, + { + "epoch": 0.17873108740375518, + "grad_norm": 0.546875, + "learning_rate": 0.0018176739052532885, + "loss": 0.1104, + "step": 20590 + }, + { + "epoch": 0.17873976788395934, + "grad_norm": 0.21484375, + "learning_rate": 0.0018176559452810295, + "loss": 0.1089, + "step": 20591 + }, + { + "epoch": 0.1787484483641635, + "grad_norm": 0.2265625, + "learning_rate": 0.0018176379845239532, + "loss": 0.1006, + "step": 20592 + }, + { + "epoch": 0.17875712884436767, + "grad_norm": 0.236328125, + "learning_rate": 0.0018176200229820785, + "loss": 0.1357, + "step": 20593 + }, + { + "epoch": 0.17876580932457184, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018176020606554253, + "loss": 0.083, + "step": 20594 + }, + { + "epoch": 0.178774489804776, + "grad_norm": 0.25, + "learning_rate": 0.0018175840975440131, + "loss": 0.1118, + "step": 20595 + }, + { + "epoch": 0.17878317028498017, + "grad_norm": 0.3203125, + "learning_rate": 0.0018175661336478624, + "loss": 0.1016, + "step": 20596 + }, + { + "epoch": 0.17879185076518433, + "grad_norm": 0.140625, + "learning_rate": 0.0018175481689669914, + "loss": 0.1216, + "step": 20597 + }, + { + "epoch": 0.1788005312453885, + "grad_norm": 0.1513671875, + "learning_rate": 0.001817530203501421, + "loss": 0.1021, + "step": 20598 + }, + { + "epoch": 0.17880921172559266, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018175122372511704, + "loss": 0.1235, + "step": 20599 + }, + { + "epoch": 0.17881789220579683, + "grad_norm": 0.408203125, + "learning_rate": 0.0018174942702162589, + "loss": 0.1016, + "step": 20600 + }, + { + "epoch": 0.178826572686001, + "grad_norm": 0.486328125, + "learning_rate": 0.0018174763023967066, + "loss": 0.1162, + "step": 20601 + }, + { + "epoch": 0.17883525316620516, + "grad_norm": 0.0810546875, + "learning_rate": 0.0018174583337925326, + "loss": 0.0923, + "step": 20602 + }, + { + "epoch": 0.17884393364640933, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018174403644037572, + "loss": 0.1631, + "step": 20603 + }, + { + "epoch": 0.1788526141266135, + "grad_norm": 0.310546875, + "learning_rate": 0.0018174223942303998, + "loss": 0.1045, + "step": 20604 + }, + { + "epoch": 0.17886129460681766, + "grad_norm": 0.7109375, + "learning_rate": 0.0018174044232724803, + "loss": 0.1318, + "step": 20605 + }, + { + "epoch": 0.17886997508702182, + "grad_norm": 0.11767578125, + "learning_rate": 0.0018173864515300178, + "loss": 0.0835, + "step": 20606 + }, + { + "epoch": 0.17887865556722599, + "grad_norm": 0.333984375, + "learning_rate": 0.0018173684790030323, + "loss": 0.1064, + "step": 20607 + }, + { + "epoch": 0.17888733604743015, + "grad_norm": 0.53515625, + "learning_rate": 0.0018173505056915437, + "loss": 0.1123, + "step": 20608 + }, + { + "epoch": 0.17889601652763432, + "grad_norm": 0.74609375, + "learning_rate": 0.0018173325315955708, + "loss": 0.1084, + "step": 20609 + }, + { + "epoch": 0.17890469700783848, + "grad_norm": 0.29296875, + "learning_rate": 0.0018173145567151343, + "loss": 0.1484, + "step": 20610 + }, + { + "epoch": 0.17891337748804265, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018172965810502534, + "loss": 0.1123, + "step": 20611 + }, + { + "epoch": 0.1789220579682468, + "grad_norm": 0.79296875, + "learning_rate": 0.0018172786046009473, + "loss": 0.1245, + "step": 20612 + }, + { + "epoch": 0.17893073844845098, + "grad_norm": 0.4765625, + "learning_rate": 0.0018172606273672365, + "loss": 0.1348, + "step": 20613 + }, + { + "epoch": 0.17893941892865514, + "grad_norm": 0.365234375, + "learning_rate": 0.0018172426493491402, + "loss": 0.123, + "step": 20614 + }, + { + "epoch": 0.1789480994088593, + "grad_norm": 0.1708984375, + "learning_rate": 0.001817224670546678, + "loss": 0.0845, + "step": 20615 + }, + { + "epoch": 0.17895677988906347, + "grad_norm": 0.6484375, + "learning_rate": 0.0018172066909598696, + "loss": 0.1206, + "step": 20616 + }, + { + "epoch": 0.17896546036926764, + "grad_norm": 0.419921875, + "learning_rate": 0.0018171887105887347, + "loss": 0.1543, + "step": 20617 + }, + { + "epoch": 0.1789741408494718, + "grad_norm": 1.078125, + "learning_rate": 0.0018171707294332934, + "loss": 0.2754, + "step": 20618 + }, + { + "epoch": 0.17898282132967597, + "grad_norm": 0.39453125, + "learning_rate": 0.0018171527474935647, + "loss": 0.1338, + "step": 20619 + }, + { + "epoch": 0.17899150180988013, + "grad_norm": 0.40625, + "learning_rate": 0.0018171347647695684, + "loss": 0.3359, + "step": 20620 + }, + { + "epoch": 0.1790001822900843, + "grad_norm": 0.36328125, + "learning_rate": 0.0018171167812613244, + "loss": 0.1201, + "step": 20621 + }, + { + "epoch": 0.17900886277028846, + "grad_norm": 0.52734375, + "learning_rate": 0.0018170987969688523, + "loss": 0.1055, + "step": 20622 + }, + { + "epoch": 0.17901754325049263, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018170808118921718, + "loss": 0.124, + "step": 20623 + }, + { + "epoch": 0.1790262237306968, + "grad_norm": 0.6328125, + "learning_rate": 0.0018170628260313025, + "loss": 0.1504, + "step": 20624 + }, + { + "epoch": 0.17903490421090096, + "grad_norm": 0.12158203125, + "learning_rate": 0.001817044839386264, + "loss": 0.1299, + "step": 20625 + }, + { + "epoch": 0.17904358469110512, + "grad_norm": 0.984375, + "learning_rate": 0.001817026851957076, + "loss": 0.1045, + "step": 20626 + }, + { + "epoch": 0.1790522651713093, + "grad_norm": 0.427734375, + "learning_rate": 0.001817008863743758, + "loss": 0.1035, + "step": 20627 + }, + { + "epoch": 0.17906094565151345, + "grad_norm": 0.66015625, + "learning_rate": 0.0018169908747463304, + "loss": 0.1504, + "step": 20628 + }, + { + "epoch": 0.17906962613171762, + "grad_norm": 0.333984375, + "learning_rate": 0.001816972884964812, + "loss": 0.0859, + "step": 20629 + }, + { + "epoch": 0.17907830661192178, + "grad_norm": 0.177734375, + "learning_rate": 0.0018169548943992228, + "loss": 0.0781, + "step": 20630 + }, + { + "epoch": 0.17908698709212595, + "grad_norm": 0.08740234375, + "learning_rate": 0.0018169369030495825, + "loss": 0.1128, + "step": 20631 + }, + { + "epoch": 0.17909566757233011, + "grad_norm": 0.6796875, + "learning_rate": 0.0018169189109159111, + "loss": 0.1211, + "step": 20632 + }, + { + "epoch": 0.17910434805253428, + "grad_norm": 0.1337890625, + "learning_rate": 0.001816900917998228, + "loss": 0.1748, + "step": 20633 + }, + { + "epoch": 0.17911302853273844, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018168829242965524, + "loss": 0.0884, + "step": 20634 + }, + { + "epoch": 0.1791217090129426, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018168649298109043, + "loss": 0.1001, + "step": 20635 + }, + { + "epoch": 0.17913038949314677, + "grad_norm": 0.7578125, + "learning_rate": 0.001816846934541304, + "loss": 0.1738, + "step": 20636 + }, + { + "epoch": 0.17913906997335094, + "grad_norm": 0.2421875, + "learning_rate": 0.00181682893848777, + "loss": 0.1299, + "step": 20637 + }, + { + "epoch": 0.17914775045355508, + "grad_norm": 0.328125, + "learning_rate": 0.0018168109416503233, + "loss": 0.1104, + "step": 20638 + }, + { + "epoch": 0.17915643093375924, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018167929440289826, + "loss": 0.1641, + "step": 20639 + }, + { + "epoch": 0.1791651114139634, + "grad_norm": 0.1884765625, + "learning_rate": 0.001816774945623768, + "loss": 0.1064, + "step": 20640 + }, + { + "epoch": 0.17917379189416757, + "grad_norm": 0.166015625, + "learning_rate": 0.0018167569464346992, + "loss": 0.1162, + "step": 20641 + }, + { + "epoch": 0.17918247237437174, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018167389464617955, + "loss": 0.1211, + "step": 20642 + }, + { + "epoch": 0.1791911528545759, + "grad_norm": 0.142578125, + "learning_rate": 0.0018167209457050771, + "loss": 0.0986, + "step": 20643 + }, + { + "epoch": 0.17919983333478007, + "grad_norm": 0.26953125, + "learning_rate": 0.0018167029441645633, + "loss": 0.123, + "step": 20644 + }, + { + "epoch": 0.17920851381498423, + "grad_norm": 0.81640625, + "learning_rate": 0.001816684941840274, + "loss": 0.0947, + "step": 20645 + }, + { + "epoch": 0.1792171942951884, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018166669387322289, + "loss": 0.1104, + "step": 20646 + }, + { + "epoch": 0.17922587477539256, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018166489348404474, + "loss": 0.0879, + "step": 20647 + }, + { + "epoch": 0.17923455525559673, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018166309301649493, + "loss": 0.0762, + "step": 20648 + }, + { + "epoch": 0.1792432357358009, + "grad_norm": 0.09375, + "learning_rate": 0.0018166129247057545, + "loss": 0.1016, + "step": 20649 + }, + { + "epoch": 0.17925191621600506, + "grad_norm": 0.10546875, + "learning_rate": 0.001816594918462883, + "loss": 0.1025, + "step": 20650 + }, + { + "epoch": 0.17926059669620922, + "grad_norm": 0.076171875, + "learning_rate": 0.0018165769114363536, + "loss": 0.1104, + "step": 20651 + }, + { + "epoch": 0.1792692771764134, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018165589036261865, + "loss": 0.1387, + "step": 20652 + }, + { + "epoch": 0.17927795765661755, + "grad_norm": 0.75, + "learning_rate": 0.0018165408950324014, + "loss": 0.1001, + "step": 20653 + }, + { + "epoch": 0.17928663813682172, + "grad_norm": 0.2578125, + "learning_rate": 0.001816522885655018, + "loss": 0.0825, + "step": 20654 + }, + { + "epoch": 0.17929531861702588, + "grad_norm": 0.1611328125, + "learning_rate": 0.0018165048754940558, + "loss": 0.1191, + "step": 20655 + }, + { + "epoch": 0.17930399909723005, + "grad_norm": 0.33203125, + "learning_rate": 0.0018164868645495346, + "loss": 0.0903, + "step": 20656 + }, + { + "epoch": 0.17931267957743421, + "grad_norm": 0.08642578125, + "learning_rate": 0.0018164688528214744, + "loss": 0.1055, + "step": 20657 + }, + { + "epoch": 0.17932136005763838, + "grad_norm": 0.5, + "learning_rate": 0.0018164508403098942, + "loss": 0.1211, + "step": 20658 + }, + { + "epoch": 0.17933004053784254, + "grad_norm": 0.59765625, + "learning_rate": 0.0018164328270148145, + "loss": 0.1074, + "step": 20659 + }, + { + "epoch": 0.1793387210180467, + "grad_norm": 0.2333984375, + "learning_rate": 0.0018164148129362547, + "loss": 0.0869, + "step": 20660 + }, + { + "epoch": 0.17934740149825087, + "grad_norm": 0.765625, + "learning_rate": 0.0018163967980742342, + "loss": 0.1338, + "step": 20661 + }, + { + "epoch": 0.17935608197845504, + "grad_norm": 0.416015625, + "learning_rate": 0.001816378782428773, + "loss": 0.1289, + "step": 20662 + }, + { + "epoch": 0.1793647624586592, + "grad_norm": 0.337890625, + "learning_rate": 0.0018163607659998906, + "loss": 0.0845, + "step": 20663 + }, + { + "epoch": 0.17937344293886337, + "grad_norm": 0.1640625, + "learning_rate": 0.0018163427487876069, + "loss": 0.1465, + "step": 20664 + }, + { + "epoch": 0.17938212341906754, + "grad_norm": 0.396484375, + "learning_rate": 0.0018163247307919415, + "loss": 0.1406, + "step": 20665 + }, + { + "epoch": 0.1793908038992717, + "grad_norm": 0.216796875, + "learning_rate": 0.0018163067120129143, + "loss": 0.1157, + "step": 20666 + }, + { + "epoch": 0.17939948437947587, + "grad_norm": 0.119140625, + "learning_rate": 0.001816288692450545, + "loss": 0.0859, + "step": 20667 + }, + { + "epoch": 0.17940816485968003, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018162706721048527, + "loss": 0.1006, + "step": 20668 + }, + { + "epoch": 0.1794168453398842, + "grad_norm": 0.3359375, + "learning_rate": 0.0018162526509758576, + "loss": 0.0918, + "step": 20669 + }, + { + "epoch": 0.17942552582008836, + "grad_norm": 0.11328125, + "learning_rate": 0.0018162346290635796, + "loss": 0.1191, + "step": 20670 + }, + { + "epoch": 0.17943420630029253, + "grad_norm": 0.3046875, + "learning_rate": 0.001816216606368038, + "loss": 0.0811, + "step": 20671 + }, + { + "epoch": 0.1794428867804967, + "grad_norm": 0.439453125, + "learning_rate": 0.0018161985828892531, + "loss": 0.1289, + "step": 20672 + }, + { + "epoch": 0.17945156726070086, + "grad_norm": 0.63671875, + "learning_rate": 0.0018161805586272438, + "loss": 0.123, + "step": 20673 + }, + { + "epoch": 0.17946024774090502, + "grad_norm": 0.2021484375, + "learning_rate": 0.00181616253358203, + "loss": 0.1064, + "step": 20674 + }, + { + "epoch": 0.1794689282211092, + "grad_norm": 0.11376953125, + "learning_rate": 0.001816144507753632, + "loss": 0.1436, + "step": 20675 + }, + { + "epoch": 0.17947760870131335, + "grad_norm": 0.279296875, + "learning_rate": 0.0018161264811420688, + "loss": 0.1328, + "step": 20676 + }, + { + "epoch": 0.17948628918151752, + "grad_norm": 0.0849609375, + "learning_rate": 0.001816108453747361, + "loss": 0.0879, + "step": 20677 + }, + { + "epoch": 0.17949496966172168, + "grad_norm": 0.1962890625, + "learning_rate": 0.001816090425569527, + "loss": 0.1133, + "step": 20678 + }, + { + "epoch": 0.17950365014192585, + "grad_norm": 0.06298828125, + "learning_rate": 0.0018160723966085877, + "loss": 0.0767, + "step": 20679 + }, + { + "epoch": 0.17951233062213, + "grad_norm": 0.71484375, + "learning_rate": 0.0018160543668645623, + "loss": 0.1211, + "step": 20680 + }, + { + "epoch": 0.17952101110233418, + "grad_norm": 0.486328125, + "learning_rate": 0.0018160363363374707, + "loss": 0.0928, + "step": 20681 + }, + { + "epoch": 0.17952969158253834, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018160183050273324, + "loss": 0.0928, + "step": 20682 + }, + { + "epoch": 0.1795383720627425, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018160002729341674, + "loss": 0.1406, + "step": 20683 + }, + { + "epoch": 0.17954705254294667, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018159822400579952, + "loss": 0.0898, + "step": 20684 + }, + { + "epoch": 0.17955573302315084, + "grad_norm": 0.20703125, + "learning_rate": 0.0018159642063988355, + "loss": 0.0928, + "step": 20685 + }, + { + "epoch": 0.179564413503355, + "grad_norm": 0.125, + "learning_rate": 0.0018159461719567081, + "loss": 0.127, + "step": 20686 + }, + { + "epoch": 0.17957309398355917, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018159281367316326, + "loss": 0.0986, + "step": 20687 + }, + { + "epoch": 0.17958177446376333, + "grad_norm": 0.1337890625, + "learning_rate": 0.0018159101007236293, + "loss": 0.1074, + "step": 20688 + }, + { + "epoch": 0.1795904549439675, + "grad_norm": 0.3984375, + "learning_rate": 0.0018158920639327171, + "loss": 0.166, + "step": 20689 + }, + { + "epoch": 0.17959913542417166, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018158740263589162, + "loss": 0.1035, + "step": 20690 + }, + { + "epoch": 0.17960781590437583, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018158559880022463, + "loss": 0.0938, + "step": 20691 + }, + { + "epoch": 0.17961649638458, + "grad_norm": 0.1142578125, + "learning_rate": 0.001815837948862727, + "loss": 0.1562, + "step": 20692 + }, + { + "epoch": 0.17962517686478416, + "grad_norm": 0.08154296875, + "learning_rate": 0.0018158199089403778, + "loss": 0.1201, + "step": 20693 + }, + { + "epoch": 0.17963385734498832, + "grad_norm": 0.1201171875, + "learning_rate": 0.001815801868235219, + "loss": 0.123, + "step": 20694 + }, + { + "epoch": 0.1796425378251925, + "grad_norm": 0.251953125, + "learning_rate": 0.00181578382674727, + "loss": 0.1377, + "step": 20695 + }, + { + "epoch": 0.17965121830539665, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018157657844765508, + "loss": 0.1152, + "step": 20696 + }, + { + "epoch": 0.17965989878560082, + "grad_norm": 0.64453125, + "learning_rate": 0.0018157477414230802, + "loss": 0.1562, + "step": 20697 + }, + { + "epoch": 0.17966857926580498, + "grad_norm": 0.46875, + "learning_rate": 0.0018157296975868792, + "loss": 0.1328, + "step": 20698 + }, + { + "epoch": 0.17967725974600915, + "grad_norm": 0.3046875, + "learning_rate": 0.0018157116529679666, + "loss": 0.1074, + "step": 20699 + }, + { + "epoch": 0.17968594022621331, + "grad_norm": 0.2890625, + "learning_rate": 0.0018156936075663628, + "loss": 0.1152, + "step": 20700 + }, + { + "epoch": 0.17969462070641748, + "grad_norm": 0.5546875, + "learning_rate": 0.0018156755613820872, + "loss": 0.1699, + "step": 20701 + }, + { + "epoch": 0.17970330118662164, + "grad_norm": 0.388671875, + "learning_rate": 0.0018156575144151594, + "loss": 0.1104, + "step": 20702 + }, + { + "epoch": 0.1797119816668258, + "grad_norm": 0.71484375, + "learning_rate": 0.0018156394666655996, + "loss": 0.1064, + "step": 20703 + }, + { + "epoch": 0.17972066214702997, + "grad_norm": 0.330078125, + "learning_rate": 0.0018156214181334267, + "loss": 0.0952, + "step": 20704 + }, + { + "epoch": 0.17972934262723414, + "grad_norm": 0.78515625, + "learning_rate": 0.0018156033688186614, + "loss": 0.1465, + "step": 20705 + }, + { + "epoch": 0.1797380231074383, + "grad_norm": 0.1845703125, + "learning_rate": 0.001815585318721323, + "loss": 0.104, + "step": 20706 + }, + { + "epoch": 0.17974670358764247, + "grad_norm": 0.32421875, + "learning_rate": 0.001815567267841431, + "loss": 0.1309, + "step": 20707 + }, + { + "epoch": 0.17975538406784664, + "grad_norm": 0.1552734375, + "learning_rate": 0.0018155492161790055, + "loss": 0.1484, + "step": 20708 + }, + { + "epoch": 0.1797640645480508, + "grad_norm": 0.29296875, + "learning_rate": 0.001815531163734066, + "loss": 0.1328, + "step": 20709 + }, + { + "epoch": 0.17977274502825497, + "grad_norm": 0.10107421875, + "learning_rate": 0.0018155131105066324, + "loss": 0.0986, + "step": 20710 + }, + { + "epoch": 0.17978142550845913, + "grad_norm": 0.40234375, + "learning_rate": 0.0018154950564967247, + "loss": 0.1289, + "step": 20711 + }, + { + "epoch": 0.1797901059886633, + "grad_norm": 1.0390625, + "learning_rate": 0.001815477001704362, + "loss": 0.1147, + "step": 20712 + }, + { + "epoch": 0.17979878646886746, + "grad_norm": 0.091796875, + "learning_rate": 0.0018154589461295647, + "loss": 0.1172, + "step": 20713 + }, + { + "epoch": 0.17980746694907163, + "grad_norm": 0.283203125, + "learning_rate": 0.0018154408897723521, + "loss": 0.123, + "step": 20714 + }, + { + "epoch": 0.1798161474292758, + "grad_norm": 0.41015625, + "learning_rate": 0.0018154228326327441, + "loss": 0.1001, + "step": 20715 + }, + { + "epoch": 0.17982482790947996, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018154047747107604, + "loss": 0.0811, + "step": 20716 + }, + { + "epoch": 0.17983350838968412, + "grad_norm": 0.154296875, + "learning_rate": 0.001815386716006421, + "loss": 0.1143, + "step": 20717 + }, + { + "epoch": 0.1798421888698883, + "grad_norm": 0.384765625, + "learning_rate": 0.0018153686565197454, + "loss": 0.127, + "step": 20718 + }, + { + "epoch": 0.17985086935009245, + "grad_norm": 0.443359375, + "learning_rate": 0.001815350596250753, + "loss": 0.1035, + "step": 20719 + }, + { + "epoch": 0.17985954983029662, + "grad_norm": 0.33984375, + "learning_rate": 0.0018153325351994644, + "loss": 0.0977, + "step": 20720 + }, + { + "epoch": 0.17986823031050078, + "grad_norm": 0.2578125, + "learning_rate": 0.0018153144733658983, + "loss": 0.123, + "step": 20721 + }, + { + "epoch": 0.17987691079070495, + "grad_norm": 0.26953125, + "learning_rate": 0.0018152964107500758, + "loss": 0.0845, + "step": 20722 + }, + { + "epoch": 0.1798855912709091, + "grad_norm": 0.12890625, + "learning_rate": 0.0018152783473520155, + "loss": 0.1523, + "step": 20723 + }, + { + "epoch": 0.17989427175111328, + "grad_norm": 0.2001953125, + "learning_rate": 0.0018152602831717375, + "loss": 0.1006, + "step": 20724 + }, + { + "epoch": 0.17990295223131744, + "grad_norm": 0.1484375, + "learning_rate": 0.0018152422182092618, + "loss": 0.1108, + "step": 20725 + }, + { + "epoch": 0.1799116327115216, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018152241524646076, + "loss": 0.1191, + "step": 20726 + }, + { + "epoch": 0.17992031319172577, + "grad_norm": 0.15625, + "learning_rate": 0.0018152060859377953, + "loss": 0.1094, + "step": 20727 + }, + { + "epoch": 0.17992899367192994, + "grad_norm": 0.107421875, + "learning_rate": 0.0018151880186288443, + "loss": 0.0996, + "step": 20728 + }, + { + "epoch": 0.1799376741521341, + "grad_norm": 0.259765625, + "learning_rate": 0.0018151699505377743, + "loss": 0.1016, + "step": 20729 + }, + { + "epoch": 0.17994635463233827, + "grad_norm": 0.119140625, + "learning_rate": 0.0018151518816646053, + "loss": 0.1719, + "step": 20730 + }, + { + "epoch": 0.17995503511254243, + "grad_norm": 0.53515625, + "learning_rate": 0.001815133812009357, + "loss": 0.0874, + "step": 20731 + }, + { + "epoch": 0.1799637155927466, + "grad_norm": 0.51171875, + "learning_rate": 0.0018151157415720492, + "loss": 0.1572, + "step": 20732 + }, + { + "epoch": 0.17997239607295076, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018150976703527016, + "loss": 0.1016, + "step": 20733 + }, + { + "epoch": 0.17998107655315493, + "grad_norm": 0.302734375, + "learning_rate": 0.0018150795983513338, + "loss": 0.1143, + "step": 20734 + }, + { + "epoch": 0.1799897570333591, + "grad_norm": 0.111328125, + "learning_rate": 0.0018150615255679654, + "loss": 0.1523, + "step": 20735 + }, + { + "epoch": 0.17999843751356326, + "grad_norm": 0.302734375, + "learning_rate": 0.0018150434520026167, + "loss": 0.1001, + "step": 20736 + }, + { + "epoch": 0.18000711799376742, + "grad_norm": 0.35546875, + "learning_rate": 0.0018150253776553073, + "loss": 0.1309, + "step": 20737 + }, + { + "epoch": 0.1800157984739716, + "grad_norm": 0.388671875, + "learning_rate": 0.001815007302526057, + "loss": 0.0908, + "step": 20738 + }, + { + "epoch": 0.18002447895417575, + "grad_norm": 0.1259765625, + "learning_rate": 0.001814989226614885, + "loss": 0.1572, + "step": 20739 + }, + { + "epoch": 0.18003315943437992, + "grad_norm": 0.54296875, + "learning_rate": 0.0018149711499218118, + "loss": 0.1235, + "step": 20740 + }, + { + "epoch": 0.18004183991458408, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018149530724468571, + "loss": 0.1543, + "step": 20741 + }, + { + "epoch": 0.18005052039478825, + "grad_norm": 0.330078125, + "learning_rate": 0.0018149349941900402, + "loss": 0.1084, + "step": 20742 + }, + { + "epoch": 0.18005920087499241, + "grad_norm": 0.37109375, + "learning_rate": 0.001814916915151381, + "loss": 0.1348, + "step": 20743 + }, + { + "epoch": 0.18006788135519658, + "grad_norm": 0.275390625, + "learning_rate": 0.0018148988353309003, + "loss": 0.0596, + "step": 20744 + }, + { + "epoch": 0.18007656183540074, + "grad_norm": 0.115234375, + "learning_rate": 0.0018148807547286158, + "loss": 0.1143, + "step": 20745 + }, + { + "epoch": 0.1800852423156049, + "grad_norm": 0.357421875, + "learning_rate": 0.001814862673344549, + "loss": 0.1235, + "step": 20746 + }, + { + "epoch": 0.18009392279580907, + "grad_norm": 0.3359375, + "learning_rate": 0.001814844591178719, + "loss": 0.1089, + "step": 20747 + }, + { + "epoch": 0.18010260327601324, + "grad_norm": 0.376953125, + "learning_rate": 0.0018148265082311458, + "loss": 0.082, + "step": 20748 + }, + { + "epoch": 0.1801112837562174, + "grad_norm": 0.546875, + "learning_rate": 0.001814808424501849, + "loss": 0.0938, + "step": 20749 + }, + { + "epoch": 0.18011996423642157, + "grad_norm": 0.08544921875, + "learning_rate": 0.0018147903399908486, + "loss": 0.0894, + "step": 20750 + }, + { + "epoch": 0.18012864471662574, + "grad_norm": 0.4921875, + "learning_rate": 0.001814772254698164, + "loss": 0.0981, + "step": 20751 + }, + { + "epoch": 0.1801373251968299, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018147541686238152, + "loss": 0.1006, + "step": 20752 + }, + { + "epoch": 0.18014600567703407, + "grad_norm": 0.171875, + "learning_rate": 0.0018147360817678223, + "loss": 0.0908, + "step": 20753 + }, + { + "epoch": 0.18015468615723823, + "grad_norm": 0.10693359375, + "learning_rate": 0.0018147179941302046, + "loss": 0.1108, + "step": 20754 + }, + { + "epoch": 0.1801633666374424, + "grad_norm": 0.890625, + "learning_rate": 0.0018146999057109816, + "loss": 0.126, + "step": 20755 + }, + { + "epoch": 0.18017204711764656, + "grad_norm": 0.11181640625, + "learning_rate": 0.0018146818165101742, + "loss": 0.1191, + "step": 20756 + }, + { + "epoch": 0.18018072759785073, + "grad_norm": 0.2373046875, + "learning_rate": 0.0018146637265278012, + "loss": 0.1025, + "step": 20757 + }, + { + "epoch": 0.1801894080780549, + "grad_norm": 0.400390625, + "learning_rate": 0.0018146456357638827, + "loss": 0.1758, + "step": 20758 + }, + { + "epoch": 0.18019808855825906, + "grad_norm": 0.53125, + "learning_rate": 0.0018146275442184386, + "loss": 0.1445, + "step": 20759 + }, + { + "epoch": 0.18020676903846322, + "grad_norm": 0.484375, + "learning_rate": 0.0018146094518914882, + "loss": 0.1367, + "step": 20760 + }, + { + "epoch": 0.18021544951866736, + "grad_norm": 0.271484375, + "learning_rate": 0.0018145913587830518, + "loss": 0.0962, + "step": 20761 + }, + { + "epoch": 0.18022412999887152, + "grad_norm": 0.158203125, + "learning_rate": 0.001814573264893149, + "loss": 0.1484, + "step": 20762 + }, + { + "epoch": 0.1802328104790757, + "grad_norm": 0.2294921875, + "learning_rate": 0.0018145551702217998, + "loss": 0.1035, + "step": 20763 + }, + { + "epoch": 0.18024149095927985, + "grad_norm": 0.1865234375, + "learning_rate": 0.0018145370747690238, + "loss": 0.1357, + "step": 20764 + }, + { + "epoch": 0.18025017143948402, + "grad_norm": 0.52734375, + "learning_rate": 0.0018145189785348406, + "loss": 0.083, + "step": 20765 + }, + { + "epoch": 0.18025885191968818, + "grad_norm": 0.80859375, + "learning_rate": 0.0018145008815192703, + "loss": 0.0928, + "step": 20766 + }, + { + "epoch": 0.18026753239989235, + "grad_norm": 0.2216796875, + "learning_rate": 0.0018144827837223325, + "loss": 0.0957, + "step": 20767 + }, + { + "epoch": 0.18027621288009651, + "grad_norm": 0.3984375, + "learning_rate": 0.0018144646851440472, + "loss": 0.0908, + "step": 20768 + }, + { + "epoch": 0.18028489336030068, + "grad_norm": 0.283203125, + "learning_rate": 0.0018144465857844341, + "loss": 0.0864, + "step": 20769 + }, + { + "epoch": 0.18029357384050484, + "grad_norm": 0.455078125, + "learning_rate": 0.0018144284856435125, + "loss": 0.082, + "step": 20770 + }, + { + "epoch": 0.180302254320709, + "grad_norm": 0.1591796875, + "learning_rate": 0.001814410384721303, + "loss": 0.1279, + "step": 20771 + }, + { + "epoch": 0.18031093480091318, + "grad_norm": 0.25390625, + "learning_rate": 0.001814392283017825, + "loss": 0.0996, + "step": 20772 + }, + { + "epoch": 0.18031961528111734, + "grad_norm": 0.328125, + "learning_rate": 0.0018143741805330981, + "loss": 0.1162, + "step": 20773 + }, + { + "epoch": 0.1803282957613215, + "grad_norm": 0.07080078125, + "learning_rate": 0.0018143560772671428, + "loss": 0.1084, + "step": 20774 + }, + { + "epoch": 0.18033697624152567, + "grad_norm": 0.10107421875, + "learning_rate": 0.001814337973219978, + "loss": 0.1191, + "step": 20775 + }, + { + "epoch": 0.18034565672172984, + "grad_norm": 0.41796875, + "learning_rate": 0.0018143198683916242, + "loss": 0.1182, + "step": 20776 + }, + { + "epoch": 0.180354337201934, + "grad_norm": 0.41796875, + "learning_rate": 0.0018143017627821009, + "loss": 0.1133, + "step": 20777 + }, + { + "epoch": 0.18036301768213817, + "grad_norm": 0.224609375, + "learning_rate": 0.001814283656391428, + "loss": 0.1318, + "step": 20778 + }, + { + "epoch": 0.18037169816234233, + "grad_norm": 0.322265625, + "learning_rate": 0.0018142655492196249, + "loss": 0.1152, + "step": 20779 + }, + { + "epoch": 0.1803803786425465, + "grad_norm": 0.279296875, + "learning_rate": 0.001814247441266712, + "loss": 0.127, + "step": 20780 + }, + { + "epoch": 0.18038905912275066, + "grad_norm": 0.89453125, + "learning_rate": 0.0018142293325327085, + "loss": 0.1543, + "step": 20781 + }, + { + "epoch": 0.18039773960295483, + "grad_norm": 0.2734375, + "learning_rate": 0.0018142112230176348, + "loss": 0.1064, + "step": 20782 + }, + { + "epoch": 0.180406420083159, + "grad_norm": 0.318359375, + "learning_rate": 0.0018141931127215105, + "loss": 0.1387, + "step": 20783 + }, + { + "epoch": 0.18041510056336316, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018141750016443551, + "loss": 0.0952, + "step": 20784 + }, + { + "epoch": 0.18042378104356732, + "grad_norm": 0.1162109375, + "learning_rate": 0.001814156889786189, + "loss": 0.1016, + "step": 20785 + }, + { + "epoch": 0.1804324615237715, + "grad_norm": 0.4453125, + "learning_rate": 0.0018141387771470315, + "loss": 0.1074, + "step": 20786 + }, + { + "epoch": 0.18044114200397565, + "grad_norm": 0.470703125, + "learning_rate": 0.0018141206637269022, + "loss": 0.1377, + "step": 20787 + }, + { + "epoch": 0.18044982248417982, + "grad_norm": 0.5859375, + "learning_rate": 0.001814102549525822, + "loss": 0.0786, + "step": 20788 + }, + { + "epoch": 0.18045850296438398, + "grad_norm": 0.20703125, + "learning_rate": 0.0018140844345438092, + "loss": 0.1699, + "step": 20789 + }, + { + "epoch": 0.18046718344458815, + "grad_norm": 0.298828125, + "learning_rate": 0.0018140663187808847, + "loss": 0.1377, + "step": 20790 + }, + { + "epoch": 0.1804758639247923, + "grad_norm": 0.10205078125, + "learning_rate": 0.0018140482022370684, + "loss": 0.1064, + "step": 20791 + }, + { + "epoch": 0.18048454440499648, + "grad_norm": 0.8828125, + "learning_rate": 0.001814030084912379, + "loss": 0.1221, + "step": 20792 + }, + { + "epoch": 0.18049322488520064, + "grad_norm": 0.294921875, + "learning_rate": 0.0018140119668068376, + "loss": 0.1001, + "step": 20793 + }, + { + "epoch": 0.1805019053654048, + "grad_norm": 0.306640625, + "learning_rate": 0.001813993847920463, + "loss": 0.1133, + "step": 20794 + }, + { + "epoch": 0.18051058584560897, + "grad_norm": 0.3671875, + "learning_rate": 0.001813975728253276, + "loss": 0.1221, + "step": 20795 + }, + { + "epoch": 0.18051926632581314, + "grad_norm": 0.18359375, + "learning_rate": 0.0018139576078052954, + "loss": 0.1191, + "step": 20796 + }, + { + "epoch": 0.1805279468060173, + "grad_norm": 0.71875, + "learning_rate": 0.0018139394865765417, + "loss": 0.0986, + "step": 20797 + }, + { + "epoch": 0.18053662728622147, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018139213645670346, + "loss": 0.0947, + "step": 20798 + }, + { + "epoch": 0.18054530776642563, + "grad_norm": 0.1630859375, + "learning_rate": 0.0018139032417767939, + "loss": 0.0938, + "step": 20799 + }, + { + "epoch": 0.1805539882466298, + "grad_norm": 0.302734375, + "learning_rate": 0.0018138851182058389, + "loss": 0.1221, + "step": 20800 + }, + { + "epoch": 0.18056266872683396, + "grad_norm": 0.18359375, + "learning_rate": 0.0018138669938541903, + "loss": 0.0986, + "step": 20801 + }, + { + "epoch": 0.18057134920703813, + "grad_norm": 0.1689453125, + "learning_rate": 0.0018138488687218673, + "loss": 0.1211, + "step": 20802 + }, + { + "epoch": 0.1805800296872423, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018138307428088897, + "loss": 0.1221, + "step": 20803 + }, + { + "epoch": 0.18058871016744646, + "grad_norm": 0.1015625, + "learning_rate": 0.0018138126161152777, + "loss": 0.1377, + "step": 20804 + }, + { + "epoch": 0.18059739064765062, + "grad_norm": 0.2470703125, + "learning_rate": 0.0018137944886410509, + "loss": 0.1533, + "step": 20805 + }, + { + "epoch": 0.1806060711278548, + "grad_norm": 0.52734375, + "learning_rate": 0.0018137763603862291, + "loss": 0.1118, + "step": 20806 + }, + { + "epoch": 0.18061475160805895, + "grad_norm": 0.73046875, + "learning_rate": 0.0018137582313508323, + "loss": 0.085, + "step": 20807 + }, + { + "epoch": 0.18062343208826312, + "grad_norm": 0.33984375, + "learning_rate": 0.0018137401015348802, + "loss": 0.1021, + "step": 20808 + }, + { + "epoch": 0.18063211256846728, + "grad_norm": 0.162109375, + "learning_rate": 0.0018137219709383927, + "loss": 0.0776, + "step": 20809 + }, + { + "epoch": 0.18064079304867145, + "grad_norm": 0.08935546875, + "learning_rate": 0.0018137038395613897, + "loss": 0.1221, + "step": 20810 + }, + { + "epoch": 0.18064947352887561, + "grad_norm": 0.4140625, + "learning_rate": 0.0018136857074038909, + "loss": 0.0854, + "step": 20811 + }, + { + "epoch": 0.18065815400907978, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018136675744659157, + "loss": 0.1143, + "step": 20812 + }, + { + "epoch": 0.18066683448928395, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018136494407474848, + "loss": 0.1406, + "step": 20813 + }, + { + "epoch": 0.1806755149694881, + "grad_norm": 0.37890625, + "learning_rate": 0.0018136313062486174, + "loss": 0.1177, + "step": 20814 + }, + { + "epoch": 0.18068419544969228, + "grad_norm": 0.4765625, + "learning_rate": 0.0018136131709693335, + "loss": 0.0972, + "step": 20815 + }, + { + "epoch": 0.18069287592989644, + "grad_norm": 0.5, + "learning_rate": 0.001813595034909653, + "loss": 0.084, + "step": 20816 + }, + { + "epoch": 0.1807015564101006, + "grad_norm": 0.0712890625, + "learning_rate": 0.0018135768980695962, + "loss": 0.0859, + "step": 20817 + }, + { + "epoch": 0.18071023689030477, + "grad_norm": 0.75390625, + "learning_rate": 0.0018135587604491818, + "loss": 0.1611, + "step": 20818 + }, + { + "epoch": 0.18071891737050894, + "grad_norm": 0.5546875, + "learning_rate": 0.0018135406220484303, + "loss": 0.1289, + "step": 20819 + }, + { + "epoch": 0.1807275978507131, + "grad_norm": 0.48828125, + "learning_rate": 0.0018135224828673615, + "loss": 0.1572, + "step": 20820 + }, + { + "epoch": 0.18073627833091727, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018135043429059951, + "loss": 0.0889, + "step": 20821 + }, + { + "epoch": 0.18074495881112143, + "grad_norm": 0.5, + "learning_rate": 0.0018134862021643517, + "loss": 0.103, + "step": 20822 + }, + { + "epoch": 0.1807536392913256, + "grad_norm": 0.87109375, + "learning_rate": 0.00181346806064245, + "loss": 0.1113, + "step": 20823 + }, + { + "epoch": 0.18076231977152976, + "grad_norm": 0.310546875, + "learning_rate": 0.0018134499183403104, + "loss": 0.168, + "step": 20824 + }, + { + "epoch": 0.18077100025173393, + "grad_norm": 0.3671875, + "learning_rate": 0.0018134317752579528, + "loss": 0.1079, + "step": 20825 + }, + { + "epoch": 0.1807796807319381, + "grad_norm": 0.302734375, + "learning_rate": 0.001813413631395397, + "loss": 0.0986, + "step": 20826 + }, + { + "epoch": 0.18078836121214226, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018133954867526627, + "loss": 0.1055, + "step": 20827 + }, + { + "epoch": 0.18079704169234642, + "grad_norm": 0.189453125, + "learning_rate": 0.0018133773413297697, + "loss": 0.1328, + "step": 20828 + }, + { + "epoch": 0.1808057221725506, + "grad_norm": 0.5234375, + "learning_rate": 0.001813359195126738, + "loss": 0.1099, + "step": 20829 + }, + { + "epoch": 0.18081440265275475, + "grad_norm": 0.4375, + "learning_rate": 0.0018133410481435874, + "loss": 0.1377, + "step": 20830 + }, + { + "epoch": 0.18082308313295892, + "grad_norm": 0.97265625, + "learning_rate": 0.001813322900380338, + "loss": 0.126, + "step": 20831 + }, + { + "epoch": 0.18083176361316308, + "grad_norm": 0.119140625, + "learning_rate": 0.0018133047518370088, + "loss": 0.1348, + "step": 20832 + }, + { + "epoch": 0.18084044409336725, + "grad_norm": 0.07958984375, + "learning_rate": 0.0018132866025136207, + "loss": 0.0864, + "step": 20833 + }, + { + "epoch": 0.1808491245735714, + "grad_norm": 0.173828125, + "learning_rate": 0.001813268452410193, + "loss": 0.0923, + "step": 20834 + }, + { + "epoch": 0.18085780505377558, + "grad_norm": 0.76953125, + "learning_rate": 0.0018132503015267458, + "loss": 0.126, + "step": 20835 + }, + { + "epoch": 0.18086648553397974, + "grad_norm": 0.341796875, + "learning_rate": 0.0018132321498632985, + "loss": 0.0581, + "step": 20836 + }, + { + "epoch": 0.1808751660141839, + "grad_norm": 0.1064453125, + "learning_rate": 0.0018132139974198717, + "loss": 0.0918, + "step": 20837 + }, + { + "epoch": 0.18088384649438807, + "grad_norm": 0.259765625, + "learning_rate": 0.0018131958441964845, + "loss": 0.1094, + "step": 20838 + }, + { + "epoch": 0.18089252697459224, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018131776901931568, + "loss": 0.0977, + "step": 20839 + }, + { + "epoch": 0.1809012074547964, + "grad_norm": 0.2158203125, + "learning_rate": 0.001813159535409909, + "loss": 0.1128, + "step": 20840 + }, + { + "epoch": 0.18090988793500057, + "grad_norm": 0.396484375, + "learning_rate": 0.0018131413798467607, + "loss": 0.1523, + "step": 20841 + }, + { + "epoch": 0.18091856841520473, + "grad_norm": 0.73046875, + "learning_rate": 0.0018131232235037312, + "loss": 0.1641, + "step": 20842 + }, + { + "epoch": 0.1809272488954089, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018131050663808413, + "loss": 0.1016, + "step": 20843 + }, + { + "epoch": 0.18093592937561306, + "grad_norm": 0.4453125, + "learning_rate": 0.0018130869084781102, + "loss": 0.127, + "step": 20844 + }, + { + "epoch": 0.18094460985581723, + "grad_norm": 0.11181640625, + "learning_rate": 0.001813068749795558, + "loss": 0.1006, + "step": 20845 + }, + { + "epoch": 0.1809532903360214, + "grad_norm": 0.6796875, + "learning_rate": 0.0018130505903332047, + "loss": 0.1621, + "step": 20846 + }, + { + "epoch": 0.18096197081622556, + "grad_norm": 0.1025390625, + "learning_rate": 0.00181303243009107, + "loss": 0.1201, + "step": 20847 + }, + { + "epoch": 0.18097065129642972, + "grad_norm": 0.138671875, + "learning_rate": 0.0018130142690691736, + "loss": 0.1113, + "step": 20848 + }, + { + "epoch": 0.1809793317766339, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018129961072675357, + "loss": 0.104, + "step": 20849 + }, + { + "epoch": 0.18098801225683805, + "grad_norm": 0.2734375, + "learning_rate": 0.0018129779446861755, + "loss": 0.0957, + "step": 20850 + }, + { + "epoch": 0.18099669273704222, + "grad_norm": 0.09765625, + "learning_rate": 0.001812959781325114, + "loss": 0.1543, + "step": 20851 + }, + { + "epoch": 0.18100537321724638, + "grad_norm": 0.3984375, + "learning_rate": 0.00181294161718437, + "loss": 0.084, + "step": 20852 + }, + { + "epoch": 0.18101405369745055, + "grad_norm": 0.3515625, + "learning_rate": 0.0018129234522639636, + "loss": 0.0942, + "step": 20853 + }, + { + "epoch": 0.18102273417765471, + "grad_norm": 0.2451171875, + "learning_rate": 0.001812905286563915, + "loss": 0.0864, + "step": 20854 + }, + { + "epoch": 0.18103141465785888, + "grad_norm": 0.8515625, + "learning_rate": 0.0018128871200842442, + "loss": 0.1143, + "step": 20855 + }, + { + "epoch": 0.18104009513806305, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018128689528249704, + "loss": 0.0991, + "step": 20856 + }, + { + "epoch": 0.1810487756182672, + "grad_norm": 0.48046875, + "learning_rate": 0.0018128507847861139, + "loss": 0.1104, + "step": 20857 + }, + { + "epoch": 0.18105745609847138, + "grad_norm": 0.30078125, + "learning_rate": 0.0018128326159676945, + "loss": 0.1523, + "step": 20858 + }, + { + "epoch": 0.18106613657867554, + "grad_norm": 0.1005859375, + "learning_rate": 0.001812814446369732, + "loss": 0.1514, + "step": 20859 + }, + { + "epoch": 0.1810748170588797, + "grad_norm": 1.4375, + "learning_rate": 0.0018127962759922464, + "loss": 0.1572, + "step": 20860 + }, + { + "epoch": 0.18108349753908387, + "grad_norm": 0.400390625, + "learning_rate": 0.0018127781048352576, + "loss": 0.1211, + "step": 20861 + }, + { + "epoch": 0.18109217801928804, + "grad_norm": 0.271484375, + "learning_rate": 0.001812759932898785, + "loss": 0.1299, + "step": 20862 + }, + { + "epoch": 0.1811008584994922, + "grad_norm": 0.08837890625, + "learning_rate": 0.0018127417601828493, + "loss": 0.1367, + "step": 20863 + }, + { + "epoch": 0.18110953897969637, + "grad_norm": 0.58203125, + "learning_rate": 0.00181272358668747, + "loss": 0.0996, + "step": 20864 + }, + { + "epoch": 0.18111821945990053, + "grad_norm": 0.1875, + "learning_rate": 0.0018127054124126665, + "loss": 0.0889, + "step": 20865 + }, + { + "epoch": 0.1811268999401047, + "grad_norm": 0.06884765625, + "learning_rate": 0.0018126872373584594, + "loss": 0.0967, + "step": 20866 + }, + { + "epoch": 0.18113558042030886, + "grad_norm": 0.1875, + "learning_rate": 0.0018126690615248677, + "loss": 0.1172, + "step": 20867 + }, + { + "epoch": 0.18114426090051303, + "grad_norm": 0.255859375, + "learning_rate": 0.0018126508849119123, + "loss": 0.1182, + "step": 20868 + }, + { + "epoch": 0.1811529413807172, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018126327075196126, + "loss": 0.105, + "step": 20869 + }, + { + "epoch": 0.18116162186092136, + "grad_norm": 0.87890625, + "learning_rate": 0.0018126145293479881, + "loss": 0.1357, + "step": 20870 + }, + { + "epoch": 0.18117030234112552, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018125963503970593, + "loss": 0.127, + "step": 20871 + }, + { + "epoch": 0.1811789828213297, + "grad_norm": 0.484375, + "learning_rate": 0.001812578170666846, + "loss": 0.0923, + "step": 20872 + }, + { + "epoch": 0.18118766330153385, + "grad_norm": 0.10791015625, + "learning_rate": 0.0018125599901573676, + "loss": 0.1216, + "step": 20873 + }, + { + "epoch": 0.18119634378173802, + "grad_norm": 0.494140625, + "learning_rate": 0.0018125418088686446, + "loss": 0.166, + "step": 20874 + }, + { + "epoch": 0.18120502426194218, + "grad_norm": 0.478515625, + "learning_rate": 0.0018125236268006961, + "loss": 0.1055, + "step": 20875 + }, + { + "epoch": 0.18121370474214635, + "grad_norm": 0.1748046875, + "learning_rate": 0.001812505443953543, + "loss": 0.123, + "step": 20876 + }, + { + "epoch": 0.1812223852223505, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018124872603272045, + "loss": 0.1279, + "step": 20877 + }, + { + "epoch": 0.18123106570255468, + "grad_norm": 0.16796875, + "learning_rate": 0.0018124690759217007, + "loss": 0.1182, + "step": 20878 + }, + { + "epoch": 0.18123974618275884, + "grad_norm": 0.69921875, + "learning_rate": 0.0018124508907370513, + "loss": 0.1025, + "step": 20879 + }, + { + "epoch": 0.181248426662963, + "grad_norm": 0.1875, + "learning_rate": 0.0018124327047732765, + "loss": 0.126, + "step": 20880 + }, + { + "epoch": 0.18125710714316717, + "grad_norm": 0.73828125, + "learning_rate": 0.0018124145180303955, + "loss": 0.1104, + "step": 20881 + }, + { + "epoch": 0.18126578762337134, + "grad_norm": 0.1435546875, + "learning_rate": 0.001812396330508429, + "loss": 0.0977, + "step": 20882 + }, + { + "epoch": 0.1812744681035755, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018123781422073968, + "loss": 0.2061, + "step": 20883 + }, + { + "epoch": 0.18128314858377967, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018123599531273185, + "loss": 0.1475, + "step": 20884 + }, + { + "epoch": 0.1812918290639838, + "grad_norm": 0.1953125, + "learning_rate": 0.0018123417632682137, + "loss": 0.0967, + "step": 20885 + }, + { + "epoch": 0.18130050954418797, + "grad_norm": 0.29296875, + "learning_rate": 0.0018123235726301029, + "loss": 0.0996, + "step": 20886 + }, + { + "epoch": 0.18130919002439214, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018123053812130058, + "loss": 0.126, + "step": 20887 + }, + { + "epoch": 0.1813178705045963, + "grad_norm": 0.375, + "learning_rate": 0.0018122871890169422, + "loss": 0.0947, + "step": 20888 + }, + { + "epoch": 0.18132655098480047, + "grad_norm": 0.75390625, + "learning_rate": 0.0018122689960419316, + "loss": 0.1064, + "step": 20889 + }, + { + "epoch": 0.18133523146500463, + "grad_norm": 0.4921875, + "learning_rate": 0.0018122508022879948, + "loss": 0.1128, + "step": 20890 + }, + { + "epoch": 0.1813439119452088, + "grad_norm": 0.10205078125, + "learning_rate": 0.001812232607755151, + "loss": 0.0889, + "step": 20891 + }, + { + "epoch": 0.18135259242541296, + "grad_norm": 0.13671875, + "learning_rate": 0.0018122144124434205, + "loss": 0.1113, + "step": 20892 + }, + { + "epoch": 0.18136127290561713, + "grad_norm": 0.0810546875, + "learning_rate": 0.001812196216352823, + "loss": 0.0879, + "step": 20893 + }, + { + "epoch": 0.1813699533858213, + "grad_norm": 0.333984375, + "learning_rate": 0.001812178019483379, + "loss": 0.0928, + "step": 20894 + }, + { + "epoch": 0.18137863386602546, + "grad_norm": 0.703125, + "learning_rate": 0.0018121598218351067, + "loss": 0.0898, + "step": 20895 + }, + { + "epoch": 0.18138731434622962, + "grad_norm": 0.5390625, + "learning_rate": 0.0018121416234080276, + "loss": 0.1504, + "step": 20896 + }, + { + "epoch": 0.1813959948264338, + "grad_norm": 0.1650390625, + "learning_rate": 0.0018121234242021611, + "loss": 0.0991, + "step": 20897 + }, + { + "epoch": 0.18140467530663795, + "grad_norm": 0.474609375, + "learning_rate": 0.0018121052242175276, + "loss": 0.1099, + "step": 20898 + }, + { + "epoch": 0.18141335578684212, + "grad_norm": 0.2314453125, + "learning_rate": 0.0018120870234541457, + "loss": 0.1162, + "step": 20899 + }, + { + "epoch": 0.18142203626704628, + "grad_norm": 0.58203125, + "learning_rate": 0.0018120688219120367, + "loss": 0.123, + "step": 20900 + }, + { + "epoch": 0.18143071674725045, + "grad_norm": 0.060302734375, + "learning_rate": 0.0018120506195912196, + "loss": 0.0869, + "step": 20901 + }, + { + "epoch": 0.1814393972274546, + "grad_norm": 0.11328125, + "learning_rate": 0.0018120324164917153, + "loss": 0.1064, + "step": 20902 + }, + { + "epoch": 0.18144807770765878, + "grad_norm": 0.474609375, + "learning_rate": 0.0018120142126135425, + "loss": 0.1445, + "step": 20903 + }, + { + "epoch": 0.18145675818786294, + "grad_norm": 0.1298828125, + "learning_rate": 0.0018119960079567214, + "loss": 0.1152, + "step": 20904 + }, + { + "epoch": 0.1814654386680671, + "grad_norm": 0.1640625, + "learning_rate": 0.0018119778025212725, + "loss": 0.0928, + "step": 20905 + }, + { + "epoch": 0.18147411914827127, + "grad_norm": 0.50390625, + "learning_rate": 0.0018119595963072156, + "loss": 0.103, + "step": 20906 + }, + { + "epoch": 0.18148279962847544, + "grad_norm": 0.59375, + "learning_rate": 0.00181194138931457, + "loss": 0.125, + "step": 20907 + }, + { + "epoch": 0.1814914801086796, + "grad_norm": 0.25, + "learning_rate": 0.001811923181543356, + "loss": 0.1011, + "step": 20908 + }, + { + "epoch": 0.18150016058888377, + "grad_norm": 0.1220703125, + "learning_rate": 0.0018119049729935937, + "loss": 0.1309, + "step": 20909 + }, + { + "epoch": 0.18150884106908793, + "grad_norm": 0.15625, + "learning_rate": 0.0018118867636653028, + "loss": 0.1162, + "step": 20910 + }, + { + "epoch": 0.1815175215492921, + "grad_norm": 0.283203125, + "learning_rate": 0.0018118685535585032, + "loss": 0.1553, + "step": 20911 + }, + { + "epoch": 0.18152620202949626, + "grad_norm": 0.380859375, + "learning_rate": 0.0018118503426732147, + "loss": 0.0903, + "step": 20912 + }, + { + "epoch": 0.18153488250970043, + "grad_norm": 0.69921875, + "learning_rate": 0.0018118321310094576, + "loss": 0.1357, + "step": 20913 + }, + { + "epoch": 0.1815435629899046, + "grad_norm": 0.1708984375, + "learning_rate": 0.0018118139185672514, + "loss": 0.0796, + "step": 20914 + }, + { + "epoch": 0.18155224347010876, + "grad_norm": 0.099609375, + "learning_rate": 0.0018117957053466161, + "loss": 0.1074, + "step": 20915 + }, + { + "epoch": 0.18156092395031292, + "grad_norm": 0.404296875, + "learning_rate": 0.0018117774913475719, + "loss": 0.1079, + "step": 20916 + }, + { + "epoch": 0.1815696044305171, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018117592765701388, + "loss": 0.1006, + "step": 20917 + }, + { + "epoch": 0.18157828491072125, + "grad_norm": 0.236328125, + "learning_rate": 0.0018117410610143361, + "loss": 0.0952, + "step": 20918 + }, + { + "epoch": 0.18158696539092542, + "grad_norm": 0.84375, + "learning_rate": 0.0018117228446801839, + "loss": 0.1035, + "step": 20919 + }, + { + "epoch": 0.18159564587112959, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018117046275677026, + "loss": 0.126, + "step": 20920 + }, + { + "epoch": 0.18160432635133375, + "grad_norm": 0.08203125, + "learning_rate": 0.0018116864096769116, + "loss": 0.1221, + "step": 20921 + }, + { + "epoch": 0.18161300683153792, + "grad_norm": 0.51953125, + "learning_rate": 0.0018116681910078313, + "loss": 0.085, + "step": 20922 + }, + { + "epoch": 0.18162168731174208, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018116499715604812, + "loss": 0.1348, + "step": 20923 + }, + { + "epoch": 0.18163036779194625, + "grad_norm": 0.3125, + "learning_rate": 0.0018116317513348812, + "loss": 0.0923, + "step": 20924 + }, + { + "epoch": 0.1816390482721504, + "grad_norm": 0.357421875, + "learning_rate": 0.0018116135303310518, + "loss": 0.0977, + "step": 20925 + }, + { + "epoch": 0.18164772875235458, + "grad_norm": 0.13671875, + "learning_rate": 0.0018115953085490121, + "loss": 0.0903, + "step": 20926 + }, + { + "epoch": 0.18165640923255874, + "grad_norm": 0.1982421875, + "learning_rate": 0.001811577085988783, + "loss": 0.1084, + "step": 20927 + }, + { + "epoch": 0.1816650897127629, + "grad_norm": 0.39453125, + "learning_rate": 0.0018115588626503836, + "loss": 0.0986, + "step": 20928 + }, + { + "epoch": 0.18167377019296707, + "grad_norm": 0.1689453125, + "learning_rate": 0.001811540638533834, + "loss": 0.0967, + "step": 20929 + }, + { + "epoch": 0.18168245067317124, + "grad_norm": 0.41015625, + "learning_rate": 0.0018115224136391545, + "loss": 0.0986, + "step": 20930 + }, + { + "epoch": 0.1816911311533754, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018115041879663647, + "loss": 0.0996, + "step": 20931 + }, + { + "epoch": 0.18169981163357957, + "grad_norm": 1.1875, + "learning_rate": 0.0018114859615154845, + "loss": 0.1455, + "step": 20932 + }, + { + "epoch": 0.18170849211378373, + "grad_norm": 0.28125, + "learning_rate": 0.001811467734286534, + "loss": 0.0903, + "step": 20933 + }, + { + "epoch": 0.1817171725939879, + "grad_norm": 0.435546875, + "learning_rate": 0.001811449506279533, + "loss": 0.1621, + "step": 20934 + }, + { + "epoch": 0.18172585307419206, + "grad_norm": 0.4921875, + "learning_rate": 0.0018114312774945014, + "loss": 0.1001, + "step": 20935 + }, + { + "epoch": 0.18173453355439623, + "grad_norm": 0.283203125, + "learning_rate": 0.0018114130479314596, + "loss": 0.1396, + "step": 20936 + }, + { + "epoch": 0.1817432140346004, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018113948175904268, + "loss": 0.1191, + "step": 20937 + }, + { + "epoch": 0.18175189451480456, + "grad_norm": 0.68359375, + "learning_rate": 0.0018113765864714235, + "loss": 0.126, + "step": 20938 + }, + { + "epoch": 0.18176057499500872, + "grad_norm": 0.43359375, + "learning_rate": 0.0018113583545744697, + "loss": 0.0859, + "step": 20939 + }, + { + "epoch": 0.1817692554752129, + "grad_norm": 0.173828125, + "learning_rate": 0.0018113401218995847, + "loss": 0.1201, + "step": 20940 + }, + { + "epoch": 0.18177793595541705, + "grad_norm": 0.306640625, + "learning_rate": 0.001811321888446789, + "loss": 0.1025, + "step": 20941 + }, + { + "epoch": 0.18178661643562122, + "grad_norm": 0.412109375, + "learning_rate": 0.0018113036542161025, + "loss": 0.1279, + "step": 20942 + }, + { + "epoch": 0.18179529691582538, + "grad_norm": 0.349609375, + "learning_rate": 0.0018112854192075447, + "loss": 0.0947, + "step": 20943 + }, + { + "epoch": 0.18180397739602955, + "grad_norm": 0.10107421875, + "learning_rate": 0.001811267183421136, + "loss": 0.0771, + "step": 20944 + }, + { + "epoch": 0.1818126578762337, + "grad_norm": 0.30859375, + "learning_rate": 0.0018112489468568964, + "loss": 0.0957, + "step": 20945 + }, + { + "epoch": 0.18182133835643788, + "grad_norm": 0.0830078125, + "learning_rate": 0.0018112307095148456, + "loss": 0.1089, + "step": 20946 + }, + { + "epoch": 0.18183001883664204, + "grad_norm": 0.10498046875, + "learning_rate": 0.0018112124713950033, + "loss": 0.1465, + "step": 20947 + }, + { + "epoch": 0.1818386993168462, + "grad_norm": 1.984375, + "learning_rate": 0.0018111942324973897, + "loss": 0.1455, + "step": 20948 + }, + { + "epoch": 0.18184737979705037, + "grad_norm": 0.2255859375, + "learning_rate": 0.001811175992822025, + "loss": 0.1455, + "step": 20949 + }, + { + "epoch": 0.18185606027725454, + "grad_norm": 0.369140625, + "learning_rate": 0.0018111577523689292, + "loss": 0.1299, + "step": 20950 + }, + { + "epoch": 0.1818647407574587, + "grad_norm": 1.25, + "learning_rate": 0.0018111395111381214, + "loss": 0.125, + "step": 20951 + }, + { + "epoch": 0.18187342123766287, + "grad_norm": 0.67578125, + "learning_rate": 0.0018111212691296222, + "loss": 0.3516, + "step": 20952 + }, + { + "epoch": 0.18188210171786703, + "grad_norm": 0.376953125, + "learning_rate": 0.0018111030263434518, + "loss": 0.1787, + "step": 20953 + }, + { + "epoch": 0.1818907821980712, + "grad_norm": 0.26171875, + "learning_rate": 0.0018110847827796294, + "loss": 0.1543, + "step": 20954 + }, + { + "epoch": 0.18189946267827536, + "grad_norm": 0.298828125, + "learning_rate": 0.0018110665384381756, + "loss": 0.0908, + "step": 20955 + }, + { + "epoch": 0.18190814315847953, + "grad_norm": 0.22265625, + "learning_rate": 0.00181104829331911, + "loss": 0.1118, + "step": 20956 + }, + { + "epoch": 0.1819168236386837, + "grad_norm": 0.74609375, + "learning_rate": 0.001811030047422453, + "loss": 0.1182, + "step": 20957 + }, + { + "epoch": 0.18192550411888786, + "grad_norm": 0.384765625, + "learning_rate": 0.001811011800748224, + "loss": 0.1426, + "step": 20958 + }, + { + "epoch": 0.18193418459909202, + "grad_norm": 0.39453125, + "learning_rate": 0.001810993553296443, + "loss": 0.0918, + "step": 20959 + }, + { + "epoch": 0.1819428650792962, + "grad_norm": 0.69921875, + "learning_rate": 0.0018109753050671307, + "loss": 0.084, + "step": 20960 + }, + { + "epoch": 0.18195154555950036, + "grad_norm": 1.0234375, + "learning_rate": 0.001810957056060306, + "loss": 0.1504, + "step": 20961 + }, + { + "epoch": 0.18196022603970452, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018109388062759893, + "loss": 0.0771, + "step": 20962 + }, + { + "epoch": 0.18196890651990869, + "grad_norm": 0.365234375, + "learning_rate": 0.0018109205557142007, + "loss": 0.1118, + "step": 20963 + }, + { + "epoch": 0.18197758700011285, + "grad_norm": 0.0927734375, + "learning_rate": 0.0018109023043749602, + "loss": 0.1621, + "step": 20964 + }, + { + "epoch": 0.18198626748031702, + "grad_norm": 0.130859375, + "learning_rate": 0.0018108840522582876, + "loss": 0.0996, + "step": 20965 + }, + { + "epoch": 0.18199494796052118, + "grad_norm": 0.294921875, + "learning_rate": 0.001810865799364203, + "loss": 0.0874, + "step": 20966 + }, + { + "epoch": 0.18200362844072535, + "grad_norm": 0.1044921875, + "learning_rate": 0.0018108475456927256, + "loss": 0.1504, + "step": 20967 + }, + { + "epoch": 0.1820123089209295, + "grad_norm": 0.310546875, + "learning_rate": 0.0018108292912438768, + "loss": 0.0977, + "step": 20968 + }, + { + "epoch": 0.18202098940113368, + "grad_norm": 0.87890625, + "learning_rate": 0.0018108110360176752, + "loss": 0.1191, + "step": 20969 + }, + { + "epoch": 0.18202966988133784, + "grad_norm": 0.54296875, + "learning_rate": 0.0018107927800141415, + "loss": 0.0977, + "step": 20970 + }, + { + "epoch": 0.182038350361542, + "grad_norm": 0.59375, + "learning_rate": 0.0018107745232332957, + "loss": 0.1182, + "step": 20971 + }, + { + "epoch": 0.18204703084174617, + "grad_norm": 0.7109375, + "learning_rate": 0.0018107562656751574, + "loss": 0.1523, + "step": 20972 + }, + { + "epoch": 0.18205571132195034, + "grad_norm": 0.271484375, + "learning_rate": 0.0018107380073397468, + "loss": 0.1279, + "step": 20973 + }, + { + "epoch": 0.1820643918021545, + "grad_norm": 0.86328125, + "learning_rate": 0.0018107197482270837, + "loss": 0.1016, + "step": 20974 + }, + { + "epoch": 0.18207307228235867, + "grad_norm": 0.79296875, + "learning_rate": 0.0018107014883371882, + "loss": 0.127, + "step": 20975 + }, + { + "epoch": 0.18208175276256283, + "grad_norm": 0.279296875, + "learning_rate": 0.0018106832276700805, + "loss": 0.1416, + "step": 20976 + }, + { + "epoch": 0.182090433242767, + "grad_norm": 0.10546875, + "learning_rate": 0.0018106649662257797, + "loss": 0.0742, + "step": 20977 + }, + { + "epoch": 0.18209911372297116, + "grad_norm": 0.4453125, + "learning_rate": 0.0018106467040043066, + "loss": 0.0952, + "step": 20978 + }, + { + "epoch": 0.18210779420317533, + "grad_norm": 0.40234375, + "learning_rate": 0.0018106284410056813, + "loss": 0.1445, + "step": 20979 + }, + { + "epoch": 0.1821164746833795, + "grad_norm": 0.2431640625, + "learning_rate": 0.001810610177229923, + "loss": 0.0918, + "step": 20980 + }, + { + "epoch": 0.18212515516358366, + "grad_norm": 0.08251953125, + "learning_rate": 0.001810591912677052, + "loss": 0.0854, + "step": 20981 + }, + { + "epoch": 0.18213383564378782, + "grad_norm": 0.287109375, + "learning_rate": 0.001810573647347089, + "loss": 0.0796, + "step": 20982 + }, + { + "epoch": 0.182142516123992, + "grad_norm": 0.2158203125, + "learning_rate": 0.0018105553812400527, + "loss": 0.1611, + "step": 20983 + }, + { + "epoch": 0.18215119660419615, + "grad_norm": 0.31640625, + "learning_rate": 0.0018105371143559642, + "loss": 0.1299, + "step": 20984 + }, + { + "epoch": 0.18215987708440032, + "grad_norm": 0.40625, + "learning_rate": 0.0018105188466948428, + "loss": 0.1445, + "step": 20985 + }, + { + "epoch": 0.18216855756460448, + "grad_norm": 0.546875, + "learning_rate": 0.0018105005782567086, + "loss": 0.1196, + "step": 20986 + }, + { + "epoch": 0.18217723804480865, + "grad_norm": 0.56640625, + "learning_rate": 0.0018104823090415814, + "loss": 0.1104, + "step": 20987 + }, + { + "epoch": 0.1821859185250128, + "grad_norm": 0.10888671875, + "learning_rate": 0.0018104640390494817, + "loss": 0.1108, + "step": 20988 + }, + { + "epoch": 0.18219459900521698, + "grad_norm": 0.36328125, + "learning_rate": 0.0018104457682804294, + "loss": 0.0869, + "step": 20989 + }, + { + "epoch": 0.18220327948542114, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018104274967344438, + "loss": 0.1172, + "step": 20990 + }, + { + "epoch": 0.1822119599656253, + "grad_norm": 0.1435546875, + "learning_rate": 0.0018104092244115456, + "loss": 0.1523, + "step": 20991 + }, + { + "epoch": 0.18222064044582947, + "grad_norm": 0.265625, + "learning_rate": 0.0018103909513117547, + "loss": 0.1143, + "step": 20992 + }, + { + "epoch": 0.18222932092603364, + "grad_norm": 0.46875, + "learning_rate": 0.0018103726774350909, + "loss": 0.1348, + "step": 20993 + }, + { + "epoch": 0.1822380014062378, + "grad_norm": 0.8515625, + "learning_rate": 0.0018103544027815738, + "loss": 0.1553, + "step": 20994 + }, + { + "epoch": 0.18224668188644197, + "grad_norm": 0.671875, + "learning_rate": 0.001810336127351224, + "loss": 0.1113, + "step": 20995 + }, + { + "epoch": 0.18225536236664613, + "grad_norm": 0.38671875, + "learning_rate": 0.0018103178511440616, + "loss": 0.082, + "step": 20996 + }, + { + "epoch": 0.1822640428468503, + "grad_norm": 0.1845703125, + "learning_rate": 0.0018102995741601058, + "loss": 0.0942, + "step": 20997 + }, + { + "epoch": 0.18227272332705446, + "grad_norm": 0.77734375, + "learning_rate": 0.0018102812963993777, + "loss": 0.1128, + "step": 20998 + }, + { + "epoch": 0.18228140380725863, + "grad_norm": 0.359375, + "learning_rate": 0.0018102630178618959, + "loss": 0.082, + "step": 20999 + }, + { + "epoch": 0.1822900842874628, + "grad_norm": 0.2890625, + "learning_rate": 0.0018102447385476817, + "loss": 0.0967, + "step": 21000 + }, + { + "epoch": 0.18229876476766696, + "grad_norm": 0.427734375, + "learning_rate": 0.0018102264584567542, + "loss": 0.1094, + "step": 21001 + }, + { + "epoch": 0.18230744524787112, + "grad_norm": 0.73046875, + "learning_rate": 0.001810208177589134, + "loss": 0.0938, + "step": 21002 + }, + { + "epoch": 0.1823161257280753, + "grad_norm": 0.3125, + "learning_rate": 0.0018101898959448407, + "loss": 0.0928, + "step": 21003 + }, + { + "epoch": 0.18232480620827946, + "grad_norm": 0.1015625, + "learning_rate": 0.0018101716135238944, + "loss": 0.1113, + "step": 21004 + }, + { + "epoch": 0.18233348668848362, + "grad_norm": 0.10400390625, + "learning_rate": 0.0018101533303263148, + "loss": 0.0972, + "step": 21005 + }, + { + "epoch": 0.18234216716868779, + "grad_norm": 0.1201171875, + "learning_rate": 0.0018101350463521225, + "loss": 0.1475, + "step": 21006 + }, + { + "epoch": 0.18235084764889195, + "grad_norm": 0.1484375, + "learning_rate": 0.001810116761601337, + "loss": 0.1089, + "step": 21007 + }, + { + "epoch": 0.1823595281290961, + "grad_norm": 0.8203125, + "learning_rate": 0.0018100984760739786, + "loss": 0.1172, + "step": 21008 + }, + { + "epoch": 0.18236820860930025, + "grad_norm": 0.279296875, + "learning_rate": 0.0018100801897700674, + "loss": 0.1279, + "step": 21009 + }, + { + "epoch": 0.18237688908950442, + "grad_norm": 0.431640625, + "learning_rate": 0.0018100619026896227, + "loss": 0.1602, + "step": 21010 + }, + { + "epoch": 0.18238556956970858, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018100436148326654, + "loss": 0.1221, + "step": 21011 + }, + { + "epoch": 0.18239425004991275, + "grad_norm": 0.171875, + "learning_rate": 0.0018100253261992148, + "loss": 0.1387, + "step": 21012 + }, + { + "epoch": 0.1824029305301169, + "grad_norm": 0.6640625, + "learning_rate": 0.0018100070367892912, + "loss": 0.0845, + "step": 21013 + }, + { + "epoch": 0.18241161101032108, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018099887466029145, + "loss": 0.0879, + "step": 21014 + }, + { + "epoch": 0.18242029149052524, + "grad_norm": 0.197265625, + "learning_rate": 0.0018099704556401048, + "loss": 0.0933, + "step": 21015 + }, + { + "epoch": 0.1824289719707294, + "grad_norm": 0.60546875, + "learning_rate": 0.0018099521639008825, + "loss": 0.0723, + "step": 21016 + }, + { + "epoch": 0.18243765245093357, + "grad_norm": 0.4921875, + "learning_rate": 0.0018099338713852666, + "loss": 0.1143, + "step": 21017 + }, + { + "epoch": 0.18244633293113774, + "grad_norm": 0.349609375, + "learning_rate": 0.0018099155780932782, + "loss": 0.1133, + "step": 21018 + }, + { + "epoch": 0.1824550134113419, + "grad_norm": 0.765625, + "learning_rate": 0.0018098972840249364, + "loss": 0.1182, + "step": 21019 + }, + { + "epoch": 0.18246369389154607, + "grad_norm": 0.341796875, + "learning_rate": 0.0018098789891802616, + "loss": 0.1152, + "step": 21020 + }, + { + "epoch": 0.18247237437175023, + "grad_norm": 0.373046875, + "learning_rate": 0.001809860693559274, + "loss": 0.1162, + "step": 21021 + }, + { + "epoch": 0.1824810548519544, + "grad_norm": 1.7734375, + "learning_rate": 0.0018098423971619933, + "loss": 0.0771, + "step": 21022 + }, + { + "epoch": 0.18248973533215856, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018098240999884399, + "loss": 0.0918, + "step": 21023 + }, + { + "epoch": 0.18249841581236273, + "grad_norm": 1.1484375, + "learning_rate": 0.0018098058020386333, + "loss": 0.1104, + "step": 21024 + }, + { + "epoch": 0.1825070962925669, + "grad_norm": 0.79296875, + "learning_rate": 0.0018097875033125934, + "loss": 0.1348, + "step": 21025 + }, + { + "epoch": 0.18251577677277106, + "grad_norm": 0.201171875, + "learning_rate": 0.0018097692038103411, + "loss": 0.085, + "step": 21026 + }, + { + "epoch": 0.18252445725297523, + "grad_norm": 0.240234375, + "learning_rate": 0.0018097509035318955, + "loss": 0.1387, + "step": 21027 + }, + { + "epoch": 0.1825331377331794, + "grad_norm": 0.166015625, + "learning_rate": 0.0018097326024772771, + "loss": 0.1406, + "step": 21028 + }, + { + "epoch": 0.18254181821338356, + "grad_norm": 1.0234375, + "learning_rate": 0.001809714300646506, + "loss": 0.167, + "step": 21029 + }, + { + "epoch": 0.18255049869358772, + "grad_norm": 0.298828125, + "learning_rate": 0.0018096959980396018, + "loss": 0.0728, + "step": 21030 + }, + { + "epoch": 0.18255917917379189, + "grad_norm": 0.64453125, + "learning_rate": 0.0018096776946565846, + "loss": 0.1309, + "step": 21031 + }, + { + "epoch": 0.18256785965399605, + "grad_norm": 0.205078125, + "learning_rate": 0.0018096593904974748, + "loss": 0.1328, + "step": 21032 + }, + { + "epoch": 0.18257654013420022, + "grad_norm": 0.158203125, + "learning_rate": 0.001809641085562292, + "loss": 0.0869, + "step": 21033 + }, + { + "epoch": 0.18258522061440438, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018096227798510562, + "loss": 0.1167, + "step": 21034 + }, + { + "epoch": 0.18259390109460855, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018096044733637879, + "loss": 0.1021, + "step": 21035 + }, + { + "epoch": 0.1826025815748127, + "grad_norm": 0.2392578125, + "learning_rate": 0.0018095861661005066, + "loss": 0.1025, + "step": 21036 + }, + { + "epoch": 0.18261126205501688, + "grad_norm": 0.08837890625, + "learning_rate": 0.0018095678580612327, + "loss": 0.1543, + "step": 21037 + }, + { + "epoch": 0.18261994253522104, + "grad_norm": 0.1875, + "learning_rate": 0.001809549549245986, + "loss": 0.0957, + "step": 21038 + }, + { + "epoch": 0.1826286230154252, + "grad_norm": 0.30859375, + "learning_rate": 0.0018095312396547865, + "loss": 0.123, + "step": 21039 + }, + { + "epoch": 0.18263730349562937, + "grad_norm": 0.83203125, + "learning_rate": 0.0018095129292876545, + "loss": 0.1279, + "step": 21040 + }, + { + "epoch": 0.18264598397583354, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018094946181446098, + "loss": 0.0957, + "step": 21041 + }, + { + "epoch": 0.1826546644560377, + "grad_norm": 0.146484375, + "learning_rate": 0.001809476306225672, + "loss": 0.127, + "step": 21042 + }, + { + "epoch": 0.18266334493624187, + "grad_norm": 0.083984375, + "learning_rate": 0.001809457993530862, + "loss": 0.1001, + "step": 21043 + }, + { + "epoch": 0.18267202541644603, + "grad_norm": 0.169921875, + "learning_rate": 0.001809439680060199, + "loss": 0.1055, + "step": 21044 + }, + { + "epoch": 0.1826807058966502, + "grad_norm": 0.07861328125, + "learning_rate": 0.0018094213658137038, + "loss": 0.103, + "step": 21045 + }, + { + "epoch": 0.18268938637685436, + "grad_norm": 0.94140625, + "learning_rate": 0.001809403050791396, + "loss": 0.0942, + "step": 21046 + }, + { + "epoch": 0.18269806685705853, + "grad_norm": 0.375, + "learning_rate": 0.0018093847349932956, + "loss": 0.1182, + "step": 21047 + }, + { + "epoch": 0.1827067473372627, + "grad_norm": 0.18359375, + "learning_rate": 0.001809366418419423, + "loss": 0.1299, + "step": 21048 + }, + { + "epoch": 0.18271542781746686, + "grad_norm": 0.466796875, + "learning_rate": 0.0018093481010697976, + "loss": 0.1006, + "step": 21049 + }, + { + "epoch": 0.18272410829767102, + "grad_norm": 0.359375, + "learning_rate": 0.0018093297829444397, + "loss": 0.1309, + "step": 21050 + }, + { + "epoch": 0.1827327887778752, + "grad_norm": 0.26171875, + "learning_rate": 0.0018093114640433696, + "loss": 0.1211, + "step": 21051 + }, + { + "epoch": 0.18274146925807935, + "grad_norm": 0.09130859375, + "learning_rate": 0.0018092931443666072, + "loss": 0.1016, + "step": 21052 + }, + { + "epoch": 0.18275014973828352, + "grad_norm": 0.09423828125, + "learning_rate": 0.0018092748239141726, + "loss": 0.0938, + "step": 21053 + }, + { + "epoch": 0.18275883021848768, + "grad_norm": 0.25, + "learning_rate": 0.0018092565026860854, + "loss": 0.1084, + "step": 21054 + }, + { + "epoch": 0.18276751069869185, + "grad_norm": 0.322265625, + "learning_rate": 0.001809238180682366, + "loss": 0.0806, + "step": 21055 + }, + { + "epoch": 0.182776191178896, + "grad_norm": 0.85546875, + "learning_rate": 0.0018092198579030344, + "loss": 0.1201, + "step": 21056 + }, + { + "epoch": 0.18278487165910018, + "grad_norm": 0.2158203125, + "learning_rate": 0.0018092015343481108, + "loss": 0.0996, + "step": 21057 + }, + { + "epoch": 0.18279355213930434, + "grad_norm": 0.32421875, + "learning_rate": 0.0018091832100176146, + "loss": 0.1128, + "step": 21058 + }, + { + "epoch": 0.1828022326195085, + "grad_norm": 0.126953125, + "learning_rate": 0.001809164884911567, + "loss": 0.0952, + "step": 21059 + }, + { + "epoch": 0.18281091309971267, + "grad_norm": 0.419921875, + "learning_rate": 0.0018091465590299868, + "loss": 0.126, + "step": 21060 + }, + { + "epoch": 0.18281959357991684, + "grad_norm": 0.111328125, + "learning_rate": 0.0018091282323728947, + "loss": 0.084, + "step": 21061 + }, + { + "epoch": 0.182828274060121, + "grad_norm": 0.12451171875, + "learning_rate": 0.0018091099049403106, + "loss": 0.0981, + "step": 21062 + }, + { + "epoch": 0.18283695454032517, + "grad_norm": 0.17578125, + "learning_rate": 0.0018090915767322545, + "loss": 0.0684, + "step": 21063 + }, + { + "epoch": 0.18284563502052933, + "grad_norm": 0.34765625, + "learning_rate": 0.0018090732477487467, + "loss": 0.0947, + "step": 21064 + }, + { + "epoch": 0.1828543155007335, + "grad_norm": 0.38671875, + "learning_rate": 0.0018090549179898068, + "loss": 0.1138, + "step": 21065 + }, + { + "epoch": 0.18286299598093766, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018090365874554556, + "loss": 0.106, + "step": 21066 + }, + { + "epoch": 0.18287167646114183, + "grad_norm": 0.25390625, + "learning_rate": 0.001809018256145712, + "loss": 0.1377, + "step": 21067 + }, + { + "epoch": 0.182880356941346, + "grad_norm": 0.318359375, + "learning_rate": 0.001808999924060597, + "loss": 0.1104, + "step": 21068 + }, + { + "epoch": 0.18288903742155016, + "grad_norm": 0.130859375, + "learning_rate": 0.0018089815912001302, + "loss": 0.1533, + "step": 21069 + }, + { + "epoch": 0.18289771790175433, + "grad_norm": 0.52734375, + "learning_rate": 0.0018089632575643323, + "loss": 0.1377, + "step": 21070 + }, + { + "epoch": 0.1829063983819585, + "grad_norm": 0.1767578125, + "learning_rate": 0.001808944923153222, + "loss": 0.1387, + "step": 21071 + }, + { + "epoch": 0.18291507886216266, + "grad_norm": 0.2275390625, + "learning_rate": 0.0018089265879668208, + "loss": 0.1299, + "step": 21072 + }, + { + "epoch": 0.18292375934236682, + "grad_norm": 0.380859375, + "learning_rate": 0.0018089082520051478, + "loss": 0.0864, + "step": 21073 + }, + { + "epoch": 0.18293243982257099, + "grad_norm": 0.197265625, + "learning_rate": 0.0018088899152682236, + "loss": 0.124, + "step": 21074 + }, + { + "epoch": 0.18294112030277515, + "grad_norm": 0.52734375, + "learning_rate": 0.001808871577756068, + "loss": 0.1172, + "step": 21075 + }, + { + "epoch": 0.18294980078297932, + "grad_norm": 0.265625, + "learning_rate": 0.0018088532394687012, + "loss": 0.1045, + "step": 21076 + }, + { + "epoch": 0.18295848126318348, + "grad_norm": 0.53125, + "learning_rate": 0.001808834900406143, + "loss": 0.0967, + "step": 21077 + }, + { + "epoch": 0.18296716174338765, + "grad_norm": 0.1884765625, + "learning_rate": 0.0018088165605684135, + "loss": 0.127, + "step": 21078 + }, + { + "epoch": 0.1829758422235918, + "grad_norm": 0.2275390625, + "learning_rate": 0.001808798219955533, + "loss": 0.0938, + "step": 21079 + }, + { + "epoch": 0.18298452270379598, + "grad_norm": 0.33984375, + "learning_rate": 0.0018087798785675214, + "loss": 0.0933, + "step": 21080 + }, + { + "epoch": 0.18299320318400014, + "grad_norm": 0.359375, + "learning_rate": 0.0018087615364043989, + "loss": 0.1162, + "step": 21081 + }, + { + "epoch": 0.1830018836642043, + "grad_norm": 0.302734375, + "learning_rate": 0.0018087431934661852, + "loss": 0.1318, + "step": 21082 + }, + { + "epoch": 0.18301056414440847, + "grad_norm": 0.0849609375, + "learning_rate": 0.0018087248497529007, + "loss": 0.1123, + "step": 21083 + }, + { + "epoch": 0.18301924462461264, + "grad_norm": 0.423828125, + "learning_rate": 0.0018087065052645652, + "loss": 0.166, + "step": 21084 + }, + { + "epoch": 0.1830279251048168, + "grad_norm": 0.234375, + "learning_rate": 0.0018086881600011993, + "loss": 0.1162, + "step": 21085 + }, + { + "epoch": 0.18303660558502097, + "grad_norm": 0.333984375, + "learning_rate": 0.0018086698139628224, + "loss": 0.0947, + "step": 21086 + }, + { + "epoch": 0.18304528606522513, + "grad_norm": 0.12109375, + "learning_rate": 0.0018086514671494546, + "loss": 0.1182, + "step": 21087 + }, + { + "epoch": 0.1830539665454293, + "grad_norm": 0.671875, + "learning_rate": 0.0018086331195611167, + "loss": 0.104, + "step": 21088 + }, + { + "epoch": 0.18306264702563346, + "grad_norm": 0.71875, + "learning_rate": 0.001808614771197828, + "loss": 0.1064, + "step": 21089 + }, + { + "epoch": 0.18307132750583763, + "grad_norm": 0.423828125, + "learning_rate": 0.001808596422059609, + "loss": 0.1279, + "step": 21090 + }, + { + "epoch": 0.1830800079860418, + "grad_norm": 0.7578125, + "learning_rate": 0.0018085780721464793, + "loss": 0.1484, + "step": 21091 + }, + { + "epoch": 0.18308868846624596, + "grad_norm": 1.171875, + "learning_rate": 0.0018085597214584595, + "loss": 0.1074, + "step": 21092 + }, + { + "epoch": 0.18309736894645012, + "grad_norm": 0.40625, + "learning_rate": 0.0018085413699955692, + "loss": 0.0962, + "step": 21093 + }, + { + "epoch": 0.1831060494266543, + "grad_norm": 0.19921875, + "learning_rate": 0.001808523017757829, + "loss": 0.0918, + "step": 21094 + }, + { + "epoch": 0.18311472990685845, + "grad_norm": 0.5546875, + "learning_rate": 0.0018085046647452587, + "loss": 0.1089, + "step": 21095 + }, + { + "epoch": 0.18312341038706262, + "grad_norm": 0.49609375, + "learning_rate": 0.0018084863109578781, + "loss": 0.0801, + "step": 21096 + }, + { + "epoch": 0.18313209086726678, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018084679563957077, + "loss": 0.1289, + "step": 21097 + }, + { + "epoch": 0.18314077134747095, + "grad_norm": 0.109375, + "learning_rate": 0.0018084496010587672, + "loss": 0.0889, + "step": 21098 + }, + { + "epoch": 0.1831494518276751, + "grad_norm": 0.279296875, + "learning_rate": 0.001808431244947077, + "loss": 0.1172, + "step": 21099 + }, + { + "epoch": 0.18315813230787928, + "grad_norm": 0.484375, + "learning_rate": 0.0018084128880606573, + "loss": 0.1123, + "step": 21100 + }, + { + "epoch": 0.18316681278808344, + "grad_norm": 0.0654296875, + "learning_rate": 0.0018083945303995273, + "loss": 0.1001, + "step": 21101 + }, + { + "epoch": 0.1831754932682876, + "grad_norm": 0.29296875, + "learning_rate": 0.0018083761719637082, + "loss": 0.1187, + "step": 21102 + }, + { + "epoch": 0.18318417374849177, + "grad_norm": 0.130859375, + "learning_rate": 0.0018083578127532191, + "loss": 0.1235, + "step": 21103 + }, + { + "epoch": 0.18319285422869594, + "grad_norm": 0.2236328125, + "learning_rate": 0.0018083394527680808, + "loss": 0.0938, + "step": 21104 + }, + { + "epoch": 0.1832015347089001, + "grad_norm": 0.283203125, + "learning_rate": 0.0018083210920083134, + "loss": 0.1289, + "step": 21105 + }, + { + "epoch": 0.18321021518910427, + "grad_norm": 0.47265625, + "learning_rate": 0.0018083027304739362, + "loss": 0.104, + "step": 21106 + }, + { + "epoch": 0.18321889566930843, + "grad_norm": 0.2314453125, + "learning_rate": 0.00180828436816497, + "loss": 0.083, + "step": 21107 + }, + { + "epoch": 0.1832275761495126, + "grad_norm": 0.326171875, + "learning_rate": 0.0018082660050814346, + "loss": 0.0684, + "step": 21108 + }, + { + "epoch": 0.18323625662971676, + "grad_norm": 0.53515625, + "learning_rate": 0.0018082476412233505, + "loss": 0.0928, + "step": 21109 + }, + { + "epoch": 0.18324493710992093, + "grad_norm": 0.73046875, + "learning_rate": 0.001808229276590737, + "loss": 0.1465, + "step": 21110 + }, + { + "epoch": 0.1832536175901251, + "grad_norm": 2.59375, + "learning_rate": 0.0018082109111836147, + "loss": 0.2188, + "step": 21111 + }, + { + "epoch": 0.18326229807032926, + "grad_norm": 0.185546875, + "learning_rate": 0.0018081925450020035, + "loss": 0.1406, + "step": 21112 + }, + { + "epoch": 0.18327097855053343, + "grad_norm": 0.1796875, + "learning_rate": 0.0018081741780459238, + "loss": 0.0923, + "step": 21113 + }, + { + "epoch": 0.1832796590307376, + "grad_norm": 0.478515625, + "learning_rate": 0.0018081558103153952, + "loss": 0.1611, + "step": 21114 + }, + { + "epoch": 0.18328833951094176, + "grad_norm": 0.1396484375, + "learning_rate": 0.001808137441810438, + "loss": 0.0913, + "step": 21115 + }, + { + "epoch": 0.18329701999114592, + "grad_norm": 0.2177734375, + "learning_rate": 0.0018081190725310727, + "loss": 0.0967, + "step": 21116 + }, + { + "epoch": 0.18330570047135009, + "grad_norm": 0.390625, + "learning_rate": 0.0018081007024773191, + "loss": 0.165, + "step": 21117 + }, + { + "epoch": 0.18331438095155425, + "grad_norm": 0.279296875, + "learning_rate": 0.0018080823316491965, + "loss": 0.1338, + "step": 21118 + }, + { + "epoch": 0.18332306143175842, + "grad_norm": 0.16015625, + "learning_rate": 0.0018080639600467263, + "loss": 0.1069, + "step": 21119 + }, + { + "epoch": 0.18333174191196258, + "grad_norm": 0.330078125, + "learning_rate": 0.0018080455876699278, + "loss": 0.1045, + "step": 21120 + }, + { + "epoch": 0.18334042239216675, + "grad_norm": 0.76953125, + "learning_rate": 0.0018080272145188211, + "loss": 0.1318, + "step": 21121 + }, + { + "epoch": 0.1833491028723709, + "grad_norm": 0.71875, + "learning_rate": 0.001808008840593427, + "loss": 0.1016, + "step": 21122 + }, + { + "epoch": 0.18335778335257508, + "grad_norm": 0.60546875, + "learning_rate": 0.0018079904658937647, + "loss": 0.085, + "step": 21123 + }, + { + "epoch": 0.18336646383277924, + "grad_norm": 0.1962890625, + "learning_rate": 0.001807972090419855, + "loss": 0.1138, + "step": 21124 + }, + { + "epoch": 0.1833751443129834, + "grad_norm": 0.216796875, + "learning_rate": 0.0018079537141717172, + "loss": 0.124, + "step": 21125 + }, + { + "epoch": 0.18338382479318757, + "grad_norm": 0.1201171875, + "learning_rate": 0.001807935337149372, + "loss": 0.1289, + "step": 21126 + }, + { + "epoch": 0.18339250527339174, + "grad_norm": 0.154296875, + "learning_rate": 0.0018079169593528395, + "loss": 0.1602, + "step": 21127 + }, + { + "epoch": 0.1834011857535959, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018078985807821392, + "loss": 0.1006, + "step": 21128 + }, + { + "epoch": 0.18340986623380007, + "grad_norm": 0.1376953125, + "learning_rate": 0.0018078802014372922, + "loss": 0.0923, + "step": 21129 + }, + { + "epoch": 0.18341854671400423, + "grad_norm": 0.220703125, + "learning_rate": 0.0018078618213183178, + "loss": 0.1069, + "step": 21130 + }, + { + "epoch": 0.18342722719420837, + "grad_norm": 0.458984375, + "learning_rate": 0.0018078434404252364, + "loss": 0.1299, + "step": 21131 + }, + { + "epoch": 0.18343590767441253, + "grad_norm": 0.271484375, + "learning_rate": 0.001807825058758068, + "loss": 0.1201, + "step": 21132 + }, + { + "epoch": 0.1834445881546167, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018078066763168327, + "loss": 0.1377, + "step": 21133 + }, + { + "epoch": 0.18345326863482087, + "grad_norm": 0.1904296875, + "learning_rate": 0.001807788293101551, + "loss": 0.0977, + "step": 21134 + }, + { + "epoch": 0.18346194911502503, + "grad_norm": 0.2060546875, + "learning_rate": 0.0018077699091122424, + "loss": 0.0957, + "step": 21135 + }, + { + "epoch": 0.1834706295952292, + "grad_norm": 0.28125, + "learning_rate": 0.0018077515243489271, + "loss": 0.1455, + "step": 21136 + }, + { + "epoch": 0.18347931007543336, + "grad_norm": 0.296875, + "learning_rate": 0.0018077331388116256, + "loss": 0.1055, + "step": 21137 + }, + { + "epoch": 0.18348799055563753, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018077147525003576, + "loss": 0.123, + "step": 21138 + }, + { + "epoch": 0.1834966710358417, + "grad_norm": 0.06005859375, + "learning_rate": 0.0018076963654151436, + "loss": 0.1084, + "step": 21139 + }, + { + "epoch": 0.18350535151604586, + "grad_norm": 0.142578125, + "learning_rate": 0.0018076779775560033, + "loss": 0.1543, + "step": 21140 + }, + { + "epoch": 0.18351403199625002, + "grad_norm": 0.2255859375, + "learning_rate": 0.001807659588922957, + "loss": 0.0967, + "step": 21141 + }, + { + "epoch": 0.1835227124764542, + "grad_norm": 0.98046875, + "learning_rate": 0.0018076411995160248, + "loss": 0.1309, + "step": 21142 + }, + { + "epoch": 0.18353139295665835, + "grad_norm": 0.185546875, + "learning_rate": 0.0018076228093352271, + "loss": 0.1201, + "step": 21143 + }, + { + "epoch": 0.18354007343686252, + "grad_norm": 0.1982421875, + "learning_rate": 0.0018076044183805836, + "loss": 0.1328, + "step": 21144 + }, + { + "epoch": 0.18354875391706668, + "grad_norm": 0.134765625, + "learning_rate": 0.0018075860266521142, + "loss": 0.0933, + "step": 21145 + }, + { + "epoch": 0.18355743439727085, + "grad_norm": 0.60546875, + "learning_rate": 0.0018075676341498396, + "loss": 0.1602, + "step": 21146 + }, + { + "epoch": 0.183566114877475, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018075492408737798, + "loss": 0.1182, + "step": 21147 + }, + { + "epoch": 0.18357479535767918, + "grad_norm": 0.2412109375, + "learning_rate": 0.0018075308468239546, + "loss": 0.1045, + "step": 21148 + }, + { + "epoch": 0.18358347583788334, + "grad_norm": 0.283203125, + "learning_rate": 0.0018075124520003842, + "loss": 0.1074, + "step": 21149 + }, + { + "epoch": 0.1835921563180875, + "grad_norm": 0.2080078125, + "learning_rate": 0.001807494056403089, + "loss": 0.0723, + "step": 21150 + }, + { + "epoch": 0.18360083679829167, + "grad_norm": 0.39453125, + "learning_rate": 0.001807475660032089, + "loss": 0.1338, + "step": 21151 + }, + { + "epoch": 0.18360951727849584, + "grad_norm": 0.57421875, + "learning_rate": 0.0018074572628874043, + "loss": 0.084, + "step": 21152 + }, + { + "epoch": 0.1836181977587, + "grad_norm": 0.7421875, + "learning_rate": 0.0018074388649690547, + "loss": 0.1172, + "step": 21153 + }, + { + "epoch": 0.18362687823890417, + "grad_norm": 0.0859375, + "learning_rate": 0.0018074204662770607, + "loss": 0.1309, + "step": 21154 + }, + { + "epoch": 0.18363555871910833, + "grad_norm": 0.1162109375, + "learning_rate": 0.001807402066811442, + "loss": 0.123, + "step": 21155 + }, + { + "epoch": 0.1836442391993125, + "grad_norm": 0.08056640625, + "learning_rate": 0.0018073836665722196, + "loss": 0.1074, + "step": 21156 + }, + { + "epoch": 0.18365291967951666, + "grad_norm": 0.32421875, + "learning_rate": 0.0018073652655594126, + "loss": 0.1289, + "step": 21157 + }, + { + "epoch": 0.18366160015972083, + "grad_norm": 1.4453125, + "learning_rate": 0.0018073468637730418, + "loss": 0.0947, + "step": 21158 + }, + { + "epoch": 0.183670280639925, + "grad_norm": 0.421875, + "learning_rate": 0.0018073284612131271, + "loss": 0.1025, + "step": 21159 + }, + { + "epoch": 0.18367896112012916, + "grad_norm": 1.0078125, + "learning_rate": 0.0018073100578796886, + "loss": 0.2109, + "step": 21160 + }, + { + "epoch": 0.18368764160033332, + "grad_norm": 0.09765625, + "learning_rate": 0.0018072916537727464, + "loss": 0.083, + "step": 21161 + }, + { + "epoch": 0.1836963220805375, + "grad_norm": 0.6484375, + "learning_rate": 0.0018072732488923207, + "loss": 0.1543, + "step": 21162 + }, + { + "epoch": 0.18370500256074165, + "grad_norm": 0.37109375, + "learning_rate": 0.0018072548432384314, + "loss": 0.0869, + "step": 21163 + }, + { + "epoch": 0.18371368304094582, + "grad_norm": 0.8828125, + "learning_rate": 0.0018072364368110993, + "loss": 0.085, + "step": 21164 + }, + { + "epoch": 0.18372236352114998, + "grad_norm": 0.6640625, + "learning_rate": 0.0018072180296103437, + "loss": 0.0996, + "step": 21165 + }, + { + "epoch": 0.18373104400135415, + "grad_norm": 0.08154296875, + "learning_rate": 0.0018071996216361852, + "loss": 0.0898, + "step": 21166 + }, + { + "epoch": 0.18373972448155831, + "grad_norm": 0.85546875, + "learning_rate": 0.0018071812128886437, + "loss": 0.1152, + "step": 21167 + }, + { + "epoch": 0.18374840496176248, + "grad_norm": 0.271484375, + "learning_rate": 0.0018071628033677397, + "loss": 0.1455, + "step": 21168 + }, + { + "epoch": 0.18375708544196664, + "grad_norm": 0.484375, + "learning_rate": 0.001807144393073493, + "loss": 0.1143, + "step": 21169 + }, + { + "epoch": 0.1837657659221708, + "grad_norm": 0.6015625, + "learning_rate": 0.0018071259820059235, + "loss": 0.1064, + "step": 21170 + }, + { + "epoch": 0.18377444640237497, + "grad_norm": 0.337890625, + "learning_rate": 0.0018071075701650522, + "loss": 0.1216, + "step": 21171 + }, + { + "epoch": 0.18378312688257914, + "grad_norm": 0.109375, + "learning_rate": 0.0018070891575508981, + "loss": 0.1436, + "step": 21172 + }, + { + "epoch": 0.1837918073627833, + "grad_norm": 0.07666015625, + "learning_rate": 0.0018070707441634824, + "loss": 0.0791, + "step": 21173 + }, + { + "epoch": 0.18380048784298747, + "grad_norm": 0.55859375, + "learning_rate": 0.0018070523300028245, + "loss": 0.1807, + "step": 21174 + }, + { + "epoch": 0.18380916832319164, + "grad_norm": 0.7890625, + "learning_rate": 0.0018070339150689449, + "loss": 0.1289, + "step": 21175 + }, + { + "epoch": 0.1838178488033958, + "grad_norm": 0.75390625, + "learning_rate": 0.0018070154993618634, + "loss": 0.1455, + "step": 21176 + }, + { + "epoch": 0.18382652928359997, + "grad_norm": 0.3203125, + "learning_rate": 0.0018069970828816007, + "loss": 0.1377, + "step": 21177 + }, + { + "epoch": 0.18383520976380413, + "grad_norm": 0.1875, + "learning_rate": 0.0018069786656281764, + "loss": 0.1338, + "step": 21178 + }, + { + "epoch": 0.1838438902440083, + "grad_norm": 0.318359375, + "learning_rate": 0.0018069602476016107, + "loss": 0.1426, + "step": 21179 + }, + { + "epoch": 0.18385257072421246, + "grad_norm": 0.2314453125, + "learning_rate": 0.0018069418288019245, + "loss": 0.083, + "step": 21180 + }, + { + "epoch": 0.18386125120441663, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018069234092291367, + "loss": 0.1416, + "step": 21181 + }, + { + "epoch": 0.1838699316846208, + "grad_norm": 0.24609375, + "learning_rate": 0.0018069049888832685, + "loss": 0.1221, + "step": 21182 + }, + { + "epoch": 0.18387861216482496, + "grad_norm": 0.326171875, + "learning_rate": 0.0018068865677643394, + "loss": 0.1455, + "step": 21183 + }, + { + "epoch": 0.18388729264502912, + "grad_norm": 0.76171875, + "learning_rate": 0.0018068681458723698, + "loss": 0.1318, + "step": 21184 + }, + { + "epoch": 0.1838959731252333, + "grad_norm": 0.37890625, + "learning_rate": 0.00180684972320738, + "loss": 0.105, + "step": 21185 + }, + { + "epoch": 0.18390465360543745, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018068312997693897, + "loss": 0.0859, + "step": 21186 + }, + { + "epoch": 0.18391333408564162, + "grad_norm": 0.458984375, + "learning_rate": 0.0018068128755584197, + "loss": 0.123, + "step": 21187 + }, + { + "epoch": 0.18392201456584578, + "grad_norm": 0.3125, + "learning_rate": 0.0018067944505744892, + "loss": 0.1084, + "step": 21188 + }, + { + "epoch": 0.18393069504604995, + "grad_norm": 0.412109375, + "learning_rate": 0.0018067760248176193, + "loss": 0.1143, + "step": 21189 + }, + { + "epoch": 0.1839393755262541, + "grad_norm": 0.1748046875, + "learning_rate": 0.0018067575982878294, + "loss": 0.0913, + "step": 21190 + }, + { + "epoch": 0.18394805600645828, + "grad_norm": 0.2392578125, + "learning_rate": 0.0018067391709851407, + "loss": 0.0889, + "step": 21191 + }, + { + "epoch": 0.18395673648666244, + "grad_norm": 0.40234375, + "learning_rate": 0.001806720742909572, + "loss": 0.1182, + "step": 21192 + }, + { + "epoch": 0.1839654169668666, + "grad_norm": 0.306640625, + "learning_rate": 0.0018067023140611444, + "loss": 0.1191, + "step": 21193 + }, + { + "epoch": 0.18397409744707077, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018066838844398774, + "loss": 0.084, + "step": 21194 + }, + { + "epoch": 0.18398277792727494, + "grad_norm": 0.7890625, + "learning_rate": 0.0018066654540457923, + "loss": 0.207, + "step": 21195 + }, + { + "epoch": 0.1839914584074791, + "grad_norm": 0.23828125, + "learning_rate": 0.0018066470228789077, + "loss": 0.1309, + "step": 21196 + }, + { + "epoch": 0.18400013888768327, + "grad_norm": 0.265625, + "learning_rate": 0.001806628590939245, + "loss": 0.1187, + "step": 21197 + }, + { + "epoch": 0.18400881936788743, + "grad_norm": 0.66015625, + "learning_rate": 0.0018066101582268237, + "loss": 0.1836, + "step": 21198 + }, + { + "epoch": 0.1840174998480916, + "grad_norm": 0.41015625, + "learning_rate": 0.0018065917247416644, + "loss": 0.125, + "step": 21199 + }, + { + "epoch": 0.18402618032829576, + "grad_norm": 0.25390625, + "learning_rate": 0.0018065732904837865, + "loss": 0.0938, + "step": 21200 + }, + { + "epoch": 0.18403486080849993, + "grad_norm": 0.07080078125, + "learning_rate": 0.001806554855453211, + "loss": 0.0938, + "step": 21201 + }, + { + "epoch": 0.1840435412887041, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018065364196499577, + "loss": 0.0869, + "step": 21202 + }, + { + "epoch": 0.18405222176890826, + "grad_norm": 0.302734375, + "learning_rate": 0.0018065179830740465, + "loss": 0.1006, + "step": 21203 + }, + { + "epoch": 0.18406090224911242, + "grad_norm": 0.09326171875, + "learning_rate": 0.0018064995457254982, + "loss": 0.1221, + "step": 21204 + }, + { + "epoch": 0.1840695827293166, + "grad_norm": 0.314453125, + "learning_rate": 0.0018064811076043325, + "loss": 0.0825, + "step": 21205 + }, + { + "epoch": 0.18407826320952075, + "grad_norm": 0.1142578125, + "learning_rate": 0.0018064626687105698, + "loss": 0.1133, + "step": 21206 + }, + { + "epoch": 0.18408694368972492, + "grad_norm": 0.251953125, + "learning_rate": 0.00180644422904423, + "loss": 0.1226, + "step": 21207 + }, + { + "epoch": 0.18409562416992908, + "grad_norm": 0.2470703125, + "learning_rate": 0.0018064257886053336, + "loss": 0.0947, + "step": 21208 + }, + { + "epoch": 0.18410430465013325, + "grad_norm": 0.0634765625, + "learning_rate": 0.0018064073473939004, + "loss": 0.1064, + "step": 21209 + }, + { + "epoch": 0.18411298513033741, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018063889054099507, + "loss": 0.1318, + "step": 21210 + }, + { + "epoch": 0.18412166561054158, + "grad_norm": 1.28125, + "learning_rate": 0.001806370462653505, + "loss": 0.1367, + "step": 21211 + }, + { + "epoch": 0.18413034609074574, + "grad_norm": 0.5859375, + "learning_rate": 0.0018063520191245828, + "loss": 0.1572, + "step": 21212 + }, + { + "epoch": 0.1841390265709499, + "grad_norm": 0.1259765625, + "learning_rate": 0.0018063335748232047, + "loss": 0.1006, + "step": 21213 + }, + { + "epoch": 0.18414770705115407, + "grad_norm": 0.244140625, + "learning_rate": 0.0018063151297493907, + "loss": 0.0854, + "step": 21214 + }, + { + "epoch": 0.18415638753135824, + "grad_norm": 0.42578125, + "learning_rate": 0.0018062966839031613, + "loss": 0.0879, + "step": 21215 + }, + { + "epoch": 0.1841650680115624, + "grad_norm": 0.578125, + "learning_rate": 0.0018062782372845366, + "loss": 0.1084, + "step": 21216 + }, + { + "epoch": 0.18417374849176657, + "grad_norm": 0.5859375, + "learning_rate": 0.0018062597898935364, + "loss": 0.1074, + "step": 21217 + }, + { + "epoch": 0.18418242897197074, + "grad_norm": 0.65625, + "learning_rate": 0.0018062413417301811, + "loss": 0.1113, + "step": 21218 + }, + { + "epoch": 0.1841911094521749, + "grad_norm": 0.26171875, + "learning_rate": 0.001806222892794491, + "loss": 0.1089, + "step": 21219 + }, + { + "epoch": 0.18419978993237907, + "grad_norm": 0.130859375, + "learning_rate": 0.0018062044430864862, + "loss": 0.125, + "step": 21220 + }, + { + "epoch": 0.18420847041258323, + "grad_norm": 0.8359375, + "learning_rate": 0.0018061859926061866, + "loss": 0.0908, + "step": 21221 + }, + { + "epoch": 0.1842171508927874, + "grad_norm": 0.484375, + "learning_rate": 0.0018061675413536128, + "loss": 0.1011, + "step": 21222 + }, + { + "epoch": 0.18422583137299156, + "grad_norm": 0.07958984375, + "learning_rate": 0.0018061490893287848, + "loss": 0.083, + "step": 21223 + }, + { + "epoch": 0.18423451185319573, + "grad_norm": 0.09912109375, + "learning_rate": 0.0018061306365317227, + "loss": 0.0986, + "step": 21224 + }, + { + "epoch": 0.1842431923333999, + "grad_norm": 0.318359375, + "learning_rate": 0.0018061121829624468, + "loss": 0.0732, + "step": 21225 + }, + { + "epoch": 0.18425187281360406, + "grad_norm": 0.51171875, + "learning_rate": 0.001806093728620977, + "loss": 0.1484, + "step": 21226 + }, + { + "epoch": 0.18426055329380822, + "grad_norm": 0.095703125, + "learning_rate": 0.001806075273507334, + "loss": 0.0864, + "step": 21227 + }, + { + "epoch": 0.1842692337740124, + "grad_norm": 0.279296875, + "learning_rate": 0.0018060568176215376, + "loss": 0.1118, + "step": 21228 + }, + { + "epoch": 0.18427791425421655, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018060383609636082, + "loss": 0.0791, + "step": 21229 + }, + { + "epoch": 0.18428659473442072, + "grad_norm": 1.265625, + "learning_rate": 0.0018060199035335653, + "loss": 0.127, + "step": 21230 + }, + { + "epoch": 0.18429527521462488, + "grad_norm": 0.263671875, + "learning_rate": 0.0018060014453314302, + "loss": 0.0752, + "step": 21231 + }, + { + "epoch": 0.18430395569482905, + "grad_norm": 0.482421875, + "learning_rate": 0.0018059829863572223, + "loss": 0.1055, + "step": 21232 + }, + { + "epoch": 0.1843126361750332, + "grad_norm": 0.099609375, + "learning_rate": 0.0018059645266109623, + "loss": 0.1104, + "step": 21233 + }, + { + "epoch": 0.18432131665523738, + "grad_norm": 0.09619140625, + "learning_rate": 0.0018059460660926699, + "loss": 0.1367, + "step": 21234 + }, + { + "epoch": 0.18432999713544154, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018059276048023654, + "loss": 0.1045, + "step": 21235 + }, + { + "epoch": 0.1843386776156457, + "grad_norm": 0.3046875, + "learning_rate": 0.0018059091427400692, + "loss": 0.1279, + "step": 21236 + }, + { + "epoch": 0.18434735809584987, + "grad_norm": 0.24609375, + "learning_rate": 0.0018058906799058012, + "loss": 0.1318, + "step": 21237 + }, + { + "epoch": 0.18435603857605404, + "grad_norm": 0.30078125, + "learning_rate": 0.001805872216299582, + "loss": 0.0889, + "step": 21238 + }, + { + "epoch": 0.1843647190562582, + "grad_norm": 0.306640625, + "learning_rate": 0.0018058537519214314, + "loss": 0.1348, + "step": 21239 + }, + { + "epoch": 0.18437339953646237, + "grad_norm": 0.09033203125, + "learning_rate": 0.0018058352867713697, + "loss": 0.0996, + "step": 21240 + }, + { + "epoch": 0.18438208001666653, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018058168208494173, + "loss": 0.1157, + "step": 21241 + }, + { + "epoch": 0.1843907604968707, + "grad_norm": 0.490234375, + "learning_rate": 0.001805798354155594, + "loss": 0.103, + "step": 21242 + }, + { + "epoch": 0.18439944097707486, + "grad_norm": 0.4765625, + "learning_rate": 0.0018057798866899205, + "loss": 0.0708, + "step": 21243 + }, + { + "epoch": 0.18440812145727903, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018057614184524165, + "loss": 0.1016, + "step": 21244 + }, + { + "epoch": 0.1844168019374832, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018057429494431026, + "loss": 0.1299, + "step": 21245 + }, + { + "epoch": 0.18442548241768736, + "grad_norm": 0.333984375, + "learning_rate": 0.0018057244796619986, + "loss": 0.1348, + "step": 21246 + }, + { + "epoch": 0.18443416289789152, + "grad_norm": 0.10009765625, + "learning_rate": 0.0018057060091091252, + "loss": 0.1387, + "step": 21247 + }, + { + "epoch": 0.1844428433780957, + "grad_norm": 0.0947265625, + "learning_rate": 0.0018056875377845019, + "loss": 0.1045, + "step": 21248 + }, + { + "epoch": 0.18445152385829985, + "grad_norm": 0.13671875, + "learning_rate": 0.0018056690656881496, + "loss": 0.0864, + "step": 21249 + }, + { + "epoch": 0.18446020433850402, + "grad_norm": 0.10400390625, + "learning_rate": 0.001805650592820088, + "loss": 0.1299, + "step": 21250 + }, + { + "epoch": 0.18446888481870818, + "grad_norm": 0.91796875, + "learning_rate": 0.0018056321191803378, + "loss": 0.1719, + "step": 21251 + }, + { + "epoch": 0.18447756529891235, + "grad_norm": 0.177734375, + "learning_rate": 0.0018056136447689188, + "loss": 0.1045, + "step": 21252 + }, + { + "epoch": 0.18448624577911651, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018055951695858513, + "loss": 0.1436, + "step": 21253 + }, + { + "epoch": 0.18449492625932065, + "grad_norm": 0.49609375, + "learning_rate": 0.0018055766936311555, + "loss": 0.127, + "step": 21254 + }, + { + "epoch": 0.18450360673952482, + "grad_norm": 0.255859375, + "learning_rate": 0.0018055582169048516, + "loss": 0.1079, + "step": 21255 + }, + { + "epoch": 0.18451228721972898, + "grad_norm": 0.404296875, + "learning_rate": 0.00180553973940696, + "loss": 0.1396, + "step": 21256 + }, + { + "epoch": 0.18452096769993315, + "grad_norm": 0.96484375, + "learning_rate": 0.0018055212611375005, + "loss": 0.166, + "step": 21257 + }, + { + "epoch": 0.1845296481801373, + "grad_norm": 0.33203125, + "learning_rate": 0.001805502782096494, + "loss": 0.0815, + "step": 21258 + }, + { + "epoch": 0.18453832866034148, + "grad_norm": 0.55078125, + "learning_rate": 0.0018054843022839598, + "loss": 0.0786, + "step": 21259 + }, + { + "epoch": 0.18454700914054564, + "grad_norm": 0.32421875, + "learning_rate": 0.0018054658216999186, + "loss": 0.1064, + "step": 21260 + }, + { + "epoch": 0.1845556896207498, + "grad_norm": 0.2255859375, + "learning_rate": 0.0018054473403443906, + "loss": 0.1025, + "step": 21261 + }, + { + "epoch": 0.18456437010095397, + "grad_norm": 0.31640625, + "learning_rate": 0.0018054288582173964, + "loss": 0.1367, + "step": 21262 + }, + { + "epoch": 0.18457305058115814, + "grad_norm": 0.07421875, + "learning_rate": 0.0018054103753189553, + "loss": 0.1226, + "step": 21263 + }, + { + "epoch": 0.1845817310613623, + "grad_norm": 0.482421875, + "learning_rate": 0.0018053918916490883, + "loss": 0.0864, + "step": 21264 + }, + { + "epoch": 0.18459041154156647, + "grad_norm": 0.72265625, + "learning_rate": 0.0018053734072078149, + "loss": 0.1055, + "step": 21265 + }, + { + "epoch": 0.18459909202177063, + "grad_norm": 0.85546875, + "learning_rate": 0.0018053549219951561, + "loss": 0.1836, + "step": 21266 + }, + { + "epoch": 0.1846077725019748, + "grad_norm": 0.1943359375, + "learning_rate": 0.0018053364360111318, + "loss": 0.1055, + "step": 21267 + }, + { + "epoch": 0.18461645298217896, + "grad_norm": 0.322265625, + "learning_rate": 0.0018053179492557622, + "loss": 0.105, + "step": 21268 + }, + { + "epoch": 0.18462513346238313, + "grad_norm": 0.39453125, + "learning_rate": 0.0018052994617290673, + "loss": 0.1025, + "step": 21269 + }, + { + "epoch": 0.1846338139425873, + "grad_norm": 1.0078125, + "learning_rate": 0.0018052809734310676, + "loss": 0.125, + "step": 21270 + }, + { + "epoch": 0.18464249442279146, + "grad_norm": 0.33984375, + "learning_rate": 0.001805262484361783, + "loss": 0.1182, + "step": 21271 + }, + { + "epoch": 0.18465117490299562, + "grad_norm": 0.234375, + "learning_rate": 0.0018052439945212345, + "loss": 0.1055, + "step": 21272 + }, + { + "epoch": 0.1846598553831998, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018052255039094412, + "loss": 0.1133, + "step": 21273 + }, + { + "epoch": 0.18466853586340395, + "grad_norm": 0.244140625, + "learning_rate": 0.001805207012526424, + "loss": 0.0859, + "step": 21274 + }, + { + "epoch": 0.18467721634360812, + "grad_norm": 0.1201171875, + "learning_rate": 0.0018051885203722033, + "loss": 0.2168, + "step": 21275 + }, + { + "epoch": 0.18468589682381228, + "grad_norm": 0.51953125, + "learning_rate": 0.0018051700274467986, + "loss": 0.1084, + "step": 21276 + }, + { + "epoch": 0.18469457730401645, + "grad_norm": 0.2216796875, + "learning_rate": 0.001805151533750231, + "loss": 0.1504, + "step": 21277 + }, + { + "epoch": 0.18470325778422061, + "grad_norm": 0.8046875, + "learning_rate": 0.00180513303928252, + "loss": 0.1289, + "step": 21278 + }, + { + "epoch": 0.18471193826442478, + "grad_norm": 0.1328125, + "learning_rate": 0.0018051145440436862, + "loss": 0.127, + "step": 21279 + }, + { + "epoch": 0.18472061874462894, + "grad_norm": 0.2490234375, + "learning_rate": 0.0018050960480337495, + "loss": 0.104, + "step": 21280 + }, + { + "epoch": 0.1847292992248331, + "grad_norm": 0.83984375, + "learning_rate": 0.0018050775512527307, + "loss": 0.1484, + "step": 21281 + }, + { + "epoch": 0.18473797970503728, + "grad_norm": 1.375, + "learning_rate": 0.0018050590537006495, + "loss": 0.1523, + "step": 21282 + }, + { + "epoch": 0.18474666018524144, + "grad_norm": 0.7109375, + "learning_rate": 0.0018050405553775265, + "loss": 0.1221, + "step": 21283 + }, + { + "epoch": 0.1847553406654456, + "grad_norm": 0.357421875, + "learning_rate": 0.0018050220562833815, + "loss": 0.1021, + "step": 21284 + }, + { + "epoch": 0.18476402114564977, + "grad_norm": 0.09716796875, + "learning_rate": 0.001805003556418235, + "loss": 0.1396, + "step": 21285 + }, + { + "epoch": 0.18477270162585394, + "grad_norm": 0.2119140625, + "learning_rate": 0.0018049850557821073, + "loss": 0.0928, + "step": 21286 + }, + { + "epoch": 0.1847813821060581, + "grad_norm": 0.2138671875, + "learning_rate": 0.0018049665543750184, + "loss": 0.1465, + "step": 21287 + }, + { + "epoch": 0.18479006258626227, + "grad_norm": 0.1044921875, + "learning_rate": 0.0018049480521969887, + "loss": 0.1162, + "step": 21288 + }, + { + "epoch": 0.18479874306646643, + "grad_norm": 0.142578125, + "learning_rate": 0.0018049295492480384, + "loss": 0.084, + "step": 21289 + }, + { + "epoch": 0.1848074235466706, + "grad_norm": 0.65234375, + "learning_rate": 0.001804911045528188, + "loss": 0.1426, + "step": 21290 + }, + { + "epoch": 0.18481610402687476, + "grad_norm": 0.12060546875, + "learning_rate": 0.0018048925410374572, + "loss": 0.0879, + "step": 21291 + }, + { + "epoch": 0.18482478450707893, + "grad_norm": 0.474609375, + "learning_rate": 0.0018048740357758666, + "loss": 0.0996, + "step": 21292 + }, + { + "epoch": 0.1848334649872831, + "grad_norm": 0.5859375, + "learning_rate": 0.0018048555297434366, + "loss": 0.0977, + "step": 21293 + }, + { + "epoch": 0.18484214546748726, + "grad_norm": 0.11669921875, + "learning_rate": 0.0018048370229401868, + "loss": 0.0698, + "step": 21294 + }, + { + "epoch": 0.18485082594769142, + "grad_norm": 0.431640625, + "learning_rate": 0.0018048185153661377, + "loss": 0.1484, + "step": 21295 + }, + { + "epoch": 0.1848595064278956, + "grad_norm": 0.1669921875, + "learning_rate": 0.0018048000070213102, + "loss": 0.085, + "step": 21296 + }, + { + "epoch": 0.18486818690809975, + "grad_norm": 0.35546875, + "learning_rate": 0.0018047814979057235, + "loss": 0.1025, + "step": 21297 + }, + { + "epoch": 0.18487686738830392, + "grad_norm": 0.208984375, + "learning_rate": 0.0018047629880193986, + "loss": 0.1523, + "step": 21298 + }, + { + "epoch": 0.18488554786850808, + "grad_norm": 0.1005859375, + "learning_rate": 0.0018047444773623556, + "loss": 0.1406, + "step": 21299 + }, + { + "epoch": 0.18489422834871225, + "grad_norm": 0.1796875, + "learning_rate": 0.0018047259659346143, + "loss": 0.1279, + "step": 21300 + }, + { + "epoch": 0.1849029088289164, + "grad_norm": 0.251953125, + "learning_rate": 0.0018047074537361953, + "loss": 0.1348, + "step": 21301 + }, + { + "epoch": 0.18491158930912058, + "grad_norm": 0.43359375, + "learning_rate": 0.001804688940767119, + "loss": 0.0815, + "step": 21302 + }, + { + "epoch": 0.18492026978932474, + "grad_norm": 0.48828125, + "learning_rate": 0.0018046704270274054, + "loss": 0.084, + "step": 21303 + }, + { + "epoch": 0.1849289502695289, + "grad_norm": 0.54296875, + "learning_rate": 0.001804651912517075, + "loss": 0.123, + "step": 21304 + }, + { + "epoch": 0.18493763074973307, + "grad_norm": 0.490234375, + "learning_rate": 0.0018046333972361476, + "loss": 0.085, + "step": 21305 + }, + { + "epoch": 0.18494631122993724, + "grad_norm": 0.09033203125, + "learning_rate": 0.0018046148811846437, + "loss": 0.1338, + "step": 21306 + }, + { + "epoch": 0.1849549917101414, + "grad_norm": 0.58203125, + "learning_rate": 0.0018045963643625836, + "loss": 0.1152, + "step": 21307 + }, + { + "epoch": 0.18496367219034557, + "grad_norm": 0.58203125, + "learning_rate": 0.0018045778467699874, + "loss": 0.1377, + "step": 21308 + }, + { + "epoch": 0.18497235267054973, + "grad_norm": 0.79296875, + "learning_rate": 0.0018045593284068755, + "loss": 0.1133, + "step": 21309 + }, + { + "epoch": 0.1849810331507539, + "grad_norm": 0.08349609375, + "learning_rate": 0.0018045408092732683, + "loss": 0.1113, + "step": 21310 + }, + { + "epoch": 0.18498971363095806, + "grad_norm": 0.55859375, + "learning_rate": 0.0018045222893691858, + "loss": 0.0938, + "step": 21311 + }, + { + "epoch": 0.18499839411116223, + "grad_norm": 0.26953125, + "learning_rate": 0.0018045037686946481, + "loss": 0.0771, + "step": 21312 + }, + { + "epoch": 0.1850070745913664, + "grad_norm": 0.34375, + "learning_rate": 0.0018044852472496758, + "loss": 0.125, + "step": 21313 + }, + { + "epoch": 0.18501575507157056, + "grad_norm": 0.181640625, + "learning_rate": 0.001804466725034289, + "loss": 0.1582, + "step": 21314 + }, + { + "epoch": 0.18502443555177472, + "grad_norm": 0.130859375, + "learning_rate": 0.0018044482020485083, + "loss": 0.1348, + "step": 21315 + }, + { + "epoch": 0.1850331160319789, + "grad_norm": 0.90234375, + "learning_rate": 0.0018044296782923532, + "loss": 0.208, + "step": 21316 + }, + { + "epoch": 0.18504179651218305, + "grad_norm": 0.28125, + "learning_rate": 0.0018044111537658446, + "loss": 0.0962, + "step": 21317 + }, + { + "epoch": 0.18505047699238722, + "grad_norm": 0.515625, + "learning_rate": 0.0018043926284690025, + "loss": 0.063, + "step": 21318 + }, + { + "epoch": 0.18505915747259138, + "grad_norm": 0.0869140625, + "learning_rate": 0.001804374102401847, + "loss": 0.0791, + "step": 21319 + }, + { + "epoch": 0.18506783795279555, + "grad_norm": 0.330078125, + "learning_rate": 0.0018043555755643988, + "loss": 0.1084, + "step": 21320 + }, + { + "epoch": 0.18507651843299971, + "grad_norm": 0.765625, + "learning_rate": 0.001804337047956678, + "loss": 0.1064, + "step": 21321 + }, + { + "epoch": 0.18508519891320388, + "grad_norm": 0.173828125, + "learning_rate": 0.0018043185195787043, + "loss": 0.1357, + "step": 21322 + }, + { + "epoch": 0.18509387939340805, + "grad_norm": 0.10986328125, + "learning_rate": 0.0018042999904304988, + "loss": 0.1172, + "step": 21323 + }, + { + "epoch": 0.1851025598736122, + "grad_norm": 0.0859375, + "learning_rate": 0.0018042814605120816, + "loss": 0.1055, + "step": 21324 + }, + { + "epoch": 0.18511124035381638, + "grad_norm": 0.1416015625, + "learning_rate": 0.0018042629298234728, + "loss": 0.0967, + "step": 21325 + }, + { + "epoch": 0.18511992083402054, + "grad_norm": 0.92578125, + "learning_rate": 0.0018042443983646921, + "loss": 0.1113, + "step": 21326 + }, + { + "epoch": 0.1851286013142247, + "grad_norm": 0.546875, + "learning_rate": 0.0018042258661357608, + "loss": 0.1406, + "step": 21327 + }, + { + "epoch": 0.18513728179442887, + "grad_norm": 0.2275390625, + "learning_rate": 0.0018042073331366985, + "loss": 0.1021, + "step": 21328 + }, + { + "epoch": 0.18514596227463304, + "grad_norm": 0.2177734375, + "learning_rate": 0.0018041887993675257, + "loss": 0.0898, + "step": 21329 + }, + { + "epoch": 0.1851546427548372, + "grad_norm": 0.0791015625, + "learning_rate": 0.0018041702648282625, + "loss": 0.0859, + "step": 21330 + }, + { + "epoch": 0.18516332323504137, + "grad_norm": 0.0986328125, + "learning_rate": 0.0018041517295189291, + "loss": 0.0942, + "step": 21331 + }, + { + "epoch": 0.18517200371524553, + "grad_norm": 0.3046875, + "learning_rate": 0.0018041331934395464, + "loss": 0.1201, + "step": 21332 + }, + { + "epoch": 0.1851806841954497, + "grad_norm": 0.27734375, + "learning_rate": 0.0018041146565901342, + "loss": 0.1152, + "step": 21333 + }, + { + "epoch": 0.18518936467565386, + "grad_norm": 0.2470703125, + "learning_rate": 0.0018040961189707127, + "loss": 0.0869, + "step": 21334 + }, + { + "epoch": 0.18519804515585803, + "grad_norm": 0.43359375, + "learning_rate": 0.0018040775805813019, + "loss": 0.1099, + "step": 21335 + }, + { + "epoch": 0.1852067256360622, + "grad_norm": 0.07373046875, + "learning_rate": 0.001804059041421923, + "loss": 0.0869, + "step": 21336 + }, + { + "epoch": 0.18521540611626636, + "grad_norm": 0.1201171875, + "learning_rate": 0.001804040501492595, + "loss": 0.1289, + "step": 21337 + }, + { + "epoch": 0.18522408659647052, + "grad_norm": 0.30078125, + "learning_rate": 0.0018040219607933395, + "loss": 0.0986, + "step": 21338 + }, + { + "epoch": 0.1852327670766747, + "grad_norm": 0.3046875, + "learning_rate": 0.0018040034193241758, + "loss": 0.1582, + "step": 21339 + }, + { + "epoch": 0.18524144755687885, + "grad_norm": 0.0869140625, + "learning_rate": 0.0018039848770851248, + "loss": 0.1113, + "step": 21340 + }, + { + "epoch": 0.18525012803708302, + "grad_norm": 0.13671875, + "learning_rate": 0.0018039663340762063, + "loss": 0.0801, + "step": 21341 + }, + { + "epoch": 0.18525880851728718, + "grad_norm": 0.1318359375, + "learning_rate": 0.0018039477902974407, + "loss": 0.1152, + "step": 21342 + }, + { + "epoch": 0.18526748899749135, + "grad_norm": 0.11328125, + "learning_rate": 0.001803929245748849, + "loss": 0.1235, + "step": 21343 + }, + { + "epoch": 0.1852761694776955, + "grad_norm": 0.34765625, + "learning_rate": 0.0018039107004304506, + "loss": 0.1182, + "step": 21344 + }, + { + "epoch": 0.18528484995789968, + "grad_norm": 0.115234375, + "learning_rate": 0.0018038921543422658, + "loss": 0.1064, + "step": 21345 + }, + { + "epoch": 0.18529353043810384, + "grad_norm": 0.087890625, + "learning_rate": 0.001803873607484315, + "loss": 0.0918, + "step": 21346 + }, + { + "epoch": 0.185302210918308, + "grad_norm": 0.083984375, + "learning_rate": 0.0018038550598566186, + "loss": 0.1196, + "step": 21347 + }, + { + "epoch": 0.18531089139851217, + "grad_norm": 0.470703125, + "learning_rate": 0.0018038365114591973, + "loss": 0.103, + "step": 21348 + }, + { + "epoch": 0.18531957187871634, + "grad_norm": 0.384765625, + "learning_rate": 0.0018038179622920706, + "loss": 0.127, + "step": 21349 + }, + { + "epoch": 0.1853282523589205, + "grad_norm": 0.3671875, + "learning_rate": 0.0018037994123552593, + "loss": 0.0742, + "step": 21350 + }, + { + "epoch": 0.18533693283912467, + "grad_norm": 0.15234375, + "learning_rate": 0.0018037808616487834, + "loss": 0.126, + "step": 21351 + }, + { + "epoch": 0.18534561331932883, + "grad_norm": 0.6015625, + "learning_rate": 0.0018037623101726634, + "loss": 0.1514, + "step": 21352 + }, + { + "epoch": 0.185354293799533, + "grad_norm": 0.1083984375, + "learning_rate": 0.0018037437579269198, + "loss": 0.1211, + "step": 21353 + }, + { + "epoch": 0.18536297427973716, + "grad_norm": 0.486328125, + "learning_rate": 0.0018037252049115722, + "loss": 0.1064, + "step": 21354 + }, + { + "epoch": 0.18537165475994133, + "grad_norm": 0.6640625, + "learning_rate": 0.0018037066511266413, + "loss": 0.1162, + "step": 21355 + }, + { + "epoch": 0.1853803352401455, + "grad_norm": 0.447265625, + "learning_rate": 0.0018036880965721475, + "loss": 0.0879, + "step": 21356 + }, + { + "epoch": 0.18538901572034966, + "grad_norm": 0.2275390625, + "learning_rate": 0.0018036695412481111, + "loss": 0.1357, + "step": 21357 + }, + { + "epoch": 0.18539769620055382, + "grad_norm": 0.1201171875, + "learning_rate": 0.001803650985154552, + "loss": 0.103, + "step": 21358 + }, + { + "epoch": 0.185406376680758, + "grad_norm": 0.80078125, + "learning_rate": 0.0018036324282914914, + "loss": 0.1221, + "step": 21359 + }, + { + "epoch": 0.18541505716096215, + "grad_norm": 0.3046875, + "learning_rate": 0.0018036138706589483, + "loss": 0.1187, + "step": 21360 + }, + { + "epoch": 0.18542373764116632, + "grad_norm": 0.1533203125, + "learning_rate": 0.0018035953122569437, + "loss": 0.1138, + "step": 21361 + }, + { + "epoch": 0.18543241812137048, + "grad_norm": 0.345703125, + "learning_rate": 0.0018035767530854981, + "loss": 0.1406, + "step": 21362 + }, + { + "epoch": 0.18544109860157465, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018035581931446312, + "loss": 0.084, + "step": 21363 + }, + { + "epoch": 0.18544977908177881, + "grad_norm": 0.1279296875, + "learning_rate": 0.001803539632434364, + "loss": 0.1055, + "step": 21364 + }, + { + "epoch": 0.18545845956198298, + "grad_norm": 1.328125, + "learning_rate": 0.0018035210709547158, + "loss": 0.127, + "step": 21365 + }, + { + "epoch": 0.18546714004218715, + "grad_norm": 0.216796875, + "learning_rate": 0.0018035025087057082, + "loss": 0.0942, + "step": 21366 + }, + { + "epoch": 0.1854758205223913, + "grad_norm": 0.154296875, + "learning_rate": 0.0018034839456873607, + "loss": 0.1025, + "step": 21367 + }, + { + "epoch": 0.18548450100259548, + "grad_norm": 0.43359375, + "learning_rate": 0.0018034653818996935, + "loss": 0.1064, + "step": 21368 + }, + { + "epoch": 0.18549318148279964, + "grad_norm": 0.220703125, + "learning_rate": 0.0018034468173427271, + "loss": 0.1104, + "step": 21369 + }, + { + "epoch": 0.1855018619630038, + "grad_norm": 0.228515625, + "learning_rate": 0.0018034282520164823, + "loss": 0.1152, + "step": 21370 + }, + { + "epoch": 0.18551054244320797, + "grad_norm": 0.275390625, + "learning_rate": 0.0018034096859209784, + "loss": 0.0708, + "step": 21371 + }, + { + "epoch": 0.18551922292341214, + "grad_norm": 0.32421875, + "learning_rate": 0.0018033911190562365, + "loss": 0.0908, + "step": 21372 + }, + { + "epoch": 0.1855279034036163, + "grad_norm": 0.396484375, + "learning_rate": 0.0018033725514222767, + "loss": 0.1016, + "step": 21373 + }, + { + "epoch": 0.18553658388382047, + "grad_norm": 0.275390625, + "learning_rate": 0.001803353983019119, + "loss": 0.1533, + "step": 21374 + }, + { + "epoch": 0.18554526436402463, + "grad_norm": 0.2197265625, + "learning_rate": 0.0018033354138467842, + "loss": 0.1387, + "step": 21375 + }, + { + "epoch": 0.1855539448442288, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018033168439052921, + "loss": 0.0962, + "step": 21376 + }, + { + "epoch": 0.18556262532443293, + "grad_norm": 0.09521484375, + "learning_rate": 0.0018032982731946633, + "loss": 0.1299, + "step": 21377 + }, + { + "epoch": 0.1855713058046371, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018032797017149185, + "loss": 0.1309, + "step": 21378 + }, + { + "epoch": 0.18557998628484126, + "grad_norm": 0.119140625, + "learning_rate": 0.0018032611294660772, + "loss": 0.1357, + "step": 21379 + }, + { + "epoch": 0.18558866676504543, + "grad_norm": 0.1748046875, + "learning_rate": 0.00180324255644816, + "loss": 0.1162, + "step": 21380 + }, + { + "epoch": 0.1855973472452496, + "grad_norm": 0.3046875, + "learning_rate": 0.0018032239826611875, + "loss": 0.124, + "step": 21381 + }, + { + "epoch": 0.18560602772545376, + "grad_norm": 0.361328125, + "learning_rate": 0.0018032054081051799, + "loss": 0.1221, + "step": 21382 + }, + { + "epoch": 0.18561470820565792, + "grad_norm": 0.453125, + "learning_rate": 0.0018031868327801574, + "loss": 0.082, + "step": 21383 + }, + { + "epoch": 0.1856233886858621, + "grad_norm": 0.283203125, + "learning_rate": 0.00180316825668614, + "loss": 0.1084, + "step": 21384 + }, + { + "epoch": 0.18563206916606625, + "grad_norm": 0.15234375, + "learning_rate": 0.0018031496798231487, + "loss": 0.1216, + "step": 21385 + }, + { + "epoch": 0.18564074964627042, + "grad_norm": 0.11328125, + "learning_rate": 0.0018031311021912031, + "loss": 0.0889, + "step": 21386 + }, + { + "epoch": 0.18564943012647458, + "grad_norm": 0.142578125, + "learning_rate": 0.0018031125237903244, + "loss": 0.0908, + "step": 21387 + }, + { + "epoch": 0.18565811060667875, + "grad_norm": 0.296875, + "learning_rate": 0.0018030939446205322, + "loss": 0.1221, + "step": 21388 + }, + { + "epoch": 0.18566679108688292, + "grad_norm": 0.77734375, + "learning_rate": 0.001803075364681847, + "loss": 0.0996, + "step": 21389 + }, + { + "epoch": 0.18567547156708708, + "grad_norm": 0.4375, + "learning_rate": 0.001803056783974289, + "loss": 0.1069, + "step": 21390 + }, + { + "epoch": 0.18568415204729125, + "grad_norm": 0.578125, + "learning_rate": 0.001803038202497879, + "loss": 0.1074, + "step": 21391 + }, + { + "epoch": 0.1856928325274954, + "grad_norm": 0.34765625, + "learning_rate": 0.001803019620252637, + "loss": 0.1162, + "step": 21392 + }, + { + "epoch": 0.18570151300769958, + "grad_norm": 0.30078125, + "learning_rate": 0.001803001037238583, + "loss": 0.1094, + "step": 21393 + }, + { + "epoch": 0.18571019348790374, + "grad_norm": 0.79296875, + "learning_rate": 0.0018029824534557377, + "loss": 0.0981, + "step": 21394 + }, + { + "epoch": 0.1857188739681079, + "grad_norm": 0.1806640625, + "learning_rate": 0.0018029638689041215, + "loss": 0.1201, + "step": 21395 + }, + { + "epoch": 0.18572755444831207, + "grad_norm": 0.298828125, + "learning_rate": 0.0018029452835837546, + "loss": 0.1162, + "step": 21396 + }, + { + "epoch": 0.18573623492851624, + "grad_norm": 0.6875, + "learning_rate": 0.001802926697494657, + "loss": 0.1084, + "step": 21397 + }, + { + "epoch": 0.1857449154087204, + "grad_norm": 0.306640625, + "learning_rate": 0.0018029081106368495, + "loss": 0.1152, + "step": 21398 + }, + { + "epoch": 0.18575359588892457, + "grad_norm": 0.259765625, + "learning_rate": 0.001802889523010352, + "loss": 0.1045, + "step": 21399 + }, + { + "epoch": 0.18576227636912873, + "grad_norm": 0.48828125, + "learning_rate": 0.0018028709346151856, + "loss": 0.1074, + "step": 21400 + }, + { + "epoch": 0.1857709568493329, + "grad_norm": 0.1279296875, + "learning_rate": 0.0018028523454513697, + "loss": 0.1416, + "step": 21401 + }, + { + "epoch": 0.18577963732953706, + "grad_norm": 0.6328125, + "learning_rate": 0.0018028337555189252, + "loss": 0.1309, + "step": 21402 + }, + { + "epoch": 0.18578831780974123, + "grad_norm": 0.2099609375, + "learning_rate": 0.0018028151648178722, + "loss": 0.0947, + "step": 21403 + }, + { + "epoch": 0.1857969982899454, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018027965733482313, + "loss": 0.0908, + "step": 21404 + }, + { + "epoch": 0.18580567877014956, + "grad_norm": 0.126953125, + "learning_rate": 0.0018027779811100224, + "loss": 0.125, + "step": 21405 + }, + { + "epoch": 0.18581435925035372, + "grad_norm": 0.6796875, + "learning_rate": 0.001802759388103266, + "loss": 0.1553, + "step": 21406 + }, + { + "epoch": 0.1858230397305579, + "grad_norm": 0.26953125, + "learning_rate": 0.0018027407943279827, + "loss": 0.1221, + "step": 21407 + }, + { + "epoch": 0.18583172021076205, + "grad_norm": 0.2578125, + "learning_rate": 0.0018027221997841925, + "loss": 0.1069, + "step": 21408 + }, + { + "epoch": 0.18584040069096622, + "grad_norm": 0.490234375, + "learning_rate": 0.0018027036044719158, + "loss": 0.166, + "step": 21409 + }, + { + "epoch": 0.18584908117117038, + "grad_norm": 0.66015625, + "learning_rate": 0.0018026850083911732, + "loss": 0.0981, + "step": 21410 + }, + { + "epoch": 0.18585776165137455, + "grad_norm": 0.298828125, + "learning_rate": 0.0018026664115419846, + "loss": 0.106, + "step": 21411 + }, + { + "epoch": 0.1858664421315787, + "grad_norm": 0.1357421875, + "learning_rate": 0.0018026478139243707, + "loss": 0.0898, + "step": 21412 + }, + { + "epoch": 0.18587512261178288, + "grad_norm": 0.75, + "learning_rate": 0.001802629215538352, + "loss": 0.1836, + "step": 21413 + }, + { + "epoch": 0.18588380309198704, + "grad_norm": 0.158203125, + "learning_rate": 0.0018026106163839483, + "loss": 0.1191, + "step": 21414 + }, + { + "epoch": 0.1858924835721912, + "grad_norm": 0.12890625, + "learning_rate": 0.0018025920164611799, + "loss": 0.1426, + "step": 21415 + }, + { + "epoch": 0.18590116405239537, + "grad_norm": 0.4765625, + "learning_rate": 0.0018025734157700675, + "loss": 0.0703, + "step": 21416 + }, + { + "epoch": 0.18590984453259954, + "grad_norm": 0.5078125, + "learning_rate": 0.0018025548143106316, + "loss": 0.125, + "step": 21417 + }, + { + "epoch": 0.1859185250128037, + "grad_norm": 0.84765625, + "learning_rate": 0.0018025362120828922, + "loss": 0.083, + "step": 21418 + }, + { + "epoch": 0.18592720549300787, + "grad_norm": 0.65234375, + "learning_rate": 0.0018025176090868697, + "loss": 0.1152, + "step": 21419 + }, + { + "epoch": 0.18593588597321203, + "grad_norm": 0.3515625, + "learning_rate": 0.0018024990053225848, + "loss": 0.1104, + "step": 21420 + }, + { + "epoch": 0.1859445664534162, + "grad_norm": 0.38671875, + "learning_rate": 0.0018024804007900574, + "loss": 0.1055, + "step": 21421 + }, + { + "epoch": 0.18595324693362036, + "grad_norm": 0.123046875, + "learning_rate": 0.0018024617954893079, + "loss": 0.1201, + "step": 21422 + }, + { + "epoch": 0.18596192741382453, + "grad_norm": 0.330078125, + "learning_rate": 0.0018024431894203565, + "loss": 0.0913, + "step": 21423 + }, + { + "epoch": 0.1859706078940287, + "grad_norm": 0.12158203125, + "learning_rate": 0.0018024245825832245, + "loss": 0.0986, + "step": 21424 + }, + { + "epoch": 0.18597928837423286, + "grad_norm": 0.173828125, + "learning_rate": 0.0018024059749779308, + "loss": 0.1289, + "step": 21425 + }, + { + "epoch": 0.18598796885443702, + "grad_norm": 0.06298828125, + "learning_rate": 0.0018023873666044967, + "loss": 0.0811, + "step": 21426 + }, + { + "epoch": 0.1859966493346412, + "grad_norm": 0.3125, + "learning_rate": 0.0018023687574629426, + "loss": 0.1108, + "step": 21427 + }, + { + "epoch": 0.18600532981484535, + "grad_norm": 0.435546875, + "learning_rate": 0.0018023501475532883, + "loss": 0.084, + "step": 21428 + }, + { + "epoch": 0.18601401029504952, + "grad_norm": 0.44140625, + "learning_rate": 0.0018023315368755543, + "loss": 0.1318, + "step": 21429 + }, + { + "epoch": 0.18602269077525369, + "grad_norm": 0.205078125, + "learning_rate": 0.0018023129254297614, + "loss": 0.1377, + "step": 21430 + }, + { + "epoch": 0.18603137125545785, + "grad_norm": 0.251953125, + "learning_rate": 0.0018022943132159296, + "loss": 0.1006, + "step": 21431 + }, + { + "epoch": 0.18604005173566202, + "grad_norm": 0.2353515625, + "learning_rate": 0.0018022757002340794, + "loss": 0.0957, + "step": 21432 + }, + { + "epoch": 0.18604873221586618, + "grad_norm": 0.44140625, + "learning_rate": 0.0018022570864842304, + "loss": 0.124, + "step": 21433 + }, + { + "epoch": 0.18605741269607035, + "grad_norm": 0.2265625, + "learning_rate": 0.0018022384719664042, + "loss": 0.0967, + "step": 21434 + }, + { + "epoch": 0.1860660931762745, + "grad_norm": 0.51171875, + "learning_rate": 0.0018022198566806204, + "loss": 0.1045, + "step": 21435 + }, + { + "epoch": 0.18607477365647868, + "grad_norm": 0.40625, + "learning_rate": 0.0018022012406268995, + "loss": 0.0962, + "step": 21436 + }, + { + "epoch": 0.18608345413668284, + "grad_norm": 0.12158203125, + "learning_rate": 0.001802182623805262, + "loss": 0.1094, + "step": 21437 + }, + { + "epoch": 0.186092134616887, + "grad_norm": 0.3203125, + "learning_rate": 0.001802164006215728, + "loss": 0.1001, + "step": 21438 + }, + { + "epoch": 0.18610081509709117, + "grad_norm": 0.267578125, + "learning_rate": 0.001802145387858318, + "loss": 0.0742, + "step": 21439 + }, + { + "epoch": 0.18610949557729534, + "grad_norm": 0.5234375, + "learning_rate": 0.0018021267687330523, + "loss": 0.0908, + "step": 21440 + }, + { + "epoch": 0.1861181760574995, + "grad_norm": 0.330078125, + "learning_rate": 0.0018021081488399516, + "loss": 0.1104, + "step": 21441 + }, + { + "epoch": 0.18612685653770367, + "grad_norm": 0.34375, + "learning_rate": 0.0018020895281790357, + "loss": 0.127, + "step": 21442 + }, + { + "epoch": 0.18613553701790783, + "grad_norm": 0.8828125, + "learning_rate": 0.0018020709067503252, + "loss": 0.0967, + "step": 21443 + }, + { + "epoch": 0.186144217498112, + "grad_norm": 0.54296875, + "learning_rate": 0.0018020522845538407, + "loss": 0.1006, + "step": 21444 + }, + { + "epoch": 0.18615289797831616, + "grad_norm": 0.298828125, + "learning_rate": 0.0018020336615896025, + "loss": 0.0908, + "step": 21445 + }, + { + "epoch": 0.18616157845852033, + "grad_norm": 0.2021484375, + "learning_rate": 0.0018020150378576304, + "loss": 0.1387, + "step": 21446 + }, + { + "epoch": 0.1861702589387245, + "grad_norm": 0.36328125, + "learning_rate": 0.0018019964133579455, + "loss": 0.1143, + "step": 21447 + }, + { + "epoch": 0.18617893941892866, + "grad_norm": 0.24609375, + "learning_rate": 0.0018019777880905676, + "loss": 0.0933, + "step": 21448 + }, + { + "epoch": 0.18618761989913282, + "grad_norm": 0.765625, + "learning_rate": 0.0018019591620555176, + "loss": 0.1025, + "step": 21449 + }, + { + "epoch": 0.186196300379337, + "grad_norm": 0.56640625, + "learning_rate": 0.0018019405352528155, + "loss": 0.1289, + "step": 21450 + }, + { + "epoch": 0.18620498085954115, + "grad_norm": 0.69921875, + "learning_rate": 0.001801921907682482, + "loss": 0.1035, + "step": 21451 + }, + { + "epoch": 0.18621366133974532, + "grad_norm": 0.28515625, + "learning_rate": 0.001801903279344537, + "loss": 0.0918, + "step": 21452 + }, + { + "epoch": 0.18622234181994948, + "grad_norm": 0.357421875, + "learning_rate": 0.0018018846502390012, + "loss": 0.1289, + "step": 21453 + }, + { + "epoch": 0.18623102230015365, + "grad_norm": 0.40625, + "learning_rate": 0.0018018660203658948, + "loss": 0.0947, + "step": 21454 + }, + { + "epoch": 0.1862397027803578, + "grad_norm": 0.11474609375, + "learning_rate": 0.0018018473897252383, + "loss": 0.125, + "step": 21455 + }, + { + "epoch": 0.18624838326056198, + "grad_norm": 0.10986328125, + "learning_rate": 0.001801828758317052, + "loss": 0.0679, + "step": 21456 + }, + { + "epoch": 0.18625706374076614, + "grad_norm": 0.380859375, + "learning_rate": 0.0018018101261413567, + "loss": 0.1104, + "step": 21457 + }, + { + "epoch": 0.1862657442209703, + "grad_norm": 0.66015625, + "learning_rate": 0.001801791493198172, + "loss": 0.1562, + "step": 21458 + }, + { + "epoch": 0.18627442470117447, + "grad_norm": 0.498046875, + "learning_rate": 0.0018017728594875186, + "loss": 0.1055, + "step": 21459 + }, + { + "epoch": 0.18628310518137864, + "grad_norm": 0.1923828125, + "learning_rate": 0.0018017542250094174, + "loss": 0.1128, + "step": 21460 + }, + { + "epoch": 0.1862917856615828, + "grad_norm": 0.2314453125, + "learning_rate": 0.001801735589763888, + "loss": 0.1162, + "step": 21461 + }, + { + "epoch": 0.18630046614178697, + "grad_norm": 0.1650390625, + "learning_rate": 0.001801716953750951, + "loss": 0.1592, + "step": 21462 + }, + { + "epoch": 0.18630914662199113, + "grad_norm": 0.376953125, + "learning_rate": 0.0018016983169706268, + "loss": 0.1621, + "step": 21463 + }, + { + "epoch": 0.1863178271021953, + "grad_norm": 0.2578125, + "learning_rate": 0.001801679679422936, + "loss": 0.125, + "step": 21464 + }, + { + "epoch": 0.18632650758239946, + "grad_norm": 0.373046875, + "learning_rate": 0.0018016610411078992, + "loss": 0.1206, + "step": 21465 + }, + { + "epoch": 0.18633518806260363, + "grad_norm": 0.09228515625, + "learning_rate": 0.001801642402025536, + "loss": 0.1016, + "step": 21466 + }, + { + "epoch": 0.1863438685428078, + "grad_norm": 0.375, + "learning_rate": 0.0018016237621758674, + "loss": 0.0825, + "step": 21467 + }, + { + "epoch": 0.18635254902301196, + "grad_norm": 0.61328125, + "learning_rate": 0.0018016051215589133, + "loss": 0.1504, + "step": 21468 + }, + { + "epoch": 0.18636122950321612, + "grad_norm": 0.107421875, + "learning_rate": 0.001801586480174695, + "loss": 0.0918, + "step": 21469 + }, + { + "epoch": 0.1863699099834203, + "grad_norm": 0.0791015625, + "learning_rate": 0.001801567838023232, + "loss": 0.1172, + "step": 21470 + }, + { + "epoch": 0.18637859046362446, + "grad_norm": 0.08837890625, + "learning_rate": 0.0018015491951045447, + "loss": 0.1152, + "step": 21471 + }, + { + "epoch": 0.18638727094382862, + "grad_norm": 0.80859375, + "learning_rate": 0.001801530551418654, + "loss": 0.1318, + "step": 21472 + }, + { + "epoch": 0.18639595142403279, + "grad_norm": 0.46875, + "learning_rate": 0.0018015119069655797, + "loss": 0.1221, + "step": 21473 + }, + { + "epoch": 0.18640463190423695, + "grad_norm": 0.1640625, + "learning_rate": 0.0018014932617453428, + "loss": 0.1191, + "step": 21474 + }, + { + "epoch": 0.18641331238444112, + "grad_norm": 0.484375, + "learning_rate": 0.0018014746157579632, + "loss": 0.0938, + "step": 21475 + }, + { + "epoch": 0.18642199286464528, + "grad_norm": 0.498046875, + "learning_rate": 0.0018014559690034617, + "loss": 0.0728, + "step": 21476 + }, + { + "epoch": 0.18643067334484945, + "grad_norm": 0.365234375, + "learning_rate": 0.0018014373214818584, + "loss": 0.0996, + "step": 21477 + }, + { + "epoch": 0.1864393538250536, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018014186731931741, + "loss": 0.1543, + "step": 21478 + }, + { + "epoch": 0.18644803430525778, + "grad_norm": 0.63671875, + "learning_rate": 0.0018014000241374285, + "loss": 0.1172, + "step": 21479 + }, + { + "epoch": 0.18645671478546194, + "grad_norm": 0.1591796875, + "learning_rate": 0.0018013813743146426, + "loss": 0.105, + "step": 21480 + }, + { + "epoch": 0.1864653952656661, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018013627237248363, + "loss": 0.168, + "step": 21481 + }, + { + "epoch": 0.18647407574587027, + "grad_norm": 1.1328125, + "learning_rate": 0.0018013440723680303, + "loss": 0.1245, + "step": 21482 + }, + { + "epoch": 0.18648275622607444, + "grad_norm": 0.294921875, + "learning_rate": 0.0018013254202442453, + "loss": 0.1338, + "step": 21483 + }, + { + "epoch": 0.1864914367062786, + "grad_norm": 0.271484375, + "learning_rate": 0.001801306767353501, + "loss": 0.1201, + "step": 21484 + }, + { + "epoch": 0.18650011718648277, + "grad_norm": 0.51953125, + "learning_rate": 0.0018012881136958184, + "loss": 0.1377, + "step": 21485 + }, + { + "epoch": 0.18650879766668693, + "grad_norm": 0.1474609375, + "learning_rate": 0.0018012694592712173, + "loss": 0.1631, + "step": 21486 + }, + { + "epoch": 0.1865174781468911, + "grad_norm": 0.1875, + "learning_rate": 0.001801250804079719, + "loss": 0.0977, + "step": 21487 + }, + { + "epoch": 0.18652615862709526, + "grad_norm": 0.23828125, + "learning_rate": 0.0018012321481213427, + "loss": 0.1377, + "step": 21488 + }, + { + "epoch": 0.18653483910729943, + "grad_norm": 0.81640625, + "learning_rate": 0.00180121349139611, + "loss": 0.1777, + "step": 21489 + }, + { + "epoch": 0.1865435195875036, + "grad_norm": 0.33984375, + "learning_rate": 0.0018011948339040405, + "loss": 0.0879, + "step": 21490 + }, + { + "epoch": 0.18655220006770776, + "grad_norm": 0.490234375, + "learning_rate": 0.0018011761756451547, + "loss": 0.0889, + "step": 21491 + }, + { + "epoch": 0.18656088054791192, + "grad_norm": 0.353515625, + "learning_rate": 0.0018011575166194736, + "loss": 0.1133, + "step": 21492 + }, + { + "epoch": 0.1865695610281161, + "grad_norm": 0.189453125, + "learning_rate": 0.001801138856827017, + "loss": 0.1069, + "step": 21493 + }, + { + "epoch": 0.18657824150832025, + "grad_norm": 0.310546875, + "learning_rate": 0.0018011201962678054, + "loss": 0.0967, + "step": 21494 + }, + { + "epoch": 0.18658692198852442, + "grad_norm": 0.91015625, + "learning_rate": 0.0018011015349418595, + "loss": 0.0903, + "step": 21495 + }, + { + "epoch": 0.18659560246872858, + "grad_norm": 0.240234375, + "learning_rate": 0.0018010828728491992, + "loss": 0.0957, + "step": 21496 + }, + { + "epoch": 0.18660428294893275, + "grad_norm": 0.318359375, + "learning_rate": 0.0018010642099898454, + "loss": 0.1406, + "step": 21497 + }, + { + "epoch": 0.1866129634291369, + "grad_norm": 0.53515625, + "learning_rate": 0.0018010455463638182, + "loss": 0.1133, + "step": 21498 + }, + { + "epoch": 0.18662164390934108, + "grad_norm": 0.1494140625, + "learning_rate": 0.0018010268819711383, + "loss": 0.0962, + "step": 21499 + }, + { + "epoch": 0.18663032438954524, + "grad_norm": 0.078125, + "learning_rate": 0.0018010082168118259, + "loss": 0.0947, + "step": 21500 + }, + { + "epoch": 0.18663900486974938, + "grad_norm": 0.56640625, + "learning_rate": 0.0018009895508859014, + "loss": 0.085, + "step": 21501 + }, + { + "epoch": 0.18664768534995355, + "grad_norm": 0.154296875, + "learning_rate": 0.0018009708841933852, + "loss": 0.125, + "step": 21502 + }, + { + "epoch": 0.1866563658301577, + "grad_norm": 0.064453125, + "learning_rate": 0.0018009522167342978, + "loss": 0.1055, + "step": 21503 + }, + { + "epoch": 0.18666504631036188, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018009335485086593, + "loss": 0.0996, + "step": 21504 + }, + { + "epoch": 0.18667372679056604, + "grad_norm": 0.1015625, + "learning_rate": 0.0018009148795164907, + "loss": 0.125, + "step": 21505 + }, + { + "epoch": 0.1866824072707702, + "grad_norm": 0.6640625, + "learning_rate": 0.001800896209757812, + "loss": 0.1416, + "step": 21506 + }, + { + "epoch": 0.18669108775097437, + "grad_norm": 0.142578125, + "learning_rate": 0.001800877539232644, + "loss": 0.0977, + "step": 21507 + }, + { + "epoch": 0.18669976823117854, + "grad_norm": 0.12255859375, + "learning_rate": 0.0018008588679410066, + "loss": 0.1016, + "step": 21508 + }, + { + "epoch": 0.1867084487113827, + "grad_norm": 0.796875, + "learning_rate": 0.0018008401958829207, + "loss": 0.1191, + "step": 21509 + }, + { + "epoch": 0.18671712919158687, + "grad_norm": 0.23828125, + "learning_rate": 0.0018008215230584065, + "loss": 0.1133, + "step": 21510 + }, + { + "epoch": 0.18672580967179103, + "grad_norm": 0.2294921875, + "learning_rate": 0.0018008028494674842, + "loss": 0.1309, + "step": 21511 + }, + { + "epoch": 0.1867344901519952, + "grad_norm": 0.44921875, + "learning_rate": 0.0018007841751101743, + "loss": 0.1055, + "step": 21512 + }, + { + "epoch": 0.18674317063219936, + "grad_norm": 0.142578125, + "learning_rate": 0.0018007654999864977, + "loss": 0.0977, + "step": 21513 + }, + { + "epoch": 0.18675185111240353, + "grad_norm": 0.10595703125, + "learning_rate": 0.0018007468240964744, + "loss": 0.1211, + "step": 21514 + }, + { + "epoch": 0.1867605315926077, + "grad_norm": 0.39453125, + "learning_rate": 0.001800728147440125, + "loss": 0.1318, + "step": 21515 + }, + { + "epoch": 0.18676921207281186, + "grad_norm": 0.2373046875, + "learning_rate": 0.0018007094700174696, + "loss": 0.0864, + "step": 21516 + }, + { + "epoch": 0.18677789255301602, + "grad_norm": 0.92578125, + "learning_rate": 0.001800690791828529, + "loss": 0.1201, + "step": 21517 + }, + { + "epoch": 0.1867865730332202, + "grad_norm": 0.443359375, + "learning_rate": 0.0018006721128733232, + "loss": 0.1035, + "step": 21518 + }, + { + "epoch": 0.18679525351342435, + "grad_norm": 0.392578125, + "learning_rate": 0.0018006534331518728, + "loss": 0.0967, + "step": 21519 + }, + { + "epoch": 0.18680393399362852, + "grad_norm": 0.1767578125, + "learning_rate": 0.001800634752664199, + "loss": 0.1387, + "step": 21520 + }, + { + "epoch": 0.18681261447383268, + "grad_norm": 0.3203125, + "learning_rate": 0.0018006160714103213, + "loss": 0.0938, + "step": 21521 + }, + { + "epoch": 0.18682129495403685, + "grad_norm": 0.267578125, + "learning_rate": 0.00180059738939026, + "loss": 0.126, + "step": 21522 + }, + { + "epoch": 0.186829975434241, + "grad_norm": 0.09716796875, + "learning_rate": 0.0018005787066040363, + "loss": 0.1133, + "step": 21523 + }, + { + "epoch": 0.18683865591444518, + "grad_norm": 0.56640625, + "learning_rate": 0.00180056002305167, + "loss": 0.0845, + "step": 21524 + }, + { + "epoch": 0.18684733639464934, + "grad_norm": 0.2890625, + "learning_rate": 0.0018005413387331822, + "loss": 0.1016, + "step": 21525 + }, + { + "epoch": 0.1868560168748535, + "grad_norm": 0.0966796875, + "learning_rate": 0.0018005226536485926, + "loss": 0.1191, + "step": 21526 + }, + { + "epoch": 0.18686469735505767, + "grad_norm": 0.390625, + "learning_rate": 0.001800503967797922, + "loss": 0.1196, + "step": 21527 + }, + { + "epoch": 0.18687337783526184, + "grad_norm": 0.171875, + "learning_rate": 0.0018004852811811906, + "loss": 0.0884, + "step": 21528 + }, + { + "epoch": 0.186882058315466, + "grad_norm": 0.12060546875, + "learning_rate": 0.0018004665937984194, + "loss": 0.1025, + "step": 21529 + }, + { + "epoch": 0.18689073879567017, + "grad_norm": 0.296875, + "learning_rate": 0.0018004479056496281, + "loss": 0.1021, + "step": 21530 + }, + { + "epoch": 0.18689941927587433, + "grad_norm": 0.16015625, + "learning_rate": 0.0018004292167348376, + "loss": 0.1147, + "step": 21531 + }, + { + "epoch": 0.1869080997560785, + "grad_norm": 0.451171875, + "learning_rate": 0.0018004105270540687, + "loss": 0.0928, + "step": 21532 + }, + { + "epoch": 0.18691678023628266, + "grad_norm": 0.2041015625, + "learning_rate": 0.0018003918366073408, + "loss": 0.103, + "step": 21533 + }, + { + "epoch": 0.18692546071648683, + "grad_norm": 0.73046875, + "learning_rate": 0.0018003731453946751, + "loss": 0.1367, + "step": 21534 + }, + { + "epoch": 0.186934141196691, + "grad_norm": 0.11865234375, + "learning_rate": 0.0018003544534160919, + "loss": 0.1143, + "step": 21535 + }, + { + "epoch": 0.18694282167689516, + "grad_norm": 0.189453125, + "learning_rate": 0.0018003357606716115, + "loss": 0.1074, + "step": 21536 + }, + { + "epoch": 0.18695150215709933, + "grad_norm": 0.1962890625, + "learning_rate": 0.0018003170671612545, + "loss": 0.1172, + "step": 21537 + }, + { + "epoch": 0.1869601826373035, + "grad_norm": 0.58203125, + "learning_rate": 0.001800298372885041, + "loss": 0.1064, + "step": 21538 + }, + { + "epoch": 0.18696886311750766, + "grad_norm": 0.427734375, + "learning_rate": 0.0018002796778429924, + "loss": 0.1191, + "step": 21539 + }, + { + "epoch": 0.18697754359771182, + "grad_norm": 0.1513671875, + "learning_rate": 0.0018002609820351276, + "loss": 0.1152, + "step": 21540 + }, + { + "epoch": 0.18698622407791599, + "grad_norm": 0.169921875, + "learning_rate": 0.0018002422854614686, + "loss": 0.1089, + "step": 21541 + }, + { + "epoch": 0.18699490455812015, + "grad_norm": 0.1220703125, + "learning_rate": 0.001800223588122035, + "loss": 0.1416, + "step": 21542 + }, + { + "epoch": 0.18700358503832432, + "grad_norm": 0.345703125, + "learning_rate": 0.001800204890016847, + "loss": 0.0938, + "step": 21543 + }, + { + "epoch": 0.18701226551852848, + "grad_norm": 0.462890625, + "learning_rate": 0.0018001861911459258, + "loss": 0.1182, + "step": 21544 + }, + { + "epoch": 0.18702094599873265, + "grad_norm": 0.478515625, + "learning_rate": 0.0018001674915092917, + "loss": 0.1074, + "step": 21545 + }, + { + "epoch": 0.1870296264789368, + "grad_norm": 0.34375, + "learning_rate": 0.0018001487911069646, + "loss": 0.1426, + "step": 21546 + }, + { + "epoch": 0.18703830695914098, + "grad_norm": 0.06787109375, + "learning_rate": 0.0018001300899389654, + "loss": 0.0703, + "step": 21547 + }, + { + "epoch": 0.18704698743934514, + "grad_norm": 0.6640625, + "learning_rate": 0.0018001113880053146, + "loss": 0.1118, + "step": 21548 + }, + { + "epoch": 0.1870556679195493, + "grad_norm": 0.0908203125, + "learning_rate": 0.0018000926853060323, + "loss": 0.1143, + "step": 21549 + }, + { + "epoch": 0.18706434839975347, + "grad_norm": 0.171875, + "learning_rate": 0.0018000739818411391, + "loss": 0.0869, + "step": 21550 + }, + { + "epoch": 0.18707302887995764, + "grad_norm": 0.5390625, + "learning_rate": 0.0018000552776106557, + "loss": 0.1104, + "step": 21551 + }, + { + "epoch": 0.1870817093601618, + "grad_norm": 0.466796875, + "learning_rate": 0.0018000365726146023, + "loss": 0.1064, + "step": 21552 + }, + { + "epoch": 0.18709038984036597, + "grad_norm": 0.08544921875, + "learning_rate": 0.0018000178668529994, + "loss": 0.084, + "step": 21553 + }, + { + "epoch": 0.18709907032057013, + "grad_norm": 1.2109375, + "learning_rate": 0.0017999991603258674, + "loss": 0.0986, + "step": 21554 + }, + { + "epoch": 0.1871077508007743, + "grad_norm": 0.1689453125, + "learning_rate": 0.001799980453033227, + "loss": 0.1895, + "step": 21555 + }, + { + "epoch": 0.18711643128097846, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017999617449750984, + "loss": 0.1152, + "step": 21556 + }, + { + "epoch": 0.18712511176118263, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017999430361515022, + "loss": 0.1182, + "step": 21557 + }, + { + "epoch": 0.1871337922413868, + "grad_norm": 0.29296875, + "learning_rate": 0.001799924326562459, + "loss": 0.1133, + "step": 21558 + }, + { + "epoch": 0.18714247272159096, + "grad_norm": 0.333984375, + "learning_rate": 0.0017999056162079888, + "loss": 0.125, + "step": 21559 + }, + { + "epoch": 0.18715115320179512, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017998869050881124, + "loss": 0.1055, + "step": 21560 + }, + { + "epoch": 0.1871598336819993, + "grad_norm": 0.478515625, + "learning_rate": 0.0017998681932028502, + "loss": 0.084, + "step": 21561 + }, + { + "epoch": 0.18716851416220345, + "grad_norm": 0.228515625, + "learning_rate": 0.0017998494805522224, + "loss": 0.0991, + "step": 21562 + }, + { + "epoch": 0.18717719464240762, + "grad_norm": 0.224609375, + "learning_rate": 0.0017998307671362502, + "loss": 0.1196, + "step": 21563 + }, + { + "epoch": 0.18718587512261178, + "grad_norm": 0.158203125, + "learning_rate": 0.0017998120529549536, + "loss": 0.1016, + "step": 21564 + }, + { + "epoch": 0.18719455560281595, + "grad_norm": 0.373046875, + "learning_rate": 0.0017997933380083527, + "loss": 0.1074, + "step": 21565 + }, + { + "epoch": 0.1872032360830201, + "grad_norm": 1.0, + "learning_rate": 0.0017997746222964683, + "loss": 0.1309, + "step": 21566 + }, + { + "epoch": 0.18721191656322428, + "grad_norm": 0.10205078125, + "learning_rate": 0.0017997559058193211, + "loss": 0.085, + "step": 21567 + }, + { + "epoch": 0.18722059704342844, + "grad_norm": 0.546875, + "learning_rate": 0.0017997371885769315, + "loss": 0.1143, + "step": 21568 + }, + { + "epoch": 0.1872292775236326, + "grad_norm": 0.90625, + "learning_rate": 0.0017997184705693197, + "loss": 0.127, + "step": 21569 + }, + { + "epoch": 0.18723795800383677, + "grad_norm": 0.2255859375, + "learning_rate": 0.001799699751796506, + "loss": 0.1152, + "step": 21570 + }, + { + "epoch": 0.18724663848404094, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017996810322585114, + "loss": 0.127, + "step": 21571 + }, + { + "epoch": 0.1872553189642451, + "grad_norm": 0.359375, + "learning_rate": 0.0017996623119553561, + "loss": 0.1143, + "step": 21572 + }, + { + "epoch": 0.18726399944444927, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017996435908870605, + "loss": 0.0864, + "step": 21573 + }, + { + "epoch": 0.18727267992465343, + "grad_norm": 0.44140625, + "learning_rate": 0.0017996248690536453, + "loss": 0.1104, + "step": 21574 + }, + { + "epoch": 0.1872813604048576, + "grad_norm": 0.1142578125, + "learning_rate": 0.001799606146455131, + "loss": 0.1152, + "step": 21575 + }, + { + "epoch": 0.18729004088506176, + "grad_norm": 0.130859375, + "learning_rate": 0.0017995874230915376, + "loss": 0.1582, + "step": 21576 + }, + { + "epoch": 0.18729872136526593, + "grad_norm": 0.333984375, + "learning_rate": 0.0017995686989628864, + "loss": 0.082, + "step": 21577 + }, + { + "epoch": 0.1873074018454701, + "grad_norm": 0.703125, + "learning_rate": 0.001799549974069197, + "loss": 0.0801, + "step": 21578 + }, + { + "epoch": 0.18731608232567426, + "grad_norm": 0.84375, + "learning_rate": 0.0017995312484104901, + "loss": 0.1108, + "step": 21579 + }, + { + "epoch": 0.18732476280587843, + "grad_norm": 0.130859375, + "learning_rate": 0.0017995125219867866, + "loss": 0.0986, + "step": 21580 + }, + { + "epoch": 0.1873334432860826, + "grad_norm": 0.578125, + "learning_rate": 0.0017994937947981068, + "loss": 0.1226, + "step": 21581 + }, + { + "epoch": 0.18734212376628676, + "grad_norm": 0.271484375, + "learning_rate": 0.0017994750668444707, + "loss": 0.0991, + "step": 21582 + }, + { + "epoch": 0.18735080424649092, + "grad_norm": 0.4609375, + "learning_rate": 0.0017994563381258997, + "loss": 0.1377, + "step": 21583 + }, + { + "epoch": 0.18735948472669509, + "grad_norm": 0.2421875, + "learning_rate": 0.0017994376086424135, + "loss": 0.1445, + "step": 21584 + }, + { + "epoch": 0.18736816520689925, + "grad_norm": 0.44921875, + "learning_rate": 0.0017994188783940328, + "loss": 0.1138, + "step": 21585 + }, + { + "epoch": 0.18737684568710342, + "grad_norm": 0.091796875, + "learning_rate": 0.0017994001473807782, + "loss": 0.0981, + "step": 21586 + }, + { + "epoch": 0.18738552616730758, + "grad_norm": 0.43359375, + "learning_rate": 0.0017993814156026702, + "loss": 0.1367, + "step": 21587 + }, + { + "epoch": 0.18739420664751175, + "grad_norm": 0.2265625, + "learning_rate": 0.0017993626830597289, + "loss": 0.1221, + "step": 21588 + }, + { + "epoch": 0.1874028871277159, + "grad_norm": 0.189453125, + "learning_rate": 0.0017993439497519753, + "loss": 0.1816, + "step": 21589 + }, + { + "epoch": 0.18741156760792008, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017993252156794295, + "loss": 0.0913, + "step": 21590 + }, + { + "epoch": 0.18742024808812424, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017993064808421123, + "loss": 0.1191, + "step": 21591 + }, + { + "epoch": 0.1874289285683284, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017992877452400443, + "loss": 0.0889, + "step": 21592 + }, + { + "epoch": 0.18743760904853257, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017992690088732453, + "loss": 0.1152, + "step": 21593 + }, + { + "epoch": 0.18744628952873674, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017992502717417364, + "loss": 0.0928, + "step": 21594 + }, + { + "epoch": 0.1874549700089409, + "grad_norm": 0.72265625, + "learning_rate": 0.0017992315338455377, + "loss": 0.105, + "step": 21595 + }, + { + "epoch": 0.18746365048914507, + "grad_norm": 0.12255859375, + "learning_rate": 0.0017992127951846702, + "loss": 0.1055, + "step": 21596 + }, + { + "epoch": 0.18747233096934923, + "grad_norm": 0.4765625, + "learning_rate": 0.0017991940557591539, + "loss": 0.0938, + "step": 21597 + }, + { + "epoch": 0.1874810114495534, + "grad_norm": 0.333984375, + "learning_rate": 0.0017991753155690094, + "loss": 0.1245, + "step": 21598 + }, + { + "epoch": 0.18748969192975756, + "grad_norm": 0.169921875, + "learning_rate": 0.0017991565746142575, + "loss": 0.0762, + "step": 21599 + }, + { + "epoch": 0.18749837240996173, + "grad_norm": 0.18359375, + "learning_rate": 0.0017991378328949185, + "loss": 0.124, + "step": 21600 + }, + { + "epoch": 0.1875070528901659, + "grad_norm": 0.18359375, + "learning_rate": 0.0017991190904110127, + "loss": 0.1167, + "step": 21601 + }, + { + "epoch": 0.18751573337037006, + "grad_norm": 0.17578125, + "learning_rate": 0.0017991003471625607, + "loss": 0.1025, + "step": 21602 + }, + { + "epoch": 0.18752441385057422, + "grad_norm": 0.474609375, + "learning_rate": 0.0017990816031495833, + "loss": 0.127, + "step": 21603 + }, + { + "epoch": 0.1875330943307784, + "grad_norm": 0.369140625, + "learning_rate": 0.0017990628583721008, + "loss": 0.1143, + "step": 21604 + }, + { + "epoch": 0.18754177481098255, + "grad_norm": 0.4609375, + "learning_rate": 0.0017990441128301333, + "loss": 0.1885, + "step": 21605 + }, + { + "epoch": 0.18755045529118672, + "grad_norm": 0.390625, + "learning_rate": 0.0017990253665237021, + "loss": 0.1196, + "step": 21606 + }, + { + "epoch": 0.18755913577139088, + "grad_norm": 0.2451171875, + "learning_rate": 0.001799006619452827, + "loss": 0.127, + "step": 21607 + }, + { + "epoch": 0.18756781625159505, + "grad_norm": 0.86328125, + "learning_rate": 0.0017989878716175287, + "loss": 0.1152, + "step": 21608 + }, + { + "epoch": 0.1875764967317992, + "grad_norm": 0.123046875, + "learning_rate": 0.001798969123017828, + "loss": 0.1064, + "step": 21609 + }, + { + "epoch": 0.18758517721200338, + "grad_norm": 0.51171875, + "learning_rate": 0.001798950373653745, + "loss": 0.1748, + "step": 21610 + }, + { + "epoch": 0.18759385769220754, + "grad_norm": 0.498046875, + "learning_rate": 0.0017989316235253003, + "loss": 0.0986, + "step": 21611 + }, + { + "epoch": 0.1876025381724117, + "grad_norm": 0.13671875, + "learning_rate": 0.0017989128726325146, + "loss": 0.1084, + "step": 21612 + }, + { + "epoch": 0.18761121865261587, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017988941209754084, + "loss": 0.1182, + "step": 21613 + }, + { + "epoch": 0.18761989913282004, + "grad_norm": 0.12109375, + "learning_rate": 0.001798875368554002, + "loss": 0.1348, + "step": 21614 + }, + { + "epoch": 0.1876285796130242, + "grad_norm": 0.400390625, + "learning_rate": 0.001798856615368316, + "loss": 0.0986, + "step": 21615 + }, + { + "epoch": 0.18763726009322837, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017988378614183706, + "loss": 0.1064, + "step": 21616 + }, + { + "epoch": 0.18764594057343253, + "grad_norm": 0.48828125, + "learning_rate": 0.001798819106704187, + "loss": 0.1621, + "step": 21617 + }, + { + "epoch": 0.1876546210536367, + "grad_norm": 0.263671875, + "learning_rate": 0.0017988003512257854, + "loss": 0.0679, + "step": 21618 + }, + { + "epoch": 0.18766330153384086, + "grad_norm": 0.232421875, + "learning_rate": 0.001798781594983186, + "loss": 0.0972, + "step": 21619 + }, + { + "epoch": 0.18767198201404503, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017987628379764094, + "loss": 0.1338, + "step": 21620 + }, + { + "epoch": 0.1876806624942492, + "grad_norm": 0.51953125, + "learning_rate": 0.0017987440802054767, + "loss": 0.0854, + "step": 21621 + }, + { + "epoch": 0.18768934297445336, + "grad_norm": 0.60546875, + "learning_rate": 0.0017987253216704077, + "loss": 0.1016, + "step": 21622 + }, + { + "epoch": 0.18769802345465753, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017987065623712231, + "loss": 0.0957, + "step": 21623 + }, + { + "epoch": 0.18770670393486166, + "grad_norm": 0.4296875, + "learning_rate": 0.0017986878023079436, + "loss": 0.1367, + "step": 21624 + }, + { + "epoch": 0.18771538441506583, + "grad_norm": 0.3203125, + "learning_rate": 0.0017986690414805897, + "loss": 0.082, + "step": 21625 + }, + { + "epoch": 0.18772406489527, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017986502798891817, + "loss": 0.1387, + "step": 21626 + }, + { + "epoch": 0.18773274537547416, + "grad_norm": 0.46875, + "learning_rate": 0.0017986315175337404, + "loss": 0.0942, + "step": 21627 + }, + { + "epoch": 0.18774142585567832, + "grad_norm": 0.294921875, + "learning_rate": 0.0017986127544142858, + "loss": 0.1064, + "step": 21628 + }, + { + "epoch": 0.1877501063358825, + "grad_norm": 0.32421875, + "learning_rate": 0.0017985939905308392, + "loss": 0.1055, + "step": 21629 + }, + { + "epoch": 0.18775878681608665, + "grad_norm": 0.443359375, + "learning_rate": 0.0017985752258834204, + "loss": 0.0977, + "step": 21630 + }, + { + "epoch": 0.18776746729629082, + "grad_norm": 1.2421875, + "learning_rate": 0.0017985564604720507, + "loss": 0.1738, + "step": 21631 + }, + { + "epoch": 0.18777614777649498, + "grad_norm": 0.24609375, + "learning_rate": 0.0017985376942967497, + "loss": 0.1055, + "step": 21632 + }, + { + "epoch": 0.18778482825669915, + "grad_norm": 0.546875, + "learning_rate": 0.0017985189273575387, + "loss": 0.0742, + "step": 21633 + }, + { + "epoch": 0.18779350873690331, + "grad_norm": 0.318359375, + "learning_rate": 0.0017985001596544376, + "loss": 0.1348, + "step": 21634 + }, + { + "epoch": 0.18780218921710748, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017984813911874674, + "loss": 0.1416, + "step": 21635 + }, + { + "epoch": 0.18781086969731164, + "grad_norm": 0.359375, + "learning_rate": 0.0017984626219566485, + "loss": 0.124, + "step": 21636 + }, + { + "epoch": 0.1878195501775158, + "grad_norm": 0.1708984375, + "learning_rate": 0.001798443851962001, + "loss": 0.0869, + "step": 21637 + }, + { + "epoch": 0.18782823065771997, + "grad_norm": 0.2353515625, + "learning_rate": 0.0017984250812035463, + "loss": 0.082, + "step": 21638 + }, + { + "epoch": 0.18783691113792414, + "grad_norm": 0.50390625, + "learning_rate": 0.001798406309681304, + "loss": 0.104, + "step": 21639 + }, + { + "epoch": 0.1878455916181283, + "grad_norm": 0.2470703125, + "learning_rate": 0.0017983875373952954, + "loss": 0.1094, + "step": 21640 + }, + { + "epoch": 0.18785427209833247, + "grad_norm": 0.375, + "learning_rate": 0.0017983687643455405, + "loss": 0.1084, + "step": 21641 + }, + { + "epoch": 0.18786295257853663, + "grad_norm": 0.25, + "learning_rate": 0.00179834999053206, + "loss": 0.1465, + "step": 21642 + }, + { + "epoch": 0.1878716330587408, + "grad_norm": 0.435546875, + "learning_rate": 0.0017983312159548747, + "loss": 0.127, + "step": 21643 + }, + { + "epoch": 0.18788031353894497, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017983124406140045, + "loss": 0.1035, + "step": 21644 + }, + { + "epoch": 0.18788899401914913, + "grad_norm": 0.38671875, + "learning_rate": 0.0017982936645094705, + "loss": 0.1387, + "step": 21645 + }, + { + "epoch": 0.1878976744993533, + "grad_norm": 0.76953125, + "learning_rate": 0.001798274887641293, + "loss": 0.0903, + "step": 21646 + }, + { + "epoch": 0.18790635497955746, + "grad_norm": 0.271484375, + "learning_rate": 0.0017982561100094924, + "loss": 0.1143, + "step": 21647 + }, + { + "epoch": 0.18791503545976163, + "grad_norm": 0.66015625, + "learning_rate": 0.0017982373316140898, + "loss": 0.1445, + "step": 21648 + }, + { + "epoch": 0.1879237159399658, + "grad_norm": 0.765625, + "learning_rate": 0.0017982185524551052, + "loss": 0.082, + "step": 21649 + }, + { + "epoch": 0.18793239642016996, + "grad_norm": 0.3671875, + "learning_rate": 0.0017981997725325589, + "loss": 0.1289, + "step": 21650 + }, + { + "epoch": 0.18794107690037412, + "grad_norm": 0.80078125, + "learning_rate": 0.0017981809918464725, + "loss": 0.1104, + "step": 21651 + }, + { + "epoch": 0.1879497573805783, + "grad_norm": 0.466796875, + "learning_rate": 0.0017981622103968654, + "loss": 0.1118, + "step": 21652 + }, + { + "epoch": 0.18795843786078245, + "grad_norm": 0.56640625, + "learning_rate": 0.0017981434281837585, + "loss": 0.105, + "step": 21653 + }, + { + "epoch": 0.18796711834098662, + "grad_norm": 0.80859375, + "learning_rate": 0.0017981246452071725, + "loss": 0.1094, + "step": 21654 + }, + { + "epoch": 0.18797579882119078, + "grad_norm": 0.0859375, + "learning_rate": 0.0017981058614671285, + "loss": 0.0835, + "step": 21655 + }, + { + "epoch": 0.18798447930139495, + "grad_norm": 0.33984375, + "learning_rate": 0.0017980870769636456, + "loss": 0.1104, + "step": 21656 + }, + { + "epoch": 0.1879931597815991, + "grad_norm": 0.26953125, + "learning_rate": 0.0017980682916967456, + "loss": 0.0986, + "step": 21657 + }, + { + "epoch": 0.18800184026180328, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017980495056664483, + "loss": 0.1191, + "step": 21658 + }, + { + "epoch": 0.18801052074200744, + "grad_norm": 0.20703125, + "learning_rate": 0.0017980307188727746, + "loss": 0.1309, + "step": 21659 + }, + { + "epoch": 0.1880192012222116, + "grad_norm": 0.6640625, + "learning_rate": 0.0017980119313157454, + "loss": 0.1074, + "step": 21660 + }, + { + "epoch": 0.18802788170241577, + "grad_norm": 0.58984375, + "learning_rate": 0.0017979931429953804, + "loss": 0.1133, + "step": 21661 + }, + { + "epoch": 0.18803656218261994, + "grad_norm": 0.33203125, + "learning_rate": 0.0017979743539117009, + "loss": 0.1187, + "step": 21662 + }, + { + "epoch": 0.1880452426628241, + "grad_norm": 0.404296875, + "learning_rate": 0.0017979555640647272, + "loss": 0.1074, + "step": 21663 + }, + { + "epoch": 0.18805392314302827, + "grad_norm": 0.369140625, + "learning_rate": 0.0017979367734544793, + "loss": 0.0889, + "step": 21664 + }, + { + "epoch": 0.18806260362323243, + "grad_norm": 0.25390625, + "learning_rate": 0.0017979179820809784, + "loss": 0.1016, + "step": 21665 + }, + { + "epoch": 0.1880712841034366, + "grad_norm": 1.2734375, + "learning_rate": 0.0017978991899442452, + "loss": 0.1162, + "step": 21666 + }, + { + "epoch": 0.18807996458364076, + "grad_norm": 0.181640625, + "learning_rate": 0.0017978803970442993, + "loss": 0.0928, + "step": 21667 + }, + { + "epoch": 0.18808864506384493, + "grad_norm": 0.71875, + "learning_rate": 0.0017978616033811625, + "loss": 0.0938, + "step": 21668 + }, + { + "epoch": 0.1880973255440491, + "grad_norm": 0.412109375, + "learning_rate": 0.0017978428089548545, + "loss": 0.0898, + "step": 21669 + }, + { + "epoch": 0.18810600602425326, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017978240137653963, + "loss": 0.1201, + "step": 21670 + }, + { + "epoch": 0.18811468650445742, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017978052178128081, + "loss": 0.1465, + "step": 21671 + }, + { + "epoch": 0.1881233669846616, + "grad_norm": 0.07470703125, + "learning_rate": 0.0017977864210971105, + "loss": 0.0986, + "step": 21672 + }, + { + "epoch": 0.18813204746486575, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017977676236183243, + "loss": 0.1055, + "step": 21673 + }, + { + "epoch": 0.18814072794506992, + "grad_norm": 0.09033203125, + "learning_rate": 0.0017977488253764697, + "loss": 0.1079, + "step": 21674 + }, + { + "epoch": 0.18814940842527408, + "grad_norm": 0.185546875, + "learning_rate": 0.0017977300263715677, + "loss": 0.103, + "step": 21675 + }, + { + "epoch": 0.18815808890547825, + "grad_norm": 0.77734375, + "learning_rate": 0.0017977112266036384, + "loss": 0.0996, + "step": 21676 + }, + { + "epoch": 0.18816676938568241, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017976924260727028, + "loss": 0.1118, + "step": 21677 + }, + { + "epoch": 0.18817544986588658, + "grad_norm": 0.462890625, + "learning_rate": 0.001797673624778781, + "loss": 0.1162, + "step": 21678 + }, + { + "epoch": 0.18818413034609074, + "grad_norm": 0.408203125, + "learning_rate": 0.0017976548227218942, + "loss": 0.1118, + "step": 21679 + }, + { + "epoch": 0.1881928108262949, + "grad_norm": 0.212890625, + "learning_rate": 0.001797636019902062, + "loss": 0.105, + "step": 21680 + }, + { + "epoch": 0.18820149130649907, + "grad_norm": 0.228515625, + "learning_rate": 0.001797617216319306, + "loss": 0.0913, + "step": 21681 + }, + { + "epoch": 0.18821017178670324, + "grad_norm": 0.185546875, + "learning_rate": 0.0017975984119736462, + "loss": 0.127, + "step": 21682 + }, + { + "epoch": 0.1882188522669074, + "grad_norm": 0.3359375, + "learning_rate": 0.0017975796068651032, + "loss": 0.1221, + "step": 21683 + }, + { + "epoch": 0.18822753274711157, + "grad_norm": 0.478515625, + "learning_rate": 0.0017975608009936977, + "loss": 0.0996, + "step": 21684 + }, + { + "epoch": 0.18823621322731574, + "grad_norm": 1.734375, + "learning_rate": 0.0017975419943594502, + "loss": 0.1562, + "step": 21685 + }, + { + "epoch": 0.1882448937075199, + "grad_norm": 0.203125, + "learning_rate": 0.001797523186962381, + "loss": 0.125, + "step": 21686 + }, + { + "epoch": 0.18825357418772407, + "grad_norm": 0.142578125, + "learning_rate": 0.0017975043788025112, + "loss": 0.0869, + "step": 21687 + }, + { + "epoch": 0.18826225466792823, + "grad_norm": 0.423828125, + "learning_rate": 0.0017974855698798609, + "loss": 0.1387, + "step": 21688 + }, + { + "epoch": 0.1882709351481324, + "grad_norm": 0.2255859375, + "learning_rate": 0.001797466760194451, + "loss": 0.1182, + "step": 21689 + }, + { + "epoch": 0.18827961562833656, + "grad_norm": 0.79296875, + "learning_rate": 0.0017974479497463022, + "loss": 0.0884, + "step": 21690 + }, + { + "epoch": 0.18828829610854073, + "grad_norm": 0.34375, + "learning_rate": 0.0017974291385354341, + "loss": 0.1123, + "step": 21691 + }, + { + "epoch": 0.1882969765887449, + "grad_norm": 0.2138671875, + "learning_rate": 0.0017974103265618684, + "loss": 0.0889, + "step": 21692 + }, + { + "epoch": 0.18830565706894906, + "grad_norm": 0.51171875, + "learning_rate": 0.0017973915138256255, + "loss": 0.1201, + "step": 21693 + }, + { + "epoch": 0.18831433754915322, + "grad_norm": 0.10693359375, + "learning_rate": 0.001797372700326725, + "loss": 0.0889, + "step": 21694 + }, + { + "epoch": 0.1883230180293574, + "grad_norm": 0.21484375, + "learning_rate": 0.0017973538860651885, + "loss": 0.0845, + "step": 21695 + }, + { + "epoch": 0.18833169850956155, + "grad_norm": 0.10205078125, + "learning_rate": 0.0017973350710410365, + "loss": 0.1357, + "step": 21696 + }, + { + "epoch": 0.18834037898976572, + "grad_norm": 0.33203125, + "learning_rate": 0.0017973162552542893, + "loss": 0.1235, + "step": 21697 + }, + { + "epoch": 0.18834905946996988, + "grad_norm": 0.1044921875, + "learning_rate": 0.0017972974387049673, + "loss": 0.1074, + "step": 21698 + }, + { + "epoch": 0.18835773995017405, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017972786213930915, + "loss": 0.1416, + "step": 21699 + }, + { + "epoch": 0.1883664204303782, + "grad_norm": 0.33203125, + "learning_rate": 0.0017972598033186825, + "loss": 0.1289, + "step": 21700 + }, + { + "epoch": 0.18837510091058238, + "grad_norm": 0.435546875, + "learning_rate": 0.00179724098448176, + "loss": 0.1123, + "step": 21701 + }, + { + "epoch": 0.18838378139078654, + "grad_norm": 0.322265625, + "learning_rate": 0.0017972221648823459, + "loss": 0.1426, + "step": 21702 + }, + { + "epoch": 0.1883924618709907, + "grad_norm": 0.09716796875, + "learning_rate": 0.00179720334452046, + "loss": 0.125, + "step": 21703 + }, + { + "epoch": 0.18840114235119487, + "grad_norm": 1.3359375, + "learning_rate": 0.0017971845233961225, + "loss": 0.1426, + "step": 21704 + }, + { + "epoch": 0.18840982283139904, + "grad_norm": 0.296875, + "learning_rate": 0.0017971657015093548, + "loss": 0.106, + "step": 21705 + }, + { + "epoch": 0.1884185033116032, + "grad_norm": 1.8828125, + "learning_rate": 0.001797146878860177, + "loss": 0.124, + "step": 21706 + }, + { + "epoch": 0.18842718379180737, + "grad_norm": 0.21875, + "learning_rate": 0.0017971280554486102, + "loss": 0.0962, + "step": 21707 + }, + { + "epoch": 0.18843586427201153, + "grad_norm": 0.59375, + "learning_rate": 0.0017971092312746744, + "loss": 0.127, + "step": 21708 + }, + { + "epoch": 0.1884445447522157, + "grad_norm": 0.4609375, + "learning_rate": 0.0017970904063383906, + "loss": 0.207, + "step": 21709 + }, + { + "epoch": 0.18845322523241986, + "grad_norm": 0.216796875, + "learning_rate": 0.001797071580639779, + "loss": 0.0894, + "step": 21710 + }, + { + "epoch": 0.18846190571262403, + "grad_norm": 0.140625, + "learning_rate": 0.0017970527541788604, + "loss": 0.082, + "step": 21711 + }, + { + "epoch": 0.1884705861928282, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017970339269556554, + "loss": 0.123, + "step": 21712 + }, + { + "epoch": 0.18847926667303236, + "grad_norm": 1.5, + "learning_rate": 0.0017970150989701848, + "loss": 0.207, + "step": 21713 + }, + { + "epoch": 0.18848794715323652, + "grad_norm": 0.3046875, + "learning_rate": 0.0017969962702224687, + "loss": 0.0835, + "step": 21714 + }, + { + "epoch": 0.1884966276334407, + "grad_norm": 0.0693359375, + "learning_rate": 0.001796977440712528, + "loss": 0.0874, + "step": 21715 + }, + { + "epoch": 0.18850530811364485, + "grad_norm": 0.08935546875, + "learning_rate": 0.0017969586104403835, + "loss": 0.1074, + "step": 21716 + }, + { + "epoch": 0.18851398859384902, + "grad_norm": 0.310546875, + "learning_rate": 0.001796939779406055, + "loss": 0.0938, + "step": 21717 + }, + { + "epoch": 0.18852266907405318, + "grad_norm": 0.62109375, + "learning_rate": 0.0017969209476095642, + "loss": 0.1221, + "step": 21718 + }, + { + "epoch": 0.18853134955425735, + "grad_norm": 0.52734375, + "learning_rate": 0.0017969021150509308, + "loss": 0.0938, + "step": 21719 + }, + { + "epoch": 0.18854003003446151, + "grad_norm": 0.12890625, + "learning_rate": 0.0017968832817301756, + "loss": 0.1465, + "step": 21720 + }, + { + "epoch": 0.18854871051466568, + "grad_norm": 0.921875, + "learning_rate": 0.0017968644476473194, + "loss": 0.0991, + "step": 21721 + }, + { + "epoch": 0.18855739099486984, + "grad_norm": 0.08544921875, + "learning_rate": 0.0017968456128023829, + "loss": 0.1143, + "step": 21722 + }, + { + "epoch": 0.188566071475074, + "grad_norm": 0.23828125, + "learning_rate": 0.0017968267771953862, + "loss": 0.0986, + "step": 21723 + }, + { + "epoch": 0.18857475195527817, + "grad_norm": 0.1611328125, + "learning_rate": 0.0017968079408263507, + "loss": 0.1328, + "step": 21724 + }, + { + "epoch": 0.18858343243548234, + "grad_norm": 0.47265625, + "learning_rate": 0.001796789103695296, + "loss": 0.1172, + "step": 21725 + }, + { + "epoch": 0.1885921129156865, + "grad_norm": 0.396484375, + "learning_rate": 0.0017967702658022438, + "loss": 0.1602, + "step": 21726 + }, + { + "epoch": 0.18860079339589067, + "grad_norm": 0.306640625, + "learning_rate": 0.0017967514271472137, + "loss": 0.1084, + "step": 21727 + }, + { + "epoch": 0.18860947387609484, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017967325877302266, + "loss": 0.1084, + "step": 21728 + }, + { + "epoch": 0.188618154356299, + "grad_norm": 0.20703125, + "learning_rate": 0.0017967137475513033, + "loss": 0.1064, + "step": 21729 + }, + { + "epoch": 0.18862683483650317, + "grad_norm": 0.455078125, + "learning_rate": 0.0017966949066104644, + "loss": 0.0967, + "step": 21730 + }, + { + "epoch": 0.18863551531670733, + "grad_norm": 0.08935546875, + "learning_rate": 0.0017966760649077305, + "loss": 0.0996, + "step": 21731 + }, + { + "epoch": 0.1886441957969115, + "grad_norm": 0.431640625, + "learning_rate": 0.001796657222443122, + "loss": 0.1494, + "step": 21732 + }, + { + "epoch": 0.18865287627711566, + "grad_norm": 0.126953125, + "learning_rate": 0.0017966383792166595, + "loss": 0.1187, + "step": 21733 + }, + { + "epoch": 0.18866155675731983, + "grad_norm": 0.41796875, + "learning_rate": 0.0017966195352283637, + "loss": 0.0889, + "step": 21734 + }, + { + "epoch": 0.188670237237524, + "grad_norm": 0.57421875, + "learning_rate": 0.0017966006904782557, + "loss": 0.1221, + "step": 21735 + }, + { + "epoch": 0.18867891771772816, + "grad_norm": 0.81640625, + "learning_rate": 0.0017965818449663551, + "loss": 0.124, + "step": 21736 + }, + { + "epoch": 0.18868759819793232, + "grad_norm": 0.11767578125, + "learning_rate": 0.0017965629986926835, + "loss": 0.1357, + "step": 21737 + }, + { + "epoch": 0.1886962786781365, + "grad_norm": 0.12451171875, + "learning_rate": 0.001796544151657261, + "loss": 0.0938, + "step": 21738 + }, + { + "epoch": 0.18870495915834065, + "grad_norm": 0.3125, + "learning_rate": 0.001796525303860108, + "loss": 0.085, + "step": 21739 + }, + { + "epoch": 0.18871363963854482, + "grad_norm": 0.55859375, + "learning_rate": 0.0017965064553012457, + "loss": 0.0972, + "step": 21740 + }, + { + "epoch": 0.18872232011874898, + "grad_norm": 0.357421875, + "learning_rate": 0.001796487605980694, + "loss": 0.1196, + "step": 21741 + }, + { + "epoch": 0.18873100059895315, + "grad_norm": 1.8671875, + "learning_rate": 0.0017964687558984745, + "loss": 0.1699, + "step": 21742 + }, + { + "epoch": 0.1887396810791573, + "grad_norm": 0.3984375, + "learning_rate": 0.0017964499050546069, + "loss": 0.1514, + "step": 21743 + }, + { + "epoch": 0.18874836155936148, + "grad_norm": 0.1611328125, + "learning_rate": 0.0017964310534491121, + "loss": 0.1104, + "step": 21744 + }, + { + "epoch": 0.18875704203956564, + "grad_norm": 0.09423828125, + "learning_rate": 0.0017964122010820107, + "loss": 0.0703, + "step": 21745 + }, + { + "epoch": 0.1887657225197698, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017963933479533236, + "loss": 0.1123, + "step": 21746 + }, + { + "epoch": 0.18877440299997394, + "grad_norm": 0.24609375, + "learning_rate": 0.0017963744940630712, + "loss": 0.1865, + "step": 21747 + }, + { + "epoch": 0.1887830834801781, + "grad_norm": 0.099609375, + "learning_rate": 0.0017963556394112738, + "loss": 0.0693, + "step": 21748 + }, + { + "epoch": 0.18879176396038228, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017963367839979529, + "loss": 0.0957, + "step": 21749 + }, + { + "epoch": 0.18880044444058644, + "grad_norm": 0.66796875, + "learning_rate": 0.0017963179278231282, + "loss": 0.1245, + "step": 21750 + }, + { + "epoch": 0.1888091249207906, + "grad_norm": 0.2578125, + "learning_rate": 0.0017962990708868206, + "loss": 0.0977, + "step": 21751 + }, + { + "epoch": 0.18881780540099477, + "grad_norm": 0.8046875, + "learning_rate": 0.0017962802131890508, + "loss": 0.1035, + "step": 21752 + }, + { + "epoch": 0.18882648588119894, + "grad_norm": 1.3359375, + "learning_rate": 0.0017962613547298396, + "loss": 0.1357, + "step": 21753 + }, + { + "epoch": 0.1888351663614031, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017962424955092074, + "loss": 0.0962, + "step": 21754 + }, + { + "epoch": 0.18884384684160727, + "grad_norm": 0.0673828125, + "learning_rate": 0.0017962236355271747, + "loss": 0.0796, + "step": 21755 + }, + { + "epoch": 0.18885252732181143, + "grad_norm": 0.291015625, + "learning_rate": 0.0017962047747837626, + "loss": 0.1123, + "step": 21756 + }, + { + "epoch": 0.1888612078020156, + "grad_norm": 0.19140625, + "learning_rate": 0.0017961859132789911, + "loss": 0.1001, + "step": 21757 + }, + { + "epoch": 0.18886988828221976, + "grad_norm": 0.072265625, + "learning_rate": 0.0017961670510128814, + "loss": 0.1006, + "step": 21758 + }, + { + "epoch": 0.18887856876242393, + "grad_norm": 0.333984375, + "learning_rate": 0.0017961481879854538, + "loss": 0.127, + "step": 21759 + }, + { + "epoch": 0.1888872492426281, + "grad_norm": 0.470703125, + "learning_rate": 0.001796129324196729, + "loss": 0.0918, + "step": 21760 + }, + { + "epoch": 0.18889592972283226, + "grad_norm": 0.435546875, + "learning_rate": 0.0017961104596467274, + "loss": 0.1631, + "step": 21761 + }, + { + "epoch": 0.18890461020303642, + "grad_norm": 0.5625, + "learning_rate": 0.0017960915943354698, + "loss": 0.1152, + "step": 21762 + }, + { + "epoch": 0.1889132906832406, + "grad_norm": 0.111328125, + "learning_rate": 0.0017960727282629774, + "loss": 0.0957, + "step": 21763 + }, + { + "epoch": 0.18892197116344475, + "grad_norm": 0.11669921875, + "learning_rate": 0.00179605386142927, + "loss": 0.0845, + "step": 21764 + }, + { + "epoch": 0.18893065164364892, + "grad_norm": 0.203125, + "learning_rate": 0.0017960349938343687, + "loss": 0.0786, + "step": 21765 + }, + { + "epoch": 0.18893933212385308, + "grad_norm": 0.1552734375, + "learning_rate": 0.001796016125478294, + "loss": 0.1235, + "step": 21766 + }, + { + "epoch": 0.18894801260405725, + "grad_norm": 0.142578125, + "learning_rate": 0.0017959972563610664, + "loss": 0.085, + "step": 21767 + }, + { + "epoch": 0.1889566930842614, + "grad_norm": 0.111328125, + "learning_rate": 0.0017959783864827067, + "loss": 0.0918, + "step": 21768 + }, + { + "epoch": 0.18896537356446558, + "grad_norm": 0.091796875, + "learning_rate": 0.0017959595158432353, + "loss": 0.0664, + "step": 21769 + }, + { + "epoch": 0.18897405404466974, + "grad_norm": 0.224609375, + "learning_rate": 0.0017959406444426733, + "loss": 0.0938, + "step": 21770 + }, + { + "epoch": 0.1889827345248739, + "grad_norm": 0.37890625, + "learning_rate": 0.001795921772281041, + "loss": 0.1904, + "step": 21771 + }, + { + "epoch": 0.18899141500507807, + "grad_norm": 0.50390625, + "learning_rate": 0.0017959028993583593, + "loss": 0.1289, + "step": 21772 + }, + { + "epoch": 0.18900009548528224, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017958840256746484, + "loss": 0.1016, + "step": 21773 + }, + { + "epoch": 0.1890087759654864, + "grad_norm": 0.50390625, + "learning_rate": 0.0017958651512299292, + "loss": 0.1216, + "step": 21774 + }, + { + "epoch": 0.18901745644569057, + "grad_norm": 0.2392578125, + "learning_rate": 0.0017958462760242222, + "loss": 0.1465, + "step": 21775 + }, + { + "epoch": 0.18902613692589473, + "grad_norm": 0.51953125, + "learning_rate": 0.0017958274000575483, + "loss": 0.0972, + "step": 21776 + }, + { + "epoch": 0.1890348174060989, + "grad_norm": 0.29296875, + "learning_rate": 0.0017958085233299279, + "loss": 0.1128, + "step": 21777 + }, + { + "epoch": 0.18904349788630306, + "grad_norm": 0.3828125, + "learning_rate": 0.0017957896458413818, + "loss": 0.1396, + "step": 21778 + }, + { + "epoch": 0.18905217836650723, + "grad_norm": 0.18359375, + "learning_rate": 0.001795770767591931, + "loss": 0.0908, + "step": 21779 + }, + { + "epoch": 0.1890608588467114, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017957518885815952, + "loss": 0.103, + "step": 21780 + }, + { + "epoch": 0.18906953932691556, + "grad_norm": 0.12255859375, + "learning_rate": 0.001795733008810396, + "loss": 0.0811, + "step": 21781 + }, + { + "epoch": 0.18907821980711972, + "grad_norm": 0.30859375, + "learning_rate": 0.0017957141282783533, + "loss": 0.1396, + "step": 21782 + }, + { + "epoch": 0.1890869002873239, + "grad_norm": 0.091796875, + "learning_rate": 0.0017956952469854882, + "loss": 0.1001, + "step": 21783 + }, + { + "epoch": 0.18909558076752805, + "grad_norm": 0.4140625, + "learning_rate": 0.001795676364931821, + "loss": 0.1069, + "step": 21784 + }, + { + "epoch": 0.18910426124773222, + "grad_norm": 0.55078125, + "learning_rate": 0.001795657482117373, + "loss": 0.1523, + "step": 21785 + }, + { + "epoch": 0.18911294172793638, + "grad_norm": 0.09619140625, + "learning_rate": 0.0017956385985421644, + "loss": 0.1045, + "step": 21786 + }, + { + "epoch": 0.18912162220814055, + "grad_norm": 0.267578125, + "learning_rate": 0.001795619714206216, + "loss": 0.1113, + "step": 21787 + }, + { + "epoch": 0.18913030268834471, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017956008291095478, + "loss": 0.1201, + "step": 21788 + }, + { + "epoch": 0.18913898316854888, + "grad_norm": 0.4609375, + "learning_rate": 0.0017955819432521813, + "loss": 0.0781, + "step": 21789 + }, + { + "epoch": 0.18914766364875304, + "grad_norm": 0.859375, + "learning_rate": 0.0017955630566341366, + "loss": 0.0933, + "step": 21790 + }, + { + "epoch": 0.1891563441289572, + "grad_norm": 0.419921875, + "learning_rate": 0.0017955441692554348, + "loss": 0.1465, + "step": 21791 + }, + { + "epoch": 0.18916502460916138, + "grad_norm": 0.52734375, + "learning_rate": 0.0017955252811160964, + "loss": 0.1211, + "step": 21792 + }, + { + "epoch": 0.18917370508936554, + "grad_norm": 0.9453125, + "learning_rate": 0.001795506392216142, + "loss": 0.1016, + "step": 21793 + }, + { + "epoch": 0.1891823855695697, + "grad_norm": 0.1796875, + "learning_rate": 0.0017954875025555923, + "loss": 0.1221, + "step": 21794 + }, + { + "epoch": 0.18919106604977387, + "grad_norm": 0.2109375, + "learning_rate": 0.0017954686121344676, + "loss": 0.1533, + "step": 21795 + }, + { + "epoch": 0.18919974652997804, + "grad_norm": 0.353515625, + "learning_rate": 0.0017954497209527892, + "loss": 0.0776, + "step": 21796 + }, + { + "epoch": 0.1892084270101822, + "grad_norm": 0.625, + "learning_rate": 0.0017954308290105772, + "loss": 0.1064, + "step": 21797 + }, + { + "epoch": 0.18921710749038637, + "grad_norm": 0.365234375, + "learning_rate": 0.0017954119363078527, + "loss": 0.1006, + "step": 21798 + }, + { + "epoch": 0.18922578797059053, + "grad_norm": 0.37109375, + "learning_rate": 0.001795393042844636, + "loss": 0.1045, + "step": 21799 + }, + { + "epoch": 0.1892344684507947, + "grad_norm": 0.451171875, + "learning_rate": 0.001795374148620948, + "loss": 0.0693, + "step": 21800 + }, + { + "epoch": 0.18924314893099886, + "grad_norm": 0.365234375, + "learning_rate": 0.0017953552536368093, + "loss": 0.1309, + "step": 21801 + }, + { + "epoch": 0.18925182941120303, + "grad_norm": 1.1328125, + "learning_rate": 0.0017953363578922404, + "loss": 0.1025, + "step": 21802 + }, + { + "epoch": 0.1892605098914072, + "grad_norm": 0.41796875, + "learning_rate": 0.0017953174613872622, + "loss": 0.126, + "step": 21803 + }, + { + "epoch": 0.18926919037161136, + "grad_norm": 0.2734375, + "learning_rate": 0.0017952985641218955, + "loss": 0.0947, + "step": 21804 + }, + { + "epoch": 0.18927787085181552, + "grad_norm": 0.451171875, + "learning_rate": 0.0017952796660961603, + "loss": 0.1357, + "step": 21805 + }, + { + "epoch": 0.1892865513320197, + "grad_norm": 0.6953125, + "learning_rate": 0.001795260767310078, + "loss": 0.1152, + "step": 21806 + }, + { + "epoch": 0.18929523181222385, + "grad_norm": 0.130859375, + "learning_rate": 0.0017952418677636688, + "loss": 0.0796, + "step": 21807 + }, + { + "epoch": 0.18930391229242802, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017952229674569537, + "loss": 0.084, + "step": 21808 + }, + { + "epoch": 0.18931259277263218, + "grad_norm": 0.45703125, + "learning_rate": 0.0017952040663899531, + "loss": 0.1187, + "step": 21809 + }, + { + "epoch": 0.18932127325283635, + "grad_norm": 0.1005859375, + "learning_rate": 0.0017951851645626878, + "loss": 0.0986, + "step": 21810 + }, + { + "epoch": 0.1893299537330405, + "grad_norm": 0.255859375, + "learning_rate": 0.0017951662619751785, + "loss": 0.1045, + "step": 21811 + }, + { + "epoch": 0.18933863421324468, + "grad_norm": 0.275390625, + "learning_rate": 0.0017951473586274459, + "loss": 0.0933, + "step": 21812 + }, + { + "epoch": 0.18934731469344884, + "grad_norm": 0.35546875, + "learning_rate": 0.0017951284545195104, + "loss": 0.1143, + "step": 21813 + }, + { + "epoch": 0.189355995173653, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017951095496513932, + "loss": 0.0952, + "step": 21814 + }, + { + "epoch": 0.18936467565385717, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017950906440231142, + "loss": 0.1025, + "step": 21815 + }, + { + "epoch": 0.18937335613406134, + "grad_norm": 1.03125, + "learning_rate": 0.0017950717376346947, + "loss": 0.1084, + "step": 21816 + }, + { + "epoch": 0.1893820366142655, + "grad_norm": 0.2109375, + "learning_rate": 0.0017950528304861552, + "loss": 0.0938, + "step": 21817 + }, + { + "epoch": 0.18939071709446967, + "grad_norm": 0.07080078125, + "learning_rate": 0.0017950339225775163, + "loss": 0.0938, + "step": 21818 + }, + { + "epoch": 0.18939939757467383, + "grad_norm": 0.314453125, + "learning_rate": 0.001795015013908799, + "loss": 0.1143, + "step": 21819 + }, + { + "epoch": 0.189408078054878, + "grad_norm": 0.095703125, + "learning_rate": 0.0017949961044800235, + "loss": 0.0986, + "step": 21820 + }, + { + "epoch": 0.18941675853508216, + "grad_norm": 0.97265625, + "learning_rate": 0.0017949771942912106, + "loss": 0.1729, + "step": 21821 + }, + { + "epoch": 0.18942543901528633, + "grad_norm": 0.4375, + "learning_rate": 0.0017949582833423812, + "loss": 0.1211, + "step": 21822 + }, + { + "epoch": 0.1894341194954905, + "grad_norm": 0.3046875, + "learning_rate": 0.001794939371633556, + "loss": 0.1206, + "step": 21823 + }, + { + "epoch": 0.18944279997569466, + "grad_norm": 0.09228515625, + "learning_rate": 0.0017949204591647555, + "loss": 0.1064, + "step": 21824 + }, + { + "epoch": 0.18945148045589882, + "grad_norm": 0.2138671875, + "learning_rate": 0.0017949015459360004, + "loss": 0.1001, + "step": 21825 + }, + { + "epoch": 0.189460160936103, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017948826319473114, + "loss": 0.1699, + "step": 21826 + }, + { + "epoch": 0.18946884141630715, + "grad_norm": 0.267578125, + "learning_rate": 0.001794863717198709, + "loss": 0.1338, + "step": 21827 + }, + { + "epoch": 0.18947752189651132, + "grad_norm": 0.38671875, + "learning_rate": 0.0017948448016902143, + "loss": 0.125, + "step": 21828 + }, + { + "epoch": 0.18948620237671548, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017948258854218475, + "loss": 0.0786, + "step": 21829 + }, + { + "epoch": 0.18949488285691965, + "grad_norm": 0.546875, + "learning_rate": 0.00179480696839363, + "loss": 0.1191, + "step": 21830 + }, + { + "epoch": 0.18950356333712381, + "grad_norm": 0.275390625, + "learning_rate": 0.0017947880506055816, + "loss": 0.1021, + "step": 21831 + }, + { + "epoch": 0.18951224381732798, + "grad_norm": 0.58984375, + "learning_rate": 0.0017947691320577236, + "loss": 0.1216, + "step": 21832 + }, + { + "epoch": 0.18952092429753215, + "grad_norm": 0.07568359375, + "learning_rate": 0.0017947502127500768, + "loss": 0.1279, + "step": 21833 + }, + { + "epoch": 0.1895296047777363, + "grad_norm": 0.1767578125, + "learning_rate": 0.001794731292682661, + "loss": 0.1406, + "step": 21834 + }, + { + "epoch": 0.18953828525794048, + "grad_norm": 0.470703125, + "learning_rate": 0.0017947123718554982, + "loss": 0.1279, + "step": 21835 + }, + { + "epoch": 0.18954696573814464, + "grad_norm": 0.263671875, + "learning_rate": 0.001794693450268608, + "loss": 0.0957, + "step": 21836 + }, + { + "epoch": 0.1895556462183488, + "grad_norm": 0.23046875, + "learning_rate": 0.0017946745279220115, + "loss": 0.1089, + "step": 21837 + }, + { + "epoch": 0.18956432669855297, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017946556048157292, + "loss": 0.1094, + "step": 21838 + }, + { + "epoch": 0.18957300717875714, + "grad_norm": 1.046875, + "learning_rate": 0.001794636680949782, + "loss": 0.105, + "step": 21839 + }, + { + "epoch": 0.1895816876589613, + "grad_norm": 0.21875, + "learning_rate": 0.001794617756324191, + "loss": 0.1279, + "step": 21840 + }, + { + "epoch": 0.18959036813916547, + "grad_norm": 0.26171875, + "learning_rate": 0.0017945988309389762, + "loss": 0.0996, + "step": 21841 + }, + { + "epoch": 0.18959904861936963, + "grad_norm": 0.1064453125, + "learning_rate": 0.0017945799047941584, + "loss": 0.0957, + "step": 21842 + }, + { + "epoch": 0.1896077290995738, + "grad_norm": 0.68359375, + "learning_rate": 0.0017945609778897587, + "loss": 0.0972, + "step": 21843 + }, + { + "epoch": 0.18961640957977796, + "grad_norm": 0.57421875, + "learning_rate": 0.0017945420502257971, + "loss": 0.1396, + "step": 21844 + }, + { + "epoch": 0.18962509005998213, + "grad_norm": 0.51953125, + "learning_rate": 0.0017945231218022954, + "loss": 0.1289, + "step": 21845 + }, + { + "epoch": 0.1896337705401863, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017945041926192733, + "loss": 0.0615, + "step": 21846 + }, + { + "epoch": 0.18964245102039046, + "grad_norm": 0.1953125, + "learning_rate": 0.0017944852626767518, + "loss": 0.103, + "step": 21847 + }, + { + "epoch": 0.18965113150059462, + "grad_norm": 0.1103515625, + "learning_rate": 0.001794466331974752, + "loss": 0.103, + "step": 21848 + }, + { + "epoch": 0.1896598119807988, + "grad_norm": 0.146484375, + "learning_rate": 0.0017944474005132935, + "loss": 0.0898, + "step": 21849 + }, + { + "epoch": 0.18966849246100295, + "grad_norm": 0.26953125, + "learning_rate": 0.0017944284682923984, + "loss": 0.1138, + "step": 21850 + }, + { + "epoch": 0.18967717294120712, + "grad_norm": 0.34375, + "learning_rate": 0.0017944095353120866, + "loss": 0.1172, + "step": 21851 + }, + { + "epoch": 0.18968585342141128, + "grad_norm": 0.34375, + "learning_rate": 0.001794390601572379, + "loss": 0.0859, + "step": 21852 + }, + { + "epoch": 0.18969453390161545, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017943716670732962, + "loss": 0.1104, + "step": 21853 + }, + { + "epoch": 0.1897032143818196, + "grad_norm": 0.3125, + "learning_rate": 0.0017943527318148589, + "loss": 0.0923, + "step": 21854 + }, + { + "epoch": 0.18971189486202378, + "grad_norm": 0.095703125, + "learning_rate": 0.0017943337957970881, + "loss": 0.1289, + "step": 21855 + }, + { + "epoch": 0.18972057534222794, + "grad_norm": 0.162109375, + "learning_rate": 0.0017943148590200042, + "loss": 0.0889, + "step": 21856 + }, + { + "epoch": 0.1897292558224321, + "grad_norm": 0.412109375, + "learning_rate": 0.001794295921483628, + "loss": 0.1562, + "step": 21857 + }, + { + "epoch": 0.18973793630263627, + "grad_norm": 0.640625, + "learning_rate": 0.00179427698318798, + "loss": 0.1064, + "step": 21858 + }, + { + "epoch": 0.18974661678284044, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017942580441330815, + "loss": 0.1055, + "step": 21859 + }, + { + "epoch": 0.1897552972630446, + "grad_norm": 0.48828125, + "learning_rate": 0.0017942391043189526, + "loss": 0.0957, + "step": 21860 + }, + { + "epoch": 0.18976397774324877, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017942201637456143, + "loss": 0.1357, + "step": 21861 + }, + { + "epoch": 0.18977265822345293, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017942012224130876, + "loss": 0.0996, + "step": 21862 + }, + { + "epoch": 0.1897813387036571, + "grad_norm": 0.0712890625, + "learning_rate": 0.0017941822803213923, + "loss": 0.0996, + "step": 21863 + }, + { + "epoch": 0.18979001918386126, + "grad_norm": 0.703125, + "learning_rate": 0.0017941633374705499, + "loss": 0.0889, + "step": 21864 + }, + { + "epoch": 0.18979869966406543, + "grad_norm": 0.2109375, + "learning_rate": 0.0017941443938605808, + "loss": 0.1367, + "step": 21865 + }, + { + "epoch": 0.1898073801442696, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017941254494915062, + "loss": 0.1006, + "step": 21866 + }, + { + "epoch": 0.18981606062447376, + "grad_norm": 0.7734375, + "learning_rate": 0.0017941065043633462, + "loss": 0.0859, + "step": 21867 + }, + { + "epoch": 0.18982474110467792, + "grad_norm": 0.130859375, + "learning_rate": 0.0017940875584761219, + "loss": 0.0928, + "step": 21868 + }, + { + "epoch": 0.1898334215848821, + "grad_norm": 0.474609375, + "learning_rate": 0.0017940686118298535, + "loss": 0.125, + "step": 21869 + }, + { + "epoch": 0.18984210206508623, + "grad_norm": 0.4765625, + "learning_rate": 0.0017940496644245623, + "loss": 0.1504, + "step": 21870 + }, + { + "epoch": 0.1898507825452904, + "grad_norm": 0.095703125, + "learning_rate": 0.0017940307162602686, + "loss": 0.0889, + "step": 21871 + }, + { + "epoch": 0.18985946302549456, + "grad_norm": 0.0869140625, + "learning_rate": 0.0017940117673369937, + "loss": 0.1104, + "step": 21872 + }, + { + "epoch": 0.18986814350569872, + "grad_norm": 0.3046875, + "learning_rate": 0.001793992817654758, + "loss": 0.1426, + "step": 21873 + }, + { + "epoch": 0.1898768239859029, + "grad_norm": 0.1279296875, + "learning_rate": 0.001793973867213582, + "loss": 0.1113, + "step": 21874 + }, + { + "epoch": 0.18988550446610705, + "grad_norm": 0.173828125, + "learning_rate": 0.0017939549160134866, + "loss": 0.0952, + "step": 21875 + }, + { + "epoch": 0.18989418494631122, + "grad_norm": 0.37109375, + "learning_rate": 0.0017939359640544922, + "loss": 0.1641, + "step": 21876 + }, + { + "epoch": 0.18990286542651538, + "grad_norm": 0.294921875, + "learning_rate": 0.0017939170113366206, + "loss": 0.123, + "step": 21877 + }, + { + "epoch": 0.18991154590671955, + "grad_norm": 0.390625, + "learning_rate": 0.001793898057859891, + "loss": 0.0957, + "step": 21878 + }, + { + "epoch": 0.1899202263869237, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017938791036243257, + "loss": 0.1221, + "step": 21879 + }, + { + "epoch": 0.18992890686712788, + "grad_norm": 0.349609375, + "learning_rate": 0.001793860148629944, + "loss": 0.1582, + "step": 21880 + }, + { + "epoch": 0.18993758734733204, + "grad_norm": 0.43359375, + "learning_rate": 0.0017938411928767677, + "loss": 0.1191, + "step": 21881 + }, + { + "epoch": 0.1899462678275362, + "grad_norm": 0.5, + "learning_rate": 0.0017938222363648166, + "loss": 0.126, + "step": 21882 + }, + { + "epoch": 0.18995494830774037, + "grad_norm": 0.07763671875, + "learning_rate": 0.0017938032790941123, + "loss": 0.0884, + "step": 21883 + }, + { + "epoch": 0.18996362878794454, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017937843210646752, + "loss": 0.0977, + "step": 21884 + }, + { + "epoch": 0.1899723092681487, + "grad_norm": 0.392578125, + "learning_rate": 0.0017937653622765257, + "loss": 0.0869, + "step": 21885 + }, + { + "epoch": 0.18998098974835287, + "grad_norm": 0.451171875, + "learning_rate": 0.001793746402729685, + "loss": 0.0977, + "step": 21886 + }, + { + "epoch": 0.18998967022855703, + "grad_norm": 0.25390625, + "learning_rate": 0.0017937274424241736, + "loss": 0.0977, + "step": 21887 + }, + { + "epoch": 0.1899983507087612, + "grad_norm": 0.416015625, + "learning_rate": 0.0017937084813600126, + "loss": 0.1211, + "step": 21888 + }, + { + "epoch": 0.19000703118896536, + "grad_norm": 0.474609375, + "learning_rate": 0.0017936895195372223, + "loss": 0.0825, + "step": 21889 + }, + { + "epoch": 0.19001571166916953, + "grad_norm": 0.255859375, + "learning_rate": 0.001793670556955823, + "loss": 0.1309, + "step": 21890 + }, + { + "epoch": 0.1900243921493737, + "grad_norm": 0.236328125, + "learning_rate": 0.0017936515936158369, + "loss": 0.1562, + "step": 21891 + }, + { + "epoch": 0.19003307262957786, + "grad_norm": 0.1875, + "learning_rate": 0.0017936326295172834, + "loss": 0.0859, + "step": 21892 + }, + { + "epoch": 0.19004175310978202, + "grad_norm": 0.255859375, + "learning_rate": 0.0017936136646601833, + "loss": 0.1523, + "step": 21893 + }, + { + "epoch": 0.1900504335899862, + "grad_norm": 0.19921875, + "learning_rate": 0.0017935946990445583, + "loss": 0.0928, + "step": 21894 + }, + { + "epoch": 0.19005911407019035, + "grad_norm": 0.59765625, + "learning_rate": 0.001793575732670428, + "loss": 0.0864, + "step": 21895 + }, + { + "epoch": 0.19006779455039452, + "grad_norm": 0.115234375, + "learning_rate": 0.0017935567655378143, + "loss": 0.0908, + "step": 21896 + }, + { + "epoch": 0.19007647503059868, + "grad_norm": 0.6328125, + "learning_rate": 0.001793537797646737, + "loss": 0.1235, + "step": 21897 + }, + { + "epoch": 0.19008515551080285, + "grad_norm": 0.6015625, + "learning_rate": 0.001793518828997217, + "loss": 0.1396, + "step": 21898 + }, + { + "epoch": 0.19009383599100702, + "grad_norm": 0.08349609375, + "learning_rate": 0.0017934998595892757, + "loss": 0.1055, + "step": 21899 + }, + { + "epoch": 0.19010251647121118, + "grad_norm": 0.427734375, + "learning_rate": 0.0017934808894229334, + "loss": 0.082, + "step": 21900 + }, + { + "epoch": 0.19011119695141535, + "grad_norm": 0.65625, + "learning_rate": 0.0017934619184982103, + "loss": 0.1113, + "step": 21901 + }, + { + "epoch": 0.1901198774316195, + "grad_norm": 0.421875, + "learning_rate": 0.0017934429468151282, + "loss": 0.1099, + "step": 21902 + }, + { + "epoch": 0.19012855791182368, + "grad_norm": 0.369140625, + "learning_rate": 0.0017934239743737069, + "loss": 0.0889, + "step": 21903 + }, + { + "epoch": 0.19013723839202784, + "grad_norm": 0.298828125, + "learning_rate": 0.0017934050011739675, + "loss": 0.126, + "step": 21904 + }, + { + "epoch": 0.190145918872232, + "grad_norm": 0.9296875, + "learning_rate": 0.0017933860272159312, + "loss": 0.0967, + "step": 21905 + }, + { + "epoch": 0.19015459935243617, + "grad_norm": 0.09033203125, + "learning_rate": 0.0017933670524996185, + "loss": 0.1016, + "step": 21906 + }, + { + "epoch": 0.19016327983264034, + "grad_norm": 0.6171875, + "learning_rate": 0.0017933480770250495, + "loss": 0.1162, + "step": 21907 + }, + { + "epoch": 0.1901719603128445, + "grad_norm": 0.12158203125, + "learning_rate": 0.0017933291007922456, + "loss": 0.0913, + "step": 21908 + }, + { + "epoch": 0.19018064079304867, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017933101238012273, + "loss": 0.127, + "step": 21909 + }, + { + "epoch": 0.19018932127325283, + "grad_norm": 0.12109375, + "learning_rate": 0.0017932911460520156, + "loss": 0.1118, + "step": 21910 + }, + { + "epoch": 0.190198001753457, + "grad_norm": 0.1171875, + "learning_rate": 0.0017932721675446312, + "loss": 0.1162, + "step": 21911 + }, + { + "epoch": 0.19020668223366116, + "grad_norm": 0.375, + "learning_rate": 0.0017932531882790949, + "loss": 0.1025, + "step": 21912 + }, + { + "epoch": 0.19021536271386533, + "grad_norm": 0.337890625, + "learning_rate": 0.001793234208255427, + "loss": 0.0894, + "step": 21913 + }, + { + "epoch": 0.1902240431940695, + "grad_norm": 0.349609375, + "learning_rate": 0.0017932152274736488, + "loss": 0.126, + "step": 21914 + }, + { + "epoch": 0.19023272367427366, + "grad_norm": 0.451171875, + "learning_rate": 0.0017931962459337807, + "loss": 0.0884, + "step": 21915 + }, + { + "epoch": 0.19024140415447782, + "grad_norm": 0.15625, + "learning_rate": 0.0017931772636358436, + "loss": 0.0996, + "step": 21916 + }, + { + "epoch": 0.190250084634682, + "grad_norm": 0.384765625, + "learning_rate": 0.0017931582805798586, + "loss": 0.1797, + "step": 21917 + }, + { + "epoch": 0.19025876511488615, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017931392967658459, + "loss": 0.1133, + "step": 21918 + }, + { + "epoch": 0.19026744559509032, + "grad_norm": 0.11083984375, + "learning_rate": 0.0017931203121938262, + "loss": 0.1147, + "step": 21919 + }, + { + "epoch": 0.19027612607529448, + "grad_norm": 0.10107421875, + "learning_rate": 0.0017931013268638208, + "loss": 0.103, + "step": 21920 + }, + { + "epoch": 0.19028480655549865, + "grad_norm": 0.51953125, + "learning_rate": 0.0017930823407758501, + "loss": 0.1045, + "step": 21921 + }, + { + "epoch": 0.1902934870357028, + "grad_norm": 0.62109375, + "learning_rate": 0.001793063353929935, + "loss": 0.1216, + "step": 21922 + }, + { + "epoch": 0.19030216751590698, + "grad_norm": 0.126953125, + "learning_rate": 0.0017930443663260966, + "loss": 0.1172, + "step": 21923 + }, + { + "epoch": 0.19031084799611114, + "grad_norm": 0.306640625, + "learning_rate": 0.0017930253779643549, + "loss": 0.1377, + "step": 21924 + }, + { + "epoch": 0.1903195284763153, + "grad_norm": 0.201171875, + "learning_rate": 0.001793006388844731, + "loss": 0.0996, + "step": 21925 + }, + { + "epoch": 0.19032820895651947, + "grad_norm": 0.515625, + "learning_rate": 0.001792987398967246, + "loss": 0.1543, + "step": 21926 + }, + { + "epoch": 0.19033688943672364, + "grad_norm": 0.302734375, + "learning_rate": 0.0017929684083319202, + "loss": 0.1318, + "step": 21927 + }, + { + "epoch": 0.1903455699169278, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017929494169387744, + "loss": 0.1177, + "step": 21928 + }, + { + "epoch": 0.19035425039713197, + "grad_norm": 0.345703125, + "learning_rate": 0.00179293042478783, + "loss": 0.1475, + "step": 21929 + }, + { + "epoch": 0.19036293087733613, + "grad_norm": 0.330078125, + "learning_rate": 0.0017929114318791068, + "loss": 0.125, + "step": 21930 + }, + { + "epoch": 0.1903716113575403, + "grad_norm": 0.6796875, + "learning_rate": 0.0017928924382126265, + "loss": 0.1152, + "step": 21931 + }, + { + "epoch": 0.19038029183774446, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017928734437884094, + "loss": 0.1484, + "step": 21932 + }, + { + "epoch": 0.19038897231794863, + "grad_norm": 1.3828125, + "learning_rate": 0.001792854448606476, + "loss": 0.1562, + "step": 21933 + }, + { + "epoch": 0.1903976527981528, + "grad_norm": 0.78515625, + "learning_rate": 0.0017928354526668474, + "loss": 0.1055, + "step": 21934 + }, + { + "epoch": 0.19040633327835696, + "grad_norm": 0.171875, + "learning_rate": 0.0017928164559695445, + "loss": 0.0835, + "step": 21935 + }, + { + "epoch": 0.19041501375856112, + "grad_norm": 0.287109375, + "learning_rate": 0.0017927974585145881, + "loss": 0.1133, + "step": 21936 + }, + { + "epoch": 0.1904236942387653, + "grad_norm": 0.11181640625, + "learning_rate": 0.0017927784603019986, + "loss": 0.1475, + "step": 21937 + }, + { + "epoch": 0.19043237471896945, + "grad_norm": 0.474609375, + "learning_rate": 0.0017927594613317969, + "loss": 0.1387, + "step": 21938 + }, + { + "epoch": 0.19044105519917362, + "grad_norm": 0.6171875, + "learning_rate": 0.001792740461604004, + "loss": 0.1143, + "step": 21939 + }, + { + "epoch": 0.19044973567937779, + "grad_norm": 1.078125, + "learning_rate": 0.0017927214611186406, + "loss": 0.1113, + "step": 21940 + }, + { + "epoch": 0.19045841615958195, + "grad_norm": 0.37890625, + "learning_rate": 0.001792702459875727, + "loss": 0.1221, + "step": 21941 + }, + { + "epoch": 0.19046709663978612, + "grad_norm": 0.095703125, + "learning_rate": 0.001792683457875285, + "loss": 0.083, + "step": 21942 + }, + { + "epoch": 0.19047577711999028, + "grad_norm": 0.19140625, + "learning_rate": 0.0017926644551173345, + "loss": 0.1128, + "step": 21943 + }, + { + "epoch": 0.19048445760019445, + "grad_norm": 0.55859375, + "learning_rate": 0.0017926454516018968, + "loss": 0.0977, + "step": 21944 + }, + { + "epoch": 0.1904931380803986, + "grad_norm": 0.98046875, + "learning_rate": 0.001792626447328992, + "loss": 0.0884, + "step": 21945 + }, + { + "epoch": 0.19050181856060278, + "grad_norm": 0.345703125, + "learning_rate": 0.0017926074422986416, + "loss": 0.1914, + "step": 21946 + }, + { + "epoch": 0.19051049904080694, + "grad_norm": 0.73046875, + "learning_rate": 0.001792588436510866, + "loss": 0.1113, + "step": 21947 + }, + { + "epoch": 0.1905191795210111, + "grad_norm": 0.2099609375, + "learning_rate": 0.0017925694299656863, + "loss": 0.1104, + "step": 21948 + }, + { + "epoch": 0.19052786000121527, + "grad_norm": 0.09912109375, + "learning_rate": 0.0017925504226631228, + "loss": 0.127, + "step": 21949 + }, + { + "epoch": 0.19053654048141944, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017925314146031968, + "loss": 0.1084, + "step": 21950 + }, + { + "epoch": 0.1905452209616236, + "grad_norm": 0.384765625, + "learning_rate": 0.0017925124057859287, + "loss": 0.082, + "step": 21951 + }, + { + "epoch": 0.19055390144182777, + "grad_norm": 0.58984375, + "learning_rate": 0.0017924933962113394, + "loss": 0.1602, + "step": 21952 + }, + { + "epoch": 0.19056258192203193, + "grad_norm": 0.353515625, + "learning_rate": 0.00179247438587945, + "loss": 0.1113, + "step": 21953 + }, + { + "epoch": 0.1905712624022361, + "grad_norm": 0.150390625, + "learning_rate": 0.001792455374790281, + "loss": 0.0928, + "step": 21954 + }, + { + "epoch": 0.19057994288244026, + "grad_norm": 0.83203125, + "learning_rate": 0.0017924363629438526, + "loss": 0.2812, + "step": 21955 + }, + { + "epoch": 0.19058862336264443, + "grad_norm": 1.21875, + "learning_rate": 0.0017924173503401867, + "loss": 0.1758, + "step": 21956 + }, + { + "epoch": 0.1905973038428486, + "grad_norm": 0.328125, + "learning_rate": 0.0017923983369793036, + "loss": 0.1279, + "step": 21957 + }, + { + "epoch": 0.19060598432305276, + "grad_norm": 0.431640625, + "learning_rate": 0.001792379322861224, + "loss": 0.0938, + "step": 21958 + }, + { + "epoch": 0.19061466480325692, + "grad_norm": 0.921875, + "learning_rate": 0.0017923603079859685, + "loss": 0.0986, + "step": 21959 + }, + { + "epoch": 0.1906233452834611, + "grad_norm": 0.33984375, + "learning_rate": 0.0017923412923535587, + "loss": 0.125, + "step": 21960 + }, + { + "epoch": 0.19063202576366525, + "grad_norm": 0.34375, + "learning_rate": 0.0017923222759640145, + "loss": 0.1387, + "step": 21961 + }, + { + "epoch": 0.19064070624386942, + "grad_norm": 0.49609375, + "learning_rate": 0.0017923032588173572, + "loss": 0.0898, + "step": 21962 + }, + { + "epoch": 0.19064938672407358, + "grad_norm": 0.490234375, + "learning_rate": 0.0017922842409136073, + "loss": 0.1064, + "step": 21963 + }, + { + "epoch": 0.19065806720427775, + "grad_norm": 0.93359375, + "learning_rate": 0.001792265222252786, + "loss": 0.1299, + "step": 21964 + }, + { + "epoch": 0.1906667476844819, + "grad_norm": 0.380859375, + "learning_rate": 0.0017922462028349137, + "loss": 0.0806, + "step": 21965 + }, + { + "epoch": 0.19067542816468608, + "grad_norm": 0.6484375, + "learning_rate": 0.0017922271826600114, + "loss": 0.1084, + "step": 21966 + }, + { + "epoch": 0.19068410864489024, + "grad_norm": 0.130859375, + "learning_rate": 0.0017922081617280999, + "loss": 0.1016, + "step": 21967 + }, + { + "epoch": 0.1906927891250944, + "grad_norm": 0.4453125, + "learning_rate": 0.0017921891400391999, + "loss": 0.0874, + "step": 21968 + }, + { + "epoch": 0.19070146960529857, + "grad_norm": 0.421875, + "learning_rate": 0.0017921701175933323, + "loss": 0.0972, + "step": 21969 + }, + { + "epoch": 0.19071015008550274, + "grad_norm": 1.390625, + "learning_rate": 0.0017921510943905176, + "loss": 0.1074, + "step": 21970 + }, + { + "epoch": 0.1907188305657069, + "grad_norm": 0.166015625, + "learning_rate": 0.001792132070430777, + "loss": 0.1084, + "step": 21971 + }, + { + "epoch": 0.19072751104591107, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017921130457141314, + "loss": 0.1621, + "step": 21972 + }, + { + "epoch": 0.19073619152611523, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017920940202406012, + "loss": 0.1123, + "step": 21973 + }, + { + "epoch": 0.1907448720063194, + "grad_norm": 0.37109375, + "learning_rate": 0.0017920749940102073, + "loss": 0.1289, + "step": 21974 + }, + { + "epoch": 0.19075355248652356, + "grad_norm": 0.150390625, + "learning_rate": 0.001792055967022971, + "loss": 0.1182, + "step": 21975 + }, + { + "epoch": 0.19076223296672773, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017920369392789121, + "loss": 0.0918, + "step": 21976 + }, + { + "epoch": 0.1907709134469319, + "grad_norm": 0.37109375, + "learning_rate": 0.0017920179107780523, + "loss": 0.0967, + "step": 21977 + }, + { + "epoch": 0.19077959392713606, + "grad_norm": 0.23828125, + "learning_rate": 0.0017919988815204122, + "loss": 0.0986, + "step": 21978 + }, + { + "epoch": 0.19078827440734022, + "grad_norm": 0.244140625, + "learning_rate": 0.0017919798515060122, + "loss": 0.0996, + "step": 21979 + }, + { + "epoch": 0.1907969548875444, + "grad_norm": 0.1748046875, + "learning_rate": 0.0017919608207348737, + "loss": 0.1211, + "step": 21980 + }, + { + "epoch": 0.19080563536774856, + "grad_norm": 0.6328125, + "learning_rate": 0.0017919417892070172, + "loss": 0.1406, + "step": 21981 + }, + { + "epoch": 0.19081431584795272, + "grad_norm": 0.419921875, + "learning_rate": 0.0017919227569224636, + "loss": 0.0898, + "step": 21982 + }, + { + "epoch": 0.19082299632815689, + "grad_norm": 0.75, + "learning_rate": 0.0017919037238812335, + "loss": 0.1138, + "step": 21983 + }, + { + "epoch": 0.19083167680836105, + "grad_norm": 0.0732421875, + "learning_rate": 0.001791884690083348, + "loss": 0.0732, + "step": 21984 + }, + { + "epoch": 0.19084035728856522, + "grad_norm": 0.46484375, + "learning_rate": 0.0017918656555288277, + "loss": 0.1123, + "step": 21985 + }, + { + "epoch": 0.19084903776876938, + "grad_norm": 0.2578125, + "learning_rate": 0.0017918466202176938, + "loss": 0.126, + "step": 21986 + }, + { + "epoch": 0.19085771824897355, + "grad_norm": 0.60546875, + "learning_rate": 0.0017918275841499664, + "loss": 0.0986, + "step": 21987 + }, + { + "epoch": 0.1908663987291777, + "grad_norm": 0.8359375, + "learning_rate": 0.0017918085473256672, + "loss": 0.1338, + "step": 21988 + }, + { + "epoch": 0.19087507920938188, + "grad_norm": 0.259765625, + "learning_rate": 0.001791789509744816, + "loss": 0.123, + "step": 21989 + }, + { + "epoch": 0.19088375968958604, + "grad_norm": 0.5859375, + "learning_rate": 0.0017917704714074348, + "loss": 0.1357, + "step": 21990 + }, + { + "epoch": 0.1908924401697902, + "grad_norm": 0.5234375, + "learning_rate": 0.0017917514323135433, + "loss": 0.125, + "step": 21991 + }, + { + "epoch": 0.19090112064999437, + "grad_norm": 0.4453125, + "learning_rate": 0.0017917323924631632, + "loss": 0.1562, + "step": 21992 + }, + { + "epoch": 0.1909098011301985, + "grad_norm": 0.41796875, + "learning_rate": 0.0017917133518563148, + "loss": 0.105, + "step": 21993 + }, + { + "epoch": 0.19091848161040267, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017916943104930192, + "loss": 0.1338, + "step": 21994 + }, + { + "epoch": 0.19092716209060684, + "grad_norm": 0.28125, + "learning_rate": 0.0017916752683732968, + "loss": 0.0942, + "step": 21995 + }, + { + "epoch": 0.190935842570811, + "grad_norm": 0.189453125, + "learning_rate": 0.0017916562254971689, + "loss": 0.0986, + "step": 21996 + }, + { + "epoch": 0.19094452305101517, + "grad_norm": 0.68359375, + "learning_rate": 0.0017916371818646562, + "loss": 0.127, + "step": 21997 + }, + { + "epoch": 0.19095320353121933, + "grad_norm": 0.8203125, + "learning_rate": 0.0017916181374757795, + "loss": 0.1318, + "step": 21998 + }, + { + "epoch": 0.1909618840114235, + "grad_norm": 0.197265625, + "learning_rate": 0.0017915990923305595, + "loss": 0.1719, + "step": 21999 + }, + { + "epoch": 0.19097056449162766, + "grad_norm": 0.5859375, + "learning_rate": 0.001791580046429017, + "loss": 0.1367, + "step": 22000 + }, + { + "epoch": 0.19097924497183183, + "grad_norm": 0.0771484375, + "learning_rate": 0.0017915609997711732, + "loss": 0.0952, + "step": 22001 + }, + { + "epoch": 0.190987925452036, + "grad_norm": 0.353515625, + "learning_rate": 0.0017915419523570485, + "loss": 0.105, + "step": 22002 + }, + { + "epoch": 0.19099660593224016, + "grad_norm": 0.2041015625, + "learning_rate": 0.001791522904186664, + "loss": 0.1709, + "step": 22003 + }, + { + "epoch": 0.19100528641244433, + "grad_norm": 0.2890625, + "learning_rate": 0.0017915038552600402, + "loss": 0.1084, + "step": 22004 + }, + { + "epoch": 0.1910139668926485, + "grad_norm": 0.392578125, + "learning_rate": 0.0017914848055771986, + "loss": 0.1494, + "step": 22005 + }, + { + "epoch": 0.19102264737285266, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017914657551381594, + "loss": 0.1099, + "step": 22006 + }, + { + "epoch": 0.19103132785305682, + "grad_norm": 0.5703125, + "learning_rate": 0.0017914467039429436, + "loss": 0.1108, + "step": 22007 + }, + { + "epoch": 0.19104000833326099, + "grad_norm": 0.10888671875, + "learning_rate": 0.001791427651991572, + "loss": 0.1377, + "step": 22008 + }, + { + "epoch": 0.19104868881346515, + "grad_norm": 0.59375, + "learning_rate": 0.0017914085992840657, + "loss": 0.124, + "step": 22009 + }, + { + "epoch": 0.19105736929366932, + "grad_norm": 0.34375, + "learning_rate": 0.0017913895458204453, + "loss": 0.1196, + "step": 22010 + }, + { + "epoch": 0.19106604977387348, + "grad_norm": 0.23046875, + "learning_rate": 0.0017913704916007317, + "loss": 0.0972, + "step": 22011 + }, + { + "epoch": 0.19107473025407765, + "grad_norm": 0.306640625, + "learning_rate": 0.0017913514366249457, + "loss": 0.1309, + "step": 22012 + }, + { + "epoch": 0.1910834107342818, + "grad_norm": 0.54296875, + "learning_rate": 0.001791332380893108, + "loss": 0.1143, + "step": 22013 + }, + { + "epoch": 0.19109209121448598, + "grad_norm": 0.61328125, + "learning_rate": 0.0017913133244052398, + "loss": 0.0898, + "step": 22014 + }, + { + "epoch": 0.19110077169469014, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017912942671613615, + "loss": 0.1021, + "step": 22015 + }, + { + "epoch": 0.1911094521748943, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017912752091614945, + "loss": 0.0928, + "step": 22016 + }, + { + "epoch": 0.19111813265509847, + "grad_norm": 0.48828125, + "learning_rate": 0.001791256150405659, + "loss": 0.6172, + "step": 22017 + }, + { + "epoch": 0.19112681313530264, + "grad_norm": 0.498046875, + "learning_rate": 0.0017912370908938762, + "loss": 0.1064, + "step": 22018 + }, + { + "epoch": 0.1911354936155068, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017912180306261672, + "loss": 0.1445, + "step": 22019 + }, + { + "epoch": 0.19114417409571097, + "grad_norm": 4.21875, + "learning_rate": 0.0017911989696025522, + "loss": 0.2539, + "step": 22020 + }, + { + "epoch": 0.19115285457591513, + "grad_norm": 0.09765625, + "learning_rate": 0.0017911799078230526, + "loss": 0.1289, + "step": 22021 + }, + { + "epoch": 0.1911615350561193, + "grad_norm": 0.10400390625, + "learning_rate": 0.001791160845287689, + "loss": 0.1035, + "step": 22022 + }, + { + "epoch": 0.19117021553632346, + "grad_norm": 0.80859375, + "learning_rate": 0.0017911417819964821, + "loss": 0.1318, + "step": 22023 + }, + { + "epoch": 0.19117889601652763, + "grad_norm": 0.51171875, + "learning_rate": 0.0017911227179494532, + "loss": 0.1084, + "step": 22024 + }, + { + "epoch": 0.1911875764967318, + "grad_norm": 0.291015625, + "learning_rate": 0.0017911036531466228, + "loss": 0.1162, + "step": 22025 + }, + { + "epoch": 0.19119625697693596, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017910845875880117, + "loss": 0.1172, + "step": 22026 + }, + { + "epoch": 0.19120493745714012, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017910655212736413, + "loss": 0.1172, + "step": 22027 + }, + { + "epoch": 0.1912136179373443, + "grad_norm": 0.7890625, + "learning_rate": 0.0017910464542035317, + "loss": 0.0938, + "step": 22028 + }, + { + "epoch": 0.19122229841754845, + "grad_norm": 0.94140625, + "learning_rate": 0.001791027386377704, + "loss": 0.1582, + "step": 22029 + }, + { + "epoch": 0.19123097889775262, + "grad_norm": 0.212890625, + "learning_rate": 0.001791008317796179, + "loss": 0.1191, + "step": 22030 + }, + { + "epoch": 0.19123965937795678, + "grad_norm": 0.1943359375, + "learning_rate": 0.001790989248458978, + "loss": 0.1074, + "step": 22031 + }, + { + "epoch": 0.19124833985816095, + "grad_norm": 0.197265625, + "learning_rate": 0.0017909701783661214, + "loss": 0.1035, + "step": 22032 + }, + { + "epoch": 0.1912570203383651, + "grad_norm": 0.33203125, + "learning_rate": 0.00179095110751763, + "loss": 0.1465, + "step": 22033 + }, + { + "epoch": 0.19126570081856928, + "grad_norm": 0.515625, + "learning_rate": 0.0017909320359135255, + "loss": 0.105, + "step": 22034 + }, + { + "epoch": 0.19127438129877344, + "grad_norm": 0.53125, + "learning_rate": 0.0017909129635538276, + "loss": 0.1094, + "step": 22035 + }, + { + "epoch": 0.1912830617789776, + "grad_norm": 0.345703125, + "learning_rate": 0.001790893890438558, + "loss": 0.1147, + "step": 22036 + }, + { + "epoch": 0.19129174225918177, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017908748165677366, + "loss": 0.124, + "step": 22037 + }, + { + "epoch": 0.19130042273938594, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017908557419413856, + "loss": 0.1123, + "step": 22038 + }, + { + "epoch": 0.1913091032195901, + "grad_norm": 0.427734375, + "learning_rate": 0.0017908366665595247, + "loss": 0.1133, + "step": 22039 + }, + { + "epoch": 0.19131778369979427, + "grad_norm": 0.72265625, + "learning_rate": 0.0017908175904221752, + "loss": 0.0962, + "step": 22040 + }, + { + "epoch": 0.19132646417999843, + "grad_norm": 0.30859375, + "learning_rate": 0.0017907985135293583, + "loss": 0.124, + "step": 22041 + }, + { + "epoch": 0.1913351446602026, + "grad_norm": 0.140625, + "learning_rate": 0.001790779435881094, + "loss": 0.0957, + "step": 22042 + }, + { + "epoch": 0.19134382514040676, + "grad_norm": 0.09375, + "learning_rate": 0.0017907603574774042, + "loss": 0.0898, + "step": 22043 + }, + { + "epoch": 0.19135250562061093, + "grad_norm": 0.330078125, + "learning_rate": 0.0017907412783183087, + "loss": 0.0869, + "step": 22044 + }, + { + "epoch": 0.1913611861008151, + "grad_norm": 0.15625, + "learning_rate": 0.0017907221984038296, + "loss": 0.0957, + "step": 22045 + }, + { + "epoch": 0.19136986658101926, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017907031177339867, + "loss": 0.1035, + "step": 22046 + }, + { + "epoch": 0.19137854706122343, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017906840363088013, + "loss": 0.1001, + "step": 22047 + }, + { + "epoch": 0.1913872275414276, + "grad_norm": 0.271484375, + "learning_rate": 0.001790664954128294, + "loss": 0.0898, + "step": 22048 + }, + { + "epoch": 0.19139590802163176, + "grad_norm": 0.083984375, + "learning_rate": 0.0017906458711924864, + "loss": 0.1025, + "step": 22049 + }, + { + "epoch": 0.19140458850183592, + "grad_norm": 0.150390625, + "learning_rate": 0.0017906267875013983, + "loss": 0.1006, + "step": 22050 + }, + { + "epoch": 0.19141326898204009, + "grad_norm": 0.12060546875, + "learning_rate": 0.0017906077030550513, + "loss": 0.1553, + "step": 22051 + }, + { + "epoch": 0.19142194946224425, + "grad_norm": 0.07861328125, + "learning_rate": 0.0017905886178534662, + "loss": 0.1094, + "step": 22052 + }, + { + "epoch": 0.19143062994244842, + "grad_norm": 0.1953125, + "learning_rate": 0.0017905695318966638, + "loss": 0.1367, + "step": 22053 + }, + { + "epoch": 0.19143931042265258, + "grad_norm": 0.265625, + "learning_rate": 0.0017905504451846649, + "loss": 0.1279, + "step": 22054 + }, + { + "epoch": 0.19144799090285675, + "grad_norm": 0.06640625, + "learning_rate": 0.0017905313577174901, + "loss": 0.0762, + "step": 22055 + }, + { + "epoch": 0.1914566713830609, + "grad_norm": 0.6875, + "learning_rate": 0.0017905122694951612, + "loss": 0.1172, + "step": 22056 + }, + { + "epoch": 0.19146535186326508, + "grad_norm": 0.4765625, + "learning_rate": 0.001790493180517698, + "loss": 0.0815, + "step": 22057 + }, + { + "epoch": 0.19147403234346924, + "grad_norm": 0.08154296875, + "learning_rate": 0.0017904740907851218, + "loss": 0.0923, + "step": 22058 + }, + { + "epoch": 0.1914827128236734, + "grad_norm": 0.150390625, + "learning_rate": 0.0017904550002974537, + "loss": 0.1357, + "step": 22059 + }, + { + "epoch": 0.19149139330387757, + "grad_norm": 0.083984375, + "learning_rate": 0.0017904359090547145, + "loss": 0.0894, + "step": 22060 + }, + { + "epoch": 0.19150007378408174, + "grad_norm": 0.228515625, + "learning_rate": 0.0017904168170569245, + "loss": 0.1021, + "step": 22061 + }, + { + "epoch": 0.1915087542642859, + "grad_norm": 0.58203125, + "learning_rate": 0.0017903977243041054, + "loss": 0.1377, + "step": 22062 + }, + { + "epoch": 0.19151743474449007, + "grad_norm": 0.39453125, + "learning_rate": 0.0017903786307962776, + "loss": 0.1475, + "step": 22063 + }, + { + "epoch": 0.19152611522469423, + "grad_norm": 0.166015625, + "learning_rate": 0.0017903595365334625, + "loss": 0.1108, + "step": 22064 + }, + { + "epoch": 0.1915347957048984, + "grad_norm": 0.443359375, + "learning_rate": 0.00179034044151568, + "loss": 0.0918, + "step": 22065 + }, + { + "epoch": 0.19154347618510256, + "grad_norm": 0.0751953125, + "learning_rate": 0.0017903213457429517, + "loss": 0.0767, + "step": 22066 + }, + { + "epoch": 0.19155215666530673, + "grad_norm": 0.400390625, + "learning_rate": 0.0017903022492152983, + "loss": 0.0947, + "step": 22067 + }, + { + "epoch": 0.1915608371455109, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017902831519327408, + "loss": 0.1299, + "step": 22068 + }, + { + "epoch": 0.19156951762571506, + "grad_norm": 0.73828125, + "learning_rate": 0.0017902640538953003, + "loss": 0.1289, + "step": 22069 + }, + { + "epoch": 0.19157819810591922, + "grad_norm": 0.41015625, + "learning_rate": 0.0017902449551029976, + "loss": 0.0884, + "step": 22070 + }, + { + "epoch": 0.1915868785861234, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017902258555558526, + "loss": 0.0981, + "step": 22071 + }, + { + "epoch": 0.19159555906632755, + "grad_norm": 0.201171875, + "learning_rate": 0.0017902067552538876, + "loss": 0.085, + "step": 22072 + }, + { + "epoch": 0.19160423954653172, + "grad_norm": 0.71484375, + "learning_rate": 0.0017901876541971224, + "loss": 0.124, + "step": 22073 + }, + { + "epoch": 0.19161292002673588, + "grad_norm": 1.03125, + "learning_rate": 0.0017901685523855789, + "loss": 0.0874, + "step": 22074 + }, + { + "epoch": 0.19162160050694005, + "grad_norm": 0.28125, + "learning_rate": 0.0017901494498192767, + "loss": 0.1133, + "step": 22075 + }, + { + "epoch": 0.1916302809871442, + "grad_norm": 0.3984375, + "learning_rate": 0.001790130346498238, + "loss": 0.1177, + "step": 22076 + }, + { + "epoch": 0.19163896146734838, + "grad_norm": 0.1279296875, + "learning_rate": 0.001790111242422483, + "loss": 0.1123, + "step": 22077 + }, + { + "epoch": 0.19164764194755254, + "grad_norm": 0.07763671875, + "learning_rate": 0.0017900921375920327, + "loss": 0.0791, + "step": 22078 + }, + { + "epoch": 0.1916563224277567, + "grad_norm": 0.5703125, + "learning_rate": 0.001790073032006908, + "loss": 0.1055, + "step": 22079 + }, + { + "epoch": 0.19166500290796087, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017900539256671298, + "loss": 0.1191, + "step": 22080 + }, + { + "epoch": 0.19167368338816504, + "grad_norm": 0.314453125, + "learning_rate": 0.0017900348185727192, + "loss": 0.1055, + "step": 22081 + }, + { + "epoch": 0.1916823638683692, + "grad_norm": 0.51953125, + "learning_rate": 0.0017900157107236963, + "loss": 0.1582, + "step": 22082 + }, + { + "epoch": 0.19169104434857337, + "grad_norm": 0.53125, + "learning_rate": 0.001789996602120083, + "loss": 0.0952, + "step": 22083 + }, + { + "epoch": 0.19169972482877753, + "grad_norm": 0.150390625, + "learning_rate": 0.0017899774927619, + "loss": 0.1777, + "step": 22084 + }, + { + "epoch": 0.1917084053089817, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017899583826491677, + "loss": 0.127, + "step": 22085 + }, + { + "epoch": 0.19171708578918586, + "grad_norm": 0.357421875, + "learning_rate": 0.0017899392717819072, + "loss": 0.1064, + "step": 22086 + }, + { + "epoch": 0.19172576626939003, + "grad_norm": 0.734375, + "learning_rate": 0.0017899201601601397, + "loss": 0.0835, + "step": 22087 + }, + { + "epoch": 0.1917344467495942, + "grad_norm": 0.462890625, + "learning_rate": 0.0017899010477838856, + "loss": 0.1133, + "step": 22088 + }, + { + "epoch": 0.19174312722979836, + "grad_norm": 0.64453125, + "learning_rate": 0.0017898819346531664, + "loss": 0.125, + "step": 22089 + }, + { + "epoch": 0.19175180771000253, + "grad_norm": 0.107421875, + "learning_rate": 0.0017898628207680023, + "loss": 0.1182, + "step": 22090 + }, + { + "epoch": 0.1917604881902067, + "grad_norm": 0.33984375, + "learning_rate": 0.001789843706128415, + "loss": 0.1064, + "step": 22091 + }, + { + "epoch": 0.19176916867041086, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017898245907344247, + "loss": 0.0923, + "step": 22092 + }, + { + "epoch": 0.19177784915061502, + "grad_norm": 1.125, + "learning_rate": 0.0017898054745860527, + "loss": 0.1172, + "step": 22093 + }, + { + "epoch": 0.19178652963081919, + "grad_norm": 0.14453125, + "learning_rate": 0.0017897863576833198, + "loss": 0.0791, + "step": 22094 + }, + { + "epoch": 0.19179521011102335, + "grad_norm": 0.451171875, + "learning_rate": 0.001789767240026247, + "loss": 0.0938, + "step": 22095 + }, + { + "epoch": 0.19180389059122752, + "grad_norm": 0.5859375, + "learning_rate": 0.001789748121614855, + "loss": 0.1289, + "step": 22096 + }, + { + "epoch": 0.19181257107143168, + "grad_norm": 0.478515625, + "learning_rate": 0.0017897290024491647, + "loss": 0.0767, + "step": 22097 + }, + { + "epoch": 0.19182125155163585, + "grad_norm": 0.36328125, + "learning_rate": 0.0017897098825291973, + "loss": 0.1348, + "step": 22098 + }, + { + "epoch": 0.19182993203184, + "grad_norm": 0.12109375, + "learning_rate": 0.0017896907618549735, + "loss": 0.0796, + "step": 22099 + }, + { + "epoch": 0.19183861251204418, + "grad_norm": 0.306640625, + "learning_rate": 0.0017896716404265142, + "loss": 0.1426, + "step": 22100 + }, + { + "epoch": 0.19184729299224834, + "grad_norm": 0.337890625, + "learning_rate": 0.0017896525182438403, + "loss": 0.1123, + "step": 22101 + }, + { + "epoch": 0.1918559734724525, + "grad_norm": 0.24609375, + "learning_rate": 0.001789633395306973, + "loss": 0.0986, + "step": 22102 + }, + { + "epoch": 0.19186465395265667, + "grad_norm": 0.515625, + "learning_rate": 0.0017896142716159328, + "loss": 0.3125, + "step": 22103 + }, + { + "epoch": 0.19187333443286084, + "grad_norm": 0.314453125, + "learning_rate": 0.0017895951471707408, + "loss": 0.1328, + "step": 22104 + }, + { + "epoch": 0.191882014913065, + "grad_norm": 0.091796875, + "learning_rate": 0.001789576021971418, + "loss": 0.0889, + "step": 22105 + }, + { + "epoch": 0.19189069539326917, + "grad_norm": 0.439453125, + "learning_rate": 0.001789556896017985, + "loss": 0.0889, + "step": 22106 + }, + { + "epoch": 0.19189937587347333, + "grad_norm": 0.427734375, + "learning_rate": 0.001789537769310463, + "loss": 0.1279, + "step": 22107 + }, + { + "epoch": 0.1919080563536775, + "grad_norm": 0.486328125, + "learning_rate": 0.0017895186418488732, + "loss": 0.1426, + "step": 22108 + }, + { + "epoch": 0.19191673683388166, + "grad_norm": 0.181640625, + "learning_rate": 0.001789499513633236, + "loss": 0.1162, + "step": 22109 + }, + { + "epoch": 0.19192541731408583, + "grad_norm": 0.140625, + "learning_rate": 0.001789480384663572, + "loss": 0.1094, + "step": 22110 + }, + { + "epoch": 0.19193409779429, + "grad_norm": 0.13671875, + "learning_rate": 0.0017894612549399032, + "loss": 0.0957, + "step": 22111 + }, + { + "epoch": 0.19194277827449416, + "grad_norm": 0.5390625, + "learning_rate": 0.0017894421244622496, + "loss": 0.0845, + "step": 22112 + }, + { + "epoch": 0.19195145875469832, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017894229932306327, + "loss": 0.1211, + "step": 22113 + }, + { + "epoch": 0.1919601392349025, + "grad_norm": 0.296875, + "learning_rate": 0.001789403861245073, + "loss": 0.1074, + "step": 22114 + }, + { + "epoch": 0.19196881971510665, + "grad_norm": 0.2099609375, + "learning_rate": 0.0017893847285055918, + "loss": 0.1562, + "step": 22115 + }, + { + "epoch": 0.19197750019531082, + "grad_norm": 0.6640625, + "learning_rate": 0.0017893655950122097, + "loss": 0.1055, + "step": 22116 + }, + { + "epoch": 0.19198618067551496, + "grad_norm": 0.09423828125, + "learning_rate": 0.001789346460764948, + "loss": 0.1011, + "step": 22117 + }, + { + "epoch": 0.19199486115571912, + "grad_norm": 0.150390625, + "learning_rate": 0.001789327325763827, + "loss": 0.1016, + "step": 22118 + }, + { + "epoch": 0.19200354163592329, + "grad_norm": 0.306640625, + "learning_rate": 0.001789308190008868, + "loss": 0.1309, + "step": 22119 + }, + { + "epoch": 0.19201222211612745, + "grad_norm": 0.21484375, + "learning_rate": 0.001789289053500092, + "loss": 0.0972, + "step": 22120 + }, + { + "epoch": 0.19202090259633162, + "grad_norm": 0.298828125, + "learning_rate": 0.00178926991623752, + "loss": 0.1084, + "step": 22121 + }, + { + "epoch": 0.19202958307653578, + "grad_norm": 1.09375, + "learning_rate": 0.0017892507782211727, + "loss": 0.1221, + "step": 22122 + }, + { + "epoch": 0.19203826355673995, + "grad_norm": 0.578125, + "learning_rate": 0.0017892316394510707, + "loss": 0.1357, + "step": 22123 + }, + { + "epoch": 0.1920469440369441, + "grad_norm": 0.189453125, + "learning_rate": 0.001789212499927236, + "loss": 0.0918, + "step": 22124 + }, + { + "epoch": 0.19205562451714828, + "grad_norm": 0.50390625, + "learning_rate": 0.0017891933596496884, + "loss": 0.106, + "step": 22125 + }, + { + "epoch": 0.19206430499735244, + "grad_norm": 0.44140625, + "learning_rate": 0.0017891742186184494, + "loss": 0.1162, + "step": 22126 + }, + { + "epoch": 0.1920729854775566, + "grad_norm": 0.09912109375, + "learning_rate": 0.0017891550768335401, + "loss": 0.1113, + "step": 22127 + }, + { + "epoch": 0.19208166595776077, + "grad_norm": 0.6328125, + "learning_rate": 0.0017891359342949806, + "loss": 0.1064, + "step": 22128 + }, + { + "epoch": 0.19209034643796494, + "grad_norm": 0.5546875, + "learning_rate": 0.001789116791002793, + "loss": 0.1797, + "step": 22129 + }, + { + "epoch": 0.1920990269181691, + "grad_norm": 0.12890625, + "learning_rate": 0.0017890976469569972, + "loss": 0.1045, + "step": 22130 + }, + { + "epoch": 0.19210770739837327, + "grad_norm": 0.271484375, + "learning_rate": 0.001789078502157615, + "loss": 0.0718, + "step": 22131 + }, + { + "epoch": 0.19211638787857743, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017890593566046665, + "loss": 0.1035, + "step": 22132 + }, + { + "epoch": 0.1921250683587816, + "grad_norm": 0.470703125, + "learning_rate": 0.001789040210298173, + "loss": 0.1504, + "step": 22133 + }, + { + "epoch": 0.19213374883898576, + "grad_norm": 1.0, + "learning_rate": 0.0017890210632381558, + "loss": 0.1348, + "step": 22134 + }, + { + "epoch": 0.19214242931918993, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017890019154246353, + "loss": 0.1108, + "step": 22135 + }, + { + "epoch": 0.1921511097993941, + "grad_norm": 0.74609375, + "learning_rate": 0.0017889827668576327, + "loss": 0.1201, + "step": 22136 + }, + { + "epoch": 0.19215979027959826, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017889636175371692, + "loss": 0.1035, + "step": 22137 + }, + { + "epoch": 0.19216847075980242, + "grad_norm": 0.4296875, + "learning_rate": 0.001788944467463265, + "loss": 0.1133, + "step": 22138 + }, + { + "epoch": 0.1921771512400066, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017889253166359417, + "loss": 0.125, + "step": 22139 + }, + { + "epoch": 0.19218583172021075, + "grad_norm": 0.21875, + "learning_rate": 0.0017889061650552198, + "loss": 0.1064, + "step": 22140 + }, + { + "epoch": 0.19219451220041492, + "grad_norm": 0.119140625, + "learning_rate": 0.0017888870127211207, + "loss": 0.1201, + "step": 22141 + }, + { + "epoch": 0.19220319268061908, + "grad_norm": 0.5390625, + "learning_rate": 0.0017888678596336652, + "loss": 0.1113, + "step": 22142 + }, + { + "epoch": 0.19221187316082325, + "grad_norm": 0.1142578125, + "learning_rate": 0.0017888487057928738, + "loss": 0.0874, + "step": 22143 + }, + { + "epoch": 0.19222055364102741, + "grad_norm": 0.11328125, + "learning_rate": 0.0017888295511987683, + "loss": 0.0879, + "step": 22144 + }, + { + "epoch": 0.19222923412123158, + "grad_norm": 0.40234375, + "learning_rate": 0.0017888103958513687, + "loss": 0.1328, + "step": 22145 + }, + { + "epoch": 0.19223791460143574, + "grad_norm": 0.45703125, + "learning_rate": 0.0017887912397506964, + "loss": 0.1582, + "step": 22146 + }, + { + "epoch": 0.1922465950816399, + "grad_norm": 1.109375, + "learning_rate": 0.0017887720828967727, + "loss": 0.1147, + "step": 22147 + }, + { + "epoch": 0.19225527556184407, + "grad_norm": 0.56640625, + "learning_rate": 0.001788752925289618, + "loss": 0.0918, + "step": 22148 + }, + { + "epoch": 0.19226395604204824, + "grad_norm": 0.0947265625, + "learning_rate": 0.0017887337669292534, + "loss": 0.1152, + "step": 22149 + }, + { + "epoch": 0.1922726365222524, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017887146078157002, + "loss": 0.1074, + "step": 22150 + }, + { + "epoch": 0.19228131700245657, + "grad_norm": 0.279296875, + "learning_rate": 0.0017886954479489786, + "loss": 0.1118, + "step": 22151 + }, + { + "epoch": 0.19228999748266073, + "grad_norm": 0.97265625, + "learning_rate": 0.0017886762873291103, + "loss": 0.1387, + "step": 22152 + }, + { + "epoch": 0.1922986779628649, + "grad_norm": 0.09521484375, + "learning_rate": 0.001788657125956116, + "loss": 0.1396, + "step": 22153 + }, + { + "epoch": 0.19230735844306907, + "grad_norm": 0.236328125, + "learning_rate": 0.0017886379638300165, + "loss": 0.1016, + "step": 22154 + }, + { + "epoch": 0.19231603892327323, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017886188009508328, + "loss": 0.125, + "step": 22155 + }, + { + "epoch": 0.1923247194034774, + "grad_norm": 0.263671875, + "learning_rate": 0.001788599637318586, + "loss": 0.1328, + "step": 22156 + }, + { + "epoch": 0.19233339988368156, + "grad_norm": 1.609375, + "learning_rate": 0.0017885804729332968, + "loss": 0.127, + "step": 22157 + }, + { + "epoch": 0.19234208036388573, + "grad_norm": 0.66015625, + "learning_rate": 0.0017885613077949867, + "loss": 0.1201, + "step": 22158 + }, + { + "epoch": 0.1923507608440899, + "grad_norm": 0.56640625, + "learning_rate": 0.001788542141903676, + "loss": 0.0913, + "step": 22159 + }, + { + "epoch": 0.19235944132429406, + "grad_norm": 0.3125, + "learning_rate": 0.0017885229752593861, + "loss": 0.1201, + "step": 22160 + }, + { + "epoch": 0.19236812180449822, + "grad_norm": 0.50390625, + "learning_rate": 0.0017885038078621377, + "loss": 0.1147, + "step": 22161 + }, + { + "epoch": 0.1923768022847024, + "grad_norm": 0.228515625, + "learning_rate": 0.0017884846397119524, + "loss": 0.126, + "step": 22162 + }, + { + "epoch": 0.19238548276490655, + "grad_norm": 0.69921875, + "learning_rate": 0.00178846547080885, + "loss": 0.1377, + "step": 22163 + }, + { + "epoch": 0.19239416324511072, + "grad_norm": 0.1748046875, + "learning_rate": 0.0017884463011528522, + "loss": 0.1128, + "step": 22164 + }, + { + "epoch": 0.19240284372531488, + "grad_norm": 0.1728515625, + "learning_rate": 0.00178842713074398, + "loss": 0.1338, + "step": 22165 + }, + { + "epoch": 0.19241152420551905, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017884079595822545, + "loss": 0.0728, + "step": 22166 + }, + { + "epoch": 0.1924202046857232, + "grad_norm": 0.236328125, + "learning_rate": 0.001788388787667696, + "loss": 0.126, + "step": 22167 + }, + { + "epoch": 0.19242888516592738, + "grad_norm": 0.87890625, + "learning_rate": 0.001788369615000326, + "loss": 0.0977, + "step": 22168 + }, + { + "epoch": 0.19243756564613154, + "grad_norm": 0.0859375, + "learning_rate": 0.0017883504415801652, + "loss": 0.0859, + "step": 22169 + }, + { + "epoch": 0.1924462461263357, + "grad_norm": 0.1455078125, + "learning_rate": 0.001788331267407235, + "loss": 0.123, + "step": 22170 + }, + { + "epoch": 0.19245492660653987, + "grad_norm": 0.16015625, + "learning_rate": 0.0017883120924815556, + "loss": 0.1201, + "step": 22171 + }, + { + "epoch": 0.19246360708674404, + "grad_norm": 0.1962890625, + "learning_rate": 0.001788292916803149, + "loss": 0.1465, + "step": 22172 + }, + { + "epoch": 0.1924722875669482, + "grad_norm": 0.60546875, + "learning_rate": 0.0017882737403720351, + "loss": 0.1016, + "step": 22173 + }, + { + "epoch": 0.19248096804715237, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017882545631882358, + "loss": 0.1426, + "step": 22174 + }, + { + "epoch": 0.19248964852735653, + "grad_norm": 0.11962890625, + "learning_rate": 0.0017882353852517712, + "loss": 0.0845, + "step": 22175 + }, + { + "epoch": 0.1924983290075607, + "grad_norm": 0.23828125, + "learning_rate": 0.0017882162065626633, + "loss": 0.1079, + "step": 22176 + }, + { + "epoch": 0.19250700948776486, + "grad_norm": 0.369140625, + "learning_rate": 0.001788197027120932, + "loss": 0.1094, + "step": 22177 + }, + { + "epoch": 0.19251568996796903, + "grad_norm": 0.1826171875, + "learning_rate": 0.0017881778469265991, + "loss": 0.126, + "step": 22178 + }, + { + "epoch": 0.1925243704481732, + "grad_norm": 0.072265625, + "learning_rate": 0.0017881586659796852, + "loss": 0.0962, + "step": 22179 + }, + { + "epoch": 0.19253305092837736, + "grad_norm": 0.111328125, + "learning_rate": 0.001788139484280211, + "loss": 0.0713, + "step": 22180 + }, + { + "epoch": 0.19254173140858152, + "grad_norm": 0.326171875, + "learning_rate": 0.001788120301828198, + "loss": 0.1235, + "step": 22181 + }, + { + "epoch": 0.1925504118887857, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017881011186236674, + "loss": 0.1045, + "step": 22182 + }, + { + "epoch": 0.19255909236898985, + "grad_norm": 0.134765625, + "learning_rate": 0.001788081934666639, + "loss": 0.1094, + "step": 22183 + }, + { + "epoch": 0.19256777284919402, + "grad_norm": 0.10791015625, + "learning_rate": 0.001788062749957135, + "loss": 0.085, + "step": 22184 + }, + { + "epoch": 0.19257645332939818, + "grad_norm": 0.330078125, + "learning_rate": 0.0017880435644951759, + "loss": 0.0771, + "step": 22185 + }, + { + "epoch": 0.19258513380960235, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017880243782807826, + "loss": 0.0981, + "step": 22186 + }, + { + "epoch": 0.19259381428980651, + "grad_norm": 0.34765625, + "learning_rate": 0.0017880051913139758, + "loss": 0.1104, + "step": 22187 + }, + { + "epoch": 0.19260249477001068, + "grad_norm": 0.28125, + "learning_rate": 0.0017879860035947778, + "loss": 0.0908, + "step": 22188 + }, + { + "epoch": 0.19261117525021484, + "grad_norm": 0.482421875, + "learning_rate": 0.001787966815123208, + "loss": 0.1069, + "step": 22189 + }, + { + "epoch": 0.192619855730419, + "grad_norm": 0.25390625, + "learning_rate": 0.001787947625899288, + "loss": 0.0918, + "step": 22190 + }, + { + "epoch": 0.19262853621062317, + "grad_norm": 0.4921875, + "learning_rate": 0.0017879284359230388, + "loss": 0.1191, + "step": 22191 + }, + { + "epoch": 0.19263721669082734, + "grad_norm": 0.66796875, + "learning_rate": 0.0017879092451944813, + "loss": 0.124, + "step": 22192 + }, + { + "epoch": 0.1926458971710315, + "grad_norm": 0.455078125, + "learning_rate": 0.0017878900537136371, + "loss": 0.165, + "step": 22193 + }, + { + "epoch": 0.19265457765123567, + "grad_norm": 0.283203125, + "learning_rate": 0.0017878708614805265, + "loss": 0.0903, + "step": 22194 + }, + { + "epoch": 0.19266325813143984, + "grad_norm": 1.640625, + "learning_rate": 0.0017878516684951703, + "loss": 0.1484, + "step": 22195 + }, + { + "epoch": 0.192671938611644, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017878324747575903, + "loss": 0.1226, + "step": 22196 + }, + { + "epoch": 0.19268061909184817, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017878132802678067, + "loss": 0.1621, + "step": 22197 + }, + { + "epoch": 0.19268929957205233, + "grad_norm": 0.09375, + "learning_rate": 0.001787794085025841, + "loss": 0.0698, + "step": 22198 + }, + { + "epoch": 0.1926979800522565, + "grad_norm": 0.62109375, + "learning_rate": 0.0017877748890317141, + "loss": 0.1099, + "step": 22199 + }, + { + "epoch": 0.19270666053246066, + "grad_norm": 0.08544921875, + "learning_rate": 0.0017877556922854464, + "loss": 0.1211, + "step": 22200 + }, + { + "epoch": 0.19271534101266483, + "grad_norm": 0.1826171875, + "learning_rate": 0.0017877364947870601, + "loss": 0.1211, + "step": 22201 + }, + { + "epoch": 0.192724021492869, + "grad_norm": 0.193359375, + "learning_rate": 0.001787717296536575, + "loss": 0.0996, + "step": 22202 + }, + { + "epoch": 0.19273270197307316, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017876980975340129, + "loss": 0.1089, + "step": 22203 + }, + { + "epoch": 0.19274138245327732, + "grad_norm": 0.345703125, + "learning_rate": 0.0017876788977793945, + "loss": 0.1084, + "step": 22204 + }, + { + "epoch": 0.1927500629334815, + "grad_norm": 0.59765625, + "learning_rate": 0.0017876596972727407, + "loss": 0.1494, + "step": 22205 + }, + { + "epoch": 0.19275874341368565, + "grad_norm": 0.1328125, + "learning_rate": 0.0017876404960140725, + "loss": 0.1187, + "step": 22206 + }, + { + "epoch": 0.19276742389388982, + "grad_norm": 0.115234375, + "learning_rate": 0.0017876212940034111, + "loss": 0.124, + "step": 22207 + }, + { + "epoch": 0.19277610437409398, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017876020912407774, + "loss": 0.1021, + "step": 22208 + }, + { + "epoch": 0.19278478485429815, + "grad_norm": 0.28515625, + "learning_rate": 0.0017875828877261923, + "loss": 0.1162, + "step": 22209 + }, + { + "epoch": 0.1927934653345023, + "grad_norm": 0.1162109375, + "learning_rate": 0.001787563683459677, + "loss": 0.0957, + "step": 22210 + }, + { + "epoch": 0.19280214581470648, + "grad_norm": 0.435546875, + "learning_rate": 0.0017875444784412522, + "loss": 0.166, + "step": 22211 + }, + { + "epoch": 0.19281082629491064, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017875252726709393, + "loss": 0.1025, + "step": 22212 + }, + { + "epoch": 0.1928195067751148, + "grad_norm": 0.255859375, + "learning_rate": 0.001787506066148759, + "loss": 0.1216, + "step": 22213 + }, + { + "epoch": 0.19282818725531897, + "grad_norm": 0.18359375, + "learning_rate": 0.0017874868588747324, + "loss": 0.1016, + "step": 22214 + }, + { + "epoch": 0.19283686773552314, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017874676508488803, + "loss": 0.1182, + "step": 22215 + }, + { + "epoch": 0.1928455482157273, + "grad_norm": 0.22265625, + "learning_rate": 0.001787448442071224, + "loss": 0.1162, + "step": 22216 + }, + { + "epoch": 0.19285422869593147, + "grad_norm": 0.431640625, + "learning_rate": 0.0017874292325417848, + "loss": 0.127, + "step": 22217 + }, + { + "epoch": 0.19286290917613563, + "grad_norm": 0.35546875, + "learning_rate": 0.0017874100222605826, + "loss": 0.127, + "step": 22218 + }, + { + "epoch": 0.1928715896563398, + "grad_norm": 0.1845703125, + "learning_rate": 0.00178739081122764, + "loss": 0.0986, + "step": 22219 + }, + { + "epoch": 0.19288027013654396, + "grad_norm": 0.3828125, + "learning_rate": 0.0017873715994429764, + "loss": 0.166, + "step": 22220 + }, + { + "epoch": 0.19288895061674813, + "grad_norm": 0.146484375, + "learning_rate": 0.001787352386906614, + "loss": 0.1069, + "step": 22221 + }, + { + "epoch": 0.1928976310969523, + "grad_norm": 0.3203125, + "learning_rate": 0.0017873331736185732, + "loss": 0.0767, + "step": 22222 + }, + { + "epoch": 0.19290631157715646, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017873139595788751, + "loss": 0.0688, + "step": 22223 + }, + { + "epoch": 0.19291499205736062, + "grad_norm": 0.86328125, + "learning_rate": 0.001787294744787541, + "loss": 0.0869, + "step": 22224 + }, + { + "epoch": 0.1929236725375648, + "grad_norm": 0.51953125, + "learning_rate": 0.0017872755292445916, + "loss": 0.1367, + "step": 22225 + }, + { + "epoch": 0.19293235301776895, + "grad_norm": 0.234375, + "learning_rate": 0.001787256312950048, + "loss": 0.1055, + "step": 22226 + }, + { + "epoch": 0.19294103349797312, + "grad_norm": 0.498046875, + "learning_rate": 0.0017872370959039307, + "loss": 0.1826, + "step": 22227 + }, + { + "epoch": 0.19294971397817728, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017872178781062618, + "loss": 0.1328, + "step": 22228 + }, + { + "epoch": 0.19295839445838145, + "grad_norm": 0.6328125, + "learning_rate": 0.0017871986595570618, + "loss": 0.1338, + "step": 22229 + }, + { + "epoch": 0.19296707493858561, + "grad_norm": 0.12109375, + "learning_rate": 0.0017871794402563514, + "loss": 0.0957, + "step": 22230 + }, + { + "epoch": 0.19297575541878978, + "grad_norm": 0.96484375, + "learning_rate": 0.001787160220204152, + "loss": 0.1084, + "step": 22231 + }, + { + "epoch": 0.19298443589899394, + "grad_norm": 0.5, + "learning_rate": 0.0017871409994004847, + "loss": 0.1377, + "step": 22232 + }, + { + "epoch": 0.1929931163791981, + "grad_norm": 0.68359375, + "learning_rate": 0.0017871217778453702, + "loss": 0.1289, + "step": 22233 + }, + { + "epoch": 0.19300179685940227, + "grad_norm": 0.1328125, + "learning_rate": 0.0017871025555388296, + "loss": 0.1025, + "step": 22234 + }, + { + "epoch": 0.19301047733960644, + "grad_norm": 0.26171875, + "learning_rate": 0.0017870833324808838, + "loss": 0.1123, + "step": 22235 + }, + { + "epoch": 0.1930191578198106, + "grad_norm": 0.671875, + "learning_rate": 0.001787064108671554, + "loss": 0.124, + "step": 22236 + }, + { + "epoch": 0.19302783830001477, + "grad_norm": 0.4609375, + "learning_rate": 0.0017870448841108616, + "loss": 0.0996, + "step": 22237 + }, + { + "epoch": 0.19303651878021894, + "grad_norm": 0.291015625, + "learning_rate": 0.001787025658798827, + "loss": 0.1328, + "step": 22238 + }, + { + "epoch": 0.1930451992604231, + "grad_norm": 0.4375, + "learning_rate": 0.0017870064327354714, + "loss": 0.0889, + "step": 22239 + }, + { + "epoch": 0.19305387974062724, + "grad_norm": 0.306640625, + "learning_rate": 0.0017869872059208159, + "loss": 0.1035, + "step": 22240 + }, + { + "epoch": 0.1930625602208314, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017869679783548815, + "loss": 0.1201, + "step": 22241 + }, + { + "epoch": 0.19307124070103557, + "grad_norm": 0.9921875, + "learning_rate": 0.0017869487500376895, + "loss": 0.1182, + "step": 22242 + }, + { + "epoch": 0.19307992118123973, + "grad_norm": 0.90234375, + "learning_rate": 0.0017869295209692604, + "loss": 0.1309, + "step": 22243 + }, + { + "epoch": 0.1930886016614439, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017869102911496155, + "loss": 0.1006, + "step": 22244 + }, + { + "epoch": 0.19309728214164806, + "grad_norm": 0.1357421875, + "learning_rate": 0.001786891060578776, + "loss": 0.1162, + "step": 22245 + }, + { + "epoch": 0.19310596262185223, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017868718292567626, + "loss": 0.1387, + "step": 22246 + }, + { + "epoch": 0.1931146431020564, + "grad_norm": 0.373046875, + "learning_rate": 0.0017868525971835966, + "loss": 0.1138, + "step": 22247 + }, + { + "epoch": 0.19312332358226056, + "grad_norm": 0.2001953125, + "learning_rate": 0.0017868333643592986, + "loss": 0.0996, + "step": 22248 + }, + { + "epoch": 0.19313200406246472, + "grad_norm": 0.30859375, + "learning_rate": 0.0017868141307838903, + "loss": 0.1055, + "step": 22249 + }, + { + "epoch": 0.1931406845426689, + "grad_norm": 0.400390625, + "learning_rate": 0.0017867948964573927, + "loss": 0.0728, + "step": 22250 + }, + { + "epoch": 0.19314936502287305, + "grad_norm": 0.2255859375, + "learning_rate": 0.001786775661379826, + "loss": 0.1045, + "step": 22251 + }, + { + "epoch": 0.19315804550307722, + "grad_norm": 0.07763671875, + "learning_rate": 0.001786756425551212, + "loss": 0.0884, + "step": 22252 + }, + { + "epoch": 0.19316672598328138, + "grad_norm": 0.4140625, + "learning_rate": 0.0017867371889715709, + "loss": 0.0869, + "step": 22253 + }, + { + "epoch": 0.19317540646348555, + "grad_norm": 1.0234375, + "learning_rate": 0.0017867179516409249, + "loss": 0.1992, + "step": 22254 + }, + { + "epoch": 0.19318408694368971, + "grad_norm": 0.796875, + "learning_rate": 0.0017866987135592945, + "loss": 0.1104, + "step": 22255 + }, + { + "epoch": 0.19319276742389388, + "grad_norm": 0.197265625, + "learning_rate": 0.0017866794747267007, + "loss": 0.1309, + "step": 22256 + }, + { + "epoch": 0.19320144790409804, + "grad_norm": 0.255859375, + "learning_rate": 0.0017866602351431643, + "loss": 0.1074, + "step": 22257 + }, + { + "epoch": 0.1932101283843022, + "grad_norm": 0.171875, + "learning_rate": 0.0017866409948087065, + "loss": 0.127, + "step": 22258 + }, + { + "epoch": 0.19321880886450638, + "grad_norm": 0.5078125, + "learning_rate": 0.0017866217537233487, + "loss": 0.1025, + "step": 22259 + }, + { + "epoch": 0.19322748934471054, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017866025118871117, + "loss": 0.1475, + "step": 22260 + }, + { + "epoch": 0.1932361698249147, + "grad_norm": 0.4140625, + "learning_rate": 0.0017865832693000163, + "loss": 0.0947, + "step": 22261 + }, + { + "epoch": 0.19324485030511887, + "grad_norm": 0.50390625, + "learning_rate": 0.0017865640259620839, + "loss": 0.123, + "step": 22262 + }, + { + "epoch": 0.19325353078532304, + "grad_norm": 0.1640625, + "learning_rate": 0.0017865447818733353, + "loss": 0.0928, + "step": 22263 + }, + { + "epoch": 0.1932622112655272, + "grad_norm": 0.333984375, + "learning_rate": 0.0017865255370337917, + "loss": 0.1099, + "step": 22264 + }, + { + "epoch": 0.19327089174573137, + "grad_norm": 0.369140625, + "learning_rate": 0.0017865062914434742, + "loss": 0.1094, + "step": 22265 + }, + { + "epoch": 0.19327957222593553, + "grad_norm": 0.51953125, + "learning_rate": 0.0017864870451024036, + "loss": 0.1016, + "step": 22266 + }, + { + "epoch": 0.1932882527061397, + "grad_norm": 0.8828125, + "learning_rate": 0.001786467798010601, + "loss": 0.1592, + "step": 22267 + }, + { + "epoch": 0.19329693318634386, + "grad_norm": 0.75, + "learning_rate": 0.001786448550168088, + "loss": 0.1406, + "step": 22268 + }, + { + "epoch": 0.19330561366654803, + "grad_norm": 0.1611328125, + "learning_rate": 0.0017864293015748844, + "loss": 0.1182, + "step": 22269 + }, + { + "epoch": 0.1933142941467522, + "grad_norm": 0.142578125, + "learning_rate": 0.0017864100522310125, + "loss": 0.1182, + "step": 22270 + }, + { + "epoch": 0.19332297462695636, + "grad_norm": 0.07958984375, + "learning_rate": 0.0017863908021364928, + "loss": 0.0923, + "step": 22271 + }, + { + "epoch": 0.19333165510716052, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017863715512913463, + "loss": 0.0938, + "step": 22272 + }, + { + "epoch": 0.1933403355873647, + "grad_norm": 0.384765625, + "learning_rate": 0.0017863522996955942, + "loss": 0.126, + "step": 22273 + }, + { + "epoch": 0.19334901606756885, + "grad_norm": 0.275390625, + "learning_rate": 0.001786333047349258, + "loss": 0.126, + "step": 22274 + }, + { + "epoch": 0.19335769654777302, + "grad_norm": 0.64453125, + "learning_rate": 0.0017863137942523582, + "loss": 0.1035, + "step": 22275 + }, + { + "epoch": 0.19336637702797718, + "grad_norm": 0.146484375, + "learning_rate": 0.0017862945404049154, + "loss": 0.1001, + "step": 22276 + }, + { + "epoch": 0.19337505750818135, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017862752858069518, + "loss": 0.1133, + "step": 22277 + }, + { + "epoch": 0.1933837379883855, + "grad_norm": 0.150390625, + "learning_rate": 0.0017862560304584875, + "loss": 0.0938, + "step": 22278 + }, + { + "epoch": 0.19339241846858968, + "grad_norm": 0.408203125, + "learning_rate": 0.0017862367743595438, + "loss": 0.1152, + "step": 22279 + }, + { + "epoch": 0.19340109894879384, + "grad_norm": 0.1513671875, + "learning_rate": 0.001786217517510142, + "loss": 0.0791, + "step": 22280 + }, + { + "epoch": 0.193409779428998, + "grad_norm": 1.0625, + "learning_rate": 0.001786198259910303, + "loss": 0.105, + "step": 22281 + }, + { + "epoch": 0.19341845990920217, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017861790015600482, + "loss": 0.1187, + "step": 22282 + }, + { + "epoch": 0.19342714038940634, + "grad_norm": 0.16796875, + "learning_rate": 0.001786159742459398, + "loss": 0.0801, + "step": 22283 + }, + { + "epoch": 0.1934358208696105, + "grad_norm": 0.259765625, + "learning_rate": 0.0017861404826083741, + "loss": 0.1201, + "step": 22284 + }, + { + "epoch": 0.19344450134981467, + "grad_norm": 0.205078125, + "learning_rate": 0.0017861212220069969, + "loss": 0.1445, + "step": 22285 + }, + { + "epoch": 0.19345318183001883, + "grad_norm": 0.287109375, + "learning_rate": 0.0017861019606552882, + "loss": 0.0786, + "step": 22286 + }, + { + "epoch": 0.193461862310223, + "grad_norm": 0.099609375, + "learning_rate": 0.0017860826985532686, + "loss": 0.1055, + "step": 22287 + }, + { + "epoch": 0.19347054279042716, + "grad_norm": 0.08984375, + "learning_rate": 0.001786063435700959, + "loss": 0.1445, + "step": 22288 + }, + { + "epoch": 0.19347922327063133, + "grad_norm": 0.34375, + "learning_rate": 0.0017860441720983812, + "loss": 0.1309, + "step": 22289 + }, + { + "epoch": 0.1934879037508355, + "grad_norm": 0.384765625, + "learning_rate": 0.0017860249077455556, + "loss": 0.0928, + "step": 22290 + }, + { + "epoch": 0.19349658423103966, + "grad_norm": 0.169921875, + "learning_rate": 0.0017860056426425032, + "loss": 0.0957, + "step": 22291 + }, + { + "epoch": 0.19350526471124382, + "grad_norm": 0.373046875, + "learning_rate": 0.001785986376789246, + "loss": 0.1216, + "step": 22292 + }, + { + "epoch": 0.193513945191448, + "grad_norm": 0.26953125, + "learning_rate": 0.0017859671101858037, + "loss": 0.1318, + "step": 22293 + }, + { + "epoch": 0.19352262567165215, + "grad_norm": 0.130859375, + "learning_rate": 0.0017859478428321985, + "loss": 0.1084, + "step": 22294 + }, + { + "epoch": 0.19353130615185632, + "grad_norm": 0.84375, + "learning_rate": 0.0017859285747284508, + "loss": 0.1299, + "step": 22295 + }, + { + "epoch": 0.19353998663206048, + "grad_norm": 0.66796875, + "learning_rate": 0.0017859093058745822, + "loss": 0.1079, + "step": 22296 + }, + { + "epoch": 0.19354866711226465, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017858900362706133, + "loss": 0.1309, + "step": 22297 + }, + { + "epoch": 0.19355734759246881, + "grad_norm": 0.15234375, + "learning_rate": 0.0017858707659165651, + "loss": 0.0879, + "step": 22298 + }, + { + "epoch": 0.19356602807267298, + "grad_norm": 0.4375, + "learning_rate": 0.0017858514948124594, + "loss": 0.0952, + "step": 22299 + }, + { + "epoch": 0.19357470855287714, + "grad_norm": 0.185546875, + "learning_rate": 0.0017858322229583163, + "loss": 0.1777, + "step": 22300 + }, + { + "epoch": 0.1935833890330813, + "grad_norm": 0.84375, + "learning_rate": 0.001785812950354158, + "loss": 0.1348, + "step": 22301 + }, + { + "epoch": 0.19359206951328548, + "grad_norm": 0.9765625, + "learning_rate": 0.0017857936770000046, + "loss": 0.1611, + "step": 22302 + }, + { + "epoch": 0.19360074999348964, + "grad_norm": 0.453125, + "learning_rate": 0.0017857744028958779, + "loss": 0.0991, + "step": 22303 + }, + { + "epoch": 0.1936094304736938, + "grad_norm": 0.279296875, + "learning_rate": 0.001785755128041798, + "loss": 0.0884, + "step": 22304 + }, + { + "epoch": 0.19361811095389797, + "grad_norm": 0.5625, + "learning_rate": 0.001785735852437787, + "loss": 0.1162, + "step": 22305 + }, + { + "epoch": 0.19362679143410214, + "grad_norm": 0.25, + "learning_rate": 0.0017857165760838654, + "loss": 0.1079, + "step": 22306 + }, + { + "epoch": 0.1936354719143063, + "grad_norm": 0.5078125, + "learning_rate": 0.0017856972989800544, + "loss": 0.1084, + "step": 22307 + }, + { + "epoch": 0.19364415239451047, + "grad_norm": 0.3984375, + "learning_rate": 0.0017856780211263754, + "loss": 0.0859, + "step": 22308 + }, + { + "epoch": 0.19365283287471463, + "grad_norm": 0.416015625, + "learning_rate": 0.0017856587425228493, + "loss": 0.0947, + "step": 22309 + }, + { + "epoch": 0.1936615133549188, + "grad_norm": 0.490234375, + "learning_rate": 0.001785639463169497, + "loss": 0.1455, + "step": 22310 + }, + { + "epoch": 0.19367019383512296, + "grad_norm": 0.8125, + "learning_rate": 0.0017856201830663393, + "loss": 0.1455, + "step": 22311 + }, + { + "epoch": 0.19367887431532713, + "grad_norm": 0.10205078125, + "learning_rate": 0.0017856009022133982, + "loss": 0.0996, + "step": 22312 + }, + { + "epoch": 0.1936875547955313, + "grad_norm": 0.134765625, + "learning_rate": 0.0017855816206106938, + "loss": 0.0947, + "step": 22313 + }, + { + "epoch": 0.19369623527573546, + "grad_norm": 0.57421875, + "learning_rate": 0.0017855623382582479, + "loss": 0.1182, + "step": 22314 + }, + { + "epoch": 0.19370491575593962, + "grad_norm": 0.419921875, + "learning_rate": 0.0017855430551560812, + "loss": 0.1484, + "step": 22315 + }, + { + "epoch": 0.1937135962361438, + "grad_norm": 0.443359375, + "learning_rate": 0.0017855237713042152, + "loss": 0.0947, + "step": 22316 + }, + { + "epoch": 0.19372227671634795, + "grad_norm": 0.40625, + "learning_rate": 0.0017855044867026704, + "loss": 0.0903, + "step": 22317 + }, + { + "epoch": 0.19373095719655212, + "grad_norm": 0.1875, + "learning_rate": 0.0017854852013514681, + "loss": 0.1455, + "step": 22318 + }, + { + "epoch": 0.19373963767675628, + "grad_norm": 0.142578125, + "learning_rate": 0.0017854659152506298, + "loss": 0.0879, + "step": 22319 + }, + { + "epoch": 0.19374831815696045, + "grad_norm": 0.271484375, + "learning_rate": 0.0017854466284001763, + "loss": 0.0942, + "step": 22320 + }, + { + "epoch": 0.1937569986371646, + "grad_norm": 1.046875, + "learning_rate": 0.0017854273408001284, + "loss": 0.1152, + "step": 22321 + }, + { + "epoch": 0.19376567911736878, + "grad_norm": 0.52734375, + "learning_rate": 0.0017854080524505077, + "loss": 0.1128, + "step": 22322 + }, + { + "epoch": 0.19377435959757294, + "grad_norm": 0.2109375, + "learning_rate": 0.0017853887633513348, + "loss": 0.1191, + "step": 22323 + }, + { + "epoch": 0.1937830400777771, + "grad_norm": 0.3671875, + "learning_rate": 0.0017853694735026312, + "loss": 0.1338, + "step": 22324 + }, + { + "epoch": 0.19379172055798127, + "grad_norm": 0.1630859375, + "learning_rate": 0.001785350182904418, + "loss": 0.1445, + "step": 22325 + }, + { + "epoch": 0.19380040103818544, + "grad_norm": 0.412109375, + "learning_rate": 0.001785330891556716, + "loss": 0.1118, + "step": 22326 + }, + { + "epoch": 0.1938090815183896, + "grad_norm": 0.0966796875, + "learning_rate": 0.0017853115994595462, + "loss": 0.1016, + "step": 22327 + }, + { + "epoch": 0.19381776199859377, + "grad_norm": 0.255859375, + "learning_rate": 0.00178529230661293, + "loss": 0.1523, + "step": 22328 + }, + { + "epoch": 0.19382644247879793, + "grad_norm": 0.83203125, + "learning_rate": 0.0017852730130168885, + "loss": 0.0791, + "step": 22329 + }, + { + "epoch": 0.1938351229590021, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017852537186714428, + "loss": 0.0942, + "step": 22330 + }, + { + "epoch": 0.19384380343920626, + "grad_norm": 1.0625, + "learning_rate": 0.001785234423576614, + "loss": 0.1367, + "step": 22331 + }, + { + "epoch": 0.19385248391941043, + "grad_norm": 0.69921875, + "learning_rate": 0.001785215127732423, + "loss": 0.1084, + "step": 22332 + }, + { + "epoch": 0.1938611643996146, + "grad_norm": 0.57421875, + "learning_rate": 0.0017851958311388908, + "loss": 0.1523, + "step": 22333 + }, + { + "epoch": 0.19386984487981876, + "grad_norm": 0.1826171875, + "learning_rate": 0.0017851765337960394, + "loss": 0.1099, + "step": 22334 + }, + { + "epoch": 0.19387852536002292, + "grad_norm": 0.291015625, + "learning_rate": 0.0017851572357038886, + "loss": 0.0986, + "step": 22335 + }, + { + "epoch": 0.1938872058402271, + "grad_norm": 0.578125, + "learning_rate": 0.0017851379368624604, + "loss": 0.1113, + "step": 22336 + }, + { + "epoch": 0.19389588632043125, + "grad_norm": 0.1201171875, + "learning_rate": 0.0017851186372717757, + "loss": 0.1055, + "step": 22337 + }, + { + "epoch": 0.19390456680063542, + "grad_norm": 0.3125, + "learning_rate": 0.0017850993369318555, + "loss": 0.0981, + "step": 22338 + }, + { + "epoch": 0.19391324728083958, + "grad_norm": 0.14453125, + "learning_rate": 0.0017850800358427208, + "loss": 0.127, + "step": 22339 + }, + { + "epoch": 0.19392192776104375, + "grad_norm": 1.6328125, + "learning_rate": 0.0017850607340043933, + "loss": 0.1084, + "step": 22340 + }, + { + "epoch": 0.19393060824124791, + "grad_norm": 0.55859375, + "learning_rate": 0.0017850414314168933, + "loss": 0.1138, + "step": 22341 + }, + { + "epoch": 0.19393928872145208, + "grad_norm": 0.236328125, + "learning_rate": 0.0017850221280802424, + "loss": 0.0771, + "step": 22342 + }, + { + "epoch": 0.19394796920165625, + "grad_norm": 0.14453125, + "learning_rate": 0.0017850028239944616, + "loss": 0.1367, + "step": 22343 + }, + { + "epoch": 0.1939566496818604, + "grad_norm": 0.447265625, + "learning_rate": 0.0017849835191595719, + "loss": 0.0918, + "step": 22344 + }, + { + "epoch": 0.19396533016206458, + "grad_norm": 0.58203125, + "learning_rate": 0.0017849642135755945, + "loss": 0.1279, + "step": 22345 + }, + { + "epoch": 0.19397401064226874, + "grad_norm": 0.28125, + "learning_rate": 0.0017849449072425507, + "loss": 0.1836, + "step": 22346 + }, + { + "epoch": 0.1939826911224729, + "grad_norm": 0.71875, + "learning_rate": 0.0017849256001604612, + "loss": 0.1182, + "step": 22347 + }, + { + "epoch": 0.19399137160267707, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017849062923293475, + "loss": 0.1104, + "step": 22348 + }, + { + "epoch": 0.19400005208288124, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017848869837492306, + "loss": 0.0801, + "step": 22349 + }, + { + "epoch": 0.1940087325630854, + "grad_norm": 0.298828125, + "learning_rate": 0.0017848676744201316, + "loss": 0.126, + "step": 22350 + }, + { + "epoch": 0.19401741304328957, + "grad_norm": 1.0078125, + "learning_rate": 0.0017848483643420714, + "loss": 0.0742, + "step": 22351 + }, + { + "epoch": 0.19402609352349373, + "grad_norm": 0.19140625, + "learning_rate": 0.0017848290535150716, + "loss": 0.1367, + "step": 22352 + }, + { + "epoch": 0.1940347740036979, + "grad_norm": 0.37890625, + "learning_rate": 0.001784809741939153, + "loss": 0.0879, + "step": 22353 + }, + { + "epoch": 0.19404345448390206, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017847904296143362, + "loss": 0.0933, + "step": 22354 + }, + { + "epoch": 0.19405213496410623, + "grad_norm": 0.1796875, + "learning_rate": 0.0017847711165406438, + "loss": 0.1084, + "step": 22355 + }, + { + "epoch": 0.1940608154443104, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017847518027180953, + "loss": 0.1709, + "step": 22356 + }, + { + "epoch": 0.19406949592451456, + "grad_norm": 0.8046875, + "learning_rate": 0.0017847324881467128, + "loss": 0.1104, + "step": 22357 + }, + { + "epoch": 0.19407817640471872, + "grad_norm": 0.283203125, + "learning_rate": 0.001784713172826517, + "loss": 0.1025, + "step": 22358 + }, + { + "epoch": 0.1940868568849229, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017846938567575292, + "loss": 0.1172, + "step": 22359 + }, + { + "epoch": 0.19409553736512705, + "grad_norm": 0.220703125, + "learning_rate": 0.0017846745399397708, + "loss": 0.0879, + "step": 22360 + }, + { + "epoch": 0.19410421784533122, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017846552223732624, + "loss": 0.1436, + "step": 22361 + }, + { + "epoch": 0.19411289832553538, + "grad_norm": 0.1220703125, + "learning_rate": 0.0017846359040580254, + "loss": 0.0718, + "step": 22362 + }, + { + "epoch": 0.19412157880573952, + "grad_norm": 0.5703125, + "learning_rate": 0.0017846165849940805, + "loss": 0.1621, + "step": 22363 + }, + { + "epoch": 0.19413025928594368, + "grad_norm": 0.314453125, + "learning_rate": 0.0017845972651814492, + "loss": 0.1221, + "step": 22364 + }, + { + "epoch": 0.19413893976614785, + "grad_norm": 0.15234375, + "learning_rate": 0.0017845779446201528, + "loss": 0.1221, + "step": 22365 + }, + { + "epoch": 0.19414762024635202, + "grad_norm": 0.55859375, + "learning_rate": 0.0017845586233102124, + "loss": 0.127, + "step": 22366 + }, + { + "epoch": 0.19415630072655618, + "grad_norm": 0.80859375, + "learning_rate": 0.0017845393012516492, + "loss": 0.1152, + "step": 22367 + }, + { + "epoch": 0.19416498120676035, + "grad_norm": 0.091796875, + "learning_rate": 0.0017845199784444835, + "loss": 0.1396, + "step": 22368 + }, + { + "epoch": 0.1941736616869645, + "grad_norm": 0.146484375, + "learning_rate": 0.0017845006548887373, + "loss": 0.1318, + "step": 22369 + }, + { + "epoch": 0.19418234216716868, + "grad_norm": 0.52734375, + "learning_rate": 0.0017844813305844317, + "loss": 0.0859, + "step": 22370 + }, + { + "epoch": 0.19419102264737284, + "grad_norm": 0.54296875, + "learning_rate": 0.0017844620055315872, + "loss": 0.125, + "step": 22371 + }, + { + "epoch": 0.194199703127577, + "grad_norm": 0.3125, + "learning_rate": 0.0017844426797302255, + "loss": 0.0728, + "step": 22372 + }, + { + "epoch": 0.19420838360778117, + "grad_norm": 1.0859375, + "learning_rate": 0.0017844233531803676, + "loss": 0.1387, + "step": 22373 + }, + { + "epoch": 0.19421706408798534, + "grad_norm": 0.3125, + "learning_rate": 0.0017844040258820346, + "loss": 0.1143, + "step": 22374 + }, + { + "epoch": 0.1942257445681895, + "grad_norm": 0.4296875, + "learning_rate": 0.0017843846978352479, + "loss": 0.1494, + "step": 22375 + }, + { + "epoch": 0.19423442504839367, + "grad_norm": 0.1640625, + "learning_rate": 0.001784365369040028, + "loss": 0.1201, + "step": 22376 + }, + { + "epoch": 0.19424310552859783, + "grad_norm": 0.279296875, + "learning_rate": 0.0017843460394963964, + "loss": 0.1035, + "step": 22377 + }, + { + "epoch": 0.194251786008802, + "grad_norm": 0.1142578125, + "learning_rate": 0.0017843267092043744, + "loss": 0.1211, + "step": 22378 + }, + { + "epoch": 0.19426046648900616, + "grad_norm": 0.24609375, + "learning_rate": 0.001784307378163983, + "loss": 0.1133, + "step": 22379 + }, + { + "epoch": 0.19426914696921033, + "grad_norm": 0.65625, + "learning_rate": 0.0017842880463752432, + "loss": 0.1445, + "step": 22380 + }, + { + "epoch": 0.1942778274494145, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017842687138381766, + "loss": 0.1504, + "step": 22381 + }, + { + "epoch": 0.19428650792961866, + "grad_norm": 0.359375, + "learning_rate": 0.0017842493805528036, + "loss": 0.127, + "step": 22382 + }, + { + "epoch": 0.19429518840982282, + "grad_norm": 0.439453125, + "learning_rate": 0.0017842300465191459, + "loss": 0.1406, + "step": 22383 + }, + { + "epoch": 0.194303868890027, + "grad_norm": 0.08984375, + "learning_rate": 0.0017842107117372245, + "loss": 0.125, + "step": 22384 + }, + { + "epoch": 0.19431254937023115, + "grad_norm": 0.10205078125, + "learning_rate": 0.0017841913762070606, + "loss": 0.1299, + "step": 22385 + }, + { + "epoch": 0.19432122985043532, + "grad_norm": 0.19140625, + "learning_rate": 0.0017841720399286755, + "loss": 0.0957, + "step": 22386 + }, + { + "epoch": 0.19432991033063948, + "grad_norm": 0.0771484375, + "learning_rate": 0.0017841527029020898, + "loss": 0.1094, + "step": 22387 + }, + { + "epoch": 0.19433859081084365, + "grad_norm": 0.059814453125, + "learning_rate": 0.001784133365127325, + "loss": 0.084, + "step": 22388 + }, + { + "epoch": 0.1943472712910478, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017841140266044025, + "loss": 0.0708, + "step": 22389 + }, + { + "epoch": 0.19435595177125198, + "grad_norm": 0.46484375, + "learning_rate": 0.001784094687333343, + "loss": 0.1182, + "step": 22390 + }, + { + "epoch": 0.19436463225145614, + "grad_norm": 0.8515625, + "learning_rate": 0.0017840753473141679, + "loss": 0.1387, + "step": 22391 + }, + { + "epoch": 0.1943733127316603, + "grad_norm": 0.76171875, + "learning_rate": 0.001784056006546898, + "loss": 0.1367, + "step": 22392 + }, + { + "epoch": 0.19438199321186447, + "grad_norm": 0.42578125, + "learning_rate": 0.0017840366650315551, + "loss": 0.105, + "step": 22393 + }, + { + "epoch": 0.19439067369206864, + "grad_norm": 0.328125, + "learning_rate": 0.0017840173227681598, + "loss": 0.1006, + "step": 22394 + }, + { + "epoch": 0.1943993541722728, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017839979797567337, + "loss": 0.1001, + "step": 22395 + }, + { + "epoch": 0.19440803465247697, + "grad_norm": 0.1162109375, + "learning_rate": 0.0017839786359972974, + "loss": 0.1011, + "step": 22396 + }, + { + "epoch": 0.19441671513268113, + "grad_norm": 0.28125, + "learning_rate": 0.001783959291489872, + "loss": 0.1206, + "step": 22397 + }, + { + "epoch": 0.1944253956128853, + "grad_norm": 0.33203125, + "learning_rate": 0.0017839399462344797, + "loss": 0.0835, + "step": 22398 + }, + { + "epoch": 0.19443407609308946, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017839206002311407, + "loss": 0.1436, + "step": 22399 + }, + { + "epoch": 0.19444275657329363, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017839012534798763, + "loss": 0.1006, + "step": 22400 + }, + { + "epoch": 0.1944514370534978, + "grad_norm": 0.1435546875, + "learning_rate": 0.001783881905980708, + "loss": 0.0967, + "step": 22401 + }, + { + "epoch": 0.19446011753370196, + "grad_norm": 0.30859375, + "learning_rate": 0.0017838625577336563, + "loss": 0.0938, + "step": 22402 + }, + { + "epoch": 0.19446879801390612, + "grad_norm": 0.228515625, + "learning_rate": 0.001783843208738743, + "loss": 0.1631, + "step": 22403 + }, + { + "epoch": 0.1944774784941103, + "grad_norm": 0.12158203125, + "learning_rate": 0.001783823858995989, + "loss": 0.1123, + "step": 22404 + }, + { + "epoch": 0.19448615897431445, + "grad_norm": 0.53125, + "learning_rate": 0.0017838045085054155, + "loss": 0.1191, + "step": 22405 + }, + { + "epoch": 0.19449483945451862, + "grad_norm": 0.294921875, + "learning_rate": 0.0017837851572670438, + "loss": 0.1523, + "step": 22406 + }, + { + "epoch": 0.19450351993472279, + "grad_norm": 0.208984375, + "learning_rate": 0.001783765805280895, + "loss": 0.1201, + "step": 22407 + }, + { + "epoch": 0.19451220041492695, + "grad_norm": 0.439453125, + "learning_rate": 0.0017837464525469898, + "loss": 0.0718, + "step": 22408 + }, + { + "epoch": 0.19452088089513112, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017837270990653502, + "loss": 0.1221, + "step": 22409 + }, + { + "epoch": 0.19452956137533528, + "grad_norm": 0.07177734375, + "learning_rate": 0.0017837077448359965, + "loss": 0.0854, + "step": 22410 + }, + { + "epoch": 0.19453824185553945, + "grad_norm": 0.1318359375, + "learning_rate": 0.0017836883898589505, + "loss": 0.1069, + "step": 22411 + }, + { + "epoch": 0.1945469223357436, + "grad_norm": 0.1474609375, + "learning_rate": 0.0017836690341342333, + "loss": 0.0903, + "step": 22412 + }, + { + "epoch": 0.19455560281594778, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017836496776618659, + "loss": 0.1357, + "step": 22413 + }, + { + "epoch": 0.19456428329615194, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017836303204418692, + "loss": 0.1426, + "step": 22414 + }, + { + "epoch": 0.1945729637763561, + "grad_norm": 1.015625, + "learning_rate": 0.0017836109624742647, + "loss": 0.1182, + "step": 22415 + }, + { + "epoch": 0.19458164425656027, + "grad_norm": 0.2578125, + "learning_rate": 0.0017835916037590735, + "loss": 0.1328, + "step": 22416 + }, + { + "epoch": 0.19459032473676444, + "grad_norm": 0.263671875, + "learning_rate": 0.0017835722442963169, + "loss": 0.1016, + "step": 22417 + }, + { + "epoch": 0.1945990052169686, + "grad_norm": 0.1083984375, + "learning_rate": 0.001783552884086016, + "loss": 0.1089, + "step": 22418 + }, + { + "epoch": 0.19460768569717277, + "grad_norm": 0.2060546875, + "learning_rate": 0.001783533523128192, + "loss": 0.1147, + "step": 22419 + }, + { + "epoch": 0.19461636617737693, + "grad_norm": 0.419921875, + "learning_rate": 0.0017835141614228661, + "loss": 0.1416, + "step": 22420 + }, + { + "epoch": 0.1946250466575811, + "grad_norm": 0.21875, + "learning_rate": 0.001783494798970059, + "loss": 0.1152, + "step": 22421 + }, + { + "epoch": 0.19463372713778526, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017834754357697927, + "loss": 0.0908, + "step": 22422 + }, + { + "epoch": 0.19464240761798943, + "grad_norm": 0.396484375, + "learning_rate": 0.0017834560718220877, + "loss": 0.1074, + "step": 22423 + }, + { + "epoch": 0.1946510880981936, + "grad_norm": 0.177734375, + "learning_rate": 0.001783436707126965, + "loss": 0.0977, + "step": 22424 + }, + { + "epoch": 0.19465976857839776, + "grad_norm": 0.357421875, + "learning_rate": 0.0017834173416844468, + "loss": 0.1104, + "step": 22425 + }, + { + "epoch": 0.19466844905860192, + "grad_norm": 0.8203125, + "learning_rate": 0.0017833979754945536, + "loss": 0.0947, + "step": 22426 + }, + { + "epoch": 0.1946771295388061, + "grad_norm": 0.486328125, + "learning_rate": 0.0017833786085573063, + "loss": 0.126, + "step": 22427 + }, + { + "epoch": 0.19468581001901025, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017833592408727267, + "loss": 0.1523, + "step": 22428 + }, + { + "epoch": 0.19469449049921442, + "grad_norm": 0.9140625, + "learning_rate": 0.0017833398724408356, + "loss": 0.1011, + "step": 22429 + }, + { + "epoch": 0.19470317097941858, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017833205032616545, + "loss": 0.1328, + "step": 22430 + }, + { + "epoch": 0.19471185145962275, + "grad_norm": 0.59375, + "learning_rate": 0.001783301133335204, + "loss": 0.1182, + "step": 22431 + }, + { + "epoch": 0.1947205319398269, + "grad_norm": 0.1181640625, + "learning_rate": 0.0017832817626615058, + "loss": 0.1094, + "step": 22432 + }, + { + "epoch": 0.19472921242003108, + "grad_norm": 0.28125, + "learning_rate": 0.001783262391240581, + "loss": 0.1035, + "step": 22433 + }, + { + "epoch": 0.19473789290023524, + "grad_norm": 0.232421875, + "learning_rate": 0.0017832430190724505, + "loss": 0.0781, + "step": 22434 + }, + { + "epoch": 0.1947465733804394, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017832236461571358, + "loss": 0.1289, + "step": 22435 + }, + { + "epoch": 0.19475525386064357, + "grad_norm": 0.111328125, + "learning_rate": 0.0017832042724946582, + "loss": 0.1211, + "step": 22436 + }, + { + "epoch": 0.19476393434084774, + "grad_norm": 0.6796875, + "learning_rate": 0.0017831848980850386, + "loss": 0.1279, + "step": 22437 + }, + { + "epoch": 0.1947726148210519, + "grad_norm": 0.1484375, + "learning_rate": 0.001783165522928298, + "loss": 0.1016, + "step": 22438 + }, + { + "epoch": 0.19478129530125607, + "grad_norm": 0.81640625, + "learning_rate": 0.0017831461470244581, + "loss": 0.0957, + "step": 22439 + }, + { + "epoch": 0.19478997578146023, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017831267703735398, + "loss": 0.1191, + "step": 22440 + }, + { + "epoch": 0.1947986562616644, + "grad_norm": 0.263671875, + "learning_rate": 0.0017831073929755641, + "loss": 0.0957, + "step": 22441 + }, + { + "epoch": 0.19480733674186856, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017830880148305525, + "loss": 0.0908, + "step": 22442 + }, + { + "epoch": 0.19481601722207273, + "grad_norm": 0.28515625, + "learning_rate": 0.0017830686359385265, + "loss": 0.1016, + "step": 22443 + }, + { + "epoch": 0.1948246977022769, + "grad_norm": 0.1044921875, + "learning_rate": 0.0017830492562995067, + "loss": 0.0977, + "step": 22444 + }, + { + "epoch": 0.19483337818248106, + "grad_norm": 0.6171875, + "learning_rate": 0.0017830298759135142, + "loss": 0.1187, + "step": 22445 + }, + { + "epoch": 0.19484205866268522, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017830104947805707, + "loss": 0.1104, + "step": 22446 + }, + { + "epoch": 0.1948507391428894, + "grad_norm": 0.78125, + "learning_rate": 0.0017829911129006973, + "loss": 0.1089, + "step": 22447 + }, + { + "epoch": 0.19485941962309355, + "grad_norm": 0.154296875, + "learning_rate": 0.0017829717302739149, + "loss": 0.0957, + "step": 22448 + }, + { + "epoch": 0.19486810010329772, + "grad_norm": 0.1064453125, + "learning_rate": 0.001782952346900245, + "loss": 0.0942, + "step": 22449 + }, + { + "epoch": 0.19487678058350189, + "grad_norm": 0.478515625, + "learning_rate": 0.0017829329627797087, + "loss": 0.1113, + "step": 22450 + }, + { + "epoch": 0.19488546106370605, + "grad_norm": 0.69140625, + "learning_rate": 0.001782913577912327, + "loss": 0.0933, + "step": 22451 + }, + { + "epoch": 0.19489414154391022, + "grad_norm": 0.0654296875, + "learning_rate": 0.0017828941922981214, + "loss": 0.0684, + "step": 22452 + }, + { + "epoch": 0.19490282202411438, + "grad_norm": 0.14453125, + "learning_rate": 0.001782874805937113, + "loss": 0.1357, + "step": 22453 + }, + { + "epoch": 0.19491150250431855, + "grad_norm": 0.12060546875, + "learning_rate": 0.0017828554188293226, + "loss": 0.0977, + "step": 22454 + }, + { + "epoch": 0.1949201829845227, + "grad_norm": 0.52734375, + "learning_rate": 0.0017828360309747723, + "loss": 0.1396, + "step": 22455 + }, + { + "epoch": 0.19492886346472688, + "grad_norm": 0.75390625, + "learning_rate": 0.0017828166423734823, + "loss": 0.1152, + "step": 22456 + }, + { + "epoch": 0.19493754394493104, + "grad_norm": 0.333984375, + "learning_rate": 0.001782797253025475, + "loss": 0.1123, + "step": 22457 + }, + { + "epoch": 0.1949462244251352, + "grad_norm": 0.494140625, + "learning_rate": 0.0017827778629307702, + "loss": 0.1191, + "step": 22458 + }, + { + "epoch": 0.19495490490533937, + "grad_norm": 0.326171875, + "learning_rate": 0.00178275847208939, + "loss": 0.1299, + "step": 22459 + }, + { + "epoch": 0.19496358538554354, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017827390805013556, + "loss": 0.1299, + "step": 22460 + }, + { + "epoch": 0.1949722658657477, + "grad_norm": 0.455078125, + "learning_rate": 0.0017827196881666876, + "loss": 0.1104, + "step": 22461 + }, + { + "epoch": 0.19498094634595187, + "grad_norm": 0.4609375, + "learning_rate": 0.0017827002950854078, + "loss": 0.1309, + "step": 22462 + }, + { + "epoch": 0.19498962682615603, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017826809012575374, + "loss": 0.1279, + "step": 22463 + }, + { + "epoch": 0.1949983073063602, + "grad_norm": 0.50390625, + "learning_rate": 0.0017826615066830972, + "loss": 0.0913, + "step": 22464 + }, + { + "epoch": 0.19500698778656436, + "grad_norm": 0.203125, + "learning_rate": 0.0017826421113621086, + "loss": 0.0903, + "step": 22465 + }, + { + "epoch": 0.19501566826676853, + "grad_norm": 0.177734375, + "learning_rate": 0.001782622715294593, + "loss": 0.1328, + "step": 22466 + }, + { + "epoch": 0.1950243487469727, + "grad_norm": 0.27734375, + "learning_rate": 0.0017826033184805714, + "loss": 0.0889, + "step": 22467 + }, + { + "epoch": 0.19503302922717686, + "grad_norm": 0.1396484375, + "learning_rate": 0.0017825839209200652, + "loss": 0.1279, + "step": 22468 + }, + { + "epoch": 0.19504170970738102, + "grad_norm": 0.373046875, + "learning_rate": 0.0017825645226130953, + "loss": 0.103, + "step": 22469 + }, + { + "epoch": 0.1950503901875852, + "grad_norm": 0.3671875, + "learning_rate": 0.001782545123559683, + "loss": 0.0801, + "step": 22470 + }, + { + "epoch": 0.19505907066778935, + "grad_norm": 0.859375, + "learning_rate": 0.0017825257237598499, + "loss": 0.1147, + "step": 22471 + }, + { + "epoch": 0.19506775114799352, + "grad_norm": 0.6640625, + "learning_rate": 0.0017825063232136168, + "loss": 0.1025, + "step": 22472 + }, + { + "epoch": 0.19507643162819768, + "grad_norm": 0.11181640625, + "learning_rate": 0.0017824869219210046, + "loss": 0.1123, + "step": 22473 + }, + { + "epoch": 0.19508511210840185, + "grad_norm": 0.087890625, + "learning_rate": 0.0017824675198820357, + "loss": 0.0869, + "step": 22474 + }, + { + "epoch": 0.195093792588606, + "grad_norm": 0.490234375, + "learning_rate": 0.00178244811709673, + "loss": 0.1426, + "step": 22475 + }, + { + "epoch": 0.19510247306881018, + "grad_norm": 0.455078125, + "learning_rate": 0.0017824287135651095, + "loss": 0.1133, + "step": 22476 + }, + { + "epoch": 0.19511115354901434, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017824093092871955, + "loss": 0.0869, + "step": 22477 + }, + { + "epoch": 0.1951198340292185, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017823899042630085, + "loss": 0.1074, + "step": 22478 + }, + { + "epoch": 0.19512851450942267, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017823704984925703, + "loss": 0.103, + "step": 22479 + }, + { + "epoch": 0.19513719498962684, + "grad_norm": 0.154296875, + "learning_rate": 0.0017823510919759019, + "loss": 0.1143, + "step": 22480 + }, + { + "epoch": 0.195145875469831, + "grad_norm": 0.478515625, + "learning_rate": 0.0017823316847130248, + "loss": 0.0747, + "step": 22481 + }, + { + "epoch": 0.19515455595003517, + "grad_norm": 0.267578125, + "learning_rate": 0.0017823122767039597, + "loss": 0.1133, + "step": 22482 + }, + { + "epoch": 0.19516323643023933, + "grad_norm": 0.6328125, + "learning_rate": 0.0017822928679487285, + "loss": 0.1016, + "step": 22483 + }, + { + "epoch": 0.1951719169104435, + "grad_norm": 0.404296875, + "learning_rate": 0.001782273458447352, + "loss": 0.0757, + "step": 22484 + }, + { + "epoch": 0.19518059739064766, + "grad_norm": 0.671875, + "learning_rate": 0.0017822540481998513, + "loss": 0.127, + "step": 22485 + }, + { + "epoch": 0.1951892778708518, + "grad_norm": 0.099609375, + "learning_rate": 0.0017822346372062482, + "loss": 0.1104, + "step": 22486 + }, + { + "epoch": 0.19519795835105597, + "grad_norm": 0.3671875, + "learning_rate": 0.0017822152254665631, + "loss": 0.0669, + "step": 22487 + }, + { + "epoch": 0.19520663883126013, + "grad_norm": 0.39453125, + "learning_rate": 0.0017821958129808177, + "loss": 0.1562, + "step": 22488 + }, + { + "epoch": 0.1952153193114643, + "grad_norm": 0.08740234375, + "learning_rate": 0.0017821763997490335, + "loss": 0.083, + "step": 22489 + }, + { + "epoch": 0.19522399979166846, + "grad_norm": 0.10546875, + "learning_rate": 0.0017821569857712314, + "loss": 0.1377, + "step": 22490 + }, + { + "epoch": 0.19523268027187263, + "grad_norm": 0.201171875, + "learning_rate": 0.0017821375710474325, + "loss": 0.0898, + "step": 22491 + }, + { + "epoch": 0.1952413607520768, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017821181555776585, + "loss": 0.0889, + "step": 22492 + }, + { + "epoch": 0.19525004123228096, + "grad_norm": 0.8515625, + "learning_rate": 0.0017820987393619302, + "loss": 0.0947, + "step": 22493 + }, + { + "epoch": 0.19525872171248512, + "grad_norm": 0.146484375, + "learning_rate": 0.0017820793224002688, + "loss": 0.0952, + "step": 22494 + }, + { + "epoch": 0.1952674021926893, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017820599046926959, + "loss": 0.0947, + "step": 22495 + }, + { + "epoch": 0.19527608267289345, + "grad_norm": 0.43359375, + "learning_rate": 0.0017820404862392324, + "loss": 0.1455, + "step": 22496 + }, + { + "epoch": 0.19528476315309762, + "grad_norm": 0.5625, + "learning_rate": 0.0017820210670398998, + "loss": 0.0928, + "step": 22497 + }, + { + "epoch": 0.19529344363330178, + "grad_norm": 0.220703125, + "learning_rate": 0.0017820016470947188, + "loss": 0.1143, + "step": 22498 + }, + { + "epoch": 0.19530212411350595, + "grad_norm": 0.26171875, + "learning_rate": 0.0017819822264037114, + "loss": 0.0938, + "step": 22499 + }, + { + "epoch": 0.1953108045937101, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017819628049668983, + "loss": 0.0752, + "step": 22500 + }, + { + "epoch": 0.19531948507391428, + "grad_norm": 0.283203125, + "learning_rate": 0.0017819433827843015, + "loss": 0.1211, + "step": 22501 + }, + { + "epoch": 0.19532816555411844, + "grad_norm": 0.310546875, + "learning_rate": 0.001781923959855941, + "loss": 0.0991, + "step": 22502 + }, + { + "epoch": 0.1953368460343226, + "grad_norm": 0.9453125, + "learning_rate": 0.0017819045361818391, + "loss": 0.1318, + "step": 22503 + }, + { + "epoch": 0.19534552651452677, + "grad_norm": 0.6328125, + "learning_rate": 0.0017818851117620165, + "loss": 0.0864, + "step": 22504 + }, + { + "epoch": 0.19535420699473094, + "grad_norm": 0.26953125, + "learning_rate": 0.0017818656865964944, + "loss": 0.104, + "step": 22505 + }, + { + "epoch": 0.1953628874749351, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017818462606852942, + "loss": 0.1167, + "step": 22506 + }, + { + "epoch": 0.19537156795513927, + "grad_norm": 0.37890625, + "learning_rate": 0.0017818268340284375, + "loss": 0.125, + "step": 22507 + }, + { + "epoch": 0.19538024843534343, + "grad_norm": 0.09814453125, + "learning_rate": 0.001781807406625945, + "loss": 0.1079, + "step": 22508 + }, + { + "epoch": 0.1953889289155476, + "grad_norm": 0.251953125, + "learning_rate": 0.0017817879784778386, + "loss": 0.1123, + "step": 22509 + }, + { + "epoch": 0.19539760939575176, + "grad_norm": 0.458984375, + "learning_rate": 0.0017817685495841386, + "loss": 0.1162, + "step": 22510 + }, + { + "epoch": 0.19540628987595593, + "grad_norm": 0.11572265625, + "learning_rate": 0.001781749119944867, + "loss": 0.1357, + "step": 22511 + }, + { + "epoch": 0.1954149703561601, + "grad_norm": 2.046875, + "learning_rate": 0.0017817296895600446, + "loss": 0.4629, + "step": 22512 + }, + { + "epoch": 0.19542365083636426, + "grad_norm": 0.1474609375, + "learning_rate": 0.001781710258429693, + "loss": 0.0962, + "step": 22513 + }, + { + "epoch": 0.19543233131656843, + "grad_norm": 0.2353515625, + "learning_rate": 0.0017816908265538332, + "loss": 0.1001, + "step": 22514 + }, + { + "epoch": 0.1954410117967726, + "grad_norm": 0.251953125, + "learning_rate": 0.001781671393932487, + "loss": 0.1328, + "step": 22515 + }, + { + "epoch": 0.19544969227697676, + "grad_norm": 0.408203125, + "learning_rate": 0.0017816519605656747, + "loss": 0.1084, + "step": 22516 + }, + { + "epoch": 0.19545837275718092, + "grad_norm": 1.0390625, + "learning_rate": 0.001781632526453418, + "loss": 0.106, + "step": 22517 + }, + { + "epoch": 0.19546705323738509, + "grad_norm": 0.80078125, + "learning_rate": 0.0017816130915957384, + "loss": 0.1094, + "step": 22518 + }, + { + "epoch": 0.19547573371758925, + "grad_norm": 0.71484375, + "learning_rate": 0.001781593655992657, + "loss": 0.1221, + "step": 22519 + }, + { + "epoch": 0.19548441419779342, + "grad_norm": 0.9453125, + "learning_rate": 0.0017815742196441952, + "loss": 0.1504, + "step": 22520 + }, + { + "epoch": 0.19549309467799758, + "grad_norm": 0.2734375, + "learning_rate": 0.001781554782550374, + "loss": 0.0918, + "step": 22521 + }, + { + "epoch": 0.19550177515820175, + "grad_norm": 0.419921875, + "learning_rate": 0.0017815353447112143, + "loss": 0.1104, + "step": 22522 + }, + { + "epoch": 0.1955104556384059, + "grad_norm": 0.09912109375, + "learning_rate": 0.001781515906126738, + "loss": 0.1084, + "step": 22523 + }, + { + "epoch": 0.19551913611861008, + "grad_norm": 0.11328125, + "learning_rate": 0.0017814964667969663, + "loss": 0.1309, + "step": 22524 + }, + { + "epoch": 0.19552781659881424, + "grad_norm": 0.310546875, + "learning_rate": 0.0017814770267219205, + "loss": 0.1221, + "step": 22525 + }, + { + "epoch": 0.1955364970790184, + "grad_norm": 0.1162109375, + "learning_rate": 0.001781457585901621, + "loss": 0.1182, + "step": 22526 + }, + { + "epoch": 0.19554517755922257, + "grad_norm": 0.8515625, + "learning_rate": 0.0017814381443360904, + "loss": 0.0981, + "step": 22527 + }, + { + "epoch": 0.19555385803942674, + "grad_norm": 0.08447265625, + "learning_rate": 0.001781418702025349, + "loss": 0.082, + "step": 22528 + }, + { + "epoch": 0.1955625385196309, + "grad_norm": 0.400390625, + "learning_rate": 0.0017813992589694183, + "loss": 0.1201, + "step": 22529 + }, + { + "epoch": 0.19557121899983507, + "grad_norm": 0.09765625, + "learning_rate": 0.0017813798151683196, + "loss": 0.1328, + "step": 22530 + }, + { + "epoch": 0.19557989948003923, + "grad_norm": 0.5859375, + "learning_rate": 0.0017813603706220745, + "loss": 0.127, + "step": 22531 + }, + { + "epoch": 0.1955885799602434, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017813409253307033, + "loss": 0.1387, + "step": 22532 + }, + { + "epoch": 0.19559726044044756, + "grad_norm": 0.16796875, + "learning_rate": 0.0017813214792942284, + "loss": 0.1367, + "step": 22533 + }, + { + "epoch": 0.19560594092065173, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017813020325126705, + "loss": 0.1104, + "step": 22534 + }, + { + "epoch": 0.1956146214008559, + "grad_norm": 0.2353515625, + "learning_rate": 0.0017812825849860508, + "loss": 0.1504, + "step": 22535 + }, + { + "epoch": 0.19562330188106006, + "grad_norm": 0.298828125, + "learning_rate": 0.001781263136714391, + "loss": 0.0737, + "step": 22536 + }, + { + "epoch": 0.19563198236126422, + "grad_norm": 0.146484375, + "learning_rate": 0.0017812436876977118, + "loss": 0.0957, + "step": 22537 + }, + { + "epoch": 0.1956406628414684, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017812242379360347, + "loss": 0.1387, + "step": 22538 + }, + { + "epoch": 0.19564934332167255, + "grad_norm": 0.275390625, + "learning_rate": 0.001781204787429381, + "loss": 0.1143, + "step": 22539 + }, + { + "epoch": 0.19565802380187672, + "grad_norm": 0.1552734375, + "learning_rate": 0.001781185336177772, + "loss": 0.1035, + "step": 22540 + }, + { + "epoch": 0.19566670428208088, + "grad_norm": 0.1982421875, + "learning_rate": 0.001781165884181229, + "loss": 0.1016, + "step": 22541 + }, + { + "epoch": 0.19567538476228505, + "grad_norm": 0.2578125, + "learning_rate": 0.0017811464314397734, + "loss": 0.1045, + "step": 22542 + }, + { + "epoch": 0.1956840652424892, + "grad_norm": 0.255859375, + "learning_rate": 0.001781126977953426, + "loss": 0.127, + "step": 22543 + }, + { + "epoch": 0.19569274572269338, + "grad_norm": 0.5390625, + "learning_rate": 0.0017811075237222085, + "loss": 0.1465, + "step": 22544 + }, + { + "epoch": 0.19570142620289754, + "grad_norm": 0.1328125, + "learning_rate": 0.0017810880687461424, + "loss": 0.0938, + "step": 22545 + }, + { + "epoch": 0.1957101066831017, + "grad_norm": 0.2353515625, + "learning_rate": 0.001781068613025248, + "loss": 0.0977, + "step": 22546 + }, + { + "epoch": 0.19571878716330587, + "grad_norm": 0.1640625, + "learning_rate": 0.0017810491565595475, + "loss": 0.1992, + "step": 22547 + }, + { + "epoch": 0.19572746764351004, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017810296993490617, + "loss": 0.1289, + "step": 22548 + }, + { + "epoch": 0.1957361481237142, + "grad_norm": 0.2216796875, + "learning_rate": 0.0017810102413938122, + "loss": 0.1025, + "step": 22549 + }, + { + "epoch": 0.19574482860391837, + "grad_norm": 0.40234375, + "learning_rate": 0.0017809907826938202, + "loss": 0.1816, + "step": 22550 + }, + { + "epoch": 0.19575350908412253, + "grad_norm": 0.8671875, + "learning_rate": 0.0017809713232491069, + "loss": 0.0967, + "step": 22551 + }, + { + "epoch": 0.1957621895643267, + "grad_norm": 0.58203125, + "learning_rate": 0.0017809518630596934, + "loss": 0.1055, + "step": 22552 + }, + { + "epoch": 0.19577087004453086, + "grad_norm": 0.87109375, + "learning_rate": 0.0017809324021256013, + "loss": 0.084, + "step": 22553 + }, + { + "epoch": 0.19577955052473503, + "grad_norm": 0.07080078125, + "learning_rate": 0.001780912940446852, + "loss": 0.0737, + "step": 22554 + }, + { + "epoch": 0.1957882310049392, + "grad_norm": 0.546875, + "learning_rate": 0.0017808934780234663, + "loss": 0.1055, + "step": 22555 + }, + { + "epoch": 0.19579691148514336, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017808740148554658, + "loss": 0.1084, + "step": 22556 + }, + { + "epoch": 0.19580559196534753, + "grad_norm": 0.478515625, + "learning_rate": 0.0017808545509428715, + "loss": 0.1387, + "step": 22557 + }, + { + "epoch": 0.1958142724455517, + "grad_norm": 0.267578125, + "learning_rate": 0.001780835086285705, + "loss": 0.1089, + "step": 22558 + }, + { + "epoch": 0.19582295292575586, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017808156208839877, + "loss": 0.1084, + "step": 22559 + }, + { + "epoch": 0.19583163340596002, + "grad_norm": 0.423828125, + "learning_rate": 0.0017807961547377403, + "loss": 0.0938, + "step": 22560 + }, + { + "epoch": 0.19584031388616419, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017807766878469846, + "loss": 0.1206, + "step": 22561 + }, + { + "epoch": 0.19584899436636835, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017807572202117419, + "loss": 0.1187, + "step": 22562 + }, + { + "epoch": 0.19585767484657252, + "grad_norm": 0.796875, + "learning_rate": 0.0017807377518320332, + "loss": 0.1035, + "step": 22563 + }, + { + "epoch": 0.19586635532677668, + "grad_norm": 0.5234375, + "learning_rate": 0.0017807182827078798, + "loss": 0.1045, + "step": 22564 + }, + { + "epoch": 0.19587503580698085, + "grad_norm": 0.443359375, + "learning_rate": 0.0017806988128393033, + "loss": 0.1133, + "step": 22565 + }, + { + "epoch": 0.195883716287185, + "grad_norm": 0.349609375, + "learning_rate": 0.0017806793422263248, + "loss": 0.1011, + "step": 22566 + }, + { + "epoch": 0.19589239676738918, + "grad_norm": 0.70703125, + "learning_rate": 0.0017806598708689657, + "loss": 0.1279, + "step": 22567 + }, + { + "epoch": 0.19590107724759334, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017806403987672469, + "loss": 0.1426, + "step": 22568 + }, + { + "epoch": 0.1959097577277975, + "grad_norm": 0.28125, + "learning_rate": 0.0017806209259211903, + "loss": 0.0996, + "step": 22569 + }, + { + "epoch": 0.19591843820800167, + "grad_norm": 0.154296875, + "learning_rate": 0.0017806014523308162, + "loss": 0.1123, + "step": 22570 + }, + { + "epoch": 0.19592711868820584, + "grad_norm": 0.58984375, + "learning_rate": 0.0017805819779961474, + "loss": 0.0854, + "step": 22571 + }, + { + "epoch": 0.19593579916841, + "grad_norm": 0.138671875, + "learning_rate": 0.0017805625029172038, + "loss": 0.1758, + "step": 22572 + }, + { + "epoch": 0.19594447964861417, + "grad_norm": 0.240234375, + "learning_rate": 0.0017805430270940077, + "loss": 0.1436, + "step": 22573 + }, + { + "epoch": 0.19595316012881833, + "grad_norm": 0.6640625, + "learning_rate": 0.0017805235505265798, + "loss": 0.1885, + "step": 22574 + }, + { + "epoch": 0.1959618406090225, + "grad_norm": 0.08740234375, + "learning_rate": 0.0017805040732149415, + "loss": 0.0986, + "step": 22575 + }, + { + "epoch": 0.19597052108922666, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017804845951591144, + "loss": 0.0918, + "step": 22576 + }, + { + "epoch": 0.19597920156943083, + "grad_norm": 0.341796875, + "learning_rate": 0.0017804651163591193, + "loss": 0.1445, + "step": 22577 + }, + { + "epoch": 0.195987882049635, + "grad_norm": 0.146484375, + "learning_rate": 0.0017804456368149782, + "loss": 0.1367, + "step": 22578 + }, + { + "epoch": 0.19599656252983916, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017804261565267116, + "loss": 0.1079, + "step": 22579 + }, + { + "epoch": 0.19600524301004332, + "grad_norm": 0.8046875, + "learning_rate": 0.001780406675494341, + "loss": 0.1934, + "step": 22580 + }, + { + "epoch": 0.1960139234902475, + "grad_norm": 0.3828125, + "learning_rate": 0.0017803871937178883, + "loss": 0.1123, + "step": 22581 + }, + { + "epoch": 0.19602260397045165, + "grad_norm": 0.06396484375, + "learning_rate": 0.0017803677111973741, + "loss": 0.1123, + "step": 22582 + }, + { + "epoch": 0.19603128445065582, + "grad_norm": 0.248046875, + "learning_rate": 0.00178034822793282, + "loss": 0.1182, + "step": 22583 + }, + { + "epoch": 0.19603996493085998, + "grad_norm": 0.412109375, + "learning_rate": 0.0017803287439242476, + "loss": 0.127, + "step": 22584 + }, + { + "epoch": 0.19604864541106415, + "grad_norm": 0.119140625, + "learning_rate": 0.001780309259171678, + "loss": 0.104, + "step": 22585 + }, + { + "epoch": 0.1960573258912683, + "grad_norm": 0.365234375, + "learning_rate": 0.0017802897736751318, + "loss": 0.1172, + "step": 22586 + }, + { + "epoch": 0.19606600637147248, + "grad_norm": 0.0849609375, + "learning_rate": 0.0017802702874346312, + "loss": 0.127, + "step": 22587 + }, + { + "epoch": 0.19607468685167664, + "grad_norm": 0.205078125, + "learning_rate": 0.0017802508004501974, + "loss": 0.105, + "step": 22588 + }, + { + "epoch": 0.1960833673318808, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017802313127218512, + "loss": 0.1123, + "step": 22589 + }, + { + "epoch": 0.19609204781208497, + "grad_norm": 0.734375, + "learning_rate": 0.0017802118242496148, + "loss": 0.125, + "step": 22590 + }, + { + "epoch": 0.19610072829228914, + "grad_norm": 0.97265625, + "learning_rate": 0.0017801923350335084, + "loss": 0.127, + "step": 22591 + }, + { + "epoch": 0.1961094087724933, + "grad_norm": 0.337890625, + "learning_rate": 0.0017801728450735543, + "loss": 0.0791, + "step": 22592 + }, + { + "epoch": 0.19611808925269747, + "grad_norm": 0.408203125, + "learning_rate": 0.0017801533543697734, + "loss": 0.1133, + "step": 22593 + }, + { + "epoch": 0.19612676973290163, + "grad_norm": 0.337890625, + "learning_rate": 0.0017801338629221865, + "loss": 0.1055, + "step": 22594 + }, + { + "epoch": 0.1961354502131058, + "grad_norm": 0.169921875, + "learning_rate": 0.001780114370730816, + "loss": 0.0781, + "step": 22595 + }, + { + "epoch": 0.19614413069330996, + "grad_norm": 0.349609375, + "learning_rate": 0.0017800948777956826, + "loss": 0.1611, + "step": 22596 + }, + { + "epoch": 0.19615281117351413, + "grad_norm": 0.2109375, + "learning_rate": 0.0017800753841168073, + "loss": 0.1436, + "step": 22597 + }, + { + "epoch": 0.1961614916537183, + "grad_norm": 0.2197265625, + "learning_rate": 0.0017800558896942118, + "loss": 0.0811, + "step": 22598 + }, + { + "epoch": 0.19617017213392246, + "grad_norm": 0.1611328125, + "learning_rate": 0.0017800363945279177, + "loss": 0.0593, + "step": 22599 + }, + { + "epoch": 0.19617885261412663, + "grad_norm": 0.470703125, + "learning_rate": 0.0017800168986179456, + "loss": 0.1025, + "step": 22600 + }, + { + "epoch": 0.1961875330943308, + "grad_norm": 0.12890625, + "learning_rate": 0.0017799974019643177, + "loss": 0.0786, + "step": 22601 + }, + { + "epoch": 0.19619621357453496, + "grad_norm": 0.076171875, + "learning_rate": 0.0017799779045670548, + "loss": 0.0889, + "step": 22602 + }, + { + "epoch": 0.19620489405473912, + "grad_norm": 0.11865234375, + "learning_rate": 0.001779958406426178, + "loss": 0.1045, + "step": 22603 + }, + { + "epoch": 0.19621357453494329, + "grad_norm": 0.404296875, + "learning_rate": 0.0017799389075417089, + "loss": 0.2148, + "step": 22604 + }, + { + "epoch": 0.19622225501514745, + "grad_norm": 0.203125, + "learning_rate": 0.0017799194079136692, + "loss": 0.1006, + "step": 22605 + }, + { + "epoch": 0.19623093549535162, + "grad_norm": 0.27734375, + "learning_rate": 0.0017798999075420796, + "loss": 0.0811, + "step": 22606 + }, + { + "epoch": 0.19623961597555578, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017798804064269617, + "loss": 0.126, + "step": 22607 + }, + { + "epoch": 0.19624829645575995, + "grad_norm": 0.64453125, + "learning_rate": 0.0017798609045683368, + "loss": 0.0938, + "step": 22608 + }, + { + "epoch": 0.19625697693596408, + "grad_norm": 0.25, + "learning_rate": 0.001779841401966226, + "loss": 0.166, + "step": 22609 + }, + { + "epoch": 0.19626565741616825, + "grad_norm": 0.2314453125, + "learning_rate": 0.0017798218986206512, + "loss": 0.106, + "step": 22610 + }, + { + "epoch": 0.19627433789637241, + "grad_norm": 0.1826171875, + "learning_rate": 0.0017798023945316333, + "loss": 0.1035, + "step": 22611 + }, + { + "epoch": 0.19628301837657658, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017797828896991936, + "loss": 0.1084, + "step": 22612 + }, + { + "epoch": 0.19629169885678074, + "grad_norm": 0.1005859375, + "learning_rate": 0.0017797633841233537, + "loss": 0.1055, + "step": 22613 + }, + { + "epoch": 0.1963003793369849, + "grad_norm": 0.3359375, + "learning_rate": 0.0017797438778041346, + "loss": 0.0898, + "step": 22614 + }, + { + "epoch": 0.19630905981718907, + "grad_norm": 0.28125, + "learning_rate": 0.001779724370741558, + "loss": 0.1104, + "step": 22615 + }, + { + "epoch": 0.19631774029739324, + "grad_norm": 0.369140625, + "learning_rate": 0.0017797048629356448, + "loss": 0.1338, + "step": 22616 + }, + { + "epoch": 0.1963264207775974, + "grad_norm": 0.1630859375, + "learning_rate": 0.001779685354386417, + "loss": 0.1108, + "step": 22617 + }, + { + "epoch": 0.19633510125780157, + "grad_norm": 0.17578125, + "learning_rate": 0.0017796658450938952, + "loss": 0.1182, + "step": 22618 + }, + { + "epoch": 0.19634378173800573, + "grad_norm": 0.46484375, + "learning_rate": 0.0017796463350581012, + "loss": 0.1108, + "step": 22619 + }, + { + "epoch": 0.1963524622182099, + "grad_norm": 0.423828125, + "learning_rate": 0.001779626824279056, + "loss": 0.1221, + "step": 22620 + }, + { + "epoch": 0.19636114269841407, + "grad_norm": 0.359375, + "learning_rate": 0.001779607312756781, + "loss": 0.1514, + "step": 22621 + }, + { + "epoch": 0.19636982317861823, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017795878004912983, + "loss": 0.123, + "step": 22622 + }, + { + "epoch": 0.1963785036588224, + "grad_norm": 0.2421875, + "learning_rate": 0.0017795682874826278, + "loss": 0.1201, + "step": 22623 + }, + { + "epoch": 0.19638718413902656, + "grad_norm": 0.341796875, + "learning_rate": 0.001779548773730792, + "loss": 0.125, + "step": 22624 + }, + { + "epoch": 0.19639586461923073, + "grad_norm": 0.234375, + "learning_rate": 0.0017795292592358116, + "loss": 0.0977, + "step": 22625 + }, + { + "epoch": 0.1964045450994349, + "grad_norm": 1.09375, + "learning_rate": 0.0017795097439977086, + "loss": 0.1504, + "step": 22626 + }, + { + "epoch": 0.19641322557963906, + "grad_norm": 0.78515625, + "learning_rate": 0.001779490228016504, + "loss": 0.1074, + "step": 22627 + }, + { + "epoch": 0.19642190605984322, + "grad_norm": 0.0771484375, + "learning_rate": 0.001779470711292219, + "loss": 0.082, + "step": 22628 + }, + { + "epoch": 0.19643058654004739, + "grad_norm": 0.63671875, + "learning_rate": 0.0017794511938248746, + "loss": 0.1025, + "step": 22629 + }, + { + "epoch": 0.19643926702025155, + "grad_norm": 0.390625, + "learning_rate": 0.001779431675614493, + "loss": 0.0732, + "step": 22630 + }, + { + "epoch": 0.19644794750045572, + "grad_norm": 0.6015625, + "learning_rate": 0.001779412156661095, + "loss": 0.063, + "step": 22631 + }, + { + "epoch": 0.19645662798065988, + "grad_norm": 0.90625, + "learning_rate": 0.0017793926369647022, + "loss": 0.168, + "step": 22632 + }, + { + "epoch": 0.19646530846086405, + "grad_norm": 0.07373046875, + "learning_rate": 0.0017793731165253357, + "loss": 0.0996, + "step": 22633 + }, + { + "epoch": 0.1964739889410682, + "grad_norm": 0.53515625, + "learning_rate": 0.001779353595343017, + "loss": 0.1191, + "step": 22634 + }, + { + "epoch": 0.19648266942127238, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017793340734177673, + "loss": 0.0977, + "step": 22635 + }, + { + "epoch": 0.19649134990147654, + "grad_norm": 0.9765625, + "learning_rate": 0.0017793145507496082, + "loss": 0.1484, + "step": 22636 + }, + { + "epoch": 0.1965000303816807, + "grad_norm": 0.953125, + "learning_rate": 0.001779295027338561, + "loss": 0.1025, + "step": 22637 + }, + { + "epoch": 0.19650871086188487, + "grad_norm": 0.359375, + "learning_rate": 0.0017792755031846468, + "loss": 0.1328, + "step": 22638 + }, + { + "epoch": 0.19651739134208904, + "grad_norm": 0.6796875, + "learning_rate": 0.0017792559782878873, + "loss": 0.0791, + "step": 22639 + }, + { + "epoch": 0.1965260718222932, + "grad_norm": 0.291015625, + "learning_rate": 0.0017792364526483034, + "loss": 0.1123, + "step": 22640 + }, + { + "epoch": 0.19653475230249737, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017792169262659167, + "loss": 0.1099, + "step": 22641 + }, + { + "epoch": 0.19654343278270153, + "grad_norm": 1.109375, + "learning_rate": 0.001779197399140749, + "loss": 0.1553, + "step": 22642 + }, + { + "epoch": 0.1965521132629057, + "grad_norm": 0.6640625, + "learning_rate": 0.0017791778712728209, + "loss": 0.1172, + "step": 22643 + }, + { + "epoch": 0.19656079374310986, + "grad_norm": 0.6875, + "learning_rate": 0.0017791583426621542, + "loss": 0.1143, + "step": 22644 + }, + { + "epoch": 0.19656947422331403, + "grad_norm": 0.376953125, + "learning_rate": 0.0017791388133087704, + "loss": 0.0942, + "step": 22645 + }, + { + "epoch": 0.1965781547035182, + "grad_norm": 0.451171875, + "learning_rate": 0.0017791192832126899, + "loss": 0.1328, + "step": 22646 + }, + { + "epoch": 0.19658683518372236, + "grad_norm": 0.341796875, + "learning_rate": 0.0017790997523739352, + "loss": 0.1025, + "step": 22647 + }, + { + "epoch": 0.19659551566392652, + "grad_norm": 0.796875, + "learning_rate": 0.0017790802207925272, + "loss": 0.1484, + "step": 22648 + }, + { + "epoch": 0.1966041961441307, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017790606884684871, + "loss": 0.1318, + "step": 22649 + }, + { + "epoch": 0.19661287662433485, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017790411554018367, + "loss": 0.1426, + "step": 22650 + }, + { + "epoch": 0.19662155710453902, + "grad_norm": 0.2109375, + "learning_rate": 0.001779021621592597, + "loss": 0.1055, + "step": 22651 + }, + { + "epoch": 0.19663023758474318, + "grad_norm": 0.76171875, + "learning_rate": 0.0017790020870407895, + "loss": 0.1318, + "step": 22652 + }, + { + "epoch": 0.19663891806494735, + "grad_norm": 0.310546875, + "learning_rate": 0.0017789825517464356, + "loss": 0.0928, + "step": 22653 + }, + { + "epoch": 0.19664759854515151, + "grad_norm": 0.396484375, + "learning_rate": 0.0017789630157095562, + "loss": 0.1377, + "step": 22654 + }, + { + "epoch": 0.19665627902535568, + "grad_norm": 0.296875, + "learning_rate": 0.0017789434789301736, + "loss": 0.0732, + "step": 22655 + }, + { + "epoch": 0.19666495950555984, + "grad_norm": 0.09423828125, + "learning_rate": 0.001778923941408308, + "loss": 0.1021, + "step": 22656 + }, + { + "epoch": 0.196673639985764, + "grad_norm": 0.08203125, + "learning_rate": 0.0017789044031439815, + "loss": 0.1138, + "step": 22657 + }, + { + "epoch": 0.19668232046596817, + "grad_norm": 0.12890625, + "learning_rate": 0.0017788848641372157, + "loss": 0.1211, + "step": 22658 + }, + { + "epoch": 0.19669100094617234, + "grad_norm": 0.205078125, + "learning_rate": 0.0017788653243880315, + "loss": 0.083, + "step": 22659 + }, + { + "epoch": 0.1966996814263765, + "grad_norm": 0.26171875, + "learning_rate": 0.0017788457838964502, + "loss": 0.1006, + "step": 22660 + }, + { + "epoch": 0.19670836190658067, + "grad_norm": 0.435546875, + "learning_rate": 0.0017788262426624935, + "loss": 0.1016, + "step": 22661 + }, + { + "epoch": 0.19671704238678484, + "grad_norm": 0.4140625, + "learning_rate": 0.0017788067006861826, + "loss": 0.1123, + "step": 22662 + }, + { + "epoch": 0.196725722866989, + "grad_norm": 0.2275390625, + "learning_rate": 0.0017787871579675388, + "loss": 0.1299, + "step": 22663 + }, + { + "epoch": 0.19673440334719317, + "grad_norm": 0.138671875, + "learning_rate": 0.0017787676145065837, + "loss": 0.1016, + "step": 22664 + }, + { + "epoch": 0.19674308382739733, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017787480703033386, + "loss": 0.0962, + "step": 22665 + }, + { + "epoch": 0.1967517643076015, + "grad_norm": 0.1474609375, + "learning_rate": 0.0017787285253578246, + "loss": 0.1016, + "step": 22666 + }, + { + "epoch": 0.19676044478780566, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017787089796700633, + "loss": 0.085, + "step": 22667 + }, + { + "epoch": 0.19676912526800983, + "grad_norm": 0.0966796875, + "learning_rate": 0.001778689433240076, + "loss": 0.1187, + "step": 22668 + }, + { + "epoch": 0.196777805748214, + "grad_norm": 0.9765625, + "learning_rate": 0.0017786698860678843, + "loss": 0.125, + "step": 22669 + }, + { + "epoch": 0.19678648622841816, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017786503381535092, + "loss": 0.0889, + "step": 22670 + }, + { + "epoch": 0.19679516670862232, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017786307894969726, + "loss": 0.0957, + "step": 22671 + }, + { + "epoch": 0.1968038471888265, + "grad_norm": 0.11962890625, + "learning_rate": 0.0017786112400982954, + "loss": 0.0967, + "step": 22672 + }, + { + "epoch": 0.19681252766903065, + "grad_norm": 0.20703125, + "learning_rate": 0.0017785916899574993, + "loss": 0.1006, + "step": 22673 + }, + { + "epoch": 0.19682120814923482, + "grad_norm": 0.17578125, + "learning_rate": 0.0017785721390746053, + "loss": 0.1465, + "step": 22674 + }, + { + "epoch": 0.19682988862943898, + "grad_norm": 0.5390625, + "learning_rate": 0.001778552587449635, + "loss": 0.1445, + "step": 22675 + }, + { + "epoch": 0.19683856910964315, + "grad_norm": 0.41015625, + "learning_rate": 0.0017785330350826099, + "loss": 0.0972, + "step": 22676 + }, + { + "epoch": 0.1968472495898473, + "grad_norm": 0.51953125, + "learning_rate": 0.0017785134819735514, + "loss": 0.1025, + "step": 22677 + }, + { + "epoch": 0.19685593007005148, + "grad_norm": 0.97265625, + "learning_rate": 0.0017784939281224803, + "loss": 0.1348, + "step": 22678 + }, + { + "epoch": 0.19686461055025564, + "grad_norm": 0.2099609375, + "learning_rate": 0.001778474373529419, + "loss": 0.1377, + "step": 22679 + }, + { + "epoch": 0.1968732910304598, + "grad_norm": 0.34375, + "learning_rate": 0.0017784548181943877, + "loss": 0.0977, + "step": 22680 + }, + { + "epoch": 0.19688197151066397, + "grad_norm": 0.09765625, + "learning_rate": 0.0017784352621174086, + "loss": 0.1357, + "step": 22681 + }, + { + "epoch": 0.19689065199086814, + "grad_norm": 0.2109375, + "learning_rate": 0.0017784157052985032, + "loss": 0.0884, + "step": 22682 + }, + { + "epoch": 0.1968993324710723, + "grad_norm": 0.474609375, + "learning_rate": 0.0017783961477376928, + "loss": 0.1465, + "step": 22683 + }, + { + "epoch": 0.19690801295127647, + "grad_norm": 0.63671875, + "learning_rate": 0.001778376589434998, + "loss": 0.085, + "step": 22684 + }, + { + "epoch": 0.19691669343148063, + "grad_norm": 0.39453125, + "learning_rate": 0.0017783570303904408, + "loss": 0.126, + "step": 22685 + }, + { + "epoch": 0.1969253739116848, + "grad_norm": 0.2021484375, + "learning_rate": 0.001778337470604043, + "loss": 0.1113, + "step": 22686 + }, + { + "epoch": 0.19693405439188896, + "grad_norm": 0.61328125, + "learning_rate": 0.001778317910075825, + "loss": 0.0947, + "step": 22687 + }, + { + "epoch": 0.19694273487209313, + "grad_norm": 0.71484375, + "learning_rate": 0.0017782983488058089, + "loss": 0.1406, + "step": 22688 + }, + { + "epoch": 0.1969514153522973, + "grad_norm": 0.298828125, + "learning_rate": 0.0017782787867940157, + "loss": 0.1016, + "step": 22689 + }, + { + "epoch": 0.19696009583250146, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017782592240404675, + "loss": 0.0859, + "step": 22690 + }, + { + "epoch": 0.19696877631270562, + "grad_norm": 0.357421875, + "learning_rate": 0.001778239660545185, + "loss": 0.1025, + "step": 22691 + }, + { + "epoch": 0.1969774567929098, + "grad_norm": 0.423828125, + "learning_rate": 0.0017782200963081898, + "loss": 0.085, + "step": 22692 + }, + { + "epoch": 0.19698613727311395, + "grad_norm": 0.349609375, + "learning_rate": 0.0017782005313295031, + "loss": 0.1196, + "step": 22693 + }, + { + "epoch": 0.19699481775331812, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017781809656091468, + "loss": 0.0986, + "step": 22694 + }, + { + "epoch": 0.19700349823352228, + "grad_norm": 0.171875, + "learning_rate": 0.0017781613991471417, + "loss": 0.1211, + "step": 22695 + }, + { + "epoch": 0.19701217871372645, + "grad_norm": 0.51171875, + "learning_rate": 0.0017781418319435093, + "loss": 0.1167, + "step": 22696 + }, + { + "epoch": 0.19702085919393061, + "grad_norm": 0.1796875, + "learning_rate": 0.0017781222639982716, + "loss": 0.0977, + "step": 22697 + }, + { + "epoch": 0.19702953967413478, + "grad_norm": 0.66796875, + "learning_rate": 0.0017781026953114496, + "loss": 0.1201, + "step": 22698 + }, + { + "epoch": 0.19703822015433894, + "grad_norm": 0.78515625, + "learning_rate": 0.0017780831258830646, + "loss": 0.0977, + "step": 22699 + }, + { + "epoch": 0.1970469006345431, + "grad_norm": 0.384765625, + "learning_rate": 0.001778063555713138, + "loss": 0.1484, + "step": 22700 + }, + { + "epoch": 0.19705558111474727, + "grad_norm": 0.16015625, + "learning_rate": 0.0017780439848016914, + "loss": 0.127, + "step": 22701 + }, + { + "epoch": 0.19706426159495144, + "grad_norm": 0.640625, + "learning_rate": 0.001778024413148746, + "loss": 0.0811, + "step": 22702 + }, + { + "epoch": 0.1970729420751556, + "grad_norm": 0.62109375, + "learning_rate": 0.0017780048407543233, + "loss": 0.1602, + "step": 22703 + }, + { + "epoch": 0.19708162255535977, + "grad_norm": 0.5703125, + "learning_rate": 0.0017779852676184447, + "loss": 0.1035, + "step": 22704 + }, + { + "epoch": 0.19709030303556394, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017779656937411317, + "loss": 0.1299, + "step": 22705 + }, + { + "epoch": 0.1970989835157681, + "grad_norm": 0.171875, + "learning_rate": 0.0017779461191224054, + "loss": 0.1113, + "step": 22706 + }, + { + "epoch": 0.19710766399597227, + "grad_norm": 0.298828125, + "learning_rate": 0.0017779265437622876, + "loss": 0.1279, + "step": 22707 + }, + { + "epoch": 0.19711634447617643, + "grad_norm": 0.328125, + "learning_rate": 0.0017779069676607992, + "loss": 0.064, + "step": 22708 + }, + { + "epoch": 0.1971250249563806, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017778873908179623, + "loss": 0.1328, + "step": 22709 + }, + { + "epoch": 0.19713370543658476, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017778678132337978, + "loss": 0.1128, + "step": 22710 + }, + { + "epoch": 0.19714238591678893, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017778482349083272, + "loss": 0.1074, + "step": 22711 + }, + { + "epoch": 0.1971510663969931, + "grad_norm": 0.251953125, + "learning_rate": 0.001777828655841572, + "loss": 0.1777, + "step": 22712 + }, + { + "epoch": 0.19715974687719726, + "grad_norm": 0.29296875, + "learning_rate": 0.0017778090760335534, + "loss": 0.1084, + "step": 22713 + }, + { + "epoch": 0.19716842735740142, + "grad_norm": 0.259765625, + "learning_rate": 0.001777789495484293, + "loss": 0.1094, + "step": 22714 + }, + { + "epoch": 0.1971771078376056, + "grad_norm": 0.07568359375, + "learning_rate": 0.0017777699141938125, + "loss": 0.0859, + "step": 22715 + }, + { + "epoch": 0.19718578831780975, + "grad_norm": 0.458984375, + "learning_rate": 0.0017777503321621327, + "loss": 0.1562, + "step": 22716 + }, + { + "epoch": 0.19719446879801392, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017777307493892756, + "loss": 0.0811, + "step": 22717 + }, + { + "epoch": 0.19720314927821808, + "grad_norm": 0.1630859375, + "learning_rate": 0.001777711165875262, + "loss": 0.1113, + "step": 22718 + }, + { + "epoch": 0.19721182975842225, + "grad_norm": 0.12890625, + "learning_rate": 0.0017776915816201137, + "loss": 0.1133, + "step": 22719 + }, + { + "epoch": 0.1972205102386264, + "grad_norm": 0.5078125, + "learning_rate": 0.0017776719966238524, + "loss": 0.0977, + "step": 22720 + }, + { + "epoch": 0.19722919071883058, + "grad_norm": 0.0859375, + "learning_rate": 0.001777652410886499, + "loss": 0.1123, + "step": 22721 + }, + { + "epoch": 0.19723787119903474, + "grad_norm": 0.35546875, + "learning_rate": 0.0017776328244080751, + "loss": 0.0967, + "step": 22722 + }, + { + "epoch": 0.1972465516792389, + "grad_norm": 0.10107421875, + "learning_rate": 0.001777613237188602, + "loss": 0.1182, + "step": 22723 + }, + { + "epoch": 0.19725523215944307, + "grad_norm": 0.53125, + "learning_rate": 0.0017775936492281015, + "loss": 0.0767, + "step": 22724 + }, + { + "epoch": 0.19726391263964724, + "grad_norm": 0.404296875, + "learning_rate": 0.0017775740605265947, + "loss": 0.1143, + "step": 22725 + }, + { + "epoch": 0.1972725931198514, + "grad_norm": 0.48828125, + "learning_rate": 0.001777554471084103, + "loss": 0.1187, + "step": 22726 + }, + { + "epoch": 0.19728127360005557, + "grad_norm": 0.30859375, + "learning_rate": 0.0017775348809006479, + "loss": 0.1035, + "step": 22727 + }, + { + "epoch": 0.19728995408025973, + "grad_norm": 0.119140625, + "learning_rate": 0.0017775152899762508, + "loss": 0.0854, + "step": 22728 + }, + { + "epoch": 0.1972986345604639, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017774956983109332, + "loss": 0.0967, + "step": 22729 + }, + { + "epoch": 0.19730731504066806, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017774761059047165, + "loss": 0.0693, + "step": 22730 + }, + { + "epoch": 0.19731599552087223, + "grad_norm": 0.408203125, + "learning_rate": 0.0017774565127576221, + "loss": 0.0747, + "step": 22731 + }, + { + "epoch": 0.1973246760010764, + "grad_norm": 0.82421875, + "learning_rate": 0.0017774369188696714, + "loss": 0.0938, + "step": 22732 + }, + { + "epoch": 0.19733335648128053, + "grad_norm": 0.2001953125, + "learning_rate": 0.001777417324240886, + "loss": 0.0859, + "step": 22733 + }, + { + "epoch": 0.1973420369614847, + "grad_norm": 0.2265625, + "learning_rate": 0.0017773977288712871, + "loss": 0.1094, + "step": 22734 + }, + { + "epoch": 0.19735071744168886, + "grad_norm": 0.291015625, + "learning_rate": 0.001777378132760896, + "loss": 0.0938, + "step": 22735 + }, + { + "epoch": 0.19735939792189303, + "grad_norm": 0.5390625, + "learning_rate": 0.001777358535909735, + "loss": 0.1177, + "step": 22736 + }, + { + "epoch": 0.1973680784020972, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017773389383178243, + "loss": 0.0996, + "step": 22737 + }, + { + "epoch": 0.19737675888230136, + "grad_norm": 0.12060546875, + "learning_rate": 0.0017773193399851861, + "loss": 0.0859, + "step": 22738 + }, + { + "epoch": 0.19738543936250552, + "grad_norm": 0.138671875, + "learning_rate": 0.0017772997409118414, + "loss": 0.0742, + "step": 22739 + }, + { + "epoch": 0.1973941198427097, + "grad_norm": 0.1923828125, + "learning_rate": 0.001777280141097812, + "loss": 0.0796, + "step": 22740 + }, + { + "epoch": 0.19740280032291385, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017772605405431195, + "loss": 0.0708, + "step": 22741 + }, + { + "epoch": 0.19741148080311802, + "grad_norm": 0.154296875, + "learning_rate": 0.0017772409392477848, + "loss": 0.1475, + "step": 22742 + }, + { + "epoch": 0.19742016128332218, + "grad_norm": 0.181640625, + "learning_rate": 0.0017772213372118295, + "loss": 0.0806, + "step": 22743 + }, + { + "epoch": 0.19742884176352635, + "grad_norm": 0.498046875, + "learning_rate": 0.0017772017344352756, + "loss": 0.125, + "step": 22744 + }, + { + "epoch": 0.1974375222437305, + "grad_norm": 0.314453125, + "learning_rate": 0.0017771821309181435, + "loss": 0.1406, + "step": 22745 + }, + { + "epoch": 0.19744620272393468, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017771625266604554, + "loss": 0.1064, + "step": 22746 + }, + { + "epoch": 0.19745488320413884, + "grad_norm": 0.73828125, + "learning_rate": 0.0017771429216622326, + "loss": 0.1387, + "step": 22747 + }, + { + "epoch": 0.197463563684343, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017771233159234966, + "loss": 0.0967, + "step": 22748 + }, + { + "epoch": 0.19747224416454717, + "grad_norm": 0.4609375, + "learning_rate": 0.0017771037094442685, + "loss": 0.1216, + "step": 22749 + }, + { + "epoch": 0.19748092464475134, + "grad_norm": 0.283203125, + "learning_rate": 0.0017770841022245697, + "loss": 0.0791, + "step": 22750 + }, + { + "epoch": 0.1974896051249555, + "grad_norm": 0.77734375, + "learning_rate": 0.0017770644942644222, + "loss": 0.1621, + "step": 22751 + }, + { + "epoch": 0.19749828560515967, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017770448855638473, + "loss": 0.1045, + "step": 22752 + }, + { + "epoch": 0.19750696608536383, + "grad_norm": 0.77734375, + "learning_rate": 0.0017770252761228661, + "loss": 0.1367, + "step": 22753 + }, + { + "epoch": 0.197515646565568, + "grad_norm": 0.70703125, + "learning_rate": 0.0017770056659415004, + "loss": 0.123, + "step": 22754 + }, + { + "epoch": 0.19752432704577216, + "grad_norm": 0.478515625, + "learning_rate": 0.0017769860550197712, + "loss": 0.106, + "step": 22755 + }, + { + "epoch": 0.19753300752597633, + "grad_norm": 0.48828125, + "learning_rate": 0.0017769664433577002, + "loss": 0.0815, + "step": 22756 + }, + { + "epoch": 0.1975416880061805, + "grad_norm": 0.255859375, + "learning_rate": 0.0017769468309553093, + "loss": 0.1279, + "step": 22757 + }, + { + "epoch": 0.19755036848638466, + "grad_norm": 0.2001953125, + "learning_rate": 0.001776927217812619, + "loss": 0.0923, + "step": 22758 + }, + { + "epoch": 0.19755904896658882, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017769076039296516, + "loss": 0.0977, + "step": 22759 + }, + { + "epoch": 0.197567729446793, + "grad_norm": 0.99609375, + "learning_rate": 0.001776887989306428, + "loss": 0.1875, + "step": 22760 + }, + { + "epoch": 0.19757640992699715, + "grad_norm": 0.162109375, + "learning_rate": 0.0017768683739429701, + "loss": 0.0908, + "step": 22761 + }, + { + "epoch": 0.19758509040720132, + "grad_norm": 0.6796875, + "learning_rate": 0.001776848757839299, + "loss": 0.1104, + "step": 22762 + }, + { + "epoch": 0.19759377088740548, + "grad_norm": 0.4765625, + "learning_rate": 0.0017768291409954362, + "loss": 0.1191, + "step": 22763 + }, + { + "epoch": 0.19760245136760965, + "grad_norm": 0.55859375, + "learning_rate": 0.0017768095234114033, + "loss": 0.1055, + "step": 22764 + }, + { + "epoch": 0.19761113184781381, + "grad_norm": 0.158203125, + "learning_rate": 0.0017767899050872217, + "loss": 0.1357, + "step": 22765 + }, + { + "epoch": 0.19761981232801798, + "grad_norm": 0.4453125, + "learning_rate": 0.0017767702860229126, + "loss": 0.0781, + "step": 22766 + }, + { + "epoch": 0.19762849280822214, + "grad_norm": 0.294921875, + "learning_rate": 0.0017767506662184978, + "loss": 0.1045, + "step": 22767 + }, + { + "epoch": 0.1976371732884263, + "grad_norm": 0.1826171875, + "learning_rate": 0.0017767310456739988, + "loss": 0.0874, + "step": 22768 + }, + { + "epoch": 0.19764585376863048, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017767114243894368, + "loss": 0.0693, + "step": 22769 + }, + { + "epoch": 0.19765453424883464, + "grad_norm": 0.39453125, + "learning_rate": 0.0017766918023648338, + "loss": 0.0957, + "step": 22770 + }, + { + "epoch": 0.1976632147290388, + "grad_norm": 1.0078125, + "learning_rate": 0.0017766721796002101, + "loss": 0.0967, + "step": 22771 + }, + { + "epoch": 0.19767189520924297, + "grad_norm": 0.96484375, + "learning_rate": 0.0017766525560955879, + "loss": 0.1006, + "step": 22772 + }, + { + "epoch": 0.19768057568944714, + "grad_norm": 0.55078125, + "learning_rate": 0.0017766329318509892, + "loss": 0.1162, + "step": 22773 + }, + { + "epoch": 0.1976892561696513, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017766133068664346, + "loss": 0.1113, + "step": 22774 + }, + { + "epoch": 0.19769793664985547, + "grad_norm": 0.388671875, + "learning_rate": 0.0017765936811419457, + "loss": 0.1279, + "step": 22775 + }, + { + "epoch": 0.19770661713005963, + "grad_norm": 0.26953125, + "learning_rate": 0.0017765740546775441, + "loss": 0.1182, + "step": 22776 + }, + { + "epoch": 0.1977152976102638, + "grad_norm": 0.46484375, + "learning_rate": 0.0017765544274732517, + "loss": 0.0796, + "step": 22777 + }, + { + "epoch": 0.19772397809046796, + "grad_norm": 0.28125, + "learning_rate": 0.001776534799529089, + "loss": 0.1084, + "step": 22778 + }, + { + "epoch": 0.19773265857067213, + "grad_norm": 0.15625, + "learning_rate": 0.0017765151708450783, + "loss": 0.1328, + "step": 22779 + }, + { + "epoch": 0.1977413390508763, + "grad_norm": 0.482421875, + "learning_rate": 0.0017764955414212409, + "loss": 0.1152, + "step": 22780 + }, + { + "epoch": 0.19775001953108046, + "grad_norm": 0.64453125, + "learning_rate": 0.0017764759112575978, + "loss": 0.1118, + "step": 22781 + }, + { + "epoch": 0.19775870001128462, + "grad_norm": 0.296875, + "learning_rate": 0.0017764562803541711, + "loss": 0.1523, + "step": 22782 + }, + { + "epoch": 0.1977673804914888, + "grad_norm": 0.416015625, + "learning_rate": 0.001776436648710982, + "loss": 0.1104, + "step": 22783 + }, + { + "epoch": 0.19777606097169295, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017764170163280517, + "loss": 0.0688, + "step": 22784 + }, + { + "epoch": 0.19778474145189712, + "grad_norm": 0.66796875, + "learning_rate": 0.0017763973832054022, + "loss": 0.1416, + "step": 22785 + }, + { + "epoch": 0.19779342193210128, + "grad_norm": 0.14453125, + "learning_rate": 0.0017763777493430542, + "loss": 0.1416, + "step": 22786 + }, + { + "epoch": 0.19780210241230545, + "grad_norm": 0.50390625, + "learning_rate": 0.0017763581147410298, + "loss": 0.1367, + "step": 22787 + }, + { + "epoch": 0.1978107828925096, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017763384793993509, + "loss": 0.1074, + "step": 22788 + }, + { + "epoch": 0.19781946337271378, + "grad_norm": 0.6640625, + "learning_rate": 0.0017763188433180383, + "loss": 0.1045, + "step": 22789 + }, + { + "epoch": 0.19782814385291794, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017762992064971129, + "loss": 0.1523, + "step": 22790 + }, + { + "epoch": 0.1978368243331221, + "grad_norm": 0.15625, + "learning_rate": 0.0017762795689365972, + "loss": 0.1045, + "step": 22791 + }, + { + "epoch": 0.19784550481332627, + "grad_norm": 0.251953125, + "learning_rate": 0.0017762599306365124, + "loss": 0.0884, + "step": 22792 + }, + { + "epoch": 0.19785418529353044, + "grad_norm": 0.23828125, + "learning_rate": 0.0017762402915968798, + "loss": 0.124, + "step": 22793 + }, + { + "epoch": 0.1978628657737346, + "grad_norm": 0.515625, + "learning_rate": 0.0017762206518177211, + "loss": 0.0859, + "step": 22794 + }, + { + "epoch": 0.19787154625393877, + "grad_norm": 0.58984375, + "learning_rate": 0.0017762010112990576, + "loss": 0.0996, + "step": 22795 + }, + { + "epoch": 0.19788022673414293, + "grad_norm": 0.41796875, + "learning_rate": 0.0017761813700409106, + "loss": 0.1279, + "step": 22796 + }, + { + "epoch": 0.1978889072143471, + "grad_norm": 0.376953125, + "learning_rate": 0.001776161728043302, + "loss": 0.1079, + "step": 22797 + }, + { + "epoch": 0.19789758769455126, + "grad_norm": 0.408203125, + "learning_rate": 0.001776142085306253, + "loss": 0.1299, + "step": 22798 + }, + { + "epoch": 0.19790626817475543, + "grad_norm": 0.54296875, + "learning_rate": 0.0017761224418297854, + "loss": 0.1289, + "step": 22799 + }, + { + "epoch": 0.1979149486549596, + "grad_norm": 0.416015625, + "learning_rate": 0.0017761027976139205, + "loss": 0.1592, + "step": 22800 + }, + { + "epoch": 0.19792362913516376, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017760831526586797, + "loss": 0.1104, + "step": 22801 + }, + { + "epoch": 0.19793230961536792, + "grad_norm": 0.154296875, + "learning_rate": 0.0017760635069640842, + "loss": 0.1426, + "step": 22802 + }, + { + "epoch": 0.1979409900955721, + "grad_norm": 0.90234375, + "learning_rate": 0.0017760438605301562, + "loss": 0.0869, + "step": 22803 + }, + { + "epoch": 0.19794967057577625, + "grad_norm": 0.3984375, + "learning_rate": 0.0017760242133569165, + "loss": 0.1021, + "step": 22804 + }, + { + "epoch": 0.19795835105598042, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017760045654443871, + "loss": 0.1201, + "step": 22805 + }, + { + "epoch": 0.19796703153618458, + "grad_norm": 0.9765625, + "learning_rate": 0.0017759849167925893, + "loss": 0.1797, + "step": 22806 + }, + { + "epoch": 0.19797571201638875, + "grad_norm": 0.609375, + "learning_rate": 0.001775965267401544, + "loss": 0.1016, + "step": 22807 + }, + { + "epoch": 0.19798439249659291, + "grad_norm": 0.2041015625, + "learning_rate": 0.001775945617271274, + "loss": 0.0801, + "step": 22808 + }, + { + "epoch": 0.19799307297679708, + "grad_norm": 0.07763671875, + "learning_rate": 0.0017759259664017995, + "loss": 0.1094, + "step": 22809 + }, + { + "epoch": 0.19800175345700124, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017759063147931428, + "loss": 0.0854, + "step": 22810 + }, + { + "epoch": 0.1980104339372054, + "grad_norm": 0.1455078125, + "learning_rate": 0.001775886662445325, + "loss": 0.1045, + "step": 22811 + }, + { + "epoch": 0.19801911441740958, + "grad_norm": 0.201171875, + "learning_rate": 0.0017758670093583674, + "loss": 0.0957, + "step": 22812 + }, + { + "epoch": 0.19802779489761374, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017758473555322925, + "loss": 0.1162, + "step": 22813 + }, + { + "epoch": 0.1980364753778179, + "grad_norm": 0.296875, + "learning_rate": 0.0017758277009671205, + "loss": 0.1055, + "step": 22814 + }, + { + "epoch": 0.19804515585802207, + "grad_norm": 0.578125, + "learning_rate": 0.0017758080456628735, + "loss": 0.1021, + "step": 22815 + }, + { + "epoch": 0.19805383633822624, + "grad_norm": 0.82421875, + "learning_rate": 0.0017757883896195733, + "loss": 0.1787, + "step": 22816 + }, + { + "epoch": 0.1980625168184304, + "grad_norm": 0.259765625, + "learning_rate": 0.0017757687328372408, + "loss": 0.0796, + "step": 22817 + }, + { + "epoch": 0.19807119729863457, + "grad_norm": 0.0859375, + "learning_rate": 0.0017757490753158981, + "loss": 0.1055, + "step": 22818 + }, + { + "epoch": 0.19807987777883873, + "grad_norm": 0.287109375, + "learning_rate": 0.001775729417055566, + "loss": 0.1641, + "step": 22819 + }, + { + "epoch": 0.1980885582590429, + "grad_norm": 0.474609375, + "learning_rate": 0.0017757097580562665, + "loss": 0.0996, + "step": 22820 + }, + { + "epoch": 0.19809723873924706, + "grad_norm": 0.2216796875, + "learning_rate": 0.001775690098318021, + "loss": 0.0938, + "step": 22821 + }, + { + "epoch": 0.19810591921945123, + "grad_norm": 0.8515625, + "learning_rate": 0.0017756704378408508, + "loss": 0.127, + "step": 22822 + }, + { + "epoch": 0.1981145996996554, + "grad_norm": 0.30859375, + "learning_rate": 0.0017756507766247777, + "loss": 0.1006, + "step": 22823 + }, + { + "epoch": 0.19812328017985956, + "grad_norm": 0.345703125, + "learning_rate": 0.001775631114669823, + "loss": 0.1025, + "step": 22824 + }, + { + "epoch": 0.19813196066006372, + "grad_norm": 0.451171875, + "learning_rate": 0.0017756114519760083, + "loss": 0.0981, + "step": 22825 + }, + { + "epoch": 0.1981406411402679, + "grad_norm": 0.1455078125, + "learning_rate": 0.001775591788543355, + "loss": 0.1094, + "step": 22826 + }, + { + "epoch": 0.19814932162047205, + "grad_norm": 0.15234375, + "learning_rate": 0.001775572124371885, + "loss": 0.1079, + "step": 22827 + }, + { + "epoch": 0.19815800210067622, + "grad_norm": 0.279296875, + "learning_rate": 0.0017755524594616193, + "loss": 0.125, + "step": 22828 + }, + { + "epoch": 0.19816668258088038, + "grad_norm": 0.8359375, + "learning_rate": 0.001775532793812579, + "loss": 0.5391, + "step": 22829 + }, + { + "epoch": 0.19817536306108455, + "grad_norm": 0.34375, + "learning_rate": 0.001775513127424787, + "loss": 0.1172, + "step": 22830 + }, + { + "epoch": 0.1981840435412887, + "grad_norm": 0.09228515625, + "learning_rate": 0.0017754934602982634, + "loss": 0.1006, + "step": 22831 + }, + { + "epoch": 0.19819272402149288, + "grad_norm": 0.2314453125, + "learning_rate": 0.0017754737924330307, + "loss": 0.106, + "step": 22832 + }, + { + "epoch": 0.19820140450169704, + "grad_norm": 0.306640625, + "learning_rate": 0.00177545412382911, + "loss": 0.1084, + "step": 22833 + }, + { + "epoch": 0.1982100849819012, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017754344544865226, + "loss": 0.106, + "step": 22834 + }, + { + "epoch": 0.19821876546210537, + "grad_norm": 0.51171875, + "learning_rate": 0.0017754147844052904, + "loss": 0.0879, + "step": 22835 + }, + { + "epoch": 0.19822744594230954, + "grad_norm": 0.2138671875, + "learning_rate": 0.001775395113585435, + "loss": 0.1191, + "step": 22836 + }, + { + "epoch": 0.1982361264225137, + "grad_norm": 0.625, + "learning_rate": 0.0017753754420269771, + "loss": 0.0884, + "step": 22837 + }, + { + "epoch": 0.19824480690271787, + "grad_norm": 0.1669921875, + "learning_rate": 0.001775355769729939, + "loss": 0.1025, + "step": 22838 + }, + { + "epoch": 0.19825348738292203, + "grad_norm": 0.4453125, + "learning_rate": 0.001775336096694342, + "loss": 0.1123, + "step": 22839 + }, + { + "epoch": 0.1982621678631262, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017753164229202074, + "loss": 0.0889, + "step": 22840 + }, + { + "epoch": 0.19827084834333036, + "grad_norm": 0.091796875, + "learning_rate": 0.001775296748407557, + "loss": 0.1348, + "step": 22841 + }, + { + "epoch": 0.19827952882353453, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017752770731564126, + "loss": 0.104, + "step": 22842 + }, + { + "epoch": 0.1982882093037387, + "grad_norm": 0.400390625, + "learning_rate": 0.0017752573971667948, + "loss": 0.1445, + "step": 22843 + }, + { + "epoch": 0.19829688978394286, + "grad_norm": 0.419921875, + "learning_rate": 0.001775237720438726, + "loss": 0.1475, + "step": 22844 + }, + { + "epoch": 0.19830557026414702, + "grad_norm": 0.294921875, + "learning_rate": 0.0017752180429722274, + "loss": 0.0742, + "step": 22845 + }, + { + "epoch": 0.1983142507443512, + "grad_norm": 0.298828125, + "learning_rate": 0.0017751983647673203, + "loss": 0.1006, + "step": 22846 + }, + { + "epoch": 0.19832293122455535, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017751786858240263, + "loss": 0.1357, + "step": 22847 + }, + { + "epoch": 0.19833161170475952, + "grad_norm": 0.283203125, + "learning_rate": 0.0017751590061423673, + "loss": 0.083, + "step": 22848 + }, + { + "epoch": 0.19834029218496368, + "grad_norm": 0.341796875, + "learning_rate": 0.0017751393257223645, + "loss": 0.0913, + "step": 22849 + }, + { + "epoch": 0.19834897266516785, + "grad_norm": 0.322265625, + "learning_rate": 0.0017751196445640394, + "loss": 0.1182, + "step": 22850 + }, + { + "epoch": 0.19835765314537201, + "grad_norm": 0.328125, + "learning_rate": 0.0017750999626674135, + "loss": 0.1562, + "step": 22851 + }, + { + "epoch": 0.19836633362557618, + "grad_norm": 0.1484375, + "learning_rate": 0.0017750802800325087, + "loss": 0.1025, + "step": 22852 + }, + { + "epoch": 0.19837501410578035, + "grad_norm": 0.62109375, + "learning_rate": 0.001775060596659346, + "loss": 0.1328, + "step": 22853 + }, + { + "epoch": 0.1983836945859845, + "grad_norm": 0.419921875, + "learning_rate": 0.0017750409125479474, + "loss": 0.0928, + "step": 22854 + }, + { + "epoch": 0.19839237506618868, + "grad_norm": 0.271484375, + "learning_rate": 0.001775021227698334, + "loss": 0.1035, + "step": 22855 + }, + { + "epoch": 0.1984010555463928, + "grad_norm": 0.267578125, + "learning_rate": 0.0017750015421105277, + "loss": 0.1206, + "step": 22856 + }, + { + "epoch": 0.19840973602659698, + "grad_norm": 0.150390625, + "learning_rate": 0.00177498185578455, + "loss": 0.1162, + "step": 22857 + }, + { + "epoch": 0.19841841650680114, + "grad_norm": 0.46875, + "learning_rate": 0.001774962168720422, + "loss": 0.1128, + "step": 22858 + }, + { + "epoch": 0.1984270969870053, + "grad_norm": 0.1162109375, + "learning_rate": 0.0017749424809181654, + "loss": 0.1279, + "step": 22859 + }, + { + "epoch": 0.19843577746720947, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017749227923778022, + "loss": 0.1182, + "step": 22860 + }, + { + "epoch": 0.19844445794741364, + "grad_norm": 0.29296875, + "learning_rate": 0.0017749031030993533, + "loss": 0.0874, + "step": 22861 + }, + { + "epoch": 0.1984531384276178, + "grad_norm": 0.30859375, + "learning_rate": 0.0017748834130828404, + "loss": 0.0737, + "step": 22862 + }, + { + "epoch": 0.19846181890782197, + "grad_norm": 0.359375, + "learning_rate": 0.0017748637223282853, + "loss": 0.1226, + "step": 22863 + }, + { + "epoch": 0.19847049938802613, + "grad_norm": 0.46875, + "learning_rate": 0.0017748440308357095, + "loss": 0.1099, + "step": 22864 + }, + { + "epoch": 0.1984791798682303, + "grad_norm": 0.091796875, + "learning_rate": 0.0017748243386051343, + "loss": 0.1348, + "step": 22865 + }, + { + "epoch": 0.19848786034843446, + "grad_norm": 0.359375, + "learning_rate": 0.0017748046456365814, + "loss": 0.1465, + "step": 22866 + }, + { + "epoch": 0.19849654082863863, + "grad_norm": 0.14453125, + "learning_rate": 0.0017747849519300723, + "loss": 0.1108, + "step": 22867 + }, + { + "epoch": 0.1985052213088428, + "grad_norm": 1.09375, + "learning_rate": 0.001774765257485628, + "loss": 0.1377, + "step": 22868 + }, + { + "epoch": 0.19851390178904696, + "grad_norm": 0.55078125, + "learning_rate": 0.0017747455623032713, + "loss": 0.1064, + "step": 22869 + }, + { + "epoch": 0.19852258226925112, + "grad_norm": 0.21484375, + "learning_rate": 0.0017747258663830228, + "loss": 0.1162, + "step": 22870 + }, + { + "epoch": 0.1985312627494553, + "grad_norm": 0.61328125, + "learning_rate": 0.001774706169724904, + "loss": 0.1055, + "step": 22871 + }, + { + "epoch": 0.19853994322965945, + "grad_norm": 0.6484375, + "learning_rate": 0.0017746864723289367, + "loss": 0.1074, + "step": 22872 + }, + { + "epoch": 0.19854862370986362, + "grad_norm": 0.349609375, + "learning_rate": 0.0017746667741951425, + "loss": 0.105, + "step": 22873 + }, + { + "epoch": 0.19855730419006778, + "grad_norm": 1.1171875, + "learning_rate": 0.0017746470753235428, + "loss": 0.1084, + "step": 22874 + }, + { + "epoch": 0.19856598467027195, + "grad_norm": 0.515625, + "learning_rate": 0.0017746273757141594, + "loss": 0.0771, + "step": 22875 + }, + { + "epoch": 0.19857466515047612, + "grad_norm": 0.29296875, + "learning_rate": 0.0017746076753670135, + "loss": 0.0854, + "step": 22876 + }, + { + "epoch": 0.19858334563068028, + "grad_norm": 0.53515625, + "learning_rate": 0.0017745879742821268, + "loss": 0.1016, + "step": 22877 + }, + { + "epoch": 0.19859202611088445, + "grad_norm": 0.68359375, + "learning_rate": 0.0017745682724595207, + "loss": 0.123, + "step": 22878 + }, + { + "epoch": 0.1986007065910886, + "grad_norm": 0.515625, + "learning_rate": 0.001774548569899217, + "loss": 0.1094, + "step": 22879 + }, + { + "epoch": 0.19860938707129278, + "grad_norm": 0.08056640625, + "learning_rate": 0.0017745288666012372, + "loss": 0.0933, + "step": 22880 + }, + { + "epoch": 0.19861806755149694, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017745091625656028, + "loss": 0.0708, + "step": 22881 + }, + { + "epoch": 0.1986267480317011, + "grad_norm": 0.1884765625, + "learning_rate": 0.001774489457792335, + "loss": 0.1104, + "step": 22882 + }, + { + "epoch": 0.19863542851190527, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017744697522814558, + "loss": 0.1504, + "step": 22883 + }, + { + "epoch": 0.19864410899210944, + "grad_norm": 0.59765625, + "learning_rate": 0.0017744500460329866, + "loss": 0.1045, + "step": 22884 + }, + { + "epoch": 0.1986527894723136, + "grad_norm": 0.1435546875, + "learning_rate": 0.001774430339046949, + "loss": 0.0898, + "step": 22885 + }, + { + "epoch": 0.19866146995251777, + "grad_norm": 0.197265625, + "learning_rate": 0.0017744106313233646, + "loss": 0.0767, + "step": 22886 + }, + { + "epoch": 0.19867015043272193, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017743909228622545, + "loss": 0.1436, + "step": 22887 + }, + { + "epoch": 0.1986788309129261, + "grad_norm": 0.1650390625, + "learning_rate": 0.001774371213663641, + "loss": 0.1021, + "step": 22888 + }, + { + "epoch": 0.19868751139313026, + "grad_norm": 0.3359375, + "learning_rate": 0.0017743515037275455, + "loss": 0.1074, + "step": 22889 + }, + { + "epoch": 0.19869619187333443, + "grad_norm": 0.251953125, + "learning_rate": 0.0017743317930539889, + "loss": 0.1328, + "step": 22890 + }, + { + "epoch": 0.1987048723535386, + "grad_norm": 0.447265625, + "learning_rate": 0.0017743120816429933, + "loss": 0.1562, + "step": 22891 + }, + { + "epoch": 0.19871355283374276, + "grad_norm": 0.44140625, + "learning_rate": 0.00177429236949458, + "loss": 0.0859, + "step": 22892 + }, + { + "epoch": 0.19872223331394692, + "grad_norm": 0.33203125, + "learning_rate": 0.0017742726566087707, + "loss": 0.0889, + "step": 22893 + }, + { + "epoch": 0.1987309137941511, + "grad_norm": 0.12255859375, + "learning_rate": 0.0017742529429855872, + "loss": 0.1475, + "step": 22894 + }, + { + "epoch": 0.19873959427435525, + "grad_norm": 0.84375, + "learning_rate": 0.0017742332286250506, + "loss": 0.1108, + "step": 22895 + }, + { + "epoch": 0.19874827475455942, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017742135135271828, + "loss": 0.1128, + "step": 22896 + }, + { + "epoch": 0.19875695523476358, + "grad_norm": 0.39453125, + "learning_rate": 0.0017741937976920052, + "loss": 0.1279, + "step": 22897 + }, + { + "epoch": 0.19876563571496775, + "grad_norm": 0.30859375, + "learning_rate": 0.0017741740811195396, + "loss": 0.1045, + "step": 22898 + }, + { + "epoch": 0.1987743161951719, + "grad_norm": 0.11083984375, + "learning_rate": 0.001774154363809807, + "loss": 0.1377, + "step": 22899 + }, + { + "epoch": 0.19878299667537608, + "grad_norm": 0.31640625, + "learning_rate": 0.0017741346457628296, + "loss": 0.1289, + "step": 22900 + }, + { + "epoch": 0.19879167715558024, + "grad_norm": 0.267578125, + "learning_rate": 0.0017741149269786284, + "loss": 0.0957, + "step": 22901 + }, + { + "epoch": 0.1988003576357844, + "grad_norm": 0.32421875, + "learning_rate": 0.0017740952074572255, + "loss": 0.0894, + "step": 22902 + }, + { + "epoch": 0.19880903811598857, + "grad_norm": 0.5546875, + "learning_rate": 0.0017740754871986418, + "loss": 0.1406, + "step": 22903 + }, + { + "epoch": 0.19881771859619274, + "grad_norm": 0.13671875, + "learning_rate": 0.0017740557662028998, + "loss": 0.0693, + "step": 22904 + }, + { + "epoch": 0.1988263990763969, + "grad_norm": 0.1025390625, + "learning_rate": 0.00177403604447002, + "loss": 0.1318, + "step": 22905 + }, + { + "epoch": 0.19883507955660107, + "grad_norm": 0.083984375, + "learning_rate": 0.001774016322000025, + "loss": 0.1177, + "step": 22906 + }, + { + "epoch": 0.19884376003680523, + "grad_norm": 0.306640625, + "learning_rate": 0.0017739965987929357, + "loss": 0.1011, + "step": 22907 + }, + { + "epoch": 0.1988524405170094, + "grad_norm": 0.291015625, + "learning_rate": 0.001773976874848774, + "loss": 0.3203, + "step": 22908 + }, + { + "epoch": 0.19886112099721356, + "grad_norm": 0.494140625, + "learning_rate": 0.0017739571501675612, + "loss": 0.0737, + "step": 22909 + }, + { + "epoch": 0.19886980147741773, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017739374247493188, + "loss": 0.1123, + "step": 22910 + }, + { + "epoch": 0.1988784819576219, + "grad_norm": 0.244140625, + "learning_rate": 0.0017739176985940687, + "loss": 0.1357, + "step": 22911 + }, + { + "epoch": 0.19888716243782606, + "grad_norm": 0.2275390625, + "learning_rate": 0.0017738979717018323, + "loss": 0.1094, + "step": 22912 + }, + { + "epoch": 0.19889584291803022, + "grad_norm": 0.265625, + "learning_rate": 0.001773878244072631, + "loss": 0.1089, + "step": 22913 + }, + { + "epoch": 0.1989045233982344, + "grad_norm": 0.201171875, + "learning_rate": 0.001773858515706487, + "loss": 0.063, + "step": 22914 + }, + { + "epoch": 0.19891320387843855, + "grad_norm": 1.3359375, + "learning_rate": 0.001773838786603421, + "loss": 0.1128, + "step": 22915 + }, + { + "epoch": 0.19892188435864272, + "grad_norm": 0.41796875, + "learning_rate": 0.0017738190567634555, + "loss": 0.1396, + "step": 22916 + }, + { + "epoch": 0.19893056483884689, + "grad_norm": 0.416015625, + "learning_rate": 0.0017737993261866115, + "loss": 0.0942, + "step": 22917 + }, + { + "epoch": 0.19893924531905105, + "grad_norm": 0.345703125, + "learning_rate": 0.0017737795948729105, + "loss": 0.0938, + "step": 22918 + }, + { + "epoch": 0.19894792579925522, + "grad_norm": 0.09765625, + "learning_rate": 0.0017737598628223743, + "loss": 0.1055, + "step": 22919 + }, + { + "epoch": 0.19895660627945938, + "grad_norm": 0.8671875, + "learning_rate": 0.0017737401300350244, + "loss": 0.1133, + "step": 22920 + }, + { + "epoch": 0.19896528675966355, + "grad_norm": 0.130859375, + "learning_rate": 0.0017737203965108826, + "loss": 0.127, + "step": 22921 + }, + { + "epoch": 0.1989739672398677, + "grad_norm": 0.1171875, + "learning_rate": 0.0017737006622499704, + "loss": 0.1152, + "step": 22922 + }, + { + "epoch": 0.19898264772007188, + "grad_norm": 0.65234375, + "learning_rate": 0.001773680927252309, + "loss": 0.1162, + "step": 22923 + }, + { + "epoch": 0.19899132820027604, + "grad_norm": 0.953125, + "learning_rate": 0.0017736611915179204, + "loss": 0.1079, + "step": 22924 + }, + { + "epoch": 0.1990000086804802, + "grad_norm": 0.62890625, + "learning_rate": 0.0017736414550468258, + "loss": 0.1064, + "step": 22925 + }, + { + "epoch": 0.19900868916068437, + "grad_norm": 0.96875, + "learning_rate": 0.0017736217178390473, + "loss": 0.1475, + "step": 22926 + }, + { + "epoch": 0.19901736964088854, + "grad_norm": 0.365234375, + "learning_rate": 0.0017736019798946059, + "loss": 0.1475, + "step": 22927 + }, + { + "epoch": 0.1990260501210927, + "grad_norm": 0.0986328125, + "learning_rate": 0.0017735822412135239, + "loss": 0.0952, + "step": 22928 + }, + { + "epoch": 0.19903473060129687, + "grad_norm": 0.458984375, + "learning_rate": 0.0017735625017958224, + "loss": 0.1055, + "step": 22929 + }, + { + "epoch": 0.19904341108150103, + "grad_norm": 0.349609375, + "learning_rate": 0.0017735427616415228, + "loss": 0.1016, + "step": 22930 + }, + { + "epoch": 0.1990520915617052, + "grad_norm": 0.75390625, + "learning_rate": 0.0017735230207506472, + "loss": 0.0728, + "step": 22931 + }, + { + "epoch": 0.19906077204190936, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017735032791232168, + "loss": 0.1201, + "step": 22932 + }, + { + "epoch": 0.19906945252211353, + "grad_norm": 0.6875, + "learning_rate": 0.0017734835367592533, + "loss": 0.1113, + "step": 22933 + }, + { + "epoch": 0.1990781330023177, + "grad_norm": 0.1064453125, + "learning_rate": 0.0017734637936587785, + "loss": 0.1216, + "step": 22934 + }, + { + "epoch": 0.19908681348252186, + "grad_norm": 0.23828125, + "learning_rate": 0.0017734440498218137, + "loss": 0.1084, + "step": 22935 + }, + { + "epoch": 0.19909549396272602, + "grad_norm": 0.625, + "learning_rate": 0.0017734243052483806, + "loss": 0.1426, + "step": 22936 + }, + { + "epoch": 0.1991041744429302, + "grad_norm": 0.2138671875, + "learning_rate": 0.0017734045599385006, + "loss": 0.1079, + "step": 22937 + }, + { + "epoch": 0.19911285492313435, + "grad_norm": 0.498046875, + "learning_rate": 0.001773384813892196, + "loss": 0.0908, + "step": 22938 + }, + { + "epoch": 0.19912153540333852, + "grad_norm": 0.419921875, + "learning_rate": 0.0017733650671094875, + "loss": 0.0908, + "step": 22939 + }, + { + "epoch": 0.19913021588354268, + "grad_norm": 0.6875, + "learning_rate": 0.001773345319590397, + "loss": 0.1211, + "step": 22940 + }, + { + "epoch": 0.19913889636374685, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017733255713349462, + "loss": 0.125, + "step": 22941 + }, + { + "epoch": 0.199147576843951, + "grad_norm": 0.34375, + "learning_rate": 0.001773305822343157, + "loss": 0.1416, + "step": 22942 + }, + { + "epoch": 0.19915625732415518, + "grad_norm": 0.11962890625, + "learning_rate": 0.001773286072615051, + "loss": 0.1128, + "step": 22943 + }, + { + "epoch": 0.19916493780435934, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017732663221506487, + "loss": 0.1318, + "step": 22944 + }, + { + "epoch": 0.1991736182845635, + "grad_norm": 0.3203125, + "learning_rate": 0.0017732465709499727, + "loss": 0.1064, + "step": 22945 + }, + { + "epoch": 0.19918229876476767, + "grad_norm": 0.07080078125, + "learning_rate": 0.0017732268190130444, + "loss": 0.083, + "step": 22946 + }, + { + "epoch": 0.19919097924497184, + "grad_norm": 0.9609375, + "learning_rate": 0.0017732070663398851, + "loss": 0.1826, + "step": 22947 + }, + { + "epoch": 0.199199659725176, + "grad_norm": 0.1708984375, + "learning_rate": 0.0017731873129305169, + "loss": 0.0718, + "step": 22948 + }, + { + "epoch": 0.19920834020538017, + "grad_norm": 0.416015625, + "learning_rate": 0.0017731675587849614, + "loss": 0.1099, + "step": 22949 + }, + { + "epoch": 0.19921702068558433, + "grad_norm": 0.58203125, + "learning_rate": 0.0017731478039032399, + "loss": 0.0986, + "step": 22950 + }, + { + "epoch": 0.1992257011657885, + "grad_norm": 0.71875, + "learning_rate": 0.0017731280482853739, + "loss": 0.1299, + "step": 22951 + }, + { + "epoch": 0.19923438164599266, + "grad_norm": 0.2060546875, + "learning_rate": 0.001773108291931385, + "loss": 0.0791, + "step": 22952 + }, + { + "epoch": 0.19924306212619683, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017730885348412953, + "loss": 0.0957, + "step": 22953 + }, + { + "epoch": 0.199251742606401, + "grad_norm": 0.51953125, + "learning_rate": 0.001773068777015126, + "loss": 0.1094, + "step": 22954 + }, + { + "epoch": 0.19926042308660516, + "grad_norm": 0.283203125, + "learning_rate": 0.001773049018452899, + "loss": 0.1299, + "step": 22955 + }, + { + "epoch": 0.19926910356680932, + "grad_norm": 0.6484375, + "learning_rate": 0.0017730292591546352, + "loss": 0.1914, + "step": 22956 + }, + { + "epoch": 0.1992777840470135, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017730094991203571, + "loss": 0.1475, + "step": 22957 + }, + { + "epoch": 0.19928646452721765, + "grad_norm": 0.640625, + "learning_rate": 0.001772989738350086, + "loss": 0.125, + "step": 22958 + }, + { + "epoch": 0.19929514500742182, + "grad_norm": 0.8125, + "learning_rate": 0.0017729699768438431, + "loss": 0.0947, + "step": 22959 + }, + { + "epoch": 0.19930382548762599, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017729502146016508, + "loss": 0.1206, + "step": 22960 + }, + { + "epoch": 0.19931250596783015, + "grad_norm": 0.08544921875, + "learning_rate": 0.0017729304516235298, + "loss": 0.1123, + "step": 22961 + }, + { + "epoch": 0.19932118644803432, + "grad_norm": 0.375, + "learning_rate": 0.0017729106879095023, + "loss": 0.1133, + "step": 22962 + }, + { + "epoch": 0.19932986692823848, + "grad_norm": 0.453125, + "learning_rate": 0.0017728909234595899, + "loss": 0.1504, + "step": 22963 + }, + { + "epoch": 0.19933854740844265, + "grad_norm": 0.1259765625, + "learning_rate": 0.001772871158273814, + "loss": 0.0762, + "step": 22964 + }, + { + "epoch": 0.1993472278886468, + "grad_norm": 0.400390625, + "learning_rate": 0.0017728513923521963, + "loss": 0.0806, + "step": 22965 + }, + { + "epoch": 0.19935590836885098, + "grad_norm": 0.13671875, + "learning_rate": 0.0017728316256947583, + "loss": 0.1045, + "step": 22966 + }, + { + "epoch": 0.19936458884905514, + "grad_norm": 0.115234375, + "learning_rate": 0.001772811858301522, + "loss": 0.1221, + "step": 22967 + }, + { + "epoch": 0.1993732693292593, + "grad_norm": 0.33984375, + "learning_rate": 0.0017727920901725086, + "loss": 0.1045, + "step": 22968 + }, + { + "epoch": 0.19938194980946347, + "grad_norm": 0.353515625, + "learning_rate": 0.0017727723213077401, + "loss": 0.0977, + "step": 22969 + }, + { + "epoch": 0.19939063028966764, + "grad_norm": 0.451171875, + "learning_rate": 0.0017727525517072377, + "loss": 0.0918, + "step": 22970 + }, + { + "epoch": 0.1993993107698718, + "grad_norm": 0.6640625, + "learning_rate": 0.0017727327813710232, + "loss": 0.0981, + "step": 22971 + }, + { + "epoch": 0.19940799125007597, + "grad_norm": 0.97265625, + "learning_rate": 0.0017727130102991185, + "loss": 0.1641, + "step": 22972 + }, + { + "epoch": 0.19941667173028013, + "grad_norm": 0.396484375, + "learning_rate": 0.0017726932384915445, + "loss": 0.0996, + "step": 22973 + }, + { + "epoch": 0.1994253522104843, + "grad_norm": 0.58203125, + "learning_rate": 0.0017726734659483235, + "loss": 0.1182, + "step": 22974 + }, + { + "epoch": 0.19943403269068846, + "grad_norm": 0.22265625, + "learning_rate": 0.0017726536926694772, + "loss": 0.0928, + "step": 22975 + }, + { + "epoch": 0.19944271317089263, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017726339186550266, + "loss": 0.1533, + "step": 22976 + }, + { + "epoch": 0.1994513936510968, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017726141439049937, + "loss": 0.1021, + "step": 22977 + }, + { + "epoch": 0.19946007413130096, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017725943684194002, + "loss": 0.1016, + "step": 22978 + }, + { + "epoch": 0.1994687546115051, + "grad_norm": 0.314453125, + "learning_rate": 0.0017725745921982675, + "loss": 0.1187, + "step": 22979 + }, + { + "epoch": 0.19947743509170926, + "grad_norm": 0.248046875, + "learning_rate": 0.0017725548152416172, + "loss": 0.085, + "step": 22980 + }, + { + "epoch": 0.19948611557191342, + "grad_norm": 0.3671875, + "learning_rate": 0.0017725350375494711, + "loss": 0.0923, + "step": 22981 + }, + { + "epoch": 0.1994947960521176, + "grad_norm": 0.75, + "learning_rate": 0.001772515259121851, + "loss": 0.1436, + "step": 22982 + }, + { + "epoch": 0.19950347653232176, + "grad_norm": 0.65625, + "learning_rate": 0.0017724954799587783, + "loss": 0.1211, + "step": 22983 + }, + { + "epoch": 0.19951215701252592, + "grad_norm": 0.326171875, + "learning_rate": 0.0017724757000602744, + "loss": 0.1396, + "step": 22984 + }, + { + "epoch": 0.19952083749273009, + "grad_norm": 0.671875, + "learning_rate": 0.0017724559194263615, + "loss": 0.0972, + "step": 22985 + }, + { + "epoch": 0.19952951797293425, + "grad_norm": 0.302734375, + "learning_rate": 0.0017724361380570606, + "loss": 0.0762, + "step": 22986 + }, + { + "epoch": 0.19953819845313842, + "grad_norm": 0.59375, + "learning_rate": 0.0017724163559523938, + "loss": 0.127, + "step": 22987 + }, + { + "epoch": 0.19954687893334258, + "grad_norm": 0.6640625, + "learning_rate": 0.0017723965731123825, + "loss": 0.1133, + "step": 22988 + }, + { + "epoch": 0.19955555941354675, + "grad_norm": 0.361328125, + "learning_rate": 0.0017723767895370486, + "loss": 0.1143, + "step": 22989 + }, + { + "epoch": 0.1995642398937509, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017723570052264133, + "loss": 0.1162, + "step": 22990 + }, + { + "epoch": 0.19957292037395508, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017723372201804987, + "loss": 0.1104, + "step": 22991 + }, + { + "epoch": 0.19958160085415924, + "grad_norm": 0.56640625, + "learning_rate": 0.0017723174343993262, + "loss": 0.0938, + "step": 22992 + }, + { + "epoch": 0.1995902813343634, + "grad_norm": 0.234375, + "learning_rate": 0.0017722976478829172, + "loss": 0.1016, + "step": 22993 + }, + { + "epoch": 0.19959896181456757, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017722778606312937, + "loss": 0.1885, + "step": 22994 + }, + { + "epoch": 0.19960764229477174, + "grad_norm": 0.9609375, + "learning_rate": 0.0017722580726444773, + "loss": 0.1133, + "step": 22995 + }, + { + "epoch": 0.1996163227749759, + "grad_norm": 0.5625, + "learning_rate": 0.0017722382839224895, + "loss": 0.1079, + "step": 22996 + }, + { + "epoch": 0.19962500325518007, + "grad_norm": 0.330078125, + "learning_rate": 0.0017722184944653522, + "loss": 0.0981, + "step": 22997 + }, + { + "epoch": 0.19963368373538423, + "grad_norm": 0.416015625, + "learning_rate": 0.0017721987042730867, + "loss": 0.1309, + "step": 22998 + }, + { + "epoch": 0.1996423642155884, + "grad_norm": 0.1435546875, + "learning_rate": 0.001772178913345715, + "loss": 0.1177, + "step": 22999 + }, + { + "epoch": 0.19965104469579256, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017721591216832582, + "loss": 0.1172, + "step": 23000 + }, + { + "epoch": 0.19965972517599673, + "grad_norm": 0.12255859375, + "learning_rate": 0.0017721393292857387, + "loss": 0.1152, + "step": 23001 + }, + { + "epoch": 0.1996684056562009, + "grad_norm": 0.84375, + "learning_rate": 0.0017721195361531771, + "loss": 0.1406, + "step": 23002 + }, + { + "epoch": 0.19967708613640506, + "grad_norm": 0.19921875, + "learning_rate": 0.0017720997422855961, + "loss": 0.1011, + "step": 23003 + }, + { + "epoch": 0.19968576661660922, + "grad_norm": 0.32421875, + "learning_rate": 0.0017720799476830169, + "loss": 0.1299, + "step": 23004 + }, + { + "epoch": 0.1996944470968134, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017720601523454613, + "loss": 0.0962, + "step": 23005 + }, + { + "epoch": 0.19970312757701755, + "grad_norm": 0.2265625, + "learning_rate": 0.0017720403562729505, + "loss": 0.0879, + "step": 23006 + }, + { + "epoch": 0.19971180805722172, + "grad_norm": 0.248046875, + "learning_rate": 0.0017720205594655067, + "loss": 0.085, + "step": 23007 + }, + { + "epoch": 0.19972048853742588, + "grad_norm": 0.123046875, + "learning_rate": 0.0017720007619231512, + "loss": 0.0889, + "step": 23008 + }, + { + "epoch": 0.19972916901763005, + "grad_norm": 0.310546875, + "learning_rate": 0.001771980963645906, + "loss": 0.1074, + "step": 23009 + }, + { + "epoch": 0.1997378494978342, + "grad_norm": 0.130859375, + "learning_rate": 0.0017719611646337923, + "loss": 0.123, + "step": 23010 + }, + { + "epoch": 0.19974652997803838, + "grad_norm": 2.34375, + "learning_rate": 0.0017719413648868319, + "loss": 0.6016, + "step": 23011 + }, + { + "epoch": 0.19975521045824254, + "grad_norm": 0.06982421875, + "learning_rate": 0.0017719215644050467, + "loss": 0.0918, + "step": 23012 + }, + { + "epoch": 0.1997638909384467, + "grad_norm": 0.279296875, + "learning_rate": 0.001771901763188458, + "loss": 0.1167, + "step": 23013 + }, + { + "epoch": 0.19977257141865087, + "grad_norm": 0.404296875, + "learning_rate": 0.0017718819612370876, + "loss": 0.1387, + "step": 23014 + }, + { + "epoch": 0.19978125189885504, + "grad_norm": 0.171875, + "learning_rate": 0.0017718621585509574, + "loss": 0.1104, + "step": 23015 + }, + { + "epoch": 0.1997899323790592, + "grad_norm": 0.52734375, + "learning_rate": 0.0017718423551300886, + "loss": 0.0986, + "step": 23016 + }, + { + "epoch": 0.19979861285926337, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017718225509745032, + "loss": 0.1475, + "step": 23017 + }, + { + "epoch": 0.19980729333946753, + "grad_norm": 0.44921875, + "learning_rate": 0.0017718027460842227, + "loss": 0.1035, + "step": 23018 + }, + { + "epoch": 0.1998159738196717, + "grad_norm": 0.1826171875, + "learning_rate": 0.001771782940459269, + "loss": 0.0947, + "step": 23019 + }, + { + "epoch": 0.19982465429987586, + "grad_norm": 0.142578125, + "learning_rate": 0.0017717631340996634, + "loss": 0.1064, + "step": 23020 + }, + { + "epoch": 0.19983333478008003, + "grad_norm": 0.21484375, + "learning_rate": 0.0017717433270054278, + "loss": 0.0835, + "step": 23021 + }, + { + "epoch": 0.1998420152602842, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017717235191765839, + "loss": 0.1162, + "step": 23022 + }, + { + "epoch": 0.19985069574048836, + "grad_norm": 0.369140625, + "learning_rate": 0.001771703710613153, + "loss": 0.1289, + "step": 23023 + }, + { + "epoch": 0.19985937622069253, + "grad_norm": 0.4453125, + "learning_rate": 0.001771683901315157, + "loss": 0.1235, + "step": 23024 + }, + { + "epoch": 0.1998680567008967, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017716640912826178, + "loss": 0.0908, + "step": 23025 + }, + { + "epoch": 0.19987673718110086, + "grad_norm": 0.107421875, + "learning_rate": 0.001771644280515557, + "loss": 0.0669, + "step": 23026 + }, + { + "epoch": 0.19988541766130502, + "grad_norm": 0.361328125, + "learning_rate": 0.0017716244690139955, + "loss": 0.0933, + "step": 23027 + }, + { + "epoch": 0.19989409814150919, + "grad_norm": 0.07861328125, + "learning_rate": 0.0017716046567779562, + "loss": 0.0947, + "step": 23028 + }, + { + "epoch": 0.19990277862171335, + "grad_norm": 0.41796875, + "learning_rate": 0.0017715848438074598, + "loss": 0.0762, + "step": 23029 + }, + { + "epoch": 0.19991145910191752, + "grad_norm": 0.466796875, + "learning_rate": 0.0017715650301025284, + "loss": 0.1152, + "step": 23030 + }, + { + "epoch": 0.19992013958212168, + "grad_norm": 0.28125, + "learning_rate": 0.0017715452156631835, + "loss": 0.1592, + "step": 23031 + }, + { + "epoch": 0.19992882006232585, + "grad_norm": 0.3828125, + "learning_rate": 0.0017715254004894469, + "loss": 0.1011, + "step": 23032 + }, + { + "epoch": 0.19993750054253, + "grad_norm": 0.0908203125, + "learning_rate": 0.0017715055845813402, + "loss": 0.0664, + "step": 23033 + }, + { + "epoch": 0.19994618102273418, + "grad_norm": 0.14453125, + "learning_rate": 0.0017714857679388852, + "loss": 0.0972, + "step": 23034 + }, + { + "epoch": 0.19995486150293834, + "grad_norm": 0.10888671875, + "learning_rate": 0.001771465950562103, + "loss": 0.084, + "step": 23035 + }, + { + "epoch": 0.1999635419831425, + "grad_norm": 0.6953125, + "learning_rate": 0.001771446132451016, + "loss": 0.1289, + "step": 23036 + }, + { + "epoch": 0.19997222246334667, + "grad_norm": 0.359375, + "learning_rate": 0.0017714263136056458, + "loss": 0.1016, + "step": 23037 + }, + { + "epoch": 0.19998090294355084, + "grad_norm": 0.2138671875, + "learning_rate": 0.0017714064940260138, + "loss": 0.1641, + "step": 23038 + }, + { + "epoch": 0.199989583423755, + "grad_norm": 0.2138671875, + "learning_rate": 0.0017713866737121416, + "loss": 0.1602, + "step": 23039 + }, + { + "epoch": 0.19999826390395917, + "grad_norm": 0.3671875, + "learning_rate": 0.0017713668526640513, + "loss": 0.0801, + "step": 23040 + }, + { + "epoch": 0.20000694438416333, + "grad_norm": 0.1396484375, + "learning_rate": 0.001771347030881764, + "loss": 0.126, + "step": 23041 + }, + { + "epoch": 0.2000156248643675, + "grad_norm": 0.341796875, + "learning_rate": 0.0017713272083653018, + "loss": 0.1396, + "step": 23042 + }, + { + "epoch": 0.20002430534457166, + "grad_norm": 0.41796875, + "learning_rate": 0.0017713073851146862, + "loss": 0.1182, + "step": 23043 + }, + { + "epoch": 0.20003298582477583, + "grad_norm": 0.55078125, + "learning_rate": 0.001771287561129939, + "loss": 0.084, + "step": 23044 + }, + { + "epoch": 0.20004166630498, + "grad_norm": 0.146484375, + "learning_rate": 0.0017712677364110816, + "loss": 0.0996, + "step": 23045 + }, + { + "epoch": 0.20005034678518416, + "grad_norm": 0.123046875, + "learning_rate": 0.0017712479109581358, + "loss": 0.1191, + "step": 23046 + }, + { + "epoch": 0.20005902726538832, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017712280847711238, + "loss": 0.1045, + "step": 23047 + }, + { + "epoch": 0.2000677077455925, + "grad_norm": 0.267578125, + "learning_rate": 0.0017712082578500668, + "loss": 0.0898, + "step": 23048 + }, + { + "epoch": 0.20007638822579665, + "grad_norm": 0.419921875, + "learning_rate": 0.0017711884301949866, + "loss": 0.0928, + "step": 23049 + }, + { + "epoch": 0.20008506870600082, + "grad_norm": 0.455078125, + "learning_rate": 0.0017711686018059044, + "loss": 0.1357, + "step": 23050 + }, + { + "epoch": 0.20009374918620498, + "grad_norm": 0.330078125, + "learning_rate": 0.0017711487726828426, + "loss": 0.0957, + "step": 23051 + }, + { + "epoch": 0.20010242966640915, + "grad_norm": 0.267578125, + "learning_rate": 0.0017711289428258225, + "loss": 0.1621, + "step": 23052 + }, + { + "epoch": 0.2001111101466133, + "grad_norm": 0.2119140625, + "learning_rate": 0.001771109112234866, + "loss": 0.0869, + "step": 23053 + }, + { + "epoch": 0.20011979062681748, + "grad_norm": 1.390625, + "learning_rate": 0.0017710892809099945, + "loss": 0.1387, + "step": 23054 + }, + { + "epoch": 0.20012847110702164, + "grad_norm": 0.15234375, + "learning_rate": 0.0017710694488512295, + "loss": 0.0908, + "step": 23055 + }, + { + "epoch": 0.2001371515872258, + "grad_norm": 0.52734375, + "learning_rate": 0.0017710496160585937, + "loss": 0.1387, + "step": 23056 + }, + { + "epoch": 0.20014583206742997, + "grad_norm": 0.388671875, + "learning_rate": 0.001771029782532108, + "loss": 0.1738, + "step": 23057 + }, + { + "epoch": 0.20015451254763414, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017710099482717938, + "loss": 0.1216, + "step": 23058 + }, + { + "epoch": 0.2001631930278383, + "grad_norm": 1.640625, + "learning_rate": 0.0017709901132776735, + "loss": 0.1289, + "step": 23059 + }, + { + "epoch": 0.20017187350804247, + "grad_norm": 0.5390625, + "learning_rate": 0.0017709702775497685, + "loss": 0.1484, + "step": 23060 + }, + { + "epoch": 0.20018055398824663, + "grad_norm": 0.8046875, + "learning_rate": 0.0017709504410881004, + "loss": 0.1289, + "step": 23061 + }, + { + "epoch": 0.2001892344684508, + "grad_norm": 0.5078125, + "learning_rate": 0.0017709306038926908, + "loss": 0.1191, + "step": 23062 + }, + { + "epoch": 0.20019791494865496, + "grad_norm": 0.34765625, + "learning_rate": 0.001770910765963562, + "loss": 0.125, + "step": 23063 + }, + { + "epoch": 0.20020659542885913, + "grad_norm": 0.318359375, + "learning_rate": 0.001770890927300735, + "loss": 0.0859, + "step": 23064 + }, + { + "epoch": 0.2002152759090633, + "grad_norm": 0.09765625, + "learning_rate": 0.0017708710879042319, + "loss": 0.124, + "step": 23065 + }, + { + "epoch": 0.20022395638926746, + "grad_norm": 0.337890625, + "learning_rate": 0.0017708512477740742, + "loss": 0.1377, + "step": 23066 + }, + { + "epoch": 0.20023263686947163, + "grad_norm": 0.421875, + "learning_rate": 0.0017708314069102835, + "loss": 0.0962, + "step": 23067 + }, + { + "epoch": 0.2002413173496758, + "grad_norm": 0.51171875, + "learning_rate": 0.0017708115653128817, + "loss": 0.1006, + "step": 23068 + }, + { + "epoch": 0.20024999782987996, + "grad_norm": 0.66015625, + "learning_rate": 0.0017707917229818908, + "loss": 0.127, + "step": 23069 + }, + { + "epoch": 0.20025867831008412, + "grad_norm": 0.1533203125, + "learning_rate": 0.001770771879917332, + "loss": 0.1094, + "step": 23070 + }, + { + "epoch": 0.20026735879028829, + "grad_norm": 0.478515625, + "learning_rate": 0.0017707520361192268, + "loss": 0.1055, + "step": 23071 + }, + { + "epoch": 0.20027603927049245, + "grad_norm": 1.0390625, + "learning_rate": 0.0017707321915875973, + "loss": 0.1387, + "step": 23072 + }, + { + "epoch": 0.20028471975069662, + "grad_norm": 0.376953125, + "learning_rate": 0.0017707123463224655, + "loss": 0.1191, + "step": 23073 + }, + { + "epoch": 0.20029340023090078, + "grad_norm": 0.515625, + "learning_rate": 0.0017706925003238526, + "loss": 0.1064, + "step": 23074 + }, + { + "epoch": 0.20030208071110495, + "grad_norm": 0.345703125, + "learning_rate": 0.0017706726535917804, + "loss": 0.0933, + "step": 23075 + }, + { + "epoch": 0.2003107611913091, + "grad_norm": 0.71484375, + "learning_rate": 0.0017706528061262707, + "loss": 0.0898, + "step": 23076 + }, + { + "epoch": 0.20031944167151328, + "grad_norm": 0.126953125, + "learning_rate": 0.0017706329579273452, + "loss": 0.1582, + "step": 23077 + }, + { + "epoch": 0.20032812215171744, + "grad_norm": 0.78515625, + "learning_rate": 0.0017706131089950256, + "loss": 0.1182, + "step": 23078 + }, + { + "epoch": 0.2003368026319216, + "grad_norm": 0.44921875, + "learning_rate": 0.0017705932593293334, + "loss": 0.0874, + "step": 23079 + }, + { + "epoch": 0.20034548311212577, + "grad_norm": 1.0234375, + "learning_rate": 0.0017705734089302908, + "loss": 0.125, + "step": 23080 + }, + { + "epoch": 0.20035416359232994, + "grad_norm": 0.361328125, + "learning_rate": 0.0017705535577979188, + "loss": 0.123, + "step": 23081 + }, + { + "epoch": 0.2003628440725341, + "grad_norm": 0.408203125, + "learning_rate": 0.0017705337059322397, + "loss": 0.0879, + "step": 23082 + }, + { + "epoch": 0.20037152455273827, + "grad_norm": 0.390625, + "learning_rate": 0.0017705138533332752, + "loss": 0.1006, + "step": 23083 + }, + { + "epoch": 0.20038020503294243, + "grad_norm": 0.2734375, + "learning_rate": 0.0017704940000010465, + "loss": 0.1113, + "step": 23084 + }, + { + "epoch": 0.2003888855131466, + "grad_norm": 0.2197265625, + "learning_rate": 0.001770474145935576, + "loss": 0.0947, + "step": 23085 + }, + { + "epoch": 0.20039756599335076, + "grad_norm": 0.173828125, + "learning_rate": 0.0017704542911368849, + "loss": 0.0859, + "step": 23086 + }, + { + "epoch": 0.20040624647355493, + "grad_norm": 0.71484375, + "learning_rate": 0.001770434435604995, + "loss": 0.1309, + "step": 23087 + }, + { + "epoch": 0.2004149269537591, + "grad_norm": 0.439453125, + "learning_rate": 0.001770414579339928, + "loss": 0.1777, + "step": 23088 + }, + { + "epoch": 0.20042360743396326, + "grad_norm": 0.271484375, + "learning_rate": 0.0017703947223417056, + "loss": 0.123, + "step": 23089 + }, + { + "epoch": 0.20043228791416742, + "grad_norm": 0.2451171875, + "learning_rate": 0.00177037486461035, + "loss": 0.1094, + "step": 23090 + }, + { + "epoch": 0.2004409683943716, + "grad_norm": 0.490234375, + "learning_rate": 0.0017703550061458821, + "loss": 0.1006, + "step": 23091 + }, + { + "epoch": 0.20044964887457575, + "grad_norm": 0.31640625, + "learning_rate": 0.0017703351469483244, + "loss": 0.0938, + "step": 23092 + }, + { + "epoch": 0.20045832935477992, + "grad_norm": 0.265625, + "learning_rate": 0.001770315287017698, + "loss": 0.1152, + "step": 23093 + }, + { + "epoch": 0.20046700983498408, + "grad_norm": 0.65234375, + "learning_rate": 0.0017702954263540249, + "loss": 0.1309, + "step": 23094 + }, + { + "epoch": 0.20047569031518825, + "grad_norm": 0.4921875, + "learning_rate": 0.0017702755649573268, + "loss": 0.1562, + "step": 23095 + }, + { + "epoch": 0.2004843707953924, + "grad_norm": 0.5078125, + "learning_rate": 0.0017702557028276254, + "loss": 0.1221, + "step": 23096 + }, + { + "epoch": 0.20049305127559658, + "grad_norm": 0.484375, + "learning_rate": 0.0017702358399649426, + "loss": 0.0781, + "step": 23097 + }, + { + "epoch": 0.20050173175580074, + "grad_norm": 1.4296875, + "learning_rate": 0.0017702159763692998, + "loss": 0.1426, + "step": 23098 + }, + { + "epoch": 0.2005104122360049, + "grad_norm": 0.412109375, + "learning_rate": 0.001770196112040719, + "loss": 0.1123, + "step": 23099 + }, + { + "epoch": 0.20051909271620907, + "grad_norm": 0.48046875, + "learning_rate": 0.001770176246979222, + "loss": 0.0825, + "step": 23100 + }, + { + "epoch": 0.20052777319641324, + "grad_norm": 0.07666015625, + "learning_rate": 0.00177015638118483, + "loss": 0.1099, + "step": 23101 + }, + { + "epoch": 0.20053645367661738, + "grad_norm": 0.19921875, + "learning_rate": 0.0017701365146575649, + "loss": 0.1104, + "step": 23102 + }, + { + "epoch": 0.20054513415682154, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017701166473974488, + "loss": 0.1309, + "step": 23103 + }, + { + "epoch": 0.2005538146370257, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017700967794045033, + "loss": 0.0942, + "step": 23104 + }, + { + "epoch": 0.20056249511722987, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017700769106787498, + "loss": 0.1016, + "step": 23105 + }, + { + "epoch": 0.20057117559743404, + "grad_norm": 0.197265625, + "learning_rate": 0.0017700570412202104, + "loss": 0.1182, + "step": 23106 + }, + { + "epoch": 0.2005798560776382, + "grad_norm": 0.8828125, + "learning_rate": 0.0017700371710289064, + "loss": 0.1367, + "step": 23107 + }, + { + "epoch": 0.20058853655784237, + "grad_norm": 0.443359375, + "learning_rate": 0.0017700173001048603, + "loss": 0.1367, + "step": 23108 + }, + { + "epoch": 0.20059721703804653, + "grad_norm": 0.51171875, + "learning_rate": 0.0017699974284480931, + "loss": 0.0913, + "step": 23109 + }, + { + "epoch": 0.2006058975182507, + "grad_norm": 0.2158203125, + "learning_rate": 0.001769977556058627, + "loss": 0.1357, + "step": 23110 + }, + { + "epoch": 0.20061457799845486, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017699576829364831, + "loss": 0.083, + "step": 23111 + }, + { + "epoch": 0.20062325847865903, + "grad_norm": 1.0703125, + "learning_rate": 0.0017699378090816839, + "loss": 0.1094, + "step": 23112 + }, + { + "epoch": 0.2006319389588632, + "grad_norm": 0.7265625, + "learning_rate": 0.0017699179344942506, + "loss": 0.0698, + "step": 23113 + }, + { + "epoch": 0.20064061943906736, + "grad_norm": 0.2578125, + "learning_rate": 0.001769898059174205, + "loss": 0.0894, + "step": 23114 + }, + { + "epoch": 0.20064929991927152, + "grad_norm": 0.84375, + "learning_rate": 0.001769878183121569, + "loss": 0.0859, + "step": 23115 + }, + { + "epoch": 0.2006579803994757, + "grad_norm": 0.142578125, + "learning_rate": 0.0017698583063363647, + "loss": 0.1064, + "step": 23116 + }, + { + "epoch": 0.20066666087967985, + "grad_norm": 0.55078125, + "learning_rate": 0.001769838428818613, + "loss": 0.1133, + "step": 23117 + }, + { + "epoch": 0.20067534135988402, + "grad_norm": 0.53515625, + "learning_rate": 0.0017698185505683362, + "loss": 0.104, + "step": 23118 + }, + { + "epoch": 0.20068402184008818, + "grad_norm": 0.318359375, + "learning_rate": 0.0017697986715855558, + "loss": 0.1064, + "step": 23119 + }, + { + "epoch": 0.20069270232029235, + "grad_norm": 1.1484375, + "learning_rate": 0.0017697787918702936, + "loss": 0.1611, + "step": 23120 + }, + { + "epoch": 0.20070138280049651, + "grad_norm": 0.90625, + "learning_rate": 0.0017697589114225715, + "loss": 0.0903, + "step": 23121 + }, + { + "epoch": 0.20071006328070068, + "grad_norm": 0.2138671875, + "learning_rate": 0.001769739030242411, + "loss": 0.1055, + "step": 23122 + }, + { + "epoch": 0.20071874376090484, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017697191483298342, + "loss": 0.0811, + "step": 23123 + }, + { + "epoch": 0.200727424241109, + "grad_norm": 0.67578125, + "learning_rate": 0.0017696992656848626, + "loss": 0.1445, + "step": 23124 + }, + { + "epoch": 0.20073610472131317, + "grad_norm": 0.7421875, + "learning_rate": 0.0017696793823075176, + "loss": 0.1279, + "step": 23125 + }, + { + "epoch": 0.20074478520151734, + "grad_norm": 0.16796875, + "learning_rate": 0.0017696594981978216, + "loss": 0.0972, + "step": 23126 + }, + { + "epoch": 0.2007534656817215, + "grad_norm": 0.09814453125, + "learning_rate": 0.001769639613355796, + "loss": 0.0884, + "step": 23127 + }, + { + "epoch": 0.20076214616192567, + "grad_norm": 0.51171875, + "learning_rate": 0.0017696197277814623, + "loss": 0.127, + "step": 23128 + }, + { + "epoch": 0.20077082664212983, + "grad_norm": 0.216796875, + "learning_rate": 0.0017695998414748429, + "loss": 0.0977, + "step": 23129 + }, + { + "epoch": 0.200779507122334, + "grad_norm": 0.173828125, + "learning_rate": 0.0017695799544359587, + "loss": 0.1279, + "step": 23130 + }, + { + "epoch": 0.20078818760253817, + "grad_norm": 0.86328125, + "learning_rate": 0.0017695600666648324, + "loss": 0.1318, + "step": 23131 + }, + { + "epoch": 0.20079686808274233, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017695401781614851, + "loss": 0.1162, + "step": 23132 + }, + { + "epoch": 0.2008055485629465, + "grad_norm": 0.232421875, + "learning_rate": 0.001769520288925939, + "loss": 0.1309, + "step": 23133 + }, + { + "epoch": 0.20081422904315066, + "grad_norm": 0.53515625, + "learning_rate": 0.001769500398958215, + "loss": 0.0962, + "step": 23134 + }, + { + "epoch": 0.20082290952335483, + "grad_norm": 0.294921875, + "learning_rate": 0.0017694805082583355, + "loss": 0.165, + "step": 23135 + }, + { + "epoch": 0.200831590003559, + "grad_norm": 0.1708984375, + "learning_rate": 0.0017694606168263225, + "loss": 0.0908, + "step": 23136 + }, + { + "epoch": 0.20084027048376316, + "grad_norm": 0.443359375, + "learning_rate": 0.0017694407246621976, + "loss": 0.0977, + "step": 23137 + }, + { + "epoch": 0.20084895096396732, + "grad_norm": 0.78515625, + "learning_rate": 0.001769420831765982, + "loss": 0.1182, + "step": 23138 + }, + { + "epoch": 0.20085763144417149, + "grad_norm": 0.2236328125, + "learning_rate": 0.001769400938137698, + "loss": 0.1436, + "step": 23139 + }, + { + "epoch": 0.20086631192437565, + "grad_norm": 2.1875, + "learning_rate": 0.001769381043777367, + "loss": 0.3145, + "step": 23140 + }, + { + "epoch": 0.20087499240457982, + "grad_norm": 0.77734375, + "learning_rate": 0.001769361148685011, + "loss": 0.1348, + "step": 23141 + }, + { + "epoch": 0.20088367288478398, + "grad_norm": 0.6484375, + "learning_rate": 0.001769341252860652, + "loss": 0.1074, + "step": 23142 + }, + { + "epoch": 0.20089235336498815, + "grad_norm": 0.205078125, + "learning_rate": 0.001769321356304311, + "loss": 0.0898, + "step": 23143 + }, + { + "epoch": 0.2009010338451923, + "grad_norm": 0.51171875, + "learning_rate": 0.0017693014590160106, + "loss": 0.168, + "step": 23144 + }, + { + "epoch": 0.20090971432539648, + "grad_norm": 0.296875, + "learning_rate": 0.001769281560995772, + "loss": 0.127, + "step": 23145 + }, + { + "epoch": 0.20091839480560064, + "grad_norm": 0.1318359375, + "learning_rate": 0.001769261662243617, + "loss": 0.1514, + "step": 23146 + }, + { + "epoch": 0.2009270752858048, + "grad_norm": 0.1796875, + "learning_rate": 0.001769241762759568, + "loss": 0.126, + "step": 23147 + }, + { + "epoch": 0.20093575576600897, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017692218625436457, + "loss": 0.1631, + "step": 23148 + }, + { + "epoch": 0.20094443624621314, + "grad_norm": 0.49609375, + "learning_rate": 0.0017692019615958728, + "loss": 0.0864, + "step": 23149 + }, + { + "epoch": 0.2009531167264173, + "grad_norm": 0.70703125, + "learning_rate": 0.0017691820599162707, + "loss": 0.1094, + "step": 23150 + }, + { + "epoch": 0.20096179720662147, + "grad_norm": 0.416015625, + "learning_rate": 0.0017691621575048608, + "loss": 0.0952, + "step": 23151 + }, + { + "epoch": 0.20097047768682563, + "grad_norm": 0.212890625, + "learning_rate": 0.0017691422543616652, + "loss": 0.1021, + "step": 23152 + }, + { + "epoch": 0.2009791581670298, + "grad_norm": 0.26171875, + "learning_rate": 0.0017691223504867062, + "loss": 0.1069, + "step": 23153 + }, + { + "epoch": 0.20098783864723396, + "grad_norm": 0.52734375, + "learning_rate": 0.0017691024458800046, + "loss": 0.1328, + "step": 23154 + }, + { + "epoch": 0.20099651912743813, + "grad_norm": 0.07763671875, + "learning_rate": 0.001769082540541583, + "loss": 0.1309, + "step": 23155 + }, + { + "epoch": 0.2010051996076423, + "grad_norm": 0.2314453125, + "learning_rate": 0.0017690626344714624, + "loss": 0.0811, + "step": 23156 + }, + { + "epoch": 0.20101388008784646, + "grad_norm": 0.5625, + "learning_rate": 0.001769042727669665, + "loss": 0.1006, + "step": 23157 + }, + { + "epoch": 0.20102256056805062, + "grad_norm": 0.291015625, + "learning_rate": 0.0017690228201362132, + "loss": 0.0854, + "step": 23158 + }, + { + "epoch": 0.2010312410482548, + "grad_norm": 0.291015625, + "learning_rate": 0.0017690029118711272, + "loss": 0.1357, + "step": 23159 + }, + { + "epoch": 0.20103992152845895, + "grad_norm": 0.228515625, + "learning_rate": 0.00176898300287443, + "loss": 0.1084, + "step": 23160 + }, + { + "epoch": 0.20104860200866312, + "grad_norm": 1.203125, + "learning_rate": 0.001768963093146143, + "loss": 0.1084, + "step": 23161 + }, + { + "epoch": 0.20105728248886728, + "grad_norm": 0.158203125, + "learning_rate": 0.001768943182686288, + "loss": 0.1367, + "step": 23162 + }, + { + "epoch": 0.20106596296907145, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017689232714948868, + "loss": 0.1016, + "step": 23163 + }, + { + "epoch": 0.20107464344927561, + "grad_norm": 0.19921875, + "learning_rate": 0.0017689033595719615, + "loss": 0.0806, + "step": 23164 + }, + { + "epoch": 0.20108332392947978, + "grad_norm": 0.490234375, + "learning_rate": 0.001768883446917533, + "loss": 0.1504, + "step": 23165 + }, + { + "epoch": 0.20109200440968394, + "grad_norm": 0.3046875, + "learning_rate": 0.0017688635335316241, + "loss": 0.0791, + "step": 23166 + }, + { + "epoch": 0.2011006848898881, + "grad_norm": 1.0546875, + "learning_rate": 0.0017688436194142555, + "loss": 0.1602, + "step": 23167 + }, + { + "epoch": 0.20110936537009227, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017688237045654503, + "loss": 0.1309, + "step": 23168 + }, + { + "epoch": 0.20111804585029644, + "grad_norm": 0.43359375, + "learning_rate": 0.001768803788985229, + "loss": 0.1367, + "step": 23169 + }, + { + "epoch": 0.2011267263305006, + "grad_norm": 0.2216796875, + "learning_rate": 0.0017687838726736143, + "loss": 0.1016, + "step": 23170 + }, + { + "epoch": 0.20113540681070477, + "grad_norm": 0.88671875, + "learning_rate": 0.0017687639556306274, + "loss": 0.0879, + "step": 23171 + }, + { + "epoch": 0.20114408729090894, + "grad_norm": 0.1962890625, + "learning_rate": 0.00176874403785629, + "loss": 0.1484, + "step": 23172 + }, + { + "epoch": 0.2011527677711131, + "grad_norm": 0.330078125, + "learning_rate": 0.0017687241193506248, + "loss": 0.0869, + "step": 23173 + }, + { + "epoch": 0.20116144825131727, + "grad_norm": 0.65234375, + "learning_rate": 0.0017687042001136525, + "loss": 0.1182, + "step": 23174 + }, + { + "epoch": 0.20117012873152143, + "grad_norm": 0.19921875, + "learning_rate": 0.0017686842801453953, + "loss": 0.0938, + "step": 23175 + }, + { + "epoch": 0.2011788092117256, + "grad_norm": 0.06982421875, + "learning_rate": 0.001768664359445875, + "loss": 0.0786, + "step": 23176 + }, + { + "epoch": 0.20118748969192976, + "grad_norm": 0.828125, + "learning_rate": 0.0017686444380151138, + "loss": 0.1699, + "step": 23177 + }, + { + "epoch": 0.20119617017213393, + "grad_norm": 0.154296875, + "learning_rate": 0.0017686245158531329, + "loss": 0.1045, + "step": 23178 + }, + { + "epoch": 0.2012048506523381, + "grad_norm": 0.4140625, + "learning_rate": 0.0017686045929599542, + "loss": 0.1338, + "step": 23179 + }, + { + "epoch": 0.20121353113254226, + "grad_norm": 0.22265625, + "learning_rate": 0.0017685846693355997, + "loss": 0.1074, + "step": 23180 + }, + { + "epoch": 0.20122221161274642, + "grad_norm": 0.10693359375, + "learning_rate": 0.0017685647449800907, + "loss": 0.1235, + "step": 23181 + }, + { + "epoch": 0.2012308920929506, + "grad_norm": 0.10791015625, + "learning_rate": 0.0017685448198934497, + "loss": 0.0786, + "step": 23182 + }, + { + "epoch": 0.20123957257315475, + "grad_norm": 0.23828125, + "learning_rate": 0.0017685248940756979, + "loss": 0.1201, + "step": 23183 + }, + { + "epoch": 0.20124825305335892, + "grad_norm": 0.4765625, + "learning_rate": 0.0017685049675268574, + "loss": 0.1221, + "step": 23184 + }, + { + "epoch": 0.20125693353356308, + "grad_norm": 0.220703125, + "learning_rate": 0.0017684850402469498, + "loss": 0.0771, + "step": 23185 + }, + { + "epoch": 0.20126561401376725, + "grad_norm": 0.376953125, + "learning_rate": 0.0017684651122359973, + "loss": 0.123, + "step": 23186 + }, + { + "epoch": 0.2012742944939714, + "grad_norm": 0.158203125, + "learning_rate": 0.0017684451834940212, + "loss": 0.1157, + "step": 23187 + }, + { + "epoch": 0.20128297497417558, + "grad_norm": 0.8203125, + "learning_rate": 0.0017684252540210435, + "loss": 0.1299, + "step": 23188 + }, + { + "epoch": 0.20129165545437974, + "grad_norm": 0.07666015625, + "learning_rate": 0.0017684053238170858, + "loss": 0.0957, + "step": 23189 + }, + { + "epoch": 0.2013003359345839, + "grad_norm": 0.609375, + "learning_rate": 0.0017683853928821707, + "loss": 0.1006, + "step": 23190 + }, + { + "epoch": 0.20130901641478807, + "grad_norm": 0.515625, + "learning_rate": 0.0017683654612163189, + "loss": 0.1167, + "step": 23191 + }, + { + "epoch": 0.20131769689499224, + "grad_norm": 0.37890625, + "learning_rate": 0.0017683455288195528, + "loss": 0.1172, + "step": 23192 + }, + { + "epoch": 0.2013263773751964, + "grad_norm": 1.28125, + "learning_rate": 0.001768325595691894, + "loss": 0.1143, + "step": 23193 + }, + { + "epoch": 0.20133505785540057, + "grad_norm": 0.35546875, + "learning_rate": 0.0017683056618333641, + "loss": 0.0903, + "step": 23194 + }, + { + "epoch": 0.20134373833560473, + "grad_norm": 0.609375, + "learning_rate": 0.0017682857272439856, + "loss": 0.1045, + "step": 23195 + }, + { + "epoch": 0.2013524188158089, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017682657919237798, + "loss": 0.082, + "step": 23196 + }, + { + "epoch": 0.20136109929601306, + "grad_norm": 0.54296875, + "learning_rate": 0.0017682458558727684, + "loss": 0.1143, + "step": 23197 + }, + { + "epoch": 0.20136977977621723, + "grad_norm": 0.64453125, + "learning_rate": 0.0017682259190909735, + "loss": 0.1348, + "step": 23198 + }, + { + "epoch": 0.2013784602564214, + "grad_norm": 0.4140625, + "learning_rate": 0.0017682059815784168, + "loss": 0.1108, + "step": 23199 + }, + { + "epoch": 0.20138714073662556, + "grad_norm": 0.515625, + "learning_rate": 0.00176818604333512, + "loss": 0.1836, + "step": 23200 + }, + { + "epoch": 0.20139582121682972, + "grad_norm": 0.51171875, + "learning_rate": 0.001768166104361105, + "loss": 0.1182, + "step": 23201 + }, + { + "epoch": 0.2014045016970339, + "grad_norm": 0.1005859375, + "learning_rate": 0.0017681461646563937, + "loss": 0.1299, + "step": 23202 + }, + { + "epoch": 0.20141318217723805, + "grad_norm": 0.609375, + "learning_rate": 0.0017681262242210075, + "loss": 0.127, + "step": 23203 + }, + { + "epoch": 0.20142186265744222, + "grad_norm": 0.1396484375, + "learning_rate": 0.0017681062830549688, + "loss": 0.1523, + "step": 23204 + }, + { + "epoch": 0.20143054313764638, + "grad_norm": 0.11962890625, + "learning_rate": 0.001768086341158299, + "loss": 0.1328, + "step": 23205 + }, + { + "epoch": 0.20143922361785055, + "grad_norm": 0.17578125, + "learning_rate": 0.0017680663985310198, + "loss": 0.1357, + "step": 23206 + }, + { + "epoch": 0.20144790409805471, + "grad_norm": 1.234375, + "learning_rate": 0.0017680464551731536, + "loss": 0.1504, + "step": 23207 + }, + { + "epoch": 0.20145658457825888, + "grad_norm": 0.91015625, + "learning_rate": 0.0017680265110847217, + "loss": 0.1084, + "step": 23208 + }, + { + "epoch": 0.20146526505846304, + "grad_norm": 0.7109375, + "learning_rate": 0.001768006566265746, + "loss": 0.1245, + "step": 23209 + }, + { + "epoch": 0.2014739455386672, + "grad_norm": 0.361328125, + "learning_rate": 0.0017679866207162484, + "loss": 0.104, + "step": 23210 + }, + { + "epoch": 0.20148262601887137, + "grad_norm": 1.1796875, + "learning_rate": 0.0017679666744362507, + "loss": 0.1064, + "step": 23211 + }, + { + "epoch": 0.20149130649907554, + "grad_norm": 0.486328125, + "learning_rate": 0.0017679467274257746, + "loss": 0.1128, + "step": 23212 + }, + { + "epoch": 0.2014999869792797, + "grad_norm": 0.236328125, + "learning_rate": 0.001767926779684842, + "loss": 0.1182, + "step": 23213 + }, + { + "epoch": 0.20150866745948387, + "grad_norm": 0.15625, + "learning_rate": 0.0017679068312134744, + "loss": 0.1094, + "step": 23214 + }, + { + "epoch": 0.20151734793968804, + "grad_norm": 0.3359375, + "learning_rate": 0.0017678868820116943, + "loss": 0.1309, + "step": 23215 + }, + { + "epoch": 0.2015260284198922, + "grad_norm": 0.8828125, + "learning_rate": 0.0017678669320795233, + "loss": 0.0928, + "step": 23216 + }, + { + "epoch": 0.20153470890009637, + "grad_norm": 0.263671875, + "learning_rate": 0.0017678469814169828, + "loss": 0.124, + "step": 23217 + }, + { + "epoch": 0.20154338938030053, + "grad_norm": 0.19921875, + "learning_rate": 0.001767827030024095, + "loss": 0.1084, + "step": 23218 + }, + { + "epoch": 0.2015520698605047, + "grad_norm": 0.11865234375, + "learning_rate": 0.0017678070779008814, + "loss": 0.1299, + "step": 23219 + }, + { + "epoch": 0.20156075034070886, + "grad_norm": 0.109375, + "learning_rate": 0.0017677871250473642, + "loss": 0.1064, + "step": 23220 + }, + { + "epoch": 0.20156943082091303, + "grad_norm": 0.796875, + "learning_rate": 0.001767767171463565, + "loss": 0.1211, + "step": 23221 + }, + { + "epoch": 0.2015781113011172, + "grad_norm": 0.15234375, + "learning_rate": 0.0017677472171495057, + "loss": 0.1406, + "step": 23222 + }, + { + "epoch": 0.20158679178132136, + "grad_norm": 0.12060546875, + "learning_rate": 0.001767727262105208, + "loss": 0.1118, + "step": 23223 + }, + { + "epoch": 0.20159547226152552, + "grad_norm": 0.330078125, + "learning_rate": 0.0017677073063306938, + "loss": 0.1074, + "step": 23224 + }, + { + "epoch": 0.2016041527417297, + "grad_norm": 0.205078125, + "learning_rate": 0.0017676873498259846, + "loss": 0.0552, + "step": 23225 + }, + { + "epoch": 0.20161283322193382, + "grad_norm": 0.384765625, + "learning_rate": 0.001767667392591103, + "loss": 0.1211, + "step": 23226 + }, + { + "epoch": 0.201621513702138, + "grad_norm": 0.392578125, + "learning_rate": 0.0017676474346260702, + "loss": 0.0908, + "step": 23227 + }, + { + "epoch": 0.20163019418234215, + "grad_norm": 0.318359375, + "learning_rate": 0.001767627475930908, + "loss": 0.0991, + "step": 23228 + }, + { + "epoch": 0.20163887466254632, + "grad_norm": 0.39453125, + "learning_rate": 0.0017676075165056389, + "loss": 0.1138, + "step": 23229 + }, + { + "epoch": 0.20164755514275048, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017675875563502841, + "loss": 0.1006, + "step": 23230 + }, + { + "epoch": 0.20165623562295465, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017675675954648654, + "loss": 0.0962, + "step": 23231 + }, + { + "epoch": 0.20166491610315881, + "grad_norm": 0.84765625, + "learning_rate": 0.001767547633849405, + "loss": 0.1001, + "step": 23232 + }, + { + "epoch": 0.20167359658336298, + "grad_norm": 0.11328125, + "learning_rate": 0.0017675276715039244, + "loss": 0.0894, + "step": 23233 + }, + { + "epoch": 0.20168227706356714, + "grad_norm": 0.1484375, + "learning_rate": 0.0017675077084284454, + "loss": 0.1172, + "step": 23234 + }, + { + "epoch": 0.2016909575437713, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017674877446229903, + "loss": 0.0986, + "step": 23235 + }, + { + "epoch": 0.20169963802397547, + "grad_norm": 0.34765625, + "learning_rate": 0.0017674677800875805, + "loss": 0.1006, + "step": 23236 + }, + { + "epoch": 0.20170831850417964, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017674478148222378, + "loss": 0.1201, + "step": 23237 + }, + { + "epoch": 0.2017169989843838, + "grad_norm": 0.2890625, + "learning_rate": 0.0017674278488269844, + "loss": 0.0859, + "step": 23238 + }, + { + "epoch": 0.20172567946458797, + "grad_norm": 0.490234375, + "learning_rate": 0.001767407882101842, + "loss": 0.2637, + "step": 23239 + }, + { + "epoch": 0.20173435994479214, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017673879146468323, + "loss": 0.0996, + "step": 23240 + }, + { + "epoch": 0.2017430404249963, + "grad_norm": 0.328125, + "learning_rate": 0.001767367946461977, + "loss": 0.0801, + "step": 23241 + }, + { + "epoch": 0.20175172090520047, + "grad_norm": 0.2578125, + "learning_rate": 0.0017673479775472984, + "loss": 0.1074, + "step": 23242 + }, + { + "epoch": 0.20176040138540463, + "grad_norm": 0.392578125, + "learning_rate": 0.0017673280079028182, + "loss": 0.1182, + "step": 23243 + }, + { + "epoch": 0.2017690818656088, + "grad_norm": 0.69140625, + "learning_rate": 0.001767308037528558, + "loss": 0.1201, + "step": 23244 + }, + { + "epoch": 0.20177776234581296, + "grad_norm": 0.349609375, + "learning_rate": 0.0017672880664245394, + "loss": 0.1182, + "step": 23245 + }, + { + "epoch": 0.20178644282601713, + "grad_norm": 0.404296875, + "learning_rate": 0.0017672680945907849, + "loss": 0.1211, + "step": 23246 + }, + { + "epoch": 0.2017951233062213, + "grad_norm": 1.6171875, + "learning_rate": 0.001767248122027316, + "loss": 0.1338, + "step": 23247 + }, + { + "epoch": 0.20180380378642546, + "grad_norm": 0.28125, + "learning_rate": 0.0017672281487341548, + "loss": 0.1025, + "step": 23248 + }, + { + "epoch": 0.20181248426662962, + "grad_norm": 0.390625, + "learning_rate": 0.0017672081747113226, + "loss": 0.124, + "step": 23249 + }, + { + "epoch": 0.2018211647468338, + "grad_norm": 0.60546875, + "learning_rate": 0.0017671881999588419, + "loss": 0.1172, + "step": 23250 + }, + { + "epoch": 0.20182984522703795, + "grad_norm": 0.314453125, + "learning_rate": 0.001767168224476734, + "loss": 0.1445, + "step": 23251 + }, + { + "epoch": 0.20183852570724212, + "grad_norm": 0.349609375, + "learning_rate": 0.0017671482482650208, + "loss": 0.1338, + "step": 23252 + }, + { + "epoch": 0.20184720618744628, + "grad_norm": 0.267578125, + "learning_rate": 0.0017671282713237243, + "loss": 0.0967, + "step": 23253 + }, + { + "epoch": 0.20185588666765045, + "grad_norm": 0.15234375, + "learning_rate": 0.0017671082936528667, + "loss": 0.0942, + "step": 23254 + }, + { + "epoch": 0.2018645671478546, + "grad_norm": 0.0859375, + "learning_rate": 0.001767088315252469, + "loss": 0.1045, + "step": 23255 + }, + { + "epoch": 0.20187324762805878, + "grad_norm": 0.1796875, + "learning_rate": 0.001767068336122554, + "loss": 0.1011, + "step": 23256 + }, + { + "epoch": 0.20188192810826294, + "grad_norm": 0.294921875, + "learning_rate": 0.001767048356263143, + "loss": 0.127, + "step": 23257 + }, + { + "epoch": 0.2018906085884671, + "grad_norm": 0.4453125, + "learning_rate": 0.0017670283756742577, + "loss": 0.0864, + "step": 23258 + }, + { + "epoch": 0.20189928906867127, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017670083943559203, + "loss": 0.0845, + "step": 23259 + }, + { + "epoch": 0.20190796954887544, + "grad_norm": 0.57421875, + "learning_rate": 0.0017669884123081528, + "loss": 0.0889, + "step": 23260 + }, + { + "epoch": 0.2019166500290796, + "grad_norm": 0.15234375, + "learning_rate": 0.0017669684295309764, + "loss": 0.0869, + "step": 23261 + }, + { + "epoch": 0.20192533050928377, + "grad_norm": 0.08740234375, + "learning_rate": 0.0017669484460244133, + "loss": 0.1445, + "step": 23262 + }, + { + "epoch": 0.20193401098948793, + "grad_norm": 0.259765625, + "learning_rate": 0.001766928461788486, + "loss": 0.1211, + "step": 23263 + }, + { + "epoch": 0.2019426914696921, + "grad_norm": 0.337890625, + "learning_rate": 0.0017669084768232152, + "loss": 0.1143, + "step": 23264 + }, + { + "epoch": 0.20195137194989626, + "grad_norm": 0.208984375, + "learning_rate": 0.0017668884911286235, + "loss": 0.0874, + "step": 23265 + }, + { + "epoch": 0.20196005243010043, + "grad_norm": 0.91015625, + "learning_rate": 0.0017668685047047329, + "loss": 0.0732, + "step": 23266 + }, + { + "epoch": 0.2019687329103046, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017668485175515642, + "loss": 0.1738, + "step": 23267 + }, + { + "epoch": 0.20197741339050876, + "grad_norm": 0.166015625, + "learning_rate": 0.0017668285296691404, + "loss": 0.1064, + "step": 23268 + }, + { + "epoch": 0.20198609387071292, + "grad_norm": 0.1650390625, + "learning_rate": 0.001766808541057483, + "loss": 0.1074, + "step": 23269 + }, + { + "epoch": 0.2019947743509171, + "grad_norm": 0.53515625, + "learning_rate": 0.0017667885517166138, + "loss": 0.1055, + "step": 23270 + }, + { + "epoch": 0.20200345483112125, + "grad_norm": 0.1328125, + "learning_rate": 0.0017667685616465545, + "loss": 0.0962, + "step": 23271 + }, + { + "epoch": 0.20201213531132542, + "grad_norm": 0.4609375, + "learning_rate": 0.0017667485708473272, + "loss": 0.0811, + "step": 23272 + }, + { + "epoch": 0.20202081579152958, + "grad_norm": 0.322265625, + "learning_rate": 0.0017667285793189534, + "loss": 0.1104, + "step": 23273 + }, + { + "epoch": 0.20202949627173375, + "grad_norm": 0.2314453125, + "learning_rate": 0.0017667085870614557, + "loss": 0.1055, + "step": 23274 + }, + { + "epoch": 0.20203817675193791, + "grad_norm": 0.0908203125, + "learning_rate": 0.0017666885940748553, + "loss": 0.0933, + "step": 23275 + }, + { + "epoch": 0.20204685723214208, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017666686003591742, + "loss": 0.1387, + "step": 23276 + }, + { + "epoch": 0.20205553771234624, + "grad_norm": 0.49609375, + "learning_rate": 0.0017666486059144343, + "loss": 0.1104, + "step": 23277 + }, + { + "epoch": 0.2020642181925504, + "grad_norm": 0.341796875, + "learning_rate": 0.0017666286107406578, + "loss": 0.1289, + "step": 23278 + }, + { + "epoch": 0.20207289867275458, + "grad_norm": 0.1748046875, + "learning_rate": 0.001766608614837866, + "loss": 0.1064, + "step": 23279 + }, + { + "epoch": 0.20208157915295874, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017665886182060813, + "loss": 0.1152, + "step": 23280 + }, + { + "epoch": 0.2020902596331629, + "grad_norm": 0.146484375, + "learning_rate": 0.0017665686208453248, + "loss": 0.0742, + "step": 23281 + }, + { + "epoch": 0.20209894011336707, + "grad_norm": 1.046875, + "learning_rate": 0.0017665486227556194, + "loss": 0.5703, + "step": 23282 + }, + { + "epoch": 0.20210762059357124, + "grad_norm": 0.2578125, + "learning_rate": 0.0017665286239369861, + "loss": 0.0981, + "step": 23283 + }, + { + "epoch": 0.2021163010737754, + "grad_norm": 0.267578125, + "learning_rate": 0.0017665086243894472, + "loss": 0.1162, + "step": 23284 + }, + { + "epoch": 0.20212498155397957, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017664886241130246, + "loss": 0.1162, + "step": 23285 + }, + { + "epoch": 0.20213366203418373, + "grad_norm": 0.31640625, + "learning_rate": 0.0017664686231077399, + "loss": 0.1147, + "step": 23286 + }, + { + "epoch": 0.2021423425143879, + "grad_norm": 0.373046875, + "learning_rate": 0.0017664486213736152, + "loss": 0.1074, + "step": 23287 + }, + { + "epoch": 0.20215102299459206, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017664286189106727, + "loss": 0.127, + "step": 23288 + }, + { + "epoch": 0.20215970347479623, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017664086157189332, + "loss": 0.0986, + "step": 23289 + }, + { + "epoch": 0.2021683839550004, + "grad_norm": 0.255859375, + "learning_rate": 0.0017663886117984197, + "loss": 0.0977, + "step": 23290 + }, + { + "epoch": 0.20217706443520456, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017663686071491532, + "loss": 0.1182, + "step": 23291 + }, + { + "epoch": 0.20218574491540872, + "grad_norm": 0.109375, + "learning_rate": 0.0017663486017711564, + "loss": 0.0889, + "step": 23292 + }, + { + "epoch": 0.2021944253956129, + "grad_norm": 0.83203125, + "learning_rate": 0.0017663285956644507, + "loss": 0.124, + "step": 23293 + }, + { + "epoch": 0.20220310587581705, + "grad_norm": 0.435546875, + "learning_rate": 0.0017663085888290578, + "loss": 0.1182, + "step": 23294 + }, + { + "epoch": 0.20221178635602122, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017662885812650003, + "loss": 0.1084, + "step": 23295 + }, + { + "epoch": 0.20222046683622538, + "grad_norm": 0.1328125, + "learning_rate": 0.0017662685729722993, + "loss": 0.1348, + "step": 23296 + }, + { + "epoch": 0.20222914731642955, + "grad_norm": 0.326171875, + "learning_rate": 0.0017662485639509768, + "loss": 0.0869, + "step": 23297 + }, + { + "epoch": 0.2022378277966337, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017662285542010552, + "loss": 0.1191, + "step": 23298 + }, + { + "epoch": 0.20224650827683788, + "grad_norm": 0.37890625, + "learning_rate": 0.0017662085437225561, + "loss": 0.1182, + "step": 23299 + }, + { + "epoch": 0.20225518875704204, + "grad_norm": 0.6328125, + "learning_rate": 0.0017661885325155012, + "loss": 0.1172, + "step": 23300 + }, + { + "epoch": 0.2022638692372462, + "grad_norm": 0.234375, + "learning_rate": 0.0017661685205799125, + "loss": 0.1465, + "step": 23301 + }, + { + "epoch": 0.20227254971745037, + "grad_norm": 0.71484375, + "learning_rate": 0.0017661485079158122, + "loss": 0.1855, + "step": 23302 + }, + { + "epoch": 0.20228123019765454, + "grad_norm": 0.2275390625, + "learning_rate": 0.0017661284945232217, + "loss": 0.0869, + "step": 23303 + }, + { + "epoch": 0.2022899106778587, + "grad_norm": 0.208984375, + "learning_rate": 0.0017661084804021629, + "loss": 0.1021, + "step": 23304 + }, + { + "epoch": 0.20229859115806287, + "grad_norm": 0.431640625, + "learning_rate": 0.0017660884655526581, + "loss": 0.1309, + "step": 23305 + }, + { + "epoch": 0.20230727163826703, + "grad_norm": 0.5078125, + "learning_rate": 0.0017660684499747291, + "loss": 0.0903, + "step": 23306 + }, + { + "epoch": 0.2023159521184712, + "grad_norm": 0.10302734375, + "learning_rate": 0.0017660484336683973, + "loss": 0.123, + "step": 23307 + }, + { + "epoch": 0.20232463259867536, + "grad_norm": 0.291015625, + "learning_rate": 0.001766028416633685, + "loss": 0.1338, + "step": 23308 + }, + { + "epoch": 0.20233331307887953, + "grad_norm": 0.359375, + "learning_rate": 0.0017660083988706144, + "loss": 0.0752, + "step": 23309 + }, + { + "epoch": 0.2023419935590837, + "grad_norm": 0.36328125, + "learning_rate": 0.0017659883803792067, + "loss": 0.082, + "step": 23310 + }, + { + "epoch": 0.20235067403928786, + "grad_norm": 0.1513671875, + "learning_rate": 0.001765968361159484, + "loss": 0.127, + "step": 23311 + }, + { + "epoch": 0.20235935451949202, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017659483412114686, + "loss": 0.1011, + "step": 23312 + }, + { + "epoch": 0.2023680349996962, + "grad_norm": 0.400390625, + "learning_rate": 0.0017659283205351819, + "loss": 0.1309, + "step": 23313 + }, + { + "epoch": 0.20237671547990035, + "grad_norm": 0.1474609375, + "learning_rate": 0.001765908299130646, + "loss": 0.1396, + "step": 23314 + }, + { + "epoch": 0.20238539596010452, + "grad_norm": 0.1845703125, + "learning_rate": 0.001765888276997883, + "loss": 0.0908, + "step": 23315 + }, + { + "epoch": 0.20239407644030868, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017658682541369145, + "loss": 0.1084, + "step": 23316 + }, + { + "epoch": 0.20240275692051285, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017658482305477625, + "loss": 0.1162, + "step": 23317 + }, + { + "epoch": 0.20241143740071701, + "grad_norm": 0.64453125, + "learning_rate": 0.0017658282062304486, + "loss": 0.1143, + "step": 23318 + }, + { + "epoch": 0.20242011788092118, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017658081811849952, + "loss": 0.1143, + "step": 23319 + }, + { + "epoch": 0.20242879836112534, + "grad_norm": 0.2421875, + "learning_rate": 0.0017657881554114239, + "loss": 0.0938, + "step": 23320 + }, + { + "epoch": 0.2024374788413295, + "grad_norm": 0.267578125, + "learning_rate": 0.0017657681289097566, + "loss": 0.1094, + "step": 23321 + }, + { + "epoch": 0.20244615932153368, + "grad_norm": 0.09619140625, + "learning_rate": 0.001765748101680016, + "loss": 0.1318, + "step": 23322 + }, + { + "epoch": 0.20245483980173784, + "grad_norm": 0.87109375, + "learning_rate": 0.0017657280737222224, + "loss": 0.1484, + "step": 23323 + }, + { + "epoch": 0.202463520281942, + "grad_norm": 0.298828125, + "learning_rate": 0.0017657080450363988, + "loss": 0.1182, + "step": 23324 + }, + { + "epoch": 0.20247220076214617, + "grad_norm": 0.384765625, + "learning_rate": 0.001765688015622567, + "loss": 0.1074, + "step": 23325 + }, + { + "epoch": 0.20248088124235034, + "grad_norm": 0.232421875, + "learning_rate": 0.001765667985480749, + "loss": 0.1016, + "step": 23326 + }, + { + "epoch": 0.2024895617225545, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017656479546109662, + "loss": 0.1289, + "step": 23327 + }, + { + "epoch": 0.20249824220275867, + "grad_norm": 0.455078125, + "learning_rate": 0.0017656279230132406, + "loss": 0.0938, + "step": 23328 + }, + { + "epoch": 0.20250692268296283, + "grad_norm": 0.10546875, + "learning_rate": 0.0017656078906875946, + "loss": 0.1191, + "step": 23329 + }, + { + "epoch": 0.202515603163167, + "grad_norm": 0.498046875, + "learning_rate": 0.0017655878576340496, + "loss": 0.0928, + "step": 23330 + }, + { + "epoch": 0.20252428364337116, + "grad_norm": 0.56640625, + "learning_rate": 0.0017655678238526278, + "loss": 0.1221, + "step": 23331 + }, + { + "epoch": 0.20253296412357533, + "grad_norm": 0.0732421875, + "learning_rate": 0.0017655477893433514, + "loss": 0.0884, + "step": 23332 + }, + { + "epoch": 0.2025416446037795, + "grad_norm": 0.1064453125, + "learning_rate": 0.0017655277541062415, + "loss": 0.1406, + "step": 23333 + }, + { + "epoch": 0.20255032508398366, + "grad_norm": 0.337890625, + "learning_rate": 0.0017655077181413206, + "loss": 0.0938, + "step": 23334 + }, + { + "epoch": 0.20255900556418782, + "grad_norm": 0.30859375, + "learning_rate": 0.0017654876814486105, + "loss": 0.126, + "step": 23335 + }, + { + "epoch": 0.202567686044392, + "grad_norm": 0.57421875, + "learning_rate": 0.0017654676440281333, + "loss": 0.1069, + "step": 23336 + }, + { + "epoch": 0.20257636652459615, + "grad_norm": 0.3671875, + "learning_rate": 0.0017654476058799101, + "loss": 0.124, + "step": 23337 + }, + { + "epoch": 0.20258504700480032, + "grad_norm": 0.1708984375, + "learning_rate": 0.001765427567003964, + "loss": 0.1157, + "step": 23338 + }, + { + "epoch": 0.20259372748500448, + "grad_norm": 0.640625, + "learning_rate": 0.0017654075274003158, + "loss": 0.1221, + "step": 23339 + }, + { + "epoch": 0.20260240796520865, + "grad_norm": 0.515625, + "learning_rate": 0.0017653874870689882, + "loss": 0.0815, + "step": 23340 + }, + { + "epoch": 0.2026110884454128, + "grad_norm": 0.265625, + "learning_rate": 0.0017653674460100027, + "loss": 0.0986, + "step": 23341 + }, + { + "epoch": 0.20261976892561698, + "grad_norm": 0.09912109375, + "learning_rate": 0.0017653474042233816, + "loss": 0.1016, + "step": 23342 + }, + { + "epoch": 0.20262844940582114, + "grad_norm": 0.06298828125, + "learning_rate": 0.0017653273617091463, + "loss": 0.0801, + "step": 23343 + }, + { + "epoch": 0.2026371298860253, + "grad_norm": 1.1640625, + "learning_rate": 0.001765307318467319, + "loss": 0.1855, + "step": 23344 + }, + { + "epoch": 0.20264581036622947, + "grad_norm": 0.455078125, + "learning_rate": 0.0017652872744979217, + "loss": 0.1084, + "step": 23345 + }, + { + "epoch": 0.20265449084643364, + "grad_norm": 0.0849609375, + "learning_rate": 0.0017652672298009763, + "loss": 0.1006, + "step": 23346 + }, + { + "epoch": 0.2026631713266378, + "grad_norm": 0.75390625, + "learning_rate": 0.0017652471843765045, + "loss": 0.1108, + "step": 23347 + }, + { + "epoch": 0.20267185180684197, + "grad_norm": 1.015625, + "learning_rate": 0.0017652271382245287, + "loss": 0.1641, + "step": 23348 + }, + { + "epoch": 0.2026805322870461, + "grad_norm": 0.115234375, + "learning_rate": 0.0017652070913450702, + "loss": 0.1191, + "step": 23349 + }, + { + "epoch": 0.20268921276725027, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017651870437381512, + "loss": 0.1406, + "step": 23350 + }, + { + "epoch": 0.20269789324745444, + "grad_norm": 0.20703125, + "learning_rate": 0.0017651669954037938, + "loss": 0.1152, + "step": 23351 + }, + { + "epoch": 0.2027065737276586, + "grad_norm": 0.4609375, + "learning_rate": 0.0017651469463420197, + "loss": 0.1162, + "step": 23352 + }, + { + "epoch": 0.20271525420786277, + "grad_norm": 0.494140625, + "learning_rate": 0.001765126896552851, + "loss": 0.1001, + "step": 23353 + }, + { + "epoch": 0.20272393468806693, + "grad_norm": 0.11865234375, + "learning_rate": 0.0017651068460363094, + "loss": 0.1182, + "step": 23354 + }, + { + "epoch": 0.2027326151682711, + "grad_norm": 0.1845703125, + "learning_rate": 0.001765086794792417, + "loss": 0.1309, + "step": 23355 + }, + { + "epoch": 0.20274129564847526, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017650667428211954, + "loss": 0.1191, + "step": 23356 + }, + { + "epoch": 0.20274997612867943, + "grad_norm": 0.72265625, + "learning_rate": 0.0017650466901226668, + "loss": 0.1064, + "step": 23357 + }, + { + "epoch": 0.2027586566088836, + "grad_norm": 0.390625, + "learning_rate": 0.0017650266366968537, + "loss": 0.0854, + "step": 23358 + }, + { + "epoch": 0.20276733708908776, + "grad_norm": 0.92578125, + "learning_rate": 0.001765006582543777, + "loss": 0.127, + "step": 23359 + }, + { + "epoch": 0.20277601756929192, + "grad_norm": 0.349609375, + "learning_rate": 0.0017649865276634594, + "loss": 0.0928, + "step": 23360 + }, + { + "epoch": 0.2027846980494961, + "grad_norm": 0.373046875, + "learning_rate": 0.001764966472055922, + "loss": 0.1318, + "step": 23361 + }, + { + "epoch": 0.20279337852970025, + "grad_norm": 0.20703125, + "learning_rate": 0.001764946415721188, + "loss": 0.1113, + "step": 23362 + }, + { + "epoch": 0.20280205900990442, + "grad_norm": 0.9296875, + "learning_rate": 0.001764926358659278, + "loss": 0.1396, + "step": 23363 + }, + { + "epoch": 0.20281073949010858, + "grad_norm": 0.427734375, + "learning_rate": 0.001764906300870215, + "loss": 0.1592, + "step": 23364 + }, + { + "epoch": 0.20281941997031275, + "grad_norm": 0.10986328125, + "learning_rate": 0.00176488624235402, + "loss": 0.0938, + "step": 23365 + }, + { + "epoch": 0.2028281004505169, + "grad_norm": 0.5859375, + "learning_rate": 0.0017648661831107155, + "loss": 0.0928, + "step": 23366 + }, + { + "epoch": 0.20283678093072108, + "grad_norm": 0.71484375, + "learning_rate": 0.0017648461231403235, + "loss": 0.1699, + "step": 23367 + }, + { + "epoch": 0.20284546141092524, + "grad_norm": 0.275390625, + "learning_rate": 0.0017648260624428655, + "loss": 0.0791, + "step": 23368 + }, + { + "epoch": 0.2028541418911294, + "grad_norm": 0.142578125, + "learning_rate": 0.0017648060010183643, + "loss": 0.0854, + "step": 23369 + }, + { + "epoch": 0.20286282237133357, + "grad_norm": 1.2734375, + "learning_rate": 0.0017647859388668407, + "loss": 0.1108, + "step": 23370 + }, + { + "epoch": 0.20287150285153774, + "grad_norm": 0.265625, + "learning_rate": 0.0017647658759883172, + "loss": 0.1074, + "step": 23371 + }, + { + "epoch": 0.2028801833317419, + "grad_norm": 0.26953125, + "learning_rate": 0.0017647458123828158, + "loss": 0.1104, + "step": 23372 + }, + { + "epoch": 0.20288886381194607, + "grad_norm": 0.98046875, + "learning_rate": 0.0017647257480503582, + "loss": 0.1143, + "step": 23373 + }, + { + "epoch": 0.20289754429215023, + "grad_norm": 0.52734375, + "learning_rate": 0.0017647056829909668, + "loss": 0.1094, + "step": 23374 + }, + { + "epoch": 0.2029062247723544, + "grad_norm": 0.359375, + "learning_rate": 0.0017646856172046635, + "loss": 0.085, + "step": 23375 + }, + { + "epoch": 0.20291490525255856, + "grad_norm": 0.44921875, + "learning_rate": 0.0017646655506914695, + "loss": 0.1191, + "step": 23376 + }, + { + "epoch": 0.20292358573276273, + "grad_norm": 0.21875, + "learning_rate": 0.0017646454834514074, + "loss": 0.123, + "step": 23377 + }, + { + "epoch": 0.2029322662129669, + "grad_norm": 0.1005859375, + "learning_rate": 0.0017646254154844987, + "loss": 0.0972, + "step": 23378 + }, + { + "epoch": 0.20294094669317106, + "grad_norm": 0.0869140625, + "learning_rate": 0.0017646053467907663, + "loss": 0.1201, + "step": 23379 + }, + { + "epoch": 0.20294962717337522, + "grad_norm": 0.373046875, + "learning_rate": 0.001764585277370231, + "loss": 0.0879, + "step": 23380 + }, + { + "epoch": 0.2029583076535794, + "grad_norm": 0.82421875, + "learning_rate": 0.0017645652072229153, + "loss": 0.0952, + "step": 23381 + }, + { + "epoch": 0.20296698813378355, + "grad_norm": 0.330078125, + "learning_rate": 0.0017645451363488415, + "loss": 0.1436, + "step": 23382 + }, + { + "epoch": 0.20297566861398772, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017645250647480301, + "loss": 0.1182, + "step": 23383 + }, + { + "epoch": 0.20298434909419188, + "grad_norm": 0.07080078125, + "learning_rate": 0.001764504992420505, + "loss": 0.1045, + "step": 23384 + }, + { + "epoch": 0.20299302957439605, + "grad_norm": 0.380859375, + "learning_rate": 0.001764484919366287, + "loss": 0.1152, + "step": 23385 + }, + { + "epoch": 0.20300171005460022, + "grad_norm": 0.2392578125, + "learning_rate": 0.001764464845585398, + "loss": 0.0645, + "step": 23386 + }, + { + "epoch": 0.20301039053480438, + "grad_norm": 0.453125, + "learning_rate": 0.0017644447710778604, + "loss": 0.0938, + "step": 23387 + }, + { + "epoch": 0.20301907101500855, + "grad_norm": 1.0625, + "learning_rate": 0.0017644246958436961, + "loss": 0.1152, + "step": 23388 + }, + { + "epoch": 0.2030277514952127, + "grad_norm": 0.64453125, + "learning_rate": 0.0017644046198829267, + "loss": 0.1367, + "step": 23389 + }, + { + "epoch": 0.20303643197541688, + "grad_norm": 0.09033203125, + "learning_rate": 0.0017643845431955748, + "loss": 0.0938, + "step": 23390 + }, + { + "epoch": 0.20304511245562104, + "grad_norm": 0.310546875, + "learning_rate": 0.0017643644657816615, + "loss": 0.0908, + "step": 23391 + }, + { + "epoch": 0.2030537929358252, + "grad_norm": 0.09912109375, + "learning_rate": 0.0017643443876412094, + "loss": 0.1211, + "step": 23392 + }, + { + "epoch": 0.20306247341602937, + "grad_norm": 0.29296875, + "learning_rate": 0.0017643243087742404, + "loss": 0.105, + "step": 23393 + }, + { + "epoch": 0.20307115389623354, + "grad_norm": 0.310546875, + "learning_rate": 0.0017643042291807763, + "loss": 0.1167, + "step": 23394 + }, + { + "epoch": 0.2030798343764377, + "grad_norm": 0.2890625, + "learning_rate": 0.0017642841488608384, + "loss": 0.1089, + "step": 23395 + }, + { + "epoch": 0.20308851485664187, + "grad_norm": 0.10986328125, + "learning_rate": 0.0017642640678144502, + "loss": 0.0718, + "step": 23396 + }, + { + "epoch": 0.20309719533684603, + "grad_norm": 0.50390625, + "learning_rate": 0.0017642439860416327, + "loss": 0.0977, + "step": 23397 + }, + { + "epoch": 0.2031058758170502, + "grad_norm": 0.8125, + "learning_rate": 0.0017642239035424074, + "loss": 0.0972, + "step": 23398 + }, + { + "epoch": 0.20311455629725436, + "grad_norm": 0.58984375, + "learning_rate": 0.0017642038203167973, + "loss": 0.0981, + "step": 23399 + }, + { + "epoch": 0.20312323677745853, + "grad_norm": 0.921875, + "learning_rate": 0.0017641837363648238, + "loss": 0.1104, + "step": 23400 + }, + { + "epoch": 0.2031319172576627, + "grad_norm": 0.125, + "learning_rate": 0.0017641636516865088, + "loss": 0.1299, + "step": 23401 + }, + { + "epoch": 0.20314059773786686, + "grad_norm": 1.15625, + "learning_rate": 0.0017641435662818744, + "loss": 0.1123, + "step": 23402 + }, + { + "epoch": 0.20314927821807102, + "grad_norm": 0.53515625, + "learning_rate": 0.0017641234801509427, + "loss": 0.124, + "step": 23403 + }, + { + "epoch": 0.2031579586982752, + "grad_norm": 0.609375, + "learning_rate": 0.0017641033932937354, + "loss": 0.1289, + "step": 23404 + }, + { + "epoch": 0.20316663917847935, + "grad_norm": 0.259765625, + "learning_rate": 0.0017640833057102749, + "loss": 0.1445, + "step": 23405 + }, + { + "epoch": 0.20317531965868352, + "grad_norm": 0.48828125, + "learning_rate": 0.0017640632174005825, + "loss": 0.1338, + "step": 23406 + }, + { + "epoch": 0.20318400013888768, + "grad_norm": 0.78125, + "learning_rate": 0.0017640431283646807, + "loss": 0.1157, + "step": 23407 + }, + { + "epoch": 0.20319268061909185, + "grad_norm": 0.416015625, + "learning_rate": 0.0017640230386025916, + "loss": 0.0928, + "step": 23408 + }, + { + "epoch": 0.203201361099296, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017640029481143366, + "loss": 0.0845, + "step": 23409 + }, + { + "epoch": 0.20321004157950018, + "grad_norm": 0.33203125, + "learning_rate": 0.0017639828568999378, + "loss": 0.0894, + "step": 23410 + }, + { + "epoch": 0.20321872205970434, + "grad_norm": 0.380859375, + "learning_rate": 0.0017639627649594174, + "loss": 0.1006, + "step": 23411 + }, + { + "epoch": 0.2032274025399085, + "grad_norm": 0.1123046875, + "learning_rate": 0.0017639426722927975, + "loss": 0.1152, + "step": 23412 + }, + { + "epoch": 0.20323608302011267, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017639225789000997, + "loss": 0.1377, + "step": 23413 + }, + { + "epoch": 0.20324476350031684, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017639024847813463, + "loss": 0.0981, + "step": 23414 + }, + { + "epoch": 0.203253443980521, + "grad_norm": 0.41015625, + "learning_rate": 0.001763882389936559, + "loss": 0.1172, + "step": 23415 + }, + { + "epoch": 0.20326212446072517, + "grad_norm": 0.39453125, + "learning_rate": 0.00176386229436576, + "loss": 0.0811, + "step": 23416 + }, + { + "epoch": 0.20327080494092933, + "grad_norm": 0.375, + "learning_rate": 0.001763842198068971, + "loss": 0.1064, + "step": 23417 + }, + { + "epoch": 0.2032794854211335, + "grad_norm": 0.169921875, + "learning_rate": 0.001763822101046214, + "loss": 0.0933, + "step": 23418 + }, + { + "epoch": 0.20328816590133766, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017638020032975116, + "loss": 0.1152, + "step": 23419 + }, + { + "epoch": 0.20329684638154183, + "grad_norm": 0.18359375, + "learning_rate": 0.001763781904822885, + "loss": 0.1016, + "step": 23420 + }, + { + "epoch": 0.203305526861746, + "grad_norm": 0.5234375, + "learning_rate": 0.0017637618056223566, + "loss": 0.085, + "step": 23421 + }, + { + "epoch": 0.20331420734195016, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017637417056959484, + "loss": 0.1348, + "step": 23422 + }, + { + "epoch": 0.20332288782215432, + "grad_norm": 0.34375, + "learning_rate": 0.0017637216050436818, + "loss": 0.1094, + "step": 23423 + }, + { + "epoch": 0.2033315683023585, + "grad_norm": 0.32421875, + "learning_rate": 0.0017637015036655794, + "loss": 0.1104, + "step": 23424 + }, + { + "epoch": 0.20334024878256265, + "grad_norm": 0.609375, + "learning_rate": 0.001763681401561663, + "loss": 0.1396, + "step": 23425 + }, + { + "epoch": 0.20334892926276682, + "grad_norm": 0.19921875, + "learning_rate": 0.0017636612987319549, + "loss": 0.1143, + "step": 23426 + }, + { + "epoch": 0.20335760974297099, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017636411951764766, + "loss": 0.1162, + "step": 23427 + }, + { + "epoch": 0.20336629022317515, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017636210908952498, + "loss": 0.1211, + "step": 23428 + }, + { + "epoch": 0.20337497070337932, + "grad_norm": 0.45703125, + "learning_rate": 0.0017636009858882976, + "loss": 0.1855, + "step": 23429 + }, + { + "epoch": 0.20338365118358348, + "grad_norm": 0.197265625, + "learning_rate": 0.0017635808801556411, + "loss": 0.1138, + "step": 23430 + }, + { + "epoch": 0.20339233166378765, + "grad_norm": 0.271484375, + "learning_rate": 0.0017635607736973026, + "loss": 0.1162, + "step": 23431 + }, + { + "epoch": 0.2034010121439918, + "grad_norm": 0.365234375, + "learning_rate": 0.0017635406665133038, + "loss": 0.1533, + "step": 23432 + }, + { + "epoch": 0.20340969262419598, + "grad_norm": 0.10009765625, + "learning_rate": 0.001763520558603667, + "loss": 0.1328, + "step": 23433 + }, + { + "epoch": 0.20341837310440014, + "grad_norm": 0.123046875, + "learning_rate": 0.001763500449968414, + "loss": 0.1172, + "step": 23434 + }, + { + "epoch": 0.2034270535846043, + "grad_norm": 0.1845703125, + "learning_rate": 0.001763480340607567, + "loss": 0.1055, + "step": 23435 + }, + { + "epoch": 0.20343573406480847, + "grad_norm": 0.08837890625, + "learning_rate": 0.001763460230521148, + "loss": 0.0957, + "step": 23436 + }, + { + "epoch": 0.20344441454501264, + "grad_norm": 0.36328125, + "learning_rate": 0.0017634401197091786, + "loss": 0.1045, + "step": 23437 + }, + { + "epoch": 0.2034530950252168, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017634200081716811, + "loss": 0.1123, + "step": 23438 + }, + { + "epoch": 0.20346177550542097, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017633998959086777, + "loss": 0.1084, + "step": 23439 + }, + { + "epoch": 0.20347045598562513, + "grad_norm": 0.69921875, + "learning_rate": 0.0017633797829201896, + "loss": 0.1079, + "step": 23440 + }, + { + "epoch": 0.2034791364658293, + "grad_norm": 0.0908203125, + "learning_rate": 0.0017633596692062396, + "loss": 0.084, + "step": 23441 + }, + { + "epoch": 0.20348781694603346, + "grad_norm": 0.283203125, + "learning_rate": 0.0017633395547668497, + "loss": 0.1338, + "step": 23442 + }, + { + "epoch": 0.20349649742623763, + "grad_norm": 0.181640625, + "learning_rate": 0.0017633194396020414, + "loss": 0.0996, + "step": 23443 + }, + { + "epoch": 0.2035051779064418, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017632993237118371, + "loss": 0.1074, + "step": 23444 + }, + { + "epoch": 0.20351385838664596, + "grad_norm": 0.4921875, + "learning_rate": 0.0017632792070962585, + "loss": 0.123, + "step": 23445 + }, + { + "epoch": 0.20352253886685012, + "grad_norm": 0.42578125, + "learning_rate": 0.0017632590897553276, + "loss": 0.0898, + "step": 23446 + }, + { + "epoch": 0.2035312193470543, + "grad_norm": 0.189453125, + "learning_rate": 0.0017632389716890668, + "loss": 0.1133, + "step": 23447 + }, + { + "epoch": 0.20353989982725845, + "grad_norm": 0.06884765625, + "learning_rate": 0.0017632188528974974, + "loss": 0.0815, + "step": 23448 + }, + { + "epoch": 0.20354858030746262, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017631987333806423, + "loss": 0.1182, + "step": 23449 + }, + { + "epoch": 0.20355726078766678, + "grad_norm": 0.375, + "learning_rate": 0.0017631786131385228, + "loss": 0.1079, + "step": 23450 + }, + { + "epoch": 0.20356594126787095, + "grad_norm": 0.126953125, + "learning_rate": 0.001763158492171161, + "loss": 0.1338, + "step": 23451 + }, + { + "epoch": 0.2035746217480751, + "grad_norm": 0.55078125, + "learning_rate": 0.0017631383704785793, + "loss": 0.0913, + "step": 23452 + }, + { + "epoch": 0.20358330222827928, + "grad_norm": 0.53125, + "learning_rate": 0.0017631182480607993, + "loss": 0.0938, + "step": 23453 + }, + { + "epoch": 0.20359198270848344, + "grad_norm": 0.1474609375, + "learning_rate": 0.0017630981249178433, + "loss": 0.1143, + "step": 23454 + }, + { + "epoch": 0.2036006631886876, + "grad_norm": 0.625, + "learning_rate": 0.001763078001049733, + "loss": 0.1309, + "step": 23455 + }, + { + "epoch": 0.20360934366889177, + "grad_norm": 0.609375, + "learning_rate": 0.0017630578764564904, + "loss": 0.1201, + "step": 23456 + }, + { + "epoch": 0.20361802414909594, + "grad_norm": 0.1435546875, + "learning_rate": 0.001763037751138138, + "loss": 0.1289, + "step": 23457 + }, + { + "epoch": 0.2036267046293001, + "grad_norm": 0.08154296875, + "learning_rate": 0.0017630176250946971, + "loss": 0.1152, + "step": 23458 + }, + { + "epoch": 0.20363538510950427, + "grad_norm": 0.359375, + "learning_rate": 0.0017629974983261905, + "loss": 0.1367, + "step": 23459 + }, + { + "epoch": 0.20364406558970843, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017629773708326398, + "loss": 0.0898, + "step": 23460 + }, + { + "epoch": 0.2036527460699126, + "grad_norm": 0.1181640625, + "learning_rate": 0.0017629572426140667, + "loss": 0.1094, + "step": 23461 + }, + { + "epoch": 0.20366142655011676, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017629371136704934, + "loss": 0.084, + "step": 23462 + }, + { + "epoch": 0.20367010703032093, + "grad_norm": 0.44921875, + "learning_rate": 0.0017629169840019427, + "loss": 0.1177, + "step": 23463 + }, + { + "epoch": 0.2036787875105251, + "grad_norm": 0.13671875, + "learning_rate": 0.0017628968536084354, + "loss": 0.1475, + "step": 23464 + }, + { + "epoch": 0.20368746799072926, + "grad_norm": 0.310546875, + "learning_rate": 0.0017628767224899943, + "loss": 0.1367, + "step": 23465 + }, + { + "epoch": 0.20369614847093342, + "grad_norm": 0.10791015625, + "learning_rate": 0.001762856590646641, + "loss": 0.124, + "step": 23466 + }, + { + "epoch": 0.2037048289511376, + "grad_norm": 0.8125, + "learning_rate": 0.001762836458078398, + "loss": 0.1328, + "step": 23467 + }, + { + "epoch": 0.20371350943134175, + "grad_norm": 0.21875, + "learning_rate": 0.0017628163247852865, + "loss": 0.1367, + "step": 23468 + }, + { + "epoch": 0.20372218991154592, + "grad_norm": 0.232421875, + "learning_rate": 0.001762796190767329, + "loss": 0.1108, + "step": 23469 + }, + { + "epoch": 0.20373087039175009, + "grad_norm": 0.330078125, + "learning_rate": 0.001762776056024548, + "loss": 0.1299, + "step": 23470 + }, + { + "epoch": 0.20373955087195425, + "grad_norm": 0.255859375, + "learning_rate": 0.001762755920556965, + "loss": 0.0903, + "step": 23471 + }, + { + "epoch": 0.2037482313521584, + "grad_norm": 0.28515625, + "learning_rate": 0.0017627357843646022, + "loss": 0.1504, + "step": 23472 + }, + { + "epoch": 0.20375691183236255, + "grad_norm": 0.337890625, + "learning_rate": 0.001762715647447481, + "loss": 0.1035, + "step": 23473 + }, + { + "epoch": 0.20376559231256672, + "grad_norm": 0.07421875, + "learning_rate": 0.0017626955098056243, + "loss": 0.0825, + "step": 23474 + }, + { + "epoch": 0.20377427279277088, + "grad_norm": 0.3828125, + "learning_rate": 0.0017626753714390538, + "loss": 0.1182, + "step": 23475 + }, + { + "epoch": 0.20378295327297505, + "grad_norm": 0.359375, + "learning_rate": 0.0017626552323477913, + "loss": 0.1099, + "step": 23476 + }, + { + "epoch": 0.2037916337531792, + "grad_norm": 0.255859375, + "learning_rate": 0.001762635092531859, + "loss": 0.1025, + "step": 23477 + }, + { + "epoch": 0.20380031423338338, + "grad_norm": 0.1103515625, + "learning_rate": 0.001762614951991279, + "loss": 0.1064, + "step": 23478 + }, + { + "epoch": 0.20380899471358754, + "grad_norm": 0.333984375, + "learning_rate": 0.0017625948107260732, + "loss": 0.1152, + "step": 23479 + }, + { + "epoch": 0.2038176751937917, + "grad_norm": 0.162109375, + "learning_rate": 0.0017625746687362637, + "loss": 0.1025, + "step": 23480 + }, + { + "epoch": 0.20382635567399587, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017625545260218725, + "loss": 0.1162, + "step": 23481 + }, + { + "epoch": 0.20383503615420004, + "grad_norm": 0.251953125, + "learning_rate": 0.0017625343825829211, + "loss": 0.0928, + "step": 23482 + }, + { + "epoch": 0.2038437166344042, + "grad_norm": 0.15625, + "learning_rate": 0.0017625142384194328, + "loss": 0.1377, + "step": 23483 + }, + { + "epoch": 0.20385239711460837, + "grad_norm": 0.130859375, + "learning_rate": 0.0017624940935314286, + "loss": 0.1475, + "step": 23484 + }, + { + "epoch": 0.20386107759481253, + "grad_norm": 0.1826171875, + "learning_rate": 0.001762473947918931, + "loss": 0.1064, + "step": 23485 + }, + { + "epoch": 0.2038697580750167, + "grad_norm": 0.3046875, + "learning_rate": 0.0017624538015819616, + "loss": 0.124, + "step": 23486 + }, + { + "epoch": 0.20387843855522086, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017624336545205427, + "loss": 0.1094, + "step": 23487 + }, + { + "epoch": 0.20388711903542503, + "grad_norm": 0.16796875, + "learning_rate": 0.0017624135067346965, + "loss": 0.1211, + "step": 23488 + }, + { + "epoch": 0.2038957995156292, + "grad_norm": 0.3671875, + "learning_rate": 0.0017623933582244446, + "loss": 0.105, + "step": 23489 + }, + { + "epoch": 0.20390447999583336, + "grad_norm": 0.91796875, + "learning_rate": 0.0017623732089898094, + "loss": 0.106, + "step": 23490 + }, + { + "epoch": 0.20391316047603752, + "grad_norm": 0.2275390625, + "learning_rate": 0.0017623530590308127, + "loss": 0.1006, + "step": 23491 + }, + { + "epoch": 0.2039218409562417, + "grad_norm": 0.357421875, + "learning_rate": 0.0017623329083474768, + "loss": 0.126, + "step": 23492 + }, + { + "epoch": 0.20393052143644586, + "grad_norm": 0.283203125, + "learning_rate": 0.0017623127569398234, + "loss": 0.1226, + "step": 23493 + }, + { + "epoch": 0.20393920191665002, + "grad_norm": 0.7421875, + "learning_rate": 0.001762292604807875, + "loss": 0.1064, + "step": 23494 + }, + { + "epoch": 0.20394788239685419, + "grad_norm": 0.58203125, + "learning_rate": 0.001762272451951653, + "loss": 0.1279, + "step": 23495 + }, + { + "epoch": 0.20395656287705835, + "grad_norm": 1.4765625, + "learning_rate": 0.0017622522983711802, + "loss": 0.1123, + "step": 23496 + }, + { + "epoch": 0.20396524335726252, + "grad_norm": 0.390625, + "learning_rate": 0.001762232144066478, + "loss": 0.1162, + "step": 23497 + }, + { + "epoch": 0.20397392383746668, + "grad_norm": 0.625, + "learning_rate": 0.0017622119890375687, + "loss": 0.1621, + "step": 23498 + }, + { + "epoch": 0.20398260431767085, + "grad_norm": 0.30078125, + "learning_rate": 0.0017621918332844742, + "loss": 0.1172, + "step": 23499 + }, + { + "epoch": 0.203991284797875, + "grad_norm": 0.25390625, + "learning_rate": 0.001762171676807217, + "loss": 0.1201, + "step": 23500 + }, + { + "epoch": 0.20399996527807918, + "grad_norm": 0.126953125, + "learning_rate": 0.0017621515196058187, + "loss": 0.1484, + "step": 23501 + }, + { + "epoch": 0.20400864575828334, + "grad_norm": 0.66015625, + "learning_rate": 0.0017621313616803013, + "loss": 0.1074, + "step": 23502 + }, + { + "epoch": 0.2040173262384875, + "grad_norm": 0.08984375, + "learning_rate": 0.001762111203030687, + "loss": 0.1016, + "step": 23503 + }, + { + "epoch": 0.20402600671869167, + "grad_norm": 0.74609375, + "learning_rate": 0.0017620910436569979, + "loss": 0.0977, + "step": 23504 + }, + { + "epoch": 0.20403468719889584, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017620708835592558, + "loss": 0.084, + "step": 23505 + }, + { + "epoch": 0.2040433676791, + "grad_norm": 0.412109375, + "learning_rate": 0.001762050722737483, + "loss": 0.0977, + "step": 23506 + }, + { + "epoch": 0.20405204815930417, + "grad_norm": 0.19921875, + "learning_rate": 0.0017620305611917018, + "loss": 0.1104, + "step": 23507 + }, + { + "epoch": 0.20406072863950833, + "grad_norm": 1.1015625, + "learning_rate": 0.0017620103989219335, + "loss": 0.1133, + "step": 23508 + }, + { + "epoch": 0.2040694091197125, + "grad_norm": 0.578125, + "learning_rate": 0.001761990235928201, + "loss": 0.1143, + "step": 23509 + }, + { + "epoch": 0.20407808959991666, + "grad_norm": 0.099609375, + "learning_rate": 0.0017619700722105254, + "loss": 0.1143, + "step": 23510 + }, + { + "epoch": 0.20408677008012083, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017619499077689295, + "loss": 0.1191, + "step": 23511 + }, + { + "epoch": 0.204095450560325, + "grad_norm": 0.90625, + "learning_rate": 0.0017619297426034353, + "loss": 0.1445, + "step": 23512 + }, + { + "epoch": 0.20410413104052916, + "grad_norm": 0.6484375, + "learning_rate": 0.0017619095767140644, + "loss": 0.0967, + "step": 23513 + }, + { + "epoch": 0.20411281152073332, + "grad_norm": 0.1484375, + "learning_rate": 0.0017618894101008393, + "loss": 0.1221, + "step": 23514 + }, + { + "epoch": 0.2041214920009375, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017618692427637818, + "loss": 0.0913, + "step": 23515 + }, + { + "epoch": 0.20413017248114165, + "grad_norm": 0.291015625, + "learning_rate": 0.0017618490747029147, + "loss": 0.0835, + "step": 23516 + }, + { + "epoch": 0.20413885296134582, + "grad_norm": 0.431640625, + "learning_rate": 0.0017618289059182584, + "loss": 0.085, + "step": 23517 + }, + { + "epoch": 0.20414753344154998, + "grad_norm": 0.201171875, + "learning_rate": 0.0017618087364098364, + "loss": 0.1006, + "step": 23518 + }, + { + "epoch": 0.20415621392175415, + "grad_norm": 0.359375, + "learning_rate": 0.0017617885661776703, + "loss": 0.104, + "step": 23519 + }, + { + "epoch": 0.2041648944019583, + "grad_norm": 0.1796875, + "learning_rate": 0.0017617683952217823, + "loss": 0.1123, + "step": 23520 + }, + { + "epoch": 0.20417357488216248, + "grad_norm": 0.76953125, + "learning_rate": 0.001761748223542194, + "loss": 0.0986, + "step": 23521 + }, + { + "epoch": 0.20418225536236664, + "grad_norm": 0.390625, + "learning_rate": 0.0017617280511389281, + "loss": 0.1123, + "step": 23522 + }, + { + "epoch": 0.2041909358425708, + "grad_norm": 0.65234375, + "learning_rate": 0.001761707878012006, + "loss": 0.1348, + "step": 23523 + }, + { + "epoch": 0.20419961632277497, + "grad_norm": 0.087890625, + "learning_rate": 0.0017616877041614503, + "loss": 0.0825, + "step": 23524 + }, + { + "epoch": 0.20420829680297914, + "grad_norm": 1.03125, + "learning_rate": 0.001761667529587283, + "loss": 0.0991, + "step": 23525 + }, + { + "epoch": 0.2042169772831833, + "grad_norm": 0.265625, + "learning_rate": 0.0017616473542895258, + "loss": 0.125, + "step": 23526 + }, + { + "epoch": 0.20422565776338747, + "grad_norm": 0.384765625, + "learning_rate": 0.0017616271782682012, + "loss": 0.1211, + "step": 23527 + }, + { + "epoch": 0.20423433824359163, + "grad_norm": 0.33203125, + "learning_rate": 0.001761607001523331, + "loss": 0.1143, + "step": 23528 + }, + { + "epoch": 0.2042430187237958, + "grad_norm": 0.2060546875, + "learning_rate": 0.001761586824054937, + "loss": 0.1631, + "step": 23529 + }, + { + "epoch": 0.20425169920399996, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017615666458630419, + "loss": 0.0718, + "step": 23530 + }, + { + "epoch": 0.20426037968420413, + "grad_norm": 0.59765625, + "learning_rate": 0.0017615464669476675, + "loss": 0.1157, + "step": 23531 + }, + { + "epoch": 0.2042690601644083, + "grad_norm": 0.70703125, + "learning_rate": 0.0017615262873088358, + "loss": 0.1216, + "step": 23532 + }, + { + "epoch": 0.20427774064461246, + "grad_norm": 0.421875, + "learning_rate": 0.001761506106946569, + "loss": 0.0991, + "step": 23533 + }, + { + "epoch": 0.20428642112481663, + "grad_norm": 0.11865234375, + "learning_rate": 0.0017614859258608887, + "loss": 0.1055, + "step": 23534 + }, + { + "epoch": 0.2042951016050208, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017614657440518176, + "loss": 0.1201, + "step": 23535 + }, + { + "epoch": 0.20430378208522496, + "grad_norm": 0.98046875, + "learning_rate": 0.0017614455615193774, + "loss": 0.0942, + "step": 23536 + }, + { + "epoch": 0.20431246256542912, + "grad_norm": 0.302734375, + "learning_rate": 0.0017614253782635902, + "loss": 0.0918, + "step": 23537 + }, + { + "epoch": 0.20432114304563329, + "grad_norm": 0.53125, + "learning_rate": 0.0017614051942844784, + "loss": 0.0732, + "step": 23538 + }, + { + "epoch": 0.20432982352583745, + "grad_norm": 0.53125, + "learning_rate": 0.0017613850095820635, + "loss": 0.1406, + "step": 23539 + }, + { + "epoch": 0.20433850400604162, + "grad_norm": 0.421875, + "learning_rate": 0.0017613648241563682, + "loss": 0.1445, + "step": 23540 + }, + { + "epoch": 0.20434718448624578, + "grad_norm": 0.6953125, + "learning_rate": 0.0017613446380074142, + "loss": 0.0889, + "step": 23541 + }, + { + "epoch": 0.20435586496644995, + "grad_norm": 0.240234375, + "learning_rate": 0.0017613244511352235, + "loss": 0.1426, + "step": 23542 + }, + { + "epoch": 0.2043645454466541, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017613042635398187, + "loss": 0.0864, + "step": 23543 + }, + { + "epoch": 0.20437322592685828, + "grad_norm": 1.03125, + "learning_rate": 0.0017612840752212209, + "loss": 0.1914, + "step": 23544 + }, + { + "epoch": 0.20438190640706244, + "grad_norm": 0.42578125, + "learning_rate": 0.001761263886179453, + "loss": 0.1128, + "step": 23545 + }, + { + "epoch": 0.2043905868872666, + "grad_norm": 0.1396484375, + "learning_rate": 0.0017612436964145372, + "loss": 0.1387, + "step": 23546 + }, + { + "epoch": 0.20439926736747077, + "grad_norm": 0.212890625, + "learning_rate": 0.0017612235059264946, + "loss": 0.1025, + "step": 23547 + }, + { + "epoch": 0.20440794784767494, + "grad_norm": 0.4140625, + "learning_rate": 0.0017612033147153484, + "loss": 0.1357, + "step": 23548 + }, + { + "epoch": 0.2044166283278791, + "grad_norm": 0.56640625, + "learning_rate": 0.0017611831227811198, + "loss": 0.1016, + "step": 23549 + }, + { + "epoch": 0.20442530880808327, + "grad_norm": 0.091796875, + "learning_rate": 0.0017611629301238316, + "loss": 0.1143, + "step": 23550 + }, + { + "epoch": 0.20443398928828743, + "grad_norm": 0.41796875, + "learning_rate": 0.0017611427367435054, + "loss": 0.1152, + "step": 23551 + }, + { + "epoch": 0.2044426697684916, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017611225426401637, + "loss": 0.1128, + "step": 23552 + }, + { + "epoch": 0.20445135024869576, + "grad_norm": 0.1953125, + "learning_rate": 0.001761102347813828, + "loss": 0.1021, + "step": 23553 + }, + { + "epoch": 0.20446003072889993, + "grad_norm": 0.462890625, + "learning_rate": 0.0017610821522645208, + "loss": 0.1055, + "step": 23554 + }, + { + "epoch": 0.2044687112091041, + "grad_norm": 0.08642578125, + "learning_rate": 0.001761061955992264, + "loss": 0.1133, + "step": 23555 + }, + { + "epoch": 0.20447739168930826, + "grad_norm": 0.5546875, + "learning_rate": 0.0017610417589970801, + "loss": 0.1396, + "step": 23556 + }, + { + "epoch": 0.20448607216951242, + "grad_norm": 0.83203125, + "learning_rate": 0.0017610215612789903, + "loss": 0.1143, + "step": 23557 + }, + { + "epoch": 0.2044947526497166, + "grad_norm": 0.8125, + "learning_rate": 0.001761001362838018, + "loss": 0.3965, + "step": 23558 + }, + { + "epoch": 0.20450343312992075, + "grad_norm": 0.2177734375, + "learning_rate": 0.001760981163674184, + "loss": 0.1299, + "step": 23559 + }, + { + "epoch": 0.20451211361012492, + "grad_norm": 0.408203125, + "learning_rate": 0.001760960963787511, + "loss": 0.0801, + "step": 23560 + }, + { + "epoch": 0.20452079409032908, + "grad_norm": 0.205078125, + "learning_rate": 0.001760940763178021, + "loss": 0.123, + "step": 23561 + }, + { + "epoch": 0.20452947457053325, + "grad_norm": 0.3515625, + "learning_rate": 0.0017609205618457364, + "loss": 0.0791, + "step": 23562 + }, + { + "epoch": 0.2045381550507374, + "grad_norm": 0.376953125, + "learning_rate": 0.0017609003597906787, + "loss": 0.0889, + "step": 23563 + }, + { + "epoch": 0.20454683553094158, + "grad_norm": 0.25, + "learning_rate": 0.0017608801570128702, + "loss": 0.1016, + "step": 23564 + }, + { + "epoch": 0.20455551601114574, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017608599535123336, + "loss": 0.106, + "step": 23565 + }, + { + "epoch": 0.2045641964913499, + "grad_norm": 0.12353515625, + "learning_rate": 0.00176083974928909, + "loss": 0.1016, + "step": 23566 + }, + { + "epoch": 0.20457287697155407, + "grad_norm": 0.130859375, + "learning_rate": 0.0017608195443431624, + "loss": 0.1196, + "step": 23567 + }, + { + "epoch": 0.20458155745175824, + "grad_norm": 0.302734375, + "learning_rate": 0.0017607993386745723, + "loss": 0.082, + "step": 23568 + }, + { + "epoch": 0.2045902379319624, + "grad_norm": 0.330078125, + "learning_rate": 0.0017607791322833417, + "loss": 0.1338, + "step": 23569 + }, + { + "epoch": 0.20459891841216657, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017607589251694934, + "loss": 0.1221, + "step": 23570 + }, + { + "epoch": 0.20460759889237073, + "grad_norm": 0.1015625, + "learning_rate": 0.0017607387173330486, + "loss": 0.0957, + "step": 23571 + }, + { + "epoch": 0.2046162793725749, + "grad_norm": 0.259765625, + "learning_rate": 0.0017607185087740302, + "loss": 0.1069, + "step": 23572 + }, + { + "epoch": 0.20462495985277906, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017606982994924597, + "loss": 0.1377, + "step": 23573 + }, + { + "epoch": 0.20463364033298323, + "grad_norm": 0.126953125, + "learning_rate": 0.0017606780894883595, + "loss": 0.124, + "step": 23574 + }, + { + "epoch": 0.2046423208131874, + "grad_norm": 0.38671875, + "learning_rate": 0.001760657878761752, + "loss": 0.0869, + "step": 23575 + }, + { + "epoch": 0.20465100129339156, + "grad_norm": 0.126953125, + "learning_rate": 0.0017606376673126587, + "loss": 0.1104, + "step": 23576 + }, + { + "epoch": 0.20465968177359573, + "grad_norm": 1.3203125, + "learning_rate": 0.001760617455141102, + "loss": 0.126, + "step": 23577 + }, + { + "epoch": 0.2046683622537999, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017605972422471041, + "loss": 0.123, + "step": 23578 + }, + { + "epoch": 0.20467704273400406, + "grad_norm": 0.447265625, + "learning_rate": 0.0017605770286306867, + "loss": 0.124, + "step": 23579 + }, + { + "epoch": 0.20468572321420822, + "grad_norm": 0.173828125, + "learning_rate": 0.0017605568142918726, + "loss": 0.1035, + "step": 23580 + }, + { + "epoch": 0.20469440369441239, + "grad_norm": 0.25, + "learning_rate": 0.0017605365992306833, + "loss": 0.0972, + "step": 23581 + }, + { + "epoch": 0.20470308417461655, + "grad_norm": 0.30859375, + "learning_rate": 0.0017605163834471412, + "loss": 0.1055, + "step": 23582 + }, + { + "epoch": 0.20471176465482072, + "grad_norm": 0.11572265625, + "learning_rate": 0.0017604961669412682, + "loss": 0.123, + "step": 23583 + }, + { + "epoch": 0.20472044513502488, + "grad_norm": 0.07861328125, + "learning_rate": 0.0017604759497130864, + "loss": 0.0962, + "step": 23584 + }, + { + "epoch": 0.20472912561522905, + "grad_norm": 0.3125, + "learning_rate": 0.0017604557317626181, + "loss": 0.1001, + "step": 23585 + }, + { + "epoch": 0.2047378060954332, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017604355130898855, + "loss": 0.1172, + "step": 23586 + }, + { + "epoch": 0.20474648657563738, + "grad_norm": 0.37890625, + "learning_rate": 0.0017604152936949104, + "loss": 0.1152, + "step": 23587 + }, + { + "epoch": 0.20475516705584154, + "grad_norm": 0.1533203125, + "learning_rate": 0.001760395073577715, + "loss": 0.1182, + "step": 23588 + }, + { + "epoch": 0.2047638475360457, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017603748527383217, + "loss": 0.0977, + "step": 23589 + }, + { + "epoch": 0.20477252801624987, + "grad_norm": 0.640625, + "learning_rate": 0.001760354631176752, + "loss": 0.127, + "step": 23590 + }, + { + "epoch": 0.20478120849645404, + "grad_norm": 0.70703125, + "learning_rate": 0.0017603344088930287, + "loss": 0.1064, + "step": 23591 + }, + { + "epoch": 0.2047898889766582, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017603141858871733, + "loss": 0.0967, + "step": 23592 + }, + { + "epoch": 0.20479856945686237, + "grad_norm": 0.34765625, + "learning_rate": 0.0017602939621592087, + "loss": 0.082, + "step": 23593 + }, + { + "epoch": 0.20480724993706653, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017602737377091563, + "loss": 0.0967, + "step": 23594 + }, + { + "epoch": 0.20481593041727067, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017602535125370385, + "loss": 0.1182, + "step": 23595 + }, + { + "epoch": 0.20482461089747483, + "grad_norm": 0.251953125, + "learning_rate": 0.0017602332866428773, + "loss": 0.1064, + "step": 23596 + }, + { + "epoch": 0.204833291377679, + "grad_norm": 0.388671875, + "learning_rate": 0.0017602130600266948, + "loss": 0.1074, + "step": 23597 + }, + { + "epoch": 0.20484197185788316, + "grad_norm": 0.78125, + "learning_rate": 0.0017601928326885135, + "loss": 0.1689, + "step": 23598 + }, + { + "epoch": 0.20485065233808733, + "grad_norm": 0.16015625, + "learning_rate": 0.0017601726046283551, + "loss": 0.1123, + "step": 23599 + }, + { + "epoch": 0.2048593328182915, + "grad_norm": 0.158203125, + "learning_rate": 0.001760152375846242, + "loss": 0.0928, + "step": 23600 + }, + { + "epoch": 0.20486801329849566, + "grad_norm": 0.25, + "learning_rate": 0.001760132146342196, + "loss": 0.1001, + "step": 23601 + }, + { + "epoch": 0.20487669377869983, + "grad_norm": 0.271484375, + "learning_rate": 0.0017601119161162393, + "loss": 0.1953, + "step": 23602 + }, + { + "epoch": 0.204885374258904, + "grad_norm": 0.2216796875, + "learning_rate": 0.0017600916851683943, + "loss": 0.0879, + "step": 23603 + }, + { + "epoch": 0.20489405473910816, + "grad_norm": 0.859375, + "learning_rate": 0.001760071453498683, + "loss": 0.0928, + "step": 23604 + }, + { + "epoch": 0.20490273521931232, + "grad_norm": 0.31640625, + "learning_rate": 0.0017600512211071273, + "loss": 0.0703, + "step": 23605 + }, + { + "epoch": 0.20491141569951649, + "grad_norm": 0.283203125, + "learning_rate": 0.0017600309879937494, + "loss": 0.105, + "step": 23606 + }, + { + "epoch": 0.20492009617972065, + "grad_norm": 0.2197265625, + "learning_rate": 0.0017600107541585717, + "loss": 0.0659, + "step": 23607 + }, + { + "epoch": 0.20492877665992482, + "grad_norm": 0.46875, + "learning_rate": 0.0017599905196016164, + "loss": 0.1104, + "step": 23608 + }, + { + "epoch": 0.20493745714012898, + "grad_norm": 0.244140625, + "learning_rate": 0.001759970284322905, + "loss": 0.0952, + "step": 23609 + }, + { + "epoch": 0.20494613762033315, + "grad_norm": 0.201171875, + "learning_rate": 0.0017599500483224603, + "loss": 0.0947, + "step": 23610 + }, + { + "epoch": 0.2049548181005373, + "grad_norm": 0.44921875, + "learning_rate": 0.0017599298116003042, + "loss": 0.1172, + "step": 23611 + }, + { + "epoch": 0.20496349858074148, + "grad_norm": 0.9453125, + "learning_rate": 0.0017599095741564583, + "loss": 0.0908, + "step": 23612 + }, + { + "epoch": 0.20497217906094564, + "grad_norm": 0.263671875, + "learning_rate": 0.0017598893359909455, + "loss": 0.1108, + "step": 23613 + }, + { + "epoch": 0.2049808595411498, + "grad_norm": 0.671875, + "learning_rate": 0.0017598690971037877, + "loss": 0.1045, + "step": 23614 + }, + { + "epoch": 0.20498954002135397, + "grad_norm": 0.298828125, + "learning_rate": 0.001759848857495007, + "loss": 0.1035, + "step": 23615 + }, + { + "epoch": 0.20499822050155814, + "grad_norm": 0.453125, + "learning_rate": 0.0017598286171646256, + "loss": 0.1143, + "step": 23616 + }, + { + "epoch": 0.2050069009817623, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017598083761126652, + "loss": 0.0806, + "step": 23617 + }, + { + "epoch": 0.20501558146196647, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017597881343391483, + "loss": 0.1123, + "step": 23618 + }, + { + "epoch": 0.20502426194217063, + "grad_norm": 0.318359375, + "learning_rate": 0.0017597678918440972, + "loss": 0.1279, + "step": 23619 + }, + { + "epoch": 0.2050329424223748, + "grad_norm": 0.95703125, + "learning_rate": 0.0017597476486275337, + "loss": 0.1221, + "step": 23620 + }, + { + "epoch": 0.20504162290257896, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017597274046894803, + "loss": 0.1328, + "step": 23621 + }, + { + "epoch": 0.20505030338278313, + "grad_norm": 0.7421875, + "learning_rate": 0.0017597071600299586, + "loss": 0.1074, + "step": 23622 + }, + { + "epoch": 0.2050589838629873, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017596869146489914, + "loss": 0.105, + "step": 23623 + }, + { + "epoch": 0.20506766434319146, + "grad_norm": 0.546875, + "learning_rate": 0.0017596666685466002, + "loss": 0.1289, + "step": 23624 + }, + { + "epoch": 0.20507634482339562, + "grad_norm": 0.359375, + "learning_rate": 0.0017596464217228076, + "loss": 0.1226, + "step": 23625 + }, + { + "epoch": 0.2050850253035998, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017596261741776357, + "loss": 0.1406, + "step": 23626 + }, + { + "epoch": 0.20509370578380395, + "grad_norm": 0.1201171875, + "learning_rate": 0.0017596059259111065, + "loss": 0.1201, + "step": 23627 + }, + { + "epoch": 0.20510238626400812, + "grad_norm": 1.7578125, + "learning_rate": 0.0017595856769232422, + "loss": 0.1465, + "step": 23628 + }, + { + "epoch": 0.20511106674421228, + "grad_norm": 0.12109375, + "learning_rate": 0.0017595654272140647, + "loss": 0.1045, + "step": 23629 + }, + { + "epoch": 0.20511974722441645, + "grad_norm": 0.0859375, + "learning_rate": 0.0017595451767835962, + "loss": 0.0835, + "step": 23630 + }, + { + "epoch": 0.20512842770462061, + "grad_norm": 0.3203125, + "learning_rate": 0.0017595249256318593, + "loss": 0.0952, + "step": 23631 + }, + { + "epoch": 0.20513710818482478, + "grad_norm": 0.6328125, + "learning_rate": 0.001759504673758876, + "loss": 0.1025, + "step": 23632 + }, + { + "epoch": 0.20514578866502894, + "grad_norm": 0.59375, + "learning_rate": 0.001759484421164668, + "loss": 0.1426, + "step": 23633 + }, + { + "epoch": 0.2051544691452331, + "grad_norm": 0.13671875, + "learning_rate": 0.0017594641678492578, + "loss": 0.123, + "step": 23634 + }, + { + "epoch": 0.20516314962543727, + "grad_norm": 0.546875, + "learning_rate": 0.0017594439138126678, + "loss": 0.1348, + "step": 23635 + }, + { + "epoch": 0.20517183010564144, + "grad_norm": 0.65625, + "learning_rate": 0.0017594236590549193, + "loss": 0.2178, + "step": 23636 + }, + { + "epoch": 0.2051805105858456, + "grad_norm": 0.1884765625, + "learning_rate": 0.001759403403576035, + "loss": 0.1064, + "step": 23637 + }, + { + "epoch": 0.20518919106604977, + "grad_norm": 0.359375, + "learning_rate": 0.0017593831473760374, + "loss": 0.0903, + "step": 23638 + }, + { + "epoch": 0.20519787154625393, + "grad_norm": 0.283203125, + "learning_rate": 0.0017593628904549485, + "loss": 0.1143, + "step": 23639 + }, + { + "epoch": 0.2052065520264581, + "grad_norm": 0.462890625, + "learning_rate": 0.00175934263281279, + "loss": 0.1143, + "step": 23640 + }, + { + "epoch": 0.20521523250666227, + "grad_norm": 0.33984375, + "learning_rate": 0.0017593223744495842, + "loss": 0.1387, + "step": 23641 + }, + { + "epoch": 0.20522391298686643, + "grad_norm": 0.255859375, + "learning_rate": 0.0017593021153653533, + "loss": 0.1045, + "step": 23642 + }, + { + "epoch": 0.2052325934670706, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017592818555601198, + "loss": 0.0933, + "step": 23643 + }, + { + "epoch": 0.20524127394727476, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017592615950339054, + "loss": 0.1328, + "step": 23644 + }, + { + "epoch": 0.20524995442747893, + "grad_norm": 0.349609375, + "learning_rate": 0.0017592413337867321, + "loss": 0.0947, + "step": 23645 + }, + { + "epoch": 0.2052586349076831, + "grad_norm": 0.388671875, + "learning_rate": 0.0017592210718186227, + "loss": 0.1133, + "step": 23646 + }, + { + "epoch": 0.20526731538788726, + "grad_norm": 0.41796875, + "learning_rate": 0.0017592008091295995, + "loss": 0.1846, + "step": 23647 + }, + { + "epoch": 0.20527599586809142, + "grad_norm": 0.48828125, + "learning_rate": 0.0017591805457196834, + "loss": 0.1094, + "step": 23648 + }, + { + "epoch": 0.20528467634829559, + "grad_norm": 0.251953125, + "learning_rate": 0.001759160281588898, + "loss": 0.0967, + "step": 23649 + }, + { + "epoch": 0.20529335682849975, + "grad_norm": 0.7265625, + "learning_rate": 0.0017591400167372643, + "loss": 0.1006, + "step": 23650 + }, + { + "epoch": 0.20530203730870392, + "grad_norm": 0.21484375, + "learning_rate": 0.0017591197511648052, + "loss": 0.1426, + "step": 23651 + }, + { + "epoch": 0.20531071778890808, + "grad_norm": 0.59765625, + "learning_rate": 0.0017590994848715427, + "loss": 0.1064, + "step": 23652 + }, + { + "epoch": 0.20531939826911225, + "grad_norm": 0.65234375, + "learning_rate": 0.0017590792178574987, + "loss": 0.103, + "step": 23653 + }, + { + "epoch": 0.2053280787493164, + "grad_norm": 0.8515625, + "learning_rate": 0.0017590589501226962, + "loss": 0.127, + "step": 23654 + }, + { + "epoch": 0.20533675922952058, + "grad_norm": 0.875, + "learning_rate": 0.0017590386816671562, + "loss": 0.1123, + "step": 23655 + }, + { + "epoch": 0.20534543970972474, + "grad_norm": 0.232421875, + "learning_rate": 0.0017590184124909012, + "loss": 0.1533, + "step": 23656 + }, + { + "epoch": 0.2053541201899289, + "grad_norm": 0.3984375, + "learning_rate": 0.001758998142593954, + "loss": 0.1143, + "step": 23657 + }, + { + "epoch": 0.20536280067013307, + "grad_norm": 0.3359375, + "learning_rate": 0.001758977871976336, + "loss": 0.0898, + "step": 23658 + }, + { + "epoch": 0.20537148115033724, + "grad_norm": 0.63671875, + "learning_rate": 0.00175895760063807, + "loss": 0.1055, + "step": 23659 + }, + { + "epoch": 0.2053801616305414, + "grad_norm": 0.06494140625, + "learning_rate": 0.0017589373285791779, + "loss": 0.1025, + "step": 23660 + }, + { + "epoch": 0.20538884211074557, + "grad_norm": 0.22265625, + "learning_rate": 0.0017589170557996813, + "loss": 0.0967, + "step": 23661 + }, + { + "epoch": 0.20539752259094973, + "grad_norm": 0.357421875, + "learning_rate": 0.0017588967822996036, + "loss": 0.1133, + "step": 23662 + }, + { + "epoch": 0.2054062030711539, + "grad_norm": 0.8515625, + "learning_rate": 0.0017588765080789658, + "loss": 0.1348, + "step": 23663 + }, + { + "epoch": 0.20541488355135806, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017588562331377906, + "loss": 0.2002, + "step": 23664 + }, + { + "epoch": 0.20542356403156223, + "grad_norm": 0.11181640625, + "learning_rate": 0.0017588359574761004, + "loss": 0.0938, + "step": 23665 + }, + { + "epoch": 0.2054322445117664, + "grad_norm": 0.365234375, + "learning_rate": 0.0017588156810939169, + "loss": 0.0938, + "step": 23666 + }, + { + "epoch": 0.20544092499197056, + "grad_norm": 0.291015625, + "learning_rate": 0.0017587954039912626, + "loss": 0.106, + "step": 23667 + }, + { + "epoch": 0.20544960547217472, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017587751261681592, + "loss": 0.1123, + "step": 23668 + }, + { + "epoch": 0.2054582859523789, + "grad_norm": 0.341796875, + "learning_rate": 0.0017587548476246295, + "loss": 0.1484, + "step": 23669 + }, + { + "epoch": 0.20546696643258305, + "grad_norm": 0.1796875, + "learning_rate": 0.0017587345683606954, + "loss": 0.127, + "step": 23670 + }, + { + "epoch": 0.20547564691278722, + "grad_norm": 0.07666015625, + "learning_rate": 0.0017587142883763792, + "loss": 0.1064, + "step": 23671 + }, + { + "epoch": 0.20548432739299138, + "grad_norm": 0.173828125, + "learning_rate": 0.0017586940076717028, + "loss": 0.1152, + "step": 23672 + }, + { + "epoch": 0.20549300787319555, + "grad_norm": 0.21484375, + "learning_rate": 0.0017586737262466883, + "loss": 0.1309, + "step": 23673 + }, + { + "epoch": 0.20550168835339971, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017586534441013583, + "loss": 0.1211, + "step": 23674 + }, + { + "epoch": 0.20551036883360388, + "grad_norm": 0.271484375, + "learning_rate": 0.001758633161235735, + "loss": 0.1035, + "step": 23675 + }, + { + "epoch": 0.20551904931380804, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017586128776498403, + "loss": 0.1108, + "step": 23676 + }, + { + "epoch": 0.2055277297940122, + "grad_norm": 0.193359375, + "learning_rate": 0.0017585925933436962, + "loss": 0.1387, + "step": 23677 + }, + { + "epoch": 0.20553641027421637, + "grad_norm": 0.28125, + "learning_rate": 0.0017585723083173254, + "loss": 0.1147, + "step": 23678 + }, + { + "epoch": 0.20554509075442054, + "grad_norm": 0.259765625, + "learning_rate": 0.0017585520225707493, + "loss": 0.0771, + "step": 23679 + }, + { + "epoch": 0.2055537712346247, + "grad_norm": 0.248046875, + "learning_rate": 0.0017585317361039912, + "loss": 0.0854, + "step": 23680 + }, + { + "epoch": 0.20556245171482887, + "grad_norm": 0.18359375, + "learning_rate": 0.0017585114489170723, + "loss": 0.105, + "step": 23681 + }, + { + "epoch": 0.20557113219503304, + "grad_norm": 0.640625, + "learning_rate": 0.0017584911610100157, + "loss": 0.1367, + "step": 23682 + }, + { + "epoch": 0.2055798126752372, + "grad_norm": 0.263671875, + "learning_rate": 0.0017584708723828424, + "loss": 0.0781, + "step": 23683 + }, + { + "epoch": 0.20558849315544137, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017584505830355757, + "loss": 0.0908, + "step": 23684 + }, + { + "epoch": 0.20559717363564553, + "grad_norm": 0.48828125, + "learning_rate": 0.001758430292968237, + "loss": 0.0898, + "step": 23685 + }, + { + "epoch": 0.2056058541158497, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017584100021808492, + "loss": 0.0869, + "step": 23686 + }, + { + "epoch": 0.20561453459605386, + "grad_norm": 0.0908203125, + "learning_rate": 0.0017583897106734341, + "loss": 0.1011, + "step": 23687 + }, + { + "epoch": 0.20562321507625803, + "grad_norm": 0.41796875, + "learning_rate": 0.0017583694184460138, + "loss": 0.0811, + "step": 23688 + }, + { + "epoch": 0.2056318955564622, + "grad_norm": 0.08154296875, + "learning_rate": 0.0017583491254986104, + "loss": 0.084, + "step": 23689 + }, + { + "epoch": 0.20564057603666636, + "grad_norm": 0.21484375, + "learning_rate": 0.0017583288318312464, + "loss": 0.1191, + "step": 23690 + }, + { + "epoch": 0.20564925651687052, + "grad_norm": 0.166015625, + "learning_rate": 0.001758308537443944, + "loss": 0.1738, + "step": 23691 + }, + { + "epoch": 0.2056579369970747, + "grad_norm": 0.4296875, + "learning_rate": 0.0017582882423367253, + "loss": 0.1094, + "step": 23692 + }, + { + "epoch": 0.20566661747727885, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017582679465096122, + "loss": 0.0913, + "step": 23693 + }, + { + "epoch": 0.20567529795748302, + "grad_norm": 0.251953125, + "learning_rate": 0.0017582476499626272, + "loss": 0.1104, + "step": 23694 + }, + { + "epoch": 0.20568397843768718, + "grad_norm": 0.51953125, + "learning_rate": 0.0017582273526957925, + "loss": 0.1074, + "step": 23695 + }, + { + "epoch": 0.20569265891789135, + "grad_norm": 0.11328125, + "learning_rate": 0.0017582070547091304, + "loss": 0.1016, + "step": 23696 + }, + { + "epoch": 0.2057013393980955, + "grad_norm": 0.1396484375, + "learning_rate": 0.001758186756002663, + "loss": 0.0918, + "step": 23697 + }, + { + "epoch": 0.20571001987829968, + "grad_norm": 0.3984375, + "learning_rate": 0.0017581664565764123, + "loss": 0.1001, + "step": 23698 + }, + { + "epoch": 0.20571870035850384, + "grad_norm": 0.4453125, + "learning_rate": 0.0017581461564304007, + "loss": 0.0864, + "step": 23699 + }, + { + "epoch": 0.205727380838708, + "grad_norm": 0.88671875, + "learning_rate": 0.0017581258555646503, + "loss": 0.1426, + "step": 23700 + }, + { + "epoch": 0.20573606131891217, + "grad_norm": 0.6015625, + "learning_rate": 0.0017581055539791835, + "loss": 0.1196, + "step": 23701 + }, + { + "epoch": 0.20574474179911634, + "grad_norm": 0.412109375, + "learning_rate": 0.0017580852516740223, + "loss": 0.1006, + "step": 23702 + }, + { + "epoch": 0.2057534222793205, + "grad_norm": 0.33984375, + "learning_rate": 0.0017580649486491887, + "loss": 0.0762, + "step": 23703 + }, + { + "epoch": 0.20576210275952467, + "grad_norm": 0.2001953125, + "learning_rate": 0.0017580446449047054, + "loss": 0.0928, + "step": 23704 + }, + { + "epoch": 0.20577078323972883, + "grad_norm": 0.29296875, + "learning_rate": 0.0017580243404405945, + "loss": 0.1523, + "step": 23705 + }, + { + "epoch": 0.205779463719933, + "grad_norm": 0.181640625, + "learning_rate": 0.0017580040352568778, + "loss": 0.103, + "step": 23706 + }, + { + "epoch": 0.20578814420013716, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017579837293535778, + "loss": 0.1201, + "step": 23707 + }, + { + "epoch": 0.20579682468034133, + "grad_norm": 0.1708984375, + "learning_rate": 0.001757963422730717, + "loss": 0.0874, + "step": 23708 + }, + { + "epoch": 0.2058055051605455, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017579431153883168, + "loss": 0.1416, + "step": 23709 + }, + { + "epoch": 0.20581418564074966, + "grad_norm": 0.412109375, + "learning_rate": 0.0017579228073264003, + "loss": 0.0942, + "step": 23710 + }, + { + "epoch": 0.20582286612095382, + "grad_norm": 0.1279296875, + "learning_rate": 0.001757902498544989, + "loss": 0.0923, + "step": 23711 + }, + { + "epoch": 0.205831546601158, + "grad_norm": 0.255859375, + "learning_rate": 0.0017578821890441056, + "loss": 0.1191, + "step": 23712 + }, + { + "epoch": 0.20584022708136215, + "grad_norm": 0.373046875, + "learning_rate": 0.0017578618788237719, + "loss": 0.0879, + "step": 23713 + }, + { + "epoch": 0.20584890756156632, + "grad_norm": 0.466796875, + "learning_rate": 0.0017578415678840104, + "loss": 0.1245, + "step": 23714 + }, + { + "epoch": 0.20585758804177048, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017578212562248435, + "loss": 0.1621, + "step": 23715 + }, + { + "epoch": 0.20586626852197465, + "grad_norm": 0.10546875, + "learning_rate": 0.0017578009438462931, + "loss": 0.1426, + "step": 23716 + }, + { + "epoch": 0.20587494900217881, + "grad_norm": 0.2373046875, + "learning_rate": 0.0017577806307483815, + "loss": 0.1211, + "step": 23717 + }, + { + "epoch": 0.20588362948238295, + "grad_norm": 0.46875, + "learning_rate": 0.0017577603169311306, + "loss": 0.0781, + "step": 23718 + }, + { + "epoch": 0.20589230996258712, + "grad_norm": 0.1005859375, + "learning_rate": 0.001757740002394563, + "loss": 0.0957, + "step": 23719 + }, + { + "epoch": 0.20590099044279128, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017577196871387012, + "loss": 0.0693, + "step": 23720 + }, + { + "epoch": 0.20590967092299545, + "grad_norm": 0.162109375, + "learning_rate": 0.0017576993711635664, + "loss": 0.1079, + "step": 23721 + }, + { + "epoch": 0.2059183514031996, + "grad_norm": 0.46875, + "learning_rate": 0.0017576790544691816, + "loss": 0.0864, + "step": 23722 + }, + { + "epoch": 0.20592703188340378, + "grad_norm": 0.447265625, + "learning_rate": 0.0017576587370555692, + "loss": 0.0762, + "step": 23723 + }, + { + "epoch": 0.20593571236360794, + "grad_norm": 0.50390625, + "learning_rate": 0.0017576384189227508, + "loss": 0.1182, + "step": 23724 + }, + { + "epoch": 0.2059443928438121, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017576181000707493, + "loss": 0.0938, + "step": 23725 + }, + { + "epoch": 0.20595307332401627, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017575977804995863, + "loss": 0.1235, + "step": 23726 + }, + { + "epoch": 0.20596175380422044, + "grad_norm": 0.349609375, + "learning_rate": 0.001757577460209284, + "loss": 0.1118, + "step": 23727 + }, + { + "epoch": 0.2059704342844246, + "grad_norm": 0.2578125, + "learning_rate": 0.0017575571391998648, + "loss": 0.1309, + "step": 23728 + }, + { + "epoch": 0.20597911476462877, + "grad_norm": 0.7265625, + "learning_rate": 0.0017575368174713515, + "loss": 0.1074, + "step": 23729 + }, + { + "epoch": 0.20598779524483293, + "grad_norm": 0.357421875, + "learning_rate": 0.0017575164950237659, + "loss": 0.123, + "step": 23730 + }, + { + "epoch": 0.2059964757250371, + "grad_norm": 1.1328125, + "learning_rate": 0.0017574961718571295, + "loss": 0.127, + "step": 23731 + }, + { + "epoch": 0.20600515620524126, + "grad_norm": 0.28125, + "learning_rate": 0.0017574758479714655, + "loss": 0.0962, + "step": 23732 + }, + { + "epoch": 0.20601383668544543, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017574555233667956, + "loss": 0.0928, + "step": 23733 + }, + { + "epoch": 0.2060225171656496, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017574351980431425, + "loss": 0.1016, + "step": 23734 + }, + { + "epoch": 0.20603119764585376, + "grad_norm": 0.2001953125, + "learning_rate": 0.001757414872000528, + "loss": 0.1152, + "step": 23735 + }, + { + "epoch": 0.20603987812605792, + "grad_norm": 0.8046875, + "learning_rate": 0.0017573945452389746, + "loss": 0.1025, + "step": 23736 + }, + { + "epoch": 0.2060485586062621, + "grad_norm": 0.310546875, + "learning_rate": 0.0017573742177585043, + "loss": 0.0806, + "step": 23737 + }, + { + "epoch": 0.20605723908646625, + "grad_norm": 0.310546875, + "learning_rate": 0.0017573538895591394, + "loss": 0.1279, + "step": 23738 + }, + { + "epoch": 0.20606591956667042, + "grad_norm": 0.28515625, + "learning_rate": 0.0017573335606409022, + "loss": 0.0815, + "step": 23739 + }, + { + "epoch": 0.20607460004687458, + "grad_norm": 0.69921875, + "learning_rate": 0.001757313231003815, + "loss": 0.0889, + "step": 23740 + }, + { + "epoch": 0.20608328052707875, + "grad_norm": 1.078125, + "learning_rate": 0.0017572929006478997, + "loss": 0.2949, + "step": 23741 + }, + { + "epoch": 0.20609196100728291, + "grad_norm": 0.11572265625, + "learning_rate": 0.0017572725695731788, + "loss": 0.0825, + "step": 23742 + }, + { + "epoch": 0.20610064148748708, + "grad_norm": 0.53125, + "learning_rate": 0.001757252237779675, + "loss": 0.0918, + "step": 23743 + }, + { + "epoch": 0.20610932196769124, + "grad_norm": 0.36328125, + "learning_rate": 0.0017572319052674094, + "loss": 0.0972, + "step": 23744 + }, + { + "epoch": 0.2061180024478954, + "grad_norm": 0.099609375, + "learning_rate": 0.001757211572036405, + "loss": 0.0947, + "step": 23745 + }, + { + "epoch": 0.20612668292809957, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017571912380866842, + "loss": 0.1719, + "step": 23746 + }, + { + "epoch": 0.20613536340830374, + "grad_norm": 0.333984375, + "learning_rate": 0.0017571709034182685, + "loss": 0.1055, + "step": 23747 + }, + { + "epoch": 0.2061440438885079, + "grad_norm": 0.310546875, + "learning_rate": 0.0017571505680311808, + "loss": 0.0723, + "step": 23748 + }, + { + "epoch": 0.20615272436871207, + "grad_norm": 0.5234375, + "learning_rate": 0.0017571302319254432, + "loss": 0.0957, + "step": 23749 + }, + { + "epoch": 0.20616140484891624, + "grad_norm": 0.294921875, + "learning_rate": 0.0017571098951010776, + "loss": 0.1035, + "step": 23750 + }, + { + "epoch": 0.2061700853291204, + "grad_norm": 0.67578125, + "learning_rate": 0.0017570895575581068, + "loss": 0.1377, + "step": 23751 + }, + { + "epoch": 0.20617876580932457, + "grad_norm": 0.13671875, + "learning_rate": 0.0017570692192965523, + "loss": 0.1055, + "step": 23752 + }, + { + "epoch": 0.20618744628952873, + "grad_norm": 0.08447265625, + "learning_rate": 0.0017570488803164374, + "loss": 0.0996, + "step": 23753 + }, + { + "epoch": 0.2061961267697329, + "grad_norm": 0.435546875, + "learning_rate": 0.001757028540617783, + "loss": 0.2344, + "step": 23754 + }, + { + "epoch": 0.20620480724993706, + "grad_norm": 0.193359375, + "learning_rate": 0.0017570082002006124, + "loss": 0.1016, + "step": 23755 + }, + { + "epoch": 0.20621348773014123, + "grad_norm": 0.1640625, + "learning_rate": 0.0017569878590649477, + "loss": 0.0967, + "step": 23756 + }, + { + "epoch": 0.2062221682103454, + "grad_norm": 0.287109375, + "learning_rate": 0.0017569675172108108, + "loss": 0.1147, + "step": 23757 + }, + { + "epoch": 0.20623084869054956, + "grad_norm": 0.294921875, + "learning_rate": 0.001756947174638224, + "loss": 0.1118, + "step": 23758 + }, + { + "epoch": 0.20623952917075372, + "grad_norm": 1.1640625, + "learning_rate": 0.0017569268313472098, + "loss": 0.1699, + "step": 23759 + }, + { + "epoch": 0.2062482096509579, + "grad_norm": 0.11376953125, + "learning_rate": 0.00175690648733779, + "loss": 0.1084, + "step": 23760 + }, + { + "epoch": 0.20625689013116205, + "grad_norm": 0.498046875, + "learning_rate": 0.0017568861426099877, + "loss": 0.0967, + "step": 23761 + }, + { + "epoch": 0.20626557061136622, + "grad_norm": 0.236328125, + "learning_rate": 0.001756865797163824, + "loss": 0.104, + "step": 23762 + }, + { + "epoch": 0.20627425109157038, + "grad_norm": 0.9375, + "learning_rate": 0.0017568454509993218, + "loss": 0.1543, + "step": 23763 + }, + { + "epoch": 0.20628293157177455, + "grad_norm": 0.2470703125, + "learning_rate": 0.0017568251041165034, + "loss": 0.0913, + "step": 23764 + }, + { + "epoch": 0.2062916120519787, + "grad_norm": 0.63671875, + "learning_rate": 0.001756804756515391, + "loss": 0.1562, + "step": 23765 + }, + { + "epoch": 0.20630029253218288, + "grad_norm": 0.10498046875, + "learning_rate": 0.0017567844081960067, + "loss": 0.0986, + "step": 23766 + }, + { + "epoch": 0.20630897301238704, + "grad_norm": 0.87890625, + "learning_rate": 0.0017567640591583731, + "loss": 0.1182, + "step": 23767 + }, + { + "epoch": 0.2063176534925912, + "grad_norm": 0.7734375, + "learning_rate": 0.0017567437094025118, + "loss": 0.0967, + "step": 23768 + }, + { + "epoch": 0.20632633397279537, + "grad_norm": 0.189453125, + "learning_rate": 0.0017567233589284455, + "loss": 0.0859, + "step": 23769 + }, + { + "epoch": 0.20633501445299954, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017567030077361967, + "loss": 0.0771, + "step": 23770 + }, + { + "epoch": 0.2063436949332037, + "grad_norm": 0.6015625, + "learning_rate": 0.0017566826558257872, + "loss": 0.1035, + "step": 23771 + }, + { + "epoch": 0.20635237541340787, + "grad_norm": 0.34765625, + "learning_rate": 0.0017566623031972394, + "loss": 0.1064, + "step": 23772 + }, + { + "epoch": 0.20636105589361203, + "grad_norm": 0.390625, + "learning_rate": 0.0017566419498505758, + "loss": 0.1235, + "step": 23773 + }, + { + "epoch": 0.2063697363738162, + "grad_norm": 0.1953125, + "learning_rate": 0.0017566215957858181, + "loss": 0.1172, + "step": 23774 + }, + { + "epoch": 0.20637841685402036, + "grad_norm": 0.41015625, + "learning_rate": 0.001756601241002989, + "loss": 0.1011, + "step": 23775 + }, + { + "epoch": 0.20638709733422453, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017565808855021106, + "loss": 0.0986, + "step": 23776 + }, + { + "epoch": 0.2063957778144287, + "grad_norm": 0.625, + "learning_rate": 0.0017565605292832053, + "loss": 0.1318, + "step": 23777 + }, + { + "epoch": 0.20640445829463286, + "grad_norm": 0.162109375, + "learning_rate": 0.001756540172346295, + "loss": 0.1094, + "step": 23778 + }, + { + "epoch": 0.20641313877483702, + "grad_norm": 0.46875, + "learning_rate": 0.0017565198146914025, + "loss": 0.0806, + "step": 23779 + }, + { + "epoch": 0.2064218192550412, + "grad_norm": 1.5078125, + "learning_rate": 0.0017564994563185497, + "loss": 0.2578, + "step": 23780 + }, + { + "epoch": 0.20643049973524535, + "grad_norm": 0.384765625, + "learning_rate": 0.0017564790972277593, + "loss": 0.1001, + "step": 23781 + }, + { + "epoch": 0.20643918021544952, + "grad_norm": 0.30859375, + "learning_rate": 0.0017564587374190525, + "loss": 0.127, + "step": 23782 + }, + { + "epoch": 0.20644786069565368, + "grad_norm": 0.06982421875, + "learning_rate": 0.001756438376892453, + "loss": 0.0718, + "step": 23783 + }, + { + "epoch": 0.20645654117585785, + "grad_norm": 0.1611328125, + "learning_rate": 0.0017564180156479819, + "loss": 0.0903, + "step": 23784 + }, + { + "epoch": 0.20646522165606201, + "grad_norm": 0.255859375, + "learning_rate": 0.0017563976536856623, + "loss": 0.124, + "step": 23785 + }, + { + "epoch": 0.20647390213626618, + "grad_norm": 0.4921875, + "learning_rate": 0.0017563772910055158, + "loss": 0.0977, + "step": 23786 + }, + { + "epoch": 0.20648258261647034, + "grad_norm": 0.490234375, + "learning_rate": 0.0017563569276075651, + "loss": 0.0752, + "step": 23787 + }, + { + "epoch": 0.2064912630966745, + "grad_norm": 0.67578125, + "learning_rate": 0.0017563365634918322, + "loss": 0.0957, + "step": 23788 + }, + { + "epoch": 0.20649994357687868, + "grad_norm": 0.78515625, + "learning_rate": 0.0017563161986583395, + "loss": 0.1162, + "step": 23789 + }, + { + "epoch": 0.20650862405708284, + "grad_norm": 0.279296875, + "learning_rate": 0.0017562958331071095, + "loss": 0.0859, + "step": 23790 + }, + { + "epoch": 0.206517304537287, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017562754668381637, + "loss": 0.0874, + "step": 23791 + }, + { + "epoch": 0.20652598501749117, + "grad_norm": 0.177734375, + "learning_rate": 0.0017562550998515257, + "loss": 0.0977, + "step": 23792 + }, + { + "epoch": 0.20653466549769534, + "grad_norm": 0.62109375, + "learning_rate": 0.0017562347321472163, + "loss": 0.1157, + "step": 23793 + }, + { + "epoch": 0.2065433459778995, + "grad_norm": 0.20703125, + "learning_rate": 0.0017562143637252588, + "loss": 0.1221, + "step": 23794 + }, + { + "epoch": 0.20655202645810367, + "grad_norm": 0.474609375, + "learning_rate": 0.001756193994585675, + "loss": 0.1099, + "step": 23795 + }, + { + "epoch": 0.20656070693830783, + "grad_norm": 0.298828125, + "learning_rate": 0.0017561736247284875, + "loss": 0.1191, + "step": 23796 + }, + { + "epoch": 0.206569387418512, + "grad_norm": 0.314453125, + "learning_rate": 0.0017561532541537181, + "loss": 0.1006, + "step": 23797 + }, + { + "epoch": 0.20657806789871616, + "grad_norm": 0.0810546875, + "learning_rate": 0.0017561328828613896, + "loss": 0.1157, + "step": 23798 + }, + { + "epoch": 0.20658674837892033, + "grad_norm": 0.11181640625, + "learning_rate": 0.0017561125108515238, + "loss": 0.085, + "step": 23799 + }, + { + "epoch": 0.2065954288591245, + "grad_norm": 0.130859375, + "learning_rate": 0.0017560921381241435, + "loss": 0.1162, + "step": 23800 + }, + { + "epoch": 0.20660410933932866, + "grad_norm": 0.46875, + "learning_rate": 0.0017560717646792703, + "loss": 0.1074, + "step": 23801 + }, + { + "epoch": 0.20661278981953282, + "grad_norm": 0.216796875, + "learning_rate": 0.0017560513905169274, + "loss": 0.1064, + "step": 23802 + }, + { + "epoch": 0.206621470299737, + "grad_norm": 0.65234375, + "learning_rate": 0.001756031015637136, + "loss": 0.1191, + "step": 23803 + }, + { + "epoch": 0.20663015077994115, + "grad_norm": 0.7421875, + "learning_rate": 0.0017560106400399194, + "loss": 0.1406, + "step": 23804 + }, + { + "epoch": 0.20663883126014532, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017559902637252992, + "loss": 0.1172, + "step": 23805 + }, + { + "epoch": 0.20664751174034948, + "grad_norm": 0.1640625, + "learning_rate": 0.0017559698866932978, + "loss": 0.0859, + "step": 23806 + }, + { + "epoch": 0.20665619222055365, + "grad_norm": 0.12890625, + "learning_rate": 0.0017559495089439377, + "loss": 0.106, + "step": 23807 + }, + { + "epoch": 0.2066648727007578, + "grad_norm": 0.26953125, + "learning_rate": 0.001755929130477241, + "loss": 0.1357, + "step": 23808 + }, + { + "epoch": 0.20667355318096198, + "grad_norm": 0.15234375, + "learning_rate": 0.0017559087512932306, + "loss": 0.1182, + "step": 23809 + }, + { + "epoch": 0.20668223366116614, + "grad_norm": 0.48828125, + "learning_rate": 0.0017558883713919276, + "loss": 0.126, + "step": 23810 + }, + { + "epoch": 0.2066909141413703, + "grad_norm": 0.52734375, + "learning_rate": 0.0017558679907733552, + "loss": 0.1045, + "step": 23811 + }, + { + "epoch": 0.20669959462157447, + "grad_norm": 0.224609375, + "learning_rate": 0.0017558476094375352, + "loss": 0.124, + "step": 23812 + }, + { + "epoch": 0.20670827510177864, + "grad_norm": 0.095703125, + "learning_rate": 0.0017558272273844903, + "loss": 0.124, + "step": 23813 + }, + { + "epoch": 0.2067169555819828, + "grad_norm": 1.5625, + "learning_rate": 0.0017558068446142426, + "loss": 0.2383, + "step": 23814 + }, + { + "epoch": 0.20672563606218697, + "grad_norm": 0.796875, + "learning_rate": 0.0017557864611268143, + "loss": 0.1309, + "step": 23815 + }, + { + "epoch": 0.20673431654239113, + "grad_norm": 0.220703125, + "learning_rate": 0.0017557660769222277, + "loss": 0.1025, + "step": 23816 + }, + { + "epoch": 0.2067429970225953, + "grad_norm": 0.236328125, + "learning_rate": 0.0017557456920005055, + "loss": 0.0825, + "step": 23817 + }, + { + "epoch": 0.20675167750279946, + "grad_norm": 0.439453125, + "learning_rate": 0.0017557253063616694, + "loss": 0.1157, + "step": 23818 + }, + { + "epoch": 0.20676035798300363, + "grad_norm": 0.26953125, + "learning_rate": 0.001755704920005742, + "loss": 0.1709, + "step": 23819 + }, + { + "epoch": 0.2067690384632078, + "grad_norm": 0.58984375, + "learning_rate": 0.0017556845329327457, + "loss": 0.1094, + "step": 23820 + }, + { + "epoch": 0.20677771894341196, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017556641451427022, + "loss": 0.1177, + "step": 23821 + }, + { + "epoch": 0.20678639942361612, + "grad_norm": 0.154296875, + "learning_rate": 0.0017556437566356349, + "loss": 0.127, + "step": 23822 + }, + { + "epoch": 0.2067950799038203, + "grad_norm": 0.5703125, + "learning_rate": 0.001755623367411565, + "loss": 0.1211, + "step": 23823 + }, + { + "epoch": 0.20680376038402445, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017556029774705153, + "loss": 0.1406, + "step": 23824 + }, + { + "epoch": 0.20681244086422862, + "grad_norm": 0.1484375, + "learning_rate": 0.0017555825868125082, + "loss": 0.1104, + "step": 23825 + }, + { + "epoch": 0.20682112134443278, + "grad_norm": 0.34765625, + "learning_rate": 0.0017555621954375657, + "loss": 0.1094, + "step": 23826 + }, + { + "epoch": 0.20682980182463695, + "grad_norm": 0.796875, + "learning_rate": 0.00175554180334571, + "loss": 0.1016, + "step": 23827 + }, + { + "epoch": 0.20683848230484111, + "grad_norm": 0.2490234375, + "learning_rate": 0.001755521410536964, + "loss": 0.0742, + "step": 23828 + }, + { + "epoch": 0.20684716278504528, + "grad_norm": 0.56640625, + "learning_rate": 0.0017555010170113492, + "loss": 0.1152, + "step": 23829 + }, + { + "epoch": 0.20685584326524944, + "grad_norm": 0.109375, + "learning_rate": 0.0017554806227688888, + "loss": 0.1328, + "step": 23830 + }, + { + "epoch": 0.2068645237454536, + "grad_norm": 0.1474609375, + "learning_rate": 0.0017554602278096049, + "loss": 0.1465, + "step": 23831 + }, + { + "epoch": 0.20687320422565778, + "grad_norm": 0.28515625, + "learning_rate": 0.0017554398321335187, + "loss": 0.1011, + "step": 23832 + }, + { + "epoch": 0.20688188470586194, + "grad_norm": 0.400390625, + "learning_rate": 0.0017554194357406539, + "loss": 0.0972, + "step": 23833 + }, + { + "epoch": 0.2068905651860661, + "grad_norm": 0.6953125, + "learning_rate": 0.001755399038631032, + "loss": 0.1309, + "step": 23834 + }, + { + "epoch": 0.20689924566627027, + "grad_norm": 0.58984375, + "learning_rate": 0.0017553786408046757, + "loss": 0.1289, + "step": 23835 + }, + { + "epoch": 0.20690792614647444, + "grad_norm": 0.88671875, + "learning_rate": 0.0017553582422616073, + "loss": 0.1143, + "step": 23836 + }, + { + "epoch": 0.2069166066266786, + "grad_norm": 0.34375, + "learning_rate": 0.0017553378430018487, + "loss": 0.1748, + "step": 23837 + }, + { + "epoch": 0.20692528710688277, + "grad_norm": 0.71875, + "learning_rate": 0.0017553174430254227, + "loss": 0.1099, + "step": 23838 + }, + { + "epoch": 0.20693396758708693, + "grad_norm": 0.146484375, + "learning_rate": 0.001755297042332351, + "loss": 0.1318, + "step": 23839 + }, + { + "epoch": 0.2069426480672911, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017552766409226568, + "loss": 0.0947, + "step": 23840 + }, + { + "epoch": 0.20695132854749526, + "grad_norm": 0.55078125, + "learning_rate": 0.0017552562387963618, + "loss": 0.1445, + "step": 23841 + }, + { + "epoch": 0.2069600090276994, + "grad_norm": 0.5625, + "learning_rate": 0.001755235835953488, + "loss": 0.1309, + "step": 23842 + }, + { + "epoch": 0.20696868950790356, + "grad_norm": 0.1044921875, + "learning_rate": 0.0017552154323940586, + "loss": 0.1152, + "step": 23843 + }, + { + "epoch": 0.20697736998810773, + "grad_norm": 0.55078125, + "learning_rate": 0.0017551950281180953, + "loss": 0.1582, + "step": 23844 + }, + { + "epoch": 0.2069860504683119, + "grad_norm": 0.39453125, + "learning_rate": 0.0017551746231256205, + "loss": 0.126, + "step": 23845 + }, + { + "epoch": 0.20699473094851606, + "grad_norm": 0.173828125, + "learning_rate": 0.0017551542174166566, + "loss": 0.1289, + "step": 23846 + }, + { + "epoch": 0.20700341142872022, + "grad_norm": 0.1572265625, + "learning_rate": 0.001755133810991226, + "loss": 0.0845, + "step": 23847 + }, + { + "epoch": 0.2070120919089244, + "grad_norm": 0.3125, + "learning_rate": 0.0017551134038493509, + "loss": 0.126, + "step": 23848 + }, + { + "epoch": 0.20702077238912855, + "grad_norm": 0.318359375, + "learning_rate": 0.0017550929959910535, + "loss": 0.1035, + "step": 23849 + }, + { + "epoch": 0.20702945286933272, + "grad_norm": 0.08349609375, + "learning_rate": 0.0017550725874163563, + "loss": 0.125, + "step": 23850 + }, + { + "epoch": 0.20703813334953688, + "grad_norm": 1.203125, + "learning_rate": 0.0017550521781252815, + "loss": 0.1084, + "step": 23851 + }, + { + "epoch": 0.20704681382974105, + "grad_norm": 0.486328125, + "learning_rate": 0.0017550317681178517, + "loss": 0.1201, + "step": 23852 + }, + { + "epoch": 0.20705549430994521, + "grad_norm": 0.33984375, + "learning_rate": 0.0017550113573940888, + "loss": 0.1104, + "step": 23853 + }, + { + "epoch": 0.20706417479014938, + "grad_norm": 0.73046875, + "learning_rate": 0.0017549909459540155, + "loss": 0.126, + "step": 23854 + }, + { + "epoch": 0.20707285527035355, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017549705337976537, + "loss": 0.1074, + "step": 23855 + }, + { + "epoch": 0.2070815357505577, + "grad_norm": 0.51171875, + "learning_rate": 0.0017549501209250262, + "loss": 0.1104, + "step": 23856 + }, + { + "epoch": 0.20709021623076188, + "grad_norm": 0.16796875, + "learning_rate": 0.001754929707336155, + "loss": 0.0854, + "step": 23857 + }, + { + "epoch": 0.20709889671096604, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017549092930310621, + "loss": 0.1201, + "step": 23858 + }, + { + "epoch": 0.2071075771911702, + "grad_norm": 0.0869140625, + "learning_rate": 0.001754888878009771, + "loss": 0.1094, + "step": 23859 + }, + { + "epoch": 0.20711625767137437, + "grad_norm": 0.1796875, + "learning_rate": 0.0017548684622723026, + "loss": 0.1377, + "step": 23860 + }, + { + "epoch": 0.20712493815157854, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017548480458186805, + "loss": 0.1045, + "step": 23861 + }, + { + "epoch": 0.2071336186317827, + "grad_norm": 0.51171875, + "learning_rate": 0.001754827628648926, + "loss": 0.0742, + "step": 23862 + }, + { + "epoch": 0.20714229911198687, + "grad_norm": 0.31640625, + "learning_rate": 0.0017548072107630619, + "loss": 0.0918, + "step": 23863 + }, + { + "epoch": 0.20715097959219103, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017547867921611105, + "loss": 0.1348, + "step": 23864 + }, + { + "epoch": 0.2071596600723952, + "grad_norm": 0.265625, + "learning_rate": 0.0017547663728430941, + "loss": 0.1377, + "step": 23865 + }, + { + "epoch": 0.20716834055259936, + "grad_norm": 0.4453125, + "learning_rate": 0.0017547459528090348, + "loss": 0.1021, + "step": 23866 + }, + { + "epoch": 0.20717702103280353, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017547255320589557, + "loss": 0.0854, + "step": 23867 + }, + { + "epoch": 0.2071857015130077, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017547051105928782, + "loss": 0.1089, + "step": 23868 + }, + { + "epoch": 0.20719438199321186, + "grad_norm": 0.0908203125, + "learning_rate": 0.001754684688410825, + "loss": 0.1167, + "step": 23869 + }, + { + "epoch": 0.20720306247341602, + "grad_norm": 0.1142578125, + "learning_rate": 0.0017546642655128185, + "loss": 0.1475, + "step": 23870 + }, + { + "epoch": 0.2072117429536202, + "grad_norm": 0.2578125, + "learning_rate": 0.0017546438418988813, + "loss": 0.1138, + "step": 23871 + }, + { + "epoch": 0.20722042343382435, + "grad_norm": 0.21484375, + "learning_rate": 0.001754623417569035, + "loss": 0.1089, + "step": 23872 + }, + { + "epoch": 0.20722910391402852, + "grad_norm": 0.6015625, + "learning_rate": 0.0017546029925233028, + "loss": 0.1719, + "step": 23873 + }, + { + "epoch": 0.20723778439423268, + "grad_norm": 0.357421875, + "learning_rate": 0.0017545825667617061, + "loss": 0.0688, + "step": 23874 + }, + { + "epoch": 0.20724646487443685, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017545621402842682, + "loss": 0.124, + "step": 23875 + }, + { + "epoch": 0.207255145354641, + "grad_norm": 0.09619140625, + "learning_rate": 0.0017545417130910106, + "loss": 0.1177, + "step": 23876 + }, + { + "epoch": 0.20726382583484518, + "grad_norm": 0.6171875, + "learning_rate": 0.0017545212851819564, + "loss": 0.1113, + "step": 23877 + }, + { + "epoch": 0.20727250631504934, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017545008565571273, + "loss": 0.1445, + "step": 23878 + }, + { + "epoch": 0.2072811867952535, + "grad_norm": 0.1455078125, + "learning_rate": 0.001754480427216546, + "loss": 0.1279, + "step": 23879 + }, + { + "epoch": 0.20728986727545767, + "grad_norm": 0.1220703125, + "learning_rate": 0.0017544599971602347, + "loss": 0.1211, + "step": 23880 + }, + { + "epoch": 0.20729854775566184, + "grad_norm": 0.193359375, + "learning_rate": 0.0017544395663882158, + "loss": 0.1074, + "step": 23881 + }, + { + "epoch": 0.207307228235866, + "grad_norm": 0.09423828125, + "learning_rate": 0.0017544191349005115, + "loss": 0.1123, + "step": 23882 + }, + { + "epoch": 0.20731590871607017, + "grad_norm": 0.625, + "learning_rate": 0.0017543987026971445, + "loss": 0.1094, + "step": 23883 + }, + { + "epoch": 0.20732458919627433, + "grad_norm": 0.4140625, + "learning_rate": 0.0017543782697781367, + "loss": 0.0977, + "step": 23884 + }, + { + "epoch": 0.2073332696764785, + "grad_norm": 0.18359375, + "learning_rate": 0.0017543578361435106, + "loss": 0.1006, + "step": 23885 + }, + { + "epoch": 0.20734195015668266, + "grad_norm": 0.416015625, + "learning_rate": 0.0017543374017932887, + "loss": 0.1123, + "step": 23886 + }, + { + "epoch": 0.20735063063688683, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017543169667274933, + "loss": 0.1084, + "step": 23887 + }, + { + "epoch": 0.207359311117091, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017542965309461469, + "loss": 0.1133, + "step": 23888 + }, + { + "epoch": 0.20736799159729516, + "grad_norm": 0.455078125, + "learning_rate": 0.0017542760944492712, + "loss": 0.1074, + "step": 23889 + }, + { + "epoch": 0.20737667207749932, + "grad_norm": 0.4609375, + "learning_rate": 0.0017542556572368894, + "loss": 0.1367, + "step": 23890 + }, + { + "epoch": 0.2073853525577035, + "grad_norm": 0.58984375, + "learning_rate": 0.0017542352193090232, + "loss": 0.1035, + "step": 23891 + }, + { + "epoch": 0.20739403303790765, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017542147806656953, + "loss": 0.1089, + "step": 23892 + }, + { + "epoch": 0.20740271351811182, + "grad_norm": 0.609375, + "learning_rate": 0.001754194341306928, + "loss": 0.1367, + "step": 23893 + }, + { + "epoch": 0.20741139399831598, + "grad_norm": 0.076171875, + "learning_rate": 0.0017541739012327435, + "loss": 0.1157, + "step": 23894 + }, + { + "epoch": 0.20742007447852015, + "grad_norm": 0.380859375, + "learning_rate": 0.0017541534604431641, + "loss": 0.1582, + "step": 23895 + }, + { + "epoch": 0.20742875495872432, + "grad_norm": 0.66796875, + "learning_rate": 0.0017541330189382125, + "loss": 0.1035, + "step": 23896 + }, + { + "epoch": 0.20743743543892848, + "grad_norm": 0.1875, + "learning_rate": 0.001754112576717911, + "loss": 0.1543, + "step": 23897 + }, + { + "epoch": 0.20744611591913265, + "grad_norm": 0.12255859375, + "learning_rate": 0.0017540921337822815, + "loss": 0.1118, + "step": 23898 + }, + { + "epoch": 0.2074547963993368, + "grad_norm": 0.138671875, + "learning_rate": 0.0017540716901313469, + "loss": 0.1357, + "step": 23899 + }, + { + "epoch": 0.20746347687954098, + "grad_norm": 0.12890625, + "learning_rate": 0.0017540512457651291, + "loss": 0.0918, + "step": 23900 + }, + { + "epoch": 0.20747215735974514, + "grad_norm": 0.23828125, + "learning_rate": 0.0017540308006836506, + "loss": 0.0903, + "step": 23901 + }, + { + "epoch": 0.2074808378399493, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017540103548869342, + "loss": 0.1006, + "step": 23902 + }, + { + "epoch": 0.20748951832015347, + "grad_norm": 0.5078125, + "learning_rate": 0.001753989908375002, + "loss": 0.0947, + "step": 23903 + }, + { + "epoch": 0.20749819880035764, + "grad_norm": 0.291015625, + "learning_rate": 0.001753969461147876, + "loss": 0.1113, + "step": 23904 + }, + { + "epoch": 0.2075068792805618, + "grad_norm": 0.52734375, + "learning_rate": 0.0017539490132055789, + "loss": 0.1133, + "step": 23905 + }, + { + "epoch": 0.20751555976076597, + "grad_norm": 0.1337890625, + "learning_rate": 0.001753928564548133, + "loss": 0.1348, + "step": 23906 + }, + { + "epoch": 0.20752424024097013, + "grad_norm": 0.1220703125, + "learning_rate": 0.0017539081151755605, + "loss": 0.1035, + "step": 23907 + }, + { + "epoch": 0.2075329207211743, + "grad_norm": 0.1328125, + "learning_rate": 0.0017538876650878838, + "loss": 0.1035, + "step": 23908 + }, + { + "epoch": 0.20754160120137846, + "grad_norm": 0.2578125, + "learning_rate": 0.0017538672142851256, + "loss": 0.1143, + "step": 23909 + }, + { + "epoch": 0.20755028168158263, + "grad_norm": 0.47265625, + "learning_rate": 0.001753846762767308, + "loss": 0.1123, + "step": 23910 + }, + { + "epoch": 0.2075589621617868, + "grad_norm": 0.302734375, + "learning_rate": 0.0017538263105344534, + "loss": 0.127, + "step": 23911 + }, + { + "epoch": 0.20756764264199096, + "grad_norm": 0.134765625, + "learning_rate": 0.0017538058575865842, + "loss": 0.1084, + "step": 23912 + }, + { + "epoch": 0.20757632312219512, + "grad_norm": 0.19140625, + "learning_rate": 0.001753785403923723, + "loss": 0.1108, + "step": 23913 + }, + { + "epoch": 0.2075850036023993, + "grad_norm": 0.291015625, + "learning_rate": 0.0017537649495458915, + "loss": 0.0996, + "step": 23914 + }, + { + "epoch": 0.20759368408260345, + "grad_norm": 0.46484375, + "learning_rate": 0.0017537444944531121, + "loss": 0.1113, + "step": 23915 + }, + { + "epoch": 0.20760236456280762, + "grad_norm": 0.640625, + "learning_rate": 0.0017537240386454084, + "loss": 0.104, + "step": 23916 + }, + { + "epoch": 0.20761104504301178, + "grad_norm": 0.29296875, + "learning_rate": 0.0017537035821228015, + "loss": 0.1436, + "step": 23917 + }, + { + "epoch": 0.20761972552321595, + "grad_norm": 0.5390625, + "learning_rate": 0.0017536831248853144, + "loss": 0.082, + "step": 23918 + }, + { + "epoch": 0.2076284060034201, + "grad_norm": 0.0966796875, + "learning_rate": 0.001753662666932969, + "loss": 0.1309, + "step": 23919 + }, + { + "epoch": 0.20763708648362428, + "grad_norm": 0.162109375, + "learning_rate": 0.001753642208265788, + "loss": 0.0913, + "step": 23920 + }, + { + "epoch": 0.20764576696382844, + "grad_norm": 0.5390625, + "learning_rate": 0.0017536217488837936, + "loss": 0.1143, + "step": 23921 + }, + { + "epoch": 0.2076544474440326, + "grad_norm": 0.98046875, + "learning_rate": 0.0017536012887870084, + "loss": 0.0938, + "step": 23922 + }, + { + "epoch": 0.20766312792423677, + "grad_norm": 1.171875, + "learning_rate": 0.0017535808279754547, + "loss": 0.1348, + "step": 23923 + }, + { + "epoch": 0.20767180840444094, + "grad_norm": 0.109375, + "learning_rate": 0.0017535603664491548, + "loss": 0.1152, + "step": 23924 + }, + { + "epoch": 0.2076804888846451, + "grad_norm": 0.09521484375, + "learning_rate": 0.001753539904208131, + "loss": 0.0957, + "step": 23925 + }, + { + "epoch": 0.20768916936484927, + "grad_norm": 0.4375, + "learning_rate": 0.001753519441252406, + "loss": 0.1006, + "step": 23926 + }, + { + "epoch": 0.20769784984505343, + "grad_norm": 0.2275390625, + "learning_rate": 0.001753498977582002, + "loss": 0.1367, + "step": 23927 + }, + { + "epoch": 0.2077065303252576, + "grad_norm": 0.232421875, + "learning_rate": 0.0017534785131969408, + "loss": 0.0698, + "step": 23928 + }, + { + "epoch": 0.20771521080546176, + "grad_norm": 0.8125, + "learning_rate": 0.0017534580480972456, + "loss": 0.1211, + "step": 23929 + }, + { + "epoch": 0.20772389128566593, + "grad_norm": 0.43359375, + "learning_rate": 0.0017534375822829387, + "loss": 0.0986, + "step": 23930 + }, + { + "epoch": 0.2077325717658701, + "grad_norm": 0.376953125, + "learning_rate": 0.0017534171157540422, + "loss": 0.1133, + "step": 23931 + }, + { + "epoch": 0.20774125224607426, + "grad_norm": 0.78515625, + "learning_rate": 0.0017533966485105783, + "loss": 0.1318, + "step": 23932 + }, + { + "epoch": 0.20774993272627842, + "grad_norm": 0.333984375, + "learning_rate": 0.00175337618055257, + "loss": 0.0845, + "step": 23933 + }, + { + "epoch": 0.2077586132064826, + "grad_norm": 0.314453125, + "learning_rate": 0.0017533557118800391, + "loss": 0.1641, + "step": 23934 + }, + { + "epoch": 0.20776729368668675, + "grad_norm": 0.107421875, + "learning_rate": 0.0017533352424930085, + "loss": 0.127, + "step": 23935 + }, + { + "epoch": 0.20777597416689092, + "grad_norm": 0.158203125, + "learning_rate": 0.0017533147723915, + "loss": 0.1211, + "step": 23936 + }, + { + "epoch": 0.20778465464709509, + "grad_norm": 0.099609375, + "learning_rate": 0.0017532943015755363, + "loss": 0.0947, + "step": 23937 + }, + { + "epoch": 0.20779333512729925, + "grad_norm": 0.482421875, + "learning_rate": 0.00175327383004514, + "loss": 0.0908, + "step": 23938 + }, + { + "epoch": 0.20780201560750342, + "grad_norm": 0.25390625, + "learning_rate": 0.001753253357800333, + "loss": 0.0933, + "step": 23939 + }, + { + "epoch": 0.20781069608770758, + "grad_norm": 0.373046875, + "learning_rate": 0.001753232884841138, + "loss": 0.1836, + "step": 23940 + }, + { + "epoch": 0.20781937656791175, + "grad_norm": 0.15625, + "learning_rate": 0.0017532124111675773, + "loss": 0.0708, + "step": 23941 + }, + { + "epoch": 0.2078280570481159, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017531919367796734, + "loss": 0.125, + "step": 23942 + }, + { + "epoch": 0.20783673752832008, + "grad_norm": 0.109375, + "learning_rate": 0.0017531714616774486, + "loss": 0.1021, + "step": 23943 + }, + { + "epoch": 0.20784541800852424, + "grad_norm": 0.1181640625, + "learning_rate": 0.0017531509858609254, + "loss": 0.1079, + "step": 23944 + }, + { + "epoch": 0.2078540984887284, + "grad_norm": 0.2412109375, + "learning_rate": 0.001753130509330126, + "loss": 0.0859, + "step": 23945 + }, + { + "epoch": 0.20786277896893257, + "grad_norm": 0.212890625, + "learning_rate": 0.0017531100320850729, + "loss": 0.0732, + "step": 23946 + }, + { + "epoch": 0.20787145944913674, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017530895541257884, + "loss": 0.0977, + "step": 23947 + }, + { + "epoch": 0.2078801399293409, + "grad_norm": 0.478515625, + "learning_rate": 0.0017530690754522951, + "loss": 0.0894, + "step": 23948 + }, + { + "epoch": 0.20788882040954507, + "grad_norm": 0.43359375, + "learning_rate": 0.0017530485960646149, + "loss": 0.0947, + "step": 23949 + }, + { + "epoch": 0.20789750088974923, + "grad_norm": 0.08056640625, + "learning_rate": 0.001753028115962771, + "loss": 0.0757, + "step": 23950 + }, + { + "epoch": 0.2079061813699534, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017530076351467852, + "loss": 0.1133, + "step": 23951 + }, + { + "epoch": 0.20791486185015756, + "grad_norm": 0.62890625, + "learning_rate": 0.00175298715361668, + "loss": 0.083, + "step": 23952 + }, + { + "epoch": 0.20792354233036173, + "grad_norm": 0.53125, + "learning_rate": 0.001752966671372478, + "loss": 0.124, + "step": 23953 + }, + { + "epoch": 0.2079322228105659, + "grad_norm": 0.67578125, + "learning_rate": 0.0017529461884142016, + "loss": 0.1289, + "step": 23954 + }, + { + "epoch": 0.20794090329077006, + "grad_norm": 0.65625, + "learning_rate": 0.0017529257047418727, + "loss": 0.125, + "step": 23955 + }, + { + "epoch": 0.20794958377097422, + "grad_norm": 0.1044921875, + "learning_rate": 0.0017529052203555143, + "loss": 0.1338, + "step": 23956 + }, + { + "epoch": 0.2079582642511784, + "grad_norm": 0.2275390625, + "learning_rate": 0.0017528847352551488, + "loss": 0.1133, + "step": 23957 + }, + { + "epoch": 0.20796694473138255, + "grad_norm": 0.09228515625, + "learning_rate": 0.001752864249440798, + "loss": 0.0898, + "step": 23958 + }, + { + "epoch": 0.20797562521158672, + "grad_norm": 0.146484375, + "learning_rate": 0.0017528437629124845, + "loss": 0.1201, + "step": 23959 + }, + { + "epoch": 0.20798430569179088, + "grad_norm": 0.14453125, + "learning_rate": 0.0017528232756702311, + "loss": 0.123, + "step": 23960 + }, + { + "epoch": 0.20799298617199505, + "grad_norm": 0.2451171875, + "learning_rate": 0.00175280278771406, + "loss": 0.126, + "step": 23961 + }, + { + "epoch": 0.2080016666521992, + "grad_norm": 0.251953125, + "learning_rate": 0.0017527822990439934, + "loss": 0.0996, + "step": 23962 + }, + { + "epoch": 0.20801034713240338, + "grad_norm": 0.31640625, + "learning_rate": 0.0017527618096600539, + "loss": 0.1338, + "step": 23963 + }, + { + "epoch": 0.20801902761260754, + "grad_norm": 0.890625, + "learning_rate": 0.0017527413195622644, + "loss": 0.1396, + "step": 23964 + }, + { + "epoch": 0.20802770809281168, + "grad_norm": 0.1328125, + "learning_rate": 0.0017527208287506464, + "loss": 0.0659, + "step": 23965 + }, + { + "epoch": 0.20803638857301585, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017527003372252228, + "loss": 0.1221, + "step": 23966 + }, + { + "epoch": 0.20804506905322, + "grad_norm": 0.54296875, + "learning_rate": 0.0017526798449860157, + "loss": 0.1191, + "step": 23967 + }, + { + "epoch": 0.20805374953342418, + "grad_norm": 0.4296875, + "learning_rate": 0.0017526593520330478, + "loss": 0.1216, + "step": 23968 + }, + { + "epoch": 0.20806243001362834, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017526388583663415, + "loss": 0.0981, + "step": 23969 + }, + { + "epoch": 0.2080711104938325, + "grad_norm": 0.1474609375, + "learning_rate": 0.001752618363985919, + "loss": 0.1582, + "step": 23970 + }, + { + "epoch": 0.20807979097403667, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017525978688918032, + "loss": 0.1309, + "step": 23971 + }, + { + "epoch": 0.20808847145424084, + "grad_norm": 0.2099609375, + "learning_rate": 0.001752577373084016, + "loss": 0.0825, + "step": 23972 + }, + { + "epoch": 0.208097151934445, + "grad_norm": 0.373046875, + "learning_rate": 0.00175255687656258, + "loss": 0.0884, + "step": 23973 + }, + { + "epoch": 0.20810583241464917, + "grad_norm": 0.37890625, + "learning_rate": 0.0017525363793275178, + "loss": 0.1006, + "step": 23974 + }, + { + "epoch": 0.20811451289485333, + "grad_norm": 0.486328125, + "learning_rate": 0.0017525158813788513, + "loss": 0.127, + "step": 23975 + }, + { + "epoch": 0.2081231933750575, + "grad_norm": 0.20703125, + "learning_rate": 0.0017524953827166032, + "loss": 0.1865, + "step": 23976 + }, + { + "epoch": 0.20813187385526166, + "grad_norm": 0.203125, + "learning_rate": 0.001752474883340796, + "loss": 0.1235, + "step": 23977 + }, + { + "epoch": 0.20814055433546583, + "grad_norm": 0.474609375, + "learning_rate": 0.0017524543832514524, + "loss": 0.127, + "step": 23978 + }, + { + "epoch": 0.20814923481567, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017524338824485943, + "loss": 0.0977, + "step": 23979 + }, + { + "epoch": 0.20815791529587416, + "grad_norm": 0.33203125, + "learning_rate": 0.0017524133809322445, + "loss": 0.1279, + "step": 23980 + }, + { + "epoch": 0.20816659577607832, + "grad_norm": 0.5703125, + "learning_rate": 0.0017523928787024248, + "loss": 0.123, + "step": 23981 + }, + { + "epoch": 0.2081752762562825, + "grad_norm": 0.50390625, + "learning_rate": 0.0017523723757591583, + "loss": 0.0957, + "step": 23982 + }, + { + "epoch": 0.20818395673648665, + "grad_norm": 0.2119140625, + "learning_rate": 0.001752351872102467, + "loss": 0.1328, + "step": 23983 + }, + { + "epoch": 0.20819263721669082, + "grad_norm": 0.232421875, + "learning_rate": 0.0017523313677323738, + "loss": 0.0713, + "step": 23984 + }, + { + "epoch": 0.20820131769689498, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017523108626489006, + "loss": 0.1191, + "step": 23985 + }, + { + "epoch": 0.20820999817709915, + "grad_norm": 0.7109375, + "learning_rate": 0.0017522903568520701, + "loss": 0.1138, + "step": 23986 + }, + { + "epoch": 0.2082186786573033, + "grad_norm": 0.875, + "learning_rate": 0.0017522698503419048, + "loss": 0.1328, + "step": 23987 + }, + { + "epoch": 0.20822735913750748, + "grad_norm": 0.64453125, + "learning_rate": 0.0017522493431184265, + "loss": 0.1738, + "step": 23988 + }, + { + "epoch": 0.20823603961771164, + "grad_norm": 0.1015625, + "learning_rate": 0.0017522288351816586, + "loss": 0.1377, + "step": 23989 + }, + { + "epoch": 0.2082447200979158, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017522083265316228, + "loss": 0.1016, + "step": 23990 + }, + { + "epoch": 0.20825340057811997, + "grad_norm": 0.296875, + "learning_rate": 0.001752187817168342, + "loss": 0.1147, + "step": 23991 + }, + { + "epoch": 0.20826208105832414, + "grad_norm": 0.11083984375, + "learning_rate": 0.0017521673070918383, + "loss": 0.1455, + "step": 23992 + }, + { + "epoch": 0.2082707615385283, + "grad_norm": 0.484375, + "learning_rate": 0.0017521467963021343, + "loss": 0.1084, + "step": 23993 + }, + { + "epoch": 0.20827944201873247, + "grad_norm": 0.232421875, + "learning_rate": 0.0017521262847992522, + "loss": 0.0903, + "step": 23994 + }, + { + "epoch": 0.20828812249893663, + "grad_norm": 0.32421875, + "learning_rate": 0.0017521057725832145, + "loss": 0.1113, + "step": 23995 + }, + { + "epoch": 0.2082968029791408, + "grad_norm": 0.09228515625, + "learning_rate": 0.001752085259654044, + "loss": 0.1079, + "step": 23996 + }, + { + "epoch": 0.20830548345934496, + "grad_norm": 0.201171875, + "learning_rate": 0.001752064746011763, + "loss": 0.1221, + "step": 23997 + }, + { + "epoch": 0.20831416393954913, + "grad_norm": 0.1748046875, + "learning_rate": 0.0017520442316563934, + "loss": 0.1157, + "step": 23998 + }, + { + "epoch": 0.2083228444197533, + "grad_norm": 0.546875, + "learning_rate": 0.0017520237165879581, + "loss": 0.1318, + "step": 23999 + }, + { + "epoch": 0.20833152489995746, + "grad_norm": 0.19921875, + "learning_rate": 0.0017520032008064795, + "loss": 0.0977, + "step": 24000 + }, + { + "epoch": 0.20834020538016162, + "grad_norm": 0.271484375, + "learning_rate": 0.0017519826843119801, + "loss": 0.1318, + "step": 24001 + }, + { + "epoch": 0.2083488858603658, + "grad_norm": 0.1025390625, + "learning_rate": 0.001751962167104482, + "loss": 0.1162, + "step": 24002 + }, + { + "epoch": 0.20835756634056996, + "grad_norm": 0.244140625, + "learning_rate": 0.0017519416491840081, + "loss": 0.1172, + "step": 24003 + }, + { + "epoch": 0.20836624682077412, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017519211305505806, + "loss": 0.0879, + "step": 24004 + }, + { + "epoch": 0.20837492730097829, + "grad_norm": 0.83203125, + "learning_rate": 0.0017519006112042218, + "loss": 0.1211, + "step": 24005 + }, + { + "epoch": 0.20838360778118245, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017518800911449544, + "loss": 0.083, + "step": 24006 + }, + { + "epoch": 0.20839228826138662, + "grad_norm": 0.294921875, + "learning_rate": 0.0017518595703728006, + "loss": 0.0991, + "step": 24007 + }, + { + "epoch": 0.20840096874159078, + "grad_norm": 0.17578125, + "learning_rate": 0.0017518390488877832, + "loss": 0.1108, + "step": 24008 + }, + { + "epoch": 0.20840964922179495, + "grad_norm": 0.435546875, + "learning_rate": 0.0017518185266899244, + "loss": 0.1221, + "step": 24009 + }, + { + "epoch": 0.2084183297019991, + "grad_norm": 0.09521484375, + "learning_rate": 0.0017517980037792464, + "loss": 0.0967, + "step": 24010 + }, + { + "epoch": 0.20842701018220328, + "grad_norm": 0.19921875, + "learning_rate": 0.0017517774801557723, + "loss": 0.1016, + "step": 24011 + }, + { + "epoch": 0.20843569066240744, + "grad_norm": 0.328125, + "learning_rate": 0.0017517569558195237, + "loss": 0.1045, + "step": 24012 + }, + { + "epoch": 0.2084443711426116, + "grad_norm": 0.40625, + "learning_rate": 0.0017517364307705235, + "loss": 0.1426, + "step": 24013 + }, + { + "epoch": 0.20845305162281577, + "grad_norm": 1.0, + "learning_rate": 0.0017517159050087945, + "loss": 0.1309, + "step": 24014 + }, + { + "epoch": 0.20846173210301994, + "grad_norm": 0.10302734375, + "learning_rate": 0.0017516953785343583, + "loss": 0.0786, + "step": 24015 + }, + { + "epoch": 0.2084704125832241, + "grad_norm": 0.291015625, + "learning_rate": 0.0017516748513472383, + "loss": 0.0723, + "step": 24016 + }, + { + "epoch": 0.20847909306342827, + "grad_norm": 0.310546875, + "learning_rate": 0.0017516543234474562, + "loss": 0.1084, + "step": 24017 + }, + { + "epoch": 0.20848777354363243, + "grad_norm": 0.44921875, + "learning_rate": 0.0017516337948350346, + "loss": 0.123, + "step": 24018 + }, + { + "epoch": 0.2084964540238366, + "grad_norm": 0.14453125, + "learning_rate": 0.0017516132655099963, + "loss": 0.1367, + "step": 24019 + }, + { + "epoch": 0.20850513450404076, + "grad_norm": 0.45703125, + "learning_rate": 0.0017515927354723635, + "loss": 0.1162, + "step": 24020 + }, + { + "epoch": 0.20851381498424493, + "grad_norm": 0.37890625, + "learning_rate": 0.0017515722047221588, + "loss": 0.1309, + "step": 24021 + }, + { + "epoch": 0.2085224954644491, + "grad_norm": 0.0859375, + "learning_rate": 0.001751551673259404, + "loss": 0.084, + "step": 24022 + }, + { + "epoch": 0.20853117594465326, + "grad_norm": 0.07861328125, + "learning_rate": 0.0017515311410841229, + "loss": 0.0977, + "step": 24023 + }, + { + "epoch": 0.20853985642485742, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017515106081963366, + "loss": 0.1162, + "step": 24024 + }, + { + "epoch": 0.2085485369050616, + "grad_norm": 0.369140625, + "learning_rate": 0.0017514900745960681, + "loss": 0.0986, + "step": 24025 + }, + { + "epoch": 0.20855721738526575, + "grad_norm": 0.3984375, + "learning_rate": 0.0017514695402833398, + "loss": 0.1211, + "step": 24026 + }, + { + "epoch": 0.20856589786546992, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017514490052581745, + "loss": 0.1357, + "step": 24027 + }, + { + "epoch": 0.20857457834567408, + "grad_norm": 0.10302734375, + "learning_rate": 0.0017514284695205942, + "loss": 0.1094, + "step": 24028 + }, + { + "epoch": 0.20858325882587825, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017514079330706214, + "loss": 0.0845, + "step": 24029 + }, + { + "epoch": 0.2085919393060824, + "grad_norm": 0.0751953125, + "learning_rate": 0.0017513873959082787, + "loss": 0.0894, + "step": 24030 + }, + { + "epoch": 0.20860061978628658, + "grad_norm": 0.74609375, + "learning_rate": 0.0017513668580335887, + "loss": 0.1172, + "step": 24031 + }, + { + "epoch": 0.20860930026649074, + "grad_norm": 0.345703125, + "learning_rate": 0.0017513463194465736, + "loss": 0.1006, + "step": 24032 + }, + { + "epoch": 0.2086179807466949, + "grad_norm": 0.09716796875, + "learning_rate": 0.001751325780147256, + "loss": 0.1104, + "step": 24033 + }, + { + "epoch": 0.20862666122689907, + "grad_norm": 1.0390625, + "learning_rate": 0.001751305240135658, + "loss": 0.1104, + "step": 24034 + }, + { + "epoch": 0.20863534170710324, + "grad_norm": 0.703125, + "learning_rate": 0.0017512846994118026, + "loss": 0.1006, + "step": 24035 + }, + { + "epoch": 0.2086440221873074, + "grad_norm": 0.1220703125, + "learning_rate": 0.001751264157975712, + "loss": 0.106, + "step": 24036 + }, + { + "epoch": 0.20865270266751157, + "grad_norm": 0.384765625, + "learning_rate": 0.001751243615827409, + "loss": 0.1045, + "step": 24037 + }, + { + "epoch": 0.20866138314771573, + "grad_norm": 0.52734375, + "learning_rate": 0.0017512230729669154, + "loss": 0.083, + "step": 24038 + }, + { + "epoch": 0.2086700636279199, + "grad_norm": 0.0986328125, + "learning_rate": 0.0017512025293942542, + "loss": 0.1201, + "step": 24039 + }, + { + "epoch": 0.20867874410812406, + "grad_norm": 0.08154296875, + "learning_rate": 0.0017511819851094477, + "loss": 0.0908, + "step": 24040 + }, + { + "epoch": 0.20868742458832823, + "grad_norm": 0.255859375, + "learning_rate": 0.0017511614401125185, + "loss": 0.1484, + "step": 24041 + }, + { + "epoch": 0.2086961050685324, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017511408944034885, + "loss": 0.1157, + "step": 24042 + }, + { + "epoch": 0.20870478554873656, + "grad_norm": 0.2265625, + "learning_rate": 0.0017511203479823809, + "loss": 0.0859, + "step": 24043 + }, + { + "epoch": 0.20871346602894073, + "grad_norm": 0.6484375, + "learning_rate": 0.0017510998008492178, + "loss": 0.0938, + "step": 24044 + }, + { + "epoch": 0.2087221465091449, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017510792530040218, + "loss": 0.166, + "step": 24045 + }, + { + "epoch": 0.20873082698934906, + "grad_norm": 0.427734375, + "learning_rate": 0.0017510587044468153, + "loss": 0.0859, + "step": 24046 + }, + { + "epoch": 0.20873950746955322, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017510381551776209, + "loss": 0.1099, + "step": 24047 + }, + { + "epoch": 0.20874818794975739, + "grad_norm": 0.216796875, + "learning_rate": 0.0017510176051964608, + "loss": 0.1201, + "step": 24048 + }, + { + "epoch": 0.20875686842996155, + "grad_norm": 0.21875, + "learning_rate": 0.0017509970545033578, + "loss": 0.1099, + "step": 24049 + }, + { + "epoch": 0.20876554891016572, + "grad_norm": 0.8203125, + "learning_rate": 0.001750976503098334, + "loss": 0.1035, + "step": 24050 + }, + { + "epoch": 0.20877422939036988, + "grad_norm": 0.625, + "learning_rate": 0.001750955950981412, + "loss": 0.0972, + "step": 24051 + }, + { + "epoch": 0.20878290987057405, + "grad_norm": 1.2734375, + "learning_rate": 0.0017509353981526145, + "loss": 0.1123, + "step": 24052 + }, + { + "epoch": 0.2087915903507782, + "grad_norm": 0.63671875, + "learning_rate": 0.0017509148446119636, + "loss": 0.0996, + "step": 24053 + }, + { + "epoch": 0.20880027083098238, + "grad_norm": 0.6796875, + "learning_rate": 0.0017508942903594824, + "loss": 0.0908, + "step": 24054 + }, + { + "epoch": 0.20880895131118654, + "grad_norm": 0.92578125, + "learning_rate": 0.0017508737353951927, + "loss": 0.1348, + "step": 24055 + }, + { + "epoch": 0.2088176317913907, + "grad_norm": 0.146484375, + "learning_rate": 0.0017508531797191173, + "loss": 0.1035, + "step": 24056 + }, + { + "epoch": 0.20882631227159487, + "grad_norm": 0.390625, + "learning_rate": 0.001750832623331279, + "loss": 0.1367, + "step": 24057 + }, + { + "epoch": 0.20883499275179904, + "grad_norm": 0.2470703125, + "learning_rate": 0.0017508120662316996, + "loss": 0.123, + "step": 24058 + }, + { + "epoch": 0.2088436732320032, + "grad_norm": 0.154296875, + "learning_rate": 0.0017507915084204017, + "loss": 0.1172, + "step": 24059 + }, + { + "epoch": 0.20885235371220737, + "grad_norm": 0.46875, + "learning_rate": 0.0017507709498974084, + "loss": 0.1592, + "step": 24060 + }, + { + "epoch": 0.20886103419241153, + "grad_norm": 0.498046875, + "learning_rate": 0.0017507503906627417, + "loss": 0.1387, + "step": 24061 + }, + { + "epoch": 0.2088697146726157, + "grad_norm": 1.625, + "learning_rate": 0.0017507298307164237, + "loss": 0.1035, + "step": 24062 + }, + { + "epoch": 0.20887839515281986, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017507092700584776, + "loss": 0.1143, + "step": 24063 + }, + { + "epoch": 0.20888707563302403, + "grad_norm": 0.5703125, + "learning_rate": 0.001750688708688926, + "loss": 0.0854, + "step": 24064 + }, + { + "epoch": 0.2088957561132282, + "grad_norm": 1.15625, + "learning_rate": 0.0017506681466077905, + "loss": 0.1562, + "step": 24065 + }, + { + "epoch": 0.20890443659343236, + "grad_norm": 0.53515625, + "learning_rate": 0.001750647583815094, + "loss": 0.1162, + "step": 24066 + }, + { + "epoch": 0.20891311707363652, + "grad_norm": 0.0849609375, + "learning_rate": 0.0017506270203108596, + "loss": 0.0879, + "step": 24067 + }, + { + "epoch": 0.2089217975538407, + "grad_norm": 0.462890625, + "learning_rate": 0.001750606456095109, + "loss": 0.0918, + "step": 24068 + }, + { + "epoch": 0.20893047803404485, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017505858911678652, + "loss": 0.1211, + "step": 24069 + }, + { + "epoch": 0.20893915851424902, + "grad_norm": 0.8046875, + "learning_rate": 0.00175056532552915, + "loss": 0.0991, + "step": 24070 + }, + { + "epoch": 0.20894783899445318, + "grad_norm": 0.84375, + "learning_rate": 0.0017505447591789865, + "loss": 0.0933, + "step": 24071 + }, + { + "epoch": 0.20895651947465735, + "grad_norm": 0.490234375, + "learning_rate": 0.0017505241921173971, + "loss": 0.0874, + "step": 24072 + }, + { + "epoch": 0.2089651999548615, + "grad_norm": 0.13671875, + "learning_rate": 0.0017505036243444043, + "loss": 0.082, + "step": 24073 + }, + { + "epoch": 0.20897388043506568, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017504830558600303, + "loss": 0.1113, + "step": 24074 + }, + { + "epoch": 0.20898256091526984, + "grad_norm": 0.71875, + "learning_rate": 0.001750462486664298, + "loss": 0.1338, + "step": 24075 + }, + { + "epoch": 0.208991241395474, + "grad_norm": 0.32421875, + "learning_rate": 0.0017504419167572297, + "loss": 0.0879, + "step": 24076 + }, + { + "epoch": 0.20899992187567817, + "grad_norm": 0.3203125, + "learning_rate": 0.0017504213461388477, + "loss": 0.0933, + "step": 24077 + }, + { + "epoch": 0.20900860235588234, + "grad_norm": 0.296875, + "learning_rate": 0.0017504007748091748, + "loss": 0.1011, + "step": 24078 + }, + { + "epoch": 0.2090172828360865, + "grad_norm": 1.265625, + "learning_rate": 0.0017503802027682334, + "loss": 0.1904, + "step": 24079 + }, + { + "epoch": 0.20902596331629067, + "grad_norm": 0.12255859375, + "learning_rate": 0.001750359630016046, + "loss": 0.1094, + "step": 24080 + }, + { + "epoch": 0.20903464379649483, + "grad_norm": 0.4140625, + "learning_rate": 0.0017503390565526348, + "loss": 0.1211, + "step": 24081 + }, + { + "epoch": 0.209043324276699, + "grad_norm": 0.087890625, + "learning_rate": 0.001750318482378023, + "loss": 0.083, + "step": 24082 + }, + { + "epoch": 0.20905200475690316, + "grad_norm": 0.14453125, + "learning_rate": 0.0017502979074922327, + "loss": 0.1016, + "step": 24083 + }, + { + "epoch": 0.20906068523710733, + "grad_norm": 0.1650390625, + "learning_rate": 0.001750277331895286, + "loss": 0.0752, + "step": 24084 + }, + { + "epoch": 0.2090693657173115, + "grad_norm": 0.11376953125, + "learning_rate": 0.001750256755587206, + "loss": 0.1104, + "step": 24085 + }, + { + "epoch": 0.20907804619751566, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017502361785680148, + "loss": 0.1221, + "step": 24086 + }, + { + "epoch": 0.20908672667771983, + "grad_norm": 0.1123046875, + "learning_rate": 0.0017502156008377353, + "loss": 0.1221, + "step": 24087 + }, + { + "epoch": 0.20909540715792396, + "grad_norm": 0.173828125, + "learning_rate": 0.0017501950223963898, + "loss": 0.1074, + "step": 24088 + }, + { + "epoch": 0.20910408763812813, + "grad_norm": 0.3671875, + "learning_rate": 0.0017501744432440005, + "loss": 0.1152, + "step": 24089 + }, + { + "epoch": 0.2091127681183323, + "grad_norm": 0.625, + "learning_rate": 0.0017501538633805905, + "loss": 0.1045, + "step": 24090 + }, + { + "epoch": 0.20912144859853646, + "grad_norm": 0.640625, + "learning_rate": 0.0017501332828061818, + "loss": 0.0967, + "step": 24091 + }, + { + "epoch": 0.20913012907874062, + "grad_norm": 0.123046875, + "learning_rate": 0.0017501127015207975, + "loss": 0.1533, + "step": 24092 + }, + { + "epoch": 0.2091388095589448, + "grad_norm": 0.35546875, + "learning_rate": 0.0017500921195244595, + "loss": 0.105, + "step": 24093 + }, + { + "epoch": 0.20914749003914895, + "grad_norm": 0.0703125, + "learning_rate": 0.0017500715368171903, + "loss": 0.0952, + "step": 24094 + }, + { + "epoch": 0.20915617051935312, + "grad_norm": 0.76953125, + "learning_rate": 0.0017500509533990127, + "loss": 0.1533, + "step": 24095 + }, + { + "epoch": 0.20916485099955728, + "grad_norm": 0.40234375, + "learning_rate": 0.0017500303692699493, + "loss": 0.2227, + "step": 24096 + }, + { + "epoch": 0.20917353147976145, + "grad_norm": 0.2392578125, + "learning_rate": 0.0017500097844300223, + "loss": 0.1074, + "step": 24097 + }, + { + "epoch": 0.2091822119599656, + "grad_norm": 0.609375, + "learning_rate": 0.0017499891988792544, + "loss": 0.0996, + "step": 24098 + }, + { + "epoch": 0.20919089244016978, + "grad_norm": 0.494140625, + "learning_rate": 0.0017499686126176684, + "loss": 0.1113, + "step": 24099 + }, + { + "epoch": 0.20919957292037394, + "grad_norm": 0.337890625, + "learning_rate": 0.001749948025645286, + "loss": 0.0972, + "step": 24100 + }, + { + "epoch": 0.2092082534005781, + "grad_norm": 0.119140625, + "learning_rate": 0.0017499274379621306, + "loss": 0.1182, + "step": 24101 + }, + { + "epoch": 0.20921693388078227, + "grad_norm": 0.18359375, + "learning_rate": 0.001749906849568224, + "loss": 0.1006, + "step": 24102 + }, + { + "epoch": 0.20922561436098644, + "grad_norm": 0.1044921875, + "learning_rate": 0.0017498862604635893, + "loss": 0.1182, + "step": 24103 + }, + { + "epoch": 0.2092342948411906, + "grad_norm": 0.6875, + "learning_rate": 0.0017498656706482485, + "loss": 0.0884, + "step": 24104 + }, + { + "epoch": 0.20924297532139477, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017498450801222247, + "loss": 0.0835, + "step": 24105 + }, + { + "epoch": 0.20925165580159893, + "grad_norm": 0.19140625, + "learning_rate": 0.00174982448888554, + "loss": 0.123, + "step": 24106 + }, + { + "epoch": 0.2092603362818031, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017498038969382167, + "loss": 0.0952, + "step": 24107 + }, + { + "epoch": 0.20926901676200726, + "grad_norm": 0.1171875, + "learning_rate": 0.0017497833042802778, + "loss": 0.123, + "step": 24108 + }, + { + "epoch": 0.20927769724221143, + "grad_norm": 0.5703125, + "learning_rate": 0.0017497627109117455, + "loss": 0.1133, + "step": 24109 + }, + { + "epoch": 0.2092863777224156, + "grad_norm": 0.65234375, + "learning_rate": 0.0017497421168326428, + "loss": 0.1162, + "step": 24110 + }, + { + "epoch": 0.20929505820261976, + "grad_norm": 0.083984375, + "learning_rate": 0.0017497215220429916, + "loss": 0.1064, + "step": 24111 + }, + { + "epoch": 0.20930373868282393, + "grad_norm": 0.6171875, + "learning_rate": 0.0017497009265428148, + "loss": 0.1055, + "step": 24112 + }, + { + "epoch": 0.2093124191630281, + "grad_norm": 0.130859375, + "learning_rate": 0.001749680330332135, + "loss": 0.1318, + "step": 24113 + }, + { + "epoch": 0.20932109964323226, + "grad_norm": 0.146484375, + "learning_rate": 0.0017496597334109743, + "loss": 0.0898, + "step": 24114 + }, + { + "epoch": 0.20932978012343642, + "grad_norm": 0.212890625, + "learning_rate": 0.0017496391357793555, + "loss": 0.104, + "step": 24115 + }, + { + "epoch": 0.20933846060364059, + "grad_norm": 0.212890625, + "learning_rate": 0.0017496185374373011, + "loss": 0.0786, + "step": 24116 + }, + { + "epoch": 0.20934714108384475, + "grad_norm": 0.341796875, + "learning_rate": 0.0017495979383848337, + "loss": 0.1465, + "step": 24117 + }, + { + "epoch": 0.20935582156404892, + "grad_norm": 0.06640625, + "learning_rate": 0.0017495773386219757, + "loss": 0.0889, + "step": 24118 + }, + { + "epoch": 0.20936450204425308, + "grad_norm": 0.2578125, + "learning_rate": 0.0017495567381487498, + "loss": 0.1108, + "step": 24119 + }, + { + "epoch": 0.20937318252445725, + "grad_norm": 0.27734375, + "learning_rate": 0.0017495361369651782, + "loss": 0.0859, + "step": 24120 + }, + { + "epoch": 0.2093818630046614, + "grad_norm": 0.1318359375, + "learning_rate": 0.0017495155350712838, + "loss": 0.1162, + "step": 24121 + }, + { + "epoch": 0.20939054348486558, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017494949324670893, + "loss": 0.1016, + "step": 24122 + }, + { + "epoch": 0.20939922396506974, + "grad_norm": 0.37109375, + "learning_rate": 0.0017494743291526162, + "loss": 0.0796, + "step": 24123 + }, + { + "epoch": 0.2094079044452739, + "grad_norm": 0.2578125, + "learning_rate": 0.0017494537251278883, + "loss": 0.126, + "step": 24124 + }, + { + "epoch": 0.20941658492547807, + "grad_norm": 0.443359375, + "learning_rate": 0.0017494331203929273, + "loss": 0.0718, + "step": 24125 + }, + { + "epoch": 0.20942526540568224, + "grad_norm": 0.546875, + "learning_rate": 0.001749412514947756, + "loss": 0.1123, + "step": 24126 + }, + { + "epoch": 0.2094339458858864, + "grad_norm": 0.09521484375, + "learning_rate": 0.0017493919087923971, + "loss": 0.1094, + "step": 24127 + }, + { + "epoch": 0.20944262636609057, + "grad_norm": 0.44921875, + "learning_rate": 0.0017493713019268726, + "loss": 0.1221, + "step": 24128 + }, + { + "epoch": 0.20945130684629473, + "grad_norm": 0.08154296875, + "learning_rate": 0.0017493506943512059, + "loss": 0.085, + "step": 24129 + }, + { + "epoch": 0.2094599873264989, + "grad_norm": 1.5390625, + "learning_rate": 0.0017493300860654184, + "loss": 0.1211, + "step": 24130 + }, + { + "epoch": 0.20946866780670306, + "grad_norm": 0.1220703125, + "learning_rate": 0.001749309477069534, + "loss": 0.1011, + "step": 24131 + }, + { + "epoch": 0.20947734828690723, + "grad_norm": 0.119140625, + "learning_rate": 0.001749288867363574, + "loss": 0.0903, + "step": 24132 + }, + { + "epoch": 0.2094860287671114, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017492682569475618, + "loss": 0.1064, + "step": 24133 + }, + { + "epoch": 0.20949470924731556, + "grad_norm": 0.1875, + "learning_rate": 0.001749247645821519, + "loss": 0.0996, + "step": 24134 + }, + { + "epoch": 0.20950338972751972, + "grad_norm": 0.6953125, + "learning_rate": 0.0017492270339854695, + "loss": 0.1211, + "step": 24135 + }, + { + "epoch": 0.2095120702077239, + "grad_norm": 0.109375, + "learning_rate": 0.0017492064214394347, + "loss": 0.1143, + "step": 24136 + }, + { + "epoch": 0.20952075068792805, + "grad_norm": 0.314453125, + "learning_rate": 0.0017491858081834374, + "loss": 0.0874, + "step": 24137 + }, + { + "epoch": 0.20952943116813222, + "grad_norm": 0.94921875, + "learning_rate": 0.0017491651942175004, + "loss": 0.5234, + "step": 24138 + }, + { + "epoch": 0.20953811164833638, + "grad_norm": 0.63671875, + "learning_rate": 0.001749144579541646, + "loss": 0.0811, + "step": 24139 + }, + { + "epoch": 0.20954679212854055, + "grad_norm": 0.48046875, + "learning_rate": 0.0017491239641558968, + "loss": 0.1816, + "step": 24140 + }, + { + "epoch": 0.20955547260874471, + "grad_norm": 0.41796875, + "learning_rate": 0.0017491033480602756, + "loss": 0.0752, + "step": 24141 + }, + { + "epoch": 0.20956415308894888, + "grad_norm": 0.48046875, + "learning_rate": 0.0017490827312548046, + "loss": 0.1172, + "step": 24142 + }, + { + "epoch": 0.20957283356915304, + "grad_norm": 0.6328125, + "learning_rate": 0.0017490621137395065, + "loss": 0.1172, + "step": 24143 + }, + { + "epoch": 0.2095815140493572, + "grad_norm": 0.458984375, + "learning_rate": 0.0017490414955144038, + "loss": 0.1245, + "step": 24144 + }, + { + "epoch": 0.20959019452956137, + "grad_norm": 0.671875, + "learning_rate": 0.001749020876579519, + "loss": 0.1069, + "step": 24145 + }, + { + "epoch": 0.20959887500976554, + "grad_norm": 0.71484375, + "learning_rate": 0.0017490002569348748, + "loss": 0.0957, + "step": 24146 + }, + { + "epoch": 0.2096075554899697, + "grad_norm": 0.34375, + "learning_rate": 0.001748979636580494, + "loss": 0.1201, + "step": 24147 + }, + { + "epoch": 0.20961623597017387, + "grad_norm": 0.435546875, + "learning_rate": 0.0017489590155163983, + "loss": 0.1201, + "step": 24148 + }, + { + "epoch": 0.20962491645037803, + "grad_norm": 0.2314453125, + "learning_rate": 0.001748938393742611, + "loss": 0.1416, + "step": 24149 + }, + { + "epoch": 0.2096335969305822, + "grad_norm": 0.1474609375, + "learning_rate": 0.0017489177712591544, + "loss": 0.0859, + "step": 24150 + }, + { + "epoch": 0.20964227741078637, + "grad_norm": 0.6171875, + "learning_rate": 0.001748897148066051, + "loss": 0.1318, + "step": 24151 + }, + { + "epoch": 0.20965095789099053, + "grad_norm": 0.65234375, + "learning_rate": 0.0017488765241633235, + "loss": 0.1123, + "step": 24152 + }, + { + "epoch": 0.2096596383711947, + "grad_norm": 0.396484375, + "learning_rate": 0.0017488558995509944, + "loss": 0.1357, + "step": 24153 + }, + { + "epoch": 0.20966831885139886, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017488352742290863, + "loss": 0.1006, + "step": 24154 + }, + { + "epoch": 0.20967699933160303, + "grad_norm": 0.65625, + "learning_rate": 0.0017488146481976215, + "loss": 0.1309, + "step": 24155 + }, + { + "epoch": 0.2096856798118072, + "grad_norm": 0.58984375, + "learning_rate": 0.001748794021456623, + "loss": 0.1025, + "step": 24156 + }, + { + "epoch": 0.20969436029201136, + "grad_norm": 0.55859375, + "learning_rate": 0.001748773394006113, + "loss": 0.0928, + "step": 24157 + }, + { + "epoch": 0.20970304077221552, + "grad_norm": 0.08642578125, + "learning_rate": 0.001748752765846114, + "loss": 0.082, + "step": 24158 + }, + { + "epoch": 0.20971172125241969, + "grad_norm": 0.63671875, + "learning_rate": 0.0017487321369766492, + "loss": 0.1406, + "step": 24159 + }, + { + "epoch": 0.20972040173262385, + "grad_norm": 0.62890625, + "learning_rate": 0.0017487115073977402, + "loss": 0.1387, + "step": 24160 + }, + { + "epoch": 0.20972908221282802, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017486908771094103, + "loss": 0.1074, + "step": 24161 + }, + { + "epoch": 0.20973776269303218, + "grad_norm": 1.234375, + "learning_rate": 0.0017486702461116817, + "loss": 0.1035, + "step": 24162 + }, + { + "epoch": 0.20974644317323635, + "grad_norm": 0.09130859375, + "learning_rate": 0.001748649614404577, + "loss": 0.1108, + "step": 24163 + }, + { + "epoch": 0.2097551236534405, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017486289819881188, + "loss": 0.1118, + "step": 24164 + }, + { + "epoch": 0.20976380413364468, + "grad_norm": 0.52734375, + "learning_rate": 0.0017486083488623297, + "loss": 0.1816, + "step": 24165 + }, + { + "epoch": 0.20977248461384884, + "grad_norm": 0.494140625, + "learning_rate": 0.0017485877150272323, + "loss": 0.0947, + "step": 24166 + }, + { + "epoch": 0.209781165094053, + "grad_norm": 0.232421875, + "learning_rate": 0.001748567080482849, + "loss": 0.1074, + "step": 24167 + }, + { + "epoch": 0.20978984557425717, + "grad_norm": 0.123046875, + "learning_rate": 0.0017485464452292028, + "loss": 0.1006, + "step": 24168 + }, + { + "epoch": 0.20979852605446134, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017485258092663158, + "loss": 0.0752, + "step": 24169 + }, + { + "epoch": 0.2098072065346655, + "grad_norm": 0.150390625, + "learning_rate": 0.0017485051725942106, + "loss": 0.1914, + "step": 24170 + }, + { + "epoch": 0.20981588701486967, + "grad_norm": 0.07666015625, + "learning_rate": 0.0017484845352129103, + "loss": 0.105, + "step": 24171 + }, + { + "epoch": 0.20982456749507383, + "grad_norm": 0.234375, + "learning_rate": 0.0017484638971224365, + "loss": 0.0996, + "step": 24172 + }, + { + "epoch": 0.209833247975278, + "grad_norm": 0.5234375, + "learning_rate": 0.0017484432583228127, + "loss": 0.1211, + "step": 24173 + }, + { + "epoch": 0.20984192845548216, + "grad_norm": 0.0810546875, + "learning_rate": 0.001748422618814061, + "loss": 0.1162, + "step": 24174 + }, + { + "epoch": 0.20985060893568633, + "grad_norm": 0.640625, + "learning_rate": 0.001748401978596204, + "loss": 0.1123, + "step": 24175 + }, + { + "epoch": 0.2098592894158905, + "grad_norm": 0.4296875, + "learning_rate": 0.0017483813376692643, + "loss": 0.1045, + "step": 24176 + }, + { + "epoch": 0.20986796989609466, + "grad_norm": 0.35546875, + "learning_rate": 0.0017483606960332645, + "loss": 0.126, + "step": 24177 + }, + { + "epoch": 0.20987665037629882, + "grad_norm": 0.1142578125, + "learning_rate": 0.0017483400536882272, + "loss": 0.1064, + "step": 24178 + }, + { + "epoch": 0.209885330856503, + "grad_norm": 0.388671875, + "learning_rate": 0.001748319410634175, + "loss": 0.083, + "step": 24179 + }, + { + "epoch": 0.20989401133670715, + "grad_norm": 0.173828125, + "learning_rate": 0.0017482987668711305, + "loss": 0.1621, + "step": 24180 + }, + { + "epoch": 0.20990269181691132, + "grad_norm": 0.1103515625, + "learning_rate": 0.0017482781223991163, + "loss": 0.126, + "step": 24181 + }, + { + "epoch": 0.20991137229711548, + "grad_norm": 0.146484375, + "learning_rate": 0.0017482574772181548, + "loss": 0.1309, + "step": 24182 + }, + { + "epoch": 0.20992005277731965, + "grad_norm": 0.2421875, + "learning_rate": 0.0017482368313282684, + "loss": 0.1182, + "step": 24183 + }, + { + "epoch": 0.20992873325752381, + "grad_norm": 0.357421875, + "learning_rate": 0.0017482161847294801, + "loss": 0.0986, + "step": 24184 + }, + { + "epoch": 0.20993741373772798, + "grad_norm": 0.416015625, + "learning_rate": 0.0017481955374218127, + "loss": 0.106, + "step": 24185 + }, + { + "epoch": 0.20994609421793214, + "grad_norm": 0.765625, + "learning_rate": 0.001748174889405288, + "loss": 0.1025, + "step": 24186 + }, + { + "epoch": 0.2099547746981363, + "grad_norm": 0.1494140625, + "learning_rate": 0.001748154240679929, + "loss": 0.1318, + "step": 24187 + }, + { + "epoch": 0.20996345517834047, + "grad_norm": 0.49609375, + "learning_rate": 0.0017481335912457583, + "loss": 0.1367, + "step": 24188 + }, + { + "epoch": 0.20997213565854464, + "grad_norm": 0.236328125, + "learning_rate": 0.0017481129411027986, + "loss": 0.1035, + "step": 24189 + }, + { + "epoch": 0.2099808161387488, + "grad_norm": 0.6484375, + "learning_rate": 0.001748092290251072, + "loss": 0.0889, + "step": 24190 + }, + { + "epoch": 0.20998949661895297, + "grad_norm": 0.1328125, + "learning_rate": 0.0017480716386906016, + "loss": 0.1006, + "step": 24191 + }, + { + "epoch": 0.20999817709915714, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017480509864214097, + "loss": 0.1318, + "step": 24192 + }, + { + "epoch": 0.2100068575793613, + "grad_norm": 0.25390625, + "learning_rate": 0.0017480303334435194, + "loss": 0.0894, + "step": 24193 + }, + { + "epoch": 0.21001553805956547, + "grad_norm": 0.40234375, + "learning_rate": 0.0017480096797569521, + "loss": 0.1035, + "step": 24194 + }, + { + "epoch": 0.21002421853976963, + "grad_norm": 0.27734375, + "learning_rate": 0.0017479890253617317, + "loss": 0.0972, + "step": 24195 + }, + { + "epoch": 0.2100328990199738, + "grad_norm": 0.111328125, + "learning_rate": 0.0017479683702578803, + "loss": 0.126, + "step": 24196 + }, + { + "epoch": 0.21004157950017796, + "grad_norm": 0.2275390625, + "learning_rate": 0.0017479477144454203, + "loss": 0.0996, + "step": 24197 + }, + { + "epoch": 0.21005025998038213, + "grad_norm": 0.140625, + "learning_rate": 0.0017479270579243743, + "loss": 0.1016, + "step": 24198 + }, + { + "epoch": 0.2100589404605863, + "grad_norm": 0.2294921875, + "learning_rate": 0.001747906400694765, + "loss": 0.1006, + "step": 24199 + }, + { + "epoch": 0.21006762094079046, + "grad_norm": 0.6015625, + "learning_rate": 0.0017478857427566151, + "loss": 0.1152, + "step": 24200 + }, + { + "epoch": 0.21007630142099462, + "grad_norm": 0.50390625, + "learning_rate": 0.0017478650841099468, + "loss": 0.1377, + "step": 24201 + }, + { + "epoch": 0.2100849819011988, + "grad_norm": 0.53515625, + "learning_rate": 0.0017478444247547833, + "loss": 0.1035, + "step": 24202 + }, + { + "epoch": 0.21009366238140295, + "grad_norm": 0.65625, + "learning_rate": 0.0017478237646911467, + "loss": 0.1309, + "step": 24203 + }, + { + "epoch": 0.21010234286160712, + "grad_norm": 0.12255859375, + "learning_rate": 0.00174780310391906, + "loss": 0.1094, + "step": 24204 + }, + { + "epoch": 0.21011102334181128, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017477824424385452, + "loss": 0.0967, + "step": 24205 + }, + { + "epoch": 0.21011970382201545, + "grad_norm": 0.302734375, + "learning_rate": 0.0017477617802496258, + "loss": 0.127, + "step": 24206 + }, + { + "epoch": 0.2101283843022196, + "grad_norm": 0.11767578125, + "learning_rate": 0.0017477411173523237, + "loss": 0.1138, + "step": 24207 + }, + { + "epoch": 0.21013706478242378, + "grad_norm": 0.890625, + "learning_rate": 0.001747720453746661, + "loss": 0.1523, + "step": 24208 + }, + { + "epoch": 0.21014574526262794, + "grad_norm": 0.9296875, + "learning_rate": 0.0017476997894326615, + "loss": 0.1152, + "step": 24209 + }, + { + "epoch": 0.2101544257428321, + "grad_norm": 0.267578125, + "learning_rate": 0.0017476791244103473, + "loss": 0.1104, + "step": 24210 + }, + { + "epoch": 0.21016310622303624, + "grad_norm": 0.10400390625, + "learning_rate": 0.001747658458679741, + "loss": 0.1299, + "step": 24211 + }, + { + "epoch": 0.2101717867032404, + "grad_norm": 0.267578125, + "learning_rate": 0.0017476377922408647, + "loss": 0.1064, + "step": 24212 + }, + { + "epoch": 0.21018046718344457, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017476171250937415, + "loss": 0.0859, + "step": 24213 + }, + { + "epoch": 0.21018914766364874, + "grad_norm": 0.4609375, + "learning_rate": 0.0017475964572383942, + "loss": 0.125, + "step": 24214 + }, + { + "epoch": 0.2101978281438529, + "grad_norm": 0.5, + "learning_rate": 0.0017475757886748454, + "loss": 0.1084, + "step": 24215 + }, + { + "epoch": 0.21020650862405707, + "grad_norm": 0.357421875, + "learning_rate": 0.001747555119403117, + "loss": 0.1221, + "step": 24216 + }, + { + "epoch": 0.21021518910426124, + "grad_norm": 0.076171875, + "learning_rate": 0.001747534449423232, + "loss": 0.1035, + "step": 24217 + }, + { + "epoch": 0.2102238695844654, + "grad_norm": 0.68359375, + "learning_rate": 0.0017475137787352133, + "loss": 0.1289, + "step": 24218 + }, + { + "epoch": 0.21023255006466957, + "grad_norm": 0.291015625, + "learning_rate": 0.001747493107339083, + "loss": 0.1602, + "step": 24219 + }, + { + "epoch": 0.21024123054487373, + "grad_norm": 0.15234375, + "learning_rate": 0.0017474724352348643, + "loss": 0.0928, + "step": 24220 + }, + { + "epoch": 0.2102499110250779, + "grad_norm": 0.10205078125, + "learning_rate": 0.0017474517624225792, + "loss": 0.1328, + "step": 24221 + }, + { + "epoch": 0.21025859150528206, + "grad_norm": 0.177734375, + "learning_rate": 0.0017474310889022508, + "loss": 0.0933, + "step": 24222 + }, + { + "epoch": 0.21026727198548623, + "grad_norm": 0.41015625, + "learning_rate": 0.0017474104146739011, + "loss": 0.1279, + "step": 24223 + }, + { + "epoch": 0.2102759524656904, + "grad_norm": 0.70703125, + "learning_rate": 0.0017473897397375536, + "loss": 0.0806, + "step": 24224 + }, + { + "epoch": 0.21028463294589456, + "grad_norm": 0.58984375, + "learning_rate": 0.0017473690640932302, + "loss": 0.103, + "step": 24225 + }, + { + "epoch": 0.21029331342609872, + "grad_norm": 0.5546875, + "learning_rate": 0.0017473483877409536, + "loss": 0.1182, + "step": 24226 + }, + { + "epoch": 0.2103019939063029, + "grad_norm": 0.11328125, + "learning_rate": 0.0017473277106807468, + "loss": 0.0928, + "step": 24227 + }, + { + "epoch": 0.21031067438650705, + "grad_norm": 0.8359375, + "learning_rate": 0.0017473070329126318, + "loss": 0.1064, + "step": 24228 + }, + { + "epoch": 0.21031935486671122, + "grad_norm": 0.1611328125, + "learning_rate": 0.001747286354436632, + "loss": 0.105, + "step": 24229 + }, + { + "epoch": 0.21032803534691538, + "grad_norm": 0.2158203125, + "learning_rate": 0.001747265675252769, + "loss": 0.1084, + "step": 24230 + }, + { + "epoch": 0.21033671582711955, + "grad_norm": 0.5390625, + "learning_rate": 0.0017472449953610666, + "loss": 0.1133, + "step": 24231 + }, + { + "epoch": 0.2103453963073237, + "grad_norm": 0.447265625, + "learning_rate": 0.0017472243147615464, + "loss": 0.1514, + "step": 24232 + }, + { + "epoch": 0.21035407678752788, + "grad_norm": 1.2578125, + "learning_rate": 0.0017472036334542315, + "loss": 0.1104, + "step": 24233 + }, + { + "epoch": 0.21036275726773204, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017471829514391444, + "loss": 0.1211, + "step": 24234 + }, + { + "epoch": 0.2103714377479362, + "grad_norm": 1.03125, + "learning_rate": 0.0017471622687163077, + "loss": 0.1562, + "step": 24235 + }, + { + "epoch": 0.21038011822814037, + "grad_norm": 0.1474609375, + "learning_rate": 0.001747141585285744, + "loss": 0.1133, + "step": 24236 + }, + { + "epoch": 0.21038879870834454, + "grad_norm": 0.451171875, + "learning_rate": 0.0017471209011474763, + "loss": 0.0957, + "step": 24237 + }, + { + "epoch": 0.2103974791885487, + "grad_norm": 0.55078125, + "learning_rate": 0.0017471002163015265, + "loss": 0.0874, + "step": 24238 + }, + { + "epoch": 0.21040615966875287, + "grad_norm": 0.28515625, + "learning_rate": 0.0017470795307479182, + "loss": 0.1172, + "step": 24239 + }, + { + "epoch": 0.21041484014895703, + "grad_norm": 0.255859375, + "learning_rate": 0.001747058844486673, + "loss": 0.1172, + "step": 24240 + }, + { + "epoch": 0.2104235206291612, + "grad_norm": 0.73046875, + "learning_rate": 0.001747038157517814, + "loss": 0.1143, + "step": 24241 + }, + { + "epoch": 0.21043220110936536, + "grad_norm": 0.1865234375, + "learning_rate": 0.001747017469841364, + "loss": 0.1426, + "step": 24242 + }, + { + "epoch": 0.21044088158956953, + "grad_norm": 0.28125, + "learning_rate": 0.0017469967814573452, + "loss": 0.1235, + "step": 24243 + }, + { + "epoch": 0.2104495620697737, + "grad_norm": 0.267578125, + "learning_rate": 0.0017469760923657805, + "loss": 0.1162, + "step": 24244 + }, + { + "epoch": 0.21045824254997786, + "grad_norm": 0.71484375, + "learning_rate": 0.0017469554025666926, + "loss": 0.1396, + "step": 24245 + }, + { + "epoch": 0.21046692303018202, + "grad_norm": 0.69921875, + "learning_rate": 0.0017469347120601037, + "loss": 0.1055, + "step": 24246 + }, + { + "epoch": 0.2104756035103862, + "grad_norm": 0.1318359375, + "learning_rate": 0.001746914020846037, + "loss": 0.1064, + "step": 24247 + }, + { + "epoch": 0.21048428399059035, + "grad_norm": 0.265625, + "learning_rate": 0.0017468933289245148, + "loss": 0.0986, + "step": 24248 + }, + { + "epoch": 0.21049296447079452, + "grad_norm": 0.2001953125, + "learning_rate": 0.0017468726362955597, + "loss": 0.1172, + "step": 24249 + }, + { + "epoch": 0.21050164495099868, + "grad_norm": 3.28125, + "learning_rate": 0.0017468519429591945, + "loss": 0.1455, + "step": 24250 + }, + { + "epoch": 0.21051032543120285, + "grad_norm": 0.166015625, + "learning_rate": 0.0017468312489154415, + "loss": 0.1045, + "step": 24251 + }, + { + "epoch": 0.21051900591140701, + "grad_norm": 0.1484375, + "learning_rate": 0.0017468105541643235, + "loss": 0.1416, + "step": 24252 + }, + { + "epoch": 0.21052768639161118, + "grad_norm": 0.70703125, + "learning_rate": 0.0017467898587058637, + "loss": 0.1138, + "step": 24253 + }, + { + "epoch": 0.21053636687181534, + "grad_norm": 0.40234375, + "learning_rate": 0.0017467691625400838, + "loss": 0.1006, + "step": 24254 + }, + { + "epoch": 0.2105450473520195, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017467484656670071, + "loss": 0.082, + "step": 24255 + }, + { + "epoch": 0.21055372783222367, + "grad_norm": 0.1396484375, + "learning_rate": 0.001746727768086656, + "loss": 0.1445, + "step": 24256 + }, + { + "epoch": 0.21056240831242784, + "grad_norm": 0.087890625, + "learning_rate": 0.001746707069799053, + "loss": 0.0796, + "step": 24257 + }, + { + "epoch": 0.210571088792632, + "grad_norm": 0.115234375, + "learning_rate": 0.0017466863708042208, + "loss": 0.1143, + "step": 24258 + }, + { + "epoch": 0.21057976927283617, + "grad_norm": 0.42578125, + "learning_rate": 0.0017466656711021822, + "loss": 0.1553, + "step": 24259 + }, + { + "epoch": 0.21058844975304034, + "grad_norm": 0.474609375, + "learning_rate": 0.0017466449706929597, + "loss": 0.1328, + "step": 24260 + }, + { + "epoch": 0.2105971302332445, + "grad_norm": 0.1796875, + "learning_rate": 0.0017466242695765761, + "loss": 0.0874, + "step": 24261 + }, + { + "epoch": 0.21060581071344867, + "grad_norm": 0.19921875, + "learning_rate": 0.001746603567753054, + "loss": 0.1309, + "step": 24262 + }, + { + "epoch": 0.21061449119365283, + "grad_norm": 0.373046875, + "learning_rate": 0.0017465828652224158, + "loss": 0.0854, + "step": 24263 + }, + { + "epoch": 0.210623171673857, + "grad_norm": 0.8359375, + "learning_rate": 0.001746562161984684, + "loss": 0.1475, + "step": 24264 + }, + { + "epoch": 0.21063185215406116, + "grad_norm": 0.1259765625, + "learning_rate": 0.001746541458039882, + "loss": 0.1016, + "step": 24265 + }, + { + "epoch": 0.21064053263426533, + "grad_norm": 0.81640625, + "learning_rate": 0.0017465207533880315, + "loss": 0.124, + "step": 24266 + }, + { + "epoch": 0.2106492131144695, + "grad_norm": 0.427734375, + "learning_rate": 0.0017465000480291562, + "loss": 0.0938, + "step": 24267 + }, + { + "epoch": 0.21065789359467366, + "grad_norm": 0.61328125, + "learning_rate": 0.0017464793419632778, + "loss": 0.1123, + "step": 24268 + }, + { + "epoch": 0.21066657407487782, + "grad_norm": 0.125, + "learning_rate": 0.0017464586351904194, + "loss": 0.1006, + "step": 24269 + }, + { + "epoch": 0.210675254555082, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017464379277106034, + "loss": 0.127, + "step": 24270 + }, + { + "epoch": 0.21068393503528615, + "grad_norm": 0.2734375, + "learning_rate": 0.001746417219523853, + "loss": 0.104, + "step": 24271 + }, + { + "epoch": 0.21069261551549032, + "grad_norm": 0.279296875, + "learning_rate": 0.00174639651063019, + "loss": 0.1406, + "step": 24272 + }, + { + "epoch": 0.21070129599569448, + "grad_norm": 0.2109375, + "learning_rate": 0.0017463758010296377, + "loss": 0.1123, + "step": 24273 + }, + { + "epoch": 0.21070997647589865, + "grad_norm": 0.82421875, + "learning_rate": 0.0017463550907222184, + "loss": 0.1387, + "step": 24274 + }, + { + "epoch": 0.2107186569561028, + "grad_norm": 0.451171875, + "learning_rate": 0.001746334379707955, + "loss": 0.1758, + "step": 24275 + }, + { + "epoch": 0.21072733743630698, + "grad_norm": 0.111328125, + "learning_rate": 0.0017463136679868698, + "loss": 0.0889, + "step": 24276 + }, + { + "epoch": 0.21073601791651114, + "grad_norm": 0.466796875, + "learning_rate": 0.001746292955558986, + "loss": 0.0859, + "step": 24277 + }, + { + "epoch": 0.2107446983967153, + "grad_norm": 0.2001953125, + "learning_rate": 0.0017462722424243257, + "loss": 0.1006, + "step": 24278 + }, + { + "epoch": 0.21075337887691947, + "grad_norm": 0.220703125, + "learning_rate": 0.0017462515285829121, + "loss": 0.1074, + "step": 24279 + }, + { + "epoch": 0.21076205935712364, + "grad_norm": 0.375, + "learning_rate": 0.0017462308140347674, + "loss": 0.1025, + "step": 24280 + }, + { + "epoch": 0.2107707398373278, + "grad_norm": 0.27734375, + "learning_rate": 0.0017462100987799144, + "loss": 0.1514, + "step": 24281 + }, + { + "epoch": 0.21077942031753197, + "grad_norm": 0.072265625, + "learning_rate": 0.0017461893828183755, + "loss": 0.0996, + "step": 24282 + }, + { + "epoch": 0.21078810079773613, + "grad_norm": 0.412109375, + "learning_rate": 0.0017461686661501739, + "loss": 0.1445, + "step": 24283 + }, + { + "epoch": 0.2107967812779403, + "grad_norm": 0.197265625, + "learning_rate": 0.0017461479487753317, + "loss": 0.1562, + "step": 24284 + }, + { + "epoch": 0.21080546175814446, + "grad_norm": 1.140625, + "learning_rate": 0.001746127230693872, + "loss": 0.1211, + "step": 24285 + }, + { + "epoch": 0.21081414223834863, + "grad_norm": 1.328125, + "learning_rate": 0.001746106511905817, + "loss": 0.1035, + "step": 24286 + }, + { + "epoch": 0.2108228227185528, + "grad_norm": 0.1875, + "learning_rate": 0.0017460857924111901, + "loss": 0.0918, + "step": 24287 + }, + { + "epoch": 0.21083150319875696, + "grad_norm": 0.421875, + "learning_rate": 0.0017460650722100131, + "loss": 0.1143, + "step": 24288 + }, + { + "epoch": 0.21084018367896112, + "grad_norm": 0.123046875, + "learning_rate": 0.0017460443513023093, + "loss": 0.083, + "step": 24289 + }, + { + "epoch": 0.2108488641591653, + "grad_norm": 0.169921875, + "learning_rate": 0.001746023629688101, + "loss": 0.1191, + "step": 24290 + }, + { + "epoch": 0.21085754463936945, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017460029073674108, + "loss": 0.0835, + "step": 24291 + }, + { + "epoch": 0.21086622511957362, + "grad_norm": 0.158203125, + "learning_rate": 0.0017459821843402617, + "loss": 0.1621, + "step": 24292 + }, + { + "epoch": 0.21087490559977778, + "grad_norm": 0.6484375, + "learning_rate": 0.0017459614606066761, + "loss": 0.1172, + "step": 24293 + }, + { + "epoch": 0.21088358607998195, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017459407361666767, + "loss": 0.1367, + "step": 24294 + }, + { + "epoch": 0.21089226656018611, + "grad_norm": 0.455078125, + "learning_rate": 0.0017459200110202864, + "loss": 0.1465, + "step": 24295 + }, + { + "epoch": 0.21090094704039028, + "grad_norm": 0.45703125, + "learning_rate": 0.0017458992851675278, + "loss": 0.083, + "step": 24296 + }, + { + "epoch": 0.21090962752059444, + "grad_norm": 0.2099609375, + "learning_rate": 0.0017458785586084232, + "loss": 0.1113, + "step": 24297 + }, + { + "epoch": 0.2109183080007986, + "grad_norm": 0.1640625, + "learning_rate": 0.0017458578313429953, + "loss": 0.1104, + "step": 24298 + }, + { + "epoch": 0.21092698848100278, + "grad_norm": 0.357421875, + "learning_rate": 0.0017458371033712676, + "loss": 0.1172, + "step": 24299 + }, + { + "epoch": 0.21093566896120694, + "grad_norm": 0.404296875, + "learning_rate": 0.0017458163746932615, + "loss": 0.1074, + "step": 24300 + }, + { + "epoch": 0.2109443494414111, + "grad_norm": 0.373046875, + "learning_rate": 0.0017457956453090008, + "loss": 0.0967, + "step": 24301 + }, + { + "epoch": 0.21095302992161527, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017457749152185074, + "loss": 0.1328, + "step": 24302 + }, + { + "epoch": 0.21096171040181944, + "grad_norm": 0.412109375, + "learning_rate": 0.0017457541844218044, + "loss": 0.1182, + "step": 24303 + }, + { + "epoch": 0.2109703908820236, + "grad_norm": 0.2470703125, + "learning_rate": 0.0017457334529189141, + "loss": 0.0898, + "step": 24304 + }, + { + "epoch": 0.21097907136222777, + "grad_norm": 0.150390625, + "learning_rate": 0.0017457127207098598, + "loss": 0.0977, + "step": 24305 + }, + { + "epoch": 0.21098775184243193, + "grad_norm": 0.54296875, + "learning_rate": 0.0017456919877946635, + "loss": 0.0938, + "step": 24306 + }, + { + "epoch": 0.2109964323226361, + "grad_norm": 0.275390625, + "learning_rate": 0.0017456712541733482, + "loss": 0.082, + "step": 24307 + }, + { + "epoch": 0.21100511280284026, + "grad_norm": 0.283203125, + "learning_rate": 0.0017456505198459363, + "loss": 0.1299, + "step": 24308 + }, + { + "epoch": 0.21101379328304443, + "grad_norm": 0.90234375, + "learning_rate": 0.0017456297848124512, + "loss": 0.1099, + "step": 24309 + }, + { + "epoch": 0.2110224737632486, + "grad_norm": 0.3828125, + "learning_rate": 0.0017456090490729145, + "loss": 0.1104, + "step": 24310 + }, + { + "epoch": 0.21103115424345276, + "grad_norm": 0.078125, + "learning_rate": 0.00174558831262735, + "loss": 0.0762, + "step": 24311 + }, + { + "epoch": 0.21103983472365692, + "grad_norm": 0.111328125, + "learning_rate": 0.0017455675754757794, + "loss": 0.1187, + "step": 24312 + }, + { + "epoch": 0.2110485152038611, + "grad_norm": 0.2373046875, + "learning_rate": 0.001745546837618226, + "loss": 0.1328, + "step": 24313 + }, + { + "epoch": 0.21105719568406525, + "grad_norm": 0.44140625, + "learning_rate": 0.001745526099054712, + "loss": 0.1094, + "step": 24314 + }, + { + "epoch": 0.21106587616426942, + "grad_norm": 0.140625, + "learning_rate": 0.0017455053597852607, + "loss": 0.1035, + "step": 24315 + }, + { + "epoch": 0.21107455664447358, + "grad_norm": 0.34375, + "learning_rate": 0.0017454846198098944, + "loss": 0.1621, + "step": 24316 + }, + { + "epoch": 0.21108323712467775, + "grad_norm": 0.07861328125, + "learning_rate": 0.0017454638791286354, + "loss": 0.0908, + "step": 24317 + }, + { + "epoch": 0.2110919176048819, + "grad_norm": 0.21484375, + "learning_rate": 0.0017454431377415072, + "loss": 0.1025, + "step": 24318 + }, + { + "epoch": 0.21110059808508608, + "grad_norm": 0.2373046875, + "learning_rate": 0.001745422395648532, + "loss": 0.0938, + "step": 24319 + }, + { + "epoch": 0.21110927856529024, + "grad_norm": 0.06884765625, + "learning_rate": 0.0017454016528497328, + "loss": 0.0928, + "step": 24320 + }, + { + "epoch": 0.2111179590454944, + "grad_norm": 0.451171875, + "learning_rate": 0.0017453809093451317, + "loss": 0.1465, + "step": 24321 + }, + { + "epoch": 0.21112663952569857, + "grad_norm": 0.30078125, + "learning_rate": 0.0017453601651347516, + "loss": 0.1396, + "step": 24322 + }, + { + "epoch": 0.21113532000590274, + "grad_norm": 0.216796875, + "learning_rate": 0.0017453394202186158, + "loss": 0.0991, + "step": 24323 + }, + { + "epoch": 0.2111440004861069, + "grad_norm": 0.5625, + "learning_rate": 0.0017453186745967461, + "loss": 0.0752, + "step": 24324 + }, + { + "epoch": 0.21115268096631107, + "grad_norm": 0.34765625, + "learning_rate": 0.0017452979282691662, + "loss": 0.1592, + "step": 24325 + }, + { + "epoch": 0.21116136144651523, + "grad_norm": 0.0849609375, + "learning_rate": 0.0017452771812358973, + "loss": 0.0776, + "step": 24326 + }, + { + "epoch": 0.2111700419267194, + "grad_norm": 0.10986328125, + "learning_rate": 0.0017452564334969635, + "loss": 0.1396, + "step": 24327 + }, + { + "epoch": 0.21117872240692356, + "grad_norm": 0.10498046875, + "learning_rate": 0.001745235685052387, + "loss": 0.1299, + "step": 24328 + }, + { + "epoch": 0.21118740288712773, + "grad_norm": 0.10107421875, + "learning_rate": 0.0017452149359021903, + "loss": 0.085, + "step": 24329 + }, + { + "epoch": 0.2111960833673319, + "grad_norm": 0.357421875, + "learning_rate": 0.0017451941860463962, + "loss": 0.1221, + "step": 24330 + }, + { + "epoch": 0.21120476384753606, + "grad_norm": 0.2099609375, + "learning_rate": 0.0017451734354850274, + "loss": 0.0986, + "step": 24331 + }, + { + "epoch": 0.21121344432774022, + "grad_norm": 0.0810546875, + "learning_rate": 0.001745152684218107, + "loss": 0.1035, + "step": 24332 + }, + { + "epoch": 0.2112221248079444, + "grad_norm": 0.73046875, + "learning_rate": 0.001745131932245657, + "loss": 0.0781, + "step": 24333 + }, + { + "epoch": 0.21123080528814853, + "grad_norm": 0.09423828125, + "learning_rate": 0.0017451111795677005, + "loss": 0.0845, + "step": 24334 + }, + { + "epoch": 0.2112394857683527, + "grad_norm": 0.10205078125, + "learning_rate": 0.00174509042618426, + "loss": 0.1523, + "step": 24335 + }, + { + "epoch": 0.21124816624855686, + "grad_norm": 0.259765625, + "learning_rate": 0.0017450696720953582, + "loss": 0.0859, + "step": 24336 + }, + { + "epoch": 0.21125684672876102, + "grad_norm": 0.345703125, + "learning_rate": 0.001745048917301018, + "loss": 0.1016, + "step": 24337 + }, + { + "epoch": 0.2112655272089652, + "grad_norm": 0.1953125, + "learning_rate": 0.0017450281618012623, + "loss": 0.0967, + "step": 24338 + }, + { + "epoch": 0.21127420768916935, + "grad_norm": 0.6953125, + "learning_rate": 0.0017450074055961134, + "loss": 0.1367, + "step": 24339 + }, + { + "epoch": 0.21128288816937352, + "grad_norm": 0.92578125, + "learning_rate": 0.0017449866486855935, + "loss": 0.1152, + "step": 24340 + }, + { + "epoch": 0.21129156864957768, + "grad_norm": 0.12109375, + "learning_rate": 0.0017449658910697265, + "loss": 0.1436, + "step": 24341 + }, + { + "epoch": 0.21130024912978185, + "grad_norm": 0.1201171875, + "learning_rate": 0.0017449451327485343, + "loss": 0.105, + "step": 24342 + }, + { + "epoch": 0.211308929609986, + "grad_norm": 0.353515625, + "learning_rate": 0.0017449243737220398, + "loss": 0.1006, + "step": 24343 + }, + { + "epoch": 0.21131761009019018, + "grad_norm": 0.24609375, + "learning_rate": 0.0017449036139902657, + "loss": 0.1001, + "step": 24344 + }, + { + "epoch": 0.21132629057039434, + "grad_norm": 0.4609375, + "learning_rate": 0.001744882853553235, + "loss": 0.0908, + "step": 24345 + }, + { + "epoch": 0.2113349710505985, + "grad_norm": 0.302734375, + "learning_rate": 0.0017448620924109695, + "loss": 0.1445, + "step": 24346 + }, + { + "epoch": 0.21134365153080267, + "grad_norm": 0.234375, + "learning_rate": 0.0017448413305634928, + "loss": 0.1074, + "step": 24347 + }, + { + "epoch": 0.21135233201100684, + "grad_norm": 0.6015625, + "learning_rate": 0.0017448205680108276, + "loss": 0.1016, + "step": 24348 + }, + { + "epoch": 0.211361012491211, + "grad_norm": 0.087890625, + "learning_rate": 0.0017447998047529958, + "loss": 0.1006, + "step": 24349 + }, + { + "epoch": 0.21136969297141517, + "grad_norm": 0.375, + "learning_rate": 0.0017447790407900208, + "loss": 0.0938, + "step": 24350 + }, + { + "epoch": 0.21137837345161933, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017447582761219251, + "loss": 0.0996, + "step": 24351 + }, + { + "epoch": 0.2113870539318235, + "grad_norm": 0.52734375, + "learning_rate": 0.0017447375107487318, + "loss": 0.1133, + "step": 24352 + }, + { + "epoch": 0.21139573441202766, + "grad_norm": 0.40234375, + "learning_rate": 0.001744716744670463, + "loss": 0.0879, + "step": 24353 + }, + { + "epoch": 0.21140441489223183, + "grad_norm": 0.162109375, + "learning_rate": 0.0017446959778871416, + "loss": 0.0928, + "step": 24354 + }, + { + "epoch": 0.211413095372436, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017446752103987904, + "loss": 0.1123, + "step": 24355 + }, + { + "epoch": 0.21142177585264016, + "grad_norm": 0.1640625, + "learning_rate": 0.0017446544422054322, + "loss": 0.1206, + "step": 24356 + }, + { + "epoch": 0.21143045633284432, + "grad_norm": 0.53515625, + "learning_rate": 0.0017446336733070893, + "loss": 0.1094, + "step": 24357 + }, + { + "epoch": 0.2114391368130485, + "grad_norm": 0.1728515625, + "learning_rate": 0.001744612903703785, + "loss": 0.1147, + "step": 24358 + }, + { + "epoch": 0.21144781729325265, + "grad_norm": 0.54296875, + "learning_rate": 0.0017445921333955415, + "loss": 0.0952, + "step": 24359 + }, + { + "epoch": 0.21145649777345682, + "grad_norm": 0.2470703125, + "learning_rate": 0.0017445713623823818, + "loss": 0.1406, + "step": 24360 + }, + { + "epoch": 0.21146517825366098, + "grad_norm": 0.5390625, + "learning_rate": 0.0017445505906643287, + "loss": 0.1367, + "step": 24361 + }, + { + "epoch": 0.21147385873386515, + "grad_norm": 0.158203125, + "learning_rate": 0.0017445298182414044, + "loss": 0.1025, + "step": 24362 + }, + { + "epoch": 0.21148253921406931, + "grad_norm": 0.07568359375, + "learning_rate": 0.001744509045113632, + "loss": 0.0942, + "step": 24363 + }, + { + "epoch": 0.21149121969427348, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017444882712810343, + "loss": 0.1191, + "step": 24364 + }, + { + "epoch": 0.21149990017447765, + "grad_norm": 0.51171875, + "learning_rate": 0.001744467496743634, + "loss": 0.1221, + "step": 24365 + }, + { + "epoch": 0.2115085806546818, + "grad_norm": 0.6328125, + "learning_rate": 0.0017444467215014538, + "loss": 0.126, + "step": 24366 + }, + { + "epoch": 0.21151726113488598, + "grad_norm": 0.578125, + "learning_rate": 0.0017444259455545162, + "loss": 0.0874, + "step": 24367 + }, + { + "epoch": 0.21152594161509014, + "grad_norm": 0.384765625, + "learning_rate": 0.0017444051689028445, + "loss": 0.1211, + "step": 24368 + }, + { + "epoch": 0.2115346220952943, + "grad_norm": 0.359375, + "learning_rate": 0.0017443843915464605, + "loss": 0.1021, + "step": 24369 + }, + { + "epoch": 0.21154330257549847, + "grad_norm": 0.96875, + "learning_rate": 0.0017443636134853874, + "loss": 0.1084, + "step": 24370 + }, + { + "epoch": 0.21155198305570264, + "grad_norm": 0.185546875, + "learning_rate": 0.0017443428347196479, + "loss": 0.1162, + "step": 24371 + }, + { + "epoch": 0.2115606635359068, + "grad_norm": 0.84375, + "learning_rate": 0.0017443220552492652, + "loss": 0.1089, + "step": 24372 + }, + { + "epoch": 0.21156934401611097, + "grad_norm": 0.16796875, + "learning_rate": 0.0017443012750742613, + "loss": 0.0996, + "step": 24373 + }, + { + "epoch": 0.21157802449631513, + "grad_norm": 0.58984375, + "learning_rate": 0.0017442804941946591, + "loss": 0.1074, + "step": 24374 + }, + { + "epoch": 0.2115867049765193, + "grad_norm": 0.734375, + "learning_rate": 0.0017442597126104816, + "loss": 0.1328, + "step": 24375 + }, + { + "epoch": 0.21159538545672346, + "grad_norm": 0.138671875, + "learning_rate": 0.0017442389303217512, + "loss": 0.1016, + "step": 24376 + }, + { + "epoch": 0.21160406593692763, + "grad_norm": 0.435546875, + "learning_rate": 0.0017442181473284908, + "loss": 0.1719, + "step": 24377 + }, + { + "epoch": 0.2116127464171318, + "grad_norm": 0.408203125, + "learning_rate": 0.001744197363630723, + "loss": 0.1475, + "step": 24378 + }, + { + "epoch": 0.21162142689733596, + "grad_norm": 0.6171875, + "learning_rate": 0.0017441765792284712, + "loss": 0.1113, + "step": 24379 + }, + { + "epoch": 0.21163010737754012, + "grad_norm": 0.123046875, + "learning_rate": 0.0017441557941217571, + "loss": 0.1289, + "step": 24380 + }, + { + "epoch": 0.2116387878577443, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017441350083106041, + "loss": 0.1221, + "step": 24381 + }, + { + "epoch": 0.21164746833794845, + "grad_norm": 0.50390625, + "learning_rate": 0.0017441142217950348, + "loss": 0.1328, + "step": 24382 + }, + { + "epoch": 0.21165614881815262, + "grad_norm": 0.09375, + "learning_rate": 0.0017440934345750714, + "loss": 0.0991, + "step": 24383 + }, + { + "epoch": 0.21166482929835678, + "grad_norm": 0.11083984375, + "learning_rate": 0.0017440726466507375, + "loss": 0.1074, + "step": 24384 + }, + { + "epoch": 0.21167350977856095, + "grad_norm": 1.0546875, + "learning_rate": 0.0017440518580220552, + "loss": 0.4121, + "step": 24385 + }, + { + "epoch": 0.2116821902587651, + "grad_norm": 0.53515625, + "learning_rate": 0.001744031068689048, + "loss": 0.0825, + "step": 24386 + }, + { + "epoch": 0.21169087073896928, + "grad_norm": 0.349609375, + "learning_rate": 0.0017440102786517374, + "loss": 0.0933, + "step": 24387 + }, + { + "epoch": 0.21169955121917344, + "grad_norm": 0.08203125, + "learning_rate": 0.0017439894879101472, + "loss": 0.0669, + "step": 24388 + }, + { + "epoch": 0.2117082316993776, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017439686964642995, + "loss": 0.1167, + "step": 24389 + }, + { + "epoch": 0.21171691217958177, + "grad_norm": 0.23828125, + "learning_rate": 0.0017439479043142175, + "loss": 0.0967, + "step": 24390 + }, + { + "epoch": 0.21172559265978594, + "grad_norm": 0.8671875, + "learning_rate": 0.001743927111459924, + "loss": 0.1816, + "step": 24391 + }, + { + "epoch": 0.2117342731399901, + "grad_norm": 0.291015625, + "learning_rate": 0.0017439063179014414, + "loss": 0.1094, + "step": 24392 + }, + { + "epoch": 0.21174295362019427, + "grad_norm": 0.19921875, + "learning_rate": 0.0017438855236387922, + "loss": 0.1045, + "step": 24393 + }, + { + "epoch": 0.21175163410039843, + "grad_norm": 0.07763671875, + "learning_rate": 0.0017438647286719996, + "loss": 0.0977, + "step": 24394 + }, + { + "epoch": 0.2117603145806026, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017438439330010862, + "loss": 0.0928, + "step": 24395 + }, + { + "epoch": 0.21176899506080676, + "grad_norm": 0.83203125, + "learning_rate": 0.0017438231366260749, + "loss": 0.1216, + "step": 24396 + }, + { + "epoch": 0.21177767554101093, + "grad_norm": 0.578125, + "learning_rate": 0.001743802339546988, + "loss": 0.104, + "step": 24397 + }, + { + "epoch": 0.2117863560212151, + "grad_norm": 0.09375, + "learning_rate": 0.0017437815417638486, + "loss": 0.1025, + "step": 24398 + }, + { + "epoch": 0.21179503650141926, + "grad_norm": 0.32421875, + "learning_rate": 0.0017437607432766798, + "loss": 0.1367, + "step": 24399 + }, + { + "epoch": 0.21180371698162342, + "grad_norm": 0.416015625, + "learning_rate": 0.0017437399440855038, + "loss": 0.1055, + "step": 24400 + }, + { + "epoch": 0.2118123974618276, + "grad_norm": 0.337890625, + "learning_rate": 0.001743719144190343, + "loss": 0.1104, + "step": 24401 + }, + { + "epoch": 0.21182107794203175, + "grad_norm": 0.2470703125, + "learning_rate": 0.0017436983435912211, + "loss": 0.1206, + "step": 24402 + }, + { + "epoch": 0.21182975842223592, + "grad_norm": 0.142578125, + "learning_rate": 0.00174367754228816, + "loss": 0.0918, + "step": 24403 + }, + { + "epoch": 0.21183843890244008, + "grad_norm": 0.69140625, + "learning_rate": 0.001743656740281183, + "loss": 0.1328, + "step": 24404 + }, + { + "epoch": 0.21184711938264425, + "grad_norm": 0.263671875, + "learning_rate": 0.0017436359375703125, + "loss": 0.0815, + "step": 24405 + }, + { + "epoch": 0.21185579986284842, + "grad_norm": 0.16796875, + "learning_rate": 0.0017436151341555716, + "loss": 0.105, + "step": 24406 + }, + { + "epoch": 0.21186448034305258, + "grad_norm": 0.08056640625, + "learning_rate": 0.001743594330036983, + "loss": 0.0947, + "step": 24407 + }, + { + "epoch": 0.21187316082325675, + "grad_norm": 0.515625, + "learning_rate": 0.001743573525214569, + "loss": 0.1641, + "step": 24408 + }, + { + "epoch": 0.2118818413034609, + "grad_norm": 0.4921875, + "learning_rate": 0.0017435527196883527, + "loss": 0.1445, + "step": 24409 + }, + { + "epoch": 0.21189052178366508, + "grad_norm": 0.16015625, + "learning_rate": 0.0017435319134583571, + "loss": 0.1338, + "step": 24410 + }, + { + "epoch": 0.21189920226386924, + "grad_norm": 0.439453125, + "learning_rate": 0.0017435111065246043, + "loss": 0.103, + "step": 24411 + }, + { + "epoch": 0.2119078827440734, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017434902988871178, + "loss": 0.1196, + "step": 24412 + }, + { + "epoch": 0.21191656322427757, + "grad_norm": 0.259765625, + "learning_rate": 0.0017434694905459197, + "loss": 0.0908, + "step": 24413 + }, + { + "epoch": 0.21192524370448174, + "grad_norm": 0.4765625, + "learning_rate": 0.001743448681501033, + "loss": 0.0869, + "step": 24414 + }, + { + "epoch": 0.2119339241846859, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017434278717524806, + "loss": 0.1064, + "step": 24415 + }, + { + "epoch": 0.21194260466489007, + "grad_norm": 0.19921875, + "learning_rate": 0.001743407061300285, + "loss": 0.1787, + "step": 24416 + }, + { + "epoch": 0.21195128514509423, + "grad_norm": 0.51953125, + "learning_rate": 0.0017433862501444696, + "loss": 0.1152, + "step": 24417 + }, + { + "epoch": 0.2119599656252984, + "grad_norm": 0.18359375, + "learning_rate": 0.001743365438285056, + "loss": 0.0977, + "step": 24418 + }, + { + "epoch": 0.21196864610550256, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017433446257220679, + "loss": 0.1099, + "step": 24419 + }, + { + "epoch": 0.21197732658570673, + "grad_norm": 0.279296875, + "learning_rate": 0.001743323812455528, + "loss": 0.0845, + "step": 24420 + }, + { + "epoch": 0.2119860070659109, + "grad_norm": 0.1328125, + "learning_rate": 0.0017433029984854586, + "loss": 0.0859, + "step": 24421 + }, + { + "epoch": 0.21199468754611506, + "grad_norm": 0.2080078125, + "learning_rate": 0.001743282183811883, + "loss": 0.1367, + "step": 24422 + }, + { + "epoch": 0.21200336802631922, + "grad_norm": 0.3671875, + "learning_rate": 0.001743261368434823, + "loss": 0.0967, + "step": 24423 + }, + { + "epoch": 0.2120120485065234, + "grad_norm": 0.5390625, + "learning_rate": 0.0017432405523543026, + "loss": 0.085, + "step": 24424 + }, + { + "epoch": 0.21202072898672755, + "grad_norm": 0.240234375, + "learning_rate": 0.0017432197355703438, + "loss": 0.0859, + "step": 24425 + }, + { + "epoch": 0.21202940946693172, + "grad_norm": 0.369140625, + "learning_rate": 0.0017431989180829697, + "loss": 0.123, + "step": 24426 + }, + { + "epoch": 0.21203808994713588, + "grad_norm": 0.53515625, + "learning_rate": 0.0017431780998922028, + "loss": 0.1377, + "step": 24427 + }, + { + "epoch": 0.21204677042734005, + "grad_norm": 0.19921875, + "learning_rate": 0.0017431572809980661, + "loss": 0.1289, + "step": 24428 + }, + { + "epoch": 0.2120554509075442, + "grad_norm": 0.37109375, + "learning_rate": 0.0017431364614005822, + "loss": 0.0947, + "step": 24429 + }, + { + "epoch": 0.21206413138774838, + "grad_norm": 0.28515625, + "learning_rate": 0.001743115641099774, + "loss": 0.0928, + "step": 24430 + }, + { + "epoch": 0.21207281186795254, + "grad_norm": 0.8046875, + "learning_rate": 0.0017430948200956643, + "loss": 0.1768, + "step": 24431 + }, + { + "epoch": 0.2120814923481567, + "grad_norm": 0.357421875, + "learning_rate": 0.0017430739983882752, + "loss": 0.0664, + "step": 24432 + }, + { + "epoch": 0.21209017282836087, + "grad_norm": 0.201171875, + "learning_rate": 0.0017430531759776307, + "loss": 0.1396, + "step": 24433 + }, + { + "epoch": 0.21209885330856504, + "grad_norm": 0.1875, + "learning_rate": 0.0017430323528637526, + "loss": 0.1416, + "step": 24434 + }, + { + "epoch": 0.2121075337887692, + "grad_norm": 0.44140625, + "learning_rate": 0.001743011529046664, + "loss": 0.0796, + "step": 24435 + }, + { + "epoch": 0.21211621426897337, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017429907045263878, + "loss": 0.0996, + "step": 24436 + }, + { + "epoch": 0.21212489474917753, + "grad_norm": 0.44921875, + "learning_rate": 0.0017429698793029462, + "loss": 0.1162, + "step": 24437 + }, + { + "epoch": 0.2121335752293817, + "grad_norm": 0.087890625, + "learning_rate": 0.0017429490533763627, + "loss": 0.1221, + "step": 24438 + }, + { + "epoch": 0.21214225570958586, + "grad_norm": 0.462890625, + "learning_rate": 0.00174292822674666, + "loss": 0.0986, + "step": 24439 + }, + { + "epoch": 0.21215093618979003, + "grad_norm": 0.22265625, + "learning_rate": 0.0017429073994138604, + "loss": 0.1348, + "step": 24440 + }, + { + "epoch": 0.2121596166699942, + "grad_norm": 0.171875, + "learning_rate": 0.001742886571377987, + "loss": 0.124, + "step": 24441 + }, + { + "epoch": 0.21216829715019836, + "grad_norm": 0.447265625, + "learning_rate": 0.0017428657426390624, + "loss": 0.1514, + "step": 24442 + }, + { + "epoch": 0.21217697763040252, + "grad_norm": 0.10400390625, + "learning_rate": 0.0017428449131971096, + "loss": 0.0996, + "step": 24443 + }, + { + "epoch": 0.2121856581106067, + "grad_norm": 0.33203125, + "learning_rate": 0.0017428240830521511, + "loss": 0.1279, + "step": 24444 + }, + { + "epoch": 0.21219433859081085, + "grad_norm": 0.59375, + "learning_rate": 0.00174280325220421, + "loss": 0.1079, + "step": 24445 + }, + { + "epoch": 0.21220301907101502, + "grad_norm": 0.341796875, + "learning_rate": 0.0017427824206533089, + "loss": 0.0952, + "step": 24446 + }, + { + "epoch": 0.21221169955121919, + "grad_norm": 0.41015625, + "learning_rate": 0.0017427615883994704, + "loss": 0.1001, + "step": 24447 + }, + { + "epoch": 0.21222038003142335, + "grad_norm": 0.07275390625, + "learning_rate": 0.0017427407554427177, + "loss": 0.0835, + "step": 24448 + }, + { + "epoch": 0.21222906051162752, + "grad_norm": 0.31640625, + "learning_rate": 0.0017427199217830735, + "loss": 0.1523, + "step": 24449 + }, + { + "epoch": 0.21223774099183168, + "grad_norm": 0.79296875, + "learning_rate": 0.00174269908742056, + "loss": 0.0923, + "step": 24450 + }, + { + "epoch": 0.21224642147203585, + "grad_norm": 0.08447265625, + "learning_rate": 0.0017426782523552009, + "loss": 0.0981, + "step": 24451 + }, + { + "epoch": 0.21225510195224, + "grad_norm": 0.15625, + "learning_rate": 0.0017426574165870182, + "loss": 0.0908, + "step": 24452 + }, + { + "epoch": 0.21226378243244418, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017426365801160354, + "loss": 0.1162, + "step": 24453 + }, + { + "epoch": 0.21227246291264834, + "grad_norm": 1.40625, + "learning_rate": 0.0017426157429422746, + "loss": 0.2793, + "step": 24454 + }, + { + "epoch": 0.2122811433928525, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017425949050657588, + "loss": 0.0986, + "step": 24455 + }, + { + "epoch": 0.21228982387305667, + "grad_norm": 0.146484375, + "learning_rate": 0.001742574066486511, + "loss": 0.1104, + "step": 24456 + }, + { + "epoch": 0.21229850435326084, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017425532272045538, + "loss": 0.0859, + "step": 24457 + }, + { + "epoch": 0.21230718483346497, + "grad_norm": 0.255859375, + "learning_rate": 0.0017425323872199102, + "loss": 0.1138, + "step": 24458 + }, + { + "epoch": 0.21231586531366914, + "grad_norm": 0.06982421875, + "learning_rate": 0.0017425115465326028, + "loss": 0.0889, + "step": 24459 + }, + { + "epoch": 0.2123245457938733, + "grad_norm": 0.50390625, + "learning_rate": 0.001742490705142654, + "loss": 0.1543, + "step": 24460 + }, + { + "epoch": 0.21233322627407747, + "grad_norm": 0.53515625, + "learning_rate": 0.0017424698630500873, + "loss": 0.0889, + "step": 24461 + }, + { + "epoch": 0.21234190675428163, + "grad_norm": 0.236328125, + "learning_rate": 0.0017424490202549254, + "loss": 0.1279, + "step": 24462 + }, + { + "epoch": 0.2123505872344858, + "grad_norm": 0.1171875, + "learning_rate": 0.0017424281767571908, + "loss": 0.104, + "step": 24463 + }, + { + "epoch": 0.21235926771468996, + "grad_norm": 0.404296875, + "learning_rate": 0.0017424073325569063, + "loss": 0.0972, + "step": 24464 + }, + { + "epoch": 0.21236794819489413, + "grad_norm": 0.10595703125, + "learning_rate": 0.001742386487654095, + "loss": 0.1279, + "step": 24465 + }, + { + "epoch": 0.2123766286750983, + "grad_norm": 0.3828125, + "learning_rate": 0.0017423656420487792, + "loss": 0.0957, + "step": 24466 + }, + { + "epoch": 0.21238530915530246, + "grad_norm": 0.462890625, + "learning_rate": 0.001742344795740982, + "loss": 0.252, + "step": 24467 + }, + { + "epoch": 0.21239398963550662, + "grad_norm": 0.796875, + "learning_rate": 0.0017423239487307266, + "loss": 0.1084, + "step": 24468 + }, + { + "epoch": 0.2124026701157108, + "grad_norm": 0.283203125, + "learning_rate": 0.001742303101018035, + "loss": 0.0879, + "step": 24469 + }, + { + "epoch": 0.21241135059591496, + "grad_norm": 0.154296875, + "learning_rate": 0.0017422822526029305, + "loss": 0.0928, + "step": 24470 + }, + { + "epoch": 0.21242003107611912, + "grad_norm": 0.4453125, + "learning_rate": 0.0017422614034854355, + "loss": 0.125, + "step": 24471 + }, + { + "epoch": 0.21242871155632329, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017422405536655734, + "loss": 0.1226, + "step": 24472 + }, + { + "epoch": 0.21243739203652745, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017422197031433666, + "loss": 0.124, + "step": 24473 + }, + { + "epoch": 0.21244607251673162, + "grad_norm": 0.1826171875, + "learning_rate": 0.0017421988519188381, + "loss": 0.127, + "step": 24474 + }, + { + "epoch": 0.21245475299693578, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017421779999920103, + "loss": 0.1108, + "step": 24475 + }, + { + "epoch": 0.21246343347713995, + "grad_norm": 0.8359375, + "learning_rate": 0.0017421571473629066, + "loss": 0.1318, + "step": 24476 + }, + { + "epoch": 0.2124721139573441, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017421362940315496, + "loss": 0.166, + "step": 24477 + }, + { + "epoch": 0.21248079443754828, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017421154399979617, + "loss": 0.1729, + "step": 24478 + }, + { + "epoch": 0.21248947491775244, + "grad_norm": 0.1748046875, + "learning_rate": 0.001742094585262166, + "loss": 0.1182, + "step": 24479 + }, + { + "epoch": 0.2124981553979566, + "grad_norm": 0.259765625, + "learning_rate": 0.0017420737298241853, + "loss": 0.1484, + "step": 24480 + }, + { + "epoch": 0.21250683587816077, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017420528736840424, + "loss": 0.1094, + "step": 24481 + }, + { + "epoch": 0.21251551635836494, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017420320168417603, + "loss": 0.0894, + "step": 24482 + }, + { + "epoch": 0.2125241968385691, + "grad_norm": 0.43359375, + "learning_rate": 0.0017420111592973614, + "loss": 0.1172, + "step": 24483 + }, + { + "epoch": 0.21253287731877327, + "grad_norm": 0.345703125, + "learning_rate": 0.001741990301050869, + "loss": 0.1025, + "step": 24484 + }, + { + "epoch": 0.21254155779897743, + "grad_norm": 0.703125, + "learning_rate": 0.0017419694421023054, + "loss": 0.1377, + "step": 24485 + }, + { + "epoch": 0.2125502382791816, + "grad_norm": 0.1904296875, + "learning_rate": 0.001741948582451694, + "loss": 0.0786, + "step": 24486 + }, + { + "epoch": 0.21255891875938576, + "grad_norm": 0.1328125, + "learning_rate": 0.001741927722099057, + "loss": 0.1182, + "step": 24487 + }, + { + "epoch": 0.21256759923958993, + "grad_norm": 1.359375, + "learning_rate": 0.0017419068610444175, + "loss": 0.3262, + "step": 24488 + }, + { + "epoch": 0.2125762797197941, + "grad_norm": 0.115234375, + "learning_rate": 0.0017418859992877984, + "loss": 0.1084, + "step": 24489 + }, + { + "epoch": 0.21258496019999826, + "grad_norm": 0.173828125, + "learning_rate": 0.0017418651368292224, + "loss": 0.0967, + "step": 24490 + }, + { + "epoch": 0.21259364068020242, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017418442736687122, + "loss": 0.1084, + "step": 24491 + }, + { + "epoch": 0.2126023211604066, + "grad_norm": 0.33984375, + "learning_rate": 0.001741823409806291, + "loss": 0.1055, + "step": 24492 + }, + { + "epoch": 0.21261100164061075, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017418025452419813, + "loss": 0.1309, + "step": 24493 + }, + { + "epoch": 0.21261968212081492, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017417816799758059, + "loss": 0.0918, + "step": 24494 + }, + { + "epoch": 0.21262836260101908, + "grad_norm": 0.140625, + "learning_rate": 0.0017417608140077876, + "loss": 0.1045, + "step": 24495 + }, + { + "epoch": 0.21263704308122325, + "grad_norm": 0.474609375, + "learning_rate": 0.0017417399473379494, + "loss": 0.0889, + "step": 24496 + }, + { + "epoch": 0.2126457235614274, + "grad_norm": 0.67578125, + "learning_rate": 0.0017417190799663142, + "loss": 0.0977, + "step": 24497 + }, + { + "epoch": 0.21265440404163158, + "grad_norm": 1.265625, + "learning_rate": 0.0017416982118929046, + "loss": 0.166, + "step": 24498 + }, + { + "epoch": 0.21266308452183574, + "grad_norm": 0.376953125, + "learning_rate": 0.001741677343117743, + "loss": 0.084, + "step": 24499 + }, + { + "epoch": 0.2126717650020399, + "grad_norm": 0.337890625, + "learning_rate": 0.0017416564736408532, + "loss": 0.0776, + "step": 24500 + }, + { + "epoch": 0.21268044548224407, + "grad_norm": 0.3515625, + "learning_rate": 0.0017416356034622574, + "loss": 0.1123, + "step": 24501 + }, + { + "epoch": 0.21268912596244824, + "grad_norm": 0.220703125, + "learning_rate": 0.0017416147325819786, + "loss": 0.0747, + "step": 24502 + }, + { + "epoch": 0.2126978064426524, + "grad_norm": 0.322265625, + "learning_rate": 0.0017415938610000395, + "loss": 0.0981, + "step": 24503 + }, + { + "epoch": 0.21270648692285657, + "grad_norm": 0.2353515625, + "learning_rate": 0.001741572988716463, + "loss": 0.0972, + "step": 24504 + }, + { + "epoch": 0.21271516740306073, + "grad_norm": 0.296875, + "learning_rate": 0.001741552115731272, + "loss": 0.0864, + "step": 24505 + }, + { + "epoch": 0.2127238478832649, + "grad_norm": 1.5078125, + "learning_rate": 0.001741531242044489, + "loss": 0.1387, + "step": 24506 + }, + { + "epoch": 0.21273252836346906, + "grad_norm": 0.353515625, + "learning_rate": 0.0017415103676561372, + "loss": 0.1055, + "step": 24507 + }, + { + "epoch": 0.21274120884367323, + "grad_norm": 0.1796875, + "learning_rate": 0.0017414894925662392, + "loss": 0.0908, + "step": 24508 + }, + { + "epoch": 0.2127498893238774, + "grad_norm": 0.10693359375, + "learning_rate": 0.0017414686167748182, + "loss": 0.1094, + "step": 24509 + }, + { + "epoch": 0.21275856980408156, + "grad_norm": 1.0, + "learning_rate": 0.001741447740281897, + "loss": 0.1162, + "step": 24510 + }, + { + "epoch": 0.21276725028428572, + "grad_norm": 0.46484375, + "learning_rate": 0.0017414268630874976, + "loss": 0.1592, + "step": 24511 + }, + { + "epoch": 0.2127759307644899, + "grad_norm": 0.220703125, + "learning_rate": 0.0017414059851916437, + "loss": 0.1426, + "step": 24512 + }, + { + "epoch": 0.21278461124469406, + "grad_norm": 0.349609375, + "learning_rate": 0.0017413851065943577, + "loss": 0.1226, + "step": 24513 + }, + { + "epoch": 0.21279329172489822, + "grad_norm": 0.388671875, + "learning_rate": 0.0017413642272956628, + "loss": 0.1035, + "step": 24514 + }, + { + "epoch": 0.21280197220510239, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017413433472955812, + "loss": 0.0791, + "step": 24515 + }, + { + "epoch": 0.21281065268530655, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017413224665941366, + "loss": 0.126, + "step": 24516 + }, + { + "epoch": 0.21281933316551072, + "grad_norm": 0.546875, + "learning_rate": 0.0017413015851913511, + "loss": 0.0879, + "step": 24517 + }, + { + "epoch": 0.21282801364571488, + "grad_norm": 0.072265625, + "learning_rate": 0.0017412807030872481, + "loss": 0.0801, + "step": 24518 + }, + { + "epoch": 0.21283669412591905, + "grad_norm": 0.0966796875, + "learning_rate": 0.0017412598202818502, + "loss": 0.0767, + "step": 24519 + }, + { + "epoch": 0.2128453746061232, + "grad_norm": 0.314453125, + "learning_rate": 0.0017412389367751798, + "loss": 0.1328, + "step": 24520 + }, + { + "epoch": 0.21285405508632738, + "grad_norm": 0.0859375, + "learning_rate": 0.0017412180525672603, + "loss": 0.0811, + "step": 24521 + }, + { + "epoch": 0.21286273556653154, + "grad_norm": 0.0947265625, + "learning_rate": 0.0017411971676581147, + "loss": 0.125, + "step": 24522 + }, + { + "epoch": 0.2128714160467357, + "grad_norm": 0.22265625, + "learning_rate": 0.001741176282047765, + "loss": 0.1084, + "step": 24523 + }, + { + "epoch": 0.21288009652693987, + "grad_norm": 0.259765625, + "learning_rate": 0.0017411553957362347, + "loss": 0.1143, + "step": 24524 + }, + { + "epoch": 0.21288877700714404, + "grad_norm": 0.07958984375, + "learning_rate": 0.0017411345087235467, + "loss": 0.0952, + "step": 24525 + }, + { + "epoch": 0.2128974574873482, + "grad_norm": 0.201171875, + "learning_rate": 0.0017411136210097235, + "loss": 0.0933, + "step": 24526 + }, + { + "epoch": 0.21290613796755237, + "grad_norm": 0.1240234375, + "learning_rate": 0.001741092732594788, + "loss": 0.0688, + "step": 24527 + }, + { + "epoch": 0.21291481844775653, + "grad_norm": 0.50390625, + "learning_rate": 0.001741071843478763, + "loss": 0.1045, + "step": 24528 + }, + { + "epoch": 0.2129234989279607, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017410509536616718, + "loss": 0.1245, + "step": 24529 + }, + { + "epoch": 0.21293217940816486, + "grad_norm": 0.76953125, + "learning_rate": 0.001741030063143537, + "loss": 0.1533, + "step": 24530 + }, + { + "epoch": 0.21294085988836903, + "grad_norm": 0.482421875, + "learning_rate": 0.001741009171924381, + "loss": 0.1191, + "step": 24531 + }, + { + "epoch": 0.2129495403685732, + "grad_norm": 0.130859375, + "learning_rate": 0.0017409882800042272, + "loss": 0.0991, + "step": 24532 + }, + { + "epoch": 0.21295822084877736, + "grad_norm": 0.16015625, + "learning_rate": 0.001740967387383098, + "loss": 0.0991, + "step": 24533 + }, + { + "epoch": 0.21296690132898152, + "grad_norm": 0.267578125, + "learning_rate": 0.0017409464940610169, + "loss": 0.1147, + "step": 24534 + }, + { + "epoch": 0.2129755818091857, + "grad_norm": 0.60546875, + "learning_rate": 0.0017409256000380058, + "loss": 0.1201, + "step": 24535 + }, + { + "epoch": 0.21298426228938985, + "grad_norm": 0.8203125, + "learning_rate": 0.0017409047053140885, + "loss": 0.1494, + "step": 24536 + }, + { + "epoch": 0.21299294276959402, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017408838098892877, + "loss": 0.1045, + "step": 24537 + }, + { + "epoch": 0.21300162324979818, + "grad_norm": 0.35546875, + "learning_rate": 0.0017408629137636255, + "loss": 0.1172, + "step": 24538 + }, + { + "epoch": 0.21301030373000235, + "grad_norm": 0.2734375, + "learning_rate": 0.0017408420169371254, + "loss": 0.1094, + "step": 24539 + }, + { + "epoch": 0.2130189842102065, + "grad_norm": 0.1669921875, + "learning_rate": 0.00174082111940981, + "loss": 0.0977, + "step": 24540 + }, + { + "epoch": 0.21302766469041068, + "grad_norm": 0.376953125, + "learning_rate": 0.0017408002211817025, + "loss": 0.0811, + "step": 24541 + }, + { + "epoch": 0.21303634517061484, + "grad_norm": 0.224609375, + "learning_rate": 0.001740779322252825, + "loss": 0.0962, + "step": 24542 + }, + { + "epoch": 0.213045025650819, + "grad_norm": 0.6171875, + "learning_rate": 0.0017407584226232014, + "loss": 0.125, + "step": 24543 + }, + { + "epoch": 0.21305370613102317, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017407375222928537, + "loss": 0.123, + "step": 24544 + }, + { + "epoch": 0.21306238661122734, + "grad_norm": 0.14453125, + "learning_rate": 0.0017407166212618052, + "loss": 0.1309, + "step": 24545 + }, + { + "epoch": 0.2130710670914315, + "grad_norm": 0.361328125, + "learning_rate": 0.0017406957195300785, + "loss": 0.0928, + "step": 24546 + }, + { + "epoch": 0.21307974757163567, + "grad_norm": 0.8515625, + "learning_rate": 0.001740674817097697, + "loss": 0.2422, + "step": 24547 + }, + { + "epoch": 0.21308842805183983, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017406539139646828, + "loss": 0.1006, + "step": 24548 + }, + { + "epoch": 0.213097108532044, + "grad_norm": 0.267578125, + "learning_rate": 0.001740633010131059, + "loss": 0.1699, + "step": 24549 + }, + { + "epoch": 0.21310578901224816, + "grad_norm": 0.24609375, + "learning_rate": 0.001740612105596849, + "loss": 0.1211, + "step": 24550 + }, + { + "epoch": 0.21311446949245233, + "grad_norm": 0.6796875, + "learning_rate": 0.001740591200362075, + "loss": 0.0874, + "step": 24551 + }, + { + "epoch": 0.2131231499726565, + "grad_norm": 0.373046875, + "learning_rate": 0.0017405702944267598, + "loss": 0.0972, + "step": 24552 + }, + { + "epoch": 0.21313183045286066, + "grad_norm": 0.12158203125, + "learning_rate": 0.0017405493877909268, + "loss": 0.1338, + "step": 24553 + }, + { + "epoch": 0.21314051093306483, + "grad_norm": 0.07568359375, + "learning_rate": 0.0017405284804545987, + "loss": 0.0879, + "step": 24554 + }, + { + "epoch": 0.213149191413269, + "grad_norm": 0.40234375, + "learning_rate": 0.0017405075724177984, + "loss": 0.1064, + "step": 24555 + }, + { + "epoch": 0.21315787189347316, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017404866636805483, + "loss": 0.0781, + "step": 24556 + }, + { + "epoch": 0.21316655237367732, + "grad_norm": 0.7265625, + "learning_rate": 0.0017404657542428718, + "loss": 0.1348, + "step": 24557 + }, + { + "epoch": 0.21317523285388149, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017404448441047913, + "loss": 0.1348, + "step": 24558 + }, + { + "epoch": 0.21318391333408565, + "grad_norm": 0.244140625, + "learning_rate": 0.0017404239332663304, + "loss": 0.0913, + "step": 24559 + }, + { + "epoch": 0.21319259381428982, + "grad_norm": 0.83203125, + "learning_rate": 0.001740403021727511, + "loss": 0.124, + "step": 24560 + }, + { + "epoch": 0.21320127429449398, + "grad_norm": 0.072265625, + "learning_rate": 0.001740382109488357, + "loss": 0.1035, + "step": 24561 + }, + { + "epoch": 0.21320995477469815, + "grad_norm": 0.76953125, + "learning_rate": 0.0017403611965488904, + "loss": 0.1289, + "step": 24562 + }, + { + "epoch": 0.2132186352549023, + "grad_norm": 0.466796875, + "learning_rate": 0.0017403402829091346, + "loss": 0.1045, + "step": 24563 + }, + { + "epoch": 0.21322731573510648, + "grad_norm": 0.51953125, + "learning_rate": 0.001740319368569112, + "loss": 0.1387, + "step": 24564 + }, + { + "epoch": 0.21323599621531064, + "grad_norm": 0.3203125, + "learning_rate": 0.001740298453528846, + "loss": 0.1641, + "step": 24565 + }, + { + "epoch": 0.2132446766955148, + "grad_norm": 0.31640625, + "learning_rate": 0.0017402775377883593, + "loss": 0.1128, + "step": 24566 + }, + { + "epoch": 0.21325335717571897, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017402566213476742, + "loss": 0.1172, + "step": 24567 + }, + { + "epoch": 0.21326203765592314, + "grad_norm": 0.310546875, + "learning_rate": 0.0017402357042068147, + "loss": 0.1328, + "step": 24568 + }, + { + "epoch": 0.2132707181361273, + "grad_norm": 0.09228515625, + "learning_rate": 0.001740214786365803, + "loss": 0.125, + "step": 24569 + }, + { + "epoch": 0.21327939861633147, + "grad_norm": 1.046875, + "learning_rate": 0.0017401938678246619, + "loss": 0.1738, + "step": 24570 + }, + { + "epoch": 0.21328807909653563, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017401729485834144, + "loss": 0.1025, + "step": 24571 + }, + { + "epoch": 0.2132967595767398, + "grad_norm": 1.859375, + "learning_rate": 0.001740152028642083, + "loss": 0.1445, + "step": 24572 + }, + { + "epoch": 0.21330544005694396, + "grad_norm": 0.5078125, + "learning_rate": 0.0017401311080006913, + "loss": 0.1387, + "step": 24573 + }, + { + "epoch": 0.21331412053714813, + "grad_norm": 0.67578125, + "learning_rate": 0.0017401101866592618, + "loss": 0.0864, + "step": 24574 + }, + { + "epoch": 0.2133228010173523, + "grad_norm": 0.59375, + "learning_rate": 0.0017400892646178177, + "loss": 0.1191, + "step": 24575 + }, + { + "epoch": 0.21333148149755646, + "grad_norm": 0.279296875, + "learning_rate": 0.0017400683418763812, + "loss": 0.1348, + "step": 24576 + }, + { + "epoch": 0.21334016197776062, + "grad_norm": 0.291015625, + "learning_rate": 0.0017400474184349757, + "loss": 0.1396, + "step": 24577 + }, + { + "epoch": 0.2133488424579648, + "grad_norm": 0.08447265625, + "learning_rate": 0.0017400264942936238, + "loss": 0.1074, + "step": 24578 + }, + { + "epoch": 0.21335752293816895, + "grad_norm": 0.240234375, + "learning_rate": 0.0017400055694523492, + "loss": 0.1279, + "step": 24579 + }, + { + "epoch": 0.21336620341837312, + "grad_norm": 0.396484375, + "learning_rate": 0.0017399846439111735, + "loss": 0.1206, + "step": 24580 + }, + { + "epoch": 0.21337488389857726, + "grad_norm": 0.400390625, + "learning_rate": 0.0017399637176701201, + "loss": 0.1016, + "step": 24581 + }, + { + "epoch": 0.21338356437878142, + "grad_norm": 0.30078125, + "learning_rate": 0.0017399427907292123, + "loss": 0.0859, + "step": 24582 + }, + { + "epoch": 0.21339224485898559, + "grad_norm": 0.10986328125, + "learning_rate": 0.0017399218630884728, + "loss": 0.0786, + "step": 24583 + }, + { + "epoch": 0.21340092533918975, + "grad_norm": 0.2265625, + "learning_rate": 0.0017399009347479242, + "loss": 0.1299, + "step": 24584 + }, + { + "epoch": 0.21340960581939392, + "grad_norm": 0.34375, + "learning_rate": 0.0017398800057075896, + "loss": 0.0991, + "step": 24585 + }, + { + "epoch": 0.21341828629959808, + "grad_norm": 0.392578125, + "learning_rate": 0.0017398590759674917, + "loss": 0.126, + "step": 24586 + }, + { + "epoch": 0.21342696677980225, + "grad_norm": 0.609375, + "learning_rate": 0.0017398381455276536, + "loss": 0.1113, + "step": 24587 + }, + { + "epoch": 0.2134356472600064, + "grad_norm": 0.353515625, + "learning_rate": 0.001739817214388098, + "loss": 0.1523, + "step": 24588 + }, + { + "epoch": 0.21344432774021058, + "grad_norm": 0.07763671875, + "learning_rate": 0.0017397962825488483, + "loss": 0.0898, + "step": 24589 + }, + { + "epoch": 0.21345300822041474, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017397753500099267, + "loss": 0.1143, + "step": 24590 + }, + { + "epoch": 0.2134616887006189, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017397544167713564, + "loss": 0.0771, + "step": 24591 + }, + { + "epoch": 0.21347036918082307, + "grad_norm": 0.298828125, + "learning_rate": 0.0017397334828331604, + "loss": 0.0991, + "step": 24592 + }, + { + "epoch": 0.21347904966102724, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017397125481953613, + "loss": 0.1406, + "step": 24593 + }, + { + "epoch": 0.2134877301412314, + "grad_norm": 0.0703125, + "learning_rate": 0.0017396916128579825, + "loss": 0.0957, + "step": 24594 + }, + { + "epoch": 0.21349641062143557, + "grad_norm": 0.0986328125, + "learning_rate": 0.0017396706768210463, + "loss": 0.127, + "step": 24595 + }, + { + "epoch": 0.21350509110163973, + "grad_norm": 0.578125, + "learning_rate": 0.0017396497400845757, + "loss": 0.1289, + "step": 24596 + }, + { + "epoch": 0.2135137715818439, + "grad_norm": 0.328125, + "learning_rate": 0.0017396288026485939, + "loss": 0.0732, + "step": 24597 + }, + { + "epoch": 0.21352245206204806, + "grad_norm": 0.65625, + "learning_rate": 0.0017396078645131239, + "loss": 0.1553, + "step": 24598 + }, + { + "epoch": 0.21353113254225223, + "grad_norm": 0.11376953125, + "learning_rate": 0.001739586925678188, + "loss": 0.1221, + "step": 24599 + }, + { + "epoch": 0.2135398130224564, + "grad_norm": 0.58984375, + "learning_rate": 0.0017395659861438096, + "loss": 0.1084, + "step": 24600 + }, + { + "epoch": 0.21354849350266056, + "grad_norm": 0.42578125, + "learning_rate": 0.0017395450459100114, + "loss": 0.1357, + "step": 24601 + }, + { + "epoch": 0.21355717398286472, + "grad_norm": 0.42578125, + "learning_rate": 0.0017395241049768162, + "loss": 0.083, + "step": 24602 + }, + { + "epoch": 0.2135658544630689, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017395031633442475, + "loss": 0.1143, + "step": 24603 + }, + { + "epoch": 0.21357453494327305, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017394822210123272, + "loss": 0.1406, + "step": 24604 + }, + { + "epoch": 0.21358321542347722, + "grad_norm": 0.08251953125, + "learning_rate": 0.0017394612779810792, + "loss": 0.1128, + "step": 24605 + }, + { + "epoch": 0.21359189590368138, + "grad_norm": 0.400390625, + "learning_rate": 0.0017394403342505258, + "loss": 0.1104, + "step": 24606 + }, + { + "epoch": 0.21360057638388555, + "grad_norm": 0.259765625, + "learning_rate": 0.0017394193898206901, + "loss": 0.0889, + "step": 24607 + }, + { + "epoch": 0.2136092568640897, + "grad_norm": 0.169921875, + "learning_rate": 0.0017393984446915948, + "loss": 0.1299, + "step": 24608 + }, + { + "epoch": 0.21361793734429388, + "grad_norm": 0.119140625, + "learning_rate": 0.001739377498863263, + "loss": 0.1152, + "step": 24609 + }, + { + "epoch": 0.21362661782449804, + "grad_norm": 0.2216796875, + "learning_rate": 0.001739356552335718, + "loss": 0.0811, + "step": 24610 + }, + { + "epoch": 0.2136352983047022, + "grad_norm": 0.30078125, + "learning_rate": 0.0017393356051089816, + "loss": 0.1245, + "step": 24611 + }, + { + "epoch": 0.21364397878490637, + "grad_norm": 0.36328125, + "learning_rate": 0.0017393146571830778, + "loss": 0.0781, + "step": 24612 + }, + { + "epoch": 0.21365265926511054, + "grad_norm": 0.1953125, + "learning_rate": 0.001739293708558029, + "loss": 0.0991, + "step": 24613 + }, + { + "epoch": 0.2136613397453147, + "grad_norm": 0.26953125, + "learning_rate": 0.0017392727592338583, + "loss": 0.1201, + "step": 24614 + }, + { + "epoch": 0.21367002022551887, + "grad_norm": 0.416015625, + "learning_rate": 0.0017392518092105885, + "loss": 0.1396, + "step": 24615 + }, + { + "epoch": 0.21367870070572303, + "grad_norm": 0.3125, + "learning_rate": 0.0017392308584882423, + "loss": 0.1074, + "step": 24616 + }, + { + "epoch": 0.2136873811859272, + "grad_norm": 0.33984375, + "learning_rate": 0.001739209907066843, + "loss": 0.1133, + "step": 24617 + }, + { + "epoch": 0.21369606166613136, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017391889549464134, + "loss": 0.0903, + "step": 24618 + }, + { + "epoch": 0.21370474214633553, + "grad_norm": 0.115234375, + "learning_rate": 0.0017391680021269763, + "loss": 0.0806, + "step": 24619 + }, + { + "epoch": 0.2137134226265397, + "grad_norm": 0.56640625, + "learning_rate": 0.0017391470486085548, + "loss": 0.1094, + "step": 24620 + }, + { + "epoch": 0.21372210310674386, + "grad_norm": 0.421875, + "learning_rate": 0.0017391260943911714, + "loss": 0.1113, + "step": 24621 + }, + { + "epoch": 0.21373078358694803, + "grad_norm": 0.1396484375, + "learning_rate": 0.0017391051394748495, + "loss": 0.1108, + "step": 24622 + }, + { + "epoch": 0.2137394640671522, + "grad_norm": 0.7109375, + "learning_rate": 0.001739084183859612, + "loss": 0.2539, + "step": 24623 + }, + { + "epoch": 0.21374814454735636, + "grad_norm": 0.263671875, + "learning_rate": 0.0017390632275454813, + "loss": 0.0977, + "step": 24624 + }, + { + "epoch": 0.21375682502756052, + "grad_norm": 0.271484375, + "learning_rate": 0.0017390422705324806, + "loss": 0.1045, + "step": 24625 + }, + { + "epoch": 0.21376550550776469, + "grad_norm": 0.0947265625, + "learning_rate": 0.0017390213128206334, + "loss": 0.0854, + "step": 24626 + }, + { + "epoch": 0.21377418598796885, + "grad_norm": 0.251953125, + "learning_rate": 0.0017390003544099618, + "loss": 0.1523, + "step": 24627 + }, + { + "epoch": 0.21378286646817302, + "grad_norm": 0.47265625, + "learning_rate": 0.001738979395300489, + "loss": 0.0806, + "step": 24628 + }, + { + "epoch": 0.21379154694837718, + "grad_norm": 0.08544921875, + "learning_rate": 0.0017389584354922378, + "loss": 0.0898, + "step": 24629 + }, + { + "epoch": 0.21380022742858135, + "grad_norm": 0.294921875, + "learning_rate": 0.0017389374749852313, + "loss": 0.0977, + "step": 24630 + }, + { + "epoch": 0.2138089079087855, + "grad_norm": 0.36328125, + "learning_rate": 0.0017389165137794925, + "loss": 0.1318, + "step": 24631 + }, + { + "epoch": 0.21381758838898968, + "grad_norm": 0.255859375, + "learning_rate": 0.0017388955518750442, + "loss": 0.1206, + "step": 24632 + }, + { + "epoch": 0.21382626886919384, + "grad_norm": 0.275390625, + "learning_rate": 0.0017388745892719093, + "loss": 0.0933, + "step": 24633 + }, + { + "epoch": 0.213834949349398, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017388536259701107, + "loss": 0.1348, + "step": 24634 + }, + { + "epoch": 0.21384362982960217, + "grad_norm": 0.419921875, + "learning_rate": 0.0017388326619696712, + "loss": 0.0879, + "step": 24635 + }, + { + "epoch": 0.21385231030980634, + "grad_norm": 0.1748046875, + "learning_rate": 0.0017388116972706143, + "loss": 0.1079, + "step": 24636 + }, + { + "epoch": 0.2138609907900105, + "grad_norm": 0.0859375, + "learning_rate": 0.0017387907318729622, + "loss": 0.0708, + "step": 24637 + }, + { + "epoch": 0.21386967127021467, + "grad_norm": 0.76171875, + "learning_rate": 0.0017387697657767383, + "loss": 0.1045, + "step": 24638 + }, + { + "epoch": 0.21387835175041883, + "grad_norm": 0.255859375, + "learning_rate": 0.0017387487989819653, + "loss": 0.0859, + "step": 24639 + }, + { + "epoch": 0.213887032230623, + "grad_norm": 0.51171875, + "learning_rate": 0.0017387278314886661, + "loss": 0.0957, + "step": 24640 + }, + { + "epoch": 0.21389571271082716, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017387068632968641, + "loss": 0.1035, + "step": 24641 + }, + { + "epoch": 0.21390439319103133, + "grad_norm": 0.1796875, + "learning_rate": 0.0017386858944065816, + "loss": 0.1016, + "step": 24642 + }, + { + "epoch": 0.2139130736712355, + "grad_norm": 1.0, + "learning_rate": 0.0017386649248178419, + "loss": 0.1318, + "step": 24643 + }, + { + "epoch": 0.21392175415143966, + "grad_norm": 0.64453125, + "learning_rate": 0.0017386439545306678, + "loss": 0.1099, + "step": 24644 + }, + { + "epoch": 0.21393043463164382, + "grad_norm": 0.390625, + "learning_rate": 0.0017386229835450822, + "loss": 0.124, + "step": 24645 + }, + { + "epoch": 0.213939115111848, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017386020118611083, + "loss": 0.124, + "step": 24646 + }, + { + "epoch": 0.21394779559205215, + "grad_norm": 0.08642578125, + "learning_rate": 0.0017385810394787687, + "loss": 0.1133, + "step": 24647 + }, + { + "epoch": 0.21395647607225632, + "grad_norm": 0.46484375, + "learning_rate": 0.0017385600663980866, + "loss": 0.0898, + "step": 24648 + }, + { + "epoch": 0.21396515655246048, + "grad_norm": 0.578125, + "learning_rate": 0.0017385390926190846, + "loss": 0.1187, + "step": 24649 + }, + { + "epoch": 0.21397383703266465, + "grad_norm": 0.1787109375, + "learning_rate": 0.001738518118141786, + "loss": 0.1094, + "step": 24650 + }, + { + "epoch": 0.21398251751286881, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017384971429662136, + "loss": 0.0859, + "step": 24651 + }, + { + "epoch": 0.21399119799307298, + "grad_norm": 0.34375, + "learning_rate": 0.0017384761670923901, + "loss": 0.0918, + "step": 24652 + }, + { + "epoch": 0.21399987847327714, + "grad_norm": 0.45703125, + "learning_rate": 0.0017384551905203387, + "loss": 0.0913, + "step": 24653 + }, + { + "epoch": 0.2140085589534813, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017384342132500824, + "loss": 0.1191, + "step": 24654 + }, + { + "epoch": 0.21401723943368547, + "grad_norm": 0.48046875, + "learning_rate": 0.001738413235281644, + "loss": 0.1182, + "step": 24655 + }, + { + "epoch": 0.21402591991388964, + "grad_norm": 0.09912109375, + "learning_rate": 0.0017383922566150465, + "loss": 0.0977, + "step": 24656 + }, + { + "epoch": 0.2140346003940938, + "grad_norm": 0.07763671875, + "learning_rate": 0.001738371277250313, + "loss": 0.0889, + "step": 24657 + }, + { + "epoch": 0.21404328087429797, + "grad_norm": 0.111328125, + "learning_rate": 0.001738350297187466, + "loss": 0.0933, + "step": 24658 + }, + { + "epoch": 0.21405196135450213, + "grad_norm": 0.48046875, + "learning_rate": 0.001738329316426529, + "loss": 0.1836, + "step": 24659 + }, + { + "epoch": 0.2140606418347063, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017383083349675242, + "loss": 0.0781, + "step": 24660 + }, + { + "epoch": 0.21406932231491047, + "grad_norm": 0.9609375, + "learning_rate": 0.0017382873528104755, + "loss": 0.1211, + "step": 24661 + }, + { + "epoch": 0.21407800279511463, + "grad_norm": 0.287109375, + "learning_rate": 0.0017382663699554047, + "loss": 0.085, + "step": 24662 + }, + { + "epoch": 0.2140866832753188, + "grad_norm": 0.439453125, + "learning_rate": 0.0017382453864023357, + "loss": 0.1279, + "step": 24663 + }, + { + "epoch": 0.21409536375552296, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017382244021512912, + "loss": 0.1348, + "step": 24664 + }, + { + "epoch": 0.21410404423572713, + "grad_norm": 0.4140625, + "learning_rate": 0.0017382034172022942, + "loss": 0.0957, + "step": 24665 + }, + { + "epoch": 0.2141127247159313, + "grad_norm": 1.2890625, + "learning_rate": 0.0017381824315553674, + "loss": 0.1021, + "step": 24666 + }, + { + "epoch": 0.21412140519613546, + "grad_norm": 0.439453125, + "learning_rate": 0.0017381614452105336, + "loss": 0.1025, + "step": 24667 + }, + { + "epoch": 0.21413008567633962, + "grad_norm": 0.6015625, + "learning_rate": 0.0017381404581678164, + "loss": 0.0942, + "step": 24668 + }, + { + "epoch": 0.21413876615654379, + "grad_norm": 0.138671875, + "learning_rate": 0.0017381194704272386, + "loss": 0.0874, + "step": 24669 + }, + { + "epoch": 0.21414744663674795, + "grad_norm": 0.162109375, + "learning_rate": 0.0017380984819888225, + "loss": 0.1299, + "step": 24670 + }, + { + "epoch": 0.21415612711695212, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017380774928525918, + "loss": 0.0908, + "step": 24671 + }, + { + "epoch": 0.21416480759715628, + "grad_norm": 0.515625, + "learning_rate": 0.001738056503018569, + "loss": 0.1406, + "step": 24672 + }, + { + "epoch": 0.21417348807736045, + "grad_norm": 0.0888671875, + "learning_rate": 0.001738035512486777, + "loss": 0.0986, + "step": 24673 + }, + { + "epoch": 0.2141821685575646, + "grad_norm": 0.578125, + "learning_rate": 0.0017380145212572394, + "loss": 0.1299, + "step": 24674 + }, + { + "epoch": 0.21419084903776878, + "grad_norm": 0.5, + "learning_rate": 0.001737993529329978, + "loss": 0.1089, + "step": 24675 + }, + { + "epoch": 0.21419952951797294, + "grad_norm": 0.369140625, + "learning_rate": 0.001737972536705017, + "loss": 0.0786, + "step": 24676 + }, + { + "epoch": 0.2142082099981771, + "grad_norm": 0.208984375, + "learning_rate": 0.0017379515433823788, + "loss": 0.0879, + "step": 24677 + }, + { + "epoch": 0.21421689047838127, + "grad_norm": 0.0908203125, + "learning_rate": 0.0017379305493620864, + "loss": 0.0996, + "step": 24678 + }, + { + "epoch": 0.21422557095858544, + "grad_norm": 0.23828125, + "learning_rate": 0.0017379095546441624, + "loss": 0.1104, + "step": 24679 + }, + { + "epoch": 0.2142342514387896, + "grad_norm": 0.07568359375, + "learning_rate": 0.0017378885592286304, + "loss": 0.0986, + "step": 24680 + }, + { + "epoch": 0.21424293191899377, + "grad_norm": 0.298828125, + "learning_rate": 0.0017378675631155127, + "loss": 0.1416, + "step": 24681 + }, + { + "epoch": 0.21425161239919793, + "grad_norm": 0.33203125, + "learning_rate": 0.001737846566304833, + "loss": 0.0889, + "step": 24682 + }, + { + "epoch": 0.2142602928794021, + "grad_norm": 0.1591796875, + "learning_rate": 0.001737825568796614, + "loss": 0.1631, + "step": 24683 + }, + { + "epoch": 0.21426897335960626, + "grad_norm": 0.162109375, + "learning_rate": 0.0017378045705908783, + "loss": 0.1436, + "step": 24684 + }, + { + "epoch": 0.21427765383981043, + "grad_norm": 0.98046875, + "learning_rate": 0.0017377835716876492, + "loss": 0.1201, + "step": 24685 + }, + { + "epoch": 0.2142863343200146, + "grad_norm": 0.56640625, + "learning_rate": 0.0017377625720869491, + "loss": 0.0991, + "step": 24686 + }, + { + "epoch": 0.21429501480021876, + "grad_norm": 0.197265625, + "learning_rate": 0.0017377415717888022, + "loss": 0.1934, + "step": 24687 + }, + { + "epoch": 0.21430369528042292, + "grad_norm": 0.318359375, + "learning_rate": 0.00173772057079323, + "loss": 0.1416, + "step": 24688 + }, + { + "epoch": 0.2143123757606271, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017376995691002566, + "loss": 0.0811, + "step": 24689 + }, + { + "epoch": 0.21432105624083125, + "grad_norm": 0.1064453125, + "learning_rate": 0.0017376785667099044, + "loss": 0.0713, + "step": 24690 + }, + { + "epoch": 0.21432973672103542, + "grad_norm": 0.435546875, + "learning_rate": 0.0017376575636221965, + "loss": 0.1123, + "step": 24691 + }, + { + "epoch": 0.21433841720123958, + "grad_norm": 0.3984375, + "learning_rate": 0.0017376365598371557, + "loss": 0.1187, + "step": 24692 + }, + { + "epoch": 0.21434709768144375, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017376155553548054, + "loss": 0.0664, + "step": 24693 + }, + { + "epoch": 0.21435577816164791, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017375945501751683, + "loss": 0.1104, + "step": 24694 + }, + { + "epoch": 0.21436445864185208, + "grad_norm": 0.287109375, + "learning_rate": 0.0017375735442982671, + "loss": 0.1289, + "step": 24695 + }, + { + "epoch": 0.21437313912205624, + "grad_norm": 0.330078125, + "learning_rate": 0.0017375525377241253, + "loss": 0.1182, + "step": 24696 + }, + { + "epoch": 0.2143818196022604, + "grad_norm": 0.0859375, + "learning_rate": 0.0017375315304527658, + "loss": 0.0889, + "step": 24697 + }, + { + "epoch": 0.21439050008246457, + "grad_norm": 1.0078125, + "learning_rate": 0.0017375105224842111, + "loss": 0.1216, + "step": 24698 + }, + { + "epoch": 0.21439918056266874, + "grad_norm": 1.03125, + "learning_rate": 0.0017374895138184848, + "loss": 0.1143, + "step": 24699 + }, + { + "epoch": 0.2144078610428729, + "grad_norm": 0.0830078125, + "learning_rate": 0.001737468504455609, + "loss": 0.0796, + "step": 24700 + }, + { + "epoch": 0.21441654152307707, + "grad_norm": 0.5703125, + "learning_rate": 0.0017374474943956076, + "loss": 0.1191, + "step": 24701 + }, + { + "epoch": 0.21442522200328124, + "grad_norm": 0.357421875, + "learning_rate": 0.001737426483638503, + "loss": 0.062, + "step": 24702 + }, + { + "epoch": 0.2144339024834854, + "grad_norm": 0.5, + "learning_rate": 0.0017374054721843184, + "loss": 0.1055, + "step": 24703 + }, + { + "epoch": 0.21444258296368954, + "grad_norm": 0.228515625, + "learning_rate": 0.001737384460033077, + "loss": 0.0952, + "step": 24704 + }, + { + "epoch": 0.2144512634438937, + "grad_norm": 0.341796875, + "learning_rate": 0.0017373634471848011, + "loss": 0.126, + "step": 24705 + }, + { + "epoch": 0.21445994392409787, + "grad_norm": 0.302734375, + "learning_rate": 0.0017373424336395147, + "loss": 0.0786, + "step": 24706 + }, + { + "epoch": 0.21446862440430203, + "grad_norm": 0.7109375, + "learning_rate": 0.0017373214193972398, + "loss": 0.0986, + "step": 24707 + }, + { + "epoch": 0.2144773048845062, + "grad_norm": 0.875, + "learning_rate": 0.0017373004044579999, + "loss": 0.0884, + "step": 24708 + }, + { + "epoch": 0.21448598536471036, + "grad_norm": 0.08740234375, + "learning_rate": 0.0017372793888218178, + "loss": 0.0723, + "step": 24709 + }, + { + "epoch": 0.21449466584491453, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017372583724887165, + "loss": 0.0933, + "step": 24710 + }, + { + "epoch": 0.2145033463251187, + "grad_norm": 0.1611328125, + "learning_rate": 0.0017372373554587189, + "loss": 0.1074, + "step": 24711 + }, + { + "epoch": 0.21451202680532286, + "grad_norm": 0.150390625, + "learning_rate": 0.0017372163377318483, + "loss": 0.1406, + "step": 24712 + }, + { + "epoch": 0.21452070728552702, + "grad_norm": 0.29296875, + "learning_rate": 0.0017371953193081277, + "loss": 0.0952, + "step": 24713 + }, + { + "epoch": 0.2145293877657312, + "grad_norm": 0.234375, + "learning_rate": 0.0017371743001875794, + "loss": 0.0923, + "step": 24714 + }, + { + "epoch": 0.21453806824593535, + "grad_norm": 0.06884765625, + "learning_rate": 0.001737153280370227, + "loss": 0.0908, + "step": 24715 + }, + { + "epoch": 0.21454674872613952, + "grad_norm": 0.318359375, + "learning_rate": 0.0017371322598560936, + "loss": 0.0908, + "step": 24716 + }, + { + "epoch": 0.21455542920634368, + "grad_norm": 0.171875, + "learning_rate": 0.0017371112386452016, + "loss": 0.1123, + "step": 24717 + }, + { + "epoch": 0.21456410968654785, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017370902167375747, + "loss": 0.0801, + "step": 24718 + }, + { + "epoch": 0.21457279016675201, + "grad_norm": 0.1484375, + "learning_rate": 0.0017370691941332353, + "loss": 0.0967, + "step": 24719 + }, + { + "epoch": 0.21458147064695618, + "grad_norm": 0.314453125, + "learning_rate": 0.0017370481708322064, + "loss": 0.1035, + "step": 24720 + }, + { + "epoch": 0.21459015112716034, + "grad_norm": 0.369140625, + "learning_rate": 0.0017370271468345113, + "loss": 0.127, + "step": 24721 + }, + { + "epoch": 0.2145988316073645, + "grad_norm": 0.53125, + "learning_rate": 0.001737006122140173, + "loss": 0.1445, + "step": 24722 + }, + { + "epoch": 0.21460751208756867, + "grad_norm": 0.26953125, + "learning_rate": 0.0017369850967492145, + "loss": 0.0923, + "step": 24723 + }, + { + "epoch": 0.21461619256777284, + "grad_norm": 0.1826171875, + "learning_rate": 0.0017369640706616583, + "loss": 0.1406, + "step": 24724 + }, + { + "epoch": 0.214624873047977, + "grad_norm": 0.173828125, + "learning_rate": 0.0017369430438775283, + "loss": 0.0947, + "step": 24725 + }, + { + "epoch": 0.21463355352818117, + "grad_norm": 0.40625, + "learning_rate": 0.0017369220163968462, + "loss": 0.0957, + "step": 24726 + }, + { + "epoch": 0.21464223400838534, + "grad_norm": 0.275390625, + "learning_rate": 0.0017369009882196363, + "loss": 0.0898, + "step": 24727 + }, + { + "epoch": 0.2146509144885895, + "grad_norm": 0.47265625, + "learning_rate": 0.001736879959345921, + "loss": 0.1055, + "step": 24728 + }, + { + "epoch": 0.21465959496879367, + "grad_norm": 0.11083984375, + "learning_rate": 0.0017368589297757232, + "loss": 0.1104, + "step": 24729 + }, + { + "epoch": 0.21466827544899783, + "grad_norm": 0.078125, + "learning_rate": 0.0017368378995090664, + "loss": 0.0688, + "step": 24730 + }, + { + "epoch": 0.214676955929202, + "grad_norm": 0.4140625, + "learning_rate": 0.001736816868545973, + "loss": 0.1084, + "step": 24731 + }, + { + "epoch": 0.21468563640940616, + "grad_norm": 0.1318359375, + "learning_rate": 0.0017367958368864661, + "loss": 0.1162, + "step": 24732 + }, + { + "epoch": 0.21469431688961033, + "grad_norm": 0.2109375, + "learning_rate": 0.0017367748045305686, + "loss": 0.1196, + "step": 24733 + }, + { + "epoch": 0.2147029973698145, + "grad_norm": 0.22265625, + "learning_rate": 0.0017367537714783044, + "loss": 0.082, + "step": 24734 + }, + { + "epoch": 0.21471167785001866, + "grad_norm": 0.11767578125, + "learning_rate": 0.0017367327377296952, + "loss": 0.1113, + "step": 24735 + }, + { + "epoch": 0.21472035833022282, + "grad_norm": 0.23046875, + "learning_rate": 0.001736711703284765, + "loss": 0.1465, + "step": 24736 + }, + { + "epoch": 0.214729038810427, + "grad_norm": 0.1640625, + "learning_rate": 0.0017366906681435365, + "loss": 0.1445, + "step": 24737 + }, + { + "epoch": 0.21473771929063115, + "grad_norm": 0.6953125, + "learning_rate": 0.0017366696323060328, + "loss": 0.085, + "step": 24738 + }, + { + "epoch": 0.21474639977083532, + "grad_norm": 0.18359375, + "learning_rate": 0.0017366485957722763, + "loss": 0.1543, + "step": 24739 + }, + { + "epoch": 0.21475508025103948, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017366275585422907, + "loss": 0.1206, + "step": 24740 + }, + { + "epoch": 0.21476376073124365, + "grad_norm": 0.3125, + "learning_rate": 0.0017366065206160988, + "loss": 0.0928, + "step": 24741 + }, + { + "epoch": 0.2147724412114478, + "grad_norm": 0.34765625, + "learning_rate": 0.0017365854819937235, + "loss": 0.1001, + "step": 24742 + }, + { + "epoch": 0.21478112169165198, + "grad_norm": 0.13671875, + "learning_rate": 0.0017365644426751879, + "loss": 0.0801, + "step": 24743 + }, + { + "epoch": 0.21478980217185614, + "grad_norm": 0.359375, + "learning_rate": 0.0017365434026605149, + "loss": 0.127, + "step": 24744 + }, + { + "epoch": 0.2147984826520603, + "grad_norm": 0.486328125, + "learning_rate": 0.0017365223619497277, + "loss": 0.1299, + "step": 24745 + }, + { + "epoch": 0.21480716313226447, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017365013205428493, + "loss": 0.1025, + "step": 24746 + }, + { + "epoch": 0.21481584361246864, + "grad_norm": 0.396484375, + "learning_rate": 0.0017364802784399025, + "loss": 0.1191, + "step": 24747 + }, + { + "epoch": 0.2148245240926728, + "grad_norm": 0.1474609375, + "learning_rate": 0.0017364592356409105, + "loss": 0.1289, + "step": 24748 + }, + { + "epoch": 0.21483320457287697, + "grad_norm": 0.2734375, + "learning_rate": 0.001736438192145896, + "loss": 0.0938, + "step": 24749 + }, + { + "epoch": 0.21484188505308113, + "grad_norm": 0.45703125, + "learning_rate": 0.0017364171479548828, + "loss": 0.1074, + "step": 24750 + }, + { + "epoch": 0.2148505655332853, + "grad_norm": 0.79296875, + "learning_rate": 0.0017363961030678928, + "loss": 0.1455, + "step": 24751 + }, + { + "epoch": 0.21485924601348946, + "grad_norm": 0.078125, + "learning_rate": 0.00173637505748495, + "loss": 0.082, + "step": 24752 + }, + { + "epoch": 0.21486792649369363, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017363540112060767, + "loss": 0.1191, + "step": 24753 + }, + { + "epoch": 0.2148766069738978, + "grad_norm": 0.140625, + "learning_rate": 0.001736332964231296, + "loss": 0.1279, + "step": 24754 + }, + { + "epoch": 0.21488528745410196, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017363119165606318, + "loss": 0.1064, + "step": 24755 + }, + { + "epoch": 0.21489396793430612, + "grad_norm": 0.302734375, + "learning_rate": 0.0017362908681941062, + "loss": 0.0967, + "step": 24756 + }, + { + "epoch": 0.2149026484145103, + "grad_norm": 0.52734375, + "learning_rate": 0.0017362698191317422, + "loss": 0.0703, + "step": 24757 + }, + { + "epoch": 0.21491132889471445, + "grad_norm": 0.60546875, + "learning_rate": 0.001736248769373563, + "loss": 0.1436, + "step": 24758 + }, + { + "epoch": 0.21492000937491862, + "grad_norm": 0.146484375, + "learning_rate": 0.0017362277189195922, + "loss": 0.0933, + "step": 24759 + }, + { + "epoch": 0.21492868985512278, + "grad_norm": 0.2890625, + "learning_rate": 0.001736206667769852, + "loss": 0.1104, + "step": 24760 + }, + { + "epoch": 0.21493737033532695, + "grad_norm": 0.251953125, + "learning_rate": 0.0017361856159243659, + "loss": 0.0889, + "step": 24761 + }, + { + "epoch": 0.21494605081553111, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017361645633831566, + "loss": 0.0918, + "step": 24762 + }, + { + "epoch": 0.21495473129573528, + "grad_norm": 0.49609375, + "learning_rate": 0.0017361435101462472, + "loss": 0.1484, + "step": 24763 + }, + { + "epoch": 0.21496341177593944, + "grad_norm": 0.28515625, + "learning_rate": 0.0017361224562136613, + "loss": 0.105, + "step": 24764 + }, + { + "epoch": 0.2149720922561436, + "grad_norm": 0.57421875, + "learning_rate": 0.001736101401585421, + "loss": 0.1182, + "step": 24765 + }, + { + "epoch": 0.21498077273634777, + "grad_norm": 0.2236328125, + "learning_rate": 0.00173608034626155, + "loss": 0.1064, + "step": 24766 + }, + { + "epoch": 0.21498945321655194, + "grad_norm": 0.32421875, + "learning_rate": 0.0017360592902420711, + "loss": 0.0811, + "step": 24767 + }, + { + "epoch": 0.2149981336967561, + "grad_norm": 0.220703125, + "learning_rate": 0.001736038233527007, + "loss": 0.0957, + "step": 24768 + }, + { + "epoch": 0.21500681417696027, + "grad_norm": 0.17578125, + "learning_rate": 0.0017360171761163815, + "loss": 0.1221, + "step": 24769 + }, + { + "epoch": 0.21501549465716444, + "grad_norm": 0.1748046875, + "learning_rate": 0.001735996118010217, + "loss": 0.0938, + "step": 24770 + }, + { + "epoch": 0.2150241751373686, + "grad_norm": 0.33203125, + "learning_rate": 0.0017359750592085364, + "loss": 0.1348, + "step": 24771 + }, + { + "epoch": 0.21503285561757277, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017359539997113633, + "loss": 0.1025, + "step": 24772 + }, + { + "epoch": 0.21504153609777693, + "grad_norm": 0.10546875, + "learning_rate": 0.0017359329395187201, + "loss": 0.125, + "step": 24773 + }, + { + "epoch": 0.2150502165779811, + "grad_norm": 0.12109375, + "learning_rate": 0.0017359118786306306, + "loss": 0.1328, + "step": 24774 + }, + { + "epoch": 0.21505889705818526, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017358908170471176, + "loss": 0.1025, + "step": 24775 + }, + { + "epoch": 0.21506757753838943, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017358697547682038, + "loss": 0.127, + "step": 24776 + }, + { + "epoch": 0.2150762580185936, + "grad_norm": 1.734375, + "learning_rate": 0.0017358486917939122, + "loss": 0.1641, + "step": 24777 + }, + { + "epoch": 0.21508493849879776, + "grad_norm": 0.1005859375, + "learning_rate": 0.0017358276281242663, + "loss": 0.1191, + "step": 24778 + }, + { + "epoch": 0.21509361897900192, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017358065637592885, + "loss": 0.0903, + "step": 24779 + }, + { + "epoch": 0.2151022994592061, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017357854986990023, + "loss": 0.0908, + "step": 24780 + }, + { + "epoch": 0.21511097993941025, + "grad_norm": 0.111328125, + "learning_rate": 0.0017357644329434308, + "loss": 0.0864, + "step": 24781 + }, + { + "epoch": 0.21511966041961442, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017357433664925966, + "loss": 0.103, + "step": 24782 + }, + { + "epoch": 0.21512834089981858, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017357222993465234, + "loss": 0.1064, + "step": 24783 + }, + { + "epoch": 0.21513702138002275, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017357012315052335, + "loss": 0.1133, + "step": 24784 + }, + { + "epoch": 0.2151457018602269, + "grad_norm": 0.6171875, + "learning_rate": 0.0017356801629687503, + "loss": 0.1016, + "step": 24785 + }, + { + "epoch": 0.21515438234043108, + "grad_norm": 0.8125, + "learning_rate": 0.0017356590937370972, + "loss": 0.124, + "step": 24786 + }, + { + "epoch": 0.21516306282063524, + "grad_norm": 0.51953125, + "learning_rate": 0.0017356380238102965, + "loss": 0.0967, + "step": 24787 + }, + { + "epoch": 0.2151717433008394, + "grad_norm": 0.283203125, + "learning_rate": 0.0017356169531883717, + "loss": 0.1279, + "step": 24788 + }, + { + "epoch": 0.21518042378104357, + "grad_norm": 0.07568359375, + "learning_rate": 0.0017355958818713456, + "loss": 0.0957, + "step": 24789 + }, + { + "epoch": 0.21518910426124774, + "grad_norm": 0.4140625, + "learning_rate": 0.0017355748098592417, + "loss": 0.0752, + "step": 24790 + }, + { + "epoch": 0.2151977847414519, + "grad_norm": 0.08984375, + "learning_rate": 0.0017355537371520824, + "loss": 0.1123, + "step": 24791 + }, + { + "epoch": 0.21520646522165607, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017355326637498913, + "loss": 0.082, + "step": 24792 + }, + { + "epoch": 0.21521514570186023, + "grad_norm": 0.11181640625, + "learning_rate": 0.0017355115896526912, + "loss": 0.0972, + "step": 24793 + }, + { + "epoch": 0.2152238261820644, + "grad_norm": 0.1767578125, + "learning_rate": 0.001735490514860505, + "loss": 0.1426, + "step": 24794 + }, + { + "epoch": 0.21523250666226856, + "grad_norm": 0.57421875, + "learning_rate": 0.0017354694393733564, + "loss": 0.1172, + "step": 24795 + }, + { + "epoch": 0.21524118714247273, + "grad_norm": 0.2421875, + "learning_rate": 0.0017354483631912675, + "loss": 0.1191, + "step": 24796 + }, + { + "epoch": 0.2152498676226769, + "grad_norm": 0.263671875, + "learning_rate": 0.0017354272863142621, + "loss": 0.1074, + "step": 24797 + }, + { + "epoch": 0.21525854810288106, + "grad_norm": 0.255859375, + "learning_rate": 0.0017354062087423625, + "loss": 0.1348, + "step": 24798 + }, + { + "epoch": 0.21526722858308522, + "grad_norm": 0.447265625, + "learning_rate": 0.0017353851304755927, + "loss": 0.1289, + "step": 24799 + }, + { + "epoch": 0.2152759090632894, + "grad_norm": 0.314453125, + "learning_rate": 0.0017353640515139753, + "loss": 0.0923, + "step": 24800 + }, + { + "epoch": 0.21528458954349355, + "grad_norm": 0.58984375, + "learning_rate": 0.001735342971857533, + "loss": 0.0952, + "step": 24801 + }, + { + "epoch": 0.21529327002369772, + "grad_norm": 0.1806640625, + "learning_rate": 0.001735321891506289, + "loss": 0.1167, + "step": 24802 + }, + { + "epoch": 0.21530195050390188, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017353008104602672, + "loss": 0.1445, + "step": 24803 + }, + { + "epoch": 0.21531063098410605, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017352797287194894, + "loss": 0.1309, + "step": 24804 + }, + { + "epoch": 0.21531931146431021, + "grad_norm": 0.26953125, + "learning_rate": 0.0017352586462839798, + "loss": 0.124, + "step": 24805 + }, + { + "epoch": 0.21532799194451438, + "grad_norm": 1.109375, + "learning_rate": 0.0017352375631537605, + "loss": 0.2891, + "step": 24806 + }, + { + "epoch": 0.21533667242471854, + "grad_norm": 0.095703125, + "learning_rate": 0.001735216479328855, + "loss": 0.0898, + "step": 24807 + }, + { + "epoch": 0.2153453529049227, + "grad_norm": 0.267578125, + "learning_rate": 0.0017351953948092863, + "loss": 0.0952, + "step": 24808 + }, + { + "epoch": 0.21535403338512688, + "grad_norm": 0.09716796875, + "learning_rate": 0.0017351743095950774, + "loss": 0.1172, + "step": 24809 + }, + { + "epoch": 0.21536271386533104, + "grad_norm": 0.5, + "learning_rate": 0.0017351532236862515, + "loss": 0.1055, + "step": 24810 + }, + { + "epoch": 0.2153713943455352, + "grad_norm": 0.2265625, + "learning_rate": 0.0017351321370828315, + "loss": 0.1045, + "step": 24811 + }, + { + "epoch": 0.21538007482573937, + "grad_norm": 0.71875, + "learning_rate": 0.0017351110497848406, + "loss": 0.1162, + "step": 24812 + }, + { + "epoch": 0.21538875530594354, + "grad_norm": 0.265625, + "learning_rate": 0.0017350899617923018, + "loss": 0.0928, + "step": 24813 + }, + { + "epoch": 0.2153974357861477, + "grad_norm": 0.296875, + "learning_rate": 0.001735068873105238, + "loss": 0.1187, + "step": 24814 + }, + { + "epoch": 0.21540611626635187, + "grad_norm": 0.349609375, + "learning_rate": 0.0017350477837236727, + "loss": 0.1191, + "step": 24815 + }, + { + "epoch": 0.21541479674655603, + "grad_norm": 0.28125, + "learning_rate": 0.0017350266936476285, + "loss": 0.0947, + "step": 24816 + }, + { + "epoch": 0.2154234772267602, + "grad_norm": 0.201171875, + "learning_rate": 0.0017350056028771288, + "loss": 0.0869, + "step": 24817 + }, + { + "epoch": 0.21543215770696436, + "grad_norm": 0.357421875, + "learning_rate": 0.0017349845114121962, + "loss": 0.1797, + "step": 24818 + }, + { + "epoch": 0.21544083818716853, + "grad_norm": 0.0693359375, + "learning_rate": 0.0017349634192528542, + "loss": 0.1152, + "step": 24819 + }, + { + "epoch": 0.2154495186673727, + "grad_norm": 0.38671875, + "learning_rate": 0.001734942326399126, + "loss": 0.0889, + "step": 24820 + }, + { + "epoch": 0.21545819914757686, + "grad_norm": 0.65234375, + "learning_rate": 0.0017349212328510342, + "loss": 0.165, + "step": 24821 + }, + { + "epoch": 0.21546687962778102, + "grad_norm": 0.09912109375, + "learning_rate": 0.001734900138608602, + "loss": 0.1289, + "step": 24822 + }, + { + "epoch": 0.2154755601079852, + "grad_norm": 0.609375, + "learning_rate": 0.0017348790436718527, + "loss": 0.0918, + "step": 24823 + }, + { + "epoch": 0.21548424058818935, + "grad_norm": 0.11328125, + "learning_rate": 0.0017348579480408095, + "loss": 0.1367, + "step": 24824 + }, + { + "epoch": 0.21549292106839352, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017348368517154948, + "loss": 0.1074, + "step": 24825 + }, + { + "epoch": 0.21550160154859768, + "grad_norm": 0.478515625, + "learning_rate": 0.001734815754695932, + "loss": 0.0981, + "step": 24826 + }, + { + "epoch": 0.21551028202880182, + "grad_norm": 0.515625, + "learning_rate": 0.0017347946569821443, + "loss": 0.1104, + "step": 24827 + }, + { + "epoch": 0.21551896250900598, + "grad_norm": 0.23046875, + "learning_rate": 0.0017347735585741548, + "loss": 0.0786, + "step": 24828 + }, + { + "epoch": 0.21552764298921015, + "grad_norm": 0.640625, + "learning_rate": 0.0017347524594719865, + "loss": 0.0771, + "step": 24829 + }, + { + "epoch": 0.21553632346941431, + "grad_norm": 0.16796875, + "learning_rate": 0.0017347313596756625, + "loss": 0.1699, + "step": 24830 + }, + { + "epoch": 0.21554500394961848, + "grad_norm": 1.3125, + "learning_rate": 0.0017347102591852056, + "loss": 0.1099, + "step": 24831 + }, + { + "epoch": 0.21555368442982265, + "grad_norm": 0.123046875, + "learning_rate": 0.0017346891580006392, + "loss": 0.1094, + "step": 24832 + }, + { + "epoch": 0.2155623649100268, + "grad_norm": 0.341796875, + "learning_rate": 0.0017346680561219864, + "loss": 0.1406, + "step": 24833 + }, + { + "epoch": 0.21557104539023098, + "grad_norm": 0.92578125, + "learning_rate": 0.00173464695354927, + "loss": 0.1011, + "step": 24834 + }, + { + "epoch": 0.21557972587043514, + "grad_norm": 0.2578125, + "learning_rate": 0.0017346258502825135, + "loss": 0.0908, + "step": 24835 + }, + { + "epoch": 0.2155884063506393, + "grad_norm": 0.2431640625, + "learning_rate": 0.001734604746321739, + "loss": 0.1201, + "step": 24836 + }, + { + "epoch": 0.21559708683084347, + "grad_norm": 0.435546875, + "learning_rate": 0.001734583641666971, + "loss": 0.1011, + "step": 24837 + }, + { + "epoch": 0.21560576731104764, + "grad_norm": 0.06591796875, + "learning_rate": 0.0017345625363182316, + "loss": 0.0713, + "step": 24838 + }, + { + "epoch": 0.2156144477912518, + "grad_norm": 0.376953125, + "learning_rate": 0.0017345414302755442, + "loss": 0.168, + "step": 24839 + }, + { + "epoch": 0.21562312827145597, + "grad_norm": 0.10546875, + "learning_rate": 0.0017345203235389318, + "loss": 0.1201, + "step": 24840 + }, + { + "epoch": 0.21563180875166013, + "grad_norm": 0.28515625, + "learning_rate": 0.0017344992161084172, + "loss": 0.1484, + "step": 24841 + }, + { + "epoch": 0.2156404892318643, + "grad_norm": 0.10546875, + "learning_rate": 0.001734478107984024, + "loss": 0.1309, + "step": 24842 + }, + { + "epoch": 0.21564916971206846, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017344569991657754, + "loss": 0.127, + "step": 24843 + }, + { + "epoch": 0.21565785019227263, + "grad_norm": 0.8046875, + "learning_rate": 0.0017344358896536941, + "loss": 0.2402, + "step": 24844 + }, + { + "epoch": 0.2156665306724768, + "grad_norm": 0.287109375, + "learning_rate": 0.0017344147794478028, + "loss": 0.0928, + "step": 24845 + }, + { + "epoch": 0.21567521115268096, + "grad_norm": 0.26953125, + "learning_rate": 0.0017343936685481252, + "loss": 0.1138, + "step": 24846 + }, + { + "epoch": 0.21568389163288512, + "grad_norm": 0.12890625, + "learning_rate": 0.0017343725569546845, + "loss": 0.1006, + "step": 24847 + }, + { + "epoch": 0.2156925721130893, + "grad_norm": 0.07763671875, + "learning_rate": 0.0017343514446675032, + "loss": 0.0811, + "step": 24848 + }, + { + "epoch": 0.21570125259329345, + "grad_norm": 0.66796875, + "learning_rate": 0.001734330331686605, + "loss": 0.1426, + "step": 24849 + }, + { + "epoch": 0.21570993307349762, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017343092180120124, + "loss": 0.0947, + "step": 24850 + }, + { + "epoch": 0.21571861355370178, + "grad_norm": 0.65625, + "learning_rate": 0.0017342881036437488, + "loss": 0.1035, + "step": 24851 + }, + { + "epoch": 0.21572729403390595, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017342669885818374, + "loss": 0.1377, + "step": 24852 + }, + { + "epoch": 0.2157359745141101, + "grad_norm": 0.45703125, + "learning_rate": 0.001734245872826301, + "loss": 0.0957, + "step": 24853 + }, + { + "epoch": 0.21574465499431428, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017342247563771626, + "loss": 0.0996, + "step": 24854 + }, + { + "epoch": 0.21575333547451844, + "grad_norm": 0.09033203125, + "learning_rate": 0.001734203639234446, + "loss": 0.1748, + "step": 24855 + }, + { + "epoch": 0.2157620159547226, + "grad_norm": 0.85546875, + "learning_rate": 0.001734182521398174, + "loss": 0.1069, + "step": 24856 + }, + { + "epoch": 0.21577069643492677, + "grad_norm": 0.369140625, + "learning_rate": 0.001734161402868369, + "loss": 0.1338, + "step": 24857 + }, + { + "epoch": 0.21577937691513094, + "grad_norm": 0.75390625, + "learning_rate": 0.0017341402836450548, + "loss": 0.0952, + "step": 24858 + }, + { + "epoch": 0.2157880573953351, + "grad_norm": 0.380859375, + "learning_rate": 0.0017341191637282542, + "loss": 0.1074, + "step": 24859 + }, + { + "epoch": 0.21579673787553927, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017340980431179906, + "loss": 0.0938, + "step": 24860 + }, + { + "epoch": 0.21580541835574343, + "grad_norm": 0.11669921875, + "learning_rate": 0.001734076921814287, + "loss": 0.0986, + "step": 24861 + }, + { + "epoch": 0.2158140988359476, + "grad_norm": 0.1474609375, + "learning_rate": 0.001734055799817166, + "loss": 0.0938, + "step": 24862 + }, + { + "epoch": 0.21582277931615176, + "grad_norm": 0.3046875, + "learning_rate": 0.0017340346771266514, + "loss": 0.0791, + "step": 24863 + }, + { + "epoch": 0.21583145979635593, + "grad_norm": 0.55078125, + "learning_rate": 0.0017340135537427658, + "loss": 0.1367, + "step": 24864 + }, + { + "epoch": 0.2158401402765601, + "grad_norm": 0.103515625, + "learning_rate": 0.0017339924296655327, + "loss": 0.0767, + "step": 24865 + }, + { + "epoch": 0.21584882075676426, + "grad_norm": 0.150390625, + "learning_rate": 0.0017339713048949749, + "loss": 0.0767, + "step": 24866 + }, + { + "epoch": 0.21585750123696842, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017339501794311157, + "loss": 0.0791, + "step": 24867 + }, + { + "epoch": 0.2158661817171726, + "grad_norm": 0.703125, + "learning_rate": 0.0017339290532739783, + "loss": 0.1367, + "step": 24868 + }, + { + "epoch": 0.21587486219737675, + "grad_norm": 0.361328125, + "learning_rate": 0.0017339079264235852, + "loss": 0.1206, + "step": 24869 + }, + { + "epoch": 0.21588354267758092, + "grad_norm": 0.5390625, + "learning_rate": 0.0017338867988799598, + "loss": 0.083, + "step": 24870 + }, + { + "epoch": 0.21589222315778508, + "grad_norm": 0.8828125, + "learning_rate": 0.0017338656706431257, + "loss": 0.0918, + "step": 24871 + }, + { + "epoch": 0.21590090363798925, + "grad_norm": 0.4921875, + "learning_rate": 0.0017338445417131056, + "loss": 0.1055, + "step": 24872 + }, + { + "epoch": 0.21590958411819342, + "grad_norm": 0.51171875, + "learning_rate": 0.0017338234120899225, + "loss": 0.1016, + "step": 24873 + }, + { + "epoch": 0.21591826459839758, + "grad_norm": 0.326171875, + "learning_rate": 0.0017338022817736, + "loss": 0.1104, + "step": 24874 + }, + { + "epoch": 0.21592694507860175, + "grad_norm": 0.470703125, + "learning_rate": 0.0017337811507641605, + "loss": 0.0952, + "step": 24875 + }, + { + "epoch": 0.2159356255588059, + "grad_norm": 0.130859375, + "learning_rate": 0.0017337600190616277, + "loss": 0.0928, + "step": 24876 + }, + { + "epoch": 0.21594430603901008, + "grad_norm": 0.08447265625, + "learning_rate": 0.001733738886666024, + "loss": 0.1021, + "step": 24877 + }, + { + "epoch": 0.21595298651921424, + "grad_norm": 0.53515625, + "learning_rate": 0.0017337177535773733, + "loss": 0.0889, + "step": 24878 + }, + { + "epoch": 0.2159616669994184, + "grad_norm": 0.12060546875, + "learning_rate": 0.0017336966197956986, + "loss": 0.0742, + "step": 24879 + }, + { + "epoch": 0.21597034747962257, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017336754853210224, + "loss": 0.1387, + "step": 24880 + }, + { + "epoch": 0.21597902795982674, + "grad_norm": 0.24609375, + "learning_rate": 0.0017336543501533684, + "loss": 0.0996, + "step": 24881 + }, + { + "epoch": 0.2159877084400309, + "grad_norm": 0.10400390625, + "learning_rate": 0.0017336332142927596, + "loss": 0.1113, + "step": 24882 + }, + { + "epoch": 0.21599638892023507, + "grad_norm": 0.4375, + "learning_rate": 0.0017336120777392185, + "loss": 0.0825, + "step": 24883 + }, + { + "epoch": 0.21600506940043923, + "grad_norm": 0.0947265625, + "learning_rate": 0.0017335909404927695, + "loss": 0.1001, + "step": 24884 + }, + { + "epoch": 0.2160137498806434, + "grad_norm": 0.0712890625, + "learning_rate": 0.0017335698025534346, + "loss": 0.1201, + "step": 24885 + }, + { + "epoch": 0.21602243036084756, + "grad_norm": 0.388671875, + "learning_rate": 0.0017335486639212371, + "loss": 0.105, + "step": 24886 + }, + { + "epoch": 0.21603111084105173, + "grad_norm": 0.212890625, + "learning_rate": 0.0017335275245962006, + "loss": 0.0986, + "step": 24887 + }, + { + "epoch": 0.2160397913212559, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017335063845783478, + "loss": 0.1328, + "step": 24888 + }, + { + "epoch": 0.21604847180146006, + "grad_norm": 0.255859375, + "learning_rate": 0.001733485243867702, + "loss": 0.1309, + "step": 24889 + }, + { + "epoch": 0.21605715228166422, + "grad_norm": 0.29296875, + "learning_rate": 0.0017334641024642863, + "loss": 0.1191, + "step": 24890 + }, + { + "epoch": 0.2160658327618684, + "grad_norm": 0.17578125, + "learning_rate": 0.001733442960368124, + "loss": 0.1133, + "step": 24891 + }, + { + "epoch": 0.21607451324207255, + "grad_norm": 0.3203125, + "learning_rate": 0.0017334218175792373, + "loss": 0.0664, + "step": 24892 + }, + { + "epoch": 0.21608319372227672, + "grad_norm": 0.07373046875, + "learning_rate": 0.0017334006740976507, + "loss": 0.103, + "step": 24893 + }, + { + "epoch": 0.21609187420248088, + "grad_norm": 0.671875, + "learning_rate": 0.0017333795299233862, + "loss": 0.1172, + "step": 24894 + }, + { + "epoch": 0.21610055468268505, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017333583850564676, + "loss": 0.1553, + "step": 24895 + }, + { + "epoch": 0.2161092351628892, + "grad_norm": 0.1650390625, + "learning_rate": 0.0017333372394969175, + "loss": 0.1436, + "step": 24896 + }, + { + "epoch": 0.21611791564309338, + "grad_norm": 0.1640625, + "learning_rate": 0.0017333160932447593, + "loss": 0.1133, + "step": 24897 + }, + { + "epoch": 0.21612659612329754, + "grad_norm": 0.349609375, + "learning_rate": 0.0017332949463000165, + "loss": 0.1055, + "step": 24898 + }, + { + "epoch": 0.2161352766035017, + "grad_norm": 0.142578125, + "learning_rate": 0.0017332737986627122, + "loss": 0.1167, + "step": 24899 + }, + { + "epoch": 0.21614395708370587, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017332526503328685, + "loss": 0.0908, + "step": 24900 + }, + { + "epoch": 0.21615263756391004, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017332315013105094, + "loss": 0.1069, + "step": 24901 + }, + { + "epoch": 0.2161613180441142, + "grad_norm": 0.25, + "learning_rate": 0.001733210351595658, + "loss": 0.1084, + "step": 24902 + }, + { + "epoch": 0.21616999852431837, + "grad_norm": 0.40234375, + "learning_rate": 0.0017331892011883368, + "loss": 0.0713, + "step": 24903 + }, + { + "epoch": 0.21617867900452253, + "grad_norm": 0.13671875, + "learning_rate": 0.0017331680500885698, + "loss": 0.0967, + "step": 24904 + }, + { + "epoch": 0.2161873594847267, + "grad_norm": 0.06982421875, + "learning_rate": 0.0017331468982963796, + "loss": 0.0752, + "step": 24905 + }, + { + "epoch": 0.21619603996493086, + "grad_norm": 0.197265625, + "learning_rate": 0.0017331257458117896, + "loss": 0.0693, + "step": 24906 + }, + { + "epoch": 0.21620472044513503, + "grad_norm": 0.31640625, + "learning_rate": 0.0017331045926348228, + "loss": 0.1416, + "step": 24907 + }, + { + "epoch": 0.2162134009253392, + "grad_norm": 0.50390625, + "learning_rate": 0.0017330834387655023, + "loss": 0.1318, + "step": 24908 + }, + { + "epoch": 0.21622208140554336, + "grad_norm": 0.85546875, + "learning_rate": 0.001733062284203851, + "loss": 0.1279, + "step": 24909 + }, + { + "epoch": 0.21623076188574752, + "grad_norm": 0.453125, + "learning_rate": 0.0017330411289498927, + "loss": 0.0947, + "step": 24910 + }, + { + "epoch": 0.2162394423659517, + "grad_norm": 0.48828125, + "learning_rate": 0.00173301997300365, + "loss": 0.0874, + "step": 24911 + }, + { + "epoch": 0.21624812284615585, + "grad_norm": 0.93359375, + "learning_rate": 0.0017329988163651462, + "loss": 0.1475, + "step": 24912 + }, + { + "epoch": 0.21625680332636002, + "grad_norm": 0.14453125, + "learning_rate": 0.0017329776590344044, + "loss": 0.1064, + "step": 24913 + }, + { + "epoch": 0.21626548380656418, + "grad_norm": 0.53125, + "learning_rate": 0.0017329565010114475, + "loss": 0.125, + "step": 24914 + }, + { + "epoch": 0.21627416428676835, + "grad_norm": 0.12158203125, + "learning_rate": 0.0017329353422962993, + "loss": 0.0859, + "step": 24915 + }, + { + "epoch": 0.21628284476697252, + "grad_norm": 0.06884765625, + "learning_rate": 0.001732914182888982, + "loss": 0.0928, + "step": 24916 + }, + { + "epoch": 0.21629152524717668, + "grad_norm": 0.76171875, + "learning_rate": 0.0017328930227895195, + "loss": 0.0977, + "step": 24917 + }, + { + "epoch": 0.21630020572738085, + "grad_norm": 0.353515625, + "learning_rate": 0.0017328718619979348, + "loss": 0.1162, + "step": 24918 + }, + { + "epoch": 0.216308886207585, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017328507005142507, + "loss": 0.1099, + "step": 24919 + }, + { + "epoch": 0.21631756668778918, + "grad_norm": 0.470703125, + "learning_rate": 0.001732829538338491, + "loss": 0.1074, + "step": 24920 + }, + { + "epoch": 0.21632624716799334, + "grad_norm": 0.154296875, + "learning_rate": 0.001732808375470678, + "loss": 0.168, + "step": 24921 + }, + { + "epoch": 0.2163349276481975, + "grad_norm": 0.34375, + "learning_rate": 0.0017327872119108354, + "loss": 0.1045, + "step": 24922 + }, + { + "epoch": 0.21634360812840167, + "grad_norm": 0.279296875, + "learning_rate": 0.0017327660476589863, + "loss": 0.1025, + "step": 24923 + }, + { + "epoch": 0.21635228860860584, + "grad_norm": 0.314453125, + "learning_rate": 0.001732744882715154, + "loss": 0.0938, + "step": 24924 + }, + { + "epoch": 0.21636096908881, + "grad_norm": 0.42578125, + "learning_rate": 0.001732723717079361, + "loss": 0.1167, + "step": 24925 + }, + { + "epoch": 0.21636964956901417, + "grad_norm": 0.369140625, + "learning_rate": 0.001732702550751631, + "loss": 0.0723, + "step": 24926 + }, + { + "epoch": 0.21637833004921833, + "grad_norm": 0.2421875, + "learning_rate": 0.0017326813837319867, + "loss": 0.1074, + "step": 24927 + }, + { + "epoch": 0.2163870105294225, + "grad_norm": 0.306640625, + "learning_rate": 0.001732660216020452, + "loss": 0.0854, + "step": 24928 + }, + { + "epoch": 0.21639569100962666, + "grad_norm": 0.126953125, + "learning_rate": 0.0017326390476170494, + "loss": 0.0977, + "step": 24929 + }, + { + "epoch": 0.21640437148983083, + "grad_norm": 0.140625, + "learning_rate": 0.0017326178785218023, + "loss": 0.0996, + "step": 24930 + }, + { + "epoch": 0.216413051970035, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017325967087347335, + "loss": 0.0913, + "step": 24931 + }, + { + "epoch": 0.21642173245023916, + "grad_norm": 0.103515625, + "learning_rate": 0.001732575538255867, + "loss": 0.0967, + "step": 24932 + }, + { + "epoch": 0.21643041293044332, + "grad_norm": 0.310546875, + "learning_rate": 0.001732554367085225, + "loss": 0.1094, + "step": 24933 + }, + { + "epoch": 0.2164390934106475, + "grad_norm": 0.380859375, + "learning_rate": 0.0017325331952228312, + "loss": 0.1357, + "step": 24934 + }, + { + "epoch": 0.21644777389085165, + "grad_norm": 0.0986328125, + "learning_rate": 0.0017325120226687083, + "loss": 0.124, + "step": 24935 + }, + { + "epoch": 0.21645645437105582, + "grad_norm": 0.46875, + "learning_rate": 0.00173249084942288, + "loss": 0.1172, + "step": 24936 + }, + { + "epoch": 0.21646513485125998, + "grad_norm": 0.263671875, + "learning_rate": 0.0017324696754853692, + "loss": 0.0928, + "step": 24937 + }, + { + "epoch": 0.21647381533146415, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017324485008561991, + "loss": 0.0732, + "step": 24938 + }, + { + "epoch": 0.2164824958116683, + "grad_norm": 0.220703125, + "learning_rate": 0.0017324273255353928, + "loss": 0.104, + "step": 24939 + }, + { + "epoch": 0.21649117629187248, + "grad_norm": 0.376953125, + "learning_rate": 0.0017324061495229736, + "loss": 0.1309, + "step": 24940 + }, + { + "epoch": 0.21649985677207664, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017323849728189645, + "loss": 0.1016, + "step": 24941 + }, + { + "epoch": 0.2165085372522808, + "grad_norm": 0.150390625, + "learning_rate": 0.0017323637954233884, + "loss": 0.1787, + "step": 24942 + }, + { + "epoch": 0.21651721773248497, + "grad_norm": 0.11328125, + "learning_rate": 0.0017323426173362691, + "loss": 0.1138, + "step": 24943 + }, + { + "epoch": 0.21652589821268914, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017323214385576292, + "loss": 0.1279, + "step": 24944 + }, + { + "epoch": 0.2165345786928933, + "grad_norm": 0.265625, + "learning_rate": 0.0017323002590874921, + "loss": 0.1055, + "step": 24945 + }, + { + "epoch": 0.21654325917309747, + "grad_norm": 0.234375, + "learning_rate": 0.0017322790789258808, + "loss": 0.1182, + "step": 24946 + }, + { + "epoch": 0.21655193965330163, + "grad_norm": 0.37890625, + "learning_rate": 0.001732257898072819, + "loss": 0.124, + "step": 24947 + }, + { + "epoch": 0.2165606201335058, + "grad_norm": 0.1875, + "learning_rate": 0.0017322367165283291, + "loss": 0.1074, + "step": 24948 + }, + { + "epoch": 0.21656930061370996, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017322155342924348, + "loss": 0.0845, + "step": 24949 + }, + { + "epoch": 0.2165779810939141, + "grad_norm": 0.55859375, + "learning_rate": 0.0017321943513651592, + "loss": 0.166, + "step": 24950 + }, + { + "epoch": 0.21658666157411827, + "grad_norm": 0.3203125, + "learning_rate": 0.001732173167746525, + "loss": 0.123, + "step": 24951 + }, + { + "epoch": 0.21659534205432243, + "grad_norm": 0.16796875, + "learning_rate": 0.001732151983436556, + "loss": 0.1167, + "step": 24952 + }, + { + "epoch": 0.2166040225345266, + "grad_norm": 0.07470703125, + "learning_rate": 0.001732130798435275, + "loss": 0.0879, + "step": 24953 + }, + { + "epoch": 0.21661270301473076, + "grad_norm": 0.384765625, + "learning_rate": 0.0017321096127427054, + "loss": 0.082, + "step": 24954 + }, + { + "epoch": 0.21662138349493493, + "grad_norm": 0.32421875, + "learning_rate": 0.0017320884263588702, + "loss": 0.1396, + "step": 24955 + }, + { + "epoch": 0.2166300639751391, + "grad_norm": 0.29296875, + "learning_rate": 0.0017320672392837925, + "loss": 0.105, + "step": 24956 + }, + { + "epoch": 0.21663874445534326, + "grad_norm": 0.11572265625, + "learning_rate": 0.0017320460515174955, + "loss": 0.084, + "step": 24957 + }, + { + "epoch": 0.21664742493554742, + "grad_norm": 0.06494140625, + "learning_rate": 0.0017320248630600026, + "loss": 0.0913, + "step": 24958 + }, + { + "epoch": 0.2166561054157516, + "grad_norm": 0.330078125, + "learning_rate": 0.0017320036739113364, + "loss": 0.1055, + "step": 24959 + }, + { + "epoch": 0.21666478589595575, + "grad_norm": 0.546875, + "learning_rate": 0.001731982484071521, + "loss": 0.106, + "step": 24960 + }, + { + "epoch": 0.21667346637615992, + "grad_norm": 0.07177734375, + "learning_rate": 0.0017319612935405785, + "loss": 0.0947, + "step": 24961 + }, + { + "epoch": 0.21668214685636408, + "grad_norm": 0.3125, + "learning_rate": 0.0017319401023185332, + "loss": 0.0894, + "step": 24962 + }, + { + "epoch": 0.21669082733656825, + "grad_norm": 0.455078125, + "learning_rate": 0.0017319189104054074, + "loss": 0.1104, + "step": 24963 + }, + { + "epoch": 0.2166995078167724, + "grad_norm": 0.345703125, + "learning_rate": 0.0017318977178012247, + "loss": 0.1099, + "step": 24964 + }, + { + "epoch": 0.21670818829697658, + "grad_norm": 0.11083984375, + "learning_rate": 0.0017318765245060079, + "loss": 0.1094, + "step": 24965 + }, + { + "epoch": 0.21671686877718074, + "grad_norm": 0.1494140625, + "learning_rate": 0.0017318553305197806, + "loss": 0.1045, + "step": 24966 + }, + { + "epoch": 0.2167255492573849, + "grad_norm": 0.181640625, + "learning_rate": 0.001731834135842566, + "loss": 0.1133, + "step": 24967 + }, + { + "epoch": 0.21673422973758907, + "grad_norm": 0.259765625, + "learning_rate": 0.0017318129404743868, + "loss": 0.0752, + "step": 24968 + }, + { + "epoch": 0.21674291021779324, + "grad_norm": 0.1767578125, + "learning_rate": 0.0017317917444152667, + "loss": 0.0898, + "step": 24969 + }, + { + "epoch": 0.2167515906979974, + "grad_norm": 0.125, + "learning_rate": 0.0017317705476652286, + "loss": 0.1289, + "step": 24970 + }, + { + "epoch": 0.21676027117820157, + "grad_norm": 0.48046875, + "learning_rate": 0.0017317493502242954, + "loss": 0.124, + "step": 24971 + }, + { + "epoch": 0.21676895165840573, + "grad_norm": 0.1650390625, + "learning_rate": 0.001731728152092491, + "loss": 0.1797, + "step": 24972 + }, + { + "epoch": 0.2167776321386099, + "grad_norm": 0.06396484375, + "learning_rate": 0.001731706953269838, + "loss": 0.1035, + "step": 24973 + }, + { + "epoch": 0.21678631261881406, + "grad_norm": 0.11767578125, + "learning_rate": 0.0017316857537563598, + "loss": 0.0874, + "step": 24974 + }, + { + "epoch": 0.21679499309901823, + "grad_norm": 0.283203125, + "learning_rate": 0.0017316645535520791, + "loss": 0.0947, + "step": 24975 + }, + { + "epoch": 0.2168036735792224, + "grad_norm": 0.28515625, + "learning_rate": 0.00173164335265702, + "loss": 0.0869, + "step": 24976 + }, + { + "epoch": 0.21681235405942656, + "grad_norm": 0.5234375, + "learning_rate": 0.0017316221510712056, + "loss": 0.0986, + "step": 24977 + }, + { + "epoch": 0.21682103453963072, + "grad_norm": 0.32421875, + "learning_rate": 0.001731600948794658, + "loss": 0.1025, + "step": 24978 + }, + { + "epoch": 0.2168297150198349, + "grad_norm": 0.197265625, + "learning_rate": 0.0017315797458274016, + "loss": 0.1289, + "step": 24979 + }, + { + "epoch": 0.21683839550003906, + "grad_norm": 0.435546875, + "learning_rate": 0.0017315585421694586, + "loss": 0.126, + "step": 24980 + }, + { + "epoch": 0.21684707598024322, + "grad_norm": 0.21484375, + "learning_rate": 0.0017315373378208531, + "loss": 0.1138, + "step": 24981 + }, + { + "epoch": 0.21685575646044739, + "grad_norm": 0.47265625, + "learning_rate": 0.0017315161327816075, + "loss": 0.1064, + "step": 24982 + }, + { + "epoch": 0.21686443694065155, + "grad_norm": 0.455078125, + "learning_rate": 0.0017314949270517458, + "loss": 0.1211, + "step": 24983 + }, + { + "epoch": 0.21687311742085572, + "grad_norm": 0.091796875, + "learning_rate": 0.0017314737206312902, + "loss": 0.0928, + "step": 24984 + }, + { + "epoch": 0.21688179790105988, + "grad_norm": 0.376953125, + "learning_rate": 0.0017314525135202648, + "loss": 0.0947, + "step": 24985 + }, + { + "epoch": 0.21689047838126405, + "grad_norm": 0.42578125, + "learning_rate": 0.0017314313057186924, + "loss": 0.0762, + "step": 24986 + }, + { + "epoch": 0.2168991588614682, + "grad_norm": 0.09375, + "learning_rate": 0.0017314100972265958, + "loss": 0.1123, + "step": 24987 + }, + { + "epoch": 0.21690783934167238, + "grad_norm": 0.4375, + "learning_rate": 0.0017313888880439992, + "loss": 0.1572, + "step": 24988 + }, + { + "epoch": 0.21691651982187654, + "grad_norm": 0.166015625, + "learning_rate": 0.001731367678170925, + "loss": 0.1025, + "step": 24989 + }, + { + "epoch": 0.2169252003020807, + "grad_norm": 0.2138671875, + "learning_rate": 0.0017313464676073965, + "loss": 0.1191, + "step": 24990 + }, + { + "epoch": 0.21693388078228487, + "grad_norm": 0.90625, + "learning_rate": 0.0017313252563534367, + "loss": 0.084, + "step": 24991 + }, + { + "epoch": 0.21694256126248904, + "grad_norm": 0.40625, + "learning_rate": 0.0017313040444090694, + "loss": 0.0889, + "step": 24992 + }, + { + "epoch": 0.2169512417426932, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017312828317743176, + "loss": 0.1309, + "step": 24993 + }, + { + "epoch": 0.21695992222289737, + "grad_norm": 0.94921875, + "learning_rate": 0.0017312616184492042, + "loss": 0.1006, + "step": 24994 + }, + { + "epoch": 0.21696860270310153, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017312404044337528, + "loss": 0.1279, + "step": 24995 + }, + { + "epoch": 0.2169772831833057, + "grad_norm": 0.193359375, + "learning_rate": 0.001731219189727986, + "loss": 0.082, + "step": 24996 + }, + { + "epoch": 0.21698596366350986, + "grad_norm": 0.55859375, + "learning_rate": 0.0017311979743319275, + "loss": 0.3008, + "step": 24997 + }, + { + "epoch": 0.21699464414371403, + "grad_norm": 0.29296875, + "learning_rate": 0.0017311767582456007, + "loss": 0.0894, + "step": 24998 + }, + { + "epoch": 0.2170033246239182, + "grad_norm": 0.1982421875, + "learning_rate": 0.001731155541469028, + "loss": 0.1021, + "step": 24999 + }, + { + "epoch": 0.21701200510412236, + "grad_norm": 0.45703125, + "learning_rate": 0.0017311343240022335, + "loss": 0.1533, + "step": 25000 + }, + { + "epoch": 0.21702068558432652, + "grad_norm": 0.125, + "learning_rate": 0.0017311131058452395, + "loss": 0.1143, + "step": 25001 + }, + { + "epoch": 0.2170293660645307, + "grad_norm": 0.2734375, + "learning_rate": 0.0017310918869980704, + "loss": 0.0923, + "step": 25002 + }, + { + "epoch": 0.21703804654473485, + "grad_norm": 0.2470703125, + "learning_rate": 0.0017310706674607483, + "loss": 0.1299, + "step": 25003 + }, + { + "epoch": 0.21704672702493902, + "grad_norm": 0.294921875, + "learning_rate": 0.0017310494472332967, + "loss": 0.1235, + "step": 25004 + }, + { + "epoch": 0.21705540750514318, + "grad_norm": 0.12890625, + "learning_rate": 0.001731028226315739, + "loss": 0.1309, + "step": 25005 + }, + { + "epoch": 0.21706408798534735, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017310070047080984, + "loss": 0.1074, + "step": 25006 + }, + { + "epoch": 0.2170727684655515, + "grad_norm": 0.2734375, + "learning_rate": 0.0017309857824103977, + "loss": 0.1182, + "step": 25007 + }, + { + "epoch": 0.21708144894575568, + "grad_norm": 0.8203125, + "learning_rate": 0.0017309645594226607, + "loss": 0.1445, + "step": 25008 + }, + { + "epoch": 0.21709012942595984, + "grad_norm": 0.6171875, + "learning_rate": 0.0017309433357449102, + "loss": 0.1162, + "step": 25009 + }, + { + "epoch": 0.217098809906164, + "grad_norm": 0.291015625, + "learning_rate": 0.0017309221113771698, + "loss": 0.1562, + "step": 25010 + }, + { + "epoch": 0.21710749038636817, + "grad_norm": 0.494140625, + "learning_rate": 0.0017309008863194623, + "loss": 0.0913, + "step": 25011 + }, + { + "epoch": 0.21711617086657234, + "grad_norm": 0.2294921875, + "learning_rate": 0.001730879660571811, + "loss": 0.0864, + "step": 25012 + }, + { + "epoch": 0.2171248513467765, + "grad_norm": 0.76171875, + "learning_rate": 0.0017308584341342395, + "loss": 0.1406, + "step": 25013 + }, + { + "epoch": 0.21713353182698067, + "grad_norm": 0.451171875, + "learning_rate": 0.0017308372070067705, + "loss": 0.1406, + "step": 25014 + }, + { + "epoch": 0.21714221230718483, + "grad_norm": 0.474609375, + "learning_rate": 0.0017308159791894274, + "loss": 0.1055, + "step": 25015 + }, + { + "epoch": 0.217150892787389, + "grad_norm": 0.546875, + "learning_rate": 0.0017307947506822335, + "loss": 0.1895, + "step": 25016 + }, + { + "epoch": 0.21715957326759316, + "grad_norm": 0.28125, + "learning_rate": 0.0017307735214852115, + "loss": 0.0967, + "step": 25017 + }, + { + "epoch": 0.21716825374779733, + "grad_norm": 0.1748046875, + "learning_rate": 0.0017307522915983853, + "loss": 0.125, + "step": 25018 + }, + { + "epoch": 0.2171769342280015, + "grad_norm": 0.2021484375, + "learning_rate": 0.0017307310610217782, + "loss": 0.1084, + "step": 25019 + }, + { + "epoch": 0.21718561470820566, + "grad_norm": 0.1083984375, + "learning_rate": 0.001730709829755413, + "loss": 0.0962, + "step": 25020 + }, + { + "epoch": 0.21719429518840982, + "grad_norm": 1.28125, + "learning_rate": 0.0017306885977993128, + "loss": 0.1104, + "step": 25021 + }, + { + "epoch": 0.217202975668614, + "grad_norm": 0.56640625, + "learning_rate": 0.0017306673651535007, + "loss": 0.126, + "step": 25022 + }, + { + "epoch": 0.21721165614881816, + "grad_norm": 1.3984375, + "learning_rate": 0.0017306461318180007, + "loss": 0.1201, + "step": 25023 + }, + { + "epoch": 0.21722033662902232, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017306248977928356, + "loss": 0.0845, + "step": 25024 + }, + { + "epoch": 0.21722901710922649, + "grad_norm": 0.357421875, + "learning_rate": 0.0017306036630780287, + "loss": 0.1025, + "step": 25025 + }, + { + "epoch": 0.21723769758943065, + "grad_norm": 0.4296875, + "learning_rate": 0.0017305824276736027, + "loss": 0.1543, + "step": 25026 + }, + { + "epoch": 0.21724637806963482, + "grad_norm": 0.53515625, + "learning_rate": 0.0017305611915795814, + "loss": 0.1475, + "step": 25027 + }, + { + "epoch": 0.21725505854983898, + "grad_norm": 0.5, + "learning_rate": 0.0017305399547959876, + "loss": 0.0957, + "step": 25028 + }, + { + "epoch": 0.21726373903004315, + "grad_norm": 0.255859375, + "learning_rate": 0.0017305187173228453, + "loss": 0.0957, + "step": 25029 + }, + { + "epoch": 0.2172724195102473, + "grad_norm": 0.578125, + "learning_rate": 0.0017304974791601771, + "loss": 0.127, + "step": 25030 + }, + { + "epoch": 0.21728109999045148, + "grad_norm": 0.0732421875, + "learning_rate": 0.001730476240308006, + "loss": 0.085, + "step": 25031 + }, + { + "epoch": 0.21728978047065564, + "grad_norm": 0.404296875, + "learning_rate": 0.0017304550007663558, + "loss": 0.0981, + "step": 25032 + }, + { + "epoch": 0.2172984609508598, + "grad_norm": 0.5703125, + "learning_rate": 0.0017304337605352493, + "loss": 0.0776, + "step": 25033 + }, + { + "epoch": 0.21730714143106397, + "grad_norm": 0.57421875, + "learning_rate": 0.0017304125196147102, + "loss": 0.1494, + "step": 25034 + }, + { + "epoch": 0.21731582191126814, + "grad_norm": 0.294921875, + "learning_rate": 0.0017303912780047613, + "loss": 0.0913, + "step": 25035 + }, + { + "epoch": 0.2173245023914723, + "grad_norm": 0.083984375, + "learning_rate": 0.0017303700357054258, + "loss": 0.0986, + "step": 25036 + }, + { + "epoch": 0.21733318287167647, + "grad_norm": 0.41015625, + "learning_rate": 0.0017303487927167272, + "loss": 0.0981, + "step": 25037 + }, + { + "epoch": 0.21734186335188063, + "grad_norm": 0.361328125, + "learning_rate": 0.001730327549038689, + "loss": 0.0864, + "step": 25038 + }, + { + "epoch": 0.2173505438320848, + "grad_norm": 0.177734375, + "learning_rate": 0.0017303063046713333, + "loss": 0.0957, + "step": 25039 + }, + { + "epoch": 0.21735922431228896, + "grad_norm": 0.625, + "learning_rate": 0.0017302850596146845, + "loss": 0.0796, + "step": 25040 + }, + { + "epoch": 0.21736790479249313, + "grad_norm": 0.150390625, + "learning_rate": 0.0017302638138687657, + "loss": 0.1123, + "step": 25041 + }, + { + "epoch": 0.2173765852726973, + "grad_norm": 0.125, + "learning_rate": 0.0017302425674335996, + "loss": 0.1006, + "step": 25042 + }, + { + "epoch": 0.21738526575290146, + "grad_norm": 0.515625, + "learning_rate": 0.0017302213203092096, + "loss": 0.1885, + "step": 25043 + }, + { + "epoch": 0.21739394623310562, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017302000724956192, + "loss": 0.0908, + "step": 25044 + }, + { + "epoch": 0.2174026267133098, + "grad_norm": 0.310546875, + "learning_rate": 0.0017301788239928513, + "loss": 0.0771, + "step": 25045 + }, + { + "epoch": 0.21741130719351395, + "grad_norm": 0.271484375, + "learning_rate": 0.0017301575748009297, + "loss": 0.1113, + "step": 25046 + }, + { + "epoch": 0.21741998767371812, + "grad_norm": 0.33203125, + "learning_rate": 0.001730136324919877, + "loss": 0.207, + "step": 25047 + }, + { + "epoch": 0.21742866815392228, + "grad_norm": 0.70703125, + "learning_rate": 0.0017301150743497164, + "loss": 0.1045, + "step": 25048 + }, + { + "epoch": 0.21743734863412645, + "grad_norm": 0.31640625, + "learning_rate": 0.0017300938230904717, + "loss": 0.1289, + "step": 25049 + }, + { + "epoch": 0.2174460291143306, + "grad_norm": 0.65625, + "learning_rate": 0.001730072571142166, + "loss": 0.1387, + "step": 25050 + }, + { + "epoch": 0.21745470959453478, + "grad_norm": 0.2578125, + "learning_rate": 0.0017300513185048224, + "loss": 0.0986, + "step": 25051 + }, + { + "epoch": 0.21746339007473894, + "grad_norm": 0.47265625, + "learning_rate": 0.0017300300651784636, + "loss": 0.1187, + "step": 25052 + }, + { + "epoch": 0.2174720705549431, + "grad_norm": 0.10302734375, + "learning_rate": 0.0017300088111631138, + "loss": 0.0767, + "step": 25053 + }, + { + "epoch": 0.21748075103514727, + "grad_norm": 0.23046875, + "learning_rate": 0.0017299875564587958, + "loss": 0.123, + "step": 25054 + }, + { + "epoch": 0.21748943151535144, + "grad_norm": 0.60546875, + "learning_rate": 0.001729966301065533, + "loss": 0.1162, + "step": 25055 + }, + { + "epoch": 0.2174981119955556, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017299450449833485, + "loss": 0.1006, + "step": 25056 + }, + { + "epoch": 0.21750679247575977, + "grad_norm": 0.240234375, + "learning_rate": 0.0017299237882122654, + "loss": 0.0957, + "step": 25057 + }, + { + "epoch": 0.21751547295596393, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017299025307523073, + "loss": 0.1064, + "step": 25058 + }, + { + "epoch": 0.2175241534361681, + "grad_norm": 0.416015625, + "learning_rate": 0.001729881272603497, + "loss": 0.0991, + "step": 25059 + }, + { + "epoch": 0.21753283391637226, + "grad_norm": 0.2734375, + "learning_rate": 0.0017298600137658582, + "loss": 0.1055, + "step": 25060 + }, + { + "epoch": 0.21754151439657643, + "grad_norm": 0.392578125, + "learning_rate": 0.0017298387542394136, + "loss": 0.1143, + "step": 25061 + }, + { + "epoch": 0.2175501948767806, + "grad_norm": 0.2392578125, + "learning_rate": 0.0017298174940241873, + "loss": 0.123, + "step": 25062 + }, + { + "epoch": 0.21755887535698476, + "grad_norm": 0.41015625, + "learning_rate": 0.0017297962331202018, + "loss": 0.1221, + "step": 25063 + }, + { + "epoch": 0.21756755583718893, + "grad_norm": 0.0703125, + "learning_rate": 0.001729774971527481, + "loss": 0.0879, + "step": 25064 + }, + { + "epoch": 0.2175762363173931, + "grad_norm": 0.78125, + "learning_rate": 0.001729753709246047, + "loss": 0.1143, + "step": 25065 + }, + { + "epoch": 0.21758491679759726, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017297324462759243, + "loss": 0.0967, + "step": 25066 + }, + { + "epoch": 0.21759359727780142, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017297111826171355, + "loss": 0.0898, + "step": 25067 + }, + { + "epoch": 0.21760227775800559, + "grad_norm": 0.125, + "learning_rate": 0.0017296899182697042, + "loss": 0.1211, + "step": 25068 + }, + { + "epoch": 0.21761095823820975, + "grad_norm": 1.296875, + "learning_rate": 0.0017296686532336531, + "loss": 0.1445, + "step": 25069 + }, + { + "epoch": 0.21761963871841392, + "grad_norm": 0.89453125, + "learning_rate": 0.0017296473875090062, + "loss": 0.0752, + "step": 25070 + }, + { + "epoch": 0.21762831919861808, + "grad_norm": 0.23046875, + "learning_rate": 0.0017296261210957861, + "loss": 0.104, + "step": 25071 + }, + { + "epoch": 0.21763699967882225, + "grad_norm": 0.2421875, + "learning_rate": 0.0017296048539940168, + "loss": 0.127, + "step": 25072 + }, + { + "epoch": 0.2176456801590264, + "grad_norm": 0.255859375, + "learning_rate": 0.0017295835862037204, + "loss": 0.0752, + "step": 25073 + }, + { + "epoch": 0.21765436063923055, + "grad_norm": 0.16796875, + "learning_rate": 0.0017295623177249213, + "loss": 0.1787, + "step": 25074 + }, + { + "epoch": 0.2176630411194347, + "grad_norm": 0.279296875, + "learning_rate": 0.0017295410485576423, + "loss": 0.125, + "step": 25075 + }, + { + "epoch": 0.21767172159963888, + "grad_norm": 0.640625, + "learning_rate": 0.0017295197787019065, + "loss": 0.1094, + "step": 25076 + }, + { + "epoch": 0.21768040207984304, + "grad_norm": 0.09228515625, + "learning_rate": 0.0017294985081577378, + "loss": 0.0923, + "step": 25077 + }, + { + "epoch": 0.2176890825600472, + "grad_norm": 0.25390625, + "learning_rate": 0.0017294772369251584, + "loss": 0.0981, + "step": 25078 + }, + { + "epoch": 0.21769776304025137, + "grad_norm": 0.11572265625, + "learning_rate": 0.0017294559650041922, + "loss": 0.0977, + "step": 25079 + }, + { + "epoch": 0.21770644352045554, + "grad_norm": 0.203125, + "learning_rate": 0.0017294346923948625, + "loss": 0.1172, + "step": 25080 + }, + { + "epoch": 0.2177151240006597, + "grad_norm": 0.54296875, + "learning_rate": 0.0017294134190971925, + "loss": 0.1367, + "step": 25081 + }, + { + "epoch": 0.21772380448086387, + "grad_norm": 0.21875, + "learning_rate": 0.0017293921451112053, + "loss": 0.082, + "step": 25082 + }, + { + "epoch": 0.21773248496106803, + "grad_norm": 0.9140625, + "learning_rate": 0.0017293708704369246, + "loss": 0.126, + "step": 25083 + }, + { + "epoch": 0.2177411654412722, + "grad_norm": 0.439453125, + "learning_rate": 0.0017293495950743735, + "loss": 0.1079, + "step": 25084 + }, + { + "epoch": 0.21774984592147636, + "grad_norm": 1.078125, + "learning_rate": 0.001729328319023575, + "loss": 0.0991, + "step": 25085 + }, + { + "epoch": 0.21775852640168053, + "grad_norm": 0.7890625, + "learning_rate": 0.0017293070422845522, + "loss": 0.1201, + "step": 25086 + }, + { + "epoch": 0.2177672068818847, + "grad_norm": 0.421875, + "learning_rate": 0.001729285764857329, + "loss": 0.1084, + "step": 25087 + }, + { + "epoch": 0.21777588736208886, + "grad_norm": 0.330078125, + "learning_rate": 0.0017292644867419281, + "loss": 0.1348, + "step": 25088 + }, + { + "epoch": 0.21778456784229303, + "grad_norm": 0.400390625, + "learning_rate": 0.001729243207938373, + "loss": 0.1348, + "step": 25089 + }, + { + "epoch": 0.2177932483224972, + "grad_norm": 0.328125, + "learning_rate": 0.0017292219284466876, + "loss": 0.1172, + "step": 25090 + }, + { + "epoch": 0.21780192880270136, + "grad_norm": 0.326171875, + "learning_rate": 0.0017292006482668937, + "loss": 0.0811, + "step": 25091 + }, + { + "epoch": 0.21781060928290552, + "grad_norm": 0.365234375, + "learning_rate": 0.001729179367399016, + "loss": 0.1201, + "step": 25092 + }, + { + "epoch": 0.21781928976310969, + "grad_norm": 0.380859375, + "learning_rate": 0.001729158085843077, + "loss": 0.0962, + "step": 25093 + }, + { + "epoch": 0.21782797024331385, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017291368035991003, + "loss": 0.1338, + "step": 25094 + }, + { + "epoch": 0.21783665072351802, + "grad_norm": 0.181640625, + "learning_rate": 0.0017291155206671086, + "loss": 0.0879, + "step": 25095 + }, + { + "epoch": 0.21784533120372218, + "grad_norm": 0.111328125, + "learning_rate": 0.0017290942370471259, + "loss": 0.0972, + "step": 25096 + }, + { + "epoch": 0.21785401168392635, + "grad_norm": 0.296875, + "learning_rate": 0.0017290729527391752, + "loss": 0.1387, + "step": 25097 + }, + { + "epoch": 0.2178626921641305, + "grad_norm": 0.396484375, + "learning_rate": 0.0017290516677432798, + "loss": 0.125, + "step": 25098 + }, + { + "epoch": 0.21787137264433468, + "grad_norm": 0.12890625, + "learning_rate": 0.0017290303820594632, + "loss": 0.124, + "step": 25099 + }, + { + "epoch": 0.21788005312453884, + "grad_norm": 0.298828125, + "learning_rate": 0.001729009095687748, + "loss": 0.0967, + "step": 25100 + }, + { + "epoch": 0.217888733604743, + "grad_norm": 0.10791015625, + "learning_rate": 0.0017289878086281581, + "loss": 0.125, + "step": 25101 + }, + { + "epoch": 0.21789741408494717, + "grad_norm": 0.34765625, + "learning_rate": 0.0017289665208807166, + "loss": 0.1133, + "step": 25102 + }, + { + "epoch": 0.21790609456515134, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017289452324454467, + "loss": 0.1553, + "step": 25103 + }, + { + "epoch": 0.2179147750453555, + "grad_norm": 0.12109375, + "learning_rate": 0.0017289239433223719, + "loss": 0.0947, + "step": 25104 + }, + { + "epoch": 0.21792345552555967, + "grad_norm": 0.34765625, + "learning_rate": 0.0017289026535115151, + "loss": 0.0928, + "step": 25105 + }, + { + "epoch": 0.21793213600576383, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017288813630129, + "loss": 0.0972, + "step": 25106 + }, + { + "epoch": 0.217940816485968, + "grad_norm": 0.388671875, + "learning_rate": 0.0017288600718265495, + "loss": 0.1172, + "step": 25107 + }, + { + "epoch": 0.21794949696617216, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017288387799524873, + "loss": 0.1045, + "step": 25108 + }, + { + "epoch": 0.21795817744637633, + "grad_norm": 0.10888671875, + "learning_rate": 0.0017288174873907362, + "loss": 0.1094, + "step": 25109 + }, + { + "epoch": 0.2179668579265805, + "grad_norm": 0.095703125, + "learning_rate": 0.0017287961941413197, + "loss": 0.0928, + "step": 25110 + }, + { + "epoch": 0.21797553840678466, + "grad_norm": 0.07275390625, + "learning_rate": 0.0017287749002042612, + "loss": 0.0898, + "step": 25111 + }, + { + "epoch": 0.21798421888698882, + "grad_norm": 0.1630859375, + "learning_rate": 0.001728753605579584, + "loss": 0.1445, + "step": 25112 + }, + { + "epoch": 0.217992899367193, + "grad_norm": 0.51171875, + "learning_rate": 0.0017287323102673114, + "loss": 0.0913, + "step": 25113 + }, + { + "epoch": 0.21800157984739715, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017287110142674665, + "loss": 0.1035, + "step": 25114 + }, + { + "epoch": 0.21801026032760132, + "grad_norm": 0.365234375, + "learning_rate": 0.0017286897175800727, + "loss": 0.1235, + "step": 25115 + }, + { + "epoch": 0.21801894080780548, + "grad_norm": 0.33203125, + "learning_rate": 0.0017286684202051533, + "loss": 0.0874, + "step": 25116 + }, + { + "epoch": 0.21802762128800965, + "grad_norm": 0.376953125, + "learning_rate": 0.0017286471221427313, + "loss": 0.1055, + "step": 25117 + }, + { + "epoch": 0.2180363017682138, + "grad_norm": 0.609375, + "learning_rate": 0.0017286258233928306, + "loss": 0.127, + "step": 25118 + }, + { + "epoch": 0.21804498224841798, + "grad_norm": 0.09375, + "learning_rate": 0.001728604523955474, + "loss": 0.0884, + "step": 25119 + }, + { + "epoch": 0.21805366272862214, + "grad_norm": 0.21875, + "learning_rate": 0.0017285832238306852, + "loss": 0.0815, + "step": 25120 + }, + { + "epoch": 0.2180623432088263, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017285619230184867, + "loss": 0.1182, + "step": 25121 + }, + { + "epoch": 0.21807102368903047, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017285406215189027, + "loss": 0.1099, + "step": 25122 + }, + { + "epoch": 0.21807970416923464, + "grad_norm": 0.08349609375, + "learning_rate": 0.0017285193193319563, + "loss": 0.0723, + "step": 25123 + }, + { + "epoch": 0.2180883846494388, + "grad_norm": 0.734375, + "learning_rate": 0.0017284980164576703, + "loss": 0.1348, + "step": 25124 + }, + { + "epoch": 0.21809706512964297, + "grad_norm": 0.3515625, + "learning_rate": 0.001728476712896068, + "loss": 0.0986, + "step": 25125 + }, + { + "epoch": 0.21810574560984713, + "grad_norm": 0.12451171875, + "learning_rate": 0.0017284554086471734, + "loss": 0.0967, + "step": 25126 + }, + { + "epoch": 0.2181144260900513, + "grad_norm": 0.107421875, + "learning_rate": 0.0017284341037110094, + "loss": 0.0928, + "step": 25127 + }, + { + "epoch": 0.21812310657025547, + "grad_norm": 0.1328125, + "learning_rate": 0.0017284127980875994, + "loss": 0.1084, + "step": 25128 + }, + { + "epoch": 0.21813178705045963, + "grad_norm": 0.287109375, + "learning_rate": 0.0017283914917769663, + "loss": 0.084, + "step": 25129 + }, + { + "epoch": 0.2181404675306638, + "grad_norm": 0.82421875, + "learning_rate": 0.001728370184779134, + "loss": 0.1216, + "step": 25130 + }, + { + "epoch": 0.21814914801086796, + "grad_norm": 0.8125, + "learning_rate": 0.0017283488770941256, + "loss": 0.1001, + "step": 25131 + }, + { + "epoch": 0.21815782849107213, + "grad_norm": 0.236328125, + "learning_rate": 0.001728327568721964, + "loss": 0.0967, + "step": 25132 + }, + { + "epoch": 0.2181665089712763, + "grad_norm": 0.189453125, + "learning_rate": 0.0017283062596626729, + "loss": 0.1387, + "step": 25133 + }, + { + "epoch": 0.21817518945148046, + "grad_norm": 0.2353515625, + "learning_rate": 0.0017282849499162755, + "loss": 0.126, + "step": 25134 + }, + { + "epoch": 0.21818386993168462, + "grad_norm": 1.0859375, + "learning_rate": 0.0017282636394827951, + "loss": 0.1143, + "step": 25135 + }, + { + "epoch": 0.21819255041188879, + "grad_norm": 1.9765625, + "learning_rate": 0.0017282423283622552, + "loss": 0.1621, + "step": 25136 + }, + { + "epoch": 0.21820123089209295, + "grad_norm": 0.4453125, + "learning_rate": 0.0017282210165546787, + "loss": 0.104, + "step": 25137 + }, + { + "epoch": 0.21820991137229712, + "grad_norm": 0.5625, + "learning_rate": 0.0017281997040600895, + "loss": 0.1484, + "step": 25138 + }, + { + "epoch": 0.21821859185250128, + "grad_norm": 0.49609375, + "learning_rate": 0.00172817839087851, + "loss": 0.0996, + "step": 25139 + }, + { + "epoch": 0.21822727233270545, + "grad_norm": 0.1123046875, + "learning_rate": 0.0017281570770099645, + "loss": 0.0947, + "step": 25140 + }, + { + "epoch": 0.2182359528129096, + "grad_norm": 0.4765625, + "learning_rate": 0.0017281357624544756, + "loss": 0.0918, + "step": 25141 + }, + { + "epoch": 0.21824463329311378, + "grad_norm": 0.232421875, + "learning_rate": 0.0017281144472120669, + "loss": 0.0996, + "step": 25142 + }, + { + "epoch": 0.21825331377331794, + "grad_norm": 0.189453125, + "learning_rate": 0.0017280931312827618, + "loss": 0.1104, + "step": 25143 + }, + { + "epoch": 0.2182619942535221, + "grad_norm": 0.423828125, + "learning_rate": 0.0017280718146665832, + "loss": 0.1396, + "step": 25144 + }, + { + "epoch": 0.21827067473372627, + "grad_norm": 0.16015625, + "learning_rate": 0.0017280504973635551, + "loss": 0.1138, + "step": 25145 + }, + { + "epoch": 0.21827935521393044, + "grad_norm": 0.34765625, + "learning_rate": 0.0017280291793737004, + "loss": 0.1055, + "step": 25146 + }, + { + "epoch": 0.2182880356941346, + "grad_norm": 0.7109375, + "learning_rate": 0.0017280078606970422, + "loss": 0.2227, + "step": 25147 + }, + { + "epoch": 0.21829671617433877, + "grad_norm": 0.3515625, + "learning_rate": 0.0017279865413336042, + "loss": 0.1235, + "step": 25148 + }, + { + "epoch": 0.21830539665454293, + "grad_norm": 0.25, + "learning_rate": 0.001727965221283409, + "loss": 0.1348, + "step": 25149 + }, + { + "epoch": 0.2183140771347471, + "grad_norm": 0.13671875, + "learning_rate": 0.0017279439005464811, + "loss": 0.1387, + "step": 25150 + }, + { + "epoch": 0.21832275761495126, + "grad_norm": 0.169921875, + "learning_rate": 0.001727922579122843, + "loss": 0.1016, + "step": 25151 + }, + { + "epoch": 0.21833143809515543, + "grad_norm": 0.30078125, + "learning_rate": 0.0017279012570125185, + "loss": 0.1104, + "step": 25152 + }, + { + "epoch": 0.2183401185753596, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017278799342155301, + "loss": 0.0957, + "step": 25153 + }, + { + "epoch": 0.21834879905556376, + "grad_norm": 0.11767578125, + "learning_rate": 0.001727858610731902, + "loss": 0.1348, + "step": 25154 + }, + { + "epoch": 0.21835747953576792, + "grad_norm": 0.26953125, + "learning_rate": 0.0017278372865616569, + "loss": 0.0918, + "step": 25155 + }, + { + "epoch": 0.2183661600159721, + "grad_norm": 0.150390625, + "learning_rate": 0.0017278159617048184, + "loss": 0.0918, + "step": 25156 + }, + { + "epoch": 0.21837484049617625, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017277946361614102, + "loss": 0.1084, + "step": 25157 + }, + { + "epoch": 0.21838352097638042, + "grad_norm": 1.1640625, + "learning_rate": 0.001727773309931455, + "loss": 0.1387, + "step": 25158 + }, + { + "epoch": 0.21839220145658458, + "grad_norm": 0.72265625, + "learning_rate": 0.001727751983014976, + "loss": 0.0947, + "step": 25159 + }, + { + "epoch": 0.21840088193678875, + "grad_norm": 0.2451171875, + "learning_rate": 0.001727730655411997, + "loss": 0.0942, + "step": 25160 + }, + { + "epoch": 0.21840956241699291, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017277093271225415, + "loss": 0.0903, + "step": 25161 + }, + { + "epoch": 0.21841824289719708, + "grad_norm": 0.0849609375, + "learning_rate": 0.0017276879981466325, + "loss": 0.0986, + "step": 25162 + }, + { + "epoch": 0.21842692337740124, + "grad_norm": 0.109375, + "learning_rate": 0.001727666668484293, + "loss": 0.1113, + "step": 25163 + }, + { + "epoch": 0.2184356038576054, + "grad_norm": 0.83203125, + "learning_rate": 0.0017276453381355468, + "loss": 0.1084, + "step": 25164 + }, + { + "epoch": 0.21844428433780957, + "grad_norm": 0.423828125, + "learning_rate": 0.001727624007100417, + "loss": 0.0933, + "step": 25165 + }, + { + "epoch": 0.21845296481801374, + "grad_norm": 0.640625, + "learning_rate": 0.0017276026753789274, + "loss": 0.0977, + "step": 25166 + }, + { + "epoch": 0.2184616452982179, + "grad_norm": 0.12255859375, + "learning_rate": 0.0017275813429711005, + "loss": 0.0986, + "step": 25167 + }, + { + "epoch": 0.21847032577842207, + "grad_norm": 0.115234375, + "learning_rate": 0.00172756000987696, + "loss": 0.1094, + "step": 25168 + }, + { + "epoch": 0.21847900625862623, + "grad_norm": 0.322265625, + "learning_rate": 0.0017275386760965297, + "loss": 0.1123, + "step": 25169 + }, + { + "epoch": 0.2184876867388304, + "grad_norm": 0.341796875, + "learning_rate": 0.0017275173416298325, + "loss": 0.1025, + "step": 25170 + }, + { + "epoch": 0.21849636721903457, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017274960064768917, + "loss": 0.0933, + "step": 25171 + }, + { + "epoch": 0.21850504769923873, + "grad_norm": 0.11328125, + "learning_rate": 0.0017274746706377306, + "loss": 0.0962, + "step": 25172 + }, + { + "epoch": 0.2185137281794429, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017274533341123724, + "loss": 0.1123, + "step": 25173 + }, + { + "epoch": 0.21852240865964706, + "grad_norm": 0.486328125, + "learning_rate": 0.0017274319969008411, + "loss": 0.1182, + "step": 25174 + }, + { + "epoch": 0.21853108913985123, + "grad_norm": 0.12890625, + "learning_rate": 0.0017274106590031595, + "loss": 0.0747, + "step": 25175 + }, + { + "epoch": 0.2185397696200554, + "grad_norm": 0.408203125, + "learning_rate": 0.001727389320419351, + "loss": 0.1064, + "step": 25176 + }, + { + "epoch": 0.21854845010025956, + "grad_norm": 0.3359375, + "learning_rate": 0.0017273679811494387, + "loss": 0.1133, + "step": 25177 + }, + { + "epoch": 0.21855713058046372, + "grad_norm": 0.25390625, + "learning_rate": 0.0017273466411934465, + "loss": 0.1074, + "step": 25178 + }, + { + "epoch": 0.21856581106066789, + "grad_norm": 0.328125, + "learning_rate": 0.0017273253005513972, + "loss": 0.0752, + "step": 25179 + }, + { + "epoch": 0.21857449154087205, + "grad_norm": 0.78515625, + "learning_rate": 0.0017273039592233146, + "loss": 0.1279, + "step": 25180 + }, + { + "epoch": 0.21858317202107622, + "grad_norm": 0.82421875, + "learning_rate": 0.0017272826172092218, + "loss": 0.166, + "step": 25181 + }, + { + "epoch": 0.21859185250128038, + "grad_norm": 0.1923828125, + "learning_rate": 0.001727261274509142, + "loss": 0.1582, + "step": 25182 + }, + { + "epoch": 0.21860053298148455, + "grad_norm": 0.609375, + "learning_rate": 0.0017272399311230988, + "loss": 0.124, + "step": 25183 + }, + { + "epoch": 0.2186092134616887, + "grad_norm": 0.15625, + "learning_rate": 0.0017272185870511156, + "loss": 0.085, + "step": 25184 + }, + { + "epoch": 0.21861789394189288, + "grad_norm": 0.59765625, + "learning_rate": 0.0017271972422932154, + "loss": 0.0908, + "step": 25185 + }, + { + "epoch": 0.21862657442209704, + "grad_norm": 0.51953125, + "learning_rate": 0.0017271758968494216, + "loss": 0.1934, + "step": 25186 + }, + { + "epoch": 0.2186352549023012, + "grad_norm": 0.09619140625, + "learning_rate": 0.0017271545507197576, + "loss": 0.126, + "step": 25187 + }, + { + "epoch": 0.21864393538250537, + "grad_norm": 0.451171875, + "learning_rate": 0.0017271332039042473, + "loss": 0.0664, + "step": 25188 + }, + { + "epoch": 0.21865261586270954, + "grad_norm": 0.6640625, + "learning_rate": 0.0017271118564029131, + "loss": 0.0918, + "step": 25189 + }, + { + "epoch": 0.2186612963429137, + "grad_norm": 0.640625, + "learning_rate": 0.001727090508215779, + "loss": 0.0947, + "step": 25190 + }, + { + "epoch": 0.21866997682311787, + "grad_norm": 0.703125, + "learning_rate": 0.001727069159342868, + "loss": 0.1143, + "step": 25191 + }, + { + "epoch": 0.21867865730332203, + "grad_norm": 3.34375, + "learning_rate": 0.0017270478097842037, + "loss": 0.208, + "step": 25192 + }, + { + "epoch": 0.2186873377835262, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017270264595398096, + "loss": 0.0854, + "step": 25193 + }, + { + "epoch": 0.21869601826373036, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017270051086097082, + "loss": 0.1016, + "step": 25194 + }, + { + "epoch": 0.21870469874393453, + "grad_norm": 0.55078125, + "learning_rate": 0.001726983756993924, + "loss": 0.1289, + "step": 25195 + }, + { + "epoch": 0.2187133792241387, + "grad_norm": 0.416015625, + "learning_rate": 0.0017269624046924795, + "loss": 0.0874, + "step": 25196 + }, + { + "epoch": 0.21872205970434283, + "grad_norm": 0.1611328125, + "learning_rate": 0.001726941051705398, + "loss": 0.0879, + "step": 25197 + }, + { + "epoch": 0.218730740184547, + "grad_norm": 0.11572265625, + "learning_rate": 0.0017269196980327038, + "loss": 0.1162, + "step": 25198 + }, + { + "epoch": 0.21873942066475116, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017268983436744191, + "loss": 0.1104, + "step": 25199 + }, + { + "epoch": 0.21874810114495533, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017268769886305685, + "loss": 0.1279, + "step": 25200 + }, + { + "epoch": 0.2187567816251595, + "grad_norm": 0.91015625, + "learning_rate": 0.001726855632901174, + "loss": 0.1182, + "step": 25201 + }, + { + "epoch": 0.21876546210536366, + "grad_norm": 0.6015625, + "learning_rate": 0.0017268342764862597, + "loss": 0.1025, + "step": 25202 + }, + { + "epoch": 0.21877414258556782, + "grad_norm": 0.255859375, + "learning_rate": 0.001726812919385849, + "loss": 0.0771, + "step": 25203 + }, + { + "epoch": 0.218782823065772, + "grad_norm": 0.2578125, + "learning_rate": 0.001726791561599965, + "loss": 0.1318, + "step": 25204 + }, + { + "epoch": 0.21879150354597615, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017267702031286314, + "loss": 0.1484, + "step": 25205 + }, + { + "epoch": 0.21880018402618032, + "grad_norm": 0.236328125, + "learning_rate": 0.0017267488439718713, + "loss": 0.0986, + "step": 25206 + }, + { + "epoch": 0.21880886450638448, + "grad_norm": 0.08642578125, + "learning_rate": 0.0017267274841297076, + "loss": 0.1445, + "step": 25207 + }, + { + "epoch": 0.21881754498658865, + "grad_norm": 0.62109375, + "learning_rate": 0.0017267061236021647, + "loss": 0.1406, + "step": 25208 + }, + { + "epoch": 0.2188262254667928, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017266847623892652, + "loss": 0.1338, + "step": 25209 + }, + { + "epoch": 0.21883490594699698, + "grad_norm": 0.6640625, + "learning_rate": 0.0017266634004910327, + "loss": 0.1367, + "step": 25210 + }, + { + "epoch": 0.21884358642720114, + "grad_norm": 0.42578125, + "learning_rate": 0.0017266420379074903, + "loss": 0.1553, + "step": 25211 + }, + { + "epoch": 0.2188522669074053, + "grad_norm": 0.5703125, + "learning_rate": 0.0017266206746386617, + "loss": 0.127, + "step": 25212 + }, + { + "epoch": 0.21886094738760947, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017265993106845704, + "loss": 0.1582, + "step": 25213 + }, + { + "epoch": 0.21886962786781364, + "grad_norm": 1.328125, + "learning_rate": 0.001726577946045239, + "loss": 0.1123, + "step": 25214 + }, + { + "epoch": 0.2188783083480178, + "grad_norm": 0.123046875, + "learning_rate": 0.001726556580720692, + "loss": 0.0776, + "step": 25215 + }, + { + "epoch": 0.21888698882822197, + "grad_norm": 0.39453125, + "learning_rate": 0.0017265352147109515, + "loss": 0.083, + "step": 25216 + }, + { + "epoch": 0.21889566930842613, + "grad_norm": 0.1806640625, + "learning_rate": 0.0017265138480160418, + "loss": 0.0898, + "step": 25217 + }, + { + "epoch": 0.2189043497886303, + "grad_norm": 0.94921875, + "learning_rate": 0.0017264924806359863, + "loss": 0.0898, + "step": 25218 + }, + { + "epoch": 0.21891303026883446, + "grad_norm": 0.1328125, + "learning_rate": 0.0017264711125708073, + "loss": 0.1011, + "step": 25219 + }, + { + "epoch": 0.21892171074903863, + "grad_norm": 0.095703125, + "learning_rate": 0.0017264497438205294, + "loss": 0.1123, + "step": 25220 + }, + { + "epoch": 0.2189303912292428, + "grad_norm": 0.154296875, + "learning_rate": 0.0017264283743851753, + "loss": 0.0986, + "step": 25221 + }, + { + "epoch": 0.21893907170944696, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017264070042647688, + "loss": 0.1016, + "step": 25222 + }, + { + "epoch": 0.21894775218965112, + "grad_norm": 0.466796875, + "learning_rate": 0.001726385633459333, + "loss": 0.1104, + "step": 25223 + }, + { + "epoch": 0.2189564326698553, + "grad_norm": 0.37109375, + "learning_rate": 0.0017263642619688908, + "loss": 0.1226, + "step": 25224 + }, + { + "epoch": 0.21896511315005945, + "grad_norm": 0.216796875, + "learning_rate": 0.0017263428897934666, + "loss": 0.1445, + "step": 25225 + }, + { + "epoch": 0.21897379363026362, + "grad_norm": 0.09765625, + "learning_rate": 0.0017263215169330828, + "loss": 0.123, + "step": 25226 + }, + { + "epoch": 0.21898247411046778, + "grad_norm": 0.11328125, + "learning_rate": 0.0017263001433877637, + "loss": 0.1045, + "step": 25227 + }, + { + "epoch": 0.21899115459067195, + "grad_norm": 0.5546875, + "learning_rate": 0.0017262787691575319, + "loss": 0.125, + "step": 25228 + }, + { + "epoch": 0.21899983507087611, + "grad_norm": 0.140625, + "learning_rate": 0.0017262573942424108, + "loss": 0.0918, + "step": 25229 + }, + { + "epoch": 0.21900851555108028, + "grad_norm": 0.32421875, + "learning_rate": 0.0017262360186424243, + "loss": 0.0996, + "step": 25230 + }, + { + "epoch": 0.21901719603128444, + "grad_norm": 0.921875, + "learning_rate": 0.0017262146423575955, + "loss": 0.1621, + "step": 25231 + }, + { + "epoch": 0.2190258765114886, + "grad_norm": 0.51953125, + "learning_rate": 0.0017261932653879477, + "loss": 0.1328, + "step": 25232 + }, + { + "epoch": 0.21903455699169277, + "grad_norm": 0.97265625, + "learning_rate": 0.0017261718877335042, + "loss": 0.1348, + "step": 25233 + }, + { + "epoch": 0.21904323747189694, + "grad_norm": 0.224609375, + "learning_rate": 0.0017261505093942888, + "loss": 0.0874, + "step": 25234 + }, + { + "epoch": 0.2190519179521011, + "grad_norm": 0.298828125, + "learning_rate": 0.0017261291303703246, + "loss": 0.1143, + "step": 25235 + }, + { + "epoch": 0.21906059843230527, + "grad_norm": 0.2275390625, + "learning_rate": 0.001726107750661635, + "loss": 0.1514, + "step": 25236 + }, + { + "epoch": 0.21906927891250944, + "grad_norm": 0.158203125, + "learning_rate": 0.0017260863702682433, + "loss": 0.1279, + "step": 25237 + }, + { + "epoch": 0.2190779593927136, + "grad_norm": 0.162109375, + "learning_rate": 0.0017260649891901727, + "loss": 0.1045, + "step": 25238 + }, + { + "epoch": 0.21908663987291777, + "grad_norm": 0.11572265625, + "learning_rate": 0.0017260436074274471, + "loss": 0.0845, + "step": 25239 + }, + { + "epoch": 0.21909532035312193, + "grad_norm": 0.470703125, + "learning_rate": 0.0017260222249800897, + "loss": 0.1191, + "step": 25240 + }, + { + "epoch": 0.2191040008333261, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017260008418481237, + "loss": 0.0874, + "step": 25241 + }, + { + "epoch": 0.21911268131353026, + "grad_norm": 0.294921875, + "learning_rate": 0.0017259794580315724, + "loss": 0.1011, + "step": 25242 + }, + { + "epoch": 0.21912136179373443, + "grad_norm": 0.390625, + "learning_rate": 0.0017259580735304596, + "loss": 0.1133, + "step": 25243 + }, + { + "epoch": 0.2191300422739386, + "grad_norm": 0.404296875, + "learning_rate": 0.0017259366883448082, + "loss": 0.1006, + "step": 25244 + }, + { + "epoch": 0.21913872275414276, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017259153024746423, + "loss": 0.1235, + "step": 25245 + }, + { + "epoch": 0.21914740323434692, + "grad_norm": 0.376953125, + "learning_rate": 0.0017258939159199844, + "loss": 0.1201, + "step": 25246 + }, + { + "epoch": 0.2191560837145511, + "grad_norm": 0.375, + "learning_rate": 0.0017258725286808585, + "loss": 0.1465, + "step": 25247 + }, + { + "epoch": 0.21916476419475525, + "grad_norm": 0.287109375, + "learning_rate": 0.001725851140757288, + "loss": 0.0996, + "step": 25248 + }, + { + "epoch": 0.21917344467495942, + "grad_norm": 0.15625, + "learning_rate": 0.0017258297521492957, + "loss": 0.1133, + "step": 25249 + }, + { + "epoch": 0.21918212515516358, + "grad_norm": 0.1484375, + "learning_rate": 0.0017258083628569057, + "loss": 0.1069, + "step": 25250 + }, + { + "epoch": 0.21919080563536775, + "grad_norm": 0.7890625, + "learning_rate": 0.001725786972880141, + "loss": 0.1167, + "step": 25251 + }, + { + "epoch": 0.2191994861155719, + "grad_norm": 0.40625, + "learning_rate": 0.001725765582219025, + "loss": 0.0996, + "step": 25252 + }, + { + "epoch": 0.21920816659577608, + "grad_norm": 0.84765625, + "learning_rate": 0.0017257441908735812, + "loss": 0.1416, + "step": 25253 + }, + { + "epoch": 0.21921684707598024, + "grad_norm": 0.625, + "learning_rate": 0.001725722798843833, + "loss": 0.1143, + "step": 25254 + }, + { + "epoch": 0.2192255275561844, + "grad_norm": 0.1728515625, + "learning_rate": 0.0017257014061298032, + "loss": 0.0908, + "step": 25255 + }, + { + "epoch": 0.21923420803638857, + "grad_norm": 0.228515625, + "learning_rate": 0.0017256800127315164, + "loss": 0.1309, + "step": 25256 + }, + { + "epoch": 0.21924288851659274, + "grad_norm": 0.158203125, + "learning_rate": 0.0017256586186489952, + "loss": 0.0791, + "step": 25257 + }, + { + "epoch": 0.2192515689967969, + "grad_norm": 0.36328125, + "learning_rate": 0.0017256372238822631, + "loss": 0.0942, + "step": 25258 + }, + { + "epoch": 0.21926024947700107, + "grad_norm": 0.365234375, + "learning_rate": 0.0017256158284313435, + "loss": 0.1182, + "step": 25259 + }, + { + "epoch": 0.21926892995720523, + "grad_norm": 0.43359375, + "learning_rate": 0.00172559443229626, + "loss": 0.1206, + "step": 25260 + }, + { + "epoch": 0.2192776104374094, + "grad_norm": 0.275390625, + "learning_rate": 0.0017255730354770356, + "loss": 0.1113, + "step": 25261 + }, + { + "epoch": 0.21928629091761356, + "grad_norm": 0.41015625, + "learning_rate": 0.0017255516379736941, + "loss": 0.1094, + "step": 25262 + }, + { + "epoch": 0.21929497139781773, + "grad_norm": 0.08984375, + "learning_rate": 0.0017255302397862586, + "loss": 0.1123, + "step": 25263 + }, + { + "epoch": 0.2193036518780219, + "grad_norm": 0.380859375, + "learning_rate": 0.0017255088409147527, + "loss": 0.1133, + "step": 25264 + }, + { + "epoch": 0.21931233235822606, + "grad_norm": 0.859375, + "learning_rate": 0.0017254874413591998, + "loss": 0.0967, + "step": 25265 + }, + { + "epoch": 0.21932101283843022, + "grad_norm": 0.283203125, + "learning_rate": 0.0017254660411196233, + "loss": 0.083, + "step": 25266 + }, + { + "epoch": 0.2193296933186344, + "grad_norm": 0.2578125, + "learning_rate": 0.0017254446401960462, + "loss": 0.1001, + "step": 25267 + }, + { + "epoch": 0.21933837379883855, + "grad_norm": 1.046875, + "learning_rate": 0.0017254232385884923, + "loss": 0.2246, + "step": 25268 + }, + { + "epoch": 0.21934705427904272, + "grad_norm": 0.1845703125, + "learning_rate": 0.001725401836296985, + "loss": 0.1104, + "step": 25269 + }, + { + "epoch": 0.21935573475924688, + "grad_norm": 0.4921875, + "learning_rate": 0.0017253804333215478, + "loss": 0.105, + "step": 25270 + }, + { + "epoch": 0.21936441523945105, + "grad_norm": 0.09521484375, + "learning_rate": 0.0017253590296622042, + "loss": 0.1084, + "step": 25271 + }, + { + "epoch": 0.21937309571965521, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017253376253189765, + "loss": 0.1299, + "step": 25272 + }, + { + "epoch": 0.21938177619985938, + "grad_norm": 0.416015625, + "learning_rate": 0.0017253162202918897, + "loss": 0.0977, + "step": 25273 + }, + { + "epoch": 0.21939045668006354, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017252948145809663, + "loss": 0.1187, + "step": 25274 + }, + { + "epoch": 0.2193991371602677, + "grad_norm": 0.2373046875, + "learning_rate": 0.00172527340818623, + "loss": 0.1055, + "step": 25275 + }, + { + "epoch": 0.21940781764047187, + "grad_norm": 0.140625, + "learning_rate": 0.0017252520011077038, + "loss": 0.1133, + "step": 25276 + }, + { + "epoch": 0.21941649812067604, + "grad_norm": 0.5703125, + "learning_rate": 0.0017252305933454116, + "loss": 0.1001, + "step": 25277 + }, + { + "epoch": 0.2194251786008802, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017252091848993765, + "loss": 0.1318, + "step": 25278 + }, + { + "epoch": 0.21943385908108437, + "grad_norm": 0.26953125, + "learning_rate": 0.001725187775769622, + "loss": 0.1211, + "step": 25279 + }, + { + "epoch": 0.21944253956128854, + "grad_norm": 0.345703125, + "learning_rate": 0.0017251663659561718, + "loss": 0.1367, + "step": 25280 + }, + { + "epoch": 0.2194512200414927, + "grad_norm": 0.287109375, + "learning_rate": 0.0017251449554590488, + "loss": 0.0991, + "step": 25281 + }, + { + "epoch": 0.21945990052169687, + "grad_norm": 0.68359375, + "learning_rate": 0.0017251235442782766, + "loss": 0.1182, + "step": 25282 + }, + { + "epoch": 0.21946858100190103, + "grad_norm": 1.296875, + "learning_rate": 0.0017251021324138792, + "loss": 0.0908, + "step": 25283 + }, + { + "epoch": 0.2194772614821052, + "grad_norm": 0.490234375, + "learning_rate": 0.001725080719865879, + "loss": 0.168, + "step": 25284 + }, + { + "epoch": 0.21948594196230936, + "grad_norm": 0.30078125, + "learning_rate": 0.0017250593066343, + "loss": 0.1035, + "step": 25285 + }, + { + "epoch": 0.21949462244251353, + "grad_norm": 1.1015625, + "learning_rate": 0.0017250378927191657, + "loss": 0.1045, + "step": 25286 + }, + { + "epoch": 0.2195033029227177, + "grad_norm": 0.37890625, + "learning_rate": 0.0017250164781204991, + "loss": 0.0972, + "step": 25287 + }, + { + "epoch": 0.21951198340292186, + "grad_norm": 0.3125, + "learning_rate": 0.0017249950628383242, + "loss": 0.0986, + "step": 25288 + }, + { + "epoch": 0.21952066388312602, + "grad_norm": 0.08349609375, + "learning_rate": 0.0017249736468726638, + "loss": 0.0889, + "step": 25289 + }, + { + "epoch": 0.2195293443633302, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017249522302235417, + "loss": 0.123, + "step": 25290 + }, + { + "epoch": 0.21953802484353435, + "grad_norm": 0.34765625, + "learning_rate": 0.0017249308128909813, + "loss": 0.126, + "step": 25291 + }, + { + "epoch": 0.21954670532373852, + "grad_norm": 0.5390625, + "learning_rate": 0.001724909394875006, + "loss": 0.1699, + "step": 25292 + }, + { + "epoch": 0.21955538580394268, + "grad_norm": 0.1318359375, + "learning_rate": 0.001724887976175639, + "loss": 0.0918, + "step": 25293 + }, + { + "epoch": 0.21956406628414685, + "grad_norm": 0.10595703125, + "learning_rate": 0.001724866556792904, + "loss": 0.0718, + "step": 25294 + }, + { + "epoch": 0.219572746764351, + "grad_norm": 0.421875, + "learning_rate": 0.0017248451367268243, + "loss": 0.0991, + "step": 25295 + }, + { + "epoch": 0.21958142724455518, + "grad_norm": 0.181640625, + "learning_rate": 0.001724823715977423, + "loss": 0.1157, + "step": 25296 + }, + { + "epoch": 0.21959010772475934, + "grad_norm": 0.2265625, + "learning_rate": 0.0017248022945447247, + "loss": 0.0981, + "step": 25297 + }, + { + "epoch": 0.2195987882049635, + "grad_norm": 0.416015625, + "learning_rate": 0.0017247808724287512, + "loss": 0.0903, + "step": 25298 + }, + { + "epoch": 0.21960746868516767, + "grad_norm": 0.3828125, + "learning_rate": 0.0017247594496295272, + "loss": 0.0947, + "step": 25299 + }, + { + "epoch": 0.21961614916537184, + "grad_norm": 0.341796875, + "learning_rate": 0.0017247380261470754, + "loss": 0.1025, + "step": 25300 + }, + { + "epoch": 0.219624829645576, + "grad_norm": 0.337890625, + "learning_rate": 0.0017247166019814196, + "loss": 0.1152, + "step": 25301 + }, + { + "epoch": 0.21963351012578017, + "grad_norm": 0.22265625, + "learning_rate": 0.001724695177132583, + "loss": 0.0957, + "step": 25302 + }, + { + "epoch": 0.21964219060598433, + "grad_norm": 0.10595703125, + "learning_rate": 0.0017246737516005894, + "loss": 0.1426, + "step": 25303 + }, + { + "epoch": 0.2196508710861885, + "grad_norm": 0.13671875, + "learning_rate": 0.0017246523253854618, + "loss": 0.1084, + "step": 25304 + }, + { + "epoch": 0.21965955156639266, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017246308984872236, + "loss": 0.1084, + "step": 25305 + }, + { + "epoch": 0.21966823204659683, + "grad_norm": 0.376953125, + "learning_rate": 0.001724609470905899, + "loss": 0.0938, + "step": 25306 + }, + { + "epoch": 0.219676912526801, + "grad_norm": 0.2138671875, + "learning_rate": 0.00172458804264151, + "loss": 0.1475, + "step": 25307 + }, + { + "epoch": 0.21968559300700516, + "grad_norm": 0.341796875, + "learning_rate": 0.0017245666136940817, + "loss": 0.1543, + "step": 25308 + }, + { + "epoch": 0.21969427348720932, + "grad_norm": 0.51953125, + "learning_rate": 0.0017245451840636364, + "loss": 0.1016, + "step": 25309 + }, + { + "epoch": 0.2197029539674135, + "grad_norm": 0.302734375, + "learning_rate": 0.0017245237537501978, + "loss": 0.126, + "step": 25310 + }, + { + "epoch": 0.21971163444761765, + "grad_norm": 0.1708984375, + "learning_rate": 0.0017245023227537898, + "loss": 0.1035, + "step": 25311 + }, + { + "epoch": 0.21972031492782182, + "grad_norm": 1.015625, + "learning_rate": 0.001724480891074435, + "loss": 0.1143, + "step": 25312 + }, + { + "epoch": 0.21972899540802598, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017244594587121575, + "loss": 0.1016, + "step": 25313 + }, + { + "epoch": 0.21973767588823015, + "grad_norm": 0.80859375, + "learning_rate": 0.0017244380256669805, + "loss": 0.332, + "step": 25314 + }, + { + "epoch": 0.21974635636843431, + "grad_norm": 0.765625, + "learning_rate": 0.0017244165919389272, + "loss": 0.1064, + "step": 25315 + }, + { + "epoch": 0.21975503684863848, + "grad_norm": 0.271484375, + "learning_rate": 0.0017243951575280218, + "loss": 0.1191, + "step": 25316 + }, + { + "epoch": 0.21976371732884264, + "grad_norm": 0.322265625, + "learning_rate": 0.0017243737224342866, + "loss": 0.1055, + "step": 25317 + }, + { + "epoch": 0.2197723978090468, + "grad_norm": 0.421875, + "learning_rate": 0.0017243522866577463, + "loss": 0.084, + "step": 25318 + }, + { + "epoch": 0.21978107828925098, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017243308501984232, + "loss": 0.0923, + "step": 25319 + }, + { + "epoch": 0.2197897587694551, + "grad_norm": 0.263671875, + "learning_rate": 0.0017243094130563418, + "loss": 0.1001, + "step": 25320 + }, + { + "epoch": 0.21979843924965928, + "grad_norm": 0.19140625, + "learning_rate": 0.0017242879752315244, + "loss": 0.1309, + "step": 25321 + }, + { + "epoch": 0.21980711972986344, + "grad_norm": 0.671875, + "learning_rate": 0.0017242665367239954, + "loss": 0.1045, + "step": 25322 + }, + { + "epoch": 0.2198158002100676, + "grad_norm": 0.6171875, + "learning_rate": 0.001724245097533778, + "loss": 0.0903, + "step": 25323 + }, + { + "epoch": 0.21982448069027177, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017242236576608951, + "loss": 0.1172, + "step": 25324 + }, + { + "epoch": 0.21983316117047594, + "grad_norm": 0.5625, + "learning_rate": 0.001724202217105371, + "loss": 0.1328, + "step": 25325 + }, + { + "epoch": 0.2198418416506801, + "grad_norm": 0.130859375, + "learning_rate": 0.0017241807758672288, + "loss": 0.1123, + "step": 25326 + }, + { + "epoch": 0.21985052213088427, + "grad_norm": 0.24609375, + "learning_rate": 0.0017241593339464916, + "loss": 0.124, + "step": 25327 + }, + { + "epoch": 0.21985920261108843, + "grad_norm": 0.14453125, + "learning_rate": 0.001724137891343183, + "loss": 0.1064, + "step": 25328 + }, + { + "epoch": 0.2198678830912926, + "grad_norm": 0.490234375, + "learning_rate": 0.001724116448057327, + "loss": 0.0977, + "step": 25329 + }, + { + "epoch": 0.21987656357149676, + "grad_norm": 0.7109375, + "learning_rate": 0.0017240950040889468, + "loss": 0.168, + "step": 25330 + }, + { + "epoch": 0.21988524405170093, + "grad_norm": 0.81640625, + "learning_rate": 0.0017240735594380651, + "loss": 0.1025, + "step": 25331 + }, + { + "epoch": 0.2198939245319051, + "grad_norm": 0.6484375, + "learning_rate": 0.0017240521141047062, + "loss": 0.0938, + "step": 25332 + }, + { + "epoch": 0.21990260501210926, + "grad_norm": 0.228515625, + "learning_rate": 0.0017240306680888933, + "loss": 0.1113, + "step": 25333 + }, + { + "epoch": 0.21991128549231342, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017240092213906496, + "loss": 0.1104, + "step": 25334 + }, + { + "epoch": 0.2199199659725176, + "grad_norm": 0.298828125, + "learning_rate": 0.0017239877740099988, + "loss": 0.126, + "step": 25335 + }, + { + "epoch": 0.21992864645272175, + "grad_norm": 0.251953125, + "learning_rate": 0.001723966325946965, + "loss": 0.1162, + "step": 25336 + }, + { + "epoch": 0.21993732693292592, + "grad_norm": 0.220703125, + "learning_rate": 0.0017239448772015703, + "loss": 0.1328, + "step": 25337 + }, + { + "epoch": 0.21994600741313008, + "grad_norm": 0.58203125, + "learning_rate": 0.0017239234277738392, + "loss": 0.1152, + "step": 25338 + }, + { + "epoch": 0.21995468789333425, + "grad_norm": 0.365234375, + "learning_rate": 0.0017239019776637944, + "loss": 0.1201, + "step": 25339 + }, + { + "epoch": 0.21996336837353841, + "grad_norm": 0.35546875, + "learning_rate": 0.0017238805268714602, + "loss": 0.1523, + "step": 25340 + }, + { + "epoch": 0.21997204885374258, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017238590753968594, + "loss": 0.0962, + "step": 25341 + }, + { + "epoch": 0.21998072933394675, + "grad_norm": 0.271484375, + "learning_rate": 0.0017238376232400158, + "loss": 0.1162, + "step": 25342 + }, + { + "epoch": 0.2199894098141509, + "grad_norm": 0.20703125, + "learning_rate": 0.0017238161704009524, + "loss": 0.126, + "step": 25343 + }, + { + "epoch": 0.21999809029435508, + "grad_norm": 0.173828125, + "learning_rate": 0.0017237947168796935, + "loss": 0.0957, + "step": 25344 + }, + { + "epoch": 0.22000677077455924, + "grad_norm": 0.10595703125, + "learning_rate": 0.001723773262676262, + "loss": 0.1699, + "step": 25345 + }, + { + "epoch": 0.2200154512547634, + "grad_norm": 0.8515625, + "learning_rate": 0.0017237518077906813, + "loss": 0.1289, + "step": 25346 + }, + { + "epoch": 0.22002413173496757, + "grad_norm": 0.2578125, + "learning_rate": 0.001723730352222975, + "loss": 0.0938, + "step": 25347 + }, + { + "epoch": 0.22003281221517174, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017237088959731665, + "loss": 0.0977, + "step": 25348 + }, + { + "epoch": 0.2200414926953759, + "grad_norm": 0.484375, + "learning_rate": 0.0017236874390412797, + "loss": 0.0898, + "step": 25349 + }, + { + "epoch": 0.22005017317558007, + "grad_norm": 0.51953125, + "learning_rate": 0.0017236659814273373, + "loss": 0.1187, + "step": 25350 + }, + { + "epoch": 0.22005885365578423, + "grad_norm": 0.6171875, + "learning_rate": 0.0017236445231313632, + "loss": 0.126, + "step": 25351 + }, + { + "epoch": 0.2200675341359884, + "grad_norm": 0.119140625, + "learning_rate": 0.001723623064153381, + "loss": 0.1084, + "step": 25352 + }, + { + "epoch": 0.22007621461619256, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017236016044934138, + "loss": 0.1113, + "step": 25353 + }, + { + "epoch": 0.22008489509639673, + "grad_norm": 0.0810546875, + "learning_rate": 0.0017235801441514854, + "loss": 0.0791, + "step": 25354 + }, + { + "epoch": 0.2200935755766009, + "grad_norm": 0.2236328125, + "learning_rate": 0.001723558683127619, + "loss": 0.1235, + "step": 25355 + }, + { + "epoch": 0.22010225605680506, + "grad_norm": 0.36328125, + "learning_rate": 0.0017235372214218384, + "loss": 0.1055, + "step": 25356 + }, + { + "epoch": 0.22011093653700922, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017235157590341664, + "loss": 0.1357, + "step": 25357 + }, + { + "epoch": 0.2201196170172134, + "grad_norm": 0.2275390625, + "learning_rate": 0.001723494295964627, + "loss": 0.1348, + "step": 25358 + }, + { + "epoch": 0.22012829749741755, + "grad_norm": 0.443359375, + "learning_rate": 0.001723472832213244, + "loss": 0.1035, + "step": 25359 + }, + { + "epoch": 0.22013697797762172, + "grad_norm": 0.4609375, + "learning_rate": 0.0017234513677800406, + "loss": 0.1416, + "step": 25360 + }, + { + "epoch": 0.22014565845782588, + "grad_norm": 0.79296875, + "learning_rate": 0.0017234299026650398, + "loss": 0.1172, + "step": 25361 + }, + { + "epoch": 0.22015433893803005, + "grad_norm": 0.09228515625, + "learning_rate": 0.0017234084368682654, + "loss": 0.1094, + "step": 25362 + }, + { + "epoch": 0.2201630194182342, + "grad_norm": 0.71484375, + "learning_rate": 0.001723386970389741, + "loss": 0.1348, + "step": 25363 + }, + { + "epoch": 0.22017169989843838, + "grad_norm": 0.0859375, + "learning_rate": 0.00172336550322949, + "loss": 0.0952, + "step": 25364 + }, + { + "epoch": 0.22018038037864254, + "grad_norm": 0.478515625, + "learning_rate": 0.001723344035387536, + "loss": 0.1055, + "step": 25365 + }, + { + "epoch": 0.2201890608588467, + "grad_norm": 0.37109375, + "learning_rate": 0.001723322566863902, + "loss": 0.1074, + "step": 25366 + }, + { + "epoch": 0.22019774133905087, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017233010976586119, + "loss": 0.1338, + "step": 25367 + }, + { + "epoch": 0.22020642181925504, + "grad_norm": 0.0986328125, + "learning_rate": 0.001723279627771689, + "loss": 0.1226, + "step": 25368 + }, + { + "epoch": 0.2202151022994592, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017232581572031568, + "loss": 0.1357, + "step": 25369 + }, + { + "epoch": 0.22022378277966337, + "grad_norm": 0.1123046875, + "learning_rate": 0.0017232366859530395, + "loss": 0.103, + "step": 25370 + }, + { + "epoch": 0.22023246325986753, + "grad_norm": 0.453125, + "learning_rate": 0.0017232152140213592, + "loss": 0.1084, + "step": 25371 + }, + { + "epoch": 0.2202411437400717, + "grad_norm": 0.07080078125, + "learning_rate": 0.0017231937414081404, + "loss": 0.1582, + "step": 25372 + }, + { + "epoch": 0.22024982422027586, + "grad_norm": 0.349609375, + "learning_rate": 0.001723172268113406, + "loss": 0.1211, + "step": 25373 + }, + { + "epoch": 0.22025850470048003, + "grad_norm": 0.275390625, + "learning_rate": 0.0017231507941371802, + "loss": 0.1074, + "step": 25374 + }, + { + "epoch": 0.2202671851806842, + "grad_norm": 0.439453125, + "learning_rate": 0.001723129319479486, + "loss": 0.0923, + "step": 25375 + }, + { + "epoch": 0.22027586566088836, + "grad_norm": 0.2001953125, + "learning_rate": 0.0017231078441403468, + "loss": 0.1641, + "step": 25376 + }, + { + "epoch": 0.22028454614109252, + "grad_norm": 0.64453125, + "learning_rate": 0.0017230863681197864, + "loss": 0.1318, + "step": 25377 + }, + { + "epoch": 0.2202932266212967, + "grad_norm": 0.1376953125, + "learning_rate": 0.001723064891417828, + "loss": 0.085, + "step": 25378 + }, + { + "epoch": 0.22030190710150085, + "grad_norm": 0.1767578125, + "learning_rate": 0.001723043414034495, + "loss": 0.104, + "step": 25379 + }, + { + "epoch": 0.22031058758170502, + "grad_norm": 0.33203125, + "learning_rate": 0.0017230219359698111, + "loss": 0.1523, + "step": 25380 + }, + { + "epoch": 0.22031926806190918, + "grad_norm": 0.1044921875, + "learning_rate": 0.0017230004572237998, + "loss": 0.0967, + "step": 25381 + }, + { + "epoch": 0.22032794854211335, + "grad_norm": 0.33203125, + "learning_rate": 0.0017229789777964848, + "loss": 0.0986, + "step": 25382 + }, + { + "epoch": 0.22033662902231752, + "grad_norm": 0.58203125, + "learning_rate": 0.0017229574976878892, + "loss": 0.1357, + "step": 25383 + }, + { + "epoch": 0.22034530950252168, + "grad_norm": 0.130859375, + "learning_rate": 0.0017229360168980368, + "loss": 0.0698, + "step": 25384 + }, + { + "epoch": 0.22035398998272585, + "grad_norm": 0.203125, + "learning_rate": 0.0017229145354269507, + "loss": 0.1289, + "step": 25385 + }, + { + "epoch": 0.22036267046293, + "grad_norm": 0.2421875, + "learning_rate": 0.0017228930532746547, + "loss": 0.1118, + "step": 25386 + }, + { + "epoch": 0.22037135094313418, + "grad_norm": 0.0771484375, + "learning_rate": 0.0017228715704411721, + "loss": 0.1079, + "step": 25387 + }, + { + "epoch": 0.22038003142333834, + "grad_norm": 0.1318359375, + "learning_rate": 0.001722850086926527, + "loss": 0.125, + "step": 25388 + }, + { + "epoch": 0.2203887119035425, + "grad_norm": 0.44140625, + "learning_rate": 0.0017228286027307417, + "loss": 0.0786, + "step": 25389 + }, + { + "epoch": 0.22039739238374667, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017228071178538409, + "loss": 0.0869, + "step": 25390 + }, + { + "epoch": 0.22040607286395084, + "grad_norm": 0.55078125, + "learning_rate": 0.0017227856322958473, + "loss": 0.0977, + "step": 25391 + }, + { + "epoch": 0.220414753344155, + "grad_norm": 0.291015625, + "learning_rate": 0.001722764146056785, + "loss": 0.1562, + "step": 25392 + }, + { + "epoch": 0.22042343382435917, + "grad_norm": 0.19140625, + "learning_rate": 0.0017227426591366769, + "loss": 0.1001, + "step": 25393 + }, + { + "epoch": 0.22043211430456333, + "grad_norm": 0.79296875, + "learning_rate": 0.001722721171535547, + "loss": 0.127, + "step": 25394 + }, + { + "epoch": 0.2204407947847675, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017226996832534185, + "loss": 0.1035, + "step": 25395 + }, + { + "epoch": 0.22044947526497166, + "grad_norm": 0.244140625, + "learning_rate": 0.001722678194290315, + "loss": 0.1094, + "step": 25396 + }, + { + "epoch": 0.22045815574517583, + "grad_norm": 0.44921875, + "learning_rate": 0.0017226567046462599, + "loss": 0.1279, + "step": 25397 + }, + { + "epoch": 0.22046683622538, + "grad_norm": 0.11767578125, + "learning_rate": 0.0017226352143212766, + "loss": 0.1133, + "step": 25398 + }, + { + "epoch": 0.22047551670558416, + "grad_norm": 0.3515625, + "learning_rate": 0.0017226137233153891, + "loss": 0.0894, + "step": 25399 + }, + { + "epoch": 0.22048419718578832, + "grad_norm": 0.30859375, + "learning_rate": 0.0017225922316286206, + "loss": 0.124, + "step": 25400 + }, + { + "epoch": 0.2204928776659925, + "grad_norm": 0.11669921875, + "learning_rate": 0.0017225707392609943, + "loss": 0.0908, + "step": 25401 + }, + { + "epoch": 0.22050155814619665, + "grad_norm": 0.205078125, + "learning_rate": 0.0017225492462125344, + "loss": 0.1328, + "step": 25402 + }, + { + "epoch": 0.22051023862640082, + "grad_norm": 0.3671875, + "learning_rate": 0.0017225277524832636, + "loss": 0.1426, + "step": 25403 + }, + { + "epoch": 0.22051891910660498, + "grad_norm": 0.1162109375, + "learning_rate": 0.001722506258073206, + "loss": 0.1006, + "step": 25404 + }, + { + "epoch": 0.22052759958680915, + "grad_norm": 0.26953125, + "learning_rate": 0.001722484762982385, + "loss": 0.1279, + "step": 25405 + }, + { + "epoch": 0.2205362800670133, + "grad_norm": 0.1611328125, + "learning_rate": 0.0017224632672108238, + "loss": 0.0928, + "step": 25406 + }, + { + "epoch": 0.22054496054721748, + "grad_norm": 1.2265625, + "learning_rate": 0.0017224417707585463, + "loss": 0.1484, + "step": 25407 + }, + { + "epoch": 0.22055364102742164, + "grad_norm": 0.322265625, + "learning_rate": 0.0017224202736255759, + "loss": 0.0977, + "step": 25408 + }, + { + "epoch": 0.2205623215076258, + "grad_norm": 0.36328125, + "learning_rate": 0.001722398775811936, + "loss": 0.123, + "step": 25409 + }, + { + "epoch": 0.22057100198782997, + "grad_norm": 0.63671875, + "learning_rate": 0.0017223772773176497, + "loss": 0.127, + "step": 25410 + }, + { + "epoch": 0.22057968246803414, + "grad_norm": 0.314453125, + "learning_rate": 0.0017223557781427415, + "loss": 0.1367, + "step": 25411 + }, + { + "epoch": 0.2205883629482383, + "grad_norm": 0.30078125, + "learning_rate": 0.001722334278287234, + "loss": 0.0923, + "step": 25412 + }, + { + "epoch": 0.22059704342844247, + "grad_norm": 0.498046875, + "learning_rate": 0.0017223127777511514, + "loss": 0.1113, + "step": 25413 + }, + { + "epoch": 0.22060572390864663, + "grad_norm": 0.765625, + "learning_rate": 0.001722291276534517, + "loss": 0.1055, + "step": 25414 + }, + { + "epoch": 0.2206144043888508, + "grad_norm": 0.28515625, + "learning_rate": 0.0017222697746373536, + "loss": 0.0879, + "step": 25415 + }, + { + "epoch": 0.22062308486905496, + "grad_norm": 0.70703125, + "learning_rate": 0.0017222482720596858, + "loss": 0.1079, + "step": 25416 + }, + { + "epoch": 0.22063176534925913, + "grad_norm": 0.5234375, + "learning_rate": 0.0017222267688015364, + "loss": 0.1348, + "step": 25417 + }, + { + "epoch": 0.2206404458294633, + "grad_norm": 0.177734375, + "learning_rate": 0.0017222052648629294, + "loss": 0.1025, + "step": 25418 + }, + { + "epoch": 0.22064912630966746, + "grad_norm": 0.2177734375, + "learning_rate": 0.001722183760243888, + "loss": 0.0918, + "step": 25419 + }, + { + "epoch": 0.22065780678987162, + "grad_norm": 0.431640625, + "learning_rate": 0.0017221622549444357, + "loss": 0.0898, + "step": 25420 + }, + { + "epoch": 0.2206664872700758, + "grad_norm": 0.75390625, + "learning_rate": 0.0017221407489645961, + "loss": 0.0908, + "step": 25421 + }, + { + "epoch": 0.22067516775027995, + "grad_norm": 0.330078125, + "learning_rate": 0.001722119242304393, + "loss": 0.0898, + "step": 25422 + }, + { + "epoch": 0.22068384823048412, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017220977349638495, + "loss": 0.0601, + "step": 25423 + }, + { + "epoch": 0.22069252871068828, + "grad_norm": 0.2578125, + "learning_rate": 0.001722076226942989, + "loss": 0.1035, + "step": 25424 + }, + { + "epoch": 0.22070120919089245, + "grad_norm": 0.455078125, + "learning_rate": 0.0017220547182418356, + "loss": 0.1348, + "step": 25425 + }, + { + "epoch": 0.22070988967109662, + "grad_norm": 0.400390625, + "learning_rate": 0.0017220332088604122, + "loss": 0.165, + "step": 25426 + }, + { + "epoch": 0.22071857015130078, + "grad_norm": 1.0234375, + "learning_rate": 0.0017220116987987431, + "loss": 0.1348, + "step": 25427 + }, + { + "epoch": 0.22072725063150495, + "grad_norm": 0.9140625, + "learning_rate": 0.0017219901880568508, + "loss": 0.0918, + "step": 25428 + }, + { + "epoch": 0.2207359311117091, + "grad_norm": 0.26171875, + "learning_rate": 0.0017219686766347597, + "loss": 0.1396, + "step": 25429 + }, + { + "epoch": 0.22074461159191328, + "grad_norm": 0.1103515625, + "learning_rate": 0.001721947164532493, + "loss": 0.1162, + "step": 25430 + }, + { + "epoch": 0.22075329207211744, + "grad_norm": 0.77734375, + "learning_rate": 0.001721925651750074, + "loss": 0.1201, + "step": 25431 + }, + { + "epoch": 0.2207619725523216, + "grad_norm": 0.6875, + "learning_rate": 0.0017219041382875264, + "loss": 0.1465, + "step": 25432 + }, + { + "epoch": 0.22077065303252577, + "grad_norm": 0.1640625, + "learning_rate": 0.0017218826241448743, + "loss": 0.1465, + "step": 25433 + }, + { + "epoch": 0.22077933351272994, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017218611093221406, + "loss": 0.1016, + "step": 25434 + }, + { + "epoch": 0.2207880139929341, + "grad_norm": 0.16015625, + "learning_rate": 0.0017218395938193485, + "loss": 0.0879, + "step": 25435 + }, + { + "epoch": 0.22079669447313827, + "grad_norm": 0.150390625, + "learning_rate": 0.0017218180776365222, + "loss": 0.1245, + "step": 25436 + }, + { + "epoch": 0.22080537495334243, + "grad_norm": 0.50390625, + "learning_rate": 0.0017217965607736851, + "loss": 0.1055, + "step": 25437 + }, + { + "epoch": 0.2208140554335466, + "grad_norm": 0.66796875, + "learning_rate": 0.0017217750432308603, + "loss": 0.1309, + "step": 25438 + }, + { + "epoch": 0.22082273591375076, + "grad_norm": 0.2197265625, + "learning_rate": 0.001721753525008072, + "loss": 0.0972, + "step": 25439 + }, + { + "epoch": 0.22083141639395493, + "grad_norm": 0.08203125, + "learning_rate": 0.0017217320061053434, + "loss": 0.0898, + "step": 25440 + }, + { + "epoch": 0.2208400968741591, + "grad_norm": 0.37890625, + "learning_rate": 0.0017217104865226976, + "loss": 0.1328, + "step": 25441 + }, + { + "epoch": 0.22084877735436326, + "grad_norm": 0.314453125, + "learning_rate": 0.001721688966260159, + "loss": 0.1025, + "step": 25442 + }, + { + "epoch": 0.2208574578345674, + "grad_norm": 0.0703125, + "learning_rate": 0.0017216674453177506, + "loss": 0.0903, + "step": 25443 + }, + { + "epoch": 0.22086613831477156, + "grad_norm": 0.2216796875, + "learning_rate": 0.001721645923695496, + "loss": 0.0854, + "step": 25444 + }, + { + "epoch": 0.22087481879497572, + "grad_norm": 0.19921875, + "learning_rate": 0.0017216244013934187, + "loss": 0.125, + "step": 25445 + }, + { + "epoch": 0.2208834992751799, + "grad_norm": 0.09912109375, + "learning_rate": 0.001721602878411542, + "loss": 0.0796, + "step": 25446 + }, + { + "epoch": 0.22089217975538405, + "grad_norm": 0.5, + "learning_rate": 0.00172158135474989, + "loss": 0.0967, + "step": 25447 + }, + { + "epoch": 0.22090086023558822, + "grad_norm": 0.1435546875, + "learning_rate": 0.001721559830408486, + "loss": 0.1455, + "step": 25448 + }, + { + "epoch": 0.22090954071579239, + "grad_norm": 0.328125, + "learning_rate": 0.0017215383053873535, + "loss": 0.1221, + "step": 25449 + }, + { + "epoch": 0.22091822119599655, + "grad_norm": 0.609375, + "learning_rate": 0.001721516779686516, + "loss": 0.1582, + "step": 25450 + }, + { + "epoch": 0.22092690167620072, + "grad_norm": 0.146484375, + "learning_rate": 0.0017214952533059972, + "loss": 0.1133, + "step": 25451 + }, + { + "epoch": 0.22093558215640488, + "grad_norm": 0.3203125, + "learning_rate": 0.0017214737262458207, + "loss": 0.0977, + "step": 25452 + }, + { + "epoch": 0.22094426263660905, + "grad_norm": 0.58203125, + "learning_rate": 0.0017214521985060096, + "loss": 0.0723, + "step": 25453 + }, + { + "epoch": 0.2209529431168132, + "grad_norm": 0.15234375, + "learning_rate": 0.0017214306700865877, + "loss": 0.085, + "step": 25454 + }, + { + "epoch": 0.22096162359701738, + "grad_norm": 0.365234375, + "learning_rate": 0.0017214091409875785, + "loss": 0.123, + "step": 25455 + }, + { + "epoch": 0.22097030407722154, + "grad_norm": 0.07421875, + "learning_rate": 0.0017213876112090057, + "loss": 0.0918, + "step": 25456 + }, + { + "epoch": 0.2209789845574257, + "grad_norm": 0.25390625, + "learning_rate": 0.001721366080750893, + "loss": 0.1045, + "step": 25457 + }, + { + "epoch": 0.22098766503762987, + "grad_norm": 0.0869140625, + "learning_rate": 0.0017213445496132632, + "loss": 0.0889, + "step": 25458 + }, + { + "epoch": 0.22099634551783404, + "grad_norm": 0.12060546875, + "learning_rate": 0.0017213230177961407, + "loss": 0.1006, + "step": 25459 + }, + { + "epoch": 0.2210050259980382, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017213014852995485, + "loss": 0.0986, + "step": 25460 + }, + { + "epoch": 0.22101370647824237, + "grad_norm": 0.416015625, + "learning_rate": 0.0017212799521235103, + "loss": 0.1543, + "step": 25461 + }, + { + "epoch": 0.22102238695844653, + "grad_norm": 0.287109375, + "learning_rate": 0.00172125841826805, + "loss": 0.1846, + "step": 25462 + }, + { + "epoch": 0.2210310674386507, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017212368837331905, + "loss": 0.0903, + "step": 25463 + }, + { + "epoch": 0.22103974791885486, + "grad_norm": 0.4921875, + "learning_rate": 0.0017212153485189556, + "loss": 0.1055, + "step": 25464 + }, + { + "epoch": 0.22104842839905903, + "grad_norm": 0.0849609375, + "learning_rate": 0.0017211938126253693, + "loss": 0.0986, + "step": 25465 + }, + { + "epoch": 0.2210571088792632, + "grad_norm": 0.11181640625, + "learning_rate": 0.0017211722760524543, + "loss": 0.1172, + "step": 25466 + }, + { + "epoch": 0.22106578935946736, + "grad_norm": 0.103515625, + "learning_rate": 0.0017211507388002352, + "loss": 0.0947, + "step": 25467 + }, + { + "epoch": 0.22107446983967152, + "grad_norm": 0.478515625, + "learning_rate": 0.0017211292008687345, + "loss": 0.0767, + "step": 25468 + }, + { + "epoch": 0.2210831503198757, + "grad_norm": 0.259765625, + "learning_rate": 0.0017211076622579765, + "loss": 0.0928, + "step": 25469 + }, + { + "epoch": 0.22109183080007985, + "grad_norm": 1.171875, + "learning_rate": 0.0017210861229679845, + "loss": 0.1006, + "step": 25470 + }, + { + "epoch": 0.22110051128028402, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017210645829987819, + "loss": 0.1128, + "step": 25471 + }, + { + "epoch": 0.22110919176048818, + "grad_norm": 0.11328125, + "learning_rate": 0.0017210430423503925, + "loss": 0.126, + "step": 25472 + }, + { + "epoch": 0.22111787224069235, + "grad_norm": 0.11474609375, + "learning_rate": 0.0017210215010228398, + "loss": 0.0938, + "step": 25473 + }, + { + "epoch": 0.2211265527208965, + "grad_norm": 0.208984375, + "learning_rate": 0.0017209999590161473, + "loss": 0.127, + "step": 25474 + }, + { + "epoch": 0.22113523320110068, + "grad_norm": 0.3046875, + "learning_rate": 0.0017209784163303384, + "loss": 0.1299, + "step": 25475 + }, + { + "epoch": 0.22114391368130484, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017209568729654367, + "loss": 0.0938, + "step": 25476 + }, + { + "epoch": 0.221152594161509, + "grad_norm": 0.93359375, + "learning_rate": 0.0017209353289214665, + "loss": 0.1035, + "step": 25477 + }, + { + "epoch": 0.22116127464171317, + "grad_norm": 0.83203125, + "learning_rate": 0.0017209137841984504, + "loss": 0.1895, + "step": 25478 + }, + { + "epoch": 0.22116995512191734, + "grad_norm": 0.240234375, + "learning_rate": 0.0017208922387964122, + "loss": 0.0986, + "step": 25479 + }, + { + "epoch": 0.2211786356021215, + "grad_norm": 0.734375, + "learning_rate": 0.001720870692715376, + "loss": 0.1426, + "step": 25480 + }, + { + "epoch": 0.22118731608232567, + "grad_norm": 0.5390625, + "learning_rate": 0.0017208491459553643, + "loss": 0.1221, + "step": 25481 + }, + { + "epoch": 0.22119599656252983, + "grad_norm": 0.7734375, + "learning_rate": 0.0017208275985164019, + "loss": 0.1182, + "step": 25482 + }, + { + "epoch": 0.221204677042734, + "grad_norm": 0.234375, + "learning_rate": 0.0017208060503985114, + "loss": 0.0928, + "step": 25483 + }, + { + "epoch": 0.22121335752293816, + "grad_norm": 0.20703125, + "learning_rate": 0.0017207845016017167, + "loss": 0.1074, + "step": 25484 + }, + { + "epoch": 0.22122203800314233, + "grad_norm": 0.208984375, + "learning_rate": 0.0017207629521260417, + "loss": 0.1025, + "step": 25485 + }, + { + "epoch": 0.2212307184833465, + "grad_norm": 0.51171875, + "learning_rate": 0.0017207414019715095, + "loss": 0.1172, + "step": 25486 + }, + { + "epoch": 0.22123939896355066, + "grad_norm": 0.12353515625, + "learning_rate": 0.0017207198511381438, + "loss": 0.1006, + "step": 25487 + }, + { + "epoch": 0.22124807944375482, + "grad_norm": 0.3046875, + "learning_rate": 0.0017206982996259682, + "loss": 0.1387, + "step": 25488 + }, + { + "epoch": 0.221256759923959, + "grad_norm": 0.23828125, + "learning_rate": 0.0017206767474350062, + "loss": 0.0801, + "step": 25489 + }, + { + "epoch": 0.22126544040416316, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017206551945652816, + "loss": 0.0908, + "step": 25490 + }, + { + "epoch": 0.22127412088436732, + "grad_norm": 0.09814453125, + "learning_rate": 0.001720633641016818, + "loss": 0.0981, + "step": 25491 + }, + { + "epoch": 0.22128280136457149, + "grad_norm": 0.2734375, + "learning_rate": 0.0017206120867896385, + "loss": 0.1113, + "step": 25492 + }, + { + "epoch": 0.22129148184477565, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017205905318837668, + "loss": 0.0801, + "step": 25493 + }, + { + "epoch": 0.22130016232497982, + "grad_norm": 0.19921875, + "learning_rate": 0.0017205689762992268, + "loss": 0.166, + "step": 25494 + }, + { + "epoch": 0.22130884280518398, + "grad_norm": 0.443359375, + "learning_rate": 0.0017205474200360418, + "loss": 0.1094, + "step": 25495 + }, + { + "epoch": 0.22131752328538815, + "grad_norm": 0.1787109375, + "learning_rate": 0.0017205258630942357, + "loss": 0.1299, + "step": 25496 + }, + { + "epoch": 0.2213262037655923, + "grad_norm": 0.412109375, + "learning_rate": 0.0017205043054738316, + "loss": 0.0957, + "step": 25497 + }, + { + "epoch": 0.22133488424579648, + "grad_norm": 0.11572265625, + "learning_rate": 0.0017204827471748534, + "loss": 0.1133, + "step": 25498 + }, + { + "epoch": 0.22134356472600064, + "grad_norm": 0.71875, + "learning_rate": 0.0017204611881973246, + "loss": 0.1206, + "step": 25499 + }, + { + "epoch": 0.2213522452062048, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017204396285412688, + "loss": 0.1367, + "step": 25500 + }, + { + "epoch": 0.22136092568640897, + "grad_norm": 0.87890625, + "learning_rate": 0.0017204180682067092, + "loss": 0.1133, + "step": 25501 + }, + { + "epoch": 0.22136960616661314, + "grad_norm": 0.40625, + "learning_rate": 0.0017203965071936702, + "loss": 0.1328, + "step": 25502 + }, + { + "epoch": 0.2213782866468173, + "grad_norm": 0.333984375, + "learning_rate": 0.0017203749455021746, + "loss": 0.0898, + "step": 25503 + }, + { + "epoch": 0.22138696712702147, + "grad_norm": 0.330078125, + "learning_rate": 0.0017203533831322464, + "loss": 0.1035, + "step": 25504 + }, + { + "epoch": 0.22139564760722563, + "grad_norm": 0.796875, + "learning_rate": 0.001720331820083909, + "loss": 0.1396, + "step": 25505 + }, + { + "epoch": 0.2214043280874298, + "grad_norm": 0.1767578125, + "learning_rate": 0.0017203102563571862, + "loss": 0.0869, + "step": 25506 + }, + { + "epoch": 0.22141300856763396, + "grad_norm": 0.0830078125, + "learning_rate": 0.0017202886919521012, + "loss": 0.0894, + "step": 25507 + }, + { + "epoch": 0.22142168904783813, + "grad_norm": 0.185546875, + "learning_rate": 0.001720267126868678, + "loss": 0.0972, + "step": 25508 + }, + { + "epoch": 0.2214303695280423, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017202455611069398, + "loss": 0.1157, + "step": 25509 + }, + { + "epoch": 0.22143905000824646, + "grad_norm": 0.08544921875, + "learning_rate": 0.0017202239946669106, + "loss": 0.0732, + "step": 25510 + }, + { + "epoch": 0.22144773048845062, + "grad_norm": 0.0859375, + "learning_rate": 0.0017202024275486136, + "loss": 0.0947, + "step": 25511 + }, + { + "epoch": 0.2214564109686548, + "grad_norm": 0.5078125, + "learning_rate": 0.0017201808597520725, + "loss": 0.1504, + "step": 25512 + }, + { + "epoch": 0.22146509144885895, + "grad_norm": 0.248046875, + "learning_rate": 0.0017201592912773113, + "loss": 0.0864, + "step": 25513 + }, + { + "epoch": 0.22147377192906312, + "grad_norm": 0.10546875, + "learning_rate": 0.0017201377221243526, + "loss": 0.0771, + "step": 25514 + }, + { + "epoch": 0.22148245240926728, + "grad_norm": 0.248046875, + "learning_rate": 0.0017201161522932208, + "loss": 0.1074, + "step": 25515 + }, + { + "epoch": 0.22149113288947145, + "grad_norm": 0.291015625, + "learning_rate": 0.00172009458178394, + "loss": 0.1162, + "step": 25516 + }, + { + "epoch": 0.2214998133696756, + "grad_norm": 0.447265625, + "learning_rate": 0.0017200730105965322, + "loss": 0.1396, + "step": 25517 + }, + { + "epoch": 0.22150849384987978, + "grad_norm": 0.34375, + "learning_rate": 0.0017200514387310224, + "loss": 0.1104, + "step": 25518 + }, + { + "epoch": 0.22151717433008394, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017200298661874335, + "loss": 0.1074, + "step": 25519 + }, + { + "epoch": 0.2215258548102881, + "grad_norm": 0.236328125, + "learning_rate": 0.0017200082929657893, + "loss": 0.1152, + "step": 25520 + }, + { + "epoch": 0.22153453529049227, + "grad_norm": 0.1748046875, + "learning_rate": 0.0017199867190661132, + "loss": 0.106, + "step": 25521 + }, + { + "epoch": 0.22154321577069644, + "grad_norm": 0.119140625, + "learning_rate": 0.0017199651444884288, + "loss": 0.0811, + "step": 25522 + }, + { + "epoch": 0.2215518962509006, + "grad_norm": 0.12158203125, + "learning_rate": 0.0017199435692327601, + "loss": 0.1152, + "step": 25523 + }, + { + "epoch": 0.22156057673110477, + "grad_norm": 0.3046875, + "learning_rate": 0.0017199219932991304, + "loss": 0.1191, + "step": 25524 + }, + { + "epoch": 0.22156925721130893, + "grad_norm": 0.2197265625, + "learning_rate": 0.001719900416687563, + "loss": 0.1621, + "step": 25525 + }, + { + "epoch": 0.2215779376915131, + "grad_norm": 1.265625, + "learning_rate": 0.0017198788393980823, + "loss": 0.1982, + "step": 25526 + }, + { + "epoch": 0.22158661817171726, + "grad_norm": 0.1484375, + "learning_rate": 0.001719857261430711, + "loss": 0.1133, + "step": 25527 + }, + { + "epoch": 0.22159529865192143, + "grad_norm": 0.71484375, + "learning_rate": 0.0017198356827854734, + "loss": 0.0752, + "step": 25528 + }, + { + "epoch": 0.2216039791321256, + "grad_norm": 0.60546875, + "learning_rate": 0.0017198141034623928, + "loss": 0.1064, + "step": 25529 + }, + { + "epoch": 0.22161265961232976, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017197925234614928, + "loss": 0.123, + "step": 25530 + }, + { + "epoch": 0.22162134009253392, + "grad_norm": 1.4453125, + "learning_rate": 0.001719770942782797, + "loss": 0.1816, + "step": 25531 + }, + { + "epoch": 0.2216300205727381, + "grad_norm": 0.53125, + "learning_rate": 0.0017197493614263286, + "loss": 0.0977, + "step": 25532 + }, + { + "epoch": 0.22163870105294226, + "grad_norm": 0.419921875, + "learning_rate": 0.001719727779392112, + "loss": 0.1709, + "step": 25533 + }, + { + "epoch": 0.22164738153314642, + "grad_norm": 0.373046875, + "learning_rate": 0.0017197061966801705, + "loss": 0.1201, + "step": 25534 + }, + { + "epoch": 0.22165606201335059, + "grad_norm": 0.150390625, + "learning_rate": 0.0017196846132905277, + "loss": 0.1182, + "step": 25535 + }, + { + "epoch": 0.22166474249355475, + "grad_norm": 0.16015625, + "learning_rate": 0.0017196630292232067, + "loss": 0.1035, + "step": 25536 + }, + { + "epoch": 0.22167342297375892, + "grad_norm": 0.1201171875, + "learning_rate": 0.0017196414444782318, + "loss": 0.1133, + "step": 25537 + }, + { + "epoch": 0.22168210345396308, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017196198590556265, + "loss": 0.0791, + "step": 25538 + }, + { + "epoch": 0.22169078393416725, + "grad_norm": 0.384765625, + "learning_rate": 0.0017195982729554138, + "loss": 0.0928, + "step": 25539 + }, + { + "epoch": 0.2216994644143714, + "grad_norm": 0.73046875, + "learning_rate": 0.0017195766861776178, + "loss": 0.1279, + "step": 25540 + }, + { + "epoch": 0.22170814489457558, + "grad_norm": 1.40625, + "learning_rate": 0.0017195550987222621, + "loss": 0.1465, + "step": 25541 + }, + { + "epoch": 0.22171682537477974, + "grad_norm": 0.267578125, + "learning_rate": 0.0017195335105893703, + "loss": 0.1016, + "step": 25542 + }, + { + "epoch": 0.2217255058549839, + "grad_norm": 0.09814453125, + "learning_rate": 0.001719511921778966, + "loss": 0.0957, + "step": 25543 + }, + { + "epoch": 0.22173418633518807, + "grad_norm": 0.640625, + "learning_rate": 0.0017194903322910726, + "loss": 0.1426, + "step": 25544 + }, + { + "epoch": 0.22174286681539224, + "grad_norm": 0.1640625, + "learning_rate": 0.0017194687421257144, + "loss": 0.1045, + "step": 25545 + }, + { + "epoch": 0.2217515472955964, + "grad_norm": 0.34765625, + "learning_rate": 0.001719447151282914, + "loss": 0.1416, + "step": 25546 + }, + { + "epoch": 0.22176022777580057, + "grad_norm": 0.419921875, + "learning_rate": 0.0017194255597626957, + "loss": 0.1104, + "step": 25547 + }, + { + "epoch": 0.22176890825600473, + "grad_norm": 0.306640625, + "learning_rate": 0.001719403967565083, + "loss": 0.123, + "step": 25548 + }, + { + "epoch": 0.2217775887362089, + "grad_norm": 0.515625, + "learning_rate": 0.0017193823746900992, + "loss": 0.1089, + "step": 25549 + }, + { + "epoch": 0.22178626921641306, + "grad_norm": 0.349609375, + "learning_rate": 0.001719360781137768, + "loss": 0.1143, + "step": 25550 + }, + { + "epoch": 0.22179494969661723, + "grad_norm": 0.1474609375, + "learning_rate": 0.0017193391869081134, + "loss": 0.1191, + "step": 25551 + }, + { + "epoch": 0.2218036301768214, + "grad_norm": 0.2314453125, + "learning_rate": 0.0017193175920011588, + "loss": 0.1211, + "step": 25552 + }, + { + "epoch": 0.22181231065702556, + "grad_norm": 0.349609375, + "learning_rate": 0.0017192959964169277, + "loss": 0.1064, + "step": 25553 + }, + { + "epoch": 0.22182099113722972, + "grad_norm": 0.12451171875, + "learning_rate": 0.001719274400155444, + "loss": 0.1147, + "step": 25554 + }, + { + "epoch": 0.2218296716174339, + "grad_norm": 0.73828125, + "learning_rate": 0.0017192528032167309, + "loss": 0.0854, + "step": 25555 + }, + { + "epoch": 0.22183835209763805, + "grad_norm": 0.67578125, + "learning_rate": 0.0017192312056008123, + "loss": 0.0981, + "step": 25556 + }, + { + "epoch": 0.22184703257784222, + "grad_norm": 0.447265625, + "learning_rate": 0.0017192096073077115, + "loss": 0.127, + "step": 25557 + }, + { + "epoch": 0.22185571305804638, + "grad_norm": 0.4765625, + "learning_rate": 0.0017191880083374529, + "loss": 0.1299, + "step": 25558 + }, + { + "epoch": 0.22186439353825055, + "grad_norm": 0.361328125, + "learning_rate": 0.0017191664086900593, + "loss": 0.1279, + "step": 25559 + }, + { + "epoch": 0.2218730740184547, + "grad_norm": 0.1064453125, + "learning_rate": 0.0017191448083655548, + "loss": 0.1289, + "step": 25560 + }, + { + "epoch": 0.22188175449865888, + "grad_norm": 0.138671875, + "learning_rate": 0.0017191232073639625, + "loss": 0.0806, + "step": 25561 + }, + { + "epoch": 0.22189043497886304, + "grad_norm": 0.20703125, + "learning_rate": 0.0017191016056853064, + "loss": 0.1167, + "step": 25562 + }, + { + "epoch": 0.2218991154590672, + "grad_norm": 0.302734375, + "learning_rate": 0.0017190800033296102, + "loss": 0.0879, + "step": 25563 + }, + { + "epoch": 0.22190779593927137, + "grad_norm": 0.80859375, + "learning_rate": 0.0017190584002968977, + "loss": 0.2031, + "step": 25564 + }, + { + "epoch": 0.22191647641947554, + "grad_norm": 0.482421875, + "learning_rate": 0.0017190367965871918, + "loss": 0.1367, + "step": 25565 + }, + { + "epoch": 0.22192515689967968, + "grad_norm": 0.271484375, + "learning_rate": 0.0017190151922005165, + "loss": 0.0869, + "step": 25566 + }, + { + "epoch": 0.22193383737988384, + "grad_norm": 0.5859375, + "learning_rate": 0.001718993587136896, + "loss": 0.1143, + "step": 25567 + }, + { + "epoch": 0.221942517860088, + "grad_norm": 0.953125, + "learning_rate": 0.0017189719813963534, + "loss": 0.0986, + "step": 25568 + }, + { + "epoch": 0.22195119834029217, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017189503749789118, + "loss": 0.0938, + "step": 25569 + }, + { + "epoch": 0.22195987882049634, + "grad_norm": 0.8046875, + "learning_rate": 0.0017189287678845955, + "loss": 0.1309, + "step": 25570 + }, + { + "epoch": 0.2219685593007005, + "grad_norm": 0.310546875, + "learning_rate": 0.0017189071601134284, + "loss": 0.1338, + "step": 25571 + }, + { + "epoch": 0.22197723978090467, + "grad_norm": 0.3828125, + "learning_rate": 0.0017188855516654335, + "loss": 0.1006, + "step": 25572 + }, + { + "epoch": 0.22198592026110883, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017188639425406345, + "loss": 0.0889, + "step": 25573 + }, + { + "epoch": 0.221994600741313, + "grad_norm": 0.2421875, + "learning_rate": 0.0017188423327390554, + "loss": 0.0811, + "step": 25574 + }, + { + "epoch": 0.22200328122151716, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017188207222607194, + "loss": 0.1396, + "step": 25575 + }, + { + "epoch": 0.22201196170172133, + "grad_norm": 0.52734375, + "learning_rate": 0.0017187991111056507, + "loss": 0.1455, + "step": 25576 + }, + { + "epoch": 0.2220206421819255, + "grad_norm": 0.162109375, + "learning_rate": 0.0017187774992738724, + "loss": 0.0991, + "step": 25577 + }, + { + "epoch": 0.22202932266212966, + "grad_norm": 0.154296875, + "learning_rate": 0.0017187558867654083, + "loss": 0.126, + "step": 25578 + }, + { + "epoch": 0.22203800314233382, + "grad_norm": 0.154296875, + "learning_rate": 0.001718734273580282, + "loss": 0.1436, + "step": 25579 + }, + { + "epoch": 0.222046683622538, + "grad_norm": 0.1328125, + "learning_rate": 0.0017187126597185173, + "loss": 0.0952, + "step": 25580 + }, + { + "epoch": 0.22205536410274215, + "grad_norm": 0.451171875, + "learning_rate": 0.0017186910451801377, + "loss": 0.0928, + "step": 25581 + }, + { + "epoch": 0.22206404458294632, + "grad_norm": 0.484375, + "learning_rate": 0.001718669429965167, + "loss": 0.1074, + "step": 25582 + }, + { + "epoch": 0.22207272506315048, + "grad_norm": 0.205078125, + "learning_rate": 0.0017186478140736284, + "loss": 0.0903, + "step": 25583 + }, + { + "epoch": 0.22208140554335465, + "grad_norm": 0.9375, + "learning_rate": 0.001718626197505546, + "loss": 0.106, + "step": 25584 + }, + { + "epoch": 0.2220900860235588, + "grad_norm": 0.90234375, + "learning_rate": 0.0017186045802609431, + "loss": 0.1328, + "step": 25585 + }, + { + "epoch": 0.22209876650376298, + "grad_norm": 0.224609375, + "learning_rate": 0.0017185829623398438, + "loss": 0.1416, + "step": 25586 + }, + { + "epoch": 0.22210744698396714, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017185613437422714, + "loss": 0.0996, + "step": 25587 + }, + { + "epoch": 0.2221161274641713, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017185397244682498, + "loss": 0.125, + "step": 25588 + }, + { + "epoch": 0.22212480794437547, + "grad_norm": 0.41796875, + "learning_rate": 0.0017185181045178022, + "loss": 0.0938, + "step": 25589 + }, + { + "epoch": 0.22213348842457964, + "grad_norm": 0.48046875, + "learning_rate": 0.0017184964838909524, + "loss": 0.105, + "step": 25590 + }, + { + "epoch": 0.2221421689047838, + "grad_norm": 0.439453125, + "learning_rate": 0.0017184748625877246, + "loss": 0.125, + "step": 25591 + }, + { + "epoch": 0.22215084938498797, + "grad_norm": 0.466796875, + "learning_rate": 0.0017184532406081415, + "loss": 0.1216, + "step": 25592 + }, + { + "epoch": 0.22215952986519213, + "grad_norm": 0.310546875, + "learning_rate": 0.0017184316179522272, + "loss": 0.1191, + "step": 25593 + }, + { + "epoch": 0.2221682103453963, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017184099946200059, + "loss": 0.0918, + "step": 25594 + }, + { + "epoch": 0.22217689082560046, + "grad_norm": 0.0869140625, + "learning_rate": 0.0017183883706115004, + "loss": 0.1206, + "step": 25595 + }, + { + "epoch": 0.22218557130580463, + "grad_norm": 0.44921875, + "learning_rate": 0.0017183667459267343, + "loss": 0.0898, + "step": 25596 + }, + { + "epoch": 0.2221942517860088, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017183451205657321, + "loss": 0.0972, + "step": 25597 + }, + { + "epoch": 0.22220293226621296, + "grad_norm": 0.1875, + "learning_rate": 0.0017183234945285168, + "loss": 0.0801, + "step": 25598 + }, + { + "epoch": 0.22221161274641713, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017183018678151125, + "loss": 0.1191, + "step": 25599 + }, + { + "epoch": 0.2222202932266213, + "grad_norm": 0.369140625, + "learning_rate": 0.0017182802404255425, + "loss": 0.1182, + "step": 25600 + }, + { + "epoch": 0.22222897370682546, + "grad_norm": 0.48046875, + "learning_rate": 0.00171825861235983, + "loss": 0.1128, + "step": 25601 + }, + { + "epoch": 0.22223765418702962, + "grad_norm": 0.1318359375, + "learning_rate": 0.0017182369836179997, + "loss": 0.1523, + "step": 25602 + }, + { + "epoch": 0.22224633466723379, + "grad_norm": 0.9296875, + "learning_rate": 0.0017182153542000744, + "loss": 0.1084, + "step": 25603 + }, + { + "epoch": 0.22225501514743795, + "grad_norm": 0.1318359375, + "learning_rate": 0.0017181937241060783, + "loss": 0.2246, + "step": 25604 + }, + { + "epoch": 0.22226369562764212, + "grad_norm": 0.373046875, + "learning_rate": 0.0017181720933360347, + "loss": 0.0815, + "step": 25605 + }, + { + "epoch": 0.22227237610784628, + "grad_norm": 0.263671875, + "learning_rate": 0.0017181504618899676, + "loss": 0.0957, + "step": 25606 + }, + { + "epoch": 0.22228105658805045, + "grad_norm": 0.392578125, + "learning_rate": 0.0017181288297679002, + "loss": 0.1123, + "step": 25607 + }, + { + "epoch": 0.2222897370682546, + "grad_norm": 0.30078125, + "learning_rate": 0.0017181071969698564, + "loss": 0.1035, + "step": 25608 + }, + { + "epoch": 0.22229841754845878, + "grad_norm": 0.08251953125, + "learning_rate": 0.00171808556349586, + "loss": 0.1279, + "step": 25609 + }, + { + "epoch": 0.22230709802866294, + "grad_norm": 0.34765625, + "learning_rate": 0.0017180639293459346, + "loss": 0.0977, + "step": 25610 + }, + { + "epoch": 0.2223157785088671, + "grad_norm": 0.11669921875, + "learning_rate": 0.0017180422945201035, + "loss": 0.1094, + "step": 25611 + }, + { + "epoch": 0.22232445898907127, + "grad_norm": 0.4921875, + "learning_rate": 0.0017180206590183906, + "loss": 0.1011, + "step": 25612 + }, + { + "epoch": 0.22233313946927544, + "grad_norm": 0.59375, + "learning_rate": 0.00171799902284082, + "loss": 0.125, + "step": 25613 + }, + { + "epoch": 0.2223418199494796, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017179773859874146, + "loss": 0.1123, + "step": 25614 + }, + { + "epoch": 0.22235050042968377, + "grad_norm": 0.29296875, + "learning_rate": 0.0017179557484581986, + "loss": 0.1279, + "step": 25615 + }, + { + "epoch": 0.22235918090988793, + "grad_norm": 0.41015625, + "learning_rate": 0.0017179341102531954, + "loss": 0.0723, + "step": 25616 + }, + { + "epoch": 0.2223678613900921, + "grad_norm": 0.130859375, + "learning_rate": 0.0017179124713724285, + "loss": 0.124, + "step": 25617 + }, + { + "epoch": 0.22237654187029626, + "grad_norm": 0.486328125, + "learning_rate": 0.001717890831815922, + "loss": 0.1245, + "step": 25618 + }, + { + "epoch": 0.22238522235050043, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017178691915836995, + "loss": 0.0879, + "step": 25619 + }, + { + "epoch": 0.2223939028307046, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017178475506757847, + "loss": 0.1084, + "step": 25620 + }, + { + "epoch": 0.22240258331090876, + "grad_norm": 0.294921875, + "learning_rate": 0.0017178259090922007, + "loss": 0.0674, + "step": 25621 + }, + { + "epoch": 0.22241126379111292, + "grad_norm": 0.466796875, + "learning_rate": 0.0017178042668329716, + "loss": 0.0776, + "step": 25622 + }, + { + "epoch": 0.2224199442713171, + "grad_norm": 1.125, + "learning_rate": 0.0017177826238981213, + "loss": 0.1143, + "step": 25623 + }, + { + "epoch": 0.22242862475152125, + "grad_norm": 0.267578125, + "learning_rate": 0.0017177609802876731, + "loss": 0.1543, + "step": 25624 + }, + { + "epoch": 0.22243730523172542, + "grad_norm": 0.12109375, + "learning_rate": 0.0017177393360016506, + "loss": 0.1396, + "step": 25625 + }, + { + "epoch": 0.22244598571192958, + "grad_norm": 0.236328125, + "learning_rate": 0.001717717691040078, + "loss": 0.1011, + "step": 25626 + }, + { + "epoch": 0.22245466619213375, + "grad_norm": 0.51171875, + "learning_rate": 0.0017176960454029781, + "loss": 0.0928, + "step": 25627 + }, + { + "epoch": 0.2224633466723379, + "grad_norm": 0.69921875, + "learning_rate": 0.0017176743990903755, + "loss": 0.105, + "step": 25628 + }, + { + "epoch": 0.22247202715254208, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017176527521022935, + "loss": 0.1157, + "step": 25629 + }, + { + "epoch": 0.22248070763274624, + "grad_norm": 0.12158203125, + "learning_rate": 0.0017176311044387557, + "loss": 0.0898, + "step": 25630 + }, + { + "epoch": 0.2224893881129504, + "grad_norm": 0.1201171875, + "learning_rate": 0.0017176094560997853, + "loss": 0.1152, + "step": 25631 + }, + { + "epoch": 0.22249806859315457, + "grad_norm": 0.306640625, + "learning_rate": 0.001717587807085407, + "loss": 0.1025, + "step": 25632 + }, + { + "epoch": 0.22250674907335874, + "grad_norm": 0.10791015625, + "learning_rate": 0.0017175661573956438, + "loss": 0.0854, + "step": 25633 + }, + { + "epoch": 0.2225154295535629, + "grad_norm": 0.333984375, + "learning_rate": 0.0017175445070305199, + "loss": 0.1001, + "step": 25634 + }, + { + "epoch": 0.22252411003376707, + "grad_norm": 0.10302734375, + "learning_rate": 0.0017175228559900577, + "loss": 0.1689, + "step": 25635 + }, + { + "epoch": 0.22253279051397123, + "grad_norm": 0.41796875, + "learning_rate": 0.0017175012042742826, + "loss": 0.0659, + "step": 25636 + }, + { + "epoch": 0.2225414709941754, + "grad_norm": 0.2216796875, + "learning_rate": 0.0017174795518832171, + "loss": 0.0977, + "step": 25637 + }, + { + "epoch": 0.22255015147437957, + "grad_norm": 0.08935546875, + "learning_rate": 0.0017174578988168852, + "loss": 0.0918, + "step": 25638 + }, + { + "epoch": 0.22255883195458373, + "grad_norm": 0.298828125, + "learning_rate": 0.001717436245075311, + "loss": 0.0806, + "step": 25639 + }, + { + "epoch": 0.2225675124347879, + "grad_norm": 0.373046875, + "learning_rate": 0.0017174145906585175, + "loss": 0.1094, + "step": 25640 + }, + { + "epoch": 0.22257619291499206, + "grad_norm": 0.70703125, + "learning_rate": 0.0017173929355665286, + "loss": 0.1113, + "step": 25641 + }, + { + "epoch": 0.22258487339519623, + "grad_norm": 0.296875, + "learning_rate": 0.0017173712797993684, + "loss": 0.1514, + "step": 25642 + }, + { + "epoch": 0.2225935538754004, + "grad_norm": 0.126953125, + "learning_rate": 0.00171734962335706, + "loss": 0.166, + "step": 25643 + }, + { + "epoch": 0.22260223435560456, + "grad_norm": 0.53515625, + "learning_rate": 0.001717327966239627, + "loss": 0.0845, + "step": 25644 + }, + { + "epoch": 0.22261091483580872, + "grad_norm": 0.267578125, + "learning_rate": 0.001717306308447094, + "loss": 0.1758, + "step": 25645 + }, + { + "epoch": 0.22261959531601289, + "grad_norm": 0.453125, + "learning_rate": 0.0017172846499794838, + "loss": 0.0884, + "step": 25646 + }, + { + "epoch": 0.22262827579621705, + "grad_norm": 0.68359375, + "learning_rate": 0.0017172629908368206, + "loss": 0.0845, + "step": 25647 + }, + { + "epoch": 0.22263695627642122, + "grad_norm": 0.890625, + "learning_rate": 0.0017172413310191275, + "loss": 0.123, + "step": 25648 + }, + { + "epoch": 0.22264563675662538, + "grad_norm": 0.6484375, + "learning_rate": 0.0017172196705264287, + "loss": 0.1074, + "step": 25649 + }, + { + "epoch": 0.22265431723682955, + "grad_norm": 0.28125, + "learning_rate": 0.001717198009358748, + "loss": 0.0859, + "step": 25650 + }, + { + "epoch": 0.2226629977170337, + "grad_norm": 0.11328125, + "learning_rate": 0.0017171763475161087, + "loss": 0.123, + "step": 25651 + }, + { + "epoch": 0.22267167819723788, + "grad_norm": 1.3828125, + "learning_rate": 0.0017171546849985345, + "loss": 0.1367, + "step": 25652 + }, + { + "epoch": 0.22268035867744204, + "grad_norm": 0.6953125, + "learning_rate": 0.0017171330218060492, + "loss": 0.0776, + "step": 25653 + }, + { + "epoch": 0.2226890391576462, + "grad_norm": 0.400390625, + "learning_rate": 0.0017171113579386766, + "loss": 0.0806, + "step": 25654 + }, + { + "epoch": 0.22269771963785037, + "grad_norm": 0.1376953125, + "learning_rate": 0.00171708969339644, + "loss": 0.1182, + "step": 25655 + }, + { + "epoch": 0.22270640011805454, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017170680281793638, + "loss": 0.1084, + "step": 25656 + }, + { + "epoch": 0.2227150805982587, + "grad_norm": 0.322265625, + "learning_rate": 0.0017170463622874711, + "loss": 0.1084, + "step": 25657 + }, + { + "epoch": 0.22272376107846287, + "grad_norm": 0.0693359375, + "learning_rate": 0.0017170246957207858, + "loss": 0.0835, + "step": 25658 + }, + { + "epoch": 0.22273244155866703, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017170030284793317, + "loss": 0.106, + "step": 25659 + }, + { + "epoch": 0.2227411220388712, + "grad_norm": 0.30078125, + "learning_rate": 0.001716981360563132, + "loss": 0.0918, + "step": 25660 + }, + { + "epoch": 0.22274980251907536, + "grad_norm": 0.1318359375, + "learning_rate": 0.001716959691972211, + "loss": 0.1235, + "step": 25661 + }, + { + "epoch": 0.22275848299927953, + "grad_norm": 0.09521484375, + "learning_rate": 0.0017169380227065924, + "loss": 0.105, + "step": 25662 + }, + { + "epoch": 0.2227671634794837, + "grad_norm": 0.470703125, + "learning_rate": 0.0017169163527662992, + "loss": 0.1211, + "step": 25663 + }, + { + "epoch": 0.22277584395968786, + "grad_norm": 0.28515625, + "learning_rate": 0.0017168946821513556, + "loss": 0.1045, + "step": 25664 + }, + { + "epoch": 0.22278452443989202, + "grad_norm": 0.3515625, + "learning_rate": 0.0017168730108617855, + "loss": 0.085, + "step": 25665 + }, + { + "epoch": 0.2227932049200962, + "grad_norm": 0.322265625, + "learning_rate": 0.0017168513388976122, + "loss": 0.1719, + "step": 25666 + }, + { + "epoch": 0.22280188540030035, + "grad_norm": 0.2734375, + "learning_rate": 0.0017168296662588595, + "loss": 0.1045, + "step": 25667 + }, + { + "epoch": 0.22281056588050452, + "grad_norm": 0.20703125, + "learning_rate": 0.0017168079929455514, + "loss": 0.0864, + "step": 25668 + }, + { + "epoch": 0.22281924636070868, + "grad_norm": 0.162109375, + "learning_rate": 0.001716786318957711, + "loss": 0.1445, + "step": 25669 + }, + { + "epoch": 0.22282792684091285, + "grad_norm": 0.79296875, + "learning_rate": 0.0017167646442953625, + "loss": 0.1699, + "step": 25670 + }, + { + "epoch": 0.22283660732111701, + "grad_norm": 0.1142578125, + "learning_rate": 0.0017167429689585291, + "loss": 0.0996, + "step": 25671 + }, + { + "epoch": 0.22284528780132118, + "grad_norm": 0.3359375, + "learning_rate": 0.0017167212929472353, + "loss": 0.1216, + "step": 25672 + }, + { + "epoch": 0.22285396828152534, + "grad_norm": 0.8515625, + "learning_rate": 0.0017166996162615044, + "loss": 0.1133, + "step": 25673 + }, + { + "epoch": 0.2228626487617295, + "grad_norm": 0.54296875, + "learning_rate": 0.0017166779389013599, + "loss": 0.1348, + "step": 25674 + }, + { + "epoch": 0.22287132924193367, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017166562608668258, + "loss": 0.125, + "step": 25675 + }, + { + "epoch": 0.22288000972213784, + "grad_norm": 0.169921875, + "learning_rate": 0.0017166345821579255, + "loss": 0.1436, + "step": 25676 + }, + { + "epoch": 0.222888690202342, + "grad_norm": 0.248046875, + "learning_rate": 0.001716612902774683, + "loss": 0.0938, + "step": 25677 + }, + { + "epoch": 0.22289737068254617, + "grad_norm": 0.283203125, + "learning_rate": 0.0017165912227171216, + "loss": 0.1006, + "step": 25678 + }, + { + "epoch": 0.22290605116275033, + "grad_norm": 0.5546875, + "learning_rate": 0.0017165695419852655, + "loss": 0.0977, + "step": 25679 + }, + { + "epoch": 0.2229147316429545, + "grad_norm": 0.515625, + "learning_rate": 0.0017165478605791382, + "loss": 0.0796, + "step": 25680 + }, + { + "epoch": 0.22292341212315867, + "grad_norm": 0.435546875, + "learning_rate": 0.0017165261784987636, + "loss": 0.0981, + "step": 25681 + }, + { + "epoch": 0.22293209260336283, + "grad_norm": 0.52734375, + "learning_rate": 0.0017165044957441653, + "loss": 0.1133, + "step": 25682 + }, + { + "epoch": 0.222940773083567, + "grad_norm": 0.48828125, + "learning_rate": 0.0017164828123153665, + "loss": 0.1279, + "step": 25683 + }, + { + "epoch": 0.22294945356377116, + "grad_norm": 0.31640625, + "learning_rate": 0.0017164611282123917, + "loss": 0.0986, + "step": 25684 + }, + { + "epoch": 0.22295813404397533, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017164394434352641, + "loss": 0.1025, + "step": 25685 + }, + { + "epoch": 0.2229668145241795, + "grad_norm": 0.484375, + "learning_rate": 0.0017164177579840078, + "loss": 0.0972, + "step": 25686 + }, + { + "epoch": 0.22297549500438366, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017163960718586462, + "loss": 0.1182, + "step": 25687 + }, + { + "epoch": 0.22298417548458782, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017163743850592032, + "loss": 0.1064, + "step": 25688 + }, + { + "epoch": 0.222992855964792, + "grad_norm": 0.34375, + "learning_rate": 0.0017163526975857023, + "loss": 0.082, + "step": 25689 + }, + { + "epoch": 0.22300153644499612, + "grad_norm": 0.6953125, + "learning_rate": 0.0017163310094381675, + "loss": 0.0933, + "step": 25690 + }, + { + "epoch": 0.2230102169252003, + "grad_norm": 0.25390625, + "learning_rate": 0.001716309320616622, + "loss": 0.1011, + "step": 25691 + }, + { + "epoch": 0.22301889740540445, + "grad_norm": 0.439453125, + "learning_rate": 0.0017162876311210902, + "loss": 0.1025, + "step": 25692 + }, + { + "epoch": 0.22302757788560862, + "grad_norm": 0.10400390625, + "learning_rate": 0.0017162659409515955, + "loss": 0.125, + "step": 25693 + }, + { + "epoch": 0.22303625836581278, + "grad_norm": 0.349609375, + "learning_rate": 0.0017162442501081616, + "loss": 0.1367, + "step": 25694 + }, + { + "epoch": 0.22304493884601695, + "grad_norm": 0.79296875, + "learning_rate": 0.0017162225585908122, + "loss": 0.105, + "step": 25695 + }, + { + "epoch": 0.22305361932622111, + "grad_norm": 0.07763671875, + "learning_rate": 0.001716200866399571, + "loss": 0.105, + "step": 25696 + }, + { + "epoch": 0.22306229980642528, + "grad_norm": 0.8671875, + "learning_rate": 0.0017161791735344622, + "loss": 0.1006, + "step": 25697 + }, + { + "epoch": 0.22307098028662944, + "grad_norm": 1.0390625, + "learning_rate": 0.0017161574799955087, + "loss": 0.1777, + "step": 25698 + }, + { + "epoch": 0.2230796607668336, + "grad_norm": 0.255859375, + "learning_rate": 0.0017161357857827346, + "loss": 0.1055, + "step": 25699 + }, + { + "epoch": 0.22308834124703777, + "grad_norm": 0.068359375, + "learning_rate": 0.0017161140908961636, + "loss": 0.0874, + "step": 25700 + }, + { + "epoch": 0.22309702172724194, + "grad_norm": 0.1875, + "learning_rate": 0.0017160923953358199, + "loss": 0.1016, + "step": 25701 + }, + { + "epoch": 0.2231057022074461, + "grad_norm": 0.37890625, + "learning_rate": 0.0017160706991017265, + "loss": 0.1387, + "step": 25702 + }, + { + "epoch": 0.22311438268765027, + "grad_norm": 0.146484375, + "learning_rate": 0.0017160490021939077, + "loss": 0.1235, + "step": 25703 + }, + { + "epoch": 0.22312306316785444, + "grad_norm": 0.2578125, + "learning_rate": 0.0017160273046123867, + "loss": 0.0991, + "step": 25704 + }, + { + "epoch": 0.2231317436480586, + "grad_norm": 0.125, + "learning_rate": 0.0017160056063571874, + "loss": 0.1406, + "step": 25705 + }, + { + "epoch": 0.22314042412826277, + "grad_norm": 0.21484375, + "learning_rate": 0.0017159839074283338, + "loss": 0.123, + "step": 25706 + }, + { + "epoch": 0.22314910460846693, + "grad_norm": 0.54296875, + "learning_rate": 0.0017159622078258498, + "loss": 0.0845, + "step": 25707 + }, + { + "epoch": 0.2231577850886711, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017159405075497583, + "loss": 0.0918, + "step": 25708 + }, + { + "epoch": 0.22316646556887526, + "grad_norm": 0.17578125, + "learning_rate": 0.0017159188066000837, + "loss": 0.123, + "step": 25709 + }, + { + "epoch": 0.22317514604907943, + "grad_norm": 0.173828125, + "learning_rate": 0.0017158971049768493, + "loss": 0.0972, + "step": 25710 + }, + { + "epoch": 0.2231838265292836, + "grad_norm": 0.44140625, + "learning_rate": 0.0017158754026800796, + "loss": 0.1465, + "step": 25711 + }, + { + "epoch": 0.22319250700948776, + "grad_norm": 0.1328125, + "learning_rate": 0.0017158536997097975, + "loss": 0.0732, + "step": 25712 + }, + { + "epoch": 0.22320118748969192, + "grad_norm": 0.10009765625, + "learning_rate": 0.0017158319960660269, + "loss": 0.0918, + "step": 25713 + }, + { + "epoch": 0.2232098679698961, + "grad_norm": 0.1484375, + "learning_rate": 0.001715810291748792, + "loss": 0.1602, + "step": 25714 + }, + { + "epoch": 0.22321854845010025, + "grad_norm": 0.640625, + "learning_rate": 0.001715788586758116, + "loss": 0.1035, + "step": 25715 + }, + { + "epoch": 0.22322722893030442, + "grad_norm": 0.0966796875, + "learning_rate": 0.001715766881094023, + "loss": 0.085, + "step": 25716 + }, + { + "epoch": 0.22323590941050858, + "grad_norm": 0.26953125, + "learning_rate": 0.0017157451747565366, + "loss": 0.1025, + "step": 25717 + }, + { + "epoch": 0.22324458989071275, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017157234677456802, + "loss": 0.123, + "step": 25718 + }, + { + "epoch": 0.2232532703709169, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017157017600614782, + "loss": 0.0928, + "step": 25719 + }, + { + "epoch": 0.22326195085112108, + "grad_norm": 0.0869140625, + "learning_rate": 0.0017156800517039539, + "loss": 0.082, + "step": 25720 + }, + { + "epoch": 0.22327063133132524, + "grad_norm": 0.1640625, + "learning_rate": 0.001715658342673131, + "loss": 0.0972, + "step": 25721 + }, + { + "epoch": 0.2232793118115294, + "grad_norm": 0.07568359375, + "learning_rate": 0.0017156366329690335, + "loss": 0.1045, + "step": 25722 + }, + { + "epoch": 0.22328799229173357, + "grad_norm": 0.224609375, + "learning_rate": 0.0017156149225916852, + "loss": 0.1191, + "step": 25723 + }, + { + "epoch": 0.22329667277193774, + "grad_norm": 0.11962890625, + "learning_rate": 0.0017155932115411096, + "loss": 0.1016, + "step": 25724 + }, + { + "epoch": 0.2233053532521419, + "grad_norm": 0.15625, + "learning_rate": 0.0017155714998173305, + "loss": 0.1143, + "step": 25725 + }, + { + "epoch": 0.22331403373234607, + "grad_norm": 0.3984375, + "learning_rate": 0.0017155497874203716, + "loss": 0.0898, + "step": 25726 + }, + { + "epoch": 0.22332271421255023, + "grad_norm": 0.349609375, + "learning_rate": 0.0017155280743502564, + "loss": 0.1201, + "step": 25727 + }, + { + "epoch": 0.2233313946927544, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017155063606070094, + "loss": 0.1289, + "step": 25728 + }, + { + "epoch": 0.22334007517295856, + "grad_norm": 0.91015625, + "learning_rate": 0.0017154846461906537, + "loss": 0.1562, + "step": 25729 + }, + { + "epoch": 0.22334875565316273, + "grad_norm": 0.11328125, + "learning_rate": 0.0017154629311012133, + "loss": 0.1201, + "step": 25730 + }, + { + "epoch": 0.2233574361333669, + "grad_norm": 0.2216796875, + "learning_rate": 0.001715441215338712, + "loss": 0.1123, + "step": 25731 + }, + { + "epoch": 0.22336611661357106, + "grad_norm": 0.130859375, + "learning_rate": 0.0017154194989031733, + "loss": 0.0864, + "step": 25732 + }, + { + "epoch": 0.22337479709377522, + "grad_norm": 0.09521484375, + "learning_rate": 0.0017153977817946212, + "loss": 0.1406, + "step": 25733 + }, + { + "epoch": 0.2233834775739794, + "grad_norm": 0.330078125, + "learning_rate": 0.0017153760640130791, + "loss": 0.0996, + "step": 25734 + }, + { + "epoch": 0.22339215805418355, + "grad_norm": 0.48046875, + "learning_rate": 0.0017153543455585712, + "loss": 0.1006, + "step": 25735 + }, + { + "epoch": 0.22340083853438772, + "grad_norm": 0.53125, + "learning_rate": 0.001715332626431121, + "loss": 0.0679, + "step": 25736 + }, + { + "epoch": 0.22340951901459188, + "grad_norm": 0.23828125, + "learning_rate": 0.0017153109066307519, + "loss": 0.1523, + "step": 25737 + }, + { + "epoch": 0.22341819949479605, + "grad_norm": 0.32421875, + "learning_rate": 0.0017152891861574883, + "loss": 0.0933, + "step": 25738 + }, + { + "epoch": 0.22342687997500021, + "grad_norm": 1.109375, + "learning_rate": 0.0017152674650113538, + "loss": 0.1133, + "step": 25739 + }, + { + "epoch": 0.22343556045520438, + "grad_norm": 0.3203125, + "learning_rate": 0.001715245743192372, + "loss": 0.1309, + "step": 25740 + }, + { + "epoch": 0.22344424093540854, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017152240207005667, + "loss": 0.0947, + "step": 25741 + }, + { + "epoch": 0.2234529214156127, + "grad_norm": 0.08349609375, + "learning_rate": 0.0017152022975359619, + "loss": 0.0757, + "step": 25742 + }, + { + "epoch": 0.22346160189581687, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017151805736985806, + "loss": 0.124, + "step": 25743 + }, + { + "epoch": 0.22347028237602104, + "grad_norm": 0.8359375, + "learning_rate": 0.0017151588491884473, + "loss": 0.1475, + "step": 25744 + }, + { + "epoch": 0.2234789628562252, + "grad_norm": 0.10693359375, + "learning_rate": 0.0017151371240055858, + "loss": 0.1128, + "step": 25745 + }, + { + "epoch": 0.22348764333642937, + "grad_norm": 0.416015625, + "learning_rate": 0.0017151153981500192, + "loss": 0.1201, + "step": 25746 + }, + { + "epoch": 0.22349632381663354, + "grad_norm": 0.365234375, + "learning_rate": 0.0017150936716217719, + "loss": 0.1416, + "step": 25747 + }, + { + "epoch": 0.2235050042968377, + "grad_norm": 0.1767578125, + "learning_rate": 0.0017150719444208671, + "loss": 0.1187, + "step": 25748 + }, + { + "epoch": 0.22351368477704187, + "grad_norm": 0.1904296875, + "learning_rate": 0.001715050216547329, + "loss": 0.1436, + "step": 25749 + }, + { + "epoch": 0.22352236525724603, + "grad_norm": 0.9921875, + "learning_rate": 0.0017150284880011815, + "loss": 0.1602, + "step": 25750 + }, + { + "epoch": 0.2235310457374502, + "grad_norm": 0.53515625, + "learning_rate": 0.0017150067587824475, + "loss": 0.1201, + "step": 25751 + }, + { + "epoch": 0.22353972621765436, + "grad_norm": 1.1015625, + "learning_rate": 0.0017149850288911519, + "loss": 0.1367, + "step": 25752 + }, + { + "epoch": 0.22354840669785853, + "grad_norm": 0.3671875, + "learning_rate": 0.0017149632983273176, + "loss": 0.0781, + "step": 25753 + }, + { + "epoch": 0.2235570871780627, + "grad_norm": 0.84375, + "learning_rate": 0.001714941567090969, + "loss": 0.0991, + "step": 25754 + }, + { + "epoch": 0.22356576765826686, + "grad_norm": 0.5859375, + "learning_rate": 0.001714919835182129, + "loss": 0.1074, + "step": 25755 + }, + { + "epoch": 0.22357444813847102, + "grad_norm": 0.392578125, + "learning_rate": 0.0017148981026008224, + "loss": 0.1201, + "step": 25756 + }, + { + "epoch": 0.2235831286186752, + "grad_norm": 0.53125, + "learning_rate": 0.0017148763693470724, + "loss": 0.0918, + "step": 25757 + }, + { + "epoch": 0.22359180909887935, + "grad_norm": 0.388671875, + "learning_rate": 0.0017148546354209028, + "loss": 0.0918, + "step": 25758 + }, + { + "epoch": 0.22360048957908352, + "grad_norm": 0.69921875, + "learning_rate": 0.0017148329008223374, + "loss": 0.0996, + "step": 25759 + }, + { + "epoch": 0.22360917005928768, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017148111655514, + "loss": 0.1309, + "step": 25760 + }, + { + "epoch": 0.22361785053949185, + "grad_norm": 0.2236328125, + "learning_rate": 0.0017147894296081138, + "loss": 0.1436, + "step": 25761 + }, + { + "epoch": 0.223626531019696, + "grad_norm": 0.0849609375, + "learning_rate": 0.0017147676929925037, + "loss": 0.0859, + "step": 25762 + }, + { + "epoch": 0.22363521149990018, + "grad_norm": 0.2470703125, + "learning_rate": 0.001714745955704593, + "loss": 0.1006, + "step": 25763 + }, + { + "epoch": 0.22364389198010434, + "grad_norm": 0.416015625, + "learning_rate": 0.0017147242177444053, + "loss": 0.1074, + "step": 25764 + }, + { + "epoch": 0.2236525724603085, + "grad_norm": 0.251953125, + "learning_rate": 0.001714702479111964, + "loss": 0.0806, + "step": 25765 + }, + { + "epoch": 0.22366125294051267, + "grad_norm": 0.15625, + "learning_rate": 0.0017146807398072937, + "loss": 0.0762, + "step": 25766 + }, + { + "epoch": 0.22366993342071684, + "grad_norm": 0.80078125, + "learning_rate": 0.0017146589998304177, + "loss": 0.1289, + "step": 25767 + }, + { + "epoch": 0.223678613900921, + "grad_norm": 0.2353515625, + "learning_rate": 0.00171463725918136, + "loss": 0.1045, + "step": 25768 + }, + { + "epoch": 0.22368729438112517, + "grad_norm": 0.251953125, + "learning_rate": 0.0017146155178601437, + "loss": 0.1484, + "step": 25769 + }, + { + "epoch": 0.22369597486132933, + "grad_norm": 0.177734375, + "learning_rate": 0.0017145937758667935, + "loss": 0.1309, + "step": 25770 + }, + { + "epoch": 0.2237046553415335, + "grad_norm": 0.61328125, + "learning_rate": 0.0017145720332013329, + "loss": 0.1006, + "step": 25771 + }, + { + "epoch": 0.22371333582173766, + "grad_norm": 0.25390625, + "learning_rate": 0.0017145502898637855, + "loss": 0.1011, + "step": 25772 + }, + { + "epoch": 0.22372201630194183, + "grad_norm": 0.2421875, + "learning_rate": 0.0017145285458541747, + "loss": 0.1162, + "step": 25773 + }, + { + "epoch": 0.223730696782146, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017145068011725252, + "loss": 0.1006, + "step": 25774 + }, + { + "epoch": 0.22373937726235016, + "grad_norm": 0.228515625, + "learning_rate": 0.0017144850558188601, + "loss": 0.123, + "step": 25775 + }, + { + "epoch": 0.22374805774255432, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017144633097932035, + "loss": 0.1865, + "step": 25776 + }, + { + "epoch": 0.2237567382227585, + "grad_norm": 0.498046875, + "learning_rate": 0.001714441563095579, + "loss": 0.1211, + "step": 25777 + }, + { + "epoch": 0.22376541870296265, + "grad_norm": 0.392578125, + "learning_rate": 0.0017144198157260104, + "loss": 0.1094, + "step": 25778 + }, + { + "epoch": 0.22377409918316682, + "grad_norm": 0.6015625, + "learning_rate": 0.0017143980676845214, + "loss": 0.1133, + "step": 25779 + }, + { + "epoch": 0.22378277966337098, + "grad_norm": 0.109375, + "learning_rate": 0.001714376318971136, + "loss": 0.1123, + "step": 25780 + }, + { + "epoch": 0.22379146014357515, + "grad_norm": 0.31640625, + "learning_rate": 0.001714354569585878, + "loss": 0.1221, + "step": 25781 + }, + { + "epoch": 0.22380014062377931, + "grad_norm": 0.3046875, + "learning_rate": 0.0017143328195287709, + "loss": 0.1523, + "step": 25782 + }, + { + "epoch": 0.22380882110398348, + "grad_norm": 0.84375, + "learning_rate": 0.0017143110687998387, + "loss": 0.1357, + "step": 25783 + }, + { + "epoch": 0.22381750158418764, + "grad_norm": 0.154296875, + "learning_rate": 0.0017142893173991054, + "loss": 0.1089, + "step": 25784 + }, + { + "epoch": 0.2238261820643918, + "grad_norm": 0.490234375, + "learning_rate": 0.0017142675653265941, + "loss": 0.0864, + "step": 25785 + }, + { + "epoch": 0.22383486254459597, + "grad_norm": 0.30859375, + "learning_rate": 0.0017142458125823294, + "loss": 0.1582, + "step": 25786 + }, + { + "epoch": 0.22384354302480014, + "grad_norm": 0.0810546875, + "learning_rate": 0.0017142240591663345, + "loss": 0.0884, + "step": 25787 + }, + { + "epoch": 0.2238522235050043, + "grad_norm": 0.455078125, + "learning_rate": 0.0017142023050786335, + "loss": 0.1157, + "step": 25788 + }, + { + "epoch": 0.22386090398520847, + "grad_norm": 0.07080078125, + "learning_rate": 0.0017141805503192503, + "loss": 0.1309, + "step": 25789 + }, + { + "epoch": 0.22386958446541264, + "grad_norm": 0.2099609375, + "learning_rate": 0.0017141587948882081, + "loss": 0.0879, + "step": 25790 + }, + { + "epoch": 0.2238782649456168, + "grad_norm": 0.16015625, + "learning_rate": 0.0017141370387855311, + "loss": 0.1309, + "step": 25791 + }, + { + "epoch": 0.22388694542582097, + "grad_norm": 0.1767578125, + "learning_rate": 0.0017141152820112431, + "loss": 0.125, + "step": 25792 + }, + { + "epoch": 0.22389562590602513, + "grad_norm": 0.24609375, + "learning_rate": 0.001714093524565368, + "loss": 0.124, + "step": 25793 + }, + { + "epoch": 0.2239043063862293, + "grad_norm": 0.26171875, + "learning_rate": 0.0017140717664479294, + "loss": 0.1045, + "step": 25794 + }, + { + "epoch": 0.22391298686643346, + "grad_norm": 0.482421875, + "learning_rate": 0.0017140500076589511, + "loss": 0.1143, + "step": 25795 + }, + { + "epoch": 0.22392166734663763, + "grad_norm": 0.5390625, + "learning_rate": 0.0017140282481984568, + "loss": 0.106, + "step": 25796 + }, + { + "epoch": 0.2239303478268418, + "grad_norm": 0.4375, + "learning_rate": 0.0017140064880664706, + "loss": 0.0918, + "step": 25797 + }, + { + "epoch": 0.22393902830704596, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017139847272630163, + "loss": 0.1221, + "step": 25798 + }, + { + "epoch": 0.22394770878725012, + "grad_norm": 0.53125, + "learning_rate": 0.001713962965788117, + "loss": 0.0918, + "step": 25799 + }, + { + "epoch": 0.2239563892674543, + "grad_norm": 0.07275390625, + "learning_rate": 0.0017139412036417977, + "loss": 0.1074, + "step": 25800 + }, + { + "epoch": 0.22396506974765845, + "grad_norm": 0.169921875, + "learning_rate": 0.001713919440824081, + "loss": 0.1094, + "step": 25801 + }, + { + "epoch": 0.22397375022786262, + "grad_norm": 0.375, + "learning_rate": 0.0017138976773349915, + "loss": 0.1133, + "step": 25802 + }, + { + "epoch": 0.22398243070806678, + "grad_norm": 0.154296875, + "learning_rate": 0.0017138759131745526, + "loss": 0.123, + "step": 25803 + }, + { + "epoch": 0.22399111118827095, + "grad_norm": 0.09765625, + "learning_rate": 0.001713854148342788, + "loss": 0.123, + "step": 25804 + }, + { + "epoch": 0.2239997916684751, + "grad_norm": 0.625, + "learning_rate": 0.0017138323828397221, + "loss": 0.1465, + "step": 25805 + }, + { + "epoch": 0.22400847214867928, + "grad_norm": 0.443359375, + "learning_rate": 0.0017138106166653781, + "loss": 0.0732, + "step": 25806 + }, + { + "epoch": 0.22401715262888344, + "grad_norm": 0.3828125, + "learning_rate": 0.0017137888498197805, + "loss": 0.0981, + "step": 25807 + }, + { + "epoch": 0.2240258331090876, + "grad_norm": 0.310546875, + "learning_rate": 0.0017137670823029522, + "loss": 0.0698, + "step": 25808 + }, + { + "epoch": 0.22403451358929177, + "grad_norm": 0.0869140625, + "learning_rate": 0.0017137453141149173, + "loss": 0.0815, + "step": 25809 + }, + { + "epoch": 0.22404319406949594, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017137235452557, + "loss": 0.0928, + "step": 25810 + }, + { + "epoch": 0.2240518745497001, + "grad_norm": 0.20703125, + "learning_rate": 0.0017137017757253238, + "loss": 0.1143, + "step": 25811 + }, + { + "epoch": 0.22406055502990427, + "grad_norm": 0.2353515625, + "learning_rate": 0.0017136800055238127, + "loss": 0.1309, + "step": 25812 + }, + { + "epoch": 0.2240692355101084, + "grad_norm": 0.1484375, + "learning_rate": 0.0017136582346511903, + "loss": 0.1299, + "step": 25813 + }, + { + "epoch": 0.22407791599031257, + "grad_norm": 0.341796875, + "learning_rate": 0.0017136364631074803, + "loss": 0.1021, + "step": 25814 + }, + { + "epoch": 0.22408659647051674, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017136146908927066, + "loss": 0.1069, + "step": 25815 + }, + { + "epoch": 0.2240952769507209, + "grad_norm": 0.162109375, + "learning_rate": 0.0017135929180068935, + "loss": 0.1445, + "step": 25816 + }, + { + "epoch": 0.22410395743092507, + "grad_norm": 0.53125, + "learning_rate": 0.001713571144450064, + "loss": 0.1113, + "step": 25817 + }, + { + "epoch": 0.22411263791112923, + "grad_norm": 0.1484375, + "learning_rate": 0.0017135493702222426, + "loss": 0.1348, + "step": 25818 + }, + { + "epoch": 0.2241213183913334, + "grad_norm": 0.35546875, + "learning_rate": 0.001713527595323453, + "loss": 0.1099, + "step": 25819 + }, + { + "epoch": 0.22412999887153756, + "grad_norm": 0.296875, + "learning_rate": 0.0017135058197537184, + "loss": 0.1016, + "step": 25820 + }, + { + "epoch": 0.22413867935174173, + "grad_norm": 0.4140625, + "learning_rate": 0.0017134840435130629, + "loss": 0.1128, + "step": 25821 + }, + { + "epoch": 0.2241473598319459, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017134622666015108, + "loss": 0.0879, + "step": 25822 + }, + { + "epoch": 0.22415604031215006, + "grad_norm": 0.1875, + "learning_rate": 0.0017134404890190857, + "loss": 0.0757, + "step": 25823 + }, + { + "epoch": 0.22416472079235422, + "grad_norm": 0.53515625, + "learning_rate": 0.0017134187107658112, + "loss": 0.124, + "step": 25824 + }, + { + "epoch": 0.2241734012725584, + "grad_norm": 0.1279296875, + "learning_rate": 0.001713396931841711, + "loss": 0.1816, + "step": 25825 + }, + { + "epoch": 0.22418208175276255, + "grad_norm": 0.08203125, + "learning_rate": 0.0017133751522468096, + "loss": 0.1089, + "step": 25826 + }, + { + "epoch": 0.22419076223296672, + "grad_norm": 0.447265625, + "learning_rate": 0.00171335337198113, + "loss": 0.0996, + "step": 25827 + }, + { + "epoch": 0.22419944271317088, + "grad_norm": 0.361328125, + "learning_rate": 0.0017133315910446965, + "loss": 0.083, + "step": 25828 + }, + { + "epoch": 0.22420812319337505, + "grad_norm": 0.125, + "learning_rate": 0.0017133098094375327, + "loss": 0.0718, + "step": 25829 + }, + { + "epoch": 0.2242168036735792, + "grad_norm": 0.3046875, + "learning_rate": 0.0017132880271596624, + "loss": 0.1084, + "step": 25830 + }, + { + "epoch": 0.22422548415378338, + "grad_norm": 0.32421875, + "learning_rate": 0.0017132662442111098, + "loss": 0.1074, + "step": 25831 + }, + { + "epoch": 0.22423416463398754, + "grad_norm": 0.4296875, + "learning_rate": 0.0017132444605918987, + "loss": 0.1104, + "step": 25832 + }, + { + "epoch": 0.2242428451141917, + "grad_norm": 0.265625, + "learning_rate": 0.001713222676302052, + "loss": 0.1016, + "step": 25833 + }, + { + "epoch": 0.22425152559439587, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017132008913415945, + "loss": 0.103, + "step": 25834 + }, + { + "epoch": 0.22426020607460004, + "grad_norm": 0.291015625, + "learning_rate": 0.0017131791057105497, + "loss": 0.1123, + "step": 25835 + }, + { + "epoch": 0.2242688865548042, + "grad_norm": 0.447265625, + "learning_rate": 0.0017131573194089414, + "loss": 0.1133, + "step": 25836 + }, + { + "epoch": 0.22427756703500837, + "grad_norm": 0.0771484375, + "learning_rate": 0.0017131355324367934, + "loss": 0.0884, + "step": 25837 + }, + { + "epoch": 0.22428624751521253, + "grad_norm": 0.47265625, + "learning_rate": 0.0017131137447941298, + "loss": 0.084, + "step": 25838 + }, + { + "epoch": 0.2242949279954167, + "grad_norm": 0.189453125, + "learning_rate": 0.0017130919564809742, + "loss": 0.0913, + "step": 25839 + }, + { + "epoch": 0.22430360847562086, + "grad_norm": 0.1357421875, + "learning_rate": 0.0017130701674973503, + "loss": 0.1152, + "step": 25840 + }, + { + "epoch": 0.22431228895582503, + "grad_norm": 0.98828125, + "learning_rate": 0.001713048377843282, + "loss": 0.1172, + "step": 25841 + }, + { + "epoch": 0.2243209694360292, + "grad_norm": 0.390625, + "learning_rate": 0.001713026587518794, + "loss": 0.1064, + "step": 25842 + }, + { + "epoch": 0.22432964991623336, + "grad_norm": 0.06982421875, + "learning_rate": 0.0017130047965239082, + "loss": 0.0889, + "step": 25843 + }, + { + "epoch": 0.22433833039643752, + "grad_norm": 0.357421875, + "learning_rate": 0.0017129830048586501, + "loss": 0.0918, + "step": 25844 + }, + { + "epoch": 0.2243470108766417, + "grad_norm": 0.1904296875, + "learning_rate": 0.001712961212523043, + "loss": 0.1309, + "step": 25845 + }, + { + "epoch": 0.22435569135684585, + "grad_norm": 0.431640625, + "learning_rate": 0.0017129394195171106, + "loss": 0.1279, + "step": 25846 + }, + { + "epoch": 0.22436437183705002, + "grad_norm": 0.06787109375, + "learning_rate": 0.001712917625840877, + "loss": 0.0986, + "step": 25847 + }, + { + "epoch": 0.22437305231725418, + "grad_norm": 0.890625, + "learning_rate": 0.0017128958314943659, + "loss": 0.1807, + "step": 25848 + }, + { + "epoch": 0.22438173279745835, + "grad_norm": 0.44140625, + "learning_rate": 0.0017128740364776008, + "loss": 0.1045, + "step": 25849 + }, + { + "epoch": 0.22439041327766251, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017128522407906062, + "loss": 0.1016, + "step": 25850 + }, + { + "epoch": 0.22439909375786668, + "grad_norm": 0.220703125, + "learning_rate": 0.0017128304444334053, + "loss": 0.126, + "step": 25851 + }, + { + "epoch": 0.22440777423807085, + "grad_norm": 0.412109375, + "learning_rate": 0.0017128086474060228, + "loss": 0.1055, + "step": 25852 + }, + { + "epoch": 0.224416454718275, + "grad_norm": 0.197265625, + "learning_rate": 0.0017127868497084813, + "loss": 0.1143, + "step": 25853 + }, + { + "epoch": 0.22442513519847918, + "grad_norm": 0.119140625, + "learning_rate": 0.0017127650513408057, + "loss": 0.1133, + "step": 25854 + }, + { + "epoch": 0.22443381567868334, + "grad_norm": 0.158203125, + "learning_rate": 0.0017127432523030195, + "loss": 0.1089, + "step": 25855 + }, + { + "epoch": 0.2244424961588875, + "grad_norm": 0.59765625, + "learning_rate": 0.0017127214525951462, + "loss": 0.1172, + "step": 25856 + }, + { + "epoch": 0.22445117663909167, + "grad_norm": 0.294921875, + "learning_rate": 0.0017126996522172102, + "loss": 0.1177, + "step": 25857 + }, + { + "epoch": 0.22445985711929584, + "grad_norm": 0.302734375, + "learning_rate": 0.0017126778511692348, + "loss": 0.105, + "step": 25858 + }, + { + "epoch": 0.2244685375995, + "grad_norm": 0.23828125, + "learning_rate": 0.0017126560494512444, + "loss": 0.0981, + "step": 25859 + }, + { + "epoch": 0.22447721807970417, + "grad_norm": 0.22265625, + "learning_rate": 0.0017126342470632624, + "loss": 0.0786, + "step": 25860 + }, + { + "epoch": 0.22448589855990833, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017126124440053128, + "loss": 0.1006, + "step": 25861 + }, + { + "epoch": 0.2244945790401125, + "grad_norm": 0.1083984375, + "learning_rate": 0.0017125906402774196, + "loss": 0.1025, + "step": 25862 + }, + { + "epoch": 0.22450325952031666, + "grad_norm": 0.640625, + "learning_rate": 0.0017125688358796062, + "loss": 0.0835, + "step": 25863 + }, + { + "epoch": 0.22451194000052083, + "grad_norm": 0.251953125, + "learning_rate": 0.001712547030811897, + "loss": 0.0889, + "step": 25864 + }, + { + "epoch": 0.224520620480725, + "grad_norm": 0.40234375, + "learning_rate": 0.0017125252250743152, + "loss": 0.1309, + "step": 25865 + }, + { + "epoch": 0.22452930096092916, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017125034186668854, + "loss": 0.168, + "step": 25866 + }, + { + "epoch": 0.22453798144113332, + "grad_norm": 0.30078125, + "learning_rate": 0.0017124816115896306, + "loss": 0.1387, + "step": 25867 + }, + { + "epoch": 0.2245466619213375, + "grad_norm": 0.369140625, + "learning_rate": 0.0017124598038425756, + "loss": 0.0688, + "step": 25868 + }, + { + "epoch": 0.22455534240154165, + "grad_norm": 0.322265625, + "learning_rate": 0.0017124379954257434, + "loss": 0.0952, + "step": 25869 + }, + { + "epoch": 0.22456402288174582, + "grad_norm": 0.431640625, + "learning_rate": 0.0017124161863391585, + "loss": 0.1187, + "step": 25870 + }, + { + "epoch": 0.22457270336194998, + "grad_norm": 0.4921875, + "learning_rate": 0.0017123943765828444, + "loss": 0.0884, + "step": 25871 + }, + { + "epoch": 0.22458138384215415, + "grad_norm": 0.248046875, + "learning_rate": 0.0017123725661568247, + "loss": 0.1016, + "step": 25872 + }, + { + "epoch": 0.2245900643223583, + "grad_norm": 0.1015625, + "learning_rate": 0.001712350755061124, + "loss": 0.1001, + "step": 25873 + }, + { + "epoch": 0.22459874480256248, + "grad_norm": 0.185546875, + "learning_rate": 0.0017123289432957654, + "loss": 0.0947, + "step": 25874 + }, + { + "epoch": 0.22460742528276664, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017123071308607733, + "loss": 0.126, + "step": 25875 + }, + { + "epoch": 0.2246161057629708, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017122853177561712, + "loss": 0.2168, + "step": 25876 + }, + { + "epoch": 0.22462478624317497, + "grad_norm": 0.08740234375, + "learning_rate": 0.0017122635039819832, + "loss": 0.1025, + "step": 25877 + }, + { + "epoch": 0.22463346672337914, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017122416895382327, + "loss": 0.0908, + "step": 25878 + }, + { + "epoch": 0.2246421472035833, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017122198744249442, + "loss": 0.0903, + "step": 25879 + }, + { + "epoch": 0.22465082768378747, + "grad_norm": 0.734375, + "learning_rate": 0.0017121980586421416, + "loss": 0.1182, + "step": 25880 + }, + { + "epoch": 0.22465950816399163, + "grad_norm": 0.130859375, + "learning_rate": 0.0017121762421898475, + "loss": 0.0786, + "step": 25881 + }, + { + "epoch": 0.2246681886441958, + "grad_norm": 0.353515625, + "learning_rate": 0.001712154425068087, + "loss": 0.126, + "step": 25882 + }, + { + "epoch": 0.22467686912439996, + "grad_norm": 0.087890625, + "learning_rate": 0.0017121326072768838, + "loss": 0.1226, + "step": 25883 + }, + { + "epoch": 0.22468554960460413, + "grad_norm": 0.62890625, + "learning_rate": 0.0017121107888162616, + "loss": 0.0879, + "step": 25884 + }, + { + "epoch": 0.2246942300848083, + "grad_norm": 0.86328125, + "learning_rate": 0.0017120889696862443, + "loss": 0.1348, + "step": 25885 + }, + { + "epoch": 0.22470291056501246, + "grad_norm": 0.1259765625, + "learning_rate": 0.0017120671498868556, + "loss": 0.0854, + "step": 25886 + }, + { + "epoch": 0.22471159104521662, + "grad_norm": 0.3125, + "learning_rate": 0.0017120453294181192, + "loss": 0.1621, + "step": 25887 + }, + { + "epoch": 0.2247202715254208, + "grad_norm": 0.287109375, + "learning_rate": 0.0017120235082800596, + "loss": 0.0967, + "step": 25888 + }, + { + "epoch": 0.22472895200562495, + "grad_norm": 0.404296875, + "learning_rate": 0.0017120016864727002, + "loss": 0.1621, + "step": 25889 + }, + { + "epoch": 0.22473763248582912, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017119798639960646, + "loss": 0.0869, + "step": 25890 + }, + { + "epoch": 0.22474631296603328, + "grad_norm": 0.17578125, + "learning_rate": 0.0017119580408501774, + "loss": 0.1143, + "step": 25891 + }, + { + "epoch": 0.22475499344623745, + "grad_norm": 0.08056640625, + "learning_rate": 0.0017119362170350619, + "loss": 0.0703, + "step": 25892 + }, + { + "epoch": 0.22476367392644162, + "grad_norm": 0.16015625, + "learning_rate": 0.0017119143925507423, + "loss": 0.2383, + "step": 25893 + }, + { + "epoch": 0.22477235440664578, + "grad_norm": 0.390625, + "learning_rate": 0.001711892567397242, + "loss": 0.1211, + "step": 25894 + }, + { + "epoch": 0.22478103488684995, + "grad_norm": 0.4921875, + "learning_rate": 0.0017118707415745855, + "loss": 0.1045, + "step": 25895 + }, + { + "epoch": 0.2247897153670541, + "grad_norm": 0.3671875, + "learning_rate": 0.0017118489150827962, + "loss": 0.0918, + "step": 25896 + }, + { + "epoch": 0.22479839584725828, + "grad_norm": 0.5078125, + "learning_rate": 0.0017118270879218985, + "loss": 0.1021, + "step": 25897 + }, + { + "epoch": 0.22480707632746244, + "grad_norm": 0.66796875, + "learning_rate": 0.0017118052600919153, + "loss": 0.1021, + "step": 25898 + }, + { + "epoch": 0.2248157568076666, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017117834315928714, + "loss": 0.1079, + "step": 25899 + }, + { + "epoch": 0.22482443728787077, + "grad_norm": 0.71875, + "learning_rate": 0.0017117616024247902, + "loss": 0.1299, + "step": 25900 + }, + { + "epoch": 0.22483311776807494, + "grad_norm": 0.671875, + "learning_rate": 0.001711739772587696, + "loss": 0.1475, + "step": 25901 + }, + { + "epoch": 0.2248417982482791, + "grad_norm": 0.97265625, + "learning_rate": 0.0017117179420816121, + "loss": 0.125, + "step": 25902 + }, + { + "epoch": 0.22485047872848327, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017116961109065625, + "loss": 0.1426, + "step": 25903 + }, + { + "epoch": 0.22485915920868743, + "grad_norm": 0.232421875, + "learning_rate": 0.0017116742790625715, + "loss": 0.1089, + "step": 25904 + }, + { + "epoch": 0.2248678396888916, + "grad_norm": 0.375, + "learning_rate": 0.0017116524465496628, + "loss": 0.124, + "step": 25905 + }, + { + "epoch": 0.22487652016909576, + "grad_norm": 0.353515625, + "learning_rate": 0.0017116306133678598, + "loss": 0.0986, + "step": 25906 + }, + { + "epoch": 0.22488520064929993, + "grad_norm": 0.337890625, + "learning_rate": 0.0017116087795171868, + "loss": 0.1182, + "step": 25907 + }, + { + "epoch": 0.2248938811295041, + "grad_norm": 0.3515625, + "learning_rate": 0.0017115869449976678, + "loss": 0.1299, + "step": 25908 + }, + { + "epoch": 0.22490256160970826, + "grad_norm": 0.2431640625, + "learning_rate": 0.0017115651098093265, + "loss": 0.0776, + "step": 25909 + }, + { + "epoch": 0.22491124208991242, + "grad_norm": 0.19140625, + "learning_rate": 0.001711543273952187, + "loss": 0.1016, + "step": 25910 + }, + { + "epoch": 0.2249199225701166, + "grad_norm": 0.4609375, + "learning_rate": 0.0017115214374262726, + "loss": 0.2129, + "step": 25911 + }, + { + "epoch": 0.22492860305032075, + "grad_norm": 0.0888671875, + "learning_rate": 0.0017114996002316075, + "loss": 0.1172, + "step": 25912 + }, + { + "epoch": 0.22493728353052492, + "grad_norm": 0.19140625, + "learning_rate": 0.0017114777623682155, + "loss": 0.0962, + "step": 25913 + }, + { + "epoch": 0.22494596401072908, + "grad_norm": 0.3359375, + "learning_rate": 0.0017114559238361213, + "loss": 0.0889, + "step": 25914 + }, + { + "epoch": 0.22495464449093325, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017114340846353475, + "loss": 0.0806, + "step": 25915 + }, + { + "epoch": 0.2249633249711374, + "grad_norm": 0.189453125, + "learning_rate": 0.0017114122447659187, + "loss": 0.1064, + "step": 25916 + }, + { + "epoch": 0.22497200545134158, + "grad_norm": 0.326171875, + "learning_rate": 0.001711390404227859, + "loss": 0.1128, + "step": 25917 + }, + { + "epoch": 0.22498068593154574, + "grad_norm": 0.1240234375, + "learning_rate": 0.0017113685630211917, + "loss": 0.1035, + "step": 25918 + }, + { + "epoch": 0.2249893664117499, + "grad_norm": 0.333984375, + "learning_rate": 0.0017113467211459408, + "loss": 0.1416, + "step": 25919 + }, + { + "epoch": 0.22499804689195407, + "grad_norm": 0.28515625, + "learning_rate": 0.0017113248786021305, + "loss": 0.1006, + "step": 25920 + }, + { + "epoch": 0.22500672737215824, + "grad_norm": 0.474609375, + "learning_rate": 0.001711303035389784, + "loss": 0.1206, + "step": 25921 + }, + { + "epoch": 0.2250154078523624, + "grad_norm": 0.296875, + "learning_rate": 0.0017112811915089262, + "loss": 0.0859, + "step": 25922 + }, + { + "epoch": 0.22502408833256657, + "grad_norm": 0.53515625, + "learning_rate": 0.0017112593469595803, + "loss": 0.1128, + "step": 25923 + }, + { + "epoch": 0.22503276881277073, + "grad_norm": 0.0947265625, + "learning_rate": 0.0017112375017417705, + "loss": 0.1157, + "step": 25924 + }, + { + "epoch": 0.2250414492929749, + "grad_norm": 0.25, + "learning_rate": 0.0017112156558555206, + "loss": 0.0898, + "step": 25925 + }, + { + "epoch": 0.22505012977317906, + "grad_norm": 0.255859375, + "learning_rate": 0.0017111938093008542, + "loss": 0.106, + "step": 25926 + }, + { + "epoch": 0.22505881025338323, + "grad_norm": 0.328125, + "learning_rate": 0.0017111719620777955, + "loss": 0.0767, + "step": 25927 + }, + { + "epoch": 0.2250674907335874, + "grad_norm": 0.296875, + "learning_rate": 0.0017111501141863685, + "loss": 0.1025, + "step": 25928 + }, + { + "epoch": 0.22507617121379156, + "grad_norm": 0.15625, + "learning_rate": 0.0017111282656265966, + "loss": 0.0918, + "step": 25929 + }, + { + "epoch": 0.22508485169399572, + "grad_norm": 0.57421875, + "learning_rate": 0.0017111064163985042, + "loss": 0.1396, + "step": 25930 + }, + { + "epoch": 0.2250935321741999, + "grad_norm": 0.44921875, + "learning_rate": 0.001711084566502115, + "loss": 0.1211, + "step": 25931 + }, + { + "epoch": 0.22510221265440405, + "grad_norm": 0.41015625, + "learning_rate": 0.001711062715937453, + "loss": 0.1553, + "step": 25932 + }, + { + "epoch": 0.22511089313460822, + "grad_norm": 0.259765625, + "learning_rate": 0.001711040864704542, + "loss": 0.1069, + "step": 25933 + }, + { + "epoch": 0.22511957361481238, + "grad_norm": 0.4375, + "learning_rate": 0.0017110190128034057, + "loss": 0.1113, + "step": 25934 + }, + { + "epoch": 0.22512825409501655, + "grad_norm": 0.2138671875, + "learning_rate": 0.001710997160234068, + "loss": 0.1309, + "step": 25935 + }, + { + "epoch": 0.2251369345752207, + "grad_norm": 0.365234375, + "learning_rate": 0.0017109753069965534, + "loss": 0.0811, + "step": 25936 + }, + { + "epoch": 0.22514561505542485, + "grad_norm": 0.55078125, + "learning_rate": 0.0017109534530908852, + "loss": 0.0869, + "step": 25937 + }, + { + "epoch": 0.22515429553562902, + "grad_norm": 0.208984375, + "learning_rate": 0.0017109315985170874, + "loss": 0.106, + "step": 25938 + }, + { + "epoch": 0.22516297601583318, + "grad_norm": 0.2421875, + "learning_rate": 0.0017109097432751843, + "loss": 0.1152, + "step": 25939 + }, + { + "epoch": 0.22517165649603735, + "grad_norm": 0.1484375, + "learning_rate": 0.0017108878873651995, + "loss": 0.0986, + "step": 25940 + }, + { + "epoch": 0.2251803369762415, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017108660307871564, + "loss": 0.1152, + "step": 25941 + }, + { + "epoch": 0.22518901745644568, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017108441735410797, + "loss": 0.1475, + "step": 25942 + }, + { + "epoch": 0.22519769793664984, + "grad_norm": 0.1484375, + "learning_rate": 0.0017108223156269932, + "loss": 0.103, + "step": 25943 + }, + { + "epoch": 0.225206378416854, + "grad_norm": 0.953125, + "learning_rate": 0.00171080045704492, + "loss": 0.1035, + "step": 25944 + }, + { + "epoch": 0.22521505889705817, + "grad_norm": 0.67578125, + "learning_rate": 0.0017107785977948848, + "loss": 0.127, + "step": 25945 + }, + { + "epoch": 0.22522373937726234, + "grad_norm": 0.2060546875, + "learning_rate": 0.0017107567378769116, + "loss": 0.085, + "step": 25946 + }, + { + "epoch": 0.2252324198574665, + "grad_norm": 0.76953125, + "learning_rate": 0.001710734877291024, + "loss": 0.0654, + "step": 25947 + }, + { + "epoch": 0.22524110033767067, + "grad_norm": 0.5, + "learning_rate": 0.0017107130160372457, + "loss": 0.0845, + "step": 25948 + }, + { + "epoch": 0.22524978081787483, + "grad_norm": 0.162109375, + "learning_rate": 0.001710691154115601, + "loss": 0.1045, + "step": 25949 + }, + { + "epoch": 0.225258461298079, + "grad_norm": 0.609375, + "learning_rate": 0.0017106692915261135, + "loss": 0.1016, + "step": 25950 + }, + { + "epoch": 0.22526714177828316, + "grad_norm": 0.1591796875, + "learning_rate": 0.001710647428268807, + "loss": 0.0986, + "step": 25951 + }, + { + "epoch": 0.22527582225848733, + "grad_norm": 1.0, + "learning_rate": 0.001710625564343706, + "loss": 0.1309, + "step": 25952 + }, + { + "epoch": 0.2252845027386915, + "grad_norm": 0.384765625, + "learning_rate": 0.0017106036997508339, + "loss": 0.0811, + "step": 25953 + }, + { + "epoch": 0.22529318321889566, + "grad_norm": 0.294921875, + "learning_rate": 0.001710581834490215, + "loss": 0.1279, + "step": 25954 + }, + { + "epoch": 0.22530186369909982, + "grad_norm": 0.09130859375, + "learning_rate": 0.0017105599685618726, + "loss": 0.0879, + "step": 25955 + }, + { + "epoch": 0.225310544179304, + "grad_norm": 0.298828125, + "learning_rate": 0.0017105381019658313, + "loss": 0.083, + "step": 25956 + }, + { + "epoch": 0.22531922465950815, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017105162347021145, + "loss": 0.1484, + "step": 25957 + }, + { + "epoch": 0.22532790513971232, + "grad_norm": 0.26953125, + "learning_rate": 0.0017104943667707465, + "loss": 0.1309, + "step": 25958 + }, + { + "epoch": 0.22533658561991649, + "grad_norm": 0.16015625, + "learning_rate": 0.0017104724981717508, + "loss": 0.1196, + "step": 25959 + }, + { + "epoch": 0.22534526610012065, + "grad_norm": 0.5859375, + "learning_rate": 0.001710450628905152, + "loss": 0.1035, + "step": 25960 + }, + { + "epoch": 0.22535394658032482, + "grad_norm": 0.2216796875, + "learning_rate": 0.001710428758970973, + "loss": 0.1494, + "step": 25961 + }, + { + "epoch": 0.22536262706052898, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017104068883692387, + "loss": 0.1064, + "step": 25962 + }, + { + "epoch": 0.22537130754073315, + "grad_norm": 0.345703125, + "learning_rate": 0.0017103850170999725, + "loss": 0.1235, + "step": 25963 + }, + { + "epoch": 0.2253799880209373, + "grad_norm": 0.154296875, + "learning_rate": 0.001710363145163198, + "loss": 0.1025, + "step": 25964 + }, + { + "epoch": 0.22538866850114148, + "grad_norm": 0.109375, + "learning_rate": 0.00171034127255894, + "loss": 0.1406, + "step": 25965 + }, + { + "epoch": 0.22539734898134564, + "grad_norm": 0.4453125, + "learning_rate": 0.001710319399287222, + "loss": 0.085, + "step": 25966 + }, + { + "epoch": 0.2254060294615498, + "grad_norm": 0.14453125, + "learning_rate": 0.0017102975253480677, + "loss": 0.1299, + "step": 25967 + }, + { + "epoch": 0.22541470994175397, + "grad_norm": 0.083984375, + "learning_rate": 0.001710275650741501, + "loss": 0.126, + "step": 25968 + }, + { + "epoch": 0.22542339042195814, + "grad_norm": 0.62890625, + "learning_rate": 0.0017102537754675464, + "loss": 0.1055, + "step": 25969 + }, + { + "epoch": 0.2254320709021623, + "grad_norm": 0.1220703125, + "learning_rate": 0.0017102318995262274, + "loss": 0.1338, + "step": 25970 + }, + { + "epoch": 0.22544075138236647, + "grad_norm": 0.384765625, + "learning_rate": 0.0017102100229175678, + "loss": 0.1226, + "step": 25971 + }, + { + "epoch": 0.22544943186257063, + "grad_norm": 0.111328125, + "learning_rate": 0.0017101881456415917, + "loss": 0.1104, + "step": 25972 + }, + { + "epoch": 0.2254581123427748, + "grad_norm": 0.44921875, + "learning_rate": 0.001710166267698323, + "loss": 0.1211, + "step": 25973 + }, + { + "epoch": 0.22546679282297896, + "grad_norm": 0.46484375, + "learning_rate": 0.0017101443890877857, + "loss": 0.1318, + "step": 25974 + }, + { + "epoch": 0.22547547330318313, + "grad_norm": 0.466796875, + "learning_rate": 0.0017101225098100038, + "loss": 0.0801, + "step": 25975 + }, + { + "epoch": 0.2254841537833873, + "grad_norm": 0.322265625, + "learning_rate": 0.001710100629865001, + "loss": 0.1201, + "step": 25976 + }, + { + "epoch": 0.22549283426359146, + "grad_norm": 0.10986328125, + "learning_rate": 0.001710078749252801, + "loss": 0.125, + "step": 25977 + }, + { + "epoch": 0.22550151474379562, + "grad_norm": 0.09375, + "learning_rate": 0.0017100568679734283, + "loss": 0.0933, + "step": 25978 + }, + { + "epoch": 0.2255101952239998, + "grad_norm": 0.451171875, + "learning_rate": 0.0017100349860269066, + "loss": 0.1006, + "step": 25979 + }, + { + "epoch": 0.22551887570420395, + "grad_norm": 0.1015625, + "learning_rate": 0.00171001310341326, + "loss": 0.1221, + "step": 25980 + }, + { + "epoch": 0.22552755618440812, + "grad_norm": 0.1181640625, + "learning_rate": 0.001709991220132512, + "loss": 0.0972, + "step": 25981 + }, + { + "epoch": 0.22553623666461228, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017099693361846866, + "loss": 0.1396, + "step": 25982 + }, + { + "epoch": 0.22554491714481645, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017099474515698083, + "loss": 0.1143, + "step": 25983 + }, + { + "epoch": 0.2255535976250206, + "grad_norm": 0.1640625, + "learning_rate": 0.0017099255662879002, + "loss": 0.1074, + "step": 25984 + }, + { + "epoch": 0.22556227810522478, + "grad_norm": 0.0791015625, + "learning_rate": 0.0017099036803389868, + "loss": 0.1001, + "step": 25985 + }, + { + "epoch": 0.22557095858542894, + "grad_norm": 0.32421875, + "learning_rate": 0.0017098817937230921, + "loss": 0.1016, + "step": 25986 + }, + { + "epoch": 0.2255796390656331, + "grad_norm": 0.25, + "learning_rate": 0.00170985990644024, + "loss": 0.168, + "step": 25987 + }, + { + "epoch": 0.22558831954583727, + "grad_norm": 0.341796875, + "learning_rate": 0.0017098380184904537, + "loss": 0.1445, + "step": 25988 + }, + { + "epoch": 0.22559700002604144, + "grad_norm": 0.09765625, + "learning_rate": 0.001709816129873758, + "loss": 0.1001, + "step": 25989 + }, + { + "epoch": 0.2256056805062456, + "grad_norm": 0.341796875, + "learning_rate": 0.0017097942405901766, + "loss": 0.0996, + "step": 25990 + }, + { + "epoch": 0.22561436098644977, + "grad_norm": 0.37109375, + "learning_rate": 0.0017097723506397332, + "loss": 0.1367, + "step": 25991 + }, + { + "epoch": 0.22562304146665393, + "grad_norm": 0.267578125, + "learning_rate": 0.001709750460022452, + "loss": 0.0767, + "step": 25992 + }, + { + "epoch": 0.2256317219468581, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017097285687383567, + "loss": 0.1201, + "step": 25993 + }, + { + "epoch": 0.22564040242706226, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017097066767874716, + "loss": 0.0981, + "step": 25994 + }, + { + "epoch": 0.22564908290726643, + "grad_norm": 0.17578125, + "learning_rate": 0.0017096847841698204, + "loss": 0.0947, + "step": 25995 + }, + { + "epoch": 0.2256577633874706, + "grad_norm": 0.1328125, + "learning_rate": 0.0017096628908854275, + "loss": 0.1152, + "step": 25996 + }, + { + "epoch": 0.22566644386767476, + "grad_norm": 0.3125, + "learning_rate": 0.0017096409969343158, + "loss": 0.0811, + "step": 25997 + }, + { + "epoch": 0.22567512434787892, + "grad_norm": 0.81640625, + "learning_rate": 0.00170961910231651, + "loss": 0.1143, + "step": 25998 + }, + { + "epoch": 0.2256838048280831, + "grad_norm": 0.1572265625, + "learning_rate": 0.001709597207032034, + "loss": 0.1475, + "step": 25999 + }, + { + "epoch": 0.22569248530828726, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017095753110809118, + "loss": 0.0879, + "step": 26000 + }, + { + "epoch": 0.22570116578849142, + "grad_norm": 0.267578125, + "learning_rate": 0.001709553414463167, + "loss": 0.1064, + "step": 26001 + }, + { + "epoch": 0.22570984626869559, + "grad_norm": 0.255859375, + "learning_rate": 0.0017095315171788237, + "loss": 0.1484, + "step": 26002 + }, + { + "epoch": 0.22571852674889975, + "grad_norm": 0.240234375, + "learning_rate": 0.001709509619227906, + "loss": 0.1211, + "step": 26003 + }, + { + "epoch": 0.22572720722910392, + "grad_norm": 0.17578125, + "learning_rate": 0.0017094877206104378, + "loss": 0.1221, + "step": 26004 + }, + { + "epoch": 0.22573588770930808, + "grad_norm": 0.53125, + "learning_rate": 0.001709465821326443, + "loss": 0.0923, + "step": 26005 + }, + { + "epoch": 0.22574456818951225, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017094439213759453, + "loss": 0.085, + "step": 26006 + }, + { + "epoch": 0.2257532486697164, + "grad_norm": 0.251953125, + "learning_rate": 0.001709422020758969, + "loss": 0.1201, + "step": 26007 + }, + { + "epoch": 0.22576192914992058, + "grad_norm": 0.33203125, + "learning_rate": 0.0017094001194755382, + "loss": 0.1216, + "step": 26008 + }, + { + "epoch": 0.22577060963012474, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017093782175256762, + "loss": 0.085, + "step": 26009 + }, + { + "epoch": 0.2257792901103289, + "grad_norm": 0.3359375, + "learning_rate": 0.0017093563149094076, + "loss": 0.0781, + "step": 26010 + }, + { + "epoch": 0.22578797059053307, + "grad_norm": 0.09326171875, + "learning_rate": 0.0017093344116267558, + "loss": 0.0977, + "step": 26011 + }, + { + "epoch": 0.22579665107073724, + "grad_norm": 0.30859375, + "learning_rate": 0.001709312507677745, + "loss": 0.1006, + "step": 26012 + }, + { + "epoch": 0.2258053315509414, + "grad_norm": 0.11767578125, + "learning_rate": 0.0017092906030623995, + "loss": 0.0728, + "step": 26013 + }, + { + "epoch": 0.22581401203114557, + "grad_norm": 0.48046875, + "learning_rate": 0.001709268697780743, + "loss": 0.1064, + "step": 26014 + }, + { + "epoch": 0.22582269251134973, + "grad_norm": 0.134765625, + "learning_rate": 0.0017092467918327994, + "loss": 0.1348, + "step": 26015 + }, + { + "epoch": 0.2258313729915539, + "grad_norm": 0.365234375, + "learning_rate": 0.0017092248852185923, + "loss": 0.0918, + "step": 26016 + }, + { + "epoch": 0.22584005347175806, + "grad_norm": 0.7578125, + "learning_rate": 0.0017092029779381464, + "loss": 0.106, + "step": 26017 + }, + { + "epoch": 0.22584873395196223, + "grad_norm": 0.56640625, + "learning_rate": 0.0017091810699914852, + "loss": 0.0996, + "step": 26018 + }, + { + "epoch": 0.2258574144321664, + "grad_norm": 0.80859375, + "learning_rate": 0.0017091591613786327, + "loss": 0.1069, + "step": 26019 + }, + { + "epoch": 0.22586609491237056, + "grad_norm": 0.1533203125, + "learning_rate": 0.0017091372520996128, + "loss": 0.0923, + "step": 26020 + }, + { + "epoch": 0.22587477539257472, + "grad_norm": 0.70703125, + "learning_rate": 0.0017091153421544497, + "loss": 0.1064, + "step": 26021 + }, + { + "epoch": 0.2258834558727789, + "grad_norm": 0.52734375, + "learning_rate": 0.0017090934315431672, + "loss": 0.1187, + "step": 26022 + }, + { + "epoch": 0.22589213635298305, + "grad_norm": 0.470703125, + "learning_rate": 0.001709071520265789, + "loss": 0.084, + "step": 26023 + }, + { + "epoch": 0.22590081683318722, + "grad_norm": 0.470703125, + "learning_rate": 0.0017090496083223394, + "loss": 0.1377, + "step": 26024 + }, + { + "epoch": 0.22590949731339138, + "grad_norm": 0.248046875, + "learning_rate": 0.0017090276957128425, + "loss": 0.168, + "step": 26025 + }, + { + "epoch": 0.22591817779359555, + "grad_norm": 0.341796875, + "learning_rate": 0.001709005782437322, + "loss": 0.0869, + "step": 26026 + }, + { + "epoch": 0.2259268582737997, + "grad_norm": 0.12890625, + "learning_rate": 0.001708983868495802, + "loss": 0.1289, + "step": 26027 + }, + { + "epoch": 0.22593553875400388, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017089619538883062, + "loss": 0.1348, + "step": 26028 + }, + { + "epoch": 0.22594421923420804, + "grad_norm": 0.32421875, + "learning_rate": 0.001708940038614859, + "loss": 0.106, + "step": 26029 + }, + { + "epoch": 0.2259528997144122, + "grad_norm": 0.2294921875, + "learning_rate": 0.0017089181226754842, + "loss": 0.0786, + "step": 26030 + }, + { + "epoch": 0.22596158019461637, + "grad_norm": 0.275390625, + "learning_rate": 0.0017088962060702053, + "loss": 0.0737, + "step": 26031 + }, + { + "epoch": 0.22597026067482054, + "grad_norm": 0.21875, + "learning_rate": 0.001708874288799047, + "loss": 0.1191, + "step": 26032 + }, + { + "epoch": 0.2259789411550247, + "grad_norm": 0.294921875, + "learning_rate": 0.0017088523708620328, + "loss": 0.0938, + "step": 26033 + }, + { + "epoch": 0.22598762163522887, + "grad_norm": 0.671875, + "learning_rate": 0.0017088304522591866, + "loss": 0.0815, + "step": 26034 + }, + { + "epoch": 0.22599630211543303, + "grad_norm": 0.228515625, + "learning_rate": 0.0017088085329905328, + "loss": 0.0703, + "step": 26035 + }, + { + "epoch": 0.2260049825956372, + "grad_norm": 0.2021484375, + "learning_rate": 0.001708786613056095, + "loss": 0.0942, + "step": 26036 + }, + { + "epoch": 0.22601366307584136, + "grad_norm": 0.1845703125, + "learning_rate": 0.0017087646924558973, + "loss": 0.0928, + "step": 26037 + }, + { + "epoch": 0.22602234355604553, + "grad_norm": 0.1533203125, + "learning_rate": 0.001708742771189964, + "loss": 0.1221, + "step": 26038 + }, + { + "epoch": 0.2260310240362497, + "grad_norm": 0.158203125, + "learning_rate": 0.0017087208492583183, + "loss": 0.1621, + "step": 26039 + }, + { + "epoch": 0.22603970451645386, + "grad_norm": 0.1240234375, + "learning_rate": 0.001708698926660985, + "loss": 0.1016, + "step": 26040 + }, + { + "epoch": 0.22604838499665802, + "grad_norm": 0.328125, + "learning_rate": 0.0017086770033979872, + "loss": 0.105, + "step": 26041 + }, + { + "epoch": 0.2260570654768622, + "grad_norm": 0.5078125, + "learning_rate": 0.0017086550794693498, + "loss": 0.1011, + "step": 26042 + }, + { + "epoch": 0.22606574595706636, + "grad_norm": 0.4453125, + "learning_rate": 0.0017086331548750965, + "loss": 0.0786, + "step": 26043 + }, + { + "epoch": 0.22607442643727052, + "grad_norm": 0.228515625, + "learning_rate": 0.0017086112296152507, + "loss": 0.0972, + "step": 26044 + }, + { + "epoch": 0.22608310691747469, + "grad_norm": 0.1650390625, + "learning_rate": 0.001708589303689837, + "loss": 0.0947, + "step": 26045 + }, + { + "epoch": 0.22609178739767885, + "grad_norm": 0.84375, + "learning_rate": 0.0017085673770988793, + "loss": 0.0908, + "step": 26046 + }, + { + "epoch": 0.22610046787788302, + "grad_norm": 0.30078125, + "learning_rate": 0.0017085454498424013, + "loss": 0.1025, + "step": 26047 + }, + { + "epoch": 0.22610914835808718, + "grad_norm": 0.83203125, + "learning_rate": 0.0017085235219204272, + "loss": 0.1152, + "step": 26048 + }, + { + "epoch": 0.22611782883829135, + "grad_norm": 0.57421875, + "learning_rate": 0.001708501593332981, + "loss": 0.1128, + "step": 26049 + }, + { + "epoch": 0.2261265093184955, + "grad_norm": 0.47265625, + "learning_rate": 0.0017084796640800862, + "loss": 0.1182, + "step": 26050 + }, + { + "epoch": 0.22613518979869968, + "grad_norm": 1.03125, + "learning_rate": 0.0017084577341617674, + "loss": 0.1406, + "step": 26051 + }, + { + "epoch": 0.22614387027890384, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017084358035780484, + "loss": 0.1133, + "step": 26052 + }, + { + "epoch": 0.226152550759108, + "grad_norm": 0.267578125, + "learning_rate": 0.0017084138723289532, + "loss": 0.0742, + "step": 26053 + }, + { + "epoch": 0.22616123123931217, + "grad_norm": 0.88671875, + "learning_rate": 0.001708391940414506, + "loss": 0.1279, + "step": 26054 + }, + { + "epoch": 0.22616991171951634, + "grad_norm": 0.181640625, + "learning_rate": 0.00170837000783473, + "loss": 0.1387, + "step": 26055 + }, + { + "epoch": 0.2261785921997205, + "grad_norm": 0.50390625, + "learning_rate": 0.0017083480745896497, + "loss": 0.1006, + "step": 26056 + }, + { + "epoch": 0.22618727267992467, + "grad_norm": 0.2578125, + "learning_rate": 0.0017083261406792896, + "loss": 0.1074, + "step": 26057 + }, + { + "epoch": 0.22619595316012883, + "grad_norm": 0.52734375, + "learning_rate": 0.0017083042061036729, + "loss": 0.123, + "step": 26058 + }, + { + "epoch": 0.22620463364033297, + "grad_norm": 0.2470703125, + "learning_rate": 0.001708282270862824, + "loss": 0.1445, + "step": 26059 + }, + { + "epoch": 0.22621331412053713, + "grad_norm": 0.15625, + "learning_rate": 0.0017082603349567664, + "loss": 0.0825, + "step": 26060 + }, + { + "epoch": 0.2262219946007413, + "grad_norm": 0.83203125, + "learning_rate": 0.0017082383983855248, + "loss": 0.1289, + "step": 26061 + }, + { + "epoch": 0.22623067508094546, + "grad_norm": 0.29296875, + "learning_rate": 0.0017082164611491226, + "loss": 0.1167, + "step": 26062 + }, + { + "epoch": 0.22623935556114963, + "grad_norm": 0.216796875, + "learning_rate": 0.0017081945232475843, + "loss": 0.0986, + "step": 26063 + }, + { + "epoch": 0.2262480360413538, + "grad_norm": 0.75390625, + "learning_rate": 0.001708172584680933, + "loss": 0.0967, + "step": 26064 + }, + { + "epoch": 0.22625671652155796, + "grad_norm": 0.115234375, + "learning_rate": 0.0017081506454491937, + "loss": 0.1045, + "step": 26065 + }, + { + "epoch": 0.22626539700176213, + "grad_norm": 1.0234375, + "learning_rate": 0.0017081287055523901, + "loss": 0.0986, + "step": 26066 + }, + { + "epoch": 0.2262740774819663, + "grad_norm": 0.2353515625, + "learning_rate": 0.0017081067649905458, + "loss": 0.1177, + "step": 26067 + }, + { + "epoch": 0.22628275796217046, + "grad_norm": 0.51953125, + "learning_rate": 0.0017080848237636854, + "loss": 0.0991, + "step": 26068 + }, + { + "epoch": 0.22629143844237462, + "grad_norm": 0.486328125, + "learning_rate": 0.0017080628818718323, + "loss": 0.0967, + "step": 26069 + }, + { + "epoch": 0.22630011892257879, + "grad_norm": 0.244140625, + "learning_rate": 0.0017080409393150106, + "loss": 0.1621, + "step": 26070 + }, + { + "epoch": 0.22630879940278295, + "grad_norm": 0.216796875, + "learning_rate": 0.0017080189960932452, + "loss": 0.1055, + "step": 26071 + }, + { + "epoch": 0.22631747988298712, + "grad_norm": 0.1318359375, + "learning_rate": 0.0017079970522065587, + "loss": 0.106, + "step": 26072 + }, + { + "epoch": 0.22632616036319128, + "grad_norm": 0.87890625, + "learning_rate": 0.001707975107654976, + "loss": 0.1094, + "step": 26073 + }, + { + "epoch": 0.22633484084339545, + "grad_norm": 0.30859375, + "learning_rate": 0.0017079531624385207, + "loss": 0.0933, + "step": 26074 + }, + { + "epoch": 0.2263435213235996, + "grad_norm": 0.1005859375, + "learning_rate": 0.0017079312165572172, + "loss": 0.1221, + "step": 26075 + }, + { + "epoch": 0.22635220180380378, + "grad_norm": 0.201171875, + "learning_rate": 0.001707909270011089, + "loss": 0.0938, + "step": 26076 + }, + { + "epoch": 0.22636088228400794, + "grad_norm": 0.123046875, + "learning_rate": 0.0017078873228001605, + "loss": 0.1045, + "step": 26077 + }, + { + "epoch": 0.2263695627642121, + "grad_norm": 0.55859375, + "learning_rate": 0.0017078653749244557, + "loss": 0.1045, + "step": 26078 + }, + { + "epoch": 0.22637824324441627, + "grad_norm": 0.59765625, + "learning_rate": 0.0017078434263839981, + "loss": 0.1021, + "step": 26079 + }, + { + "epoch": 0.22638692372462044, + "grad_norm": 0.208984375, + "learning_rate": 0.0017078214771788123, + "loss": 0.1172, + "step": 26080 + }, + { + "epoch": 0.2263956042048246, + "grad_norm": 0.087890625, + "learning_rate": 0.0017077995273089219, + "loss": 0.1064, + "step": 26081 + }, + { + "epoch": 0.22640428468502877, + "grad_norm": 0.65625, + "learning_rate": 0.0017077775767743513, + "loss": 0.1279, + "step": 26082 + }, + { + "epoch": 0.22641296516523293, + "grad_norm": 0.20703125, + "learning_rate": 0.0017077556255751243, + "loss": 0.1113, + "step": 26083 + }, + { + "epoch": 0.2264216456454371, + "grad_norm": 0.2177734375, + "learning_rate": 0.0017077336737112646, + "loss": 0.126, + "step": 26084 + }, + { + "epoch": 0.22643032612564126, + "grad_norm": 0.5390625, + "learning_rate": 0.0017077117211827969, + "loss": 0.0928, + "step": 26085 + }, + { + "epoch": 0.22643900660584543, + "grad_norm": 0.1279296875, + "learning_rate": 0.0017076897679897444, + "loss": 0.0903, + "step": 26086 + }, + { + "epoch": 0.2264476870860496, + "grad_norm": 0.2578125, + "learning_rate": 0.0017076678141321317, + "loss": 0.0898, + "step": 26087 + }, + { + "epoch": 0.22645636756625376, + "grad_norm": 0.0966796875, + "learning_rate": 0.0017076458596099824, + "loss": 0.0835, + "step": 26088 + }, + { + "epoch": 0.22646504804645792, + "grad_norm": 0.130859375, + "learning_rate": 0.0017076239044233207, + "loss": 0.0967, + "step": 26089 + }, + { + "epoch": 0.2264737285266621, + "grad_norm": 0.13671875, + "learning_rate": 0.001707601948572171, + "loss": 0.1816, + "step": 26090 + }, + { + "epoch": 0.22648240900686625, + "grad_norm": 0.12890625, + "learning_rate": 0.0017075799920565568, + "loss": 0.1211, + "step": 26091 + }, + { + "epoch": 0.22649108948707042, + "grad_norm": 0.431640625, + "learning_rate": 0.001707558034876502, + "loss": 0.0977, + "step": 26092 + }, + { + "epoch": 0.22649976996727458, + "grad_norm": 0.443359375, + "learning_rate": 0.0017075360770320309, + "loss": 0.1006, + "step": 26093 + }, + { + "epoch": 0.22650845044747875, + "grad_norm": 1.859375, + "learning_rate": 0.0017075141185231677, + "loss": 0.4004, + "step": 26094 + }, + { + "epoch": 0.2265171309276829, + "grad_norm": 0.291015625, + "learning_rate": 0.001707492159349936, + "loss": 0.1084, + "step": 26095 + }, + { + "epoch": 0.22652581140788708, + "grad_norm": 0.314453125, + "learning_rate": 0.0017074701995123602, + "loss": 0.1104, + "step": 26096 + }, + { + "epoch": 0.22653449188809124, + "grad_norm": 0.99609375, + "learning_rate": 0.0017074482390104643, + "loss": 0.124, + "step": 26097 + }, + { + "epoch": 0.2265431723682954, + "grad_norm": 0.80078125, + "learning_rate": 0.0017074262778442712, + "loss": 0.0967, + "step": 26098 + }, + { + "epoch": 0.22655185284849957, + "grad_norm": 0.625, + "learning_rate": 0.001707404316013807, + "loss": 0.1074, + "step": 26099 + }, + { + "epoch": 0.22656053332870374, + "grad_norm": 0.0947265625, + "learning_rate": 0.001707382353519094, + "loss": 0.1011, + "step": 26100 + }, + { + "epoch": 0.2265692138089079, + "grad_norm": 0.333984375, + "learning_rate": 0.0017073603903601567, + "loss": 0.1118, + "step": 26101 + }, + { + "epoch": 0.22657789428911207, + "grad_norm": 0.50390625, + "learning_rate": 0.0017073384265370193, + "loss": 0.1787, + "step": 26102 + }, + { + "epoch": 0.22658657476931623, + "grad_norm": 0.412109375, + "learning_rate": 0.001707316462049706, + "loss": 0.0903, + "step": 26103 + }, + { + "epoch": 0.2265952552495204, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017072944968982403, + "loss": 0.0859, + "step": 26104 + }, + { + "epoch": 0.22660393572972456, + "grad_norm": 0.6484375, + "learning_rate": 0.0017072725310826465, + "loss": 0.0967, + "step": 26105 + }, + { + "epoch": 0.22661261620992873, + "grad_norm": 0.228515625, + "learning_rate": 0.0017072505646029482, + "loss": 0.0825, + "step": 26106 + }, + { + "epoch": 0.2266212966901329, + "grad_norm": 0.18359375, + "learning_rate": 0.0017072285974591701, + "loss": 0.0693, + "step": 26107 + }, + { + "epoch": 0.22662997717033706, + "grad_norm": 0.400390625, + "learning_rate": 0.0017072066296513362, + "loss": 0.0708, + "step": 26108 + }, + { + "epoch": 0.22663865765054123, + "grad_norm": 0.17578125, + "learning_rate": 0.0017071846611794702, + "loss": 0.0703, + "step": 26109 + }, + { + "epoch": 0.2266473381307454, + "grad_norm": 0.56640625, + "learning_rate": 0.0017071626920435962, + "loss": 0.1035, + "step": 26110 + }, + { + "epoch": 0.22665601861094956, + "grad_norm": 0.3828125, + "learning_rate": 0.0017071407222437377, + "loss": 0.1533, + "step": 26111 + }, + { + "epoch": 0.22666469909115372, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017071187517799197, + "loss": 0.1074, + "step": 26112 + }, + { + "epoch": 0.22667337957135789, + "grad_norm": 0.07666015625, + "learning_rate": 0.0017070967806521655, + "loss": 0.0957, + "step": 26113 + }, + { + "epoch": 0.22668206005156205, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017070748088604997, + "loss": 0.0791, + "step": 26114 + }, + { + "epoch": 0.22669074053176622, + "grad_norm": 0.271484375, + "learning_rate": 0.0017070528364049454, + "loss": 0.1074, + "step": 26115 + }, + { + "epoch": 0.22669942101197038, + "grad_norm": 0.37109375, + "learning_rate": 0.0017070308632855277, + "loss": 0.0947, + "step": 26116 + }, + { + "epoch": 0.22670810149217455, + "grad_norm": 0.1591796875, + "learning_rate": 0.0017070088895022701, + "loss": 0.1045, + "step": 26117 + }, + { + "epoch": 0.2267167819723787, + "grad_norm": 0.2236328125, + "learning_rate": 0.001706986915055197, + "loss": 0.0801, + "step": 26118 + }, + { + "epoch": 0.22672546245258288, + "grad_norm": 0.1865234375, + "learning_rate": 0.0017069649399443314, + "loss": 0.1143, + "step": 26119 + }, + { + "epoch": 0.22673414293278704, + "grad_norm": 0.19921875, + "learning_rate": 0.0017069429641696986, + "loss": 0.0981, + "step": 26120 + }, + { + "epoch": 0.2267428234129912, + "grad_norm": 0.53125, + "learning_rate": 0.001706920987731322, + "loss": 0.1074, + "step": 26121 + }, + { + "epoch": 0.22675150389319537, + "grad_norm": 0.126953125, + "learning_rate": 0.001706899010629226, + "loss": 0.0693, + "step": 26122 + }, + { + "epoch": 0.22676018437339954, + "grad_norm": 0.2275390625, + "learning_rate": 0.0017068770328634343, + "loss": 0.125, + "step": 26123 + }, + { + "epoch": 0.2267688648536037, + "grad_norm": 0.126953125, + "learning_rate": 0.0017068550544339703, + "loss": 0.124, + "step": 26124 + }, + { + "epoch": 0.22677754533380787, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017068330753408595, + "loss": 0.125, + "step": 26125 + }, + { + "epoch": 0.22678622581401203, + "grad_norm": 0.13671875, + "learning_rate": 0.0017068110955841247, + "loss": 0.0967, + "step": 26126 + }, + { + "epoch": 0.2267949062942162, + "grad_norm": 1.078125, + "learning_rate": 0.0017067891151637904, + "loss": 0.1904, + "step": 26127 + }, + { + "epoch": 0.22680358677442036, + "grad_norm": 0.55859375, + "learning_rate": 0.0017067671340798812, + "loss": 0.1602, + "step": 26128 + }, + { + "epoch": 0.22681226725462453, + "grad_norm": 0.103515625, + "learning_rate": 0.00170674515233242, + "loss": 0.1064, + "step": 26129 + }, + { + "epoch": 0.2268209477348287, + "grad_norm": 0.1171875, + "learning_rate": 0.0017067231699214316, + "loss": 0.125, + "step": 26130 + }, + { + "epoch": 0.22682962821503286, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017067011868469398, + "loss": 0.0986, + "step": 26131 + }, + { + "epoch": 0.22683830869523702, + "grad_norm": 0.357421875, + "learning_rate": 0.0017066792031089687, + "loss": 0.1094, + "step": 26132 + }, + { + "epoch": 0.2268469891754412, + "grad_norm": 0.5859375, + "learning_rate": 0.0017066572187075424, + "loss": 0.125, + "step": 26133 + }, + { + "epoch": 0.22685566965564535, + "grad_norm": 0.66796875, + "learning_rate": 0.0017066352336426848, + "loss": 0.1025, + "step": 26134 + }, + { + "epoch": 0.22686435013584952, + "grad_norm": 0.171875, + "learning_rate": 0.0017066132479144204, + "loss": 0.1162, + "step": 26135 + }, + { + "epoch": 0.22687303061605368, + "grad_norm": 0.2353515625, + "learning_rate": 0.0017065912615227725, + "loss": 0.0938, + "step": 26136 + }, + { + "epoch": 0.22688171109625785, + "grad_norm": 1.015625, + "learning_rate": 0.0017065692744677654, + "loss": 0.0986, + "step": 26137 + }, + { + "epoch": 0.226890391576462, + "grad_norm": 0.55078125, + "learning_rate": 0.0017065472867494236, + "loss": 0.0806, + "step": 26138 + }, + { + "epoch": 0.22689907205666618, + "grad_norm": 0.333984375, + "learning_rate": 0.0017065252983677708, + "loss": 0.1055, + "step": 26139 + }, + { + "epoch": 0.22690775253687034, + "grad_norm": 0.11767578125, + "learning_rate": 0.0017065033093228306, + "loss": 0.1221, + "step": 26140 + }, + { + "epoch": 0.2269164330170745, + "grad_norm": 0.1982421875, + "learning_rate": 0.001706481319614628, + "loss": 0.0996, + "step": 26141 + }, + { + "epoch": 0.22692511349727867, + "grad_norm": 0.42578125, + "learning_rate": 0.0017064593292431863, + "loss": 0.0903, + "step": 26142 + }, + { + "epoch": 0.22693379397748284, + "grad_norm": 0.380859375, + "learning_rate": 0.0017064373382085296, + "loss": 0.0986, + "step": 26143 + }, + { + "epoch": 0.226942474457687, + "grad_norm": 0.2041015625, + "learning_rate": 0.0017064153465106823, + "loss": 0.085, + "step": 26144 + }, + { + "epoch": 0.22695115493789117, + "grad_norm": 0.59375, + "learning_rate": 0.0017063933541496686, + "loss": 0.1387, + "step": 26145 + }, + { + "epoch": 0.22695983541809533, + "grad_norm": 1.0078125, + "learning_rate": 0.001706371361125512, + "loss": 0.1699, + "step": 26146 + }, + { + "epoch": 0.2269685158982995, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017063493674382368, + "loss": 0.0918, + "step": 26147 + }, + { + "epoch": 0.22697719637850367, + "grad_norm": 0.140625, + "learning_rate": 0.001706327373087867, + "loss": 0.1396, + "step": 26148 + }, + { + "epoch": 0.22698587685870783, + "grad_norm": 0.134765625, + "learning_rate": 0.0017063053780744266, + "loss": 0.1377, + "step": 26149 + }, + { + "epoch": 0.226994557338912, + "grad_norm": 0.255859375, + "learning_rate": 0.00170628338239794, + "loss": 0.1631, + "step": 26150 + }, + { + "epoch": 0.22700323781911616, + "grad_norm": 0.2099609375, + "learning_rate": 0.0017062613860584306, + "loss": 0.1206, + "step": 26151 + }, + { + "epoch": 0.22701191829932033, + "grad_norm": 0.279296875, + "learning_rate": 0.0017062393890559233, + "loss": 0.1641, + "step": 26152 + }, + { + "epoch": 0.2270205987795245, + "grad_norm": 0.322265625, + "learning_rate": 0.0017062173913904414, + "loss": 0.1021, + "step": 26153 + }, + { + "epoch": 0.22702927925972866, + "grad_norm": 0.66796875, + "learning_rate": 0.0017061953930620096, + "loss": 0.1621, + "step": 26154 + }, + { + "epoch": 0.22703795973993282, + "grad_norm": 0.162109375, + "learning_rate": 0.0017061733940706512, + "loss": 0.1318, + "step": 26155 + }, + { + "epoch": 0.22704664022013699, + "grad_norm": 0.5078125, + "learning_rate": 0.001706151394416391, + "loss": 0.1064, + "step": 26156 + }, + { + "epoch": 0.22705532070034115, + "grad_norm": 0.251953125, + "learning_rate": 0.0017061293940992526, + "loss": 0.1426, + "step": 26157 + }, + { + "epoch": 0.22706400118054532, + "grad_norm": 0.462890625, + "learning_rate": 0.0017061073931192603, + "loss": 0.1016, + "step": 26158 + }, + { + "epoch": 0.22707268166074948, + "grad_norm": 0.46484375, + "learning_rate": 0.0017060853914764378, + "loss": 0.1221, + "step": 26159 + }, + { + "epoch": 0.22708136214095365, + "grad_norm": 0.224609375, + "learning_rate": 0.0017060633891708098, + "loss": 0.0913, + "step": 26160 + }, + { + "epoch": 0.2270900426211578, + "grad_norm": 0.080078125, + "learning_rate": 0.0017060413862024, + "loss": 0.0913, + "step": 26161 + }, + { + "epoch": 0.22709872310136198, + "grad_norm": 0.361328125, + "learning_rate": 0.001706019382571232, + "loss": 0.1021, + "step": 26162 + }, + { + "epoch": 0.22710740358156614, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017059973782773306, + "loss": 0.1367, + "step": 26163 + }, + { + "epoch": 0.2271160840617703, + "grad_norm": 0.427734375, + "learning_rate": 0.0017059753733207196, + "loss": 0.1289, + "step": 26164 + }, + { + "epoch": 0.22712476454197447, + "grad_norm": 0.173828125, + "learning_rate": 0.001705953367701423, + "loss": 0.1338, + "step": 26165 + }, + { + "epoch": 0.22713344502217864, + "grad_norm": 0.11376953125, + "learning_rate": 0.0017059313614194648, + "loss": 0.1328, + "step": 26166 + }, + { + "epoch": 0.2271421255023828, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017059093544748694, + "loss": 0.1162, + "step": 26167 + }, + { + "epoch": 0.22715080598258697, + "grad_norm": 0.2578125, + "learning_rate": 0.0017058873468676604, + "loss": 0.1006, + "step": 26168 + }, + { + "epoch": 0.22715948646279113, + "grad_norm": 0.484375, + "learning_rate": 0.0017058653385978622, + "loss": 0.1064, + "step": 26169 + }, + { + "epoch": 0.2271681669429953, + "grad_norm": 0.61328125, + "learning_rate": 0.0017058433296654987, + "loss": 0.1279, + "step": 26170 + }, + { + "epoch": 0.22717684742319946, + "grad_norm": 0.1962890625, + "learning_rate": 0.001705821320070594, + "loss": 0.1162, + "step": 26171 + }, + { + "epoch": 0.22718552790340363, + "grad_norm": 0.1708984375, + "learning_rate": 0.0017057993098131723, + "loss": 0.0879, + "step": 26172 + }, + { + "epoch": 0.2271942083836078, + "grad_norm": 0.12060546875, + "learning_rate": 0.0017057772988932577, + "loss": 0.0928, + "step": 26173 + }, + { + "epoch": 0.22720288886381196, + "grad_norm": 0.291015625, + "learning_rate": 0.0017057552873108739, + "loss": 0.1045, + "step": 26174 + }, + { + "epoch": 0.22721156934401612, + "grad_norm": 0.201171875, + "learning_rate": 0.0017057332750660455, + "loss": 0.123, + "step": 26175 + }, + { + "epoch": 0.2272202498242203, + "grad_norm": 0.3984375, + "learning_rate": 0.001705711262158796, + "loss": 0.1562, + "step": 26176 + }, + { + "epoch": 0.22722893030442445, + "grad_norm": 0.443359375, + "learning_rate": 0.00170568924858915, + "loss": 0.1084, + "step": 26177 + }, + { + "epoch": 0.22723761078462862, + "grad_norm": 0.55078125, + "learning_rate": 0.001705667234357131, + "loss": 0.082, + "step": 26178 + }, + { + "epoch": 0.22724629126483278, + "grad_norm": 0.103515625, + "learning_rate": 0.001705645219462764, + "loss": 0.1348, + "step": 26179 + }, + { + "epoch": 0.22725497174503695, + "grad_norm": 0.6328125, + "learning_rate": 0.001705623203906072, + "loss": 0.1592, + "step": 26180 + }, + { + "epoch": 0.22726365222524111, + "grad_norm": 0.39453125, + "learning_rate": 0.0017056011876870796, + "loss": 0.1104, + "step": 26181 + }, + { + "epoch": 0.22727233270544528, + "grad_norm": 0.177734375, + "learning_rate": 0.0017055791708058114, + "loss": 0.0747, + "step": 26182 + }, + { + "epoch": 0.22728101318564942, + "grad_norm": 0.373046875, + "learning_rate": 0.00170555715326229, + "loss": 0.103, + "step": 26183 + }, + { + "epoch": 0.22728969366585358, + "grad_norm": 0.28125, + "learning_rate": 0.0017055351350565408, + "loss": 0.0908, + "step": 26184 + }, + { + "epoch": 0.22729837414605775, + "grad_norm": 0.08837890625, + "learning_rate": 0.0017055131161885875, + "loss": 0.1484, + "step": 26185 + }, + { + "epoch": 0.2273070546262619, + "grad_norm": 0.375, + "learning_rate": 0.0017054910966584542, + "loss": 0.0864, + "step": 26186 + }, + { + "epoch": 0.22731573510646608, + "grad_norm": 0.236328125, + "learning_rate": 0.001705469076466165, + "loss": 0.0996, + "step": 26187 + }, + { + "epoch": 0.22732441558667024, + "grad_norm": 0.34765625, + "learning_rate": 0.0017054470556117439, + "loss": 0.166, + "step": 26188 + }, + { + "epoch": 0.2273330960668744, + "grad_norm": 0.09375, + "learning_rate": 0.001705425034095215, + "loss": 0.1045, + "step": 26189 + }, + { + "epoch": 0.22734177654707857, + "grad_norm": 0.1904296875, + "learning_rate": 0.0017054030119166021, + "loss": 0.1167, + "step": 26190 + }, + { + "epoch": 0.22735045702728274, + "grad_norm": 0.46875, + "learning_rate": 0.00170538098907593, + "loss": 0.0938, + "step": 26191 + }, + { + "epoch": 0.2273591375074869, + "grad_norm": 0.15625, + "learning_rate": 0.001705358965573222, + "loss": 0.1084, + "step": 26192 + }, + { + "epoch": 0.22736781798769107, + "grad_norm": 0.193359375, + "learning_rate": 0.0017053369414085026, + "loss": 0.0835, + "step": 26193 + }, + { + "epoch": 0.22737649846789523, + "grad_norm": 1.0703125, + "learning_rate": 0.0017053149165817958, + "loss": 0.125, + "step": 26194 + }, + { + "epoch": 0.2273851789480994, + "grad_norm": 0.2265625, + "learning_rate": 0.001705292891093126, + "loss": 0.1089, + "step": 26195 + }, + { + "epoch": 0.22739385942830356, + "grad_norm": 0.224609375, + "learning_rate": 0.0017052708649425165, + "loss": 0.0649, + "step": 26196 + }, + { + "epoch": 0.22740253990850773, + "grad_norm": 0.08203125, + "learning_rate": 0.0017052488381299922, + "loss": 0.084, + "step": 26197 + }, + { + "epoch": 0.2274112203887119, + "grad_norm": 0.10205078125, + "learning_rate": 0.0017052268106555767, + "loss": 0.1309, + "step": 26198 + }, + { + "epoch": 0.22741990086891606, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017052047825192945, + "loss": 0.125, + "step": 26199 + }, + { + "epoch": 0.22742858134912022, + "grad_norm": 0.2119140625, + "learning_rate": 0.0017051827537211692, + "loss": 0.1104, + "step": 26200 + }, + { + "epoch": 0.2274372618293244, + "grad_norm": 0.39453125, + "learning_rate": 0.001705160724261225, + "loss": 0.0957, + "step": 26201 + }, + { + "epoch": 0.22744594230952855, + "grad_norm": 0.0908203125, + "learning_rate": 0.0017051386941394867, + "loss": 0.1045, + "step": 26202 + }, + { + "epoch": 0.22745462278973272, + "grad_norm": 0.2216796875, + "learning_rate": 0.0017051166633559771, + "loss": 0.1035, + "step": 26203 + }, + { + "epoch": 0.22746330326993688, + "grad_norm": 0.474609375, + "learning_rate": 0.0017050946319107218, + "loss": 0.0957, + "step": 26204 + }, + { + "epoch": 0.22747198375014105, + "grad_norm": 0.294921875, + "learning_rate": 0.0017050725998037435, + "loss": 0.1055, + "step": 26205 + }, + { + "epoch": 0.22748066423034521, + "grad_norm": 0.353515625, + "learning_rate": 0.0017050505670350672, + "loss": 0.1455, + "step": 26206 + }, + { + "epoch": 0.22748934471054938, + "grad_norm": 0.15625, + "learning_rate": 0.0017050285336047163, + "loss": 0.0698, + "step": 26207 + }, + { + "epoch": 0.22749802519075354, + "grad_norm": 0.140625, + "learning_rate": 0.0017050064995127155, + "loss": 0.1904, + "step": 26208 + }, + { + "epoch": 0.2275067056709577, + "grad_norm": 0.189453125, + "learning_rate": 0.0017049844647590892, + "loss": 0.0981, + "step": 26209 + }, + { + "epoch": 0.22751538615116187, + "grad_norm": 0.255859375, + "learning_rate": 0.0017049624293438601, + "loss": 0.1162, + "step": 26210 + }, + { + "epoch": 0.22752406663136604, + "grad_norm": 0.291015625, + "learning_rate": 0.0017049403932670535, + "loss": 0.1011, + "step": 26211 + }, + { + "epoch": 0.2275327471115702, + "grad_norm": 0.578125, + "learning_rate": 0.0017049183565286934, + "loss": 0.1006, + "step": 26212 + }, + { + "epoch": 0.22754142759177437, + "grad_norm": 1.2421875, + "learning_rate": 0.0017048963191288037, + "loss": 0.1816, + "step": 26213 + }, + { + "epoch": 0.22755010807197854, + "grad_norm": 0.310546875, + "learning_rate": 0.001704874281067408, + "loss": 0.2539, + "step": 26214 + }, + { + "epoch": 0.2275587885521827, + "grad_norm": 0.138671875, + "learning_rate": 0.0017048522423445314, + "loss": 0.1221, + "step": 26215 + }, + { + "epoch": 0.22756746903238687, + "grad_norm": 0.0673828125, + "learning_rate": 0.001704830202960197, + "loss": 0.0615, + "step": 26216 + }, + { + "epoch": 0.22757614951259103, + "grad_norm": 0.8046875, + "learning_rate": 0.00170480816291443, + "loss": 0.1113, + "step": 26217 + }, + { + "epoch": 0.2275848299927952, + "grad_norm": 0.2412109375, + "learning_rate": 0.0017047861222072534, + "loss": 0.1196, + "step": 26218 + }, + { + "epoch": 0.22759351047299936, + "grad_norm": 0.111328125, + "learning_rate": 0.0017047640808386917, + "loss": 0.1445, + "step": 26219 + }, + { + "epoch": 0.22760219095320353, + "grad_norm": 0.67578125, + "learning_rate": 0.0017047420388087693, + "loss": 0.1133, + "step": 26220 + }, + { + "epoch": 0.2276108714334077, + "grad_norm": 0.12890625, + "learning_rate": 0.0017047199961175102, + "loss": 0.0879, + "step": 26221 + }, + { + "epoch": 0.22761955191361186, + "grad_norm": 0.08544921875, + "learning_rate": 0.0017046979527649384, + "loss": 0.1299, + "step": 26222 + }, + { + "epoch": 0.22762823239381602, + "grad_norm": 0.146484375, + "learning_rate": 0.0017046759087510778, + "loss": 0.1221, + "step": 26223 + }, + { + "epoch": 0.2276369128740202, + "grad_norm": 0.267578125, + "learning_rate": 0.0017046538640759527, + "loss": 0.1436, + "step": 26224 + }, + { + "epoch": 0.22764559335422435, + "grad_norm": 0.201171875, + "learning_rate": 0.0017046318187395873, + "loss": 0.0928, + "step": 26225 + }, + { + "epoch": 0.22765427383442852, + "grad_norm": 0.150390625, + "learning_rate": 0.0017046097727420057, + "loss": 0.1084, + "step": 26226 + }, + { + "epoch": 0.22766295431463268, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017045877260832318, + "loss": 0.1348, + "step": 26227 + }, + { + "epoch": 0.22767163479483685, + "grad_norm": 0.1689453125, + "learning_rate": 0.0017045656787632896, + "loss": 0.1377, + "step": 26228 + }, + { + "epoch": 0.227680315275041, + "grad_norm": 0.0869140625, + "learning_rate": 0.0017045436307822042, + "loss": 0.0854, + "step": 26229 + }, + { + "epoch": 0.22768899575524518, + "grad_norm": 0.208984375, + "learning_rate": 0.0017045215821399986, + "loss": 0.2773, + "step": 26230 + }, + { + "epoch": 0.22769767623544934, + "grad_norm": 0.1396484375, + "learning_rate": 0.0017044995328366972, + "loss": 0.1533, + "step": 26231 + }, + { + "epoch": 0.2277063567156535, + "grad_norm": 0.203125, + "learning_rate": 0.0017044774828723242, + "loss": 0.0796, + "step": 26232 + }, + { + "epoch": 0.22771503719585767, + "grad_norm": 0.76171875, + "learning_rate": 0.0017044554322469038, + "loss": 0.166, + "step": 26233 + }, + { + "epoch": 0.22772371767606184, + "grad_norm": 0.1552734375, + "learning_rate": 0.0017044333809604598, + "loss": 0.0752, + "step": 26234 + }, + { + "epoch": 0.227732398156266, + "grad_norm": 0.2490234375, + "learning_rate": 0.0017044113290130168, + "loss": 0.1406, + "step": 26235 + }, + { + "epoch": 0.22774107863647017, + "grad_norm": 0.365234375, + "learning_rate": 0.0017043892764045986, + "loss": 0.104, + "step": 26236 + }, + { + "epoch": 0.22774975911667433, + "grad_norm": 0.333984375, + "learning_rate": 0.001704367223135229, + "loss": 0.1152, + "step": 26237 + }, + { + "epoch": 0.2277584395968785, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017043451692049327, + "loss": 0.0928, + "step": 26238 + }, + { + "epoch": 0.22776712007708266, + "grad_norm": 0.333984375, + "learning_rate": 0.0017043231146137339, + "loss": 0.1338, + "step": 26239 + }, + { + "epoch": 0.22777580055728683, + "grad_norm": 0.609375, + "learning_rate": 0.001704301059361656, + "loss": 0.0986, + "step": 26240 + }, + { + "epoch": 0.227784481037491, + "grad_norm": 0.099609375, + "learning_rate": 0.0017042790034487241, + "loss": 0.0864, + "step": 26241 + }, + { + "epoch": 0.22779316151769516, + "grad_norm": 0.87109375, + "learning_rate": 0.0017042569468749614, + "loss": 0.1201, + "step": 26242 + }, + { + "epoch": 0.22780184199789932, + "grad_norm": 0.1982421875, + "learning_rate": 0.0017042348896403924, + "loss": 0.127, + "step": 26243 + }, + { + "epoch": 0.2278105224781035, + "grad_norm": 0.25390625, + "learning_rate": 0.0017042128317450412, + "loss": 0.0869, + "step": 26244 + }, + { + "epoch": 0.22781920295830765, + "grad_norm": 0.25390625, + "learning_rate": 0.001704190773188932, + "loss": 0.1387, + "step": 26245 + }, + { + "epoch": 0.22782788343851182, + "grad_norm": 0.640625, + "learning_rate": 0.0017041687139720887, + "loss": 0.1162, + "step": 26246 + }, + { + "epoch": 0.22783656391871598, + "grad_norm": 0.123046875, + "learning_rate": 0.0017041466540945356, + "loss": 0.1074, + "step": 26247 + }, + { + "epoch": 0.22784524439892015, + "grad_norm": 0.224609375, + "learning_rate": 0.001704124593556297, + "loss": 0.0693, + "step": 26248 + }, + { + "epoch": 0.22785392487912431, + "grad_norm": 0.53515625, + "learning_rate": 0.0017041025323573968, + "loss": 0.1084, + "step": 26249 + }, + { + "epoch": 0.22786260535932848, + "grad_norm": 0.0986328125, + "learning_rate": 0.0017040804704978587, + "loss": 0.0923, + "step": 26250 + }, + { + "epoch": 0.22787128583953264, + "grad_norm": 0.490234375, + "learning_rate": 0.0017040584079777077, + "loss": 0.0869, + "step": 26251 + }, + { + "epoch": 0.2278799663197368, + "grad_norm": 0.30078125, + "learning_rate": 0.0017040363447969674, + "loss": 0.0957, + "step": 26252 + }, + { + "epoch": 0.22788864679994097, + "grad_norm": 0.35546875, + "learning_rate": 0.0017040142809556617, + "loss": 0.0918, + "step": 26253 + }, + { + "epoch": 0.22789732728014514, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017039922164538155, + "loss": 0.0742, + "step": 26254 + }, + { + "epoch": 0.2279060077603493, + "grad_norm": 0.62109375, + "learning_rate": 0.0017039701512914524, + "loss": 0.1094, + "step": 26255 + }, + { + "epoch": 0.22791468824055347, + "grad_norm": 0.375, + "learning_rate": 0.0017039480854685967, + "loss": 0.1396, + "step": 26256 + }, + { + "epoch": 0.22792336872075764, + "grad_norm": 0.1640625, + "learning_rate": 0.001703926018985272, + "loss": 0.0903, + "step": 26257 + }, + { + "epoch": 0.2279320492009618, + "grad_norm": 0.08544921875, + "learning_rate": 0.0017039039518415035, + "loss": 0.1191, + "step": 26258 + }, + { + "epoch": 0.22794072968116597, + "grad_norm": 0.248046875, + "learning_rate": 0.0017038818840373142, + "loss": 0.1055, + "step": 26259 + }, + { + "epoch": 0.22794941016137013, + "grad_norm": 0.267578125, + "learning_rate": 0.0017038598155727294, + "loss": 0.0854, + "step": 26260 + }, + { + "epoch": 0.2279580906415743, + "grad_norm": 0.2080078125, + "learning_rate": 0.0017038377464477718, + "loss": 0.082, + "step": 26261 + }, + { + "epoch": 0.22796677112177846, + "grad_norm": 0.0859375, + "learning_rate": 0.0017038156766624665, + "loss": 0.0977, + "step": 26262 + }, + { + "epoch": 0.22797545160198263, + "grad_norm": 0.2314453125, + "learning_rate": 0.001703793606216838, + "loss": 0.1543, + "step": 26263 + }, + { + "epoch": 0.2279841320821868, + "grad_norm": 0.314453125, + "learning_rate": 0.0017037715351109093, + "loss": 0.1182, + "step": 26264 + }, + { + "epoch": 0.22799281256239096, + "grad_norm": 0.205078125, + "learning_rate": 0.0017037494633447055, + "loss": 0.0957, + "step": 26265 + }, + { + "epoch": 0.22800149304259512, + "grad_norm": 0.1494140625, + "learning_rate": 0.00170372739091825, + "loss": 0.124, + "step": 26266 + }, + { + "epoch": 0.2280101735227993, + "grad_norm": 0.25390625, + "learning_rate": 0.0017037053178315673, + "loss": 0.1172, + "step": 26267 + }, + { + "epoch": 0.22801885400300345, + "grad_norm": 0.44921875, + "learning_rate": 0.0017036832440846818, + "loss": 0.1143, + "step": 26268 + }, + { + "epoch": 0.22802753448320762, + "grad_norm": 0.62109375, + "learning_rate": 0.0017036611696776174, + "loss": 0.1177, + "step": 26269 + }, + { + "epoch": 0.22803621496341178, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017036390946103981, + "loss": 0.0947, + "step": 26270 + }, + { + "epoch": 0.22804489544361595, + "grad_norm": 0.68359375, + "learning_rate": 0.001703617018883048, + "loss": 0.1094, + "step": 26271 + }, + { + "epoch": 0.2280535759238201, + "grad_norm": 0.287109375, + "learning_rate": 0.0017035949424955917, + "loss": 0.1426, + "step": 26272 + }, + { + "epoch": 0.22806225640402428, + "grad_norm": 0.5390625, + "learning_rate": 0.0017035728654480528, + "loss": 0.0728, + "step": 26273 + }, + { + "epoch": 0.22807093688422844, + "grad_norm": 0.4921875, + "learning_rate": 0.0017035507877404558, + "loss": 0.1001, + "step": 26274 + }, + { + "epoch": 0.2280796173644326, + "grad_norm": 0.12109375, + "learning_rate": 0.0017035287093728247, + "loss": 0.1377, + "step": 26275 + }, + { + "epoch": 0.22808829784463677, + "grad_norm": 0.2216796875, + "learning_rate": 0.0017035066303451838, + "loss": 0.0889, + "step": 26276 + }, + { + "epoch": 0.22809697832484094, + "grad_norm": 0.1513671875, + "learning_rate": 0.0017034845506575568, + "loss": 0.1631, + "step": 26277 + }, + { + "epoch": 0.2281056588050451, + "grad_norm": 0.41015625, + "learning_rate": 0.0017034624703099685, + "loss": 0.0864, + "step": 26278 + }, + { + "epoch": 0.22811433928524927, + "grad_norm": 0.203125, + "learning_rate": 0.0017034403893024424, + "loss": 0.0996, + "step": 26279 + }, + { + "epoch": 0.22812301976545343, + "grad_norm": 1.0703125, + "learning_rate": 0.001703418307635003, + "loss": 0.1035, + "step": 26280 + }, + { + "epoch": 0.2281317002456576, + "grad_norm": 0.234375, + "learning_rate": 0.0017033962253076747, + "loss": 0.1162, + "step": 26281 + }, + { + "epoch": 0.22814038072586176, + "grad_norm": 0.1318359375, + "learning_rate": 0.001703374142320481, + "loss": 0.1309, + "step": 26282 + }, + { + "epoch": 0.22814906120606593, + "grad_norm": 0.75, + "learning_rate": 0.0017033520586734465, + "loss": 0.1543, + "step": 26283 + }, + { + "epoch": 0.2281577416862701, + "grad_norm": 0.240234375, + "learning_rate": 0.0017033299743665956, + "loss": 0.1094, + "step": 26284 + }, + { + "epoch": 0.22816642216647426, + "grad_norm": 0.345703125, + "learning_rate": 0.0017033078893999517, + "loss": 0.1104, + "step": 26285 + }, + { + "epoch": 0.22817510264667842, + "grad_norm": 0.3203125, + "learning_rate": 0.0017032858037735394, + "loss": 0.1162, + "step": 26286 + }, + { + "epoch": 0.2281837831268826, + "grad_norm": 1.03125, + "learning_rate": 0.0017032637174873833, + "loss": 0.1211, + "step": 26287 + }, + { + "epoch": 0.22819246360708675, + "grad_norm": 0.30859375, + "learning_rate": 0.0017032416305415065, + "loss": 0.1328, + "step": 26288 + }, + { + "epoch": 0.22820114408729092, + "grad_norm": 0.427734375, + "learning_rate": 0.0017032195429359336, + "loss": 0.1348, + "step": 26289 + }, + { + "epoch": 0.22820982456749508, + "grad_norm": 0.07275390625, + "learning_rate": 0.0017031974546706892, + "loss": 0.0928, + "step": 26290 + }, + { + "epoch": 0.22821850504769925, + "grad_norm": 0.34375, + "learning_rate": 0.001703175365745797, + "loss": 0.1128, + "step": 26291 + }, + { + "epoch": 0.22822718552790341, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017031532761612813, + "loss": 0.1475, + "step": 26292 + }, + { + "epoch": 0.22823586600810758, + "grad_norm": 0.703125, + "learning_rate": 0.0017031311859171665, + "loss": 0.125, + "step": 26293 + }, + { + "epoch": 0.22824454648831174, + "grad_norm": 0.166015625, + "learning_rate": 0.0017031090950134762, + "loss": 0.0947, + "step": 26294 + }, + { + "epoch": 0.2282532269685159, + "grad_norm": 0.1611328125, + "learning_rate": 0.001703087003450235, + "loss": 0.1099, + "step": 26295 + }, + { + "epoch": 0.22826190744872007, + "grad_norm": 0.50390625, + "learning_rate": 0.0017030649112274668, + "loss": 0.1108, + "step": 26296 + }, + { + "epoch": 0.22827058792892424, + "grad_norm": 0.1171875, + "learning_rate": 0.0017030428183451957, + "loss": 0.1055, + "step": 26297 + }, + { + "epoch": 0.2282792684091284, + "grad_norm": 0.609375, + "learning_rate": 0.0017030207248034466, + "loss": 0.125, + "step": 26298 + }, + { + "epoch": 0.22828794888933257, + "grad_norm": 0.322265625, + "learning_rate": 0.0017029986306022428, + "loss": 0.1123, + "step": 26299 + }, + { + "epoch": 0.22829662936953674, + "grad_norm": 1.3515625, + "learning_rate": 0.0017029765357416088, + "loss": 0.1992, + "step": 26300 + }, + { + "epoch": 0.2283053098497409, + "grad_norm": 0.1748046875, + "learning_rate": 0.0017029544402215684, + "loss": 0.0742, + "step": 26301 + }, + { + "epoch": 0.22831399032994507, + "grad_norm": 0.181640625, + "learning_rate": 0.0017029323440421462, + "loss": 0.106, + "step": 26302 + }, + { + "epoch": 0.22832267081014923, + "grad_norm": 0.51171875, + "learning_rate": 0.001702910247203367, + "loss": 0.1064, + "step": 26303 + }, + { + "epoch": 0.2283313512903534, + "grad_norm": 0.357421875, + "learning_rate": 0.0017028881497052532, + "loss": 0.1758, + "step": 26304 + }, + { + "epoch": 0.22834003177055756, + "grad_norm": 0.373046875, + "learning_rate": 0.0017028660515478302, + "loss": 0.1104, + "step": 26305 + }, + { + "epoch": 0.2283487122507617, + "grad_norm": 0.482421875, + "learning_rate": 0.0017028439527311222, + "loss": 0.1211, + "step": 26306 + }, + { + "epoch": 0.22835739273096586, + "grad_norm": 0.181640625, + "learning_rate": 0.0017028218532551531, + "loss": 0.1133, + "step": 26307 + }, + { + "epoch": 0.22836607321117003, + "grad_norm": 0.1015625, + "learning_rate": 0.0017027997531199471, + "loss": 0.0859, + "step": 26308 + }, + { + "epoch": 0.2283747536913742, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017027776523255282, + "loss": 0.127, + "step": 26309 + }, + { + "epoch": 0.22838343417157836, + "grad_norm": 0.13671875, + "learning_rate": 0.0017027555508719206, + "loss": 0.1099, + "step": 26310 + }, + { + "epoch": 0.22839211465178252, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017027334487591489, + "loss": 0.1045, + "step": 26311 + }, + { + "epoch": 0.2284007951319867, + "grad_norm": 0.453125, + "learning_rate": 0.0017027113459872369, + "loss": 0.0796, + "step": 26312 + }, + { + "epoch": 0.22840947561219085, + "grad_norm": 0.09814453125, + "learning_rate": 0.0017026892425562087, + "loss": 0.1289, + "step": 26313 + }, + { + "epoch": 0.22841815609239502, + "grad_norm": 1.3828125, + "learning_rate": 0.0017026671384660884, + "loss": 0.1177, + "step": 26314 + }, + { + "epoch": 0.22842683657259918, + "grad_norm": 0.11083984375, + "learning_rate": 0.0017026450337169006, + "loss": 0.1211, + "step": 26315 + }, + { + "epoch": 0.22843551705280335, + "grad_norm": 0.58984375, + "learning_rate": 0.0017026229283086693, + "loss": 0.1182, + "step": 26316 + }, + { + "epoch": 0.22844419753300751, + "grad_norm": 1.6171875, + "learning_rate": 0.0017026008222414187, + "loss": 0.1299, + "step": 26317 + }, + { + "epoch": 0.22845287801321168, + "grad_norm": 0.1572265625, + "learning_rate": 0.0017025787155151727, + "loss": 0.0732, + "step": 26318 + }, + { + "epoch": 0.22846155849341584, + "grad_norm": 0.7421875, + "learning_rate": 0.0017025566081299557, + "loss": 0.0908, + "step": 26319 + }, + { + "epoch": 0.22847023897362, + "grad_norm": 0.1884765625, + "learning_rate": 0.001702534500085792, + "loss": 0.1299, + "step": 26320 + }, + { + "epoch": 0.22847891945382418, + "grad_norm": 0.8671875, + "learning_rate": 0.0017025123913827053, + "loss": 0.1289, + "step": 26321 + }, + { + "epoch": 0.22848759993402834, + "grad_norm": 0.107421875, + "learning_rate": 0.0017024902820207204, + "loss": 0.0762, + "step": 26322 + }, + { + "epoch": 0.2284962804142325, + "grad_norm": 0.1962890625, + "learning_rate": 0.0017024681719998611, + "loss": 0.0815, + "step": 26323 + }, + { + "epoch": 0.22850496089443667, + "grad_norm": 0.1943359375, + "learning_rate": 0.0017024460613201517, + "loss": 0.0645, + "step": 26324 + }, + { + "epoch": 0.22851364137464084, + "grad_norm": 0.486328125, + "learning_rate": 0.0017024239499816164, + "loss": 0.1016, + "step": 26325 + }, + { + "epoch": 0.228522321854845, + "grad_norm": 0.21875, + "learning_rate": 0.001702401837984279, + "loss": 0.1289, + "step": 26326 + }, + { + "epoch": 0.22853100233504917, + "grad_norm": 0.271484375, + "learning_rate": 0.0017023797253281642, + "loss": 0.123, + "step": 26327 + }, + { + "epoch": 0.22853968281525333, + "grad_norm": 0.90625, + "learning_rate": 0.001702357612013296, + "loss": 0.1182, + "step": 26328 + }, + { + "epoch": 0.2285483632954575, + "grad_norm": 0.63671875, + "learning_rate": 0.0017023354980396988, + "loss": 0.1338, + "step": 26329 + }, + { + "epoch": 0.22855704377566166, + "grad_norm": 0.2451171875, + "learning_rate": 0.0017023133834073961, + "loss": 0.1621, + "step": 26330 + }, + { + "epoch": 0.22856572425586583, + "grad_norm": 0.31640625, + "learning_rate": 0.0017022912681164128, + "loss": 0.1016, + "step": 26331 + }, + { + "epoch": 0.22857440473607, + "grad_norm": 0.1025390625, + "learning_rate": 0.0017022691521667728, + "loss": 0.1553, + "step": 26332 + }, + { + "epoch": 0.22858308521627416, + "grad_norm": 0.2734375, + "learning_rate": 0.0017022470355585, + "loss": 0.166, + "step": 26333 + }, + { + "epoch": 0.22859176569647832, + "grad_norm": 0.39453125, + "learning_rate": 0.0017022249182916192, + "loss": 0.0781, + "step": 26334 + }, + { + "epoch": 0.2286004461766825, + "grad_norm": 0.26953125, + "learning_rate": 0.0017022028003661543, + "loss": 0.1553, + "step": 26335 + }, + { + "epoch": 0.22860912665688665, + "grad_norm": 0.5390625, + "learning_rate": 0.0017021806817821292, + "loss": 0.1445, + "step": 26336 + }, + { + "epoch": 0.22861780713709082, + "grad_norm": 0.306640625, + "learning_rate": 0.0017021585625395688, + "loss": 0.1006, + "step": 26337 + }, + { + "epoch": 0.22862648761729498, + "grad_norm": 0.62890625, + "learning_rate": 0.0017021364426384967, + "loss": 0.0815, + "step": 26338 + }, + { + "epoch": 0.22863516809749915, + "grad_norm": 0.2890625, + "learning_rate": 0.0017021143220789369, + "loss": 0.1338, + "step": 26339 + }, + { + "epoch": 0.2286438485777033, + "grad_norm": 0.07080078125, + "learning_rate": 0.0017020922008609144, + "loss": 0.085, + "step": 26340 + }, + { + "epoch": 0.22865252905790748, + "grad_norm": 0.1318359375, + "learning_rate": 0.0017020700789844526, + "loss": 0.1099, + "step": 26341 + }, + { + "epoch": 0.22866120953811164, + "grad_norm": 0.349609375, + "learning_rate": 0.0017020479564495762, + "loss": 0.1226, + "step": 26342 + }, + { + "epoch": 0.2286698900183158, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017020258332563088, + "loss": 0.1445, + "step": 26343 + }, + { + "epoch": 0.22867857049851997, + "grad_norm": 0.1787109375, + "learning_rate": 0.001702003709404675, + "loss": 0.085, + "step": 26344 + }, + { + "epoch": 0.22868725097872414, + "grad_norm": 0.1455078125, + "learning_rate": 0.0017019815848946995, + "loss": 0.0991, + "step": 26345 + }, + { + "epoch": 0.2286959314589283, + "grad_norm": 0.4609375, + "learning_rate": 0.0017019594597264057, + "loss": 0.1631, + "step": 26346 + }, + { + "epoch": 0.22870461193913247, + "grad_norm": 0.27734375, + "learning_rate": 0.001701937333899818, + "loss": 0.0942, + "step": 26347 + }, + { + "epoch": 0.22871329241933663, + "grad_norm": 0.361328125, + "learning_rate": 0.0017019152074149609, + "loss": 0.0967, + "step": 26348 + }, + { + "epoch": 0.2287219728995408, + "grad_norm": 0.1630859375, + "learning_rate": 0.0017018930802718583, + "loss": 0.104, + "step": 26349 + }, + { + "epoch": 0.22873065337974496, + "grad_norm": 0.072265625, + "learning_rate": 0.0017018709524705346, + "loss": 0.0728, + "step": 26350 + }, + { + "epoch": 0.22873933385994913, + "grad_norm": 0.498046875, + "learning_rate": 0.0017018488240110137, + "loss": 0.123, + "step": 26351 + }, + { + "epoch": 0.2287480143401533, + "grad_norm": 0.3046875, + "learning_rate": 0.0017018266948933198, + "loss": 0.1226, + "step": 26352 + }, + { + "epoch": 0.22875669482035746, + "grad_norm": 0.24609375, + "learning_rate": 0.0017018045651174775, + "loss": 0.0913, + "step": 26353 + }, + { + "epoch": 0.22876537530056162, + "grad_norm": 0.1328125, + "learning_rate": 0.0017017824346835107, + "loss": 0.1006, + "step": 26354 + }, + { + "epoch": 0.2287740557807658, + "grad_norm": 0.1337890625, + "learning_rate": 0.0017017603035914437, + "loss": 0.1016, + "step": 26355 + }, + { + "epoch": 0.22878273626096995, + "grad_norm": 0.1669921875, + "learning_rate": 0.0017017381718413007, + "loss": 0.0952, + "step": 26356 + }, + { + "epoch": 0.22879141674117412, + "grad_norm": 0.08984375, + "learning_rate": 0.001701716039433106, + "loss": 0.1006, + "step": 26357 + }, + { + "epoch": 0.22880009722137828, + "grad_norm": 0.1435546875, + "learning_rate": 0.0017016939063668832, + "loss": 0.1738, + "step": 26358 + }, + { + "epoch": 0.22880877770158245, + "grad_norm": 0.435546875, + "learning_rate": 0.0017016717726426576, + "loss": 0.1035, + "step": 26359 + }, + { + "epoch": 0.22881745818178661, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017016496382604525, + "loss": 0.1562, + "step": 26360 + }, + { + "epoch": 0.22882613866199078, + "grad_norm": 0.11669921875, + "learning_rate": 0.001701627503220292, + "loss": 0.1157, + "step": 26361 + }, + { + "epoch": 0.22883481914219495, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017016053675222012, + "loss": 0.0806, + "step": 26362 + }, + { + "epoch": 0.2288434996223991, + "grad_norm": 0.216796875, + "learning_rate": 0.0017015832311662036, + "loss": 0.1025, + "step": 26363 + }, + { + "epoch": 0.22885218010260328, + "grad_norm": 0.72265625, + "learning_rate": 0.001701561094152324, + "loss": 0.1465, + "step": 26364 + }, + { + "epoch": 0.22886086058280744, + "grad_norm": 0.41015625, + "learning_rate": 0.001701538956480586, + "loss": 0.0708, + "step": 26365 + }, + { + "epoch": 0.2288695410630116, + "grad_norm": 0.265625, + "learning_rate": 0.0017015168181510137, + "loss": 0.084, + "step": 26366 + }, + { + "epoch": 0.22887822154321577, + "grad_norm": 0.1279296875, + "learning_rate": 0.001701494679163632, + "loss": 0.1069, + "step": 26367 + }, + { + "epoch": 0.22888690202341994, + "grad_norm": 0.1064453125, + "learning_rate": 0.0017014725395184647, + "loss": 0.0996, + "step": 26368 + }, + { + "epoch": 0.2288955825036241, + "grad_norm": 0.1162109375, + "learning_rate": 0.001701450399215536, + "loss": 0.0889, + "step": 26369 + }, + { + "epoch": 0.22890426298382827, + "grad_norm": 0.2333984375, + "learning_rate": 0.0017014282582548705, + "loss": 0.1406, + "step": 26370 + }, + { + "epoch": 0.22891294346403243, + "grad_norm": 0.490234375, + "learning_rate": 0.0017014061166364916, + "loss": 0.1182, + "step": 26371 + }, + { + "epoch": 0.2289216239442366, + "grad_norm": 0.11279296875, + "learning_rate": 0.001701383974360424, + "loss": 0.082, + "step": 26372 + }, + { + "epoch": 0.22893030442444076, + "grad_norm": 0.142578125, + "learning_rate": 0.0017013618314266922, + "loss": 0.1021, + "step": 26373 + }, + { + "epoch": 0.22893898490464493, + "grad_norm": 0.11279296875, + "learning_rate": 0.0017013396878353202, + "loss": 0.1484, + "step": 26374 + }, + { + "epoch": 0.2289476653848491, + "grad_norm": 0.44140625, + "learning_rate": 0.0017013175435863321, + "loss": 0.123, + "step": 26375 + }, + { + "epoch": 0.22895634586505326, + "grad_norm": 0.2314453125, + "learning_rate": 0.001701295398679752, + "loss": 0.1377, + "step": 26376 + }, + { + "epoch": 0.22896502634525742, + "grad_norm": 0.1376953125, + "learning_rate": 0.0017012732531156043, + "loss": 0.1465, + "step": 26377 + }, + { + "epoch": 0.2289737068254616, + "grad_norm": 3.21875, + "learning_rate": 0.0017012511068939134, + "loss": 0.377, + "step": 26378 + }, + { + "epoch": 0.22898238730566575, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017012289600147033, + "loss": 0.1011, + "step": 26379 + }, + { + "epoch": 0.22899106778586992, + "grad_norm": 0.51953125, + "learning_rate": 0.001701206812477998, + "loss": 0.1074, + "step": 26380 + }, + { + "epoch": 0.22899974826607408, + "grad_norm": 0.095703125, + "learning_rate": 0.0017011846642838223, + "loss": 0.0952, + "step": 26381 + }, + { + "epoch": 0.22900842874627825, + "grad_norm": 0.234375, + "learning_rate": 0.0017011625154322, + "loss": 0.0981, + "step": 26382 + }, + { + "epoch": 0.2290171092264824, + "grad_norm": 0.2890625, + "learning_rate": 0.0017011403659231553, + "loss": 0.126, + "step": 26383 + }, + { + "epoch": 0.22902578970668658, + "grad_norm": 0.06884765625, + "learning_rate": 0.0017011182157567124, + "loss": 0.0977, + "step": 26384 + }, + { + "epoch": 0.22903447018689074, + "grad_norm": 0.248046875, + "learning_rate": 0.0017010960649328957, + "loss": 0.1621, + "step": 26385 + }, + { + "epoch": 0.2290431506670949, + "grad_norm": 0.134765625, + "learning_rate": 0.0017010739134517296, + "loss": 0.1001, + "step": 26386 + }, + { + "epoch": 0.22905183114729907, + "grad_norm": 0.50390625, + "learning_rate": 0.001701051761313238, + "loss": 0.0942, + "step": 26387 + }, + { + "epoch": 0.22906051162750324, + "grad_norm": 0.2021484375, + "learning_rate": 0.001701029608517445, + "loss": 0.106, + "step": 26388 + }, + { + "epoch": 0.2290691921077074, + "grad_norm": 0.37109375, + "learning_rate": 0.0017010074550643755, + "loss": 0.0645, + "step": 26389 + }, + { + "epoch": 0.22907787258791157, + "grad_norm": 0.453125, + "learning_rate": 0.0017009853009540528, + "loss": 0.1206, + "step": 26390 + }, + { + "epoch": 0.22908655306811573, + "grad_norm": 0.09326171875, + "learning_rate": 0.001700963146186502, + "loss": 0.1162, + "step": 26391 + }, + { + "epoch": 0.2290952335483199, + "grad_norm": 0.189453125, + "learning_rate": 0.0017009409907617467, + "loss": 0.1006, + "step": 26392 + }, + { + "epoch": 0.22910391402852406, + "grad_norm": 0.1884765625, + "learning_rate": 0.0017009188346798114, + "loss": 0.1021, + "step": 26393 + }, + { + "epoch": 0.22911259450872823, + "grad_norm": 0.1416015625, + "learning_rate": 0.0017008966779407202, + "loss": 0.1025, + "step": 26394 + }, + { + "epoch": 0.2291212749889324, + "grad_norm": 0.384765625, + "learning_rate": 0.0017008745205444975, + "loss": 0.0898, + "step": 26395 + }, + { + "epoch": 0.22912995546913656, + "grad_norm": 0.1923828125, + "learning_rate": 0.0017008523624911674, + "loss": 0.1436, + "step": 26396 + }, + { + "epoch": 0.22913863594934072, + "grad_norm": 0.08203125, + "learning_rate": 0.0017008302037807544, + "loss": 0.0796, + "step": 26397 + }, + { + "epoch": 0.2291473164295449, + "grad_norm": 0.146484375, + "learning_rate": 0.0017008080444132826, + "loss": 0.1211, + "step": 26398 + }, + { + "epoch": 0.22915599690974905, + "grad_norm": 0.1298828125, + "learning_rate": 0.0017007858843887758, + "loss": 0.0859, + "step": 26399 + }, + { + "epoch": 0.22916467738995322, + "grad_norm": 0.80859375, + "learning_rate": 0.0017007637237072587, + "loss": 0.123, + "step": 26400 + }, + { + "epoch": 0.22917335787015738, + "grad_norm": 0.287109375, + "learning_rate": 0.0017007415623687553, + "loss": 0.1045, + "step": 26401 + }, + { + "epoch": 0.22918203835036155, + "grad_norm": 0.298828125, + "learning_rate": 0.0017007194003732901, + "loss": 0.0908, + "step": 26402 + }, + { + "epoch": 0.22919071883056572, + "grad_norm": 0.267578125, + "learning_rate": 0.0017006972377208874, + "loss": 0.085, + "step": 26403 + }, + { + "epoch": 0.22919939931076988, + "grad_norm": 0.69140625, + "learning_rate": 0.0017006750744115709, + "loss": 0.1318, + "step": 26404 + }, + { + "epoch": 0.22920807979097405, + "grad_norm": 0.326171875, + "learning_rate": 0.0017006529104453651, + "loss": 0.1143, + "step": 26405 + }, + { + "epoch": 0.2292167602711782, + "grad_norm": 0.439453125, + "learning_rate": 0.0017006307458222944, + "loss": 0.1338, + "step": 26406 + }, + { + "epoch": 0.22922544075138238, + "grad_norm": 0.203125, + "learning_rate": 0.0017006085805423832, + "loss": 0.1172, + "step": 26407 + }, + { + "epoch": 0.22923412123158654, + "grad_norm": 0.55859375, + "learning_rate": 0.0017005864146056554, + "loss": 0.1108, + "step": 26408 + }, + { + "epoch": 0.2292428017117907, + "grad_norm": 0.181640625, + "learning_rate": 0.001700564248012135, + "loss": 0.0918, + "step": 26409 + }, + { + "epoch": 0.22925148219199487, + "grad_norm": 0.1796875, + "learning_rate": 0.0017005420807618468, + "loss": 0.1045, + "step": 26410 + }, + { + "epoch": 0.22926016267219904, + "grad_norm": 0.166015625, + "learning_rate": 0.0017005199128548147, + "loss": 0.1465, + "step": 26411 + }, + { + "epoch": 0.2292688431524032, + "grad_norm": 0.205078125, + "learning_rate": 0.0017004977442910635, + "loss": 0.1055, + "step": 26412 + }, + { + "epoch": 0.22927752363260737, + "grad_norm": 0.0927734375, + "learning_rate": 0.0017004755750706162, + "loss": 0.1045, + "step": 26413 + }, + { + "epoch": 0.22928620411281153, + "grad_norm": 0.5859375, + "learning_rate": 0.0017004534051934983, + "loss": 0.1719, + "step": 26414 + }, + { + "epoch": 0.2292948845930157, + "grad_norm": 0.5078125, + "learning_rate": 0.0017004312346597336, + "loss": 0.1621, + "step": 26415 + }, + { + "epoch": 0.22930356507321986, + "grad_norm": 0.68359375, + "learning_rate": 0.0017004090634693463, + "loss": 0.0957, + "step": 26416 + }, + { + "epoch": 0.22931224555342403, + "grad_norm": 0.154296875, + "learning_rate": 0.0017003868916223603, + "loss": 0.1211, + "step": 26417 + }, + { + "epoch": 0.2293209260336282, + "grad_norm": 0.111328125, + "learning_rate": 0.0017003647191188007, + "loss": 0.0859, + "step": 26418 + }, + { + "epoch": 0.22932960651383236, + "grad_norm": 0.15625, + "learning_rate": 0.001700342545958691, + "loss": 0.1113, + "step": 26419 + }, + { + "epoch": 0.22933828699403652, + "grad_norm": 0.08984375, + "learning_rate": 0.0017003203721420558, + "loss": 0.1143, + "step": 26420 + }, + { + "epoch": 0.2293469674742407, + "grad_norm": 0.11328125, + "learning_rate": 0.0017002981976689195, + "loss": 0.1133, + "step": 26421 + }, + { + "epoch": 0.22935564795444485, + "grad_norm": 0.63671875, + "learning_rate": 0.0017002760225393056, + "loss": 0.1367, + "step": 26422 + }, + { + "epoch": 0.22936432843464902, + "grad_norm": 1.21875, + "learning_rate": 0.0017002538467532397, + "loss": 0.0825, + "step": 26423 + }, + { + "epoch": 0.22937300891485318, + "grad_norm": 0.212890625, + "learning_rate": 0.0017002316703107444, + "loss": 0.0835, + "step": 26424 + }, + { + "epoch": 0.22938168939505735, + "grad_norm": 0.248046875, + "learning_rate": 0.0017002094932118451, + "loss": 0.0869, + "step": 26425 + }, + { + "epoch": 0.2293903698752615, + "grad_norm": 0.2158203125, + "learning_rate": 0.0017001873154565655, + "loss": 0.0947, + "step": 26426 + }, + { + "epoch": 0.22939905035546568, + "grad_norm": 0.091796875, + "learning_rate": 0.0017001651370449303, + "loss": 0.0913, + "step": 26427 + }, + { + "epoch": 0.22940773083566984, + "grad_norm": 0.330078125, + "learning_rate": 0.0017001429579769634, + "loss": 0.0947, + "step": 26428 + }, + { + "epoch": 0.22941641131587398, + "grad_norm": 0.478515625, + "learning_rate": 0.0017001207782526891, + "loss": 0.1289, + "step": 26429 + }, + { + "epoch": 0.22942509179607815, + "grad_norm": 0.419921875, + "learning_rate": 0.0017000985978721317, + "loss": 0.085, + "step": 26430 + }, + { + "epoch": 0.2294337722762823, + "grad_norm": 0.3046875, + "learning_rate": 0.0017000764168353158, + "loss": 0.124, + "step": 26431 + }, + { + "epoch": 0.22944245275648648, + "grad_norm": 0.4453125, + "learning_rate": 0.0017000542351422651, + "loss": 0.1055, + "step": 26432 + }, + { + "epoch": 0.22945113323669064, + "grad_norm": 0.34765625, + "learning_rate": 0.0017000320527930039, + "loss": 0.1543, + "step": 26433 + }, + { + "epoch": 0.2294598137168948, + "grad_norm": 0.2255859375, + "learning_rate": 0.0017000098697875572, + "loss": 0.1113, + "step": 26434 + }, + { + "epoch": 0.22946849419709897, + "grad_norm": 0.1279296875, + "learning_rate": 0.0016999876861259484, + "loss": 0.1436, + "step": 26435 + }, + { + "epoch": 0.22947717467730314, + "grad_norm": 0.10546875, + "learning_rate": 0.0016999655018082023, + "loss": 0.1318, + "step": 26436 + }, + { + "epoch": 0.2294858551575073, + "grad_norm": 0.298828125, + "learning_rate": 0.0016999433168343428, + "loss": 0.1348, + "step": 26437 + }, + { + "epoch": 0.22949453563771147, + "grad_norm": 0.255859375, + "learning_rate": 0.0016999211312043942, + "loss": 0.1484, + "step": 26438 + }, + { + "epoch": 0.22950321611791563, + "grad_norm": 0.376953125, + "learning_rate": 0.0016998989449183808, + "loss": 0.0815, + "step": 26439 + }, + { + "epoch": 0.2295118965981198, + "grad_norm": 0.10595703125, + "learning_rate": 0.0016998767579763272, + "loss": 0.1104, + "step": 26440 + }, + { + "epoch": 0.22952057707832396, + "grad_norm": 0.51953125, + "learning_rate": 0.0016998545703782572, + "loss": 0.0698, + "step": 26441 + }, + { + "epoch": 0.22952925755852813, + "grad_norm": 0.337890625, + "learning_rate": 0.0016998323821241954, + "loss": 0.1162, + "step": 26442 + }, + { + "epoch": 0.2295379380387323, + "grad_norm": 1.0078125, + "learning_rate": 0.0016998101932141655, + "loss": 0.418, + "step": 26443 + }, + { + "epoch": 0.22954661851893646, + "grad_norm": 0.09228515625, + "learning_rate": 0.0016997880036481927, + "loss": 0.0806, + "step": 26444 + }, + { + "epoch": 0.22955529899914062, + "grad_norm": 0.162109375, + "learning_rate": 0.0016997658134263007, + "loss": 0.1494, + "step": 26445 + }, + { + "epoch": 0.2295639794793448, + "grad_norm": 0.25, + "learning_rate": 0.0016997436225485137, + "loss": 0.1084, + "step": 26446 + }, + { + "epoch": 0.22957265995954895, + "grad_norm": 0.37109375, + "learning_rate": 0.0016997214310148564, + "loss": 0.1416, + "step": 26447 + }, + { + "epoch": 0.22958134043975312, + "grad_norm": 0.232421875, + "learning_rate": 0.0016996992388253523, + "loss": 0.0767, + "step": 26448 + }, + { + "epoch": 0.22959002091995728, + "grad_norm": 0.1298828125, + "learning_rate": 0.001699677045980026, + "loss": 0.085, + "step": 26449 + }, + { + "epoch": 0.22959870140016145, + "grad_norm": 0.416015625, + "learning_rate": 0.0016996548524789025, + "loss": 0.0933, + "step": 26450 + }, + { + "epoch": 0.2296073818803656, + "grad_norm": 0.4296875, + "learning_rate": 0.001699632658322005, + "loss": 0.0693, + "step": 26451 + }, + { + "epoch": 0.22961606236056978, + "grad_norm": 0.111328125, + "learning_rate": 0.0016996104635093583, + "loss": 0.1309, + "step": 26452 + }, + { + "epoch": 0.22962474284077394, + "grad_norm": 1.140625, + "learning_rate": 0.0016995882680409867, + "loss": 0.1094, + "step": 26453 + }, + { + "epoch": 0.2296334233209781, + "grad_norm": 0.609375, + "learning_rate": 0.0016995660719169143, + "loss": 0.124, + "step": 26454 + }, + { + "epoch": 0.22964210380118227, + "grad_norm": 0.703125, + "learning_rate": 0.0016995438751371654, + "loss": 0.1299, + "step": 26455 + }, + { + "epoch": 0.22965078428138644, + "grad_norm": 0.6796875, + "learning_rate": 0.0016995216777017643, + "loss": 0.1289, + "step": 26456 + }, + { + "epoch": 0.2296594647615906, + "grad_norm": 0.291015625, + "learning_rate": 0.0016994994796107354, + "loss": 0.0879, + "step": 26457 + }, + { + "epoch": 0.22966814524179477, + "grad_norm": 0.53515625, + "learning_rate": 0.001699477280864103, + "loss": 0.0898, + "step": 26458 + }, + { + "epoch": 0.22967682572199893, + "grad_norm": 0.30859375, + "learning_rate": 0.0016994550814618912, + "loss": 0.0889, + "step": 26459 + }, + { + "epoch": 0.2296855062022031, + "grad_norm": 0.19140625, + "learning_rate": 0.0016994328814041242, + "loss": 0.0913, + "step": 26460 + }, + { + "epoch": 0.22969418668240726, + "grad_norm": 0.103515625, + "learning_rate": 0.0016994106806908266, + "loss": 0.1641, + "step": 26461 + }, + { + "epoch": 0.22970286716261143, + "grad_norm": 0.3046875, + "learning_rate": 0.0016993884793220224, + "loss": 0.0928, + "step": 26462 + }, + { + "epoch": 0.2297115476428156, + "grad_norm": 0.12451171875, + "learning_rate": 0.0016993662772977357, + "loss": 0.1309, + "step": 26463 + }, + { + "epoch": 0.22972022812301976, + "grad_norm": 0.3515625, + "learning_rate": 0.0016993440746179912, + "loss": 0.1143, + "step": 26464 + }, + { + "epoch": 0.22972890860322392, + "grad_norm": 0.2119140625, + "learning_rate": 0.0016993218712828134, + "loss": 0.0938, + "step": 26465 + }, + { + "epoch": 0.2297375890834281, + "grad_norm": 0.10986328125, + "learning_rate": 0.0016992996672922259, + "loss": 0.1543, + "step": 26466 + }, + { + "epoch": 0.22974626956363225, + "grad_norm": 0.28125, + "learning_rate": 0.0016992774626462532, + "loss": 0.0679, + "step": 26467 + }, + { + "epoch": 0.22975495004383642, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016992552573449199, + "loss": 0.0889, + "step": 26468 + }, + { + "epoch": 0.22976363052404059, + "grad_norm": 0.1845703125, + "learning_rate": 0.0016992330513882496, + "loss": 0.105, + "step": 26469 + }, + { + "epoch": 0.22977231100424475, + "grad_norm": 0.44140625, + "learning_rate": 0.0016992108447762674, + "loss": 0.1045, + "step": 26470 + }, + { + "epoch": 0.22978099148444892, + "grad_norm": 0.0791015625, + "learning_rate": 0.0016991886375089973, + "loss": 0.0903, + "step": 26471 + }, + { + "epoch": 0.22978967196465308, + "grad_norm": 0.25390625, + "learning_rate": 0.0016991664295864632, + "loss": 0.1025, + "step": 26472 + }, + { + "epoch": 0.22979835244485725, + "grad_norm": 0.458984375, + "learning_rate": 0.00169914422100869, + "loss": 0.1084, + "step": 26473 + }, + { + "epoch": 0.2298070329250614, + "grad_norm": 0.072265625, + "learning_rate": 0.0016991220117757017, + "loss": 0.0938, + "step": 26474 + }, + { + "epoch": 0.22981571340526558, + "grad_norm": 0.412109375, + "learning_rate": 0.0016990998018875222, + "loss": 0.1465, + "step": 26475 + }, + { + "epoch": 0.22982439388546974, + "grad_norm": 0.126953125, + "learning_rate": 0.0016990775913441768, + "loss": 0.0991, + "step": 26476 + }, + { + "epoch": 0.2298330743656739, + "grad_norm": 2.65625, + "learning_rate": 0.0016990553801456887, + "loss": 0.1279, + "step": 26477 + }, + { + "epoch": 0.22984175484587807, + "grad_norm": 0.7421875, + "learning_rate": 0.0016990331682920823, + "loss": 0.1367, + "step": 26478 + }, + { + "epoch": 0.22985043532608224, + "grad_norm": 0.29296875, + "learning_rate": 0.0016990109557833827, + "loss": 0.085, + "step": 26479 + }, + { + "epoch": 0.2298591158062864, + "grad_norm": 0.1650390625, + "learning_rate": 0.0016989887426196138, + "loss": 0.0869, + "step": 26480 + }, + { + "epoch": 0.22986779628649057, + "grad_norm": 0.11376953125, + "learning_rate": 0.0016989665288007995, + "loss": 0.1006, + "step": 26481 + }, + { + "epoch": 0.22987647676669473, + "grad_norm": 0.3671875, + "learning_rate": 0.0016989443143269648, + "loss": 0.0918, + "step": 26482 + }, + { + "epoch": 0.2298851572468989, + "grad_norm": 0.212890625, + "learning_rate": 0.0016989220991981328, + "loss": 0.0938, + "step": 26483 + }, + { + "epoch": 0.22989383772710306, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016988998834143296, + "loss": 0.0933, + "step": 26484 + }, + { + "epoch": 0.22990251820730723, + "grad_norm": 0.1923828125, + "learning_rate": 0.001698877666975578, + "loss": 0.082, + "step": 26485 + }, + { + "epoch": 0.2299111986875114, + "grad_norm": 0.4921875, + "learning_rate": 0.0016988554498819023, + "loss": 0.0903, + "step": 26486 + }, + { + "epoch": 0.22991987916771556, + "grad_norm": 0.65234375, + "learning_rate": 0.001698833232133328, + "loss": 0.1328, + "step": 26487 + }, + { + "epoch": 0.22992855964791972, + "grad_norm": 0.15625, + "learning_rate": 0.0016988110137298785, + "loss": 0.1387, + "step": 26488 + }, + { + "epoch": 0.2299372401281239, + "grad_norm": 0.17578125, + "learning_rate": 0.0016987887946715782, + "loss": 0.082, + "step": 26489 + }, + { + "epoch": 0.22994592060832805, + "grad_norm": 0.5859375, + "learning_rate": 0.0016987665749584515, + "loss": 0.0903, + "step": 26490 + }, + { + "epoch": 0.22995460108853222, + "grad_norm": 0.458984375, + "learning_rate": 0.0016987443545905226, + "loss": 0.0815, + "step": 26491 + }, + { + "epoch": 0.22996328156873638, + "grad_norm": 0.2470703125, + "learning_rate": 0.001698722133567816, + "loss": 0.1191, + "step": 26492 + }, + { + "epoch": 0.22997196204894055, + "grad_norm": 0.236328125, + "learning_rate": 0.0016986999118903559, + "loss": 0.0967, + "step": 26493 + }, + { + "epoch": 0.2299806425291447, + "grad_norm": 0.8515625, + "learning_rate": 0.0016986776895581666, + "loss": 0.0996, + "step": 26494 + }, + { + "epoch": 0.22998932300934888, + "grad_norm": 1.171875, + "learning_rate": 0.001698655466571272, + "loss": 0.2344, + "step": 26495 + }, + { + "epoch": 0.22999800348955304, + "grad_norm": 0.10302734375, + "learning_rate": 0.0016986332429296971, + "loss": 0.0957, + "step": 26496 + }, + { + "epoch": 0.2300066839697572, + "grad_norm": 0.166015625, + "learning_rate": 0.0016986110186334657, + "loss": 0.1025, + "step": 26497 + }, + { + "epoch": 0.23001536444996137, + "grad_norm": 0.984375, + "learning_rate": 0.0016985887936826025, + "loss": 0.2295, + "step": 26498 + }, + { + "epoch": 0.23002404493016554, + "grad_norm": 0.1083984375, + "learning_rate": 0.0016985665680771314, + "loss": 0.1147, + "step": 26499 + }, + { + "epoch": 0.2300327254103697, + "grad_norm": 0.08740234375, + "learning_rate": 0.001698544341817077, + "loss": 0.0879, + "step": 26500 + }, + { + "epoch": 0.23004140589057387, + "grad_norm": 0.466796875, + "learning_rate": 0.0016985221149024635, + "loss": 0.123, + "step": 26501 + }, + { + "epoch": 0.23005008637077803, + "grad_norm": 0.75390625, + "learning_rate": 0.0016984998873333153, + "loss": 0.1055, + "step": 26502 + }, + { + "epoch": 0.2300587668509822, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016984776591096565, + "loss": 0.1089, + "step": 26503 + }, + { + "epoch": 0.23006744733118636, + "grad_norm": 0.921875, + "learning_rate": 0.0016984554302315114, + "loss": 0.126, + "step": 26504 + }, + { + "epoch": 0.23007612781139053, + "grad_norm": 0.421875, + "learning_rate": 0.0016984332006989046, + "loss": 0.1104, + "step": 26505 + }, + { + "epoch": 0.2300848082915947, + "grad_norm": 0.2734375, + "learning_rate": 0.0016984109705118601, + "loss": 0.1436, + "step": 26506 + }, + { + "epoch": 0.23009348877179886, + "grad_norm": 0.5859375, + "learning_rate": 0.0016983887396704025, + "loss": 0.1172, + "step": 26507 + }, + { + "epoch": 0.23010216925200302, + "grad_norm": 0.3515625, + "learning_rate": 0.0016983665081745558, + "loss": 0.1602, + "step": 26508 + }, + { + "epoch": 0.2301108497322072, + "grad_norm": 0.21484375, + "learning_rate": 0.0016983442760243445, + "loss": 0.1006, + "step": 26509 + }, + { + "epoch": 0.23011953021241136, + "grad_norm": 0.58984375, + "learning_rate": 0.001698322043219793, + "loss": 0.1377, + "step": 26510 + }, + { + "epoch": 0.23012821069261552, + "grad_norm": 0.12060546875, + "learning_rate": 0.0016982998097609257, + "loss": 0.1104, + "step": 26511 + }, + { + "epoch": 0.23013689117281969, + "grad_norm": 0.07861328125, + "learning_rate": 0.0016982775756477664, + "loss": 0.0845, + "step": 26512 + }, + { + "epoch": 0.23014557165302385, + "grad_norm": 0.10595703125, + "learning_rate": 0.0016982553408803399, + "loss": 0.1562, + "step": 26513 + }, + { + "epoch": 0.23015425213322802, + "grad_norm": 0.34765625, + "learning_rate": 0.0016982331054586704, + "loss": 0.1143, + "step": 26514 + }, + { + "epoch": 0.23016293261343218, + "grad_norm": 0.07373046875, + "learning_rate": 0.001698210869382782, + "loss": 0.0825, + "step": 26515 + }, + { + "epoch": 0.23017161309363635, + "grad_norm": 0.82421875, + "learning_rate": 0.0016981886326526989, + "loss": 0.1465, + "step": 26516 + }, + { + "epoch": 0.2301802935738405, + "grad_norm": 0.30078125, + "learning_rate": 0.0016981663952684463, + "loss": 0.1445, + "step": 26517 + }, + { + "epoch": 0.23018897405404468, + "grad_norm": 0.3828125, + "learning_rate": 0.0016981441572300475, + "loss": 0.1138, + "step": 26518 + }, + { + "epoch": 0.23019765453424884, + "grad_norm": 0.83984375, + "learning_rate": 0.0016981219185375276, + "loss": 0.1611, + "step": 26519 + }, + { + "epoch": 0.230206335014453, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016980996791909103, + "loss": 0.1162, + "step": 26520 + }, + { + "epoch": 0.23021501549465717, + "grad_norm": 0.3359375, + "learning_rate": 0.0016980774391902199, + "loss": 0.1074, + "step": 26521 + }, + { + "epoch": 0.23022369597486134, + "grad_norm": 0.51171875, + "learning_rate": 0.0016980551985354814, + "loss": 0.2461, + "step": 26522 + }, + { + "epoch": 0.2302323764550655, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016980329572267188, + "loss": 0.1104, + "step": 26523 + }, + { + "epoch": 0.23024105693526967, + "grad_norm": 1.03125, + "learning_rate": 0.0016980107152639562, + "loss": 0.2695, + "step": 26524 + }, + { + "epoch": 0.23024973741547383, + "grad_norm": 0.51953125, + "learning_rate": 0.0016979884726472181, + "loss": 0.1104, + "step": 26525 + }, + { + "epoch": 0.230258417895678, + "grad_norm": 0.32421875, + "learning_rate": 0.0016979662293765286, + "loss": 0.1235, + "step": 26526 + }, + { + "epoch": 0.23026709837588216, + "grad_norm": 0.78515625, + "learning_rate": 0.0016979439854519125, + "loss": 0.1631, + "step": 26527 + }, + { + "epoch": 0.23027577885608633, + "grad_norm": 0.2001953125, + "learning_rate": 0.0016979217408733937, + "loss": 0.0811, + "step": 26528 + }, + { + "epoch": 0.2302844593362905, + "grad_norm": 1.1171875, + "learning_rate": 0.0016978994956409965, + "loss": 0.1289, + "step": 26529 + }, + { + "epoch": 0.23029313981649466, + "grad_norm": 1.4296875, + "learning_rate": 0.0016978772497547457, + "loss": 0.1191, + "step": 26530 + }, + { + "epoch": 0.23030182029669882, + "grad_norm": 0.392578125, + "learning_rate": 0.001697855003214665, + "loss": 0.123, + "step": 26531 + }, + { + "epoch": 0.230310500776903, + "grad_norm": 0.283203125, + "learning_rate": 0.0016978327560207794, + "loss": 0.1089, + "step": 26532 + }, + { + "epoch": 0.23031918125710715, + "grad_norm": 1.2734375, + "learning_rate": 0.0016978105081731129, + "loss": 0.1069, + "step": 26533 + }, + { + "epoch": 0.23032786173731132, + "grad_norm": 0.140625, + "learning_rate": 0.001697788259671689, + "loss": 0.1318, + "step": 26534 + }, + { + "epoch": 0.23033654221751548, + "grad_norm": 1.0390625, + "learning_rate": 0.0016977660105165337, + "loss": 0.103, + "step": 26535 + }, + { + "epoch": 0.23034522269771965, + "grad_norm": 0.189453125, + "learning_rate": 0.0016977437607076705, + "loss": 0.0718, + "step": 26536 + }, + { + "epoch": 0.2303539031779238, + "grad_norm": 0.18359375, + "learning_rate": 0.0016977215102451234, + "loss": 0.1738, + "step": 26537 + }, + { + "epoch": 0.23036258365812798, + "grad_norm": 0.57421875, + "learning_rate": 0.0016976992591289167, + "loss": 0.0952, + "step": 26538 + }, + { + "epoch": 0.23037126413833214, + "grad_norm": 0.140625, + "learning_rate": 0.0016976770073590757, + "loss": 0.1055, + "step": 26539 + }, + { + "epoch": 0.2303799446185363, + "grad_norm": 0.77734375, + "learning_rate": 0.0016976547549356238, + "loss": 0.0811, + "step": 26540 + }, + { + "epoch": 0.23038862509874047, + "grad_norm": 0.83984375, + "learning_rate": 0.0016976325018585856, + "loss": 0.0942, + "step": 26541 + }, + { + "epoch": 0.23039730557894464, + "grad_norm": 0.6953125, + "learning_rate": 0.0016976102481279856, + "loss": 0.123, + "step": 26542 + }, + { + "epoch": 0.2304059860591488, + "grad_norm": 0.74609375, + "learning_rate": 0.001697587993743848, + "loss": 0.1494, + "step": 26543 + }, + { + "epoch": 0.23041466653935297, + "grad_norm": 0.58203125, + "learning_rate": 0.001697565738706197, + "loss": 0.1074, + "step": 26544 + }, + { + "epoch": 0.23042334701955713, + "grad_norm": 0.1796875, + "learning_rate": 0.0016975434830150573, + "loss": 0.0996, + "step": 26545 + }, + { + "epoch": 0.2304320274997613, + "grad_norm": 0.130859375, + "learning_rate": 0.001697521226670453, + "loss": 0.1357, + "step": 26546 + }, + { + "epoch": 0.23044070797996546, + "grad_norm": 0.1640625, + "learning_rate": 0.0016974989696724083, + "loss": 0.0957, + "step": 26547 + }, + { + "epoch": 0.23044938846016963, + "grad_norm": 0.1240234375, + "learning_rate": 0.001697476712020948, + "loss": 0.104, + "step": 26548 + }, + { + "epoch": 0.2304580689403738, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016974544537160959, + "loss": 0.1416, + "step": 26549 + }, + { + "epoch": 0.23046674942057796, + "grad_norm": 0.6484375, + "learning_rate": 0.0016974321947578765, + "loss": 0.0879, + "step": 26550 + }, + { + "epoch": 0.23047542990078212, + "grad_norm": 0.50390625, + "learning_rate": 0.0016974099351463145, + "loss": 0.1641, + "step": 26551 + }, + { + "epoch": 0.23048411038098626, + "grad_norm": 0.130859375, + "learning_rate": 0.0016973876748814336, + "loss": 0.1387, + "step": 26552 + }, + { + "epoch": 0.23049279086119043, + "grad_norm": 0.58203125, + "learning_rate": 0.001697365413963259, + "loss": 0.1562, + "step": 26553 + }, + { + "epoch": 0.2305014713413946, + "grad_norm": 0.421875, + "learning_rate": 0.001697343152391814, + "loss": 0.1084, + "step": 26554 + }, + { + "epoch": 0.23051015182159876, + "grad_norm": 0.21484375, + "learning_rate": 0.0016973208901671242, + "loss": 0.0972, + "step": 26555 + }, + { + "epoch": 0.23051883230180292, + "grad_norm": 0.373046875, + "learning_rate": 0.0016972986272892128, + "loss": 0.1089, + "step": 26556 + }, + { + "epoch": 0.2305275127820071, + "grad_norm": 0.50390625, + "learning_rate": 0.0016972763637581045, + "loss": 0.1523, + "step": 26557 + }, + { + "epoch": 0.23053619326221125, + "grad_norm": 0.1640625, + "learning_rate": 0.0016972540995738238, + "loss": 0.0898, + "step": 26558 + }, + { + "epoch": 0.23054487374241542, + "grad_norm": 0.203125, + "learning_rate": 0.0016972318347363952, + "loss": 0.0806, + "step": 26559 + }, + { + "epoch": 0.23055355422261958, + "grad_norm": 0.490234375, + "learning_rate": 0.0016972095692458426, + "loss": 0.0801, + "step": 26560 + }, + { + "epoch": 0.23056223470282375, + "grad_norm": 0.7109375, + "learning_rate": 0.0016971873031021907, + "loss": 0.1182, + "step": 26561 + }, + { + "epoch": 0.2305709151830279, + "grad_norm": 0.17578125, + "learning_rate": 0.001697165036305464, + "loss": 0.106, + "step": 26562 + }, + { + "epoch": 0.23057959566323208, + "grad_norm": 1.0859375, + "learning_rate": 0.0016971427688556862, + "loss": 0.127, + "step": 26563 + }, + { + "epoch": 0.23058827614343624, + "grad_norm": 0.408203125, + "learning_rate": 0.0016971205007528822, + "loss": 0.1123, + "step": 26564 + }, + { + "epoch": 0.2305969566236404, + "grad_norm": 0.07275390625, + "learning_rate": 0.0016970982319970762, + "loss": 0.0894, + "step": 26565 + }, + { + "epoch": 0.23060563710384457, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016970759625882926, + "loss": 0.1729, + "step": 26566 + }, + { + "epoch": 0.23061431758404874, + "grad_norm": 0.42578125, + "learning_rate": 0.0016970536925265556, + "loss": 0.0903, + "step": 26567 + }, + { + "epoch": 0.2306229980642529, + "grad_norm": 0.1044921875, + "learning_rate": 0.0016970314218118896, + "loss": 0.1064, + "step": 26568 + }, + { + "epoch": 0.23063167854445707, + "grad_norm": 0.09765625, + "learning_rate": 0.001697009150444319, + "loss": 0.0947, + "step": 26569 + }, + { + "epoch": 0.23064035902466123, + "grad_norm": 0.37109375, + "learning_rate": 0.0016969868784238685, + "loss": 0.1211, + "step": 26570 + }, + { + "epoch": 0.2306490395048654, + "grad_norm": 0.8125, + "learning_rate": 0.0016969646057505617, + "loss": 0.1162, + "step": 26571 + }, + { + "epoch": 0.23065771998506956, + "grad_norm": 0.36328125, + "learning_rate": 0.0016969423324244236, + "loss": 0.1299, + "step": 26572 + }, + { + "epoch": 0.23066640046527373, + "grad_norm": 0.2099609375, + "learning_rate": 0.0016969200584454783, + "loss": 0.1104, + "step": 26573 + }, + { + "epoch": 0.2306750809454779, + "grad_norm": 0.29296875, + "learning_rate": 0.00169689778381375, + "loss": 0.1562, + "step": 26574 + }, + { + "epoch": 0.23068376142568206, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016968755085292634, + "loss": 0.1221, + "step": 26575 + }, + { + "epoch": 0.23069244190588623, + "grad_norm": 0.255859375, + "learning_rate": 0.001696853232592043, + "loss": 0.1084, + "step": 26576 + }, + { + "epoch": 0.2307011223860904, + "grad_norm": 0.291015625, + "learning_rate": 0.0016968309560021124, + "loss": 0.1191, + "step": 26577 + }, + { + "epoch": 0.23070980286629456, + "grad_norm": 0.189453125, + "learning_rate": 0.0016968086787594965, + "loss": 0.0835, + "step": 26578 + }, + { + "epoch": 0.23071848334649872, + "grad_norm": 0.51953125, + "learning_rate": 0.0016967864008642197, + "loss": 0.1123, + "step": 26579 + }, + { + "epoch": 0.23072716382670289, + "grad_norm": 0.42578125, + "learning_rate": 0.0016967641223163065, + "loss": 0.0864, + "step": 26580 + }, + { + "epoch": 0.23073584430690705, + "grad_norm": 0.06103515625, + "learning_rate": 0.0016967418431157808, + "loss": 0.0796, + "step": 26581 + }, + { + "epoch": 0.23074452478711122, + "grad_norm": 0.080078125, + "learning_rate": 0.0016967195632626674, + "loss": 0.082, + "step": 26582 + }, + { + "epoch": 0.23075320526731538, + "grad_norm": 0.32421875, + "learning_rate": 0.00169669728275699, + "loss": 0.1138, + "step": 26583 + }, + { + "epoch": 0.23076188574751955, + "grad_norm": 0.59765625, + "learning_rate": 0.0016966750015987736, + "loss": 0.0957, + "step": 26584 + }, + { + "epoch": 0.2307705662277237, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016966527197880429, + "loss": 0.0762, + "step": 26585 + }, + { + "epoch": 0.23077924670792788, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016966304373248209, + "loss": 0.1348, + "step": 26586 + }, + { + "epoch": 0.23078792718813204, + "grad_norm": 0.255859375, + "learning_rate": 0.0016966081542091334, + "loss": 0.0918, + "step": 26587 + }, + { + "epoch": 0.2307966076683362, + "grad_norm": 0.28515625, + "learning_rate": 0.001696585870441004, + "loss": 0.1157, + "step": 26588 + }, + { + "epoch": 0.23080528814854037, + "grad_norm": 0.283203125, + "learning_rate": 0.001696563586020457, + "loss": 0.1436, + "step": 26589 + }, + { + "epoch": 0.23081396862874454, + "grad_norm": 0.45703125, + "learning_rate": 0.0016965413009475173, + "loss": 0.106, + "step": 26590 + }, + { + "epoch": 0.2308226491089487, + "grad_norm": 0.337890625, + "learning_rate": 0.001696519015222209, + "loss": 0.1138, + "step": 26591 + }, + { + "epoch": 0.23083132958915287, + "grad_norm": 0.8125, + "learning_rate": 0.0016964967288445566, + "loss": 0.0879, + "step": 26592 + }, + { + "epoch": 0.23084001006935703, + "grad_norm": 0.08251953125, + "learning_rate": 0.0016964744418145842, + "loss": 0.1035, + "step": 26593 + }, + { + "epoch": 0.2308486905495612, + "grad_norm": 0.65625, + "learning_rate": 0.0016964521541323163, + "loss": 0.1514, + "step": 26594 + }, + { + "epoch": 0.23085737102976536, + "grad_norm": 0.5703125, + "learning_rate": 0.0016964298657977772, + "loss": 0.1138, + "step": 26595 + }, + { + "epoch": 0.23086605150996953, + "grad_norm": 0.251953125, + "learning_rate": 0.0016964075768109914, + "loss": 0.1289, + "step": 26596 + }, + { + "epoch": 0.2308747319901737, + "grad_norm": 0.12890625, + "learning_rate": 0.0016963852871719831, + "loss": 0.0938, + "step": 26597 + }, + { + "epoch": 0.23088341247037786, + "grad_norm": 0.6796875, + "learning_rate": 0.0016963629968807772, + "loss": 0.1147, + "step": 26598 + }, + { + "epoch": 0.23089209295058202, + "grad_norm": 0.08837890625, + "learning_rate": 0.0016963407059373975, + "loss": 0.0908, + "step": 26599 + }, + { + "epoch": 0.2309007734307862, + "grad_norm": 0.3125, + "learning_rate": 0.0016963184143418686, + "loss": 0.0962, + "step": 26600 + }, + { + "epoch": 0.23090945391099035, + "grad_norm": 0.19140625, + "learning_rate": 0.0016962961220942149, + "loss": 0.1084, + "step": 26601 + }, + { + "epoch": 0.23091813439119452, + "grad_norm": 0.07568359375, + "learning_rate": 0.0016962738291944606, + "loss": 0.1025, + "step": 26602 + }, + { + "epoch": 0.23092681487139868, + "grad_norm": 0.515625, + "learning_rate": 0.00169625153564263, + "loss": 0.1396, + "step": 26603 + }, + { + "epoch": 0.23093549535160285, + "grad_norm": 0.302734375, + "learning_rate": 0.0016962292414387477, + "loss": 0.0918, + "step": 26604 + }, + { + "epoch": 0.230944175831807, + "grad_norm": 0.45703125, + "learning_rate": 0.0016962069465828384, + "loss": 0.1211, + "step": 26605 + }, + { + "epoch": 0.23095285631201118, + "grad_norm": 0.2265625, + "learning_rate": 0.001696184651074926, + "loss": 0.1133, + "step": 26606 + }, + { + "epoch": 0.23096153679221534, + "grad_norm": 0.2109375, + "learning_rate": 0.001696162354915035, + "loss": 0.1289, + "step": 26607 + }, + { + "epoch": 0.2309702172724195, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016961400581031899, + "loss": 0.1011, + "step": 26608 + }, + { + "epoch": 0.23097889775262367, + "grad_norm": 0.27734375, + "learning_rate": 0.001696117760639415, + "loss": 0.0859, + "step": 26609 + }, + { + "epoch": 0.23098757823282784, + "grad_norm": 0.287109375, + "learning_rate": 0.0016960954625237343, + "loss": 0.0859, + "step": 26610 + }, + { + "epoch": 0.230996258713032, + "grad_norm": 0.2275390625, + "learning_rate": 0.001696073163756173, + "loss": 0.1191, + "step": 26611 + }, + { + "epoch": 0.23100493919323617, + "grad_norm": 0.0810546875, + "learning_rate": 0.0016960508643367547, + "loss": 0.0854, + "step": 26612 + }, + { + "epoch": 0.23101361967344033, + "grad_norm": 0.216796875, + "learning_rate": 0.0016960285642655046, + "loss": 0.0918, + "step": 26613 + }, + { + "epoch": 0.2310223001536445, + "grad_norm": 2.125, + "learning_rate": 0.0016960062635424464, + "loss": 0.2217, + "step": 26614 + }, + { + "epoch": 0.23103098063384866, + "grad_norm": 0.119140625, + "learning_rate": 0.0016959839621676046, + "loss": 0.1172, + "step": 26615 + }, + { + "epoch": 0.23103966111405283, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016959616601410037, + "loss": 0.1348, + "step": 26616 + }, + { + "epoch": 0.231048341594257, + "grad_norm": 0.1103515625, + "learning_rate": 0.001695939357462668, + "loss": 0.1138, + "step": 26617 + }, + { + "epoch": 0.23105702207446116, + "grad_norm": 0.17578125, + "learning_rate": 0.001695917054132622, + "loss": 0.1016, + "step": 26618 + }, + { + "epoch": 0.23106570255466533, + "grad_norm": 0.71875, + "learning_rate": 0.0016958947501508903, + "loss": 0.1348, + "step": 26619 + }, + { + "epoch": 0.2310743830348695, + "grad_norm": 0.4140625, + "learning_rate": 0.0016958724455174968, + "loss": 0.0977, + "step": 26620 + }, + { + "epoch": 0.23108306351507366, + "grad_norm": 0.52734375, + "learning_rate": 0.0016958501402324663, + "loss": 0.0845, + "step": 26621 + }, + { + "epoch": 0.23109174399527782, + "grad_norm": 1.0, + "learning_rate": 0.001695827834295823, + "loss": 0.1367, + "step": 26622 + }, + { + "epoch": 0.23110042447548199, + "grad_norm": 1.015625, + "learning_rate": 0.0016958055277075913, + "loss": 0.1514, + "step": 26623 + }, + { + "epoch": 0.23110910495568615, + "grad_norm": 0.298828125, + "learning_rate": 0.0016957832204677957, + "loss": 0.1064, + "step": 26624 + }, + { + "epoch": 0.23111778543589032, + "grad_norm": 0.859375, + "learning_rate": 0.0016957609125764603, + "loss": 0.1162, + "step": 26625 + }, + { + "epoch": 0.23112646591609448, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016957386040336098, + "loss": 0.1201, + "step": 26626 + }, + { + "epoch": 0.23113514639629865, + "grad_norm": 0.26171875, + "learning_rate": 0.0016957162948392688, + "loss": 0.1143, + "step": 26627 + }, + { + "epoch": 0.2311438268765028, + "grad_norm": 0.14453125, + "learning_rate": 0.0016956939849934612, + "loss": 0.083, + "step": 26628 + }, + { + "epoch": 0.23115250735670698, + "grad_norm": 0.265625, + "learning_rate": 0.0016956716744962114, + "loss": 0.1006, + "step": 26629 + }, + { + "epoch": 0.23116118783691114, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016956493633475446, + "loss": 0.125, + "step": 26630 + }, + { + "epoch": 0.2311698683171153, + "grad_norm": 0.98828125, + "learning_rate": 0.0016956270515474838, + "loss": 0.1113, + "step": 26631 + }, + { + "epoch": 0.23117854879731947, + "grad_norm": 0.2734375, + "learning_rate": 0.0016956047390960546, + "loss": 0.103, + "step": 26632 + }, + { + "epoch": 0.23118722927752364, + "grad_norm": 0.1748046875, + "learning_rate": 0.0016955824259932812, + "loss": 0.0869, + "step": 26633 + }, + { + "epoch": 0.2311959097577278, + "grad_norm": 0.09912109375, + "learning_rate": 0.0016955601122391878, + "loss": 0.1016, + "step": 26634 + }, + { + "epoch": 0.23120459023793197, + "grad_norm": 0.4453125, + "learning_rate": 0.0016955377978337984, + "loss": 0.1089, + "step": 26635 + }, + { + "epoch": 0.23121327071813613, + "grad_norm": 0.55859375, + "learning_rate": 0.0016955154827771384, + "loss": 0.0957, + "step": 26636 + }, + { + "epoch": 0.2312219511983403, + "grad_norm": 0.5, + "learning_rate": 0.0016954931670692312, + "loss": 0.126, + "step": 26637 + }, + { + "epoch": 0.23123063167854446, + "grad_norm": 0.1552734375, + "learning_rate": 0.0016954708507101015, + "loss": 0.1035, + "step": 26638 + }, + { + "epoch": 0.23123931215874863, + "grad_norm": 0.10888671875, + "learning_rate": 0.001695448533699774, + "loss": 0.1006, + "step": 26639 + }, + { + "epoch": 0.2312479926389528, + "grad_norm": 0.07568359375, + "learning_rate": 0.001695426216038273, + "loss": 0.0845, + "step": 26640 + }, + { + "epoch": 0.23125667311915696, + "grad_norm": 0.220703125, + "learning_rate": 0.0016954038977256228, + "loss": 0.1006, + "step": 26641 + }, + { + "epoch": 0.23126535359936112, + "grad_norm": 0.392578125, + "learning_rate": 0.0016953815787618478, + "loss": 0.0991, + "step": 26642 + }, + { + "epoch": 0.2312740340795653, + "grad_norm": 0.0654296875, + "learning_rate": 0.0016953592591469726, + "loss": 0.0815, + "step": 26643 + }, + { + "epoch": 0.23128271455976945, + "grad_norm": 0.0908203125, + "learning_rate": 0.0016953369388810213, + "loss": 0.1104, + "step": 26644 + }, + { + "epoch": 0.23129139503997362, + "grad_norm": 0.21875, + "learning_rate": 0.0016953146179640186, + "loss": 0.0962, + "step": 26645 + }, + { + "epoch": 0.23130007552017778, + "grad_norm": 0.1416015625, + "learning_rate": 0.0016952922963959884, + "loss": 0.1221, + "step": 26646 + }, + { + "epoch": 0.23130875600038195, + "grad_norm": 0.4140625, + "learning_rate": 0.001695269974176956, + "loss": 0.1484, + "step": 26647 + }, + { + "epoch": 0.2313174364805861, + "grad_norm": 1.234375, + "learning_rate": 0.001695247651306945, + "loss": 0.0933, + "step": 26648 + }, + { + "epoch": 0.23132611696079028, + "grad_norm": 0.1162109375, + "learning_rate": 0.0016952253277859803, + "loss": 0.0791, + "step": 26649 + }, + { + "epoch": 0.23133479744099444, + "grad_norm": 0.1044921875, + "learning_rate": 0.001695203003614086, + "loss": 0.165, + "step": 26650 + }, + { + "epoch": 0.2313434779211986, + "grad_norm": 0.1640625, + "learning_rate": 0.0016951806787912868, + "loss": 0.0815, + "step": 26651 + }, + { + "epoch": 0.23135215840140277, + "grad_norm": 0.53515625, + "learning_rate": 0.0016951583533176067, + "loss": 0.1191, + "step": 26652 + }, + { + "epoch": 0.23136083888160694, + "grad_norm": 0.103515625, + "learning_rate": 0.0016951360271930708, + "loss": 0.1157, + "step": 26653 + }, + { + "epoch": 0.2313695193618111, + "grad_norm": 0.55078125, + "learning_rate": 0.0016951137004177024, + "loss": 0.1289, + "step": 26654 + }, + { + "epoch": 0.23137819984201527, + "grad_norm": 0.5234375, + "learning_rate": 0.001695091372991527, + "loss": 0.123, + "step": 26655 + }, + { + "epoch": 0.23138688032221943, + "grad_norm": 1.1328125, + "learning_rate": 0.0016950690449145686, + "loss": 0.0981, + "step": 26656 + }, + { + "epoch": 0.2313955608024236, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016950467161868518, + "loss": 0.1123, + "step": 26657 + }, + { + "epoch": 0.23140424128262777, + "grad_norm": 0.43359375, + "learning_rate": 0.0016950243868084005, + "loss": 0.127, + "step": 26658 + }, + { + "epoch": 0.23141292176283193, + "grad_norm": 0.65234375, + "learning_rate": 0.0016950020567792397, + "loss": 0.0869, + "step": 26659 + }, + { + "epoch": 0.2314216022430361, + "grad_norm": 0.29296875, + "learning_rate": 0.0016949797260993934, + "loss": 0.123, + "step": 26660 + }, + { + "epoch": 0.23143028272324026, + "grad_norm": 0.3203125, + "learning_rate": 0.0016949573947688864, + "loss": 0.123, + "step": 26661 + }, + { + "epoch": 0.23143896320344443, + "grad_norm": 1.125, + "learning_rate": 0.001694935062787743, + "loss": 0.1133, + "step": 26662 + }, + { + "epoch": 0.2314476436836486, + "grad_norm": 0.25390625, + "learning_rate": 0.0016949127301559876, + "loss": 0.0996, + "step": 26663 + }, + { + "epoch": 0.23145632416385276, + "grad_norm": 0.34765625, + "learning_rate": 0.001694890396873644, + "loss": 0.1113, + "step": 26664 + }, + { + "epoch": 0.23146500464405692, + "grad_norm": 0.166015625, + "learning_rate": 0.0016948680629407377, + "loss": 0.1055, + "step": 26665 + }, + { + "epoch": 0.23147368512426109, + "grad_norm": 0.216796875, + "learning_rate": 0.0016948457283572926, + "loss": 0.1133, + "step": 26666 + }, + { + "epoch": 0.23148236560446525, + "grad_norm": 0.16796875, + "learning_rate": 0.001694823393123333, + "loss": 0.0996, + "step": 26667 + }, + { + "epoch": 0.23149104608466942, + "grad_norm": 0.451171875, + "learning_rate": 0.0016948010572388835, + "loss": 0.0938, + "step": 26668 + }, + { + "epoch": 0.23149972656487358, + "grad_norm": 0.5078125, + "learning_rate": 0.0016947787207039686, + "loss": 0.1104, + "step": 26669 + }, + { + "epoch": 0.23150840704507775, + "grad_norm": 0.259765625, + "learning_rate": 0.0016947563835186124, + "loss": 0.0874, + "step": 26670 + }, + { + "epoch": 0.2315170875252819, + "grad_norm": 0.068359375, + "learning_rate": 0.00169473404568284, + "loss": 0.0698, + "step": 26671 + }, + { + "epoch": 0.23152576800548608, + "grad_norm": 0.59765625, + "learning_rate": 0.001694711707196675, + "loss": 0.2002, + "step": 26672 + }, + { + "epoch": 0.23153444848569024, + "grad_norm": 0.30078125, + "learning_rate": 0.0016946893680601424, + "loss": 0.0864, + "step": 26673 + }, + { + "epoch": 0.2315431289658944, + "grad_norm": 0.302734375, + "learning_rate": 0.0016946670282732663, + "loss": 0.1104, + "step": 26674 + }, + { + "epoch": 0.23155180944609854, + "grad_norm": 0.203125, + "learning_rate": 0.0016946446878360712, + "loss": 0.085, + "step": 26675 + }, + { + "epoch": 0.2315604899263027, + "grad_norm": 0.11572265625, + "learning_rate": 0.001694622346748582, + "loss": 0.1128, + "step": 26676 + }, + { + "epoch": 0.23156917040650687, + "grad_norm": 0.55859375, + "learning_rate": 0.0016946000050108224, + "loss": 0.1123, + "step": 26677 + }, + { + "epoch": 0.23157785088671104, + "grad_norm": 0.126953125, + "learning_rate": 0.0016945776626228171, + "loss": 0.1299, + "step": 26678 + }, + { + "epoch": 0.2315865313669152, + "grad_norm": 0.51953125, + "learning_rate": 0.001694555319584591, + "loss": 0.1094, + "step": 26679 + }, + { + "epoch": 0.23159521184711937, + "grad_norm": 0.08544921875, + "learning_rate": 0.0016945329758961676, + "loss": 0.1172, + "step": 26680 + }, + { + "epoch": 0.23160389232732354, + "grad_norm": 0.43359375, + "learning_rate": 0.0016945106315575723, + "loss": 0.0835, + "step": 26681 + }, + { + "epoch": 0.2316125728075277, + "grad_norm": 0.08935546875, + "learning_rate": 0.0016944882865688289, + "loss": 0.1152, + "step": 26682 + }, + { + "epoch": 0.23162125328773187, + "grad_norm": 0.0986328125, + "learning_rate": 0.0016944659409299618, + "loss": 0.1025, + "step": 26683 + }, + { + "epoch": 0.23162993376793603, + "grad_norm": 0.51953125, + "learning_rate": 0.0016944435946409965, + "loss": 0.1182, + "step": 26684 + }, + { + "epoch": 0.2316386142481402, + "grad_norm": 0.8125, + "learning_rate": 0.0016944212477019557, + "loss": 0.123, + "step": 26685 + }, + { + "epoch": 0.23164729472834436, + "grad_norm": 0.61328125, + "learning_rate": 0.0016943989001128655, + "loss": 0.0938, + "step": 26686 + }, + { + "epoch": 0.23165597520854853, + "grad_norm": 0.2734375, + "learning_rate": 0.001694376551873749, + "loss": 0.0957, + "step": 26687 + }, + { + "epoch": 0.2316646556887527, + "grad_norm": 0.52734375, + "learning_rate": 0.0016943542029846314, + "loss": 0.1143, + "step": 26688 + }, + { + "epoch": 0.23167333616895686, + "grad_norm": 0.201171875, + "learning_rate": 0.0016943318534455374, + "loss": 0.1172, + "step": 26689 + }, + { + "epoch": 0.23168201664916102, + "grad_norm": 0.259765625, + "learning_rate": 0.0016943095032564903, + "loss": 0.1221, + "step": 26690 + }, + { + "epoch": 0.2316906971293652, + "grad_norm": 0.486328125, + "learning_rate": 0.001694287152417516, + "loss": 0.1328, + "step": 26691 + }, + { + "epoch": 0.23169937760956935, + "grad_norm": 0.365234375, + "learning_rate": 0.0016942648009286377, + "loss": 0.1016, + "step": 26692 + }, + { + "epoch": 0.23170805808977352, + "grad_norm": 0.66796875, + "learning_rate": 0.0016942424487898804, + "loss": 0.1172, + "step": 26693 + }, + { + "epoch": 0.23171673856997768, + "grad_norm": 0.7109375, + "learning_rate": 0.0016942200960012685, + "loss": 0.1143, + "step": 26694 + }, + { + "epoch": 0.23172541905018185, + "grad_norm": 0.515625, + "learning_rate": 0.0016941977425628265, + "loss": 0.0957, + "step": 26695 + }, + { + "epoch": 0.231734099530386, + "grad_norm": 0.49609375, + "learning_rate": 0.0016941753884745787, + "loss": 0.1152, + "step": 26696 + }, + { + "epoch": 0.23174278001059018, + "grad_norm": 0.31640625, + "learning_rate": 0.0016941530337365497, + "loss": 0.1465, + "step": 26697 + }, + { + "epoch": 0.23175146049079434, + "grad_norm": 0.82421875, + "learning_rate": 0.0016941306783487637, + "loss": 0.1445, + "step": 26698 + }, + { + "epoch": 0.2317601409709985, + "grad_norm": 0.279296875, + "learning_rate": 0.0016941083223112458, + "loss": 0.1855, + "step": 26699 + }, + { + "epoch": 0.23176882145120267, + "grad_norm": 0.125, + "learning_rate": 0.0016940859656240195, + "loss": 0.1426, + "step": 26700 + }, + { + "epoch": 0.23177750193140684, + "grad_norm": 0.349609375, + "learning_rate": 0.0016940636082871101, + "loss": 0.0791, + "step": 26701 + }, + { + "epoch": 0.231786182411611, + "grad_norm": 0.134765625, + "learning_rate": 0.0016940412503005412, + "loss": 0.1084, + "step": 26702 + }, + { + "epoch": 0.23179486289181517, + "grad_norm": 1.265625, + "learning_rate": 0.0016940188916643377, + "loss": 0.1089, + "step": 26703 + }, + { + "epoch": 0.23180354337201933, + "grad_norm": 0.2490234375, + "learning_rate": 0.0016939965323785246, + "loss": 0.127, + "step": 26704 + }, + { + "epoch": 0.2318122238522235, + "grad_norm": 0.5625, + "learning_rate": 0.0016939741724431255, + "loss": 0.1221, + "step": 26705 + }, + { + "epoch": 0.23182090433242766, + "grad_norm": 0.1484375, + "learning_rate": 0.001693951811858165, + "loss": 0.0928, + "step": 26706 + }, + { + "epoch": 0.23182958481263183, + "grad_norm": 0.1533203125, + "learning_rate": 0.001693929450623668, + "loss": 0.1147, + "step": 26707 + }, + { + "epoch": 0.231838265292836, + "grad_norm": 0.228515625, + "learning_rate": 0.0016939070887396583, + "loss": 0.0986, + "step": 26708 + }, + { + "epoch": 0.23184694577304016, + "grad_norm": 0.4140625, + "learning_rate": 0.0016938847262061613, + "loss": 0.1387, + "step": 26709 + }, + { + "epoch": 0.23185562625324432, + "grad_norm": 0.1904296875, + "learning_rate": 0.0016938623630232007, + "loss": 0.0977, + "step": 26710 + }, + { + "epoch": 0.2318643067334485, + "grad_norm": 0.52734375, + "learning_rate": 0.001693839999190801, + "loss": 0.1035, + "step": 26711 + }, + { + "epoch": 0.23187298721365265, + "grad_norm": 0.15234375, + "learning_rate": 0.0016938176347089866, + "loss": 0.1328, + "step": 26712 + }, + { + "epoch": 0.23188166769385682, + "grad_norm": 0.2392578125, + "learning_rate": 0.0016937952695777825, + "loss": 0.1738, + "step": 26713 + }, + { + "epoch": 0.23189034817406098, + "grad_norm": 0.60546875, + "learning_rate": 0.0016937729037972127, + "loss": 0.1328, + "step": 26714 + }, + { + "epoch": 0.23189902865426515, + "grad_norm": 0.64453125, + "learning_rate": 0.0016937505373673019, + "loss": 0.1118, + "step": 26715 + }, + { + "epoch": 0.23190770913446931, + "grad_norm": 0.67578125, + "learning_rate": 0.0016937281702880743, + "loss": 0.1357, + "step": 26716 + }, + { + "epoch": 0.23191638961467348, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016937058025595542, + "loss": 0.1123, + "step": 26717 + }, + { + "epoch": 0.23192507009487764, + "grad_norm": 1.03125, + "learning_rate": 0.001693683434181767, + "loss": 0.1035, + "step": 26718 + }, + { + "epoch": 0.2319337505750818, + "grad_norm": 0.310546875, + "learning_rate": 0.0016936610651547362, + "loss": 0.123, + "step": 26719 + }, + { + "epoch": 0.23194243105528597, + "grad_norm": 0.490234375, + "learning_rate": 0.0016936386954784863, + "loss": 0.1055, + "step": 26720 + }, + { + "epoch": 0.23195111153549014, + "grad_norm": 0.23046875, + "learning_rate": 0.0016936163251530425, + "loss": 0.1191, + "step": 26721 + }, + { + "epoch": 0.2319597920156943, + "grad_norm": 0.50390625, + "learning_rate": 0.0016935939541784287, + "loss": 0.1094, + "step": 26722 + }, + { + "epoch": 0.23196847249589847, + "grad_norm": 0.1376953125, + "learning_rate": 0.001693571582554669, + "loss": 0.0674, + "step": 26723 + }, + { + "epoch": 0.23197715297610264, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016935492102817889, + "loss": 0.0981, + "step": 26724 + }, + { + "epoch": 0.2319858334563068, + "grad_norm": 0.404296875, + "learning_rate": 0.001693526837359812, + "loss": 0.1348, + "step": 26725 + }, + { + "epoch": 0.23199451393651097, + "grad_norm": 0.37890625, + "learning_rate": 0.0016935044637887632, + "loss": 0.124, + "step": 26726 + }, + { + "epoch": 0.23200319441671513, + "grad_norm": 0.1240234375, + "learning_rate": 0.001693482089568667, + "loss": 0.0815, + "step": 26727 + }, + { + "epoch": 0.2320118748969193, + "grad_norm": 0.703125, + "learning_rate": 0.0016934597146995476, + "loss": 0.0859, + "step": 26728 + }, + { + "epoch": 0.23202055537712346, + "grad_norm": 1.0390625, + "learning_rate": 0.0016934373391814296, + "loss": 0.1191, + "step": 26729 + }, + { + "epoch": 0.23202923585732763, + "grad_norm": 0.126953125, + "learning_rate": 0.001693414963014337, + "loss": 0.1602, + "step": 26730 + }, + { + "epoch": 0.2320379163375318, + "grad_norm": 0.236328125, + "learning_rate": 0.0016933925861982951, + "loss": 0.0977, + "step": 26731 + }, + { + "epoch": 0.23204659681773596, + "grad_norm": 0.416015625, + "learning_rate": 0.0016933702087333277, + "loss": 0.1094, + "step": 26732 + }, + { + "epoch": 0.23205527729794012, + "grad_norm": 0.45703125, + "learning_rate": 0.00169334783061946, + "loss": 0.0986, + "step": 26733 + }, + { + "epoch": 0.2320639577781443, + "grad_norm": 0.17578125, + "learning_rate": 0.0016933254518567158, + "loss": 0.0674, + "step": 26734 + }, + { + "epoch": 0.23207263825834845, + "grad_norm": 0.828125, + "learning_rate": 0.00169330307244512, + "loss": 0.1855, + "step": 26735 + }, + { + "epoch": 0.23208131873855262, + "grad_norm": 0.283203125, + "learning_rate": 0.0016932806923846966, + "loss": 0.0996, + "step": 26736 + }, + { + "epoch": 0.23208999921875678, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016932583116754704, + "loss": 0.0781, + "step": 26737 + }, + { + "epoch": 0.23209867969896095, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016932359303174658, + "loss": 0.1465, + "step": 26738 + }, + { + "epoch": 0.2321073601791651, + "grad_norm": 0.1708984375, + "learning_rate": 0.0016932135483107074, + "loss": 0.1318, + "step": 26739 + }, + { + "epoch": 0.23211604065936928, + "grad_norm": 0.154296875, + "learning_rate": 0.0016931911656552197, + "loss": 0.1338, + "step": 26740 + }, + { + "epoch": 0.23212472113957344, + "grad_norm": 0.486328125, + "learning_rate": 0.0016931687823510267, + "loss": 0.1123, + "step": 26741 + }, + { + "epoch": 0.2321334016197776, + "grad_norm": 1.28125, + "learning_rate": 0.0016931463983981534, + "loss": 0.126, + "step": 26742 + }, + { + "epoch": 0.23214208209998177, + "grad_norm": 0.310546875, + "learning_rate": 0.0016931240137966243, + "loss": 0.0981, + "step": 26743 + }, + { + "epoch": 0.23215076258018594, + "grad_norm": 0.734375, + "learning_rate": 0.0016931016285464636, + "loss": 0.1099, + "step": 26744 + }, + { + "epoch": 0.2321594430603901, + "grad_norm": 0.2265625, + "learning_rate": 0.0016930792426476957, + "loss": 0.0874, + "step": 26745 + }, + { + "epoch": 0.23216812354059427, + "grad_norm": 0.12890625, + "learning_rate": 0.0016930568561003454, + "loss": 0.1289, + "step": 26746 + }, + { + "epoch": 0.23217680402079843, + "grad_norm": 0.30078125, + "learning_rate": 0.0016930344689044367, + "loss": 0.1162, + "step": 26747 + }, + { + "epoch": 0.2321854845010026, + "grad_norm": 0.09765625, + "learning_rate": 0.0016930120810599947, + "loss": 0.1182, + "step": 26748 + }, + { + "epoch": 0.23219416498120676, + "grad_norm": 0.1640625, + "learning_rate": 0.0016929896925670439, + "loss": 0.0986, + "step": 26749 + }, + { + "epoch": 0.23220284546141093, + "grad_norm": 0.271484375, + "learning_rate": 0.001692967303425608, + "loss": 0.1196, + "step": 26750 + }, + { + "epoch": 0.2322115259416151, + "grad_norm": 0.2490234375, + "learning_rate": 0.001692944913635712, + "loss": 0.1006, + "step": 26751 + }, + { + "epoch": 0.23222020642181926, + "grad_norm": 0.09228515625, + "learning_rate": 0.0016929225231973805, + "loss": 0.127, + "step": 26752 + }, + { + "epoch": 0.23222888690202342, + "grad_norm": 0.3828125, + "learning_rate": 0.0016929001321106382, + "loss": 0.1123, + "step": 26753 + }, + { + "epoch": 0.2322375673822276, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016928777403755085, + "loss": 0.1084, + "step": 26754 + }, + { + "epoch": 0.23224624786243175, + "grad_norm": 0.1630859375, + "learning_rate": 0.001692855347992017, + "loss": 0.1084, + "step": 26755 + }, + { + "epoch": 0.23225492834263592, + "grad_norm": 0.2109375, + "learning_rate": 0.0016928329549601878, + "loss": 0.1025, + "step": 26756 + }, + { + "epoch": 0.23226360882284008, + "grad_norm": 0.30078125, + "learning_rate": 0.0016928105612800453, + "loss": 0.0938, + "step": 26757 + }, + { + "epoch": 0.23227228930304425, + "grad_norm": 0.2392578125, + "learning_rate": 0.0016927881669516144, + "loss": 0.1328, + "step": 26758 + }, + { + "epoch": 0.23228096978324841, + "grad_norm": 0.1220703125, + "learning_rate": 0.0016927657719749188, + "loss": 0.1299, + "step": 26759 + }, + { + "epoch": 0.23228965026345258, + "grad_norm": 0.36328125, + "learning_rate": 0.0016927433763499838, + "loss": 0.0967, + "step": 26760 + }, + { + "epoch": 0.23229833074365674, + "grad_norm": 0.0859375, + "learning_rate": 0.0016927209800768333, + "loss": 0.0894, + "step": 26761 + }, + { + "epoch": 0.2323070112238609, + "grad_norm": 0.34765625, + "learning_rate": 0.0016926985831554925, + "loss": 0.1289, + "step": 26762 + }, + { + "epoch": 0.23231569170406507, + "grad_norm": 0.2265625, + "learning_rate": 0.0016926761855859847, + "loss": 0.1211, + "step": 26763 + }, + { + "epoch": 0.23232437218426924, + "grad_norm": 0.0986328125, + "learning_rate": 0.0016926537873683354, + "loss": 0.1021, + "step": 26764 + }, + { + "epoch": 0.2323330526644734, + "grad_norm": 0.30078125, + "learning_rate": 0.0016926313885025689, + "loss": 0.1406, + "step": 26765 + }, + { + "epoch": 0.23234173314467757, + "grad_norm": 0.5, + "learning_rate": 0.00169260898898871, + "loss": 0.1494, + "step": 26766 + }, + { + "epoch": 0.23235041362488174, + "grad_norm": 0.2578125, + "learning_rate": 0.001692586588826782, + "loss": 0.168, + "step": 26767 + }, + { + "epoch": 0.2323590941050859, + "grad_norm": 0.14453125, + "learning_rate": 0.0016925641880168108, + "loss": 0.1543, + "step": 26768 + }, + { + "epoch": 0.23236777458529007, + "grad_norm": 0.10302734375, + "learning_rate": 0.0016925417865588201, + "loss": 0.0957, + "step": 26769 + }, + { + "epoch": 0.23237645506549423, + "grad_norm": 0.10498046875, + "learning_rate": 0.0016925193844528348, + "loss": 0.1064, + "step": 26770 + }, + { + "epoch": 0.2323851355456984, + "grad_norm": 0.337890625, + "learning_rate": 0.0016924969816988792, + "loss": 0.1104, + "step": 26771 + }, + { + "epoch": 0.23239381602590256, + "grad_norm": 0.37109375, + "learning_rate": 0.0016924745782969778, + "loss": 0.1074, + "step": 26772 + }, + { + "epoch": 0.23240249650610673, + "grad_norm": 0.466796875, + "learning_rate": 0.0016924521742471548, + "loss": 0.0928, + "step": 26773 + }, + { + "epoch": 0.2324111769863109, + "grad_norm": 0.318359375, + "learning_rate": 0.0016924297695494353, + "loss": 0.1074, + "step": 26774 + }, + { + "epoch": 0.23241985746651506, + "grad_norm": 0.259765625, + "learning_rate": 0.0016924073642038432, + "loss": 0.0835, + "step": 26775 + }, + { + "epoch": 0.23242853794671922, + "grad_norm": 0.3046875, + "learning_rate": 0.0016923849582104037, + "loss": 0.1211, + "step": 26776 + }, + { + "epoch": 0.2324372184269234, + "grad_norm": 0.11962890625, + "learning_rate": 0.0016923625515691406, + "loss": 0.0791, + "step": 26777 + }, + { + "epoch": 0.23244589890712755, + "grad_norm": 2.625, + "learning_rate": 0.001692340144280079, + "loss": 0.2637, + "step": 26778 + }, + { + "epoch": 0.23245457938733172, + "grad_norm": 0.1943359375, + "learning_rate": 0.001692317736343243, + "loss": 0.1143, + "step": 26779 + }, + { + "epoch": 0.23246325986753588, + "grad_norm": 0.07275390625, + "learning_rate": 0.001692295327758657, + "loss": 0.1035, + "step": 26780 + }, + { + "epoch": 0.23247194034774005, + "grad_norm": 0.2734375, + "learning_rate": 0.001692272918526346, + "loss": 0.1602, + "step": 26781 + }, + { + "epoch": 0.2324806208279442, + "grad_norm": 0.369140625, + "learning_rate": 0.0016922505086463342, + "loss": 0.0771, + "step": 26782 + }, + { + "epoch": 0.23248930130814838, + "grad_norm": 0.90625, + "learning_rate": 0.0016922280981186463, + "loss": 0.127, + "step": 26783 + }, + { + "epoch": 0.23249798178835254, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016922056869433063, + "loss": 0.1338, + "step": 26784 + }, + { + "epoch": 0.2325066622685567, + "grad_norm": 1.109375, + "learning_rate": 0.001692183275120339, + "loss": 0.084, + "step": 26785 + }, + { + "epoch": 0.23251534274876087, + "grad_norm": 0.4453125, + "learning_rate": 0.0016921608626497694, + "loss": 0.0957, + "step": 26786 + }, + { + "epoch": 0.23252402322896504, + "grad_norm": 0.44921875, + "learning_rate": 0.001692138449531621, + "loss": 0.0938, + "step": 26787 + }, + { + "epoch": 0.2325327037091692, + "grad_norm": 0.44921875, + "learning_rate": 0.0016921160357659193, + "loss": 0.1602, + "step": 26788 + }, + { + "epoch": 0.23254138418937337, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016920936213526883, + "loss": 0.1025, + "step": 26789 + }, + { + "epoch": 0.23255006466957753, + "grad_norm": 0.6171875, + "learning_rate": 0.0016920712062919526, + "loss": 0.1543, + "step": 26790 + }, + { + "epoch": 0.2325587451497817, + "grad_norm": 1.0, + "learning_rate": 0.0016920487905837367, + "loss": 0.1182, + "step": 26791 + }, + { + "epoch": 0.23256742562998586, + "grad_norm": 0.173828125, + "learning_rate": 0.0016920263742280653, + "loss": 0.1152, + "step": 26792 + }, + { + "epoch": 0.23257610611019003, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016920039572249628, + "loss": 0.1348, + "step": 26793 + }, + { + "epoch": 0.2325847865903942, + "grad_norm": 0.2578125, + "learning_rate": 0.0016919815395744534, + "loss": 0.1196, + "step": 26794 + }, + { + "epoch": 0.23259346707059836, + "grad_norm": 0.51171875, + "learning_rate": 0.001691959121276562, + "loss": 0.1631, + "step": 26795 + }, + { + "epoch": 0.23260214755080252, + "grad_norm": 0.447265625, + "learning_rate": 0.0016919367023313131, + "loss": 0.124, + "step": 26796 + }, + { + "epoch": 0.2326108280310067, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016919142827387307, + "loss": 0.124, + "step": 26797 + }, + { + "epoch": 0.23261950851121085, + "grad_norm": 0.08251953125, + "learning_rate": 0.0016918918624988402, + "loss": 0.082, + "step": 26798 + }, + { + "epoch": 0.232628188991415, + "grad_norm": 0.478515625, + "learning_rate": 0.0016918694416116653, + "loss": 0.1084, + "step": 26799 + }, + { + "epoch": 0.23263686947161916, + "grad_norm": 0.2470703125, + "learning_rate": 0.0016918470200772316, + "loss": 0.1523, + "step": 26800 + }, + { + "epoch": 0.23264554995182332, + "grad_norm": 0.07763671875, + "learning_rate": 0.0016918245978955622, + "loss": 0.1245, + "step": 26801 + }, + { + "epoch": 0.2326542304320275, + "grad_norm": 0.1083984375, + "learning_rate": 0.0016918021750666826, + "loss": 0.0762, + "step": 26802 + }, + { + "epoch": 0.23266291091223165, + "grad_norm": 0.59375, + "learning_rate": 0.0016917797515906168, + "loss": 0.1406, + "step": 26803 + }, + { + "epoch": 0.23267159139243582, + "grad_norm": 0.33203125, + "learning_rate": 0.0016917573274673895, + "loss": 0.123, + "step": 26804 + }, + { + "epoch": 0.23268027187263998, + "grad_norm": 0.28515625, + "learning_rate": 0.0016917349026970257, + "loss": 0.1035, + "step": 26805 + }, + { + "epoch": 0.23268895235284415, + "grad_norm": 0.59765625, + "learning_rate": 0.0016917124772795492, + "loss": 0.1035, + "step": 26806 + }, + { + "epoch": 0.2326976328330483, + "grad_norm": 0.09423828125, + "learning_rate": 0.001691690051214985, + "loss": 0.1006, + "step": 26807 + }, + { + "epoch": 0.23270631331325248, + "grad_norm": 0.23046875, + "learning_rate": 0.001691667624503357, + "loss": 0.1025, + "step": 26808 + }, + { + "epoch": 0.23271499379345664, + "grad_norm": 0.423828125, + "learning_rate": 0.0016916451971446904, + "loss": 0.0811, + "step": 26809 + }, + { + "epoch": 0.2327236742736608, + "grad_norm": 0.69140625, + "learning_rate": 0.00169162276913901, + "loss": 0.127, + "step": 26810 + }, + { + "epoch": 0.23273235475386497, + "grad_norm": 0.1708984375, + "learning_rate": 0.0016916003404863391, + "loss": 0.0854, + "step": 26811 + }, + { + "epoch": 0.23274103523406914, + "grad_norm": 0.796875, + "learning_rate": 0.0016915779111867034, + "loss": 0.0879, + "step": 26812 + }, + { + "epoch": 0.2327497157142733, + "grad_norm": 0.26953125, + "learning_rate": 0.0016915554812401267, + "loss": 0.0894, + "step": 26813 + }, + { + "epoch": 0.23275839619447747, + "grad_norm": 0.12353515625, + "learning_rate": 0.001691533050646634, + "loss": 0.0977, + "step": 26814 + }, + { + "epoch": 0.23276707667468163, + "grad_norm": 0.1357421875, + "learning_rate": 0.0016915106194062498, + "loss": 0.1035, + "step": 26815 + }, + { + "epoch": 0.2327757571548858, + "grad_norm": 0.8671875, + "learning_rate": 0.0016914881875189983, + "loss": 0.1445, + "step": 26816 + }, + { + "epoch": 0.23278443763508996, + "grad_norm": 0.33203125, + "learning_rate": 0.001691465754984904, + "loss": 0.0986, + "step": 26817 + }, + { + "epoch": 0.23279311811529413, + "grad_norm": 0.421875, + "learning_rate": 0.001691443321803992, + "loss": 0.1196, + "step": 26818 + }, + { + "epoch": 0.2328017985954983, + "grad_norm": 0.30859375, + "learning_rate": 0.0016914208879762862, + "loss": 0.1289, + "step": 26819 + }, + { + "epoch": 0.23281047907570246, + "grad_norm": 0.54296875, + "learning_rate": 0.0016913984535018116, + "loss": 0.124, + "step": 26820 + }, + { + "epoch": 0.23281915955590662, + "grad_norm": 0.353515625, + "learning_rate": 0.0016913760183805925, + "loss": 0.1113, + "step": 26821 + }, + { + "epoch": 0.2328278400361108, + "grad_norm": 0.4765625, + "learning_rate": 0.0016913535826126535, + "loss": 0.1245, + "step": 26822 + }, + { + "epoch": 0.23283652051631495, + "grad_norm": 0.97265625, + "learning_rate": 0.0016913311461980189, + "loss": 0.1025, + "step": 26823 + }, + { + "epoch": 0.23284520099651912, + "grad_norm": 0.28515625, + "learning_rate": 0.0016913087091367134, + "loss": 0.0928, + "step": 26824 + }, + { + "epoch": 0.23285388147672328, + "grad_norm": 0.12158203125, + "learning_rate": 0.0016912862714287618, + "loss": 0.1211, + "step": 26825 + }, + { + "epoch": 0.23286256195692745, + "grad_norm": 0.5546875, + "learning_rate": 0.0016912638330741888, + "loss": 0.1445, + "step": 26826 + }, + { + "epoch": 0.23287124243713161, + "grad_norm": 0.0908203125, + "learning_rate": 0.0016912413940730177, + "loss": 0.1064, + "step": 26827 + }, + { + "epoch": 0.23287992291733578, + "grad_norm": 0.3671875, + "learning_rate": 0.0016912189544252742, + "loss": 0.0801, + "step": 26828 + }, + { + "epoch": 0.23288860339753994, + "grad_norm": 0.08740234375, + "learning_rate": 0.0016911965141309829, + "loss": 0.1465, + "step": 26829 + }, + { + "epoch": 0.2328972838777441, + "grad_norm": 0.8203125, + "learning_rate": 0.0016911740731901676, + "loss": 0.1523, + "step": 26830 + }, + { + "epoch": 0.23290596435794828, + "grad_norm": 0.76953125, + "learning_rate": 0.0016911516316028535, + "loss": 0.1206, + "step": 26831 + }, + { + "epoch": 0.23291464483815244, + "grad_norm": 0.125, + "learning_rate": 0.0016911291893690645, + "loss": 0.1123, + "step": 26832 + }, + { + "epoch": 0.2329233253183566, + "grad_norm": 0.65625, + "learning_rate": 0.001691106746488826, + "loss": 0.0908, + "step": 26833 + }, + { + "epoch": 0.23293200579856077, + "grad_norm": 0.173828125, + "learning_rate": 0.0016910843029621617, + "loss": 0.0879, + "step": 26834 + }, + { + "epoch": 0.23294068627876494, + "grad_norm": 0.166015625, + "learning_rate": 0.0016910618587890965, + "loss": 0.0762, + "step": 26835 + }, + { + "epoch": 0.2329493667589691, + "grad_norm": 0.283203125, + "learning_rate": 0.001691039413969655, + "loss": 0.1182, + "step": 26836 + }, + { + "epoch": 0.23295804723917327, + "grad_norm": 0.55859375, + "learning_rate": 0.0016910169685038616, + "loss": 0.1191, + "step": 26837 + }, + { + "epoch": 0.23296672771937743, + "grad_norm": 0.1650390625, + "learning_rate": 0.001690994522391741, + "loss": 0.0903, + "step": 26838 + }, + { + "epoch": 0.2329754081995816, + "grad_norm": 0.55859375, + "learning_rate": 0.0016909720756333175, + "loss": 0.0947, + "step": 26839 + }, + { + "epoch": 0.23298408867978576, + "grad_norm": 0.5625, + "learning_rate": 0.001690949628228616, + "loss": 0.1445, + "step": 26840 + }, + { + "epoch": 0.23299276915998993, + "grad_norm": 0.232421875, + "learning_rate": 0.0016909271801776608, + "loss": 0.085, + "step": 26841 + }, + { + "epoch": 0.2330014496401941, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016909047314804765, + "loss": 0.125, + "step": 26842 + }, + { + "epoch": 0.23301013012039826, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016908822821370877, + "loss": 0.1172, + "step": 26843 + }, + { + "epoch": 0.23301881060060242, + "grad_norm": 0.21484375, + "learning_rate": 0.0016908598321475191, + "loss": 0.1172, + "step": 26844 + }, + { + "epoch": 0.2330274910808066, + "grad_norm": 0.2490234375, + "learning_rate": 0.0016908373815117948, + "loss": 0.1172, + "step": 26845 + }, + { + "epoch": 0.23303617156101075, + "grad_norm": 0.63671875, + "learning_rate": 0.0016908149302299396, + "loss": 0.1328, + "step": 26846 + }, + { + "epoch": 0.23304485204121492, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016907924783019783, + "loss": 0.1118, + "step": 26847 + }, + { + "epoch": 0.23305353252141908, + "grad_norm": 0.396484375, + "learning_rate": 0.0016907700257279353, + "loss": 0.1338, + "step": 26848 + }, + { + "epoch": 0.23306221300162325, + "grad_norm": 0.435546875, + "learning_rate": 0.0016907475725078348, + "loss": 0.0928, + "step": 26849 + }, + { + "epoch": 0.2330708934818274, + "grad_norm": 0.06982421875, + "learning_rate": 0.0016907251186417016, + "loss": 0.0972, + "step": 26850 + }, + { + "epoch": 0.23307957396203158, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016907026641295602, + "loss": 0.1113, + "step": 26851 + }, + { + "epoch": 0.23308825444223574, + "grad_norm": 0.162109375, + "learning_rate": 0.0016906802089714352, + "loss": 0.0757, + "step": 26852 + }, + { + "epoch": 0.2330969349224399, + "grad_norm": 0.142578125, + "learning_rate": 0.0016906577531673518, + "loss": 0.0747, + "step": 26853 + }, + { + "epoch": 0.23310561540264407, + "grad_norm": 0.10009765625, + "learning_rate": 0.0016906352967173332, + "loss": 0.0952, + "step": 26854 + }, + { + "epoch": 0.23311429588284824, + "grad_norm": 0.361328125, + "learning_rate": 0.0016906128396214054, + "loss": 0.0786, + "step": 26855 + }, + { + "epoch": 0.2331229763630524, + "grad_norm": 0.3203125, + "learning_rate": 0.0016905903818795918, + "loss": 0.1001, + "step": 26856 + }, + { + "epoch": 0.23313165684325657, + "grad_norm": 0.09619140625, + "learning_rate": 0.0016905679234919177, + "loss": 0.1055, + "step": 26857 + }, + { + "epoch": 0.23314033732346073, + "grad_norm": 0.41796875, + "learning_rate": 0.0016905454644584074, + "loss": 0.0938, + "step": 26858 + }, + { + "epoch": 0.2331490178036649, + "grad_norm": 0.390625, + "learning_rate": 0.0016905230047790853, + "loss": 0.0688, + "step": 26859 + }, + { + "epoch": 0.23315769828386906, + "grad_norm": 0.390625, + "learning_rate": 0.001690500544453976, + "loss": 0.0791, + "step": 26860 + }, + { + "epoch": 0.23316637876407323, + "grad_norm": 0.6015625, + "learning_rate": 0.0016904780834831044, + "loss": 0.1318, + "step": 26861 + }, + { + "epoch": 0.2331750592442774, + "grad_norm": 0.140625, + "learning_rate": 0.001690455621866495, + "loss": 0.1133, + "step": 26862 + }, + { + "epoch": 0.23318373972448156, + "grad_norm": 0.34375, + "learning_rate": 0.0016904331596041715, + "loss": 0.1113, + "step": 26863 + }, + { + "epoch": 0.23319242020468572, + "grad_norm": 0.32421875, + "learning_rate": 0.0016904106966961598, + "loss": 0.1016, + "step": 26864 + }, + { + "epoch": 0.2332011006848899, + "grad_norm": 0.470703125, + "learning_rate": 0.0016903882331424838, + "loss": 0.1064, + "step": 26865 + }, + { + "epoch": 0.23320978116509405, + "grad_norm": 0.0908203125, + "learning_rate": 0.001690365768943168, + "loss": 0.082, + "step": 26866 + }, + { + "epoch": 0.23321846164529822, + "grad_norm": 0.2265625, + "learning_rate": 0.0016903433040982368, + "loss": 0.1289, + "step": 26867 + }, + { + "epoch": 0.23322714212550238, + "grad_norm": 0.2158203125, + "learning_rate": 0.0016903208386077153, + "loss": 0.0952, + "step": 26868 + }, + { + "epoch": 0.23323582260570655, + "grad_norm": 0.302734375, + "learning_rate": 0.0016902983724716278, + "loss": 0.0938, + "step": 26869 + }, + { + "epoch": 0.23324450308591071, + "grad_norm": 0.2890625, + "learning_rate": 0.001690275905689999, + "loss": 0.0718, + "step": 26870 + }, + { + "epoch": 0.23325318356611488, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016902534382628532, + "loss": 0.104, + "step": 26871 + }, + { + "epoch": 0.23326186404631905, + "grad_norm": 0.5859375, + "learning_rate": 0.0016902309701902149, + "loss": 0.1504, + "step": 26872 + }, + { + "epoch": 0.2332705445265232, + "grad_norm": 0.142578125, + "learning_rate": 0.0016902085014721092, + "loss": 0.0684, + "step": 26873 + }, + { + "epoch": 0.23327922500672738, + "grad_norm": 0.169921875, + "learning_rate": 0.0016901860321085602, + "loss": 0.1299, + "step": 26874 + }, + { + "epoch": 0.23328790548693154, + "grad_norm": 0.609375, + "learning_rate": 0.0016901635620995927, + "loss": 0.1226, + "step": 26875 + }, + { + "epoch": 0.2332965859671357, + "grad_norm": 0.169921875, + "learning_rate": 0.0016901410914452312, + "loss": 0.127, + "step": 26876 + }, + { + "epoch": 0.23330526644733987, + "grad_norm": 0.0888671875, + "learning_rate": 0.0016901186201455, + "loss": 0.0889, + "step": 26877 + }, + { + "epoch": 0.23331394692754404, + "grad_norm": 0.1318359375, + "learning_rate": 0.001690096148200424, + "loss": 0.1064, + "step": 26878 + }, + { + "epoch": 0.2333226274077482, + "grad_norm": 0.306640625, + "learning_rate": 0.0016900736756100278, + "loss": 0.0952, + "step": 26879 + }, + { + "epoch": 0.23333130788795237, + "grad_norm": 0.609375, + "learning_rate": 0.0016900512023743362, + "loss": 0.1729, + "step": 26880 + }, + { + "epoch": 0.23333998836815653, + "grad_norm": 0.08935546875, + "learning_rate": 0.0016900287284933734, + "loss": 0.1445, + "step": 26881 + }, + { + "epoch": 0.2333486688483607, + "grad_norm": 0.06640625, + "learning_rate": 0.0016900062539671637, + "loss": 0.0781, + "step": 26882 + }, + { + "epoch": 0.23335734932856486, + "grad_norm": 0.359375, + "learning_rate": 0.0016899837787957325, + "loss": 0.123, + "step": 26883 + }, + { + "epoch": 0.23336602980876903, + "grad_norm": 0.2158203125, + "learning_rate": 0.0016899613029791032, + "loss": 0.1016, + "step": 26884 + }, + { + "epoch": 0.2333747102889732, + "grad_norm": 0.333984375, + "learning_rate": 0.0016899388265173014, + "loss": 0.0923, + "step": 26885 + }, + { + "epoch": 0.23338339076917736, + "grad_norm": 0.333984375, + "learning_rate": 0.0016899163494103516, + "loss": 0.1299, + "step": 26886 + }, + { + "epoch": 0.23339207124938152, + "grad_norm": 0.14453125, + "learning_rate": 0.001689893871658278, + "loss": 0.1553, + "step": 26887 + }, + { + "epoch": 0.2334007517295857, + "grad_norm": 0.11376953125, + "learning_rate": 0.0016898713932611054, + "loss": 0.104, + "step": 26888 + }, + { + "epoch": 0.23340943220978985, + "grad_norm": 0.458984375, + "learning_rate": 0.0016898489142188585, + "loss": 0.1279, + "step": 26889 + }, + { + "epoch": 0.23341811268999402, + "grad_norm": 0.22265625, + "learning_rate": 0.001689826434531561, + "loss": 0.0903, + "step": 26890 + }, + { + "epoch": 0.23342679317019818, + "grad_norm": 0.3046875, + "learning_rate": 0.0016898039541992386, + "loss": 0.1104, + "step": 26891 + }, + { + "epoch": 0.23343547365040235, + "grad_norm": 0.50390625, + "learning_rate": 0.0016897814732219157, + "loss": 0.1582, + "step": 26892 + }, + { + "epoch": 0.2334441541306065, + "grad_norm": 0.08154296875, + "learning_rate": 0.0016897589915996164, + "loss": 0.0806, + "step": 26893 + }, + { + "epoch": 0.23345283461081068, + "grad_norm": 0.185546875, + "learning_rate": 0.0016897365093323654, + "loss": 0.0742, + "step": 26894 + }, + { + "epoch": 0.23346151509101484, + "grad_norm": 0.14453125, + "learning_rate": 0.0016897140264201876, + "loss": 0.061, + "step": 26895 + }, + { + "epoch": 0.233470195571219, + "grad_norm": 0.251953125, + "learning_rate": 0.0016896915428631072, + "loss": 0.1113, + "step": 26896 + }, + { + "epoch": 0.23347887605142317, + "grad_norm": 0.400390625, + "learning_rate": 0.0016896690586611493, + "loss": 0.1113, + "step": 26897 + }, + { + "epoch": 0.23348755653162734, + "grad_norm": 0.08349609375, + "learning_rate": 0.0016896465738143378, + "loss": 0.0874, + "step": 26898 + }, + { + "epoch": 0.2334962370118315, + "grad_norm": 0.29296875, + "learning_rate": 0.001689624088322698, + "loss": 0.1348, + "step": 26899 + }, + { + "epoch": 0.23350491749203567, + "grad_norm": 0.8046875, + "learning_rate": 0.001689601602186254, + "loss": 0.1147, + "step": 26900 + }, + { + "epoch": 0.23351359797223983, + "grad_norm": 0.275390625, + "learning_rate": 0.0016895791154050304, + "loss": 0.0825, + "step": 26901 + }, + { + "epoch": 0.233522278452444, + "grad_norm": 0.181640625, + "learning_rate": 0.0016895566279790523, + "loss": 0.126, + "step": 26902 + }, + { + "epoch": 0.23353095893264816, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016895341399083437, + "loss": 0.1104, + "step": 26903 + }, + { + "epoch": 0.23353963941285233, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016895116511929295, + "loss": 0.1104, + "step": 26904 + }, + { + "epoch": 0.2335483198930565, + "grad_norm": 0.353515625, + "learning_rate": 0.001689489161832834, + "loss": 0.1094, + "step": 26905 + }, + { + "epoch": 0.23355700037326066, + "grad_norm": 0.1982421875, + "learning_rate": 0.0016894666718280822, + "loss": 0.123, + "step": 26906 + }, + { + "epoch": 0.23356568085346482, + "grad_norm": 0.06787109375, + "learning_rate": 0.0016894441811786988, + "loss": 0.0811, + "step": 26907 + }, + { + "epoch": 0.233574361333669, + "grad_norm": 0.47265625, + "learning_rate": 0.0016894216898847075, + "loss": 0.1216, + "step": 26908 + }, + { + "epoch": 0.23358304181387315, + "grad_norm": 0.130859375, + "learning_rate": 0.001689399197946134, + "loss": 0.125, + "step": 26909 + }, + { + "epoch": 0.23359172229407732, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016893767053630018, + "loss": 0.0938, + "step": 26910 + }, + { + "epoch": 0.23360040277428148, + "grad_norm": 0.130859375, + "learning_rate": 0.0016893542121353368, + "loss": 0.085, + "step": 26911 + }, + { + "epoch": 0.23360908325448565, + "grad_norm": 0.625, + "learning_rate": 0.0016893317182631624, + "loss": 0.1006, + "step": 26912 + }, + { + "epoch": 0.23361776373468982, + "grad_norm": 0.39453125, + "learning_rate": 0.001689309223746504, + "loss": 0.1211, + "step": 26913 + }, + { + "epoch": 0.23362644421489398, + "grad_norm": 0.119140625, + "learning_rate": 0.0016892867285853857, + "loss": 0.0967, + "step": 26914 + }, + { + "epoch": 0.23363512469509815, + "grad_norm": 0.267578125, + "learning_rate": 0.0016892642327798323, + "loss": 0.1016, + "step": 26915 + }, + { + "epoch": 0.2336438051753023, + "grad_norm": 0.2021484375, + "learning_rate": 0.0016892417363298684, + "loss": 0.1357, + "step": 26916 + }, + { + "epoch": 0.23365248565550648, + "grad_norm": 0.921875, + "learning_rate": 0.0016892192392355186, + "loss": 0.1001, + "step": 26917 + }, + { + "epoch": 0.23366116613571064, + "grad_norm": 0.2158203125, + "learning_rate": 0.0016891967414968075, + "loss": 0.1484, + "step": 26918 + }, + { + "epoch": 0.2336698466159148, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016891742431137595, + "loss": 0.0781, + "step": 26919 + }, + { + "epoch": 0.23367852709611897, + "grad_norm": 0.30078125, + "learning_rate": 0.0016891517440863998, + "loss": 0.1113, + "step": 26920 + }, + { + "epoch": 0.23368720757632314, + "grad_norm": 1.0, + "learning_rate": 0.0016891292444147523, + "loss": 0.1338, + "step": 26921 + }, + { + "epoch": 0.23369588805652727, + "grad_norm": 0.353515625, + "learning_rate": 0.0016891067440988421, + "loss": 0.083, + "step": 26922 + }, + { + "epoch": 0.23370456853673144, + "grad_norm": 0.10546875, + "learning_rate": 0.0016890842431386934, + "loss": 0.123, + "step": 26923 + }, + { + "epoch": 0.2337132490169356, + "grad_norm": 1.1953125, + "learning_rate": 0.001689061741534331, + "loss": 0.1367, + "step": 26924 + }, + { + "epoch": 0.23372192949713977, + "grad_norm": 0.125, + "learning_rate": 0.0016890392392857798, + "loss": 0.1045, + "step": 26925 + }, + { + "epoch": 0.23373060997734393, + "grad_norm": 0.3125, + "learning_rate": 0.001689016736393064, + "loss": 0.0947, + "step": 26926 + }, + { + "epoch": 0.2337392904575481, + "grad_norm": 0.5546875, + "learning_rate": 0.0016889942328562084, + "loss": 0.126, + "step": 26927 + }, + { + "epoch": 0.23374797093775226, + "grad_norm": 0.90234375, + "learning_rate": 0.0016889717286752376, + "loss": 0.1143, + "step": 26928 + }, + { + "epoch": 0.23375665141795643, + "grad_norm": 0.2578125, + "learning_rate": 0.0016889492238501757, + "loss": 0.0957, + "step": 26929 + }, + { + "epoch": 0.2337653318981606, + "grad_norm": 1.59375, + "learning_rate": 0.0016889267183810483, + "loss": 0.1396, + "step": 26930 + }, + { + "epoch": 0.23377401237836476, + "grad_norm": 0.1845703125, + "learning_rate": 0.0016889042122678797, + "loss": 0.0962, + "step": 26931 + }, + { + "epoch": 0.23378269285856892, + "grad_norm": 0.66796875, + "learning_rate": 0.0016888817055106939, + "loss": 0.1445, + "step": 26932 + }, + { + "epoch": 0.2337913733387731, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016888591981095156, + "loss": 0.127, + "step": 26933 + }, + { + "epoch": 0.23380005381897725, + "grad_norm": 0.177734375, + "learning_rate": 0.0016888366900643704, + "loss": 0.127, + "step": 26934 + }, + { + "epoch": 0.23380873429918142, + "grad_norm": 0.7265625, + "learning_rate": 0.0016888141813752818, + "loss": 0.1108, + "step": 26935 + }, + { + "epoch": 0.23381741477938559, + "grad_norm": 0.5, + "learning_rate": 0.0016887916720422752, + "loss": 0.1279, + "step": 26936 + }, + { + "epoch": 0.23382609525958975, + "grad_norm": 0.2451171875, + "learning_rate": 0.0016887691620653747, + "loss": 0.1113, + "step": 26937 + }, + { + "epoch": 0.23383477573979392, + "grad_norm": 0.78125, + "learning_rate": 0.0016887466514446047, + "loss": 0.1396, + "step": 26938 + }, + { + "epoch": 0.23384345621999808, + "grad_norm": 0.447265625, + "learning_rate": 0.0016887241401799908, + "loss": 0.1196, + "step": 26939 + }, + { + "epoch": 0.23385213670020225, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016887016282715567, + "loss": 0.124, + "step": 26940 + }, + { + "epoch": 0.2338608171804064, + "grad_norm": 0.2060546875, + "learning_rate": 0.0016886791157193274, + "loss": 0.124, + "step": 26941 + }, + { + "epoch": 0.23386949766061058, + "grad_norm": 0.73828125, + "learning_rate": 0.0016886566025233276, + "loss": 0.0918, + "step": 26942 + }, + { + "epoch": 0.23387817814081474, + "grad_norm": 0.220703125, + "learning_rate": 0.0016886340886835816, + "loss": 0.1338, + "step": 26943 + }, + { + "epoch": 0.2338868586210189, + "grad_norm": 0.87109375, + "learning_rate": 0.0016886115742001143, + "loss": 0.1289, + "step": 26944 + }, + { + "epoch": 0.23389553910122307, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016885890590729502, + "loss": 0.0986, + "step": 26945 + }, + { + "epoch": 0.23390421958142724, + "grad_norm": 0.076171875, + "learning_rate": 0.001688566543302114, + "loss": 0.0796, + "step": 26946 + }, + { + "epoch": 0.2339129000616314, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016885440268876303, + "loss": 0.1279, + "step": 26947 + }, + { + "epoch": 0.23392158054183557, + "grad_norm": 0.09130859375, + "learning_rate": 0.0016885215098295238, + "loss": 0.0957, + "step": 26948 + }, + { + "epoch": 0.23393026102203973, + "grad_norm": 0.18359375, + "learning_rate": 0.0016884989921278184, + "loss": 0.0835, + "step": 26949 + }, + { + "epoch": 0.2339389415022439, + "grad_norm": 0.216796875, + "learning_rate": 0.00168847647378254, + "loss": 0.1455, + "step": 26950 + }, + { + "epoch": 0.23394762198244806, + "grad_norm": 0.33984375, + "learning_rate": 0.0016884539547937123, + "loss": 0.1396, + "step": 26951 + }, + { + "epoch": 0.23395630246265223, + "grad_norm": 0.251953125, + "learning_rate": 0.00168843143516136, + "loss": 0.0942, + "step": 26952 + }, + { + "epoch": 0.2339649829428564, + "grad_norm": 0.1162109375, + "learning_rate": 0.0016884089148855085, + "loss": 0.1094, + "step": 26953 + }, + { + "epoch": 0.23397366342306056, + "grad_norm": 0.451171875, + "learning_rate": 0.0016883863939661813, + "loss": 0.0977, + "step": 26954 + }, + { + "epoch": 0.23398234390326472, + "grad_norm": 0.75390625, + "learning_rate": 0.0016883638724034039, + "loss": 0.0928, + "step": 26955 + }, + { + "epoch": 0.2339910243834689, + "grad_norm": 0.248046875, + "learning_rate": 0.0016883413501972008, + "loss": 0.1133, + "step": 26956 + }, + { + "epoch": 0.23399970486367305, + "grad_norm": 0.1474609375, + "learning_rate": 0.0016883188273475963, + "loss": 0.0845, + "step": 26957 + }, + { + "epoch": 0.23400838534387722, + "grad_norm": 0.2177734375, + "learning_rate": 0.0016882963038546146, + "loss": 0.127, + "step": 26958 + }, + { + "epoch": 0.23401706582408138, + "grad_norm": 0.7421875, + "learning_rate": 0.0016882737797182814, + "loss": 0.1104, + "step": 26959 + }, + { + "epoch": 0.23402574630428555, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016882512549386208, + "loss": 0.0977, + "step": 26960 + }, + { + "epoch": 0.2340344267844897, + "grad_norm": 0.39453125, + "learning_rate": 0.0016882287295156573, + "loss": 0.1006, + "step": 26961 + }, + { + "epoch": 0.23404310726469388, + "grad_norm": 0.3046875, + "learning_rate": 0.001688206203449416, + "loss": 0.1221, + "step": 26962 + }, + { + "epoch": 0.23405178774489804, + "grad_norm": 0.10791015625, + "learning_rate": 0.001688183676739921, + "loss": 0.1123, + "step": 26963 + }, + { + "epoch": 0.2340604682251022, + "grad_norm": 0.375, + "learning_rate": 0.0016881611493871974, + "loss": 0.0786, + "step": 26964 + }, + { + "epoch": 0.23406914870530637, + "grad_norm": 0.6796875, + "learning_rate": 0.0016881386213912692, + "loss": 0.1289, + "step": 26965 + }, + { + "epoch": 0.23407782918551054, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016881160927521618, + "loss": 0.1138, + "step": 26966 + }, + { + "epoch": 0.2340865096657147, + "grad_norm": 0.2119140625, + "learning_rate": 0.0016880935634698993, + "loss": 0.125, + "step": 26967 + }, + { + "epoch": 0.23409519014591887, + "grad_norm": 0.2197265625, + "learning_rate": 0.0016880710335445066, + "loss": 0.085, + "step": 26968 + }, + { + "epoch": 0.23410387062612303, + "grad_norm": 0.115234375, + "learning_rate": 0.0016880485029760084, + "loss": 0.1089, + "step": 26969 + }, + { + "epoch": 0.2341125511063272, + "grad_norm": 0.1396484375, + "learning_rate": 0.0016880259717644288, + "loss": 0.1035, + "step": 26970 + }, + { + "epoch": 0.23412123158653136, + "grad_norm": 0.349609375, + "learning_rate": 0.001688003439909793, + "loss": 0.1094, + "step": 26971 + }, + { + "epoch": 0.23412991206673553, + "grad_norm": 0.470703125, + "learning_rate": 0.0016879809074121256, + "loss": 0.123, + "step": 26972 + }, + { + "epoch": 0.2341385925469397, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016879583742714513, + "loss": 0.1064, + "step": 26973 + }, + { + "epoch": 0.23414727302714386, + "grad_norm": 0.53515625, + "learning_rate": 0.0016879358404877942, + "loss": 0.127, + "step": 26974 + }, + { + "epoch": 0.23415595350734802, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016879133060611793, + "loss": 0.1045, + "step": 26975 + }, + { + "epoch": 0.2341646339875522, + "grad_norm": 0.09521484375, + "learning_rate": 0.001687890770991631, + "loss": 0.1079, + "step": 26976 + }, + { + "epoch": 0.23417331446775635, + "grad_norm": 0.318359375, + "learning_rate": 0.001687868235279175, + "loss": 0.0933, + "step": 26977 + }, + { + "epoch": 0.23418199494796052, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016878456989238347, + "loss": 0.0898, + "step": 26978 + }, + { + "epoch": 0.23419067542816469, + "grad_norm": 0.1572265625, + "learning_rate": 0.001687823161925635, + "loss": 0.0874, + "step": 26979 + }, + { + "epoch": 0.23419935590836885, + "grad_norm": 0.359375, + "learning_rate": 0.0016878006242846007, + "loss": 0.1104, + "step": 26980 + }, + { + "epoch": 0.23420803638857302, + "grad_norm": 0.10986328125, + "learning_rate": 0.0016877780860007567, + "loss": 0.1113, + "step": 26981 + }, + { + "epoch": 0.23421671686877718, + "grad_norm": 0.130859375, + "learning_rate": 0.0016877555470741273, + "loss": 0.0737, + "step": 26982 + }, + { + "epoch": 0.23422539734898135, + "grad_norm": 0.6484375, + "learning_rate": 0.001687733007504737, + "loss": 0.1484, + "step": 26983 + }, + { + "epoch": 0.2342340778291855, + "grad_norm": 0.78125, + "learning_rate": 0.001687710467292611, + "loss": 0.1099, + "step": 26984 + }, + { + "epoch": 0.23424275830938968, + "grad_norm": 0.400390625, + "learning_rate": 0.001687687926437774, + "loss": 0.085, + "step": 26985 + }, + { + "epoch": 0.23425143878959384, + "grad_norm": 0.32421875, + "learning_rate": 0.00168766538494025, + "loss": 0.0859, + "step": 26986 + }, + { + "epoch": 0.234260119269798, + "grad_norm": 0.142578125, + "learning_rate": 0.001687642842800064, + "loss": 0.123, + "step": 26987 + }, + { + "epoch": 0.23426879975000217, + "grad_norm": 0.166015625, + "learning_rate": 0.0016876203000172404, + "loss": 0.1387, + "step": 26988 + }, + { + "epoch": 0.23427748023020634, + "grad_norm": 0.435546875, + "learning_rate": 0.001687597756591804, + "loss": 0.0757, + "step": 26989 + }, + { + "epoch": 0.2342861607104105, + "grad_norm": 0.546875, + "learning_rate": 0.00168757521252378, + "loss": 0.1089, + "step": 26990 + }, + { + "epoch": 0.23429484119061467, + "grad_norm": 0.7578125, + "learning_rate": 0.0016875526678131925, + "loss": 0.0898, + "step": 26991 + }, + { + "epoch": 0.23430352167081883, + "grad_norm": 0.09716796875, + "learning_rate": 0.001687530122460066, + "loss": 0.1045, + "step": 26992 + }, + { + "epoch": 0.234312202151023, + "grad_norm": 0.44140625, + "learning_rate": 0.0016875075764644254, + "loss": 0.1016, + "step": 26993 + }, + { + "epoch": 0.23432088263122716, + "grad_norm": 0.35546875, + "learning_rate": 0.0016874850298262954, + "loss": 0.1113, + "step": 26994 + }, + { + "epoch": 0.23432956311143133, + "grad_norm": 0.453125, + "learning_rate": 0.0016874624825457007, + "loss": 0.1016, + "step": 26995 + }, + { + "epoch": 0.2343382435916355, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016874399346226657, + "loss": 0.127, + "step": 26996 + }, + { + "epoch": 0.23434692407183966, + "grad_norm": 0.51171875, + "learning_rate": 0.0016874173860572154, + "loss": 0.1035, + "step": 26997 + }, + { + "epoch": 0.23435560455204382, + "grad_norm": 0.32421875, + "learning_rate": 0.001687394836849374, + "loss": 0.1094, + "step": 26998 + }, + { + "epoch": 0.234364285032248, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016873722869991665, + "loss": 0.1406, + "step": 26999 + }, + { + "epoch": 0.23437296551245215, + "grad_norm": 0.63671875, + "learning_rate": 0.0016873497365066177, + "loss": 0.1196, + "step": 27000 + }, + { + "epoch": 0.23438164599265632, + "grad_norm": 0.65234375, + "learning_rate": 0.0016873271853717518, + "loss": 0.125, + "step": 27001 + }, + { + "epoch": 0.23439032647286048, + "grad_norm": 0.6171875, + "learning_rate": 0.0016873046335945938, + "loss": 0.2041, + "step": 27002 + }, + { + "epoch": 0.23439900695306465, + "grad_norm": 0.248046875, + "learning_rate": 0.0016872820811751684, + "loss": 0.0825, + "step": 27003 + }, + { + "epoch": 0.2344076874332688, + "grad_norm": 0.26171875, + "learning_rate": 0.0016872595281135002, + "loss": 0.0957, + "step": 27004 + }, + { + "epoch": 0.23441636791347298, + "grad_norm": 0.51953125, + "learning_rate": 0.0016872369744096135, + "loss": 0.1289, + "step": 27005 + }, + { + "epoch": 0.23442504839367714, + "grad_norm": 0.298828125, + "learning_rate": 0.0016872144200635333, + "loss": 0.1104, + "step": 27006 + }, + { + "epoch": 0.2344337288738813, + "grad_norm": 0.134765625, + "learning_rate": 0.0016871918650752842, + "loss": 0.126, + "step": 27007 + }, + { + "epoch": 0.23444240935408547, + "grad_norm": 0.1171875, + "learning_rate": 0.0016871693094448912, + "loss": 0.1162, + "step": 27008 + }, + { + "epoch": 0.23445108983428964, + "grad_norm": 0.310546875, + "learning_rate": 0.0016871467531723785, + "loss": 0.0947, + "step": 27009 + }, + { + "epoch": 0.2344597703144938, + "grad_norm": 0.140625, + "learning_rate": 0.0016871241962577708, + "loss": 0.1025, + "step": 27010 + }, + { + "epoch": 0.23446845079469797, + "grad_norm": 0.6328125, + "learning_rate": 0.0016871016387010932, + "loss": 0.0664, + "step": 27011 + }, + { + "epoch": 0.23447713127490213, + "grad_norm": 0.1865234375, + "learning_rate": 0.00168707908050237, + "loss": 0.085, + "step": 27012 + }, + { + "epoch": 0.2344858117551063, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016870565216616257, + "loss": 0.0737, + "step": 27013 + }, + { + "epoch": 0.23449449223531046, + "grad_norm": 0.123046875, + "learning_rate": 0.0016870339621788855, + "loss": 0.1123, + "step": 27014 + }, + { + "epoch": 0.23450317271551463, + "grad_norm": 0.1953125, + "learning_rate": 0.0016870114020541736, + "loss": 0.1133, + "step": 27015 + }, + { + "epoch": 0.2345118531957188, + "grad_norm": 0.384765625, + "learning_rate": 0.0016869888412875146, + "loss": 0.0864, + "step": 27016 + }, + { + "epoch": 0.23452053367592296, + "grad_norm": 0.3046875, + "learning_rate": 0.001686966279878934, + "loss": 0.0938, + "step": 27017 + }, + { + "epoch": 0.23452921415612712, + "grad_norm": 0.2109375, + "learning_rate": 0.0016869437178284555, + "loss": 0.1079, + "step": 27018 + }, + { + "epoch": 0.2345378946363313, + "grad_norm": 0.50390625, + "learning_rate": 0.0016869211551361043, + "loss": 0.1572, + "step": 27019 + }, + { + "epoch": 0.23454657511653546, + "grad_norm": 0.1455078125, + "learning_rate": 0.001686898591801905, + "loss": 0.084, + "step": 27020 + }, + { + "epoch": 0.23455525559673962, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016868760278258824, + "loss": 0.1641, + "step": 27021 + }, + { + "epoch": 0.23456393607694379, + "grad_norm": 0.140625, + "learning_rate": 0.0016868534632080603, + "loss": 0.1157, + "step": 27022 + }, + { + "epoch": 0.23457261655714795, + "grad_norm": 0.57421875, + "learning_rate": 0.0016868308979484648, + "loss": 0.0972, + "step": 27023 + }, + { + "epoch": 0.23458129703735212, + "grad_norm": 0.734375, + "learning_rate": 0.0016868083320471193, + "loss": 0.1211, + "step": 27024 + }, + { + "epoch": 0.23458997751755628, + "grad_norm": 0.10205078125, + "learning_rate": 0.0016867857655040496, + "loss": 0.0996, + "step": 27025 + }, + { + "epoch": 0.23459865799776045, + "grad_norm": 0.431640625, + "learning_rate": 0.0016867631983192792, + "loss": 0.0728, + "step": 27026 + }, + { + "epoch": 0.2346073384779646, + "grad_norm": 0.09814453125, + "learning_rate": 0.0016867406304928337, + "loss": 0.0933, + "step": 27027 + }, + { + "epoch": 0.23461601895816878, + "grad_norm": 0.125, + "learning_rate": 0.0016867180620247374, + "loss": 0.1045, + "step": 27028 + }, + { + "epoch": 0.23462469943837294, + "grad_norm": 0.400390625, + "learning_rate": 0.0016866954929150152, + "loss": 0.0996, + "step": 27029 + }, + { + "epoch": 0.2346333799185771, + "grad_norm": 0.140625, + "learning_rate": 0.0016866729231636915, + "loss": 0.124, + "step": 27030 + }, + { + "epoch": 0.23464206039878127, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016866503527707911, + "loss": 0.1211, + "step": 27031 + }, + { + "epoch": 0.23465074087898544, + "grad_norm": 0.99609375, + "learning_rate": 0.0016866277817363384, + "loss": 0.1011, + "step": 27032 + }, + { + "epoch": 0.2346594213591896, + "grad_norm": 0.0869140625, + "learning_rate": 0.0016866052100603587, + "loss": 0.123, + "step": 27033 + }, + { + "epoch": 0.23466810183939377, + "grad_norm": 0.498046875, + "learning_rate": 0.0016865826377428764, + "loss": 0.1396, + "step": 27034 + }, + { + "epoch": 0.23467678231959793, + "grad_norm": 0.484375, + "learning_rate": 0.001686560064783916, + "loss": 0.1172, + "step": 27035 + }, + { + "epoch": 0.2346854627998021, + "grad_norm": 0.70703125, + "learning_rate": 0.0016865374911835024, + "loss": 0.1025, + "step": 27036 + }, + { + "epoch": 0.23469414328000626, + "grad_norm": 0.16796875, + "learning_rate": 0.00168651491694166, + "loss": 0.1074, + "step": 27037 + }, + { + "epoch": 0.23470282376021043, + "grad_norm": 0.271484375, + "learning_rate": 0.001686492342058414, + "loss": 0.084, + "step": 27038 + }, + { + "epoch": 0.2347115042404146, + "grad_norm": 0.1396484375, + "learning_rate": 0.0016864697665337888, + "loss": 0.0996, + "step": 27039 + }, + { + "epoch": 0.23472018472061876, + "grad_norm": 0.82421875, + "learning_rate": 0.001686447190367809, + "loss": 0.1338, + "step": 27040 + }, + { + "epoch": 0.23472886520082292, + "grad_norm": 0.36328125, + "learning_rate": 0.0016864246135604992, + "loss": 0.1172, + "step": 27041 + }, + { + "epoch": 0.2347375456810271, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016864020361118844, + "loss": 0.1152, + "step": 27042 + }, + { + "epoch": 0.23474622616123125, + "grad_norm": 0.26953125, + "learning_rate": 0.001686379458021989, + "loss": 0.1602, + "step": 27043 + }, + { + "epoch": 0.23475490664143542, + "grad_norm": 0.546875, + "learning_rate": 0.001686356879290838, + "loss": 0.1211, + "step": 27044 + }, + { + "epoch": 0.23476358712163956, + "grad_norm": 0.173828125, + "learning_rate": 0.0016863342999184558, + "loss": 0.1108, + "step": 27045 + }, + { + "epoch": 0.23477226760184372, + "grad_norm": 0.0869140625, + "learning_rate": 0.0016863117199048675, + "loss": 0.124, + "step": 27046 + }, + { + "epoch": 0.23478094808204789, + "grad_norm": 0.28125, + "learning_rate": 0.0016862891392500972, + "loss": 0.1064, + "step": 27047 + }, + { + "epoch": 0.23478962856225205, + "grad_norm": 0.5234375, + "learning_rate": 0.00168626655795417, + "loss": 0.0962, + "step": 27048 + }, + { + "epoch": 0.23479830904245622, + "grad_norm": 0.1357421875, + "learning_rate": 0.0016862439760171105, + "loss": 0.1055, + "step": 27049 + }, + { + "epoch": 0.23480698952266038, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016862213934389434, + "loss": 0.0952, + "step": 27050 + }, + { + "epoch": 0.23481567000286455, + "grad_norm": 0.3203125, + "learning_rate": 0.0016861988102196935, + "loss": 0.127, + "step": 27051 + }, + { + "epoch": 0.2348243504830687, + "grad_norm": 0.306640625, + "learning_rate": 0.0016861762263593854, + "loss": 0.1016, + "step": 27052 + }, + { + "epoch": 0.23483303096327288, + "grad_norm": 0.337890625, + "learning_rate": 0.0016861536418580434, + "loss": 0.1123, + "step": 27053 + }, + { + "epoch": 0.23484171144347704, + "grad_norm": 0.62890625, + "learning_rate": 0.001686131056715693, + "loss": 0.1113, + "step": 27054 + }, + { + "epoch": 0.2348503919236812, + "grad_norm": 0.54296875, + "learning_rate": 0.0016861084709323584, + "loss": 0.1035, + "step": 27055 + }, + { + "epoch": 0.23485907240388537, + "grad_norm": 0.376953125, + "learning_rate": 0.0016860858845080642, + "loss": 0.104, + "step": 27056 + }, + { + "epoch": 0.23486775288408954, + "grad_norm": 0.65234375, + "learning_rate": 0.0016860632974428354, + "loss": 0.123, + "step": 27057 + }, + { + "epoch": 0.2348764333642937, + "grad_norm": 0.134765625, + "learning_rate": 0.0016860407097366965, + "loss": 0.1211, + "step": 27058 + }, + { + "epoch": 0.23488511384449787, + "grad_norm": 0.6015625, + "learning_rate": 0.0016860181213896727, + "loss": 0.1187, + "step": 27059 + }, + { + "epoch": 0.23489379432470203, + "grad_norm": 0.419921875, + "learning_rate": 0.0016859955324017878, + "loss": 0.1172, + "step": 27060 + }, + { + "epoch": 0.2349024748049062, + "grad_norm": 0.29296875, + "learning_rate": 0.0016859729427730673, + "loss": 0.1172, + "step": 27061 + }, + { + "epoch": 0.23491115528511036, + "grad_norm": 0.87109375, + "learning_rate": 0.0016859503525035353, + "loss": 0.083, + "step": 27062 + }, + { + "epoch": 0.23491983576531453, + "grad_norm": 0.259765625, + "learning_rate": 0.001685927761593217, + "loss": 0.0859, + "step": 27063 + }, + { + "epoch": 0.2349285162455187, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016859051700421365, + "loss": 0.1094, + "step": 27064 + }, + { + "epoch": 0.23493719672572286, + "grad_norm": 0.1953125, + "learning_rate": 0.0016858825778503198, + "loss": 0.124, + "step": 27065 + }, + { + "epoch": 0.23494587720592702, + "grad_norm": 0.470703125, + "learning_rate": 0.0016858599850177898, + "loss": 0.1396, + "step": 27066 + }, + { + "epoch": 0.2349545576861312, + "grad_norm": 0.0654296875, + "learning_rate": 0.0016858373915445725, + "loss": 0.0869, + "step": 27067 + }, + { + "epoch": 0.23496323816633535, + "grad_norm": 0.80859375, + "learning_rate": 0.0016858147974306923, + "loss": 0.1299, + "step": 27068 + }, + { + "epoch": 0.23497191864653952, + "grad_norm": 0.126953125, + "learning_rate": 0.0016857922026761738, + "loss": 0.0986, + "step": 27069 + }, + { + "epoch": 0.23498059912674368, + "grad_norm": 0.212890625, + "learning_rate": 0.0016857696072810416, + "loss": 0.1562, + "step": 27070 + }, + { + "epoch": 0.23498927960694785, + "grad_norm": 0.380859375, + "learning_rate": 0.0016857470112453209, + "loss": 0.0757, + "step": 27071 + }, + { + "epoch": 0.234997960087152, + "grad_norm": 0.326171875, + "learning_rate": 0.0016857244145690358, + "loss": 0.1094, + "step": 27072 + }, + { + "epoch": 0.23500664056735618, + "grad_norm": 0.8515625, + "learning_rate": 0.0016857018172522112, + "loss": 0.1074, + "step": 27073 + }, + { + "epoch": 0.23501532104756034, + "grad_norm": 0.22265625, + "learning_rate": 0.0016856792192948718, + "loss": 0.0586, + "step": 27074 + }, + { + "epoch": 0.2350240015277645, + "grad_norm": 0.41015625, + "learning_rate": 0.0016856566206970428, + "loss": 0.1309, + "step": 27075 + }, + { + "epoch": 0.23503268200796867, + "grad_norm": 0.1572265625, + "learning_rate": 0.0016856340214587485, + "loss": 0.0913, + "step": 27076 + }, + { + "epoch": 0.23504136248817284, + "grad_norm": 0.1953125, + "learning_rate": 0.0016856114215800134, + "loss": 0.0933, + "step": 27077 + }, + { + "epoch": 0.235050042968377, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016855888210608628, + "loss": 0.125, + "step": 27078 + }, + { + "epoch": 0.23505872344858117, + "grad_norm": 0.19140625, + "learning_rate": 0.0016855662199013203, + "loss": 0.1128, + "step": 27079 + }, + { + "epoch": 0.23506740392878533, + "grad_norm": 0.87109375, + "learning_rate": 0.001685543618101412, + "loss": 0.0977, + "step": 27080 + }, + { + "epoch": 0.2350760844089895, + "grad_norm": 0.189453125, + "learning_rate": 0.001685521015661162, + "loss": 0.0933, + "step": 27081 + }, + { + "epoch": 0.23508476488919366, + "grad_norm": 0.8671875, + "learning_rate": 0.001685498412580595, + "loss": 0.0923, + "step": 27082 + }, + { + "epoch": 0.23509344536939783, + "grad_norm": 0.30078125, + "learning_rate": 0.0016854758088597356, + "loss": 0.1025, + "step": 27083 + }, + { + "epoch": 0.235102125849602, + "grad_norm": 0.125, + "learning_rate": 0.0016854532044986085, + "loss": 0.0957, + "step": 27084 + }, + { + "epoch": 0.23511080632980616, + "grad_norm": 0.130859375, + "learning_rate": 0.0016854305994972384, + "loss": 0.1377, + "step": 27085 + }, + { + "epoch": 0.23511948681001033, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016854079938556509, + "loss": 0.1064, + "step": 27086 + }, + { + "epoch": 0.2351281672902145, + "grad_norm": 0.11328125, + "learning_rate": 0.0016853853875738694, + "loss": 0.0972, + "step": 27087 + }, + { + "epoch": 0.23513684777041866, + "grad_norm": 0.244140625, + "learning_rate": 0.0016853627806519196, + "loss": 0.1147, + "step": 27088 + }, + { + "epoch": 0.23514552825062282, + "grad_norm": 2.5, + "learning_rate": 0.001685340173089826, + "loss": 0.1475, + "step": 27089 + }, + { + "epoch": 0.23515420873082699, + "grad_norm": 0.10546875, + "learning_rate": 0.0016853175648876126, + "loss": 0.1035, + "step": 27090 + }, + { + "epoch": 0.23516288921103115, + "grad_norm": 8.8125, + "learning_rate": 0.0016852949560453051, + "loss": 0.1348, + "step": 27091 + }, + { + "epoch": 0.23517156969123532, + "grad_norm": 0.1279296875, + "learning_rate": 0.0016852723465629277, + "loss": 0.1104, + "step": 27092 + }, + { + "epoch": 0.23518025017143948, + "grad_norm": 0.5703125, + "learning_rate": 0.0016852497364405054, + "loss": 0.0732, + "step": 27093 + }, + { + "epoch": 0.23518893065164365, + "grad_norm": 0.333984375, + "learning_rate": 0.0016852271256780627, + "loss": 0.1416, + "step": 27094 + }, + { + "epoch": 0.2351976111318478, + "grad_norm": 0.306640625, + "learning_rate": 0.0016852045142756243, + "loss": 0.1084, + "step": 27095 + }, + { + "epoch": 0.23520629161205198, + "grad_norm": 0.11669921875, + "learning_rate": 0.0016851819022332152, + "loss": 0.1133, + "step": 27096 + }, + { + "epoch": 0.23521497209225614, + "grad_norm": 0.333984375, + "learning_rate": 0.00168515928955086, + "loss": 0.1035, + "step": 27097 + }, + { + "epoch": 0.2352236525724603, + "grad_norm": 0.296875, + "learning_rate": 0.0016851366762285832, + "loss": 0.1289, + "step": 27098 + }, + { + "epoch": 0.23523233305266447, + "grad_norm": 0.9296875, + "learning_rate": 0.0016851140622664098, + "loss": 0.1025, + "step": 27099 + }, + { + "epoch": 0.23524101353286864, + "grad_norm": 0.09033203125, + "learning_rate": 0.0016850914476643647, + "loss": 0.0859, + "step": 27100 + }, + { + "epoch": 0.2352496940130728, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016850688324224717, + "loss": 0.0781, + "step": 27101 + }, + { + "epoch": 0.23525837449327697, + "grad_norm": 0.083984375, + "learning_rate": 0.0016850462165407568, + "loss": 0.0918, + "step": 27102 + }, + { + "epoch": 0.23526705497348113, + "grad_norm": 0.431640625, + "learning_rate": 0.0016850236000192443, + "loss": 0.0923, + "step": 27103 + }, + { + "epoch": 0.2352757354536853, + "grad_norm": 0.1767578125, + "learning_rate": 0.0016850009828579583, + "loss": 0.0947, + "step": 27104 + }, + { + "epoch": 0.23528441593388946, + "grad_norm": 1.015625, + "learning_rate": 0.001684978365056924, + "loss": 0.1484, + "step": 27105 + }, + { + "epoch": 0.23529309641409363, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016849557466161664, + "loss": 0.0801, + "step": 27106 + }, + { + "epoch": 0.2353017768942978, + "grad_norm": 0.259765625, + "learning_rate": 0.00168493312753571, + "loss": 0.0918, + "step": 27107 + }, + { + "epoch": 0.23531045737450196, + "grad_norm": 0.546875, + "learning_rate": 0.0016849105078155797, + "loss": 0.1226, + "step": 27108 + }, + { + "epoch": 0.23531913785470612, + "grad_norm": 0.37109375, + "learning_rate": 0.0016848878874557995, + "loss": 0.0938, + "step": 27109 + }, + { + "epoch": 0.2353278183349103, + "grad_norm": 0.1904296875, + "learning_rate": 0.0016848652664563952, + "loss": 0.1572, + "step": 27110 + }, + { + "epoch": 0.23533649881511445, + "grad_norm": 0.392578125, + "learning_rate": 0.0016848426448173909, + "loss": 0.126, + "step": 27111 + }, + { + "epoch": 0.23534517929531862, + "grad_norm": 0.53125, + "learning_rate": 0.0016848200225388114, + "loss": 0.1621, + "step": 27112 + }, + { + "epoch": 0.23535385977552278, + "grad_norm": 0.40234375, + "learning_rate": 0.0016847973996206816, + "loss": 0.1055, + "step": 27113 + }, + { + "epoch": 0.23536254025572695, + "grad_norm": 0.359375, + "learning_rate": 0.001684774776063026, + "loss": 0.0728, + "step": 27114 + }, + { + "epoch": 0.2353712207359311, + "grad_norm": 0.41796875, + "learning_rate": 0.0016847521518658696, + "loss": 0.1348, + "step": 27115 + }, + { + "epoch": 0.23537990121613528, + "grad_norm": 0.2734375, + "learning_rate": 0.001684729527029237, + "loss": 0.0815, + "step": 27116 + }, + { + "epoch": 0.23538858169633944, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016847069015531529, + "loss": 0.1123, + "step": 27117 + }, + { + "epoch": 0.2353972621765436, + "grad_norm": 0.1455078125, + "learning_rate": 0.0016846842754376422, + "loss": 0.1309, + "step": 27118 + }, + { + "epoch": 0.23540594265674777, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016846616486827298, + "loss": 0.1201, + "step": 27119 + }, + { + "epoch": 0.23541462313695194, + "grad_norm": 0.11328125, + "learning_rate": 0.0016846390212884397, + "loss": 0.1504, + "step": 27120 + }, + { + "epoch": 0.2354233036171561, + "grad_norm": 0.46875, + "learning_rate": 0.0016846163932547973, + "loss": 0.0742, + "step": 27121 + }, + { + "epoch": 0.23543198409736027, + "grad_norm": 0.1083984375, + "learning_rate": 0.0016845937645818272, + "loss": 0.1572, + "step": 27122 + }, + { + "epoch": 0.23544066457756443, + "grad_norm": 0.109375, + "learning_rate": 0.0016845711352695546, + "loss": 0.0928, + "step": 27123 + }, + { + "epoch": 0.2354493450577686, + "grad_norm": 0.1279296875, + "learning_rate": 0.0016845485053180033, + "loss": 0.1035, + "step": 27124 + }, + { + "epoch": 0.23545802553797276, + "grad_norm": 0.1591796875, + "learning_rate": 0.0016845258747271987, + "loss": 0.1021, + "step": 27125 + }, + { + "epoch": 0.23546670601817693, + "grad_norm": 0.5703125, + "learning_rate": 0.0016845032434971653, + "loss": 0.1348, + "step": 27126 + }, + { + "epoch": 0.2354753864983811, + "grad_norm": 0.39453125, + "learning_rate": 0.0016844806116279281, + "loss": 0.1226, + "step": 27127 + }, + { + "epoch": 0.23548406697858526, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016844579791195115, + "loss": 0.123, + "step": 27128 + }, + { + "epoch": 0.23549274745878943, + "grad_norm": 0.4296875, + "learning_rate": 0.0016844353459719405, + "loss": 0.0996, + "step": 27129 + }, + { + "epoch": 0.2355014279389936, + "grad_norm": 0.55078125, + "learning_rate": 0.0016844127121852398, + "loss": 0.1963, + "step": 27130 + }, + { + "epoch": 0.23551010841919776, + "grad_norm": 0.484375, + "learning_rate": 0.0016843900777594341, + "loss": 0.0684, + "step": 27131 + }, + { + "epoch": 0.23551878889940192, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016843674426945483, + "loss": 0.1191, + "step": 27132 + }, + { + "epoch": 0.23552746937960609, + "grad_norm": 0.0888671875, + "learning_rate": 0.001684344806990607, + "loss": 0.0981, + "step": 27133 + }, + { + "epoch": 0.23553614985981025, + "grad_norm": 0.11962890625, + "learning_rate": 0.001684322170647635, + "loss": 0.123, + "step": 27134 + }, + { + "epoch": 0.23554483034001442, + "grad_norm": 0.50390625, + "learning_rate": 0.001684299533665657, + "loss": 0.124, + "step": 27135 + }, + { + "epoch": 0.23555351082021858, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016842768960446978, + "loss": 0.1104, + "step": 27136 + }, + { + "epoch": 0.23556219130042275, + "grad_norm": 0.640625, + "learning_rate": 0.001684254257784782, + "loss": 0.1299, + "step": 27137 + }, + { + "epoch": 0.2355708717806269, + "grad_norm": 0.1318359375, + "learning_rate": 0.001684231618885935, + "loss": 0.0811, + "step": 27138 + }, + { + "epoch": 0.23557955226083108, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016842089793481806, + "loss": 0.1138, + "step": 27139 + }, + { + "epoch": 0.23558823274103524, + "grad_norm": 0.11474609375, + "learning_rate": 0.0016841863391715442, + "loss": 0.1182, + "step": 27140 + }, + { + "epoch": 0.2355969132212394, + "grad_norm": 0.640625, + "learning_rate": 0.0016841636983560507, + "loss": 0.1094, + "step": 27141 + }, + { + "epoch": 0.23560559370144357, + "grad_norm": 0.212890625, + "learning_rate": 0.0016841410569017243, + "loss": 0.1201, + "step": 27142 + }, + { + "epoch": 0.23561427418164774, + "grad_norm": 0.07470703125, + "learning_rate": 0.00168411841480859, + "loss": 0.0806, + "step": 27143 + }, + { + "epoch": 0.2356229546618519, + "grad_norm": 0.12060546875, + "learning_rate": 0.0016840957720766726, + "loss": 0.1279, + "step": 27144 + }, + { + "epoch": 0.23563163514205607, + "grad_norm": 0.52734375, + "learning_rate": 0.0016840731287059967, + "loss": 0.1426, + "step": 27145 + }, + { + "epoch": 0.23564031562226023, + "grad_norm": 0.2412109375, + "learning_rate": 0.0016840504846965874, + "loss": 0.0957, + "step": 27146 + }, + { + "epoch": 0.2356489961024644, + "grad_norm": 0.2353515625, + "learning_rate": 0.0016840278400484694, + "loss": 0.1084, + "step": 27147 + }, + { + "epoch": 0.23565767658266856, + "grad_norm": 0.490234375, + "learning_rate": 0.001684005194761667, + "loss": 0.123, + "step": 27148 + }, + { + "epoch": 0.23566635706287273, + "grad_norm": 0.30859375, + "learning_rate": 0.0016839825488362053, + "loss": 0.083, + "step": 27149 + }, + { + "epoch": 0.2356750375430769, + "grad_norm": 0.09326171875, + "learning_rate": 0.0016839599022721093, + "loss": 0.085, + "step": 27150 + }, + { + "epoch": 0.23568371802328106, + "grad_norm": 0.1845703125, + "learning_rate": 0.0016839372550694035, + "loss": 0.0928, + "step": 27151 + }, + { + "epoch": 0.23569239850348522, + "grad_norm": 0.185546875, + "learning_rate": 0.0016839146072281123, + "loss": 0.0703, + "step": 27152 + }, + { + "epoch": 0.2357010789836894, + "grad_norm": 0.2392578125, + "learning_rate": 0.0016838919587482614, + "loss": 0.103, + "step": 27153 + }, + { + "epoch": 0.23570975946389355, + "grad_norm": 0.69921875, + "learning_rate": 0.0016838693096298746, + "loss": 0.0957, + "step": 27154 + }, + { + "epoch": 0.23571843994409772, + "grad_norm": 0.251953125, + "learning_rate": 0.0016838466598729774, + "loss": 0.0864, + "step": 27155 + }, + { + "epoch": 0.23572712042430188, + "grad_norm": 0.78515625, + "learning_rate": 0.001683824009477594, + "loss": 0.1055, + "step": 27156 + }, + { + "epoch": 0.23573580090450605, + "grad_norm": 0.6171875, + "learning_rate": 0.0016838013584437495, + "loss": 0.0938, + "step": 27157 + }, + { + "epoch": 0.2357444813847102, + "grad_norm": 0.2890625, + "learning_rate": 0.001683778706771469, + "loss": 0.1553, + "step": 27158 + }, + { + "epoch": 0.23575316186491438, + "grad_norm": 0.4921875, + "learning_rate": 0.0016837560544607764, + "loss": 0.0879, + "step": 27159 + }, + { + "epoch": 0.23576184234511854, + "grad_norm": 0.609375, + "learning_rate": 0.0016837334015116971, + "loss": 0.1318, + "step": 27160 + }, + { + "epoch": 0.2357705228253227, + "grad_norm": 0.1640625, + "learning_rate": 0.0016837107479242557, + "loss": 0.1235, + "step": 27161 + }, + { + "epoch": 0.23577920330552687, + "grad_norm": 0.9453125, + "learning_rate": 0.0016836880936984772, + "loss": 0.1099, + "step": 27162 + }, + { + "epoch": 0.23578788378573104, + "grad_norm": 1.421875, + "learning_rate": 0.001683665438834386, + "loss": 0.1387, + "step": 27163 + }, + { + "epoch": 0.2357965642659352, + "grad_norm": 0.8046875, + "learning_rate": 0.001683642783332007, + "loss": 0.1309, + "step": 27164 + }, + { + "epoch": 0.23580524474613937, + "grad_norm": 0.33984375, + "learning_rate": 0.001683620127191365, + "loss": 0.1191, + "step": 27165 + }, + { + "epoch": 0.23581392522634353, + "grad_norm": 0.44921875, + "learning_rate": 0.001683597470412485, + "loss": 0.0947, + "step": 27166 + }, + { + "epoch": 0.2358226057065477, + "grad_norm": 0.1005859375, + "learning_rate": 0.0016835748129953915, + "loss": 0.0933, + "step": 27167 + }, + { + "epoch": 0.23583128618675184, + "grad_norm": 0.130859375, + "learning_rate": 0.0016835521549401092, + "loss": 0.0933, + "step": 27168 + }, + { + "epoch": 0.235839966666956, + "grad_norm": 0.1328125, + "learning_rate": 0.001683529496246663, + "loss": 0.1055, + "step": 27169 + }, + { + "epoch": 0.23584864714716017, + "grad_norm": 0.10400390625, + "learning_rate": 0.0016835068369150778, + "loss": 0.0923, + "step": 27170 + }, + { + "epoch": 0.23585732762736433, + "grad_norm": 0.115234375, + "learning_rate": 0.0016834841769453783, + "loss": 0.1187, + "step": 27171 + }, + { + "epoch": 0.2358660081075685, + "grad_norm": 0.98046875, + "learning_rate": 0.0016834615163375892, + "loss": 0.0967, + "step": 27172 + }, + { + "epoch": 0.23587468858777266, + "grad_norm": 0.3515625, + "learning_rate": 0.0016834388550917357, + "loss": 0.1133, + "step": 27173 + }, + { + "epoch": 0.23588336906797683, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016834161932078417, + "loss": 0.0957, + "step": 27174 + }, + { + "epoch": 0.235892049548181, + "grad_norm": 0.37890625, + "learning_rate": 0.001683393530685933, + "loss": 0.0859, + "step": 27175 + }, + { + "epoch": 0.23590073002838516, + "grad_norm": 0.158203125, + "learning_rate": 0.0016833708675260339, + "loss": 0.0688, + "step": 27176 + }, + { + "epoch": 0.23590941050858932, + "grad_norm": 0.220703125, + "learning_rate": 0.001683348203728169, + "loss": 0.0908, + "step": 27177 + }, + { + "epoch": 0.2359180909887935, + "grad_norm": 0.18359375, + "learning_rate": 0.001683325539292363, + "loss": 0.1104, + "step": 27178 + }, + { + "epoch": 0.23592677146899765, + "grad_norm": 0.3828125, + "learning_rate": 0.0016833028742186413, + "loss": 0.1064, + "step": 27179 + }, + { + "epoch": 0.23593545194920182, + "grad_norm": 0.275390625, + "learning_rate": 0.0016832802085070282, + "loss": 0.0977, + "step": 27180 + }, + { + "epoch": 0.23594413242940598, + "grad_norm": 0.0634765625, + "learning_rate": 0.0016832575421575487, + "loss": 0.0771, + "step": 27181 + }, + { + "epoch": 0.23595281290961015, + "grad_norm": 0.265625, + "learning_rate": 0.0016832348751702276, + "loss": 0.084, + "step": 27182 + }, + { + "epoch": 0.23596149338981431, + "grad_norm": 0.2177734375, + "learning_rate": 0.0016832122075450893, + "loss": 0.0601, + "step": 27183 + }, + { + "epoch": 0.23597017387001848, + "grad_norm": 0.69140625, + "learning_rate": 0.0016831895392821592, + "loss": 0.1875, + "step": 27184 + }, + { + "epoch": 0.23597885435022264, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016831668703814616, + "loss": 0.0908, + "step": 27185 + }, + { + "epoch": 0.2359875348304268, + "grad_norm": 0.1328125, + "learning_rate": 0.0016831442008430216, + "loss": 0.1426, + "step": 27186 + }, + { + "epoch": 0.23599621531063097, + "grad_norm": 0.466796875, + "learning_rate": 0.0016831215306668636, + "loss": 0.0928, + "step": 27187 + }, + { + "epoch": 0.23600489579083514, + "grad_norm": 0.1025390625, + "learning_rate": 0.001683098859853013, + "loss": 0.0874, + "step": 27188 + }, + { + "epoch": 0.2360135762710393, + "grad_norm": 0.5, + "learning_rate": 0.0016830761884014942, + "loss": 0.1094, + "step": 27189 + }, + { + "epoch": 0.23602225675124347, + "grad_norm": 0.1875, + "learning_rate": 0.0016830535163123322, + "loss": 0.1777, + "step": 27190 + }, + { + "epoch": 0.23603093723144764, + "grad_norm": 0.10986328125, + "learning_rate": 0.001683030843585551, + "loss": 0.0977, + "step": 27191 + }, + { + "epoch": 0.2360396177116518, + "grad_norm": 0.1953125, + "learning_rate": 0.0016830081702211763, + "loss": 0.0908, + "step": 27192 + }, + { + "epoch": 0.23604829819185597, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016829854962192327, + "loss": 0.1074, + "step": 27193 + }, + { + "epoch": 0.23605697867206013, + "grad_norm": 0.51953125, + "learning_rate": 0.0016829628215797453, + "loss": 0.166, + "step": 27194 + }, + { + "epoch": 0.2360656591522643, + "grad_norm": 0.75390625, + "learning_rate": 0.0016829401463027381, + "loss": 0.1387, + "step": 27195 + }, + { + "epoch": 0.23607433963246846, + "grad_norm": 0.345703125, + "learning_rate": 0.001682917470388236, + "loss": 0.0879, + "step": 27196 + }, + { + "epoch": 0.23608302011267263, + "grad_norm": 0.091796875, + "learning_rate": 0.0016828947938362648, + "loss": 0.0664, + "step": 27197 + }, + { + "epoch": 0.2360917005928768, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016828721166468486, + "loss": 0.127, + "step": 27198 + }, + { + "epoch": 0.23610038107308096, + "grad_norm": 0.32421875, + "learning_rate": 0.0016828494388200117, + "loss": 0.1152, + "step": 27199 + }, + { + "epoch": 0.23610906155328512, + "grad_norm": 0.65625, + "learning_rate": 0.0016828267603557795, + "loss": 0.1602, + "step": 27200 + }, + { + "epoch": 0.2361177420334893, + "grad_norm": 0.734375, + "learning_rate": 0.001682804081254177, + "loss": 0.1377, + "step": 27201 + }, + { + "epoch": 0.23612642251369345, + "grad_norm": 0.09326171875, + "learning_rate": 0.0016827814015152286, + "loss": 0.0908, + "step": 27202 + }, + { + "epoch": 0.23613510299389762, + "grad_norm": 0.181640625, + "learning_rate": 0.0016827587211389592, + "loss": 0.1045, + "step": 27203 + }, + { + "epoch": 0.23614378347410178, + "grad_norm": 0.84765625, + "learning_rate": 0.0016827360401253935, + "loss": 0.0977, + "step": 27204 + }, + { + "epoch": 0.23615246395430595, + "grad_norm": 0.7109375, + "learning_rate": 0.0016827133584745566, + "loss": 0.1875, + "step": 27205 + }, + { + "epoch": 0.2361611444345101, + "grad_norm": 0.984375, + "learning_rate": 0.001682690676186473, + "loss": 0.0928, + "step": 27206 + }, + { + "epoch": 0.23616982491471428, + "grad_norm": 0.08984375, + "learning_rate": 0.0016826679932611676, + "loss": 0.1206, + "step": 27207 + }, + { + "epoch": 0.23617850539491844, + "grad_norm": 0.1904296875, + "learning_rate": 0.0016826453096986651, + "loss": 0.1172, + "step": 27208 + }, + { + "epoch": 0.2361871858751226, + "grad_norm": 1.171875, + "learning_rate": 0.0016826226254989909, + "loss": 0.1416, + "step": 27209 + }, + { + "epoch": 0.23619586635532677, + "grad_norm": 0.3671875, + "learning_rate": 0.0016825999406621692, + "loss": 0.1226, + "step": 27210 + }, + { + "epoch": 0.23620454683553094, + "grad_norm": 0.4375, + "learning_rate": 0.0016825772551882247, + "loss": 0.1357, + "step": 27211 + }, + { + "epoch": 0.2362132273157351, + "grad_norm": 0.470703125, + "learning_rate": 0.0016825545690771827, + "loss": 0.1084, + "step": 27212 + }, + { + "epoch": 0.23622190779593927, + "grad_norm": 0.154296875, + "learning_rate": 0.0016825318823290678, + "loss": 0.1484, + "step": 27213 + }, + { + "epoch": 0.23623058827614343, + "grad_norm": 0.51171875, + "learning_rate": 0.0016825091949439046, + "loss": 0.1064, + "step": 27214 + }, + { + "epoch": 0.2362392687563476, + "grad_norm": 0.357421875, + "learning_rate": 0.0016824865069217185, + "loss": 0.1069, + "step": 27215 + }, + { + "epoch": 0.23624794923655176, + "grad_norm": 0.65625, + "learning_rate": 0.0016824638182625334, + "loss": 0.1367, + "step": 27216 + }, + { + "epoch": 0.23625662971675593, + "grad_norm": 1.140625, + "learning_rate": 0.0016824411289663747, + "loss": 0.1445, + "step": 27217 + }, + { + "epoch": 0.2362653101969601, + "grad_norm": 0.3984375, + "learning_rate": 0.0016824184390332674, + "loss": 0.1162, + "step": 27218 + }, + { + "epoch": 0.23627399067716426, + "grad_norm": 0.66015625, + "learning_rate": 0.001682395748463236, + "loss": 0.1221, + "step": 27219 + }, + { + "epoch": 0.23628267115736842, + "grad_norm": 0.41015625, + "learning_rate": 0.0016823730572563052, + "loss": 0.1196, + "step": 27220 + }, + { + "epoch": 0.2362913516375726, + "grad_norm": 0.220703125, + "learning_rate": 0.0016823503654125002, + "loss": 0.125, + "step": 27221 + }, + { + "epoch": 0.23630003211777675, + "grad_norm": 0.349609375, + "learning_rate": 0.0016823276729318452, + "loss": 0.123, + "step": 27222 + }, + { + "epoch": 0.23630871259798092, + "grad_norm": 0.66015625, + "learning_rate": 0.0016823049798143654, + "loss": 0.1387, + "step": 27223 + }, + { + "epoch": 0.23631739307818508, + "grad_norm": 1.0625, + "learning_rate": 0.0016822822860600862, + "loss": 0.457, + "step": 27224 + }, + { + "epoch": 0.23632607355838925, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016822595916690312, + "loss": 0.1113, + "step": 27225 + }, + { + "epoch": 0.23633475403859341, + "grad_norm": 0.40625, + "learning_rate": 0.0016822368966412263, + "loss": 0.1348, + "step": 27226 + }, + { + "epoch": 0.23634343451879758, + "grad_norm": 0.126953125, + "learning_rate": 0.0016822142009766958, + "loss": 0.1143, + "step": 27227 + }, + { + "epoch": 0.23635211499900174, + "grad_norm": 0.7890625, + "learning_rate": 0.0016821915046754646, + "loss": 0.125, + "step": 27228 + }, + { + "epoch": 0.2363607954792059, + "grad_norm": 1.2578125, + "learning_rate": 0.0016821688077375573, + "loss": 0.125, + "step": 27229 + }, + { + "epoch": 0.23636947595941007, + "grad_norm": 0.63671875, + "learning_rate": 0.0016821461101629992, + "loss": 0.0938, + "step": 27230 + }, + { + "epoch": 0.23637815643961424, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016821234119518145, + "loss": 0.1006, + "step": 27231 + }, + { + "epoch": 0.2363868369198184, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016821007131040287, + "loss": 0.1123, + "step": 27232 + }, + { + "epoch": 0.23639551740002257, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016820780136196662, + "loss": 0.1523, + "step": 27233 + }, + { + "epoch": 0.23640419788022674, + "grad_norm": 0.2041015625, + "learning_rate": 0.001682055313498752, + "loss": 0.1221, + "step": 27234 + }, + { + "epoch": 0.2364128783604309, + "grad_norm": 0.498046875, + "learning_rate": 0.001682032612741311, + "loss": 0.0879, + "step": 27235 + }, + { + "epoch": 0.23642155884063507, + "grad_norm": 0.1962890625, + "learning_rate": 0.0016820099113473675, + "loss": 0.1934, + "step": 27236 + }, + { + "epoch": 0.23643023932083923, + "grad_norm": 0.2099609375, + "learning_rate": 0.001681987209316947, + "loss": 0.0981, + "step": 27237 + }, + { + "epoch": 0.2364389198010434, + "grad_norm": 0.19921875, + "learning_rate": 0.0016819645066500739, + "loss": 0.0986, + "step": 27238 + }, + { + "epoch": 0.23644760028124756, + "grad_norm": 0.34765625, + "learning_rate": 0.0016819418033467732, + "loss": 0.1406, + "step": 27239 + }, + { + "epoch": 0.23645628076145173, + "grad_norm": 0.1484375, + "learning_rate": 0.0016819190994070694, + "loss": 0.1445, + "step": 27240 + }, + { + "epoch": 0.2364649612416559, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016818963948309882, + "loss": 0.1069, + "step": 27241 + }, + { + "epoch": 0.23647364172186006, + "grad_norm": 0.63671875, + "learning_rate": 0.0016818736896185538, + "loss": 0.1196, + "step": 27242 + }, + { + "epoch": 0.23648232220206422, + "grad_norm": 0.69140625, + "learning_rate": 0.0016818509837697906, + "loss": 0.0928, + "step": 27243 + }, + { + "epoch": 0.2364910026822684, + "grad_norm": 0.8046875, + "learning_rate": 0.001681828277284724, + "loss": 0.1367, + "step": 27244 + }, + { + "epoch": 0.23649968316247255, + "grad_norm": 0.33984375, + "learning_rate": 0.001681805570163379, + "loss": 0.0918, + "step": 27245 + }, + { + "epoch": 0.23650836364267672, + "grad_norm": 0.458984375, + "learning_rate": 0.00168178286240578, + "loss": 0.1582, + "step": 27246 + }, + { + "epoch": 0.23651704412288088, + "grad_norm": 0.443359375, + "learning_rate": 0.0016817601540119519, + "loss": 0.1089, + "step": 27247 + }, + { + "epoch": 0.23652572460308505, + "grad_norm": 0.1357421875, + "learning_rate": 0.0016817374449819198, + "loss": 0.1299, + "step": 27248 + }, + { + "epoch": 0.2365344050832892, + "grad_norm": 0.080078125, + "learning_rate": 0.0016817147353157082, + "loss": 0.1094, + "step": 27249 + }, + { + "epoch": 0.23654308556349338, + "grad_norm": 0.33203125, + "learning_rate": 0.0016816920250133426, + "loss": 0.1494, + "step": 27250 + }, + { + "epoch": 0.23655176604369754, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016816693140748466, + "loss": 0.1318, + "step": 27251 + }, + { + "epoch": 0.2365604465239017, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016816466025002463, + "loss": 0.124, + "step": 27252 + }, + { + "epoch": 0.23656912700410587, + "grad_norm": 0.12060546875, + "learning_rate": 0.001681623890289566, + "loss": 0.1221, + "step": 27253 + }, + { + "epoch": 0.23657780748431004, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016816011774428306, + "loss": 0.1494, + "step": 27254 + }, + { + "epoch": 0.2365864879645142, + "grad_norm": 0.75390625, + "learning_rate": 0.0016815784639600646, + "loss": 0.1055, + "step": 27255 + }, + { + "epoch": 0.23659516844471837, + "grad_norm": 0.373046875, + "learning_rate": 0.0016815557498412932, + "loss": 0.1104, + "step": 27256 + }, + { + "epoch": 0.23660384892492253, + "grad_norm": 0.080078125, + "learning_rate": 0.0016815330350865411, + "loss": 0.1113, + "step": 27257 + }, + { + "epoch": 0.2366125294051267, + "grad_norm": 0.23828125, + "learning_rate": 0.0016815103196958336, + "loss": 0.1025, + "step": 27258 + }, + { + "epoch": 0.23662120988533086, + "grad_norm": 0.353515625, + "learning_rate": 0.0016814876036691949, + "loss": 0.124, + "step": 27259 + }, + { + "epoch": 0.23662989036553503, + "grad_norm": 0.44921875, + "learning_rate": 0.0016814648870066497, + "loss": 0.1328, + "step": 27260 + }, + { + "epoch": 0.2366385708457392, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016814421697082238, + "loss": 0.0981, + "step": 27261 + }, + { + "epoch": 0.23664725132594336, + "grad_norm": 0.185546875, + "learning_rate": 0.0016814194517739416, + "loss": 0.1201, + "step": 27262 + }, + { + "epoch": 0.23665593180614752, + "grad_norm": 0.1328125, + "learning_rate": 0.0016813967332038274, + "loss": 0.0889, + "step": 27263 + }, + { + "epoch": 0.2366646122863517, + "grad_norm": 2.34375, + "learning_rate": 0.0016813740139979064, + "loss": 0.1035, + "step": 27264 + }, + { + "epoch": 0.23667329276655585, + "grad_norm": 0.1982421875, + "learning_rate": 0.001681351294156204, + "loss": 0.0913, + "step": 27265 + }, + { + "epoch": 0.23668197324676002, + "grad_norm": 0.2138671875, + "learning_rate": 0.001681328573678744, + "loss": 0.2148, + "step": 27266 + }, + { + "epoch": 0.23669065372696418, + "grad_norm": 0.380859375, + "learning_rate": 0.0016813058525655521, + "loss": 0.1064, + "step": 27267 + }, + { + "epoch": 0.23669933420716835, + "grad_norm": 0.275390625, + "learning_rate": 0.0016812831308166529, + "loss": 0.1406, + "step": 27268 + }, + { + "epoch": 0.23670801468737251, + "grad_norm": 0.302734375, + "learning_rate": 0.0016812604084320713, + "loss": 0.1328, + "step": 27269 + }, + { + "epoch": 0.23671669516757668, + "grad_norm": 0.09814453125, + "learning_rate": 0.0016812376854118316, + "loss": 0.1162, + "step": 27270 + }, + { + "epoch": 0.23672537564778084, + "grad_norm": 0.47265625, + "learning_rate": 0.0016812149617559593, + "loss": 0.0796, + "step": 27271 + }, + { + "epoch": 0.236734056127985, + "grad_norm": 0.2734375, + "learning_rate": 0.0016811922374644793, + "loss": 0.1191, + "step": 27272 + }, + { + "epoch": 0.23674273660818917, + "grad_norm": 1.0859375, + "learning_rate": 0.001681169512537416, + "loss": 0.1035, + "step": 27273 + }, + { + "epoch": 0.23675141708839334, + "grad_norm": 0.37109375, + "learning_rate": 0.0016811467869747945, + "loss": 0.1025, + "step": 27274 + }, + { + "epoch": 0.2367600975685975, + "grad_norm": 0.396484375, + "learning_rate": 0.0016811240607766393, + "loss": 0.1338, + "step": 27275 + }, + { + "epoch": 0.23676877804880167, + "grad_norm": 0.52734375, + "learning_rate": 0.001681101333942976, + "loss": 0.104, + "step": 27276 + }, + { + "epoch": 0.23677745852900584, + "grad_norm": 0.2001953125, + "learning_rate": 0.001681078606473829, + "loss": 0.0933, + "step": 27277 + }, + { + "epoch": 0.23678613900921, + "grad_norm": 0.490234375, + "learning_rate": 0.0016810558783692232, + "loss": 0.1299, + "step": 27278 + }, + { + "epoch": 0.23679481948941417, + "grad_norm": 0.11669921875, + "learning_rate": 0.0016810331496291831, + "loss": 0.1611, + "step": 27279 + }, + { + "epoch": 0.23680349996961833, + "grad_norm": 0.125, + "learning_rate": 0.0016810104202537339, + "loss": 0.1182, + "step": 27280 + }, + { + "epoch": 0.2368121804498225, + "grad_norm": 0.130859375, + "learning_rate": 0.0016809876902429007, + "loss": 0.0933, + "step": 27281 + }, + { + "epoch": 0.23682086093002666, + "grad_norm": 2.0625, + "learning_rate": 0.0016809649595967077, + "loss": 0.1309, + "step": 27282 + }, + { + "epoch": 0.23682954141023083, + "grad_norm": 0.578125, + "learning_rate": 0.0016809422283151805, + "loss": 0.0962, + "step": 27283 + }, + { + "epoch": 0.236838221890435, + "grad_norm": 0.5390625, + "learning_rate": 0.0016809194963983436, + "loss": 0.1309, + "step": 27284 + }, + { + "epoch": 0.23684690237063916, + "grad_norm": 0.310546875, + "learning_rate": 0.0016808967638462218, + "loss": 0.0923, + "step": 27285 + }, + { + "epoch": 0.23685558285084332, + "grad_norm": 0.796875, + "learning_rate": 0.00168087403065884, + "loss": 0.1328, + "step": 27286 + }, + { + "epoch": 0.2368642633310475, + "grad_norm": 0.28515625, + "learning_rate": 0.0016808512968362228, + "loss": 0.1934, + "step": 27287 + }, + { + "epoch": 0.23687294381125165, + "grad_norm": 0.79296875, + "learning_rate": 0.0016808285623783957, + "loss": 0.123, + "step": 27288 + }, + { + "epoch": 0.23688162429145582, + "grad_norm": 0.279296875, + "learning_rate": 0.0016808058272853833, + "loss": 0.1133, + "step": 27289 + }, + { + "epoch": 0.23689030477165998, + "grad_norm": 0.1005859375, + "learning_rate": 0.0016807830915572103, + "loss": 0.1064, + "step": 27290 + }, + { + "epoch": 0.23689898525186412, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016807603551939015, + "loss": 0.1001, + "step": 27291 + }, + { + "epoch": 0.23690766573206828, + "grad_norm": 0.34765625, + "learning_rate": 0.001680737618195482, + "loss": 0.1221, + "step": 27292 + }, + { + "epoch": 0.23691634621227245, + "grad_norm": 0.5625, + "learning_rate": 0.0016807148805619762, + "loss": 0.1079, + "step": 27293 + }, + { + "epoch": 0.23692502669247661, + "grad_norm": 0.1494140625, + "learning_rate": 0.00168069214229341, + "loss": 0.1416, + "step": 27294 + }, + { + "epoch": 0.23693370717268078, + "grad_norm": 0.1591796875, + "learning_rate": 0.001680669403389807, + "loss": 0.1196, + "step": 27295 + }, + { + "epoch": 0.23694238765288494, + "grad_norm": 0.54296875, + "learning_rate": 0.001680646663851193, + "loss": 0.1357, + "step": 27296 + }, + { + "epoch": 0.2369510681330891, + "grad_norm": 0.158203125, + "learning_rate": 0.0016806239236775923, + "loss": 0.1099, + "step": 27297 + }, + { + "epoch": 0.23695974861329328, + "grad_norm": 0.46875, + "learning_rate": 0.0016806011828690301, + "loss": 0.1318, + "step": 27298 + }, + { + "epoch": 0.23696842909349744, + "grad_norm": 0.82421875, + "learning_rate": 0.0016805784414255314, + "loss": 0.1084, + "step": 27299 + }, + { + "epoch": 0.2369771095737016, + "grad_norm": 0.45703125, + "learning_rate": 0.0016805556993471208, + "loss": 0.1055, + "step": 27300 + }, + { + "epoch": 0.23698579005390577, + "grad_norm": 0.71875, + "learning_rate": 0.001680532956633823, + "loss": 0.0972, + "step": 27301 + }, + { + "epoch": 0.23699447053410994, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016805102132856633, + "loss": 0.1367, + "step": 27302 + }, + { + "epoch": 0.2370031510143141, + "grad_norm": 0.1181640625, + "learning_rate": 0.001680487469302666, + "loss": 0.1543, + "step": 27303 + }, + { + "epoch": 0.23701183149451827, + "grad_norm": 0.55078125, + "learning_rate": 0.0016804647246848566, + "loss": 0.1904, + "step": 27304 + }, + { + "epoch": 0.23702051197472243, + "grad_norm": 0.162109375, + "learning_rate": 0.0016804419794322599, + "loss": 0.1245, + "step": 27305 + }, + { + "epoch": 0.2370291924549266, + "grad_norm": 0.271484375, + "learning_rate": 0.0016804192335449003, + "loss": 0.0894, + "step": 27306 + }, + { + "epoch": 0.23703787293513076, + "grad_norm": 0.7265625, + "learning_rate": 0.0016803964870228025, + "loss": 0.1069, + "step": 27307 + }, + { + "epoch": 0.23704655341533493, + "grad_norm": 0.10595703125, + "learning_rate": 0.0016803737398659925, + "loss": 0.1162, + "step": 27308 + }, + { + "epoch": 0.2370552338955391, + "grad_norm": 0.27734375, + "learning_rate": 0.0016803509920744941, + "loss": 0.1123, + "step": 27309 + }, + { + "epoch": 0.23706391437574326, + "grad_norm": 0.228515625, + "learning_rate": 0.0016803282436483329, + "loss": 0.1162, + "step": 27310 + }, + { + "epoch": 0.23707259485594742, + "grad_norm": 0.59765625, + "learning_rate": 0.0016803054945875332, + "loss": 0.0957, + "step": 27311 + }, + { + "epoch": 0.2370812753361516, + "grad_norm": 0.326171875, + "learning_rate": 0.0016802827448921204, + "loss": 0.1367, + "step": 27312 + }, + { + "epoch": 0.23708995581635575, + "grad_norm": 0.158203125, + "learning_rate": 0.0016802599945621186, + "loss": 0.1523, + "step": 27313 + }, + { + "epoch": 0.23709863629655992, + "grad_norm": 0.3359375, + "learning_rate": 0.0016802372435975536, + "loss": 0.1123, + "step": 27314 + }, + { + "epoch": 0.23710731677676408, + "grad_norm": 0.1064453125, + "learning_rate": 0.00168021449199845, + "loss": 0.0928, + "step": 27315 + }, + { + "epoch": 0.23711599725696825, + "grad_norm": 0.19921875, + "learning_rate": 0.0016801917397648322, + "loss": 0.0874, + "step": 27316 + }, + { + "epoch": 0.2371246777371724, + "grad_norm": 0.1240234375, + "learning_rate": 0.0016801689868967257, + "loss": 0.1338, + "step": 27317 + }, + { + "epoch": 0.23713335821737658, + "grad_norm": 0.4375, + "learning_rate": 0.001680146233394155, + "loss": 0.1787, + "step": 27318 + }, + { + "epoch": 0.23714203869758074, + "grad_norm": 0.4453125, + "learning_rate": 0.0016801234792571452, + "loss": 0.1289, + "step": 27319 + }, + { + "epoch": 0.2371507191777849, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016801007244857209, + "loss": 0.1338, + "step": 27320 + }, + { + "epoch": 0.23715939965798907, + "grad_norm": 0.546875, + "learning_rate": 0.0016800779690799073, + "loss": 0.124, + "step": 27321 + }, + { + "epoch": 0.23716808013819324, + "grad_norm": 0.09423828125, + "learning_rate": 0.0016800552130397288, + "loss": 0.0938, + "step": 27322 + }, + { + "epoch": 0.2371767606183974, + "grad_norm": 0.48828125, + "learning_rate": 0.001680032456365211, + "loss": 0.1016, + "step": 27323 + }, + { + "epoch": 0.23718544109860157, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016800096990563784, + "loss": 0.1172, + "step": 27324 + }, + { + "epoch": 0.23719412157880573, + "grad_norm": 0.19140625, + "learning_rate": 0.0016799869411132555, + "loss": 0.0894, + "step": 27325 + }, + { + "epoch": 0.2372028020590099, + "grad_norm": 0.76953125, + "learning_rate": 0.0016799641825358683, + "loss": 0.1104, + "step": 27326 + }, + { + "epoch": 0.23721148253921406, + "grad_norm": 1.1484375, + "learning_rate": 0.0016799414233242406, + "loss": 0.1143, + "step": 27327 + }, + { + "epoch": 0.23722016301941823, + "grad_norm": 0.4296875, + "learning_rate": 0.0016799186634783975, + "loss": 0.0918, + "step": 27328 + }, + { + "epoch": 0.2372288434996224, + "grad_norm": 0.69140625, + "learning_rate": 0.0016798959029983643, + "loss": 0.1992, + "step": 27329 + }, + { + "epoch": 0.23723752397982656, + "grad_norm": 0.08447265625, + "learning_rate": 0.0016798731418841656, + "loss": 0.0825, + "step": 27330 + }, + { + "epoch": 0.23724620446003072, + "grad_norm": 0.87109375, + "learning_rate": 0.0016798503801358263, + "loss": 0.1143, + "step": 27331 + }, + { + "epoch": 0.2372548849402349, + "grad_norm": 0.6328125, + "learning_rate": 0.0016798276177533715, + "loss": 0.0938, + "step": 27332 + }, + { + "epoch": 0.23726356542043905, + "grad_norm": 0.08349609375, + "learning_rate": 0.0016798048547368256, + "loss": 0.127, + "step": 27333 + }, + { + "epoch": 0.23727224590064322, + "grad_norm": 0.275390625, + "learning_rate": 0.0016797820910862142, + "loss": 0.0693, + "step": 27334 + }, + { + "epoch": 0.23728092638084738, + "grad_norm": 0.43359375, + "learning_rate": 0.0016797593268015618, + "loss": 0.0908, + "step": 27335 + }, + { + "epoch": 0.23728960686105155, + "grad_norm": 0.08154296875, + "learning_rate": 0.0016797365618828931, + "loss": 0.1021, + "step": 27336 + }, + { + "epoch": 0.23729828734125571, + "grad_norm": 0.875, + "learning_rate": 0.001679713796330233, + "loss": 0.1455, + "step": 27337 + }, + { + "epoch": 0.23730696782145988, + "grad_norm": 0.10107421875, + "learning_rate": 0.001679691030143607, + "loss": 0.1367, + "step": 27338 + }, + { + "epoch": 0.23731564830166405, + "grad_norm": 0.1650390625, + "learning_rate": 0.0016796682633230396, + "loss": 0.1172, + "step": 27339 + }, + { + "epoch": 0.2373243287818682, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016796454958685554, + "loss": 0.085, + "step": 27340 + }, + { + "epoch": 0.23733300926207238, + "grad_norm": 0.33984375, + "learning_rate": 0.00167962272778018, + "loss": 0.1357, + "step": 27341 + }, + { + "epoch": 0.23734168974227654, + "grad_norm": 0.98046875, + "learning_rate": 0.0016795999590579374, + "loss": 0.1357, + "step": 27342 + }, + { + "epoch": 0.2373503702224807, + "grad_norm": 0.490234375, + "learning_rate": 0.0016795771897018534, + "loss": 0.1172, + "step": 27343 + }, + { + "epoch": 0.23735905070268487, + "grad_norm": 0.302734375, + "learning_rate": 0.0016795544197119522, + "loss": 0.1025, + "step": 27344 + }, + { + "epoch": 0.23736773118288904, + "grad_norm": 0.1181640625, + "learning_rate": 0.0016795316490882592, + "loss": 0.0684, + "step": 27345 + }, + { + "epoch": 0.2373764116630932, + "grad_norm": 0.322265625, + "learning_rate": 0.0016795088778307993, + "loss": 0.1094, + "step": 27346 + }, + { + "epoch": 0.23738509214329737, + "grad_norm": 0.66796875, + "learning_rate": 0.0016794861059395967, + "loss": 0.1143, + "step": 27347 + }, + { + "epoch": 0.23739377262350153, + "grad_norm": 0.41015625, + "learning_rate": 0.0016794633334146771, + "loss": 0.0845, + "step": 27348 + }, + { + "epoch": 0.2374024531037057, + "grad_norm": 0.279296875, + "learning_rate": 0.001679440560256065, + "loss": 0.0771, + "step": 27349 + }, + { + "epoch": 0.23741113358390986, + "grad_norm": 0.263671875, + "learning_rate": 0.0016794177864637855, + "loss": 0.126, + "step": 27350 + }, + { + "epoch": 0.23741981406411403, + "grad_norm": 0.46875, + "learning_rate": 0.001679395012037863, + "loss": 0.0859, + "step": 27351 + }, + { + "epoch": 0.2374284945443182, + "grad_norm": 0.67578125, + "learning_rate": 0.0016793722369783234, + "loss": 0.1377, + "step": 27352 + }, + { + "epoch": 0.23743717502452236, + "grad_norm": 0.359375, + "learning_rate": 0.001679349461285191, + "loss": 0.1211, + "step": 27353 + }, + { + "epoch": 0.23744585550472652, + "grad_norm": 0.4140625, + "learning_rate": 0.0016793266849584904, + "loss": 0.1123, + "step": 27354 + }, + { + "epoch": 0.2374545359849307, + "grad_norm": 0.73828125, + "learning_rate": 0.0016793039079982472, + "loss": 0.1143, + "step": 27355 + }, + { + "epoch": 0.23746321646513485, + "grad_norm": 0.369140625, + "learning_rate": 0.0016792811304044855, + "loss": 0.0649, + "step": 27356 + }, + { + "epoch": 0.23747189694533902, + "grad_norm": 0.2578125, + "learning_rate": 0.0016792583521772308, + "loss": 0.0967, + "step": 27357 + }, + { + "epoch": 0.23748057742554318, + "grad_norm": 0.09716796875, + "learning_rate": 0.001679235573316508, + "loss": 0.0933, + "step": 27358 + }, + { + "epoch": 0.23748925790574735, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016792127938223419, + "loss": 0.1387, + "step": 27359 + }, + { + "epoch": 0.2374979383859515, + "grad_norm": 0.3359375, + "learning_rate": 0.0016791900136947572, + "loss": 0.1216, + "step": 27360 + }, + { + "epoch": 0.23750661886615568, + "grad_norm": 0.3984375, + "learning_rate": 0.0016791672329337793, + "loss": 0.0854, + "step": 27361 + }, + { + "epoch": 0.23751529934635984, + "grad_norm": 0.384765625, + "learning_rate": 0.0016791444515394326, + "loss": 0.127, + "step": 27362 + }, + { + "epoch": 0.237523979826564, + "grad_norm": 1.1796875, + "learning_rate": 0.0016791216695117425, + "loss": 0.0986, + "step": 27363 + }, + { + "epoch": 0.23753266030676817, + "grad_norm": 0.1650390625, + "learning_rate": 0.0016790988868507332, + "loss": 0.0967, + "step": 27364 + }, + { + "epoch": 0.23754134078697234, + "grad_norm": 0.2431640625, + "learning_rate": 0.0016790761035564304, + "loss": 0.1348, + "step": 27365 + }, + { + "epoch": 0.2375500212671765, + "grad_norm": 0.5625, + "learning_rate": 0.0016790533196288583, + "loss": 0.0972, + "step": 27366 + }, + { + "epoch": 0.23755870174738067, + "grad_norm": 0.66015625, + "learning_rate": 0.0016790305350680426, + "loss": 0.0811, + "step": 27367 + }, + { + "epoch": 0.23756738222758483, + "grad_norm": 0.25390625, + "learning_rate": 0.0016790077498740074, + "loss": 0.1416, + "step": 27368 + }, + { + "epoch": 0.237576062707789, + "grad_norm": 0.08544921875, + "learning_rate": 0.0016789849640467785, + "loss": 0.083, + "step": 27369 + }, + { + "epoch": 0.23758474318799316, + "grad_norm": 0.30078125, + "learning_rate": 0.00167896217758638, + "loss": 0.1133, + "step": 27370 + }, + { + "epoch": 0.23759342366819733, + "grad_norm": 0.37890625, + "learning_rate": 0.0016789393904928375, + "loss": 0.125, + "step": 27371 + }, + { + "epoch": 0.2376021041484015, + "grad_norm": 0.5625, + "learning_rate": 0.0016789166027661754, + "loss": 0.1387, + "step": 27372 + }, + { + "epoch": 0.23761078462860566, + "grad_norm": 0.251953125, + "learning_rate": 0.0016788938144064185, + "loss": 0.0996, + "step": 27373 + }, + { + "epoch": 0.23761946510880982, + "grad_norm": 0.26171875, + "learning_rate": 0.0016788710254135926, + "loss": 0.0947, + "step": 27374 + }, + { + "epoch": 0.237628145589014, + "grad_norm": 0.13671875, + "learning_rate": 0.0016788482357877217, + "loss": 0.1377, + "step": 27375 + }, + { + "epoch": 0.23763682606921815, + "grad_norm": 0.10888671875, + "learning_rate": 0.001678825445528831, + "loss": 0.1113, + "step": 27376 + }, + { + "epoch": 0.23764550654942232, + "grad_norm": 0.890625, + "learning_rate": 0.0016788026546369454, + "loss": 0.1196, + "step": 27377 + }, + { + "epoch": 0.23765418702962648, + "grad_norm": 0.37890625, + "learning_rate": 0.0016787798631120903, + "loss": 0.0996, + "step": 27378 + }, + { + "epoch": 0.23766286750983065, + "grad_norm": 0.24609375, + "learning_rate": 0.00167875707095429, + "loss": 0.0859, + "step": 27379 + }, + { + "epoch": 0.23767154799003481, + "grad_norm": 0.439453125, + "learning_rate": 0.00167873427816357, + "loss": 0.1143, + "step": 27380 + }, + { + "epoch": 0.23768022847023898, + "grad_norm": 0.2451171875, + "learning_rate": 0.0016787114847399542, + "loss": 0.0923, + "step": 27381 + }, + { + "epoch": 0.23768890895044315, + "grad_norm": 0.208984375, + "learning_rate": 0.0016786886906834688, + "loss": 0.1279, + "step": 27382 + }, + { + "epoch": 0.2376975894306473, + "grad_norm": 0.1328125, + "learning_rate": 0.001678665895994138, + "loss": 0.0996, + "step": 27383 + }, + { + "epoch": 0.23770626991085148, + "grad_norm": 0.306640625, + "learning_rate": 0.0016786431006719869, + "loss": 0.104, + "step": 27384 + }, + { + "epoch": 0.23771495039105564, + "grad_norm": 0.5234375, + "learning_rate": 0.00167862030471704, + "loss": 0.0938, + "step": 27385 + }, + { + "epoch": 0.2377236308712598, + "grad_norm": 0.54296875, + "learning_rate": 0.001678597508129323, + "loss": 0.0879, + "step": 27386 + }, + { + "epoch": 0.23773231135146397, + "grad_norm": 0.365234375, + "learning_rate": 0.0016785747109088606, + "loss": 0.1162, + "step": 27387 + }, + { + "epoch": 0.23774099183166814, + "grad_norm": 0.40234375, + "learning_rate": 0.0016785519130556774, + "loss": 0.0972, + "step": 27388 + }, + { + "epoch": 0.2377496723118723, + "grad_norm": 0.48046875, + "learning_rate": 0.0016785291145697983, + "loss": 0.0762, + "step": 27389 + }, + { + "epoch": 0.23775835279207647, + "grad_norm": 0.21484375, + "learning_rate": 0.001678506315451249, + "loss": 0.0825, + "step": 27390 + }, + { + "epoch": 0.23776703327228063, + "grad_norm": 0.408203125, + "learning_rate": 0.0016784835157000536, + "loss": 0.1055, + "step": 27391 + }, + { + "epoch": 0.2377757137524848, + "grad_norm": 0.314453125, + "learning_rate": 0.001678460715316237, + "loss": 0.0957, + "step": 27392 + }, + { + "epoch": 0.23778439423268896, + "grad_norm": 0.12158203125, + "learning_rate": 0.0016784379142998247, + "loss": 0.1094, + "step": 27393 + }, + { + "epoch": 0.23779307471289313, + "grad_norm": 0.294921875, + "learning_rate": 0.0016784151126508413, + "loss": 0.1182, + "step": 27394 + }, + { + "epoch": 0.2378017551930973, + "grad_norm": 0.54296875, + "learning_rate": 0.0016783923103693122, + "loss": 0.1797, + "step": 27395 + }, + { + "epoch": 0.23781043567330146, + "grad_norm": 0.33984375, + "learning_rate": 0.0016783695074552617, + "loss": 0.1895, + "step": 27396 + }, + { + "epoch": 0.23781911615350562, + "grad_norm": 0.515625, + "learning_rate": 0.0016783467039087148, + "loss": 0.1016, + "step": 27397 + }, + { + "epoch": 0.2378277966337098, + "grad_norm": 0.5234375, + "learning_rate": 0.001678323899729697, + "loss": 0.1001, + "step": 27398 + }, + { + "epoch": 0.23783647711391395, + "grad_norm": 0.1953125, + "learning_rate": 0.0016783010949182324, + "loss": 0.1147, + "step": 27399 + }, + { + "epoch": 0.23784515759411812, + "grad_norm": 0.64453125, + "learning_rate": 0.001678278289474347, + "loss": 0.1138, + "step": 27400 + }, + { + "epoch": 0.23785383807432228, + "grad_norm": 0.44140625, + "learning_rate": 0.001678255483398065, + "loss": 0.1138, + "step": 27401 + }, + { + "epoch": 0.23786251855452645, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016782326766894113, + "loss": 0.1182, + "step": 27402 + }, + { + "epoch": 0.2378711990347306, + "grad_norm": 0.28125, + "learning_rate": 0.0016782098693484111, + "loss": 0.1426, + "step": 27403 + }, + { + "epoch": 0.23787987951493478, + "grad_norm": 0.4140625, + "learning_rate": 0.001678187061375089, + "loss": 0.1196, + "step": 27404 + }, + { + "epoch": 0.23788855999513894, + "grad_norm": 0.197265625, + "learning_rate": 0.0016781642527694705, + "loss": 0.0903, + "step": 27405 + }, + { + "epoch": 0.2378972404753431, + "grad_norm": 0.453125, + "learning_rate": 0.0016781414435315806, + "loss": 0.1338, + "step": 27406 + }, + { + "epoch": 0.23790592095554727, + "grad_norm": 0.416015625, + "learning_rate": 0.0016781186336614433, + "loss": 0.123, + "step": 27407 + }, + { + "epoch": 0.23791460143575144, + "grad_norm": 0.1455078125, + "learning_rate": 0.0016780958231590844, + "loss": 0.1523, + "step": 27408 + }, + { + "epoch": 0.2379232819159556, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016780730120245285, + "loss": 0.123, + "step": 27409 + }, + { + "epoch": 0.23793196239615977, + "grad_norm": 0.12255859375, + "learning_rate": 0.0016780502002578007, + "loss": 0.0923, + "step": 27410 + }, + { + "epoch": 0.23794064287636393, + "grad_norm": 0.412109375, + "learning_rate": 0.001678027387858926, + "loss": 0.1152, + "step": 27411 + }, + { + "epoch": 0.2379493233565681, + "grad_norm": 0.248046875, + "learning_rate": 0.001678004574827929, + "loss": 0.0898, + "step": 27412 + }, + { + "epoch": 0.23795800383677226, + "grad_norm": 0.2490234375, + "learning_rate": 0.0016779817611648352, + "loss": 0.1426, + "step": 27413 + }, + { + "epoch": 0.23796668431697643, + "grad_norm": 0.09033203125, + "learning_rate": 0.001677958946869669, + "loss": 0.1016, + "step": 27414 + }, + { + "epoch": 0.23797536479718057, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016779361319424558, + "loss": 0.124, + "step": 27415 + }, + { + "epoch": 0.23798404527738473, + "grad_norm": 1.0625, + "learning_rate": 0.0016779133163832198, + "loss": 0.126, + "step": 27416 + }, + { + "epoch": 0.2379927257575889, + "grad_norm": 0.703125, + "learning_rate": 0.001677890500191987, + "loss": 0.1328, + "step": 27417 + }, + { + "epoch": 0.23800140623779306, + "grad_norm": 0.3515625, + "learning_rate": 0.001677867683368782, + "loss": 0.0703, + "step": 27418 + }, + { + "epoch": 0.23801008671799723, + "grad_norm": 0.326171875, + "learning_rate": 0.001677844865913629, + "loss": 0.1328, + "step": 27419 + }, + { + "epoch": 0.2380187671982014, + "grad_norm": 0.91015625, + "learning_rate": 0.0016778220478265535, + "loss": 0.1025, + "step": 27420 + }, + { + "epoch": 0.23802744767840556, + "grad_norm": 0.416015625, + "learning_rate": 0.001677799229107581, + "loss": 0.1973, + "step": 27421 + }, + { + "epoch": 0.23803612815860972, + "grad_norm": 1.2890625, + "learning_rate": 0.0016777764097567359, + "loss": 0.1299, + "step": 27422 + }, + { + "epoch": 0.2380448086388139, + "grad_norm": 0.1240234375, + "learning_rate": 0.001677753589774043, + "loss": 0.1035, + "step": 27423 + }, + { + "epoch": 0.23805348911901805, + "grad_norm": 0.0830078125, + "learning_rate": 0.0016777307691595274, + "loss": 0.103, + "step": 27424 + }, + { + "epoch": 0.23806216959922222, + "grad_norm": 0.515625, + "learning_rate": 0.001677707947913214, + "loss": 0.1045, + "step": 27425 + }, + { + "epoch": 0.23807085007942638, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016776851260351282, + "loss": 0.1147, + "step": 27426 + }, + { + "epoch": 0.23807953055963055, + "grad_norm": 0.275390625, + "learning_rate": 0.0016776623035252944, + "loss": 0.1113, + "step": 27427 + }, + { + "epoch": 0.2380882110398347, + "grad_norm": 0.36328125, + "learning_rate": 0.001677639480383738, + "loss": 0.0898, + "step": 27428 + }, + { + "epoch": 0.23809689152003888, + "grad_norm": 0.177734375, + "learning_rate": 0.0016776166566104835, + "loss": 0.1201, + "step": 27429 + }, + { + "epoch": 0.23810557200024304, + "grad_norm": 0.3125, + "learning_rate": 0.0016775938322055562, + "loss": 0.0791, + "step": 27430 + }, + { + "epoch": 0.2381142524804472, + "grad_norm": 0.359375, + "learning_rate": 0.0016775710071689812, + "loss": 0.0898, + "step": 27431 + }, + { + "epoch": 0.23812293296065137, + "grad_norm": 0.173828125, + "learning_rate": 0.0016775481815007832, + "loss": 0.0928, + "step": 27432 + }, + { + "epoch": 0.23813161344085554, + "grad_norm": 0.240234375, + "learning_rate": 0.001677525355200987, + "loss": 0.1006, + "step": 27433 + }, + { + "epoch": 0.2381402939210597, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016775025282696177, + "loss": 0.0898, + "step": 27434 + }, + { + "epoch": 0.23814897440126387, + "grad_norm": 0.1748046875, + "learning_rate": 0.0016774797007067003, + "loss": 0.0825, + "step": 27435 + }, + { + "epoch": 0.23815765488146803, + "grad_norm": 0.126953125, + "learning_rate": 0.0016774568725122596, + "loss": 0.1338, + "step": 27436 + }, + { + "epoch": 0.2381663353616722, + "grad_norm": 0.39453125, + "learning_rate": 0.001677434043686321, + "loss": 0.1123, + "step": 27437 + }, + { + "epoch": 0.23817501584187636, + "grad_norm": 0.0673828125, + "learning_rate": 0.0016774112142289093, + "loss": 0.0742, + "step": 27438 + }, + { + "epoch": 0.23818369632208053, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016773883841400493, + "loss": 0.0957, + "step": 27439 + }, + { + "epoch": 0.2381923768022847, + "grad_norm": 0.4921875, + "learning_rate": 0.0016773655534197658, + "loss": 0.1279, + "step": 27440 + }, + { + "epoch": 0.23820105728248886, + "grad_norm": 0.16015625, + "learning_rate": 0.0016773427220680844, + "loss": 0.1157, + "step": 27441 + }, + { + "epoch": 0.23820973776269302, + "grad_norm": 0.1123046875, + "learning_rate": 0.0016773198900850294, + "loss": 0.0981, + "step": 27442 + }, + { + "epoch": 0.2382184182428972, + "grad_norm": 0.27734375, + "learning_rate": 0.001677297057470626, + "loss": 0.0928, + "step": 27443 + }, + { + "epoch": 0.23822709872310135, + "grad_norm": 0.16015625, + "learning_rate": 0.0016772742242248996, + "loss": 0.1045, + "step": 27444 + }, + { + "epoch": 0.23823577920330552, + "grad_norm": 0.130859375, + "learning_rate": 0.0016772513903478743, + "loss": 0.1582, + "step": 27445 + }, + { + "epoch": 0.23824445968350969, + "grad_norm": 0.1796875, + "learning_rate": 0.0016772285558395758, + "loss": 0.1006, + "step": 27446 + }, + { + "epoch": 0.23825314016371385, + "grad_norm": 0.99609375, + "learning_rate": 0.0016772057207000287, + "loss": 0.1318, + "step": 27447 + }, + { + "epoch": 0.23826182064391802, + "grad_norm": 0.08984375, + "learning_rate": 0.001677182884929258, + "loss": 0.0791, + "step": 27448 + }, + { + "epoch": 0.23827050112412218, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016771600485272892, + "loss": 0.1348, + "step": 27449 + }, + { + "epoch": 0.23827918160432635, + "grad_norm": 0.212890625, + "learning_rate": 0.0016771372114941466, + "loss": 0.1299, + "step": 27450 + }, + { + "epoch": 0.2382878620845305, + "grad_norm": 0.578125, + "learning_rate": 0.0016771143738298553, + "loss": 0.0894, + "step": 27451 + }, + { + "epoch": 0.23829654256473468, + "grad_norm": 0.376953125, + "learning_rate": 0.0016770915355344405, + "loss": 0.1201, + "step": 27452 + }, + { + "epoch": 0.23830522304493884, + "grad_norm": 0.1640625, + "learning_rate": 0.0016770686966079275, + "loss": 0.1094, + "step": 27453 + }, + { + "epoch": 0.238313903525143, + "grad_norm": 0.318359375, + "learning_rate": 0.0016770458570503403, + "loss": 0.083, + "step": 27454 + }, + { + "epoch": 0.23832258400534717, + "grad_norm": 0.115234375, + "learning_rate": 0.0016770230168617042, + "loss": 0.0869, + "step": 27455 + }, + { + "epoch": 0.23833126448555134, + "grad_norm": 0.74609375, + "learning_rate": 0.0016770001760420448, + "loss": 0.1143, + "step": 27456 + }, + { + "epoch": 0.2383399449657555, + "grad_norm": 0.392578125, + "learning_rate": 0.0016769773345913868, + "loss": 0.0952, + "step": 27457 + }, + { + "epoch": 0.23834862544595967, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016769544925097546, + "loss": 0.0903, + "step": 27458 + }, + { + "epoch": 0.23835730592616383, + "grad_norm": 0.50390625, + "learning_rate": 0.0016769316497971737, + "loss": 0.1221, + "step": 27459 + }, + { + "epoch": 0.238365986406368, + "grad_norm": 0.158203125, + "learning_rate": 0.0016769088064536694, + "loss": 0.1152, + "step": 27460 + }, + { + "epoch": 0.23837466688657216, + "grad_norm": 0.09326171875, + "learning_rate": 0.0016768859624792663, + "loss": 0.1045, + "step": 27461 + }, + { + "epoch": 0.23838334736677633, + "grad_norm": 0.109375, + "learning_rate": 0.0016768631178739892, + "loss": 0.1484, + "step": 27462 + }, + { + "epoch": 0.2383920278469805, + "grad_norm": 0.2451171875, + "learning_rate": 0.001676840272637863, + "loss": 0.1299, + "step": 27463 + }, + { + "epoch": 0.23840070832718466, + "grad_norm": 0.62109375, + "learning_rate": 0.0016768174267709132, + "loss": 0.1074, + "step": 27464 + }, + { + "epoch": 0.23840938880738882, + "grad_norm": 0.40234375, + "learning_rate": 0.0016767945802731645, + "loss": 0.0908, + "step": 27465 + }, + { + "epoch": 0.238418069287593, + "grad_norm": 0.09423828125, + "learning_rate": 0.001676771733144642, + "loss": 0.1064, + "step": 27466 + }, + { + "epoch": 0.23842674976779715, + "grad_norm": 0.63671875, + "learning_rate": 0.0016767488853853705, + "loss": 0.0933, + "step": 27467 + }, + { + "epoch": 0.23843543024800132, + "grad_norm": 0.287109375, + "learning_rate": 0.0016767260369953752, + "loss": 0.1055, + "step": 27468 + }, + { + "epoch": 0.23844411072820548, + "grad_norm": 0.0888671875, + "learning_rate": 0.001676703187974681, + "loss": 0.0801, + "step": 27469 + }, + { + "epoch": 0.23845279120840965, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016766803383233128, + "loss": 0.1094, + "step": 27470 + }, + { + "epoch": 0.2384614716886138, + "grad_norm": 0.74609375, + "learning_rate": 0.0016766574880412955, + "loss": 0.0801, + "step": 27471 + }, + { + "epoch": 0.23847015216881798, + "grad_norm": 0.10693359375, + "learning_rate": 0.001676634637128654, + "loss": 0.0938, + "step": 27472 + }, + { + "epoch": 0.23847883264902214, + "grad_norm": 0.578125, + "learning_rate": 0.0016766117855854142, + "loss": 0.0889, + "step": 27473 + }, + { + "epoch": 0.2384875131292263, + "grad_norm": 0.224609375, + "learning_rate": 0.0016765889334116, + "loss": 0.1035, + "step": 27474 + }, + { + "epoch": 0.23849619360943047, + "grad_norm": 0.1572265625, + "learning_rate": 0.001676566080607237, + "loss": 0.1328, + "step": 27475 + }, + { + "epoch": 0.23850487408963464, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016765432271723497, + "loss": 0.1055, + "step": 27476 + }, + { + "epoch": 0.2385135545698388, + "grad_norm": 0.40234375, + "learning_rate": 0.0016765203731069636, + "loss": 0.1221, + "step": 27477 + }, + { + "epoch": 0.23852223505004297, + "grad_norm": 0.2412109375, + "learning_rate": 0.0016764975184111036, + "loss": 0.1133, + "step": 27478 + }, + { + "epoch": 0.23853091553024713, + "grad_norm": 0.296875, + "learning_rate": 0.0016764746630847942, + "loss": 0.0889, + "step": 27479 + }, + { + "epoch": 0.2385395960104513, + "grad_norm": 0.26953125, + "learning_rate": 0.001676451807128061, + "loss": 0.1094, + "step": 27480 + }, + { + "epoch": 0.23854827649065546, + "grad_norm": 0.169921875, + "learning_rate": 0.001676428950540929, + "loss": 0.0967, + "step": 27481 + }, + { + "epoch": 0.23855695697085963, + "grad_norm": 0.318359375, + "learning_rate": 0.0016764060933234226, + "loss": 0.0996, + "step": 27482 + }, + { + "epoch": 0.2385656374510638, + "grad_norm": 0.412109375, + "learning_rate": 0.0016763832354755673, + "loss": 0.1172, + "step": 27483 + }, + { + "epoch": 0.23857431793126796, + "grad_norm": 0.55078125, + "learning_rate": 0.001676360376997388, + "loss": 0.1191, + "step": 27484 + }, + { + "epoch": 0.23858299841147212, + "grad_norm": 0.76953125, + "learning_rate": 0.0016763375178889095, + "loss": 0.1963, + "step": 27485 + }, + { + "epoch": 0.2385916788916763, + "grad_norm": 0.48046875, + "learning_rate": 0.001676314658150157, + "loss": 0.1211, + "step": 27486 + }, + { + "epoch": 0.23860035937188045, + "grad_norm": 0.251953125, + "learning_rate": 0.0016762917977811557, + "loss": 0.1143, + "step": 27487 + }, + { + "epoch": 0.23860903985208462, + "grad_norm": 0.33984375, + "learning_rate": 0.0016762689367819299, + "loss": 0.124, + "step": 27488 + }, + { + "epoch": 0.23861772033228879, + "grad_norm": 0.443359375, + "learning_rate": 0.0016762460751525053, + "loss": 0.1914, + "step": 27489 + }, + { + "epoch": 0.23862640081249295, + "grad_norm": 0.384765625, + "learning_rate": 0.0016762232128929066, + "loss": 0.1123, + "step": 27490 + }, + { + "epoch": 0.23863508129269712, + "grad_norm": 0.185546875, + "learning_rate": 0.001676200350003159, + "loss": 0.1182, + "step": 27491 + }, + { + "epoch": 0.23864376177290128, + "grad_norm": 0.1171875, + "learning_rate": 0.001676177486483287, + "loss": 0.1113, + "step": 27492 + }, + { + "epoch": 0.23865244225310545, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016761546223333162, + "loss": 0.0977, + "step": 27493 + }, + { + "epoch": 0.2386611227333096, + "grad_norm": 0.14453125, + "learning_rate": 0.0016761317575532716, + "loss": 0.1104, + "step": 27494 + }, + { + "epoch": 0.23866980321351378, + "grad_norm": 0.14453125, + "learning_rate": 0.0016761088921431774, + "loss": 0.1064, + "step": 27495 + }, + { + "epoch": 0.23867848369371794, + "grad_norm": 0.39453125, + "learning_rate": 0.0016760860261030596, + "loss": 0.1367, + "step": 27496 + }, + { + "epoch": 0.2386871641739221, + "grad_norm": 0.54296875, + "learning_rate": 0.0016760631594329427, + "loss": 0.1084, + "step": 27497 + }, + { + "epoch": 0.23869584465412627, + "grad_norm": 0.140625, + "learning_rate": 0.0016760402921328517, + "loss": 0.1328, + "step": 27498 + }, + { + "epoch": 0.23870452513433044, + "grad_norm": 0.365234375, + "learning_rate": 0.001676017424202812, + "loss": 0.1143, + "step": 27499 + }, + { + "epoch": 0.2387132056145346, + "grad_norm": 0.14453125, + "learning_rate": 0.0016759945556428478, + "loss": 0.0874, + "step": 27500 + }, + { + "epoch": 0.23872188609473877, + "grad_norm": 0.09326171875, + "learning_rate": 0.001675971686452985, + "loss": 0.0957, + "step": 27501 + }, + { + "epoch": 0.23873056657494293, + "grad_norm": 0.14453125, + "learning_rate": 0.001675948816633248, + "loss": 0.1455, + "step": 27502 + }, + { + "epoch": 0.2387392470551471, + "grad_norm": 0.2490234375, + "learning_rate": 0.001675925946183662, + "loss": 0.1138, + "step": 27503 + }, + { + "epoch": 0.23874792753535126, + "grad_norm": 0.125, + "learning_rate": 0.0016759030751042523, + "loss": 0.1113, + "step": 27504 + }, + { + "epoch": 0.23875660801555543, + "grad_norm": 0.6953125, + "learning_rate": 0.0016758802033950432, + "loss": 0.1484, + "step": 27505 + }, + { + "epoch": 0.2387652884957596, + "grad_norm": 0.1064453125, + "learning_rate": 0.0016758573310560606, + "loss": 0.0933, + "step": 27506 + }, + { + "epoch": 0.23877396897596376, + "grad_norm": 0.10400390625, + "learning_rate": 0.0016758344580873289, + "loss": 0.104, + "step": 27507 + }, + { + "epoch": 0.23878264945616792, + "grad_norm": 0.1181640625, + "learning_rate": 0.0016758115844888735, + "loss": 0.1055, + "step": 27508 + }, + { + "epoch": 0.2387913299363721, + "grad_norm": 0.279296875, + "learning_rate": 0.001675788710260719, + "loss": 0.0806, + "step": 27509 + }, + { + "epoch": 0.23880001041657625, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016757658354028906, + "loss": 0.1436, + "step": 27510 + }, + { + "epoch": 0.23880869089678042, + "grad_norm": 0.10888671875, + "learning_rate": 0.0016757429599154132, + "loss": 0.1289, + "step": 27511 + }, + { + "epoch": 0.23881737137698458, + "grad_norm": 0.10009765625, + "learning_rate": 0.0016757200837983122, + "loss": 0.1387, + "step": 27512 + }, + { + "epoch": 0.23882605185718875, + "grad_norm": 0.31640625, + "learning_rate": 0.0016756972070516123, + "loss": 0.1719, + "step": 27513 + }, + { + "epoch": 0.2388347323373929, + "grad_norm": 0.63671875, + "learning_rate": 0.0016756743296753387, + "loss": 0.0859, + "step": 27514 + }, + { + "epoch": 0.23884341281759708, + "grad_norm": 3.28125, + "learning_rate": 0.001675651451669516, + "loss": 0.2275, + "step": 27515 + }, + { + "epoch": 0.23885209329780124, + "grad_norm": 0.1328125, + "learning_rate": 0.0016756285730341698, + "loss": 0.1172, + "step": 27516 + }, + { + "epoch": 0.2388607737780054, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016756056937693247, + "loss": 0.123, + "step": 27517 + }, + { + "epoch": 0.23886945425820957, + "grad_norm": 0.5546875, + "learning_rate": 0.0016755828138750057, + "loss": 0.167, + "step": 27518 + }, + { + "epoch": 0.23887813473841374, + "grad_norm": 0.3046875, + "learning_rate": 0.0016755599333512384, + "loss": 0.0825, + "step": 27519 + }, + { + "epoch": 0.2388868152186179, + "grad_norm": 0.09033203125, + "learning_rate": 0.001675537052198047, + "loss": 0.1338, + "step": 27520 + }, + { + "epoch": 0.23889549569882207, + "grad_norm": 0.271484375, + "learning_rate": 0.001675514170415457, + "loss": 0.1416, + "step": 27521 + }, + { + "epoch": 0.23890417617902623, + "grad_norm": 0.40625, + "learning_rate": 0.0016754912880034935, + "loss": 0.1465, + "step": 27522 + }, + { + "epoch": 0.2389128566592304, + "grad_norm": 0.74609375, + "learning_rate": 0.0016754684049621815, + "loss": 0.1348, + "step": 27523 + }, + { + "epoch": 0.23892153713943456, + "grad_norm": 0.49609375, + "learning_rate": 0.0016754455212915455, + "loss": 0.1182, + "step": 27524 + }, + { + "epoch": 0.23893021761963873, + "grad_norm": 0.56640625, + "learning_rate": 0.0016754226369916116, + "loss": 0.0923, + "step": 27525 + }, + { + "epoch": 0.2389388980998429, + "grad_norm": 0.19921875, + "learning_rate": 0.0016753997520624034, + "loss": 0.0967, + "step": 27526 + }, + { + "epoch": 0.23894757858004706, + "grad_norm": 0.1396484375, + "learning_rate": 0.001675376866503947, + "loss": 0.127, + "step": 27527 + }, + { + "epoch": 0.23895625906025122, + "grad_norm": 0.12890625, + "learning_rate": 0.001675353980316267, + "loss": 0.1211, + "step": 27528 + }, + { + "epoch": 0.2389649395404554, + "grad_norm": 0.0771484375, + "learning_rate": 0.0016753310934993887, + "loss": 0.0986, + "step": 27529 + }, + { + "epoch": 0.23897362002065956, + "grad_norm": 0.189453125, + "learning_rate": 0.0016753082060533369, + "loss": 0.0801, + "step": 27530 + }, + { + "epoch": 0.23898230050086372, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016752853179781368, + "loss": 0.1152, + "step": 27531 + }, + { + "epoch": 0.23899098098106789, + "grad_norm": 0.294921875, + "learning_rate": 0.001675262429273813, + "loss": 0.0889, + "step": 27532 + }, + { + "epoch": 0.23899966146127205, + "grad_norm": 0.1767578125, + "learning_rate": 0.0016752395399403913, + "loss": 0.1582, + "step": 27533 + }, + { + "epoch": 0.23900834194147622, + "grad_norm": 0.271484375, + "learning_rate": 0.0016752166499778957, + "loss": 0.0859, + "step": 27534 + }, + { + "epoch": 0.23901702242168038, + "grad_norm": 0.44140625, + "learning_rate": 0.0016751937593863521, + "loss": 0.1133, + "step": 27535 + }, + { + "epoch": 0.23902570290188455, + "grad_norm": 0.314453125, + "learning_rate": 0.0016751708681657855, + "loss": 0.0938, + "step": 27536 + }, + { + "epoch": 0.2390343833820887, + "grad_norm": 0.220703125, + "learning_rate": 0.0016751479763162205, + "loss": 0.0996, + "step": 27537 + }, + { + "epoch": 0.23904306386229285, + "grad_norm": 0.162109375, + "learning_rate": 0.001675125083837682, + "loss": 0.0918, + "step": 27538 + }, + { + "epoch": 0.239051744342497, + "grad_norm": 0.61328125, + "learning_rate": 0.0016751021907301959, + "loss": 0.0996, + "step": 27539 + }, + { + "epoch": 0.23906042482270118, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016750792969937864, + "loss": 0.1113, + "step": 27540 + }, + { + "epoch": 0.23906910530290534, + "grad_norm": 0.61328125, + "learning_rate": 0.0016750564026284786, + "loss": 0.125, + "step": 27541 + }, + { + "epoch": 0.2390777857831095, + "grad_norm": 0.212890625, + "learning_rate": 0.001675033507634298, + "loss": 0.0967, + "step": 27542 + }, + { + "epoch": 0.23908646626331367, + "grad_norm": 0.236328125, + "learning_rate": 0.0016750106120112694, + "loss": 0.0854, + "step": 27543 + }, + { + "epoch": 0.23909514674351784, + "grad_norm": 0.154296875, + "learning_rate": 0.001674987715759418, + "loss": 0.0918, + "step": 27544 + }, + { + "epoch": 0.239103827223722, + "grad_norm": 0.42578125, + "learning_rate": 0.0016749648188787683, + "loss": 0.1191, + "step": 27545 + }, + { + "epoch": 0.23911250770392617, + "grad_norm": 0.2216796875, + "learning_rate": 0.001674941921369346, + "loss": 0.1387, + "step": 27546 + }, + { + "epoch": 0.23912118818413033, + "grad_norm": 0.99609375, + "learning_rate": 0.0016749190232311756, + "loss": 0.1523, + "step": 27547 + }, + { + "epoch": 0.2391298686643345, + "grad_norm": 0.294921875, + "learning_rate": 0.0016748961244642826, + "loss": 0.1172, + "step": 27548 + }, + { + "epoch": 0.23913854914453866, + "grad_norm": 0.1240234375, + "learning_rate": 0.001674873225068692, + "loss": 0.124, + "step": 27549 + }, + { + "epoch": 0.23914722962474283, + "grad_norm": 0.255859375, + "learning_rate": 0.0016748503250444285, + "loss": 0.1191, + "step": 27550 + }, + { + "epoch": 0.239155910104947, + "grad_norm": 0.400390625, + "learning_rate": 0.0016748274243915172, + "loss": 0.1221, + "step": 27551 + }, + { + "epoch": 0.23916459058515116, + "grad_norm": 0.16015625, + "learning_rate": 0.0016748045231099833, + "loss": 0.123, + "step": 27552 + }, + { + "epoch": 0.23917327106535533, + "grad_norm": 0.10400390625, + "learning_rate": 0.001674781621199852, + "loss": 0.1123, + "step": 27553 + }, + { + "epoch": 0.2391819515455595, + "grad_norm": 1.40625, + "learning_rate": 0.0016747587186611477, + "loss": 0.0889, + "step": 27554 + }, + { + "epoch": 0.23919063202576366, + "grad_norm": 0.123046875, + "learning_rate": 0.0016747358154938963, + "loss": 0.0918, + "step": 27555 + }, + { + "epoch": 0.23919931250596782, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016747129116981227, + "loss": 0.0776, + "step": 27556 + }, + { + "epoch": 0.23920799298617199, + "grad_norm": 0.76171875, + "learning_rate": 0.0016746900072738512, + "loss": 0.0767, + "step": 27557 + }, + { + "epoch": 0.23921667346637615, + "grad_norm": 0.8515625, + "learning_rate": 0.0016746671022211074, + "loss": 0.1807, + "step": 27558 + }, + { + "epoch": 0.23922535394658032, + "grad_norm": 0.103515625, + "learning_rate": 0.0016746441965399164, + "loss": 0.0977, + "step": 27559 + }, + { + "epoch": 0.23923403442678448, + "grad_norm": 0.30859375, + "learning_rate": 0.0016746212902303034, + "loss": 0.1113, + "step": 27560 + }, + { + "epoch": 0.23924271490698865, + "grad_norm": 0.076171875, + "learning_rate": 0.0016745983832922927, + "loss": 0.0898, + "step": 27561 + }, + { + "epoch": 0.2392513953871928, + "grad_norm": 1.03125, + "learning_rate": 0.00167457547572591, + "loss": 0.1816, + "step": 27562 + }, + { + "epoch": 0.23926007586739698, + "grad_norm": 0.26171875, + "learning_rate": 0.0016745525675311804, + "loss": 0.1348, + "step": 27563 + }, + { + "epoch": 0.23926875634760114, + "grad_norm": 0.2734375, + "learning_rate": 0.0016745296587081285, + "loss": 0.1113, + "step": 27564 + }, + { + "epoch": 0.2392774368278053, + "grad_norm": 0.283203125, + "learning_rate": 0.0016745067492567798, + "loss": 0.1426, + "step": 27565 + }, + { + "epoch": 0.23928611730800947, + "grad_norm": 0.197265625, + "learning_rate": 0.0016744838391771588, + "loss": 0.1562, + "step": 27566 + }, + { + "epoch": 0.23929479778821364, + "grad_norm": 1.390625, + "learning_rate": 0.0016744609284692913, + "loss": 0.1504, + "step": 27567 + }, + { + "epoch": 0.2393034782684178, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016744380171332019, + "loss": 0.1172, + "step": 27568 + }, + { + "epoch": 0.23931215874862197, + "grad_norm": 0.181640625, + "learning_rate": 0.0016744151051689157, + "loss": 0.0898, + "step": 27569 + }, + { + "epoch": 0.23932083922882613, + "grad_norm": 0.3125, + "learning_rate": 0.0016743921925764576, + "loss": 0.1338, + "step": 27570 + }, + { + "epoch": 0.2393295197090303, + "grad_norm": 0.0751953125, + "learning_rate": 0.0016743692793558529, + "loss": 0.1055, + "step": 27571 + }, + { + "epoch": 0.23933820018923446, + "grad_norm": 0.16015625, + "learning_rate": 0.0016743463655071267, + "loss": 0.0811, + "step": 27572 + }, + { + "epoch": 0.23934688066943863, + "grad_norm": 1.234375, + "learning_rate": 0.0016743234510303037, + "loss": 0.0986, + "step": 27573 + }, + { + "epoch": 0.2393555611496428, + "grad_norm": 0.57421875, + "learning_rate": 0.0016743005359254096, + "loss": 0.1045, + "step": 27574 + }, + { + "epoch": 0.23936424162984696, + "grad_norm": 0.10888671875, + "learning_rate": 0.0016742776201924686, + "loss": 0.0762, + "step": 27575 + }, + { + "epoch": 0.23937292211005112, + "grad_norm": 0.39453125, + "learning_rate": 0.0016742547038315066, + "loss": 0.0977, + "step": 27576 + }, + { + "epoch": 0.2393816025902553, + "grad_norm": 0.369140625, + "learning_rate": 0.0016742317868425481, + "loss": 0.1221, + "step": 27577 + }, + { + "epoch": 0.23939028307045945, + "grad_norm": 0.291015625, + "learning_rate": 0.0016742088692256183, + "loss": 0.0806, + "step": 27578 + }, + { + "epoch": 0.23939896355066362, + "grad_norm": 0.31640625, + "learning_rate": 0.0016741859509807424, + "loss": 0.1035, + "step": 27579 + }, + { + "epoch": 0.23940764403086778, + "grad_norm": 0.5234375, + "learning_rate": 0.0016741630321079452, + "loss": 0.1201, + "step": 27580 + }, + { + "epoch": 0.23941632451107195, + "grad_norm": 0.484375, + "learning_rate": 0.0016741401126072523, + "loss": 0.0938, + "step": 27581 + }, + { + "epoch": 0.2394250049912761, + "grad_norm": 0.16796875, + "learning_rate": 0.0016741171924786883, + "loss": 0.1064, + "step": 27582 + }, + { + "epoch": 0.23943368547148028, + "grad_norm": 0.294921875, + "learning_rate": 0.0016740942717222782, + "loss": 0.0728, + "step": 27583 + }, + { + "epoch": 0.23944236595168444, + "grad_norm": 0.57421875, + "learning_rate": 0.0016740713503380473, + "loss": 0.1069, + "step": 27584 + }, + { + "epoch": 0.2394510464318886, + "grad_norm": 0.64453125, + "learning_rate": 0.0016740484283260207, + "loss": 0.0986, + "step": 27585 + }, + { + "epoch": 0.23945972691209277, + "grad_norm": 0.349609375, + "learning_rate": 0.001674025505686223, + "loss": 0.0957, + "step": 27586 + }, + { + "epoch": 0.23946840739229694, + "grad_norm": 0.65234375, + "learning_rate": 0.00167400258241868, + "loss": 0.1108, + "step": 27587 + }, + { + "epoch": 0.2394770878725011, + "grad_norm": 0.244140625, + "learning_rate": 0.0016739796585234165, + "loss": 0.0703, + "step": 27588 + }, + { + "epoch": 0.23948576835270527, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016739567340004569, + "loss": 0.1074, + "step": 27589 + }, + { + "epoch": 0.23949444883290943, + "grad_norm": 0.0771484375, + "learning_rate": 0.0016739338088498272, + "loss": 0.0737, + "step": 27590 + }, + { + "epoch": 0.2395031293131136, + "grad_norm": 0.2353515625, + "learning_rate": 0.0016739108830715523, + "loss": 0.0981, + "step": 27591 + }, + { + "epoch": 0.23951180979331776, + "grad_norm": 0.30078125, + "learning_rate": 0.001673887956665657, + "loss": 0.1543, + "step": 27592 + }, + { + "epoch": 0.23952049027352193, + "grad_norm": 0.125, + "learning_rate": 0.001673865029632166, + "loss": 0.0874, + "step": 27593 + }, + { + "epoch": 0.2395291707537261, + "grad_norm": 0.318359375, + "learning_rate": 0.0016738421019711052, + "loss": 0.1021, + "step": 27594 + }, + { + "epoch": 0.23953785123393026, + "grad_norm": 0.302734375, + "learning_rate": 0.0016738191736824994, + "loss": 0.1377, + "step": 27595 + }, + { + "epoch": 0.23954653171413443, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016737962447663736, + "loss": 0.1133, + "step": 27596 + }, + { + "epoch": 0.2395552121943386, + "grad_norm": 0.3359375, + "learning_rate": 0.0016737733152227523, + "loss": 0.084, + "step": 27597 + }, + { + "epoch": 0.23956389267454276, + "grad_norm": 0.208984375, + "learning_rate": 0.0016737503850516615, + "loss": 0.106, + "step": 27598 + }, + { + "epoch": 0.23957257315474692, + "grad_norm": 0.349609375, + "learning_rate": 0.0016737274542531263, + "loss": 0.1035, + "step": 27599 + }, + { + "epoch": 0.23958125363495109, + "grad_norm": 0.291015625, + "learning_rate": 0.0016737045228271705, + "loss": 0.1357, + "step": 27600 + }, + { + "epoch": 0.23958993411515525, + "grad_norm": 0.458984375, + "learning_rate": 0.0016736815907738207, + "loss": 0.0767, + "step": 27601 + }, + { + "epoch": 0.23959861459535942, + "grad_norm": 0.68359375, + "learning_rate": 0.0016736586580931012, + "loss": 0.0894, + "step": 27602 + }, + { + "epoch": 0.23960729507556358, + "grad_norm": 0.400390625, + "learning_rate": 0.001673635724785037, + "loss": 0.0967, + "step": 27603 + }, + { + "epoch": 0.23961597555576775, + "grad_norm": 0.326171875, + "learning_rate": 0.0016736127908496539, + "loss": 0.1143, + "step": 27604 + }, + { + "epoch": 0.2396246560359719, + "grad_norm": 0.21484375, + "learning_rate": 0.001673589856286976, + "loss": 0.1172, + "step": 27605 + }, + { + "epoch": 0.23963333651617608, + "grad_norm": 0.27734375, + "learning_rate": 0.0016735669210970288, + "loss": 0.0986, + "step": 27606 + }, + { + "epoch": 0.23964201699638024, + "grad_norm": 0.197265625, + "learning_rate": 0.0016735439852798377, + "loss": 0.1172, + "step": 27607 + }, + { + "epoch": 0.2396506974765844, + "grad_norm": 0.48046875, + "learning_rate": 0.0016735210488354275, + "loss": 0.1445, + "step": 27608 + }, + { + "epoch": 0.23965937795678857, + "grad_norm": 0.5, + "learning_rate": 0.0016734981117638233, + "loss": 0.1074, + "step": 27609 + }, + { + "epoch": 0.23966805843699274, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016734751740650501, + "loss": 0.0928, + "step": 27610 + }, + { + "epoch": 0.2396767389171969, + "grad_norm": 1.203125, + "learning_rate": 0.001673452235739133, + "loss": 0.1602, + "step": 27611 + }, + { + "epoch": 0.23968541939740107, + "grad_norm": 0.3046875, + "learning_rate": 0.0016734292967860974, + "loss": 0.126, + "step": 27612 + }, + { + "epoch": 0.23969409987760523, + "grad_norm": 0.13671875, + "learning_rate": 0.0016734063572059678, + "loss": 0.1143, + "step": 27613 + }, + { + "epoch": 0.2397027803578094, + "grad_norm": 0.54296875, + "learning_rate": 0.0016733834169987698, + "loss": 0.1157, + "step": 27614 + }, + { + "epoch": 0.23971146083801356, + "grad_norm": 0.1328125, + "learning_rate": 0.0016733604761645282, + "loss": 0.1123, + "step": 27615 + }, + { + "epoch": 0.23972014131821773, + "grad_norm": 0.61328125, + "learning_rate": 0.0016733375347032684, + "loss": 0.0942, + "step": 27616 + }, + { + "epoch": 0.2397288217984219, + "grad_norm": 0.2412109375, + "learning_rate": 0.0016733145926150152, + "loss": 0.0781, + "step": 27617 + }, + { + "epoch": 0.23973750227862606, + "grad_norm": 0.42578125, + "learning_rate": 0.0016732916498997938, + "loss": 0.1113, + "step": 27618 + }, + { + "epoch": 0.23974618275883022, + "grad_norm": 0.443359375, + "learning_rate": 0.0016732687065576296, + "loss": 0.1143, + "step": 27619 + }, + { + "epoch": 0.2397548632390344, + "grad_norm": 0.32421875, + "learning_rate": 0.0016732457625885469, + "loss": 0.083, + "step": 27620 + }, + { + "epoch": 0.23976354371923855, + "grad_norm": 0.1064453125, + "learning_rate": 0.0016732228179925713, + "loss": 0.1118, + "step": 27621 + }, + { + "epoch": 0.23977222419944272, + "grad_norm": 0.267578125, + "learning_rate": 0.001673199872769728, + "loss": 0.1226, + "step": 27622 + }, + { + "epoch": 0.23978090467964688, + "grad_norm": 0.08154296875, + "learning_rate": 0.0016731769269200416, + "loss": 0.0942, + "step": 27623 + }, + { + "epoch": 0.23978958515985105, + "grad_norm": 0.1474609375, + "learning_rate": 0.0016731539804435377, + "loss": 0.1289, + "step": 27624 + }, + { + "epoch": 0.2397982656400552, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016731310333402412, + "loss": 0.125, + "step": 27625 + }, + { + "epoch": 0.23980694612025938, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016731080856101773, + "loss": 0.0967, + "step": 27626 + }, + { + "epoch": 0.23981562660046354, + "grad_norm": 0.314453125, + "learning_rate": 0.001673085137253371, + "loss": 0.1123, + "step": 27627 + }, + { + "epoch": 0.2398243070806677, + "grad_norm": 0.83984375, + "learning_rate": 0.0016730621882698473, + "loss": 0.1001, + "step": 27628 + }, + { + "epoch": 0.23983298756087187, + "grad_norm": 0.703125, + "learning_rate": 0.0016730392386596314, + "loss": 0.1572, + "step": 27629 + }, + { + "epoch": 0.23984166804107604, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016730162884227486, + "loss": 0.1777, + "step": 27630 + }, + { + "epoch": 0.2398503485212802, + "grad_norm": 0.181640625, + "learning_rate": 0.0016729933375592239, + "loss": 0.1113, + "step": 27631 + }, + { + "epoch": 0.23985902900148437, + "grad_norm": 0.39453125, + "learning_rate": 0.0016729703860690816, + "loss": 0.084, + "step": 27632 + }, + { + "epoch": 0.23986770948168853, + "grad_norm": 0.11376953125, + "learning_rate": 0.0016729474339523484, + "loss": 0.1055, + "step": 27633 + }, + { + "epoch": 0.2398763899618927, + "grad_norm": 0.1044921875, + "learning_rate": 0.0016729244812090478, + "loss": 0.0942, + "step": 27634 + }, + { + "epoch": 0.23988507044209686, + "grad_norm": 0.6796875, + "learning_rate": 0.0016729015278392056, + "loss": 0.1523, + "step": 27635 + }, + { + "epoch": 0.23989375092230103, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016728785738428474, + "loss": 0.1016, + "step": 27636 + }, + { + "epoch": 0.2399024314025052, + "grad_norm": 0.353515625, + "learning_rate": 0.0016728556192199975, + "loss": 0.1074, + "step": 27637 + }, + { + "epoch": 0.23991111188270936, + "grad_norm": 0.3984375, + "learning_rate": 0.0016728326639706814, + "loss": 0.1016, + "step": 27638 + }, + { + "epoch": 0.23991979236291353, + "grad_norm": 0.412109375, + "learning_rate": 0.0016728097080949238, + "loss": 0.1328, + "step": 27639 + }, + { + "epoch": 0.2399284728431177, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016727867515927504, + "loss": 0.0767, + "step": 27640 + }, + { + "epoch": 0.23993715332332186, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016727637944641863, + "loss": 0.1211, + "step": 27641 + }, + { + "epoch": 0.23994583380352602, + "grad_norm": 0.443359375, + "learning_rate": 0.0016727408367092557, + "loss": 0.0967, + "step": 27642 + }, + { + "epoch": 0.23995451428373019, + "grad_norm": 0.419921875, + "learning_rate": 0.0016727178783279848, + "loss": 0.1484, + "step": 27643 + }, + { + "epoch": 0.23996319476393435, + "grad_norm": 0.62890625, + "learning_rate": 0.001672694919320398, + "loss": 0.1328, + "step": 27644 + }, + { + "epoch": 0.23997187524413852, + "grad_norm": 0.08251953125, + "learning_rate": 0.0016726719596865205, + "loss": 0.0811, + "step": 27645 + }, + { + "epoch": 0.23998055572434268, + "grad_norm": 0.36328125, + "learning_rate": 0.001672648999426378, + "loss": 0.1318, + "step": 27646 + }, + { + "epoch": 0.23998923620454685, + "grad_norm": 0.6015625, + "learning_rate": 0.0016726260385399947, + "loss": 0.2754, + "step": 27647 + }, + { + "epoch": 0.239997916684751, + "grad_norm": 0.36328125, + "learning_rate": 0.0016726030770273966, + "loss": 0.1152, + "step": 27648 + }, + { + "epoch": 0.24000659716495518, + "grad_norm": 0.287109375, + "learning_rate": 0.0016725801148886079, + "loss": 0.0869, + "step": 27649 + }, + { + "epoch": 0.24001527764515934, + "grad_norm": 0.5390625, + "learning_rate": 0.0016725571521236545, + "loss": 0.1299, + "step": 27650 + }, + { + "epoch": 0.2400239581253635, + "grad_norm": 0.58984375, + "learning_rate": 0.0016725341887325613, + "loss": 0.1055, + "step": 27651 + }, + { + "epoch": 0.24003263860556767, + "grad_norm": 0.1640625, + "learning_rate": 0.001672511224715353, + "loss": 0.1113, + "step": 27652 + }, + { + "epoch": 0.24004131908577184, + "grad_norm": 0.384765625, + "learning_rate": 0.0016724882600720554, + "loss": 0.1191, + "step": 27653 + }, + { + "epoch": 0.240049999565976, + "grad_norm": 0.1708984375, + "learning_rate": 0.0016724652948026928, + "loss": 0.1006, + "step": 27654 + }, + { + "epoch": 0.24005868004618017, + "grad_norm": 0.1513671875, + "learning_rate": 0.001672442328907291, + "loss": 0.0859, + "step": 27655 + }, + { + "epoch": 0.24006736052638433, + "grad_norm": 0.16015625, + "learning_rate": 0.001672419362385875, + "loss": 0.1025, + "step": 27656 + }, + { + "epoch": 0.2400760410065885, + "grad_norm": 0.251953125, + "learning_rate": 0.0016723963952384698, + "loss": 0.1055, + "step": 27657 + }, + { + "epoch": 0.24008472148679266, + "grad_norm": 0.515625, + "learning_rate": 0.0016723734274651002, + "loss": 0.0972, + "step": 27658 + }, + { + "epoch": 0.24009340196699683, + "grad_norm": 0.267578125, + "learning_rate": 0.0016723504590657916, + "loss": 0.1201, + "step": 27659 + }, + { + "epoch": 0.240102082447201, + "grad_norm": 0.65625, + "learning_rate": 0.0016723274900405695, + "loss": 0.1357, + "step": 27660 + }, + { + "epoch": 0.24011076292740513, + "grad_norm": 0.349609375, + "learning_rate": 0.0016723045203894582, + "loss": 0.0859, + "step": 27661 + }, + { + "epoch": 0.2401194434076093, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016722815501124836, + "loss": 0.125, + "step": 27662 + }, + { + "epoch": 0.24012812388781346, + "grad_norm": 0.17578125, + "learning_rate": 0.0016722585792096703, + "loss": 0.0879, + "step": 27663 + }, + { + "epoch": 0.24013680436801763, + "grad_norm": 1.3984375, + "learning_rate": 0.0016722356076810437, + "loss": 0.1543, + "step": 27664 + }, + { + "epoch": 0.2401454848482218, + "grad_norm": 0.19921875, + "learning_rate": 0.001672212635526629, + "loss": 0.1128, + "step": 27665 + }, + { + "epoch": 0.24015416532842596, + "grad_norm": 0.119140625, + "learning_rate": 0.0016721896627464507, + "loss": 0.1133, + "step": 27666 + }, + { + "epoch": 0.24016284580863012, + "grad_norm": 0.388671875, + "learning_rate": 0.0016721666893405348, + "loss": 0.1309, + "step": 27667 + }, + { + "epoch": 0.24017152628883429, + "grad_norm": 0.14453125, + "learning_rate": 0.0016721437153089061, + "loss": 0.1543, + "step": 27668 + }, + { + "epoch": 0.24018020676903845, + "grad_norm": 0.275390625, + "learning_rate": 0.0016721207406515896, + "loss": 0.1045, + "step": 27669 + }, + { + "epoch": 0.24018888724924262, + "grad_norm": 0.1796875, + "learning_rate": 0.0016720977653686103, + "loss": 0.106, + "step": 27670 + }, + { + "epoch": 0.24019756772944678, + "grad_norm": 0.10546875, + "learning_rate": 0.0016720747894599934, + "loss": 0.1084, + "step": 27671 + }, + { + "epoch": 0.24020624820965095, + "grad_norm": 0.482421875, + "learning_rate": 0.0016720518129257642, + "loss": 0.1436, + "step": 27672 + }, + { + "epoch": 0.2402149286898551, + "grad_norm": 0.74609375, + "learning_rate": 0.0016720288357659477, + "loss": 0.0869, + "step": 27673 + }, + { + "epoch": 0.24022360917005928, + "grad_norm": 0.2734375, + "learning_rate": 0.001672005857980569, + "loss": 0.0938, + "step": 27674 + }, + { + "epoch": 0.24023228965026344, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016719828795696532, + "loss": 0.0811, + "step": 27675 + }, + { + "epoch": 0.2402409701304676, + "grad_norm": 0.07421875, + "learning_rate": 0.001671959900533226, + "loss": 0.0869, + "step": 27676 + }, + { + "epoch": 0.24024965061067177, + "grad_norm": 0.091796875, + "learning_rate": 0.0016719369208713117, + "loss": 0.0977, + "step": 27677 + }, + { + "epoch": 0.24025833109087594, + "grad_norm": 0.1044921875, + "learning_rate": 0.0016719139405839357, + "loss": 0.1226, + "step": 27678 + }, + { + "epoch": 0.2402670115710801, + "grad_norm": 0.091796875, + "learning_rate": 0.0016718909596711233, + "loss": 0.0908, + "step": 27679 + }, + { + "epoch": 0.24027569205128427, + "grad_norm": 0.5234375, + "learning_rate": 0.0016718679781328998, + "loss": 0.209, + "step": 27680 + }, + { + "epoch": 0.24028437253148843, + "grad_norm": 0.56640625, + "learning_rate": 0.0016718449959692897, + "loss": 0.1074, + "step": 27681 + }, + { + "epoch": 0.2402930530116926, + "grad_norm": 0.365234375, + "learning_rate": 0.0016718220131803185, + "loss": 0.0864, + "step": 27682 + }, + { + "epoch": 0.24030173349189676, + "grad_norm": 0.404296875, + "learning_rate": 0.0016717990297660118, + "loss": 0.0796, + "step": 27683 + }, + { + "epoch": 0.24031041397210093, + "grad_norm": 0.306640625, + "learning_rate": 0.0016717760457263936, + "loss": 0.1025, + "step": 27684 + }, + { + "epoch": 0.2403190944523051, + "grad_norm": 0.5546875, + "learning_rate": 0.00167175306106149, + "loss": 0.1196, + "step": 27685 + }, + { + "epoch": 0.24032777493250926, + "grad_norm": 0.2734375, + "learning_rate": 0.001671730075771326, + "loss": 0.1201, + "step": 27686 + }, + { + "epoch": 0.24033645541271342, + "grad_norm": 0.09423828125, + "learning_rate": 0.0016717070898559269, + "loss": 0.1553, + "step": 27687 + }, + { + "epoch": 0.2403451358929176, + "grad_norm": 0.1328125, + "learning_rate": 0.0016716841033153165, + "loss": 0.0928, + "step": 27688 + }, + { + "epoch": 0.24035381637312175, + "grad_norm": 0.515625, + "learning_rate": 0.0016716611161495217, + "loss": 0.0957, + "step": 27689 + }, + { + "epoch": 0.24036249685332592, + "grad_norm": 0.578125, + "learning_rate": 0.0016716381283585669, + "loss": 0.1089, + "step": 27690 + }, + { + "epoch": 0.24037117733353008, + "grad_norm": 0.1845703125, + "learning_rate": 0.001671615139942477, + "loss": 0.1201, + "step": 27691 + }, + { + "epoch": 0.24037985781373425, + "grad_norm": 1.5234375, + "learning_rate": 0.0016715921509012773, + "loss": 0.1289, + "step": 27692 + }, + { + "epoch": 0.24038853829393841, + "grad_norm": 0.6015625, + "learning_rate": 0.0016715691612349932, + "loss": 0.1348, + "step": 27693 + }, + { + "epoch": 0.24039721877414258, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016715461709436495, + "loss": 0.0962, + "step": 27694 + }, + { + "epoch": 0.24040589925434674, + "grad_norm": 0.1015625, + "learning_rate": 0.0016715231800272718, + "loss": 0.1465, + "step": 27695 + }, + { + "epoch": 0.2404145797345509, + "grad_norm": 0.1591796875, + "learning_rate": 0.0016715001884858847, + "loss": 0.1201, + "step": 27696 + }, + { + "epoch": 0.24042326021475507, + "grad_norm": 0.328125, + "learning_rate": 0.0016714771963195134, + "loss": 0.1084, + "step": 27697 + }, + { + "epoch": 0.24043194069495924, + "grad_norm": 0.08251953125, + "learning_rate": 0.0016714542035281834, + "loss": 0.166, + "step": 27698 + }, + { + "epoch": 0.2404406211751634, + "grad_norm": 0.119140625, + "learning_rate": 0.00167143121011192, + "loss": 0.1084, + "step": 27699 + }, + { + "epoch": 0.24044930165536757, + "grad_norm": 0.30078125, + "learning_rate": 0.0016714082160707478, + "loss": 0.1113, + "step": 27700 + }, + { + "epoch": 0.24045798213557174, + "grad_norm": 0.58203125, + "learning_rate": 0.001671385221404692, + "loss": 0.1167, + "step": 27701 + }, + { + "epoch": 0.2404666626157759, + "grad_norm": 0.58984375, + "learning_rate": 0.0016713622261137777, + "loss": 0.0923, + "step": 27702 + }, + { + "epoch": 0.24047534309598007, + "grad_norm": 0.5078125, + "learning_rate": 0.0016713392301980305, + "loss": 0.1309, + "step": 27703 + }, + { + "epoch": 0.24048402357618423, + "grad_norm": 0.326171875, + "learning_rate": 0.0016713162336574757, + "loss": 0.0723, + "step": 27704 + }, + { + "epoch": 0.2404927040563884, + "grad_norm": 0.306640625, + "learning_rate": 0.0016712932364921376, + "loss": 0.0752, + "step": 27705 + }, + { + "epoch": 0.24050138453659256, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016712702387020417, + "loss": 0.1064, + "step": 27706 + }, + { + "epoch": 0.24051006501679673, + "grad_norm": 0.57421875, + "learning_rate": 0.0016712472402872134, + "loss": 0.1504, + "step": 27707 + }, + { + "epoch": 0.2405187454970009, + "grad_norm": 0.271484375, + "learning_rate": 0.0016712242412476779, + "loss": 0.1172, + "step": 27708 + }, + { + "epoch": 0.24052742597720506, + "grad_norm": 1.0625, + "learning_rate": 0.0016712012415834598, + "loss": 0.0859, + "step": 27709 + }, + { + "epoch": 0.24053610645740922, + "grad_norm": 0.171875, + "learning_rate": 0.0016711782412945852, + "loss": 0.0923, + "step": 27710 + }, + { + "epoch": 0.2405447869376134, + "grad_norm": 1.0390625, + "learning_rate": 0.001671155240381078, + "loss": 0.0713, + "step": 27711 + }, + { + "epoch": 0.24055346741781755, + "grad_norm": 0.1474609375, + "learning_rate": 0.0016711322388429644, + "loss": 0.1406, + "step": 27712 + }, + { + "epoch": 0.24056214789802172, + "grad_norm": 0.451171875, + "learning_rate": 0.0016711092366802692, + "loss": 0.1211, + "step": 27713 + }, + { + "epoch": 0.24057082837822588, + "grad_norm": 0.30078125, + "learning_rate": 0.0016710862338930172, + "loss": 0.0952, + "step": 27714 + }, + { + "epoch": 0.24057950885843005, + "grad_norm": 0.65234375, + "learning_rate": 0.001671063230481234, + "loss": 0.0811, + "step": 27715 + }, + { + "epoch": 0.2405881893386342, + "grad_norm": 0.15234375, + "learning_rate": 0.0016710402264449448, + "loss": 0.0869, + "step": 27716 + }, + { + "epoch": 0.24059686981883838, + "grad_norm": 0.78515625, + "learning_rate": 0.0016710172217841743, + "loss": 0.126, + "step": 27717 + }, + { + "epoch": 0.24060555029904254, + "grad_norm": 0.142578125, + "learning_rate": 0.001670994216498948, + "loss": 0.1094, + "step": 27718 + }, + { + "epoch": 0.2406142307792467, + "grad_norm": 0.6796875, + "learning_rate": 0.001670971210589291, + "loss": 0.1152, + "step": 27719 + }, + { + "epoch": 0.24062291125945087, + "grad_norm": 0.38671875, + "learning_rate": 0.0016709482040552287, + "loss": 0.0732, + "step": 27720 + }, + { + "epoch": 0.24063159173965504, + "grad_norm": 0.49609375, + "learning_rate": 0.001670925196896786, + "loss": 0.1108, + "step": 27721 + }, + { + "epoch": 0.2406402722198592, + "grad_norm": 0.95703125, + "learning_rate": 0.0016709021891139876, + "loss": 0.1162, + "step": 27722 + }, + { + "epoch": 0.24064895270006337, + "grad_norm": 0.248046875, + "learning_rate": 0.0016708791807068597, + "loss": 0.1191, + "step": 27723 + }, + { + "epoch": 0.24065763318026753, + "grad_norm": 0.26953125, + "learning_rate": 0.001670856171675427, + "loss": 0.1123, + "step": 27724 + }, + { + "epoch": 0.2406663136604717, + "grad_norm": 1.0546875, + "learning_rate": 0.001670833162019714, + "loss": 0.1348, + "step": 27725 + }, + { + "epoch": 0.24067499414067586, + "grad_norm": 0.5234375, + "learning_rate": 0.0016708101517397469, + "loss": 0.0972, + "step": 27726 + }, + { + "epoch": 0.24068367462088003, + "grad_norm": 0.376953125, + "learning_rate": 0.00167078714083555, + "loss": 0.0981, + "step": 27727 + }, + { + "epoch": 0.2406923551010842, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016707641293071491, + "loss": 0.1279, + "step": 27728 + }, + { + "epoch": 0.24070103558128836, + "grad_norm": 0.11962890625, + "learning_rate": 0.0016707411171545691, + "loss": 0.0767, + "step": 27729 + }, + { + "epoch": 0.24070971606149252, + "grad_norm": 0.2421875, + "learning_rate": 0.0016707181043778352, + "loss": 0.0718, + "step": 27730 + }, + { + "epoch": 0.2407183965416967, + "grad_norm": 0.19140625, + "learning_rate": 0.0016706950909769725, + "loss": 0.0928, + "step": 27731 + }, + { + "epoch": 0.24072707702190085, + "grad_norm": 0.337890625, + "learning_rate": 0.0016706720769520064, + "loss": 0.1562, + "step": 27732 + }, + { + "epoch": 0.24073575750210502, + "grad_norm": 0.40234375, + "learning_rate": 0.0016706490623029618, + "loss": 0.1104, + "step": 27733 + }, + { + "epoch": 0.24074443798230918, + "grad_norm": 0.2451171875, + "learning_rate": 0.0016706260470298637, + "loss": 0.1201, + "step": 27734 + }, + { + "epoch": 0.24075311846251335, + "grad_norm": 0.49609375, + "learning_rate": 0.001670603031132738, + "loss": 0.1016, + "step": 27735 + }, + { + "epoch": 0.24076179894271751, + "grad_norm": 0.3828125, + "learning_rate": 0.001670580014611609, + "loss": 0.123, + "step": 27736 + }, + { + "epoch": 0.24077047942292168, + "grad_norm": 0.23046875, + "learning_rate": 0.0016705569974665025, + "loss": 0.0684, + "step": 27737 + }, + { + "epoch": 0.24077915990312584, + "grad_norm": 0.6171875, + "learning_rate": 0.0016705339796974435, + "loss": 0.0869, + "step": 27738 + }, + { + "epoch": 0.24078784038333, + "grad_norm": 0.2734375, + "learning_rate": 0.0016705109613044567, + "loss": 0.0991, + "step": 27739 + }, + { + "epoch": 0.24079652086353417, + "grad_norm": 0.40234375, + "learning_rate": 0.001670487942287568, + "loss": 0.0889, + "step": 27740 + }, + { + "epoch": 0.24080520134373834, + "grad_norm": 0.486328125, + "learning_rate": 0.0016704649226468023, + "loss": 0.1201, + "step": 27741 + }, + { + "epoch": 0.2408138818239425, + "grad_norm": 0.0908203125, + "learning_rate": 0.0016704419023821845, + "loss": 0.0928, + "step": 27742 + }, + { + "epoch": 0.24082256230414667, + "grad_norm": 0.609375, + "learning_rate": 0.0016704188814937403, + "loss": 0.0771, + "step": 27743 + }, + { + "epoch": 0.24083124278435084, + "grad_norm": 0.439453125, + "learning_rate": 0.001670395859981494, + "loss": 0.1191, + "step": 27744 + }, + { + "epoch": 0.240839923264555, + "grad_norm": 1.21875, + "learning_rate": 0.001670372837845472, + "loss": 0.0952, + "step": 27745 + }, + { + "epoch": 0.24084860374475917, + "grad_norm": 0.099609375, + "learning_rate": 0.0016703498150856986, + "loss": 0.0923, + "step": 27746 + }, + { + "epoch": 0.24085728422496333, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016703267917021994, + "loss": 0.1152, + "step": 27747 + }, + { + "epoch": 0.2408659647051675, + "grad_norm": 1.0, + "learning_rate": 0.0016703037676949993, + "loss": 0.1064, + "step": 27748 + }, + { + "epoch": 0.24087464518537166, + "grad_norm": 0.45703125, + "learning_rate": 0.0016702807430641234, + "loss": 0.0742, + "step": 27749 + }, + { + "epoch": 0.24088332566557583, + "grad_norm": 0.09521484375, + "learning_rate": 0.001670257717809597, + "loss": 0.0942, + "step": 27750 + }, + { + "epoch": 0.24089200614578, + "grad_norm": 0.326171875, + "learning_rate": 0.0016702346919314456, + "loss": 0.1157, + "step": 27751 + }, + { + "epoch": 0.24090068662598416, + "grad_norm": 0.33984375, + "learning_rate": 0.001670211665429694, + "loss": 0.1074, + "step": 27752 + }, + { + "epoch": 0.24090936710618832, + "grad_norm": 0.0908203125, + "learning_rate": 0.0016701886383043674, + "loss": 0.106, + "step": 27753 + }, + { + "epoch": 0.2409180475863925, + "grad_norm": 0.19140625, + "learning_rate": 0.0016701656105554909, + "loss": 0.0928, + "step": 27754 + }, + { + "epoch": 0.24092672806659665, + "grad_norm": 0.265625, + "learning_rate": 0.0016701425821830903, + "loss": 0.1055, + "step": 27755 + }, + { + "epoch": 0.24093540854680082, + "grad_norm": 0.322265625, + "learning_rate": 0.00167011955318719, + "loss": 0.1299, + "step": 27756 + }, + { + "epoch": 0.24094408902700498, + "grad_norm": 0.11474609375, + "learning_rate": 0.0016700965235678159, + "loss": 0.0894, + "step": 27757 + }, + { + "epoch": 0.24095276950720915, + "grad_norm": 0.57421875, + "learning_rate": 0.0016700734933249924, + "loss": 0.1016, + "step": 27758 + }, + { + "epoch": 0.2409614499874133, + "grad_norm": 1.1875, + "learning_rate": 0.001670050462458745, + "loss": 0.1084, + "step": 27759 + }, + { + "epoch": 0.24097013046761748, + "grad_norm": 0.43359375, + "learning_rate": 0.0016700274309690995, + "loss": 0.0903, + "step": 27760 + }, + { + "epoch": 0.24097881094782164, + "grad_norm": 0.09765625, + "learning_rate": 0.0016700043988560802, + "loss": 0.0918, + "step": 27761 + }, + { + "epoch": 0.2409874914280258, + "grad_norm": 0.62109375, + "learning_rate": 0.0016699813661197127, + "loss": 0.0693, + "step": 27762 + }, + { + "epoch": 0.24099617190822997, + "grad_norm": 0.130859375, + "learning_rate": 0.001669958332760022, + "loss": 0.1182, + "step": 27763 + }, + { + "epoch": 0.24100485238843414, + "grad_norm": 0.416015625, + "learning_rate": 0.001669935298777034, + "loss": 0.1123, + "step": 27764 + }, + { + "epoch": 0.2410135328686383, + "grad_norm": 0.421875, + "learning_rate": 0.0016699122641707729, + "loss": 0.1045, + "step": 27765 + }, + { + "epoch": 0.24102221334884247, + "grad_norm": 0.357421875, + "learning_rate": 0.001669889228941264, + "loss": 0.1523, + "step": 27766 + }, + { + "epoch": 0.24103089382904663, + "grad_norm": 0.08740234375, + "learning_rate": 0.0016698661930885335, + "loss": 0.1016, + "step": 27767 + }, + { + "epoch": 0.2410395743092508, + "grad_norm": 0.2216796875, + "learning_rate": 0.0016698431566126055, + "loss": 0.1177, + "step": 27768 + }, + { + "epoch": 0.24104825478945496, + "grad_norm": 0.26171875, + "learning_rate": 0.0016698201195135055, + "loss": 0.1108, + "step": 27769 + }, + { + "epoch": 0.24105693526965913, + "grad_norm": 0.94140625, + "learning_rate": 0.0016697970817912589, + "loss": 0.127, + "step": 27770 + }, + { + "epoch": 0.2410656157498633, + "grad_norm": 1.078125, + "learning_rate": 0.0016697740434458907, + "loss": 0.1484, + "step": 27771 + }, + { + "epoch": 0.24107429623006746, + "grad_norm": 0.2578125, + "learning_rate": 0.0016697510044774264, + "loss": 0.1172, + "step": 27772 + }, + { + "epoch": 0.24108297671027162, + "grad_norm": 0.33203125, + "learning_rate": 0.0016697279648858908, + "loss": 0.1216, + "step": 27773 + }, + { + "epoch": 0.2410916571904758, + "grad_norm": 0.29296875, + "learning_rate": 0.0016697049246713093, + "loss": 0.0913, + "step": 27774 + }, + { + "epoch": 0.24110033767067995, + "grad_norm": 0.482421875, + "learning_rate": 0.0016696818838337069, + "loss": 0.1328, + "step": 27775 + }, + { + "epoch": 0.24110901815088412, + "grad_norm": 0.380859375, + "learning_rate": 0.001669658842373109, + "loss": 0.1016, + "step": 27776 + }, + { + "epoch": 0.24111769863108828, + "grad_norm": 0.150390625, + "learning_rate": 0.001669635800289541, + "loss": 0.1348, + "step": 27777 + }, + { + "epoch": 0.24112637911129245, + "grad_norm": 0.162109375, + "learning_rate": 0.0016696127575830278, + "loss": 0.103, + "step": 27778 + }, + { + "epoch": 0.24113505959149661, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016695897142535944, + "loss": 0.1211, + "step": 27779 + }, + { + "epoch": 0.24114374007170078, + "grad_norm": 0.78515625, + "learning_rate": 0.0016695666703012665, + "loss": 0.1147, + "step": 27780 + }, + { + "epoch": 0.24115242055190494, + "grad_norm": 0.3515625, + "learning_rate": 0.0016695436257260687, + "loss": 0.0815, + "step": 27781 + }, + { + "epoch": 0.2411611010321091, + "grad_norm": 0.87109375, + "learning_rate": 0.001669520580528027, + "loss": 0.103, + "step": 27782 + }, + { + "epoch": 0.24116978151231327, + "grad_norm": 0.6484375, + "learning_rate": 0.0016694975347071657, + "loss": 0.1221, + "step": 27783 + }, + { + "epoch": 0.2411784619925174, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016694744882635106, + "loss": 0.0986, + "step": 27784 + }, + { + "epoch": 0.24118714247272158, + "grad_norm": 0.255859375, + "learning_rate": 0.0016694514411970869, + "loss": 0.1484, + "step": 27785 + }, + { + "epoch": 0.24119582295292574, + "grad_norm": 0.73046875, + "learning_rate": 0.0016694283935079198, + "loss": 0.0923, + "step": 27786 + }, + { + "epoch": 0.2412045034331299, + "grad_norm": 0.2353515625, + "learning_rate": 0.0016694053451960338, + "loss": 0.0781, + "step": 27787 + }, + { + "epoch": 0.24121318391333407, + "grad_norm": 0.396484375, + "learning_rate": 0.0016693822962614551, + "loss": 0.1177, + "step": 27788 + }, + { + "epoch": 0.24122186439353824, + "grad_norm": 0.447265625, + "learning_rate": 0.0016693592467042085, + "loss": 0.1367, + "step": 27789 + }, + { + "epoch": 0.2412305448737424, + "grad_norm": 0.1806640625, + "learning_rate": 0.001669336196524319, + "loss": 0.1006, + "step": 27790 + }, + { + "epoch": 0.24123922535394657, + "grad_norm": 0.50390625, + "learning_rate": 0.001669313145721812, + "loss": 0.0913, + "step": 27791 + }, + { + "epoch": 0.24124790583415073, + "grad_norm": 0.41796875, + "learning_rate": 0.001669290094296713, + "loss": 0.1084, + "step": 27792 + }, + { + "epoch": 0.2412565863143549, + "grad_norm": 1.953125, + "learning_rate": 0.0016692670422490464, + "loss": 0.0986, + "step": 27793 + }, + { + "epoch": 0.24126526679455906, + "grad_norm": 2.125, + "learning_rate": 0.0016692439895788383, + "loss": 0.1709, + "step": 27794 + }, + { + "epoch": 0.24127394727476323, + "grad_norm": 0.330078125, + "learning_rate": 0.0016692209362861134, + "loss": 0.1113, + "step": 27795 + }, + { + "epoch": 0.2412826277549674, + "grad_norm": 0.88671875, + "learning_rate": 0.0016691978823708971, + "loss": 0.1138, + "step": 27796 + }, + { + "epoch": 0.24129130823517156, + "grad_norm": 0.16015625, + "learning_rate": 0.0016691748278332144, + "loss": 0.0674, + "step": 27797 + }, + { + "epoch": 0.24129998871537572, + "grad_norm": 0.109375, + "learning_rate": 0.0016691517726730907, + "loss": 0.1211, + "step": 27798 + }, + { + "epoch": 0.2413086691955799, + "grad_norm": 0.50390625, + "learning_rate": 0.001669128716890551, + "loss": 0.1064, + "step": 27799 + }, + { + "epoch": 0.24131734967578405, + "grad_norm": 0.2021484375, + "learning_rate": 0.0016691056604856212, + "loss": 0.1113, + "step": 27800 + }, + { + "epoch": 0.24132603015598822, + "grad_norm": 0.244140625, + "learning_rate": 0.0016690826034583257, + "loss": 0.0771, + "step": 27801 + }, + { + "epoch": 0.24133471063619238, + "grad_norm": 0.10986328125, + "learning_rate": 0.0016690595458086902, + "loss": 0.0732, + "step": 27802 + }, + { + "epoch": 0.24134339111639655, + "grad_norm": 0.390625, + "learning_rate": 0.0016690364875367393, + "loss": 0.0859, + "step": 27803 + }, + { + "epoch": 0.24135207159660071, + "grad_norm": 0.52734375, + "learning_rate": 0.0016690134286424989, + "loss": 0.103, + "step": 27804 + }, + { + "epoch": 0.24136075207680488, + "grad_norm": 0.3125, + "learning_rate": 0.001668990369125994, + "loss": 0.1084, + "step": 27805 + }, + { + "epoch": 0.24136943255700904, + "grad_norm": 0.140625, + "learning_rate": 0.00166896730898725, + "loss": 0.0938, + "step": 27806 + }, + { + "epoch": 0.2413781130372132, + "grad_norm": 0.26953125, + "learning_rate": 0.0016689442482262914, + "loss": 0.1055, + "step": 27807 + }, + { + "epoch": 0.24138679351741738, + "grad_norm": 0.302734375, + "learning_rate": 0.0016689211868431441, + "loss": 0.1289, + "step": 27808 + }, + { + "epoch": 0.24139547399762154, + "grad_norm": 0.09130859375, + "learning_rate": 0.0016688981248378333, + "loss": 0.1025, + "step": 27809 + }, + { + "epoch": 0.2414041544778257, + "grad_norm": 0.447265625, + "learning_rate": 0.001668875062210384, + "loss": 0.1182, + "step": 27810 + }, + { + "epoch": 0.24141283495802987, + "grad_norm": 0.3359375, + "learning_rate": 0.0016688519989608212, + "loss": 0.1104, + "step": 27811 + }, + { + "epoch": 0.24142151543823404, + "grad_norm": 0.380859375, + "learning_rate": 0.0016688289350891708, + "loss": 0.1699, + "step": 27812 + }, + { + "epoch": 0.2414301959184382, + "grad_norm": 0.125, + "learning_rate": 0.0016688058705954575, + "loss": 0.1143, + "step": 27813 + }, + { + "epoch": 0.24143887639864237, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016687828054797065, + "loss": 0.0845, + "step": 27814 + }, + { + "epoch": 0.24144755687884653, + "grad_norm": 0.478515625, + "learning_rate": 0.0016687597397419436, + "loss": 0.0845, + "step": 27815 + }, + { + "epoch": 0.2414562373590507, + "grad_norm": 0.17578125, + "learning_rate": 0.0016687366733821932, + "loss": 0.0938, + "step": 27816 + }, + { + "epoch": 0.24146491783925486, + "grad_norm": 0.4296875, + "learning_rate": 0.0016687136064004809, + "loss": 0.1191, + "step": 27817 + }, + { + "epoch": 0.24147359831945903, + "grad_norm": 0.427734375, + "learning_rate": 0.0016686905387968318, + "loss": 0.0791, + "step": 27818 + }, + { + "epoch": 0.2414822787996632, + "grad_norm": 0.07177734375, + "learning_rate": 0.0016686674705712716, + "loss": 0.0752, + "step": 27819 + }, + { + "epoch": 0.24149095927986736, + "grad_norm": 0.703125, + "learning_rate": 0.0016686444017238254, + "loss": 0.0996, + "step": 27820 + }, + { + "epoch": 0.24149963976007152, + "grad_norm": 0.326171875, + "learning_rate": 0.0016686213322545179, + "loss": 0.0908, + "step": 27821 + }, + { + "epoch": 0.2415083202402757, + "grad_norm": 0.423828125, + "learning_rate": 0.0016685982621633746, + "loss": 0.0977, + "step": 27822 + }, + { + "epoch": 0.24151700072047985, + "grad_norm": 0.2451171875, + "learning_rate": 0.0016685751914504209, + "loss": 0.0762, + "step": 27823 + }, + { + "epoch": 0.24152568120068402, + "grad_norm": 0.1181640625, + "learning_rate": 0.001668552120115682, + "loss": 0.1235, + "step": 27824 + }, + { + "epoch": 0.24153436168088818, + "grad_norm": 0.34765625, + "learning_rate": 0.0016685290481591828, + "loss": 0.1025, + "step": 27825 + }, + { + "epoch": 0.24154304216109235, + "grad_norm": 0.33984375, + "learning_rate": 0.001668505975580949, + "loss": 0.1191, + "step": 27826 + }, + { + "epoch": 0.2415517226412965, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016684829023810052, + "loss": 0.1162, + "step": 27827 + }, + { + "epoch": 0.24156040312150068, + "grad_norm": 0.275390625, + "learning_rate": 0.0016684598285593775, + "loss": 0.1191, + "step": 27828 + }, + { + "epoch": 0.24156908360170484, + "grad_norm": 0.62890625, + "learning_rate": 0.0016684367541160905, + "loss": 0.1445, + "step": 27829 + }, + { + "epoch": 0.241577764081909, + "grad_norm": 0.828125, + "learning_rate": 0.0016684136790511698, + "loss": 0.1562, + "step": 27830 + }, + { + "epoch": 0.24158644456211317, + "grad_norm": 0.1591796875, + "learning_rate": 0.00166839060336464, + "loss": 0.0903, + "step": 27831 + }, + { + "epoch": 0.24159512504231734, + "grad_norm": 0.7890625, + "learning_rate": 0.0016683675270565274, + "loss": 0.1211, + "step": 27832 + }, + { + "epoch": 0.2416038055225215, + "grad_norm": 0.91796875, + "learning_rate": 0.0016683444501268561, + "loss": 0.0801, + "step": 27833 + }, + { + "epoch": 0.24161248600272567, + "grad_norm": 0.58203125, + "learning_rate": 0.0016683213725756521, + "loss": 0.1113, + "step": 27834 + }, + { + "epoch": 0.24162116648292983, + "grad_norm": 0.1083984375, + "learning_rate": 0.00166829829440294, + "loss": 0.1426, + "step": 27835 + }, + { + "epoch": 0.241629846963134, + "grad_norm": 0.314453125, + "learning_rate": 0.001668275215608746, + "loss": 0.1328, + "step": 27836 + }, + { + "epoch": 0.24163852744333816, + "grad_norm": 0.6015625, + "learning_rate": 0.0016682521361930942, + "loss": 0.1035, + "step": 27837 + }, + { + "epoch": 0.24164720792354233, + "grad_norm": 0.189453125, + "learning_rate": 0.0016682290561560109, + "loss": 0.1436, + "step": 27838 + }, + { + "epoch": 0.2416558884037465, + "grad_norm": 0.1005859375, + "learning_rate": 0.0016682059754975205, + "loss": 0.0981, + "step": 27839 + }, + { + "epoch": 0.24166456888395066, + "grad_norm": 0.07470703125, + "learning_rate": 0.001668182894217649, + "loss": 0.0806, + "step": 27840 + }, + { + "epoch": 0.24167324936415482, + "grad_norm": 0.162109375, + "learning_rate": 0.0016681598123164205, + "loss": 0.126, + "step": 27841 + }, + { + "epoch": 0.241681929844359, + "grad_norm": 0.37109375, + "learning_rate": 0.0016681367297938614, + "loss": 0.0938, + "step": 27842 + }, + { + "epoch": 0.24169061032456315, + "grad_norm": 0.8125, + "learning_rate": 0.001668113646649997, + "loss": 0.1426, + "step": 27843 + }, + { + "epoch": 0.24169929080476732, + "grad_norm": 0.7734375, + "learning_rate": 0.0016680905628848515, + "loss": 0.1025, + "step": 27844 + }, + { + "epoch": 0.24170797128497148, + "grad_norm": 0.057373046875, + "learning_rate": 0.0016680674784984506, + "loss": 0.0618, + "step": 27845 + }, + { + "epoch": 0.24171665176517565, + "grad_norm": 0.30078125, + "learning_rate": 0.0016680443934908196, + "loss": 0.0747, + "step": 27846 + }, + { + "epoch": 0.24172533224537981, + "grad_norm": 0.65625, + "learning_rate": 0.0016680213078619842, + "loss": 0.1455, + "step": 27847 + }, + { + "epoch": 0.24173401272558398, + "grad_norm": 0.54296875, + "learning_rate": 0.001667998221611969, + "loss": 0.1167, + "step": 27848 + }, + { + "epoch": 0.24174269320578815, + "grad_norm": 0.37109375, + "learning_rate": 0.0016679751347407994, + "loss": 0.1309, + "step": 27849 + }, + { + "epoch": 0.2417513736859923, + "grad_norm": 0.1923828125, + "learning_rate": 0.0016679520472485008, + "loss": 0.124, + "step": 27850 + }, + { + "epoch": 0.24176005416619648, + "grad_norm": 0.328125, + "learning_rate": 0.0016679289591350985, + "loss": 0.126, + "step": 27851 + }, + { + "epoch": 0.24176873464640064, + "grad_norm": 0.34765625, + "learning_rate": 0.0016679058704006174, + "loss": 0.168, + "step": 27852 + }, + { + "epoch": 0.2417774151266048, + "grad_norm": 0.2412109375, + "learning_rate": 0.001667882781045083, + "loss": 0.0845, + "step": 27853 + }, + { + "epoch": 0.24178609560680897, + "grad_norm": 0.09228515625, + "learning_rate": 0.0016678596910685208, + "loss": 0.0786, + "step": 27854 + }, + { + "epoch": 0.24179477608701314, + "grad_norm": 0.1640625, + "learning_rate": 0.0016678366004709555, + "loss": 0.1025, + "step": 27855 + }, + { + "epoch": 0.2418034565672173, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016678135092524128, + "loss": 0.1006, + "step": 27856 + }, + { + "epoch": 0.24181213704742147, + "grad_norm": 0.55078125, + "learning_rate": 0.0016677904174129175, + "loss": 0.1318, + "step": 27857 + }, + { + "epoch": 0.24182081752762563, + "grad_norm": 0.765625, + "learning_rate": 0.0016677673249524954, + "loss": 0.0991, + "step": 27858 + }, + { + "epoch": 0.2418294980078298, + "grad_norm": 0.1240234375, + "learning_rate": 0.0016677442318711714, + "loss": 0.1328, + "step": 27859 + }, + { + "epoch": 0.24183817848803396, + "grad_norm": 0.17578125, + "learning_rate": 0.0016677211381689709, + "loss": 0.1045, + "step": 27860 + }, + { + "epoch": 0.24184685896823813, + "grad_norm": 0.52734375, + "learning_rate": 0.001667698043845919, + "loss": 0.0703, + "step": 27861 + }, + { + "epoch": 0.2418555394484423, + "grad_norm": 0.248046875, + "learning_rate": 0.001667674948902041, + "loss": 0.1079, + "step": 27862 + }, + { + "epoch": 0.24186421992864646, + "grad_norm": 0.345703125, + "learning_rate": 0.0016676518533373626, + "loss": 0.1211, + "step": 27863 + }, + { + "epoch": 0.24187290040885062, + "grad_norm": 0.37109375, + "learning_rate": 0.001667628757151908, + "loss": 0.125, + "step": 27864 + }, + { + "epoch": 0.2418815808890548, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016676056603457036, + "loss": 0.1328, + "step": 27865 + }, + { + "epoch": 0.24189026136925895, + "grad_norm": 0.19140625, + "learning_rate": 0.0016675825629187742, + "loss": 0.1211, + "step": 27866 + }, + { + "epoch": 0.24189894184946312, + "grad_norm": 0.203125, + "learning_rate": 0.0016675594648711447, + "loss": 0.1084, + "step": 27867 + }, + { + "epoch": 0.24190762232966728, + "grad_norm": 0.220703125, + "learning_rate": 0.001667536366202841, + "loss": 0.085, + "step": 27868 + }, + { + "epoch": 0.24191630280987145, + "grad_norm": 0.140625, + "learning_rate": 0.001667513266913888, + "loss": 0.0791, + "step": 27869 + }, + { + "epoch": 0.2419249832900756, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016674901670043108, + "loss": 0.1367, + "step": 27870 + }, + { + "epoch": 0.24193366377027978, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016674670664741352, + "loss": 0.1328, + "step": 27871 + }, + { + "epoch": 0.24194234425048394, + "grad_norm": 0.279296875, + "learning_rate": 0.001667443965323386, + "loss": 0.1006, + "step": 27872 + }, + { + "epoch": 0.2419510247306881, + "grad_norm": 0.50390625, + "learning_rate": 0.0016674208635520885, + "loss": 0.0806, + "step": 27873 + }, + { + "epoch": 0.24195970521089227, + "grad_norm": 0.185546875, + "learning_rate": 0.0016673977611602682, + "loss": 0.1016, + "step": 27874 + }, + { + "epoch": 0.24196838569109644, + "grad_norm": 0.2001953125, + "learning_rate": 0.00166737465814795, + "loss": 0.0947, + "step": 27875 + }, + { + "epoch": 0.2419770661713006, + "grad_norm": 0.12109375, + "learning_rate": 0.0016673515545151596, + "loss": 0.1035, + "step": 27876 + }, + { + "epoch": 0.24198574665150477, + "grad_norm": 0.484375, + "learning_rate": 0.0016673284502619222, + "loss": 0.0898, + "step": 27877 + }, + { + "epoch": 0.24199442713170893, + "grad_norm": 0.9765625, + "learning_rate": 0.0016673053453882627, + "loss": 0.1133, + "step": 27878 + }, + { + "epoch": 0.2420031076119131, + "grad_norm": 0.1923828125, + "learning_rate": 0.0016672822398942067, + "loss": 0.1543, + "step": 27879 + }, + { + "epoch": 0.24201178809211726, + "grad_norm": 0.09130859375, + "learning_rate": 0.0016672591337797793, + "loss": 0.085, + "step": 27880 + }, + { + "epoch": 0.24202046857232143, + "grad_norm": 0.34375, + "learning_rate": 0.0016672360270450059, + "loss": 0.1084, + "step": 27881 + }, + { + "epoch": 0.2420291490525256, + "grad_norm": 0.1416015625, + "learning_rate": 0.0016672129196899114, + "loss": 0.1104, + "step": 27882 + }, + { + "epoch": 0.24203782953272976, + "grad_norm": 0.16796875, + "learning_rate": 0.001667189811714522, + "loss": 0.166, + "step": 27883 + }, + { + "epoch": 0.24204651001293392, + "grad_norm": 1.15625, + "learning_rate": 0.001667166703118862, + "loss": 0.1172, + "step": 27884 + }, + { + "epoch": 0.2420551904931381, + "grad_norm": 0.1923828125, + "learning_rate": 0.0016671435939029568, + "loss": 0.0776, + "step": 27885 + }, + { + "epoch": 0.24206387097334225, + "grad_norm": 0.263671875, + "learning_rate": 0.0016671204840668324, + "loss": 0.1299, + "step": 27886 + }, + { + "epoch": 0.24207255145354642, + "grad_norm": 0.33984375, + "learning_rate": 0.001667097373610513, + "loss": 0.1387, + "step": 27887 + }, + { + "epoch": 0.24208123193375058, + "grad_norm": 0.431640625, + "learning_rate": 0.0016670742625340246, + "loss": 0.0742, + "step": 27888 + }, + { + "epoch": 0.24208991241395475, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016670511508373926, + "loss": 0.0825, + "step": 27889 + }, + { + "epoch": 0.24209859289415891, + "grad_norm": 0.2275390625, + "learning_rate": 0.001667028038520642, + "loss": 0.1465, + "step": 27890 + }, + { + "epoch": 0.24210727337436308, + "grad_norm": 0.169921875, + "learning_rate": 0.0016670049255837975, + "loss": 0.0811, + "step": 27891 + }, + { + "epoch": 0.24211595385456725, + "grad_norm": 0.09765625, + "learning_rate": 0.0016669818120268855, + "loss": 0.1396, + "step": 27892 + }, + { + "epoch": 0.2421246343347714, + "grad_norm": 0.478515625, + "learning_rate": 0.0016669586978499303, + "loss": 0.1143, + "step": 27893 + }, + { + "epoch": 0.24213331481497558, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016669355830529577, + "loss": 0.0801, + "step": 27894 + }, + { + "epoch": 0.24214199529517974, + "grad_norm": 0.51953125, + "learning_rate": 0.0016669124676359929, + "loss": 0.1191, + "step": 27895 + }, + { + "epoch": 0.2421506757753839, + "grad_norm": 0.330078125, + "learning_rate": 0.0016668893515990615, + "loss": 0.1152, + "step": 27896 + }, + { + "epoch": 0.24215935625558807, + "grad_norm": 0.66015625, + "learning_rate": 0.001666866234942188, + "loss": 0.1006, + "step": 27897 + }, + { + "epoch": 0.24216803673579224, + "grad_norm": 0.63671875, + "learning_rate": 0.0016668431176653981, + "loss": 0.1113, + "step": 27898 + }, + { + "epoch": 0.2421767172159964, + "grad_norm": 0.61328125, + "learning_rate": 0.0016668199997687174, + "loss": 0.1045, + "step": 27899 + }, + { + "epoch": 0.24218539769620057, + "grad_norm": 0.08837890625, + "learning_rate": 0.0016667968812521703, + "loss": 0.0991, + "step": 27900 + }, + { + "epoch": 0.24219407817640473, + "grad_norm": 0.5078125, + "learning_rate": 0.0016667737621157833, + "loss": 0.1123, + "step": 27901 + }, + { + "epoch": 0.2422027586566089, + "grad_norm": 0.28125, + "learning_rate": 0.0016667506423595807, + "loss": 0.1177, + "step": 27902 + }, + { + "epoch": 0.24221143913681306, + "grad_norm": 0.75390625, + "learning_rate": 0.001666727521983588, + "loss": 0.0903, + "step": 27903 + }, + { + "epoch": 0.24222011961701723, + "grad_norm": 0.431640625, + "learning_rate": 0.0016667044009878308, + "loss": 0.1016, + "step": 27904 + }, + { + "epoch": 0.2422288000972214, + "grad_norm": 0.8828125, + "learning_rate": 0.001666681279372334, + "loss": 0.0825, + "step": 27905 + }, + { + "epoch": 0.24223748057742556, + "grad_norm": 0.5390625, + "learning_rate": 0.0016666581571371233, + "loss": 0.1162, + "step": 27906 + }, + { + "epoch": 0.2422461610576297, + "grad_norm": 0.33984375, + "learning_rate": 0.0016666350342822236, + "loss": 0.0869, + "step": 27907 + }, + { + "epoch": 0.24225484153783386, + "grad_norm": 0.21484375, + "learning_rate": 0.0016666119108076603, + "loss": 0.1191, + "step": 27908 + }, + { + "epoch": 0.24226352201803802, + "grad_norm": 0.50390625, + "learning_rate": 0.0016665887867134588, + "loss": 0.1035, + "step": 27909 + }, + { + "epoch": 0.2422722024982422, + "grad_norm": 0.63671875, + "learning_rate": 0.0016665656619996444, + "loss": 0.0942, + "step": 27910 + }, + { + "epoch": 0.24228088297844635, + "grad_norm": 0.275390625, + "learning_rate": 0.0016665425366662423, + "loss": 0.1426, + "step": 27911 + }, + { + "epoch": 0.24228956345865052, + "grad_norm": 0.388671875, + "learning_rate": 0.0016665194107132778, + "loss": 0.1689, + "step": 27912 + }, + { + "epoch": 0.24229824393885468, + "grad_norm": 0.88671875, + "learning_rate": 0.001666496284140776, + "loss": 0.1113, + "step": 27913 + }, + { + "epoch": 0.24230692441905885, + "grad_norm": 0.318359375, + "learning_rate": 0.0016664731569487625, + "loss": 0.0952, + "step": 27914 + }, + { + "epoch": 0.24231560489926302, + "grad_norm": 0.359375, + "learning_rate": 0.0016664500291372626, + "loss": 0.0845, + "step": 27915 + }, + { + "epoch": 0.24232428537946718, + "grad_norm": 1.0546875, + "learning_rate": 0.0016664269007063011, + "loss": 0.1816, + "step": 27916 + }, + { + "epoch": 0.24233296585967135, + "grad_norm": 0.45703125, + "learning_rate": 0.0016664037716559038, + "loss": 0.0723, + "step": 27917 + }, + { + "epoch": 0.2423416463398755, + "grad_norm": 0.13671875, + "learning_rate": 0.0016663806419860959, + "loss": 0.0962, + "step": 27918 + }, + { + "epoch": 0.24235032682007968, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016663575116969026, + "loss": 0.1035, + "step": 27919 + }, + { + "epoch": 0.24235900730028384, + "grad_norm": 0.134765625, + "learning_rate": 0.0016663343807883494, + "loss": 0.1357, + "step": 27920 + }, + { + "epoch": 0.242367687780488, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016663112492604612, + "loss": 0.1582, + "step": 27921 + }, + { + "epoch": 0.24237636826069217, + "grad_norm": 0.453125, + "learning_rate": 0.0016662881171132635, + "loss": 0.1328, + "step": 27922 + }, + { + "epoch": 0.24238504874089634, + "grad_norm": 0.240234375, + "learning_rate": 0.001666264984346782, + "loss": 0.1021, + "step": 27923 + }, + { + "epoch": 0.2423937292211005, + "grad_norm": 0.19140625, + "learning_rate": 0.0016662418509610414, + "loss": 0.105, + "step": 27924 + }, + { + "epoch": 0.24240240970130467, + "grad_norm": 0.447265625, + "learning_rate": 0.0016662187169560673, + "loss": 0.1367, + "step": 27925 + }, + { + "epoch": 0.24241109018150883, + "grad_norm": 0.470703125, + "learning_rate": 0.0016661955823318846, + "loss": 0.0991, + "step": 27926 + }, + { + "epoch": 0.242419770661713, + "grad_norm": 0.11669921875, + "learning_rate": 0.001666172447088519, + "loss": 0.127, + "step": 27927 + }, + { + "epoch": 0.24242845114191716, + "grad_norm": 0.6953125, + "learning_rate": 0.0016661493112259962, + "loss": 0.1084, + "step": 27928 + }, + { + "epoch": 0.24243713162212133, + "grad_norm": 0.26171875, + "learning_rate": 0.0016661261747443405, + "loss": 0.085, + "step": 27929 + }, + { + "epoch": 0.2424458121023255, + "grad_norm": 0.326171875, + "learning_rate": 0.0016661030376435779, + "loss": 0.0815, + "step": 27930 + }, + { + "epoch": 0.24245449258252966, + "grad_norm": 0.259765625, + "learning_rate": 0.0016660798999237334, + "loss": 0.1211, + "step": 27931 + }, + { + "epoch": 0.24246317306273382, + "grad_norm": 0.26953125, + "learning_rate": 0.0016660567615848325, + "loss": 0.1328, + "step": 27932 + }, + { + "epoch": 0.242471853542938, + "grad_norm": 0.5078125, + "learning_rate": 0.001666033622626901, + "loss": 0.1602, + "step": 27933 + }, + { + "epoch": 0.24248053402314215, + "grad_norm": 0.19140625, + "learning_rate": 0.0016660104830499628, + "loss": 0.1328, + "step": 27934 + }, + { + "epoch": 0.24248921450334632, + "grad_norm": 0.373046875, + "learning_rate": 0.0016659873428540445, + "loss": 0.1338, + "step": 27935 + }, + { + "epoch": 0.24249789498355048, + "grad_norm": 0.12109375, + "learning_rate": 0.0016659642020391706, + "loss": 0.1143, + "step": 27936 + }, + { + "epoch": 0.24250657546375465, + "grad_norm": 0.203125, + "learning_rate": 0.0016659410606053669, + "loss": 0.0742, + "step": 27937 + }, + { + "epoch": 0.2425152559439588, + "grad_norm": 0.12158203125, + "learning_rate": 0.0016659179185526585, + "loss": 0.1016, + "step": 27938 + }, + { + "epoch": 0.24252393642416298, + "grad_norm": 0.1064453125, + "learning_rate": 0.001665894775881071, + "loss": 0.123, + "step": 27939 + }, + { + "epoch": 0.24253261690436714, + "grad_norm": 0.7734375, + "learning_rate": 0.0016658716325906294, + "loss": 0.0703, + "step": 27940 + }, + { + "epoch": 0.2425412973845713, + "grad_norm": 0.30859375, + "learning_rate": 0.001665848488681359, + "loss": 0.0864, + "step": 27941 + }, + { + "epoch": 0.24254997786477547, + "grad_norm": 0.7265625, + "learning_rate": 0.0016658253441532854, + "loss": 0.1133, + "step": 27942 + }, + { + "epoch": 0.24255865834497964, + "grad_norm": 0.63671875, + "learning_rate": 0.0016658021990064334, + "loss": 0.0933, + "step": 27943 + }, + { + "epoch": 0.2425673388251838, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016657790532408289, + "loss": 0.0771, + "step": 27944 + }, + { + "epoch": 0.24257601930538797, + "grad_norm": 0.10546875, + "learning_rate": 0.0016657559068564967, + "loss": 0.1016, + "step": 27945 + }, + { + "epoch": 0.24258469978559213, + "grad_norm": 0.193359375, + "learning_rate": 0.0016657327598534623, + "loss": 0.0962, + "step": 27946 + }, + { + "epoch": 0.2425933802657963, + "grad_norm": 0.310546875, + "learning_rate": 0.0016657096122317512, + "loss": 0.0825, + "step": 27947 + }, + { + "epoch": 0.24260206074600046, + "grad_norm": 0.107421875, + "learning_rate": 0.0016656864639913887, + "loss": 0.1309, + "step": 27948 + }, + { + "epoch": 0.24261074122620463, + "grad_norm": 0.14453125, + "learning_rate": 0.0016656633151323997, + "loss": 0.0806, + "step": 27949 + }, + { + "epoch": 0.2426194217064088, + "grad_norm": 0.59375, + "learning_rate": 0.0016656401656548099, + "loss": 0.1338, + "step": 27950 + }, + { + "epoch": 0.24262810218661296, + "grad_norm": 0.251953125, + "learning_rate": 0.0016656170155586445, + "loss": 0.0967, + "step": 27951 + }, + { + "epoch": 0.24263678266681712, + "grad_norm": 0.703125, + "learning_rate": 0.0016655938648439288, + "loss": 0.0991, + "step": 27952 + }, + { + "epoch": 0.2426454631470213, + "grad_norm": 0.93359375, + "learning_rate": 0.0016655707135106884, + "loss": 0.0791, + "step": 27953 + }, + { + "epoch": 0.24265414362722545, + "grad_norm": 0.2353515625, + "learning_rate": 0.001665547561558948, + "loss": 0.125, + "step": 27954 + }, + { + "epoch": 0.24266282410742962, + "grad_norm": 0.16015625, + "learning_rate": 0.0016655244089887335, + "loss": 0.0781, + "step": 27955 + }, + { + "epoch": 0.24267150458763379, + "grad_norm": 0.609375, + "learning_rate": 0.00166550125580007, + "loss": 0.084, + "step": 27956 + }, + { + "epoch": 0.24268018506783795, + "grad_norm": 0.28125, + "learning_rate": 0.0016654781019929828, + "loss": 0.1289, + "step": 27957 + }, + { + "epoch": 0.24268886554804212, + "grad_norm": 0.322265625, + "learning_rate": 0.0016654549475674971, + "loss": 0.0923, + "step": 27958 + }, + { + "epoch": 0.24269754602824628, + "grad_norm": 0.107421875, + "learning_rate": 0.001665431792523639, + "loss": 0.0806, + "step": 27959 + }, + { + "epoch": 0.24270622650845045, + "grad_norm": 0.3828125, + "learning_rate": 0.0016654086368614322, + "loss": 0.105, + "step": 27960 + }, + { + "epoch": 0.2427149069886546, + "grad_norm": 0.67578125, + "learning_rate": 0.0016653854805809038, + "loss": 0.1045, + "step": 27961 + }, + { + "epoch": 0.24272358746885878, + "grad_norm": 0.4921875, + "learning_rate": 0.0016653623236820778, + "loss": 0.1416, + "step": 27962 + }, + { + "epoch": 0.24273226794906294, + "grad_norm": 0.16796875, + "learning_rate": 0.0016653391661649805, + "loss": 0.1133, + "step": 27963 + }, + { + "epoch": 0.2427409484292671, + "grad_norm": 0.2734375, + "learning_rate": 0.0016653160080296364, + "loss": 0.1484, + "step": 27964 + }, + { + "epoch": 0.24274962890947127, + "grad_norm": 0.3046875, + "learning_rate": 0.0016652928492760714, + "loss": 0.1025, + "step": 27965 + }, + { + "epoch": 0.24275830938967544, + "grad_norm": 0.62890625, + "learning_rate": 0.0016652696899043108, + "loss": 0.0991, + "step": 27966 + }, + { + "epoch": 0.2427669898698796, + "grad_norm": 0.1279296875, + "learning_rate": 0.0016652465299143793, + "loss": 0.123, + "step": 27967 + }, + { + "epoch": 0.24277567035008377, + "grad_norm": 0.44921875, + "learning_rate": 0.001665223369306303, + "loss": 0.0889, + "step": 27968 + }, + { + "epoch": 0.24278435083028793, + "grad_norm": 0.337890625, + "learning_rate": 0.0016652002080801069, + "loss": 0.123, + "step": 27969 + }, + { + "epoch": 0.2427930313104921, + "grad_norm": 0.173828125, + "learning_rate": 0.0016651770462358165, + "loss": 0.1211, + "step": 27970 + }, + { + "epoch": 0.24280171179069626, + "grad_norm": 0.09912109375, + "learning_rate": 0.0016651538837734565, + "loss": 0.1123, + "step": 27971 + }, + { + "epoch": 0.24281039227090043, + "grad_norm": 0.296875, + "learning_rate": 0.001665130720693053, + "loss": 0.083, + "step": 27972 + }, + { + "epoch": 0.2428190727511046, + "grad_norm": 0.427734375, + "learning_rate": 0.0016651075569946313, + "loss": 0.0645, + "step": 27973 + }, + { + "epoch": 0.24282775323130876, + "grad_norm": 0.75, + "learning_rate": 0.0016650843926782162, + "loss": 0.1299, + "step": 27974 + }, + { + "epoch": 0.24283643371151292, + "grad_norm": 0.251953125, + "learning_rate": 0.0016650612277438331, + "loss": 0.1133, + "step": 27975 + }, + { + "epoch": 0.2428451141917171, + "grad_norm": 0.96484375, + "learning_rate": 0.0016650380621915078, + "loss": 0.1211, + "step": 27976 + }, + { + "epoch": 0.24285379467192125, + "grad_norm": 0.19921875, + "learning_rate": 0.001665014896021265, + "loss": 0.0986, + "step": 27977 + }, + { + "epoch": 0.24286247515212542, + "grad_norm": 0.42578125, + "learning_rate": 0.001664991729233131, + "loss": 0.1211, + "step": 27978 + }, + { + "epoch": 0.24287115563232958, + "grad_norm": 0.287109375, + "learning_rate": 0.00166496856182713, + "loss": 0.1035, + "step": 27979 + }, + { + "epoch": 0.24287983611253375, + "grad_norm": 0.14453125, + "learning_rate": 0.001664945393803288, + "loss": 0.0923, + "step": 27980 + }, + { + "epoch": 0.2428885165927379, + "grad_norm": 0.1455078125, + "learning_rate": 0.0016649222251616305, + "loss": 0.0874, + "step": 27981 + }, + { + "epoch": 0.24289719707294208, + "grad_norm": 0.0849609375, + "learning_rate": 0.0016648990559021823, + "loss": 0.1074, + "step": 27982 + }, + { + "epoch": 0.24290587755314624, + "grad_norm": 0.57421875, + "learning_rate": 0.001664875886024969, + "loss": 0.1221, + "step": 27983 + }, + { + "epoch": 0.2429145580333504, + "grad_norm": 0.078125, + "learning_rate": 0.001664852715530016, + "loss": 0.0806, + "step": 27984 + }, + { + "epoch": 0.24292323851355457, + "grad_norm": 0.4453125, + "learning_rate": 0.0016648295444173485, + "loss": 0.0972, + "step": 27985 + }, + { + "epoch": 0.24293191899375874, + "grad_norm": 0.150390625, + "learning_rate": 0.0016648063726869917, + "loss": 0.0938, + "step": 27986 + }, + { + "epoch": 0.2429405994739629, + "grad_norm": 0.349609375, + "learning_rate": 0.0016647832003389715, + "loss": 0.1118, + "step": 27987 + }, + { + "epoch": 0.24294927995416707, + "grad_norm": 0.345703125, + "learning_rate": 0.0016647600273733128, + "loss": 0.0952, + "step": 27988 + }, + { + "epoch": 0.24295796043437123, + "grad_norm": 0.162109375, + "learning_rate": 0.0016647368537900408, + "loss": 0.0942, + "step": 27989 + }, + { + "epoch": 0.2429666409145754, + "grad_norm": 0.7890625, + "learning_rate": 0.0016647136795891811, + "loss": 0.1172, + "step": 27990 + }, + { + "epoch": 0.24297532139477956, + "grad_norm": 0.158203125, + "learning_rate": 0.0016646905047707592, + "loss": 0.0928, + "step": 27991 + }, + { + "epoch": 0.24298400187498373, + "grad_norm": 0.169921875, + "learning_rate": 0.0016646673293348, + "loss": 0.1084, + "step": 27992 + }, + { + "epoch": 0.2429926823551879, + "grad_norm": 0.6328125, + "learning_rate": 0.0016646441532813293, + "loss": 0.1436, + "step": 27993 + }, + { + "epoch": 0.24300136283539206, + "grad_norm": 0.201171875, + "learning_rate": 0.0016646209766103723, + "loss": 0.0918, + "step": 27994 + }, + { + "epoch": 0.24301004331559622, + "grad_norm": 0.408203125, + "learning_rate": 0.0016645977993219542, + "loss": 0.106, + "step": 27995 + }, + { + "epoch": 0.2430187237958004, + "grad_norm": 0.1982421875, + "learning_rate": 0.0016645746214161003, + "loss": 0.1094, + "step": 27996 + }, + { + "epoch": 0.24302740427600455, + "grad_norm": 0.10498046875, + "learning_rate": 0.0016645514428928363, + "loss": 0.123, + "step": 27997 + }, + { + "epoch": 0.24303608475620872, + "grad_norm": 0.10009765625, + "learning_rate": 0.001664528263752187, + "loss": 0.1045, + "step": 27998 + }, + { + "epoch": 0.24304476523641289, + "grad_norm": 0.185546875, + "learning_rate": 0.0016645050839941783, + "loss": 0.1191, + "step": 27999 + }, + { + "epoch": 0.24305344571661705, + "grad_norm": 0.248046875, + "learning_rate": 0.0016644819036188351, + "loss": 0.0986, + "step": 28000 + }, + { + "epoch": 0.24306212619682122, + "grad_norm": 0.31640625, + "learning_rate": 0.0016644587226261835, + "loss": 0.1182, + "step": 28001 + }, + { + "epoch": 0.24307080667702538, + "grad_norm": 0.0888671875, + "learning_rate": 0.001664435541016248, + "loss": 0.127, + "step": 28002 + }, + { + "epoch": 0.24307948715722955, + "grad_norm": 0.50390625, + "learning_rate": 0.0016644123587890541, + "loss": 0.1016, + "step": 28003 + }, + { + "epoch": 0.2430881676374337, + "grad_norm": 0.130859375, + "learning_rate": 0.0016643891759446272, + "loss": 0.1348, + "step": 28004 + }, + { + "epoch": 0.24309684811763788, + "grad_norm": 0.166015625, + "learning_rate": 0.0016643659924829933, + "loss": 0.1367, + "step": 28005 + }, + { + "epoch": 0.24310552859784204, + "grad_norm": 0.07958984375, + "learning_rate": 0.001664342808404177, + "loss": 0.0962, + "step": 28006 + }, + { + "epoch": 0.2431142090780462, + "grad_norm": 0.9765625, + "learning_rate": 0.0016643196237082038, + "loss": 0.1143, + "step": 28007 + }, + { + "epoch": 0.24312288955825037, + "grad_norm": 0.48828125, + "learning_rate": 0.001664296438395099, + "loss": 0.0762, + "step": 28008 + }, + { + "epoch": 0.24313157003845454, + "grad_norm": 0.470703125, + "learning_rate": 0.0016642732524648888, + "loss": 0.0898, + "step": 28009 + }, + { + "epoch": 0.2431402505186587, + "grad_norm": 0.1484375, + "learning_rate": 0.001664250065917597, + "loss": 0.0791, + "step": 28010 + }, + { + "epoch": 0.24314893099886287, + "grad_norm": 0.1884765625, + "learning_rate": 0.0016642268787532505, + "loss": 0.0898, + "step": 28011 + }, + { + "epoch": 0.24315761147906703, + "grad_norm": 0.90625, + "learning_rate": 0.0016642036909718734, + "loss": 0.1514, + "step": 28012 + }, + { + "epoch": 0.2431662919592712, + "grad_norm": 0.71875, + "learning_rate": 0.001664180502573492, + "loss": 0.1279, + "step": 28013 + }, + { + "epoch": 0.24317497243947536, + "grad_norm": 2.421875, + "learning_rate": 0.0016641573135581307, + "loss": 0.209, + "step": 28014 + }, + { + "epoch": 0.24318365291967953, + "grad_norm": 0.33203125, + "learning_rate": 0.0016641341239258158, + "loss": 0.1064, + "step": 28015 + }, + { + "epoch": 0.2431923333998837, + "grad_norm": 0.9140625, + "learning_rate": 0.0016641109336765728, + "loss": 0.0986, + "step": 28016 + }, + { + "epoch": 0.24320101388008786, + "grad_norm": 0.2353515625, + "learning_rate": 0.0016640877428104259, + "loss": 0.1299, + "step": 28017 + }, + { + "epoch": 0.24320969436029202, + "grad_norm": 0.125, + "learning_rate": 0.0016640645513274013, + "loss": 0.0952, + "step": 28018 + }, + { + "epoch": 0.2432183748404962, + "grad_norm": 0.890625, + "learning_rate": 0.0016640413592275242, + "loss": 0.1055, + "step": 28019 + }, + { + "epoch": 0.24322705532070035, + "grad_norm": 0.2333984375, + "learning_rate": 0.0016640181665108199, + "loss": 0.0972, + "step": 28020 + }, + { + "epoch": 0.24323573580090452, + "grad_norm": 0.8046875, + "learning_rate": 0.001663994973177314, + "loss": 0.2461, + "step": 28021 + }, + { + "epoch": 0.24324441628110868, + "grad_norm": 0.11962890625, + "learning_rate": 0.0016639717792270314, + "loss": 0.084, + "step": 28022 + }, + { + "epoch": 0.24325309676131285, + "grad_norm": 0.2041015625, + "learning_rate": 0.001663948584659998, + "loss": 0.123, + "step": 28023 + }, + { + "epoch": 0.243261777241517, + "grad_norm": 0.330078125, + "learning_rate": 0.0016639253894762387, + "loss": 0.0928, + "step": 28024 + }, + { + "epoch": 0.24327045772172118, + "grad_norm": 0.3046875, + "learning_rate": 0.001663902193675779, + "loss": 0.1094, + "step": 28025 + }, + { + "epoch": 0.24327913820192534, + "grad_norm": 0.462890625, + "learning_rate": 0.0016638789972586447, + "loss": 0.1021, + "step": 28026 + }, + { + "epoch": 0.2432878186821295, + "grad_norm": 0.146484375, + "learning_rate": 0.0016638558002248604, + "loss": 0.1084, + "step": 28027 + }, + { + "epoch": 0.24329649916233367, + "grad_norm": 0.1962890625, + "learning_rate": 0.0016638326025744524, + "loss": 0.1094, + "step": 28028 + }, + { + "epoch": 0.24330517964253784, + "grad_norm": 0.6875, + "learning_rate": 0.001663809404307445, + "loss": 0.1309, + "step": 28029 + }, + { + "epoch": 0.243313860122742, + "grad_norm": 0.115234375, + "learning_rate": 0.0016637862054238641, + "loss": 0.1074, + "step": 28030 + }, + { + "epoch": 0.24332254060294614, + "grad_norm": 0.46484375, + "learning_rate": 0.0016637630059237354, + "loss": 0.0928, + "step": 28031 + }, + { + "epoch": 0.2433312210831503, + "grad_norm": 0.053466796875, + "learning_rate": 0.0016637398058070838, + "loss": 0.0752, + "step": 28032 + }, + { + "epoch": 0.24333990156335447, + "grad_norm": 0.1513671875, + "learning_rate": 0.001663716605073935, + "loss": 0.1133, + "step": 28033 + }, + { + "epoch": 0.24334858204355864, + "grad_norm": 0.1767578125, + "learning_rate": 0.001663693403724314, + "loss": 0.1172, + "step": 28034 + }, + { + "epoch": 0.2433572625237628, + "grad_norm": 0.3203125, + "learning_rate": 0.001663670201758246, + "loss": 0.1162, + "step": 28035 + }, + { + "epoch": 0.24336594300396697, + "grad_norm": 0.69921875, + "learning_rate": 0.0016636469991757573, + "loss": 0.1201, + "step": 28036 + }, + { + "epoch": 0.24337462348417113, + "grad_norm": 0.10009765625, + "learning_rate": 0.0016636237959768724, + "loss": 0.1016, + "step": 28037 + }, + { + "epoch": 0.2433833039643753, + "grad_norm": 0.412109375, + "learning_rate": 0.0016636005921616172, + "loss": 0.106, + "step": 28038 + }, + { + "epoch": 0.24339198444457946, + "grad_norm": 0.294921875, + "learning_rate": 0.0016635773877300168, + "loss": 0.0938, + "step": 28039 + }, + { + "epoch": 0.24340066492478363, + "grad_norm": 0.259765625, + "learning_rate": 0.0016635541826820967, + "loss": 0.106, + "step": 28040 + }, + { + "epoch": 0.2434093454049878, + "grad_norm": 0.390625, + "learning_rate": 0.001663530977017882, + "loss": 0.124, + "step": 28041 + }, + { + "epoch": 0.24341802588519196, + "grad_norm": 0.345703125, + "learning_rate": 0.0016635077707373984, + "loss": 0.1196, + "step": 28042 + }, + { + "epoch": 0.24342670636539612, + "grad_norm": 0.244140625, + "learning_rate": 0.0016634845638406712, + "loss": 0.1064, + "step": 28043 + }, + { + "epoch": 0.2434353868456003, + "grad_norm": 0.416015625, + "learning_rate": 0.0016634613563277258, + "loss": 0.1084, + "step": 28044 + }, + { + "epoch": 0.24344406732580445, + "grad_norm": 0.181640625, + "learning_rate": 0.0016634381481985877, + "loss": 0.1079, + "step": 28045 + }, + { + "epoch": 0.24345274780600862, + "grad_norm": 0.19140625, + "learning_rate": 0.0016634149394532817, + "loss": 0.0854, + "step": 28046 + }, + { + "epoch": 0.24346142828621278, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016633917300918337, + "loss": 0.0981, + "step": 28047 + }, + { + "epoch": 0.24347010876641695, + "grad_norm": 0.306640625, + "learning_rate": 0.001663368520114269, + "loss": 0.0884, + "step": 28048 + }, + { + "epoch": 0.2434787892466211, + "grad_norm": 0.42578125, + "learning_rate": 0.0016633453095206128, + "loss": 0.1045, + "step": 28049 + }, + { + "epoch": 0.24348746972682528, + "grad_norm": 0.25, + "learning_rate": 0.0016633220983108907, + "loss": 0.1035, + "step": 28050 + }, + { + "epoch": 0.24349615020702944, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016632988864851283, + "loss": 0.1133, + "step": 28051 + }, + { + "epoch": 0.2435048306872336, + "grad_norm": 0.107421875, + "learning_rate": 0.0016632756740433505, + "loss": 0.1084, + "step": 28052 + }, + { + "epoch": 0.24351351116743777, + "grad_norm": 0.087890625, + "learning_rate": 0.001663252460985583, + "loss": 0.1128, + "step": 28053 + }, + { + "epoch": 0.24352219164764194, + "grad_norm": 0.5703125, + "learning_rate": 0.0016632292473118511, + "loss": 0.0854, + "step": 28054 + }, + { + "epoch": 0.2435308721278461, + "grad_norm": 0.10595703125, + "learning_rate": 0.0016632060330221798, + "loss": 0.1582, + "step": 28055 + }, + { + "epoch": 0.24353955260805027, + "grad_norm": 0.1357421875, + "learning_rate": 0.0016631828181165951, + "loss": 0.0776, + "step": 28056 + }, + { + "epoch": 0.24354823308825443, + "grad_norm": 0.375, + "learning_rate": 0.001663159602595122, + "loss": 0.1289, + "step": 28057 + }, + { + "epoch": 0.2435569135684586, + "grad_norm": 0.431640625, + "learning_rate": 0.0016631363864577865, + "loss": 0.1035, + "step": 28058 + }, + { + "epoch": 0.24356559404866276, + "grad_norm": 0.14453125, + "learning_rate": 0.001663113169704613, + "loss": 0.1211, + "step": 28059 + }, + { + "epoch": 0.24357427452886693, + "grad_norm": 0.0791015625, + "learning_rate": 0.0016630899523356276, + "loss": 0.0903, + "step": 28060 + }, + { + "epoch": 0.2435829550090711, + "grad_norm": 0.435546875, + "learning_rate": 0.0016630667343508552, + "loss": 0.1064, + "step": 28061 + }, + { + "epoch": 0.24359163548927526, + "grad_norm": 0.09814453125, + "learning_rate": 0.001663043515750322, + "loss": 0.0996, + "step": 28062 + }, + { + "epoch": 0.24360031596947943, + "grad_norm": 0.60546875, + "learning_rate": 0.0016630202965340524, + "loss": 0.1138, + "step": 28063 + }, + { + "epoch": 0.2436089964496836, + "grad_norm": 0.115234375, + "learning_rate": 0.0016629970767020723, + "loss": 0.0796, + "step": 28064 + }, + { + "epoch": 0.24361767692988776, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016629738562544073, + "loss": 0.1162, + "step": 28065 + }, + { + "epoch": 0.24362635741009192, + "grad_norm": 0.259765625, + "learning_rate": 0.0016629506351910823, + "loss": 0.0776, + "step": 28066 + }, + { + "epoch": 0.24363503789029609, + "grad_norm": 0.1552734375, + "learning_rate": 0.0016629274135121232, + "loss": 0.1025, + "step": 28067 + }, + { + "epoch": 0.24364371837050025, + "grad_norm": 0.365234375, + "learning_rate": 0.001662904191217555, + "loss": 0.1816, + "step": 28068 + }, + { + "epoch": 0.24365239885070442, + "grad_norm": 0.6171875, + "learning_rate": 0.0016628809683074033, + "loss": 0.1436, + "step": 28069 + }, + { + "epoch": 0.24366107933090858, + "grad_norm": 0.1064453125, + "learning_rate": 0.0016628577447816936, + "loss": 0.1006, + "step": 28070 + }, + { + "epoch": 0.24366975981111275, + "grad_norm": 0.08984375, + "learning_rate": 0.0016628345206404506, + "loss": 0.0645, + "step": 28071 + }, + { + "epoch": 0.2436784402913169, + "grad_norm": 0.283203125, + "learning_rate": 0.0016628112958837005, + "loss": 0.1006, + "step": 28072 + }, + { + "epoch": 0.24368712077152108, + "grad_norm": 0.3515625, + "learning_rate": 0.0016627880705114687, + "loss": 0.1514, + "step": 28073 + }, + { + "epoch": 0.24369580125172524, + "grad_norm": 0.6328125, + "learning_rate": 0.00166276484452378, + "loss": 0.1387, + "step": 28074 + }, + { + "epoch": 0.2437044817319294, + "grad_norm": 0.61328125, + "learning_rate": 0.0016627416179206603, + "loss": 0.0977, + "step": 28075 + }, + { + "epoch": 0.24371316221213357, + "grad_norm": 0.2314453125, + "learning_rate": 0.0016627183907021348, + "loss": 0.1084, + "step": 28076 + }, + { + "epoch": 0.24372184269233774, + "grad_norm": 0.578125, + "learning_rate": 0.0016626951628682285, + "loss": 0.0889, + "step": 28077 + }, + { + "epoch": 0.2437305231725419, + "grad_norm": 0.11767578125, + "learning_rate": 0.0016626719344189679, + "loss": 0.1006, + "step": 28078 + }, + { + "epoch": 0.24373920365274607, + "grad_norm": 1.0078125, + "learning_rate": 0.0016626487053543774, + "loss": 0.1855, + "step": 28079 + }, + { + "epoch": 0.24374788413295023, + "grad_norm": 0.140625, + "learning_rate": 0.0016626254756744827, + "loss": 0.1177, + "step": 28080 + }, + { + "epoch": 0.2437565646131544, + "grad_norm": 0.328125, + "learning_rate": 0.0016626022453793093, + "loss": 0.1367, + "step": 28081 + }, + { + "epoch": 0.24376524509335856, + "grad_norm": 0.9375, + "learning_rate": 0.0016625790144688824, + "loss": 0.0952, + "step": 28082 + }, + { + "epoch": 0.24377392557356273, + "grad_norm": 0.416015625, + "learning_rate": 0.0016625557829432279, + "loss": 0.0835, + "step": 28083 + }, + { + "epoch": 0.2437826060537669, + "grad_norm": 0.171875, + "learning_rate": 0.0016625325508023703, + "loss": 0.127, + "step": 28084 + }, + { + "epoch": 0.24379128653397106, + "grad_norm": 0.1455078125, + "learning_rate": 0.001662509318046336, + "loss": 0.0996, + "step": 28085 + }, + { + "epoch": 0.24379996701417522, + "grad_norm": 0.73828125, + "learning_rate": 0.0016624860846751496, + "loss": 0.1318, + "step": 28086 + }, + { + "epoch": 0.2438086474943794, + "grad_norm": 0.07763671875, + "learning_rate": 0.001662462850688837, + "loss": 0.084, + "step": 28087 + }, + { + "epoch": 0.24381732797458355, + "grad_norm": 0.23046875, + "learning_rate": 0.001662439616087424, + "loss": 0.0869, + "step": 28088 + }, + { + "epoch": 0.24382600845478772, + "grad_norm": 0.0810546875, + "learning_rate": 0.001662416380870935, + "loss": 0.1045, + "step": 28089 + }, + { + "epoch": 0.24383468893499188, + "grad_norm": 0.443359375, + "learning_rate": 0.001662393145039396, + "loss": 0.1309, + "step": 28090 + }, + { + "epoch": 0.24384336941519605, + "grad_norm": 0.115234375, + "learning_rate": 0.0016623699085928322, + "loss": 0.0894, + "step": 28091 + }, + { + "epoch": 0.2438520498954002, + "grad_norm": 0.08349609375, + "learning_rate": 0.001662346671531269, + "loss": 0.1016, + "step": 28092 + }, + { + "epoch": 0.24386073037560438, + "grad_norm": 0.263671875, + "learning_rate": 0.0016623234338547321, + "loss": 0.0996, + "step": 28093 + }, + { + "epoch": 0.24386941085580854, + "grad_norm": 0.328125, + "learning_rate": 0.0016623001955632466, + "loss": 0.1289, + "step": 28094 + }, + { + "epoch": 0.2438780913360127, + "grad_norm": 0.50390625, + "learning_rate": 0.0016622769566568382, + "loss": 0.1001, + "step": 28095 + }, + { + "epoch": 0.24388677181621687, + "grad_norm": 0.11474609375, + "learning_rate": 0.001662253717135532, + "loss": 0.1523, + "step": 28096 + }, + { + "epoch": 0.24389545229642104, + "grad_norm": 0.333984375, + "learning_rate": 0.001662230476999354, + "loss": 0.1045, + "step": 28097 + }, + { + "epoch": 0.2439041327766252, + "grad_norm": 0.6640625, + "learning_rate": 0.001662207236248329, + "loss": 0.1172, + "step": 28098 + }, + { + "epoch": 0.24391281325682937, + "grad_norm": 0.2197265625, + "learning_rate": 0.0016621839948824826, + "loss": 0.1025, + "step": 28099 + }, + { + "epoch": 0.24392149373703353, + "grad_norm": 0.1474609375, + "learning_rate": 0.00166216075290184, + "loss": 0.1025, + "step": 28100 + }, + { + "epoch": 0.2439301742172377, + "grad_norm": 0.1103515625, + "learning_rate": 0.001662137510306427, + "loss": 0.0898, + "step": 28101 + }, + { + "epoch": 0.24393885469744186, + "grad_norm": 0.31640625, + "learning_rate": 0.001662114267096269, + "loss": 0.0879, + "step": 28102 + }, + { + "epoch": 0.24394753517764603, + "grad_norm": 0.44921875, + "learning_rate": 0.0016620910232713912, + "loss": 0.085, + "step": 28103 + }, + { + "epoch": 0.2439562156578502, + "grad_norm": 0.2373046875, + "learning_rate": 0.001662067778831819, + "loss": 0.1113, + "step": 28104 + }, + { + "epoch": 0.24396489613805436, + "grad_norm": 0.3515625, + "learning_rate": 0.0016620445337775782, + "loss": 0.1006, + "step": 28105 + }, + { + "epoch": 0.24397357661825853, + "grad_norm": 0.1083984375, + "learning_rate": 0.0016620212881086939, + "loss": 0.1094, + "step": 28106 + }, + { + "epoch": 0.2439822570984627, + "grad_norm": 0.9609375, + "learning_rate": 0.0016619980418251911, + "loss": 0.0864, + "step": 28107 + }, + { + "epoch": 0.24399093757866686, + "grad_norm": 0.7265625, + "learning_rate": 0.001661974794927096, + "loss": 0.1143, + "step": 28108 + }, + { + "epoch": 0.24399961805887102, + "grad_norm": 0.466796875, + "learning_rate": 0.0016619515474144337, + "loss": 0.0898, + "step": 28109 + }, + { + "epoch": 0.24400829853907519, + "grad_norm": 0.89453125, + "learning_rate": 0.0016619282992872298, + "loss": 0.1377, + "step": 28110 + }, + { + "epoch": 0.24401697901927935, + "grad_norm": 0.16015625, + "learning_rate": 0.0016619050505455092, + "loss": 0.0801, + "step": 28111 + }, + { + "epoch": 0.24402565949948352, + "grad_norm": 0.369140625, + "learning_rate": 0.001661881801189298, + "loss": 0.125, + "step": 28112 + }, + { + "epoch": 0.24403433997968768, + "grad_norm": 0.388671875, + "learning_rate": 0.001661858551218621, + "loss": 0.1299, + "step": 28113 + }, + { + "epoch": 0.24404302045989185, + "grad_norm": 0.1162109375, + "learning_rate": 0.0016618353006335042, + "loss": 0.0913, + "step": 28114 + }, + { + "epoch": 0.244051700940096, + "grad_norm": 0.5390625, + "learning_rate": 0.001661812049433973, + "loss": 0.0874, + "step": 28115 + }, + { + "epoch": 0.24406038142030018, + "grad_norm": 0.1328125, + "learning_rate": 0.0016617887976200518, + "loss": 0.1221, + "step": 28116 + }, + { + "epoch": 0.24406906190050434, + "grad_norm": 0.52734375, + "learning_rate": 0.0016617655451917673, + "loss": 0.1396, + "step": 28117 + }, + { + "epoch": 0.2440777423807085, + "grad_norm": 0.10400390625, + "learning_rate": 0.0016617422921491446, + "loss": 0.0957, + "step": 28118 + }, + { + "epoch": 0.24408642286091267, + "grad_norm": 0.20703125, + "learning_rate": 0.0016617190384922086, + "loss": 0.1357, + "step": 28119 + }, + { + "epoch": 0.24409510334111684, + "grad_norm": 0.154296875, + "learning_rate": 0.0016616957842209852, + "loss": 0.1309, + "step": 28120 + }, + { + "epoch": 0.244103783821321, + "grad_norm": 0.10400390625, + "learning_rate": 0.0016616725293355, + "loss": 0.0938, + "step": 28121 + }, + { + "epoch": 0.24411246430152517, + "grad_norm": 0.07373046875, + "learning_rate": 0.0016616492738357777, + "loss": 0.103, + "step": 28122 + }, + { + "epoch": 0.24412114478172933, + "grad_norm": 0.57421875, + "learning_rate": 0.0016616260177218446, + "loss": 0.1025, + "step": 28123 + }, + { + "epoch": 0.2441298252619335, + "grad_norm": 0.326171875, + "learning_rate": 0.0016616027609937254, + "loss": 0.0898, + "step": 28124 + }, + { + "epoch": 0.24413850574213766, + "grad_norm": 0.09765625, + "learning_rate": 0.0016615795036514461, + "loss": 0.062, + "step": 28125 + }, + { + "epoch": 0.24414718622234183, + "grad_norm": 0.11767578125, + "learning_rate": 0.001661556245695032, + "loss": 0.0815, + "step": 28126 + }, + { + "epoch": 0.244155866702546, + "grad_norm": 0.10595703125, + "learning_rate": 0.001661532987124508, + "loss": 0.0811, + "step": 28127 + }, + { + "epoch": 0.24416454718275016, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016615097279399004, + "loss": 0.084, + "step": 28128 + }, + { + "epoch": 0.24417322766295432, + "grad_norm": 0.578125, + "learning_rate": 0.001661486468141234, + "loss": 0.125, + "step": 28129 + }, + { + "epoch": 0.2441819081431585, + "grad_norm": 0.671875, + "learning_rate": 0.0016614632077285344, + "loss": 0.1289, + "step": 28130 + }, + { + "epoch": 0.24419058862336265, + "grad_norm": 0.44140625, + "learning_rate": 0.0016614399467018274, + "loss": 0.0879, + "step": 28131 + }, + { + "epoch": 0.24419926910356682, + "grad_norm": 0.177734375, + "learning_rate": 0.0016614166850611376, + "loss": 0.1309, + "step": 28132 + }, + { + "epoch": 0.24420794958377098, + "grad_norm": 0.296875, + "learning_rate": 0.0016613934228064911, + "loss": 0.1045, + "step": 28133 + }, + { + "epoch": 0.24421663006397515, + "grad_norm": 0.7734375, + "learning_rate": 0.0016613701599379135, + "loss": 0.0645, + "step": 28134 + }, + { + "epoch": 0.2442253105441793, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016613468964554294, + "loss": 0.1133, + "step": 28135 + }, + { + "epoch": 0.24423399102438348, + "grad_norm": 0.54296875, + "learning_rate": 0.0016613236323590654, + "loss": 0.0684, + "step": 28136 + }, + { + "epoch": 0.24424267150458764, + "grad_norm": 0.60546875, + "learning_rate": 0.0016613003676488457, + "loss": 0.1172, + "step": 28137 + }, + { + "epoch": 0.2442513519847918, + "grad_norm": 0.462890625, + "learning_rate": 0.0016612771023247965, + "loss": 0.127, + "step": 28138 + }, + { + "epoch": 0.24426003246499597, + "grad_norm": 0.5234375, + "learning_rate": 0.0016612538363869435, + "loss": 0.1201, + "step": 28139 + }, + { + "epoch": 0.24426871294520014, + "grad_norm": 0.19140625, + "learning_rate": 0.0016612305698353114, + "loss": 0.0996, + "step": 28140 + }, + { + "epoch": 0.2442773934254043, + "grad_norm": 0.17578125, + "learning_rate": 0.0016612073026699263, + "loss": 0.1182, + "step": 28141 + }, + { + "epoch": 0.24428607390560847, + "grad_norm": 0.408203125, + "learning_rate": 0.001661184034890813, + "loss": 0.0737, + "step": 28142 + }, + { + "epoch": 0.24429475438581263, + "grad_norm": 0.185546875, + "learning_rate": 0.001661160766497997, + "loss": 0.1279, + "step": 28143 + }, + { + "epoch": 0.2443034348660168, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016611374974915046, + "loss": 0.0938, + "step": 28144 + }, + { + "epoch": 0.24431211534622096, + "grad_norm": 0.462890625, + "learning_rate": 0.0016611142278713602, + "loss": 0.105, + "step": 28145 + }, + { + "epoch": 0.24432079582642513, + "grad_norm": 0.11376953125, + "learning_rate": 0.0016610909576375902, + "loss": 0.0791, + "step": 28146 + }, + { + "epoch": 0.2443294763066293, + "grad_norm": 0.390625, + "learning_rate": 0.0016610676867902191, + "loss": 0.1201, + "step": 28147 + }, + { + "epoch": 0.24433815678683346, + "grad_norm": 0.17578125, + "learning_rate": 0.001661044415329273, + "loss": 0.1074, + "step": 28148 + }, + { + "epoch": 0.24434683726703763, + "grad_norm": 0.349609375, + "learning_rate": 0.0016610211432547773, + "loss": 0.1201, + "step": 28149 + }, + { + "epoch": 0.2443555177472418, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016609978705667567, + "loss": 0.0864, + "step": 28150 + }, + { + "epoch": 0.24436419822744596, + "grad_norm": 0.765625, + "learning_rate": 0.001660974597265238, + "loss": 0.0991, + "step": 28151 + }, + { + "epoch": 0.24437287870765012, + "grad_norm": 0.392578125, + "learning_rate": 0.0016609513233502456, + "loss": 0.0957, + "step": 28152 + }, + { + "epoch": 0.24438155918785429, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016609280488218052, + "loss": 0.1279, + "step": 28153 + }, + { + "epoch": 0.24439023966805842, + "grad_norm": 0.2490234375, + "learning_rate": 0.0016609047736799425, + "loss": 0.1016, + "step": 28154 + }, + { + "epoch": 0.2443989201482626, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016608814979246825, + "loss": 0.1245, + "step": 28155 + }, + { + "epoch": 0.24440760062846675, + "grad_norm": 0.373046875, + "learning_rate": 0.0016608582215560514, + "loss": 0.0972, + "step": 28156 + }, + { + "epoch": 0.24441628110867092, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016608349445740736, + "loss": 0.1025, + "step": 28157 + }, + { + "epoch": 0.24442496158887508, + "grad_norm": 0.10546875, + "learning_rate": 0.0016608116669787757, + "loss": 0.1074, + "step": 28158 + }, + { + "epoch": 0.24443364206907925, + "grad_norm": 0.10888671875, + "learning_rate": 0.001660788388770182, + "loss": 0.0957, + "step": 28159 + }, + { + "epoch": 0.24444232254928341, + "grad_norm": 0.115234375, + "learning_rate": 0.001660765109948319, + "loss": 0.0791, + "step": 28160 + }, + { + "epoch": 0.24445100302948758, + "grad_norm": 0.19921875, + "learning_rate": 0.0016607418305132117, + "loss": 0.1348, + "step": 28161 + }, + { + "epoch": 0.24445968350969174, + "grad_norm": 0.328125, + "learning_rate": 0.001660718550464885, + "loss": 0.1377, + "step": 28162 + }, + { + "epoch": 0.2444683639898959, + "grad_norm": 0.10302734375, + "learning_rate": 0.0016606952698033654, + "loss": 0.1377, + "step": 28163 + }, + { + "epoch": 0.24447704447010007, + "grad_norm": 0.27734375, + "learning_rate": 0.0016606719885286778, + "loss": 0.0737, + "step": 28164 + }, + { + "epoch": 0.24448572495030424, + "grad_norm": 1.0859375, + "learning_rate": 0.0016606487066408479, + "loss": 0.1133, + "step": 28165 + }, + { + "epoch": 0.2444944054305084, + "grad_norm": 0.359375, + "learning_rate": 0.0016606254241399005, + "loss": 0.1123, + "step": 28166 + }, + { + "epoch": 0.24450308591071257, + "grad_norm": 0.6171875, + "learning_rate": 0.0016606021410258622, + "loss": 0.104, + "step": 28167 + }, + { + "epoch": 0.24451176639091673, + "grad_norm": 0.384765625, + "learning_rate": 0.0016605788572987572, + "loss": 0.1055, + "step": 28168 + }, + { + "epoch": 0.2445204468711209, + "grad_norm": 0.376953125, + "learning_rate": 0.001660555572958612, + "loss": 0.126, + "step": 28169 + }, + { + "epoch": 0.24452912735132507, + "grad_norm": 0.490234375, + "learning_rate": 0.0016605322880054515, + "loss": 0.0996, + "step": 28170 + }, + { + "epoch": 0.24453780783152923, + "grad_norm": 0.2021484375, + "learning_rate": 0.0016605090024393013, + "loss": 0.0664, + "step": 28171 + }, + { + "epoch": 0.2445464883117334, + "grad_norm": 0.25390625, + "learning_rate": 0.0016604857162601868, + "loss": 0.1562, + "step": 28172 + }, + { + "epoch": 0.24455516879193756, + "grad_norm": 0.2099609375, + "learning_rate": 0.0016604624294681337, + "loss": 0.1055, + "step": 28173 + }, + { + "epoch": 0.24456384927214173, + "grad_norm": 0.08447265625, + "learning_rate": 0.0016604391420631673, + "loss": 0.1016, + "step": 28174 + }, + { + "epoch": 0.2445725297523459, + "grad_norm": 0.279296875, + "learning_rate": 0.0016604158540453131, + "loss": 0.0796, + "step": 28175 + }, + { + "epoch": 0.24458121023255006, + "grad_norm": 0.59765625, + "learning_rate": 0.0016603925654145966, + "loss": 0.1191, + "step": 28176 + }, + { + "epoch": 0.24458989071275422, + "grad_norm": 0.3046875, + "learning_rate": 0.001660369276171043, + "loss": 0.0889, + "step": 28177 + }, + { + "epoch": 0.2445985711929584, + "grad_norm": 0.51953125, + "learning_rate": 0.0016603459863146779, + "loss": 0.0815, + "step": 28178 + }, + { + "epoch": 0.24460725167316255, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016603226958455273, + "loss": 0.0762, + "step": 28179 + }, + { + "epoch": 0.24461593215336672, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016602994047636157, + "loss": 0.1123, + "step": 28180 + }, + { + "epoch": 0.24462461263357088, + "grad_norm": 0.0859375, + "learning_rate": 0.0016602761130689694, + "loss": 0.0854, + "step": 28181 + }, + { + "epoch": 0.24463329311377505, + "grad_norm": 0.359375, + "learning_rate": 0.0016602528207616137, + "loss": 0.1074, + "step": 28182 + }, + { + "epoch": 0.2446419735939792, + "grad_norm": 0.95703125, + "learning_rate": 0.0016602295278415734, + "loss": 0.1235, + "step": 28183 + }, + { + "epoch": 0.24465065407418338, + "grad_norm": 0.453125, + "learning_rate": 0.0016602062343088748, + "loss": 0.1191, + "step": 28184 + }, + { + "epoch": 0.24465933455438754, + "grad_norm": 0.48046875, + "learning_rate": 0.0016601829401635432, + "loss": 0.0957, + "step": 28185 + }, + { + "epoch": 0.2446680150345917, + "grad_norm": 0.166015625, + "learning_rate": 0.0016601596454056038, + "loss": 0.0957, + "step": 28186 + }, + { + "epoch": 0.24467669551479587, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016601363500350823, + "loss": 0.1143, + "step": 28187 + }, + { + "epoch": 0.24468537599500004, + "grad_norm": 0.08203125, + "learning_rate": 0.0016601130540520041, + "loss": 0.1055, + "step": 28188 + }, + { + "epoch": 0.2446940564752042, + "grad_norm": 0.07958984375, + "learning_rate": 0.0016600897574563947, + "loss": 0.0962, + "step": 28189 + }, + { + "epoch": 0.24470273695540837, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016600664602482796, + "loss": 0.0996, + "step": 28190 + }, + { + "epoch": 0.24471141743561253, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016600431624276839, + "loss": 0.0684, + "step": 28191 + }, + { + "epoch": 0.2447200979158167, + "grad_norm": 0.10205078125, + "learning_rate": 0.001660019863994634, + "loss": 0.1064, + "step": 28192 + }, + { + "epoch": 0.24472877839602086, + "grad_norm": 0.392578125, + "learning_rate": 0.0016599965649491543, + "loss": 0.1025, + "step": 28193 + }, + { + "epoch": 0.24473745887622503, + "grad_norm": 0.451171875, + "learning_rate": 0.0016599732652912709, + "loss": 0.103, + "step": 28194 + }, + { + "epoch": 0.2447461393564292, + "grad_norm": 0.189453125, + "learning_rate": 0.0016599499650210093, + "loss": 0.124, + "step": 28195 + }, + { + "epoch": 0.24475481983663336, + "grad_norm": 0.181640625, + "learning_rate": 0.0016599266641383946, + "loss": 0.1001, + "step": 28196 + }, + { + "epoch": 0.24476350031683752, + "grad_norm": 0.263671875, + "learning_rate": 0.0016599033626434527, + "loss": 0.0933, + "step": 28197 + }, + { + "epoch": 0.2447721807970417, + "grad_norm": 0.240234375, + "learning_rate": 0.001659880060536209, + "loss": 0.1455, + "step": 28198 + }, + { + "epoch": 0.24478086127724585, + "grad_norm": 0.126953125, + "learning_rate": 0.0016598567578166884, + "loss": 0.0593, + "step": 28199 + }, + { + "epoch": 0.24478954175745002, + "grad_norm": 0.416015625, + "learning_rate": 0.0016598334544849175, + "loss": 0.0933, + "step": 28200 + }, + { + "epoch": 0.24479822223765418, + "grad_norm": 0.2333984375, + "learning_rate": 0.0016598101505409207, + "loss": 0.1094, + "step": 28201 + }, + { + "epoch": 0.24480690271785835, + "grad_norm": 0.97265625, + "learning_rate": 0.0016597868459847238, + "loss": 0.1113, + "step": 28202 + }, + { + "epoch": 0.24481558319806251, + "grad_norm": 0.259765625, + "learning_rate": 0.001659763540816353, + "loss": 0.1338, + "step": 28203 + }, + { + "epoch": 0.24482426367826668, + "grad_norm": 0.060302734375, + "learning_rate": 0.0016597402350358329, + "loss": 0.0947, + "step": 28204 + }, + { + "epoch": 0.24483294415847084, + "grad_norm": 0.66015625, + "learning_rate": 0.001659716928643189, + "loss": 0.1143, + "step": 28205 + }, + { + "epoch": 0.244841624638675, + "grad_norm": 0.083984375, + "learning_rate": 0.0016596936216384474, + "loss": 0.0859, + "step": 28206 + }, + { + "epoch": 0.24485030511887917, + "grad_norm": 0.42578125, + "learning_rate": 0.0016596703140216332, + "loss": 0.1348, + "step": 28207 + }, + { + "epoch": 0.24485898559908334, + "grad_norm": 0.12890625, + "learning_rate": 0.0016596470057927718, + "loss": 0.1084, + "step": 28208 + }, + { + "epoch": 0.2448676660792875, + "grad_norm": 0.1728515625, + "learning_rate": 0.001659623696951889, + "loss": 0.1084, + "step": 28209 + }, + { + "epoch": 0.24487634655949167, + "grad_norm": 0.2294921875, + "learning_rate": 0.0016596003874990103, + "loss": 0.1445, + "step": 28210 + }, + { + "epoch": 0.24488502703969584, + "grad_norm": 0.546875, + "learning_rate": 0.0016595770774341608, + "loss": 0.0898, + "step": 28211 + }, + { + "epoch": 0.2448937075199, + "grad_norm": 0.1982421875, + "learning_rate": 0.0016595537667573664, + "loss": 0.1172, + "step": 28212 + }, + { + "epoch": 0.24490238800010417, + "grad_norm": 0.5078125, + "learning_rate": 0.0016595304554686521, + "loss": 0.1084, + "step": 28213 + }, + { + "epoch": 0.24491106848030833, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016595071435680437, + "loss": 0.0688, + "step": 28214 + }, + { + "epoch": 0.2449197489605125, + "grad_norm": 0.111328125, + "learning_rate": 0.001659483831055567, + "loss": 0.1426, + "step": 28215 + }, + { + "epoch": 0.24492842944071666, + "grad_norm": 0.07666015625, + "learning_rate": 0.0016594605179312471, + "loss": 0.1035, + "step": 28216 + }, + { + "epoch": 0.24493710992092083, + "grad_norm": 0.162109375, + "learning_rate": 0.0016594372041951095, + "loss": 0.1016, + "step": 28217 + }, + { + "epoch": 0.244945790401125, + "grad_norm": 0.1513671875, + "learning_rate": 0.00165941388984718, + "loss": 0.125, + "step": 28218 + }, + { + "epoch": 0.24495447088132916, + "grad_norm": 0.51953125, + "learning_rate": 0.0016593905748874833, + "loss": 0.0723, + "step": 28219 + }, + { + "epoch": 0.24496315136153332, + "grad_norm": 0.349609375, + "learning_rate": 0.001659367259316046, + "loss": 0.0801, + "step": 28220 + }, + { + "epoch": 0.2449718318417375, + "grad_norm": 0.326171875, + "learning_rate": 0.0016593439431328932, + "loss": 0.1289, + "step": 28221 + }, + { + "epoch": 0.24498051232194165, + "grad_norm": 0.138671875, + "learning_rate": 0.0016593206263380497, + "loss": 0.0938, + "step": 28222 + }, + { + "epoch": 0.24498919280214582, + "grad_norm": 0.28515625, + "learning_rate": 0.001659297308931542, + "loss": 0.0908, + "step": 28223 + }, + { + "epoch": 0.24499787328234998, + "grad_norm": 0.140625, + "learning_rate": 0.001659273990913395, + "loss": 0.1084, + "step": 28224 + }, + { + "epoch": 0.24500655376255415, + "grad_norm": 0.65625, + "learning_rate": 0.0016592506722836343, + "loss": 0.1299, + "step": 28225 + }, + { + "epoch": 0.2450152342427583, + "grad_norm": 0.423828125, + "learning_rate": 0.0016592273530422856, + "loss": 0.0703, + "step": 28226 + }, + { + "epoch": 0.24502391472296248, + "grad_norm": 0.39453125, + "learning_rate": 0.0016592040331893745, + "loss": 0.105, + "step": 28227 + }, + { + "epoch": 0.24503259520316664, + "grad_norm": 0.2158203125, + "learning_rate": 0.0016591807127249256, + "loss": 0.124, + "step": 28228 + }, + { + "epoch": 0.2450412756833708, + "grad_norm": 0.21875, + "learning_rate": 0.0016591573916489655, + "loss": 0.0933, + "step": 28229 + }, + { + "epoch": 0.24504995616357497, + "grad_norm": 0.376953125, + "learning_rate": 0.0016591340699615195, + "loss": 0.1191, + "step": 28230 + }, + { + "epoch": 0.24505863664377914, + "grad_norm": 0.1064453125, + "learning_rate": 0.0016591107476626127, + "loss": 0.1055, + "step": 28231 + }, + { + "epoch": 0.2450673171239833, + "grad_norm": 0.5234375, + "learning_rate": 0.0016590874247522707, + "loss": 0.1523, + "step": 28232 + }, + { + "epoch": 0.24507599760418747, + "grad_norm": 0.53125, + "learning_rate": 0.001659064101230519, + "loss": 0.0918, + "step": 28233 + }, + { + "epoch": 0.24508467808439163, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016590407770973835, + "loss": 0.0854, + "step": 28234 + }, + { + "epoch": 0.2450933585645958, + "grad_norm": 0.61328125, + "learning_rate": 0.0016590174523528891, + "loss": 0.1436, + "step": 28235 + }, + { + "epoch": 0.24510203904479996, + "grad_norm": 0.3359375, + "learning_rate": 0.001658994126997062, + "loss": 0.1045, + "step": 28236 + }, + { + "epoch": 0.24511071952500413, + "grad_norm": 0.349609375, + "learning_rate": 0.001658970801029927, + "loss": 0.1445, + "step": 28237 + }, + { + "epoch": 0.2451194000052083, + "grad_norm": 0.15625, + "learning_rate": 0.0016589474744515098, + "loss": 0.0952, + "step": 28238 + }, + { + "epoch": 0.24512808048541246, + "grad_norm": 0.1591796875, + "learning_rate": 0.0016589241472618362, + "loss": 0.0859, + "step": 28239 + }, + { + "epoch": 0.24513676096561662, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016589008194609317, + "loss": 0.1108, + "step": 28240 + }, + { + "epoch": 0.2451454414458208, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016588774910488216, + "loss": 0.082, + "step": 28241 + }, + { + "epoch": 0.24515412192602495, + "grad_norm": 0.7890625, + "learning_rate": 0.001658854162025531, + "loss": 0.1152, + "step": 28242 + }, + { + "epoch": 0.24516280240622912, + "grad_norm": 0.5859375, + "learning_rate": 0.0016588308323910864, + "loss": 0.1289, + "step": 28243 + }, + { + "epoch": 0.24517148288643328, + "grad_norm": 0.294921875, + "learning_rate": 0.0016588075021455127, + "loss": 0.0996, + "step": 28244 + }, + { + "epoch": 0.24518016336663745, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016587841712888357, + "loss": 0.0918, + "step": 28245 + }, + { + "epoch": 0.24518884384684161, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016587608398210803, + "loss": 0.0918, + "step": 28246 + }, + { + "epoch": 0.24519752432704578, + "grad_norm": 0.51953125, + "learning_rate": 0.001658737507742273, + "loss": 0.1182, + "step": 28247 + }, + { + "epoch": 0.24520620480724994, + "grad_norm": 0.25, + "learning_rate": 0.0016587141750524382, + "loss": 0.1216, + "step": 28248 + }, + { + "epoch": 0.2452148852874541, + "grad_norm": 0.142578125, + "learning_rate": 0.0016586908417516022, + "loss": 0.1055, + "step": 28249 + }, + { + "epoch": 0.24522356576765827, + "grad_norm": 0.14453125, + "learning_rate": 0.0016586675078397904, + "loss": 0.0718, + "step": 28250 + }, + { + "epoch": 0.24523224624786244, + "grad_norm": 0.2431640625, + "learning_rate": 0.0016586441733170282, + "loss": 0.1182, + "step": 28251 + }, + { + "epoch": 0.2452409267280666, + "grad_norm": 0.228515625, + "learning_rate": 0.001658620838183341, + "loss": 0.1621, + "step": 28252 + }, + { + "epoch": 0.24524960720827077, + "grad_norm": 0.625, + "learning_rate": 0.0016585975024387544, + "loss": 0.0923, + "step": 28253 + }, + { + "epoch": 0.24525828768847494, + "grad_norm": 0.83984375, + "learning_rate": 0.0016585741660832937, + "loss": 0.1543, + "step": 28254 + }, + { + "epoch": 0.2452669681686791, + "grad_norm": 0.359375, + "learning_rate": 0.0016585508291169853, + "loss": 0.0977, + "step": 28255 + }, + { + "epoch": 0.24527564864888327, + "grad_norm": 0.1669921875, + "learning_rate": 0.001658527491539854, + "loss": 0.0742, + "step": 28256 + }, + { + "epoch": 0.24528432912908743, + "grad_norm": 0.765625, + "learning_rate": 0.001658504153351925, + "loss": 0.1064, + "step": 28257 + }, + { + "epoch": 0.2452930096092916, + "grad_norm": 1.2421875, + "learning_rate": 0.0016584808145532247, + "loss": 0.1943, + "step": 28258 + }, + { + "epoch": 0.24530169008949576, + "grad_norm": 0.306640625, + "learning_rate": 0.0016584574751437777, + "loss": 0.0957, + "step": 28259 + }, + { + "epoch": 0.24531037056969993, + "grad_norm": 0.3515625, + "learning_rate": 0.0016584341351236105, + "loss": 0.1299, + "step": 28260 + }, + { + "epoch": 0.2453190510499041, + "grad_norm": 0.68359375, + "learning_rate": 0.001658410794492748, + "loss": 0.1348, + "step": 28261 + }, + { + "epoch": 0.24532773153010826, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016583874532512155, + "loss": 0.0762, + "step": 28262 + }, + { + "epoch": 0.24533641201031242, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016583641113990391, + "loss": 0.0957, + "step": 28263 + }, + { + "epoch": 0.2453450924905166, + "grad_norm": 0.32421875, + "learning_rate": 0.0016583407689362442, + "loss": 0.1514, + "step": 28264 + }, + { + "epoch": 0.24535377297072075, + "grad_norm": 0.361328125, + "learning_rate": 0.001658317425862856, + "loss": 0.1045, + "step": 28265 + }, + { + "epoch": 0.24536245345092492, + "grad_norm": 0.6953125, + "learning_rate": 0.0016582940821789004, + "loss": 0.1426, + "step": 28266 + }, + { + "epoch": 0.24537113393112908, + "grad_norm": 0.1416015625, + "learning_rate": 0.0016582707378844026, + "loss": 0.0957, + "step": 28267 + }, + { + "epoch": 0.24537981441133325, + "grad_norm": 0.498046875, + "learning_rate": 0.001658247392979389, + "loss": 0.1318, + "step": 28268 + }, + { + "epoch": 0.2453884948915374, + "grad_norm": 0.154296875, + "learning_rate": 0.0016582240474638838, + "loss": 0.0854, + "step": 28269 + }, + { + "epoch": 0.24539717537174158, + "grad_norm": 0.2890625, + "learning_rate": 0.0016582007013379132, + "loss": 0.1689, + "step": 28270 + }, + { + "epoch": 0.24540585585194574, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016581773546015027, + "loss": 0.0938, + "step": 28271 + }, + { + "epoch": 0.2454145363321499, + "grad_norm": 0.7421875, + "learning_rate": 0.001658154007254678, + "loss": 0.1328, + "step": 28272 + }, + { + "epoch": 0.24542321681235407, + "grad_norm": 0.177734375, + "learning_rate": 0.0016581306592974641, + "loss": 0.0981, + "step": 28273 + }, + { + "epoch": 0.24543189729255824, + "grad_norm": 0.56640625, + "learning_rate": 0.0016581073107298874, + "loss": 0.0952, + "step": 28274 + }, + { + "epoch": 0.2454405777727624, + "grad_norm": 0.08349609375, + "learning_rate": 0.0016580839615519727, + "loss": 0.0894, + "step": 28275 + }, + { + "epoch": 0.24544925825296657, + "grad_norm": 0.361328125, + "learning_rate": 0.0016580606117637458, + "loss": 0.25, + "step": 28276 + }, + { + "epoch": 0.2454579387331707, + "grad_norm": 0.31640625, + "learning_rate": 0.001658037261365232, + "loss": 0.1016, + "step": 28277 + }, + { + "epoch": 0.24546661921337487, + "grad_norm": 0.41015625, + "learning_rate": 0.0016580139103564574, + "loss": 0.1201, + "step": 28278 + }, + { + "epoch": 0.24547529969357904, + "grad_norm": 0.66796875, + "learning_rate": 0.001657990558737447, + "loss": 0.1104, + "step": 28279 + }, + { + "epoch": 0.2454839801737832, + "grad_norm": 0.890625, + "learning_rate": 0.0016579672065082263, + "loss": 0.127, + "step": 28280 + }, + { + "epoch": 0.24549266065398737, + "grad_norm": 0.1982421875, + "learning_rate": 0.0016579438536688212, + "loss": 0.1035, + "step": 28281 + }, + { + "epoch": 0.24550134113419153, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016579205002192573, + "loss": 0.1445, + "step": 28282 + }, + { + "epoch": 0.2455100216143957, + "grad_norm": 0.427734375, + "learning_rate": 0.0016578971461595595, + "loss": 0.0884, + "step": 28283 + }, + { + "epoch": 0.24551870209459986, + "grad_norm": 0.2197265625, + "learning_rate": 0.001657873791489754, + "loss": 0.1582, + "step": 28284 + }, + { + "epoch": 0.24552738257480403, + "grad_norm": 0.9921875, + "learning_rate": 0.0016578504362098662, + "loss": 0.125, + "step": 28285 + }, + { + "epoch": 0.2455360630550082, + "grad_norm": 0.1767578125, + "learning_rate": 0.0016578270803199215, + "loss": 0.0879, + "step": 28286 + }, + { + "epoch": 0.24554474353521236, + "grad_norm": 0.5859375, + "learning_rate": 0.0016578037238199452, + "loss": 0.1318, + "step": 28287 + }, + { + "epoch": 0.24555342401541652, + "grad_norm": 0.39453125, + "learning_rate": 0.0016577803667099636, + "loss": 0.1143, + "step": 28288 + }, + { + "epoch": 0.2455621044956207, + "grad_norm": 0.4140625, + "learning_rate": 0.0016577570089900014, + "loss": 0.0938, + "step": 28289 + }, + { + "epoch": 0.24557078497582485, + "grad_norm": 0.09375, + "learning_rate": 0.0016577336506600846, + "loss": 0.1357, + "step": 28290 + }, + { + "epoch": 0.24557946545602902, + "grad_norm": 0.357421875, + "learning_rate": 0.0016577102917202386, + "loss": 0.0874, + "step": 28291 + }, + { + "epoch": 0.24558814593623318, + "grad_norm": 0.162109375, + "learning_rate": 0.0016576869321704892, + "loss": 0.0986, + "step": 28292 + }, + { + "epoch": 0.24559682641643735, + "grad_norm": 0.478515625, + "learning_rate": 0.0016576635720108616, + "loss": 0.1143, + "step": 28293 + }, + { + "epoch": 0.2456055068966415, + "grad_norm": 0.15625, + "learning_rate": 0.0016576402112413817, + "loss": 0.1074, + "step": 28294 + }, + { + "epoch": 0.24561418737684568, + "grad_norm": 0.337890625, + "learning_rate": 0.0016576168498620748, + "loss": 0.1133, + "step": 28295 + }, + { + "epoch": 0.24562286785704984, + "grad_norm": 0.103515625, + "learning_rate": 0.0016575934878729664, + "loss": 0.0713, + "step": 28296 + }, + { + "epoch": 0.245631548337254, + "grad_norm": 0.291015625, + "learning_rate": 0.001657570125274082, + "loss": 0.0669, + "step": 28297 + }, + { + "epoch": 0.24564022881745817, + "grad_norm": 0.1005859375, + "learning_rate": 0.0016575467620654474, + "loss": 0.0938, + "step": 28298 + }, + { + "epoch": 0.24564890929766234, + "grad_norm": 0.68359375, + "learning_rate": 0.001657523398247088, + "loss": 0.1133, + "step": 28299 + }, + { + "epoch": 0.2456575897778665, + "grad_norm": 0.16796875, + "learning_rate": 0.0016575000338190295, + "loss": 0.1289, + "step": 28300 + }, + { + "epoch": 0.24566627025807067, + "grad_norm": 0.5, + "learning_rate": 0.0016574766687812974, + "loss": 0.0732, + "step": 28301 + }, + { + "epoch": 0.24567495073827483, + "grad_norm": 0.10986328125, + "learning_rate": 0.001657453303133917, + "loss": 0.084, + "step": 28302 + }, + { + "epoch": 0.245683631218479, + "grad_norm": 0.2451171875, + "learning_rate": 0.0016574299368769141, + "loss": 0.1279, + "step": 28303 + }, + { + "epoch": 0.24569231169868316, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016574065700103145, + "loss": 0.0908, + "step": 28304 + }, + { + "epoch": 0.24570099217888733, + "grad_norm": 0.125, + "learning_rate": 0.0016573832025341428, + "loss": 0.0889, + "step": 28305 + }, + { + "epoch": 0.2457096726590915, + "grad_norm": 0.08935546875, + "learning_rate": 0.0016573598344484257, + "loss": 0.0918, + "step": 28306 + }, + { + "epoch": 0.24571835313929566, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016573364657531881, + "loss": 0.0732, + "step": 28307 + }, + { + "epoch": 0.24572703361949982, + "grad_norm": 0.2099609375, + "learning_rate": 0.0016573130964484557, + "loss": 0.0825, + "step": 28308 + }, + { + "epoch": 0.245735714099704, + "grad_norm": 0.38671875, + "learning_rate": 0.0016572897265342541, + "loss": 0.1143, + "step": 28309 + }, + { + "epoch": 0.24574439457990815, + "grad_norm": 0.53515625, + "learning_rate": 0.001657266356010609, + "loss": 0.1123, + "step": 28310 + }, + { + "epoch": 0.24575307506011232, + "grad_norm": 0.076171875, + "learning_rate": 0.0016572429848775454, + "loss": 0.0957, + "step": 28311 + }, + { + "epoch": 0.24576175554031648, + "grad_norm": 0.115234375, + "learning_rate": 0.0016572196131350897, + "loss": 0.0972, + "step": 28312 + }, + { + "epoch": 0.24577043602052065, + "grad_norm": 0.16015625, + "learning_rate": 0.001657196240783267, + "loss": 0.1592, + "step": 28313 + }, + { + "epoch": 0.24577911650072481, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016571728678221025, + "loss": 0.0947, + "step": 28314 + }, + { + "epoch": 0.24578779698092898, + "grad_norm": 0.142578125, + "learning_rate": 0.0016571494942516222, + "loss": 0.1133, + "step": 28315 + }, + { + "epoch": 0.24579647746113314, + "grad_norm": 0.482421875, + "learning_rate": 0.0016571261200718516, + "loss": 0.1216, + "step": 28316 + }, + { + "epoch": 0.2458051579413373, + "grad_norm": 0.25390625, + "learning_rate": 0.0016571027452828163, + "loss": 0.1211, + "step": 28317 + }, + { + "epoch": 0.24581383842154148, + "grad_norm": 0.31640625, + "learning_rate": 0.001657079369884542, + "loss": 0.166, + "step": 28318 + }, + { + "epoch": 0.24582251890174564, + "grad_norm": 0.3828125, + "learning_rate": 0.0016570559938770537, + "loss": 0.1016, + "step": 28319 + }, + { + "epoch": 0.2458311993819498, + "grad_norm": 0.10205078125, + "learning_rate": 0.0016570326172603773, + "loss": 0.1045, + "step": 28320 + }, + { + "epoch": 0.24583987986215397, + "grad_norm": 3.28125, + "learning_rate": 0.0016570092400345388, + "loss": 0.1895, + "step": 28321 + }, + { + "epoch": 0.24584856034235814, + "grad_norm": 0.279296875, + "learning_rate": 0.0016569858621995632, + "loss": 0.0811, + "step": 28322 + }, + { + "epoch": 0.2458572408225623, + "grad_norm": 0.14453125, + "learning_rate": 0.001656962483755476, + "loss": 0.1689, + "step": 28323 + }, + { + "epoch": 0.24586592130276647, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016569391047023035, + "loss": 0.1504, + "step": 28324 + }, + { + "epoch": 0.24587460178297063, + "grad_norm": 0.5234375, + "learning_rate": 0.0016569157250400702, + "loss": 0.1416, + "step": 28325 + }, + { + "epoch": 0.2458832822631748, + "grad_norm": 0.14453125, + "learning_rate": 0.0016568923447688028, + "loss": 0.1133, + "step": 28326 + }, + { + "epoch": 0.24589196274337896, + "grad_norm": 0.109375, + "learning_rate": 0.0016568689638885258, + "loss": 0.1152, + "step": 28327 + }, + { + "epoch": 0.24590064322358313, + "grad_norm": 0.103515625, + "learning_rate": 0.0016568455823992654, + "loss": 0.0908, + "step": 28328 + }, + { + "epoch": 0.2459093237037873, + "grad_norm": 0.2294921875, + "learning_rate": 0.001656822200301047, + "loss": 0.1738, + "step": 28329 + }, + { + "epoch": 0.24591800418399146, + "grad_norm": 0.400390625, + "learning_rate": 0.0016567988175938964, + "loss": 0.0898, + "step": 28330 + }, + { + "epoch": 0.24592668466419562, + "grad_norm": 0.154296875, + "learning_rate": 0.001656775434277839, + "loss": 0.1211, + "step": 28331 + }, + { + "epoch": 0.2459353651443998, + "grad_norm": 0.19140625, + "learning_rate": 0.0016567520503529, + "loss": 0.1045, + "step": 28332 + }, + { + "epoch": 0.24594404562460395, + "grad_norm": 1.046875, + "learning_rate": 0.0016567286658191057, + "loss": 0.1328, + "step": 28333 + }, + { + "epoch": 0.24595272610480812, + "grad_norm": 0.470703125, + "learning_rate": 0.0016567052806764813, + "loss": 0.1152, + "step": 28334 + }, + { + "epoch": 0.24596140658501228, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016566818949250522, + "loss": 0.0874, + "step": 28335 + }, + { + "epoch": 0.24597008706521645, + "grad_norm": 0.8984375, + "learning_rate": 0.001656658508564844, + "loss": 0.104, + "step": 28336 + }, + { + "epoch": 0.2459787675454206, + "grad_norm": 0.421875, + "learning_rate": 0.0016566351215958825, + "loss": 0.1006, + "step": 28337 + }, + { + "epoch": 0.24598744802562478, + "grad_norm": 0.09716796875, + "learning_rate": 0.0016566117340181933, + "loss": 0.0869, + "step": 28338 + }, + { + "epoch": 0.24599612850582894, + "grad_norm": 0.318359375, + "learning_rate": 0.001656588345831802, + "loss": 0.0835, + "step": 28339 + }, + { + "epoch": 0.2460048089860331, + "grad_norm": 0.1884765625, + "learning_rate": 0.0016565649570367338, + "loss": 0.0747, + "step": 28340 + }, + { + "epoch": 0.24601348946623727, + "grad_norm": 0.11328125, + "learning_rate": 0.0016565415676330147, + "loss": 0.1104, + "step": 28341 + }, + { + "epoch": 0.24602216994644144, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016565181776206702, + "loss": 0.1055, + "step": 28342 + }, + { + "epoch": 0.2460308504266456, + "grad_norm": 0.19140625, + "learning_rate": 0.0016564947869997255, + "loss": 0.0903, + "step": 28343 + }, + { + "epoch": 0.24603953090684977, + "grad_norm": 0.49609375, + "learning_rate": 0.0016564713957702066, + "loss": 0.1338, + "step": 28344 + }, + { + "epoch": 0.24604821138705393, + "grad_norm": 0.150390625, + "learning_rate": 0.0016564480039321387, + "loss": 0.1152, + "step": 28345 + }, + { + "epoch": 0.2460568918672581, + "grad_norm": 0.2490234375, + "learning_rate": 0.0016564246114855478, + "loss": 0.127, + "step": 28346 + }, + { + "epoch": 0.24606557234746226, + "grad_norm": 0.11767578125, + "learning_rate": 0.0016564012184304595, + "loss": 0.1094, + "step": 28347 + }, + { + "epoch": 0.24607425282766643, + "grad_norm": 0.51171875, + "learning_rate": 0.001656377824766899, + "loss": 0.1182, + "step": 28348 + }, + { + "epoch": 0.2460829333078706, + "grad_norm": 0.68359375, + "learning_rate": 0.0016563544304948924, + "loss": 0.1562, + "step": 28349 + }, + { + "epoch": 0.24609161378807476, + "grad_norm": 0.2421875, + "learning_rate": 0.0016563310356144643, + "loss": 0.105, + "step": 28350 + }, + { + "epoch": 0.24610029426827892, + "grad_norm": 0.3046875, + "learning_rate": 0.0016563076401256416, + "loss": 0.125, + "step": 28351 + }, + { + "epoch": 0.2461089747484831, + "grad_norm": 0.123046875, + "learning_rate": 0.0016562842440284488, + "loss": 0.1055, + "step": 28352 + }, + { + "epoch": 0.24611765522868725, + "grad_norm": 0.0712890625, + "learning_rate": 0.001656260847322912, + "loss": 0.0889, + "step": 28353 + }, + { + "epoch": 0.24612633570889142, + "grad_norm": 0.08740234375, + "learning_rate": 0.0016562374500090568, + "loss": 0.0923, + "step": 28354 + }, + { + "epoch": 0.24613501618909558, + "grad_norm": 0.75, + "learning_rate": 0.0016562140520869084, + "loss": 0.1133, + "step": 28355 + }, + { + "epoch": 0.24614369666929975, + "grad_norm": 0.1240234375, + "learning_rate": 0.001656190653556493, + "loss": 0.1562, + "step": 28356 + }, + { + "epoch": 0.24615237714950391, + "grad_norm": 0.3125, + "learning_rate": 0.0016561672544178356, + "loss": 0.1094, + "step": 28357 + }, + { + "epoch": 0.24616105762970808, + "grad_norm": 0.12060546875, + "learning_rate": 0.0016561438546709624, + "loss": 0.0718, + "step": 28358 + }, + { + "epoch": 0.24616973810991225, + "grad_norm": 0.13671875, + "learning_rate": 0.0016561204543158983, + "loss": 0.0879, + "step": 28359 + }, + { + "epoch": 0.2461784185901164, + "grad_norm": 0.1875, + "learning_rate": 0.0016560970533526693, + "loss": 0.0977, + "step": 28360 + }, + { + "epoch": 0.24618709907032058, + "grad_norm": 1.0390625, + "learning_rate": 0.0016560736517813011, + "loss": 0.1299, + "step": 28361 + }, + { + "epoch": 0.24619577955052474, + "grad_norm": 0.08349609375, + "learning_rate": 0.0016560502496018189, + "loss": 0.1079, + "step": 28362 + }, + { + "epoch": 0.2462044600307289, + "grad_norm": 0.21484375, + "learning_rate": 0.0016560268468142486, + "loss": 0.0713, + "step": 28363 + }, + { + "epoch": 0.24621314051093307, + "grad_norm": 0.5390625, + "learning_rate": 0.0016560034434186157, + "loss": 0.0811, + "step": 28364 + }, + { + "epoch": 0.24622182099113724, + "grad_norm": 0.3671875, + "learning_rate": 0.0016559800394149457, + "loss": 0.104, + "step": 28365 + }, + { + "epoch": 0.2462305014713414, + "grad_norm": 0.0908203125, + "learning_rate": 0.0016559566348032643, + "loss": 0.1016, + "step": 28366 + }, + { + "epoch": 0.24623918195154557, + "grad_norm": 0.283203125, + "learning_rate": 0.0016559332295835972, + "loss": 0.0869, + "step": 28367 + }, + { + "epoch": 0.24624786243174973, + "grad_norm": 0.4921875, + "learning_rate": 0.0016559098237559696, + "loss": 0.1021, + "step": 28368 + }, + { + "epoch": 0.2462565429119539, + "grad_norm": 0.5546875, + "learning_rate": 0.0016558864173204077, + "loss": 0.1099, + "step": 28369 + }, + { + "epoch": 0.24626522339215806, + "grad_norm": 0.126953125, + "learning_rate": 0.0016558630102769365, + "loss": 0.0889, + "step": 28370 + }, + { + "epoch": 0.24627390387236223, + "grad_norm": 0.294921875, + "learning_rate": 0.001655839602625582, + "loss": 0.1348, + "step": 28371 + }, + { + "epoch": 0.2462825843525664, + "grad_norm": 0.31640625, + "learning_rate": 0.0016558161943663696, + "loss": 0.1406, + "step": 28372 + }, + { + "epoch": 0.24629126483277056, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016557927854993252, + "loss": 0.1084, + "step": 28373 + }, + { + "epoch": 0.24629994531297472, + "grad_norm": 0.63671875, + "learning_rate": 0.0016557693760244738, + "loss": 0.1133, + "step": 28374 + }, + { + "epoch": 0.2463086257931789, + "grad_norm": 0.25, + "learning_rate": 0.0016557459659418414, + "loss": 0.0923, + "step": 28375 + }, + { + "epoch": 0.24631730627338305, + "grad_norm": 0.51171875, + "learning_rate": 0.0016557225552514539, + "loss": 0.1455, + "step": 28376 + }, + { + "epoch": 0.24632598675358722, + "grad_norm": 0.58984375, + "learning_rate": 0.0016556991439533363, + "loss": 0.0986, + "step": 28377 + }, + { + "epoch": 0.24633466723379138, + "grad_norm": 0.1708984375, + "learning_rate": 0.0016556757320475147, + "loss": 0.1211, + "step": 28378 + }, + { + "epoch": 0.24634334771399555, + "grad_norm": 0.5625, + "learning_rate": 0.0016556523195340141, + "loss": 0.1182, + "step": 28379 + }, + { + "epoch": 0.2463520281941997, + "grad_norm": 0.2392578125, + "learning_rate": 0.0016556289064128605, + "loss": 0.1221, + "step": 28380 + }, + { + "epoch": 0.24636070867440388, + "grad_norm": 0.140625, + "learning_rate": 0.00165560549268408, + "loss": 0.1025, + "step": 28381 + }, + { + "epoch": 0.24636938915460804, + "grad_norm": 0.466796875, + "learning_rate": 0.0016555820783476969, + "loss": 0.0859, + "step": 28382 + }, + { + "epoch": 0.2463780696348122, + "grad_norm": 0.181640625, + "learning_rate": 0.0016555586634037383, + "loss": 0.1133, + "step": 28383 + }, + { + "epoch": 0.24638675011501637, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016555352478522288, + "loss": 0.084, + "step": 28384 + }, + { + "epoch": 0.24639543059522054, + "grad_norm": 0.419921875, + "learning_rate": 0.001655511831693194, + "loss": 0.1211, + "step": 28385 + }, + { + "epoch": 0.2464041110754247, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016554884149266602, + "loss": 0.0894, + "step": 28386 + }, + { + "epoch": 0.24641279155562887, + "grad_norm": 0.82421875, + "learning_rate": 0.0016554649975526528, + "loss": 0.1348, + "step": 28387 + }, + { + "epoch": 0.24642147203583303, + "grad_norm": 0.318359375, + "learning_rate": 0.0016554415795711967, + "loss": 0.1196, + "step": 28388 + }, + { + "epoch": 0.2464301525160372, + "grad_norm": 0.1796875, + "learning_rate": 0.0016554181609823183, + "loss": 0.1094, + "step": 28389 + }, + { + "epoch": 0.24643883299624136, + "grad_norm": 0.1552734375, + "learning_rate": 0.001655394741786043, + "loss": 0.1348, + "step": 28390 + }, + { + "epoch": 0.24644751347644553, + "grad_norm": 0.30859375, + "learning_rate": 0.0016553713219823966, + "loss": 0.0781, + "step": 28391 + }, + { + "epoch": 0.2464561939566497, + "grad_norm": 0.248046875, + "learning_rate": 0.0016553479015714038, + "loss": 0.1045, + "step": 28392 + }, + { + "epoch": 0.24646487443685386, + "grad_norm": 0.5234375, + "learning_rate": 0.0016553244805530911, + "loss": 0.0791, + "step": 28393 + }, + { + "epoch": 0.24647355491705802, + "grad_norm": 0.2080078125, + "learning_rate": 0.001655301058927484, + "loss": 0.1191, + "step": 28394 + }, + { + "epoch": 0.2464822353972622, + "grad_norm": 0.1552734375, + "learning_rate": 0.001655277636694608, + "loss": 0.1152, + "step": 28395 + }, + { + "epoch": 0.24649091587746635, + "grad_norm": 0.318359375, + "learning_rate": 0.0016552542138544891, + "loss": 0.0908, + "step": 28396 + }, + { + "epoch": 0.24649959635767052, + "grad_norm": 0.1005859375, + "learning_rate": 0.0016552307904071522, + "loss": 0.0957, + "step": 28397 + }, + { + "epoch": 0.24650827683787468, + "grad_norm": 0.8984375, + "learning_rate": 0.0016552073663526228, + "loss": 0.0957, + "step": 28398 + }, + { + "epoch": 0.24651695731807885, + "grad_norm": 0.255859375, + "learning_rate": 0.0016551839416909279, + "loss": 0.1069, + "step": 28399 + }, + { + "epoch": 0.246525637798283, + "grad_norm": 0.375, + "learning_rate": 0.0016551605164220912, + "loss": 0.1035, + "step": 28400 + }, + { + "epoch": 0.24653431827848715, + "grad_norm": 0.33203125, + "learning_rate": 0.0016551370905461401, + "loss": 0.0991, + "step": 28401 + }, + { + "epoch": 0.24654299875869132, + "grad_norm": 0.458984375, + "learning_rate": 0.0016551136640630987, + "loss": 0.1074, + "step": 28402 + }, + { + "epoch": 0.24655167923889548, + "grad_norm": 0.57421875, + "learning_rate": 0.0016550902369729939, + "loss": 0.1089, + "step": 28403 + }, + { + "epoch": 0.24656035971909965, + "grad_norm": 0.13671875, + "learning_rate": 0.0016550668092758507, + "loss": 0.1133, + "step": 28404 + }, + { + "epoch": 0.2465690401993038, + "grad_norm": 0.296875, + "learning_rate": 0.0016550433809716944, + "loss": 0.0996, + "step": 28405 + }, + { + "epoch": 0.24657772067950798, + "grad_norm": 0.1474609375, + "learning_rate": 0.0016550199520605513, + "loss": 0.0771, + "step": 28406 + }, + { + "epoch": 0.24658640115971214, + "grad_norm": 0.130859375, + "learning_rate": 0.001654996522542447, + "loss": 0.0874, + "step": 28407 + }, + { + "epoch": 0.2465950816399163, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016549730924174065, + "loss": 0.0918, + "step": 28408 + }, + { + "epoch": 0.24660376212012047, + "grad_norm": 0.2734375, + "learning_rate": 0.0016549496616854557, + "loss": 0.126, + "step": 28409 + }, + { + "epoch": 0.24661244260032464, + "grad_norm": 1.1328125, + "learning_rate": 0.0016549262303466202, + "loss": 0.124, + "step": 28410 + }, + { + "epoch": 0.2466211230805288, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016549027984009259, + "loss": 0.0952, + "step": 28411 + }, + { + "epoch": 0.24662980356073297, + "grad_norm": 0.69140625, + "learning_rate": 0.0016548793658483984, + "loss": 0.1074, + "step": 28412 + }, + { + "epoch": 0.24663848404093713, + "grad_norm": 0.71875, + "learning_rate": 0.0016548559326890631, + "loss": 0.1133, + "step": 28413 + }, + { + "epoch": 0.2466471645211413, + "grad_norm": 0.216796875, + "learning_rate": 0.0016548324989229452, + "loss": 0.1187, + "step": 28414 + }, + { + "epoch": 0.24665584500134546, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016548090645500714, + "loss": 0.1289, + "step": 28415 + }, + { + "epoch": 0.24666452548154963, + "grad_norm": 0.66796875, + "learning_rate": 0.0016547856295704665, + "loss": 0.0957, + "step": 28416 + }, + { + "epoch": 0.2466732059617538, + "grad_norm": 0.3046875, + "learning_rate": 0.0016547621939841564, + "loss": 0.1123, + "step": 28417 + }, + { + "epoch": 0.24668188644195796, + "grad_norm": 0.134765625, + "learning_rate": 0.0016547387577911666, + "loss": 0.0869, + "step": 28418 + }, + { + "epoch": 0.24669056692216212, + "grad_norm": 0.2431640625, + "learning_rate": 0.001654715320991523, + "loss": 0.1182, + "step": 28419 + }, + { + "epoch": 0.2466992474023663, + "grad_norm": 0.107421875, + "learning_rate": 0.001654691883585251, + "loss": 0.0967, + "step": 28420 + }, + { + "epoch": 0.24670792788257045, + "grad_norm": 0.20703125, + "learning_rate": 0.0016546684455723763, + "loss": 0.1094, + "step": 28421 + }, + { + "epoch": 0.24671660836277462, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016546450069529245, + "loss": 0.0894, + "step": 28422 + }, + { + "epoch": 0.24672528884297878, + "grad_norm": 0.50390625, + "learning_rate": 0.0016546215677269212, + "loss": 0.0879, + "step": 28423 + }, + { + "epoch": 0.24673396932318295, + "grad_norm": 0.1796875, + "learning_rate": 0.0016545981278943923, + "loss": 0.1367, + "step": 28424 + }, + { + "epoch": 0.24674264980338712, + "grad_norm": 0.59375, + "learning_rate": 0.001654574687455363, + "loss": 0.1182, + "step": 28425 + }, + { + "epoch": 0.24675133028359128, + "grad_norm": 0.140625, + "learning_rate": 0.0016545512464098593, + "loss": 0.1318, + "step": 28426 + }, + { + "epoch": 0.24676001076379545, + "grad_norm": 0.11669921875, + "learning_rate": 0.0016545278047579064, + "loss": 0.105, + "step": 28427 + }, + { + "epoch": 0.2467686912439996, + "grad_norm": 0.33984375, + "learning_rate": 0.0016545043624995306, + "loss": 0.0977, + "step": 28428 + }, + { + "epoch": 0.24677737172420378, + "grad_norm": 0.298828125, + "learning_rate": 0.001654480919634757, + "loss": 0.104, + "step": 28429 + }, + { + "epoch": 0.24678605220440794, + "grad_norm": 0.1123046875, + "learning_rate": 0.0016544574761636116, + "loss": 0.1167, + "step": 28430 + }, + { + "epoch": 0.2467947326846121, + "grad_norm": 0.1640625, + "learning_rate": 0.0016544340320861193, + "loss": 0.1328, + "step": 28431 + }, + { + "epoch": 0.24680341316481627, + "grad_norm": 0.5390625, + "learning_rate": 0.0016544105874023065, + "loss": 0.0957, + "step": 28432 + }, + { + "epoch": 0.24681209364502044, + "grad_norm": 0.058837890625, + "learning_rate": 0.0016543871421121987, + "loss": 0.0659, + "step": 28433 + }, + { + "epoch": 0.2468207741252246, + "grad_norm": 1.3359375, + "learning_rate": 0.0016543636962158218, + "loss": 0.124, + "step": 28434 + }, + { + "epoch": 0.24682945460542877, + "grad_norm": 0.1142578125, + "learning_rate": 0.0016543402497132007, + "loss": 0.0757, + "step": 28435 + }, + { + "epoch": 0.24683813508563293, + "grad_norm": 0.59375, + "learning_rate": 0.0016543168026043613, + "loss": 0.1318, + "step": 28436 + }, + { + "epoch": 0.2468468155658371, + "grad_norm": 0.11328125, + "learning_rate": 0.0016542933548893298, + "loss": 0.1045, + "step": 28437 + }, + { + "epoch": 0.24685549604604126, + "grad_norm": 0.1474609375, + "learning_rate": 0.001654269906568131, + "loss": 0.0977, + "step": 28438 + }, + { + "epoch": 0.24686417652624543, + "grad_norm": 0.162109375, + "learning_rate": 0.0016542464576407912, + "loss": 0.0791, + "step": 28439 + }, + { + "epoch": 0.2468728570064496, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016542230081073358, + "loss": 0.1396, + "step": 28440 + }, + { + "epoch": 0.24688153748665376, + "grad_norm": 0.263671875, + "learning_rate": 0.0016541995579677903, + "loss": 0.1084, + "step": 28441 + }, + { + "epoch": 0.24689021796685792, + "grad_norm": 0.2177734375, + "learning_rate": 0.0016541761072221806, + "loss": 0.124, + "step": 28442 + }, + { + "epoch": 0.2468988984470621, + "grad_norm": 0.1171875, + "learning_rate": 0.0016541526558705322, + "loss": 0.0972, + "step": 28443 + }, + { + "epoch": 0.24690757892726625, + "grad_norm": 0.40234375, + "learning_rate": 0.0016541292039128708, + "loss": 0.1436, + "step": 28444 + }, + { + "epoch": 0.24691625940747042, + "grad_norm": 0.12060546875, + "learning_rate": 0.0016541057513492219, + "loss": 0.1582, + "step": 28445 + }, + { + "epoch": 0.24692493988767458, + "grad_norm": 0.92578125, + "learning_rate": 0.0016540822981796115, + "loss": 0.1367, + "step": 28446 + }, + { + "epoch": 0.24693362036787875, + "grad_norm": 0.671875, + "learning_rate": 0.001654058844404065, + "loss": 0.0776, + "step": 28447 + }, + { + "epoch": 0.2469423008480829, + "grad_norm": 0.734375, + "learning_rate": 0.0016540353900226076, + "loss": 0.0894, + "step": 28448 + }, + { + "epoch": 0.24695098132828708, + "grad_norm": 0.087890625, + "learning_rate": 0.001654011935035266, + "loss": 0.1182, + "step": 28449 + }, + { + "epoch": 0.24695966180849124, + "grad_norm": 0.44921875, + "learning_rate": 0.001653988479442065, + "loss": 0.106, + "step": 28450 + }, + { + "epoch": 0.2469683422886954, + "grad_norm": 0.16796875, + "learning_rate": 0.0016539650232430308, + "loss": 0.0854, + "step": 28451 + }, + { + "epoch": 0.24697702276889957, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016539415664381888, + "loss": 0.0977, + "step": 28452 + }, + { + "epoch": 0.24698570324910374, + "grad_norm": 0.6015625, + "learning_rate": 0.001653918109027564, + "loss": 0.1338, + "step": 28453 + }, + { + "epoch": 0.2469943837293079, + "grad_norm": 0.26171875, + "learning_rate": 0.0016538946510111833, + "loss": 0.1113, + "step": 28454 + }, + { + "epoch": 0.24700306420951207, + "grad_norm": 0.244140625, + "learning_rate": 0.0016538711923890712, + "loss": 0.1035, + "step": 28455 + }, + { + "epoch": 0.24701174468971623, + "grad_norm": 0.0888671875, + "learning_rate": 0.0016538477331612545, + "loss": 0.0713, + "step": 28456 + }, + { + "epoch": 0.2470204251699204, + "grad_norm": 0.220703125, + "learning_rate": 0.0016538242733277577, + "loss": 0.1025, + "step": 28457 + }, + { + "epoch": 0.24702910565012456, + "grad_norm": 0.2412109375, + "learning_rate": 0.0016538008128886075, + "loss": 0.0884, + "step": 28458 + }, + { + "epoch": 0.24703778613032873, + "grad_norm": 0.275390625, + "learning_rate": 0.0016537773518438285, + "loss": 0.1108, + "step": 28459 + }, + { + "epoch": 0.2470464666105329, + "grad_norm": 0.2060546875, + "learning_rate": 0.0016537538901934472, + "loss": 0.1309, + "step": 28460 + }, + { + "epoch": 0.24705514709073706, + "grad_norm": 0.5390625, + "learning_rate": 0.0016537304279374892, + "loss": 0.1074, + "step": 28461 + }, + { + "epoch": 0.24706382757094122, + "grad_norm": 0.388671875, + "learning_rate": 0.0016537069650759794, + "loss": 0.1299, + "step": 28462 + }, + { + "epoch": 0.2470725080511454, + "grad_norm": 0.400390625, + "learning_rate": 0.0016536835016089447, + "loss": 0.0708, + "step": 28463 + }, + { + "epoch": 0.24708118853134955, + "grad_norm": 0.486328125, + "learning_rate": 0.0016536600375364092, + "loss": 0.0903, + "step": 28464 + }, + { + "epoch": 0.24708986901155372, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016536365728584, + "loss": 0.0957, + "step": 28465 + }, + { + "epoch": 0.24709854949175789, + "grad_norm": 0.25390625, + "learning_rate": 0.001653613107574942, + "loss": 0.1172, + "step": 28466 + }, + { + "epoch": 0.24710722997196205, + "grad_norm": 0.60546875, + "learning_rate": 0.0016535896416860613, + "loss": 0.083, + "step": 28467 + }, + { + "epoch": 0.24711591045216622, + "grad_norm": 0.259765625, + "learning_rate": 0.0016535661751917827, + "loss": 0.1475, + "step": 28468 + }, + { + "epoch": 0.24712459093237038, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016535427080921325, + "loss": 0.106, + "step": 28469 + }, + { + "epoch": 0.24713327141257455, + "grad_norm": 0.333984375, + "learning_rate": 0.0016535192403871366, + "loss": 0.0913, + "step": 28470 + }, + { + "epoch": 0.2471419518927787, + "grad_norm": 0.76953125, + "learning_rate": 0.0016534957720768203, + "loss": 0.1357, + "step": 28471 + }, + { + "epoch": 0.24715063237298288, + "grad_norm": 0.4140625, + "learning_rate": 0.0016534723031612093, + "loss": 0.0801, + "step": 28472 + }, + { + "epoch": 0.24715931285318704, + "grad_norm": 0.408203125, + "learning_rate": 0.0016534488336403296, + "loss": 0.1719, + "step": 28473 + }, + { + "epoch": 0.2471679933333912, + "grad_norm": 0.10888671875, + "learning_rate": 0.0016534253635142062, + "loss": 0.1133, + "step": 28474 + }, + { + "epoch": 0.24717667381359537, + "grad_norm": 0.1513671875, + "learning_rate": 0.001653401892782865, + "loss": 0.1147, + "step": 28475 + }, + { + "epoch": 0.24718535429379954, + "grad_norm": 0.58984375, + "learning_rate": 0.0016533784214463325, + "loss": 0.1504, + "step": 28476 + }, + { + "epoch": 0.2471940347740037, + "grad_norm": 0.146484375, + "learning_rate": 0.0016533549495046332, + "loss": 0.1167, + "step": 28477 + }, + { + "epoch": 0.24720271525420787, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016533314769577932, + "loss": 0.1445, + "step": 28478 + }, + { + "epoch": 0.24721139573441203, + "grad_norm": 0.0966796875, + "learning_rate": 0.001653308003805838, + "loss": 0.1357, + "step": 28479 + }, + { + "epoch": 0.2472200762146162, + "grad_norm": 0.1240234375, + "learning_rate": 0.001653284530048794, + "loss": 0.1396, + "step": 28480 + }, + { + "epoch": 0.24722875669482036, + "grad_norm": 0.466796875, + "learning_rate": 0.0016532610556866862, + "loss": 0.1465, + "step": 28481 + }, + { + "epoch": 0.24723743717502453, + "grad_norm": 0.26171875, + "learning_rate": 0.0016532375807195405, + "loss": 0.1167, + "step": 28482 + }, + { + "epoch": 0.2472461176552287, + "grad_norm": 0.5, + "learning_rate": 0.0016532141051473824, + "loss": 0.1191, + "step": 28483 + }, + { + "epoch": 0.24725479813543286, + "grad_norm": 0.1357421875, + "learning_rate": 0.0016531906289702371, + "loss": 0.0874, + "step": 28484 + }, + { + "epoch": 0.24726347861563702, + "grad_norm": 0.31640625, + "learning_rate": 0.0016531671521881318, + "loss": 0.1416, + "step": 28485 + }, + { + "epoch": 0.2472721590958412, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016531436748010908, + "loss": 0.0938, + "step": 28486 + }, + { + "epoch": 0.24728083957604535, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016531201968091398, + "loss": 0.1084, + "step": 28487 + }, + { + "epoch": 0.24728952005624952, + "grad_norm": 0.1982421875, + "learning_rate": 0.0016530967182123055, + "loss": 0.0693, + "step": 28488 + }, + { + "epoch": 0.24729820053645368, + "grad_norm": 0.30859375, + "learning_rate": 0.0016530732390106126, + "loss": 0.1299, + "step": 28489 + }, + { + "epoch": 0.24730688101665785, + "grad_norm": 0.63671875, + "learning_rate": 0.0016530497592040872, + "loss": 0.0859, + "step": 28490 + }, + { + "epoch": 0.247315561496862, + "grad_norm": 0.26953125, + "learning_rate": 0.001653026278792755, + "loss": 0.0996, + "step": 28491 + }, + { + "epoch": 0.24732424197706618, + "grad_norm": 0.59765625, + "learning_rate": 0.0016530027977766414, + "loss": 0.0967, + "step": 28492 + }, + { + "epoch": 0.24733292245727034, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016529793161557723, + "loss": 0.125, + "step": 28493 + }, + { + "epoch": 0.2473416029374745, + "grad_norm": 0.125, + "learning_rate": 0.0016529558339301732, + "loss": 0.1396, + "step": 28494 + }, + { + "epoch": 0.24735028341767867, + "grad_norm": 0.1591796875, + "learning_rate": 0.00165293235109987, + "loss": 0.1094, + "step": 28495 + }, + { + "epoch": 0.24735896389788284, + "grad_norm": 0.421875, + "learning_rate": 0.0016529088676648883, + "loss": 0.0913, + "step": 28496 + }, + { + "epoch": 0.247367644378087, + "grad_norm": 0.55078125, + "learning_rate": 0.0016528853836252537, + "loss": 0.1074, + "step": 28497 + }, + { + "epoch": 0.24737632485829117, + "grad_norm": 0.205078125, + "learning_rate": 0.0016528618989809921, + "loss": 0.1045, + "step": 28498 + }, + { + "epoch": 0.24738500533849533, + "grad_norm": 0.3125, + "learning_rate": 0.001652838413732129, + "loss": 0.1089, + "step": 28499 + }, + { + "epoch": 0.2473936858186995, + "grad_norm": 0.271484375, + "learning_rate": 0.00165281492787869, + "loss": 0.1162, + "step": 28500 + }, + { + "epoch": 0.24740236629890366, + "grad_norm": 0.2412109375, + "learning_rate": 0.0016527914414207012, + "loss": 0.1279, + "step": 28501 + }, + { + "epoch": 0.24741104677910783, + "grad_norm": 0.2109375, + "learning_rate": 0.0016527679543581878, + "loss": 0.1016, + "step": 28502 + }, + { + "epoch": 0.247419727259312, + "grad_norm": 0.296875, + "learning_rate": 0.0016527444666911753, + "loss": 0.0864, + "step": 28503 + }, + { + "epoch": 0.24742840773951616, + "grad_norm": 0.54296875, + "learning_rate": 0.0016527209784196904, + "loss": 0.1289, + "step": 28504 + }, + { + "epoch": 0.24743708821972032, + "grad_norm": 0.44921875, + "learning_rate": 0.0016526974895437576, + "loss": 0.1143, + "step": 28505 + }, + { + "epoch": 0.2474457686999245, + "grad_norm": 0.416015625, + "learning_rate": 0.0016526740000634035, + "loss": 0.1167, + "step": 28506 + }, + { + "epoch": 0.24745444918012865, + "grad_norm": 0.23828125, + "learning_rate": 0.0016526505099786531, + "loss": 0.1465, + "step": 28507 + }, + { + "epoch": 0.24746312966033282, + "grad_norm": 0.291015625, + "learning_rate": 0.001652627019289533, + "loss": 0.1104, + "step": 28508 + }, + { + "epoch": 0.24747181014053699, + "grad_norm": 0.56640625, + "learning_rate": 0.0016526035279960675, + "loss": 0.1367, + "step": 28509 + }, + { + "epoch": 0.24748049062074115, + "grad_norm": 0.251953125, + "learning_rate": 0.0016525800360982837, + "loss": 0.0933, + "step": 28510 + }, + { + "epoch": 0.24748917110094532, + "grad_norm": 0.138671875, + "learning_rate": 0.0016525565435962064, + "loss": 0.1348, + "step": 28511 + }, + { + "epoch": 0.24749785158114948, + "grad_norm": 0.4375, + "learning_rate": 0.0016525330504898615, + "loss": 0.1113, + "step": 28512 + }, + { + "epoch": 0.24750653206135365, + "grad_norm": 0.5, + "learning_rate": 0.0016525095567792747, + "loss": 0.1338, + "step": 28513 + }, + { + "epoch": 0.2475152125415578, + "grad_norm": 0.10595703125, + "learning_rate": 0.0016524860624644723, + "loss": 0.082, + "step": 28514 + }, + { + "epoch": 0.24752389302176198, + "grad_norm": 1.25, + "learning_rate": 0.0016524625675454787, + "loss": 0.1099, + "step": 28515 + }, + { + "epoch": 0.24753257350196614, + "grad_norm": 0.486328125, + "learning_rate": 0.001652439072022321, + "loss": 0.0649, + "step": 28516 + }, + { + "epoch": 0.2475412539821703, + "grad_norm": 0.1142578125, + "learning_rate": 0.001652415575895024, + "loss": 0.0986, + "step": 28517 + }, + { + "epoch": 0.24754993446237447, + "grad_norm": 0.0986328125, + "learning_rate": 0.0016523920791636133, + "loss": 0.0811, + "step": 28518 + }, + { + "epoch": 0.24755861494257864, + "grad_norm": 0.76953125, + "learning_rate": 0.0016523685818281154, + "loss": 0.0879, + "step": 28519 + }, + { + "epoch": 0.2475672954227828, + "grad_norm": 0.16796875, + "learning_rate": 0.0016523450838885551, + "loss": 0.0811, + "step": 28520 + }, + { + "epoch": 0.24757597590298697, + "grad_norm": 0.171875, + "learning_rate": 0.001652321585344959, + "loss": 0.1221, + "step": 28521 + }, + { + "epoch": 0.24758465638319113, + "grad_norm": 0.19921875, + "learning_rate": 0.001652298086197352, + "loss": 0.0908, + "step": 28522 + }, + { + "epoch": 0.2475933368633953, + "grad_norm": 0.146484375, + "learning_rate": 0.00165227458644576, + "loss": 0.0996, + "step": 28523 + }, + { + "epoch": 0.24760201734359943, + "grad_norm": 0.37890625, + "learning_rate": 0.001652251086090209, + "loss": 0.123, + "step": 28524 + }, + { + "epoch": 0.2476106978238036, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016522275851307245, + "loss": 0.0938, + "step": 28525 + }, + { + "epoch": 0.24761937830400776, + "grad_norm": 0.26171875, + "learning_rate": 0.001652204083567332, + "loss": 0.1245, + "step": 28526 + }, + { + "epoch": 0.24762805878421193, + "grad_norm": 0.54296875, + "learning_rate": 0.0016521805814000577, + "loss": 0.1196, + "step": 28527 + }, + { + "epoch": 0.2476367392644161, + "grad_norm": 0.1142578125, + "learning_rate": 0.0016521570786289267, + "loss": 0.0938, + "step": 28528 + }, + { + "epoch": 0.24764541974462026, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016521335752539652, + "loss": 0.0889, + "step": 28529 + }, + { + "epoch": 0.24765410022482442, + "grad_norm": 0.2890625, + "learning_rate": 0.0016521100712751989, + "loss": 0.0859, + "step": 28530 + }, + { + "epoch": 0.2476627807050286, + "grad_norm": 0.6796875, + "learning_rate": 0.0016520865666926532, + "loss": 0.1338, + "step": 28531 + }, + { + "epoch": 0.24767146118523276, + "grad_norm": 0.2578125, + "learning_rate": 0.0016520630615063537, + "loss": 0.0737, + "step": 28532 + }, + { + "epoch": 0.24768014166543692, + "grad_norm": 0.40234375, + "learning_rate": 0.0016520395557163266, + "loss": 0.085, + "step": 28533 + }, + { + "epoch": 0.24768882214564109, + "grad_norm": 0.29296875, + "learning_rate": 0.0016520160493225973, + "loss": 0.1001, + "step": 28534 + }, + { + "epoch": 0.24769750262584525, + "grad_norm": 0.10546875, + "learning_rate": 0.0016519925423251916, + "loss": 0.1226, + "step": 28535 + }, + { + "epoch": 0.24770618310604942, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016519690347241347, + "loss": 0.0762, + "step": 28536 + }, + { + "epoch": 0.24771486358625358, + "grad_norm": 0.228515625, + "learning_rate": 0.0016519455265194532, + "loss": 0.1367, + "step": 28537 + }, + { + "epoch": 0.24772354406645775, + "grad_norm": 0.47265625, + "learning_rate": 0.0016519220177111725, + "loss": 0.1177, + "step": 28538 + }, + { + "epoch": 0.2477322245466619, + "grad_norm": 0.0791015625, + "learning_rate": 0.0016518985082993177, + "loss": 0.1006, + "step": 28539 + }, + { + "epoch": 0.24774090502686608, + "grad_norm": 0.0771484375, + "learning_rate": 0.0016518749982839152, + "loss": 0.0884, + "step": 28540 + }, + { + "epoch": 0.24774958550707024, + "grad_norm": 0.447265625, + "learning_rate": 0.0016518514876649906, + "loss": 0.1436, + "step": 28541 + }, + { + "epoch": 0.2477582659872744, + "grad_norm": 0.6953125, + "learning_rate": 0.0016518279764425692, + "loss": 0.1245, + "step": 28542 + }, + { + "epoch": 0.24776694646747857, + "grad_norm": 0.10498046875, + "learning_rate": 0.0016518044646166772, + "loss": 0.1152, + "step": 28543 + }, + { + "epoch": 0.24777562694768274, + "grad_norm": 0.25390625, + "learning_rate": 0.0016517809521873403, + "loss": 0.0752, + "step": 28544 + }, + { + "epoch": 0.2477843074278869, + "grad_norm": 0.1533203125, + "learning_rate": 0.001651757439154584, + "loss": 0.1089, + "step": 28545 + }, + { + "epoch": 0.24779298790809107, + "grad_norm": 0.439453125, + "learning_rate": 0.0016517339255184339, + "loss": 0.0869, + "step": 28546 + }, + { + "epoch": 0.24780166838829523, + "grad_norm": 0.6796875, + "learning_rate": 0.001651710411278916, + "loss": 0.0967, + "step": 28547 + }, + { + "epoch": 0.2478103488684994, + "grad_norm": 0.16015625, + "learning_rate": 0.0016516868964360557, + "loss": 0.1299, + "step": 28548 + }, + { + "epoch": 0.24781902934870356, + "grad_norm": 0.09375, + "learning_rate": 0.001651663380989879, + "loss": 0.1055, + "step": 28549 + }, + { + "epoch": 0.24782770982890773, + "grad_norm": 0.306640625, + "learning_rate": 0.0016516398649404115, + "loss": 0.1201, + "step": 28550 + }, + { + "epoch": 0.2478363903091119, + "grad_norm": 0.10791015625, + "learning_rate": 0.001651616348287679, + "loss": 0.1123, + "step": 28551 + }, + { + "epoch": 0.24784507078931606, + "grad_norm": 0.1650390625, + "learning_rate": 0.001651592831031707, + "loss": 0.1001, + "step": 28552 + }, + { + "epoch": 0.24785375126952022, + "grad_norm": 0.2578125, + "learning_rate": 0.0016515693131725216, + "loss": 0.1123, + "step": 28553 + }, + { + "epoch": 0.2478624317497244, + "grad_norm": 0.201171875, + "learning_rate": 0.001651545794710148, + "loss": 0.1162, + "step": 28554 + }, + { + "epoch": 0.24787111222992855, + "grad_norm": 0.255859375, + "learning_rate": 0.0016515222756446125, + "loss": 0.1147, + "step": 28555 + }, + { + "epoch": 0.24787979271013272, + "grad_norm": 1.140625, + "learning_rate": 0.0016514987559759402, + "loss": 0.1621, + "step": 28556 + }, + { + "epoch": 0.24788847319033688, + "grad_norm": 0.5859375, + "learning_rate": 0.0016514752357041573, + "loss": 0.0928, + "step": 28557 + }, + { + "epoch": 0.24789715367054105, + "grad_norm": 0.462890625, + "learning_rate": 0.0016514517148292893, + "loss": 0.1396, + "step": 28558 + }, + { + "epoch": 0.2479058341507452, + "grad_norm": 0.416015625, + "learning_rate": 0.0016514281933513624, + "loss": 0.1172, + "step": 28559 + }, + { + "epoch": 0.24791451463094938, + "grad_norm": 8.125, + "learning_rate": 0.0016514046712704014, + "loss": 0.707, + "step": 28560 + }, + { + "epoch": 0.24792319511115354, + "grad_norm": 0.2333984375, + "learning_rate": 0.0016513811485864328, + "loss": 0.083, + "step": 28561 + }, + { + "epoch": 0.2479318755913577, + "grad_norm": 0.345703125, + "learning_rate": 0.001651357625299482, + "loss": 0.1055, + "step": 28562 + }, + { + "epoch": 0.24794055607156187, + "grad_norm": 0.1884765625, + "learning_rate": 0.001651334101409575, + "loss": 0.0796, + "step": 28563 + }, + { + "epoch": 0.24794923655176604, + "grad_norm": 0.451171875, + "learning_rate": 0.0016513105769167368, + "loss": 0.0967, + "step": 28564 + }, + { + "epoch": 0.2479579170319702, + "grad_norm": 0.42578125, + "learning_rate": 0.0016512870518209939, + "loss": 0.0894, + "step": 28565 + }, + { + "epoch": 0.24796659751217437, + "grad_norm": 0.291015625, + "learning_rate": 0.0016512635261223718, + "loss": 0.0703, + "step": 28566 + }, + { + "epoch": 0.24797527799237853, + "grad_norm": 0.43359375, + "learning_rate": 0.0016512399998208963, + "loss": 0.1035, + "step": 28567 + }, + { + "epoch": 0.2479839584725827, + "grad_norm": 1.0703125, + "learning_rate": 0.0016512164729165929, + "loss": 0.125, + "step": 28568 + }, + { + "epoch": 0.24799263895278686, + "grad_norm": 0.1640625, + "learning_rate": 0.0016511929454094874, + "loss": 0.0889, + "step": 28569 + }, + { + "epoch": 0.24800131943299103, + "grad_norm": 0.17578125, + "learning_rate": 0.0016511694172996058, + "loss": 0.0947, + "step": 28570 + }, + { + "epoch": 0.2480099999131952, + "grad_norm": 0.56640625, + "learning_rate": 0.0016511458885869732, + "loss": 0.0879, + "step": 28571 + }, + { + "epoch": 0.24801868039339936, + "grad_norm": 0.1328125, + "learning_rate": 0.001651122359271616, + "loss": 0.103, + "step": 28572 + }, + { + "epoch": 0.24802736087360353, + "grad_norm": 0.294921875, + "learning_rate": 0.0016510988293535597, + "loss": 0.0811, + "step": 28573 + }, + { + "epoch": 0.2480360413538077, + "grad_norm": 0.546875, + "learning_rate": 0.00165107529883283, + "loss": 0.1309, + "step": 28574 + }, + { + "epoch": 0.24804472183401186, + "grad_norm": 0.6171875, + "learning_rate": 0.0016510517677094526, + "loss": 0.0845, + "step": 28575 + }, + { + "epoch": 0.24805340231421602, + "grad_norm": 0.0888671875, + "learning_rate": 0.0016510282359834532, + "loss": 0.1089, + "step": 28576 + }, + { + "epoch": 0.24806208279442019, + "grad_norm": 0.1044921875, + "learning_rate": 0.0016510047036548577, + "loss": 0.1309, + "step": 28577 + }, + { + "epoch": 0.24807076327462435, + "grad_norm": 0.166015625, + "learning_rate": 0.0016509811707236914, + "loss": 0.0981, + "step": 28578 + }, + { + "epoch": 0.24807944375482852, + "grad_norm": 0.306640625, + "learning_rate": 0.0016509576371899809, + "loss": 0.1104, + "step": 28579 + }, + { + "epoch": 0.24808812423503268, + "grad_norm": 0.17578125, + "learning_rate": 0.001650934103053751, + "loss": 0.1084, + "step": 28580 + }, + { + "epoch": 0.24809680471523685, + "grad_norm": 0.55078125, + "learning_rate": 0.001650910568315028, + "loss": 0.0986, + "step": 28581 + }, + { + "epoch": 0.248105485195441, + "grad_norm": 0.267578125, + "learning_rate": 0.0016508870329738372, + "loss": 0.1348, + "step": 28582 + }, + { + "epoch": 0.24811416567564518, + "grad_norm": 0.345703125, + "learning_rate": 0.0016508634970302052, + "loss": 0.1177, + "step": 28583 + }, + { + "epoch": 0.24812284615584934, + "grad_norm": 0.435546875, + "learning_rate": 0.001650839960484157, + "loss": 0.1016, + "step": 28584 + }, + { + "epoch": 0.2481315266360535, + "grad_norm": 0.30859375, + "learning_rate": 0.001650816423335718, + "loss": 0.1396, + "step": 28585 + }, + { + "epoch": 0.24814020711625767, + "grad_norm": 0.171875, + "learning_rate": 0.001650792885584915, + "loss": 0.1006, + "step": 28586 + }, + { + "epoch": 0.24814888759646184, + "grad_norm": 0.30859375, + "learning_rate": 0.001650769347231773, + "loss": 0.0938, + "step": 28587 + }, + { + "epoch": 0.248157568076666, + "grad_norm": 0.17578125, + "learning_rate": 0.0016507458082763175, + "loss": 0.0947, + "step": 28588 + }, + { + "epoch": 0.24816624855687017, + "grad_norm": 0.6171875, + "learning_rate": 0.0016507222687185753, + "loss": 0.1035, + "step": 28589 + }, + { + "epoch": 0.24817492903707433, + "grad_norm": 0.30859375, + "learning_rate": 0.0016506987285585713, + "loss": 0.1006, + "step": 28590 + }, + { + "epoch": 0.2481836095172785, + "grad_norm": 0.310546875, + "learning_rate": 0.001650675187796331, + "loss": 0.1582, + "step": 28591 + }, + { + "epoch": 0.24819228999748266, + "grad_norm": 0.53125, + "learning_rate": 0.0016506516464318813, + "loss": 0.1191, + "step": 28592 + }, + { + "epoch": 0.24820097047768683, + "grad_norm": 0.380859375, + "learning_rate": 0.001650628104465247, + "loss": 0.1006, + "step": 28593 + }, + { + "epoch": 0.248209650957891, + "grad_norm": 0.12158203125, + "learning_rate": 0.0016506045618964538, + "loss": 0.0928, + "step": 28594 + }, + { + "epoch": 0.24821833143809516, + "grad_norm": 0.208984375, + "learning_rate": 0.001650581018725528, + "loss": 0.0913, + "step": 28595 + }, + { + "epoch": 0.24822701191829932, + "grad_norm": 0.31640625, + "learning_rate": 0.0016505574749524948, + "loss": 0.0903, + "step": 28596 + }, + { + "epoch": 0.2482356923985035, + "grad_norm": 0.7265625, + "learning_rate": 0.0016505339305773804, + "loss": 0.1196, + "step": 28597 + }, + { + "epoch": 0.24824437287870765, + "grad_norm": 0.470703125, + "learning_rate": 0.0016505103856002104, + "loss": 0.0654, + "step": 28598 + }, + { + "epoch": 0.24825305335891182, + "grad_norm": 0.255859375, + "learning_rate": 0.0016504868400210105, + "loss": 0.1143, + "step": 28599 + }, + { + "epoch": 0.24826173383911598, + "grad_norm": 0.9765625, + "learning_rate": 0.0016504632938398062, + "loss": 0.085, + "step": 28600 + }, + { + "epoch": 0.24827041431932015, + "grad_norm": 0.28125, + "learning_rate": 0.0016504397470566239, + "loss": 0.127, + "step": 28601 + }, + { + "epoch": 0.2482790947995243, + "grad_norm": 0.2353515625, + "learning_rate": 0.001650416199671489, + "loss": 0.1221, + "step": 28602 + }, + { + "epoch": 0.24828777527972848, + "grad_norm": 0.328125, + "learning_rate": 0.0016503926516844268, + "loss": 0.1318, + "step": 28603 + }, + { + "epoch": 0.24829645575993264, + "grad_norm": 0.16015625, + "learning_rate": 0.0016503691030954636, + "loss": 0.1406, + "step": 28604 + }, + { + "epoch": 0.2483051362401368, + "grad_norm": 1.1015625, + "learning_rate": 0.0016503455539046252, + "loss": 0.0889, + "step": 28605 + }, + { + "epoch": 0.24831381672034097, + "grad_norm": 0.375, + "learning_rate": 0.001650322004111937, + "loss": 0.0957, + "step": 28606 + }, + { + "epoch": 0.24832249720054514, + "grad_norm": 0.11474609375, + "learning_rate": 0.0016502984537174253, + "loss": 0.1201, + "step": 28607 + }, + { + "epoch": 0.2483311776807493, + "grad_norm": 0.1279296875, + "learning_rate": 0.001650274902721115, + "loss": 0.1191, + "step": 28608 + }, + { + "epoch": 0.24833985816095347, + "grad_norm": 0.42578125, + "learning_rate": 0.0016502513511230323, + "loss": 0.1328, + "step": 28609 + }, + { + "epoch": 0.24834853864115763, + "grad_norm": 0.0966796875, + "learning_rate": 0.0016502277989232032, + "loss": 0.1152, + "step": 28610 + }, + { + "epoch": 0.2483572191213618, + "grad_norm": 0.400390625, + "learning_rate": 0.0016502042461216536, + "loss": 0.0791, + "step": 28611 + }, + { + "epoch": 0.24836589960156596, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016501806927184082, + "loss": 0.1299, + "step": 28612 + }, + { + "epoch": 0.24837458008177013, + "grad_norm": 0.427734375, + "learning_rate": 0.001650157138713494, + "loss": 0.1055, + "step": 28613 + }, + { + "epoch": 0.2483832605619743, + "grad_norm": 0.10986328125, + "learning_rate": 0.001650133584106936, + "loss": 0.0908, + "step": 28614 + }, + { + "epoch": 0.24839194104217846, + "grad_norm": 0.81640625, + "learning_rate": 0.0016501100288987603, + "loss": 0.1104, + "step": 28615 + }, + { + "epoch": 0.24840062152238263, + "grad_norm": 0.212890625, + "learning_rate": 0.0016500864730889925, + "loss": 0.1201, + "step": 28616 + }, + { + "epoch": 0.2484093020025868, + "grad_norm": 0.55859375, + "learning_rate": 0.0016500629166776581, + "loss": 0.0859, + "step": 28617 + }, + { + "epoch": 0.24841798248279096, + "grad_norm": 0.60546875, + "learning_rate": 0.0016500393596647835, + "loss": 0.1045, + "step": 28618 + }, + { + "epoch": 0.24842666296299512, + "grad_norm": 0.111328125, + "learning_rate": 0.0016500158020503939, + "loss": 0.1055, + "step": 28619 + }, + { + "epoch": 0.24843534344319929, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016499922438345156, + "loss": 0.1201, + "step": 28620 + }, + { + "epoch": 0.24844402392340345, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016499686850171735, + "loss": 0.0874, + "step": 28621 + }, + { + "epoch": 0.24845270440360762, + "grad_norm": 0.2890625, + "learning_rate": 0.0016499451255983946, + "loss": 0.1299, + "step": 28622 + }, + { + "epoch": 0.24846138488381178, + "grad_norm": 0.291015625, + "learning_rate": 0.0016499215655782034, + "loss": 0.106, + "step": 28623 + }, + { + "epoch": 0.24847006536401595, + "grad_norm": 0.365234375, + "learning_rate": 0.0016498980049566266, + "loss": 0.0889, + "step": 28624 + }, + { + "epoch": 0.2484787458442201, + "grad_norm": 1.0546875, + "learning_rate": 0.0016498744437336895, + "loss": 0.1406, + "step": 28625 + }, + { + "epoch": 0.24848742632442428, + "grad_norm": 0.09912109375, + "learning_rate": 0.0016498508819094177, + "loss": 0.0781, + "step": 28626 + }, + { + "epoch": 0.24849610680462844, + "grad_norm": 0.3046875, + "learning_rate": 0.0016498273194838378, + "loss": 0.1367, + "step": 28627 + }, + { + "epoch": 0.2485047872848326, + "grad_norm": 0.361328125, + "learning_rate": 0.0016498037564569746, + "loss": 0.0962, + "step": 28628 + }, + { + "epoch": 0.24851346776503677, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016497801928288542, + "loss": 0.1172, + "step": 28629 + }, + { + "epoch": 0.24852214824524094, + "grad_norm": 0.408203125, + "learning_rate": 0.0016497566285995028, + "loss": 0.1108, + "step": 28630 + }, + { + "epoch": 0.2485308287254451, + "grad_norm": 0.07568359375, + "learning_rate": 0.0016497330637689457, + "loss": 0.103, + "step": 28631 + }, + { + "epoch": 0.24853950920564927, + "grad_norm": 0.546875, + "learning_rate": 0.0016497094983372086, + "loss": 0.0972, + "step": 28632 + }, + { + "epoch": 0.24854818968585343, + "grad_norm": 0.484375, + "learning_rate": 0.0016496859323043176, + "loss": 0.1133, + "step": 28633 + }, + { + "epoch": 0.2485568701660576, + "grad_norm": 0.32421875, + "learning_rate": 0.001649662365670298, + "loss": 0.1445, + "step": 28634 + }, + { + "epoch": 0.24856555064626176, + "grad_norm": 0.56640625, + "learning_rate": 0.0016496387984351762, + "loss": 0.1543, + "step": 28635 + }, + { + "epoch": 0.24857423112646593, + "grad_norm": 0.65625, + "learning_rate": 0.0016496152305989777, + "loss": 0.1172, + "step": 28636 + }, + { + "epoch": 0.2485829116066701, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016495916621617282, + "loss": 0.0884, + "step": 28637 + }, + { + "epoch": 0.24859159208687426, + "grad_norm": 0.25390625, + "learning_rate": 0.0016495680931234535, + "loss": 0.0913, + "step": 28638 + }, + { + "epoch": 0.24860027256707842, + "grad_norm": 0.1279296875, + "learning_rate": 0.0016495445234841796, + "loss": 0.0942, + "step": 28639 + }, + { + "epoch": 0.2486089530472826, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016495209532439317, + "loss": 0.1055, + "step": 28640 + }, + { + "epoch": 0.24861763352748675, + "grad_norm": 0.380859375, + "learning_rate": 0.001649497382402736, + "loss": 0.1182, + "step": 28641 + }, + { + "epoch": 0.24862631400769092, + "grad_norm": 0.62890625, + "learning_rate": 0.0016494738109606184, + "loss": 0.082, + "step": 28642 + }, + { + "epoch": 0.24863499448789508, + "grad_norm": 0.66015625, + "learning_rate": 0.0016494502389176042, + "loss": 0.1235, + "step": 28643 + }, + { + "epoch": 0.24864367496809925, + "grad_norm": 0.10986328125, + "learning_rate": 0.0016494266662737196, + "loss": 0.1641, + "step": 28644 + }, + { + "epoch": 0.2486523554483034, + "grad_norm": 0.185546875, + "learning_rate": 0.0016494030930289908, + "loss": 0.082, + "step": 28645 + }, + { + "epoch": 0.24866103592850758, + "grad_norm": 0.1162109375, + "learning_rate": 0.0016493795191834422, + "loss": 0.1377, + "step": 28646 + }, + { + "epoch": 0.24866971640871172, + "grad_norm": 0.20703125, + "learning_rate": 0.0016493559447371008, + "loss": 0.106, + "step": 28647 + }, + { + "epoch": 0.24867839688891588, + "grad_norm": 0.1728515625, + "learning_rate": 0.001649332369689992, + "loss": 0.1426, + "step": 28648 + }, + { + "epoch": 0.24868707736912005, + "grad_norm": 0.34375, + "learning_rate": 0.0016493087940421417, + "loss": 0.1113, + "step": 28649 + }, + { + "epoch": 0.2486957578493242, + "grad_norm": 0.470703125, + "learning_rate": 0.0016492852177935754, + "loss": 0.1221, + "step": 28650 + }, + { + "epoch": 0.24870443832952838, + "grad_norm": 0.140625, + "learning_rate": 0.0016492616409443188, + "loss": 0.1279, + "step": 28651 + }, + { + "epoch": 0.24871311880973254, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016492380634943982, + "loss": 0.084, + "step": 28652 + }, + { + "epoch": 0.2487217992899367, + "grad_norm": 0.2265625, + "learning_rate": 0.0016492144854438394, + "loss": 0.0854, + "step": 28653 + }, + { + "epoch": 0.24873047977014087, + "grad_norm": 0.328125, + "learning_rate": 0.0016491909067926675, + "loss": 0.0942, + "step": 28654 + }, + { + "epoch": 0.24873916025034504, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016491673275409087, + "loss": 0.1172, + "step": 28655 + }, + { + "epoch": 0.2487478407305492, + "grad_norm": 0.07080078125, + "learning_rate": 0.0016491437476885886, + "loss": 0.0693, + "step": 28656 + }, + { + "epoch": 0.24875652121075337, + "grad_norm": 0.15625, + "learning_rate": 0.0016491201672357335, + "loss": 0.0742, + "step": 28657 + }, + { + "epoch": 0.24876520169095753, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016490965861823686, + "loss": 0.1025, + "step": 28658 + }, + { + "epoch": 0.2487738821711617, + "grad_norm": 0.330078125, + "learning_rate": 0.00164907300452852, + "loss": 0.1182, + "step": 28659 + }, + { + "epoch": 0.24878256265136586, + "grad_norm": 0.15234375, + "learning_rate": 0.0016490494222742133, + "loss": 0.082, + "step": 28660 + }, + { + "epoch": 0.24879124313157003, + "grad_norm": 0.3046875, + "learning_rate": 0.0016490258394194744, + "loss": 0.1113, + "step": 28661 + }, + { + "epoch": 0.2487999236117742, + "grad_norm": 0.341796875, + "learning_rate": 0.0016490022559643293, + "loss": 0.1279, + "step": 28662 + }, + { + "epoch": 0.24880860409197836, + "grad_norm": 0.10546875, + "learning_rate": 0.0016489786719088035, + "loss": 0.1143, + "step": 28663 + }, + { + "epoch": 0.24881728457218252, + "grad_norm": 0.578125, + "learning_rate": 0.001648955087252923, + "loss": 0.1035, + "step": 28664 + }, + { + "epoch": 0.2488259650523867, + "grad_norm": 0.1748046875, + "learning_rate": 0.0016489315019967133, + "loss": 0.1035, + "step": 28665 + }, + { + "epoch": 0.24883464553259085, + "grad_norm": 0.625, + "learning_rate": 0.0016489079161402004, + "loss": 0.1069, + "step": 28666 + }, + { + "epoch": 0.24884332601279502, + "grad_norm": 0.546875, + "learning_rate": 0.0016488843296834101, + "loss": 0.1045, + "step": 28667 + }, + { + "epoch": 0.24885200649299918, + "grad_norm": 0.75, + "learning_rate": 0.001648860742626368, + "loss": 0.1123, + "step": 28668 + }, + { + "epoch": 0.24886068697320335, + "grad_norm": 0.091796875, + "learning_rate": 0.0016488371549690998, + "loss": 0.0879, + "step": 28669 + }, + { + "epoch": 0.24886936745340751, + "grad_norm": 0.236328125, + "learning_rate": 0.001648813566711632, + "loss": 0.0908, + "step": 28670 + }, + { + "epoch": 0.24887804793361168, + "grad_norm": 0.478515625, + "learning_rate": 0.00164878997785399, + "loss": 0.0889, + "step": 28671 + }, + { + "epoch": 0.24888672841381584, + "grad_norm": 0.6328125, + "learning_rate": 0.0016487663883961992, + "loss": 0.1182, + "step": 28672 + }, + { + "epoch": 0.24889540889402, + "grad_norm": 0.388671875, + "learning_rate": 0.0016487427983382856, + "loss": 0.1465, + "step": 28673 + }, + { + "epoch": 0.24890408937422417, + "grad_norm": 0.33984375, + "learning_rate": 0.0016487192076802754, + "loss": 0.0947, + "step": 28674 + }, + { + "epoch": 0.24891276985442834, + "grad_norm": 0.24609375, + "learning_rate": 0.0016486956164221943, + "loss": 0.1216, + "step": 28675 + }, + { + "epoch": 0.2489214503346325, + "grad_norm": 0.40625, + "learning_rate": 0.001648672024564068, + "loss": 0.106, + "step": 28676 + }, + { + "epoch": 0.24893013081483667, + "grad_norm": 0.337890625, + "learning_rate": 0.0016486484321059213, + "loss": 0.125, + "step": 28677 + }, + { + "epoch": 0.24893881129504083, + "grad_norm": 0.3984375, + "learning_rate": 0.0016486248390477816, + "loss": 0.1348, + "step": 28678 + }, + { + "epoch": 0.248947491775245, + "grad_norm": 0.4453125, + "learning_rate": 0.001648601245389674, + "loss": 0.0879, + "step": 28679 + }, + { + "epoch": 0.24895617225544917, + "grad_norm": 0.322265625, + "learning_rate": 0.0016485776511316243, + "loss": 0.0967, + "step": 28680 + }, + { + "epoch": 0.24896485273565333, + "grad_norm": 0.287109375, + "learning_rate": 0.0016485540562736583, + "loss": 0.082, + "step": 28681 + }, + { + "epoch": 0.2489735332158575, + "grad_norm": 0.13671875, + "learning_rate": 0.0016485304608158018, + "loss": 0.0796, + "step": 28682 + }, + { + "epoch": 0.24898221369606166, + "grad_norm": 0.48828125, + "learning_rate": 0.0016485068647580806, + "loss": 0.1016, + "step": 28683 + }, + { + "epoch": 0.24899089417626583, + "grad_norm": 0.2890625, + "learning_rate": 0.0016484832681005206, + "loss": 0.123, + "step": 28684 + }, + { + "epoch": 0.24899957465647, + "grad_norm": 0.78515625, + "learning_rate": 0.0016484596708431476, + "loss": 0.0986, + "step": 28685 + }, + { + "epoch": 0.24900825513667416, + "grad_norm": 0.125, + "learning_rate": 0.0016484360729859871, + "loss": 0.1289, + "step": 28686 + }, + { + "epoch": 0.24901693561687832, + "grad_norm": 0.53125, + "learning_rate": 0.0016484124745290655, + "loss": 0.1118, + "step": 28687 + }, + { + "epoch": 0.2490256160970825, + "grad_norm": 0.2021484375, + "learning_rate": 0.001648388875472408, + "loss": 0.0967, + "step": 28688 + }, + { + "epoch": 0.24903429657728665, + "grad_norm": 0.12451171875, + "learning_rate": 0.0016483652758160405, + "loss": 0.0762, + "step": 28689 + }, + { + "epoch": 0.24904297705749082, + "grad_norm": 0.310546875, + "learning_rate": 0.0016483416755599895, + "loss": 0.1191, + "step": 28690 + }, + { + "epoch": 0.24905165753769498, + "grad_norm": 0.09765625, + "learning_rate": 0.0016483180747042798, + "loss": 0.0889, + "step": 28691 + }, + { + "epoch": 0.24906033801789915, + "grad_norm": 0.166015625, + "learning_rate": 0.001648294473248938, + "loss": 0.1152, + "step": 28692 + }, + { + "epoch": 0.2490690184981033, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016482708711939896, + "loss": 0.0933, + "step": 28693 + }, + { + "epoch": 0.24907769897830748, + "grad_norm": 0.232421875, + "learning_rate": 0.00164824726853946, + "loss": 0.1162, + "step": 28694 + }, + { + "epoch": 0.24908637945851164, + "grad_norm": 0.2294921875, + "learning_rate": 0.0016482236652853757, + "loss": 0.1123, + "step": 28695 + }, + { + "epoch": 0.2490950599387158, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016482000614317623, + "loss": 0.1592, + "step": 28696 + }, + { + "epoch": 0.24910374041891997, + "grad_norm": 0.5, + "learning_rate": 0.0016481764569786458, + "loss": 0.0996, + "step": 28697 + }, + { + "epoch": 0.24911242089912414, + "grad_norm": 0.4296875, + "learning_rate": 0.0016481528519260514, + "loss": 0.1094, + "step": 28698 + }, + { + "epoch": 0.2491211013793283, + "grad_norm": 0.08837890625, + "learning_rate": 0.0016481292462740052, + "loss": 0.0835, + "step": 28699 + }, + { + "epoch": 0.24912978185953247, + "grad_norm": 0.490234375, + "learning_rate": 0.001648105640022533, + "loss": 0.1055, + "step": 28700 + }, + { + "epoch": 0.24913846233973663, + "grad_norm": 0.275390625, + "learning_rate": 0.001648082033171661, + "loss": 0.1157, + "step": 28701 + }, + { + "epoch": 0.2491471428199408, + "grad_norm": 0.6953125, + "learning_rate": 0.001648058425721415, + "loss": 0.0894, + "step": 28702 + }, + { + "epoch": 0.24915582330014496, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016480348176718198, + "loss": 0.1416, + "step": 28703 + }, + { + "epoch": 0.24916450378034913, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016480112090229027, + "loss": 0.1108, + "step": 28704 + }, + { + "epoch": 0.2491731842605533, + "grad_norm": 0.76171875, + "learning_rate": 0.0016479875997746885, + "loss": 0.1133, + "step": 28705 + }, + { + "epoch": 0.24918186474075746, + "grad_norm": 0.498046875, + "learning_rate": 0.0016479639899272033, + "loss": 0.1621, + "step": 28706 + }, + { + "epoch": 0.24919054522096162, + "grad_norm": 0.578125, + "learning_rate": 0.0016479403794804727, + "loss": 0.1123, + "step": 28707 + }, + { + "epoch": 0.2491992257011658, + "grad_norm": 0.08935546875, + "learning_rate": 0.001647916768434523, + "loss": 0.0894, + "step": 28708 + }, + { + "epoch": 0.24920790618136995, + "grad_norm": 0.08935546875, + "learning_rate": 0.0016478931567893793, + "loss": 0.0825, + "step": 28709 + }, + { + "epoch": 0.24921658666157412, + "grad_norm": 0.478515625, + "learning_rate": 0.0016478695445450685, + "loss": 0.126, + "step": 28710 + }, + { + "epoch": 0.24922526714177828, + "grad_norm": 0.263671875, + "learning_rate": 0.0016478459317016154, + "loss": 0.0703, + "step": 28711 + }, + { + "epoch": 0.24923394762198245, + "grad_norm": 0.5390625, + "learning_rate": 0.0016478223182590462, + "loss": 0.126, + "step": 28712 + }, + { + "epoch": 0.24924262810218661, + "grad_norm": 0.58984375, + "learning_rate": 0.0016477987042173867, + "loss": 0.0928, + "step": 28713 + }, + { + "epoch": 0.24925130858239078, + "grad_norm": 0.08447265625, + "learning_rate": 0.0016477750895766631, + "loss": 0.0977, + "step": 28714 + }, + { + "epoch": 0.24925998906259494, + "grad_norm": 0.263671875, + "learning_rate": 0.0016477514743369007, + "loss": 0.1138, + "step": 28715 + }, + { + "epoch": 0.2492686695427991, + "grad_norm": 0.671875, + "learning_rate": 0.0016477278584981255, + "loss": 0.0762, + "step": 28716 + }, + { + "epoch": 0.24927735002300327, + "grad_norm": 0.392578125, + "learning_rate": 0.0016477042420603636, + "loss": 0.1133, + "step": 28717 + }, + { + "epoch": 0.24928603050320744, + "grad_norm": 0.404296875, + "learning_rate": 0.0016476806250236402, + "loss": 0.1016, + "step": 28718 + }, + { + "epoch": 0.2492947109834116, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016476570073879818, + "loss": 0.082, + "step": 28719 + }, + { + "epoch": 0.24930339146361577, + "grad_norm": 0.4140625, + "learning_rate": 0.0016476333891534138, + "loss": 0.1348, + "step": 28720 + }, + { + "epoch": 0.24931207194381994, + "grad_norm": 0.380859375, + "learning_rate": 0.0016476097703199619, + "loss": 0.0996, + "step": 28721 + }, + { + "epoch": 0.2493207524240241, + "grad_norm": 0.37890625, + "learning_rate": 0.0016475861508876526, + "loss": 0.1006, + "step": 28722 + }, + { + "epoch": 0.24932943290422827, + "grad_norm": 0.75390625, + "learning_rate": 0.0016475625308565113, + "loss": 0.0918, + "step": 28723 + }, + { + "epoch": 0.24933811338443243, + "grad_norm": 0.47265625, + "learning_rate": 0.0016475389102265637, + "loss": 0.0967, + "step": 28724 + }, + { + "epoch": 0.2493467938646366, + "grad_norm": 0.66796875, + "learning_rate": 0.0016475152889978354, + "loss": 0.1553, + "step": 28725 + }, + { + "epoch": 0.24935547434484076, + "grad_norm": 0.8671875, + "learning_rate": 0.001647491667170353, + "loss": 0.1533, + "step": 28726 + }, + { + "epoch": 0.24936415482504493, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016474680447441422, + "loss": 0.0889, + "step": 28727 + }, + { + "epoch": 0.2493728353052491, + "grad_norm": 0.5078125, + "learning_rate": 0.0016474444217192282, + "loss": 0.1089, + "step": 28728 + }, + { + "epoch": 0.24938151578545326, + "grad_norm": 0.625, + "learning_rate": 0.001647420798095637, + "loss": 0.0776, + "step": 28729 + }, + { + "epoch": 0.24939019626565742, + "grad_norm": 0.3984375, + "learning_rate": 0.001647397173873395, + "loss": 0.0874, + "step": 28730 + }, + { + "epoch": 0.2493988767458616, + "grad_norm": 0.92578125, + "learning_rate": 0.0016473735490525278, + "loss": 0.1357, + "step": 28731 + }, + { + "epoch": 0.24940755722606575, + "grad_norm": 0.326171875, + "learning_rate": 0.0016473499236330607, + "loss": 0.0957, + "step": 28732 + }, + { + "epoch": 0.24941623770626992, + "grad_norm": 0.302734375, + "learning_rate": 0.0016473262976150203, + "loss": 0.124, + "step": 28733 + }, + { + "epoch": 0.24942491818647408, + "grad_norm": 0.412109375, + "learning_rate": 0.0016473026709984317, + "loss": 0.063, + "step": 28734 + }, + { + "epoch": 0.24943359866667825, + "grad_norm": 0.412109375, + "learning_rate": 0.0016472790437833216, + "loss": 0.1211, + "step": 28735 + }, + { + "epoch": 0.2494422791468824, + "grad_norm": 0.396484375, + "learning_rate": 0.001647255415969715, + "loss": 0.1021, + "step": 28736 + }, + { + "epoch": 0.24945095962708658, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016472317875576381, + "loss": 0.1367, + "step": 28737 + }, + { + "epoch": 0.24945964010729074, + "grad_norm": 0.216796875, + "learning_rate": 0.001647208158547117, + "loss": 0.0942, + "step": 28738 + }, + { + "epoch": 0.2494683205874949, + "grad_norm": 0.1083984375, + "learning_rate": 0.001647184528938177, + "loss": 0.1113, + "step": 28739 + }, + { + "epoch": 0.24947700106769907, + "grad_norm": 0.07958984375, + "learning_rate": 0.0016471608987308445, + "loss": 0.084, + "step": 28740 + }, + { + "epoch": 0.24948568154790324, + "grad_norm": 0.80078125, + "learning_rate": 0.001647137267925145, + "loss": 0.0869, + "step": 28741 + }, + { + "epoch": 0.2494943620281074, + "grad_norm": 0.21484375, + "learning_rate": 0.0016471136365211042, + "loss": 0.0815, + "step": 28742 + }, + { + "epoch": 0.24950304250831157, + "grad_norm": 0.1123046875, + "learning_rate": 0.0016470900045187484, + "loss": 0.0923, + "step": 28743 + }, + { + "epoch": 0.24951172298851573, + "grad_norm": 0.29296875, + "learning_rate": 0.0016470663719181029, + "loss": 0.1191, + "step": 28744 + }, + { + "epoch": 0.2495204034687199, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016470427387191941, + "loss": 0.1055, + "step": 28745 + }, + { + "epoch": 0.24952908394892406, + "grad_norm": 0.4765625, + "learning_rate": 0.0016470191049220476, + "loss": 0.1445, + "step": 28746 + }, + { + "epoch": 0.24953776442912823, + "grad_norm": 0.294921875, + "learning_rate": 0.0016469954705266888, + "loss": 0.0933, + "step": 28747 + }, + { + "epoch": 0.2495464449093324, + "grad_norm": 0.439453125, + "learning_rate": 0.0016469718355331445, + "loss": 0.1055, + "step": 28748 + }, + { + "epoch": 0.24955512538953656, + "grad_norm": 0.1201171875, + "learning_rate": 0.00164694819994144, + "loss": 0.1025, + "step": 28749 + }, + { + "epoch": 0.24956380586974072, + "grad_norm": 0.236328125, + "learning_rate": 0.0016469245637516007, + "loss": 0.0933, + "step": 28750 + }, + { + "epoch": 0.2495724863499449, + "grad_norm": 0.1103515625, + "learning_rate": 0.0016469009269636532, + "loss": 0.0996, + "step": 28751 + }, + { + "epoch": 0.24958116683014905, + "grad_norm": 0.4375, + "learning_rate": 0.0016468772895776229, + "loss": 0.0781, + "step": 28752 + }, + { + "epoch": 0.24958984731035322, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016468536515935358, + "loss": 0.0972, + "step": 28753 + }, + { + "epoch": 0.24959852779055738, + "grad_norm": 0.80078125, + "learning_rate": 0.001646830013011418, + "loss": 0.1055, + "step": 28754 + }, + { + "epoch": 0.24960720827076155, + "grad_norm": 0.37890625, + "learning_rate": 0.0016468063738312951, + "loss": 0.166, + "step": 28755 + }, + { + "epoch": 0.24961588875096571, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016467827340531929, + "loss": 0.0613, + "step": 28756 + }, + { + "epoch": 0.24962456923116988, + "grad_norm": 0.68359375, + "learning_rate": 0.0016467590936771373, + "loss": 0.0928, + "step": 28757 + }, + { + "epoch": 0.24963324971137404, + "grad_norm": 0.07080078125, + "learning_rate": 0.0016467354527031541, + "loss": 0.0767, + "step": 28758 + }, + { + "epoch": 0.2496419301915782, + "grad_norm": 0.08935546875, + "learning_rate": 0.0016467118111312695, + "loss": 0.1201, + "step": 28759 + }, + { + "epoch": 0.24965061067178237, + "grad_norm": 0.90625, + "learning_rate": 0.0016466881689615087, + "loss": 0.1084, + "step": 28760 + }, + { + "epoch": 0.24965929115198654, + "grad_norm": 1.21875, + "learning_rate": 0.001646664526193898, + "loss": 0.0869, + "step": 28761 + }, + { + "epoch": 0.2496679716321907, + "grad_norm": 0.62109375, + "learning_rate": 0.0016466408828284635, + "loss": 0.1172, + "step": 28762 + }, + { + "epoch": 0.24967665211239487, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016466172388652307, + "loss": 0.0977, + "step": 28763 + }, + { + "epoch": 0.24968533259259904, + "grad_norm": 0.439453125, + "learning_rate": 0.0016465935943042252, + "loss": 0.0874, + "step": 28764 + }, + { + "epoch": 0.2496940130728032, + "grad_norm": 0.671875, + "learning_rate": 0.0016465699491454732, + "loss": 0.0874, + "step": 28765 + }, + { + "epoch": 0.24970269355300737, + "grad_norm": 1.0078125, + "learning_rate": 0.0016465463033890006, + "loss": 0.1973, + "step": 28766 + }, + { + "epoch": 0.24971137403321153, + "grad_norm": 0.205078125, + "learning_rate": 0.0016465226570348335, + "loss": 0.1289, + "step": 28767 + }, + { + "epoch": 0.2497200545134157, + "grad_norm": 0.12890625, + "learning_rate": 0.0016464990100829968, + "loss": 0.1138, + "step": 28768 + }, + { + "epoch": 0.24972873499361986, + "grad_norm": 0.44140625, + "learning_rate": 0.0016464753625335174, + "loss": 0.082, + "step": 28769 + }, + { + "epoch": 0.249737415473824, + "grad_norm": 0.15234375, + "learning_rate": 0.0016464517143864208, + "loss": 0.1426, + "step": 28770 + }, + { + "epoch": 0.24974609595402816, + "grad_norm": 0.2060546875, + "learning_rate": 0.0016464280656417328, + "loss": 0.1602, + "step": 28771 + }, + { + "epoch": 0.24975477643423233, + "grad_norm": 0.09814453125, + "learning_rate": 0.0016464044162994788, + "loss": 0.104, + "step": 28772 + }, + { + "epoch": 0.2497634569144365, + "grad_norm": 0.267578125, + "learning_rate": 0.0016463807663596858, + "loss": 0.0869, + "step": 28773 + }, + { + "epoch": 0.24977213739464066, + "grad_norm": 0.4765625, + "learning_rate": 0.0016463571158223786, + "loss": 0.1196, + "step": 28774 + }, + { + "epoch": 0.24978081787484482, + "grad_norm": 0.162109375, + "learning_rate": 0.0016463334646875836, + "loss": 0.0718, + "step": 28775 + }, + { + "epoch": 0.249789498355049, + "grad_norm": 0.3359375, + "learning_rate": 0.0016463098129553264, + "loss": 0.1094, + "step": 28776 + }, + { + "epoch": 0.24979817883525315, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016462861606256333, + "loss": 0.0918, + "step": 28777 + }, + { + "epoch": 0.24980685931545732, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016462625076985296, + "loss": 0.1172, + "step": 28778 + }, + { + "epoch": 0.24981553979566148, + "grad_norm": 0.119140625, + "learning_rate": 0.0016462388541740413, + "loss": 0.0801, + "step": 28779 + }, + { + "epoch": 0.24982422027586565, + "grad_norm": 0.5625, + "learning_rate": 0.0016462152000521948, + "loss": 0.1064, + "step": 28780 + }, + { + "epoch": 0.24983290075606981, + "grad_norm": 0.41796875, + "learning_rate": 0.0016461915453330155, + "loss": 0.125, + "step": 28781 + }, + { + "epoch": 0.24984158123627398, + "grad_norm": 0.1396484375, + "learning_rate": 0.001646167890016529, + "loss": 0.0933, + "step": 28782 + }, + { + "epoch": 0.24985026171647814, + "grad_norm": 0.490234375, + "learning_rate": 0.0016461442341027617, + "loss": 0.1426, + "step": 28783 + }, + { + "epoch": 0.2498589421966823, + "grad_norm": 0.15625, + "learning_rate": 0.0016461205775917395, + "loss": 0.1348, + "step": 28784 + }, + { + "epoch": 0.24986762267688647, + "grad_norm": 0.310546875, + "learning_rate": 0.0016460969204834875, + "loss": 0.0879, + "step": 28785 + }, + { + "epoch": 0.24987630315709064, + "grad_norm": 0.287109375, + "learning_rate": 0.0016460732627780325, + "loss": 0.1133, + "step": 28786 + }, + { + "epoch": 0.2498849836372948, + "grad_norm": 0.64453125, + "learning_rate": 0.0016460496044753998, + "loss": 0.1426, + "step": 28787 + }, + { + "epoch": 0.24989366411749897, + "grad_norm": 0.1455078125, + "learning_rate": 0.0016460259455756158, + "loss": 0.1201, + "step": 28788 + }, + { + "epoch": 0.24990234459770314, + "grad_norm": 0.34375, + "learning_rate": 0.0016460022860787058, + "loss": 0.0996, + "step": 28789 + }, + { + "epoch": 0.2499110250779073, + "grad_norm": 0.126953125, + "learning_rate": 0.001645978625984696, + "loss": 0.1011, + "step": 28790 + }, + { + "epoch": 0.24991970555811147, + "grad_norm": 0.22265625, + "learning_rate": 0.0016459549652936118, + "loss": 0.1416, + "step": 28791 + }, + { + "epoch": 0.24992838603831563, + "grad_norm": 0.150390625, + "learning_rate": 0.0016459313040054796, + "loss": 0.1094, + "step": 28792 + }, + { + "epoch": 0.2499370665185198, + "grad_norm": 0.306640625, + "learning_rate": 0.0016459076421203255, + "loss": 0.1348, + "step": 28793 + }, + { + "epoch": 0.24994574699872396, + "grad_norm": 0.396484375, + "learning_rate": 0.0016458839796381745, + "loss": 0.1211, + "step": 28794 + }, + { + "epoch": 0.24995442747892813, + "grad_norm": 0.1455078125, + "learning_rate": 0.0016458603165590536, + "loss": 0.0757, + "step": 28795 + }, + { + "epoch": 0.2499631079591323, + "grad_norm": 0.134765625, + "learning_rate": 0.0016458366528829874, + "loss": 0.1182, + "step": 28796 + }, + { + "epoch": 0.24997178843933646, + "grad_norm": 0.5625, + "learning_rate": 0.0016458129886100029, + "loss": 0.1084, + "step": 28797 + }, + { + "epoch": 0.24998046891954062, + "grad_norm": 0.1474609375, + "learning_rate": 0.0016457893237401251, + "loss": 0.0811, + "step": 28798 + }, + { + "epoch": 0.2499891493997448, + "grad_norm": 0.109375, + "learning_rate": 0.0016457656582733807, + "loss": 0.0991, + "step": 28799 + }, + { + "epoch": 0.24999782987994895, + "grad_norm": 0.69921875, + "learning_rate": 0.0016457419922097948, + "loss": 0.1504, + "step": 28800 + }, + { + "epoch": 0.25000651036015314, + "grad_norm": 0.2236328125, + "learning_rate": 0.001645718325549394, + "loss": 0.0884, + "step": 28801 + }, + { + "epoch": 0.2500151908403573, + "grad_norm": 0.185546875, + "learning_rate": 0.0016456946582922038, + "loss": 0.1035, + "step": 28802 + }, + { + "epoch": 0.2500238713205615, + "grad_norm": 0.310546875, + "learning_rate": 0.00164567099043825, + "loss": 0.1182, + "step": 28803 + }, + { + "epoch": 0.2500325518007656, + "grad_norm": 0.06591796875, + "learning_rate": 0.0016456473219875587, + "loss": 0.0781, + "step": 28804 + }, + { + "epoch": 0.2500412322809698, + "grad_norm": 0.310546875, + "learning_rate": 0.0016456236529401559, + "loss": 0.0977, + "step": 28805 + }, + { + "epoch": 0.25004991276117394, + "grad_norm": 0.14453125, + "learning_rate": 0.001645599983296067, + "loss": 0.1426, + "step": 28806 + }, + { + "epoch": 0.25005859324137814, + "grad_norm": 0.279296875, + "learning_rate": 0.001645576313055318, + "loss": 0.1035, + "step": 28807 + }, + { + "epoch": 0.2500672737215823, + "grad_norm": 0.1279296875, + "learning_rate": 0.001645552642217935, + "loss": 0.0977, + "step": 28808 + }, + { + "epoch": 0.25007595420178647, + "grad_norm": 0.12158203125, + "learning_rate": 0.001645528970783944, + "loss": 0.1387, + "step": 28809 + }, + { + "epoch": 0.2500846346819906, + "grad_norm": 0.216796875, + "learning_rate": 0.0016455052987533708, + "loss": 0.1084, + "step": 28810 + }, + { + "epoch": 0.2500933151621948, + "grad_norm": 0.20703125, + "learning_rate": 0.0016454816261262408, + "loss": 0.1001, + "step": 28811 + }, + { + "epoch": 0.25010199564239893, + "grad_norm": 0.138671875, + "learning_rate": 0.0016454579529025805, + "loss": 0.1191, + "step": 28812 + }, + { + "epoch": 0.2501106761226031, + "grad_norm": 0.353515625, + "learning_rate": 0.0016454342790824157, + "loss": 0.1299, + "step": 28813 + }, + { + "epoch": 0.25011935660280726, + "grad_norm": 0.20703125, + "learning_rate": 0.001645410604665772, + "loss": 0.123, + "step": 28814 + }, + { + "epoch": 0.25012803708301146, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016453869296526756, + "loss": 0.0693, + "step": 28815 + }, + { + "epoch": 0.2501367175632156, + "grad_norm": 0.11376953125, + "learning_rate": 0.0016453632540431521, + "loss": 0.1201, + "step": 28816 + }, + { + "epoch": 0.2501453980434198, + "grad_norm": 0.26953125, + "learning_rate": 0.0016453395778372275, + "loss": 0.1143, + "step": 28817 + }, + { + "epoch": 0.2501540785236239, + "grad_norm": 0.2021484375, + "learning_rate": 0.0016453159010349282, + "loss": 0.0859, + "step": 28818 + }, + { + "epoch": 0.2501627590038281, + "grad_norm": 0.244140625, + "learning_rate": 0.001645292223636279, + "loss": 0.1016, + "step": 28819 + }, + { + "epoch": 0.25017143948403225, + "grad_norm": 0.173828125, + "learning_rate": 0.0016452685456413067, + "loss": 0.124, + "step": 28820 + }, + { + "epoch": 0.25018011996423645, + "grad_norm": 0.375, + "learning_rate": 0.0016452448670500372, + "loss": 0.106, + "step": 28821 + }, + { + "epoch": 0.2501888004444406, + "grad_norm": 0.2392578125, + "learning_rate": 0.0016452211878624956, + "loss": 0.0898, + "step": 28822 + }, + { + "epoch": 0.2501974809246448, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016451975080787088, + "loss": 0.1113, + "step": 28823 + }, + { + "epoch": 0.2502061614048489, + "grad_norm": 0.58984375, + "learning_rate": 0.0016451738276987017, + "loss": 0.1094, + "step": 28824 + }, + { + "epoch": 0.2502148418850531, + "grad_norm": 0.173828125, + "learning_rate": 0.0016451501467225006, + "loss": 0.0947, + "step": 28825 + }, + { + "epoch": 0.25022352236525724, + "grad_norm": 0.2158203125, + "learning_rate": 0.0016451264651501322, + "loss": 0.0806, + "step": 28826 + }, + { + "epoch": 0.25023220284546144, + "grad_norm": 0.205078125, + "learning_rate": 0.0016451027829816212, + "loss": 0.1338, + "step": 28827 + }, + { + "epoch": 0.2502408833256656, + "grad_norm": 0.77734375, + "learning_rate": 0.001645079100216994, + "loss": 0.1055, + "step": 28828 + }, + { + "epoch": 0.25024956380586977, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016450554168562766, + "loss": 0.1348, + "step": 28829 + }, + { + "epoch": 0.2502582442860739, + "grad_norm": 0.24609375, + "learning_rate": 0.0016450317328994945, + "loss": 0.0977, + "step": 28830 + }, + { + "epoch": 0.2502669247662781, + "grad_norm": 0.123046875, + "learning_rate": 0.0016450080483466744, + "loss": 0.127, + "step": 28831 + }, + { + "epoch": 0.25027560524648224, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016449843631978413, + "loss": 0.0996, + "step": 28832 + }, + { + "epoch": 0.2502842857266864, + "grad_norm": 0.08447265625, + "learning_rate": 0.0016449606774530214, + "loss": 0.0967, + "step": 28833 + }, + { + "epoch": 0.25029296620689057, + "grad_norm": 0.220703125, + "learning_rate": 0.0016449369911122412, + "loss": 0.0854, + "step": 28834 + }, + { + "epoch": 0.2503016466870947, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016449133041755257, + "loss": 0.0938, + "step": 28835 + }, + { + "epoch": 0.2503103271672989, + "grad_norm": 0.134765625, + "learning_rate": 0.0016448896166429013, + "loss": 0.0684, + "step": 28836 + }, + { + "epoch": 0.25031900764750303, + "grad_norm": 0.287109375, + "learning_rate": 0.001644865928514394, + "loss": 0.1191, + "step": 28837 + }, + { + "epoch": 0.2503276881277072, + "grad_norm": 0.220703125, + "learning_rate": 0.001644842239790029, + "loss": 0.1074, + "step": 28838 + }, + { + "epoch": 0.25033636860791136, + "grad_norm": 0.091796875, + "learning_rate": 0.001644818550469833, + "loss": 0.1279, + "step": 28839 + }, + { + "epoch": 0.25034504908811556, + "grad_norm": 0.322265625, + "learning_rate": 0.0016447948605538317, + "loss": 0.1001, + "step": 28840 + }, + { + "epoch": 0.2503537295683197, + "grad_norm": 0.1953125, + "learning_rate": 0.0016447711700420511, + "loss": 0.1001, + "step": 28841 + }, + { + "epoch": 0.2503624100485239, + "grad_norm": 0.6171875, + "learning_rate": 0.0016447474789345164, + "loss": 0.1162, + "step": 28842 + }, + { + "epoch": 0.250371090528728, + "grad_norm": 0.30859375, + "learning_rate": 0.0016447237872312544, + "loss": 0.1465, + "step": 28843 + }, + { + "epoch": 0.2503797710089322, + "grad_norm": 0.275390625, + "learning_rate": 0.0016447000949322906, + "loss": 0.1172, + "step": 28844 + }, + { + "epoch": 0.25038845148913635, + "grad_norm": 0.283203125, + "learning_rate": 0.001644676402037651, + "loss": 0.0723, + "step": 28845 + }, + { + "epoch": 0.25039713196934055, + "grad_norm": 0.203125, + "learning_rate": 0.0016446527085473615, + "loss": 0.1016, + "step": 28846 + }, + { + "epoch": 0.2504058124495447, + "grad_norm": 0.39453125, + "learning_rate": 0.0016446290144614478, + "loss": 0.085, + "step": 28847 + }, + { + "epoch": 0.2504144929297489, + "grad_norm": 0.49609375, + "learning_rate": 0.001644605319779936, + "loss": 0.2109, + "step": 28848 + }, + { + "epoch": 0.250423173409953, + "grad_norm": 0.0888671875, + "learning_rate": 0.001644581624502852, + "loss": 0.1069, + "step": 28849 + }, + { + "epoch": 0.2504318538901572, + "grad_norm": 0.73046875, + "learning_rate": 0.0016445579286302218, + "loss": 0.1484, + "step": 28850 + }, + { + "epoch": 0.25044053437036135, + "grad_norm": 0.265625, + "learning_rate": 0.0016445342321620713, + "loss": 0.1367, + "step": 28851 + }, + { + "epoch": 0.25044921485056554, + "grad_norm": 0.29296875, + "learning_rate": 0.0016445105350984264, + "loss": 0.1299, + "step": 28852 + }, + { + "epoch": 0.2504578953307697, + "grad_norm": 0.140625, + "learning_rate": 0.0016444868374393127, + "loss": 0.1426, + "step": 28853 + }, + { + "epoch": 0.25046657581097387, + "grad_norm": 0.416015625, + "learning_rate": 0.0016444631391847567, + "loss": 0.0908, + "step": 28854 + }, + { + "epoch": 0.250475256291178, + "grad_norm": 0.4140625, + "learning_rate": 0.0016444394403347835, + "loss": 0.0977, + "step": 28855 + }, + { + "epoch": 0.2504839367713822, + "grad_norm": 0.3671875, + "learning_rate": 0.0016444157408894201, + "loss": 0.125, + "step": 28856 + }, + { + "epoch": 0.25049261725158634, + "grad_norm": 0.51171875, + "learning_rate": 0.0016443920408486913, + "loss": 0.0933, + "step": 28857 + }, + { + "epoch": 0.25050129773179053, + "grad_norm": 0.5859375, + "learning_rate": 0.001644368340212624, + "loss": 0.0869, + "step": 28858 + }, + { + "epoch": 0.25050997821199467, + "grad_norm": 0.166015625, + "learning_rate": 0.0016443446389812434, + "loss": 0.1465, + "step": 28859 + }, + { + "epoch": 0.25051865869219886, + "grad_norm": 0.57421875, + "learning_rate": 0.0016443209371545756, + "loss": 0.1113, + "step": 28860 + }, + { + "epoch": 0.250527339172403, + "grad_norm": 0.61328125, + "learning_rate": 0.0016442972347326468, + "loss": 0.1055, + "step": 28861 + }, + { + "epoch": 0.2505360196526072, + "grad_norm": 0.578125, + "learning_rate": 0.0016442735317154828, + "loss": 0.0884, + "step": 28862 + }, + { + "epoch": 0.2505447001328113, + "grad_norm": 0.33203125, + "learning_rate": 0.0016442498281031093, + "loss": 0.1084, + "step": 28863 + }, + { + "epoch": 0.2505533806130155, + "grad_norm": 0.41015625, + "learning_rate": 0.001644226123895552, + "loss": 0.1021, + "step": 28864 + }, + { + "epoch": 0.25056206109321966, + "grad_norm": 1.109375, + "learning_rate": 0.0016442024190928377, + "loss": 0.1406, + "step": 28865 + }, + { + "epoch": 0.25057074157342385, + "grad_norm": 0.130859375, + "learning_rate": 0.0016441787136949918, + "loss": 0.0942, + "step": 28866 + }, + { + "epoch": 0.250579422053628, + "grad_norm": 0.40234375, + "learning_rate": 0.00164415500770204, + "loss": 0.1309, + "step": 28867 + }, + { + "epoch": 0.2505881025338322, + "grad_norm": 0.423828125, + "learning_rate": 0.0016441313011140084, + "loss": 0.0947, + "step": 28868 + }, + { + "epoch": 0.2505967830140363, + "grad_norm": 0.1396484375, + "learning_rate": 0.0016441075939309234, + "loss": 0.0894, + "step": 28869 + }, + { + "epoch": 0.2506054634942405, + "grad_norm": 0.373046875, + "learning_rate": 0.0016440838861528102, + "loss": 0.1201, + "step": 28870 + }, + { + "epoch": 0.25061414397444465, + "grad_norm": 1.0, + "learning_rate": 0.0016440601777796952, + "loss": 0.1045, + "step": 28871 + }, + { + "epoch": 0.25062282445464884, + "grad_norm": 0.5390625, + "learning_rate": 0.0016440364688116039, + "loss": 0.1084, + "step": 28872 + }, + { + "epoch": 0.250631504934853, + "grad_norm": 0.79296875, + "learning_rate": 0.0016440127592485625, + "loss": 0.0894, + "step": 28873 + }, + { + "epoch": 0.25064018541505717, + "grad_norm": 0.5625, + "learning_rate": 0.0016439890490905973, + "loss": 0.0864, + "step": 28874 + }, + { + "epoch": 0.2506488658952613, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016439653383377335, + "loss": 0.1211, + "step": 28875 + }, + { + "epoch": 0.2506575463754655, + "grad_norm": 0.34375, + "learning_rate": 0.0016439416269899974, + "loss": 0.127, + "step": 28876 + }, + { + "epoch": 0.25066622685566964, + "grad_norm": 0.0791015625, + "learning_rate": 0.001643917915047415, + "loss": 0.105, + "step": 28877 + }, + { + "epoch": 0.25067490733587383, + "grad_norm": 0.72265625, + "learning_rate": 0.0016438942025100121, + "loss": 0.1289, + "step": 28878 + }, + { + "epoch": 0.25068358781607797, + "grad_norm": 0.10791015625, + "learning_rate": 0.001643870489377815, + "loss": 0.1025, + "step": 28879 + }, + { + "epoch": 0.25069226829628216, + "grad_norm": 0.271484375, + "learning_rate": 0.001643846775650849, + "loss": 0.0996, + "step": 28880 + }, + { + "epoch": 0.2507009487764863, + "grad_norm": 0.103515625, + "learning_rate": 0.00164382306132914, + "loss": 0.0869, + "step": 28881 + }, + { + "epoch": 0.2507096292566905, + "grad_norm": 0.111328125, + "learning_rate": 0.0016437993464127147, + "loss": 0.1035, + "step": 28882 + }, + { + "epoch": 0.25071830973689463, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016437756309015986, + "loss": 0.1108, + "step": 28883 + }, + { + "epoch": 0.2507269902170988, + "grad_norm": 0.140625, + "learning_rate": 0.0016437519147958178, + "loss": 0.1113, + "step": 28884 + }, + { + "epoch": 0.25073567069730296, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016437281980953976, + "loss": 0.1016, + "step": 28885 + }, + { + "epoch": 0.25074435117750715, + "grad_norm": 0.1474609375, + "learning_rate": 0.0016437044808003648, + "loss": 0.168, + "step": 28886 + }, + { + "epoch": 0.2507530316577113, + "grad_norm": 0.310546875, + "learning_rate": 0.0016436807629107448, + "loss": 0.1367, + "step": 28887 + }, + { + "epoch": 0.2507617121379155, + "grad_norm": 1.0859375, + "learning_rate": 0.0016436570444265637, + "loss": 0.0996, + "step": 28888 + }, + { + "epoch": 0.2507703926181196, + "grad_norm": 0.1640625, + "learning_rate": 0.0016436333253478474, + "loss": 0.0679, + "step": 28889 + }, + { + "epoch": 0.2507790730983238, + "grad_norm": 0.419921875, + "learning_rate": 0.0016436096056746219, + "loss": 0.0957, + "step": 28890 + }, + { + "epoch": 0.25078775357852795, + "grad_norm": 0.11572265625, + "learning_rate": 0.001643585885406913, + "loss": 0.1289, + "step": 28891 + }, + { + "epoch": 0.25079643405873214, + "grad_norm": 0.46484375, + "learning_rate": 0.0016435621645447469, + "loss": 0.1172, + "step": 28892 + }, + { + "epoch": 0.2508051145389363, + "grad_norm": 0.314453125, + "learning_rate": 0.001643538443088149, + "loss": 0.0947, + "step": 28893 + }, + { + "epoch": 0.2508137950191405, + "grad_norm": 0.25390625, + "learning_rate": 0.0016435147210371464, + "loss": 0.1123, + "step": 28894 + }, + { + "epoch": 0.2508224754993446, + "grad_norm": 0.1875, + "learning_rate": 0.0016434909983917636, + "loss": 0.1533, + "step": 28895 + }, + { + "epoch": 0.2508311559795488, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016434672751520277, + "loss": 0.0884, + "step": 28896 + }, + { + "epoch": 0.25083983645975294, + "grad_norm": 0.56640625, + "learning_rate": 0.0016434435513179636, + "loss": 0.0854, + "step": 28897 + }, + { + "epoch": 0.25084851693995713, + "grad_norm": 0.267578125, + "learning_rate": 0.0016434198268895982, + "loss": 0.085, + "step": 28898 + }, + { + "epoch": 0.25085719742016127, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016433961018669568, + "loss": 0.0991, + "step": 28899 + }, + { + "epoch": 0.25086587790036546, + "grad_norm": 0.2578125, + "learning_rate": 0.0016433723762500655, + "loss": 0.1055, + "step": 28900 + }, + { + "epoch": 0.2508745583805696, + "grad_norm": 0.7109375, + "learning_rate": 0.0016433486500389508, + "loss": 0.1191, + "step": 28901 + }, + { + "epoch": 0.2508832388607738, + "grad_norm": 0.63671875, + "learning_rate": 0.0016433249232336378, + "loss": 0.1099, + "step": 28902 + }, + { + "epoch": 0.25089191934097793, + "grad_norm": 0.57421875, + "learning_rate": 0.0016433011958341532, + "loss": 0.1377, + "step": 28903 + }, + { + "epoch": 0.2509005998211821, + "grad_norm": 0.271484375, + "learning_rate": 0.001643277467840522, + "loss": 0.1006, + "step": 28904 + }, + { + "epoch": 0.25090928030138626, + "grad_norm": 0.1171875, + "learning_rate": 0.0016432537392527713, + "loss": 0.1094, + "step": 28905 + }, + { + "epoch": 0.25091796078159045, + "grad_norm": 0.4296875, + "learning_rate": 0.001643230010070926, + "loss": 0.1094, + "step": 28906 + }, + { + "epoch": 0.2509266412617946, + "grad_norm": 0.36328125, + "learning_rate": 0.0016432062802950128, + "loss": 0.1328, + "step": 28907 + }, + { + "epoch": 0.2509353217419988, + "grad_norm": 0.58203125, + "learning_rate": 0.0016431825499250573, + "loss": 0.1279, + "step": 28908 + }, + { + "epoch": 0.2509440022222029, + "grad_norm": 0.19140625, + "learning_rate": 0.0016431588189610856, + "loss": 0.0996, + "step": 28909 + }, + { + "epoch": 0.2509526827024071, + "grad_norm": 0.220703125, + "learning_rate": 0.0016431350874031234, + "loss": 0.124, + "step": 28910 + }, + { + "epoch": 0.25096136318261125, + "grad_norm": 0.09716796875, + "learning_rate": 0.001643111355251197, + "loss": 0.0869, + "step": 28911 + }, + { + "epoch": 0.25097004366281545, + "grad_norm": 0.1748046875, + "learning_rate": 0.001643087622505332, + "loss": 0.1138, + "step": 28912 + }, + { + "epoch": 0.2509787241430196, + "grad_norm": 0.291015625, + "learning_rate": 0.0016430638891655547, + "loss": 0.1084, + "step": 28913 + }, + { + "epoch": 0.2509874046232238, + "grad_norm": 0.263671875, + "learning_rate": 0.001643040155231891, + "loss": 0.1318, + "step": 28914 + }, + { + "epoch": 0.2509960851034279, + "grad_norm": 0.3359375, + "learning_rate": 0.0016430164207043665, + "loss": 0.123, + "step": 28915 + }, + { + "epoch": 0.2510047655836321, + "grad_norm": 0.62890625, + "learning_rate": 0.0016429926855830073, + "loss": 0.1172, + "step": 28916 + }, + { + "epoch": 0.25101344606383624, + "grad_norm": 0.9296875, + "learning_rate": 0.0016429689498678398, + "loss": 0.063, + "step": 28917 + }, + { + "epoch": 0.25102212654404044, + "grad_norm": 0.09033203125, + "learning_rate": 0.0016429452135588896, + "loss": 0.0718, + "step": 28918 + }, + { + "epoch": 0.2510308070242446, + "grad_norm": 0.302734375, + "learning_rate": 0.0016429214766561823, + "loss": 0.1084, + "step": 28919 + }, + { + "epoch": 0.25103948750444877, + "grad_norm": 0.091796875, + "learning_rate": 0.0016428977391597446, + "loss": 0.0996, + "step": 28920 + }, + { + "epoch": 0.2510481679846529, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016428740010696018, + "loss": 0.0762, + "step": 28921 + }, + { + "epoch": 0.2510568484648571, + "grad_norm": 0.115234375, + "learning_rate": 0.0016428502623857804, + "loss": 0.1226, + "step": 28922 + }, + { + "epoch": 0.25106552894506123, + "grad_norm": 0.111328125, + "learning_rate": 0.0016428265231083062, + "loss": 0.1074, + "step": 28923 + }, + { + "epoch": 0.2510742094252654, + "grad_norm": 0.380859375, + "learning_rate": 0.0016428027832372048, + "loss": 0.1035, + "step": 28924 + }, + { + "epoch": 0.25108288990546956, + "grad_norm": 0.2451171875, + "learning_rate": 0.0016427790427725021, + "loss": 0.0776, + "step": 28925 + }, + { + "epoch": 0.25109157038567376, + "grad_norm": 0.205078125, + "learning_rate": 0.0016427553017142249, + "loss": 0.123, + "step": 28926 + }, + { + "epoch": 0.2511002508658779, + "grad_norm": 0.28515625, + "learning_rate": 0.001642731560062399, + "loss": 0.0972, + "step": 28927 + }, + { + "epoch": 0.2511089313460821, + "grad_norm": 0.369140625, + "learning_rate": 0.0016427078178170493, + "loss": 0.1133, + "step": 28928 + }, + { + "epoch": 0.2511176118262862, + "grad_norm": 0.396484375, + "learning_rate": 0.001642684074978203, + "loss": 0.0933, + "step": 28929 + }, + { + "epoch": 0.2511262923064904, + "grad_norm": 0.94140625, + "learning_rate": 0.001642660331545885, + "loss": 0.1016, + "step": 28930 + }, + { + "epoch": 0.25113497278669455, + "grad_norm": 0.3359375, + "learning_rate": 0.0016426365875201224, + "loss": 0.1084, + "step": 28931 + }, + { + "epoch": 0.25114365326689875, + "grad_norm": 0.70703125, + "learning_rate": 0.0016426128429009403, + "loss": 0.104, + "step": 28932 + }, + { + "epoch": 0.2511523337471029, + "grad_norm": 0.361328125, + "learning_rate": 0.0016425890976883648, + "loss": 0.0894, + "step": 28933 + }, + { + "epoch": 0.2511610142273071, + "grad_norm": 0.80859375, + "learning_rate": 0.0016425653518824224, + "loss": 0.1094, + "step": 28934 + }, + { + "epoch": 0.2511696947075112, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016425416054831384, + "loss": 0.0776, + "step": 28935 + }, + { + "epoch": 0.2511783751877154, + "grad_norm": 0.546875, + "learning_rate": 0.001642517858490539, + "loss": 0.0977, + "step": 28936 + }, + { + "epoch": 0.25118705566791955, + "grad_norm": 0.09814453125, + "learning_rate": 0.0016424941109046502, + "loss": 0.0972, + "step": 28937 + }, + { + "epoch": 0.25119573614812374, + "grad_norm": 0.5, + "learning_rate": 0.0016424703627254983, + "loss": 0.1025, + "step": 28938 + }, + { + "epoch": 0.2512044166283279, + "grad_norm": 0.1826171875, + "learning_rate": 0.001642446613953109, + "loss": 0.1152, + "step": 28939 + }, + { + "epoch": 0.25121309710853207, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016424228645875079, + "loss": 0.1162, + "step": 28940 + }, + { + "epoch": 0.2512217775887362, + "grad_norm": 0.10107421875, + "learning_rate": 0.0016423991146287211, + "loss": 0.1191, + "step": 28941 + }, + { + "epoch": 0.2512304580689404, + "grad_norm": 0.345703125, + "learning_rate": 0.0016423753640767752, + "loss": 0.0981, + "step": 28942 + }, + { + "epoch": 0.25123913854914454, + "grad_norm": 0.298828125, + "learning_rate": 0.0016423516129316959, + "loss": 0.1475, + "step": 28943 + }, + { + "epoch": 0.25124781902934873, + "grad_norm": 0.2421875, + "learning_rate": 0.0016423278611935087, + "loss": 0.1094, + "step": 28944 + }, + { + "epoch": 0.25125649950955287, + "grad_norm": 2.453125, + "learning_rate": 0.00164230410886224, + "loss": 0.2188, + "step": 28945 + }, + { + "epoch": 0.25126517998975706, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016422803559379155, + "loss": 0.1367, + "step": 28946 + }, + { + "epoch": 0.2512738604699612, + "grad_norm": 0.1328125, + "learning_rate": 0.0016422566024205617, + "loss": 0.1543, + "step": 28947 + }, + { + "epoch": 0.2512825409501654, + "grad_norm": 0.259765625, + "learning_rate": 0.001642232848310204, + "loss": 0.1133, + "step": 28948 + }, + { + "epoch": 0.2512912214303695, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016422090936068685, + "loss": 0.1016, + "step": 28949 + }, + { + "epoch": 0.2512999019105737, + "grad_norm": 0.6640625, + "learning_rate": 0.0016421853383105814, + "loss": 0.1123, + "step": 28950 + }, + { + "epoch": 0.25130858239077786, + "grad_norm": 0.0859375, + "learning_rate": 0.0016421615824213688, + "loss": 0.1108, + "step": 28951 + }, + { + "epoch": 0.25131726287098205, + "grad_norm": 0.34765625, + "learning_rate": 0.0016421378259392561, + "loss": 0.0928, + "step": 28952 + }, + { + "epoch": 0.2513259433511862, + "grad_norm": 0.43359375, + "learning_rate": 0.00164211406886427, + "loss": 0.1309, + "step": 28953 + }, + { + "epoch": 0.2513346238313904, + "grad_norm": 0.4921875, + "learning_rate": 0.0016420903111964355, + "loss": 0.0898, + "step": 28954 + }, + { + "epoch": 0.2513433043115945, + "grad_norm": 0.244140625, + "learning_rate": 0.0016420665529357796, + "loss": 0.0908, + "step": 28955 + }, + { + "epoch": 0.25135198479179865, + "grad_norm": 0.20703125, + "learning_rate": 0.0016420427940823277, + "loss": 0.0981, + "step": 28956 + }, + { + "epoch": 0.25136066527200285, + "grad_norm": 0.37109375, + "learning_rate": 0.001642019034636106, + "loss": 0.1162, + "step": 28957 + }, + { + "epoch": 0.251369345752207, + "grad_norm": 0.5546875, + "learning_rate": 0.0016419952745971406, + "loss": 0.127, + "step": 28958 + }, + { + "epoch": 0.2513780262324112, + "grad_norm": 0.177734375, + "learning_rate": 0.0016419715139654569, + "loss": 0.105, + "step": 28959 + }, + { + "epoch": 0.2513867067126153, + "grad_norm": 0.9375, + "learning_rate": 0.0016419477527410819, + "loss": 0.1836, + "step": 28960 + }, + { + "epoch": 0.2513953871928195, + "grad_norm": 0.115234375, + "learning_rate": 0.0016419239909240405, + "loss": 0.1523, + "step": 28961 + }, + { + "epoch": 0.25140406767302365, + "grad_norm": 0.404296875, + "learning_rate": 0.001641900228514359, + "loss": 0.1309, + "step": 28962 + }, + { + "epoch": 0.25141274815322784, + "grad_norm": 0.65625, + "learning_rate": 0.0016418764655120642, + "loss": 0.0898, + "step": 28963 + }, + { + "epoch": 0.251421428633432, + "grad_norm": 0.1513671875, + "learning_rate": 0.001641852701917181, + "loss": 0.1338, + "step": 28964 + }, + { + "epoch": 0.25143010911363617, + "grad_norm": 0.099609375, + "learning_rate": 0.001641828937729736, + "loss": 0.0728, + "step": 28965 + }, + { + "epoch": 0.2514387895938403, + "grad_norm": 0.1962890625, + "learning_rate": 0.0016418051729497549, + "loss": 0.1523, + "step": 28966 + }, + { + "epoch": 0.2514474700740445, + "grad_norm": 0.0908203125, + "learning_rate": 0.0016417814075772637, + "loss": 0.0874, + "step": 28967 + }, + { + "epoch": 0.25145615055424864, + "grad_norm": 0.169921875, + "learning_rate": 0.0016417576416122889, + "loss": 0.125, + "step": 28968 + }, + { + "epoch": 0.25146483103445283, + "grad_norm": 0.3984375, + "learning_rate": 0.0016417338750548557, + "loss": 0.1147, + "step": 28969 + }, + { + "epoch": 0.25147351151465697, + "grad_norm": 0.15625, + "learning_rate": 0.0016417101079049907, + "loss": 0.1035, + "step": 28970 + }, + { + "epoch": 0.25148219199486116, + "grad_norm": 0.162109375, + "learning_rate": 0.0016416863401627194, + "loss": 0.1113, + "step": 28971 + }, + { + "epoch": 0.2514908724750653, + "grad_norm": 0.349609375, + "learning_rate": 0.0016416625718280683, + "loss": 0.1162, + "step": 28972 + }, + { + "epoch": 0.2514995529552695, + "grad_norm": 0.35546875, + "learning_rate": 0.0016416388029010629, + "loss": 0.0996, + "step": 28973 + }, + { + "epoch": 0.2515082334354736, + "grad_norm": 0.2734375, + "learning_rate": 0.0016416150333817296, + "loss": 0.1348, + "step": 28974 + }, + { + "epoch": 0.2515169139156778, + "grad_norm": 0.072265625, + "learning_rate": 0.0016415912632700943, + "loss": 0.0781, + "step": 28975 + }, + { + "epoch": 0.25152559439588196, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016415674925661827, + "loss": 0.1143, + "step": 28976 + }, + { + "epoch": 0.25153427487608615, + "grad_norm": 0.08251953125, + "learning_rate": 0.0016415437212700214, + "loss": 0.1128, + "step": 28977 + }, + { + "epoch": 0.2515429553562903, + "grad_norm": 0.216796875, + "learning_rate": 0.0016415199493816357, + "loss": 0.0845, + "step": 28978 + }, + { + "epoch": 0.2515516358364945, + "grad_norm": 0.6171875, + "learning_rate": 0.001641496176901052, + "loss": 0.1162, + "step": 28979 + }, + { + "epoch": 0.2515603163166986, + "grad_norm": 0.21875, + "learning_rate": 0.0016414724038282965, + "loss": 0.1152, + "step": 28980 + }, + { + "epoch": 0.2515689967969028, + "grad_norm": 0.30859375, + "learning_rate": 0.0016414486301633946, + "loss": 0.1045, + "step": 28981 + }, + { + "epoch": 0.25157767727710695, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016414248559063727, + "loss": 0.0835, + "step": 28982 + }, + { + "epoch": 0.25158635775731114, + "grad_norm": 0.26953125, + "learning_rate": 0.0016414010810572566, + "loss": 0.104, + "step": 28983 + }, + { + "epoch": 0.2515950382375153, + "grad_norm": 0.59765625, + "learning_rate": 0.0016413773056160724, + "loss": 0.1338, + "step": 28984 + }, + { + "epoch": 0.25160371871771947, + "grad_norm": 0.0751953125, + "learning_rate": 0.0016413535295828462, + "loss": 0.082, + "step": 28985 + }, + { + "epoch": 0.2516123991979236, + "grad_norm": 0.8125, + "learning_rate": 0.0016413297529576044, + "loss": 0.1289, + "step": 28986 + }, + { + "epoch": 0.2516210796781278, + "grad_norm": 0.11669921875, + "learning_rate": 0.0016413059757403718, + "loss": 0.1152, + "step": 28987 + }, + { + "epoch": 0.25162976015833194, + "grad_norm": 0.76953125, + "learning_rate": 0.0016412821979311756, + "loss": 0.1001, + "step": 28988 + }, + { + "epoch": 0.25163844063853613, + "grad_norm": 0.14453125, + "learning_rate": 0.001641258419530041, + "loss": 0.1011, + "step": 28989 + }, + { + "epoch": 0.25164712111874027, + "grad_norm": 0.138671875, + "learning_rate": 0.0016412346405369944, + "loss": 0.0879, + "step": 28990 + }, + { + "epoch": 0.25165580159894446, + "grad_norm": 0.220703125, + "learning_rate": 0.0016412108609520617, + "loss": 0.0952, + "step": 28991 + }, + { + "epoch": 0.2516644820791486, + "grad_norm": 0.205078125, + "learning_rate": 0.001641187080775269, + "loss": 0.0806, + "step": 28992 + }, + { + "epoch": 0.2516731625593528, + "grad_norm": 0.390625, + "learning_rate": 0.001641163300006642, + "loss": 0.1011, + "step": 28993 + }, + { + "epoch": 0.25168184303955693, + "grad_norm": 0.8203125, + "learning_rate": 0.0016411395186462072, + "loss": 0.3164, + "step": 28994 + }, + { + "epoch": 0.2516905235197611, + "grad_norm": 0.41015625, + "learning_rate": 0.0016411157366939904, + "loss": 0.1641, + "step": 28995 + }, + { + "epoch": 0.25169920399996526, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016410919541500176, + "loss": 0.0796, + "step": 28996 + }, + { + "epoch": 0.25170788448016945, + "grad_norm": 0.10546875, + "learning_rate": 0.0016410681710143148, + "loss": 0.0835, + "step": 28997 + }, + { + "epoch": 0.2517165649603736, + "grad_norm": 0.10498046875, + "learning_rate": 0.0016410443872869073, + "loss": 0.0718, + "step": 28998 + }, + { + "epoch": 0.2517252454405778, + "grad_norm": 0.27734375, + "learning_rate": 0.0016410206029678226, + "loss": 0.084, + "step": 28999 + }, + { + "epoch": 0.2517339259207819, + "grad_norm": 0.0859375, + "learning_rate": 0.0016409968180570859, + "loss": 0.0811, + "step": 29000 + }, + { + "epoch": 0.2517426064009861, + "grad_norm": 0.375, + "learning_rate": 0.0016409730325547226, + "loss": 0.0933, + "step": 29001 + }, + { + "epoch": 0.25175128688119025, + "grad_norm": 0.373046875, + "learning_rate": 0.0016409492464607598, + "loss": 0.083, + "step": 29002 + }, + { + "epoch": 0.25175996736139444, + "grad_norm": 0.45703125, + "learning_rate": 0.0016409254597752228, + "loss": 0.0918, + "step": 29003 + }, + { + "epoch": 0.2517686478415986, + "grad_norm": 0.7265625, + "learning_rate": 0.001640901672498138, + "loss": 0.1338, + "step": 29004 + }, + { + "epoch": 0.2517773283218028, + "grad_norm": 0.25, + "learning_rate": 0.001640877884629531, + "loss": 0.1338, + "step": 29005 + }, + { + "epoch": 0.2517860088020069, + "grad_norm": 0.34765625, + "learning_rate": 0.001640854096169428, + "loss": 0.1099, + "step": 29006 + }, + { + "epoch": 0.2517946892822111, + "grad_norm": 0.421875, + "learning_rate": 0.0016408303071178555, + "loss": 0.1348, + "step": 29007 + }, + { + "epoch": 0.25180336976241524, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016408065174748393, + "loss": 0.1128, + "step": 29008 + }, + { + "epoch": 0.25181205024261943, + "grad_norm": 0.380859375, + "learning_rate": 0.0016407827272404047, + "loss": 0.1543, + "step": 29009 + }, + { + "epoch": 0.25182073072282357, + "grad_norm": 0.111328125, + "learning_rate": 0.001640758936414578, + "loss": 0.1133, + "step": 29010 + }, + { + "epoch": 0.25182941120302776, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016407351449973862, + "loss": 0.1279, + "step": 29011 + }, + { + "epoch": 0.2518380916832319, + "grad_norm": 0.3359375, + "learning_rate": 0.0016407113529888543, + "loss": 0.1016, + "step": 29012 + }, + { + "epoch": 0.2518467721634361, + "grad_norm": 0.6640625, + "learning_rate": 0.0016406875603890082, + "loss": 0.1196, + "step": 29013 + }, + { + "epoch": 0.25185545264364023, + "grad_norm": 0.123046875, + "learning_rate": 0.0016406637671978746, + "loss": 0.1055, + "step": 29014 + }, + { + "epoch": 0.2518641331238444, + "grad_norm": 0.2265625, + "learning_rate": 0.0016406399734154792, + "loss": 0.127, + "step": 29015 + }, + { + "epoch": 0.25187281360404856, + "grad_norm": 0.318359375, + "learning_rate": 0.0016406161790418483, + "loss": 0.0928, + "step": 29016 + }, + { + "epoch": 0.25188149408425275, + "grad_norm": 0.61328125, + "learning_rate": 0.0016405923840770074, + "loss": 0.1367, + "step": 29017 + }, + { + "epoch": 0.2518901745644569, + "grad_norm": 0.59765625, + "learning_rate": 0.0016405685885209828, + "loss": 0.1191, + "step": 29018 + }, + { + "epoch": 0.2518988550446611, + "grad_norm": 0.2197265625, + "learning_rate": 0.0016405447923738005, + "loss": 0.0996, + "step": 29019 + }, + { + "epoch": 0.2519075355248652, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016405209956354867, + "loss": 0.1201, + "step": 29020 + }, + { + "epoch": 0.2519162160050694, + "grad_norm": 0.1923828125, + "learning_rate": 0.001640497198306067, + "loss": 0.1855, + "step": 29021 + }, + { + "epoch": 0.25192489648527355, + "grad_norm": 0.439453125, + "learning_rate": 0.001640473400385568, + "loss": 0.0859, + "step": 29022 + }, + { + "epoch": 0.25193357696547775, + "grad_norm": 0.291015625, + "learning_rate": 0.0016404496018740155, + "loss": 0.1523, + "step": 29023 + }, + { + "epoch": 0.2519422574456819, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016404258027714349, + "loss": 0.0977, + "step": 29024 + }, + { + "epoch": 0.2519509379258861, + "grad_norm": 0.1298828125, + "learning_rate": 0.001640402003077853, + "loss": 0.0806, + "step": 29025 + }, + { + "epoch": 0.2519596184060902, + "grad_norm": 0.416015625, + "learning_rate": 0.0016403782027932957, + "loss": 0.0918, + "step": 29026 + }, + { + "epoch": 0.2519682988862944, + "grad_norm": 0.126953125, + "learning_rate": 0.0016403544019177887, + "loss": 0.125, + "step": 29027 + }, + { + "epoch": 0.25197697936649854, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016403306004513585, + "loss": 0.0781, + "step": 29028 + }, + { + "epoch": 0.25198565984670274, + "grad_norm": 0.291015625, + "learning_rate": 0.0016403067983940308, + "loss": 0.1006, + "step": 29029 + }, + { + "epoch": 0.2519943403269069, + "grad_norm": 0.205078125, + "learning_rate": 0.0016402829957458318, + "loss": 0.1182, + "step": 29030 + }, + { + "epoch": 0.25200302080711107, + "grad_norm": 0.177734375, + "learning_rate": 0.001640259192506787, + "loss": 0.0898, + "step": 29031 + }, + { + "epoch": 0.2520117012873152, + "grad_norm": 0.373046875, + "learning_rate": 0.0016402353886769232, + "loss": 0.0806, + "step": 29032 + }, + { + "epoch": 0.2520203817675194, + "grad_norm": 0.32421875, + "learning_rate": 0.0016402115842562661, + "loss": 0.1055, + "step": 29033 + }, + { + "epoch": 0.25202906224772353, + "grad_norm": 0.150390625, + "learning_rate": 0.001640187779244842, + "loss": 0.1611, + "step": 29034 + }, + { + "epoch": 0.2520377427279277, + "grad_norm": 0.09814453125, + "learning_rate": 0.001640163973642676, + "loss": 0.104, + "step": 29035 + }, + { + "epoch": 0.25204642320813186, + "grad_norm": 0.310546875, + "learning_rate": 0.0016401401674497953, + "loss": 0.1099, + "step": 29036 + }, + { + "epoch": 0.25205510368833606, + "grad_norm": 0.361328125, + "learning_rate": 0.0016401163606662252, + "loss": 0.1279, + "step": 29037 + }, + { + "epoch": 0.2520637841685402, + "grad_norm": 0.1357421875, + "learning_rate": 0.0016400925532919922, + "loss": 0.1016, + "step": 29038 + }, + { + "epoch": 0.2520724646487444, + "grad_norm": 0.765625, + "learning_rate": 0.001640068745327122, + "loss": 0.127, + "step": 29039 + }, + { + "epoch": 0.2520811451289485, + "grad_norm": 0.306640625, + "learning_rate": 0.0016400449367716405, + "loss": 0.1289, + "step": 29040 + }, + { + "epoch": 0.2520898256091527, + "grad_norm": 0.341796875, + "learning_rate": 0.0016400211276255741, + "loss": 0.1631, + "step": 29041 + }, + { + "epoch": 0.25209850608935686, + "grad_norm": 0.9296875, + "learning_rate": 0.001639997317888949, + "loss": 0.1318, + "step": 29042 + }, + { + "epoch": 0.25210718656956105, + "grad_norm": 0.23046875, + "learning_rate": 0.0016399735075617906, + "loss": 0.0811, + "step": 29043 + }, + { + "epoch": 0.2521158670497652, + "grad_norm": 0.25, + "learning_rate": 0.0016399496966441253, + "loss": 0.0923, + "step": 29044 + }, + { + "epoch": 0.2521245475299694, + "grad_norm": 0.2451171875, + "learning_rate": 0.0016399258851359792, + "loss": 0.1074, + "step": 29045 + }, + { + "epoch": 0.2521332280101735, + "grad_norm": 0.228515625, + "learning_rate": 0.0016399020730373784, + "loss": 0.1309, + "step": 29046 + }, + { + "epoch": 0.2521419084903777, + "grad_norm": 0.65625, + "learning_rate": 0.0016398782603483486, + "loss": 0.106, + "step": 29047 + }, + { + "epoch": 0.25215058897058185, + "grad_norm": 0.216796875, + "learning_rate": 0.0016398544470689162, + "loss": 0.1055, + "step": 29048 + }, + { + "epoch": 0.25215926945078604, + "grad_norm": 0.3203125, + "learning_rate": 0.0016398306331991068, + "loss": 0.1011, + "step": 29049 + }, + { + "epoch": 0.2521679499309902, + "grad_norm": 0.125, + "learning_rate": 0.001639806818738947, + "loss": 0.1172, + "step": 29050 + }, + { + "epoch": 0.25217663041119437, + "grad_norm": 0.10205078125, + "learning_rate": 0.0016397830036884626, + "loss": 0.1055, + "step": 29051 + }, + { + "epoch": 0.2521853108913985, + "grad_norm": 0.326171875, + "learning_rate": 0.0016397591880476796, + "loss": 0.0957, + "step": 29052 + }, + { + "epoch": 0.2521939913716027, + "grad_norm": 0.27734375, + "learning_rate": 0.001639735371816624, + "loss": 0.1104, + "step": 29053 + }, + { + "epoch": 0.25220267185180684, + "grad_norm": 0.578125, + "learning_rate": 0.001639711554995322, + "loss": 0.0815, + "step": 29054 + }, + { + "epoch": 0.25221135233201103, + "grad_norm": 0.2890625, + "learning_rate": 0.0016396877375837993, + "loss": 0.0996, + "step": 29055 + }, + { + "epoch": 0.25222003281221517, + "grad_norm": 0.283203125, + "learning_rate": 0.0016396639195820823, + "loss": 0.0713, + "step": 29056 + }, + { + "epoch": 0.25222871329241936, + "grad_norm": 0.416015625, + "learning_rate": 0.001639640100990197, + "loss": 0.0854, + "step": 29057 + }, + { + "epoch": 0.2522373937726235, + "grad_norm": 0.369140625, + "learning_rate": 0.0016396162818081695, + "loss": 0.1221, + "step": 29058 + }, + { + "epoch": 0.2522460742528277, + "grad_norm": 0.828125, + "learning_rate": 0.0016395924620360256, + "loss": 0.125, + "step": 29059 + }, + { + "epoch": 0.2522547547330318, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016395686416737915, + "loss": 0.0996, + "step": 29060 + }, + { + "epoch": 0.252263435213236, + "grad_norm": 0.1005859375, + "learning_rate": 0.0016395448207214935, + "loss": 0.0854, + "step": 29061 + }, + { + "epoch": 0.25227211569344016, + "grad_norm": 0.380859375, + "learning_rate": 0.001639520999179157, + "loss": 0.0957, + "step": 29062 + }, + { + "epoch": 0.25228079617364435, + "grad_norm": 0.19140625, + "learning_rate": 0.0016394971770468084, + "loss": 0.1611, + "step": 29063 + }, + { + "epoch": 0.2522894766538485, + "grad_norm": 0.201171875, + "learning_rate": 0.001639473354324474, + "loss": 0.127, + "step": 29064 + }, + { + "epoch": 0.2522981571340527, + "grad_norm": 0.1826171875, + "learning_rate": 0.00163944953101218, + "loss": 0.1143, + "step": 29065 + }, + { + "epoch": 0.2523068376142568, + "grad_norm": 0.130859375, + "learning_rate": 0.0016394257071099514, + "loss": 0.1494, + "step": 29066 + }, + { + "epoch": 0.252315518094461, + "grad_norm": 0.171875, + "learning_rate": 0.0016394018826178154, + "loss": 0.0938, + "step": 29067 + }, + { + "epoch": 0.25232419857466515, + "grad_norm": 0.216796875, + "learning_rate": 0.0016393780575357975, + "loss": 0.1201, + "step": 29068 + }, + { + "epoch": 0.25233287905486934, + "grad_norm": 0.08740234375, + "learning_rate": 0.0016393542318639239, + "loss": 0.0874, + "step": 29069 + }, + { + "epoch": 0.2523415595350735, + "grad_norm": 0.875, + "learning_rate": 0.0016393304056022207, + "loss": 0.1465, + "step": 29070 + }, + { + "epoch": 0.25235024001527767, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016393065787507135, + "loss": 0.1162, + "step": 29071 + }, + { + "epoch": 0.2523589204954818, + "grad_norm": 0.12890625, + "learning_rate": 0.001639282751309429, + "loss": 0.124, + "step": 29072 + }, + { + "epoch": 0.252367600975686, + "grad_norm": 4.9375, + "learning_rate": 0.001639258923278393, + "loss": 0.2773, + "step": 29073 + }, + { + "epoch": 0.25237628145589014, + "grad_norm": 0.408203125, + "learning_rate": 0.0016392350946576315, + "loss": 0.0918, + "step": 29074 + }, + { + "epoch": 0.25238496193609433, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016392112654471705, + "loss": 0.1196, + "step": 29075 + }, + { + "epoch": 0.25239364241629847, + "grad_norm": 0.5390625, + "learning_rate": 0.0016391874356470365, + "loss": 0.0752, + "step": 29076 + }, + { + "epoch": 0.25240232289650266, + "grad_norm": 0.2578125, + "learning_rate": 0.001639163605257255, + "loss": 0.1025, + "step": 29077 + }, + { + "epoch": 0.2524110033767068, + "grad_norm": 0.154296875, + "learning_rate": 0.001639139774277852, + "loss": 0.165, + "step": 29078 + }, + { + "epoch": 0.25241968385691094, + "grad_norm": 0.1611328125, + "learning_rate": 0.001639115942708854, + "loss": 0.1045, + "step": 29079 + }, + { + "epoch": 0.25242836433711513, + "grad_norm": 0.43359375, + "learning_rate": 0.0016390921105502872, + "loss": 0.1074, + "step": 29080 + }, + { + "epoch": 0.25243704481731927, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016390682778021773, + "loss": 0.1055, + "step": 29081 + }, + { + "epoch": 0.25244572529752346, + "grad_norm": 0.51171875, + "learning_rate": 0.0016390444444645503, + "loss": 0.0791, + "step": 29082 + }, + { + "epoch": 0.2524544057777276, + "grad_norm": 0.353515625, + "learning_rate": 0.0016390206105374324, + "loss": 0.1172, + "step": 29083 + }, + { + "epoch": 0.2524630862579318, + "grad_norm": 0.232421875, + "learning_rate": 0.0016389967760208496, + "loss": 0.0752, + "step": 29084 + }, + { + "epoch": 0.2524717667381359, + "grad_norm": 0.46484375, + "learning_rate": 0.001638972940914828, + "loss": 0.1084, + "step": 29085 + }, + { + "epoch": 0.2524804472183401, + "grad_norm": 0.53515625, + "learning_rate": 0.0016389491052193942, + "loss": 0.1475, + "step": 29086 + }, + { + "epoch": 0.25248912769854426, + "grad_norm": 0.2080078125, + "learning_rate": 0.001638925268934573, + "loss": 0.1445, + "step": 29087 + }, + { + "epoch": 0.25249780817874845, + "grad_norm": 0.62890625, + "learning_rate": 0.0016389014320603917, + "loss": 0.1084, + "step": 29088 + }, + { + "epoch": 0.2525064886589526, + "grad_norm": 0.72265625, + "learning_rate": 0.0016388775945968756, + "loss": 0.0762, + "step": 29089 + }, + { + "epoch": 0.2525151691391568, + "grad_norm": 0.193359375, + "learning_rate": 0.0016388537565440512, + "loss": 0.1426, + "step": 29090 + }, + { + "epoch": 0.2525238496193609, + "grad_norm": 0.828125, + "learning_rate": 0.0016388299179019446, + "loss": 0.125, + "step": 29091 + }, + { + "epoch": 0.2525325300995651, + "grad_norm": 0.08740234375, + "learning_rate": 0.0016388060786705816, + "loss": 0.1123, + "step": 29092 + }, + { + "epoch": 0.25254121057976925, + "grad_norm": 0.1904296875, + "learning_rate": 0.001638782238849988, + "loss": 0.0845, + "step": 29093 + }, + { + "epoch": 0.25254989105997344, + "grad_norm": 0.3828125, + "learning_rate": 0.0016387583984401907, + "loss": 0.1177, + "step": 29094 + }, + { + "epoch": 0.2525585715401776, + "grad_norm": 0.74609375, + "learning_rate": 0.0016387345574412151, + "loss": 0.1016, + "step": 29095 + }, + { + "epoch": 0.25256725202038177, + "grad_norm": 0.80859375, + "learning_rate": 0.0016387107158530876, + "loss": 0.0952, + "step": 29096 + }, + { + "epoch": 0.2525759325005859, + "grad_norm": 0.64453125, + "learning_rate": 0.001638686873675834, + "loss": 0.2285, + "step": 29097 + }, + { + "epoch": 0.2525846129807901, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016386630309094808, + "loss": 0.1191, + "step": 29098 + }, + { + "epoch": 0.25259329346099424, + "grad_norm": 0.81640625, + "learning_rate": 0.0016386391875540533, + "loss": 0.2207, + "step": 29099 + }, + { + "epoch": 0.25260197394119843, + "grad_norm": 0.2001953125, + "learning_rate": 0.0016386153436095785, + "loss": 0.0947, + "step": 29100 + }, + { + "epoch": 0.25261065442140257, + "grad_norm": 0.05712890625, + "learning_rate": 0.0016385914990760821, + "loss": 0.0601, + "step": 29101 + }, + { + "epoch": 0.25261933490160676, + "grad_norm": 0.33203125, + "learning_rate": 0.0016385676539535901, + "loss": 0.0898, + "step": 29102 + }, + { + "epoch": 0.2526280153818109, + "grad_norm": 0.35546875, + "learning_rate": 0.0016385438082421285, + "loss": 0.1123, + "step": 29103 + }, + { + "epoch": 0.2526366958620151, + "grad_norm": 0.07421875, + "learning_rate": 0.0016385199619417235, + "loss": 0.0874, + "step": 29104 + }, + { + "epoch": 0.25264537634221923, + "grad_norm": 0.36328125, + "learning_rate": 0.001638496115052401, + "loss": 0.1211, + "step": 29105 + }, + { + "epoch": 0.2526540568224234, + "grad_norm": 0.2890625, + "learning_rate": 0.0016384722675741872, + "loss": 0.0752, + "step": 29106 + }, + { + "epoch": 0.25266273730262756, + "grad_norm": 0.353515625, + "learning_rate": 0.0016384484195071086, + "loss": 0.0938, + "step": 29107 + }, + { + "epoch": 0.25267141778283175, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016384245708511904, + "loss": 0.0942, + "step": 29108 + }, + { + "epoch": 0.2526800982630359, + "grad_norm": 0.10498046875, + "learning_rate": 0.0016384007216064598, + "loss": 0.1309, + "step": 29109 + }, + { + "epoch": 0.2526887787432401, + "grad_norm": 0.162109375, + "learning_rate": 0.0016383768717729418, + "loss": 0.1436, + "step": 29110 + }, + { + "epoch": 0.2526974592234442, + "grad_norm": 0.80859375, + "learning_rate": 0.001638353021350663, + "loss": 0.1152, + "step": 29111 + }, + { + "epoch": 0.2527061397036484, + "grad_norm": 0.314453125, + "learning_rate": 0.0016383291703396495, + "loss": 0.1279, + "step": 29112 + }, + { + "epoch": 0.25271482018385255, + "grad_norm": 0.515625, + "learning_rate": 0.0016383053187399275, + "loss": 0.0723, + "step": 29113 + }, + { + "epoch": 0.25272350066405674, + "grad_norm": 1.015625, + "learning_rate": 0.0016382814665515224, + "loss": 0.1328, + "step": 29114 + }, + { + "epoch": 0.2527321811442609, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016382576137744611, + "loss": 0.0977, + "step": 29115 + }, + { + "epoch": 0.2527408616244651, + "grad_norm": 0.51953125, + "learning_rate": 0.0016382337604087694, + "loss": 0.0908, + "step": 29116 + }, + { + "epoch": 0.2527495421046692, + "grad_norm": 0.384765625, + "learning_rate": 0.0016382099064544736, + "loss": 0.0918, + "step": 29117 + }, + { + "epoch": 0.2527582225848734, + "grad_norm": 0.2890625, + "learning_rate": 0.0016381860519115987, + "loss": 0.0967, + "step": 29118 + }, + { + "epoch": 0.25276690306507754, + "grad_norm": 1.3359375, + "learning_rate": 0.0016381621967801723, + "loss": 0.1582, + "step": 29119 + }, + { + "epoch": 0.25277558354528173, + "grad_norm": 1.6796875, + "learning_rate": 0.0016381383410602197, + "loss": 0.3418, + "step": 29120 + }, + { + "epoch": 0.25278426402548587, + "grad_norm": 0.318359375, + "learning_rate": 0.0016381144847517671, + "loss": 0.0977, + "step": 29121 + }, + { + "epoch": 0.25279294450569006, + "grad_norm": 0.0986328125, + "learning_rate": 0.0016380906278548404, + "loss": 0.1123, + "step": 29122 + }, + { + "epoch": 0.2528016249858942, + "grad_norm": 0.76171875, + "learning_rate": 0.001638066770369466, + "loss": 0.0654, + "step": 29123 + }, + { + "epoch": 0.2528103054660984, + "grad_norm": 0.6640625, + "learning_rate": 0.00163804291229567, + "loss": 0.125, + "step": 29124 + }, + { + "epoch": 0.25281898594630253, + "grad_norm": 0.333984375, + "learning_rate": 0.0016380190536334782, + "loss": 0.0605, + "step": 29125 + }, + { + "epoch": 0.2528276664265067, + "grad_norm": 0.201171875, + "learning_rate": 0.0016379951943829168, + "loss": 0.1182, + "step": 29126 + }, + { + "epoch": 0.25283634690671086, + "grad_norm": 0.62109375, + "learning_rate": 0.0016379713345440119, + "loss": 0.0806, + "step": 29127 + }, + { + "epoch": 0.25284502738691506, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016379474741167897, + "loss": 0.0874, + "step": 29128 + }, + { + "epoch": 0.2528537078671192, + "grad_norm": 0.39453125, + "learning_rate": 0.0016379236131012762, + "loss": 0.0977, + "step": 29129 + }, + { + "epoch": 0.2528623883473234, + "grad_norm": 0.392578125, + "learning_rate": 0.0016378997514974979, + "loss": 0.0889, + "step": 29130 + }, + { + "epoch": 0.2528710688275275, + "grad_norm": 0.236328125, + "learning_rate": 0.00163787588930548, + "loss": 0.1211, + "step": 29131 + }, + { + "epoch": 0.2528797493077317, + "grad_norm": 0.1123046875, + "learning_rate": 0.0016378520265252492, + "loss": 0.1074, + "step": 29132 + }, + { + "epoch": 0.25288842978793585, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016378281631568316, + "loss": 0.0889, + "step": 29133 + }, + { + "epoch": 0.25289711026814005, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016378042992002536, + "loss": 0.1055, + "step": 29134 + }, + { + "epoch": 0.2529057907483442, + "grad_norm": 0.44921875, + "learning_rate": 0.00163778043465554, + "loss": 0.0815, + "step": 29135 + }, + { + "epoch": 0.2529144712285484, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016377565695227182, + "loss": 0.1035, + "step": 29136 + }, + { + "epoch": 0.2529231517087525, + "grad_norm": 0.166015625, + "learning_rate": 0.0016377327038018142, + "loss": 0.1143, + "step": 29137 + }, + { + "epoch": 0.2529318321889567, + "grad_norm": 0.361328125, + "learning_rate": 0.0016377088374928538, + "loss": 0.0986, + "step": 29138 + }, + { + "epoch": 0.25294051266916084, + "grad_norm": 0.177734375, + "learning_rate": 0.0016376849705958627, + "loss": 0.0981, + "step": 29139 + }, + { + "epoch": 0.25294919314936504, + "grad_norm": 0.25, + "learning_rate": 0.0016376611031108676, + "loss": 0.0742, + "step": 29140 + }, + { + "epoch": 0.2529578736295692, + "grad_norm": 0.1953125, + "learning_rate": 0.0016376372350378943, + "loss": 0.0923, + "step": 29141 + }, + { + "epoch": 0.25296655410977337, + "grad_norm": 0.396484375, + "learning_rate": 0.0016376133663769691, + "loss": 0.1021, + "step": 29142 + }, + { + "epoch": 0.2529752345899775, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016375894971281182, + "loss": 0.083, + "step": 29143 + }, + { + "epoch": 0.2529839150701817, + "grad_norm": 0.0830078125, + "learning_rate": 0.001637565627291367, + "loss": 0.0859, + "step": 29144 + }, + { + "epoch": 0.25299259555038583, + "grad_norm": 0.4609375, + "learning_rate": 0.0016375417568667426, + "loss": 0.1113, + "step": 29145 + }, + { + "epoch": 0.25300127603059, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016375178858542702, + "loss": 0.123, + "step": 29146 + }, + { + "epoch": 0.25300995651079417, + "grad_norm": 0.81640625, + "learning_rate": 0.0016374940142539767, + "loss": 0.1719, + "step": 29147 + }, + { + "epoch": 0.25301863699099836, + "grad_norm": 0.322265625, + "learning_rate": 0.0016374701420658873, + "loss": 0.1289, + "step": 29148 + }, + { + "epoch": 0.2530273174712025, + "grad_norm": 0.37890625, + "learning_rate": 0.0016374462692900292, + "loss": 0.1104, + "step": 29149 + }, + { + "epoch": 0.2530359979514067, + "grad_norm": 0.271484375, + "learning_rate": 0.0016374223959264276, + "loss": 0.1045, + "step": 29150 + }, + { + "epoch": 0.2530446784316108, + "grad_norm": 0.138671875, + "learning_rate": 0.0016373985219751092, + "loss": 0.1138, + "step": 29151 + }, + { + "epoch": 0.253053358911815, + "grad_norm": 0.1259765625, + "learning_rate": 0.0016373746474360996, + "loss": 0.1699, + "step": 29152 + }, + { + "epoch": 0.25306203939201916, + "grad_norm": 0.2353515625, + "learning_rate": 0.0016373507723094253, + "loss": 0.1162, + "step": 29153 + }, + { + "epoch": 0.25307071987222335, + "grad_norm": 0.2041015625, + "learning_rate": 0.001637326896595112, + "loss": 0.0981, + "step": 29154 + }, + { + "epoch": 0.2530794003524275, + "grad_norm": 0.6640625, + "learning_rate": 0.0016373030202931865, + "loss": 0.1074, + "step": 29155 + }, + { + "epoch": 0.2530880808326317, + "grad_norm": 0.251953125, + "learning_rate": 0.0016372791434036745, + "loss": 0.1099, + "step": 29156 + }, + { + "epoch": 0.2530967613128358, + "grad_norm": 0.6171875, + "learning_rate": 0.0016372552659266015, + "loss": 0.1279, + "step": 29157 + }, + { + "epoch": 0.25310544179304, + "grad_norm": 0.10302734375, + "learning_rate": 0.0016372313878619948, + "loss": 0.0786, + "step": 29158 + }, + { + "epoch": 0.25311412227324415, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016372075092098797, + "loss": 0.0674, + "step": 29159 + }, + { + "epoch": 0.25312280275344834, + "grad_norm": 0.39453125, + "learning_rate": 0.0016371836299702825, + "loss": 0.0977, + "step": 29160 + }, + { + "epoch": 0.2531314832336525, + "grad_norm": 0.48828125, + "learning_rate": 0.0016371597501432294, + "loss": 0.124, + "step": 29161 + }, + { + "epoch": 0.25314016371385667, + "grad_norm": 0.2001953125, + "learning_rate": 0.0016371358697287463, + "loss": 0.1045, + "step": 29162 + }, + { + "epoch": 0.2531488441940608, + "grad_norm": 0.412109375, + "learning_rate": 0.00163711198872686, + "loss": 0.1206, + "step": 29163 + }, + { + "epoch": 0.253157524674265, + "grad_norm": 0.2890625, + "learning_rate": 0.0016370881071375958, + "loss": 0.085, + "step": 29164 + }, + { + "epoch": 0.25316620515446914, + "grad_norm": 0.1455078125, + "learning_rate": 0.0016370642249609799, + "loss": 0.0996, + "step": 29165 + }, + { + "epoch": 0.25317488563467333, + "grad_norm": 0.259765625, + "learning_rate": 0.0016370403421970387, + "loss": 0.125, + "step": 29166 + }, + { + "epoch": 0.25318356611487747, + "grad_norm": 0.09716796875, + "learning_rate": 0.0016370164588457986, + "loss": 0.0728, + "step": 29167 + }, + { + "epoch": 0.25319224659508166, + "grad_norm": 0.08154296875, + "learning_rate": 0.0016369925749072848, + "loss": 0.0723, + "step": 29168 + }, + { + "epoch": 0.2532009270752858, + "grad_norm": 0.361328125, + "learning_rate": 0.0016369686903815247, + "loss": 0.1504, + "step": 29169 + }, + { + "epoch": 0.25320960755549, + "grad_norm": 0.84765625, + "learning_rate": 0.0016369448052685432, + "loss": 0.1143, + "step": 29170 + }, + { + "epoch": 0.25321828803569413, + "grad_norm": 0.2158203125, + "learning_rate": 0.001636920919568367, + "loss": 0.1182, + "step": 29171 + }, + { + "epoch": 0.2532269685158983, + "grad_norm": 0.177734375, + "learning_rate": 0.0016368970332810223, + "loss": 0.0859, + "step": 29172 + }, + { + "epoch": 0.25323564899610246, + "grad_norm": 0.1884765625, + "learning_rate": 0.001636873146406535, + "loss": 0.0884, + "step": 29173 + }, + { + "epoch": 0.25324432947630665, + "grad_norm": 0.2890625, + "learning_rate": 0.0016368492589449312, + "loss": 0.1006, + "step": 29174 + }, + { + "epoch": 0.2532530099565108, + "grad_norm": 0.455078125, + "learning_rate": 0.001636825370896237, + "loss": 0.1104, + "step": 29175 + }, + { + "epoch": 0.253261690436715, + "grad_norm": 0.306640625, + "learning_rate": 0.0016368014822604792, + "loss": 0.1104, + "step": 29176 + }, + { + "epoch": 0.2532703709169191, + "grad_norm": 0.77734375, + "learning_rate": 0.0016367775930376828, + "loss": 0.1035, + "step": 29177 + }, + { + "epoch": 0.2532790513971233, + "grad_norm": 0.0712890625, + "learning_rate": 0.0016367537032278747, + "loss": 0.0767, + "step": 29178 + }, + { + "epoch": 0.25328773187732745, + "grad_norm": 0.51953125, + "learning_rate": 0.0016367298128310808, + "loss": 0.1055, + "step": 29179 + }, + { + "epoch": 0.25329641235753164, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016367059218473274, + "loss": 0.1055, + "step": 29180 + }, + { + "epoch": 0.2533050928377358, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016366820302766403, + "loss": 0.1167, + "step": 29181 + }, + { + "epoch": 0.25331377331793997, + "grad_norm": 0.109375, + "learning_rate": 0.0016366581381190458, + "loss": 0.1035, + "step": 29182 + }, + { + "epoch": 0.2533224537981441, + "grad_norm": 0.302734375, + "learning_rate": 0.00163663424537457, + "loss": 0.1094, + "step": 29183 + }, + { + "epoch": 0.2533311342783483, + "grad_norm": 0.244140625, + "learning_rate": 0.001636610352043239, + "loss": 0.1299, + "step": 29184 + }, + { + "epoch": 0.25333981475855244, + "grad_norm": 0.197265625, + "learning_rate": 0.001636586458125079, + "loss": 0.0767, + "step": 29185 + }, + { + "epoch": 0.25334849523875663, + "grad_norm": 0.439453125, + "learning_rate": 0.0016365625636201163, + "loss": 0.1157, + "step": 29186 + }, + { + "epoch": 0.25335717571896077, + "grad_norm": 0.146484375, + "learning_rate": 0.0016365386685283764, + "loss": 0.1523, + "step": 29187 + }, + { + "epoch": 0.25336585619916496, + "grad_norm": 0.1083984375, + "learning_rate": 0.0016365147728498867, + "loss": 0.0903, + "step": 29188 + }, + { + "epoch": 0.2533745366793691, + "grad_norm": 0.185546875, + "learning_rate": 0.0016364908765846716, + "loss": 0.1387, + "step": 29189 + }, + { + "epoch": 0.2533832171595733, + "grad_norm": 0.490234375, + "learning_rate": 0.0016364669797327587, + "loss": 0.0781, + "step": 29190 + }, + { + "epoch": 0.25339189763977743, + "grad_norm": 0.07763671875, + "learning_rate": 0.0016364430822941734, + "loss": 0.0811, + "step": 29191 + }, + { + "epoch": 0.2534005781199816, + "grad_norm": 0.283203125, + "learning_rate": 0.001636419184268942, + "loss": 0.0938, + "step": 29192 + }, + { + "epoch": 0.25340925860018576, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016363952856570908, + "loss": 0.1191, + "step": 29193 + }, + { + "epoch": 0.25341793908038995, + "grad_norm": 0.2734375, + "learning_rate": 0.0016363713864586454, + "loss": 0.1094, + "step": 29194 + }, + { + "epoch": 0.2534266195605941, + "grad_norm": 0.640625, + "learning_rate": 0.0016363474866736328, + "loss": 0.0928, + "step": 29195 + }, + { + "epoch": 0.2534353000407983, + "grad_norm": 0.302734375, + "learning_rate": 0.001636323586302078, + "loss": 0.0815, + "step": 29196 + }, + { + "epoch": 0.2534439805210024, + "grad_norm": 0.435546875, + "learning_rate": 0.0016362996853440082, + "loss": 0.1289, + "step": 29197 + }, + { + "epoch": 0.2534526610012066, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016362757837994494, + "loss": 0.0996, + "step": 29198 + }, + { + "epoch": 0.25346134148141075, + "grad_norm": 0.1630859375, + "learning_rate": 0.001636251881668427, + "loss": 0.0869, + "step": 29199 + }, + { + "epoch": 0.25347002196161494, + "grad_norm": 0.326171875, + "learning_rate": 0.0016362279789509677, + "loss": 0.0864, + "step": 29200 + }, + { + "epoch": 0.2534787024418191, + "grad_norm": 0.28515625, + "learning_rate": 0.0016362040756470976, + "loss": 0.0938, + "step": 29201 + }, + { + "epoch": 0.2534873829220232, + "grad_norm": 0.267578125, + "learning_rate": 0.0016361801717568428, + "loss": 0.1357, + "step": 29202 + }, + { + "epoch": 0.2534960634022274, + "grad_norm": 0.224609375, + "learning_rate": 0.0016361562672802294, + "loss": 0.0947, + "step": 29203 + }, + { + "epoch": 0.25350474388243155, + "grad_norm": 0.306640625, + "learning_rate": 0.0016361323622172835, + "loss": 0.1338, + "step": 29204 + }, + { + "epoch": 0.25351342436263574, + "grad_norm": 0.1826171875, + "learning_rate": 0.001636108456568031, + "loss": 0.0957, + "step": 29205 + }, + { + "epoch": 0.2535221048428399, + "grad_norm": 0.28515625, + "learning_rate": 0.001636084550332499, + "loss": 0.1108, + "step": 29206 + }, + { + "epoch": 0.25353078532304407, + "grad_norm": 0.74609375, + "learning_rate": 0.0016360606435107128, + "loss": 0.1069, + "step": 29207 + }, + { + "epoch": 0.2535394658032482, + "grad_norm": 0.36328125, + "learning_rate": 0.0016360367361026983, + "loss": 0.0776, + "step": 29208 + }, + { + "epoch": 0.2535481462834524, + "grad_norm": 0.474609375, + "learning_rate": 0.0016360128281084825, + "loss": 0.1455, + "step": 29209 + }, + { + "epoch": 0.25355682676365654, + "grad_norm": 0.80859375, + "learning_rate": 0.001635988919528091, + "loss": 0.082, + "step": 29210 + }, + { + "epoch": 0.25356550724386073, + "grad_norm": 0.2578125, + "learning_rate": 0.0016359650103615502, + "loss": 0.1035, + "step": 29211 + }, + { + "epoch": 0.25357418772406487, + "grad_norm": 0.11181640625, + "learning_rate": 0.001635941100608886, + "loss": 0.0928, + "step": 29212 + }, + { + "epoch": 0.25358286820426906, + "grad_norm": 0.4375, + "learning_rate": 0.001635917190270125, + "loss": 0.0776, + "step": 29213 + }, + { + "epoch": 0.2535915486844732, + "grad_norm": 0.27734375, + "learning_rate": 0.0016358932793452926, + "loss": 0.0776, + "step": 29214 + }, + { + "epoch": 0.2536002291646774, + "grad_norm": 0.357421875, + "learning_rate": 0.0016358693678344154, + "loss": 0.0654, + "step": 29215 + }, + { + "epoch": 0.25360890964488153, + "grad_norm": 0.16015625, + "learning_rate": 0.0016358454557375195, + "loss": 0.0728, + "step": 29216 + }, + { + "epoch": 0.2536175901250857, + "grad_norm": 0.224609375, + "learning_rate": 0.0016358215430546312, + "loss": 0.1079, + "step": 29217 + }, + { + "epoch": 0.25362627060528986, + "grad_norm": 0.2265625, + "learning_rate": 0.001635797629785776, + "loss": 0.1328, + "step": 29218 + }, + { + "epoch": 0.25363495108549405, + "grad_norm": 0.5625, + "learning_rate": 0.0016357737159309813, + "loss": 0.0703, + "step": 29219 + }, + { + "epoch": 0.2536436315656982, + "grad_norm": 0.171875, + "learning_rate": 0.0016357498014902723, + "loss": 0.1055, + "step": 29220 + }, + { + "epoch": 0.2536523120459024, + "grad_norm": 0.291015625, + "learning_rate": 0.0016357258864636752, + "loss": 0.1167, + "step": 29221 + }, + { + "epoch": 0.2536609925261065, + "grad_norm": 0.255859375, + "learning_rate": 0.0016357019708512167, + "loss": 0.1177, + "step": 29222 + }, + { + "epoch": 0.2536696730063107, + "grad_norm": 0.1259765625, + "learning_rate": 0.001635678054652922, + "loss": 0.1133, + "step": 29223 + }, + { + "epoch": 0.25367835348651485, + "grad_norm": 0.58984375, + "learning_rate": 0.0016356541378688182, + "loss": 0.1484, + "step": 29224 + }, + { + "epoch": 0.25368703396671904, + "grad_norm": 0.458984375, + "learning_rate": 0.001635630220498931, + "loss": 0.0869, + "step": 29225 + }, + { + "epoch": 0.2536957144469232, + "grad_norm": 0.3125, + "learning_rate": 0.0016356063025432866, + "loss": 0.1128, + "step": 29226 + }, + { + "epoch": 0.2537043949271274, + "grad_norm": 0.6328125, + "learning_rate": 0.0016355823840019112, + "loss": 0.1211, + "step": 29227 + }, + { + "epoch": 0.2537130754073315, + "grad_norm": 0.5703125, + "learning_rate": 0.0016355584648748308, + "loss": 0.1348, + "step": 29228 + }, + { + "epoch": 0.2537217558875357, + "grad_norm": 0.2236328125, + "learning_rate": 0.001635534545162072, + "loss": 0.1006, + "step": 29229 + }, + { + "epoch": 0.25373043636773984, + "grad_norm": 0.38671875, + "learning_rate": 0.0016355106248636606, + "loss": 0.1016, + "step": 29230 + }, + { + "epoch": 0.25373911684794404, + "grad_norm": 0.34375, + "learning_rate": 0.0016354867039796227, + "loss": 0.0771, + "step": 29231 + }, + { + "epoch": 0.2537477973281482, + "grad_norm": 1.8671875, + "learning_rate": 0.0016354627825099848, + "loss": 0.1992, + "step": 29232 + }, + { + "epoch": 0.25375647780835237, + "grad_norm": 0.109375, + "learning_rate": 0.0016354388604547724, + "loss": 0.084, + "step": 29233 + }, + { + "epoch": 0.2537651582885565, + "grad_norm": 0.87890625, + "learning_rate": 0.0016354149378140126, + "loss": 0.1045, + "step": 29234 + }, + { + "epoch": 0.2537738387687607, + "grad_norm": 0.126953125, + "learning_rate": 0.001635391014587731, + "loss": 0.0864, + "step": 29235 + }, + { + "epoch": 0.25378251924896483, + "grad_norm": 0.1875, + "learning_rate": 0.0016353670907759538, + "loss": 0.1299, + "step": 29236 + }, + { + "epoch": 0.253791199729169, + "grad_norm": 0.1357421875, + "learning_rate": 0.0016353431663787069, + "loss": 0.1025, + "step": 29237 + }, + { + "epoch": 0.25379988020937316, + "grad_norm": 0.171875, + "learning_rate": 0.001635319241396017, + "loss": 0.1016, + "step": 29238 + }, + { + "epoch": 0.25380856068957736, + "grad_norm": 0.07568359375, + "learning_rate": 0.0016352953158279098, + "loss": 0.0879, + "step": 29239 + }, + { + "epoch": 0.2538172411697815, + "grad_norm": 0.2197265625, + "learning_rate": 0.001635271389674412, + "loss": 0.1035, + "step": 29240 + }, + { + "epoch": 0.2538259216499857, + "grad_norm": 0.51953125, + "learning_rate": 0.001635247462935549, + "loss": 0.0908, + "step": 29241 + }, + { + "epoch": 0.2538346021301898, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016352235356113481, + "loss": 0.0767, + "step": 29242 + }, + { + "epoch": 0.253843282610394, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016351996077018345, + "loss": 0.0859, + "step": 29243 + }, + { + "epoch": 0.25385196309059815, + "grad_norm": 0.357421875, + "learning_rate": 0.0016351756792070341, + "loss": 0.1064, + "step": 29244 + }, + { + "epoch": 0.25386064357080235, + "grad_norm": 0.2734375, + "learning_rate": 0.0016351517501269745, + "loss": 0.1206, + "step": 29245 + }, + { + "epoch": 0.2538693240510065, + "grad_norm": 0.3046875, + "learning_rate": 0.0016351278204616807, + "loss": 0.1309, + "step": 29246 + }, + { + "epoch": 0.2538780045312107, + "grad_norm": 0.119140625, + "learning_rate": 0.0016351038902111788, + "loss": 0.1211, + "step": 29247 + }, + { + "epoch": 0.2538866850114148, + "grad_norm": 0.193359375, + "learning_rate": 0.0016350799593754957, + "loss": 0.1055, + "step": 29248 + }, + { + "epoch": 0.253895365491619, + "grad_norm": 0.216796875, + "learning_rate": 0.0016350560279546572, + "loss": 0.1035, + "step": 29249 + }, + { + "epoch": 0.25390404597182314, + "grad_norm": 0.138671875, + "learning_rate": 0.0016350320959486891, + "loss": 0.0879, + "step": 29250 + }, + { + "epoch": 0.25391272645202734, + "grad_norm": 0.78515625, + "learning_rate": 0.001635008163357618, + "loss": 0.0742, + "step": 29251 + }, + { + "epoch": 0.2539214069322315, + "grad_norm": 0.107421875, + "learning_rate": 0.0016349842301814704, + "loss": 0.1016, + "step": 29252 + }, + { + "epoch": 0.25393008741243567, + "grad_norm": 0.259765625, + "learning_rate": 0.001634960296420272, + "loss": 0.0728, + "step": 29253 + }, + { + "epoch": 0.2539387678926398, + "grad_norm": 0.373046875, + "learning_rate": 0.001634936362074049, + "loss": 0.0747, + "step": 29254 + }, + { + "epoch": 0.253947448372844, + "grad_norm": 0.310546875, + "learning_rate": 0.0016349124271428273, + "loss": 0.125, + "step": 29255 + }, + { + "epoch": 0.25395612885304814, + "grad_norm": 0.15234375, + "learning_rate": 0.0016348884916266336, + "loss": 0.0879, + "step": 29256 + }, + { + "epoch": 0.25396480933325233, + "grad_norm": 0.12158203125, + "learning_rate": 0.0016348645555254936, + "loss": 0.0938, + "step": 29257 + }, + { + "epoch": 0.25397348981345647, + "grad_norm": 0.396484375, + "learning_rate": 0.0016348406188394344, + "loss": 0.1021, + "step": 29258 + }, + { + "epoch": 0.25398217029366066, + "grad_norm": 0.3828125, + "learning_rate": 0.0016348166815684812, + "loss": 0.1006, + "step": 29259 + }, + { + "epoch": 0.2539908507738648, + "grad_norm": 0.189453125, + "learning_rate": 0.0016347927437126606, + "loss": 0.0962, + "step": 29260 + }, + { + "epoch": 0.253999531254069, + "grad_norm": 0.2060546875, + "learning_rate": 0.0016347688052719984, + "loss": 0.1553, + "step": 29261 + }, + { + "epoch": 0.2540082117342731, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016347448662465214, + "loss": 0.1011, + "step": 29262 + }, + { + "epoch": 0.2540168922144773, + "grad_norm": 0.64453125, + "learning_rate": 0.0016347209266362555, + "loss": 0.0967, + "step": 29263 + }, + { + "epoch": 0.25402557269468146, + "grad_norm": 0.404296875, + "learning_rate": 0.0016346969864412266, + "loss": 0.0623, + "step": 29264 + }, + { + "epoch": 0.25403425317488565, + "grad_norm": 0.109375, + "learning_rate": 0.001634673045661461, + "loss": 0.084, + "step": 29265 + }, + { + "epoch": 0.2540429336550898, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016346491042969853, + "loss": 0.0874, + "step": 29266 + }, + { + "epoch": 0.254051614135294, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016346251623478253, + "loss": 0.1113, + "step": 29267 + }, + { + "epoch": 0.2540602946154981, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016346012198140071, + "loss": 0.0918, + "step": 29268 + }, + { + "epoch": 0.2540689750957023, + "grad_norm": 0.310546875, + "learning_rate": 0.0016345772766955572, + "loss": 0.124, + "step": 29269 + }, + { + "epoch": 0.25407765557590645, + "grad_norm": 0.80859375, + "learning_rate": 0.0016345533329925016, + "loss": 0.0674, + "step": 29270 + }, + { + "epoch": 0.25408633605611064, + "grad_norm": 0.287109375, + "learning_rate": 0.0016345293887048665, + "loss": 0.0898, + "step": 29271 + }, + { + "epoch": 0.2540950165363148, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016345054438326781, + "loss": 0.0776, + "step": 29272 + }, + { + "epoch": 0.25410369701651897, + "grad_norm": 0.154296875, + "learning_rate": 0.0016344814983759624, + "loss": 0.1001, + "step": 29273 + }, + { + "epoch": 0.2541123774967231, + "grad_norm": 0.09619140625, + "learning_rate": 0.001634457552334746, + "loss": 0.0781, + "step": 29274 + }, + { + "epoch": 0.2541210579769273, + "grad_norm": 0.1923828125, + "learning_rate": 0.001634433605709055, + "loss": 0.0874, + "step": 29275 + }, + { + "epoch": 0.25412973845713144, + "grad_norm": 0.2353515625, + "learning_rate": 0.0016344096584989148, + "loss": 0.1406, + "step": 29276 + }, + { + "epoch": 0.25413841893733563, + "grad_norm": 0.7265625, + "learning_rate": 0.001634385710704353, + "loss": 0.1172, + "step": 29277 + }, + { + "epoch": 0.25414709941753977, + "grad_norm": 0.1650390625, + "learning_rate": 0.0016343617623253944, + "loss": 0.0801, + "step": 29278 + }, + { + "epoch": 0.25415577989774396, + "grad_norm": 0.09423828125, + "learning_rate": 0.001634337813362066, + "loss": 0.0996, + "step": 29279 + }, + { + "epoch": 0.2541644603779481, + "grad_norm": 0.376953125, + "learning_rate": 0.0016343138638143938, + "loss": 0.0991, + "step": 29280 + }, + { + "epoch": 0.2541731408581523, + "grad_norm": 0.291015625, + "learning_rate": 0.0016342899136824042, + "loss": 0.1064, + "step": 29281 + }, + { + "epoch": 0.25418182133835643, + "grad_norm": 0.07177734375, + "learning_rate": 0.001634265962966123, + "loss": 0.0698, + "step": 29282 + }, + { + "epoch": 0.2541905018185606, + "grad_norm": 0.1962890625, + "learning_rate": 0.0016342420116655764, + "loss": 0.0811, + "step": 29283 + }, + { + "epoch": 0.25419918229876476, + "grad_norm": 0.474609375, + "learning_rate": 0.001634218059780791, + "loss": 0.085, + "step": 29284 + }, + { + "epoch": 0.25420786277896895, + "grad_norm": 0.412109375, + "learning_rate": 0.0016341941073117928, + "loss": 0.1113, + "step": 29285 + }, + { + "epoch": 0.2542165432591731, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016341701542586077, + "loss": 0.0986, + "step": 29286 + }, + { + "epoch": 0.2542252237393773, + "grad_norm": 0.259765625, + "learning_rate": 0.0016341462006212622, + "loss": 0.0933, + "step": 29287 + }, + { + "epoch": 0.2542339042195814, + "grad_norm": 0.1318359375, + "learning_rate": 0.0016341222463997825, + "loss": 0.0986, + "step": 29288 + }, + { + "epoch": 0.2542425846997856, + "grad_norm": 0.14453125, + "learning_rate": 0.001634098291594195, + "loss": 0.1191, + "step": 29289 + }, + { + "epoch": 0.25425126517998975, + "grad_norm": 0.3828125, + "learning_rate": 0.0016340743362045255, + "loss": 0.1133, + "step": 29290 + }, + { + "epoch": 0.25425994566019394, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016340503802308, + "loss": 0.0962, + "step": 29291 + }, + { + "epoch": 0.2542686261403981, + "grad_norm": 0.2421875, + "learning_rate": 0.001634026423673045, + "loss": 0.1094, + "step": 29292 + }, + { + "epoch": 0.2542773066206023, + "grad_norm": 0.494140625, + "learning_rate": 0.0016340024665312872, + "loss": 0.1113, + "step": 29293 + }, + { + "epoch": 0.2542859871008064, + "grad_norm": 0.314453125, + "learning_rate": 0.0016339785088055523, + "loss": 0.0938, + "step": 29294 + }, + { + "epoch": 0.2542946675810106, + "grad_norm": 0.2138671875, + "learning_rate": 0.001633954550495866, + "loss": 0.0801, + "step": 29295 + }, + { + "epoch": 0.25430334806121474, + "grad_norm": 1.1328125, + "learning_rate": 0.0016339305916022555, + "loss": 0.1196, + "step": 29296 + }, + { + "epoch": 0.25431202854141893, + "grad_norm": 0.40234375, + "learning_rate": 0.0016339066321247465, + "loss": 0.1309, + "step": 29297 + }, + { + "epoch": 0.25432070902162307, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016338826720633652, + "loss": 0.0942, + "step": 29298 + }, + { + "epoch": 0.25432938950182726, + "grad_norm": 0.3984375, + "learning_rate": 0.0016338587114181375, + "loss": 0.0928, + "step": 29299 + }, + { + "epoch": 0.2543380699820314, + "grad_norm": 0.380859375, + "learning_rate": 0.0016338347501890901, + "loss": 0.106, + "step": 29300 + }, + { + "epoch": 0.2543467504622356, + "grad_norm": 0.3828125, + "learning_rate": 0.0016338107883762493, + "loss": 0.0889, + "step": 29301 + }, + { + "epoch": 0.25435543094243973, + "grad_norm": 0.1416015625, + "learning_rate": 0.001633786825979641, + "loss": 0.124, + "step": 29302 + }, + { + "epoch": 0.2543641114226439, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016337628629992914, + "loss": 0.0791, + "step": 29303 + }, + { + "epoch": 0.25437279190284806, + "grad_norm": 0.30859375, + "learning_rate": 0.0016337388994352264, + "loss": 0.1025, + "step": 29304 + }, + { + "epoch": 0.25438147238305225, + "grad_norm": 0.09228515625, + "learning_rate": 0.0016337149352874728, + "loss": 0.0967, + "step": 29305 + }, + { + "epoch": 0.2543901528632564, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016336909705560567, + "loss": 0.0977, + "step": 29306 + }, + { + "epoch": 0.2543988333434606, + "grad_norm": 0.330078125, + "learning_rate": 0.0016336670052410044, + "loss": 0.1016, + "step": 29307 + }, + { + "epoch": 0.2544075138236647, + "grad_norm": 0.1103515625, + "learning_rate": 0.001633643039342341, + "loss": 0.1143, + "step": 29308 + }, + { + "epoch": 0.2544161943038689, + "grad_norm": 0.14453125, + "learning_rate": 0.001633619072860094, + "loss": 0.0889, + "step": 29309 + }, + { + "epoch": 0.25442487478407305, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016335951057942896, + "loss": 0.1348, + "step": 29310 + }, + { + "epoch": 0.25443355526427724, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016335711381449535, + "loss": 0.1123, + "step": 29311 + }, + { + "epoch": 0.2544422357444814, + "grad_norm": 0.37890625, + "learning_rate": 0.0016335471699121119, + "loss": 0.0898, + "step": 29312 + }, + { + "epoch": 0.2544509162246856, + "grad_norm": 0.1982421875, + "learning_rate": 0.0016335232010957909, + "loss": 0.1055, + "step": 29313 + }, + { + "epoch": 0.2544595967048897, + "grad_norm": 0.6484375, + "learning_rate": 0.001633499231696017, + "loss": 0.0903, + "step": 29314 + }, + { + "epoch": 0.2544682771850939, + "grad_norm": 0.125, + "learning_rate": 0.001633475261712817, + "loss": 0.1021, + "step": 29315 + }, + { + "epoch": 0.25447695766529804, + "grad_norm": 0.3671875, + "learning_rate": 0.0016334512911462156, + "loss": 0.1011, + "step": 29316 + }, + { + "epoch": 0.25448563814550224, + "grad_norm": 0.76171875, + "learning_rate": 0.0016334273199962402, + "loss": 0.2715, + "step": 29317 + }, + { + "epoch": 0.2544943186257064, + "grad_norm": 0.25390625, + "learning_rate": 0.0016334033482629168, + "loss": 0.125, + "step": 29318 + }, + { + "epoch": 0.25450299910591057, + "grad_norm": 0.26171875, + "learning_rate": 0.0016333793759462714, + "loss": 0.1768, + "step": 29319 + }, + { + "epoch": 0.2545116795861147, + "grad_norm": 0.2578125, + "learning_rate": 0.0016333554030463304, + "loss": 0.0957, + "step": 29320 + }, + { + "epoch": 0.2545203600663189, + "grad_norm": 0.1474609375, + "learning_rate": 0.00163333142956312, + "loss": 0.1128, + "step": 29321 + }, + { + "epoch": 0.25452904054652303, + "grad_norm": 0.515625, + "learning_rate": 0.0016333074554966663, + "loss": 0.1035, + "step": 29322 + }, + { + "epoch": 0.2545377210267272, + "grad_norm": 0.10888671875, + "learning_rate": 0.0016332834808469954, + "loss": 0.0947, + "step": 29323 + }, + { + "epoch": 0.25454640150693136, + "grad_norm": 0.23828125, + "learning_rate": 0.001633259505614134, + "loss": 0.1338, + "step": 29324 + }, + { + "epoch": 0.2545550819871355, + "grad_norm": 0.21875, + "learning_rate": 0.0016332355297981078, + "loss": 0.0786, + "step": 29325 + }, + { + "epoch": 0.2545637624673397, + "grad_norm": 0.189453125, + "learning_rate": 0.0016332115533989435, + "loss": 0.1406, + "step": 29326 + }, + { + "epoch": 0.25457244294754383, + "grad_norm": 0.361328125, + "learning_rate": 0.001633187576416667, + "loss": 0.1582, + "step": 29327 + }, + { + "epoch": 0.254581123427748, + "grad_norm": 0.48828125, + "learning_rate": 0.0016331635988513042, + "loss": 0.1426, + "step": 29328 + }, + { + "epoch": 0.25458980390795216, + "grad_norm": 0.0986328125, + "learning_rate": 0.001633139620702882, + "loss": 0.1289, + "step": 29329 + }, + { + "epoch": 0.25459848438815635, + "grad_norm": 0.45703125, + "learning_rate": 0.0016331156419714262, + "loss": 0.1006, + "step": 29330 + }, + { + "epoch": 0.2546071648683605, + "grad_norm": 0.88671875, + "learning_rate": 0.0016330916626569633, + "loss": 0.1143, + "step": 29331 + }, + { + "epoch": 0.2546158453485647, + "grad_norm": 0.76953125, + "learning_rate": 0.0016330676827595193, + "loss": 0.0898, + "step": 29332 + }, + { + "epoch": 0.2546245258287688, + "grad_norm": 0.18359375, + "learning_rate": 0.0016330437022791205, + "loss": 0.0957, + "step": 29333 + }, + { + "epoch": 0.254633206308973, + "grad_norm": 0.21875, + "learning_rate": 0.0016330197212157931, + "loss": 0.1108, + "step": 29334 + }, + { + "epoch": 0.25464188678917715, + "grad_norm": 0.380859375, + "learning_rate": 0.0016329957395695633, + "loss": 0.1113, + "step": 29335 + }, + { + "epoch": 0.25465056726938134, + "grad_norm": 0.1220703125, + "learning_rate": 0.0016329717573404575, + "loss": 0.1074, + "step": 29336 + }, + { + "epoch": 0.2546592477495855, + "grad_norm": 0.31640625, + "learning_rate": 0.0016329477745285015, + "loss": 0.1133, + "step": 29337 + }, + { + "epoch": 0.2546679282297897, + "grad_norm": 0.59375, + "learning_rate": 0.0016329237911337222, + "loss": 0.1152, + "step": 29338 + }, + { + "epoch": 0.2546766087099938, + "grad_norm": 0.470703125, + "learning_rate": 0.0016328998071561452, + "loss": 0.0898, + "step": 29339 + }, + { + "epoch": 0.254685289190198, + "grad_norm": 0.388671875, + "learning_rate": 0.001632875822595797, + "loss": 0.0874, + "step": 29340 + }, + { + "epoch": 0.25469396967040214, + "grad_norm": 0.140625, + "learning_rate": 0.001632851837452704, + "loss": 0.1328, + "step": 29341 + }, + { + "epoch": 0.25470265015060634, + "grad_norm": 0.177734375, + "learning_rate": 0.0016328278517268922, + "loss": 0.0967, + "step": 29342 + }, + { + "epoch": 0.2547113306308105, + "grad_norm": 0.490234375, + "learning_rate": 0.0016328038654183875, + "loss": 0.0752, + "step": 29343 + }, + { + "epoch": 0.25472001111101467, + "grad_norm": 0.2890625, + "learning_rate": 0.0016327798785272167, + "loss": 0.084, + "step": 29344 + }, + { + "epoch": 0.2547286915912188, + "grad_norm": 0.1796875, + "learning_rate": 0.0016327558910534062, + "loss": 0.1055, + "step": 29345 + }, + { + "epoch": 0.254737372071423, + "grad_norm": 0.2197265625, + "learning_rate": 0.0016327319029969811, + "loss": 0.1235, + "step": 29346 + }, + { + "epoch": 0.25474605255162713, + "grad_norm": 0.30078125, + "learning_rate": 0.0016327079143579688, + "loss": 0.0635, + "step": 29347 + }, + { + "epoch": 0.2547547330318313, + "grad_norm": 0.72265625, + "learning_rate": 0.0016326839251363952, + "loss": 0.1426, + "step": 29348 + }, + { + "epoch": 0.25476341351203546, + "grad_norm": 0.1220703125, + "learning_rate": 0.0016326599353322867, + "loss": 0.1079, + "step": 29349 + }, + { + "epoch": 0.25477209399223966, + "grad_norm": 0.36328125, + "learning_rate": 0.0016326359449456691, + "loss": 0.1455, + "step": 29350 + }, + { + "epoch": 0.2547807744724438, + "grad_norm": 0.12109375, + "learning_rate": 0.0016326119539765687, + "loss": 0.0825, + "step": 29351 + }, + { + "epoch": 0.254789454952648, + "grad_norm": 0.310546875, + "learning_rate": 0.0016325879624250117, + "loss": 0.0913, + "step": 29352 + }, + { + "epoch": 0.2547981354328521, + "grad_norm": 0.44140625, + "learning_rate": 0.001632563970291025, + "loss": 0.0781, + "step": 29353 + }, + { + "epoch": 0.2548068159130563, + "grad_norm": 0.7890625, + "learning_rate": 0.0016325399775746342, + "loss": 0.1064, + "step": 29354 + }, + { + "epoch": 0.25481549639326045, + "grad_norm": 0.1708984375, + "learning_rate": 0.0016325159842758654, + "loss": 0.1196, + "step": 29355 + }, + { + "epoch": 0.25482417687346465, + "grad_norm": 0.53125, + "learning_rate": 0.001632491990394745, + "loss": 0.0889, + "step": 29356 + }, + { + "epoch": 0.2548328573536688, + "grad_norm": 0.90625, + "learning_rate": 0.0016324679959312997, + "loss": 0.1113, + "step": 29357 + }, + { + "epoch": 0.254841537833873, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016324440008855553, + "loss": 0.1055, + "step": 29358 + }, + { + "epoch": 0.2548502183140771, + "grad_norm": 0.1103515625, + "learning_rate": 0.0016324200052575384, + "loss": 0.0967, + "step": 29359 + }, + { + "epoch": 0.2548588987942813, + "grad_norm": 0.625, + "learning_rate": 0.0016323960090472748, + "loss": 0.1328, + "step": 29360 + }, + { + "epoch": 0.25486757927448545, + "grad_norm": 0.26171875, + "learning_rate": 0.001632372012254791, + "loss": 0.0918, + "step": 29361 + }, + { + "epoch": 0.25487625975468964, + "grad_norm": 0.171875, + "learning_rate": 0.0016323480148801128, + "loss": 0.0933, + "step": 29362 + }, + { + "epoch": 0.2548849402348938, + "grad_norm": 0.55078125, + "learning_rate": 0.0016323240169232671, + "loss": 0.1338, + "step": 29363 + }, + { + "epoch": 0.25489362071509797, + "grad_norm": 0.08203125, + "learning_rate": 0.00163230001838428, + "loss": 0.0938, + "step": 29364 + }, + { + "epoch": 0.2549023011953021, + "grad_norm": 0.08984375, + "learning_rate": 0.0016322760192631774, + "loss": 0.0972, + "step": 29365 + }, + { + "epoch": 0.2549109816755063, + "grad_norm": 0.52734375, + "learning_rate": 0.0016322520195599856, + "loss": 0.1484, + "step": 29366 + }, + { + "epoch": 0.25491966215571044, + "grad_norm": 0.1015625, + "learning_rate": 0.0016322280192747314, + "loss": 0.1143, + "step": 29367 + }, + { + "epoch": 0.25492834263591463, + "grad_norm": 0.0791015625, + "learning_rate": 0.0016322040184074402, + "loss": 0.0903, + "step": 29368 + }, + { + "epoch": 0.25493702311611877, + "grad_norm": 0.26953125, + "learning_rate": 0.0016321800169581389, + "loss": 0.1016, + "step": 29369 + }, + { + "epoch": 0.25494570359632296, + "grad_norm": 0.66796875, + "learning_rate": 0.0016321560149268534, + "loss": 0.1108, + "step": 29370 + }, + { + "epoch": 0.2549543840765271, + "grad_norm": 0.34375, + "learning_rate": 0.0016321320123136103, + "loss": 0.082, + "step": 29371 + }, + { + "epoch": 0.2549630645567313, + "grad_norm": 0.345703125, + "learning_rate": 0.0016321080091184356, + "loss": 0.0752, + "step": 29372 + }, + { + "epoch": 0.2549717450369354, + "grad_norm": 0.3828125, + "learning_rate": 0.0016320840053413552, + "loss": 0.1221, + "step": 29373 + }, + { + "epoch": 0.2549804255171396, + "grad_norm": 0.1240234375, + "learning_rate": 0.0016320600009823963, + "loss": 0.1543, + "step": 29374 + }, + { + "epoch": 0.25498910599734376, + "grad_norm": 0.09130859375, + "learning_rate": 0.001632035996041584, + "loss": 0.123, + "step": 29375 + }, + { + "epoch": 0.25499778647754795, + "grad_norm": 0.279296875, + "learning_rate": 0.0016320119905189453, + "loss": 0.1621, + "step": 29376 + }, + { + "epoch": 0.2550064669577521, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016319879844145064, + "loss": 0.1074, + "step": 29377 + }, + { + "epoch": 0.2550151474379563, + "grad_norm": 0.2265625, + "learning_rate": 0.0016319639777282935, + "loss": 0.1006, + "step": 29378 + }, + { + "epoch": 0.2550238279181604, + "grad_norm": 0.330078125, + "learning_rate": 0.001631939970460333, + "loss": 0.0996, + "step": 29379 + }, + { + "epoch": 0.2550325083983646, + "grad_norm": 0.423828125, + "learning_rate": 0.0016319159626106505, + "loss": 0.1006, + "step": 29380 + }, + { + "epoch": 0.25504118887856875, + "grad_norm": 0.134765625, + "learning_rate": 0.001631891954179273, + "loss": 0.0908, + "step": 29381 + }, + { + "epoch": 0.25504986935877294, + "grad_norm": 0.67578125, + "learning_rate": 0.0016318679451662263, + "loss": 0.1172, + "step": 29382 + }, + { + "epoch": 0.2550585498389771, + "grad_norm": 0.66796875, + "learning_rate": 0.0016318439355715365, + "loss": 0.1543, + "step": 29383 + }, + { + "epoch": 0.25506723031918127, + "grad_norm": 0.29296875, + "learning_rate": 0.0016318199253952306, + "loss": 0.1035, + "step": 29384 + }, + { + "epoch": 0.2550759107993854, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016317959146373343, + "loss": 0.1719, + "step": 29385 + }, + { + "epoch": 0.2550845912795896, + "grad_norm": 0.318359375, + "learning_rate": 0.001631771903297874, + "loss": 0.1182, + "step": 29386 + }, + { + "epoch": 0.25509327175979374, + "grad_norm": 0.58203125, + "learning_rate": 0.0016317478913768758, + "loss": 0.1001, + "step": 29387 + }, + { + "epoch": 0.25510195223999793, + "grad_norm": 0.365234375, + "learning_rate": 0.0016317238788743661, + "loss": 0.1133, + "step": 29388 + }, + { + "epoch": 0.25511063272020207, + "grad_norm": 0.2314453125, + "learning_rate": 0.0016316998657903714, + "loss": 0.1172, + "step": 29389 + }, + { + "epoch": 0.25511931320040626, + "grad_norm": 0.396484375, + "learning_rate": 0.0016316758521249173, + "loss": 0.1777, + "step": 29390 + }, + { + "epoch": 0.2551279936806104, + "grad_norm": 0.09130859375, + "learning_rate": 0.0016316518378780309, + "loss": 0.1011, + "step": 29391 + }, + { + "epoch": 0.2551366741608146, + "grad_norm": 0.58203125, + "learning_rate": 0.0016316278230497378, + "loss": 0.1348, + "step": 29392 + }, + { + "epoch": 0.25514535464101873, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016316038076400648, + "loss": 0.0928, + "step": 29393 + }, + { + "epoch": 0.2551540351212229, + "grad_norm": 0.396484375, + "learning_rate": 0.0016315797916490375, + "loss": 0.1216, + "step": 29394 + }, + { + "epoch": 0.25516271560142706, + "grad_norm": 0.73828125, + "learning_rate": 0.0016315557750766828, + "loss": 0.1133, + "step": 29395 + }, + { + "epoch": 0.25517139608163125, + "grad_norm": 0.3203125, + "learning_rate": 0.0016315317579230265, + "loss": 0.1484, + "step": 29396 + }, + { + "epoch": 0.2551800765618354, + "grad_norm": 0.111328125, + "learning_rate": 0.0016315077401880954, + "loss": 0.1152, + "step": 29397 + }, + { + "epoch": 0.2551887570420396, + "grad_norm": 0.470703125, + "learning_rate": 0.0016314837218719149, + "loss": 0.1055, + "step": 29398 + }, + { + "epoch": 0.2551974375222437, + "grad_norm": 0.1328125, + "learning_rate": 0.001631459702974512, + "loss": 0.0723, + "step": 29399 + }, + { + "epoch": 0.2552061180024479, + "grad_norm": 0.984375, + "learning_rate": 0.0016314356834959125, + "loss": 0.1016, + "step": 29400 + }, + { + "epoch": 0.25521479848265205, + "grad_norm": 0.251953125, + "learning_rate": 0.0016314116634361434, + "loss": 0.125, + "step": 29401 + }, + { + "epoch": 0.25522347896285624, + "grad_norm": 0.390625, + "learning_rate": 0.0016313876427952304, + "loss": 0.1074, + "step": 29402 + }, + { + "epoch": 0.2552321594430604, + "grad_norm": 0.173828125, + "learning_rate": 0.0016313636215731994, + "loss": 0.1128, + "step": 29403 + }, + { + "epoch": 0.2552408399232646, + "grad_norm": 0.412109375, + "learning_rate": 0.0016313395997700777, + "loss": 0.124, + "step": 29404 + }, + { + "epoch": 0.2552495204034687, + "grad_norm": 0.310546875, + "learning_rate": 0.0016313155773858908, + "loss": 0.1367, + "step": 29405 + }, + { + "epoch": 0.2552582008836729, + "grad_norm": 0.09716796875, + "learning_rate": 0.0016312915544206652, + "loss": 0.085, + "step": 29406 + }, + { + "epoch": 0.25526688136387704, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016312675308744271, + "loss": 0.1514, + "step": 29407 + }, + { + "epoch": 0.25527556184408123, + "grad_norm": 0.578125, + "learning_rate": 0.0016312435067472027, + "loss": 0.1309, + "step": 29408 + }, + { + "epoch": 0.25528424232428537, + "grad_norm": 0.97265625, + "learning_rate": 0.0016312194820390185, + "loss": 0.1494, + "step": 29409 + }, + { + "epoch": 0.25529292280448956, + "grad_norm": 0.279296875, + "learning_rate": 0.0016311954567499007, + "loss": 0.1094, + "step": 29410 + }, + { + "epoch": 0.2553016032846937, + "grad_norm": 0.31640625, + "learning_rate": 0.0016311714308798752, + "loss": 0.1069, + "step": 29411 + }, + { + "epoch": 0.2553102837648979, + "grad_norm": 0.1435546875, + "learning_rate": 0.001631147404428969, + "loss": 0.0977, + "step": 29412 + }, + { + "epoch": 0.25531896424510203, + "grad_norm": 0.7578125, + "learning_rate": 0.0016311233773972078, + "loss": 0.1328, + "step": 29413 + }, + { + "epoch": 0.2553276447253062, + "grad_norm": 0.8359375, + "learning_rate": 0.0016310993497846184, + "loss": 0.127, + "step": 29414 + }, + { + "epoch": 0.25533632520551036, + "grad_norm": 0.126953125, + "learning_rate": 0.0016310753215912261, + "loss": 0.1494, + "step": 29415 + }, + { + "epoch": 0.25534500568571455, + "grad_norm": 0.12890625, + "learning_rate": 0.001631051292817058, + "loss": 0.0889, + "step": 29416 + }, + { + "epoch": 0.2553536861659187, + "grad_norm": 0.296875, + "learning_rate": 0.0016310272634621405, + "loss": 0.0884, + "step": 29417 + }, + { + "epoch": 0.2553623666461229, + "grad_norm": 0.25390625, + "learning_rate": 0.0016310032335264996, + "loss": 0.1055, + "step": 29418 + }, + { + "epoch": 0.255371047126327, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016309792030101613, + "loss": 0.1299, + "step": 29419 + }, + { + "epoch": 0.2553797276065312, + "grad_norm": 0.279296875, + "learning_rate": 0.001630955171913152, + "loss": 0.0879, + "step": 29420 + }, + { + "epoch": 0.25538840808673535, + "grad_norm": 0.1572265625, + "learning_rate": 0.0016309311402354983, + "loss": 0.0928, + "step": 29421 + }, + { + "epoch": 0.25539708856693955, + "grad_norm": 0.58203125, + "learning_rate": 0.001630907107977226, + "loss": 0.1108, + "step": 29422 + }, + { + "epoch": 0.2554057690471437, + "grad_norm": 0.1201171875, + "learning_rate": 0.0016308830751383622, + "loss": 0.1187, + "step": 29423 + }, + { + "epoch": 0.2554144495273479, + "grad_norm": 0.44140625, + "learning_rate": 0.0016308590417189323, + "loss": 0.0874, + "step": 29424 + }, + { + "epoch": 0.255423130007552, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016308350077189626, + "loss": 0.1035, + "step": 29425 + }, + { + "epoch": 0.2554318104877562, + "grad_norm": 0.1474609375, + "learning_rate": 0.00163081097313848, + "loss": 0.1172, + "step": 29426 + }, + { + "epoch": 0.25544049096796034, + "grad_norm": 0.25390625, + "learning_rate": 0.001630786937977511, + "loss": 0.1445, + "step": 29427 + }, + { + "epoch": 0.25544917144816454, + "grad_norm": 0.171875, + "learning_rate": 0.001630762902236081, + "loss": 0.0806, + "step": 29428 + }, + { + "epoch": 0.2554578519283687, + "grad_norm": 0.07763671875, + "learning_rate": 0.0016307388659142167, + "loss": 0.0713, + "step": 29429 + }, + { + "epoch": 0.25546653240857287, + "grad_norm": 0.68359375, + "learning_rate": 0.001630714829011944, + "loss": 0.1123, + "step": 29430 + }, + { + "epoch": 0.255475212888777, + "grad_norm": 0.205078125, + "learning_rate": 0.00163069079152929, + "loss": 0.1348, + "step": 29431 + }, + { + "epoch": 0.2554838933689812, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016306667534662803, + "loss": 0.0801, + "step": 29432 + }, + { + "epoch": 0.25549257384918533, + "grad_norm": 0.3203125, + "learning_rate": 0.0016306427148229411, + "loss": 0.0957, + "step": 29433 + }, + { + "epoch": 0.2555012543293895, + "grad_norm": 0.25, + "learning_rate": 0.0016306186755992995, + "loss": 0.1172, + "step": 29434 + }, + { + "epoch": 0.25550993480959366, + "grad_norm": 0.1591796875, + "learning_rate": 0.0016305946357953812, + "loss": 0.1123, + "step": 29435 + }, + { + "epoch": 0.25551861528979786, + "grad_norm": 0.1875, + "learning_rate": 0.0016305705954112128, + "loss": 0.1006, + "step": 29436 + }, + { + "epoch": 0.255527295770002, + "grad_norm": 0.1796875, + "learning_rate": 0.00163054655444682, + "loss": 0.1309, + "step": 29437 + }, + { + "epoch": 0.2555359762502062, + "grad_norm": 0.5859375, + "learning_rate": 0.0016305225129022293, + "loss": 0.1309, + "step": 29438 + }, + { + "epoch": 0.2555446567304103, + "grad_norm": 0.58984375, + "learning_rate": 0.0016304984707774673, + "loss": 0.0918, + "step": 29439 + }, + { + "epoch": 0.2555533372106145, + "grad_norm": 0.83203125, + "learning_rate": 0.0016304744280725603, + "loss": 0.1035, + "step": 29440 + }, + { + "epoch": 0.25556201769081865, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016304503847875344, + "loss": 0.0938, + "step": 29441 + }, + { + "epoch": 0.25557069817102285, + "grad_norm": 0.208984375, + "learning_rate": 0.0016304263409224159, + "loss": 0.1328, + "step": 29442 + }, + { + "epoch": 0.255579378651227, + "grad_norm": 0.43359375, + "learning_rate": 0.0016304022964772312, + "loss": 0.1094, + "step": 29443 + }, + { + "epoch": 0.2555880591314312, + "grad_norm": 0.474609375, + "learning_rate": 0.0016303782514520062, + "loss": 0.1289, + "step": 29444 + }, + { + "epoch": 0.2555967396116353, + "grad_norm": 0.83984375, + "learning_rate": 0.0016303542058467679, + "loss": 0.1206, + "step": 29445 + }, + { + "epoch": 0.2556054200918395, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016303301596615422, + "loss": 0.1006, + "step": 29446 + }, + { + "epoch": 0.25561410057204365, + "grad_norm": 0.173828125, + "learning_rate": 0.001630306112896355, + "loss": 0.167, + "step": 29447 + }, + { + "epoch": 0.2556227810522478, + "grad_norm": 0.2099609375, + "learning_rate": 0.001630282065551233, + "loss": 0.0986, + "step": 29448 + }, + { + "epoch": 0.255631461532452, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016302580176262031, + "loss": 0.0815, + "step": 29449 + }, + { + "epoch": 0.2556401420126561, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016302339691212906, + "loss": 0.1211, + "step": 29450 + }, + { + "epoch": 0.2556488224928603, + "grad_norm": 0.103515625, + "learning_rate": 0.0016302099200365223, + "loss": 0.1279, + "step": 29451 + }, + { + "epoch": 0.25565750297306444, + "grad_norm": 0.12060546875, + "learning_rate": 0.0016301858703719244, + "loss": 0.0825, + "step": 29452 + }, + { + "epoch": 0.25566618345326864, + "grad_norm": 0.1220703125, + "learning_rate": 0.001630161820127523, + "loss": 0.0967, + "step": 29453 + }, + { + "epoch": 0.2556748639334728, + "grad_norm": 0.259765625, + "learning_rate": 0.0016301377693033448, + "loss": 0.0952, + "step": 29454 + }, + { + "epoch": 0.25568354441367697, + "grad_norm": 0.1513671875, + "learning_rate": 0.001630113717899416, + "loss": 0.1074, + "step": 29455 + }, + { + "epoch": 0.2556922248938811, + "grad_norm": 0.58203125, + "learning_rate": 0.0016300896659157628, + "loss": 0.125, + "step": 29456 + }, + { + "epoch": 0.2557009053740853, + "grad_norm": 3.171875, + "learning_rate": 0.0016300656133524112, + "loss": 0.4102, + "step": 29457 + }, + { + "epoch": 0.25570958585428943, + "grad_norm": 0.431640625, + "learning_rate": 0.001630041560209388, + "loss": 0.0952, + "step": 29458 + }, + { + "epoch": 0.2557182663344936, + "grad_norm": 1.234375, + "learning_rate": 0.0016300175064867194, + "loss": 0.0854, + "step": 29459 + }, + { + "epoch": 0.25572694681469776, + "grad_norm": 0.162109375, + "learning_rate": 0.0016299934521844313, + "loss": 0.0879, + "step": 29460 + }, + { + "epoch": 0.25573562729490196, + "grad_norm": 0.07177734375, + "learning_rate": 0.0016299693973025507, + "loss": 0.0742, + "step": 29461 + }, + { + "epoch": 0.2557443077751061, + "grad_norm": 0.3359375, + "learning_rate": 0.0016299453418411032, + "loss": 0.1182, + "step": 29462 + }, + { + "epoch": 0.2557529882553103, + "grad_norm": 0.427734375, + "learning_rate": 0.0016299212858001157, + "loss": 0.1338, + "step": 29463 + }, + { + "epoch": 0.2557616687355144, + "grad_norm": 0.435546875, + "learning_rate": 0.0016298972291796138, + "loss": 0.0947, + "step": 29464 + }, + { + "epoch": 0.2557703492157186, + "grad_norm": 0.1455078125, + "learning_rate": 0.001629873171979625, + "loss": 0.0991, + "step": 29465 + }, + { + "epoch": 0.25577902969592275, + "grad_norm": 0.34765625, + "learning_rate": 0.0016298491142001745, + "loss": 0.1177, + "step": 29466 + }, + { + "epoch": 0.25578771017612695, + "grad_norm": 0.416015625, + "learning_rate": 0.0016298250558412887, + "loss": 0.1069, + "step": 29467 + }, + { + "epoch": 0.2557963906563311, + "grad_norm": 0.6796875, + "learning_rate": 0.0016298009969029942, + "loss": 0.1235, + "step": 29468 + }, + { + "epoch": 0.2558050711365353, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016297769373853177, + "loss": 0.1523, + "step": 29469 + }, + { + "epoch": 0.2558137516167394, + "grad_norm": 0.236328125, + "learning_rate": 0.001629752877288285, + "loss": 0.1221, + "step": 29470 + }, + { + "epoch": 0.2558224320969436, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016297288166119224, + "loss": 0.0747, + "step": 29471 + }, + { + "epoch": 0.25583111257714775, + "grad_norm": 0.10107421875, + "learning_rate": 0.0016297047553562563, + "loss": 0.1104, + "step": 29472 + }, + { + "epoch": 0.25583979305735194, + "grad_norm": 0.154296875, + "learning_rate": 0.0016296806935213132, + "loss": 0.0967, + "step": 29473 + }, + { + "epoch": 0.2558484735375561, + "grad_norm": 0.56640625, + "learning_rate": 0.001629656631107119, + "loss": 0.1426, + "step": 29474 + }, + { + "epoch": 0.25585715401776027, + "grad_norm": 0.09326171875, + "learning_rate": 0.0016296325681137004, + "loss": 0.1235, + "step": 29475 + }, + { + "epoch": 0.2558658344979644, + "grad_norm": 0.43359375, + "learning_rate": 0.0016296085045410839, + "loss": 0.0933, + "step": 29476 + }, + { + "epoch": 0.2558745149781686, + "grad_norm": 0.193359375, + "learning_rate": 0.0016295844403892948, + "loss": 0.1338, + "step": 29477 + }, + { + "epoch": 0.25588319545837274, + "grad_norm": 0.369140625, + "learning_rate": 0.0016295603756583605, + "loss": 0.1377, + "step": 29478 + }, + { + "epoch": 0.25589187593857693, + "grad_norm": 0.43359375, + "learning_rate": 0.001629536310348307, + "loss": 0.1387, + "step": 29479 + }, + { + "epoch": 0.25590055641878107, + "grad_norm": 0.173828125, + "learning_rate": 0.0016295122444591604, + "loss": 0.1221, + "step": 29480 + }, + { + "epoch": 0.25590923689898526, + "grad_norm": 0.3125, + "learning_rate": 0.0016294881779909472, + "loss": 0.0898, + "step": 29481 + }, + { + "epoch": 0.2559179173791894, + "grad_norm": 0.4140625, + "learning_rate": 0.0016294641109436936, + "loss": 0.0845, + "step": 29482 + }, + { + "epoch": 0.2559265978593936, + "grad_norm": 0.474609375, + "learning_rate": 0.001629440043317426, + "loss": 0.0908, + "step": 29483 + }, + { + "epoch": 0.2559352783395977, + "grad_norm": 0.291015625, + "learning_rate": 0.001629415975112171, + "loss": 0.1143, + "step": 29484 + }, + { + "epoch": 0.2559439588198019, + "grad_norm": 0.4921875, + "learning_rate": 0.0016293919063279545, + "loss": 0.1172, + "step": 29485 + }, + { + "epoch": 0.25595263930000606, + "grad_norm": 0.42578125, + "learning_rate": 0.001629367836964803, + "loss": 0.1152, + "step": 29486 + }, + { + "epoch": 0.25596131978021025, + "grad_norm": 0.21484375, + "learning_rate": 0.0016293437670227427, + "loss": 0.085, + "step": 29487 + }, + { + "epoch": 0.2559700002604144, + "grad_norm": 0.07080078125, + "learning_rate": 0.0016293196965018, + "loss": 0.1064, + "step": 29488 + }, + { + "epoch": 0.2559786807406186, + "grad_norm": 0.359375, + "learning_rate": 0.0016292956254020012, + "loss": 0.1416, + "step": 29489 + }, + { + "epoch": 0.2559873612208227, + "grad_norm": 0.4765625, + "learning_rate": 0.001629271553723373, + "loss": 0.1084, + "step": 29490 + }, + { + "epoch": 0.2559960417010269, + "grad_norm": 0.11279296875, + "learning_rate": 0.001629247481465941, + "loss": 0.1279, + "step": 29491 + }, + { + "epoch": 0.25600472218123105, + "grad_norm": 0.439453125, + "learning_rate": 0.001629223408629732, + "loss": 0.127, + "step": 29492 + }, + { + "epoch": 0.25601340266143524, + "grad_norm": 0.5234375, + "learning_rate": 0.0016291993352147724, + "loss": 0.0781, + "step": 29493 + }, + { + "epoch": 0.2560220831416394, + "grad_norm": 0.3984375, + "learning_rate": 0.001629175261221088, + "loss": 0.1533, + "step": 29494 + }, + { + "epoch": 0.25603076362184357, + "grad_norm": 0.1162109375, + "learning_rate": 0.001629151186648706, + "loss": 0.1099, + "step": 29495 + }, + { + "epoch": 0.2560394441020477, + "grad_norm": 0.189453125, + "learning_rate": 0.0016291271114976516, + "loss": 0.1406, + "step": 29496 + }, + { + "epoch": 0.2560481245822519, + "grad_norm": 0.25, + "learning_rate": 0.0016291030357679522, + "loss": 0.0879, + "step": 29497 + }, + { + "epoch": 0.25605680506245604, + "grad_norm": 0.11474609375, + "learning_rate": 0.0016290789594596334, + "loss": 0.1055, + "step": 29498 + }, + { + "epoch": 0.25606548554266023, + "grad_norm": 0.162109375, + "learning_rate": 0.0016290548825727217, + "loss": 0.1016, + "step": 29499 + }, + { + "epoch": 0.25607416602286437, + "grad_norm": 0.255859375, + "learning_rate": 0.0016290308051072437, + "loss": 0.1055, + "step": 29500 + }, + { + "epoch": 0.25608284650306856, + "grad_norm": 0.255859375, + "learning_rate": 0.001629006727063226, + "loss": 0.1084, + "step": 29501 + }, + { + "epoch": 0.2560915269832727, + "grad_norm": 0.10107421875, + "learning_rate": 0.0016289826484406939, + "loss": 0.1172, + "step": 29502 + }, + { + "epoch": 0.2561002074634769, + "grad_norm": 0.486328125, + "learning_rate": 0.0016289585692396745, + "loss": 0.1035, + "step": 29503 + }, + { + "epoch": 0.25610888794368103, + "grad_norm": 0.158203125, + "learning_rate": 0.0016289344894601938, + "loss": 0.1367, + "step": 29504 + }, + { + "epoch": 0.2561175684238852, + "grad_norm": 0.255859375, + "learning_rate": 0.0016289104091022787, + "loss": 0.1455, + "step": 29505 + }, + { + "epoch": 0.25612624890408936, + "grad_norm": 0.416015625, + "learning_rate": 0.0016288863281659549, + "loss": 0.0923, + "step": 29506 + }, + { + "epoch": 0.25613492938429355, + "grad_norm": 0.1435546875, + "learning_rate": 0.001628862246651249, + "loss": 0.0942, + "step": 29507 + }, + { + "epoch": 0.2561436098644977, + "grad_norm": 0.458984375, + "learning_rate": 0.001628838164558187, + "loss": 0.0762, + "step": 29508 + }, + { + "epoch": 0.2561522903447019, + "grad_norm": 0.1015625, + "learning_rate": 0.001628814081886796, + "loss": 0.0938, + "step": 29509 + }, + { + "epoch": 0.256160970824906, + "grad_norm": 0.10595703125, + "learning_rate": 0.0016287899986371015, + "loss": 0.1113, + "step": 29510 + }, + { + "epoch": 0.2561696513051102, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016287659148091305, + "loss": 0.0996, + "step": 29511 + }, + { + "epoch": 0.25617833178531435, + "grad_norm": 0.6015625, + "learning_rate": 0.0016287418304029087, + "loss": 0.0889, + "step": 29512 + }, + { + "epoch": 0.25618701226551854, + "grad_norm": 0.087890625, + "learning_rate": 0.001628717745418463, + "loss": 0.1191, + "step": 29513 + }, + { + "epoch": 0.2561956927457227, + "grad_norm": 0.326171875, + "learning_rate": 0.0016286936598558196, + "loss": 0.1011, + "step": 29514 + }, + { + "epoch": 0.2562043732259269, + "grad_norm": 0.400390625, + "learning_rate": 0.0016286695737150047, + "loss": 0.0928, + "step": 29515 + }, + { + "epoch": 0.256213053706131, + "grad_norm": 0.65234375, + "learning_rate": 0.0016286454869960444, + "loss": 0.1055, + "step": 29516 + }, + { + "epoch": 0.2562217341863352, + "grad_norm": 0.6328125, + "learning_rate": 0.0016286213996989656, + "loss": 0.1719, + "step": 29517 + }, + { + "epoch": 0.25623041466653934, + "grad_norm": 0.1875, + "learning_rate": 0.0016285973118237945, + "loss": 0.0972, + "step": 29518 + }, + { + "epoch": 0.25623909514674353, + "grad_norm": 0.1962890625, + "learning_rate": 0.0016285732233705572, + "loss": 0.1387, + "step": 29519 + }, + { + "epoch": 0.25624777562694767, + "grad_norm": 0.185546875, + "learning_rate": 0.00162854913433928, + "loss": 0.1328, + "step": 29520 + }, + { + "epoch": 0.25625645610715186, + "grad_norm": 0.52734375, + "learning_rate": 0.0016285250447299893, + "loss": 0.1143, + "step": 29521 + }, + { + "epoch": 0.256265136587356, + "grad_norm": 0.27734375, + "learning_rate": 0.001628500954542712, + "loss": 0.1309, + "step": 29522 + }, + { + "epoch": 0.2562738170675602, + "grad_norm": 0.125, + "learning_rate": 0.0016284768637774738, + "loss": 0.1064, + "step": 29523 + }, + { + "epoch": 0.25628249754776433, + "grad_norm": 0.080078125, + "learning_rate": 0.0016284527724343013, + "loss": 0.0825, + "step": 29524 + }, + { + "epoch": 0.2562911780279685, + "grad_norm": 0.0986328125, + "learning_rate": 0.001628428680513221, + "loss": 0.0986, + "step": 29525 + }, + { + "epoch": 0.25629985850817266, + "grad_norm": 0.81640625, + "learning_rate": 0.0016284045880142586, + "loss": 0.0977, + "step": 29526 + }, + { + "epoch": 0.25630853898837686, + "grad_norm": 0.1123046875, + "learning_rate": 0.001628380494937441, + "loss": 0.125, + "step": 29527 + }, + { + "epoch": 0.256317219468581, + "grad_norm": 0.1552734375, + "learning_rate": 0.0016283564012827942, + "loss": 0.1113, + "step": 29528 + }, + { + "epoch": 0.2563258999487852, + "grad_norm": 0.63671875, + "learning_rate": 0.0016283323070503451, + "loss": 0.1172, + "step": 29529 + }, + { + "epoch": 0.2563345804289893, + "grad_norm": 0.5703125, + "learning_rate": 0.0016283082122401198, + "loss": 0.1328, + "step": 29530 + }, + { + "epoch": 0.2563432609091935, + "grad_norm": 0.2333984375, + "learning_rate": 0.0016282841168521445, + "loss": 0.1182, + "step": 29531 + }, + { + "epoch": 0.25635194138939765, + "grad_norm": 0.25390625, + "learning_rate": 0.0016282600208864459, + "loss": 0.1514, + "step": 29532 + }, + { + "epoch": 0.25636062186960185, + "grad_norm": 0.0732421875, + "learning_rate": 0.0016282359243430494, + "loss": 0.082, + "step": 29533 + }, + { + "epoch": 0.256369302349806, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016282118272219824, + "loss": 0.103, + "step": 29534 + }, + { + "epoch": 0.2563779828300102, + "grad_norm": 0.080078125, + "learning_rate": 0.0016281877295232707, + "loss": 0.0952, + "step": 29535 + }, + { + "epoch": 0.2563866633102143, + "grad_norm": 0.09033203125, + "learning_rate": 0.0016281636312469412, + "loss": 0.1201, + "step": 29536 + }, + { + "epoch": 0.2563953437904185, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016281395323930197, + "loss": 0.1118, + "step": 29537 + }, + { + "epoch": 0.25640402427062264, + "grad_norm": 0.240234375, + "learning_rate": 0.0016281154329615326, + "loss": 0.1216, + "step": 29538 + }, + { + "epoch": 0.25641270475082684, + "grad_norm": 0.140625, + "learning_rate": 0.0016280913329525064, + "loss": 0.0938, + "step": 29539 + }, + { + "epoch": 0.256421385231031, + "grad_norm": 1.1796875, + "learning_rate": 0.0016280672323659678, + "loss": 0.1289, + "step": 29540 + }, + { + "epoch": 0.25643006571123517, + "grad_norm": 0.7890625, + "learning_rate": 0.0016280431312019426, + "loss": 0.0923, + "step": 29541 + }, + { + "epoch": 0.2564387461914393, + "grad_norm": 0.11865234375, + "learning_rate": 0.0016280190294604574, + "loss": 0.0967, + "step": 29542 + }, + { + "epoch": 0.2564474266716435, + "grad_norm": 0.068359375, + "learning_rate": 0.0016279949271415385, + "loss": 0.0923, + "step": 29543 + }, + { + "epoch": 0.25645610715184763, + "grad_norm": 0.40625, + "learning_rate": 0.001627970824245212, + "loss": 0.0781, + "step": 29544 + }, + { + "epoch": 0.2564647876320518, + "grad_norm": 0.48828125, + "learning_rate": 0.0016279467207715049, + "loss": 0.0942, + "step": 29545 + }, + { + "epoch": 0.25647346811225596, + "grad_norm": 0.380859375, + "learning_rate": 0.0016279226167204431, + "loss": 0.1191, + "step": 29546 + }, + { + "epoch": 0.25648214859246016, + "grad_norm": 0.265625, + "learning_rate": 0.001627898512092053, + "loss": 0.1465, + "step": 29547 + }, + { + "epoch": 0.2564908290726643, + "grad_norm": 0.220703125, + "learning_rate": 0.001627874406886361, + "loss": 0.124, + "step": 29548 + }, + { + "epoch": 0.2564995095528685, + "grad_norm": 0.703125, + "learning_rate": 0.0016278503011033937, + "loss": 0.1553, + "step": 29549 + }, + { + "epoch": 0.2565081900330726, + "grad_norm": 0.462890625, + "learning_rate": 0.001627826194743177, + "loss": 0.1123, + "step": 29550 + }, + { + "epoch": 0.2565168705132768, + "grad_norm": 0.09716796875, + "learning_rate": 0.0016278020878057375, + "loss": 0.1162, + "step": 29551 + }, + { + "epoch": 0.25652555099348096, + "grad_norm": 0.349609375, + "learning_rate": 0.001627777980291102, + "loss": 0.0752, + "step": 29552 + }, + { + "epoch": 0.25653423147368515, + "grad_norm": 0.2197265625, + "learning_rate": 0.0016277538721992957, + "loss": 0.0845, + "step": 29553 + }, + { + "epoch": 0.2565429119538893, + "grad_norm": 0.2265625, + "learning_rate": 0.0016277297635303462, + "loss": 0.0771, + "step": 29554 + }, + { + "epoch": 0.2565515924340935, + "grad_norm": 0.0966796875, + "learning_rate": 0.0016277056542842794, + "loss": 0.0996, + "step": 29555 + }, + { + "epoch": 0.2565602729142976, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016276815444611215, + "loss": 0.0747, + "step": 29556 + }, + { + "epoch": 0.2565689533945018, + "grad_norm": 0.0771484375, + "learning_rate": 0.001627657434060899, + "loss": 0.1118, + "step": 29557 + }, + { + "epoch": 0.25657763387470595, + "grad_norm": 0.134765625, + "learning_rate": 0.001627633323083638, + "loss": 0.0791, + "step": 29558 + }, + { + "epoch": 0.25658631435491014, + "grad_norm": 0.423828125, + "learning_rate": 0.0016276092115293653, + "loss": 0.0811, + "step": 29559 + }, + { + "epoch": 0.2565949948351143, + "grad_norm": 0.7265625, + "learning_rate": 0.0016275850993981071, + "loss": 0.1177, + "step": 29560 + }, + { + "epoch": 0.25660367531531847, + "grad_norm": 0.35546875, + "learning_rate": 0.00162756098668989, + "loss": 0.084, + "step": 29561 + }, + { + "epoch": 0.2566123557955226, + "grad_norm": 0.298828125, + "learning_rate": 0.0016275368734047402, + "loss": 0.0703, + "step": 29562 + }, + { + "epoch": 0.2566210362757268, + "grad_norm": 0.57421875, + "learning_rate": 0.0016275127595426833, + "loss": 0.1113, + "step": 29563 + }, + { + "epoch": 0.25662971675593094, + "grad_norm": 0.318359375, + "learning_rate": 0.001627488645103747, + "loss": 0.1167, + "step": 29564 + }, + { + "epoch": 0.25663839723613513, + "grad_norm": 0.419921875, + "learning_rate": 0.0016274645300879572, + "loss": 0.1074, + "step": 29565 + }, + { + "epoch": 0.25664707771633927, + "grad_norm": 0.193359375, + "learning_rate": 0.0016274404144953395, + "loss": 0.0869, + "step": 29566 + }, + { + "epoch": 0.25665575819654346, + "grad_norm": 0.150390625, + "learning_rate": 0.001627416298325921, + "loss": 0.1104, + "step": 29567 + }, + { + "epoch": 0.2566644386767476, + "grad_norm": 0.080078125, + "learning_rate": 0.0016273921815797283, + "loss": 0.0742, + "step": 29568 + }, + { + "epoch": 0.2566731191569518, + "grad_norm": 0.10400390625, + "learning_rate": 0.001627368064256787, + "loss": 0.0981, + "step": 29569 + }, + { + "epoch": 0.2566817996371559, + "grad_norm": 0.146484375, + "learning_rate": 0.0016273439463571245, + "loss": 0.1143, + "step": 29570 + }, + { + "epoch": 0.25669048011736006, + "grad_norm": 0.703125, + "learning_rate": 0.0016273198278807662, + "loss": 0.0874, + "step": 29571 + }, + { + "epoch": 0.25669916059756426, + "grad_norm": 0.353515625, + "learning_rate": 0.0016272957088277389, + "loss": 0.1064, + "step": 29572 + }, + { + "epoch": 0.2567078410777684, + "grad_norm": 0.333984375, + "learning_rate": 0.0016272715891980688, + "loss": 0.0938, + "step": 29573 + }, + { + "epoch": 0.2567165215579726, + "grad_norm": 0.9453125, + "learning_rate": 0.0016272474689917828, + "loss": 0.1104, + "step": 29574 + }, + { + "epoch": 0.2567252020381767, + "grad_norm": 0.177734375, + "learning_rate": 0.0016272233482089065, + "loss": 0.1377, + "step": 29575 + }, + { + "epoch": 0.2567338825183809, + "grad_norm": 0.26171875, + "learning_rate": 0.0016271992268494667, + "loss": 0.127, + "step": 29576 + }, + { + "epoch": 0.25674256299858506, + "grad_norm": 0.12060546875, + "learning_rate": 0.00162717510491349, + "loss": 0.127, + "step": 29577 + }, + { + "epoch": 0.25675124347878925, + "grad_norm": 0.38671875, + "learning_rate": 0.0016271509824010025, + "loss": 0.124, + "step": 29578 + }, + { + "epoch": 0.2567599239589934, + "grad_norm": 0.07470703125, + "learning_rate": 0.0016271268593120306, + "loss": 0.084, + "step": 29579 + }, + { + "epoch": 0.2567686044391976, + "grad_norm": 0.265625, + "learning_rate": 0.0016271027356466003, + "loss": 0.0918, + "step": 29580 + }, + { + "epoch": 0.2567772849194017, + "grad_norm": 0.89453125, + "learning_rate": 0.0016270786114047386, + "loss": 0.0957, + "step": 29581 + }, + { + "epoch": 0.2567859653996059, + "grad_norm": 0.2001953125, + "learning_rate": 0.0016270544865864718, + "loss": 0.127, + "step": 29582 + }, + { + "epoch": 0.25679464587981005, + "grad_norm": 0.2275390625, + "learning_rate": 0.001627030361191826, + "loss": 0.1475, + "step": 29583 + }, + { + "epoch": 0.25680332636001424, + "grad_norm": 0.10498046875, + "learning_rate": 0.0016270062352208277, + "loss": 0.1406, + "step": 29584 + }, + { + "epoch": 0.2568120068402184, + "grad_norm": 0.251953125, + "learning_rate": 0.0016269821086735033, + "loss": 0.0635, + "step": 29585 + }, + { + "epoch": 0.25682068732042257, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016269579815498792, + "loss": 0.103, + "step": 29586 + }, + { + "epoch": 0.2568293678006267, + "grad_norm": 0.7265625, + "learning_rate": 0.0016269338538499821, + "loss": 0.1123, + "step": 29587 + }, + { + "epoch": 0.2568380482808309, + "grad_norm": 0.515625, + "learning_rate": 0.0016269097255738375, + "loss": 0.1143, + "step": 29588 + }, + { + "epoch": 0.25684672876103504, + "grad_norm": 0.251953125, + "learning_rate": 0.0016268855967214726, + "loss": 0.0747, + "step": 29589 + }, + { + "epoch": 0.25685540924123923, + "grad_norm": 0.2158203125, + "learning_rate": 0.0016268614672929136, + "loss": 0.0981, + "step": 29590 + }, + { + "epoch": 0.25686408972144337, + "grad_norm": 0.13671875, + "learning_rate": 0.0016268373372881869, + "loss": 0.127, + "step": 29591 + }, + { + "epoch": 0.25687277020164756, + "grad_norm": 0.6875, + "learning_rate": 0.0016268132067073185, + "loss": 0.104, + "step": 29592 + }, + { + "epoch": 0.2568814506818517, + "grad_norm": 0.1279296875, + "learning_rate": 0.001626789075550335, + "loss": 0.124, + "step": 29593 + }, + { + "epoch": 0.2568901311620559, + "grad_norm": 0.220703125, + "learning_rate": 0.0016267649438172634, + "loss": 0.1133, + "step": 29594 + }, + { + "epoch": 0.25689881164226, + "grad_norm": 0.44921875, + "learning_rate": 0.001626740811508129, + "loss": 0.0796, + "step": 29595 + }, + { + "epoch": 0.2569074921224642, + "grad_norm": 0.99609375, + "learning_rate": 0.0016267166786229593, + "loss": 0.0825, + "step": 29596 + }, + { + "epoch": 0.25691617260266836, + "grad_norm": 0.6875, + "learning_rate": 0.0016266925451617798, + "loss": 0.1309, + "step": 29597 + }, + { + "epoch": 0.25692485308287255, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016266684111246174, + "loss": 0.1543, + "step": 29598 + }, + { + "epoch": 0.2569335335630767, + "grad_norm": 0.1591796875, + "learning_rate": 0.0016266442765114982, + "loss": 0.1338, + "step": 29599 + }, + { + "epoch": 0.2569422140432809, + "grad_norm": 0.6875, + "learning_rate": 0.0016266201413224488, + "loss": 0.0942, + "step": 29600 + }, + { + "epoch": 0.256950894523485, + "grad_norm": 0.388671875, + "learning_rate": 0.0016265960055574956, + "loss": 0.0977, + "step": 29601 + }, + { + "epoch": 0.2569595750036892, + "grad_norm": 0.64453125, + "learning_rate": 0.0016265718692166646, + "loss": 0.1885, + "step": 29602 + }, + { + "epoch": 0.25696825548389335, + "grad_norm": 0.140625, + "learning_rate": 0.001626547732299983, + "loss": 0.1504, + "step": 29603 + }, + { + "epoch": 0.25697693596409754, + "grad_norm": 0.3359375, + "learning_rate": 0.0016265235948074764, + "loss": 0.1641, + "step": 29604 + }, + { + "epoch": 0.2569856164443017, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016264994567391719, + "loss": 0.083, + "step": 29605 + }, + { + "epoch": 0.25699429692450587, + "grad_norm": 0.78125, + "learning_rate": 0.0016264753180950948, + "loss": 0.1514, + "step": 29606 + }, + { + "epoch": 0.25700297740471, + "grad_norm": 0.09716796875, + "learning_rate": 0.0016264511788752728, + "loss": 0.1084, + "step": 29607 + }, + { + "epoch": 0.2570116578849142, + "grad_norm": 0.1318359375, + "learning_rate": 0.0016264270390797314, + "loss": 0.0942, + "step": 29608 + }, + { + "epoch": 0.25702033836511834, + "grad_norm": 0.427734375, + "learning_rate": 0.0016264028987084973, + "loss": 0.1123, + "step": 29609 + }, + { + "epoch": 0.25702901884532253, + "grad_norm": 0.53515625, + "learning_rate": 0.0016263787577615971, + "loss": 0.0859, + "step": 29610 + }, + { + "epoch": 0.25703769932552667, + "grad_norm": 0.263671875, + "learning_rate": 0.0016263546162390567, + "loss": 0.0918, + "step": 29611 + }, + { + "epoch": 0.25704637980573086, + "grad_norm": 0.51171875, + "learning_rate": 0.0016263304741409032, + "loss": 0.1147, + "step": 29612 + }, + { + "epoch": 0.257055060285935, + "grad_norm": 0.1328125, + "learning_rate": 0.0016263063314671624, + "loss": 0.0918, + "step": 29613 + }, + { + "epoch": 0.2570637407661392, + "grad_norm": 1.1015625, + "learning_rate": 0.0016262821882178612, + "loss": 0.127, + "step": 29614 + }, + { + "epoch": 0.25707242124634333, + "grad_norm": 0.0869140625, + "learning_rate": 0.0016262580443930253, + "loss": 0.0908, + "step": 29615 + }, + { + "epoch": 0.2570811017265475, + "grad_norm": 0.2119140625, + "learning_rate": 0.0016262338999926814, + "loss": 0.1426, + "step": 29616 + }, + { + "epoch": 0.25708978220675166, + "grad_norm": 0.15234375, + "learning_rate": 0.0016262097550168563, + "loss": 0.0981, + "step": 29617 + }, + { + "epoch": 0.25709846268695585, + "grad_norm": 0.21484375, + "learning_rate": 0.0016261856094655761, + "loss": 0.1035, + "step": 29618 + }, + { + "epoch": 0.25710714316716, + "grad_norm": 0.33984375, + "learning_rate": 0.0016261614633388672, + "loss": 0.0938, + "step": 29619 + }, + { + "epoch": 0.2571158236473642, + "grad_norm": 0.369140625, + "learning_rate": 0.001626137316636756, + "loss": 0.1001, + "step": 29620 + }, + { + "epoch": 0.2571245041275683, + "grad_norm": 0.11767578125, + "learning_rate": 0.0016261131693592688, + "loss": 0.0967, + "step": 29621 + }, + { + "epoch": 0.2571331846077725, + "grad_norm": 0.318359375, + "learning_rate": 0.0016260890215064323, + "loss": 0.1387, + "step": 29622 + }, + { + "epoch": 0.25714186508797665, + "grad_norm": 0.59375, + "learning_rate": 0.0016260648730782729, + "loss": 0.104, + "step": 29623 + }, + { + "epoch": 0.25715054556818084, + "grad_norm": 0.337890625, + "learning_rate": 0.0016260407240748166, + "loss": 0.0957, + "step": 29624 + }, + { + "epoch": 0.257159226048385, + "grad_norm": 0.51171875, + "learning_rate": 0.0016260165744960903, + "loss": 0.0986, + "step": 29625 + }, + { + "epoch": 0.2571679065285892, + "grad_norm": 0.154296875, + "learning_rate": 0.0016259924243421198, + "loss": 0.082, + "step": 29626 + }, + { + "epoch": 0.2571765870087933, + "grad_norm": 0.2470703125, + "learning_rate": 0.0016259682736129324, + "loss": 0.1045, + "step": 29627 + }, + { + "epoch": 0.2571852674889975, + "grad_norm": 0.443359375, + "learning_rate": 0.0016259441223085535, + "loss": 0.1367, + "step": 29628 + }, + { + "epoch": 0.25719394796920164, + "grad_norm": 0.2060546875, + "learning_rate": 0.0016259199704290103, + "loss": 0.1182, + "step": 29629 + }, + { + "epoch": 0.25720262844940583, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016258958179743291, + "loss": 0.082, + "step": 29630 + }, + { + "epoch": 0.25721130892960997, + "grad_norm": 0.1591796875, + "learning_rate": 0.0016258716649445357, + "loss": 0.0654, + "step": 29631 + }, + { + "epoch": 0.25721998940981416, + "grad_norm": 0.1962890625, + "learning_rate": 0.0016258475113396572, + "loss": 0.103, + "step": 29632 + }, + { + "epoch": 0.2572286698900183, + "grad_norm": 0.578125, + "learning_rate": 0.0016258233571597197, + "loss": 0.1055, + "step": 29633 + }, + { + "epoch": 0.2572373503702225, + "grad_norm": 0.54296875, + "learning_rate": 0.0016257992024047497, + "loss": 0.1201, + "step": 29634 + }, + { + "epoch": 0.25724603085042663, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016257750470747734, + "loss": 0.1025, + "step": 29635 + }, + { + "epoch": 0.2572547113306308, + "grad_norm": 0.373046875, + "learning_rate": 0.0016257508911698179, + "loss": 0.0625, + "step": 29636 + }, + { + "epoch": 0.25726339181083496, + "grad_norm": 0.349609375, + "learning_rate": 0.0016257267346899088, + "loss": 0.123, + "step": 29637 + }, + { + "epoch": 0.25727207229103916, + "grad_norm": 0.46875, + "learning_rate": 0.0016257025776350725, + "loss": 0.0898, + "step": 29638 + }, + { + "epoch": 0.2572807527712433, + "grad_norm": 0.275390625, + "learning_rate": 0.0016256784200053363, + "loss": 0.083, + "step": 29639 + }, + { + "epoch": 0.2572894332514475, + "grad_norm": 0.44921875, + "learning_rate": 0.0016256542618007259, + "loss": 0.1172, + "step": 29640 + }, + { + "epoch": 0.2572981137316516, + "grad_norm": 0.1474609375, + "learning_rate": 0.0016256301030212678, + "loss": 0.1133, + "step": 29641 + }, + { + "epoch": 0.2573067942118558, + "grad_norm": 0.6015625, + "learning_rate": 0.0016256059436669888, + "loss": 0.1406, + "step": 29642 + }, + { + "epoch": 0.25731547469205995, + "grad_norm": 0.0859375, + "learning_rate": 0.0016255817837379147, + "loss": 0.0859, + "step": 29643 + }, + { + "epoch": 0.25732415517226415, + "grad_norm": 0.119140625, + "learning_rate": 0.0016255576232340723, + "loss": 0.1221, + "step": 29644 + }, + { + "epoch": 0.2573328356524683, + "grad_norm": 0.4296875, + "learning_rate": 0.0016255334621554881, + "loss": 0.0859, + "step": 29645 + }, + { + "epoch": 0.2573415161326725, + "grad_norm": 0.341796875, + "learning_rate": 0.001625509300502188, + "loss": 0.1201, + "step": 29646 + }, + { + "epoch": 0.2573501966128766, + "grad_norm": 0.8671875, + "learning_rate": 0.0016254851382741995, + "loss": 0.1172, + "step": 29647 + }, + { + "epoch": 0.2573588770930808, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016254609754715484, + "loss": 0.1055, + "step": 29648 + }, + { + "epoch": 0.25736755757328494, + "grad_norm": 0.56640625, + "learning_rate": 0.0016254368120942604, + "loss": 0.0977, + "step": 29649 + }, + { + "epoch": 0.25737623805348914, + "grad_norm": 0.259765625, + "learning_rate": 0.001625412648142363, + "loss": 0.0635, + "step": 29650 + }, + { + "epoch": 0.2573849185336933, + "grad_norm": 0.73046875, + "learning_rate": 0.0016253884836158821, + "loss": 0.0796, + "step": 29651 + }, + { + "epoch": 0.25739359901389747, + "grad_norm": 0.5546875, + "learning_rate": 0.0016253643185148443, + "loss": 0.0977, + "step": 29652 + }, + { + "epoch": 0.2574022794941016, + "grad_norm": 0.11572265625, + "learning_rate": 0.001625340152839276, + "loss": 0.1025, + "step": 29653 + }, + { + "epoch": 0.2574109599743058, + "grad_norm": 1.6640625, + "learning_rate": 0.0016253159865892036, + "loss": 0.1182, + "step": 29654 + }, + { + "epoch": 0.25741964045450993, + "grad_norm": 0.70703125, + "learning_rate": 0.0016252918197646534, + "loss": 0.1514, + "step": 29655 + }, + { + "epoch": 0.2574283209347141, + "grad_norm": 1.046875, + "learning_rate": 0.0016252676523656523, + "loss": 0.0903, + "step": 29656 + }, + { + "epoch": 0.25743700141491827, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016252434843922262, + "loss": 0.1143, + "step": 29657 + }, + { + "epoch": 0.25744568189512246, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016252193158444014, + "loss": 0.0918, + "step": 29658 + }, + { + "epoch": 0.2574543623753266, + "grad_norm": 0.1650390625, + "learning_rate": 0.001625195146722205, + "loss": 0.125, + "step": 29659 + }, + { + "epoch": 0.2574630428555308, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016251709770256628, + "loss": 0.1348, + "step": 29660 + }, + { + "epoch": 0.2574717233357349, + "grad_norm": 0.1025390625, + "learning_rate": 0.001625146806754802, + "loss": 0.1172, + "step": 29661 + }, + { + "epoch": 0.2574804038159391, + "grad_norm": 0.2099609375, + "learning_rate": 0.0016251226359096482, + "loss": 0.1123, + "step": 29662 + }, + { + "epoch": 0.25748908429614326, + "grad_norm": 1.4296875, + "learning_rate": 0.0016250984644902281, + "loss": 0.124, + "step": 29663 + }, + { + "epoch": 0.25749776477634745, + "grad_norm": 0.58984375, + "learning_rate": 0.0016250742924965684, + "loss": 0.1021, + "step": 29664 + }, + { + "epoch": 0.2575064452565516, + "grad_norm": 0.1015625, + "learning_rate": 0.001625050119928695, + "loss": 0.0674, + "step": 29665 + }, + { + "epoch": 0.2575151257367558, + "grad_norm": 0.482421875, + "learning_rate": 0.0016250259467866352, + "loss": 0.1089, + "step": 29666 + }, + { + "epoch": 0.2575238062169599, + "grad_norm": 0.83984375, + "learning_rate": 0.0016250017730704148, + "loss": 0.0859, + "step": 29667 + }, + { + "epoch": 0.2575324866971641, + "grad_norm": 0.265625, + "learning_rate": 0.00162497759878006, + "loss": 0.0825, + "step": 29668 + }, + { + "epoch": 0.25754116717736825, + "grad_norm": 0.1748046875, + "learning_rate": 0.0016249534239155977, + "loss": 0.0747, + "step": 29669 + }, + { + "epoch": 0.25754984765757244, + "grad_norm": 0.5390625, + "learning_rate": 0.0016249292484770545, + "loss": 0.0913, + "step": 29670 + }, + { + "epoch": 0.2575585281377766, + "grad_norm": 0.392578125, + "learning_rate": 0.0016249050724644561, + "loss": 0.0913, + "step": 29671 + }, + { + "epoch": 0.25756720861798077, + "grad_norm": 0.43359375, + "learning_rate": 0.0016248808958778296, + "loss": 0.1064, + "step": 29672 + }, + { + "epoch": 0.2575758890981849, + "grad_norm": 0.375, + "learning_rate": 0.0016248567187172017, + "loss": 0.1094, + "step": 29673 + }, + { + "epoch": 0.2575845695783891, + "grad_norm": 0.1904296875, + "learning_rate": 0.0016248325409825978, + "loss": 0.1104, + "step": 29674 + }, + { + "epoch": 0.25759325005859324, + "grad_norm": 0.1748046875, + "learning_rate": 0.0016248083626740452, + "loss": 0.1143, + "step": 29675 + }, + { + "epoch": 0.25760193053879743, + "grad_norm": 0.07763671875, + "learning_rate": 0.00162478418379157, + "loss": 0.1143, + "step": 29676 + }, + { + "epoch": 0.25761061101900157, + "grad_norm": 0.353515625, + "learning_rate": 0.0016247600043351987, + "loss": 0.0923, + "step": 29677 + }, + { + "epoch": 0.25761929149920576, + "grad_norm": 0.76171875, + "learning_rate": 0.0016247358243049574, + "loss": 0.1465, + "step": 29678 + }, + { + "epoch": 0.2576279719794099, + "grad_norm": 0.62109375, + "learning_rate": 0.0016247116437008735, + "loss": 0.1475, + "step": 29679 + }, + { + "epoch": 0.2576366524596141, + "grad_norm": 0.291015625, + "learning_rate": 0.0016246874625229725, + "loss": 0.0991, + "step": 29680 + }, + { + "epoch": 0.25764533293981823, + "grad_norm": 0.06884765625, + "learning_rate": 0.0016246632807712812, + "loss": 0.0996, + "step": 29681 + }, + { + "epoch": 0.2576540134200224, + "grad_norm": 0.19921875, + "learning_rate": 0.0016246390984458261, + "loss": 0.1279, + "step": 29682 + }, + { + "epoch": 0.25766269390022656, + "grad_norm": 0.859375, + "learning_rate": 0.0016246149155466334, + "loss": 0.1064, + "step": 29683 + }, + { + "epoch": 0.25767137438043075, + "grad_norm": 0.40625, + "learning_rate": 0.0016245907320737297, + "loss": 0.1011, + "step": 29684 + }, + { + "epoch": 0.2576800548606349, + "grad_norm": 0.3515625, + "learning_rate": 0.0016245665480271417, + "loss": 0.1191, + "step": 29685 + }, + { + "epoch": 0.2576887353408391, + "grad_norm": 0.78515625, + "learning_rate": 0.0016245423634068954, + "loss": 0.0723, + "step": 29686 + }, + { + "epoch": 0.2576974158210432, + "grad_norm": 0.27734375, + "learning_rate": 0.0016245181782130178, + "loss": 0.1123, + "step": 29687 + }, + { + "epoch": 0.2577060963012474, + "grad_norm": 0.208984375, + "learning_rate": 0.0016244939924455347, + "loss": 0.0859, + "step": 29688 + }, + { + "epoch": 0.25771477678145155, + "grad_norm": 0.546875, + "learning_rate": 0.0016244698061044728, + "loss": 0.1104, + "step": 29689 + }, + { + "epoch": 0.25772345726165574, + "grad_norm": 0.390625, + "learning_rate": 0.0016244456191898584, + "loss": 0.085, + "step": 29690 + }, + { + "epoch": 0.2577321377418599, + "grad_norm": 0.359375, + "learning_rate": 0.0016244214317017185, + "loss": 0.126, + "step": 29691 + }, + { + "epoch": 0.25774081822206407, + "grad_norm": 0.361328125, + "learning_rate": 0.0016243972436400791, + "loss": 0.1113, + "step": 29692 + }, + { + "epoch": 0.2577494987022682, + "grad_norm": 0.255859375, + "learning_rate": 0.0016243730550049669, + "loss": 0.0986, + "step": 29693 + }, + { + "epoch": 0.25775817918247235, + "grad_norm": 0.390625, + "learning_rate": 0.0016243488657964082, + "loss": 0.1133, + "step": 29694 + }, + { + "epoch": 0.25776685966267654, + "grad_norm": 0.1982421875, + "learning_rate": 0.0016243246760144292, + "loss": 0.0923, + "step": 29695 + }, + { + "epoch": 0.2577755401428807, + "grad_norm": 0.1220703125, + "learning_rate": 0.0016243004856590566, + "loss": 0.082, + "step": 29696 + }, + { + "epoch": 0.25778422062308487, + "grad_norm": 0.90234375, + "learning_rate": 0.0016242762947303172, + "loss": 0.1211, + "step": 29697 + }, + { + "epoch": 0.257792901103289, + "grad_norm": 0.166015625, + "learning_rate": 0.001624252103228237, + "loss": 0.0815, + "step": 29698 + }, + { + "epoch": 0.2578015815834932, + "grad_norm": 0.65625, + "learning_rate": 0.0016242279111528422, + "loss": 0.0884, + "step": 29699 + }, + { + "epoch": 0.25781026206369734, + "grad_norm": 0.61328125, + "learning_rate": 0.00162420371850416, + "loss": 0.1143, + "step": 29700 + }, + { + "epoch": 0.25781894254390153, + "grad_norm": 0.09619140625, + "learning_rate": 0.0016241795252822165, + "loss": 0.1128, + "step": 29701 + }, + { + "epoch": 0.25782762302410567, + "grad_norm": 0.1123046875, + "learning_rate": 0.001624155331487038, + "loss": 0.0747, + "step": 29702 + }, + { + "epoch": 0.25783630350430986, + "grad_norm": 0.19140625, + "learning_rate": 0.0016241311371186513, + "loss": 0.1001, + "step": 29703 + }, + { + "epoch": 0.257844983984514, + "grad_norm": 0.341796875, + "learning_rate": 0.0016241069421770826, + "loss": 0.1133, + "step": 29704 + }, + { + "epoch": 0.2578536644647182, + "grad_norm": 0.72265625, + "learning_rate": 0.0016240827466623585, + "loss": 0.1055, + "step": 29705 + }, + { + "epoch": 0.25786234494492233, + "grad_norm": 0.259765625, + "learning_rate": 0.001624058550574505, + "loss": 0.1221, + "step": 29706 + }, + { + "epoch": 0.2578710254251265, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016240343539135494, + "loss": 0.1035, + "step": 29707 + }, + { + "epoch": 0.25787970590533066, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016240101566795174, + "loss": 0.1396, + "step": 29708 + }, + { + "epoch": 0.25788838638553485, + "grad_norm": 0.2158203125, + "learning_rate": 0.001623985958872436, + "loss": 0.1436, + "step": 29709 + }, + { + "epoch": 0.257897066865739, + "grad_norm": 0.28515625, + "learning_rate": 0.0016239617604923312, + "loss": 0.165, + "step": 29710 + }, + { + "epoch": 0.2579057473459432, + "grad_norm": 0.279296875, + "learning_rate": 0.0016239375615392298, + "loss": 0.1318, + "step": 29711 + }, + { + "epoch": 0.2579144278261473, + "grad_norm": 0.33203125, + "learning_rate": 0.001623913362013158, + "loss": 0.0942, + "step": 29712 + }, + { + "epoch": 0.2579231083063515, + "grad_norm": 0.3984375, + "learning_rate": 0.0016238891619141426, + "loss": 0.0762, + "step": 29713 + }, + { + "epoch": 0.25793178878655565, + "grad_norm": 0.318359375, + "learning_rate": 0.0016238649612422098, + "loss": 0.0938, + "step": 29714 + }, + { + "epoch": 0.25794046926675984, + "grad_norm": 0.091796875, + "learning_rate": 0.0016238407599973863, + "loss": 0.0986, + "step": 29715 + }, + { + "epoch": 0.257949149746964, + "grad_norm": 0.50390625, + "learning_rate": 0.0016238165581796984, + "loss": 0.1201, + "step": 29716 + }, + { + "epoch": 0.25795783022716817, + "grad_norm": 0.328125, + "learning_rate": 0.0016237923557891722, + "loss": 0.0903, + "step": 29717 + }, + { + "epoch": 0.2579665107073723, + "grad_norm": 0.24609375, + "learning_rate": 0.001623768152825835, + "loss": 0.126, + "step": 29718 + }, + { + "epoch": 0.2579751911875765, + "grad_norm": 0.27734375, + "learning_rate": 0.0016237439492897128, + "loss": 0.1104, + "step": 29719 + }, + { + "epoch": 0.25798387166778064, + "grad_norm": 0.29296875, + "learning_rate": 0.001623719745180832, + "loss": 0.0742, + "step": 29720 + }, + { + "epoch": 0.25799255214798483, + "grad_norm": 0.10107421875, + "learning_rate": 0.001623695540499219, + "loss": 0.1099, + "step": 29721 + }, + { + "epoch": 0.25800123262818897, + "grad_norm": 0.169921875, + "learning_rate": 0.0016236713352449003, + "loss": 0.0947, + "step": 29722 + }, + { + "epoch": 0.25800991310839316, + "grad_norm": 0.115234375, + "learning_rate": 0.001623647129417903, + "loss": 0.1709, + "step": 29723 + }, + { + "epoch": 0.2580185935885973, + "grad_norm": 0.1162109375, + "learning_rate": 0.0016236229230182527, + "loss": 0.1162, + "step": 29724 + }, + { + "epoch": 0.2580272740688015, + "grad_norm": 0.119140625, + "learning_rate": 0.0016235987160459762, + "loss": 0.1143, + "step": 29725 + }, + { + "epoch": 0.25803595454900563, + "grad_norm": 0.52734375, + "learning_rate": 0.0016235745085011004, + "loss": 0.0908, + "step": 29726 + }, + { + "epoch": 0.2580446350292098, + "grad_norm": 0.29296875, + "learning_rate": 0.001623550300383651, + "loss": 0.1104, + "step": 29727 + }, + { + "epoch": 0.25805331550941396, + "grad_norm": 0.53125, + "learning_rate": 0.001623526091693655, + "loss": 0.1079, + "step": 29728 + }, + { + "epoch": 0.25806199598961815, + "grad_norm": 0.494140625, + "learning_rate": 0.0016235018824311386, + "loss": 0.0649, + "step": 29729 + }, + { + "epoch": 0.2580706764698223, + "grad_norm": 0.12109375, + "learning_rate": 0.0016234776725961287, + "loss": 0.0898, + "step": 29730 + }, + { + "epoch": 0.2580793569500265, + "grad_norm": 0.12451171875, + "learning_rate": 0.0016234534621886515, + "loss": 0.0859, + "step": 29731 + }, + { + "epoch": 0.2580880374302306, + "grad_norm": 0.50390625, + "learning_rate": 0.0016234292512087329, + "loss": 0.0938, + "step": 29732 + }, + { + "epoch": 0.2580967179104348, + "grad_norm": 0.2373046875, + "learning_rate": 0.0016234050396564002, + "loss": 0.0918, + "step": 29733 + }, + { + "epoch": 0.25810539839063895, + "grad_norm": 0.1318359375, + "learning_rate": 0.00162338082753168, + "loss": 0.083, + "step": 29734 + }, + { + "epoch": 0.25811407887084314, + "grad_norm": 0.10595703125, + "learning_rate": 0.001623356614834598, + "loss": 0.0679, + "step": 29735 + }, + { + "epoch": 0.2581227593510473, + "grad_norm": 0.53515625, + "learning_rate": 0.0016233324015651812, + "loss": 0.0742, + "step": 29736 + }, + { + "epoch": 0.2581314398312515, + "grad_norm": 1.6484375, + "learning_rate": 0.0016233081877234559, + "loss": 0.165, + "step": 29737 + }, + { + "epoch": 0.2581401203114556, + "grad_norm": 0.298828125, + "learning_rate": 0.0016232839733094485, + "loss": 0.0977, + "step": 29738 + }, + { + "epoch": 0.2581488007916598, + "grad_norm": 0.16015625, + "learning_rate": 0.001623259758323186, + "loss": 0.1104, + "step": 29739 + }, + { + "epoch": 0.25815748127186394, + "grad_norm": 0.287109375, + "learning_rate": 0.001623235542764694, + "loss": 0.1084, + "step": 29740 + }, + { + "epoch": 0.25816616175206814, + "grad_norm": 0.0830078125, + "learning_rate": 0.0016232113266339999, + "loss": 0.0654, + "step": 29741 + }, + { + "epoch": 0.2581748422322723, + "grad_norm": 0.63671875, + "learning_rate": 0.0016231871099311296, + "loss": 0.0908, + "step": 29742 + }, + { + "epoch": 0.25818352271247647, + "grad_norm": 0.640625, + "learning_rate": 0.0016231628926561097, + "loss": 0.2598, + "step": 29743 + }, + { + "epoch": 0.2581922031926806, + "grad_norm": 0.51953125, + "learning_rate": 0.0016231386748089668, + "loss": 0.1064, + "step": 29744 + }, + { + "epoch": 0.2582008836728848, + "grad_norm": 0.10107421875, + "learning_rate": 0.0016231144563897272, + "loss": 0.1367, + "step": 29745 + }, + { + "epoch": 0.25820956415308893, + "grad_norm": 0.48828125, + "learning_rate": 0.0016230902373984173, + "loss": 0.1279, + "step": 29746 + }, + { + "epoch": 0.2582182446332931, + "grad_norm": 0.083984375, + "learning_rate": 0.001623066017835064, + "loss": 0.0942, + "step": 29747 + }, + { + "epoch": 0.25822692511349726, + "grad_norm": 0.146484375, + "learning_rate": 0.0016230417976996938, + "loss": 0.1094, + "step": 29748 + }, + { + "epoch": 0.25823560559370146, + "grad_norm": 0.0810546875, + "learning_rate": 0.0016230175769923329, + "loss": 0.0928, + "step": 29749 + }, + { + "epoch": 0.2582442860739056, + "grad_norm": 0.380859375, + "learning_rate": 0.0016229933557130075, + "loss": 0.0942, + "step": 29750 + }, + { + "epoch": 0.2582529665541098, + "grad_norm": 0.375, + "learning_rate": 0.0016229691338617445, + "loss": 0.1001, + "step": 29751 + }, + { + "epoch": 0.2582616470343139, + "grad_norm": 0.3515625, + "learning_rate": 0.0016229449114385705, + "loss": 0.1089, + "step": 29752 + }, + { + "epoch": 0.2582703275145181, + "grad_norm": 0.267578125, + "learning_rate": 0.0016229206884435115, + "loss": 0.1104, + "step": 29753 + }, + { + "epoch": 0.25827900799472225, + "grad_norm": 0.318359375, + "learning_rate": 0.0016228964648765944, + "loss": 0.1211, + "step": 29754 + }, + { + "epoch": 0.25828768847492645, + "grad_norm": 0.314453125, + "learning_rate": 0.0016228722407378457, + "loss": 0.0942, + "step": 29755 + }, + { + "epoch": 0.2582963689551306, + "grad_norm": 0.1298828125, + "learning_rate": 0.001622848016027292, + "loss": 0.0928, + "step": 29756 + }, + { + "epoch": 0.2583050494353348, + "grad_norm": 0.6640625, + "learning_rate": 0.0016228237907449593, + "loss": 0.1357, + "step": 29757 + }, + { + "epoch": 0.2583137299155389, + "grad_norm": 0.63671875, + "learning_rate": 0.0016227995648908742, + "loss": 0.0762, + "step": 29758 + }, + { + "epoch": 0.2583224103957431, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016227753384650636, + "loss": 0.0879, + "step": 29759 + }, + { + "epoch": 0.25833109087594724, + "grad_norm": 0.388671875, + "learning_rate": 0.0016227511114675535, + "loss": 0.1328, + "step": 29760 + }, + { + "epoch": 0.25833977135615144, + "grad_norm": 0.7890625, + "learning_rate": 0.0016227268838983708, + "loss": 0.1582, + "step": 29761 + }, + { + "epoch": 0.2583484518363556, + "grad_norm": 1.0078125, + "learning_rate": 0.0016227026557575418, + "loss": 0.1328, + "step": 29762 + }, + { + "epoch": 0.25835713231655977, + "grad_norm": 0.1435546875, + "learning_rate": 0.001622678427045093, + "loss": 0.0854, + "step": 29763 + }, + { + "epoch": 0.2583658127967639, + "grad_norm": 0.0947265625, + "learning_rate": 0.001622654197761051, + "loss": 0.0859, + "step": 29764 + }, + { + "epoch": 0.2583744932769681, + "grad_norm": 0.337890625, + "learning_rate": 0.0016226299679054424, + "loss": 0.1045, + "step": 29765 + }, + { + "epoch": 0.25838317375717224, + "grad_norm": 0.3671875, + "learning_rate": 0.0016226057374782931, + "loss": 0.1133, + "step": 29766 + }, + { + "epoch": 0.25839185423737643, + "grad_norm": 0.25390625, + "learning_rate": 0.0016225815064796305, + "loss": 0.0859, + "step": 29767 + }, + { + "epoch": 0.25840053471758057, + "grad_norm": 0.208984375, + "learning_rate": 0.0016225572749094803, + "loss": 0.1094, + "step": 29768 + }, + { + "epoch": 0.25840921519778476, + "grad_norm": 0.39453125, + "learning_rate": 0.0016225330427678695, + "loss": 0.0879, + "step": 29769 + }, + { + "epoch": 0.2584178956779889, + "grad_norm": 0.1923828125, + "learning_rate": 0.0016225088100548242, + "loss": 0.1162, + "step": 29770 + }, + { + "epoch": 0.2584265761581931, + "grad_norm": 0.39453125, + "learning_rate": 0.001622484576770371, + "loss": 0.1089, + "step": 29771 + }, + { + "epoch": 0.2584352566383972, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016224603429145369, + "loss": 0.1064, + "step": 29772 + }, + { + "epoch": 0.2584439371186014, + "grad_norm": 0.6484375, + "learning_rate": 0.0016224361084873481, + "loss": 0.1562, + "step": 29773 + }, + { + "epoch": 0.25845261759880556, + "grad_norm": 0.18359375, + "learning_rate": 0.0016224118734888306, + "loss": 0.1387, + "step": 29774 + }, + { + "epoch": 0.25846129807900975, + "grad_norm": 0.140625, + "learning_rate": 0.0016223876379190115, + "loss": 0.1025, + "step": 29775 + }, + { + "epoch": 0.2584699785592139, + "grad_norm": 0.271484375, + "learning_rate": 0.001622363401777917, + "loss": 0.1094, + "step": 29776 + }, + { + "epoch": 0.2584786590394181, + "grad_norm": 0.08056640625, + "learning_rate": 0.0016223391650655738, + "loss": 0.0918, + "step": 29777 + }, + { + "epoch": 0.2584873395196222, + "grad_norm": 0.201171875, + "learning_rate": 0.0016223149277820087, + "loss": 0.0942, + "step": 29778 + }, + { + "epoch": 0.2584960199998264, + "grad_norm": 0.0859375, + "learning_rate": 0.0016222906899272475, + "loss": 0.1182, + "step": 29779 + }, + { + "epoch": 0.25850470048003055, + "grad_norm": 0.29296875, + "learning_rate": 0.0016222664515013168, + "loss": 0.1143, + "step": 29780 + }, + { + "epoch": 0.25851338096023474, + "grad_norm": 0.2578125, + "learning_rate": 0.0016222422125042437, + "loss": 0.0737, + "step": 29781 + }, + { + "epoch": 0.2585220614404389, + "grad_norm": 0.35546875, + "learning_rate": 0.0016222179729360544, + "loss": 0.0615, + "step": 29782 + }, + { + "epoch": 0.25853074192064307, + "grad_norm": 0.7109375, + "learning_rate": 0.0016221937327967754, + "loss": 0.1064, + "step": 29783 + }, + { + "epoch": 0.2585394224008472, + "grad_norm": 0.34765625, + "learning_rate": 0.0016221694920864327, + "loss": 0.0864, + "step": 29784 + }, + { + "epoch": 0.2585481028810514, + "grad_norm": 0.58203125, + "learning_rate": 0.0016221452508050538, + "loss": 0.1045, + "step": 29785 + }, + { + "epoch": 0.25855678336125554, + "grad_norm": 0.359375, + "learning_rate": 0.0016221210089526646, + "loss": 0.0879, + "step": 29786 + }, + { + "epoch": 0.25856546384145973, + "grad_norm": 0.35546875, + "learning_rate": 0.0016220967665292916, + "loss": 0.1104, + "step": 29787 + }, + { + "epoch": 0.25857414432166387, + "grad_norm": 0.421875, + "learning_rate": 0.0016220725235349613, + "loss": 0.0845, + "step": 29788 + }, + { + "epoch": 0.25858282480186806, + "grad_norm": 0.267578125, + "learning_rate": 0.0016220482799697007, + "loss": 0.103, + "step": 29789 + }, + { + "epoch": 0.2585915052820722, + "grad_norm": 0.65234375, + "learning_rate": 0.0016220240358335357, + "loss": 0.1064, + "step": 29790 + }, + { + "epoch": 0.2586001857622764, + "grad_norm": 0.44921875, + "learning_rate": 0.0016219997911264931, + "loss": 0.1113, + "step": 29791 + }, + { + "epoch": 0.25860886624248053, + "grad_norm": 0.4765625, + "learning_rate": 0.0016219755458485993, + "loss": 0.0874, + "step": 29792 + }, + { + "epoch": 0.2586175467226847, + "grad_norm": 0.427734375, + "learning_rate": 0.001621951299999881, + "loss": 0.1016, + "step": 29793 + }, + { + "epoch": 0.25862622720288886, + "grad_norm": 0.53515625, + "learning_rate": 0.0016219270535803645, + "loss": 0.1035, + "step": 29794 + }, + { + "epoch": 0.25863490768309305, + "grad_norm": 0.62109375, + "learning_rate": 0.0016219028065900766, + "loss": 0.123, + "step": 29795 + }, + { + "epoch": 0.2586435881632972, + "grad_norm": 0.69921875, + "learning_rate": 0.0016218785590290435, + "loss": 0.1338, + "step": 29796 + }, + { + "epoch": 0.2586522686435014, + "grad_norm": 0.322265625, + "learning_rate": 0.001621854310897292, + "loss": 0.0986, + "step": 29797 + }, + { + "epoch": 0.2586609491237055, + "grad_norm": 0.35546875, + "learning_rate": 0.0016218300621948483, + "loss": 0.125, + "step": 29798 + }, + { + "epoch": 0.2586696296039097, + "grad_norm": 0.26171875, + "learning_rate": 0.001621805812921739, + "loss": 0.0825, + "step": 29799 + }, + { + "epoch": 0.25867831008411385, + "grad_norm": 0.60546875, + "learning_rate": 0.0016217815630779908, + "loss": 0.1348, + "step": 29800 + }, + { + "epoch": 0.25868699056431804, + "grad_norm": 0.8125, + "learning_rate": 0.0016217573126636297, + "loss": 0.1187, + "step": 29801 + }, + { + "epoch": 0.2586956710445222, + "grad_norm": 0.4296875, + "learning_rate": 0.0016217330616786834, + "loss": 0.1001, + "step": 29802 + }, + { + "epoch": 0.2587043515247264, + "grad_norm": 0.451171875, + "learning_rate": 0.0016217088101231774, + "loss": 0.0938, + "step": 29803 + }, + { + "epoch": 0.2587130320049305, + "grad_norm": 0.310546875, + "learning_rate": 0.0016216845579971385, + "loss": 0.0962, + "step": 29804 + }, + { + "epoch": 0.2587217124851347, + "grad_norm": 0.12353515625, + "learning_rate": 0.001621660305300593, + "loss": 0.0679, + "step": 29805 + }, + { + "epoch": 0.25873039296533884, + "grad_norm": 0.185546875, + "learning_rate": 0.0016216360520335679, + "loss": 0.1084, + "step": 29806 + }, + { + "epoch": 0.25873907344554303, + "grad_norm": 1.171875, + "learning_rate": 0.001621611798196089, + "loss": 0.1484, + "step": 29807 + }, + { + "epoch": 0.25874775392574717, + "grad_norm": 0.1259765625, + "learning_rate": 0.001621587543788184, + "loss": 0.105, + "step": 29808 + }, + { + "epoch": 0.25875643440595136, + "grad_norm": 0.166015625, + "learning_rate": 0.001621563288809878, + "loss": 0.1055, + "step": 29809 + }, + { + "epoch": 0.2587651148861555, + "grad_norm": 0.51171875, + "learning_rate": 0.0016215390332611987, + "loss": 0.1094, + "step": 29810 + }, + { + "epoch": 0.2587737953663597, + "grad_norm": 0.232421875, + "learning_rate": 0.0016215147771421717, + "loss": 0.0938, + "step": 29811 + }, + { + "epoch": 0.25878247584656383, + "grad_norm": 0.173828125, + "learning_rate": 0.0016214905204528242, + "loss": 0.1162, + "step": 29812 + }, + { + "epoch": 0.258791156326768, + "grad_norm": 0.55078125, + "learning_rate": 0.0016214662631931828, + "loss": 0.1318, + "step": 29813 + }, + { + "epoch": 0.25879983680697216, + "grad_norm": 0.53515625, + "learning_rate": 0.001621442005363273, + "loss": 0.124, + "step": 29814 + }, + { + "epoch": 0.25880851728717635, + "grad_norm": 0.37109375, + "learning_rate": 0.0016214177469631229, + "loss": 0.0918, + "step": 29815 + }, + { + "epoch": 0.2588171977673805, + "grad_norm": 0.25, + "learning_rate": 0.0016213934879927576, + "loss": 0.1089, + "step": 29816 + }, + { + "epoch": 0.25882587824758463, + "grad_norm": 0.1083984375, + "learning_rate": 0.0016213692284522045, + "loss": 0.1201, + "step": 29817 + }, + { + "epoch": 0.2588345587277888, + "grad_norm": 0.30078125, + "learning_rate": 0.0016213449683414897, + "loss": 0.085, + "step": 29818 + }, + { + "epoch": 0.25884323920799296, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016213207076606403, + "loss": 0.1465, + "step": 29819 + }, + { + "epoch": 0.25885191968819715, + "grad_norm": 0.3828125, + "learning_rate": 0.001621296446409682, + "loss": 0.127, + "step": 29820 + }, + { + "epoch": 0.2588606001684013, + "grad_norm": 0.30859375, + "learning_rate": 0.0016212721845886419, + "loss": 0.1504, + "step": 29821 + }, + { + "epoch": 0.2588692806486055, + "grad_norm": 0.98828125, + "learning_rate": 0.001621247922197546, + "loss": 0.1406, + "step": 29822 + }, + { + "epoch": 0.2588779611288096, + "grad_norm": 0.2890625, + "learning_rate": 0.0016212236592364213, + "loss": 0.0996, + "step": 29823 + }, + { + "epoch": 0.2588866416090138, + "grad_norm": 0.16015625, + "learning_rate": 0.0016211993957052945, + "loss": 0.1084, + "step": 29824 + }, + { + "epoch": 0.25889532208921795, + "grad_norm": 0.06884765625, + "learning_rate": 0.0016211751316041918, + "loss": 0.0776, + "step": 29825 + }, + { + "epoch": 0.25890400256942214, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016211508669331399, + "loss": 0.1055, + "step": 29826 + }, + { + "epoch": 0.2589126830496263, + "grad_norm": 0.322265625, + "learning_rate": 0.001621126601692165, + "loss": 0.1387, + "step": 29827 + }, + { + "epoch": 0.2589213635298305, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016211023358812938, + "loss": 0.0889, + "step": 29828 + }, + { + "epoch": 0.2589300440100346, + "grad_norm": 0.369140625, + "learning_rate": 0.0016210780695005533, + "loss": 0.123, + "step": 29829 + }, + { + "epoch": 0.2589387244902388, + "grad_norm": 0.69921875, + "learning_rate": 0.0016210538025499695, + "loss": 0.1025, + "step": 29830 + }, + { + "epoch": 0.25894740497044294, + "grad_norm": 0.443359375, + "learning_rate": 0.0016210295350295688, + "loss": 0.1279, + "step": 29831 + }, + { + "epoch": 0.25895608545064713, + "grad_norm": 0.16796875, + "learning_rate": 0.0016210052669393786, + "loss": 0.0869, + "step": 29832 + }, + { + "epoch": 0.25896476593085127, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016209809982794243, + "loss": 0.0859, + "step": 29833 + }, + { + "epoch": 0.25897344641105546, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016209567290497333, + "loss": 0.1055, + "step": 29834 + }, + { + "epoch": 0.2589821268912596, + "grad_norm": 0.08056640625, + "learning_rate": 0.001620932459250332, + "loss": 0.0698, + "step": 29835 + }, + { + "epoch": 0.2589908073714638, + "grad_norm": 0.189453125, + "learning_rate": 0.0016209081888812465, + "loss": 0.0806, + "step": 29836 + }, + { + "epoch": 0.25899948785166793, + "grad_norm": 0.55078125, + "learning_rate": 0.0016208839179425037, + "loss": 0.126, + "step": 29837 + }, + { + "epoch": 0.2590081683318721, + "grad_norm": 0.197265625, + "learning_rate": 0.00162085964643413, + "loss": 0.0859, + "step": 29838 + }, + { + "epoch": 0.25901684881207626, + "grad_norm": 0.5390625, + "learning_rate": 0.0016208353743561521, + "loss": 0.0898, + "step": 29839 + }, + { + "epoch": 0.25902552929228045, + "grad_norm": 0.44921875, + "learning_rate": 0.0016208111017085966, + "loss": 0.1006, + "step": 29840 + }, + { + "epoch": 0.2590342097724846, + "grad_norm": 0.138671875, + "learning_rate": 0.0016207868284914898, + "loss": 0.1133, + "step": 29841 + }, + { + "epoch": 0.2590428902526888, + "grad_norm": 0.177734375, + "learning_rate": 0.0016207625547048584, + "loss": 0.1465, + "step": 29842 + }, + { + "epoch": 0.2590515707328929, + "grad_norm": 0.3359375, + "learning_rate": 0.0016207382803487289, + "loss": 0.0879, + "step": 29843 + }, + { + "epoch": 0.2590602512130971, + "grad_norm": 0.1328125, + "learning_rate": 0.0016207140054231275, + "loss": 0.1348, + "step": 29844 + }, + { + "epoch": 0.25906893169330125, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016206897299280813, + "loss": 0.085, + "step": 29845 + }, + { + "epoch": 0.25907761217350544, + "grad_norm": 0.212890625, + "learning_rate": 0.0016206654538636167, + "loss": 0.1094, + "step": 29846 + }, + { + "epoch": 0.2590862926537096, + "grad_norm": 0.640625, + "learning_rate": 0.0016206411772297605, + "loss": 0.105, + "step": 29847 + }, + { + "epoch": 0.2590949731339138, + "grad_norm": 0.353515625, + "learning_rate": 0.0016206169000265384, + "loss": 0.1025, + "step": 29848 + }, + { + "epoch": 0.2591036536141179, + "grad_norm": 0.130859375, + "learning_rate": 0.0016205926222539774, + "loss": 0.1123, + "step": 29849 + }, + { + "epoch": 0.2591123340943221, + "grad_norm": 0.330078125, + "learning_rate": 0.0016205683439121047, + "loss": 0.1104, + "step": 29850 + }, + { + "epoch": 0.25912101457452624, + "grad_norm": 0.267578125, + "learning_rate": 0.001620544065000946, + "loss": 0.0962, + "step": 29851 + }, + { + "epoch": 0.25912969505473044, + "grad_norm": 0.10986328125, + "learning_rate": 0.0016205197855205284, + "loss": 0.1055, + "step": 29852 + }, + { + "epoch": 0.2591383755349346, + "grad_norm": 0.1484375, + "learning_rate": 0.0016204955054708776, + "loss": 0.0576, + "step": 29853 + }, + { + "epoch": 0.25914705601513877, + "grad_norm": 0.27734375, + "learning_rate": 0.001620471224852021, + "loss": 0.1133, + "step": 29854 + }, + { + "epoch": 0.2591557364953429, + "grad_norm": 0.19140625, + "learning_rate": 0.001620446943663985, + "loss": 0.0869, + "step": 29855 + }, + { + "epoch": 0.2591644169755471, + "grad_norm": 0.205078125, + "learning_rate": 0.001620422661906796, + "loss": 0.1133, + "step": 29856 + }, + { + "epoch": 0.25917309745575123, + "grad_norm": 0.09765625, + "learning_rate": 0.0016203983795804804, + "loss": 0.1299, + "step": 29857 + }, + { + "epoch": 0.2591817779359554, + "grad_norm": 0.10693359375, + "learning_rate": 0.0016203740966850653, + "loss": 0.1045, + "step": 29858 + }, + { + "epoch": 0.25919045841615956, + "grad_norm": 0.486328125, + "learning_rate": 0.0016203498132205767, + "loss": 0.0977, + "step": 29859 + }, + { + "epoch": 0.25919913889636376, + "grad_norm": 0.353515625, + "learning_rate": 0.0016203255291870413, + "loss": 0.1436, + "step": 29860 + }, + { + "epoch": 0.2592078193765679, + "grad_norm": 1.0703125, + "learning_rate": 0.0016203012445844861, + "loss": 0.123, + "step": 29861 + }, + { + "epoch": 0.2592164998567721, + "grad_norm": 0.255859375, + "learning_rate": 0.0016202769594129368, + "loss": 0.0908, + "step": 29862 + }, + { + "epoch": 0.2592251803369762, + "grad_norm": 0.158203125, + "learning_rate": 0.0016202526736724206, + "loss": 0.1279, + "step": 29863 + }, + { + "epoch": 0.2592338608171804, + "grad_norm": 0.1748046875, + "learning_rate": 0.001620228387362964, + "loss": 0.0908, + "step": 29864 + }, + { + "epoch": 0.25924254129738455, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016202041004845934, + "loss": 0.1406, + "step": 29865 + }, + { + "epoch": 0.25925122177758875, + "grad_norm": 0.7578125, + "learning_rate": 0.0016201798130373354, + "loss": 0.207, + "step": 29866 + }, + { + "epoch": 0.2592599022577929, + "grad_norm": 0.34375, + "learning_rate": 0.0016201555250212165, + "loss": 0.1016, + "step": 29867 + }, + { + "epoch": 0.2592685827379971, + "grad_norm": 0.267578125, + "learning_rate": 0.0016201312364362633, + "loss": 0.1123, + "step": 29868 + }, + { + "epoch": 0.2592772632182012, + "grad_norm": 0.333984375, + "learning_rate": 0.0016201069472825025, + "loss": 0.0947, + "step": 29869 + }, + { + "epoch": 0.2592859436984054, + "grad_norm": 0.09619140625, + "learning_rate": 0.0016200826575599604, + "loss": 0.1064, + "step": 29870 + }, + { + "epoch": 0.25929462417860955, + "grad_norm": 0.5546875, + "learning_rate": 0.001620058367268664, + "loss": 0.0981, + "step": 29871 + }, + { + "epoch": 0.25930330465881374, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016200340764086394, + "loss": 0.1084, + "step": 29872 + }, + { + "epoch": 0.2593119851390179, + "grad_norm": 0.146484375, + "learning_rate": 0.0016200097849799133, + "loss": 0.1113, + "step": 29873 + }, + { + "epoch": 0.25932066561922207, + "grad_norm": 0.22265625, + "learning_rate": 0.0016199854929825125, + "loss": 0.083, + "step": 29874 + }, + { + "epoch": 0.2593293460994262, + "grad_norm": 0.265625, + "learning_rate": 0.0016199612004164634, + "loss": 0.1504, + "step": 29875 + }, + { + "epoch": 0.2593380265796304, + "grad_norm": 0.185546875, + "learning_rate": 0.0016199369072817922, + "loss": 0.1006, + "step": 29876 + }, + { + "epoch": 0.25934670705983454, + "grad_norm": 0.115234375, + "learning_rate": 0.0016199126135785264, + "loss": 0.0933, + "step": 29877 + }, + { + "epoch": 0.25935538754003873, + "grad_norm": 0.181640625, + "learning_rate": 0.0016198883193066914, + "loss": 0.1045, + "step": 29878 + }, + { + "epoch": 0.25936406802024287, + "grad_norm": 0.25390625, + "learning_rate": 0.0016198640244663145, + "loss": 0.1162, + "step": 29879 + }, + { + "epoch": 0.25937274850044706, + "grad_norm": 0.345703125, + "learning_rate": 0.001619839729057422, + "loss": 0.1309, + "step": 29880 + }, + { + "epoch": 0.2593814289806512, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016198154330800407, + "loss": 0.1035, + "step": 29881 + }, + { + "epoch": 0.2593901094608554, + "grad_norm": 0.39453125, + "learning_rate": 0.0016197911365341973, + "loss": 0.127, + "step": 29882 + }, + { + "epoch": 0.2593987899410595, + "grad_norm": 0.69140625, + "learning_rate": 0.0016197668394199178, + "loss": 0.1069, + "step": 29883 + }, + { + "epoch": 0.2594074704212637, + "grad_norm": 0.09814453125, + "learning_rate": 0.0016197425417372293, + "loss": 0.0781, + "step": 29884 + }, + { + "epoch": 0.25941615090146786, + "grad_norm": 0.30859375, + "learning_rate": 0.001619718243486158, + "loss": 0.1699, + "step": 29885 + }, + { + "epoch": 0.25942483138167205, + "grad_norm": 0.1220703125, + "learning_rate": 0.0016196939446667309, + "loss": 0.1118, + "step": 29886 + }, + { + "epoch": 0.2594335118618762, + "grad_norm": 0.22265625, + "learning_rate": 0.001619669645278974, + "loss": 0.1396, + "step": 29887 + }, + { + "epoch": 0.2594421923420804, + "grad_norm": 0.330078125, + "learning_rate": 0.0016196453453229142, + "loss": 0.0835, + "step": 29888 + }, + { + "epoch": 0.2594508728222845, + "grad_norm": 0.283203125, + "learning_rate": 0.0016196210447985784, + "loss": 0.0967, + "step": 29889 + }, + { + "epoch": 0.2594595533024887, + "grad_norm": 0.353515625, + "learning_rate": 0.0016195967437059925, + "loss": 0.1089, + "step": 29890 + }, + { + "epoch": 0.25946823378269285, + "grad_norm": 1.3359375, + "learning_rate": 0.0016195724420451836, + "loss": 0.1162, + "step": 29891 + }, + { + "epoch": 0.25947691426289704, + "grad_norm": 0.12890625, + "learning_rate": 0.0016195481398161777, + "loss": 0.0996, + "step": 29892 + }, + { + "epoch": 0.2594855947431012, + "grad_norm": 0.11328125, + "learning_rate": 0.0016195238370190024, + "loss": 0.0923, + "step": 29893 + }, + { + "epoch": 0.25949427522330537, + "grad_norm": 0.2099609375, + "learning_rate": 0.0016194995336536832, + "loss": 0.1152, + "step": 29894 + }, + { + "epoch": 0.2595029557035095, + "grad_norm": 0.474609375, + "learning_rate": 0.0016194752297202472, + "loss": 0.1055, + "step": 29895 + }, + { + "epoch": 0.2595116361837137, + "grad_norm": 0.51171875, + "learning_rate": 0.001619450925218721, + "loss": 0.1172, + "step": 29896 + }, + { + "epoch": 0.25952031666391784, + "grad_norm": 0.57421875, + "learning_rate": 0.0016194266201491308, + "loss": 0.0952, + "step": 29897 + }, + { + "epoch": 0.25952899714412203, + "grad_norm": 0.34375, + "learning_rate": 0.001619402314511504, + "loss": 0.1133, + "step": 29898 + }, + { + "epoch": 0.25953767762432617, + "grad_norm": 0.70703125, + "learning_rate": 0.001619378008305866, + "loss": 0.0854, + "step": 29899 + }, + { + "epoch": 0.25954635810453036, + "grad_norm": 0.45703125, + "learning_rate": 0.0016193537015322442, + "loss": 0.1123, + "step": 29900 + }, + { + "epoch": 0.2595550385847345, + "grad_norm": 0.578125, + "learning_rate": 0.001619329394190665, + "loss": 0.1201, + "step": 29901 + }, + { + "epoch": 0.2595637190649387, + "grad_norm": 0.78515625, + "learning_rate": 0.001619305086281155, + "loss": 0.0918, + "step": 29902 + }, + { + "epoch": 0.25957239954514283, + "grad_norm": 0.5078125, + "learning_rate": 0.001619280777803741, + "loss": 0.127, + "step": 29903 + }, + { + "epoch": 0.259581080025347, + "grad_norm": 0.392578125, + "learning_rate": 0.0016192564687584495, + "loss": 0.1172, + "step": 29904 + }, + { + "epoch": 0.25958976050555116, + "grad_norm": 0.5546875, + "learning_rate": 0.001619232159145306, + "loss": 0.0923, + "step": 29905 + }, + { + "epoch": 0.25959844098575535, + "grad_norm": 0.109375, + "learning_rate": 0.0016192078489643386, + "loss": 0.1201, + "step": 29906 + }, + { + "epoch": 0.2596071214659595, + "grad_norm": 0.4296875, + "learning_rate": 0.0016191835382155738, + "loss": 0.0703, + "step": 29907 + }, + { + "epoch": 0.2596158019461637, + "grad_norm": 0.42578125, + "learning_rate": 0.001619159226899037, + "loss": 0.1875, + "step": 29908 + }, + { + "epoch": 0.2596244824263678, + "grad_norm": 0.158203125, + "learning_rate": 0.0016191349150147555, + "loss": 0.0967, + "step": 29909 + }, + { + "epoch": 0.259633162906572, + "grad_norm": 0.12890625, + "learning_rate": 0.001619110602562756, + "loss": 0.0889, + "step": 29910 + }, + { + "epoch": 0.25964184338677615, + "grad_norm": 0.80859375, + "learning_rate": 0.001619086289543065, + "loss": 0.0981, + "step": 29911 + }, + { + "epoch": 0.25965052386698034, + "grad_norm": 0.1318359375, + "learning_rate": 0.001619061975955709, + "loss": 0.1279, + "step": 29912 + }, + { + "epoch": 0.2596592043471845, + "grad_norm": 0.6484375, + "learning_rate": 0.0016190376618007145, + "loss": 0.1011, + "step": 29913 + }, + { + "epoch": 0.2596678848273887, + "grad_norm": 0.08447265625, + "learning_rate": 0.0016190133470781083, + "loss": 0.0781, + "step": 29914 + }, + { + "epoch": 0.2596765653075928, + "grad_norm": 0.1494140625, + "learning_rate": 0.001618989031787917, + "loss": 0.0708, + "step": 29915 + }, + { + "epoch": 0.259685245787797, + "grad_norm": 0.12158203125, + "learning_rate": 0.001618964715930167, + "loss": 0.0649, + "step": 29916 + }, + { + "epoch": 0.25969392626800114, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016189403995048853, + "loss": 0.1533, + "step": 29917 + }, + { + "epoch": 0.25970260674820533, + "grad_norm": 0.46484375, + "learning_rate": 0.0016189160825120974, + "loss": 0.1465, + "step": 29918 + }, + { + "epoch": 0.25971128722840947, + "grad_norm": 0.162109375, + "learning_rate": 0.0016188917649518312, + "loss": 0.1182, + "step": 29919 + }, + { + "epoch": 0.25971996770861366, + "grad_norm": 0.6875, + "learning_rate": 0.0016188674468241128, + "loss": 0.0757, + "step": 29920 + }, + { + "epoch": 0.2597286481888178, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016188431281289688, + "loss": 0.0991, + "step": 29921 + }, + { + "epoch": 0.259737328669022, + "grad_norm": 0.0849609375, + "learning_rate": 0.0016188188088664256, + "loss": 0.0947, + "step": 29922 + }, + { + "epoch": 0.25974600914922613, + "grad_norm": 0.095703125, + "learning_rate": 0.00161879448903651, + "loss": 0.0747, + "step": 29923 + }, + { + "epoch": 0.2597546896294303, + "grad_norm": 0.107421875, + "learning_rate": 0.0016187701686392482, + "loss": 0.0898, + "step": 29924 + }, + { + "epoch": 0.25976337010963446, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016187458476746676, + "loss": 0.1064, + "step": 29925 + }, + { + "epoch": 0.25977205058983865, + "grad_norm": 0.69140625, + "learning_rate": 0.0016187215261427939, + "loss": 0.1006, + "step": 29926 + }, + { + "epoch": 0.2597807310700428, + "grad_norm": 0.291015625, + "learning_rate": 0.0016186972040436545, + "loss": 0.127, + "step": 29927 + }, + { + "epoch": 0.259789411550247, + "grad_norm": 0.396484375, + "learning_rate": 0.0016186728813772755, + "loss": 0.0537, + "step": 29928 + }, + { + "epoch": 0.2597980920304511, + "grad_norm": 0.298828125, + "learning_rate": 0.001618648558143684, + "loss": 0.0962, + "step": 29929 + }, + { + "epoch": 0.2598067725106553, + "grad_norm": 0.10986328125, + "learning_rate": 0.001618624234342906, + "loss": 0.0918, + "step": 29930 + }, + { + "epoch": 0.25981545299085945, + "grad_norm": 1.1484375, + "learning_rate": 0.001618599909974968, + "loss": 0.085, + "step": 29931 + }, + { + "epoch": 0.25982413347106365, + "grad_norm": 0.1328125, + "learning_rate": 0.0016185755850398969, + "loss": 0.0947, + "step": 29932 + }, + { + "epoch": 0.2598328139512678, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016185512595377196, + "loss": 0.125, + "step": 29933 + }, + { + "epoch": 0.259841494431472, + "grad_norm": 0.53515625, + "learning_rate": 0.0016185269334684626, + "loss": 0.0815, + "step": 29934 + }, + { + "epoch": 0.2598501749116761, + "grad_norm": 0.328125, + "learning_rate": 0.001618502606832152, + "loss": 0.1328, + "step": 29935 + }, + { + "epoch": 0.2598588553918803, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016184782796288151, + "loss": 0.1143, + "step": 29936 + }, + { + "epoch": 0.25986753587208444, + "grad_norm": 0.25390625, + "learning_rate": 0.0016184539518584778, + "loss": 0.1602, + "step": 29937 + }, + { + "epoch": 0.25987621635228864, + "grad_norm": 0.326171875, + "learning_rate": 0.001618429623521167, + "loss": 0.1055, + "step": 29938 + }, + { + "epoch": 0.2598848968324928, + "grad_norm": 0.29296875, + "learning_rate": 0.0016184052946169097, + "loss": 0.1445, + "step": 29939 + }, + { + "epoch": 0.2598935773126969, + "grad_norm": 0.189453125, + "learning_rate": 0.0016183809651457322, + "loss": 0.1152, + "step": 29940 + }, + { + "epoch": 0.2599022577929011, + "grad_norm": 0.5546875, + "learning_rate": 0.0016183566351076605, + "loss": 0.0889, + "step": 29941 + }, + { + "epoch": 0.25991093827310524, + "grad_norm": 0.181640625, + "learning_rate": 0.0016183323045027222, + "loss": 0.0835, + "step": 29942 + }, + { + "epoch": 0.25991961875330943, + "grad_norm": 0.384765625, + "learning_rate": 0.0016183079733309437, + "loss": 0.1245, + "step": 29943 + }, + { + "epoch": 0.25992829923351357, + "grad_norm": 0.10595703125, + "learning_rate": 0.001618283641592351, + "loss": 0.0894, + "step": 29944 + }, + { + "epoch": 0.25993697971371776, + "grad_norm": 0.6171875, + "learning_rate": 0.0016182593092869714, + "loss": 0.1064, + "step": 29945 + }, + { + "epoch": 0.2599456601939219, + "grad_norm": 0.7734375, + "learning_rate": 0.0016182349764148308, + "loss": 0.1152, + "step": 29946 + }, + { + "epoch": 0.2599543406741261, + "grad_norm": 0.201171875, + "learning_rate": 0.0016182106429759562, + "loss": 0.0918, + "step": 29947 + }, + { + "epoch": 0.25996302115433023, + "grad_norm": 0.2890625, + "learning_rate": 0.0016181863089703744, + "loss": 0.1494, + "step": 29948 + }, + { + "epoch": 0.2599717016345344, + "grad_norm": 0.361328125, + "learning_rate": 0.001618161974398112, + "loss": 0.0928, + "step": 29949 + }, + { + "epoch": 0.25998038211473856, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016181376392591955, + "loss": 0.0962, + "step": 29950 + }, + { + "epoch": 0.25998906259494275, + "grad_norm": 0.06884765625, + "learning_rate": 0.0016181133035536512, + "loss": 0.0869, + "step": 29951 + }, + { + "epoch": 0.2599977430751469, + "grad_norm": 1.7734375, + "learning_rate": 0.0016180889672815061, + "loss": 0.2021, + "step": 29952 + }, + { + "epoch": 0.2600064235553511, + "grad_norm": 0.392578125, + "learning_rate": 0.0016180646304427865, + "loss": 0.1206, + "step": 29953 + }, + { + "epoch": 0.2600151040355552, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016180402930375194, + "loss": 0.0996, + "step": 29954 + }, + { + "epoch": 0.2600237845157594, + "grad_norm": 0.087890625, + "learning_rate": 0.0016180159550657311, + "loss": 0.103, + "step": 29955 + }, + { + "epoch": 0.26003246499596355, + "grad_norm": 0.111328125, + "learning_rate": 0.0016179916165274485, + "loss": 0.1045, + "step": 29956 + }, + { + "epoch": 0.26004114547616775, + "grad_norm": 0.578125, + "learning_rate": 0.0016179672774226982, + "loss": 0.1094, + "step": 29957 + }, + { + "epoch": 0.2600498259563719, + "grad_norm": 0.64453125, + "learning_rate": 0.0016179429377515063, + "loss": 0.1055, + "step": 29958 + }, + { + "epoch": 0.2600585064365761, + "grad_norm": 0.07568359375, + "learning_rate": 0.0016179185975139, + "loss": 0.0986, + "step": 29959 + }, + { + "epoch": 0.2600671869167802, + "grad_norm": 0.0869140625, + "learning_rate": 0.0016178942567099056, + "loss": 0.0781, + "step": 29960 + }, + { + "epoch": 0.2600758673969844, + "grad_norm": 0.138671875, + "learning_rate": 0.0016178699153395497, + "loss": 0.0996, + "step": 29961 + }, + { + "epoch": 0.26008454787718854, + "grad_norm": 0.23828125, + "learning_rate": 0.0016178455734028592, + "loss": 0.085, + "step": 29962 + }, + { + "epoch": 0.26009322835739274, + "grad_norm": 0.47265625, + "learning_rate": 0.0016178212308998608, + "loss": 0.0898, + "step": 29963 + }, + { + "epoch": 0.2601019088375969, + "grad_norm": 0.318359375, + "learning_rate": 0.0016177968878305804, + "loss": 0.0654, + "step": 29964 + }, + { + "epoch": 0.26011058931780107, + "grad_norm": 0.298828125, + "learning_rate": 0.0016177725441950455, + "loss": 0.1201, + "step": 29965 + }, + { + "epoch": 0.2601192697980052, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016177481999932822, + "loss": 0.0913, + "step": 29966 + }, + { + "epoch": 0.2601279502782094, + "grad_norm": 0.271484375, + "learning_rate": 0.0016177238552253172, + "loss": 0.1143, + "step": 29967 + }, + { + "epoch": 0.26013663075841353, + "grad_norm": 0.40625, + "learning_rate": 0.001617699509891177, + "loss": 0.103, + "step": 29968 + }, + { + "epoch": 0.2601453112386177, + "grad_norm": 1.1015625, + "learning_rate": 0.0016176751639908886, + "loss": 0.0894, + "step": 29969 + }, + { + "epoch": 0.26015399171882186, + "grad_norm": 0.275390625, + "learning_rate": 0.0016176508175244784, + "loss": 0.0703, + "step": 29970 + }, + { + "epoch": 0.26016267219902606, + "grad_norm": 1.0546875, + "learning_rate": 0.001617626470491973, + "loss": 0.1279, + "step": 29971 + }, + { + "epoch": 0.2601713526792302, + "grad_norm": 0.2021484375, + "learning_rate": 0.0016176021228933991, + "loss": 0.1182, + "step": 29972 + }, + { + "epoch": 0.2601800331594344, + "grad_norm": 0.2109375, + "learning_rate": 0.0016175777747287833, + "loss": 0.0811, + "step": 29973 + }, + { + "epoch": 0.2601887136396385, + "grad_norm": 0.3515625, + "learning_rate": 0.0016175534259981522, + "loss": 0.1016, + "step": 29974 + }, + { + "epoch": 0.2601973941198427, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016175290767015324, + "loss": 0.1016, + "step": 29975 + }, + { + "epoch": 0.26020607460004685, + "grad_norm": 0.5546875, + "learning_rate": 0.0016175047268389509, + "loss": 0.0908, + "step": 29976 + }, + { + "epoch": 0.26021475508025105, + "grad_norm": 0.328125, + "learning_rate": 0.0016174803764104338, + "loss": 0.0796, + "step": 29977 + }, + { + "epoch": 0.2602234355604552, + "grad_norm": 0.365234375, + "learning_rate": 0.0016174560254160076, + "loss": 0.0859, + "step": 29978 + }, + { + "epoch": 0.2602321160406594, + "grad_norm": 1.21875, + "learning_rate": 0.0016174316738556999, + "loss": 0.1211, + "step": 29979 + }, + { + "epoch": 0.2602407965208635, + "grad_norm": 0.232421875, + "learning_rate": 0.0016174073217295362, + "loss": 0.1416, + "step": 29980 + }, + { + "epoch": 0.2602494770010677, + "grad_norm": 0.369140625, + "learning_rate": 0.0016173829690375437, + "loss": 0.0977, + "step": 29981 + }, + { + "epoch": 0.26025815748127185, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016173586157797488, + "loss": 0.1143, + "step": 29982 + }, + { + "epoch": 0.26026683796147604, + "grad_norm": 1.125, + "learning_rate": 0.0016173342619561785, + "loss": 0.2295, + "step": 29983 + }, + { + "epoch": 0.2602755184416802, + "grad_norm": 0.18359375, + "learning_rate": 0.0016173099075668593, + "loss": 0.0869, + "step": 29984 + }, + { + "epoch": 0.26028419892188437, + "grad_norm": 0.25390625, + "learning_rate": 0.001617285552611818, + "loss": 0.1533, + "step": 29985 + }, + { + "epoch": 0.2602928794020885, + "grad_norm": 0.140625, + "learning_rate": 0.0016172611970910807, + "loss": 0.1006, + "step": 29986 + }, + { + "epoch": 0.2603015598822927, + "grad_norm": 1.546875, + "learning_rate": 0.001617236841004674, + "loss": 0.1055, + "step": 29987 + }, + { + "epoch": 0.26031024036249684, + "grad_norm": 0.46484375, + "learning_rate": 0.0016172124843526255, + "loss": 0.0684, + "step": 29988 + }, + { + "epoch": 0.26031892084270103, + "grad_norm": 0.392578125, + "learning_rate": 0.0016171881271349609, + "loss": 0.0996, + "step": 29989 + }, + { + "epoch": 0.26032760132290517, + "grad_norm": 0.54296875, + "learning_rate": 0.0016171637693517071, + "loss": 0.1016, + "step": 29990 + }, + { + "epoch": 0.26033628180310936, + "grad_norm": 0.1513671875, + "learning_rate": 0.0016171394110028907, + "loss": 0.0898, + "step": 29991 + }, + { + "epoch": 0.2603449622833135, + "grad_norm": 0.328125, + "learning_rate": 0.0016171150520885385, + "loss": 0.0908, + "step": 29992 + }, + { + "epoch": 0.2603536427635177, + "grad_norm": 0.2158203125, + "learning_rate": 0.001617090692608677, + "loss": 0.0918, + "step": 29993 + }, + { + "epoch": 0.2603623232437218, + "grad_norm": 0.1064453125, + "learning_rate": 0.0016170663325633331, + "loss": 0.1123, + "step": 29994 + }, + { + "epoch": 0.260371003723926, + "grad_norm": 0.234375, + "learning_rate": 0.001617041971952533, + "loss": 0.0791, + "step": 29995 + }, + { + "epoch": 0.26037968420413016, + "grad_norm": 0.30078125, + "learning_rate": 0.0016170176107763038, + "loss": 0.1465, + "step": 29996 + }, + { + "epoch": 0.26038836468433435, + "grad_norm": 0.228515625, + "learning_rate": 0.0016169932490346717, + "loss": 0.1162, + "step": 29997 + }, + { + "epoch": 0.2603970451645385, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016169688867276636, + "loss": 0.0938, + "step": 29998 + }, + { + "epoch": 0.2604057256447427, + "grad_norm": 0.2158203125, + "learning_rate": 0.001616944523855306, + "loss": 0.1035, + "step": 29999 + }, + { + "epoch": 0.2604144061249468, + "grad_norm": 0.61328125, + "learning_rate": 0.0016169201604176258, + "loss": 0.083, + "step": 30000 + }, + { + "epoch": 0.260423086605151, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016168957964146494, + "loss": 0.1172, + "step": 30001 + }, + { + "epoch": 0.26043176708535515, + "grad_norm": 0.1962890625, + "learning_rate": 0.0016168714318464034, + "loss": 0.1094, + "step": 30002 + }, + { + "epoch": 0.26044044756555934, + "grad_norm": 0.291015625, + "learning_rate": 0.001616847066712915, + "loss": 0.1035, + "step": 30003 + }, + { + "epoch": 0.2604491280457635, + "grad_norm": 0.1318359375, + "learning_rate": 0.0016168227010142098, + "loss": 0.125, + "step": 30004 + }, + { + "epoch": 0.26045780852596767, + "grad_norm": 0.123046875, + "learning_rate": 0.0016167983347503154, + "loss": 0.0718, + "step": 30005 + }, + { + "epoch": 0.2604664890061718, + "grad_norm": 0.306640625, + "learning_rate": 0.001616773967921258, + "loss": 0.1045, + "step": 30006 + }, + { + "epoch": 0.260475169486376, + "grad_norm": 0.240234375, + "learning_rate": 0.0016167496005270645, + "loss": 0.1226, + "step": 30007 + }, + { + "epoch": 0.26048384996658014, + "grad_norm": 0.380859375, + "learning_rate": 0.0016167252325677613, + "loss": 0.0811, + "step": 30008 + }, + { + "epoch": 0.26049253044678433, + "grad_norm": 0.091796875, + "learning_rate": 0.0016167008640433753, + "loss": 0.0771, + "step": 30009 + }, + { + "epoch": 0.26050121092698847, + "grad_norm": 0.279296875, + "learning_rate": 0.0016166764949539327, + "loss": 0.106, + "step": 30010 + }, + { + "epoch": 0.26050989140719266, + "grad_norm": 0.8359375, + "learning_rate": 0.0016166521252994608, + "loss": 0.1045, + "step": 30011 + }, + { + "epoch": 0.2605185718873968, + "grad_norm": 0.205078125, + "learning_rate": 0.0016166277550799858, + "loss": 0.104, + "step": 30012 + }, + { + "epoch": 0.260527252367601, + "grad_norm": 0.21484375, + "learning_rate": 0.001616603384295534, + "loss": 0.0623, + "step": 30013 + }, + { + "epoch": 0.26053593284780513, + "grad_norm": 0.232421875, + "learning_rate": 0.001616579012946133, + "loss": 0.1396, + "step": 30014 + }, + { + "epoch": 0.2605446133280093, + "grad_norm": 0.435546875, + "learning_rate": 0.001616554641031809, + "loss": 0.1162, + "step": 30015 + }, + { + "epoch": 0.26055329380821346, + "grad_norm": 0.6796875, + "learning_rate": 0.0016165302685525885, + "loss": 0.1133, + "step": 30016 + }, + { + "epoch": 0.26056197428841765, + "grad_norm": 2.359375, + "learning_rate": 0.001616505895508498, + "loss": 0.3574, + "step": 30017 + }, + { + "epoch": 0.2605706547686218, + "grad_norm": 0.76171875, + "learning_rate": 0.0016164815218995647, + "loss": 0.1143, + "step": 30018 + }, + { + "epoch": 0.260579335248826, + "grad_norm": 0.5234375, + "learning_rate": 0.001616457147725815, + "loss": 0.1128, + "step": 30019 + }, + { + "epoch": 0.2605880157290301, + "grad_norm": 0.251953125, + "learning_rate": 0.0016164327729872754, + "loss": 0.0879, + "step": 30020 + }, + { + "epoch": 0.2605966962092343, + "grad_norm": 0.09130859375, + "learning_rate": 0.0016164083976839723, + "loss": 0.0845, + "step": 30021 + }, + { + "epoch": 0.26060537668943845, + "grad_norm": 0.4921875, + "learning_rate": 0.0016163840218159334, + "loss": 0.1025, + "step": 30022 + }, + { + "epoch": 0.26061405716964264, + "grad_norm": 0.75, + "learning_rate": 0.0016163596453831843, + "loss": 0.1289, + "step": 30023 + }, + { + "epoch": 0.2606227376498468, + "grad_norm": 0.30078125, + "learning_rate": 0.0016163352683857522, + "loss": 0.1328, + "step": 30024 + }, + { + "epoch": 0.260631418130051, + "grad_norm": 0.31640625, + "learning_rate": 0.0016163108908236637, + "loss": 0.0786, + "step": 30025 + }, + { + "epoch": 0.2606400986102551, + "grad_norm": 0.671875, + "learning_rate": 0.001616286512696945, + "loss": 0.0869, + "step": 30026 + }, + { + "epoch": 0.2606487790904593, + "grad_norm": 0.267578125, + "learning_rate": 0.0016162621340056235, + "loss": 0.1494, + "step": 30027 + }, + { + "epoch": 0.26065745957066344, + "grad_norm": 0.2265625, + "learning_rate": 0.0016162377547497254, + "loss": 0.1338, + "step": 30028 + }, + { + "epoch": 0.26066614005086763, + "grad_norm": 0.55859375, + "learning_rate": 0.0016162133749292773, + "loss": 0.1099, + "step": 30029 + }, + { + "epoch": 0.26067482053107177, + "grad_norm": 0.72265625, + "learning_rate": 0.001616188994544306, + "loss": 0.1133, + "step": 30030 + }, + { + "epoch": 0.26068350101127596, + "grad_norm": 0.220703125, + "learning_rate": 0.0016161646135948384, + "loss": 0.1025, + "step": 30031 + }, + { + "epoch": 0.2606921814914801, + "grad_norm": 0.16015625, + "learning_rate": 0.0016161402320809008, + "loss": 0.1108, + "step": 30032 + }, + { + "epoch": 0.2607008619716843, + "grad_norm": 0.2001953125, + "learning_rate": 0.00161611585000252, + "loss": 0.1064, + "step": 30033 + }, + { + "epoch": 0.26070954245188843, + "grad_norm": 0.162109375, + "learning_rate": 0.0016160914673597226, + "loss": 0.1016, + "step": 30034 + }, + { + "epoch": 0.2607182229320926, + "grad_norm": 0.310546875, + "learning_rate": 0.0016160670841525353, + "loss": 0.0781, + "step": 30035 + }, + { + "epoch": 0.26072690341229676, + "grad_norm": 0.14453125, + "learning_rate": 0.001616042700380985, + "loss": 0.0728, + "step": 30036 + }, + { + "epoch": 0.26073558389250096, + "grad_norm": 0.5234375, + "learning_rate": 0.0016160183160450982, + "loss": 0.1484, + "step": 30037 + }, + { + "epoch": 0.2607442643727051, + "grad_norm": 0.09033203125, + "learning_rate": 0.0016159939311449012, + "loss": 0.103, + "step": 30038 + }, + { + "epoch": 0.2607529448529093, + "grad_norm": 0.35546875, + "learning_rate": 0.001615969545680421, + "loss": 0.0791, + "step": 30039 + }, + { + "epoch": 0.2607616253331134, + "grad_norm": 0.41796875, + "learning_rate": 0.0016159451596516848, + "loss": 0.0947, + "step": 30040 + }, + { + "epoch": 0.2607703058133176, + "grad_norm": 0.2216796875, + "learning_rate": 0.001615920773058718, + "loss": 0.1152, + "step": 30041 + }, + { + "epoch": 0.26077898629352175, + "grad_norm": 0.28515625, + "learning_rate": 0.0016158963859015483, + "loss": 0.1113, + "step": 30042 + }, + { + "epoch": 0.26078766677372595, + "grad_norm": 0.275390625, + "learning_rate": 0.0016158719981802022, + "loss": 0.0977, + "step": 30043 + }, + { + "epoch": 0.2607963472539301, + "grad_norm": 0.125, + "learning_rate": 0.001615847609894706, + "loss": 0.0776, + "step": 30044 + }, + { + "epoch": 0.2608050277341343, + "grad_norm": 0.18359375, + "learning_rate": 0.001615823221045087, + "loss": 0.0991, + "step": 30045 + }, + { + "epoch": 0.2608137082143384, + "grad_norm": 0.150390625, + "learning_rate": 0.0016157988316313713, + "loss": 0.0942, + "step": 30046 + }, + { + "epoch": 0.2608223886945426, + "grad_norm": 0.353515625, + "learning_rate": 0.0016157744416535854, + "loss": 0.1006, + "step": 30047 + }, + { + "epoch": 0.26083106917474674, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016157500511117568, + "loss": 0.0889, + "step": 30048 + }, + { + "epoch": 0.26083974965495094, + "grad_norm": 0.60546875, + "learning_rate": 0.0016157256600059114, + "loss": 0.085, + "step": 30049 + }, + { + "epoch": 0.2608484301351551, + "grad_norm": 0.154296875, + "learning_rate": 0.0016157012683360762, + "loss": 0.1289, + "step": 30050 + }, + { + "epoch": 0.26085711061535927, + "grad_norm": 0.40625, + "learning_rate": 0.0016156768761022777, + "loss": 0.1172, + "step": 30051 + }, + { + "epoch": 0.2608657910955634, + "grad_norm": 0.31640625, + "learning_rate": 0.0016156524833045428, + "loss": 0.1221, + "step": 30052 + }, + { + "epoch": 0.2608744715757676, + "grad_norm": 0.0966796875, + "learning_rate": 0.0016156280899428984, + "loss": 0.1191, + "step": 30053 + }, + { + "epoch": 0.26088315205597173, + "grad_norm": 0.125, + "learning_rate": 0.0016156036960173709, + "loss": 0.084, + "step": 30054 + }, + { + "epoch": 0.2608918325361759, + "grad_norm": 0.28125, + "learning_rate": 0.0016155793015279864, + "loss": 0.0615, + "step": 30055 + }, + { + "epoch": 0.26090051301638006, + "grad_norm": 0.384765625, + "learning_rate": 0.0016155549064747724, + "loss": 0.1396, + "step": 30056 + }, + { + "epoch": 0.26090919349658426, + "grad_norm": 0.49609375, + "learning_rate": 0.0016155305108577554, + "loss": 0.0776, + "step": 30057 + }, + { + "epoch": 0.2609178739767884, + "grad_norm": 0.212890625, + "learning_rate": 0.0016155061146769618, + "loss": 0.1436, + "step": 30058 + }, + { + "epoch": 0.2609265544569926, + "grad_norm": 0.1318359375, + "learning_rate": 0.0016154817179324184, + "loss": 0.1064, + "step": 30059 + }, + { + "epoch": 0.2609352349371967, + "grad_norm": 0.193359375, + "learning_rate": 0.0016154573206241523, + "loss": 0.1123, + "step": 30060 + }, + { + "epoch": 0.2609439154174009, + "grad_norm": 0.7890625, + "learning_rate": 0.0016154329227521895, + "loss": 0.1426, + "step": 30061 + }, + { + "epoch": 0.26095259589760506, + "grad_norm": 0.5703125, + "learning_rate": 0.0016154085243165572, + "loss": 0.0972, + "step": 30062 + }, + { + "epoch": 0.26096127637780925, + "grad_norm": 0.478515625, + "learning_rate": 0.0016153841253172816, + "loss": 0.1133, + "step": 30063 + }, + { + "epoch": 0.2609699568580134, + "grad_norm": 0.099609375, + "learning_rate": 0.00161535972575439, + "loss": 0.0884, + "step": 30064 + }, + { + "epoch": 0.2609786373382175, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016153353256279086, + "loss": 0.0791, + "step": 30065 + }, + { + "epoch": 0.2609873178184217, + "grad_norm": 0.443359375, + "learning_rate": 0.0016153109249378642, + "loss": 0.1123, + "step": 30066 + }, + { + "epoch": 0.26099599829862585, + "grad_norm": 0.099609375, + "learning_rate": 0.0016152865236842835, + "loss": 0.0986, + "step": 30067 + }, + { + "epoch": 0.26100467877883005, + "grad_norm": 0.1533203125, + "learning_rate": 0.0016152621218671935, + "loss": 0.1309, + "step": 30068 + }, + { + "epoch": 0.2610133592590342, + "grad_norm": 0.13671875, + "learning_rate": 0.00161523771948662, + "loss": 0.1074, + "step": 30069 + }, + { + "epoch": 0.2610220397392384, + "grad_norm": 1.625, + "learning_rate": 0.0016152133165425906, + "loss": 0.2217, + "step": 30070 + }, + { + "epoch": 0.2610307202194425, + "grad_norm": 0.26171875, + "learning_rate": 0.001615188913035132, + "loss": 0.1055, + "step": 30071 + }, + { + "epoch": 0.2610394006996467, + "grad_norm": 0.1611328125, + "learning_rate": 0.00161516450896427, + "loss": 0.1309, + "step": 30072 + }, + { + "epoch": 0.26104808117985084, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016151401043300322, + "loss": 0.1006, + "step": 30073 + }, + { + "epoch": 0.26105676166005504, + "grad_norm": 0.6640625, + "learning_rate": 0.0016151156991324445, + "loss": 0.1104, + "step": 30074 + }, + { + "epoch": 0.2610654421402592, + "grad_norm": 0.76171875, + "learning_rate": 0.0016150912933715345, + "loss": 0.1387, + "step": 30075 + }, + { + "epoch": 0.26107412262046337, + "grad_norm": 0.251953125, + "learning_rate": 0.0016150668870473282, + "loss": 0.0908, + "step": 30076 + }, + { + "epoch": 0.2610828031006675, + "grad_norm": 0.953125, + "learning_rate": 0.0016150424801598525, + "loss": 0.2051, + "step": 30077 + }, + { + "epoch": 0.2610914835808717, + "grad_norm": 0.1181640625, + "learning_rate": 0.0016150180727091338, + "loss": 0.0903, + "step": 30078 + }, + { + "epoch": 0.26110016406107583, + "grad_norm": 0.380859375, + "learning_rate": 0.0016149936646951994, + "loss": 0.1123, + "step": 30079 + }, + { + "epoch": 0.26110884454128, + "grad_norm": 0.193359375, + "learning_rate": 0.0016149692561180757, + "loss": 0.1235, + "step": 30080 + }, + { + "epoch": 0.26111752502148416, + "grad_norm": 0.279296875, + "learning_rate": 0.0016149448469777894, + "loss": 0.1123, + "step": 30081 + }, + { + "epoch": 0.26112620550168836, + "grad_norm": 0.111328125, + "learning_rate": 0.0016149204372743668, + "loss": 0.0918, + "step": 30082 + }, + { + "epoch": 0.2611348859818925, + "grad_norm": 0.189453125, + "learning_rate": 0.001614896027007835, + "loss": 0.1289, + "step": 30083 + }, + { + "epoch": 0.2611435664620967, + "grad_norm": 0.4921875, + "learning_rate": 0.001614871616178221, + "loss": 0.1211, + "step": 30084 + }, + { + "epoch": 0.2611522469423008, + "grad_norm": 0.361328125, + "learning_rate": 0.001614847204785551, + "loss": 0.127, + "step": 30085 + }, + { + "epoch": 0.261160927422505, + "grad_norm": 0.205078125, + "learning_rate": 0.0016148227928298516, + "loss": 0.0923, + "step": 30086 + }, + { + "epoch": 0.26116960790270916, + "grad_norm": 0.314453125, + "learning_rate": 0.00161479838031115, + "loss": 0.1289, + "step": 30087 + }, + { + "epoch": 0.26117828838291335, + "grad_norm": 0.609375, + "learning_rate": 0.0016147739672294724, + "loss": 0.1475, + "step": 30088 + }, + { + "epoch": 0.2611869688631175, + "grad_norm": 0.259765625, + "learning_rate": 0.0016147495535848459, + "loss": 0.1055, + "step": 30089 + }, + { + "epoch": 0.2611956493433217, + "grad_norm": 0.671875, + "learning_rate": 0.0016147251393772968, + "loss": 0.0718, + "step": 30090 + }, + { + "epoch": 0.2612043298235258, + "grad_norm": 0.75390625, + "learning_rate": 0.0016147007246068523, + "loss": 0.1553, + "step": 30091 + }, + { + "epoch": 0.26121301030373, + "grad_norm": 0.125, + "learning_rate": 0.0016146763092735386, + "loss": 0.0718, + "step": 30092 + }, + { + "epoch": 0.26122169078393415, + "grad_norm": 0.486328125, + "learning_rate": 0.0016146518933773829, + "loss": 0.0801, + "step": 30093 + }, + { + "epoch": 0.26123037126413834, + "grad_norm": 0.080078125, + "learning_rate": 0.0016146274769184112, + "loss": 0.0811, + "step": 30094 + }, + { + "epoch": 0.2612390517443425, + "grad_norm": 0.123046875, + "learning_rate": 0.001614603059896651, + "loss": 0.0835, + "step": 30095 + }, + { + "epoch": 0.26124773222454667, + "grad_norm": 0.5078125, + "learning_rate": 0.0016145786423121284, + "loss": 0.0996, + "step": 30096 + }, + { + "epoch": 0.2612564127047508, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016145542241648705, + "loss": 0.1309, + "step": 30097 + }, + { + "epoch": 0.261265093184955, + "grad_norm": 0.244140625, + "learning_rate": 0.0016145298054549037, + "loss": 0.0737, + "step": 30098 + }, + { + "epoch": 0.26127377366515914, + "grad_norm": 0.8515625, + "learning_rate": 0.001614505386182255, + "loss": 0.0967, + "step": 30099 + }, + { + "epoch": 0.26128245414536333, + "grad_norm": 0.09033203125, + "learning_rate": 0.0016144809663469508, + "loss": 0.1074, + "step": 30100 + }, + { + "epoch": 0.26129113462556747, + "grad_norm": 0.2216796875, + "learning_rate": 0.001614456545949018, + "loss": 0.0928, + "step": 30101 + }, + { + "epoch": 0.26129981510577166, + "grad_norm": 0.470703125, + "learning_rate": 0.0016144321249884832, + "loss": 0.1084, + "step": 30102 + }, + { + "epoch": 0.2613084955859758, + "grad_norm": 0.302734375, + "learning_rate": 0.001614407703465373, + "loss": 0.0859, + "step": 30103 + }, + { + "epoch": 0.26131717606618, + "grad_norm": 0.45703125, + "learning_rate": 0.0016143832813797145, + "loss": 0.124, + "step": 30104 + }, + { + "epoch": 0.2613258565463841, + "grad_norm": 0.42578125, + "learning_rate": 0.0016143588587315342, + "loss": 0.0659, + "step": 30105 + }, + { + "epoch": 0.2613345370265883, + "grad_norm": 0.431640625, + "learning_rate": 0.001614334435520859, + "loss": 0.126, + "step": 30106 + }, + { + "epoch": 0.26134321750679246, + "grad_norm": 0.4140625, + "learning_rate": 0.0016143100117477149, + "loss": 0.1138, + "step": 30107 + }, + { + "epoch": 0.26135189798699665, + "grad_norm": 0.609375, + "learning_rate": 0.0016142855874121292, + "loss": 0.1602, + "step": 30108 + }, + { + "epoch": 0.2613605784672008, + "grad_norm": 0.92578125, + "learning_rate": 0.0016142611625141288, + "loss": 0.1777, + "step": 30109 + }, + { + "epoch": 0.261369258947405, + "grad_norm": 0.6015625, + "learning_rate": 0.0016142367370537397, + "loss": 0.1484, + "step": 30110 + }, + { + "epoch": 0.2613779394276091, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016142123110309892, + "loss": 0.1396, + "step": 30111 + }, + { + "epoch": 0.2613866199078133, + "grad_norm": 0.333984375, + "learning_rate": 0.0016141878844459037, + "loss": 0.1172, + "step": 30112 + }, + { + "epoch": 0.26139530038801745, + "grad_norm": 0.07958984375, + "learning_rate": 0.0016141634572985107, + "loss": 0.0908, + "step": 30113 + }, + { + "epoch": 0.26140398086822164, + "grad_norm": 0.15234375, + "learning_rate": 0.0016141390295888356, + "loss": 0.1377, + "step": 30114 + }, + { + "epoch": 0.2614126613484258, + "grad_norm": 0.240234375, + "learning_rate": 0.001614114601316906, + "loss": 0.082, + "step": 30115 + }, + { + "epoch": 0.26142134182862997, + "grad_norm": 0.212890625, + "learning_rate": 0.0016140901724827485, + "loss": 0.1162, + "step": 30116 + }, + { + "epoch": 0.2614300223088341, + "grad_norm": 0.0791015625, + "learning_rate": 0.0016140657430863893, + "loss": 0.0996, + "step": 30117 + }, + { + "epoch": 0.2614387027890383, + "grad_norm": 0.39453125, + "learning_rate": 0.001614041313127856, + "loss": 0.084, + "step": 30118 + }, + { + "epoch": 0.26144738326924244, + "grad_norm": 0.10205078125, + "learning_rate": 0.0016140168826071746, + "loss": 0.1006, + "step": 30119 + }, + { + "epoch": 0.26145606374944663, + "grad_norm": 0.4453125, + "learning_rate": 0.0016139924515243722, + "loss": 0.1387, + "step": 30120 + }, + { + "epoch": 0.26146474422965077, + "grad_norm": 0.10986328125, + "learning_rate": 0.0016139680198794754, + "loss": 0.0825, + "step": 30121 + }, + { + "epoch": 0.26147342470985496, + "grad_norm": 0.48828125, + "learning_rate": 0.0016139435876725108, + "loss": 0.082, + "step": 30122 + }, + { + "epoch": 0.2614821051900591, + "grad_norm": 0.326171875, + "learning_rate": 0.001613919154903505, + "loss": 0.1074, + "step": 30123 + }, + { + "epoch": 0.2614907856702633, + "grad_norm": 0.271484375, + "learning_rate": 0.0016138947215724852, + "loss": 0.1455, + "step": 30124 + }, + { + "epoch": 0.26149946615046743, + "grad_norm": 0.103515625, + "learning_rate": 0.0016138702876794778, + "loss": 0.1035, + "step": 30125 + }, + { + "epoch": 0.2615081466306716, + "grad_norm": 0.29296875, + "learning_rate": 0.0016138458532245097, + "loss": 0.1367, + "step": 30126 + }, + { + "epoch": 0.26151682711087576, + "grad_norm": 0.166015625, + "learning_rate": 0.0016138214182076073, + "loss": 0.1084, + "step": 30127 + }, + { + "epoch": 0.26152550759107995, + "grad_norm": 0.40234375, + "learning_rate": 0.001613796982628798, + "loss": 0.127, + "step": 30128 + }, + { + "epoch": 0.2615341880712841, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016137725464881072, + "loss": 0.1045, + "step": 30129 + }, + { + "epoch": 0.2615428685514883, + "grad_norm": 0.40234375, + "learning_rate": 0.0016137481097855632, + "loss": 0.0898, + "step": 30130 + }, + { + "epoch": 0.2615515490316924, + "grad_norm": 0.279296875, + "learning_rate": 0.0016137236725211914, + "loss": 0.0938, + "step": 30131 + }, + { + "epoch": 0.2615602295118966, + "grad_norm": 0.57421875, + "learning_rate": 0.0016136992346950194, + "loss": 0.0977, + "step": 30132 + }, + { + "epoch": 0.26156890999210075, + "grad_norm": 0.51953125, + "learning_rate": 0.0016136747963070737, + "loss": 0.0928, + "step": 30133 + }, + { + "epoch": 0.26157759047230494, + "grad_norm": 0.2099609375, + "learning_rate": 0.001613650357357381, + "loss": 0.1094, + "step": 30134 + }, + { + "epoch": 0.2615862709525091, + "grad_norm": 0.30078125, + "learning_rate": 0.0016136259178459677, + "loss": 0.1465, + "step": 30135 + }, + { + "epoch": 0.2615949514327133, + "grad_norm": 1.5859375, + "learning_rate": 0.001613601477772861, + "loss": 0.4355, + "step": 30136 + }, + { + "epoch": 0.2616036319129174, + "grad_norm": 0.484375, + "learning_rate": 0.0016135770371380875, + "loss": 0.1504, + "step": 30137 + }, + { + "epoch": 0.2616123123931216, + "grad_norm": 0.373046875, + "learning_rate": 0.0016135525959416737, + "loss": 0.1299, + "step": 30138 + }, + { + "epoch": 0.26162099287332574, + "grad_norm": 0.07275390625, + "learning_rate": 0.001613528154183647, + "loss": 0.1045, + "step": 30139 + }, + { + "epoch": 0.26162967335352993, + "grad_norm": 0.1005859375, + "learning_rate": 0.001613503711864033, + "loss": 0.0781, + "step": 30140 + }, + { + "epoch": 0.26163835383373407, + "grad_norm": 0.2080078125, + "learning_rate": 0.0016134792689828593, + "loss": 0.105, + "step": 30141 + }, + { + "epoch": 0.26164703431393826, + "grad_norm": 0.37109375, + "learning_rate": 0.0016134548255401521, + "loss": 0.1436, + "step": 30142 + }, + { + "epoch": 0.2616557147941424, + "grad_norm": 0.279296875, + "learning_rate": 0.0016134303815359388, + "loss": 0.0928, + "step": 30143 + }, + { + "epoch": 0.2616643952743466, + "grad_norm": 0.08544921875, + "learning_rate": 0.0016134059369702456, + "loss": 0.1143, + "step": 30144 + }, + { + "epoch": 0.26167307575455073, + "grad_norm": 0.57421875, + "learning_rate": 0.0016133814918430993, + "loss": 0.1621, + "step": 30145 + }, + { + "epoch": 0.2616817562347549, + "grad_norm": 0.18359375, + "learning_rate": 0.0016133570461545272, + "loss": 0.1113, + "step": 30146 + }, + { + "epoch": 0.26169043671495906, + "grad_norm": 0.1689453125, + "learning_rate": 0.001613332599904555, + "loss": 0.1118, + "step": 30147 + }, + { + "epoch": 0.26169911719516326, + "grad_norm": 0.09375, + "learning_rate": 0.0016133081530932101, + "loss": 0.1001, + "step": 30148 + }, + { + "epoch": 0.2617077976753674, + "grad_norm": 0.28125, + "learning_rate": 0.0016132837057205192, + "loss": 0.0957, + "step": 30149 + }, + { + "epoch": 0.2617164781555716, + "grad_norm": 0.19140625, + "learning_rate": 0.0016132592577865089, + "loss": 0.0884, + "step": 30150 + }, + { + "epoch": 0.2617251586357757, + "grad_norm": 0.21484375, + "learning_rate": 0.001613234809291206, + "loss": 0.0977, + "step": 30151 + }, + { + "epoch": 0.2617338391159799, + "grad_norm": 0.23828125, + "learning_rate": 0.0016132103602346374, + "loss": 0.1631, + "step": 30152 + }, + { + "epoch": 0.26174251959618405, + "grad_norm": 0.2890625, + "learning_rate": 0.0016131859106168294, + "loss": 0.124, + "step": 30153 + }, + { + "epoch": 0.26175120007638825, + "grad_norm": 0.390625, + "learning_rate": 0.0016131614604378093, + "loss": 0.0796, + "step": 30154 + }, + { + "epoch": 0.2617598805565924, + "grad_norm": 0.24609375, + "learning_rate": 0.001613137009697603, + "loss": 0.0981, + "step": 30155 + }, + { + "epoch": 0.2617685610367966, + "grad_norm": 0.609375, + "learning_rate": 0.0016131125583962382, + "loss": 0.1079, + "step": 30156 + }, + { + "epoch": 0.2617772415170007, + "grad_norm": 0.29296875, + "learning_rate": 0.001613088106533741, + "loss": 0.103, + "step": 30157 + }, + { + "epoch": 0.2617859219972049, + "grad_norm": 0.498046875, + "learning_rate": 0.0016130636541101385, + "loss": 0.1562, + "step": 30158 + }, + { + "epoch": 0.26179460247740904, + "grad_norm": 0.306640625, + "learning_rate": 0.0016130392011254574, + "loss": 0.0957, + "step": 30159 + }, + { + "epoch": 0.26180328295761324, + "grad_norm": 0.150390625, + "learning_rate": 0.001613014747579724, + "loss": 0.105, + "step": 30160 + }, + { + "epoch": 0.2618119634378174, + "grad_norm": 0.451171875, + "learning_rate": 0.0016129902934729657, + "loss": 0.0859, + "step": 30161 + }, + { + "epoch": 0.26182064391802157, + "grad_norm": 0.5234375, + "learning_rate": 0.001612965838805209, + "loss": 0.1641, + "step": 30162 + }, + { + "epoch": 0.2618293243982257, + "grad_norm": 0.26171875, + "learning_rate": 0.0016129413835764804, + "loss": 0.1338, + "step": 30163 + }, + { + "epoch": 0.2618380048784299, + "grad_norm": 0.10595703125, + "learning_rate": 0.0016129169277868068, + "loss": 0.1094, + "step": 30164 + }, + { + "epoch": 0.26184668535863403, + "grad_norm": 0.2197265625, + "learning_rate": 0.0016128924714362148, + "loss": 0.1016, + "step": 30165 + }, + { + "epoch": 0.2618553658388382, + "grad_norm": 0.10888671875, + "learning_rate": 0.0016128680145247314, + "loss": 0.1221, + "step": 30166 + }, + { + "epoch": 0.26186404631904237, + "grad_norm": 0.333984375, + "learning_rate": 0.0016128435570523835, + "loss": 0.1045, + "step": 30167 + }, + { + "epoch": 0.26187272679924656, + "grad_norm": 0.298828125, + "learning_rate": 0.001612819099019197, + "loss": 0.0791, + "step": 30168 + }, + { + "epoch": 0.2618814072794507, + "grad_norm": 0.35546875, + "learning_rate": 0.0016127946404252, + "loss": 0.1367, + "step": 30169 + }, + { + "epoch": 0.2618900877596549, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016127701812704179, + "loss": 0.083, + "step": 30170 + }, + { + "epoch": 0.261898768239859, + "grad_norm": 0.439453125, + "learning_rate": 0.0016127457215548784, + "loss": 0.1079, + "step": 30171 + }, + { + "epoch": 0.2619074487200632, + "grad_norm": 0.333984375, + "learning_rate": 0.0016127212612786076, + "loss": 0.1187, + "step": 30172 + }, + { + "epoch": 0.26191612920026736, + "grad_norm": 0.5625, + "learning_rate": 0.0016126968004416327, + "loss": 0.1758, + "step": 30173 + }, + { + "epoch": 0.26192480968047155, + "grad_norm": 0.384765625, + "learning_rate": 0.00161267233904398, + "loss": 0.0938, + "step": 30174 + }, + { + "epoch": 0.2619334901606757, + "grad_norm": 0.06787109375, + "learning_rate": 0.001612647877085677, + "loss": 0.0747, + "step": 30175 + }, + { + "epoch": 0.2619421706408799, + "grad_norm": 0.10107421875, + "learning_rate": 0.00161262341456675, + "loss": 0.1309, + "step": 30176 + }, + { + "epoch": 0.261950851121084, + "grad_norm": 0.291015625, + "learning_rate": 0.001612598951487225, + "loss": 0.0986, + "step": 30177 + }, + { + "epoch": 0.2619595316012882, + "grad_norm": 0.07958984375, + "learning_rate": 0.0016125744878471302, + "loss": 0.1201, + "step": 30178 + }, + { + "epoch": 0.26196821208149235, + "grad_norm": 0.1845703125, + "learning_rate": 0.0016125500236464914, + "loss": 0.1104, + "step": 30179 + }, + { + "epoch": 0.26197689256169654, + "grad_norm": 0.38671875, + "learning_rate": 0.0016125255588853356, + "loss": 0.1113, + "step": 30180 + }, + { + "epoch": 0.2619855730419007, + "grad_norm": 0.1826171875, + "learning_rate": 0.0016125010935636895, + "loss": 0.0845, + "step": 30181 + }, + { + "epoch": 0.26199425352210487, + "grad_norm": 0.142578125, + "learning_rate": 0.00161247662768158, + "loss": 0.0962, + "step": 30182 + }, + { + "epoch": 0.262002934002309, + "grad_norm": 0.15234375, + "learning_rate": 0.0016124521612390337, + "loss": 0.0928, + "step": 30183 + }, + { + "epoch": 0.2620116144825132, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016124276942360775, + "loss": 0.1079, + "step": 30184 + }, + { + "epoch": 0.26202029496271734, + "grad_norm": 0.103515625, + "learning_rate": 0.0016124032266727382, + "loss": 0.1133, + "step": 30185 + }, + { + "epoch": 0.26202897544292153, + "grad_norm": 0.353515625, + "learning_rate": 0.0016123787585490422, + "loss": 0.0898, + "step": 30186 + }, + { + "epoch": 0.26203765592312567, + "grad_norm": 1.046875, + "learning_rate": 0.0016123542898650166, + "loss": 0.1387, + "step": 30187 + }, + { + "epoch": 0.2620463364033298, + "grad_norm": 0.1689453125, + "learning_rate": 0.0016123298206206877, + "loss": 0.0918, + "step": 30188 + }, + { + "epoch": 0.262055016883534, + "grad_norm": 0.609375, + "learning_rate": 0.0016123053508160833, + "loss": 0.1602, + "step": 30189 + }, + { + "epoch": 0.26206369736373814, + "grad_norm": 0.1455078125, + "learning_rate": 0.0016122808804512289, + "loss": 0.1201, + "step": 30190 + }, + { + "epoch": 0.26207237784394233, + "grad_norm": 0.27734375, + "learning_rate": 0.001612256409526152, + "loss": 0.1221, + "step": 30191 + }, + { + "epoch": 0.26208105832414647, + "grad_norm": 0.1474609375, + "learning_rate": 0.001612231938040879, + "loss": 0.0996, + "step": 30192 + }, + { + "epoch": 0.26208973880435066, + "grad_norm": 0.236328125, + "learning_rate": 0.0016122074659954372, + "loss": 0.1172, + "step": 30193 + }, + { + "epoch": 0.2620984192845548, + "grad_norm": 0.2021484375, + "learning_rate": 0.0016121829933898523, + "loss": 0.1021, + "step": 30194 + }, + { + "epoch": 0.262107099764759, + "grad_norm": 0.1201171875, + "learning_rate": 0.0016121585202241524, + "loss": 0.1104, + "step": 30195 + }, + { + "epoch": 0.2621157802449631, + "grad_norm": 0.333984375, + "learning_rate": 0.0016121340464983638, + "loss": 0.0962, + "step": 30196 + }, + { + "epoch": 0.2621244607251673, + "grad_norm": 0.248046875, + "learning_rate": 0.0016121095722125129, + "loss": 0.1152, + "step": 30197 + }, + { + "epoch": 0.26213314120537146, + "grad_norm": 0.279296875, + "learning_rate": 0.0016120850973666266, + "loss": 0.083, + "step": 30198 + }, + { + "epoch": 0.26214182168557565, + "grad_norm": 0.376953125, + "learning_rate": 0.0016120606219607316, + "loss": 0.0859, + "step": 30199 + }, + { + "epoch": 0.2621505021657798, + "grad_norm": 0.578125, + "learning_rate": 0.001612036145994855, + "loss": 0.1079, + "step": 30200 + }, + { + "epoch": 0.262159182645984, + "grad_norm": 0.28515625, + "learning_rate": 0.0016120116694690232, + "loss": 0.0845, + "step": 30201 + }, + { + "epoch": 0.2621678631261881, + "grad_norm": 0.44140625, + "learning_rate": 0.0016119871923832635, + "loss": 0.0776, + "step": 30202 + }, + { + "epoch": 0.2621765436063923, + "grad_norm": 0.13671875, + "learning_rate": 0.0016119627147376018, + "loss": 0.1113, + "step": 30203 + }, + { + "epoch": 0.26218522408659645, + "grad_norm": 0.279296875, + "learning_rate": 0.0016119382365320657, + "loss": 0.1348, + "step": 30204 + }, + { + "epoch": 0.26219390456680064, + "grad_norm": 0.40234375, + "learning_rate": 0.0016119137577666816, + "loss": 0.1079, + "step": 30205 + }, + { + "epoch": 0.2622025850470048, + "grad_norm": 0.125, + "learning_rate": 0.001611889278441476, + "loss": 0.085, + "step": 30206 + }, + { + "epoch": 0.26221126552720897, + "grad_norm": 0.453125, + "learning_rate": 0.0016118647985564763, + "loss": 0.1133, + "step": 30207 + }, + { + "epoch": 0.2622199460074131, + "grad_norm": 0.1650390625, + "learning_rate": 0.0016118403181117089, + "loss": 0.105, + "step": 30208 + }, + { + "epoch": 0.2622286264876173, + "grad_norm": 0.10546875, + "learning_rate": 0.0016118158371072008, + "loss": 0.0713, + "step": 30209 + }, + { + "epoch": 0.26223730696782144, + "grad_norm": 1.0859375, + "learning_rate": 0.001611791355542978, + "loss": 0.1582, + "step": 30210 + }, + { + "epoch": 0.26224598744802563, + "grad_norm": 0.2236328125, + "learning_rate": 0.0016117668734190686, + "loss": 0.0898, + "step": 30211 + }, + { + "epoch": 0.26225466792822977, + "grad_norm": 0.302734375, + "learning_rate": 0.001611742390735498, + "loss": 0.1006, + "step": 30212 + }, + { + "epoch": 0.26226334840843396, + "grad_norm": 0.16796875, + "learning_rate": 0.0016117179074922941, + "loss": 0.1084, + "step": 30213 + }, + { + "epoch": 0.2622720288886381, + "grad_norm": 0.298828125, + "learning_rate": 0.001611693423689483, + "loss": 0.0908, + "step": 30214 + }, + { + "epoch": 0.2622807093688423, + "grad_norm": 0.19140625, + "learning_rate": 0.0016116689393270913, + "loss": 0.1006, + "step": 30215 + }, + { + "epoch": 0.26228938984904643, + "grad_norm": 0.11767578125, + "learning_rate": 0.0016116444544051465, + "loss": 0.0903, + "step": 30216 + }, + { + "epoch": 0.2622980703292506, + "grad_norm": 0.1962890625, + "learning_rate": 0.001611619968923675, + "loss": 0.1191, + "step": 30217 + }, + { + "epoch": 0.26230675080945476, + "grad_norm": 0.41796875, + "learning_rate": 0.0016115954828827038, + "loss": 0.1226, + "step": 30218 + }, + { + "epoch": 0.26231543128965895, + "grad_norm": 0.58984375, + "learning_rate": 0.001611570996282259, + "loss": 0.0996, + "step": 30219 + }, + { + "epoch": 0.2623241117698631, + "grad_norm": 0.4453125, + "learning_rate": 0.0016115465091223681, + "loss": 0.084, + "step": 30220 + }, + { + "epoch": 0.2623327922500673, + "grad_norm": 0.30859375, + "learning_rate": 0.0016115220214030575, + "loss": 0.125, + "step": 30221 + }, + { + "epoch": 0.2623414727302714, + "grad_norm": 0.099609375, + "learning_rate": 0.0016114975331243543, + "loss": 0.0908, + "step": 30222 + }, + { + "epoch": 0.2623501532104756, + "grad_norm": 0.62109375, + "learning_rate": 0.0016114730442862851, + "loss": 0.0674, + "step": 30223 + }, + { + "epoch": 0.26235883369067975, + "grad_norm": 0.30078125, + "learning_rate": 0.0016114485548888767, + "loss": 0.1011, + "step": 30224 + }, + { + "epoch": 0.26236751417088394, + "grad_norm": 0.1865234375, + "learning_rate": 0.0016114240649321555, + "loss": 0.1084, + "step": 30225 + }, + { + "epoch": 0.2623761946510881, + "grad_norm": 0.71875, + "learning_rate": 0.0016113995744161489, + "loss": 0.1406, + "step": 30226 + }, + { + "epoch": 0.26238487513129227, + "grad_norm": 0.423828125, + "learning_rate": 0.0016113750833408832, + "loss": 0.124, + "step": 30227 + }, + { + "epoch": 0.2623935556114964, + "grad_norm": 0.0966796875, + "learning_rate": 0.0016113505917063858, + "loss": 0.127, + "step": 30228 + }, + { + "epoch": 0.2624022360917006, + "grad_norm": 0.154296875, + "learning_rate": 0.0016113260995126825, + "loss": 0.1016, + "step": 30229 + }, + { + "epoch": 0.26241091657190474, + "grad_norm": 0.56640625, + "learning_rate": 0.0016113016067598011, + "loss": 0.083, + "step": 30230 + }, + { + "epoch": 0.26241959705210893, + "grad_norm": 0.08740234375, + "learning_rate": 0.0016112771134477678, + "loss": 0.1162, + "step": 30231 + }, + { + "epoch": 0.26242827753231307, + "grad_norm": 0.275390625, + "learning_rate": 0.0016112526195766097, + "loss": 0.0786, + "step": 30232 + }, + { + "epoch": 0.26243695801251726, + "grad_norm": 0.2080078125, + "learning_rate": 0.001611228125146353, + "loss": 0.1104, + "step": 30233 + }, + { + "epoch": 0.2624456384927214, + "grad_norm": 0.25390625, + "learning_rate": 0.001611203630157025, + "loss": 0.1553, + "step": 30234 + }, + { + "epoch": 0.2624543189729256, + "grad_norm": 0.271484375, + "learning_rate": 0.0016111791346086527, + "loss": 0.1079, + "step": 30235 + }, + { + "epoch": 0.26246299945312973, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016111546385012626, + "loss": 0.0806, + "step": 30236 + }, + { + "epoch": 0.2624716799333339, + "grad_norm": 0.408203125, + "learning_rate": 0.0016111301418348812, + "loss": 0.1143, + "step": 30237 + }, + { + "epoch": 0.26248036041353806, + "grad_norm": 0.146484375, + "learning_rate": 0.0016111056446095355, + "loss": 0.0791, + "step": 30238 + }, + { + "epoch": 0.26248904089374225, + "grad_norm": 0.33984375, + "learning_rate": 0.0016110811468252524, + "loss": 0.0938, + "step": 30239 + }, + { + "epoch": 0.2624977213739464, + "grad_norm": 0.5625, + "learning_rate": 0.0016110566484820589, + "loss": 0.1045, + "step": 30240 + }, + { + "epoch": 0.2625064018541506, + "grad_norm": 0.4765625, + "learning_rate": 0.0016110321495799813, + "loss": 0.1777, + "step": 30241 + }, + { + "epoch": 0.2625150823343547, + "grad_norm": 0.1875, + "learning_rate": 0.0016110076501190467, + "loss": 0.1387, + "step": 30242 + }, + { + "epoch": 0.2625237628145589, + "grad_norm": 0.11376953125, + "learning_rate": 0.0016109831500992818, + "loss": 0.0884, + "step": 30243 + }, + { + "epoch": 0.26253244329476305, + "grad_norm": 1.1171875, + "learning_rate": 0.0016109586495207133, + "loss": 0.1201, + "step": 30244 + }, + { + "epoch": 0.26254112377496724, + "grad_norm": 0.296875, + "learning_rate": 0.0016109341483833684, + "loss": 0.0977, + "step": 30245 + }, + { + "epoch": 0.2625498042551714, + "grad_norm": 0.37109375, + "learning_rate": 0.001610909646687273, + "loss": 0.1187, + "step": 30246 + }, + { + "epoch": 0.2625584847353756, + "grad_norm": 0.5703125, + "learning_rate": 0.0016108851444324552, + "loss": 0.1426, + "step": 30247 + }, + { + "epoch": 0.2625671652155797, + "grad_norm": 0.2470703125, + "learning_rate": 0.0016108606416189407, + "loss": 0.1108, + "step": 30248 + }, + { + "epoch": 0.2625758456957839, + "grad_norm": 0.5, + "learning_rate": 0.0016108361382467567, + "loss": 0.1104, + "step": 30249 + }, + { + "epoch": 0.26258452617598804, + "grad_norm": 0.15625, + "learning_rate": 0.00161081163431593, + "loss": 0.0869, + "step": 30250 + }, + { + "epoch": 0.26259320665619224, + "grad_norm": 0.466796875, + "learning_rate": 0.0016107871298264873, + "loss": 0.1035, + "step": 30251 + }, + { + "epoch": 0.2626018871363964, + "grad_norm": 0.1220703125, + "learning_rate": 0.0016107626247784557, + "loss": 0.1191, + "step": 30252 + }, + { + "epoch": 0.26261056761660057, + "grad_norm": 0.455078125, + "learning_rate": 0.0016107381191718616, + "loss": 0.1025, + "step": 30253 + }, + { + "epoch": 0.2626192480968047, + "grad_norm": 0.212890625, + "learning_rate": 0.001610713613006732, + "loss": 0.0737, + "step": 30254 + }, + { + "epoch": 0.2626279285770089, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016106891062830934, + "loss": 0.1396, + "step": 30255 + }, + { + "epoch": 0.26263660905721303, + "grad_norm": 0.177734375, + "learning_rate": 0.0016106645990009733, + "loss": 0.1006, + "step": 30256 + }, + { + "epoch": 0.2626452895374172, + "grad_norm": 0.2294921875, + "learning_rate": 0.001610640091160398, + "loss": 0.0986, + "step": 30257 + }, + { + "epoch": 0.26265397001762136, + "grad_norm": 0.482421875, + "learning_rate": 0.0016106155827613942, + "loss": 0.1309, + "step": 30258 + }, + { + "epoch": 0.26266265049782556, + "grad_norm": 0.09326171875, + "learning_rate": 0.001610591073803989, + "loss": 0.1289, + "step": 30259 + }, + { + "epoch": 0.2626713309780297, + "grad_norm": 0.54296875, + "learning_rate": 0.0016105665642882087, + "loss": 0.0869, + "step": 30260 + }, + { + "epoch": 0.2626800114582339, + "grad_norm": 0.10400390625, + "learning_rate": 0.0016105420542140809, + "loss": 0.083, + "step": 30261 + }, + { + "epoch": 0.262688691938438, + "grad_norm": 0.154296875, + "learning_rate": 0.0016105175435816319, + "loss": 0.0903, + "step": 30262 + }, + { + "epoch": 0.2626973724186422, + "grad_norm": 0.423828125, + "learning_rate": 0.0016104930323908883, + "loss": 0.1602, + "step": 30263 + }, + { + "epoch": 0.26270605289884635, + "grad_norm": 0.208984375, + "learning_rate": 0.0016104685206418776, + "loss": 0.0703, + "step": 30264 + }, + { + "epoch": 0.26271473337905055, + "grad_norm": 0.166015625, + "learning_rate": 0.0016104440083346261, + "loss": 0.1309, + "step": 30265 + }, + { + "epoch": 0.2627234138592547, + "grad_norm": 0.470703125, + "learning_rate": 0.0016104194954691606, + "loss": 0.1055, + "step": 30266 + }, + { + "epoch": 0.2627320943394589, + "grad_norm": 0.50390625, + "learning_rate": 0.001610394982045508, + "loss": 0.1152, + "step": 30267 + }, + { + "epoch": 0.262740774819663, + "grad_norm": 0.3671875, + "learning_rate": 0.0016103704680636953, + "loss": 0.1318, + "step": 30268 + }, + { + "epoch": 0.2627494552998672, + "grad_norm": 0.1298828125, + "learning_rate": 0.001610345953523749, + "loss": 0.0986, + "step": 30269 + }, + { + "epoch": 0.26275813578007134, + "grad_norm": 0.1669921875, + "learning_rate": 0.0016103214384256958, + "loss": 0.1191, + "step": 30270 + }, + { + "epoch": 0.26276681626027554, + "grad_norm": 0.19921875, + "learning_rate": 0.001610296922769563, + "loss": 0.1113, + "step": 30271 + }, + { + "epoch": 0.2627754967404797, + "grad_norm": 0.56640625, + "learning_rate": 0.0016102724065553769, + "loss": 0.1025, + "step": 30272 + }, + { + "epoch": 0.26278417722068387, + "grad_norm": 0.33203125, + "learning_rate": 0.0016102478897831649, + "loss": 0.1221, + "step": 30273 + }, + { + "epoch": 0.262792857700888, + "grad_norm": 0.259765625, + "learning_rate": 0.0016102233724529533, + "loss": 0.0762, + "step": 30274 + }, + { + "epoch": 0.2628015381810922, + "grad_norm": 0.08447265625, + "learning_rate": 0.0016101988545647692, + "loss": 0.0957, + "step": 30275 + }, + { + "epoch": 0.26281021866129634, + "grad_norm": 0.10009765625, + "learning_rate": 0.0016101743361186392, + "loss": 0.0879, + "step": 30276 + }, + { + "epoch": 0.26281889914150053, + "grad_norm": 0.291015625, + "learning_rate": 0.0016101498171145901, + "loss": 0.0986, + "step": 30277 + }, + { + "epoch": 0.26282757962170467, + "grad_norm": 0.138671875, + "learning_rate": 0.0016101252975526492, + "loss": 0.127, + "step": 30278 + }, + { + "epoch": 0.26283626010190886, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016101007774328427, + "loss": 0.0977, + "step": 30279 + }, + { + "epoch": 0.262844940582113, + "grad_norm": 0.66796875, + "learning_rate": 0.0016100762567551976, + "loss": 0.105, + "step": 30280 + }, + { + "epoch": 0.2628536210623172, + "grad_norm": 0.166015625, + "learning_rate": 0.0016100517355197405, + "loss": 0.103, + "step": 30281 + }, + { + "epoch": 0.2628623015425213, + "grad_norm": 0.25, + "learning_rate": 0.0016100272137264992, + "loss": 0.124, + "step": 30282 + }, + { + "epoch": 0.2628709820227255, + "grad_norm": 0.333984375, + "learning_rate": 0.0016100026913754993, + "loss": 0.127, + "step": 30283 + }, + { + "epoch": 0.26287966250292966, + "grad_norm": 0.1640625, + "learning_rate": 0.0016099781684667683, + "loss": 0.1079, + "step": 30284 + }, + { + "epoch": 0.26288834298313385, + "grad_norm": 0.2060546875, + "learning_rate": 0.0016099536450003328, + "loss": 0.1084, + "step": 30285 + }, + { + "epoch": 0.262897023463338, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016099291209762195, + "loss": 0.0996, + "step": 30286 + }, + { + "epoch": 0.2629057039435422, + "grad_norm": 1.1953125, + "learning_rate": 0.0016099045963944554, + "loss": 0.1787, + "step": 30287 + }, + { + "epoch": 0.2629143844237463, + "grad_norm": 0.0859375, + "learning_rate": 0.0016098800712550673, + "loss": 0.0654, + "step": 30288 + }, + { + "epoch": 0.2629230649039505, + "grad_norm": 0.1181640625, + "learning_rate": 0.001609855545558082, + "loss": 0.1289, + "step": 30289 + }, + { + "epoch": 0.26293174538415465, + "grad_norm": 0.265625, + "learning_rate": 0.0016098310193035266, + "loss": 0.1318, + "step": 30290 + }, + { + "epoch": 0.26294042586435884, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016098064924914275, + "loss": 0.0977, + "step": 30291 + }, + { + "epoch": 0.262949106344563, + "grad_norm": 0.431640625, + "learning_rate": 0.0016097819651218117, + "loss": 0.1211, + "step": 30292 + }, + { + "epoch": 0.26295778682476717, + "grad_norm": 0.177734375, + "learning_rate": 0.001609757437194706, + "loss": 0.0791, + "step": 30293 + }, + { + "epoch": 0.2629664673049713, + "grad_norm": 0.37109375, + "learning_rate": 0.001609732908710137, + "loss": 0.0977, + "step": 30294 + }, + { + "epoch": 0.2629751477851755, + "grad_norm": 0.4140625, + "learning_rate": 0.001609708379668132, + "loss": 0.126, + "step": 30295 + }, + { + "epoch": 0.26298382826537964, + "grad_norm": 0.20703125, + "learning_rate": 0.0016096838500687177, + "loss": 0.1006, + "step": 30296 + }, + { + "epoch": 0.26299250874558383, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016096593199119206, + "loss": 0.1016, + "step": 30297 + }, + { + "epoch": 0.26300118922578797, + "grad_norm": 0.0908203125, + "learning_rate": 0.0016096347891977679, + "loss": 0.0781, + "step": 30298 + }, + { + "epoch": 0.26300986970599216, + "grad_norm": 0.322265625, + "learning_rate": 0.0016096102579262858, + "loss": 0.1143, + "step": 30299 + }, + { + "epoch": 0.2630185501861963, + "grad_norm": 0.2314453125, + "learning_rate": 0.001609585726097502, + "loss": 0.0703, + "step": 30300 + }, + { + "epoch": 0.2630272306664005, + "grad_norm": 0.32421875, + "learning_rate": 0.001609561193711443, + "loss": 0.1123, + "step": 30301 + }, + { + "epoch": 0.26303591114660463, + "grad_norm": 0.1201171875, + "learning_rate": 0.0016095366607681353, + "loss": 0.1211, + "step": 30302 + }, + { + "epoch": 0.2630445916268088, + "grad_norm": 0.46484375, + "learning_rate": 0.001609512127267606, + "loss": 0.0977, + "step": 30303 + }, + { + "epoch": 0.26305327210701296, + "grad_norm": 0.216796875, + "learning_rate": 0.0016094875932098815, + "loss": 0.0986, + "step": 30304 + }, + { + "epoch": 0.26306195258721715, + "grad_norm": 0.4609375, + "learning_rate": 0.0016094630585949898, + "loss": 0.0942, + "step": 30305 + }, + { + "epoch": 0.2630706330674213, + "grad_norm": 0.36328125, + "learning_rate": 0.0016094385234229566, + "loss": 0.1328, + "step": 30306 + }, + { + "epoch": 0.2630793135476255, + "grad_norm": 0.39453125, + "learning_rate": 0.001609413987693809, + "loss": 0.1035, + "step": 30307 + }, + { + "epoch": 0.2630879940278296, + "grad_norm": 0.2177734375, + "learning_rate": 0.001609389451407574, + "loss": 0.0986, + "step": 30308 + }, + { + "epoch": 0.2630966745080338, + "grad_norm": 0.330078125, + "learning_rate": 0.001609364914564278, + "loss": 0.0879, + "step": 30309 + }, + { + "epoch": 0.26310535498823795, + "grad_norm": 0.17578125, + "learning_rate": 0.0016093403771639488, + "loss": 0.085, + "step": 30310 + }, + { + "epoch": 0.2631140354684421, + "grad_norm": 0.298828125, + "learning_rate": 0.001609315839206612, + "loss": 0.0908, + "step": 30311 + }, + { + "epoch": 0.2631227159486463, + "grad_norm": 0.365234375, + "learning_rate": 0.0016092913006922954, + "loss": 0.0698, + "step": 30312 + }, + { + "epoch": 0.2631313964288504, + "grad_norm": 0.546875, + "learning_rate": 0.0016092667616210256, + "loss": 0.165, + "step": 30313 + }, + { + "epoch": 0.2631400769090546, + "grad_norm": 0.19921875, + "learning_rate": 0.0016092422219928292, + "loss": 0.0649, + "step": 30314 + }, + { + "epoch": 0.26314875738925875, + "grad_norm": 0.203125, + "learning_rate": 0.001609217681807733, + "loss": 0.0918, + "step": 30315 + }, + { + "epoch": 0.26315743786946294, + "grad_norm": 0.208984375, + "learning_rate": 0.0016091931410657642, + "loss": 0.123, + "step": 30316 + }, + { + "epoch": 0.2631661183496671, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016091685997669492, + "loss": 0.0962, + "step": 30317 + }, + { + "epoch": 0.26317479882987127, + "grad_norm": 0.07666015625, + "learning_rate": 0.0016091440579113151, + "loss": 0.0918, + "step": 30318 + }, + { + "epoch": 0.2631834793100754, + "grad_norm": 0.365234375, + "learning_rate": 0.0016091195154988892, + "loss": 0.0986, + "step": 30319 + }, + { + "epoch": 0.2631921597902796, + "grad_norm": 0.4140625, + "learning_rate": 0.0016090949725296975, + "loss": 0.1084, + "step": 30320 + }, + { + "epoch": 0.26320084027048374, + "grad_norm": 0.455078125, + "learning_rate": 0.001609070429003767, + "loss": 0.1377, + "step": 30321 + }, + { + "epoch": 0.26320952075068793, + "grad_norm": 0.76171875, + "learning_rate": 0.001609045884921125, + "loss": 0.0781, + "step": 30322 + }, + { + "epoch": 0.26321820123089207, + "grad_norm": 0.62890625, + "learning_rate": 0.001609021340281798, + "loss": 0.1738, + "step": 30323 + }, + { + "epoch": 0.26322688171109626, + "grad_norm": 0.259765625, + "learning_rate": 0.0016089967950858126, + "loss": 0.1309, + "step": 30324 + }, + { + "epoch": 0.2632355621913004, + "grad_norm": 0.15234375, + "learning_rate": 0.0016089722493331964, + "loss": 0.1045, + "step": 30325 + }, + { + "epoch": 0.2632442426715046, + "grad_norm": 0.166015625, + "learning_rate": 0.0016089477030239757, + "loss": 0.1055, + "step": 30326 + }, + { + "epoch": 0.26325292315170873, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016089231561581773, + "loss": 0.1152, + "step": 30327 + }, + { + "epoch": 0.2632616036319129, + "grad_norm": 0.69140625, + "learning_rate": 0.001608898608735828, + "loss": 0.085, + "step": 30328 + }, + { + "epoch": 0.26327028411211706, + "grad_norm": 0.0927734375, + "learning_rate": 0.0016088740607569552, + "loss": 0.1582, + "step": 30329 + }, + { + "epoch": 0.26327896459232125, + "grad_norm": 0.134765625, + "learning_rate": 0.0016088495122215853, + "loss": 0.0938, + "step": 30330 + }, + { + "epoch": 0.2632876450725254, + "grad_norm": 0.17578125, + "learning_rate": 0.0016088249631297455, + "loss": 0.0938, + "step": 30331 + }, + { + "epoch": 0.2632963255527296, + "grad_norm": 0.08056640625, + "learning_rate": 0.0016088004134814614, + "loss": 0.0967, + "step": 30332 + }, + { + "epoch": 0.2633050060329337, + "grad_norm": 0.447265625, + "learning_rate": 0.0016087758632767617, + "loss": 0.1484, + "step": 30333 + }, + { + "epoch": 0.2633136865131379, + "grad_norm": 0.3515625, + "learning_rate": 0.001608751312515672, + "loss": 0.1475, + "step": 30334 + }, + { + "epoch": 0.26332236699334205, + "grad_norm": 0.240234375, + "learning_rate": 0.0016087267611982197, + "loss": 0.1123, + "step": 30335 + }, + { + "epoch": 0.26333104747354624, + "grad_norm": 0.2158203125, + "learning_rate": 0.0016087022093244315, + "loss": 0.0776, + "step": 30336 + }, + { + "epoch": 0.2633397279537504, + "grad_norm": 0.375, + "learning_rate": 0.0016086776568943338, + "loss": 0.0752, + "step": 30337 + }, + { + "epoch": 0.2633484084339546, + "grad_norm": 0.337890625, + "learning_rate": 0.0016086531039079541, + "loss": 0.1025, + "step": 30338 + }, + { + "epoch": 0.2633570889141587, + "grad_norm": 0.33203125, + "learning_rate": 0.0016086285503653193, + "loss": 0.1338, + "step": 30339 + }, + { + "epoch": 0.2633657693943629, + "grad_norm": 0.24609375, + "learning_rate": 0.0016086039962664555, + "loss": 0.1187, + "step": 30340 + }, + { + "epoch": 0.26337444987456704, + "grad_norm": 0.12353515625, + "learning_rate": 0.0016085794416113904, + "loss": 0.1055, + "step": 30341 + }, + { + "epoch": 0.26338313035477123, + "grad_norm": 0.0927734375, + "learning_rate": 0.00160855488640015, + "loss": 0.0879, + "step": 30342 + }, + { + "epoch": 0.26339181083497537, + "grad_norm": 0.2890625, + "learning_rate": 0.0016085303306327622, + "loss": 0.106, + "step": 30343 + }, + { + "epoch": 0.26340049131517956, + "grad_norm": 0.82421875, + "learning_rate": 0.0016085057743092529, + "loss": 0.1152, + "step": 30344 + }, + { + "epoch": 0.2634091717953837, + "grad_norm": 0.1796875, + "learning_rate": 0.0016084812174296493, + "loss": 0.1152, + "step": 30345 + }, + { + "epoch": 0.2634178522755879, + "grad_norm": 0.3046875, + "learning_rate": 0.0016084566599939784, + "loss": 0.1167, + "step": 30346 + }, + { + "epoch": 0.26342653275579203, + "grad_norm": 0.828125, + "learning_rate": 0.0016084321020022668, + "loss": 0.1318, + "step": 30347 + }, + { + "epoch": 0.2634352132359962, + "grad_norm": 0.5390625, + "learning_rate": 0.0016084075434545416, + "loss": 0.1543, + "step": 30348 + }, + { + "epoch": 0.26344389371620036, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016083829843508298, + "loss": 0.1562, + "step": 30349 + }, + { + "epoch": 0.26345257419640455, + "grad_norm": 0.17578125, + "learning_rate": 0.0016083584246911575, + "loss": 0.0835, + "step": 30350 + }, + { + "epoch": 0.2634612546766087, + "grad_norm": 0.07958984375, + "learning_rate": 0.0016083338644755523, + "loss": 0.0889, + "step": 30351 + }, + { + "epoch": 0.2634699351568129, + "grad_norm": 0.421875, + "learning_rate": 0.001608309303704041, + "loss": 0.0933, + "step": 30352 + }, + { + "epoch": 0.263478615637017, + "grad_norm": 0.12158203125, + "learning_rate": 0.00160828474237665, + "loss": 0.0947, + "step": 30353 + }, + { + "epoch": 0.2634872961172212, + "grad_norm": 0.78125, + "learning_rate": 0.0016082601804934067, + "loss": 0.1118, + "step": 30354 + }, + { + "epoch": 0.26349597659742535, + "grad_norm": 0.40625, + "learning_rate": 0.0016082356180543375, + "loss": 0.1191, + "step": 30355 + }, + { + "epoch": 0.26350465707762954, + "grad_norm": 0.1845703125, + "learning_rate": 0.0016082110550594696, + "loss": 0.1089, + "step": 30356 + }, + { + "epoch": 0.2635133375578337, + "grad_norm": 0.482421875, + "learning_rate": 0.0016081864915088295, + "loss": 0.0869, + "step": 30357 + }, + { + "epoch": 0.2635220180380379, + "grad_norm": 0.2138671875, + "learning_rate": 0.0016081619274024442, + "loss": 0.123, + "step": 30358 + }, + { + "epoch": 0.263530698518242, + "grad_norm": 0.11376953125, + "learning_rate": 0.001608137362740341, + "loss": 0.1348, + "step": 30359 + }, + { + "epoch": 0.2635393789984462, + "grad_norm": 0.138671875, + "learning_rate": 0.0016081127975225464, + "loss": 0.1133, + "step": 30360 + }, + { + "epoch": 0.26354805947865034, + "grad_norm": 0.11474609375, + "learning_rate": 0.0016080882317490873, + "loss": 0.1133, + "step": 30361 + }, + { + "epoch": 0.26355673995885454, + "grad_norm": 0.11279296875, + "learning_rate": 0.0016080636654199904, + "loss": 0.0613, + "step": 30362 + }, + { + "epoch": 0.2635654204390587, + "grad_norm": 0.126953125, + "learning_rate": 0.0016080390985352825, + "loss": 0.2109, + "step": 30363 + }, + { + "epoch": 0.26357410091926287, + "grad_norm": 0.0966796875, + "learning_rate": 0.0016080145310949912, + "loss": 0.082, + "step": 30364 + }, + { + "epoch": 0.263582781399467, + "grad_norm": 0.294921875, + "learning_rate": 0.0016079899630991423, + "loss": 0.0674, + "step": 30365 + }, + { + "epoch": 0.2635914618796712, + "grad_norm": 0.09423828125, + "learning_rate": 0.0016079653945477636, + "loss": 0.0635, + "step": 30366 + }, + { + "epoch": 0.26360014235987533, + "grad_norm": 0.16796875, + "learning_rate": 0.0016079408254408814, + "loss": 0.1357, + "step": 30367 + }, + { + "epoch": 0.2636088228400795, + "grad_norm": 0.1162109375, + "learning_rate": 0.001607916255778523, + "loss": 0.1191, + "step": 30368 + }, + { + "epoch": 0.26361750332028366, + "grad_norm": 0.1640625, + "learning_rate": 0.0016078916855607147, + "loss": 0.0938, + "step": 30369 + }, + { + "epoch": 0.26362618380048786, + "grad_norm": 0.09521484375, + "learning_rate": 0.0016078671147874838, + "loss": 0.0879, + "step": 30370 + }, + { + "epoch": 0.263634864280692, + "grad_norm": 1.1328125, + "learning_rate": 0.0016078425434588569, + "loss": 0.1367, + "step": 30371 + }, + { + "epoch": 0.2636435447608962, + "grad_norm": 0.341796875, + "learning_rate": 0.0016078179715748612, + "loss": 0.0732, + "step": 30372 + }, + { + "epoch": 0.2636522252411003, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016077933991355234, + "loss": 0.0674, + "step": 30373 + }, + { + "epoch": 0.2636609057213045, + "grad_norm": 0.77734375, + "learning_rate": 0.0016077688261408702, + "loss": 0.1309, + "step": 30374 + }, + { + "epoch": 0.26366958620150865, + "grad_norm": 0.419921875, + "learning_rate": 0.001607744252590929, + "loss": 0.104, + "step": 30375 + }, + { + "epoch": 0.26367826668171285, + "grad_norm": 0.494140625, + "learning_rate": 0.001607719678485726, + "loss": 0.1084, + "step": 30376 + }, + { + "epoch": 0.263686947161917, + "grad_norm": 0.55859375, + "learning_rate": 0.0016076951038252886, + "loss": 0.1289, + "step": 30377 + }, + { + "epoch": 0.2636956276421212, + "grad_norm": 0.453125, + "learning_rate": 0.001607670528609643, + "loss": 0.1416, + "step": 30378 + }, + { + "epoch": 0.2637043081223253, + "grad_norm": 0.578125, + "learning_rate": 0.001607645952838817, + "loss": 0.0986, + "step": 30379 + }, + { + "epoch": 0.2637129886025295, + "grad_norm": 0.33203125, + "learning_rate": 0.0016076213765128366, + "loss": 0.1123, + "step": 30380 + }, + { + "epoch": 0.26372166908273365, + "grad_norm": 0.31640625, + "learning_rate": 0.0016075967996317297, + "loss": 0.0781, + "step": 30381 + }, + { + "epoch": 0.26373034956293784, + "grad_norm": 0.09228515625, + "learning_rate": 0.0016075722221955222, + "loss": 0.0845, + "step": 30382 + }, + { + "epoch": 0.263739030043142, + "grad_norm": 0.5234375, + "learning_rate": 0.0016075476442042416, + "loss": 0.0835, + "step": 30383 + }, + { + "epoch": 0.26374771052334617, + "grad_norm": 0.08447265625, + "learning_rate": 0.0016075230656579142, + "loss": 0.0859, + "step": 30384 + }, + { + "epoch": 0.2637563910035503, + "grad_norm": 0.5390625, + "learning_rate": 0.0016074984865565671, + "loss": 0.0933, + "step": 30385 + }, + { + "epoch": 0.2637650714837545, + "grad_norm": 0.0673828125, + "learning_rate": 0.0016074739069002278, + "loss": 0.0938, + "step": 30386 + }, + { + "epoch": 0.26377375196395864, + "grad_norm": 0.1015625, + "learning_rate": 0.0016074493266889224, + "loss": 0.1279, + "step": 30387 + }, + { + "epoch": 0.26378243244416283, + "grad_norm": 0.5234375, + "learning_rate": 0.0016074247459226779, + "loss": 0.0977, + "step": 30388 + }, + { + "epoch": 0.26379111292436697, + "grad_norm": 0.10107421875, + "learning_rate": 0.0016074001646015214, + "loss": 0.1025, + "step": 30389 + }, + { + "epoch": 0.26379979340457116, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016073755827254797, + "loss": 0.1094, + "step": 30390 + }, + { + "epoch": 0.2638084738847753, + "grad_norm": 0.1123046875, + "learning_rate": 0.0016073510002945796, + "loss": 0.1152, + "step": 30391 + }, + { + "epoch": 0.2638171543649795, + "grad_norm": 0.212890625, + "learning_rate": 0.0016073264173088481, + "loss": 0.1504, + "step": 30392 + }, + { + "epoch": 0.2638258348451836, + "grad_norm": 5.78125, + "learning_rate": 0.001607301833768312, + "loss": 0.4023, + "step": 30393 + }, + { + "epoch": 0.2638345153253878, + "grad_norm": 0.1064453125, + "learning_rate": 0.0016072772496729987, + "loss": 0.1143, + "step": 30394 + }, + { + "epoch": 0.26384319580559196, + "grad_norm": 0.51953125, + "learning_rate": 0.0016072526650229343, + "loss": 0.1113, + "step": 30395 + }, + { + "epoch": 0.26385187628579615, + "grad_norm": 0.431640625, + "learning_rate": 0.001607228079818146, + "loss": 0.1172, + "step": 30396 + }, + { + "epoch": 0.2638605567660003, + "grad_norm": 0.361328125, + "learning_rate": 0.0016072034940586605, + "loss": 0.1318, + "step": 30397 + }, + { + "epoch": 0.2638692372462045, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016071789077445051, + "loss": 0.105, + "step": 30398 + }, + { + "epoch": 0.2638779177264086, + "grad_norm": 0.48046875, + "learning_rate": 0.0016071543208757064, + "loss": 0.1016, + "step": 30399 + }, + { + "epoch": 0.2638865982066128, + "grad_norm": 0.44921875, + "learning_rate": 0.0016071297334522914, + "loss": 0.1011, + "step": 30400 + }, + { + "epoch": 0.26389527868681695, + "grad_norm": 0.30078125, + "learning_rate": 0.001607105145474287, + "loss": 0.0962, + "step": 30401 + }, + { + "epoch": 0.26390395916702114, + "grad_norm": 0.546875, + "learning_rate": 0.0016070805569417198, + "loss": 0.1177, + "step": 30402 + }, + { + "epoch": 0.2639126396472253, + "grad_norm": 0.443359375, + "learning_rate": 0.001607055967854617, + "loss": 0.1299, + "step": 30403 + }, + { + "epoch": 0.26392132012742947, + "grad_norm": 0.3359375, + "learning_rate": 0.0016070313782130057, + "loss": 0.1045, + "step": 30404 + }, + { + "epoch": 0.2639300006076336, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016070067880169124, + "loss": 0.1152, + "step": 30405 + }, + { + "epoch": 0.2639386810878378, + "grad_norm": 0.18359375, + "learning_rate": 0.0016069821972663637, + "loss": 0.1201, + "step": 30406 + }, + { + "epoch": 0.26394736156804194, + "grad_norm": 0.404296875, + "learning_rate": 0.0016069576059613872, + "loss": 0.1143, + "step": 30407 + }, + { + "epoch": 0.26395604204824613, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016069330141020098, + "loss": 0.0801, + "step": 30408 + }, + { + "epoch": 0.26396472252845027, + "grad_norm": 0.162109375, + "learning_rate": 0.0016069084216882574, + "loss": 0.127, + "step": 30409 + }, + { + "epoch": 0.26397340300865446, + "grad_norm": 0.2392578125, + "learning_rate": 0.001606883828720158, + "loss": 0.1035, + "step": 30410 + }, + { + "epoch": 0.2639820834888586, + "grad_norm": 0.51171875, + "learning_rate": 0.0016068592351977378, + "loss": 0.1064, + "step": 30411 + }, + { + "epoch": 0.2639907639690628, + "grad_norm": 0.1279296875, + "learning_rate": 0.0016068346411210245, + "loss": 0.1001, + "step": 30412 + }, + { + "epoch": 0.26399944444926693, + "grad_norm": 0.09912109375, + "learning_rate": 0.001606810046490044, + "loss": 0.0957, + "step": 30413 + }, + { + "epoch": 0.2640081249294711, + "grad_norm": 0.251953125, + "learning_rate": 0.0016067854513048235, + "loss": 0.1064, + "step": 30414 + }, + { + "epoch": 0.26401680540967526, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016067608555653903, + "loss": 0.1191, + "step": 30415 + }, + { + "epoch": 0.26402548588987945, + "grad_norm": 0.0947265625, + "learning_rate": 0.0016067362592717708, + "loss": 0.1099, + "step": 30416 + }, + { + "epoch": 0.2640341663700836, + "grad_norm": 0.2421875, + "learning_rate": 0.0016067116624239926, + "loss": 0.1543, + "step": 30417 + }, + { + "epoch": 0.2640428468502878, + "grad_norm": 0.73046875, + "learning_rate": 0.0016066870650220818, + "loss": 0.1553, + "step": 30418 + }, + { + "epoch": 0.2640515273304919, + "grad_norm": 0.58203125, + "learning_rate": 0.0016066624670660657, + "loss": 0.1279, + "step": 30419 + }, + { + "epoch": 0.2640602078106961, + "grad_norm": 0.46484375, + "learning_rate": 0.001606637868555971, + "loss": 0.0869, + "step": 30420 + }, + { + "epoch": 0.26406888829090025, + "grad_norm": 0.400390625, + "learning_rate": 0.001606613269491825, + "loss": 0.1396, + "step": 30421 + }, + { + "epoch": 0.26407756877110444, + "grad_norm": 0.349609375, + "learning_rate": 0.0016065886698736545, + "loss": 0.1157, + "step": 30422 + }, + { + "epoch": 0.2640862492513086, + "grad_norm": 0.162109375, + "learning_rate": 0.001606564069701486, + "loss": 0.0859, + "step": 30423 + }, + { + "epoch": 0.2640949297315128, + "grad_norm": 0.11572265625, + "learning_rate": 0.0016065394689753465, + "loss": 0.1025, + "step": 30424 + }, + { + "epoch": 0.2641036102117169, + "grad_norm": 0.5703125, + "learning_rate": 0.001606514867695263, + "loss": 0.1201, + "step": 30425 + }, + { + "epoch": 0.2641122906919211, + "grad_norm": 0.3359375, + "learning_rate": 0.0016064902658612626, + "loss": 0.1162, + "step": 30426 + }, + { + "epoch": 0.26412097117212524, + "grad_norm": 0.361328125, + "learning_rate": 0.001606465663473372, + "loss": 0.1035, + "step": 30427 + }, + { + "epoch": 0.26412965165232943, + "grad_norm": 0.12451171875, + "learning_rate": 0.0016064410605316182, + "loss": 0.125, + "step": 30428 + }, + { + "epoch": 0.26413833213253357, + "grad_norm": 0.140625, + "learning_rate": 0.0016064164570360282, + "loss": 0.1055, + "step": 30429 + }, + { + "epoch": 0.26414701261273776, + "grad_norm": 0.2255859375, + "learning_rate": 0.0016063918529866286, + "loss": 0.1689, + "step": 30430 + }, + { + "epoch": 0.2641556930929419, + "grad_norm": 0.15625, + "learning_rate": 0.0016063672483834467, + "loss": 0.1504, + "step": 30431 + }, + { + "epoch": 0.2641643735731461, + "grad_norm": 0.1875, + "learning_rate": 0.0016063426432265091, + "loss": 0.1055, + "step": 30432 + }, + { + "epoch": 0.26417305405335023, + "grad_norm": 0.224609375, + "learning_rate": 0.0016063180375158427, + "loss": 0.1143, + "step": 30433 + }, + { + "epoch": 0.26418173453355437, + "grad_norm": 0.138671875, + "learning_rate": 0.0016062934312514746, + "loss": 0.1201, + "step": 30434 + }, + { + "epoch": 0.26419041501375856, + "grad_norm": 0.189453125, + "learning_rate": 0.0016062688244334314, + "loss": 0.1406, + "step": 30435 + }, + { + "epoch": 0.2641990954939627, + "grad_norm": 0.298828125, + "learning_rate": 0.0016062442170617405, + "loss": 0.1177, + "step": 30436 + }, + { + "epoch": 0.2642077759741669, + "grad_norm": 0.53515625, + "learning_rate": 0.0016062196091364282, + "loss": 0.084, + "step": 30437 + }, + { + "epoch": 0.26421645645437103, + "grad_norm": 0.3203125, + "learning_rate": 0.0016061950006575221, + "loss": 0.168, + "step": 30438 + }, + { + "epoch": 0.2642251369345752, + "grad_norm": 0.251953125, + "learning_rate": 0.0016061703916250484, + "loss": 0.083, + "step": 30439 + }, + { + "epoch": 0.26423381741477936, + "grad_norm": 0.59765625, + "learning_rate": 0.0016061457820390349, + "loss": 0.0874, + "step": 30440 + }, + { + "epoch": 0.26424249789498355, + "grad_norm": 0.0986328125, + "learning_rate": 0.0016061211718995075, + "loss": 0.0928, + "step": 30441 + }, + { + "epoch": 0.2642511783751877, + "grad_norm": 0.4609375, + "learning_rate": 0.0016060965612064936, + "loss": 0.126, + "step": 30442 + }, + { + "epoch": 0.2642598588553919, + "grad_norm": 0.34765625, + "learning_rate": 0.0016060719499600203, + "loss": 0.1523, + "step": 30443 + }, + { + "epoch": 0.264268539335596, + "grad_norm": 0.224609375, + "learning_rate": 0.0016060473381601143, + "loss": 0.1074, + "step": 30444 + }, + { + "epoch": 0.2642772198158002, + "grad_norm": 0.48828125, + "learning_rate": 0.0016060227258068024, + "loss": 0.0928, + "step": 30445 + }, + { + "epoch": 0.26428590029600435, + "grad_norm": 0.59765625, + "learning_rate": 0.0016059981129001115, + "loss": 0.1104, + "step": 30446 + }, + { + "epoch": 0.26429458077620854, + "grad_norm": 0.1875, + "learning_rate": 0.0016059734994400693, + "loss": 0.0645, + "step": 30447 + }, + { + "epoch": 0.2643032612564127, + "grad_norm": 0.62109375, + "learning_rate": 0.0016059488854267014, + "loss": 0.0898, + "step": 30448 + }, + { + "epoch": 0.2643119417366169, + "grad_norm": 0.232421875, + "learning_rate": 0.0016059242708600355, + "loss": 0.104, + "step": 30449 + }, + { + "epoch": 0.264320622216821, + "grad_norm": 0.1630859375, + "learning_rate": 0.0016058996557400988, + "loss": 0.0962, + "step": 30450 + }, + { + "epoch": 0.2643293026970252, + "grad_norm": 0.6171875, + "learning_rate": 0.0016058750400669175, + "loss": 0.2891, + "step": 30451 + }, + { + "epoch": 0.26433798317722934, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016058504238405191, + "loss": 0.0854, + "step": 30452 + }, + { + "epoch": 0.26434666365743353, + "grad_norm": 0.107421875, + "learning_rate": 0.0016058258070609302, + "loss": 0.0918, + "step": 30453 + }, + { + "epoch": 0.26435534413763767, + "grad_norm": 0.09130859375, + "learning_rate": 0.0016058011897281776, + "loss": 0.1006, + "step": 30454 + }, + { + "epoch": 0.26436402461784186, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016057765718422887, + "loss": 0.1592, + "step": 30455 + }, + { + "epoch": 0.264372705098046, + "grad_norm": 0.283203125, + "learning_rate": 0.0016057519534032898, + "loss": 0.1094, + "step": 30456 + }, + { + "epoch": 0.2643813855782502, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016057273344112085, + "loss": 0.1123, + "step": 30457 + }, + { + "epoch": 0.26439006605845433, + "grad_norm": 0.330078125, + "learning_rate": 0.0016057027148660712, + "loss": 0.1699, + "step": 30458 + }, + { + "epoch": 0.2643987465386585, + "grad_norm": 0.06787109375, + "learning_rate": 0.0016056780947679053, + "loss": 0.1016, + "step": 30459 + }, + { + "epoch": 0.26440742701886266, + "grad_norm": 0.15234375, + "learning_rate": 0.0016056534741167373, + "loss": 0.084, + "step": 30460 + }, + { + "epoch": 0.26441610749906685, + "grad_norm": 0.1494140625, + "learning_rate": 0.0016056288529125943, + "loss": 0.1216, + "step": 30461 + }, + { + "epoch": 0.264424787979271, + "grad_norm": 0.09228515625, + "learning_rate": 0.0016056042311555028, + "loss": 0.1113, + "step": 30462 + }, + { + "epoch": 0.2644334684594752, + "grad_norm": 0.443359375, + "learning_rate": 0.0016055796088454903, + "loss": 0.1216, + "step": 30463 + }, + { + "epoch": 0.2644421489396793, + "grad_norm": 0.271484375, + "learning_rate": 0.0016055549859825835, + "loss": 0.0688, + "step": 30464 + }, + { + "epoch": 0.2644508294198835, + "grad_norm": 0.080078125, + "learning_rate": 0.0016055303625668095, + "loss": 0.1211, + "step": 30465 + }, + { + "epoch": 0.26445950990008765, + "grad_norm": 0.244140625, + "learning_rate": 0.001605505738598195, + "loss": 0.0996, + "step": 30466 + }, + { + "epoch": 0.26446819038029185, + "grad_norm": 0.263671875, + "learning_rate": 0.001605481114076767, + "loss": 0.1279, + "step": 30467 + }, + { + "epoch": 0.264476870860496, + "grad_norm": 0.1767578125, + "learning_rate": 0.001605456489002553, + "loss": 0.1123, + "step": 30468 + }, + { + "epoch": 0.2644855513407002, + "grad_norm": 0.2578125, + "learning_rate": 0.0016054318633755788, + "loss": 0.1592, + "step": 30469 + }, + { + "epoch": 0.2644942318209043, + "grad_norm": 0.1474609375, + "learning_rate": 0.001605407237195872, + "loss": 0.1074, + "step": 30470 + }, + { + "epoch": 0.2645029123011085, + "grad_norm": 0.1162109375, + "learning_rate": 0.0016053826104634593, + "loss": 0.1123, + "step": 30471 + }, + { + "epoch": 0.26451159278131264, + "grad_norm": 0.5703125, + "learning_rate": 0.001605357983178368, + "loss": 0.1221, + "step": 30472 + }, + { + "epoch": 0.26452027326151684, + "grad_norm": 0.109375, + "learning_rate": 0.0016053333553406246, + "loss": 0.0928, + "step": 30473 + }, + { + "epoch": 0.264528953741721, + "grad_norm": 0.408203125, + "learning_rate": 0.0016053087269502567, + "loss": 0.0854, + "step": 30474 + }, + { + "epoch": 0.26453763422192517, + "grad_norm": 0.345703125, + "learning_rate": 0.00160528409800729, + "loss": 0.0986, + "step": 30475 + }, + { + "epoch": 0.2645463147021293, + "grad_norm": 0.431640625, + "learning_rate": 0.0016052594685117528, + "loss": 0.0898, + "step": 30476 + }, + { + "epoch": 0.2645549951823335, + "grad_norm": 0.4921875, + "learning_rate": 0.0016052348384636712, + "loss": 0.1084, + "step": 30477 + }, + { + "epoch": 0.26456367566253763, + "grad_norm": 0.18359375, + "learning_rate": 0.0016052102078630726, + "loss": 0.1367, + "step": 30478 + }, + { + "epoch": 0.2645723561427418, + "grad_norm": 0.248046875, + "learning_rate": 0.0016051855767099832, + "loss": 0.1104, + "step": 30479 + }, + { + "epoch": 0.26458103662294596, + "grad_norm": 0.11767578125, + "learning_rate": 0.0016051609450044308, + "loss": 0.0703, + "step": 30480 + }, + { + "epoch": 0.26458971710315016, + "grad_norm": 0.44140625, + "learning_rate": 0.0016051363127464423, + "loss": 0.1289, + "step": 30481 + }, + { + "epoch": 0.2645983975833543, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016051116799360439, + "loss": 0.1099, + "step": 30482 + }, + { + "epoch": 0.2646070780635585, + "grad_norm": 0.125, + "learning_rate": 0.001605087046573263, + "loss": 0.1152, + "step": 30483 + }, + { + "epoch": 0.2646157585437626, + "grad_norm": 0.435546875, + "learning_rate": 0.0016050624126581266, + "loss": 0.1309, + "step": 30484 + }, + { + "epoch": 0.2646244390239668, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016050377781906611, + "loss": 0.1123, + "step": 30485 + }, + { + "epoch": 0.26463311950417095, + "grad_norm": 0.5234375, + "learning_rate": 0.0016050131431708948, + "loss": 0.1709, + "step": 30486 + }, + { + "epoch": 0.26464179998437515, + "grad_norm": 1.296875, + "learning_rate": 0.0016049885075988528, + "loss": 0.0967, + "step": 30487 + }, + { + "epoch": 0.2646504804645793, + "grad_norm": 0.34375, + "learning_rate": 0.0016049638714745634, + "loss": 0.1055, + "step": 30488 + }, + { + "epoch": 0.2646591609447835, + "grad_norm": 0.58203125, + "learning_rate": 0.0016049392347980529, + "loss": 0.0889, + "step": 30489 + }, + { + "epoch": 0.2646678414249876, + "grad_norm": 0.1796875, + "learning_rate": 0.0016049145975693486, + "loss": 0.1162, + "step": 30490 + }, + { + "epoch": 0.2646765219051918, + "grad_norm": 0.146484375, + "learning_rate": 0.0016048899597884774, + "loss": 0.0908, + "step": 30491 + }, + { + "epoch": 0.26468520238539595, + "grad_norm": 0.482421875, + "learning_rate": 0.0016048653214554658, + "loss": 0.0752, + "step": 30492 + }, + { + "epoch": 0.26469388286560014, + "grad_norm": 0.095703125, + "learning_rate": 0.0016048406825703411, + "loss": 0.0757, + "step": 30493 + }, + { + "epoch": 0.2647025633458043, + "grad_norm": 0.625, + "learning_rate": 0.0016048160431331305, + "loss": 0.123, + "step": 30494 + }, + { + "epoch": 0.26471124382600847, + "grad_norm": 0.26953125, + "learning_rate": 0.0016047914031438607, + "loss": 0.1289, + "step": 30495 + }, + { + "epoch": 0.2647199243062126, + "grad_norm": 0.59375, + "learning_rate": 0.0016047667626025582, + "loss": 0.1094, + "step": 30496 + }, + { + "epoch": 0.2647286047864168, + "grad_norm": 0.06396484375, + "learning_rate": 0.0016047421215092503, + "loss": 0.0908, + "step": 30497 + }, + { + "epoch": 0.26473728526662094, + "grad_norm": 0.1943359375, + "learning_rate": 0.0016047174798639646, + "loss": 0.1045, + "step": 30498 + }, + { + "epoch": 0.26474596574682513, + "grad_norm": 0.828125, + "learning_rate": 0.001604692837666727, + "loss": 0.0625, + "step": 30499 + }, + { + "epoch": 0.26475464622702927, + "grad_norm": 0.482421875, + "learning_rate": 0.0016046681949175652, + "loss": 0.1094, + "step": 30500 + }, + { + "epoch": 0.26476332670723346, + "grad_norm": 0.34765625, + "learning_rate": 0.0016046435516165057, + "loss": 0.0781, + "step": 30501 + }, + { + "epoch": 0.2647720071874376, + "grad_norm": 0.236328125, + "learning_rate": 0.0016046189077635754, + "loss": 0.127, + "step": 30502 + }, + { + "epoch": 0.2647806876676418, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016045942633588015, + "loss": 0.127, + "step": 30503 + }, + { + "epoch": 0.2647893681478459, + "grad_norm": 0.1806640625, + "learning_rate": 0.0016045696184022109, + "loss": 0.1523, + "step": 30504 + }, + { + "epoch": 0.2647980486280501, + "grad_norm": 0.59375, + "learning_rate": 0.0016045449728938309, + "loss": 0.0986, + "step": 30505 + }, + { + "epoch": 0.26480672910825426, + "grad_norm": 0.376953125, + "learning_rate": 0.0016045203268336876, + "loss": 0.1143, + "step": 30506 + }, + { + "epoch": 0.26481540958845845, + "grad_norm": 0.8046875, + "learning_rate": 0.0016044956802218086, + "loss": 0.1025, + "step": 30507 + }, + { + "epoch": 0.2648240900686626, + "grad_norm": 0.359375, + "learning_rate": 0.0016044710330582208, + "loss": 0.0928, + "step": 30508 + }, + { + "epoch": 0.2648327705488668, + "grad_norm": 0.48828125, + "learning_rate": 0.0016044463853429514, + "loss": 0.1211, + "step": 30509 + }, + { + "epoch": 0.2648414510290709, + "grad_norm": 0.283203125, + "learning_rate": 0.0016044217370760263, + "loss": 0.1348, + "step": 30510 + }, + { + "epoch": 0.2648501315092751, + "grad_norm": 0.41796875, + "learning_rate": 0.0016043970882574736, + "loss": 0.0854, + "step": 30511 + }, + { + "epoch": 0.26485881198947925, + "grad_norm": 0.1572265625, + "learning_rate": 0.0016043724388873197, + "loss": 0.0845, + "step": 30512 + }, + { + "epoch": 0.26486749246968344, + "grad_norm": 0.1884765625, + "learning_rate": 0.0016043477889655918, + "loss": 0.0952, + "step": 30513 + }, + { + "epoch": 0.2648761729498876, + "grad_norm": 0.486328125, + "learning_rate": 0.0016043231384923167, + "loss": 0.104, + "step": 30514 + }, + { + "epoch": 0.26488485343009177, + "grad_norm": 0.2041015625, + "learning_rate": 0.0016042984874675213, + "loss": 0.106, + "step": 30515 + }, + { + "epoch": 0.2648935339102959, + "grad_norm": 0.1298828125, + "learning_rate": 0.0016042738358912328, + "loss": 0.1348, + "step": 30516 + }, + { + "epoch": 0.2649022143905001, + "grad_norm": 0.2333984375, + "learning_rate": 0.0016042491837634778, + "loss": 0.0811, + "step": 30517 + }, + { + "epoch": 0.26491089487070424, + "grad_norm": 0.1767578125, + "learning_rate": 0.0016042245310842838, + "loss": 0.1147, + "step": 30518 + }, + { + "epoch": 0.26491957535090843, + "grad_norm": 0.6015625, + "learning_rate": 0.0016041998778536768, + "loss": 0.0645, + "step": 30519 + }, + { + "epoch": 0.26492825583111257, + "grad_norm": 0.11962890625, + "learning_rate": 0.0016041752240716848, + "loss": 0.124, + "step": 30520 + }, + { + "epoch": 0.26493693631131676, + "grad_norm": 0.12060546875, + "learning_rate": 0.0016041505697383346, + "loss": 0.1016, + "step": 30521 + }, + { + "epoch": 0.2649456167915209, + "grad_norm": 0.193359375, + "learning_rate": 0.0016041259148536526, + "loss": 0.1426, + "step": 30522 + }, + { + "epoch": 0.2649542972717251, + "grad_norm": 0.11328125, + "learning_rate": 0.001604101259417666, + "loss": 0.0732, + "step": 30523 + }, + { + "epoch": 0.26496297775192923, + "grad_norm": 0.240234375, + "learning_rate": 0.0016040766034304023, + "loss": 0.1797, + "step": 30524 + }, + { + "epoch": 0.2649716582321334, + "grad_norm": 0.37890625, + "learning_rate": 0.0016040519468918876, + "loss": 0.0791, + "step": 30525 + }, + { + "epoch": 0.26498033871233756, + "grad_norm": 0.703125, + "learning_rate": 0.0016040272898021493, + "loss": 0.1011, + "step": 30526 + }, + { + "epoch": 0.26498901919254175, + "grad_norm": 0.337890625, + "learning_rate": 0.0016040026321612142, + "loss": 0.1582, + "step": 30527 + }, + { + "epoch": 0.2649976996727459, + "grad_norm": 0.267578125, + "learning_rate": 0.0016039779739691097, + "loss": 0.0835, + "step": 30528 + }, + { + "epoch": 0.2650063801529501, + "grad_norm": 0.2275390625, + "learning_rate": 0.0016039533152258623, + "loss": 0.1074, + "step": 30529 + }, + { + "epoch": 0.2650150606331542, + "grad_norm": 0.310546875, + "learning_rate": 0.0016039286559314991, + "loss": 0.1074, + "step": 30530 + }, + { + "epoch": 0.2650237411133584, + "grad_norm": 0.8359375, + "learning_rate": 0.001603903996086047, + "loss": 0.1084, + "step": 30531 + }, + { + "epoch": 0.26503242159356255, + "grad_norm": 0.69921875, + "learning_rate": 0.0016038793356895334, + "loss": 0.1387, + "step": 30532 + }, + { + "epoch": 0.26504110207376674, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016038546747419846, + "loss": 0.1367, + "step": 30533 + }, + { + "epoch": 0.2650497825539709, + "grad_norm": 0.337890625, + "learning_rate": 0.001603830013243428, + "loss": 0.1025, + "step": 30534 + }, + { + "epoch": 0.2650584630341751, + "grad_norm": 0.2734375, + "learning_rate": 0.0016038053511938908, + "loss": 0.0654, + "step": 30535 + }, + { + "epoch": 0.2650671435143792, + "grad_norm": 0.212890625, + "learning_rate": 0.0016037806885933993, + "loss": 0.0977, + "step": 30536 + }, + { + "epoch": 0.2650758239945834, + "grad_norm": 0.2333984375, + "learning_rate": 0.0016037560254419806, + "loss": 0.1182, + "step": 30537 + }, + { + "epoch": 0.26508450447478754, + "grad_norm": 0.609375, + "learning_rate": 0.001603731361739662, + "loss": 0.0938, + "step": 30538 + }, + { + "epoch": 0.26509318495499173, + "grad_norm": 0.12890625, + "learning_rate": 0.0016037066974864706, + "loss": 0.1143, + "step": 30539 + }, + { + "epoch": 0.26510186543519587, + "grad_norm": 0.46484375, + "learning_rate": 0.0016036820326824329, + "loss": 0.0947, + "step": 30540 + }, + { + "epoch": 0.26511054591540006, + "grad_norm": 0.48828125, + "learning_rate": 0.0016036573673275762, + "loss": 0.1104, + "step": 30541 + }, + { + "epoch": 0.2651192263956042, + "grad_norm": 0.357421875, + "learning_rate": 0.0016036327014219275, + "loss": 0.0889, + "step": 30542 + }, + { + "epoch": 0.2651279068758084, + "grad_norm": 0.26171875, + "learning_rate": 0.0016036080349655136, + "loss": 0.1118, + "step": 30543 + }, + { + "epoch": 0.26513658735601253, + "grad_norm": 0.474609375, + "learning_rate": 0.001603583367958361, + "loss": 0.0957, + "step": 30544 + }, + { + "epoch": 0.2651452678362167, + "grad_norm": 0.0859375, + "learning_rate": 0.001603558700400498, + "loss": 0.0972, + "step": 30545 + }, + { + "epoch": 0.26515394831642086, + "grad_norm": 0.1611328125, + "learning_rate": 0.0016035340322919504, + "loss": 0.082, + "step": 30546 + }, + { + "epoch": 0.26516262879662506, + "grad_norm": 0.59765625, + "learning_rate": 0.0016035093636327456, + "loss": 0.1113, + "step": 30547 + }, + { + "epoch": 0.2651713092768292, + "grad_norm": 0.50390625, + "learning_rate": 0.0016034846944229103, + "loss": 0.1396, + "step": 30548 + }, + { + "epoch": 0.2651799897570334, + "grad_norm": 0.4765625, + "learning_rate": 0.0016034600246624718, + "loss": 0.1006, + "step": 30549 + }, + { + "epoch": 0.2651886702372375, + "grad_norm": 0.181640625, + "learning_rate": 0.001603435354351457, + "loss": 0.1079, + "step": 30550 + }, + { + "epoch": 0.2651973507174417, + "grad_norm": 0.3046875, + "learning_rate": 0.0016034106834898931, + "loss": 0.1025, + "step": 30551 + }, + { + "epoch": 0.26520603119764585, + "grad_norm": 0.287109375, + "learning_rate": 0.0016033860120778065, + "loss": 0.0967, + "step": 30552 + }, + { + "epoch": 0.26521471167785005, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016033613401152248, + "loss": 0.0825, + "step": 30553 + }, + { + "epoch": 0.2652233921580542, + "grad_norm": 0.5859375, + "learning_rate": 0.0016033366676021746, + "loss": 0.085, + "step": 30554 + }, + { + "epoch": 0.2652320726382584, + "grad_norm": 0.37109375, + "learning_rate": 0.001603311994538683, + "loss": 0.1074, + "step": 30555 + }, + { + "epoch": 0.2652407531184625, + "grad_norm": 0.1787109375, + "learning_rate": 0.0016032873209247769, + "loss": 0.1143, + "step": 30556 + }, + { + "epoch": 0.26524943359866665, + "grad_norm": 0.234375, + "learning_rate": 0.001603262646760483, + "loss": 0.0767, + "step": 30557 + }, + { + "epoch": 0.26525811407887084, + "grad_norm": 0.11083984375, + "learning_rate": 0.0016032379720458292, + "loss": 0.1118, + "step": 30558 + }, + { + "epoch": 0.265266794559075, + "grad_norm": 0.1728515625, + "learning_rate": 0.0016032132967808417, + "loss": 0.1201, + "step": 30559 + }, + { + "epoch": 0.2652754750392792, + "grad_norm": 0.142578125, + "learning_rate": 0.0016031886209655478, + "loss": 0.1387, + "step": 30560 + }, + { + "epoch": 0.2652841555194833, + "grad_norm": 0.4375, + "learning_rate": 0.0016031639445999743, + "loss": 0.1289, + "step": 30561 + }, + { + "epoch": 0.2652928359996875, + "grad_norm": 0.65234375, + "learning_rate": 0.001603139267684148, + "loss": 0.0913, + "step": 30562 + }, + { + "epoch": 0.26530151647989164, + "grad_norm": 0.357421875, + "learning_rate": 0.0016031145902180965, + "loss": 0.2295, + "step": 30563 + }, + { + "epoch": 0.26531019696009583, + "grad_norm": 0.1552734375, + "learning_rate": 0.0016030899122018464, + "loss": 0.1143, + "step": 30564 + }, + { + "epoch": 0.26531887744029997, + "grad_norm": 0.162109375, + "learning_rate": 0.001603065233635425, + "loss": 0.0879, + "step": 30565 + }, + { + "epoch": 0.26532755792050416, + "grad_norm": 0.4921875, + "learning_rate": 0.0016030405545188586, + "loss": 0.1162, + "step": 30566 + }, + { + "epoch": 0.2653362384007083, + "grad_norm": 0.47265625, + "learning_rate": 0.0016030158748521748, + "loss": 0.1602, + "step": 30567 + }, + { + "epoch": 0.2653449188809125, + "grad_norm": 0.416015625, + "learning_rate": 0.0016029911946354003, + "loss": 0.1187, + "step": 30568 + }, + { + "epoch": 0.26535359936111663, + "grad_norm": 0.5625, + "learning_rate": 0.0016029665138685622, + "loss": 0.1816, + "step": 30569 + }, + { + "epoch": 0.2653622798413208, + "grad_norm": 0.875, + "learning_rate": 0.0016029418325516876, + "loss": 0.1318, + "step": 30570 + }, + { + "epoch": 0.26537096032152496, + "grad_norm": 0.138671875, + "learning_rate": 0.0016029171506848033, + "loss": 0.0947, + "step": 30571 + }, + { + "epoch": 0.26537964080172916, + "grad_norm": 0.10888671875, + "learning_rate": 0.001602892468267936, + "loss": 0.0811, + "step": 30572 + }, + { + "epoch": 0.2653883212819333, + "grad_norm": 0.431640625, + "learning_rate": 0.0016028677853011136, + "loss": 0.103, + "step": 30573 + }, + { + "epoch": 0.2653970017621375, + "grad_norm": 0.375, + "learning_rate": 0.0016028431017843623, + "loss": 0.1406, + "step": 30574 + }, + { + "epoch": 0.2654056822423416, + "grad_norm": 0.431640625, + "learning_rate": 0.0016028184177177092, + "loss": 0.1162, + "step": 30575 + }, + { + "epoch": 0.2654143627225458, + "grad_norm": 1.875, + "learning_rate": 0.0016027937331011815, + "loss": 0.3105, + "step": 30576 + }, + { + "epoch": 0.26542304320274995, + "grad_norm": 0.15625, + "learning_rate": 0.0016027690479348063, + "loss": 0.0737, + "step": 30577 + }, + { + "epoch": 0.26543172368295415, + "grad_norm": 0.51953125, + "learning_rate": 0.0016027443622186107, + "loss": 0.1187, + "step": 30578 + }, + { + "epoch": 0.2654404041631583, + "grad_norm": 0.255859375, + "learning_rate": 0.0016027196759526208, + "loss": 0.0986, + "step": 30579 + }, + { + "epoch": 0.2654490846433625, + "grad_norm": 0.443359375, + "learning_rate": 0.0016026949891368642, + "loss": 0.127, + "step": 30580 + }, + { + "epoch": 0.2654577651235666, + "grad_norm": 0.51953125, + "learning_rate": 0.0016026703017713684, + "loss": 0.0913, + "step": 30581 + }, + { + "epoch": 0.2654664456037708, + "grad_norm": 0.423828125, + "learning_rate": 0.0016026456138561597, + "loss": 0.1172, + "step": 30582 + }, + { + "epoch": 0.26547512608397494, + "grad_norm": 0.2158203125, + "learning_rate": 0.001602620925391265, + "loss": 0.0908, + "step": 30583 + }, + { + "epoch": 0.26548380656417914, + "grad_norm": 0.421875, + "learning_rate": 0.0016025962363767124, + "loss": 0.0752, + "step": 30584 + }, + { + "epoch": 0.2654924870443833, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016025715468125274, + "loss": 0.1514, + "step": 30585 + }, + { + "epoch": 0.26550116752458747, + "grad_norm": 0.298828125, + "learning_rate": 0.0016025468566987378, + "loss": 0.1787, + "step": 30586 + }, + { + "epoch": 0.2655098480047916, + "grad_norm": 0.18359375, + "learning_rate": 0.0016025221660353707, + "loss": 0.1025, + "step": 30587 + }, + { + "epoch": 0.2655185284849958, + "grad_norm": 0.2890625, + "learning_rate": 0.0016024974748224529, + "loss": 0.1016, + "step": 30588 + }, + { + "epoch": 0.26552720896519993, + "grad_norm": 0.0712890625, + "learning_rate": 0.0016024727830600111, + "loss": 0.0752, + "step": 30589 + }, + { + "epoch": 0.2655358894454041, + "grad_norm": 0.10791015625, + "learning_rate": 0.0016024480907480731, + "loss": 0.0864, + "step": 30590 + }, + { + "epoch": 0.26554456992560826, + "grad_norm": 0.11181640625, + "learning_rate": 0.0016024233978866654, + "loss": 0.1025, + "step": 30591 + }, + { + "epoch": 0.26555325040581246, + "grad_norm": 0.3515625, + "learning_rate": 0.0016023987044758144, + "loss": 0.1621, + "step": 30592 + }, + { + "epoch": 0.2655619308860166, + "grad_norm": 0.11767578125, + "learning_rate": 0.0016023740105155482, + "loss": 0.1133, + "step": 30593 + }, + { + "epoch": 0.2655706113662208, + "grad_norm": 0.6640625, + "learning_rate": 0.0016023493160058934, + "loss": 0.085, + "step": 30594 + }, + { + "epoch": 0.2655792918464249, + "grad_norm": 2.375, + "learning_rate": 0.0016023246209468766, + "loss": 0.4629, + "step": 30595 + }, + { + "epoch": 0.2655879723266291, + "grad_norm": 0.3984375, + "learning_rate": 0.0016022999253385256, + "loss": 0.0928, + "step": 30596 + }, + { + "epoch": 0.26559665280683326, + "grad_norm": 0.2890625, + "learning_rate": 0.0016022752291808666, + "loss": 0.1387, + "step": 30597 + }, + { + "epoch": 0.26560533328703745, + "grad_norm": 0.26171875, + "learning_rate": 0.001602250532473927, + "loss": 0.1309, + "step": 30598 + }, + { + "epoch": 0.2656140137672416, + "grad_norm": 0.08642578125, + "learning_rate": 0.0016022258352177338, + "loss": 0.0938, + "step": 30599 + }, + { + "epoch": 0.2656226942474458, + "grad_norm": 0.328125, + "learning_rate": 0.001602201137412314, + "loss": 0.1162, + "step": 30600 + }, + { + "epoch": 0.2656313747276499, + "grad_norm": 0.07763671875, + "learning_rate": 0.0016021764390576948, + "loss": 0.1104, + "step": 30601 + }, + { + "epoch": 0.2656400552078541, + "grad_norm": 0.09326171875, + "learning_rate": 0.0016021517401539026, + "loss": 0.0908, + "step": 30602 + }, + { + "epoch": 0.26564873568805825, + "grad_norm": 0.2734375, + "learning_rate": 0.001602127040700965, + "loss": 0.1064, + "step": 30603 + }, + { + "epoch": 0.26565741616826244, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016021023406989088, + "loss": 0.0903, + "step": 30604 + }, + { + "epoch": 0.2656660966484666, + "grad_norm": 0.353515625, + "learning_rate": 0.001602077640147761, + "loss": 0.0977, + "step": 30605 + }, + { + "epoch": 0.26567477712867077, + "grad_norm": 0.62109375, + "learning_rate": 0.0016020529390475488, + "loss": 0.0967, + "step": 30606 + }, + { + "epoch": 0.2656834576088749, + "grad_norm": 0.1416015625, + "learning_rate": 0.0016020282373982986, + "loss": 0.1143, + "step": 30607 + }, + { + "epoch": 0.2656921380890791, + "grad_norm": 0.384765625, + "learning_rate": 0.0016020035352000382, + "loss": 0.0977, + "step": 30608 + }, + { + "epoch": 0.26570081856928324, + "grad_norm": 0.8359375, + "learning_rate": 0.0016019788324527942, + "loss": 0.1167, + "step": 30609 + }, + { + "epoch": 0.26570949904948743, + "grad_norm": 0.23046875, + "learning_rate": 0.0016019541291565937, + "loss": 0.084, + "step": 30610 + }, + { + "epoch": 0.26571817952969157, + "grad_norm": 0.072265625, + "learning_rate": 0.0016019294253114638, + "loss": 0.0884, + "step": 30611 + }, + { + "epoch": 0.26572686000989576, + "grad_norm": 0.2001953125, + "learning_rate": 0.0016019047209174314, + "loss": 0.0884, + "step": 30612 + }, + { + "epoch": 0.2657355404900999, + "grad_norm": 0.2353515625, + "learning_rate": 0.0016018800159745233, + "loss": 0.0811, + "step": 30613 + }, + { + "epoch": 0.2657442209703041, + "grad_norm": 0.1240234375, + "learning_rate": 0.0016018553104827668, + "loss": 0.083, + "step": 30614 + }, + { + "epoch": 0.2657529014505082, + "grad_norm": 0.08837890625, + "learning_rate": 0.001601830604442189, + "loss": 0.0918, + "step": 30615 + }, + { + "epoch": 0.2657615819307124, + "grad_norm": 0.58203125, + "learning_rate": 0.001601805897852817, + "loss": 0.1738, + "step": 30616 + }, + { + "epoch": 0.26577026241091656, + "grad_norm": 0.09765625, + "learning_rate": 0.0016017811907146776, + "loss": 0.0913, + "step": 30617 + }, + { + "epoch": 0.26577894289112075, + "grad_norm": 0.134765625, + "learning_rate": 0.0016017564830277972, + "loss": 0.1201, + "step": 30618 + }, + { + "epoch": 0.2657876233713249, + "grad_norm": 0.1728515625, + "learning_rate": 0.001601731774792204, + "loss": 0.0957, + "step": 30619 + }, + { + "epoch": 0.2657963038515291, + "grad_norm": 0.330078125, + "learning_rate": 0.0016017070660079246, + "loss": 0.104, + "step": 30620 + }, + { + "epoch": 0.2658049843317332, + "grad_norm": 0.19921875, + "learning_rate": 0.0016016823566749857, + "loss": 0.1621, + "step": 30621 + }, + { + "epoch": 0.2658136648119374, + "grad_norm": 0.07373046875, + "learning_rate": 0.001601657646793414, + "loss": 0.0879, + "step": 30622 + }, + { + "epoch": 0.26582234529214155, + "grad_norm": 0.5, + "learning_rate": 0.0016016329363632376, + "loss": 0.1123, + "step": 30623 + }, + { + "epoch": 0.26583102577234574, + "grad_norm": 0.08984375, + "learning_rate": 0.001601608225384483, + "loss": 0.1162, + "step": 30624 + }, + { + "epoch": 0.2658397062525499, + "grad_norm": 0.263671875, + "learning_rate": 0.001601583513857177, + "loss": 0.1465, + "step": 30625 + }, + { + "epoch": 0.26584838673275407, + "grad_norm": 0.55078125, + "learning_rate": 0.0016015588017813467, + "loss": 0.1123, + "step": 30626 + }, + { + "epoch": 0.2658570672129582, + "grad_norm": 0.236328125, + "learning_rate": 0.0016015340891570194, + "loss": 0.1387, + "step": 30627 + }, + { + "epoch": 0.2658657476931624, + "grad_norm": 0.2578125, + "learning_rate": 0.001601509375984222, + "loss": 0.0786, + "step": 30628 + }, + { + "epoch": 0.26587442817336654, + "grad_norm": 0.3046875, + "learning_rate": 0.0016014846622629815, + "loss": 0.0811, + "step": 30629 + }, + { + "epoch": 0.26588310865357073, + "grad_norm": 0.205078125, + "learning_rate": 0.0016014599479933249, + "loss": 0.0957, + "step": 30630 + }, + { + "epoch": 0.26589178913377487, + "grad_norm": 0.333984375, + "learning_rate": 0.0016014352331752788, + "loss": 0.0879, + "step": 30631 + }, + { + "epoch": 0.26590046961397906, + "grad_norm": 0.18359375, + "learning_rate": 0.0016014105178088715, + "loss": 0.1235, + "step": 30632 + }, + { + "epoch": 0.2659091500941832, + "grad_norm": 0.0859375, + "learning_rate": 0.0016013858018941284, + "loss": 0.124, + "step": 30633 + }, + { + "epoch": 0.2659178305743874, + "grad_norm": 0.109375, + "learning_rate": 0.0016013610854310775, + "loss": 0.1367, + "step": 30634 + }, + { + "epoch": 0.26592651105459153, + "grad_norm": 0.365234375, + "learning_rate": 0.0016013363684197457, + "loss": 0.1069, + "step": 30635 + }, + { + "epoch": 0.2659351915347957, + "grad_norm": 0.27734375, + "learning_rate": 0.0016013116508601603, + "loss": 0.0933, + "step": 30636 + }, + { + "epoch": 0.26594387201499986, + "grad_norm": 0.609375, + "learning_rate": 0.0016012869327523475, + "loss": 0.0913, + "step": 30637 + }, + { + "epoch": 0.26595255249520405, + "grad_norm": 0.1572265625, + "learning_rate": 0.001601262214096335, + "loss": 0.1147, + "step": 30638 + }, + { + "epoch": 0.2659612329754082, + "grad_norm": 0.111328125, + "learning_rate": 0.00160123749489215, + "loss": 0.0977, + "step": 30639 + }, + { + "epoch": 0.2659699134556124, + "grad_norm": 0.50390625, + "learning_rate": 0.001601212775139819, + "loss": 0.0806, + "step": 30640 + }, + { + "epoch": 0.2659785939358165, + "grad_norm": 0.3984375, + "learning_rate": 0.0016011880548393693, + "loss": 0.123, + "step": 30641 + }, + { + "epoch": 0.2659872744160207, + "grad_norm": 0.111328125, + "learning_rate": 0.001601163333990828, + "loss": 0.0791, + "step": 30642 + }, + { + "epoch": 0.26599595489622485, + "grad_norm": 0.083984375, + "learning_rate": 0.0016011386125942217, + "loss": 0.0903, + "step": 30643 + }, + { + "epoch": 0.26600463537642904, + "grad_norm": 0.1337890625, + "learning_rate": 0.0016011138906495777, + "loss": 0.0923, + "step": 30644 + }, + { + "epoch": 0.2660133158566332, + "grad_norm": 0.1630859375, + "learning_rate": 0.001601089168156923, + "loss": 0.1328, + "step": 30645 + }, + { + "epoch": 0.2660219963368374, + "grad_norm": 0.267578125, + "learning_rate": 0.0016010644451162853, + "loss": 0.0972, + "step": 30646 + }, + { + "epoch": 0.2660306768170415, + "grad_norm": 0.359375, + "learning_rate": 0.0016010397215276905, + "loss": 0.0693, + "step": 30647 + }, + { + "epoch": 0.2660393572972457, + "grad_norm": 0.57421875, + "learning_rate": 0.0016010149973911666, + "loss": 0.082, + "step": 30648 + }, + { + "epoch": 0.26604803777744984, + "grad_norm": 0.279296875, + "learning_rate": 0.00160099027270674, + "loss": 0.1089, + "step": 30649 + }, + { + "epoch": 0.26605671825765403, + "grad_norm": 0.15625, + "learning_rate": 0.001600965547474438, + "loss": 0.1226, + "step": 30650 + }, + { + "epoch": 0.26606539873785817, + "grad_norm": 0.1025390625, + "learning_rate": 0.0016009408216942874, + "loss": 0.0811, + "step": 30651 + }, + { + "epoch": 0.26607407921806236, + "grad_norm": 0.89453125, + "learning_rate": 0.0016009160953663155, + "loss": 0.0908, + "step": 30652 + }, + { + "epoch": 0.2660827596982665, + "grad_norm": 0.55078125, + "learning_rate": 0.0016008913684905492, + "loss": 0.0908, + "step": 30653 + }, + { + "epoch": 0.2660914401784707, + "grad_norm": 0.53125, + "learning_rate": 0.0016008666410670164, + "loss": 0.1094, + "step": 30654 + }, + { + "epoch": 0.26610012065867483, + "grad_norm": 0.1435546875, + "learning_rate": 0.0016008419130957427, + "loss": 0.0938, + "step": 30655 + }, + { + "epoch": 0.266108801138879, + "grad_norm": 0.158203125, + "learning_rate": 0.0016008171845767558, + "loss": 0.0879, + "step": 30656 + }, + { + "epoch": 0.26611748161908316, + "grad_norm": 0.431640625, + "learning_rate": 0.0016007924555100827, + "loss": 0.1406, + "step": 30657 + }, + { + "epoch": 0.26612616209928736, + "grad_norm": 0.2099609375, + "learning_rate": 0.0016007677258957507, + "loss": 0.1079, + "step": 30658 + }, + { + "epoch": 0.2661348425794915, + "grad_norm": 0.27734375, + "learning_rate": 0.0016007429957337865, + "loss": 0.1035, + "step": 30659 + }, + { + "epoch": 0.2661435230596957, + "grad_norm": 0.458984375, + "learning_rate": 0.0016007182650242175, + "loss": 0.1172, + "step": 30660 + }, + { + "epoch": 0.2661522035398998, + "grad_norm": 0.267578125, + "learning_rate": 0.0016006935337670705, + "loss": 0.0771, + "step": 30661 + }, + { + "epoch": 0.266160884020104, + "grad_norm": 0.546875, + "learning_rate": 0.0016006688019623723, + "loss": 0.0977, + "step": 30662 + }, + { + "epoch": 0.26616956450030815, + "grad_norm": 0.31640625, + "learning_rate": 0.0016006440696101506, + "loss": 0.1055, + "step": 30663 + }, + { + "epoch": 0.26617824498051235, + "grad_norm": 0.8125, + "learning_rate": 0.0016006193367104318, + "loss": 0.1328, + "step": 30664 + }, + { + "epoch": 0.2661869254607165, + "grad_norm": 0.09619140625, + "learning_rate": 0.001600594603263243, + "loss": 0.1069, + "step": 30665 + }, + { + "epoch": 0.2661956059409207, + "grad_norm": 0.1513671875, + "learning_rate": 0.001600569869268612, + "loss": 0.1216, + "step": 30666 + }, + { + "epoch": 0.2662042864211248, + "grad_norm": 0.322265625, + "learning_rate": 0.0016005451347265648, + "loss": 0.084, + "step": 30667 + }, + { + "epoch": 0.266212966901329, + "grad_norm": 0.53515625, + "learning_rate": 0.0016005203996371294, + "loss": 0.1025, + "step": 30668 + }, + { + "epoch": 0.26622164738153314, + "grad_norm": 0.1376953125, + "learning_rate": 0.0016004956640003323, + "loss": 0.0708, + "step": 30669 + }, + { + "epoch": 0.26623032786173734, + "grad_norm": 0.173828125, + "learning_rate": 0.0016004709278162007, + "loss": 0.1162, + "step": 30670 + }, + { + "epoch": 0.2662390083419415, + "grad_norm": 0.08984375, + "learning_rate": 0.0016004461910847615, + "loss": 0.0962, + "step": 30671 + }, + { + "epoch": 0.26624768882214567, + "grad_norm": 1.1015625, + "learning_rate": 0.0016004214538060417, + "loss": 0.0972, + "step": 30672 + }, + { + "epoch": 0.2662563693023498, + "grad_norm": 0.69140625, + "learning_rate": 0.0016003967159800689, + "loss": 0.1133, + "step": 30673 + }, + { + "epoch": 0.266265049782554, + "grad_norm": 0.515625, + "learning_rate": 0.0016003719776068693, + "loss": 0.1094, + "step": 30674 + }, + { + "epoch": 0.26627373026275813, + "grad_norm": 0.07421875, + "learning_rate": 0.0016003472386864708, + "loss": 0.0864, + "step": 30675 + }, + { + "epoch": 0.2662824107429623, + "grad_norm": 0.123046875, + "learning_rate": 0.0016003224992189, + "loss": 0.1426, + "step": 30676 + }, + { + "epoch": 0.26629109122316647, + "grad_norm": 0.251953125, + "learning_rate": 0.0016002977592041838, + "loss": 0.0854, + "step": 30677 + }, + { + "epoch": 0.26629977170337066, + "grad_norm": 0.197265625, + "learning_rate": 0.0016002730186423497, + "loss": 0.0879, + "step": 30678 + }, + { + "epoch": 0.2663084521835748, + "grad_norm": 0.10107421875, + "learning_rate": 0.0016002482775334247, + "loss": 0.1182, + "step": 30679 + }, + { + "epoch": 0.26631713266377893, + "grad_norm": 0.10205078125, + "learning_rate": 0.0016002235358774353, + "loss": 0.1221, + "step": 30680 + }, + { + "epoch": 0.2663258131439831, + "grad_norm": 0.498046875, + "learning_rate": 0.0016001987936744093, + "loss": 0.0996, + "step": 30681 + }, + { + "epoch": 0.26633449362418726, + "grad_norm": 0.8125, + "learning_rate": 0.0016001740509243734, + "loss": 0.0952, + "step": 30682 + }, + { + "epoch": 0.26634317410439146, + "grad_norm": 0.095703125, + "learning_rate": 0.0016001493076273543, + "loss": 0.0967, + "step": 30683 + }, + { + "epoch": 0.2663518545845956, + "grad_norm": 0.421875, + "learning_rate": 0.0016001245637833799, + "loss": 0.0762, + "step": 30684 + }, + { + "epoch": 0.2663605350647998, + "grad_norm": 0.103515625, + "learning_rate": 0.0016000998193924764, + "loss": 0.1011, + "step": 30685 + }, + { + "epoch": 0.2663692155450039, + "grad_norm": 0.291015625, + "learning_rate": 0.0016000750744546714, + "loss": 0.103, + "step": 30686 + }, + { + "epoch": 0.2663778960252081, + "grad_norm": 0.2021484375, + "learning_rate": 0.0016000503289699919, + "loss": 0.1152, + "step": 30687 + }, + { + "epoch": 0.26638657650541225, + "grad_norm": 1.4453125, + "learning_rate": 0.0016000255829384648, + "loss": 0.2969, + "step": 30688 + }, + { + "epoch": 0.26639525698561645, + "grad_norm": 0.53515625, + "learning_rate": 0.0016000008363601175, + "loss": 0.1055, + "step": 30689 + }, + { + "epoch": 0.2664039374658206, + "grad_norm": 0.474609375, + "learning_rate": 0.0015999760892349765, + "loss": 0.0806, + "step": 30690 + }, + { + "epoch": 0.2664126179460248, + "grad_norm": 0.111328125, + "learning_rate": 0.0015999513415630691, + "loss": 0.1523, + "step": 30691 + }, + { + "epoch": 0.2664212984262289, + "grad_norm": 0.44921875, + "learning_rate": 0.0015999265933444224, + "loss": 0.1768, + "step": 30692 + }, + { + "epoch": 0.2664299789064331, + "grad_norm": 0.1806640625, + "learning_rate": 0.001599901844579064, + "loss": 0.1641, + "step": 30693 + }, + { + "epoch": 0.26643865938663724, + "grad_norm": 0.796875, + "learning_rate": 0.0015998770952670202, + "loss": 0.1011, + "step": 30694 + }, + { + "epoch": 0.26644733986684144, + "grad_norm": 0.1845703125, + "learning_rate": 0.001599852345408318, + "loss": 0.1167, + "step": 30695 + }, + { + "epoch": 0.2664560203470456, + "grad_norm": 0.1044921875, + "learning_rate": 0.001599827595002985, + "loss": 0.0952, + "step": 30696 + }, + { + "epoch": 0.26646470082724977, + "grad_norm": 0.0947265625, + "learning_rate": 0.001599802844051048, + "loss": 0.1299, + "step": 30697 + }, + { + "epoch": 0.2664733813074539, + "grad_norm": 0.1630859375, + "learning_rate": 0.001599778092552534, + "loss": 0.1045, + "step": 30698 + }, + { + "epoch": 0.2664820617876581, + "grad_norm": 0.375, + "learning_rate": 0.0015997533405074705, + "loss": 0.085, + "step": 30699 + }, + { + "epoch": 0.26649074226786224, + "grad_norm": 0.419921875, + "learning_rate": 0.001599728587915884, + "loss": 0.1143, + "step": 30700 + }, + { + "epoch": 0.26649942274806643, + "grad_norm": 0.189453125, + "learning_rate": 0.0015997038347778019, + "loss": 0.0996, + "step": 30701 + }, + { + "epoch": 0.26650810322827057, + "grad_norm": 0.24609375, + "learning_rate": 0.0015996790810932516, + "loss": 0.1211, + "step": 30702 + }, + { + "epoch": 0.26651678370847476, + "grad_norm": 0.6015625, + "learning_rate": 0.001599654326862259, + "loss": 0.1348, + "step": 30703 + }, + { + "epoch": 0.2665254641886789, + "grad_norm": 0.2470703125, + "learning_rate": 0.0015996295720848526, + "loss": 0.124, + "step": 30704 + }, + { + "epoch": 0.2665341446688831, + "grad_norm": 0.353515625, + "learning_rate": 0.0015996048167610584, + "loss": 0.1201, + "step": 30705 + }, + { + "epoch": 0.2665428251490872, + "grad_norm": 0.2265625, + "learning_rate": 0.0015995800608909038, + "loss": 0.1182, + "step": 30706 + }, + { + "epoch": 0.2665515056292914, + "grad_norm": 0.0830078125, + "learning_rate": 0.0015995553044744165, + "loss": 0.1064, + "step": 30707 + }, + { + "epoch": 0.26656018610949556, + "grad_norm": 0.4296875, + "learning_rate": 0.0015995305475116225, + "loss": 0.1416, + "step": 30708 + }, + { + "epoch": 0.26656886658969975, + "grad_norm": 0.119140625, + "learning_rate": 0.0015995057900025497, + "loss": 0.0996, + "step": 30709 + }, + { + "epoch": 0.2665775470699039, + "grad_norm": 0.330078125, + "learning_rate": 0.0015994810319472246, + "loss": 0.1484, + "step": 30710 + }, + { + "epoch": 0.2665862275501081, + "grad_norm": 0.22265625, + "learning_rate": 0.0015994562733456748, + "loss": 0.1094, + "step": 30711 + }, + { + "epoch": 0.2665949080303122, + "grad_norm": 0.47265625, + "learning_rate": 0.0015994315141979272, + "loss": 0.0977, + "step": 30712 + }, + { + "epoch": 0.2666035885105164, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015994067545040086, + "loss": 0.1172, + "step": 30713 + }, + { + "epoch": 0.26661226899072055, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015993819942639463, + "loss": 0.1045, + "step": 30714 + }, + { + "epoch": 0.26662094947092474, + "grad_norm": 0.2294921875, + "learning_rate": 0.001599357233477767, + "loss": 0.1138, + "step": 30715 + }, + { + "epoch": 0.2666296299511289, + "grad_norm": 0.1064453125, + "learning_rate": 0.0015993324721454987, + "loss": 0.0972, + "step": 30716 + }, + { + "epoch": 0.26663831043133307, + "grad_norm": 0.138671875, + "learning_rate": 0.0015993077102671677, + "loss": 0.125, + "step": 30717 + }, + { + "epoch": 0.2666469909115372, + "grad_norm": 0.099609375, + "learning_rate": 0.0015992829478428015, + "loss": 0.0962, + "step": 30718 + }, + { + "epoch": 0.2666556713917414, + "grad_norm": 0.2392578125, + "learning_rate": 0.0015992581848724265, + "loss": 0.0903, + "step": 30719 + }, + { + "epoch": 0.26666435187194554, + "grad_norm": 0.169921875, + "learning_rate": 0.0015992334213560705, + "loss": 0.1084, + "step": 30720 + }, + { + "epoch": 0.26667303235214973, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015992086572937604, + "loss": 0.0806, + "step": 30721 + }, + { + "epoch": 0.26668171283235387, + "grad_norm": 0.314453125, + "learning_rate": 0.001599183892685523, + "loss": 0.0854, + "step": 30722 + }, + { + "epoch": 0.26669039331255806, + "grad_norm": 0.07958984375, + "learning_rate": 0.0015991591275313858, + "loss": 0.1152, + "step": 30723 + }, + { + "epoch": 0.2666990737927622, + "grad_norm": 0.0673828125, + "learning_rate": 0.001599134361831376, + "loss": 0.0693, + "step": 30724 + }, + { + "epoch": 0.2667077542729664, + "grad_norm": 0.083984375, + "learning_rate": 0.0015991095955855197, + "loss": 0.0776, + "step": 30725 + }, + { + "epoch": 0.26671643475317053, + "grad_norm": 0.58203125, + "learning_rate": 0.0015990848287938452, + "loss": 0.2539, + "step": 30726 + }, + { + "epoch": 0.2667251152333747, + "grad_norm": 0.515625, + "learning_rate": 0.0015990600614563784, + "loss": 0.1172, + "step": 30727 + }, + { + "epoch": 0.26673379571357886, + "grad_norm": 0.79296875, + "learning_rate": 0.0015990352935731475, + "loss": 0.1138, + "step": 30728 + }, + { + "epoch": 0.26674247619378305, + "grad_norm": 0.45703125, + "learning_rate": 0.0015990105251441791, + "loss": 0.1416, + "step": 30729 + }, + { + "epoch": 0.2667511566739872, + "grad_norm": 0.26953125, + "learning_rate": 0.0015989857561695, + "loss": 0.1089, + "step": 30730 + }, + { + "epoch": 0.2667598371541914, + "grad_norm": 0.353515625, + "learning_rate": 0.001598960986649138, + "loss": 0.1064, + "step": 30731 + }, + { + "epoch": 0.2667685176343955, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015989362165831193, + "loss": 0.1201, + "step": 30732 + }, + { + "epoch": 0.2667771981145997, + "grad_norm": 0.18359375, + "learning_rate": 0.0015989114459714717, + "loss": 0.1289, + "step": 30733 + }, + { + "epoch": 0.26678587859480385, + "grad_norm": 0.1904296875, + "learning_rate": 0.001598886674814222, + "loss": 0.1069, + "step": 30734 + }, + { + "epoch": 0.26679455907500804, + "grad_norm": 0.236328125, + "learning_rate": 0.0015988619031113973, + "loss": 0.1465, + "step": 30735 + }, + { + "epoch": 0.2668032395552122, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015988371308630247, + "loss": 0.1318, + "step": 30736 + }, + { + "epoch": 0.2668119200354164, + "grad_norm": 0.283203125, + "learning_rate": 0.0015988123580691314, + "loss": 0.1006, + "step": 30737 + }, + { + "epoch": 0.2668206005156205, + "grad_norm": 0.1044921875, + "learning_rate": 0.0015987875847297444, + "loss": 0.105, + "step": 30738 + }, + { + "epoch": 0.2668292809958247, + "grad_norm": 0.115234375, + "learning_rate": 0.0015987628108448907, + "loss": 0.1123, + "step": 30739 + }, + { + "epoch": 0.26683796147602884, + "grad_norm": 0.142578125, + "learning_rate": 0.0015987380364145974, + "loss": 0.106, + "step": 30740 + }, + { + "epoch": 0.26684664195623303, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015987132614388919, + "loss": 0.1641, + "step": 30741 + }, + { + "epoch": 0.26685532243643717, + "grad_norm": 0.392578125, + "learning_rate": 0.001598688485917801, + "loss": 0.0791, + "step": 30742 + }, + { + "epoch": 0.26686400291664136, + "grad_norm": 0.298828125, + "learning_rate": 0.0015986637098513517, + "loss": 0.1162, + "step": 30743 + }, + { + "epoch": 0.2668726833968455, + "grad_norm": 0.21484375, + "learning_rate": 0.0015986389332395714, + "loss": 0.1338, + "step": 30744 + }, + { + "epoch": 0.2668813638770497, + "grad_norm": 0.625, + "learning_rate": 0.0015986141560824872, + "loss": 0.1338, + "step": 30745 + }, + { + "epoch": 0.26689004435725383, + "grad_norm": 0.1015625, + "learning_rate": 0.0015985893783801257, + "loss": 0.1128, + "step": 30746 + }, + { + "epoch": 0.266898724837458, + "grad_norm": 0.138671875, + "learning_rate": 0.0015985646001325144, + "loss": 0.127, + "step": 30747 + }, + { + "epoch": 0.26690740531766216, + "grad_norm": 0.244140625, + "learning_rate": 0.0015985398213396807, + "loss": 0.0913, + "step": 30748 + }, + { + "epoch": 0.26691608579786635, + "grad_norm": 4.03125, + "learning_rate": 0.001598515042001651, + "loss": 0.1484, + "step": 30749 + }, + { + "epoch": 0.2669247662780705, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015984902621184532, + "loss": 0.0981, + "step": 30750 + }, + { + "epoch": 0.2669334467582747, + "grad_norm": 0.61328125, + "learning_rate": 0.0015984654816901137, + "loss": 0.127, + "step": 30751 + }, + { + "epoch": 0.2669421272384788, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015984407007166594, + "loss": 0.1289, + "step": 30752 + }, + { + "epoch": 0.266950807718683, + "grad_norm": 0.271484375, + "learning_rate": 0.0015984159191981182, + "loss": 0.0894, + "step": 30753 + }, + { + "epoch": 0.26695948819888715, + "grad_norm": 0.10546875, + "learning_rate": 0.001598391137134517, + "loss": 0.0947, + "step": 30754 + }, + { + "epoch": 0.26696816867909134, + "grad_norm": 0.25390625, + "learning_rate": 0.0015983663545258825, + "loss": 0.1367, + "step": 30755 + }, + { + "epoch": 0.2669768491592955, + "grad_norm": 0.54296875, + "learning_rate": 0.001598341571372242, + "loss": 0.1611, + "step": 30756 + }, + { + "epoch": 0.2669855296394997, + "grad_norm": 0.7578125, + "learning_rate": 0.001598316787673623, + "loss": 0.0854, + "step": 30757 + }, + { + "epoch": 0.2669942101197038, + "grad_norm": 0.12890625, + "learning_rate": 0.0015982920034300519, + "loss": 0.0854, + "step": 30758 + }, + { + "epoch": 0.267002890599908, + "grad_norm": 0.3515625, + "learning_rate": 0.0015982672186415563, + "loss": 0.1299, + "step": 30759 + }, + { + "epoch": 0.26701157108011214, + "grad_norm": 1.078125, + "learning_rate": 0.001598242433308163, + "loss": 0.1201, + "step": 30760 + }, + { + "epoch": 0.26702025156031634, + "grad_norm": 0.6953125, + "learning_rate": 0.0015982176474298995, + "loss": 0.1001, + "step": 30761 + }, + { + "epoch": 0.2670289320405205, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015981928610067926, + "loss": 0.0693, + "step": 30762 + }, + { + "epoch": 0.26703761252072467, + "grad_norm": 0.08935546875, + "learning_rate": 0.0015981680740388693, + "loss": 0.0811, + "step": 30763 + }, + { + "epoch": 0.2670462930009288, + "grad_norm": 0.4765625, + "learning_rate": 0.0015981432865261572, + "loss": 0.0962, + "step": 30764 + }, + { + "epoch": 0.267054973481133, + "grad_norm": 0.14453125, + "learning_rate": 0.0015981184984686827, + "loss": 0.1001, + "step": 30765 + }, + { + "epoch": 0.26706365396133713, + "grad_norm": 0.34765625, + "learning_rate": 0.0015980937098664734, + "loss": 0.0942, + "step": 30766 + }, + { + "epoch": 0.2670723344415413, + "grad_norm": 0.115234375, + "learning_rate": 0.0015980689207195566, + "loss": 0.1113, + "step": 30767 + }, + { + "epoch": 0.26708101492174546, + "grad_norm": 0.12451171875, + "learning_rate": 0.0015980441310279587, + "loss": 0.0986, + "step": 30768 + }, + { + "epoch": 0.26708969540194966, + "grad_norm": 0.451171875, + "learning_rate": 0.0015980193407917074, + "loss": 0.0884, + "step": 30769 + }, + { + "epoch": 0.2670983758821538, + "grad_norm": 0.453125, + "learning_rate": 0.0015979945500108296, + "loss": 0.0854, + "step": 30770 + }, + { + "epoch": 0.267107056362358, + "grad_norm": 0.369140625, + "learning_rate": 0.0015979697586853527, + "loss": 0.105, + "step": 30771 + }, + { + "epoch": 0.2671157368425621, + "grad_norm": 0.216796875, + "learning_rate": 0.0015979449668153032, + "loss": 0.1445, + "step": 30772 + }, + { + "epoch": 0.2671244173227663, + "grad_norm": 0.0634765625, + "learning_rate": 0.0015979201744007086, + "loss": 0.0598, + "step": 30773 + }, + { + "epoch": 0.26713309780297045, + "grad_norm": 0.310546875, + "learning_rate": 0.0015978953814415962, + "loss": 0.1094, + "step": 30774 + }, + { + "epoch": 0.26714177828317465, + "grad_norm": 0.53125, + "learning_rate": 0.0015978705879379927, + "loss": 0.1006, + "step": 30775 + }, + { + "epoch": 0.2671504587633788, + "grad_norm": 0.92578125, + "learning_rate": 0.0015978457938899257, + "loss": 0.1172, + "step": 30776 + }, + { + "epoch": 0.267159139243583, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015978209992974215, + "loss": 0.085, + "step": 30777 + }, + { + "epoch": 0.2671678197237871, + "grad_norm": 0.6484375, + "learning_rate": 0.001597796204160508, + "loss": 0.1221, + "step": 30778 + }, + { + "epoch": 0.2671765002039913, + "grad_norm": 0.1416015625, + "learning_rate": 0.0015977714084792123, + "loss": 0.1162, + "step": 30779 + }, + { + "epoch": 0.26718518068419544, + "grad_norm": 0.173828125, + "learning_rate": 0.0015977466122535611, + "loss": 0.0938, + "step": 30780 + }, + { + "epoch": 0.26719386116439964, + "grad_norm": 0.412109375, + "learning_rate": 0.0015977218154835816, + "loss": 0.1094, + "step": 30781 + }, + { + "epoch": 0.2672025416446038, + "grad_norm": 0.51171875, + "learning_rate": 0.0015976970181693008, + "loss": 0.1094, + "step": 30782 + }, + { + "epoch": 0.26721122212480797, + "grad_norm": 0.208984375, + "learning_rate": 0.0015976722203107463, + "loss": 0.0947, + "step": 30783 + }, + { + "epoch": 0.2672199026050121, + "grad_norm": 0.76953125, + "learning_rate": 0.001597647421907945, + "loss": 0.1025, + "step": 30784 + }, + { + "epoch": 0.2672285830852163, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015976226229609239, + "loss": 0.1406, + "step": 30785 + }, + { + "epoch": 0.26723726356542044, + "grad_norm": 0.15625, + "learning_rate": 0.0015975978234697103, + "loss": 0.1367, + "step": 30786 + }, + { + "epoch": 0.26724594404562463, + "grad_norm": 0.08935546875, + "learning_rate": 0.0015975730234343308, + "loss": 0.0708, + "step": 30787 + }, + { + "epoch": 0.26725462452582877, + "grad_norm": 0.578125, + "learning_rate": 0.0015975482228548132, + "loss": 0.0889, + "step": 30788 + }, + { + "epoch": 0.26726330500603296, + "grad_norm": 0.1123046875, + "learning_rate": 0.0015975234217311845, + "loss": 0.1055, + "step": 30789 + }, + { + "epoch": 0.2672719854862371, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015974986200634712, + "loss": 0.0928, + "step": 30790 + }, + { + "epoch": 0.2672806659664413, + "grad_norm": 0.51953125, + "learning_rate": 0.0015974738178517012, + "loss": 0.1094, + "step": 30791 + }, + { + "epoch": 0.2672893464466454, + "grad_norm": 0.11474609375, + "learning_rate": 0.0015974490150959014, + "loss": 0.0591, + "step": 30792 + }, + { + "epoch": 0.2672980269268496, + "grad_norm": 0.33984375, + "learning_rate": 0.0015974242117960985, + "loss": 0.0986, + "step": 30793 + }, + { + "epoch": 0.26730670740705376, + "grad_norm": 0.26953125, + "learning_rate": 0.0015973994079523201, + "loss": 0.2119, + "step": 30794 + }, + { + "epoch": 0.26731538788725795, + "grad_norm": 0.1943359375, + "learning_rate": 0.001597374603564593, + "loss": 0.1455, + "step": 30795 + }, + { + "epoch": 0.2673240683674621, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015973497986329448, + "loss": 0.1016, + "step": 30796 + }, + { + "epoch": 0.2673327488476663, + "grad_norm": 0.09228515625, + "learning_rate": 0.0015973249931574022, + "loss": 0.0918, + "step": 30797 + }, + { + "epoch": 0.2673414293278704, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015973001871379927, + "loss": 0.0967, + "step": 30798 + }, + { + "epoch": 0.2673501098080746, + "grad_norm": 0.482421875, + "learning_rate": 0.001597275380574743, + "loss": 0.2188, + "step": 30799 + }, + { + "epoch": 0.26735879028827875, + "grad_norm": 0.2255859375, + "learning_rate": 0.0015972505734676804, + "loss": 0.0889, + "step": 30800 + }, + { + "epoch": 0.26736747076848294, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015972257658168321, + "loss": 0.1035, + "step": 30801 + }, + { + "epoch": 0.2673761512486871, + "grad_norm": 0.333984375, + "learning_rate": 0.001597200957622225, + "loss": 0.082, + "step": 30802 + }, + { + "epoch": 0.2673848317288912, + "grad_norm": 0.150390625, + "learning_rate": 0.0015971761488838866, + "loss": 0.0718, + "step": 30803 + }, + { + "epoch": 0.2673935122090954, + "grad_norm": 0.33984375, + "learning_rate": 0.0015971513396018434, + "loss": 0.0913, + "step": 30804 + }, + { + "epoch": 0.26740219268929954, + "grad_norm": 0.2353515625, + "learning_rate": 0.0015971265297761235, + "loss": 0.0908, + "step": 30805 + }, + { + "epoch": 0.26741087316950374, + "grad_norm": 0.35546875, + "learning_rate": 0.0015971017194067534, + "loss": 0.127, + "step": 30806 + }, + { + "epoch": 0.2674195536497079, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015970769084937603, + "loss": 0.1445, + "step": 30807 + }, + { + "epoch": 0.26742823412991207, + "grad_norm": 3.03125, + "learning_rate": 0.0015970520970371711, + "loss": 0.124, + "step": 30808 + }, + { + "epoch": 0.2674369146101162, + "grad_norm": 0.12109375, + "learning_rate": 0.0015970272850370135, + "loss": 0.1045, + "step": 30809 + }, + { + "epoch": 0.2674455950903204, + "grad_norm": 0.2119140625, + "learning_rate": 0.0015970024724933144, + "loss": 0.0771, + "step": 30810 + }, + { + "epoch": 0.26745427557052454, + "grad_norm": 0.318359375, + "learning_rate": 0.0015969776594061005, + "loss": 0.1191, + "step": 30811 + }, + { + "epoch": 0.26746295605072873, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015969528457753994, + "loss": 0.1074, + "step": 30812 + }, + { + "epoch": 0.26747163653093287, + "grad_norm": 0.115234375, + "learning_rate": 0.0015969280316012382, + "loss": 0.1387, + "step": 30813 + }, + { + "epoch": 0.26748031701113706, + "grad_norm": 0.66015625, + "learning_rate": 0.001596903216883644, + "loss": 0.1064, + "step": 30814 + }, + { + "epoch": 0.2674889974913412, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015968784016226441, + "loss": 0.0908, + "step": 30815 + }, + { + "epoch": 0.2674976779715454, + "grad_norm": 0.447265625, + "learning_rate": 0.001596853585818265, + "loss": 0.082, + "step": 30816 + }, + { + "epoch": 0.2675063584517495, + "grad_norm": 0.2265625, + "learning_rate": 0.0015968287694705348, + "loss": 0.0938, + "step": 30817 + }, + { + "epoch": 0.2675150389319537, + "grad_norm": 0.10888671875, + "learning_rate": 0.0015968039525794796, + "loss": 0.0786, + "step": 30818 + }, + { + "epoch": 0.26752371941215786, + "grad_norm": 0.341796875, + "learning_rate": 0.0015967791351451275, + "loss": 0.0908, + "step": 30819 + }, + { + "epoch": 0.26753239989236205, + "grad_norm": 0.07861328125, + "learning_rate": 0.0015967543171675049, + "loss": 0.0981, + "step": 30820 + }, + { + "epoch": 0.2675410803725662, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015967294986466397, + "loss": 0.0957, + "step": 30821 + }, + { + "epoch": 0.2675497608527704, + "grad_norm": 0.34765625, + "learning_rate": 0.0015967046795825582, + "loss": 0.1152, + "step": 30822 + }, + { + "epoch": 0.2675584413329745, + "grad_norm": 0.1064453125, + "learning_rate": 0.001596679859975288, + "loss": 0.1069, + "step": 30823 + }, + { + "epoch": 0.2675671218131787, + "grad_norm": 0.109375, + "learning_rate": 0.001596655039824856, + "loss": 0.0806, + "step": 30824 + }, + { + "epoch": 0.26757580229338285, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015966302191312896, + "loss": 0.0903, + "step": 30825 + }, + { + "epoch": 0.26758448277358704, + "grad_norm": 0.53125, + "learning_rate": 0.0015966053978946158, + "loss": 0.1084, + "step": 30826 + }, + { + "epoch": 0.2675931632537912, + "grad_norm": 0.154296875, + "learning_rate": 0.001596580576114862, + "loss": 0.1191, + "step": 30827 + }, + { + "epoch": 0.26760184373399537, + "grad_norm": 0.5078125, + "learning_rate": 0.001596555753792055, + "loss": 0.0938, + "step": 30828 + }, + { + "epoch": 0.2676105242141995, + "grad_norm": 0.15625, + "learning_rate": 0.0015965309309262218, + "loss": 0.1162, + "step": 30829 + }, + { + "epoch": 0.2676192046944037, + "grad_norm": 0.431640625, + "learning_rate": 0.0015965061075173905, + "loss": 0.1104, + "step": 30830 + }, + { + "epoch": 0.26762788517460784, + "grad_norm": 0.349609375, + "learning_rate": 0.0015964812835655873, + "loss": 0.1289, + "step": 30831 + }, + { + "epoch": 0.26763656565481203, + "grad_norm": 0.0673828125, + "learning_rate": 0.0015964564590708395, + "loss": 0.0679, + "step": 30832 + }, + { + "epoch": 0.26764524613501617, + "grad_norm": 0.18359375, + "learning_rate": 0.0015964316340331746, + "loss": 0.0815, + "step": 30833 + }, + { + "epoch": 0.26765392661522036, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015964068084526192, + "loss": 0.1128, + "step": 30834 + }, + { + "epoch": 0.2676626070954245, + "grad_norm": 0.396484375, + "learning_rate": 0.0015963819823292009, + "loss": 0.126, + "step": 30835 + }, + { + "epoch": 0.2676712875756287, + "grad_norm": 0.72265625, + "learning_rate": 0.0015963571556629468, + "loss": 0.1553, + "step": 30836 + }, + { + "epoch": 0.26767996805583283, + "grad_norm": 0.6328125, + "learning_rate": 0.0015963323284538839, + "loss": 0.0859, + "step": 30837 + }, + { + "epoch": 0.267688648536037, + "grad_norm": 0.06640625, + "learning_rate": 0.0015963075007020397, + "loss": 0.0884, + "step": 30838 + }, + { + "epoch": 0.26769732901624116, + "grad_norm": 0.2119140625, + "learning_rate": 0.0015962826724074408, + "loss": 0.1289, + "step": 30839 + }, + { + "epoch": 0.26770600949644535, + "grad_norm": 0.330078125, + "learning_rate": 0.0015962578435701148, + "loss": 0.1318, + "step": 30840 + }, + { + "epoch": 0.2677146899766495, + "grad_norm": 1.1484375, + "learning_rate": 0.0015962330141900882, + "loss": 0.1328, + "step": 30841 + }, + { + "epoch": 0.2677233704568537, + "grad_norm": 0.55859375, + "learning_rate": 0.0015962081842673893, + "loss": 0.1143, + "step": 30842 + }, + { + "epoch": 0.2677320509370578, + "grad_norm": 0.154296875, + "learning_rate": 0.001596183353802044, + "loss": 0.167, + "step": 30843 + }, + { + "epoch": 0.267740731417262, + "grad_norm": 0.67578125, + "learning_rate": 0.0015961585227940802, + "loss": 0.1279, + "step": 30844 + }, + { + "epoch": 0.26774941189746615, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015961336912435253, + "loss": 0.1299, + "step": 30845 + }, + { + "epoch": 0.26775809237767034, + "grad_norm": 0.6796875, + "learning_rate": 0.0015961088591504055, + "loss": 0.1396, + "step": 30846 + }, + { + "epoch": 0.2677667728578745, + "grad_norm": 0.26171875, + "learning_rate": 0.001596084026514749, + "loss": 0.1035, + "step": 30847 + }, + { + "epoch": 0.2677754533380787, + "grad_norm": 0.306640625, + "learning_rate": 0.001596059193336582, + "loss": 0.127, + "step": 30848 + }, + { + "epoch": 0.2677841338182828, + "grad_norm": 0.37890625, + "learning_rate": 0.0015960343596159326, + "loss": 0.0908, + "step": 30849 + }, + { + "epoch": 0.267792814298487, + "grad_norm": 0.79296875, + "learning_rate": 0.0015960095253528268, + "loss": 0.1113, + "step": 30850 + }, + { + "epoch": 0.26780149477869114, + "grad_norm": 1.0234375, + "learning_rate": 0.0015959846905472927, + "loss": 0.1494, + "step": 30851 + }, + { + "epoch": 0.26781017525889533, + "grad_norm": 0.21484375, + "learning_rate": 0.0015959598551993578, + "loss": 0.1387, + "step": 30852 + }, + { + "epoch": 0.26781885573909947, + "grad_norm": 0.330078125, + "learning_rate": 0.001595935019309048, + "loss": 0.0708, + "step": 30853 + }, + { + "epoch": 0.26782753621930366, + "grad_norm": 0.59375, + "learning_rate": 0.0015959101828763916, + "loss": 0.1592, + "step": 30854 + }, + { + "epoch": 0.2678362166995078, + "grad_norm": 0.46484375, + "learning_rate": 0.001595885345901415, + "loss": 0.1338, + "step": 30855 + }, + { + "epoch": 0.267844897179712, + "grad_norm": 0.73046875, + "learning_rate": 0.0015958605083841455, + "loss": 0.1172, + "step": 30856 + }, + { + "epoch": 0.26785357765991613, + "grad_norm": 0.2236328125, + "learning_rate": 0.0015958356703246104, + "loss": 0.1006, + "step": 30857 + }, + { + "epoch": 0.2678622581401203, + "grad_norm": 0.62109375, + "learning_rate": 0.0015958108317228371, + "loss": 0.0845, + "step": 30858 + }, + { + "epoch": 0.26787093862032446, + "grad_norm": 0.62109375, + "learning_rate": 0.0015957859925788522, + "loss": 0.1348, + "step": 30859 + }, + { + "epoch": 0.26787961910052865, + "grad_norm": 0.443359375, + "learning_rate": 0.0015957611528926833, + "loss": 0.1089, + "step": 30860 + }, + { + "epoch": 0.2678882995807328, + "grad_norm": 1.015625, + "learning_rate": 0.0015957363126643577, + "loss": 0.1416, + "step": 30861 + }, + { + "epoch": 0.267896980060937, + "grad_norm": 0.81640625, + "learning_rate": 0.0015957114718939023, + "loss": 0.1475, + "step": 30862 + }, + { + "epoch": 0.2679056605411411, + "grad_norm": 0.26953125, + "learning_rate": 0.001595686630581344, + "loss": 0.1318, + "step": 30863 + }, + { + "epoch": 0.2679143410213453, + "grad_norm": 0.3515625, + "learning_rate": 0.0015956617887267106, + "loss": 0.1279, + "step": 30864 + }, + { + "epoch": 0.26792302150154945, + "grad_norm": 0.251953125, + "learning_rate": 0.0015956369463300283, + "loss": 0.0854, + "step": 30865 + }, + { + "epoch": 0.26793170198175364, + "grad_norm": 0.2451171875, + "learning_rate": 0.0015956121033913256, + "loss": 0.0903, + "step": 30866 + }, + { + "epoch": 0.2679403824619578, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015955872599106287, + "loss": 0.1133, + "step": 30867 + }, + { + "epoch": 0.267949062942162, + "grad_norm": 0.5078125, + "learning_rate": 0.001595562415887965, + "loss": 0.1016, + "step": 30868 + }, + { + "epoch": 0.2679577434223661, + "grad_norm": 0.5859375, + "learning_rate": 0.0015955375713233614, + "loss": 0.0825, + "step": 30869 + }, + { + "epoch": 0.2679664239025703, + "grad_norm": 0.109375, + "learning_rate": 0.0015955127262168457, + "loss": 0.1436, + "step": 30870 + }, + { + "epoch": 0.26797510438277444, + "grad_norm": 0.73046875, + "learning_rate": 0.0015954878805684446, + "loss": 0.1279, + "step": 30871 + }, + { + "epoch": 0.26798378486297864, + "grad_norm": 0.337890625, + "learning_rate": 0.0015954630343781855, + "loss": 0.0894, + "step": 30872 + }, + { + "epoch": 0.2679924653431828, + "grad_norm": 0.77734375, + "learning_rate": 0.0015954381876460955, + "loss": 0.0986, + "step": 30873 + }, + { + "epoch": 0.26800114582338697, + "grad_norm": 0.474609375, + "learning_rate": 0.0015954133403722013, + "loss": 0.1045, + "step": 30874 + }, + { + "epoch": 0.2680098263035911, + "grad_norm": 0.1416015625, + "learning_rate": 0.0015953884925565313, + "loss": 0.1201, + "step": 30875 + }, + { + "epoch": 0.2680185067837953, + "grad_norm": 0.1875, + "learning_rate": 0.001595363644199111, + "loss": 0.1338, + "step": 30876 + }, + { + "epoch": 0.26802718726399943, + "grad_norm": 0.48828125, + "learning_rate": 0.001595338795299969, + "loss": 0.0967, + "step": 30877 + }, + { + "epoch": 0.2680358677442036, + "grad_norm": 0.53125, + "learning_rate": 0.001595313945859132, + "loss": 0.1436, + "step": 30878 + }, + { + "epoch": 0.26804454822440776, + "grad_norm": 0.353515625, + "learning_rate": 0.001595289095876627, + "loss": 0.0977, + "step": 30879 + }, + { + "epoch": 0.26805322870461196, + "grad_norm": 0.86328125, + "learning_rate": 0.001595264245352481, + "loss": 0.1079, + "step": 30880 + }, + { + "epoch": 0.2680619091848161, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015952393942867216, + "loss": 0.0908, + "step": 30881 + }, + { + "epoch": 0.2680705896650203, + "grad_norm": 0.0810546875, + "learning_rate": 0.001595214542679376, + "loss": 0.0869, + "step": 30882 + }, + { + "epoch": 0.2680792701452244, + "grad_norm": 0.3125, + "learning_rate": 0.001595189690530471, + "loss": 0.0918, + "step": 30883 + }, + { + "epoch": 0.2680879506254286, + "grad_norm": 0.2578125, + "learning_rate": 0.0015951648378400346, + "loss": 0.0923, + "step": 30884 + }, + { + "epoch": 0.26809663110563275, + "grad_norm": 0.3046875, + "learning_rate": 0.0015951399846080927, + "loss": 0.0986, + "step": 30885 + }, + { + "epoch": 0.26810531158583695, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015951151308346734, + "loss": 0.1074, + "step": 30886 + }, + { + "epoch": 0.2681139920660411, + "grad_norm": 0.0751953125, + "learning_rate": 0.0015950902765198033, + "loss": 0.0859, + "step": 30887 + }, + { + "epoch": 0.2681226725462453, + "grad_norm": 0.83984375, + "learning_rate": 0.0015950654216635104, + "loss": 0.1436, + "step": 30888 + }, + { + "epoch": 0.2681313530264494, + "grad_norm": 0.515625, + "learning_rate": 0.0015950405662658214, + "loss": 0.124, + "step": 30889 + }, + { + "epoch": 0.2681400335066536, + "grad_norm": 0.10302734375, + "learning_rate": 0.001595015710326763, + "loss": 0.0879, + "step": 30890 + }, + { + "epoch": 0.26814871398685775, + "grad_norm": 0.10693359375, + "learning_rate": 0.0015949908538463633, + "loss": 0.0669, + "step": 30891 + }, + { + "epoch": 0.26815739446706194, + "grad_norm": 0.328125, + "learning_rate": 0.0015949659968246491, + "loss": 0.1074, + "step": 30892 + }, + { + "epoch": 0.2681660749472661, + "grad_norm": 0.32421875, + "learning_rate": 0.0015949411392616472, + "loss": 0.0938, + "step": 30893 + }, + { + "epoch": 0.26817475542747027, + "grad_norm": 0.314453125, + "learning_rate": 0.0015949162811573852, + "loss": 0.0796, + "step": 30894 + }, + { + "epoch": 0.2681834359076744, + "grad_norm": 0.392578125, + "learning_rate": 0.0015948914225118902, + "loss": 0.1021, + "step": 30895 + }, + { + "epoch": 0.2681921163878786, + "grad_norm": 0.345703125, + "learning_rate": 0.0015948665633251896, + "loss": 0.1172, + "step": 30896 + }, + { + "epoch": 0.26820079686808274, + "grad_norm": 0.3359375, + "learning_rate": 0.00159484170359731, + "loss": 0.1221, + "step": 30897 + }, + { + "epoch": 0.26820947734828693, + "grad_norm": 0.330078125, + "learning_rate": 0.001594816843328279, + "loss": 0.0991, + "step": 30898 + }, + { + "epoch": 0.26821815782849107, + "grad_norm": 0.0830078125, + "learning_rate": 0.0015947919825181239, + "loss": 0.1143, + "step": 30899 + }, + { + "epoch": 0.26822683830869526, + "grad_norm": 0.287109375, + "learning_rate": 0.0015947671211668716, + "loss": 0.0962, + "step": 30900 + }, + { + "epoch": 0.2682355187888994, + "grad_norm": 0.953125, + "learning_rate": 0.0015947422592745494, + "loss": 0.1396, + "step": 30901 + }, + { + "epoch": 0.2682441992691036, + "grad_norm": 0.09423828125, + "learning_rate": 0.0015947173968411847, + "loss": 0.0723, + "step": 30902 + }, + { + "epoch": 0.2682528797493077, + "grad_norm": 0.6328125, + "learning_rate": 0.001594692533866804, + "loss": 0.1035, + "step": 30903 + }, + { + "epoch": 0.2682615602295119, + "grad_norm": 0.13671875, + "learning_rate": 0.0015946676703514355, + "loss": 0.1162, + "step": 30904 + }, + { + "epoch": 0.26827024070971606, + "grad_norm": 0.306640625, + "learning_rate": 0.0015946428062951056, + "loss": 0.1064, + "step": 30905 + }, + { + "epoch": 0.26827892118992025, + "grad_norm": 0.47265625, + "learning_rate": 0.001594617941697842, + "loss": 0.1445, + "step": 30906 + }, + { + "epoch": 0.2682876016701244, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015945930765596714, + "loss": 0.1455, + "step": 30907 + }, + { + "epoch": 0.2682962821503286, + "grad_norm": 0.193359375, + "learning_rate": 0.0015945682108806213, + "loss": 0.0869, + "step": 30908 + }, + { + "epoch": 0.2683049626305327, + "grad_norm": 0.23828125, + "learning_rate": 0.001594543344660719, + "loss": 0.0898, + "step": 30909 + }, + { + "epoch": 0.2683136431107369, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015945184778999916, + "loss": 0.0967, + "step": 30910 + }, + { + "epoch": 0.26832232359094105, + "grad_norm": 0.12158203125, + "learning_rate": 0.0015944936105984659, + "loss": 0.0977, + "step": 30911 + }, + { + "epoch": 0.26833100407114524, + "grad_norm": 0.75390625, + "learning_rate": 0.0015944687427561692, + "loss": 0.1426, + "step": 30912 + }, + { + "epoch": 0.2683396845513494, + "grad_norm": 0.328125, + "learning_rate": 0.0015944438743731295, + "loss": 0.0723, + "step": 30913 + }, + { + "epoch": 0.26834836503155357, + "grad_norm": 0.341796875, + "learning_rate": 0.0015944190054493733, + "loss": 0.1406, + "step": 30914 + }, + { + "epoch": 0.2683570455117577, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015943941359849276, + "loss": 0.1416, + "step": 30915 + }, + { + "epoch": 0.2683657259919619, + "grad_norm": 0.287109375, + "learning_rate": 0.0015943692659798202, + "loss": 0.0918, + "step": 30916 + }, + { + "epoch": 0.26837440647216604, + "grad_norm": 0.65625, + "learning_rate": 0.001594344395434078, + "loss": 0.1396, + "step": 30917 + }, + { + "epoch": 0.26838308695237023, + "grad_norm": 0.19921875, + "learning_rate": 0.0015943195243477279, + "loss": 0.1089, + "step": 30918 + }, + { + "epoch": 0.26839176743257437, + "grad_norm": 0.326171875, + "learning_rate": 0.0015942946527207976, + "loss": 0.125, + "step": 30919 + }, + { + "epoch": 0.26840044791277856, + "grad_norm": 0.5078125, + "learning_rate": 0.001594269780553314, + "loss": 0.1138, + "step": 30920 + }, + { + "epoch": 0.2684091283929827, + "grad_norm": 0.326171875, + "learning_rate": 0.0015942449078453044, + "loss": 0.1367, + "step": 30921 + }, + { + "epoch": 0.2684178088731869, + "grad_norm": 0.234375, + "learning_rate": 0.001594220034596796, + "loss": 0.1099, + "step": 30922 + }, + { + "epoch": 0.26842648935339103, + "grad_norm": 0.08251953125, + "learning_rate": 0.001594195160807816, + "loss": 0.1113, + "step": 30923 + }, + { + "epoch": 0.2684351698335952, + "grad_norm": 0.09228515625, + "learning_rate": 0.0015941702864783918, + "loss": 0.103, + "step": 30924 + }, + { + "epoch": 0.26844385031379936, + "grad_norm": 0.353515625, + "learning_rate": 0.0015941454116085503, + "loss": 0.125, + "step": 30925 + }, + { + "epoch": 0.2684525307940035, + "grad_norm": 0.5, + "learning_rate": 0.0015941205361983187, + "loss": 0.1348, + "step": 30926 + }, + { + "epoch": 0.2684612112742077, + "grad_norm": 0.177734375, + "learning_rate": 0.0015940956602477241, + "loss": 0.0879, + "step": 30927 + }, + { + "epoch": 0.2684698917544118, + "grad_norm": 0.2353515625, + "learning_rate": 0.0015940707837567944, + "loss": 0.1021, + "step": 30928 + }, + { + "epoch": 0.268478572234616, + "grad_norm": 0.83984375, + "learning_rate": 0.0015940459067255558, + "loss": 0.0903, + "step": 30929 + }, + { + "epoch": 0.26848725271482016, + "grad_norm": 0.1572265625, + "learning_rate": 0.0015940210291540368, + "loss": 0.1406, + "step": 30930 + }, + { + "epoch": 0.26849593319502435, + "grad_norm": 0.47265625, + "learning_rate": 0.001593996151042263, + "loss": 0.0762, + "step": 30931 + }, + { + "epoch": 0.2685046136752285, + "grad_norm": 0.82421875, + "learning_rate": 0.0015939712723902627, + "loss": 0.083, + "step": 30932 + }, + { + "epoch": 0.2685132941554327, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015939463931980628, + "loss": 0.124, + "step": 30933 + }, + { + "epoch": 0.2685219746356368, + "grad_norm": 0.27734375, + "learning_rate": 0.0015939215134656909, + "loss": 0.1211, + "step": 30934 + }, + { + "epoch": 0.268530655115841, + "grad_norm": 0.640625, + "learning_rate": 0.0015938966331931733, + "loss": 0.0854, + "step": 30935 + }, + { + "epoch": 0.26853933559604515, + "grad_norm": 0.462890625, + "learning_rate": 0.0015938717523805385, + "loss": 0.1074, + "step": 30936 + }, + { + "epoch": 0.26854801607624934, + "grad_norm": 0.1982421875, + "learning_rate": 0.0015938468710278123, + "loss": 0.127, + "step": 30937 + }, + { + "epoch": 0.2685566965564535, + "grad_norm": 0.298828125, + "learning_rate": 0.0015938219891350226, + "loss": 0.0957, + "step": 30938 + }, + { + "epoch": 0.26856537703665767, + "grad_norm": 0.15625, + "learning_rate": 0.0015937971067021969, + "loss": 0.1504, + "step": 30939 + }, + { + "epoch": 0.2685740575168618, + "grad_norm": 0.373046875, + "learning_rate": 0.001593772223729362, + "loss": 0.103, + "step": 30940 + }, + { + "epoch": 0.268582737997066, + "grad_norm": 0.453125, + "learning_rate": 0.0015937473402165452, + "loss": 0.0928, + "step": 30941 + }, + { + "epoch": 0.26859141847727014, + "grad_norm": 0.16796875, + "learning_rate": 0.0015937224561637735, + "loss": 0.0996, + "step": 30942 + }, + { + "epoch": 0.26860009895747433, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015936975715710748, + "loss": 0.0898, + "step": 30943 + }, + { + "epoch": 0.26860877943767847, + "grad_norm": 0.279296875, + "learning_rate": 0.0015936726864384757, + "loss": 0.1328, + "step": 30944 + }, + { + "epoch": 0.26861745991788266, + "grad_norm": 0.296875, + "learning_rate": 0.0015936478007660033, + "loss": 0.1299, + "step": 30945 + }, + { + "epoch": 0.2686261403980868, + "grad_norm": 0.427734375, + "learning_rate": 0.0015936229145536854, + "loss": 0.125, + "step": 30946 + }, + { + "epoch": 0.268634820878291, + "grad_norm": 1.0, + "learning_rate": 0.0015935980278015487, + "loss": 0.0967, + "step": 30947 + }, + { + "epoch": 0.26864350135849513, + "grad_norm": 0.296875, + "learning_rate": 0.0015935731405096207, + "loss": 0.1211, + "step": 30948 + }, + { + "epoch": 0.2686521818386993, + "grad_norm": 0.10986328125, + "learning_rate": 0.001593548252677929, + "loss": 0.1377, + "step": 30949 + }, + { + "epoch": 0.26866086231890346, + "grad_norm": 0.365234375, + "learning_rate": 0.0015935233643064994, + "loss": 0.082, + "step": 30950 + }, + { + "epoch": 0.26866954279910765, + "grad_norm": 0.138671875, + "learning_rate": 0.0015934984753953605, + "loss": 0.1377, + "step": 30951 + }, + { + "epoch": 0.2686782232793118, + "grad_norm": 0.2578125, + "learning_rate": 0.001593473585944539, + "loss": 0.1445, + "step": 30952 + }, + { + "epoch": 0.268686903759516, + "grad_norm": 0.49609375, + "learning_rate": 0.0015934486959540623, + "loss": 0.0947, + "step": 30953 + }, + { + "epoch": 0.2686955842397201, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015934238054239578, + "loss": 0.125, + "step": 30954 + }, + { + "epoch": 0.2687042647199243, + "grad_norm": 0.28125, + "learning_rate": 0.001593398914354252, + "loss": 0.0879, + "step": 30955 + }, + { + "epoch": 0.26871294520012845, + "grad_norm": 0.158203125, + "learning_rate": 0.0015933740227449726, + "loss": 0.0728, + "step": 30956 + }, + { + "epoch": 0.26872162568033264, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015933491305961474, + "loss": 0.0649, + "step": 30957 + }, + { + "epoch": 0.2687303061605368, + "grad_norm": 0.6484375, + "learning_rate": 0.0015933242379078023, + "loss": 0.0898, + "step": 30958 + }, + { + "epoch": 0.268738986640741, + "grad_norm": 0.16796875, + "learning_rate": 0.0015932993446799652, + "loss": 0.1113, + "step": 30959 + }, + { + "epoch": 0.2687476671209451, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015932744509126638, + "loss": 0.1367, + "step": 30960 + }, + { + "epoch": 0.2687563476011493, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015932495566059247, + "loss": 0.1211, + "step": 30961 + }, + { + "epoch": 0.26876502808135344, + "grad_norm": 0.412109375, + "learning_rate": 0.0015932246617597753, + "loss": 0.1064, + "step": 30962 + }, + { + "epoch": 0.26877370856155763, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015931997663742427, + "loss": 0.084, + "step": 30963 + }, + { + "epoch": 0.26878238904176177, + "grad_norm": 1.6015625, + "learning_rate": 0.0015931748704493545, + "loss": 0.1279, + "step": 30964 + }, + { + "epoch": 0.26879106952196596, + "grad_norm": 0.11767578125, + "learning_rate": 0.0015931499739851374, + "loss": 0.0762, + "step": 30965 + }, + { + "epoch": 0.2687997500021701, + "grad_norm": 0.62109375, + "learning_rate": 0.001593125076981619, + "loss": 0.1504, + "step": 30966 + }, + { + "epoch": 0.2688084304823743, + "grad_norm": 0.126953125, + "learning_rate": 0.0015931001794388265, + "loss": 0.1055, + "step": 30967 + }, + { + "epoch": 0.26881711096257843, + "grad_norm": 0.2158203125, + "learning_rate": 0.0015930752813567868, + "loss": 0.0923, + "step": 30968 + }, + { + "epoch": 0.2688257914427826, + "grad_norm": 0.45703125, + "learning_rate": 0.0015930503827355277, + "loss": 0.1074, + "step": 30969 + }, + { + "epoch": 0.26883447192298676, + "grad_norm": 0.08251953125, + "learning_rate": 0.001593025483575076, + "loss": 0.0918, + "step": 30970 + }, + { + "epoch": 0.26884315240319095, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015930005838754588, + "loss": 0.0752, + "step": 30971 + }, + { + "epoch": 0.2688518328833951, + "grad_norm": 0.25, + "learning_rate": 0.0015929756836367038, + "loss": 0.1152, + "step": 30972 + }, + { + "epoch": 0.2688605133635993, + "grad_norm": 0.62109375, + "learning_rate": 0.0015929507828588382, + "loss": 0.1104, + "step": 30973 + }, + { + "epoch": 0.2688691938438034, + "grad_norm": 0.32421875, + "learning_rate": 0.0015929258815418886, + "loss": 0.1289, + "step": 30974 + }, + { + "epoch": 0.2688778743240076, + "grad_norm": 0.09228515625, + "learning_rate": 0.001592900979685883, + "loss": 0.0991, + "step": 30975 + }, + { + "epoch": 0.26888655480421175, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015928760772908482, + "loss": 0.166, + "step": 30976 + }, + { + "epoch": 0.26889523528441595, + "grad_norm": 0.36328125, + "learning_rate": 0.0015928511743568113, + "loss": 0.1641, + "step": 30977 + }, + { + "epoch": 0.2689039157646201, + "grad_norm": 0.62109375, + "learning_rate": 0.0015928262708837999, + "loss": 0.1396, + "step": 30978 + }, + { + "epoch": 0.2689125962448243, + "grad_norm": 0.11376953125, + "learning_rate": 0.001592801366871841, + "loss": 0.0952, + "step": 30979 + }, + { + "epoch": 0.2689212767250284, + "grad_norm": 0.2158203125, + "learning_rate": 0.0015927764623209625, + "loss": 0.1543, + "step": 30980 + }, + { + "epoch": 0.2689299572052326, + "grad_norm": 0.30078125, + "learning_rate": 0.0015927515572311903, + "loss": 0.1191, + "step": 30981 + }, + { + "epoch": 0.26893863768543674, + "grad_norm": 0.154296875, + "learning_rate": 0.001592726651602553, + "loss": 0.1182, + "step": 30982 + }, + { + "epoch": 0.26894731816564094, + "grad_norm": 0.0869140625, + "learning_rate": 0.001592701745435077, + "loss": 0.0879, + "step": 30983 + }, + { + "epoch": 0.2689559986458451, + "grad_norm": 0.193359375, + "learning_rate": 0.0015926768387287894, + "loss": 0.0898, + "step": 30984 + }, + { + "epoch": 0.26896467912604927, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015926519314837183, + "loss": 0.1006, + "step": 30985 + }, + { + "epoch": 0.2689733596062534, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015926270236998902, + "loss": 0.0977, + "step": 30986 + }, + { + "epoch": 0.2689820400864576, + "grad_norm": 0.5078125, + "learning_rate": 0.0015926021153773332, + "loss": 0.083, + "step": 30987 + }, + { + "epoch": 0.26899072056666173, + "grad_norm": 0.20703125, + "learning_rate": 0.001592577206516073, + "loss": 0.1094, + "step": 30988 + }, + { + "epoch": 0.2689994010468659, + "grad_norm": 0.3671875, + "learning_rate": 0.0015925522971161384, + "loss": 0.1094, + "step": 30989 + }, + { + "epoch": 0.26900808152707006, + "grad_norm": 0.29296875, + "learning_rate": 0.0015925273871775557, + "loss": 0.1084, + "step": 30990 + }, + { + "epoch": 0.26901676200727426, + "grad_norm": 0.326171875, + "learning_rate": 0.0015925024767003526, + "loss": 0.1172, + "step": 30991 + }, + { + "epoch": 0.2690254424874784, + "grad_norm": 0.11962890625, + "learning_rate": 0.001592477565684556, + "loss": 0.1504, + "step": 30992 + }, + { + "epoch": 0.2690341229676826, + "grad_norm": 0.9453125, + "learning_rate": 0.0015924526541301936, + "loss": 0.123, + "step": 30993 + }, + { + "epoch": 0.2690428034478867, + "grad_norm": 0.88671875, + "learning_rate": 0.001592427742037292, + "loss": 0.1357, + "step": 30994 + }, + { + "epoch": 0.2690514839280909, + "grad_norm": 0.373046875, + "learning_rate": 0.0015924028294058792, + "loss": 0.1348, + "step": 30995 + }, + { + "epoch": 0.26906016440829505, + "grad_norm": 0.12060546875, + "learning_rate": 0.0015923779162359821, + "loss": 0.1172, + "step": 30996 + }, + { + "epoch": 0.26906884488849925, + "grad_norm": 0.10107421875, + "learning_rate": 0.0015923530025276278, + "loss": 0.1182, + "step": 30997 + }, + { + "epoch": 0.2690775253687034, + "grad_norm": 0.251953125, + "learning_rate": 0.0015923280882808432, + "loss": 0.1758, + "step": 30998 + }, + { + "epoch": 0.2690862058489076, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015923031734956566, + "loss": 0.0952, + "step": 30999 + }, + { + "epoch": 0.2690948863291117, + "grad_norm": 0.197265625, + "learning_rate": 0.0015922782581720945, + "loss": 0.125, + "step": 31000 + }, + { + "epoch": 0.2691035668093159, + "grad_norm": 0.14453125, + "learning_rate": 0.0015922533423101844, + "loss": 0.0928, + "step": 31001 + }, + { + "epoch": 0.26911224728952005, + "grad_norm": 0.197265625, + "learning_rate": 0.0015922284259099531, + "loss": 0.1162, + "step": 31002 + }, + { + "epoch": 0.26912092776972424, + "grad_norm": 0.294921875, + "learning_rate": 0.0015922035089714285, + "loss": 0.1973, + "step": 31003 + }, + { + "epoch": 0.2691296082499284, + "grad_norm": 0.126953125, + "learning_rate": 0.0015921785914946372, + "loss": 0.104, + "step": 31004 + }, + { + "epoch": 0.26913828873013257, + "grad_norm": 0.578125, + "learning_rate": 0.0015921536734796072, + "loss": 0.1182, + "step": 31005 + }, + { + "epoch": 0.2691469692103367, + "grad_norm": 0.3515625, + "learning_rate": 0.0015921287549263653, + "loss": 0.0835, + "step": 31006 + }, + { + "epoch": 0.2691556496905409, + "grad_norm": 0.220703125, + "learning_rate": 0.0015921038358349382, + "loss": 0.127, + "step": 31007 + }, + { + "epoch": 0.26916433017074504, + "grad_norm": 0.177734375, + "learning_rate": 0.0015920789162053544, + "loss": 0.1211, + "step": 31008 + }, + { + "epoch": 0.26917301065094923, + "grad_norm": 0.53515625, + "learning_rate": 0.0015920539960376401, + "loss": 0.0845, + "step": 31009 + }, + { + "epoch": 0.26918169113115337, + "grad_norm": 0.392578125, + "learning_rate": 0.0015920290753318233, + "loss": 0.1094, + "step": 31010 + }, + { + "epoch": 0.26919037161135756, + "grad_norm": 0.296875, + "learning_rate": 0.0015920041540879305, + "loss": 0.1279, + "step": 31011 + }, + { + "epoch": 0.2691990520915617, + "grad_norm": 0.177734375, + "learning_rate": 0.0015919792323059896, + "loss": 0.0981, + "step": 31012 + }, + { + "epoch": 0.2692077325717659, + "grad_norm": 0.13671875, + "learning_rate": 0.0015919543099860275, + "loss": 0.085, + "step": 31013 + }, + { + "epoch": 0.26921641305197, + "grad_norm": 0.205078125, + "learning_rate": 0.0015919293871280717, + "loss": 0.1064, + "step": 31014 + }, + { + "epoch": 0.2692250935321742, + "grad_norm": 0.345703125, + "learning_rate": 0.0015919044637321491, + "loss": 0.124, + "step": 31015 + }, + { + "epoch": 0.26923377401237836, + "grad_norm": 0.197265625, + "learning_rate": 0.0015918795397982873, + "loss": 0.1123, + "step": 31016 + }, + { + "epoch": 0.26924245449258255, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015918546153265135, + "loss": 0.0986, + "step": 31017 + }, + { + "epoch": 0.2692511349727867, + "grad_norm": 0.34765625, + "learning_rate": 0.0015918296903168548, + "loss": 0.1172, + "step": 31018 + }, + { + "epoch": 0.2692598154529909, + "grad_norm": 0.62109375, + "learning_rate": 0.0015918047647693384, + "loss": 0.1221, + "step": 31019 + }, + { + "epoch": 0.269268495933195, + "grad_norm": 0.380859375, + "learning_rate": 0.0015917798386839918, + "loss": 0.1201, + "step": 31020 + }, + { + "epoch": 0.2692771764133992, + "grad_norm": 0.1044921875, + "learning_rate": 0.0015917549120608423, + "loss": 0.1069, + "step": 31021 + }, + { + "epoch": 0.26928585689360335, + "grad_norm": 0.38671875, + "learning_rate": 0.0015917299848999169, + "loss": 0.0898, + "step": 31022 + }, + { + "epoch": 0.26929453737380754, + "grad_norm": 0.1923828125, + "learning_rate": 0.001591705057201243, + "loss": 0.1572, + "step": 31023 + }, + { + "epoch": 0.2693032178540117, + "grad_norm": 0.16015625, + "learning_rate": 0.0015916801289648477, + "loss": 0.083, + "step": 31024 + }, + { + "epoch": 0.26931189833421587, + "grad_norm": 0.349609375, + "learning_rate": 0.0015916552001907585, + "loss": 0.0791, + "step": 31025 + }, + { + "epoch": 0.26932057881442, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015916302708790027, + "loss": 0.1143, + "step": 31026 + }, + { + "epoch": 0.2693292592946242, + "grad_norm": 0.455078125, + "learning_rate": 0.0015916053410296074, + "loss": 0.1387, + "step": 31027 + }, + { + "epoch": 0.26933793977482834, + "grad_norm": 0.08984375, + "learning_rate": 0.0015915804106425993, + "loss": 0.0898, + "step": 31028 + }, + { + "epoch": 0.26934662025503253, + "grad_norm": 0.275390625, + "learning_rate": 0.001591555479718007, + "loss": 0.0762, + "step": 31029 + }, + { + "epoch": 0.26935530073523667, + "grad_norm": 0.458984375, + "learning_rate": 0.0015915305482558568, + "loss": 0.125, + "step": 31030 + }, + { + "epoch": 0.26936398121544086, + "grad_norm": 0.412109375, + "learning_rate": 0.0015915056162561763, + "loss": 0.0801, + "step": 31031 + }, + { + "epoch": 0.269372661695645, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015914806837189922, + "loss": 0.125, + "step": 31032 + }, + { + "epoch": 0.2693813421758492, + "grad_norm": 0.318359375, + "learning_rate": 0.0015914557506443326, + "loss": 0.1729, + "step": 31033 + }, + { + "epoch": 0.26939002265605333, + "grad_norm": 0.263671875, + "learning_rate": 0.001591430817032224, + "loss": 0.1221, + "step": 31034 + }, + { + "epoch": 0.2693987031362575, + "grad_norm": 0.41015625, + "learning_rate": 0.0015914058828826944, + "loss": 0.1094, + "step": 31035 + }, + { + "epoch": 0.26940738361646166, + "grad_norm": 0.109375, + "learning_rate": 0.001591380948195771, + "loss": 0.1035, + "step": 31036 + }, + { + "epoch": 0.26941606409666585, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015913560129714802, + "loss": 0.1309, + "step": 31037 + }, + { + "epoch": 0.26942474457687, + "grad_norm": 0.33203125, + "learning_rate": 0.00159133107720985, + "loss": 0.0791, + "step": 31038 + }, + { + "epoch": 0.2694334250570742, + "grad_norm": 0.609375, + "learning_rate": 0.0015913061409109073, + "loss": 0.126, + "step": 31039 + }, + { + "epoch": 0.2694421055372783, + "grad_norm": 0.13671875, + "learning_rate": 0.00159128120407468, + "loss": 0.1016, + "step": 31040 + }, + { + "epoch": 0.2694507860174825, + "grad_norm": 0.396484375, + "learning_rate": 0.0015912562667011949, + "loss": 0.0938, + "step": 31041 + }, + { + "epoch": 0.26945946649768665, + "grad_norm": 0.1767578125, + "learning_rate": 0.001591231328790479, + "loss": 0.1211, + "step": 31042 + }, + { + "epoch": 0.26946814697789084, + "grad_norm": 0.1923828125, + "learning_rate": 0.0015912063903425599, + "loss": 0.1348, + "step": 31043 + }, + { + "epoch": 0.269476827458095, + "grad_norm": 0.146484375, + "learning_rate": 0.0015911814513574654, + "loss": 0.0918, + "step": 31044 + }, + { + "epoch": 0.2694855079382992, + "grad_norm": 0.189453125, + "learning_rate": 0.001591156511835222, + "loss": 0.1484, + "step": 31045 + }, + { + "epoch": 0.2694941884185033, + "grad_norm": 0.2109375, + "learning_rate": 0.0015911315717758573, + "loss": 0.1094, + "step": 31046 + }, + { + "epoch": 0.2695028688987075, + "grad_norm": 0.58984375, + "learning_rate": 0.001591106631179398, + "loss": 0.0928, + "step": 31047 + }, + { + "epoch": 0.26951154937891164, + "grad_norm": 0.25390625, + "learning_rate": 0.0015910816900458723, + "loss": 0.0781, + "step": 31048 + }, + { + "epoch": 0.2695202298591158, + "grad_norm": 0.10791015625, + "learning_rate": 0.0015910567483753068, + "loss": 0.0991, + "step": 31049 + }, + { + "epoch": 0.26952891033931997, + "grad_norm": 0.12353515625, + "learning_rate": 0.001591031806167729, + "loss": 0.0977, + "step": 31050 + }, + { + "epoch": 0.2695375908195241, + "grad_norm": 0.099609375, + "learning_rate": 0.0015910068634231668, + "loss": 0.0542, + "step": 31051 + }, + { + "epoch": 0.2695462712997283, + "grad_norm": 0.58203125, + "learning_rate": 0.0015909819201416465, + "loss": 0.1406, + "step": 31052 + }, + { + "epoch": 0.26955495177993244, + "grad_norm": 0.431640625, + "learning_rate": 0.0015909569763231956, + "loss": 0.0781, + "step": 31053 + }, + { + "epoch": 0.26956363226013663, + "grad_norm": 0.275390625, + "learning_rate": 0.0015909320319678413, + "loss": 0.1113, + "step": 31054 + }, + { + "epoch": 0.26957231274034077, + "grad_norm": 0.4609375, + "learning_rate": 0.0015909070870756114, + "loss": 0.1172, + "step": 31055 + }, + { + "epoch": 0.26958099322054496, + "grad_norm": 0.205078125, + "learning_rate": 0.0015908821416465329, + "loss": 0.0864, + "step": 31056 + }, + { + "epoch": 0.2695896737007491, + "grad_norm": 0.2353515625, + "learning_rate": 0.0015908571956806331, + "loss": 0.103, + "step": 31057 + }, + { + "epoch": 0.2695983541809533, + "grad_norm": 0.30078125, + "learning_rate": 0.0015908322491779393, + "loss": 0.0825, + "step": 31058 + }, + { + "epoch": 0.26960703466115743, + "grad_norm": 0.26171875, + "learning_rate": 0.0015908073021384786, + "loss": 0.1211, + "step": 31059 + }, + { + "epoch": 0.2696157151413616, + "grad_norm": 0.267578125, + "learning_rate": 0.0015907823545622786, + "loss": 0.0972, + "step": 31060 + }, + { + "epoch": 0.26962439562156576, + "grad_norm": 0.55078125, + "learning_rate": 0.001590757406449366, + "loss": 0.124, + "step": 31061 + }, + { + "epoch": 0.26963307610176995, + "grad_norm": 0.5, + "learning_rate": 0.001590732457799769, + "loss": 0.1011, + "step": 31062 + }, + { + "epoch": 0.2696417565819741, + "grad_norm": 0.09814453125, + "learning_rate": 0.001590707508613514, + "loss": 0.0903, + "step": 31063 + }, + { + "epoch": 0.2696504370621783, + "grad_norm": 0.46484375, + "learning_rate": 0.0015906825588906286, + "loss": 0.0986, + "step": 31064 + }, + { + "epoch": 0.2696591175423824, + "grad_norm": 0.12109375, + "learning_rate": 0.0015906576086311404, + "loss": 0.0684, + "step": 31065 + }, + { + "epoch": 0.2696677980225866, + "grad_norm": 0.0986328125, + "learning_rate": 0.0015906326578350764, + "loss": 0.0957, + "step": 31066 + }, + { + "epoch": 0.26967647850279075, + "grad_norm": 0.2734375, + "learning_rate": 0.0015906077065024633, + "loss": 0.1113, + "step": 31067 + }, + { + "epoch": 0.26968515898299494, + "grad_norm": 0.478515625, + "learning_rate": 0.0015905827546333297, + "loss": 0.126, + "step": 31068 + }, + { + "epoch": 0.2696938394631991, + "grad_norm": 0.431640625, + "learning_rate": 0.0015905578022277018, + "loss": 0.1123, + "step": 31069 + }, + { + "epoch": 0.2697025199434033, + "grad_norm": 0.494140625, + "learning_rate": 0.0015905328492856075, + "loss": 0.1289, + "step": 31070 + }, + { + "epoch": 0.2697112004236074, + "grad_norm": 0.546875, + "learning_rate": 0.0015905078958070738, + "loss": 0.0859, + "step": 31071 + }, + { + "epoch": 0.2697198809038116, + "grad_norm": 0.283203125, + "learning_rate": 0.0015904829417921278, + "loss": 0.1152, + "step": 31072 + }, + { + "epoch": 0.26972856138401574, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015904579872407971, + "loss": 0.1152, + "step": 31073 + }, + { + "epoch": 0.26973724186421993, + "grad_norm": 0.388671875, + "learning_rate": 0.001590433032153109, + "loss": 0.1328, + "step": 31074 + }, + { + "epoch": 0.26974592234442407, + "grad_norm": 0.1201171875, + "learning_rate": 0.001590408076529091, + "loss": 0.1328, + "step": 31075 + }, + { + "epoch": 0.26975460282462826, + "grad_norm": 0.43359375, + "learning_rate": 0.0015903831203687693, + "loss": 0.106, + "step": 31076 + }, + { + "epoch": 0.2697632833048324, + "grad_norm": 0.318359375, + "learning_rate": 0.0015903581636721724, + "loss": 0.1582, + "step": 31077 + }, + { + "epoch": 0.2697719637850366, + "grad_norm": 0.44921875, + "learning_rate": 0.0015903332064393274, + "loss": 0.1387, + "step": 31078 + }, + { + "epoch": 0.26978064426524073, + "grad_norm": 0.419921875, + "learning_rate": 0.0015903082486702614, + "loss": 0.127, + "step": 31079 + }, + { + "epoch": 0.2697893247454449, + "grad_norm": 1.515625, + "learning_rate": 0.0015902832903650012, + "loss": 0.166, + "step": 31080 + }, + { + "epoch": 0.26979800522564906, + "grad_norm": 0.265625, + "learning_rate": 0.001590258331523575, + "loss": 0.1523, + "step": 31081 + }, + { + "epoch": 0.26980668570585326, + "grad_norm": 0.427734375, + "learning_rate": 0.0015902333721460095, + "loss": 0.1445, + "step": 31082 + }, + { + "epoch": 0.2698153661860574, + "grad_norm": 0.126953125, + "learning_rate": 0.0015902084122323323, + "loss": 0.0815, + "step": 31083 + }, + { + "epoch": 0.2698240466662616, + "grad_norm": 0.59375, + "learning_rate": 0.0015901834517825701, + "loss": 0.105, + "step": 31084 + }, + { + "epoch": 0.2698327271464657, + "grad_norm": 0.228515625, + "learning_rate": 0.001590158490796751, + "loss": 0.0874, + "step": 31085 + }, + { + "epoch": 0.2698414076266699, + "grad_norm": 0.0859375, + "learning_rate": 0.001590133529274902, + "loss": 0.0981, + "step": 31086 + }, + { + "epoch": 0.26985008810687405, + "grad_norm": 0.50390625, + "learning_rate": 0.0015901085672170503, + "loss": 0.1514, + "step": 31087 + }, + { + "epoch": 0.26985876858707825, + "grad_norm": 0.5703125, + "learning_rate": 0.0015900836046232232, + "loss": 0.1816, + "step": 31088 + }, + { + "epoch": 0.2698674490672824, + "grad_norm": 0.267578125, + "learning_rate": 0.0015900586414934479, + "loss": 0.0918, + "step": 31089 + }, + { + "epoch": 0.2698761295474866, + "grad_norm": 0.57421875, + "learning_rate": 0.0015900336778277516, + "loss": 0.1025, + "step": 31090 + }, + { + "epoch": 0.2698848100276907, + "grad_norm": 0.275390625, + "learning_rate": 0.0015900087136261625, + "loss": 0.1079, + "step": 31091 + }, + { + "epoch": 0.2698934905078949, + "grad_norm": 0.78125, + "learning_rate": 0.0015899837488887068, + "loss": 0.1147, + "step": 31092 + }, + { + "epoch": 0.26990217098809904, + "grad_norm": 0.76171875, + "learning_rate": 0.0015899587836154122, + "loss": 0.0996, + "step": 31093 + }, + { + "epoch": 0.26991085146830324, + "grad_norm": 0.65234375, + "learning_rate": 0.0015899338178063062, + "loss": 0.0874, + "step": 31094 + }, + { + "epoch": 0.2699195319485074, + "grad_norm": 0.79296875, + "learning_rate": 0.001589908851461416, + "loss": 0.103, + "step": 31095 + }, + { + "epoch": 0.26992821242871157, + "grad_norm": 0.2001953125, + "learning_rate": 0.0015898838845807685, + "loss": 0.0947, + "step": 31096 + }, + { + "epoch": 0.2699368929089157, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015898589171643915, + "loss": 0.0815, + "step": 31097 + }, + { + "epoch": 0.2699455733891199, + "grad_norm": 0.1982421875, + "learning_rate": 0.0015898339492123124, + "loss": 0.0977, + "step": 31098 + }, + { + "epoch": 0.26995425386932403, + "grad_norm": 0.546875, + "learning_rate": 0.001589808980724558, + "loss": 0.1094, + "step": 31099 + }, + { + "epoch": 0.2699629343495282, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015897840117011562, + "loss": 0.1245, + "step": 31100 + }, + { + "epoch": 0.26997161482973236, + "grad_norm": 0.53125, + "learning_rate": 0.0015897590421421338, + "loss": 0.0996, + "step": 31101 + }, + { + "epoch": 0.26998029530993656, + "grad_norm": 0.51171875, + "learning_rate": 0.0015897340720475177, + "loss": 0.1133, + "step": 31102 + }, + { + "epoch": 0.2699889757901407, + "grad_norm": 0.271484375, + "learning_rate": 0.0015897091014173367, + "loss": 0.1016, + "step": 31103 + }, + { + "epoch": 0.2699976562703449, + "grad_norm": 0.5078125, + "learning_rate": 0.0015896841302516164, + "loss": 0.0801, + "step": 31104 + }, + { + "epoch": 0.270006336750549, + "grad_norm": 0.7734375, + "learning_rate": 0.0015896591585503853, + "loss": 0.127, + "step": 31105 + }, + { + "epoch": 0.2700150172307532, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015896341863136701, + "loss": 0.0859, + "step": 31106 + }, + { + "epoch": 0.27002369771095736, + "grad_norm": 0.357421875, + "learning_rate": 0.0015896092135414986, + "loss": 0.126, + "step": 31107 + }, + { + "epoch": 0.27003237819116155, + "grad_norm": 1.4453125, + "learning_rate": 0.0015895842402338977, + "loss": 0.2021, + "step": 31108 + }, + { + "epoch": 0.2700410586713657, + "grad_norm": 0.326171875, + "learning_rate": 0.0015895592663908947, + "loss": 0.1162, + "step": 31109 + }, + { + "epoch": 0.2700497391515699, + "grad_norm": 0.0732421875, + "learning_rate": 0.0015895342920125171, + "loss": 0.0737, + "step": 31110 + }, + { + "epoch": 0.270058419631774, + "grad_norm": 0.416015625, + "learning_rate": 0.001589509317098792, + "loss": 0.0762, + "step": 31111 + }, + { + "epoch": 0.2700671001119782, + "grad_norm": 0.6015625, + "learning_rate": 0.001589484341649747, + "loss": 0.1299, + "step": 31112 + }, + { + "epoch": 0.27007578059218235, + "grad_norm": 0.193359375, + "learning_rate": 0.0015894593656654095, + "loss": 0.0781, + "step": 31113 + }, + { + "epoch": 0.27008446107238654, + "grad_norm": 0.1015625, + "learning_rate": 0.0015894343891458065, + "loss": 0.1309, + "step": 31114 + }, + { + "epoch": 0.2700931415525907, + "grad_norm": 0.6171875, + "learning_rate": 0.0015894094120909653, + "loss": 0.1602, + "step": 31115 + }, + { + "epoch": 0.27010182203279487, + "grad_norm": 0.201171875, + "learning_rate": 0.0015893844345009132, + "loss": 0.0869, + "step": 31116 + }, + { + "epoch": 0.270110502512999, + "grad_norm": 0.115234375, + "learning_rate": 0.001589359456375678, + "loss": 0.1104, + "step": 31117 + }, + { + "epoch": 0.2701191829932032, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015893344777152866, + "loss": 0.1069, + "step": 31118 + }, + { + "epoch": 0.27012786347340734, + "grad_norm": 0.166015625, + "learning_rate": 0.0015893094985197657, + "loss": 0.0864, + "step": 31119 + }, + { + "epoch": 0.27013654395361153, + "grad_norm": 0.08837890625, + "learning_rate": 0.001589284518789144, + "loss": 0.1182, + "step": 31120 + }, + { + "epoch": 0.27014522443381567, + "grad_norm": 0.384765625, + "learning_rate": 0.0015892595385234478, + "loss": 0.2119, + "step": 31121 + }, + { + "epoch": 0.27015390491401986, + "grad_norm": 0.080078125, + "learning_rate": 0.001589234557722705, + "loss": 0.0811, + "step": 31122 + }, + { + "epoch": 0.270162585394224, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015892095763869424, + "loss": 0.1182, + "step": 31123 + }, + { + "epoch": 0.2701712658744282, + "grad_norm": 0.2158203125, + "learning_rate": 0.0015891845945161874, + "loss": 0.0957, + "step": 31124 + }, + { + "epoch": 0.2701799463546323, + "grad_norm": 0.302734375, + "learning_rate": 0.001589159612110468, + "loss": 0.1016, + "step": 31125 + }, + { + "epoch": 0.2701886268348365, + "grad_norm": 0.73828125, + "learning_rate": 0.0015891346291698105, + "loss": 0.1367, + "step": 31126 + }, + { + "epoch": 0.27019730731504066, + "grad_norm": 0.43359375, + "learning_rate": 0.0015891096456942429, + "loss": 0.1074, + "step": 31127 + }, + { + "epoch": 0.27020598779524485, + "grad_norm": 0.2373046875, + "learning_rate": 0.0015890846616837921, + "loss": 0.1289, + "step": 31128 + }, + { + "epoch": 0.270214668275449, + "grad_norm": 0.2451171875, + "learning_rate": 0.001589059677138486, + "loss": 0.0986, + "step": 31129 + }, + { + "epoch": 0.2702233487556532, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015890346920583516, + "loss": 0.0928, + "step": 31130 + }, + { + "epoch": 0.2702320292358573, + "grad_norm": 0.72265625, + "learning_rate": 0.0015890097064434159, + "loss": 0.1406, + "step": 31131 + }, + { + "epoch": 0.2702407097160615, + "grad_norm": 1.359375, + "learning_rate": 0.0015889847202937067, + "loss": 0.1914, + "step": 31132 + }, + { + "epoch": 0.27024939019626565, + "grad_norm": 0.400390625, + "learning_rate": 0.001588959733609251, + "loss": 0.0928, + "step": 31133 + }, + { + "epoch": 0.27025807067646984, + "grad_norm": 0.23828125, + "learning_rate": 0.0015889347463900767, + "loss": 0.1084, + "step": 31134 + }, + { + "epoch": 0.270266751156674, + "grad_norm": 0.515625, + "learning_rate": 0.0015889097586362106, + "loss": 0.0767, + "step": 31135 + }, + { + "epoch": 0.27027543163687817, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015888847703476798, + "loss": 0.1182, + "step": 31136 + }, + { + "epoch": 0.2702841121170823, + "grad_norm": 0.263671875, + "learning_rate": 0.0015888597815245122, + "loss": 0.1309, + "step": 31137 + }, + { + "epoch": 0.2702927925972865, + "grad_norm": 0.53515625, + "learning_rate": 0.0015888347921667347, + "loss": 0.1172, + "step": 31138 + }, + { + "epoch": 0.27030147307749064, + "grad_norm": 0.1611328125, + "learning_rate": 0.001588809802274375, + "loss": 0.1221, + "step": 31139 + }, + { + "epoch": 0.27031015355769483, + "grad_norm": 0.337890625, + "learning_rate": 0.0015887848118474603, + "loss": 0.1357, + "step": 31140 + }, + { + "epoch": 0.27031883403789897, + "grad_norm": 0.353515625, + "learning_rate": 0.0015887598208860177, + "loss": 0.1152, + "step": 31141 + }, + { + "epoch": 0.27032751451810316, + "grad_norm": 0.734375, + "learning_rate": 0.001588734829390075, + "loss": 0.1318, + "step": 31142 + }, + { + "epoch": 0.2703361949983073, + "grad_norm": 0.296875, + "learning_rate": 0.001588709837359659, + "loss": 0.1377, + "step": 31143 + }, + { + "epoch": 0.2703448754785115, + "grad_norm": 0.1923828125, + "learning_rate": 0.0015886848447947974, + "loss": 0.1211, + "step": 31144 + }, + { + "epoch": 0.27035355595871563, + "grad_norm": 0.392578125, + "learning_rate": 0.0015886598516955174, + "loss": 0.0986, + "step": 31145 + }, + { + "epoch": 0.2703622364389198, + "grad_norm": 0.455078125, + "learning_rate": 0.001588634858061846, + "loss": 0.0781, + "step": 31146 + }, + { + "epoch": 0.27037091691912396, + "grad_norm": 0.5234375, + "learning_rate": 0.0015886098638938108, + "loss": 0.1631, + "step": 31147 + }, + { + "epoch": 0.27037959739932815, + "grad_norm": 0.296875, + "learning_rate": 0.0015885848691914398, + "loss": 0.104, + "step": 31148 + }, + { + "epoch": 0.2703882778795323, + "grad_norm": 0.59765625, + "learning_rate": 0.001588559873954759, + "loss": 0.1245, + "step": 31149 + }, + { + "epoch": 0.2703969583597365, + "grad_norm": 0.177734375, + "learning_rate": 0.0015885348781837971, + "loss": 0.1357, + "step": 31150 + }, + { + "epoch": 0.2704056388399406, + "grad_norm": 0.263671875, + "learning_rate": 0.0015885098818785807, + "loss": 0.124, + "step": 31151 + }, + { + "epoch": 0.2704143193201448, + "grad_norm": 0.357421875, + "learning_rate": 0.0015884848850391368, + "loss": 0.0957, + "step": 31152 + }, + { + "epoch": 0.27042299980034895, + "grad_norm": 0.2890625, + "learning_rate": 0.0015884598876654936, + "loss": 0.0801, + "step": 31153 + }, + { + "epoch": 0.27043168028055314, + "grad_norm": 0.298828125, + "learning_rate": 0.0015884348897576776, + "loss": 0.1064, + "step": 31154 + }, + { + "epoch": 0.2704403607607573, + "grad_norm": 0.6796875, + "learning_rate": 0.001588409891315717, + "loss": 0.0942, + "step": 31155 + }, + { + "epoch": 0.2704490412409615, + "grad_norm": 0.466796875, + "learning_rate": 0.0015883848923396386, + "loss": 0.1113, + "step": 31156 + }, + { + "epoch": 0.2704577217211656, + "grad_norm": 0.10400390625, + "learning_rate": 0.0015883598928294696, + "loss": 0.1436, + "step": 31157 + }, + { + "epoch": 0.2704664022013698, + "grad_norm": 0.546875, + "learning_rate": 0.0015883348927852375, + "loss": 0.085, + "step": 31158 + }, + { + "epoch": 0.27047508268157394, + "grad_norm": 0.298828125, + "learning_rate": 0.00158830989220697, + "loss": 0.1445, + "step": 31159 + }, + { + "epoch": 0.27048376316177813, + "grad_norm": 0.224609375, + "learning_rate": 0.001588284891094694, + "loss": 0.1108, + "step": 31160 + }, + { + "epoch": 0.27049244364198227, + "grad_norm": 0.2138671875, + "learning_rate": 0.001588259889448437, + "loss": 0.083, + "step": 31161 + }, + { + "epoch": 0.27050112412218646, + "grad_norm": 0.41015625, + "learning_rate": 0.0015882348872682263, + "loss": 0.1211, + "step": 31162 + }, + { + "epoch": 0.2705098046023906, + "grad_norm": 0.2109375, + "learning_rate": 0.0015882098845540894, + "loss": 0.084, + "step": 31163 + }, + { + "epoch": 0.2705184850825948, + "grad_norm": 0.2314453125, + "learning_rate": 0.0015881848813060534, + "loss": 0.0786, + "step": 31164 + }, + { + "epoch": 0.27052716556279893, + "grad_norm": 0.0830078125, + "learning_rate": 0.0015881598775241454, + "loss": 0.0791, + "step": 31165 + }, + { + "epoch": 0.2705358460430031, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015881348732083936, + "loss": 0.0947, + "step": 31166 + }, + { + "epoch": 0.27054452652320726, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015881098683588244, + "loss": 0.1025, + "step": 31167 + }, + { + "epoch": 0.27055320700341146, + "grad_norm": 0.287109375, + "learning_rate": 0.001588084862975466, + "loss": 0.1152, + "step": 31168 + }, + { + "epoch": 0.2705618874836156, + "grad_norm": 0.376953125, + "learning_rate": 0.0015880598570583453, + "loss": 0.0679, + "step": 31169 + }, + { + "epoch": 0.2705705679638198, + "grad_norm": 0.54296875, + "learning_rate": 0.0015880348506074895, + "loss": 0.1113, + "step": 31170 + }, + { + "epoch": 0.2705792484440239, + "grad_norm": 0.326171875, + "learning_rate": 0.001588009843622926, + "loss": 0.1143, + "step": 31171 + }, + { + "epoch": 0.2705879289242281, + "grad_norm": 0.154296875, + "learning_rate": 0.0015879848361046824, + "loss": 0.1016, + "step": 31172 + }, + { + "epoch": 0.27059660940443225, + "grad_norm": 0.068359375, + "learning_rate": 0.001587959828052786, + "loss": 0.0737, + "step": 31173 + }, + { + "epoch": 0.2706052898846364, + "grad_norm": 0.162109375, + "learning_rate": 0.001587934819467264, + "loss": 0.0913, + "step": 31174 + }, + { + "epoch": 0.2706139703648406, + "grad_norm": 0.12109375, + "learning_rate": 0.0015879098103481438, + "loss": 0.0923, + "step": 31175 + }, + { + "epoch": 0.2706226508450447, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015878848006954528, + "loss": 0.1309, + "step": 31176 + }, + { + "epoch": 0.2706313313252489, + "grad_norm": 0.1025390625, + "learning_rate": 0.0015878597905092183, + "loss": 0.0806, + "step": 31177 + }, + { + "epoch": 0.27064001180545305, + "grad_norm": 0.126953125, + "learning_rate": 0.0015878347797894676, + "loss": 0.1143, + "step": 31178 + }, + { + "epoch": 0.27064869228565724, + "grad_norm": 0.306640625, + "learning_rate": 0.0015878097685362283, + "loss": 0.125, + "step": 31179 + }, + { + "epoch": 0.2706573727658614, + "grad_norm": 0.12060546875, + "learning_rate": 0.0015877847567495273, + "loss": 0.0967, + "step": 31180 + }, + { + "epoch": 0.2706660532460656, + "grad_norm": 0.16796875, + "learning_rate": 0.0015877597444293923, + "loss": 0.1138, + "step": 31181 + }, + { + "epoch": 0.2706747337262697, + "grad_norm": 0.08349609375, + "learning_rate": 0.0015877347315758508, + "loss": 0.0967, + "step": 31182 + }, + { + "epoch": 0.2706834142064739, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015877097181889294, + "loss": 0.0923, + "step": 31183 + }, + { + "epoch": 0.27069209468667804, + "grad_norm": 0.9140625, + "learning_rate": 0.0015876847042686564, + "loss": 0.1338, + "step": 31184 + }, + { + "epoch": 0.27070077516688223, + "grad_norm": 0.38671875, + "learning_rate": 0.0015876596898150588, + "loss": 0.0884, + "step": 31185 + }, + { + "epoch": 0.27070945564708637, + "grad_norm": 0.484375, + "learning_rate": 0.0015876346748281637, + "loss": 0.1396, + "step": 31186 + }, + { + "epoch": 0.27071813612729057, + "grad_norm": 0.10400390625, + "learning_rate": 0.0015876096593079986, + "loss": 0.1416, + "step": 31187 + }, + { + "epoch": 0.2707268166074947, + "grad_norm": 0.6484375, + "learning_rate": 0.001587584643254591, + "loss": 0.0938, + "step": 31188 + }, + { + "epoch": 0.2707354970876989, + "grad_norm": 0.404296875, + "learning_rate": 0.001587559626667968, + "loss": 0.0918, + "step": 31189 + }, + { + "epoch": 0.27074417756790303, + "grad_norm": 0.515625, + "learning_rate": 0.0015875346095481579, + "loss": 0.1011, + "step": 31190 + }, + { + "epoch": 0.2707528580481072, + "grad_norm": 0.31640625, + "learning_rate": 0.0015875095918951865, + "loss": 0.0913, + "step": 31191 + }, + { + "epoch": 0.27076153852831136, + "grad_norm": 0.54296875, + "learning_rate": 0.0015874845737090817, + "loss": 0.0811, + "step": 31192 + }, + { + "epoch": 0.27077021900851556, + "grad_norm": 0.33984375, + "learning_rate": 0.0015874595549898716, + "loss": 0.0884, + "step": 31193 + }, + { + "epoch": 0.2707788994887197, + "grad_norm": 0.51171875, + "learning_rate": 0.001587434535737583, + "loss": 0.1309, + "step": 31194 + }, + { + "epoch": 0.2707875799689239, + "grad_norm": 0.439453125, + "learning_rate": 0.0015874095159522433, + "loss": 0.0781, + "step": 31195 + }, + { + "epoch": 0.270796260449128, + "grad_norm": 0.4375, + "learning_rate": 0.00158738449563388, + "loss": 0.1436, + "step": 31196 + }, + { + "epoch": 0.2708049409293322, + "grad_norm": 0.400390625, + "learning_rate": 0.00158735947478252, + "loss": 0.1143, + "step": 31197 + }, + { + "epoch": 0.27081362140953635, + "grad_norm": 0.470703125, + "learning_rate": 0.001587334453398191, + "loss": 0.1289, + "step": 31198 + }, + { + "epoch": 0.27082230188974055, + "grad_norm": 0.21484375, + "learning_rate": 0.001587309431480921, + "loss": 0.1055, + "step": 31199 + }, + { + "epoch": 0.2708309823699447, + "grad_norm": 0.79296875, + "learning_rate": 0.001587284409030736, + "loss": 0.1211, + "step": 31200 + }, + { + "epoch": 0.2708396628501489, + "grad_norm": 0.345703125, + "learning_rate": 0.001587259386047664, + "loss": 0.1035, + "step": 31201 + }, + { + "epoch": 0.270848343330353, + "grad_norm": 0.1328125, + "learning_rate": 0.001587234362531733, + "loss": 0.123, + "step": 31202 + }, + { + "epoch": 0.2708570238105572, + "grad_norm": 0.298828125, + "learning_rate": 0.0015872093384829695, + "loss": 0.1025, + "step": 31203 + }, + { + "epoch": 0.27086570429076134, + "grad_norm": 0.12890625, + "learning_rate": 0.0015871843139014016, + "loss": 0.0894, + "step": 31204 + }, + { + "epoch": 0.27087438477096554, + "grad_norm": 0.9140625, + "learning_rate": 0.0015871592887870558, + "loss": 0.1709, + "step": 31205 + }, + { + "epoch": 0.2708830652511697, + "grad_norm": 0.333984375, + "learning_rate": 0.0015871342631399602, + "loss": 0.084, + "step": 31206 + }, + { + "epoch": 0.27089174573137387, + "grad_norm": 0.115234375, + "learning_rate": 0.0015871092369601416, + "loss": 0.083, + "step": 31207 + }, + { + "epoch": 0.270900426211578, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015870842102476282, + "loss": 0.0977, + "step": 31208 + }, + { + "epoch": 0.2709091066917822, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015870591830024461, + "loss": 0.0996, + "step": 31209 + }, + { + "epoch": 0.27091778717198634, + "grad_norm": 0.5703125, + "learning_rate": 0.0015870341552246237, + "loss": 0.0957, + "step": 31210 + }, + { + "epoch": 0.27092646765219053, + "grad_norm": 0.228515625, + "learning_rate": 0.0015870091269141883, + "loss": 0.1445, + "step": 31211 + }, + { + "epoch": 0.27093514813239467, + "grad_norm": 0.126953125, + "learning_rate": 0.001586984098071167, + "loss": 0.1201, + "step": 31212 + }, + { + "epoch": 0.27094382861259886, + "grad_norm": 0.57421875, + "learning_rate": 0.0015869590686955872, + "loss": 0.1221, + "step": 31213 + }, + { + "epoch": 0.270952509092803, + "grad_norm": 0.1171875, + "learning_rate": 0.0015869340387874757, + "loss": 0.1064, + "step": 31214 + }, + { + "epoch": 0.2709611895730072, + "grad_norm": 0.19140625, + "learning_rate": 0.0015869090083468608, + "loss": 0.1074, + "step": 31215 + }, + { + "epoch": 0.2709698700532113, + "grad_norm": 0.232421875, + "learning_rate": 0.0015868839773737697, + "loss": 0.1104, + "step": 31216 + }, + { + "epoch": 0.2709785505334155, + "grad_norm": 0.177734375, + "learning_rate": 0.00158685894586823, + "loss": 0.1123, + "step": 31217 + }, + { + "epoch": 0.27098723101361966, + "grad_norm": 0.388671875, + "learning_rate": 0.001586833913830268, + "loss": 0.1099, + "step": 31218 + }, + { + "epoch": 0.27099591149382385, + "grad_norm": 0.09716796875, + "learning_rate": 0.001586808881259912, + "loss": 0.1074, + "step": 31219 + }, + { + "epoch": 0.271004591974028, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015867838481571888, + "loss": 0.1025, + "step": 31220 + }, + { + "epoch": 0.2710132724542322, + "grad_norm": 0.404296875, + "learning_rate": 0.0015867588145221264, + "loss": 0.1426, + "step": 31221 + }, + { + "epoch": 0.2710219529344363, + "grad_norm": 0.57421875, + "learning_rate": 0.001586733780354752, + "loss": 0.1182, + "step": 31222 + }, + { + "epoch": 0.2710306334146405, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015867087456550926, + "loss": 0.0845, + "step": 31223 + }, + { + "epoch": 0.27103931389484465, + "grad_norm": 0.69921875, + "learning_rate": 0.0015866837104231758, + "loss": 0.0977, + "step": 31224 + }, + { + "epoch": 0.27104799437504884, + "grad_norm": 0.08984375, + "learning_rate": 0.0015866586746590293, + "loss": 0.1094, + "step": 31225 + }, + { + "epoch": 0.271056674855253, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015866336383626798, + "loss": 0.0869, + "step": 31226 + }, + { + "epoch": 0.27106535533545717, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015866086015341554, + "loss": 0.1035, + "step": 31227 + }, + { + "epoch": 0.2710740358156613, + "grad_norm": 0.62890625, + "learning_rate": 0.001586583564173483, + "loss": 0.1089, + "step": 31228 + }, + { + "epoch": 0.2710827162958655, + "grad_norm": 0.1201171875, + "learning_rate": 0.00158655852628069, + "loss": 0.0913, + "step": 31229 + }, + { + "epoch": 0.27109139677606964, + "grad_norm": 0.36328125, + "learning_rate": 0.0015865334878558041, + "loss": 0.1152, + "step": 31230 + }, + { + "epoch": 0.27110007725627383, + "grad_norm": 0.08447265625, + "learning_rate": 0.0015865084488988525, + "loss": 0.0903, + "step": 31231 + }, + { + "epoch": 0.27110875773647797, + "grad_norm": 0.2333984375, + "learning_rate": 0.0015864834094098624, + "loss": 0.1201, + "step": 31232 + }, + { + "epoch": 0.27111743821668216, + "grad_norm": 0.1328125, + "learning_rate": 0.0015864583693888615, + "loss": 0.0771, + "step": 31233 + }, + { + "epoch": 0.2711261186968863, + "grad_norm": 0.43359375, + "learning_rate": 0.001586433328835877, + "loss": 0.1543, + "step": 31234 + }, + { + "epoch": 0.2711347991770905, + "grad_norm": 0.80859375, + "learning_rate": 0.0015864082877509364, + "loss": 0.1133, + "step": 31235 + }, + { + "epoch": 0.27114347965729463, + "grad_norm": 0.58984375, + "learning_rate": 0.0015863832461340667, + "loss": 0.1094, + "step": 31236 + }, + { + "epoch": 0.2711521601374988, + "grad_norm": 0.08447265625, + "learning_rate": 0.001586358203985296, + "loss": 0.0835, + "step": 31237 + }, + { + "epoch": 0.27116084061770296, + "grad_norm": 0.10546875, + "learning_rate": 0.001586333161304651, + "loss": 0.1187, + "step": 31238 + }, + { + "epoch": 0.27116952109790715, + "grad_norm": 0.703125, + "learning_rate": 0.0015863081180921596, + "loss": 0.1562, + "step": 31239 + }, + { + "epoch": 0.2711782015781113, + "grad_norm": 0.30078125, + "learning_rate": 0.0015862830743478486, + "loss": 0.1025, + "step": 31240 + }, + { + "epoch": 0.2711868820583155, + "grad_norm": 0.421875, + "learning_rate": 0.001586258030071746, + "loss": 0.1143, + "step": 31241 + }, + { + "epoch": 0.2711955625385196, + "grad_norm": 0.296875, + "learning_rate": 0.0015862329852638789, + "loss": 0.0869, + "step": 31242 + }, + { + "epoch": 0.2712042430187238, + "grad_norm": 1.0859375, + "learning_rate": 0.0015862079399242745, + "loss": 0.125, + "step": 31243 + }, + { + "epoch": 0.27121292349892795, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015861828940529608, + "loss": 0.1719, + "step": 31244 + }, + { + "epoch": 0.27122160397913214, + "grad_norm": 0.125, + "learning_rate": 0.0015861578476499645, + "loss": 0.0957, + "step": 31245 + }, + { + "epoch": 0.2712302844593363, + "grad_norm": 0.2197265625, + "learning_rate": 0.0015861328007153131, + "loss": 0.1162, + "step": 31246 + }, + { + "epoch": 0.2712389649395405, + "grad_norm": 0.126953125, + "learning_rate": 0.0015861077532490348, + "loss": 0.1216, + "step": 31247 + }, + { + "epoch": 0.2712476454197446, + "grad_norm": 0.173828125, + "learning_rate": 0.001586082705251156, + "loss": 0.1328, + "step": 31248 + }, + { + "epoch": 0.2712563258999488, + "grad_norm": 0.140625, + "learning_rate": 0.0015860576567217042, + "loss": 0.1348, + "step": 31249 + }, + { + "epoch": 0.27126500638015294, + "grad_norm": 0.23046875, + "learning_rate": 0.0015860326076607071, + "loss": 0.1289, + "step": 31250 + }, + { + "epoch": 0.27127368686035713, + "grad_norm": 0.265625, + "learning_rate": 0.0015860075580681925, + "loss": 0.0884, + "step": 31251 + }, + { + "epoch": 0.27128236734056127, + "grad_norm": 0.55859375, + "learning_rate": 0.001585982507944187, + "loss": 0.0898, + "step": 31252 + }, + { + "epoch": 0.27129104782076546, + "grad_norm": 0.1845703125, + "learning_rate": 0.0015859574572887182, + "loss": 0.1514, + "step": 31253 + }, + { + "epoch": 0.2712997283009696, + "grad_norm": 0.1640625, + "learning_rate": 0.001585932406101814, + "loss": 0.1011, + "step": 31254 + }, + { + "epoch": 0.2713084087811738, + "grad_norm": 0.384765625, + "learning_rate": 0.0015859073543835012, + "loss": 0.1201, + "step": 31255 + }, + { + "epoch": 0.27131708926137793, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015858823021338076, + "loss": 0.0908, + "step": 31256 + }, + { + "epoch": 0.2713257697415821, + "grad_norm": 0.314453125, + "learning_rate": 0.0015858572493527598, + "loss": 0.1328, + "step": 31257 + }, + { + "epoch": 0.27133445022178626, + "grad_norm": 0.3203125, + "learning_rate": 0.0015858321960403865, + "loss": 0.0874, + "step": 31258 + }, + { + "epoch": 0.27134313070199045, + "grad_norm": 0.123046875, + "learning_rate": 0.0015858071421967143, + "loss": 0.0898, + "step": 31259 + }, + { + "epoch": 0.2713518111821946, + "grad_norm": 0.5546875, + "learning_rate": 0.001585782087821771, + "loss": 0.1387, + "step": 31260 + }, + { + "epoch": 0.2713604916623988, + "grad_norm": 0.388671875, + "learning_rate": 0.001585757032915583, + "loss": 0.125, + "step": 31261 + }, + { + "epoch": 0.2713691721426029, + "grad_norm": 0.138671875, + "learning_rate": 0.0015857319774781786, + "loss": 0.0918, + "step": 31262 + }, + { + "epoch": 0.2713778526228071, + "grad_norm": 0.47265625, + "learning_rate": 0.001585706921509585, + "loss": 0.1089, + "step": 31263 + }, + { + "epoch": 0.27138653310301125, + "grad_norm": 0.37109375, + "learning_rate": 0.00158568186500983, + "loss": 0.0928, + "step": 31264 + }, + { + "epoch": 0.27139521358321544, + "grad_norm": 0.38671875, + "learning_rate": 0.0015856568079789404, + "loss": 0.1338, + "step": 31265 + }, + { + "epoch": 0.2714038940634196, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015856317504169435, + "loss": 0.0918, + "step": 31266 + }, + { + "epoch": 0.2714125745436238, + "grad_norm": 0.49609375, + "learning_rate": 0.0015856066923238673, + "loss": 0.1074, + "step": 31267 + }, + { + "epoch": 0.2714212550238279, + "grad_norm": 0.177734375, + "learning_rate": 0.0015855816336997394, + "loss": 0.1328, + "step": 31268 + }, + { + "epoch": 0.2714299355040321, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015855565745445859, + "loss": 0.1074, + "step": 31269 + }, + { + "epoch": 0.27143861598423624, + "grad_norm": 0.9453125, + "learning_rate": 0.0015855315148584352, + "loss": 0.1016, + "step": 31270 + }, + { + "epoch": 0.27144729646444044, + "grad_norm": 0.109375, + "learning_rate": 0.0015855064546413148, + "loss": 0.0928, + "step": 31271 + }, + { + "epoch": 0.2714559769446446, + "grad_norm": 0.34765625, + "learning_rate": 0.001585481393893252, + "loss": 0.0771, + "step": 31272 + }, + { + "epoch": 0.27146465742484877, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015854563326142734, + "loss": 0.084, + "step": 31273 + }, + { + "epoch": 0.2714733379050529, + "grad_norm": 0.3359375, + "learning_rate": 0.0015854312708044073, + "loss": 0.1162, + "step": 31274 + }, + { + "epoch": 0.2714820183852571, + "grad_norm": 3.546875, + "learning_rate": 0.001585406208463681, + "loss": 0.2812, + "step": 31275 + }, + { + "epoch": 0.27149069886546123, + "grad_norm": 0.267578125, + "learning_rate": 0.0015853811455921215, + "loss": 0.1309, + "step": 31276 + }, + { + "epoch": 0.2714993793456654, + "grad_norm": 0.310546875, + "learning_rate": 0.0015853560821897567, + "loss": 0.0908, + "step": 31277 + }, + { + "epoch": 0.27150805982586956, + "grad_norm": 0.119140625, + "learning_rate": 0.001585331018256614, + "loss": 0.0898, + "step": 31278 + }, + { + "epoch": 0.27151674030607376, + "grad_norm": 0.546875, + "learning_rate": 0.0015853059537927203, + "loss": 0.0938, + "step": 31279 + }, + { + "epoch": 0.2715254207862779, + "grad_norm": 0.734375, + "learning_rate": 0.0015852808887981034, + "loss": 0.1094, + "step": 31280 + }, + { + "epoch": 0.2715341012664821, + "grad_norm": 0.8046875, + "learning_rate": 0.0015852558232727907, + "loss": 0.0986, + "step": 31281 + }, + { + "epoch": 0.2715427817466862, + "grad_norm": 0.09912109375, + "learning_rate": 0.0015852307572168094, + "loss": 0.1152, + "step": 31282 + }, + { + "epoch": 0.2715514622268904, + "grad_norm": 0.162109375, + "learning_rate": 0.001585205690630187, + "loss": 0.1084, + "step": 31283 + }, + { + "epoch": 0.27156014270709455, + "grad_norm": 0.33203125, + "learning_rate": 0.001585180623512951, + "loss": 0.1211, + "step": 31284 + }, + { + "epoch": 0.27156882318729875, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015851555558651287, + "loss": 0.0845, + "step": 31285 + }, + { + "epoch": 0.2715775036675029, + "grad_norm": 0.5390625, + "learning_rate": 0.0015851304876867477, + "loss": 0.0996, + "step": 31286 + }, + { + "epoch": 0.2715861841477071, + "grad_norm": 0.22265625, + "learning_rate": 0.0015851054189778351, + "loss": 0.0811, + "step": 31287 + }, + { + "epoch": 0.2715948646279112, + "grad_norm": 0.451171875, + "learning_rate": 0.0015850803497384188, + "loss": 0.1123, + "step": 31288 + }, + { + "epoch": 0.2716035451081154, + "grad_norm": 0.71875, + "learning_rate": 0.0015850552799685253, + "loss": 0.1006, + "step": 31289 + }, + { + "epoch": 0.27161222558831954, + "grad_norm": 0.71484375, + "learning_rate": 0.0015850302096681833, + "loss": 0.1147, + "step": 31290 + }, + { + "epoch": 0.27162090606852374, + "grad_norm": 0.134765625, + "learning_rate": 0.0015850051388374194, + "loss": 0.0947, + "step": 31291 + }, + { + "epoch": 0.2716295865487279, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015849800674762606, + "loss": 0.0938, + "step": 31292 + }, + { + "epoch": 0.27163826702893207, + "grad_norm": 0.291015625, + "learning_rate": 0.0015849549955847355, + "loss": 0.1172, + "step": 31293 + }, + { + "epoch": 0.2716469475091362, + "grad_norm": 0.140625, + "learning_rate": 0.001584929923162871, + "loss": 0.0972, + "step": 31294 + }, + { + "epoch": 0.2716556279893404, + "grad_norm": 0.60546875, + "learning_rate": 0.0015849048502106942, + "loss": 0.1309, + "step": 31295 + }, + { + "epoch": 0.27166430846954454, + "grad_norm": 0.287109375, + "learning_rate": 0.0015848797767282323, + "loss": 0.124, + "step": 31296 + }, + { + "epoch": 0.2716729889497487, + "grad_norm": 0.1142578125, + "learning_rate": 0.001584854702715514, + "loss": 0.1104, + "step": 31297 + }, + { + "epoch": 0.27168166942995287, + "grad_norm": 0.1083984375, + "learning_rate": 0.0015848296281725655, + "loss": 0.0796, + "step": 31298 + }, + { + "epoch": 0.271690349910157, + "grad_norm": 0.2578125, + "learning_rate": 0.0015848045530994143, + "loss": 0.1104, + "step": 31299 + }, + { + "epoch": 0.2716990303903612, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015847794774960884, + "loss": 0.126, + "step": 31300 + }, + { + "epoch": 0.27170771087056533, + "grad_norm": 0.279296875, + "learning_rate": 0.0015847544013626149, + "loss": 0.0957, + "step": 31301 + }, + { + "epoch": 0.2717163913507695, + "grad_norm": 0.66796875, + "learning_rate": 0.0015847293246990214, + "loss": 0.1084, + "step": 31302 + }, + { + "epoch": 0.27172507183097366, + "grad_norm": 0.30078125, + "learning_rate": 0.001584704247505335, + "loss": 0.1182, + "step": 31303 + }, + { + "epoch": 0.27173375231117786, + "grad_norm": 0.365234375, + "learning_rate": 0.0015846791697815836, + "loss": 0.0781, + "step": 31304 + }, + { + "epoch": 0.271742432791382, + "grad_norm": 0.1416015625, + "learning_rate": 0.001584654091527794, + "loss": 0.0874, + "step": 31305 + }, + { + "epoch": 0.2717511132715862, + "grad_norm": 0.9375, + "learning_rate": 0.0015846290127439942, + "loss": 0.1367, + "step": 31306 + }, + { + "epoch": 0.2717597937517903, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015846039334302111, + "loss": 0.0781, + "step": 31307 + }, + { + "epoch": 0.2717684742319945, + "grad_norm": 0.263671875, + "learning_rate": 0.001584578853586473, + "loss": 0.0981, + "step": 31308 + }, + { + "epoch": 0.27177715471219865, + "grad_norm": 0.68359375, + "learning_rate": 0.0015845537732128063, + "loss": 0.0986, + "step": 31309 + }, + { + "epoch": 0.27178583519240285, + "grad_norm": 0.125, + "learning_rate": 0.001584528692309239, + "loss": 0.0908, + "step": 31310 + }, + { + "epoch": 0.271794515672607, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015845036108757984, + "loss": 0.1445, + "step": 31311 + }, + { + "epoch": 0.2718031961528112, + "grad_norm": 0.435546875, + "learning_rate": 0.001584478528912512, + "loss": 0.1064, + "step": 31312 + }, + { + "epoch": 0.2718118766330153, + "grad_norm": 0.384765625, + "learning_rate": 0.0015844534464194069, + "loss": 0.0752, + "step": 31313 + }, + { + "epoch": 0.2718205571132195, + "grad_norm": 0.189453125, + "learning_rate": 0.001584428363396511, + "loss": 0.126, + "step": 31314 + }, + { + "epoch": 0.27182923759342364, + "grad_norm": 0.396484375, + "learning_rate": 0.0015844032798438513, + "loss": 0.1099, + "step": 31315 + }, + { + "epoch": 0.27183791807362784, + "grad_norm": 0.146484375, + "learning_rate": 0.0015843781957614557, + "loss": 0.1289, + "step": 31316 + }, + { + "epoch": 0.271846598553832, + "grad_norm": 0.59375, + "learning_rate": 0.0015843531111493514, + "loss": 0.1074, + "step": 31317 + }, + { + "epoch": 0.27185527903403617, + "grad_norm": 0.09716796875, + "learning_rate": 0.0015843280260075655, + "loss": 0.084, + "step": 31318 + }, + { + "epoch": 0.2718639595142403, + "grad_norm": 0.392578125, + "learning_rate": 0.001584302940336126, + "loss": 0.1011, + "step": 31319 + }, + { + "epoch": 0.2718726399944445, + "grad_norm": 0.48828125, + "learning_rate": 0.0015842778541350602, + "loss": 0.1006, + "step": 31320 + }, + { + "epoch": 0.27188132047464864, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015842527674043952, + "loss": 0.0981, + "step": 31321 + }, + { + "epoch": 0.27189000095485283, + "grad_norm": 0.349609375, + "learning_rate": 0.0015842276801441586, + "loss": 0.1279, + "step": 31322 + }, + { + "epoch": 0.27189868143505697, + "grad_norm": 0.205078125, + "learning_rate": 0.0015842025923543778, + "loss": 0.1143, + "step": 31323 + }, + { + "epoch": 0.27190736191526116, + "grad_norm": 0.0908203125, + "learning_rate": 0.0015841775040350806, + "loss": 0.1289, + "step": 31324 + }, + { + "epoch": 0.2719160423954653, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015841524151862941, + "loss": 0.0933, + "step": 31325 + }, + { + "epoch": 0.2719247228756695, + "grad_norm": 0.45703125, + "learning_rate": 0.001584127325808046, + "loss": 0.0918, + "step": 31326 + }, + { + "epoch": 0.2719334033558736, + "grad_norm": 0.083984375, + "learning_rate": 0.0015841022359003633, + "loss": 0.0967, + "step": 31327 + }, + { + "epoch": 0.2719420838360778, + "grad_norm": 0.19140625, + "learning_rate": 0.0015840771454632737, + "loss": 0.1152, + "step": 31328 + }, + { + "epoch": 0.27195076431628196, + "grad_norm": 0.7265625, + "learning_rate": 0.0015840520544968045, + "loss": 0.1406, + "step": 31329 + }, + { + "epoch": 0.27195944479648615, + "grad_norm": 0.224609375, + "learning_rate": 0.0015840269630009833, + "loss": 0.0908, + "step": 31330 + }, + { + "epoch": 0.2719681252766903, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015840018709758375, + "loss": 0.0747, + "step": 31331 + }, + { + "epoch": 0.2719768057568945, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015839767784213944, + "loss": 0.1338, + "step": 31332 + }, + { + "epoch": 0.2719854862370986, + "grad_norm": 0.2236328125, + "learning_rate": 0.0015839516853376817, + "loss": 0.0967, + "step": 31333 + }, + { + "epoch": 0.2719941667173028, + "grad_norm": 0.47265625, + "learning_rate": 0.0015839265917247268, + "loss": 0.1914, + "step": 31334 + }, + { + "epoch": 0.27200284719750695, + "grad_norm": 0.318359375, + "learning_rate": 0.0015839014975825567, + "loss": 0.1445, + "step": 31335 + }, + { + "epoch": 0.27201152767771114, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015838764029111995, + "loss": 0.0801, + "step": 31336 + }, + { + "epoch": 0.2720202081579153, + "grad_norm": 0.083984375, + "learning_rate": 0.0015838513077106825, + "loss": 0.1016, + "step": 31337 + }, + { + "epoch": 0.27202888863811947, + "grad_norm": 0.302734375, + "learning_rate": 0.0015838262119810328, + "loss": 0.1279, + "step": 31338 + }, + { + "epoch": 0.2720375691183236, + "grad_norm": 3.75, + "learning_rate": 0.0015838011157222778, + "loss": 0.2578, + "step": 31339 + }, + { + "epoch": 0.2720462495985278, + "grad_norm": 0.1240234375, + "learning_rate": 0.0015837760189344454, + "loss": 0.0693, + "step": 31340 + }, + { + "epoch": 0.27205493007873194, + "grad_norm": 0.08056640625, + "learning_rate": 0.0015837509216175627, + "loss": 0.1123, + "step": 31341 + }, + { + "epoch": 0.27206361055893613, + "grad_norm": 0.08203125, + "learning_rate": 0.0015837258237716572, + "loss": 0.1211, + "step": 31342 + }, + { + "epoch": 0.27207229103914027, + "grad_norm": 0.353515625, + "learning_rate": 0.0015837007253967567, + "loss": 0.0967, + "step": 31343 + }, + { + "epoch": 0.27208097151934446, + "grad_norm": 0.177734375, + "learning_rate": 0.001583675626492888, + "loss": 0.1074, + "step": 31344 + }, + { + "epoch": 0.2720896519995486, + "grad_norm": 0.39453125, + "learning_rate": 0.0015836505270600792, + "loss": 0.1016, + "step": 31345 + }, + { + "epoch": 0.2720983324797528, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015836254270983577, + "loss": 0.0918, + "step": 31346 + }, + { + "epoch": 0.27210701295995693, + "grad_norm": 0.25390625, + "learning_rate": 0.0015836003266077499, + "loss": 0.1221, + "step": 31347 + }, + { + "epoch": 0.2721156934401611, + "grad_norm": 0.19140625, + "learning_rate": 0.0015835752255882846, + "loss": 0.1123, + "step": 31348 + }, + { + "epoch": 0.27212437392036526, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015835501240399888, + "loss": 0.0996, + "step": 31349 + }, + { + "epoch": 0.27213305440056945, + "grad_norm": 0.11865234375, + "learning_rate": 0.0015835250219628895, + "loss": 0.1045, + "step": 31350 + }, + { + "epoch": 0.2721417348807736, + "grad_norm": 0.275390625, + "learning_rate": 0.0015834999193570145, + "loss": 0.1709, + "step": 31351 + }, + { + "epoch": 0.2721504153609778, + "grad_norm": 0.9453125, + "learning_rate": 0.0015834748162223914, + "loss": 0.1147, + "step": 31352 + }, + { + "epoch": 0.2721590958411819, + "grad_norm": 0.158203125, + "learning_rate": 0.0015834497125590474, + "loss": 0.1099, + "step": 31353 + }, + { + "epoch": 0.2721677763213861, + "grad_norm": 3.140625, + "learning_rate": 0.0015834246083670104, + "loss": 0.3105, + "step": 31354 + }, + { + "epoch": 0.27217645680159025, + "grad_norm": 0.216796875, + "learning_rate": 0.0015833995036463073, + "loss": 0.0859, + "step": 31355 + }, + { + "epoch": 0.27218513728179444, + "grad_norm": 0.26171875, + "learning_rate": 0.0015833743983969658, + "loss": 0.1143, + "step": 31356 + }, + { + "epoch": 0.2721938177619986, + "grad_norm": 0.59375, + "learning_rate": 0.001583349292619013, + "loss": 0.0884, + "step": 31357 + }, + { + "epoch": 0.2722024982422028, + "grad_norm": 0.06982421875, + "learning_rate": 0.0015833241863124773, + "loss": 0.0845, + "step": 31358 + }, + { + "epoch": 0.2722111787224069, + "grad_norm": 0.353515625, + "learning_rate": 0.001583299079477385, + "loss": 0.083, + "step": 31359 + }, + { + "epoch": 0.2722198592026111, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015832739721137641, + "loss": 0.0825, + "step": 31360 + }, + { + "epoch": 0.27222853968281524, + "grad_norm": 0.30078125, + "learning_rate": 0.0015832488642216424, + "loss": 0.0918, + "step": 31361 + }, + { + "epoch": 0.27223722016301943, + "grad_norm": 0.2060546875, + "learning_rate": 0.0015832237558010465, + "loss": 0.1406, + "step": 31362 + }, + { + "epoch": 0.27224590064322357, + "grad_norm": 0.357421875, + "learning_rate": 0.0015831986468520048, + "loss": 0.1221, + "step": 31363 + }, + { + "epoch": 0.27225458112342776, + "grad_norm": 0.2734375, + "learning_rate": 0.0015831735373745442, + "loss": 0.0869, + "step": 31364 + }, + { + "epoch": 0.2722632616036319, + "grad_norm": 0.546875, + "learning_rate": 0.0015831484273686922, + "loss": 0.0942, + "step": 31365 + }, + { + "epoch": 0.2722719420838361, + "grad_norm": 0.1337890625, + "learning_rate": 0.0015831233168344763, + "loss": 0.0962, + "step": 31366 + }, + { + "epoch": 0.27228062256404023, + "grad_norm": 0.2158203125, + "learning_rate": 0.001583098205771924, + "loss": 0.1367, + "step": 31367 + }, + { + "epoch": 0.2722893030442444, + "grad_norm": 0.546875, + "learning_rate": 0.0015830730941810629, + "loss": 0.0864, + "step": 31368 + }, + { + "epoch": 0.27229798352444856, + "grad_norm": 0.09326171875, + "learning_rate": 0.0015830479820619205, + "loss": 0.0781, + "step": 31369 + }, + { + "epoch": 0.27230666400465275, + "grad_norm": 0.359375, + "learning_rate": 0.0015830228694145236, + "loss": 0.1348, + "step": 31370 + }, + { + "epoch": 0.2723153444848569, + "grad_norm": 0.51171875, + "learning_rate": 0.0015829977562389003, + "loss": 0.1201, + "step": 31371 + }, + { + "epoch": 0.2723240249650611, + "grad_norm": 0.59765625, + "learning_rate": 0.001582972642535078, + "loss": 0.0947, + "step": 31372 + }, + { + "epoch": 0.2723327054452652, + "grad_norm": 0.259765625, + "learning_rate": 0.0015829475283030846, + "loss": 0.0737, + "step": 31373 + }, + { + "epoch": 0.2723413859254694, + "grad_norm": 0.419921875, + "learning_rate": 0.001582922413542946, + "loss": 0.1191, + "step": 31374 + }, + { + "epoch": 0.27235006640567355, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015828972982546913, + "loss": 0.1182, + "step": 31375 + }, + { + "epoch": 0.27235874688587774, + "grad_norm": 0.0595703125, + "learning_rate": 0.0015828721824383476, + "loss": 0.0615, + "step": 31376 + }, + { + "epoch": 0.2723674273660819, + "grad_norm": 0.1064453125, + "learning_rate": 0.0015828470660939418, + "loss": 0.1338, + "step": 31377 + }, + { + "epoch": 0.2723761078462861, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015828219492215018, + "loss": 0.0908, + "step": 31378 + }, + { + "epoch": 0.2723847883264902, + "grad_norm": 0.5078125, + "learning_rate": 0.0015827968318210548, + "loss": 0.1396, + "step": 31379 + }, + { + "epoch": 0.2723934688066944, + "grad_norm": 0.333984375, + "learning_rate": 0.0015827717138926286, + "loss": 0.082, + "step": 31380 + }, + { + "epoch": 0.27240214928689854, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015827465954362503, + "loss": 0.1113, + "step": 31381 + }, + { + "epoch": 0.27241082976710274, + "grad_norm": 0.2421875, + "learning_rate": 0.001582721476451948, + "loss": 0.0708, + "step": 31382 + }, + { + "epoch": 0.2724195102473069, + "grad_norm": 0.19140625, + "learning_rate": 0.0015826963569397483, + "loss": 0.1104, + "step": 31383 + }, + { + "epoch": 0.27242819072751107, + "grad_norm": 0.396484375, + "learning_rate": 0.0015826712368996795, + "loss": 0.0894, + "step": 31384 + }, + { + "epoch": 0.2724368712077152, + "grad_norm": 0.72265625, + "learning_rate": 0.0015826461163317683, + "loss": 0.0962, + "step": 31385 + }, + { + "epoch": 0.2724455516879194, + "grad_norm": 0.07958984375, + "learning_rate": 0.0015826209952360432, + "loss": 0.0967, + "step": 31386 + }, + { + "epoch": 0.27245423216812353, + "grad_norm": 0.376953125, + "learning_rate": 0.0015825958736125302, + "loss": 0.1079, + "step": 31387 + }, + { + "epoch": 0.2724629126483277, + "grad_norm": 0.447265625, + "learning_rate": 0.001582570751461258, + "loss": 0.0864, + "step": 31388 + }, + { + "epoch": 0.27247159312853186, + "grad_norm": 0.40625, + "learning_rate": 0.0015825456287822538, + "loss": 0.1113, + "step": 31389 + }, + { + "epoch": 0.27248027360873606, + "grad_norm": 0.11181640625, + "learning_rate": 0.0015825205055755448, + "loss": 0.1318, + "step": 31390 + }, + { + "epoch": 0.2724889540889402, + "grad_norm": 0.45703125, + "learning_rate": 0.0015824953818411583, + "loss": 0.1211, + "step": 31391 + }, + { + "epoch": 0.2724976345691444, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015824702575791226, + "loss": 0.1143, + "step": 31392 + }, + { + "epoch": 0.2725063150493485, + "grad_norm": 0.072265625, + "learning_rate": 0.0015824451327894646, + "loss": 0.085, + "step": 31393 + }, + { + "epoch": 0.2725149955295527, + "grad_norm": 0.1396484375, + "learning_rate": 0.0015824200074722118, + "loss": 0.1182, + "step": 31394 + }, + { + "epoch": 0.27252367600975685, + "grad_norm": 0.1552734375, + "learning_rate": 0.0015823948816273917, + "loss": 0.1289, + "step": 31395 + }, + { + "epoch": 0.27253235648996105, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015823697552550316, + "loss": 0.1064, + "step": 31396 + }, + { + "epoch": 0.2725410369701652, + "grad_norm": 0.12158203125, + "learning_rate": 0.0015823446283551596, + "loss": 0.1128, + "step": 31397 + }, + { + "epoch": 0.2725497174503694, + "grad_norm": 0.23828125, + "learning_rate": 0.0015823195009278027, + "loss": 0.1011, + "step": 31398 + }, + { + "epoch": 0.2725583979305735, + "grad_norm": 0.275390625, + "learning_rate": 0.001582294372972988, + "loss": 0.1143, + "step": 31399 + }, + { + "epoch": 0.2725670784107777, + "grad_norm": 0.353515625, + "learning_rate": 0.0015822692444907437, + "loss": 0.0928, + "step": 31400 + }, + { + "epoch": 0.27257575889098185, + "grad_norm": 0.330078125, + "learning_rate": 0.001582244115481097, + "loss": 0.1064, + "step": 31401 + }, + { + "epoch": 0.27258443937118604, + "grad_norm": 0.35546875, + "learning_rate": 0.0015822189859440751, + "loss": 0.127, + "step": 31402 + }, + { + "epoch": 0.2725931198513902, + "grad_norm": 0.1689453125, + "learning_rate": 0.0015821938558797065, + "loss": 0.1064, + "step": 31403 + }, + { + "epoch": 0.27260180033159437, + "grad_norm": 0.376953125, + "learning_rate": 0.0015821687252880175, + "loss": 0.0879, + "step": 31404 + }, + { + "epoch": 0.2726104808117985, + "grad_norm": 0.431640625, + "learning_rate": 0.0015821435941690359, + "loss": 0.1318, + "step": 31405 + }, + { + "epoch": 0.2726191612920027, + "grad_norm": 0.2412109375, + "learning_rate": 0.0015821184625227896, + "loss": 0.1113, + "step": 31406 + }, + { + "epoch": 0.27262784177220684, + "grad_norm": 0.09130859375, + "learning_rate": 0.0015820933303493055, + "loss": 0.1006, + "step": 31407 + }, + { + "epoch": 0.27263652225241103, + "grad_norm": 0.0810546875, + "learning_rate": 0.0015820681976486113, + "loss": 0.0771, + "step": 31408 + }, + { + "epoch": 0.27264520273261517, + "grad_norm": 0.07861328125, + "learning_rate": 0.0015820430644207348, + "loss": 0.1025, + "step": 31409 + }, + { + "epoch": 0.27265388321281936, + "grad_norm": 0.93359375, + "learning_rate": 0.0015820179306657032, + "loss": 0.0908, + "step": 31410 + }, + { + "epoch": 0.2726625636930235, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015819927963835441, + "loss": 0.0811, + "step": 31411 + }, + { + "epoch": 0.2726712441732277, + "grad_norm": 0.65625, + "learning_rate": 0.001581967661574285, + "loss": 0.1738, + "step": 31412 + }, + { + "epoch": 0.2726799246534318, + "grad_norm": 0.3671875, + "learning_rate": 0.001581942526237953, + "loss": 0.0908, + "step": 31413 + }, + { + "epoch": 0.272688605133636, + "grad_norm": 0.29296875, + "learning_rate": 0.0015819173903745762, + "loss": 0.125, + "step": 31414 + }, + { + "epoch": 0.27269728561384016, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015818922539841817, + "loss": 0.1074, + "step": 31415 + }, + { + "epoch": 0.27270596609404435, + "grad_norm": 0.1181640625, + "learning_rate": 0.0015818671170667968, + "loss": 0.1133, + "step": 31416 + }, + { + "epoch": 0.2727146465742485, + "grad_norm": 0.330078125, + "learning_rate": 0.0015818419796224494, + "loss": 0.0957, + "step": 31417 + }, + { + "epoch": 0.2727233270544527, + "grad_norm": 0.5546875, + "learning_rate": 0.001581816841651167, + "loss": 0.0928, + "step": 31418 + }, + { + "epoch": 0.2727320075346568, + "grad_norm": 0.419921875, + "learning_rate": 0.0015817917031529767, + "loss": 0.1602, + "step": 31419 + }, + { + "epoch": 0.27274068801486095, + "grad_norm": 0.76953125, + "learning_rate": 0.0015817665641279066, + "loss": 0.084, + "step": 31420 + }, + { + "epoch": 0.27274936849506515, + "grad_norm": 0.470703125, + "learning_rate": 0.0015817414245759835, + "loss": 0.0957, + "step": 31421 + }, + { + "epoch": 0.2727580489752693, + "grad_norm": 0.66015625, + "learning_rate": 0.0015817162844972352, + "loss": 0.0942, + "step": 31422 + }, + { + "epoch": 0.2727667294554735, + "grad_norm": 0.498046875, + "learning_rate": 0.0015816911438916897, + "loss": 0.123, + "step": 31423 + }, + { + "epoch": 0.2727754099356776, + "grad_norm": 0.08203125, + "learning_rate": 0.0015816660027593737, + "loss": 0.0889, + "step": 31424 + }, + { + "epoch": 0.2727840904158818, + "grad_norm": 0.57421875, + "learning_rate": 0.001581640861100315, + "loss": 0.0938, + "step": 31425 + }, + { + "epoch": 0.27279277089608595, + "grad_norm": 0.73046875, + "learning_rate": 0.0015816157189145407, + "loss": 0.1172, + "step": 31426 + }, + { + "epoch": 0.27280145137629014, + "grad_norm": 0.421875, + "learning_rate": 0.0015815905762020792, + "loss": 0.1074, + "step": 31427 + }, + { + "epoch": 0.2728101318564943, + "grad_norm": 0.1875, + "learning_rate": 0.0015815654329629572, + "loss": 0.1006, + "step": 31428 + }, + { + "epoch": 0.27281881233669847, + "grad_norm": 0.115234375, + "learning_rate": 0.0015815402891972026, + "loss": 0.1191, + "step": 31429 + }, + { + "epoch": 0.2728274928169026, + "grad_norm": 0.62890625, + "learning_rate": 0.0015815151449048428, + "loss": 0.1523, + "step": 31430 + }, + { + "epoch": 0.2728361732971068, + "grad_norm": 0.2734375, + "learning_rate": 0.001581490000085905, + "loss": 0.1455, + "step": 31431 + }, + { + "epoch": 0.27284485377731094, + "grad_norm": 0.115234375, + "learning_rate": 0.0015814648547404176, + "loss": 0.1182, + "step": 31432 + }, + { + "epoch": 0.27285353425751513, + "grad_norm": 0.4921875, + "learning_rate": 0.001581439708868407, + "loss": 0.1064, + "step": 31433 + }, + { + "epoch": 0.27286221473771927, + "grad_norm": 0.5390625, + "learning_rate": 0.001581414562469901, + "loss": 0.1211, + "step": 31434 + }, + { + "epoch": 0.27287089521792346, + "grad_norm": 1.109375, + "learning_rate": 0.0015813894155449278, + "loss": 0.0977, + "step": 31435 + }, + { + "epoch": 0.2728795756981276, + "grad_norm": 0.33203125, + "learning_rate": 0.0015813642680935138, + "loss": 0.1338, + "step": 31436 + }, + { + "epoch": 0.2728882561783318, + "grad_norm": 0.09228515625, + "learning_rate": 0.0015813391201156873, + "loss": 0.1211, + "step": 31437 + }, + { + "epoch": 0.2728969366585359, + "grad_norm": 0.56640625, + "learning_rate": 0.0015813139716114757, + "loss": 0.1094, + "step": 31438 + }, + { + "epoch": 0.2729056171387401, + "grad_norm": 0.275390625, + "learning_rate": 0.001581288822580906, + "loss": 0.1162, + "step": 31439 + }, + { + "epoch": 0.27291429761894426, + "grad_norm": 0.12255859375, + "learning_rate": 0.0015812636730240066, + "loss": 0.103, + "step": 31440 + }, + { + "epoch": 0.27292297809914845, + "grad_norm": 0.375, + "learning_rate": 0.0015812385229408044, + "loss": 0.0854, + "step": 31441 + }, + { + "epoch": 0.2729316585793526, + "grad_norm": 0.326171875, + "learning_rate": 0.001581213372331327, + "loss": 0.1348, + "step": 31442 + }, + { + "epoch": 0.2729403390595568, + "grad_norm": 0.33203125, + "learning_rate": 0.0015811882211956014, + "loss": 0.0806, + "step": 31443 + }, + { + "epoch": 0.2729490195397609, + "grad_norm": 0.2890625, + "learning_rate": 0.001581163069533656, + "loss": 0.1172, + "step": 31444 + }, + { + "epoch": 0.2729577000199651, + "grad_norm": 0.119140625, + "learning_rate": 0.0015811379173455177, + "loss": 0.1055, + "step": 31445 + }, + { + "epoch": 0.27296638050016925, + "grad_norm": 0.171875, + "learning_rate": 0.0015811127646312144, + "loss": 0.0996, + "step": 31446 + }, + { + "epoch": 0.27297506098037344, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015810876113907732, + "loss": 0.0869, + "step": 31447 + }, + { + "epoch": 0.2729837414605776, + "grad_norm": 0.498046875, + "learning_rate": 0.0015810624576242218, + "loss": 0.1084, + "step": 31448 + }, + { + "epoch": 0.27299242194078177, + "grad_norm": 0.119140625, + "learning_rate": 0.0015810373033315881, + "loss": 0.0791, + "step": 31449 + }, + { + "epoch": 0.2730011024209859, + "grad_norm": 0.3359375, + "learning_rate": 0.001581012148512899, + "loss": 0.123, + "step": 31450 + }, + { + "epoch": 0.2730097829011901, + "grad_norm": 0.1708984375, + "learning_rate": 0.0015809869931681822, + "loss": 0.062, + "step": 31451 + }, + { + "epoch": 0.27301846338139424, + "grad_norm": 0.453125, + "learning_rate": 0.0015809618372974652, + "loss": 0.1182, + "step": 31452 + }, + { + "epoch": 0.27302714386159843, + "grad_norm": 0.90625, + "learning_rate": 0.0015809366809007759, + "loss": 0.1123, + "step": 31453 + }, + { + "epoch": 0.27303582434180257, + "grad_norm": 0.50390625, + "learning_rate": 0.001580911523978141, + "loss": 0.123, + "step": 31454 + }, + { + "epoch": 0.27304450482200676, + "grad_norm": 0.5703125, + "learning_rate": 0.0015808863665295887, + "loss": 0.1245, + "step": 31455 + }, + { + "epoch": 0.2730531853022109, + "grad_norm": 0.2099609375, + "learning_rate": 0.0015808612085551463, + "loss": 0.0947, + "step": 31456 + }, + { + "epoch": 0.2730618657824151, + "grad_norm": 0.58984375, + "learning_rate": 0.0015808360500548413, + "loss": 0.0894, + "step": 31457 + }, + { + "epoch": 0.27307054626261923, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015808108910287011, + "loss": 0.1133, + "step": 31458 + }, + { + "epoch": 0.2730792267428234, + "grad_norm": 0.396484375, + "learning_rate": 0.0015807857314767534, + "loss": 0.0732, + "step": 31459 + }, + { + "epoch": 0.27308790722302756, + "grad_norm": 0.283203125, + "learning_rate": 0.0015807605713990258, + "loss": 0.1089, + "step": 31460 + }, + { + "epoch": 0.27309658770323175, + "grad_norm": 0.87109375, + "learning_rate": 0.0015807354107955456, + "loss": 0.0942, + "step": 31461 + }, + { + "epoch": 0.2731052681834359, + "grad_norm": 0.48046875, + "learning_rate": 0.00158071024966634, + "loss": 0.0977, + "step": 31462 + }, + { + "epoch": 0.2731139486636401, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015806850880114376, + "loss": 0.1074, + "step": 31463 + }, + { + "epoch": 0.2731226291438442, + "grad_norm": 0.3359375, + "learning_rate": 0.0015806599258308648, + "loss": 0.1387, + "step": 31464 + }, + { + "epoch": 0.2731313096240484, + "grad_norm": 0.48046875, + "learning_rate": 0.0015806347631246497, + "loss": 0.0737, + "step": 31465 + }, + { + "epoch": 0.27313999010425255, + "grad_norm": 0.310546875, + "learning_rate": 0.0015806095998928194, + "loss": 0.084, + "step": 31466 + }, + { + "epoch": 0.27314867058445674, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015805844361354016, + "loss": 0.1055, + "step": 31467 + }, + { + "epoch": 0.2731573510646609, + "grad_norm": 0.310546875, + "learning_rate": 0.0015805592718524243, + "loss": 0.1562, + "step": 31468 + }, + { + "epoch": 0.2731660315448651, + "grad_norm": 0.451171875, + "learning_rate": 0.0015805341070439139, + "loss": 0.1182, + "step": 31469 + }, + { + "epoch": 0.2731747120250692, + "grad_norm": 0.11669921875, + "learning_rate": 0.0015805089417098993, + "loss": 0.1211, + "step": 31470 + }, + { + "epoch": 0.2731833925052734, + "grad_norm": 0.19140625, + "learning_rate": 0.001580483775850407, + "loss": 0.0869, + "step": 31471 + }, + { + "epoch": 0.27319207298547754, + "grad_norm": 0.408203125, + "learning_rate": 0.0015804586094654652, + "loss": 0.1123, + "step": 31472 + }, + { + "epoch": 0.27320075346568173, + "grad_norm": 0.56640625, + "learning_rate": 0.0015804334425551009, + "loss": 0.0869, + "step": 31473 + }, + { + "epoch": 0.27320943394588587, + "grad_norm": 0.34765625, + "learning_rate": 0.0015804082751193415, + "loss": 0.0918, + "step": 31474 + }, + { + "epoch": 0.27321811442609006, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015803831071582153, + "loss": 0.0825, + "step": 31475 + }, + { + "epoch": 0.2732267949062942, + "grad_norm": 0.1025390625, + "learning_rate": 0.001580357938671749, + "loss": 0.1138, + "step": 31476 + }, + { + "epoch": 0.2732354753864984, + "grad_norm": 0.578125, + "learning_rate": 0.0015803327696599708, + "loss": 0.0898, + "step": 31477 + }, + { + "epoch": 0.27324415586670253, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015803076001229075, + "loss": 0.0986, + "step": 31478 + }, + { + "epoch": 0.2732528363469067, + "grad_norm": 0.1826171875, + "learning_rate": 0.0015802824300605873, + "loss": 0.1064, + "step": 31479 + }, + { + "epoch": 0.27326151682711086, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015802572594730375, + "loss": 0.0957, + "step": 31480 + }, + { + "epoch": 0.27327019730731505, + "grad_norm": 0.51171875, + "learning_rate": 0.0015802320883602854, + "loss": 0.1279, + "step": 31481 + }, + { + "epoch": 0.2732788777875192, + "grad_norm": 0.10107421875, + "learning_rate": 0.0015802069167223588, + "loss": 0.1035, + "step": 31482 + }, + { + "epoch": 0.2732875582677234, + "grad_norm": 0.326171875, + "learning_rate": 0.0015801817445592853, + "loss": 0.1113, + "step": 31483 + }, + { + "epoch": 0.2732962387479275, + "grad_norm": 0.314453125, + "learning_rate": 0.001580156571871092, + "loss": 0.0967, + "step": 31484 + }, + { + "epoch": 0.2733049192281317, + "grad_norm": 0.349609375, + "learning_rate": 0.0015801313986578067, + "loss": 0.085, + "step": 31485 + }, + { + "epoch": 0.27331359970833585, + "grad_norm": 0.1630859375, + "learning_rate": 0.001580106224919457, + "loss": 0.1079, + "step": 31486 + }, + { + "epoch": 0.27332228018854005, + "grad_norm": 0.279296875, + "learning_rate": 0.0015800810506560702, + "loss": 0.1245, + "step": 31487 + }, + { + "epoch": 0.2733309606687442, + "grad_norm": 0.091796875, + "learning_rate": 0.0015800558758676739, + "loss": 0.1216, + "step": 31488 + }, + { + "epoch": 0.2733396411489484, + "grad_norm": 0.271484375, + "learning_rate": 0.0015800307005542963, + "loss": 0.1094, + "step": 31489 + }, + { + "epoch": 0.2733483216291525, + "grad_norm": 0.119140625, + "learning_rate": 0.001580005524715964, + "loss": 0.0957, + "step": 31490 + }, + { + "epoch": 0.2733570021093567, + "grad_norm": 0.142578125, + "learning_rate": 0.0015799803483527045, + "loss": 0.084, + "step": 31491 + }, + { + "epoch": 0.27336568258956084, + "grad_norm": 0.314453125, + "learning_rate": 0.0015799551714645462, + "loss": 0.1084, + "step": 31492 + }, + { + "epoch": 0.27337436306976504, + "grad_norm": 0.1796875, + "learning_rate": 0.001579929994051516, + "loss": 0.0771, + "step": 31493 + }, + { + "epoch": 0.2733830435499692, + "grad_norm": 0.091796875, + "learning_rate": 0.0015799048161136415, + "loss": 0.1025, + "step": 31494 + }, + { + "epoch": 0.27339172403017337, + "grad_norm": 0.16015625, + "learning_rate": 0.00157987963765095, + "loss": 0.1162, + "step": 31495 + }, + { + "epoch": 0.2734004045103775, + "grad_norm": 0.1484375, + "learning_rate": 0.0015798544586634697, + "loss": 0.1436, + "step": 31496 + }, + { + "epoch": 0.2734090849905817, + "grad_norm": 0.60546875, + "learning_rate": 0.0015798292791512276, + "loss": 0.1104, + "step": 31497 + }, + { + "epoch": 0.27341776547078583, + "grad_norm": 0.310546875, + "learning_rate": 0.001579804099114252, + "loss": 0.1006, + "step": 31498 + }, + { + "epoch": 0.27342644595099, + "grad_norm": 0.11181640625, + "learning_rate": 0.001579778918552569, + "loss": 0.1143, + "step": 31499 + }, + { + "epoch": 0.27343512643119416, + "grad_norm": 0.365234375, + "learning_rate": 0.0015797537374662076, + "loss": 0.0928, + "step": 31500 + }, + { + "epoch": 0.27344380691139836, + "grad_norm": 0.388671875, + "learning_rate": 0.0015797285558551944, + "loss": 0.1094, + "step": 31501 + }, + { + "epoch": 0.2734524873916025, + "grad_norm": 0.09423828125, + "learning_rate": 0.0015797033737195575, + "loss": 0.106, + "step": 31502 + }, + { + "epoch": 0.2734611678718067, + "grad_norm": 0.1240234375, + "learning_rate": 0.0015796781910593241, + "loss": 0.165, + "step": 31503 + }, + { + "epoch": 0.2734698483520108, + "grad_norm": 0.52734375, + "learning_rate": 0.0015796530078745217, + "loss": 0.125, + "step": 31504 + }, + { + "epoch": 0.273478528832215, + "grad_norm": 0.376953125, + "learning_rate": 0.001579627824165178, + "loss": 0.1377, + "step": 31505 + }, + { + "epoch": 0.27348720931241915, + "grad_norm": 0.5703125, + "learning_rate": 0.0015796026399313204, + "loss": 0.1152, + "step": 31506 + }, + { + "epoch": 0.27349588979262335, + "grad_norm": 0.73828125, + "learning_rate": 0.001579577455172977, + "loss": 0.1084, + "step": 31507 + }, + { + "epoch": 0.2735045702728275, + "grad_norm": 0.08935546875, + "learning_rate": 0.0015795522698901744, + "loss": 0.0903, + "step": 31508 + }, + { + "epoch": 0.2735132507530317, + "grad_norm": 0.53515625, + "learning_rate": 0.001579527084082941, + "loss": 0.1172, + "step": 31509 + }, + { + "epoch": 0.2735219312332358, + "grad_norm": 0.26953125, + "learning_rate": 0.001579501897751304, + "loss": 0.1221, + "step": 31510 + }, + { + "epoch": 0.27353061171344, + "grad_norm": 0.466796875, + "learning_rate": 0.0015794767108952908, + "loss": 0.1118, + "step": 31511 + }, + { + "epoch": 0.27353929219364415, + "grad_norm": 0.796875, + "learning_rate": 0.0015794515235149288, + "loss": 0.1484, + "step": 31512 + }, + { + "epoch": 0.27354797267384834, + "grad_norm": 0.353515625, + "learning_rate": 0.001579426335610246, + "loss": 0.166, + "step": 31513 + }, + { + "epoch": 0.2735566531540525, + "grad_norm": 0.5390625, + "learning_rate": 0.0015794011471812702, + "loss": 0.0933, + "step": 31514 + }, + { + "epoch": 0.27356533363425667, + "grad_norm": 0.1240234375, + "learning_rate": 0.001579375958228028, + "loss": 0.0938, + "step": 31515 + }, + { + "epoch": 0.2735740141144608, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015793507687505478, + "loss": 0.0947, + "step": 31516 + }, + { + "epoch": 0.273582694594665, + "grad_norm": 0.82421875, + "learning_rate": 0.0015793255787488563, + "loss": 0.0771, + "step": 31517 + }, + { + "epoch": 0.27359137507486914, + "grad_norm": 0.8671875, + "learning_rate": 0.0015793003882229823, + "loss": 0.1123, + "step": 31518 + }, + { + "epoch": 0.27360005555507333, + "grad_norm": 0.333984375, + "learning_rate": 0.001579275197172952, + "loss": 0.0933, + "step": 31519 + }, + { + "epoch": 0.27360873603527747, + "grad_norm": 0.46484375, + "learning_rate": 0.0015792500055987937, + "loss": 0.0928, + "step": 31520 + }, + { + "epoch": 0.27361741651548166, + "grad_norm": 0.2314453125, + "learning_rate": 0.0015792248135005347, + "loss": 0.1328, + "step": 31521 + }, + { + "epoch": 0.2736260969956858, + "grad_norm": 0.30859375, + "learning_rate": 0.0015791996208782027, + "loss": 0.1777, + "step": 31522 + }, + { + "epoch": 0.27363477747589, + "grad_norm": 0.146484375, + "learning_rate": 0.0015791744277318255, + "loss": 0.1133, + "step": 31523 + }, + { + "epoch": 0.2736434579560941, + "grad_norm": 0.9140625, + "learning_rate": 0.0015791492340614297, + "loss": 0.166, + "step": 31524 + }, + { + "epoch": 0.2736521384362983, + "grad_norm": 0.29296875, + "learning_rate": 0.0015791240398670438, + "loss": 0.103, + "step": 31525 + }, + { + "epoch": 0.27366081891650246, + "grad_norm": 0.259765625, + "learning_rate": 0.0015790988451486953, + "loss": 0.0815, + "step": 31526 + }, + { + "epoch": 0.27366949939670665, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015790736499064113, + "loss": 0.1133, + "step": 31527 + }, + { + "epoch": 0.2736781798769108, + "grad_norm": 0.1142578125, + "learning_rate": 0.00157904845414022, + "loss": 0.0786, + "step": 31528 + }, + { + "epoch": 0.273686860357115, + "grad_norm": 0.1328125, + "learning_rate": 0.0015790232578501478, + "loss": 0.1025, + "step": 31529 + }, + { + "epoch": 0.2736955408373191, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015789980610362232, + "loss": 0.124, + "step": 31530 + }, + { + "epoch": 0.2737042213175233, + "grad_norm": 0.35546875, + "learning_rate": 0.0015789728636984737, + "loss": 0.0732, + "step": 31531 + }, + { + "epoch": 0.27371290179772745, + "grad_norm": 0.38671875, + "learning_rate": 0.0015789476658369266, + "loss": 0.103, + "step": 31532 + }, + { + "epoch": 0.27372158227793164, + "grad_norm": 0.6328125, + "learning_rate": 0.0015789224674516093, + "loss": 0.1074, + "step": 31533 + }, + { + "epoch": 0.2737302627581358, + "grad_norm": 0.40234375, + "learning_rate": 0.0015788972685425499, + "loss": 0.103, + "step": 31534 + }, + { + "epoch": 0.27373894323833997, + "grad_norm": 0.7109375, + "learning_rate": 0.0015788720691097752, + "loss": 0.1055, + "step": 31535 + }, + { + "epoch": 0.2737476237185441, + "grad_norm": 0.515625, + "learning_rate": 0.0015788468691533138, + "loss": 0.0933, + "step": 31536 + }, + { + "epoch": 0.2737563041987483, + "grad_norm": 0.376953125, + "learning_rate": 0.0015788216686731923, + "loss": 0.1226, + "step": 31537 + }, + { + "epoch": 0.27376498467895244, + "grad_norm": 0.67578125, + "learning_rate": 0.001578796467669439, + "loss": 0.1211, + "step": 31538 + }, + { + "epoch": 0.27377366515915663, + "grad_norm": 0.412109375, + "learning_rate": 0.0015787712661420802, + "loss": 0.1445, + "step": 31539 + }, + { + "epoch": 0.27378234563936077, + "grad_norm": 0.9921875, + "learning_rate": 0.0015787460640911454, + "loss": 0.1416, + "step": 31540 + }, + { + "epoch": 0.27379102611956496, + "grad_norm": 0.35546875, + "learning_rate": 0.0015787208615166603, + "loss": 0.0742, + "step": 31541 + }, + { + "epoch": 0.2737997065997691, + "grad_norm": 0.18359375, + "learning_rate": 0.0015786956584186538, + "loss": 0.1436, + "step": 31542 + }, + { + "epoch": 0.27380838707997324, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015786704547971525, + "loss": 0.1025, + "step": 31543 + }, + { + "epoch": 0.27381706756017743, + "grad_norm": 0.98828125, + "learning_rate": 0.0015786452506521844, + "loss": 0.1328, + "step": 31544 + }, + { + "epoch": 0.27382574804038157, + "grad_norm": 0.8203125, + "learning_rate": 0.001578620045983777, + "loss": 0.085, + "step": 31545 + }, + { + "epoch": 0.27383442852058576, + "grad_norm": 0.82421875, + "learning_rate": 0.0015785948407919584, + "loss": 0.0806, + "step": 31546 + }, + { + "epoch": 0.2738431090007899, + "grad_norm": 0.2109375, + "learning_rate": 0.0015785696350767552, + "loss": 0.1011, + "step": 31547 + }, + { + "epoch": 0.2738517894809941, + "grad_norm": 0.63671875, + "learning_rate": 0.0015785444288381957, + "loss": 0.0791, + "step": 31548 + }, + { + "epoch": 0.2738604699611982, + "grad_norm": 0.302734375, + "learning_rate": 0.0015785192220763074, + "loss": 0.105, + "step": 31549 + }, + { + "epoch": 0.2738691504414024, + "grad_norm": 0.1142578125, + "learning_rate": 0.0015784940147911173, + "loss": 0.0957, + "step": 31550 + }, + { + "epoch": 0.27387783092160656, + "grad_norm": 0.263671875, + "learning_rate": 0.0015784688069826534, + "loss": 0.1465, + "step": 31551 + }, + { + "epoch": 0.27388651140181075, + "grad_norm": 0.1484375, + "learning_rate": 0.0015784435986509434, + "loss": 0.1143, + "step": 31552 + }, + { + "epoch": 0.2738951918820149, + "grad_norm": 0.65625, + "learning_rate": 0.0015784183897960143, + "loss": 0.0928, + "step": 31553 + }, + { + "epoch": 0.2739038723622191, + "grad_norm": 0.0947265625, + "learning_rate": 0.0015783931804178941, + "loss": 0.0859, + "step": 31554 + }, + { + "epoch": 0.2739125528424232, + "grad_norm": 0.2255859375, + "learning_rate": 0.0015783679705166104, + "loss": 0.0796, + "step": 31555 + }, + { + "epoch": 0.2739212333226274, + "grad_norm": 0.57421875, + "learning_rate": 0.001578342760092191, + "loss": 0.1133, + "step": 31556 + }, + { + "epoch": 0.27392991380283155, + "grad_norm": 0.08935546875, + "learning_rate": 0.0015783175491446626, + "loss": 0.1104, + "step": 31557 + }, + { + "epoch": 0.27393859428303574, + "grad_norm": 0.275390625, + "learning_rate": 0.0015782923376740539, + "loss": 0.1221, + "step": 31558 + }, + { + "epoch": 0.2739472747632399, + "grad_norm": 0.29296875, + "learning_rate": 0.0015782671256803916, + "loss": 0.1157, + "step": 31559 + }, + { + "epoch": 0.27395595524344407, + "grad_norm": 0.42578125, + "learning_rate": 0.0015782419131637033, + "loss": 0.0957, + "step": 31560 + }, + { + "epoch": 0.2739646357236482, + "grad_norm": 0.1318359375, + "learning_rate": 0.001578216700124017, + "loss": 0.1191, + "step": 31561 + }, + { + "epoch": 0.2739733162038524, + "grad_norm": 0.0966796875, + "learning_rate": 0.00157819148656136, + "loss": 0.1108, + "step": 31562 + }, + { + "epoch": 0.27398199668405654, + "grad_norm": 0.10107421875, + "learning_rate": 0.0015781662724757603, + "loss": 0.1182, + "step": 31563 + }, + { + "epoch": 0.27399067716426073, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015781410578672448, + "loss": 0.1123, + "step": 31564 + }, + { + "epoch": 0.27399935764446487, + "grad_norm": 0.58984375, + "learning_rate": 0.001578115842735842, + "loss": 0.124, + "step": 31565 + }, + { + "epoch": 0.27400803812466906, + "grad_norm": 0.09228515625, + "learning_rate": 0.0015780906270815782, + "loss": 0.1133, + "step": 31566 + }, + { + "epoch": 0.2740167186048732, + "grad_norm": 0.298828125, + "learning_rate": 0.0015780654109044821, + "loss": 0.0806, + "step": 31567 + }, + { + "epoch": 0.2740253990850774, + "grad_norm": 0.267578125, + "learning_rate": 0.0015780401942045808, + "loss": 0.1279, + "step": 31568 + }, + { + "epoch": 0.27403407956528153, + "grad_norm": 0.53125, + "learning_rate": 0.0015780149769819016, + "loss": 0.1348, + "step": 31569 + }, + { + "epoch": 0.2740427600454857, + "grad_norm": 0.28125, + "learning_rate": 0.0015779897592364729, + "loss": 0.1177, + "step": 31570 + }, + { + "epoch": 0.27405144052568986, + "grad_norm": 1.6875, + "learning_rate": 0.0015779645409683214, + "loss": 0.4141, + "step": 31571 + }, + { + "epoch": 0.27406012100589405, + "grad_norm": 0.330078125, + "learning_rate": 0.0015779393221774752, + "loss": 0.0996, + "step": 31572 + }, + { + "epoch": 0.2740688014860982, + "grad_norm": 0.65625, + "learning_rate": 0.0015779141028639616, + "loss": 0.1523, + "step": 31573 + }, + { + "epoch": 0.2740774819663024, + "grad_norm": 0.40625, + "learning_rate": 0.0015778888830278086, + "loss": 0.1006, + "step": 31574 + }, + { + "epoch": 0.2740861624465065, + "grad_norm": 0.5078125, + "learning_rate": 0.0015778636626690432, + "loss": 0.1064, + "step": 31575 + }, + { + "epoch": 0.2740948429267107, + "grad_norm": 0.294921875, + "learning_rate": 0.0015778384417876937, + "loss": 0.1162, + "step": 31576 + }, + { + "epoch": 0.27410352340691485, + "grad_norm": 0.0693359375, + "learning_rate": 0.0015778132203837867, + "loss": 0.0781, + "step": 31577 + }, + { + "epoch": 0.27411220388711904, + "grad_norm": 0.470703125, + "learning_rate": 0.0015777879984573506, + "loss": 0.0918, + "step": 31578 + }, + { + "epoch": 0.2741208843673232, + "grad_norm": 0.69921875, + "learning_rate": 0.0015777627760084128, + "loss": 0.105, + "step": 31579 + }, + { + "epoch": 0.2741295648475274, + "grad_norm": 0.265625, + "learning_rate": 0.0015777375530370008, + "loss": 0.0952, + "step": 31580 + }, + { + "epoch": 0.2741382453277315, + "grad_norm": 0.40234375, + "learning_rate": 0.0015777123295431421, + "loss": 0.1094, + "step": 31581 + }, + { + "epoch": 0.2741469258079357, + "grad_norm": 0.189453125, + "learning_rate": 0.0015776871055268645, + "loss": 0.0957, + "step": 31582 + }, + { + "epoch": 0.27415560628813984, + "grad_norm": 0.1748046875, + "learning_rate": 0.0015776618809881953, + "loss": 0.1602, + "step": 31583 + }, + { + "epoch": 0.27416428676834403, + "grad_norm": 0.55078125, + "learning_rate": 0.0015776366559271623, + "loss": 0.0938, + "step": 31584 + }, + { + "epoch": 0.27417296724854817, + "grad_norm": 0.2041015625, + "learning_rate": 0.001577611430343793, + "loss": 0.1074, + "step": 31585 + }, + { + "epoch": 0.27418164772875236, + "grad_norm": 0.380859375, + "learning_rate": 0.0015775862042381149, + "loss": 0.127, + "step": 31586 + }, + { + "epoch": 0.2741903282089565, + "grad_norm": 0.328125, + "learning_rate": 0.001577560977610156, + "loss": 0.0942, + "step": 31587 + }, + { + "epoch": 0.2741990086891607, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015775357504599435, + "loss": 0.106, + "step": 31588 + }, + { + "epoch": 0.27420768916936483, + "grad_norm": 0.07763671875, + "learning_rate": 0.0015775105227875052, + "loss": 0.0879, + "step": 31589 + }, + { + "epoch": 0.274216369649569, + "grad_norm": 0.1982421875, + "learning_rate": 0.0015774852945928683, + "loss": 0.0898, + "step": 31590 + }, + { + "epoch": 0.27422505012977316, + "grad_norm": 0.31640625, + "learning_rate": 0.0015774600658760608, + "loss": 0.126, + "step": 31591 + }, + { + "epoch": 0.27423373060997736, + "grad_norm": 0.353515625, + "learning_rate": 0.0015774348366371098, + "loss": 0.0771, + "step": 31592 + }, + { + "epoch": 0.2742424110901815, + "grad_norm": 0.396484375, + "learning_rate": 0.0015774096068760437, + "loss": 0.0889, + "step": 31593 + }, + { + "epoch": 0.2742510915703857, + "grad_norm": 0.365234375, + "learning_rate": 0.0015773843765928895, + "loss": 0.1289, + "step": 31594 + }, + { + "epoch": 0.2742597720505898, + "grad_norm": 0.373046875, + "learning_rate": 0.001577359145787675, + "loss": 0.0898, + "step": 31595 + }, + { + "epoch": 0.274268452530794, + "grad_norm": 0.1484375, + "learning_rate": 0.0015773339144604275, + "loss": 0.0742, + "step": 31596 + }, + { + "epoch": 0.27427713301099815, + "grad_norm": 0.64453125, + "learning_rate": 0.0015773086826111752, + "loss": 0.1504, + "step": 31597 + }, + { + "epoch": 0.27428581349120235, + "grad_norm": 0.15234375, + "learning_rate": 0.0015772834502399448, + "loss": 0.1162, + "step": 31598 + }, + { + "epoch": 0.2742944939714065, + "grad_norm": 0.734375, + "learning_rate": 0.0015772582173467645, + "loss": 0.1128, + "step": 31599 + }, + { + "epoch": 0.2743031744516107, + "grad_norm": 0.31640625, + "learning_rate": 0.0015772329839316619, + "loss": 0.0762, + "step": 31600 + }, + { + "epoch": 0.2743118549318148, + "grad_norm": 0.09228515625, + "learning_rate": 0.0015772077499946644, + "loss": 0.0718, + "step": 31601 + }, + { + "epoch": 0.274320535412019, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015771825155357997, + "loss": 0.0986, + "step": 31602 + }, + { + "epoch": 0.27432921589222314, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015771572805550953, + "loss": 0.0913, + "step": 31603 + }, + { + "epoch": 0.27433789637242734, + "grad_norm": 0.162109375, + "learning_rate": 0.001577132045052579, + "loss": 0.1094, + "step": 31604 + }, + { + "epoch": 0.2743465768526315, + "grad_norm": 0.236328125, + "learning_rate": 0.0015771068090282782, + "loss": 0.1074, + "step": 31605 + }, + { + "epoch": 0.27435525733283567, + "grad_norm": 0.78125, + "learning_rate": 0.0015770815724822206, + "loss": 0.0986, + "step": 31606 + }, + { + "epoch": 0.2743639378130398, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015770563354144337, + "loss": 0.1348, + "step": 31607 + }, + { + "epoch": 0.274372618293244, + "grad_norm": 0.33984375, + "learning_rate": 0.0015770310978249454, + "loss": 0.1201, + "step": 31608 + }, + { + "epoch": 0.27438129877344813, + "grad_norm": 0.2216796875, + "learning_rate": 0.0015770058597137828, + "loss": 0.0908, + "step": 31609 + }, + { + "epoch": 0.2743899792536523, + "grad_norm": 0.3203125, + "learning_rate": 0.0015769806210809737, + "loss": 0.0869, + "step": 31610 + }, + { + "epoch": 0.27439865973385646, + "grad_norm": 0.330078125, + "learning_rate": 0.0015769553819265458, + "loss": 0.082, + "step": 31611 + }, + { + "epoch": 0.27440734021406066, + "grad_norm": 0.0947265625, + "learning_rate": 0.0015769301422505267, + "loss": 0.1084, + "step": 31612 + }, + { + "epoch": 0.2744160206942648, + "grad_norm": 0.6328125, + "learning_rate": 0.0015769049020529443, + "loss": 0.2871, + "step": 31613 + }, + { + "epoch": 0.274424701174469, + "grad_norm": 0.56640625, + "learning_rate": 0.0015768796613338256, + "loss": 0.1104, + "step": 31614 + }, + { + "epoch": 0.2744333816546731, + "grad_norm": 0.45703125, + "learning_rate": 0.0015768544200931982, + "loss": 0.105, + "step": 31615 + }, + { + "epoch": 0.2744420621348773, + "grad_norm": 0.57421875, + "learning_rate": 0.0015768291783310902, + "loss": 0.0962, + "step": 31616 + }, + { + "epoch": 0.27445074261508146, + "grad_norm": 0.2890625, + "learning_rate": 0.001576803936047529, + "loss": 0.1011, + "step": 31617 + }, + { + "epoch": 0.27445942309528565, + "grad_norm": 0.259765625, + "learning_rate": 0.0015767786932425422, + "loss": 0.1201, + "step": 31618 + }, + { + "epoch": 0.2744681035754898, + "grad_norm": 0.08447265625, + "learning_rate": 0.0015767534499161572, + "loss": 0.0928, + "step": 31619 + }, + { + "epoch": 0.274476784055694, + "grad_norm": 0.4609375, + "learning_rate": 0.001576728206068402, + "loss": 0.1113, + "step": 31620 + }, + { + "epoch": 0.2744854645358981, + "grad_norm": 0.45703125, + "learning_rate": 0.0015767029616993039, + "loss": 0.1221, + "step": 31621 + }, + { + "epoch": 0.2744941450161023, + "grad_norm": 0.38671875, + "learning_rate": 0.0015766777168088907, + "loss": 0.1572, + "step": 31622 + }, + { + "epoch": 0.27450282549630645, + "grad_norm": 0.44140625, + "learning_rate": 0.0015766524713971902, + "loss": 0.0762, + "step": 31623 + }, + { + "epoch": 0.27451150597651064, + "grad_norm": 0.091796875, + "learning_rate": 0.001576627225464229, + "loss": 0.0615, + "step": 31624 + }, + { + "epoch": 0.2745201864567148, + "grad_norm": 0.28125, + "learning_rate": 0.001576601979010036, + "loss": 0.0957, + "step": 31625 + }, + { + "epoch": 0.27452886693691897, + "grad_norm": 0.60546875, + "learning_rate": 0.001576576732034638, + "loss": 0.1162, + "step": 31626 + }, + { + "epoch": 0.2745375474171231, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015765514845380631, + "loss": 0.1133, + "step": 31627 + }, + { + "epoch": 0.2745462278973273, + "grad_norm": 0.28515625, + "learning_rate": 0.0015765262365203382, + "loss": 0.0967, + "step": 31628 + }, + { + "epoch": 0.27455490837753144, + "grad_norm": 0.296875, + "learning_rate": 0.0015765009879814915, + "loss": 0.1553, + "step": 31629 + }, + { + "epoch": 0.27456358885773563, + "grad_norm": 0.451171875, + "learning_rate": 0.0015764757389215502, + "loss": 0.1162, + "step": 31630 + }, + { + "epoch": 0.27457226933793977, + "grad_norm": 0.416015625, + "learning_rate": 0.001576450489340543, + "loss": 0.1162, + "step": 31631 + }, + { + "epoch": 0.27458094981814396, + "grad_norm": 0.46484375, + "learning_rate": 0.0015764252392384964, + "loss": 0.105, + "step": 31632 + }, + { + "epoch": 0.2745896302983481, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015763999886154378, + "loss": 0.1162, + "step": 31633 + }, + { + "epoch": 0.2745983107785523, + "grad_norm": 0.62890625, + "learning_rate": 0.001576374737471396, + "loss": 0.125, + "step": 31634 + }, + { + "epoch": 0.2746069912587564, + "grad_norm": 0.26953125, + "learning_rate": 0.0015763494858063976, + "loss": 0.1245, + "step": 31635 + }, + { + "epoch": 0.2746156717389606, + "grad_norm": 0.208984375, + "learning_rate": 0.0015763242336204709, + "loss": 0.0776, + "step": 31636 + }, + { + "epoch": 0.27462435221916476, + "grad_norm": 0.400390625, + "learning_rate": 0.0015762989809136427, + "loss": 0.1289, + "step": 31637 + }, + { + "epoch": 0.27463303269936895, + "grad_norm": 0.2265625, + "learning_rate": 0.0015762737276859412, + "loss": 0.085, + "step": 31638 + }, + { + "epoch": 0.2746417131795731, + "grad_norm": 0.171875, + "learning_rate": 0.0015762484739373938, + "loss": 0.0791, + "step": 31639 + }, + { + "epoch": 0.2746503936597773, + "grad_norm": 0.33984375, + "learning_rate": 0.0015762232196680286, + "loss": 0.1006, + "step": 31640 + }, + { + "epoch": 0.2746590741399814, + "grad_norm": 0.67578125, + "learning_rate": 0.0015761979648778726, + "loss": 0.4883, + "step": 31641 + }, + { + "epoch": 0.2746677546201856, + "grad_norm": 0.4921875, + "learning_rate": 0.0015761727095669536, + "loss": 0.1025, + "step": 31642 + }, + { + "epoch": 0.27467643510038975, + "grad_norm": 0.5078125, + "learning_rate": 0.0015761474537352995, + "loss": 0.125, + "step": 31643 + }, + { + "epoch": 0.27468511558059394, + "grad_norm": 0.3203125, + "learning_rate": 0.0015761221973829376, + "loss": 0.1089, + "step": 31644 + }, + { + "epoch": 0.2746937960607981, + "grad_norm": 0.419921875, + "learning_rate": 0.001576096940509896, + "loss": 0.0991, + "step": 31645 + }, + { + "epoch": 0.27470247654100227, + "grad_norm": 0.10791015625, + "learning_rate": 0.0015760716831162015, + "loss": 0.0957, + "step": 31646 + }, + { + "epoch": 0.2747111570212064, + "grad_norm": 0.236328125, + "learning_rate": 0.001576046425201882, + "loss": 0.0957, + "step": 31647 + }, + { + "epoch": 0.2747198375014106, + "grad_norm": 0.328125, + "learning_rate": 0.0015760211667669656, + "loss": 0.1406, + "step": 31648 + }, + { + "epoch": 0.27472851798161474, + "grad_norm": 0.451171875, + "learning_rate": 0.0015759959078114794, + "loss": 0.0786, + "step": 31649 + }, + { + "epoch": 0.27473719846181893, + "grad_norm": 0.337890625, + "learning_rate": 0.0015759706483354512, + "loss": 0.126, + "step": 31650 + }, + { + "epoch": 0.27474587894202307, + "grad_norm": 0.10791015625, + "learning_rate": 0.0015759453883389088, + "loss": 0.124, + "step": 31651 + }, + { + "epoch": 0.27475455942222726, + "grad_norm": 0.09619140625, + "learning_rate": 0.0015759201278218798, + "loss": 0.0869, + "step": 31652 + }, + { + "epoch": 0.2747632399024314, + "grad_norm": 0.703125, + "learning_rate": 0.0015758948667843917, + "loss": 0.1221, + "step": 31653 + }, + { + "epoch": 0.2747719203826356, + "grad_norm": 0.11279296875, + "learning_rate": 0.001575869605226472, + "loss": 0.1006, + "step": 31654 + }, + { + "epoch": 0.27478060086283973, + "grad_norm": 0.263671875, + "learning_rate": 0.0015758443431481482, + "loss": 0.0776, + "step": 31655 + }, + { + "epoch": 0.2747892813430439, + "grad_norm": 0.154296875, + "learning_rate": 0.0015758190805494484, + "loss": 0.0918, + "step": 31656 + }, + { + "epoch": 0.27479796182324806, + "grad_norm": 0.51171875, + "learning_rate": 0.0015757938174304, + "loss": 0.166, + "step": 31657 + }, + { + "epoch": 0.27480664230345225, + "grad_norm": 0.458984375, + "learning_rate": 0.001575768553791031, + "loss": 0.1104, + "step": 31658 + }, + { + "epoch": 0.2748153227836564, + "grad_norm": 0.06689453125, + "learning_rate": 0.0015757432896313678, + "loss": 0.1079, + "step": 31659 + }, + { + "epoch": 0.2748240032638606, + "grad_norm": 0.2392578125, + "learning_rate": 0.0015757180249514396, + "loss": 0.1436, + "step": 31660 + }, + { + "epoch": 0.2748326837440647, + "grad_norm": 0.890625, + "learning_rate": 0.001575692759751273, + "loss": 0.123, + "step": 31661 + }, + { + "epoch": 0.2748413642242689, + "grad_norm": 1.25, + "learning_rate": 0.001575667494030896, + "loss": 0.123, + "step": 31662 + }, + { + "epoch": 0.27485004470447305, + "grad_norm": 0.26171875, + "learning_rate": 0.0015756422277903364, + "loss": 0.1094, + "step": 31663 + }, + { + "epoch": 0.27485872518467724, + "grad_norm": 0.0908203125, + "learning_rate": 0.0015756169610296213, + "loss": 0.104, + "step": 31664 + }, + { + "epoch": 0.2748674056648814, + "grad_norm": 0.10498046875, + "learning_rate": 0.001575591693748779, + "loss": 0.127, + "step": 31665 + }, + { + "epoch": 0.2748760861450855, + "grad_norm": 0.6328125, + "learning_rate": 0.0015755664259478365, + "loss": 0.105, + "step": 31666 + }, + { + "epoch": 0.2748847666252897, + "grad_norm": 0.625, + "learning_rate": 0.0015755411576268215, + "loss": 0.0884, + "step": 31667 + }, + { + "epoch": 0.27489344710549385, + "grad_norm": 0.271484375, + "learning_rate": 0.0015755158887857624, + "loss": 0.0952, + "step": 31668 + }, + { + "epoch": 0.27490212758569804, + "grad_norm": 0.48046875, + "learning_rate": 0.0015754906194246857, + "loss": 0.1309, + "step": 31669 + }, + { + "epoch": 0.2749108080659022, + "grad_norm": 0.091796875, + "learning_rate": 0.0015754653495436198, + "loss": 0.0825, + "step": 31670 + }, + { + "epoch": 0.27491948854610637, + "grad_norm": 0.1083984375, + "learning_rate": 0.0015754400791425922, + "loss": 0.1152, + "step": 31671 + }, + { + "epoch": 0.2749281690263105, + "grad_norm": 0.240234375, + "learning_rate": 0.0015754148082216305, + "loss": 0.0796, + "step": 31672 + }, + { + "epoch": 0.2749368495065147, + "grad_norm": 0.0830078125, + "learning_rate": 0.0015753895367807622, + "loss": 0.127, + "step": 31673 + }, + { + "epoch": 0.27494552998671884, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015753642648200152, + "loss": 0.0996, + "step": 31674 + }, + { + "epoch": 0.27495421046692303, + "grad_norm": 0.2060546875, + "learning_rate": 0.001575338992339417, + "loss": 0.1045, + "step": 31675 + }, + { + "epoch": 0.27496289094712717, + "grad_norm": 0.431640625, + "learning_rate": 0.0015753137193389951, + "loss": 0.1128, + "step": 31676 + }, + { + "epoch": 0.27497157142733136, + "grad_norm": 0.30078125, + "learning_rate": 0.001575288445818777, + "loss": 0.1475, + "step": 31677 + }, + { + "epoch": 0.2749802519075355, + "grad_norm": 0.2451171875, + "learning_rate": 0.0015752631717787908, + "loss": 0.1025, + "step": 31678 + }, + { + "epoch": 0.2749889323877397, + "grad_norm": 0.498046875, + "learning_rate": 0.0015752378972190642, + "loss": 0.0781, + "step": 31679 + }, + { + "epoch": 0.27499761286794383, + "grad_norm": 0.384765625, + "learning_rate": 0.0015752126221396244, + "loss": 0.1064, + "step": 31680 + }, + { + "epoch": 0.275006293348148, + "grad_norm": 0.1005859375, + "learning_rate": 0.001575187346540499, + "loss": 0.125, + "step": 31681 + }, + { + "epoch": 0.27501497382835216, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015751620704217162, + "loss": 0.1211, + "step": 31682 + }, + { + "epoch": 0.27502365430855635, + "grad_norm": 0.228515625, + "learning_rate": 0.0015751367937833032, + "loss": 0.0801, + "step": 31683 + }, + { + "epoch": 0.2750323347887605, + "grad_norm": 0.123046875, + "learning_rate": 0.0015751115166252876, + "loss": 0.0845, + "step": 31684 + }, + { + "epoch": 0.2750410152689647, + "grad_norm": 0.203125, + "learning_rate": 0.0015750862389476974, + "loss": 0.0688, + "step": 31685 + }, + { + "epoch": 0.2750496957491688, + "grad_norm": 0.375, + "learning_rate": 0.00157506096075056, + "loss": 0.1177, + "step": 31686 + }, + { + "epoch": 0.275058376229373, + "grad_norm": 0.703125, + "learning_rate": 0.0015750356820339027, + "loss": 0.1494, + "step": 31687 + }, + { + "epoch": 0.27506705670957715, + "grad_norm": 0.39453125, + "learning_rate": 0.0015750104027977539, + "loss": 0.1113, + "step": 31688 + }, + { + "epoch": 0.27507573718978134, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015749851230421408, + "loss": 0.0981, + "step": 31689 + }, + { + "epoch": 0.2750844176699855, + "grad_norm": 0.08203125, + "learning_rate": 0.001574959842767091, + "loss": 0.0835, + "step": 31690 + }, + { + "epoch": 0.2750930981501897, + "grad_norm": 0.44140625, + "learning_rate": 0.0015749345619726323, + "loss": 0.1182, + "step": 31691 + }, + { + "epoch": 0.2751017786303938, + "grad_norm": 0.10986328125, + "learning_rate": 0.0015749092806587924, + "loss": 0.0684, + "step": 31692 + }, + { + "epoch": 0.275110459110598, + "grad_norm": 0.228515625, + "learning_rate": 0.0015748839988255984, + "loss": 0.084, + "step": 31693 + }, + { + "epoch": 0.27511913959080214, + "grad_norm": 0.248046875, + "learning_rate": 0.001574858716473079, + "loss": 0.1016, + "step": 31694 + }, + { + "epoch": 0.27512782007100633, + "grad_norm": 0.11083984375, + "learning_rate": 0.0015748334336012609, + "loss": 0.1064, + "step": 31695 + }, + { + "epoch": 0.27513650055121047, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015748081502101722, + "loss": 0.1152, + "step": 31696 + }, + { + "epoch": 0.27514518103141467, + "grad_norm": 0.22265625, + "learning_rate": 0.0015747828662998405, + "loss": 0.1113, + "step": 31697 + }, + { + "epoch": 0.2751538615116188, + "grad_norm": 0.416015625, + "learning_rate": 0.001574757581870293, + "loss": 0.1084, + "step": 31698 + }, + { + "epoch": 0.275162541991823, + "grad_norm": 0.162109375, + "learning_rate": 0.0015747322969215582, + "loss": 0.0796, + "step": 31699 + }, + { + "epoch": 0.27517122247202713, + "grad_norm": 0.07958984375, + "learning_rate": 0.0015747070114536632, + "loss": 0.0996, + "step": 31700 + }, + { + "epoch": 0.2751799029522313, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015746817254666359, + "loss": 0.1318, + "step": 31701 + }, + { + "epoch": 0.27518858343243546, + "grad_norm": 0.2373046875, + "learning_rate": 0.0015746564389605032, + "loss": 0.124, + "step": 31702 + }, + { + "epoch": 0.27519726391263966, + "grad_norm": 0.09228515625, + "learning_rate": 0.0015746311519352937, + "loss": 0.0972, + "step": 31703 + }, + { + "epoch": 0.2752059443928438, + "grad_norm": 0.578125, + "learning_rate": 0.001574605864391035, + "loss": 0.0962, + "step": 31704 + }, + { + "epoch": 0.275214624873048, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015745805763277543, + "loss": 0.1011, + "step": 31705 + }, + { + "epoch": 0.2752233053532521, + "grad_norm": 0.46875, + "learning_rate": 0.0015745552877454792, + "loss": 0.0918, + "step": 31706 + }, + { + "epoch": 0.2752319858334563, + "grad_norm": 0.53515625, + "learning_rate": 0.0015745299986442377, + "loss": 0.1196, + "step": 31707 + }, + { + "epoch": 0.27524066631366045, + "grad_norm": 0.1796875, + "learning_rate": 0.0015745047090240574, + "loss": 0.1045, + "step": 31708 + }, + { + "epoch": 0.27524934679386465, + "grad_norm": 0.328125, + "learning_rate": 0.0015744794188849655, + "loss": 0.0967, + "step": 31709 + }, + { + "epoch": 0.2752580272740688, + "grad_norm": 0.15625, + "learning_rate": 0.0015744541282269905, + "loss": 0.0664, + "step": 31710 + }, + { + "epoch": 0.275266707754273, + "grad_norm": 0.447265625, + "learning_rate": 0.0015744288370501593, + "loss": 0.0981, + "step": 31711 + }, + { + "epoch": 0.2752753882344771, + "grad_norm": 0.431640625, + "learning_rate": 0.0015744035453545, + "loss": 0.0933, + "step": 31712 + }, + { + "epoch": 0.2752840687146813, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015743782531400402, + "loss": 0.0859, + "step": 31713 + }, + { + "epoch": 0.27529274919488544, + "grad_norm": 0.08740234375, + "learning_rate": 0.0015743529604068075, + "loss": 0.1055, + "step": 31714 + }, + { + "epoch": 0.27530142967508964, + "grad_norm": 0.6328125, + "learning_rate": 0.0015743276671548291, + "loss": 0.0967, + "step": 31715 + }, + { + "epoch": 0.2753101101552938, + "grad_norm": 0.21875, + "learning_rate": 0.0015743023733841334, + "loss": 0.0967, + "step": 31716 + }, + { + "epoch": 0.27531879063549797, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015742770790947477, + "loss": 0.1543, + "step": 31717 + }, + { + "epoch": 0.2753274711157021, + "grad_norm": 0.421875, + "learning_rate": 0.0015742517842866999, + "loss": 0.1045, + "step": 31718 + }, + { + "epoch": 0.2753361515959063, + "grad_norm": 0.185546875, + "learning_rate": 0.0015742264889600174, + "loss": 0.0859, + "step": 31719 + }, + { + "epoch": 0.27534483207611044, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015742011931147275, + "loss": 0.1309, + "step": 31720 + }, + { + "epoch": 0.27535351255631463, + "grad_norm": 0.314453125, + "learning_rate": 0.0015741758967508586, + "loss": 0.0566, + "step": 31721 + }, + { + "epoch": 0.27536219303651877, + "grad_norm": 0.50390625, + "learning_rate": 0.0015741505998684382, + "loss": 0.1001, + "step": 31722 + }, + { + "epoch": 0.27537087351672296, + "grad_norm": 0.9375, + "learning_rate": 0.0015741253024674938, + "loss": 0.1045, + "step": 31723 + }, + { + "epoch": 0.2753795539969271, + "grad_norm": 0.275390625, + "learning_rate": 0.0015741000045480528, + "loss": 0.1631, + "step": 31724 + }, + { + "epoch": 0.2753882344771313, + "grad_norm": 0.322265625, + "learning_rate": 0.0015740747061101436, + "loss": 0.0869, + "step": 31725 + }, + { + "epoch": 0.2753969149573354, + "grad_norm": 0.6640625, + "learning_rate": 0.0015740494071537932, + "loss": 0.3359, + "step": 31726 + }, + { + "epoch": 0.2754055954375396, + "grad_norm": 0.07958984375, + "learning_rate": 0.0015740241076790294, + "loss": 0.0869, + "step": 31727 + }, + { + "epoch": 0.27541427591774376, + "grad_norm": 0.6875, + "learning_rate": 0.00157399880768588, + "loss": 0.1562, + "step": 31728 + }, + { + "epoch": 0.27542295639794795, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015739735071743726, + "loss": 0.1006, + "step": 31729 + }, + { + "epoch": 0.2754316368781521, + "grad_norm": 0.48828125, + "learning_rate": 0.0015739482061445354, + "loss": 0.1172, + "step": 31730 + }, + { + "epoch": 0.2754403173583563, + "grad_norm": 0.494140625, + "learning_rate": 0.001573922904596395, + "loss": 0.0845, + "step": 31731 + }, + { + "epoch": 0.2754489978385604, + "grad_norm": 0.3359375, + "learning_rate": 0.0015738976025299798, + "loss": 0.124, + "step": 31732 + }, + { + "epoch": 0.2754576783187646, + "grad_norm": 0.48828125, + "learning_rate": 0.0015738722999453173, + "loss": 0.0977, + "step": 31733 + }, + { + "epoch": 0.27546635879896875, + "grad_norm": 0.640625, + "learning_rate": 0.0015738469968424348, + "loss": 0.0928, + "step": 31734 + }, + { + "epoch": 0.27547503927917294, + "grad_norm": 0.60546875, + "learning_rate": 0.001573821693221361, + "loss": 0.1289, + "step": 31735 + }, + { + "epoch": 0.2754837197593771, + "grad_norm": 0.10400390625, + "learning_rate": 0.0015737963890821226, + "loss": 0.1006, + "step": 31736 + }, + { + "epoch": 0.27549240023958127, + "grad_norm": 0.2197265625, + "learning_rate": 0.0015737710844247474, + "loss": 0.1172, + "step": 31737 + }, + { + "epoch": 0.2755010807197854, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015737457792492633, + "loss": 0.1523, + "step": 31738 + }, + { + "epoch": 0.2755097611999896, + "grad_norm": 0.11767578125, + "learning_rate": 0.0015737204735556982, + "loss": 0.1138, + "step": 31739 + }, + { + "epoch": 0.27551844168019374, + "grad_norm": 0.171875, + "learning_rate": 0.0015736951673440795, + "loss": 0.0869, + "step": 31740 + }, + { + "epoch": 0.27552712216039793, + "grad_norm": 0.2734375, + "learning_rate": 0.0015736698606144347, + "loss": 0.0938, + "step": 31741 + }, + { + "epoch": 0.27553580264060207, + "grad_norm": 0.16015625, + "learning_rate": 0.0015736445533667918, + "loss": 0.127, + "step": 31742 + }, + { + "epoch": 0.27554448312080626, + "grad_norm": 1.265625, + "learning_rate": 0.0015736192456011781, + "loss": 0.1367, + "step": 31743 + }, + { + "epoch": 0.2755531636010104, + "grad_norm": 0.16015625, + "learning_rate": 0.0015735939373176218, + "loss": 0.0898, + "step": 31744 + }, + { + "epoch": 0.2755618440812146, + "grad_norm": 0.255859375, + "learning_rate": 0.0015735686285161502, + "loss": 0.1221, + "step": 31745 + }, + { + "epoch": 0.27557052456141873, + "grad_norm": 0.38671875, + "learning_rate": 0.001573543319196791, + "loss": 0.1196, + "step": 31746 + }, + { + "epoch": 0.2755792050416229, + "grad_norm": 0.65625, + "learning_rate": 0.001573518009359572, + "loss": 0.1084, + "step": 31747 + }, + { + "epoch": 0.27558788552182706, + "grad_norm": 0.494140625, + "learning_rate": 0.0015734926990045206, + "loss": 0.1025, + "step": 31748 + }, + { + "epoch": 0.27559656600203125, + "grad_norm": 0.5234375, + "learning_rate": 0.001573467388131665, + "loss": 0.1191, + "step": 31749 + }, + { + "epoch": 0.2756052464822354, + "grad_norm": 0.1826171875, + "learning_rate": 0.0015734420767410326, + "loss": 0.1055, + "step": 31750 + }, + { + "epoch": 0.2756139269624396, + "grad_norm": 0.2451171875, + "learning_rate": 0.0015734167648326508, + "loss": 0.1064, + "step": 31751 + }, + { + "epoch": 0.2756226074426437, + "grad_norm": 0.51953125, + "learning_rate": 0.0015733914524065475, + "loss": 0.1035, + "step": 31752 + }, + { + "epoch": 0.2756312879228479, + "grad_norm": 0.1884765625, + "learning_rate": 0.001573366139462751, + "loss": 0.0898, + "step": 31753 + }, + { + "epoch": 0.27563996840305205, + "grad_norm": 0.38671875, + "learning_rate": 0.0015733408260012877, + "loss": 0.1289, + "step": 31754 + }, + { + "epoch": 0.27564864888325624, + "grad_norm": 0.3828125, + "learning_rate": 0.001573315512022186, + "loss": 0.0698, + "step": 31755 + }, + { + "epoch": 0.2756573293634604, + "grad_norm": 0.2373046875, + "learning_rate": 0.001573290197525474, + "loss": 0.1123, + "step": 31756 + }, + { + "epoch": 0.2756660098436646, + "grad_norm": 0.18359375, + "learning_rate": 0.001573264882511179, + "loss": 0.0864, + "step": 31757 + }, + { + "epoch": 0.2756746903238687, + "grad_norm": 0.5546875, + "learning_rate": 0.0015732395669793282, + "loss": 0.1465, + "step": 31758 + }, + { + "epoch": 0.2756833708040729, + "grad_norm": 0.2021484375, + "learning_rate": 0.00157321425092995, + "loss": 0.0981, + "step": 31759 + }, + { + "epoch": 0.27569205128427704, + "grad_norm": 0.208984375, + "learning_rate": 0.0015731889343630718, + "loss": 0.1108, + "step": 31760 + }, + { + "epoch": 0.27570073176448123, + "grad_norm": 0.4296875, + "learning_rate": 0.0015731636172787212, + "loss": 0.0713, + "step": 31761 + }, + { + "epoch": 0.27570941224468537, + "grad_norm": 0.380859375, + "learning_rate": 0.001573138299676926, + "loss": 0.0938, + "step": 31762 + }, + { + "epoch": 0.27571809272488956, + "grad_norm": 0.2734375, + "learning_rate": 0.0015731129815577138, + "loss": 0.0898, + "step": 31763 + }, + { + "epoch": 0.2757267732050937, + "grad_norm": 0.59765625, + "learning_rate": 0.0015730876629211127, + "loss": 0.0864, + "step": 31764 + }, + { + "epoch": 0.2757354536852979, + "grad_norm": 0.318359375, + "learning_rate": 0.0015730623437671496, + "loss": 0.105, + "step": 31765 + }, + { + "epoch": 0.27574413416550203, + "grad_norm": 0.06787109375, + "learning_rate": 0.0015730370240958527, + "loss": 0.0645, + "step": 31766 + }, + { + "epoch": 0.2757528146457062, + "grad_norm": 0.08544921875, + "learning_rate": 0.0015730117039072498, + "loss": 0.0869, + "step": 31767 + }, + { + "epoch": 0.27576149512591036, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015729863832013682, + "loss": 0.0796, + "step": 31768 + }, + { + "epoch": 0.27577017560611455, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015729610619782362, + "loss": 0.0815, + "step": 31769 + }, + { + "epoch": 0.2757788560863187, + "grad_norm": 0.0986328125, + "learning_rate": 0.0015729357402378807, + "loss": 0.0986, + "step": 31770 + }, + { + "epoch": 0.2757875365665229, + "grad_norm": 0.32421875, + "learning_rate": 0.0015729104179803297, + "loss": 0.1221, + "step": 31771 + }, + { + "epoch": 0.275796217046727, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015728850952056116, + "loss": 0.1504, + "step": 31772 + }, + { + "epoch": 0.2758048975269312, + "grad_norm": 0.376953125, + "learning_rate": 0.0015728597719137528, + "loss": 0.1113, + "step": 31773 + }, + { + "epoch": 0.27581357800713535, + "grad_norm": 0.12451171875, + "learning_rate": 0.0015728344481047817, + "loss": 0.1006, + "step": 31774 + }, + { + "epoch": 0.27582225848733954, + "grad_norm": 0.359375, + "learning_rate": 0.0015728091237787263, + "loss": 0.0645, + "step": 31775 + }, + { + "epoch": 0.2758309389675437, + "grad_norm": 0.0869140625, + "learning_rate": 0.0015727837989356138, + "loss": 0.1094, + "step": 31776 + }, + { + "epoch": 0.2758396194477479, + "grad_norm": 0.322265625, + "learning_rate": 0.001572758473575472, + "loss": 0.1182, + "step": 31777 + }, + { + "epoch": 0.275848299927952, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015727331476983286, + "loss": 0.0898, + "step": 31778 + }, + { + "epoch": 0.2758569804081562, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015727078213042117, + "loss": 0.1196, + "step": 31779 + }, + { + "epoch": 0.27586566088836034, + "grad_norm": 0.07373046875, + "learning_rate": 0.0015726824943931479, + "loss": 0.0903, + "step": 31780 + }, + { + "epoch": 0.27587434136856454, + "grad_norm": 0.0966796875, + "learning_rate": 0.0015726571669651664, + "loss": 0.1152, + "step": 31781 + }, + { + "epoch": 0.2758830218487687, + "grad_norm": 0.6484375, + "learning_rate": 0.0015726318390202937, + "loss": 0.0603, + "step": 31782 + }, + { + "epoch": 0.27589170232897287, + "grad_norm": 0.28125, + "learning_rate": 0.0015726065105585577, + "loss": 0.1523, + "step": 31783 + }, + { + "epoch": 0.275900382809177, + "grad_norm": 0.337890625, + "learning_rate": 0.0015725811815799867, + "loss": 0.1211, + "step": 31784 + }, + { + "epoch": 0.2759090632893812, + "grad_norm": 1.0234375, + "learning_rate": 0.001572555852084608, + "loss": 0.0933, + "step": 31785 + }, + { + "epoch": 0.27591774376958533, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015725305220724492, + "loss": 0.0879, + "step": 31786 + }, + { + "epoch": 0.2759264242497895, + "grad_norm": 0.6796875, + "learning_rate": 0.0015725051915435383, + "loss": 0.1279, + "step": 31787 + }, + { + "epoch": 0.27593510472999366, + "grad_norm": 0.322265625, + "learning_rate": 0.0015724798604979026, + "loss": 0.1309, + "step": 31788 + }, + { + "epoch": 0.2759437852101978, + "grad_norm": 0.29296875, + "learning_rate": 0.0015724545289355703, + "loss": 0.0918, + "step": 31789 + }, + { + "epoch": 0.275952465690402, + "grad_norm": 0.26171875, + "learning_rate": 0.0015724291968565686, + "loss": 0.1064, + "step": 31790 + }, + { + "epoch": 0.27596114617060613, + "grad_norm": 0.3671875, + "learning_rate": 0.0015724038642609252, + "loss": 0.2021, + "step": 31791 + }, + { + "epoch": 0.2759698266508103, + "grad_norm": 0.2353515625, + "learning_rate": 0.0015723785311486684, + "loss": 0.0742, + "step": 31792 + }, + { + "epoch": 0.27597850713101446, + "grad_norm": 0.251953125, + "learning_rate": 0.0015723531975198254, + "loss": 0.0884, + "step": 31793 + }, + { + "epoch": 0.27598718761121865, + "grad_norm": 0.66015625, + "learning_rate": 0.001572327863374424, + "loss": 0.1748, + "step": 31794 + }, + { + "epoch": 0.2759958680914228, + "grad_norm": 0.0947265625, + "learning_rate": 0.0015723025287124922, + "loss": 0.1182, + "step": 31795 + }, + { + "epoch": 0.276004548571627, + "grad_norm": 0.66796875, + "learning_rate": 0.0015722771935340572, + "loss": 0.1211, + "step": 31796 + }, + { + "epoch": 0.2760132290518311, + "grad_norm": 0.11181640625, + "learning_rate": 0.001572251857839147, + "loss": 0.1118, + "step": 31797 + }, + { + "epoch": 0.2760219095320353, + "grad_norm": 0.345703125, + "learning_rate": 0.0015722265216277895, + "loss": 0.1094, + "step": 31798 + }, + { + "epoch": 0.27603059001223945, + "grad_norm": 0.30078125, + "learning_rate": 0.001572201184900012, + "loss": 0.0698, + "step": 31799 + }, + { + "epoch": 0.27603927049244364, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015721758476558423, + "loss": 0.125, + "step": 31800 + }, + { + "epoch": 0.2760479509726478, + "grad_norm": 0.27734375, + "learning_rate": 0.0015721505098953083, + "loss": 0.1582, + "step": 31801 + }, + { + "epoch": 0.276056631452852, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015721251716184373, + "loss": 0.0957, + "step": 31802 + }, + { + "epoch": 0.2760653119330561, + "grad_norm": 0.19921875, + "learning_rate": 0.0015720998328252575, + "loss": 0.1016, + "step": 31803 + }, + { + "epoch": 0.2760739924132603, + "grad_norm": 0.404296875, + "learning_rate": 0.0015720744935157967, + "loss": 0.1118, + "step": 31804 + }, + { + "epoch": 0.27608267289346444, + "grad_norm": 0.091796875, + "learning_rate": 0.001572049153690082, + "loss": 0.1201, + "step": 31805 + }, + { + "epoch": 0.27609135337366864, + "grad_norm": 0.48046875, + "learning_rate": 0.0015720238133481415, + "loss": 0.2031, + "step": 31806 + }, + { + "epoch": 0.2761000338538728, + "grad_norm": 0.29296875, + "learning_rate": 0.0015719984724900029, + "loss": 0.124, + "step": 31807 + }, + { + "epoch": 0.27610871433407697, + "grad_norm": 0.09130859375, + "learning_rate": 0.001571973131115694, + "loss": 0.1064, + "step": 31808 + }, + { + "epoch": 0.2761173948142811, + "grad_norm": 0.1484375, + "learning_rate": 0.001571947789225242, + "loss": 0.1104, + "step": 31809 + }, + { + "epoch": 0.2761260752944853, + "grad_norm": 0.314453125, + "learning_rate": 0.0015719224468186753, + "loss": 0.1172, + "step": 31810 + }, + { + "epoch": 0.27613475577468943, + "grad_norm": 0.296875, + "learning_rate": 0.0015718971038960212, + "loss": 0.085, + "step": 31811 + }, + { + "epoch": 0.2761434362548936, + "grad_norm": 0.09716796875, + "learning_rate": 0.0015718717604573074, + "loss": 0.0864, + "step": 31812 + }, + { + "epoch": 0.27615211673509776, + "grad_norm": 0.314453125, + "learning_rate": 0.0015718464165025618, + "loss": 0.0908, + "step": 31813 + }, + { + "epoch": 0.27616079721530196, + "grad_norm": 0.271484375, + "learning_rate": 0.0015718210720318123, + "loss": 0.124, + "step": 31814 + }, + { + "epoch": 0.2761694776955061, + "grad_norm": 0.130859375, + "learning_rate": 0.0015717957270450858, + "loss": 0.0991, + "step": 31815 + }, + { + "epoch": 0.2761781581757103, + "grad_norm": 0.2021484375, + "learning_rate": 0.001571770381542411, + "loss": 0.1069, + "step": 31816 + }, + { + "epoch": 0.2761868386559144, + "grad_norm": 0.8359375, + "learning_rate": 0.0015717450355238153, + "loss": 0.1074, + "step": 31817 + }, + { + "epoch": 0.2761955191361186, + "grad_norm": 0.1953125, + "learning_rate": 0.0015717196889893264, + "loss": 0.0908, + "step": 31818 + }, + { + "epoch": 0.27620419961632275, + "grad_norm": 0.25390625, + "learning_rate": 0.0015716943419389713, + "loss": 0.1377, + "step": 31819 + }, + { + "epoch": 0.27621288009652695, + "grad_norm": 0.322265625, + "learning_rate": 0.001571668994372779, + "loss": 0.1338, + "step": 31820 + }, + { + "epoch": 0.2762215605767311, + "grad_norm": 0.41015625, + "learning_rate": 0.0015716436462907765, + "loss": 0.1182, + "step": 31821 + }, + { + "epoch": 0.2762302410569353, + "grad_norm": 0.51171875, + "learning_rate": 0.0015716182976929912, + "loss": 0.0869, + "step": 31822 + }, + { + "epoch": 0.2762389215371394, + "grad_norm": 0.267578125, + "learning_rate": 0.0015715929485794514, + "loss": 0.1465, + "step": 31823 + }, + { + "epoch": 0.2762476020173436, + "grad_norm": 0.5703125, + "learning_rate": 0.0015715675989501844, + "loss": 0.0869, + "step": 31824 + }, + { + "epoch": 0.27625628249754774, + "grad_norm": 0.416015625, + "learning_rate": 0.0015715422488052186, + "loss": 0.1167, + "step": 31825 + }, + { + "epoch": 0.27626496297775194, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015715168981445813, + "loss": 0.1123, + "step": 31826 + }, + { + "epoch": 0.2762736434579561, + "grad_norm": 0.35546875, + "learning_rate": 0.0015714915469683, + "loss": 0.0806, + "step": 31827 + }, + { + "epoch": 0.27628232393816027, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015714661952764026, + "loss": 0.0884, + "step": 31828 + }, + { + "epoch": 0.2762910044183644, + "grad_norm": 0.1826171875, + "learning_rate": 0.0015714408430689169, + "loss": 0.0933, + "step": 31829 + }, + { + "epoch": 0.2762996848985686, + "grad_norm": 0.228515625, + "learning_rate": 0.0015714154903458707, + "loss": 0.1328, + "step": 31830 + }, + { + "epoch": 0.27630836537877274, + "grad_norm": 0.1171875, + "learning_rate": 0.0015713901371072914, + "loss": 0.0825, + "step": 31831 + }, + { + "epoch": 0.27631704585897693, + "grad_norm": 0.86328125, + "learning_rate": 0.0015713647833532067, + "loss": 0.127, + "step": 31832 + }, + { + "epoch": 0.27632572633918107, + "grad_norm": 0.1748046875, + "learning_rate": 0.0015713394290836449, + "loss": 0.0869, + "step": 31833 + }, + { + "epoch": 0.27633440681938526, + "grad_norm": 0.169921875, + "learning_rate": 0.0015713140742986333, + "loss": 0.1099, + "step": 31834 + }, + { + "epoch": 0.2763430872995894, + "grad_norm": 0.08740234375, + "learning_rate": 0.0015712887189982, + "loss": 0.0859, + "step": 31835 + }, + { + "epoch": 0.2763517677797936, + "grad_norm": 0.32421875, + "learning_rate": 0.001571263363182372, + "loss": 0.1191, + "step": 31836 + }, + { + "epoch": 0.2763604482599977, + "grad_norm": 0.322265625, + "learning_rate": 0.0015712380068511778, + "loss": 0.0791, + "step": 31837 + }, + { + "epoch": 0.2763691287402019, + "grad_norm": 0.234375, + "learning_rate": 0.001571212650004645, + "loss": 0.1045, + "step": 31838 + }, + { + "epoch": 0.27637780922040606, + "grad_norm": 0.42578125, + "learning_rate": 0.0015711872926428005, + "loss": 0.1182, + "step": 31839 + }, + { + "epoch": 0.27638648970061025, + "grad_norm": 0.1279296875, + "learning_rate": 0.001571161934765673, + "loss": 0.0889, + "step": 31840 + }, + { + "epoch": 0.2763951701808144, + "grad_norm": 0.2080078125, + "learning_rate": 0.0015711365763732897, + "loss": 0.1445, + "step": 31841 + }, + { + "epoch": 0.2764038506610186, + "grad_norm": 0.62890625, + "learning_rate": 0.0015711112174656787, + "loss": 0.1035, + "step": 31842 + }, + { + "epoch": 0.2764125311412227, + "grad_norm": 0.2734375, + "learning_rate": 0.0015710858580428675, + "loss": 0.0977, + "step": 31843 + }, + { + "epoch": 0.2764212116214269, + "grad_norm": 0.5078125, + "learning_rate": 0.001571060498104884, + "loss": 0.1152, + "step": 31844 + }, + { + "epoch": 0.27642989210163105, + "grad_norm": 0.4296875, + "learning_rate": 0.0015710351376517556, + "loss": 0.1758, + "step": 31845 + }, + { + "epoch": 0.27643857258183524, + "grad_norm": 0.349609375, + "learning_rate": 0.0015710097766835102, + "loss": 0.1377, + "step": 31846 + }, + { + "epoch": 0.2764472530620394, + "grad_norm": 0.0908203125, + "learning_rate": 0.0015709844152001758, + "loss": 0.0889, + "step": 31847 + }, + { + "epoch": 0.27645593354224357, + "grad_norm": 0.326171875, + "learning_rate": 0.00157095905320178, + "loss": 0.062, + "step": 31848 + }, + { + "epoch": 0.2764646140224477, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015709336906883503, + "loss": 0.0698, + "step": 31849 + }, + { + "epoch": 0.2764732945026519, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015709083276599148, + "loss": 0.0835, + "step": 31850 + }, + { + "epoch": 0.27648197498285604, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015708829641165004, + "loss": 0.062, + "step": 31851 + }, + { + "epoch": 0.27649065546306023, + "grad_norm": 0.376953125, + "learning_rate": 0.0015708576000581361, + "loss": 0.1582, + "step": 31852 + }, + { + "epoch": 0.27649933594326437, + "grad_norm": 0.1845703125, + "learning_rate": 0.0015708322354848485, + "loss": 0.1445, + "step": 31853 + }, + { + "epoch": 0.27650801642346856, + "grad_norm": 0.09912109375, + "learning_rate": 0.0015708068703966665, + "loss": 0.0781, + "step": 31854 + }, + { + "epoch": 0.2765166969036727, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015707815047936167, + "loss": 0.0835, + "step": 31855 + }, + { + "epoch": 0.2765253773838769, + "grad_norm": 0.294921875, + "learning_rate": 0.0015707561386757273, + "loss": 0.1709, + "step": 31856 + }, + { + "epoch": 0.27653405786408103, + "grad_norm": 0.08203125, + "learning_rate": 0.0015707307720430263, + "loss": 0.0889, + "step": 31857 + }, + { + "epoch": 0.2765427383442852, + "grad_norm": 0.2080078125, + "learning_rate": 0.001570705404895541, + "loss": 0.1245, + "step": 31858 + }, + { + "epoch": 0.27655141882448936, + "grad_norm": 0.173828125, + "learning_rate": 0.0015706800372332996, + "loss": 0.1094, + "step": 31859 + }, + { + "epoch": 0.27656009930469355, + "grad_norm": 0.859375, + "learning_rate": 0.0015706546690563295, + "loss": 0.1279, + "step": 31860 + }, + { + "epoch": 0.2765687797848977, + "grad_norm": 0.33984375, + "learning_rate": 0.0015706293003646583, + "loss": 0.1191, + "step": 31861 + }, + { + "epoch": 0.2765774602651019, + "grad_norm": 0.11865234375, + "learning_rate": 0.001570603931158314, + "loss": 0.1094, + "step": 31862 + }, + { + "epoch": 0.276586140745306, + "grad_norm": 0.625, + "learning_rate": 0.0015705785614373246, + "loss": 0.1025, + "step": 31863 + }, + { + "epoch": 0.2765948212255102, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015705531912017171, + "loss": 0.0869, + "step": 31864 + }, + { + "epoch": 0.27660350170571435, + "grad_norm": 0.1572265625, + "learning_rate": 0.00157052782045152, + "loss": 0.0698, + "step": 31865 + }, + { + "epoch": 0.27661218218591854, + "grad_norm": 0.36328125, + "learning_rate": 0.001570502449186761, + "loss": 0.1045, + "step": 31866 + }, + { + "epoch": 0.2766208626661227, + "grad_norm": 0.5390625, + "learning_rate": 0.0015704770774074672, + "loss": 0.0967, + "step": 31867 + }, + { + "epoch": 0.2766295431463269, + "grad_norm": 0.27734375, + "learning_rate": 0.001570451705113667, + "loss": 0.0972, + "step": 31868 + }, + { + "epoch": 0.276638223626531, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015704263323053876, + "loss": 0.1011, + "step": 31869 + }, + { + "epoch": 0.2766469041067352, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015704009589826574, + "loss": 0.1235, + "step": 31870 + }, + { + "epoch": 0.27665558458693934, + "grad_norm": 0.60546875, + "learning_rate": 0.0015703755851455032, + "loss": 0.1279, + "step": 31871 + }, + { + "epoch": 0.27666426506714353, + "grad_norm": 0.361328125, + "learning_rate": 0.0015703502107939538, + "loss": 0.1133, + "step": 31872 + }, + { + "epoch": 0.27667294554734767, + "grad_norm": 0.068359375, + "learning_rate": 0.0015703248359280365, + "loss": 0.0972, + "step": 31873 + }, + { + "epoch": 0.27668162602755186, + "grad_norm": 0.146484375, + "learning_rate": 0.0015702994605477788, + "loss": 0.084, + "step": 31874 + }, + { + "epoch": 0.276690306507756, + "grad_norm": 0.328125, + "learning_rate": 0.0015702740846532087, + "loss": 0.1172, + "step": 31875 + }, + { + "epoch": 0.2766989869879602, + "grad_norm": 0.234375, + "learning_rate": 0.001570248708244354, + "loss": 0.0947, + "step": 31876 + }, + { + "epoch": 0.27670766746816433, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015702233313212423, + "loss": 0.1089, + "step": 31877 + }, + { + "epoch": 0.2767163479483685, + "grad_norm": 0.3671875, + "learning_rate": 0.0015701979538839017, + "loss": 0.1201, + "step": 31878 + }, + { + "epoch": 0.27672502842857266, + "grad_norm": 0.7890625, + "learning_rate": 0.0015701725759323592, + "loss": 0.1064, + "step": 31879 + }, + { + "epoch": 0.27673370890877685, + "grad_norm": 0.474609375, + "learning_rate": 0.0015701471974666433, + "loss": 0.1187, + "step": 31880 + }, + { + "epoch": 0.276742389388981, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015701218184867817, + "loss": 0.0762, + "step": 31881 + }, + { + "epoch": 0.2767510698691852, + "grad_norm": 0.453125, + "learning_rate": 0.0015700964389928017, + "loss": 0.0991, + "step": 31882 + }, + { + "epoch": 0.2767597503493893, + "grad_norm": 0.30859375, + "learning_rate": 0.0015700710589847315, + "loss": 0.1475, + "step": 31883 + }, + { + "epoch": 0.2767684308295935, + "grad_norm": 0.1123046875, + "learning_rate": 0.0015700456784625983, + "loss": 0.1118, + "step": 31884 + }, + { + "epoch": 0.27677711130979765, + "grad_norm": 0.33984375, + "learning_rate": 0.0015700202974264304, + "loss": 0.1289, + "step": 31885 + }, + { + "epoch": 0.27678579179000184, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015699949158762556, + "loss": 0.0854, + "step": 31886 + }, + { + "epoch": 0.276794472270206, + "grad_norm": 0.375, + "learning_rate": 0.0015699695338121011, + "loss": 0.1025, + "step": 31887 + }, + { + "epoch": 0.2768031527504102, + "grad_norm": 0.236328125, + "learning_rate": 0.001569944151233995, + "loss": 0.1094, + "step": 31888 + }, + { + "epoch": 0.2768118332306143, + "grad_norm": 0.4453125, + "learning_rate": 0.0015699187681419652, + "loss": 0.0996, + "step": 31889 + }, + { + "epoch": 0.2768205137108185, + "grad_norm": 0.21484375, + "learning_rate": 0.001569893384536039, + "loss": 0.1045, + "step": 31890 + }, + { + "epoch": 0.27682919419102264, + "grad_norm": 0.43359375, + "learning_rate": 0.0015698680004162446, + "loss": 0.1592, + "step": 31891 + }, + { + "epoch": 0.27683787467122684, + "grad_norm": 0.14453125, + "learning_rate": 0.0015698426157826095, + "loss": 0.0894, + "step": 31892 + }, + { + "epoch": 0.276846555151431, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015698172306351618, + "loss": 0.1289, + "step": 31893 + }, + { + "epoch": 0.27685523563163517, + "grad_norm": 0.32421875, + "learning_rate": 0.0015697918449739292, + "loss": 0.106, + "step": 31894 + }, + { + "epoch": 0.2768639161118393, + "grad_norm": 0.220703125, + "learning_rate": 0.0015697664587989392, + "loss": 0.0801, + "step": 31895 + }, + { + "epoch": 0.2768725965920435, + "grad_norm": 0.1552734375, + "learning_rate": 0.0015697410721102194, + "loss": 0.1328, + "step": 31896 + }, + { + "epoch": 0.27688127707224763, + "grad_norm": 0.12353515625, + "learning_rate": 0.001569715684907798, + "loss": 0.126, + "step": 31897 + }, + { + "epoch": 0.2768899575524518, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015696902971917026, + "loss": 0.1152, + "step": 31898 + }, + { + "epoch": 0.27689863803265596, + "grad_norm": 0.51171875, + "learning_rate": 0.0015696649089619605, + "loss": 0.0737, + "step": 31899 + }, + { + "epoch": 0.27690731851286016, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015696395202186007, + "loss": 0.0884, + "step": 31900 + }, + { + "epoch": 0.2769159989930643, + "grad_norm": 0.60546875, + "learning_rate": 0.0015696141309616497, + "loss": 0.1045, + "step": 31901 + }, + { + "epoch": 0.2769246794732685, + "grad_norm": 0.734375, + "learning_rate": 0.0015695887411911358, + "loss": 0.1309, + "step": 31902 + }, + { + "epoch": 0.2769333599534726, + "grad_norm": 0.263671875, + "learning_rate": 0.0015695633509070865, + "loss": 0.1465, + "step": 31903 + }, + { + "epoch": 0.2769420404336768, + "grad_norm": 0.115234375, + "learning_rate": 0.00156953796010953, + "loss": 0.1201, + "step": 31904 + }, + { + "epoch": 0.27695072091388095, + "grad_norm": 0.251953125, + "learning_rate": 0.0015695125687984939, + "loss": 0.0698, + "step": 31905 + }, + { + "epoch": 0.27695940139408515, + "grad_norm": 0.1953125, + "learning_rate": 0.001569487176974006, + "loss": 0.064, + "step": 31906 + }, + { + "epoch": 0.2769680818742893, + "grad_norm": 0.27734375, + "learning_rate": 0.0015694617846360936, + "loss": 0.1211, + "step": 31907 + }, + { + "epoch": 0.2769767623544935, + "grad_norm": 0.53125, + "learning_rate": 0.0015694363917847854, + "loss": 0.1289, + "step": 31908 + }, + { + "epoch": 0.2769854428346976, + "grad_norm": 0.47265625, + "learning_rate": 0.0015694109984201083, + "loss": 0.1011, + "step": 31909 + }, + { + "epoch": 0.2769941233149018, + "grad_norm": 0.453125, + "learning_rate": 0.00156938560454209, + "loss": 0.1289, + "step": 31910 + }, + { + "epoch": 0.27700280379510595, + "grad_norm": 0.29296875, + "learning_rate": 0.001569360210150759, + "loss": 0.0889, + "step": 31911 + }, + { + "epoch": 0.2770114842753101, + "grad_norm": 0.255859375, + "learning_rate": 0.001569334815246143, + "loss": 0.1191, + "step": 31912 + }, + { + "epoch": 0.2770201647555143, + "grad_norm": 0.8203125, + "learning_rate": 0.0015693094198282694, + "loss": 0.2432, + "step": 31913 + }, + { + "epoch": 0.2770288452357184, + "grad_norm": 0.248046875, + "learning_rate": 0.0015692840238971657, + "loss": 0.0938, + "step": 31914 + }, + { + "epoch": 0.2770375257159226, + "grad_norm": 0.66796875, + "learning_rate": 0.0015692586274528603, + "loss": 0.1133, + "step": 31915 + }, + { + "epoch": 0.27704620619612674, + "grad_norm": 0.291015625, + "learning_rate": 0.001569233230495381, + "loss": 0.0747, + "step": 31916 + }, + { + "epoch": 0.27705488667633094, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015692078330247548, + "loss": 0.1006, + "step": 31917 + }, + { + "epoch": 0.2770635671565351, + "grad_norm": 0.1298828125, + "learning_rate": 0.00156918243504101, + "loss": 0.1328, + "step": 31918 + }, + { + "epoch": 0.27707224763673927, + "grad_norm": 0.130859375, + "learning_rate": 0.0015691570365441744, + "loss": 0.0962, + "step": 31919 + }, + { + "epoch": 0.2770809281169434, + "grad_norm": 0.361328125, + "learning_rate": 0.0015691316375342759, + "loss": 0.0669, + "step": 31920 + }, + { + "epoch": 0.2770896085971476, + "grad_norm": 0.369140625, + "learning_rate": 0.001569106238011342, + "loss": 0.1074, + "step": 31921 + }, + { + "epoch": 0.27709828907735173, + "grad_norm": 0.1416015625, + "learning_rate": 0.0015690808379754007, + "loss": 0.1162, + "step": 31922 + }, + { + "epoch": 0.2771069695575559, + "grad_norm": 0.1025390625, + "learning_rate": 0.0015690554374264791, + "loss": 0.1191, + "step": 31923 + }, + { + "epoch": 0.27711565003776006, + "grad_norm": 0.08203125, + "learning_rate": 0.0015690300363646061, + "loss": 0.1011, + "step": 31924 + }, + { + "epoch": 0.27712433051796426, + "grad_norm": 0.2373046875, + "learning_rate": 0.0015690046347898088, + "loss": 0.0913, + "step": 31925 + }, + { + "epoch": 0.2771330109981684, + "grad_norm": 0.330078125, + "learning_rate": 0.001568979232702115, + "loss": 0.1309, + "step": 31926 + }, + { + "epoch": 0.2771416914783726, + "grad_norm": 0.12890625, + "learning_rate": 0.0015689538301015527, + "loss": 0.084, + "step": 31927 + }, + { + "epoch": 0.2771503719585767, + "grad_norm": 0.240234375, + "learning_rate": 0.0015689284269881489, + "loss": 0.1133, + "step": 31928 + }, + { + "epoch": 0.2771590524387809, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015689030233619328, + "loss": 0.0928, + "step": 31929 + }, + { + "epoch": 0.27716773291898505, + "grad_norm": 0.140625, + "learning_rate": 0.001568877619222931, + "loss": 0.1396, + "step": 31930 + }, + { + "epoch": 0.27717641339918925, + "grad_norm": 0.279296875, + "learning_rate": 0.0015688522145711717, + "loss": 0.0928, + "step": 31931 + }, + { + "epoch": 0.2771850938793934, + "grad_norm": 0.27734375, + "learning_rate": 0.0015688268094066827, + "loss": 0.0869, + "step": 31932 + }, + { + "epoch": 0.2771937743595976, + "grad_norm": 0.0947265625, + "learning_rate": 0.001568801403729492, + "loss": 0.1279, + "step": 31933 + }, + { + "epoch": 0.2772024548398017, + "grad_norm": 0.322265625, + "learning_rate": 0.0015687759975396268, + "loss": 0.0713, + "step": 31934 + }, + { + "epoch": 0.2772111353200059, + "grad_norm": 0.6484375, + "learning_rate": 0.0015687505908371154, + "loss": 0.1582, + "step": 31935 + }, + { + "epoch": 0.27721981580021005, + "grad_norm": 0.353515625, + "learning_rate": 0.0015687251836219852, + "loss": 0.1465, + "step": 31936 + }, + { + "epoch": 0.27722849628041424, + "grad_norm": 0.1220703125, + "learning_rate": 0.0015686997758942642, + "loss": 0.1079, + "step": 31937 + }, + { + "epoch": 0.2772371767606184, + "grad_norm": 0.185546875, + "learning_rate": 0.0015686743676539805, + "loss": 0.1016, + "step": 31938 + }, + { + "epoch": 0.27724585724082257, + "grad_norm": 0.1826171875, + "learning_rate": 0.0015686489589011611, + "loss": 0.1211, + "step": 31939 + }, + { + "epoch": 0.2772545377210267, + "grad_norm": 0.1142578125, + "learning_rate": 0.0015686235496358346, + "loss": 0.1006, + "step": 31940 + }, + { + "epoch": 0.2772632182012309, + "grad_norm": 0.25, + "learning_rate": 0.0015685981398580282, + "loss": 0.0996, + "step": 31941 + }, + { + "epoch": 0.27727189868143504, + "grad_norm": 0.310546875, + "learning_rate": 0.00156857272956777, + "loss": 0.1138, + "step": 31942 + }, + { + "epoch": 0.27728057916163923, + "grad_norm": 0.310546875, + "learning_rate": 0.001568547318765088, + "loss": 0.1357, + "step": 31943 + }, + { + "epoch": 0.27728925964184337, + "grad_norm": 0.2373046875, + "learning_rate": 0.0015685219074500093, + "loss": 0.0898, + "step": 31944 + }, + { + "epoch": 0.27729794012204756, + "grad_norm": 0.1220703125, + "learning_rate": 0.0015684964956225623, + "loss": 0.1543, + "step": 31945 + }, + { + "epoch": 0.2773066206022517, + "grad_norm": 0.189453125, + "learning_rate": 0.0015684710832827744, + "loss": 0.0791, + "step": 31946 + }, + { + "epoch": 0.2773153010824559, + "grad_norm": 0.1240234375, + "learning_rate": 0.0015684456704306736, + "loss": 0.1099, + "step": 31947 + }, + { + "epoch": 0.27732398156266, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015684202570662876, + "loss": 0.1152, + "step": 31948 + }, + { + "epoch": 0.2773326620428642, + "grad_norm": 0.1416015625, + "learning_rate": 0.0015683948431896445, + "loss": 0.0947, + "step": 31949 + }, + { + "epoch": 0.27734134252306836, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015683694288007712, + "loss": 0.0962, + "step": 31950 + }, + { + "epoch": 0.27735002300327255, + "grad_norm": 0.8828125, + "learning_rate": 0.0015683440138996969, + "loss": 0.1226, + "step": 31951 + }, + { + "epoch": 0.2773587034834767, + "grad_norm": 0.412109375, + "learning_rate": 0.0015683185984864483, + "loss": 0.1582, + "step": 31952 + }, + { + "epoch": 0.2773673839636809, + "grad_norm": 0.140625, + "learning_rate": 0.0015682931825610534, + "loss": 0.1436, + "step": 31953 + }, + { + "epoch": 0.277376064443885, + "grad_norm": 0.37890625, + "learning_rate": 0.0015682677661235404, + "loss": 0.1084, + "step": 31954 + }, + { + "epoch": 0.2773847449240892, + "grad_norm": 0.072265625, + "learning_rate": 0.0015682423491739366, + "loss": 0.0986, + "step": 31955 + }, + { + "epoch": 0.27739342540429335, + "grad_norm": 0.27734375, + "learning_rate": 0.0015682169317122702, + "loss": 0.1094, + "step": 31956 + }, + { + "epoch": 0.27740210588449754, + "grad_norm": 0.53515625, + "learning_rate": 0.0015681915137385682, + "loss": 0.1177, + "step": 31957 + }, + { + "epoch": 0.2774107863647017, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015681660952528596, + "loss": 0.0859, + "step": 31958 + }, + { + "epoch": 0.27741946684490587, + "grad_norm": 0.296875, + "learning_rate": 0.0015681406762551714, + "loss": 0.1191, + "step": 31959 + }, + { + "epoch": 0.27742814732511, + "grad_norm": 0.212890625, + "learning_rate": 0.0015681152567455317, + "loss": 0.0913, + "step": 31960 + }, + { + "epoch": 0.2774368278053142, + "grad_norm": 0.31640625, + "learning_rate": 0.0015680898367239683, + "loss": 0.0918, + "step": 31961 + }, + { + "epoch": 0.27744550828551834, + "grad_norm": 0.166015625, + "learning_rate": 0.0015680644161905084, + "loss": 0.0737, + "step": 31962 + }, + { + "epoch": 0.27745418876572253, + "grad_norm": 0.2578125, + "learning_rate": 0.0015680389951451808, + "loss": 0.1143, + "step": 31963 + }, + { + "epoch": 0.27746286924592667, + "grad_norm": 0.42578125, + "learning_rate": 0.0015680135735880124, + "loss": 0.1201, + "step": 31964 + }, + { + "epoch": 0.27747154972613086, + "grad_norm": 0.490234375, + "learning_rate": 0.0015679881515190317, + "loss": 0.1006, + "step": 31965 + }, + { + "epoch": 0.277480230206335, + "grad_norm": 0.58203125, + "learning_rate": 0.0015679627289382659, + "loss": 0.1074, + "step": 31966 + }, + { + "epoch": 0.2774889106865392, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015679373058457432, + "loss": 0.124, + "step": 31967 + }, + { + "epoch": 0.27749759116674333, + "grad_norm": 0.25390625, + "learning_rate": 0.0015679118822414912, + "loss": 0.1045, + "step": 31968 + }, + { + "epoch": 0.2775062716469475, + "grad_norm": 0.57421875, + "learning_rate": 0.001567886458125538, + "loss": 0.1104, + "step": 31969 + }, + { + "epoch": 0.27751495212715166, + "grad_norm": 0.296875, + "learning_rate": 0.001567861033497911, + "loss": 0.0928, + "step": 31970 + }, + { + "epoch": 0.27752363260735585, + "grad_norm": 0.318359375, + "learning_rate": 0.0015678356083586385, + "loss": 0.0898, + "step": 31971 + }, + { + "epoch": 0.27753231308756, + "grad_norm": 0.69140625, + "learning_rate": 0.001567810182707748, + "loss": 0.1152, + "step": 31972 + }, + { + "epoch": 0.2775409935677642, + "grad_norm": 0.140625, + "learning_rate": 0.001567784756545267, + "loss": 0.0991, + "step": 31973 + }, + { + "epoch": 0.2775496740479683, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015677593298712238, + "loss": 0.1465, + "step": 31974 + }, + { + "epoch": 0.2775583545281725, + "grad_norm": 0.1962890625, + "learning_rate": 0.001567733902685646, + "loss": 0.1045, + "step": 31975 + }, + { + "epoch": 0.27756703500837665, + "grad_norm": 0.326171875, + "learning_rate": 0.0015677084749885614, + "loss": 0.1035, + "step": 31976 + }, + { + "epoch": 0.27757571548858084, + "grad_norm": 0.11767578125, + "learning_rate": 0.001567683046779998, + "loss": 0.1045, + "step": 31977 + }, + { + "epoch": 0.277584395968785, + "grad_norm": 0.32421875, + "learning_rate": 0.0015676576180599833, + "loss": 0.0918, + "step": 31978 + }, + { + "epoch": 0.2775930764489892, + "grad_norm": 0.244140625, + "learning_rate": 0.0015676321888285453, + "loss": 0.168, + "step": 31979 + }, + { + "epoch": 0.2776017569291933, + "grad_norm": 0.3203125, + "learning_rate": 0.0015676067590857116, + "loss": 0.0742, + "step": 31980 + }, + { + "epoch": 0.2776104374093975, + "grad_norm": 0.1083984375, + "learning_rate": 0.0015675813288315106, + "loss": 0.1074, + "step": 31981 + }, + { + "epoch": 0.27761911788960164, + "grad_norm": 0.3984375, + "learning_rate": 0.0015675558980659692, + "loss": 0.0986, + "step": 31982 + }, + { + "epoch": 0.27762779836980583, + "grad_norm": 0.21484375, + "learning_rate": 0.001567530466789116, + "loss": 0.0938, + "step": 31983 + }, + { + "epoch": 0.27763647885000997, + "grad_norm": 0.34375, + "learning_rate": 0.001567505035000978, + "loss": 0.0938, + "step": 31984 + }, + { + "epoch": 0.27764515933021416, + "grad_norm": 0.71875, + "learning_rate": 0.001567479602701584, + "loss": 0.2734, + "step": 31985 + }, + { + "epoch": 0.2776538398104183, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015674541698909614, + "loss": 0.0791, + "step": 31986 + }, + { + "epoch": 0.2776625202906225, + "grad_norm": 0.166015625, + "learning_rate": 0.0015674287365691378, + "loss": 0.104, + "step": 31987 + }, + { + "epoch": 0.27767120077082663, + "grad_norm": 0.212890625, + "learning_rate": 0.0015674033027361413, + "loss": 0.0859, + "step": 31988 + }, + { + "epoch": 0.2776798812510308, + "grad_norm": 0.1875, + "learning_rate": 0.001567377868391999, + "loss": 0.0947, + "step": 31989 + }, + { + "epoch": 0.27768856173123496, + "grad_norm": 1.2578125, + "learning_rate": 0.0015673524335367399, + "loss": 0.1211, + "step": 31990 + }, + { + "epoch": 0.27769724221143915, + "grad_norm": 0.0966796875, + "learning_rate": 0.0015673269981703914, + "loss": 0.1523, + "step": 31991 + }, + { + "epoch": 0.2777059226916433, + "grad_norm": 0.421875, + "learning_rate": 0.0015673015622929803, + "loss": 0.1113, + "step": 31992 + }, + { + "epoch": 0.2777146031718475, + "grad_norm": 0.67578125, + "learning_rate": 0.0015672761259045355, + "loss": 0.1797, + "step": 31993 + }, + { + "epoch": 0.2777232836520516, + "grad_norm": 0.51953125, + "learning_rate": 0.0015672506890050846, + "loss": 0.0908, + "step": 31994 + }, + { + "epoch": 0.2777319641322558, + "grad_norm": 0.60546875, + "learning_rate": 0.0015672252515946557, + "loss": 0.1582, + "step": 31995 + }, + { + "epoch": 0.27774064461245995, + "grad_norm": 0.2734375, + "learning_rate": 0.001567199813673276, + "loss": 0.0942, + "step": 31996 + }, + { + "epoch": 0.27774932509266415, + "grad_norm": 1.203125, + "learning_rate": 0.0015671743752409737, + "loss": 0.3672, + "step": 31997 + }, + { + "epoch": 0.2777580055728683, + "grad_norm": 0.201171875, + "learning_rate": 0.0015671489362977767, + "loss": 0.0854, + "step": 31998 + }, + { + "epoch": 0.2777666860530725, + "grad_norm": 0.1953125, + "learning_rate": 0.0015671234968437121, + "loss": 0.0967, + "step": 31999 + }, + { + "epoch": 0.2777753665332766, + "grad_norm": 0.087890625, + "learning_rate": 0.0015670980568788086, + "loss": 0.1011, + "step": 32000 + }, + { + "epoch": 0.2777840470134808, + "grad_norm": 0.27734375, + "learning_rate": 0.0015670726164030938, + "loss": 0.0918, + "step": 32001 + }, + { + "epoch": 0.27779272749368494, + "grad_norm": 0.63671875, + "learning_rate": 0.0015670471754165951, + "loss": 0.103, + "step": 32002 + }, + { + "epoch": 0.27780140797388914, + "grad_norm": 0.14453125, + "learning_rate": 0.001567021733919341, + "loss": 0.1162, + "step": 32003 + }, + { + "epoch": 0.2778100884540933, + "grad_norm": 0.22265625, + "learning_rate": 0.0015669962919113586, + "loss": 0.103, + "step": 32004 + }, + { + "epoch": 0.27781876893429747, + "grad_norm": 0.08837890625, + "learning_rate": 0.001566970849392676, + "loss": 0.0898, + "step": 32005 + }, + { + "epoch": 0.2778274494145016, + "grad_norm": 0.1123046875, + "learning_rate": 0.0015669454063633212, + "loss": 0.1211, + "step": 32006 + }, + { + "epoch": 0.2778361298947058, + "grad_norm": 0.099609375, + "learning_rate": 0.0015669199628233223, + "loss": 0.0757, + "step": 32007 + }, + { + "epoch": 0.27784481037490993, + "grad_norm": 0.515625, + "learning_rate": 0.0015668945187727063, + "loss": 0.1279, + "step": 32008 + }, + { + "epoch": 0.2778534908551141, + "grad_norm": 0.7265625, + "learning_rate": 0.0015668690742115018, + "loss": 0.1299, + "step": 32009 + }, + { + "epoch": 0.27786217133531826, + "grad_norm": 0.279296875, + "learning_rate": 0.001566843629139736, + "loss": 0.1064, + "step": 32010 + }, + { + "epoch": 0.27787085181552246, + "grad_norm": 0.2890625, + "learning_rate": 0.0015668181835574372, + "loss": 0.0864, + "step": 32011 + }, + { + "epoch": 0.2778795322957266, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015667927374646332, + "loss": 0.0952, + "step": 32012 + }, + { + "epoch": 0.2778882127759308, + "grad_norm": 0.37109375, + "learning_rate": 0.0015667672908613512, + "loss": 0.0928, + "step": 32013 + }, + { + "epoch": 0.2778968932561349, + "grad_norm": 0.2294921875, + "learning_rate": 0.00156674184374762, + "loss": 0.1377, + "step": 32014 + }, + { + "epoch": 0.2779055737363391, + "grad_norm": 0.470703125, + "learning_rate": 0.0015667163961234668, + "loss": 0.0957, + "step": 32015 + }, + { + "epoch": 0.27791425421654326, + "grad_norm": 1.078125, + "learning_rate": 0.0015666909479889193, + "loss": 0.1465, + "step": 32016 + }, + { + "epoch": 0.27792293469674745, + "grad_norm": 0.8359375, + "learning_rate": 0.0015666654993440059, + "loss": 0.4688, + "step": 32017 + }, + { + "epoch": 0.2779316151769516, + "grad_norm": 0.123046875, + "learning_rate": 0.001566640050188754, + "loss": 0.0928, + "step": 32018 + }, + { + "epoch": 0.2779402956571558, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015666146005231913, + "loss": 0.082, + "step": 32019 + }, + { + "epoch": 0.2779489761373599, + "grad_norm": 0.283203125, + "learning_rate": 0.0015665891503473463, + "loss": 0.1118, + "step": 32020 + }, + { + "epoch": 0.2779576566175641, + "grad_norm": 0.392578125, + "learning_rate": 0.0015665636996612465, + "loss": 0.1113, + "step": 32021 + }, + { + "epoch": 0.27796633709776825, + "grad_norm": 0.240234375, + "learning_rate": 0.0015665382484649191, + "loss": 0.1621, + "step": 32022 + }, + { + "epoch": 0.27797501757797244, + "grad_norm": 0.287109375, + "learning_rate": 0.0015665127967583927, + "loss": 0.0928, + "step": 32023 + }, + { + "epoch": 0.2779836980581766, + "grad_norm": 0.470703125, + "learning_rate": 0.001566487344541695, + "loss": 0.0898, + "step": 32024 + }, + { + "epoch": 0.27799237853838077, + "grad_norm": 1.28125, + "learning_rate": 0.0015664618918148538, + "loss": 0.1816, + "step": 32025 + }, + { + "epoch": 0.2780010590185849, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015664364385778967, + "loss": 0.0723, + "step": 32026 + }, + { + "epoch": 0.2780097394987891, + "grad_norm": 0.5703125, + "learning_rate": 0.001566410984830852, + "loss": 0.209, + "step": 32027 + }, + { + "epoch": 0.27801841997899324, + "grad_norm": 0.0703125, + "learning_rate": 0.001566385530573747, + "loss": 0.0835, + "step": 32028 + }, + { + "epoch": 0.27802710045919743, + "grad_norm": 0.1748046875, + "learning_rate": 0.0015663600758066098, + "loss": 0.1055, + "step": 32029 + }, + { + "epoch": 0.27803578093940157, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015663346205294687, + "loss": 0.0996, + "step": 32030 + }, + { + "epoch": 0.27804446141960576, + "grad_norm": 0.06884765625, + "learning_rate": 0.0015663091647423505, + "loss": 0.0747, + "step": 32031 + }, + { + "epoch": 0.2780531418998099, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015662837084452837, + "loss": 0.0869, + "step": 32032 + }, + { + "epoch": 0.2780618223800141, + "grad_norm": 0.625, + "learning_rate": 0.001566258251638296, + "loss": 0.123, + "step": 32033 + }, + { + "epoch": 0.2780705028602182, + "grad_norm": 0.390625, + "learning_rate": 0.0015662327943214154, + "loss": 0.1221, + "step": 32034 + }, + { + "epoch": 0.27807918334042236, + "grad_norm": 0.251953125, + "learning_rate": 0.0015662073364946696, + "loss": 0.1426, + "step": 32035 + }, + { + "epoch": 0.27808786382062656, + "grad_norm": 0.5234375, + "learning_rate": 0.0015661818781580867, + "loss": 0.0977, + "step": 32036 + }, + { + "epoch": 0.2780965443008307, + "grad_norm": 0.1416015625, + "learning_rate": 0.001566156419311694, + "loss": 0.1113, + "step": 32037 + }, + { + "epoch": 0.2781052247810349, + "grad_norm": 0.62890625, + "learning_rate": 0.0015661309599555198, + "loss": 0.1133, + "step": 32038 + }, + { + "epoch": 0.278113905261239, + "grad_norm": 0.373046875, + "learning_rate": 0.0015661055000895918, + "loss": 0.1201, + "step": 32039 + }, + { + "epoch": 0.2781225857414432, + "grad_norm": 0.162109375, + "learning_rate": 0.0015660800397139377, + "loss": 0.0786, + "step": 32040 + }, + { + "epoch": 0.27813126622164736, + "grad_norm": 0.125, + "learning_rate": 0.0015660545788285856, + "loss": 0.1147, + "step": 32041 + }, + { + "epoch": 0.27813994670185155, + "grad_norm": 0.84375, + "learning_rate": 0.001566029117433563, + "loss": 0.1128, + "step": 32042 + }, + { + "epoch": 0.2781486271820557, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015660036555288981, + "loss": 0.168, + "step": 32043 + }, + { + "epoch": 0.2781573076622599, + "grad_norm": 0.5234375, + "learning_rate": 0.0015659781931146185, + "loss": 0.0981, + "step": 32044 + }, + { + "epoch": 0.278165988142464, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015659527301907523, + "loss": 0.1387, + "step": 32045 + }, + { + "epoch": 0.2781746686226682, + "grad_norm": 0.404296875, + "learning_rate": 0.001565927266757327, + "loss": 0.082, + "step": 32046 + }, + { + "epoch": 0.27818334910287235, + "grad_norm": 0.375, + "learning_rate": 0.0015659018028143707, + "loss": 0.1006, + "step": 32047 + }, + { + "epoch": 0.27819202958307654, + "grad_norm": 0.6640625, + "learning_rate": 0.0015658763383619115, + "loss": 0.1025, + "step": 32048 + }, + { + "epoch": 0.2782007100632807, + "grad_norm": 0.294921875, + "learning_rate": 0.0015658508733999768, + "loss": 0.0854, + "step": 32049 + }, + { + "epoch": 0.27820939054348487, + "grad_norm": 0.091796875, + "learning_rate": 0.0015658254079285946, + "loss": 0.0991, + "step": 32050 + }, + { + "epoch": 0.278218071023689, + "grad_norm": 0.4375, + "learning_rate": 0.0015657999419477926, + "loss": 0.1133, + "step": 32051 + }, + { + "epoch": 0.2782267515038932, + "grad_norm": 0.375, + "learning_rate": 0.0015657744754575987, + "loss": 0.1348, + "step": 32052 + }, + { + "epoch": 0.27823543198409734, + "grad_norm": 0.322265625, + "learning_rate": 0.0015657490084580409, + "loss": 0.1279, + "step": 32053 + }, + { + "epoch": 0.27824411246430153, + "grad_norm": 0.2431640625, + "learning_rate": 0.001565723540949147, + "loss": 0.1309, + "step": 32054 + }, + { + "epoch": 0.27825279294450567, + "grad_norm": 0.10546875, + "learning_rate": 0.0015656980729309448, + "loss": 0.1084, + "step": 32055 + }, + { + "epoch": 0.27826147342470986, + "grad_norm": 0.32421875, + "learning_rate": 0.0015656726044034623, + "loss": 0.0889, + "step": 32056 + }, + { + "epoch": 0.278270153904914, + "grad_norm": 0.310546875, + "learning_rate": 0.0015656471353667272, + "loss": 0.0669, + "step": 32057 + }, + { + "epoch": 0.2782788343851182, + "grad_norm": 0.15234375, + "learning_rate": 0.0015656216658207673, + "loss": 0.0879, + "step": 32058 + }, + { + "epoch": 0.2782875148653223, + "grad_norm": 0.392578125, + "learning_rate": 0.0015655961957656108, + "loss": 0.1133, + "step": 32059 + }, + { + "epoch": 0.2782961953455265, + "grad_norm": 0.119140625, + "learning_rate": 0.0015655707252012852, + "loss": 0.1147, + "step": 32060 + }, + { + "epoch": 0.27830487582573066, + "grad_norm": 0.271484375, + "learning_rate": 0.0015655452541278182, + "loss": 0.1172, + "step": 32061 + }, + { + "epoch": 0.27831355630593485, + "grad_norm": 0.5234375, + "learning_rate": 0.001565519782545238, + "loss": 0.1123, + "step": 32062 + }, + { + "epoch": 0.278322236786139, + "grad_norm": 0.10400390625, + "learning_rate": 0.001565494310453572, + "loss": 0.0981, + "step": 32063 + }, + { + "epoch": 0.2783309172663432, + "grad_norm": 0.1982421875, + "learning_rate": 0.0015654688378528494, + "loss": 0.0869, + "step": 32064 + }, + { + "epoch": 0.2783395977465473, + "grad_norm": 0.302734375, + "learning_rate": 0.001565443364743096, + "loss": 0.1016, + "step": 32065 + }, + { + "epoch": 0.2783482782267515, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015654178911243416, + "loss": 0.1465, + "step": 32066 + }, + { + "epoch": 0.27835695870695565, + "grad_norm": 0.055419921875, + "learning_rate": 0.0015653924169966127, + "loss": 0.0625, + "step": 32067 + }, + { + "epoch": 0.27836563918715984, + "grad_norm": 0.244140625, + "learning_rate": 0.001565366942359938, + "loss": 0.1084, + "step": 32068 + }, + { + "epoch": 0.278374319667364, + "grad_norm": 0.375, + "learning_rate": 0.0015653414672143447, + "loss": 0.1221, + "step": 32069 + }, + { + "epoch": 0.27838300014756817, + "grad_norm": 0.609375, + "learning_rate": 0.0015653159915598612, + "loss": 0.0928, + "step": 32070 + }, + { + "epoch": 0.2783916806277723, + "grad_norm": 0.1396484375, + "learning_rate": 0.0015652905153965147, + "loss": 0.1133, + "step": 32071 + }, + { + "epoch": 0.2784003611079765, + "grad_norm": 0.52734375, + "learning_rate": 0.0015652650387243338, + "loss": 0.1152, + "step": 32072 + }, + { + "epoch": 0.27840904158818064, + "grad_norm": 0.474609375, + "learning_rate": 0.0015652395615433458, + "loss": 0.1187, + "step": 32073 + }, + { + "epoch": 0.27841772206838483, + "grad_norm": 0.142578125, + "learning_rate": 0.0015652140838535791, + "loss": 0.1914, + "step": 32074 + }, + { + "epoch": 0.27842640254858897, + "grad_norm": 0.306640625, + "learning_rate": 0.001565188605655061, + "loss": 0.103, + "step": 32075 + }, + { + "epoch": 0.27843508302879316, + "grad_norm": 0.1484375, + "learning_rate": 0.0015651631269478202, + "loss": 0.1094, + "step": 32076 + }, + { + "epoch": 0.2784437635089973, + "grad_norm": 0.271484375, + "learning_rate": 0.0015651376477318834, + "loss": 0.1104, + "step": 32077 + }, + { + "epoch": 0.2784524439892015, + "grad_norm": 0.24609375, + "learning_rate": 0.0015651121680072793, + "loss": 0.0967, + "step": 32078 + }, + { + "epoch": 0.27846112446940563, + "grad_norm": 0.44921875, + "learning_rate": 0.0015650866877740355, + "loss": 0.0811, + "step": 32079 + }, + { + "epoch": 0.2784698049496098, + "grad_norm": 0.421875, + "learning_rate": 0.0015650612070321798, + "loss": 0.1348, + "step": 32080 + }, + { + "epoch": 0.27847848542981396, + "grad_norm": 0.4921875, + "learning_rate": 0.00156503572578174, + "loss": 0.1147, + "step": 32081 + }, + { + "epoch": 0.27848716591001815, + "grad_norm": 0.287109375, + "learning_rate": 0.0015650102440227444, + "loss": 0.1787, + "step": 32082 + }, + { + "epoch": 0.2784958463902223, + "grad_norm": 0.130859375, + "learning_rate": 0.0015649847617552205, + "loss": 0.1152, + "step": 32083 + }, + { + "epoch": 0.2785045268704265, + "grad_norm": 0.09423828125, + "learning_rate": 0.0015649592789791962, + "loss": 0.0928, + "step": 32084 + }, + { + "epoch": 0.2785132073506306, + "grad_norm": 0.392578125, + "learning_rate": 0.0015649337956946997, + "loss": 0.1045, + "step": 32085 + }, + { + "epoch": 0.2785218878308348, + "grad_norm": 0.490234375, + "learning_rate": 0.0015649083119017582, + "loss": 0.085, + "step": 32086 + }, + { + "epoch": 0.27853056831103895, + "grad_norm": 0.3046875, + "learning_rate": 0.0015648828276004, + "loss": 0.0864, + "step": 32087 + }, + { + "epoch": 0.27853924879124314, + "grad_norm": 0.36328125, + "learning_rate": 0.001564857342790653, + "loss": 0.1064, + "step": 32088 + }, + { + "epoch": 0.2785479292714473, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015648318574725453, + "loss": 0.0806, + "step": 32089 + }, + { + "epoch": 0.2785566097516515, + "grad_norm": 0.291015625, + "learning_rate": 0.001564806371646104, + "loss": 0.0825, + "step": 32090 + }, + { + "epoch": 0.2785652902318556, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015647808853113576, + "loss": 0.0869, + "step": 32091 + }, + { + "epoch": 0.2785739707120598, + "grad_norm": 0.193359375, + "learning_rate": 0.0015647553984683338, + "loss": 0.0908, + "step": 32092 + }, + { + "epoch": 0.27858265119226394, + "grad_norm": 0.3125, + "learning_rate": 0.0015647299111170604, + "loss": 0.1504, + "step": 32093 + }, + { + "epoch": 0.27859133167246813, + "grad_norm": 0.318359375, + "learning_rate": 0.0015647044232575658, + "loss": 0.0767, + "step": 32094 + }, + { + "epoch": 0.27860001215267227, + "grad_norm": 0.412109375, + "learning_rate": 0.0015646789348898768, + "loss": 0.1514, + "step": 32095 + }, + { + "epoch": 0.27860869263287646, + "grad_norm": 0.2099609375, + "learning_rate": 0.0015646534460140222, + "loss": 0.1484, + "step": 32096 + }, + { + "epoch": 0.2786173731130806, + "grad_norm": 0.462890625, + "learning_rate": 0.0015646279566300296, + "loss": 0.1177, + "step": 32097 + }, + { + "epoch": 0.2786260535932848, + "grad_norm": 0.37890625, + "learning_rate": 0.0015646024667379267, + "loss": 0.1055, + "step": 32098 + }, + { + "epoch": 0.27863473407348893, + "grad_norm": 0.55859375, + "learning_rate": 0.0015645769763377418, + "loss": 0.1157, + "step": 32099 + }, + { + "epoch": 0.2786434145536931, + "grad_norm": 0.67578125, + "learning_rate": 0.0015645514854295023, + "loss": 0.1157, + "step": 32100 + }, + { + "epoch": 0.27865209503389726, + "grad_norm": 0.48046875, + "learning_rate": 0.001564525994013236, + "loss": 0.0957, + "step": 32101 + }, + { + "epoch": 0.27866077551410146, + "grad_norm": 0.60546875, + "learning_rate": 0.001564500502088972, + "loss": 0.1582, + "step": 32102 + }, + { + "epoch": 0.2786694559943056, + "grad_norm": 0.3515625, + "learning_rate": 0.0015644750096567364, + "loss": 0.1025, + "step": 32103 + }, + { + "epoch": 0.2786781364745098, + "grad_norm": 0.208984375, + "learning_rate": 0.001564449516716558, + "loss": 0.0801, + "step": 32104 + }, + { + "epoch": 0.2786868169547139, + "grad_norm": 0.62109375, + "learning_rate": 0.0015644240232684649, + "loss": 0.1719, + "step": 32105 + }, + { + "epoch": 0.2786954974349181, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015643985293124843, + "loss": 0.0762, + "step": 32106 + }, + { + "epoch": 0.27870417791512225, + "grad_norm": 0.33203125, + "learning_rate": 0.0015643730348486448, + "loss": 0.0718, + "step": 32107 + }, + { + "epoch": 0.27871285839532645, + "grad_norm": 0.4140625, + "learning_rate": 0.0015643475398769736, + "loss": 0.1055, + "step": 32108 + }, + { + "epoch": 0.2787215388755306, + "grad_norm": 0.189453125, + "learning_rate": 0.001564322044397499, + "loss": 0.1172, + "step": 32109 + }, + { + "epoch": 0.2787302193557348, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015642965484102486, + "loss": 0.1611, + "step": 32110 + }, + { + "epoch": 0.2787388998359389, + "grad_norm": 0.1796875, + "learning_rate": 0.001564271051915251, + "loss": 0.0957, + "step": 32111 + }, + { + "epoch": 0.2787475803161431, + "grad_norm": 0.0869140625, + "learning_rate": 0.0015642455549125332, + "loss": 0.123, + "step": 32112 + }, + { + "epoch": 0.27875626079634724, + "grad_norm": 0.466796875, + "learning_rate": 0.0015642200574021233, + "loss": 0.1182, + "step": 32113 + }, + { + "epoch": 0.27876494127655144, + "grad_norm": 0.546875, + "learning_rate": 0.0015641945593840496, + "loss": 0.1113, + "step": 32114 + }, + { + "epoch": 0.2787736217567556, + "grad_norm": 0.48828125, + "learning_rate": 0.0015641690608583397, + "loss": 0.1367, + "step": 32115 + }, + { + "epoch": 0.27878230223695977, + "grad_norm": 0.2470703125, + "learning_rate": 0.0015641435618250212, + "loss": 0.1211, + "step": 32116 + }, + { + "epoch": 0.2787909827171639, + "grad_norm": 0.08447265625, + "learning_rate": 0.0015641180622841225, + "loss": 0.0898, + "step": 32117 + }, + { + "epoch": 0.2787996631973681, + "grad_norm": 0.11181640625, + "learning_rate": 0.0015640925622356712, + "loss": 0.1182, + "step": 32118 + }, + { + "epoch": 0.27880834367757223, + "grad_norm": 0.435546875, + "learning_rate": 0.0015640670616796954, + "loss": 0.083, + "step": 32119 + }, + { + "epoch": 0.2788170241577764, + "grad_norm": 0.07958984375, + "learning_rate": 0.0015640415606162229, + "loss": 0.0903, + "step": 32120 + }, + { + "epoch": 0.27882570463798056, + "grad_norm": 0.275390625, + "learning_rate": 0.0015640160590452811, + "loss": 0.0889, + "step": 32121 + }, + { + "epoch": 0.27883438511818476, + "grad_norm": 0.68359375, + "learning_rate": 0.0015639905569668985, + "loss": 0.1348, + "step": 32122 + }, + { + "epoch": 0.2788430655983889, + "grad_norm": 0.26953125, + "learning_rate": 0.0015639650543811028, + "loss": 0.1895, + "step": 32123 + }, + { + "epoch": 0.2788517460785931, + "grad_norm": 0.1875, + "learning_rate": 0.001563939551287922, + "loss": 0.1152, + "step": 32124 + }, + { + "epoch": 0.2788604265587972, + "grad_norm": 0.80078125, + "learning_rate": 0.0015639140476873838, + "loss": 0.1357, + "step": 32125 + }, + { + "epoch": 0.2788691070390014, + "grad_norm": 0.326171875, + "learning_rate": 0.001563888543579516, + "loss": 0.1289, + "step": 32126 + }, + { + "epoch": 0.27887778751920556, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015638630389643471, + "loss": 0.0967, + "step": 32127 + }, + { + "epoch": 0.27888646799940975, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015638375338419043, + "loss": 0.1133, + "step": 32128 + }, + { + "epoch": 0.2788951484796139, + "grad_norm": 0.8828125, + "learning_rate": 0.0015638120282122155, + "loss": 0.1328, + "step": 32129 + }, + { + "epoch": 0.2789038289598181, + "grad_norm": 1.09375, + "learning_rate": 0.001563786522075309, + "loss": 0.1484, + "step": 32130 + }, + { + "epoch": 0.2789125094400222, + "grad_norm": 0.1845703125, + "learning_rate": 0.0015637610154312126, + "loss": 0.1377, + "step": 32131 + }, + { + "epoch": 0.2789211899202264, + "grad_norm": 0.0771484375, + "learning_rate": 0.0015637355082799543, + "loss": 0.0708, + "step": 32132 + }, + { + "epoch": 0.27892987040043055, + "grad_norm": 0.474609375, + "learning_rate": 0.0015637100006215615, + "loss": 0.207, + "step": 32133 + }, + { + "epoch": 0.27893855088063474, + "grad_norm": 0.4296875, + "learning_rate": 0.0015636844924560626, + "loss": 0.0854, + "step": 32134 + }, + { + "epoch": 0.2789472313608389, + "grad_norm": 0.55859375, + "learning_rate": 0.0015636589837834851, + "loss": 0.1016, + "step": 32135 + }, + { + "epoch": 0.27895591184104307, + "grad_norm": 0.6796875, + "learning_rate": 0.0015636334746038573, + "loss": 0.1221, + "step": 32136 + }, + { + "epoch": 0.2789645923212472, + "grad_norm": 0.16796875, + "learning_rate": 0.001563607964917207, + "loss": 0.1162, + "step": 32137 + }, + { + "epoch": 0.2789732728014514, + "grad_norm": 0.380859375, + "learning_rate": 0.001563582454723562, + "loss": 0.0938, + "step": 32138 + }, + { + "epoch": 0.27898195328165554, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015635569440229499, + "loss": 0.1396, + "step": 32139 + }, + { + "epoch": 0.27899063376185973, + "grad_norm": 0.515625, + "learning_rate": 0.001563531432815399, + "loss": 0.0977, + "step": 32140 + }, + { + "epoch": 0.27899931424206387, + "grad_norm": 0.330078125, + "learning_rate": 0.0015635059211009368, + "loss": 0.1099, + "step": 32141 + }, + { + "epoch": 0.27900799472226806, + "grad_norm": 0.453125, + "learning_rate": 0.0015634804088795922, + "loss": 0.1367, + "step": 32142 + }, + { + "epoch": 0.2790166752024722, + "grad_norm": 0.53515625, + "learning_rate": 0.0015634548961513917, + "loss": 0.0957, + "step": 32143 + }, + { + "epoch": 0.2790253556826764, + "grad_norm": 0.1640625, + "learning_rate": 0.0015634293829163645, + "loss": 0.1328, + "step": 32144 + }, + { + "epoch": 0.2790340361628805, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015634038691745373, + "loss": 0.1216, + "step": 32145 + }, + { + "epoch": 0.2790427166430847, + "grad_norm": 0.41015625, + "learning_rate": 0.0015633783549259391, + "loss": 0.0991, + "step": 32146 + }, + { + "epoch": 0.27905139712328886, + "grad_norm": 0.11181640625, + "learning_rate": 0.0015633528401705968, + "loss": 0.1055, + "step": 32147 + }, + { + "epoch": 0.27906007760349305, + "grad_norm": 0.1103515625, + "learning_rate": 0.001563327324908539, + "loss": 0.0947, + "step": 32148 + }, + { + "epoch": 0.2790687580836972, + "grad_norm": 0.0927734375, + "learning_rate": 0.0015633018091397936, + "loss": 0.126, + "step": 32149 + }, + { + "epoch": 0.2790774385639014, + "grad_norm": 0.953125, + "learning_rate": 0.0015632762928643883, + "loss": 0.1416, + "step": 32150 + }, + { + "epoch": 0.2790861190441055, + "grad_norm": 0.24609375, + "learning_rate": 0.0015632507760823507, + "loss": 0.0791, + "step": 32151 + }, + { + "epoch": 0.2790947995243097, + "grad_norm": 0.1494140625, + "learning_rate": 0.001563225258793709, + "loss": 0.1113, + "step": 32152 + }, + { + "epoch": 0.27910348000451385, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015631997409984912, + "loss": 0.1152, + "step": 32153 + }, + { + "epoch": 0.27911216048471804, + "grad_norm": 0.427734375, + "learning_rate": 0.0015631742226967253, + "loss": 0.1104, + "step": 32154 + }, + { + "epoch": 0.2791208409649222, + "grad_norm": 0.1875, + "learning_rate": 0.0015631487038884392, + "loss": 0.1162, + "step": 32155 + }, + { + "epoch": 0.27912952144512637, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015631231845736602, + "loss": 0.1084, + "step": 32156 + }, + { + "epoch": 0.2791382019253305, + "grad_norm": 0.08544921875, + "learning_rate": 0.0015630976647524168, + "loss": 0.0625, + "step": 32157 + }, + { + "epoch": 0.27914688240553465, + "grad_norm": 0.7890625, + "learning_rate": 0.0015630721444247367, + "loss": 0.1396, + "step": 32158 + }, + { + "epoch": 0.27915556288573884, + "grad_norm": 0.1044921875, + "learning_rate": 0.0015630466235906479, + "loss": 0.1104, + "step": 32159 + }, + { + "epoch": 0.279164243365943, + "grad_norm": 0.56640625, + "learning_rate": 0.001563021102250178, + "loss": 0.0947, + "step": 32160 + }, + { + "epoch": 0.27917292384614717, + "grad_norm": 0.10986328125, + "learning_rate": 0.0015629955804033558, + "loss": 0.0845, + "step": 32161 + }, + { + "epoch": 0.2791816043263513, + "grad_norm": 0.171875, + "learning_rate": 0.0015629700580502082, + "loss": 0.0684, + "step": 32162 + }, + { + "epoch": 0.2791902848065555, + "grad_norm": 0.09130859375, + "learning_rate": 0.0015629445351907638, + "loss": 0.1289, + "step": 32163 + }, + { + "epoch": 0.27919896528675964, + "grad_norm": 1.375, + "learning_rate": 0.0015629190118250501, + "loss": 0.1309, + "step": 32164 + }, + { + "epoch": 0.27920764576696383, + "grad_norm": 0.1611328125, + "learning_rate": 0.001562893487953095, + "loss": 0.126, + "step": 32165 + }, + { + "epoch": 0.27921632624716797, + "grad_norm": 0.103515625, + "learning_rate": 0.0015628679635749263, + "loss": 0.127, + "step": 32166 + }, + { + "epoch": 0.27922500672737216, + "grad_norm": 0.390625, + "learning_rate": 0.0015628424386905722, + "loss": 0.0908, + "step": 32167 + }, + { + "epoch": 0.2792336872075763, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015628169133000609, + "loss": 0.1084, + "step": 32168 + }, + { + "epoch": 0.2792423676877805, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015627913874034197, + "loss": 0.1123, + "step": 32169 + }, + { + "epoch": 0.27925104816798463, + "grad_norm": 0.46875, + "learning_rate": 0.001562765861000677, + "loss": 0.0962, + "step": 32170 + }, + { + "epoch": 0.2792597286481888, + "grad_norm": 0.171875, + "learning_rate": 0.0015627403340918605, + "loss": 0.1045, + "step": 32171 + }, + { + "epoch": 0.27926840912839296, + "grad_norm": 0.63671875, + "learning_rate": 0.0015627148066769982, + "loss": 0.082, + "step": 32172 + }, + { + "epoch": 0.27927708960859715, + "grad_norm": 0.400390625, + "learning_rate": 0.0015626892787561175, + "loss": 0.1079, + "step": 32173 + }, + { + "epoch": 0.2792857700888013, + "grad_norm": 0.703125, + "learning_rate": 0.001562663750329247, + "loss": 0.0806, + "step": 32174 + }, + { + "epoch": 0.2792944505690055, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015626382213964147, + "loss": 0.0928, + "step": 32175 + }, + { + "epoch": 0.2793031310492096, + "grad_norm": 0.1015625, + "learning_rate": 0.001562612691957648, + "loss": 0.082, + "step": 32176 + }, + { + "epoch": 0.2793118115294138, + "grad_norm": 0.4453125, + "learning_rate": 0.0015625871620129747, + "loss": 0.1221, + "step": 32177 + }, + { + "epoch": 0.27932049200961795, + "grad_norm": 0.36328125, + "learning_rate": 0.001562561631562423, + "loss": 0.1167, + "step": 32178 + }, + { + "epoch": 0.27932917248982214, + "grad_norm": 0.232421875, + "learning_rate": 0.0015625361006060213, + "loss": 0.1279, + "step": 32179 + }, + { + "epoch": 0.2793378529700263, + "grad_norm": 0.12255859375, + "learning_rate": 0.0015625105691437968, + "loss": 0.1133, + "step": 32180 + }, + { + "epoch": 0.27934653345023047, + "grad_norm": 0.6328125, + "learning_rate": 0.001562485037175778, + "loss": 0.1289, + "step": 32181 + }, + { + "epoch": 0.2793552139304346, + "grad_norm": 0.1748046875, + "learning_rate": 0.0015624595047019922, + "loss": 0.1406, + "step": 32182 + }, + { + "epoch": 0.2793638944106388, + "grad_norm": 0.90625, + "learning_rate": 0.0015624339717224677, + "loss": 0.0894, + "step": 32183 + }, + { + "epoch": 0.27937257489084294, + "grad_norm": 0.201171875, + "learning_rate": 0.0015624084382372325, + "loss": 0.1187, + "step": 32184 + }, + { + "epoch": 0.27938125537104713, + "grad_norm": 0.58203125, + "learning_rate": 0.0015623829042463143, + "loss": 0.1826, + "step": 32185 + }, + { + "epoch": 0.27938993585125127, + "grad_norm": 0.17578125, + "learning_rate": 0.001562357369749741, + "loss": 0.1094, + "step": 32186 + }, + { + "epoch": 0.27939861633145546, + "grad_norm": 0.408203125, + "learning_rate": 0.0015623318347475403, + "loss": 0.1484, + "step": 32187 + }, + { + "epoch": 0.2794072968116596, + "grad_norm": 0.51953125, + "learning_rate": 0.001562306299239741, + "loss": 0.0991, + "step": 32188 + }, + { + "epoch": 0.2794159772918638, + "grad_norm": 0.205078125, + "learning_rate": 0.0015622807632263704, + "loss": 0.0762, + "step": 32189 + }, + { + "epoch": 0.27942465777206793, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015622552267074564, + "loss": 0.1016, + "step": 32190 + }, + { + "epoch": 0.2794333382522721, + "grad_norm": 0.39453125, + "learning_rate": 0.0015622296896830269, + "loss": 0.1074, + "step": 32191 + }, + { + "epoch": 0.27944201873247626, + "grad_norm": 0.291015625, + "learning_rate": 0.00156220415215311, + "loss": 0.0986, + "step": 32192 + }, + { + "epoch": 0.27945069921268045, + "grad_norm": 0.3046875, + "learning_rate": 0.001562178614117734, + "loss": 0.1328, + "step": 32193 + }, + { + "epoch": 0.2794593796928846, + "grad_norm": 0.1572265625, + "learning_rate": 0.001562153075576926, + "loss": 0.0845, + "step": 32194 + }, + { + "epoch": 0.2794680601730888, + "grad_norm": 0.609375, + "learning_rate": 0.0015621275365307146, + "loss": 0.1748, + "step": 32195 + }, + { + "epoch": 0.2794767406532929, + "grad_norm": 0.138671875, + "learning_rate": 0.001562101996979127, + "loss": 0.0889, + "step": 32196 + }, + { + "epoch": 0.2794854211334971, + "grad_norm": 0.5703125, + "learning_rate": 0.001562076456922192, + "loss": 0.1611, + "step": 32197 + }, + { + "epoch": 0.27949410161370125, + "grad_norm": 0.2490234375, + "learning_rate": 0.001562050916359937, + "loss": 0.1069, + "step": 32198 + }, + { + "epoch": 0.27950278209390544, + "grad_norm": 0.60546875, + "learning_rate": 0.0015620253752923901, + "loss": 0.1006, + "step": 32199 + }, + { + "epoch": 0.2795114625741096, + "grad_norm": 0.75390625, + "learning_rate": 0.0015619998337195794, + "loss": 0.0908, + "step": 32200 + }, + { + "epoch": 0.2795201430543138, + "grad_norm": 0.4296875, + "learning_rate": 0.0015619742916415324, + "loss": 0.1172, + "step": 32201 + }, + { + "epoch": 0.2795288235345179, + "grad_norm": 0.57421875, + "learning_rate": 0.001561948749058277, + "loss": 0.0889, + "step": 32202 + }, + { + "epoch": 0.2795375040147221, + "grad_norm": 0.8125, + "learning_rate": 0.0015619232059698419, + "loss": 0.1113, + "step": 32203 + }, + { + "epoch": 0.27954618449492624, + "grad_norm": 0.55078125, + "learning_rate": 0.0015618976623762541, + "loss": 0.0962, + "step": 32204 + }, + { + "epoch": 0.27955486497513043, + "grad_norm": 0.25, + "learning_rate": 0.0015618721182775422, + "loss": 0.0991, + "step": 32205 + }, + { + "epoch": 0.27956354545533457, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015618465736737337, + "loss": 0.1504, + "step": 32206 + }, + { + "epoch": 0.27957222593553877, + "grad_norm": 0.134765625, + "learning_rate": 0.0015618210285648568, + "loss": 0.0703, + "step": 32207 + }, + { + "epoch": 0.2795809064157429, + "grad_norm": 0.291015625, + "learning_rate": 0.0015617954829509394, + "loss": 0.0806, + "step": 32208 + }, + { + "epoch": 0.2795895868959471, + "grad_norm": 0.080078125, + "learning_rate": 0.0015617699368320095, + "loss": 0.0947, + "step": 32209 + }, + { + "epoch": 0.27959826737615123, + "grad_norm": 0.3515625, + "learning_rate": 0.0015617443902080947, + "loss": 0.1465, + "step": 32210 + }, + { + "epoch": 0.2796069478563554, + "grad_norm": 0.203125, + "learning_rate": 0.0015617188430792238, + "loss": 0.126, + "step": 32211 + }, + { + "epoch": 0.27961562833655956, + "grad_norm": 0.68359375, + "learning_rate": 0.0015616932954454232, + "loss": 0.1279, + "step": 32212 + }, + { + "epoch": 0.27962430881676376, + "grad_norm": 0.71484375, + "learning_rate": 0.0015616677473067222, + "loss": 0.1113, + "step": 32213 + }, + { + "epoch": 0.2796329892969679, + "grad_norm": 0.16796875, + "learning_rate": 0.0015616421986631485, + "loss": 0.1016, + "step": 32214 + }, + { + "epoch": 0.2796416697771721, + "grad_norm": 0.205078125, + "learning_rate": 0.0015616166495147296, + "loss": 0.1143, + "step": 32215 + }, + { + "epoch": 0.2796503502573762, + "grad_norm": 0.162109375, + "learning_rate": 0.0015615910998614934, + "loss": 0.085, + "step": 32216 + }, + { + "epoch": 0.2796590307375804, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015615655497034683, + "loss": 0.1182, + "step": 32217 + }, + { + "epoch": 0.27966771121778455, + "grad_norm": 0.380859375, + "learning_rate": 0.0015615399990406822, + "loss": 0.1035, + "step": 32218 + }, + { + "epoch": 0.27967639169798875, + "grad_norm": 0.3125, + "learning_rate": 0.0015615144478731628, + "loss": 0.1201, + "step": 32219 + }, + { + "epoch": 0.2796850721781929, + "grad_norm": 0.294921875, + "learning_rate": 0.0015614888962009383, + "loss": 0.0938, + "step": 32220 + }, + { + "epoch": 0.2796937526583971, + "grad_norm": 0.306640625, + "learning_rate": 0.001561463344024036, + "loss": 0.0967, + "step": 32221 + }, + { + "epoch": 0.2797024331386012, + "grad_norm": 0.119140625, + "learning_rate": 0.001561437791342485, + "loss": 0.0811, + "step": 32222 + }, + { + "epoch": 0.2797111136188054, + "grad_norm": 0.287109375, + "learning_rate": 0.001561412238156312, + "loss": 0.1084, + "step": 32223 + }, + { + "epoch": 0.27971979409900954, + "grad_norm": 0.359375, + "learning_rate": 0.0015613866844655458, + "loss": 0.1855, + "step": 32224 + }, + { + "epoch": 0.27972847457921374, + "grad_norm": 0.1279296875, + "learning_rate": 0.001561361130270214, + "loss": 0.1416, + "step": 32225 + }, + { + "epoch": 0.2797371550594179, + "grad_norm": 0.326171875, + "learning_rate": 0.0015613355755703445, + "loss": 0.123, + "step": 32226 + }, + { + "epoch": 0.27974583553962207, + "grad_norm": 0.3671875, + "learning_rate": 0.0015613100203659656, + "loss": 0.0923, + "step": 32227 + }, + { + "epoch": 0.2797545160198262, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015612844646571045, + "loss": 0.1021, + "step": 32228 + }, + { + "epoch": 0.2797631965000304, + "grad_norm": 1.15625, + "learning_rate": 0.00156125890844379, + "loss": 0.1621, + "step": 32229 + }, + { + "epoch": 0.27977187698023454, + "grad_norm": 0.171875, + "learning_rate": 0.0015612333517260495, + "loss": 0.1182, + "step": 32230 + }, + { + "epoch": 0.27978055746043873, + "grad_norm": 0.291015625, + "learning_rate": 0.0015612077945039114, + "loss": 0.082, + "step": 32231 + }, + { + "epoch": 0.27978923794064287, + "grad_norm": 0.162109375, + "learning_rate": 0.0015611822367774034, + "loss": 0.1133, + "step": 32232 + }, + { + "epoch": 0.27979791842084706, + "grad_norm": 0.0771484375, + "learning_rate": 0.0015611566785465533, + "loss": 0.0869, + "step": 32233 + }, + { + "epoch": 0.2798065989010512, + "grad_norm": 0.3046875, + "learning_rate": 0.0015611311198113892, + "loss": 0.1348, + "step": 32234 + }, + { + "epoch": 0.2798152793812554, + "grad_norm": 0.1181640625, + "learning_rate": 0.0015611055605719388, + "loss": 0.1099, + "step": 32235 + }, + { + "epoch": 0.2798239598614595, + "grad_norm": 0.150390625, + "learning_rate": 0.0015610800008282306, + "loss": 0.0972, + "step": 32236 + }, + { + "epoch": 0.2798326403416637, + "grad_norm": 0.169921875, + "learning_rate": 0.0015610544405802922, + "loss": 0.1299, + "step": 32237 + }, + { + "epoch": 0.27984132082186786, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015610288798281516, + "loss": 0.0918, + "step": 32238 + }, + { + "epoch": 0.27985000130207205, + "grad_norm": 0.1923828125, + "learning_rate": 0.0015610033185718365, + "loss": 0.0991, + "step": 32239 + }, + { + "epoch": 0.2798586817822762, + "grad_norm": 0.14453125, + "learning_rate": 0.0015609777568113751, + "loss": 0.1025, + "step": 32240 + }, + { + "epoch": 0.2798673622624804, + "grad_norm": 0.31640625, + "learning_rate": 0.0015609521945467958, + "loss": 0.0996, + "step": 32241 + }, + { + "epoch": 0.2798760427426845, + "grad_norm": 0.490234375, + "learning_rate": 0.001560926631778126, + "loss": 0.0962, + "step": 32242 + }, + { + "epoch": 0.2798847232228887, + "grad_norm": 0.65625, + "learning_rate": 0.0015609010685053934, + "loss": 0.1016, + "step": 32243 + }, + { + "epoch": 0.27989340370309285, + "grad_norm": 0.173828125, + "learning_rate": 0.0015608755047286263, + "loss": 0.0938, + "step": 32244 + }, + { + "epoch": 0.27990208418329704, + "grad_norm": 0.091796875, + "learning_rate": 0.0015608499404478528, + "loss": 0.1172, + "step": 32245 + }, + { + "epoch": 0.2799107646635012, + "grad_norm": 0.333984375, + "learning_rate": 0.001560824375663101, + "loss": 0.1309, + "step": 32246 + }, + { + "epoch": 0.27991944514370537, + "grad_norm": 0.09423828125, + "learning_rate": 0.0015607988103743984, + "loss": 0.1011, + "step": 32247 + }, + { + "epoch": 0.2799281256239095, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015607732445817731, + "loss": 0.124, + "step": 32248 + }, + { + "epoch": 0.2799368061041137, + "grad_norm": 0.162109375, + "learning_rate": 0.0015607476782852532, + "loss": 0.1123, + "step": 32249 + }, + { + "epoch": 0.27994548658431784, + "grad_norm": 0.11865234375, + "learning_rate": 0.0015607221114848663, + "loss": 0.1211, + "step": 32250 + }, + { + "epoch": 0.27995416706452203, + "grad_norm": 0.12255859375, + "learning_rate": 0.001560696544180641, + "loss": 0.0957, + "step": 32251 + }, + { + "epoch": 0.27996284754472617, + "grad_norm": 0.287109375, + "learning_rate": 0.0015606709763726048, + "loss": 0.1133, + "step": 32252 + }, + { + "epoch": 0.27997152802493036, + "grad_norm": 0.306640625, + "learning_rate": 0.0015606454080607855, + "loss": 0.0889, + "step": 32253 + }, + { + "epoch": 0.2799802085051345, + "grad_norm": 0.130859375, + "learning_rate": 0.0015606198392452114, + "loss": 0.0972, + "step": 32254 + }, + { + "epoch": 0.2799888889853387, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015605942699259104, + "loss": 0.1338, + "step": 32255 + }, + { + "epoch": 0.27999756946554283, + "grad_norm": 0.193359375, + "learning_rate": 0.0015605687001029106, + "loss": 0.0713, + "step": 32256 + }, + { + "epoch": 0.280006249945747, + "grad_norm": 0.1640625, + "learning_rate": 0.0015605431297762396, + "loss": 0.1348, + "step": 32257 + }, + { + "epoch": 0.28001493042595116, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015605175589459255, + "loss": 0.0806, + "step": 32258 + }, + { + "epoch": 0.28002361090615535, + "grad_norm": 0.23046875, + "learning_rate": 0.0015604919876119965, + "loss": 0.0986, + "step": 32259 + }, + { + "epoch": 0.2800322913863595, + "grad_norm": 0.41796875, + "learning_rate": 0.0015604664157744803, + "loss": 0.105, + "step": 32260 + }, + { + "epoch": 0.2800409718665637, + "grad_norm": 0.0771484375, + "learning_rate": 0.0015604408434334047, + "loss": 0.0957, + "step": 32261 + }, + { + "epoch": 0.2800496523467678, + "grad_norm": 0.2353515625, + "learning_rate": 0.0015604152705887984, + "loss": 0.1143, + "step": 32262 + }, + { + "epoch": 0.280058332826972, + "grad_norm": 0.546875, + "learning_rate": 0.0015603896972406885, + "loss": 0.1475, + "step": 32263 + }, + { + "epoch": 0.28006701330717615, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015603641233891033, + "loss": 0.1328, + "step": 32264 + }, + { + "epoch": 0.28007569378738034, + "grad_norm": 0.26171875, + "learning_rate": 0.001560338549034071, + "loss": 0.1094, + "step": 32265 + }, + { + "epoch": 0.2800843742675845, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015603129741756194, + "loss": 0.0762, + "step": 32266 + }, + { + "epoch": 0.2800930547477887, + "grad_norm": 0.2236328125, + "learning_rate": 0.0015602873988137765, + "loss": 0.0923, + "step": 32267 + }, + { + "epoch": 0.2801017352279928, + "grad_norm": 0.30078125, + "learning_rate": 0.0015602618229485698, + "loss": 0.0752, + "step": 32268 + }, + { + "epoch": 0.280110415708197, + "grad_norm": 0.42578125, + "learning_rate": 0.0015602362465800283, + "loss": 0.0747, + "step": 32269 + }, + { + "epoch": 0.28011909618840114, + "grad_norm": 0.9140625, + "learning_rate": 0.0015602106697081791, + "loss": 0.1172, + "step": 32270 + }, + { + "epoch": 0.28012777666860533, + "grad_norm": 0.224609375, + "learning_rate": 0.0015601850923330505, + "loss": 0.083, + "step": 32271 + }, + { + "epoch": 0.28013645714880947, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015601595144546698, + "loss": 0.0923, + "step": 32272 + }, + { + "epoch": 0.28014513762901366, + "grad_norm": 0.2177734375, + "learning_rate": 0.0015601339360730665, + "loss": 0.0811, + "step": 32273 + }, + { + "epoch": 0.2801538181092178, + "grad_norm": 0.25390625, + "learning_rate": 0.0015601083571882667, + "loss": 0.0762, + "step": 32274 + }, + { + "epoch": 0.280162498589422, + "grad_norm": 0.4921875, + "learning_rate": 0.0015600827778003003, + "loss": 0.1064, + "step": 32275 + }, + { + "epoch": 0.28017117906962613, + "grad_norm": 0.64453125, + "learning_rate": 0.0015600571979091935, + "loss": 0.1152, + "step": 32276 + }, + { + "epoch": 0.2801798595498303, + "grad_norm": 0.080078125, + "learning_rate": 0.0015600316175149753, + "loss": 0.0869, + "step": 32277 + }, + { + "epoch": 0.28018854003003446, + "grad_norm": 0.32421875, + "learning_rate": 0.0015600060366176738, + "loss": 0.0825, + "step": 32278 + }, + { + "epoch": 0.28019722051023865, + "grad_norm": 0.09814453125, + "learning_rate": 0.001559980455217316, + "loss": 0.127, + "step": 32279 + }, + { + "epoch": 0.2802059009904428, + "grad_norm": 0.12890625, + "learning_rate": 0.0015599548733139308, + "loss": 0.1797, + "step": 32280 + }, + { + "epoch": 0.28021458147064693, + "grad_norm": 0.09765625, + "learning_rate": 0.001559929290907546, + "loss": 0.1055, + "step": 32281 + }, + { + "epoch": 0.2802232619508511, + "grad_norm": 0.9453125, + "learning_rate": 0.0015599037079981892, + "loss": 0.123, + "step": 32282 + }, + { + "epoch": 0.28023194243105526, + "grad_norm": 0.109375, + "learning_rate": 0.0015598781245858887, + "loss": 0.1074, + "step": 32283 + }, + { + "epoch": 0.28024062291125945, + "grad_norm": 0.423828125, + "learning_rate": 0.0015598525406706725, + "loss": 0.1221, + "step": 32284 + }, + { + "epoch": 0.2802493033914636, + "grad_norm": 1.875, + "learning_rate": 0.0015598269562525685, + "loss": 0.0991, + "step": 32285 + }, + { + "epoch": 0.2802579838716678, + "grad_norm": 0.169921875, + "learning_rate": 0.0015598013713316041, + "loss": 0.1055, + "step": 32286 + }, + { + "epoch": 0.2802666643518719, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015597757859078082, + "loss": 0.1064, + "step": 32287 + }, + { + "epoch": 0.2802753448320761, + "grad_norm": 0.1064453125, + "learning_rate": 0.0015597501999812086, + "loss": 0.0977, + "step": 32288 + }, + { + "epoch": 0.28028402531228025, + "grad_norm": 0.404296875, + "learning_rate": 0.001559724613551833, + "loss": 0.1201, + "step": 32289 + }, + { + "epoch": 0.28029270579248444, + "grad_norm": 0.45703125, + "learning_rate": 0.0015596990266197094, + "loss": 0.1055, + "step": 32290 + }, + { + "epoch": 0.2803013862726886, + "grad_norm": 0.1787109375, + "learning_rate": 0.001559673439184866, + "loss": 0.1045, + "step": 32291 + }, + { + "epoch": 0.2803100667528928, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015596478512473302, + "loss": 0.1367, + "step": 32292 + }, + { + "epoch": 0.2803187472330969, + "grad_norm": 0.23828125, + "learning_rate": 0.0015596222628071307, + "loss": 0.125, + "step": 32293 + }, + { + "epoch": 0.2803274277133011, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015595966738642952, + "loss": 0.0635, + "step": 32294 + }, + { + "epoch": 0.28033610819350524, + "grad_norm": 0.79296875, + "learning_rate": 0.001559571084418852, + "loss": 0.104, + "step": 32295 + }, + { + "epoch": 0.28034478867370943, + "grad_norm": 0.37109375, + "learning_rate": 0.0015595454944708283, + "loss": 0.0986, + "step": 32296 + }, + { + "epoch": 0.28035346915391357, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015595199040202528, + "loss": 0.0786, + "step": 32297 + }, + { + "epoch": 0.28036214963411776, + "grad_norm": 0.29296875, + "learning_rate": 0.001559494313067153, + "loss": 0.1387, + "step": 32298 + }, + { + "epoch": 0.2803708301143219, + "grad_norm": 0.333984375, + "learning_rate": 0.0015594687216115575, + "loss": 0.0688, + "step": 32299 + }, + { + "epoch": 0.2803795105945261, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015594431296534934, + "loss": 0.1016, + "step": 32300 + }, + { + "epoch": 0.28038819107473023, + "grad_norm": 0.294921875, + "learning_rate": 0.0015594175371929895, + "loss": 0.0879, + "step": 32301 + }, + { + "epoch": 0.2803968715549344, + "grad_norm": 0.11669921875, + "learning_rate": 0.0015593919442300733, + "loss": 0.1011, + "step": 32302 + }, + { + "epoch": 0.28040555203513856, + "grad_norm": 0.10986328125, + "learning_rate": 0.001559366350764773, + "loss": 0.0874, + "step": 32303 + }, + { + "epoch": 0.28041423251534275, + "grad_norm": 0.24609375, + "learning_rate": 0.001559340756797117, + "loss": 0.0918, + "step": 32304 + }, + { + "epoch": 0.2804229129955469, + "grad_norm": 0.162109375, + "learning_rate": 0.0015593151623271323, + "loss": 0.1206, + "step": 32305 + }, + { + "epoch": 0.2804315934757511, + "grad_norm": 0.6015625, + "learning_rate": 0.001559289567354848, + "loss": 0.1079, + "step": 32306 + }, + { + "epoch": 0.2804402739559552, + "grad_norm": 0.1845703125, + "learning_rate": 0.001559263971880291, + "loss": 0.0967, + "step": 32307 + }, + { + "epoch": 0.2804489544361594, + "grad_norm": 0.18359375, + "learning_rate": 0.00155923837590349, + "loss": 0.0947, + "step": 32308 + }, + { + "epoch": 0.28045763491636355, + "grad_norm": 0.3984375, + "learning_rate": 0.001559212779424473, + "loss": 0.1406, + "step": 32309 + }, + { + "epoch": 0.28046631539656774, + "grad_norm": 0.076171875, + "learning_rate": 0.0015591871824432678, + "loss": 0.1069, + "step": 32310 + }, + { + "epoch": 0.2804749958767719, + "grad_norm": 0.162109375, + "learning_rate": 0.0015591615849599022, + "loss": 0.0928, + "step": 32311 + }, + { + "epoch": 0.2804836763569761, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015591359869744042, + "loss": 0.084, + "step": 32312 + }, + { + "epoch": 0.2804923568371802, + "grad_norm": 0.09423828125, + "learning_rate": 0.0015591103884868022, + "loss": 0.1201, + "step": 32313 + }, + { + "epoch": 0.2805010373173844, + "grad_norm": 0.220703125, + "learning_rate": 0.001559084789497124, + "loss": 0.1104, + "step": 32314 + }, + { + "epoch": 0.28050971779758854, + "grad_norm": 0.298828125, + "learning_rate": 0.0015590591900053977, + "loss": 0.0557, + "step": 32315 + }, + { + "epoch": 0.28051839827779274, + "grad_norm": 0.34765625, + "learning_rate": 0.0015590335900116511, + "loss": 0.1426, + "step": 32316 + }, + { + "epoch": 0.2805270787579969, + "grad_norm": 0.498046875, + "learning_rate": 0.0015590079895159122, + "loss": 0.123, + "step": 32317 + }, + { + "epoch": 0.28053575923820107, + "grad_norm": 0.205078125, + "learning_rate": 0.001558982388518209, + "loss": 0.082, + "step": 32318 + }, + { + "epoch": 0.2805444397184052, + "grad_norm": 1.2421875, + "learning_rate": 0.0015589567870185697, + "loss": 0.1855, + "step": 32319 + }, + { + "epoch": 0.2805531201986094, + "grad_norm": 0.625, + "learning_rate": 0.0015589311850170218, + "loss": 0.1162, + "step": 32320 + }, + { + "epoch": 0.28056180067881353, + "grad_norm": 0.1328125, + "learning_rate": 0.001558905582513594, + "loss": 0.1367, + "step": 32321 + }, + { + "epoch": 0.2805704811590177, + "grad_norm": 0.13671875, + "learning_rate": 0.0015588799795083139, + "loss": 0.1309, + "step": 32322 + }, + { + "epoch": 0.28057916163922186, + "grad_norm": 0.369140625, + "learning_rate": 0.0015588543760012095, + "loss": 0.0928, + "step": 32323 + }, + { + "epoch": 0.28058784211942606, + "grad_norm": 0.3828125, + "learning_rate": 0.0015588287719923088, + "loss": 0.0923, + "step": 32324 + }, + { + "epoch": 0.2805965225996302, + "grad_norm": 0.1845703125, + "learning_rate": 0.0015588031674816398, + "loss": 0.1504, + "step": 32325 + }, + { + "epoch": 0.2806052030798344, + "grad_norm": 0.30859375, + "learning_rate": 0.0015587775624692308, + "loss": 0.1523, + "step": 32326 + }, + { + "epoch": 0.2806138835600385, + "grad_norm": 0.64453125, + "learning_rate": 0.0015587519569551097, + "loss": 0.0938, + "step": 32327 + }, + { + "epoch": 0.2806225640402427, + "grad_norm": 0.248046875, + "learning_rate": 0.0015587263509393042, + "loss": 0.1084, + "step": 32328 + }, + { + "epoch": 0.28063124452044685, + "grad_norm": 0.31640625, + "learning_rate": 0.001558700744421842, + "loss": 0.1099, + "step": 32329 + }, + { + "epoch": 0.28063992500065105, + "grad_norm": 0.162109375, + "learning_rate": 0.001558675137402752, + "loss": 0.1426, + "step": 32330 + }, + { + "epoch": 0.2806486054808552, + "grad_norm": 0.443359375, + "learning_rate": 0.0015586495298820612, + "loss": 0.0967, + "step": 32331 + }, + { + "epoch": 0.2806572859610594, + "grad_norm": 0.251953125, + "learning_rate": 0.0015586239218597989, + "loss": 0.0747, + "step": 32332 + }, + { + "epoch": 0.2806659664412635, + "grad_norm": 0.388671875, + "learning_rate": 0.001558598313335992, + "loss": 0.1348, + "step": 32333 + }, + { + "epoch": 0.2806746469214677, + "grad_norm": 0.09521484375, + "learning_rate": 0.0015585727043106689, + "loss": 0.0874, + "step": 32334 + }, + { + "epoch": 0.28068332740167184, + "grad_norm": 0.125, + "learning_rate": 0.0015585470947838576, + "loss": 0.1196, + "step": 32335 + }, + { + "epoch": 0.28069200788187604, + "grad_norm": 0.400390625, + "learning_rate": 0.0015585214847555864, + "loss": 0.0908, + "step": 32336 + }, + { + "epoch": 0.2807006883620802, + "grad_norm": 0.8515625, + "learning_rate": 0.0015584958742258827, + "loss": 0.1445, + "step": 32337 + }, + { + "epoch": 0.28070936884228437, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015584702631947747, + "loss": 0.124, + "step": 32338 + }, + { + "epoch": 0.2807180493224885, + "grad_norm": 0.78515625, + "learning_rate": 0.0015584446516622907, + "loss": 0.103, + "step": 32339 + }, + { + "epoch": 0.2807267298026927, + "grad_norm": 0.71875, + "learning_rate": 0.0015584190396284582, + "loss": 0.1123, + "step": 32340 + }, + { + "epoch": 0.28073541028289684, + "grad_norm": 0.2314453125, + "learning_rate": 0.0015583934270933057, + "loss": 0.1074, + "step": 32341 + }, + { + "epoch": 0.28074409076310103, + "grad_norm": 0.09814453125, + "learning_rate": 0.0015583678140568612, + "loss": 0.0928, + "step": 32342 + }, + { + "epoch": 0.28075277124330517, + "grad_norm": 0.90234375, + "learning_rate": 0.0015583422005191524, + "loss": 0.1465, + "step": 32343 + }, + { + "epoch": 0.28076145172350936, + "grad_norm": 0.296875, + "learning_rate": 0.0015583165864802076, + "loss": 0.1084, + "step": 32344 + }, + { + "epoch": 0.2807701322037135, + "grad_norm": 0.734375, + "learning_rate": 0.0015582909719400546, + "loss": 0.0996, + "step": 32345 + }, + { + "epoch": 0.2807788126839177, + "grad_norm": 0.455078125, + "learning_rate": 0.0015582653568987214, + "loss": 0.0864, + "step": 32346 + }, + { + "epoch": 0.2807874931641218, + "grad_norm": 0.349609375, + "learning_rate": 0.001558239741356236, + "loss": 0.0483, + "step": 32347 + }, + { + "epoch": 0.280796173644326, + "grad_norm": 0.25, + "learning_rate": 0.001558214125312627, + "loss": 0.127, + "step": 32348 + }, + { + "epoch": 0.28080485412453016, + "grad_norm": 0.3671875, + "learning_rate": 0.0015581885087679213, + "loss": 0.1836, + "step": 32349 + }, + { + "epoch": 0.28081353460473435, + "grad_norm": 0.41796875, + "learning_rate": 0.0015581628917221478, + "loss": 0.085, + "step": 32350 + }, + { + "epoch": 0.2808222150849385, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015581372741753341, + "loss": 0.0613, + "step": 32351 + }, + { + "epoch": 0.2808308955651427, + "grad_norm": 0.2353515625, + "learning_rate": 0.0015581116561275087, + "loss": 0.0898, + "step": 32352 + }, + { + "epoch": 0.2808395760453468, + "grad_norm": 0.166015625, + "learning_rate": 0.001558086037578699, + "loss": 0.084, + "step": 32353 + }, + { + "epoch": 0.280848256525551, + "grad_norm": 0.205078125, + "learning_rate": 0.0015580604185289333, + "loss": 0.1216, + "step": 32354 + }, + { + "epoch": 0.28085693700575515, + "grad_norm": 0.193359375, + "learning_rate": 0.0015580347989782397, + "loss": 0.1099, + "step": 32355 + }, + { + "epoch": 0.28086561748595934, + "grad_norm": 0.72265625, + "learning_rate": 0.001558009178926646, + "loss": 0.1074, + "step": 32356 + }, + { + "epoch": 0.2808742979661635, + "grad_norm": 0.408203125, + "learning_rate": 0.0015579835583741806, + "loss": 0.106, + "step": 32357 + }, + { + "epoch": 0.28088297844636767, + "grad_norm": 0.279296875, + "learning_rate": 0.0015579579373208713, + "loss": 0.1064, + "step": 32358 + }, + { + "epoch": 0.2808916589265718, + "grad_norm": 0.78515625, + "learning_rate": 0.0015579323157667456, + "loss": 0.1494, + "step": 32359 + }, + { + "epoch": 0.280900339406776, + "grad_norm": 0.27734375, + "learning_rate": 0.0015579066937118323, + "loss": 0.0918, + "step": 32360 + }, + { + "epoch": 0.28090901988698014, + "grad_norm": 0.255859375, + "learning_rate": 0.001557881071156159, + "loss": 0.0947, + "step": 32361 + }, + { + "epoch": 0.28091770036718433, + "grad_norm": 0.68359375, + "learning_rate": 0.001557855448099754, + "loss": 0.1074, + "step": 32362 + }, + { + "epoch": 0.28092638084738847, + "grad_norm": 0.8984375, + "learning_rate": 0.0015578298245426453, + "loss": 0.1108, + "step": 32363 + }, + { + "epoch": 0.28093506132759266, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015578042004848604, + "loss": 0.1094, + "step": 32364 + }, + { + "epoch": 0.2809437418077968, + "grad_norm": 0.57421875, + "learning_rate": 0.001557778575926428, + "loss": 0.1338, + "step": 32365 + }, + { + "epoch": 0.280952422288001, + "grad_norm": 2.390625, + "learning_rate": 0.0015577529508673759, + "loss": 0.2715, + "step": 32366 + }, + { + "epoch": 0.28096110276820513, + "grad_norm": 0.30078125, + "learning_rate": 0.0015577273253077318, + "loss": 0.0928, + "step": 32367 + }, + { + "epoch": 0.2809697832484093, + "grad_norm": 1.296875, + "learning_rate": 0.001557701699247524, + "loss": 0.0908, + "step": 32368 + }, + { + "epoch": 0.28097846372861346, + "grad_norm": 0.5390625, + "learning_rate": 0.0015576760726867802, + "loss": 0.1074, + "step": 32369 + }, + { + "epoch": 0.28098714420881765, + "grad_norm": 0.205078125, + "learning_rate": 0.0015576504456255291, + "loss": 0.0908, + "step": 32370 + }, + { + "epoch": 0.2809958246890218, + "grad_norm": 0.2392578125, + "learning_rate": 0.0015576248180637984, + "loss": 0.1016, + "step": 32371 + }, + { + "epoch": 0.281004505169226, + "grad_norm": 0.408203125, + "learning_rate": 0.001557599190001616, + "loss": 0.1416, + "step": 32372 + }, + { + "epoch": 0.2810131856494301, + "grad_norm": 0.1962890625, + "learning_rate": 0.00155757356143901, + "loss": 0.0942, + "step": 32373 + }, + { + "epoch": 0.2810218661296343, + "grad_norm": 0.88671875, + "learning_rate": 0.0015575479323760085, + "loss": 0.1328, + "step": 32374 + }, + { + "epoch": 0.28103054660983845, + "grad_norm": 0.7265625, + "learning_rate": 0.0015575223028126395, + "loss": 0.106, + "step": 32375 + }, + { + "epoch": 0.28103922709004264, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015574966727489306, + "loss": 0.1011, + "step": 32376 + }, + { + "epoch": 0.2810479075702468, + "grad_norm": 0.34375, + "learning_rate": 0.0015574710421849105, + "loss": 0.0977, + "step": 32377 + }, + { + "epoch": 0.281056588050451, + "grad_norm": 0.419921875, + "learning_rate": 0.001557445411120607, + "loss": 0.1143, + "step": 32378 + }, + { + "epoch": 0.2810652685306551, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015574197795560482, + "loss": 0.1035, + "step": 32379 + }, + { + "epoch": 0.2810739490108593, + "grad_norm": 0.228515625, + "learning_rate": 0.0015573941474912613, + "loss": 0.085, + "step": 32380 + }, + { + "epoch": 0.28108262949106344, + "grad_norm": 0.1484375, + "learning_rate": 0.0015573685149262756, + "loss": 0.103, + "step": 32381 + }, + { + "epoch": 0.28109130997126763, + "grad_norm": 0.72265625, + "learning_rate": 0.0015573428818611186, + "loss": 0.1699, + "step": 32382 + }, + { + "epoch": 0.28109999045147177, + "grad_norm": 0.396484375, + "learning_rate": 0.0015573172482958183, + "loss": 0.0952, + "step": 32383 + }, + { + "epoch": 0.28110867093167596, + "grad_norm": 0.14453125, + "learning_rate": 0.0015572916142304026, + "loss": 0.125, + "step": 32384 + }, + { + "epoch": 0.2811173514118801, + "grad_norm": 0.68359375, + "learning_rate": 0.0015572659796648995, + "loss": 0.0732, + "step": 32385 + }, + { + "epoch": 0.2811260318920843, + "grad_norm": 0.30078125, + "learning_rate": 0.0015572403445993372, + "loss": 0.1011, + "step": 32386 + }, + { + "epoch": 0.28113471237228843, + "grad_norm": 0.4453125, + "learning_rate": 0.0015572147090337444, + "loss": 0.0947, + "step": 32387 + }, + { + "epoch": 0.2811433928524926, + "grad_norm": 0.345703125, + "learning_rate": 0.0015571890729681476, + "loss": 0.1152, + "step": 32388 + }, + { + "epoch": 0.28115207333269676, + "grad_norm": 0.189453125, + "learning_rate": 0.001557163436402576, + "loss": 0.0889, + "step": 32389 + }, + { + "epoch": 0.28116075381290095, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015571377993370574, + "loss": 0.0947, + "step": 32390 + }, + { + "epoch": 0.2811694342931051, + "grad_norm": 0.205078125, + "learning_rate": 0.00155711216177162, + "loss": 0.1348, + "step": 32391 + }, + { + "epoch": 0.2811781147733093, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015570865237062913, + "loss": 0.1201, + "step": 32392 + }, + { + "epoch": 0.2811867952535134, + "grad_norm": 0.287109375, + "learning_rate": 0.0015570608851410996, + "loss": 0.0752, + "step": 32393 + }, + { + "epoch": 0.2811954757337176, + "grad_norm": 0.263671875, + "learning_rate": 0.0015570352460760734, + "loss": 0.0938, + "step": 32394 + }, + { + "epoch": 0.28120415621392175, + "grad_norm": 0.404296875, + "learning_rate": 0.0015570096065112397, + "loss": 0.0991, + "step": 32395 + }, + { + "epoch": 0.28121283669412594, + "grad_norm": 0.07666015625, + "learning_rate": 0.0015569839664466277, + "loss": 0.0859, + "step": 32396 + }, + { + "epoch": 0.2812215171743301, + "grad_norm": 0.158203125, + "learning_rate": 0.0015569583258822643, + "loss": 0.084, + "step": 32397 + }, + { + "epoch": 0.2812301976545343, + "grad_norm": 0.1181640625, + "learning_rate": 0.0015569326848181787, + "loss": 0.127, + "step": 32398 + }, + { + "epoch": 0.2812388781347384, + "grad_norm": 0.376953125, + "learning_rate": 0.001556907043254398, + "loss": 0.127, + "step": 32399 + }, + { + "epoch": 0.2812475586149426, + "grad_norm": 0.427734375, + "learning_rate": 0.0015568814011909509, + "loss": 0.0796, + "step": 32400 + }, + { + "epoch": 0.28125623909514674, + "grad_norm": 0.2197265625, + "learning_rate": 0.0015568557586278654, + "loss": 0.0889, + "step": 32401 + }, + { + "epoch": 0.28126491957535094, + "grad_norm": 0.0888671875, + "learning_rate": 0.0015568301155651687, + "loss": 0.104, + "step": 32402 + }, + { + "epoch": 0.2812736000555551, + "grad_norm": 0.859375, + "learning_rate": 0.0015568044720028899, + "loss": 0.1699, + "step": 32403 + }, + { + "epoch": 0.28128228053575927, + "grad_norm": 0.1396484375, + "learning_rate": 0.0015567788279410566, + "loss": 0.0884, + "step": 32404 + }, + { + "epoch": 0.2812909610159634, + "grad_norm": 0.578125, + "learning_rate": 0.0015567531833796965, + "loss": 0.1133, + "step": 32405 + }, + { + "epoch": 0.28129964149616754, + "grad_norm": 0.27734375, + "learning_rate": 0.001556727538318838, + "loss": 0.1006, + "step": 32406 + }, + { + "epoch": 0.28130832197637173, + "grad_norm": 0.32421875, + "learning_rate": 0.0015567018927585094, + "loss": 0.0903, + "step": 32407 + }, + { + "epoch": 0.28131700245657587, + "grad_norm": 0.10986328125, + "learning_rate": 0.0015566762466987386, + "loss": 0.0947, + "step": 32408 + }, + { + "epoch": 0.28132568293678006, + "grad_norm": 0.365234375, + "learning_rate": 0.001556650600139553, + "loss": 0.1211, + "step": 32409 + }, + { + "epoch": 0.2813343634169842, + "grad_norm": 0.333984375, + "learning_rate": 0.0015566249530809815, + "loss": 0.0835, + "step": 32410 + }, + { + "epoch": 0.2813430438971884, + "grad_norm": 0.220703125, + "learning_rate": 0.0015565993055230518, + "loss": 0.0898, + "step": 32411 + }, + { + "epoch": 0.28135172437739253, + "grad_norm": 0.130859375, + "learning_rate": 0.0015565736574657918, + "loss": 0.1064, + "step": 32412 + }, + { + "epoch": 0.2813604048575967, + "grad_norm": 0.73828125, + "learning_rate": 0.00155654800890923, + "loss": 0.0884, + "step": 32413 + }, + { + "epoch": 0.28136908533780086, + "grad_norm": 0.1474609375, + "learning_rate": 0.001556522359853394, + "loss": 0.1846, + "step": 32414 + }, + { + "epoch": 0.28137776581800505, + "grad_norm": 0.359375, + "learning_rate": 0.0015564967102983118, + "loss": 0.1309, + "step": 32415 + }, + { + "epoch": 0.2813864462982092, + "grad_norm": 0.546875, + "learning_rate": 0.0015564710602440118, + "loss": 0.1094, + "step": 32416 + }, + { + "epoch": 0.2813951267784134, + "grad_norm": 0.306640625, + "learning_rate": 0.0015564454096905218, + "loss": 0.1006, + "step": 32417 + }, + { + "epoch": 0.2814038072586175, + "grad_norm": 0.45703125, + "learning_rate": 0.0015564197586378703, + "loss": 0.1113, + "step": 32418 + }, + { + "epoch": 0.2814124877388217, + "grad_norm": 0.91015625, + "learning_rate": 0.0015563941070860848, + "loss": 0.1777, + "step": 32419 + }, + { + "epoch": 0.28142116821902585, + "grad_norm": 0.216796875, + "learning_rate": 0.0015563684550351936, + "loss": 0.0811, + "step": 32420 + }, + { + "epoch": 0.28142984869923005, + "grad_norm": 0.09765625, + "learning_rate": 0.0015563428024852246, + "loss": 0.125, + "step": 32421 + }, + { + "epoch": 0.2814385291794342, + "grad_norm": 0.1875, + "learning_rate": 0.001556317149436206, + "loss": 0.1084, + "step": 32422 + }, + { + "epoch": 0.2814472096596384, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015562914958881663, + "loss": 0.1387, + "step": 32423 + }, + { + "epoch": 0.2814558901398425, + "grad_norm": 0.091796875, + "learning_rate": 0.0015562658418411324, + "loss": 0.0894, + "step": 32424 + }, + { + "epoch": 0.2814645706200467, + "grad_norm": 0.25390625, + "learning_rate": 0.0015562401872951331, + "loss": 0.1162, + "step": 32425 + }, + { + "epoch": 0.28147325110025084, + "grad_norm": 0.5546875, + "learning_rate": 0.0015562145322501966, + "loss": 0.0923, + "step": 32426 + }, + { + "epoch": 0.28148193158045504, + "grad_norm": 0.72265625, + "learning_rate": 0.001556188876706351, + "loss": 0.1084, + "step": 32427 + }, + { + "epoch": 0.2814906120606592, + "grad_norm": 0.95703125, + "learning_rate": 0.0015561632206636234, + "loss": 0.0879, + "step": 32428 + }, + { + "epoch": 0.28149929254086337, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015561375641220433, + "loss": 0.0742, + "step": 32429 + }, + { + "epoch": 0.2815079730210675, + "grad_norm": 0.8203125, + "learning_rate": 0.0015561119070816374, + "loss": 0.0996, + "step": 32430 + }, + { + "epoch": 0.2815166535012717, + "grad_norm": 0.236328125, + "learning_rate": 0.0015560862495424348, + "loss": 0.1387, + "step": 32431 + }, + { + "epoch": 0.28152533398147583, + "grad_norm": 0.33984375, + "learning_rate": 0.0015560605915044632, + "loss": 0.0947, + "step": 32432 + }, + { + "epoch": 0.28153401446168, + "grad_norm": 1.1015625, + "learning_rate": 0.0015560349329677503, + "loss": 0.1582, + "step": 32433 + }, + { + "epoch": 0.28154269494188416, + "grad_norm": 0.11669921875, + "learning_rate": 0.0015560092739323244, + "loss": 0.1406, + "step": 32434 + }, + { + "epoch": 0.28155137542208836, + "grad_norm": 0.2333984375, + "learning_rate": 0.001555983614398214, + "loss": 0.0957, + "step": 32435 + }, + { + "epoch": 0.2815600559022925, + "grad_norm": 0.263671875, + "learning_rate": 0.001555957954365446, + "loss": 0.0986, + "step": 32436 + }, + { + "epoch": 0.2815687363824967, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015559322938340499, + "loss": 0.0791, + "step": 32437 + }, + { + "epoch": 0.2815774168627008, + "grad_norm": 0.1181640625, + "learning_rate": 0.001555906632804053, + "loss": 0.1211, + "step": 32438 + }, + { + "epoch": 0.281586097342905, + "grad_norm": 0.42578125, + "learning_rate": 0.0015558809712754834, + "loss": 0.1069, + "step": 32439 + }, + { + "epoch": 0.28159477782310915, + "grad_norm": 0.328125, + "learning_rate": 0.0015558553092483694, + "loss": 0.0874, + "step": 32440 + }, + { + "epoch": 0.28160345830331335, + "grad_norm": 0.1484375, + "learning_rate": 0.0015558296467227384, + "loss": 0.085, + "step": 32441 + }, + { + "epoch": 0.2816121387835175, + "grad_norm": 1.078125, + "learning_rate": 0.0015558039836986192, + "loss": 0.126, + "step": 32442 + }, + { + "epoch": 0.2816208192637217, + "grad_norm": 0.296875, + "learning_rate": 0.0015557783201760398, + "loss": 0.0796, + "step": 32443 + }, + { + "epoch": 0.2816294997439258, + "grad_norm": 0.345703125, + "learning_rate": 0.0015557526561550282, + "loss": 0.1035, + "step": 32444 + }, + { + "epoch": 0.28163818022413, + "grad_norm": 0.18359375, + "learning_rate": 0.0015557269916356116, + "loss": 0.0938, + "step": 32445 + }, + { + "epoch": 0.28164686070433415, + "grad_norm": 0.41015625, + "learning_rate": 0.0015557013266178192, + "loss": 0.1006, + "step": 32446 + }, + { + "epoch": 0.28165554118453834, + "grad_norm": 0.30859375, + "learning_rate": 0.001555675661101679, + "loss": 0.1406, + "step": 32447 + }, + { + "epoch": 0.2816642216647425, + "grad_norm": 0.0791015625, + "learning_rate": 0.0015556499950872184, + "loss": 0.0854, + "step": 32448 + }, + { + "epoch": 0.28167290214494667, + "grad_norm": 0.69140625, + "learning_rate": 0.001555624328574466, + "loss": 0.1201, + "step": 32449 + }, + { + "epoch": 0.2816815826251508, + "grad_norm": 0.345703125, + "learning_rate": 0.0015555986615634495, + "loss": 0.0703, + "step": 32450 + }, + { + "epoch": 0.281690263105355, + "grad_norm": 0.36328125, + "learning_rate": 0.001555572994054197, + "loss": 0.1064, + "step": 32451 + }, + { + "epoch": 0.28169894358555914, + "grad_norm": 0.0986328125, + "learning_rate": 0.0015555473260467371, + "loss": 0.1143, + "step": 32452 + }, + { + "epoch": 0.28170762406576333, + "grad_norm": 1.125, + "learning_rate": 0.0015555216575410973, + "loss": 0.0967, + "step": 32453 + }, + { + "epoch": 0.28171630454596747, + "grad_norm": 0.2275390625, + "learning_rate": 0.001555495988537306, + "loss": 0.1143, + "step": 32454 + }, + { + "epoch": 0.28172498502617166, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015554703190353912, + "loss": 0.1182, + "step": 32455 + }, + { + "epoch": 0.2817336655063758, + "grad_norm": 0.130859375, + "learning_rate": 0.0015554446490353804, + "loss": 0.0737, + "step": 32456 + }, + { + "epoch": 0.28174234598658, + "grad_norm": 0.10693359375, + "learning_rate": 0.0015554189785373027, + "loss": 0.1416, + "step": 32457 + }, + { + "epoch": 0.2817510264667841, + "grad_norm": 0.119140625, + "learning_rate": 0.0015553933075411854, + "loss": 0.123, + "step": 32458 + }, + { + "epoch": 0.2817597069469883, + "grad_norm": 0.47265625, + "learning_rate": 0.0015553676360470568, + "loss": 0.1074, + "step": 32459 + }, + { + "epoch": 0.28176838742719246, + "grad_norm": 0.208984375, + "learning_rate": 0.001555341964054945, + "loss": 0.1211, + "step": 32460 + }, + { + "epoch": 0.28177706790739665, + "grad_norm": 0.37109375, + "learning_rate": 0.0015553162915648785, + "loss": 0.0869, + "step": 32461 + }, + { + "epoch": 0.2817857483876008, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015552906185768846, + "loss": 0.1631, + "step": 32462 + }, + { + "epoch": 0.281794428867805, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015552649450909918, + "loss": 0.1211, + "step": 32463 + }, + { + "epoch": 0.2818031093480091, + "grad_norm": 0.71875, + "learning_rate": 0.001555239271107228, + "loss": 0.1221, + "step": 32464 + }, + { + "epoch": 0.2818117898282133, + "grad_norm": 0.5390625, + "learning_rate": 0.001555213596625621, + "loss": 0.0723, + "step": 32465 + }, + { + "epoch": 0.28182047030841745, + "grad_norm": 0.5078125, + "learning_rate": 0.0015551879216461997, + "loss": 0.0952, + "step": 32466 + }, + { + "epoch": 0.28182915078862164, + "grad_norm": 0.123046875, + "learning_rate": 0.0015551622461689913, + "loss": 0.0967, + "step": 32467 + }, + { + "epoch": 0.2818378312688258, + "grad_norm": 0.369140625, + "learning_rate": 0.001555136570194025, + "loss": 0.1299, + "step": 32468 + }, + { + "epoch": 0.28184651174902997, + "grad_norm": 0.177734375, + "learning_rate": 0.0015551108937213276, + "loss": 0.1396, + "step": 32469 + }, + { + "epoch": 0.2818551922292341, + "grad_norm": 0.341796875, + "learning_rate": 0.001555085216750928, + "loss": 0.1084, + "step": 32470 + }, + { + "epoch": 0.2818638727094383, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015550595392828544, + "loss": 0.1162, + "step": 32471 + }, + { + "epoch": 0.28187255318964244, + "grad_norm": 0.625, + "learning_rate": 0.001555033861317134, + "loss": 0.1123, + "step": 32472 + }, + { + "epoch": 0.28188123366984663, + "grad_norm": 0.330078125, + "learning_rate": 0.0015550081828537958, + "loss": 0.127, + "step": 32473 + }, + { + "epoch": 0.28188991415005077, + "grad_norm": 0.494140625, + "learning_rate": 0.0015549825038928668, + "loss": 0.0894, + "step": 32474 + }, + { + "epoch": 0.28189859463025496, + "grad_norm": 0.435546875, + "learning_rate": 0.0015549568244343766, + "loss": 0.1465, + "step": 32475 + }, + { + "epoch": 0.2819072751104591, + "grad_norm": 0.1357421875, + "learning_rate": 0.001554931144478352, + "loss": 0.1035, + "step": 32476 + }, + { + "epoch": 0.2819159555906633, + "grad_norm": 0.2470703125, + "learning_rate": 0.0015549054640248215, + "loss": 0.1436, + "step": 32477 + }, + { + "epoch": 0.28192463607086743, + "grad_norm": 0.171875, + "learning_rate": 0.0015548797830738136, + "loss": 0.1123, + "step": 32478 + }, + { + "epoch": 0.2819333165510716, + "grad_norm": 0.294921875, + "learning_rate": 0.0015548541016253557, + "loss": 0.1328, + "step": 32479 + }, + { + "epoch": 0.28194199703127576, + "grad_norm": 0.384765625, + "learning_rate": 0.001554828419679476, + "loss": 0.1118, + "step": 32480 + }, + { + "epoch": 0.28195067751147995, + "grad_norm": 0.50390625, + "learning_rate": 0.0015548027372362032, + "loss": 0.1055, + "step": 32481 + }, + { + "epoch": 0.2819593579916841, + "grad_norm": 0.447265625, + "learning_rate": 0.001554777054295565, + "loss": 0.0947, + "step": 32482 + }, + { + "epoch": 0.2819680384718883, + "grad_norm": 0.283203125, + "learning_rate": 0.001554751370857589, + "loss": 0.1089, + "step": 32483 + }, + { + "epoch": 0.2819767189520924, + "grad_norm": 0.341796875, + "learning_rate": 0.001554725686922304, + "loss": 0.1328, + "step": 32484 + }, + { + "epoch": 0.2819853994322966, + "grad_norm": 0.294921875, + "learning_rate": 0.0015547000024897378, + "loss": 0.1035, + "step": 32485 + }, + { + "epoch": 0.28199407991250075, + "grad_norm": 0.068359375, + "learning_rate": 0.0015546743175599185, + "loss": 0.0957, + "step": 32486 + }, + { + "epoch": 0.28200276039270494, + "grad_norm": 0.08740234375, + "learning_rate": 0.001554648632132874, + "loss": 0.126, + "step": 32487 + }, + { + "epoch": 0.2820114408729091, + "grad_norm": 0.2578125, + "learning_rate": 0.001554622946208633, + "loss": 0.1631, + "step": 32488 + }, + { + "epoch": 0.2820201213531133, + "grad_norm": 0.09423828125, + "learning_rate": 0.0015545972597872226, + "loss": 0.1406, + "step": 32489 + }, + { + "epoch": 0.2820288018333174, + "grad_norm": 0.1259765625, + "learning_rate": 0.001554571572868672, + "loss": 0.1475, + "step": 32490 + }, + { + "epoch": 0.2820374823135216, + "grad_norm": 0.087890625, + "learning_rate": 0.0015545458854530087, + "loss": 0.0991, + "step": 32491 + }, + { + "epoch": 0.28204616279372574, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015545201975402608, + "loss": 0.1211, + "step": 32492 + }, + { + "epoch": 0.28205484327392993, + "grad_norm": 0.201171875, + "learning_rate": 0.0015544945091304561, + "loss": 0.1055, + "step": 32493 + }, + { + "epoch": 0.28206352375413407, + "grad_norm": 0.404296875, + "learning_rate": 0.0015544688202236232, + "loss": 0.1406, + "step": 32494 + }, + { + "epoch": 0.28207220423433826, + "grad_norm": 0.875, + "learning_rate": 0.00155444313081979, + "loss": 0.1162, + "step": 32495 + }, + { + "epoch": 0.2820808847145424, + "grad_norm": 0.322265625, + "learning_rate": 0.0015544174409189851, + "loss": 0.1104, + "step": 32496 + }, + { + "epoch": 0.2820895651947466, + "grad_norm": 0.3984375, + "learning_rate": 0.0015543917505212355, + "loss": 0.123, + "step": 32497 + }, + { + "epoch": 0.28209824567495073, + "grad_norm": 0.53125, + "learning_rate": 0.0015543660596265703, + "loss": 0.1299, + "step": 32498 + }, + { + "epoch": 0.2821069261551549, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015543403682350173, + "loss": 0.1211, + "step": 32499 + }, + { + "epoch": 0.28211560663535906, + "grad_norm": 0.5390625, + "learning_rate": 0.0015543146763466044, + "loss": 0.1104, + "step": 32500 + }, + { + "epoch": 0.28212428711556325, + "grad_norm": 0.427734375, + "learning_rate": 0.0015542889839613595, + "loss": 0.1094, + "step": 32501 + }, + { + "epoch": 0.2821329675957674, + "grad_norm": 0.09912109375, + "learning_rate": 0.001554263291079311, + "loss": 0.1045, + "step": 32502 + }, + { + "epoch": 0.2821416480759716, + "grad_norm": 0.0888671875, + "learning_rate": 0.0015542375977004872, + "loss": 0.0967, + "step": 32503 + }, + { + "epoch": 0.2821503285561757, + "grad_norm": 0.181640625, + "learning_rate": 0.0015542119038249157, + "loss": 0.1074, + "step": 32504 + }, + { + "epoch": 0.2821590090363799, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015541862094526255, + "loss": 0.0752, + "step": 32505 + }, + { + "epoch": 0.28216768951658405, + "grad_norm": 0.337890625, + "learning_rate": 0.0015541605145836435, + "loss": 0.1191, + "step": 32506 + }, + { + "epoch": 0.28217636999678825, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015541348192179985, + "loss": 0.105, + "step": 32507 + }, + { + "epoch": 0.2821850504769924, + "grad_norm": 0.3203125, + "learning_rate": 0.0015541091233557186, + "loss": 0.1001, + "step": 32508 + }, + { + "epoch": 0.2821937309571966, + "grad_norm": 0.32421875, + "learning_rate": 0.0015540834269968317, + "loss": 0.0879, + "step": 32509 + }, + { + "epoch": 0.2822024114374007, + "grad_norm": 0.12109375, + "learning_rate": 0.0015540577301413662, + "loss": 0.1836, + "step": 32510 + }, + { + "epoch": 0.2822110919176049, + "grad_norm": 0.0908203125, + "learning_rate": 0.0015540320327893497, + "loss": 0.0918, + "step": 32511 + }, + { + "epoch": 0.28221977239780904, + "grad_norm": 0.3671875, + "learning_rate": 0.001554006334940811, + "loss": 0.1641, + "step": 32512 + }, + { + "epoch": 0.28222845287801324, + "grad_norm": 0.203125, + "learning_rate": 0.0015539806365957773, + "loss": 0.1074, + "step": 32513 + }, + { + "epoch": 0.2822371333582174, + "grad_norm": 0.357421875, + "learning_rate": 0.0015539549377542776, + "loss": 0.126, + "step": 32514 + }, + { + "epoch": 0.28224581383842157, + "grad_norm": 0.09765625, + "learning_rate": 0.0015539292384163393, + "loss": 0.0767, + "step": 32515 + }, + { + "epoch": 0.2822544943186257, + "grad_norm": 0.0908203125, + "learning_rate": 0.001553903538581991, + "loss": 0.125, + "step": 32516 + }, + { + "epoch": 0.2822631747988299, + "grad_norm": 0.375, + "learning_rate": 0.0015538778382512607, + "loss": 0.1045, + "step": 32517 + }, + { + "epoch": 0.28227185527903403, + "grad_norm": 0.52734375, + "learning_rate": 0.001553852137424176, + "loss": 0.1035, + "step": 32518 + }, + { + "epoch": 0.2822805357592382, + "grad_norm": 0.25390625, + "learning_rate": 0.0015538264361007659, + "loss": 0.1289, + "step": 32519 + }, + { + "epoch": 0.28228921623944236, + "grad_norm": 0.28515625, + "learning_rate": 0.0015538007342810575, + "loss": 0.1074, + "step": 32520 + }, + { + "epoch": 0.28229789671964656, + "grad_norm": 0.5546875, + "learning_rate": 0.0015537750319650797, + "loss": 0.0776, + "step": 32521 + }, + { + "epoch": 0.2823065771998507, + "grad_norm": 0.11572265625, + "learning_rate": 0.0015537493291528607, + "loss": 0.124, + "step": 32522 + }, + { + "epoch": 0.2823152576800549, + "grad_norm": 0.10986328125, + "learning_rate": 0.0015537236258444278, + "loss": 0.1523, + "step": 32523 + }, + { + "epoch": 0.282323938160259, + "grad_norm": 0.130859375, + "learning_rate": 0.0015536979220398095, + "loss": 0.0908, + "step": 32524 + }, + { + "epoch": 0.2823326186404632, + "grad_norm": 0.3671875, + "learning_rate": 0.0015536722177390344, + "loss": 0.0918, + "step": 32525 + }, + { + "epoch": 0.28234129912066736, + "grad_norm": 0.15625, + "learning_rate": 0.00155364651294213, + "loss": 0.0732, + "step": 32526 + }, + { + "epoch": 0.28234997960087155, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015536208076491246, + "loss": 0.0767, + "step": 32527 + }, + { + "epoch": 0.2823586600810757, + "grad_norm": 0.3671875, + "learning_rate": 0.001553595101860046, + "loss": 0.1035, + "step": 32528 + }, + { + "epoch": 0.2823673405612798, + "grad_norm": 0.546875, + "learning_rate": 0.001553569395574923, + "loss": 0.1001, + "step": 32529 + }, + { + "epoch": 0.282376021041484, + "grad_norm": 0.1826171875, + "learning_rate": 0.0015535436887937832, + "loss": 0.0957, + "step": 32530 + }, + { + "epoch": 0.28238470152168815, + "grad_norm": 0.08984375, + "learning_rate": 0.0015535179815166548, + "loss": 0.1055, + "step": 32531 + }, + { + "epoch": 0.28239338200189235, + "grad_norm": 0.228515625, + "learning_rate": 0.0015534922737435659, + "loss": 0.1123, + "step": 32532 + }, + { + "epoch": 0.2824020624820965, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015534665654745447, + "loss": 0.1279, + "step": 32533 + }, + { + "epoch": 0.2824107429623007, + "grad_norm": 0.1337890625, + "learning_rate": 0.0015534408567096192, + "loss": 0.105, + "step": 32534 + }, + { + "epoch": 0.2824194234425048, + "grad_norm": 0.609375, + "learning_rate": 0.0015534151474488175, + "loss": 0.1035, + "step": 32535 + }, + { + "epoch": 0.282428103922709, + "grad_norm": 0.1162109375, + "learning_rate": 0.001553389437692168, + "loss": 0.1021, + "step": 32536 + }, + { + "epoch": 0.28243678440291314, + "grad_norm": 0.1328125, + "learning_rate": 0.0015533637274396989, + "loss": 0.1074, + "step": 32537 + }, + { + "epoch": 0.28244546488311734, + "grad_norm": 0.70703125, + "learning_rate": 0.0015533380166914378, + "loss": 0.0898, + "step": 32538 + }, + { + "epoch": 0.2824541453633215, + "grad_norm": 0.06298828125, + "learning_rate": 0.001553312305447413, + "loss": 0.064, + "step": 32539 + }, + { + "epoch": 0.28246282584352567, + "grad_norm": 0.55078125, + "learning_rate": 0.0015532865937076526, + "loss": 0.0923, + "step": 32540 + }, + { + "epoch": 0.2824715063237298, + "grad_norm": 0.5703125, + "learning_rate": 0.0015532608814721848, + "loss": 0.0864, + "step": 32541 + }, + { + "epoch": 0.282480186803934, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015532351687410375, + "loss": 0.1699, + "step": 32542 + }, + { + "epoch": 0.28248886728413813, + "grad_norm": 0.0908203125, + "learning_rate": 0.0015532094555142394, + "loss": 0.0957, + "step": 32543 + }, + { + "epoch": 0.2824975477643423, + "grad_norm": 0.474609375, + "learning_rate": 0.0015531837417918184, + "loss": 0.0894, + "step": 32544 + }, + { + "epoch": 0.28250622824454646, + "grad_norm": 0.1650390625, + "learning_rate": 0.001553158027573802, + "loss": 0.1289, + "step": 32545 + }, + { + "epoch": 0.28251490872475066, + "grad_norm": 0.42578125, + "learning_rate": 0.001553132312860219, + "loss": 0.0986, + "step": 32546 + }, + { + "epoch": 0.2825235892049548, + "grad_norm": 0.15625, + "learning_rate": 0.0015531065976510977, + "loss": 0.1279, + "step": 32547 + }, + { + "epoch": 0.282532269685159, + "grad_norm": 0.1328125, + "learning_rate": 0.0015530808819464652, + "loss": 0.1328, + "step": 32548 + }, + { + "epoch": 0.2825409501653631, + "grad_norm": 0.51171875, + "learning_rate": 0.0015530551657463507, + "loss": 0.1279, + "step": 32549 + }, + { + "epoch": 0.2825496306455673, + "grad_norm": 0.11767578125, + "learning_rate": 0.0015530294490507816, + "loss": 0.0962, + "step": 32550 + }, + { + "epoch": 0.28255831112577146, + "grad_norm": 0.236328125, + "learning_rate": 0.0015530037318597866, + "loss": 0.1025, + "step": 32551 + }, + { + "epoch": 0.28256699160597565, + "grad_norm": 0.302734375, + "learning_rate": 0.0015529780141733931, + "loss": 0.1074, + "step": 32552 + }, + { + "epoch": 0.2825756720861798, + "grad_norm": 0.267578125, + "learning_rate": 0.00155295229599163, + "loss": 0.0791, + "step": 32553 + }, + { + "epoch": 0.282584352566384, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015529265773145248, + "loss": 0.0835, + "step": 32554 + }, + { + "epoch": 0.2825930330465881, + "grad_norm": 0.3046875, + "learning_rate": 0.001552900858142106, + "loss": 0.0957, + "step": 32555 + }, + { + "epoch": 0.2826017135267923, + "grad_norm": 0.244140625, + "learning_rate": 0.0015528751384744019, + "loss": 0.1396, + "step": 32556 + }, + { + "epoch": 0.28261039400699645, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015528494183114402, + "loss": 0.1118, + "step": 32557 + }, + { + "epoch": 0.28261907448720064, + "grad_norm": 0.224609375, + "learning_rate": 0.0015528236976532488, + "loss": 0.0928, + "step": 32558 + }, + { + "epoch": 0.2826277549674048, + "grad_norm": 0.181640625, + "learning_rate": 0.0015527979764998565, + "loss": 0.0918, + "step": 32559 + }, + { + "epoch": 0.28263643544760897, + "grad_norm": 0.10791015625, + "learning_rate": 0.0015527722548512916, + "loss": 0.1025, + "step": 32560 + }, + { + "epoch": 0.2826451159278131, + "grad_norm": 0.1416015625, + "learning_rate": 0.001552746532707581, + "loss": 0.1211, + "step": 32561 + }, + { + "epoch": 0.2826537964080173, + "grad_norm": 0.74609375, + "learning_rate": 0.0015527208100687539, + "loss": 0.0786, + "step": 32562 + }, + { + "epoch": 0.28266247688822144, + "grad_norm": 0.51171875, + "learning_rate": 0.0015526950869348382, + "loss": 0.1099, + "step": 32563 + }, + { + "epoch": 0.28267115736842563, + "grad_norm": 0.357421875, + "learning_rate": 0.0015526693633058618, + "loss": 0.1094, + "step": 32564 + }, + { + "epoch": 0.28267983784862977, + "grad_norm": 0.6796875, + "learning_rate": 0.0015526436391818533, + "loss": 0.0879, + "step": 32565 + }, + { + "epoch": 0.28268851832883396, + "grad_norm": 0.447265625, + "learning_rate": 0.0015526179145628402, + "loss": 0.1045, + "step": 32566 + }, + { + "epoch": 0.2826971988090381, + "grad_norm": 0.640625, + "learning_rate": 0.001552592189448851, + "loss": 0.0928, + "step": 32567 + }, + { + "epoch": 0.2827058792892423, + "grad_norm": 0.205078125, + "learning_rate": 0.0015525664638399138, + "loss": 0.0952, + "step": 32568 + }, + { + "epoch": 0.2827145597694464, + "grad_norm": 0.61328125, + "learning_rate": 0.001552540737736057, + "loss": 0.1191, + "step": 32569 + }, + { + "epoch": 0.2827232402496506, + "grad_norm": 0.10009765625, + "learning_rate": 0.001552515011137308, + "loss": 0.0918, + "step": 32570 + }, + { + "epoch": 0.28273192072985476, + "grad_norm": 0.84765625, + "learning_rate": 0.0015524892840436952, + "loss": 0.1465, + "step": 32571 + }, + { + "epoch": 0.28274060121005895, + "grad_norm": 0.474609375, + "learning_rate": 0.0015524635564552474, + "loss": 0.0957, + "step": 32572 + }, + { + "epoch": 0.2827492816902631, + "grad_norm": 0.498046875, + "learning_rate": 0.0015524378283719921, + "loss": 0.0864, + "step": 32573 + }, + { + "epoch": 0.2827579621704673, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015524120997939578, + "loss": 0.127, + "step": 32574 + }, + { + "epoch": 0.2827666426506714, + "grad_norm": 0.42578125, + "learning_rate": 0.0015523863707211723, + "loss": 0.1235, + "step": 32575 + }, + { + "epoch": 0.2827753231308756, + "grad_norm": 0.71875, + "learning_rate": 0.0015523606411536636, + "loss": 0.1221, + "step": 32576 + }, + { + "epoch": 0.28278400361107975, + "grad_norm": 0.310546875, + "learning_rate": 0.0015523349110914606, + "loss": 0.0889, + "step": 32577 + }, + { + "epoch": 0.28279268409128394, + "grad_norm": 0.146484375, + "learning_rate": 0.0015523091805345905, + "loss": 0.1172, + "step": 32578 + }, + { + "epoch": 0.2828013645714881, + "grad_norm": 0.41796875, + "learning_rate": 0.0015522834494830823, + "loss": 0.083, + "step": 32579 + }, + { + "epoch": 0.28281004505169227, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015522577179369632, + "loss": 0.0781, + "step": 32580 + }, + { + "epoch": 0.2828187255318964, + "grad_norm": 0.3125, + "learning_rate": 0.001552231985896262, + "loss": 0.085, + "step": 32581 + }, + { + "epoch": 0.2828274060121006, + "grad_norm": 0.08837890625, + "learning_rate": 0.001552206253361007, + "loss": 0.0938, + "step": 32582 + }, + { + "epoch": 0.28283608649230474, + "grad_norm": 0.08740234375, + "learning_rate": 0.0015521805203312258, + "loss": 0.1104, + "step": 32583 + }, + { + "epoch": 0.28284476697250893, + "grad_norm": 0.3359375, + "learning_rate": 0.0015521547868069467, + "loss": 0.1035, + "step": 32584 + }, + { + "epoch": 0.28285344745271307, + "grad_norm": 0.63671875, + "learning_rate": 0.0015521290527881983, + "loss": 0.0913, + "step": 32585 + }, + { + "epoch": 0.28286212793291726, + "grad_norm": 0.4375, + "learning_rate": 0.001552103318275008, + "loss": 0.0996, + "step": 32586 + }, + { + "epoch": 0.2828708084131214, + "grad_norm": 0.09375, + "learning_rate": 0.0015520775832674044, + "loss": 0.0879, + "step": 32587 + }, + { + "epoch": 0.2828794888933256, + "grad_norm": 0.23046875, + "learning_rate": 0.0015520518477654152, + "loss": 0.0781, + "step": 32588 + }, + { + "epoch": 0.28288816937352973, + "grad_norm": 0.357421875, + "learning_rate": 0.0015520261117690694, + "loss": 0.0879, + "step": 32589 + }, + { + "epoch": 0.2828968498537339, + "grad_norm": 0.765625, + "learning_rate": 0.0015520003752783945, + "loss": 0.1465, + "step": 32590 + }, + { + "epoch": 0.28290553033393806, + "grad_norm": 0.515625, + "learning_rate": 0.0015519746382934188, + "loss": 0.1357, + "step": 32591 + }, + { + "epoch": 0.28291421081414225, + "grad_norm": 0.1845703125, + "learning_rate": 0.0015519489008141705, + "loss": 0.1191, + "step": 32592 + }, + { + "epoch": 0.2829228912943464, + "grad_norm": 0.42578125, + "learning_rate": 0.0015519231628406771, + "loss": 0.1836, + "step": 32593 + }, + { + "epoch": 0.2829315717745506, + "grad_norm": 0.60546875, + "learning_rate": 0.001551897424372968, + "loss": 0.1152, + "step": 32594 + }, + { + "epoch": 0.2829402522547547, + "grad_norm": 0.50390625, + "learning_rate": 0.0015518716854110706, + "loss": 0.1128, + "step": 32595 + }, + { + "epoch": 0.2829489327349589, + "grad_norm": 0.330078125, + "learning_rate": 0.001551845945955013, + "loss": 0.1348, + "step": 32596 + }, + { + "epoch": 0.28295761321516305, + "grad_norm": 0.078125, + "learning_rate": 0.0015518202060048232, + "loss": 0.0898, + "step": 32597 + }, + { + "epoch": 0.28296629369536724, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015517944655605298, + "loss": 0.1021, + "step": 32598 + }, + { + "epoch": 0.2829749741755714, + "grad_norm": 0.109375, + "learning_rate": 0.0015517687246221607, + "loss": 0.0908, + "step": 32599 + }, + { + "epoch": 0.2829836546557756, + "grad_norm": 0.53125, + "learning_rate": 0.0015517429831897443, + "loss": 0.166, + "step": 32600 + }, + { + "epoch": 0.2829923351359797, + "grad_norm": 0.142578125, + "learning_rate": 0.0015517172412633084, + "loss": 0.1455, + "step": 32601 + }, + { + "epoch": 0.2830010156161839, + "grad_norm": 0.279296875, + "learning_rate": 0.0015516914988428815, + "loss": 0.0762, + "step": 32602 + }, + { + "epoch": 0.28300969609638804, + "grad_norm": 0.1875, + "learning_rate": 0.0015516657559284913, + "loss": 0.1484, + "step": 32603 + }, + { + "epoch": 0.28301837657659223, + "grad_norm": 0.373046875, + "learning_rate": 0.0015516400125201665, + "loss": 0.0947, + "step": 32604 + }, + { + "epoch": 0.28302705705679637, + "grad_norm": 0.1904296875, + "learning_rate": 0.001551614268617935, + "loss": 0.0811, + "step": 32605 + }, + { + "epoch": 0.28303573753700056, + "grad_norm": 0.28125, + "learning_rate": 0.0015515885242218244, + "loss": 0.1504, + "step": 32606 + }, + { + "epoch": 0.2830444180172047, + "grad_norm": 0.3515625, + "learning_rate": 0.0015515627793318636, + "loss": 0.0923, + "step": 32607 + }, + { + "epoch": 0.2830530984974089, + "grad_norm": 0.15234375, + "learning_rate": 0.0015515370339480809, + "loss": 0.123, + "step": 32608 + }, + { + "epoch": 0.28306177897761303, + "grad_norm": 0.251953125, + "learning_rate": 0.0015515112880705039, + "loss": 0.0869, + "step": 32609 + }, + { + "epoch": 0.2830704594578172, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015514855416991604, + "loss": 0.0791, + "step": 32610 + }, + { + "epoch": 0.28307913993802136, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015514597948340799, + "loss": 0.1064, + "step": 32611 + }, + { + "epoch": 0.28308782041822556, + "grad_norm": 0.2392578125, + "learning_rate": 0.0015514340474752893, + "loss": 0.124, + "step": 32612 + }, + { + "epoch": 0.2830965008984297, + "grad_norm": 0.158203125, + "learning_rate": 0.0015514082996228174, + "loss": 0.0854, + "step": 32613 + }, + { + "epoch": 0.2831051813786339, + "grad_norm": 0.244140625, + "learning_rate": 0.0015513825512766919, + "loss": 0.1182, + "step": 32614 + }, + { + "epoch": 0.283113861858838, + "grad_norm": 0.6484375, + "learning_rate": 0.0015513568024369414, + "loss": 0.1963, + "step": 32615 + }, + { + "epoch": 0.2831225423390422, + "grad_norm": 0.1806640625, + "learning_rate": 0.001551331053103594, + "loss": 0.0703, + "step": 32616 + }, + { + "epoch": 0.28313122281924635, + "grad_norm": 0.1875, + "learning_rate": 0.0015513053032766776, + "loss": 0.0957, + "step": 32617 + }, + { + "epoch": 0.28313990329945055, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015512795529562206, + "loss": 0.0977, + "step": 32618 + }, + { + "epoch": 0.2831485837796547, + "grad_norm": 0.64453125, + "learning_rate": 0.001551253802142251, + "loss": 0.0981, + "step": 32619 + }, + { + "epoch": 0.2831572642598589, + "grad_norm": 0.283203125, + "learning_rate": 0.0015512280508347969, + "loss": 0.1006, + "step": 32620 + }, + { + "epoch": 0.283165944740063, + "grad_norm": 0.875, + "learning_rate": 0.0015512022990338869, + "loss": 0.1416, + "step": 32621 + }, + { + "epoch": 0.2831746252202672, + "grad_norm": 0.75390625, + "learning_rate": 0.0015511765467395486, + "loss": 0.1079, + "step": 32622 + }, + { + "epoch": 0.28318330570047134, + "grad_norm": 0.271484375, + "learning_rate": 0.0015511507939518104, + "loss": 0.1074, + "step": 32623 + }, + { + "epoch": 0.28319198618067554, + "grad_norm": 0.279296875, + "learning_rate": 0.0015511250406707005, + "loss": 0.1025, + "step": 32624 + }, + { + "epoch": 0.2832006666608797, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015510992868962472, + "loss": 0.1035, + "step": 32625 + }, + { + "epoch": 0.28320934714108387, + "grad_norm": 0.11767578125, + "learning_rate": 0.0015510735326284785, + "loss": 0.1045, + "step": 32626 + }, + { + "epoch": 0.283218027621288, + "grad_norm": 0.232421875, + "learning_rate": 0.0015510477778674222, + "loss": 0.0894, + "step": 32627 + }, + { + "epoch": 0.2832267081014922, + "grad_norm": 0.1201171875, + "learning_rate": 0.001551022022613107, + "loss": 0.1211, + "step": 32628 + }, + { + "epoch": 0.28323538858169633, + "grad_norm": 0.1142578125, + "learning_rate": 0.001550996266865561, + "loss": 0.0874, + "step": 32629 + }, + { + "epoch": 0.2832440690619005, + "grad_norm": 0.470703125, + "learning_rate": 0.001550970510624812, + "loss": 0.126, + "step": 32630 + }, + { + "epoch": 0.28325274954210466, + "grad_norm": 0.1328125, + "learning_rate": 0.0015509447538908886, + "loss": 0.1279, + "step": 32631 + }, + { + "epoch": 0.28326143002230886, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015509189966638186, + "loss": 0.1396, + "step": 32632 + }, + { + "epoch": 0.283270110502513, + "grad_norm": 0.1796875, + "learning_rate": 0.0015508932389436307, + "loss": 0.1055, + "step": 32633 + }, + { + "epoch": 0.2832787909827172, + "grad_norm": 0.60546875, + "learning_rate": 0.001550867480730353, + "loss": 0.1162, + "step": 32634 + }, + { + "epoch": 0.2832874714629213, + "grad_norm": 0.66015625, + "learning_rate": 0.001550841722024013, + "loss": 0.1162, + "step": 32635 + }, + { + "epoch": 0.2832961519431255, + "grad_norm": 0.10986328125, + "learning_rate": 0.001550815962824639, + "loss": 0.0786, + "step": 32636 + }, + { + "epoch": 0.28330483242332966, + "grad_norm": 0.1337890625, + "learning_rate": 0.0015507902031322597, + "loss": 0.1055, + "step": 32637 + }, + { + "epoch": 0.28331351290353385, + "grad_norm": 0.33984375, + "learning_rate": 0.001550764442946903, + "loss": 0.1084, + "step": 32638 + }, + { + "epoch": 0.283322193383738, + "grad_norm": 0.65234375, + "learning_rate": 0.0015507386822685972, + "loss": 0.1099, + "step": 32639 + }, + { + "epoch": 0.2833308738639422, + "grad_norm": 0.73828125, + "learning_rate": 0.00155071292109737, + "loss": 0.0977, + "step": 32640 + }, + { + "epoch": 0.2833395543441463, + "grad_norm": 0.07470703125, + "learning_rate": 0.0015506871594332503, + "loss": 0.082, + "step": 32641 + }, + { + "epoch": 0.2833482348243505, + "grad_norm": 0.380859375, + "learning_rate": 0.0015506613972762655, + "loss": 0.1021, + "step": 32642 + }, + { + "epoch": 0.28335691530455465, + "grad_norm": 0.375, + "learning_rate": 0.0015506356346264448, + "loss": 0.0918, + "step": 32643 + }, + { + "epoch": 0.28336559578475884, + "grad_norm": 0.388671875, + "learning_rate": 0.0015506098714838151, + "loss": 0.127, + "step": 32644 + }, + { + "epoch": 0.283374276264963, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015505841078484058, + "loss": 0.1133, + "step": 32645 + }, + { + "epoch": 0.28338295674516717, + "grad_norm": 0.10107421875, + "learning_rate": 0.001550558343720244, + "loss": 0.126, + "step": 32646 + }, + { + "epoch": 0.2833916372253713, + "grad_norm": 0.76171875, + "learning_rate": 0.0015505325790993584, + "loss": 0.0747, + "step": 32647 + }, + { + "epoch": 0.2834003177055755, + "grad_norm": 0.287109375, + "learning_rate": 0.0015505068139857773, + "loss": 0.106, + "step": 32648 + }, + { + "epoch": 0.28340899818577964, + "grad_norm": 0.357421875, + "learning_rate": 0.0015504810483795287, + "loss": 0.0884, + "step": 32649 + }, + { + "epoch": 0.28341767866598383, + "grad_norm": 0.26171875, + "learning_rate": 0.0015504552822806405, + "loss": 0.1201, + "step": 32650 + }, + { + "epoch": 0.28342635914618797, + "grad_norm": 0.32421875, + "learning_rate": 0.0015504295156891419, + "loss": 0.0967, + "step": 32651 + }, + { + "epoch": 0.2834350396263921, + "grad_norm": 0.2734375, + "learning_rate": 0.0015504037486050598, + "loss": 0.0957, + "step": 32652 + }, + { + "epoch": 0.2834437201065963, + "grad_norm": 0.078125, + "learning_rate": 0.001550377981028423, + "loss": 0.0874, + "step": 32653 + }, + { + "epoch": 0.28345240058680043, + "grad_norm": 0.08935546875, + "learning_rate": 0.0015503522129592595, + "loss": 0.1064, + "step": 32654 + }, + { + "epoch": 0.2834610810670046, + "grad_norm": 0.25390625, + "learning_rate": 0.0015503264443975978, + "loss": 0.1079, + "step": 32655 + }, + { + "epoch": 0.28346976154720877, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015503006753434657, + "loss": 0.0889, + "step": 32656 + }, + { + "epoch": 0.28347844202741296, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015502749057968916, + "loss": 0.1084, + "step": 32657 + }, + { + "epoch": 0.2834871225076171, + "grad_norm": 0.0810546875, + "learning_rate": 0.0015502491357579037, + "loss": 0.1089, + "step": 32658 + }, + { + "epoch": 0.2834958029878213, + "grad_norm": 0.10498046875, + "learning_rate": 0.00155022336522653, + "loss": 0.0874, + "step": 32659 + }, + { + "epoch": 0.2835044834680254, + "grad_norm": 0.578125, + "learning_rate": 0.001550197594202799, + "loss": 0.1162, + "step": 32660 + }, + { + "epoch": 0.2835131639482296, + "grad_norm": 0.130859375, + "learning_rate": 0.0015501718226867388, + "loss": 0.1484, + "step": 32661 + }, + { + "epoch": 0.28352184442843376, + "grad_norm": 0.2138671875, + "learning_rate": 0.001550146050678377, + "loss": 0.0957, + "step": 32662 + }, + { + "epoch": 0.28353052490863795, + "grad_norm": 0.1923828125, + "learning_rate": 0.0015501202781777424, + "loss": 0.0781, + "step": 32663 + }, + { + "epoch": 0.2835392053888421, + "grad_norm": 0.271484375, + "learning_rate": 0.001550094505184863, + "loss": 0.1123, + "step": 32664 + }, + { + "epoch": 0.2835478858690463, + "grad_norm": 0.25390625, + "learning_rate": 0.0015500687316997673, + "loss": 0.0938, + "step": 32665 + }, + { + "epoch": 0.2835565663492504, + "grad_norm": 0.392578125, + "learning_rate": 0.0015500429577224829, + "loss": 0.1416, + "step": 32666 + }, + { + "epoch": 0.2835652468294546, + "grad_norm": 0.07470703125, + "learning_rate": 0.001550017183253038, + "loss": 0.0859, + "step": 32667 + }, + { + "epoch": 0.28357392730965875, + "grad_norm": 0.8515625, + "learning_rate": 0.0015499914082914617, + "loss": 0.1152, + "step": 32668 + }, + { + "epoch": 0.28358260778986294, + "grad_norm": 1.3046875, + "learning_rate": 0.001549965632837781, + "loss": 0.1953, + "step": 32669 + }, + { + "epoch": 0.2835912882700671, + "grad_norm": 0.1513671875, + "learning_rate": 0.001549939856892025, + "loss": 0.1152, + "step": 32670 + }, + { + "epoch": 0.28359996875027127, + "grad_norm": 0.08203125, + "learning_rate": 0.0015499140804542215, + "loss": 0.1025, + "step": 32671 + }, + { + "epoch": 0.2836086492304754, + "grad_norm": 0.30078125, + "learning_rate": 0.0015498883035243987, + "loss": 0.1064, + "step": 32672 + }, + { + "epoch": 0.2836173297106796, + "grad_norm": 0.1689453125, + "learning_rate": 0.0015498625261025848, + "loss": 0.0854, + "step": 32673 + }, + { + "epoch": 0.28362601019088374, + "grad_norm": 0.1328125, + "learning_rate": 0.0015498367481888082, + "loss": 0.0981, + "step": 32674 + }, + { + "epoch": 0.28363469067108793, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015498109697830967, + "loss": 0.1143, + "step": 32675 + }, + { + "epoch": 0.28364337115129207, + "grad_norm": 0.71875, + "learning_rate": 0.0015497851908854786, + "loss": 0.123, + "step": 32676 + }, + { + "epoch": 0.28365205163149626, + "grad_norm": 0.185546875, + "learning_rate": 0.0015497594114959822, + "loss": 0.0928, + "step": 32677 + }, + { + "epoch": 0.2836607321117004, + "grad_norm": 0.6953125, + "learning_rate": 0.0015497336316146358, + "loss": 0.1123, + "step": 32678 + }, + { + "epoch": 0.2836694125919046, + "grad_norm": 0.193359375, + "learning_rate": 0.001549707851241467, + "loss": 0.126, + "step": 32679 + }, + { + "epoch": 0.28367809307210873, + "grad_norm": 0.271484375, + "learning_rate": 0.001549682070376505, + "loss": 0.0913, + "step": 32680 + }, + { + "epoch": 0.2836867735523129, + "grad_norm": 0.326171875, + "learning_rate": 0.0015496562890197772, + "loss": 0.084, + "step": 32681 + }, + { + "epoch": 0.28369545403251706, + "grad_norm": 0.306640625, + "learning_rate": 0.001549630507171312, + "loss": 0.0942, + "step": 32682 + }, + { + "epoch": 0.28370413451272125, + "grad_norm": 0.0947265625, + "learning_rate": 0.001549604724831138, + "loss": 0.1113, + "step": 32683 + }, + { + "epoch": 0.2837128149929254, + "grad_norm": 0.28125, + "learning_rate": 0.0015495789419992826, + "loss": 0.0947, + "step": 32684 + }, + { + "epoch": 0.2837214954731296, + "grad_norm": 0.2265625, + "learning_rate": 0.0015495531586757747, + "loss": 0.1152, + "step": 32685 + }, + { + "epoch": 0.2837301759533337, + "grad_norm": 0.40234375, + "learning_rate": 0.001549527374860642, + "loss": 0.1309, + "step": 32686 + }, + { + "epoch": 0.2837388564335379, + "grad_norm": 0.123046875, + "learning_rate": 0.001549501590553913, + "loss": 0.105, + "step": 32687 + }, + { + "epoch": 0.28374753691374205, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015494758057556157, + "loss": 0.0928, + "step": 32688 + }, + { + "epoch": 0.28375621739394624, + "grad_norm": 0.41015625, + "learning_rate": 0.0015494500204657786, + "loss": 0.1094, + "step": 32689 + }, + { + "epoch": 0.2837648978741504, + "grad_norm": 0.2255859375, + "learning_rate": 0.0015494242346844299, + "loss": 0.124, + "step": 32690 + }, + { + "epoch": 0.28377357835435457, + "grad_norm": 0.640625, + "learning_rate": 0.0015493984484115974, + "loss": 0.0986, + "step": 32691 + }, + { + "epoch": 0.2837822588345587, + "grad_norm": 0.4609375, + "learning_rate": 0.0015493726616473092, + "loss": 0.0625, + "step": 32692 + }, + { + "epoch": 0.2837909393147629, + "grad_norm": 0.74609375, + "learning_rate": 0.0015493468743915941, + "loss": 0.1133, + "step": 32693 + }, + { + "epoch": 0.28379961979496704, + "grad_norm": 0.73046875, + "learning_rate": 0.0015493210866444802, + "loss": 0.0967, + "step": 32694 + }, + { + "epoch": 0.28380830027517123, + "grad_norm": 0.265625, + "learning_rate": 0.0015492952984059952, + "loss": 0.1152, + "step": 32695 + }, + { + "epoch": 0.28381698075537537, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015492695096761678, + "loss": 0.1133, + "step": 32696 + }, + { + "epoch": 0.28382566123557956, + "grad_norm": 0.92578125, + "learning_rate": 0.0015492437204550256, + "loss": 0.1318, + "step": 32697 + }, + { + "epoch": 0.2838343417157837, + "grad_norm": 0.181640625, + "learning_rate": 0.0015492179307425979, + "loss": 0.1318, + "step": 32698 + }, + { + "epoch": 0.2838430221959879, + "grad_norm": 0.294921875, + "learning_rate": 0.0015491921405389117, + "loss": 0.1113, + "step": 32699 + }, + { + "epoch": 0.28385170267619203, + "grad_norm": 0.5546875, + "learning_rate": 0.0015491663498439958, + "loss": 0.1128, + "step": 32700 + }, + { + "epoch": 0.2838603831563962, + "grad_norm": 0.44140625, + "learning_rate": 0.0015491405586578787, + "loss": 0.0791, + "step": 32701 + }, + { + "epoch": 0.28386906363660036, + "grad_norm": 0.3125, + "learning_rate": 0.0015491147669805879, + "loss": 0.0801, + "step": 32702 + }, + { + "epoch": 0.28387774411680455, + "grad_norm": 0.22265625, + "learning_rate": 0.001549088974812152, + "loss": 0.1338, + "step": 32703 + }, + { + "epoch": 0.2838864245970087, + "grad_norm": 0.5234375, + "learning_rate": 0.001549063182152599, + "loss": 0.1172, + "step": 32704 + }, + { + "epoch": 0.2838951050772129, + "grad_norm": 0.498046875, + "learning_rate": 0.0015490373890019577, + "loss": 0.1367, + "step": 32705 + }, + { + "epoch": 0.283903785557417, + "grad_norm": 0.74609375, + "learning_rate": 0.0015490115953602553, + "loss": 0.126, + "step": 32706 + }, + { + "epoch": 0.2839124660376212, + "grad_norm": 0.65625, + "learning_rate": 0.0015489858012275209, + "loss": 0.0898, + "step": 32707 + }, + { + "epoch": 0.28392114651782535, + "grad_norm": 0.31640625, + "learning_rate": 0.0015489600066037822, + "loss": 0.127, + "step": 32708 + }, + { + "epoch": 0.28392982699802954, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015489342114890677, + "loss": 0.1162, + "step": 32709 + }, + { + "epoch": 0.2839385074782337, + "grad_norm": 0.2373046875, + "learning_rate": 0.0015489084158834055, + "loss": 0.1172, + "step": 32710 + }, + { + "epoch": 0.2839471879584379, + "grad_norm": 0.181640625, + "learning_rate": 0.0015488826197868238, + "loss": 0.1504, + "step": 32711 + }, + { + "epoch": 0.283955868438642, + "grad_norm": 0.265625, + "learning_rate": 0.0015488568231993506, + "loss": 0.0942, + "step": 32712 + }, + { + "epoch": 0.2839645489188462, + "grad_norm": 0.1748046875, + "learning_rate": 0.0015488310261210147, + "loss": 0.1152, + "step": 32713 + }, + { + "epoch": 0.28397322939905034, + "grad_norm": 0.318359375, + "learning_rate": 0.0015488052285518438, + "loss": 0.123, + "step": 32714 + }, + { + "epoch": 0.28398190987925453, + "grad_norm": 0.357421875, + "learning_rate": 0.0015487794304918658, + "loss": 0.0898, + "step": 32715 + }, + { + "epoch": 0.28399059035945867, + "grad_norm": 0.1669921875, + "learning_rate": 0.00154875363194111, + "loss": 0.1426, + "step": 32716 + }, + { + "epoch": 0.28399927083966287, + "grad_norm": 0.21875, + "learning_rate": 0.0015487278328996034, + "loss": 0.1094, + "step": 32717 + }, + { + "epoch": 0.284007951319867, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015487020333673752, + "loss": 0.1045, + "step": 32718 + }, + { + "epoch": 0.2840166318000712, + "grad_norm": 0.16796875, + "learning_rate": 0.001548676233344453, + "loss": 0.0835, + "step": 32719 + }, + { + "epoch": 0.28402531228027533, + "grad_norm": 0.1572265625, + "learning_rate": 0.0015486504328308655, + "loss": 0.0708, + "step": 32720 + }, + { + "epoch": 0.2840339927604795, + "grad_norm": 0.22265625, + "learning_rate": 0.0015486246318266405, + "loss": 0.1094, + "step": 32721 + }, + { + "epoch": 0.28404267324068366, + "grad_norm": 0.30078125, + "learning_rate": 0.0015485988303318061, + "loss": 0.0957, + "step": 32722 + }, + { + "epoch": 0.28405135372088786, + "grad_norm": 0.33203125, + "learning_rate": 0.0015485730283463911, + "loss": 0.0752, + "step": 32723 + }, + { + "epoch": 0.284060034201092, + "grad_norm": 0.1259765625, + "learning_rate": 0.001548547225870423, + "loss": 0.1211, + "step": 32724 + }, + { + "epoch": 0.2840687146812962, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015485214229039306, + "loss": 0.1216, + "step": 32725 + }, + { + "epoch": 0.2840773951615003, + "grad_norm": 0.6171875, + "learning_rate": 0.0015484956194469419, + "loss": 0.1094, + "step": 32726 + }, + { + "epoch": 0.2840860756417045, + "grad_norm": 0.166015625, + "learning_rate": 0.0015484698154994852, + "loss": 0.1084, + "step": 32727 + }, + { + "epoch": 0.28409475612190865, + "grad_norm": 0.3515625, + "learning_rate": 0.0015484440110615887, + "loss": 0.1533, + "step": 32728 + }, + { + "epoch": 0.28410343660211285, + "grad_norm": 0.099609375, + "learning_rate": 0.0015484182061332803, + "loss": 0.1396, + "step": 32729 + }, + { + "epoch": 0.284112117082317, + "grad_norm": 0.2353515625, + "learning_rate": 0.001548392400714589, + "loss": 0.1416, + "step": 32730 + }, + { + "epoch": 0.2841207975625212, + "grad_norm": 0.51171875, + "learning_rate": 0.0015483665948055416, + "loss": 0.123, + "step": 32731 + }, + { + "epoch": 0.2841294780427253, + "grad_norm": 0.37890625, + "learning_rate": 0.001548340788406168, + "loss": 0.104, + "step": 32732 + }, + { + "epoch": 0.2841381585229295, + "grad_norm": 0.31640625, + "learning_rate": 0.0015483149815164953, + "loss": 0.1367, + "step": 32733 + }, + { + "epoch": 0.28414683900313364, + "grad_norm": 0.287109375, + "learning_rate": 0.0015482891741365524, + "loss": 0.1064, + "step": 32734 + }, + { + "epoch": 0.28415551948333784, + "grad_norm": 0.39453125, + "learning_rate": 0.0015482633662663669, + "loss": 0.123, + "step": 32735 + }, + { + "epoch": 0.284164199963542, + "grad_norm": 0.44921875, + "learning_rate": 0.0015482375579059674, + "loss": 0.1572, + "step": 32736 + }, + { + "epoch": 0.28417288044374617, + "grad_norm": 0.203125, + "learning_rate": 0.001548211749055382, + "loss": 0.0908, + "step": 32737 + }, + { + "epoch": 0.2841815609239503, + "grad_norm": 5.46875, + "learning_rate": 0.0015481859397146389, + "loss": 0.3379, + "step": 32738 + }, + { + "epoch": 0.2841902414041545, + "grad_norm": 0.220703125, + "learning_rate": 0.0015481601298837663, + "loss": 0.0684, + "step": 32739 + }, + { + "epoch": 0.28419892188435864, + "grad_norm": 0.2236328125, + "learning_rate": 0.001548134319562793, + "loss": 0.1162, + "step": 32740 + }, + { + "epoch": 0.28420760236456283, + "grad_norm": 0.205078125, + "learning_rate": 0.0015481085087517465, + "loss": 0.0928, + "step": 32741 + }, + { + "epoch": 0.28421628284476697, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015480826974506555, + "loss": 0.0942, + "step": 32742 + }, + { + "epoch": 0.28422496332497116, + "grad_norm": 0.7578125, + "learning_rate": 0.0015480568856595473, + "loss": 0.1001, + "step": 32743 + }, + { + "epoch": 0.2842336438051753, + "grad_norm": 0.302734375, + "learning_rate": 0.0015480310733784514, + "loss": 0.0762, + "step": 32744 + }, + { + "epoch": 0.2842423242853795, + "grad_norm": 0.44140625, + "learning_rate": 0.0015480052606073953, + "loss": 0.1025, + "step": 32745 + }, + { + "epoch": 0.2842510047655836, + "grad_norm": 0.1328125, + "learning_rate": 0.0015479794473464074, + "loss": 0.1406, + "step": 32746 + }, + { + "epoch": 0.2842596852457878, + "grad_norm": 0.353515625, + "learning_rate": 0.0015479536335955158, + "loss": 0.1543, + "step": 32747 + }, + { + "epoch": 0.28426836572599196, + "grad_norm": 0.220703125, + "learning_rate": 0.0015479278193547487, + "loss": 0.0918, + "step": 32748 + }, + { + "epoch": 0.28427704620619615, + "grad_norm": 0.2392578125, + "learning_rate": 0.001547902004624135, + "loss": 0.1289, + "step": 32749 + }, + { + "epoch": 0.2842857266864003, + "grad_norm": 0.1650390625, + "learning_rate": 0.001547876189403702, + "loss": 0.1494, + "step": 32750 + }, + { + "epoch": 0.2842944071666045, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015478503736934786, + "loss": 0.1328, + "step": 32751 + }, + { + "epoch": 0.2843030876468086, + "grad_norm": 0.20703125, + "learning_rate": 0.0015478245574934925, + "loss": 0.1709, + "step": 32752 + }, + { + "epoch": 0.2843117681270128, + "grad_norm": 0.435546875, + "learning_rate": 0.0015477987408037722, + "loss": 0.0854, + "step": 32753 + }, + { + "epoch": 0.28432044860721695, + "grad_norm": 0.30859375, + "learning_rate": 0.001547772923624346, + "loss": 0.082, + "step": 32754 + }, + { + "epoch": 0.28432912908742114, + "grad_norm": 0.81640625, + "learning_rate": 0.001547747105955242, + "loss": 0.1162, + "step": 32755 + }, + { + "epoch": 0.2843378095676253, + "grad_norm": 0.26953125, + "learning_rate": 0.0015477212877964888, + "loss": 0.0879, + "step": 32756 + }, + { + "epoch": 0.28434649004782947, + "grad_norm": 0.134765625, + "learning_rate": 0.0015476954691481138, + "loss": 0.0879, + "step": 32757 + }, + { + "epoch": 0.2843551705280336, + "grad_norm": 0.2216796875, + "learning_rate": 0.0015476696500101462, + "loss": 0.0791, + "step": 32758 + }, + { + "epoch": 0.2843638510082378, + "grad_norm": 0.1640625, + "learning_rate": 0.0015476438303826136, + "loss": 0.0918, + "step": 32759 + }, + { + "epoch": 0.28437253148844194, + "grad_norm": 0.828125, + "learning_rate": 0.0015476180102655444, + "loss": 0.1099, + "step": 32760 + }, + { + "epoch": 0.28438121196864613, + "grad_norm": 0.08642578125, + "learning_rate": 0.001547592189658967, + "loss": 0.1094, + "step": 32761 + }, + { + "epoch": 0.28438989244885027, + "grad_norm": 0.40234375, + "learning_rate": 0.0015475663685629096, + "loss": 0.0996, + "step": 32762 + }, + { + "epoch": 0.28439857292905446, + "grad_norm": 0.177734375, + "learning_rate": 0.0015475405469773998, + "loss": 0.1055, + "step": 32763 + }, + { + "epoch": 0.2844072534092586, + "grad_norm": 0.09619140625, + "learning_rate": 0.001547514724902467, + "loss": 0.1211, + "step": 32764 + }, + { + "epoch": 0.2844159338894628, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015474889023381385, + "loss": 0.0801, + "step": 32765 + }, + { + "epoch": 0.28442461436966693, + "grad_norm": 0.1328125, + "learning_rate": 0.001547463079284443, + "loss": 0.1172, + "step": 32766 + }, + { + "epoch": 0.2844332948498711, + "grad_norm": 0.0927734375, + "learning_rate": 0.0015474372557414086, + "loss": 0.1035, + "step": 32767 + }, + { + "epoch": 0.28444197533007526, + "grad_norm": 0.255859375, + "learning_rate": 0.0015474114317090633, + "loss": 0.1035, + "step": 32768 + }, + { + "epoch": 0.28445065581027945, + "grad_norm": 0.64453125, + "learning_rate": 0.0015473856071874358, + "loss": 0.1123, + "step": 32769 + }, + { + "epoch": 0.2844593362904836, + "grad_norm": 0.09619140625, + "learning_rate": 0.0015473597821765539, + "loss": 0.1406, + "step": 32770 + }, + { + "epoch": 0.2844680167706878, + "grad_norm": 0.251953125, + "learning_rate": 0.0015473339566764465, + "loss": 0.1045, + "step": 32771 + }, + { + "epoch": 0.2844766972508919, + "grad_norm": 0.171875, + "learning_rate": 0.001547308130687141, + "loss": 0.0771, + "step": 32772 + }, + { + "epoch": 0.2844853777310961, + "grad_norm": 0.0693359375, + "learning_rate": 0.0015472823042086664, + "loss": 0.0811, + "step": 32773 + }, + { + "epoch": 0.28449405821130025, + "grad_norm": 0.6484375, + "learning_rate": 0.00154725647724105, + "loss": 0.1055, + "step": 32774 + }, + { + "epoch": 0.2845027386915044, + "grad_norm": 0.7265625, + "learning_rate": 0.001547230649784321, + "loss": 0.0967, + "step": 32775 + }, + { + "epoch": 0.2845114191717086, + "grad_norm": 0.166015625, + "learning_rate": 0.0015472048218385078, + "loss": 0.1328, + "step": 32776 + }, + { + "epoch": 0.2845200996519127, + "grad_norm": 0.134765625, + "learning_rate": 0.0015471789934036375, + "loss": 0.1104, + "step": 32777 + }, + { + "epoch": 0.2845287801321169, + "grad_norm": 0.271484375, + "learning_rate": 0.001547153164479739, + "loss": 0.103, + "step": 32778 + }, + { + "epoch": 0.28453746061232105, + "grad_norm": 1.4453125, + "learning_rate": 0.0015471273350668407, + "loss": 0.1523, + "step": 32779 + }, + { + "epoch": 0.28454614109252524, + "grad_norm": 0.19921875, + "learning_rate": 0.0015471015051649706, + "loss": 0.1064, + "step": 32780 + }, + { + "epoch": 0.2845548215727294, + "grad_norm": 0.1728515625, + "learning_rate": 0.001547075674774157, + "loss": 0.1025, + "step": 32781 + }, + { + "epoch": 0.28456350205293357, + "grad_norm": 0.13671875, + "learning_rate": 0.0015470498438944283, + "loss": 0.1328, + "step": 32782 + }, + { + "epoch": 0.2845721825331377, + "grad_norm": 0.134765625, + "learning_rate": 0.0015470240125258121, + "loss": 0.0913, + "step": 32783 + }, + { + "epoch": 0.2845808630133419, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015469981806683374, + "loss": 0.0898, + "step": 32784 + }, + { + "epoch": 0.28458954349354604, + "grad_norm": 0.06396484375, + "learning_rate": 0.0015469723483220326, + "loss": 0.0811, + "step": 32785 + }, + { + "epoch": 0.28459822397375023, + "grad_norm": 0.1484375, + "learning_rate": 0.0015469465154869255, + "loss": 0.1406, + "step": 32786 + }, + { + "epoch": 0.28460690445395437, + "grad_norm": 0.0693359375, + "learning_rate": 0.001546920682163044, + "loss": 0.1279, + "step": 32787 + }, + { + "epoch": 0.28461558493415856, + "grad_norm": 0.265625, + "learning_rate": 0.001546894848350417, + "loss": 0.0957, + "step": 32788 + }, + { + "epoch": 0.2846242654143627, + "grad_norm": 0.08203125, + "learning_rate": 0.0015468690140490726, + "loss": 0.0908, + "step": 32789 + }, + { + "epoch": 0.2846329458945669, + "grad_norm": 0.55859375, + "learning_rate": 0.0015468431792590388, + "loss": 0.103, + "step": 32790 + }, + { + "epoch": 0.28464162637477103, + "grad_norm": 0.255859375, + "learning_rate": 0.0015468173439803438, + "loss": 0.1045, + "step": 32791 + }, + { + "epoch": 0.2846503068549752, + "grad_norm": 0.416015625, + "learning_rate": 0.0015467915082130162, + "loss": 0.1035, + "step": 32792 + }, + { + "epoch": 0.28465898733517936, + "grad_norm": 0.197265625, + "learning_rate": 0.0015467656719570845, + "loss": 0.1162, + "step": 32793 + }, + { + "epoch": 0.28466766781538355, + "grad_norm": 0.70703125, + "learning_rate": 0.001546739835212576, + "loss": 0.2402, + "step": 32794 + }, + { + "epoch": 0.2846763482955877, + "grad_norm": 0.17578125, + "learning_rate": 0.0015467139979795201, + "loss": 0.0938, + "step": 32795 + }, + { + "epoch": 0.2846850287757919, + "grad_norm": 0.158203125, + "learning_rate": 0.0015466881602579441, + "loss": 0.1187, + "step": 32796 + }, + { + "epoch": 0.284693709255996, + "grad_norm": 0.244140625, + "learning_rate": 0.0015466623220478767, + "loss": 0.0918, + "step": 32797 + }, + { + "epoch": 0.2847023897362002, + "grad_norm": 0.671875, + "learning_rate": 0.0015466364833493463, + "loss": 0.125, + "step": 32798 + }, + { + "epoch": 0.28471107021640435, + "grad_norm": 0.2578125, + "learning_rate": 0.001546610644162381, + "loss": 0.127, + "step": 32799 + }, + { + "epoch": 0.28471975069660854, + "grad_norm": 0.3515625, + "learning_rate": 0.0015465848044870086, + "loss": 0.1465, + "step": 32800 + }, + { + "epoch": 0.2847284311768127, + "grad_norm": 0.84765625, + "learning_rate": 0.001546558964323258, + "loss": 0.1543, + "step": 32801 + }, + { + "epoch": 0.2847371116570169, + "grad_norm": 0.19921875, + "learning_rate": 0.001546533123671157, + "loss": 0.0835, + "step": 32802 + }, + { + "epoch": 0.284745792137221, + "grad_norm": 0.1708984375, + "learning_rate": 0.0015465072825307342, + "loss": 0.1147, + "step": 32803 + }, + { + "epoch": 0.2847544726174252, + "grad_norm": 0.75, + "learning_rate": 0.0015464814409020177, + "loss": 0.1152, + "step": 32804 + }, + { + "epoch": 0.28476315309762934, + "grad_norm": 0.12890625, + "learning_rate": 0.001546455598785036, + "loss": 0.1143, + "step": 32805 + }, + { + "epoch": 0.28477183357783353, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015464297561798173, + "loss": 0.0947, + "step": 32806 + }, + { + "epoch": 0.28478051405803767, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015464039130863897, + "loss": 0.0918, + "step": 32807 + }, + { + "epoch": 0.28478919453824186, + "grad_norm": 0.73828125, + "learning_rate": 0.001546378069504781, + "loss": 0.1582, + "step": 32808 + }, + { + "epoch": 0.284797875018446, + "grad_norm": 0.35546875, + "learning_rate": 0.0015463522254350204, + "loss": 0.1191, + "step": 32809 + }, + { + "epoch": 0.2848065554986502, + "grad_norm": 0.404296875, + "learning_rate": 0.0015463263808771356, + "loss": 0.125, + "step": 32810 + }, + { + "epoch": 0.28481523597885433, + "grad_norm": 0.1015625, + "learning_rate": 0.0015463005358311549, + "loss": 0.0879, + "step": 32811 + }, + { + "epoch": 0.2848239164590585, + "grad_norm": 0.1328125, + "learning_rate": 0.0015462746902971068, + "loss": 0.082, + "step": 32812 + }, + { + "epoch": 0.28483259693926266, + "grad_norm": 0.2578125, + "learning_rate": 0.001546248844275019, + "loss": 0.0684, + "step": 32813 + }, + { + "epoch": 0.28484127741946685, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015462229977649205, + "loss": 0.1094, + "step": 32814 + }, + { + "epoch": 0.284849957899671, + "grad_norm": 0.66796875, + "learning_rate": 0.0015461971507668394, + "loss": 0.1514, + "step": 32815 + }, + { + "epoch": 0.2848586383798752, + "grad_norm": 0.263671875, + "learning_rate": 0.0015461713032808035, + "loss": 0.1123, + "step": 32816 + }, + { + "epoch": 0.2848673188600793, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015461454553068416, + "loss": 0.0957, + "step": 32817 + }, + { + "epoch": 0.2848759993402835, + "grad_norm": 0.18359375, + "learning_rate": 0.0015461196068449815, + "loss": 0.127, + "step": 32818 + }, + { + "epoch": 0.28488467982048765, + "grad_norm": 0.169921875, + "learning_rate": 0.0015460937578952519, + "loss": 0.1211, + "step": 32819 + }, + { + "epoch": 0.28489336030069184, + "grad_norm": 0.2109375, + "learning_rate": 0.0015460679084576805, + "loss": 0.1006, + "step": 32820 + }, + { + "epoch": 0.284902040780896, + "grad_norm": 0.09912109375, + "learning_rate": 0.0015460420585322963, + "loss": 0.083, + "step": 32821 + }, + { + "epoch": 0.2849107212611002, + "grad_norm": 0.251953125, + "learning_rate": 0.001546016208119127, + "loss": 0.1016, + "step": 32822 + }, + { + "epoch": 0.2849194017413043, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015459903572182013, + "loss": 0.082, + "step": 32823 + }, + { + "epoch": 0.2849280822215085, + "grad_norm": 0.353515625, + "learning_rate": 0.0015459645058295472, + "loss": 0.1196, + "step": 32824 + }, + { + "epoch": 0.28493676270171264, + "grad_norm": 0.58203125, + "learning_rate": 0.001545938653953193, + "loss": 0.1289, + "step": 32825 + }, + { + "epoch": 0.28494544318191684, + "grad_norm": 0.46484375, + "learning_rate": 0.0015459128015891668, + "loss": 0.0854, + "step": 32826 + }, + { + "epoch": 0.284954123662121, + "grad_norm": 0.427734375, + "learning_rate": 0.0015458869487374972, + "loss": 0.0801, + "step": 32827 + }, + { + "epoch": 0.28496280414232517, + "grad_norm": 0.166015625, + "learning_rate": 0.0015458610953982125, + "loss": 0.1504, + "step": 32828 + }, + { + "epoch": 0.2849714846225293, + "grad_norm": 0.5703125, + "learning_rate": 0.0015458352415713404, + "loss": 0.1094, + "step": 32829 + }, + { + "epoch": 0.2849801651027335, + "grad_norm": 0.1455078125, + "learning_rate": 0.00154580938725691, + "loss": 0.0981, + "step": 32830 + }, + { + "epoch": 0.28498884558293763, + "grad_norm": 0.609375, + "learning_rate": 0.0015457835324549486, + "loss": 0.1055, + "step": 32831 + }, + { + "epoch": 0.2849975260631418, + "grad_norm": 0.37890625, + "learning_rate": 0.0015457576771654858, + "loss": 0.1455, + "step": 32832 + }, + { + "epoch": 0.28500620654334596, + "grad_norm": 0.09765625, + "learning_rate": 0.0015457318213885486, + "loss": 0.0933, + "step": 32833 + }, + { + "epoch": 0.28501488702355016, + "grad_norm": 0.345703125, + "learning_rate": 0.0015457059651241658, + "loss": 0.1064, + "step": 32834 + }, + { + "epoch": 0.2850235675037543, + "grad_norm": 0.515625, + "learning_rate": 0.0015456801083723656, + "loss": 0.0762, + "step": 32835 + }, + { + "epoch": 0.2850322479839585, + "grad_norm": 0.51171875, + "learning_rate": 0.0015456542511331766, + "loss": 0.0908, + "step": 32836 + }, + { + "epoch": 0.2850409284641626, + "grad_norm": 0.73828125, + "learning_rate": 0.0015456283934066266, + "loss": 0.0952, + "step": 32837 + }, + { + "epoch": 0.2850496089443668, + "grad_norm": 0.396484375, + "learning_rate": 0.0015456025351927444, + "loss": 0.1289, + "step": 32838 + }, + { + "epoch": 0.28505828942457095, + "grad_norm": 0.404296875, + "learning_rate": 0.0015455766764915575, + "loss": 0.3066, + "step": 32839 + }, + { + "epoch": 0.28506696990477515, + "grad_norm": 0.09326171875, + "learning_rate": 0.0015455508173030947, + "loss": 0.0762, + "step": 32840 + }, + { + "epoch": 0.2850756503849793, + "grad_norm": 0.56640625, + "learning_rate": 0.0015455249576273845, + "loss": 0.1045, + "step": 32841 + }, + { + "epoch": 0.2850843308651835, + "grad_norm": 0.09716796875, + "learning_rate": 0.0015454990974644548, + "loss": 0.085, + "step": 32842 + }, + { + "epoch": 0.2850930113453876, + "grad_norm": 0.09765625, + "learning_rate": 0.0015454732368143339, + "loss": 0.1201, + "step": 32843 + }, + { + "epoch": 0.2851016918255918, + "grad_norm": 0.494140625, + "learning_rate": 0.0015454473756770502, + "loss": 0.123, + "step": 32844 + }, + { + "epoch": 0.28511037230579594, + "grad_norm": 0.095703125, + "learning_rate": 0.001545421514052632, + "loss": 0.104, + "step": 32845 + }, + { + "epoch": 0.28511905278600014, + "grad_norm": 1.875, + "learning_rate": 0.0015453956519411074, + "loss": 0.1973, + "step": 32846 + }, + { + "epoch": 0.2851277332662043, + "grad_norm": 0.33203125, + "learning_rate": 0.001545369789342505, + "loss": 0.0928, + "step": 32847 + }, + { + "epoch": 0.28513641374640847, + "grad_norm": 0.1953125, + "learning_rate": 0.0015453439262568527, + "loss": 0.0986, + "step": 32848 + }, + { + "epoch": 0.2851450942266126, + "grad_norm": 0.33984375, + "learning_rate": 0.001545318062684179, + "loss": 0.1406, + "step": 32849 + }, + { + "epoch": 0.2851537747068168, + "grad_norm": 0.66796875, + "learning_rate": 0.001545292198624512, + "loss": 0.1445, + "step": 32850 + }, + { + "epoch": 0.28516245518702094, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015452663340778808, + "loss": 0.0845, + "step": 32851 + }, + { + "epoch": 0.28517113566722513, + "grad_norm": 0.119140625, + "learning_rate": 0.0015452404690443123, + "loss": 0.1162, + "step": 32852 + }, + { + "epoch": 0.28517981614742927, + "grad_norm": 0.4765625, + "learning_rate": 0.001545214603523836, + "loss": 0.0928, + "step": 32853 + }, + { + "epoch": 0.28518849662763346, + "grad_norm": 0.4765625, + "learning_rate": 0.0015451887375164792, + "loss": 0.1055, + "step": 32854 + }, + { + "epoch": 0.2851971771078376, + "grad_norm": 0.314453125, + "learning_rate": 0.0015451628710222711, + "loss": 0.1084, + "step": 32855 + }, + { + "epoch": 0.2852058575880418, + "grad_norm": 0.423828125, + "learning_rate": 0.0015451370040412396, + "loss": 0.1387, + "step": 32856 + }, + { + "epoch": 0.2852145380682459, + "grad_norm": 0.259765625, + "learning_rate": 0.0015451111365734128, + "loss": 0.1133, + "step": 32857 + }, + { + "epoch": 0.2852232185484501, + "grad_norm": 0.380859375, + "learning_rate": 0.001545085268618819, + "loss": 0.0967, + "step": 32858 + }, + { + "epoch": 0.28523189902865426, + "grad_norm": 0.23828125, + "learning_rate": 0.0015450594001774867, + "loss": 0.0947, + "step": 32859 + }, + { + "epoch": 0.28524057950885845, + "grad_norm": 0.328125, + "learning_rate": 0.0015450335312494443, + "loss": 0.0864, + "step": 32860 + }, + { + "epoch": 0.2852492599890626, + "grad_norm": 0.298828125, + "learning_rate": 0.00154500766183472, + "loss": 0.0698, + "step": 32861 + }, + { + "epoch": 0.2852579404692668, + "grad_norm": 0.306640625, + "learning_rate": 0.0015449817919333416, + "loss": 0.1387, + "step": 32862 + }, + { + "epoch": 0.2852666209494709, + "grad_norm": 0.44921875, + "learning_rate": 0.001544955921545338, + "loss": 0.0991, + "step": 32863 + }, + { + "epoch": 0.2852753014296751, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015449300506707375, + "loss": 0.0962, + "step": 32864 + }, + { + "epoch": 0.28528398190987925, + "grad_norm": 0.12890625, + "learning_rate": 0.001544904179309568, + "loss": 0.0898, + "step": 32865 + }, + { + "epoch": 0.28529266239008344, + "grad_norm": 0.63671875, + "learning_rate": 0.001544878307461858, + "loss": 0.1201, + "step": 32866 + }, + { + "epoch": 0.2853013428702876, + "grad_norm": 0.220703125, + "learning_rate": 0.001544852435127636, + "loss": 0.1221, + "step": 32867 + }, + { + "epoch": 0.28531002335049177, + "grad_norm": 0.326171875, + "learning_rate": 0.0015448265623069295, + "loss": 0.0923, + "step": 32868 + }, + { + "epoch": 0.2853187038306959, + "grad_norm": 0.546875, + "learning_rate": 0.0015448006889997676, + "loss": 0.1035, + "step": 32869 + }, + { + "epoch": 0.2853273843109001, + "grad_norm": 0.73828125, + "learning_rate": 0.0015447748152061784, + "loss": 0.1143, + "step": 32870 + }, + { + "epoch": 0.28533606479110424, + "grad_norm": 1.2109375, + "learning_rate": 0.0015447489409261903, + "loss": 0.0801, + "step": 32871 + }, + { + "epoch": 0.28534474527130843, + "grad_norm": 0.2421875, + "learning_rate": 0.0015447230661598315, + "loss": 0.1055, + "step": 32872 + }, + { + "epoch": 0.28535342575151257, + "grad_norm": 0.43359375, + "learning_rate": 0.0015446971909071298, + "loss": 0.1104, + "step": 32873 + }, + { + "epoch": 0.28536210623171676, + "grad_norm": 0.2412109375, + "learning_rate": 0.0015446713151681141, + "loss": 0.085, + "step": 32874 + }, + { + "epoch": 0.2853707867119209, + "grad_norm": 0.07177734375, + "learning_rate": 0.0015446454389428127, + "loss": 0.083, + "step": 32875 + }, + { + "epoch": 0.2853794671921251, + "grad_norm": 0.10888671875, + "learning_rate": 0.0015446195622312539, + "loss": 0.0889, + "step": 32876 + }, + { + "epoch": 0.28538814767232923, + "grad_norm": 0.3125, + "learning_rate": 0.0015445936850334654, + "loss": 0.1104, + "step": 32877 + }, + { + "epoch": 0.2853968281525334, + "grad_norm": 0.310546875, + "learning_rate": 0.0015445678073494758, + "loss": 0.085, + "step": 32878 + }, + { + "epoch": 0.28540550863273756, + "grad_norm": 0.55078125, + "learning_rate": 0.001544541929179314, + "loss": 0.1118, + "step": 32879 + }, + { + "epoch": 0.28541418911294175, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015445160505230076, + "loss": 0.0825, + "step": 32880 + }, + { + "epoch": 0.2854228695931459, + "grad_norm": 0.8125, + "learning_rate": 0.0015444901713805852, + "loss": 0.1084, + "step": 32881 + }, + { + "epoch": 0.2854315500733501, + "grad_norm": 0.5, + "learning_rate": 0.0015444642917520751, + "loss": 0.0986, + "step": 32882 + }, + { + "epoch": 0.2854402305535542, + "grad_norm": 0.546875, + "learning_rate": 0.0015444384116375053, + "loss": 0.1143, + "step": 32883 + }, + { + "epoch": 0.2854489110337584, + "grad_norm": 0.5625, + "learning_rate": 0.0015444125310369046, + "loss": 0.0845, + "step": 32884 + }, + { + "epoch": 0.28545759151396255, + "grad_norm": 0.15234375, + "learning_rate": 0.0015443866499503009, + "loss": 0.0908, + "step": 32885 + }, + { + "epoch": 0.28546627199416674, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015443607683777226, + "loss": 0.1016, + "step": 32886 + }, + { + "epoch": 0.2854749524743709, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015443348863191982, + "loss": 0.166, + "step": 32887 + }, + { + "epoch": 0.2854836329545751, + "grad_norm": 0.2216796875, + "learning_rate": 0.0015443090037747555, + "loss": 0.0806, + "step": 32888 + }, + { + "epoch": 0.2854923134347792, + "grad_norm": 0.30078125, + "learning_rate": 0.0015442831207444234, + "loss": 0.0898, + "step": 32889 + }, + { + "epoch": 0.2855009939149834, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015442572372282297, + "loss": 0.127, + "step": 32890 + }, + { + "epoch": 0.28550967439518754, + "grad_norm": 0.142578125, + "learning_rate": 0.0015442313532262034, + "loss": 0.0928, + "step": 32891 + }, + { + "epoch": 0.28551835487539173, + "grad_norm": 0.1318359375, + "learning_rate": 0.001544205468738372, + "loss": 0.0869, + "step": 32892 + }, + { + "epoch": 0.28552703535559587, + "grad_norm": 0.37890625, + "learning_rate": 0.0015441795837647643, + "loss": 0.207, + "step": 32893 + }, + { + "epoch": 0.28553571583580006, + "grad_norm": 0.328125, + "learning_rate": 0.0015441536983054084, + "loss": 0.123, + "step": 32894 + }, + { + "epoch": 0.2855443963160042, + "grad_norm": 0.1396484375, + "learning_rate": 0.0015441278123603328, + "loss": 0.0771, + "step": 32895 + }, + { + "epoch": 0.2855530767962084, + "grad_norm": 0.34765625, + "learning_rate": 0.0015441019259295656, + "loss": 0.1113, + "step": 32896 + }, + { + "epoch": 0.28556175727641253, + "grad_norm": 0.1875, + "learning_rate": 0.0015440760390131353, + "loss": 0.1006, + "step": 32897 + }, + { + "epoch": 0.28557043775661667, + "grad_norm": 0.94140625, + "learning_rate": 0.0015440501516110704, + "loss": 0.3672, + "step": 32898 + }, + { + "epoch": 0.28557911823682086, + "grad_norm": 0.15625, + "learning_rate": 0.0015440242637233983, + "loss": 0.084, + "step": 32899 + }, + { + "epoch": 0.285587798717025, + "grad_norm": 0.30859375, + "learning_rate": 0.0015439983753501485, + "loss": 0.0947, + "step": 32900 + }, + { + "epoch": 0.2855964791972292, + "grad_norm": 0.8046875, + "learning_rate": 0.0015439724864913485, + "loss": 0.1377, + "step": 32901 + }, + { + "epoch": 0.28560515967743333, + "grad_norm": 0.3203125, + "learning_rate": 0.001543946597147027, + "loss": 0.1152, + "step": 32902 + }, + { + "epoch": 0.2856138401576375, + "grad_norm": 0.462890625, + "learning_rate": 0.0015439207073172118, + "loss": 0.0815, + "step": 32903 + }, + { + "epoch": 0.28562252063784166, + "grad_norm": 0.48828125, + "learning_rate": 0.001543894817001932, + "loss": 0.1094, + "step": 32904 + }, + { + "epoch": 0.28563120111804585, + "grad_norm": 0.3359375, + "learning_rate": 0.001543868926201215, + "loss": 0.0747, + "step": 32905 + }, + { + "epoch": 0.28563988159825, + "grad_norm": 0.1669921875, + "learning_rate": 0.00154384303491509, + "loss": 0.1045, + "step": 32906 + }, + { + "epoch": 0.2856485620784542, + "grad_norm": 0.345703125, + "learning_rate": 0.0015438171431435847, + "loss": 0.0918, + "step": 32907 + }, + { + "epoch": 0.2856572425586583, + "grad_norm": 0.45703125, + "learning_rate": 0.001543791250886728, + "loss": 0.0757, + "step": 32908 + }, + { + "epoch": 0.2856659230388625, + "grad_norm": 0.1689453125, + "learning_rate": 0.0015437653581445478, + "loss": 0.1318, + "step": 32909 + }, + { + "epoch": 0.28567460351906665, + "grad_norm": 0.357421875, + "learning_rate": 0.0015437394649170722, + "loss": 0.082, + "step": 32910 + }, + { + "epoch": 0.28568328399927084, + "grad_norm": 0.1640625, + "learning_rate": 0.00154371357120433, + "loss": 0.1064, + "step": 32911 + }, + { + "epoch": 0.285691964479475, + "grad_norm": 0.0908203125, + "learning_rate": 0.0015436876770063492, + "loss": 0.1143, + "step": 32912 + }, + { + "epoch": 0.2857006449596792, + "grad_norm": 0.63671875, + "learning_rate": 0.0015436617823231583, + "loss": 0.1035, + "step": 32913 + }, + { + "epoch": 0.2857093254398833, + "grad_norm": 0.41015625, + "learning_rate": 0.0015436358871547856, + "loss": 0.1226, + "step": 32914 + }, + { + "epoch": 0.2857180059200875, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015436099915012592, + "loss": 0.1416, + "step": 32915 + }, + { + "epoch": 0.28572668640029164, + "grad_norm": 0.2451171875, + "learning_rate": 0.0015435840953626075, + "loss": 0.1064, + "step": 32916 + }, + { + "epoch": 0.28573536688049583, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015435581987388587, + "loss": 0.1562, + "step": 32917 + }, + { + "epoch": 0.28574404736069997, + "grad_norm": 0.6796875, + "learning_rate": 0.0015435323016300417, + "loss": 0.1299, + "step": 32918 + }, + { + "epoch": 0.28575272784090416, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015435064040361845, + "loss": 0.0698, + "step": 32919 + }, + { + "epoch": 0.2857614083211083, + "grad_norm": 0.54296875, + "learning_rate": 0.0015434805059573154, + "loss": 0.1016, + "step": 32920 + }, + { + "epoch": 0.2857700888013125, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015434546073934622, + "loss": 0.1123, + "step": 32921 + }, + { + "epoch": 0.28577876928151663, + "grad_norm": 0.2060546875, + "learning_rate": 0.0015434287083446543, + "loss": 0.0962, + "step": 32922 + }, + { + "epoch": 0.2857874497617208, + "grad_norm": 0.40625, + "learning_rate": 0.0015434028088109192, + "loss": 0.1211, + "step": 32923 + }, + { + "epoch": 0.28579613024192496, + "grad_norm": 0.08544921875, + "learning_rate": 0.0015433769087922853, + "loss": 0.0781, + "step": 32924 + }, + { + "epoch": 0.28580481072212915, + "grad_norm": 0.224609375, + "learning_rate": 0.0015433510082887814, + "loss": 0.1309, + "step": 32925 + }, + { + "epoch": 0.2858134912023333, + "grad_norm": 2.421875, + "learning_rate": 0.0015433251073004352, + "loss": 0.2139, + "step": 32926 + }, + { + "epoch": 0.2858221716825375, + "grad_norm": 0.328125, + "learning_rate": 0.0015432992058272753, + "loss": 0.1064, + "step": 32927 + }, + { + "epoch": 0.2858308521627416, + "grad_norm": 0.0771484375, + "learning_rate": 0.00154327330386933, + "loss": 0.0801, + "step": 32928 + }, + { + "epoch": 0.2858395326429458, + "grad_norm": 0.3046875, + "learning_rate": 0.001543247401426628, + "loss": 0.1396, + "step": 32929 + }, + { + "epoch": 0.28584821312314995, + "grad_norm": 0.357421875, + "learning_rate": 0.001543221498499197, + "loss": 0.1182, + "step": 32930 + }, + { + "epoch": 0.28585689360335415, + "grad_norm": 0.376953125, + "learning_rate": 0.0015431955950870658, + "loss": 0.124, + "step": 32931 + }, + { + "epoch": 0.2858655740835583, + "grad_norm": 0.224609375, + "learning_rate": 0.0015431696911902626, + "loss": 0.0884, + "step": 32932 + }, + { + "epoch": 0.2858742545637625, + "grad_norm": 0.125, + "learning_rate": 0.0015431437868088157, + "loss": 0.1152, + "step": 32933 + }, + { + "epoch": 0.2858829350439666, + "grad_norm": 0.1943359375, + "learning_rate": 0.001543117881942753, + "loss": 0.0874, + "step": 32934 + }, + { + "epoch": 0.2858916155241708, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015430919765921038, + "loss": 0.1094, + "step": 32935 + }, + { + "epoch": 0.28590029600437494, + "grad_norm": 0.208984375, + "learning_rate": 0.0015430660707568956, + "loss": 0.1108, + "step": 32936 + }, + { + "epoch": 0.28590897648457914, + "grad_norm": 0.1455078125, + "learning_rate": 0.0015430401644371567, + "loss": 0.1436, + "step": 32937 + }, + { + "epoch": 0.2859176569647833, + "grad_norm": 0.74609375, + "learning_rate": 0.0015430142576329162, + "loss": 0.0918, + "step": 32938 + }, + { + "epoch": 0.28592633744498747, + "grad_norm": 0.09033203125, + "learning_rate": 0.001542988350344202, + "loss": 0.0874, + "step": 32939 + }, + { + "epoch": 0.2859350179251916, + "grad_norm": 0.37890625, + "learning_rate": 0.001542962442571042, + "loss": 0.1279, + "step": 32940 + }, + { + "epoch": 0.2859436984053958, + "grad_norm": 0.1201171875, + "learning_rate": 0.0015429365343134654, + "loss": 0.1025, + "step": 32941 + }, + { + "epoch": 0.28595237888559993, + "grad_norm": 0.08740234375, + "learning_rate": 0.0015429106255714998, + "loss": 0.0918, + "step": 32942 + }, + { + "epoch": 0.2859610593658041, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015428847163451737, + "loss": 0.084, + "step": 32943 + }, + { + "epoch": 0.28596973984600826, + "grad_norm": 0.1552734375, + "learning_rate": 0.0015428588066345155, + "loss": 0.1465, + "step": 32944 + }, + { + "epoch": 0.28597842032621246, + "grad_norm": 1.109375, + "learning_rate": 0.0015428328964395537, + "loss": 0.1084, + "step": 32945 + }, + { + "epoch": 0.2859871008064166, + "grad_norm": 0.421875, + "learning_rate": 0.0015428069857603166, + "loss": 0.1045, + "step": 32946 + }, + { + "epoch": 0.2859957812866208, + "grad_norm": 0.166015625, + "learning_rate": 0.0015427810745968322, + "loss": 0.1406, + "step": 32947 + }, + { + "epoch": 0.2860044617668249, + "grad_norm": 0.0712890625, + "learning_rate": 0.0015427551629491293, + "loss": 0.1143, + "step": 32948 + }, + { + "epoch": 0.2860131422470291, + "grad_norm": 0.23828125, + "learning_rate": 0.001542729250817236, + "loss": 0.1084, + "step": 32949 + }, + { + "epoch": 0.28602182272723325, + "grad_norm": 0.1396484375, + "learning_rate": 0.0015427033382011803, + "loss": 0.1484, + "step": 32950 + }, + { + "epoch": 0.28603050320743745, + "grad_norm": 0.8046875, + "learning_rate": 0.0015426774251009912, + "loss": 0.1084, + "step": 32951 + }, + { + "epoch": 0.2860391836876416, + "grad_norm": 0.2314453125, + "learning_rate": 0.0015426515115166965, + "loss": 0.0884, + "step": 32952 + }, + { + "epoch": 0.2860478641678458, + "grad_norm": 0.6484375, + "learning_rate": 0.001542625597448325, + "loss": 0.1147, + "step": 32953 + }, + { + "epoch": 0.2860565446480499, + "grad_norm": 0.51171875, + "learning_rate": 0.0015425996828959047, + "loss": 0.0894, + "step": 32954 + }, + { + "epoch": 0.2860652251282541, + "grad_norm": 0.703125, + "learning_rate": 0.0015425737678594642, + "loss": 0.1328, + "step": 32955 + }, + { + "epoch": 0.28607390560845825, + "grad_norm": 0.337890625, + "learning_rate": 0.0015425478523390312, + "loss": 0.1562, + "step": 32956 + }, + { + "epoch": 0.28608258608866244, + "grad_norm": 0.34375, + "learning_rate": 0.001542521936334635, + "loss": 0.0938, + "step": 32957 + }, + { + "epoch": 0.2860912665688666, + "grad_norm": 0.115234375, + "learning_rate": 0.0015424960198463034, + "loss": 0.1396, + "step": 32958 + }, + { + "epoch": 0.28609994704907077, + "grad_norm": 0.189453125, + "learning_rate": 0.0015424701028740647, + "loss": 0.0967, + "step": 32959 + }, + { + "epoch": 0.2861086275292749, + "grad_norm": 0.1484375, + "learning_rate": 0.001542444185417947, + "loss": 0.1182, + "step": 32960 + }, + { + "epoch": 0.2861173080094791, + "grad_norm": 0.061279296875, + "learning_rate": 0.0015424182674779797, + "loss": 0.0815, + "step": 32961 + }, + { + "epoch": 0.28612598848968324, + "grad_norm": 0.47265625, + "learning_rate": 0.0015423923490541897, + "loss": 0.1006, + "step": 32962 + }, + { + "epoch": 0.28613466896988743, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015423664301466066, + "loss": 0.104, + "step": 32963 + }, + { + "epoch": 0.28614334945009157, + "grad_norm": 0.16796875, + "learning_rate": 0.001542340510755258, + "loss": 0.0884, + "step": 32964 + }, + { + "epoch": 0.28615202993029576, + "grad_norm": 0.95703125, + "learning_rate": 0.0015423145908801724, + "loss": 0.1211, + "step": 32965 + }, + { + "epoch": 0.2861607104104999, + "grad_norm": 0.12060546875, + "learning_rate": 0.0015422886705213786, + "loss": 0.0889, + "step": 32966 + }, + { + "epoch": 0.2861693908907041, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015422627496789044, + "loss": 0.1104, + "step": 32967 + }, + { + "epoch": 0.2861780713709082, + "grad_norm": 0.19921875, + "learning_rate": 0.0015422368283527779, + "loss": 0.0869, + "step": 32968 + }, + { + "epoch": 0.2861867518511124, + "grad_norm": 0.25390625, + "learning_rate": 0.0015422109065430282, + "loss": 0.1299, + "step": 32969 + }, + { + "epoch": 0.28619543233131656, + "grad_norm": 0.16015625, + "learning_rate": 0.001542184984249683, + "loss": 0.0903, + "step": 32970 + }, + { + "epoch": 0.28620411281152075, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015421590614727715, + "loss": 0.1074, + "step": 32971 + }, + { + "epoch": 0.2862127932917249, + "grad_norm": 0.439453125, + "learning_rate": 0.001542133138212321, + "loss": 0.1445, + "step": 32972 + }, + { + "epoch": 0.2862214737719291, + "grad_norm": 0.265625, + "learning_rate": 0.0015421072144683606, + "loss": 0.1328, + "step": 32973 + }, + { + "epoch": 0.2862301542521332, + "grad_norm": 0.4609375, + "learning_rate": 0.001542081290240918, + "loss": 0.1235, + "step": 32974 + }, + { + "epoch": 0.2862388347323374, + "grad_norm": 0.474609375, + "learning_rate": 0.0015420553655300226, + "loss": 0.124, + "step": 32975 + }, + { + "epoch": 0.28624751521254155, + "grad_norm": 0.09619140625, + "learning_rate": 0.0015420294403357014, + "loss": 0.0962, + "step": 32976 + }, + { + "epoch": 0.28625619569274574, + "grad_norm": 0.287109375, + "learning_rate": 0.0015420035146579839, + "loss": 0.0869, + "step": 32977 + }, + { + "epoch": 0.2862648761729499, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015419775884968978, + "loss": 0.1094, + "step": 32978 + }, + { + "epoch": 0.28627355665315407, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015419516618524717, + "loss": 0.083, + "step": 32979 + }, + { + "epoch": 0.2862822371333582, + "grad_norm": 0.08056640625, + "learning_rate": 0.0015419257347247342, + "loss": 0.0977, + "step": 32980 + }, + { + "epoch": 0.2862909176135624, + "grad_norm": 0.484375, + "learning_rate": 0.0015418998071137127, + "loss": 0.0938, + "step": 32981 + }, + { + "epoch": 0.28629959809376654, + "grad_norm": 0.25, + "learning_rate": 0.001541873879019437, + "loss": 0.1001, + "step": 32982 + }, + { + "epoch": 0.28630827857397073, + "grad_norm": 0.0966796875, + "learning_rate": 0.0015418479504419338, + "loss": 0.0869, + "step": 32983 + }, + { + "epoch": 0.28631695905417487, + "grad_norm": 0.19140625, + "learning_rate": 0.0015418220213812331, + "loss": 0.0771, + "step": 32984 + }, + { + "epoch": 0.28632563953437906, + "grad_norm": 0.34375, + "learning_rate": 0.001541796091837362, + "loss": 0.0933, + "step": 32985 + }, + { + "epoch": 0.2863343200145832, + "grad_norm": 0.310546875, + "learning_rate": 0.0015417701618103494, + "loss": 0.126, + "step": 32986 + }, + { + "epoch": 0.2863430004947874, + "grad_norm": 0.150390625, + "learning_rate": 0.0015417442313002236, + "loss": 0.1172, + "step": 32987 + }, + { + "epoch": 0.28635168097499153, + "grad_norm": 0.10546875, + "learning_rate": 0.0015417183003070133, + "loss": 0.0957, + "step": 32988 + }, + { + "epoch": 0.2863603614551957, + "grad_norm": 0.5546875, + "learning_rate": 0.0015416923688307462, + "loss": 0.0942, + "step": 32989 + }, + { + "epoch": 0.28636904193539986, + "grad_norm": 0.08349609375, + "learning_rate": 0.0015416664368714508, + "loss": 0.1309, + "step": 32990 + }, + { + "epoch": 0.28637772241560405, + "grad_norm": 0.37109375, + "learning_rate": 0.001541640504429156, + "loss": 0.1157, + "step": 32991 + }, + { + "epoch": 0.2863864028958082, + "grad_norm": 0.671875, + "learning_rate": 0.0015416145715038895, + "loss": 0.1079, + "step": 32992 + }, + { + "epoch": 0.2863950833760124, + "grad_norm": 0.10302734375, + "learning_rate": 0.00154158863809568, + "loss": 0.1099, + "step": 32993 + }, + { + "epoch": 0.2864037638562165, + "grad_norm": 0.54296875, + "learning_rate": 0.0015415627042045558, + "loss": 0.1084, + "step": 32994 + }, + { + "epoch": 0.2864124443364207, + "grad_norm": 0.306640625, + "learning_rate": 0.001541536769830545, + "loss": 0.1094, + "step": 32995 + }, + { + "epoch": 0.28642112481662485, + "grad_norm": 0.1640625, + "learning_rate": 0.0015415108349736768, + "loss": 0.083, + "step": 32996 + }, + { + "epoch": 0.28642980529682904, + "grad_norm": 0.67578125, + "learning_rate": 0.0015414848996339787, + "loss": 0.0991, + "step": 32997 + }, + { + "epoch": 0.2864384857770332, + "grad_norm": 0.134765625, + "learning_rate": 0.0015414589638114795, + "loss": 0.0947, + "step": 32998 + }, + { + "epoch": 0.2864471662572374, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015414330275062072, + "loss": 0.0967, + "step": 32999 + }, + { + "epoch": 0.2864558467374415, + "grad_norm": 0.150390625, + "learning_rate": 0.0015414070907181904, + "loss": 0.1279, + "step": 33000 + }, + { + "epoch": 0.2864645272176457, + "grad_norm": 0.423828125, + "learning_rate": 0.0015413811534474575, + "loss": 0.1094, + "step": 33001 + }, + { + "epoch": 0.28647320769784984, + "grad_norm": 0.77734375, + "learning_rate": 0.0015413552156940365, + "loss": 0.0747, + "step": 33002 + }, + { + "epoch": 0.28648188817805403, + "grad_norm": 0.396484375, + "learning_rate": 0.0015413292774579563, + "loss": 0.125, + "step": 33003 + }, + { + "epoch": 0.28649056865825817, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015413033387392451, + "loss": 0.083, + "step": 33004 + }, + { + "epoch": 0.28649924913846236, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015412773995379311, + "loss": 0.083, + "step": 33005 + }, + { + "epoch": 0.2865079296186665, + "grad_norm": 0.205078125, + "learning_rate": 0.001541251459854043, + "loss": 0.1099, + "step": 33006 + }, + { + "epoch": 0.2865166100988707, + "grad_norm": 0.8125, + "learning_rate": 0.0015412255196876085, + "loss": 0.1553, + "step": 33007 + }, + { + "epoch": 0.28652529057907483, + "grad_norm": 0.265625, + "learning_rate": 0.0015411995790386571, + "loss": 0.0781, + "step": 33008 + }, + { + "epoch": 0.286533971059279, + "grad_norm": 0.111328125, + "learning_rate": 0.001541173637907216, + "loss": 0.1108, + "step": 33009 + }, + { + "epoch": 0.28654265153948316, + "grad_norm": 0.216796875, + "learning_rate": 0.001541147696293314, + "loss": 0.104, + "step": 33010 + }, + { + "epoch": 0.28655133201968735, + "grad_norm": 0.57421875, + "learning_rate": 0.0015411217541969795, + "loss": 0.1465, + "step": 33011 + }, + { + "epoch": 0.2865600124998915, + "grad_norm": 0.1650390625, + "learning_rate": 0.001541095811618241, + "loss": 0.0933, + "step": 33012 + }, + { + "epoch": 0.2865686929800957, + "grad_norm": 0.333984375, + "learning_rate": 0.0015410698685571266, + "loss": 0.0967, + "step": 33013 + }, + { + "epoch": 0.2865773734602998, + "grad_norm": 0.115234375, + "learning_rate": 0.001541043925013665, + "loss": 0.124, + "step": 33014 + }, + { + "epoch": 0.286586053940504, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015410179809878845, + "loss": 0.1133, + "step": 33015 + }, + { + "epoch": 0.28659473442070815, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015409920364798128, + "loss": 0.1094, + "step": 33016 + }, + { + "epoch": 0.28660341490091235, + "grad_norm": 0.9140625, + "learning_rate": 0.0015409660914894794, + "loss": 0.0869, + "step": 33017 + }, + { + "epoch": 0.2866120953811165, + "grad_norm": 0.14453125, + "learning_rate": 0.0015409401460169119, + "loss": 0.1484, + "step": 33018 + }, + { + "epoch": 0.2866207758613207, + "grad_norm": 0.275390625, + "learning_rate": 0.001540914200062139, + "loss": 0.0869, + "step": 33019 + }, + { + "epoch": 0.2866294563415248, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015408882536251888, + "loss": 0.0879, + "step": 33020 + }, + { + "epoch": 0.28663813682172895, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015408623067060898, + "loss": 0.0967, + "step": 33021 + }, + { + "epoch": 0.28664681730193314, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015408363593048704, + "loss": 0.0601, + "step": 33022 + }, + { + "epoch": 0.2866554977821373, + "grad_norm": 0.0869140625, + "learning_rate": 0.001540810411421559, + "loss": 0.0854, + "step": 33023 + }, + { + "epoch": 0.2866641782623415, + "grad_norm": 0.55859375, + "learning_rate": 0.0015407844630561842, + "loss": 0.084, + "step": 33024 + }, + { + "epoch": 0.2866728587425456, + "grad_norm": 0.55859375, + "learning_rate": 0.0015407585142087738, + "loss": 0.0811, + "step": 33025 + }, + { + "epoch": 0.2866815392227498, + "grad_norm": 0.130859375, + "learning_rate": 0.0015407325648793568, + "loss": 0.0918, + "step": 33026 + }, + { + "epoch": 0.28669021970295394, + "grad_norm": 0.30859375, + "learning_rate": 0.0015407066150679612, + "loss": 0.1055, + "step": 33027 + }, + { + "epoch": 0.28669890018315813, + "grad_norm": 0.12451171875, + "learning_rate": 0.0015406806647746153, + "loss": 0.0659, + "step": 33028 + }, + { + "epoch": 0.28670758066336227, + "grad_norm": 0.578125, + "learning_rate": 0.0015406547139993482, + "loss": 0.1147, + "step": 33029 + }, + { + "epoch": 0.28671626114356646, + "grad_norm": 2.140625, + "learning_rate": 0.0015406287627421872, + "loss": 0.0918, + "step": 33030 + }, + { + "epoch": 0.2867249416237706, + "grad_norm": 0.28515625, + "learning_rate": 0.0015406028110031612, + "loss": 0.085, + "step": 33031 + }, + { + "epoch": 0.2867336221039748, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015405768587822988, + "loss": 0.1113, + "step": 33032 + }, + { + "epoch": 0.28674230258417893, + "grad_norm": 0.25390625, + "learning_rate": 0.001540550906079628, + "loss": 0.0933, + "step": 33033 + }, + { + "epoch": 0.2867509830643831, + "grad_norm": 1.2890625, + "learning_rate": 0.0015405249528951772, + "loss": 0.1128, + "step": 33034 + }, + { + "epoch": 0.28675966354458726, + "grad_norm": 0.498046875, + "learning_rate": 0.0015404989992289753, + "loss": 0.126, + "step": 33035 + }, + { + "epoch": 0.28676834402479146, + "grad_norm": 0.43359375, + "learning_rate": 0.00154047304508105, + "loss": 0.1152, + "step": 33036 + }, + { + "epoch": 0.2867770245049956, + "grad_norm": 0.474609375, + "learning_rate": 0.0015404470904514303, + "loss": 0.0933, + "step": 33037 + }, + { + "epoch": 0.2867857049851998, + "grad_norm": 0.169921875, + "learning_rate": 0.001540421135340144, + "loss": 0.0991, + "step": 33038 + }, + { + "epoch": 0.2867943854654039, + "grad_norm": 0.369140625, + "learning_rate": 0.0015403951797472199, + "loss": 0.1055, + "step": 33039 + }, + { + "epoch": 0.2868030659456081, + "grad_norm": 0.392578125, + "learning_rate": 0.0015403692236726865, + "loss": 0.0898, + "step": 33040 + }, + { + "epoch": 0.28681174642581225, + "grad_norm": 0.330078125, + "learning_rate": 0.0015403432671165713, + "loss": 0.1299, + "step": 33041 + }, + { + "epoch": 0.28682042690601645, + "grad_norm": 0.11474609375, + "learning_rate": 0.0015403173100789036, + "loss": 0.125, + "step": 33042 + }, + { + "epoch": 0.2868291073862206, + "grad_norm": 0.53125, + "learning_rate": 0.0015402913525597117, + "loss": 0.1387, + "step": 33043 + }, + { + "epoch": 0.2868377878664248, + "grad_norm": 0.10205078125, + "learning_rate": 0.001540265394559024, + "loss": 0.0845, + "step": 33044 + }, + { + "epoch": 0.2868464683466289, + "grad_norm": 0.15625, + "learning_rate": 0.0015402394360768683, + "loss": 0.1084, + "step": 33045 + }, + { + "epoch": 0.2868551488268331, + "grad_norm": 0.8828125, + "learning_rate": 0.0015402134771132733, + "loss": 0.083, + "step": 33046 + }, + { + "epoch": 0.28686382930703724, + "grad_norm": 0.10546875, + "learning_rate": 0.0015401875176682676, + "loss": 0.0879, + "step": 33047 + }, + { + "epoch": 0.28687250978724144, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015401615577418796, + "loss": 0.1025, + "step": 33048 + }, + { + "epoch": 0.2868811902674456, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015401355973341374, + "loss": 0.1055, + "step": 33049 + }, + { + "epoch": 0.28688987074764977, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015401096364450696, + "loss": 0.1084, + "step": 33050 + }, + { + "epoch": 0.2868985512278539, + "grad_norm": 0.1796875, + "learning_rate": 0.001540083675074704, + "loss": 0.064, + "step": 33051 + }, + { + "epoch": 0.2869072317080581, + "grad_norm": 0.154296875, + "learning_rate": 0.0015400577132230699, + "loss": 0.1025, + "step": 33052 + }, + { + "epoch": 0.28691591218826223, + "grad_norm": 0.86328125, + "learning_rate": 0.0015400317508901956, + "loss": 0.1079, + "step": 33053 + }, + { + "epoch": 0.2869245926684664, + "grad_norm": 0.310546875, + "learning_rate": 0.0015400057880761088, + "loss": 0.104, + "step": 33054 + }, + { + "epoch": 0.28693327314867056, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015399798247808384, + "loss": 0.085, + "step": 33055 + }, + { + "epoch": 0.28694195362887476, + "grad_norm": 0.66796875, + "learning_rate": 0.0015399538610044127, + "loss": 0.1318, + "step": 33056 + }, + { + "epoch": 0.2869506341090789, + "grad_norm": 0.1484375, + "learning_rate": 0.0015399278967468599, + "loss": 0.1016, + "step": 33057 + }, + { + "epoch": 0.2869593145892831, + "grad_norm": 0.1552734375, + "learning_rate": 0.0015399019320082089, + "loss": 0.0903, + "step": 33058 + }, + { + "epoch": 0.2869679950694872, + "grad_norm": 0.36328125, + "learning_rate": 0.0015398759667884874, + "loss": 0.1523, + "step": 33059 + }, + { + "epoch": 0.2869766755496914, + "grad_norm": 0.333984375, + "learning_rate": 0.001539850001087724, + "loss": 0.103, + "step": 33060 + }, + { + "epoch": 0.28698535602989556, + "grad_norm": 0.32421875, + "learning_rate": 0.0015398240349059477, + "loss": 0.1055, + "step": 33061 + }, + { + "epoch": 0.28699403651009975, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015397980682431863, + "loss": 0.1011, + "step": 33062 + }, + { + "epoch": 0.2870027169903039, + "grad_norm": 0.083984375, + "learning_rate": 0.0015397721010994682, + "loss": 0.0923, + "step": 33063 + }, + { + "epoch": 0.2870113974705081, + "grad_norm": 0.2373046875, + "learning_rate": 0.0015397461334748222, + "loss": 0.1035, + "step": 33064 + }, + { + "epoch": 0.2870200779507122, + "grad_norm": 0.69921875, + "learning_rate": 0.0015397201653692763, + "loss": 0.1631, + "step": 33065 + }, + { + "epoch": 0.2870287584309164, + "grad_norm": 0.6171875, + "learning_rate": 0.001539694196782859, + "loss": 0.1152, + "step": 33066 + }, + { + "epoch": 0.28703743891112055, + "grad_norm": 0.11767578125, + "learning_rate": 0.0015396682277155989, + "loss": 0.0933, + "step": 33067 + }, + { + "epoch": 0.28704611939132474, + "grad_norm": 0.123046875, + "learning_rate": 0.0015396422581675237, + "loss": 0.1025, + "step": 33068 + }, + { + "epoch": 0.2870547998715289, + "grad_norm": 0.28515625, + "learning_rate": 0.001539616288138663, + "loss": 0.0928, + "step": 33069 + }, + { + "epoch": 0.28706348035173307, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015395903176290441, + "loss": 0.1045, + "step": 33070 + }, + { + "epoch": 0.2870721608319372, + "grad_norm": 0.5234375, + "learning_rate": 0.0015395643466386963, + "loss": 0.1035, + "step": 33071 + }, + { + "epoch": 0.2870808413121414, + "grad_norm": 0.466796875, + "learning_rate": 0.001539538375167647, + "loss": 0.1055, + "step": 33072 + }, + { + "epoch": 0.28708952179234554, + "grad_norm": 0.427734375, + "learning_rate": 0.0015395124032159255, + "loss": 0.0781, + "step": 33073 + }, + { + "epoch": 0.28709820227254973, + "grad_norm": 0.16015625, + "learning_rate": 0.0015394864307835597, + "loss": 0.1064, + "step": 33074 + }, + { + "epoch": 0.28710688275275387, + "grad_norm": 0.294921875, + "learning_rate": 0.0015394604578705781, + "loss": 0.0884, + "step": 33075 + }, + { + "epoch": 0.28711556323295806, + "grad_norm": 0.388671875, + "learning_rate": 0.0015394344844770092, + "loss": 0.0908, + "step": 33076 + }, + { + "epoch": 0.2871242437131622, + "grad_norm": 0.154296875, + "learning_rate": 0.0015394085106028815, + "loss": 0.1006, + "step": 33077 + }, + { + "epoch": 0.2871329241933664, + "grad_norm": 0.220703125, + "learning_rate": 0.001539382536248223, + "loss": 0.1143, + "step": 33078 + }, + { + "epoch": 0.2871416046735705, + "grad_norm": 0.09375, + "learning_rate": 0.0015393565614130623, + "loss": 0.0776, + "step": 33079 + }, + { + "epoch": 0.2871502851537747, + "grad_norm": 0.32421875, + "learning_rate": 0.001539330586097428, + "loss": 0.1011, + "step": 33080 + }, + { + "epoch": 0.28715896563397886, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015393046103013483, + "loss": 0.0889, + "step": 33081 + }, + { + "epoch": 0.28716764611418305, + "grad_norm": 0.140625, + "learning_rate": 0.001539278634024852, + "loss": 0.0986, + "step": 33082 + }, + { + "epoch": 0.2871763265943872, + "grad_norm": 0.1142578125, + "learning_rate": 0.0015392526572679671, + "loss": 0.0791, + "step": 33083 + }, + { + "epoch": 0.2871850070745914, + "grad_norm": 0.37109375, + "learning_rate": 0.001539226680030722, + "loss": 0.0737, + "step": 33084 + }, + { + "epoch": 0.2871936875547955, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015392007023131448, + "loss": 0.0928, + "step": 33085 + }, + { + "epoch": 0.2872023680349997, + "grad_norm": 0.69921875, + "learning_rate": 0.0015391747241152646, + "loss": 0.1152, + "step": 33086 + }, + { + "epoch": 0.28721104851520385, + "grad_norm": 0.27734375, + "learning_rate": 0.0015391487454371098, + "loss": 0.1064, + "step": 33087 + }, + { + "epoch": 0.28721972899540804, + "grad_norm": 0.10302734375, + "learning_rate": 0.0015391227662787084, + "loss": 0.1226, + "step": 33088 + }, + { + "epoch": 0.2872284094756122, + "grad_norm": 0.19140625, + "learning_rate": 0.001539096786640089, + "loss": 0.123, + "step": 33089 + }, + { + "epoch": 0.28723708995581637, + "grad_norm": 0.248046875, + "learning_rate": 0.0015390708065212795, + "loss": 0.123, + "step": 33090 + }, + { + "epoch": 0.2872457704360205, + "grad_norm": 0.52734375, + "learning_rate": 0.001539044825922309, + "loss": 0.1035, + "step": 33091 + }, + { + "epoch": 0.2872544509162247, + "grad_norm": 0.404296875, + "learning_rate": 0.0015390188448432059, + "loss": 0.1143, + "step": 33092 + }, + { + "epoch": 0.28726313139642884, + "grad_norm": 0.255859375, + "learning_rate": 0.0015389928632839982, + "loss": 0.1055, + "step": 33093 + }, + { + "epoch": 0.28727181187663303, + "grad_norm": 0.177734375, + "learning_rate": 0.0015389668812447145, + "loss": 0.1021, + "step": 33094 + }, + { + "epoch": 0.28728049235683717, + "grad_norm": 0.5390625, + "learning_rate": 0.0015389408987253833, + "loss": 0.0947, + "step": 33095 + }, + { + "epoch": 0.28728917283704136, + "grad_norm": 0.09228515625, + "learning_rate": 0.001538914915726033, + "loss": 0.1279, + "step": 33096 + }, + { + "epoch": 0.2872978533172455, + "grad_norm": 0.5859375, + "learning_rate": 0.001538888932246692, + "loss": 0.0708, + "step": 33097 + }, + { + "epoch": 0.2873065337974497, + "grad_norm": 0.9453125, + "learning_rate": 0.001538862948287388, + "loss": 0.1055, + "step": 33098 + }, + { + "epoch": 0.28731521427765383, + "grad_norm": 0.494140625, + "learning_rate": 0.0015388369638481508, + "loss": 0.1387, + "step": 33099 + }, + { + "epoch": 0.287323894757858, + "grad_norm": 0.25390625, + "learning_rate": 0.0015388109789290078, + "loss": 0.1006, + "step": 33100 + }, + { + "epoch": 0.28733257523806216, + "grad_norm": 0.310546875, + "learning_rate": 0.0015387849935299877, + "loss": 0.0747, + "step": 33101 + }, + { + "epoch": 0.28734125571826635, + "grad_norm": 0.267578125, + "learning_rate": 0.0015387590076511189, + "loss": 0.1074, + "step": 33102 + }, + { + "epoch": 0.2873499361984705, + "grad_norm": 0.29296875, + "learning_rate": 0.00153873302129243, + "loss": 0.1035, + "step": 33103 + }, + { + "epoch": 0.2873586166786747, + "grad_norm": 0.291015625, + "learning_rate": 0.0015387070344539493, + "loss": 0.1094, + "step": 33104 + }, + { + "epoch": 0.2873672971588788, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015386810471357051, + "loss": 0.0811, + "step": 33105 + }, + { + "epoch": 0.287375977639083, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015386550593377258, + "loss": 0.1104, + "step": 33106 + }, + { + "epoch": 0.28738465811928715, + "grad_norm": 0.10107421875, + "learning_rate": 0.00153862907106004, + "loss": 0.0718, + "step": 33107 + }, + { + "epoch": 0.28739333859949134, + "grad_norm": 0.6015625, + "learning_rate": 0.0015386030823026757, + "loss": 0.1416, + "step": 33108 + }, + { + "epoch": 0.2874020190796955, + "grad_norm": 0.37109375, + "learning_rate": 0.0015385770930656621, + "loss": 0.0957, + "step": 33109 + }, + { + "epoch": 0.2874106995598997, + "grad_norm": 1.015625, + "learning_rate": 0.0015385511033490272, + "loss": 0.1689, + "step": 33110 + }, + { + "epoch": 0.2874193800401038, + "grad_norm": 0.455078125, + "learning_rate": 0.001538525113152799, + "loss": 0.1543, + "step": 33111 + }, + { + "epoch": 0.287428060520308, + "grad_norm": 0.490234375, + "learning_rate": 0.0015384991224770069, + "loss": 0.1055, + "step": 33112 + }, + { + "epoch": 0.28743674100051214, + "grad_norm": 0.07421875, + "learning_rate": 0.0015384731313216784, + "loss": 0.0859, + "step": 33113 + }, + { + "epoch": 0.28744542148071633, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015384471396868422, + "loss": 0.085, + "step": 33114 + }, + { + "epoch": 0.28745410196092047, + "grad_norm": 0.2109375, + "learning_rate": 0.0015384211475725268, + "loss": 0.0957, + "step": 33115 + }, + { + "epoch": 0.28746278244112466, + "grad_norm": 0.271484375, + "learning_rate": 0.0015383951549787606, + "loss": 0.0864, + "step": 33116 + }, + { + "epoch": 0.2874714629213288, + "grad_norm": 0.2451171875, + "learning_rate": 0.0015383691619055724, + "loss": 0.1191, + "step": 33117 + }, + { + "epoch": 0.287480143401533, + "grad_norm": 0.267578125, + "learning_rate": 0.0015383431683529901, + "loss": 0.0972, + "step": 33118 + }, + { + "epoch": 0.28748882388173713, + "grad_norm": 0.1376953125, + "learning_rate": 0.001538317174321042, + "loss": 0.123, + "step": 33119 + }, + { + "epoch": 0.2874975043619413, + "grad_norm": 0.90234375, + "learning_rate": 0.001538291179809757, + "loss": 0.1064, + "step": 33120 + }, + { + "epoch": 0.28750618484214546, + "grad_norm": 0.32421875, + "learning_rate": 0.0015382651848191635, + "loss": 0.1211, + "step": 33121 + }, + { + "epoch": 0.28751486532234966, + "grad_norm": 0.53125, + "learning_rate": 0.0015382391893492896, + "loss": 0.1055, + "step": 33122 + }, + { + "epoch": 0.2875235458025538, + "grad_norm": 0.3125, + "learning_rate": 0.0015382131934001642, + "loss": 0.2324, + "step": 33123 + }, + { + "epoch": 0.287532226282758, + "grad_norm": 0.2294921875, + "learning_rate": 0.001538187196971815, + "loss": 0.0923, + "step": 33124 + }, + { + "epoch": 0.2875409067629621, + "grad_norm": 0.26171875, + "learning_rate": 0.001538161200064271, + "loss": 0.126, + "step": 33125 + }, + { + "epoch": 0.2875495872431663, + "grad_norm": 0.271484375, + "learning_rate": 0.0015381352026775609, + "loss": 0.0908, + "step": 33126 + }, + { + "epoch": 0.28755826772337045, + "grad_norm": 0.337890625, + "learning_rate": 0.0015381092048117123, + "loss": 0.126, + "step": 33127 + }, + { + "epoch": 0.28756694820357465, + "grad_norm": 0.22265625, + "learning_rate": 0.001538083206466754, + "loss": 0.1177, + "step": 33128 + }, + { + "epoch": 0.2875756286837788, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015380572076427149, + "loss": 0.0708, + "step": 33129 + }, + { + "epoch": 0.287584309163983, + "grad_norm": 0.40234375, + "learning_rate": 0.0015380312083396225, + "loss": 0.1094, + "step": 33130 + }, + { + "epoch": 0.2875929896441871, + "grad_norm": 0.1181640625, + "learning_rate": 0.0015380052085575062, + "loss": 0.0781, + "step": 33131 + }, + { + "epoch": 0.2876016701243913, + "grad_norm": 0.1181640625, + "learning_rate": 0.0015379792082963937, + "loss": 0.1196, + "step": 33132 + }, + { + "epoch": 0.28761035060459544, + "grad_norm": 0.1064453125, + "learning_rate": 0.0015379532075563138, + "loss": 0.1172, + "step": 33133 + }, + { + "epoch": 0.28761903108479964, + "grad_norm": 0.1923828125, + "learning_rate": 0.0015379272063372948, + "loss": 0.1191, + "step": 33134 + }, + { + "epoch": 0.2876277115650038, + "grad_norm": 0.6015625, + "learning_rate": 0.0015379012046393655, + "loss": 0.1152, + "step": 33135 + }, + { + "epoch": 0.28763639204520797, + "grad_norm": 0.48828125, + "learning_rate": 0.0015378752024625538, + "loss": 0.1162, + "step": 33136 + }, + { + "epoch": 0.2876450725254121, + "grad_norm": 0.119140625, + "learning_rate": 0.0015378491998068883, + "loss": 0.1064, + "step": 33137 + }, + { + "epoch": 0.2876537530056163, + "grad_norm": 0.3828125, + "learning_rate": 0.0015378231966723973, + "loss": 0.0679, + "step": 33138 + }, + { + "epoch": 0.28766243348582043, + "grad_norm": 0.11669921875, + "learning_rate": 0.0015377971930591097, + "loss": 0.1182, + "step": 33139 + }, + { + "epoch": 0.2876711139660246, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015377711889670539, + "loss": 0.1133, + "step": 33140 + }, + { + "epoch": 0.28767979444622876, + "grad_norm": 0.65234375, + "learning_rate": 0.0015377451843962575, + "loss": 0.1221, + "step": 33141 + }, + { + "epoch": 0.28768847492643296, + "grad_norm": 0.375, + "learning_rate": 0.0015377191793467502, + "loss": 0.0879, + "step": 33142 + }, + { + "epoch": 0.2876971554066371, + "grad_norm": 1.0390625, + "learning_rate": 0.0015376931738185593, + "loss": 0.0957, + "step": 33143 + }, + { + "epoch": 0.28770583588684123, + "grad_norm": 0.20703125, + "learning_rate": 0.001537667167811714, + "loss": 0.1123, + "step": 33144 + }, + { + "epoch": 0.2877145163670454, + "grad_norm": 0.6015625, + "learning_rate": 0.0015376411613262425, + "loss": 0.1162, + "step": 33145 + }, + { + "epoch": 0.28772319684724956, + "grad_norm": 0.1123046875, + "learning_rate": 0.0015376151543621728, + "loss": 0.0908, + "step": 33146 + }, + { + "epoch": 0.28773187732745376, + "grad_norm": 0.72265625, + "learning_rate": 0.001537589146919534, + "loss": 0.0938, + "step": 33147 + }, + { + "epoch": 0.2877405578076579, + "grad_norm": 0.1474609375, + "learning_rate": 0.001537563138998354, + "loss": 0.124, + "step": 33148 + }, + { + "epoch": 0.2877492382878621, + "grad_norm": 0.66796875, + "learning_rate": 0.0015375371305986624, + "loss": 0.1182, + "step": 33149 + }, + { + "epoch": 0.2877579187680662, + "grad_norm": 0.427734375, + "learning_rate": 0.0015375111217204858, + "loss": 0.0986, + "step": 33150 + }, + { + "epoch": 0.2877665992482704, + "grad_norm": 0.36328125, + "learning_rate": 0.001537485112363854, + "loss": 0.1069, + "step": 33151 + }, + { + "epoch": 0.28777527972847455, + "grad_norm": 0.5703125, + "learning_rate": 0.001537459102528795, + "loss": 0.1328, + "step": 33152 + }, + { + "epoch": 0.28778396020867875, + "grad_norm": 0.107421875, + "learning_rate": 0.0015374330922153375, + "loss": 0.0884, + "step": 33153 + }, + { + "epoch": 0.2877926406888829, + "grad_norm": 0.345703125, + "learning_rate": 0.0015374070814235094, + "loss": 0.0845, + "step": 33154 + }, + { + "epoch": 0.2878013211690871, + "grad_norm": 0.77734375, + "learning_rate": 0.0015373810701533397, + "loss": 0.1016, + "step": 33155 + }, + { + "epoch": 0.2878100016492912, + "grad_norm": 0.203125, + "learning_rate": 0.0015373550584048565, + "loss": 0.0791, + "step": 33156 + }, + { + "epoch": 0.2878186821294954, + "grad_norm": 0.2578125, + "learning_rate": 0.0015373290461780887, + "loss": 0.1318, + "step": 33157 + }, + { + "epoch": 0.28782736260969954, + "grad_norm": 0.44921875, + "learning_rate": 0.001537303033473064, + "loss": 0.105, + "step": 33158 + }, + { + "epoch": 0.28783604308990374, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015372770202898114, + "loss": 0.1021, + "step": 33159 + }, + { + "epoch": 0.2878447235701079, + "grad_norm": 1.3515625, + "learning_rate": 0.0015372510066283596, + "loss": 0.1025, + "step": 33160 + }, + { + "epoch": 0.28785340405031207, + "grad_norm": 0.326171875, + "learning_rate": 0.0015372249924887362, + "loss": 0.1089, + "step": 33161 + }, + { + "epoch": 0.2878620845305162, + "grad_norm": 0.13671875, + "learning_rate": 0.0015371989778709702, + "loss": 0.1001, + "step": 33162 + }, + { + "epoch": 0.2878707650107204, + "grad_norm": 0.388671875, + "learning_rate": 0.0015371729627750902, + "loss": 0.1064, + "step": 33163 + }, + { + "epoch": 0.28787944549092453, + "grad_norm": 0.09912109375, + "learning_rate": 0.0015371469472011243, + "loss": 0.0703, + "step": 33164 + }, + { + "epoch": 0.2878881259711287, + "grad_norm": 0.296875, + "learning_rate": 0.001537120931149101, + "loss": 0.1484, + "step": 33165 + }, + { + "epoch": 0.28789680645133287, + "grad_norm": 0.6953125, + "learning_rate": 0.0015370949146190488, + "loss": 0.1367, + "step": 33166 + }, + { + "epoch": 0.28790548693153706, + "grad_norm": 0.357421875, + "learning_rate": 0.0015370688976109959, + "loss": 0.1196, + "step": 33167 + }, + { + "epoch": 0.2879141674117412, + "grad_norm": 0.890625, + "learning_rate": 0.0015370428801249716, + "loss": 0.0938, + "step": 33168 + }, + { + "epoch": 0.2879228478919454, + "grad_norm": 0.13671875, + "learning_rate": 0.0015370168621610034, + "loss": 0.1445, + "step": 33169 + }, + { + "epoch": 0.2879315283721495, + "grad_norm": 0.2109375, + "learning_rate": 0.0015369908437191204, + "loss": 0.1123, + "step": 33170 + }, + { + "epoch": 0.2879402088523537, + "grad_norm": 0.0947265625, + "learning_rate": 0.0015369648247993508, + "loss": 0.084, + "step": 33171 + }, + { + "epoch": 0.28794888933255786, + "grad_norm": 0.451171875, + "learning_rate": 0.0015369388054017226, + "loss": 0.0947, + "step": 33172 + }, + { + "epoch": 0.28795756981276205, + "grad_norm": 0.177734375, + "learning_rate": 0.001536912785526265, + "loss": 0.125, + "step": 33173 + }, + { + "epoch": 0.2879662502929662, + "grad_norm": 0.1201171875, + "learning_rate": 0.0015368867651730062, + "loss": 0.0996, + "step": 33174 + }, + { + "epoch": 0.2879749307731704, + "grad_norm": 0.2578125, + "learning_rate": 0.0015368607443419745, + "loss": 0.1123, + "step": 33175 + }, + { + "epoch": 0.2879836112533745, + "grad_norm": 0.26171875, + "learning_rate": 0.0015368347230331984, + "loss": 0.1523, + "step": 33176 + }, + { + "epoch": 0.2879922917335787, + "grad_norm": 0.162109375, + "learning_rate": 0.0015368087012467063, + "loss": 0.0986, + "step": 33177 + }, + { + "epoch": 0.28800097221378285, + "grad_norm": 0.78515625, + "learning_rate": 0.001536782678982527, + "loss": 0.1406, + "step": 33178 + }, + { + "epoch": 0.28800965269398704, + "grad_norm": 0.11572265625, + "learning_rate": 0.0015367566562406885, + "loss": 0.0952, + "step": 33179 + }, + { + "epoch": 0.2880183331741912, + "grad_norm": 0.66796875, + "learning_rate": 0.0015367306330212197, + "loss": 0.1162, + "step": 33180 + }, + { + "epoch": 0.28802701365439537, + "grad_norm": 0.0859375, + "learning_rate": 0.0015367046093241487, + "loss": 0.103, + "step": 33181 + }, + { + "epoch": 0.2880356941345995, + "grad_norm": 0.185546875, + "learning_rate": 0.0015366785851495045, + "loss": 0.1055, + "step": 33182 + }, + { + "epoch": 0.2880443746148037, + "grad_norm": 0.54296875, + "learning_rate": 0.0015366525604973147, + "loss": 0.1016, + "step": 33183 + }, + { + "epoch": 0.28805305509500784, + "grad_norm": 0.3359375, + "learning_rate": 0.001536626535367608, + "loss": 0.1035, + "step": 33184 + }, + { + "epoch": 0.28806173557521203, + "grad_norm": 0.37890625, + "learning_rate": 0.0015366005097604137, + "loss": 0.0952, + "step": 33185 + }, + { + "epoch": 0.28807041605541617, + "grad_norm": 0.251953125, + "learning_rate": 0.0015365744836757591, + "loss": 0.1045, + "step": 33186 + }, + { + "epoch": 0.28807909653562036, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015365484571136737, + "loss": 0.1113, + "step": 33187 + }, + { + "epoch": 0.2880877770158245, + "grad_norm": 0.64453125, + "learning_rate": 0.001536522430074185, + "loss": 0.1475, + "step": 33188 + }, + { + "epoch": 0.2880964574960287, + "grad_norm": 0.2451171875, + "learning_rate": 0.0015364964025573222, + "loss": 0.0977, + "step": 33189 + }, + { + "epoch": 0.28810513797623283, + "grad_norm": 0.67578125, + "learning_rate": 0.0015364703745631134, + "loss": 0.1182, + "step": 33190 + }, + { + "epoch": 0.288113818456437, + "grad_norm": 0.435546875, + "learning_rate": 0.0015364443460915873, + "loss": 0.0986, + "step": 33191 + }, + { + "epoch": 0.28812249893664116, + "grad_norm": 0.1455078125, + "learning_rate": 0.0015364183171427722, + "loss": 0.1436, + "step": 33192 + }, + { + "epoch": 0.28813117941684535, + "grad_norm": 0.361328125, + "learning_rate": 0.0015363922877166965, + "loss": 0.1016, + "step": 33193 + }, + { + "epoch": 0.2881398598970495, + "grad_norm": 0.640625, + "learning_rate": 0.0015363662578133889, + "loss": 0.0815, + "step": 33194 + }, + { + "epoch": 0.2881485403772537, + "grad_norm": 0.78515625, + "learning_rate": 0.0015363402274328774, + "loss": 0.1191, + "step": 33195 + }, + { + "epoch": 0.2881572208574578, + "grad_norm": 0.82421875, + "learning_rate": 0.001536314196575191, + "loss": 0.1084, + "step": 33196 + }, + { + "epoch": 0.288165901337662, + "grad_norm": 0.314453125, + "learning_rate": 0.0015362881652403579, + "loss": 0.0898, + "step": 33197 + }, + { + "epoch": 0.28817458181786615, + "grad_norm": 0.1796875, + "learning_rate": 0.0015362621334284066, + "loss": 0.1006, + "step": 33198 + }, + { + "epoch": 0.28818326229807034, + "grad_norm": 0.2109375, + "learning_rate": 0.0015362361011393658, + "loss": 0.0776, + "step": 33199 + }, + { + "epoch": 0.2881919427782745, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015362100683732633, + "loss": 0.1001, + "step": 33200 + }, + { + "epoch": 0.28820062325847867, + "grad_norm": 0.5234375, + "learning_rate": 0.0015361840351301285, + "loss": 0.1074, + "step": 33201 + }, + { + "epoch": 0.2882093037386828, + "grad_norm": 0.333984375, + "learning_rate": 0.0015361580014099892, + "loss": 0.1094, + "step": 33202 + }, + { + "epoch": 0.288217984218887, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015361319672128743, + "loss": 0.0986, + "step": 33203 + }, + { + "epoch": 0.28822666469909114, + "grad_norm": 0.197265625, + "learning_rate": 0.0015361059325388116, + "loss": 0.124, + "step": 33204 + }, + { + "epoch": 0.28823534517929533, + "grad_norm": 0.48828125, + "learning_rate": 0.0015360798973878302, + "loss": 0.0972, + "step": 33205 + }, + { + "epoch": 0.28824402565949947, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015360538617599582, + "loss": 0.1211, + "step": 33206 + }, + { + "epoch": 0.28825270613970366, + "grad_norm": 0.408203125, + "learning_rate": 0.0015360278256552245, + "loss": 0.1162, + "step": 33207 + }, + { + "epoch": 0.2882613866199078, + "grad_norm": 0.259765625, + "learning_rate": 0.001536001789073657, + "loss": 0.0835, + "step": 33208 + }, + { + "epoch": 0.288270067100112, + "grad_norm": 0.1826171875, + "learning_rate": 0.0015359757520152849, + "loss": 0.1035, + "step": 33209 + }, + { + "epoch": 0.28827874758031613, + "grad_norm": 0.5234375, + "learning_rate": 0.0015359497144801363, + "loss": 0.0938, + "step": 33210 + }, + { + "epoch": 0.2882874280605203, + "grad_norm": 0.1103515625, + "learning_rate": 0.001535923676468239, + "loss": 0.084, + "step": 33211 + }, + { + "epoch": 0.28829610854072446, + "grad_norm": 0.22265625, + "learning_rate": 0.0015358976379796227, + "loss": 0.1123, + "step": 33212 + }, + { + "epoch": 0.28830478902092865, + "grad_norm": 0.275390625, + "learning_rate": 0.001535871599014315, + "loss": 0.1182, + "step": 33213 + }, + { + "epoch": 0.2883134695011328, + "grad_norm": 0.2099609375, + "learning_rate": 0.0015358455595723451, + "loss": 0.0889, + "step": 33214 + }, + { + "epoch": 0.288322149981337, + "grad_norm": 0.3125, + "learning_rate": 0.0015358195196537407, + "loss": 0.0825, + "step": 33215 + }, + { + "epoch": 0.2883308304615411, + "grad_norm": 0.134765625, + "learning_rate": 0.0015357934792585306, + "loss": 0.083, + "step": 33216 + }, + { + "epoch": 0.2883395109417453, + "grad_norm": 0.150390625, + "learning_rate": 0.0015357674383867437, + "loss": 0.1299, + "step": 33217 + }, + { + "epoch": 0.28834819142194945, + "grad_norm": 0.43359375, + "learning_rate": 0.0015357413970384078, + "loss": 0.126, + "step": 33218 + }, + { + "epoch": 0.28835687190215364, + "grad_norm": 0.26953125, + "learning_rate": 0.0015357153552135515, + "loss": 0.1338, + "step": 33219 + }, + { + "epoch": 0.2883655523823578, + "grad_norm": 0.244140625, + "learning_rate": 0.0015356893129122035, + "loss": 0.1123, + "step": 33220 + }, + { + "epoch": 0.288374232862562, + "grad_norm": 0.10302734375, + "learning_rate": 0.0015356632701343921, + "loss": 0.1211, + "step": 33221 + }, + { + "epoch": 0.2883829133427661, + "grad_norm": 0.09521484375, + "learning_rate": 0.0015356372268801465, + "loss": 0.1104, + "step": 33222 + }, + { + "epoch": 0.2883915938229703, + "grad_norm": 0.150390625, + "learning_rate": 0.001535611183149494, + "loss": 0.1074, + "step": 33223 + }, + { + "epoch": 0.28840027430317444, + "grad_norm": 0.515625, + "learning_rate": 0.0015355851389424636, + "loss": 0.1147, + "step": 33224 + }, + { + "epoch": 0.28840895478337863, + "grad_norm": 0.64453125, + "learning_rate": 0.0015355590942590838, + "loss": 0.0908, + "step": 33225 + }, + { + "epoch": 0.2884176352635828, + "grad_norm": 0.23828125, + "learning_rate": 0.0015355330490993836, + "loss": 0.0996, + "step": 33226 + }, + { + "epoch": 0.28842631574378697, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015355070034633906, + "loss": 0.1162, + "step": 33227 + }, + { + "epoch": 0.2884349962239911, + "grad_norm": 0.6484375, + "learning_rate": 0.001535480957351134, + "loss": 0.1025, + "step": 33228 + }, + { + "epoch": 0.2884436767041953, + "grad_norm": 0.2333984375, + "learning_rate": 0.0015354549107626416, + "loss": 0.1396, + "step": 33229 + }, + { + "epoch": 0.28845235718439943, + "grad_norm": 0.875, + "learning_rate": 0.0015354288636979425, + "loss": 0.0898, + "step": 33230 + }, + { + "epoch": 0.2884610376646036, + "grad_norm": 1.6015625, + "learning_rate": 0.001535402816157065, + "loss": 0.3789, + "step": 33231 + }, + { + "epoch": 0.28846971814480776, + "grad_norm": 0.298828125, + "learning_rate": 0.0015353767681400372, + "loss": 0.1035, + "step": 33232 + }, + { + "epoch": 0.28847839862501196, + "grad_norm": 0.55078125, + "learning_rate": 0.001535350719646888, + "loss": 0.1514, + "step": 33233 + }, + { + "epoch": 0.2884870791052161, + "grad_norm": 0.1875, + "learning_rate": 0.001535324670677646, + "loss": 0.0889, + "step": 33234 + }, + { + "epoch": 0.2884957595854203, + "grad_norm": 0.13671875, + "learning_rate": 0.0015352986212323397, + "loss": 0.1367, + "step": 33235 + }, + { + "epoch": 0.2885044400656244, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015352725713109966, + "loss": 0.1562, + "step": 33236 + }, + { + "epoch": 0.2885131205458286, + "grad_norm": 0.73828125, + "learning_rate": 0.0015352465209136467, + "loss": 0.1182, + "step": 33237 + }, + { + "epoch": 0.28852180102603275, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015352204700403175, + "loss": 0.1172, + "step": 33238 + }, + { + "epoch": 0.28853048150623695, + "grad_norm": 0.8671875, + "learning_rate": 0.0015351944186910377, + "loss": 0.0894, + "step": 33239 + }, + { + "epoch": 0.2885391619864411, + "grad_norm": 0.88671875, + "learning_rate": 0.0015351683668658358, + "loss": 0.123, + "step": 33240 + }, + { + "epoch": 0.2885478424666453, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015351423145647405, + "loss": 0.0977, + "step": 33241 + }, + { + "epoch": 0.2885565229468494, + "grad_norm": 1.0078125, + "learning_rate": 0.0015351162617877798, + "loss": 0.0967, + "step": 33242 + }, + { + "epoch": 0.2885652034270536, + "grad_norm": 0.2177734375, + "learning_rate": 0.0015350902085349825, + "loss": 0.127, + "step": 33243 + }, + { + "epoch": 0.28857388390725774, + "grad_norm": 0.2109375, + "learning_rate": 0.0015350641548063776, + "loss": 0.0928, + "step": 33244 + }, + { + "epoch": 0.28858256438746194, + "grad_norm": 0.263671875, + "learning_rate": 0.0015350381006019923, + "loss": 0.0752, + "step": 33245 + }, + { + "epoch": 0.2885912448676661, + "grad_norm": 0.10107421875, + "learning_rate": 0.0015350120459218563, + "loss": 0.1143, + "step": 33246 + }, + { + "epoch": 0.28859992534787027, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015349859907659975, + "loss": 0.0664, + "step": 33247 + }, + { + "epoch": 0.2886086058280744, + "grad_norm": 0.5, + "learning_rate": 0.001534959935134445, + "loss": 0.1426, + "step": 33248 + }, + { + "epoch": 0.2886172863082786, + "grad_norm": 0.08935546875, + "learning_rate": 0.0015349338790272263, + "loss": 0.1045, + "step": 33249 + }, + { + "epoch": 0.28862596678848274, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015349078224443707, + "loss": 0.1182, + "step": 33250 + }, + { + "epoch": 0.28863464726868693, + "grad_norm": 0.10986328125, + "learning_rate": 0.0015348817653859067, + "loss": 0.1162, + "step": 33251 + }, + { + "epoch": 0.28864332774889107, + "grad_norm": 0.2265625, + "learning_rate": 0.0015348557078518619, + "loss": 0.0913, + "step": 33252 + }, + { + "epoch": 0.28865200822909526, + "grad_norm": 0.7265625, + "learning_rate": 0.0015348296498422659, + "loss": 0.1562, + "step": 33253 + }, + { + "epoch": 0.2886606887092994, + "grad_norm": 0.35546875, + "learning_rate": 0.0015348035913571466, + "loss": 0.1348, + "step": 33254 + }, + { + "epoch": 0.2886693691895036, + "grad_norm": 0.302734375, + "learning_rate": 0.0015347775323965326, + "loss": 0.0942, + "step": 33255 + }, + { + "epoch": 0.2886780496697077, + "grad_norm": 0.50390625, + "learning_rate": 0.0015347514729604523, + "loss": 0.0957, + "step": 33256 + }, + { + "epoch": 0.2886867301499119, + "grad_norm": 0.1083984375, + "learning_rate": 0.0015347254130489347, + "loss": 0.1016, + "step": 33257 + }, + { + "epoch": 0.28869541063011606, + "grad_norm": 0.3984375, + "learning_rate": 0.0015346993526620077, + "loss": 0.1191, + "step": 33258 + }, + { + "epoch": 0.28870409111032025, + "grad_norm": 0.39453125, + "learning_rate": 0.0015346732917997, + "loss": 0.105, + "step": 33259 + }, + { + "epoch": 0.2887127715905244, + "grad_norm": 0.27734375, + "learning_rate": 0.00153464723046204, + "loss": 0.0752, + "step": 33260 + }, + { + "epoch": 0.2887214520707286, + "grad_norm": 0.4765625, + "learning_rate": 0.0015346211686490566, + "loss": 0.0898, + "step": 33261 + }, + { + "epoch": 0.2887301325509327, + "grad_norm": 0.482421875, + "learning_rate": 0.0015345951063607778, + "loss": 0.1846, + "step": 33262 + }, + { + "epoch": 0.2887388130311369, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015345690435972323, + "loss": 0.1387, + "step": 33263 + }, + { + "epoch": 0.28874749351134105, + "grad_norm": 0.10791015625, + "learning_rate": 0.0015345429803584488, + "loss": 0.1035, + "step": 33264 + }, + { + "epoch": 0.28875617399154524, + "grad_norm": 0.283203125, + "learning_rate": 0.0015345169166444557, + "loss": 0.123, + "step": 33265 + }, + { + "epoch": 0.2887648544717494, + "grad_norm": 0.263671875, + "learning_rate": 0.0015344908524552812, + "loss": 0.1279, + "step": 33266 + }, + { + "epoch": 0.2887735349519535, + "grad_norm": 0.259765625, + "learning_rate": 0.0015344647877909541, + "loss": 0.0957, + "step": 33267 + }, + { + "epoch": 0.2887822154321577, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015344387226515028, + "loss": 0.0811, + "step": 33268 + }, + { + "epoch": 0.28879089591236184, + "grad_norm": 0.265625, + "learning_rate": 0.0015344126570369558, + "loss": 0.1123, + "step": 33269 + }, + { + "epoch": 0.28879957639256604, + "grad_norm": 0.2197265625, + "learning_rate": 0.001534386590947342, + "loss": 0.0981, + "step": 33270 + }, + { + "epoch": 0.2888082568727702, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015343605243826892, + "loss": 0.0825, + "step": 33271 + }, + { + "epoch": 0.28881693735297437, + "grad_norm": 0.703125, + "learning_rate": 0.001534334457343026, + "loss": 0.1445, + "step": 33272 + }, + { + "epoch": 0.2888256178331785, + "grad_norm": 0.203125, + "learning_rate": 0.0015343083898283817, + "loss": 0.1006, + "step": 33273 + }, + { + "epoch": 0.2888342983133827, + "grad_norm": 0.1953125, + "learning_rate": 0.001534282321838784, + "loss": 0.0957, + "step": 33274 + }, + { + "epoch": 0.28884297879358684, + "grad_norm": 0.158203125, + "learning_rate": 0.0015342562533742616, + "loss": 0.1006, + "step": 33275 + }, + { + "epoch": 0.28885165927379103, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015342301844348432, + "loss": 0.0996, + "step": 33276 + }, + { + "epoch": 0.28886033975399517, + "grad_norm": 0.439453125, + "learning_rate": 0.0015342041150205575, + "loss": 0.1011, + "step": 33277 + }, + { + "epoch": 0.28886902023419936, + "grad_norm": 0.263671875, + "learning_rate": 0.0015341780451314324, + "loss": 0.083, + "step": 33278 + }, + { + "epoch": 0.2888777007144035, + "grad_norm": 0.11328125, + "learning_rate": 0.0015341519747674968, + "loss": 0.1084, + "step": 33279 + }, + { + "epoch": 0.2888863811946077, + "grad_norm": 0.86328125, + "learning_rate": 0.001534125903928779, + "loss": 0.1055, + "step": 33280 + }, + { + "epoch": 0.2888950616748118, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015340998326153076, + "loss": 0.0903, + "step": 33281 + }, + { + "epoch": 0.288903742155016, + "grad_norm": 0.4921875, + "learning_rate": 0.0015340737608271113, + "loss": 0.1191, + "step": 33282 + }, + { + "epoch": 0.28891242263522016, + "grad_norm": 0.3359375, + "learning_rate": 0.0015340476885642184, + "loss": 0.124, + "step": 33283 + }, + { + "epoch": 0.28892110311542435, + "grad_norm": 0.162109375, + "learning_rate": 0.0015340216158266572, + "loss": 0.1118, + "step": 33284 + }, + { + "epoch": 0.2889297835956285, + "grad_norm": 0.392578125, + "learning_rate": 0.0015339955426144569, + "loss": 0.104, + "step": 33285 + }, + { + "epoch": 0.2889384640758327, + "grad_norm": 0.107421875, + "learning_rate": 0.0015339694689276455, + "loss": 0.0928, + "step": 33286 + }, + { + "epoch": 0.2889471445560368, + "grad_norm": 0.515625, + "learning_rate": 0.0015339433947662517, + "loss": 0.0986, + "step": 33287 + }, + { + "epoch": 0.288955825036241, + "grad_norm": 0.0927734375, + "learning_rate": 0.0015339173201303037, + "loss": 0.0811, + "step": 33288 + }, + { + "epoch": 0.28896450551644515, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015338912450198303, + "loss": 0.1816, + "step": 33289 + }, + { + "epoch": 0.28897318599664934, + "grad_norm": 0.1708984375, + "learning_rate": 0.00153386516943486, + "loss": 0.1128, + "step": 33290 + }, + { + "epoch": 0.2889818664768535, + "grad_norm": 0.1982421875, + "learning_rate": 0.001533839093375421, + "loss": 0.0859, + "step": 33291 + }, + { + "epoch": 0.28899054695705767, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015338130168415425, + "loss": 0.0713, + "step": 33292 + }, + { + "epoch": 0.2889992274372618, + "grad_norm": 0.181640625, + "learning_rate": 0.0015337869398332523, + "loss": 0.1016, + "step": 33293 + }, + { + "epoch": 0.289007907917466, + "grad_norm": 0.2216796875, + "learning_rate": 0.0015337608623505795, + "loss": 0.1182, + "step": 33294 + }, + { + "epoch": 0.28901658839767014, + "grad_norm": 0.60546875, + "learning_rate": 0.0015337347843935522, + "loss": 0.1104, + "step": 33295 + }, + { + "epoch": 0.28902526887787433, + "grad_norm": 0.1884765625, + "learning_rate": 0.001533708705962199, + "loss": 0.1396, + "step": 33296 + }, + { + "epoch": 0.28903394935807847, + "grad_norm": 0.283203125, + "learning_rate": 0.0015336826270565486, + "loss": 0.1177, + "step": 33297 + }, + { + "epoch": 0.28904262983828266, + "grad_norm": 0.462890625, + "learning_rate": 0.0015336565476766293, + "loss": 0.0928, + "step": 33298 + }, + { + "epoch": 0.2890513103184868, + "grad_norm": 0.2158203125, + "learning_rate": 0.00153363046782247, + "loss": 0.1582, + "step": 33299 + }, + { + "epoch": 0.289059990798691, + "grad_norm": 0.6015625, + "learning_rate": 0.0015336043874940983, + "loss": 0.084, + "step": 33300 + }, + { + "epoch": 0.28906867127889513, + "grad_norm": 0.67578125, + "learning_rate": 0.0015335783066915435, + "loss": 0.1572, + "step": 33301 + }, + { + "epoch": 0.2890773517590993, + "grad_norm": 0.3125, + "learning_rate": 0.0015335522254148345, + "loss": 0.0693, + "step": 33302 + }, + { + "epoch": 0.28908603223930346, + "grad_norm": 0.126953125, + "learning_rate": 0.001533526143663999, + "loss": 0.083, + "step": 33303 + }, + { + "epoch": 0.28909471271950765, + "grad_norm": 0.1328125, + "learning_rate": 0.001533500061439066, + "loss": 0.1011, + "step": 33304 + }, + { + "epoch": 0.2891033931997118, + "grad_norm": 0.3671875, + "learning_rate": 0.0015334739787400636, + "loss": 0.0986, + "step": 33305 + }, + { + "epoch": 0.289112073679916, + "grad_norm": 0.171875, + "learning_rate": 0.0015334478955670203, + "loss": 0.1025, + "step": 33306 + }, + { + "epoch": 0.2891207541601201, + "grad_norm": 0.18359375, + "learning_rate": 0.0015334218119199656, + "loss": 0.0986, + "step": 33307 + }, + { + "epoch": 0.2891294346403243, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015333957277989267, + "loss": 0.1162, + "step": 33308 + }, + { + "epoch": 0.28913811512052845, + "grad_norm": 0.50390625, + "learning_rate": 0.001533369643203933, + "loss": 0.0967, + "step": 33309 + }, + { + "epoch": 0.28914679560073264, + "grad_norm": 0.28125, + "learning_rate": 0.0015333435581350128, + "loss": 0.0781, + "step": 33310 + }, + { + "epoch": 0.2891554760809368, + "grad_norm": 0.40234375, + "learning_rate": 0.0015333174725921945, + "loss": 0.0874, + "step": 33311 + }, + { + "epoch": 0.289164156561141, + "grad_norm": 0.232421875, + "learning_rate": 0.0015332913865755068, + "loss": 0.0898, + "step": 33312 + }, + { + "epoch": 0.2891728370413451, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015332653000849784, + "loss": 0.0918, + "step": 33313 + }, + { + "epoch": 0.2891815175215493, + "grad_norm": 0.1630859375, + "learning_rate": 0.001533239213120637, + "loss": 0.1084, + "step": 33314 + }, + { + "epoch": 0.28919019800175344, + "grad_norm": 0.4296875, + "learning_rate": 0.0015332131256825122, + "loss": 0.1533, + "step": 33315 + }, + { + "epoch": 0.28919887848195763, + "grad_norm": 0.115234375, + "learning_rate": 0.001533187037770632, + "loss": 0.1289, + "step": 33316 + }, + { + "epoch": 0.28920755896216177, + "grad_norm": 0.296875, + "learning_rate": 0.0015331609493850248, + "loss": 0.1104, + "step": 33317 + }, + { + "epoch": 0.28921623944236596, + "grad_norm": 0.796875, + "learning_rate": 0.001533134860525719, + "loss": 0.1016, + "step": 33318 + }, + { + "epoch": 0.2892249199225701, + "grad_norm": 0.1552734375, + "learning_rate": 0.0015331087711927438, + "loss": 0.0928, + "step": 33319 + }, + { + "epoch": 0.2892336004027743, + "grad_norm": 0.298828125, + "learning_rate": 0.0015330826813861275, + "loss": 0.1328, + "step": 33320 + }, + { + "epoch": 0.28924228088297843, + "grad_norm": 0.201171875, + "learning_rate": 0.0015330565911058983, + "loss": 0.0942, + "step": 33321 + }, + { + "epoch": 0.2892509613631826, + "grad_norm": 0.111328125, + "learning_rate": 0.001533030500352085, + "loss": 0.1221, + "step": 33322 + }, + { + "epoch": 0.28925964184338676, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015330044091247158, + "loss": 0.1123, + "step": 33323 + }, + { + "epoch": 0.28926832232359095, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015329783174238196, + "loss": 0.1289, + "step": 33324 + }, + { + "epoch": 0.2892770028037951, + "grad_norm": 0.1796875, + "learning_rate": 0.001532952225249425, + "loss": 0.0996, + "step": 33325 + }, + { + "epoch": 0.2892856832839993, + "grad_norm": 0.470703125, + "learning_rate": 0.0015329261326015602, + "loss": 0.1064, + "step": 33326 + }, + { + "epoch": 0.2892943637642034, + "grad_norm": 0.44921875, + "learning_rate": 0.0015329000394802537, + "loss": 0.0918, + "step": 33327 + }, + { + "epoch": 0.2893030442444076, + "grad_norm": 0.7109375, + "learning_rate": 0.0015328739458855345, + "loss": 0.0811, + "step": 33328 + }, + { + "epoch": 0.28931172472461175, + "grad_norm": 0.478515625, + "learning_rate": 0.0015328478518174306, + "loss": 0.1211, + "step": 33329 + }, + { + "epoch": 0.28932040520481594, + "grad_norm": 0.3046875, + "learning_rate": 0.0015328217572759708, + "loss": 0.1016, + "step": 33330 + }, + { + "epoch": 0.2893290856850201, + "grad_norm": 0.4609375, + "learning_rate": 0.0015327956622611838, + "loss": 0.085, + "step": 33331 + }, + { + "epoch": 0.2893377661652243, + "grad_norm": 0.76171875, + "learning_rate": 0.001532769566773098, + "loss": 0.1006, + "step": 33332 + }, + { + "epoch": 0.2893464466454284, + "grad_norm": 0.578125, + "learning_rate": 0.001532743470811742, + "loss": 0.209, + "step": 33333 + }, + { + "epoch": 0.2893551271256326, + "grad_norm": 0.30859375, + "learning_rate": 0.0015327173743771441, + "loss": 0.1055, + "step": 33334 + }, + { + "epoch": 0.28936380760583674, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015326912774693332, + "loss": 0.1553, + "step": 33335 + }, + { + "epoch": 0.28937248808604094, + "grad_norm": 0.205078125, + "learning_rate": 0.0015326651800883372, + "loss": 0.0957, + "step": 33336 + }, + { + "epoch": 0.2893811685662451, + "grad_norm": 0.3984375, + "learning_rate": 0.0015326390822341853, + "loss": 0.1123, + "step": 33337 + }, + { + "epoch": 0.28938984904644927, + "grad_norm": 0.58203125, + "learning_rate": 0.0015326129839069056, + "loss": 0.1309, + "step": 33338 + }, + { + "epoch": 0.2893985295266534, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015325868851065272, + "loss": 0.1094, + "step": 33339 + }, + { + "epoch": 0.2894072100068576, + "grad_norm": 0.10205078125, + "learning_rate": 0.001532560785833078, + "loss": 0.1011, + "step": 33340 + }, + { + "epoch": 0.28941589048706173, + "grad_norm": 1.09375, + "learning_rate": 0.0015325346860865867, + "loss": 0.2305, + "step": 33341 + }, + { + "epoch": 0.2894245709672659, + "grad_norm": 0.16796875, + "learning_rate": 0.0015325085858670824, + "loss": 0.0898, + "step": 33342 + }, + { + "epoch": 0.28943325144747006, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015324824851745932, + "loss": 0.0879, + "step": 33343 + }, + { + "epoch": 0.28944193192767426, + "grad_norm": 0.0830078125, + "learning_rate": 0.001532456384009147, + "loss": 0.1064, + "step": 33344 + }, + { + "epoch": 0.2894506124078784, + "grad_norm": 0.43359375, + "learning_rate": 0.0015324302823707737, + "loss": 0.1084, + "step": 33345 + }, + { + "epoch": 0.2894592928880826, + "grad_norm": 0.1875, + "learning_rate": 0.0015324041802595007, + "loss": 0.0603, + "step": 33346 + }, + { + "epoch": 0.2894679733682867, + "grad_norm": 0.25, + "learning_rate": 0.0015323780776753575, + "loss": 0.1099, + "step": 33347 + }, + { + "epoch": 0.2894766538484909, + "grad_norm": 1.5234375, + "learning_rate": 0.0015323519746183716, + "loss": 0.2695, + "step": 33348 + }, + { + "epoch": 0.28948533432869505, + "grad_norm": 0.099609375, + "learning_rate": 0.001532325871088572, + "loss": 0.0986, + "step": 33349 + }, + { + "epoch": 0.28949401480889925, + "grad_norm": 0.349609375, + "learning_rate": 0.0015322997670859878, + "loss": 0.1562, + "step": 33350 + }, + { + "epoch": 0.2895026952891034, + "grad_norm": 0.341796875, + "learning_rate": 0.001532273662610647, + "loss": 0.0737, + "step": 33351 + }, + { + "epoch": 0.2895113757693076, + "grad_norm": 0.1201171875, + "learning_rate": 0.0015322475576625778, + "loss": 0.0728, + "step": 33352 + }, + { + "epoch": 0.2895200562495117, + "grad_norm": 0.240234375, + "learning_rate": 0.0015322214522418094, + "loss": 0.0879, + "step": 33353 + }, + { + "epoch": 0.2895287367297159, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015321953463483702, + "loss": 0.1045, + "step": 33354 + }, + { + "epoch": 0.28953741720992004, + "grad_norm": 1.0625, + "learning_rate": 0.001532169239982289, + "loss": 0.1387, + "step": 33355 + }, + { + "epoch": 0.28954609769012424, + "grad_norm": 0.5078125, + "learning_rate": 0.0015321431331435935, + "loss": 0.0938, + "step": 33356 + }, + { + "epoch": 0.2895547781703284, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015321170258323128, + "loss": 0.1226, + "step": 33357 + }, + { + "epoch": 0.28956345865053257, + "grad_norm": 0.296875, + "learning_rate": 0.0015320909180484752, + "loss": 0.1025, + "step": 33358 + }, + { + "epoch": 0.2895721391307367, + "grad_norm": 0.1396484375, + "learning_rate": 0.00153206480979211, + "loss": 0.1416, + "step": 33359 + }, + { + "epoch": 0.2895808196109409, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015320387010632449, + "loss": 0.1328, + "step": 33360 + }, + { + "epoch": 0.28958950009114504, + "grad_norm": 0.3203125, + "learning_rate": 0.001532012591861909, + "loss": 0.0791, + "step": 33361 + }, + { + "epoch": 0.28959818057134923, + "grad_norm": 0.4921875, + "learning_rate": 0.0015319864821881305, + "loss": 0.1152, + "step": 33362 + }, + { + "epoch": 0.28960686105155337, + "grad_norm": 0.2294921875, + "learning_rate": 0.001531960372041938, + "loss": 0.0845, + "step": 33363 + }, + { + "epoch": 0.28961554153175756, + "grad_norm": 0.19140625, + "learning_rate": 0.0015319342614233602, + "loss": 0.1387, + "step": 33364 + }, + { + "epoch": 0.2896242220119617, + "grad_norm": 0.37109375, + "learning_rate": 0.0015319081503324256, + "loss": 0.0859, + "step": 33365 + }, + { + "epoch": 0.2896329024921659, + "grad_norm": 0.326171875, + "learning_rate": 0.0015318820387691626, + "loss": 0.1182, + "step": 33366 + }, + { + "epoch": 0.28964158297237, + "grad_norm": 0.3359375, + "learning_rate": 0.0015318559267336, + "loss": 0.0889, + "step": 33367 + }, + { + "epoch": 0.2896502634525742, + "grad_norm": 0.53125, + "learning_rate": 0.001531829814225766, + "loss": 0.1172, + "step": 33368 + }, + { + "epoch": 0.28965894393277836, + "grad_norm": 0.314453125, + "learning_rate": 0.0015318037012456898, + "loss": 0.1104, + "step": 33369 + }, + { + "epoch": 0.28966762441298255, + "grad_norm": 0.33984375, + "learning_rate": 0.0015317775877933992, + "loss": 0.123, + "step": 33370 + }, + { + "epoch": 0.2896763048931867, + "grad_norm": 0.1572265625, + "learning_rate": 0.0015317514738689235, + "loss": 0.0991, + "step": 33371 + }, + { + "epoch": 0.2896849853733909, + "grad_norm": 0.2255859375, + "learning_rate": 0.0015317253594722905, + "loss": 0.0776, + "step": 33372 + }, + { + "epoch": 0.289693665853595, + "grad_norm": 0.365234375, + "learning_rate": 0.0015316992446035294, + "loss": 0.1104, + "step": 33373 + }, + { + "epoch": 0.2897023463337992, + "grad_norm": 0.39453125, + "learning_rate": 0.0015316731292626686, + "loss": 0.0928, + "step": 33374 + }, + { + "epoch": 0.28971102681400335, + "grad_norm": 0.1181640625, + "learning_rate": 0.0015316470134497362, + "loss": 0.1055, + "step": 33375 + }, + { + "epoch": 0.28971970729420754, + "grad_norm": 0.51953125, + "learning_rate": 0.0015316208971647613, + "loss": 0.0908, + "step": 33376 + }, + { + "epoch": 0.2897283877744117, + "grad_norm": 0.369140625, + "learning_rate": 0.001531594780407772, + "loss": 0.124, + "step": 33377 + }, + { + "epoch": 0.28973706825461587, + "grad_norm": 0.12109375, + "learning_rate": 0.0015315686631787978, + "loss": 0.1348, + "step": 33378 + }, + { + "epoch": 0.28974574873482, + "grad_norm": 0.1640625, + "learning_rate": 0.0015315425454778659, + "loss": 0.125, + "step": 33379 + }, + { + "epoch": 0.2897544292150242, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015315164273050057, + "loss": 0.1367, + "step": 33380 + }, + { + "epoch": 0.28976310969522834, + "grad_norm": 0.5390625, + "learning_rate": 0.0015314903086602462, + "loss": 0.0947, + "step": 33381 + }, + { + "epoch": 0.28977179017543253, + "grad_norm": 0.171875, + "learning_rate": 0.001531464189543615, + "loss": 0.1055, + "step": 33382 + }, + { + "epoch": 0.28978047065563667, + "grad_norm": 0.4609375, + "learning_rate": 0.0015314380699551406, + "loss": 0.1211, + "step": 33383 + }, + { + "epoch": 0.28978915113584086, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015314119498948525, + "loss": 0.1289, + "step": 33384 + }, + { + "epoch": 0.289797831616045, + "grad_norm": 0.79296875, + "learning_rate": 0.0015313858293627789, + "loss": 0.1494, + "step": 33385 + }, + { + "epoch": 0.2898065120962492, + "grad_norm": 0.1171875, + "learning_rate": 0.0015313597083589476, + "loss": 0.1289, + "step": 33386 + }, + { + "epoch": 0.28981519257645333, + "grad_norm": 0.466796875, + "learning_rate": 0.0015313335868833884, + "loss": 0.0967, + "step": 33387 + }, + { + "epoch": 0.2898238730566575, + "grad_norm": 0.330078125, + "learning_rate": 0.0015313074649361289, + "loss": 0.0845, + "step": 33388 + }, + { + "epoch": 0.28983255353686166, + "grad_norm": 0.275390625, + "learning_rate": 0.0015312813425171982, + "loss": 0.127, + "step": 33389 + }, + { + "epoch": 0.2898412340170658, + "grad_norm": 0.3125, + "learning_rate": 0.0015312552196266246, + "loss": 0.1338, + "step": 33390 + }, + { + "epoch": 0.28984991449727, + "grad_norm": 0.15234375, + "learning_rate": 0.001531229096264437, + "loss": 0.1143, + "step": 33391 + }, + { + "epoch": 0.2898585949774741, + "grad_norm": 0.4140625, + "learning_rate": 0.0015312029724306635, + "loss": 0.0811, + "step": 33392 + }, + { + "epoch": 0.2898672754576783, + "grad_norm": 0.73828125, + "learning_rate": 0.0015311768481253334, + "loss": 0.106, + "step": 33393 + }, + { + "epoch": 0.28987595593788246, + "grad_norm": 0.095703125, + "learning_rate": 0.0015311507233484742, + "loss": 0.1064, + "step": 33394 + }, + { + "epoch": 0.28988463641808665, + "grad_norm": 0.4609375, + "learning_rate": 0.0015311245981001155, + "loss": 0.1089, + "step": 33395 + }, + { + "epoch": 0.2898933168982908, + "grad_norm": 0.58984375, + "learning_rate": 0.001531098472380285, + "loss": 0.167, + "step": 33396 + }, + { + "epoch": 0.289901997378495, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015310723461890118, + "loss": 0.0859, + "step": 33397 + }, + { + "epoch": 0.2899106778586991, + "grad_norm": 0.369140625, + "learning_rate": 0.0015310462195263243, + "loss": 0.1045, + "step": 33398 + }, + { + "epoch": 0.2899193583389033, + "grad_norm": 0.0859375, + "learning_rate": 0.0015310200923922515, + "loss": 0.103, + "step": 33399 + }, + { + "epoch": 0.28992803881910745, + "grad_norm": 0.76953125, + "learning_rate": 0.0015309939647868214, + "loss": 0.125, + "step": 33400 + }, + { + "epoch": 0.28993671929931164, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015309678367100626, + "loss": 0.0505, + "step": 33401 + }, + { + "epoch": 0.2899453997795158, + "grad_norm": 0.17578125, + "learning_rate": 0.001530941708162004, + "loss": 0.0801, + "step": 33402 + }, + { + "epoch": 0.28995408025971997, + "grad_norm": 0.75390625, + "learning_rate": 0.0015309155791426738, + "loss": 0.1089, + "step": 33403 + }, + { + "epoch": 0.2899627607399241, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015308894496521013, + "loss": 0.1016, + "step": 33404 + }, + { + "epoch": 0.2899714412201283, + "grad_norm": 0.2412109375, + "learning_rate": 0.001530863319690314, + "loss": 0.1055, + "step": 33405 + }, + { + "epoch": 0.28998012170033244, + "grad_norm": 0.91796875, + "learning_rate": 0.0015308371892573413, + "loss": 0.1484, + "step": 33406 + }, + { + "epoch": 0.28998880218053663, + "grad_norm": 0.111328125, + "learning_rate": 0.0015308110583532118, + "loss": 0.0947, + "step": 33407 + }, + { + "epoch": 0.28999748266074077, + "grad_norm": 0.34375, + "learning_rate": 0.0015307849269779535, + "loss": 0.085, + "step": 33408 + }, + { + "epoch": 0.29000616314094496, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015307587951315952, + "loss": 0.1074, + "step": 33409 + }, + { + "epoch": 0.2900148436211491, + "grad_norm": 0.302734375, + "learning_rate": 0.0015307326628141655, + "loss": 0.1235, + "step": 33410 + }, + { + "epoch": 0.2900235241013533, + "grad_norm": 0.390625, + "learning_rate": 0.0015307065300256934, + "loss": 0.1011, + "step": 33411 + }, + { + "epoch": 0.29003220458155743, + "grad_norm": 0.58203125, + "learning_rate": 0.001530680396766207, + "loss": 0.1079, + "step": 33412 + }, + { + "epoch": 0.2900408850617616, + "grad_norm": 0.09912109375, + "learning_rate": 0.001530654263035735, + "loss": 0.1021, + "step": 33413 + }, + { + "epoch": 0.29004956554196576, + "grad_norm": 0.2734375, + "learning_rate": 0.0015306281288343057, + "loss": 0.1416, + "step": 33414 + }, + { + "epoch": 0.29005824602216995, + "grad_norm": 0.130859375, + "learning_rate": 0.001530601994161948, + "loss": 0.1094, + "step": 33415 + }, + { + "epoch": 0.2900669265023741, + "grad_norm": 0.146484375, + "learning_rate": 0.0015305758590186908, + "loss": 0.1289, + "step": 33416 + }, + { + "epoch": 0.2900756069825783, + "grad_norm": 0.46875, + "learning_rate": 0.001530549723404562, + "loss": 0.0996, + "step": 33417 + }, + { + "epoch": 0.2900842874627824, + "grad_norm": 0.359375, + "learning_rate": 0.00153052358731959, + "loss": 0.1055, + "step": 33418 + }, + { + "epoch": 0.2900929679429866, + "grad_norm": 0.53125, + "learning_rate": 0.0015304974507638048, + "loss": 0.1035, + "step": 33419 + }, + { + "epoch": 0.29010164842319075, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015304713137372337, + "loss": 0.1016, + "step": 33420 + }, + { + "epoch": 0.29011032890339494, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015304451762399056, + "loss": 0.1279, + "step": 33421 + }, + { + "epoch": 0.2901190093835991, + "grad_norm": 0.357421875, + "learning_rate": 0.001530419038271849, + "loss": 0.1504, + "step": 33422 + }, + { + "epoch": 0.2901276898638033, + "grad_norm": 0.302734375, + "learning_rate": 0.001530392899833093, + "loss": 0.0859, + "step": 33423 + }, + { + "epoch": 0.2901363703440074, + "grad_norm": 0.138671875, + "learning_rate": 0.0015303667609236655, + "loss": 0.123, + "step": 33424 + }, + { + "epoch": 0.2901450508242116, + "grad_norm": 0.361328125, + "learning_rate": 0.0015303406215435953, + "loss": 0.085, + "step": 33425 + }, + { + "epoch": 0.29015373130441574, + "grad_norm": 0.263671875, + "learning_rate": 0.0015303144816929113, + "loss": 0.1089, + "step": 33426 + }, + { + "epoch": 0.29016241178461993, + "grad_norm": 0.384765625, + "learning_rate": 0.0015302883413716415, + "loss": 0.0952, + "step": 33427 + }, + { + "epoch": 0.29017109226482407, + "grad_norm": 0.0986328125, + "learning_rate": 0.0015302622005798152, + "loss": 0.0957, + "step": 33428 + }, + { + "epoch": 0.29017977274502826, + "grad_norm": 0.55078125, + "learning_rate": 0.0015302360593174606, + "loss": 0.1191, + "step": 33429 + }, + { + "epoch": 0.2901884532252324, + "grad_norm": 0.328125, + "learning_rate": 0.0015302099175846062, + "loss": 0.0859, + "step": 33430 + }, + { + "epoch": 0.2901971337054366, + "grad_norm": 0.126953125, + "learning_rate": 0.0015301837753812803, + "loss": 0.0815, + "step": 33431 + }, + { + "epoch": 0.29020581418564073, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015301576327075123, + "loss": 0.1074, + "step": 33432 + }, + { + "epoch": 0.2902144946658449, + "grad_norm": 0.6796875, + "learning_rate": 0.0015301314895633307, + "loss": 0.0767, + "step": 33433 + }, + { + "epoch": 0.29022317514604906, + "grad_norm": 0.2001953125, + "learning_rate": 0.0015301053459487633, + "loss": 0.0894, + "step": 33434 + }, + { + "epoch": 0.29023185562625325, + "grad_norm": 0.330078125, + "learning_rate": 0.001530079201863839, + "loss": 0.082, + "step": 33435 + }, + { + "epoch": 0.2902405361064574, + "grad_norm": 0.3671875, + "learning_rate": 0.0015300530573085869, + "loss": 0.1035, + "step": 33436 + }, + { + "epoch": 0.2902492165866616, + "grad_norm": 0.2392578125, + "learning_rate": 0.0015300269122830354, + "loss": 0.0986, + "step": 33437 + }, + { + "epoch": 0.2902578970668657, + "grad_norm": 0.953125, + "learning_rate": 0.0015300007667872126, + "loss": 0.1152, + "step": 33438 + }, + { + "epoch": 0.2902665775470699, + "grad_norm": 0.18359375, + "learning_rate": 0.0015299746208211477, + "loss": 0.1006, + "step": 33439 + }, + { + "epoch": 0.29027525802727405, + "grad_norm": 0.546875, + "learning_rate": 0.0015299484743848688, + "loss": 0.1016, + "step": 33440 + }, + { + "epoch": 0.29028393850747825, + "grad_norm": 0.1484375, + "learning_rate": 0.0015299223274784046, + "loss": 0.1045, + "step": 33441 + }, + { + "epoch": 0.2902926189876824, + "grad_norm": 0.54296875, + "learning_rate": 0.0015298961801017842, + "loss": 0.1064, + "step": 33442 + }, + { + "epoch": 0.2903012994678866, + "grad_norm": 0.2333984375, + "learning_rate": 0.0015298700322550352, + "loss": 0.1855, + "step": 33443 + }, + { + "epoch": 0.2903099799480907, + "grad_norm": 0.27734375, + "learning_rate": 0.0015298438839381873, + "loss": 0.0708, + "step": 33444 + }, + { + "epoch": 0.2903186604282949, + "grad_norm": 0.2734375, + "learning_rate": 0.0015298177351512681, + "loss": 0.1602, + "step": 33445 + }, + { + "epoch": 0.29032734090849904, + "grad_norm": 0.546875, + "learning_rate": 0.0015297915858943072, + "loss": 0.0913, + "step": 33446 + }, + { + "epoch": 0.29033602138870324, + "grad_norm": 0.306640625, + "learning_rate": 0.0015297654361673323, + "loss": 0.1118, + "step": 33447 + }, + { + "epoch": 0.2903447018689074, + "grad_norm": 1.4296875, + "learning_rate": 0.0015297392859703724, + "loss": 0.0947, + "step": 33448 + }, + { + "epoch": 0.29035338234911157, + "grad_norm": 0.1953125, + "learning_rate": 0.0015297131353034564, + "loss": 0.1562, + "step": 33449 + }, + { + "epoch": 0.2903620628293157, + "grad_norm": 0.58203125, + "learning_rate": 0.0015296869841666125, + "loss": 0.1328, + "step": 33450 + }, + { + "epoch": 0.2903707433095199, + "grad_norm": 0.1875, + "learning_rate": 0.0015296608325598693, + "loss": 0.1152, + "step": 33451 + }, + { + "epoch": 0.29037942378972403, + "grad_norm": 0.73046875, + "learning_rate": 0.0015296346804832554, + "loss": 0.0938, + "step": 33452 + }, + { + "epoch": 0.2903881042699282, + "grad_norm": 0.55078125, + "learning_rate": 0.0015296085279367995, + "loss": 0.0903, + "step": 33453 + }, + { + "epoch": 0.29039678475013236, + "grad_norm": 0.291015625, + "learning_rate": 0.00152958237492053, + "loss": 0.1543, + "step": 33454 + }, + { + "epoch": 0.29040546523033656, + "grad_norm": 1.0078125, + "learning_rate": 0.001529556221434476, + "loss": 0.332, + "step": 33455 + }, + { + "epoch": 0.2904141457105407, + "grad_norm": 0.2158203125, + "learning_rate": 0.0015295300674786655, + "loss": 0.1182, + "step": 33456 + }, + { + "epoch": 0.2904228261907449, + "grad_norm": 0.57421875, + "learning_rate": 0.0015295039130531277, + "loss": 0.127, + "step": 33457 + }, + { + "epoch": 0.290431506670949, + "grad_norm": 0.32421875, + "learning_rate": 0.0015294777581578904, + "loss": 0.1113, + "step": 33458 + }, + { + "epoch": 0.2904401871511532, + "grad_norm": 0.1806640625, + "learning_rate": 0.001529451602792983, + "loss": 0.1191, + "step": 33459 + }, + { + "epoch": 0.29044886763135735, + "grad_norm": 0.2158203125, + "learning_rate": 0.0015294254469584339, + "loss": 0.1113, + "step": 33460 + }, + { + "epoch": 0.29045754811156155, + "grad_norm": 0.44140625, + "learning_rate": 0.0015293992906542713, + "loss": 0.1318, + "step": 33461 + }, + { + "epoch": 0.2904662285917657, + "grad_norm": 0.146484375, + "learning_rate": 0.0015293731338805244, + "loss": 0.1416, + "step": 33462 + }, + { + "epoch": 0.2904749090719699, + "grad_norm": 0.21484375, + "learning_rate": 0.001529346976637221, + "loss": 0.1104, + "step": 33463 + }, + { + "epoch": 0.290483589552174, + "grad_norm": 0.51171875, + "learning_rate": 0.0015293208189243904, + "loss": 0.0635, + "step": 33464 + }, + { + "epoch": 0.2904922700323782, + "grad_norm": 0.83203125, + "learning_rate": 0.001529294660742061, + "loss": 0.1045, + "step": 33465 + }, + { + "epoch": 0.29050095051258235, + "grad_norm": 0.337890625, + "learning_rate": 0.0015292685020902614, + "loss": 0.0977, + "step": 33466 + }, + { + "epoch": 0.29050963099278654, + "grad_norm": 0.53515625, + "learning_rate": 0.0015292423429690205, + "loss": 0.1484, + "step": 33467 + }, + { + "epoch": 0.2905183114729907, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015292161833783664, + "loss": 0.126, + "step": 33468 + }, + { + "epoch": 0.29052699195319487, + "grad_norm": 0.29296875, + "learning_rate": 0.001529190023318328, + "loss": 0.1201, + "step": 33469 + }, + { + "epoch": 0.290535672433399, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015291638627889337, + "loss": 0.1045, + "step": 33470 + }, + { + "epoch": 0.2905443529136032, + "grad_norm": 0.20703125, + "learning_rate": 0.0015291377017902124, + "loss": 0.1875, + "step": 33471 + }, + { + "epoch": 0.29055303339380734, + "grad_norm": 0.138671875, + "learning_rate": 0.0015291115403221927, + "loss": 0.0859, + "step": 33472 + }, + { + "epoch": 0.29056171387401153, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015290853783849027, + "loss": 0.1279, + "step": 33473 + }, + { + "epoch": 0.29057039435421567, + "grad_norm": 0.11474609375, + "learning_rate": 0.0015290592159783716, + "loss": 0.1221, + "step": 33474 + }, + { + "epoch": 0.29057907483441986, + "grad_norm": 0.16796875, + "learning_rate": 0.0015290330531026277, + "loss": 0.1006, + "step": 33475 + }, + { + "epoch": 0.290587755314624, + "grad_norm": 0.41015625, + "learning_rate": 0.0015290068897576997, + "loss": 0.1182, + "step": 33476 + }, + { + "epoch": 0.2905964357948282, + "grad_norm": 0.228515625, + "learning_rate": 0.001528980725943616, + "loss": 0.105, + "step": 33477 + }, + { + "epoch": 0.2906051162750323, + "grad_norm": 0.0908203125, + "learning_rate": 0.001528954561660406, + "loss": 0.0942, + "step": 33478 + }, + { + "epoch": 0.2906137967552365, + "grad_norm": 0.11572265625, + "learning_rate": 0.0015289283969080972, + "loss": 0.1099, + "step": 33479 + }, + { + "epoch": 0.29062247723544066, + "grad_norm": 0.31640625, + "learning_rate": 0.0015289022316867192, + "loss": 0.0986, + "step": 33480 + }, + { + "epoch": 0.29063115771564485, + "grad_norm": 0.0673828125, + "learning_rate": 0.0015288760659963, + "loss": 0.0718, + "step": 33481 + }, + { + "epoch": 0.290639838195849, + "grad_norm": 0.37109375, + "learning_rate": 0.001528849899836868, + "loss": 0.1367, + "step": 33482 + }, + { + "epoch": 0.2906485186760532, + "grad_norm": 0.10302734375, + "learning_rate": 0.0015288237332084522, + "loss": 0.0801, + "step": 33483 + }, + { + "epoch": 0.2906571991562573, + "grad_norm": 0.1201171875, + "learning_rate": 0.0015287975661110818, + "loss": 0.1279, + "step": 33484 + }, + { + "epoch": 0.2906658796364615, + "grad_norm": 0.45703125, + "learning_rate": 0.0015287713985447844, + "loss": 0.1211, + "step": 33485 + }, + { + "epoch": 0.29067456011666565, + "grad_norm": 0.09521484375, + "learning_rate": 0.0015287452305095893, + "loss": 0.0938, + "step": 33486 + }, + { + "epoch": 0.29068324059686984, + "grad_norm": 0.75, + "learning_rate": 0.0015287190620055246, + "loss": 0.1387, + "step": 33487 + }, + { + "epoch": 0.290691921077074, + "grad_norm": 0.10400390625, + "learning_rate": 0.0015286928930326193, + "loss": 0.1162, + "step": 33488 + }, + { + "epoch": 0.29070060155727817, + "grad_norm": 0.1455078125, + "learning_rate": 0.001528666723590902, + "loss": 0.0981, + "step": 33489 + }, + { + "epoch": 0.2907092820374823, + "grad_norm": 0.48828125, + "learning_rate": 0.0015286405536804012, + "loss": 0.1025, + "step": 33490 + }, + { + "epoch": 0.2907179625176865, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015286143833011451, + "loss": 0.1084, + "step": 33491 + }, + { + "epoch": 0.29072664299789064, + "grad_norm": 0.2890625, + "learning_rate": 0.0015285882124531633, + "loss": 0.1211, + "step": 33492 + }, + { + "epoch": 0.29073532347809483, + "grad_norm": 0.609375, + "learning_rate": 0.0015285620411364836, + "loss": 0.1348, + "step": 33493 + }, + { + "epoch": 0.29074400395829897, + "grad_norm": 0.2080078125, + "learning_rate": 0.0015285358693511348, + "loss": 0.1045, + "step": 33494 + }, + { + "epoch": 0.29075268443850316, + "grad_norm": 0.10205078125, + "learning_rate": 0.001528509697097146, + "loss": 0.0835, + "step": 33495 + }, + { + "epoch": 0.2907613649187073, + "grad_norm": 0.0791015625, + "learning_rate": 0.001528483524374545, + "loss": 0.1011, + "step": 33496 + }, + { + "epoch": 0.2907700453989115, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015284573511833612, + "loss": 0.103, + "step": 33497 + }, + { + "epoch": 0.29077872587911563, + "grad_norm": 0.40234375, + "learning_rate": 0.0015284311775236226, + "loss": 0.1133, + "step": 33498 + }, + { + "epoch": 0.2907874063593198, + "grad_norm": 0.259765625, + "learning_rate": 0.0015284050033953585, + "loss": 0.0869, + "step": 33499 + }, + { + "epoch": 0.29079608683952396, + "grad_norm": 0.7265625, + "learning_rate": 0.0015283788287985968, + "loss": 0.0869, + "step": 33500 + }, + { + "epoch": 0.29080476731972815, + "grad_norm": 0.271484375, + "learning_rate": 0.0015283526537333662, + "loss": 0.0791, + "step": 33501 + }, + { + "epoch": 0.2908134477999323, + "grad_norm": 0.453125, + "learning_rate": 0.0015283264781996958, + "loss": 0.124, + "step": 33502 + }, + { + "epoch": 0.2908221282801365, + "grad_norm": 0.255859375, + "learning_rate": 0.0015283003021976144, + "loss": 0.1094, + "step": 33503 + }, + { + "epoch": 0.2908308087603406, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015282741257271494, + "loss": 0.0903, + "step": 33504 + }, + { + "epoch": 0.2908394892405448, + "grad_norm": 0.10791015625, + "learning_rate": 0.001528247948788331, + "loss": 0.0703, + "step": 33505 + }, + { + "epoch": 0.29084816972074895, + "grad_norm": 0.486328125, + "learning_rate": 0.0015282217713811868, + "loss": 0.1445, + "step": 33506 + }, + { + "epoch": 0.29085685020095314, + "grad_norm": 0.271484375, + "learning_rate": 0.0015281955935057458, + "loss": 0.1143, + "step": 33507 + }, + { + "epoch": 0.2908655306811573, + "grad_norm": 0.330078125, + "learning_rate": 0.0015281694151620365, + "loss": 0.1416, + "step": 33508 + }, + { + "epoch": 0.2908742111613615, + "grad_norm": 0.412109375, + "learning_rate": 0.0015281432363500873, + "loss": 0.0801, + "step": 33509 + }, + { + "epoch": 0.2908828916415656, + "grad_norm": 0.171875, + "learning_rate": 0.0015281170570699274, + "loss": 0.1113, + "step": 33510 + }, + { + "epoch": 0.2908915721217698, + "grad_norm": 0.2578125, + "learning_rate": 0.001528090877321585, + "loss": 0.0986, + "step": 33511 + }, + { + "epoch": 0.29090025260197394, + "grad_norm": 0.36328125, + "learning_rate": 0.001528064697105089, + "loss": 0.1211, + "step": 33512 + }, + { + "epoch": 0.29090893308217813, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015280385164204676, + "loss": 0.1465, + "step": 33513 + }, + { + "epoch": 0.29091761356238227, + "grad_norm": 0.5234375, + "learning_rate": 0.0015280123352677498, + "loss": 0.0811, + "step": 33514 + }, + { + "epoch": 0.2909262940425864, + "grad_norm": 0.515625, + "learning_rate": 0.0015279861536469643, + "loss": 0.1177, + "step": 33515 + }, + { + "epoch": 0.2909349745227906, + "grad_norm": 0.12451171875, + "learning_rate": 0.0015279599715581395, + "loss": 0.1104, + "step": 33516 + }, + { + "epoch": 0.29094365500299474, + "grad_norm": 0.51953125, + "learning_rate": 0.0015279337890013042, + "loss": 0.1582, + "step": 33517 + }, + { + "epoch": 0.29095233548319893, + "grad_norm": 0.1044921875, + "learning_rate": 0.0015279076059764867, + "loss": 0.1172, + "step": 33518 + }, + { + "epoch": 0.29096101596340307, + "grad_norm": 0.28515625, + "learning_rate": 0.001527881422483716, + "loss": 0.1123, + "step": 33519 + }, + { + "epoch": 0.29096969644360726, + "grad_norm": 0.55078125, + "learning_rate": 0.0015278552385230206, + "loss": 0.1406, + "step": 33520 + }, + { + "epoch": 0.2909783769238114, + "grad_norm": 0.2265625, + "learning_rate": 0.0015278290540944293, + "loss": 0.0913, + "step": 33521 + }, + { + "epoch": 0.2909870574040156, + "grad_norm": 0.318359375, + "learning_rate": 0.00152780286919797, + "loss": 0.0791, + "step": 33522 + }, + { + "epoch": 0.29099573788421973, + "grad_norm": 0.55078125, + "learning_rate": 0.0015277766838336724, + "loss": 0.1777, + "step": 33523 + }, + { + "epoch": 0.2910044183644239, + "grad_norm": 0.310546875, + "learning_rate": 0.0015277504980015647, + "loss": 0.0996, + "step": 33524 + }, + { + "epoch": 0.29101309884462806, + "grad_norm": 0.353515625, + "learning_rate": 0.0015277243117016754, + "loss": 0.1025, + "step": 33525 + }, + { + "epoch": 0.29102177932483225, + "grad_norm": 0.84765625, + "learning_rate": 0.0015276981249340332, + "loss": 0.1543, + "step": 33526 + }, + { + "epoch": 0.2910304598050364, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015276719376986667, + "loss": 0.1094, + "step": 33527 + }, + { + "epoch": 0.2910391402852406, + "grad_norm": 0.283203125, + "learning_rate": 0.0015276457499956047, + "loss": 0.1426, + "step": 33528 + }, + { + "epoch": 0.2910478207654447, + "grad_norm": 0.2158203125, + "learning_rate": 0.0015276195618248757, + "loss": 0.1201, + "step": 33529 + }, + { + "epoch": 0.2910565012456489, + "grad_norm": 0.306640625, + "learning_rate": 0.0015275933731865082, + "loss": 0.0996, + "step": 33530 + }, + { + "epoch": 0.29106518172585305, + "grad_norm": 0.1865234375, + "learning_rate": 0.001527567184080531, + "loss": 0.0801, + "step": 33531 + }, + { + "epoch": 0.29107386220605724, + "grad_norm": 0.1875, + "learning_rate": 0.0015275409945069728, + "loss": 0.1084, + "step": 33532 + }, + { + "epoch": 0.2910825426862614, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015275148044658624, + "loss": 0.1113, + "step": 33533 + }, + { + "epoch": 0.2910912231664656, + "grad_norm": 0.19140625, + "learning_rate": 0.0015274886139572284, + "loss": 0.1357, + "step": 33534 + }, + { + "epoch": 0.2910999036466697, + "grad_norm": 0.52734375, + "learning_rate": 0.0015274624229810985, + "loss": 0.1221, + "step": 33535 + }, + { + "epoch": 0.2911085841268739, + "grad_norm": 0.1396484375, + "learning_rate": 0.0015274362315375028, + "loss": 0.0981, + "step": 33536 + }, + { + "epoch": 0.29111726460707804, + "grad_norm": 0.498046875, + "learning_rate": 0.001527410039626469, + "loss": 0.083, + "step": 33537 + }, + { + "epoch": 0.29112594508728223, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015273838472480263, + "loss": 0.1064, + "step": 33538 + }, + { + "epoch": 0.29113462556748637, + "grad_norm": 0.0859375, + "learning_rate": 0.0015273576544022028, + "loss": 0.0918, + "step": 33539 + }, + { + "epoch": 0.29114330604769056, + "grad_norm": 0.138671875, + "learning_rate": 0.0015273314610890272, + "loss": 0.1064, + "step": 33540 + }, + { + "epoch": 0.2911519865278947, + "grad_norm": 0.18359375, + "learning_rate": 0.0015273052673085287, + "loss": 0.0947, + "step": 33541 + }, + { + "epoch": 0.2911606670080989, + "grad_norm": 0.2431640625, + "learning_rate": 0.0015272790730607353, + "loss": 0.0879, + "step": 33542 + }, + { + "epoch": 0.29116934748830303, + "grad_norm": 0.439453125, + "learning_rate": 0.001527252878345676, + "loss": 0.1406, + "step": 33543 + }, + { + "epoch": 0.2911780279685072, + "grad_norm": 0.275390625, + "learning_rate": 0.0015272266831633794, + "loss": 0.1016, + "step": 33544 + }, + { + "epoch": 0.29118670844871136, + "grad_norm": 0.126953125, + "learning_rate": 0.0015272004875138744, + "loss": 0.0752, + "step": 33545 + }, + { + "epoch": 0.29119538892891556, + "grad_norm": 0.26171875, + "learning_rate": 0.0015271742913971888, + "loss": 0.1035, + "step": 33546 + }, + { + "epoch": 0.2912040694091197, + "grad_norm": 0.236328125, + "learning_rate": 0.0015271480948133524, + "loss": 0.0986, + "step": 33547 + }, + { + "epoch": 0.2912127498893239, + "grad_norm": 0.177734375, + "learning_rate": 0.0015271218977623928, + "loss": 0.1357, + "step": 33548 + }, + { + "epoch": 0.291221430369528, + "grad_norm": 0.5078125, + "learning_rate": 0.0015270957002443393, + "loss": 0.1289, + "step": 33549 + }, + { + "epoch": 0.2912301108497322, + "grad_norm": 0.2099609375, + "learning_rate": 0.0015270695022592201, + "loss": 0.1123, + "step": 33550 + }, + { + "epoch": 0.29123879132993635, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015270433038070646, + "loss": 0.1123, + "step": 33551 + }, + { + "epoch": 0.29124747181014055, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015270171048879005, + "loss": 0.085, + "step": 33552 + }, + { + "epoch": 0.2912561522903447, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015269909055017573, + "loss": 0.0938, + "step": 33553 + }, + { + "epoch": 0.2912648327705489, + "grad_norm": 0.09912109375, + "learning_rate": 0.001526964705648663, + "loss": 0.1113, + "step": 33554 + }, + { + "epoch": 0.291273513250753, + "grad_norm": 0.17578125, + "learning_rate": 0.0015269385053286466, + "loss": 0.1074, + "step": 33555 + }, + { + "epoch": 0.2912821937309572, + "grad_norm": 0.1533203125, + "learning_rate": 0.0015269123045417365, + "loss": 0.0991, + "step": 33556 + }, + { + "epoch": 0.29129087421116134, + "grad_norm": 0.10791015625, + "learning_rate": 0.001526886103287962, + "loss": 0.085, + "step": 33557 + }, + { + "epoch": 0.29129955469136554, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015268599015673507, + "loss": 0.1055, + "step": 33558 + }, + { + "epoch": 0.2913082351715697, + "grad_norm": 0.28515625, + "learning_rate": 0.001526833699379932, + "loss": 0.1016, + "step": 33559 + }, + { + "epoch": 0.29131691565177387, + "grad_norm": 0.34375, + "learning_rate": 0.0015268074967257346, + "loss": 0.124, + "step": 33560 + }, + { + "epoch": 0.291325596131978, + "grad_norm": 0.1455078125, + "learning_rate": 0.0015267812936047866, + "loss": 0.1104, + "step": 33561 + }, + { + "epoch": 0.2913342766121822, + "grad_norm": 0.142578125, + "learning_rate": 0.0015267550900171173, + "loss": 0.0947, + "step": 33562 + }, + { + "epoch": 0.29134295709238633, + "grad_norm": 0.296875, + "learning_rate": 0.001526728885962755, + "loss": 0.1904, + "step": 33563 + }, + { + "epoch": 0.2913516375725905, + "grad_norm": 0.1015625, + "learning_rate": 0.0015267026814417284, + "loss": 0.0972, + "step": 33564 + }, + { + "epoch": 0.29136031805279466, + "grad_norm": 0.291015625, + "learning_rate": 0.0015266764764540661, + "loss": 0.123, + "step": 33565 + }, + { + "epoch": 0.29136899853299886, + "grad_norm": 0.244140625, + "learning_rate": 0.0015266502709997967, + "loss": 0.0903, + "step": 33566 + }, + { + "epoch": 0.291377679013203, + "grad_norm": 0.10546875, + "learning_rate": 0.001526624065078949, + "loss": 0.1221, + "step": 33567 + }, + { + "epoch": 0.2913863594934072, + "grad_norm": 0.109375, + "learning_rate": 0.0015265978586915519, + "loss": 0.0918, + "step": 33568 + }, + { + "epoch": 0.2913950399736113, + "grad_norm": 0.21875, + "learning_rate": 0.0015265716518376336, + "loss": 0.103, + "step": 33569 + }, + { + "epoch": 0.2914037204538155, + "grad_norm": 0.80078125, + "learning_rate": 0.001526545444517223, + "loss": 0.1182, + "step": 33570 + }, + { + "epoch": 0.29141240093401966, + "grad_norm": 0.263671875, + "learning_rate": 0.0015265192367303487, + "loss": 0.127, + "step": 33571 + }, + { + "epoch": 0.29142108141422385, + "grad_norm": 0.1328125, + "learning_rate": 0.0015264930284770394, + "loss": 0.1348, + "step": 33572 + }, + { + "epoch": 0.291429761894428, + "grad_norm": 0.32421875, + "learning_rate": 0.0015264668197573232, + "loss": 0.0908, + "step": 33573 + }, + { + "epoch": 0.2914384423746322, + "grad_norm": 0.267578125, + "learning_rate": 0.0015264406105712299, + "loss": 0.1055, + "step": 33574 + }, + { + "epoch": 0.2914471228548363, + "grad_norm": 0.1337890625, + "learning_rate": 0.0015264144009187876, + "loss": 0.0981, + "step": 33575 + }, + { + "epoch": 0.2914558033350405, + "grad_norm": 0.435546875, + "learning_rate": 0.0015263881908000245, + "loss": 0.1201, + "step": 33576 + }, + { + "epoch": 0.29146448381524465, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015263619802149701, + "loss": 0.0952, + "step": 33577 + }, + { + "epoch": 0.29147316429544884, + "grad_norm": 0.166015625, + "learning_rate": 0.0015263357691636524, + "loss": 0.063, + "step": 33578 + }, + { + "epoch": 0.291481844775653, + "grad_norm": 0.40625, + "learning_rate": 0.0015263095576461, + "loss": 0.0986, + "step": 33579 + }, + { + "epoch": 0.29149052525585717, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015262833456623423, + "loss": 0.1025, + "step": 33580 + }, + { + "epoch": 0.2914992057360613, + "grad_norm": 1.078125, + "learning_rate": 0.0015262571332124077, + "loss": 0.2461, + "step": 33581 + }, + { + "epoch": 0.2915078862162655, + "grad_norm": 0.490234375, + "learning_rate": 0.0015262309202963246, + "loss": 0.0913, + "step": 33582 + }, + { + "epoch": 0.29151656669646964, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015262047069141216, + "loss": 0.1436, + "step": 33583 + }, + { + "epoch": 0.29152524717667383, + "grad_norm": 0.2001953125, + "learning_rate": 0.0015261784930658275, + "loss": 0.1562, + "step": 33584 + }, + { + "epoch": 0.29153392765687797, + "grad_norm": 0.53515625, + "learning_rate": 0.0015261522787514711, + "loss": 0.0825, + "step": 33585 + }, + { + "epoch": 0.29154260813708216, + "grad_norm": 0.7109375, + "learning_rate": 0.001526126063971081, + "loss": 0.0723, + "step": 33586 + }, + { + "epoch": 0.2915512886172863, + "grad_norm": 0.24609375, + "learning_rate": 0.0015260998487246857, + "loss": 0.0933, + "step": 33587 + }, + { + "epoch": 0.2915599690974905, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015260736330123142, + "loss": 0.1055, + "step": 33588 + }, + { + "epoch": 0.2915686495776946, + "grad_norm": 0.40625, + "learning_rate": 0.0015260474168339948, + "loss": 0.1094, + "step": 33589 + }, + { + "epoch": 0.2915773300578988, + "grad_norm": 0.2001953125, + "learning_rate": 0.0015260212001897567, + "loss": 0.0771, + "step": 33590 + }, + { + "epoch": 0.29158601053810296, + "grad_norm": 0.44140625, + "learning_rate": 0.0015259949830796276, + "loss": 0.1084, + "step": 33591 + }, + { + "epoch": 0.29159469101830715, + "grad_norm": 0.890625, + "learning_rate": 0.0015259687655036374, + "loss": 0.125, + "step": 33592 + }, + { + "epoch": 0.2916033714985113, + "grad_norm": 0.296875, + "learning_rate": 0.001525942547461814, + "loss": 0.0947, + "step": 33593 + }, + { + "epoch": 0.2916120519787155, + "grad_norm": 0.11767578125, + "learning_rate": 0.001525916328954186, + "loss": 0.1074, + "step": 33594 + }, + { + "epoch": 0.2916207324589196, + "grad_norm": 0.19140625, + "learning_rate": 0.0015258901099807823, + "loss": 0.1128, + "step": 33595 + }, + { + "epoch": 0.2916294129391238, + "grad_norm": 0.220703125, + "learning_rate": 0.0015258638905416317, + "loss": 0.0972, + "step": 33596 + }, + { + "epoch": 0.29163809341932795, + "grad_norm": 0.10546875, + "learning_rate": 0.0015258376706367629, + "loss": 0.0698, + "step": 33597 + }, + { + "epoch": 0.29164677389953214, + "grad_norm": 0.361328125, + "learning_rate": 0.001525811450266204, + "loss": 0.1338, + "step": 33598 + }, + { + "epoch": 0.2916554543797363, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015257852294299843, + "loss": 0.106, + "step": 33599 + }, + { + "epoch": 0.29166413485994047, + "grad_norm": 1.046875, + "learning_rate": 0.0015257590081281325, + "loss": 0.1064, + "step": 33600 + }, + { + "epoch": 0.2916728153401446, + "grad_norm": 0.11572265625, + "learning_rate": 0.001525732786360677, + "loss": 0.0913, + "step": 33601 + }, + { + "epoch": 0.2916814958203488, + "grad_norm": 0.19140625, + "learning_rate": 0.0015257065641276464, + "loss": 0.1123, + "step": 33602 + }, + { + "epoch": 0.29169017630055294, + "grad_norm": 0.27734375, + "learning_rate": 0.0015256803414290694, + "loss": 0.1406, + "step": 33603 + }, + { + "epoch": 0.29169885678075713, + "grad_norm": 1.1015625, + "learning_rate": 0.0015256541182649748, + "loss": 0.1152, + "step": 33604 + }, + { + "epoch": 0.29170753726096127, + "grad_norm": 0.095703125, + "learning_rate": 0.0015256278946353915, + "loss": 0.127, + "step": 33605 + }, + { + "epoch": 0.29171621774116546, + "grad_norm": 0.72265625, + "learning_rate": 0.0015256016705403478, + "loss": 0.0986, + "step": 33606 + }, + { + "epoch": 0.2917248982213696, + "grad_norm": 0.1171875, + "learning_rate": 0.001525575445979872, + "loss": 0.1235, + "step": 33607 + }, + { + "epoch": 0.2917335787015738, + "grad_norm": 0.0791015625, + "learning_rate": 0.0015255492209539938, + "loss": 0.0938, + "step": 33608 + }, + { + "epoch": 0.29174225918177793, + "grad_norm": 0.431640625, + "learning_rate": 0.0015255229954627414, + "loss": 0.1357, + "step": 33609 + }, + { + "epoch": 0.2917509396619821, + "grad_norm": 0.73828125, + "learning_rate": 0.0015254967695061434, + "loss": 0.1123, + "step": 33610 + }, + { + "epoch": 0.29175962014218626, + "grad_norm": 0.228515625, + "learning_rate": 0.0015254705430842285, + "loss": 0.1377, + "step": 33611 + }, + { + "epoch": 0.29176830062239045, + "grad_norm": 0.62109375, + "learning_rate": 0.0015254443161970253, + "loss": 0.125, + "step": 33612 + }, + { + "epoch": 0.2917769811025946, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015254180888445627, + "loss": 0.0825, + "step": 33613 + }, + { + "epoch": 0.2917856615827988, + "grad_norm": 0.1396484375, + "learning_rate": 0.0015253918610268692, + "loss": 0.0933, + "step": 33614 + }, + { + "epoch": 0.2917943420630029, + "grad_norm": 0.337890625, + "learning_rate": 0.0015253656327439739, + "loss": 0.0801, + "step": 33615 + }, + { + "epoch": 0.2918030225432071, + "grad_norm": 0.20703125, + "learning_rate": 0.0015253394039959048, + "loss": 0.0981, + "step": 33616 + }, + { + "epoch": 0.29181170302341125, + "grad_norm": 0.3671875, + "learning_rate": 0.0015253131747826909, + "loss": 0.1318, + "step": 33617 + }, + { + "epoch": 0.29182038350361544, + "grad_norm": 0.49609375, + "learning_rate": 0.0015252869451043607, + "loss": 0.1162, + "step": 33618 + }, + { + "epoch": 0.2918290639838196, + "grad_norm": 0.216796875, + "learning_rate": 0.0015252607149609434, + "loss": 0.1396, + "step": 33619 + }, + { + "epoch": 0.2918377444640238, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015252344843524675, + "loss": 0.1021, + "step": 33620 + }, + { + "epoch": 0.2918464249442279, + "grad_norm": 0.6484375, + "learning_rate": 0.0015252082532789613, + "loss": 0.0923, + "step": 33621 + }, + { + "epoch": 0.2918551054244321, + "grad_norm": 0.083984375, + "learning_rate": 0.0015251820217404538, + "loss": 0.0879, + "step": 33622 + }, + { + "epoch": 0.29186378590463624, + "grad_norm": 0.228515625, + "learning_rate": 0.0015251557897369736, + "loss": 0.1504, + "step": 33623 + }, + { + "epoch": 0.29187246638484043, + "grad_norm": 0.2734375, + "learning_rate": 0.0015251295572685497, + "loss": 0.0928, + "step": 33624 + }, + { + "epoch": 0.29188114686504457, + "grad_norm": 0.35546875, + "learning_rate": 0.0015251033243352103, + "loss": 0.127, + "step": 33625 + }, + { + "epoch": 0.29188982734524876, + "grad_norm": 0.201171875, + "learning_rate": 0.0015250770909369842, + "loss": 0.1143, + "step": 33626 + }, + { + "epoch": 0.2918985078254529, + "grad_norm": 0.140625, + "learning_rate": 0.0015250508570739, + "loss": 0.1445, + "step": 33627 + }, + { + "epoch": 0.2919071883056571, + "grad_norm": 0.416015625, + "learning_rate": 0.0015250246227459868, + "loss": 0.1221, + "step": 33628 + }, + { + "epoch": 0.29191586878586123, + "grad_norm": 0.287109375, + "learning_rate": 0.001524998387953273, + "loss": 0.123, + "step": 33629 + }, + { + "epoch": 0.2919245492660654, + "grad_norm": 0.671875, + "learning_rate": 0.0015249721526957875, + "loss": 0.1426, + "step": 33630 + }, + { + "epoch": 0.29193322974626956, + "grad_norm": 0.37109375, + "learning_rate": 0.001524945916973559, + "loss": 0.1084, + "step": 33631 + }, + { + "epoch": 0.29194191022647376, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015249196807866157, + "loss": 0.0913, + "step": 33632 + }, + { + "epoch": 0.2919505907066779, + "grad_norm": 1.4765625, + "learning_rate": 0.0015248934441349866, + "loss": 0.1289, + "step": 33633 + }, + { + "epoch": 0.2919592711868821, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015248672070187004, + "loss": 0.0898, + "step": 33634 + }, + { + "epoch": 0.2919679516670862, + "grad_norm": 0.38671875, + "learning_rate": 0.001524840969437786, + "loss": 0.0879, + "step": 33635 + }, + { + "epoch": 0.2919766321472904, + "grad_norm": 0.421875, + "learning_rate": 0.0015248147313922716, + "loss": 0.1084, + "step": 33636 + }, + { + "epoch": 0.29198531262749455, + "grad_norm": 0.3125, + "learning_rate": 0.0015247884928821867, + "loss": 0.0981, + "step": 33637 + }, + { + "epoch": 0.2919939931076987, + "grad_norm": 0.412109375, + "learning_rate": 0.001524762253907559, + "loss": 0.1104, + "step": 33638 + }, + { + "epoch": 0.2920026735879029, + "grad_norm": 0.224609375, + "learning_rate": 0.001524736014468418, + "loss": 0.1025, + "step": 33639 + }, + { + "epoch": 0.292011354068107, + "grad_norm": 0.287109375, + "learning_rate": 0.001524709774564792, + "loss": 0.1787, + "step": 33640 + }, + { + "epoch": 0.2920200345483112, + "grad_norm": 0.291015625, + "learning_rate": 0.0015246835341967099, + "loss": 0.1328, + "step": 33641 + }, + { + "epoch": 0.29202871502851535, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015246572933642, + "loss": 0.0918, + "step": 33642 + }, + { + "epoch": 0.29203739550871954, + "grad_norm": 0.08984375, + "learning_rate": 0.0015246310520672913, + "loss": 0.082, + "step": 33643 + }, + { + "epoch": 0.2920460759889237, + "grad_norm": 0.1337890625, + "learning_rate": 0.0015246048103060125, + "loss": 0.084, + "step": 33644 + }, + { + "epoch": 0.2920547564691279, + "grad_norm": 0.1015625, + "learning_rate": 0.0015245785680803921, + "loss": 0.084, + "step": 33645 + }, + { + "epoch": 0.292063436949332, + "grad_norm": 0.4453125, + "learning_rate": 0.0015245523253904594, + "loss": 0.0854, + "step": 33646 + }, + { + "epoch": 0.2920721174295362, + "grad_norm": 0.482421875, + "learning_rate": 0.0015245260822362421, + "loss": 0.1475, + "step": 33647 + }, + { + "epoch": 0.29208079790974034, + "grad_norm": 0.36328125, + "learning_rate": 0.0015244998386177698, + "loss": 0.0957, + "step": 33648 + }, + { + "epoch": 0.29208947838994453, + "grad_norm": 0.62890625, + "learning_rate": 0.0015244735945350708, + "loss": 0.1104, + "step": 33649 + }, + { + "epoch": 0.29209815887014867, + "grad_norm": 0.265625, + "learning_rate": 0.001524447349988174, + "loss": 0.0825, + "step": 33650 + }, + { + "epoch": 0.29210683935035286, + "grad_norm": 1.359375, + "learning_rate": 0.0015244211049771077, + "loss": 0.1245, + "step": 33651 + }, + { + "epoch": 0.292115519830557, + "grad_norm": 0.1748046875, + "learning_rate": 0.001524394859501901, + "loss": 0.0737, + "step": 33652 + }, + { + "epoch": 0.2921242003107612, + "grad_norm": 0.80078125, + "learning_rate": 0.0015243686135625826, + "loss": 0.166, + "step": 33653 + }, + { + "epoch": 0.29213288079096533, + "grad_norm": 0.08984375, + "learning_rate": 0.001524342367159181, + "loss": 0.1133, + "step": 33654 + }, + { + "epoch": 0.2921415612711695, + "grad_norm": 0.2578125, + "learning_rate": 0.0015243161202917247, + "loss": 0.1318, + "step": 33655 + }, + { + "epoch": 0.29215024175137366, + "grad_norm": 0.203125, + "learning_rate": 0.0015242898729602428, + "loss": 0.123, + "step": 33656 + }, + { + "epoch": 0.29215892223157786, + "grad_norm": 0.1484375, + "learning_rate": 0.001524263625164764, + "loss": 0.0889, + "step": 33657 + }, + { + "epoch": 0.292167602711782, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015242373769053165, + "loss": 0.0986, + "step": 33658 + }, + { + "epoch": 0.2921762831919862, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015242111281819298, + "loss": 0.0933, + "step": 33659 + }, + { + "epoch": 0.2921849636721903, + "grad_norm": 0.154296875, + "learning_rate": 0.0015241848789946317, + "loss": 0.0918, + "step": 33660 + }, + { + "epoch": 0.2921936441523945, + "grad_norm": 0.255859375, + "learning_rate": 0.001524158629343452, + "loss": 0.1133, + "step": 33661 + }, + { + "epoch": 0.29220232463259865, + "grad_norm": 0.38671875, + "learning_rate": 0.0015241323792284185, + "loss": 0.1035, + "step": 33662 + }, + { + "epoch": 0.29221100511280285, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015241061286495603, + "loss": 0.0928, + "step": 33663 + }, + { + "epoch": 0.292219685593007, + "grad_norm": 0.66015625, + "learning_rate": 0.0015240798776069058, + "loss": 0.1348, + "step": 33664 + }, + { + "epoch": 0.2922283660732112, + "grad_norm": 0.2451171875, + "learning_rate": 0.001524053626100484, + "loss": 0.1211, + "step": 33665 + }, + { + "epoch": 0.2922370465534153, + "grad_norm": 0.29296875, + "learning_rate": 0.0015240273741303236, + "loss": 0.1074, + "step": 33666 + }, + { + "epoch": 0.2922457270336195, + "grad_norm": 0.5625, + "learning_rate": 0.0015240011216964534, + "loss": 0.1533, + "step": 33667 + }, + { + "epoch": 0.29225440751382364, + "grad_norm": 0.1552734375, + "learning_rate": 0.0015239748687989017, + "loss": 0.1094, + "step": 33668 + }, + { + "epoch": 0.29226308799402784, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015239486154376971, + "loss": 0.0991, + "step": 33669 + }, + { + "epoch": 0.292271768474232, + "grad_norm": 0.134765625, + "learning_rate": 0.0015239223616128692, + "loss": 0.1045, + "step": 33670 + }, + { + "epoch": 0.29228044895443617, + "grad_norm": 0.083984375, + "learning_rate": 0.0015238961073244463, + "loss": 0.0859, + "step": 33671 + }, + { + "epoch": 0.2922891294346403, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015238698525724568, + "loss": 0.1006, + "step": 33672 + }, + { + "epoch": 0.2922978099148445, + "grad_norm": 0.87890625, + "learning_rate": 0.0015238435973569293, + "loss": 0.1367, + "step": 33673 + }, + { + "epoch": 0.29230649039504863, + "grad_norm": 0.171875, + "learning_rate": 0.001523817341677893, + "loss": 0.1152, + "step": 33674 + }, + { + "epoch": 0.2923151708752528, + "grad_norm": 0.265625, + "learning_rate": 0.0015237910855353763, + "loss": 0.1069, + "step": 33675 + }, + { + "epoch": 0.29232385135545697, + "grad_norm": 0.53125, + "learning_rate": 0.0015237648289294083, + "loss": 0.1025, + "step": 33676 + }, + { + "epoch": 0.29233253183566116, + "grad_norm": 0.4296875, + "learning_rate": 0.0015237385718600172, + "loss": 0.0933, + "step": 33677 + }, + { + "epoch": 0.2923412123158653, + "grad_norm": 0.359375, + "learning_rate": 0.0015237123143272322, + "loss": 0.1104, + "step": 33678 + }, + { + "epoch": 0.2923498927960695, + "grad_norm": 0.06884765625, + "learning_rate": 0.0015236860563310817, + "loss": 0.1064, + "step": 33679 + }, + { + "epoch": 0.2923585732762736, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015236597978715947, + "loss": 0.1099, + "step": 33680 + }, + { + "epoch": 0.2923672537564778, + "grad_norm": 0.2109375, + "learning_rate": 0.0015236335389487996, + "loss": 0.1143, + "step": 33681 + }, + { + "epoch": 0.29237593423668196, + "grad_norm": 0.271484375, + "learning_rate": 0.001523607279562725, + "loss": 0.084, + "step": 33682 + }, + { + "epoch": 0.29238461471688615, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015235810197134, + "loss": 0.1094, + "step": 33683 + }, + { + "epoch": 0.2923932951970903, + "grad_norm": 0.11669921875, + "learning_rate": 0.0015235547594008528, + "loss": 0.0938, + "step": 33684 + }, + { + "epoch": 0.2924019756772945, + "grad_norm": 0.2392578125, + "learning_rate": 0.0015235284986251127, + "loss": 0.0801, + "step": 33685 + }, + { + "epoch": 0.2924106561574986, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015235022373862085, + "loss": 0.1221, + "step": 33686 + }, + { + "epoch": 0.2924193366377028, + "grad_norm": 0.34375, + "learning_rate": 0.0015234759756841683, + "loss": 0.1377, + "step": 33687 + }, + { + "epoch": 0.29242801711790695, + "grad_norm": 0.283203125, + "learning_rate": 0.0015234497135190215, + "loss": 0.1064, + "step": 33688 + }, + { + "epoch": 0.29243669759811114, + "grad_norm": 0.1796875, + "learning_rate": 0.001523423450890796, + "loss": 0.1318, + "step": 33689 + }, + { + "epoch": 0.2924453780783153, + "grad_norm": 0.11083984375, + "learning_rate": 0.0015233971877995213, + "loss": 0.1484, + "step": 33690 + }, + { + "epoch": 0.29245405855851947, + "grad_norm": 0.17578125, + "learning_rate": 0.0015233709242452257, + "loss": 0.1279, + "step": 33691 + }, + { + "epoch": 0.2924627390387236, + "grad_norm": 0.318359375, + "learning_rate": 0.001523344660227938, + "loss": 0.1094, + "step": 33692 + }, + { + "epoch": 0.2924714195189278, + "grad_norm": 0.119140625, + "learning_rate": 0.0015233183957476869, + "loss": 0.1221, + "step": 33693 + }, + { + "epoch": 0.29248009999913194, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015232921308045012, + "loss": 0.1016, + "step": 33694 + }, + { + "epoch": 0.29248878047933613, + "grad_norm": 0.2109375, + "learning_rate": 0.0015232658653984094, + "loss": 0.0928, + "step": 33695 + }, + { + "epoch": 0.29249746095954027, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015232395995294406, + "loss": 0.1064, + "step": 33696 + }, + { + "epoch": 0.29250614143974446, + "grad_norm": 0.16796875, + "learning_rate": 0.0015232133331976234, + "loss": 0.0889, + "step": 33697 + }, + { + "epoch": 0.2925148219199486, + "grad_norm": 0.55078125, + "learning_rate": 0.0015231870664029865, + "loss": 0.1191, + "step": 33698 + }, + { + "epoch": 0.2925235024001528, + "grad_norm": 0.6640625, + "learning_rate": 0.0015231607991455583, + "loss": 0.0903, + "step": 33699 + }, + { + "epoch": 0.29253218288035693, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015231345314253681, + "loss": 0.0762, + "step": 33700 + }, + { + "epoch": 0.2925408633605611, + "grad_norm": 0.154296875, + "learning_rate": 0.0015231082632424443, + "loss": 0.1719, + "step": 33701 + }, + { + "epoch": 0.29254954384076526, + "grad_norm": 0.53125, + "learning_rate": 0.0015230819945968156, + "loss": 0.1035, + "step": 33702 + }, + { + "epoch": 0.29255822432096945, + "grad_norm": 0.85546875, + "learning_rate": 0.0015230557254885108, + "loss": 0.4023, + "step": 33703 + }, + { + "epoch": 0.2925669048011736, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015230294559175584, + "loss": 0.082, + "step": 33704 + }, + { + "epoch": 0.2925755852813778, + "grad_norm": 0.37890625, + "learning_rate": 0.0015230031858839877, + "loss": 0.1494, + "step": 33705 + }, + { + "epoch": 0.2925842657615819, + "grad_norm": 0.5234375, + "learning_rate": 0.0015229769153878268, + "loss": 0.0869, + "step": 33706 + }, + { + "epoch": 0.2925929462417861, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015229506444291046, + "loss": 0.1045, + "step": 33707 + }, + { + "epoch": 0.29260162672199025, + "grad_norm": 0.5078125, + "learning_rate": 0.0015229243730078505, + "loss": 0.0679, + "step": 33708 + }, + { + "epoch": 0.29261030720219444, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015228981011240921, + "loss": 0.1001, + "step": 33709 + }, + { + "epoch": 0.2926189876823986, + "grad_norm": 0.56640625, + "learning_rate": 0.001522871828777859, + "loss": 0.0806, + "step": 33710 + }, + { + "epoch": 0.29262766816260277, + "grad_norm": 1.0234375, + "learning_rate": 0.0015228455559691794, + "loss": 0.1128, + "step": 33711 + }, + { + "epoch": 0.2926363486428069, + "grad_norm": 0.08984375, + "learning_rate": 0.0015228192826980823, + "loss": 0.103, + "step": 33712 + }, + { + "epoch": 0.2926450291230111, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015227930089645964, + "loss": 0.0918, + "step": 33713 + }, + { + "epoch": 0.29265370960321524, + "grad_norm": 0.77734375, + "learning_rate": 0.0015227667347687505, + "loss": 0.1328, + "step": 33714 + }, + { + "epoch": 0.29266239008341943, + "grad_norm": 0.408203125, + "learning_rate": 0.0015227404601105732, + "loss": 0.123, + "step": 33715 + }, + { + "epoch": 0.29267107056362357, + "grad_norm": 0.1357421875, + "learning_rate": 0.0015227141849900932, + "loss": 0.1162, + "step": 33716 + }, + { + "epoch": 0.29267975104382776, + "grad_norm": 0.09228515625, + "learning_rate": 0.0015226879094073394, + "loss": 0.0762, + "step": 33717 + }, + { + "epoch": 0.2926884315240319, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015226616333623405, + "loss": 0.0894, + "step": 33718 + }, + { + "epoch": 0.2926971120042361, + "grad_norm": 0.181640625, + "learning_rate": 0.001522635356855125, + "loss": 0.0791, + "step": 33719 + }, + { + "epoch": 0.29270579248444023, + "grad_norm": 0.36328125, + "learning_rate": 0.0015226090798857222, + "loss": 0.085, + "step": 33720 + }, + { + "epoch": 0.2927144729646444, + "grad_norm": 0.08203125, + "learning_rate": 0.00152258280245416, + "loss": 0.1299, + "step": 33721 + }, + { + "epoch": 0.29272315344484856, + "grad_norm": 0.171875, + "learning_rate": 0.001522556524560468, + "loss": 0.0918, + "step": 33722 + }, + { + "epoch": 0.29273183392505275, + "grad_norm": 0.1025390625, + "learning_rate": 0.001522530246204674, + "loss": 0.0947, + "step": 33723 + }, + { + "epoch": 0.2927405144052569, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015225039673868077, + "loss": 0.1484, + "step": 33724 + }, + { + "epoch": 0.2927491948854611, + "grad_norm": 0.2080078125, + "learning_rate": 0.0015224776881068974, + "loss": 0.1123, + "step": 33725 + }, + { + "epoch": 0.2927578753656652, + "grad_norm": 0.31640625, + "learning_rate": 0.0015224514083649718, + "loss": 0.1172, + "step": 33726 + }, + { + "epoch": 0.2927665558458694, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015224251281610597, + "loss": 0.0806, + "step": 33727 + }, + { + "epoch": 0.29277523632607355, + "grad_norm": 0.1533203125, + "learning_rate": 0.00152239884749519, + "loss": 0.0938, + "step": 33728 + }, + { + "epoch": 0.29278391680627774, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015223725663673908, + "loss": 0.0923, + "step": 33729 + }, + { + "epoch": 0.2927925972864819, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015223462847776916, + "loss": 0.1211, + "step": 33730 + }, + { + "epoch": 0.2928012777666861, + "grad_norm": 0.138671875, + "learning_rate": 0.0015223200027261207, + "loss": 0.0967, + "step": 33731 + }, + { + "epoch": 0.2928099582468902, + "grad_norm": 0.57421875, + "learning_rate": 0.0015222937202127073, + "loss": 0.127, + "step": 33732 + }, + { + "epoch": 0.2928186387270944, + "grad_norm": 0.30078125, + "learning_rate": 0.0015222674372374794, + "loss": 0.1387, + "step": 33733 + }, + { + "epoch": 0.29282731920729854, + "grad_norm": 0.15234375, + "learning_rate": 0.0015222411538004666, + "loss": 0.0898, + "step": 33734 + }, + { + "epoch": 0.29283599968750273, + "grad_norm": 0.1240234375, + "learning_rate": 0.001522214869901697, + "loss": 0.1084, + "step": 33735 + }, + { + "epoch": 0.2928446801677069, + "grad_norm": 0.2314453125, + "learning_rate": 0.0015221885855411996, + "loss": 0.125, + "step": 33736 + }, + { + "epoch": 0.29285336064791107, + "grad_norm": 0.263671875, + "learning_rate": 0.0015221623007190034, + "loss": 0.1338, + "step": 33737 + }, + { + "epoch": 0.2928620411281152, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015221360154351363, + "loss": 0.0918, + "step": 33738 + }, + { + "epoch": 0.2928707216083194, + "grad_norm": 0.380859375, + "learning_rate": 0.0015221097296896277, + "loss": 0.0791, + "step": 33739 + }, + { + "epoch": 0.29287940208852353, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015220834434825066, + "loss": 0.1328, + "step": 33740 + }, + { + "epoch": 0.2928880825687277, + "grad_norm": 0.7578125, + "learning_rate": 0.0015220571568138011, + "loss": 0.1445, + "step": 33741 + }, + { + "epoch": 0.29289676304893186, + "grad_norm": 0.1123046875, + "learning_rate": 0.0015220308696835404, + "loss": 0.1133, + "step": 33742 + }, + { + "epoch": 0.29290544352913606, + "grad_norm": 0.7890625, + "learning_rate": 0.001522004582091753, + "loss": 0.1055, + "step": 33743 + }, + { + "epoch": 0.2929141240093402, + "grad_norm": 0.224609375, + "learning_rate": 0.0015219782940384676, + "loss": 0.1367, + "step": 33744 + }, + { + "epoch": 0.2929228044895444, + "grad_norm": 0.251953125, + "learning_rate": 0.0015219520055237132, + "loss": 0.1074, + "step": 33745 + }, + { + "epoch": 0.2929314849697485, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015219257165475185, + "loss": 0.0791, + "step": 33746 + }, + { + "epoch": 0.2929401654499527, + "grad_norm": 0.625, + "learning_rate": 0.001521899427109912, + "loss": 0.1328, + "step": 33747 + }, + { + "epoch": 0.29294884593015685, + "grad_norm": 0.421875, + "learning_rate": 0.001521873137210923, + "loss": 0.0996, + "step": 33748 + }, + { + "epoch": 0.29295752641036105, + "grad_norm": 0.2177734375, + "learning_rate": 0.0015218468468505796, + "loss": 0.126, + "step": 33749 + }, + { + "epoch": 0.2929662068905652, + "grad_norm": 0.1953125, + "learning_rate": 0.0015218205560289107, + "loss": 0.0859, + "step": 33750 + }, + { + "epoch": 0.2929748873707694, + "grad_norm": 0.32421875, + "learning_rate": 0.0015217942647459456, + "loss": 0.0933, + "step": 33751 + }, + { + "epoch": 0.2929835678509735, + "grad_norm": 0.703125, + "learning_rate": 0.001521767973001712, + "loss": 0.123, + "step": 33752 + }, + { + "epoch": 0.2929922483311777, + "grad_norm": 0.58203125, + "learning_rate": 0.0015217416807962395, + "loss": 0.0986, + "step": 33753 + }, + { + "epoch": 0.29300092881138184, + "grad_norm": 0.1845703125, + "learning_rate": 0.001521715388129557, + "loss": 0.0781, + "step": 33754 + }, + { + "epoch": 0.29300960929158604, + "grad_norm": 0.357421875, + "learning_rate": 0.0015216890950016926, + "loss": 0.1025, + "step": 33755 + }, + { + "epoch": 0.2930182897717902, + "grad_norm": 0.11083984375, + "learning_rate": 0.0015216628014126752, + "loss": 0.0928, + "step": 33756 + }, + { + "epoch": 0.29302697025199437, + "grad_norm": 0.65234375, + "learning_rate": 0.001521636507362534, + "loss": 0.1089, + "step": 33757 + }, + { + "epoch": 0.2930356507321985, + "grad_norm": 0.359375, + "learning_rate": 0.0015216102128512974, + "loss": 0.0884, + "step": 33758 + }, + { + "epoch": 0.2930443312124027, + "grad_norm": 0.357421875, + "learning_rate": 0.0015215839178789941, + "loss": 0.1055, + "step": 33759 + }, + { + "epoch": 0.29305301169260684, + "grad_norm": 0.1328125, + "learning_rate": 0.001521557622445653, + "loss": 0.0918, + "step": 33760 + }, + { + "epoch": 0.293061692172811, + "grad_norm": 0.37890625, + "learning_rate": 0.0015215313265513027, + "loss": 0.1152, + "step": 33761 + }, + { + "epoch": 0.29307037265301517, + "grad_norm": 0.419921875, + "learning_rate": 0.0015215050301959723, + "loss": 0.0806, + "step": 33762 + }, + { + "epoch": 0.2930790531332193, + "grad_norm": 0.58203125, + "learning_rate": 0.0015214787333796905, + "loss": 0.0698, + "step": 33763 + }, + { + "epoch": 0.2930877336134235, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015214524361024852, + "loss": 0.0962, + "step": 33764 + }, + { + "epoch": 0.29309641409362763, + "grad_norm": 0.11572265625, + "learning_rate": 0.0015214261383643863, + "loss": 0.1084, + "step": 33765 + }, + { + "epoch": 0.2931050945738318, + "grad_norm": 0.515625, + "learning_rate": 0.0015213998401654221, + "loss": 0.0879, + "step": 33766 + }, + { + "epoch": 0.29311377505403596, + "grad_norm": 0.142578125, + "learning_rate": 0.0015213735415056215, + "loss": 0.1016, + "step": 33767 + }, + { + "epoch": 0.29312245553424016, + "grad_norm": 0.48828125, + "learning_rate": 0.0015213472423850128, + "loss": 0.126, + "step": 33768 + }, + { + "epoch": 0.2931311360144443, + "grad_norm": 1.3515625, + "learning_rate": 0.0015213209428036254, + "loss": 0.1216, + "step": 33769 + }, + { + "epoch": 0.2931398164946485, + "grad_norm": 0.73828125, + "learning_rate": 0.0015212946427614876, + "loss": 0.1543, + "step": 33770 + }, + { + "epoch": 0.2931484969748526, + "grad_norm": 0.3515625, + "learning_rate": 0.0015212683422586282, + "loss": 0.0889, + "step": 33771 + }, + { + "epoch": 0.2931571774550568, + "grad_norm": 0.9296875, + "learning_rate": 0.001521242041295076, + "loss": 0.1387, + "step": 33772 + }, + { + "epoch": 0.29316585793526095, + "grad_norm": 0.130859375, + "learning_rate": 0.0015212157398708603, + "loss": 0.1069, + "step": 33773 + }, + { + "epoch": 0.29317453841546515, + "grad_norm": 0.21484375, + "learning_rate": 0.001521189437986009, + "loss": 0.0913, + "step": 33774 + }, + { + "epoch": 0.2931832188956693, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015211631356405514, + "loss": 0.1328, + "step": 33775 + }, + { + "epoch": 0.2931918993758735, + "grad_norm": 0.2265625, + "learning_rate": 0.0015211368328345162, + "loss": 0.1045, + "step": 33776 + }, + { + "epoch": 0.2932005798560776, + "grad_norm": 0.2412109375, + "learning_rate": 0.001521110529567932, + "loss": 0.167, + "step": 33777 + }, + { + "epoch": 0.2932092603362818, + "grad_norm": 0.423828125, + "learning_rate": 0.0015210842258408276, + "loss": 0.1069, + "step": 33778 + }, + { + "epoch": 0.29321794081648594, + "grad_norm": 0.220703125, + "learning_rate": 0.001521057921653232, + "loss": 0.1543, + "step": 33779 + }, + { + "epoch": 0.29322662129669014, + "grad_norm": 0.337890625, + "learning_rate": 0.0015210316170051735, + "loss": 0.1406, + "step": 33780 + }, + { + "epoch": 0.2932353017768943, + "grad_norm": 0.51171875, + "learning_rate": 0.001521005311896681, + "loss": 0.1211, + "step": 33781 + }, + { + "epoch": 0.29324398225709847, + "grad_norm": 0.291015625, + "learning_rate": 0.0015209790063277838, + "loss": 0.1152, + "step": 33782 + }, + { + "epoch": 0.2932526627373026, + "grad_norm": 0.177734375, + "learning_rate": 0.0015209527002985103, + "loss": 0.1016, + "step": 33783 + }, + { + "epoch": 0.2932613432175068, + "grad_norm": 0.5078125, + "learning_rate": 0.001520926393808889, + "loss": 0.1592, + "step": 33784 + }, + { + "epoch": 0.29327002369771094, + "grad_norm": 0.1904296875, + "learning_rate": 0.001520900086858949, + "loss": 0.104, + "step": 33785 + }, + { + "epoch": 0.29327870417791513, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015208737794487192, + "loss": 0.1357, + "step": 33786 + }, + { + "epoch": 0.29328738465811927, + "grad_norm": 0.30859375, + "learning_rate": 0.001520847471578228, + "loss": 0.1191, + "step": 33787 + }, + { + "epoch": 0.29329606513832346, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015208211632475042, + "loss": 0.1143, + "step": 33788 + }, + { + "epoch": 0.2933047456185276, + "grad_norm": 0.11962890625, + "learning_rate": 0.001520794854456577, + "loss": 0.1357, + "step": 33789 + }, + { + "epoch": 0.2933134260987318, + "grad_norm": 0.17578125, + "learning_rate": 0.0015207685452054743, + "loss": 0.1182, + "step": 33790 + }, + { + "epoch": 0.2933221065789359, + "grad_norm": 0.201171875, + "learning_rate": 0.001520742235494226, + "loss": 0.1182, + "step": 33791 + }, + { + "epoch": 0.2933307870591401, + "grad_norm": 0.3046875, + "learning_rate": 0.0015207159253228602, + "loss": 0.1074, + "step": 33792 + }, + { + "epoch": 0.29333946753934426, + "grad_norm": 0.18359375, + "learning_rate": 0.0015206896146914058, + "loss": 0.1162, + "step": 33793 + }, + { + "epoch": 0.29334814801954845, + "grad_norm": 0.296875, + "learning_rate": 0.0015206633035998913, + "loss": 0.1108, + "step": 33794 + }, + { + "epoch": 0.2933568284997526, + "grad_norm": 0.1875, + "learning_rate": 0.0015206369920483464, + "loss": 0.1201, + "step": 33795 + }, + { + "epoch": 0.2933655089799568, + "grad_norm": 0.59375, + "learning_rate": 0.0015206106800367985, + "loss": 0.1094, + "step": 33796 + }, + { + "epoch": 0.2933741894601609, + "grad_norm": 0.7421875, + "learning_rate": 0.0015205843675652772, + "loss": 0.0928, + "step": 33797 + }, + { + "epoch": 0.2933828699403651, + "grad_norm": 0.154296875, + "learning_rate": 0.0015205580546338113, + "loss": 0.1426, + "step": 33798 + }, + { + "epoch": 0.29339155042056925, + "grad_norm": 0.16015625, + "learning_rate": 0.0015205317412424295, + "loss": 0.0977, + "step": 33799 + }, + { + "epoch": 0.29340023090077344, + "grad_norm": 0.50390625, + "learning_rate": 0.0015205054273911603, + "loss": 0.1035, + "step": 33800 + }, + { + "epoch": 0.2934089113809776, + "grad_norm": 0.12060546875, + "learning_rate": 0.0015204791130800327, + "loss": 0.0698, + "step": 33801 + }, + { + "epoch": 0.29341759186118177, + "grad_norm": 0.26953125, + "learning_rate": 0.0015204527983090754, + "loss": 0.0854, + "step": 33802 + }, + { + "epoch": 0.2934262723413859, + "grad_norm": 0.283203125, + "learning_rate": 0.0015204264830783174, + "loss": 0.0947, + "step": 33803 + }, + { + "epoch": 0.2934349528215901, + "grad_norm": 0.134765625, + "learning_rate": 0.0015204001673877873, + "loss": 0.1074, + "step": 33804 + }, + { + "epoch": 0.29344363330179424, + "grad_norm": 0.1533203125, + "learning_rate": 0.001520373851237514, + "loss": 0.1084, + "step": 33805 + }, + { + "epoch": 0.29345231378199843, + "grad_norm": 0.4453125, + "learning_rate": 0.001520347534627526, + "loss": 0.0859, + "step": 33806 + }, + { + "epoch": 0.29346099426220257, + "grad_norm": 0.265625, + "learning_rate": 0.0015203212175578524, + "loss": 0.0879, + "step": 33807 + }, + { + "epoch": 0.29346967474240676, + "grad_norm": 0.2333984375, + "learning_rate": 0.0015202949000285217, + "loss": 0.1445, + "step": 33808 + }, + { + "epoch": 0.2934783552226109, + "grad_norm": 0.2431640625, + "learning_rate": 0.0015202685820395628, + "loss": 0.0986, + "step": 33809 + }, + { + "epoch": 0.2934870357028151, + "grad_norm": 0.4296875, + "learning_rate": 0.0015202422635910047, + "loss": 0.1196, + "step": 33810 + }, + { + "epoch": 0.29349571618301923, + "grad_norm": 0.232421875, + "learning_rate": 0.0015202159446828757, + "loss": 0.0728, + "step": 33811 + }, + { + "epoch": 0.2935043966632234, + "grad_norm": 0.421875, + "learning_rate": 0.001520189625315205, + "loss": 0.082, + "step": 33812 + }, + { + "epoch": 0.29351307714342756, + "grad_norm": 0.169921875, + "learning_rate": 0.0015201633054880213, + "loss": 0.1221, + "step": 33813 + }, + { + "epoch": 0.29352175762363175, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015201369852013533, + "loss": 0.1055, + "step": 33814 + }, + { + "epoch": 0.2935304381038359, + "grad_norm": 0.126953125, + "learning_rate": 0.0015201106644552296, + "loss": 0.1011, + "step": 33815 + }, + { + "epoch": 0.2935391185840401, + "grad_norm": 1.3125, + "learning_rate": 0.0015200843432496794, + "loss": 0.106, + "step": 33816 + }, + { + "epoch": 0.2935477990642442, + "grad_norm": 0.16796875, + "learning_rate": 0.0015200580215847313, + "loss": 0.0957, + "step": 33817 + }, + { + "epoch": 0.2935564795444484, + "grad_norm": 0.0888671875, + "learning_rate": 0.0015200316994604137, + "loss": 0.0669, + "step": 33818 + }, + { + "epoch": 0.29356516002465255, + "grad_norm": 1.078125, + "learning_rate": 0.0015200053768767562, + "loss": 0.0781, + "step": 33819 + }, + { + "epoch": 0.29357384050485674, + "grad_norm": 0.072265625, + "learning_rate": 0.0015199790538337867, + "loss": 0.0771, + "step": 33820 + }, + { + "epoch": 0.2935825209850609, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015199527303315348, + "loss": 0.1064, + "step": 33821 + }, + { + "epoch": 0.2935912014652651, + "grad_norm": 0.76953125, + "learning_rate": 0.0015199264063700288, + "loss": 0.1299, + "step": 33822 + }, + { + "epoch": 0.2935998819454692, + "grad_norm": 0.8515625, + "learning_rate": 0.0015199000819492976, + "loss": 0.1064, + "step": 33823 + }, + { + "epoch": 0.2936085624256734, + "grad_norm": 0.07763671875, + "learning_rate": 0.00151987375706937, + "loss": 0.0908, + "step": 33824 + }, + { + "epoch": 0.29361724290587754, + "grad_norm": 0.154296875, + "learning_rate": 0.0015198474317302745, + "loss": 0.1182, + "step": 33825 + }, + { + "epoch": 0.29362592338608173, + "grad_norm": 0.138671875, + "learning_rate": 0.0015198211059320407, + "loss": 0.0986, + "step": 33826 + }, + { + "epoch": 0.29363460386628587, + "grad_norm": 0.11474609375, + "learning_rate": 0.0015197947796746964, + "loss": 0.126, + "step": 33827 + }, + { + "epoch": 0.29364328434649006, + "grad_norm": 0.60546875, + "learning_rate": 0.0015197684529582706, + "loss": 0.0713, + "step": 33828 + }, + { + "epoch": 0.2936519648266942, + "grad_norm": 0.40625, + "learning_rate": 0.0015197421257827927, + "loss": 0.0913, + "step": 33829 + }, + { + "epoch": 0.2936606453068984, + "grad_norm": 0.216796875, + "learning_rate": 0.0015197157981482914, + "loss": 0.1162, + "step": 33830 + }, + { + "epoch": 0.29366932578710253, + "grad_norm": 0.12890625, + "learning_rate": 0.0015196894700547948, + "loss": 0.1426, + "step": 33831 + }, + { + "epoch": 0.2936780062673067, + "grad_norm": 0.498046875, + "learning_rate": 0.001519663141502332, + "loss": 0.1069, + "step": 33832 + }, + { + "epoch": 0.29368668674751086, + "grad_norm": 1.0625, + "learning_rate": 0.0015196368124909324, + "loss": 0.123, + "step": 33833 + }, + { + "epoch": 0.29369536722771505, + "grad_norm": 0.31640625, + "learning_rate": 0.001519610483020624, + "loss": 0.083, + "step": 33834 + }, + { + "epoch": 0.2937040477079192, + "grad_norm": 0.38671875, + "learning_rate": 0.0015195841530914358, + "loss": 0.123, + "step": 33835 + }, + { + "epoch": 0.2937127281881234, + "grad_norm": 0.294921875, + "learning_rate": 0.0015195578227033967, + "loss": 0.0698, + "step": 33836 + }, + { + "epoch": 0.2937214086683275, + "grad_norm": 0.236328125, + "learning_rate": 0.0015195314918565353, + "loss": 0.1011, + "step": 33837 + }, + { + "epoch": 0.2937300891485317, + "grad_norm": 0.515625, + "learning_rate": 0.0015195051605508807, + "loss": 0.0898, + "step": 33838 + }, + { + "epoch": 0.29373876962873585, + "grad_norm": 0.55078125, + "learning_rate": 0.0015194788287864619, + "loss": 0.1816, + "step": 33839 + }, + { + "epoch": 0.29374745010894004, + "grad_norm": 0.271484375, + "learning_rate": 0.0015194524965633069, + "loss": 0.0928, + "step": 33840 + }, + { + "epoch": 0.2937561305891442, + "grad_norm": 0.578125, + "learning_rate": 0.0015194261638814452, + "loss": 0.1006, + "step": 33841 + }, + { + "epoch": 0.2937648110693484, + "grad_norm": 0.236328125, + "learning_rate": 0.0015193998307409052, + "loss": 0.0864, + "step": 33842 + }, + { + "epoch": 0.2937734915495525, + "grad_norm": 0.21484375, + "learning_rate": 0.001519373497141716, + "loss": 0.1465, + "step": 33843 + }, + { + "epoch": 0.2937821720297567, + "grad_norm": 0.59375, + "learning_rate": 0.0015193471630839063, + "loss": 0.1084, + "step": 33844 + }, + { + "epoch": 0.29379085250996084, + "grad_norm": 0.54296875, + "learning_rate": 0.0015193208285675045, + "loss": 0.1543, + "step": 33845 + }, + { + "epoch": 0.29379953299016504, + "grad_norm": 0.5625, + "learning_rate": 0.00151929449359254, + "loss": 0.125, + "step": 33846 + }, + { + "epoch": 0.2938082134703692, + "grad_norm": 0.578125, + "learning_rate": 0.0015192681581590413, + "loss": 0.1348, + "step": 33847 + }, + { + "epoch": 0.29381689395057337, + "grad_norm": 0.33984375, + "learning_rate": 0.0015192418222670372, + "loss": 0.0854, + "step": 33848 + }, + { + "epoch": 0.2938255744307775, + "grad_norm": 0.66015625, + "learning_rate": 0.0015192154859165567, + "loss": 0.1162, + "step": 33849 + }, + { + "epoch": 0.2938342549109817, + "grad_norm": 0.369140625, + "learning_rate": 0.0015191891491076282, + "loss": 0.106, + "step": 33850 + }, + { + "epoch": 0.29384293539118583, + "grad_norm": 0.671875, + "learning_rate": 0.0015191628118402809, + "loss": 0.1309, + "step": 33851 + }, + { + "epoch": 0.29385161587139, + "grad_norm": 0.25390625, + "learning_rate": 0.0015191364741145436, + "loss": 0.0986, + "step": 33852 + }, + { + "epoch": 0.29386029635159416, + "grad_norm": 0.29296875, + "learning_rate": 0.0015191101359304448, + "loss": 0.126, + "step": 33853 + }, + { + "epoch": 0.29386897683179836, + "grad_norm": 0.1201171875, + "learning_rate": 0.0015190837972880135, + "loss": 0.1035, + "step": 33854 + }, + { + "epoch": 0.2938776573120025, + "grad_norm": 0.359375, + "learning_rate": 0.001519057458187278, + "loss": 0.1216, + "step": 33855 + }, + { + "epoch": 0.2938863377922067, + "grad_norm": 0.365234375, + "learning_rate": 0.0015190311186282681, + "loss": 0.0815, + "step": 33856 + }, + { + "epoch": 0.2938950182724108, + "grad_norm": 0.296875, + "learning_rate": 0.0015190047786110118, + "loss": 0.1211, + "step": 33857 + }, + { + "epoch": 0.293903698752615, + "grad_norm": 0.3671875, + "learning_rate": 0.0015189784381355383, + "loss": 0.1191, + "step": 33858 + }, + { + "epoch": 0.29391237923281915, + "grad_norm": 0.51953125, + "learning_rate": 0.0015189520972018762, + "loss": 0.0845, + "step": 33859 + }, + { + "epoch": 0.29392105971302335, + "grad_norm": 0.259765625, + "learning_rate": 0.0015189257558100545, + "loss": 0.1016, + "step": 33860 + }, + { + "epoch": 0.2939297401932275, + "grad_norm": 1.4375, + "learning_rate": 0.001518899413960102, + "loss": 0.1533, + "step": 33861 + }, + { + "epoch": 0.2939384206734317, + "grad_norm": 0.1923828125, + "learning_rate": 0.0015188730716520472, + "loss": 0.1201, + "step": 33862 + }, + { + "epoch": 0.2939471011536358, + "grad_norm": 0.298828125, + "learning_rate": 0.001518846728885919, + "loss": 0.1162, + "step": 33863 + }, + { + "epoch": 0.29395578163384, + "grad_norm": 0.10693359375, + "learning_rate": 0.0015188203856617464, + "loss": 0.1064, + "step": 33864 + }, + { + "epoch": 0.29396446211404414, + "grad_norm": 0.1953125, + "learning_rate": 0.001518794041979558, + "loss": 0.1689, + "step": 33865 + }, + { + "epoch": 0.29397314259424834, + "grad_norm": 0.1689453125, + "learning_rate": 0.001518767697839383, + "loss": 0.0835, + "step": 33866 + }, + { + "epoch": 0.2939818230744525, + "grad_norm": 0.349609375, + "learning_rate": 0.0015187413532412497, + "loss": 0.0854, + "step": 33867 + }, + { + "epoch": 0.29399050355465667, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015187150081851872, + "loss": 0.1221, + "step": 33868 + }, + { + "epoch": 0.2939991840348608, + "grad_norm": 0.201171875, + "learning_rate": 0.0015186886626712243, + "loss": 0.1035, + "step": 33869 + }, + { + "epoch": 0.294007864515065, + "grad_norm": 0.3046875, + "learning_rate": 0.0015186623166993899, + "loss": 0.1123, + "step": 33870 + }, + { + "epoch": 0.29401654499526914, + "grad_norm": 0.078125, + "learning_rate": 0.0015186359702697126, + "loss": 0.0811, + "step": 33871 + }, + { + "epoch": 0.29402522547547333, + "grad_norm": 0.65625, + "learning_rate": 0.0015186096233822213, + "loss": 0.1221, + "step": 33872 + }, + { + "epoch": 0.29403390595567747, + "grad_norm": 0.330078125, + "learning_rate": 0.0015185832760369446, + "loss": 0.0684, + "step": 33873 + }, + { + "epoch": 0.29404258643588166, + "grad_norm": 0.10498046875, + "learning_rate": 0.0015185569282339117, + "loss": 0.0996, + "step": 33874 + }, + { + "epoch": 0.2940512669160858, + "grad_norm": 0.263671875, + "learning_rate": 0.0015185305799731512, + "loss": 0.1167, + "step": 33875 + }, + { + "epoch": 0.29405994739629, + "grad_norm": 0.0869140625, + "learning_rate": 0.0015185042312546922, + "loss": 0.0845, + "step": 33876 + }, + { + "epoch": 0.2940686278764941, + "grad_norm": 0.39453125, + "learning_rate": 0.0015184778820785628, + "loss": 0.0903, + "step": 33877 + }, + { + "epoch": 0.2940773083566983, + "grad_norm": 0.6171875, + "learning_rate": 0.0015184515324447928, + "loss": 0.0947, + "step": 33878 + }, + { + "epoch": 0.29408598883690246, + "grad_norm": 0.234375, + "learning_rate": 0.0015184251823534098, + "loss": 0.1309, + "step": 33879 + }, + { + "epoch": 0.29409466931710665, + "grad_norm": 0.58203125, + "learning_rate": 0.0015183988318044437, + "loss": 0.1113, + "step": 33880 + }, + { + "epoch": 0.2941033497973108, + "grad_norm": 0.5234375, + "learning_rate": 0.001518372480797923, + "loss": 0.1143, + "step": 33881 + }, + { + "epoch": 0.294112030277515, + "grad_norm": 0.263671875, + "learning_rate": 0.0015183461293338766, + "loss": 0.1377, + "step": 33882 + }, + { + "epoch": 0.2941207107577191, + "grad_norm": 0.46875, + "learning_rate": 0.0015183197774123328, + "loss": 0.123, + "step": 33883 + }, + { + "epoch": 0.29412939123792325, + "grad_norm": 0.12451171875, + "learning_rate": 0.0015182934250333208, + "loss": 0.1118, + "step": 33884 + }, + { + "epoch": 0.29413807171812745, + "grad_norm": 0.142578125, + "learning_rate": 0.0015182670721968694, + "loss": 0.1021, + "step": 33885 + }, + { + "epoch": 0.2941467521983316, + "grad_norm": 0.37109375, + "learning_rate": 0.0015182407189030073, + "loss": 0.1162, + "step": 33886 + }, + { + "epoch": 0.2941554326785358, + "grad_norm": 0.2197265625, + "learning_rate": 0.001518214365151764, + "loss": 0.0894, + "step": 33887 + }, + { + "epoch": 0.2941641131587399, + "grad_norm": 0.09765625, + "learning_rate": 0.0015181880109431673, + "loss": 0.0811, + "step": 33888 + }, + { + "epoch": 0.2941727936389441, + "grad_norm": 0.404296875, + "learning_rate": 0.0015181616562772464, + "loss": 0.0991, + "step": 33889 + }, + { + "epoch": 0.29418147411914825, + "grad_norm": 0.33203125, + "learning_rate": 0.0015181353011540304, + "loss": 0.1084, + "step": 33890 + }, + { + "epoch": 0.29419015459935244, + "grad_norm": 0.2158203125, + "learning_rate": 0.0015181089455735478, + "loss": 0.084, + "step": 33891 + }, + { + "epoch": 0.2941988350795566, + "grad_norm": 0.56640625, + "learning_rate": 0.001518082589535828, + "loss": 0.127, + "step": 33892 + }, + { + "epoch": 0.29420751555976077, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015180562330408987, + "loss": 0.1172, + "step": 33893 + }, + { + "epoch": 0.2942161960399649, + "grad_norm": 0.486328125, + "learning_rate": 0.0015180298760887896, + "loss": 0.0918, + "step": 33894 + }, + { + "epoch": 0.2942248765201691, + "grad_norm": 0.1455078125, + "learning_rate": 0.0015180035186795296, + "loss": 0.1113, + "step": 33895 + }, + { + "epoch": 0.29423355700037324, + "grad_norm": 0.11767578125, + "learning_rate": 0.0015179771608131469, + "loss": 0.0869, + "step": 33896 + }, + { + "epoch": 0.29424223748057743, + "grad_norm": 0.34765625, + "learning_rate": 0.001517950802489671, + "loss": 0.1143, + "step": 33897 + }, + { + "epoch": 0.29425091796078157, + "grad_norm": 0.1484375, + "learning_rate": 0.00151792444370913, + "loss": 0.1035, + "step": 33898 + }, + { + "epoch": 0.29425959844098576, + "grad_norm": 0.1181640625, + "learning_rate": 0.0015178980844715532, + "loss": 0.0903, + "step": 33899 + }, + { + "epoch": 0.2942682789211899, + "grad_norm": 0.09423828125, + "learning_rate": 0.0015178717247769694, + "loss": 0.1035, + "step": 33900 + }, + { + "epoch": 0.2942769594013941, + "grad_norm": 0.177734375, + "learning_rate": 0.0015178453646254076, + "loss": 0.168, + "step": 33901 + }, + { + "epoch": 0.2942856398815982, + "grad_norm": 0.1484375, + "learning_rate": 0.0015178190040168959, + "loss": 0.0635, + "step": 33902 + }, + { + "epoch": 0.2942943203618024, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015177926429514638, + "loss": 0.0947, + "step": 33903 + }, + { + "epoch": 0.29430300084200656, + "grad_norm": 0.4453125, + "learning_rate": 0.00151776628142914, + "loss": 0.1094, + "step": 33904 + }, + { + "epoch": 0.29431168132221075, + "grad_norm": 0.30078125, + "learning_rate": 0.001517739919449953, + "loss": 0.1035, + "step": 33905 + }, + { + "epoch": 0.2943203618024149, + "grad_norm": 0.78515625, + "learning_rate": 0.0015177135570139321, + "loss": 0.0923, + "step": 33906 + }, + { + "epoch": 0.2943290422826191, + "grad_norm": 0.1435546875, + "learning_rate": 0.0015176871941211061, + "loss": 0.0981, + "step": 33907 + }, + { + "epoch": 0.2943377227628232, + "grad_norm": 0.126953125, + "learning_rate": 0.0015176608307715038, + "loss": 0.0801, + "step": 33908 + }, + { + "epoch": 0.2943464032430274, + "grad_norm": 0.134765625, + "learning_rate": 0.0015176344669651535, + "loss": 0.085, + "step": 33909 + }, + { + "epoch": 0.29435508372323155, + "grad_norm": 0.4296875, + "learning_rate": 0.0015176081027020846, + "loss": 0.0869, + "step": 33910 + }, + { + "epoch": 0.29436376420343574, + "grad_norm": 0.365234375, + "learning_rate": 0.0015175817379823257, + "loss": 0.104, + "step": 33911 + }, + { + "epoch": 0.2943724446836399, + "grad_norm": 0.63671875, + "learning_rate": 0.001517555372805906, + "loss": 0.1494, + "step": 33912 + }, + { + "epoch": 0.29438112516384407, + "grad_norm": 0.2353515625, + "learning_rate": 0.0015175290071728536, + "loss": 0.0752, + "step": 33913 + }, + { + "epoch": 0.2943898056440482, + "grad_norm": 0.19921875, + "learning_rate": 0.0015175026410831976, + "loss": 0.1289, + "step": 33914 + }, + { + "epoch": 0.2943984861242524, + "grad_norm": 0.1416015625, + "learning_rate": 0.0015174762745369672, + "loss": 0.0981, + "step": 33915 + }, + { + "epoch": 0.29440716660445654, + "grad_norm": 0.236328125, + "learning_rate": 0.0015174499075341912, + "loss": 0.1094, + "step": 33916 + }, + { + "epoch": 0.29441584708466073, + "grad_norm": 0.158203125, + "learning_rate": 0.0015174235400748982, + "loss": 0.0957, + "step": 33917 + }, + { + "epoch": 0.29442452756486487, + "grad_norm": 0.2265625, + "learning_rate": 0.001517397172159117, + "loss": 0.0898, + "step": 33918 + }, + { + "epoch": 0.29443320804506906, + "grad_norm": 0.2216796875, + "learning_rate": 0.0015173708037868766, + "loss": 0.1123, + "step": 33919 + }, + { + "epoch": 0.2944418885252732, + "grad_norm": 0.2001953125, + "learning_rate": 0.0015173444349582054, + "loss": 0.1016, + "step": 33920 + }, + { + "epoch": 0.2944505690054774, + "grad_norm": 0.85546875, + "learning_rate": 0.001517318065673133, + "loss": 0.1299, + "step": 33921 + }, + { + "epoch": 0.29445924948568153, + "grad_norm": 0.412109375, + "learning_rate": 0.001517291695931688, + "loss": 0.1221, + "step": 33922 + }, + { + "epoch": 0.2944679299658857, + "grad_norm": 0.91015625, + "learning_rate": 0.0015172653257338987, + "loss": 0.0898, + "step": 33923 + }, + { + "epoch": 0.29447661044608986, + "grad_norm": 0.1708984375, + "learning_rate": 0.0015172389550797945, + "loss": 0.123, + "step": 33924 + }, + { + "epoch": 0.29448529092629405, + "grad_norm": 0.7265625, + "learning_rate": 0.001517212583969404, + "loss": 0.1113, + "step": 33925 + }, + { + "epoch": 0.2944939714064982, + "grad_norm": 0.4921875, + "learning_rate": 0.0015171862124027558, + "loss": 0.1074, + "step": 33926 + }, + { + "epoch": 0.2945026518867024, + "grad_norm": 0.12890625, + "learning_rate": 0.0015171598403798793, + "loss": 0.1318, + "step": 33927 + }, + { + "epoch": 0.2945113323669065, + "grad_norm": 0.578125, + "learning_rate": 0.0015171334679008031, + "loss": 0.1084, + "step": 33928 + }, + { + "epoch": 0.2945200128471107, + "grad_norm": 0.1328125, + "learning_rate": 0.0015171070949655558, + "loss": 0.0938, + "step": 33929 + }, + { + "epoch": 0.29452869332731485, + "grad_norm": 0.56640625, + "learning_rate": 0.0015170807215741668, + "loss": 0.1133, + "step": 33930 + }, + { + "epoch": 0.29453737380751904, + "grad_norm": 0.490234375, + "learning_rate": 0.0015170543477266644, + "loss": 0.1167, + "step": 33931 + }, + { + "epoch": 0.2945460542877232, + "grad_norm": 0.431640625, + "learning_rate": 0.0015170279734230776, + "loss": 0.1055, + "step": 33932 + }, + { + "epoch": 0.2945547347679274, + "grad_norm": 0.47265625, + "learning_rate": 0.0015170015986634353, + "loss": 0.127, + "step": 33933 + }, + { + "epoch": 0.2945634152481315, + "grad_norm": 0.609375, + "learning_rate": 0.0015169752234477662, + "loss": 0.0928, + "step": 33934 + }, + { + "epoch": 0.2945720957283357, + "grad_norm": 0.150390625, + "learning_rate": 0.0015169488477760995, + "loss": 0.1226, + "step": 33935 + }, + { + "epoch": 0.29458077620853984, + "grad_norm": 0.640625, + "learning_rate": 0.0015169224716484635, + "loss": 0.0801, + "step": 33936 + }, + { + "epoch": 0.29458945668874403, + "grad_norm": 0.6953125, + "learning_rate": 0.0015168960950648876, + "loss": 0.0854, + "step": 33937 + }, + { + "epoch": 0.29459813716894817, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015168697180254003, + "loss": 0.0977, + "step": 33938 + }, + { + "epoch": 0.29460681764915236, + "grad_norm": 0.181640625, + "learning_rate": 0.0015168433405300305, + "loss": 0.1152, + "step": 33939 + }, + { + "epoch": 0.2946154981293565, + "grad_norm": 0.5859375, + "learning_rate": 0.0015168169625788073, + "loss": 0.1738, + "step": 33940 + }, + { + "epoch": 0.2946241786095607, + "grad_norm": 0.1669921875, + "learning_rate": 0.001516790584171759, + "loss": 0.105, + "step": 33941 + }, + { + "epoch": 0.29463285908976483, + "grad_norm": 0.078125, + "learning_rate": 0.0015167642053089151, + "loss": 0.0801, + "step": 33942 + }, + { + "epoch": 0.294641539569969, + "grad_norm": 0.609375, + "learning_rate": 0.001516737825990304, + "loss": 0.126, + "step": 33943 + }, + { + "epoch": 0.29465022005017316, + "grad_norm": 0.09619140625, + "learning_rate": 0.0015167114462159546, + "loss": 0.0874, + "step": 33944 + }, + { + "epoch": 0.29465890053037735, + "grad_norm": 0.11083984375, + "learning_rate": 0.0015166850659858958, + "loss": 0.0928, + "step": 33945 + }, + { + "epoch": 0.2946675810105815, + "grad_norm": 0.408203125, + "learning_rate": 0.0015166586853001564, + "loss": 0.0811, + "step": 33946 + }, + { + "epoch": 0.2946762614907857, + "grad_norm": 0.287109375, + "learning_rate": 0.0015166323041587654, + "loss": 0.1206, + "step": 33947 + }, + { + "epoch": 0.2946849419709898, + "grad_norm": 0.248046875, + "learning_rate": 0.001516605922561752, + "loss": 0.0986, + "step": 33948 + }, + { + "epoch": 0.294693622451194, + "grad_norm": 0.51171875, + "learning_rate": 0.001516579540509144, + "loss": 0.0908, + "step": 33949 + }, + { + "epoch": 0.29470230293139815, + "grad_norm": 0.07421875, + "learning_rate": 0.0015165531580009712, + "loss": 0.082, + "step": 33950 + }, + { + "epoch": 0.29471098341160235, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015165267750372623, + "loss": 0.1118, + "step": 33951 + }, + { + "epoch": 0.2947196638918065, + "grad_norm": 0.20703125, + "learning_rate": 0.0015165003916180458, + "loss": 0.1162, + "step": 33952 + }, + { + "epoch": 0.2947283443720107, + "grad_norm": 0.392578125, + "learning_rate": 0.0015164740077433505, + "loss": 0.1152, + "step": 33953 + }, + { + "epoch": 0.2947370248522148, + "grad_norm": 0.26953125, + "learning_rate": 0.0015164476234132054, + "loss": 0.0835, + "step": 33954 + }, + { + "epoch": 0.294745705332419, + "grad_norm": 0.67578125, + "learning_rate": 0.00151642123862764, + "loss": 0.1201, + "step": 33955 + }, + { + "epoch": 0.29475438581262314, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015163948533866822, + "loss": 0.1113, + "step": 33956 + }, + { + "epoch": 0.29476306629282734, + "grad_norm": 0.1728515625, + "learning_rate": 0.0015163684676903612, + "loss": 0.1138, + "step": 33957 + }, + { + "epoch": 0.2947717467730315, + "grad_norm": 0.34765625, + "learning_rate": 0.001516342081538706, + "loss": 0.0981, + "step": 33958 + }, + { + "epoch": 0.29478042725323567, + "grad_norm": 0.08056640625, + "learning_rate": 0.0015163156949317457, + "loss": 0.0854, + "step": 33959 + }, + { + "epoch": 0.2947891077334398, + "grad_norm": 0.34375, + "learning_rate": 0.0015162893078695084, + "loss": 0.0967, + "step": 33960 + }, + { + "epoch": 0.294797788213644, + "grad_norm": 0.50390625, + "learning_rate": 0.0015162629203520235, + "loss": 0.124, + "step": 33961 + }, + { + "epoch": 0.29480646869384813, + "grad_norm": 0.130859375, + "learning_rate": 0.0015162365323793195, + "loss": 0.1123, + "step": 33962 + }, + { + "epoch": 0.2948151491740523, + "grad_norm": 0.087890625, + "learning_rate": 0.0015162101439514257, + "loss": 0.1001, + "step": 33963 + }, + { + "epoch": 0.29482382965425646, + "grad_norm": 0.138671875, + "learning_rate": 0.0015161837550683706, + "loss": 0.0923, + "step": 33964 + }, + { + "epoch": 0.29483251013446066, + "grad_norm": 0.67578125, + "learning_rate": 0.0015161573657301835, + "loss": 0.0908, + "step": 33965 + }, + { + "epoch": 0.2948411906146648, + "grad_norm": 0.31640625, + "learning_rate": 0.0015161309759368926, + "loss": 0.0947, + "step": 33966 + }, + { + "epoch": 0.294849871094869, + "grad_norm": 0.494140625, + "learning_rate": 0.001516104585688527, + "loss": 0.1094, + "step": 33967 + }, + { + "epoch": 0.2948585515750731, + "grad_norm": 0.275390625, + "learning_rate": 0.0015160781949851163, + "loss": 0.0771, + "step": 33968 + }, + { + "epoch": 0.2948672320552773, + "grad_norm": 0.111328125, + "learning_rate": 0.0015160518038266884, + "loss": 0.0845, + "step": 33969 + }, + { + "epoch": 0.29487591253548145, + "grad_norm": 0.2255859375, + "learning_rate": 0.0015160254122132727, + "loss": 0.1064, + "step": 33970 + }, + { + "epoch": 0.29488459301568565, + "grad_norm": 0.4921875, + "learning_rate": 0.0015159990201448973, + "loss": 0.0923, + "step": 33971 + }, + { + "epoch": 0.2948932734958898, + "grad_norm": 0.123046875, + "learning_rate": 0.0015159726276215918, + "loss": 0.0864, + "step": 33972 + }, + { + "epoch": 0.294901953976094, + "grad_norm": 0.18359375, + "learning_rate": 0.0015159462346433852, + "loss": 0.1187, + "step": 33973 + }, + { + "epoch": 0.2949106344562981, + "grad_norm": 0.57421875, + "learning_rate": 0.001515919841210306, + "loss": 0.0898, + "step": 33974 + }, + { + "epoch": 0.2949193149365023, + "grad_norm": 0.427734375, + "learning_rate": 0.0015158934473223828, + "loss": 0.0928, + "step": 33975 + }, + { + "epoch": 0.29492799541670645, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015158670529796451, + "loss": 0.1133, + "step": 33976 + }, + { + "epoch": 0.29493667589691064, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015158406581821213, + "loss": 0.1094, + "step": 33977 + }, + { + "epoch": 0.2949453563771148, + "grad_norm": 0.0869140625, + "learning_rate": 0.0015158142629298405, + "loss": 0.1182, + "step": 33978 + }, + { + "epoch": 0.29495403685731897, + "grad_norm": 0.1640625, + "learning_rate": 0.0015157878672228315, + "loss": 0.1104, + "step": 33979 + }, + { + "epoch": 0.2949627173375231, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015157614710611229, + "loss": 0.167, + "step": 33980 + }, + { + "epoch": 0.2949713978177273, + "grad_norm": 0.515625, + "learning_rate": 0.0015157350744447442, + "loss": 0.0957, + "step": 33981 + }, + { + "epoch": 0.29498007829793144, + "grad_norm": 0.18359375, + "learning_rate": 0.0015157086773737235, + "loss": 0.127, + "step": 33982 + }, + { + "epoch": 0.29498875877813563, + "grad_norm": 0.1298828125, + "learning_rate": 0.00151568227984809, + "loss": 0.1035, + "step": 33983 + }, + { + "epoch": 0.29499743925833977, + "grad_norm": 0.1640625, + "learning_rate": 0.0015156558818678726, + "loss": 0.106, + "step": 33984 + }, + { + "epoch": 0.29500611973854396, + "grad_norm": 0.126953125, + "learning_rate": 0.0015156294834331005, + "loss": 0.0781, + "step": 33985 + }, + { + "epoch": 0.2950148002187481, + "grad_norm": 0.263671875, + "learning_rate": 0.0015156030845438022, + "loss": 0.1025, + "step": 33986 + }, + { + "epoch": 0.2950234806989523, + "grad_norm": 0.380859375, + "learning_rate": 0.0015155766852000066, + "loss": 0.0938, + "step": 33987 + }, + { + "epoch": 0.2950321611791564, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015155502854017424, + "loss": 0.1514, + "step": 33988 + }, + { + "epoch": 0.2950408416593606, + "grad_norm": 0.259765625, + "learning_rate": 0.0015155238851490387, + "loss": 0.0664, + "step": 33989 + }, + { + "epoch": 0.29504952213956476, + "grad_norm": 0.185546875, + "learning_rate": 0.0015154974844419245, + "loss": 0.1172, + "step": 33990 + }, + { + "epoch": 0.29505820261976895, + "grad_norm": 0.41015625, + "learning_rate": 0.001515471083280428, + "loss": 0.1621, + "step": 33991 + }, + { + "epoch": 0.2950668830999731, + "grad_norm": 0.32421875, + "learning_rate": 0.001515444681664579, + "loss": 0.1011, + "step": 33992 + }, + { + "epoch": 0.2950755635801773, + "grad_norm": 0.482421875, + "learning_rate": 0.0015154182795944057, + "loss": 0.1406, + "step": 33993 + }, + { + "epoch": 0.2950842440603814, + "grad_norm": 0.08544921875, + "learning_rate": 0.0015153918770699373, + "loss": 0.0947, + "step": 33994 + }, + { + "epoch": 0.2950929245405856, + "grad_norm": 0.55078125, + "learning_rate": 0.0015153654740912028, + "loss": 0.1484, + "step": 33995 + }, + { + "epoch": 0.29510160502078975, + "grad_norm": 0.154296875, + "learning_rate": 0.0015153390706582309, + "loss": 0.0732, + "step": 33996 + }, + { + "epoch": 0.29511028550099394, + "grad_norm": 0.216796875, + "learning_rate": 0.00151531266677105, + "loss": 0.1167, + "step": 33997 + }, + { + "epoch": 0.2951189659811981, + "grad_norm": 0.12109375, + "learning_rate": 0.0015152862624296898, + "loss": 0.1289, + "step": 33998 + }, + { + "epoch": 0.29512764646140227, + "grad_norm": 0.318359375, + "learning_rate": 0.0015152598576341784, + "loss": 0.0947, + "step": 33999 + }, + { + "epoch": 0.2951363269416064, + "grad_norm": 0.185546875, + "learning_rate": 0.0015152334523845451, + "loss": 0.1523, + "step": 34000 + }, + { + "epoch": 0.2951450074218106, + "grad_norm": 0.51171875, + "learning_rate": 0.0015152070466808191, + "loss": 0.1777, + "step": 34001 + }, + { + "epoch": 0.29515368790201474, + "grad_norm": 0.515625, + "learning_rate": 0.0015151806405230286, + "loss": 0.123, + "step": 34002 + }, + { + "epoch": 0.29516236838221893, + "grad_norm": 0.24609375, + "learning_rate": 0.001515154233911203, + "loss": 0.1494, + "step": 34003 + }, + { + "epoch": 0.29517104886242307, + "grad_norm": 0.1484375, + "learning_rate": 0.0015151278268453708, + "loss": 0.1035, + "step": 34004 + }, + { + "epoch": 0.29517972934262726, + "grad_norm": 0.51171875, + "learning_rate": 0.0015151014193255613, + "loss": 0.085, + "step": 34005 + }, + { + "epoch": 0.2951884098228314, + "grad_norm": 0.19140625, + "learning_rate": 0.0015150750113518028, + "loss": 0.0684, + "step": 34006 + }, + { + "epoch": 0.29519709030303554, + "grad_norm": 0.494140625, + "learning_rate": 0.0015150486029241246, + "loss": 0.1465, + "step": 34007 + }, + { + "epoch": 0.29520577078323973, + "grad_norm": 0.1767578125, + "learning_rate": 0.0015150221940425557, + "loss": 0.125, + "step": 34008 + }, + { + "epoch": 0.29521445126344387, + "grad_norm": 0.8671875, + "learning_rate": 0.0015149957847071245, + "loss": 0.1108, + "step": 34009 + }, + { + "epoch": 0.29522313174364806, + "grad_norm": 0.0849609375, + "learning_rate": 0.0015149693749178603, + "loss": 0.1133, + "step": 34010 + }, + { + "epoch": 0.2952318122238522, + "grad_norm": 0.69921875, + "learning_rate": 0.0015149429646747917, + "loss": 0.1123, + "step": 34011 + }, + { + "epoch": 0.2952404927040564, + "grad_norm": 0.1689453125, + "learning_rate": 0.0015149165539779478, + "loss": 0.125, + "step": 34012 + }, + { + "epoch": 0.2952491731842605, + "grad_norm": 0.484375, + "learning_rate": 0.0015148901428273574, + "loss": 0.0894, + "step": 34013 + }, + { + "epoch": 0.2952578536644647, + "grad_norm": 1.3671875, + "learning_rate": 0.001514863731223049, + "loss": 0.1357, + "step": 34014 + }, + { + "epoch": 0.29526653414466886, + "grad_norm": 0.3671875, + "learning_rate": 0.0015148373191650525, + "loss": 0.1465, + "step": 34015 + }, + { + "epoch": 0.29527521462487305, + "grad_norm": 0.302734375, + "learning_rate": 0.001514810906653396, + "loss": 0.1025, + "step": 34016 + }, + { + "epoch": 0.2952838951050772, + "grad_norm": 0.671875, + "learning_rate": 0.0015147844936881083, + "loss": 0.1484, + "step": 34017 + }, + { + "epoch": 0.2952925755852814, + "grad_norm": 0.08203125, + "learning_rate": 0.0015147580802692187, + "loss": 0.0977, + "step": 34018 + }, + { + "epoch": 0.2953012560654855, + "grad_norm": 0.478515625, + "learning_rate": 0.0015147316663967553, + "loss": 0.1533, + "step": 34019 + }, + { + "epoch": 0.2953099365456897, + "grad_norm": 0.1796875, + "learning_rate": 0.001514705252070748, + "loss": 0.084, + "step": 34020 + }, + { + "epoch": 0.29531861702589385, + "grad_norm": 0.796875, + "learning_rate": 0.0015146788372912257, + "loss": 0.0996, + "step": 34021 + }, + { + "epoch": 0.29532729750609804, + "grad_norm": 0.314453125, + "learning_rate": 0.0015146524220582163, + "loss": 0.105, + "step": 34022 + }, + { + "epoch": 0.2953359779863022, + "grad_norm": 0.373046875, + "learning_rate": 0.001514626006371749, + "loss": 0.1235, + "step": 34023 + }, + { + "epoch": 0.29534465846650637, + "grad_norm": 1.078125, + "learning_rate": 0.0015145995902318538, + "loss": 0.4238, + "step": 34024 + }, + { + "epoch": 0.2953533389467105, + "grad_norm": 0.37890625, + "learning_rate": 0.0015145731736385581, + "loss": 0.0918, + "step": 34025 + }, + { + "epoch": 0.2953620194269147, + "grad_norm": 0.140625, + "learning_rate": 0.0015145467565918919, + "loss": 0.1201, + "step": 34026 + }, + { + "epoch": 0.29537069990711884, + "grad_norm": 0.240234375, + "learning_rate": 0.001514520339091883, + "loss": 0.1069, + "step": 34027 + }, + { + "epoch": 0.29537938038732303, + "grad_norm": 0.326171875, + "learning_rate": 0.001514493921138561, + "loss": 0.1338, + "step": 34028 + }, + { + "epoch": 0.29538806086752717, + "grad_norm": 0.337890625, + "learning_rate": 0.0015144675027319553, + "loss": 0.0908, + "step": 34029 + }, + { + "epoch": 0.29539674134773136, + "grad_norm": 0.34375, + "learning_rate": 0.0015144410838720933, + "loss": 0.1074, + "step": 34030 + }, + { + "epoch": 0.2954054218279355, + "grad_norm": 0.357421875, + "learning_rate": 0.0015144146645590052, + "loss": 0.1289, + "step": 34031 + }, + { + "epoch": 0.2954141023081397, + "grad_norm": 0.58984375, + "learning_rate": 0.0015143882447927193, + "loss": 0.1172, + "step": 34032 + }, + { + "epoch": 0.29542278278834383, + "grad_norm": 0.39453125, + "learning_rate": 0.0015143618245732648, + "loss": 0.0859, + "step": 34033 + }, + { + "epoch": 0.295431463268548, + "grad_norm": 0.2080078125, + "learning_rate": 0.0015143354039006702, + "loss": 0.1006, + "step": 34034 + }, + { + "epoch": 0.29544014374875216, + "grad_norm": 0.302734375, + "learning_rate": 0.001514308982774965, + "loss": 0.1064, + "step": 34035 + }, + { + "epoch": 0.29544882422895635, + "grad_norm": 0.259765625, + "learning_rate": 0.0015142825611961775, + "loss": 0.1416, + "step": 34036 + }, + { + "epoch": 0.2954575047091605, + "grad_norm": 0.34765625, + "learning_rate": 0.001514256139164337, + "loss": 0.1475, + "step": 34037 + }, + { + "epoch": 0.2954661851893647, + "grad_norm": 0.07763671875, + "learning_rate": 0.0015142297166794718, + "loss": 0.1118, + "step": 34038 + }, + { + "epoch": 0.2954748656695688, + "grad_norm": 0.53515625, + "learning_rate": 0.0015142032937416115, + "loss": 0.0742, + "step": 34039 + }, + { + "epoch": 0.295483546149773, + "grad_norm": 0.1923828125, + "learning_rate": 0.0015141768703507846, + "loss": 0.0742, + "step": 34040 + }, + { + "epoch": 0.29549222662997715, + "grad_norm": 0.3203125, + "learning_rate": 0.00151415044650702, + "loss": 0.0869, + "step": 34041 + }, + { + "epoch": 0.29550090711018134, + "grad_norm": 0.2197265625, + "learning_rate": 0.001514124022210347, + "loss": 0.0918, + "step": 34042 + }, + { + "epoch": 0.2955095875903855, + "grad_norm": 0.134765625, + "learning_rate": 0.001514097597460794, + "loss": 0.1104, + "step": 34043 + }, + { + "epoch": 0.2955182680705897, + "grad_norm": 0.6484375, + "learning_rate": 0.0015140711722583901, + "loss": 0.0947, + "step": 34044 + }, + { + "epoch": 0.2955269485507938, + "grad_norm": 0.703125, + "learning_rate": 0.0015140447466031643, + "loss": 0.1201, + "step": 34045 + }, + { + "epoch": 0.295535629030998, + "grad_norm": 0.5546875, + "learning_rate": 0.0015140183204951454, + "loss": 0.0928, + "step": 34046 + }, + { + "epoch": 0.29554430951120214, + "grad_norm": 0.0693359375, + "learning_rate": 0.001513991893934362, + "loss": 0.0815, + "step": 34047 + }, + { + "epoch": 0.29555298999140633, + "grad_norm": 0.69140625, + "learning_rate": 0.0015139654669208435, + "loss": 0.1025, + "step": 34048 + }, + { + "epoch": 0.29556167047161047, + "grad_norm": 0.28125, + "learning_rate": 0.0015139390394546186, + "loss": 0.1348, + "step": 34049 + }, + { + "epoch": 0.29557035095181466, + "grad_norm": 0.27734375, + "learning_rate": 0.001513912611535716, + "loss": 0.1094, + "step": 34050 + }, + { + "epoch": 0.2955790314320188, + "grad_norm": 0.12109375, + "learning_rate": 0.0015138861831641651, + "loss": 0.1011, + "step": 34051 + }, + { + "epoch": 0.295587711912223, + "grad_norm": 0.1337890625, + "learning_rate": 0.0015138597543399943, + "loss": 0.127, + "step": 34052 + }, + { + "epoch": 0.29559639239242713, + "grad_norm": 0.25390625, + "learning_rate": 0.0015138333250632324, + "loss": 0.0825, + "step": 34053 + }, + { + "epoch": 0.2956050728726313, + "grad_norm": 0.1923828125, + "learning_rate": 0.001513806895333909, + "loss": 0.1191, + "step": 34054 + }, + { + "epoch": 0.29561375335283546, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015137804651520527, + "loss": 0.1201, + "step": 34055 + }, + { + "epoch": 0.29562243383303966, + "grad_norm": 0.392578125, + "learning_rate": 0.001513754034517692, + "loss": 0.1074, + "step": 34056 + }, + { + "epoch": 0.2956311143132438, + "grad_norm": 0.318359375, + "learning_rate": 0.0015137276034308561, + "loss": 0.0923, + "step": 34057 + }, + { + "epoch": 0.295639794793448, + "grad_norm": 0.9609375, + "learning_rate": 0.0015137011718915742, + "loss": 0.1348, + "step": 34058 + }, + { + "epoch": 0.2956484752736521, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015136747398998745, + "loss": 0.1084, + "step": 34059 + }, + { + "epoch": 0.2956571557538563, + "grad_norm": 0.1611328125, + "learning_rate": 0.0015136483074557868, + "loss": 0.1211, + "step": 34060 + }, + { + "epoch": 0.29566583623406045, + "grad_norm": 0.1416015625, + "learning_rate": 0.0015136218745593394, + "loss": 0.1406, + "step": 34061 + }, + { + "epoch": 0.29567451671426465, + "grad_norm": 0.12255859375, + "learning_rate": 0.0015135954412105609, + "loss": 0.1016, + "step": 34062 + }, + { + "epoch": 0.2956831971944688, + "grad_norm": 0.265625, + "learning_rate": 0.001513569007409481, + "loss": 0.1177, + "step": 34063 + }, + { + "epoch": 0.295691877674673, + "grad_norm": 0.2275390625, + "learning_rate": 0.0015135425731561281, + "loss": 0.0923, + "step": 34064 + }, + { + "epoch": 0.2957005581548771, + "grad_norm": 0.2294921875, + "learning_rate": 0.0015135161384505313, + "loss": 0.1562, + "step": 34065 + }, + { + "epoch": 0.2957092386350813, + "grad_norm": 0.171875, + "learning_rate": 0.0015134897032927194, + "loss": 0.0762, + "step": 34066 + }, + { + "epoch": 0.29571791911528544, + "grad_norm": 0.6328125, + "learning_rate": 0.0015134632676827215, + "loss": 0.1069, + "step": 34067 + }, + { + "epoch": 0.29572659959548964, + "grad_norm": 0.08935546875, + "learning_rate": 0.0015134368316205664, + "loss": 0.0889, + "step": 34068 + }, + { + "epoch": 0.2957352800756938, + "grad_norm": 0.0888671875, + "learning_rate": 0.001513410395106283, + "loss": 0.1089, + "step": 34069 + }, + { + "epoch": 0.29574396055589797, + "grad_norm": 0.7734375, + "learning_rate": 0.0015133839581399002, + "loss": 0.0962, + "step": 34070 + }, + { + "epoch": 0.2957526410361021, + "grad_norm": 0.10888671875, + "learning_rate": 0.001513357520721447, + "loss": 0.1348, + "step": 34071 + }, + { + "epoch": 0.2957613215163063, + "grad_norm": 0.263671875, + "learning_rate": 0.0015133310828509521, + "loss": 0.1123, + "step": 34072 + }, + { + "epoch": 0.29577000199651043, + "grad_norm": 0.271484375, + "learning_rate": 0.0015133046445284443, + "loss": 0.123, + "step": 34073 + }, + { + "epoch": 0.2957786824767146, + "grad_norm": 0.2080078125, + "learning_rate": 0.0015132782057539532, + "loss": 0.0918, + "step": 34074 + }, + { + "epoch": 0.29578736295691876, + "grad_norm": 0.828125, + "learning_rate": 0.001513251766527507, + "loss": 0.1152, + "step": 34075 + }, + { + "epoch": 0.29579604343712296, + "grad_norm": 0.16796875, + "learning_rate": 0.0015132253268491352, + "loss": 0.0986, + "step": 34076 + }, + { + "epoch": 0.2958047239173271, + "grad_norm": 0.1416015625, + "learning_rate": 0.001513198886718866, + "loss": 0.0781, + "step": 34077 + }, + { + "epoch": 0.2958134043975313, + "grad_norm": 0.126953125, + "learning_rate": 0.0015131724461367284, + "loss": 0.1035, + "step": 34078 + }, + { + "epoch": 0.2958220848777354, + "grad_norm": 0.462890625, + "learning_rate": 0.0015131460051027521, + "loss": 0.0879, + "step": 34079 + }, + { + "epoch": 0.2958307653579396, + "grad_norm": 0.1796875, + "learning_rate": 0.0015131195636169657, + "loss": 0.1387, + "step": 34080 + }, + { + "epoch": 0.29583944583814376, + "grad_norm": 0.3828125, + "learning_rate": 0.0015130931216793978, + "loss": 0.0845, + "step": 34081 + }, + { + "epoch": 0.29584812631834795, + "grad_norm": 0.07568359375, + "learning_rate": 0.0015130666792900773, + "loss": 0.0918, + "step": 34082 + }, + { + "epoch": 0.2958568067985521, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015130402364490333, + "loss": 0.0659, + "step": 34083 + }, + { + "epoch": 0.2958654872787563, + "grad_norm": 0.1748046875, + "learning_rate": 0.001513013793156295, + "loss": 0.1006, + "step": 34084 + }, + { + "epoch": 0.2958741677589604, + "grad_norm": 0.11376953125, + "learning_rate": 0.0015129873494118908, + "loss": 0.0977, + "step": 34085 + }, + { + "epoch": 0.2958828482391646, + "grad_norm": 0.099609375, + "learning_rate": 0.00151296090521585, + "loss": 0.0933, + "step": 34086 + }, + { + "epoch": 0.29589152871936875, + "grad_norm": 0.390625, + "learning_rate": 0.001512934460568201, + "loss": 0.1689, + "step": 34087 + }, + { + "epoch": 0.29590020919957294, + "grad_norm": 0.83203125, + "learning_rate": 0.0015129080154689732, + "loss": 0.1299, + "step": 34088 + }, + { + "epoch": 0.2959088896797771, + "grad_norm": 0.41796875, + "learning_rate": 0.0015128815699181955, + "loss": 0.1079, + "step": 34089 + }, + { + "epoch": 0.29591757015998127, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015128551239158967, + "loss": 0.0713, + "step": 34090 + }, + { + "epoch": 0.2959262506401854, + "grad_norm": 0.68359375, + "learning_rate": 0.0015128286774621059, + "loss": 0.0947, + "step": 34091 + }, + { + "epoch": 0.2959349311203896, + "grad_norm": 0.51953125, + "learning_rate": 0.0015128022305568515, + "loss": 0.082, + "step": 34092 + }, + { + "epoch": 0.29594361160059374, + "grad_norm": 0.34765625, + "learning_rate": 0.0015127757832001632, + "loss": 0.1484, + "step": 34093 + }, + { + "epoch": 0.29595229208079793, + "grad_norm": 0.3203125, + "learning_rate": 0.0015127493353920695, + "loss": 0.0962, + "step": 34094 + }, + { + "epoch": 0.29596097256100207, + "grad_norm": 0.095703125, + "learning_rate": 0.0015127228871325991, + "loss": 0.0952, + "step": 34095 + }, + { + "epoch": 0.29596965304120626, + "grad_norm": 0.263671875, + "learning_rate": 0.0015126964384217809, + "loss": 0.0713, + "step": 34096 + }, + { + "epoch": 0.2959783335214104, + "grad_norm": 0.09033203125, + "learning_rate": 0.0015126699892596444, + "loss": 0.0889, + "step": 34097 + }, + { + "epoch": 0.2959870140016146, + "grad_norm": 0.35546875, + "learning_rate": 0.0015126435396462178, + "loss": 0.1074, + "step": 34098 + }, + { + "epoch": 0.2959956944818187, + "grad_norm": 0.1142578125, + "learning_rate": 0.0015126170895815312, + "loss": 0.1074, + "step": 34099 + }, + { + "epoch": 0.2960043749620229, + "grad_norm": 0.107421875, + "learning_rate": 0.0015125906390656121, + "loss": 0.0938, + "step": 34100 + }, + { + "epoch": 0.29601305544222706, + "grad_norm": 0.1455078125, + "learning_rate": 0.0015125641880984904, + "loss": 0.1016, + "step": 34101 + }, + { + "epoch": 0.29602173592243125, + "grad_norm": 0.07763671875, + "learning_rate": 0.0015125377366801947, + "loss": 0.0718, + "step": 34102 + }, + { + "epoch": 0.2960304164026354, + "grad_norm": 0.240234375, + "learning_rate": 0.0015125112848107538, + "loss": 0.0996, + "step": 34103 + }, + { + "epoch": 0.2960390968828396, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015124848324901967, + "loss": 0.0957, + "step": 34104 + }, + { + "epoch": 0.2960477773630437, + "grad_norm": 0.345703125, + "learning_rate": 0.0015124583797185526, + "loss": 0.1201, + "step": 34105 + }, + { + "epoch": 0.2960564578432479, + "grad_norm": 0.376953125, + "learning_rate": 0.0015124319264958502, + "loss": 0.0938, + "step": 34106 + }, + { + "epoch": 0.29606513832345205, + "grad_norm": 0.11767578125, + "learning_rate": 0.001512405472822118, + "loss": 0.1152, + "step": 34107 + }, + { + "epoch": 0.29607381880365624, + "grad_norm": 0.197265625, + "learning_rate": 0.0015123790186973858, + "loss": 0.1504, + "step": 34108 + }, + { + "epoch": 0.2960824992838604, + "grad_norm": 0.208984375, + "learning_rate": 0.0015123525641216819, + "loss": 0.1182, + "step": 34109 + }, + { + "epoch": 0.29609117976406457, + "grad_norm": 0.380859375, + "learning_rate": 0.0015123261090950356, + "loss": 0.084, + "step": 34110 + }, + { + "epoch": 0.2960998602442687, + "grad_norm": 0.08642578125, + "learning_rate": 0.0015122996536174756, + "loss": 0.0894, + "step": 34111 + }, + { + "epoch": 0.2961085407244729, + "grad_norm": 0.578125, + "learning_rate": 0.001512273197689031, + "loss": 0.0933, + "step": 34112 + }, + { + "epoch": 0.29611722120467704, + "grad_norm": 0.11083984375, + "learning_rate": 0.0015122467413097305, + "loss": 0.1211, + "step": 34113 + }, + { + "epoch": 0.29612590168488123, + "grad_norm": 0.1044921875, + "learning_rate": 0.0015122202844796032, + "loss": 0.1182, + "step": 34114 + }, + { + "epoch": 0.29613458216508537, + "grad_norm": 0.10595703125, + "learning_rate": 0.0015121938271986778, + "loss": 0.1348, + "step": 34115 + }, + { + "epoch": 0.29614326264528956, + "grad_norm": 0.4921875, + "learning_rate": 0.0015121673694669835, + "loss": 0.1309, + "step": 34116 + }, + { + "epoch": 0.2961519431254937, + "grad_norm": 0.625, + "learning_rate": 0.0015121409112845495, + "loss": 0.1006, + "step": 34117 + }, + { + "epoch": 0.2961606236056979, + "grad_norm": 0.1953125, + "learning_rate": 0.001512114452651404, + "loss": 0.0913, + "step": 34118 + }, + { + "epoch": 0.29616930408590203, + "grad_norm": 0.1123046875, + "learning_rate": 0.0015120879935675764, + "loss": 0.0752, + "step": 34119 + }, + { + "epoch": 0.2961779845661062, + "grad_norm": 0.2099609375, + "learning_rate": 0.0015120615340330958, + "loss": 0.1113, + "step": 34120 + }, + { + "epoch": 0.29618666504631036, + "grad_norm": 0.1416015625, + "learning_rate": 0.001512035074047991, + "loss": 0.1196, + "step": 34121 + }, + { + "epoch": 0.29619534552651455, + "grad_norm": 0.1162109375, + "learning_rate": 0.0015120086136122903, + "loss": 0.1055, + "step": 34122 + }, + { + "epoch": 0.2962040260067187, + "grad_norm": 0.25390625, + "learning_rate": 0.0015119821527260233, + "loss": 0.1016, + "step": 34123 + }, + { + "epoch": 0.2962127064869229, + "grad_norm": 0.119140625, + "learning_rate": 0.001511955691389219, + "loss": 0.0928, + "step": 34124 + }, + { + "epoch": 0.296221386967127, + "grad_norm": 0.255859375, + "learning_rate": 0.001511929229601906, + "loss": 0.0913, + "step": 34125 + }, + { + "epoch": 0.2962300674473312, + "grad_norm": 0.310546875, + "learning_rate": 0.0015119027673641135, + "loss": 0.1182, + "step": 34126 + }, + { + "epoch": 0.29623874792753535, + "grad_norm": 0.75390625, + "learning_rate": 0.0015118763046758701, + "loss": 0.1758, + "step": 34127 + }, + { + "epoch": 0.29624742840773954, + "grad_norm": 0.302734375, + "learning_rate": 0.0015118498415372056, + "loss": 0.085, + "step": 34128 + }, + { + "epoch": 0.2962561088879437, + "grad_norm": 0.4765625, + "learning_rate": 0.0015118233779481476, + "loss": 0.125, + "step": 34129 + }, + { + "epoch": 0.2962647893681478, + "grad_norm": 0.1884765625, + "learning_rate": 0.001511796913908726, + "loss": 0.0952, + "step": 34130 + }, + { + "epoch": 0.296273469848352, + "grad_norm": 0.138671875, + "learning_rate": 0.0015117704494189694, + "loss": 0.1172, + "step": 34131 + }, + { + "epoch": 0.29628215032855615, + "grad_norm": 0.27734375, + "learning_rate": 0.0015117439844789069, + "loss": 0.1064, + "step": 34132 + }, + { + "epoch": 0.29629083080876034, + "grad_norm": 0.2109375, + "learning_rate": 0.0015117175190885674, + "loss": 0.0835, + "step": 34133 + }, + { + "epoch": 0.2962995112889645, + "grad_norm": 0.53125, + "learning_rate": 0.0015116910532479801, + "loss": 0.1387, + "step": 34134 + }, + { + "epoch": 0.29630819176916867, + "grad_norm": 0.115234375, + "learning_rate": 0.0015116645869571732, + "loss": 0.1133, + "step": 34135 + }, + { + "epoch": 0.2963168722493728, + "grad_norm": 0.2080078125, + "learning_rate": 0.001511638120216176, + "loss": 0.0835, + "step": 34136 + }, + { + "epoch": 0.296325552729577, + "grad_norm": 0.44140625, + "learning_rate": 0.0015116116530250177, + "loss": 0.085, + "step": 34137 + }, + { + "epoch": 0.29633423320978114, + "grad_norm": 0.119140625, + "learning_rate": 0.0015115851853837272, + "loss": 0.1025, + "step": 34138 + }, + { + "epoch": 0.29634291368998533, + "grad_norm": 0.10009765625, + "learning_rate": 0.0015115587172923333, + "loss": 0.1113, + "step": 34139 + }, + { + "epoch": 0.29635159417018947, + "grad_norm": 0.201171875, + "learning_rate": 0.0015115322487508649, + "loss": 0.1025, + "step": 34140 + }, + { + "epoch": 0.29636027465039366, + "grad_norm": 0.279296875, + "learning_rate": 0.0015115057797593513, + "loss": 0.0898, + "step": 34141 + }, + { + "epoch": 0.2963689551305978, + "grad_norm": 0.1513671875, + "learning_rate": 0.0015114793103178205, + "loss": 0.0669, + "step": 34142 + }, + { + "epoch": 0.296377635610802, + "grad_norm": 0.150390625, + "learning_rate": 0.001511452840426303, + "loss": 0.1011, + "step": 34143 + }, + { + "epoch": 0.29638631609100613, + "grad_norm": 0.73828125, + "learning_rate": 0.001511426370084826, + "loss": 0.1455, + "step": 34144 + }, + { + "epoch": 0.2963949965712103, + "grad_norm": 0.140625, + "learning_rate": 0.0015113998992934198, + "loss": 0.1191, + "step": 34145 + }, + { + "epoch": 0.29640367705141446, + "grad_norm": 0.318359375, + "learning_rate": 0.0015113734280521125, + "loss": 0.1367, + "step": 34146 + }, + { + "epoch": 0.29641235753161865, + "grad_norm": 0.22265625, + "learning_rate": 0.0015113469563609338, + "loss": 0.0879, + "step": 34147 + }, + { + "epoch": 0.2964210380118228, + "grad_norm": 0.1181640625, + "learning_rate": 0.001511320484219912, + "loss": 0.1167, + "step": 34148 + }, + { + "epoch": 0.296429718492027, + "grad_norm": 0.08154296875, + "learning_rate": 0.0015112940116290762, + "loss": 0.1172, + "step": 34149 + }, + { + "epoch": 0.2964383989722311, + "grad_norm": 0.13671875, + "learning_rate": 0.001511267538588456, + "loss": 0.085, + "step": 34150 + }, + { + "epoch": 0.2964470794524353, + "grad_norm": 0.2080078125, + "learning_rate": 0.0015112410650980792, + "loss": 0.1152, + "step": 34151 + }, + { + "epoch": 0.29645575993263945, + "grad_norm": 0.1044921875, + "learning_rate": 0.0015112145911579757, + "loss": 0.1064, + "step": 34152 + }, + { + "epoch": 0.29646444041284364, + "grad_norm": 0.474609375, + "learning_rate": 0.0015111881167681738, + "loss": 0.082, + "step": 34153 + }, + { + "epoch": 0.2964731208930478, + "grad_norm": 0.2333984375, + "learning_rate": 0.0015111616419287031, + "loss": 0.1104, + "step": 34154 + }, + { + "epoch": 0.296481801373252, + "grad_norm": 0.46875, + "learning_rate": 0.001511135166639592, + "loss": 0.1465, + "step": 34155 + }, + { + "epoch": 0.2964904818534561, + "grad_norm": 0.267578125, + "learning_rate": 0.0015111086909008697, + "loss": 0.1191, + "step": 34156 + }, + { + "epoch": 0.2964991623336603, + "grad_norm": 0.11474609375, + "learning_rate": 0.001511082214712565, + "loss": 0.0869, + "step": 34157 + }, + { + "epoch": 0.29650784281386444, + "grad_norm": 0.09619140625, + "learning_rate": 0.001511055738074707, + "loss": 0.0659, + "step": 34158 + }, + { + "epoch": 0.29651652329406863, + "grad_norm": 0.380859375, + "learning_rate": 0.0015110292609873245, + "loss": 0.0918, + "step": 34159 + }, + { + "epoch": 0.29652520377427277, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015110027834504469, + "loss": 0.125, + "step": 34160 + }, + { + "epoch": 0.29653388425447696, + "grad_norm": 0.19140625, + "learning_rate": 0.0015109763054641028, + "loss": 0.0879, + "step": 34161 + }, + { + "epoch": 0.2965425647346811, + "grad_norm": 0.1689453125, + "learning_rate": 0.001510949827028321, + "loss": 0.0996, + "step": 34162 + }, + { + "epoch": 0.2965512452148853, + "grad_norm": 0.55859375, + "learning_rate": 0.001510923348143131, + "loss": 0.106, + "step": 34163 + }, + { + "epoch": 0.29655992569508943, + "grad_norm": 0.322265625, + "learning_rate": 0.001510896868808561, + "loss": 0.0967, + "step": 34164 + }, + { + "epoch": 0.2965686061752936, + "grad_norm": 0.1318359375, + "learning_rate": 0.0015108703890246405, + "loss": 0.0928, + "step": 34165 + }, + { + "epoch": 0.29657728665549776, + "grad_norm": 0.5390625, + "learning_rate": 0.0015108439087913983, + "loss": 0.1279, + "step": 34166 + }, + { + "epoch": 0.29658596713570196, + "grad_norm": 0.14453125, + "learning_rate": 0.0015108174281088633, + "loss": 0.0747, + "step": 34167 + }, + { + "epoch": 0.2965946476159061, + "grad_norm": 0.318359375, + "learning_rate": 0.0015107909469770644, + "loss": 0.1016, + "step": 34168 + }, + { + "epoch": 0.2966033280961103, + "grad_norm": 0.2734375, + "learning_rate": 0.0015107644653960313, + "loss": 0.1011, + "step": 34169 + }, + { + "epoch": 0.2966120085763144, + "grad_norm": 0.25, + "learning_rate": 0.0015107379833657916, + "loss": 0.0859, + "step": 34170 + }, + { + "epoch": 0.2966206890565186, + "grad_norm": 0.86328125, + "learning_rate": 0.0015107115008863756, + "loss": 0.0928, + "step": 34171 + }, + { + "epoch": 0.29662936953672275, + "grad_norm": 0.79296875, + "learning_rate": 0.0015106850179578116, + "loss": 0.1289, + "step": 34172 + }, + { + "epoch": 0.29663805001692695, + "grad_norm": 0.376953125, + "learning_rate": 0.0015106585345801282, + "loss": 0.1484, + "step": 34173 + }, + { + "epoch": 0.2966467304971311, + "grad_norm": 0.71484375, + "learning_rate": 0.0015106320507533551, + "loss": 0.1553, + "step": 34174 + }, + { + "epoch": 0.2966554109773353, + "grad_norm": 0.263671875, + "learning_rate": 0.0015106055664775209, + "loss": 0.0908, + "step": 34175 + }, + { + "epoch": 0.2966640914575394, + "grad_norm": 0.58203125, + "learning_rate": 0.0015105790817526545, + "loss": 0.0742, + "step": 34176 + }, + { + "epoch": 0.2966727719377436, + "grad_norm": 0.388671875, + "learning_rate": 0.0015105525965787854, + "loss": 0.0908, + "step": 34177 + }, + { + "epoch": 0.29668145241794774, + "grad_norm": 0.35546875, + "learning_rate": 0.0015105261109559416, + "loss": 0.0684, + "step": 34178 + }, + { + "epoch": 0.29669013289815194, + "grad_norm": 0.1845703125, + "learning_rate": 0.001510499624884153, + "loss": 0.0957, + "step": 34179 + }, + { + "epoch": 0.2966988133783561, + "grad_norm": 0.134765625, + "learning_rate": 0.0015104731383634483, + "loss": 0.0825, + "step": 34180 + }, + { + "epoch": 0.29670749385856027, + "grad_norm": 0.1123046875, + "learning_rate": 0.001510446651393856, + "loss": 0.0977, + "step": 34181 + }, + { + "epoch": 0.2967161743387644, + "grad_norm": 0.8046875, + "learning_rate": 0.0015104201639754055, + "loss": 0.0996, + "step": 34182 + }, + { + "epoch": 0.2967248548189686, + "grad_norm": 0.466796875, + "learning_rate": 0.001510393676108126, + "loss": 0.1089, + "step": 34183 + }, + { + "epoch": 0.29673353529917273, + "grad_norm": 0.30078125, + "learning_rate": 0.0015103671877920455, + "loss": 0.1211, + "step": 34184 + }, + { + "epoch": 0.2967422157793769, + "grad_norm": 0.423828125, + "learning_rate": 0.001510340699027194, + "loss": 0.1309, + "step": 34185 + }, + { + "epoch": 0.29675089625958107, + "grad_norm": 0.216796875, + "learning_rate": 0.0015103142098136, + "loss": 0.0698, + "step": 34186 + }, + { + "epoch": 0.29675957673978526, + "grad_norm": 0.67578125, + "learning_rate": 0.0015102877201512925, + "loss": 0.1143, + "step": 34187 + }, + { + "epoch": 0.2967682572199894, + "grad_norm": 0.0869140625, + "learning_rate": 0.0015102612300403008, + "loss": 0.104, + "step": 34188 + }, + { + "epoch": 0.2967769377001936, + "grad_norm": 0.2412109375, + "learning_rate": 0.0015102347394806532, + "loss": 0.0835, + "step": 34189 + }, + { + "epoch": 0.2967856181803977, + "grad_norm": 0.1806640625, + "learning_rate": 0.0015102082484723795, + "loss": 0.1104, + "step": 34190 + }, + { + "epoch": 0.2967942986606019, + "grad_norm": 0.29296875, + "learning_rate": 0.0015101817570155076, + "loss": 0.1201, + "step": 34191 + }, + { + "epoch": 0.29680297914080606, + "grad_norm": 0.21484375, + "learning_rate": 0.0015101552651100676, + "loss": 0.0928, + "step": 34192 + }, + { + "epoch": 0.29681165962101025, + "grad_norm": 0.318359375, + "learning_rate": 0.0015101287727560878, + "loss": 0.1055, + "step": 34193 + }, + { + "epoch": 0.2968203401012144, + "grad_norm": 0.390625, + "learning_rate": 0.0015101022799535973, + "loss": 0.1504, + "step": 34194 + }, + { + "epoch": 0.2968290205814186, + "grad_norm": 0.15625, + "learning_rate": 0.0015100757867026254, + "loss": 0.1001, + "step": 34195 + }, + { + "epoch": 0.2968377010616227, + "grad_norm": 0.40625, + "learning_rate": 0.0015100492930032001, + "loss": 0.1719, + "step": 34196 + }, + { + "epoch": 0.2968463815418269, + "grad_norm": 0.28125, + "learning_rate": 0.0015100227988553518, + "loss": 0.0815, + "step": 34197 + }, + { + "epoch": 0.29685506202203105, + "grad_norm": 0.6015625, + "learning_rate": 0.0015099963042591086, + "loss": 0.127, + "step": 34198 + }, + { + "epoch": 0.29686374250223524, + "grad_norm": 0.09326171875, + "learning_rate": 0.0015099698092144993, + "loss": 0.1245, + "step": 34199 + }, + { + "epoch": 0.2968724229824394, + "grad_norm": 0.62890625, + "learning_rate": 0.001509943313721553, + "loss": 0.0938, + "step": 34200 + }, + { + "epoch": 0.29688110346264357, + "grad_norm": 0.380859375, + "learning_rate": 0.0015099168177802992, + "loss": 0.0771, + "step": 34201 + }, + { + "epoch": 0.2968897839428477, + "grad_norm": 0.15234375, + "learning_rate": 0.0015098903213907665, + "loss": 0.1143, + "step": 34202 + }, + { + "epoch": 0.2968984644230519, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015098638245529838, + "loss": 0.0967, + "step": 34203 + }, + { + "epoch": 0.29690714490325604, + "grad_norm": 0.38671875, + "learning_rate": 0.0015098373272669802, + "loss": 0.1309, + "step": 34204 + }, + { + "epoch": 0.29691582538346023, + "grad_norm": 0.58203125, + "learning_rate": 0.0015098108295327848, + "loss": 0.0986, + "step": 34205 + }, + { + "epoch": 0.29692450586366437, + "grad_norm": 0.66796875, + "learning_rate": 0.0015097843313504264, + "loss": 0.1084, + "step": 34206 + }, + { + "epoch": 0.29693318634386856, + "grad_norm": 0.1123046875, + "learning_rate": 0.001509757832719934, + "loss": 0.1108, + "step": 34207 + }, + { + "epoch": 0.2969418668240727, + "grad_norm": 0.33203125, + "learning_rate": 0.0015097313336413366, + "loss": 0.1182, + "step": 34208 + }, + { + "epoch": 0.2969505473042769, + "grad_norm": 0.33984375, + "learning_rate": 0.0015097048341146631, + "loss": 0.1016, + "step": 34209 + }, + { + "epoch": 0.29695922778448103, + "grad_norm": 0.51171875, + "learning_rate": 0.0015096783341399425, + "loss": 0.1426, + "step": 34210 + }, + { + "epoch": 0.2969679082646852, + "grad_norm": 0.2451171875, + "learning_rate": 0.001509651833717204, + "loss": 0.0962, + "step": 34211 + }, + { + "epoch": 0.29697658874488936, + "grad_norm": 0.072265625, + "learning_rate": 0.0015096253328464762, + "loss": 0.0752, + "step": 34212 + }, + { + "epoch": 0.29698526922509355, + "grad_norm": 0.16796875, + "learning_rate": 0.0015095988315277885, + "loss": 0.0889, + "step": 34213 + }, + { + "epoch": 0.2969939497052977, + "grad_norm": 0.0869140625, + "learning_rate": 0.0015095723297611694, + "loss": 0.0957, + "step": 34214 + }, + { + "epoch": 0.2970026301855019, + "grad_norm": 0.2255859375, + "learning_rate": 0.0015095458275466486, + "loss": 0.0723, + "step": 34215 + }, + { + "epoch": 0.297011310665706, + "grad_norm": 0.84765625, + "learning_rate": 0.001509519324884254, + "loss": 0.0986, + "step": 34216 + }, + { + "epoch": 0.2970199911459102, + "grad_norm": 0.126953125, + "learning_rate": 0.0015094928217740157, + "loss": 0.1074, + "step": 34217 + }, + { + "epoch": 0.29702867162611435, + "grad_norm": 0.56640625, + "learning_rate": 0.001509466318215962, + "loss": 0.1074, + "step": 34218 + }, + { + "epoch": 0.29703735210631854, + "grad_norm": 0.1025390625, + "learning_rate": 0.0015094398142101223, + "loss": 0.0947, + "step": 34219 + }, + { + "epoch": 0.2970460325865227, + "grad_norm": 0.1787109375, + "learning_rate": 0.0015094133097565251, + "loss": 0.1328, + "step": 34220 + }, + { + "epoch": 0.29705471306672687, + "grad_norm": 0.2578125, + "learning_rate": 0.0015093868048551998, + "loss": 0.1006, + "step": 34221 + }, + { + "epoch": 0.297063393546931, + "grad_norm": 0.6015625, + "learning_rate": 0.0015093602995061752, + "loss": 0.1074, + "step": 34222 + }, + { + "epoch": 0.2970720740271352, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015093337937094804, + "loss": 0.1055, + "step": 34223 + }, + { + "epoch": 0.29708075450733934, + "grad_norm": 0.2099609375, + "learning_rate": 0.0015093072874651442, + "loss": 0.1123, + "step": 34224 + }, + { + "epoch": 0.29708943498754353, + "grad_norm": 0.1689453125, + "learning_rate": 0.0015092807807731959, + "loss": 0.1021, + "step": 34225 + }, + { + "epoch": 0.29709811546774767, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015092542736336644, + "loss": 0.0552, + "step": 34226 + }, + { + "epoch": 0.29710679594795186, + "grad_norm": 0.41015625, + "learning_rate": 0.001509227766046578, + "loss": 0.0908, + "step": 34227 + }, + { + "epoch": 0.297115476428156, + "grad_norm": 0.546875, + "learning_rate": 0.0015092012580119668, + "loss": 0.1113, + "step": 34228 + }, + { + "epoch": 0.2971241569083602, + "grad_norm": 0.5078125, + "learning_rate": 0.001509174749529859, + "loss": 0.1045, + "step": 34229 + }, + { + "epoch": 0.29713283738856433, + "grad_norm": 0.09912109375, + "learning_rate": 0.001509148240600284, + "loss": 0.1416, + "step": 34230 + }, + { + "epoch": 0.2971415178687685, + "grad_norm": 0.41796875, + "learning_rate": 0.0015091217312232705, + "loss": 0.0889, + "step": 34231 + }, + { + "epoch": 0.29715019834897266, + "grad_norm": 0.185546875, + "learning_rate": 0.0015090952213988476, + "loss": 0.0986, + "step": 34232 + }, + { + "epoch": 0.29715887882917685, + "grad_norm": 0.18359375, + "learning_rate": 0.0015090687111270443, + "loss": 0.1484, + "step": 34233 + }, + { + "epoch": 0.297167559309381, + "grad_norm": 0.453125, + "learning_rate": 0.0015090422004078899, + "loss": 0.1416, + "step": 34234 + }, + { + "epoch": 0.2971762397895852, + "grad_norm": 0.3828125, + "learning_rate": 0.0015090156892414127, + "loss": 0.1484, + "step": 34235 + }, + { + "epoch": 0.2971849202697893, + "grad_norm": 0.5, + "learning_rate": 0.001508989177627642, + "loss": 0.1128, + "step": 34236 + }, + { + "epoch": 0.2971936007499935, + "grad_norm": 0.17578125, + "learning_rate": 0.0015089626655666072, + "loss": 0.104, + "step": 34237 + }, + { + "epoch": 0.29720228123019765, + "grad_norm": 0.48046875, + "learning_rate": 0.0015089361530583373, + "loss": 0.0898, + "step": 34238 + }, + { + "epoch": 0.29721096171040184, + "grad_norm": 0.41796875, + "learning_rate": 0.0015089096401028604, + "loss": 0.0898, + "step": 34239 + }, + { + "epoch": 0.297219642190606, + "grad_norm": 0.20703125, + "learning_rate": 0.001508883126700206, + "loss": 0.0879, + "step": 34240 + }, + { + "epoch": 0.2972283226708102, + "grad_norm": 0.14453125, + "learning_rate": 0.0015088566128504035, + "loss": 0.0879, + "step": 34241 + }, + { + "epoch": 0.2972370031510143, + "grad_norm": 0.10302734375, + "learning_rate": 0.0015088300985534814, + "loss": 0.1191, + "step": 34242 + }, + { + "epoch": 0.2972456836312185, + "grad_norm": 0.30859375, + "learning_rate": 0.0015088035838094688, + "loss": 0.1089, + "step": 34243 + }, + { + "epoch": 0.29725436411142264, + "grad_norm": 0.453125, + "learning_rate": 0.0015087770686183948, + "loss": 0.1128, + "step": 34244 + }, + { + "epoch": 0.29726304459162683, + "grad_norm": 0.57421875, + "learning_rate": 0.0015087505529802886, + "loss": 0.0815, + "step": 34245 + }, + { + "epoch": 0.297271725071831, + "grad_norm": 0.283203125, + "learning_rate": 0.0015087240368951787, + "loss": 0.1221, + "step": 34246 + }, + { + "epoch": 0.29728040555203517, + "grad_norm": 0.13671875, + "learning_rate": 0.0015086975203630944, + "loss": 0.0674, + "step": 34247 + }, + { + "epoch": 0.2972890860322393, + "grad_norm": 0.171875, + "learning_rate": 0.0015086710033840646, + "loss": 0.1133, + "step": 34248 + }, + { + "epoch": 0.2972977665124435, + "grad_norm": 0.1962890625, + "learning_rate": 0.0015086444859581183, + "loss": 0.1602, + "step": 34249 + }, + { + "epoch": 0.29730644699264763, + "grad_norm": 0.7890625, + "learning_rate": 0.0015086179680852843, + "loss": 0.0996, + "step": 34250 + }, + { + "epoch": 0.2973151274728518, + "grad_norm": 0.404296875, + "learning_rate": 0.001508591449765592, + "loss": 0.0811, + "step": 34251 + }, + { + "epoch": 0.29732380795305596, + "grad_norm": 0.267578125, + "learning_rate": 0.0015085649309990704, + "loss": 0.0879, + "step": 34252 + }, + { + "epoch": 0.2973324884332601, + "grad_norm": 1.234375, + "learning_rate": 0.0015085384117857483, + "loss": 0.124, + "step": 34253 + }, + { + "epoch": 0.2973411689134643, + "grad_norm": 0.197265625, + "learning_rate": 0.0015085118921256547, + "loss": 0.1211, + "step": 34254 + }, + { + "epoch": 0.29734984939366843, + "grad_norm": 0.45703125, + "learning_rate": 0.0015084853720188185, + "loss": 0.0757, + "step": 34255 + }, + { + "epoch": 0.2973585298738726, + "grad_norm": 0.1865234375, + "learning_rate": 0.0015084588514652688, + "loss": 0.1055, + "step": 34256 + }, + { + "epoch": 0.29736721035407676, + "grad_norm": 0.0966796875, + "learning_rate": 0.001508432330465035, + "loss": 0.1025, + "step": 34257 + }, + { + "epoch": 0.29737589083428095, + "grad_norm": 0.1220703125, + "learning_rate": 0.0015084058090181454, + "loss": 0.1455, + "step": 34258 + }, + { + "epoch": 0.2973845713144851, + "grad_norm": 0.310546875, + "learning_rate": 0.0015083792871246296, + "loss": 0.0825, + "step": 34259 + }, + { + "epoch": 0.2973932517946893, + "grad_norm": 0.2890625, + "learning_rate": 0.001508352764784516, + "loss": 0.1406, + "step": 34260 + }, + { + "epoch": 0.2974019322748934, + "grad_norm": 0.1279296875, + "learning_rate": 0.001508326241997834, + "loss": 0.0957, + "step": 34261 + }, + { + "epoch": 0.2974106127550976, + "grad_norm": 0.0732421875, + "learning_rate": 0.0015082997187646128, + "loss": 0.0981, + "step": 34262 + }, + { + "epoch": 0.29741929323530175, + "grad_norm": 0.2470703125, + "learning_rate": 0.001508273195084881, + "loss": 0.0991, + "step": 34263 + }, + { + "epoch": 0.29742797371550594, + "grad_norm": 0.103515625, + "learning_rate": 0.0015082466709586682, + "loss": 0.1504, + "step": 34264 + }, + { + "epoch": 0.2974366541957101, + "grad_norm": 0.5390625, + "learning_rate": 0.001508220146386002, + "loss": 0.083, + "step": 34265 + }, + { + "epoch": 0.2974453346759143, + "grad_norm": 0.38671875, + "learning_rate": 0.0015081936213669133, + "loss": 0.1123, + "step": 34266 + }, + { + "epoch": 0.2974540151561184, + "grad_norm": 0.15234375, + "learning_rate": 0.00150816709590143, + "loss": 0.0991, + "step": 34267 + }, + { + "epoch": 0.2974626956363226, + "grad_norm": 0.376953125, + "learning_rate": 0.0015081405699895812, + "loss": 0.0879, + "step": 34268 + }, + { + "epoch": 0.29747137611652674, + "grad_norm": 0.08642578125, + "learning_rate": 0.0015081140436313959, + "loss": 0.1211, + "step": 34269 + }, + { + "epoch": 0.29748005659673094, + "grad_norm": 0.1767578125, + "learning_rate": 0.001508087516826903, + "loss": 0.1299, + "step": 34270 + }, + { + "epoch": 0.2974887370769351, + "grad_norm": 0.4296875, + "learning_rate": 0.001508060989576132, + "loss": 0.0977, + "step": 34271 + }, + { + "epoch": 0.29749741755713927, + "grad_norm": 0.67578125, + "learning_rate": 0.0015080344618791117, + "loss": 0.0859, + "step": 34272 + }, + { + "epoch": 0.2975060980373434, + "grad_norm": 0.146484375, + "learning_rate": 0.0015080079337358714, + "loss": 0.123, + "step": 34273 + }, + { + "epoch": 0.2975147785175476, + "grad_norm": 0.462890625, + "learning_rate": 0.0015079814051464392, + "loss": 0.1113, + "step": 34274 + }, + { + "epoch": 0.29752345899775173, + "grad_norm": 0.330078125, + "learning_rate": 0.0015079548761108446, + "loss": 0.0928, + "step": 34275 + }, + { + "epoch": 0.2975321394779559, + "grad_norm": 0.48828125, + "learning_rate": 0.0015079283466291169, + "loss": 0.1069, + "step": 34276 + }, + { + "epoch": 0.29754081995816006, + "grad_norm": 0.486328125, + "learning_rate": 0.001507901816701285, + "loss": 0.1738, + "step": 34277 + }, + { + "epoch": 0.29754950043836426, + "grad_norm": 0.3203125, + "learning_rate": 0.0015078752863273777, + "loss": 0.127, + "step": 34278 + }, + { + "epoch": 0.2975581809185684, + "grad_norm": 1.0625, + "learning_rate": 0.001507848755507424, + "loss": 0.1367, + "step": 34279 + }, + { + "epoch": 0.2975668613987726, + "grad_norm": 0.1162109375, + "learning_rate": 0.001507822224241453, + "loss": 0.1396, + "step": 34280 + }, + { + "epoch": 0.2975755418789767, + "grad_norm": 0.53125, + "learning_rate": 0.001507795692529494, + "loss": 0.1011, + "step": 34281 + }, + { + "epoch": 0.2975842223591809, + "grad_norm": 0.130859375, + "learning_rate": 0.0015077691603715756, + "loss": 0.084, + "step": 34282 + }, + { + "epoch": 0.29759290283938505, + "grad_norm": 0.2216796875, + "learning_rate": 0.0015077426277677268, + "loss": 0.0703, + "step": 34283 + }, + { + "epoch": 0.29760158331958925, + "grad_norm": 0.23046875, + "learning_rate": 0.0015077160947179773, + "loss": 0.0859, + "step": 34284 + }, + { + "epoch": 0.2976102637997934, + "grad_norm": 0.1708984375, + "learning_rate": 0.001507689561222355, + "loss": 0.1055, + "step": 34285 + }, + { + "epoch": 0.2976189442799976, + "grad_norm": 0.1474609375, + "learning_rate": 0.00150766302728089, + "loss": 0.1182, + "step": 34286 + }, + { + "epoch": 0.2976276247602017, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015076364928936106, + "loss": 0.0801, + "step": 34287 + }, + { + "epoch": 0.2976363052404059, + "grad_norm": 0.162109375, + "learning_rate": 0.0015076099580605463, + "loss": 0.1035, + "step": 34288 + }, + { + "epoch": 0.29764498572061004, + "grad_norm": 0.10400390625, + "learning_rate": 0.0015075834227817255, + "loss": 0.0913, + "step": 34289 + }, + { + "epoch": 0.29765366620081424, + "grad_norm": 0.146484375, + "learning_rate": 0.0015075568870571779, + "loss": 0.124, + "step": 34290 + }, + { + "epoch": 0.2976623466810184, + "grad_norm": 0.396484375, + "learning_rate": 0.001507530350886932, + "loss": 0.083, + "step": 34291 + }, + { + "epoch": 0.29767102716122257, + "grad_norm": 0.23046875, + "learning_rate": 0.0015075038142710168, + "loss": 0.1289, + "step": 34292 + }, + { + "epoch": 0.2976797076414267, + "grad_norm": 0.12255859375, + "learning_rate": 0.0015074772772094623, + "loss": 0.1318, + "step": 34293 + }, + { + "epoch": 0.2976883881216309, + "grad_norm": 0.251953125, + "learning_rate": 0.0015074507397022964, + "loss": 0.1172, + "step": 34294 + }, + { + "epoch": 0.29769706860183504, + "grad_norm": 0.353515625, + "learning_rate": 0.001507424201749548, + "loss": 0.0864, + "step": 34295 + }, + { + "epoch": 0.29770574908203923, + "grad_norm": 0.07080078125, + "learning_rate": 0.0015073976633512471, + "loss": 0.0923, + "step": 34296 + }, + { + "epoch": 0.29771442956224337, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015073711245074221, + "loss": 0.1328, + "step": 34297 + }, + { + "epoch": 0.29772311004244756, + "grad_norm": 0.203125, + "learning_rate": 0.0015073445852181021, + "loss": 0.1182, + "step": 34298 + }, + { + "epoch": 0.2977317905226517, + "grad_norm": 0.224609375, + "learning_rate": 0.0015073180454833167, + "loss": 0.1113, + "step": 34299 + }, + { + "epoch": 0.2977404710028559, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015072915053030939, + "loss": 0.082, + "step": 34300 + }, + { + "epoch": 0.29774915148306, + "grad_norm": 0.38671875, + "learning_rate": 0.0015072649646774635, + "loss": 0.1289, + "step": 34301 + }, + { + "epoch": 0.2977578319632642, + "grad_norm": 0.32421875, + "learning_rate": 0.001507238423606454, + "loss": 0.0815, + "step": 34302 + }, + { + "epoch": 0.29776651244346836, + "grad_norm": 0.1884765625, + "learning_rate": 0.0015072118820900948, + "loss": 0.1001, + "step": 34303 + }, + { + "epoch": 0.29777519292367255, + "grad_norm": 0.5078125, + "learning_rate": 0.0015071853401284149, + "loss": 0.0913, + "step": 34304 + }, + { + "epoch": 0.2977838734038767, + "grad_norm": 0.294921875, + "learning_rate": 0.001507158797721443, + "loss": 0.0928, + "step": 34305 + }, + { + "epoch": 0.2977925538840809, + "grad_norm": 0.255859375, + "learning_rate": 0.0015071322548692089, + "loss": 0.1084, + "step": 34306 + }, + { + "epoch": 0.297801234364285, + "grad_norm": 0.12158203125, + "learning_rate": 0.0015071057115717405, + "loss": 0.0583, + "step": 34307 + }, + { + "epoch": 0.2978099148444892, + "grad_norm": 0.55078125, + "learning_rate": 0.0015070791678290672, + "loss": 0.1016, + "step": 34308 + }, + { + "epoch": 0.29781859532469335, + "grad_norm": 0.546875, + "learning_rate": 0.0015070526236412188, + "loss": 0.082, + "step": 34309 + }, + { + "epoch": 0.29782727580489754, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015070260790082234, + "loss": 0.1084, + "step": 34310 + }, + { + "epoch": 0.2978359562851017, + "grad_norm": 0.91015625, + "learning_rate": 0.0015069995339301106, + "loss": 0.1172, + "step": 34311 + }, + { + "epoch": 0.29784463676530587, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015069729884069091, + "loss": 0.127, + "step": 34312 + }, + { + "epoch": 0.29785331724551, + "grad_norm": 0.318359375, + "learning_rate": 0.0015069464424386483, + "loss": 0.123, + "step": 34313 + }, + { + "epoch": 0.2978619977257142, + "grad_norm": 0.1318359375, + "learning_rate": 0.001506919896025357, + "loss": 0.1504, + "step": 34314 + }, + { + "epoch": 0.29787067820591834, + "grad_norm": 0.6875, + "learning_rate": 0.001506893349167064, + "loss": 0.1396, + "step": 34315 + }, + { + "epoch": 0.29787935868612253, + "grad_norm": 0.197265625, + "learning_rate": 0.0015068668018637986, + "loss": 0.0928, + "step": 34316 + }, + { + "epoch": 0.29788803916632667, + "grad_norm": 0.435546875, + "learning_rate": 0.0015068402541155896, + "loss": 0.1074, + "step": 34317 + }, + { + "epoch": 0.29789671964653086, + "grad_norm": 0.1123046875, + "learning_rate": 0.0015068137059224662, + "loss": 0.0879, + "step": 34318 + }, + { + "epoch": 0.297905400126735, + "grad_norm": 0.171875, + "learning_rate": 0.001506787157284458, + "loss": 0.0908, + "step": 34319 + }, + { + "epoch": 0.2979140806069392, + "grad_norm": 0.1435546875, + "learning_rate": 0.001506760608201593, + "loss": 0.0698, + "step": 34320 + }, + { + "epoch": 0.29792276108714333, + "grad_norm": 0.421875, + "learning_rate": 0.0015067340586739006, + "loss": 0.1001, + "step": 34321 + }, + { + "epoch": 0.2979314415673475, + "grad_norm": 0.52734375, + "learning_rate": 0.0015067075087014103, + "loss": 0.2285, + "step": 34322 + }, + { + "epoch": 0.29794012204755166, + "grad_norm": 0.17578125, + "learning_rate": 0.0015066809582841506, + "loss": 0.0986, + "step": 34323 + }, + { + "epoch": 0.29794880252775585, + "grad_norm": 0.181640625, + "learning_rate": 0.0015066544074221512, + "loss": 0.126, + "step": 34324 + }, + { + "epoch": 0.29795748300796, + "grad_norm": 0.416015625, + "learning_rate": 0.0015066278561154398, + "loss": 0.106, + "step": 34325 + }, + { + "epoch": 0.2979661634881642, + "grad_norm": 0.2265625, + "learning_rate": 0.0015066013043640468, + "loss": 0.103, + "step": 34326 + }, + { + "epoch": 0.2979748439683683, + "grad_norm": 0.59375, + "learning_rate": 0.0015065747521680005, + "loss": 0.1001, + "step": 34327 + }, + { + "epoch": 0.2979835244485725, + "grad_norm": 0.328125, + "learning_rate": 0.0015065481995273306, + "loss": 0.1416, + "step": 34328 + }, + { + "epoch": 0.29799220492877665, + "grad_norm": 0.109375, + "learning_rate": 0.0015065216464420653, + "loss": 0.0869, + "step": 34329 + }, + { + "epoch": 0.29800088540898084, + "grad_norm": 0.447265625, + "learning_rate": 0.001506495092912234, + "loss": 0.1196, + "step": 34330 + }, + { + "epoch": 0.298009565889185, + "grad_norm": 0.1328125, + "learning_rate": 0.0015064685389378659, + "loss": 0.0938, + "step": 34331 + }, + { + "epoch": 0.2980182463693892, + "grad_norm": 0.091796875, + "learning_rate": 0.00150644198451899, + "loss": 0.1201, + "step": 34332 + }, + { + "epoch": 0.2980269268495933, + "grad_norm": 0.1083984375, + "learning_rate": 0.001506415429655635, + "loss": 0.1182, + "step": 34333 + }, + { + "epoch": 0.2980356073297975, + "grad_norm": 0.390625, + "learning_rate": 0.0015063888743478302, + "loss": 0.1182, + "step": 34334 + }, + { + "epoch": 0.29804428781000164, + "grad_norm": 0.2177734375, + "learning_rate": 0.001506362318595605, + "loss": 0.0811, + "step": 34335 + }, + { + "epoch": 0.29805296829020583, + "grad_norm": 0.55078125, + "learning_rate": 0.001506335762398988, + "loss": 0.126, + "step": 34336 + }, + { + "epoch": 0.29806164877040997, + "grad_norm": 0.55078125, + "learning_rate": 0.0015063092057580078, + "loss": 0.1143, + "step": 34337 + }, + { + "epoch": 0.29807032925061416, + "grad_norm": 0.1689453125, + "learning_rate": 0.0015062826486726942, + "loss": 0.0957, + "step": 34338 + }, + { + "epoch": 0.2980790097308183, + "grad_norm": 0.6953125, + "learning_rate": 0.0015062560911430762, + "loss": 0.0967, + "step": 34339 + }, + { + "epoch": 0.2980876902110225, + "grad_norm": 0.267578125, + "learning_rate": 0.0015062295331691826, + "loss": 0.1006, + "step": 34340 + }, + { + "epoch": 0.29809637069122663, + "grad_norm": 0.11474609375, + "learning_rate": 0.0015062029747510426, + "loss": 0.1064, + "step": 34341 + }, + { + "epoch": 0.2981050511714308, + "grad_norm": 0.259765625, + "learning_rate": 0.001506176415888685, + "loss": 0.0928, + "step": 34342 + }, + { + "epoch": 0.29811373165163496, + "grad_norm": 0.12255859375, + "learning_rate": 0.001506149856582139, + "loss": 0.0991, + "step": 34343 + }, + { + "epoch": 0.29812241213183915, + "grad_norm": 0.361328125, + "learning_rate": 0.0015061232968314334, + "loss": 0.0977, + "step": 34344 + }, + { + "epoch": 0.2981310926120433, + "grad_norm": 0.142578125, + "learning_rate": 0.0015060967366365976, + "loss": 0.1128, + "step": 34345 + }, + { + "epoch": 0.2981397730922475, + "grad_norm": 0.26953125, + "learning_rate": 0.0015060701759976607, + "loss": 0.0894, + "step": 34346 + }, + { + "epoch": 0.2981484535724516, + "grad_norm": 0.134765625, + "learning_rate": 0.0015060436149146512, + "loss": 0.0845, + "step": 34347 + }, + { + "epoch": 0.2981571340526558, + "grad_norm": 0.15625, + "learning_rate": 0.0015060170533875987, + "loss": 0.0806, + "step": 34348 + }, + { + "epoch": 0.29816581453285995, + "grad_norm": 0.326171875, + "learning_rate": 0.001505990491416532, + "loss": 0.1553, + "step": 34349 + }, + { + "epoch": 0.29817449501306414, + "grad_norm": 0.33984375, + "learning_rate": 0.0015059639290014805, + "loss": 0.0874, + "step": 34350 + }, + { + "epoch": 0.2981831754932683, + "grad_norm": 0.060546875, + "learning_rate": 0.0015059373661424724, + "loss": 0.0684, + "step": 34351 + }, + { + "epoch": 0.2981918559734725, + "grad_norm": 0.3828125, + "learning_rate": 0.0015059108028395377, + "loss": 0.1602, + "step": 34352 + }, + { + "epoch": 0.2982005364536766, + "grad_norm": 0.302734375, + "learning_rate": 0.0015058842390927051, + "loss": 0.0781, + "step": 34353 + }, + { + "epoch": 0.2982092169338808, + "grad_norm": 0.173828125, + "learning_rate": 0.0015058576749020034, + "loss": 0.1416, + "step": 34354 + }, + { + "epoch": 0.29821789741408494, + "grad_norm": 0.12890625, + "learning_rate": 0.0015058311102674618, + "loss": 0.0884, + "step": 34355 + }, + { + "epoch": 0.29822657789428914, + "grad_norm": 0.236328125, + "learning_rate": 0.00150580454518911, + "loss": 0.0864, + "step": 34356 + }, + { + "epoch": 0.2982352583744933, + "grad_norm": 0.1943359375, + "learning_rate": 0.0015057779796669754, + "loss": 0.1709, + "step": 34357 + }, + { + "epoch": 0.29824393885469747, + "grad_norm": 0.66796875, + "learning_rate": 0.0015057514137010886, + "loss": 0.1064, + "step": 34358 + }, + { + "epoch": 0.2982526193349016, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015057248472914785, + "loss": 0.1025, + "step": 34359 + }, + { + "epoch": 0.2982612998151058, + "grad_norm": 0.203125, + "learning_rate": 0.001505698280438173, + "loss": 0.1152, + "step": 34360 + }, + { + "epoch": 0.29826998029530993, + "grad_norm": 0.30078125, + "learning_rate": 0.0015056717131412029, + "loss": 0.124, + "step": 34361 + }, + { + "epoch": 0.2982786607755141, + "grad_norm": 0.359375, + "learning_rate": 0.0015056451454005955, + "loss": 0.1191, + "step": 34362 + }, + { + "epoch": 0.29828734125571826, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015056185772163813, + "loss": 0.0884, + "step": 34363 + }, + { + "epoch": 0.29829602173592246, + "grad_norm": 0.375, + "learning_rate": 0.0015055920085885884, + "loss": 0.0947, + "step": 34364 + }, + { + "epoch": 0.2983047022161266, + "grad_norm": 0.259765625, + "learning_rate": 0.001505565439517246, + "loss": 0.0542, + "step": 34365 + }, + { + "epoch": 0.2983133826963308, + "grad_norm": 0.1826171875, + "learning_rate": 0.0015055388700023836, + "loss": 0.1201, + "step": 34366 + }, + { + "epoch": 0.2983220631765349, + "grad_norm": 0.333984375, + "learning_rate": 0.00150551230004403, + "loss": 0.1177, + "step": 34367 + }, + { + "epoch": 0.2983307436567391, + "grad_norm": 0.341796875, + "learning_rate": 0.001505485729642214, + "loss": 0.1084, + "step": 34368 + }, + { + "epoch": 0.29833942413694325, + "grad_norm": 0.578125, + "learning_rate": 0.001505459158796965, + "loss": 0.1064, + "step": 34369 + }, + { + "epoch": 0.29834810461714745, + "grad_norm": 0.384765625, + "learning_rate": 0.0015054325875083119, + "loss": 0.1104, + "step": 34370 + }, + { + "epoch": 0.2983567850973516, + "grad_norm": 0.2216796875, + "learning_rate": 0.001505406015776284, + "loss": 0.0918, + "step": 34371 + }, + { + "epoch": 0.2983654655775558, + "grad_norm": 0.08349609375, + "learning_rate": 0.0015053794436009102, + "loss": 0.1069, + "step": 34372 + }, + { + "epoch": 0.2983741460577599, + "grad_norm": 0.28125, + "learning_rate": 0.0015053528709822195, + "loss": 0.0928, + "step": 34373 + }, + { + "epoch": 0.2983828265379641, + "grad_norm": 0.53125, + "learning_rate": 0.001505326297920241, + "loss": 0.1001, + "step": 34374 + }, + { + "epoch": 0.29839150701816824, + "grad_norm": 0.1279296875, + "learning_rate": 0.0015052997244150033, + "loss": 0.1045, + "step": 34375 + }, + { + "epoch": 0.2984001874983724, + "grad_norm": 0.08642578125, + "learning_rate": 0.0015052731504665365, + "loss": 0.0947, + "step": 34376 + }, + { + "epoch": 0.2984088679785766, + "grad_norm": 0.09716796875, + "learning_rate": 0.0015052465760748688, + "loss": 0.0742, + "step": 34377 + }, + { + "epoch": 0.2984175484587807, + "grad_norm": 0.8203125, + "learning_rate": 0.0015052200012400294, + "loss": 0.1162, + "step": 34378 + }, + { + "epoch": 0.2984262289389849, + "grad_norm": 0.169921875, + "learning_rate": 0.001505193425962048, + "loss": 0.1021, + "step": 34379 + }, + { + "epoch": 0.29843490941918904, + "grad_norm": 0.11181640625, + "learning_rate": 0.0015051668502409528, + "loss": 0.0977, + "step": 34380 + }, + { + "epoch": 0.29844358989939324, + "grad_norm": 0.30859375, + "learning_rate": 0.0015051402740767733, + "loss": 0.1494, + "step": 34381 + }, + { + "epoch": 0.2984522703795974, + "grad_norm": 0.27734375, + "learning_rate": 0.0015051136974695383, + "loss": 0.0874, + "step": 34382 + }, + { + "epoch": 0.29846095085980157, + "grad_norm": 0.494140625, + "learning_rate": 0.0015050871204192774, + "loss": 0.0728, + "step": 34383 + }, + { + "epoch": 0.2984696313400057, + "grad_norm": 0.59765625, + "learning_rate": 0.001505060542926019, + "loss": 0.1328, + "step": 34384 + }, + { + "epoch": 0.2984783118202099, + "grad_norm": 0.427734375, + "learning_rate": 0.0015050339649897923, + "loss": 0.1113, + "step": 34385 + }, + { + "epoch": 0.29848699230041403, + "grad_norm": 0.2890625, + "learning_rate": 0.0015050073866106268, + "loss": 0.1191, + "step": 34386 + }, + { + "epoch": 0.2984956727806182, + "grad_norm": 0.08740234375, + "learning_rate": 0.0015049808077885515, + "loss": 0.0723, + "step": 34387 + }, + { + "epoch": 0.29850435326082236, + "grad_norm": 0.1904296875, + "learning_rate": 0.0015049542285235952, + "loss": 0.1074, + "step": 34388 + }, + { + "epoch": 0.29851303374102656, + "grad_norm": 0.2041015625, + "learning_rate": 0.0015049276488157868, + "loss": 0.0869, + "step": 34389 + }, + { + "epoch": 0.2985217142212307, + "grad_norm": 0.419921875, + "learning_rate": 0.0015049010686651557, + "loss": 0.1006, + "step": 34390 + }, + { + "epoch": 0.2985303947014349, + "grad_norm": 0.296875, + "learning_rate": 0.0015048744880717308, + "loss": 0.1064, + "step": 34391 + }, + { + "epoch": 0.298539075181639, + "grad_norm": 0.240234375, + "learning_rate": 0.0015048479070355414, + "loss": 0.1079, + "step": 34392 + }, + { + "epoch": 0.2985477556618432, + "grad_norm": 0.306640625, + "learning_rate": 0.0015048213255566168, + "loss": 0.2598, + "step": 34393 + }, + { + "epoch": 0.29855643614204735, + "grad_norm": 0.1396484375, + "learning_rate": 0.001504794743634985, + "loss": 0.0957, + "step": 34394 + }, + { + "epoch": 0.29856511662225155, + "grad_norm": 0.40234375, + "learning_rate": 0.001504768161270676, + "loss": 0.1045, + "step": 34395 + }, + { + "epoch": 0.2985737971024557, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015047415784637185, + "loss": 0.0942, + "step": 34396 + }, + { + "epoch": 0.2985824775826599, + "grad_norm": 0.6484375, + "learning_rate": 0.0015047149952141419, + "loss": 0.0786, + "step": 34397 + }, + { + "epoch": 0.298591158062864, + "grad_norm": 0.1708984375, + "learning_rate": 0.001504688411521975, + "loss": 0.0923, + "step": 34398 + }, + { + "epoch": 0.2985998385430682, + "grad_norm": 0.390625, + "learning_rate": 0.0015046618273872468, + "loss": 0.0918, + "step": 34399 + }, + { + "epoch": 0.29860851902327235, + "grad_norm": 0.2021484375, + "learning_rate": 0.0015046352428099868, + "loss": 0.1465, + "step": 34400 + }, + { + "epoch": 0.29861719950347654, + "grad_norm": 0.126953125, + "learning_rate": 0.0015046086577902235, + "loss": 0.1299, + "step": 34401 + }, + { + "epoch": 0.2986258799836807, + "grad_norm": 0.1220703125, + "learning_rate": 0.0015045820723279865, + "loss": 0.1182, + "step": 34402 + }, + { + "epoch": 0.29863456046388487, + "grad_norm": 0.515625, + "learning_rate": 0.0015045554864233043, + "loss": 0.1045, + "step": 34403 + }, + { + "epoch": 0.298643240944089, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015045289000762063, + "loss": 0.1157, + "step": 34404 + }, + { + "epoch": 0.2986519214242932, + "grad_norm": 0.494140625, + "learning_rate": 0.0015045023132867219, + "loss": 0.2246, + "step": 34405 + }, + { + "epoch": 0.29866060190449734, + "grad_norm": 0.1875, + "learning_rate": 0.0015044757260548798, + "loss": 0.125, + "step": 34406 + }, + { + "epoch": 0.29866928238470153, + "grad_norm": 0.259765625, + "learning_rate": 0.0015044491383807085, + "loss": 0.1074, + "step": 34407 + }, + { + "epoch": 0.29867796286490567, + "grad_norm": 0.361328125, + "learning_rate": 0.0015044225502642385, + "loss": 0.0913, + "step": 34408 + }, + { + "epoch": 0.29868664334510986, + "grad_norm": 0.2412109375, + "learning_rate": 0.001504395961705498, + "loss": 0.0791, + "step": 34409 + }, + { + "epoch": 0.298695323825314, + "grad_norm": 0.90234375, + "learning_rate": 0.0015043693727045159, + "loss": 0.0938, + "step": 34410 + }, + { + "epoch": 0.2987040043055182, + "grad_norm": 0.640625, + "learning_rate": 0.0015043427832613218, + "loss": 0.1113, + "step": 34411 + }, + { + "epoch": 0.2987126847857223, + "grad_norm": 0.21484375, + "learning_rate": 0.0015043161933759438, + "loss": 0.1455, + "step": 34412 + }, + { + "epoch": 0.2987213652659265, + "grad_norm": 0.2138671875, + "learning_rate": 0.0015042896030484124, + "loss": 0.1143, + "step": 34413 + }, + { + "epoch": 0.29873004574613066, + "grad_norm": 0.22265625, + "learning_rate": 0.0015042630122787555, + "loss": 0.0967, + "step": 34414 + }, + { + "epoch": 0.29873872622633485, + "grad_norm": 0.1435546875, + "learning_rate": 0.001504236421067003, + "loss": 0.1377, + "step": 34415 + }, + { + "epoch": 0.298747406706539, + "grad_norm": 0.48828125, + "learning_rate": 0.0015042098294131834, + "loss": 0.0986, + "step": 34416 + }, + { + "epoch": 0.2987560871867432, + "grad_norm": 0.486328125, + "learning_rate": 0.0015041832373173262, + "loss": 0.1187, + "step": 34417 + }, + { + "epoch": 0.2987647676669473, + "grad_norm": 0.1513671875, + "learning_rate": 0.00150415664477946, + "loss": 0.0908, + "step": 34418 + }, + { + "epoch": 0.2987734481471515, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015041300517996145, + "loss": 0.0664, + "step": 34419 + }, + { + "epoch": 0.29878212862735565, + "grad_norm": 0.79296875, + "learning_rate": 0.0015041034583778182, + "loss": 0.2451, + "step": 34420 + }, + { + "epoch": 0.29879080910755984, + "grad_norm": 0.2119140625, + "learning_rate": 0.0015040768645141005, + "loss": 0.0986, + "step": 34421 + }, + { + "epoch": 0.298799489587764, + "grad_norm": 0.224609375, + "learning_rate": 0.0015040502702084908, + "loss": 0.0938, + "step": 34422 + }, + { + "epoch": 0.29880817006796817, + "grad_norm": 0.287109375, + "learning_rate": 0.0015040236754610171, + "loss": 0.125, + "step": 34423 + }, + { + "epoch": 0.2988168505481723, + "grad_norm": 0.8203125, + "learning_rate": 0.0015039970802717098, + "loss": 0.1396, + "step": 34424 + }, + { + "epoch": 0.2988255310283765, + "grad_norm": 0.1650390625, + "learning_rate": 0.0015039704846405966, + "loss": 0.0962, + "step": 34425 + }, + { + "epoch": 0.29883421150858064, + "grad_norm": 0.58984375, + "learning_rate": 0.0015039438885677081, + "loss": 0.1006, + "step": 34426 + }, + { + "epoch": 0.29884289198878483, + "grad_norm": 0.40234375, + "learning_rate": 0.0015039172920530723, + "loss": 0.0996, + "step": 34427 + }, + { + "epoch": 0.29885157246898897, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015038906950967186, + "loss": 0.105, + "step": 34428 + }, + { + "epoch": 0.29886025294919316, + "grad_norm": 0.58984375, + "learning_rate": 0.0015038640976986763, + "loss": 0.1143, + "step": 34429 + }, + { + "epoch": 0.2988689334293973, + "grad_norm": 0.296875, + "learning_rate": 0.0015038374998589741, + "loss": 0.123, + "step": 34430 + }, + { + "epoch": 0.2988776139096015, + "grad_norm": 0.373046875, + "learning_rate": 0.0015038109015776415, + "loss": 0.1123, + "step": 34431 + }, + { + "epoch": 0.29888629438980563, + "grad_norm": 0.099609375, + "learning_rate": 0.0015037843028547072, + "loss": 0.1157, + "step": 34432 + }, + { + "epoch": 0.2988949748700098, + "grad_norm": 0.0908203125, + "learning_rate": 0.0015037577036902004, + "loss": 0.0957, + "step": 34433 + }, + { + "epoch": 0.29890365535021396, + "grad_norm": 0.11767578125, + "learning_rate": 0.00150373110408415, + "loss": 0.1299, + "step": 34434 + }, + { + "epoch": 0.29891233583041815, + "grad_norm": 0.251953125, + "learning_rate": 0.0015037045040365857, + "loss": 0.124, + "step": 34435 + }, + { + "epoch": 0.2989210163106223, + "grad_norm": 0.64453125, + "learning_rate": 0.0015036779035475364, + "loss": 0.1138, + "step": 34436 + }, + { + "epoch": 0.2989296967908265, + "grad_norm": 0.1494140625, + "learning_rate": 0.0015036513026170308, + "loss": 0.0938, + "step": 34437 + }, + { + "epoch": 0.2989383772710306, + "grad_norm": 0.10302734375, + "learning_rate": 0.0015036247012450979, + "loss": 0.0957, + "step": 34438 + }, + { + "epoch": 0.2989470577512348, + "grad_norm": 0.283203125, + "learning_rate": 0.0015035980994317676, + "loss": 0.1396, + "step": 34439 + }, + { + "epoch": 0.29895573823143895, + "grad_norm": 0.4453125, + "learning_rate": 0.0015035714971770684, + "loss": 0.064, + "step": 34440 + }, + { + "epoch": 0.29896441871164314, + "grad_norm": 0.10693359375, + "learning_rate": 0.0015035448944810293, + "loss": 0.0938, + "step": 34441 + }, + { + "epoch": 0.2989730991918473, + "grad_norm": 0.45703125, + "learning_rate": 0.0015035182913436793, + "loss": 0.0747, + "step": 34442 + }, + { + "epoch": 0.2989817796720515, + "grad_norm": 0.1171875, + "learning_rate": 0.001503491687765048, + "loss": 0.1162, + "step": 34443 + }, + { + "epoch": 0.2989904601522556, + "grad_norm": 0.1474609375, + "learning_rate": 0.0015034650837451643, + "loss": 0.0864, + "step": 34444 + }, + { + "epoch": 0.2989991406324598, + "grad_norm": 0.447265625, + "learning_rate": 0.0015034384792840575, + "loss": 0.0952, + "step": 34445 + }, + { + "epoch": 0.29900782111266394, + "grad_norm": 1.2890625, + "learning_rate": 0.0015034118743817559, + "loss": 0.1045, + "step": 34446 + }, + { + "epoch": 0.29901650159286813, + "grad_norm": 0.384765625, + "learning_rate": 0.0015033852690382894, + "loss": 0.1318, + "step": 34447 + }, + { + "epoch": 0.29902518207307227, + "grad_norm": 0.271484375, + "learning_rate": 0.001503358663253687, + "loss": 0.0762, + "step": 34448 + }, + { + "epoch": 0.29903386255327646, + "grad_norm": 0.55859375, + "learning_rate": 0.0015033320570279776, + "loss": 0.1147, + "step": 34449 + }, + { + "epoch": 0.2990425430334806, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015033054503611903, + "loss": 0.0825, + "step": 34450 + }, + { + "epoch": 0.2990512235136848, + "grad_norm": 0.16796875, + "learning_rate": 0.0015032788432533537, + "loss": 0.1123, + "step": 34451 + }, + { + "epoch": 0.29905990399388893, + "grad_norm": 0.56640625, + "learning_rate": 0.0015032522357044979, + "loss": 0.1396, + "step": 34452 + }, + { + "epoch": 0.2990685844740931, + "grad_norm": 0.255859375, + "learning_rate": 0.0015032256277146515, + "loss": 0.0991, + "step": 34453 + }, + { + "epoch": 0.29907726495429726, + "grad_norm": 0.470703125, + "learning_rate": 0.0015031990192838436, + "loss": 0.0654, + "step": 34454 + }, + { + "epoch": 0.29908594543450145, + "grad_norm": 0.1630859375, + "learning_rate": 0.0015031724104121033, + "loss": 0.125, + "step": 34455 + }, + { + "epoch": 0.2990946259147056, + "grad_norm": 0.55078125, + "learning_rate": 0.0015031458010994596, + "loss": 0.0981, + "step": 34456 + }, + { + "epoch": 0.2991033063949098, + "grad_norm": 0.37109375, + "learning_rate": 0.0015031191913459417, + "loss": 0.0825, + "step": 34457 + }, + { + "epoch": 0.2991119868751139, + "grad_norm": 0.2236328125, + "learning_rate": 0.001503092581151579, + "loss": 0.124, + "step": 34458 + }, + { + "epoch": 0.2991206673553181, + "grad_norm": 0.384765625, + "learning_rate": 0.0015030659705164001, + "loss": 0.1084, + "step": 34459 + }, + { + "epoch": 0.29912934783552225, + "grad_norm": 0.1591796875, + "learning_rate": 0.0015030393594404343, + "loss": 0.103, + "step": 34460 + }, + { + "epoch": 0.29913802831572645, + "grad_norm": 1.265625, + "learning_rate": 0.0015030127479237108, + "loss": 0.3125, + "step": 34461 + }, + { + "epoch": 0.2991467087959306, + "grad_norm": 0.9609375, + "learning_rate": 0.0015029861359662584, + "loss": 0.1348, + "step": 34462 + }, + { + "epoch": 0.2991553892761348, + "grad_norm": 0.73828125, + "learning_rate": 0.0015029595235681063, + "loss": 0.0889, + "step": 34463 + }, + { + "epoch": 0.2991640697563389, + "grad_norm": 0.08544921875, + "learning_rate": 0.001502932910729284, + "loss": 0.0962, + "step": 34464 + }, + { + "epoch": 0.2991727502365431, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015029062974498205, + "loss": 0.124, + "step": 34465 + }, + { + "epoch": 0.29918143071674724, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015028796837297444, + "loss": 0.1094, + "step": 34466 + }, + { + "epoch": 0.29919011119695144, + "grad_norm": 0.095703125, + "learning_rate": 0.0015028530695690855, + "loss": 0.0767, + "step": 34467 + }, + { + "epoch": 0.2991987916771556, + "grad_norm": 0.08984375, + "learning_rate": 0.001502826454967872, + "loss": 0.1074, + "step": 34468 + }, + { + "epoch": 0.29920747215735977, + "grad_norm": 0.18359375, + "learning_rate": 0.001502799839926134, + "loss": 0.0815, + "step": 34469 + }, + { + "epoch": 0.2992161526375639, + "grad_norm": 0.232421875, + "learning_rate": 0.0015027732244439, + "loss": 0.1211, + "step": 34470 + }, + { + "epoch": 0.2992248331177681, + "grad_norm": 0.1298828125, + "learning_rate": 0.0015027466085211993, + "loss": 0.0879, + "step": 34471 + }, + { + "epoch": 0.29923351359797223, + "grad_norm": 0.10205078125, + "learning_rate": 0.0015027199921580606, + "loss": 0.1113, + "step": 34472 + }, + { + "epoch": 0.2992421940781764, + "grad_norm": 0.2890625, + "learning_rate": 0.0015026933753545138, + "loss": 0.1016, + "step": 34473 + }, + { + "epoch": 0.29925087455838056, + "grad_norm": 0.12353515625, + "learning_rate": 0.0015026667581105875, + "loss": 0.1094, + "step": 34474 + }, + { + "epoch": 0.29925955503858476, + "grad_norm": 0.53515625, + "learning_rate": 0.0015026401404263107, + "loss": 0.1143, + "step": 34475 + }, + { + "epoch": 0.2992682355187889, + "grad_norm": 0.345703125, + "learning_rate": 0.001502613522301713, + "loss": 0.1152, + "step": 34476 + }, + { + "epoch": 0.2992769159989931, + "grad_norm": 0.73046875, + "learning_rate": 0.001502586903736823, + "loss": 0.0889, + "step": 34477 + }, + { + "epoch": 0.2992855964791972, + "grad_norm": 0.1376953125, + "learning_rate": 0.0015025602847316699, + "loss": 0.0869, + "step": 34478 + }, + { + "epoch": 0.2992942769594014, + "grad_norm": 0.26953125, + "learning_rate": 0.0015025336652862832, + "loss": 0.1455, + "step": 34479 + }, + { + "epoch": 0.29930295743960555, + "grad_norm": 0.11279296875, + "learning_rate": 0.0015025070454006916, + "loss": 0.1162, + "step": 34480 + }, + { + "epoch": 0.29931163791980975, + "grad_norm": 0.078125, + "learning_rate": 0.0015024804250749243, + "loss": 0.0708, + "step": 34481 + }, + { + "epoch": 0.2993203184000139, + "grad_norm": 0.07666015625, + "learning_rate": 0.0015024538043090104, + "loss": 0.0996, + "step": 34482 + }, + { + "epoch": 0.2993289988802181, + "grad_norm": 0.38671875, + "learning_rate": 0.001502427183102979, + "loss": 0.1328, + "step": 34483 + }, + { + "epoch": 0.2993376793604222, + "grad_norm": 0.26171875, + "learning_rate": 0.0015024005614568594, + "loss": 0.1221, + "step": 34484 + }, + { + "epoch": 0.2993463598406264, + "grad_norm": 0.87109375, + "learning_rate": 0.0015023739393706806, + "loss": 0.1113, + "step": 34485 + }, + { + "epoch": 0.29935504032083055, + "grad_norm": 0.1953125, + "learning_rate": 0.001502347316844472, + "loss": 0.0869, + "step": 34486 + }, + { + "epoch": 0.29936372080103474, + "grad_norm": 0.0908203125, + "learning_rate": 0.001502320693878262, + "loss": 0.0967, + "step": 34487 + }, + { + "epoch": 0.2993724012812389, + "grad_norm": 0.3203125, + "learning_rate": 0.0015022940704720804, + "loss": 0.0767, + "step": 34488 + }, + { + "epoch": 0.29938108176144307, + "grad_norm": 0.091796875, + "learning_rate": 0.001502267446625956, + "loss": 0.0898, + "step": 34489 + }, + { + "epoch": 0.2993897622416472, + "grad_norm": 0.171875, + "learning_rate": 0.0015022408223399175, + "loss": 0.085, + "step": 34490 + }, + { + "epoch": 0.2993984427218514, + "grad_norm": 0.08349609375, + "learning_rate": 0.0015022141976139948, + "loss": 0.0996, + "step": 34491 + }, + { + "epoch": 0.29940712320205554, + "grad_norm": 0.39453125, + "learning_rate": 0.001502187572448217, + "loss": 0.1416, + "step": 34492 + }, + { + "epoch": 0.29941580368225973, + "grad_norm": 0.1005859375, + "learning_rate": 0.0015021609468426128, + "loss": 0.0986, + "step": 34493 + }, + { + "epoch": 0.29942448416246387, + "grad_norm": 0.6875, + "learning_rate": 0.0015021343207972109, + "loss": 0.1172, + "step": 34494 + }, + { + "epoch": 0.29943316464266806, + "grad_norm": 0.494140625, + "learning_rate": 0.0015021076943120418, + "loss": 0.1309, + "step": 34495 + }, + { + "epoch": 0.2994418451228722, + "grad_norm": 0.294921875, + "learning_rate": 0.001502081067387133, + "loss": 0.1309, + "step": 34496 + }, + { + "epoch": 0.2994505256030764, + "grad_norm": 0.26171875, + "learning_rate": 0.0015020544400225148, + "loss": 0.1299, + "step": 34497 + }, + { + "epoch": 0.2994592060832805, + "grad_norm": 0.279296875, + "learning_rate": 0.0015020278122182158, + "loss": 0.1309, + "step": 34498 + }, + { + "epoch": 0.29946788656348466, + "grad_norm": 0.08056640625, + "learning_rate": 0.0015020011839742653, + "loss": 0.0991, + "step": 34499 + }, + { + "epoch": 0.29947656704368886, + "grad_norm": 0.12158203125, + "learning_rate": 0.0015019745552906922, + "loss": 0.0801, + "step": 34500 + }, + { + "epoch": 0.299485247523893, + "grad_norm": 0.37109375, + "learning_rate": 0.001501947926167526, + "loss": 0.1504, + "step": 34501 + }, + { + "epoch": 0.2994939280040972, + "grad_norm": 0.5703125, + "learning_rate": 0.0015019212966047954, + "loss": 0.0913, + "step": 34502 + }, + { + "epoch": 0.2995026084843013, + "grad_norm": 0.138671875, + "learning_rate": 0.0015018946666025294, + "loss": 0.1104, + "step": 34503 + }, + { + "epoch": 0.2995112889645055, + "grad_norm": 0.0849609375, + "learning_rate": 0.001501868036160758, + "loss": 0.1055, + "step": 34504 + }, + { + "epoch": 0.29951996944470966, + "grad_norm": 0.7578125, + "learning_rate": 0.0015018414052795095, + "loss": 0.1216, + "step": 34505 + }, + { + "epoch": 0.29952864992491385, + "grad_norm": 0.6328125, + "learning_rate": 0.0015018147739588135, + "loss": 0.0713, + "step": 34506 + }, + { + "epoch": 0.299537330405118, + "grad_norm": 0.083984375, + "learning_rate": 0.0015017881421986988, + "loss": 0.0742, + "step": 34507 + }, + { + "epoch": 0.2995460108853222, + "grad_norm": 0.51953125, + "learning_rate": 0.0015017615099991945, + "loss": 0.1299, + "step": 34508 + }, + { + "epoch": 0.2995546913655263, + "grad_norm": 0.6171875, + "learning_rate": 0.0015017348773603298, + "loss": 0.1162, + "step": 34509 + }, + { + "epoch": 0.2995633718457305, + "grad_norm": 0.1669921875, + "learning_rate": 0.0015017082442821338, + "loss": 0.1055, + "step": 34510 + }, + { + "epoch": 0.29957205232593465, + "grad_norm": 0.13671875, + "learning_rate": 0.0015016816107646359, + "loss": 0.1084, + "step": 34511 + }, + { + "epoch": 0.29958073280613884, + "grad_norm": 0.515625, + "learning_rate": 0.001501654976807865, + "loss": 0.1562, + "step": 34512 + }, + { + "epoch": 0.299589413286343, + "grad_norm": 0.412109375, + "learning_rate": 0.0015016283424118503, + "loss": 0.1113, + "step": 34513 + }, + { + "epoch": 0.29959809376654717, + "grad_norm": 0.357421875, + "learning_rate": 0.0015016017075766208, + "loss": 0.1357, + "step": 34514 + }, + { + "epoch": 0.2996067742467513, + "grad_norm": 0.10400390625, + "learning_rate": 0.0015015750723022057, + "loss": 0.0996, + "step": 34515 + }, + { + "epoch": 0.2996154547269555, + "grad_norm": 0.08740234375, + "learning_rate": 0.0015015484365886343, + "loss": 0.0967, + "step": 34516 + }, + { + "epoch": 0.29962413520715964, + "grad_norm": 0.2216796875, + "learning_rate": 0.0015015218004359352, + "loss": 0.0879, + "step": 34517 + }, + { + "epoch": 0.29963281568736383, + "grad_norm": 0.169921875, + "learning_rate": 0.0015014951638441384, + "loss": 0.1514, + "step": 34518 + }, + { + "epoch": 0.29964149616756797, + "grad_norm": 0.15234375, + "learning_rate": 0.001501468526813272, + "loss": 0.1016, + "step": 34519 + }, + { + "epoch": 0.29965017664777216, + "grad_norm": 0.62109375, + "learning_rate": 0.001501441889343366, + "loss": 0.1196, + "step": 34520 + }, + { + "epoch": 0.2996588571279763, + "grad_norm": 0.39453125, + "learning_rate": 0.001501415251434449, + "loss": 0.1201, + "step": 34521 + }, + { + "epoch": 0.2996675376081805, + "grad_norm": 0.64453125, + "learning_rate": 0.0015013886130865503, + "loss": 0.1045, + "step": 34522 + }, + { + "epoch": 0.2996762180883846, + "grad_norm": 0.11962890625, + "learning_rate": 0.0015013619742996994, + "loss": 0.1299, + "step": 34523 + }, + { + "epoch": 0.2996848985685888, + "grad_norm": 0.1103515625, + "learning_rate": 0.0015013353350739242, + "loss": 0.0923, + "step": 34524 + }, + { + "epoch": 0.29969357904879296, + "grad_norm": 0.390625, + "learning_rate": 0.0015013086954092556, + "loss": 0.1357, + "step": 34525 + }, + { + "epoch": 0.29970225952899715, + "grad_norm": 0.2490234375, + "learning_rate": 0.0015012820553057216, + "loss": 0.1006, + "step": 34526 + }, + { + "epoch": 0.2997109400092013, + "grad_norm": 0.396484375, + "learning_rate": 0.0015012554147633517, + "loss": 0.1025, + "step": 34527 + }, + { + "epoch": 0.2997196204894055, + "grad_norm": 0.3671875, + "learning_rate": 0.0015012287737821747, + "loss": 0.0835, + "step": 34528 + }, + { + "epoch": 0.2997283009696096, + "grad_norm": 0.84375, + "learning_rate": 0.0015012021323622196, + "loss": 0.123, + "step": 34529 + }, + { + "epoch": 0.2997369814498138, + "grad_norm": 0.1416015625, + "learning_rate": 0.0015011754905035163, + "loss": 0.1025, + "step": 34530 + }, + { + "epoch": 0.29974566193001795, + "grad_norm": 0.326171875, + "learning_rate": 0.0015011488482060936, + "loss": 0.0938, + "step": 34531 + }, + { + "epoch": 0.29975434241022214, + "grad_norm": 0.2265625, + "learning_rate": 0.0015011222054699804, + "loss": 0.0986, + "step": 34532 + }, + { + "epoch": 0.2997630228904263, + "grad_norm": 0.1259765625, + "learning_rate": 0.0015010955622952059, + "loss": 0.0723, + "step": 34533 + }, + { + "epoch": 0.29977170337063047, + "grad_norm": 0.67578125, + "learning_rate": 0.0015010689186817994, + "loss": 0.1216, + "step": 34534 + }, + { + "epoch": 0.2997803838508346, + "grad_norm": 0.185546875, + "learning_rate": 0.00150104227462979, + "loss": 0.1055, + "step": 34535 + }, + { + "epoch": 0.2997890643310388, + "grad_norm": 0.26953125, + "learning_rate": 0.0015010156301392067, + "loss": 0.1367, + "step": 34536 + }, + { + "epoch": 0.29979774481124294, + "grad_norm": 0.2314453125, + "learning_rate": 0.0015009889852100789, + "loss": 0.103, + "step": 34537 + }, + { + "epoch": 0.29980642529144713, + "grad_norm": 0.08837890625, + "learning_rate": 0.0015009623398424354, + "loss": 0.0938, + "step": 34538 + }, + { + "epoch": 0.29981510577165127, + "grad_norm": 0.55859375, + "learning_rate": 0.0015009356940363055, + "loss": 0.0918, + "step": 34539 + }, + { + "epoch": 0.29982378625185546, + "grad_norm": 0.138671875, + "learning_rate": 0.0015009090477917184, + "loss": 0.1084, + "step": 34540 + }, + { + "epoch": 0.2998324667320596, + "grad_norm": 0.57421875, + "learning_rate": 0.001500882401108703, + "loss": 0.0918, + "step": 34541 + }, + { + "epoch": 0.2998411472122638, + "grad_norm": 0.58203125, + "learning_rate": 0.001500855753987289, + "loss": 0.1533, + "step": 34542 + }, + { + "epoch": 0.29984982769246793, + "grad_norm": 0.38671875, + "learning_rate": 0.001500829106427505, + "loss": 0.1016, + "step": 34543 + }, + { + "epoch": 0.2998585081726721, + "grad_norm": 1.53125, + "learning_rate": 0.0015008024584293803, + "loss": 0.1357, + "step": 34544 + }, + { + "epoch": 0.29986718865287626, + "grad_norm": 0.26171875, + "learning_rate": 0.0015007758099929442, + "loss": 0.0962, + "step": 34545 + }, + { + "epoch": 0.29987586913308045, + "grad_norm": 0.765625, + "learning_rate": 0.0015007491611182255, + "loss": 0.1123, + "step": 34546 + }, + { + "epoch": 0.2998845496132846, + "grad_norm": 0.267578125, + "learning_rate": 0.0015007225118052537, + "loss": 0.0923, + "step": 34547 + }, + { + "epoch": 0.2998932300934888, + "grad_norm": 0.1455078125, + "learning_rate": 0.0015006958620540573, + "loss": 0.126, + "step": 34548 + }, + { + "epoch": 0.2999019105736929, + "grad_norm": 0.197265625, + "learning_rate": 0.0015006692118646665, + "loss": 0.1084, + "step": 34549 + }, + { + "epoch": 0.2999105910538971, + "grad_norm": 0.095703125, + "learning_rate": 0.0015006425612371095, + "loss": 0.0972, + "step": 34550 + }, + { + "epoch": 0.29991927153410125, + "grad_norm": 0.361328125, + "learning_rate": 0.0015006159101714162, + "loss": 0.1172, + "step": 34551 + }, + { + "epoch": 0.29992795201430544, + "grad_norm": 0.4765625, + "learning_rate": 0.0015005892586676152, + "loss": 0.1069, + "step": 34552 + }, + { + "epoch": 0.2999366324945096, + "grad_norm": 0.27734375, + "learning_rate": 0.0015005626067257356, + "loss": 0.1035, + "step": 34553 + }, + { + "epoch": 0.2999453129747138, + "grad_norm": 0.2021484375, + "learning_rate": 0.001500535954345807, + "loss": 0.1094, + "step": 34554 + }, + { + "epoch": 0.2999539934549179, + "grad_norm": 0.169921875, + "learning_rate": 0.0015005093015278582, + "loss": 0.1094, + "step": 34555 + }, + { + "epoch": 0.2999626739351221, + "grad_norm": 0.376953125, + "learning_rate": 0.0015004826482719186, + "loss": 0.1152, + "step": 34556 + }, + { + "epoch": 0.29997135441532624, + "grad_norm": 0.134765625, + "learning_rate": 0.0015004559945780169, + "loss": 0.0752, + "step": 34557 + }, + { + "epoch": 0.29998003489553043, + "grad_norm": 0.09716796875, + "learning_rate": 0.0015004293404461827, + "loss": 0.1514, + "step": 34558 + }, + { + "epoch": 0.29998871537573457, + "grad_norm": 0.5625, + "learning_rate": 0.0015004026858764447, + "loss": 0.0859, + "step": 34559 + }, + { + "epoch": 0.29999739585593876, + "grad_norm": 0.54296875, + "learning_rate": 0.001500376030868833, + "loss": 0.0767, + "step": 34560 + }, + { + "epoch": 0.3000060763361429, + "grad_norm": 0.26953125, + "learning_rate": 0.0015003493754233758, + "loss": 0.0903, + "step": 34561 + }, + { + "epoch": 0.3000147568163471, + "grad_norm": 0.2734375, + "learning_rate": 0.0015003227195401023, + "loss": 0.123, + "step": 34562 + }, + { + "epoch": 0.30002343729655123, + "grad_norm": 0.39453125, + "learning_rate": 0.001500296063219042, + "loss": 0.1416, + "step": 34563 + }, + { + "epoch": 0.3000321177767554, + "grad_norm": 0.404296875, + "learning_rate": 0.001500269406460224, + "loss": 0.0874, + "step": 34564 + }, + { + "epoch": 0.30004079825695956, + "grad_norm": 0.251953125, + "learning_rate": 0.0015002427492636773, + "loss": 0.1152, + "step": 34565 + }, + { + "epoch": 0.30004947873716376, + "grad_norm": 0.0830078125, + "learning_rate": 0.0015002160916294312, + "loss": 0.0889, + "step": 34566 + }, + { + "epoch": 0.3000581592173679, + "grad_norm": 0.25, + "learning_rate": 0.001500189433557515, + "loss": 0.0967, + "step": 34567 + }, + { + "epoch": 0.3000668396975721, + "grad_norm": 0.333984375, + "learning_rate": 0.0015001627750479572, + "loss": 0.2324, + "step": 34568 + }, + { + "epoch": 0.3000755201777762, + "grad_norm": 0.20703125, + "learning_rate": 0.0015001361161007876, + "loss": 0.1021, + "step": 34569 + }, + { + "epoch": 0.3000842006579804, + "grad_norm": 0.421875, + "learning_rate": 0.0015001094567160353, + "loss": 0.1582, + "step": 34570 + }, + { + "epoch": 0.30009288113818455, + "grad_norm": 0.369140625, + "learning_rate": 0.0015000827968937294, + "loss": 0.083, + "step": 34571 + }, + { + "epoch": 0.30010156161838875, + "grad_norm": 0.62109375, + "learning_rate": 0.0015000561366338985, + "loss": 0.0869, + "step": 34572 + }, + { + "epoch": 0.3001102420985929, + "grad_norm": 0.138671875, + "learning_rate": 0.0015000294759365724, + "loss": 0.1797, + "step": 34573 + }, + { + "epoch": 0.3001189225787971, + "grad_norm": 0.126953125, + "learning_rate": 0.0015000028148017804, + "loss": 0.0869, + "step": 34574 + }, + { + "epoch": 0.3001276030590012, + "grad_norm": 0.306640625, + "learning_rate": 0.0014999761532295513, + "loss": 0.1084, + "step": 34575 + }, + { + "epoch": 0.3001362835392054, + "grad_norm": 0.36328125, + "learning_rate": 0.001499949491219914, + "loss": 0.123, + "step": 34576 + }, + { + "epoch": 0.30014496401940954, + "grad_norm": 0.53125, + "learning_rate": 0.0014999228287728979, + "loss": 0.1167, + "step": 34577 + }, + { + "epoch": 0.30015364449961374, + "grad_norm": 0.28515625, + "learning_rate": 0.0014998961658885324, + "loss": 0.1182, + "step": 34578 + }, + { + "epoch": 0.3001623249798179, + "grad_norm": 0.12890625, + "learning_rate": 0.0014998695025668462, + "loss": 0.0972, + "step": 34579 + }, + { + "epoch": 0.30017100546002207, + "grad_norm": 0.189453125, + "learning_rate": 0.0014998428388078688, + "loss": 0.0698, + "step": 34580 + }, + { + "epoch": 0.3001796859402262, + "grad_norm": 0.138671875, + "learning_rate": 0.0014998161746116296, + "loss": 0.1211, + "step": 34581 + }, + { + "epoch": 0.3001883664204304, + "grad_norm": 0.12451171875, + "learning_rate": 0.0014997895099781574, + "loss": 0.106, + "step": 34582 + }, + { + "epoch": 0.30019704690063453, + "grad_norm": 0.42578125, + "learning_rate": 0.0014997628449074812, + "loss": 0.0981, + "step": 34583 + }, + { + "epoch": 0.3002057273808387, + "grad_norm": 0.1328125, + "learning_rate": 0.0014997361793996306, + "loss": 0.1108, + "step": 34584 + }, + { + "epoch": 0.30021440786104286, + "grad_norm": 0.41015625, + "learning_rate": 0.0014997095134546342, + "loss": 0.1602, + "step": 34585 + }, + { + "epoch": 0.30022308834124706, + "grad_norm": 0.1484375, + "learning_rate": 0.0014996828470725217, + "loss": 0.0991, + "step": 34586 + }, + { + "epoch": 0.3002317688214512, + "grad_norm": 0.1279296875, + "learning_rate": 0.0014996561802533216, + "loss": 0.0986, + "step": 34587 + }, + { + "epoch": 0.3002404493016554, + "grad_norm": 0.451171875, + "learning_rate": 0.001499629512997064, + "loss": 0.1206, + "step": 34588 + }, + { + "epoch": 0.3002491297818595, + "grad_norm": 0.126953125, + "learning_rate": 0.0014996028453037773, + "loss": 0.1279, + "step": 34589 + }, + { + "epoch": 0.3002578102620637, + "grad_norm": 0.2890625, + "learning_rate": 0.001499576177173491, + "loss": 0.0898, + "step": 34590 + }, + { + "epoch": 0.30026649074226786, + "grad_norm": 0.220703125, + "learning_rate": 0.0014995495086062344, + "loss": 0.0771, + "step": 34591 + }, + { + "epoch": 0.30027517122247205, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014995228396020364, + "loss": 0.1123, + "step": 34592 + }, + { + "epoch": 0.3002838517026762, + "grad_norm": 0.08056640625, + "learning_rate": 0.001499496170160926, + "loss": 0.0908, + "step": 34593 + }, + { + "epoch": 0.3002925321828804, + "grad_norm": 0.337890625, + "learning_rate": 0.001499469500282933, + "loss": 0.1123, + "step": 34594 + }, + { + "epoch": 0.3003012126630845, + "grad_norm": 0.279296875, + "learning_rate": 0.0014994428299680857, + "loss": 0.1367, + "step": 34595 + }, + { + "epoch": 0.3003098931432887, + "grad_norm": 0.09130859375, + "learning_rate": 0.0014994161592164137, + "loss": 0.0854, + "step": 34596 + }, + { + "epoch": 0.30031857362349285, + "grad_norm": 0.419921875, + "learning_rate": 0.0014993894880279467, + "loss": 0.1064, + "step": 34597 + }, + { + "epoch": 0.30032725410369704, + "grad_norm": 0.248046875, + "learning_rate": 0.001499362816402713, + "loss": 0.0698, + "step": 34598 + }, + { + "epoch": 0.3003359345839012, + "grad_norm": 0.29296875, + "learning_rate": 0.001499336144340742, + "loss": 0.0825, + "step": 34599 + }, + { + "epoch": 0.30034461506410537, + "grad_norm": 0.6875, + "learning_rate": 0.0014993094718420631, + "loss": 0.127, + "step": 34600 + }, + { + "epoch": 0.3003532955443095, + "grad_norm": 0.625, + "learning_rate": 0.0014992827989067058, + "loss": 0.0923, + "step": 34601 + }, + { + "epoch": 0.3003619760245137, + "grad_norm": 0.3359375, + "learning_rate": 0.0014992561255346981, + "loss": 0.0869, + "step": 34602 + }, + { + "epoch": 0.30037065650471784, + "grad_norm": 0.51171875, + "learning_rate": 0.0014992294517260705, + "loss": 0.126, + "step": 34603 + }, + { + "epoch": 0.30037933698492203, + "grad_norm": 0.609375, + "learning_rate": 0.0014992027774808512, + "loss": 0.1045, + "step": 34604 + }, + { + "epoch": 0.30038801746512617, + "grad_norm": 0.578125, + "learning_rate": 0.0014991761027990699, + "loss": 0.1357, + "step": 34605 + }, + { + "epoch": 0.30039669794533036, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014991494276807556, + "loss": 0.0781, + "step": 34606 + }, + { + "epoch": 0.3004053784255345, + "grad_norm": 0.17578125, + "learning_rate": 0.0014991227521259374, + "loss": 0.084, + "step": 34607 + }, + { + "epoch": 0.3004140589057387, + "grad_norm": 0.177734375, + "learning_rate": 0.0014990960761346447, + "loss": 0.0825, + "step": 34608 + }, + { + "epoch": 0.3004227393859428, + "grad_norm": 0.71484375, + "learning_rate": 0.0014990693997069065, + "loss": 0.0894, + "step": 34609 + }, + { + "epoch": 0.300431419866147, + "grad_norm": 0.1337890625, + "learning_rate": 0.001499042722842752, + "loss": 0.1084, + "step": 34610 + }, + { + "epoch": 0.30044010034635116, + "grad_norm": 0.21484375, + "learning_rate": 0.0014990160455422101, + "loss": 0.1221, + "step": 34611 + }, + { + "epoch": 0.30044878082655535, + "grad_norm": 0.130859375, + "learning_rate": 0.0014989893678053106, + "loss": 0.1074, + "step": 34612 + }, + { + "epoch": 0.3004574613067595, + "grad_norm": 0.52734375, + "learning_rate": 0.0014989626896320823, + "loss": 0.127, + "step": 34613 + }, + { + "epoch": 0.3004661417869637, + "grad_norm": 0.18359375, + "learning_rate": 0.0014989360110225544, + "loss": 0.1387, + "step": 34614 + }, + { + "epoch": 0.3004748222671678, + "grad_norm": 0.39453125, + "learning_rate": 0.0014989093319767556, + "loss": 0.0669, + "step": 34615 + }, + { + "epoch": 0.300483502747372, + "grad_norm": 1.6171875, + "learning_rate": 0.0014988826524947158, + "loss": 0.0977, + "step": 34616 + }, + { + "epoch": 0.30049218322757615, + "grad_norm": 0.3125, + "learning_rate": 0.0014988559725764644, + "loss": 0.0791, + "step": 34617 + }, + { + "epoch": 0.30050086370778034, + "grad_norm": 0.97265625, + "learning_rate": 0.0014988292922220295, + "loss": 0.1572, + "step": 34618 + }, + { + "epoch": 0.3005095441879845, + "grad_norm": 0.2294921875, + "learning_rate": 0.001498802611431441, + "loss": 0.0977, + "step": 34619 + }, + { + "epoch": 0.30051822466818867, + "grad_norm": 1.421875, + "learning_rate": 0.001498775930204728, + "loss": 0.1309, + "step": 34620 + }, + { + "epoch": 0.3005269051483928, + "grad_norm": 0.1884765625, + "learning_rate": 0.00149874924854192, + "loss": 0.1221, + "step": 34621 + }, + { + "epoch": 0.30053558562859695, + "grad_norm": 0.345703125, + "learning_rate": 0.0014987225664430455, + "loss": 0.1221, + "step": 34622 + }, + { + "epoch": 0.30054426610880114, + "grad_norm": 0.70703125, + "learning_rate": 0.0014986958839081342, + "loss": 0.1074, + "step": 34623 + }, + { + "epoch": 0.3005529465890053, + "grad_norm": 0.455078125, + "learning_rate": 0.0014986692009372148, + "loss": 0.1025, + "step": 34624 + }, + { + "epoch": 0.30056162706920947, + "grad_norm": 0.3125, + "learning_rate": 0.0014986425175303167, + "loss": 0.1162, + "step": 34625 + }, + { + "epoch": 0.3005703075494136, + "grad_norm": 0.09375, + "learning_rate": 0.0014986158336874691, + "loss": 0.1104, + "step": 34626 + }, + { + "epoch": 0.3005789880296178, + "grad_norm": 0.26171875, + "learning_rate": 0.0014985891494087016, + "loss": 0.1079, + "step": 34627 + }, + { + "epoch": 0.30058766850982194, + "grad_norm": 0.6171875, + "learning_rate": 0.0014985624646940425, + "loss": 0.0942, + "step": 34628 + }, + { + "epoch": 0.30059634899002613, + "grad_norm": 0.7265625, + "learning_rate": 0.001498535779543522, + "loss": 0.1572, + "step": 34629 + }, + { + "epoch": 0.30060502947023027, + "grad_norm": 0.81640625, + "learning_rate": 0.0014985090939571683, + "loss": 0.1079, + "step": 34630 + }, + { + "epoch": 0.30061370995043446, + "grad_norm": 0.2197265625, + "learning_rate": 0.0014984824079350117, + "loss": 0.0972, + "step": 34631 + }, + { + "epoch": 0.3006223904306386, + "grad_norm": 0.142578125, + "learning_rate": 0.00149845572147708, + "loss": 0.0835, + "step": 34632 + }, + { + "epoch": 0.3006310709108428, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014984290345834035, + "loss": 0.0991, + "step": 34633 + }, + { + "epoch": 0.3006397513910469, + "grad_norm": 0.091796875, + "learning_rate": 0.0014984023472540107, + "loss": 0.1055, + "step": 34634 + }, + { + "epoch": 0.3006484318712511, + "grad_norm": 0.55078125, + "learning_rate": 0.0014983756594889315, + "loss": 0.0933, + "step": 34635 + }, + { + "epoch": 0.30065711235145526, + "grad_norm": 0.7109375, + "learning_rate": 0.0014983489712881942, + "loss": 0.0781, + "step": 34636 + }, + { + "epoch": 0.30066579283165945, + "grad_norm": 0.40234375, + "learning_rate": 0.0014983222826518289, + "loss": 0.0967, + "step": 34637 + }, + { + "epoch": 0.3006744733118636, + "grad_norm": 0.109375, + "learning_rate": 0.0014982955935798638, + "loss": 0.064, + "step": 34638 + }, + { + "epoch": 0.3006831537920678, + "grad_norm": 0.69921875, + "learning_rate": 0.001498268904072329, + "loss": 0.1455, + "step": 34639 + }, + { + "epoch": 0.3006918342722719, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014982422141292533, + "loss": 0.1045, + "step": 34640 + }, + { + "epoch": 0.3007005147524761, + "grad_norm": 0.5546875, + "learning_rate": 0.0014982155237506659, + "loss": 0.1113, + "step": 34641 + }, + { + "epoch": 0.30070919523268025, + "grad_norm": 0.337890625, + "learning_rate": 0.0014981888329365961, + "loss": 0.0947, + "step": 34642 + }, + { + "epoch": 0.30071787571288444, + "grad_norm": 0.140625, + "learning_rate": 0.0014981621416870726, + "loss": 0.084, + "step": 34643 + }, + { + "epoch": 0.3007265561930886, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014981354500021252, + "loss": 0.126, + "step": 34644 + }, + { + "epoch": 0.30073523667329277, + "grad_norm": 0.08984375, + "learning_rate": 0.0014981087578817827, + "loss": 0.0762, + "step": 34645 + }, + { + "epoch": 0.3007439171534969, + "grad_norm": 0.087890625, + "learning_rate": 0.0014980820653260746, + "loss": 0.1172, + "step": 34646 + }, + { + "epoch": 0.3007525976337011, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014980553723350299, + "loss": 0.0776, + "step": 34647 + }, + { + "epoch": 0.30076127811390524, + "grad_norm": 0.447265625, + "learning_rate": 0.001498028678908678, + "loss": 0.082, + "step": 34648 + }, + { + "epoch": 0.30076995859410943, + "grad_norm": 0.302734375, + "learning_rate": 0.0014980019850470477, + "loss": 0.1357, + "step": 34649 + }, + { + "epoch": 0.30077863907431357, + "grad_norm": 0.291015625, + "learning_rate": 0.0014979752907501684, + "loss": 0.1494, + "step": 34650 + }, + { + "epoch": 0.30078731955451776, + "grad_norm": 0.302734375, + "learning_rate": 0.0014979485960180693, + "loss": 0.1079, + "step": 34651 + }, + { + "epoch": 0.3007960000347219, + "grad_norm": 0.12109375, + "learning_rate": 0.0014979219008507798, + "loss": 0.1504, + "step": 34652 + }, + { + "epoch": 0.3008046805149261, + "grad_norm": 0.91796875, + "learning_rate": 0.0014978952052483285, + "loss": 0.1191, + "step": 34653 + }, + { + "epoch": 0.30081336099513023, + "grad_norm": 0.203125, + "learning_rate": 0.001497868509210745, + "loss": 0.0898, + "step": 34654 + }, + { + "epoch": 0.3008220414753344, + "grad_norm": 0.37109375, + "learning_rate": 0.001497841812738059, + "loss": 0.1328, + "step": 34655 + }, + { + "epoch": 0.30083072195553856, + "grad_norm": 0.349609375, + "learning_rate": 0.001497815115830299, + "loss": 0.0728, + "step": 34656 + }, + { + "epoch": 0.30083940243574275, + "grad_norm": 0.294921875, + "learning_rate": 0.0014977884184874941, + "loss": 0.0898, + "step": 34657 + }, + { + "epoch": 0.3008480829159469, + "grad_norm": 0.6953125, + "learning_rate": 0.0014977617207096738, + "loss": 0.0845, + "step": 34658 + }, + { + "epoch": 0.3008567633961511, + "grad_norm": 0.11328125, + "learning_rate": 0.0014977350224968675, + "loss": 0.0879, + "step": 34659 + }, + { + "epoch": 0.3008654438763552, + "grad_norm": 0.07421875, + "learning_rate": 0.0014977083238491042, + "loss": 0.0781, + "step": 34660 + }, + { + "epoch": 0.3008741243565594, + "grad_norm": 0.0908203125, + "learning_rate": 0.001497681624766413, + "loss": 0.1279, + "step": 34661 + }, + { + "epoch": 0.30088280483676355, + "grad_norm": 0.40234375, + "learning_rate": 0.0014976549252488226, + "loss": 0.1777, + "step": 34662 + }, + { + "epoch": 0.30089148531696774, + "grad_norm": 0.326171875, + "learning_rate": 0.0014976282252963631, + "loss": 0.1406, + "step": 34663 + }, + { + "epoch": 0.3009001657971719, + "grad_norm": 0.224609375, + "learning_rate": 0.0014976015249090636, + "loss": 0.085, + "step": 34664 + }, + { + "epoch": 0.3009088462773761, + "grad_norm": 0.302734375, + "learning_rate": 0.0014975748240869528, + "loss": 0.0967, + "step": 34665 + }, + { + "epoch": 0.3009175267575802, + "grad_norm": 0.115234375, + "learning_rate": 0.0014975481228300602, + "loss": 0.1123, + "step": 34666 + }, + { + "epoch": 0.3009262072377844, + "grad_norm": 0.78515625, + "learning_rate": 0.0014975214211384149, + "loss": 0.1201, + "step": 34667 + }, + { + "epoch": 0.30093488771798854, + "grad_norm": 0.318359375, + "learning_rate": 0.0014974947190120463, + "loss": 0.1309, + "step": 34668 + }, + { + "epoch": 0.30094356819819273, + "grad_norm": 0.390625, + "learning_rate": 0.0014974680164509832, + "loss": 0.0986, + "step": 34669 + }, + { + "epoch": 0.30095224867839687, + "grad_norm": 0.443359375, + "learning_rate": 0.0014974413134552554, + "loss": 0.1152, + "step": 34670 + }, + { + "epoch": 0.30096092915860106, + "grad_norm": 0.06298828125, + "learning_rate": 0.0014974146100248916, + "loss": 0.0757, + "step": 34671 + }, + { + "epoch": 0.3009696096388052, + "grad_norm": 0.412109375, + "learning_rate": 0.001497387906159921, + "loss": 0.0972, + "step": 34672 + }, + { + "epoch": 0.3009782901190094, + "grad_norm": 0.23046875, + "learning_rate": 0.0014973612018603728, + "loss": 0.0781, + "step": 34673 + }, + { + "epoch": 0.30098697059921353, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014973344971262765, + "loss": 0.0962, + "step": 34674 + }, + { + "epoch": 0.3009956510794177, + "grad_norm": 0.462890625, + "learning_rate": 0.0014973077919576614, + "loss": 0.1152, + "step": 34675 + }, + { + "epoch": 0.30100433155962186, + "grad_norm": 0.166015625, + "learning_rate": 0.0014972810863545564, + "loss": 0.0796, + "step": 34676 + }, + { + "epoch": 0.30101301203982606, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014972543803169907, + "loss": 0.0962, + "step": 34677 + }, + { + "epoch": 0.3010216925200302, + "grad_norm": 0.390625, + "learning_rate": 0.001497227673844994, + "loss": 0.0771, + "step": 34678 + }, + { + "epoch": 0.3010303730002344, + "grad_norm": 0.267578125, + "learning_rate": 0.0014972009669385945, + "loss": 0.0908, + "step": 34679 + }, + { + "epoch": 0.3010390534804385, + "grad_norm": 0.83203125, + "learning_rate": 0.0014971742595978222, + "loss": 0.1426, + "step": 34680 + }, + { + "epoch": 0.3010477339606427, + "grad_norm": 0.061767578125, + "learning_rate": 0.0014971475518227057, + "loss": 0.0723, + "step": 34681 + }, + { + "epoch": 0.30105641444084685, + "grad_norm": 0.0927734375, + "learning_rate": 0.001497120843613275, + "loss": 0.1045, + "step": 34682 + }, + { + "epoch": 0.30106509492105105, + "grad_norm": 0.40625, + "learning_rate": 0.001497094134969559, + "loss": 0.1113, + "step": 34683 + }, + { + "epoch": 0.3010737754012552, + "grad_norm": 0.51171875, + "learning_rate": 0.0014970674258915867, + "loss": 0.1357, + "step": 34684 + }, + { + "epoch": 0.3010824558814594, + "grad_norm": 0.59765625, + "learning_rate": 0.0014970407163793873, + "loss": 0.1206, + "step": 34685 + }, + { + "epoch": 0.3010911363616635, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014970140064329905, + "loss": 0.0889, + "step": 34686 + }, + { + "epoch": 0.3010998168418677, + "grad_norm": 0.119140625, + "learning_rate": 0.0014969872960524252, + "loss": 0.0747, + "step": 34687 + }, + { + "epoch": 0.30110849732207184, + "grad_norm": 0.185546875, + "learning_rate": 0.0014969605852377197, + "loss": 0.1152, + "step": 34688 + }, + { + "epoch": 0.30111717780227604, + "grad_norm": 0.3359375, + "learning_rate": 0.0014969338739889049, + "loss": 0.0776, + "step": 34689 + }, + { + "epoch": 0.3011258582824802, + "grad_norm": 0.162109375, + "learning_rate": 0.001496907162306009, + "loss": 0.1099, + "step": 34690 + }, + { + "epoch": 0.30113453876268437, + "grad_norm": 0.25, + "learning_rate": 0.0014968804501890613, + "loss": 0.1133, + "step": 34691 + }, + { + "epoch": 0.3011432192428885, + "grad_norm": 0.451171875, + "learning_rate": 0.001496853737638091, + "loss": 0.0894, + "step": 34692 + }, + { + "epoch": 0.3011518997230927, + "grad_norm": 0.87890625, + "learning_rate": 0.0014968270246531272, + "loss": 0.1045, + "step": 34693 + }, + { + "epoch": 0.30116058020329683, + "grad_norm": 0.109375, + "learning_rate": 0.0014968003112342, + "loss": 0.1025, + "step": 34694 + }, + { + "epoch": 0.301169260683501, + "grad_norm": 0.208984375, + "learning_rate": 0.0014967735973813376, + "loss": 0.0884, + "step": 34695 + }, + { + "epoch": 0.30117794116370517, + "grad_norm": 3.65625, + "learning_rate": 0.0014967468830945697, + "loss": 0.3359, + "step": 34696 + }, + { + "epoch": 0.30118662164390936, + "grad_norm": 0.3203125, + "learning_rate": 0.001496720168373925, + "loss": 0.127, + "step": 34697 + }, + { + "epoch": 0.3011953021241135, + "grad_norm": 0.11962890625, + "learning_rate": 0.0014966934532194334, + "loss": 0.1172, + "step": 34698 + }, + { + "epoch": 0.3012039826043177, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014966667376311237, + "loss": 0.0854, + "step": 34699 + }, + { + "epoch": 0.3012126630845218, + "grad_norm": 0.28515625, + "learning_rate": 0.0014966400216090252, + "loss": 0.1045, + "step": 34700 + }, + { + "epoch": 0.301221343564726, + "grad_norm": 0.2451171875, + "learning_rate": 0.0014966133051531672, + "loss": 0.127, + "step": 34701 + }, + { + "epoch": 0.30123002404493016, + "grad_norm": 0.224609375, + "learning_rate": 0.0014965865882635787, + "loss": 0.0752, + "step": 34702 + }, + { + "epoch": 0.30123870452513435, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014965598709402892, + "loss": 0.125, + "step": 34703 + }, + { + "epoch": 0.3012473850053385, + "grad_norm": 0.423828125, + "learning_rate": 0.001496533153183328, + "loss": 0.0957, + "step": 34704 + }, + { + "epoch": 0.3012560654855427, + "grad_norm": 0.421875, + "learning_rate": 0.0014965064349927238, + "loss": 0.0791, + "step": 34705 + }, + { + "epoch": 0.3012647459657468, + "grad_norm": 0.1865234375, + "learning_rate": 0.001496479716368506, + "loss": 0.127, + "step": 34706 + }, + { + "epoch": 0.301273426445951, + "grad_norm": 0.41796875, + "learning_rate": 0.0014964529973107042, + "loss": 0.1011, + "step": 34707 + }, + { + "epoch": 0.30128210692615515, + "grad_norm": 0.431640625, + "learning_rate": 0.0014964262778193473, + "loss": 0.1543, + "step": 34708 + }, + { + "epoch": 0.30129078740635934, + "grad_norm": 1.0234375, + "learning_rate": 0.0014963995578944646, + "loss": 0.1235, + "step": 34709 + }, + { + "epoch": 0.3012994678865635, + "grad_norm": 0.2353515625, + "learning_rate": 0.001496372837536085, + "loss": 0.1387, + "step": 34710 + }, + { + "epoch": 0.30130814836676767, + "grad_norm": 0.21875, + "learning_rate": 0.0014963461167442386, + "loss": 0.1191, + "step": 34711 + }, + { + "epoch": 0.3013168288469718, + "grad_norm": 0.421875, + "learning_rate": 0.0014963193955189534, + "loss": 0.0767, + "step": 34712 + }, + { + "epoch": 0.301325509327176, + "grad_norm": 0.546875, + "learning_rate": 0.0014962926738602597, + "loss": 0.1562, + "step": 34713 + }, + { + "epoch": 0.30133418980738014, + "grad_norm": 0.671875, + "learning_rate": 0.001496265951768186, + "loss": 0.1006, + "step": 34714 + }, + { + "epoch": 0.30134287028758433, + "grad_norm": 0.0947265625, + "learning_rate": 0.001496239229242762, + "loss": 0.0977, + "step": 34715 + }, + { + "epoch": 0.30135155076778847, + "grad_norm": 0.248046875, + "learning_rate": 0.001496212506284017, + "loss": 0.083, + "step": 34716 + }, + { + "epoch": 0.30136023124799266, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014961857828919797, + "loss": 0.1055, + "step": 34717 + }, + { + "epoch": 0.3013689117281968, + "grad_norm": 0.103515625, + "learning_rate": 0.0014961590590666797, + "loss": 0.1367, + "step": 34718 + }, + { + "epoch": 0.301377592208401, + "grad_norm": 0.255859375, + "learning_rate": 0.0014961323348081457, + "loss": 0.1211, + "step": 34719 + }, + { + "epoch": 0.30138627268860513, + "grad_norm": 0.3828125, + "learning_rate": 0.0014961056101164077, + "loss": 0.0684, + "step": 34720 + }, + { + "epoch": 0.3013949531688093, + "grad_norm": 0.75, + "learning_rate": 0.0014960788849914942, + "loss": 0.1357, + "step": 34721 + }, + { + "epoch": 0.30140363364901346, + "grad_norm": 0.3984375, + "learning_rate": 0.001496052159433435, + "loss": 0.1001, + "step": 34722 + }, + { + "epoch": 0.30141231412921765, + "grad_norm": 0.3046875, + "learning_rate": 0.0014960254334422589, + "loss": 0.0601, + "step": 34723 + }, + { + "epoch": 0.3014209946094218, + "grad_norm": 0.29296875, + "learning_rate": 0.0014959987070179957, + "loss": 0.104, + "step": 34724 + }, + { + "epoch": 0.301429675089626, + "grad_norm": 0.314453125, + "learning_rate": 0.0014959719801606742, + "loss": 0.0732, + "step": 34725 + }, + { + "epoch": 0.3014383555698301, + "grad_norm": 0.267578125, + "learning_rate": 0.0014959452528703237, + "loss": 0.1465, + "step": 34726 + }, + { + "epoch": 0.3014470360500343, + "grad_norm": 0.58984375, + "learning_rate": 0.0014959185251469734, + "loss": 0.127, + "step": 34727 + }, + { + "epoch": 0.30145571653023845, + "grad_norm": 0.248046875, + "learning_rate": 0.0014958917969906524, + "loss": 0.1436, + "step": 34728 + }, + { + "epoch": 0.30146439701044264, + "grad_norm": 0.138671875, + "learning_rate": 0.00149586506840139, + "loss": 0.0771, + "step": 34729 + }, + { + "epoch": 0.3014730774906468, + "grad_norm": 0.076171875, + "learning_rate": 0.001495838339379216, + "loss": 0.0713, + "step": 34730 + }, + { + "epoch": 0.30148175797085097, + "grad_norm": 0.45703125, + "learning_rate": 0.0014958116099241588, + "loss": 0.1309, + "step": 34731 + }, + { + "epoch": 0.3014904384510551, + "grad_norm": 0.08349609375, + "learning_rate": 0.0014957848800362477, + "loss": 0.084, + "step": 34732 + }, + { + "epoch": 0.3014991189312593, + "grad_norm": 0.11279296875, + "learning_rate": 0.0014957581497155126, + "loss": 0.0879, + "step": 34733 + }, + { + "epoch": 0.30150779941146344, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014957314189619824, + "loss": 0.084, + "step": 34734 + }, + { + "epoch": 0.30151647989166763, + "grad_norm": 0.271484375, + "learning_rate": 0.001495704687775686, + "loss": 0.1182, + "step": 34735 + }, + { + "epoch": 0.30152516037187177, + "grad_norm": 0.482421875, + "learning_rate": 0.001495677956156653, + "loss": 0.0918, + "step": 34736 + }, + { + "epoch": 0.30153384085207596, + "grad_norm": 0.83203125, + "learning_rate": 0.0014956512241049125, + "loss": 0.1074, + "step": 34737 + }, + { + "epoch": 0.3015425213322801, + "grad_norm": 0.83203125, + "learning_rate": 0.0014956244916204937, + "loss": 0.0645, + "step": 34738 + }, + { + "epoch": 0.3015512018124843, + "grad_norm": 0.435546875, + "learning_rate": 0.001495597758703426, + "loss": 0.0884, + "step": 34739 + }, + { + "epoch": 0.30155988229268843, + "grad_norm": 0.3984375, + "learning_rate": 0.0014955710253537386, + "loss": 0.1235, + "step": 34740 + }, + { + "epoch": 0.3015685627728926, + "grad_norm": 0.458984375, + "learning_rate": 0.0014955442915714602, + "loss": 0.0898, + "step": 34741 + }, + { + "epoch": 0.30157724325309676, + "grad_norm": 0.087890625, + "learning_rate": 0.0014955175573566214, + "loss": 0.0869, + "step": 34742 + }, + { + "epoch": 0.30158592373330095, + "grad_norm": 0.267578125, + "learning_rate": 0.0014954908227092498, + "loss": 0.1455, + "step": 34743 + }, + { + "epoch": 0.3015946042135051, + "grad_norm": 0.4609375, + "learning_rate": 0.001495464087629376, + "loss": 0.0986, + "step": 34744 + }, + { + "epoch": 0.3016032846937093, + "grad_norm": 0.7890625, + "learning_rate": 0.0014954373521170279, + "loss": 0.123, + "step": 34745 + }, + { + "epoch": 0.3016119651739134, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014954106161722358, + "loss": 0.1123, + "step": 34746 + }, + { + "epoch": 0.30162064565411756, + "grad_norm": 1.609375, + "learning_rate": 0.0014953838797950288, + "loss": 0.0962, + "step": 34747 + }, + { + "epoch": 0.30162932613432175, + "grad_norm": 0.7890625, + "learning_rate": 0.0014953571429854357, + "loss": 0.1279, + "step": 34748 + }, + { + "epoch": 0.3016380066145259, + "grad_norm": 0.279296875, + "learning_rate": 0.0014953304057434857, + "loss": 0.0713, + "step": 34749 + }, + { + "epoch": 0.3016466870947301, + "grad_norm": 0.423828125, + "learning_rate": 0.0014953036680692084, + "loss": 0.0776, + "step": 34750 + }, + { + "epoch": 0.3016553675749342, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014952769299626334, + "loss": 0.0811, + "step": 34751 + }, + { + "epoch": 0.3016640480551384, + "grad_norm": 0.6875, + "learning_rate": 0.0014952501914237892, + "loss": 0.1055, + "step": 34752 + }, + { + "epoch": 0.30167272853534255, + "grad_norm": 0.498046875, + "learning_rate": 0.0014952234524527052, + "loss": 0.0806, + "step": 34753 + }, + { + "epoch": 0.30168140901554674, + "grad_norm": 0.158203125, + "learning_rate": 0.001495196713049411, + "loss": 0.0801, + "step": 34754 + }, + { + "epoch": 0.3016900894957509, + "grad_norm": 0.173828125, + "learning_rate": 0.0014951699732139356, + "loss": 0.1069, + "step": 34755 + }, + { + "epoch": 0.3016987699759551, + "grad_norm": 0.09423828125, + "learning_rate": 0.001495143232946308, + "loss": 0.123, + "step": 34756 + }, + { + "epoch": 0.3017074504561592, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014951164922465583, + "loss": 0.0767, + "step": 34757 + }, + { + "epoch": 0.3017161309363634, + "grad_norm": 0.4765625, + "learning_rate": 0.0014950897511147143, + "loss": 0.125, + "step": 34758 + }, + { + "epoch": 0.30172481141656754, + "grad_norm": 0.4296875, + "learning_rate": 0.0014950630095508066, + "loss": 0.0903, + "step": 34759 + }, + { + "epoch": 0.30173349189677173, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014950362675548638, + "loss": 0.1387, + "step": 34760 + }, + { + "epoch": 0.30174217237697587, + "grad_norm": 0.396484375, + "learning_rate": 0.0014950095251269153, + "loss": 0.1719, + "step": 34761 + }, + { + "epoch": 0.30175085285718006, + "grad_norm": 0.2138671875, + "learning_rate": 0.0014949827822669905, + "loss": 0.1055, + "step": 34762 + }, + { + "epoch": 0.3017595333373842, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014949560389751182, + "loss": 0.0908, + "step": 34763 + }, + { + "epoch": 0.3017682138175884, + "grad_norm": 0.16015625, + "learning_rate": 0.001494929295251328, + "loss": 0.0703, + "step": 34764 + }, + { + "epoch": 0.30177689429779253, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014949025510956494, + "loss": 0.1055, + "step": 34765 + }, + { + "epoch": 0.3017855747779967, + "grad_norm": 0.23828125, + "learning_rate": 0.0014948758065081106, + "loss": 0.0986, + "step": 34766 + }, + { + "epoch": 0.30179425525820086, + "grad_norm": 0.12060546875, + "learning_rate": 0.0014948490614887422, + "loss": 0.0986, + "step": 34767 + }, + { + "epoch": 0.30180293573840505, + "grad_norm": 0.5859375, + "learning_rate": 0.0014948223160375724, + "loss": 0.1011, + "step": 34768 + }, + { + "epoch": 0.3018116162186092, + "grad_norm": 0.232421875, + "learning_rate": 0.0014947955701546307, + "loss": 0.1172, + "step": 34769 + }, + { + "epoch": 0.3018202966988134, + "grad_norm": 0.357421875, + "learning_rate": 0.0014947688238399468, + "loss": 0.1641, + "step": 34770 + }, + { + "epoch": 0.3018289771790175, + "grad_norm": 0.1083984375, + "learning_rate": 0.0014947420770935495, + "loss": 0.0884, + "step": 34771 + }, + { + "epoch": 0.3018376576592217, + "grad_norm": 0.306640625, + "learning_rate": 0.0014947153299154686, + "loss": 0.1152, + "step": 34772 + }, + { + "epoch": 0.30184633813942585, + "grad_norm": 0.302734375, + "learning_rate": 0.0014946885823057326, + "loss": 0.0825, + "step": 34773 + }, + { + "epoch": 0.30185501861963004, + "grad_norm": 0.263671875, + "learning_rate": 0.0014946618342643711, + "loss": 0.0781, + "step": 34774 + }, + { + "epoch": 0.3018636990998342, + "grad_norm": 0.09619140625, + "learning_rate": 0.0014946350857914135, + "loss": 0.1133, + "step": 34775 + }, + { + "epoch": 0.3018723795800384, + "grad_norm": 0.123046875, + "learning_rate": 0.0014946083368868887, + "loss": 0.1084, + "step": 34776 + }, + { + "epoch": 0.3018810600602425, + "grad_norm": 0.1904296875, + "learning_rate": 0.0014945815875508262, + "loss": 0.1201, + "step": 34777 + }, + { + "epoch": 0.3018897405404467, + "grad_norm": 0.259765625, + "learning_rate": 0.0014945548377832552, + "loss": 0.1113, + "step": 34778 + }, + { + "epoch": 0.30189842102065084, + "grad_norm": 1.4453125, + "learning_rate": 0.001494528087584205, + "loss": 0.1406, + "step": 34779 + }, + { + "epoch": 0.30190710150085504, + "grad_norm": 0.11962890625, + "learning_rate": 0.0014945013369537046, + "loss": 0.1069, + "step": 34780 + }, + { + "epoch": 0.3019157819810592, + "grad_norm": 0.267578125, + "learning_rate": 0.0014944745858917838, + "loss": 0.0986, + "step": 34781 + }, + { + "epoch": 0.30192446246126337, + "grad_norm": 0.19140625, + "learning_rate": 0.0014944478343984719, + "loss": 0.0801, + "step": 34782 + }, + { + "epoch": 0.3019331429414675, + "grad_norm": 0.357421875, + "learning_rate": 0.0014944210824737968, + "loss": 0.1025, + "step": 34783 + }, + { + "epoch": 0.3019418234216717, + "grad_norm": 0.2109375, + "learning_rate": 0.0014943943301177895, + "loss": 0.1602, + "step": 34784 + }, + { + "epoch": 0.30195050390187583, + "grad_norm": 0.34765625, + "learning_rate": 0.001494367577330478, + "loss": 0.1006, + "step": 34785 + }, + { + "epoch": 0.30195918438208, + "grad_norm": 0.875, + "learning_rate": 0.0014943408241118923, + "loss": 0.3633, + "step": 34786 + }, + { + "epoch": 0.30196786486228416, + "grad_norm": 0.263671875, + "learning_rate": 0.0014943140704620612, + "loss": 0.1387, + "step": 34787 + }, + { + "epoch": 0.30197654534248836, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014942873163810144, + "loss": 0.0825, + "step": 34788 + }, + { + "epoch": 0.3019852258226925, + "grad_norm": 0.466796875, + "learning_rate": 0.0014942605618687805, + "loss": 0.0835, + "step": 34789 + }, + { + "epoch": 0.3019939063028967, + "grad_norm": 0.31640625, + "learning_rate": 0.0014942338069253898, + "loss": 0.085, + "step": 34790 + }, + { + "epoch": 0.3020025867831008, + "grad_norm": 0.34375, + "learning_rate": 0.0014942070515508706, + "loss": 0.0972, + "step": 34791 + }, + { + "epoch": 0.302011267263305, + "grad_norm": 0.44921875, + "learning_rate": 0.0014941802957452524, + "loss": 0.0947, + "step": 34792 + }, + { + "epoch": 0.30201994774350915, + "grad_norm": 0.28515625, + "learning_rate": 0.0014941535395085646, + "loss": 0.1045, + "step": 34793 + }, + { + "epoch": 0.30202862822371335, + "grad_norm": 0.78125, + "learning_rate": 0.0014941267828408368, + "loss": 0.1562, + "step": 34794 + }, + { + "epoch": 0.3020373087039175, + "grad_norm": 0.1953125, + "learning_rate": 0.0014941000257420974, + "loss": 0.1011, + "step": 34795 + }, + { + "epoch": 0.3020459891841217, + "grad_norm": 0.12353515625, + "learning_rate": 0.0014940732682123762, + "loss": 0.0791, + "step": 34796 + }, + { + "epoch": 0.3020546696643258, + "grad_norm": 0.2392578125, + "learning_rate": 0.0014940465102517024, + "loss": 0.0967, + "step": 34797 + }, + { + "epoch": 0.30206335014453, + "grad_norm": 0.1494140625, + "learning_rate": 0.001494019751860105, + "loss": 0.0835, + "step": 34798 + }, + { + "epoch": 0.30207203062473414, + "grad_norm": 0.28125, + "learning_rate": 0.0014939929930376141, + "loss": 0.082, + "step": 34799 + }, + { + "epoch": 0.30208071110493834, + "grad_norm": 0.10107421875, + "learning_rate": 0.001493966233784258, + "loss": 0.1187, + "step": 34800 + }, + { + "epoch": 0.3020893915851425, + "grad_norm": 0.27734375, + "learning_rate": 0.0014939394741000666, + "loss": 0.1133, + "step": 34801 + }, + { + "epoch": 0.30209807206534667, + "grad_norm": 0.1630859375, + "learning_rate": 0.001493912713985069, + "loss": 0.0854, + "step": 34802 + }, + { + "epoch": 0.3021067525455508, + "grad_norm": 0.408203125, + "learning_rate": 0.001493885953439294, + "loss": 0.1016, + "step": 34803 + }, + { + "epoch": 0.302115433025755, + "grad_norm": 0.11376953125, + "learning_rate": 0.0014938591924627714, + "loss": 0.0947, + "step": 34804 + }, + { + "epoch": 0.30212411350595914, + "grad_norm": 0.0576171875, + "learning_rate": 0.0014938324310555303, + "loss": 0.0771, + "step": 34805 + }, + { + "epoch": 0.30213279398616333, + "grad_norm": 0.244140625, + "learning_rate": 0.0014938056692176001, + "loss": 0.1191, + "step": 34806 + }, + { + "epoch": 0.30214147446636747, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014937789069490097, + "loss": 0.0752, + "step": 34807 + }, + { + "epoch": 0.30215015494657166, + "grad_norm": 0.44921875, + "learning_rate": 0.0014937521442497886, + "loss": 0.127, + "step": 34808 + }, + { + "epoch": 0.3021588354267758, + "grad_norm": 0.11962890625, + "learning_rate": 0.001493725381119966, + "loss": 0.1357, + "step": 34809 + }, + { + "epoch": 0.30216751590698, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014936986175595715, + "loss": 0.0972, + "step": 34810 + }, + { + "epoch": 0.3021761963871841, + "grad_norm": 0.404296875, + "learning_rate": 0.001493671853568634, + "loss": 0.1289, + "step": 34811 + }, + { + "epoch": 0.3021848768673883, + "grad_norm": 0.1826171875, + "learning_rate": 0.0014936450891471832, + "loss": 0.0977, + "step": 34812 + }, + { + "epoch": 0.30219355734759246, + "grad_norm": 0.09228515625, + "learning_rate": 0.001493618324295248, + "loss": 0.0659, + "step": 34813 + }, + { + "epoch": 0.30220223782779665, + "grad_norm": 0.578125, + "learning_rate": 0.0014935915590128571, + "loss": 0.0894, + "step": 34814 + }, + { + "epoch": 0.3022109183080008, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014935647933000409, + "loss": 0.0977, + "step": 34815 + }, + { + "epoch": 0.302219598788205, + "grad_norm": 0.1640625, + "learning_rate": 0.001493538027156828, + "loss": 0.1113, + "step": 34816 + }, + { + "epoch": 0.3022282792684091, + "grad_norm": 0.466796875, + "learning_rate": 0.0014935112605832478, + "loss": 0.0825, + "step": 34817 + }, + { + "epoch": 0.3022369597486133, + "grad_norm": 0.12109375, + "learning_rate": 0.0014934844935793293, + "loss": 0.1084, + "step": 34818 + }, + { + "epoch": 0.30224564022881745, + "grad_norm": 0.515625, + "learning_rate": 0.0014934577261451025, + "loss": 0.1021, + "step": 34819 + }, + { + "epoch": 0.30225432070902164, + "grad_norm": 0.4765625, + "learning_rate": 0.0014934309582805962, + "loss": 0.1074, + "step": 34820 + }, + { + "epoch": 0.3022630011892258, + "grad_norm": 0.326171875, + "learning_rate": 0.0014934041899858398, + "loss": 0.0859, + "step": 34821 + }, + { + "epoch": 0.30227168166942997, + "grad_norm": 0.205078125, + "learning_rate": 0.001493377421260862, + "loss": 0.0908, + "step": 34822 + }, + { + "epoch": 0.3022803621496341, + "grad_norm": 0.302734375, + "learning_rate": 0.001493350652105693, + "loss": 0.1064, + "step": 34823 + }, + { + "epoch": 0.3022890426298383, + "grad_norm": 0.451171875, + "learning_rate": 0.0014933238825203617, + "loss": 0.1035, + "step": 34824 + }, + { + "epoch": 0.30229772311004244, + "grad_norm": 0.271484375, + "learning_rate": 0.0014932971125048973, + "loss": 0.1074, + "step": 34825 + }, + { + "epoch": 0.30230640359024663, + "grad_norm": 0.6796875, + "learning_rate": 0.0014932703420593287, + "loss": 0.084, + "step": 34826 + }, + { + "epoch": 0.30231508407045077, + "grad_norm": 0.2578125, + "learning_rate": 0.0014932435711836856, + "loss": 0.1182, + "step": 34827 + }, + { + "epoch": 0.30232376455065496, + "grad_norm": 0.0810546875, + "learning_rate": 0.0014932167998779975, + "loss": 0.0942, + "step": 34828 + }, + { + "epoch": 0.3023324450308591, + "grad_norm": 0.369140625, + "learning_rate": 0.0014931900281422938, + "loss": 0.1777, + "step": 34829 + }, + { + "epoch": 0.3023411255110633, + "grad_norm": 0.0908203125, + "learning_rate": 0.0014931632559766026, + "loss": 0.082, + "step": 34830 + }, + { + "epoch": 0.30234980599126743, + "grad_norm": 0.341796875, + "learning_rate": 0.0014931364833809543, + "loss": 0.1064, + "step": 34831 + }, + { + "epoch": 0.3023584864714716, + "grad_norm": 0.056884765625, + "learning_rate": 0.0014931097103553779, + "loss": 0.0718, + "step": 34832 + }, + { + "epoch": 0.30236716695167576, + "grad_norm": 0.326171875, + "learning_rate": 0.0014930829368999027, + "loss": 0.1025, + "step": 34833 + }, + { + "epoch": 0.30237584743187995, + "grad_norm": 0.36328125, + "learning_rate": 0.0014930561630145578, + "loss": 0.1064, + "step": 34834 + }, + { + "epoch": 0.3023845279120841, + "grad_norm": 0.412109375, + "learning_rate": 0.0014930293886993727, + "loss": 0.1006, + "step": 34835 + }, + { + "epoch": 0.3023932083922883, + "grad_norm": 0.412109375, + "learning_rate": 0.0014930026139543765, + "loss": 0.1001, + "step": 34836 + }, + { + "epoch": 0.3024018888724924, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014929758387795984, + "loss": 0.1611, + "step": 34837 + }, + { + "epoch": 0.3024105693526966, + "grad_norm": 0.349609375, + "learning_rate": 0.001492949063175068, + "loss": 0.0986, + "step": 34838 + }, + { + "epoch": 0.30241924983290075, + "grad_norm": 0.5703125, + "learning_rate": 0.0014929222871408142, + "loss": 0.0859, + "step": 34839 + }, + { + "epoch": 0.30242793031310494, + "grad_norm": 0.33203125, + "learning_rate": 0.0014928955106768666, + "loss": 0.1016, + "step": 34840 + }, + { + "epoch": 0.3024366107933091, + "grad_norm": 0.287109375, + "learning_rate": 0.0014928687337832545, + "loss": 0.0957, + "step": 34841 + }, + { + "epoch": 0.3024452912735133, + "grad_norm": 0.73828125, + "learning_rate": 0.0014928419564600069, + "loss": 0.1133, + "step": 34842 + }, + { + "epoch": 0.3024539717537174, + "grad_norm": 0.357421875, + "learning_rate": 0.0014928151787071532, + "loss": 0.0967, + "step": 34843 + }, + { + "epoch": 0.3024626522339216, + "grad_norm": 0.380859375, + "learning_rate": 0.001492788400524723, + "loss": 0.0996, + "step": 34844 + }, + { + "epoch": 0.30247133271412574, + "grad_norm": 0.09619140625, + "learning_rate": 0.001492761621912745, + "loss": 0.0918, + "step": 34845 + }, + { + "epoch": 0.30248001319432993, + "grad_norm": 0.11474609375, + "learning_rate": 0.001492734842871249, + "loss": 0.1123, + "step": 34846 + }, + { + "epoch": 0.30248869367453407, + "grad_norm": 0.462890625, + "learning_rate": 0.001492708063400264, + "loss": 0.0786, + "step": 34847 + }, + { + "epoch": 0.30249737415473826, + "grad_norm": 0.126953125, + "learning_rate": 0.0014926812834998191, + "loss": 0.1484, + "step": 34848 + }, + { + "epoch": 0.3025060546349424, + "grad_norm": 0.359375, + "learning_rate": 0.001492654503169944, + "loss": 0.0986, + "step": 34849 + }, + { + "epoch": 0.3025147351151466, + "grad_norm": 0.193359375, + "learning_rate": 0.0014926277224106682, + "loss": 0.1011, + "step": 34850 + }, + { + "epoch": 0.30252341559535073, + "grad_norm": 0.5, + "learning_rate": 0.0014926009412220204, + "loss": 0.1328, + "step": 34851 + }, + { + "epoch": 0.3025320960755549, + "grad_norm": 0.23828125, + "learning_rate": 0.00149257415960403, + "loss": 0.1279, + "step": 34852 + }, + { + "epoch": 0.30254077655575906, + "grad_norm": 0.125, + "learning_rate": 0.0014925473775567264, + "loss": 0.0962, + "step": 34853 + }, + { + "epoch": 0.30254945703596325, + "grad_norm": 0.33203125, + "learning_rate": 0.001492520595080139, + "loss": 0.0869, + "step": 34854 + }, + { + "epoch": 0.3025581375161674, + "grad_norm": 0.41796875, + "learning_rate": 0.001492493812174297, + "loss": 0.0811, + "step": 34855 + }, + { + "epoch": 0.3025668179963716, + "grad_norm": 0.22265625, + "learning_rate": 0.0014924670288392295, + "loss": 0.1006, + "step": 34856 + }, + { + "epoch": 0.3025754984765757, + "grad_norm": 0.2421875, + "learning_rate": 0.0014924402450749658, + "loss": 0.1299, + "step": 34857 + }, + { + "epoch": 0.3025841789567799, + "grad_norm": 0.337890625, + "learning_rate": 0.0014924134608815358, + "loss": 0.1279, + "step": 34858 + }, + { + "epoch": 0.30259285943698405, + "grad_norm": 0.5859375, + "learning_rate": 0.0014923866762589684, + "loss": 0.1133, + "step": 34859 + }, + { + "epoch": 0.30260153991718824, + "grad_norm": 0.240234375, + "learning_rate": 0.0014923598912072924, + "loss": 0.1167, + "step": 34860 + }, + { + "epoch": 0.3026102203973924, + "grad_norm": 0.478515625, + "learning_rate": 0.0014923331057265375, + "loss": 0.064, + "step": 34861 + }, + { + "epoch": 0.3026189008775966, + "grad_norm": 0.1640625, + "learning_rate": 0.0014923063198167333, + "loss": 0.1387, + "step": 34862 + }, + { + "epoch": 0.3026275813578007, + "grad_norm": 0.62109375, + "learning_rate": 0.0014922795334779088, + "loss": 0.1055, + "step": 34863 + }, + { + "epoch": 0.3026362618380049, + "grad_norm": 0.2021484375, + "learning_rate": 0.001492252746710093, + "loss": 0.0908, + "step": 34864 + }, + { + "epoch": 0.30264494231820904, + "grad_norm": 0.265625, + "learning_rate": 0.0014922259595133156, + "loss": 0.082, + "step": 34865 + }, + { + "epoch": 0.30265362279841324, + "grad_norm": 0.2421875, + "learning_rate": 0.0014921991718876058, + "loss": 0.0869, + "step": 34866 + }, + { + "epoch": 0.3026623032786174, + "grad_norm": 0.095703125, + "learning_rate": 0.0014921723838329931, + "loss": 0.0913, + "step": 34867 + }, + { + "epoch": 0.30267098375882157, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014921455953495065, + "loss": 0.083, + "step": 34868 + }, + { + "epoch": 0.3026796642390257, + "grad_norm": 0.12890625, + "learning_rate": 0.001492118806437175, + "loss": 0.1152, + "step": 34869 + }, + { + "epoch": 0.30268834471922984, + "grad_norm": 1.09375, + "learning_rate": 0.0014920920170960286, + "loss": 0.1719, + "step": 34870 + }, + { + "epoch": 0.30269702519943403, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014920652273260964, + "loss": 0.0713, + "step": 34871 + }, + { + "epoch": 0.30270570567963817, + "grad_norm": 0.4609375, + "learning_rate": 0.0014920384371274072, + "loss": 0.1289, + "step": 34872 + }, + { + "epoch": 0.30271438615984236, + "grad_norm": 0.1650390625, + "learning_rate": 0.0014920116464999908, + "loss": 0.1025, + "step": 34873 + }, + { + "epoch": 0.3027230666400465, + "grad_norm": 0.15234375, + "learning_rate": 0.0014919848554438763, + "loss": 0.0918, + "step": 34874 + }, + { + "epoch": 0.3027317471202507, + "grad_norm": 0.302734375, + "learning_rate": 0.001491958063959093, + "loss": 0.1357, + "step": 34875 + }, + { + "epoch": 0.30274042760045483, + "grad_norm": 0.7578125, + "learning_rate": 0.0014919312720456707, + "loss": 0.1221, + "step": 34876 + }, + { + "epoch": 0.302749108080659, + "grad_norm": 0.72265625, + "learning_rate": 0.0014919044797036378, + "loss": 0.0898, + "step": 34877 + }, + { + "epoch": 0.30275778856086316, + "grad_norm": 0.6640625, + "learning_rate": 0.001491877686933024, + "loss": 0.0986, + "step": 34878 + }, + { + "epoch": 0.30276646904106735, + "grad_norm": 0.43359375, + "learning_rate": 0.001491850893733859, + "loss": 0.1221, + "step": 34879 + }, + { + "epoch": 0.3027751495212715, + "grad_norm": 0.25390625, + "learning_rate": 0.0014918241001061717, + "loss": 0.0742, + "step": 34880 + }, + { + "epoch": 0.3027838300014757, + "grad_norm": 1.296875, + "learning_rate": 0.0014917973060499911, + "loss": 0.1318, + "step": 34881 + }, + { + "epoch": 0.3027925104816798, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014917705115653473, + "loss": 0.1113, + "step": 34882 + }, + { + "epoch": 0.302801190961884, + "grad_norm": 0.3671875, + "learning_rate": 0.0014917437166522688, + "loss": 0.0688, + "step": 34883 + }, + { + "epoch": 0.30280987144208815, + "grad_norm": 0.41015625, + "learning_rate": 0.0014917169213107852, + "loss": 0.082, + "step": 34884 + }, + { + "epoch": 0.30281855192229234, + "grad_norm": 0.287109375, + "learning_rate": 0.0014916901255409261, + "loss": 0.0938, + "step": 34885 + }, + { + "epoch": 0.3028272324024965, + "grad_norm": 0.146484375, + "learning_rate": 0.0014916633293427207, + "loss": 0.1064, + "step": 34886 + }, + { + "epoch": 0.3028359128827007, + "grad_norm": 0.3515625, + "learning_rate": 0.0014916365327161977, + "loss": 0.1123, + "step": 34887 + }, + { + "epoch": 0.3028445933629048, + "grad_norm": 0.2421875, + "learning_rate": 0.0014916097356613871, + "loss": 0.1094, + "step": 34888 + }, + { + "epoch": 0.302853273843109, + "grad_norm": 0.296875, + "learning_rate": 0.0014915829381783182, + "loss": 0.0781, + "step": 34889 + }, + { + "epoch": 0.30286195432331314, + "grad_norm": 0.5234375, + "learning_rate": 0.0014915561402670202, + "loss": 0.1113, + "step": 34890 + }, + { + "epoch": 0.30287063480351734, + "grad_norm": 0.267578125, + "learning_rate": 0.0014915293419275218, + "loss": 0.1064, + "step": 34891 + }, + { + "epoch": 0.3028793152837215, + "grad_norm": 0.515625, + "learning_rate": 0.001491502543159853, + "loss": 0.1465, + "step": 34892 + }, + { + "epoch": 0.30288799576392567, + "grad_norm": 0.79296875, + "learning_rate": 0.0014914757439640429, + "loss": 0.0801, + "step": 34893 + }, + { + "epoch": 0.3028966762441298, + "grad_norm": 1.015625, + "learning_rate": 0.0014914489443401207, + "loss": 0.0742, + "step": 34894 + }, + { + "epoch": 0.302905356724334, + "grad_norm": 0.1796875, + "learning_rate": 0.0014914221442881158, + "loss": 0.1318, + "step": 34895 + }, + { + "epoch": 0.30291403720453813, + "grad_norm": 0.37890625, + "learning_rate": 0.0014913953438080577, + "loss": 0.1387, + "step": 34896 + }, + { + "epoch": 0.3029227176847423, + "grad_norm": 0.69921875, + "learning_rate": 0.0014913685428999755, + "loss": 0.0947, + "step": 34897 + }, + { + "epoch": 0.30293139816494646, + "grad_norm": 0.380859375, + "learning_rate": 0.0014913417415638986, + "loss": 0.1279, + "step": 34898 + }, + { + "epoch": 0.30294007864515066, + "grad_norm": 0.474609375, + "learning_rate": 0.0014913149397998563, + "loss": 0.1289, + "step": 34899 + }, + { + "epoch": 0.3029487591253548, + "grad_norm": 0.306640625, + "learning_rate": 0.0014912881376078773, + "loss": 0.0854, + "step": 34900 + }, + { + "epoch": 0.302957439605559, + "grad_norm": 0.134765625, + "learning_rate": 0.0014912613349879922, + "loss": 0.123, + "step": 34901 + }, + { + "epoch": 0.3029661200857631, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014912345319402292, + "loss": 0.0737, + "step": 34902 + }, + { + "epoch": 0.3029748005659673, + "grad_norm": 0.5078125, + "learning_rate": 0.0014912077284646179, + "loss": 0.1025, + "step": 34903 + }, + { + "epoch": 0.30298348104617145, + "grad_norm": 0.357421875, + "learning_rate": 0.001491180924561188, + "loss": 0.0796, + "step": 34904 + }, + { + "epoch": 0.30299216152637565, + "grad_norm": 0.32421875, + "learning_rate": 0.0014911541202299683, + "loss": 0.0869, + "step": 34905 + }, + { + "epoch": 0.3030008420065798, + "grad_norm": 0.208984375, + "learning_rate": 0.0014911273154709886, + "loss": 0.1094, + "step": 34906 + }, + { + "epoch": 0.303009522486784, + "grad_norm": 0.1904296875, + "learning_rate": 0.0014911005102842777, + "loss": 0.0713, + "step": 34907 + }, + { + "epoch": 0.3030182029669881, + "grad_norm": 0.228515625, + "learning_rate": 0.0014910737046698653, + "loss": 0.1436, + "step": 34908 + }, + { + "epoch": 0.3030268834471923, + "grad_norm": 0.240234375, + "learning_rate": 0.0014910468986277805, + "loss": 0.1123, + "step": 34909 + }, + { + "epoch": 0.30303556392739645, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014910200921580527, + "loss": 0.0933, + "step": 34910 + }, + { + "epoch": 0.30304424440760064, + "grad_norm": 0.1318359375, + "learning_rate": 0.001490993285260711, + "loss": 0.1182, + "step": 34911 + }, + { + "epoch": 0.3030529248878048, + "grad_norm": 0.171875, + "learning_rate": 0.0014909664779357852, + "loss": 0.1055, + "step": 34912 + }, + { + "epoch": 0.30306160536800897, + "grad_norm": 0.3125, + "learning_rate": 0.0014909396701833044, + "loss": 0.1191, + "step": 34913 + }, + { + "epoch": 0.3030702858482131, + "grad_norm": 0.51171875, + "learning_rate": 0.0014909128620032974, + "loss": 0.1152, + "step": 34914 + }, + { + "epoch": 0.3030789663284173, + "grad_norm": 0.64453125, + "learning_rate": 0.0014908860533957944, + "loss": 0.1328, + "step": 34915 + }, + { + "epoch": 0.30308764680862144, + "grad_norm": 0.2333984375, + "learning_rate": 0.0014908592443608243, + "loss": 0.0957, + "step": 34916 + }, + { + "epoch": 0.30309632728882563, + "grad_norm": 0.06689453125, + "learning_rate": 0.0014908324348984162, + "loss": 0.0723, + "step": 34917 + }, + { + "epoch": 0.30310500776902977, + "grad_norm": 0.0791015625, + "learning_rate": 0.0014908056250085996, + "loss": 0.1426, + "step": 34918 + }, + { + "epoch": 0.30311368824923396, + "grad_norm": 0.2021484375, + "learning_rate": 0.001490778814691404, + "loss": 0.0967, + "step": 34919 + }, + { + "epoch": 0.3031223687294381, + "grad_norm": 0.357421875, + "learning_rate": 0.0014907520039468586, + "loss": 0.0918, + "step": 34920 + }, + { + "epoch": 0.3031310492096423, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014907251927749926, + "loss": 0.1143, + "step": 34921 + }, + { + "epoch": 0.3031397296898464, + "grad_norm": 0.71484375, + "learning_rate": 0.001490698381175835, + "loss": 0.123, + "step": 34922 + }, + { + "epoch": 0.3031484101700506, + "grad_norm": 0.203125, + "learning_rate": 0.001490671569149416, + "loss": 0.0942, + "step": 34923 + }, + { + "epoch": 0.30315709065025476, + "grad_norm": 0.494140625, + "learning_rate": 0.0014906447566957644, + "loss": 0.0962, + "step": 34924 + }, + { + "epoch": 0.30316577113045895, + "grad_norm": 0.07666015625, + "learning_rate": 0.0014906179438149093, + "loss": 0.1533, + "step": 34925 + }, + { + "epoch": 0.3031744516106631, + "grad_norm": 0.236328125, + "learning_rate": 0.0014905911305068804, + "loss": 0.0625, + "step": 34926 + }, + { + "epoch": 0.3031831320908673, + "grad_norm": 0.2119140625, + "learning_rate": 0.001490564316771707, + "loss": 0.0884, + "step": 34927 + }, + { + "epoch": 0.3031918125710714, + "grad_norm": 0.2119140625, + "learning_rate": 0.0014905375026094184, + "loss": 0.0938, + "step": 34928 + }, + { + "epoch": 0.3032004930512756, + "grad_norm": 0.1640625, + "learning_rate": 0.001490510688020044, + "loss": 0.0864, + "step": 34929 + }, + { + "epoch": 0.30320917353147975, + "grad_norm": 0.103515625, + "learning_rate": 0.0014904838730036122, + "loss": 0.1011, + "step": 34930 + }, + { + "epoch": 0.30321785401168394, + "grad_norm": 0.248046875, + "learning_rate": 0.0014904570575601538, + "loss": 0.1187, + "step": 34931 + }, + { + "epoch": 0.3032265344918881, + "grad_norm": 0.2119140625, + "learning_rate": 0.0014904302416896974, + "loss": 0.1104, + "step": 34932 + }, + { + "epoch": 0.30323521497209227, + "grad_norm": 0.2392578125, + "learning_rate": 0.0014904034253922721, + "loss": 0.0859, + "step": 34933 + }, + { + "epoch": 0.3032438954522964, + "grad_norm": 0.224609375, + "learning_rate": 0.0014903766086679078, + "loss": 0.1143, + "step": 34934 + }, + { + "epoch": 0.3032525759325006, + "grad_norm": 0.30078125, + "learning_rate": 0.001490349791516633, + "loss": 0.106, + "step": 34935 + }, + { + "epoch": 0.30326125641270474, + "grad_norm": 0.54296875, + "learning_rate": 0.0014903229739384781, + "loss": 0.1074, + "step": 34936 + }, + { + "epoch": 0.30326993689290893, + "grad_norm": 0.15234375, + "learning_rate": 0.0014902961559334715, + "loss": 0.0908, + "step": 34937 + }, + { + "epoch": 0.30327861737311307, + "grad_norm": 0.46875, + "learning_rate": 0.001490269337501643, + "loss": 0.0918, + "step": 34938 + }, + { + "epoch": 0.30328729785331726, + "grad_norm": 0.4921875, + "learning_rate": 0.0014902425186430219, + "loss": 0.1338, + "step": 34939 + }, + { + "epoch": 0.3032959783335214, + "grad_norm": 1.125, + "learning_rate": 0.0014902156993576374, + "loss": 0.1631, + "step": 34940 + }, + { + "epoch": 0.3033046588137256, + "grad_norm": 0.396484375, + "learning_rate": 0.001490188879645519, + "loss": 0.0825, + "step": 34941 + }, + { + "epoch": 0.30331333929392973, + "grad_norm": 0.25, + "learning_rate": 0.0014901620595066953, + "loss": 0.0698, + "step": 34942 + }, + { + "epoch": 0.3033220197741339, + "grad_norm": 0.271484375, + "learning_rate": 0.0014901352389411968, + "loss": 0.1367, + "step": 34943 + }, + { + "epoch": 0.30333070025433806, + "grad_norm": 0.4609375, + "learning_rate": 0.001490108417949052, + "loss": 0.0869, + "step": 34944 + }, + { + "epoch": 0.30333938073454225, + "grad_norm": 0.322265625, + "learning_rate": 0.0014900815965302907, + "loss": 0.1025, + "step": 34945 + }, + { + "epoch": 0.3033480612147464, + "grad_norm": 0.087890625, + "learning_rate": 0.001490054774684942, + "loss": 0.0859, + "step": 34946 + }, + { + "epoch": 0.3033567416949506, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014900279524130353, + "loss": 0.1035, + "step": 34947 + }, + { + "epoch": 0.3033654221751547, + "grad_norm": 0.29296875, + "learning_rate": 0.0014900011297145997, + "loss": 0.1025, + "step": 34948 + }, + { + "epoch": 0.3033741026553589, + "grad_norm": 0.6875, + "learning_rate": 0.001489974306589665, + "loss": 0.0815, + "step": 34949 + }, + { + "epoch": 0.30338278313556305, + "grad_norm": 0.1279296875, + "learning_rate": 0.00148994748303826, + "loss": 0.1084, + "step": 34950 + }, + { + "epoch": 0.30339146361576724, + "grad_norm": 0.28515625, + "learning_rate": 0.0014899206590604144, + "loss": 0.0742, + "step": 34951 + }, + { + "epoch": 0.3034001440959714, + "grad_norm": 0.353515625, + "learning_rate": 0.001489893834656157, + "loss": 0.0938, + "step": 34952 + }, + { + "epoch": 0.3034088245761756, + "grad_norm": 0.11669921875, + "learning_rate": 0.0014898670098255182, + "loss": 0.0908, + "step": 34953 + }, + { + "epoch": 0.3034175050563797, + "grad_norm": 0.0927734375, + "learning_rate": 0.0014898401845685266, + "loss": 0.127, + "step": 34954 + }, + { + "epoch": 0.3034261855365839, + "grad_norm": 0.3203125, + "learning_rate": 0.0014898133588852115, + "loss": 0.1143, + "step": 34955 + }, + { + "epoch": 0.30343486601678804, + "grad_norm": 0.59765625, + "learning_rate": 0.0014897865327756022, + "loss": 0.1641, + "step": 34956 + }, + { + "epoch": 0.30344354649699223, + "grad_norm": 0.08251953125, + "learning_rate": 0.0014897597062397284, + "loss": 0.0825, + "step": 34957 + }, + { + "epoch": 0.30345222697719637, + "grad_norm": 0.19140625, + "learning_rate": 0.0014897328792776192, + "loss": 0.0952, + "step": 34958 + }, + { + "epoch": 0.30346090745740056, + "grad_norm": 0.3515625, + "learning_rate": 0.001489706051889304, + "loss": 0.25, + "step": 34959 + }, + { + "epoch": 0.3034695879376047, + "grad_norm": 0.27734375, + "learning_rate": 0.0014896792240748121, + "loss": 0.1387, + "step": 34960 + }, + { + "epoch": 0.3034782684178089, + "grad_norm": 0.2080078125, + "learning_rate": 0.001489652395834173, + "loss": 0.0972, + "step": 34961 + }, + { + "epoch": 0.30348694889801303, + "grad_norm": 0.46875, + "learning_rate": 0.0014896255671674157, + "loss": 0.1113, + "step": 34962 + }, + { + "epoch": 0.3034956293782172, + "grad_norm": 0.474609375, + "learning_rate": 0.0014895987380745697, + "loss": 0.0967, + "step": 34963 + }, + { + "epoch": 0.30350430985842136, + "grad_norm": 0.1279296875, + "learning_rate": 0.0014895719085556646, + "loss": 0.0845, + "step": 34964 + }, + { + "epoch": 0.30351299033862555, + "grad_norm": 0.26171875, + "learning_rate": 0.0014895450786107291, + "loss": 0.1016, + "step": 34965 + }, + { + "epoch": 0.3035216708188297, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014895182482397933, + "loss": 0.1289, + "step": 34966 + }, + { + "epoch": 0.3035303512990339, + "grad_norm": 0.0966796875, + "learning_rate": 0.0014894914174428862, + "loss": 0.0728, + "step": 34967 + }, + { + "epoch": 0.303539031779238, + "grad_norm": 0.375, + "learning_rate": 0.0014894645862200372, + "loss": 0.0947, + "step": 34968 + }, + { + "epoch": 0.3035477122594422, + "grad_norm": 0.34375, + "learning_rate": 0.0014894377545712752, + "loss": 0.0806, + "step": 34969 + }, + { + "epoch": 0.30355639273964635, + "grad_norm": 0.138671875, + "learning_rate": 0.0014894109224966303, + "loss": 0.1074, + "step": 34970 + }, + { + "epoch": 0.30356507321985055, + "grad_norm": 0.48828125, + "learning_rate": 0.001489384089996131, + "loss": 0.1045, + "step": 34971 + }, + { + "epoch": 0.3035737537000547, + "grad_norm": 0.228515625, + "learning_rate": 0.0014893572570698077, + "loss": 0.0913, + "step": 34972 + }, + { + "epoch": 0.3035824341802589, + "grad_norm": 1.28125, + "learning_rate": 0.0014893304237176888, + "loss": 0.207, + "step": 34973 + }, + { + "epoch": 0.303591114660463, + "grad_norm": 0.07373046875, + "learning_rate": 0.001489303589939804, + "loss": 0.0835, + "step": 34974 + }, + { + "epoch": 0.3035997951406672, + "grad_norm": 0.11962890625, + "learning_rate": 0.0014892767557361827, + "loss": 0.0967, + "step": 34975 + }, + { + "epoch": 0.30360847562087134, + "grad_norm": 0.78515625, + "learning_rate": 0.0014892499211068544, + "loss": 0.1523, + "step": 34976 + }, + { + "epoch": 0.30361715610107554, + "grad_norm": 0.0771484375, + "learning_rate": 0.0014892230860518482, + "loss": 0.0947, + "step": 34977 + }, + { + "epoch": 0.3036258365812797, + "grad_norm": 0.1689453125, + "learning_rate": 0.001489196250571193, + "loss": 0.0977, + "step": 34978 + }, + { + "epoch": 0.30363451706148387, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014891694146649192, + "loss": 0.0879, + "step": 34979 + }, + { + "epoch": 0.303643197541688, + "grad_norm": 0.30859375, + "learning_rate": 0.0014891425783330552, + "loss": 0.0957, + "step": 34980 + }, + { + "epoch": 0.3036518780218922, + "grad_norm": 0.609375, + "learning_rate": 0.0014891157415756308, + "loss": 0.0732, + "step": 34981 + }, + { + "epoch": 0.30366055850209633, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014890889043926755, + "loss": 0.0781, + "step": 34982 + }, + { + "epoch": 0.3036692389823005, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014890620667842181, + "loss": 0.1035, + "step": 34983 + }, + { + "epoch": 0.30367791946250466, + "grad_norm": 0.10546875, + "learning_rate": 0.0014890352287502883, + "loss": 0.1387, + "step": 34984 + }, + { + "epoch": 0.30368659994270886, + "grad_norm": 0.3125, + "learning_rate": 0.0014890083902909157, + "loss": 0.1309, + "step": 34985 + }, + { + "epoch": 0.303695280422913, + "grad_norm": 0.23828125, + "learning_rate": 0.0014889815514061294, + "loss": 0.1221, + "step": 34986 + }, + { + "epoch": 0.3037039609031172, + "grad_norm": 0.671875, + "learning_rate": 0.0014889547120959584, + "loss": 0.0693, + "step": 34987 + }, + { + "epoch": 0.3037126413833213, + "grad_norm": 0.22265625, + "learning_rate": 0.0014889278723604325, + "loss": 0.0972, + "step": 34988 + }, + { + "epoch": 0.3037213218635255, + "grad_norm": 0.25390625, + "learning_rate": 0.0014889010321995808, + "loss": 0.1396, + "step": 34989 + }, + { + "epoch": 0.30373000234372965, + "grad_norm": 0.419921875, + "learning_rate": 0.0014888741916134329, + "loss": 0.1406, + "step": 34990 + }, + { + "epoch": 0.30373868282393385, + "grad_norm": 0.1650390625, + "learning_rate": 0.001488847350602018, + "loss": 0.1416, + "step": 34991 + }, + { + "epoch": 0.303747363304138, + "grad_norm": 0.3515625, + "learning_rate": 0.0014888205091653655, + "loss": 0.0747, + "step": 34992 + }, + { + "epoch": 0.3037560437843421, + "grad_norm": 0.1474609375, + "learning_rate": 0.0014887936673035047, + "loss": 0.105, + "step": 34993 + }, + { + "epoch": 0.3037647242645463, + "grad_norm": 0.578125, + "learning_rate": 0.0014887668250164652, + "loss": 0.2871, + "step": 34994 + }, + { + "epoch": 0.30377340474475045, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014887399823042759, + "loss": 0.0991, + "step": 34995 + }, + { + "epoch": 0.30378208522495465, + "grad_norm": 0.73046875, + "learning_rate": 0.0014887131391669666, + "loss": 0.084, + "step": 34996 + }, + { + "epoch": 0.3037907657051588, + "grad_norm": 0.2060546875, + "learning_rate": 0.0014886862956045661, + "loss": 0.0884, + "step": 34997 + }, + { + "epoch": 0.303799446185363, + "grad_norm": 0.22265625, + "learning_rate": 0.0014886594516171046, + "loss": 0.1143, + "step": 34998 + }, + { + "epoch": 0.3038081266655671, + "grad_norm": 0.412109375, + "learning_rate": 0.0014886326072046103, + "loss": 0.1113, + "step": 34999 + }, + { + "epoch": 0.3038168071457713, + "grad_norm": 0.1845703125, + "learning_rate": 0.0014886057623671137, + "loss": 0.0703, + "step": 35000 + }, + { + "epoch": 0.30382548762597544, + "grad_norm": 0.455078125, + "learning_rate": 0.0014885789171046435, + "loss": 0.1719, + "step": 35001 + }, + { + "epoch": 0.30383416810617964, + "grad_norm": 0.28125, + "learning_rate": 0.0014885520714172293, + "loss": 0.1562, + "step": 35002 + }, + { + "epoch": 0.3038428485863838, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014885252253049007, + "loss": 0.126, + "step": 35003 + }, + { + "epoch": 0.30385152906658797, + "grad_norm": 0.19921875, + "learning_rate": 0.0014884983787676865, + "loss": 0.1299, + "step": 35004 + }, + { + "epoch": 0.3038602095467921, + "grad_norm": 0.10009765625, + "learning_rate": 0.0014884715318056161, + "loss": 0.124, + "step": 35005 + }, + { + "epoch": 0.3038688900269963, + "grad_norm": 0.427734375, + "learning_rate": 0.0014884446844187194, + "loss": 0.0806, + "step": 35006 + }, + { + "epoch": 0.30387757050720043, + "grad_norm": 0.20703125, + "learning_rate": 0.0014884178366070253, + "loss": 0.1035, + "step": 35007 + }, + { + "epoch": 0.3038862509874046, + "grad_norm": 0.1328125, + "learning_rate": 0.0014883909883705632, + "loss": 0.1079, + "step": 35008 + }, + { + "epoch": 0.30389493146760876, + "grad_norm": 0.388671875, + "learning_rate": 0.0014883641397093624, + "loss": 0.0972, + "step": 35009 + }, + { + "epoch": 0.30390361194781296, + "grad_norm": 0.21875, + "learning_rate": 0.0014883372906234528, + "loss": 0.0703, + "step": 35010 + }, + { + "epoch": 0.3039122924280171, + "grad_norm": 0.1630859375, + "learning_rate": 0.001488310441112863, + "loss": 0.1191, + "step": 35011 + }, + { + "epoch": 0.3039209729082213, + "grad_norm": 0.1923828125, + "learning_rate": 0.001488283591177623, + "loss": 0.1113, + "step": 35012 + }, + { + "epoch": 0.3039296533884254, + "grad_norm": 0.169921875, + "learning_rate": 0.0014882567408177618, + "loss": 0.0645, + "step": 35013 + }, + { + "epoch": 0.3039383338686296, + "grad_norm": 0.6015625, + "learning_rate": 0.001488229890033309, + "loss": 0.0918, + "step": 35014 + }, + { + "epoch": 0.30394701434883376, + "grad_norm": 0.3203125, + "learning_rate": 0.0014882030388242938, + "loss": 0.0869, + "step": 35015 + }, + { + "epoch": 0.30395569482903795, + "grad_norm": 0.2578125, + "learning_rate": 0.0014881761871907455, + "loss": 0.0703, + "step": 35016 + }, + { + "epoch": 0.3039643753092421, + "grad_norm": 0.0703125, + "learning_rate": 0.0014881493351326935, + "loss": 0.0737, + "step": 35017 + }, + { + "epoch": 0.3039730557894463, + "grad_norm": 0.18359375, + "learning_rate": 0.0014881224826501674, + "loss": 0.1426, + "step": 35018 + }, + { + "epoch": 0.3039817362696504, + "grad_norm": 0.20703125, + "learning_rate": 0.0014880956297431962, + "loss": 0.1221, + "step": 35019 + }, + { + "epoch": 0.3039904167498546, + "grad_norm": 0.6328125, + "learning_rate": 0.0014880687764118096, + "loss": 0.0898, + "step": 35020 + }, + { + "epoch": 0.30399909723005875, + "grad_norm": 0.515625, + "learning_rate": 0.0014880419226560367, + "loss": 0.1289, + "step": 35021 + }, + { + "epoch": 0.30400777771026294, + "grad_norm": 0.6875, + "learning_rate": 0.0014880150684759069, + "loss": 0.1455, + "step": 35022 + }, + { + "epoch": 0.3040164581904671, + "grad_norm": 0.310546875, + "learning_rate": 0.0014879882138714498, + "loss": 0.106, + "step": 35023 + }, + { + "epoch": 0.30402513867067127, + "grad_norm": 0.546875, + "learning_rate": 0.0014879613588426946, + "loss": 0.0796, + "step": 35024 + }, + { + "epoch": 0.3040338191508754, + "grad_norm": 0.3671875, + "learning_rate": 0.0014879345033896707, + "loss": 0.1118, + "step": 35025 + }, + { + "epoch": 0.3040424996310796, + "grad_norm": 0.51953125, + "learning_rate": 0.0014879076475124073, + "loss": 0.1206, + "step": 35026 + }, + { + "epoch": 0.30405118011128374, + "grad_norm": 0.0966796875, + "learning_rate": 0.001487880791210934, + "loss": 0.0933, + "step": 35027 + }, + { + "epoch": 0.30405986059148793, + "grad_norm": 0.2431640625, + "learning_rate": 0.00148785393448528, + "loss": 0.0957, + "step": 35028 + }, + { + "epoch": 0.30406854107169207, + "grad_norm": 0.9375, + "learning_rate": 0.0014878270773354752, + "loss": 0.0806, + "step": 35029 + }, + { + "epoch": 0.30407722155189626, + "grad_norm": 0.14453125, + "learning_rate": 0.001487800219761548, + "loss": 0.0654, + "step": 35030 + }, + { + "epoch": 0.3040859020321004, + "grad_norm": 0.12158203125, + "learning_rate": 0.0014877733617635288, + "loss": 0.1084, + "step": 35031 + }, + { + "epoch": 0.3040945825123046, + "grad_norm": 0.419921875, + "learning_rate": 0.001487746503341446, + "loss": 0.0713, + "step": 35032 + }, + { + "epoch": 0.3041032629925087, + "grad_norm": 0.08447265625, + "learning_rate": 0.0014877196444953301, + "loss": 0.0815, + "step": 35033 + }, + { + "epoch": 0.3041119434727129, + "grad_norm": 0.447265625, + "learning_rate": 0.0014876927852252092, + "loss": 0.1221, + "step": 35034 + }, + { + "epoch": 0.30412062395291706, + "grad_norm": 0.0830078125, + "learning_rate": 0.0014876659255311135, + "loss": 0.0781, + "step": 35035 + }, + { + "epoch": 0.30412930443312125, + "grad_norm": 0.1181640625, + "learning_rate": 0.0014876390654130721, + "loss": 0.1201, + "step": 35036 + }, + { + "epoch": 0.3041379849133254, + "grad_norm": 0.33203125, + "learning_rate": 0.0014876122048711146, + "loss": 0.1621, + "step": 35037 + }, + { + "epoch": 0.3041466653935296, + "grad_norm": 0.359375, + "learning_rate": 0.00148758534390527, + "loss": 0.1426, + "step": 35038 + }, + { + "epoch": 0.3041553458737337, + "grad_norm": 0.171875, + "learning_rate": 0.001487558482515568, + "loss": 0.1201, + "step": 35039 + }, + { + "epoch": 0.3041640263539379, + "grad_norm": 0.734375, + "learning_rate": 0.0014875316207020378, + "loss": 0.1016, + "step": 35040 + }, + { + "epoch": 0.30417270683414205, + "grad_norm": 0.357421875, + "learning_rate": 0.001487504758464709, + "loss": 0.1074, + "step": 35041 + }, + { + "epoch": 0.30418138731434624, + "grad_norm": 0.53125, + "learning_rate": 0.001487477895803611, + "loss": 0.105, + "step": 35042 + }, + { + "epoch": 0.3041900677945504, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014874510327187724, + "loss": 0.1074, + "step": 35043 + }, + { + "epoch": 0.30419874827475457, + "grad_norm": 0.27734375, + "learning_rate": 0.0014874241692102235, + "loss": 0.1045, + "step": 35044 + }, + { + "epoch": 0.3042074287549587, + "grad_norm": 0.0869140625, + "learning_rate": 0.0014873973052779938, + "loss": 0.0762, + "step": 35045 + }, + { + "epoch": 0.3042161092351629, + "grad_norm": 0.08349609375, + "learning_rate": 0.0014873704409221116, + "loss": 0.083, + "step": 35046 + }, + { + "epoch": 0.30422478971536704, + "grad_norm": 0.1455078125, + "learning_rate": 0.001487343576142607, + "loss": 0.0776, + "step": 35047 + }, + { + "epoch": 0.30423347019557123, + "grad_norm": 0.18359375, + "learning_rate": 0.0014873167109395093, + "loss": 0.1196, + "step": 35048 + }, + { + "epoch": 0.30424215067577537, + "grad_norm": 0.150390625, + "learning_rate": 0.001487289845312848, + "loss": 0.0869, + "step": 35049 + }, + { + "epoch": 0.30425083115597956, + "grad_norm": 0.27734375, + "learning_rate": 0.0014872629792626521, + "loss": 0.0684, + "step": 35050 + }, + { + "epoch": 0.3042595116361837, + "grad_norm": 0.416015625, + "learning_rate": 0.0014872361127889516, + "loss": 0.0903, + "step": 35051 + }, + { + "epoch": 0.3042681921163879, + "grad_norm": 0.26171875, + "learning_rate": 0.0014872092458917752, + "loss": 0.1367, + "step": 35052 + }, + { + "epoch": 0.30427687259659203, + "grad_norm": 0.09228515625, + "learning_rate": 0.0014871823785711526, + "loss": 0.1143, + "step": 35053 + }, + { + "epoch": 0.3042855530767962, + "grad_norm": 0.5546875, + "learning_rate": 0.0014871555108271135, + "loss": 0.1201, + "step": 35054 + }, + { + "epoch": 0.30429423355700036, + "grad_norm": 0.9765625, + "learning_rate": 0.0014871286426596869, + "loss": 0.0957, + "step": 35055 + }, + { + "epoch": 0.30430291403720455, + "grad_norm": 0.28515625, + "learning_rate": 0.0014871017740689019, + "loss": 0.1377, + "step": 35056 + }, + { + "epoch": 0.3043115945174087, + "grad_norm": 0.294921875, + "learning_rate": 0.0014870749050547882, + "loss": 0.1367, + "step": 35057 + }, + { + "epoch": 0.3043202749976129, + "grad_norm": 0.92578125, + "learning_rate": 0.0014870480356173755, + "loss": 0.1094, + "step": 35058 + }, + { + "epoch": 0.304328955477817, + "grad_norm": 0.3125, + "learning_rate": 0.0014870211657566927, + "loss": 0.0938, + "step": 35059 + }, + { + "epoch": 0.3043376359580212, + "grad_norm": 0.0966796875, + "learning_rate": 0.0014869942954727692, + "loss": 0.1045, + "step": 35060 + }, + { + "epoch": 0.30434631643822535, + "grad_norm": 0.427734375, + "learning_rate": 0.001486967424765635, + "loss": 0.1113, + "step": 35061 + }, + { + "epoch": 0.30435499691842954, + "grad_norm": 0.2216796875, + "learning_rate": 0.001486940553635319, + "loss": 0.1318, + "step": 35062 + }, + { + "epoch": 0.3043636773986337, + "grad_norm": 0.6484375, + "learning_rate": 0.0014869136820818503, + "loss": 0.1221, + "step": 35063 + }, + { + "epoch": 0.3043723578788379, + "grad_norm": 0.07763671875, + "learning_rate": 0.0014868868101052587, + "loss": 0.1104, + "step": 35064 + }, + { + "epoch": 0.304381038359042, + "grad_norm": 0.33203125, + "learning_rate": 0.0014868599377055737, + "loss": 0.0723, + "step": 35065 + }, + { + "epoch": 0.3043897188392462, + "grad_norm": 0.2255859375, + "learning_rate": 0.001486833064882824, + "loss": 0.166, + "step": 35066 + }, + { + "epoch": 0.30439839931945034, + "grad_norm": 0.177734375, + "learning_rate": 0.0014868061916370398, + "loss": 0.1523, + "step": 35067 + }, + { + "epoch": 0.30440707979965453, + "grad_norm": 0.0849609375, + "learning_rate": 0.0014867793179682504, + "loss": 0.0811, + "step": 35068 + }, + { + "epoch": 0.30441576027985867, + "grad_norm": 0.11181640625, + "learning_rate": 0.0014867524438764847, + "loss": 0.1182, + "step": 35069 + }, + { + "epoch": 0.30442444076006286, + "grad_norm": 0.259765625, + "learning_rate": 0.0014867255693617723, + "loss": 0.0732, + "step": 35070 + }, + { + "epoch": 0.304433121240267, + "grad_norm": 0.09326171875, + "learning_rate": 0.0014866986944241428, + "loss": 0.085, + "step": 35071 + }, + { + "epoch": 0.3044418017204712, + "grad_norm": 0.53125, + "learning_rate": 0.0014866718190636253, + "loss": 0.0908, + "step": 35072 + }, + { + "epoch": 0.30445048220067533, + "grad_norm": 0.435546875, + "learning_rate": 0.0014866449432802491, + "loss": 0.1113, + "step": 35073 + }, + { + "epoch": 0.3044591626808795, + "grad_norm": 0.263671875, + "learning_rate": 0.0014866180670740441, + "loss": 0.083, + "step": 35074 + }, + { + "epoch": 0.30446784316108366, + "grad_norm": 0.578125, + "learning_rate": 0.0014865911904450393, + "loss": 0.0942, + "step": 35075 + }, + { + "epoch": 0.30447652364128786, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014865643133932644, + "loss": 0.127, + "step": 35076 + }, + { + "epoch": 0.304485204121492, + "grad_norm": 0.82421875, + "learning_rate": 0.001486537435918748, + "loss": 0.1309, + "step": 35077 + }, + { + "epoch": 0.3044938846016962, + "grad_norm": 0.38671875, + "learning_rate": 0.0014865105580215206, + "loss": 0.0757, + "step": 35078 + }, + { + "epoch": 0.3045025650819003, + "grad_norm": 0.384765625, + "learning_rate": 0.0014864836797016108, + "loss": 0.1172, + "step": 35079 + }, + { + "epoch": 0.3045112455621045, + "grad_norm": 0.10498046875, + "learning_rate": 0.0014864568009590484, + "loss": 0.0874, + "step": 35080 + }, + { + "epoch": 0.30451992604230865, + "grad_norm": 0.146484375, + "learning_rate": 0.0014864299217938626, + "loss": 0.1309, + "step": 35081 + }, + { + "epoch": 0.30452860652251285, + "grad_norm": 0.142578125, + "learning_rate": 0.0014864030422060829, + "loss": 0.1113, + "step": 35082 + }, + { + "epoch": 0.304537287002717, + "grad_norm": 0.29296875, + "learning_rate": 0.0014863761621957386, + "loss": 0.1299, + "step": 35083 + }, + { + "epoch": 0.3045459674829212, + "grad_norm": 0.34375, + "learning_rate": 0.0014863492817628592, + "loss": 0.1206, + "step": 35084 + }, + { + "epoch": 0.3045546479631253, + "grad_norm": 0.65625, + "learning_rate": 0.0014863224009074739, + "loss": 0.0986, + "step": 35085 + }, + { + "epoch": 0.3045633284433295, + "grad_norm": 0.9296875, + "learning_rate": 0.001486295519629612, + "loss": 0.1338, + "step": 35086 + }, + { + "epoch": 0.30457200892353364, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014862686379293034, + "loss": 0.0801, + "step": 35087 + }, + { + "epoch": 0.30458068940373784, + "grad_norm": 0.28125, + "learning_rate": 0.0014862417558065772, + "loss": 0.1523, + "step": 35088 + }, + { + "epoch": 0.304589369883942, + "grad_norm": 0.1357421875, + "learning_rate": 0.001486214873261463, + "loss": 0.1147, + "step": 35089 + }, + { + "epoch": 0.30459805036414617, + "grad_norm": 0.0869140625, + "learning_rate": 0.0014861879902939897, + "loss": 0.1357, + "step": 35090 + }, + { + "epoch": 0.3046067308443503, + "grad_norm": 0.45703125, + "learning_rate": 0.001486161106904187, + "loss": 0.1113, + "step": 35091 + }, + { + "epoch": 0.3046154113245545, + "grad_norm": 0.54296875, + "learning_rate": 0.0014861342230920846, + "loss": 0.1758, + "step": 35092 + }, + { + "epoch": 0.30462409180475863, + "grad_norm": 0.181640625, + "learning_rate": 0.0014861073388577116, + "loss": 0.0967, + "step": 35093 + }, + { + "epoch": 0.3046327722849628, + "grad_norm": 0.5859375, + "learning_rate": 0.0014860804542010972, + "loss": 0.103, + "step": 35094 + }, + { + "epoch": 0.30464145276516696, + "grad_norm": 0.54296875, + "learning_rate": 0.0014860535691222712, + "loss": 0.1035, + "step": 35095 + }, + { + "epoch": 0.30465013324537116, + "grad_norm": 0.5546875, + "learning_rate": 0.0014860266836212626, + "loss": 0.1221, + "step": 35096 + }, + { + "epoch": 0.3046588137255753, + "grad_norm": 0.275390625, + "learning_rate": 0.0014859997976981012, + "loss": 0.1104, + "step": 35097 + }, + { + "epoch": 0.3046674942057795, + "grad_norm": 0.12255859375, + "learning_rate": 0.0014859729113528162, + "loss": 0.1338, + "step": 35098 + }, + { + "epoch": 0.3046761746859836, + "grad_norm": 0.400390625, + "learning_rate": 0.0014859460245854371, + "loss": 0.1025, + "step": 35099 + }, + { + "epoch": 0.3046848551661878, + "grad_norm": 0.953125, + "learning_rate": 0.001485919137395993, + "loss": 0.1113, + "step": 35100 + }, + { + "epoch": 0.30469353564639196, + "grad_norm": 0.283203125, + "learning_rate": 0.0014858922497845138, + "loss": 0.0947, + "step": 35101 + }, + { + "epoch": 0.30470221612659615, + "grad_norm": 0.609375, + "learning_rate": 0.0014858653617510285, + "loss": 0.1348, + "step": 35102 + }, + { + "epoch": 0.3047108966068003, + "grad_norm": 0.1455078125, + "learning_rate": 0.0014858384732955665, + "loss": 0.1055, + "step": 35103 + }, + { + "epoch": 0.3047195770870045, + "grad_norm": 0.212890625, + "learning_rate": 0.0014858115844181574, + "loss": 0.1367, + "step": 35104 + }, + { + "epoch": 0.3047282575672086, + "grad_norm": 0.53125, + "learning_rate": 0.0014857846951188307, + "loss": 0.0977, + "step": 35105 + }, + { + "epoch": 0.3047369380474128, + "grad_norm": 0.255859375, + "learning_rate": 0.0014857578053976154, + "loss": 0.1543, + "step": 35106 + }, + { + "epoch": 0.30474561852761695, + "grad_norm": 0.275390625, + "learning_rate": 0.0014857309152545414, + "loss": 0.1025, + "step": 35107 + }, + { + "epoch": 0.30475429900782114, + "grad_norm": 0.51171875, + "learning_rate": 0.001485704024689638, + "loss": 0.1748, + "step": 35108 + }, + { + "epoch": 0.3047629794880253, + "grad_norm": 0.5859375, + "learning_rate": 0.0014856771337029343, + "loss": 0.2461, + "step": 35109 + }, + { + "epoch": 0.30477165996822947, + "grad_norm": 0.44921875, + "learning_rate": 0.0014856502422944597, + "loss": 0.0815, + "step": 35110 + }, + { + "epoch": 0.3047803404484336, + "grad_norm": 0.294921875, + "learning_rate": 0.001485623350464244, + "loss": 0.1133, + "step": 35111 + }, + { + "epoch": 0.3047890209286378, + "grad_norm": 1.3515625, + "learning_rate": 0.0014855964582123162, + "loss": 0.1143, + "step": 35112 + }, + { + "epoch": 0.30479770140884194, + "grad_norm": 0.23828125, + "learning_rate": 0.0014855695655387063, + "loss": 0.1357, + "step": 35113 + }, + { + "epoch": 0.30480638188904613, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014855426724434426, + "loss": 0.1113, + "step": 35114 + }, + { + "epoch": 0.30481506236925027, + "grad_norm": 0.30859375, + "learning_rate": 0.001485515778926556, + "loss": 0.1279, + "step": 35115 + }, + { + "epoch": 0.3048237428494544, + "grad_norm": 0.1328125, + "learning_rate": 0.0014854888849880749, + "loss": 0.1055, + "step": 35116 + }, + { + "epoch": 0.3048324233296586, + "grad_norm": 0.166015625, + "learning_rate": 0.0014854619906280286, + "loss": 0.1309, + "step": 35117 + }, + { + "epoch": 0.30484110380986273, + "grad_norm": 0.42578125, + "learning_rate": 0.0014854350958464473, + "loss": 0.1191, + "step": 35118 + }, + { + "epoch": 0.3048497842900669, + "grad_norm": 0.28125, + "learning_rate": 0.0014854082006433597, + "loss": 0.1084, + "step": 35119 + }, + { + "epoch": 0.30485846477027106, + "grad_norm": 0.267578125, + "learning_rate": 0.0014853813050187958, + "loss": 0.0791, + "step": 35120 + }, + { + "epoch": 0.30486714525047526, + "grad_norm": 0.2421875, + "learning_rate": 0.0014853544089727844, + "loss": 0.1396, + "step": 35121 + }, + { + "epoch": 0.3048758257306794, + "grad_norm": 0.1064453125, + "learning_rate": 0.0014853275125053553, + "loss": 0.1074, + "step": 35122 + }, + { + "epoch": 0.3048845062108836, + "grad_norm": 0.6015625, + "learning_rate": 0.0014853006156165382, + "loss": 0.1279, + "step": 35123 + }, + { + "epoch": 0.3048931866910877, + "grad_norm": 0.2431640625, + "learning_rate": 0.001485273718306361, + "loss": 0.1196, + "step": 35124 + }, + { + "epoch": 0.3049018671712919, + "grad_norm": 0.1767578125, + "learning_rate": 0.001485246820574855, + "loss": 0.1133, + "step": 35125 + }, + { + "epoch": 0.30491054765149606, + "grad_norm": 0.2470703125, + "learning_rate": 0.0014852199224220492, + "loss": 0.127, + "step": 35126 + }, + { + "epoch": 0.30491922813170025, + "grad_norm": 0.5, + "learning_rate": 0.0014851930238479723, + "loss": 0.104, + "step": 35127 + }, + { + "epoch": 0.3049279086119044, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014851661248526545, + "loss": 0.1123, + "step": 35128 + }, + { + "epoch": 0.3049365890921086, + "grad_norm": 0.6171875, + "learning_rate": 0.0014851392254361243, + "loss": 0.0947, + "step": 35129 + }, + { + "epoch": 0.3049452695723127, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014851123255984115, + "loss": 0.1299, + "step": 35130 + }, + { + "epoch": 0.3049539500525169, + "grad_norm": 0.251953125, + "learning_rate": 0.001485085425339546, + "loss": 0.0801, + "step": 35131 + }, + { + "epoch": 0.30496263053272105, + "grad_norm": 0.64453125, + "learning_rate": 0.001485058524659557, + "loss": 0.0918, + "step": 35132 + }, + { + "epoch": 0.30497131101292524, + "grad_norm": 0.36328125, + "learning_rate": 0.0014850316235584732, + "loss": 0.1084, + "step": 35133 + }, + { + "epoch": 0.3049799914931294, + "grad_norm": 0.30078125, + "learning_rate": 0.0014850047220363249, + "loss": 0.0684, + "step": 35134 + }, + { + "epoch": 0.30498867197333357, + "grad_norm": 0.322265625, + "learning_rate": 0.0014849778200931414, + "loss": 0.0693, + "step": 35135 + }, + { + "epoch": 0.3049973524535377, + "grad_norm": 0.111328125, + "learning_rate": 0.001484950917728952, + "loss": 0.1245, + "step": 35136 + }, + { + "epoch": 0.3050060329337419, + "grad_norm": 0.53125, + "learning_rate": 0.0014849240149437859, + "loss": 0.0791, + "step": 35137 + }, + { + "epoch": 0.30501471341394604, + "grad_norm": 0.1435546875, + "learning_rate": 0.0014848971117376727, + "loss": 0.1084, + "step": 35138 + }, + { + "epoch": 0.30502339389415023, + "grad_norm": 0.7421875, + "learning_rate": 0.0014848702081106417, + "loss": 0.125, + "step": 35139 + }, + { + "epoch": 0.30503207437435437, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014848433040627227, + "loss": 0.0977, + "step": 35140 + }, + { + "epoch": 0.30504075485455856, + "grad_norm": 0.33203125, + "learning_rate": 0.0014848163995939447, + "loss": 0.0967, + "step": 35141 + }, + { + "epoch": 0.3050494353347627, + "grad_norm": 0.306640625, + "learning_rate": 0.0014847894947043372, + "loss": 0.1084, + "step": 35142 + }, + { + "epoch": 0.3050581158149669, + "grad_norm": 0.31640625, + "learning_rate": 0.0014847625893939296, + "loss": 0.0571, + "step": 35143 + }, + { + "epoch": 0.305066796295171, + "grad_norm": 0.32421875, + "learning_rate": 0.0014847356836627512, + "loss": 0.1094, + "step": 35144 + }, + { + "epoch": 0.3050754767753752, + "grad_norm": 0.08642578125, + "learning_rate": 0.0014847087775108324, + "loss": 0.0771, + "step": 35145 + }, + { + "epoch": 0.30508415725557936, + "grad_norm": 0.333984375, + "learning_rate": 0.001484681870938201, + "loss": 0.1152, + "step": 35146 + }, + { + "epoch": 0.30509283773578355, + "grad_norm": 0.51171875, + "learning_rate": 0.0014846549639448879, + "loss": 0.124, + "step": 35147 + }, + { + "epoch": 0.3051015182159877, + "grad_norm": 0.30078125, + "learning_rate": 0.0014846280565309219, + "loss": 0.0708, + "step": 35148 + }, + { + "epoch": 0.3051101986961919, + "grad_norm": 0.3203125, + "learning_rate": 0.001484601148696332, + "loss": 0.0806, + "step": 35149 + }, + { + "epoch": 0.305118879176396, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014845742404411483, + "loss": 0.1206, + "step": 35150 + }, + { + "epoch": 0.3051275596566002, + "grad_norm": 0.5390625, + "learning_rate": 0.0014845473317654002, + "loss": 0.0801, + "step": 35151 + }, + { + "epoch": 0.30513624013680435, + "grad_norm": 0.10400390625, + "learning_rate": 0.001484520422669117, + "loss": 0.0767, + "step": 35152 + }, + { + "epoch": 0.30514492061700854, + "grad_norm": 0.1689453125, + "learning_rate": 0.0014844935131523274, + "loss": 0.1104, + "step": 35153 + }, + { + "epoch": 0.3051536010972127, + "grad_norm": 0.58984375, + "learning_rate": 0.0014844666032150617, + "loss": 0.1104, + "step": 35154 + }, + { + "epoch": 0.30516228157741687, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014844396928573493, + "loss": 0.0986, + "step": 35155 + }, + { + "epoch": 0.305170962057621, + "grad_norm": 0.12353515625, + "learning_rate": 0.0014844127820792196, + "loss": 0.1001, + "step": 35156 + }, + { + "epoch": 0.3051796425378252, + "grad_norm": 0.51953125, + "learning_rate": 0.0014843858708807015, + "loss": 0.105, + "step": 35157 + }, + { + "epoch": 0.30518832301802934, + "grad_norm": 0.203125, + "learning_rate": 0.0014843589592618248, + "loss": 0.0913, + "step": 35158 + }, + { + "epoch": 0.30519700349823353, + "grad_norm": 0.58984375, + "learning_rate": 0.0014843320472226191, + "loss": 0.1602, + "step": 35159 + }, + { + "epoch": 0.30520568397843767, + "grad_norm": 0.275390625, + "learning_rate": 0.0014843051347631136, + "loss": 0.1133, + "step": 35160 + }, + { + "epoch": 0.30521436445864186, + "grad_norm": 0.431640625, + "learning_rate": 0.0014842782218833378, + "loss": 0.085, + "step": 35161 + }, + { + "epoch": 0.305223044938846, + "grad_norm": 0.2490234375, + "learning_rate": 0.001484251308583321, + "loss": 0.0981, + "step": 35162 + }, + { + "epoch": 0.3052317254190502, + "grad_norm": 0.115234375, + "learning_rate": 0.0014842243948630927, + "loss": 0.1074, + "step": 35163 + }, + { + "epoch": 0.30524040589925433, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014841974807226824, + "loss": 0.1006, + "step": 35164 + }, + { + "epoch": 0.3052490863794585, + "grad_norm": 0.458984375, + "learning_rate": 0.0014841705661621197, + "loss": 0.0771, + "step": 35165 + }, + { + "epoch": 0.30525776685966266, + "grad_norm": 0.34375, + "learning_rate": 0.001484143651181434, + "loss": 0.1416, + "step": 35166 + }, + { + "epoch": 0.30526644733986685, + "grad_norm": 0.4140625, + "learning_rate": 0.001484116735780654, + "loss": 0.0635, + "step": 35167 + }, + { + "epoch": 0.305275127820071, + "grad_norm": 0.5234375, + "learning_rate": 0.0014840898199598098, + "loss": 0.1021, + "step": 35168 + }, + { + "epoch": 0.3052838083002752, + "grad_norm": 0.47265625, + "learning_rate": 0.0014840629037189313, + "loss": 0.1064, + "step": 35169 + }, + { + "epoch": 0.3052924887804793, + "grad_norm": 0.162109375, + "learning_rate": 0.0014840359870580471, + "loss": 0.1089, + "step": 35170 + }, + { + "epoch": 0.3053011692606835, + "grad_norm": 0.255859375, + "learning_rate": 0.001484009069977187, + "loss": 0.0938, + "step": 35171 + }, + { + "epoch": 0.30530984974088765, + "grad_norm": 0.115234375, + "learning_rate": 0.0014839821524763798, + "loss": 0.0957, + "step": 35172 + }, + { + "epoch": 0.30531853022109184, + "grad_norm": 0.09375, + "learning_rate": 0.001483955234555656, + "loss": 0.1016, + "step": 35173 + }, + { + "epoch": 0.305327210701296, + "grad_norm": 0.361328125, + "learning_rate": 0.0014839283162150442, + "loss": 0.0723, + "step": 35174 + }, + { + "epoch": 0.3053358911815002, + "grad_norm": 0.359375, + "learning_rate": 0.0014839013974545746, + "loss": 0.0908, + "step": 35175 + }, + { + "epoch": 0.3053445716617043, + "grad_norm": 0.1865234375, + "learning_rate": 0.001483874478274276, + "loss": 0.125, + "step": 35176 + }, + { + "epoch": 0.3053532521419085, + "grad_norm": 0.181640625, + "learning_rate": 0.0014838475586741778, + "loss": 0.0957, + "step": 35177 + }, + { + "epoch": 0.30536193262211264, + "grad_norm": 0.3125, + "learning_rate": 0.00148382063865431, + "loss": 0.1182, + "step": 35178 + }, + { + "epoch": 0.30537061310231683, + "grad_norm": 0.25390625, + "learning_rate": 0.0014837937182147016, + "loss": 0.1172, + "step": 35179 + }, + { + "epoch": 0.30537929358252097, + "grad_norm": 0.326171875, + "learning_rate": 0.0014837667973553823, + "loss": 0.127, + "step": 35180 + }, + { + "epoch": 0.30538797406272516, + "grad_norm": 0.16796875, + "learning_rate": 0.001483739876076381, + "loss": 0.1562, + "step": 35181 + }, + { + "epoch": 0.3053966545429293, + "grad_norm": 0.375, + "learning_rate": 0.001483712954377728, + "loss": 0.1406, + "step": 35182 + }, + { + "epoch": 0.3054053350231335, + "grad_norm": 0.48046875, + "learning_rate": 0.001483686032259452, + "loss": 0.0752, + "step": 35183 + }, + { + "epoch": 0.30541401550333763, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014836591097215825, + "loss": 0.0947, + "step": 35184 + }, + { + "epoch": 0.3054226959835418, + "grad_norm": 0.326171875, + "learning_rate": 0.0014836321867641496, + "loss": 0.168, + "step": 35185 + }, + { + "epoch": 0.30543137646374596, + "grad_norm": 0.1689453125, + "learning_rate": 0.0014836052633871821, + "loss": 0.1045, + "step": 35186 + }, + { + "epoch": 0.30544005694395016, + "grad_norm": 0.59765625, + "learning_rate": 0.0014835783395907099, + "loss": 0.0742, + "step": 35187 + }, + { + "epoch": 0.3054487374241543, + "grad_norm": 0.064453125, + "learning_rate": 0.001483551415374762, + "loss": 0.0742, + "step": 35188 + }, + { + "epoch": 0.3054574179043585, + "grad_norm": 0.248046875, + "learning_rate": 0.0014835244907393681, + "loss": 0.0791, + "step": 35189 + }, + { + "epoch": 0.3054660983845626, + "grad_norm": 0.5234375, + "learning_rate": 0.0014834975656845572, + "loss": 0.0815, + "step": 35190 + }, + { + "epoch": 0.3054747788647668, + "grad_norm": 0.14453125, + "learning_rate": 0.0014834706402103595, + "loss": 0.0903, + "step": 35191 + }, + { + "epoch": 0.30548345934497095, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014834437143168041, + "loss": 0.0786, + "step": 35192 + }, + { + "epoch": 0.30549213982517515, + "grad_norm": 0.58203125, + "learning_rate": 0.0014834167880039204, + "loss": 0.0859, + "step": 35193 + }, + { + "epoch": 0.3055008203053793, + "grad_norm": 0.5234375, + "learning_rate": 0.0014833898612717376, + "loss": 0.0791, + "step": 35194 + }, + { + "epoch": 0.3055095007855835, + "grad_norm": 0.70703125, + "learning_rate": 0.0014833629341202857, + "loss": 0.1172, + "step": 35195 + }, + { + "epoch": 0.3055181812657876, + "grad_norm": 0.4765625, + "learning_rate": 0.0014833360065495938, + "loss": 0.0854, + "step": 35196 + }, + { + "epoch": 0.3055268617459918, + "grad_norm": 0.53125, + "learning_rate": 0.0014833090785596912, + "loss": 0.1182, + "step": 35197 + }, + { + "epoch": 0.30553554222619594, + "grad_norm": 0.6796875, + "learning_rate": 0.0014832821501506077, + "loss": 0.0957, + "step": 35198 + }, + { + "epoch": 0.30554422270640014, + "grad_norm": 0.25390625, + "learning_rate": 0.0014832552213223727, + "loss": 0.085, + "step": 35199 + }, + { + "epoch": 0.3055529031866043, + "grad_norm": 0.10693359375, + "learning_rate": 0.0014832282920750155, + "loss": 0.0918, + "step": 35200 + }, + { + "epoch": 0.30556158366680847, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014832013624085654, + "loss": 0.0854, + "step": 35201 + }, + { + "epoch": 0.3055702641470126, + "grad_norm": 0.115234375, + "learning_rate": 0.0014831744323230522, + "loss": 0.0737, + "step": 35202 + }, + { + "epoch": 0.3055789446272168, + "grad_norm": 0.142578125, + "learning_rate": 0.001483147501818505, + "loss": 0.0801, + "step": 35203 + }, + { + "epoch": 0.30558762510742093, + "grad_norm": 0.1640625, + "learning_rate": 0.0014831205708949538, + "loss": 0.1641, + "step": 35204 + }, + { + "epoch": 0.3055963055876251, + "grad_norm": 0.5625, + "learning_rate": 0.0014830936395524275, + "loss": 0.1426, + "step": 35205 + }, + { + "epoch": 0.30560498606782927, + "grad_norm": 0.28515625, + "learning_rate": 0.0014830667077909562, + "loss": 0.0986, + "step": 35206 + }, + { + "epoch": 0.30561366654803346, + "grad_norm": 0.2578125, + "learning_rate": 0.0014830397756105682, + "loss": 0.0991, + "step": 35207 + }, + { + "epoch": 0.3056223470282376, + "grad_norm": 0.42578125, + "learning_rate": 0.001483012843011294, + "loss": 0.0933, + "step": 35208 + }, + { + "epoch": 0.3056310275084418, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014829859099931627, + "loss": 0.0923, + "step": 35209 + }, + { + "epoch": 0.3056397079886459, + "grad_norm": 0.71875, + "learning_rate": 0.0014829589765562035, + "loss": 0.1387, + "step": 35210 + }, + { + "epoch": 0.3056483884688501, + "grad_norm": 0.1552734375, + "learning_rate": 0.0014829320427004462, + "loss": 0.105, + "step": 35211 + }, + { + "epoch": 0.30565706894905426, + "grad_norm": 0.365234375, + "learning_rate": 0.0014829051084259204, + "loss": 0.1143, + "step": 35212 + }, + { + "epoch": 0.30566574942925845, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014828781737326554, + "loss": 0.0938, + "step": 35213 + }, + { + "epoch": 0.3056744299094626, + "grad_norm": 0.29296875, + "learning_rate": 0.0014828512386206804, + "loss": 0.0918, + "step": 35214 + }, + { + "epoch": 0.3056831103896668, + "grad_norm": 0.1953125, + "learning_rate": 0.001482824303090025, + "loss": 0.1104, + "step": 35215 + }, + { + "epoch": 0.3056917908698709, + "grad_norm": 0.40234375, + "learning_rate": 0.001482797367140719, + "loss": 0.0752, + "step": 35216 + }, + { + "epoch": 0.3057004713500751, + "grad_norm": 0.953125, + "learning_rate": 0.0014827704307727911, + "loss": 0.0986, + "step": 35217 + }, + { + "epoch": 0.30570915183027925, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014827434939862717, + "loss": 0.1035, + "step": 35218 + }, + { + "epoch": 0.30571783231048344, + "grad_norm": 0.1171875, + "learning_rate": 0.0014827165567811893, + "loss": 0.124, + "step": 35219 + }, + { + "epoch": 0.3057265127906876, + "grad_norm": 0.55859375, + "learning_rate": 0.0014826896191575739, + "loss": 0.0952, + "step": 35220 + }, + { + "epoch": 0.30573519327089177, + "grad_norm": 0.4453125, + "learning_rate": 0.0014826626811154549, + "loss": 0.1738, + "step": 35221 + }, + { + "epoch": 0.3057438737510959, + "grad_norm": 0.216796875, + "learning_rate": 0.001482635742654862, + "loss": 0.0986, + "step": 35222 + }, + { + "epoch": 0.3057525542313001, + "grad_norm": 0.1474609375, + "learning_rate": 0.0014826088037758242, + "loss": 0.1221, + "step": 35223 + }, + { + "epoch": 0.30576123471150424, + "grad_norm": 0.224609375, + "learning_rate": 0.0014825818644783708, + "loss": 0.0938, + "step": 35224 + }, + { + "epoch": 0.30576991519170843, + "grad_norm": 0.1484375, + "learning_rate": 0.0014825549247625323, + "loss": 0.1328, + "step": 35225 + }, + { + "epoch": 0.30577859567191257, + "grad_norm": 0.3828125, + "learning_rate": 0.0014825279846283372, + "loss": 0.085, + "step": 35226 + }, + { + "epoch": 0.30578727615211676, + "grad_norm": 0.5, + "learning_rate": 0.0014825010440758153, + "loss": 0.0957, + "step": 35227 + }, + { + "epoch": 0.3057959566323209, + "grad_norm": 0.435546875, + "learning_rate": 0.001482474103104996, + "loss": 0.0962, + "step": 35228 + }, + { + "epoch": 0.3058046371125251, + "grad_norm": 0.07666015625, + "learning_rate": 0.0014824471617159085, + "loss": 0.0713, + "step": 35229 + }, + { + "epoch": 0.30581331759272923, + "grad_norm": 0.09423828125, + "learning_rate": 0.0014824202199085828, + "loss": 0.1123, + "step": 35230 + }, + { + "epoch": 0.3058219980729334, + "grad_norm": 0.416015625, + "learning_rate": 0.001482393277683048, + "loss": 0.1001, + "step": 35231 + }, + { + "epoch": 0.30583067855313756, + "grad_norm": 0.23828125, + "learning_rate": 0.0014823663350393337, + "loss": 0.1235, + "step": 35232 + }, + { + "epoch": 0.30583935903334175, + "grad_norm": 0.1474609375, + "learning_rate": 0.0014823393919774692, + "loss": 0.082, + "step": 35233 + }, + { + "epoch": 0.3058480395135459, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014823124484974841, + "loss": 0.1211, + "step": 35234 + }, + { + "epoch": 0.3058567199937501, + "grad_norm": 0.283203125, + "learning_rate": 0.0014822855045994081, + "loss": 0.085, + "step": 35235 + }, + { + "epoch": 0.3058654004739542, + "grad_norm": 0.240234375, + "learning_rate": 0.0014822585602832705, + "loss": 0.0874, + "step": 35236 + }, + { + "epoch": 0.3058740809541584, + "grad_norm": 0.259765625, + "learning_rate": 0.0014822316155491003, + "loss": 0.1357, + "step": 35237 + }, + { + "epoch": 0.30588276143436255, + "grad_norm": 0.25390625, + "learning_rate": 0.0014822046703969274, + "loss": 0.0728, + "step": 35238 + }, + { + "epoch": 0.3058914419145667, + "grad_norm": 0.69140625, + "learning_rate": 0.0014821777248267812, + "loss": 0.0962, + "step": 35239 + }, + { + "epoch": 0.3059001223947709, + "grad_norm": 0.333984375, + "learning_rate": 0.0014821507788386912, + "loss": 0.1162, + "step": 35240 + }, + { + "epoch": 0.305908802874975, + "grad_norm": 0.06689453125, + "learning_rate": 0.0014821238324326871, + "loss": 0.0801, + "step": 35241 + }, + { + "epoch": 0.3059174833551792, + "grad_norm": 0.263671875, + "learning_rate": 0.001482096885608798, + "loss": 0.083, + "step": 35242 + }, + { + "epoch": 0.30592616383538335, + "grad_norm": 0.333984375, + "learning_rate": 0.0014820699383670535, + "loss": 0.0967, + "step": 35243 + }, + { + "epoch": 0.30593484431558754, + "grad_norm": 0.314453125, + "learning_rate": 0.0014820429907074829, + "loss": 0.1504, + "step": 35244 + }, + { + "epoch": 0.3059435247957917, + "grad_norm": 0.2373046875, + "learning_rate": 0.001482016042630116, + "loss": 0.0938, + "step": 35245 + }, + { + "epoch": 0.30595220527599587, + "grad_norm": 0.3359375, + "learning_rate": 0.0014819890941349819, + "loss": 0.0942, + "step": 35246 + }, + { + "epoch": 0.3059608857562, + "grad_norm": 0.333984375, + "learning_rate": 0.0014819621452221105, + "loss": 0.1064, + "step": 35247 + }, + { + "epoch": 0.3059695662364042, + "grad_norm": 0.5078125, + "learning_rate": 0.001481935195891531, + "loss": 0.1055, + "step": 35248 + }, + { + "epoch": 0.30597824671660834, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014819082461432725, + "loss": 0.0742, + "step": 35249 + }, + { + "epoch": 0.30598692719681253, + "grad_norm": 0.16015625, + "learning_rate": 0.0014818812959773652, + "loss": 0.1055, + "step": 35250 + }, + { + "epoch": 0.30599560767701667, + "grad_norm": 0.9296875, + "learning_rate": 0.0014818543453938384, + "loss": 0.0869, + "step": 35251 + }, + { + "epoch": 0.30600428815722086, + "grad_norm": 0.1328125, + "learning_rate": 0.0014818273943927214, + "loss": 0.1133, + "step": 35252 + }, + { + "epoch": 0.306012968637425, + "grad_norm": 0.1015625, + "learning_rate": 0.0014818004429740434, + "loss": 0.0933, + "step": 35253 + }, + { + "epoch": 0.3060216491176292, + "grad_norm": 0.2431640625, + "learning_rate": 0.0014817734911378342, + "loss": 0.0742, + "step": 35254 + }, + { + "epoch": 0.30603032959783333, + "grad_norm": 0.1484375, + "learning_rate": 0.0014817465388841237, + "loss": 0.1162, + "step": 35255 + }, + { + "epoch": 0.3060390100780375, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014817195862129404, + "loss": 0.1094, + "step": 35256 + }, + { + "epoch": 0.30604769055824166, + "grad_norm": 0.11376953125, + "learning_rate": 0.0014816926331243149, + "loss": 0.0835, + "step": 35257 + }, + { + "epoch": 0.30605637103844585, + "grad_norm": 0.15234375, + "learning_rate": 0.0014816656796182756, + "loss": 0.0913, + "step": 35258 + }, + { + "epoch": 0.30606505151865, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014816387256948523, + "loss": 0.085, + "step": 35259 + }, + { + "epoch": 0.3060737319988542, + "grad_norm": 0.083984375, + "learning_rate": 0.0014816117713540749, + "loss": 0.0835, + "step": 35260 + }, + { + "epoch": 0.3060824124790583, + "grad_norm": 0.328125, + "learning_rate": 0.0014815848165959727, + "loss": 0.1055, + "step": 35261 + }, + { + "epoch": 0.3060910929592625, + "grad_norm": 0.1201171875, + "learning_rate": 0.001481557861420575, + "loss": 0.1001, + "step": 35262 + }, + { + "epoch": 0.30609977343946665, + "grad_norm": 0.1357421875, + "learning_rate": 0.001481530905827911, + "loss": 0.1348, + "step": 35263 + }, + { + "epoch": 0.30610845391967084, + "grad_norm": 0.36328125, + "learning_rate": 0.001481503949818011, + "loss": 0.125, + "step": 35264 + }, + { + "epoch": 0.306117134399875, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014814769933909038, + "loss": 0.0674, + "step": 35265 + }, + { + "epoch": 0.3061258148800792, + "grad_norm": 0.08544921875, + "learning_rate": 0.0014814500365466193, + "loss": 0.0713, + "step": 35266 + }, + { + "epoch": 0.3061344953602833, + "grad_norm": 0.1845703125, + "learning_rate": 0.0014814230792851864, + "loss": 0.0879, + "step": 35267 + }, + { + "epoch": 0.3061431758404875, + "grad_norm": 0.51953125, + "learning_rate": 0.0014813961216066354, + "loss": 0.0781, + "step": 35268 + }, + { + "epoch": 0.30615185632069164, + "grad_norm": 0.125, + "learning_rate": 0.0014813691635109946, + "loss": 0.1118, + "step": 35269 + }, + { + "epoch": 0.30616053680089583, + "grad_norm": 0.10302734375, + "learning_rate": 0.0014813422049982949, + "loss": 0.0977, + "step": 35270 + }, + { + "epoch": 0.30616921728109997, + "grad_norm": 0.0869140625, + "learning_rate": 0.001481315246068565, + "loss": 0.0918, + "step": 35271 + }, + { + "epoch": 0.30617789776130416, + "grad_norm": 0.287109375, + "learning_rate": 0.0014812882867218345, + "loss": 0.0879, + "step": 35272 + }, + { + "epoch": 0.3061865782415083, + "grad_norm": 0.31640625, + "learning_rate": 0.0014812613269581327, + "loss": 0.1152, + "step": 35273 + }, + { + "epoch": 0.3061952587217125, + "grad_norm": 0.154296875, + "learning_rate": 0.0014812343667774891, + "loss": 0.0898, + "step": 35274 + }, + { + "epoch": 0.30620393920191663, + "grad_norm": 0.318359375, + "learning_rate": 0.0014812074061799337, + "loss": 0.0879, + "step": 35275 + }, + { + "epoch": 0.3062126196821208, + "grad_norm": 0.212890625, + "learning_rate": 0.0014811804451654953, + "loss": 0.1328, + "step": 35276 + }, + { + "epoch": 0.30622130016232496, + "grad_norm": 0.443359375, + "learning_rate": 0.0014811534837342038, + "loss": 0.1162, + "step": 35277 + }, + { + "epoch": 0.30622998064252915, + "grad_norm": 0.40234375, + "learning_rate": 0.0014811265218860883, + "loss": 0.0645, + "step": 35278 + }, + { + "epoch": 0.3062386611227333, + "grad_norm": 0.2373046875, + "learning_rate": 0.001481099559621179, + "loss": 0.0967, + "step": 35279 + }, + { + "epoch": 0.3062473416029375, + "grad_norm": 0.220703125, + "learning_rate": 0.0014810725969395047, + "loss": 0.0723, + "step": 35280 + }, + { + "epoch": 0.3062560220831416, + "grad_norm": 0.08935546875, + "learning_rate": 0.0014810456338410953, + "loss": 0.0864, + "step": 35281 + }, + { + "epoch": 0.3062647025633458, + "grad_norm": 0.126953125, + "learning_rate": 0.00148101867032598, + "loss": 0.0981, + "step": 35282 + }, + { + "epoch": 0.30627338304354995, + "grad_norm": 0.37109375, + "learning_rate": 0.0014809917063941887, + "loss": 0.1118, + "step": 35283 + }, + { + "epoch": 0.30628206352375414, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014809647420457501, + "loss": 0.1191, + "step": 35284 + }, + { + "epoch": 0.3062907440039583, + "grad_norm": 0.1005859375, + "learning_rate": 0.0014809377772806948, + "loss": 0.106, + "step": 35285 + }, + { + "epoch": 0.3062994244841625, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014809108120990512, + "loss": 0.0757, + "step": 35286 + }, + { + "epoch": 0.3063081049643666, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014808838465008491, + "loss": 0.0952, + "step": 35287 + }, + { + "epoch": 0.3063167854445708, + "grad_norm": 0.10595703125, + "learning_rate": 0.0014808568804861185, + "loss": 0.123, + "step": 35288 + }, + { + "epoch": 0.30632546592477494, + "grad_norm": 0.1279296875, + "learning_rate": 0.0014808299140548881, + "loss": 0.1182, + "step": 35289 + }, + { + "epoch": 0.30633414640497914, + "grad_norm": 0.0849609375, + "learning_rate": 0.0014808029472071884, + "loss": 0.124, + "step": 35290 + }, + { + "epoch": 0.3063428268851833, + "grad_norm": 0.388671875, + "learning_rate": 0.001480775979943048, + "loss": 0.0815, + "step": 35291 + }, + { + "epoch": 0.30635150736538747, + "grad_norm": 0.4453125, + "learning_rate": 0.001480749012262497, + "loss": 0.1025, + "step": 35292 + }, + { + "epoch": 0.3063601878455916, + "grad_norm": 0.4296875, + "learning_rate": 0.0014807220441655643, + "loss": 0.1045, + "step": 35293 + }, + { + "epoch": 0.3063688683257958, + "grad_norm": 0.765625, + "learning_rate": 0.0014806950756522798, + "loss": 0.1035, + "step": 35294 + }, + { + "epoch": 0.30637754880599993, + "grad_norm": 0.3046875, + "learning_rate": 0.001480668106722673, + "loss": 0.1445, + "step": 35295 + }, + { + "epoch": 0.3063862292862041, + "grad_norm": 0.28515625, + "learning_rate": 0.0014806411373767727, + "loss": 0.1387, + "step": 35296 + }, + { + "epoch": 0.30639490976640826, + "grad_norm": 0.12890625, + "learning_rate": 0.0014806141676146096, + "loss": 0.127, + "step": 35297 + }, + { + "epoch": 0.30640359024661246, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014805871974362123, + "loss": 0.0918, + "step": 35298 + }, + { + "epoch": 0.3064122707268166, + "grad_norm": 0.263671875, + "learning_rate": 0.0014805602268416107, + "loss": 0.0762, + "step": 35299 + }, + { + "epoch": 0.3064209512070208, + "grad_norm": 0.17578125, + "learning_rate": 0.0014805332558308342, + "loss": 0.1191, + "step": 35300 + }, + { + "epoch": 0.3064296316872249, + "grad_norm": 0.38671875, + "learning_rate": 0.001480506284403912, + "loss": 0.0957, + "step": 35301 + }, + { + "epoch": 0.3064383121674291, + "grad_norm": 0.08251953125, + "learning_rate": 0.0014804793125608741, + "loss": 0.0923, + "step": 35302 + }, + { + "epoch": 0.30644699264763325, + "grad_norm": 0.224609375, + "learning_rate": 0.0014804523403017496, + "loss": 0.0996, + "step": 35303 + }, + { + "epoch": 0.30645567312783745, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014804253676265681, + "loss": 0.082, + "step": 35304 + }, + { + "epoch": 0.3064643536080416, + "grad_norm": 0.2890625, + "learning_rate": 0.0014803983945353594, + "loss": 0.0586, + "step": 35305 + }, + { + "epoch": 0.3064730340882458, + "grad_norm": 0.13671875, + "learning_rate": 0.0014803714210281525, + "loss": 0.1602, + "step": 35306 + }, + { + "epoch": 0.3064817145684499, + "grad_norm": 0.09765625, + "learning_rate": 0.001480344447104977, + "loss": 0.127, + "step": 35307 + }, + { + "epoch": 0.3064903950486541, + "grad_norm": 0.0986328125, + "learning_rate": 0.0014803174727658628, + "loss": 0.0996, + "step": 35308 + }, + { + "epoch": 0.30649907552885824, + "grad_norm": 0.447265625, + "learning_rate": 0.0014802904980108392, + "loss": 0.0996, + "step": 35309 + }, + { + "epoch": 0.30650775600906244, + "grad_norm": 1.0390625, + "learning_rate": 0.0014802635228399353, + "loss": 0.1484, + "step": 35310 + }, + { + "epoch": 0.3065164364892666, + "grad_norm": 0.1455078125, + "learning_rate": 0.001480236547253181, + "loss": 0.1104, + "step": 35311 + }, + { + "epoch": 0.30652511696947077, + "grad_norm": 0.380859375, + "learning_rate": 0.0014802095712506058, + "loss": 0.0908, + "step": 35312 + }, + { + "epoch": 0.3065337974496749, + "grad_norm": 0.84765625, + "learning_rate": 0.0014801825948322395, + "loss": 0.0977, + "step": 35313 + }, + { + "epoch": 0.3065424779298791, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014801556179981105, + "loss": 0.0752, + "step": 35314 + }, + { + "epoch": 0.30655115841008324, + "grad_norm": 0.5234375, + "learning_rate": 0.0014801286407482492, + "loss": 0.1074, + "step": 35315 + }, + { + "epoch": 0.30655983889028743, + "grad_norm": 0.310546875, + "learning_rate": 0.001480101663082685, + "loss": 0.1152, + "step": 35316 + }, + { + "epoch": 0.30656851937049157, + "grad_norm": 0.189453125, + "learning_rate": 0.0014800746850014478, + "loss": 0.062, + "step": 35317 + }, + { + "epoch": 0.30657719985069576, + "grad_norm": 0.375, + "learning_rate": 0.0014800477065045664, + "loss": 0.1011, + "step": 35318 + }, + { + "epoch": 0.3065858803308999, + "grad_norm": 0.4609375, + "learning_rate": 0.00148002072759207, + "loss": 0.0693, + "step": 35319 + }, + { + "epoch": 0.3065945608111041, + "grad_norm": 0.81640625, + "learning_rate": 0.0014799937482639892, + "loss": 0.3398, + "step": 35320 + }, + { + "epoch": 0.3066032412913082, + "grad_norm": 0.318359375, + "learning_rate": 0.001479966768520353, + "loss": 0.0835, + "step": 35321 + }, + { + "epoch": 0.3066119217715124, + "grad_norm": 0.216796875, + "learning_rate": 0.0014799397883611908, + "loss": 0.0894, + "step": 35322 + }, + { + "epoch": 0.30662060225171656, + "grad_norm": 0.515625, + "learning_rate": 0.0014799128077865318, + "loss": 0.1143, + "step": 35323 + }, + { + "epoch": 0.30662928273192075, + "grad_norm": 0.435546875, + "learning_rate": 0.0014798858267964059, + "loss": 0.083, + "step": 35324 + }, + { + "epoch": 0.3066379632121249, + "grad_norm": 0.1796875, + "learning_rate": 0.0014798588453908427, + "loss": 0.0947, + "step": 35325 + }, + { + "epoch": 0.3066466436923291, + "grad_norm": 0.482421875, + "learning_rate": 0.0014798318635698715, + "loss": 0.1523, + "step": 35326 + }, + { + "epoch": 0.3066553241725332, + "grad_norm": 0.42578125, + "learning_rate": 0.0014798048813335221, + "loss": 0.1367, + "step": 35327 + }, + { + "epoch": 0.3066640046527374, + "grad_norm": 0.19140625, + "learning_rate": 0.0014797778986818234, + "loss": 0.1055, + "step": 35328 + }, + { + "epoch": 0.30667268513294155, + "grad_norm": 0.27734375, + "learning_rate": 0.001479750915614806, + "loss": 0.0693, + "step": 35329 + }, + { + "epoch": 0.30668136561314574, + "grad_norm": 0.2451171875, + "learning_rate": 0.001479723932132498, + "loss": 0.0791, + "step": 35330 + }, + { + "epoch": 0.3066900460933499, + "grad_norm": 0.384765625, + "learning_rate": 0.00147969694823493, + "loss": 0.1289, + "step": 35331 + }, + { + "epoch": 0.30669872657355407, + "grad_norm": 0.150390625, + "learning_rate": 0.001479669963922131, + "loss": 0.104, + "step": 35332 + }, + { + "epoch": 0.3067074070537582, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014796429791941307, + "loss": 0.1055, + "step": 35333 + }, + { + "epoch": 0.3067160875339624, + "grad_norm": 0.466796875, + "learning_rate": 0.0014796159940509586, + "loss": 0.1152, + "step": 35334 + }, + { + "epoch": 0.30672476801416654, + "grad_norm": 0.373046875, + "learning_rate": 0.0014795890084926438, + "loss": 0.0913, + "step": 35335 + }, + { + "epoch": 0.30673344849437073, + "grad_norm": 0.3515625, + "learning_rate": 0.0014795620225192165, + "loss": 0.0811, + "step": 35336 + }, + { + "epoch": 0.30674212897457487, + "grad_norm": 0.412109375, + "learning_rate": 0.0014795350361307056, + "loss": 0.1152, + "step": 35337 + }, + { + "epoch": 0.30675080945477906, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014795080493271412, + "loss": 0.1138, + "step": 35338 + }, + { + "epoch": 0.3067594899349832, + "grad_norm": 0.08056640625, + "learning_rate": 0.0014794810621085524, + "loss": 0.1328, + "step": 35339 + }, + { + "epoch": 0.3067681704151874, + "grad_norm": 0.29296875, + "learning_rate": 0.001479454074474969, + "loss": 0.1235, + "step": 35340 + }, + { + "epoch": 0.30677685089539153, + "grad_norm": 0.494140625, + "learning_rate": 0.0014794270864264196, + "loss": 0.0854, + "step": 35341 + }, + { + "epoch": 0.3067855313755957, + "grad_norm": 0.0732421875, + "learning_rate": 0.0014794000979629354, + "loss": 0.0791, + "step": 35342 + }, + { + "epoch": 0.30679421185579986, + "grad_norm": 0.14453125, + "learning_rate": 0.0014793731090845443, + "loss": 0.1602, + "step": 35343 + }, + { + "epoch": 0.30680289233600405, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014793461197912766, + "loss": 0.0781, + "step": 35344 + }, + { + "epoch": 0.3068115728162082, + "grad_norm": 0.57421875, + "learning_rate": 0.001479319130083162, + "loss": 0.1475, + "step": 35345 + }, + { + "epoch": 0.3068202532964124, + "grad_norm": 0.6875, + "learning_rate": 0.001479292139960229, + "loss": 0.082, + "step": 35346 + }, + { + "epoch": 0.3068289337766165, + "grad_norm": 0.376953125, + "learning_rate": 0.0014792651494225084, + "loss": 0.1069, + "step": 35347 + }, + { + "epoch": 0.3068376142568207, + "grad_norm": 0.296875, + "learning_rate": 0.001479238158470029, + "loss": 0.1318, + "step": 35348 + }, + { + "epoch": 0.30684629473702485, + "grad_norm": 0.09033203125, + "learning_rate": 0.0014792111671028208, + "loss": 0.0693, + "step": 35349 + }, + { + "epoch": 0.30685497521722904, + "grad_norm": 0.484375, + "learning_rate": 0.0014791841753209122, + "loss": 0.125, + "step": 35350 + }, + { + "epoch": 0.3068636556974332, + "grad_norm": 0.3828125, + "learning_rate": 0.001479157183124334, + "loss": 0.1357, + "step": 35351 + }, + { + "epoch": 0.3068723361776374, + "grad_norm": 0.8359375, + "learning_rate": 0.0014791301905131152, + "loss": 0.0894, + "step": 35352 + }, + { + "epoch": 0.3068810166578415, + "grad_norm": 0.8203125, + "learning_rate": 0.0014791031974872853, + "loss": 0.0894, + "step": 35353 + }, + { + "epoch": 0.3068896971380457, + "grad_norm": 0.5390625, + "learning_rate": 0.0014790762040468738, + "loss": 0.0635, + "step": 35354 + }, + { + "epoch": 0.30689837761824984, + "grad_norm": 0.197265625, + "learning_rate": 0.00147904921019191, + "loss": 0.166, + "step": 35355 + }, + { + "epoch": 0.30690705809845403, + "grad_norm": 1.359375, + "learning_rate": 0.0014790222159224238, + "loss": 0.0771, + "step": 35356 + }, + { + "epoch": 0.30691573857865817, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014789952212384445, + "loss": 0.103, + "step": 35357 + }, + { + "epoch": 0.30692441905886236, + "grad_norm": 0.28515625, + "learning_rate": 0.001478968226140002, + "loss": 0.1016, + "step": 35358 + }, + { + "epoch": 0.3069330995390665, + "grad_norm": 0.1259765625, + "learning_rate": 0.0014789412306271256, + "loss": 0.1011, + "step": 35359 + }, + { + "epoch": 0.3069417800192707, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014789142346998446, + "loss": 0.105, + "step": 35360 + }, + { + "epoch": 0.30695046049947483, + "grad_norm": 0.267578125, + "learning_rate": 0.0014788872383581886, + "loss": 0.0898, + "step": 35361 + }, + { + "epoch": 0.30695914097967897, + "grad_norm": 0.1201171875, + "learning_rate": 0.0014788602416021875, + "loss": 0.0859, + "step": 35362 + }, + { + "epoch": 0.30696782145988316, + "grad_norm": 0.306640625, + "learning_rate": 0.00147883324443187, + "loss": 0.0713, + "step": 35363 + }, + { + "epoch": 0.3069765019400873, + "grad_norm": 0.74609375, + "learning_rate": 0.0014788062468472664, + "loss": 0.1123, + "step": 35364 + }, + { + "epoch": 0.3069851824202915, + "grad_norm": 0.287109375, + "learning_rate": 0.0014787792488484062, + "loss": 0.1055, + "step": 35365 + }, + { + "epoch": 0.30699386290049563, + "grad_norm": 0.25, + "learning_rate": 0.0014787522504353186, + "loss": 0.0962, + "step": 35366 + }, + { + "epoch": 0.3070025433806998, + "grad_norm": 0.5234375, + "learning_rate": 0.0014787252516080332, + "loss": 0.0908, + "step": 35367 + }, + { + "epoch": 0.30701122386090396, + "grad_norm": 0.109375, + "learning_rate": 0.0014786982523665795, + "loss": 0.1104, + "step": 35368 + }, + { + "epoch": 0.30701990434110815, + "grad_norm": 0.5546875, + "learning_rate": 0.0014786712527109873, + "loss": 0.1592, + "step": 35369 + }, + { + "epoch": 0.3070285848213123, + "grad_norm": 0.201171875, + "learning_rate": 0.0014786442526412858, + "loss": 0.1021, + "step": 35370 + }, + { + "epoch": 0.3070372653015165, + "grad_norm": 0.158203125, + "learning_rate": 0.0014786172521575045, + "loss": 0.1035, + "step": 35371 + }, + { + "epoch": 0.3070459457817206, + "grad_norm": 0.333984375, + "learning_rate": 0.0014785902512596731, + "loss": 0.0752, + "step": 35372 + }, + { + "epoch": 0.3070546262619248, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014785632499478211, + "loss": 0.0894, + "step": 35373 + }, + { + "epoch": 0.30706330674212895, + "grad_norm": 0.0947265625, + "learning_rate": 0.001478536248221978, + "loss": 0.0986, + "step": 35374 + }, + { + "epoch": 0.30707198722233314, + "grad_norm": 0.48828125, + "learning_rate": 0.0014785092460821733, + "loss": 0.1113, + "step": 35375 + }, + { + "epoch": 0.3070806677025373, + "grad_norm": 0.306640625, + "learning_rate": 0.0014784822435284367, + "loss": 0.0918, + "step": 35376 + }, + { + "epoch": 0.3070893481827415, + "grad_norm": 0.4140625, + "learning_rate": 0.0014784552405607979, + "loss": 0.0884, + "step": 35377 + }, + { + "epoch": 0.3070980286629456, + "grad_norm": 0.11474609375, + "learning_rate": 0.001478428237179286, + "loss": 0.0928, + "step": 35378 + }, + { + "epoch": 0.3071067091431498, + "grad_norm": 0.11767578125, + "learning_rate": 0.00147840123338393, + "loss": 0.0889, + "step": 35379 + }, + { + "epoch": 0.30711538962335394, + "grad_norm": 0.09814453125, + "learning_rate": 0.0014783742291747607, + "loss": 0.1108, + "step": 35380 + }, + { + "epoch": 0.30712407010355813, + "grad_norm": 0.2197265625, + "learning_rate": 0.0014783472245518071, + "loss": 0.1191, + "step": 35381 + }, + { + "epoch": 0.30713275058376227, + "grad_norm": 0.58203125, + "learning_rate": 0.0014783202195150984, + "loss": 0.1406, + "step": 35382 + }, + { + "epoch": 0.30714143106396646, + "grad_norm": 0.13671875, + "learning_rate": 0.0014782932140646643, + "loss": 0.1045, + "step": 35383 + }, + { + "epoch": 0.3071501115441706, + "grad_norm": 0.45703125, + "learning_rate": 0.0014782662082005345, + "loss": 0.0977, + "step": 35384 + }, + { + "epoch": 0.3071587920243748, + "grad_norm": 0.353515625, + "learning_rate": 0.0014782392019227384, + "loss": 0.0933, + "step": 35385 + }, + { + "epoch": 0.30716747250457893, + "grad_norm": 0.17578125, + "learning_rate": 0.001478212195231306, + "loss": 0.1562, + "step": 35386 + }, + { + "epoch": 0.3071761529847831, + "grad_norm": 0.498046875, + "learning_rate": 0.0014781851881262662, + "loss": 0.0869, + "step": 35387 + }, + { + "epoch": 0.30718483346498726, + "grad_norm": 0.09765625, + "learning_rate": 0.0014781581806076487, + "loss": 0.0737, + "step": 35388 + }, + { + "epoch": 0.30719351394519145, + "grad_norm": 0.2236328125, + "learning_rate": 0.0014781311726754832, + "loss": 0.1201, + "step": 35389 + }, + { + "epoch": 0.3072021944253956, + "grad_norm": 0.34375, + "learning_rate": 0.0014781041643297987, + "loss": 0.0879, + "step": 35390 + }, + { + "epoch": 0.3072108749055998, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014780771555706258, + "loss": 0.0864, + "step": 35391 + }, + { + "epoch": 0.3072195553858039, + "grad_norm": 0.65625, + "learning_rate": 0.001478050146397993, + "loss": 0.0791, + "step": 35392 + }, + { + "epoch": 0.3072282358660081, + "grad_norm": 0.81640625, + "learning_rate": 0.0014780231368119302, + "loss": 0.1064, + "step": 35393 + }, + { + "epoch": 0.30723691634621225, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014779961268124673, + "loss": 0.1152, + "step": 35394 + }, + { + "epoch": 0.30724559682641644, + "grad_norm": 0.263671875, + "learning_rate": 0.0014779691163996334, + "loss": 0.1475, + "step": 35395 + }, + { + "epoch": 0.3072542773066206, + "grad_norm": 0.443359375, + "learning_rate": 0.0014779421055734578, + "loss": 0.0737, + "step": 35396 + }, + { + "epoch": 0.3072629577868248, + "grad_norm": 0.58203125, + "learning_rate": 0.0014779150943339708, + "loss": 0.084, + "step": 35397 + }, + { + "epoch": 0.3072716382670289, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014778880826812016, + "loss": 0.0869, + "step": 35398 + }, + { + "epoch": 0.3072803187472331, + "grad_norm": 0.98828125, + "learning_rate": 0.0014778610706151795, + "loss": 0.1172, + "step": 35399 + }, + { + "epoch": 0.30728899922743724, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014778340581359341, + "loss": 0.0986, + "step": 35400 + }, + { + "epoch": 0.30729767970764144, + "grad_norm": 0.259765625, + "learning_rate": 0.0014778070452434954, + "loss": 0.1035, + "step": 35401 + }, + { + "epoch": 0.3073063601878456, + "grad_norm": 0.3828125, + "learning_rate": 0.001477780031937892, + "loss": 0.1006, + "step": 35402 + }, + { + "epoch": 0.30731504066804977, + "grad_norm": 0.458984375, + "learning_rate": 0.0014777530182191545, + "loss": 0.1055, + "step": 35403 + }, + { + "epoch": 0.3073237211482539, + "grad_norm": 0.169921875, + "learning_rate": 0.001477726004087312, + "loss": 0.0933, + "step": 35404 + }, + { + "epoch": 0.3073324016284581, + "grad_norm": 0.3828125, + "learning_rate": 0.0014776989895423937, + "loss": 0.0962, + "step": 35405 + }, + { + "epoch": 0.30734108210866223, + "grad_norm": 0.6015625, + "learning_rate": 0.0014776719745844295, + "loss": 0.1338, + "step": 35406 + }, + { + "epoch": 0.3073497625888664, + "grad_norm": 0.080078125, + "learning_rate": 0.001477644959213449, + "loss": 0.0835, + "step": 35407 + }, + { + "epoch": 0.30735844306907056, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014776179434294822, + "loss": 0.1025, + "step": 35408 + }, + { + "epoch": 0.30736712354927476, + "grad_norm": 0.185546875, + "learning_rate": 0.0014775909272325574, + "loss": 0.0898, + "step": 35409 + }, + { + "epoch": 0.3073758040294789, + "grad_norm": 0.2294921875, + "learning_rate": 0.0014775639106227047, + "loss": 0.1377, + "step": 35410 + }, + { + "epoch": 0.3073844845096831, + "grad_norm": 0.69921875, + "learning_rate": 0.0014775368935999542, + "loss": 0.1338, + "step": 35411 + }, + { + "epoch": 0.3073931649898872, + "grad_norm": 0.1943359375, + "learning_rate": 0.0014775098761643348, + "loss": 0.0957, + "step": 35412 + }, + { + "epoch": 0.3074018454700914, + "grad_norm": 0.1806640625, + "learning_rate": 0.0014774828583158765, + "loss": 0.1143, + "step": 35413 + }, + { + "epoch": 0.30741052595029555, + "grad_norm": 0.4453125, + "learning_rate": 0.0014774558400546083, + "loss": 0.1025, + "step": 35414 + }, + { + "epoch": 0.30741920643049975, + "grad_norm": 0.224609375, + "learning_rate": 0.0014774288213805604, + "loss": 0.0752, + "step": 35415 + }, + { + "epoch": 0.3074278869107039, + "grad_norm": 0.08544921875, + "learning_rate": 0.001477401802293762, + "loss": 0.0977, + "step": 35416 + }, + { + "epoch": 0.3074365673909081, + "grad_norm": 1.171875, + "learning_rate": 0.0014773747827942425, + "loss": 0.1318, + "step": 35417 + }, + { + "epoch": 0.3074452478711122, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014773477628820316, + "loss": 0.1001, + "step": 35418 + }, + { + "epoch": 0.3074539283513164, + "grad_norm": 0.42578125, + "learning_rate": 0.0014773207425571588, + "loss": 0.1113, + "step": 35419 + }, + { + "epoch": 0.30746260883152055, + "grad_norm": 0.158203125, + "learning_rate": 0.0014772937218196541, + "loss": 0.0806, + "step": 35420 + }, + { + "epoch": 0.30747128931172474, + "grad_norm": 0.380859375, + "learning_rate": 0.0014772667006695461, + "loss": 0.0771, + "step": 35421 + }, + { + "epoch": 0.3074799697919289, + "grad_norm": 0.2099609375, + "learning_rate": 0.0014772396791068653, + "loss": 0.1045, + "step": 35422 + }, + { + "epoch": 0.30748865027213307, + "grad_norm": 0.09228515625, + "learning_rate": 0.0014772126571316404, + "loss": 0.0957, + "step": 35423 + }, + { + "epoch": 0.3074973307523372, + "grad_norm": 0.1455078125, + "learning_rate": 0.0014771856347439022, + "loss": 0.1094, + "step": 35424 + }, + { + "epoch": 0.3075060112325414, + "grad_norm": 0.15234375, + "learning_rate": 0.0014771586119436788, + "loss": 0.0859, + "step": 35425 + }, + { + "epoch": 0.30751469171274554, + "grad_norm": 0.1640625, + "learning_rate": 0.0014771315887310006, + "loss": 0.2734, + "step": 35426 + }, + { + "epoch": 0.30752337219294973, + "grad_norm": 0.279296875, + "learning_rate": 0.0014771045651058969, + "loss": 0.0791, + "step": 35427 + }, + { + "epoch": 0.30753205267315387, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014770775410683974, + "loss": 0.0933, + "step": 35428 + }, + { + "epoch": 0.30754073315335806, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014770505166185317, + "loss": 0.0967, + "step": 35429 + }, + { + "epoch": 0.3075494136335622, + "grad_norm": 0.25, + "learning_rate": 0.001477023491756329, + "loss": 0.0796, + "step": 35430 + }, + { + "epoch": 0.3075580941137664, + "grad_norm": 1.8671875, + "learning_rate": 0.001476996466481819, + "loss": 0.1309, + "step": 35431 + }, + { + "epoch": 0.3075667745939705, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014769694407950314, + "loss": 0.1357, + "step": 35432 + }, + { + "epoch": 0.3075754550741747, + "grad_norm": 0.28125, + "learning_rate": 0.0014769424146959958, + "loss": 0.1201, + "step": 35433 + }, + { + "epoch": 0.30758413555437886, + "grad_norm": 0.10205078125, + "learning_rate": 0.0014769153881847416, + "loss": 0.1025, + "step": 35434 + }, + { + "epoch": 0.30759281603458305, + "grad_norm": 0.291015625, + "learning_rate": 0.0014768883612612984, + "loss": 0.1016, + "step": 35435 + }, + { + "epoch": 0.3076014965147872, + "grad_norm": 0.11767578125, + "learning_rate": 0.0014768613339256956, + "loss": 0.0986, + "step": 35436 + }, + { + "epoch": 0.3076101769949914, + "grad_norm": 0.189453125, + "learning_rate": 0.001476834306177963, + "loss": 0.1152, + "step": 35437 + }, + { + "epoch": 0.3076188574751955, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014768072780181303, + "loss": 0.0864, + "step": 35438 + }, + { + "epoch": 0.3076275379553997, + "grad_norm": 0.197265625, + "learning_rate": 0.0014767802494462265, + "loss": 0.0962, + "step": 35439 + }, + { + "epoch": 0.30763621843560385, + "grad_norm": 0.60546875, + "learning_rate": 0.0014767532204622816, + "loss": 0.1045, + "step": 35440 + }, + { + "epoch": 0.30764489891580804, + "grad_norm": 0.2314453125, + "learning_rate": 0.001476726191066325, + "loss": 0.0913, + "step": 35441 + }, + { + "epoch": 0.3076535793960122, + "grad_norm": 0.322265625, + "learning_rate": 0.0014766991612583861, + "loss": 0.1016, + "step": 35442 + }, + { + "epoch": 0.30766225987621637, + "grad_norm": 0.4921875, + "learning_rate": 0.0014766721310384952, + "loss": 0.1279, + "step": 35443 + }, + { + "epoch": 0.3076709403564205, + "grad_norm": 0.1689453125, + "learning_rate": 0.001476645100406681, + "loss": 0.1006, + "step": 35444 + }, + { + "epoch": 0.3076796208366247, + "grad_norm": 0.515625, + "learning_rate": 0.0014766180693629731, + "loss": 0.123, + "step": 35445 + }, + { + "epoch": 0.30768830131682884, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014765910379074018, + "loss": 0.1406, + "step": 35446 + }, + { + "epoch": 0.30769698179703303, + "grad_norm": 0.6484375, + "learning_rate": 0.0014765640060399963, + "loss": 0.1338, + "step": 35447 + }, + { + "epoch": 0.30770566227723717, + "grad_norm": 0.12158203125, + "learning_rate": 0.0014765369737607856, + "loss": 0.1094, + "step": 35448 + }, + { + "epoch": 0.30771434275744136, + "grad_norm": 0.8515625, + "learning_rate": 0.0014765099410698, + "loss": 0.0859, + "step": 35449 + }, + { + "epoch": 0.3077230232376455, + "grad_norm": 0.24609375, + "learning_rate": 0.0014764829079670688, + "loss": 0.1152, + "step": 35450 + }, + { + "epoch": 0.3077317037178497, + "grad_norm": 0.07080078125, + "learning_rate": 0.001476455874452621, + "loss": 0.0664, + "step": 35451 + }, + { + "epoch": 0.30774038419805383, + "grad_norm": 0.765625, + "learning_rate": 0.0014764288405264873, + "loss": 0.0913, + "step": 35452 + }, + { + "epoch": 0.307749064678258, + "grad_norm": 0.43359375, + "learning_rate": 0.0014764018061886965, + "loss": 0.1201, + "step": 35453 + }, + { + "epoch": 0.30775774515846216, + "grad_norm": 0.91796875, + "learning_rate": 0.0014763747714392783, + "loss": 0.1211, + "step": 35454 + }, + { + "epoch": 0.30776642563866635, + "grad_norm": 0.0859375, + "learning_rate": 0.0014763477362782626, + "loss": 0.0947, + "step": 35455 + }, + { + "epoch": 0.3077751061188705, + "grad_norm": 0.26953125, + "learning_rate": 0.0014763207007056785, + "loss": 0.0874, + "step": 35456 + }, + { + "epoch": 0.3077837865990747, + "grad_norm": 0.185546875, + "learning_rate": 0.0014762936647215556, + "loss": 0.1387, + "step": 35457 + }, + { + "epoch": 0.3077924670792788, + "grad_norm": 0.12890625, + "learning_rate": 0.0014762666283259236, + "loss": 0.0903, + "step": 35458 + }, + { + "epoch": 0.307801147559483, + "grad_norm": 0.27734375, + "learning_rate": 0.0014762395915188123, + "loss": 0.0977, + "step": 35459 + }, + { + "epoch": 0.30780982803968715, + "grad_norm": 0.1025390625, + "learning_rate": 0.0014762125543002506, + "loss": 0.0898, + "step": 35460 + }, + { + "epoch": 0.30781850851989134, + "grad_norm": 0.703125, + "learning_rate": 0.0014761855166702692, + "loss": 0.085, + "step": 35461 + }, + { + "epoch": 0.3078271890000955, + "grad_norm": 0.388671875, + "learning_rate": 0.0014761584786288963, + "loss": 0.0635, + "step": 35462 + }, + { + "epoch": 0.3078358694802997, + "grad_norm": 0.1171875, + "learning_rate": 0.0014761314401761625, + "loss": 0.0938, + "step": 35463 + }, + { + "epoch": 0.3078445499605038, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014761044013120968, + "loss": 0.1289, + "step": 35464 + }, + { + "epoch": 0.307853230440708, + "grad_norm": 0.2353515625, + "learning_rate": 0.001476077362036729, + "loss": 0.0732, + "step": 35465 + }, + { + "epoch": 0.30786191092091214, + "grad_norm": 0.11767578125, + "learning_rate": 0.0014760503223500887, + "loss": 0.0898, + "step": 35466 + }, + { + "epoch": 0.30787059140111633, + "grad_norm": 0.208984375, + "learning_rate": 0.0014760232822522055, + "loss": 0.0781, + "step": 35467 + }, + { + "epoch": 0.30787927188132047, + "grad_norm": 0.80859375, + "learning_rate": 0.001475996241743109, + "loss": 0.1064, + "step": 35468 + }, + { + "epoch": 0.30788795236152466, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014759692008228286, + "loss": 0.085, + "step": 35469 + }, + { + "epoch": 0.3078966328417288, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014759421594913937, + "loss": 0.0977, + "step": 35470 + }, + { + "epoch": 0.307905313321933, + "grad_norm": 0.1611328125, + "learning_rate": 0.001475915117748834, + "loss": 0.1104, + "step": 35471 + }, + { + "epoch": 0.30791399380213713, + "grad_norm": 0.5234375, + "learning_rate": 0.0014758880755951793, + "loss": 0.105, + "step": 35472 + }, + { + "epoch": 0.3079226742823413, + "grad_norm": 0.2314453125, + "learning_rate": 0.0014758610330304596, + "loss": 0.082, + "step": 35473 + }, + { + "epoch": 0.30793135476254546, + "grad_norm": 0.38671875, + "learning_rate": 0.0014758339900547033, + "loss": 0.0952, + "step": 35474 + }, + { + "epoch": 0.30794003524274965, + "grad_norm": 0.47265625, + "learning_rate": 0.001475806946667941, + "loss": 0.1504, + "step": 35475 + }, + { + "epoch": 0.3079487157229538, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014757799028702015, + "loss": 0.1221, + "step": 35476 + }, + { + "epoch": 0.307957396203158, + "grad_norm": 0.1025390625, + "learning_rate": 0.001475752858661515, + "loss": 0.1021, + "step": 35477 + }, + { + "epoch": 0.3079660766833621, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014757258140419107, + "loss": 0.1162, + "step": 35478 + }, + { + "epoch": 0.3079747571635663, + "grad_norm": 0.333984375, + "learning_rate": 0.0014756987690114182, + "loss": 0.0967, + "step": 35479 + }, + { + "epoch": 0.30798343764377045, + "grad_norm": 0.5078125, + "learning_rate": 0.0014756717235700672, + "loss": 0.0981, + "step": 35480 + }, + { + "epoch": 0.30799211812397465, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014756446777178874, + "loss": 0.1143, + "step": 35481 + }, + { + "epoch": 0.3080007986041788, + "grad_norm": 0.08642578125, + "learning_rate": 0.0014756176314549081, + "loss": 0.1211, + "step": 35482 + }, + { + "epoch": 0.308009479084383, + "grad_norm": 0.142578125, + "learning_rate": 0.001475590584781159, + "loss": 0.084, + "step": 35483 + }, + { + "epoch": 0.3080181595645871, + "grad_norm": 0.267578125, + "learning_rate": 0.0014755635376966696, + "loss": 0.0986, + "step": 35484 + }, + { + "epoch": 0.30802684004479125, + "grad_norm": 0.796875, + "learning_rate": 0.0014755364902014699, + "loss": 0.1138, + "step": 35485 + }, + { + "epoch": 0.30803552052499544, + "grad_norm": 0.58984375, + "learning_rate": 0.001475509442295589, + "loss": 0.124, + "step": 35486 + }, + { + "epoch": 0.3080442010051996, + "grad_norm": 0.140625, + "learning_rate": 0.0014754823939790564, + "loss": 0.0986, + "step": 35487 + }, + { + "epoch": 0.3080528814854038, + "grad_norm": 0.189453125, + "learning_rate": 0.001475455345251902, + "loss": 0.0723, + "step": 35488 + }, + { + "epoch": 0.3080615619656079, + "grad_norm": 0.1513671875, + "learning_rate": 0.001475428296114155, + "loss": 0.0938, + "step": 35489 + }, + { + "epoch": 0.3080702424458121, + "grad_norm": 0.31640625, + "learning_rate": 0.001475401246565846, + "loss": 0.1035, + "step": 35490 + }, + { + "epoch": 0.30807892292601624, + "grad_norm": 0.26171875, + "learning_rate": 0.001475374196607003, + "loss": 0.1133, + "step": 35491 + }, + { + "epoch": 0.30808760340622043, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014753471462376569, + "loss": 0.0688, + "step": 35492 + }, + { + "epoch": 0.30809628388642457, + "grad_norm": 0.33984375, + "learning_rate": 0.0014753200954578367, + "loss": 0.1426, + "step": 35493 + }, + { + "epoch": 0.30810496436662876, + "grad_norm": 0.4609375, + "learning_rate": 0.0014752930442675722, + "loss": 0.0908, + "step": 35494 + }, + { + "epoch": 0.3081136448468329, + "grad_norm": 0.486328125, + "learning_rate": 0.001475265992666893, + "loss": 0.0923, + "step": 35495 + }, + { + "epoch": 0.3081223253270371, + "grad_norm": 0.37890625, + "learning_rate": 0.001475238940655828, + "loss": 0.1025, + "step": 35496 + }, + { + "epoch": 0.30813100580724123, + "grad_norm": 0.369140625, + "learning_rate": 0.0014752118882344078, + "loss": 0.1021, + "step": 35497 + }, + { + "epoch": 0.3081396862874454, + "grad_norm": 0.09375, + "learning_rate": 0.0014751848354026614, + "loss": 0.0776, + "step": 35498 + }, + { + "epoch": 0.30814836676764956, + "grad_norm": 0.1650390625, + "learning_rate": 0.0014751577821606183, + "loss": 0.1924, + "step": 35499 + }, + { + "epoch": 0.30815704724785375, + "grad_norm": 0.44140625, + "learning_rate": 0.0014751307285083086, + "loss": 0.0771, + "step": 35500 + }, + { + "epoch": 0.3081657277280579, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014751036744457614, + "loss": 0.1357, + "step": 35501 + }, + { + "epoch": 0.3081744082082621, + "grad_norm": 0.314453125, + "learning_rate": 0.0014750766199730066, + "loss": 0.0684, + "step": 35502 + }, + { + "epoch": 0.3081830886884662, + "grad_norm": 0.388671875, + "learning_rate": 0.0014750495650900732, + "loss": 0.1133, + "step": 35503 + }, + { + "epoch": 0.3081917691686704, + "grad_norm": 0.23828125, + "learning_rate": 0.0014750225097969918, + "loss": 0.1309, + "step": 35504 + }, + { + "epoch": 0.30820044964887455, + "grad_norm": 0.53515625, + "learning_rate": 0.0014749954540937914, + "loss": 0.0825, + "step": 35505 + }, + { + "epoch": 0.30820913012907875, + "grad_norm": 0.208984375, + "learning_rate": 0.0014749683979805013, + "loss": 0.0806, + "step": 35506 + }, + { + "epoch": 0.3082178106092829, + "grad_norm": 0.10400390625, + "learning_rate": 0.0014749413414571516, + "loss": 0.0879, + "step": 35507 + }, + { + "epoch": 0.3082264910894871, + "grad_norm": 0.08154296875, + "learning_rate": 0.0014749142845237712, + "loss": 0.0942, + "step": 35508 + }, + { + "epoch": 0.3082351715696912, + "grad_norm": 0.1142578125, + "learning_rate": 0.0014748872271803904, + "loss": 0.1309, + "step": 35509 + }, + { + "epoch": 0.3082438520498954, + "grad_norm": 0.6875, + "learning_rate": 0.0014748601694270387, + "loss": 0.1006, + "step": 35510 + }, + { + "epoch": 0.30825253253009954, + "grad_norm": 0.1865234375, + "learning_rate": 0.0014748331112637457, + "loss": 0.1094, + "step": 35511 + }, + { + "epoch": 0.30826121301030374, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014748060526905407, + "loss": 0.1426, + "step": 35512 + }, + { + "epoch": 0.3082698934905079, + "grad_norm": 0.22265625, + "learning_rate": 0.0014747789937074535, + "loss": 0.0752, + "step": 35513 + }, + { + "epoch": 0.30827857397071207, + "grad_norm": 0.609375, + "learning_rate": 0.0014747519343145134, + "loss": 0.1113, + "step": 35514 + }, + { + "epoch": 0.3082872544509162, + "grad_norm": 0.185546875, + "learning_rate": 0.0014747248745117505, + "loss": 0.1055, + "step": 35515 + }, + { + "epoch": 0.3082959349311204, + "grad_norm": 0.56640625, + "learning_rate": 0.0014746978142991941, + "loss": 0.1172, + "step": 35516 + }, + { + "epoch": 0.30830461541132453, + "grad_norm": 1.078125, + "learning_rate": 0.0014746707536768735, + "loss": 0.1768, + "step": 35517 + }, + { + "epoch": 0.3083132958915287, + "grad_norm": 0.494140625, + "learning_rate": 0.0014746436926448185, + "loss": 0.0786, + "step": 35518 + }, + { + "epoch": 0.30832197637173286, + "grad_norm": 0.263671875, + "learning_rate": 0.0014746166312030592, + "loss": 0.1074, + "step": 35519 + }, + { + "epoch": 0.30833065685193706, + "grad_norm": 0.609375, + "learning_rate": 0.0014745895693516248, + "loss": 0.1104, + "step": 35520 + }, + { + "epoch": 0.3083393373321412, + "grad_norm": 0.76953125, + "learning_rate": 0.0014745625070905447, + "loss": 0.1152, + "step": 35521 + }, + { + "epoch": 0.3083480178123454, + "grad_norm": 0.146484375, + "learning_rate": 0.0014745354444198485, + "loss": 0.1455, + "step": 35522 + }, + { + "epoch": 0.3083566982925495, + "grad_norm": 0.1435546875, + "learning_rate": 0.001474508381339566, + "loss": 0.1128, + "step": 35523 + }, + { + "epoch": 0.3083653787727537, + "grad_norm": 0.439453125, + "learning_rate": 0.001474481317849727, + "loss": 0.0752, + "step": 35524 + }, + { + "epoch": 0.30837405925295786, + "grad_norm": 0.2041015625, + "learning_rate": 0.001474454253950361, + "loss": 0.1162, + "step": 35525 + }, + { + "epoch": 0.30838273973316205, + "grad_norm": 0.267578125, + "learning_rate": 0.0014744271896414971, + "loss": 0.0928, + "step": 35526 + }, + { + "epoch": 0.3083914202133662, + "grad_norm": 1.6328125, + "learning_rate": 0.0014744001249231652, + "loss": 0.1484, + "step": 35527 + }, + { + "epoch": 0.3084001006935704, + "grad_norm": 0.1337890625, + "learning_rate": 0.001474373059795395, + "loss": 0.0791, + "step": 35528 + }, + { + "epoch": 0.3084087811737745, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014743459942582163, + "loss": 0.1182, + "step": 35529 + }, + { + "epoch": 0.3084174616539787, + "grad_norm": 0.47265625, + "learning_rate": 0.0014743189283116583, + "loss": 0.1299, + "step": 35530 + }, + { + "epoch": 0.30842614213418285, + "grad_norm": 0.240234375, + "learning_rate": 0.0014742918619557507, + "loss": 0.1001, + "step": 35531 + }, + { + "epoch": 0.30843482261438704, + "grad_norm": 0.298828125, + "learning_rate": 0.001474264795190523, + "loss": 0.0977, + "step": 35532 + }, + { + "epoch": 0.3084435030945912, + "grad_norm": 0.5859375, + "learning_rate": 0.0014742377280160053, + "loss": 0.1543, + "step": 35533 + }, + { + "epoch": 0.30845218357479537, + "grad_norm": 0.271484375, + "learning_rate": 0.0014742106604322268, + "loss": 0.125, + "step": 35534 + }, + { + "epoch": 0.3084608640549995, + "grad_norm": 0.51953125, + "learning_rate": 0.0014741835924392168, + "loss": 0.1133, + "step": 35535 + }, + { + "epoch": 0.3084695445352037, + "grad_norm": 0.142578125, + "learning_rate": 0.0014741565240370055, + "loss": 0.0767, + "step": 35536 + }, + { + "epoch": 0.30847822501540784, + "grad_norm": 0.361328125, + "learning_rate": 0.0014741294552256222, + "loss": 0.105, + "step": 35537 + }, + { + "epoch": 0.30848690549561203, + "grad_norm": 0.125, + "learning_rate": 0.0014741023860050965, + "loss": 0.0903, + "step": 35538 + }, + { + "epoch": 0.30849558597581617, + "grad_norm": 0.0673828125, + "learning_rate": 0.0014740753163754584, + "loss": 0.0967, + "step": 35539 + }, + { + "epoch": 0.30850426645602036, + "grad_norm": 0.48828125, + "learning_rate": 0.0014740482463367365, + "loss": 0.083, + "step": 35540 + }, + { + "epoch": 0.3085129469362245, + "grad_norm": 0.27734375, + "learning_rate": 0.0014740211758889617, + "loss": 0.124, + "step": 35541 + }, + { + "epoch": 0.3085216274164287, + "grad_norm": 0.099609375, + "learning_rate": 0.0014739941050321626, + "loss": 0.1182, + "step": 35542 + }, + { + "epoch": 0.3085303078966328, + "grad_norm": 0.451171875, + "learning_rate": 0.0014739670337663693, + "loss": 0.1011, + "step": 35543 + }, + { + "epoch": 0.308538988376837, + "grad_norm": 0.51953125, + "learning_rate": 0.001473939962091611, + "loss": 0.0938, + "step": 35544 + }, + { + "epoch": 0.30854766885704116, + "grad_norm": 0.177734375, + "learning_rate": 0.001473912890007918, + "loss": 0.1152, + "step": 35545 + }, + { + "epoch": 0.30855634933724535, + "grad_norm": 0.07080078125, + "learning_rate": 0.0014738858175153192, + "loss": 0.0786, + "step": 35546 + }, + { + "epoch": 0.3085650298174495, + "grad_norm": 0.77734375, + "learning_rate": 0.0014738587446138445, + "loss": 0.0889, + "step": 35547 + }, + { + "epoch": 0.3085737102976537, + "grad_norm": 0.703125, + "learning_rate": 0.0014738316713035233, + "loss": 0.1494, + "step": 35548 + }, + { + "epoch": 0.3085823907778578, + "grad_norm": 0.388671875, + "learning_rate": 0.0014738045975843858, + "loss": 0.0771, + "step": 35549 + }, + { + "epoch": 0.308591071258062, + "grad_norm": 0.255859375, + "learning_rate": 0.0014737775234564612, + "loss": 0.0859, + "step": 35550 + }, + { + "epoch": 0.30859975173826615, + "grad_norm": 0.4375, + "learning_rate": 0.001473750448919779, + "loss": 0.1113, + "step": 35551 + }, + { + "epoch": 0.30860843221847034, + "grad_norm": 0.087890625, + "learning_rate": 0.001473723373974369, + "loss": 0.1118, + "step": 35552 + }, + { + "epoch": 0.3086171126986745, + "grad_norm": 0.130859375, + "learning_rate": 0.0014736962986202605, + "loss": 0.0771, + "step": 35553 + }, + { + "epoch": 0.30862579317887867, + "grad_norm": 0.203125, + "learning_rate": 0.0014736692228574837, + "loss": 0.0815, + "step": 35554 + }, + { + "epoch": 0.3086344736590828, + "grad_norm": 0.341796875, + "learning_rate": 0.0014736421466860674, + "loss": 0.1738, + "step": 35555 + }, + { + "epoch": 0.308643154139287, + "grad_norm": 0.09619140625, + "learning_rate": 0.001473615070106042, + "loss": 0.1406, + "step": 35556 + }, + { + "epoch": 0.30865183461949114, + "grad_norm": 0.1552734375, + "learning_rate": 0.0014735879931174365, + "loss": 0.0942, + "step": 35557 + }, + { + "epoch": 0.30866051509969533, + "grad_norm": 0.390625, + "learning_rate": 0.001473560915720281, + "loss": 0.1045, + "step": 35558 + }, + { + "epoch": 0.30866919557989947, + "grad_norm": 0.357421875, + "learning_rate": 0.001473533837914605, + "loss": 0.0972, + "step": 35559 + }, + { + "epoch": 0.30867787606010366, + "grad_norm": 0.078125, + "learning_rate": 0.0014735067597004378, + "loss": 0.0654, + "step": 35560 + }, + { + "epoch": 0.3086865565403078, + "grad_norm": 0.171875, + "learning_rate": 0.0014734796810778094, + "loss": 0.1465, + "step": 35561 + }, + { + "epoch": 0.308695237020512, + "grad_norm": 0.06396484375, + "learning_rate": 0.001473452602046749, + "loss": 0.0752, + "step": 35562 + }, + { + "epoch": 0.30870391750071613, + "grad_norm": 0.119140625, + "learning_rate": 0.0014734255226072867, + "loss": 0.1367, + "step": 35563 + }, + { + "epoch": 0.3087125979809203, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014733984427594518, + "loss": 0.1025, + "step": 35564 + }, + { + "epoch": 0.30872127846112446, + "grad_norm": 0.55078125, + "learning_rate": 0.0014733713625032737, + "loss": 0.1001, + "step": 35565 + }, + { + "epoch": 0.30872995894132865, + "grad_norm": 0.24609375, + "learning_rate": 0.001473344281838782, + "loss": 0.1177, + "step": 35566 + }, + { + "epoch": 0.3087386394215328, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014733172007660074, + "loss": 0.1128, + "step": 35567 + }, + { + "epoch": 0.308747319901737, + "grad_norm": 0.55859375, + "learning_rate": 0.0014732901192849783, + "loss": 0.0854, + "step": 35568 + }, + { + "epoch": 0.3087560003819411, + "grad_norm": 0.09619140625, + "learning_rate": 0.001473263037395725, + "loss": 0.1064, + "step": 35569 + }, + { + "epoch": 0.3087646808621453, + "grad_norm": 0.416015625, + "learning_rate": 0.0014732359550982765, + "loss": 0.0967, + "step": 35570 + }, + { + "epoch": 0.30877336134234945, + "grad_norm": 0.51953125, + "learning_rate": 0.001473208872392663, + "loss": 0.0913, + "step": 35571 + }, + { + "epoch": 0.30878204182255364, + "grad_norm": 0.16796875, + "learning_rate": 0.001473181789278914, + "loss": 0.0728, + "step": 35572 + }, + { + "epoch": 0.3087907223027578, + "grad_norm": 0.130859375, + "learning_rate": 0.0014731547057570589, + "loss": 0.0933, + "step": 35573 + }, + { + "epoch": 0.308799402782962, + "grad_norm": 0.11083984375, + "learning_rate": 0.0014731276218271272, + "loss": 0.1152, + "step": 35574 + }, + { + "epoch": 0.3088080832631661, + "grad_norm": 1.3046875, + "learning_rate": 0.001473100537489149, + "loss": 0.1816, + "step": 35575 + }, + { + "epoch": 0.3088167637433703, + "grad_norm": 0.2890625, + "learning_rate": 0.0014730734527431536, + "loss": 0.0918, + "step": 35576 + }, + { + "epoch": 0.30882544422357444, + "grad_norm": 0.3359375, + "learning_rate": 0.0014730463675891705, + "loss": 0.0942, + "step": 35577 + }, + { + "epoch": 0.30883412470377863, + "grad_norm": 0.5, + "learning_rate": 0.0014730192820272296, + "loss": 0.1001, + "step": 35578 + }, + { + "epoch": 0.30884280518398277, + "grad_norm": 0.3359375, + "learning_rate": 0.0014729921960573602, + "loss": 0.1045, + "step": 35579 + }, + { + "epoch": 0.30885148566418696, + "grad_norm": 0.326171875, + "learning_rate": 0.0014729651096795928, + "loss": 0.1602, + "step": 35580 + }, + { + "epoch": 0.3088601661443911, + "grad_norm": 0.1943359375, + "learning_rate": 0.0014729380228939559, + "loss": 0.1387, + "step": 35581 + }, + { + "epoch": 0.3088688466245953, + "grad_norm": 0.251953125, + "learning_rate": 0.0014729109357004798, + "loss": 0.1416, + "step": 35582 + }, + { + "epoch": 0.30887752710479943, + "grad_norm": 0.4375, + "learning_rate": 0.0014728838480991934, + "loss": 0.1172, + "step": 35583 + }, + { + "epoch": 0.3088862075850036, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014728567600901272, + "loss": 0.0972, + "step": 35584 + }, + { + "epoch": 0.30889488806520776, + "grad_norm": 0.07763671875, + "learning_rate": 0.0014728296716733102, + "loss": 0.0869, + "step": 35585 + }, + { + "epoch": 0.30890356854541196, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014728025828487728, + "loss": 0.0986, + "step": 35586 + }, + { + "epoch": 0.3089122490256161, + "grad_norm": 0.15625, + "learning_rate": 0.0014727754936165436, + "loss": 0.0967, + "step": 35587 + }, + { + "epoch": 0.3089209295058203, + "grad_norm": 0.1640625, + "learning_rate": 0.0014727484039766528, + "loss": 0.1309, + "step": 35588 + }, + { + "epoch": 0.3089296099860244, + "grad_norm": 0.298828125, + "learning_rate": 0.0014727213139291303, + "loss": 0.1123, + "step": 35589 + }, + { + "epoch": 0.3089382904662286, + "grad_norm": 0.50390625, + "learning_rate": 0.001472694223474005, + "loss": 0.1602, + "step": 35590 + }, + { + "epoch": 0.30894697094643275, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014726671326113071, + "loss": 0.1211, + "step": 35591 + }, + { + "epoch": 0.30895565142663695, + "grad_norm": 0.2255859375, + "learning_rate": 0.0014726400413410659, + "loss": 0.0845, + "step": 35592 + }, + { + "epoch": 0.3089643319068411, + "grad_norm": 0.0771484375, + "learning_rate": 0.0014726129496633112, + "loss": 0.1147, + "step": 35593 + }, + { + "epoch": 0.3089730123870453, + "grad_norm": 0.423828125, + "learning_rate": 0.0014725858575780724, + "loss": 0.1426, + "step": 35594 + }, + { + "epoch": 0.3089816928672494, + "grad_norm": 0.50390625, + "learning_rate": 0.0014725587650853795, + "loss": 0.1582, + "step": 35595 + }, + { + "epoch": 0.3089903733474536, + "grad_norm": 0.2255859375, + "learning_rate": 0.0014725316721852617, + "loss": 0.0879, + "step": 35596 + }, + { + "epoch": 0.30899905382765774, + "grad_norm": 0.17578125, + "learning_rate": 0.0014725045788777493, + "loss": 0.1104, + "step": 35597 + }, + { + "epoch": 0.30900773430786194, + "grad_norm": 0.337890625, + "learning_rate": 0.0014724774851628713, + "loss": 0.0747, + "step": 35598 + }, + { + "epoch": 0.3090164147880661, + "grad_norm": 0.2333984375, + "learning_rate": 0.0014724503910406575, + "loss": 0.1016, + "step": 35599 + }, + { + "epoch": 0.30902509526827027, + "grad_norm": 0.328125, + "learning_rate": 0.0014724232965111373, + "loss": 0.0933, + "step": 35600 + }, + { + "epoch": 0.3090337757484744, + "grad_norm": 0.2275390625, + "learning_rate": 0.001472396201574341, + "loss": 0.1279, + "step": 35601 + }, + { + "epoch": 0.3090424562286786, + "grad_norm": 0.71875, + "learning_rate": 0.0014723691062302974, + "loss": 0.1182, + "step": 35602 + }, + { + "epoch": 0.30905113670888273, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014723420104790369, + "loss": 0.0854, + "step": 35603 + }, + { + "epoch": 0.3090598171890869, + "grad_norm": 0.462890625, + "learning_rate": 0.0014723149143205886, + "loss": 0.1123, + "step": 35604 + }, + { + "epoch": 0.30906849766929106, + "grad_norm": 0.1767578125, + "learning_rate": 0.001472287817754982, + "loss": 0.083, + "step": 35605 + }, + { + "epoch": 0.30907717814949526, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014722607207822474, + "loss": 0.0864, + "step": 35606 + }, + { + "epoch": 0.3090858586296994, + "grad_norm": 0.251953125, + "learning_rate": 0.0014722336234024142, + "loss": 0.0913, + "step": 35607 + }, + { + "epoch": 0.30909453910990353, + "grad_norm": 0.2109375, + "learning_rate": 0.001472206525615512, + "loss": 0.1147, + "step": 35608 + }, + { + "epoch": 0.3091032195901077, + "grad_norm": 0.11181640625, + "learning_rate": 0.0014721794274215696, + "loss": 0.1602, + "step": 35609 + }, + { + "epoch": 0.30911190007031186, + "grad_norm": 0.427734375, + "learning_rate": 0.001472152328820618, + "loss": 0.1055, + "step": 35610 + }, + { + "epoch": 0.30912058055051606, + "grad_norm": 0.50390625, + "learning_rate": 0.0014721252298126862, + "loss": 0.0869, + "step": 35611 + }, + { + "epoch": 0.3091292610307202, + "grad_norm": 0.271484375, + "learning_rate": 0.0014720981303978037, + "loss": 0.1367, + "step": 35612 + }, + { + "epoch": 0.3091379415109244, + "grad_norm": 0.359375, + "learning_rate": 0.0014720710305760002, + "loss": 0.0928, + "step": 35613 + }, + { + "epoch": 0.3091466219911285, + "grad_norm": 0.3671875, + "learning_rate": 0.0014720439303473052, + "loss": 0.1309, + "step": 35614 + }, + { + "epoch": 0.3091553024713327, + "grad_norm": 0.68359375, + "learning_rate": 0.0014720168297117488, + "loss": 0.1035, + "step": 35615 + }, + { + "epoch": 0.30916398295153685, + "grad_norm": 0.25390625, + "learning_rate": 0.0014719897286693607, + "loss": 0.0767, + "step": 35616 + }, + { + "epoch": 0.30917266343174105, + "grad_norm": 0.4140625, + "learning_rate": 0.0014719626272201695, + "loss": 0.103, + "step": 35617 + }, + { + "epoch": 0.3091813439119452, + "grad_norm": 0.90234375, + "learning_rate": 0.001471935525364206, + "loss": 0.1182, + "step": 35618 + }, + { + "epoch": 0.3091900243921494, + "grad_norm": 0.1904296875, + "learning_rate": 0.0014719084231014994, + "loss": 0.0742, + "step": 35619 + }, + { + "epoch": 0.3091987048723535, + "grad_norm": 0.16015625, + "learning_rate": 0.0014718813204320795, + "loss": 0.0825, + "step": 35620 + }, + { + "epoch": 0.3092073853525577, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014718542173559758, + "loss": 0.0854, + "step": 35621 + }, + { + "epoch": 0.30921606583276184, + "grad_norm": 0.14453125, + "learning_rate": 0.0014718271138732175, + "loss": 0.1064, + "step": 35622 + }, + { + "epoch": 0.30922474631296604, + "grad_norm": 0.68359375, + "learning_rate": 0.0014718000099838348, + "loss": 0.0928, + "step": 35623 + }, + { + "epoch": 0.3092334267931702, + "grad_norm": 0.2353515625, + "learning_rate": 0.0014717729056878573, + "loss": 0.1045, + "step": 35624 + }, + { + "epoch": 0.30924210727337437, + "grad_norm": 0.171875, + "learning_rate": 0.0014717458009853143, + "loss": 0.0942, + "step": 35625 + }, + { + "epoch": 0.3092507877535785, + "grad_norm": 0.1328125, + "learning_rate": 0.0014717186958762359, + "loss": 0.0967, + "step": 35626 + }, + { + "epoch": 0.3092594682337827, + "grad_norm": 0.35546875, + "learning_rate": 0.0014716915903606517, + "loss": 0.1328, + "step": 35627 + }, + { + "epoch": 0.30926814871398683, + "grad_norm": 0.10986328125, + "learning_rate": 0.001471664484438591, + "loss": 0.1064, + "step": 35628 + }, + { + "epoch": 0.309276829194191, + "grad_norm": 0.1640625, + "learning_rate": 0.0014716373781100836, + "loss": 0.064, + "step": 35629 + }, + { + "epoch": 0.30928550967439516, + "grad_norm": 0.53125, + "learning_rate": 0.0014716102713751588, + "loss": 0.126, + "step": 35630 + }, + { + "epoch": 0.30929419015459936, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014715831642338468, + "loss": 0.106, + "step": 35631 + }, + { + "epoch": 0.3093028706348035, + "grad_norm": 0.09521484375, + "learning_rate": 0.0014715560566861772, + "loss": 0.0708, + "step": 35632 + }, + { + "epoch": 0.3093115511150077, + "grad_norm": 0.6171875, + "learning_rate": 0.0014715289487321793, + "loss": 0.1289, + "step": 35633 + }, + { + "epoch": 0.3093202315952118, + "grad_norm": 0.27734375, + "learning_rate": 0.001471501840371883, + "loss": 0.1128, + "step": 35634 + }, + { + "epoch": 0.309328912075416, + "grad_norm": 0.2021484375, + "learning_rate": 0.001471474731605318, + "loss": 0.0618, + "step": 35635 + }, + { + "epoch": 0.30933759255562016, + "grad_norm": 0.25390625, + "learning_rate": 0.0014714476224325138, + "loss": 0.1182, + "step": 35636 + }, + { + "epoch": 0.30934627303582435, + "grad_norm": 0.173828125, + "learning_rate": 0.0014714205128534999, + "loss": 0.0942, + "step": 35637 + }, + { + "epoch": 0.3093549535160285, + "grad_norm": 0.53515625, + "learning_rate": 0.0014713934028683063, + "loss": 0.0942, + "step": 35638 + }, + { + "epoch": 0.3093636339962327, + "grad_norm": 0.125, + "learning_rate": 0.0014713662924769621, + "loss": 0.0879, + "step": 35639 + }, + { + "epoch": 0.3093723144764368, + "grad_norm": 0.1689453125, + "learning_rate": 0.0014713391816794977, + "loss": 0.1079, + "step": 35640 + }, + { + "epoch": 0.309380994956641, + "grad_norm": 0.16796875, + "learning_rate": 0.0014713120704759421, + "loss": 0.1426, + "step": 35641 + }, + { + "epoch": 0.30938967543684515, + "grad_norm": 0.08154296875, + "learning_rate": 0.0014712849588663253, + "loss": 0.1108, + "step": 35642 + }, + { + "epoch": 0.30939835591704934, + "grad_norm": 0.75390625, + "learning_rate": 0.0014712578468506765, + "loss": 0.1475, + "step": 35643 + }, + { + "epoch": 0.3094070363972535, + "grad_norm": 0.380859375, + "learning_rate": 0.001471230734429026, + "loss": 0.1021, + "step": 35644 + }, + { + "epoch": 0.30941571687745767, + "grad_norm": 0.51953125, + "learning_rate": 0.0014712036216014033, + "loss": 0.126, + "step": 35645 + }, + { + "epoch": 0.3094243973576618, + "grad_norm": 0.38671875, + "learning_rate": 0.001471176508367838, + "loss": 0.0732, + "step": 35646 + }, + { + "epoch": 0.309433077837866, + "grad_norm": 0.50390625, + "learning_rate": 0.0014711493947283593, + "loss": 0.085, + "step": 35647 + }, + { + "epoch": 0.30944175831807014, + "grad_norm": 0.10400390625, + "learning_rate": 0.0014711222806829973, + "loss": 0.0811, + "step": 35648 + }, + { + "epoch": 0.30945043879827433, + "grad_norm": 0.3203125, + "learning_rate": 0.0014710951662317815, + "loss": 0.0967, + "step": 35649 + }, + { + "epoch": 0.30945911927847847, + "grad_norm": 0.1103515625, + "learning_rate": 0.0014710680513747418, + "loss": 0.0728, + "step": 35650 + }, + { + "epoch": 0.30946779975868266, + "grad_norm": 0.27734375, + "learning_rate": 0.0014710409361119075, + "loss": 0.1221, + "step": 35651 + }, + { + "epoch": 0.3094764802388868, + "grad_norm": 0.671875, + "learning_rate": 0.0014710138204433083, + "loss": 0.1484, + "step": 35652 + }, + { + "epoch": 0.309485160719091, + "grad_norm": 0.11376953125, + "learning_rate": 0.0014709867043689743, + "loss": 0.0806, + "step": 35653 + }, + { + "epoch": 0.3094938411992951, + "grad_norm": 0.1650390625, + "learning_rate": 0.0014709595878889346, + "loss": 0.127, + "step": 35654 + }, + { + "epoch": 0.3095025216794993, + "grad_norm": 0.06640625, + "learning_rate": 0.001470932471003219, + "loss": 0.0781, + "step": 35655 + }, + { + "epoch": 0.30951120215970346, + "grad_norm": 0.1953125, + "learning_rate": 0.0014709053537118573, + "loss": 0.0737, + "step": 35656 + }, + { + "epoch": 0.30951988263990765, + "grad_norm": 0.1865234375, + "learning_rate": 0.001470878236014879, + "loss": 0.0747, + "step": 35657 + }, + { + "epoch": 0.3095285631201118, + "grad_norm": 0.119140625, + "learning_rate": 0.001470851117912314, + "loss": 0.1152, + "step": 35658 + }, + { + "epoch": 0.309537243600316, + "grad_norm": 0.255859375, + "learning_rate": 0.0014708239994041916, + "loss": 0.1133, + "step": 35659 + }, + { + "epoch": 0.3095459240805201, + "grad_norm": 0.1015625, + "learning_rate": 0.0014707968804905421, + "loss": 0.1152, + "step": 35660 + }, + { + "epoch": 0.3095546045607243, + "grad_norm": 0.69140625, + "learning_rate": 0.0014707697611713942, + "loss": 0.0859, + "step": 35661 + }, + { + "epoch": 0.30956328504092845, + "grad_norm": 0.11328125, + "learning_rate": 0.001470742641446778, + "loss": 0.1206, + "step": 35662 + }, + { + "epoch": 0.30957196552113264, + "grad_norm": 0.244140625, + "learning_rate": 0.0014707155213167233, + "loss": 0.1094, + "step": 35663 + }, + { + "epoch": 0.3095806460013368, + "grad_norm": 0.07373046875, + "learning_rate": 0.00147068840078126, + "loss": 0.0879, + "step": 35664 + }, + { + "epoch": 0.30958932648154097, + "grad_norm": 0.1953125, + "learning_rate": 0.0014706612798404172, + "loss": 0.0767, + "step": 35665 + }, + { + "epoch": 0.3095980069617451, + "grad_norm": 0.1044921875, + "learning_rate": 0.0014706341584942249, + "loss": 0.1138, + "step": 35666 + }, + { + "epoch": 0.3096066874419493, + "grad_norm": 0.447265625, + "learning_rate": 0.0014706070367427125, + "loss": 0.0854, + "step": 35667 + }, + { + "epoch": 0.30961536792215344, + "grad_norm": 0.2431640625, + "learning_rate": 0.0014705799145859101, + "loss": 0.1318, + "step": 35668 + }, + { + "epoch": 0.30962404840235763, + "grad_norm": 0.169921875, + "learning_rate": 0.0014705527920238467, + "loss": 0.1436, + "step": 35669 + }, + { + "epoch": 0.30963272888256177, + "grad_norm": 0.466796875, + "learning_rate": 0.0014705256690565524, + "loss": 0.1221, + "step": 35670 + }, + { + "epoch": 0.30964140936276596, + "grad_norm": 0.1640625, + "learning_rate": 0.0014704985456840567, + "loss": 0.0962, + "step": 35671 + }, + { + "epoch": 0.3096500898429701, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014704714219063897, + "loss": 0.1035, + "step": 35672 + }, + { + "epoch": 0.3096587703231743, + "grad_norm": 0.263671875, + "learning_rate": 0.0014704442977235806, + "loss": 0.0996, + "step": 35673 + }, + { + "epoch": 0.30966745080337843, + "grad_norm": 0.490234375, + "learning_rate": 0.0014704171731356591, + "loss": 0.1211, + "step": 35674 + }, + { + "epoch": 0.3096761312835826, + "grad_norm": 0.2333984375, + "learning_rate": 0.001470390048142655, + "loss": 0.1104, + "step": 35675 + }, + { + "epoch": 0.30968481176378676, + "grad_norm": 0.1962890625, + "learning_rate": 0.001470362922744598, + "loss": 0.0991, + "step": 35676 + }, + { + "epoch": 0.30969349224399095, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014703357969415177, + "loss": 0.0747, + "step": 35677 + }, + { + "epoch": 0.3097021727241951, + "grad_norm": 0.109375, + "learning_rate": 0.0014703086707334434, + "loss": 0.1123, + "step": 35678 + }, + { + "epoch": 0.3097108532043993, + "grad_norm": 0.455078125, + "learning_rate": 0.0014702815441204053, + "loss": 0.0913, + "step": 35679 + }, + { + "epoch": 0.3097195336846034, + "grad_norm": 0.12255859375, + "learning_rate": 0.001470254417102433, + "loss": 0.0723, + "step": 35680 + }, + { + "epoch": 0.3097282141648076, + "grad_norm": 0.314453125, + "learning_rate": 0.0014702272896795557, + "loss": 0.1113, + "step": 35681 + }, + { + "epoch": 0.30973689464501175, + "grad_norm": 0.484375, + "learning_rate": 0.0014702001618518035, + "loss": 0.0986, + "step": 35682 + }, + { + "epoch": 0.30974557512521594, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014701730336192063, + "loss": 0.1245, + "step": 35683 + }, + { + "epoch": 0.3097542556054201, + "grad_norm": 0.111328125, + "learning_rate": 0.001470145904981793, + "loss": 0.1001, + "step": 35684 + }, + { + "epoch": 0.3097629360856243, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014701187759395941, + "loss": 0.0898, + "step": 35685 + }, + { + "epoch": 0.3097716165658284, + "grad_norm": 0.279296875, + "learning_rate": 0.001470091646492639, + "loss": 0.0732, + "step": 35686 + }, + { + "epoch": 0.3097802970460326, + "grad_norm": 1.0234375, + "learning_rate": 0.0014700645166409567, + "loss": 0.1201, + "step": 35687 + }, + { + "epoch": 0.30978897752623674, + "grad_norm": 0.10400390625, + "learning_rate": 0.0014700373863845778, + "loss": 0.0688, + "step": 35688 + }, + { + "epoch": 0.30979765800644093, + "grad_norm": 0.466796875, + "learning_rate": 0.001470010255723531, + "loss": 0.1211, + "step": 35689 + }, + { + "epoch": 0.30980633848664507, + "grad_norm": 0.236328125, + "learning_rate": 0.001469983124657847, + "loss": 0.1035, + "step": 35690 + }, + { + "epoch": 0.30981501896684926, + "grad_norm": 0.09814453125, + "learning_rate": 0.001469955993187555, + "loss": 0.1216, + "step": 35691 + }, + { + "epoch": 0.3098236994470534, + "grad_norm": 0.35546875, + "learning_rate": 0.0014699288613126847, + "loss": 0.0918, + "step": 35692 + }, + { + "epoch": 0.3098323799272576, + "grad_norm": 0.94921875, + "learning_rate": 0.0014699017290332656, + "loss": 0.1143, + "step": 35693 + }, + { + "epoch": 0.30984106040746173, + "grad_norm": 0.458984375, + "learning_rate": 0.001469874596349328, + "loss": 0.1211, + "step": 35694 + }, + { + "epoch": 0.3098497408876659, + "grad_norm": 0.1044921875, + "learning_rate": 0.0014698474632609002, + "loss": 0.0859, + "step": 35695 + }, + { + "epoch": 0.30985842136787006, + "grad_norm": 0.12353515625, + "learning_rate": 0.0014698203297680134, + "loss": 0.1191, + "step": 35696 + }, + { + "epoch": 0.30986710184807426, + "grad_norm": 0.5859375, + "learning_rate": 0.0014697931958706967, + "loss": 0.0947, + "step": 35697 + }, + { + "epoch": 0.3098757823282784, + "grad_norm": 0.2578125, + "learning_rate": 0.0014697660615689795, + "loss": 0.0879, + "step": 35698 + }, + { + "epoch": 0.3098844628084826, + "grad_norm": 0.671875, + "learning_rate": 0.0014697389268628915, + "loss": 0.1367, + "step": 35699 + }, + { + "epoch": 0.3098931432886867, + "grad_norm": 0.4375, + "learning_rate": 0.001469711791752463, + "loss": 0.0737, + "step": 35700 + }, + { + "epoch": 0.3099018237688909, + "grad_norm": 0.609375, + "learning_rate": 0.001469684656237723, + "loss": 0.1016, + "step": 35701 + }, + { + "epoch": 0.30991050424909505, + "grad_norm": 0.28515625, + "learning_rate": 0.0014696575203187014, + "loss": 0.1143, + "step": 35702 + }, + { + "epoch": 0.30991918472929925, + "grad_norm": 0.287109375, + "learning_rate": 0.001469630383995428, + "loss": 0.0996, + "step": 35703 + }, + { + "epoch": 0.3099278652095034, + "grad_norm": 0.625, + "learning_rate": 0.0014696032472679321, + "loss": 0.0791, + "step": 35704 + }, + { + "epoch": 0.3099365456897076, + "grad_norm": 0.75, + "learning_rate": 0.001469576110136244, + "loss": 0.1113, + "step": 35705 + }, + { + "epoch": 0.3099452261699117, + "grad_norm": 0.26171875, + "learning_rate": 0.001469548972600393, + "loss": 0.0845, + "step": 35706 + }, + { + "epoch": 0.3099539066501159, + "grad_norm": 0.251953125, + "learning_rate": 0.0014695218346604086, + "loss": 0.1006, + "step": 35707 + }, + { + "epoch": 0.30996258713032004, + "grad_norm": 0.27734375, + "learning_rate": 0.0014694946963163208, + "loss": 0.127, + "step": 35708 + }, + { + "epoch": 0.30997126761052424, + "grad_norm": 0.2109375, + "learning_rate": 0.001469467557568159, + "loss": 0.1133, + "step": 35709 + }, + { + "epoch": 0.3099799480907284, + "grad_norm": 0.201171875, + "learning_rate": 0.0014694404184159532, + "loss": 0.2305, + "step": 35710 + }, + { + "epoch": 0.30998862857093257, + "grad_norm": 0.392578125, + "learning_rate": 0.0014694132788597328, + "loss": 0.0938, + "step": 35711 + }, + { + "epoch": 0.3099973090511367, + "grad_norm": 0.275390625, + "learning_rate": 0.0014693861388995278, + "loss": 0.1406, + "step": 35712 + }, + { + "epoch": 0.3100059895313409, + "grad_norm": 0.16015625, + "learning_rate": 0.0014693589985353673, + "loss": 0.0688, + "step": 35713 + }, + { + "epoch": 0.31001467001154503, + "grad_norm": 0.46875, + "learning_rate": 0.0014693318577672816, + "loss": 0.082, + "step": 35714 + }, + { + "epoch": 0.31002335049174923, + "grad_norm": 0.2392578125, + "learning_rate": 0.0014693047165952999, + "loss": 0.0884, + "step": 35715 + }, + { + "epoch": 0.31003203097195337, + "grad_norm": 0.201171875, + "learning_rate": 0.0014692775750194526, + "loss": 0.1465, + "step": 35716 + }, + { + "epoch": 0.31004071145215756, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014692504330397684, + "loss": 0.1055, + "step": 35717 + }, + { + "epoch": 0.3100493919323617, + "grad_norm": 0.2314453125, + "learning_rate": 0.0014692232906562777, + "loss": 0.1387, + "step": 35718 + }, + { + "epoch": 0.3100580724125659, + "grad_norm": 0.1845703125, + "learning_rate": 0.0014691961478690097, + "loss": 0.1426, + "step": 35719 + }, + { + "epoch": 0.31006675289277, + "grad_norm": 0.75, + "learning_rate": 0.0014691690046779947, + "loss": 0.1094, + "step": 35720 + }, + { + "epoch": 0.3100754333729742, + "grad_norm": 0.1123046875, + "learning_rate": 0.001469141861083262, + "loss": 0.0801, + "step": 35721 + }, + { + "epoch": 0.31008411385317836, + "grad_norm": 0.10888671875, + "learning_rate": 0.0014691147170848408, + "loss": 0.1191, + "step": 35722 + }, + { + "epoch": 0.31009279433338255, + "grad_norm": 0.171875, + "learning_rate": 0.0014690875726827619, + "loss": 0.0771, + "step": 35723 + }, + { + "epoch": 0.3101014748135867, + "grad_norm": 0.25390625, + "learning_rate": 0.001469060427877054, + "loss": 0.0942, + "step": 35724 + }, + { + "epoch": 0.3101101552937909, + "grad_norm": 0.37890625, + "learning_rate": 0.0014690332826677475, + "loss": 0.1465, + "step": 35725 + }, + { + "epoch": 0.310118835773995, + "grad_norm": 0.50390625, + "learning_rate": 0.0014690061370548715, + "loss": 0.0986, + "step": 35726 + }, + { + "epoch": 0.3101275162541992, + "grad_norm": 0.49609375, + "learning_rate": 0.001468978991038456, + "loss": 0.0972, + "step": 35727 + }, + { + "epoch": 0.31013619673440335, + "grad_norm": 0.32421875, + "learning_rate": 0.0014689518446185305, + "loss": 0.0903, + "step": 35728 + }, + { + "epoch": 0.31014487721460754, + "grad_norm": 0.05859375, + "learning_rate": 0.001468924697795125, + "loss": 0.0732, + "step": 35729 + }, + { + "epoch": 0.3101535576948117, + "grad_norm": 0.423828125, + "learning_rate": 0.0014688975505682688, + "loss": 0.0718, + "step": 35730 + }, + { + "epoch": 0.3101622381750158, + "grad_norm": 0.16015625, + "learning_rate": 0.0014688704029379917, + "loss": 0.0854, + "step": 35731 + }, + { + "epoch": 0.31017091865522, + "grad_norm": 0.1494140625, + "learning_rate": 0.001468843254904324, + "loss": 0.0908, + "step": 35732 + }, + { + "epoch": 0.31017959913542414, + "grad_norm": 0.427734375, + "learning_rate": 0.0014688161064672941, + "loss": 0.1172, + "step": 35733 + }, + { + "epoch": 0.31018827961562834, + "grad_norm": 0.431640625, + "learning_rate": 0.001468788957626933, + "loss": 0.1416, + "step": 35734 + }, + { + "epoch": 0.3101969600958325, + "grad_norm": 0.24609375, + "learning_rate": 0.0014687618083832697, + "loss": 0.0835, + "step": 35735 + }, + { + "epoch": 0.31020564057603667, + "grad_norm": 0.126953125, + "learning_rate": 0.0014687346587363343, + "loss": 0.1147, + "step": 35736 + }, + { + "epoch": 0.3102143210562408, + "grad_norm": 0.162109375, + "learning_rate": 0.0014687075086861558, + "loss": 0.0811, + "step": 35737 + }, + { + "epoch": 0.310223001536445, + "grad_norm": 0.296875, + "learning_rate": 0.0014686803582327644, + "loss": 0.0957, + "step": 35738 + }, + { + "epoch": 0.31023168201664914, + "grad_norm": 0.23046875, + "learning_rate": 0.0014686532073761899, + "loss": 0.1182, + "step": 35739 + }, + { + "epoch": 0.31024036249685333, + "grad_norm": 0.05859375, + "learning_rate": 0.0014686260561164613, + "loss": 0.0796, + "step": 35740 + }, + { + "epoch": 0.31024904297705747, + "grad_norm": 0.26953125, + "learning_rate": 0.0014685989044536092, + "loss": 0.1035, + "step": 35741 + }, + { + "epoch": 0.31025772345726166, + "grad_norm": 0.8203125, + "learning_rate": 0.0014685717523876631, + "loss": 0.1387, + "step": 35742 + }, + { + "epoch": 0.3102664039374658, + "grad_norm": 1.640625, + "learning_rate": 0.0014685445999186519, + "loss": 0.1494, + "step": 35743 + }, + { + "epoch": 0.31027508441767, + "grad_norm": 0.484375, + "learning_rate": 0.0014685174470466062, + "loss": 0.1099, + "step": 35744 + }, + { + "epoch": 0.3102837648978741, + "grad_norm": 0.322265625, + "learning_rate": 0.0014684902937715555, + "loss": 0.0835, + "step": 35745 + }, + { + "epoch": 0.3102924453780783, + "grad_norm": 0.11572265625, + "learning_rate": 0.0014684631400935292, + "loss": 0.103, + "step": 35746 + }, + { + "epoch": 0.31030112585828246, + "grad_norm": 0.498046875, + "learning_rate": 0.001468435986012557, + "loss": 0.1123, + "step": 35747 + }, + { + "epoch": 0.31030980633848665, + "grad_norm": 0.3203125, + "learning_rate": 0.0014684088315286687, + "loss": 0.1113, + "step": 35748 + }, + { + "epoch": 0.3103184868186908, + "grad_norm": 0.42578125, + "learning_rate": 0.001468381676641894, + "loss": 0.1875, + "step": 35749 + }, + { + "epoch": 0.310327167298895, + "grad_norm": 0.5078125, + "learning_rate": 0.0014683545213522629, + "loss": 0.1069, + "step": 35750 + }, + { + "epoch": 0.3103358477790991, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014683273656598047, + "loss": 0.0986, + "step": 35751 + }, + { + "epoch": 0.3103445282593033, + "grad_norm": 0.298828125, + "learning_rate": 0.0014683002095645492, + "loss": 0.1006, + "step": 35752 + }, + { + "epoch": 0.31035320873950745, + "grad_norm": 0.2119140625, + "learning_rate": 0.0014682730530665262, + "loss": 0.1025, + "step": 35753 + }, + { + "epoch": 0.31036188921971164, + "grad_norm": 0.43359375, + "learning_rate": 0.0014682458961657655, + "loss": 0.0972, + "step": 35754 + }, + { + "epoch": 0.3103705696999158, + "grad_norm": 1.15625, + "learning_rate": 0.0014682187388622963, + "loss": 0.1128, + "step": 35755 + }, + { + "epoch": 0.31037925018011997, + "grad_norm": 0.76953125, + "learning_rate": 0.0014681915811561486, + "loss": 0.0986, + "step": 35756 + }, + { + "epoch": 0.3103879306603241, + "grad_norm": 0.67578125, + "learning_rate": 0.0014681644230473522, + "loss": 0.1104, + "step": 35757 + }, + { + "epoch": 0.3103966111405283, + "grad_norm": 0.490234375, + "learning_rate": 0.0014681372645359368, + "loss": 0.1348, + "step": 35758 + }, + { + "epoch": 0.31040529162073244, + "grad_norm": 0.419921875, + "learning_rate": 0.0014681101056219321, + "loss": 0.1406, + "step": 35759 + }, + { + "epoch": 0.31041397210093663, + "grad_norm": 0.5234375, + "learning_rate": 0.0014680829463053675, + "loss": 0.0938, + "step": 35760 + }, + { + "epoch": 0.31042265258114077, + "grad_norm": 0.302734375, + "learning_rate": 0.0014680557865862731, + "loss": 0.0859, + "step": 35761 + }, + { + "epoch": 0.31043133306134496, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014680286264646784, + "loss": 0.1182, + "step": 35762 + }, + { + "epoch": 0.3104400135415491, + "grad_norm": 1.015625, + "learning_rate": 0.0014680014659406134, + "loss": 0.1187, + "step": 35763 + }, + { + "epoch": 0.3104486940217533, + "grad_norm": 0.16796875, + "learning_rate": 0.001467974305014107, + "loss": 0.1055, + "step": 35764 + }, + { + "epoch": 0.31045737450195743, + "grad_norm": 0.091796875, + "learning_rate": 0.0014679471436851897, + "loss": 0.085, + "step": 35765 + }, + { + "epoch": 0.3104660549821616, + "grad_norm": 0.076171875, + "learning_rate": 0.0014679199819538909, + "loss": 0.085, + "step": 35766 + }, + { + "epoch": 0.31047473546236576, + "grad_norm": 0.49609375, + "learning_rate": 0.00146789281982024, + "loss": 0.1064, + "step": 35767 + }, + { + "epoch": 0.31048341594256995, + "grad_norm": 0.44140625, + "learning_rate": 0.0014678656572842674, + "loss": 0.1055, + "step": 35768 + }, + { + "epoch": 0.3104920964227741, + "grad_norm": 0.10302734375, + "learning_rate": 0.0014678384943460025, + "loss": 0.1216, + "step": 35769 + }, + { + "epoch": 0.3105007769029783, + "grad_norm": 0.3125, + "learning_rate": 0.0014678113310054748, + "loss": 0.1357, + "step": 35770 + }, + { + "epoch": 0.3105094573831824, + "grad_norm": 0.16015625, + "learning_rate": 0.0014677841672627143, + "loss": 0.0791, + "step": 35771 + }, + { + "epoch": 0.3105181378633866, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014677570031177505, + "loss": 0.0962, + "step": 35772 + }, + { + "epoch": 0.31052681834359075, + "grad_norm": 0.35546875, + "learning_rate": 0.001467729838570613, + "loss": 0.0928, + "step": 35773 + }, + { + "epoch": 0.31053549882379494, + "grad_norm": 0.3671875, + "learning_rate": 0.0014677026736213317, + "loss": 0.1621, + "step": 35774 + }, + { + "epoch": 0.3105441793039991, + "grad_norm": 0.99609375, + "learning_rate": 0.0014676755082699364, + "loss": 0.1787, + "step": 35775 + }, + { + "epoch": 0.3105528597842033, + "grad_norm": 0.365234375, + "learning_rate": 0.0014676483425164566, + "loss": 0.0737, + "step": 35776 + }, + { + "epoch": 0.3105615402644074, + "grad_norm": 0.12255859375, + "learning_rate": 0.001467621176360922, + "loss": 0.0693, + "step": 35777 + }, + { + "epoch": 0.3105702207446116, + "grad_norm": 0.58203125, + "learning_rate": 0.0014675940098033626, + "loss": 0.0889, + "step": 35778 + }, + { + "epoch": 0.31057890122481574, + "grad_norm": 1.046875, + "learning_rate": 0.001467566842843808, + "loss": 0.1514, + "step": 35779 + }, + { + "epoch": 0.31058758170501993, + "grad_norm": 2.890625, + "learning_rate": 0.0014675396754822879, + "loss": 0.1826, + "step": 35780 + }, + { + "epoch": 0.31059626218522407, + "grad_norm": 0.5625, + "learning_rate": 0.0014675125077188318, + "loss": 0.1064, + "step": 35781 + }, + { + "epoch": 0.31060494266542826, + "grad_norm": 0.423828125, + "learning_rate": 0.001467485339553469, + "loss": 0.0972, + "step": 35782 + }, + { + "epoch": 0.3106136231456324, + "grad_norm": 0.2890625, + "learning_rate": 0.0014674581709862306, + "loss": 0.1104, + "step": 35783 + }, + { + "epoch": 0.3106223036258366, + "grad_norm": 0.2412109375, + "learning_rate": 0.001467431002017145, + "loss": 0.0918, + "step": 35784 + }, + { + "epoch": 0.31063098410604073, + "grad_norm": 0.07568359375, + "learning_rate": 0.0014674038326462427, + "loss": 0.1011, + "step": 35785 + }, + { + "epoch": 0.3106396645862449, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014673766628735528, + "loss": 0.1221, + "step": 35786 + }, + { + "epoch": 0.31064834506644906, + "grad_norm": 0.10498046875, + "learning_rate": 0.0014673494926991054, + "loss": 0.1069, + "step": 35787 + }, + { + "epoch": 0.31065702554665325, + "grad_norm": 0.08447265625, + "learning_rate": 0.00146732232212293, + "loss": 0.0854, + "step": 35788 + }, + { + "epoch": 0.3106657060268574, + "grad_norm": 0.3203125, + "learning_rate": 0.0014672951511450565, + "loss": 0.1089, + "step": 35789 + }, + { + "epoch": 0.3106743865070616, + "grad_norm": 0.3046875, + "learning_rate": 0.0014672679797655143, + "loss": 0.1309, + "step": 35790 + }, + { + "epoch": 0.3106830669872657, + "grad_norm": 0.1064453125, + "learning_rate": 0.0014672408079843336, + "loss": 0.0957, + "step": 35791 + }, + { + "epoch": 0.3106917474674699, + "grad_norm": 0.12890625, + "learning_rate": 0.001467213635801544, + "loss": 0.1279, + "step": 35792 + }, + { + "epoch": 0.31070042794767405, + "grad_norm": 0.1845703125, + "learning_rate": 0.001467186463217175, + "loss": 0.1035, + "step": 35793 + }, + { + "epoch": 0.31070910842787824, + "grad_norm": 0.330078125, + "learning_rate": 0.0014671592902312566, + "loss": 0.1104, + "step": 35794 + }, + { + "epoch": 0.3107177889080824, + "grad_norm": 0.48828125, + "learning_rate": 0.001467132116843818, + "loss": 0.0732, + "step": 35795 + }, + { + "epoch": 0.3107264693882866, + "grad_norm": 0.19140625, + "learning_rate": 0.001467104943054889, + "loss": 0.0718, + "step": 35796 + }, + { + "epoch": 0.3107351498684907, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014670777688645, + "loss": 0.0967, + "step": 35797 + }, + { + "epoch": 0.3107438303486949, + "grad_norm": 0.5703125, + "learning_rate": 0.00146705059427268, + "loss": 0.1138, + "step": 35798 + }, + { + "epoch": 0.31075251082889904, + "grad_norm": 0.6328125, + "learning_rate": 0.001467023419279459, + "loss": 0.0928, + "step": 35799 + }, + { + "epoch": 0.31076119130910324, + "grad_norm": 0.1953125, + "learning_rate": 0.0014669962438848669, + "loss": 0.1016, + "step": 35800 + }, + { + "epoch": 0.3107698717893074, + "grad_norm": 0.236328125, + "learning_rate": 0.0014669690680889334, + "loss": 0.1797, + "step": 35801 + }, + { + "epoch": 0.31077855226951157, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014669418918916877, + "loss": 0.0688, + "step": 35802 + }, + { + "epoch": 0.3107872327497157, + "grad_norm": 0.1044921875, + "learning_rate": 0.00146691471529316, + "loss": 0.1074, + "step": 35803 + }, + { + "epoch": 0.3107959132299199, + "grad_norm": 0.435546875, + "learning_rate": 0.0014668875382933798, + "loss": 0.1543, + "step": 35804 + }, + { + "epoch": 0.31080459371012403, + "grad_norm": 1.453125, + "learning_rate": 0.001466860360892377, + "loss": 0.1338, + "step": 35805 + }, + { + "epoch": 0.3108132741903282, + "grad_norm": 0.640625, + "learning_rate": 0.0014668331830901809, + "loss": 0.0674, + "step": 35806 + }, + { + "epoch": 0.31082195467053236, + "grad_norm": 0.10009765625, + "learning_rate": 0.0014668060048868218, + "loss": 0.1074, + "step": 35807 + }, + { + "epoch": 0.31083063515073656, + "grad_norm": 1.96875, + "learning_rate": 0.0014667788262823292, + "loss": 0.126, + "step": 35808 + }, + { + "epoch": 0.3108393156309407, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014667516472767325, + "loss": 0.0894, + "step": 35809 + }, + { + "epoch": 0.3108479961111449, + "grad_norm": 0.1630859375, + "learning_rate": 0.0014667244678700621, + "loss": 0.0977, + "step": 35810 + }, + { + "epoch": 0.310856676591349, + "grad_norm": 0.11572265625, + "learning_rate": 0.0014666972880623475, + "loss": 0.0981, + "step": 35811 + }, + { + "epoch": 0.3108653570715532, + "grad_norm": 0.0830078125, + "learning_rate": 0.0014666701078536177, + "loss": 0.0879, + "step": 35812 + }, + { + "epoch": 0.31087403755175735, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014666429272439032, + "loss": 0.1045, + "step": 35813 + }, + { + "epoch": 0.31088271803196155, + "grad_norm": 0.52734375, + "learning_rate": 0.0014666157462332335, + "loss": 0.0942, + "step": 35814 + }, + { + "epoch": 0.3108913985121657, + "grad_norm": 0.33203125, + "learning_rate": 0.0014665885648216383, + "loss": 0.083, + "step": 35815 + }, + { + "epoch": 0.3109000789923699, + "grad_norm": 0.345703125, + "learning_rate": 0.0014665613830091473, + "loss": 0.1245, + "step": 35816 + }, + { + "epoch": 0.310908759472574, + "grad_norm": 0.0859375, + "learning_rate": 0.0014665342007957904, + "loss": 0.0938, + "step": 35817 + }, + { + "epoch": 0.3109174399527782, + "grad_norm": 0.267578125, + "learning_rate": 0.0014665070181815972, + "loss": 0.1006, + "step": 35818 + }, + { + "epoch": 0.31092612043298234, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014664798351665975, + "loss": 0.1094, + "step": 35819 + }, + { + "epoch": 0.31093480091318654, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014664526517508209, + "loss": 0.0938, + "step": 35820 + }, + { + "epoch": 0.3109434813933907, + "grad_norm": 0.0830078125, + "learning_rate": 0.001466425467934297, + "loss": 0.0894, + "step": 35821 + }, + { + "epoch": 0.31095216187359487, + "grad_norm": 0.134765625, + "learning_rate": 0.001466398283717056, + "loss": 0.126, + "step": 35822 + }, + { + "epoch": 0.310960842353799, + "grad_norm": 0.451171875, + "learning_rate": 0.0014663710990991273, + "loss": 0.1084, + "step": 35823 + }, + { + "epoch": 0.3109695228340032, + "grad_norm": 0.134765625, + "learning_rate": 0.0014663439140805404, + "loss": 0.1089, + "step": 35824 + }, + { + "epoch": 0.31097820331420734, + "grad_norm": 0.12109375, + "learning_rate": 0.0014663167286613256, + "loss": 0.0791, + "step": 35825 + }, + { + "epoch": 0.31098688379441153, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014662895428415122, + "loss": 0.0967, + "step": 35826 + }, + { + "epoch": 0.31099556427461567, + "grad_norm": 0.2421875, + "learning_rate": 0.0014662623566211298, + "loss": 0.084, + "step": 35827 + }, + { + "epoch": 0.31100424475481986, + "grad_norm": 0.41796875, + "learning_rate": 0.001466235170000209, + "loss": 0.0713, + "step": 35828 + }, + { + "epoch": 0.311012925235024, + "grad_norm": 0.162109375, + "learning_rate": 0.0014662079829787784, + "loss": 0.0894, + "step": 35829 + }, + { + "epoch": 0.3110216057152282, + "grad_norm": 0.59375, + "learning_rate": 0.0014661807955568685, + "loss": 0.1191, + "step": 35830 + }, + { + "epoch": 0.3110302861954323, + "grad_norm": 0.08935546875, + "learning_rate": 0.0014661536077345085, + "loss": 0.0986, + "step": 35831 + }, + { + "epoch": 0.3110389666756365, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014661264195117288, + "loss": 0.1104, + "step": 35832 + }, + { + "epoch": 0.31104764715584066, + "grad_norm": 0.1005859375, + "learning_rate": 0.0014660992308885587, + "loss": 0.1016, + "step": 35833 + }, + { + "epoch": 0.31105632763604485, + "grad_norm": 0.287109375, + "learning_rate": 0.0014660720418650278, + "loss": 0.1143, + "step": 35834 + }, + { + "epoch": 0.311065008116249, + "grad_norm": 0.345703125, + "learning_rate": 0.001466044852441166, + "loss": 0.0654, + "step": 35835 + }, + { + "epoch": 0.3110736885964532, + "grad_norm": 0.1025390625, + "learning_rate": 0.001466017662617003, + "loss": 0.0918, + "step": 35836 + }, + { + "epoch": 0.3110823690766573, + "grad_norm": 0.326171875, + "learning_rate": 0.0014659904723925688, + "loss": 0.0864, + "step": 35837 + }, + { + "epoch": 0.3110910495568615, + "grad_norm": 0.2109375, + "learning_rate": 0.001465963281767893, + "loss": 0.0586, + "step": 35838 + }, + { + "epoch": 0.31109973003706565, + "grad_norm": 0.2275390625, + "learning_rate": 0.001465936090743005, + "loss": 0.0879, + "step": 35839 + }, + { + "epoch": 0.31110841051726984, + "grad_norm": 0.228515625, + "learning_rate": 0.0014659088993179347, + "loss": 0.0889, + "step": 35840 + }, + { + "epoch": 0.311117090997474, + "grad_norm": 0.25, + "learning_rate": 0.0014658817074927122, + "loss": 0.1006, + "step": 35841 + }, + { + "epoch": 0.31112577147767817, + "grad_norm": 1.7734375, + "learning_rate": 0.0014658545152673667, + "loss": 0.1338, + "step": 35842 + }, + { + "epoch": 0.3111344519578823, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014658273226419284, + "loss": 0.0923, + "step": 35843 + }, + { + "epoch": 0.3111431324380865, + "grad_norm": 0.283203125, + "learning_rate": 0.0014658001296164267, + "loss": 0.1045, + "step": 35844 + }, + { + "epoch": 0.31115181291829064, + "grad_norm": 0.193359375, + "learning_rate": 0.001465772936190892, + "loss": 0.1055, + "step": 35845 + }, + { + "epoch": 0.31116049339849483, + "grad_norm": 0.55859375, + "learning_rate": 0.001465745742365353, + "loss": 0.1006, + "step": 35846 + }, + { + "epoch": 0.31116917387869897, + "grad_norm": 0.1826171875, + "learning_rate": 0.00146571854813984, + "loss": 0.1758, + "step": 35847 + }, + { + "epoch": 0.31117785435890316, + "grad_norm": 0.6953125, + "learning_rate": 0.0014656913535143826, + "loss": 0.1211, + "step": 35848 + }, + { + "epoch": 0.3111865348391073, + "grad_norm": 0.09423828125, + "learning_rate": 0.0014656641584890107, + "loss": 0.0767, + "step": 35849 + }, + { + "epoch": 0.3111952153193115, + "grad_norm": 0.39453125, + "learning_rate": 0.0014656369630637544, + "loss": 0.1064, + "step": 35850 + }, + { + "epoch": 0.31120389579951563, + "grad_norm": 0.30078125, + "learning_rate": 0.0014656097672386423, + "loss": 0.1172, + "step": 35851 + }, + { + "epoch": 0.3112125762797198, + "grad_norm": 0.291015625, + "learning_rate": 0.0014655825710137053, + "loss": 0.0815, + "step": 35852 + }, + { + "epoch": 0.31122125675992396, + "grad_norm": 0.42578125, + "learning_rate": 0.0014655553743889724, + "loss": 0.0879, + "step": 35853 + }, + { + "epoch": 0.31122993724012815, + "grad_norm": 0.353515625, + "learning_rate": 0.001465528177364474, + "loss": 0.1299, + "step": 35854 + }, + { + "epoch": 0.3112386177203323, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014655009799402392, + "loss": 0.0967, + "step": 35855 + }, + { + "epoch": 0.3112472982005364, + "grad_norm": 0.1376953125, + "learning_rate": 0.001465473782116298, + "loss": 0.083, + "step": 35856 + }, + { + "epoch": 0.3112559786807406, + "grad_norm": 0.1328125, + "learning_rate": 0.0014654465838926805, + "loss": 0.0913, + "step": 35857 + }, + { + "epoch": 0.31126465916094476, + "grad_norm": 0.0693359375, + "learning_rate": 0.0014654193852694157, + "loss": 0.0771, + "step": 35858 + }, + { + "epoch": 0.31127333964114895, + "grad_norm": 0.4140625, + "learning_rate": 0.0014653921862465339, + "loss": 0.0908, + "step": 35859 + }, + { + "epoch": 0.3112820201213531, + "grad_norm": 0.83203125, + "learning_rate": 0.0014653649868240648, + "loss": 0.1387, + "step": 35860 + }, + { + "epoch": 0.3112907006015573, + "grad_norm": 0.1259765625, + "learning_rate": 0.0014653377870020381, + "loss": 0.125, + "step": 35861 + }, + { + "epoch": 0.3112993810817614, + "grad_norm": 0.109375, + "learning_rate": 0.0014653105867804834, + "loss": 0.0938, + "step": 35862 + }, + { + "epoch": 0.3113080615619656, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014652833861594301, + "loss": 0.1025, + "step": 35863 + }, + { + "epoch": 0.31131674204216975, + "grad_norm": 0.267578125, + "learning_rate": 0.0014652561851389088, + "loss": 0.1348, + "step": 35864 + }, + { + "epoch": 0.31132542252237394, + "grad_norm": 0.3203125, + "learning_rate": 0.001465228983718949, + "loss": 0.1025, + "step": 35865 + }, + { + "epoch": 0.3113341030025781, + "grad_norm": 0.220703125, + "learning_rate": 0.00146520178189958, + "loss": 0.1104, + "step": 35866 + }, + { + "epoch": 0.31134278348278227, + "grad_norm": 0.3984375, + "learning_rate": 0.001465174579680832, + "loss": 0.1025, + "step": 35867 + }, + { + "epoch": 0.3113514639629864, + "grad_norm": 0.0791015625, + "learning_rate": 0.0014651473770627343, + "loss": 0.0771, + "step": 35868 + }, + { + "epoch": 0.3113601444431906, + "grad_norm": 0.1083984375, + "learning_rate": 0.001465120174045317, + "loss": 0.1128, + "step": 35869 + }, + { + "epoch": 0.31136882492339474, + "grad_norm": 0.45703125, + "learning_rate": 0.0014650929706286101, + "loss": 0.0776, + "step": 35870 + }, + { + "epoch": 0.31137750540359893, + "grad_norm": 0.283203125, + "learning_rate": 0.0014650657668126429, + "loss": 0.0869, + "step": 35871 + }, + { + "epoch": 0.31138618588380307, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014650385625974446, + "loss": 0.1182, + "step": 35872 + }, + { + "epoch": 0.31139486636400726, + "grad_norm": 0.287109375, + "learning_rate": 0.001465011357983046, + "loss": 0.0898, + "step": 35873 + }, + { + "epoch": 0.3114035468442114, + "grad_norm": 0.23046875, + "learning_rate": 0.0014649841529694768, + "loss": 0.1094, + "step": 35874 + }, + { + "epoch": 0.3114122273244156, + "grad_norm": 0.82421875, + "learning_rate": 0.0014649569475567662, + "loss": 0.1123, + "step": 35875 + }, + { + "epoch": 0.31142090780461973, + "grad_norm": 0.4609375, + "learning_rate": 0.0014649297417449441, + "loss": 0.0962, + "step": 35876 + }, + { + "epoch": 0.3114295882848239, + "grad_norm": 0.69140625, + "learning_rate": 0.0014649025355340404, + "loss": 0.0967, + "step": 35877 + }, + { + "epoch": 0.31143826876502806, + "grad_norm": 0.44921875, + "learning_rate": 0.0014648753289240846, + "loss": 0.1279, + "step": 35878 + }, + { + "epoch": 0.31144694924523225, + "grad_norm": 0.306640625, + "learning_rate": 0.0014648481219151068, + "loss": 0.1011, + "step": 35879 + }, + { + "epoch": 0.3114556297254364, + "grad_norm": 0.4453125, + "learning_rate": 0.0014648209145071368, + "loss": 0.126, + "step": 35880 + }, + { + "epoch": 0.3114643102056406, + "grad_norm": 0.421875, + "learning_rate": 0.0014647937067002038, + "loss": 0.0957, + "step": 35881 + }, + { + "epoch": 0.3114729906858447, + "grad_norm": 0.220703125, + "learning_rate": 0.0014647664984943378, + "loss": 0.1123, + "step": 35882 + }, + { + "epoch": 0.3114816711660489, + "grad_norm": 0.232421875, + "learning_rate": 0.0014647392898895688, + "loss": 0.1113, + "step": 35883 + }, + { + "epoch": 0.31149035164625305, + "grad_norm": 0.375, + "learning_rate": 0.0014647120808859264, + "loss": 0.1133, + "step": 35884 + }, + { + "epoch": 0.31149903212645724, + "grad_norm": 0.185546875, + "learning_rate": 0.0014646848714834403, + "loss": 0.103, + "step": 35885 + }, + { + "epoch": 0.3115077126066614, + "grad_norm": 0.392578125, + "learning_rate": 0.0014646576616821404, + "loss": 0.1006, + "step": 35886 + }, + { + "epoch": 0.3115163930868656, + "grad_norm": 0.1455078125, + "learning_rate": 0.0014646304514820564, + "loss": 0.0918, + "step": 35887 + }, + { + "epoch": 0.3115250735670697, + "grad_norm": 2.328125, + "learning_rate": 0.001464603240883218, + "loss": 0.2246, + "step": 35888 + }, + { + "epoch": 0.3115337540472739, + "grad_norm": 0.42578125, + "learning_rate": 0.0014645760298856547, + "loss": 0.083, + "step": 35889 + }, + { + "epoch": 0.31154243452747804, + "grad_norm": 0.404296875, + "learning_rate": 0.0014645488184893967, + "loss": 0.1543, + "step": 35890 + }, + { + "epoch": 0.31155111500768223, + "grad_norm": 0.12060546875, + "learning_rate": 0.0014645216066944736, + "loss": 0.1089, + "step": 35891 + }, + { + "epoch": 0.31155979548788637, + "grad_norm": 1.5625, + "learning_rate": 0.001464494394500915, + "loss": 0.2324, + "step": 35892 + }, + { + "epoch": 0.31156847596809056, + "grad_norm": 0.11279296875, + "learning_rate": 0.0014644671819087512, + "loss": 0.0864, + "step": 35893 + }, + { + "epoch": 0.3115771564482947, + "grad_norm": 0.1708984375, + "learning_rate": 0.001464439968918011, + "loss": 0.1777, + "step": 35894 + }, + { + "epoch": 0.3115858369284989, + "grad_norm": 0.31640625, + "learning_rate": 0.0014644127555287253, + "loss": 0.1074, + "step": 35895 + }, + { + "epoch": 0.31159451740870303, + "grad_norm": 0.3125, + "learning_rate": 0.001464385541740923, + "loss": 0.0732, + "step": 35896 + }, + { + "epoch": 0.3116031978889072, + "grad_norm": 0.388671875, + "learning_rate": 0.0014643583275546345, + "loss": 0.1123, + "step": 35897 + }, + { + "epoch": 0.31161187836911136, + "grad_norm": 0.08740234375, + "learning_rate": 0.0014643311129698886, + "loss": 0.1152, + "step": 35898 + }, + { + "epoch": 0.31162055884931555, + "grad_norm": 0.2099609375, + "learning_rate": 0.001464303897986716, + "loss": 0.104, + "step": 35899 + }, + { + "epoch": 0.3116292393295197, + "grad_norm": 0.12890625, + "learning_rate": 0.0014642766826051458, + "loss": 0.0918, + "step": 35900 + }, + { + "epoch": 0.3116379198097239, + "grad_norm": 0.345703125, + "learning_rate": 0.0014642494668252084, + "loss": 0.1162, + "step": 35901 + }, + { + "epoch": 0.311646600289928, + "grad_norm": 0.11181640625, + "learning_rate": 0.0014642222506469334, + "loss": 0.0918, + "step": 35902 + }, + { + "epoch": 0.3116552807701322, + "grad_norm": 0.205078125, + "learning_rate": 0.0014641950340703502, + "loss": 0.0938, + "step": 35903 + }, + { + "epoch": 0.31166396125033635, + "grad_norm": 0.283203125, + "learning_rate": 0.0014641678170954889, + "loss": 0.0581, + "step": 35904 + }, + { + "epoch": 0.31167264173054054, + "grad_norm": 0.58203125, + "learning_rate": 0.001464140599722379, + "loss": 0.0967, + "step": 35905 + }, + { + "epoch": 0.3116813222107447, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014641133819510506, + "loss": 0.1221, + "step": 35906 + }, + { + "epoch": 0.3116900026909489, + "grad_norm": 0.57421875, + "learning_rate": 0.0014640861637815331, + "loss": 0.0781, + "step": 35907 + }, + { + "epoch": 0.311698683171153, + "grad_norm": 0.326171875, + "learning_rate": 0.0014640589452138567, + "loss": 0.125, + "step": 35908 + }, + { + "epoch": 0.3117073636513572, + "grad_norm": 0.23828125, + "learning_rate": 0.0014640317262480506, + "loss": 0.1162, + "step": 35909 + }, + { + "epoch": 0.31171604413156134, + "grad_norm": 0.09912109375, + "learning_rate": 0.001464004506884145, + "loss": 0.0757, + "step": 35910 + }, + { + "epoch": 0.31172472461176554, + "grad_norm": 1.484375, + "learning_rate": 0.0014639772871221696, + "loss": 0.3457, + "step": 35911 + }, + { + "epoch": 0.3117334050919697, + "grad_norm": 0.0947265625, + "learning_rate": 0.001463950066962154, + "loss": 0.1074, + "step": 35912 + }, + { + "epoch": 0.31174208557217387, + "grad_norm": 0.2080078125, + "learning_rate": 0.001463922846404128, + "loss": 0.0654, + "step": 35913 + }, + { + "epoch": 0.311750766052378, + "grad_norm": 0.06494140625, + "learning_rate": 0.0014638956254481215, + "loss": 0.0752, + "step": 35914 + }, + { + "epoch": 0.3117594465325822, + "grad_norm": 0.3984375, + "learning_rate": 0.0014638684040941643, + "loss": 0.1318, + "step": 35915 + }, + { + "epoch": 0.31176812701278633, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014638411823422858, + "loss": 0.0859, + "step": 35916 + }, + { + "epoch": 0.3117768074929905, + "grad_norm": 0.09375, + "learning_rate": 0.0014638139601925166, + "loss": 0.1099, + "step": 35917 + }, + { + "epoch": 0.31178548797319466, + "grad_norm": 1.2109375, + "learning_rate": 0.0014637867376448856, + "loss": 0.1797, + "step": 35918 + }, + { + "epoch": 0.31179416845339886, + "grad_norm": 0.2421875, + "learning_rate": 0.001463759514699423, + "loss": 0.1025, + "step": 35919 + }, + { + "epoch": 0.311802848933603, + "grad_norm": 0.62890625, + "learning_rate": 0.0014637322913561583, + "loss": 0.0845, + "step": 35920 + }, + { + "epoch": 0.3118115294138072, + "grad_norm": 0.419921875, + "learning_rate": 0.0014637050676151214, + "loss": 0.103, + "step": 35921 + }, + { + "epoch": 0.3118202098940113, + "grad_norm": 0.306640625, + "learning_rate": 0.0014636778434763424, + "loss": 0.1348, + "step": 35922 + }, + { + "epoch": 0.3118288903742155, + "grad_norm": 0.6015625, + "learning_rate": 0.0014636506189398505, + "loss": 0.1582, + "step": 35923 + }, + { + "epoch": 0.31183757085441965, + "grad_norm": 0.486328125, + "learning_rate": 0.0014636233940056758, + "loss": 0.0879, + "step": 35924 + }, + { + "epoch": 0.31184625133462385, + "grad_norm": 1.15625, + "learning_rate": 0.0014635961686738478, + "loss": 0.1206, + "step": 35925 + }, + { + "epoch": 0.311854931814828, + "grad_norm": 0.4375, + "learning_rate": 0.001463568942944397, + "loss": 0.1289, + "step": 35926 + }, + { + "epoch": 0.3118636122950322, + "grad_norm": 0.51953125, + "learning_rate": 0.0014635417168173525, + "loss": 0.1172, + "step": 35927 + }, + { + "epoch": 0.3118722927752363, + "grad_norm": 0.2099609375, + "learning_rate": 0.0014635144902927443, + "loss": 0.0962, + "step": 35928 + }, + { + "epoch": 0.3118809732554405, + "grad_norm": 0.0830078125, + "learning_rate": 0.0014634872633706019, + "loss": 0.0933, + "step": 35929 + }, + { + "epoch": 0.31188965373564465, + "grad_norm": 0.1015625, + "learning_rate": 0.0014634600360509554, + "loss": 0.1006, + "step": 35930 + }, + { + "epoch": 0.31189833421584884, + "grad_norm": 0.53125, + "learning_rate": 0.0014634328083338345, + "loss": 0.1279, + "step": 35931 + }, + { + "epoch": 0.311907014696053, + "grad_norm": 0.09423828125, + "learning_rate": 0.0014634055802192688, + "loss": 0.1167, + "step": 35932 + }, + { + "epoch": 0.31191569517625717, + "grad_norm": 0.296875, + "learning_rate": 0.001463378351707288, + "loss": 0.1211, + "step": 35933 + }, + { + "epoch": 0.3119243756564613, + "grad_norm": 0.1435546875, + "learning_rate": 0.0014633511227979223, + "loss": 0.0752, + "step": 35934 + }, + { + "epoch": 0.3119330561366655, + "grad_norm": 0.56640625, + "learning_rate": 0.0014633238934912017, + "loss": 0.1309, + "step": 35935 + }, + { + "epoch": 0.31194173661686964, + "grad_norm": 0.4765625, + "learning_rate": 0.0014632966637871552, + "loss": 0.0742, + "step": 35936 + }, + { + "epoch": 0.31195041709707383, + "grad_norm": 0.1044921875, + "learning_rate": 0.0014632694336858126, + "loss": 0.0791, + "step": 35937 + }, + { + "epoch": 0.31195909757727797, + "grad_norm": 0.271484375, + "learning_rate": 0.0014632422031872043, + "loss": 0.0835, + "step": 35938 + }, + { + "epoch": 0.31196777805748216, + "grad_norm": 0.103515625, + "learning_rate": 0.00146321497229136, + "loss": 0.1367, + "step": 35939 + }, + { + "epoch": 0.3119764585376863, + "grad_norm": 0.3671875, + "learning_rate": 0.001463187740998309, + "loss": 0.0972, + "step": 35940 + }, + { + "epoch": 0.3119851390178905, + "grad_norm": 0.19921875, + "learning_rate": 0.0014631605093080816, + "loss": 0.082, + "step": 35941 + }, + { + "epoch": 0.3119938194980946, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014631332772207068, + "loss": 0.0942, + "step": 35942 + }, + { + "epoch": 0.3120024999782988, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014631060447362154, + "loss": 0.1396, + "step": 35943 + }, + { + "epoch": 0.31201118045850296, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014630788118546365, + "loss": 0.1211, + "step": 35944 + }, + { + "epoch": 0.31201986093870715, + "grad_norm": 0.515625, + "learning_rate": 0.0014630515785760004, + "loss": 0.0869, + "step": 35945 + }, + { + "epoch": 0.3120285414189113, + "grad_norm": 0.1328125, + "learning_rate": 0.0014630243449003358, + "loss": 0.0781, + "step": 35946 + }, + { + "epoch": 0.3120372218991155, + "grad_norm": 0.1484375, + "learning_rate": 0.0014629971108276738, + "loss": 0.0986, + "step": 35947 + }, + { + "epoch": 0.3120459023793196, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014629698763580434, + "loss": 0.0913, + "step": 35948 + }, + { + "epoch": 0.3120545828595238, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014629426414914748, + "loss": 0.123, + "step": 35949 + }, + { + "epoch": 0.31206326333972795, + "grad_norm": 0.1328125, + "learning_rate": 0.0014629154062279974, + "loss": 0.0928, + "step": 35950 + }, + { + "epoch": 0.31207194381993214, + "grad_norm": 0.1201171875, + "learning_rate": 0.0014628881705676413, + "loss": 0.084, + "step": 35951 + }, + { + "epoch": 0.3120806243001363, + "grad_norm": 0.396484375, + "learning_rate": 0.0014628609345104361, + "loss": 0.0879, + "step": 35952 + }, + { + "epoch": 0.31208930478034047, + "grad_norm": 0.1943359375, + "learning_rate": 0.0014628336980564117, + "loss": 0.1064, + "step": 35953 + }, + { + "epoch": 0.3120979852605446, + "grad_norm": 0.3359375, + "learning_rate": 0.0014628064612055977, + "loss": 0.1123, + "step": 35954 + }, + { + "epoch": 0.3121066657407488, + "grad_norm": 0.5859375, + "learning_rate": 0.001462779223958024, + "loss": 0.0913, + "step": 35955 + }, + { + "epoch": 0.31211534622095294, + "grad_norm": 0.09423828125, + "learning_rate": 0.0014627519863137203, + "loss": 0.1025, + "step": 35956 + }, + { + "epoch": 0.31212402670115713, + "grad_norm": 0.078125, + "learning_rate": 0.0014627247482727169, + "loss": 0.0747, + "step": 35957 + }, + { + "epoch": 0.31213270718136127, + "grad_norm": 0.0869140625, + "learning_rate": 0.0014626975098350427, + "loss": 0.0874, + "step": 35958 + }, + { + "epoch": 0.31214138766156546, + "grad_norm": 0.33203125, + "learning_rate": 0.001462670271000728, + "loss": 0.1006, + "step": 35959 + }, + { + "epoch": 0.3121500681417696, + "grad_norm": 0.5390625, + "learning_rate": 0.0014626430317698027, + "loss": 0.1035, + "step": 35960 + }, + { + "epoch": 0.3121587486219738, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014626157921422964, + "loss": 0.083, + "step": 35961 + }, + { + "epoch": 0.31216742910217793, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014625885521182388, + "loss": 0.0938, + "step": 35962 + }, + { + "epoch": 0.3121761095823821, + "grad_norm": 0.09375, + "learning_rate": 0.00146256131169766, + "loss": 0.0908, + "step": 35963 + }, + { + "epoch": 0.31218479006258626, + "grad_norm": 0.65234375, + "learning_rate": 0.0014625340708805897, + "loss": 0.1406, + "step": 35964 + }, + { + "epoch": 0.31219347054279045, + "grad_norm": 0.26171875, + "learning_rate": 0.0014625068296670574, + "loss": 0.0996, + "step": 35965 + }, + { + "epoch": 0.3122021510229946, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014624795880570929, + "loss": 0.0977, + "step": 35966 + }, + { + "epoch": 0.3122108315031988, + "grad_norm": 0.267578125, + "learning_rate": 0.0014624523460507264, + "loss": 0.0869, + "step": 35967 + }, + { + "epoch": 0.3122195119834029, + "grad_norm": 0.12353515625, + "learning_rate": 0.0014624251036479874, + "loss": 0.0791, + "step": 35968 + }, + { + "epoch": 0.3122281924636071, + "grad_norm": 0.349609375, + "learning_rate": 0.0014623978608489059, + "loss": 0.0874, + "step": 35969 + }, + { + "epoch": 0.31223687294381125, + "grad_norm": 0.1328125, + "learning_rate": 0.0014623706176535112, + "loss": 0.085, + "step": 35970 + }, + { + "epoch": 0.31224555342401544, + "grad_norm": 0.291015625, + "learning_rate": 0.0014623433740618337, + "loss": 0.085, + "step": 35971 + }, + { + "epoch": 0.3122542339042196, + "grad_norm": 0.58984375, + "learning_rate": 0.0014623161300739028, + "loss": 0.1104, + "step": 35972 + }, + { + "epoch": 0.3122629143844238, + "grad_norm": 0.451171875, + "learning_rate": 0.0014622888856897485, + "loss": 0.1221, + "step": 35973 + }, + { + "epoch": 0.3122715948646279, + "grad_norm": 0.16015625, + "learning_rate": 0.0014622616409094007, + "loss": 0.0991, + "step": 35974 + }, + { + "epoch": 0.3122802753448321, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014622343957328887, + "loss": 0.1494, + "step": 35975 + }, + { + "epoch": 0.31228895582503624, + "grad_norm": 0.220703125, + "learning_rate": 0.0014622071501602428, + "loss": 0.0972, + "step": 35976 + }, + { + "epoch": 0.31229763630524043, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014621799041914926, + "loss": 0.1309, + "step": 35977 + }, + { + "epoch": 0.31230631678544457, + "grad_norm": 0.26953125, + "learning_rate": 0.0014621526578266675, + "loss": 0.1406, + "step": 35978 + }, + { + "epoch": 0.3123149972656487, + "grad_norm": 0.216796875, + "learning_rate": 0.0014621254110657983, + "loss": 0.0996, + "step": 35979 + }, + { + "epoch": 0.3123236777458529, + "grad_norm": 0.294921875, + "learning_rate": 0.0014620981639089138, + "loss": 0.1074, + "step": 35980 + }, + { + "epoch": 0.31233235822605704, + "grad_norm": 0.7578125, + "learning_rate": 0.0014620709163560444, + "loss": 0.0869, + "step": 35981 + }, + { + "epoch": 0.31234103870626123, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014620436684072199, + "loss": 0.1025, + "step": 35982 + }, + { + "epoch": 0.31234971918646537, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014620164200624692, + "loss": 0.0923, + "step": 35983 + }, + { + "epoch": 0.31235839966666956, + "grad_norm": 0.08935546875, + "learning_rate": 0.0014619891713218234, + "loss": 0.1182, + "step": 35984 + }, + { + "epoch": 0.3123670801468737, + "grad_norm": 0.1865234375, + "learning_rate": 0.0014619619221853113, + "loss": 0.1177, + "step": 35985 + }, + { + "epoch": 0.3123757606270779, + "grad_norm": 1.0390625, + "learning_rate": 0.0014619346726529635, + "loss": 0.1079, + "step": 35986 + }, + { + "epoch": 0.31238444110728203, + "grad_norm": 0.216796875, + "learning_rate": 0.001461907422724809, + "loss": 0.1079, + "step": 35987 + }, + { + "epoch": 0.3123931215874862, + "grad_norm": 0.328125, + "learning_rate": 0.001461880172400878, + "loss": 0.084, + "step": 35988 + }, + { + "epoch": 0.31240180206769036, + "grad_norm": 0.1103515625, + "learning_rate": 0.0014618529216812001, + "loss": 0.0972, + "step": 35989 + }, + { + "epoch": 0.31241048254789455, + "grad_norm": 0.361328125, + "learning_rate": 0.0014618256705658055, + "loss": 0.1338, + "step": 35990 + }, + { + "epoch": 0.3124191630280987, + "grad_norm": 0.1318359375, + "learning_rate": 0.001461798419054724, + "loss": 0.1079, + "step": 35991 + }, + { + "epoch": 0.3124278435083029, + "grad_norm": 0.12451171875, + "learning_rate": 0.0014617711671479853, + "loss": 0.0996, + "step": 35992 + }, + { + "epoch": 0.312436523988507, + "grad_norm": 0.1279296875, + "learning_rate": 0.0014617439148456187, + "loss": 0.1367, + "step": 35993 + }, + { + "epoch": 0.3124452044687112, + "grad_norm": 0.2109375, + "learning_rate": 0.0014617166621476543, + "loss": 0.0879, + "step": 35994 + }, + { + "epoch": 0.31245388494891535, + "grad_norm": 0.365234375, + "learning_rate": 0.0014616894090541223, + "loss": 0.1108, + "step": 35995 + }, + { + "epoch": 0.31246256542911954, + "grad_norm": 0.37109375, + "learning_rate": 0.001461662155565052, + "loss": 0.0688, + "step": 35996 + }, + { + "epoch": 0.3124712459093237, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014616349016804736, + "loss": 0.1094, + "step": 35997 + }, + { + "epoch": 0.3124799263895279, + "grad_norm": 0.349609375, + "learning_rate": 0.0014616076474004162, + "loss": 0.0986, + "step": 35998 + }, + { + "epoch": 0.312488606869732, + "grad_norm": 0.09033203125, + "learning_rate": 0.0014615803927249106, + "loss": 0.1016, + "step": 35999 + }, + { + "epoch": 0.3124972873499362, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014615531376539863, + "loss": 0.0996, + "step": 36000 + }, + { + "epoch": 0.31250596783014034, + "grad_norm": 0.224609375, + "learning_rate": 0.0014615258821876727, + "loss": 0.123, + "step": 36001 + }, + { + "epoch": 0.31251464831034453, + "grad_norm": 0.267578125, + "learning_rate": 0.0014614986263259996, + "loss": 0.1162, + "step": 36002 + }, + { + "epoch": 0.31252332879054867, + "grad_norm": 0.09228515625, + "learning_rate": 0.0014614713700689971, + "loss": 0.126, + "step": 36003 + }, + { + "epoch": 0.31253200927075286, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014614441134166952, + "loss": 0.1094, + "step": 36004 + }, + { + "epoch": 0.312540689750957, + "grad_norm": 0.283203125, + "learning_rate": 0.0014614168563691233, + "loss": 0.0845, + "step": 36005 + }, + { + "epoch": 0.3125493702311612, + "grad_norm": 0.251953125, + "learning_rate": 0.0014613895989263113, + "loss": 0.0762, + "step": 36006 + }, + { + "epoch": 0.31255805071136533, + "grad_norm": 0.46875, + "learning_rate": 0.001461362341088289, + "loss": 0.0898, + "step": 36007 + }, + { + "epoch": 0.3125667311915695, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014613350828550863, + "loss": 0.0986, + "step": 36008 + }, + { + "epoch": 0.31257541167177366, + "grad_norm": 0.09326171875, + "learning_rate": 0.0014613078242267333, + "loss": 0.1191, + "step": 36009 + }, + { + "epoch": 0.31258409215197785, + "grad_norm": 0.416015625, + "learning_rate": 0.0014612805652032593, + "loss": 0.0732, + "step": 36010 + }, + { + "epoch": 0.312592772632182, + "grad_norm": 0.240234375, + "learning_rate": 0.0014612533057846943, + "loss": 0.1143, + "step": 36011 + }, + { + "epoch": 0.3126014531123862, + "grad_norm": 0.5, + "learning_rate": 0.001461226045971068, + "loss": 0.1309, + "step": 36012 + }, + { + "epoch": 0.3126101335925903, + "grad_norm": 0.05908203125, + "learning_rate": 0.0014611987857624104, + "loss": 0.0583, + "step": 36013 + }, + { + "epoch": 0.3126188140727945, + "grad_norm": 0.8125, + "learning_rate": 0.0014611715251587514, + "loss": 0.0786, + "step": 36014 + }, + { + "epoch": 0.31262749455299865, + "grad_norm": 0.294921875, + "learning_rate": 0.0014611442641601202, + "loss": 0.1406, + "step": 36015 + }, + { + "epoch": 0.31263617503320285, + "grad_norm": 0.62109375, + "learning_rate": 0.0014611170027665475, + "loss": 0.082, + "step": 36016 + }, + { + "epoch": 0.312644855513407, + "grad_norm": 0.28125, + "learning_rate": 0.0014610897409780622, + "loss": 0.0835, + "step": 36017 + }, + { + "epoch": 0.3126535359936112, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014610624787946948, + "loss": 0.0986, + "step": 36018 + }, + { + "epoch": 0.3126622164738153, + "grad_norm": 0.55078125, + "learning_rate": 0.0014610352162164751, + "loss": 0.1187, + "step": 36019 + }, + { + "epoch": 0.3126708969540195, + "grad_norm": 0.1767578125, + "learning_rate": 0.0014610079532434328, + "loss": 0.1006, + "step": 36020 + }, + { + "epoch": 0.31267957743422364, + "grad_norm": 0.1142578125, + "learning_rate": 0.0014609806898755976, + "loss": 0.1328, + "step": 36021 + }, + { + "epoch": 0.31268825791442784, + "grad_norm": 0.28125, + "learning_rate": 0.001460953426112999, + "loss": 0.0898, + "step": 36022 + }, + { + "epoch": 0.312696938394632, + "grad_norm": 0.1171875, + "learning_rate": 0.0014609261619556674, + "loss": 0.123, + "step": 36023 + }, + { + "epoch": 0.31270561887483617, + "grad_norm": 0.1494140625, + "learning_rate": 0.001460898897403632, + "loss": 0.103, + "step": 36024 + }, + { + "epoch": 0.3127142993550403, + "grad_norm": 0.294921875, + "learning_rate": 0.0014608716324569235, + "loss": 0.0977, + "step": 36025 + }, + { + "epoch": 0.3127229798352445, + "grad_norm": 0.201171875, + "learning_rate": 0.0014608443671155708, + "loss": 0.1055, + "step": 36026 + }, + { + "epoch": 0.31273166031544863, + "grad_norm": 0.3203125, + "learning_rate": 0.0014608171013796043, + "loss": 0.1768, + "step": 36027 + }, + { + "epoch": 0.3127403407956528, + "grad_norm": 0.2099609375, + "learning_rate": 0.0014607898352490536, + "loss": 0.082, + "step": 36028 + }, + { + "epoch": 0.31274902127585696, + "grad_norm": 0.40234375, + "learning_rate": 0.0014607625687239488, + "loss": 0.1123, + "step": 36029 + }, + { + "epoch": 0.31275770175606116, + "grad_norm": 0.26171875, + "learning_rate": 0.001460735301804319, + "loss": 0.0898, + "step": 36030 + }, + { + "epoch": 0.3127663822362653, + "grad_norm": 0.50390625, + "learning_rate": 0.001460708034490195, + "loss": 0.0977, + "step": 36031 + }, + { + "epoch": 0.3127750627164695, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014606807667816059, + "loss": 0.1035, + "step": 36032 + }, + { + "epoch": 0.3127837431966736, + "grad_norm": 0.314453125, + "learning_rate": 0.0014606534986785816, + "loss": 0.1299, + "step": 36033 + }, + { + "epoch": 0.3127924236768778, + "grad_norm": 0.1796875, + "learning_rate": 0.0014606262301811522, + "loss": 0.1035, + "step": 36034 + }, + { + "epoch": 0.31280110415708196, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014605989612893476, + "loss": 0.1162, + "step": 36035 + }, + { + "epoch": 0.31280978463728615, + "grad_norm": 0.6484375, + "learning_rate": 0.0014605716920031969, + "loss": 0.0859, + "step": 36036 + }, + { + "epoch": 0.3128184651174903, + "grad_norm": 0.1630859375, + "learning_rate": 0.0014605444223227304, + "loss": 0.0996, + "step": 36037 + }, + { + "epoch": 0.3128271455976945, + "grad_norm": 1.0703125, + "learning_rate": 0.0014605171522479783, + "loss": 0.1006, + "step": 36038 + }, + { + "epoch": 0.3128358260778986, + "grad_norm": 0.49609375, + "learning_rate": 0.0014604898817789703, + "loss": 0.1357, + "step": 36039 + }, + { + "epoch": 0.3128445065581028, + "grad_norm": 0.11083984375, + "learning_rate": 0.0014604626109157356, + "loss": 0.0996, + "step": 36040 + }, + { + "epoch": 0.31285318703830695, + "grad_norm": 0.062255859375, + "learning_rate": 0.0014604353396583042, + "loss": 0.0908, + "step": 36041 + }, + { + "epoch": 0.31286186751851114, + "grad_norm": 0.61328125, + "learning_rate": 0.0014604080680067064, + "loss": 0.1523, + "step": 36042 + }, + { + "epoch": 0.3128705479987153, + "grad_norm": 0.75390625, + "learning_rate": 0.0014603807959609719, + "loss": 0.1035, + "step": 36043 + }, + { + "epoch": 0.31287922847891947, + "grad_norm": 0.1484375, + "learning_rate": 0.0014603535235211303, + "loss": 0.0654, + "step": 36044 + }, + { + "epoch": 0.3128879089591236, + "grad_norm": 0.6328125, + "learning_rate": 0.001460326250687211, + "loss": 0.0918, + "step": 36045 + }, + { + "epoch": 0.3128965894393278, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014602989774592445, + "loss": 0.1211, + "step": 36046 + }, + { + "epoch": 0.31290526991953194, + "grad_norm": 0.5703125, + "learning_rate": 0.001460271703837261, + "loss": 0.0996, + "step": 36047 + }, + { + "epoch": 0.31291395039973613, + "grad_norm": 0.0810546875, + "learning_rate": 0.0014602444298212895, + "loss": 0.1016, + "step": 36048 + }, + { + "epoch": 0.31292263087994027, + "grad_norm": 1.25, + "learning_rate": 0.0014602171554113602, + "loss": 0.1582, + "step": 36049 + }, + { + "epoch": 0.31293131136014446, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014601898806075025, + "loss": 0.064, + "step": 36050 + }, + { + "epoch": 0.3129399918403486, + "grad_norm": 0.46484375, + "learning_rate": 0.0014601626054097467, + "loss": 0.1016, + "step": 36051 + }, + { + "epoch": 0.3129486723205528, + "grad_norm": 0.1943359375, + "learning_rate": 0.0014601353298181227, + "loss": 0.0815, + "step": 36052 + }, + { + "epoch": 0.3129573528007569, + "grad_norm": 0.3359375, + "learning_rate": 0.0014601080538326598, + "loss": 0.1338, + "step": 36053 + }, + { + "epoch": 0.3129660332809611, + "grad_norm": 0.095703125, + "learning_rate": 0.0014600807774533884, + "loss": 0.1133, + "step": 36054 + }, + { + "epoch": 0.31297471376116526, + "grad_norm": 0.1806640625, + "learning_rate": 0.001460053500680338, + "loss": 0.1152, + "step": 36055 + }, + { + "epoch": 0.31298339424136945, + "grad_norm": 0.12109375, + "learning_rate": 0.001460026223513538, + "loss": 0.1123, + "step": 36056 + }, + { + "epoch": 0.3129920747215736, + "grad_norm": 0.203125, + "learning_rate": 0.0014599989459530194, + "loss": 0.1055, + "step": 36057 + }, + { + "epoch": 0.3130007552017778, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014599716679988112, + "loss": 0.1621, + "step": 36058 + }, + { + "epoch": 0.3130094356819819, + "grad_norm": 0.1865234375, + "learning_rate": 0.0014599443896509434, + "loss": 0.0771, + "step": 36059 + }, + { + "epoch": 0.3130181161621861, + "grad_norm": 0.216796875, + "learning_rate": 0.0014599171109094457, + "loss": 0.1216, + "step": 36060 + }, + { + "epoch": 0.31302679664239025, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014598898317743482, + "loss": 0.1406, + "step": 36061 + }, + { + "epoch": 0.31303547712259444, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014598625522456805, + "loss": 0.1201, + "step": 36062 + }, + { + "epoch": 0.3130441576027986, + "grad_norm": 0.142578125, + "learning_rate": 0.0014598352723234726, + "loss": 0.0952, + "step": 36063 + }, + { + "epoch": 0.31305283808300277, + "grad_norm": 0.2138671875, + "learning_rate": 0.0014598079920077542, + "loss": 0.125, + "step": 36064 + }, + { + "epoch": 0.3130615185632069, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014597807112985551, + "loss": 0.123, + "step": 36065 + }, + { + "epoch": 0.3130701990434111, + "grad_norm": 0.259765625, + "learning_rate": 0.0014597534301959052, + "loss": 0.1128, + "step": 36066 + }, + { + "epoch": 0.31307887952361524, + "grad_norm": 0.23046875, + "learning_rate": 0.0014597261486998343, + "loss": 0.1562, + "step": 36067 + }, + { + "epoch": 0.31308756000381943, + "grad_norm": 0.10205078125, + "learning_rate": 0.0014596988668103726, + "loss": 0.0918, + "step": 36068 + }, + { + "epoch": 0.31309624048402357, + "grad_norm": 0.0908203125, + "learning_rate": 0.0014596715845275495, + "loss": 0.0947, + "step": 36069 + }, + { + "epoch": 0.31310492096422776, + "grad_norm": 0.15234375, + "learning_rate": 0.0014596443018513952, + "loss": 0.1187, + "step": 36070 + }, + { + "epoch": 0.3131136014444319, + "grad_norm": 0.6953125, + "learning_rate": 0.0014596170187819385, + "loss": 0.0938, + "step": 36071 + }, + { + "epoch": 0.3131222819246361, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014595897353192108, + "loss": 0.0996, + "step": 36072 + }, + { + "epoch": 0.31313096240484023, + "grad_norm": 0.54296875, + "learning_rate": 0.0014595624514632411, + "loss": 0.0928, + "step": 36073 + }, + { + "epoch": 0.3131396428850444, + "grad_norm": 0.6015625, + "learning_rate": 0.0014595351672140589, + "loss": 0.1094, + "step": 36074 + }, + { + "epoch": 0.31314832336524856, + "grad_norm": 0.3046875, + "learning_rate": 0.0014595078825716946, + "loss": 0.1108, + "step": 36075 + }, + { + "epoch": 0.31315700384545275, + "grad_norm": 0.396484375, + "learning_rate": 0.0014594805975361779, + "loss": 0.1035, + "step": 36076 + }, + { + "epoch": 0.3131656843256569, + "grad_norm": 0.1455078125, + "learning_rate": 0.001459453312107539, + "loss": 0.0908, + "step": 36077 + }, + { + "epoch": 0.3131743648058611, + "grad_norm": 0.244140625, + "learning_rate": 0.001459426026285807, + "loss": 0.1221, + "step": 36078 + }, + { + "epoch": 0.3131830452860652, + "grad_norm": 0.890625, + "learning_rate": 0.001459398740071012, + "loss": 0.1016, + "step": 36079 + }, + { + "epoch": 0.3131917257662694, + "grad_norm": 0.18359375, + "learning_rate": 0.0014593714534631842, + "loss": 0.0898, + "step": 36080 + }, + { + "epoch": 0.31320040624647355, + "grad_norm": 0.185546875, + "learning_rate": 0.0014593441664623528, + "loss": 0.105, + "step": 36081 + }, + { + "epoch": 0.31320908672667774, + "grad_norm": 0.232421875, + "learning_rate": 0.0014593168790685482, + "loss": 0.0898, + "step": 36082 + }, + { + "epoch": 0.3132177672068819, + "grad_norm": 0.396484375, + "learning_rate": 0.0014592895912818001, + "loss": 0.123, + "step": 36083 + }, + { + "epoch": 0.3132264476870861, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014592623031021387, + "loss": 0.1055, + "step": 36084 + }, + { + "epoch": 0.3132351281672902, + "grad_norm": 0.9765625, + "learning_rate": 0.0014592350145295927, + "loss": 0.0942, + "step": 36085 + }, + { + "epoch": 0.3132438086474944, + "grad_norm": 0.1669921875, + "learning_rate": 0.001459207725564193, + "loss": 0.0923, + "step": 36086 + }, + { + "epoch": 0.31325248912769854, + "grad_norm": 0.220703125, + "learning_rate": 0.0014591804362059694, + "loss": 0.0674, + "step": 36087 + }, + { + "epoch": 0.31326116960790273, + "grad_norm": 0.255859375, + "learning_rate": 0.001459153146454951, + "loss": 0.0713, + "step": 36088 + }, + { + "epoch": 0.31326985008810687, + "grad_norm": 0.609375, + "learning_rate": 0.0014591258563111684, + "loss": 0.1074, + "step": 36089 + }, + { + "epoch": 0.31327853056831106, + "grad_norm": 0.34375, + "learning_rate": 0.0014590985657746511, + "loss": 0.127, + "step": 36090 + }, + { + "epoch": 0.3132872110485152, + "grad_norm": 0.408203125, + "learning_rate": 0.0014590712748454291, + "loss": 0.1133, + "step": 36091 + }, + { + "epoch": 0.3132958915287194, + "grad_norm": 0.3046875, + "learning_rate": 0.0014590439835235323, + "loss": 0.0801, + "step": 36092 + }, + { + "epoch": 0.31330457200892353, + "grad_norm": 0.1865234375, + "learning_rate": 0.00145901669180899, + "loss": 0.0962, + "step": 36093 + }, + { + "epoch": 0.3133132524891277, + "grad_norm": 0.263671875, + "learning_rate": 0.0014589893997018323, + "loss": 0.0825, + "step": 36094 + }, + { + "epoch": 0.31332193296933186, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014589621072020895, + "loss": 0.103, + "step": 36095 + }, + { + "epoch": 0.31333061344953606, + "grad_norm": 0.341796875, + "learning_rate": 0.0014589348143097913, + "loss": 0.0757, + "step": 36096 + }, + { + "epoch": 0.3133392939297402, + "grad_norm": 0.349609375, + "learning_rate": 0.001458907521024967, + "loss": 0.1006, + "step": 36097 + }, + { + "epoch": 0.3133479744099444, + "grad_norm": 0.42578125, + "learning_rate": 0.001458880227347647, + "loss": 0.1035, + "step": 36098 + }, + { + "epoch": 0.3133566548901485, + "grad_norm": 0.353515625, + "learning_rate": 0.0014588529332778613, + "loss": 0.0762, + "step": 36099 + }, + { + "epoch": 0.3133653353703527, + "grad_norm": 0.2578125, + "learning_rate": 0.001458825638815639, + "loss": 0.1445, + "step": 36100 + }, + { + "epoch": 0.31337401585055685, + "grad_norm": 0.23828125, + "learning_rate": 0.0014587983439610106, + "loss": 0.1426, + "step": 36101 + }, + { + "epoch": 0.313382696330761, + "grad_norm": 0.13671875, + "learning_rate": 0.0014587710487140055, + "loss": 0.082, + "step": 36102 + }, + { + "epoch": 0.3133913768109652, + "grad_norm": 0.58203125, + "learning_rate": 0.0014587437530746538, + "loss": 0.1465, + "step": 36103 + }, + { + "epoch": 0.3134000572911693, + "grad_norm": 0.498046875, + "learning_rate": 0.0014587164570429856, + "loss": 0.1299, + "step": 36104 + }, + { + "epoch": 0.3134087377713735, + "grad_norm": 0.466796875, + "learning_rate": 0.0014586891606190302, + "loss": 0.0801, + "step": 36105 + }, + { + "epoch": 0.31341741825157765, + "grad_norm": 0.3203125, + "learning_rate": 0.0014586618638028177, + "loss": 0.1035, + "step": 36106 + }, + { + "epoch": 0.31342609873178184, + "grad_norm": 0.1640625, + "learning_rate": 0.001458634566594378, + "loss": 0.0967, + "step": 36107 + }, + { + "epoch": 0.313434779211986, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014586072689937413, + "loss": 0.1133, + "step": 36108 + }, + { + "epoch": 0.3134434596921902, + "grad_norm": 0.171875, + "learning_rate": 0.0014585799710009368, + "loss": 0.1162, + "step": 36109 + }, + { + "epoch": 0.3134521401723943, + "grad_norm": 1.0625, + "learning_rate": 0.0014585526726159945, + "loss": 0.2441, + "step": 36110 + }, + { + "epoch": 0.3134608206525985, + "grad_norm": 0.1875, + "learning_rate": 0.0014585253738389446, + "loss": 0.0771, + "step": 36111 + }, + { + "epoch": 0.31346950113280264, + "grad_norm": 0.69140625, + "learning_rate": 0.001458498074669817, + "loss": 0.1113, + "step": 36112 + }, + { + "epoch": 0.31347818161300683, + "grad_norm": 0.83203125, + "learning_rate": 0.0014584707751086406, + "loss": 0.1318, + "step": 36113 + }, + { + "epoch": 0.31348686209321097, + "grad_norm": 1.265625, + "learning_rate": 0.001458443475155446, + "loss": 0.124, + "step": 36114 + }, + { + "epoch": 0.31349554257341516, + "grad_norm": 0.2890625, + "learning_rate": 0.0014584161748102634, + "loss": 0.1562, + "step": 36115 + }, + { + "epoch": 0.3135042230536193, + "grad_norm": 0.197265625, + "learning_rate": 0.0014583888740731222, + "loss": 0.1055, + "step": 36116 + }, + { + "epoch": 0.3135129035338235, + "grad_norm": 0.416015625, + "learning_rate": 0.0014583615729440525, + "loss": 0.0938, + "step": 36117 + }, + { + "epoch": 0.31352158401402763, + "grad_norm": 0.244140625, + "learning_rate": 0.0014583342714230836, + "loss": 0.0913, + "step": 36118 + }, + { + "epoch": 0.3135302644942318, + "grad_norm": 0.6171875, + "learning_rate": 0.0014583069695102457, + "loss": 0.1533, + "step": 36119 + }, + { + "epoch": 0.31353894497443596, + "grad_norm": 0.08642578125, + "learning_rate": 0.001458279667205569, + "loss": 0.0962, + "step": 36120 + }, + { + "epoch": 0.31354762545464016, + "grad_norm": 1.046875, + "learning_rate": 0.001458252364509083, + "loss": 0.168, + "step": 36121 + }, + { + "epoch": 0.3135563059348443, + "grad_norm": 0.345703125, + "learning_rate": 0.0014582250614208173, + "loss": 0.1475, + "step": 36122 + }, + { + "epoch": 0.3135649864150485, + "grad_norm": 0.09130859375, + "learning_rate": 0.0014581977579408022, + "loss": 0.0903, + "step": 36123 + }, + { + "epoch": 0.3135736668952526, + "grad_norm": 0.169921875, + "learning_rate": 0.0014581704540690672, + "loss": 0.1055, + "step": 36124 + }, + { + "epoch": 0.3135823473754568, + "grad_norm": 0.0654296875, + "learning_rate": 0.0014581431498056428, + "loss": 0.085, + "step": 36125 + }, + { + "epoch": 0.31359102785566095, + "grad_norm": 0.140625, + "learning_rate": 0.0014581158451505583, + "loss": 0.1123, + "step": 36126 + }, + { + "epoch": 0.31359970833586515, + "grad_norm": 0.1201171875, + "learning_rate": 0.0014580885401038436, + "loss": 0.0986, + "step": 36127 + }, + { + "epoch": 0.3136083888160693, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014580612346655286, + "loss": 0.1738, + "step": 36128 + }, + { + "epoch": 0.3136170692962735, + "grad_norm": 0.9375, + "learning_rate": 0.0014580339288356435, + "loss": 0.1016, + "step": 36129 + }, + { + "epoch": 0.3136257497764776, + "grad_norm": 0.302734375, + "learning_rate": 0.0014580066226142177, + "loss": 0.105, + "step": 36130 + }, + { + "epoch": 0.3136344302566818, + "grad_norm": 0.498046875, + "learning_rate": 0.001457979316001281, + "loss": 0.1104, + "step": 36131 + }, + { + "epoch": 0.31364311073688594, + "grad_norm": 0.1064453125, + "learning_rate": 0.0014579520089968638, + "loss": 0.1167, + "step": 36132 + }, + { + "epoch": 0.31365179121709014, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014579247016009954, + "loss": 0.0986, + "step": 36133 + }, + { + "epoch": 0.3136604716972943, + "grad_norm": 0.154296875, + "learning_rate": 0.0014578973938137061, + "loss": 0.0977, + "step": 36134 + }, + { + "epoch": 0.31366915217749847, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014578700856350255, + "loss": 0.1123, + "step": 36135 + }, + { + "epoch": 0.3136778326577026, + "grad_norm": 0.341796875, + "learning_rate": 0.0014578427770649837, + "loss": 0.082, + "step": 36136 + }, + { + "epoch": 0.3136865131379068, + "grad_norm": 0.65234375, + "learning_rate": 0.0014578154681036105, + "loss": 0.1094, + "step": 36137 + }, + { + "epoch": 0.31369519361811093, + "grad_norm": 0.4609375, + "learning_rate": 0.0014577881587509354, + "loss": 0.1089, + "step": 36138 + }, + { + "epoch": 0.3137038740983151, + "grad_norm": 0.109375, + "learning_rate": 0.0014577608490069886, + "loss": 0.1104, + "step": 36139 + }, + { + "epoch": 0.31371255457851926, + "grad_norm": 0.1142578125, + "learning_rate": 0.0014577335388718, + "loss": 0.1084, + "step": 36140 + }, + { + "epoch": 0.31372123505872346, + "grad_norm": 0.314453125, + "learning_rate": 0.0014577062283453995, + "loss": 0.0889, + "step": 36141 + }, + { + "epoch": 0.3137299155389276, + "grad_norm": 0.423828125, + "learning_rate": 0.0014576789174278167, + "loss": 0.1055, + "step": 36142 + }, + { + "epoch": 0.3137385960191318, + "grad_norm": 0.365234375, + "learning_rate": 0.0014576516061190815, + "loss": 0.1074, + "step": 36143 + }, + { + "epoch": 0.3137472764993359, + "grad_norm": 0.2119140625, + "learning_rate": 0.0014576242944192243, + "loss": 0.0879, + "step": 36144 + }, + { + "epoch": 0.3137559569795401, + "grad_norm": 0.197265625, + "learning_rate": 0.001457596982328274, + "loss": 0.1177, + "step": 36145 + }, + { + "epoch": 0.31376463745974426, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014575696698462613, + "loss": 0.0889, + "step": 36146 + }, + { + "epoch": 0.31377331793994845, + "grad_norm": 0.419921875, + "learning_rate": 0.0014575423569732158, + "loss": 0.1328, + "step": 36147 + }, + { + "epoch": 0.3137819984201526, + "grad_norm": 0.25390625, + "learning_rate": 0.0014575150437091673, + "loss": 0.1113, + "step": 36148 + }, + { + "epoch": 0.3137906789003568, + "grad_norm": 0.3125, + "learning_rate": 0.0014574877300541455, + "loss": 0.1152, + "step": 36149 + }, + { + "epoch": 0.3137993593805609, + "grad_norm": 0.1142578125, + "learning_rate": 0.0014574604160081809, + "loss": 0.0771, + "step": 36150 + }, + { + "epoch": 0.3138080398607651, + "grad_norm": 0.33984375, + "learning_rate": 0.0014574331015713025, + "loss": 0.1074, + "step": 36151 + }, + { + "epoch": 0.31381672034096925, + "grad_norm": 0.322265625, + "learning_rate": 0.0014574057867435411, + "loss": 0.1357, + "step": 36152 + }, + { + "epoch": 0.31382540082117344, + "grad_norm": 0.40234375, + "learning_rate": 0.0014573784715249257, + "loss": 0.1001, + "step": 36153 + }, + { + "epoch": 0.3138340813013776, + "grad_norm": 0.359375, + "learning_rate": 0.0014573511559154866, + "loss": 0.1191, + "step": 36154 + }, + { + "epoch": 0.31384276178158177, + "grad_norm": 0.0888671875, + "learning_rate": 0.0014573238399152543, + "loss": 0.0898, + "step": 36155 + }, + { + "epoch": 0.3138514422617859, + "grad_norm": 0.298828125, + "learning_rate": 0.0014572965235242574, + "loss": 0.0801, + "step": 36156 + }, + { + "epoch": 0.3138601227419901, + "grad_norm": 0.353515625, + "learning_rate": 0.0014572692067425263, + "loss": 0.1084, + "step": 36157 + }, + { + "epoch": 0.31386880322219424, + "grad_norm": 0.39453125, + "learning_rate": 0.0014572418895700912, + "loss": 0.0884, + "step": 36158 + }, + { + "epoch": 0.31387748370239843, + "grad_norm": 0.443359375, + "learning_rate": 0.0014572145720069818, + "loss": 0.0913, + "step": 36159 + }, + { + "epoch": 0.31388616418260257, + "grad_norm": 0.66796875, + "learning_rate": 0.001457187254053228, + "loss": 0.1699, + "step": 36160 + }, + { + "epoch": 0.31389484466280676, + "grad_norm": 0.259765625, + "learning_rate": 0.0014571599357088591, + "loss": 0.0923, + "step": 36161 + }, + { + "epoch": 0.3139035251430109, + "grad_norm": 0.2490234375, + "learning_rate": 0.0014571326169739056, + "loss": 0.1094, + "step": 36162 + }, + { + "epoch": 0.3139122056232151, + "grad_norm": 0.5234375, + "learning_rate": 0.0014571052978483974, + "loss": 0.1094, + "step": 36163 + }, + { + "epoch": 0.3139208861034192, + "grad_norm": 0.10302734375, + "learning_rate": 0.0014570779783323643, + "loss": 0.0869, + "step": 36164 + }, + { + "epoch": 0.3139295665836234, + "grad_norm": 0.404296875, + "learning_rate": 0.001457050658425836, + "loss": 0.1143, + "step": 36165 + }, + { + "epoch": 0.31393824706382756, + "grad_norm": 0.10546875, + "learning_rate": 0.0014570233381288424, + "loss": 0.0923, + "step": 36166 + }, + { + "epoch": 0.31394692754403175, + "grad_norm": 0.38671875, + "learning_rate": 0.0014569960174414139, + "loss": 0.0776, + "step": 36167 + }, + { + "epoch": 0.3139556080242359, + "grad_norm": 0.2890625, + "learning_rate": 0.0014569686963635793, + "loss": 0.1182, + "step": 36168 + }, + { + "epoch": 0.3139642885044401, + "grad_norm": 0.5078125, + "learning_rate": 0.0014569413748953695, + "loss": 0.1387, + "step": 36169 + }, + { + "epoch": 0.3139729689846442, + "grad_norm": 0.1064453125, + "learning_rate": 0.0014569140530368138, + "loss": 0.1299, + "step": 36170 + }, + { + "epoch": 0.3139816494648484, + "grad_norm": 0.2177734375, + "learning_rate": 0.001456886730787942, + "loss": 0.0908, + "step": 36171 + }, + { + "epoch": 0.31399032994505255, + "grad_norm": 0.11328125, + "learning_rate": 0.0014568594081487845, + "loss": 0.1025, + "step": 36172 + }, + { + "epoch": 0.31399901042525674, + "grad_norm": 0.37890625, + "learning_rate": 0.0014568320851193709, + "loss": 0.1162, + "step": 36173 + }, + { + "epoch": 0.3140076909054609, + "grad_norm": 0.390625, + "learning_rate": 0.0014568047616997314, + "loss": 0.082, + "step": 36174 + }, + { + "epoch": 0.31401637138566507, + "grad_norm": 0.1162109375, + "learning_rate": 0.001456777437889895, + "loss": 0.0752, + "step": 36175 + }, + { + "epoch": 0.3140250518658692, + "grad_norm": 0.578125, + "learning_rate": 0.0014567501136898927, + "loss": 0.0908, + "step": 36176 + }, + { + "epoch": 0.3140337323460734, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014567227890997533, + "loss": 0.1123, + "step": 36177 + }, + { + "epoch": 0.31404241282627754, + "grad_norm": 0.359375, + "learning_rate": 0.0014566954641195076, + "loss": 0.0947, + "step": 36178 + }, + { + "epoch": 0.31405109330648173, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014566681387491848, + "loss": 0.0928, + "step": 36179 + }, + { + "epoch": 0.31405977378668587, + "grad_norm": 0.1240234375, + "learning_rate": 0.0014566408129888153, + "loss": 0.1016, + "step": 36180 + }, + { + "epoch": 0.31406845426689006, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014566134868384286, + "loss": 0.0918, + "step": 36181 + }, + { + "epoch": 0.3140771347470942, + "grad_norm": 0.1552734375, + "learning_rate": 0.001456586160298055, + "loss": 0.0757, + "step": 36182 + }, + { + "epoch": 0.3140858152272984, + "grad_norm": 0.4140625, + "learning_rate": 0.001456558833367724, + "loss": 0.0869, + "step": 36183 + }, + { + "epoch": 0.31409449570750253, + "grad_norm": 0.89453125, + "learning_rate": 0.0014565315060474655, + "loss": 0.0898, + "step": 36184 + }, + { + "epoch": 0.3141031761877067, + "grad_norm": 0.11376953125, + "learning_rate": 0.0014565041783373097, + "loss": 0.0854, + "step": 36185 + }, + { + "epoch": 0.31411185666791086, + "grad_norm": 0.1171875, + "learning_rate": 0.0014564768502372863, + "loss": 0.1104, + "step": 36186 + }, + { + "epoch": 0.31412053714811505, + "grad_norm": 0.390625, + "learning_rate": 0.0014564495217474253, + "loss": 0.0713, + "step": 36187 + }, + { + "epoch": 0.3141292176283192, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014564221928677558, + "loss": 0.0942, + "step": 36188 + }, + { + "epoch": 0.3141378981085234, + "grad_norm": 0.416015625, + "learning_rate": 0.001456394863598309, + "loss": 0.0913, + "step": 36189 + }, + { + "epoch": 0.3141465785887275, + "grad_norm": 0.3125, + "learning_rate": 0.0014563675339391137, + "loss": 0.1201, + "step": 36190 + }, + { + "epoch": 0.3141552590689317, + "grad_norm": 0.1806640625, + "learning_rate": 0.0014563402038902007, + "loss": 0.085, + "step": 36191 + }, + { + "epoch": 0.31416393954913585, + "grad_norm": 0.458984375, + "learning_rate": 0.001456312873451599, + "loss": 0.1182, + "step": 36192 + }, + { + "epoch": 0.31417262002934004, + "grad_norm": 0.08203125, + "learning_rate": 0.001456285542623339, + "loss": 0.0742, + "step": 36193 + }, + { + "epoch": 0.3141813005095442, + "grad_norm": 0.439453125, + "learning_rate": 0.0014562582114054506, + "loss": 0.0742, + "step": 36194 + }, + { + "epoch": 0.3141899809897484, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014562308797979636, + "loss": 0.1182, + "step": 36195 + }, + { + "epoch": 0.3141986614699525, + "grad_norm": 1.3046875, + "learning_rate": 0.0014562035478009076, + "loss": 0.1387, + "step": 36196 + }, + { + "epoch": 0.3142073419501567, + "grad_norm": 0.2890625, + "learning_rate": 0.0014561762154143127, + "loss": 0.1113, + "step": 36197 + }, + { + "epoch": 0.31421602243036084, + "grad_norm": 0.130859375, + "learning_rate": 0.0014561488826382095, + "loss": 0.0947, + "step": 36198 + }, + { + "epoch": 0.31422470291056503, + "grad_norm": 0.55859375, + "learning_rate": 0.0014561215494726266, + "loss": 0.1621, + "step": 36199 + }, + { + "epoch": 0.31423338339076917, + "grad_norm": 0.2392578125, + "learning_rate": 0.0014560942159175948, + "loss": 0.0972, + "step": 36200 + }, + { + "epoch": 0.31424206387097336, + "grad_norm": 0.208984375, + "learning_rate": 0.0014560668819731436, + "loss": 0.1289, + "step": 36201 + }, + { + "epoch": 0.3142507443511775, + "grad_norm": 0.8046875, + "learning_rate": 0.001456039547639303, + "loss": 0.103, + "step": 36202 + }, + { + "epoch": 0.3142594248313817, + "grad_norm": 0.1640625, + "learning_rate": 0.001456012212916103, + "loss": 0.1152, + "step": 36203 + }, + { + "epoch": 0.31426810531158583, + "grad_norm": 0.388671875, + "learning_rate": 0.0014559848778035731, + "loss": 0.1396, + "step": 36204 + }, + { + "epoch": 0.31427678579179, + "grad_norm": 0.37109375, + "learning_rate": 0.0014559575423017438, + "loss": 0.1162, + "step": 36205 + }, + { + "epoch": 0.31428546627199416, + "grad_norm": 0.09765625, + "learning_rate": 0.0014559302064106445, + "loss": 0.1123, + "step": 36206 + }, + { + "epoch": 0.31429414675219836, + "grad_norm": 0.1455078125, + "learning_rate": 0.0014559028701303057, + "loss": 0.0776, + "step": 36207 + }, + { + "epoch": 0.3143028272324025, + "grad_norm": 0.240234375, + "learning_rate": 0.0014558755334607564, + "loss": 0.1201, + "step": 36208 + }, + { + "epoch": 0.3143115077126067, + "grad_norm": 0.5078125, + "learning_rate": 0.0014558481964020268, + "loss": 0.1245, + "step": 36209 + }, + { + "epoch": 0.3143201881928108, + "grad_norm": 0.318359375, + "learning_rate": 0.0014558208589541472, + "loss": 0.1484, + "step": 36210 + }, + { + "epoch": 0.314328868673015, + "grad_norm": 0.345703125, + "learning_rate": 0.0014557935211171476, + "loss": 0.1406, + "step": 36211 + }, + { + "epoch": 0.31433754915321915, + "grad_norm": 0.431640625, + "learning_rate": 0.0014557661828910574, + "loss": 0.0869, + "step": 36212 + }, + { + "epoch": 0.31434622963342335, + "grad_norm": 0.6484375, + "learning_rate": 0.0014557388442759064, + "loss": 0.0767, + "step": 36213 + }, + { + "epoch": 0.3143549101136275, + "grad_norm": 0.23828125, + "learning_rate": 0.0014557115052717248, + "loss": 0.1221, + "step": 36214 + }, + { + "epoch": 0.3143635905938317, + "grad_norm": 0.1044921875, + "learning_rate": 0.0014556841658785425, + "loss": 0.1162, + "step": 36215 + }, + { + "epoch": 0.3143722710740358, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014556568260963895, + "loss": 0.1162, + "step": 36216 + }, + { + "epoch": 0.31438095155424, + "grad_norm": 0.1904296875, + "learning_rate": 0.0014556294859252952, + "loss": 0.0786, + "step": 36217 + }, + { + "epoch": 0.31438963203444414, + "grad_norm": 0.8515625, + "learning_rate": 0.0014556021453652902, + "loss": 0.0889, + "step": 36218 + }, + { + "epoch": 0.31439831251464834, + "grad_norm": 0.16796875, + "learning_rate": 0.0014555748044164036, + "loss": 0.0996, + "step": 36219 + }, + { + "epoch": 0.3144069929948525, + "grad_norm": 0.83984375, + "learning_rate": 0.0014555474630786663, + "loss": 0.1084, + "step": 36220 + }, + { + "epoch": 0.31441567347505667, + "grad_norm": 0.265625, + "learning_rate": 0.0014555201213521073, + "loss": 0.1152, + "step": 36221 + }, + { + "epoch": 0.3144243539552608, + "grad_norm": 0.65234375, + "learning_rate": 0.0014554927792367566, + "loss": 0.1147, + "step": 36222 + }, + { + "epoch": 0.314433034435465, + "grad_norm": 1.0390625, + "learning_rate": 0.0014554654367326448, + "loss": 0.0884, + "step": 36223 + }, + { + "epoch": 0.31444171491566913, + "grad_norm": 0.41796875, + "learning_rate": 0.0014554380938398008, + "loss": 0.1084, + "step": 36224 + }, + { + "epoch": 0.3144503953958733, + "grad_norm": 0.4140625, + "learning_rate": 0.0014554107505582556, + "loss": 0.1309, + "step": 36225 + }, + { + "epoch": 0.31445907587607747, + "grad_norm": 0.29296875, + "learning_rate": 0.0014553834068880384, + "loss": 0.1045, + "step": 36226 + }, + { + "epoch": 0.3144677563562816, + "grad_norm": 0.1630859375, + "learning_rate": 0.001455356062829179, + "loss": 0.1426, + "step": 36227 + }, + { + "epoch": 0.3144764368364858, + "grad_norm": 0.482421875, + "learning_rate": 0.0014553287183817077, + "loss": 0.1162, + "step": 36228 + }, + { + "epoch": 0.31448511731668993, + "grad_norm": 0.0966796875, + "learning_rate": 0.0014553013735456545, + "loss": 0.1123, + "step": 36229 + }, + { + "epoch": 0.3144937977968941, + "grad_norm": 0.33984375, + "learning_rate": 0.0014552740283210485, + "loss": 0.1289, + "step": 36230 + }, + { + "epoch": 0.31450247827709826, + "grad_norm": 0.1865234375, + "learning_rate": 0.0014552466827079205, + "loss": 0.1099, + "step": 36231 + }, + { + "epoch": 0.31451115875730246, + "grad_norm": 0.37890625, + "learning_rate": 0.0014552193367063002, + "loss": 0.1055, + "step": 36232 + }, + { + "epoch": 0.3145198392375066, + "grad_norm": 0.39453125, + "learning_rate": 0.0014551919903162173, + "loss": 0.0811, + "step": 36233 + }, + { + "epoch": 0.3145285197177108, + "grad_norm": 0.193359375, + "learning_rate": 0.0014551646435377018, + "loss": 0.1216, + "step": 36234 + }, + { + "epoch": 0.3145372001979149, + "grad_norm": 0.328125, + "learning_rate": 0.0014551372963707832, + "loss": 0.1226, + "step": 36235 + }, + { + "epoch": 0.3145458806781191, + "grad_norm": 0.12451171875, + "learning_rate": 0.001455109948815492, + "loss": 0.1328, + "step": 36236 + }, + { + "epoch": 0.31455456115832325, + "grad_norm": 0.126953125, + "learning_rate": 0.001455082600871858, + "loss": 0.1001, + "step": 36237 + }, + { + "epoch": 0.31456324163852745, + "grad_norm": 0.349609375, + "learning_rate": 0.001455055252539911, + "loss": 0.1201, + "step": 36238 + }, + { + "epoch": 0.3145719221187316, + "grad_norm": 0.08837890625, + "learning_rate": 0.001455027903819681, + "loss": 0.0972, + "step": 36239 + }, + { + "epoch": 0.3145806025989358, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014550005547111975, + "loss": 0.0996, + "step": 36240 + }, + { + "epoch": 0.3145892830791399, + "grad_norm": 0.205078125, + "learning_rate": 0.0014549732052144913, + "loss": 0.124, + "step": 36241 + }, + { + "epoch": 0.3145979635593441, + "grad_norm": 0.32421875, + "learning_rate": 0.001454945855329591, + "loss": 0.0967, + "step": 36242 + }, + { + "epoch": 0.31460664403954824, + "grad_norm": 0.09765625, + "learning_rate": 0.001454918505056528, + "loss": 0.0869, + "step": 36243 + }, + { + "epoch": 0.31461532451975244, + "grad_norm": 0.2451171875, + "learning_rate": 0.0014548911543953308, + "loss": 0.1807, + "step": 36244 + }, + { + "epoch": 0.3146240049999566, + "grad_norm": 0.19140625, + "learning_rate": 0.0014548638033460304, + "loss": 0.0825, + "step": 36245 + }, + { + "epoch": 0.31463268548016077, + "grad_norm": 0.126953125, + "learning_rate": 0.0014548364519086558, + "loss": 0.0708, + "step": 36246 + }, + { + "epoch": 0.3146413659603649, + "grad_norm": 0.30078125, + "learning_rate": 0.0014548091000832378, + "loss": 0.1021, + "step": 36247 + }, + { + "epoch": 0.3146500464405691, + "grad_norm": 0.58203125, + "learning_rate": 0.0014547817478698056, + "loss": 0.0869, + "step": 36248 + }, + { + "epoch": 0.31465872692077324, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014547543952683898, + "loss": 0.085, + "step": 36249 + }, + { + "epoch": 0.31466740740097743, + "grad_norm": 0.7109375, + "learning_rate": 0.0014547270422790198, + "loss": 0.1133, + "step": 36250 + }, + { + "epoch": 0.31467608788118157, + "grad_norm": 0.443359375, + "learning_rate": 0.0014546996889017256, + "loss": 0.0737, + "step": 36251 + }, + { + "epoch": 0.31468476836138576, + "grad_norm": 1.0546875, + "learning_rate": 0.0014546723351365372, + "loss": 0.2617, + "step": 36252 + }, + { + "epoch": 0.3146934488415899, + "grad_norm": 0.2734375, + "learning_rate": 0.001454644980983484, + "loss": 0.1216, + "step": 36253 + }, + { + "epoch": 0.3147021293217941, + "grad_norm": 0.515625, + "learning_rate": 0.001454617626442597, + "loss": 0.1348, + "step": 36254 + }, + { + "epoch": 0.3147108098019982, + "grad_norm": 0.240234375, + "learning_rate": 0.0014545902715139053, + "loss": 0.1182, + "step": 36255 + }, + { + "epoch": 0.3147194902822024, + "grad_norm": 0.228515625, + "learning_rate": 0.0014545629161974388, + "loss": 0.1816, + "step": 36256 + }, + { + "epoch": 0.31472817076240656, + "grad_norm": 0.259765625, + "learning_rate": 0.001454535560493228, + "loss": 0.1182, + "step": 36257 + }, + { + "epoch": 0.31473685124261075, + "grad_norm": 0.330078125, + "learning_rate": 0.0014545082044013018, + "loss": 0.1211, + "step": 36258 + }, + { + "epoch": 0.3147455317228149, + "grad_norm": 0.515625, + "learning_rate": 0.0014544808479216914, + "loss": 0.1455, + "step": 36259 + }, + { + "epoch": 0.3147542122030191, + "grad_norm": 0.181640625, + "learning_rate": 0.0014544534910544257, + "loss": 0.0967, + "step": 36260 + }, + { + "epoch": 0.3147628926832232, + "grad_norm": 0.24609375, + "learning_rate": 0.0014544261337995352, + "loss": 0.1021, + "step": 36261 + }, + { + "epoch": 0.3147715731634274, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014543987761570494, + "loss": 0.1099, + "step": 36262 + }, + { + "epoch": 0.31478025364363155, + "grad_norm": 0.115234375, + "learning_rate": 0.0014543714181269988, + "loss": 0.1064, + "step": 36263 + }, + { + "epoch": 0.31478893412383574, + "grad_norm": 0.1640625, + "learning_rate": 0.0014543440597094125, + "loss": 0.0811, + "step": 36264 + }, + { + "epoch": 0.3147976146040399, + "grad_norm": 0.2890625, + "learning_rate": 0.0014543167009043211, + "loss": 0.0957, + "step": 36265 + }, + { + "epoch": 0.31480629508424407, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014542893417117545, + "loss": 0.1055, + "step": 36266 + }, + { + "epoch": 0.3148149755644482, + "grad_norm": 0.09814453125, + "learning_rate": 0.0014542619821317417, + "loss": 0.1113, + "step": 36267 + }, + { + "epoch": 0.3148236560446524, + "grad_norm": 0.28515625, + "learning_rate": 0.001454234622164314, + "loss": 0.0918, + "step": 36268 + }, + { + "epoch": 0.31483233652485654, + "grad_norm": 0.2490234375, + "learning_rate": 0.0014542072618095004, + "loss": 0.0947, + "step": 36269 + }, + { + "epoch": 0.31484101700506073, + "grad_norm": 0.146484375, + "learning_rate": 0.0014541799010673309, + "loss": 0.0903, + "step": 36270 + }, + { + "epoch": 0.31484969748526487, + "grad_norm": 0.27734375, + "learning_rate": 0.001454152539937836, + "loss": 0.0864, + "step": 36271 + }, + { + "epoch": 0.31485837796546906, + "grad_norm": 0.236328125, + "learning_rate": 0.0014541251784210448, + "loss": 0.0918, + "step": 36272 + }, + { + "epoch": 0.3148670584456732, + "grad_norm": 0.267578125, + "learning_rate": 0.0014540978165169878, + "loss": 0.1719, + "step": 36273 + }, + { + "epoch": 0.3148757389258774, + "grad_norm": 0.6328125, + "learning_rate": 0.0014540704542256946, + "loss": 0.1191, + "step": 36274 + }, + { + "epoch": 0.31488441940608153, + "grad_norm": 0.41015625, + "learning_rate": 0.0014540430915471955, + "loss": 0.1523, + "step": 36275 + }, + { + "epoch": 0.3148930998862857, + "grad_norm": 0.158203125, + "learning_rate": 0.0014540157284815198, + "loss": 0.1328, + "step": 36276 + }, + { + "epoch": 0.31490178036648986, + "grad_norm": 0.396484375, + "learning_rate": 0.0014539883650286983, + "loss": 0.1235, + "step": 36277 + }, + { + "epoch": 0.31491046084669405, + "grad_norm": 0.310546875, + "learning_rate": 0.0014539610011887604, + "loss": 0.0806, + "step": 36278 + }, + { + "epoch": 0.3149191413268982, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014539336369617354, + "loss": 0.1011, + "step": 36279 + }, + { + "epoch": 0.3149278218071024, + "grad_norm": 1.703125, + "learning_rate": 0.0014539062723476549, + "loss": 0.125, + "step": 36280 + }, + { + "epoch": 0.3149365022873065, + "grad_norm": 0.10205078125, + "learning_rate": 0.001453878907346547, + "loss": 0.0933, + "step": 36281 + }, + { + "epoch": 0.3149451827675107, + "grad_norm": 0.40234375, + "learning_rate": 0.0014538515419584431, + "loss": 0.1152, + "step": 36282 + }, + { + "epoch": 0.31495386324771485, + "grad_norm": 0.8359375, + "learning_rate": 0.001453824176183372, + "loss": 0.0698, + "step": 36283 + }, + { + "epoch": 0.31496254372791904, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014537968100213643, + "loss": 0.0957, + "step": 36284 + }, + { + "epoch": 0.3149712242081232, + "grad_norm": 0.27734375, + "learning_rate": 0.0014537694434724495, + "loss": 0.1074, + "step": 36285 + }, + { + "epoch": 0.3149799046883274, + "grad_norm": 0.62890625, + "learning_rate": 0.001453742076536658, + "loss": 0.1172, + "step": 36286 + }, + { + "epoch": 0.3149885851685315, + "grad_norm": 0.0791015625, + "learning_rate": 0.001453714709214019, + "loss": 0.0913, + "step": 36287 + }, + { + "epoch": 0.3149972656487357, + "grad_norm": 0.337890625, + "learning_rate": 0.0014536873415045635, + "loss": 0.0801, + "step": 36288 + }, + { + "epoch": 0.31500594612893984, + "grad_norm": 0.35546875, + "learning_rate": 0.0014536599734083207, + "loss": 0.0947, + "step": 36289 + }, + { + "epoch": 0.31501462660914403, + "grad_norm": 0.205078125, + "learning_rate": 0.0014536326049253206, + "loss": 0.1123, + "step": 36290 + }, + { + "epoch": 0.31502330708934817, + "grad_norm": 0.0927734375, + "learning_rate": 0.0014536052360555934, + "loss": 0.1196, + "step": 36291 + }, + { + "epoch": 0.31503198756955236, + "grad_norm": 0.263671875, + "learning_rate": 0.0014535778667991684, + "loss": 0.1289, + "step": 36292 + }, + { + "epoch": 0.3150406680497565, + "grad_norm": 0.13671875, + "learning_rate": 0.0014535504971560761, + "loss": 0.1865, + "step": 36293 + }, + { + "epoch": 0.3150493485299607, + "grad_norm": 0.1171875, + "learning_rate": 0.0014535231271263465, + "loss": 0.0928, + "step": 36294 + }, + { + "epoch": 0.31505802901016483, + "grad_norm": 0.263671875, + "learning_rate": 0.001453495756710009, + "loss": 0.1104, + "step": 36295 + }, + { + "epoch": 0.315066709490369, + "grad_norm": 0.138671875, + "learning_rate": 0.0014534683859070941, + "loss": 0.0977, + "step": 36296 + }, + { + "epoch": 0.31507538997057316, + "grad_norm": 0.87109375, + "learning_rate": 0.0014534410147176314, + "loss": 0.1025, + "step": 36297 + }, + { + "epoch": 0.31508407045077735, + "grad_norm": 0.484375, + "learning_rate": 0.0014534136431416512, + "loss": 0.1318, + "step": 36298 + }, + { + "epoch": 0.3150927509309815, + "grad_norm": 0.11572265625, + "learning_rate": 0.0014533862711791827, + "loss": 0.1045, + "step": 36299 + }, + { + "epoch": 0.3151014314111857, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014533588988302564, + "loss": 0.1533, + "step": 36300 + }, + { + "epoch": 0.3151101118913898, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014533315260949022, + "loss": 0.1016, + "step": 36301 + }, + { + "epoch": 0.315118792371594, + "grad_norm": 0.146484375, + "learning_rate": 0.0014533041529731499, + "loss": 0.1016, + "step": 36302 + }, + { + "epoch": 0.31512747285179815, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014532767794650297, + "loss": 0.0732, + "step": 36303 + }, + { + "epoch": 0.31513615333200234, + "grad_norm": 0.126953125, + "learning_rate": 0.001453249405570571, + "loss": 0.1045, + "step": 36304 + }, + { + "epoch": 0.3151448338122065, + "grad_norm": 0.369140625, + "learning_rate": 0.0014532220312898042, + "loss": 0.1011, + "step": 36305 + }, + { + "epoch": 0.3151535142924107, + "grad_norm": 0.63671875, + "learning_rate": 0.001453194656622759, + "loss": 0.1309, + "step": 36306 + }, + { + "epoch": 0.3151621947726148, + "grad_norm": 0.39453125, + "learning_rate": 0.0014531672815694657, + "loss": 0.1118, + "step": 36307 + }, + { + "epoch": 0.315170875252819, + "grad_norm": 0.35546875, + "learning_rate": 0.0014531399061299538, + "loss": 0.1094, + "step": 36308 + }, + { + "epoch": 0.31517955573302314, + "grad_norm": 0.0771484375, + "learning_rate": 0.0014531125303042533, + "loss": 0.125, + "step": 36309 + }, + { + "epoch": 0.31518823621322734, + "grad_norm": 0.109375, + "learning_rate": 0.0014530851540923945, + "loss": 0.0884, + "step": 36310 + }, + { + "epoch": 0.3151969166934315, + "grad_norm": 0.34765625, + "learning_rate": 0.001453057777494407, + "loss": 0.1318, + "step": 36311 + }, + { + "epoch": 0.31520559717363567, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014530304005103208, + "loss": 0.1436, + "step": 36312 + }, + { + "epoch": 0.3152142776538398, + "grad_norm": 0.4296875, + "learning_rate": 0.0014530030231401654, + "loss": 0.085, + "step": 36313 + }, + { + "epoch": 0.315222958134044, + "grad_norm": 0.185546875, + "learning_rate": 0.0014529756453839718, + "loss": 0.1357, + "step": 36314 + }, + { + "epoch": 0.31523163861424813, + "grad_norm": 0.1796875, + "learning_rate": 0.001452948267241769, + "loss": 0.0884, + "step": 36315 + }, + { + "epoch": 0.3152403190944523, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014529208887135874, + "loss": 0.1455, + "step": 36316 + }, + { + "epoch": 0.31524899957465646, + "grad_norm": 0.45703125, + "learning_rate": 0.0014528935097994568, + "loss": 0.1006, + "step": 36317 + }, + { + "epoch": 0.31525768005486066, + "grad_norm": 0.26953125, + "learning_rate": 0.0014528661304994072, + "loss": 0.0933, + "step": 36318 + }, + { + "epoch": 0.3152663605350648, + "grad_norm": 0.482421875, + "learning_rate": 0.0014528387508134686, + "loss": 0.123, + "step": 36319 + }, + { + "epoch": 0.315275041015269, + "grad_norm": 0.0888671875, + "learning_rate": 0.001452811370741671, + "loss": 0.0664, + "step": 36320 + }, + { + "epoch": 0.3152837214954731, + "grad_norm": 0.1796875, + "learning_rate": 0.0014527839902840438, + "loss": 0.0947, + "step": 36321 + }, + { + "epoch": 0.3152924019756773, + "grad_norm": 0.56640625, + "learning_rate": 0.0014527566094406173, + "loss": 0.1387, + "step": 36322 + }, + { + "epoch": 0.31530108245588145, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014527292282114218, + "loss": 0.0986, + "step": 36323 + }, + { + "epoch": 0.31530976293608565, + "grad_norm": 0.0810546875, + "learning_rate": 0.0014527018465964865, + "loss": 0.082, + "step": 36324 + }, + { + "epoch": 0.3153184434162898, + "grad_norm": 0.380859375, + "learning_rate": 0.001452674464595842, + "loss": 0.1406, + "step": 36325 + }, + { + "epoch": 0.315327123896494, + "grad_norm": 0.1826171875, + "learning_rate": 0.001452647082209518, + "loss": 0.1128, + "step": 36326 + }, + { + "epoch": 0.3153358043766981, + "grad_norm": 0.1240234375, + "learning_rate": 0.0014526196994375444, + "loss": 0.0967, + "step": 36327 + }, + { + "epoch": 0.3153444848569023, + "grad_norm": 0.63671875, + "learning_rate": 0.0014525923162799514, + "loss": 0.085, + "step": 36328 + }, + { + "epoch": 0.31535316533710644, + "grad_norm": 0.18359375, + "learning_rate": 0.0014525649327367686, + "loss": 0.1309, + "step": 36329 + }, + { + "epoch": 0.31536184581731064, + "grad_norm": 0.294921875, + "learning_rate": 0.0014525375488080262, + "loss": 0.0918, + "step": 36330 + }, + { + "epoch": 0.3153705262975148, + "grad_norm": 0.12158203125, + "learning_rate": 0.0014525101644937539, + "loss": 0.124, + "step": 36331 + }, + { + "epoch": 0.31537920677771897, + "grad_norm": 0.31640625, + "learning_rate": 0.0014524827797939817, + "loss": 0.1279, + "step": 36332 + }, + { + "epoch": 0.3153878872579231, + "grad_norm": 0.36328125, + "learning_rate": 0.0014524553947087395, + "loss": 0.1025, + "step": 36333 + }, + { + "epoch": 0.3153965677381273, + "grad_norm": 0.21875, + "learning_rate": 0.0014524280092380578, + "loss": 0.0967, + "step": 36334 + }, + { + "epoch": 0.31540524821833144, + "grad_norm": 0.275390625, + "learning_rate": 0.0014524006233819658, + "loss": 0.0977, + "step": 36335 + }, + { + "epoch": 0.31541392869853563, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014523732371404942, + "loss": 0.1299, + "step": 36336 + }, + { + "epoch": 0.31542260917873977, + "grad_norm": 0.8203125, + "learning_rate": 0.001452345850513672, + "loss": 0.1875, + "step": 36337 + }, + { + "epoch": 0.31543128965894396, + "grad_norm": 0.267578125, + "learning_rate": 0.0014523184635015299, + "loss": 0.127, + "step": 36338 + }, + { + "epoch": 0.3154399701391481, + "grad_norm": 0.43359375, + "learning_rate": 0.0014522910761040977, + "loss": 0.0732, + "step": 36339 + }, + { + "epoch": 0.3154486506193523, + "grad_norm": 0.57421875, + "learning_rate": 0.0014522636883214052, + "loss": 0.1133, + "step": 36340 + }, + { + "epoch": 0.3154573310995564, + "grad_norm": 0.197265625, + "learning_rate": 0.0014522363001534823, + "loss": 0.127, + "step": 36341 + }, + { + "epoch": 0.3154660115797606, + "grad_norm": 0.130859375, + "learning_rate": 0.0014522089116003592, + "loss": 0.0796, + "step": 36342 + }, + { + "epoch": 0.31547469205996476, + "grad_norm": 0.51953125, + "learning_rate": 0.0014521815226620656, + "loss": 0.1104, + "step": 36343 + }, + { + "epoch": 0.31548337254016895, + "grad_norm": 0.50390625, + "learning_rate": 0.0014521541333386316, + "loss": 0.1299, + "step": 36344 + }, + { + "epoch": 0.3154920530203731, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014521267436300872, + "loss": 0.0972, + "step": 36345 + }, + { + "epoch": 0.3155007335005773, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014520993535364624, + "loss": 0.0947, + "step": 36346 + }, + { + "epoch": 0.3155094139807814, + "grad_norm": 0.306640625, + "learning_rate": 0.001452071963057787, + "loss": 0.0859, + "step": 36347 + }, + { + "epoch": 0.31551809446098555, + "grad_norm": 0.98828125, + "learning_rate": 0.0014520445721940907, + "loss": 0.1021, + "step": 36348 + }, + { + "epoch": 0.31552677494118975, + "grad_norm": 0.55078125, + "learning_rate": 0.0014520171809454042, + "loss": 0.0977, + "step": 36349 + }, + { + "epoch": 0.3155354554213939, + "grad_norm": 0.80078125, + "learning_rate": 0.001451989789311757, + "loss": 0.1157, + "step": 36350 + }, + { + "epoch": 0.3155441359015981, + "grad_norm": 0.451171875, + "learning_rate": 0.0014519623972931786, + "loss": 0.0908, + "step": 36351 + }, + { + "epoch": 0.3155528163818022, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014519350048896994, + "loss": 0.0591, + "step": 36352 + }, + { + "epoch": 0.3155614968620064, + "grad_norm": 0.2470703125, + "learning_rate": 0.0014519076121013498, + "loss": 0.0889, + "step": 36353 + }, + { + "epoch": 0.31557017734221054, + "grad_norm": 0.10888671875, + "learning_rate": 0.001451880218928159, + "loss": 0.0693, + "step": 36354 + }, + { + "epoch": 0.31557885782241474, + "grad_norm": 0.365234375, + "learning_rate": 0.0014518528253701574, + "loss": 0.1299, + "step": 36355 + }, + { + "epoch": 0.3155875383026189, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014518254314273746, + "loss": 0.1064, + "step": 36356 + }, + { + "epoch": 0.31559621878282307, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014517980370998414, + "loss": 0.0894, + "step": 36357 + }, + { + "epoch": 0.3156048992630272, + "grad_norm": 0.267578125, + "learning_rate": 0.0014517706423875867, + "loss": 0.1152, + "step": 36358 + }, + { + "epoch": 0.3156135797432314, + "grad_norm": 0.3828125, + "learning_rate": 0.001451743247290641, + "loss": 0.0981, + "step": 36359 + }, + { + "epoch": 0.31562226022343554, + "grad_norm": 0.083984375, + "learning_rate": 0.0014517158518090342, + "loss": 0.0996, + "step": 36360 + }, + { + "epoch": 0.31563094070363973, + "grad_norm": 0.3671875, + "learning_rate": 0.0014516884559427961, + "loss": 0.1099, + "step": 36361 + }, + { + "epoch": 0.31563962118384387, + "grad_norm": 0.404296875, + "learning_rate": 0.001451661059691957, + "loss": 0.1631, + "step": 36362 + }, + { + "epoch": 0.31564830166404806, + "grad_norm": 0.1142578125, + "learning_rate": 0.0014516336630565462, + "loss": 0.1094, + "step": 36363 + }, + { + "epoch": 0.3156569821442522, + "grad_norm": 0.255859375, + "learning_rate": 0.001451606266036595, + "loss": 0.1133, + "step": 36364 + }, + { + "epoch": 0.3156656626244564, + "grad_norm": 0.1552734375, + "learning_rate": 0.0014515788686321315, + "loss": 0.1006, + "step": 36365 + }, + { + "epoch": 0.3156743431046605, + "grad_norm": 0.333984375, + "learning_rate": 0.001451551470843187, + "loss": 0.1074, + "step": 36366 + }, + { + "epoch": 0.3156830235848647, + "grad_norm": 0.2294921875, + "learning_rate": 0.0014515240726697915, + "loss": 0.084, + "step": 36367 + }, + { + "epoch": 0.31569170406506886, + "grad_norm": 0.11572265625, + "learning_rate": 0.001451496674111974, + "loss": 0.1138, + "step": 36368 + }, + { + "epoch": 0.31570038454527305, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014514692751697652, + "loss": 0.1191, + "step": 36369 + }, + { + "epoch": 0.3157090650254772, + "grad_norm": 0.2275390625, + "learning_rate": 0.001451441875843195, + "loss": 0.1709, + "step": 36370 + }, + { + "epoch": 0.3157177455056814, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014514144761322931, + "loss": 0.0771, + "step": 36371 + }, + { + "epoch": 0.3157264259858855, + "grad_norm": 0.255859375, + "learning_rate": 0.0014513870760370898, + "loss": 0.1328, + "step": 36372 + }, + { + "epoch": 0.3157351064660897, + "grad_norm": 0.291015625, + "learning_rate": 0.0014513596755576146, + "loss": 0.085, + "step": 36373 + }, + { + "epoch": 0.31574378694629385, + "grad_norm": 0.291015625, + "learning_rate": 0.0014513322746938977, + "loss": 0.1211, + "step": 36374 + }, + { + "epoch": 0.31575246742649804, + "grad_norm": 0.283203125, + "learning_rate": 0.0014513048734459692, + "loss": 0.1011, + "step": 36375 + }, + { + "epoch": 0.3157611479067022, + "grad_norm": 0.361328125, + "learning_rate": 0.001451277471813859, + "loss": 0.0981, + "step": 36376 + }, + { + "epoch": 0.31576982838690637, + "grad_norm": 1.046875, + "learning_rate": 0.0014512500697975972, + "loss": 0.2969, + "step": 36377 + }, + { + "epoch": 0.3157785088671105, + "grad_norm": 0.72265625, + "learning_rate": 0.0014512226673972135, + "loss": 0.1035, + "step": 36378 + }, + { + "epoch": 0.3157871893473147, + "grad_norm": 0.13671875, + "learning_rate": 0.001451195264612738, + "loss": 0.0889, + "step": 36379 + }, + { + "epoch": 0.31579586982751884, + "grad_norm": 0.73828125, + "learning_rate": 0.0014511678614442006, + "loss": 0.0908, + "step": 36380 + }, + { + "epoch": 0.31580455030772303, + "grad_norm": 0.353515625, + "learning_rate": 0.0014511404578916311, + "loss": 0.0947, + "step": 36381 + }, + { + "epoch": 0.31581323078792717, + "grad_norm": 0.310546875, + "learning_rate": 0.00145111305395506, + "loss": 0.0996, + "step": 36382 + }, + { + "epoch": 0.31582191126813136, + "grad_norm": 0.453125, + "learning_rate": 0.0014510856496345167, + "loss": 0.1318, + "step": 36383 + }, + { + "epoch": 0.3158305917483355, + "grad_norm": 0.189453125, + "learning_rate": 0.0014510582449300316, + "loss": 0.1055, + "step": 36384 + }, + { + "epoch": 0.3158392722285397, + "grad_norm": 0.337890625, + "learning_rate": 0.0014510308398416342, + "loss": 0.1357, + "step": 36385 + }, + { + "epoch": 0.31584795270874383, + "grad_norm": 0.2177734375, + "learning_rate": 0.001451003434369355, + "loss": 0.0786, + "step": 36386 + }, + { + "epoch": 0.315856633188948, + "grad_norm": 0.5703125, + "learning_rate": 0.0014509760285132236, + "loss": 0.1035, + "step": 36387 + }, + { + "epoch": 0.31586531366915216, + "grad_norm": 0.318359375, + "learning_rate": 0.0014509486222732703, + "loss": 0.1064, + "step": 36388 + }, + { + "epoch": 0.31587399414935635, + "grad_norm": 0.412109375, + "learning_rate": 0.0014509212156495247, + "loss": 0.0918, + "step": 36389 + }, + { + "epoch": 0.3158826746295605, + "grad_norm": 0.83984375, + "learning_rate": 0.0014508938086420168, + "loss": 0.1465, + "step": 36390 + }, + { + "epoch": 0.3158913551097647, + "grad_norm": 0.1796875, + "learning_rate": 0.0014508664012507768, + "loss": 0.1206, + "step": 36391 + }, + { + "epoch": 0.3159000355899688, + "grad_norm": 0.423828125, + "learning_rate": 0.0014508389934758345, + "loss": 0.106, + "step": 36392 + }, + { + "epoch": 0.315908716070173, + "grad_norm": 0.111328125, + "learning_rate": 0.0014508115853172198, + "loss": 0.165, + "step": 36393 + }, + { + "epoch": 0.31591739655037715, + "grad_norm": 0.2470703125, + "learning_rate": 0.0014507841767749635, + "loss": 0.1064, + "step": 36394 + }, + { + "epoch": 0.31592607703058134, + "grad_norm": 0.30859375, + "learning_rate": 0.0014507567678490943, + "loss": 0.0928, + "step": 36395 + }, + { + "epoch": 0.3159347575107855, + "grad_norm": 0.46875, + "learning_rate": 0.0014507293585396428, + "loss": 0.1123, + "step": 36396 + }, + { + "epoch": 0.3159434379909897, + "grad_norm": 0.27734375, + "learning_rate": 0.0014507019488466393, + "loss": 0.1035, + "step": 36397 + }, + { + "epoch": 0.3159521184711938, + "grad_norm": 0.408203125, + "learning_rate": 0.0014506745387701133, + "loss": 0.0933, + "step": 36398 + }, + { + "epoch": 0.315960798951398, + "grad_norm": 0.60546875, + "learning_rate": 0.0014506471283100948, + "loss": 0.1006, + "step": 36399 + }, + { + "epoch": 0.31596947943160214, + "grad_norm": 0.37109375, + "learning_rate": 0.001450619717466614, + "loss": 0.1025, + "step": 36400 + }, + { + "epoch": 0.31597815991180633, + "grad_norm": 0.44921875, + "learning_rate": 0.0014505923062397002, + "loss": 0.1055, + "step": 36401 + }, + { + "epoch": 0.31598684039201047, + "grad_norm": 0.5546875, + "learning_rate": 0.0014505648946293844, + "loss": 0.0918, + "step": 36402 + }, + { + "epoch": 0.31599552087221466, + "grad_norm": 0.2490234375, + "learning_rate": 0.0014505374826356962, + "loss": 0.1191, + "step": 36403 + }, + { + "epoch": 0.3160042013524188, + "grad_norm": 0.111328125, + "learning_rate": 0.0014505100702586651, + "loss": 0.0913, + "step": 36404 + }, + { + "epoch": 0.316012881832623, + "grad_norm": 0.84765625, + "learning_rate": 0.0014504826574983215, + "loss": 0.1147, + "step": 36405 + }, + { + "epoch": 0.31602156231282713, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014504552443546956, + "loss": 0.1123, + "step": 36406 + }, + { + "epoch": 0.3160302427930313, + "grad_norm": 0.478515625, + "learning_rate": 0.0014504278308278171, + "loss": 0.167, + "step": 36407 + }, + { + "epoch": 0.31603892327323546, + "grad_norm": 0.5078125, + "learning_rate": 0.0014504004169177158, + "loss": 0.0815, + "step": 36408 + }, + { + "epoch": 0.31604760375343965, + "grad_norm": 0.1474609375, + "learning_rate": 0.001450373002624422, + "loss": 0.1113, + "step": 36409 + }, + { + "epoch": 0.3160562842336438, + "grad_norm": 0.263671875, + "learning_rate": 0.0014503455879479655, + "loss": 0.123, + "step": 36410 + }, + { + "epoch": 0.316064964713848, + "grad_norm": 0.2275390625, + "learning_rate": 0.001450318172888376, + "loss": 0.1006, + "step": 36411 + }, + { + "epoch": 0.3160736451940521, + "grad_norm": 0.14453125, + "learning_rate": 0.0014502907574456844, + "loss": 0.0947, + "step": 36412 + }, + { + "epoch": 0.3160823256742563, + "grad_norm": 0.27734375, + "learning_rate": 0.0014502633416199197, + "loss": 0.0889, + "step": 36413 + }, + { + "epoch": 0.31609100615446045, + "grad_norm": 0.25390625, + "learning_rate": 0.0014502359254111123, + "loss": 0.1221, + "step": 36414 + }, + { + "epoch": 0.31609968663466465, + "grad_norm": 0.2890625, + "learning_rate": 0.0014502085088192924, + "loss": 0.0869, + "step": 36415 + }, + { + "epoch": 0.3161083671148688, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014501810918444895, + "loss": 0.083, + "step": 36416 + }, + { + "epoch": 0.316117047595073, + "grad_norm": 0.189453125, + "learning_rate": 0.001450153674486734, + "loss": 0.1152, + "step": 36417 + }, + { + "epoch": 0.3161257280752771, + "grad_norm": 0.26171875, + "learning_rate": 0.0014501262567460556, + "loss": 0.1484, + "step": 36418 + }, + { + "epoch": 0.3161344085554813, + "grad_norm": 0.7265625, + "learning_rate": 0.0014500988386224845, + "loss": 0.1279, + "step": 36419 + }, + { + "epoch": 0.31614308903568544, + "grad_norm": 0.125, + "learning_rate": 0.0014500714201160503, + "loss": 0.0869, + "step": 36420 + }, + { + "epoch": 0.31615176951588964, + "grad_norm": 0.12255859375, + "learning_rate": 0.0014500440012267832, + "loss": 0.1064, + "step": 36421 + }, + { + "epoch": 0.3161604499960938, + "grad_norm": 0.41796875, + "learning_rate": 0.0014500165819547137, + "loss": 0.1641, + "step": 36422 + }, + { + "epoch": 0.31616913047629797, + "grad_norm": 0.91796875, + "learning_rate": 0.001449989162299871, + "loss": 0.1816, + "step": 36423 + }, + { + "epoch": 0.3161778109565021, + "grad_norm": 0.466796875, + "learning_rate": 0.0014499617422622858, + "loss": 0.1406, + "step": 36424 + }, + { + "epoch": 0.3161864914367063, + "grad_norm": 0.09619140625, + "learning_rate": 0.0014499343218419875, + "loss": 0.1064, + "step": 36425 + }, + { + "epoch": 0.31619517191691043, + "grad_norm": 0.1689453125, + "learning_rate": 0.001449906901039006, + "loss": 0.1172, + "step": 36426 + }, + { + "epoch": 0.3162038523971146, + "grad_norm": 0.1484375, + "learning_rate": 0.0014498794798533717, + "loss": 0.1021, + "step": 36427 + }, + { + "epoch": 0.31621253287731876, + "grad_norm": 0.212890625, + "learning_rate": 0.0014498520582851145, + "loss": 0.0947, + "step": 36428 + }, + { + "epoch": 0.31622121335752296, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014498246363342643, + "loss": 0.0977, + "step": 36429 + }, + { + "epoch": 0.3162298938377271, + "grad_norm": 0.13671875, + "learning_rate": 0.0014497972140008512, + "loss": 0.0977, + "step": 36430 + }, + { + "epoch": 0.3162385743179313, + "grad_norm": 0.42578125, + "learning_rate": 0.0014497697912849048, + "loss": 0.1064, + "step": 36431 + }, + { + "epoch": 0.3162472547981354, + "grad_norm": 0.08544921875, + "learning_rate": 0.001449742368186456, + "loss": 0.0771, + "step": 36432 + }, + { + "epoch": 0.3162559352783396, + "grad_norm": 0.1181640625, + "learning_rate": 0.001449714944705534, + "loss": 0.0928, + "step": 36433 + }, + { + "epoch": 0.31626461575854375, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014496875208421687, + "loss": 0.085, + "step": 36434 + }, + { + "epoch": 0.31627329623874795, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014496600965963906, + "loss": 0.1045, + "step": 36435 + }, + { + "epoch": 0.3162819767189521, + "grad_norm": 0.08984375, + "learning_rate": 0.0014496326719682295, + "loss": 0.0698, + "step": 36436 + }, + { + "epoch": 0.3162906571991563, + "grad_norm": 0.5234375, + "learning_rate": 0.0014496052469577154, + "loss": 0.0938, + "step": 36437 + }, + { + "epoch": 0.3162993376793604, + "grad_norm": 0.10693359375, + "learning_rate": 0.001449577821564878, + "loss": 0.0908, + "step": 36438 + }, + { + "epoch": 0.3163080181595646, + "grad_norm": 0.19921875, + "learning_rate": 0.001449550395789748, + "loss": 0.1045, + "step": 36439 + }, + { + "epoch": 0.31631669863976875, + "grad_norm": 0.0634765625, + "learning_rate": 0.0014495229696323544, + "loss": 0.0576, + "step": 36440 + }, + { + "epoch": 0.31632537911997294, + "grad_norm": 0.181640625, + "learning_rate": 0.0014494955430927282, + "loss": 0.1201, + "step": 36441 + }, + { + "epoch": 0.3163340596001771, + "grad_norm": 0.2158203125, + "learning_rate": 0.0014494681161708986, + "loss": 0.0835, + "step": 36442 + }, + { + "epoch": 0.31634274008038127, + "grad_norm": 0.1171875, + "learning_rate": 0.001449440688866896, + "loss": 0.1377, + "step": 36443 + }, + { + "epoch": 0.3163514205605854, + "grad_norm": 0.60546875, + "learning_rate": 0.0014494132611807505, + "loss": 0.1738, + "step": 36444 + }, + { + "epoch": 0.3163601010407896, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014493858331124918, + "loss": 0.0952, + "step": 36445 + }, + { + "epoch": 0.31636878152099374, + "grad_norm": 0.099609375, + "learning_rate": 0.00144935840466215, + "loss": 0.0825, + "step": 36446 + }, + { + "epoch": 0.31637746200119793, + "grad_norm": 0.4375, + "learning_rate": 0.001449330975829755, + "loss": 0.0928, + "step": 36447 + }, + { + "epoch": 0.31638614248140207, + "grad_norm": 0.4140625, + "learning_rate": 0.001449303546615337, + "loss": 0.0723, + "step": 36448 + }, + { + "epoch": 0.31639482296160626, + "grad_norm": 0.28125, + "learning_rate": 0.0014492761170189258, + "loss": 0.1016, + "step": 36449 + }, + { + "epoch": 0.3164035034418104, + "grad_norm": 0.419921875, + "learning_rate": 0.0014492486870405514, + "loss": 0.0947, + "step": 36450 + }, + { + "epoch": 0.3164121839220146, + "grad_norm": 0.0849609375, + "learning_rate": 0.0014492212566802438, + "loss": 0.1099, + "step": 36451 + }, + { + "epoch": 0.3164208644022187, + "grad_norm": 0.15625, + "learning_rate": 0.0014491938259380336, + "loss": 0.0898, + "step": 36452 + }, + { + "epoch": 0.3164295448824229, + "grad_norm": 0.2392578125, + "learning_rate": 0.00144916639481395, + "loss": 0.1113, + "step": 36453 + }, + { + "epoch": 0.31643822536262706, + "grad_norm": 0.1474609375, + "learning_rate": 0.0014491389633080232, + "loss": 0.1113, + "step": 36454 + }, + { + "epoch": 0.31644690584283125, + "grad_norm": 0.23828125, + "learning_rate": 0.0014491115314202835, + "loss": 0.1089, + "step": 36455 + }, + { + "epoch": 0.3164555863230354, + "grad_norm": 0.58203125, + "learning_rate": 0.0014490840991507603, + "loss": 0.1216, + "step": 36456 + }, + { + "epoch": 0.3164642668032396, + "grad_norm": 0.197265625, + "learning_rate": 0.0014490566664994843, + "loss": 0.0884, + "step": 36457 + }, + { + "epoch": 0.3164729472834437, + "grad_norm": 0.1650390625, + "learning_rate": 0.001449029233466485, + "loss": 0.0918, + "step": 36458 + }, + { + "epoch": 0.3164816277636479, + "grad_norm": 0.111328125, + "learning_rate": 0.0014490018000517925, + "loss": 0.1309, + "step": 36459 + }, + { + "epoch": 0.31649030824385205, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014489743662554368, + "loss": 0.1064, + "step": 36460 + }, + { + "epoch": 0.31649898872405624, + "grad_norm": 0.171875, + "learning_rate": 0.0014489469320774482, + "loss": 0.0771, + "step": 36461 + }, + { + "epoch": 0.3165076692042604, + "grad_norm": 0.15234375, + "learning_rate": 0.0014489194975178565, + "loss": 0.1196, + "step": 36462 + }, + { + "epoch": 0.31651634968446457, + "grad_norm": 0.111328125, + "learning_rate": 0.0014488920625766917, + "loss": 0.1084, + "step": 36463 + }, + { + "epoch": 0.3165250301646687, + "grad_norm": 1.0, + "learning_rate": 0.0014488646272539836, + "loss": 0.167, + "step": 36464 + }, + { + "epoch": 0.3165337106448729, + "grad_norm": 0.2255859375, + "learning_rate": 0.001448837191549763, + "loss": 0.0903, + "step": 36465 + }, + { + "epoch": 0.31654239112507704, + "grad_norm": 0.275390625, + "learning_rate": 0.0014488097554640586, + "loss": 0.1104, + "step": 36466 + }, + { + "epoch": 0.31655107160528123, + "grad_norm": 0.359375, + "learning_rate": 0.0014487823189969014, + "loss": 0.1152, + "step": 36467 + }, + { + "epoch": 0.31655975208548537, + "grad_norm": 0.251953125, + "learning_rate": 0.0014487548821483208, + "loss": 0.1025, + "step": 36468 + }, + { + "epoch": 0.31656843256568956, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014487274449183469, + "loss": 0.0723, + "step": 36469 + }, + { + "epoch": 0.3165771130458937, + "grad_norm": 0.1943359375, + "learning_rate": 0.0014487000073070105, + "loss": 0.1011, + "step": 36470 + }, + { + "epoch": 0.31658579352609784, + "grad_norm": 0.58984375, + "learning_rate": 0.001448672569314341, + "loss": 0.1172, + "step": 36471 + }, + { + "epoch": 0.31659447400630203, + "grad_norm": 0.275390625, + "learning_rate": 0.0014486451309403683, + "loss": 0.1387, + "step": 36472 + }, + { + "epoch": 0.31660315448650617, + "grad_norm": 0.439453125, + "learning_rate": 0.0014486176921851223, + "loss": 0.1357, + "step": 36473 + }, + { + "epoch": 0.31661183496671036, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014485902530486335, + "loss": 0.0869, + "step": 36474 + }, + { + "epoch": 0.3166205154469145, + "grad_norm": 0.61328125, + "learning_rate": 0.0014485628135309314, + "loss": 0.1123, + "step": 36475 + }, + { + "epoch": 0.3166291959271187, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014485353736320467, + "loss": 0.1289, + "step": 36476 + }, + { + "epoch": 0.3166378764073228, + "grad_norm": 0.28515625, + "learning_rate": 0.0014485079333520084, + "loss": 0.1172, + "step": 36477 + }, + { + "epoch": 0.316646556887527, + "grad_norm": 0.671875, + "learning_rate": 0.001448480492690847, + "loss": 0.123, + "step": 36478 + }, + { + "epoch": 0.31665523736773116, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014484530516485927, + "loss": 0.1406, + "step": 36479 + }, + { + "epoch": 0.31666391784793535, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014484256102252756, + "loss": 0.1182, + "step": 36480 + }, + { + "epoch": 0.3166725983281395, + "grad_norm": 0.2314453125, + "learning_rate": 0.0014483981684209255, + "loss": 0.0762, + "step": 36481 + }, + { + "epoch": 0.3166812788083437, + "grad_norm": 0.384765625, + "learning_rate": 0.0014483707262355722, + "loss": 0.1216, + "step": 36482 + }, + { + "epoch": 0.3166899592885478, + "grad_norm": 0.1884765625, + "learning_rate": 0.001448343283669246, + "loss": 0.1016, + "step": 36483 + }, + { + "epoch": 0.316698639768752, + "grad_norm": 0.408203125, + "learning_rate": 0.001448315840721977, + "loss": 0.0913, + "step": 36484 + }, + { + "epoch": 0.31670732024895615, + "grad_norm": 0.2255859375, + "learning_rate": 0.0014482883973937948, + "loss": 0.1367, + "step": 36485 + }, + { + "epoch": 0.31671600072916034, + "grad_norm": 0.85546875, + "learning_rate": 0.0014482609536847295, + "loss": 0.1445, + "step": 36486 + }, + { + "epoch": 0.3167246812093645, + "grad_norm": 0.625, + "learning_rate": 0.0014482335095948116, + "loss": 0.0986, + "step": 36487 + }, + { + "epoch": 0.31673336168956867, + "grad_norm": 0.392578125, + "learning_rate": 0.0014482060651240705, + "loss": 0.1074, + "step": 36488 + }, + { + "epoch": 0.3167420421697728, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014481786202725367, + "loss": 0.1406, + "step": 36489 + }, + { + "epoch": 0.316750722649977, + "grad_norm": 0.55859375, + "learning_rate": 0.0014481511750402399, + "loss": 0.084, + "step": 36490 + }, + { + "epoch": 0.31675940313018114, + "grad_norm": 0.197265625, + "learning_rate": 0.00144812372942721, + "loss": 0.1162, + "step": 36491 + }, + { + "epoch": 0.31676808361038533, + "grad_norm": 0.36328125, + "learning_rate": 0.0014480962834334777, + "loss": 0.0967, + "step": 36492 + }, + { + "epoch": 0.31677676409058947, + "grad_norm": 0.12158203125, + "learning_rate": 0.0014480688370590723, + "loss": 0.1079, + "step": 36493 + }, + { + "epoch": 0.31678544457079366, + "grad_norm": 0.10302734375, + "learning_rate": 0.0014480413903040239, + "loss": 0.1099, + "step": 36494 + }, + { + "epoch": 0.3167941250509978, + "grad_norm": 0.36328125, + "learning_rate": 0.0014480139431683626, + "loss": 0.1318, + "step": 36495 + }, + { + "epoch": 0.316802805531202, + "grad_norm": 0.10205078125, + "learning_rate": 0.0014479864956521188, + "loss": 0.0732, + "step": 36496 + }, + { + "epoch": 0.31681148601140613, + "grad_norm": 0.12060546875, + "learning_rate": 0.001447959047755322, + "loss": 0.0801, + "step": 36497 + }, + { + "epoch": 0.3168201664916103, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014479315994780021, + "loss": 0.1113, + "step": 36498 + }, + { + "epoch": 0.31682884697181446, + "grad_norm": 0.26171875, + "learning_rate": 0.0014479041508201897, + "loss": 0.0972, + "step": 36499 + }, + { + "epoch": 0.31683752745201865, + "grad_norm": 1.1171875, + "learning_rate": 0.0014478767017819148, + "loss": 0.1172, + "step": 36500 + }, + { + "epoch": 0.3168462079322228, + "grad_norm": 0.126953125, + "learning_rate": 0.001447849252363207, + "loss": 0.0771, + "step": 36501 + }, + { + "epoch": 0.316854888412427, + "grad_norm": 1.046875, + "learning_rate": 0.0014478218025640965, + "loss": 0.085, + "step": 36502 + }, + { + "epoch": 0.3168635688926311, + "grad_norm": 0.302734375, + "learning_rate": 0.0014477943523846133, + "loss": 0.1201, + "step": 36503 + }, + { + "epoch": 0.3168722493728353, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014477669018247874, + "loss": 0.0957, + "step": 36504 + }, + { + "epoch": 0.31688092985303945, + "grad_norm": 0.171875, + "learning_rate": 0.0014477394508846488, + "loss": 0.1504, + "step": 36505 + }, + { + "epoch": 0.31688961033324364, + "grad_norm": 0.41015625, + "learning_rate": 0.0014477119995642276, + "loss": 0.1064, + "step": 36506 + }, + { + "epoch": 0.3168982908134478, + "grad_norm": 0.62109375, + "learning_rate": 0.0014476845478635539, + "loss": 0.0918, + "step": 36507 + }, + { + "epoch": 0.316906971293652, + "grad_norm": 0.07373046875, + "learning_rate": 0.0014476570957826574, + "loss": 0.0718, + "step": 36508 + }, + { + "epoch": 0.3169156517738561, + "grad_norm": 0.177734375, + "learning_rate": 0.0014476296433215685, + "loss": 0.123, + "step": 36509 + }, + { + "epoch": 0.3169243322540603, + "grad_norm": 0.37890625, + "learning_rate": 0.001447602190480317, + "loss": 0.1406, + "step": 36510 + }, + { + "epoch": 0.31693301273426444, + "grad_norm": 0.3515625, + "learning_rate": 0.001447574737258933, + "loss": 0.0645, + "step": 36511 + }, + { + "epoch": 0.31694169321446863, + "grad_norm": 0.2177734375, + "learning_rate": 0.001447547283657446, + "loss": 0.0938, + "step": 36512 + }, + { + "epoch": 0.31695037369467277, + "grad_norm": 0.201171875, + "learning_rate": 0.0014475198296758874, + "loss": 0.0874, + "step": 36513 + }, + { + "epoch": 0.31695905417487696, + "grad_norm": 0.06396484375, + "learning_rate": 0.0014474923753142856, + "loss": 0.0898, + "step": 36514 + }, + { + "epoch": 0.3169677346550811, + "grad_norm": 0.23046875, + "learning_rate": 0.0014474649205726716, + "loss": 0.1299, + "step": 36515 + }, + { + "epoch": 0.3169764151352853, + "grad_norm": 0.4296875, + "learning_rate": 0.0014474374654510755, + "loss": 0.1162, + "step": 36516 + }, + { + "epoch": 0.31698509561548943, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014474100099495263, + "loss": 0.1055, + "step": 36517 + }, + { + "epoch": 0.3169937760956936, + "grad_norm": 0.2314453125, + "learning_rate": 0.0014473825540680552, + "loss": 0.0967, + "step": 36518 + }, + { + "epoch": 0.31700245657589776, + "grad_norm": 0.89453125, + "learning_rate": 0.0014473550978066918, + "loss": 0.1035, + "step": 36519 + }, + { + "epoch": 0.31701113705610195, + "grad_norm": 0.51171875, + "learning_rate": 0.001447327641165466, + "loss": 0.1113, + "step": 36520 + }, + { + "epoch": 0.3170198175363061, + "grad_norm": 0.1767578125, + "learning_rate": 0.0014473001841444077, + "loss": 0.1377, + "step": 36521 + }, + { + "epoch": 0.3170284980165103, + "grad_norm": 0.455078125, + "learning_rate": 0.0014472727267435474, + "loss": 0.085, + "step": 36522 + }, + { + "epoch": 0.3170371784967144, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014472452689629149, + "loss": 0.1299, + "step": 36523 + }, + { + "epoch": 0.3170458589769186, + "grad_norm": 0.216796875, + "learning_rate": 0.0014472178108025401, + "loss": 0.084, + "step": 36524 + }, + { + "epoch": 0.31705453945712275, + "grad_norm": 0.58984375, + "learning_rate": 0.0014471903522624533, + "loss": 0.1216, + "step": 36525 + }, + { + "epoch": 0.31706321993732695, + "grad_norm": 0.6171875, + "learning_rate": 0.0014471628933426838, + "loss": 0.1113, + "step": 36526 + }, + { + "epoch": 0.3170719004175311, + "grad_norm": 0.22265625, + "learning_rate": 0.0014471354340432625, + "loss": 0.1221, + "step": 36527 + }, + { + "epoch": 0.3170805808977353, + "grad_norm": 0.68359375, + "learning_rate": 0.0014471079743642193, + "loss": 0.1206, + "step": 36528 + }, + { + "epoch": 0.3170892613779394, + "grad_norm": 0.125, + "learning_rate": 0.001447080514305584, + "loss": 0.0938, + "step": 36529 + }, + { + "epoch": 0.3170979418581436, + "grad_norm": 0.13671875, + "learning_rate": 0.0014470530538673865, + "loss": 0.1377, + "step": 36530 + }, + { + "epoch": 0.31710662233834774, + "grad_norm": 0.11181640625, + "learning_rate": 0.001447025593049657, + "loss": 0.1143, + "step": 36531 + }, + { + "epoch": 0.31711530281855194, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014469981318524256, + "loss": 0.1162, + "step": 36532 + }, + { + "epoch": 0.3171239832987561, + "grad_norm": 0.09765625, + "learning_rate": 0.0014469706702757222, + "loss": 0.1309, + "step": 36533 + }, + { + "epoch": 0.31713266377896027, + "grad_norm": 0.373046875, + "learning_rate": 0.0014469432083195769, + "loss": 0.1035, + "step": 36534 + }, + { + "epoch": 0.3171413442591644, + "grad_norm": 0.12890625, + "learning_rate": 0.0014469157459840194, + "loss": 0.1167, + "step": 36535 + }, + { + "epoch": 0.3171500247393686, + "grad_norm": 0.70703125, + "learning_rate": 0.0014468882832690806, + "loss": 0.0864, + "step": 36536 + }, + { + "epoch": 0.31715870521957273, + "grad_norm": 0.255859375, + "learning_rate": 0.00144686082017479, + "loss": 0.0967, + "step": 36537 + }, + { + "epoch": 0.3171673856997769, + "grad_norm": 0.51953125, + "learning_rate": 0.001446833356701177, + "loss": 0.1084, + "step": 36538 + }, + { + "epoch": 0.31717606617998106, + "grad_norm": 0.17578125, + "learning_rate": 0.0014468058928482728, + "loss": 0.1289, + "step": 36539 + }, + { + "epoch": 0.31718474666018526, + "grad_norm": 0.236328125, + "learning_rate": 0.0014467784286161067, + "loss": 0.0898, + "step": 36540 + }, + { + "epoch": 0.3171934271403894, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014467509640047089, + "loss": 0.1221, + "step": 36541 + }, + { + "epoch": 0.3172021076205936, + "grad_norm": 0.123046875, + "learning_rate": 0.0014467234990141096, + "loss": 0.0815, + "step": 36542 + }, + { + "epoch": 0.3172107881007977, + "grad_norm": 0.130859375, + "learning_rate": 0.0014466960336443385, + "loss": 0.127, + "step": 36543 + }, + { + "epoch": 0.3172194685810019, + "grad_norm": 0.1904296875, + "learning_rate": 0.0014466685678954261, + "loss": 0.1133, + "step": 36544 + }, + { + "epoch": 0.31722814906120606, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014466411017674016, + "loss": 0.125, + "step": 36545 + }, + { + "epoch": 0.31723682954141025, + "grad_norm": 0.408203125, + "learning_rate": 0.001446613635260296, + "loss": 0.104, + "step": 36546 + }, + { + "epoch": 0.3172455100216144, + "grad_norm": 0.1083984375, + "learning_rate": 0.001446586168374139, + "loss": 0.1299, + "step": 36547 + }, + { + "epoch": 0.3172541905018186, + "grad_norm": 0.099609375, + "learning_rate": 0.0014465587011089605, + "loss": 0.0957, + "step": 36548 + }, + { + "epoch": 0.3172628709820227, + "grad_norm": 0.146484375, + "learning_rate": 0.0014465312334647907, + "loss": 0.1055, + "step": 36549 + }, + { + "epoch": 0.3172715514622269, + "grad_norm": 0.25390625, + "learning_rate": 0.0014465037654416595, + "loss": 0.1436, + "step": 36550 + }, + { + "epoch": 0.31728023194243105, + "grad_norm": 0.43359375, + "learning_rate": 0.0014464762970395968, + "loss": 0.1079, + "step": 36551 + }, + { + "epoch": 0.31728891242263524, + "grad_norm": 0.62109375, + "learning_rate": 0.001446448828258633, + "loss": 0.0845, + "step": 36552 + }, + { + "epoch": 0.3172975929028394, + "grad_norm": 0.349609375, + "learning_rate": 0.001446421359098798, + "loss": 0.0791, + "step": 36553 + }, + { + "epoch": 0.31730627338304357, + "grad_norm": 0.1953125, + "learning_rate": 0.0014463938895601213, + "loss": 0.126, + "step": 36554 + }, + { + "epoch": 0.3173149538632477, + "grad_norm": 0.232421875, + "learning_rate": 0.001446366419642634, + "loss": 0.1055, + "step": 36555 + }, + { + "epoch": 0.3173236343434519, + "grad_norm": 0.10888671875, + "learning_rate": 0.0014463389493463655, + "loss": 0.0923, + "step": 36556 + }, + { + "epoch": 0.31733231482365604, + "grad_norm": 0.1396484375, + "learning_rate": 0.001446311478671346, + "loss": 0.084, + "step": 36557 + }, + { + "epoch": 0.31734099530386023, + "grad_norm": 1.0625, + "learning_rate": 0.0014462840076176053, + "loss": 0.1162, + "step": 36558 + }, + { + "epoch": 0.31734967578406437, + "grad_norm": 0.228515625, + "learning_rate": 0.0014462565361851739, + "loss": 0.0903, + "step": 36559 + }, + { + "epoch": 0.31735835626426856, + "grad_norm": 0.58203125, + "learning_rate": 0.0014462290643740812, + "loss": 0.0869, + "step": 36560 + }, + { + "epoch": 0.3173670367444727, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014462015921843578, + "loss": 0.0894, + "step": 36561 + }, + { + "epoch": 0.3173757172246769, + "grad_norm": 0.279296875, + "learning_rate": 0.0014461741196160336, + "loss": 0.082, + "step": 36562 + }, + { + "epoch": 0.317384397704881, + "grad_norm": 0.25390625, + "learning_rate": 0.001446146646669138, + "loss": 0.124, + "step": 36563 + }, + { + "epoch": 0.3173930781850852, + "grad_norm": 0.76953125, + "learning_rate": 0.0014461191733437023, + "loss": 0.1133, + "step": 36564 + }, + { + "epoch": 0.31740175866528936, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014460916996397557, + "loss": 0.0986, + "step": 36565 + }, + { + "epoch": 0.31741043914549355, + "grad_norm": 0.28515625, + "learning_rate": 0.0014460642255573282, + "loss": 0.1152, + "step": 36566 + }, + { + "epoch": 0.3174191196256977, + "grad_norm": 0.349609375, + "learning_rate": 0.0014460367510964506, + "loss": 0.1494, + "step": 36567 + }, + { + "epoch": 0.3174278001059019, + "grad_norm": 0.07421875, + "learning_rate": 0.0014460092762571518, + "loss": 0.0786, + "step": 36568 + }, + { + "epoch": 0.317436480586106, + "grad_norm": 0.318359375, + "learning_rate": 0.0014459818010394627, + "loss": 0.1123, + "step": 36569 + }, + { + "epoch": 0.3174451610663102, + "grad_norm": 0.08837890625, + "learning_rate": 0.001445954325443413, + "loss": 0.0918, + "step": 36570 + }, + { + "epoch": 0.31745384154651435, + "grad_norm": 0.07421875, + "learning_rate": 0.0014459268494690332, + "loss": 0.0703, + "step": 36571 + }, + { + "epoch": 0.31746252202671854, + "grad_norm": 1.7890625, + "learning_rate": 0.0014458993731163525, + "loss": 0.2148, + "step": 36572 + }, + { + "epoch": 0.3174712025069227, + "grad_norm": 0.408203125, + "learning_rate": 0.001445871896385402, + "loss": 0.1006, + "step": 36573 + }, + { + "epoch": 0.31747988298712687, + "grad_norm": 0.55078125, + "learning_rate": 0.0014458444192762105, + "loss": 0.1084, + "step": 36574 + }, + { + "epoch": 0.317488563467331, + "grad_norm": 0.439453125, + "learning_rate": 0.0014458169417888092, + "loss": 0.0918, + "step": 36575 + }, + { + "epoch": 0.3174972439475352, + "grad_norm": 0.080078125, + "learning_rate": 0.0014457894639232276, + "loss": 0.0811, + "step": 36576 + }, + { + "epoch": 0.31750592442773934, + "grad_norm": 0.33203125, + "learning_rate": 0.001445761985679496, + "loss": 0.0718, + "step": 36577 + }, + { + "epoch": 0.31751460490794353, + "grad_norm": 0.2294921875, + "learning_rate": 0.0014457345070576439, + "loss": 0.0928, + "step": 36578 + }, + { + "epoch": 0.31752328538814767, + "grad_norm": 0.421875, + "learning_rate": 0.001445707028057702, + "loss": 0.0874, + "step": 36579 + }, + { + "epoch": 0.31753196586835186, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014456795486797, + "loss": 0.0593, + "step": 36580 + }, + { + "epoch": 0.317540646348556, + "grad_norm": 0.0771484375, + "learning_rate": 0.0014456520689236684, + "loss": 0.0796, + "step": 36581 + }, + { + "epoch": 0.3175493268287602, + "grad_norm": 0.97265625, + "learning_rate": 0.0014456245887896364, + "loss": 0.0869, + "step": 36582 + }, + { + "epoch": 0.31755800730896433, + "grad_norm": 0.796875, + "learning_rate": 0.0014455971082776345, + "loss": 0.0752, + "step": 36583 + }, + { + "epoch": 0.3175666877891685, + "grad_norm": 0.56640625, + "learning_rate": 0.001445569627387693, + "loss": 0.0928, + "step": 36584 + }, + { + "epoch": 0.31757536826937266, + "grad_norm": 0.71875, + "learning_rate": 0.0014455421461198417, + "loss": 0.0903, + "step": 36585 + }, + { + "epoch": 0.31758404874957685, + "grad_norm": 0.8046875, + "learning_rate": 0.0014455146644741109, + "loss": 0.1133, + "step": 36586 + }, + { + "epoch": 0.317592729229781, + "grad_norm": 0.09130859375, + "learning_rate": 0.0014454871824505305, + "loss": 0.0728, + "step": 36587 + }, + { + "epoch": 0.3176014097099852, + "grad_norm": 0.09716796875, + "learning_rate": 0.0014454597000491303, + "loss": 0.1064, + "step": 36588 + }, + { + "epoch": 0.3176100901901893, + "grad_norm": 0.306640625, + "learning_rate": 0.0014454322172699405, + "loss": 0.126, + "step": 36589 + }, + { + "epoch": 0.3176187706703935, + "grad_norm": 0.09814453125, + "learning_rate": 0.0014454047341129915, + "loss": 0.0713, + "step": 36590 + }, + { + "epoch": 0.31762745115059765, + "grad_norm": 0.54296875, + "learning_rate": 0.0014453772505783127, + "loss": 0.0928, + "step": 36591 + }, + { + "epoch": 0.31763613163080184, + "grad_norm": 0.1796875, + "learning_rate": 0.0014453497666659348, + "loss": 0.0771, + "step": 36592 + }, + { + "epoch": 0.317644812111006, + "grad_norm": 0.22265625, + "learning_rate": 0.0014453222823758874, + "loss": 0.1416, + "step": 36593 + }, + { + "epoch": 0.3176534925912101, + "grad_norm": 0.3515625, + "learning_rate": 0.0014452947977082007, + "loss": 0.0747, + "step": 36594 + }, + { + "epoch": 0.3176621730714143, + "grad_norm": 0.3515625, + "learning_rate": 0.001445267312662905, + "loss": 0.0908, + "step": 36595 + }, + { + "epoch": 0.31767085355161845, + "grad_norm": 0.26171875, + "learning_rate": 0.0014452398272400302, + "loss": 0.1523, + "step": 36596 + }, + { + "epoch": 0.31767953403182264, + "grad_norm": 0.10009765625, + "learning_rate": 0.0014452123414396064, + "loss": 0.0845, + "step": 36597 + }, + { + "epoch": 0.3176882145120268, + "grad_norm": 0.6015625, + "learning_rate": 0.001445184855261663, + "loss": 0.1118, + "step": 36598 + }, + { + "epoch": 0.31769689499223097, + "grad_norm": 0.11181640625, + "learning_rate": 0.0014451573687062312, + "loss": 0.1191, + "step": 36599 + }, + { + "epoch": 0.3177055754724351, + "grad_norm": 0.171875, + "learning_rate": 0.0014451298817733404, + "loss": 0.103, + "step": 36600 + }, + { + "epoch": 0.3177142559526393, + "grad_norm": 0.388671875, + "learning_rate": 0.0014451023944630206, + "loss": 0.0967, + "step": 36601 + }, + { + "epoch": 0.31772293643284344, + "grad_norm": 0.81640625, + "learning_rate": 0.001445074906775302, + "loss": 0.1104, + "step": 36602 + }, + { + "epoch": 0.31773161691304763, + "grad_norm": 0.361328125, + "learning_rate": 0.0014450474187102146, + "loss": 0.1128, + "step": 36603 + }, + { + "epoch": 0.31774029739325177, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014450199302677884, + "loss": 0.1523, + "step": 36604 + }, + { + "epoch": 0.31774897787345596, + "grad_norm": 0.244140625, + "learning_rate": 0.0014449924414480539, + "loss": 0.0889, + "step": 36605 + }, + { + "epoch": 0.3177576583536601, + "grad_norm": 0.12353515625, + "learning_rate": 0.001444964952251041, + "loss": 0.0981, + "step": 36606 + }, + { + "epoch": 0.3177663388338643, + "grad_norm": 0.89453125, + "learning_rate": 0.0014449374626767794, + "loss": 0.1152, + "step": 36607 + }, + { + "epoch": 0.31777501931406843, + "grad_norm": 0.271484375, + "learning_rate": 0.0014449099727252993, + "loss": 0.0664, + "step": 36608 + }, + { + "epoch": 0.3177836997942726, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014448824823966309, + "loss": 0.1055, + "step": 36609 + }, + { + "epoch": 0.31779238027447676, + "grad_norm": 0.060791015625, + "learning_rate": 0.0014448549916908043, + "loss": 0.0781, + "step": 36610 + }, + { + "epoch": 0.31780106075468095, + "grad_norm": 0.91015625, + "learning_rate": 0.0014448275006078493, + "loss": 0.1348, + "step": 36611 + }, + { + "epoch": 0.3178097412348851, + "grad_norm": 0.25390625, + "learning_rate": 0.0014448000091477962, + "loss": 0.0815, + "step": 36612 + }, + { + "epoch": 0.3178184217150893, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014447725173106747, + "loss": 0.0674, + "step": 36613 + }, + { + "epoch": 0.3178271021952934, + "grad_norm": 0.2734375, + "learning_rate": 0.0014447450250965155, + "loss": 0.1001, + "step": 36614 + }, + { + "epoch": 0.3178357826754976, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014447175325053483, + "loss": 0.105, + "step": 36615 + }, + { + "epoch": 0.31784446315570175, + "grad_norm": 0.255859375, + "learning_rate": 0.0014446900395372031, + "loss": 0.0986, + "step": 36616 + }, + { + "epoch": 0.31785314363590594, + "grad_norm": 0.353515625, + "learning_rate": 0.0014446625461921102, + "loss": 0.0918, + "step": 36617 + }, + { + "epoch": 0.3178618241161101, + "grad_norm": 0.333984375, + "learning_rate": 0.0014446350524700995, + "loss": 0.0957, + "step": 36618 + }, + { + "epoch": 0.3178705045963143, + "grad_norm": 0.330078125, + "learning_rate": 0.001444607558371201, + "loss": 0.0693, + "step": 36619 + }, + { + "epoch": 0.3178791850765184, + "grad_norm": 0.11181640625, + "learning_rate": 0.0014445800638954445, + "loss": 0.0747, + "step": 36620 + }, + { + "epoch": 0.3178878655567226, + "grad_norm": 0.404296875, + "learning_rate": 0.001444552569042861, + "loss": 0.1128, + "step": 36621 + }, + { + "epoch": 0.31789654603692674, + "grad_norm": 0.412109375, + "learning_rate": 0.0014445250738134797, + "loss": 0.123, + "step": 36622 + }, + { + "epoch": 0.31790522651713093, + "grad_norm": 0.36328125, + "learning_rate": 0.0014444975782073309, + "loss": 0.0752, + "step": 36623 + }, + { + "epoch": 0.31791390699733507, + "grad_norm": 0.169921875, + "learning_rate": 0.0014444700822244447, + "loss": 0.0591, + "step": 36624 + }, + { + "epoch": 0.31792258747753926, + "grad_norm": 0.28515625, + "learning_rate": 0.0014444425858648514, + "loss": 0.1035, + "step": 36625 + }, + { + "epoch": 0.3179312679577434, + "grad_norm": 0.466796875, + "learning_rate": 0.0014444150891285807, + "loss": 0.0996, + "step": 36626 + }, + { + "epoch": 0.3179399484379476, + "grad_norm": 0.248046875, + "learning_rate": 0.0014443875920156632, + "loss": 0.1006, + "step": 36627 + }, + { + "epoch": 0.31794862891815173, + "grad_norm": 0.099609375, + "learning_rate": 0.0014443600945261285, + "loss": 0.105, + "step": 36628 + }, + { + "epoch": 0.3179573093983559, + "grad_norm": 0.0908203125, + "learning_rate": 0.0014443325966600063, + "loss": 0.0938, + "step": 36629 + }, + { + "epoch": 0.31796598987856006, + "grad_norm": 0.10595703125, + "learning_rate": 0.0014443050984173276, + "loss": 0.1226, + "step": 36630 + }, + { + "epoch": 0.31797467035876426, + "grad_norm": 0.412109375, + "learning_rate": 0.0014442775997981216, + "loss": 0.1221, + "step": 36631 + }, + { + "epoch": 0.3179833508389684, + "grad_norm": 0.2734375, + "learning_rate": 0.0014442501008024192, + "loss": 0.0991, + "step": 36632 + }, + { + "epoch": 0.3179920313191726, + "grad_norm": 0.1845703125, + "learning_rate": 0.0014442226014302498, + "loss": 0.0962, + "step": 36633 + }, + { + "epoch": 0.3180007117993767, + "grad_norm": 0.52734375, + "learning_rate": 0.001444195101681644, + "loss": 0.0977, + "step": 36634 + }, + { + "epoch": 0.3180093922795809, + "grad_norm": 0.091796875, + "learning_rate": 0.0014441676015566312, + "loss": 0.1206, + "step": 36635 + }, + { + "epoch": 0.31801807275978505, + "grad_norm": 0.66015625, + "learning_rate": 0.0014441401010552423, + "loss": 0.1177, + "step": 36636 + }, + { + "epoch": 0.31802675323998925, + "grad_norm": 0.1552734375, + "learning_rate": 0.0014441126001775067, + "loss": 0.126, + "step": 36637 + }, + { + "epoch": 0.3180354337201934, + "grad_norm": 1.9453125, + "learning_rate": 0.0014440850989234548, + "loss": 0.126, + "step": 36638 + }, + { + "epoch": 0.3180441142003976, + "grad_norm": 0.46484375, + "learning_rate": 0.0014440575972931168, + "loss": 0.123, + "step": 36639 + }, + { + "epoch": 0.3180527946806017, + "grad_norm": 0.138671875, + "learning_rate": 0.0014440300952865224, + "loss": 0.1143, + "step": 36640 + }, + { + "epoch": 0.3180614751608059, + "grad_norm": 0.625, + "learning_rate": 0.0014440025929037018, + "loss": 0.125, + "step": 36641 + }, + { + "epoch": 0.31807015564101004, + "grad_norm": 0.63671875, + "learning_rate": 0.001443975090144685, + "loss": 0.0957, + "step": 36642 + }, + { + "epoch": 0.31807883612121424, + "grad_norm": 1.1640625, + "learning_rate": 0.0014439475870095025, + "loss": 0.1162, + "step": 36643 + }, + { + "epoch": 0.3180875166014184, + "grad_norm": 0.33203125, + "learning_rate": 0.001443920083498184, + "loss": 0.0864, + "step": 36644 + }, + { + "epoch": 0.31809619708162257, + "grad_norm": 0.294921875, + "learning_rate": 0.0014438925796107598, + "loss": 0.1602, + "step": 36645 + }, + { + "epoch": 0.3181048775618267, + "grad_norm": 0.2470703125, + "learning_rate": 0.0014438650753472596, + "loss": 0.1006, + "step": 36646 + }, + { + "epoch": 0.3181135580420309, + "grad_norm": 0.5546875, + "learning_rate": 0.0014438375707077138, + "loss": 0.1045, + "step": 36647 + }, + { + "epoch": 0.31812223852223503, + "grad_norm": 0.2431640625, + "learning_rate": 0.0014438100656921525, + "loss": 0.1235, + "step": 36648 + }, + { + "epoch": 0.3181309190024392, + "grad_norm": 0.345703125, + "learning_rate": 0.0014437825603006056, + "loss": 0.1074, + "step": 36649 + }, + { + "epoch": 0.31813959948264336, + "grad_norm": 0.79296875, + "learning_rate": 0.001443755054533103, + "loss": 0.0718, + "step": 36650 + }, + { + "epoch": 0.31814827996284756, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014437275483896752, + "loss": 0.1475, + "step": 36651 + }, + { + "epoch": 0.3181569604430517, + "grad_norm": 0.20703125, + "learning_rate": 0.0014437000418703525, + "loss": 0.1113, + "step": 36652 + }, + { + "epoch": 0.3181656409232559, + "grad_norm": 0.11962890625, + "learning_rate": 0.0014436725349751642, + "loss": 0.166, + "step": 36653 + }, + { + "epoch": 0.31817432140346, + "grad_norm": 0.3359375, + "learning_rate": 0.0014436450277041408, + "loss": 0.0967, + "step": 36654 + }, + { + "epoch": 0.3181830018836642, + "grad_norm": 0.90234375, + "learning_rate": 0.0014436175200573124, + "loss": 0.0977, + "step": 36655 + }, + { + "epoch": 0.31819168236386836, + "grad_norm": 0.1806640625, + "learning_rate": 0.0014435900120347092, + "loss": 0.0879, + "step": 36656 + }, + { + "epoch": 0.31820036284407255, + "grad_norm": 0.107421875, + "learning_rate": 0.0014435625036363611, + "loss": 0.0933, + "step": 36657 + }, + { + "epoch": 0.3182090433242767, + "grad_norm": 0.2060546875, + "learning_rate": 0.0014435349948622982, + "loss": 0.1064, + "step": 36658 + }, + { + "epoch": 0.3182177238044809, + "grad_norm": 0.236328125, + "learning_rate": 0.0014435074857125504, + "loss": 0.0806, + "step": 36659 + }, + { + "epoch": 0.318226404284685, + "grad_norm": 0.45703125, + "learning_rate": 0.001443479976187148, + "loss": 0.0776, + "step": 36660 + }, + { + "epoch": 0.3182350847648892, + "grad_norm": 0.478515625, + "learning_rate": 0.0014434524662861213, + "loss": 0.1123, + "step": 36661 + }, + { + "epoch": 0.31824376524509335, + "grad_norm": 0.158203125, + "learning_rate": 0.0014434249560095, + "loss": 0.1318, + "step": 36662 + }, + { + "epoch": 0.31825244572529754, + "grad_norm": 0.32421875, + "learning_rate": 0.0014433974453573143, + "loss": 0.0928, + "step": 36663 + }, + { + "epoch": 0.3182611262055017, + "grad_norm": 0.08544921875, + "learning_rate": 0.0014433699343295943, + "loss": 0.0786, + "step": 36664 + }, + { + "epoch": 0.31826980668570587, + "grad_norm": 0.5234375, + "learning_rate": 0.0014433424229263704, + "loss": 0.1279, + "step": 36665 + }, + { + "epoch": 0.31827848716591, + "grad_norm": 0.234375, + "learning_rate": 0.0014433149111476719, + "loss": 0.0977, + "step": 36666 + }, + { + "epoch": 0.3182871676461142, + "grad_norm": 0.216796875, + "learning_rate": 0.0014432873989935295, + "loss": 0.1211, + "step": 36667 + }, + { + "epoch": 0.31829584812631834, + "grad_norm": 0.15625, + "learning_rate": 0.0014432598864639733, + "loss": 0.0767, + "step": 36668 + }, + { + "epoch": 0.31830452860652253, + "grad_norm": 0.15234375, + "learning_rate": 0.0014432323735590333, + "loss": 0.0928, + "step": 36669 + }, + { + "epoch": 0.31831320908672667, + "grad_norm": 0.0966796875, + "learning_rate": 0.0014432048602787393, + "loss": 0.0947, + "step": 36670 + }, + { + "epoch": 0.31832188956693086, + "grad_norm": 0.09619140625, + "learning_rate": 0.001443177346623122, + "loss": 0.0884, + "step": 36671 + }, + { + "epoch": 0.318330570047135, + "grad_norm": 0.326171875, + "learning_rate": 0.0014431498325922104, + "loss": 0.1064, + "step": 36672 + }, + { + "epoch": 0.3183392505273392, + "grad_norm": 0.306640625, + "learning_rate": 0.001443122318186036, + "loss": 0.0923, + "step": 36673 + }, + { + "epoch": 0.3183479310075433, + "grad_norm": 0.384765625, + "learning_rate": 0.0014430948034046278, + "loss": 0.0918, + "step": 36674 + }, + { + "epoch": 0.3183566114877475, + "grad_norm": 0.380859375, + "learning_rate": 0.0014430672882480165, + "loss": 0.0957, + "step": 36675 + }, + { + "epoch": 0.31836529196795166, + "grad_norm": 0.45703125, + "learning_rate": 0.0014430397727162316, + "loss": 0.1299, + "step": 36676 + }, + { + "epoch": 0.31837397244815585, + "grad_norm": 0.60546875, + "learning_rate": 0.0014430122568093043, + "loss": 0.0898, + "step": 36677 + }, + { + "epoch": 0.31838265292836, + "grad_norm": 0.255859375, + "learning_rate": 0.0014429847405272637, + "loss": 0.1162, + "step": 36678 + }, + { + "epoch": 0.3183913334085642, + "grad_norm": 0.232421875, + "learning_rate": 0.0014429572238701394, + "loss": 0.0986, + "step": 36679 + }, + { + "epoch": 0.3184000138887683, + "grad_norm": 0.546875, + "learning_rate": 0.001442929706837963, + "loss": 0.0854, + "step": 36680 + }, + { + "epoch": 0.3184086943689725, + "grad_norm": 0.3515625, + "learning_rate": 0.0014429021894307632, + "loss": 0.1641, + "step": 36681 + }, + { + "epoch": 0.31841737484917665, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014428746716485714, + "loss": 0.0889, + "step": 36682 + }, + { + "epoch": 0.31842605532938084, + "grad_norm": 0.224609375, + "learning_rate": 0.0014428471534914168, + "loss": 0.0889, + "step": 36683 + }, + { + "epoch": 0.318434735809585, + "grad_norm": 0.2421875, + "learning_rate": 0.0014428196349593296, + "loss": 0.0942, + "step": 36684 + }, + { + "epoch": 0.31844341628978917, + "grad_norm": 0.375, + "learning_rate": 0.00144279211605234, + "loss": 0.0864, + "step": 36685 + }, + { + "epoch": 0.3184520967699933, + "grad_norm": 0.376953125, + "learning_rate": 0.0014427645967704781, + "loss": 0.0947, + "step": 36686 + }, + { + "epoch": 0.3184607772501975, + "grad_norm": 0.400390625, + "learning_rate": 0.0014427370771137743, + "loss": 0.0986, + "step": 36687 + }, + { + "epoch": 0.31846945773040164, + "grad_norm": 0.609375, + "learning_rate": 0.0014427095570822576, + "loss": 0.1035, + "step": 36688 + }, + { + "epoch": 0.31847813821060583, + "grad_norm": 0.28515625, + "learning_rate": 0.0014426820366759596, + "loss": 0.1084, + "step": 36689 + }, + { + "epoch": 0.31848681869080997, + "grad_norm": 0.419921875, + "learning_rate": 0.0014426545158949093, + "loss": 0.1348, + "step": 36690 + }, + { + "epoch": 0.31849549917101416, + "grad_norm": 0.76171875, + "learning_rate": 0.0014426269947391374, + "loss": 0.1289, + "step": 36691 + }, + { + "epoch": 0.3185041796512183, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014425994732086738, + "loss": 0.1328, + "step": 36692 + }, + { + "epoch": 0.3185128601314225, + "grad_norm": 0.44921875, + "learning_rate": 0.0014425719513035482, + "loss": 0.1123, + "step": 36693 + }, + { + "epoch": 0.31852154061162663, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014425444290237916, + "loss": 0.083, + "step": 36694 + }, + { + "epoch": 0.3185302210918308, + "grad_norm": 0.265625, + "learning_rate": 0.0014425169063694333, + "loss": 0.1309, + "step": 36695 + }, + { + "epoch": 0.31853890157203496, + "grad_norm": 0.4296875, + "learning_rate": 0.0014424893833405036, + "loss": 0.1299, + "step": 36696 + }, + { + "epoch": 0.31854758205223915, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014424618599370325, + "loss": 0.0908, + "step": 36697 + }, + { + "epoch": 0.3185562625324433, + "grad_norm": 0.1806640625, + "learning_rate": 0.0014424343361590505, + "loss": 0.0815, + "step": 36698 + }, + { + "epoch": 0.3185649430126475, + "grad_norm": 0.330078125, + "learning_rate": 0.0014424068120065872, + "loss": 0.0938, + "step": 36699 + }, + { + "epoch": 0.3185736234928516, + "grad_norm": 0.09521484375, + "learning_rate": 0.0014423792874796732, + "loss": 0.1094, + "step": 36700 + }, + { + "epoch": 0.3185823039730558, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014423517625783383, + "loss": 0.105, + "step": 36701 + }, + { + "epoch": 0.31859098445325995, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014423242373026127, + "loss": 0.0996, + "step": 36702 + }, + { + "epoch": 0.31859966493346414, + "grad_norm": 0.53515625, + "learning_rate": 0.0014422967116525262, + "loss": 0.0991, + "step": 36703 + }, + { + "epoch": 0.3186083454136683, + "grad_norm": 0.5625, + "learning_rate": 0.0014422691856281096, + "loss": 0.1299, + "step": 36704 + }, + { + "epoch": 0.3186170258938725, + "grad_norm": 0.376953125, + "learning_rate": 0.0014422416592293924, + "loss": 0.1123, + "step": 36705 + }, + { + "epoch": 0.3186257063740766, + "grad_norm": 0.4140625, + "learning_rate": 0.0014422141324564046, + "loss": 0.1387, + "step": 36706 + }, + { + "epoch": 0.3186343868542808, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014421866053091767, + "loss": 0.1094, + "step": 36707 + }, + { + "epoch": 0.31864306733448494, + "grad_norm": 0.3359375, + "learning_rate": 0.0014421590777877389, + "loss": 0.1138, + "step": 36708 + }, + { + "epoch": 0.31865174781468913, + "grad_norm": 0.212890625, + "learning_rate": 0.001442131549892121, + "loss": 0.1045, + "step": 36709 + }, + { + "epoch": 0.31866042829489327, + "grad_norm": 0.54296875, + "learning_rate": 0.0014421040216223528, + "loss": 0.1504, + "step": 36710 + }, + { + "epoch": 0.31866910877509746, + "grad_norm": 0.12890625, + "learning_rate": 0.0014420764929784652, + "loss": 0.0859, + "step": 36711 + }, + { + "epoch": 0.3186777892553016, + "grad_norm": 0.40625, + "learning_rate": 0.0014420489639604876, + "loss": 0.0825, + "step": 36712 + }, + { + "epoch": 0.3186864697355058, + "grad_norm": 0.1904296875, + "learning_rate": 0.0014420214345684506, + "loss": 0.0894, + "step": 36713 + }, + { + "epoch": 0.31869515021570993, + "grad_norm": 0.197265625, + "learning_rate": 0.001441993904802384, + "loss": 0.1045, + "step": 36714 + }, + { + "epoch": 0.3187038306959141, + "grad_norm": 0.08642578125, + "learning_rate": 0.0014419663746623181, + "loss": 0.0928, + "step": 36715 + }, + { + "epoch": 0.31871251117611826, + "grad_norm": 0.73046875, + "learning_rate": 0.0014419388441482828, + "loss": 0.1143, + "step": 36716 + }, + { + "epoch": 0.3187211916563224, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014419113132603085, + "loss": 0.0986, + "step": 36717 + }, + { + "epoch": 0.3187298721365266, + "grad_norm": 0.271484375, + "learning_rate": 0.001441883781998425, + "loss": 0.1064, + "step": 36718 + }, + { + "epoch": 0.31873855261673073, + "grad_norm": 0.1650390625, + "learning_rate": 0.0014418562503626625, + "loss": 0.1084, + "step": 36719 + }, + { + "epoch": 0.3187472330969349, + "grad_norm": 0.388671875, + "learning_rate": 0.001441828718353051, + "loss": 0.1177, + "step": 36720 + }, + { + "epoch": 0.31875591357713906, + "grad_norm": 0.248046875, + "learning_rate": 0.0014418011859696211, + "loss": 0.0874, + "step": 36721 + }, + { + "epoch": 0.31876459405734325, + "grad_norm": 0.44140625, + "learning_rate": 0.0014417736532124025, + "loss": 0.123, + "step": 36722 + }, + { + "epoch": 0.3187732745375474, + "grad_norm": 0.66015625, + "learning_rate": 0.001441746120081425, + "loss": 0.1118, + "step": 36723 + }, + { + "epoch": 0.3187819550177516, + "grad_norm": 0.11328125, + "learning_rate": 0.0014417185865767193, + "loss": 0.1118, + "step": 36724 + }, + { + "epoch": 0.3187906354979557, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014416910526983153, + "loss": 0.1211, + "step": 36725 + }, + { + "epoch": 0.3187993159781599, + "grad_norm": 0.310546875, + "learning_rate": 0.0014416635184462431, + "loss": 0.1035, + "step": 36726 + }, + { + "epoch": 0.31880799645836405, + "grad_norm": 0.5546875, + "learning_rate": 0.001441635983820533, + "loss": 0.0898, + "step": 36727 + }, + { + "epoch": 0.31881667693856824, + "grad_norm": 0.1865234375, + "learning_rate": 0.0014416084488212144, + "loss": 0.0752, + "step": 36728 + }, + { + "epoch": 0.3188253574187724, + "grad_norm": 0.30859375, + "learning_rate": 0.0014415809134483188, + "loss": 0.1348, + "step": 36729 + }, + { + "epoch": 0.3188340378989766, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014415533777018745, + "loss": 0.1348, + "step": 36730 + }, + { + "epoch": 0.3188427183791807, + "grad_norm": 0.51953125, + "learning_rate": 0.0014415258415819132, + "loss": 0.1206, + "step": 36731 + }, + { + "epoch": 0.3188513988593849, + "grad_norm": 0.287109375, + "learning_rate": 0.001441498305088464, + "loss": 0.0986, + "step": 36732 + }, + { + "epoch": 0.31886007933958904, + "grad_norm": 0.1767578125, + "learning_rate": 0.0014414707682215576, + "loss": 0.1201, + "step": 36733 + }, + { + "epoch": 0.31886875981979323, + "grad_norm": 0.2734375, + "learning_rate": 0.0014414432309812237, + "loss": 0.1123, + "step": 36734 + }, + { + "epoch": 0.3188774402999974, + "grad_norm": 0.197265625, + "learning_rate": 0.0014414156933674931, + "loss": 0.061, + "step": 36735 + }, + { + "epoch": 0.31888612078020157, + "grad_norm": 0.1787109375, + "learning_rate": 0.001441388155380395, + "loss": 0.0723, + "step": 36736 + }, + { + "epoch": 0.3188948012604057, + "grad_norm": 0.1875, + "learning_rate": 0.00144136061701996, + "loss": 0.1182, + "step": 36737 + }, + { + "epoch": 0.3189034817406099, + "grad_norm": 0.09765625, + "learning_rate": 0.001441333078286218, + "loss": 0.1167, + "step": 36738 + }, + { + "epoch": 0.31891216222081403, + "grad_norm": 0.138671875, + "learning_rate": 0.0014413055391791995, + "loss": 0.0938, + "step": 36739 + }, + { + "epoch": 0.3189208427010182, + "grad_norm": 0.46484375, + "learning_rate": 0.0014412779996989345, + "loss": 0.1348, + "step": 36740 + }, + { + "epoch": 0.31892952318122236, + "grad_norm": 0.26953125, + "learning_rate": 0.0014412504598454526, + "loss": 0.0762, + "step": 36741 + }, + { + "epoch": 0.31893820366142656, + "grad_norm": 0.212890625, + "learning_rate": 0.0014412229196187846, + "loss": 0.1089, + "step": 36742 + }, + { + "epoch": 0.3189468841416307, + "grad_norm": 0.228515625, + "learning_rate": 0.0014411953790189605, + "loss": 0.1123, + "step": 36743 + }, + { + "epoch": 0.3189555646218349, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014411678380460101, + "loss": 0.0967, + "step": 36744 + }, + { + "epoch": 0.318964245102039, + "grad_norm": 0.09130859375, + "learning_rate": 0.0014411402966999635, + "loss": 0.085, + "step": 36745 + }, + { + "epoch": 0.3189729255822432, + "grad_norm": 0.439453125, + "learning_rate": 0.001441112754980851, + "loss": 0.123, + "step": 36746 + }, + { + "epoch": 0.31898160606244735, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014410852128887031, + "loss": 0.0879, + "step": 36747 + }, + { + "epoch": 0.31899028654265155, + "grad_norm": 0.150390625, + "learning_rate": 0.0014410576704235493, + "loss": 0.0806, + "step": 36748 + }, + { + "epoch": 0.3189989670228557, + "grad_norm": 0.71875, + "learning_rate": 0.00144103012758542, + "loss": 0.1016, + "step": 36749 + }, + { + "epoch": 0.3190076475030599, + "grad_norm": 0.859375, + "learning_rate": 0.0014410025843743447, + "loss": 0.1108, + "step": 36750 + }, + { + "epoch": 0.319016327983264, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014409750407903548, + "loss": 0.1001, + "step": 36751 + }, + { + "epoch": 0.3190250084634682, + "grad_norm": 0.267578125, + "learning_rate": 0.0014409474968334797, + "loss": 0.127, + "step": 36752 + }, + { + "epoch": 0.31903368894367234, + "grad_norm": 0.341796875, + "learning_rate": 0.0014409199525037496, + "loss": 0.1123, + "step": 36753 + }, + { + "epoch": 0.31904236942387654, + "grad_norm": 0.07568359375, + "learning_rate": 0.0014408924078011941, + "loss": 0.0752, + "step": 36754 + }, + { + "epoch": 0.3190510499040807, + "grad_norm": 0.4765625, + "learning_rate": 0.0014408648627258441, + "loss": 0.0962, + "step": 36755 + }, + { + "epoch": 0.31905973038428487, + "grad_norm": 0.171875, + "learning_rate": 0.0014408373172777292, + "loss": 0.0918, + "step": 36756 + }, + { + "epoch": 0.319068410864489, + "grad_norm": 0.130859375, + "learning_rate": 0.00144080977145688, + "loss": 0.1143, + "step": 36757 + }, + { + "epoch": 0.3190770913446932, + "grad_norm": 1.546875, + "learning_rate": 0.001440782225263326, + "loss": 0.4395, + "step": 36758 + }, + { + "epoch": 0.31908577182489734, + "grad_norm": 0.333984375, + "learning_rate": 0.001440754678697098, + "loss": 0.0889, + "step": 36759 + }, + { + "epoch": 0.31909445230510153, + "grad_norm": 0.373046875, + "learning_rate": 0.001440727131758226, + "loss": 0.1211, + "step": 36760 + }, + { + "epoch": 0.31910313278530567, + "grad_norm": 0.38671875, + "learning_rate": 0.0014406995844467396, + "loss": 0.0957, + "step": 36761 + }, + { + "epoch": 0.31911181326550986, + "grad_norm": 0.296875, + "learning_rate": 0.0014406720367626693, + "loss": 0.1021, + "step": 36762 + }, + { + "epoch": 0.319120493745714, + "grad_norm": 0.67578125, + "learning_rate": 0.001440644488706045, + "loss": 0.0947, + "step": 36763 + }, + { + "epoch": 0.3191291742259182, + "grad_norm": 0.52734375, + "learning_rate": 0.0014406169402768972, + "loss": 0.0933, + "step": 36764 + }, + { + "epoch": 0.3191378547061223, + "grad_norm": 0.2294921875, + "learning_rate": 0.0014405893914752558, + "loss": 0.1182, + "step": 36765 + }, + { + "epoch": 0.3191465351863265, + "grad_norm": 0.32421875, + "learning_rate": 0.0014405618423011508, + "loss": 0.0806, + "step": 36766 + }, + { + "epoch": 0.31915521566653066, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014405342927546125, + "loss": 0.0947, + "step": 36767 + }, + { + "epoch": 0.31916389614673485, + "grad_norm": 0.2373046875, + "learning_rate": 0.0014405067428356715, + "loss": 0.1123, + "step": 36768 + }, + { + "epoch": 0.319172576626939, + "grad_norm": 0.85546875, + "learning_rate": 0.001440479192544357, + "loss": 0.1377, + "step": 36769 + }, + { + "epoch": 0.3191812571071432, + "grad_norm": 0.06884765625, + "learning_rate": 0.0014404516418806996, + "loss": 0.083, + "step": 36770 + }, + { + "epoch": 0.3191899375873473, + "grad_norm": 0.283203125, + "learning_rate": 0.0014404240908447296, + "loss": 0.0977, + "step": 36771 + }, + { + "epoch": 0.3191986180675515, + "grad_norm": 0.177734375, + "learning_rate": 0.0014403965394364769, + "loss": 0.1016, + "step": 36772 + }, + { + "epoch": 0.31920729854775565, + "grad_norm": 0.44140625, + "learning_rate": 0.0014403689876559714, + "loss": 0.0723, + "step": 36773 + }, + { + "epoch": 0.31921597902795984, + "grad_norm": 0.333984375, + "learning_rate": 0.0014403414355032439, + "loss": 0.1079, + "step": 36774 + }, + { + "epoch": 0.319224659508164, + "grad_norm": 0.40234375, + "learning_rate": 0.0014403138829783238, + "loss": 0.1016, + "step": 36775 + }, + { + "epoch": 0.31923333998836817, + "grad_norm": 0.494140625, + "learning_rate": 0.0014402863300812415, + "loss": 0.1084, + "step": 36776 + }, + { + "epoch": 0.3192420204685723, + "grad_norm": 0.23046875, + "learning_rate": 0.0014402587768120273, + "loss": 0.0967, + "step": 36777 + }, + { + "epoch": 0.3192507009487765, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014402312231707112, + "loss": 0.1128, + "step": 36778 + }, + { + "epoch": 0.31925938142898064, + "grad_norm": 0.1943359375, + "learning_rate": 0.0014402036691573234, + "loss": 0.1328, + "step": 36779 + }, + { + "epoch": 0.31926806190918483, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014401761147718939, + "loss": 0.0923, + "step": 36780 + }, + { + "epoch": 0.31927674238938897, + "grad_norm": 0.482421875, + "learning_rate": 0.001440148560014453, + "loss": 0.1245, + "step": 36781 + }, + { + "epoch": 0.31928542286959316, + "grad_norm": 0.48046875, + "learning_rate": 0.001440121004885031, + "loss": 0.084, + "step": 36782 + }, + { + "epoch": 0.3192941033497973, + "grad_norm": 0.06298828125, + "learning_rate": 0.0014400934493836572, + "loss": 0.0649, + "step": 36783 + }, + { + "epoch": 0.3193027838300015, + "grad_norm": 1.1484375, + "learning_rate": 0.0014400658935103626, + "loss": 0.1309, + "step": 36784 + }, + { + "epoch": 0.31931146431020563, + "grad_norm": 0.12890625, + "learning_rate": 0.0014400383372651768, + "loss": 0.1279, + "step": 36785 + }, + { + "epoch": 0.3193201447904098, + "grad_norm": 0.4921875, + "learning_rate": 0.0014400107806481306, + "loss": 0.0942, + "step": 36786 + }, + { + "epoch": 0.31932882527061396, + "grad_norm": 0.55078125, + "learning_rate": 0.0014399832236592536, + "loss": 0.1406, + "step": 36787 + }, + { + "epoch": 0.31933750575081815, + "grad_norm": 0.10302734375, + "learning_rate": 0.001439955666298576, + "loss": 0.0737, + "step": 36788 + }, + { + "epoch": 0.3193461862310223, + "grad_norm": 0.177734375, + "learning_rate": 0.001439928108566128, + "loss": 0.083, + "step": 36789 + }, + { + "epoch": 0.3193548667112265, + "grad_norm": 0.30859375, + "learning_rate": 0.0014399005504619397, + "loss": 0.1035, + "step": 36790 + }, + { + "epoch": 0.3193635471914306, + "grad_norm": 0.50390625, + "learning_rate": 0.0014398729919860413, + "loss": 0.123, + "step": 36791 + }, + { + "epoch": 0.3193722276716348, + "grad_norm": 0.1689453125, + "learning_rate": 0.0014398454331384632, + "loss": 0.0977, + "step": 36792 + }, + { + "epoch": 0.31938090815183895, + "grad_norm": 0.2890625, + "learning_rate": 0.0014398178739192346, + "loss": 0.1099, + "step": 36793 + }, + { + "epoch": 0.31938958863204314, + "grad_norm": 0.201171875, + "learning_rate": 0.0014397903143283867, + "loss": 0.1455, + "step": 36794 + }, + { + "epoch": 0.3193982691122473, + "grad_norm": 0.326171875, + "learning_rate": 0.001439762754365949, + "loss": 0.1445, + "step": 36795 + }, + { + "epoch": 0.3194069495924515, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014397351940319521, + "loss": 0.1113, + "step": 36796 + }, + { + "epoch": 0.3194156300726556, + "grad_norm": 0.27734375, + "learning_rate": 0.0014397076333264257, + "loss": 0.1348, + "step": 36797 + }, + { + "epoch": 0.3194243105528598, + "grad_norm": 0.07421875, + "learning_rate": 0.0014396800722494, + "loss": 0.083, + "step": 36798 + }, + { + "epoch": 0.31943299103306394, + "grad_norm": 0.07177734375, + "learning_rate": 0.0014396525108009057, + "loss": 0.061, + "step": 36799 + }, + { + "epoch": 0.31944167151326813, + "grad_norm": 0.1796875, + "learning_rate": 0.0014396249489809724, + "loss": 0.0923, + "step": 36800 + }, + { + "epoch": 0.31945035199347227, + "grad_norm": 0.1201171875, + "learning_rate": 0.0014395973867896304, + "loss": 0.0996, + "step": 36801 + }, + { + "epoch": 0.31945903247367646, + "grad_norm": 1.0703125, + "learning_rate": 0.0014395698242269095, + "loss": 0.1094, + "step": 36802 + }, + { + "epoch": 0.3194677129538806, + "grad_norm": 0.08349609375, + "learning_rate": 0.0014395422612928405, + "loss": 0.1084, + "step": 36803 + }, + { + "epoch": 0.3194763934340848, + "grad_norm": 0.5, + "learning_rate": 0.0014395146979874528, + "loss": 0.0835, + "step": 36804 + }, + { + "epoch": 0.31948507391428893, + "grad_norm": 0.140625, + "learning_rate": 0.001439487134310777, + "loss": 0.1406, + "step": 36805 + }, + { + "epoch": 0.3194937543944931, + "grad_norm": 0.099609375, + "learning_rate": 0.0014394595702628432, + "loss": 0.1055, + "step": 36806 + }, + { + "epoch": 0.31950243487469726, + "grad_norm": 0.451171875, + "learning_rate": 0.001439432005843682, + "loss": 0.0791, + "step": 36807 + }, + { + "epoch": 0.31951111535490145, + "grad_norm": 0.76171875, + "learning_rate": 0.0014394044410533224, + "loss": 0.1113, + "step": 36808 + }, + { + "epoch": 0.3195197958351056, + "grad_norm": 0.29296875, + "learning_rate": 0.0014393768758917955, + "loss": 0.0957, + "step": 36809 + }, + { + "epoch": 0.3195284763153098, + "grad_norm": 0.3046875, + "learning_rate": 0.001439349310359131, + "loss": 0.1089, + "step": 36810 + }, + { + "epoch": 0.3195371567955139, + "grad_norm": 0.3515625, + "learning_rate": 0.0014393217444553593, + "loss": 0.0747, + "step": 36811 + }, + { + "epoch": 0.3195458372757181, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014392941781805105, + "loss": 0.0918, + "step": 36812 + }, + { + "epoch": 0.31955451775592225, + "grad_norm": 0.380859375, + "learning_rate": 0.0014392666115346144, + "loss": 0.0962, + "step": 36813 + }, + { + "epoch": 0.31956319823612644, + "grad_norm": 0.208984375, + "learning_rate": 0.001439239044517702, + "loss": 0.0967, + "step": 36814 + }, + { + "epoch": 0.3195718787163306, + "grad_norm": 0.734375, + "learning_rate": 0.0014392114771298026, + "loss": 0.1582, + "step": 36815 + }, + { + "epoch": 0.3195805591965348, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014391839093709465, + "loss": 0.0601, + "step": 36816 + }, + { + "epoch": 0.3195892396767389, + "grad_norm": 0.482421875, + "learning_rate": 0.0014391563412411638, + "loss": 0.0991, + "step": 36817 + }, + { + "epoch": 0.3195979201569431, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014391287727404853, + "loss": 0.1191, + "step": 36818 + }, + { + "epoch": 0.31960660063714724, + "grad_norm": 0.1767578125, + "learning_rate": 0.00143910120386894, + "loss": 0.1152, + "step": 36819 + }, + { + "epoch": 0.31961528111735144, + "grad_norm": 0.470703125, + "learning_rate": 0.0014390736346265596, + "loss": 0.1201, + "step": 36820 + }, + { + "epoch": 0.3196239615975556, + "grad_norm": 0.107421875, + "learning_rate": 0.0014390460650133728, + "loss": 0.1035, + "step": 36821 + }, + { + "epoch": 0.31963264207775977, + "grad_norm": 0.09521484375, + "learning_rate": 0.0014390184950294107, + "loss": 0.1143, + "step": 36822 + }, + { + "epoch": 0.3196413225579639, + "grad_norm": 0.396484375, + "learning_rate": 0.0014389909246747027, + "loss": 0.0728, + "step": 36823 + }, + { + "epoch": 0.3196500030381681, + "grad_norm": 0.380859375, + "learning_rate": 0.001438963353949279, + "loss": 0.0835, + "step": 36824 + }, + { + "epoch": 0.31965868351837223, + "grad_norm": 2.75, + "learning_rate": 0.0014389357828531707, + "loss": 0.1445, + "step": 36825 + }, + { + "epoch": 0.3196673639985764, + "grad_norm": 0.1259765625, + "learning_rate": 0.001438908211386407, + "loss": 0.0908, + "step": 36826 + }, + { + "epoch": 0.31967604447878056, + "grad_norm": 0.17578125, + "learning_rate": 0.0014388806395490186, + "loss": 0.127, + "step": 36827 + }, + { + "epoch": 0.31968472495898476, + "grad_norm": 0.333984375, + "learning_rate": 0.0014388530673410352, + "loss": 0.0859, + "step": 36828 + }, + { + "epoch": 0.3196934054391889, + "grad_norm": 0.353515625, + "learning_rate": 0.0014388254947624874, + "loss": 0.106, + "step": 36829 + }, + { + "epoch": 0.3197020859193931, + "grad_norm": 0.330078125, + "learning_rate": 0.001438797921813405, + "loss": 0.1143, + "step": 36830 + }, + { + "epoch": 0.3197107663995972, + "grad_norm": 0.421875, + "learning_rate": 0.0014387703484938187, + "loss": 0.1504, + "step": 36831 + }, + { + "epoch": 0.3197194468798014, + "grad_norm": 0.228515625, + "learning_rate": 0.0014387427748037576, + "loss": 0.0957, + "step": 36832 + }, + { + "epoch": 0.31972812736000555, + "grad_norm": 0.58984375, + "learning_rate": 0.0014387152007432524, + "loss": 0.1748, + "step": 36833 + }, + { + "epoch": 0.31973680784020975, + "grad_norm": 0.271484375, + "learning_rate": 0.001438687626312334, + "loss": 0.1084, + "step": 36834 + }, + { + "epoch": 0.3197454883204139, + "grad_norm": 0.375, + "learning_rate": 0.0014386600515110316, + "loss": 0.1211, + "step": 36835 + }, + { + "epoch": 0.3197541688006181, + "grad_norm": 0.21875, + "learning_rate": 0.0014386324763393755, + "loss": 0.1006, + "step": 36836 + }, + { + "epoch": 0.3197628492808222, + "grad_norm": 0.453125, + "learning_rate": 0.001438604900797396, + "loss": 0.0752, + "step": 36837 + }, + { + "epoch": 0.3197715297610264, + "grad_norm": 0.306640625, + "learning_rate": 0.0014385773248851235, + "loss": 0.0928, + "step": 36838 + }, + { + "epoch": 0.31978021024123054, + "grad_norm": 0.3203125, + "learning_rate": 0.001438549748602588, + "loss": 0.0649, + "step": 36839 + }, + { + "epoch": 0.3197888907214347, + "grad_norm": 0.345703125, + "learning_rate": 0.0014385221719498192, + "loss": 0.0977, + "step": 36840 + }, + { + "epoch": 0.3197975712016389, + "grad_norm": 0.7890625, + "learning_rate": 0.001438494594926848, + "loss": 0.0967, + "step": 36841 + }, + { + "epoch": 0.319806251681843, + "grad_norm": 0.8125, + "learning_rate": 0.001438467017533704, + "loss": 0.1475, + "step": 36842 + }, + { + "epoch": 0.3198149321620472, + "grad_norm": 0.349609375, + "learning_rate": 0.0014384394397704175, + "loss": 0.1338, + "step": 36843 + }, + { + "epoch": 0.31982361264225134, + "grad_norm": 0.13671875, + "learning_rate": 0.001438411861637019, + "loss": 0.105, + "step": 36844 + }, + { + "epoch": 0.31983229312245554, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014383842831335381, + "loss": 0.0977, + "step": 36845 + }, + { + "epoch": 0.3198409736026597, + "grad_norm": 0.5625, + "learning_rate": 0.0014383567042600051, + "loss": 0.1191, + "step": 36846 + }, + { + "epoch": 0.31984965408286387, + "grad_norm": 0.267578125, + "learning_rate": 0.0014383291250164507, + "loss": 0.0986, + "step": 36847 + }, + { + "epoch": 0.319858334563068, + "grad_norm": 0.302734375, + "learning_rate": 0.0014383015454029044, + "loss": 0.1191, + "step": 36848 + }, + { + "epoch": 0.3198670150432722, + "grad_norm": 0.1484375, + "learning_rate": 0.0014382739654193969, + "loss": 0.0859, + "step": 36849 + }, + { + "epoch": 0.31987569552347633, + "grad_norm": 0.1640625, + "learning_rate": 0.0014382463850659576, + "loss": 0.0913, + "step": 36850 + }, + { + "epoch": 0.3198843760036805, + "grad_norm": 0.6015625, + "learning_rate": 0.0014382188043426175, + "loss": 0.1143, + "step": 36851 + }, + { + "epoch": 0.31989305648388466, + "grad_norm": 1.0703125, + "learning_rate": 0.001438191223249406, + "loss": 0.1299, + "step": 36852 + }, + { + "epoch": 0.31990173696408886, + "grad_norm": 0.66796875, + "learning_rate": 0.001438163641786354, + "loss": 0.126, + "step": 36853 + }, + { + "epoch": 0.319910417444293, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014381360599534914, + "loss": 0.0928, + "step": 36854 + }, + { + "epoch": 0.3199190979244972, + "grad_norm": 0.326171875, + "learning_rate": 0.0014381084777508482, + "loss": 0.0938, + "step": 36855 + }, + { + "epoch": 0.3199277784047013, + "grad_norm": 0.51953125, + "learning_rate": 0.0014380808951784546, + "loss": 0.0791, + "step": 36856 + }, + { + "epoch": 0.3199364588849055, + "grad_norm": 1.1484375, + "learning_rate": 0.0014380533122363408, + "loss": 0.1504, + "step": 36857 + }, + { + "epoch": 0.31994513936510965, + "grad_norm": 0.4609375, + "learning_rate": 0.0014380257289245372, + "loss": 0.126, + "step": 36858 + }, + { + "epoch": 0.31995381984531385, + "grad_norm": 0.400390625, + "learning_rate": 0.0014379981452430732, + "loss": 0.0864, + "step": 36859 + }, + { + "epoch": 0.319962500325518, + "grad_norm": 0.44140625, + "learning_rate": 0.00143797056119198, + "loss": 0.1143, + "step": 36860 + }, + { + "epoch": 0.3199711808057222, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014379429767712871, + "loss": 0.0908, + "step": 36861 + }, + { + "epoch": 0.3199798612859263, + "grad_norm": 0.470703125, + "learning_rate": 0.0014379153919810246, + "loss": 0.1099, + "step": 36862 + }, + { + "epoch": 0.3199885417661305, + "grad_norm": 0.2890625, + "learning_rate": 0.001437887806821223, + "loss": 0.1348, + "step": 36863 + }, + { + "epoch": 0.31999722224633464, + "grad_norm": 0.173828125, + "learning_rate": 0.0014378602212919128, + "loss": 0.1182, + "step": 36864 + }, + { + "epoch": 0.32000590272653884, + "grad_norm": 0.2265625, + "learning_rate": 0.0014378326353931235, + "loss": 0.0884, + "step": 36865 + }, + { + "epoch": 0.320014583206743, + "grad_norm": 0.10888671875, + "learning_rate": 0.0014378050491248853, + "loss": 0.1064, + "step": 36866 + }, + { + "epoch": 0.32002326368694717, + "grad_norm": 0.51171875, + "learning_rate": 0.0014377774624872287, + "loss": 0.126, + "step": 36867 + }, + { + "epoch": 0.3200319441671513, + "grad_norm": 0.275390625, + "learning_rate": 0.0014377498754801841, + "loss": 0.1055, + "step": 36868 + }, + { + "epoch": 0.3200406246473555, + "grad_norm": 0.359375, + "learning_rate": 0.0014377222881037809, + "loss": 0.1152, + "step": 36869 + }, + { + "epoch": 0.32004930512755964, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014376947003580495, + "loss": 0.084, + "step": 36870 + }, + { + "epoch": 0.32005798560776383, + "grad_norm": 0.111328125, + "learning_rate": 0.0014376671122430208, + "loss": 0.0938, + "step": 36871 + }, + { + "epoch": 0.32006666608796797, + "grad_norm": 0.15625, + "learning_rate": 0.001437639523758724, + "loss": 0.1006, + "step": 36872 + }, + { + "epoch": 0.32007534656817216, + "grad_norm": 0.330078125, + "learning_rate": 0.00143761193490519, + "loss": 0.1138, + "step": 36873 + }, + { + "epoch": 0.3200840270483763, + "grad_norm": 0.189453125, + "learning_rate": 0.0014375843456824484, + "loss": 0.125, + "step": 36874 + }, + { + "epoch": 0.3200927075285805, + "grad_norm": 0.23046875, + "learning_rate": 0.0014375567560905298, + "loss": 0.123, + "step": 36875 + }, + { + "epoch": 0.3201013880087846, + "grad_norm": 0.283203125, + "learning_rate": 0.001437529166129464, + "loss": 0.1216, + "step": 36876 + }, + { + "epoch": 0.3201100684889888, + "grad_norm": 0.455078125, + "learning_rate": 0.0014375015757992818, + "loss": 0.085, + "step": 36877 + }, + { + "epoch": 0.32011874896919296, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014374739851000127, + "loss": 0.0947, + "step": 36878 + }, + { + "epoch": 0.32012742944939715, + "grad_norm": 0.12255859375, + "learning_rate": 0.0014374463940316866, + "loss": 0.0918, + "step": 36879 + }, + { + "epoch": 0.3201361099296013, + "grad_norm": 0.208984375, + "learning_rate": 0.001437418802594335, + "loss": 0.0811, + "step": 36880 + }, + { + "epoch": 0.3201447904098055, + "grad_norm": 0.322265625, + "learning_rate": 0.0014373912107879868, + "loss": 0.0835, + "step": 36881 + }, + { + "epoch": 0.3201534708900096, + "grad_norm": 0.44140625, + "learning_rate": 0.001437363618612673, + "loss": 0.125, + "step": 36882 + }, + { + "epoch": 0.3201621513702138, + "grad_norm": 0.47265625, + "learning_rate": 0.001437336026068423, + "loss": 0.1069, + "step": 36883 + }, + { + "epoch": 0.32017083185041795, + "grad_norm": 0.27734375, + "learning_rate": 0.0014373084331552675, + "loss": 0.1211, + "step": 36884 + }, + { + "epoch": 0.32017951233062214, + "grad_norm": 0.1640625, + "learning_rate": 0.0014372808398732367, + "loss": 0.0972, + "step": 36885 + }, + { + "epoch": 0.3201881928108263, + "grad_norm": 0.451171875, + "learning_rate": 0.0014372532462223605, + "loss": 0.168, + "step": 36886 + }, + { + "epoch": 0.32019687329103047, + "grad_norm": 0.271484375, + "learning_rate": 0.0014372256522026693, + "loss": 0.1221, + "step": 36887 + }, + { + "epoch": 0.3202055537712346, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014371980578141933, + "loss": 0.0791, + "step": 36888 + }, + { + "epoch": 0.3202142342514388, + "grad_norm": 0.2060546875, + "learning_rate": 0.0014371704630569623, + "loss": 0.1309, + "step": 36889 + }, + { + "epoch": 0.32022291473164294, + "grad_norm": 0.435546875, + "learning_rate": 0.001437142867931007, + "loss": 0.126, + "step": 36890 + }, + { + "epoch": 0.32023159521184713, + "grad_norm": 1.7578125, + "learning_rate": 0.0014371152724363574, + "loss": 0.2266, + "step": 36891 + }, + { + "epoch": 0.32024027569205127, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014370876765730433, + "loss": 0.0645, + "step": 36892 + }, + { + "epoch": 0.32024895617225546, + "grad_norm": 0.1259765625, + "learning_rate": 0.0014370600803410953, + "loss": 0.0908, + "step": 36893 + }, + { + "epoch": 0.3202576366524596, + "grad_norm": 0.12890625, + "learning_rate": 0.0014370324837405435, + "loss": 0.0933, + "step": 36894 + }, + { + "epoch": 0.3202663171326638, + "grad_norm": 0.33984375, + "learning_rate": 0.001437004886771418, + "loss": 0.0942, + "step": 36895 + }, + { + "epoch": 0.32027499761286793, + "grad_norm": 0.1240234375, + "learning_rate": 0.0014369772894337492, + "loss": 0.1299, + "step": 36896 + }, + { + "epoch": 0.3202836780930721, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014369496917275668, + "loss": 0.0977, + "step": 36897 + }, + { + "epoch": 0.32029235857327626, + "grad_norm": 0.0771484375, + "learning_rate": 0.0014369220936529014, + "loss": 0.0703, + "step": 36898 + }, + { + "epoch": 0.32030103905348045, + "grad_norm": 0.1875, + "learning_rate": 0.0014368944952097833, + "loss": 0.0762, + "step": 36899 + }, + { + "epoch": 0.3203097195336846, + "grad_norm": 0.2421875, + "learning_rate": 0.001436866896398242, + "loss": 0.0894, + "step": 36900 + }, + { + "epoch": 0.3203184000138888, + "grad_norm": 0.46484375, + "learning_rate": 0.0014368392972183082, + "loss": 0.1133, + "step": 36901 + }, + { + "epoch": 0.3203270804940929, + "grad_norm": 0.28515625, + "learning_rate": 0.0014368116976700123, + "loss": 0.0938, + "step": 36902 + }, + { + "epoch": 0.3203357609742971, + "grad_norm": 0.1064453125, + "learning_rate": 0.0014367840977533841, + "loss": 0.0923, + "step": 36903 + }, + { + "epoch": 0.32034444145450125, + "grad_norm": 0.1455078125, + "learning_rate": 0.001436756497468454, + "loss": 0.1094, + "step": 36904 + }, + { + "epoch": 0.32035312193470544, + "grad_norm": 0.267578125, + "learning_rate": 0.0014367288968152516, + "loss": 0.0918, + "step": 36905 + }, + { + "epoch": 0.3203618024149096, + "grad_norm": 0.1416015625, + "learning_rate": 0.001436701295793808, + "loss": 0.1182, + "step": 36906 + }, + { + "epoch": 0.3203704828951138, + "grad_norm": 0.34375, + "learning_rate": 0.0014366736944041527, + "loss": 0.0908, + "step": 36907 + }, + { + "epoch": 0.3203791633753179, + "grad_norm": 0.2060546875, + "learning_rate": 0.0014366460926463164, + "loss": 0.1025, + "step": 36908 + }, + { + "epoch": 0.3203878438555221, + "grad_norm": 0.134765625, + "learning_rate": 0.0014366184905203283, + "loss": 0.1104, + "step": 36909 + }, + { + "epoch": 0.32039652433572624, + "grad_norm": 0.609375, + "learning_rate": 0.0014365908880262196, + "loss": 0.0747, + "step": 36910 + }, + { + "epoch": 0.32040520481593043, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014365632851640203, + "loss": 0.124, + "step": 36911 + }, + { + "epoch": 0.32041388529613457, + "grad_norm": 0.255859375, + "learning_rate": 0.0014365356819337604, + "loss": 0.1426, + "step": 36912 + }, + { + "epoch": 0.32042256577633876, + "grad_norm": 0.8125, + "learning_rate": 0.0014365080783354705, + "loss": 0.1484, + "step": 36913 + }, + { + "epoch": 0.3204312462565429, + "grad_norm": 0.1611328125, + "learning_rate": 0.00143648047436918, + "loss": 0.1348, + "step": 36914 + }, + { + "epoch": 0.3204399267367471, + "grad_norm": 0.9453125, + "learning_rate": 0.0014364528700349195, + "loss": 0.1182, + "step": 36915 + }, + { + "epoch": 0.32044860721695123, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014364252653327192, + "loss": 0.0962, + "step": 36916 + }, + { + "epoch": 0.3204572876971554, + "grad_norm": 0.3359375, + "learning_rate": 0.0014363976602626094, + "loss": 0.0728, + "step": 36917 + }, + { + "epoch": 0.32046596817735956, + "grad_norm": 0.205078125, + "learning_rate": 0.00143637005482462, + "loss": 0.085, + "step": 36918 + }, + { + "epoch": 0.32047464865756375, + "grad_norm": 0.314453125, + "learning_rate": 0.0014363424490187815, + "loss": 0.0859, + "step": 36919 + }, + { + "epoch": 0.3204833291377679, + "grad_norm": 0.1318359375, + "learning_rate": 0.001436314842845124, + "loss": 0.1387, + "step": 36920 + }, + { + "epoch": 0.3204920096179721, + "grad_norm": 0.5625, + "learning_rate": 0.0014362872363036776, + "loss": 0.1211, + "step": 36921 + }, + { + "epoch": 0.3205006900981762, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014362596293944726, + "loss": 0.0928, + "step": 36922 + }, + { + "epoch": 0.3205093705783804, + "grad_norm": 0.08447265625, + "learning_rate": 0.0014362320221175389, + "loss": 0.104, + "step": 36923 + }, + { + "epoch": 0.32051805105858455, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014362044144729071, + "loss": 0.1113, + "step": 36924 + }, + { + "epoch": 0.32052673153878875, + "grad_norm": 0.466796875, + "learning_rate": 0.0014361768064606072, + "loss": 0.0967, + "step": 36925 + }, + { + "epoch": 0.3205354120189929, + "grad_norm": 0.1513671875, + "learning_rate": 0.0014361491980806693, + "loss": 0.1211, + "step": 36926 + }, + { + "epoch": 0.3205440924991971, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014361215893331236, + "loss": 0.1152, + "step": 36927 + }, + { + "epoch": 0.3205527729794012, + "grad_norm": 0.134765625, + "learning_rate": 0.0014360939802180008, + "loss": 0.084, + "step": 36928 + }, + { + "epoch": 0.3205614534596054, + "grad_norm": 0.2353515625, + "learning_rate": 0.0014360663707353304, + "loss": 0.084, + "step": 36929 + }, + { + "epoch": 0.32057013393980954, + "grad_norm": 1.3046875, + "learning_rate": 0.0014360387608851426, + "loss": 0.1621, + "step": 36930 + }, + { + "epoch": 0.32057881442001374, + "grad_norm": 0.134765625, + "learning_rate": 0.0014360111506674681, + "loss": 0.085, + "step": 36931 + }, + { + "epoch": 0.3205874949002179, + "grad_norm": 0.298828125, + "learning_rate": 0.001435983540082337, + "loss": 0.0859, + "step": 36932 + }, + { + "epoch": 0.32059617538042207, + "grad_norm": 0.1103515625, + "learning_rate": 0.0014359559291297792, + "loss": 0.1133, + "step": 36933 + }, + { + "epoch": 0.3206048558606262, + "grad_norm": 1.1328125, + "learning_rate": 0.0014359283178098253, + "loss": 0.1299, + "step": 36934 + }, + { + "epoch": 0.3206135363408304, + "grad_norm": 0.0859375, + "learning_rate": 0.0014359007061225048, + "loss": 0.0884, + "step": 36935 + }, + { + "epoch": 0.32062221682103453, + "grad_norm": 0.78125, + "learning_rate": 0.0014358730940678484, + "loss": 0.1074, + "step": 36936 + }, + { + "epoch": 0.3206308973012387, + "grad_norm": 0.51171875, + "learning_rate": 0.0014358454816458866, + "loss": 0.1084, + "step": 36937 + }, + { + "epoch": 0.32063957778144286, + "grad_norm": 0.51953125, + "learning_rate": 0.0014358178688566488, + "loss": 0.0962, + "step": 36938 + }, + { + "epoch": 0.32064825826164706, + "grad_norm": 0.154296875, + "learning_rate": 0.0014357902557001658, + "loss": 0.0742, + "step": 36939 + }, + { + "epoch": 0.3206569387418512, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014357626421764675, + "loss": 0.0967, + "step": 36940 + }, + { + "epoch": 0.3206656192220554, + "grad_norm": 0.421875, + "learning_rate": 0.0014357350282855846, + "loss": 0.1035, + "step": 36941 + }, + { + "epoch": 0.3206742997022595, + "grad_norm": 0.126953125, + "learning_rate": 0.0014357074140275469, + "loss": 0.1064, + "step": 36942 + }, + { + "epoch": 0.3206829801824637, + "grad_norm": 0.259765625, + "learning_rate": 0.0014356797994023842, + "loss": 0.0781, + "step": 36943 + }, + { + "epoch": 0.32069166066266785, + "grad_norm": 0.1220703125, + "learning_rate": 0.0014356521844101273, + "loss": 0.0874, + "step": 36944 + }, + { + "epoch": 0.32070034114287205, + "grad_norm": 0.154296875, + "learning_rate": 0.0014356245690508062, + "loss": 0.0967, + "step": 36945 + }, + { + "epoch": 0.3207090216230762, + "grad_norm": 0.2392578125, + "learning_rate": 0.0014355969533244512, + "loss": 0.1079, + "step": 36946 + }, + { + "epoch": 0.3207177021032804, + "grad_norm": 0.091796875, + "learning_rate": 0.0014355693372310925, + "loss": 0.0918, + "step": 36947 + }, + { + "epoch": 0.3207263825834845, + "grad_norm": 0.8671875, + "learning_rate": 0.00143554172077076, + "loss": 0.0938, + "step": 36948 + }, + { + "epoch": 0.3207350630636887, + "grad_norm": 0.158203125, + "learning_rate": 0.001435514103943484, + "loss": 0.1328, + "step": 36949 + }, + { + "epoch": 0.32074374354389285, + "grad_norm": 0.08251953125, + "learning_rate": 0.0014354864867492949, + "loss": 0.0903, + "step": 36950 + }, + { + "epoch": 0.32075242402409704, + "grad_norm": 0.1728515625, + "learning_rate": 0.001435458869188223, + "loss": 0.1299, + "step": 36951 + }, + { + "epoch": 0.3207611045043012, + "grad_norm": 0.291015625, + "learning_rate": 0.0014354312512602982, + "loss": 0.0884, + "step": 36952 + }, + { + "epoch": 0.32076978498450537, + "grad_norm": 0.201171875, + "learning_rate": 0.0014354036329655509, + "loss": 0.0903, + "step": 36953 + }, + { + "epoch": 0.3207784654647095, + "grad_norm": 0.08056640625, + "learning_rate": 0.0014353760143040112, + "loss": 0.0884, + "step": 36954 + }, + { + "epoch": 0.3207871459449137, + "grad_norm": 0.4296875, + "learning_rate": 0.0014353483952757094, + "loss": 0.1289, + "step": 36955 + }, + { + "epoch": 0.32079582642511784, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014353207758806756, + "loss": 0.1621, + "step": 36956 + }, + { + "epoch": 0.32080450690532203, + "grad_norm": 0.55078125, + "learning_rate": 0.0014352931561189399, + "loss": 0.0806, + "step": 36957 + }, + { + "epoch": 0.32081318738552617, + "grad_norm": 0.345703125, + "learning_rate": 0.0014352655359905324, + "loss": 0.1328, + "step": 36958 + }, + { + "epoch": 0.32082186786573036, + "grad_norm": 0.51171875, + "learning_rate": 0.0014352379154954841, + "loss": 0.1104, + "step": 36959 + }, + { + "epoch": 0.3208305483459345, + "grad_norm": 0.296875, + "learning_rate": 0.0014352102946338246, + "loss": 0.1021, + "step": 36960 + }, + { + "epoch": 0.3208392288261387, + "grad_norm": 0.2265625, + "learning_rate": 0.0014351826734055838, + "loss": 0.1719, + "step": 36961 + }, + { + "epoch": 0.3208479093063428, + "grad_norm": 0.455078125, + "learning_rate": 0.0014351550518107924, + "loss": 0.1191, + "step": 36962 + }, + { + "epoch": 0.32085658978654696, + "grad_norm": 0.201171875, + "learning_rate": 0.0014351274298494807, + "loss": 0.0996, + "step": 36963 + }, + { + "epoch": 0.32086527026675116, + "grad_norm": 0.17578125, + "learning_rate": 0.0014350998075216787, + "loss": 0.0835, + "step": 36964 + }, + { + "epoch": 0.3208739507469553, + "grad_norm": 0.1396484375, + "learning_rate": 0.001435072184827416, + "loss": 0.1016, + "step": 36965 + }, + { + "epoch": 0.3208826312271595, + "grad_norm": 0.8359375, + "learning_rate": 0.0014350445617667238, + "loss": 0.0791, + "step": 36966 + }, + { + "epoch": 0.3208913117073636, + "grad_norm": 0.12060546875, + "learning_rate": 0.0014350169383396318, + "loss": 0.1396, + "step": 36967 + }, + { + "epoch": 0.3208999921875678, + "grad_norm": 0.259765625, + "learning_rate": 0.0014349893145461708, + "loss": 0.124, + "step": 36968 + }, + { + "epoch": 0.32090867266777195, + "grad_norm": 1.046875, + "learning_rate": 0.0014349616903863698, + "loss": 0.1001, + "step": 36969 + }, + { + "epoch": 0.32091735314797615, + "grad_norm": 1.0625, + "learning_rate": 0.00143493406586026, + "loss": 0.0854, + "step": 36970 + }, + { + "epoch": 0.3209260336281803, + "grad_norm": 0.5234375, + "learning_rate": 0.0014349064409678713, + "loss": 0.1021, + "step": 36971 + }, + { + "epoch": 0.3209347141083845, + "grad_norm": 0.1767578125, + "learning_rate": 0.0014348788157092342, + "loss": 0.1016, + "step": 36972 + }, + { + "epoch": 0.3209433945885886, + "grad_norm": 0.1796875, + "learning_rate": 0.0014348511900843786, + "loss": 0.123, + "step": 36973 + }, + { + "epoch": 0.3209520750687928, + "grad_norm": 0.248046875, + "learning_rate": 0.0014348235640933345, + "loss": 0.126, + "step": 36974 + }, + { + "epoch": 0.32096075554899695, + "grad_norm": 0.10546875, + "learning_rate": 0.0014347959377361327, + "loss": 0.0781, + "step": 36975 + }, + { + "epoch": 0.32096943602920114, + "grad_norm": 0.10302734375, + "learning_rate": 0.0014347683110128029, + "loss": 0.1138, + "step": 36976 + }, + { + "epoch": 0.3209781165094053, + "grad_norm": 0.427734375, + "learning_rate": 0.0014347406839233755, + "loss": 0.0957, + "step": 36977 + }, + { + "epoch": 0.32098679698960947, + "grad_norm": 0.21484375, + "learning_rate": 0.0014347130564678807, + "loss": 0.0854, + "step": 36978 + }, + { + "epoch": 0.3209954774698136, + "grad_norm": 0.21484375, + "learning_rate": 0.0014346854286463489, + "loss": 0.0835, + "step": 36979 + }, + { + "epoch": 0.3210041579500178, + "grad_norm": 0.287109375, + "learning_rate": 0.0014346578004588097, + "loss": 0.1543, + "step": 36980 + }, + { + "epoch": 0.32101283843022194, + "grad_norm": 0.61328125, + "learning_rate": 0.0014346301719052944, + "loss": 0.1445, + "step": 36981 + }, + { + "epoch": 0.32102151891042613, + "grad_norm": 0.6328125, + "learning_rate": 0.001434602542985832, + "loss": 0.0947, + "step": 36982 + }, + { + "epoch": 0.32103019939063027, + "grad_norm": 0.19921875, + "learning_rate": 0.0014345749137004537, + "loss": 0.0952, + "step": 36983 + }, + { + "epoch": 0.32103887987083446, + "grad_norm": 0.203125, + "learning_rate": 0.0014345472840491893, + "loss": 0.0869, + "step": 36984 + }, + { + "epoch": 0.3210475603510386, + "grad_norm": 0.34765625, + "learning_rate": 0.001434519654032069, + "loss": 0.0938, + "step": 36985 + }, + { + "epoch": 0.3210562408312428, + "grad_norm": 0.3984375, + "learning_rate": 0.0014344920236491225, + "loss": 0.1406, + "step": 36986 + }, + { + "epoch": 0.3210649213114469, + "grad_norm": 0.466796875, + "learning_rate": 0.0014344643929003807, + "loss": 0.0903, + "step": 36987 + }, + { + "epoch": 0.3210736017916511, + "grad_norm": 0.1083984375, + "learning_rate": 0.001434436761785874, + "loss": 0.082, + "step": 36988 + }, + { + "epoch": 0.32108228227185526, + "grad_norm": 0.06689453125, + "learning_rate": 0.0014344091303056324, + "loss": 0.085, + "step": 36989 + }, + { + "epoch": 0.32109096275205945, + "grad_norm": 0.5, + "learning_rate": 0.0014343814984596858, + "loss": 0.1074, + "step": 36990 + }, + { + "epoch": 0.3210996432322636, + "grad_norm": 0.32421875, + "learning_rate": 0.0014343538662480646, + "loss": 0.0718, + "step": 36991 + }, + { + "epoch": 0.3211083237124678, + "grad_norm": 0.255859375, + "learning_rate": 0.001434326233670799, + "loss": 0.1641, + "step": 36992 + }, + { + "epoch": 0.3211170041926719, + "grad_norm": 0.06689453125, + "learning_rate": 0.0014342986007279193, + "loss": 0.0811, + "step": 36993 + }, + { + "epoch": 0.3211256846728761, + "grad_norm": 0.50390625, + "learning_rate": 0.0014342709674194558, + "loss": 0.1094, + "step": 36994 + }, + { + "epoch": 0.32113436515308025, + "grad_norm": 0.1689453125, + "learning_rate": 0.0014342433337454384, + "loss": 0.0791, + "step": 36995 + }, + { + "epoch": 0.32114304563328444, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014342156997058976, + "loss": 0.1191, + "step": 36996 + }, + { + "epoch": 0.3211517261134886, + "grad_norm": 0.38671875, + "learning_rate": 0.0014341880653008631, + "loss": 0.0752, + "step": 36997 + }, + { + "epoch": 0.32116040659369277, + "grad_norm": 0.1083984375, + "learning_rate": 0.0014341604305303664, + "loss": 0.0923, + "step": 36998 + }, + { + "epoch": 0.3211690870738969, + "grad_norm": 0.7421875, + "learning_rate": 0.0014341327953944364, + "loss": 0.1504, + "step": 36999 + }, + { + "epoch": 0.3211777675541011, + "grad_norm": 0.392578125, + "learning_rate": 0.0014341051598931037, + "loss": 0.0938, + "step": 37000 + }, + { + "epoch": 0.32118644803430524, + "grad_norm": 0.3125, + "learning_rate": 0.0014340775240263986, + "loss": 0.0947, + "step": 37001 + }, + { + "epoch": 0.32119512851450943, + "grad_norm": 0.3203125, + "learning_rate": 0.0014340498877943514, + "loss": 0.1328, + "step": 37002 + }, + { + "epoch": 0.32120380899471357, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014340222511969928, + "loss": 0.126, + "step": 37003 + }, + { + "epoch": 0.32121248947491776, + "grad_norm": 0.671875, + "learning_rate": 0.001433994614234352, + "loss": 0.1348, + "step": 37004 + }, + { + "epoch": 0.3212211699551219, + "grad_norm": 0.91796875, + "learning_rate": 0.0014339669769064596, + "loss": 0.1133, + "step": 37005 + }, + { + "epoch": 0.3212298504353261, + "grad_norm": 0.10009765625, + "learning_rate": 0.001433939339213346, + "loss": 0.1011, + "step": 37006 + }, + { + "epoch": 0.32123853091553023, + "grad_norm": 0.1240234375, + "learning_rate": 0.0014339117011550413, + "loss": 0.0688, + "step": 37007 + }, + { + "epoch": 0.3212472113957344, + "grad_norm": 0.3046875, + "learning_rate": 0.001433884062731576, + "loss": 0.0967, + "step": 37008 + }, + { + "epoch": 0.32125589187593856, + "grad_norm": 0.0947265625, + "learning_rate": 0.00143385642394298, + "loss": 0.0791, + "step": 37009 + }, + { + "epoch": 0.32126457235614275, + "grad_norm": 0.421875, + "learning_rate": 0.0014338287847892835, + "loss": 0.0889, + "step": 37010 + }, + { + "epoch": 0.3212732528363469, + "grad_norm": 0.5390625, + "learning_rate": 0.0014338011452705171, + "loss": 0.0967, + "step": 37011 + }, + { + "epoch": 0.3212819333165511, + "grad_norm": 0.328125, + "learning_rate": 0.001433773505386711, + "loss": 0.0757, + "step": 37012 + }, + { + "epoch": 0.3212906137967552, + "grad_norm": 0.1806640625, + "learning_rate": 0.0014337458651378944, + "loss": 0.1138, + "step": 37013 + }, + { + "epoch": 0.3212992942769594, + "grad_norm": 0.47265625, + "learning_rate": 0.0014337182245240992, + "loss": 0.1211, + "step": 37014 + }, + { + "epoch": 0.32130797475716355, + "grad_norm": 0.109375, + "learning_rate": 0.001433690583545354, + "loss": 0.1123, + "step": 37015 + }, + { + "epoch": 0.32131665523736774, + "grad_norm": 0.2734375, + "learning_rate": 0.0014336629422016902, + "loss": 0.1074, + "step": 37016 + }, + { + "epoch": 0.3213253357175719, + "grad_norm": 0.19921875, + "learning_rate": 0.001433635300493138, + "loss": 0.103, + "step": 37017 + }, + { + "epoch": 0.3213340161977761, + "grad_norm": 0.53125, + "learning_rate": 0.0014336076584197264, + "loss": 0.1172, + "step": 37018 + }, + { + "epoch": 0.3213426966779802, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014335800159814875, + "loss": 0.0825, + "step": 37019 + }, + { + "epoch": 0.3213513771581844, + "grad_norm": 0.9296875, + "learning_rate": 0.0014335523731784496, + "loss": 0.1143, + "step": 37020 + }, + { + "epoch": 0.32136005763838854, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014335247300106442, + "loss": 0.0957, + "step": 37021 + }, + { + "epoch": 0.32136873811859273, + "grad_norm": 0.37109375, + "learning_rate": 0.0014334970864781009, + "loss": 0.1738, + "step": 37022 + }, + { + "epoch": 0.32137741859879687, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014334694425808508, + "loss": 0.1006, + "step": 37023 + }, + { + "epoch": 0.32138609907900106, + "grad_norm": 0.4296875, + "learning_rate": 0.0014334417983189233, + "loss": 0.1006, + "step": 37024 + }, + { + "epoch": 0.3213947795592052, + "grad_norm": 0.69140625, + "learning_rate": 0.0014334141536923484, + "loss": 0.1055, + "step": 37025 + }, + { + "epoch": 0.3214034600394094, + "grad_norm": 0.1904296875, + "learning_rate": 0.001433386508701157, + "loss": 0.085, + "step": 37026 + }, + { + "epoch": 0.32141214051961353, + "grad_norm": 0.1953125, + "learning_rate": 0.0014333588633453794, + "loss": 0.1162, + "step": 37027 + }, + { + "epoch": 0.3214208209998177, + "grad_norm": 0.515625, + "learning_rate": 0.0014333312176250453, + "loss": 0.1143, + "step": 37028 + }, + { + "epoch": 0.32142950148002186, + "grad_norm": 0.44921875, + "learning_rate": 0.0014333035715401851, + "loss": 0.0933, + "step": 37029 + }, + { + "epoch": 0.32143818196022605, + "grad_norm": 0.06494140625, + "learning_rate": 0.0014332759250908296, + "loss": 0.0752, + "step": 37030 + }, + { + "epoch": 0.3214468624404302, + "grad_norm": 0.4296875, + "learning_rate": 0.0014332482782770082, + "loss": 0.0977, + "step": 37031 + }, + { + "epoch": 0.3214555429206344, + "grad_norm": 0.6015625, + "learning_rate": 0.0014332206310987516, + "loss": 0.1074, + "step": 37032 + }, + { + "epoch": 0.3214642234008385, + "grad_norm": 0.29296875, + "learning_rate": 0.0014331929835560899, + "loss": 0.0854, + "step": 37033 + }, + { + "epoch": 0.3214729038810427, + "grad_norm": 0.3671875, + "learning_rate": 0.001433165335649053, + "loss": 0.0947, + "step": 37034 + }, + { + "epoch": 0.32148158436124685, + "grad_norm": 0.21875, + "learning_rate": 0.0014331376873776718, + "loss": 0.1152, + "step": 37035 + }, + { + "epoch": 0.32149026484145105, + "grad_norm": 0.248046875, + "learning_rate": 0.0014331100387419765, + "loss": 0.0757, + "step": 37036 + }, + { + "epoch": 0.3214989453216552, + "grad_norm": 0.6875, + "learning_rate": 0.0014330823897419967, + "loss": 0.1001, + "step": 37037 + }, + { + "epoch": 0.3215076258018594, + "grad_norm": 0.0830078125, + "learning_rate": 0.0014330547403777632, + "loss": 0.1084, + "step": 37038 + }, + { + "epoch": 0.3215163062820635, + "grad_norm": 0.13671875, + "learning_rate": 0.001433027090649306, + "loss": 0.1001, + "step": 37039 + }, + { + "epoch": 0.3215249867622677, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014329994405566553, + "loss": 0.1338, + "step": 37040 + }, + { + "epoch": 0.32153366724247184, + "grad_norm": 0.435546875, + "learning_rate": 0.0014329717900998416, + "loss": 0.1172, + "step": 37041 + }, + { + "epoch": 0.32154234772267604, + "grad_norm": 0.296875, + "learning_rate": 0.0014329441392788946, + "loss": 0.1055, + "step": 37042 + }, + { + "epoch": 0.3215510282028802, + "grad_norm": 0.146484375, + "learning_rate": 0.0014329164880938453, + "loss": 0.1177, + "step": 37043 + }, + { + "epoch": 0.32155970868308437, + "grad_norm": 0.23828125, + "learning_rate": 0.0014328888365447232, + "loss": 0.083, + "step": 37044 + }, + { + "epoch": 0.3215683891632885, + "grad_norm": 0.5078125, + "learning_rate": 0.0014328611846315589, + "loss": 0.0869, + "step": 37045 + }, + { + "epoch": 0.3215770696434927, + "grad_norm": 0.1943359375, + "learning_rate": 0.001432833532354383, + "loss": 0.1152, + "step": 37046 + }, + { + "epoch": 0.32158575012369683, + "grad_norm": 0.158203125, + "learning_rate": 0.001432805879713225, + "loss": 0.1445, + "step": 37047 + }, + { + "epoch": 0.321594430603901, + "grad_norm": 2.0, + "learning_rate": 0.0014327782267081154, + "loss": 0.1699, + "step": 37048 + }, + { + "epoch": 0.32160311108410516, + "grad_norm": 0.2890625, + "learning_rate": 0.0014327505733390848, + "loss": 0.0967, + "step": 37049 + }, + { + "epoch": 0.32161179156430936, + "grad_norm": 0.25390625, + "learning_rate": 0.0014327229196061631, + "loss": 0.0767, + "step": 37050 + }, + { + "epoch": 0.3216204720445135, + "grad_norm": 0.06640625, + "learning_rate": 0.0014326952655093806, + "loss": 0.0723, + "step": 37051 + }, + { + "epoch": 0.3216291525247177, + "grad_norm": 0.0869140625, + "learning_rate": 0.0014326676110487672, + "loss": 0.0884, + "step": 37052 + }, + { + "epoch": 0.3216378330049218, + "grad_norm": 0.3984375, + "learning_rate": 0.001432639956224354, + "loss": 0.085, + "step": 37053 + }, + { + "epoch": 0.321646513485126, + "grad_norm": 0.83984375, + "learning_rate": 0.0014326123010361704, + "loss": 0.0947, + "step": 37054 + }, + { + "epoch": 0.32165519396533016, + "grad_norm": 0.28515625, + "learning_rate": 0.0014325846454842474, + "loss": 0.1191, + "step": 37055 + }, + { + "epoch": 0.32166387444553435, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014325569895686143, + "loss": 0.0811, + "step": 37056 + }, + { + "epoch": 0.3216725549257385, + "grad_norm": 0.34765625, + "learning_rate": 0.0014325293332893023, + "loss": 0.1113, + "step": 37057 + }, + { + "epoch": 0.3216812354059427, + "grad_norm": 0.1162109375, + "learning_rate": 0.001432501676646341, + "loss": 0.1426, + "step": 37058 + }, + { + "epoch": 0.3216899158861468, + "grad_norm": 0.376953125, + "learning_rate": 0.001432474019639761, + "loss": 0.1167, + "step": 37059 + }, + { + "epoch": 0.321698596366351, + "grad_norm": 0.158203125, + "learning_rate": 0.0014324463622695926, + "loss": 0.1113, + "step": 37060 + }, + { + "epoch": 0.32170727684655515, + "grad_norm": 0.16796875, + "learning_rate": 0.0014324187045358654, + "loss": 0.1133, + "step": 37061 + }, + { + "epoch": 0.32171595732675934, + "grad_norm": 0.474609375, + "learning_rate": 0.0014323910464386102, + "loss": 0.0957, + "step": 37062 + }, + { + "epoch": 0.3217246378069635, + "grad_norm": 0.62890625, + "learning_rate": 0.0014323633879778574, + "loss": 0.1138, + "step": 37063 + }, + { + "epoch": 0.32173331828716767, + "grad_norm": 0.14453125, + "learning_rate": 0.0014323357291536368, + "loss": 0.0928, + "step": 37064 + }, + { + "epoch": 0.3217419987673718, + "grad_norm": 0.10009765625, + "learning_rate": 0.0014323080699659785, + "loss": 0.1069, + "step": 37065 + }, + { + "epoch": 0.321750679247576, + "grad_norm": 0.1826171875, + "learning_rate": 0.0014322804104149136, + "loss": 0.0864, + "step": 37066 + }, + { + "epoch": 0.32175935972778014, + "grad_norm": 0.1103515625, + "learning_rate": 0.0014322527505004718, + "loss": 0.1235, + "step": 37067 + }, + { + "epoch": 0.32176804020798433, + "grad_norm": 0.283203125, + "learning_rate": 0.0014322250902226834, + "loss": 0.0786, + "step": 37068 + }, + { + "epoch": 0.32177672068818847, + "grad_norm": 0.7109375, + "learning_rate": 0.0014321974295815783, + "loss": 0.0942, + "step": 37069 + }, + { + "epoch": 0.32178540116839266, + "grad_norm": 0.1181640625, + "learning_rate": 0.0014321697685771872, + "loss": 0.1104, + "step": 37070 + }, + { + "epoch": 0.3217940816485968, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014321421072095407, + "loss": 0.0732, + "step": 37071 + }, + { + "epoch": 0.321802762128801, + "grad_norm": 0.26171875, + "learning_rate": 0.001432114445478668, + "loss": 0.1035, + "step": 37072 + }, + { + "epoch": 0.3218114426090051, + "grad_norm": 0.14453125, + "learning_rate": 0.0014320867833846, + "loss": 0.1123, + "step": 37073 + }, + { + "epoch": 0.3218201230892093, + "grad_norm": 0.08349609375, + "learning_rate": 0.001432059120927367, + "loss": 0.103, + "step": 37074 + }, + { + "epoch": 0.32182880356941346, + "grad_norm": 0.212890625, + "learning_rate": 0.0014320314581069992, + "loss": 0.0679, + "step": 37075 + }, + { + "epoch": 0.32183748404961765, + "grad_norm": 0.15625, + "learning_rate": 0.001432003794923527, + "loss": 0.1094, + "step": 37076 + }, + { + "epoch": 0.3218461645298218, + "grad_norm": 0.306640625, + "learning_rate": 0.0014319761313769804, + "loss": 0.1201, + "step": 37077 + }, + { + "epoch": 0.321854845010026, + "grad_norm": 0.24609375, + "learning_rate": 0.0014319484674673892, + "loss": 0.123, + "step": 37078 + }, + { + "epoch": 0.3218635254902301, + "grad_norm": 0.490234375, + "learning_rate": 0.0014319208031947845, + "loss": 0.1094, + "step": 37079 + }, + { + "epoch": 0.3218722059704343, + "grad_norm": 0.193359375, + "learning_rate": 0.0014318931385591963, + "loss": 0.0918, + "step": 37080 + }, + { + "epoch": 0.32188088645063845, + "grad_norm": 0.423828125, + "learning_rate": 0.0014318654735606545, + "loss": 0.1016, + "step": 37081 + }, + { + "epoch": 0.32188956693084264, + "grad_norm": 0.37109375, + "learning_rate": 0.0014318378081991895, + "loss": 0.1074, + "step": 37082 + }, + { + "epoch": 0.3218982474110468, + "grad_norm": 0.09619140625, + "learning_rate": 0.0014318101424748316, + "loss": 0.0835, + "step": 37083 + }, + { + "epoch": 0.32190692789125097, + "grad_norm": 0.29296875, + "learning_rate": 0.0014317824763876116, + "loss": 0.124, + "step": 37084 + }, + { + "epoch": 0.3219156083714551, + "grad_norm": 0.11328125, + "learning_rate": 0.001431754809937559, + "loss": 0.0864, + "step": 37085 + }, + { + "epoch": 0.3219242888516593, + "grad_norm": 0.2890625, + "learning_rate": 0.0014317271431247044, + "loss": 0.083, + "step": 37086 + }, + { + "epoch": 0.32193296933186344, + "grad_norm": 0.2197265625, + "learning_rate": 0.0014316994759490777, + "loss": 0.0972, + "step": 37087 + }, + { + "epoch": 0.3219416498120676, + "grad_norm": 0.55078125, + "learning_rate": 0.00143167180841071, + "loss": 0.084, + "step": 37088 + }, + { + "epoch": 0.32195033029227177, + "grad_norm": 0.11376953125, + "learning_rate": 0.0014316441405096303, + "loss": 0.0967, + "step": 37089 + }, + { + "epoch": 0.3219590107724759, + "grad_norm": 0.44921875, + "learning_rate": 0.00143161647224587, + "loss": 0.1348, + "step": 37090 + }, + { + "epoch": 0.3219676912526801, + "grad_norm": 0.388671875, + "learning_rate": 0.0014315888036194589, + "loss": 0.0981, + "step": 37091 + }, + { + "epoch": 0.32197637173288424, + "grad_norm": 0.37890625, + "learning_rate": 0.0014315611346304268, + "loss": 0.0972, + "step": 37092 + }, + { + "epoch": 0.32198505221308843, + "grad_norm": 0.1103515625, + "learning_rate": 0.001431533465278805, + "loss": 0.1025, + "step": 37093 + }, + { + "epoch": 0.32199373269329257, + "grad_norm": 0.396484375, + "learning_rate": 0.001431505795564623, + "loss": 0.0869, + "step": 37094 + }, + { + "epoch": 0.32200241317349676, + "grad_norm": 0.361328125, + "learning_rate": 0.001431478125487911, + "loss": 0.0894, + "step": 37095 + }, + { + "epoch": 0.3220110936537009, + "grad_norm": 0.25390625, + "learning_rate": 0.0014314504550487, + "loss": 0.1045, + "step": 37096 + }, + { + "epoch": 0.3220197741339051, + "grad_norm": 0.37890625, + "learning_rate": 0.0014314227842470194, + "loss": 0.1113, + "step": 37097 + }, + { + "epoch": 0.3220284546141092, + "grad_norm": 0.10498046875, + "learning_rate": 0.0014313951130828998, + "loss": 0.0986, + "step": 37098 + }, + { + "epoch": 0.3220371350943134, + "grad_norm": 0.134765625, + "learning_rate": 0.0014313674415563718, + "loss": 0.0776, + "step": 37099 + }, + { + "epoch": 0.32204581557451756, + "grad_norm": 0.07763671875, + "learning_rate": 0.001431339769667465, + "loss": 0.0728, + "step": 37100 + }, + { + "epoch": 0.32205449605472175, + "grad_norm": 0.1650390625, + "learning_rate": 0.0014313120974162104, + "loss": 0.0957, + "step": 37101 + }, + { + "epoch": 0.3220631765349259, + "grad_norm": 0.69140625, + "learning_rate": 0.0014312844248026374, + "loss": 0.0957, + "step": 37102 + }, + { + "epoch": 0.3220718570151301, + "grad_norm": 0.267578125, + "learning_rate": 0.001431256751826777, + "loss": 0.1045, + "step": 37103 + }, + { + "epoch": 0.3220805374953342, + "grad_norm": 0.185546875, + "learning_rate": 0.0014312290784886587, + "loss": 0.1123, + "step": 37104 + }, + { + "epoch": 0.3220892179755384, + "grad_norm": 0.37890625, + "learning_rate": 0.001431201404788314, + "loss": 0.126, + "step": 37105 + }, + { + "epoch": 0.32209789845574255, + "grad_norm": 0.30078125, + "learning_rate": 0.0014311737307257722, + "loss": 0.1152, + "step": 37106 + }, + { + "epoch": 0.32210657893594674, + "grad_norm": 0.73828125, + "learning_rate": 0.001431146056301064, + "loss": 0.1006, + "step": 37107 + }, + { + "epoch": 0.3221152594161509, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014311183815142188, + "loss": 0.1157, + "step": 37108 + }, + { + "epoch": 0.32212393989635507, + "grad_norm": 0.59765625, + "learning_rate": 0.0014310907063652678, + "loss": 0.1455, + "step": 37109 + }, + { + "epoch": 0.3221326203765592, + "grad_norm": 0.447265625, + "learning_rate": 0.0014310630308542413, + "loss": 0.1123, + "step": 37110 + }, + { + "epoch": 0.3221413008567634, + "grad_norm": 0.33203125, + "learning_rate": 0.001431035354981169, + "loss": 0.0786, + "step": 37111 + }, + { + "epoch": 0.32214998133696754, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014310076787460812, + "loss": 0.0967, + "step": 37112 + }, + { + "epoch": 0.32215866181717173, + "grad_norm": 0.39453125, + "learning_rate": 0.0014309800021490085, + "loss": 0.0845, + "step": 37113 + }, + { + "epoch": 0.32216734229737587, + "grad_norm": 0.453125, + "learning_rate": 0.0014309523251899812, + "loss": 0.2266, + "step": 37114 + }, + { + "epoch": 0.32217602277758006, + "grad_norm": 0.443359375, + "learning_rate": 0.0014309246478690293, + "loss": 0.1123, + "step": 37115 + }, + { + "epoch": 0.3221847032577842, + "grad_norm": 0.40625, + "learning_rate": 0.0014308969701861834, + "loss": 0.1147, + "step": 37116 + }, + { + "epoch": 0.3221933837379884, + "grad_norm": 0.17578125, + "learning_rate": 0.0014308692921414733, + "loss": 0.1196, + "step": 37117 + }, + { + "epoch": 0.32220206421819253, + "grad_norm": 0.150390625, + "learning_rate": 0.0014308416137349294, + "loss": 0.1196, + "step": 37118 + }, + { + "epoch": 0.3222107446983967, + "grad_norm": 0.142578125, + "learning_rate": 0.0014308139349665822, + "loss": 0.1328, + "step": 37119 + }, + { + "epoch": 0.32221942517860086, + "grad_norm": 0.79296875, + "learning_rate": 0.0014307862558364617, + "loss": 0.1445, + "step": 37120 + }, + { + "epoch": 0.32222810565880505, + "grad_norm": 0.53515625, + "learning_rate": 0.0014307585763445986, + "loss": 0.1553, + "step": 37121 + }, + { + "epoch": 0.3222367861390092, + "grad_norm": 0.33984375, + "learning_rate": 0.0014307308964910228, + "loss": 0.0938, + "step": 37122 + }, + { + "epoch": 0.3222454666192134, + "grad_norm": 0.458984375, + "learning_rate": 0.0014307032162757646, + "loss": 0.085, + "step": 37123 + }, + { + "epoch": 0.3222541470994175, + "grad_norm": 0.2734375, + "learning_rate": 0.0014306755356988546, + "loss": 0.0996, + "step": 37124 + }, + { + "epoch": 0.3222628275796217, + "grad_norm": 0.1005859375, + "learning_rate": 0.0014306478547603226, + "loss": 0.0967, + "step": 37125 + }, + { + "epoch": 0.32227150805982585, + "grad_norm": 0.1337890625, + "learning_rate": 0.001430620173460199, + "loss": 0.1069, + "step": 37126 + }, + { + "epoch": 0.32228018854003004, + "grad_norm": 0.09765625, + "learning_rate": 0.0014305924917985143, + "loss": 0.123, + "step": 37127 + }, + { + "epoch": 0.3222888690202342, + "grad_norm": 0.1552734375, + "learning_rate": 0.0014305648097752985, + "loss": 0.1143, + "step": 37128 + }, + { + "epoch": 0.3222975495004384, + "grad_norm": 0.54296875, + "learning_rate": 0.001430537127390582, + "loss": 0.0986, + "step": 37129 + }, + { + "epoch": 0.3223062299806425, + "grad_norm": 0.2265625, + "learning_rate": 0.001430509444644395, + "loss": 0.1396, + "step": 37130 + }, + { + "epoch": 0.3223149104608467, + "grad_norm": 0.06884765625, + "learning_rate": 0.0014304817615367678, + "loss": 0.0581, + "step": 37131 + }, + { + "epoch": 0.32232359094105084, + "grad_norm": 0.2177734375, + "learning_rate": 0.001430454078067731, + "loss": 0.0747, + "step": 37132 + }, + { + "epoch": 0.32233227142125503, + "grad_norm": 0.1845703125, + "learning_rate": 0.0014304263942373143, + "loss": 0.1045, + "step": 37133 + }, + { + "epoch": 0.32234095190145917, + "grad_norm": 0.08056640625, + "learning_rate": 0.0014303987100455484, + "loss": 0.0859, + "step": 37134 + }, + { + "epoch": 0.32234963238166336, + "grad_norm": 0.31640625, + "learning_rate": 0.0014303710254924633, + "loss": 0.0884, + "step": 37135 + }, + { + "epoch": 0.3223583128618675, + "grad_norm": 0.181640625, + "learning_rate": 0.0014303433405780898, + "loss": 0.0938, + "step": 37136 + }, + { + "epoch": 0.3223669933420717, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014303156553024576, + "loss": 0.0879, + "step": 37137 + }, + { + "epoch": 0.32237567382227583, + "grad_norm": 0.20703125, + "learning_rate": 0.001430287969665597, + "loss": 0.1084, + "step": 37138 + }, + { + "epoch": 0.32238435430248, + "grad_norm": 0.10595703125, + "learning_rate": 0.0014302602836675385, + "loss": 0.1113, + "step": 37139 + }, + { + "epoch": 0.32239303478268416, + "grad_norm": 0.50390625, + "learning_rate": 0.0014302325973083122, + "loss": 0.0923, + "step": 37140 + }, + { + "epoch": 0.32240171526288836, + "grad_norm": 0.423828125, + "learning_rate": 0.0014302049105879486, + "loss": 0.1289, + "step": 37141 + }, + { + "epoch": 0.3224103957430925, + "grad_norm": 0.059814453125, + "learning_rate": 0.0014301772235064778, + "loss": 0.0825, + "step": 37142 + }, + { + "epoch": 0.3224190762232967, + "grad_norm": 0.1796875, + "learning_rate": 0.0014301495360639304, + "loss": 0.0703, + "step": 37143 + }, + { + "epoch": 0.3224277567035008, + "grad_norm": 0.578125, + "learning_rate": 0.0014301218482603362, + "loss": 0.1143, + "step": 37144 + }, + { + "epoch": 0.322436437183705, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014300941600957258, + "loss": 0.082, + "step": 37145 + }, + { + "epoch": 0.32244511766390915, + "grad_norm": 0.353515625, + "learning_rate": 0.0014300664715701295, + "loss": 0.1235, + "step": 37146 + }, + { + "epoch": 0.32245379814411335, + "grad_norm": 0.5, + "learning_rate": 0.0014300387826835772, + "loss": 0.0884, + "step": 37147 + }, + { + "epoch": 0.3224624786243175, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014300110934360995, + "loss": 0.1309, + "step": 37148 + }, + { + "epoch": 0.3224711591045217, + "grad_norm": 0.10595703125, + "learning_rate": 0.0014299834038277268, + "loss": 0.0684, + "step": 37149 + }, + { + "epoch": 0.3224798395847258, + "grad_norm": 0.451171875, + "learning_rate": 0.0014299557138584891, + "loss": 0.0903, + "step": 37150 + }, + { + "epoch": 0.32248852006493, + "grad_norm": 0.1640625, + "learning_rate": 0.0014299280235284169, + "loss": 0.1006, + "step": 37151 + }, + { + "epoch": 0.32249720054513414, + "grad_norm": 0.09375, + "learning_rate": 0.00142990033283754, + "loss": 0.0986, + "step": 37152 + }, + { + "epoch": 0.32250588102533834, + "grad_norm": 1.0234375, + "learning_rate": 0.0014298726417858894, + "loss": 0.1133, + "step": 37153 + }, + { + "epoch": 0.3225145615055425, + "grad_norm": 0.486328125, + "learning_rate": 0.001429844950373495, + "loss": 0.1084, + "step": 37154 + }, + { + "epoch": 0.32252324198574667, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014298172586003873, + "loss": 0.0776, + "step": 37155 + }, + { + "epoch": 0.3225319224659508, + "grad_norm": 0.578125, + "learning_rate": 0.001429789566466596, + "loss": 0.0859, + "step": 37156 + }, + { + "epoch": 0.322540602946155, + "grad_norm": 0.3359375, + "learning_rate": 0.001429761873972152, + "loss": 0.1436, + "step": 37157 + }, + { + "epoch": 0.32254928342635913, + "grad_norm": 0.259765625, + "learning_rate": 0.0014297341811170853, + "loss": 0.1553, + "step": 37158 + }, + { + "epoch": 0.3225579639065633, + "grad_norm": 0.193359375, + "learning_rate": 0.001429706487901426, + "loss": 0.1348, + "step": 37159 + }, + { + "epoch": 0.32256664438676746, + "grad_norm": 1.125, + "learning_rate": 0.001429678794325205, + "loss": 0.0854, + "step": 37160 + }, + { + "epoch": 0.32257532486697166, + "grad_norm": 0.1025390625, + "learning_rate": 0.001429651100388452, + "loss": 0.0996, + "step": 37161 + }, + { + "epoch": 0.3225840053471758, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014296234060911979, + "loss": 0.1143, + "step": 37162 + }, + { + "epoch": 0.32259268582738, + "grad_norm": 0.369140625, + "learning_rate": 0.0014295957114334723, + "loss": 0.126, + "step": 37163 + }, + { + "epoch": 0.3226013663075841, + "grad_norm": 0.1962890625, + "learning_rate": 0.001429568016415306, + "loss": 0.0928, + "step": 37164 + }, + { + "epoch": 0.3226100467877883, + "grad_norm": 1.1796875, + "learning_rate": 0.0014295403210367289, + "loss": 0.0928, + "step": 37165 + }, + { + "epoch": 0.32261872726799246, + "grad_norm": 0.45703125, + "learning_rate": 0.0014295126252977714, + "loss": 0.0791, + "step": 37166 + }, + { + "epoch": 0.32262740774819665, + "grad_norm": 0.142578125, + "learning_rate": 0.001429484929198464, + "loss": 0.1143, + "step": 37167 + }, + { + "epoch": 0.3226360882284008, + "grad_norm": 0.2255859375, + "learning_rate": 0.0014294572327388367, + "loss": 0.1108, + "step": 37168 + }, + { + "epoch": 0.322644768708605, + "grad_norm": 0.154296875, + "learning_rate": 0.00142942953591892, + "loss": 0.0957, + "step": 37169 + }, + { + "epoch": 0.3226534491888091, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014294018387387437, + "loss": 0.0889, + "step": 37170 + }, + { + "epoch": 0.3226621296690133, + "grad_norm": 0.37890625, + "learning_rate": 0.001429374141198339, + "loss": 0.1484, + "step": 37171 + }, + { + "epoch": 0.32267081014921745, + "grad_norm": 0.298828125, + "learning_rate": 0.0014293464432977357, + "loss": 0.1211, + "step": 37172 + }, + { + "epoch": 0.32267949062942164, + "grad_norm": 0.353515625, + "learning_rate": 0.0014293187450369637, + "loss": 0.126, + "step": 37173 + }, + { + "epoch": 0.3226881711096258, + "grad_norm": 0.2138671875, + "learning_rate": 0.001429291046416054, + "loss": 0.0859, + "step": 37174 + }, + { + "epoch": 0.32269685158982997, + "grad_norm": 0.353515625, + "learning_rate": 0.0014292633474350365, + "loss": 0.0874, + "step": 37175 + }, + { + "epoch": 0.3227055320700341, + "grad_norm": 0.515625, + "learning_rate": 0.0014292356480939414, + "loss": 0.1187, + "step": 37176 + }, + { + "epoch": 0.3227142125502383, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014292079483927992, + "loss": 0.1182, + "step": 37177 + }, + { + "epoch": 0.32272289303044244, + "grad_norm": 0.3046875, + "learning_rate": 0.0014291802483316404, + "loss": 0.0732, + "step": 37178 + }, + { + "epoch": 0.32273157351064663, + "grad_norm": 0.36328125, + "learning_rate": 0.0014291525479104947, + "loss": 0.0859, + "step": 37179 + }, + { + "epoch": 0.32274025399085077, + "grad_norm": 0.328125, + "learning_rate": 0.001429124847129393, + "loss": 0.1309, + "step": 37180 + }, + { + "epoch": 0.32274893447105496, + "grad_norm": 0.58984375, + "learning_rate": 0.0014290971459883651, + "loss": 0.1348, + "step": 37181 + }, + { + "epoch": 0.3227576149512591, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014290694444874415, + "loss": 0.1064, + "step": 37182 + }, + { + "epoch": 0.3227662954314633, + "grad_norm": 0.31640625, + "learning_rate": 0.0014290417426266525, + "loss": 0.1289, + "step": 37183 + }, + { + "epoch": 0.3227749759116674, + "grad_norm": 0.109375, + "learning_rate": 0.0014290140404060285, + "loss": 0.1084, + "step": 37184 + }, + { + "epoch": 0.3227836563918716, + "grad_norm": 0.279296875, + "learning_rate": 0.0014289863378255997, + "loss": 0.1089, + "step": 37185 + }, + { + "epoch": 0.32279233687207576, + "grad_norm": 0.201171875, + "learning_rate": 0.001428958634885396, + "loss": 0.1172, + "step": 37186 + }, + { + "epoch": 0.32280101735227995, + "grad_norm": 0.296875, + "learning_rate": 0.0014289309315854487, + "loss": 0.1235, + "step": 37187 + }, + { + "epoch": 0.3228096978324841, + "grad_norm": 0.1474609375, + "learning_rate": 0.001428903227925787, + "loss": 0.0786, + "step": 37188 + }, + { + "epoch": 0.3228183783126883, + "grad_norm": 0.1318359375, + "learning_rate": 0.001428875523906442, + "loss": 0.165, + "step": 37189 + }, + { + "epoch": 0.3228270587928924, + "grad_norm": 0.283203125, + "learning_rate": 0.0014288478195274433, + "loss": 0.0801, + "step": 37190 + }, + { + "epoch": 0.3228357392730966, + "grad_norm": 0.515625, + "learning_rate": 0.0014288201147888218, + "loss": 0.1104, + "step": 37191 + }, + { + "epoch": 0.32284441975330075, + "grad_norm": 0.236328125, + "learning_rate": 0.0014287924096906075, + "loss": 0.1143, + "step": 37192 + }, + { + "epoch": 0.32285310023350494, + "grad_norm": 0.390625, + "learning_rate": 0.0014287647042328308, + "loss": 0.0967, + "step": 37193 + }, + { + "epoch": 0.3228617807137091, + "grad_norm": 0.55078125, + "learning_rate": 0.001428736998415522, + "loss": 0.1299, + "step": 37194 + }, + { + "epoch": 0.32287046119391327, + "grad_norm": 0.203125, + "learning_rate": 0.0014287092922387113, + "loss": 0.0801, + "step": 37195 + }, + { + "epoch": 0.3228791416741174, + "grad_norm": 0.7578125, + "learning_rate": 0.001428681585702429, + "loss": 0.1006, + "step": 37196 + }, + { + "epoch": 0.3228878221543216, + "grad_norm": 0.6328125, + "learning_rate": 0.0014286538788067054, + "loss": 0.0957, + "step": 37197 + }, + { + "epoch": 0.32289650263452574, + "grad_norm": 0.126953125, + "learning_rate": 0.0014286261715515712, + "loss": 0.1177, + "step": 37198 + }, + { + "epoch": 0.32290518311472993, + "grad_norm": 0.49609375, + "learning_rate": 0.001428598463937056, + "loss": 0.0747, + "step": 37199 + }, + { + "epoch": 0.32291386359493407, + "grad_norm": 0.06884765625, + "learning_rate": 0.0014285707559631906, + "loss": 0.0972, + "step": 37200 + }, + { + "epoch": 0.32292254407513826, + "grad_norm": 0.13671875, + "learning_rate": 0.0014285430476300052, + "loss": 0.1064, + "step": 37201 + }, + { + "epoch": 0.3229312245553424, + "grad_norm": 0.125, + "learning_rate": 0.00142851533893753, + "loss": 0.0879, + "step": 37202 + }, + { + "epoch": 0.3229399050355466, + "grad_norm": 0.0888671875, + "learning_rate": 0.0014284876298857954, + "loss": 0.1387, + "step": 37203 + }, + { + "epoch": 0.32294858551575073, + "grad_norm": 1.8671875, + "learning_rate": 0.0014284599204748315, + "loss": 0.1357, + "step": 37204 + }, + { + "epoch": 0.3229572659959549, + "grad_norm": 0.193359375, + "learning_rate": 0.001428432210704669, + "loss": 0.1113, + "step": 37205 + }, + { + "epoch": 0.32296594647615906, + "grad_norm": 0.37109375, + "learning_rate": 0.0014284045005753378, + "loss": 0.127, + "step": 37206 + }, + { + "epoch": 0.32297462695636325, + "grad_norm": 0.71484375, + "learning_rate": 0.0014283767900868684, + "loss": 0.123, + "step": 37207 + }, + { + "epoch": 0.3229833074365674, + "grad_norm": 0.15625, + "learning_rate": 0.0014283490792392913, + "loss": 0.0874, + "step": 37208 + }, + { + "epoch": 0.3229919879167716, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014283213680326362, + "loss": 0.1074, + "step": 37209 + }, + { + "epoch": 0.3230006683969757, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014282936564669342, + "loss": 0.1406, + "step": 37210 + }, + { + "epoch": 0.32300934887717986, + "grad_norm": 0.1728515625, + "learning_rate": 0.001428265944542215, + "loss": 0.0918, + "step": 37211 + }, + { + "epoch": 0.32301802935738405, + "grad_norm": 0.26953125, + "learning_rate": 0.001428238232258509, + "loss": 0.1016, + "step": 37212 + }, + { + "epoch": 0.3230267098375882, + "grad_norm": 0.08056640625, + "learning_rate": 0.0014282105196158465, + "loss": 0.083, + "step": 37213 + }, + { + "epoch": 0.3230353903177924, + "grad_norm": 0.4453125, + "learning_rate": 0.0014281828066142583, + "loss": 0.0996, + "step": 37214 + }, + { + "epoch": 0.3230440707979965, + "grad_norm": 0.5859375, + "learning_rate": 0.0014281550932537745, + "loss": 0.1445, + "step": 37215 + }, + { + "epoch": 0.3230527512782007, + "grad_norm": 0.6015625, + "learning_rate": 0.0014281273795344245, + "loss": 0.1211, + "step": 37216 + }, + { + "epoch": 0.32306143175840485, + "grad_norm": 0.306640625, + "learning_rate": 0.0014280996654562393, + "loss": 0.0918, + "step": 37217 + }, + { + "epoch": 0.32307011223860904, + "grad_norm": 0.333984375, + "learning_rate": 0.0014280719510192496, + "loss": 0.0928, + "step": 37218 + }, + { + "epoch": 0.3230787927188132, + "grad_norm": 0.291015625, + "learning_rate": 0.0014280442362234855, + "loss": 0.0977, + "step": 37219 + }, + { + "epoch": 0.32308747319901737, + "grad_norm": 0.283203125, + "learning_rate": 0.001428016521068977, + "loss": 0.1406, + "step": 37220 + }, + { + "epoch": 0.3230961536792215, + "grad_norm": 0.224609375, + "learning_rate": 0.0014279888055557545, + "loss": 0.1182, + "step": 37221 + }, + { + "epoch": 0.3231048341594257, + "grad_norm": 0.08056640625, + "learning_rate": 0.0014279610896838484, + "loss": 0.0942, + "step": 37222 + }, + { + "epoch": 0.32311351463962984, + "grad_norm": 0.263671875, + "learning_rate": 0.001427933373453289, + "loss": 0.0884, + "step": 37223 + }, + { + "epoch": 0.32312219511983403, + "grad_norm": 0.111328125, + "learning_rate": 0.0014279056568641067, + "loss": 0.0874, + "step": 37224 + }, + { + "epoch": 0.32313087560003817, + "grad_norm": 0.158203125, + "learning_rate": 0.0014278779399163315, + "loss": 0.0957, + "step": 37225 + }, + { + "epoch": 0.32313955608024236, + "grad_norm": 0.30859375, + "learning_rate": 0.001427850222609994, + "loss": 0.1245, + "step": 37226 + }, + { + "epoch": 0.3231482365604465, + "grad_norm": 0.447265625, + "learning_rate": 0.0014278225049451244, + "loss": 0.1069, + "step": 37227 + }, + { + "epoch": 0.3231569170406507, + "grad_norm": 0.306640625, + "learning_rate": 0.0014277947869217528, + "loss": 0.0986, + "step": 37228 + }, + { + "epoch": 0.32316559752085483, + "grad_norm": 0.26953125, + "learning_rate": 0.00142776706853991, + "loss": 0.1377, + "step": 37229 + }, + { + "epoch": 0.323174278001059, + "grad_norm": 0.1513671875, + "learning_rate": 0.001427739349799626, + "loss": 0.125, + "step": 37230 + }, + { + "epoch": 0.32318295848126316, + "grad_norm": 0.10546875, + "learning_rate": 0.0014277116307009315, + "loss": 0.0928, + "step": 37231 + }, + { + "epoch": 0.32319163896146735, + "grad_norm": 0.0947265625, + "learning_rate": 0.0014276839112438563, + "loss": 0.1001, + "step": 37232 + }, + { + "epoch": 0.3232003194416715, + "grad_norm": 0.126953125, + "learning_rate": 0.0014276561914284306, + "loss": 0.1191, + "step": 37233 + }, + { + "epoch": 0.3232089999218757, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014276284712546855, + "loss": 0.1016, + "step": 37234 + }, + { + "epoch": 0.3232176804020798, + "grad_norm": 0.2392578125, + "learning_rate": 0.0014276007507226503, + "loss": 0.1172, + "step": 37235 + }, + { + "epoch": 0.323226360882284, + "grad_norm": 0.1708984375, + "learning_rate": 0.001427573029832356, + "loss": 0.0693, + "step": 37236 + }, + { + "epoch": 0.32323504136248815, + "grad_norm": 0.2265625, + "learning_rate": 0.0014275453085838329, + "loss": 0.1123, + "step": 37237 + }, + { + "epoch": 0.32324372184269234, + "grad_norm": 0.2080078125, + "learning_rate": 0.0014275175869771111, + "loss": 0.1289, + "step": 37238 + }, + { + "epoch": 0.3232524023228965, + "grad_norm": 0.158203125, + "learning_rate": 0.001427489865012221, + "loss": 0.0967, + "step": 37239 + }, + { + "epoch": 0.3232610828031007, + "grad_norm": 0.2421875, + "learning_rate": 0.001427462142689193, + "loss": 0.1201, + "step": 37240 + }, + { + "epoch": 0.3232697632833048, + "grad_norm": 0.421875, + "learning_rate": 0.0014274344200080572, + "loss": 0.1514, + "step": 37241 + }, + { + "epoch": 0.323278443763509, + "grad_norm": 1.2109375, + "learning_rate": 0.001427406696968844, + "loss": 0.0864, + "step": 37242 + }, + { + "epoch": 0.32328712424371314, + "grad_norm": 0.224609375, + "learning_rate": 0.0014273789735715838, + "loss": 0.1104, + "step": 37243 + }, + { + "epoch": 0.32329580472391733, + "grad_norm": 0.97265625, + "learning_rate": 0.0014273512498163068, + "loss": 0.0962, + "step": 37244 + }, + { + "epoch": 0.3233044852041215, + "grad_norm": 0.427734375, + "learning_rate": 0.0014273235257030437, + "loss": 0.1191, + "step": 37245 + }, + { + "epoch": 0.32331316568432567, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014272958012318242, + "loss": 0.1025, + "step": 37246 + }, + { + "epoch": 0.3233218461645298, + "grad_norm": 0.4375, + "learning_rate": 0.0014272680764026789, + "loss": 0.0967, + "step": 37247 + }, + { + "epoch": 0.323330526644734, + "grad_norm": 0.2490234375, + "learning_rate": 0.0014272403512156381, + "loss": 0.1211, + "step": 37248 + }, + { + "epoch": 0.32333920712493813, + "grad_norm": 0.275390625, + "learning_rate": 0.0014272126256707324, + "loss": 0.1055, + "step": 37249 + }, + { + "epoch": 0.3233478876051423, + "grad_norm": 0.0712890625, + "learning_rate": 0.001427184899767992, + "loss": 0.0801, + "step": 37250 + }, + { + "epoch": 0.32335656808534646, + "grad_norm": 0.2265625, + "learning_rate": 0.001427157173507447, + "loss": 0.1099, + "step": 37251 + }, + { + "epoch": 0.32336524856555066, + "grad_norm": 0.09326171875, + "learning_rate": 0.0014271294468891278, + "loss": 0.0786, + "step": 37252 + }, + { + "epoch": 0.3233739290457548, + "grad_norm": 0.498046875, + "learning_rate": 0.0014271017199130648, + "loss": 0.127, + "step": 37253 + }, + { + "epoch": 0.323382609525959, + "grad_norm": 0.19140625, + "learning_rate": 0.001427073992579288, + "loss": 0.1406, + "step": 37254 + }, + { + "epoch": 0.3233912900061631, + "grad_norm": 0.294921875, + "learning_rate": 0.0014270462648878282, + "loss": 0.0669, + "step": 37255 + }, + { + "epoch": 0.3233999704863673, + "grad_norm": 0.71875, + "learning_rate": 0.0014270185368387154, + "loss": 0.0742, + "step": 37256 + }, + { + "epoch": 0.32340865096657145, + "grad_norm": 0.1513671875, + "learning_rate": 0.00142699080843198, + "loss": 0.0957, + "step": 37257 + }, + { + "epoch": 0.32341733144677565, + "grad_norm": 0.1328125, + "learning_rate": 0.0014269630796676527, + "loss": 0.1318, + "step": 37258 + }, + { + "epoch": 0.3234260119269798, + "grad_norm": 0.1875, + "learning_rate": 0.0014269353505457635, + "loss": 0.0708, + "step": 37259 + }, + { + "epoch": 0.323434692407184, + "grad_norm": 0.2412109375, + "learning_rate": 0.0014269076210663423, + "loss": 0.1201, + "step": 37260 + }, + { + "epoch": 0.3234433728873881, + "grad_norm": 0.287109375, + "learning_rate": 0.0014268798912294202, + "loss": 0.0981, + "step": 37261 + }, + { + "epoch": 0.3234520533675923, + "grad_norm": 0.1435546875, + "learning_rate": 0.0014268521610350268, + "loss": 0.0879, + "step": 37262 + }, + { + "epoch": 0.32346073384779644, + "grad_norm": 0.09375, + "learning_rate": 0.0014268244304831933, + "loss": 0.1113, + "step": 37263 + }, + { + "epoch": 0.32346941432800064, + "grad_norm": 0.29296875, + "learning_rate": 0.001426796699573949, + "loss": 0.0977, + "step": 37264 + }, + { + "epoch": 0.3234780948082048, + "grad_norm": 0.328125, + "learning_rate": 0.0014267689683073249, + "loss": 0.1299, + "step": 37265 + }, + { + "epoch": 0.32348677528840897, + "grad_norm": 0.283203125, + "learning_rate": 0.0014267412366833511, + "loss": 0.0967, + "step": 37266 + }, + { + "epoch": 0.3234954557686131, + "grad_norm": 0.345703125, + "learning_rate": 0.0014267135047020584, + "loss": 0.1074, + "step": 37267 + }, + { + "epoch": 0.3235041362488173, + "grad_norm": 0.27734375, + "learning_rate": 0.0014266857723634764, + "loss": 0.1094, + "step": 37268 + }, + { + "epoch": 0.32351281672902144, + "grad_norm": 0.4296875, + "learning_rate": 0.0014266580396676356, + "loss": 0.1113, + "step": 37269 + }, + { + "epoch": 0.32352149720922563, + "grad_norm": 0.0791015625, + "learning_rate": 0.001426630306614567, + "loss": 0.0928, + "step": 37270 + }, + { + "epoch": 0.32353017768942977, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014266025732042997, + "loss": 0.0933, + "step": 37271 + }, + { + "epoch": 0.32353885816963396, + "grad_norm": 0.287109375, + "learning_rate": 0.0014265748394368654, + "loss": 0.0884, + "step": 37272 + }, + { + "epoch": 0.3235475386498381, + "grad_norm": 0.33203125, + "learning_rate": 0.0014265471053122934, + "loss": 0.1147, + "step": 37273 + }, + { + "epoch": 0.3235562191300423, + "grad_norm": 0.234375, + "learning_rate": 0.0014265193708306143, + "loss": 0.1357, + "step": 37274 + }, + { + "epoch": 0.3235648996102464, + "grad_norm": 0.462890625, + "learning_rate": 0.0014264916359918588, + "loss": 0.0957, + "step": 37275 + }, + { + "epoch": 0.3235735800904506, + "grad_norm": 0.353515625, + "learning_rate": 0.0014264639007960567, + "loss": 0.0874, + "step": 37276 + }, + { + "epoch": 0.32358226057065476, + "grad_norm": 0.326171875, + "learning_rate": 0.0014264361652432385, + "loss": 0.0967, + "step": 37277 + }, + { + "epoch": 0.32359094105085895, + "grad_norm": 0.515625, + "learning_rate": 0.001426408429333435, + "loss": 0.0947, + "step": 37278 + }, + { + "epoch": 0.3235996215310631, + "grad_norm": 0.67578125, + "learning_rate": 0.001426380693066676, + "loss": 0.1484, + "step": 37279 + }, + { + "epoch": 0.3236083020112673, + "grad_norm": 0.11279296875, + "learning_rate": 0.0014263529564429922, + "loss": 0.123, + "step": 37280 + }, + { + "epoch": 0.3236169824914714, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014263252194624133, + "loss": 0.0669, + "step": 37281 + }, + { + "epoch": 0.3236256629716756, + "grad_norm": 0.08349609375, + "learning_rate": 0.0014262974821249701, + "loss": 0.0981, + "step": 37282 + }, + { + "epoch": 0.32363434345187975, + "grad_norm": 0.2353515625, + "learning_rate": 0.001426269744430693, + "loss": 0.0986, + "step": 37283 + }, + { + "epoch": 0.32364302393208394, + "grad_norm": 0.53125, + "learning_rate": 0.001426242006379612, + "loss": 0.0898, + "step": 37284 + }, + { + "epoch": 0.3236517044122881, + "grad_norm": 0.2119140625, + "learning_rate": 0.001426214267971758, + "loss": 0.1167, + "step": 37285 + }, + { + "epoch": 0.32366038489249227, + "grad_norm": 0.275390625, + "learning_rate": 0.0014261865292071608, + "loss": 0.1279, + "step": 37286 + }, + { + "epoch": 0.3236690653726964, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014261587900858508, + "loss": 0.0933, + "step": 37287 + }, + { + "epoch": 0.3236777458529006, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014261310506078586, + "loss": 0.1069, + "step": 37288 + }, + { + "epoch": 0.32368642633310474, + "grad_norm": 0.205078125, + "learning_rate": 0.0014261033107732144, + "loss": 0.1069, + "step": 37289 + }, + { + "epoch": 0.32369510681330893, + "grad_norm": 0.2138671875, + "learning_rate": 0.0014260755705819484, + "loss": 0.1465, + "step": 37290 + }, + { + "epoch": 0.32370378729351307, + "grad_norm": 0.0830078125, + "learning_rate": 0.001426047830034091, + "loss": 0.082, + "step": 37291 + }, + { + "epoch": 0.32371246777371726, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014260200891296731, + "loss": 0.1221, + "step": 37292 + }, + { + "epoch": 0.3237211482539214, + "grad_norm": 0.1484375, + "learning_rate": 0.001425992347868724, + "loss": 0.1172, + "step": 37293 + }, + { + "epoch": 0.3237298287341256, + "grad_norm": 0.2890625, + "learning_rate": 0.0014259646062512747, + "loss": 0.1318, + "step": 37294 + }, + { + "epoch": 0.32373850921432973, + "grad_norm": 0.494140625, + "learning_rate": 0.0014259368642773555, + "loss": 0.0913, + "step": 37295 + }, + { + "epoch": 0.3237471896945339, + "grad_norm": 0.123046875, + "learning_rate": 0.0014259091219469965, + "loss": 0.1816, + "step": 37296 + }, + { + "epoch": 0.32375587017473806, + "grad_norm": 0.53125, + "learning_rate": 0.0014258813792602284, + "loss": 0.0908, + "step": 37297 + }, + { + "epoch": 0.32376455065494225, + "grad_norm": 0.466796875, + "learning_rate": 0.001425853636217081, + "loss": 0.0986, + "step": 37298 + }, + { + "epoch": 0.3237732311351464, + "grad_norm": 0.138671875, + "learning_rate": 0.001425825892817585, + "loss": 0.0815, + "step": 37299 + }, + { + "epoch": 0.3237819116153506, + "grad_norm": 0.48046875, + "learning_rate": 0.0014257981490617708, + "loss": 0.1104, + "step": 37300 + }, + { + "epoch": 0.3237905920955547, + "grad_norm": 0.0654296875, + "learning_rate": 0.0014257704049496687, + "loss": 0.0718, + "step": 37301 + }, + { + "epoch": 0.3237992725757589, + "grad_norm": 0.1513671875, + "learning_rate": 0.001425742660481309, + "loss": 0.1182, + "step": 37302 + }, + { + "epoch": 0.32380795305596305, + "grad_norm": 0.267578125, + "learning_rate": 0.001425714915656722, + "loss": 0.1206, + "step": 37303 + }, + { + "epoch": 0.32381663353616724, + "grad_norm": 0.10205078125, + "learning_rate": 0.0014256871704759378, + "loss": 0.1025, + "step": 37304 + }, + { + "epoch": 0.3238253140163714, + "grad_norm": 0.2451171875, + "learning_rate": 0.0014256594249389871, + "loss": 0.0864, + "step": 37305 + }, + { + "epoch": 0.3238339944965756, + "grad_norm": 0.2060546875, + "learning_rate": 0.0014256316790459006, + "loss": 0.0869, + "step": 37306 + }, + { + "epoch": 0.3238426749767797, + "grad_norm": 0.50390625, + "learning_rate": 0.0014256039327967076, + "loss": 0.1367, + "step": 37307 + }, + { + "epoch": 0.3238513554569839, + "grad_norm": 0.244140625, + "learning_rate": 0.0014255761861914394, + "loss": 0.1016, + "step": 37308 + }, + { + "epoch": 0.32386003593718804, + "grad_norm": 0.1845703125, + "learning_rate": 0.0014255484392301258, + "loss": 0.1201, + "step": 37309 + }, + { + "epoch": 0.32386871641739223, + "grad_norm": 0.11279296875, + "learning_rate": 0.0014255206919127975, + "loss": 0.127, + "step": 37310 + }, + { + "epoch": 0.32387739689759637, + "grad_norm": 0.2431640625, + "learning_rate": 0.0014254929442394843, + "loss": 0.106, + "step": 37311 + }, + { + "epoch": 0.32388607737780056, + "grad_norm": 0.53515625, + "learning_rate": 0.0014254651962102175, + "loss": 0.1084, + "step": 37312 + }, + { + "epoch": 0.3238947578580047, + "grad_norm": 0.232421875, + "learning_rate": 0.0014254374478250263, + "loss": 0.0898, + "step": 37313 + }, + { + "epoch": 0.3239034383382089, + "grad_norm": 1.1171875, + "learning_rate": 0.0014254096990839419, + "loss": 0.1328, + "step": 37314 + }, + { + "epoch": 0.32391211881841303, + "grad_norm": 0.46875, + "learning_rate": 0.0014253819499869943, + "loss": 0.0742, + "step": 37315 + }, + { + "epoch": 0.3239207992986172, + "grad_norm": 0.10009765625, + "learning_rate": 0.0014253542005342138, + "loss": 0.084, + "step": 37316 + }, + { + "epoch": 0.32392947977882136, + "grad_norm": 0.283203125, + "learning_rate": 0.0014253264507256306, + "loss": 0.0737, + "step": 37317 + }, + { + "epoch": 0.32393816025902555, + "grad_norm": 0.212890625, + "learning_rate": 0.0014252987005612758, + "loss": 0.1641, + "step": 37318 + }, + { + "epoch": 0.3239468407392297, + "grad_norm": 0.1416015625, + "learning_rate": 0.001425270950041179, + "loss": 0.0947, + "step": 37319 + }, + { + "epoch": 0.3239555212194339, + "grad_norm": 0.30078125, + "learning_rate": 0.0014252431991653707, + "loss": 0.0908, + "step": 37320 + }, + { + "epoch": 0.323964201699638, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014252154479338813, + "loss": 0.1123, + "step": 37321 + }, + { + "epoch": 0.3239728821798422, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014251876963467412, + "loss": 0.1104, + "step": 37322 + }, + { + "epoch": 0.32398156266004635, + "grad_norm": 0.20703125, + "learning_rate": 0.0014251599444039806, + "loss": 0.0957, + "step": 37323 + }, + { + "epoch": 0.32399024314025054, + "grad_norm": 0.11181640625, + "learning_rate": 0.0014251321921056302, + "loss": 0.1094, + "step": 37324 + }, + { + "epoch": 0.3239989236204547, + "grad_norm": 0.4765625, + "learning_rate": 0.00142510443945172, + "loss": 0.1406, + "step": 37325 + }, + { + "epoch": 0.3240076041006589, + "grad_norm": 0.2890625, + "learning_rate": 0.0014250766864422808, + "loss": 0.1055, + "step": 37326 + }, + { + "epoch": 0.324016284580863, + "grad_norm": 0.220703125, + "learning_rate": 0.0014250489330773421, + "loss": 0.0898, + "step": 37327 + }, + { + "epoch": 0.3240249650610672, + "grad_norm": 0.193359375, + "learning_rate": 0.0014250211793569352, + "loss": 0.0752, + "step": 37328 + }, + { + "epoch": 0.32403364554127134, + "grad_norm": 0.12158203125, + "learning_rate": 0.0014249934252810897, + "loss": 0.0908, + "step": 37329 + }, + { + "epoch": 0.32404232602147554, + "grad_norm": 0.1259765625, + "learning_rate": 0.0014249656708498364, + "loss": 0.0933, + "step": 37330 + }, + { + "epoch": 0.3240510065016797, + "grad_norm": 0.482421875, + "learning_rate": 0.0014249379160632056, + "loss": 0.1113, + "step": 37331 + }, + { + "epoch": 0.32405968698188387, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014249101609212275, + "loss": 0.1055, + "step": 37332 + }, + { + "epoch": 0.324068367462088, + "grad_norm": 0.203125, + "learning_rate": 0.0014248824054239324, + "loss": 0.249, + "step": 37333 + }, + { + "epoch": 0.32407704794229214, + "grad_norm": 0.267578125, + "learning_rate": 0.001424854649571351, + "loss": 0.1216, + "step": 37334 + }, + { + "epoch": 0.32408572842249633, + "grad_norm": 0.279296875, + "learning_rate": 0.0014248268933635135, + "loss": 0.1196, + "step": 37335 + }, + { + "epoch": 0.32409440890270047, + "grad_norm": 0.29296875, + "learning_rate": 0.0014247991368004501, + "loss": 0.0898, + "step": 37336 + }, + { + "epoch": 0.32410308938290466, + "grad_norm": 0.1533203125, + "learning_rate": 0.001424771379882191, + "loss": 0.1084, + "step": 37337 + }, + { + "epoch": 0.3241117698631088, + "grad_norm": 0.1279296875, + "learning_rate": 0.001424743622608767, + "loss": 0.1104, + "step": 37338 + }, + { + "epoch": 0.324120450343313, + "grad_norm": 0.0888671875, + "learning_rate": 0.0014247158649802086, + "loss": 0.1138, + "step": 37339 + }, + { + "epoch": 0.32412913082351713, + "grad_norm": 0.201171875, + "learning_rate": 0.0014246881069965452, + "loss": 0.0996, + "step": 37340 + }, + { + "epoch": 0.3241378113037213, + "grad_norm": 0.322265625, + "learning_rate": 0.001424660348657808, + "loss": 0.126, + "step": 37341 + }, + { + "epoch": 0.32414649178392546, + "grad_norm": 0.15625, + "learning_rate": 0.0014246325899640272, + "loss": 0.1055, + "step": 37342 + }, + { + "epoch": 0.32415517226412965, + "grad_norm": 0.5234375, + "learning_rate": 0.0014246048309152328, + "loss": 0.1582, + "step": 37343 + }, + { + "epoch": 0.3241638527443338, + "grad_norm": 0.255859375, + "learning_rate": 0.0014245770715114558, + "loss": 0.1123, + "step": 37344 + }, + { + "epoch": 0.324172533224538, + "grad_norm": 0.271484375, + "learning_rate": 0.0014245493117527263, + "loss": 0.1133, + "step": 37345 + }, + { + "epoch": 0.3241812137047421, + "grad_norm": 0.322265625, + "learning_rate": 0.001424521551639074, + "loss": 0.0952, + "step": 37346 + }, + { + "epoch": 0.3241898941849463, + "grad_norm": 0.244140625, + "learning_rate": 0.0014244937911705303, + "loss": 0.1162, + "step": 37347 + }, + { + "epoch": 0.32419857466515045, + "grad_norm": 0.181640625, + "learning_rate": 0.0014244660303471249, + "loss": 0.0903, + "step": 37348 + }, + { + "epoch": 0.32420725514535464, + "grad_norm": 0.34375, + "learning_rate": 0.0014244382691688884, + "loss": 0.0757, + "step": 37349 + }, + { + "epoch": 0.3242159356255588, + "grad_norm": 0.1025390625, + "learning_rate": 0.0014244105076358508, + "loss": 0.0894, + "step": 37350 + }, + { + "epoch": 0.324224616105763, + "grad_norm": 0.28125, + "learning_rate": 0.001424382745748043, + "loss": 0.0869, + "step": 37351 + }, + { + "epoch": 0.3242332965859671, + "grad_norm": 0.375, + "learning_rate": 0.001424354983505495, + "loss": 0.0684, + "step": 37352 + }, + { + "epoch": 0.3242419770661713, + "grad_norm": 0.3828125, + "learning_rate": 0.0014243272209082372, + "loss": 0.167, + "step": 37353 + }, + { + "epoch": 0.32425065754637544, + "grad_norm": 0.158203125, + "learning_rate": 0.0014242994579563, + "loss": 0.0928, + "step": 37354 + }, + { + "epoch": 0.32425933802657964, + "grad_norm": 0.85546875, + "learning_rate": 0.0014242716946497138, + "loss": 0.1475, + "step": 37355 + }, + { + "epoch": 0.3242680185067838, + "grad_norm": 0.1318359375, + "learning_rate": 0.001424243930988509, + "loss": 0.1104, + "step": 37356 + }, + { + "epoch": 0.32427669898698797, + "grad_norm": 0.1796875, + "learning_rate": 0.0014242161669727163, + "loss": 0.0879, + "step": 37357 + }, + { + "epoch": 0.3242853794671921, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014241884026023654, + "loss": 0.0713, + "step": 37358 + }, + { + "epoch": 0.3242940599473963, + "grad_norm": 0.09716796875, + "learning_rate": 0.0014241606378774866, + "loss": 0.0957, + "step": 37359 + }, + { + "epoch": 0.32430274042760043, + "grad_norm": 0.44921875, + "learning_rate": 0.001424132872798111, + "loss": 0.0908, + "step": 37360 + }, + { + "epoch": 0.3243114209078046, + "grad_norm": 0.10498046875, + "learning_rate": 0.001424105107364268, + "loss": 0.0771, + "step": 37361 + }, + { + "epoch": 0.32432010138800876, + "grad_norm": 0.12353515625, + "learning_rate": 0.0014240773415759892, + "loss": 0.1348, + "step": 37362 + }, + { + "epoch": 0.32432878186821296, + "grad_norm": 0.14453125, + "learning_rate": 0.0014240495754333038, + "loss": 0.0835, + "step": 37363 + }, + { + "epoch": 0.3243374623484171, + "grad_norm": 0.08935546875, + "learning_rate": 0.0014240218089362432, + "loss": 0.124, + "step": 37364 + }, + { + "epoch": 0.3243461428286213, + "grad_norm": 0.78125, + "learning_rate": 0.0014239940420848367, + "loss": 0.1357, + "step": 37365 + }, + { + "epoch": 0.3243548233088254, + "grad_norm": 0.5546875, + "learning_rate": 0.0014239662748791156, + "loss": 0.168, + "step": 37366 + }, + { + "epoch": 0.3243635037890296, + "grad_norm": 0.2333984375, + "learning_rate": 0.0014239385073191095, + "loss": 0.125, + "step": 37367 + }, + { + "epoch": 0.32437218426923375, + "grad_norm": 0.11669921875, + "learning_rate": 0.0014239107394048492, + "loss": 0.0747, + "step": 37368 + }, + { + "epoch": 0.32438086474943795, + "grad_norm": 0.10498046875, + "learning_rate": 0.001423882971136365, + "loss": 0.0894, + "step": 37369 + }, + { + "epoch": 0.3243895452296421, + "grad_norm": 0.203125, + "learning_rate": 0.0014238552025136874, + "loss": 0.1123, + "step": 37370 + }, + { + "epoch": 0.3243982257098463, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014238274335368463, + "loss": 0.0815, + "step": 37371 + }, + { + "epoch": 0.3244069061900504, + "grad_norm": 0.5859375, + "learning_rate": 0.0014237996642058725, + "loss": 0.082, + "step": 37372 + }, + { + "epoch": 0.3244155866702546, + "grad_norm": 0.20703125, + "learning_rate": 0.0014237718945207965, + "loss": 0.0742, + "step": 37373 + }, + { + "epoch": 0.32442426715045874, + "grad_norm": 0.251953125, + "learning_rate": 0.0014237441244816483, + "loss": 0.1245, + "step": 37374 + }, + { + "epoch": 0.32443294763066294, + "grad_norm": 1.765625, + "learning_rate": 0.0014237163540884584, + "loss": 0.3789, + "step": 37375 + }, + { + "epoch": 0.3244416281108671, + "grad_norm": 0.4453125, + "learning_rate": 0.001423688583341257, + "loss": 0.1348, + "step": 37376 + }, + { + "epoch": 0.32445030859107127, + "grad_norm": 0.283203125, + "learning_rate": 0.0014236608122400747, + "loss": 0.0776, + "step": 37377 + }, + { + "epoch": 0.3244589890712754, + "grad_norm": 0.25390625, + "learning_rate": 0.0014236330407849417, + "loss": 0.1299, + "step": 37378 + }, + { + "epoch": 0.3244676695514796, + "grad_norm": 0.68359375, + "learning_rate": 0.001423605268975889, + "loss": 0.1123, + "step": 37379 + }, + { + "epoch": 0.32447635003168374, + "grad_norm": 0.291015625, + "learning_rate": 0.0014235774968129457, + "loss": 0.085, + "step": 37380 + }, + { + "epoch": 0.32448503051188793, + "grad_norm": 1.0625, + "learning_rate": 0.0014235497242961433, + "loss": 0.0908, + "step": 37381 + }, + { + "epoch": 0.32449371099209207, + "grad_norm": 0.248046875, + "learning_rate": 0.0014235219514255116, + "loss": 0.1348, + "step": 37382 + }, + { + "epoch": 0.32450239147229626, + "grad_norm": 0.51953125, + "learning_rate": 0.0014234941782010813, + "loss": 0.0732, + "step": 37383 + }, + { + "epoch": 0.3245110719525004, + "grad_norm": 0.5703125, + "learning_rate": 0.001423466404622883, + "loss": 0.0859, + "step": 37384 + }, + { + "epoch": 0.3245197524327046, + "grad_norm": 0.12255859375, + "learning_rate": 0.0014234386306909458, + "loss": 0.1143, + "step": 37385 + }, + { + "epoch": 0.3245284329129087, + "grad_norm": 1.0546875, + "learning_rate": 0.0014234108564053017, + "loss": 0.1025, + "step": 37386 + }, + { + "epoch": 0.3245371133931129, + "grad_norm": 0.93359375, + "learning_rate": 0.0014233830817659803, + "loss": 0.1162, + "step": 37387 + }, + { + "epoch": 0.32454579387331706, + "grad_norm": 0.4609375, + "learning_rate": 0.0014233553067730118, + "loss": 0.1201, + "step": 37388 + }, + { + "epoch": 0.32455447435352125, + "grad_norm": 0.251953125, + "learning_rate": 0.0014233275314264266, + "loss": 0.125, + "step": 37389 + }, + { + "epoch": 0.3245631548337254, + "grad_norm": 0.220703125, + "learning_rate": 0.0014232997557262555, + "loss": 0.0811, + "step": 37390 + }, + { + "epoch": 0.3245718353139296, + "grad_norm": 0.392578125, + "learning_rate": 0.0014232719796725287, + "loss": 0.0889, + "step": 37391 + }, + { + "epoch": 0.3245805157941337, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014232442032652765, + "loss": 0.0967, + "step": 37392 + }, + { + "epoch": 0.3245891962743379, + "grad_norm": 0.11328125, + "learning_rate": 0.0014232164265045295, + "loss": 0.0889, + "step": 37393 + }, + { + "epoch": 0.32459787675454205, + "grad_norm": 0.103515625, + "learning_rate": 0.0014231886493903173, + "loss": 0.1357, + "step": 37394 + }, + { + "epoch": 0.32460655723474624, + "grad_norm": 0.2216796875, + "learning_rate": 0.0014231608719226711, + "loss": 0.1367, + "step": 37395 + }, + { + "epoch": 0.3246152377149504, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014231330941016217, + "loss": 0.0889, + "step": 37396 + }, + { + "epoch": 0.32462391819515457, + "grad_norm": 0.486328125, + "learning_rate": 0.001423105315927198, + "loss": 0.1514, + "step": 37397 + }, + { + "epoch": 0.3246325986753587, + "grad_norm": 0.42578125, + "learning_rate": 0.0014230775373994312, + "loss": 0.0752, + "step": 37398 + }, + { + "epoch": 0.3246412791555629, + "grad_norm": 0.51171875, + "learning_rate": 0.0014230497585183519, + "loss": 0.1289, + "step": 37399 + }, + { + "epoch": 0.32464995963576704, + "grad_norm": 0.34375, + "learning_rate": 0.0014230219792839903, + "loss": 0.0928, + "step": 37400 + }, + { + "epoch": 0.32465864011597123, + "grad_norm": 0.12109375, + "learning_rate": 0.0014229941996963765, + "loss": 0.1162, + "step": 37401 + }, + { + "epoch": 0.32466732059617537, + "grad_norm": 0.2138671875, + "learning_rate": 0.0014229664197555412, + "loss": 0.0669, + "step": 37402 + }, + { + "epoch": 0.32467600107637956, + "grad_norm": 0.65625, + "learning_rate": 0.0014229386394615147, + "loss": 0.1309, + "step": 37403 + }, + { + "epoch": 0.3246846815565837, + "grad_norm": 0.1865234375, + "learning_rate": 0.0014229108588143273, + "loss": 0.1299, + "step": 37404 + }, + { + "epoch": 0.3246933620367879, + "grad_norm": 0.115234375, + "learning_rate": 0.0014228830778140098, + "loss": 0.082, + "step": 37405 + }, + { + "epoch": 0.32470204251699203, + "grad_norm": 0.5859375, + "learning_rate": 0.0014228552964605917, + "loss": 0.1562, + "step": 37406 + }, + { + "epoch": 0.3247107229971962, + "grad_norm": 0.181640625, + "learning_rate": 0.0014228275147541041, + "loss": 0.0913, + "step": 37407 + }, + { + "epoch": 0.32471940347740036, + "grad_norm": 0.0986328125, + "learning_rate": 0.001422799732694577, + "loss": 0.0713, + "step": 37408 + }, + { + "epoch": 0.32472808395760455, + "grad_norm": 0.63671875, + "learning_rate": 0.0014227719502820414, + "loss": 0.0674, + "step": 37409 + }, + { + "epoch": 0.3247367644378087, + "grad_norm": 0.31640625, + "learning_rate": 0.001422744167516527, + "loss": 0.0718, + "step": 37410 + }, + { + "epoch": 0.3247454449180129, + "grad_norm": 0.33203125, + "learning_rate": 0.0014227163843980642, + "loss": 0.1025, + "step": 37411 + }, + { + "epoch": 0.324754125398217, + "grad_norm": 0.16015625, + "learning_rate": 0.001422688600926684, + "loss": 0.1123, + "step": 37412 + }, + { + "epoch": 0.3247628058784212, + "grad_norm": 0.4453125, + "learning_rate": 0.0014226608171024162, + "loss": 0.1689, + "step": 37413 + }, + { + "epoch": 0.32477148635862535, + "grad_norm": 0.10009765625, + "learning_rate": 0.0014226330329252914, + "loss": 0.0869, + "step": 37414 + }, + { + "epoch": 0.32478016683882954, + "grad_norm": 0.2294921875, + "learning_rate": 0.0014226052483953399, + "loss": 0.0918, + "step": 37415 + }, + { + "epoch": 0.3247888473190337, + "grad_norm": 0.37109375, + "learning_rate": 0.0014225774635125924, + "loss": 0.0884, + "step": 37416 + }, + { + "epoch": 0.3247975277992379, + "grad_norm": 0.296875, + "learning_rate": 0.0014225496782770788, + "loss": 0.0854, + "step": 37417 + }, + { + "epoch": 0.324806208279442, + "grad_norm": 0.6328125, + "learning_rate": 0.0014225218926888297, + "loss": 0.0947, + "step": 37418 + }, + { + "epoch": 0.3248148887596462, + "grad_norm": 0.3671875, + "learning_rate": 0.0014224941067478754, + "loss": 0.0898, + "step": 37419 + }, + { + "epoch": 0.32482356923985034, + "grad_norm": 0.275390625, + "learning_rate": 0.0014224663204542465, + "loss": 0.1328, + "step": 37420 + }, + { + "epoch": 0.32483224972005453, + "grad_norm": 0.10498046875, + "learning_rate": 0.0014224385338079736, + "loss": 0.1118, + "step": 37421 + }, + { + "epoch": 0.32484093020025867, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014224107468090866, + "loss": 0.0928, + "step": 37422 + }, + { + "epoch": 0.32484961068046286, + "grad_norm": 0.466796875, + "learning_rate": 0.0014223829594576158, + "loss": 0.1079, + "step": 37423 + }, + { + "epoch": 0.324858291160667, + "grad_norm": 0.115234375, + "learning_rate": 0.001422355171753592, + "loss": 0.0737, + "step": 37424 + }, + { + "epoch": 0.3248669716408712, + "grad_norm": 0.2470703125, + "learning_rate": 0.0014223273836970454, + "loss": 0.166, + "step": 37425 + }, + { + "epoch": 0.32487565212107533, + "grad_norm": 0.15625, + "learning_rate": 0.0014222995952880065, + "loss": 0.1152, + "step": 37426 + }, + { + "epoch": 0.3248843326012795, + "grad_norm": 0.330078125, + "learning_rate": 0.0014222718065265054, + "loss": 0.127, + "step": 37427 + }, + { + "epoch": 0.32489301308148366, + "grad_norm": 1.078125, + "learning_rate": 0.001422244017412573, + "loss": 0.1426, + "step": 37428 + }, + { + "epoch": 0.32490169356168785, + "grad_norm": 1.0703125, + "learning_rate": 0.001422216227946239, + "loss": 0.1035, + "step": 37429 + }, + { + "epoch": 0.324910374041892, + "grad_norm": 0.46875, + "learning_rate": 0.0014221884381275345, + "loss": 0.1562, + "step": 37430 + }, + { + "epoch": 0.3249190545220962, + "grad_norm": 0.42578125, + "learning_rate": 0.0014221606479564895, + "loss": 0.0889, + "step": 37431 + }, + { + "epoch": 0.3249277350023003, + "grad_norm": 0.466796875, + "learning_rate": 0.0014221328574331343, + "loss": 0.2793, + "step": 37432 + }, + { + "epoch": 0.3249364154825045, + "grad_norm": 0.2412109375, + "learning_rate": 0.0014221050665574994, + "loss": 0.1191, + "step": 37433 + }, + { + "epoch": 0.32494509596270865, + "grad_norm": 0.15234375, + "learning_rate": 0.0014220772753296156, + "loss": 0.1245, + "step": 37434 + }, + { + "epoch": 0.32495377644291285, + "grad_norm": 0.279296875, + "learning_rate": 0.0014220494837495128, + "loss": 0.1504, + "step": 37435 + }, + { + "epoch": 0.324962456923117, + "grad_norm": 0.478515625, + "learning_rate": 0.0014220216918172214, + "loss": 0.0879, + "step": 37436 + }, + { + "epoch": 0.3249711374033212, + "grad_norm": 0.78515625, + "learning_rate": 0.001421993899532772, + "loss": 0.126, + "step": 37437 + }, + { + "epoch": 0.3249798178835253, + "grad_norm": 0.2890625, + "learning_rate": 0.0014219661068961946, + "loss": 0.0806, + "step": 37438 + }, + { + "epoch": 0.3249884983637295, + "grad_norm": 0.10693359375, + "learning_rate": 0.0014219383139075204, + "loss": 0.1328, + "step": 37439 + }, + { + "epoch": 0.32499717884393364, + "grad_norm": 0.07568359375, + "learning_rate": 0.001421910520566779, + "loss": 0.0952, + "step": 37440 + }, + { + "epoch": 0.32500585932413784, + "grad_norm": 0.3828125, + "learning_rate": 0.001421882726874001, + "loss": 0.126, + "step": 37441 + }, + { + "epoch": 0.325014539804342, + "grad_norm": 0.08349609375, + "learning_rate": 0.0014218549328292175, + "loss": 0.1099, + "step": 37442 + }, + { + "epoch": 0.32502322028454617, + "grad_norm": 0.1767578125, + "learning_rate": 0.0014218271384324574, + "loss": 0.1543, + "step": 37443 + }, + { + "epoch": 0.3250319007647503, + "grad_norm": 0.177734375, + "learning_rate": 0.0014217993436837525, + "loss": 0.0913, + "step": 37444 + }, + { + "epoch": 0.3250405812449545, + "grad_norm": 0.298828125, + "learning_rate": 0.0014217715485831325, + "loss": 0.1123, + "step": 37445 + }, + { + "epoch": 0.32504926172515863, + "grad_norm": 0.70703125, + "learning_rate": 0.001421743753130628, + "loss": 0.1104, + "step": 37446 + }, + { + "epoch": 0.3250579422053628, + "grad_norm": 0.1142578125, + "learning_rate": 0.0014217159573262697, + "loss": 0.0781, + "step": 37447 + }, + { + "epoch": 0.32506662268556696, + "grad_norm": 0.12451171875, + "learning_rate": 0.0014216881611700872, + "loss": 0.0713, + "step": 37448 + }, + { + "epoch": 0.32507530316577116, + "grad_norm": 0.322265625, + "learning_rate": 0.0014216603646621116, + "loss": 0.0742, + "step": 37449 + }, + { + "epoch": 0.3250839836459753, + "grad_norm": 0.15625, + "learning_rate": 0.001421632567802373, + "loss": 0.0801, + "step": 37450 + }, + { + "epoch": 0.3250926641261795, + "grad_norm": 0.35546875, + "learning_rate": 0.0014216047705909018, + "loss": 0.0957, + "step": 37451 + }, + { + "epoch": 0.3251013446063836, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014215769730277286, + "loss": 0.124, + "step": 37452 + }, + { + "epoch": 0.3251100250865878, + "grad_norm": 0.6875, + "learning_rate": 0.0014215491751128836, + "loss": 0.1157, + "step": 37453 + }, + { + "epoch": 0.32511870556679195, + "grad_norm": 0.25, + "learning_rate": 0.001421521376846397, + "loss": 0.1021, + "step": 37454 + }, + { + "epoch": 0.32512738604699615, + "grad_norm": 0.416015625, + "learning_rate": 0.0014214935782283, + "loss": 0.0991, + "step": 37455 + }, + { + "epoch": 0.3251360665272003, + "grad_norm": 0.2490234375, + "learning_rate": 0.001421465779258622, + "loss": 0.0898, + "step": 37456 + }, + { + "epoch": 0.3251447470074044, + "grad_norm": 0.1435546875, + "learning_rate": 0.001421437979937394, + "loss": 0.1021, + "step": 37457 + }, + { + "epoch": 0.3251534274876086, + "grad_norm": 0.1796875, + "learning_rate": 0.0014214101802646462, + "loss": 0.0767, + "step": 37458 + }, + { + "epoch": 0.32516210796781275, + "grad_norm": 0.140625, + "learning_rate": 0.0014213823802404092, + "loss": 0.084, + "step": 37459 + }, + { + "epoch": 0.32517078844801695, + "grad_norm": 0.1708984375, + "learning_rate": 0.001421354579864713, + "loss": 0.0728, + "step": 37460 + }, + { + "epoch": 0.3251794689282211, + "grad_norm": 0.09326171875, + "learning_rate": 0.0014213267791375883, + "loss": 0.0889, + "step": 37461 + }, + { + "epoch": 0.3251881494084253, + "grad_norm": 0.1953125, + "learning_rate": 0.001421298978059066, + "loss": 0.1289, + "step": 37462 + }, + { + "epoch": 0.3251968298886294, + "grad_norm": 0.6171875, + "learning_rate": 0.0014212711766291753, + "loss": 0.0889, + "step": 37463 + }, + { + "epoch": 0.3252055103688336, + "grad_norm": 0.28125, + "learning_rate": 0.0014212433748479474, + "loss": 0.0811, + "step": 37464 + }, + { + "epoch": 0.32521419084903774, + "grad_norm": 0.734375, + "learning_rate": 0.0014212155727154128, + "loss": 0.0967, + "step": 37465 + }, + { + "epoch": 0.32522287132924194, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014211877702316016, + "loss": 0.0859, + "step": 37466 + }, + { + "epoch": 0.3252315518094461, + "grad_norm": 0.283203125, + "learning_rate": 0.001421159967396544, + "loss": 0.0845, + "step": 37467 + }, + { + "epoch": 0.32524023228965027, + "grad_norm": 0.146484375, + "learning_rate": 0.001421132164210271, + "loss": 0.0947, + "step": 37468 + }, + { + "epoch": 0.3252489127698544, + "grad_norm": 0.125, + "learning_rate": 0.0014211043606728124, + "loss": 0.1221, + "step": 37469 + }, + { + "epoch": 0.3252575932500586, + "grad_norm": 0.53125, + "learning_rate": 0.0014210765567841993, + "loss": 0.1074, + "step": 37470 + }, + { + "epoch": 0.32526627373026273, + "grad_norm": 0.11767578125, + "learning_rate": 0.0014210487525444615, + "loss": 0.0986, + "step": 37471 + }, + { + "epoch": 0.3252749542104669, + "grad_norm": 0.10205078125, + "learning_rate": 0.0014210209479536295, + "loss": 0.1084, + "step": 37472 + }, + { + "epoch": 0.32528363469067106, + "grad_norm": 0.4296875, + "learning_rate": 0.001420993143011734, + "loss": 0.123, + "step": 37473 + }, + { + "epoch": 0.32529231517087526, + "grad_norm": 0.255859375, + "learning_rate": 0.0014209653377188053, + "loss": 0.0859, + "step": 37474 + }, + { + "epoch": 0.3253009956510794, + "grad_norm": 0.287109375, + "learning_rate": 0.0014209375320748735, + "loss": 0.0864, + "step": 37475 + }, + { + "epoch": 0.3253096761312836, + "grad_norm": 0.244140625, + "learning_rate": 0.0014209097260799695, + "loss": 0.0986, + "step": 37476 + }, + { + "epoch": 0.3253183566114877, + "grad_norm": 0.12158203125, + "learning_rate": 0.001420881919734123, + "loss": 0.1167, + "step": 37477 + }, + { + "epoch": 0.3253270370916919, + "grad_norm": 0.1044921875, + "learning_rate": 0.001420854113037365, + "loss": 0.0752, + "step": 37478 + }, + { + "epoch": 0.32533571757189605, + "grad_norm": 0.6328125, + "learning_rate": 0.001420826305989726, + "loss": 0.1426, + "step": 37479 + }, + { + "epoch": 0.32534439805210025, + "grad_norm": 0.33203125, + "learning_rate": 0.001420798498591236, + "loss": 0.082, + "step": 37480 + }, + { + "epoch": 0.3253530785323044, + "grad_norm": 0.265625, + "learning_rate": 0.0014207706908419256, + "loss": 0.0894, + "step": 37481 + }, + { + "epoch": 0.3253617590125086, + "grad_norm": 0.515625, + "learning_rate": 0.0014207428827418254, + "loss": 0.1143, + "step": 37482 + }, + { + "epoch": 0.3253704394927127, + "grad_norm": 0.07275390625, + "learning_rate": 0.0014207150742909654, + "loss": 0.0703, + "step": 37483 + }, + { + "epoch": 0.3253791199729169, + "grad_norm": 0.53515625, + "learning_rate": 0.0014206872654893763, + "loss": 0.0918, + "step": 37484 + }, + { + "epoch": 0.32538780045312105, + "grad_norm": 0.09619140625, + "learning_rate": 0.0014206594563370882, + "loss": 0.082, + "step": 37485 + }, + { + "epoch": 0.32539648093332524, + "grad_norm": 0.4609375, + "learning_rate": 0.0014206316468341317, + "loss": 0.0928, + "step": 37486 + }, + { + "epoch": 0.3254051614135294, + "grad_norm": 0.205078125, + "learning_rate": 0.0014206038369805377, + "loss": 0.1167, + "step": 37487 + }, + { + "epoch": 0.32541384189373357, + "grad_norm": 0.4375, + "learning_rate": 0.001420576026776336, + "loss": 0.1436, + "step": 37488 + }, + { + "epoch": 0.3254225223739377, + "grad_norm": 0.2041015625, + "learning_rate": 0.001420548216221557, + "loss": 0.0771, + "step": 37489 + }, + { + "epoch": 0.3254312028541419, + "grad_norm": 0.1708984375, + "learning_rate": 0.0014205204053162313, + "loss": 0.0835, + "step": 37490 + }, + { + "epoch": 0.32543988333434604, + "grad_norm": 0.42578125, + "learning_rate": 0.0014204925940603894, + "loss": 0.1279, + "step": 37491 + }, + { + "epoch": 0.32544856381455023, + "grad_norm": 0.59765625, + "learning_rate": 0.0014204647824540618, + "loss": 0.1152, + "step": 37492 + }, + { + "epoch": 0.32545724429475437, + "grad_norm": 0.1494140625, + "learning_rate": 0.0014204369704972784, + "loss": 0.1621, + "step": 37493 + }, + { + "epoch": 0.32546592477495856, + "grad_norm": 0.1552734375, + "learning_rate": 0.00142040915819007, + "loss": 0.0869, + "step": 37494 + }, + { + "epoch": 0.3254746052551627, + "grad_norm": 0.28125, + "learning_rate": 0.0014203813455324669, + "loss": 0.0967, + "step": 37495 + }, + { + "epoch": 0.3254832857353669, + "grad_norm": 0.09228515625, + "learning_rate": 0.0014203535325244996, + "loss": 0.0894, + "step": 37496 + }, + { + "epoch": 0.325491966215571, + "grad_norm": 0.30859375, + "learning_rate": 0.0014203257191661987, + "loss": 0.0684, + "step": 37497 + }, + { + "epoch": 0.3255006466957752, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014202979054575942, + "loss": 0.0952, + "step": 37498 + }, + { + "epoch": 0.32550932717597936, + "grad_norm": 0.142578125, + "learning_rate": 0.0014202700913987167, + "loss": 0.063, + "step": 37499 + }, + { + "epoch": 0.32551800765618355, + "grad_norm": 0.431640625, + "learning_rate": 0.0014202422769895968, + "loss": 0.0791, + "step": 37500 + }, + { + "epoch": 0.3255266881363877, + "grad_norm": 0.263671875, + "learning_rate": 0.0014202144622302649, + "loss": 0.1025, + "step": 37501 + }, + { + "epoch": 0.3255353686165919, + "grad_norm": 0.109375, + "learning_rate": 0.0014201866471207512, + "loss": 0.1011, + "step": 37502 + }, + { + "epoch": 0.325544049096796, + "grad_norm": 0.314453125, + "learning_rate": 0.0014201588316610858, + "loss": 0.0894, + "step": 37503 + }, + { + "epoch": 0.3255527295770002, + "grad_norm": 0.130859375, + "learning_rate": 0.0014201310158513001, + "loss": 0.1172, + "step": 37504 + }, + { + "epoch": 0.32556141005720435, + "grad_norm": 0.2451171875, + "learning_rate": 0.0014201031996914234, + "loss": 0.0771, + "step": 37505 + }, + { + "epoch": 0.32557009053740854, + "grad_norm": 0.3984375, + "learning_rate": 0.0014200753831814868, + "loss": 0.1045, + "step": 37506 + }, + { + "epoch": 0.3255787710176127, + "grad_norm": 0.07958984375, + "learning_rate": 0.0014200475663215207, + "loss": 0.0962, + "step": 37507 + }, + { + "epoch": 0.32558745149781687, + "grad_norm": 0.3359375, + "learning_rate": 0.0014200197491115554, + "loss": 0.0879, + "step": 37508 + }, + { + "epoch": 0.325596131978021, + "grad_norm": 0.78515625, + "learning_rate": 0.0014199919315516212, + "loss": 0.1621, + "step": 37509 + }, + { + "epoch": 0.3256048124582252, + "grad_norm": 0.4375, + "learning_rate": 0.0014199641136417487, + "loss": 0.1523, + "step": 37510 + }, + { + "epoch": 0.32561349293842934, + "grad_norm": 0.189453125, + "learning_rate": 0.0014199362953819681, + "loss": 0.0698, + "step": 37511 + }, + { + "epoch": 0.32562217341863353, + "grad_norm": 0.171875, + "learning_rate": 0.0014199084767723103, + "loss": 0.0947, + "step": 37512 + }, + { + "epoch": 0.32563085389883767, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014198806578128052, + "loss": 0.1416, + "step": 37513 + }, + { + "epoch": 0.32563953437904186, + "grad_norm": 0.255859375, + "learning_rate": 0.0014198528385034835, + "loss": 0.1953, + "step": 37514 + }, + { + "epoch": 0.325648214859246, + "grad_norm": 0.404296875, + "learning_rate": 0.001419825018844375, + "loss": 0.1279, + "step": 37515 + }, + { + "epoch": 0.3256568953394502, + "grad_norm": 0.099609375, + "learning_rate": 0.0014197971988355112, + "loss": 0.0957, + "step": 37516 + }, + { + "epoch": 0.32566557581965433, + "grad_norm": 0.12060546875, + "learning_rate": 0.0014197693784769219, + "loss": 0.1465, + "step": 37517 + }, + { + "epoch": 0.3256742562998585, + "grad_norm": 0.15234375, + "learning_rate": 0.0014197415577686379, + "loss": 0.1016, + "step": 37518 + }, + { + "epoch": 0.32568293678006266, + "grad_norm": 0.25390625, + "learning_rate": 0.001419713736710689, + "loss": 0.083, + "step": 37519 + }, + { + "epoch": 0.32569161726026685, + "grad_norm": 0.111328125, + "learning_rate": 0.0014196859153031058, + "loss": 0.1104, + "step": 37520 + }, + { + "epoch": 0.325700297740471, + "grad_norm": 0.66796875, + "learning_rate": 0.0014196580935459193, + "loss": 0.1299, + "step": 37521 + }, + { + "epoch": 0.3257089782206752, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014196302714391594, + "loss": 0.0874, + "step": 37522 + }, + { + "epoch": 0.3257176587008793, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014196024489828564, + "loss": 0.0654, + "step": 37523 + }, + { + "epoch": 0.3257263391810835, + "grad_norm": 0.61328125, + "learning_rate": 0.001419574626177041, + "loss": 0.2617, + "step": 37524 + }, + { + "epoch": 0.32573501966128765, + "grad_norm": 0.53125, + "learning_rate": 0.0014195468030217438, + "loss": 0.0977, + "step": 37525 + }, + { + "epoch": 0.32574370014149184, + "grad_norm": 0.341796875, + "learning_rate": 0.001419518979516995, + "loss": 0.0938, + "step": 37526 + }, + { + "epoch": 0.325752380621696, + "grad_norm": 0.2255859375, + "learning_rate": 0.0014194911556628248, + "loss": 0.1143, + "step": 37527 + }, + { + "epoch": 0.3257610611019002, + "grad_norm": 0.2119140625, + "learning_rate": 0.0014194633314592643, + "loss": 0.0752, + "step": 37528 + }, + { + "epoch": 0.3257697415821043, + "grad_norm": 0.333984375, + "learning_rate": 0.0014194355069063432, + "loss": 0.0723, + "step": 37529 + }, + { + "epoch": 0.3257784220623085, + "grad_norm": 0.078125, + "learning_rate": 0.0014194076820040922, + "loss": 0.0938, + "step": 37530 + }, + { + "epoch": 0.32578710254251264, + "grad_norm": 0.1943359375, + "learning_rate": 0.001419379856752542, + "loss": 0.085, + "step": 37531 + }, + { + "epoch": 0.32579578302271683, + "grad_norm": 0.17578125, + "learning_rate": 0.0014193520311517225, + "loss": 0.0996, + "step": 37532 + }, + { + "epoch": 0.32580446350292097, + "grad_norm": 0.498046875, + "learning_rate": 0.0014193242052016647, + "loss": 0.1172, + "step": 37533 + }, + { + "epoch": 0.32581314398312516, + "grad_norm": 0.0849609375, + "learning_rate": 0.0014192963789023982, + "loss": 0.1064, + "step": 37534 + }, + { + "epoch": 0.3258218244633293, + "grad_norm": 0.396484375, + "learning_rate": 0.0014192685522539544, + "loss": 0.1113, + "step": 37535 + }, + { + "epoch": 0.3258305049435335, + "grad_norm": 0.1435546875, + "learning_rate": 0.0014192407252563632, + "loss": 0.103, + "step": 37536 + }, + { + "epoch": 0.32583918542373763, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014192128979096553, + "loss": 0.0776, + "step": 37537 + }, + { + "epoch": 0.3258478659039418, + "grad_norm": 0.13671875, + "learning_rate": 0.001419185070213861, + "loss": 0.0757, + "step": 37538 + }, + { + "epoch": 0.32585654638414596, + "grad_norm": 0.09716796875, + "learning_rate": 0.0014191572421690106, + "loss": 0.1094, + "step": 37539 + }, + { + "epoch": 0.32586522686435015, + "grad_norm": 0.16015625, + "learning_rate": 0.0014191294137751345, + "loss": 0.1406, + "step": 37540 + }, + { + "epoch": 0.3258739073445543, + "grad_norm": 0.1240234375, + "learning_rate": 0.0014191015850322636, + "loss": 0.1182, + "step": 37541 + }, + { + "epoch": 0.3258825878247585, + "grad_norm": 0.1318359375, + "learning_rate": 0.0014190737559404277, + "loss": 0.084, + "step": 37542 + }, + { + "epoch": 0.3258912683049626, + "grad_norm": 0.12060546875, + "learning_rate": 0.0014190459264996574, + "loss": 0.1084, + "step": 37543 + }, + { + "epoch": 0.3258999487851668, + "grad_norm": 0.10888671875, + "learning_rate": 0.0014190180967099837, + "loss": 0.0728, + "step": 37544 + }, + { + "epoch": 0.32590862926537095, + "grad_norm": 0.37890625, + "learning_rate": 0.0014189902665714363, + "loss": 0.0952, + "step": 37545 + }, + { + "epoch": 0.32591730974557515, + "grad_norm": 0.390625, + "learning_rate": 0.0014189624360840458, + "loss": 0.0869, + "step": 37546 + }, + { + "epoch": 0.3259259902257793, + "grad_norm": 0.17578125, + "learning_rate": 0.0014189346052478433, + "loss": 0.0908, + "step": 37547 + }, + { + "epoch": 0.3259346707059835, + "grad_norm": 0.84375, + "learning_rate": 0.0014189067740628582, + "loss": 0.1709, + "step": 37548 + }, + { + "epoch": 0.3259433511861876, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014188789425291217, + "loss": 0.1104, + "step": 37549 + }, + { + "epoch": 0.3259520316663918, + "grad_norm": 0.33984375, + "learning_rate": 0.001418851110646664, + "loss": 0.1099, + "step": 37550 + }, + { + "epoch": 0.32596071214659594, + "grad_norm": 0.515625, + "learning_rate": 0.0014188232784155155, + "loss": 0.1113, + "step": 37551 + }, + { + "epoch": 0.32596939262680014, + "grad_norm": 0.1875, + "learning_rate": 0.0014187954458357067, + "loss": 0.0947, + "step": 37552 + }, + { + "epoch": 0.3259780731070043, + "grad_norm": 0.189453125, + "learning_rate": 0.0014187676129072677, + "loss": 0.1162, + "step": 37553 + }, + { + "epoch": 0.32598675358720847, + "grad_norm": 0.09228515625, + "learning_rate": 0.0014187397796302295, + "loss": 0.1016, + "step": 37554 + }, + { + "epoch": 0.3259954340674126, + "grad_norm": 0.25, + "learning_rate": 0.001418711946004622, + "loss": 0.083, + "step": 37555 + }, + { + "epoch": 0.3260041145476168, + "grad_norm": 0.408203125, + "learning_rate": 0.0014186841120304764, + "loss": 0.1143, + "step": 37556 + }, + { + "epoch": 0.32601279502782093, + "grad_norm": 0.10400390625, + "learning_rate": 0.0014186562777078221, + "loss": 0.0908, + "step": 37557 + }, + { + "epoch": 0.3260214755080251, + "grad_norm": 0.322265625, + "learning_rate": 0.0014186284430366906, + "loss": 0.105, + "step": 37558 + }, + { + "epoch": 0.32603015598822926, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014186006080171114, + "loss": 0.1006, + "step": 37559 + }, + { + "epoch": 0.32603883646843346, + "grad_norm": 0.5546875, + "learning_rate": 0.0014185727726491154, + "loss": 0.126, + "step": 37560 + }, + { + "epoch": 0.3260475169486376, + "grad_norm": 0.220703125, + "learning_rate": 0.0014185449369327333, + "loss": 0.062, + "step": 37561 + }, + { + "epoch": 0.3260561974288418, + "grad_norm": 0.4765625, + "learning_rate": 0.001418517100867995, + "loss": 0.085, + "step": 37562 + }, + { + "epoch": 0.3260648779090459, + "grad_norm": 0.23046875, + "learning_rate": 0.001418489264454931, + "loss": 0.1328, + "step": 37563 + }, + { + "epoch": 0.3260735583892501, + "grad_norm": 0.267578125, + "learning_rate": 0.0014184614276935722, + "loss": 0.1279, + "step": 37564 + }, + { + "epoch": 0.32608223886945426, + "grad_norm": 0.1728515625, + "learning_rate": 0.0014184335905839487, + "loss": 0.0933, + "step": 37565 + }, + { + "epoch": 0.32609091934965845, + "grad_norm": 0.2412109375, + "learning_rate": 0.001418405753126091, + "loss": 0.0791, + "step": 37566 + }, + { + "epoch": 0.3260995998298626, + "grad_norm": 0.57421875, + "learning_rate": 0.0014183779153200296, + "loss": 0.1621, + "step": 37567 + }, + { + "epoch": 0.3261082803100668, + "grad_norm": 0.177734375, + "learning_rate": 0.001418350077165795, + "loss": 0.0918, + "step": 37568 + }, + { + "epoch": 0.3261169607902709, + "grad_norm": 0.177734375, + "learning_rate": 0.0014183222386634174, + "loss": 0.1016, + "step": 37569 + }, + { + "epoch": 0.3261256412704751, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014182943998129274, + "loss": 0.1133, + "step": 37570 + }, + { + "epoch": 0.32613432175067925, + "grad_norm": 0.14453125, + "learning_rate": 0.0014182665606143553, + "loss": 0.1089, + "step": 37571 + }, + { + "epoch": 0.32614300223088344, + "grad_norm": 0.220703125, + "learning_rate": 0.0014182387210677316, + "loss": 0.1133, + "step": 37572 + }, + { + "epoch": 0.3261516827110876, + "grad_norm": 0.73828125, + "learning_rate": 0.0014182108811730872, + "loss": 0.1211, + "step": 37573 + }, + { + "epoch": 0.32616036319129177, + "grad_norm": 0.10400390625, + "learning_rate": 0.001418183040930452, + "loss": 0.082, + "step": 37574 + }, + { + "epoch": 0.3261690436714959, + "grad_norm": 0.291015625, + "learning_rate": 0.0014181552003398566, + "loss": 0.1318, + "step": 37575 + }, + { + "epoch": 0.3261777241517001, + "grad_norm": 0.9296875, + "learning_rate": 0.0014181273594013315, + "loss": 0.1055, + "step": 37576 + }, + { + "epoch": 0.32618640463190424, + "grad_norm": 0.1806640625, + "learning_rate": 0.001418099518114907, + "loss": 0.0728, + "step": 37577 + }, + { + "epoch": 0.32619508511210843, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014180716764806138, + "loss": 0.1279, + "step": 37578 + }, + { + "epoch": 0.32620376559231257, + "grad_norm": 0.57421875, + "learning_rate": 0.0014180438344984822, + "loss": 0.1621, + "step": 37579 + }, + { + "epoch": 0.3262124460725167, + "grad_norm": 0.345703125, + "learning_rate": 0.0014180159921685425, + "loss": 0.0889, + "step": 37580 + }, + { + "epoch": 0.3262211265527209, + "grad_norm": 0.62109375, + "learning_rate": 0.0014179881494908251, + "loss": 0.1328, + "step": 37581 + }, + { + "epoch": 0.32622980703292503, + "grad_norm": 0.37109375, + "learning_rate": 0.0014179603064653607, + "loss": 0.0825, + "step": 37582 + }, + { + "epoch": 0.3262384875131292, + "grad_norm": 0.095703125, + "learning_rate": 0.0014179324630921798, + "loss": 0.0786, + "step": 37583 + }, + { + "epoch": 0.32624716799333336, + "grad_norm": 0.1103515625, + "learning_rate": 0.0014179046193713127, + "loss": 0.0947, + "step": 37584 + }, + { + "epoch": 0.32625584847353756, + "grad_norm": 0.173828125, + "learning_rate": 0.00141787677530279, + "loss": 0.0801, + "step": 37585 + }, + { + "epoch": 0.3262645289537417, + "grad_norm": 0.251953125, + "learning_rate": 0.0014178489308866423, + "loss": 0.1064, + "step": 37586 + }, + { + "epoch": 0.3262732094339459, + "grad_norm": 0.2294921875, + "learning_rate": 0.0014178210861228995, + "loss": 0.0894, + "step": 37587 + }, + { + "epoch": 0.32628188991415, + "grad_norm": 0.1845703125, + "learning_rate": 0.001417793241011592, + "loss": 0.0972, + "step": 37588 + }, + { + "epoch": 0.3262905703943542, + "grad_norm": 0.158203125, + "learning_rate": 0.001417765395552751, + "loss": 0.1133, + "step": 37589 + }, + { + "epoch": 0.32629925087455836, + "grad_norm": 0.341796875, + "learning_rate": 0.0014177375497464065, + "loss": 0.0654, + "step": 37590 + }, + { + "epoch": 0.32630793135476255, + "grad_norm": 0.69140625, + "learning_rate": 0.0014177097035925888, + "loss": 0.1338, + "step": 37591 + }, + { + "epoch": 0.3263166118349667, + "grad_norm": 0.12451171875, + "learning_rate": 0.0014176818570913285, + "loss": 0.1045, + "step": 37592 + }, + { + "epoch": 0.3263252923151709, + "grad_norm": 0.3125, + "learning_rate": 0.0014176540102426565, + "loss": 0.1196, + "step": 37593 + }, + { + "epoch": 0.326333972795375, + "grad_norm": 0.1630859375, + "learning_rate": 0.0014176261630466024, + "loss": 0.0947, + "step": 37594 + }, + { + "epoch": 0.3263426532755792, + "grad_norm": 0.42578125, + "learning_rate": 0.0014175983155031975, + "loss": 0.0947, + "step": 37595 + }, + { + "epoch": 0.32635133375578335, + "grad_norm": 0.48046875, + "learning_rate": 0.0014175704676124717, + "loss": 0.1367, + "step": 37596 + }, + { + "epoch": 0.32636001423598754, + "grad_norm": 0.1396484375, + "learning_rate": 0.0014175426193744553, + "loss": 0.0825, + "step": 37597 + }, + { + "epoch": 0.3263686947161917, + "grad_norm": 0.0859375, + "learning_rate": 0.0014175147707891796, + "loss": 0.1191, + "step": 37598 + }, + { + "epoch": 0.32637737519639587, + "grad_norm": 0.318359375, + "learning_rate": 0.0014174869218566742, + "loss": 0.1089, + "step": 37599 + }, + { + "epoch": 0.3263860556766, + "grad_norm": 0.185546875, + "learning_rate": 0.0014174590725769699, + "loss": 0.126, + "step": 37600 + }, + { + "epoch": 0.3263947361568042, + "grad_norm": 0.3046875, + "learning_rate": 0.0014174312229500972, + "loss": 0.127, + "step": 37601 + }, + { + "epoch": 0.32640341663700834, + "grad_norm": 0.59375, + "learning_rate": 0.0014174033729760862, + "loss": 0.1562, + "step": 37602 + }, + { + "epoch": 0.32641209711721253, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014173755226549681, + "loss": 0.1157, + "step": 37603 + }, + { + "epoch": 0.32642077759741667, + "grad_norm": 0.75, + "learning_rate": 0.0014173476719867727, + "loss": 0.0952, + "step": 37604 + }, + { + "epoch": 0.32642945807762086, + "grad_norm": 0.078125, + "learning_rate": 0.0014173198209715306, + "loss": 0.0791, + "step": 37605 + }, + { + "epoch": 0.326438138557825, + "grad_norm": 0.2080078125, + "learning_rate": 0.001417291969609272, + "loss": 0.1162, + "step": 37606 + }, + { + "epoch": 0.3264468190380292, + "grad_norm": 0.10009765625, + "learning_rate": 0.0014172641179000284, + "loss": 0.0962, + "step": 37607 + }, + { + "epoch": 0.3264554995182333, + "grad_norm": 0.33203125, + "learning_rate": 0.0014172362658438288, + "loss": 0.1182, + "step": 37608 + }, + { + "epoch": 0.3264641799984375, + "grad_norm": 0.16796875, + "learning_rate": 0.0014172084134407052, + "loss": 0.1104, + "step": 37609 + }, + { + "epoch": 0.32647286047864166, + "grad_norm": 0.146484375, + "learning_rate": 0.0014171805606906865, + "loss": 0.1299, + "step": 37610 + }, + { + "epoch": 0.32648154095884585, + "grad_norm": 0.169921875, + "learning_rate": 0.001417152707593804, + "loss": 0.1035, + "step": 37611 + }, + { + "epoch": 0.32649022143905, + "grad_norm": 0.357421875, + "learning_rate": 0.0014171248541500882, + "loss": 0.1289, + "step": 37612 + }, + { + "epoch": 0.3264989019192542, + "grad_norm": 0.322265625, + "learning_rate": 0.0014170970003595696, + "loss": 0.1069, + "step": 37613 + }, + { + "epoch": 0.3265075823994583, + "grad_norm": 1.53125, + "learning_rate": 0.0014170691462222782, + "loss": 0.1211, + "step": 37614 + }, + { + "epoch": 0.3265162628796625, + "grad_norm": 0.296875, + "learning_rate": 0.001417041291738245, + "loss": 0.0884, + "step": 37615 + }, + { + "epoch": 0.32652494335986665, + "grad_norm": 0.458984375, + "learning_rate": 0.0014170134369075001, + "loss": 0.1621, + "step": 37616 + }, + { + "epoch": 0.32653362384007084, + "grad_norm": 0.1767578125, + "learning_rate": 0.001416985581730074, + "loss": 0.168, + "step": 37617 + }, + { + "epoch": 0.326542304320275, + "grad_norm": 0.08984375, + "learning_rate": 0.0014169577262059972, + "loss": 0.1235, + "step": 37618 + }, + { + "epoch": 0.32655098480047917, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014169298703353005, + "loss": 0.1089, + "step": 37619 + }, + { + "epoch": 0.3265596652806833, + "grad_norm": 0.2314453125, + "learning_rate": 0.0014169020141180139, + "loss": 0.0796, + "step": 37620 + }, + { + "epoch": 0.3265683457608875, + "grad_norm": 0.4140625, + "learning_rate": 0.0014168741575541677, + "loss": 0.1016, + "step": 37621 + }, + { + "epoch": 0.32657702624109164, + "grad_norm": 0.1904296875, + "learning_rate": 0.001416846300643793, + "loss": 0.1064, + "step": 37622 + }, + { + "epoch": 0.32658570672129583, + "grad_norm": 0.5, + "learning_rate": 0.0014168184433869199, + "loss": 0.0869, + "step": 37623 + }, + { + "epoch": 0.32659438720149997, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014167905857835788, + "loss": 0.167, + "step": 37624 + }, + { + "epoch": 0.32660306768170416, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014167627278338004, + "loss": 0.1108, + "step": 37625 + }, + { + "epoch": 0.3266117481619083, + "grad_norm": 0.146484375, + "learning_rate": 0.0014167348695376153, + "loss": 0.0879, + "step": 37626 + }, + { + "epoch": 0.3266204286421125, + "grad_norm": 0.1865234375, + "learning_rate": 0.001416707010895053, + "loss": 0.1211, + "step": 37627 + }, + { + "epoch": 0.32662910912231663, + "grad_norm": 0.10693359375, + "learning_rate": 0.0014166791519061453, + "loss": 0.1104, + "step": 37628 + }, + { + "epoch": 0.3266377896025208, + "grad_norm": 0.431640625, + "learning_rate": 0.0014166512925709216, + "loss": 0.1201, + "step": 37629 + }, + { + "epoch": 0.32664647008272496, + "grad_norm": 0.2177734375, + "learning_rate": 0.001416623432889413, + "loss": 0.1084, + "step": 37630 + }, + { + "epoch": 0.32665515056292915, + "grad_norm": 0.09033203125, + "learning_rate": 0.00141659557286165, + "loss": 0.0903, + "step": 37631 + }, + { + "epoch": 0.3266638310431333, + "grad_norm": 0.0830078125, + "learning_rate": 0.0014165677124876623, + "loss": 0.0938, + "step": 37632 + }, + { + "epoch": 0.3266725115233375, + "grad_norm": 0.26171875, + "learning_rate": 0.0014165398517674815, + "loss": 0.1152, + "step": 37633 + }, + { + "epoch": 0.3266811920035416, + "grad_norm": 0.373046875, + "learning_rate": 0.0014165119907011373, + "loss": 0.1289, + "step": 37634 + }, + { + "epoch": 0.3266898724837458, + "grad_norm": 0.47265625, + "learning_rate": 0.0014164841292886603, + "loss": 0.082, + "step": 37635 + }, + { + "epoch": 0.32669855296394995, + "grad_norm": 0.123046875, + "learning_rate": 0.001416456267530081, + "loss": 0.1367, + "step": 37636 + }, + { + "epoch": 0.32670723344415414, + "grad_norm": 0.306640625, + "learning_rate": 0.00141642840542543, + "loss": 0.1045, + "step": 37637 + }, + { + "epoch": 0.3267159139243583, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014164005429747374, + "loss": 0.0981, + "step": 37638 + }, + { + "epoch": 0.3267245944045625, + "grad_norm": 0.2451171875, + "learning_rate": 0.0014163726801780338, + "loss": 0.1064, + "step": 37639 + }, + { + "epoch": 0.3267332748847666, + "grad_norm": 0.2470703125, + "learning_rate": 0.00141634481703535, + "loss": 0.0889, + "step": 37640 + }, + { + "epoch": 0.3267419553649708, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014163169535467163, + "loss": 0.1396, + "step": 37641 + }, + { + "epoch": 0.32675063584517494, + "grad_norm": 0.1474609375, + "learning_rate": 0.0014162890897121629, + "loss": 0.1504, + "step": 37642 + }, + { + "epoch": 0.32675931632537913, + "grad_norm": 0.35546875, + "learning_rate": 0.0014162612255317212, + "loss": 0.0947, + "step": 37643 + }, + { + "epoch": 0.32676799680558327, + "grad_norm": 0.1826171875, + "learning_rate": 0.00141623336100542, + "loss": 0.0996, + "step": 37644 + }, + { + "epoch": 0.32677667728578746, + "grad_norm": 0.421875, + "learning_rate": 0.0014162054961332916, + "loss": 0.1133, + "step": 37645 + }, + { + "epoch": 0.3267853577659916, + "grad_norm": 0.1845703125, + "learning_rate": 0.001416177630915365, + "loss": 0.0903, + "step": 37646 + }, + { + "epoch": 0.3267940382461958, + "grad_norm": 0.185546875, + "learning_rate": 0.0014161497653516718, + "loss": 0.0923, + "step": 37647 + }, + { + "epoch": 0.32680271872639993, + "grad_norm": 0.224609375, + "learning_rate": 0.0014161218994422416, + "loss": 0.1152, + "step": 37648 + }, + { + "epoch": 0.3268113992066041, + "grad_norm": 0.138671875, + "learning_rate": 0.001416094033187105, + "loss": 0.0864, + "step": 37649 + }, + { + "epoch": 0.32682007968680826, + "grad_norm": 0.466796875, + "learning_rate": 0.0014160661665862935, + "loss": 0.0776, + "step": 37650 + }, + { + "epoch": 0.32682876016701246, + "grad_norm": 2.0, + "learning_rate": 0.0014160382996398363, + "loss": 0.3066, + "step": 37651 + }, + { + "epoch": 0.3268374406472166, + "grad_norm": 0.34375, + "learning_rate": 0.0014160104323477645, + "loss": 0.1084, + "step": 37652 + }, + { + "epoch": 0.3268461211274208, + "grad_norm": 0.2353515625, + "learning_rate": 0.0014159825647101082, + "loss": 0.083, + "step": 37653 + }, + { + "epoch": 0.3268548016076249, + "grad_norm": 0.1845703125, + "learning_rate": 0.0014159546967268983, + "loss": 0.0884, + "step": 37654 + }, + { + "epoch": 0.3268634820878291, + "grad_norm": 0.076171875, + "learning_rate": 0.001415926828398165, + "loss": 0.0933, + "step": 37655 + }, + { + "epoch": 0.32687216256803325, + "grad_norm": 0.09814453125, + "learning_rate": 0.001415898959723939, + "loss": 0.1055, + "step": 37656 + }, + { + "epoch": 0.32688084304823745, + "grad_norm": 0.318359375, + "learning_rate": 0.0014158710907042505, + "loss": 0.1079, + "step": 37657 + }, + { + "epoch": 0.3268895235284416, + "grad_norm": 0.3828125, + "learning_rate": 0.0014158432213391304, + "loss": 0.0972, + "step": 37658 + }, + { + "epoch": 0.3268982040086458, + "grad_norm": 0.1904296875, + "learning_rate": 0.0014158153516286086, + "loss": 0.0879, + "step": 37659 + }, + { + "epoch": 0.3269068844888499, + "grad_norm": 0.11669921875, + "learning_rate": 0.001415787481572716, + "loss": 0.0947, + "step": 37660 + }, + { + "epoch": 0.3269155649690541, + "grad_norm": 0.158203125, + "learning_rate": 0.0014157596111714827, + "loss": 0.1074, + "step": 37661 + }, + { + "epoch": 0.32692424544925824, + "grad_norm": 0.22265625, + "learning_rate": 0.0014157317404249398, + "loss": 0.1318, + "step": 37662 + }, + { + "epoch": 0.32693292592946244, + "grad_norm": 0.2431640625, + "learning_rate": 0.0014157038693331174, + "loss": 0.1025, + "step": 37663 + }, + { + "epoch": 0.3269416064096666, + "grad_norm": 0.1953125, + "learning_rate": 0.0014156759978960459, + "loss": 0.1094, + "step": 37664 + }, + { + "epoch": 0.32695028688987077, + "grad_norm": 0.119140625, + "learning_rate": 0.0014156481261137558, + "loss": 0.0923, + "step": 37665 + }, + { + "epoch": 0.3269589673700749, + "grad_norm": 0.80078125, + "learning_rate": 0.0014156202539862776, + "loss": 0.2129, + "step": 37666 + }, + { + "epoch": 0.3269676478502791, + "grad_norm": 0.189453125, + "learning_rate": 0.001415592381513642, + "loss": 0.1045, + "step": 37667 + }, + { + "epoch": 0.32697632833048323, + "grad_norm": 0.1103515625, + "learning_rate": 0.0014155645086958793, + "loss": 0.0811, + "step": 37668 + }, + { + "epoch": 0.3269850088106874, + "grad_norm": 0.0966796875, + "learning_rate": 0.0014155366355330199, + "loss": 0.085, + "step": 37669 + }, + { + "epoch": 0.32699368929089156, + "grad_norm": 0.1484375, + "learning_rate": 0.001415508762025094, + "loss": 0.0864, + "step": 37670 + }, + { + "epoch": 0.32700236977109576, + "grad_norm": 0.10107421875, + "learning_rate": 0.001415480888172133, + "loss": 0.0874, + "step": 37671 + }, + { + "epoch": 0.3270110502512999, + "grad_norm": 0.66796875, + "learning_rate": 0.0014154530139741668, + "loss": 0.0938, + "step": 37672 + }, + { + "epoch": 0.3270197307315041, + "grad_norm": 0.69921875, + "learning_rate": 0.0014154251394312258, + "loss": 0.1055, + "step": 37673 + }, + { + "epoch": 0.3270284112117082, + "grad_norm": 0.3203125, + "learning_rate": 0.0014153972645433409, + "loss": 0.0869, + "step": 37674 + }, + { + "epoch": 0.3270370916919124, + "grad_norm": 0.328125, + "learning_rate": 0.001415369389310542, + "loss": 0.1123, + "step": 37675 + }, + { + "epoch": 0.32704577217211656, + "grad_norm": 0.482421875, + "learning_rate": 0.0014153415137328597, + "loss": 0.0986, + "step": 37676 + }, + { + "epoch": 0.32705445265232075, + "grad_norm": 0.1962890625, + "learning_rate": 0.0014153136378103248, + "loss": 0.0781, + "step": 37677 + }, + { + "epoch": 0.3270631331325249, + "grad_norm": 0.08056640625, + "learning_rate": 0.0014152857615429679, + "loss": 0.085, + "step": 37678 + }, + { + "epoch": 0.3270718136127291, + "grad_norm": 0.73046875, + "learning_rate": 0.0014152578849308188, + "loss": 0.1123, + "step": 37679 + }, + { + "epoch": 0.3270804940929332, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014152300079739086, + "loss": 0.0742, + "step": 37680 + }, + { + "epoch": 0.3270891745731374, + "grad_norm": 0.408203125, + "learning_rate": 0.0014152021306722679, + "loss": 0.1533, + "step": 37681 + }, + { + "epoch": 0.32709785505334155, + "grad_norm": 0.890625, + "learning_rate": 0.0014151742530259265, + "loss": 0.0894, + "step": 37682 + }, + { + "epoch": 0.32710653553354574, + "grad_norm": 0.1171875, + "learning_rate": 0.0014151463750349154, + "loss": 0.0825, + "step": 37683 + }, + { + "epoch": 0.3271152160137499, + "grad_norm": 0.515625, + "learning_rate": 0.001415118496699265, + "loss": 0.1016, + "step": 37684 + }, + { + "epoch": 0.32712389649395407, + "grad_norm": 0.13671875, + "learning_rate": 0.001415090618019006, + "loss": 0.1104, + "step": 37685 + }, + { + "epoch": 0.3271325769741582, + "grad_norm": 0.28515625, + "learning_rate": 0.001415062738994168, + "loss": 0.0869, + "step": 37686 + }, + { + "epoch": 0.3271412574543624, + "grad_norm": 0.24609375, + "learning_rate": 0.0014150348596247825, + "loss": 0.1582, + "step": 37687 + }, + { + "epoch": 0.32714993793456654, + "grad_norm": 0.3359375, + "learning_rate": 0.0014150069799108793, + "loss": 0.1123, + "step": 37688 + }, + { + "epoch": 0.32715861841477073, + "grad_norm": 0.53125, + "learning_rate": 0.0014149790998524898, + "loss": 0.0781, + "step": 37689 + }, + { + "epoch": 0.32716729889497487, + "grad_norm": 0.486328125, + "learning_rate": 0.0014149512194496436, + "loss": 0.085, + "step": 37690 + }, + { + "epoch": 0.32717597937517906, + "grad_norm": 0.1796875, + "learning_rate": 0.0014149233387023715, + "loss": 0.1123, + "step": 37691 + }, + { + "epoch": 0.3271846598553832, + "grad_norm": 0.328125, + "learning_rate": 0.0014148954576107039, + "loss": 0.0986, + "step": 37692 + }, + { + "epoch": 0.3271933403355874, + "grad_norm": 0.66796875, + "learning_rate": 0.0014148675761746711, + "loss": 0.0889, + "step": 37693 + }, + { + "epoch": 0.3272020208157915, + "grad_norm": 0.412109375, + "learning_rate": 0.0014148396943943042, + "loss": 0.0796, + "step": 37694 + }, + { + "epoch": 0.3272107012959957, + "grad_norm": 0.11181640625, + "learning_rate": 0.0014148118122696337, + "loss": 0.1079, + "step": 37695 + }, + { + "epoch": 0.32721938177619986, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014147839298006889, + "loss": 0.0942, + "step": 37696 + }, + { + "epoch": 0.32722806225640405, + "grad_norm": 0.1064453125, + "learning_rate": 0.0014147560469875014, + "loss": 0.0918, + "step": 37697 + }, + { + "epoch": 0.3272367427366082, + "grad_norm": 0.283203125, + "learning_rate": 0.0014147281638301015, + "loss": 0.0986, + "step": 37698 + }, + { + "epoch": 0.3272454232168124, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014147002803285198, + "loss": 0.1064, + "step": 37699 + }, + { + "epoch": 0.3272541036970165, + "grad_norm": 0.32421875, + "learning_rate": 0.0014146723964827863, + "loss": 0.124, + "step": 37700 + }, + { + "epoch": 0.3272627841772207, + "grad_norm": 0.08056640625, + "learning_rate": 0.001414644512292932, + "loss": 0.1006, + "step": 37701 + }, + { + "epoch": 0.32727146465742485, + "grad_norm": 0.1982421875, + "learning_rate": 0.0014146166277589872, + "loss": 0.1055, + "step": 37702 + }, + { + "epoch": 0.327280145137629, + "grad_norm": 0.353515625, + "learning_rate": 0.0014145887428809825, + "loss": 0.0786, + "step": 37703 + }, + { + "epoch": 0.3272888256178332, + "grad_norm": 0.06494140625, + "learning_rate": 0.0014145608576589482, + "loss": 0.0879, + "step": 37704 + }, + { + "epoch": 0.3272975060980373, + "grad_norm": 0.353515625, + "learning_rate": 0.0014145329720929142, + "loss": 0.084, + "step": 37705 + }, + { + "epoch": 0.3273061865782415, + "grad_norm": 0.1005859375, + "learning_rate": 0.0014145050861829127, + "loss": 0.1006, + "step": 37706 + }, + { + "epoch": 0.32731486705844565, + "grad_norm": 0.12451171875, + "learning_rate": 0.0014144771999289726, + "loss": 0.1094, + "step": 37707 + }, + { + "epoch": 0.32732354753864984, + "grad_norm": 0.076171875, + "learning_rate": 0.0014144493133311248, + "loss": 0.0801, + "step": 37708 + }, + { + "epoch": 0.327332228018854, + "grad_norm": 0.154296875, + "learning_rate": 0.0014144214263894003, + "loss": 0.1099, + "step": 37709 + }, + { + "epoch": 0.32734090849905817, + "grad_norm": 0.17578125, + "learning_rate": 0.0014143935391038294, + "loss": 0.0986, + "step": 37710 + }, + { + "epoch": 0.3273495889792623, + "grad_norm": 0.40234375, + "learning_rate": 0.0014143656514744418, + "loss": 0.1621, + "step": 37711 + }, + { + "epoch": 0.3273582694594665, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014143377635012693, + "loss": 0.1113, + "step": 37712 + }, + { + "epoch": 0.32736694993967064, + "grad_norm": 0.099609375, + "learning_rate": 0.0014143098751843413, + "loss": 0.1011, + "step": 37713 + }, + { + "epoch": 0.32737563041987483, + "grad_norm": 0.251953125, + "learning_rate": 0.0014142819865236892, + "loss": 0.0845, + "step": 37714 + }, + { + "epoch": 0.32738431090007897, + "grad_norm": 0.5, + "learning_rate": 0.0014142540975193428, + "loss": 0.1299, + "step": 37715 + }, + { + "epoch": 0.32739299138028316, + "grad_norm": 0.1279296875, + "learning_rate": 0.0014142262081713328, + "loss": 0.1074, + "step": 37716 + }, + { + "epoch": 0.3274016718604873, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014141983184796896, + "loss": 0.1279, + "step": 37717 + }, + { + "epoch": 0.3274103523406915, + "grad_norm": 0.255859375, + "learning_rate": 0.001414170428444444, + "loss": 0.082, + "step": 37718 + }, + { + "epoch": 0.32741903282089563, + "grad_norm": 0.27734375, + "learning_rate": 0.0014141425380656263, + "loss": 0.1074, + "step": 37719 + }, + { + "epoch": 0.3274277133010998, + "grad_norm": 0.21484375, + "learning_rate": 0.0014141146473432673, + "loss": 0.1318, + "step": 37720 + }, + { + "epoch": 0.32743639378130396, + "grad_norm": 0.7734375, + "learning_rate": 0.001414086756277397, + "loss": 0.1699, + "step": 37721 + }, + { + "epoch": 0.32744507426150815, + "grad_norm": 0.08447265625, + "learning_rate": 0.001414058864868046, + "loss": 0.083, + "step": 37722 + }, + { + "epoch": 0.3274537547417123, + "grad_norm": 0.41796875, + "learning_rate": 0.0014140309731152454, + "loss": 0.1133, + "step": 37723 + }, + { + "epoch": 0.3274624352219165, + "grad_norm": 0.69921875, + "learning_rate": 0.0014140030810190247, + "loss": 0.106, + "step": 37724 + }, + { + "epoch": 0.3274711157021206, + "grad_norm": 0.07080078125, + "learning_rate": 0.0014139751885794151, + "loss": 0.105, + "step": 37725 + }, + { + "epoch": 0.3274797961823248, + "grad_norm": 0.0869140625, + "learning_rate": 0.0014139472957964475, + "loss": 0.0918, + "step": 37726 + }, + { + "epoch": 0.32748847666252895, + "grad_norm": 0.29296875, + "learning_rate": 0.001413919402670151, + "loss": 0.1348, + "step": 37727 + }, + { + "epoch": 0.32749715714273314, + "grad_norm": 0.1875, + "learning_rate": 0.0014138915092005576, + "loss": 0.0742, + "step": 37728 + }, + { + "epoch": 0.3275058376229373, + "grad_norm": 0.1044921875, + "learning_rate": 0.001413863615387697, + "loss": 0.1016, + "step": 37729 + }, + { + "epoch": 0.32751451810314147, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014138357212315998, + "loss": 0.0815, + "step": 37730 + }, + { + "epoch": 0.3275231985833456, + "grad_norm": 0.455078125, + "learning_rate": 0.0014138078267322963, + "loss": 0.0918, + "step": 37731 + }, + { + "epoch": 0.3275318790635498, + "grad_norm": 0.302734375, + "learning_rate": 0.0014137799318898177, + "loss": 0.0903, + "step": 37732 + }, + { + "epoch": 0.32754055954375394, + "grad_norm": 0.11376953125, + "learning_rate": 0.001413752036704194, + "loss": 0.0986, + "step": 37733 + }, + { + "epoch": 0.32754924002395813, + "grad_norm": 0.2060546875, + "learning_rate": 0.0014137241411754555, + "loss": 0.1406, + "step": 37734 + }, + { + "epoch": 0.32755792050416227, + "grad_norm": 0.271484375, + "learning_rate": 0.0014136962453036332, + "loss": 0.1108, + "step": 37735 + }, + { + "epoch": 0.32756660098436646, + "grad_norm": 0.10498046875, + "learning_rate": 0.0014136683490887573, + "loss": 0.0781, + "step": 37736 + }, + { + "epoch": 0.3275752814645706, + "grad_norm": 0.1181640625, + "learning_rate": 0.0014136404525308588, + "loss": 0.1523, + "step": 37737 + }, + { + "epoch": 0.3275839619447748, + "grad_norm": 0.11669921875, + "learning_rate": 0.0014136125556299675, + "loss": 0.0845, + "step": 37738 + }, + { + "epoch": 0.32759264242497893, + "grad_norm": 0.234375, + "learning_rate": 0.0014135846583861141, + "loss": 0.1143, + "step": 37739 + }, + { + "epoch": 0.3276013229051831, + "grad_norm": 0.1943359375, + "learning_rate": 0.0014135567607993294, + "loss": 0.0757, + "step": 37740 + }, + { + "epoch": 0.32761000338538726, + "grad_norm": 0.318359375, + "learning_rate": 0.0014135288628696438, + "loss": 0.1191, + "step": 37741 + }, + { + "epoch": 0.32761868386559145, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014135009645970878, + "loss": 0.0845, + "step": 37742 + }, + { + "epoch": 0.3276273643457956, + "grad_norm": 0.2099609375, + "learning_rate": 0.0014134730659816916, + "loss": 0.1543, + "step": 37743 + }, + { + "epoch": 0.3276360448259998, + "grad_norm": 0.1611328125, + "learning_rate": 0.0014134451670234863, + "loss": 0.0562, + "step": 37744 + }, + { + "epoch": 0.3276447253062039, + "grad_norm": 1.25, + "learning_rate": 0.0014134172677225014, + "loss": 0.1226, + "step": 37745 + }, + { + "epoch": 0.3276534057864081, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014133893680787686, + "loss": 0.0977, + "step": 37746 + }, + { + "epoch": 0.32766208626661225, + "grad_norm": 0.3671875, + "learning_rate": 0.0014133614680923179, + "loss": 0.0981, + "step": 37747 + }, + { + "epoch": 0.32767076674681644, + "grad_norm": 0.11279296875, + "learning_rate": 0.0014133335677631795, + "loss": 0.106, + "step": 37748 + }, + { + "epoch": 0.3276794472270206, + "grad_norm": 0.2001953125, + "learning_rate": 0.0014133056670913846, + "loss": 0.0762, + "step": 37749 + }, + { + "epoch": 0.3276881277072248, + "grad_norm": 0.232421875, + "learning_rate": 0.0014132777660769633, + "loss": 0.1128, + "step": 37750 + }, + { + "epoch": 0.3276968081874289, + "grad_norm": 0.5859375, + "learning_rate": 0.0014132498647199461, + "loss": 0.1118, + "step": 37751 + }, + { + "epoch": 0.3277054886676331, + "grad_norm": 0.1767578125, + "learning_rate": 0.0014132219630203636, + "loss": 0.0889, + "step": 37752 + }, + { + "epoch": 0.32771416914783724, + "grad_norm": 0.41015625, + "learning_rate": 0.0014131940609782459, + "loss": 0.1162, + "step": 37753 + }, + { + "epoch": 0.32772284962804143, + "grad_norm": 0.240234375, + "learning_rate": 0.0014131661585936241, + "loss": 0.105, + "step": 37754 + }, + { + "epoch": 0.3277315301082456, + "grad_norm": 0.490234375, + "learning_rate": 0.0014131382558665287, + "loss": 0.0977, + "step": 37755 + }, + { + "epoch": 0.32774021058844977, + "grad_norm": 0.1630859375, + "learning_rate": 0.0014131103527969897, + "loss": 0.126, + "step": 37756 + }, + { + "epoch": 0.3277488910686539, + "grad_norm": 0.2353515625, + "learning_rate": 0.0014130824493850378, + "loss": 0.1162, + "step": 37757 + }, + { + "epoch": 0.3277575715488581, + "grad_norm": 0.08740234375, + "learning_rate": 0.0014130545456307042, + "loss": 0.1387, + "step": 37758 + }, + { + "epoch": 0.32776625202906223, + "grad_norm": 0.236328125, + "learning_rate": 0.0014130266415340188, + "loss": 0.0952, + "step": 37759 + }, + { + "epoch": 0.3277749325092664, + "grad_norm": 0.40234375, + "learning_rate": 0.0014129987370950117, + "loss": 0.0835, + "step": 37760 + }, + { + "epoch": 0.32778361298947056, + "grad_norm": 0.8671875, + "learning_rate": 0.001412970832313714, + "loss": 0.1357, + "step": 37761 + }, + { + "epoch": 0.32779229346967476, + "grad_norm": 0.2470703125, + "learning_rate": 0.0014129429271901562, + "loss": 0.1045, + "step": 37762 + }, + { + "epoch": 0.3278009739498789, + "grad_norm": 0.421875, + "learning_rate": 0.0014129150217243689, + "loss": 0.0825, + "step": 37763 + }, + { + "epoch": 0.3278096544300831, + "grad_norm": 0.294921875, + "learning_rate": 0.001412887115916382, + "loss": 0.125, + "step": 37764 + }, + { + "epoch": 0.3278183349102872, + "grad_norm": 0.24609375, + "learning_rate": 0.0014128592097662268, + "loss": 0.0874, + "step": 37765 + }, + { + "epoch": 0.3278270153904914, + "grad_norm": 0.228515625, + "learning_rate": 0.0014128313032739333, + "loss": 0.0757, + "step": 37766 + }, + { + "epoch": 0.32783569587069555, + "grad_norm": 0.265625, + "learning_rate": 0.0014128033964395325, + "loss": 0.1367, + "step": 37767 + }, + { + "epoch": 0.32784437635089975, + "grad_norm": 0.2314453125, + "learning_rate": 0.0014127754892630541, + "loss": 0.0996, + "step": 37768 + }, + { + "epoch": 0.3278530568311039, + "grad_norm": 0.408203125, + "learning_rate": 0.0014127475817445295, + "loss": 0.0747, + "step": 37769 + }, + { + "epoch": 0.3278617373113081, + "grad_norm": 0.451171875, + "learning_rate": 0.0014127196738839888, + "loss": 0.1191, + "step": 37770 + }, + { + "epoch": 0.3278704177915122, + "grad_norm": 0.10498046875, + "learning_rate": 0.0014126917656814625, + "loss": 0.1055, + "step": 37771 + }, + { + "epoch": 0.3278790982717164, + "grad_norm": 0.25390625, + "learning_rate": 0.0014126638571369812, + "loss": 0.1055, + "step": 37772 + }, + { + "epoch": 0.32788777875192054, + "grad_norm": 0.11572265625, + "learning_rate": 0.0014126359482505753, + "loss": 0.1436, + "step": 37773 + }, + { + "epoch": 0.32789645923212474, + "grad_norm": 0.2412109375, + "learning_rate": 0.0014126080390222755, + "loss": 0.1143, + "step": 37774 + }, + { + "epoch": 0.3279051397123289, + "grad_norm": 0.1748046875, + "learning_rate": 0.001412580129452112, + "loss": 0.1113, + "step": 37775 + }, + { + "epoch": 0.32791382019253307, + "grad_norm": 0.1953125, + "learning_rate": 0.0014125522195401162, + "loss": 0.1387, + "step": 37776 + }, + { + "epoch": 0.3279225006727372, + "grad_norm": 0.404296875, + "learning_rate": 0.0014125243092863175, + "loss": 0.0918, + "step": 37777 + }, + { + "epoch": 0.3279311811529414, + "grad_norm": 0.322265625, + "learning_rate": 0.001412496398690747, + "loss": 0.0996, + "step": 37778 + }, + { + "epoch": 0.32793986163314554, + "grad_norm": 0.36328125, + "learning_rate": 0.0014124684877534351, + "loss": 0.083, + "step": 37779 + }, + { + "epoch": 0.32794854211334973, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014124405764744125, + "loss": 0.0854, + "step": 37780 + }, + { + "epoch": 0.32795722259355387, + "grad_norm": 0.18359375, + "learning_rate": 0.0014124126648537097, + "loss": 0.083, + "step": 37781 + }, + { + "epoch": 0.32796590307375806, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014123847528913567, + "loss": 0.0889, + "step": 37782 + }, + { + "epoch": 0.3279745835539622, + "grad_norm": 0.1806640625, + "learning_rate": 0.0014123568405873847, + "loss": 0.0864, + "step": 37783 + }, + { + "epoch": 0.3279832640341664, + "grad_norm": 0.2470703125, + "learning_rate": 0.0014123289279418237, + "loss": 0.1289, + "step": 37784 + }, + { + "epoch": 0.3279919445143705, + "grad_norm": 0.322265625, + "learning_rate": 0.0014123010149547047, + "loss": 0.126, + "step": 37785 + }, + { + "epoch": 0.3280006249945747, + "grad_norm": 0.38671875, + "learning_rate": 0.001412273101626058, + "loss": 0.1494, + "step": 37786 + }, + { + "epoch": 0.32800930547477886, + "grad_norm": 0.373046875, + "learning_rate": 0.001412245187955914, + "loss": 0.0811, + "step": 37787 + }, + { + "epoch": 0.32801798595498305, + "grad_norm": 0.181640625, + "learning_rate": 0.0014122172739443036, + "loss": 0.0698, + "step": 37788 + }, + { + "epoch": 0.3280266664351872, + "grad_norm": 0.376953125, + "learning_rate": 0.0014121893595912568, + "loss": 0.1157, + "step": 37789 + }, + { + "epoch": 0.3280353469153914, + "grad_norm": 0.06591796875, + "learning_rate": 0.0014121614448968046, + "loss": 0.0801, + "step": 37790 + }, + { + "epoch": 0.3280440273955955, + "grad_norm": 0.0986328125, + "learning_rate": 0.001412133529860977, + "loss": 0.0845, + "step": 37791 + }, + { + "epoch": 0.3280527078757997, + "grad_norm": 0.1806640625, + "learning_rate": 0.0014121056144838053, + "loss": 0.0903, + "step": 37792 + }, + { + "epoch": 0.32806138835600385, + "grad_norm": 0.390625, + "learning_rate": 0.0014120776987653193, + "loss": 0.125, + "step": 37793 + }, + { + "epoch": 0.32807006883620804, + "grad_norm": 0.080078125, + "learning_rate": 0.0014120497827055498, + "loss": 0.0859, + "step": 37794 + }, + { + "epoch": 0.3280787493164122, + "grad_norm": 0.515625, + "learning_rate": 0.0014120218663045275, + "loss": 0.1475, + "step": 37795 + }, + { + "epoch": 0.32808742979661637, + "grad_norm": 0.306640625, + "learning_rate": 0.0014119939495622828, + "loss": 0.0928, + "step": 37796 + }, + { + "epoch": 0.3280961102768205, + "grad_norm": 0.13671875, + "learning_rate": 0.0014119660324788462, + "loss": 0.1182, + "step": 37797 + }, + { + "epoch": 0.3281047907570247, + "grad_norm": 0.1142578125, + "learning_rate": 0.001411938115054248, + "loss": 0.1025, + "step": 37798 + }, + { + "epoch": 0.32811347123722884, + "grad_norm": 0.259765625, + "learning_rate": 0.0014119101972885192, + "loss": 0.0742, + "step": 37799 + }, + { + "epoch": 0.32812215171743303, + "grad_norm": 0.3671875, + "learning_rate": 0.0014118822791816903, + "loss": 0.0854, + "step": 37800 + }, + { + "epoch": 0.32813083219763717, + "grad_norm": 0.55859375, + "learning_rate": 0.0014118543607337913, + "loss": 0.0977, + "step": 37801 + }, + { + "epoch": 0.32813951267784136, + "grad_norm": 0.1484375, + "learning_rate": 0.001411826441944853, + "loss": 0.0928, + "step": 37802 + }, + { + "epoch": 0.3281481931580455, + "grad_norm": 0.08056640625, + "learning_rate": 0.001411798522814906, + "loss": 0.0967, + "step": 37803 + }, + { + "epoch": 0.3281568736382497, + "grad_norm": 0.388671875, + "learning_rate": 0.001411770603343981, + "loss": 0.1187, + "step": 37804 + }, + { + "epoch": 0.32816555411845383, + "grad_norm": 0.435546875, + "learning_rate": 0.0014117426835321084, + "loss": 0.0767, + "step": 37805 + }, + { + "epoch": 0.328174234598658, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014117147633793186, + "loss": 0.127, + "step": 37806 + }, + { + "epoch": 0.32818291507886216, + "grad_norm": 0.703125, + "learning_rate": 0.001411686842885642, + "loss": 0.1318, + "step": 37807 + }, + { + "epoch": 0.32819159555906635, + "grad_norm": 0.23046875, + "learning_rate": 0.0014116589220511096, + "loss": 0.1543, + "step": 37808 + }, + { + "epoch": 0.3282002760392705, + "grad_norm": 0.162109375, + "learning_rate": 0.0014116310008757515, + "loss": 0.0947, + "step": 37809 + }, + { + "epoch": 0.3282089565194747, + "grad_norm": 0.07373046875, + "learning_rate": 0.0014116030793595987, + "loss": 0.1025, + "step": 37810 + }, + { + "epoch": 0.3282176369996788, + "grad_norm": 0.7578125, + "learning_rate": 0.0014115751575026815, + "loss": 0.1279, + "step": 37811 + }, + { + "epoch": 0.328226317479883, + "grad_norm": 1.7109375, + "learning_rate": 0.0014115472353050298, + "loss": 0.1162, + "step": 37812 + }, + { + "epoch": 0.32823499796008715, + "grad_norm": 0.80078125, + "learning_rate": 0.0014115193127666752, + "loss": 0.1182, + "step": 37813 + }, + { + "epoch": 0.32824367844029134, + "grad_norm": 0.79296875, + "learning_rate": 0.0014114913898876475, + "loss": 0.1367, + "step": 37814 + }, + { + "epoch": 0.3282523589204955, + "grad_norm": 0.52734375, + "learning_rate": 0.001411463466667978, + "loss": 0.1406, + "step": 37815 + }, + { + "epoch": 0.3282610394006997, + "grad_norm": 0.11865234375, + "learning_rate": 0.0014114355431076963, + "loss": 0.1367, + "step": 37816 + }, + { + "epoch": 0.3282697198809038, + "grad_norm": 0.474609375, + "learning_rate": 0.0014114076192068335, + "loss": 0.1201, + "step": 37817 + }, + { + "epoch": 0.328278400361108, + "grad_norm": 0.294921875, + "learning_rate": 0.00141137969496542, + "loss": 0.1172, + "step": 37818 + }, + { + "epoch": 0.32828708084131214, + "grad_norm": 0.2294921875, + "learning_rate": 0.0014113517703834863, + "loss": 0.1055, + "step": 37819 + }, + { + "epoch": 0.32829576132151633, + "grad_norm": 0.212890625, + "learning_rate": 0.001411323845461063, + "loss": 0.1289, + "step": 37820 + }, + { + "epoch": 0.32830444180172047, + "grad_norm": 0.41015625, + "learning_rate": 0.0014112959201981808, + "loss": 0.0859, + "step": 37821 + }, + { + "epoch": 0.32831312228192466, + "grad_norm": 0.421875, + "learning_rate": 0.0014112679945948697, + "loss": 0.1167, + "step": 37822 + }, + { + "epoch": 0.3283218027621288, + "grad_norm": 0.09228515625, + "learning_rate": 0.0014112400686511607, + "loss": 0.0938, + "step": 37823 + }, + { + "epoch": 0.328330483242333, + "grad_norm": 0.341796875, + "learning_rate": 0.0014112121423670843, + "loss": 0.0908, + "step": 37824 + }, + { + "epoch": 0.32833916372253713, + "grad_norm": 0.125, + "learning_rate": 0.0014111842157426713, + "loss": 0.1074, + "step": 37825 + }, + { + "epoch": 0.32834784420274127, + "grad_norm": 0.3125, + "learning_rate": 0.0014111562887779512, + "loss": 0.1455, + "step": 37826 + }, + { + "epoch": 0.32835652468294546, + "grad_norm": 0.072265625, + "learning_rate": 0.001411128361472956, + "loss": 0.0757, + "step": 37827 + }, + { + "epoch": 0.3283652051631496, + "grad_norm": 0.1865234375, + "learning_rate": 0.001411100433827715, + "loss": 0.1128, + "step": 37828 + }, + { + "epoch": 0.3283738856433538, + "grad_norm": 0.177734375, + "learning_rate": 0.0014110725058422595, + "loss": 0.0986, + "step": 37829 + }, + { + "epoch": 0.32838256612355793, + "grad_norm": 0.455078125, + "learning_rate": 0.0014110445775166196, + "loss": 0.0986, + "step": 37830 + }, + { + "epoch": 0.3283912466037621, + "grad_norm": 0.23046875, + "learning_rate": 0.0014110166488508263, + "loss": 0.1211, + "step": 37831 + }, + { + "epoch": 0.32839992708396626, + "grad_norm": 0.0771484375, + "learning_rate": 0.0014109887198449093, + "loss": 0.0884, + "step": 37832 + }, + { + "epoch": 0.32840860756417045, + "grad_norm": 0.2490234375, + "learning_rate": 0.0014109607904989004, + "loss": 0.1133, + "step": 37833 + }, + { + "epoch": 0.3284172880443746, + "grad_norm": 0.10009765625, + "learning_rate": 0.001410932860812829, + "loss": 0.0918, + "step": 37834 + }, + { + "epoch": 0.3284259685245788, + "grad_norm": 0.2275390625, + "learning_rate": 0.001410904930786726, + "loss": 0.0977, + "step": 37835 + }, + { + "epoch": 0.3284346490047829, + "grad_norm": 0.341796875, + "learning_rate": 0.0014108770004206226, + "loss": 0.0767, + "step": 37836 + }, + { + "epoch": 0.3284433294849871, + "grad_norm": 0.154296875, + "learning_rate": 0.0014108490697145484, + "loss": 0.1016, + "step": 37837 + }, + { + "epoch": 0.32845200996519125, + "grad_norm": 0.5390625, + "learning_rate": 0.0014108211386685346, + "loss": 0.0918, + "step": 37838 + }, + { + "epoch": 0.32846069044539544, + "grad_norm": 0.373046875, + "learning_rate": 0.001410793207282611, + "loss": 0.0742, + "step": 37839 + }, + { + "epoch": 0.3284693709255996, + "grad_norm": 0.267578125, + "learning_rate": 0.001410765275556809, + "loss": 0.0762, + "step": 37840 + }, + { + "epoch": 0.3284780514058038, + "grad_norm": 0.0986328125, + "learning_rate": 0.0014107373434911585, + "loss": 0.1108, + "step": 37841 + }, + { + "epoch": 0.3284867318860079, + "grad_norm": 0.52734375, + "learning_rate": 0.0014107094110856905, + "loss": 0.1445, + "step": 37842 + }, + { + "epoch": 0.3284954123662121, + "grad_norm": 0.240234375, + "learning_rate": 0.0014106814783404352, + "loss": 0.0933, + "step": 37843 + }, + { + "epoch": 0.32850409284641624, + "grad_norm": 0.423828125, + "learning_rate": 0.0014106535452554235, + "loss": 0.1074, + "step": 37844 + }, + { + "epoch": 0.32851277332662043, + "grad_norm": 0.291015625, + "learning_rate": 0.0014106256118306857, + "loss": 0.1211, + "step": 37845 + }, + { + "epoch": 0.32852145380682457, + "grad_norm": 0.08837890625, + "learning_rate": 0.0014105976780662524, + "loss": 0.0845, + "step": 37846 + }, + { + "epoch": 0.32853013428702876, + "grad_norm": 0.369140625, + "learning_rate": 0.0014105697439621538, + "loss": 0.084, + "step": 37847 + }, + { + "epoch": 0.3285388147672329, + "grad_norm": 0.6171875, + "learning_rate": 0.0014105418095184212, + "loss": 0.0845, + "step": 37848 + }, + { + "epoch": 0.3285474952474371, + "grad_norm": 3.46875, + "learning_rate": 0.0014105138747350847, + "loss": 0.3594, + "step": 37849 + }, + { + "epoch": 0.32855617572764123, + "grad_norm": 0.1376953125, + "learning_rate": 0.001410485939612175, + "loss": 0.1084, + "step": 37850 + }, + { + "epoch": 0.3285648562078454, + "grad_norm": 0.46484375, + "learning_rate": 0.0014104580041497223, + "loss": 0.1152, + "step": 37851 + }, + { + "epoch": 0.32857353668804956, + "grad_norm": 0.90234375, + "learning_rate": 0.0014104300683477574, + "loss": 0.1162, + "step": 37852 + }, + { + "epoch": 0.32858221716825375, + "grad_norm": 0.4765625, + "learning_rate": 0.0014104021322063108, + "loss": 0.1367, + "step": 37853 + }, + { + "epoch": 0.3285908976484579, + "grad_norm": 0.099609375, + "learning_rate": 0.0014103741957254134, + "loss": 0.0757, + "step": 37854 + }, + { + "epoch": 0.3285995781286621, + "grad_norm": 0.265625, + "learning_rate": 0.0014103462589050953, + "loss": 0.1045, + "step": 37855 + }, + { + "epoch": 0.3286082586088662, + "grad_norm": 0.1611328125, + "learning_rate": 0.001410318321745387, + "loss": 0.1074, + "step": 37856 + }, + { + "epoch": 0.3286169390890704, + "grad_norm": 0.20703125, + "learning_rate": 0.0014102903842463192, + "loss": 0.1045, + "step": 37857 + }, + { + "epoch": 0.32862561956927455, + "grad_norm": 0.1455078125, + "learning_rate": 0.0014102624464079228, + "loss": 0.0923, + "step": 37858 + }, + { + "epoch": 0.32863430004947874, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014102345082302279, + "loss": 0.1104, + "step": 37859 + }, + { + "epoch": 0.3286429805296829, + "grad_norm": 0.828125, + "learning_rate": 0.0014102065697132653, + "loss": 0.083, + "step": 37860 + }, + { + "epoch": 0.3286516610098871, + "grad_norm": 0.55859375, + "learning_rate": 0.0014101786308570652, + "loss": 0.0845, + "step": 37861 + }, + { + "epoch": 0.3286603414900912, + "grad_norm": 0.2158203125, + "learning_rate": 0.0014101506916616587, + "loss": 0.1123, + "step": 37862 + }, + { + "epoch": 0.3286690219702954, + "grad_norm": 0.09765625, + "learning_rate": 0.001410122752127076, + "loss": 0.1089, + "step": 37863 + }, + { + "epoch": 0.32867770245049954, + "grad_norm": 0.330078125, + "learning_rate": 0.0014100948122533477, + "loss": 0.1162, + "step": 37864 + }, + { + "epoch": 0.32868638293070374, + "grad_norm": 0.1455078125, + "learning_rate": 0.0014100668720405042, + "loss": 0.0938, + "step": 37865 + }, + { + "epoch": 0.3286950634109079, + "grad_norm": 0.1435546875, + "learning_rate": 0.0014100389314885766, + "loss": 0.1201, + "step": 37866 + }, + { + "epoch": 0.32870374389111207, + "grad_norm": 0.376953125, + "learning_rate": 0.0014100109905975948, + "loss": 0.166, + "step": 37867 + }, + { + "epoch": 0.3287124243713162, + "grad_norm": 0.6015625, + "learning_rate": 0.0014099830493675897, + "loss": 0.0762, + "step": 37868 + }, + { + "epoch": 0.3287211048515204, + "grad_norm": 0.34375, + "learning_rate": 0.001409955107798592, + "loss": 0.0859, + "step": 37869 + }, + { + "epoch": 0.32872978533172453, + "grad_norm": 0.25, + "learning_rate": 0.0014099271658906316, + "loss": 0.1025, + "step": 37870 + }, + { + "epoch": 0.3287384658119287, + "grad_norm": 0.341796875, + "learning_rate": 0.0014098992236437399, + "loss": 0.1084, + "step": 37871 + }, + { + "epoch": 0.32874714629213286, + "grad_norm": 0.08984375, + "learning_rate": 0.0014098712810579471, + "loss": 0.1309, + "step": 37872 + }, + { + "epoch": 0.32875582677233706, + "grad_norm": 0.10498046875, + "learning_rate": 0.0014098433381332833, + "loss": 0.0947, + "step": 37873 + }, + { + "epoch": 0.3287645072525412, + "grad_norm": 0.1015625, + "learning_rate": 0.00140981539486978, + "loss": 0.0791, + "step": 37874 + }, + { + "epoch": 0.3287731877327454, + "grad_norm": 0.197265625, + "learning_rate": 0.0014097874512674666, + "loss": 0.1162, + "step": 37875 + }, + { + "epoch": 0.3287818682129495, + "grad_norm": 0.349609375, + "learning_rate": 0.001409759507326375, + "loss": 0.0854, + "step": 37876 + }, + { + "epoch": 0.3287905486931537, + "grad_norm": 0.78515625, + "learning_rate": 0.0014097315630465345, + "loss": 0.1016, + "step": 37877 + }, + { + "epoch": 0.32879922917335785, + "grad_norm": 0.1953125, + "learning_rate": 0.0014097036184279764, + "loss": 0.1162, + "step": 37878 + }, + { + "epoch": 0.32880790965356205, + "grad_norm": 0.33984375, + "learning_rate": 0.001409675673470731, + "loss": 0.0859, + "step": 37879 + }, + { + "epoch": 0.3288165901337662, + "grad_norm": 0.08447265625, + "learning_rate": 0.001409647728174829, + "loss": 0.0679, + "step": 37880 + }, + { + "epoch": 0.3288252706139704, + "grad_norm": 0.27734375, + "learning_rate": 0.0014096197825403012, + "loss": 0.0869, + "step": 37881 + }, + { + "epoch": 0.3288339510941745, + "grad_norm": 0.2275390625, + "learning_rate": 0.0014095918365671775, + "loss": 0.0879, + "step": 37882 + }, + { + "epoch": 0.3288426315743787, + "grad_norm": 0.0712890625, + "learning_rate": 0.001409563890255489, + "loss": 0.0835, + "step": 37883 + }, + { + "epoch": 0.32885131205458284, + "grad_norm": 0.39453125, + "learning_rate": 0.001409535943605266, + "loss": 0.1377, + "step": 37884 + }, + { + "epoch": 0.32885999253478704, + "grad_norm": 0.31640625, + "learning_rate": 0.0014095079966165392, + "loss": 0.0801, + "step": 37885 + }, + { + "epoch": 0.3288686730149912, + "grad_norm": 0.302734375, + "learning_rate": 0.001409480049289339, + "loss": 0.0898, + "step": 37886 + }, + { + "epoch": 0.32887735349519537, + "grad_norm": 0.265625, + "learning_rate": 0.001409452101623696, + "loss": 0.1162, + "step": 37887 + }, + { + "epoch": 0.3288860339753995, + "grad_norm": 0.1640625, + "learning_rate": 0.0014094241536196411, + "loss": 0.0742, + "step": 37888 + }, + { + "epoch": 0.3288947144556037, + "grad_norm": 0.8046875, + "learning_rate": 0.0014093962052772043, + "loss": 0.0967, + "step": 37889 + }, + { + "epoch": 0.32890339493580784, + "grad_norm": 0.5625, + "learning_rate": 0.0014093682565964165, + "loss": 0.085, + "step": 37890 + }, + { + "epoch": 0.32891207541601203, + "grad_norm": 0.421875, + "learning_rate": 0.001409340307577308, + "loss": 0.1396, + "step": 37891 + }, + { + "epoch": 0.32892075589621617, + "grad_norm": 0.142578125, + "learning_rate": 0.0014093123582199104, + "loss": 0.0967, + "step": 37892 + }, + { + "epoch": 0.32892943637642036, + "grad_norm": 0.26171875, + "learning_rate": 0.0014092844085242529, + "loss": 0.124, + "step": 37893 + }, + { + "epoch": 0.3289381168566245, + "grad_norm": 0.54296875, + "learning_rate": 0.0014092564584903666, + "loss": 0.103, + "step": 37894 + }, + { + "epoch": 0.3289467973368287, + "grad_norm": 0.11181640625, + "learning_rate": 0.001409228508118282, + "loss": 0.1152, + "step": 37895 + }, + { + "epoch": 0.3289554778170328, + "grad_norm": 0.265625, + "learning_rate": 0.0014092005574080303, + "loss": 0.0762, + "step": 37896 + }, + { + "epoch": 0.328964158297237, + "grad_norm": 1.1640625, + "learning_rate": 0.0014091726063596413, + "loss": 0.209, + "step": 37897 + }, + { + "epoch": 0.32897283877744116, + "grad_norm": 0.169921875, + "learning_rate": 0.0014091446549731454, + "loss": 0.0918, + "step": 37898 + }, + { + "epoch": 0.32898151925764535, + "grad_norm": 0.234375, + "learning_rate": 0.0014091167032485735, + "loss": 0.1094, + "step": 37899 + }, + { + "epoch": 0.3289901997378495, + "grad_norm": 0.67578125, + "learning_rate": 0.0014090887511859565, + "loss": 0.1172, + "step": 37900 + }, + { + "epoch": 0.3289988802180537, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014090607987853247, + "loss": 0.1338, + "step": 37901 + }, + { + "epoch": 0.3290075606982578, + "grad_norm": 0.27734375, + "learning_rate": 0.0014090328460467088, + "loss": 0.0986, + "step": 37902 + }, + { + "epoch": 0.329016241178462, + "grad_norm": 0.1162109375, + "learning_rate": 0.001409004892970139, + "loss": 0.0942, + "step": 37903 + }, + { + "epoch": 0.32902492165866615, + "grad_norm": 0.74609375, + "learning_rate": 0.001408976939555646, + "loss": 0.082, + "step": 37904 + }, + { + "epoch": 0.32903360213887034, + "grad_norm": 0.1171875, + "learning_rate": 0.0014089489858032608, + "loss": 0.1289, + "step": 37905 + }, + { + "epoch": 0.3290422826190745, + "grad_norm": 0.130859375, + "learning_rate": 0.0014089210317130136, + "loss": 0.0723, + "step": 37906 + }, + { + "epoch": 0.32905096309927867, + "grad_norm": 0.2421875, + "learning_rate": 0.0014088930772849343, + "loss": 0.0947, + "step": 37907 + }, + { + "epoch": 0.3290596435794828, + "grad_norm": 1.3984375, + "learning_rate": 0.001408865122519055, + "loss": 0.1084, + "step": 37908 + }, + { + "epoch": 0.329068324059687, + "grad_norm": 0.119140625, + "learning_rate": 0.001408837167415405, + "loss": 0.1045, + "step": 37909 + }, + { + "epoch": 0.32907700453989114, + "grad_norm": 0.08349609375, + "learning_rate": 0.0014088092119740154, + "loss": 0.0996, + "step": 37910 + }, + { + "epoch": 0.32908568502009533, + "grad_norm": 0.39453125, + "learning_rate": 0.0014087812561949169, + "loss": 0.104, + "step": 37911 + }, + { + "epoch": 0.32909436550029947, + "grad_norm": 0.20703125, + "learning_rate": 0.0014087533000781396, + "loss": 0.0918, + "step": 37912 + }, + { + "epoch": 0.32910304598050366, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014087253436237143, + "loss": 0.1143, + "step": 37913 + }, + { + "epoch": 0.3291117264607078, + "grad_norm": 0.10205078125, + "learning_rate": 0.0014086973868316722, + "loss": 0.1406, + "step": 37914 + }, + { + "epoch": 0.329120406940912, + "grad_norm": 0.40625, + "learning_rate": 0.0014086694297020428, + "loss": 0.0933, + "step": 37915 + }, + { + "epoch": 0.32912908742111613, + "grad_norm": 0.69921875, + "learning_rate": 0.0014086414722348569, + "loss": 0.1084, + "step": 37916 + }, + { + "epoch": 0.3291377679013203, + "grad_norm": 0.11328125, + "learning_rate": 0.0014086135144301458, + "loss": 0.1299, + "step": 37917 + }, + { + "epoch": 0.32914644838152446, + "grad_norm": 0.09375, + "learning_rate": 0.001408585556287939, + "loss": 0.0903, + "step": 37918 + }, + { + "epoch": 0.32915512886172865, + "grad_norm": 0.359375, + "learning_rate": 0.0014085575978082681, + "loss": 0.1235, + "step": 37919 + }, + { + "epoch": 0.3291638093419328, + "grad_norm": 0.283203125, + "learning_rate": 0.0014085296389911635, + "loss": 0.1025, + "step": 37920 + }, + { + "epoch": 0.329172489822137, + "grad_norm": 0.298828125, + "learning_rate": 0.0014085016798366548, + "loss": 0.1025, + "step": 37921 + }, + { + "epoch": 0.3291811703023411, + "grad_norm": 0.69140625, + "learning_rate": 0.001408473720344774, + "loss": 0.1094, + "step": 37922 + }, + { + "epoch": 0.3291898507825453, + "grad_norm": 0.1484375, + "learning_rate": 0.0014084457605155506, + "loss": 0.1133, + "step": 37923 + }, + { + "epoch": 0.32919853126274945, + "grad_norm": 0.255859375, + "learning_rate": 0.0014084178003490157, + "loss": 0.0898, + "step": 37924 + }, + { + "epoch": 0.32920721174295364, + "grad_norm": 0.1533203125, + "learning_rate": 0.0014083898398451998, + "loss": 0.127, + "step": 37925 + }, + { + "epoch": 0.3292158922231578, + "grad_norm": 0.41015625, + "learning_rate": 0.0014083618790041333, + "loss": 0.1035, + "step": 37926 + }, + { + "epoch": 0.329224572703362, + "grad_norm": 0.0859375, + "learning_rate": 0.0014083339178258467, + "loss": 0.1064, + "step": 37927 + }, + { + "epoch": 0.3292332531835661, + "grad_norm": 1.671875, + "learning_rate": 0.0014083059563103708, + "loss": 0.1396, + "step": 37928 + }, + { + "epoch": 0.3292419336637703, + "grad_norm": 0.30859375, + "learning_rate": 0.001408277994457736, + "loss": 0.1152, + "step": 37929 + }, + { + "epoch": 0.32925061414397444, + "grad_norm": 0.10888671875, + "learning_rate": 0.0014082500322679734, + "loss": 0.1113, + "step": 37930 + }, + { + "epoch": 0.32925929462417863, + "grad_norm": 0.32421875, + "learning_rate": 0.0014082220697411135, + "loss": 0.123, + "step": 37931 + }, + { + "epoch": 0.32926797510438277, + "grad_norm": 0.189453125, + "learning_rate": 0.001408194106877186, + "loss": 0.0889, + "step": 37932 + }, + { + "epoch": 0.32927665558458696, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014081661436762224, + "loss": 0.0781, + "step": 37933 + }, + { + "epoch": 0.3292853360647911, + "grad_norm": 0.302734375, + "learning_rate": 0.0014081381801382526, + "loss": 0.1641, + "step": 37934 + }, + { + "epoch": 0.3292940165449953, + "grad_norm": 0.189453125, + "learning_rate": 0.0014081102162633076, + "loss": 0.1118, + "step": 37935 + }, + { + "epoch": 0.32930269702519943, + "grad_norm": 0.51171875, + "learning_rate": 0.001408082252051418, + "loss": 0.0938, + "step": 37936 + }, + { + "epoch": 0.3293113775054036, + "grad_norm": 0.3671875, + "learning_rate": 0.0014080542875026141, + "loss": 0.0977, + "step": 37937 + }, + { + "epoch": 0.32932005798560776, + "grad_norm": 0.265625, + "learning_rate": 0.0014080263226169268, + "loss": 0.0752, + "step": 37938 + }, + { + "epoch": 0.32932873846581195, + "grad_norm": 0.1298828125, + "learning_rate": 0.0014079983573943864, + "loss": 0.1201, + "step": 37939 + }, + { + "epoch": 0.3293374189460161, + "grad_norm": 0.2294921875, + "learning_rate": 0.001407970391835024, + "loss": 0.1221, + "step": 37940 + }, + { + "epoch": 0.3293460994262203, + "grad_norm": 0.26953125, + "learning_rate": 0.0014079424259388696, + "loss": 0.0996, + "step": 37941 + }, + { + "epoch": 0.3293547799064244, + "grad_norm": 0.1552734375, + "learning_rate": 0.0014079144597059538, + "loss": 0.1523, + "step": 37942 + }, + { + "epoch": 0.3293634603866286, + "grad_norm": 0.1884765625, + "learning_rate": 0.0014078864931363075, + "loss": 0.0928, + "step": 37943 + }, + { + "epoch": 0.32937214086683275, + "grad_norm": 0.29296875, + "learning_rate": 0.0014078585262299612, + "loss": 0.0752, + "step": 37944 + }, + { + "epoch": 0.32938082134703695, + "grad_norm": 0.318359375, + "learning_rate": 0.0014078305589869456, + "loss": 0.0723, + "step": 37945 + }, + { + "epoch": 0.3293895018272411, + "grad_norm": 0.15625, + "learning_rate": 0.0014078025914072909, + "loss": 0.1157, + "step": 37946 + }, + { + "epoch": 0.3293981823074453, + "grad_norm": 0.1328125, + "learning_rate": 0.0014077746234910278, + "loss": 0.1152, + "step": 37947 + }, + { + "epoch": 0.3294068627876494, + "grad_norm": 0.48828125, + "learning_rate": 0.001407746655238187, + "loss": 0.0952, + "step": 37948 + }, + { + "epoch": 0.32941554326785355, + "grad_norm": 0.21484375, + "learning_rate": 0.0014077186866487995, + "loss": 0.083, + "step": 37949 + }, + { + "epoch": 0.32942422374805774, + "grad_norm": 0.1826171875, + "learning_rate": 0.0014076907177228953, + "loss": 0.1035, + "step": 37950 + }, + { + "epoch": 0.3294329042282619, + "grad_norm": 0.458984375, + "learning_rate": 0.0014076627484605048, + "loss": 0.1719, + "step": 37951 + }, + { + "epoch": 0.3294415847084661, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014076347788616595, + "loss": 0.0854, + "step": 37952 + }, + { + "epoch": 0.3294502651886702, + "grad_norm": 0.08984375, + "learning_rate": 0.0014076068089263892, + "loss": 0.0811, + "step": 37953 + }, + { + "epoch": 0.3294589456688744, + "grad_norm": 0.11328125, + "learning_rate": 0.0014075788386547245, + "loss": 0.1118, + "step": 37954 + }, + { + "epoch": 0.32946762614907854, + "grad_norm": 0.07763671875, + "learning_rate": 0.0014075508680466962, + "loss": 0.0811, + "step": 37955 + }, + { + "epoch": 0.32947630662928273, + "grad_norm": 0.5234375, + "learning_rate": 0.0014075228971023352, + "loss": 0.0996, + "step": 37956 + }, + { + "epoch": 0.32948498710948687, + "grad_norm": 0.376953125, + "learning_rate": 0.0014074949258216713, + "loss": 0.1123, + "step": 37957 + }, + { + "epoch": 0.32949366758969106, + "grad_norm": 0.2890625, + "learning_rate": 0.001407466954204736, + "loss": 0.0825, + "step": 37958 + }, + { + "epoch": 0.3295023480698952, + "grad_norm": 0.478515625, + "learning_rate": 0.0014074389822515594, + "loss": 0.0859, + "step": 37959 + }, + { + "epoch": 0.3295110285500994, + "grad_norm": 0.2421875, + "learning_rate": 0.001407411009962172, + "loss": 0.0889, + "step": 37960 + }, + { + "epoch": 0.32951970903030353, + "grad_norm": 0.3125, + "learning_rate": 0.0014073830373366048, + "loss": 0.1465, + "step": 37961 + }, + { + "epoch": 0.3295283895105077, + "grad_norm": 0.1484375, + "learning_rate": 0.0014073550643748881, + "loss": 0.1084, + "step": 37962 + }, + { + "epoch": 0.32953706999071186, + "grad_norm": 0.373046875, + "learning_rate": 0.0014073270910770525, + "loss": 0.082, + "step": 37963 + }, + { + "epoch": 0.32954575047091605, + "grad_norm": 0.1728515625, + "learning_rate": 0.001407299117443128, + "loss": 0.0869, + "step": 37964 + }, + { + "epoch": 0.3295544309511202, + "grad_norm": 0.326171875, + "learning_rate": 0.0014072711434731464, + "loss": 0.123, + "step": 37965 + }, + { + "epoch": 0.3295631114313244, + "grad_norm": 0.1181640625, + "learning_rate": 0.0014072431691671374, + "loss": 0.1079, + "step": 37966 + }, + { + "epoch": 0.3295717919115285, + "grad_norm": 0.1572265625, + "learning_rate": 0.0014072151945251324, + "loss": 0.085, + "step": 37967 + }, + { + "epoch": 0.3295804723917327, + "grad_norm": 0.0908203125, + "learning_rate": 0.001407187219547161, + "loss": 0.127, + "step": 37968 + }, + { + "epoch": 0.32958915287193685, + "grad_norm": 0.11474609375, + "learning_rate": 0.0014071592442332543, + "loss": 0.1416, + "step": 37969 + }, + { + "epoch": 0.32959783335214105, + "grad_norm": 0.2890625, + "learning_rate": 0.0014071312685834433, + "loss": 0.1079, + "step": 37970 + }, + { + "epoch": 0.3296065138323452, + "grad_norm": 0.30078125, + "learning_rate": 0.001407103292597758, + "loss": 0.0791, + "step": 37971 + }, + { + "epoch": 0.3296151943125494, + "grad_norm": 0.25390625, + "learning_rate": 0.0014070753162762288, + "loss": 0.054, + "step": 37972 + }, + { + "epoch": 0.3296238747927535, + "grad_norm": 0.2890625, + "learning_rate": 0.0014070473396188869, + "loss": 0.1074, + "step": 37973 + }, + { + "epoch": 0.3296325552729577, + "grad_norm": 0.1865234375, + "learning_rate": 0.0014070193626257626, + "loss": 0.1465, + "step": 37974 + }, + { + "epoch": 0.32964123575316184, + "grad_norm": 0.298828125, + "learning_rate": 0.0014069913852968862, + "loss": 0.1172, + "step": 37975 + }, + { + "epoch": 0.32964991623336604, + "grad_norm": 0.1787109375, + "learning_rate": 0.001406963407632289, + "loss": 0.0786, + "step": 37976 + }, + { + "epoch": 0.3296585967135702, + "grad_norm": 0.1123046875, + "learning_rate": 0.0014069354296320012, + "loss": 0.1108, + "step": 37977 + }, + { + "epoch": 0.32966727719377437, + "grad_norm": 0.22265625, + "learning_rate": 0.0014069074512960535, + "loss": 0.1006, + "step": 37978 + }, + { + "epoch": 0.3296759576739785, + "grad_norm": 0.171875, + "learning_rate": 0.0014068794726244764, + "loss": 0.1021, + "step": 37979 + }, + { + "epoch": 0.3296846381541827, + "grad_norm": 0.376953125, + "learning_rate": 0.0014068514936173004, + "loss": 0.1113, + "step": 37980 + }, + { + "epoch": 0.32969331863438683, + "grad_norm": 0.13671875, + "learning_rate": 0.0014068235142745563, + "loss": 0.1025, + "step": 37981 + }, + { + "epoch": 0.329701999114591, + "grad_norm": 0.341796875, + "learning_rate": 0.0014067955345962748, + "loss": 0.0918, + "step": 37982 + }, + { + "epoch": 0.32971067959479516, + "grad_norm": 0.287109375, + "learning_rate": 0.001406767554582486, + "loss": 0.0781, + "step": 37983 + }, + { + "epoch": 0.32971936007499936, + "grad_norm": 0.0849609375, + "learning_rate": 0.001406739574233221, + "loss": 0.1338, + "step": 37984 + }, + { + "epoch": 0.3297280405552035, + "grad_norm": 0.099609375, + "learning_rate": 0.0014067115935485102, + "loss": 0.085, + "step": 37985 + }, + { + "epoch": 0.3297367210354077, + "grad_norm": 0.216796875, + "learning_rate": 0.0014066836125283836, + "loss": 0.0923, + "step": 37986 + }, + { + "epoch": 0.3297454015156118, + "grad_norm": 0.103515625, + "learning_rate": 0.0014066556311728732, + "loss": 0.0996, + "step": 37987 + }, + { + "epoch": 0.329754081995816, + "grad_norm": 0.1640625, + "learning_rate": 0.0014066276494820085, + "loss": 0.084, + "step": 37988 + }, + { + "epoch": 0.32976276247602015, + "grad_norm": 0.142578125, + "learning_rate": 0.0014065996674558206, + "loss": 0.1553, + "step": 37989 + }, + { + "epoch": 0.32977144295622435, + "grad_norm": 1.7109375, + "learning_rate": 0.0014065716850943397, + "loss": 0.2578, + "step": 37990 + }, + { + "epoch": 0.3297801234364285, + "grad_norm": 0.08837890625, + "learning_rate": 0.0014065437023975966, + "loss": 0.0903, + "step": 37991 + }, + { + "epoch": 0.3297888039166327, + "grad_norm": 0.34375, + "learning_rate": 0.0014065157193656223, + "loss": 0.0996, + "step": 37992 + }, + { + "epoch": 0.3297974843968368, + "grad_norm": 0.146484375, + "learning_rate": 0.0014064877359984468, + "loss": 0.082, + "step": 37993 + }, + { + "epoch": 0.329806164877041, + "grad_norm": 0.51953125, + "learning_rate": 0.0014064597522961007, + "loss": 0.0938, + "step": 37994 + }, + { + "epoch": 0.32981484535724515, + "grad_norm": 0.291015625, + "learning_rate": 0.0014064317682586148, + "loss": 0.064, + "step": 37995 + }, + { + "epoch": 0.32982352583744934, + "grad_norm": 0.326171875, + "learning_rate": 0.0014064037838860198, + "loss": 0.1226, + "step": 37996 + }, + { + "epoch": 0.3298322063176535, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014063757991783464, + "loss": 0.0894, + "step": 37997 + }, + { + "epoch": 0.32984088679785767, + "grad_norm": 0.353515625, + "learning_rate": 0.001406347814135625, + "loss": 0.0908, + "step": 37998 + }, + { + "epoch": 0.3298495672780618, + "grad_norm": 0.3125, + "learning_rate": 0.001406319828757886, + "loss": 0.0854, + "step": 37999 + }, + { + "epoch": 0.329858247758266, + "grad_norm": 0.25390625, + "learning_rate": 0.0014062918430451609, + "loss": 0.0898, + "step": 38000 + }, + { + "epoch": 0.32986692823847014, + "grad_norm": 0.32421875, + "learning_rate": 0.0014062638569974789, + "loss": 0.1289, + "step": 38001 + }, + { + "epoch": 0.32987560871867433, + "grad_norm": 0.1953125, + "learning_rate": 0.0014062358706148717, + "loss": 0.1299, + "step": 38002 + }, + { + "epoch": 0.32988428919887847, + "grad_norm": 0.236328125, + "learning_rate": 0.0014062078838973694, + "loss": 0.082, + "step": 38003 + }, + { + "epoch": 0.32989296967908266, + "grad_norm": 0.30078125, + "learning_rate": 0.0014061798968450026, + "loss": 0.1016, + "step": 38004 + }, + { + "epoch": 0.3299016501592868, + "grad_norm": 0.33984375, + "learning_rate": 0.0014061519094578023, + "loss": 0.1045, + "step": 38005 + }, + { + "epoch": 0.329910330639491, + "grad_norm": 0.51953125, + "learning_rate": 0.0014061239217357988, + "loss": 0.1133, + "step": 38006 + }, + { + "epoch": 0.3299190111196951, + "grad_norm": 0.1884765625, + "learning_rate": 0.001406095933679023, + "loss": 0.0679, + "step": 38007 + }, + { + "epoch": 0.3299276915998993, + "grad_norm": 0.31640625, + "learning_rate": 0.001406067945287505, + "loss": 0.0732, + "step": 38008 + }, + { + "epoch": 0.32993637208010346, + "grad_norm": 0.3359375, + "learning_rate": 0.0014060399565612757, + "loss": 0.0879, + "step": 38009 + }, + { + "epoch": 0.32994505256030765, + "grad_norm": 0.279296875, + "learning_rate": 0.001406011967500366, + "loss": 0.1484, + "step": 38010 + }, + { + "epoch": 0.3299537330405118, + "grad_norm": 0.201171875, + "learning_rate": 0.0014059839781048059, + "loss": 0.0854, + "step": 38011 + }, + { + "epoch": 0.329962413520716, + "grad_norm": 0.28125, + "learning_rate": 0.0014059559883746262, + "loss": 0.1055, + "step": 38012 + }, + { + "epoch": 0.3299710940009201, + "grad_norm": 0.361328125, + "learning_rate": 0.0014059279983098583, + "loss": 0.1201, + "step": 38013 + }, + { + "epoch": 0.3299797744811243, + "grad_norm": 0.24609375, + "learning_rate": 0.0014059000079105315, + "loss": 0.0811, + "step": 38014 + }, + { + "epoch": 0.32998845496132845, + "grad_norm": 0.291015625, + "learning_rate": 0.0014058720171766772, + "loss": 0.0942, + "step": 38015 + }, + { + "epoch": 0.32999713544153264, + "grad_norm": 0.330078125, + "learning_rate": 0.0014058440261083258, + "loss": 0.0718, + "step": 38016 + }, + { + "epoch": 0.3300058159217368, + "grad_norm": 0.1474609375, + "learning_rate": 0.0014058160347055083, + "loss": 0.0986, + "step": 38017 + }, + { + "epoch": 0.33001449640194097, + "grad_norm": 0.15234375, + "learning_rate": 0.0014057880429682547, + "loss": 0.1455, + "step": 38018 + }, + { + "epoch": 0.3300231768821451, + "grad_norm": 0.1748046875, + "learning_rate": 0.0014057600508965958, + "loss": 0.0811, + "step": 38019 + }, + { + "epoch": 0.3300318573623493, + "grad_norm": 0.6640625, + "learning_rate": 0.0014057320584905627, + "loss": 0.1064, + "step": 38020 + }, + { + "epoch": 0.33004053784255344, + "grad_norm": 0.30859375, + "learning_rate": 0.0014057040657501853, + "loss": 0.1104, + "step": 38021 + }, + { + "epoch": 0.33004921832275763, + "grad_norm": 0.12353515625, + "learning_rate": 0.0014056760726754948, + "loss": 0.1123, + "step": 38022 + }, + { + "epoch": 0.33005789880296177, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014056480792665209, + "loss": 0.0781, + "step": 38023 + }, + { + "epoch": 0.33006657928316596, + "grad_norm": 0.1220703125, + "learning_rate": 0.001405620085523295, + "loss": 0.0845, + "step": 38024 + }, + { + "epoch": 0.3300752597633701, + "grad_norm": 0.09521484375, + "learning_rate": 0.0014055920914458483, + "loss": 0.0908, + "step": 38025 + }, + { + "epoch": 0.3300839402435743, + "grad_norm": 0.1416015625, + "learning_rate": 0.00140556409703421, + "loss": 0.1064, + "step": 38026 + }, + { + "epoch": 0.33009262072377843, + "grad_norm": 0.07373046875, + "learning_rate": 0.0014055361022884119, + "loss": 0.1045, + "step": 38027 + }, + { + "epoch": 0.3301013012039826, + "grad_norm": 0.625, + "learning_rate": 0.0014055081072084836, + "loss": 0.0962, + "step": 38028 + }, + { + "epoch": 0.33010998168418676, + "grad_norm": 0.412109375, + "learning_rate": 0.0014054801117944565, + "loss": 0.0928, + "step": 38029 + }, + { + "epoch": 0.33011866216439095, + "grad_norm": 0.30078125, + "learning_rate": 0.0014054521160463608, + "loss": 0.0859, + "step": 38030 + }, + { + "epoch": 0.3301273426445951, + "grad_norm": 0.1279296875, + "learning_rate": 0.0014054241199642275, + "loss": 0.1348, + "step": 38031 + }, + { + "epoch": 0.3301360231247993, + "grad_norm": 0.07763671875, + "learning_rate": 0.0014053961235480869, + "loss": 0.0879, + "step": 38032 + }, + { + "epoch": 0.3301447036050034, + "grad_norm": 0.2060546875, + "learning_rate": 0.0014053681267979694, + "loss": 0.1182, + "step": 38033 + }, + { + "epoch": 0.3301533840852076, + "grad_norm": 0.17578125, + "learning_rate": 0.001405340129713906, + "loss": 0.0962, + "step": 38034 + }, + { + "epoch": 0.33016206456541175, + "grad_norm": 0.158203125, + "learning_rate": 0.0014053121322959274, + "loss": 0.0786, + "step": 38035 + }, + { + "epoch": 0.33017074504561594, + "grad_norm": 0.373046875, + "learning_rate": 0.0014052841345440642, + "loss": 0.0786, + "step": 38036 + }, + { + "epoch": 0.3301794255258201, + "grad_norm": 0.369140625, + "learning_rate": 0.0014052561364583465, + "loss": 0.1006, + "step": 38037 + }, + { + "epoch": 0.3301881060060243, + "grad_norm": 0.271484375, + "learning_rate": 0.0014052281380388053, + "loss": 0.1079, + "step": 38038 + }, + { + "epoch": 0.3301967864862284, + "grad_norm": 0.69140625, + "learning_rate": 0.0014052001392854715, + "loss": 0.1143, + "step": 38039 + }, + { + "epoch": 0.3302054669664326, + "grad_norm": 0.5390625, + "learning_rate": 0.0014051721401983751, + "loss": 0.1162, + "step": 38040 + }, + { + "epoch": 0.33021414744663674, + "grad_norm": 0.09912109375, + "learning_rate": 0.0014051441407775473, + "loss": 0.106, + "step": 38041 + }, + { + "epoch": 0.33022282792684093, + "grad_norm": 0.1953125, + "learning_rate": 0.0014051161410230182, + "loss": 0.1143, + "step": 38042 + }, + { + "epoch": 0.33023150840704507, + "grad_norm": 0.431640625, + "learning_rate": 0.0014050881409348186, + "loss": 0.1211, + "step": 38043 + }, + { + "epoch": 0.33024018888724926, + "grad_norm": 0.212890625, + "learning_rate": 0.0014050601405129796, + "loss": 0.1055, + "step": 38044 + }, + { + "epoch": 0.3302488693674534, + "grad_norm": 0.11279296875, + "learning_rate": 0.0014050321397575311, + "loss": 0.1089, + "step": 38045 + }, + { + "epoch": 0.3302575498476576, + "grad_norm": 0.302734375, + "learning_rate": 0.0014050041386685038, + "loss": 0.1514, + "step": 38046 + }, + { + "epoch": 0.33026623032786173, + "grad_norm": 0.416015625, + "learning_rate": 0.001404976137245929, + "loss": 0.0918, + "step": 38047 + }, + { + "epoch": 0.3302749108080659, + "grad_norm": 0.09716796875, + "learning_rate": 0.0014049481354898368, + "loss": 0.0669, + "step": 38048 + }, + { + "epoch": 0.33028359128827006, + "grad_norm": 0.09228515625, + "learning_rate": 0.001404920133400258, + "loss": 0.0996, + "step": 38049 + }, + { + "epoch": 0.33029227176847425, + "grad_norm": 0.1767578125, + "learning_rate": 0.0014048921309772226, + "loss": 0.1104, + "step": 38050 + }, + { + "epoch": 0.3303009522486784, + "grad_norm": 0.259765625, + "learning_rate": 0.0014048641282207624, + "loss": 0.0996, + "step": 38051 + }, + { + "epoch": 0.3303096327288826, + "grad_norm": 0.298828125, + "learning_rate": 0.001404836125130907, + "loss": 0.127, + "step": 38052 + }, + { + "epoch": 0.3303183132090867, + "grad_norm": 0.1376953125, + "learning_rate": 0.0014048081217076873, + "loss": 0.0859, + "step": 38053 + }, + { + "epoch": 0.3303269936892909, + "grad_norm": 0.41796875, + "learning_rate": 0.0014047801179511343, + "loss": 0.0718, + "step": 38054 + }, + { + "epoch": 0.33033567416949505, + "grad_norm": 0.4453125, + "learning_rate": 0.0014047521138612783, + "loss": 0.0938, + "step": 38055 + }, + { + "epoch": 0.33034435464969925, + "grad_norm": 0.2734375, + "learning_rate": 0.0014047241094381497, + "loss": 0.0859, + "step": 38056 + }, + { + "epoch": 0.3303530351299034, + "grad_norm": 0.08203125, + "learning_rate": 0.00140469610468178, + "loss": 0.0835, + "step": 38057 + }, + { + "epoch": 0.3303617156101076, + "grad_norm": 0.07958984375, + "learning_rate": 0.0014046680995921986, + "loss": 0.0811, + "step": 38058 + }, + { + "epoch": 0.3303703960903117, + "grad_norm": 0.1728515625, + "learning_rate": 0.001404640094169437, + "loss": 0.0659, + "step": 38059 + }, + { + "epoch": 0.3303790765705159, + "grad_norm": 0.3359375, + "learning_rate": 0.0014046120884135254, + "loss": 0.3145, + "step": 38060 + }, + { + "epoch": 0.33038775705072004, + "grad_norm": 0.205078125, + "learning_rate": 0.001404584082324495, + "loss": 0.0923, + "step": 38061 + }, + { + "epoch": 0.33039643753092424, + "grad_norm": 0.345703125, + "learning_rate": 0.0014045560759023757, + "loss": 0.0767, + "step": 38062 + }, + { + "epoch": 0.3304051180111284, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014045280691471983, + "loss": 0.1016, + "step": 38063 + }, + { + "epoch": 0.33041379849133257, + "grad_norm": 0.2041015625, + "learning_rate": 0.0014045000620589937, + "loss": 0.1021, + "step": 38064 + }, + { + "epoch": 0.3304224789715367, + "grad_norm": 0.62890625, + "learning_rate": 0.0014044720546377925, + "loss": 0.1309, + "step": 38065 + }, + { + "epoch": 0.3304311594517409, + "grad_norm": 0.306640625, + "learning_rate": 0.0014044440468836256, + "loss": 0.1367, + "step": 38066 + }, + { + "epoch": 0.33043983993194503, + "grad_norm": 0.462890625, + "learning_rate": 0.0014044160387965228, + "loss": 0.1104, + "step": 38067 + }, + { + "epoch": 0.3304485204121492, + "grad_norm": 0.76953125, + "learning_rate": 0.001404388030376515, + "loss": 0.1328, + "step": 38068 + }, + { + "epoch": 0.33045720089235336, + "grad_norm": 0.140625, + "learning_rate": 0.0014043600216236333, + "loss": 0.1377, + "step": 38069 + }, + { + "epoch": 0.33046588137255756, + "grad_norm": 0.1640625, + "learning_rate": 0.0014043320125379083, + "loss": 0.1133, + "step": 38070 + }, + { + "epoch": 0.3304745618527617, + "grad_norm": 0.48046875, + "learning_rate": 0.0014043040031193699, + "loss": 0.1172, + "step": 38071 + }, + { + "epoch": 0.33048324233296583, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014042759933680492, + "loss": 0.1152, + "step": 38072 + }, + { + "epoch": 0.33049192281317, + "grad_norm": 0.359375, + "learning_rate": 0.0014042479832839774, + "loss": 0.1221, + "step": 38073 + }, + { + "epoch": 0.33050060329337416, + "grad_norm": 0.26953125, + "learning_rate": 0.0014042199728671846, + "loss": 0.0742, + "step": 38074 + }, + { + "epoch": 0.33050928377357836, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014041919621177008, + "loss": 0.1055, + "step": 38075 + }, + { + "epoch": 0.3305179642537825, + "grad_norm": 0.357421875, + "learning_rate": 0.0014041639510355575, + "loss": 0.1055, + "step": 38076 + }, + { + "epoch": 0.3305266447339867, + "grad_norm": 0.189453125, + "learning_rate": 0.0014041359396207854, + "loss": 0.0928, + "step": 38077 + }, + { + "epoch": 0.3305353252141908, + "grad_norm": 0.4453125, + "learning_rate": 0.0014041079278734143, + "loss": 0.0889, + "step": 38078 + }, + { + "epoch": 0.330544005694395, + "grad_norm": 0.412109375, + "learning_rate": 0.0014040799157934757, + "loss": 0.0742, + "step": 38079 + }, + { + "epoch": 0.33055268617459915, + "grad_norm": 0.26171875, + "learning_rate": 0.0014040519033809995, + "loss": 0.123, + "step": 38080 + }, + { + "epoch": 0.33056136665480335, + "grad_norm": 0.1484375, + "learning_rate": 0.0014040238906360167, + "loss": 0.1094, + "step": 38081 + }, + { + "epoch": 0.3305700471350075, + "grad_norm": 0.5, + "learning_rate": 0.0014039958775585585, + "loss": 0.1104, + "step": 38082 + }, + { + "epoch": 0.3305787276152117, + "grad_norm": 0.396484375, + "learning_rate": 0.0014039678641486546, + "loss": 0.1123, + "step": 38083 + }, + { + "epoch": 0.3305874080954158, + "grad_norm": 0.193359375, + "learning_rate": 0.0014039398504063362, + "loss": 0.1367, + "step": 38084 + }, + { + "epoch": 0.33059608857562, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014039118363316334, + "loss": 0.0894, + "step": 38085 + }, + { + "epoch": 0.33060476905582414, + "grad_norm": 0.3515625, + "learning_rate": 0.0014038838219245773, + "loss": 0.123, + "step": 38086 + }, + { + "epoch": 0.33061344953602834, + "grad_norm": 0.103515625, + "learning_rate": 0.0014038558071851987, + "loss": 0.1123, + "step": 38087 + }, + { + "epoch": 0.3306221300162325, + "grad_norm": 0.478515625, + "learning_rate": 0.0014038277921135278, + "loss": 0.0913, + "step": 38088 + }, + { + "epoch": 0.33063081049643667, + "grad_norm": 0.373046875, + "learning_rate": 0.0014037997767095954, + "loss": 0.1084, + "step": 38089 + }, + { + "epoch": 0.3306394909766408, + "grad_norm": 0.703125, + "learning_rate": 0.001403771760973432, + "loss": 0.1191, + "step": 38090 + }, + { + "epoch": 0.330648171456845, + "grad_norm": 1.015625, + "learning_rate": 0.0014037437449050682, + "loss": 0.1562, + "step": 38091 + }, + { + "epoch": 0.33065685193704913, + "grad_norm": 1.1171875, + "learning_rate": 0.0014037157285045352, + "loss": 0.1123, + "step": 38092 + }, + { + "epoch": 0.3306655324172533, + "grad_norm": 0.2099609375, + "learning_rate": 0.0014036877117718633, + "loss": 0.1069, + "step": 38093 + }, + { + "epoch": 0.33067421289745746, + "grad_norm": 0.16796875, + "learning_rate": 0.0014036596947070825, + "loss": 0.0801, + "step": 38094 + }, + { + "epoch": 0.33068289337766166, + "grad_norm": 0.412109375, + "learning_rate": 0.0014036316773102246, + "loss": 0.0962, + "step": 38095 + }, + { + "epoch": 0.3306915738578658, + "grad_norm": 0.82421875, + "learning_rate": 0.0014036036595813195, + "loss": 0.0962, + "step": 38096 + }, + { + "epoch": 0.33070025433807, + "grad_norm": 0.076171875, + "learning_rate": 0.001403575641520398, + "loss": 0.1064, + "step": 38097 + }, + { + "epoch": 0.3307089348182741, + "grad_norm": 0.40234375, + "learning_rate": 0.0014035476231274906, + "loss": 0.126, + "step": 38098 + }, + { + "epoch": 0.3307176152984783, + "grad_norm": 0.3828125, + "learning_rate": 0.0014035196044026282, + "loss": 0.1211, + "step": 38099 + }, + { + "epoch": 0.33072629577868246, + "grad_norm": 0.296875, + "learning_rate": 0.0014034915853458412, + "loss": 0.1621, + "step": 38100 + }, + { + "epoch": 0.33073497625888665, + "grad_norm": 0.41015625, + "learning_rate": 0.0014034635659571605, + "loss": 0.0898, + "step": 38101 + }, + { + "epoch": 0.3307436567390908, + "grad_norm": 0.5859375, + "learning_rate": 0.0014034355462366165, + "loss": 0.1226, + "step": 38102 + }, + { + "epoch": 0.330752337219295, + "grad_norm": 0.146484375, + "learning_rate": 0.0014034075261842398, + "loss": 0.083, + "step": 38103 + }, + { + "epoch": 0.3307610176994991, + "grad_norm": 0.1484375, + "learning_rate": 0.0014033795058000615, + "loss": 0.0903, + "step": 38104 + }, + { + "epoch": 0.3307696981797033, + "grad_norm": 0.197265625, + "learning_rate": 0.001403351485084112, + "loss": 0.1001, + "step": 38105 + }, + { + "epoch": 0.33077837865990745, + "grad_norm": 0.0927734375, + "learning_rate": 0.0014033234640364219, + "loss": 0.0742, + "step": 38106 + }, + { + "epoch": 0.33078705914011164, + "grad_norm": 0.08447265625, + "learning_rate": 0.0014032954426570214, + "loss": 0.0654, + "step": 38107 + }, + { + "epoch": 0.3307957396203158, + "grad_norm": 0.318359375, + "learning_rate": 0.0014032674209459417, + "loss": 0.0913, + "step": 38108 + }, + { + "epoch": 0.33080442010051997, + "grad_norm": 0.158203125, + "learning_rate": 0.0014032393989032135, + "loss": 0.1299, + "step": 38109 + }, + { + "epoch": 0.3308131005807241, + "grad_norm": 0.0712890625, + "learning_rate": 0.001403211376528867, + "loss": 0.085, + "step": 38110 + }, + { + "epoch": 0.3308217810609283, + "grad_norm": 0.25390625, + "learning_rate": 0.0014031833538229333, + "loss": 0.1387, + "step": 38111 + }, + { + "epoch": 0.33083046154113244, + "grad_norm": 0.1455078125, + "learning_rate": 0.0014031553307854428, + "loss": 0.103, + "step": 38112 + }, + { + "epoch": 0.33083914202133663, + "grad_norm": 0.27734375, + "learning_rate": 0.0014031273074164262, + "loss": 0.1211, + "step": 38113 + }, + { + "epoch": 0.33084782250154077, + "grad_norm": 0.404296875, + "learning_rate": 0.001403099283715914, + "loss": 0.0898, + "step": 38114 + }, + { + "epoch": 0.33085650298174496, + "grad_norm": 0.1552734375, + "learning_rate": 0.001403071259683937, + "loss": 0.0967, + "step": 38115 + }, + { + "epoch": 0.3308651834619491, + "grad_norm": 0.10888671875, + "learning_rate": 0.001403043235320526, + "loss": 0.0815, + "step": 38116 + }, + { + "epoch": 0.3308738639421533, + "grad_norm": 0.10791015625, + "learning_rate": 0.0014030152106257112, + "loss": 0.083, + "step": 38117 + }, + { + "epoch": 0.3308825444223574, + "grad_norm": 0.24609375, + "learning_rate": 0.0014029871855995239, + "loss": 0.1104, + "step": 38118 + }, + { + "epoch": 0.3308912249025616, + "grad_norm": 1.125, + "learning_rate": 0.001402959160241994, + "loss": 0.3477, + "step": 38119 + }, + { + "epoch": 0.33089990538276576, + "grad_norm": 0.1669921875, + "learning_rate": 0.0014029311345531523, + "loss": 0.1045, + "step": 38120 + }, + { + "epoch": 0.33090858586296995, + "grad_norm": 0.23046875, + "learning_rate": 0.00140290310853303, + "loss": 0.0894, + "step": 38121 + }, + { + "epoch": 0.3309172663431741, + "grad_norm": 0.2451171875, + "learning_rate": 0.0014028750821816576, + "loss": 0.1387, + "step": 38122 + }, + { + "epoch": 0.3309259468233783, + "grad_norm": 0.177734375, + "learning_rate": 0.0014028470554990654, + "loss": 0.1045, + "step": 38123 + }, + { + "epoch": 0.3309346273035824, + "grad_norm": 0.64453125, + "learning_rate": 0.0014028190284852843, + "loss": 0.0815, + "step": 38124 + }, + { + "epoch": 0.3309433077837866, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014027910011403446, + "loss": 0.0742, + "step": 38125 + }, + { + "epoch": 0.33095198826399075, + "grad_norm": 0.291015625, + "learning_rate": 0.0014027629734642776, + "loss": 0.0938, + "step": 38126 + }, + { + "epoch": 0.33096066874419494, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014027349454571133, + "loss": 0.1338, + "step": 38127 + }, + { + "epoch": 0.3309693492243991, + "grad_norm": 0.7265625, + "learning_rate": 0.0014027069171188824, + "loss": 0.1562, + "step": 38128 + }, + { + "epoch": 0.33097802970460327, + "grad_norm": 0.1513671875, + "learning_rate": 0.001402678888449616, + "loss": 0.082, + "step": 38129 + }, + { + "epoch": 0.3309867101848074, + "grad_norm": 0.10205078125, + "learning_rate": 0.0014026508594493447, + "loss": 0.105, + "step": 38130 + }, + { + "epoch": 0.3309953906650116, + "grad_norm": 0.326171875, + "learning_rate": 0.0014026228301180987, + "loss": 0.0884, + "step": 38131 + }, + { + "epoch": 0.33100407114521574, + "grad_norm": 0.68359375, + "learning_rate": 0.0014025948004559089, + "loss": 0.103, + "step": 38132 + }, + { + "epoch": 0.33101275162541993, + "grad_norm": 0.1591796875, + "learning_rate": 0.0014025667704628062, + "loss": 0.1123, + "step": 38133 + }, + { + "epoch": 0.33102143210562407, + "grad_norm": 0.322265625, + "learning_rate": 0.001402538740138821, + "loss": 0.1006, + "step": 38134 + }, + { + "epoch": 0.33103011258582826, + "grad_norm": 0.1337890625, + "learning_rate": 0.0014025107094839838, + "loss": 0.0996, + "step": 38135 + }, + { + "epoch": 0.3310387930660324, + "grad_norm": 0.3828125, + "learning_rate": 0.0014024826784983257, + "loss": 0.0874, + "step": 38136 + }, + { + "epoch": 0.3310474735462366, + "grad_norm": 0.189453125, + "learning_rate": 0.0014024546471818767, + "loss": 0.1006, + "step": 38137 + }, + { + "epoch": 0.33105615402644073, + "grad_norm": 0.470703125, + "learning_rate": 0.0014024266155346682, + "loss": 0.1191, + "step": 38138 + }, + { + "epoch": 0.3310648345066449, + "grad_norm": 0.294921875, + "learning_rate": 0.00140239858355673, + "loss": 0.0859, + "step": 38139 + }, + { + "epoch": 0.33107351498684906, + "grad_norm": 0.5234375, + "learning_rate": 0.0014023705512480938, + "loss": 0.1445, + "step": 38140 + }, + { + "epoch": 0.33108219546705325, + "grad_norm": 0.2119140625, + "learning_rate": 0.0014023425186087894, + "loss": 0.0967, + "step": 38141 + }, + { + "epoch": 0.3310908759472574, + "grad_norm": 0.08935546875, + "learning_rate": 0.0014023144856388481, + "loss": 0.1011, + "step": 38142 + }, + { + "epoch": 0.3310995564274616, + "grad_norm": 0.2265625, + "learning_rate": 0.0014022864523382998, + "loss": 0.1387, + "step": 38143 + }, + { + "epoch": 0.3311082369076657, + "grad_norm": 0.1318359375, + "learning_rate": 0.001402258418707176, + "loss": 0.0972, + "step": 38144 + }, + { + "epoch": 0.3311169173878699, + "grad_norm": 0.08984375, + "learning_rate": 0.0014022303847455067, + "loss": 0.0903, + "step": 38145 + }, + { + "epoch": 0.33112559786807405, + "grad_norm": 0.224609375, + "learning_rate": 0.0014022023504533226, + "loss": 0.1045, + "step": 38146 + }, + { + "epoch": 0.33113427834827824, + "grad_norm": 0.302734375, + "learning_rate": 0.0014021743158306548, + "loss": 0.0801, + "step": 38147 + }, + { + "epoch": 0.3311429588284824, + "grad_norm": 0.462890625, + "learning_rate": 0.0014021462808775336, + "loss": 0.0806, + "step": 38148 + }, + { + "epoch": 0.3311516393086866, + "grad_norm": 0.625, + "learning_rate": 0.00140211824559399, + "loss": 0.124, + "step": 38149 + }, + { + "epoch": 0.3311603197888907, + "grad_norm": 0.1416015625, + "learning_rate": 0.0014020902099800541, + "loss": 0.0811, + "step": 38150 + }, + { + "epoch": 0.3311690002690949, + "grad_norm": 0.31640625, + "learning_rate": 0.001402062174035757, + "loss": 0.0928, + "step": 38151 + }, + { + "epoch": 0.33117768074929904, + "grad_norm": 0.203125, + "learning_rate": 0.0014020341377611295, + "loss": 0.1113, + "step": 38152 + }, + { + "epoch": 0.33118636122950323, + "grad_norm": 0.4140625, + "learning_rate": 0.0014020061011562018, + "loss": 0.1016, + "step": 38153 + }, + { + "epoch": 0.33119504170970737, + "grad_norm": 0.2021484375, + "learning_rate": 0.0014019780642210045, + "loss": 0.0996, + "step": 38154 + }, + { + "epoch": 0.33120372218991156, + "grad_norm": 0.21875, + "learning_rate": 0.0014019500269555688, + "loss": 0.0791, + "step": 38155 + }, + { + "epoch": 0.3312124026701157, + "grad_norm": 0.306640625, + "learning_rate": 0.0014019219893599251, + "loss": 0.0952, + "step": 38156 + }, + { + "epoch": 0.3312210831503199, + "grad_norm": 0.10888671875, + "learning_rate": 0.0014018939514341038, + "loss": 0.1113, + "step": 38157 + }, + { + "epoch": 0.33122976363052403, + "grad_norm": 0.0927734375, + "learning_rate": 0.0014018659131781364, + "loss": 0.0854, + "step": 38158 + }, + { + "epoch": 0.3312384441107282, + "grad_norm": 3.078125, + "learning_rate": 0.0014018378745920522, + "loss": 0.3184, + "step": 38159 + }, + { + "epoch": 0.33124712459093236, + "grad_norm": 0.38671875, + "learning_rate": 0.001401809835675883, + "loss": 0.0972, + "step": 38160 + }, + { + "epoch": 0.33125580507113656, + "grad_norm": 0.408203125, + "learning_rate": 0.0014017817964296594, + "loss": 0.0938, + "step": 38161 + }, + { + "epoch": 0.3312644855513407, + "grad_norm": 0.07861328125, + "learning_rate": 0.0014017537568534115, + "loss": 0.0977, + "step": 38162 + }, + { + "epoch": 0.3312731660315449, + "grad_norm": 0.36328125, + "learning_rate": 0.00140172571694717, + "loss": 0.1455, + "step": 38163 + }, + { + "epoch": 0.331281846511749, + "grad_norm": 0.443359375, + "learning_rate": 0.001401697676710966, + "loss": 0.0806, + "step": 38164 + }, + { + "epoch": 0.3312905269919532, + "grad_norm": 0.16015625, + "learning_rate": 0.00140166963614483, + "loss": 0.0903, + "step": 38165 + }, + { + "epoch": 0.33129920747215735, + "grad_norm": 0.2177734375, + "learning_rate": 0.0014016415952487929, + "loss": 0.0942, + "step": 38166 + }, + { + "epoch": 0.33130788795236155, + "grad_norm": 0.625, + "learning_rate": 0.0014016135540228845, + "loss": 0.1289, + "step": 38167 + }, + { + "epoch": 0.3313165684325657, + "grad_norm": 0.06591796875, + "learning_rate": 0.0014015855124671361, + "loss": 0.082, + "step": 38168 + }, + { + "epoch": 0.3313252489127699, + "grad_norm": 0.83984375, + "learning_rate": 0.0014015574705815786, + "loss": 0.1016, + "step": 38169 + }, + { + "epoch": 0.331333929392974, + "grad_norm": 0.447265625, + "learning_rate": 0.0014015294283662425, + "loss": 0.0894, + "step": 38170 + }, + { + "epoch": 0.3313426098731782, + "grad_norm": 0.345703125, + "learning_rate": 0.001401501385821158, + "loss": 0.1206, + "step": 38171 + }, + { + "epoch": 0.33135129035338234, + "grad_norm": 0.162109375, + "learning_rate": 0.0014014733429463565, + "loss": 0.127, + "step": 38172 + }, + { + "epoch": 0.33135997083358654, + "grad_norm": 0.53515625, + "learning_rate": 0.001401445299741868, + "loss": 0.0894, + "step": 38173 + }, + { + "epoch": 0.3313686513137907, + "grad_norm": 0.11328125, + "learning_rate": 0.0014014172562077235, + "loss": 0.0947, + "step": 38174 + }, + { + "epoch": 0.33137733179399487, + "grad_norm": 0.12109375, + "learning_rate": 0.0014013892123439537, + "loss": 0.1211, + "step": 38175 + }, + { + "epoch": 0.331386012274199, + "grad_norm": 0.34765625, + "learning_rate": 0.001401361168150589, + "loss": 0.0703, + "step": 38176 + }, + { + "epoch": 0.3313946927544032, + "grad_norm": 0.1259765625, + "learning_rate": 0.0014013331236276602, + "loss": 0.0903, + "step": 38177 + }, + { + "epoch": 0.33140337323460733, + "grad_norm": 0.333984375, + "learning_rate": 0.0014013050787751982, + "loss": 0.126, + "step": 38178 + }, + { + "epoch": 0.3314120537148115, + "grad_norm": 0.061767578125, + "learning_rate": 0.0014012770335932335, + "loss": 0.0742, + "step": 38179 + }, + { + "epoch": 0.33142073419501566, + "grad_norm": 0.26953125, + "learning_rate": 0.0014012489880817966, + "loss": 0.0952, + "step": 38180 + }, + { + "epoch": 0.33142941467521986, + "grad_norm": 0.09326171875, + "learning_rate": 0.0014012209422409186, + "loss": 0.1035, + "step": 38181 + }, + { + "epoch": 0.331438095155424, + "grad_norm": 0.98828125, + "learning_rate": 0.0014011928960706297, + "loss": 0.1523, + "step": 38182 + }, + { + "epoch": 0.3314467756356282, + "grad_norm": 0.29296875, + "learning_rate": 0.001401164849570961, + "loss": 0.0977, + "step": 38183 + }, + { + "epoch": 0.3314554561158323, + "grad_norm": 0.341796875, + "learning_rate": 0.0014011368027419427, + "loss": 0.1162, + "step": 38184 + }, + { + "epoch": 0.3314641365960365, + "grad_norm": 0.359375, + "learning_rate": 0.001401108755583606, + "loss": 0.0864, + "step": 38185 + }, + { + "epoch": 0.33147281707624066, + "grad_norm": 1.3671875, + "learning_rate": 0.0014010807080959806, + "loss": 0.1768, + "step": 38186 + }, + { + "epoch": 0.33148149755644485, + "grad_norm": 0.10107421875, + "learning_rate": 0.0014010526602790988, + "loss": 0.1113, + "step": 38187 + }, + { + "epoch": 0.331490178036649, + "grad_norm": 0.2099609375, + "learning_rate": 0.0014010246121329898, + "loss": 0.0884, + "step": 38188 + }, + { + "epoch": 0.3314988585168532, + "grad_norm": 0.18359375, + "learning_rate": 0.0014009965636576848, + "loss": 0.1084, + "step": 38189 + }, + { + "epoch": 0.3315075389970573, + "grad_norm": 0.291015625, + "learning_rate": 0.0014009685148532146, + "loss": 0.1445, + "step": 38190 + }, + { + "epoch": 0.3315162194772615, + "grad_norm": 0.10986328125, + "learning_rate": 0.0014009404657196098, + "loss": 0.0894, + "step": 38191 + }, + { + "epoch": 0.33152489995746565, + "grad_norm": 0.09716796875, + "learning_rate": 0.0014009124162569012, + "loss": 0.0801, + "step": 38192 + }, + { + "epoch": 0.33153358043766984, + "grad_norm": 0.1923828125, + "learning_rate": 0.0014008843664651185, + "loss": 0.0928, + "step": 38193 + }, + { + "epoch": 0.331542260917874, + "grad_norm": 0.1357421875, + "learning_rate": 0.0014008563163442943, + "loss": 0.1064, + "step": 38194 + }, + { + "epoch": 0.3315509413980781, + "grad_norm": 0.220703125, + "learning_rate": 0.0014008282658944573, + "loss": 0.1211, + "step": 38195 + }, + { + "epoch": 0.3315596218782823, + "grad_norm": 0.2099609375, + "learning_rate": 0.0014008002151156393, + "loss": 0.082, + "step": 38196 + }, + { + "epoch": 0.33156830235848644, + "grad_norm": 0.35546875, + "learning_rate": 0.0014007721640078707, + "loss": 0.0898, + "step": 38197 + }, + { + "epoch": 0.33157698283869064, + "grad_norm": 0.4296875, + "learning_rate": 0.0014007441125711822, + "loss": 0.127, + "step": 38198 + }, + { + "epoch": 0.3315856633188948, + "grad_norm": 0.4609375, + "learning_rate": 0.0014007160608056045, + "loss": 0.1426, + "step": 38199 + }, + { + "epoch": 0.33159434379909897, + "grad_norm": 0.236328125, + "learning_rate": 0.0014006880087111683, + "loss": 0.0835, + "step": 38200 + }, + { + "epoch": 0.3316030242793031, + "grad_norm": 0.408203125, + "learning_rate": 0.0014006599562879042, + "loss": 0.1064, + "step": 38201 + }, + { + "epoch": 0.3316117047595073, + "grad_norm": 0.51953125, + "learning_rate": 0.0014006319035358426, + "loss": 0.1133, + "step": 38202 + }, + { + "epoch": 0.33162038523971143, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014006038504550146, + "loss": 0.1045, + "step": 38203 + }, + { + "epoch": 0.33162906571991563, + "grad_norm": 0.578125, + "learning_rate": 0.001400575797045451, + "loss": 0.0869, + "step": 38204 + }, + { + "epoch": 0.33163774620011977, + "grad_norm": 0.1787109375, + "learning_rate": 0.0014005477433071821, + "loss": 0.1406, + "step": 38205 + }, + { + "epoch": 0.33164642668032396, + "grad_norm": 0.08935546875, + "learning_rate": 0.0014005196892402383, + "loss": 0.0752, + "step": 38206 + }, + { + "epoch": 0.3316551071605281, + "grad_norm": 0.26171875, + "learning_rate": 0.001400491634844651, + "loss": 0.0986, + "step": 38207 + }, + { + "epoch": 0.3316637876407323, + "grad_norm": 0.220703125, + "learning_rate": 0.0014004635801204506, + "loss": 0.1011, + "step": 38208 + }, + { + "epoch": 0.3316724681209364, + "grad_norm": 0.4765625, + "learning_rate": 0.001400435525067668, + "loss": 0.1816, + "step": 38209 + }, + { + "epoch": 0.3316811486011406, + "grad_norm": 0.08984375, + "learning_rate": 0.0014004074696863332, + "loss": 0.1006, + "step": 38210 + }, + { + "epoch": 0.33168982908134476, + "grad_norm": 0.37109375, + "learning_rate": 0.0014003794139764777, + "loss": 0.1001, + "step": 38211 + }, + { + "epoch": 0.33169850956154895, + "grad_norm": 0.6875, + "learning_rate": 0.0014003513579381314, + "loss": 0.1621, + "step": 38212 + }, + { + "epoch": 0.3317071900417531, + "grad_norm": 1.3671875, + "learning_rate": 0.0014003233015713254, + "loss": 0.1504, + "step": 38213 + }, + { + "epoch": 0.3317158705219573, + "grad_norm": 0.08642578125, + "learning_rate": 0.0014002952448760903, + "loss": 0.0933, + "step": 38214 + }, + { + "epoch": 0.3317245510021614, + "grad_norm": 0.25, + "learning_rate": 0.0014002671878524573, + "loss": 0.1387, + "step": 38215 + }, + { + "epoch": 0.3317332314823656, + "grad_norm": 0.10400390625, + "learning_rate": 0.0014002391305004563, + "loss": 0.127, + "step": 38216 + }, + { + "epoch": 0.33174191196256975, + "grad_norm": 0.1162109375, + "learning_rate": 0.0014002110728201184, + "loss": 0.1562, + "step": 38217 + }, + { + "epoch": 0.33175059244277394, + "grad_norm": 0.29296875, + "learning_rate": 0.001400183014811474, + "loss": 0.1221, + "step": 38218 + }, + { + "epoch": 0.3317592729229781, + "grad_norm": 0.2392578125, + "learning_rate": 0.0014001549564745539, + "loss": 0.0962, + "step": 38219 + }, + { + "epoch": 0.33176795340318227, + "grad_norm": 0.1689453125, + "learning_rate": 0.0014001268978093891, + "loss": 0.1299, + "step": 38220 + }, + { + "epoch": 0.3317766338833864, + "grad_norm": 0.11669921875, + "learning_rate": 0.00140009883881601, + "loss": 0.0728, + "step": 38221 + }, + { + "epoch": 0.3317853143635906, + "grad_norm": 0.56640625, + "learning_rate": 0.001400070779494447, + "loss": 0.1309, + "step": 38222 + }, + { + "epoch": 0.33179399484379474, + "grad_norm": 0.1103515625, + "learning_rate": 0.0014000427198447313, + "loss": 0.1133, + "step": 38223 + }, + { + "epoch": 0.33180267532399893, + "grad_norm": 0.28125, + "learning_rate": 0.0014000146598668938, + "loss": 0.1279, + "step": 38224 + }, + { + "epoch": 0.33181135580420307, + "grad_norm": 0.05029296875, + "learning_rate": 0.0013999865995609642, + "loss": 0.054, + "step": 38225 + }, + { + "epoch": 0.33182003628440726, + "grad_norm": 0.36328125, + "learning_rate": 0.0013999585389269742, + "loss": 0.1162, + "step": 38226 + }, + { + "epoch": 0.3318287167646114, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013999304779649538, + "loss": 0.1074, + "step": 38227 + }, + { + "epoch": 0.3318373972448156, + "grad_norm": 0.08642578125, + "learning_rate": 0.0013999024166749337, + "loss": 0.1191, + "step": 38228 + }, + { + "epoch": 0.33184607772501973, + "grad_norm": 0.70703125, + "learning_rate": 0.0013998743550569452, + "loss": 0.1055, + "step": 38229 + }, + { + "epoch": 0.3318547582052239, + "grad_norm": 0.154296875, + "learning_rate": 0.0013998462931110185, + "loss": 0.1279, + "step": 38230 + }, + { + "epoch": 0.33186343868542806, + "grad_norm": 0.25390625, + "learning_rate": 0.0013998182308371846, + "loss": 0.165, + "step": 38231 + }, + { + "epoch": 0.33187211916563225, + "grad_norm": 0.09375, + "learning_rate": 0.0013997901682354734, + "loss": 0.0771, + "step": 38232 + }, + { + "epoch": 0.3318807996458364, + "grad_norm": 0.271484375, + "learning_rate": 0.001399762105305917, + "loss": 0.1172, + "step": 38233 + }, + { + "epoch": 0.3318894801260406, + "grad_norm": 0.59375, + "learning_rate": 0.0013997340420485447, + "loss": 0.1602, + "step": 38234 + }, + { + "epoch": 0.3318981606062447, + "grad_norm": 0.2236328125, + "learning_rate": 0.0013997059784633876, + "loss": 0.1016, + "step": 38235 + }, + { + "epoch": 0.3319068410864489, + "grad_norm": 0.09228515625, + "learning_rate": 0.001399677914550477, + "loss": 0.1016, + "step": 38236 + }, + { + "epoch": 0.33191552156665305, + "grad_norm": 0.12890625, + "learning_rate": 0.001399649850309843, + "loss": 0.0991, + "step": 38237 + }, + { + "epoch": 0.33192420204685724, + "grad_norm": 0.271484375, + "learning_rate": 0.0013996217857415163, + "loss": 0.0864, + "step": 38238 + }, + { + "epoch": 0.3319328825270614, + "grad_norm": 0.86328125, + "learning_rate": 0.001399593720845528, + "loss": 0.1113, + "step": 38239 + }, + { + "epoch": 0.33194156300726557, + "grad_norm": 0.330078125, + "learning_rate": 0.0013995656556219083, + "loss": 0.1318, + "step": 38240 + }, + { + "epoch": 0.3319502434874697, + "grad_norm": 0.3046875, + "learning_rate": 0.001399537590070688, + "loss": 0.0884, + "step": 38241 + }, + { + "epoch": 0.3319589239676739, + "grad_norm": 0.4140625, + "learning_rate": 0.001399509524191898, + "loss": 0.0894, + "step": 38242 + }, + { + "epoch": 0.33196760444787804, + "grad_norm": 0.248046875, + "learning_rate": 0.0013994814579855691, + "loss": 0.0952, + "step": 38243 + }, + { + "epoch": 0.33197628492808223, + "grad_norm": 0.5078125, + "learning_rate": 0.0013994533914517314, + "loss": 0.1035, + "step": 38244 + }, + { + "epoch": 0.33198496540828637, + "grad_norm": 0.373046875, + "learning_rate": 0.001399425324590416, + "loss": 0.1709, + "step": 38245 + }, + { + "epoch": 0.33199364588849056, + "grad_norm": 0.26171875, + "learning_rate": 0.0013993972574016537, + "loss": 0.1025, + "step": 38246 + }, + { + "epoch": 0.3320023263686947, + "grad_norm": 0.1416015625, + "learning_rate": 0.0013993691898854753, + "loss": 0.1084, + "step": 38247 + }, + { + "epoch": 0.3320110068488989, + "grad_norm": 0.0888671875, + "learning_rate": 0.001399341122041911, + "loss": 0.0874, + "step": 38248 + }, + { + "epoch": 0.33201968732910303, + "grad_norm": 0.74609375, + "learning_rate": 0.0013993130538709918, + "loss": 0.126, + "step": 38249 + }, + { + "epoch": 0.3320283678093072, + "grad_norm": 0.255859375, + "learning_rate": 0.0013992849853727482, + "loss": 0.1226, + "step": 38250 + }, + { + "epoch": 0.33203704828951136, + "grad_norm": 0.490234375, + "learning_rate": 0.0013992569165472112, + "loss": 0.1201, + "step": 38251 + }, + { + "epoch": 0.33204572876971555, + "grad_norm": 0.3671875, + "learning_rate": 0.0013992288473944113, + "loss": 0.0918, + "step": 38252 + }, + { + "epoch": 0.3320544092499197, + "grad_norm": 0.1162109375, + "learning_rate": 0.001399200777914379, + "loss": 0.0898, + "step": 38253 + }, + { + "epoch": 0.3320630897301239, + "grad_norm": 0.91796875, + "learning_rate": 0.0013991727081071456, + "loss": 0.1245, + "step": 38254 + }, + { + "epoch": 0.332071770210328, + "grad_norm": 0.39453125, + "learning_rate": 0.0013991446379727412, + "loss": 0.1045, + "step": 38255 + }, + { + "epoch": 0.3320804506905322, + "grad_norm": 0.1640625, + "learning_rate": 0.0013991165675111968, + "loss": 0.1055, + "step": 38256 + }, + { + "epoch": 0.33208913117073635, + "grad_norm": 0.09619140625, + "learning_rate": 0.001399088496722543, + "loss": 0.0864, + "step": 38257 + }, + { + "epoch": 0.33209781165094054, + "grad_norm": 0.255859375, + "learning_rate": 0.0013990604256068104, + "loss": 0.1182, + "step": 38258 + }, + { + "epoch": 0.3321064921311447, + "grad_norm": 0.083984375, + "learning_rate": 0.00139903235416403, + "loss": 0.0898, + "step": 38259 + }, + { + "epoch": 0.3321151726113489, + "grad_norm": 0.5703125, + "learning_rate": 0.0013990042823942323, + "loss": 0.1738, + "step": 38260 + }, + { + "epoch": 0.332123853091553, + "grad_norm": 0.169921875, + "learning_rate": 0.001398976210297448, + "loss": 0.0889, + "step": 38261 + }, + { + "epoch": 0.3321325335717572, + "grad_norm": 0.41015625, + "learning_rate": 0.0013989481378737074, + "loss": 0.1719, + "step": 38262 + }, + { + "epoch": 0.33214121405196134, + "grad_norm": 0.1513671875, + "learning_rate": 0.001398920065123042, + "loss": 0.1182, + "step": 38263 + }, + { + "epoch": 0.33214989453216553, + "grad_norm": 0.1748046875, + "learning_rate": 0.0013988919920454822, + "loss": 0.106, + "step": 38264 + }, + { + "epoch": 0.3321585750123697, + "grad_norm": 0.078125, + "learning_rate": 0.0013988639186410584, + "loss": 0.1084, + "step": 38265 + }, + { + "epoch": 0.33216725549257387, + "grad_norm": 0.671875, + "learning_rate": 0.0013988358449098016, + "loss": 0.1094, + "step": 38266 + }, + { + "epoch": 0.332175935972778, + "grad_norm": 0.421875, + "learning_rate": 0.0013988077708517424, + "loss": 0.1035, + "step": 38267 + }, + { + "epoch": 0.3321846164529822, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013987796964669114, + "loss": 0.1201, + "step": 38268 + }, + { + "epoch": 0.33219329693318633, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013987516217553396, + "loss": 0.0718, + "step": 38269 + }, + { + "epoch": 0.3322019774133905, + "grad_norm": 0.2734375, + "learning_rate": 0.0013987235467170574, + "loss": 0.1069, + "step": 38270 + }, + { + "epoch": 0.33221065789359466, + "grad_norm": 0.6328125, + "learning_rate": 0.0013986954713520955, + "loss": 0.1162, + "step": 38271 + }, + { + "epoch": 0.33221933837379886, + "grad_norm": 0.267578125, + "learning_rate": 0.001398667395660485, + "loss": 0.0942, + "step": 38272 + }, + { + "epoch": 0.332228018854003, + "grad_norm": 0.25, + "learning_rate": 0.001398639319642256, + "loss": 0.0649, + "step": 38273 + }, + { + "epoch": 0.3322366993342072, + "grad_norm": 0.1982421875, + "learning_rate": 0.0013986112432974398, + "loss": 0.0986, + "step": 38274 + }, + { + "epoch": 0.3322453798144113, + "grad_norm": 0.5546875, + "learning_rate": 0.0013985831666260666, + "loss": 0.1152, + "step": 38275 + }, + { + "epoch": 0.3322540602946155, + "grad_norm": 0.30078125, + "learning_rate": 0.0013985550896281676, + "loss": 0.1045, + "step": 38276 + }, + { + "epoch": 0.33226274077481965, + "grad_norm": 0.322265625, + "learning_rate": 0.001398527012303773, + "loss": 0.0986, + "step": 38277 + }, + { + "epoch": 0.33227142125502385, + "grad_norm": 0.51171875, + "learning_rate": 0.0013984989346529143, + "loss": 0.1484, + "step": 38278 + }, + { + "epoch": 0.332280101735228, + "grad_norm": 0.88671875, + "learning_rate": 0.001398470856675621, + "loss": 0.0923, + "step": 38279 + }, + { + "epoch": 0.3322887822154322, + "grad_norm": 0.10205078125, + "learning_rate": 0.0013984427783719244, + "loss": 0.1084, + "step": 38280 + }, + { + "epoch": 0.3322974626956363, + "grad_norm": 0.3359375, + "learning_rate": 0.0013984146997418558, + "loss": 0.1621, + "step": 38281 + }, + { + "epoch": 0.3323061431758405, + "grad_norm": 0.11181640625, + "learning_rate": 0.001398386620785445, + "loss": 0.0806, + "step": 38282 + }, + { + "epoch": 0.33231482365604464, + "grad_norm": 1.1875, + "learning_rate": 0.0013983585415027232, + "loss": 0.1973, + "step": 38283 + }, + { + "epoch": 0.33232350413624884, + "grad_norm": 0.310546875, + "learning_rate": 0.001398330461893721, + "loss": 0.1045, + "step": 38284 + }, + { + "epoch": 0.332332184616453, + "grad_norm": 0.177734375, + "learning_rate": 0.001398302381958469, + "loss": 0.1045, + "step": 38285 + }, + { + "epoch": 0.33234086509665717, + "grad_norm": 0.3828125, + "learning_rate": 0.001398274301696998, + "loss": 0.1387, + "step": 38286 + }, + { + "epoch": 0.3323495455768613, + "grad_norm": 0.94140625, + "learning_rate": 0.0013982462211093386, + "loss": 0.0879, + "step": 38287 + }, + { + "epoch": 0.3323582260570655, + "grad_norm": 0.59765625, + "learning_rate": 0.0013982181401955218, + "loss": 0.167, + "step": 38288 + }, + { + "epoch": 0.33236690653726964, + "grad_norm": 0.1962890625, + "learning_rate": 0.001398190058955578, + "loss": 0.1001, + "step": 38289 + }, + { + "epoch": 0.33237558701747383, + "grad_norm": 0.7578125, + "learning_rate": 0.0013981619773895384, + "loss": 0.1045, + "step": 38290 + }, + { + "epoch": 0.33238426749767797, + "grad_norm": 0.376953125, + "learning_rate": 0.001398133895497433, + "loss": 0.1147, + "step": 38291 + }, + { + "epoch": 0.33239294797788216, + "grad_norm": 0.68359375, + "learning_rate": 0.0013981058132792928, + "loss": 0.0923, + "step": 38292 + }, + { + "epoch": 0.3324016284580863, + "grad_norm": 0.2197265625, + "learning_rate": 0.0013980777307351483, + "loss": 0.1006, + "step": 38293 + }, + { + "epoch": 0.3324103089382905, + "grad_norm": 0.07373046875, + "learning_rate": 0.001398049647865031, + "loss": 0.0996, + "step": 38294 + }, + { + "epoch": 0.3324189894184946, + "grad_norm": 0.34765625, + "learning_rate": 0.001398021564668971, + "loss": 0.0977, + "step": 38295 + }, + { + "epoch": 0.3324276698986988, + "grad_norm": 0.328125, + "learning_rate": 0.0013979934811469991, + "loss": 0.1396, + "step": 38296 + }, + { + "epoch": 0.33243635037890296, + "grad_norm": 0.40234375, + "learning_rate": 0.001397965397299146, + "loss": 0.0938, + "step": 38297 + }, + { + "epoch": 0.33244503085910715, + "grad_norm": 0.296875, + "learning_rate": 0.0013979373131254422, + "loss": 0.0918, + "step": 38298 + }, + { + "epoch": 0.3324537113393113, + "grad_norm": 0.734375, + "learning_rate": 0.001397909228625919, + "loss": 0.0913, + "step": 38299 + }, + { + "epoch": 0.3324623918195155, + "grad_norm": 0.6796875, + "learning_rate": 0.0013978811438006065, + "loss": 0.0986, + "step": 38300 + }, + { + "epoch": 0.3324710722997196, + "grad_norm": 0.68359375, + "learning_rate": 0.0013978530586495354, + "loss": 0.1504, + "step": 38301 + }, + { + "epoch": 0.3324797527799238, + "grad_norm": 0.484375, + "learning_rate": 0.001397824973172737, + "loss": 0.1211, + "step": 38302 + }, + { + "epoch": 0.33248843326012795, + "grad_norm": 0.66796875, + "learning_rate": 0.0013977968873702417, + "loss": 0.1113, + "step": 38303 + }, + { + "epoch": 0.33249711374033214, + "grad_norm": 0.5703125, + "learning_rate": 0.0013977688012420802, + "loss": 0.1289, + "step": 38304 + }, + { + "epoch": 0.3325057942205363, + "grad_norm": 0.29296875, + "learning_rate": 0.0013977407147882828, + "loss": 0.0957, + "step": 38305 + }, + { + "epoch": 0.33251447470074047, + "grad_norm": 0.578125, + "learning_rate": 0.0013977126280088812, + "loss": 0.0771, + "step": 38306 + }, + { + "epoch": 0.3325231551809446, + "grad_norm": 0.1572265625, + "learning_rate": 0.0013976845409039054, + "loss": 0.1079, + "step": 38307 + }, + { + "epoch": 0.3325318356611488, + "grad_norm": 0.0908203125, + "learning_rate": 0.0013976564534733862, + "loss": 0.0815, + "step": 38308 + }, + { + "epoch": 0.33254051614135294, + "grad_norm": 0.2265625, + "learning_rate": 0.0013976283657173543, + "loss": 0.1172, + "step": 38309 + }, + { + "epoch": 0.33254919662155713, + "grad_norm": 0.1357421875, + "learning_rate": 0.0013976002776358405, + "loss": 0.0952, + "step": 38310 + }, + { + "epoch": 0.33255787710176127, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013975721892288753, + "loss": 0.082, + "step": 38311 + }, + { + "epoch": 0.33256655758196546, + "grad_norm": 0.33203125, + "learning_rate": 0.00139754410049649, + "loss": 0.0869, + "step": 38312 + }, + { + "epoch": 0.3325752380621696, + "grad_norm": 0.279296875, + "learning_rate": 0.001397516011438715, + "loss": 0.1094, + "step": 38313 + }, + { + "epoch": 0.3325839185423738, + "grad_norm": 0.103515625, + "learning_rate": 0.0013974879220555805, + "loss": 0.0845, + "step": 38314 + }, + { + "epoch": 0.33259259902257793, + "grad_norm": 0.27734375, + "learning_rate": 0.001397459832347118, + "loss": 0.1348, + "step": 38315 + }, + { + "epoch": 0.3326012795027821, + "grad_norm": 0.12255859375, + "learning_rate": 0.001397431742313358, + "loss": 0.1055, + "step": 38316 + }, + { + "epoch": 0.33260995998298626, + "grad_norm": 0.212890625, + "learning_rate": 0.001397403651954331, + "loss": 0.1182, + "step": 38317 + }, + { + "epoch": 0.33261864046319045, + "grad_norm": 0.27734375, + "learning_rate": 0.0013973755612700677, + "loss": 0.0996, + "step": 38318 + }, + { + "epoch": 0.3326273209433946, + "grad_norm": 0.14453125, + "learning_rate": 0.001397347470260599, + "loss": 0.1025, + "step": 38319 + }, + { + "epoch": 0.3326360014235987, + "grad_norm": 1.4765625, + "learning_rate": 0.0013973193789259557, + "loss": 0.1387, + "step": 38320 + }, + { + "epoch": 0.3326446819038029, + "grad_norm": 0.1875, + "learning_rate": 0.0013972912872661683, + "loss": 0.1055, + "step": 38321 + }, + { + "epoch": 0.33265336238400706, + "grad_norm": 0.55078125, + "learning_rate": 0.0013972631952812677, + "loss": 0.0986, + "step": 38322 + }, + { + "epoch": 0.33266204286421125, + "grad_norm": 0.06982421875, + "learning_rate": 0.0013972351029712844, + "loss": 0.0918, + "step": 38323 + }, + { + "epoch": 0.3326707233444154, + "grad_norm": 0.453125, + "learning_rate": 0.0013972070103362494, + "loss": 0.1113, + "step": 38324 + }, + { + "epoch": 0.3326794038246196, + "grad_norm": 0.10302734375, + "learning_rate": 0.001397178917376193, + "loss": 0.127, + "step": 38325 + }, + { + "epoch": 0.3326880843048237, + "grad_norm": 0.10302734375, + "learning_rate": 0.0013971508240911467, + "loss": 0.0669, + "step": 38326 + }, + { + "epoch": 0.3326967647850279, + "grad_norm": 0.32421875, + "learning_rate": 0.00139712273048114, + "loss": 0.0947, + "step": 38327 + }, + { + "epoch": 0.33270544526523205, + "grad_norm": 0.296875, + "learning_rate": 0.001397094636546205, + "loss": 0.1021, + "step": 38328 + }, + { + "epoch": 0.33271412574543624, + "grad_norm": 0.16796875, + "learning_rate": 0.0013970665422863716, + "loss": 0.1221, + "step": 38329 + }, + { + "epoch": 0.3327228062256404, + "grad_norm": 0.126953125, + "learning_rate": 0.0013970384477016704, + "loss": 0.127, + "step": 38330 + }, + { + "epoch": 0.33273148670584457, + "grad_norm": 0.76171875, + "learning_rate": 0.0013970103527921327, + "loss": 0.0952, + "step": 38331 + }, + { + "epoch": 0.3327401671860487, + "grad_norm": 0.2578125, + "learning_rate": 0.0013969822575577889, + "loss": 0.1494, + "step": 38332 + }, + { + "epoch": 0.3327488476662529, + "grad_norm": 0.146484375, + "learning_rate": 0.0013969541619986699, + "loss": 0.1572, + "step": 38333 + }, + { + "epoch": 0.33275752814645704, + "grad_norm": 0.0927734375, + "learning_rate": 0.001396926066114806, + "loss": 0.0674, + "step": 38334 + }, + { + "epoch": 0.33276620862666123, + "grad_norm": 0.765625, + "learning_rate": 0.0013968979699062285, + "loss": 0.0923, + "step": 38335 + }, + { + "epoch": 0.33277488910686537, + "grad_norm": 0.66015625, + "learning_rate": 0.0013968698733729677, + "loss": 0.1357, + "step": 38336 + }, + { + "epoch": 0.33278356958706956, + "grad_norm": 0.12353515625, + "learning_rate": 0.0013968417765150545, + "loss": 0.1494, + "step": 38337 + }, + { + "epoch": 0.3327922500672737, + "grad_norm": 0.1708984375, + "learning_rate": 0.0013968136793325195, + "loss": 0.0703, + "step": 38338 + }, + { + "epoch": 0.3328009305474779, + "grad_norm": 0.365234375, + "learning_rate": 0.0013967855818253936, + "loss": 0.0811, + "step": 38339 + }, + { + "epoch": 0.33280961102768203, + "grad_norm": 0.11572265625, + "learning_rate": 0.0013967574839937073, + "loss": 0.0933, + "step": 38340 + }, + { + "epoch": 0.3328182915078862, + "grad_norm": 0.36328125, + "learning_rate": 0.0013967293858374917, + "loss": 0.082, + "step": 38341 + }, + { + "epoch": 0.33282697198809036, + "grad_norm": 0.4453125, + "learning_rate": 0.0013967012873567772, + "loss": 0.1191, + "step": 38342 + }, + { + "epoch": 0.33283565246829455, + "grad_norm": 0.59765625, + "learning_rate": 0.0013966731885515947, + "loss": 0.0923, + "step": 38343 + }, + { + "epoch": 0.3328443329484987, + "grad_norm": 0.88671875, + "learning_rate": 0.0013966450894219746, + "loss": 0.1621, + "step": 38344 + }, + { + "epoch": 0.3328530134287029, + "grad_norm": 0.115234375, + "learning_rate": 0.0013966169899679482, + "loss": 0.1104, + "step": 38345 + }, + { + "epoch": 0.332861693908907, + "grad_norm": 0.201171875, + "learning_rate": 0.0013965888901895458, + "loss": 0.1104, + "step": 38346 + }, + { + "epoch": 0.3328703743891112, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013965607900867985, + "loss": 0.0835, + "step": 38347 + }, + { + "epoch": 0.33287905486931535, + "grad_norm": 0.2392578125, + "learning_rate": 0.0013965326896597362, + "loss": 0.0649, + "step": 38348 + }, + { + "epoch": 0.33288773534951954, + "grad_norm": 0.4609375, + "learning_rate": 0.0013965045889083908, + "loss": 0.1348, + "step": 38349 + }, + { + "epoch": 0.3328964158297237, + "grad_norm": 0.0791015625, + "learning_rate": 0.0013964764878327918, + "loss": 0.0889, + "step": 38350 + }, + { + "epoch": 0.3329050963099279, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013964483864329711, + "loss": 0.106, + "step": 38351 + }, + { + "epoch": 0.332913776790132, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013964202847089587, + "loss": 0.1104, + "step": 38352 + }, + { + "epoch": 0.3329224572703362, + "grad_norm": 0.33984375, + "learning_rate": 0.0013963921826607856, + "loss": 0.1152, + "step": 38353 + }, + { + "epoch": 0.33293113775054034, + "grad_norm": 0.10205078125, + "learning_rate": 0.0013963640802884822, + "loss": 0.0815, + "step": 38354 + }, + { + "epoch": 0.33293981823074453, + "grad_norm": 0.1689453125, + "learning_rate": 0.00139633597759208, + "loss": 0.1167, + "step": 38355 + }, + { + "epoch": 0.33294849871094867, + "grad_norm": 0.490234375, + "learning_rate": 0.0013963078745716092, + "loss": 0.0947, + "step": 38356 + }, + { + "epoch": 0.33295717919115286, + "grad_norm": 0.58984375, + "learning_rate": 0.0013962797712271, + "loss": 0.1245, + "step": 38357 + }, + { + "epoch": 0.332965859671357, + "grad_norm": 0.51953125, + "learning_rate": 0.001396251667558584, + "loss": 0.1006, + "step": 38358 + }, + { + "epoch": 0.3329745401515612, + "grad_norm": 0.1796875, + "learning_rate": 0.0013962235635660916, + "loss": 0.0791, + "step": 38359 + }, + { + "epoch": 0.33298322063176533, + "grad_norm": 0.4453125, + "learning_rate": 0.0013961954592496535, + "loss": 0.0752, + "step": 38360 + }, + { + "epoch": 0.3329919011119695, + "grad_norm": 0.09130859375, + "learning_rate": 0.0013961673546093007, + "loss": 0.1079, + "step": 38361 + }, + { + "epoch": 0.33300058159217366, + "grad_norm": 0.09130859375, + "learning_rate": 0.0013961392496450635, + "loss": 0.0869, + "step": 38362 + }, + { + "epoch": 0.33300926207237785, + "grad_norm": 0.2431640625, + "learning_rate": 0.001396111144356973, + "loss": 0.0996, + "step": 38363 + }, + { + "epoch": 0.333017942552582, + "grad_norm": 0.169921875, + "learning_rate": 0.00139608303874506, + "loss": 0.1025, + "step": 38364 + }, + { + "epoch": 0.3330266230327862, + "grad_norm": 0.86328125, + "learning_rate": 0.0013960549328093548, + "loss": 0.1426, + "step": 38365 + }, + { + "epoch": 0.3330353035129903, + "grad_norm": 0.345703125, + "learning_rate": 0.0013960268265498883, + "loss": 0.0791, + "step": 38366 + }, + { + "epoch": 0.3330439839931945, + "grad_norm": 0.236328125, + "learning_rate": 0.0013959987199666914, + "loss": 0.1006, + "step": 38367 + }, + { + "epoch": 0.33305266447339865, + "grad_norm": 0.3828125, + "learning_rate": 0.0013959706130597948, + "loss": 0.1045, + "step": 38368 + }, + { + "epoch": 0.33306134495360284, + "grad_norm": 0.4921875, + "learning_rate": 0.001395942505829229, + "loss": 0.0879, + "step": 38369 + }, + { + "epoch": 0.333070025433807, + "grad_norm": 0.1162109375, + "learning_rate": 0.0013959143982750248, + "loss": 0.0742, + "step": 38370 + }, + { + "epoch": 0.3330787059140112, + "grad_norm": 0.474609375, + "learning_rate": 0.0013958862903972135, + "loss": 0.1465, + "step": 38371 + }, + { + "epoch": 0.3330873863942153, + "grad_norm": 0.55078125, + "learning_rate": 0.001395858182195825, + "loss": 0.1084, + "step": 38372 + }, + { + "epoch": 0.3330960668744195, + "grad_norm": 1.1875, + "learning_rate": 0.001395830073670891, + "loss": 0.1348, + "step": 38373 + }, + { + "epoch": 0.33310474735462364, + "grad_norm": 0.07373046875, + "learning_rate": 0.0013958019648224412, + "loss": 0.0693, + "step": 38374 + }, + { + "epoch": 0.33311342783482784, + "grad_norm": 0.263671875, + "learning_rate": 0.0013957738556505072, + "loss": 0.1387, + "step": 38375 + }, + { + "epoch": 0.333122108315032, + "grad_norm": 0.302734375, + "learning_rate": 0.001395745746155119, + "loss": 0.1001, + "step": 38376 + }, + { + "epoch": 0.33313078879523617, + "grad_norm": 0.25390625, + "learning_rate": 0.0013957176363363077, + "loss": 0.0884, + "step": 38377 + }, + { + "epoch": 0.3331394692754403, + "grad_norm": 0.12890625, + "learning_rate": 0.0013956895261941042, + "loss": 0.0933, + "step": 38378 + }, + { + "epoch": 0.3331481497556445, + "grad_norm": 0.11083984375, + "learning_rate": 0.001395661415728539, + "loss": 0.0791, + "step": 38379 + }, + { + "epoch": 0.33315683023584863, + "grad_norm": 0.10986328125, + "learning_rate": 0.0013956333049396428, + "loss": 0.1025, + "step": 38380 + }, + { + "epoch": 0.3331655107160528, + "grad_norm": 0.408203125, + "learning_rate": 0.0013956051938274466, + "loss": 0.0918, + "step": 38381 + }, + { + "epoch": 0.33317419119625696, + "grad_norm": 0.54296875, + "learning_rate": 0.001395577082391981, + "loss": 0.1006, + "step": 38382 + }, + { + "epoch": 0.33318287167646116, + "grad_norm": 0.7578125, + "learning_rate": 0.0013955489706332766, + "loss": 0.1182, + "step": 38383 + }, + { + "epoch": 0.3331915521566653, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013955208585513645, + "loss": 0.0684, + "step": 38384 + }, + { + "epoch": 0.3332002326368695, + "grad_norm": 0.7109375, + "learning_rate": 0.0013954927461462753, + "loss": 0.106, + "step": 38385 + }, + { + "epoch": 0.3332089131170736, + "grad_norm": 0.408203125, + "learning_rate": 0.0013954646334180397, + "loss": 0.0957, + "step": 38386 + }, + { + "epoch": 0.3332175935972778, + "grad_norm": 0.130859375, + "learning_rate": 0.0013954365203666878, + "loss": 0.0957, + "step": 38387 + }, + { + "epoch": 0.33322627407748195, + "grad_norm": 1.0625, + "learning_rate": 0.0013954084069922514, + "loss": 0.167, + "step": 38388 + }, + { + "epoch": 0.33323495455768615, + "grad_norm": 0.1083984375, + "learning_rate": 0.001395380293294761, + "loss": 0.1182, + "step": 38389 + }, + { + "epoch": 0.3332436350378903, + "grad_norm": 0.0947265625, + "learning_rate": 0.001395352179274247, + "loss": 0.0908, + "step": 38390 + }, + { + "epoch": 0.3332523155180945, + "grad_norm": 0.1005859375, + "learning_rate": 0.0013953240649307403, + "loss": 0.0908, + "step": 38391 + }, + { + "epoch": 0.3332609959982986, + "grad_norm": 0.458984375, + "learning_rate": 0.0013952959502642715, + "loss": 0.0562, + "step": 38392 + }, + { + "epoch": 0.3332696764785028, + "grad_norm": 0.251953125, + "learning_rate": 0.0013952678352748716, + "loss": 0.0781, + "step": 38393 + }, + { + "epoch": 0.33327835695870694, + "grad_norm": 0.48046875, + "learning_rate": 0.0013952397199625713, + "loss": 0.0791, + "step": 38394 + }, + { + "epoch": 0.33328703743891114, + "grad_norm": 0.40234375, + "learning_rate": 0.001395211604327401, + "loss": 0.0986, + "step": 38395 + }, + { + "epoch": 0.3332957179191153, + "grad_norm": 0.1591796875, + "learning_rate": 0.001395183488369392, + "loss": 0.0908, + "step": 38396 + }, + { + "epoch": 0.33330439839931947, + "grad_norm": 0.6015625, + "learning_rate": 0.0013951553720885747, + "loss": 0.0801, + "step": 38397 + }, + { + "epoch": 0.3333130788795236, + "grad_norm": 0.384765625, + "learning_rate": 0.0013951272554849799, + "loss": 0.1328, + "step": 38398 + }, + { + "epoch": 0.3333217593597278, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013950991385586382, + "loss": 0.0923, + "step": 38399 + }, + { + "epoch": 0.33333043983993194, + "grad_norm": 0.345703125, + "learning_rate": 0.0013950710213095807, + "loss": 0.125, + "step": 38400 + }, + { + "epoch": 0.33333912032013613, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013950429037378378, + "loss": 0.123, + "step": 38401 + }, + { + "epoch": 0.33334780080034027, + "grad_norm": 0.59375, + "learning_rate": 0.0013950147858434409, + "loss": 0.0869, + "step": 38402 + }, + { + "epoch": 0.33335648128054446, + "grad_norm": 0.546875, + "learning_rate": 0.0013949866676264198, + "loss": 0.0918, + "step": 38403 + }, + { + "epoch": 0.3333651617607486, + "grad_norm": 0.0751953125, + "learning_rate": 0.001394958549086806, + "loss": 0.0854, + "step": 38404 + }, + { + "epoch": 0.3333738422409528, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013949304302246297, + "loss": 0.1055, + "step": 38405 + }, + { + "epoch": 0.3333825227211569, + "grad_norm": 0.4765625, + "learning_rate": 0.0013949023110399218, + "loss": 0.1074, + "step": 38406 + }, + { + "epoch": 0.3333912032013611, + "grad_norm": 0.76171875, + "learning_rate": 0.0013948741915327135, + "loss": 0.1035, + "step": 38407 + }, + { + "epoch": 0.33339988368156526, + "grad_norm": 0.431640625, + "learning_rate": 0.0013948460717030352, + "loss": 0.1123, + "step": 38408 + }, + { + "epoch": 0.33340856416176945, + "grad_norm": 0.150390625, + "learning_rate": 0.001394817951550917, + "loss": 0.082, + "step": 38409 + }, + { + "epoch": 0.3334172446419736, + "grad_norm": 0.412109375, + "learning_rate": 0.0013947898310763913, + "loss": 0.1094, + "step": 38410 + }, + { + "epoch": 0.3334259251221778, + "grad_norm": 0.4296875, + "learning_rate": 0.0013947617102794874, + "loss": 0.1299, + "step": 38411 + }, + { + "epoch": 0.3334346056023819, + "grad_norm": 0.2734375, + "learning_rate": 0.0013947335891602367, + "loss": 0.0801, + "step": 38412 + }, + { + "epoch": 0.3334432860825861, + "grad_norm": 0.1474609375, + "learning_rate": 0.0013947054677186696, + "loss": 0.1152, + "step": 38413 + }, + { + "epoch": 0.33345196656279025, + "grad_norm": 0.1611328125, + "learning_rate": 0.001394677345954817, + "loss": 0.083, + "step": 38414 + }, + { + "epoch": 0.33346064704299444, + "grad_norm": 0.23046875, + "learning_rate": 0.0013946492238687098, + "loss": 0.0957, + "step": 38415 + }, + { + "epoch": 0.3334693275231986, + "grad_norm": 1.2265625, + "learning_rate": 0.0013946211014603785, + "loss": 0.1836, + "step": 38416 + }, + { + "epoch": 0.33347800800340277, + "grad_norm": 0.248046875, + "learning_rate": 0.0013945929787298542, + "loss": 0.123, + "step": 38417 + }, + { + "epoch": 0.3334866884836069, + "grad_norm": 0.3984375, + "learning_rate": 0.001394564855677167, + "loss": 0.1377, + "step": 38418 + }, + { + "epoch": 0.3334953689638111, + "grad_norm": 0.279296875, + "learning_rate": 0.0013945367323023486, + "loss": 0.0889, + "step": 38419 + }, + { + "epoch": 0.33350404944401524, + "grad_norm": 0.45703125, + "learning_rate": 0.001394508608605429, + "loss": 0.1221, + "step": 38420 + }, + { + "epoch": 0.33351272992421943, + "grad_norm": 0.46484375, + "learning_rate": 0.0013944804845864394, + "loss": 0.0913, + "step": 38421 + }, + { + "epoch": 0.33352141040442357, + "grad_norm": 0.15625, + "learning_rate": 0.00139445236024541, + "loss": 0.1084, + "step": 38422 + }, + { + "epoch": 0.33353009088462776, + "grad_norm": 0.279296875, + "learning_rate": 0.0013944242355823724, + "loss": 0.0811, + "step": 38423 + }, + { + "epoch": 0.3335387713648319, + "grad_norm": 0.298828125, + "learning_rate": 0.0013943961105973566, + "loss": 0.1069, + "step": 38424 + }, + { + "epoch": 0.3335474518450361, + "grad_norm": 0.65625, + "learning_rate": 0.0013943679852903939, + "loss": 0.0996, + "step": 38425 + }, + { + "epoch": 0.33355613232524023, + "grad_norm": 1.0546875, + "learning_rate": 0.0013943398596615144, + "loss": 0.1094, + "step": 38426 + }, + { + "epoch": 0.3335648128054444, + "grad_norm": 0.1513671875, + "learning_rate": 0.0013943117337107493, + "loss": 0.1338, + "step": 38427 + }, + { + "epoch": 0.33357349328564856, + "grad_norm": 0.26171875, + "learning_rate": 0.0013942836074381298, + "loss": 0.0996, + "step": 38428 + }, + { + "epoch": 0.33358217376585275, + "grad_norm": 0.16796875, + "learning_rate": 0.0013942554808436857, + "loss": 0.1299, + "step": 38429 + }, + { + "epoch": 0.3335908542460569, + "grad_norm": 0.29296875, + "learning_rate": 0.0013942273539274484, + "loss": 0.1221, + "step": 38430 + }, + { + "epoch": 0.3335995347262611, + "grad_norm": 0.087890625, + "learning_rate": 0.001394199226689448, + "loss": 0.1064, + "step": 38431 + }, + { + "epoch": 0.3336082152064652, + "grad_norm": 0.37890625, + "learning_rate": 0.0013941710991297162, + "loss": 0.2441, + "step": 38432 + }, + { + "epoch": 0.3336168956866694, + "grad_norm": 0.111328125, + "learning_rate": 0.0013941429712482835, + "loss": 0.1006, + "step": 38433 + }, + { + "epoch": 0.33362557616687355, + "grad_norm": 0.69140625, + "learning_rate": 0.0013941148430451804, + "loss": 0.0742, + "step": 38434 + }, + { + "epoch": 0.33363425664707774, + "grad_norm": 0.0625, + "learning_rate": 0.0013940867145204375, + "loss": 0.0776, + "step": 38435 + }, + { + "epoch": 0.3336429371272819, + "grad_norm": 0.443359375, + "learning_rate": 0.0013940585856740857, + "loss": 0.0938, + "step": 38436 + }, + { + "epoch": 0.3336516176074861, + "grad_norm": 0.1435546875, + "learning_rate": 0.001394030456506156, + "loss": 0.0947, + "step": 38437 + }, + { + "epoch": 0.3336602980876902, + "grad_norm": 1.625, + "learning_rate": 0.001394002327016679, + "loss": 0.1147, + "step": 38438 + }, + { + "epoch": 0.3336689785678944, + "grad_norm": 0.400390625, + "learning_rate": 0.0013939741972056857, + "loss": 0.1016, + "step": 38439 + }, + { + "epoch": 0.33367765904809854, + "grad_norm": 0.251953125, + "learning_rate": 0.0013939460670732062, + "loss": 0.0879, + "step": 38440 + }, + { + "epoch": 0.33368633952830273, + "grad_norm": 0.0791015625, + "learning_rate": 0.001393917936619272, + "loss": 0.1216, + "step": 38441 + }, + { + "epoch": 0.33369502000850687, + "grad_norm": 0.3828125, + "learning_rate": 0.0013938898058439139, + "loss": 0.0786, + "step": 38442 + }, + { + "epoch": 0.333703700488711, + "grad_norm": 0.265625, + "learning_rate": 0.0013938616747471617, + "loss": 0.1562, + "step": 38443 + }, + { + "epoch": 0.3337123809689152, + "grad_norm": 0.33203125, + "learning_rate": 0.0013938335433290468, + "loss": 0.1133, + "step": 38444 + }, + { + "epoch": 0.33372106144911934, + "grad_norm": 0.57421875, + "learning_rate": 0.0013938054115896004, + "loss": 0.0996, + "step": 38445 + }, + { + "epoch": 0.33372974192932353, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013937772795288524, + "loss": 0.084, + "step": 38446 + }, + { + "epoch": 0.33373842240952767, + "grad_norm": 0.3046875, + "learning_rate": 0.001393749147146834, + "loss": 0.1406, + "step": 38447 + }, + { + "epoch": 0.33374710288973186, + "grad_norm": 0.16796875, + "learning_rate": 0.0013937210144435762, + "loss": 0.1064, + "step": 38448 + }, + { + "epoch": 0.333755783369936, + "grad_norm": 0.1181640625, + "learning_rate": 0.0013936928814191091, + "loss": 0.0933, + "step": 38449 + }, + { + "epoch": 0.3337644638501402, + "grad_norm": 0.130859375, + "learning_rate": 0.0013936647480734645, + "loss": 0.127, + "step": 38450 + }, + { + "epoch": 0.33377314433034433, + "grad_norm": 0.1162109375, + "learning_rate": 0.0013936366144066724, + "loss": 0.1729, + "step": 38451 + }, + { + "epoch": 0.3337818248105485, + "grad_norm": 0.52734375, + "learning_rate": 0.0013936084804187633, + "loss": 0.1094, + "step": 38452 + }, + { + "epoch": 0.33379050529075266, + "grad_norm": 0.443359375, + "learning_rate": 0.0013935803461097686, + "loss": 0.1177, + "step": 38453 + }, + { + "epoch": 0.33379918577095685, + "grad_norm": 0.09912109375, + "learning_rate": 0.001393552211479719, + "loss": 0.0718, + "step": 38454 + }, + { + "epoch": 0.333807866251161, + "grad_norm": 0.5703125, + "learning_rate": 0.001393524076528645, + "loss": 0.0986, + "step": 38455 + }, + { + "epoch": 0.3338165467313652, + "grad_norm": 0.27734375, + "learning_rate": 0.0013934959412565773, + "loss": 0.1021, + "step": 38456 + }, + { + "epoch": 0.3338252272115693, + "grad_norm": 0.11279296875, + "learning_rate": 0.001393467805663547, + "loss": 0.1152, + "step": 38457 + }, + { + "epoch": 0.3338339076917735, + "grad_norm": 0.2021484375, + "learning_rate": 0.0013934396697495848, + "loss": 0.085, + "step": 38458 + }, + { + "epoch": 0.33384258817197765, + "grad_norm": 0.125, + "learning_rate": 0.0013934115335147213, + "loss": 0.1133, + "step": 38459 + }, + { + "epoch": 0.33385126865218184, + "grad_norm": 0.443359375, + "learning_rate": 0.0013933833969589876, + "loss": 0.0801, + "step": 38460 + }, + { + "epoch": 0.333859949132386, + "grad_norm": 0.11474609375, + "learning_rate": 0.0013933552600824137, + "loss": 0.0918, + "step": 38461 + }, + { + "epoch": 0.3338686296125902, + "grad_norm": 0.431640625, + "learning_rate": 0.0013933271228850311, + "loss": 0.0977, + "step": 38462 + }, + { + "epoch": 0.3338773100927943, + "grad_norm": 0.322265625, + "learning_rate": 0.0013932989853668706, + "loss": 0.125, + "step": 38463 + }, + { + "epoch": 0.3338859905729985, + "grad_norm": 0.3984375, + "learning_rate": 0.0013932708475279626, + "loss": 0.0859, + "step": 38464 + }, + { + "epoch": 0.33389467105320264, + "grad_norm": 0.1572265625, + "learning_rate": 0.001393242709368338, + "loss": 0.0889, + "step": 38465 + }, + { + "epoch": 0.33390335153340683, + "grad_norm": 0.30859375, + "learning_rate": 0.0013932145708880272, + "loss": 0.0903, + "step": 38466 + }, + { + "epoch": 0.33391203201361097, + "grad_norm": 0.63671875, + "learning_rate": 0.001393186432087062, + "loss": 0.0918, + "step": 38467 + }, + { + "epoch": 0.33392071249381516, + "grad_norm": 0.38671875, + "learning_rate": 0.0013931582929654724, + "loss": 0.1729, + "step": 38468 + }, + { + "epoch": 0.3339293929740193, + "grad_norm": 0.09326171875, + "learning_rate": 0.001393130153523289, + "loss": 0.1196, + "step": 38469 + }, + { + "epoch": 0.3339380734542235, + "grad_norm": 0.2431640625, + "learning_rate": 0.0013931020137605428, + "loss": 0.0874, + "step": 38470 + }, + { + "epoch": 0.33394675393442763, + "grad_norm": 0.392578125, + "learning_rate": 0.001393073873677265, + "loss": 0.085, + "step": 38471 + }, + { + "epoch": 0.3339554344146318, + "grad_norm": 0.51171875, + "learning_rate": 0.001393045733273486, + "loss": 0.3164, + "step": 38472 + }, + { + "epoch": 0.33396411489483596, + "grad_norm": 0.09814453125, + "learning_rate": 0.0013930175925492365, + "loss": 0.0981, + "step": 38473 + }, + { + "epoch": 0.33397279537504015, + "grad_norm": 0.2734375, + "learning_rate": 0.0013929894515045472, + "loss": 0.1191, + "step": 38474 + }, + { + "epoch": 0.3339814758552443, + "grad_norm": 0.310546875, + "learning_rate": 0.001392961310139449, + "loss": 0.0957, + "step": 38475 + }, + { + "epoch": 0.3339901563354485, + "grad_norm": 0.10009765625, + "learning_rate": 0.001392933168453973, + "loss": 0.1094, + "step": 38476 + }, + { + "epoch": 0.3339988368156526, + "grad_norm": 0.45703125, + "learning_rate": 0.0013929050264481495, + "loss": 0.0938, + "step": 38477 + }, + { + "epoch": 0.3340075172958568, + "grad_norm": 0.09228515625, + "learning_rate": 0.0013928768841220095, + "loss": 0.1191, + "step": 38478 + }, + { + "epoch": 0.33401619777606095, + "grad_norm": 0.15625, + "learning_rate": 0.0013928487414755838, + "loss": 0.0781, + "step": 38479 + }, + { + "epoch": 0.33402487825626515, + "grad_norm": 0.080078125, + "learning_rate": 0.0013928205985089032, + "loss": 0.0874, + "step": 38480 + }, + { + "epoch": 0.3340335587364693, + "grad_norm": 0.138671875, + "learning_rate": 0.0013927924552219982, + "loss": 0.0991, + "step": 38481 + }, + { + "epoch": 0.3340422392166735, + "grad_norm": 0.421875, + "learning_rate": 0.0013927643116148998, + "loss": 0.1074, + "step": 38482 + }, + { + "epoch": 0.3340509196968776, + "grad_norm": 0.412109375, + "learning_rate": 0.0013927361676876386, + "loss": 0.1133, + "step": 38483 + }, + { + "epoch": 0.3340596001770818, + "grad_norm": 0.44140625, + "learning_rate": 0.0013927080234402455, + "loss": 0.1152, + "step": 38484 + }, + { + "epoch": 0.33406828065728594, + "grad_norm": 0.36328125, + "learning_rate": 0.0013926798788727519, + "loss": 0.1001, + "step": 38485 + }, + { + "epoch": 0.33407696113749014, + "grad_norm": 0.79296875, + "learning_rate": 0.0013926517339851877, + "loss": 0.0869, + "step": 38486 + }, + { + "epoch": 0.3340856416176943, + "grad_norm": 0.107421875, + "learning_rate": 0.0013926235887775837, + "loss": 0.1289, + "step": 38487 + }, + { + "epoch": 0.33409432209789847, + "grad_norm": 0.3203125, + "learning_rate": 0.0013925954432499712, + "loss": 0.1182, + "step": 38488 + }, + { + "epoch": 0.3341030025781026, + "grad_norm": 0.388671875, + "learning_rate": 0.0013925672974023807, + "loss": 0.0713, + "step": 38489 + }, + { + "epoch": 0.3341116830583068, + "grad_norm": 0.400390625, + "learning_rate": 0.0013925391512348427, + "loss": 0.1045, + "step": 38490 + }, + { + "epoch": 0.33412036353851093, + "grad_norm": 0.10302734375, + "learning_rate": 0.0013925110047473886, + "loss": 0.1348, + "step": 38491 + }, + { + "epoch": 0.3341290440187151, + "grad_norm": 0.68359375, + "learning_rate": 0.0013924828579400487, + "loss": 0.1445, + "step": 38492 + }, + { + "epoch": 0.33413772449891926, + "grad_norm": 0.259765625, + "learning_rate": 0.001392454710812854, + "loss": 0.1133, + "step": 38493 + }, + { + "epoch": 0.33414640497912346, + "grad_norm": 0.1904296875, + "learning_rate": 0.001392426563365835, + "loss": 0.0991, + "step": 38494 + }, + { + "epoch": 0.3341550854593276, + "grad_norm": 0.12890625, + "learning_rate": 0.0013923984155990228, + "loss": 0.0918, + "step": 38495 + }, + { + "epoch": 0.3341637659395318, + "grad_norm": 0.515625, + "learning_rate": 0.0013923702675124484, + "loss": 0.0967, + "step": 38496 + }, + { + "epoch": 0.3341724464197359, + "grad_norm": 0.158203125, + "learning_rate": 0.001392342119106142, + "loss": 0.124, + "step": 38497 + }, + { + "epoch": 0.3341811268999401, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013923139703801348, + "loss": 0.1094, + "step": 38498 + }, + { + "epoch": 0.33418980738014425, + "grad_norm": 0.58984375, + "learning_rate": 0.0013922858213344575, + "loss": 0.1162, + "step": 38499 + }, + { + "epoch": 0.33419848786034845, + "grad_norm": 0.263671875, + "learning_rate": 0.0013922576719691405, + "loss": 0.0771, + "step": 38500 + }, + { + "epoch": 0.3342071683405526, + "grad_norm": 0.2392578125, + "learning_rate": 0.0013922295222842153, + "loss": 0.1279, + "step": 38501 + }, + { + "epoch": 0.3342158488207568, + "grad_norm": 0.48046875, + "learning_rate": 0.001392201372279712, + "loss": 0.0952, + "step": 38502 + }, + { + "epoch": 0.3342245293009609, + "grad_norm": 0.3046875, + "learning_rate": 0.0013921732219556618, + "loss": 0.1084, + "step": 38503 + }, + { + "epoch": 0.3342332097811651, + "grad_norm": 0.2060546875, + "learning_rate": 0.0013921450713120951, + "loss": 0.0737, + "step": 38504 + }, + { + "epoch": 0.33424189026136925, + "grad_norm": 0.37109375, + "learning_rate": 0.0013921169203490432, + "loss": 0.1143, + "step": 38505 + }, + { + "epoch": 0.33425057074157344, + "grad_norm": 0.36328125, + "learning_rate": 0.0013920887690665365, + "loss": 0.1138, + "step": 38506 + }, + { + "epoch": 0.3342592512217776, + "grad_norm": 0.1689453125, + "learning_rate": 0.001392060617464606, + "loss": 0.0913, + "step": 38507 + }, + { + "epoch": 0.33426793170198177, + "grad_norm": 0.478515625, + "learning_rate": 0.0013920324655432824, + "loss": 0.1016, + "step": 38508 + }, + { + "epoch": 0.3342766121821859, + "grad_norm": 0.142578125, + "learning_rate": 0.0013920043133025964, + "loss": 0.0771, + "step": 38509 + }, + { + "epoch": 0.3342852926623901, + "grad_norm": 0.37890625, + "learning_rate": 0.001391976160742579, + "loss": 0.1211, + "step": 38510 + }, + { + "epoch": 0.33429397314259424, + "grad_norm": 0.1826171875, + "learning_rate": 0.0013919480078632607, + "loss": 0.1289, + "step": 38511 + }, + { + "epoch": 0.33430265362279843, + "grad_norm": 0.39453125, + "learning_rate": 0.0013919198546646725, + "loss": 0.1162, + "step": 38512 + }, + { + "epoch": 0.33431133410300257, + "grad_norm": 0.435546875, + "learning_rate": 0.0013918917011468454, + "loss": 0.1445, + "step": 38513 + }, + { + "epoch": 0.33432001458320676, + "grad_norm": 0.2734375, + "learning_rate": 0.0013918635473098096, + "loss": 0.1582, + "step": 38514 + }, + { + "epoch": 0.3343286950634109, + "grad_norm": 0.2119140625, + "learning_rate": 0.0013918353931535962, + "loss": 0.1123, + "step": 38515 + }, + { + "epoch": 0.3343373755436151, + "grad_norm": 0.65625, + "learning_rate": 0.0013918072386782361, + "loss": 0.1182, + "step": 38516 + }, + { + "epoch": 0.3343460560238192, + "grad_norm": 0.4296875, + "learning_rate": 0.0013917790838837599, + "loss": 0.1074, + "step": 38517 + }, + { + "epoch": 0.3343547365040234, + "grad_norm": 0.423828125, + "learning_rate": 0.0013917509287701983, + "loss": 0.1543, + "step": 38518 + }, + { + "epoch": 0.33436341698422756, + "grad_norm": 0.310546875, + "learning_rate": 0.0013917227733375825, + "loss": 0.0889, + "step": 38519 + }, + { + "epoch": 0.33437209746443175, + "grad_norm": 0.2001953125, + "learning_rate": 0.0013916946175859433, + "loss": 0.1011, + "step": 38520 + }, + { + "epoch": 0.3343807779446359, + "grad_norm": 0.1767578125, + "learning_rate": 0.0013916664615153112, + "loss": 0.084, + "step": 38521 + }, + { + "epoch": 0.3343894584248401, + "grad_norm": 0.50390625, + "learning_rate": 0.001391638305125717, + "loss": 0.085, + "step": 38522 + }, + { + "epoch": 0.3343981389050442, + "grad_norm": 0.1787109375, + "learning_rate": 0.0013916101484171908, + "loss": 0.1123, + "step": 38523 + }, + { + "epoch": 0.3344068193852484, + "grad_norm": 0.2001953125, + "learning_rate": 0.001391581991389765, + "loss": 0.1182, + "step": 38524 + }, + { + "epoch": 0.33441549986545255, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013915538340434693, + "loss": 0.0742, + "step": 38525 + }, + { + "epoch": 0.33442418034565674, + "grad_norm": 0.453125, + "learning_rate": 0.0013915256763783346, + "loss": 0.0996, + "step": 38526 + }, + { + "epoch": 0.3344328608258609, + "grad_norm": 0.486328125, + "learning_rate": 0.0013914975183943919, + "loss": 0.0786, + "step": 38527 + }, + { + "epoch": 0.33444154130606507, + "grad_norm": 0.1748046875, + "learning_rate": 0.0013914693600916718, + "loss": 0.1182, + "step": 38528 + }, + { + "epoch": 0.3344502217862692, + "grad_norm": 0.212890625, + "learning_rate": 0.0013914412014702053, + "loss": 0.0776, + "step": 38529 + }, + { + "epoch": 0.3344589022664734, + "grad_norm": 0.216796875, + "learning_rate": 0.0013914130425300227, + "loss": 0.0825, + "step": 38530 + }, + { + "epoch": 0.33446758274667754, + "grad_norm": 0.09619140625, + "learning_rate": 0.0013913848832711557, + "loss": 0.1069, + "step": 38531 + }, + { + "epoch": 0.33447626322688173, + "grad_norm": 0.60546875, + "learning_rate": 0.0013913567236936342, + "loss": 0.1348, + "step": 38532 + }, + { + "epoch": 0.33448494370708587, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013913285637974894, + "loss": 0.1348, + "step": 38533 + }, + { + "epoch": 0.33449362418729006, + "grad_norm": 0.19921875, + "learning_rate": 0.0013913004035827521, + "loss": 0.1152, + "step": 38534 + }, + { + "epoch": 0.3345023046674942, + "grad_norm": 0.2314453125, + "learning_rate": 0.0013912722430494534, + "loss": 0.0986, + "step": 38535 + }, + { + "epoch": 0.3345109851476984, + "grad_norm": 0.396484375, + "learning_rate": 0.0013912440821976235, + "loss": 0.1172, + "step": 38536 + }, + { + "epoch": 0.33451966562790253, + "grad_norm": 0.69140625, + "learning_rate": 0.0013912159210272934, + "loss": 0.1138, + "step": 38537 + }, + { + "epoch": 0.3345283461081067, + "grad_norm": 0.28125, + "learning_rate": 0.001391187759538494, + "loss": 0.0938, + "step": 38538 + }, + { + "epoch": 0.33453702658831086, + "grad_norm": 0.08154296875, + "learning_rate": 0.0013911595977312559, + "loss": 0.085, + "step": 38539 + }, + { + "epoch": 0.33454570706851505, + "grad_norm": 0.18359375, + "learning_rate": 0.0013911314356056103, + "loss": 0.1108, + "step": 38540 + }, + { + "epoch": 0.3345543875487192, + "grad_norm": 0.177734375, + "learning_rate": 0.0013911032731615874, + "loss": 0.0957, + "step": 38541 + }, + { + "epoch": 0.3345630680289234, + "grad_norm": 0.193359375, + "learning_rate": 0.0013910751103992185, + "loss": 0.1562, + "step": 38542 + }, + { + "epoch": 0.3345717485091275, + "grad_norm": 0.08837890625, + "learning_rate": 0.0013910469473185343, + "loss": 0.1011, + "step": 38543 + }, + { + "epoch": 0.3345804289893317, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013910187839195653, + "loss": 0.1182, + "step": 38544 + }, + { + "epoch": 0.33458910946953585, + "grad_norm": 0.4765625, + "learning_rate": 0.001390990620202343, + "loss": 0.1035, + "step": 38545 + }, + { + "epoch": 0.33459778994974004, + "grad_norm": 0.50390625, + "learning_rate": 0.0013909624561668974, + "loss": 0.0703, + "step": 38546 + }, + { + "epoch": 0.3346064704299442, + "grad_norm": 0.08935546875, + "learning_rate": 0.0013909342918132595, + "loss": 0.1396, + "step": 38547 + }, + { + "epoch": 0.3346151509101484, + "grad_norm": 0.37109375, + "learning_rate": 0.0013909061271414605, + "loss": 0.105, + "step": 38548 + }, + { + "epoch": 0.3346238313903525, + "grad_norm": 0.2236328125, + "learning_rate": 0.0013908779621515308, + "loss": 0.0718, + "step": 38549 + }, + { + "epoch": 0.3346325118705567, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013908497968435013, + "loss": 0.1045, + "step": 38550 + }, + { + "epoch": 0.33464119235076084, + "grad_norm": 0.1259765625, + "learning_rate": 0.001390821631217403, + "loss": 0.0703, + "step": 38551 + }, + { + "epoch": 0.33464987283096503, + "grad_norm": 0.2333984375, + "learning_rate": 0.0013907934652732663, + "loss": 0.1128, + "step": 38552 + }, + { + "epoch": 0.33465855331116917, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013907652990111225, + "loss": 0.1592, + "step": 38553 + }, + { + "epoch": 0.33466723379137336, + "grad_norm": 0.326171875, + "learning_rate": 0.001390737132431002, + "loss": 0.105, + "step": 38554 + }, + { + "epoch": 0.3346759142715775, + "grad_norm": 0.90234375, + "learning_rate": 0.001390708965532936, + "loss": 0.1875, + "step": 38555 + }, + { + "epoch": 0.3346845947517817, + "grad_norm": 0.31640625, + "learning_rate": 0.0013906807983169545, + "loss": 0.083, + "step": 38556 + }, + { + "epoch": 0.33469327523198583, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013906526307830892, + "loss": 0.0996, + "step": 38557 + }, + { + "epoch": 0.33470195571219, + "grad_norm": 0.447265625, + "learning_rate": 0.0013906244629313706, + "loss": 0.1201, + "step": 38558 + }, + { + "epoch": 0.33471063619239416, + "grad_norm": 0.07763671875, + "learning_rate": 0.0013905962947618291, + "loss": 0.0898, + "step": 38559 + }, + { + "epoch": 0.33471931667259835, + "grad_norm": 0.17578125, + "learning_rate": 0.0013905681262744963, + "loss": 0.1016, + "step": 38560 + }, + { + "epoch": 0.3347279971528025, + "grad_norm": 0.1875, + "learning_rate": 0.001390539957469402, + "loss": 0.1143, + "step": 38561 + }, + { + "epoch": 0.3347366776330067, + "grad_norm": 0.150390625, + "learning_rate": 0.0013905117883465778, + "loss": 0.1191, + "step": 38562 + }, + { + "epoch": 0.3347453581132108, + "grad_norm": 0.255859375, + "learning_rate": 0.0013904836189060547, + "loss": 0.0918, + "step": 38563 + }, + { + "epoch": 0.334754038593415, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013904554491478629, + "loss": 0.1123, + "step": 38564 + }, + { + "epoch": 0.33476271907361915, + "grad_norm": 0.69140625, + "learning_rate": 0.0013904272790720331, + "loss": 0.084, + "step": 38565 + }, + { + "epoch": 0.3347713995538233, + "grad_norm": 0.1943359375, + "learning_rate": 0.0013903991086785966, + "loss": 0.1006, + "step": 38566 + }, + { + "epoch": 0.3347800800340275, + "grad_norm": 0.52734375, + "learning_rate": 0.001390370937967584, + "loss": 0.1406, + "step": 38567 + }, + { + "epoch": 0.3347887605142316, + "grad_norm": 0.0869140625, + "learning_rate": 0.001390342766939026, + "loss": 0.0815, + "step": 38568 + }, + { + "epoch": 0.3347974409944358, + "grad_norm": 0.3046875, + "learning_rate": 0.0013903145955929538, + "loss": 0.1123, + "step": 38569 + }, + { + "epoch": 0.33480612147463995, + "grad_norm": 0.6875, + "learning_rate": 0.0013902864239293978, + "loss": 0.1416, + "step": 38570 + }, + { + "epoch": 0.33481480195484414, + "grad_norm": 0.19140625, + "learning_rate": 0.0013902582519483885, + "loss": 0.126, + "step": 38571 + }, + { + "epoch": 0.3348234824350483, + "grad_norm": 0.21875, + "learning_rate": 0.0013902300796499575, + "loss": 0.1133, + "step": 38572 + }, + { + "epoch": 0.3348321629152525, + "grad_norm": 0.11572265625, + "learning_rate": 0.001390201907034135, + "loss": 0.0991, + "step": 38573 + }, + { + "epoch": 0.3348408433954566, + "grad_norm": 0.671875, + "learning_rate": 0.0013901737341009523, + "loss": 0.1108, + "step": 38574 + }, + { + "epoch": 0.3348495238756608, + "grad_norm": 0.1923828125, + "learning_rate": 0.00139014556085044, + "loss": 0.0957, + "step": 38575 + }, + { + "epoch": 0.33485820435586494, + "grad_norm": 0.3515625, + "learning_rate": 0.0013901173872826289, + "loss": 0.1357, + "step": 38576 + }, + { + "epoch": 0.33486688483606913, + "grad_norm": 1.125, + "learning_rate": 0.0013900892133975496, + "loss": 0.1162, + "step": 38577 + }, + { + "epoch": 0.33487556531627327, + "grad_norm": 0.349609375, + "learning_rate": 0.0013900610391952334, + "loss": 0.123, + "step": 38578 + }, + { + "epoch": 0.33488424579647746, + "grad_norm": 0.4453125, + "learning_rate": 0.0013900328646757104, + "loss": 0.127, + "step": 38579 + }, + { + "epoch": 0.3348929262766816, + "grad_norm": 0.423828125, + "learning_rate": 0.0013900046898390117, + "loss": 0.2148, + "step": 38580 + }, + { + "epoch": 0.3349016067568858, + "grad_norm": 0.1484375, + "learning_rate": 0.0013899765146851686, + "loss": 0.1348, + "step": 38581 + }, + { + "epoch": 0.33491028723708993, + "grad_norm": 0.59765625, + "learning_rate": 0.0013899483392142114, + "loss": 0.1162, + "step": 38582 + }, + { + "epoch": 0.3349189677172941, + "grad_norm": 0.1904296875, + "learning_rate": 0.0013899201634261713, + "loss": 0.1055, + "step": 38583 + }, + { + "epoch": 0.33492764819749826, + "grad_norm": 0.421875, + "learning_rate": 0.0013898919873210787, + "loss": 0.125, + "step": 38584 + }, + { + "epoch": 0.33493632867770246, + "grad_norm": 0.111328125, + "learning_rate": 0.0013898638108989648, + "loss": 0.0977, + "step": 38585 + }, + { + "epoch": 0.3349450091579066, + "grad_norm": 0.41015625, + "learning_rate": 0.0013898356341598597, + "loss": 0.1113, + "step": 38586 + }, + { + "epoch": 0.3349536896381108, + "grad_norm": 0.3203125, + "learning_rate": 0.0013898074571037948, + "loss": 0.1152, + "step": 38587 + }, + { + "epoch": 0.3349623701183149, + "grad_norm": 0.515625, + "learning_rate": 0.0013897792797308013, + "loss": 0.127, + "step": 38588 + }, + { + "epoch": 0.3349710505985191, + "grad_norm": 0.240234375, + "learning_rate": 0.001389751102040909, + "loss": 0.1143, + "step": 38589 + }, + { + "epoch": 0.33497973107872325, + "grad_norm": 0.28125, + "learning_rate": 0.0013897229240341497, + "loss": 0.082, + "step": 38590 + }, + { + "epoch": 0.33498841155892745, + "grad_norm": 0.427734375, + "learning_rate": 0.0013896947457105531, + "loss": 0.0913, + "step": 38591 + }, + { + "epoch": 0.3349970920391316, + "grad_norm": 0.80078125, + "learning_rate": 0.0013896665670701512, + "loss": 0.083, + "step": 38592 + }, + { + "epoch": 0.3350057725193358, + "grad_norm": 0.08935546875, + "learning_rate": 0.0013896383881129744, + "loss": 0.0923, + "step": 38593 + }, + { + "epoch": 0.3350144529995399, + "grad_norm": 0.69140625, + "learning_rate": 0.0013896102088390531, + "loss": 0.1025, + "step": 38594 + }, + { + "epoch": 0.3350231334797441, + "grad_norm": 0.52734375, + "learning_rate": 0.0013895820292484185, + "loss": 0.1152, + "step": 38595 + }, + { + "epoch": 0.33503181395994824, + "grad_norm": 0.34375, + "learning_rate": 0.0013895538493411015, + "loss": 0.1113, + "step": 38596 + }, + { + "epoch": 0.33504049444015244, + "grad_norm": 0.099609375, + "learning_rate": 0.0013895256691171325, + "loss": 0.0967, + "step": 38597 + }, + { + "epoch": 0.3350491749203566, + "grad_norm": 0.56640625, + "learning_rate": 0.0013894974885765431, + "loss": 0.126, + "step": 38598 + }, + { + "epoch": 0.33505785540056077, + "grad_norm": 0.1123046875, + "learning_rate": 0.001389469307719363, + "loss": 0.1055, + "step": 38599 + }, + { + "epoch": 0.3350665358807649, + "grad_norm": 0.2578125, + "learning_rate": 0.0013894411265456235, + "loss": 0.1001, + "step": 38600 + }, + { + "epoch": 0.3350752163609691, + "grad_norm": 0.376953125, + "learning_rate": 0.0013894129450553558, + "loss": 0.104, + "step": 38601 + }, + { + "epoch": 0.33508389684117323, + "grad_norm": 0.12890625, + "learning_rate": 0.0013893847632485906, + "loss": 0.1152, + "step": 38602 + }, + { + "epoch": 0.3350925773213774, + "grad_norm": 0.7421875, + "learning_rate": 0.0013893565811253586, + "loss": 0.1758, + "step": 38603 + }, + { + "epoch": 0.33510125780158156, + "grad_norm": 0.486328125, + "learning_rate": 0.0013893283986856904, + "loss": 0.0806, + "step": 38604 + }, + { + "epoch": 0.33510993828178576, + "grad_norm": 0.5, + "learning_rate": 0.001389300215929617, + "loss": 0.1582, + "step": 38605 + }, + { + "epoch": 0.3351186187619899, + "grad_norm": 0.6015625, + "learning_rate": 0.0013892720328571694, + "loss": 0.0708, + "step": 38606 + }, + { + "epoch": 0.3351272992421941, + "grad_norm": 0.31640625, + "learning_rate": 0.0013892438494683785, + "loss": 0.1309, + "step": 38607 + }, + { + "epoch": 0.3351359797223982, + "grad_norm": 0.134765625, + "learning_rate": 0.0013892156657632744, + "loss": 0.1348, + "step": 38608 + }, + { + "epoch": 0.3351446602026024, + "grad_norm": 0.6015625, + "learning_rate": 0.0013891874817418882, + "loss": 0.0957, + "step": 38609 + }, + { + "epoch": 0.33515334068280656, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013891592974042515, + "loss": 0.0898, + "step": 38610 + }, + { + "epoch": 0.33516202116301075, + "grad_norm": 0.08642578125, + "learning_rate": 0.0013891311127503944, + "loss": 0.0684, + "step": 38611 + }, + { + "epoch": 0.3351707016432149, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013891029277803475, + "loss": 0.1523, + "step": 38612 + }, + { + "epoch": 0.3351793821234191, + "grad_norm": 0.13671875, + "learning_rate": 0.0013890747424941422, + "loss": 0.1436, + "step": 38613 + }, + { + "epoch": 0.3351880626036232, + "grad_norm": 0.41796875, + "learning_rate": 0.001389046556891809, + "loss": 0.1123, + "step": 38614 + }, + { + "epoch": 0.3351967430838274, + "grad_norm": 0.3203125, + "learning_rate": 0.0013890183709733793, + "loss": 0.1055, + "step": 38615 + }, + { + "epoch": 0.33520542356403155, + "grad_norm": 0.59765625, + "learning_rate": 0.0013889901847388831, + "loss": 0.0986, + "step": 38616 + }, + { + "epoch": 0.33521410404423574, + "grad_norm": 0.1708984375, + "learning_rate": 0.0013889619981883514, + "loss": 0.1309, + "step": 38617 + }, + { + "epoch": 0.3352227845244399, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013889338113218155, + "loss": 0.106, + "step": 38618 + }, + { + "epoch": 0.33523146500464407, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013889056241393057, + "loss": 0.1006, + "step": 38619 + }, + { + "epoch": 0.3352401454848482, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013888774366408532, + "loss": 0.0928, + "step": 38620 + }, + { + "epoch": 0.3352488259650524, + "grad_norm": 0.1796875, + "learning_rate": 0.0013888492488264887, + "loss": 0.1299, + "step": 38621 + }, + { + "epoch": 0.33525750644525654, + "grad_norm": 0.625, + "learning_rate": 0.0013888210606962429, + "loss": 0.1123, + "step": 38622 + }, + { + "epoch": 0.33526618692546073, + "grad_norm": 0.181640625, + "learning_rate": 0.001388792872250147, + "loss": 0.1328, + "step": 38623 + }, + { + "epoch": 0.33527486740566487, + "grad_norm": 0.12109375, + "learning_rate": 0.001388764683488231, + "loss": 0.1426, + "step": 38624 + }, + { + "epoch": 0.33528354788586906, + "grad_norm": 0.232421875, + "learning_rate": 0.0013887364944105264, + "loss": 0.1816, + "step": 38625 + }, + { + "epoch": 0.3352922283660732, + "grad_norm": 0.10400390625, + "learning_rate": 0.0013887083050170643, + "loss": 0.1465, + "step": 38626 + }, + { + "epoch": 0.3353009088462774, + "grad_norm": 0.1923828125, + "learning_rate": 0.001388680115307875, + "loss": 0.123, + "step": 38627 + }, + { + "epoch": 0.3353095893264815, + "grad_norm": 0.5625, + "learning_rate": 0.0013886519252829894, + "loss": 0.125, + "step": 38628 + }, + { + "epoch": 0.3353182698066857, + "grad_norm": 0.1357421875, + "learning_rate": 0.001388623734942438, + "loss": 0.1211, + "step": 38629 + }, + { + "epoch": 0.33532695028688986, + "grad_norm": 0.12890625, + "learning_rate": 0.0013885955442862526, + "loss": 0.1191, + "step": 38630 + }, + { + "epoch": 0.33533563076709405, + "grad_norm": 0.2734375, + "learning_rate": 0.0013885673533144635, + "loss": 0.1172, + "step": 38631 + }, + { + "epoch": 0.3353443112472982, + "grad_norm": 0.11083984375, + "learning_rate": 0.001388539162027101, + "loss": 0.0815, + "step": 38632 + }, + { + "epoch": 0.3353529917275024, + "grad_norm": 0.416015625, + "learning_rate": 0.0013885109704241969, + "loss": 0.1016, + "step": 38633 + }, + { + "epoch": 0.3353616722077065, + "grad_norm": 0.224609375, + "learning_rate": 0.0013884827785057812, + "loss": 0.0972, + "step": 38634 + }, + { + "epoch": 0.3353703526879107, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013884545862718851, + "loss": 0.1562, + "step": 38635 + }, + { + "epoch": 0.33537903316811485, + "grad_norm": 0.208984375, + "learning_rate": 0.0013884263937225397, + "loss": 0.0801, + "step": 38636 + }, + { + "epoch": 0.33538771364831904, + "grad_norm": 0.12890625, + "learning_rate": 0.0013883982008577752, + "loss": 0.0874, + "step": 38637 + }, + { + "epoch": 0.3353963941285232, + "grad_norm": 0.2314453125, + "learning_rate": 0.0013883700076776227, + "loss": 0.0791, + "step": 38638 + }, + { + "epoch": 0.33540507460872737, + "grad_norm": 0.314453125, + "learning_rate": 0.0013883418141821132, + "loss": 0.1016, + "step": 38639 + }, + { + "epoch": 0.3354137550889315, + "grad_norm": 0.31640625, + "learning_rate": 0.0013883136203712773, + "loss": 0.0903, + "step": 38640 + }, + { + "epoch": 0.3354224355691357, + "grad_norm": 0.294921875, + "learning_rate": 0.0013882854262451463, + "loss": 0.127, + "step": 38641 + }, + { + "epoch": 0.33543111604933984, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013882572318037506, + "loss": 0.0737, + "step": 38642 + }, + { + "epoch": 0.33543979652954403, + "grad_norm": 0.345703125, + "learning_rate": 0.001388229037047121, + "loss": 0.0771, + "step": 38643 + }, + { + "epoch": 0.33544847700974817, + "grad_norm": 0.51953125, + "learning_rate": 0.0013882008419752886, + "loss": 0.0938, + "step": 38644 + }, + { + "epoch": 0.33545715748995236, + "grad_norm": 0.107421875, + "learning_rate": 0.001388172646588284, + "loss": 0.1001, + "step": 38645 + }, + { + "epoch": 0.3354658379701565, + "grad_norm": 0.314453125, + "learning_rate": 0.001388144450886138, + "loss": 0.0977, + "step": 38646 + }, + { + "epoch": 0.3354745184503607, + "grad_norm": 0.578125, + "learning_rate": 0.0013881162548688816, + "loss": 0.1445, + "step": 38647 + }, + { + "epoch": 0.33548319893056483, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013880880585365456, + "loss": 0.0732, + "step": 38648 + }, + { + "epoch": 0.335491879410769, + "grad_norm": 0.267578125, + "learning_rate": 0.0013880598618891608, + "loss": 0.1533, + "step": 38649 + }, + { + "epoch": 0.33550055989097316, + "grad_norm": 0.236328125, + "learning_rate": 0.0013880316649267583, + "loss": 0.0811, + "step": 38650 + }, + { + "epoch": 0.33550924037117735, + "grad_norm": 0.78515625, + "learning_rate": 0.0013880034676493683, + "loss": 0.0991, + "step": 38651 + }, + { + "epoch": 0.3355179208513815, + "grad_norm": 0.61328125, + "learning_rate": 0.0013879752700570226, + "loss": 0.1367, + "step": 38652 + }, + { + "epoch": 0.3355266013315857, + "grad_norm": 0.1474609375, + "learning_rate": 0.001387947072149751, + "loss": 0.1172, + "step": 38653 + }, + { + "epoch": 0.3355352818117898, + "grad_norm": 0.328125, + "learning_rate": 0.001387918873927585, + "loss": 0.1025, + "step": 38654 + }, + { + "epoch": 0.335543962291994, + "grad_norm": 0.361328125, + "learning_rate": 0.001387890675390555, + "loss": 0.0928, + "step": 38655 + }, + { + "epoch": 0.33555264277219815, + "grad_norm": 0.267578125, + "learning_rate": 0.0013878624765386925, + "loss": 0.1328, + "step": 38656 + }, + { + "epoch": 0.33556132325240234, + "grad_norm": 0.10400390625, + "learning_rate": 0.0013878342773720276, + "loss": 0.1318, + "step": 38657 + }, + { + "epoch": 0.3355700037326065, + "grad_norm": 0.2236328125, + "learning_rate": 0.0013878060778905917, + "loss": 0.1416, + "step": 38658 + }, + { + "epoch": 0.3355786842128107, + "grad_norm": 0.25, + "learning_rate": 0.0013877778780944153, + "loss": 0.1152, + "step": 38659 + }, + { + "epoch": 0.3355873646930148, + "grad_norm": 0.376953125, + "learning_rate": 0.0013877496779835293, + "loss": 0.1006, + "step": 38660 + }, + { + "epoch": 0.335596045173219, + "grad_norm": 0.875, + "learning_rate": 0.0013877214775579645, + "loss": 0.1299, + "step": 38661 + }, + { + "epoch": 0.33560472565342314, + "grad_norm": 0.396484375, + "learning_rate": 0.0013876932768177516, + "loss": 0.0967, + "step": 38662 + }, + { + "epoch": 0.33561340613362733, + "grad_norm": 1.0, + "learning_rate": 0.0013876650757629221, + "loss": 0.1299, + "step": 38663 + }, + { + "epoch": 0.33562208661383147, + "grad_norm": 0.08935546875, + "learning_rate": 0.0013876368743935064, + "loss": 0.1001, + "step": 38664 + }, + { + "epoch": 0.33563076709403566, + "grad_norm": 0.259765625, + "learning_rate": 0.001387608672709535, + "loss": 0.0947, + "step": 38665 + }, + { + "epoch": 0.3356394475742398, + "grad_norm": 0.251953125, + "learning_rate": 0.0013875804707110394, + "loss": 0.1035, + "step": 38666 + }, + { + "epoch": 0.335648128054444, + "grad_norm": 0.6015625, + "learning_rate": 0.00138755226839805, + "loss": 0.0928, + "step": 38667 + }, + { + "epoch": 0.33565680853464813, + "grad_norm": 0.369140625, + "learning_rate": 0.0013875240657705975, + "loss": 0.127, + "step": 38668 + }, + { + "epoch": 0.3356654890148523, + "grad_norm": 0.208984375, + "learning_rate": 0.0013874958628287133, + "loss": 0.0771, + "step": 38669 + }, + { + "epoch": 0.33567416949505646, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013874676595724278, + "loss": 0.0952, + "step": 38670 + }, + { + "epoch": 0.33568284997526066, + "grad_norm": 0.478515625, + "learning_rate": 0.0013874394560017722, + "loss": 0.0918, + "step": 38671 + }, + { + "epoch": 0.3356915304554648, + "grad_norm": 0.26953125, + "learning_rate": 0.001387411252116777, + "loss": 0.0952, + "step": 38672 + }, + { + "epoch": 0.335700210935669, + "grad_norm": 0.12109375, + "learning_rate": 0.001387383047917473, + "loss": 0.1011, + "step": 38673 + }, + { + "epoch": 0.3357088914158731, + "grad_norm": 0.470703125, + "learning_rate": 0.0013873548434038916, + "loss": 0.1387, + "step": 38674 + }, + { + "epoch": 0.3357175718960773, + "grad_norm": 0.345703125, + "learning_rate": 0.001387326638576063, + "loss": 0.0752, + "step": 38675 + }, + { + "epoch": 0.33572625237628145, + "grad_norm": 0.271484375, + "learning_rate": 0.0013872984334340182, + "loss": 0.1035, + "step": 38676 + }, + { + "epoch": 0.33573493285648565, + "grad_norm": 0.58203125, + "learning_rate": 0.0013872702279777882, + "loss": 0.1641, + "step": 38677 + }, + { + "epoch": 0.3357436133366898, + "grad_norm": 0.138671875, + "learning_rate": 0.001387242022207404, + "loss": 0.1377, + "step": 38678 + }, + { + "epoch": 0.335752293816894, + "grad_norm": 0.1591796875, + "learning_rate": 0.0013872138161228962, + "loss": 0.0801, + "step": 38679 + }, + { + "epoch": 0.3357609742970981, + "grad_norm": 0.09521484375, + "learning_rate": 0.0013871856097242956, + "loss": 0.1172, + "step": 38680 + }, + { + "epoch": 0.3357696547773023, + "grad_norm": 0.2890625, + "learning_rate": 0.001387157403011633, + "loss": 0.1211, + "step": 38681 + }, + { + "epoch": 0.33577833525750644, + "grad_norm": 0.30859375, + "learning_rate": 0.0013871291959849396, + "loss": 0.0996, + "step": 38682 + }, + { + "epoch": 0.33578701573771064, + "grad_norm": 0.443359375, + "learning_rate": 0.001387100988644246, + "loss": 0.0835, + "step": 38683 + }, + { + "epoch": 0.3357956962179148, + "grad_norm": 0.162109375, + "learning_rate": 0.0013870727809895832, + "loss": 0.0913, + "step": 38684 + }, + { + "epoch": 0.33580437669811897, + "grad_norm": 0.47265625, + "learning_rate": 0.0013870445730209816, + "loss": 0.0898, + "step": 38685 + }, + { + "epoch": 0.3358130571783231, + "grad_norm": 0.6015625, + "learning_rate": 0.001387016364738472, + "loss": 0.1133, + "step": 38686 + }, + { + "epoch": 0.3358217376585273, + "grad_norm": 0.3046875, + "learning_rate": 0.0013869881561420865, + "loss": 0.1191, + "step": 38687 + }, + { + "epoch": 0.33583041813873143, + "grad_norm": 0.1884765625, + "learning_rate": 0.001386959947231855, + "loss": 0.1021, + "step": 38688 + }, + { + "epoch": 0.33583909861893557, + "grad_norm": 0.115234375, + "learning_rate": 0.001386931738007808, + "loss": 0.1182, + "step": 38689 + }, + { + "epoch": 0.33584777909913976, + "grad_norm": 0.119140625, + "learning_rate": 0.0013869035284699766, + "loss": 0.1089, + "step": 38690 + }, + { + "epoch": 0.3358564595793439, + "grad_norm": 0.2578125, + "learning_rate": 0.0013868753186183922, + "loss": 0.1699, + "step": 38691 + }, + { + "epoch": 0.3358651400595481, + "grad_norm": 0.11083984375, + "learning_rate": 0.0013868471084530851, + "loss": 0.0879, + "step": 38692 + }, + { + "epoch": 0.33587382053975223, + "grad_norm": 0.1875, + "learning_rate": 0.0013868188979740863, + "loss": 0.1367, + "step": 38693 + }, + { + "epoch": 0.3358825010199564, + "grad_norm": 0.392578125, + "learning_rate": 0.001386790687181427, + "loss": 0.0845, + "step": 38694 + }, + { + "epoch": 0.33589118150016056, + "grad_norm": 0.25390625, + "learning_rate": 0.0013867624760751371, + "loss": 0.1367, + "step": 38695 + }, + { + "epoch": 0.33589986198036476, + "grad_norm": 0.21484375, + "learning_rate": 0.0013867342646552486, + "loss": 0.1084, + "step": 38696 + }, + { + "epoch": 0.3359085424605689, + "grad_norm": 0.39453125, + "learning_rate": 0.0013867060529217917, + "loss": 0.0986, + "step": 38697 + }, + { + "epoch": 0.3359172229407731, + "grad_norm": 0.15234375, + "learning_rate": 0.001386677840874797, + "loss": 0.0952, + "step": 38698 + }, + { + "epoch": 0.3359259034209772, + "grad_norm": 0.1884765625, + "learning_rate": 0.001386649628514296, + "loss": 0.0737, + "step": 38699 + }, + { + "epoch": 0.3359345839011814, + "grad_norm": 0.0791015625, + "learning_rate": 0.0013866214158403194, + "loss": 0.0996, + "step": 38700 + }, + { + "epoch": 0.33594326438138555, + "grad_norm": 0.3359375, + "learning_rate": 0.001386593202852898, + "loss": 0.0635, + "step": 38701 + }, + { + "epoch": 0.33595194486158975, + "grad_norm": 0.09326171875, + "learning_rate": 0.0013865649895520625, + "loss": 0.1318, + "step": 38702 + }, + { + "epoch": 0.3359606253417939, + "grad_norm": 0.3671875, + "learning_rate": 0.0013865367759378436, + "loss": 0.0796, + "step": 38703 + }, + { + "epoch": 0.3359693058219981, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013865085620102724, + "loss": 0.0903, + "step": 38704 + }, + { + "epoch": 0.3359779863022022, + "grad_norm": 0.126953125, + "learning_rate": 0.0013864803477693797, + "loss": 0.1465, + "step": 38705 + }, + { + "epoch": 0.3359866667824064, + "grad_norm": 0.09228515625, + "learning_rate": 0.0013864521332151967, + "loss": 0.1016, + "step": 38706 + }, + { + "epoch": 0.33599534726261054, + "grad_norm": 0.3671875, + "learning_rate": 0.0013864239183477534, + "loss": 0.1328, + "step": 38707 + }, + { + "epoch": 0.33600402774281474, + "grad_norm": 0.16796875, + "learning_rate": 0.0013863957031670818, + "loss": 0.127, + "step": 38708 + }, + { + "epoch": 0.3360127082230189, + "grad_norm": 0.1298828125, + "learning_rate": 0.001386367487673212, + "loss": 0.0996, + "step": 38709 + }, + { + "epoch": 0.33602138870322307, + "grad_norm": 0.36328125, + "learning_rate": 0.0013863392718661749, + "loss": 0.1045, + "step": 38710 + }, + { + "epoch": 0.3360300691834272, + "grad_norm": 0.51953125, + "learning_rate": 0.0013863110557460019, + "loss": 0.1001, + "step": 38711 + }, + { + "epoch": 0.3360387496636314, + "grad_norm": 0.13671875, + "learning_rate": 0.0013862828393127228, + "loss": 0.0864, + "step": 38712 + }, + { + "epoch": 0.33604743014383553, + "grad_norm": 0.298828125, + "learning_rate": 0.0013862546225663693, + "loss": 0.1797, + "step": 38713 + }, + { + "epoch": 0.33605611062403973, + "grad_norm": 0.15234375, + "learning_rate": 0.001386226405506972, + "loss": 0.1211, + "step": 38714 + }, + { + "epoch": 0.33606479110424387, + "grad_norm": 0.1669921875, + "learning_rate": 0.001386198188134562, + "loss": 0.127, + "step": 38715 + }, + { + "epoch": 0.33607347158444806, + "grad_norm": 0.2314453125, + "learning_rate": 0.0013861699704491696, + "loss": 0.0854, + "step": 38716 + }, + { + "epoch": 0.3360821520646522, + "grad_norm": 0.5546875, + "learning_rate": 0.0013861417524508267, + "loss": 0.1084, + "step": 38717 + }, + { + "epoch": 0.3360908325448564, + "grad_norm": 0.19140625, + "learning_rate": 0.0013861135341395629, + "loss": 0.1367, + "step": 38718 + }, + { + "epoch": 0.3360995130250605, + "grad_norm": 0.2197265625, + "learning_rate": 0.0013860853155154098, + "loss": 0.1465, + "step": 38719 + }, + { + "epoch": 0.3361081935052647, + "grad_norm": 0.53515625, + "learning_rate": 0.0013860570965783983, + "loss": 0.0698, + "step": 38720 + }, + { + "epoch": 0.33611687398546886, + "grad_norm": 0.3046875, + "learning_rate": 0.0013860288773285588, + "loss": 0.1357, + "step": 38721 + }, + { + "epoch": 0.33612555446567305, + "grad_norm": 0.55859375, + "learning_rate": 0.0013860006577659225, + "loss": 0.1523, + "step": 38722 + }, + { + "epoch": 0.3361342349458772, + "grad_norm": 0.1923828125, + "learning_rate": 0.0013859724378905204, + "loss": 0.0957, + "step": 38723 + }, + { + "epoch": 0.3361429154260814, + "grad_norm": 0.126953125, + "learning_rate": 0.0013859442177023824, + "loss": 0.0933, + "step": 38724 + }, + { + "epoch": 0.3361515959062855, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013859159972015407, + "loss": 0.085, + "step": 38725 + }, + { + "epoch": 0.3361602763864897, + "grad_norm": 0.349609375, + "learning_rate": 0.0013858877763880256, + "loss": 0.103, + "step": 38726 + }, + { + "epoch": 0.33616895686669385, + "grad_norm": 0.07080078125, + "learning_rate": 0.0013858595552618679, + "loss": 0.0669, + "step": 38727 + }, + { + "epoch": 0.33617763734689804, + "grad_norm": 0.431640625, + "learning_rate": 0.0013858313338230984, + "loss": 0.2598, + "step": 38728 + }, + { + "epoch": 0.3361863178271022, + "grad_norm": 0.296875, + "learning_rate": 0.0013858031120717478, + "loss": 0.1611, + "step": 38729 + }, + { + "epoch": 0.33619499830730637, + "grad_norm": 0.1328125, + "learning_rate": 0.0013857748900078476, + "loss": 0.1211, + "step": 38730 + }, + { + "epoch": 0.3362036787875105, + "grad_norm": 0.2236328125, + "learning_rate": 0.001385746667631428, + "loss": 0.1523, + "step": 38731 + }, + { + "epoch": 0.3362123592677147, + "grad_norm": 0.31640625, + "learning_rate": 0.0013857184449425205, + "loss": 0.1147, + "step": 38732 + }, + { + "epoch": 0.33622103974791884, + "grad_norm": 0.119140625, + "learning_rate": 0.0013856902219411549, + "loss": 0.0947, + "step": 38733 + }, + { + "epoch": 0.33622972022812303, + "grad_norm": 0.208984375, + "learning_rate": 0.0013856619986273634, + "loss": 0.1074, + "step": 38734 + }, + { + "epoch": 0.33623840070832717, + "grad_norm": 0.318359375, + "learning_rate": 0.0013856337750011761, + "loss": 0.0859, + "step": 38735 + }, + { + "epoch": 0.33624708118853136, + "grad_norm": 0.1240234375, + "learning_rate": 0.001385605551062624, + "loss": 0.0918, + "step": 38736 + }, + { + "epoch": 0.3362557616687355, + "grad_norm": 0.138671875, + "learning_rate": 0.0013855773268117379, + "loss": 0.1133, + "step": 38737 + }, + { + "epoch": 0.3362644421489397, + "grad_norm": 0.1318359375, + "learning_rate": 0.0013855491022485486, + "loss": 0.083, + "step": 38738 + }, + { + "epoch": 0.33627312262914383, + "grad_norm": 0.349609375, + "learning_rate": 0.0013855208773730873, + "loss": 0.0938, + "step": 38739 + }, + { + "epoch": 0.336281803109348, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013854926521853846, + "loss": 0.1279, + "step": 38740 + }, + { + "epoch": 0.33629048358955216, + "grad_norm": 0.11328125, + "learning_rate": 0.0013854644266854716, + "loss": 0.1201, + "step": 38741 + }, + { + "epoch": 0.33629916406975635, + "grad_norm": 0.19921875, + "learning_rate": 0.0013854362008733784, + "loss": 0.0996, + "step": 38742 + }, + { + "epoch": 0.3363078445499605, + "grad_norm": 0.369140625, + "learning_rate": 0.0013854079747491366, + "loss": 0.1025, + "step": 38743 + }, + { + "epoch": 0.3363165250301647, + "grad_norm": 0.369140625, + "learning_rate": 0.0013853797483127774, + "loss": 0.1133, + "step": 38744 + }, + { + "epoch": 0.3363252055103688, + "grad_norm": 1.046875, + "learning_rate": 0.0013853515215643312, + "loss": 0.1738, + "step": 38745 + }, + { + "epoch": 0.336333885990573, + "grad_norm": 0.11669921875, + "learning_rate": 0.0013853232945038286, + "loss": 0.0737, + "step": 38746 + }, + { + "epoch": 0.33634256647077715, + "grad_norm": 0.34765625, + "learning_rate": 0.0013852950671313007, + "loss": 0.0771, + "step": 38747 + }, + { + "epoch": 0.33635124695098134, + "grad_norm": 0.09228515625, + "learning_rate": 0.0013852668394467784, + "loss": 0.083, + "step": 38748 + }, + { + "epoch": 0.3363599274311855, + "grad_norm": 0.369140625, + "learning_rate": 0.0013852386114502927, + "loss": 0.0923, + "step": 38749 + }, + { + "epoch": 0.33636860791138967, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013852103831418743, + "loss": 0.0718, + "step": 38750 + }, + { + "epoch": 0.3363772883915938, + "grad_norm": 0.341796875, + "learning_rate": 0.001385182154521554, + "loss": 0.0933, + "step": 38751 + }, + { + "epoch": 0.336385968871798, + "grad_norm": 0.1953125, + "learning_rate": 0.001385153925589363, + "loss": 0.1182, + "step": 38752 + }, + { + "epoch": 0.33639464935200214, + "grad_norm": 0.0908203125, + "learning_rate": 0.0013851256963453315, + "loss": 0.1064, + "step": 38753 + }, + { + "epoch": 0.33640332983220633, + "grad_norm": 0.73046875, + "learning_rate": 0.0013850974667894912, + "loss": 0.1201, + "step": 38754 + }, + { + "epoch": 0.33641201031241047, + "grad_norm": 0.69140625, + "learning_rate": 0.0013850692369218726, + "loss": 0.0889, + "step": 38755 + }, + { + "epoch": 0.33642069079261466, + "grad_norm": 0.64453125, + "learning_rate": 0.0013850410067425066, + "loss": 0.0693, + "step": 38756 + }, + { + "epoch": 0.3364293712728188, + "grad_norm": 0.80859375, + "learning_rate": 0.0013850127762514237, + "loss": 0.0679, + "step": 38757 + }, + { + "epoch": 0.336438051753023, + "grad_norm": 0.53515625, + "learning_rate": 0.0013849845454486556, + "loss": 0.0947, + "step": 38758 + }, + { + "epoch": 0.33644673223322713, + "grad_norm": 0.33984375, + "learning_rate": 0.0013849563143342324, + "loss": 0.1055, + "step": 38759 + }, + { + "epoch": 0.3364554127134313, + "grad_norm": 0.16015625, + "learning_rate": 0.0013849280829081852, + "loss": 0.0986, + "step": 38760 + }, + { + "epoch": 0.33646409319363546, + "grad_norm": 0.50390625, + "learning_rate": 0.0013848998511705453, + "loss": 0.0962, + "step": 38761 + }, + { + "epoch": 0.33647277367383965, + "grad_norm": 0.48046875, + "learning_rate": 0.001384871619121343, + "loss": 0.0869, + "step": 38762 + }, + { + "epoch": 0.3364814541540438, + "grad_norm": 0.10498046875, + "learning_rate": 0.001384843386760609, + "loss": 0.1021, + "step": 38763 + }, + { + "epoch": 0.336490134634248, + "grad_norm": 0.11669921875, + "learning_rate": 0.0013848151540883746, + "loss": 0.0635, + "step": 38764 + }, + { + "epoch": 0.3364988151144521, + "grad_norm": 0.396484375, + "learning_rate": 0.001384786921104671, + "loss": 0.1108, + "step": 38765 + }, + { + "epoch": 0.3365074955946563, + "grad_norm": 0.39453125, + "learning_rate": 0.0013847586878095288, + "loss": 0.1191, + "step": 38766 + }, + { + "epoch": 0.33651617607486045, + "grad_norm": 0.30078125, + "learning_rate": 0.0013847304542029784, + "loss": 0.105, + "step": 38767 + }, + { + "epoch": 0.33652485655506464, + "grad_norm": 0.39453125, + "learning_rate": 0.0013847022202850513, + "loss": 0.0737, + "step": 38768 + }, + { + "epoch": 0.3365335370352688, + "grad_norm": 0.1787109375, + "learning_rate": 0.001384673986055778, + "loss": 0.0723, + "step": 38769 + }, + { + "epoch": 0.336542217515473, + "grad_norm": 0.55078125, + "learning_rate": 0.0013846457515151896, + "loss": 0.0918, + "step": 38770 + }, + { + "epoch": 0.3365508979956771, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013846175166633168, + "loss": 0.1064, + "step": 38771 + }, + { + "epoch": 0.3365595784758813, + "grad_norm": 0.59375, + "learning_rate": 0.0013845892815001907, + "loss": 0.1035, + "step": 38772 + }, + { + "epoch": 0.33656825895608544, + "grad_norm": 0.126953125, + "learning_rate": 0.0013845610460258419, + "loss": 0.0986, + "step": 38773 + }, + { + "epoch": 0.33657693943628963, + "grad_norm": 0.10888671875, + "learning_rate": 0.0013845328102403015, + "loss": 0.1113, + "step": 38774 + }, + { + "epoch": 0.3365856199164938, + "grad_norm": 0.26171875, + "learning_rate": 0.0013845045741436, + "loss": 0.1118, + "step": 38775 + }, + { + "epoch": 0.33659430039669797, + "grad_norm": 0.1435546875, + "learning_rate": 0.001384476337735769, + "loss": 0.0859, + "step": 38776 + }, + { + "epoch": 0.3366029808769021, + "grad_norm": 0.1572265625, + "learning_rate": 0.0013844481010168387, + "loss": 0.1191, + "step": 38777 + }, + { + "epoch": 0.3366116613571063, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013844198639868402, + "loss": 0.1465, + "step": 38778 + }, + { + "epoch": 0.33662034183731043, + "grad_norm": 0.259765625, + "learning_rate": 0.0013843916266458047, + "loss": 0.0938, + "step": 38779 + }, + { + "epoch": 0.3366290223175146, + "grad_norm": 0.15625, + "learning_rate": 0.0013843633889937625, + "loss": 0.0918, + "step": 38780 + }, + { + "epoch": 0.33663770279771876, + "grad_norm": 0.1875, + "learning_rate": 0.001384335151030745, + "loss": 0.1152, + "step": 38781 + }, + { + "epoch": 0.33664638327792296, + "grad_norm": 0.107421875, + "learning_rate": 0.0013843069127567823, + "loss": 0.1357, + "step": 38782 + }, + { + "epoch": 0.3366550637581271, + "grad_norm": 0.08154296875, + "learning_rate": 0.0013842786741719065, + "loss": 0.0903, + "step": 38783 + }, + { + "epoch": 0.3366637442383313, + "grad_norm": 0.224609375, + "learning_rate": 0.0013842504352761477, + "loss": 0.0928, + "step": 38784 + }, + { + "epoch": 0.3366724247185354, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013842221960695365, + "loss": 0.0835, + "step": 38785 + }, + { + "epoch": 0.3366811051987396, + "grad_norm": 0.06494140625, + "learning_rate": 0.0013841939565521044, + "loss": 0.0967, + "step": 38786 + }, + { + "epoch": 0.33668978567894375, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013841657167238822, + "loss": 0.1152, + "step": 38787 + }, + { + "epoch": 0.33669846615914795, + "grad_norm": 0.1923828125, + "learning_rate": 0.0013841374765849004, + "loss": 0.1201, + "step": 38788 + }, + { + "epoch": 0.3367071466393521, + "grad_norm": 0.10009765625, + "learning_rate": 0.0013841092361351903, + "loss": 0.085, + "step": 38789 + }, + { + "epoch": 0.3367158271195563, + "grad_norm": 0.1513671875, + "learning_rate": 0.0013840809953747828, + "loss": 0.1445, + "step": 38790 + }, + { + "epoch": 0.3367245075997604, + "grad_norm": 0.1416015625, + "learning_rate": 0.001384052754303708, + "loss": 0.126, + "step": 38791 + }, + { + "epoch": 0.3367331880799646, + "grad_norm": 0.31640625, + "learning_rate": 0.001384024512921998, + "loss": 0.1021, + "step": 38792 + }, + { + "epoch": 0.33674186856016874, + "grad_norm": 0.11767578125, + "learning_rate": 0.0013839962712296828, + "loss": 0.0854, + "step": 38793 + }, + { + "epoch": 0.33675054904037294, + "grad_norm": 0.216796875, + "learning_rate": 0.0013839680292267933, + "loss": 0.1196, + "step": 38794 + }, + { + "epoch": 0.3367592295205771, + "grad_norm": 0.6171875, + "learning_rate": 0.001383939786913361, + "loss": 0.2324, + "step": 38795 + }, + { + "epoch": 0.33676791000078127, + "grad_norm": 0.376953125, + "learning_rate": 0.001383911544289416, + "loss": 0.1387, + "step": 38796 + }, + { + "epoch": 0.3367765904809854, + "grad_norm": 0.310546875, + "learning_rate": 0.0013838833013549903, + "loss": 0.1426, + "step": 38797 + }, + { + "epoch": 0.3367852709611896, + "grad_norm": 0.2255859375, + "learning_rate": 0.0013838550581101135, + "loss": 0.0786, + "step": 38798 + }, + { + "epoch": 0.33679395144139374, + "grad_norm": 0.17578125, + "learning_rate": 0.0013838268145548174, + "loss": 0.1133, + "step": 38799 + }, + { + "epoch": 0.33680263192159793, + "grad_norm": 0.71484375, + "learning_rate": 0.0013837985706891323, + "loss": 0.1318, + "step": 38800 + }, + { + "epoch": 0.33681131240180207, + "grad_norm": 0.10205078125, + "learning_rate": 0.0013837703265130894, + "loss": 0.1133, + "step": 38801 + }, + { + "epoch": 0.33681999288200626, + "grad_norm": 0.140625, + "learning_rate": 0.0013837420820267195, + "loss": 0.123, + "step": 38802 + }, + { + "epoch": 0.3368286733622104, + "grad_norm": 0.5, + "learning_rate": 0.0013837138372300537, + "loss": 0.1309, + "step": 38803 + }, + { + "epoch": 0.3368373538424146, + "grad_norm": 0.23828125, + "learning_rate": 0.0013836855921231226, + "loss": 0.1045, + "step": 38804 + }, + { + "epoch": 0.3368460343226187, + "grad_norm": 0.267578125, + "learning_rate": 0.0013836573467059575, + "loss": 0.0889, + "step": 38805 + }, + { + "epoch": 0.3368547148028229, + "grad_norm": 0.5390625, + "learning_rate": 0.0013836291009785885, + "loss": 0.0928, + "step": 38806 + }, + { + "epoch": 0.33686339528302706, + "grad_norm": 0.11865234375, + "learning_rate": 0.001383600854941047, + "loss": 0.1221, + "step": 38807 + }, + { + "epoch": 0.33687207576323125, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013835726085933643, + "loss": 0.0811, + "step": 38808 + }, + { + "epoch": 0.3368807562434354, + "grad_norm": 0.154296875, + "learning_rate": 0.0013835443619355705, + "loss": 0.1113, + "step": 38809 + }, + { + "epoch": 0.3368894367236396, + "grad_norm": 0.267578125, + "learning_rate": 0.0013835161149676968, + "loss": 0.083, + "step": 38810 + }, + { + "epoch": 0.3368981172038437, + "grad_norm": 0.224609375, + "learning_rate": 0.0013834878676897745, + "loss": 0.0996, + "step": 38811 + }, + { + "epoch": 0.33690679768404785, + "grad_norm": 0.6640625, + "learning_rate": 0.0013834596201018337, + "loss": 0.0977, + "step": 38812 + }, + { + "epoch": 0.33691547816425205, + "grad_norm": 0.2392578125, + "learning_rate": 0.001383431372203906, + "loss": 0.1094, + "step": 38813 + }, + { + "epoch": 0.3369241586444562, + "grad_norm": 0.396484375, + "learning_rate": 0.001383403123996022, + "loss": 0.1143, + "step": 38814 + }, + { + "epoch": 0.3369328391246604, + "grad_norm": 0.2451171875, + "learning_rate": 0.0013833748754782124, + "loss": 0.0898, + "step": 38815 + }, + { + "epoch": 0.3369415196048645, + "grad_norm": 0.65234375, + "learning_rate": 0.0013833466266505083, + "loss": 0.1133, + "step": 38816 + }, + { + "epoch": 0.3369502000850687, + "grad_norm": 0.2138671875, + "learning_rate": 0.001383318377512941, + "loss": 0.0693, + "step": 38817 + }, + { + "epoch": 0.33695888056527284, + "grad_norm": 0.08984375, + "learning_rate": 0.0013832901280655407, + "loss": 0.085, + "step": 38818 + }, + { + "epoch": 0.33696756104547704, + "grad_norm": 0.42578125, + "learning_rate": 0.0013832618783083384, + "loss": 0.1074, + "step": 38819 + }, + { + "epoch": 0.3369762415256812, + "grad_norm": 0.2421875, + "learning_rate": 0.0013832336282413656, + "loss": 0.1025, + "step": 38820 + }, + { + "epoch": 0.33698492200588537, + "grad_norm": 0.482421875, + "learning_rate": 0.0013832053778646523, + "loss": 0.1475, + "step": 38821 + }, + { + "epoch": 0.3369936024860895, + "grad_norm": 0.1865234375, + "learning_rate": 0.0013831771271782304, + "loss": 0.1016, + "step": 38822 + }, + { + "epoch": 0.3370022829662937, + "grad_norm": 0.10986328125, + "learning_rate": 0.0013831488761821297, + "loss": 0.1309, + "step": 38823 + }, + { + "epoch": 0.33701096344649784, + "grad_norm": 0.1796875, + "learning_rate": 0.0013831206248763816, + "loss": 0.0938, + "step": 38824 + }, + { + "epoch": 0.33701964392670203, + "grad_norm": 0.78515625, + "learning_rate": 0.0013830923732610174, + "loss": 0.166, + "step": 38825 + }, + { + "epoch": 0.33702832440690617, + "grad_norm": 0.40625, + "learning_rate": 0.0013830641213360675, + "loss": 0.0977, + "step": 38826 + }, + { + "epoch": 0.33703700488711036, + "grad_norm": 0.193359375, + "learning_rate": 0.001383035869101563, + "loss": 0.0796, + "step": 38827 + }, + { + "epoch": 0.3370456853673145, + "grad_norm": 0.173828125, + "learning_rate": 0.0013830076165575345, + "loss": 0.1104, + "step": 38828 + }, + { + "epoch": 0.3370543658475187, + "grad_norm": 0.50390625, + "learning_rate": 0.0013829793637040136, + "loss": 0.1084, + "step": 38829 + }, + { + "epoch": 0.3370630463277228, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013829511105410304, + "loss": 0.0859, + "step": 38830 + }, + { + "epoch": 0.337071726807927, + "grad_norm": 0.439453125, + "learning_rate": 0.0013829228570686161, + "loss": 0.0645, + "step": 38831 + }, + { + "epoch": 0.33708040728813116, + "grad_norm": 0.0859375, + "learning_rate": 0.0013828946032868018, + "loss": 0.0737, + "step": 38832 + }, + { + "epoch": 0.33708908776833535, + "grad_norm": 0.09765625, + "learning_rate": 0.001382866349195618, + "loss": 0.1309, + "step": 38833 + }, + { + "epoch": 0.3370977682485395, + "grad_norm": 0.2275390625, + "learning_rate": 0.0013828380947950956, + "loss": 0.1289, + "step": 38834 + }, + { + "epoch": 0.3371064487287437, + "grad_norm": 0.3046875, + "learning_rate": 0.001382809840085266, + "loss": 0.0879, + "step": 38835 + }, + { + "epoch": 0.3371151292089478, + "grad_norm": 0.1513671875, + "learning_rate": 0.00138278158506616, + "loss": 0.0713, + "step": 38836 + }, + { + "epoch": 0.337123809689152, + "grad_norm": 0.56640625, + "learning_rate": 0.0013827533297378078, + "loss": 0.1221, + "step": 38837 + }, + { + "epoch": 0.33713249016935615, + "grad_norm": 0.11865234375, + "learning_rate": 0.0013827250741002414, + "loss": 0.1289, + "step": 38838 + }, + { + "epoch": 0.33714117064956034, + "grad_norm": 0.60546875, + "learning_rate": 0.0013826968181534904, + "loss": 0.1016, + "step": 38839 + }, + { + "epoch": 0.3371498511297645, + "grad_norm": 0.1103515625, + "learning_rate": 0.001382668561897587, + "loss": 0.1182, + "step": 38840 + }, + { + "epoch": 0.33715853160996867, + "grad_norm": 0.58984375, + "learning_rate": 0.0013826403053325612, + "loss": 0.1377, + "step": 38841 + }, + { + "epoch": 0.3371672120901728, + "grad_norm": 0.2060546875, + "learning_rate": 0.0013826120484584444, + "loss": 0.0801, + "step": 38842 + }, + { + "epoch": 0.337175892570377, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013825837912752673, + "loss": 0.0923, + "step": 38843 + }, + { + "epoch": 0.33718457305058114, + "grad_norm": 0.55078125, + "learning_rate": 0.0013825555337830607, + "loss": 0.0952, + "step": 38844 + }, + { + "epoch": 0.33719325353078533, + "grad_norm": 0.40625, + "learning_rate": 0.0013825272759818558, + "loss": 0.1309, + "step": 38845 + }, + { + "epoch": 0.33720193401098947, + "grad_norm": 0.076171875, + "learning_rate": 0.0013824990178716832, + "loss": 0.0977, + "step": 38846 + }, + { + "epoch": 0.33721061449119366, + "grad_norm": 0.5234375, + "learning_rate": 0.0013824707594525738, + "loss": 0.1143, + "step": 38847 + }, + { + "epoch": 0.3372192949713978, + "grad_norm": 0.26953125, + "learning_rate": 0.0013824425007245586, + "loss": 0.0825, + "step": 38848 + }, + { + "epoch": 0.337227975451602, + "grad_norm": 0.3984375, + "learning_rate": 0.0013824142416876688, + "loss": 0.105, + "step": 38849 + }, + { + "epoch": 0.33723665593180613, + "grad_norm": 0.1435546875, + "learning_rate": 0.0013823859823419349, + "loss": 0.0801, + "step": 38850 + }, + { + "epoch": 0.3372453364120103, + "grad_norm": 0.302734375, + "learning_rate": 0.0013823577226873879, + "loss": 0.0786, + "step": 38851 + }, + { + "epoch": 0.33725401689221446, + "grad_norm": 0.2021484375, + "learning_rate": 0.0013823294627240588, + "loss": 0.0874, + "step": 38852 + }, + { + "epoch": 0.33726269737241865, + "grad_norm": 0.2080078125, + "learning_rate": 0.0013823012024519786, + "loss": 0.0693, + "step": 38853 + }, + { + "epoch": 0.3372713778526228, + "grad_norm": 0.341796875, + "learning_rate": 0.0013822729418711776, + "loss": 0.1025, + "step": 38854 + }, + { + "epoch": 0.337280058332827, + "grad_norm": 0.515625, + "learning_rate": 0.0013822446809816874, + "loss": 0.0762, + "step": 38855 + }, + { + "epoch": 0.3372887388130311, + "grad_norm": 0.859375, + "learning_rate": 0.0013822164197835388, + "loss": 0.1348, + "step": 38856 + }, + { + "epoch": 0.3372974192932353, + "grad_norm": 0.87109375, + "learning_rate": 0.0013821881582767623, + "loss": 0.0806, + "step": 38857 + }, + { + "epoch": 0.33730609977343945, + "grad_norm": 0.11328125, + "learning_rate": 0.001382159896461389, + "loss": 0.0898, + "step": 38858 + }, + { + "epoch": 0.33731478025364364, + "grad_norm": 0.3046875, + "learning_rate": 0.00138213163433745, + "loss": 0.1045, + "step": 38859 + }, + { + "epoch": 0.3373234607338478, + "grad_norm": 0.1181640625, + "learning_rate": 0.0013821033719049762, + "loss": 0.0918, + "step": 38860 + }, + { + "epoch": 0.337332141214052, + "grad_norm": 0.201171875, + "learning_rate": 0.0013820751091639986, + "loss": 0.1016, + "step": 38861 + }, + { + "epoch": 0.3373408216942561, + "grad_norm": 0.55078125, + "learning_rate": 0.0013820468461145474, + "loss": 0.1328, + "step": 38862 + }, + { + "epoch": 0.3373495021744603, + "grad_norm": 0.283203125, + "learning_rate": 0.0013820185827566542, + "loss": 0.0996, + "step": 38863 + }, + { + "epoch": 0.33735818265466444, + "grad_norm": 0.330078125, + "learning_rate": 0.0013819903190903499, + "loss": 0.0776, + "step": 38864 + }, + { + "epoch": 0.33736686313486863, + "grad_norm": 0.8203125, + "learning_rate": 0.001381962055115665, + "loss": 0.1328, + "step": 38865 + }, + { + "epoch": 0.33737554361507277, + "grad_norm": 0.103515625, + "learning_rate": 0.0013819337908326306, + "loss": 0.0913, + "step": 38866 + }, + { + "epoch": 0.33738422409527696, + "grad_norm": 0.166015625, + "learning_rate": 0.0013819055262412779, + "loss": 0.0811, + "step": 38867 + }, + { + "epoch": 0.3373929045754811, + "grad_norm": 0.40234375, + "learning_rate": 0.0013818772613416372, + "loss": 0.1094, + "step": 38868 + }, + { + "epoch": 0.3374015850556853, + "grad_norm": 0.369140625, + "learning_rate": 0.00138184899613374, + "loss": 0.1191, + "step": 38869 + }, + { + "epoch": 0.33741026553588943, + "grad_norm": 0.275390625, + "learning_rate": 0.001381820730617617, + "loss": 0.0771, + "step": 38870 + }, + { + "epoch": 0.3374189460160936, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013817924647932989, + "loss": 0.1172, + "step": 38871 + }, + { + "epoch": 0.33742762649629776, + "grad_norm": 0.1259765625, + "learning_rate": 0.001381764198660817, + "loss": 0.0957, + "step": 38872 + }, + { + "epoch": 0.33743630697650195, + "grad_norm": 0.10888671875, + "learning_rate": 0.001381735932220202, + "loss": 0.0928, + "step": 38873 + }, + { + "epoch": 0.3374449874567061, + "grad_norm": 0.40625, + "learning_rate": 0.0013817076654714846, + "loss": 0.0898, + "step": 38874 + }, + { + "epoch": 0.3374536679369103, + "grad_norm": 0.3515625, + "learning_rate": 0.0013816793984146962, + "loss": 0.1602, + "step": 38875 + }, + { + "epoch": 0.3374623484171144, + "grad_norm": 0.12353515625, + "learning_rate": 0.0013816511310498672, + "loss": 0.1318, + "step": 38876 + }, + { + "epoch": 0.3374710288973186, + "grad_norm": 0.447265625, + "learning_rate": 0.0013816228633770288, + "loss": 0.0889, + "step": 38877 + }, + { + "epoch": 0.33747970937752275, + "grad_norm": 0.1572265625, + "learning_rate": 0.0013815945953962118, + "loss": 0.1118, + "step": 38878 + }, + { + "epoch": 0.33748838985772694, + "grad_norm": 0.072265625, + "learning_rate": 0.0013815663271074478, + "loss": 0.1055, + "step": 38879 + }, + { + "epoch": 0.3374970703379311, + "grad_norm": 0.248046875, + "learning_rate": 0.0013815380585107662, + "loss": 0.0796, + "step": 38880 + }, + { + "epoch": 0.3375057508181353, + "grad_norm": 0.171875, + "learning_rate": 0.0013815097896061994, + "loss": 0.0933, + "step": 38881 + }, + { + "epoch": 0.3375144312983394, + "grad_norm": 0.197265625, + "learning_rate": 0.0013814815203937777, + "loss": 0.125, + "step": 38882 + }, + { + "epoch": 0.3375231117785436, + "grad_norm": 0.10498046875, + "learning_rate": 0.0013814532508735317, + "loss": 0.0879, + "step": 38883 + }, + { + "epoch": 0.33753179225874774, + "grad_norm": 0.2294921875, + "learning_rate": 0.001381424981045493, + "loss": 0.0884, + "step": 38884 + }, + { + "epoch": 0.33754047273895194, + "grad_norm": 0.60546875, + "learning_rate": 0.0013813967109096918, + "loss": 0.1152, + "step": 38885 + }, + { + "epoch": 0.3375491532191561, + "grad_norm": 0.376953125, + "learning_rate": 0.0013813684404661596, + "loss": 0.1309, + "step": 38886 + }, + { + "epoch": 0.33755783369936027, + "grad_norm": 0.3125, + "learning_rate": 0.001381340169714927, + "loss": 0.1216, + "step": 38887 + }, + { + "epoch": 0.3375665141795644, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013813118986560251, + "loss": 0.085, + "step": 38888 + }, + { + "epoch": 0.3375751946597686, + "grad_norm": 0.302734375, + "learning_rate": 0.0013812836272894845, + "loss": 0.0723, + "step": 38889 + }, + { + "epoch": 0.33758387513997273, + "grad_norm": 0.07275390625, + "learning_rate": 0.001381255355615337, + "loss": 0.1245, + "step": 38890 + }, + { + "epoch": 0.3375925556201769, + "grad_norm": 0.11083984375, + "learning_rate": 0.0013812270836336123, + "loss": 0.1143, + "step": 38891 + }, + { + "epoch": 0.33760123610038106, + "grad_norm": 0.09130859375, + "learning_rate": 0.0013811988113443421, + "loss": 0.1035, + "step": 38892 + }, + { + "epoch": 0.33760991658058526, + "grad_norm": 0.08837890625, + "learning_rate": 0.0013811705387475571, + "loss": 0.0786, + "step": 38893 + }, + { + "epoch": 0.3376185970607894, + "grad_norm": 0.16796875, + "learning_rate": 0.0013811422658432883, + "loss": 0.127, + "step": 38894 + }, + { + "epoch": 0.3376272775409936, + "grad_norm": 0.2333984375, + "learning_rate": 0.0013811139926315664, + "loss": 0.0698, + "step": 38895 + }, + { + "epoch": 0.3376359580211977, + "grad_norm": 0.326171875, + "learning_rate": 0.0013810857191124224, + "loss": 0.0859, + "step": 38896 + }, + { + "epoch": 0.3376446385014019, + "grad_norm": 0.18359375, + "learning_rate": 0.0013810574452858874, + "loss": 0.1016, + "step": 38897 + }, + { + "epoch": 0.33765331898160605, + "grad_norm": 0.1328125, + "learning_rate": 0.001381029171151992, + "loss": 0.083, + "step": 38898 + }, + { + "epoch": 0.33766199946181025, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013810008967107676, + "loss": 0.1172, + "step": 38899 + }, + { + "epoch": 0.3376706799420144, + "grad_norm": 0.146484375, + "learning_rate": 0.0013809726219622448, + "loss": 0.1196, + "step": 38900 + }, + { + "epoch": 0.3376793604222186, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013809443469064544, + "loss": 0.1377, + "step": 38901 + }, + { + "epoch": 0.3376880409024227, + "grad_norm": 0.279296875, + "learning_rate": 0.0013809160715434275, + "loss": 0.0859, + "step": 38902 + }, + { + "epoch": 0.3376967213826269, + "grad_norm": 0.482421875, + "learning_rate": 0.001380887795873195, + "loss": 0.1377, + "step": 38903 + }, + { + "epoch": 0.33770540186283105, + "grad_norm": 0.1298828125, + "learning_rate": 0.001380859519895788, + "loss": 0.0981, + "step": 38904 + }, + { + "epoch": 0.33771408234303524, + "grad_norm": 0.33203125, + "learning_rate": 0.0013808312436112374, + "loss": 0.0593, + "step": 38905 + }, + { + "epoch": 0.3377227628232394, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013808029670195733, + "loss": 0.1152, + "step": 38906 + }, + { + "epoch": 0.33773144330344357, + "grad_norm": 0.08349609375, + "learning_rate": 0.0013807746901208279, + "loss": 0.1094, + "step": 38907 + }, + { + "epoch": 0.3377401237836477, + "grad_norm": 0.7578125, + "learning_rate": 0.0013807464129150313, + "loss": 0.1147, + "step": 38908 + }, + { + "epoch": 0.3377488042638519, + "grad_norm": 0.130859375, + "learning_rate": 0.0013807181354022148, + "loss": 0.0923, + "step": 38909 + }, + { + "epoch": 0.33775748474405604, + "grad_norm": 0.412109375, + "learning_rate": 0.0013806898575824088, + "loss": 0.1611, + "step": 38910 + }, + { + "epoch": 0.33776616522426023, + "grad_norm": 0.10546875, + "learning_rate": 0.0013806615794556447, + "loss": 0.1104, + "step": 38911 + }, + { + "epoch": 0.33777484570446437, + "grad_norm": 0.302734375, + "learning_rate": 0.001380633301021954, + "loss": 0.1016, + "step": 38912 + }, + { + "epoch": 0.33778352618466856, + "grad_norm": 0.10302734375, + "learning_rate": 0.0013806050222813662, + "loss": 0.0747, + "step": 38913 + }, + { + "epoch": 0.3377922066648727, + "grad_norm": 0.078125, + "learning_rate": 0.0013805767432339131, + "loss": 0.1035, + "step": 38914 + }, + { + "epoch": 0.3378008871450769, + "grad_norm": 0.2255859375, + "learning_rate": 0.0013805484638796254, + "loss": 0.1113, + "step": 38915 + }, + { + "epoch": 0.337809567625281, + "grad_norm": 0.2275390625, + "learning_rate": 0.001380520184218534, + "loss": 0.0903, + "step": 38916 + }, + { + "epoch": 0.3378182481054852, + "grad_norm": 0.46875, + "learning_rate": 0.0013804919042506702, + "loss": 0.1123, + "step": 38917 + }, + { + "epoch": 0.33782692858568936, + "grad_norm": 0.39453125, + "learning_rate": 0.0013804636239760648, + "loss": 0.1289, + "step": 38918 + }, + { + "epoch": 0.33783560906589355, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013804353433947486, + "loss": 0.0698, + "step": 38919 + }, + { + "epoch": 0.3378442895460977, + "grad_norm": 0.0830078125, + "learning_rate": 0.001380407062506752, + "loss": 0.1064, + "step": 38920 + }, + { + "epoch": 0.3378529700263019, + "grad_norm": 0.400390625, + "learning_rate": 0.001380378781312107, + "loss": 0.0918, + "step": 38921 + }, + { + "epoch": 0.337861650506506, + "grad_norm": 0.1064453125, + "learning_rate": 0.001380350499810844, + "loss": 0.0938, + "step": 38922 + }, + { + "epoch": 0.3378703309867102, + "grad_norm": 0.1689453125, + "learning_rate": 0.0013803222180029935, + "loss": 0.1191, + "step": 38923 + }, + { + "epoch": 0.33787901146691435, + "grad_norm": 0.197265625, + "learning_rate": 0.001380293935888587, + "loss": 0.1133, + "step": 38924 + }, + { + "epoch": 0.33788769194711854, + "grad_norm": 0.6015625, + "learning_rate": 0.0013802656534676557, + "loss": 0.0713, + "step": 38925 + }, + { + "epoch": 0.3378963724273227, + "grad_norm": 0.177734375, + "learning_rate": 0.0013802373707402296, + "loss": 0.125, + "step": 38926 + }, + { + "epoch": 0.33790505290752687, + "grad_norm": 0.16796875, + "learning_rate": 0.0013802090877063403, + "loss": 0.084, + "step": 38927 + }, + { + "epoch": 0.337913733387731, + "grad_norm": 0.1083984375, + "learning_rate": 0.0013801808043660182, + "loss": 0.1289, + "step": 38928 + }, + { + "epoch": 0.3379224138679352, + "grad_norm": 0.125, + "learning_rate": 0.0013801525207192949, + "loss": 0.0957, + "step": 38929 + }, + { + "epoch": 0.33793109434813934, + "grad_norm": 0.27734375, + "learning_rate": 0.0013801242367662008, + "loss": 0.1143, + "step": 38930 + }, + { + "epoch": 0.33793977482834353, + "grad_norm": 0.47265625, + "learning_rate": 0.0013800959525067675, + "loss": 0.126, + "step": 38931 + }, + { + "epoch": 0.33794845530854767, + "grad_norm": 0.162109375, + "learning_rate": 0.001380067667941025, + "loss": 0.0933, + "step": 38932 + }, + { + "epoch": 0.33795713578875186, + "grad_norm": 0.13671875, + "learning_rate": 0.001380039383069005, + "loss": 0.1074, + "step": 38933 + }, + { + "epoch": 0.337965816268956, + "grad_norm": 0.142578125, + "learning_rate": 0.001380011097890738, + "loss": 0.0977, + "step": 38934 + }, + { + "epoch": 0.33797449674916014, + "grad_norm": 0.2333984375, + "learning_rate": 0.001379982812406255, + "loss": 0.0806, + "step": 38935 + }, + { + "epoch": 0.33798317722936433, + "grad_norm": 0.27734375, + "learning_rate": 0.0013799545266155872, + "loss": 0.1338, + "step": 38936 + }, + { + "epoch": 0.33799185770956847, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013799262405187651, + "loss": 0.0996, + "step": 38937 + }, + { + "epoch": 0.33800053818977266, + "grad_norm": 0.5078125, + "learning_rate": 0.0013798979541158201, + "loss": 0.1289, + "step": 38938 + }, + { + "epoch": 0.3380092186699768, + "grad_norm": 0.1181640625, + "learning_rate": 0.0013798696674067828, + "loss": 0.0996, + "step": 38939 + }, + { + "epoch": 0.338017899150181, + "grad_norm": 0.314453125, + "learning_rate": 0.0013798413803916842, + "loss": 0.2041, + "step": 38940 + }, + { + "epoch": 0.3380265796303851, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013798130930705553, + "loss": 0.0947, + "step": 38941 + }, + { + "epoch": 0.3380352601105893, + "grad_norm": 0.15625, + "learning_rate": 0.0013797848054434272, + "loss": 0.0781, + "step": 38942 + }, + { + "epoch": 0.33804394059079346, + "grad_norm": 0.267578125, + "learning_rate": 0.0013797565175103304, + "loss": 0.1357, + "step": 38943 + }, + { + "epoch": 0.33805262107099765, + "grad_norm": 0.421875, + "learning_rate": 0.001379728229271296, + "loss": 0.0732, + "step": 38944 + }, + { + "epoch": 0.3380613015512018, + "grad_norm": 0.2333984375, + "learning_rate": 0.0013796999407263554, + "loss": 0.1016, + "step": 38945 + }, + { + "epoch": 0.338069982031406, + "grad_norm": 0.83203125, + "learning_rate": 0.0013796716518755386, + "loss": 0.0835, + "step": 38946 + }, + { + "epoch": 0.3380786625116101, + "grad_norm": 0.41015625, + "learning_rate": 0.0013796433627188773, + "loss": 0.1367, + "step": 38947 + }, + { + "epoch": 0.3380873429918143, + "grad_norm": 2.21875, + "learning_rate": 0.0013796150732564025, + "loss": 0.209, + "step": 38948 + }, + { + "epoch": 0.33809602347201845, + "grad_norm": 0.7421875, + "learning_rate": 0.0013795867834881445, + "loss": 0.0996, + "step": 38949 + }, + { + "epoch": 0.33810470395222264, + "grad_norm": 0.25390625, + "learning_rate": 0.001379558493414135, + "loss": 0.1367, + "step": 38950 + }, + { + "epoch": 0.3381133844324268, + "grad_norm": 0.416015625, + "learning_rate": 0.001379530203034404, + "loss": 0.0732, + "step": 38951 + }, + { + "epoch": 0.33812206491263097, + "grad_norm": 0.181640625, + "learning_rate": 0.0013795019123489837, + "loss": 0.1206, + "step": 38952 + }, + { + "epoch": 0.3381307453928351, + "grad_norm": 0.078125, + "learning_rate": 0.0013794736213579035, + "loss": 0.064, + "step": 38953 + }, + { + "epoch": 0.3381394258730393, + "grad_norm": 0.43359375, + "learning_rate": 0.0013794453300611956, + "loss": 0.0908, + "step": 38954 + }, + { + "epoch": 0.33814810635324344, + "grad_norm": 0.2578125, + "learning_rate": 0.0013794170384588901, + "loss": 0.084, + "step": 38955 + }, + { + "epoch": 0.33815678683344763, + "grad_norm": 0.0732421875, + "learning_rate": 0.001379388746551019, + "loss": 0.0737, + "step": 38956 + }, + { + "epoch": 0.33816546731365177, + "grad_norm": 0.10400390625, + "learning_rate": 0.001379360454337612, + "loss": 0.0986, + "step": 38957 + }, + { + "epoch": 0.33817414779385596, + "grad_norm": 0.275390625, + "learning_rate": 0.0013793321618187006, + "loss": 0.1064, + "step": 38958 + }, + { + "epoch": 0.3381828282740601, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013793038689943163, + "loss": 0.0981, + "step": 38959 + }, + { + "epoch": 0.3381915087542643, + "grad_norm": 0.326171875, + "learning_rate": 0.0013792755758644892, + "loss": 0.0801, + "step": 38960 + }, + { + "epoch": 0.33820018923446843, + "grad_norm": 0.181640625, + "learning_rate": 0.0013792472824292507, + "loss": 0.123, + "step": 38961 + }, + { + "epoch": 0.3382088697146726, + "grad_norm": 0.130859375, + "learning_rate": 0.0013792189886886307, + "loss": 0.1089, + "step": 38962 + }, + { + "epoch": 0.33821755019487676, + "grad_norm": 0.1435546875, + "learning_rate": 0.001379190694642662, + "loss": 0.0669, + "step": 38963 + }, + { + "epoch": 0.33822623067508095, + "grad_norm": 0.251953125, + "learning_rate": 0.001379162400291374, + "loss": 0.1006, + "step": 38964 + }, + { + "epoch": 0.3382349111552851, + "grad_norm": 0.142578125, + "learning_rate": 0.0013791341056347984, + "loss": 0.1001, + "step": 38965 + }, + { + "epoch": 0.3382435916354893, + "grad_norm": 0.08544921875, + "learning_rate": 0.0013791058106729656, + "loss": 0.0801, + "step": 38966 + }, + { + "epoch": 0.3382522721156934, + "grad_norm": 0.14453125, + "learning_rate": 0.0013790775154059074, + "loss": 0.0972, + "step": 38967 + }, + { + "epoch": 0.3382609525958976, + "grad_norm": 0.37890625, + "learning_rate": 0.0013790492198336541, + "loss": 0.1641, + "step": 38968 + }, + { + "epoch": 0.33826963307610175, + "grad_norm": 0.474609375, + "learning_rate": 0.0013790209239562365, + "loss": 0.0742, + "step": 38969 + }, + { + "epoch": 0.33827831355630594, + "grad_norm": 0.294921875, + "learning_rate": 0.0013789926277736862, + "loss": 0.1045, + "step": 38970 + }, + { + "epoch": 0.3382869940365101, + "grad_norm": 0.1826171875, + "learning_rate": 0.0013789643312860332, + "loss": 0.0977, + "step": 38971 + }, + { + "epoch": 0.3382956745167143, + "grad_norm": 0.451171875, + "learning_rate": 0.0013789360344933096, + "loss": 0.123, + "step": 38972 + }, + { + "epoch": 0.3383043549969184, + "grad_norm": 0.6796875, + "learning_rate": 0.0013789077373955452, + "loss": 0.2227, + "step": 38973 + }, + { + "epoch": 0.3383130354771226, + "grad_norm": 0.2021484375, + "learning_rate": 0.001378879439992772, + "loss": 0.0811, + "step": 38974 + }, + { + "epoch": 0.33832171595732674, + "grad_norm": 0.54296875, + "learning_rate": 0.00137885114228502, + "loss": 0.127, + "step": 38975 + }, + { + "epoch": 0.33833039643753093, + "grad_norm": 0.07861328125, + "learning_rate": 0.001378822844272321, + "loss": 0.0654, + "step": 38976 + }, + { + "epoch": 0.33833907691773507, + "grad_norm": 0.6953125, + "learning_rate": 0.001378794545954705, + "loss": 0.1309, + "step": 38977 + }, + { + "epoch": 0.33834775739793926, + "grad_norm": 0.2197265625, + "learning_rate": 0.001378766247332204, + "loss": 0.083, + "step": 38978 + }, + { + "epoch": 0.3383564378781434, + "grad_norm": 0.59765625, + "learning_rate": 0.001378737948404848, + "loss": 0.1191, + "step": 38979 + }, + { + "epoch": 0.3383651183583476, + "grad_norm": 0.46484375, + "learning_rate": 0.0013787096491726686, + "loss": 0.1172, + "step": 38980 + }, + { + "epoch": 0.33837379883855173, + "grad_norm": 0.1162109375, + "learning_rate": 0.0013786813496356968, + "loss": 0.1211, + "step": 38981 + }, + { + "epoch": 0.3383824793187559, + "grad_norm": 0.10888671875, + "learning_rate": 0.0013786530497939629, + "loss": 0.1094, + "step": 38982 + }, + { + "epoch": 0.33839115979896006, + "grad_norm": 0.08984375, + "learning_rate": 0.0013786247496474981, + "loss": 0.0898, + "step": 38983 + }, + { + "epoch": 0.33839984027916425, + "grad_norm": 0.322265625, + "learning_rate": 0.0013785964491963337, + "loss": 0.0728, + "step": 38984 + }, + { + "epoch": 0.3384085207593684, + "grad_norm": 0.5625, + "learning_rate": 0.0013785681484405003, + "loss": 0.1084, + "step": 38985 + }, + { + "epoch": 0.3384172012395726, + "grad_norm": 0.20703125, + "learning_rate": 0.001378539847380029, + "loss": 0.0781, + "step": 38986 + }, + { + "epoch": 0.3384258817197767, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013785115460149507, + "loss": 0.126, + "step": 38987 + }, + { + "epoch": 0.3384345621999809, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013784832443452964, + "loss": 0.0962, + "step": 38988 + }, + { + "epoch": 0.33844324268018505, + "grad_norm": 0.2197265625, + "learning_rate": 0.0013784549423710968, + "loss": 0.0996, + "step": 38989 + }, + { + "epoch": 0.33845192316038925, + "grad_norm": 0.193359375, + "learning_rate": 0.0013784266400923835, + "loss": 0.1157, + "step": 38990 + }, + { + "epoch": 0.3384606036405934, + "grad_norm": 0.1796875, + "learning_rate": 0.0013783983375091867, + "loss": 0.1562, + "step": 38991 + }, + { + "epoch": 0.3384692841207976, + "grad_norm": 0.46484375, + "learning_rate": 0.0013783700346215375, + "loss": 0.0977, + "step": 38992 + }, + { + "epoch": 0.3384779646010017, + "grad_norm": 0.1884765625, + "learning_rate": 0.001378341731429467, + "loss": 0.1064, + "step": 38993 + }, + { + "epoch": 0.3384866450812059, + "grad_norm": 0.515625, + "learning_rate": 0.0013783134279330065, + "loss": 0.0996, + "step": 38994 + }, + { + "epoch": 0.33849532556141004, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013782851241321864, + "loss": 0.1035, + "step": 38995 + }, + { + "epoch": 0.33850400604161424, + "grad_norm": 0.091796875, + "learning_rate": 0.0013782568200270383, + "loss": 0.1045, + "step": 38996 + }, + { + "epoch": 0.3385126865218184, + "grad_norm": 0.27734375, + "learning_rate": 0.001378228515617592, + "loss": 0.0649, + "step": 38997 + }, + { + "epoch": 0.33852136700202257, + "grad_norm": 0.326171875, + "learning_rate": 0.0013782002109038796, + "loss": 0.1104, + "step": 38998 + }, + { + "epoch": 0.3385300474822267, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013781719058859317, + "loss": 0.1143, + "step": 38999 + }, + { + "epoch": 0.3385387279624309, + "grad_norm": 0.203125, + "learning_rate": 0.0013781436005637791, + "loss": 0.0854, + "step": 39000 + }, + { + "epoch": 0.33854740844263503, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013781152949374528, + "loss": 0.0923, + "step": 39001 + }, + { + "epoch": 0.3385560889228392, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013780869890069839, + "loss": 0.0952, + "step": 39002 + }, + { + "epoch": 0.33856476940304336, + "grad_norm": 0.111328125, + "learning_rate": 0.0013780586827724028, + "loss": 0.0776, + "step": 39003 + }, + { + "epoch": 0.33857344988324756, + "grad_norm": 0.10498046875, + "learning_rate": 0.0013780303762337414, + "loss": 0.0889, + "step": 39004 + }, + { + "epoch": 0.3385821303634517, + "grad_norm": 0.244140625, + "learning_rate": 0.0013780020693910297, + "loss": 0.1055, + "step": 39005 + }, + { + "epoch": 0.3385908108436559, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013779737622442996, + "loss": 0.0854, + "step": 39006 + }, + { + "epoch": 0.33859949132386, + "grad_norm": 0.126953125, + "learning_rate": 0.0013779454547935815, + "loss": 0.123, + "step": 39007 + }, + { + "epoch": 0.3386081718040642, + "grad_norm": 0.2255859375, + "learning_rate": 0.0013779171470389064, + "loss": 0.0811, + "step": 39008 + }, + { + "epoch": 0.33861685228426835, + "grad_norm": 0.255859375, + "learning_rate": 0.0013778888389803052, + "loss": 0.0732, + "step": 39009 + }, + { + "epoch": 0.33862553276447255, + "grad_norm": 0.302734375, + "learning_rate": 0.0013778605306178089, + "loss": 0.0801, + "step": 39010 + }, + { + "epoch": 0.3386342132446767, + "grad_norm": 0.53515625, + "learning_rate": 0.0013778322219514486, + "loss": 0.0698, + "step": 39011 + }, + { + "epoch": 0.3386428937248809, + "grad_norm": 0.197265625, + "learning_rate": 0.0013778039129812548, + "loss": 0.0967, + "step": 39012 + }, + { + "epoch": 0.338651574205085, + "grad_norm": 0.09130859375, + "learning_rate": 0.0013777756037072592, + "loss": 0.0845, + "step": 39013 + }, + { + "epoch": 0.3386602546852892, + "grad_norm": 0.451171875, + "learning_rate": 0.0013777472941294922, + "loss": 0.1055, + "step": 39014 + }, + { + "epoch": 0.33866893516549335, + "grad_norm": 0.349609375, + "learning_rate": 0.0013777189842479852, + "loss": 0.1445, + "step": 39015 + }, + { + "epoch": 0.33867761564569754, + "grad_norm": 0.68359375, + "learning_rate": 0.0013776906740627687, + "loss": 0.123, + "step": 39016 + }, + { + "epoch": 0.3386862961259017, + "grad_norm": 0.55859375, + "learning_rate": 0.001377662363573874, + "loss": 0.1157, + "step": 39017 + }, + { + "epoch": 0.33869497660610587, + "grad_norm": 0.515625, + "learning_rate": 0.0013776340527813316, + "loss": 0.0811, + "step": 39018 + }, + { + "epoch": 0.33870365708631, + "grad_norm": 0.125, + "learning_rate": 0.001377605741685173, + "loss": 0.127, + "step": 39019 + }, + { + "epoch": 0.3387123375665142, + "grad_norm": 0.20703125, + "learning_rate": 0.0013775774302854289, + "loss": 0.1348, + "step": 39020 + }, + { + "epoch": 0.33872101804671834, + "grad_norm": 0.2294921875, + "learning_rate": 0.00137754911858213, + "loss": 0.106, + "step": 39021 + }, + { + "epoch": 0.33872969852692253, + "grad_norm": 0.138671875, + "learning_rate": 0.001377520806575308, + "loss": 0.1006, + "step": 39022 + }, + { + "epoch": 0.33873837900712667, + "grad_norm": 0.46875, + "learning_rate": 0.0013774924942649932, + "loss": 0.0908, + "step": 39023 + }, + { + "epoch": 0.33874705948733086, + "grad_norm": 0.7109375, + "learning_rate": 0.0013774641816512165, + "loss": 0.1299, + "step": 39024 + }, + { + "epoch": 0.338755739967535, + "grad_norm": 0.080078125, + "learning_rate": 0.0013774358687340098, + "loss": 0.1001, + "step": 39025 + }, + { + "epoch": 0.3387644204477392, + "grad_norm": 0.5546875, + "learning_rate": 0.001377407555513403, + "loss": 0.1182, + "step": 39026 + }, + { + "epoch": 0.3387731009279433, + "grad_norm": 0.208984375, + "learning_rate": 0.0013773792419894276, + "loss": 0.0967, + "step": 39027 + }, + { + "epoch": 0.3387817814081475, + "grad_norm": 0.134765625, + "learning_rate": 0.001377350928162114, + "loss": 0.1045, + "step": 39028 + }, + { + "epoch": 0.33879046188835166, + "grad_norm": 0.3515625, + "learning_rate": 0.0013773226140314943, + "loss": 0.1104, + "step": 39029 + }, + { + "epoch": 0.33879914236855585, + "grad_norm": 0.314453125, + "learning_rate": 0.0013772942995975983, + "loss": 0.0889, + "step": 39030 + }, + { + "epoch": 0.33880782284876, + "grad_norm": 0.3671875, + "learning_rate": 0.0013772659848604574, + "loss": 0.1426, + "step": 39031 + }, + { + "epoch": 0.3388165033289642, + "grad_norm": 0.35546875, + "learning_rate": 0.0013772376698201027, + "loss": 0.1074, + "step": 39032 + }, + { + "epoch": 0.3388251838091683, + "grad_norm": 0.232421875, + "learning_rate": 0.001377209354476565, + "loss": 0.0947, + "step": 39033 + }, + { + "epoch": 0.3388338642893725, + "grad_norm": 0.19921875, + "learning_rate": 0.0013771810388298756, + "loss": 0.1152, + "step": 39034 + }, + { + "epoch": 0.33884254476957665, + "grad_norm": 0.359375, + "learning_rate": 0.0013771527228800648, + "loss": 0.0825, + "step": 39035 + }, + { + "epoch": 0.33885122524978084, + "grad_norm": 0.052978515625, + "learning_rate": 0.0013771244066271642, + "loss": 0.0439, + "step": 39036 + }, + { + "epoch": 0.338859905729985, + "grad_norm": 0.322265625, + "learning_rate": 0.0013770960900712041, + "loss": 0.1001, + "step": 39037 + }, + { + "epoch": 0.33886858621018917, + "grad_norm": 0.16015625, + "learning_rate": 0.0013770677732122165, + "loss": 0.1162, + "step": 39038 + }, + { + "epoch": 0.3388772666903933, + "grad_norm": 0.11328125, + "learning_rate": 0.0013770394560502313, + "loss": 0.1074, + "step": 39039 + }, + { + "epoch": 0.3388859471705975, + "grad_norm": 0.302734375, + "learning_rate": 0.0013770111385852802, + "loss": 0.0859, + "step": 39040 + }, + { + "epoch": 0.33889462765080164, + "grad_norm": 0.6484375, + "learning_rate": 0.0013769828208173936, + "loss": 0.1445, + "step": 39041 + }, + { + "epoch": 0.33890330813100583, + "grad_norm": 0.408203125, + "learning_rate": 0.001376954502746603, + "loss": 0.105, + "step": 39042 + }, + { + "epoch": 0.33891198861120997, + "grad_norm": 0.5390625, + "learning_rate": 0.001376926184372939, + "loss": 0.1021, + "step": 39043 + }, + { + "epoch": 0.33892066909141416, + "grad_norm": 0.232421875, + "learning_rate": 0.0013768978656964325, + "loss": 0.1289, + "step": 39044 + }, + { + "epoch": 0.3389293495716183, + "grad_norm": 0.345703125, + "learning_rate": 0.0013768695467171147, + "loss": 0.0928, + "step": 39045 + }, + { + "epoch": 0.3389380300518225, + "grad_norm": 0.08544921875, + "learning_rate": 0.0013768412274350167, + "loss": 0.085, + "step": 39046 + }, + { + "epoch": 0.33894671053202663, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013768129078501694, + "loss": 0.0986, + "step": 39047 + }, + { + "epoch": 0.3389553910122308, + "grad_norm": 0.171875, + "learning_rate": 0.0013767845879626036, + "loss": 0.126, + "step": 39048 + }, + { + "epoch": 0.33896407149243496, + "grad_norm": 0.287109375, + "learning_rate": 0.00137675626777235, + "loss": 0.1406, + "step": 39049 + }, + { + "epoch": 0.33897275197263915, + "grad_norm": 0.318359375, + "learning_rate": 0.0013767279472794403, + "loss": 0.0771, + "step": 39050 + }, + { + "epoch": 0.3389814324528433, + "grad_norm": 0.58984375, + "learning_rate": 0.0013766996264839046, + "loss": 0.0815, + "step": 39051 + }, + { + "epoch": 0.3389901129330475, + "grad_norm": 0.115234375, + "learning_rate": 0.001376671305385775, + "loss": 0.1045, + "step": 39052 + }, + { + "epoch": 0.3389987934132516, + "grad_norm": 0.15234375, + "learning_rate": 0.001376642983985081, + "loss": 0.1084, + "step": 39053 + }, + { + "epoch": 0.3390074738934558, + "grad_norm": 0.0830078125, + "learning_rate": 0.0013766146622818552, + "loss": 0.0928, + "step": 39054 + }, + { + "epoch": 0.33901615437365995, + "grad_norm": 0.1826171875, + "learning_rate": 0.0013765863402761275, + "loss": 0.1128, + "step": 39055 + }, + { + "epoch": 0.33902483485386414, + "grad_norm": 0.1904296875, + "learning_rate": 0.001376558017967929, + "loss": 0.1172, + "step": 39056 + }, + { + "epoch": 0.3390335153340683, + "grad_norm": 0.40625, + "learning_rate": 0.0013765296953572909, + "loss": 0.106, + "step": 39057 + }, + { + "epoch": 0.3390421958142724, + "grad_norm": 0.087890625, + "learning_rate": 0.0013765013724442439, + "loss": 0.1021, + "step": 39058 + }, + { + "epoch": 0.3390508762944766, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013764730492288196, + "loss": 0.0898, + "step": 39059 + }, + { + "epoch": 0.33905955677468075, + "grad_norm": 0.234375, + "learning_rate": 0.001376444725711048, + "loss": 0.1201, + "step": 39060 + }, + { + "epoch": 0.33906823725488494, + "grad_norm": 0.59765625, + "learning_rate": 0.001376416401890961, + "loss": 0.0703, + "step": 39061 + }, + { + "epoch": 0.3390769177350891, + "grad_norm": 0.73046875, + "learning_rate": 0.0013763880777685888, + "loss": 0.0815, + "step": 39062 + }, + { + "epoch": 0.33908559821529327, + "grad_norm": 0.2021484375, + "learning_rate": 0.001376359753343963, + "loss": 0.0825, + "step": 39063 + }, + { + "epoch": 0.3390942786954974, + "grad_norm": 0.154296875, + "learning_rate": 0.0013763314286171144, + "loss": 0.1123, + "step": 39064 + }, + { + "epoch": 0.3391029591757016, + "grad_norm": 0.357421875, + "learning_rate": 0.0013763031035880738, + "loss": 0.1123, + "step": 39065 + }, + { + "epoch": 0.33911163965590574, + "grad_norm": 0.1484375, + "learning_rate": 0.0013762747782568722, + "loss": 0.0981, + "step": 39066 + }, + { + "epoch": 0.33912032013610993, + "grad_norm": 0.181640625, + "learning_rate": 0.0013762464526235407, + "loss": 0.1396, + "step": 39067 + }, + { + "epoch": 0.33912900061631407, + "grad_norm": 0.1103515625, + "learning_rate": 0.00137621812668811, + "loss": 0.1089, + "step": 39068 + }, + { + "epoch": 0.33913768109651826, + "grad_norm": 0.328125, + "learning_rate": 0.001376189800450612, + "loss": 0.1221, + "step": 39069 + }, + { + "epoch": 0.3391463615767224, + "grad_norm": 0.291015625, + "learning_rate": 0.0013761614739110763, + "loss": 0.1074, + "step": 39070 + }, + { + "epoch": 0.3391550420569266, + "grad_norm": 2.234375, + "learning_rate": 0.0013761331470695346, + "loss": 0.1436, + "step": 39071 + }, + { + "epoch": 0.33916372253713073, + "grad_norm": 0.37890625, + "learning_rate": 0.0013761048199260183, + "loss": 0.1123, + "step": 39072 + }, + { + "epoch": 0.3391724030173349, + "grad_norm": 0.8671875, + "learning_rate": 0.0013760764924805576, + "loss": 0.1406, + "step": 39073 + }, + { + "epoch": 0.33918108349753906, + "grad_norm": 0.421875, + "learning_rate": 0.0013760481647331838, + "loss": 0.1094, + "step": 39074 + }, + { + "epoch": 0.33918976397774325, + "grad_norm": 0.1103515625, + "learning_rate": 0.001376019836683928, + "loss": 0.0801, + "step": 39075 + }, + { + "epoch": 0.3391984444579474, + "grad_norm": 0.984375, + "learning_rate": 0.001375991508332821, + "loss": 0.1133, + "step": 39076 + }, + { + "epoch": 0.3392071249381516, + "grad_norm": 0.267578125, + "learning_rate": 0.001375963179679894, + "loss": 0.127, + "step": 39077 + }, + { + "epoch": 0.3392158054183557, + "grad_norm": 0.69921875, + "learning_rate": 0.0013759348507251775, + "loss": 0.1187, + "step": 39078 + }, + { + "epoch": 0.3392244858985599, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013759065214687027, + "loss": 0.0903, + "step": 39079 + }, + { + "epoch": 0.33923316637876405, + "grad_norm": 0.1416015625, + "learning_rate": 0.001375878191910501, + "loss": 0.0859, + "step": 39080 + }, + { + "epoch": 0.33924184685896824, + "grad_norm": 0.1572265625, + "learning_rate": 0.0013758498620506027, + "loss": 0.1177, + "step": 39081 + }, + { + "epoch": 0.3392505273391724, + "grad_norm": 0.8046875, + "learning_rate": 0.0013758215318890397, + "loss": 0.0864, + "step": 39082 + }, + { + "epoch": 0.3392592078193766, + "grad_norm": 0.10888671875, + "learning_rate": 0.001375793201425842, + "loss": 0.1299, + "step": 39083 + }, + { + "epoch": 0.3392678882995807, + "grad_norm": 0.27734375, + "learning_rate": 0.0013757648706610411, + "loss": 0.0752, + "step": 39084 + }, + { + "epoch": 0.3392765687797849, + "grad_norm": 0.087890625, + "learning_rate": 0.001375736539594668, + "loss": 0.1069, + "step": 39085 + }, + { + "epoch": 0.33928524925998904, + "grad_norm": 0.212890625, + "learning_rate": 0.0013757082082267536, + "loss": 0.1104, + "step": 39086 + }, + { + "epoch": 0.33929392974019323, + "grad_norm": 0.376953125, + "learning_rate": 0.0013756798765573283, + "loss": 0.1123, + "step": 39087 + }, + { + "epoch": 0.33930261022039737, + "grad_norm": 0.29296875, + "learning_rate": 0.0013756515445864242, + "loss": 0.1094, + "step": 39088 + }, + { + "epoch": 0.33931129070060156, + "grad_norm": 0.279296875, + "learning_rate": 0.0013756232123140716, + "loss": 0.1138, + "step": 39089 + }, + { + "epoch": 0.3393199711808057, + "grad_norm": 0.150390625, + "learning_rate": 0.0013755948797403016, + "loss": 0.1172, + "step": 39090 + }, + { + "epoch": 0.3393286516610099, + "grad_norm": 0.515625, + "learning_rate": 0.0013755665468651451, + "loss": 0.0986, + "step": 39091 + }, + { + "epoch": 0.33933733214121403, + "grad_norm": 0.2890625, + "learning_rate": 0.001375538213688633, + "loss": 0.0874, + "step": 39092 + }, + { + "epoch": 0.3393460126214182, + "grad_norm": 0.498046875, + "learning_rate": 0.0013755098802107966, + "loss": 0.1201, + "step": 39093 + }, + { + "epoch": 0.33935469310162236, + "grad_norm": 0.1455078125, + "learning_rate": 0.001375481546431667, + "loss": 0.1133, + "step": 39094 + }, + { + "epoch": 0.33936337358182656, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013754532123512746, + "loss": 0.1113, + "step": 39095 + }, + { + "epoch": 0.3393720540620307, + "grad_norm": 0.1064453125, + "learning_rate": 0.0013754248779696508, + "loss": 0.0566, + "step": 39096 + }, + { + "epoch": 0.3393807345422349, + "grad_norm": 0.296875, + "learning_rate": 0.0013753965432868262, + "loss": 0.1309, + "step": 39097 + }, + { + "epoch": 0.339389415022439, + "grad_norm": 0.05859375, + "learning_rate": 0.0013753682083028325, + "loss": 0.0786, + "step": 39098 + }, + { + "epoch": 0.3393980955026432, + "grad_norm": 0.232421875, + "learning_rate": 0.0013753398730177002, + "loss": 0.0742, + "step": 39099 + }, + { + "epoch": 0.33940677598284735, + "grad_norm": 0.484375, + "learning_rate": 0.00137531153743146, + "loss": 0.1182, + "step": 39100 + }, + { + "epoch": 0.33941545646305155, + "grad_norm": 0.353515625, + "learning_rate": 0.0013752832015441437, + "loss": 0.1187, + "step": 39101 + }, + { + "epoch": 0.3394241369432557, + "grad_norm": 0.32421875, + "learning_rate": 0.0013752548653557816, + "loss": 0.1309, + "step": 39102 + }, + { + "epoch": 0.3394328174234599, + "grad_norm": 0.1357421875, + "learning_rate": 0.0013752265288664049, + "loss": 0.1011, + "step": 39103 + }, + { + "epoch": 0.339441497903664, + "grad_norm": 0.234375, + "learning_rate": 0.0013751981920760447, + "loss": 0.0972, + "step": 39104 + }, + { + "epoch": 0.3394501783838682, + "grad_norm": 0.1875, + "learning_rate": 0.001375169854984732, + "loss": 0.1016, + "step": 39105 + }, + { + "epoch": 0.33945885886407234, + "grad_norm": 0.2890625, + "learning_rate": 0.0013751415175924976, + "loss": 0.0635, + "step": 39106 + }, + { + "epoch": 0.33946753934427654, + "grad_norm": 0.359375, + "learning_rate": 0.0013751131798993724, + "loss": 0.0811, + "step": 39107 + }, + { + "epoch": 0.3394762198244807, + "grad_norm": 0.55859375, + "learning_rate": 0.0013750848419053876, + "loss": 0.0825, + "step": 39108 + }, + { + "epoch": 0.33948490030468487, + "grad_norm": 0.251953125, + "learning_rate": 0.001375056503610574, + "loss": 0.0752, + "step": 39109 + }, + { + "epoch": 0.339493580784889, + "grad_norm": 0.96875, + "learning_rate": 0.0013750281650149632, + "loss": 0.1055, + "step": 39110 + }, + { + "epoch": 0.3395022612650932, + "grad_norm": 0.1943359375, + "learning_rate": 0.0013749998261185854, + "loss": 0.1523, + "step": 39111 + }, + { + "epoch": 0.33951094174529733, + "grad_norm": 0.146484375, + "learning_rate": 0.001374971486921472, + "loss": 0.0713, + "step": 39112 + }, + { + "epoch": 0.3395196222255015, + "grad_norm": 0.3203125, + "learning_rate": 0.001374943147423654, + "loss": 0.1533, + "step": 39113 + }, + { + "epoch": 0.33952830270570566, + "grad_norm": 0.33203125, + "learning_rate": 0.0013749148076251621, + "loss": 0.1406, + "step": 39114 + }, + { + "epoch": 0.33953698318590986, + "grad_norm": 0.1630859375, + "learning_rate": 0.0013748864675260277, + "loss": 0.0967, + "step": 39115 + }, + { + "epoch": 0.339545663666114, + "grad_norm": 0.1142578125, + "learning_rate": 0.0013748581271262816, + "loss": 0.1299, + "step": 39116 + }, + { + "epoch": 0.3395543441463182, + "grad_norm": 0.400390625, + "learning_rate": 0.0013748297864259545, + "loss": 0.1367, + "step": 39117 + }, + { + "epoch": 0.3395630246265223, + "grad_norm": 0.546875, + "learning_rate": 0.0013748014454250776, + "loss": 0.1416, + "step": 39118 + }, + { + "epoch": 0.3395717051067265, + "grad_norm": 0.328125, + "learning_rate": 0.0013747731041236825, + "loss": 0.1602, + "step": 39119 + }, + { + "epoch": 0.33958038558693066, + "grad_norm": 0.310546875, + "learning_rate": 0.0013747447625217994, + "loss": 0.1025, + "step": 39120 + }, + { + "epoch": 0.33958906606713485, + "grad_norm": 0.51171875, + "learning_rate": 0.0013747164206194596, + "loss": 0.0957, + "step": 39121 + }, + { + "epoch": 0.339597746547339, + "grad_norm": 0.46484375, + "learning_rate": 0.001374688078416694, + "loss": 0.1094, + "step": 39122 + }, + { + "epoch": 0.3396064270275432, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013746597359135338, + "loss": 0.0957, + "step": 39123 + }, + { + "epoch": 0.3396151075077473, + "grad_norm": 0.09765625, + "learning_rate": 0.0013746313931100097, + "loss": 0.1011, + "step": 39124 + }, + { + "epoch": 0.3396237879879515, + "grad_norm": 0.1728515625, + "learning_rate": 0.0013746030500061527, + "loss": 0.0815, + "step": 39125 + }, + { + "epoch": 0.33963246846815565, + "grad_norm": 0.07861328125, + "learning_rate": 0.001374574706601994, + "loss": 0.0981, + "step": 39126 + }, + { + "epoch": 0.33964114894835984, + "grad_norm": 0.1015625, + "learning_rate": 0.0013745463628975648, + "loss": 0.1084, + "step": 39127 + }, + { + "epoch": 0.339649829428564, + "grad_norm": 0.08251953125, + "learning_rate": 0.0013745180188928954, + "loss": 0.0898, + "step": 39128 + }, + { + "epoch": 0.33965850990876817, + "grad_norm": 0.375, + "learning_rate": 0.0013744896745880174, + "loss": 0.0918, + "step": 39129 + }, + { + "epoch": 0.3396671903889723, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013744613299829615, + "loss": 0.0767, + "step": 39130 + }, + { + "epoch": 0.3396758708691765, + "grad_norm": 0.33984375, + "learning_rate": 0.001374432985077759, + "loss": 0.0947, + "step": 39131 + }, + { + "epoch": 0.33968455134938064, + "grad_norm": 0.46484375, + "learning_rate": 0.0013744046398724405, + "loss": 0.127, + "step": 39132 + }, + { + "epoch": 0.33969323182958483, + "grad_norm": 0.431640625, + "learning_rate": 0.0013743762943670378, + "loss": 0.0762, + "step": 39133 + }, + { + "epoch": 0.33970191230978897, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013743479485615808, + "loss": 0.1182, + "step": 39134 + }, + { + "epoch": 0.33971059278999316, + "grad_norm": 0.3203125, + "learning_rate": 0.0013743196024561008, + "loss": 0.1309, + "step": 39135 + }, + { + "epoch": 0.3397192732701973, + "grad_norm": 0.39453125, + "learning_rate": 0.0013742912560506296, + "loss": 0.1099, + "step": 39136 + }, + { + "epoch": 0.3397279537504015, + "grad_norm": 0.08837890625, + "learning_rate": 0.001374262909345197, + "loss": 0.063, + "step": 39137 + }, + { + "epoch": 0.3397366342306056, + "grad_norm": 0.482421875, + "learning_rate": 0.0013742345623398349, + "loss": 0.1084, + "step": 39138 + }, + { + "epoch": 0.3397453147108098, + "grad_norm": 0.146484375, + "learning_rate": 0.001374206215034574, + "loss": 0.1172, + "step": 39139 + }, + { + "epoch": 0.33975399519101396, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013741778674294454, + "loss": 0.1309, + "step": 39140 + }, + { + "epoch": 0.33976267567121815, + "grad_norm": 0.189453125, + "learning_rate": 0.0013741495195244797, + "loss": 0.0879, + "step": 39141 + }, + { + "epoch": 0.3397713561514223, + "grad_norm": 0.44140625, + "learning_rate": 0.0013741211713197087, + "loss": 0.0859, + "step": 39142 + }, + { + "epoch": 0.3397800366316265, + "grad_norm": 0.2421875, + "learning_rate": 0.0013740928228151625, + "loss": 0.0938, + "step": 39143 + }, + { + "epoch": 0.3397887171118306, + "grad_norm": 0.37109375, + "learning_rate": 0.0013740644740108728, + "loss": 0.0889, + "step": 39144 + }, + { + "epoch": 0.3397973975920348, + "grad_norm": 0.333984375, + "learning_rate": 0.00137403612490687, + "loss": 0.1196, + "step": 39145 + }, + { + "epoch": 0.33980607807223895, + "grad_norm": 0.130859375, + "learning_rate": 0.0013740077755031855, + "loss": 0.105, + "step": 39146 + }, + { + "epoch": 0.33981475855244314, + "grad_norm": 0.37890625, + "learning_rate": 0.0013739794257998502, + "loss": 0.1357, + "step": 39147 + }, + { + "epoch": 0.3398234390326473, + "grad_norm": 0.125, + "learning_rate": 0.001373951075796895, + "loss": 0.1572, + "step": 39148 + }, + { + "epoch": 0.33983211951285147, + "grad_norm": 0.1494140625, + "learning_rate": 0.001373922725494351, + "loss": 0.1221, + "step": 39149 + }, + { + "epoch": 0.3398407999930556, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013738943748922497, + "loss": 0.0801, + "step": 39150 + }, + { + "epoch": 0.3398494804732598, + "grad_norm": 0.2890625, + "learning_rate": 0.0013738660239906216, + "loss": 0.1143, + "step": 39151 + }, + { + "epoch": 0.33985816095346394, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013738376727894972, + "loss": 0.0957, + "step": 39152 + }, + { + "epoch": 0.33986684143366813, + "grad_norm": 0.486328125, + "learning_rate": 0.0013738093212889082, + "loss": 0.1484, + "step": 39153 + }, + { + "epoch": 0.33987552191387227, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013737809694888859, + "loss": 0.1172, + "step": 39154 + }, + { + "epoch": 0.33988420239407646, + "grad_norm": 0.11669921875, + "learning_rate": 0.0013737526173894603, + "loss": 0.1089, + "step": 39155 + }, + { + "epoch": 0.3398928828742806, + "grad_norm": 0.1298828125, + "learning_rate": 0.001373724264990663, + "loss": 0.0801, + "step": 39156 + }, + { + "epoch": 0.3399015633544848, + "grad_norm": 0.291015625, + "learning_rate": 0.001373695912292525, + "loss": 0.0967, + "step": 39157 + }, + { + "epoch": 0.33991024383468893, + "grad_norm": 0.373046875, + "learning_rate": 0.0013736675592950773, + "loss": 0.1289, + "step": 39158 + }, + { + "epoch": 0.3399189243148931, + "grad_norm": 0.232421875, + "learning_rate": 0.001373639205998351, + "loss": 0.1172, + "step": 39159 + }, + { + "epoch": 0.33992760479509726, + "grad_norm": 0.333984375, + "learning_rate": 0.001373610852402377, + "loss": 0.0928, + "step": 39160 + }, + { + "epoch": 0.33993628527530145, + "grad_norm": 0.68359375, + "learning_rate": 0.001373582498507186, + "loss": 0.0859, + "step": 39161 + }, + { + "epoch": 0.3399449657555056, + "grad_norm": 0.12060546875, + "learning_rate": 0.0013735541443128097, + "loss": 0.1289, + "step": 39162 + }, + { + "epoch": 0.3399536462357098, + "grad_norm": 0.12451171875, + "learning_rate": 0.0013735257898192786, + "loss": 0.085, + "step": 39163 + }, + { + "epoch": 0.3399623267159139, + "grad_norm": 0.314453125, + "learning_rate": 0.0013734974350266237, + "loss": 0.1245, + "step": 39164 + }, + { + "epoch": 0.3399710071961181, + "grad_norm": 0.55078125, + "learning_rate": 0.001373469079934876, + "loss": 0.1069, + "step": 39165 + }, + { + "epoch": 0.33997968767632225, + "grad_norm": 0.166015625, + "learning_rate": 0.0013734407245440665, + "loss": 0.1216, + "step": 39166 + }, + { + "epoch": 0.33998836815652644, + "grad_norm": 0.3515625, + "learning_rate": 0.0013734123688542265, + "loss": 0.1104, + "step": 39167 + }, + { + "epoch": 0.3399970486367306, + "grad_norm": 0.27734375, + "learning_rate": 0.0013733840128653872, + "loss": 0.0786, + "step": 39168 + }, + { + "epoch": 0.3400057291169348, + "grad_norm": 0.087890625, + "learning_rate": 0.001373355656577579, + "loss": 0.1152, + "step": 39169 + }, + { + "epoch": 0.3400144095971389, + "grad_norm": 0.27734375, + "learning_rate": 0.0013733272999908329, + "loss": 0.1123, + "step": 39170 + }, + { + "epoch": 0.3400230900773431, + "grad_norm": 0.32421875, + "learning_rate": 0.0013732989431051803, + "loss": 0.0903, + "step": 39171 + }, + { + "epoch": 0.34003177055754724, + "grad_norm": 0.09423828125, + "learning_rate": 0.0013732705859206523, + "loss": 0.0781, + "step": 39172 + }, + { + "epoch": 0.34004045103775143, + "grad_norm": 0.10986328125, + "learning_rate": 0.0013732422284372797, + "loss": 0.1113, + "step": 39173 + }, + { + "epoch": 0.34004913151795557, + "grad_norm": 0.173828125, + "learning_rate": 0.0013732138706550932, + "loss": 0.1387, + "step": 39174 + }, + { + "epoch": 0.34005781199815976, + "grad_norm": 0.8359375, + "learning_rate": 0.0013731855125741243, + "loss": 0.1484, + "step": 39175 + }, + { + "epoch": 0.3400664924783639, + "grad_norm": 0.1767578125, + "learning_rate": 0.0013731571541944036, + "loss": 0.104, + "step": 39176 + }, + { + "epoch": 0.3400751729585681, + "grad_norm": 0.224609375, + "learning_rate": 0.0013731287955159625, + "loss": 0.0879, + "step": 39177 + }, + { + "epoch": 0.34008385343877223, + "grad_norm": 0.224609375, + "learning_rate": 0.0013731004365388317, + "loss": 0.1011, + "step": 39178 + }, + { + "epoch": 0.3400925339189764, + "grad_norm": 0.484375, + "learning_rate": 0.0013730720772630427, + "loss": 0.1201, + "step": 39179 + }, + { + "epoch": 0.34010121439918056, + "grad_norm": 0.4140625, + "learning_rate": 0.0013730437176886264, + "loss": 0.0811, + "step": 39180 + }, + { + "epoch": 0.3401098948793847, + "grad_norm": 0.21484375, + "learning_rate": 0.0013730153578156133, + "loss": 0.0913, + "step": 39181 + }, + { + "epoch": 0.3401185753595889, + "grad_norm": 0.169921875, + "learning_rate": 0.0013729869976440343, + "loss": 0.1045, + "step": 39182 + }, + { + "epoch": 0.34012725583979303, + "grad_norm": 0.203125, + "learning_rate": 0.0013729586371739213, + "loss": 0.1191, + "step": 39183 + }, + { + "epoch": 0.3401359363199972, + "grad_norm": 0.28125, + "learning_rate": 0.0013729302764053047, + "loss": 0.0947, + "step": 39184 + }, + { + "epoch": 0.34014461680020136, + "grad_norm": 0.0869140625, + "learning_rate": 0.0013729019153382153, + "loss": 0.0908, + "step": 39185 + }, + { + "epoch": 0.34015329728040555, + "grad_norm": 0.470703125, + "learning_rate": 0.0013728735539726851, + "loss": 0.0908, + "step": 39186 + }, + { + "epoch": 0.3401619777606097, + "grad_norm": 0.330078125, + "learning_rate": 0.001372845192308744, + "loss": 0.083, + "step": 39187 + }, + { + "epoch": 0.3401706582408139, + "grad_norm": 0.2734375, + "learning_rate": 0.0013728168303464238, + "loss": 0.082, + "step": 39188 + }, + { + "epoch": 0.340179338721018, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013727884680857553, + "loss": 0.1216, + "step": 39189 + }, + { + "epoch": 0.3401880192012222, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013727601055267692, + "loss": 0.1123, + "step": 39190 + }, + { + "epoch": 0.34019669968142635, + "grad_norm": 0.134765625, + "learning_rate": 0.0013727317426694967, + "loss": 0.1416, + "step": 39191 + }, + { + "epoch": 0.34020538016163054, + "grad_norm": 0.48046875, + "learning_rate": 0.0013727033795139687, + "loss": 0.1289, + "step": 39192 + }, + { + "epoch": 0.3402140606418347, + "grad_norm": 0.19140625, + "learning_rate": 0.001372675016060217, + "loss": 0.1387, + "step": 39193 + }, + { + "epoch": 0.3402227411220389, + "grad_norm": 0.47265625, + "learning_rate": 0.0013726466523082714, + "loss": 0.1167, + "step": 39194 + }, + { + "epoch": 0.340231421602243, + "grad_norm": 0.404296875, + "learning_rate": 0.0013726182882581638, + "loss": 0.1045, + "step": 39195 + }, + { + "epoch": 0.3402401020824472, + "grad_norm": 0.22265625, + "learning_rate": 0.0013725899239099247, + "loss": 0.0757, + "step": 39196 + }, + { + "epoch": 0.34024878256265134, + "grad_norm": 0.48828125, + "learning_rate": 0.0013725615592635855, + "loss": 0.106, + "step": 39197 + }, + { + "epoch": 0.34025746304285553, + "grad_norm": 0.2451171875, + "learning_rate": 0.0013725331943191773, + "loss": 0.0898, + "step": 39198 + }, + { + "epoch": 0.34026614352305967, + "grad_norm": 0.28515625, + "learning_rate": 0.0013725048290767307, + "loss": 0.1104, + "step": 39199 + }, + { + "epoch": 0.34027482400326386, + "grad_norm": 0.361328125, + "learning_rate": 0.0013724764635362767, + "loss": 0.1138, + "step": 39200 + }, + { + "epoch": 0.340283504483468, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013724480976978468, + "loss": 0.0898, + "step": 39201 + }, + { + "epoch": 0.3402921849636722, + "grad_norm": 0.310546875, + "learning_rate": 0.001372419731561472, + "loss": 0.0913, + "step": 39202 + }, + { + "epoch": 0.34030086544387633, + "grad_norm": 0.15625, + "learning_rate": 0.0013723913651271823, + "loss": 0.0718, + "step": 39203 + }, + { + "epoch": 0.3403095459240805, + "grad_norm": 0.189453125, + "learning_rate": 0.0013723629983950101, + "loss": 0.0713, + "step": 39204 + }, + { + "epoch": 0.34031822640428466, + "grad_norm": 0.123046875, + "learning_rate": 0.0013723346313649857, + "loss": 0.127, + "step": 39205 + }, + { + "epoch": 0.34032690688448886, + "grad_norm": 0.30078125, + "learning_rate": 0.0013723062640371405, + "loss": 0.1738, + "step": 39206 + }, + { + "epoch": 0.340335587364693, + "grad_norm": 0.361328125, + "learning_rate": 0.0013722778964115048, + "loss": 0.1094, + "step": 39207 + }, + { + "epoch": 0.3403442678448972, + "grad_norm": 0.302734375, + "learning_rate": 0.0013722495284881103, + "loss": 0.0752, + "step": 39208 + }, + { + "epoch": 0.3403529483251013, + "grad_norm": 0.1826171875, + "learning_rate": 0.0013722211602669877, + "loss": 0.0986, + "step": 39209 + }, + { + "epoch": 0.3403616288053055, + "grad_norm": 0.447265625, + "learning_rate": 0.0013721927917481683, + "loss": 0.0845, + "step": 39210 + }, + { + "epoch": 0.34037030928550965, + "grad_norm": 0.1728515625, + "learning_rate": 0.0013721644229316829, + "loss": 0.1182, + "step": 39211 + }, + { + "epoch": 0.34037898976571385, + "grad_norm": 1.0546875, + "learning_rate": 0.0013721360538175627, + "loss": 0.1719, + "step": 39212 + }, + { + "epoch": 0.340387670245918, + "grad_norm": 0.10009765625, + "learning_rate": 0.0013721076844058382, + "loss": 0.1387, + "step": 39213 + }, + { + "epoch": 0.3403963507261222, + "grad_norm": 0.34375, + "learning_rate": 0.001372079314696541, + "loss": 0.1562, + "step": 39214 + }, + { + "epoch": 0.3404050312063263, + "grad_norm": 0.4296875, + "learning_rate": 0.001372050944689702, + "loss": 0.1172, + "step": 39215 + }, + { + "epoch": 0.3404137116865305, + "grad_norm": 0.12890625, + "learning_rate": 0.0013720225743853523, + "loss": 0.0669, + "step": 39216 + }, + { + "epoch": 0.34042239216673464, + "grad_norm": 0.1328125, + "learning_rate": 0.0013719942037835227, + "loss": 0.1143, + "step": 39217 + }, + { + "epoch": 0.34043107264693884, + "grad_norm": 0.08740234375, + "learning_rate": 0.0013719658328842442, + "loss": 0.0898, + "step": 39218 + }, + { + "epoch": 0.340439753127143, + "grad_norm": 0.314453125, + "learning_rate": 0.0013719374616875482, + "loss": 0.1123, + "step": 39219 + }, + { + "epoch": 0.34044843360734717, + "grad_norm": 0.19140625, + "learning_rate": 0.0013719090901934653, + "loss": 0.0957, + "step": 39220 + }, + { + "epoch": 0.3404571140875513, + "grad_norm": 0.267578125, + "learning_rate": 0.0013718807184020265, + "loss": 0.085, + "step": 39221 + }, + { + "epoch": 0.3404657945677555, + "grad_norm": 0.138671875, + "learning_rate": 0.0013718523463132635, + "loss": 0.1309, + "step": 39222 + }, + { + "epoch": 0.34047447504795963, + "grad_norm": 0.15625, + "learning_rate": 0.0013718239739272066, + "loss": 0.0869, + "step": 39223 + }, + { + "epoch": 0.34048315552816383, + "grad_norm": 0.54296875, + "learning_rate": 0.001371795601243887, + "loss": 0.1309, + "step": 39224 + }, + { + "epoch": 0.34049183600836797, + "grad_norm": 0.1796875, + "learning_rate": 0.001371767228263336, + "loss": 0.082, + "step": 39225 + }, + { + "epoch": 0.34050051648857216, + "grad_norm": 0.11376953125, + "learning_rate": 0.0013717388549855842, + "loss": 0.082, + "step": 39226 + }, + { + "epoch": 0.3405091969687763, + "grad_norm": 0.169921875, + "learning_rate": 0.0013717104814106632, + "loss": 0.1084, + "step": 39227 + }, + { + "epoch": 0.3405178774489805, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013716821075386034, + "loss": 0.0977, + "step": 39228 + }, + { + "epoch": 0.3405265579291846, + "grad_norm": 0.10791015625, + "learning_rate": 0.0013716537333694366, + "loss": 0.0723, + "step": 39229 + }, + { + "epoch": 0.3405352384093888, + "grad_norm": 0.2294921875, + "learning_rate": 0.0013716253589031928, + "loss": 0.0938, + "step": 39230 + }, + { + "epoch": 0.34054391888959296, + "grad_norm": 0.357421875, + "learning_rate": 0.0013715969841399038, + "loss": 0.1182, + "step": 39231 + }, + { + "epoch": 0.34055259936979715, + "grad_norm": 0.07666015625, + "learning_rate": 0.0013715686090796004, + "loss": 0.0942, + "step": 39232 + }, + { + "epoch": 0.3405612798500013, + "grad_norm": 0.236328125, + "learning_rate": 0.001371540233722314, + "loss": 0.0747, + "step": 39233 + }, + { + "epoch": 0.3405699603302055, + "grad_norm": 0.55078125, + "learning_rate": 0.0013715118580680744, + "loss": 0.1064, + "step": 39234 + }, + { + "epoch": 0.3405786408104096, + "grad_norm": 0.2314453125, + "learning_rate": 0.0013714834821169145, + "loss": 0.127, + "step": 39235 + }, + { + "epoch": 0.3405873212906138, + "grad_norm": 0.453125, + "learning_rate": 0.0013714551058688637, + "loss": 0.0903, + "step": 39236 + }, + { + "epoch": 0.34059600177081795, + "grad_norm": 0.29296875, + "learning_rate": 0.001371426729323954, + "loss": 0.0752, + "step": 39237 + }, + { + "epoch": 0.34060468225102214, + "grad_norm": 0.240234375, + "learning_rate": 0.001371398352482216, + "loss": 0.0728, + "step": 39238 + }, + { + "epoch": 0.3406133627312263, + "grad_norm": 0.259765625, + "learning_rate": 0.0013713699753436807, + "loss": 0.0806, + "step": 39239 + }, + { + "epoch": 0.34062204321143047, + "grad_norm": 0.251953125, + "learning_rate": 0.0013713415979083796, + "loss": 0.1104, + "step": 39240 + }, + { + "epoch": 0.3406307236916346, + "grad_norm": 0.328125, + "learning_rate": 0.0013713132201763432, + "loss": 0.1025, + "step": 39241 + }, + { + "epoch": 0.3406394041718388, + "grad_norm": 0.146484375, + "learning_rate": 0.0013712848421476027, + "loss": 0.0918, + "step": 39242 + }, + { + "epoch": 0.34064808465204294, + "grad_norm": 0.5859375, + "learning_rate": 0.0013712564638221892, + "loss": 0.123, + "step": 39243 + }, + { + "epoch": 0.34065676513224713, + "grad_norm": 0.265625, + "learning_rate": 0.0013712280852001338, + "loss": 0.1172, + "step": 39244 + }, + { + "epoch": 0.34066544561245127, + "grad_norm": 0.38671875, + "learning_rate": 0.0013711997062814674, + "loss": 0.1299, + "step": 39245 + }, + { + "epoch": 0.34067412609265546, + "grad_norm": 0.09326171875, + "learning_rate": 0.0013711713270662211, + "loss": 0.1416, + "step": 39246 + }, + { + "epoch": 0.3406828065728596, + "grad_norm": 0.5234375, + "learning_rate": 0.0013711429475544259, + "loss": 0.123, + "step": 39247 + }, + { + "epoch": 0.3406914870530638, + "grad_norm": 0.07275390625, + "learning_rate": 0.0013711145677461128, + "loss": 0.0713, + "step": 39248 + }, + { + "epoch": 0.34070016753326793, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013710861876413127, + "loss": 0.0889, + "step": 39249 + }, + { + "epoch": 0.3407088480134721, + "grad_norm": 0.380859375, + "learning_rate": 0.0013710578072400573, + "loss": 0.1299, + "step": 39250 + }, + { + "epoch": 0.34071752849367626, + "grad_norm": 0.291015625, + "learning_rate": 0.001371029426542377, + "loss": 0.1143, + "step": 39251 + }, + { + "epoch": 0.34072620897388045, + "grad_norm": 0.41796875, + "learning_rate": 0.001371001045548303, + "loss": 0.0654, + "step": 39252 + }, + { + "epoch": 0.3407348894540846, + "grad_norm": 0.55859375, + "learning_rate": 0.0013709726642578658, + "loss": 0.1006, + "step": 39253 + }, + { + "epoch": 0.3407435699342888, + "grad_norm": 0.29296875, + "learning_rate": 0.0013709442826710975, + "loss": 0.0996, + "step": 39254 + }, + { + "epoch": 0.3407522504144929, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013709159007880285, + "loss": 0.1055, + "step": 39255 + }, + { + "epoch": 0.3407609308946971, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013708875186086897, + "loss": 0.1196, + "step": 39256 + }, + { + "epoch": 0.34076961137490125, + "grad_norm": 0.1064453125, + "learning_rate": 0.0013708591361331127, + "loss": 0.1084, + "step": 39257 + }, + { + "epoch": 0.34077829185510544, + "grad_norm": 0.85546875, + "learning_rate": 0.001370830753361328, + "loss": 0.1064, + "step": 39258 + }, + { + "epoch": 0.3407869723353096, + "grad_norm": 0.46875, + "learning_rate": 0.001370802370293367, + "loss": 0.0767, + "step": 39259 + }, + { + "epoch": 0.34079565281551377, + "grad_norm": 0.5859375, + "learning_rate": 0.0013707739869292604, + "loss": 0.1328, + "step": 39260 + }, + { + "epoch": 0.3408043332957179, + "grad_norm": 0.18359375, + "learning_rate": 0.0013707456032690398, + "loss": 0.1289, + "step": 39261 + }, + { + "epoch": 0.3408130137759221, + "grad_norm": 0.087890625, + "learning_rate": 0.0013707172193127355, + "loss": 0.0608, + "step": 39262 + }, + { + "epoch": 0.34082169425612624, + "grad_norm": 0.1591796875, + "learning_rate": 0.0013706888350603792, + "loss": 0.123, + "step": 39263 + }, + { + "epoch": 0.34083037473633043, + "grad_norm": 0.78515625, + "learning_rate": 0.0013706604505120015, + "loss": 0.1016, + "step": 39264 + }, + { + "epoch": 0.34083905521653457, + "grad_norm": 0.330078125, + "learning_rate": 0.0013706320656676333, + "loss": 0.1162, + "step": 39265 + }, + { + "epoch": 0.34084773569673876, + "grad_norm": 0.8046875, + "learning_rate": 0.0013706036805273066, + "loss": 0.0938, + "step": 39266 + }, + { + "epoch": 0.3408564161769429, + "grad_norm": 0.21484375, + "learning_rate": 0.0013705752950910516, + "loss": 0.0972, + "step": 39267 + }, + { + "epoch": 0.3408650966571471, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013705469093588992, + "loss": 0.1147, + "step": 39268 + }, + { + "epoch": 0.34087377713735123, + "grad_norm": 0.703125, + "learning_rate": 0.0013705185233308808, + "loss": 0.1543, + "step": 39269 + }, + { + "epoch": 0.3408824576175554, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013704901370070277, + "loss": 0.1245, + "step": 39270 + }, + { + "epoch": 0.34089113809775956, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013704617503873704, + "loss": 0.0898, + "step": 39271 + }, + { + "epoch": 0.34089981857796375, + "grad_norm": 0.84375, + "learning_rate": 0.0013704333634719404, + "loss": 0.1211, + "step": 39272 + }, + { + "epoch": 0.3409084990581679, + "grad_norm": 0.14453125, + "learning_rate": 0.0013704049762607684, + "loss": 0.1211, + "step": 39273 + }, + { + "epoch": 0.3409171795383721, + "grad_norm": 0.71484375, + "learning_rate": 0.0013703765887538855, + "loss": 0.0674, + "step": 39274 + }, + { + "epoch": 0.3409258600185762, + "grad_norm": 0.251953125, + "learning_rate": 0.001370348200951323, + "loss": 0.1064, + "step": 39275 + }, + { + "epoch": 0.3409345404987804, + "grad_norm": 0.3515625, + "learning_rate": 0.0013703198128531117, + "loss": 0.1191, + "step": 39276 + }, + { + "epoch": 0.34094322097898455, + "grad_norm": 0.2314453125, + "learning_rate": 0.0013702914244592829, + "loss": 0.0977, + "step": 39277 + }, + { + "epoch": 0.34095190145918874, + "grad_norm": 0.11767578125, + "learning_rate": 0.0013702630357698671, + "loss": 0.084, + "step": 39278 + }, + { + "epoch": 0.3409605819393929, + "grad_norm": 0.400390625, + "learning_rate": 0.0013702346467848962, + "loss": 0.0889, + "step": 39279 + }, + { + "epoch": 0.3409692624195971, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013702062575044003, + "loss": 0.0674, + "step": 39280 + }, + { + "epoch": 0.3409779428998012, + "grad_norm": 0.53125, + "learning_rate": 0.001370177867928411, + "loss": 0.1221, + "step": 39281 + }, + { + "epoch": 0.3409866233800054, + "grad_norm": 0.44140625, + "learning_rate": 0.001370149478056959, + "loss": 0.1152, + "step": 39282 + }, + { + "epoch": 0.34099530386020954, + "grad_norm": 2.046875, + "learning_rate": 0.001370121087890076, + "loss": 0.1167, + "step": 39283 + }, + { + "epoch": 0.34100398434041373, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013700926974277924, + "loss": 0.1138, + "step": 39284 + }, + { + "epoch": 0.3410126648206179, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013700643066701396, + "loss": 0.1182, + "step": 39285 + }, + { + "epoch": 0.34102134530082207, + "grad_norm": 0.328125, + "learning_rate": 0.0013700359156171486, + "loss": 0.0869, + "step": 39286 + }, + { + "epoch": 0.3410300257810262, + "grad_norm": 0.154296875, + "learning_rate": 0.0013700075242688502, + "loss": 0.084, + "step": 39287 + }, + { + "epoch": 0.3410387062612304, + "grad_norm": 0.81640625, + "learning_rate": 0.001369979132625276, + "loss": 0.123, + "step": 39288 + }, + { + "epoch": 0.34104738674143453, + "grad_norm": 0.310546875, + "learning_rate": 0.0013699507406864564, + "loss": 0.082, + "step": 39289 + }, + { + "epoch": 0.3410560672216387, + "grad_norm": 0.47265625, + "learning_rate": 0.0013699223484524225, + "loss": 0.0664, + "step": 39290 + }, + { + "epoch": 0.34106474770184286, + "grad_norm": 1.390625, + "learning_rate": 0.0013698939559232057, + "loss": 0.1133, + "step": 39291 + }, + { + "epoch": 0.34107342818204706, + "grad_norm": 0.33203125, + "learning_rate": 0.001369865563098837, + "loss": 0.1035, + "step": 39292 + }, + { + "epoch": 0.3410821086622512, + "grad_norm": 0.09765625, + "learning_rate": 0.0013698371699793475, + "loss": 0.0986, + "step": 39293 + }, + { + "epoch": 0.3410907891424554, + "grad_norm": 0.44140625, + "learning_rate": 0.0013698087765647681, + "loss": 0.0742, + "step": 39294 + }, + { + "epoch": 0.3410994696226595, + "grad_norm": 0.0810546875, + "learning_rate": 0.0013697803828551299, + "loss": 0.0747, + "step": 39295 + }, + { + "epoch": 0.3411081501028637, + "grad_norm": 0.6484375, + "learning_rate": 0.0013697519888504637, + "loss": 0.0762, + "step": 39296 + }, + { + "epoch": 0.34111683058306785, + "grad_norm": 0.392578125, + "learning_rate": 0.0013697235945508012, + "loss": 0.0684, + "step": 39297 + }, + { + "epoch": 0.34112551106327205, + "grad_norm": 0.20703125, + "learning_rate": 0.0013696951999561727, + "loss": 0.125, + "step": 39298 + }, + { + "epoch": 0.3411341915434762, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013696668050666094, + "loss": 0.0894, + "step": 39299 + }, + { + "epoch": 0.3411428720236804, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013696384098821426, + "loss": 0.1055, + "step": 39300 + }, + { + "epoch": 0.3411515525038845, + "grad_norm": 0.1787109375, + "learning_rate": 0.0013696100144028037, + "loss": 0.1279, + "step": 39301 + }, + { + "epoch": 0.3411602329840887, + "grad_norm": 0.140625, + "learning_rate": 0.0013695816186286229, + "loss": 0.0957, + "step": 39302 + }, + { + "epoch": 0.34116891346429284, + "grad_norm": 0.3359375, + "learning_rate": 0.001369553222559632, + "loss": 0.0918, + "step": 39303 + }, + { + "epoch": 0.341177593944497, + "grad_norm": 0.75, + "learning_rate": 0.0013695248261958617, + "loss": 0.1426, + "step": 39304 + }, + { + "epoch": 0.3411862744247012, + "grad_norm": 0.12158203125, + "learning_rate": 0.0013694964295373432, + "loss": 0.0889, + "step": 39305 + }, + { + "epoch": 0.3411949549049053, + "grad_norm": 0.6953125, + "learning_rate": 0.001369468032584107, + "loss": 0.1133, + "step": 39306 + }, + { + "epoch": 0.3412036353851095, + "grad_norm": 0.287109375, + "learning_rate": 0.001369439635336185, + "loss": 0.0752, + "step": 39307 + }, + { + "epoch": 0.34121231586531364, + "grad_norm": 0.27734375, + "learning_rate": 0.0013694112377936076, + "loss": 0.0933, + "step": 39308 + }, + { + "epoch": 0.34122099634551784, + "grad_norm": 0.07275390625, + "learning_rate": 0.0013693828399564062, + "loss": 0.084, + "step": 39309 + }, + { + "epoch": 0.341229676825722, + "grad_norm": 0.14453125, + "learning_rate": 0.001369354441824612, + "loss": 0.0874, + "step": 39310 + }, + { + "epoch": 0.34123835730592617, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013693260433982558, + "loss": 0.1318, + "step": 39311 + }, + { + "epoch": 0.3412470377861303, + "grad_norm": 0.2109375, + "learning_rate": 0.0013692976446773684, + "loss": 0.1016, + "step": 39312 + }, + { + "epoch": 0.3412557182663345, + "grad_norm": 0.1748046875, + "learning_rate": 0.0013692692456619812, + "loss": 0.0879, + "step": 39313 + }, + { + "epoch": 0.34126439874653863, + "grad_norm": 0.11962890625, + "learning_rate": 0.0013692408463521254, + "loss": 0.1182, + "step": 39314 + }, + { + "epoch": 0.3412730792267428, + "grad_norm": 0.275390625, + "learning_rate": 0.0013692124467478317, + "loss": 0.1211, + "step": 39315 + }, + { + "epoch": 0.34128175970694696, + "grad_norm": 0.2431640625, + "learning_rate": 0.0013691840468491314, + "loss": 0.1045, + "step": 39316 + }, + { + "epoch": 0.34129044018715116, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013691556466560555, + "loss": 0.123, + "step": 39317 + }, + { + "epoch": 0.3412991206673553, + "grad_norm": 0.171875, + "learning_rate": 0.0013691272461686349, + "loss": 0.1001, + "step": 39318 + }, + { + "epoch": 0.3413078011475595, + "grad_norm": 0.08251953125, + "learning_rate": 0.001369098845386901, + "loss": 0.0713, + "step": 39319 + }, + { + "epoch": 0.3413164816277636, + "grad_norm": 0.341796875, + "learning_rate": 0.0013690704443108843, + "loss": 0.0718, + "step": 39320 + }, + { + "epoch": 0.3413251621079678, + "grad_norm": 0.4375, + "learning_rate": 0.0013690420429406163, + "loss": 0.1108, + "step": 39321 + }, + { + "epoch": 0.34133384258817195, + "grad_norm": 0.10546875, + "learning_rate": 0.001369013641276128, + "loss": 0.1387, + "step": 39322 + }, + { + "epoch": 0.34134252306837615, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013689852393174507, + "loss": 0.1123, + "step": 39323 + }, + { + "epoch": 0.3413512035485803, + "grad_norm": 0.4140625, + "learning_rate": 0.001368956837064615, + "loss": 0.0752, + "step": 39324 + }, + { + "epoch": 0.3413598840287845, + "grad_norm": 0.07275390625, + "learning_rate": 0.0013689284345176518, + "loss": 0.0918, + "step": 39325 + }, + { + "epoch": 0.3413685645089886, + "grad_norm": 0.2177734375, + "learning_rate": 0.001368900031676593, + "loss": 0.1074, + "step": 39326 + }, + { + "epoch": 0.3413772449891928, + "grad_norm": 0.12060546875, + "learning_rate": 0.0013688716285414691, + "loss": 0.1064, + "step": 39327 + }, + { + "epoch": 0.34138592546939694, + "grad_norm": 0.076171875, + "learning_rate": 0.001368843225112311, + "loss": 0.1025, + "step": 39328 + }, + { + "epoch": 0.34139460594960114, + "grad_norm": 0.11083984375, + "learning_rate": 0.0013688148213891502, + "loss": 0.0967, + "step": 39329 + }, + { + "epoch": 0.3414032864298053, + "grad_norm": 0.2275390625, + "learning_rate": 0.0013687864173720172, + "loss": 0.0986, + "step": 39330 + }, + { + "epoch": 0.34141196691000947, + "grad_norm": 0.490234375, + "learning_rate": 0.0013687580130609437, + "loss": 0.0957, + "step": 39331 + }, + { + "epoch": 0.3414206473902136, + "grad_norm": 0.1435546875, + "learning_rate": 0.0013687296084559607, + "loss": 0.1143, + "step": 39332 + }, + { + "epoch": 0.3414293278704178, + "grad_norm": 0.625, + "learning_rate": 0.0013687012035570986, + "loss": 0.1289, + "step": 39333 + }, + { + "epoch": 0.34143800835062194, + "grad_norm": 0.1455078125, + "learning_rate": 0.001368672798364389, + "loss": 0.1006, + "step": 39334 + }, + { + "epoch": 0.34144668883082613, + "grad_norm": 0.333984375, + "learning_rate": 0.0013686443928778629, + "loss": 0.0996, + "step": 39335 + }, + { + "epoch": 0.34145536931103027, + "grad_norm": 0.59765625, + "learning_rate": 0.0013686159870975516, + "loss": 0.1113, + "step": 39336 + }, + { + "epoch": 0.34146404979123446, + "grad_norm": 0.296875, + "learning_rate": 0.0013685875810234857, + "loss": 0.126, + "step": 39337 + }, + { + "epoch": 0.3414727302714386, + "grad_norm": 0.0849609375, + "learning_rate": 0.0013685591746556963, + "loss": 0.1104, + "step": 39338 + }, + { + "epoch": 0.3414814107516428, + "grad_norm": 0.26171875, + "learning_rate": 0.0013685307679942146, + "loss": 0.0806, + "step": 39339 + }, + { + "epoch": 0.3414900912318469, + "grad_norm": 0.291015625, + "learning_rate": 0.0013685023610390719, + "loss": 0.125, + "step": 39340 + }, + { + "epoch": 0.3414987717120511, + "grad_norm": 0.1806640625, + "learning_rate": 0.001368473953790299, + "loss": 0.0815, + "step": 39341 + }, + { + "epoch": 0.34150745219225526, + "grad_norm": 0.255859375, + "learning_rate": 0.0013684455462479271, + "loss": 0.0908, + "step": 39342 + }, + { + "epoch": 0.34151613267245945, + "grad_norm": 0.30859375, + "learning_rate": 0.001368417138411987, + "loss": 0.0874, + "step": 39343 + }, + { + "epoch": 0.3415248131526636, + "grad_norm": 0.427734375, + "learning_rate": 0.0013683887302825101, + "loss": 0.25, + "step": 39344 + }, + { + "epoch": 0.3415334936328678, + "grad_norm": 0.75390625, + "learning_rate": 0.0013683603218595272, + "loss": 0.0835, + "step": 39345 + }, + { + "epoch": 0.3415421741130719, + "grad_norm": 0.294921875, + "learning_rate": 0.0013683319131430696, + "loss": 0.0898, + "step": 39346 + }, + { + "epoch": 0.3415508545932761, + "grad_norm": 0.1181640625, + "learning_rate": 0.0013683035041331681, + "loss": 0.1279, + "step": 39347 + }, + { + "epoch": 0.34155953507348025, + "grad_norm": 0.11083984375, + "learning_rate": 0.001368275094829854, + "loss": 0.1084, + "step": 39348 + }, + { + "epoch": 0.34156821555368444, + "grad_norm": 0.578125, + "learning_rate": 0.001368246685233158, + "loss": 0.1064, + "step": 39349 + }, + { + "epoch": 0.3415768960338886, + "grad_norm": 0.193359375, + "learning_rate": 0.001368218275343112, + "loss": 0.0933, + "step": 39350 + }, + { + "epoch": 0.34158557651409277, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013681898651597463, + "loss": 0.124, + "step": 39351 + }, + { + "epoch": 0.3415942569942969, + "grad_norm": 0.27734375, + "learning_rate": 0.0013681614546830923, + "loss": 0.1113, + "step": 39352 + }, + { + "epoch": 0.3416029374745011, + "grad_norm": 0.451171875, + "learning_rate": 0.0013681330439131807, + "loss": 0.1016, + "step": 39353 + }, + { + "epoch": 0.34161161795470524, + "grad_norm": 0.1689453125, + "learning_rate": 0.001368104632850043, + "loss": 0.0947, + "step": 39354 + }, + { + "epoch": 0.34162029843490943, + "grad_norm": 0.3359375, + "learning_rate": 0.0013680762214937098, + "loss": 0.1113, + "step": 39355 + }, + { + "epoch": 0.34162897891511357, + "grad_norm": 0.9296875, + "learning_rate": 0.0013680478098442128, + "loss": 0.1006, + "step": 39356 + }, + { + "epoch": 0.34163765939531776, + "grad_norm": 0.16796875, + "learning_rate": 0.0013680193979015828, + "loss": 0.0791, + "step": 39357 + }, + { + "epoch": 0.3416463398755219, + "grad_norm": 0.0810546875, + "learning_rate": 0.0013679909856658505, + "loss": 0.0864, + "step": 39358 + }, + { + "epoch": 0.3416550203557261, + "grad_norm": 0.216796875, + "learning_rate": 0.0013679625731370475, + "loss": 0.1328, + "step": 39359 + }, + { + "epoch": 0.34166370083593023, + "grad_norm": 0.298828125, + "learning_rate": 0.0013679341603152046, + "loss": 0.1279, + "step": 39360 + }, + { + "epoch": 0.3416723813161344, + "grad_norm": 0.474609375, + "learning_rate": 0.001367905747200353, + "loss": 0.1006, + "step": 39361 + }, + { + "epoch": 0.34168106179633856, + "grad_norm": 0.279296875, + "learning_rate": 0.0013678773337925236, + "loss": 0.1064, + "step": 39362 + }, + { + "epoch": 0.34168974227654275, + "grad_norm": 0.298828125, + "learning_rate": 0.0013678489200917477, + "loss": 0.0933, + "step": 39363 + }, + { + "epoch": 0.3416984227567469, + "grad_norm": 0.515625, + "learning_rate": 0.001367820506098056, + "loss": 0.1138, + "step": 39364 + }, + { + "epoch": 0.3417071032369511, + "grad_norm": 0.2265625, + "learning_rate": 0.0013677920918114801, + "loss": 0.0928, + "step": 39365 + }, + { + "epoch": 0.3417157837171552, + "grad_norm": 0.435546875, + "learning_rate": 0.0013677636772320507, + "loss": 0.1045, + "step": 39366 + }, + { + "epoch": 0.3417244641973594, + "grad_norm": 0.3203125, + "learning_rate": 0.0013677352623597992, + "loss": 0.1309, + "step": 39367 + }, + { + "epoch": 0.34173314467756355, + "grad_norm": 0.318359375, + "learning_rate": 0.001367706847194756, + "loss": 0.1001, + "step": 39368 + }, + { + "epoch": 0.34174182515776774, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013676784317369528, + "loss": 0.1162, + "step": 39369 + }, + { + "epoch": 0.3417505056379719, + "grad_norm": 0.60546875, + "learning_rate": 0.0013676500159864205, + "loss": 0.106, + "step": 39370 + }, + { + "epoch": 0.3417591861181761, + "grad_norm": 0.2021484375, + "learning_rate": 0.0013676215999431903, + "loss": 0.0918, + "step": 39371 + }, + { + "epoch": 0.3417678665983802, + "grad_norm": 0.56640625, + "learning_rate": 0.001367593183607293, + "loss": 0.1934, + "step": 39372 + }, + { + "epoch": 0.3417765470785844, + "grad_norm": 0.197265625, + "learning_rate": 0.0013675647669787597, + "loss": 0.0791, + "step": 39373 + }, + { + "epoch": 0.34178522755878854, + "grad_norm": 2.5625, + "learning_rate": 0.001367536350057622, + "loss": 0.1348, + "step": 39374 + }, + { + "epoch": 0.34179390803899273, + "grad_norm": 0.1025390625, + "learning_rate": 0.00136750793284391, + "loss": 0.0767, + "step": 39375 + }, + { + "epoch": 0.34180258851919687, + "grad_norm": 0.22265625, + "learning_rate": 0.0013674795153376556, + "loss": 0.1006, + "step": 39376 + }, + { + "epoch": 0.34181126899940106, + "grad_norm": 0.349609375, + "learning_rate": 0.0013674510975388897, + "loss": 0.1387, + "step": 39377 + }, + { + "epoch": 0.3418199494796052, + "grad_norm": 0.1796875, + "learning_rate": 0.0013674226794476432, + "loss": 0.1064, + "step": 39378 + }, + { + "epoch": 0.3418286299598094, + "grad_norm": 0.404296875, + "learning_rate": 0.0013673942610639472, + "loss": 0.0996, + "step": 39379 + }, + { + "epoch": 0.34183731044001353, + "grad_norm": 0.216796875, + "learning_rate": 0.001367365842387833, + "loss": 0.1211, + "step": 39380 + }, + { + "epoch": 0.3418459909202177, + "grad_norm": 0.177734375, + "learning_rate": 0.0013673374234193314, + "loss": 0.1084, + "step": 39381 + }, + { + "epoch": 0.34185467140042186, + "grad_norm": 0.74609375, + "learning_rate": 0.0013673090041584736, + "loss": 0.0801, + "step": 39382 + }, + { + "epoch": 0.34186335188062605, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013672805846052908, + "loss": 0.1113, + "step": 39383 + }, + { + "epoch": 0.3418720323608302, + "grad_norm": 0.1630859375, + "learning_rate": 0.001367252164759814, + "loss": 0.0996, + "step": 39384 + }, + { + "epoch": 0.3418807128410344, + "grad_norm": 0.263671875, + "learning_rate": 0.0013672237446220746, + "loss": 0.0835, + "step": 39385 + }, + { + "epoch": 0.3418893933212385, + "grad_norm": 0.138671875, + "learning_rate": 0.0013671953241921027, + "loss": 0.0938, + "step": 39386 + }, + { + "epoch": 0.3418980738014427, + "grad_norm": 0.6015625, + "learning_rate": 0.0013671669034699301, + "loss": 0.1396, + "step": 39387 + }, + { + "epoch": 0.34190675428164685, + "grad_norm": 0.1591796875, + "learning_rate": 0.001367138482455588, + "loss": 0.1104, + "step": 39388 + }, + { + "epoch": 0.34191543476185104, + "grad_norm": 0.1796875, + "learning_rate": 0.0013671100611491072, + "loss": 0.1016, + "step": 39389 + }, + { + "epoch": 0.3419241152420552, + "grad_norm": 0.4296875, + "learning_rate": 0.0013670816395505187, + "loss": 0.106, + "step": 39390 + }, + { + "epoch": 0.3419327957222594, + "grad_norm": 0.8515625, + "learning_rate": 0.001367053217659854, + "loss": 0.1201, + "step": 39391 + }, + { + "epoch": 0.3419414762024635, + "grad_norm": 0.23046875, + "learning_rate": 0.001367024795477144, + "loss": 0.0928, + "step": 39392 + }, + { + "epoch": 0.3419501566826677, + "grad_norm": 0.298828125, + "learning_rate": 0.0013669963730024196, + "loss": 0.1035, + "step": 39393 + }, + { + "epoch": 0.34195883716287184, + "grad_norm": 0.16015625, + "learning_rate": 0.0013669679502357117, + "loss": 0.0703, + "step": 39394 + }, + { + "epoch": 0.34196751764307604, + "grad_norm": 0.3203125, + "learning_rate": 0.0013669395271770518, + "loss": 0.1094, + "step": 39395 + }, + { + "epoch": 0.3419761981232802, + "grad_norm": 0.5625, + "learning_rate": 0.0013669111038264708, + "loss": 0.1162, + "step": 39396 + }, + { + "epoch": 0.34198487860348437, + "grad_norm": 0.1767578125, + "learning_rate": 0.0013668826801839999, + "loss": 0.1152, + "step": 39397 + }, + { + "epoch": 0.3419935590836885, + "grad_norm": 0.578125, + "learning_rate": 0.0013668542562496704, + "loss": 0.0962, + "step": 39398 + }, + { + "epoch": 0.3420022395638927, + "grad_norm": 0.8359375, + "learning_rate": 0.0013668258320235125, + "loss": 0.1279, + "step": 39399 + }, + { + "epoch": 0.34201092004409683, + "grad_norm": 0.408203125, + "learning_rate": 0.0013667974075055583, + "loss": 0.0928, + "step": 39400 + }, + { + "epoch": 0.342019600524301, + "grad_norm": 0.12255859375, + "learning_rate": 0.0013667689826958383, + "loss": 0.0854, + "step": 39401 + }, + { + "epoch": 0.34202828100450516, + "grad_norm": 0.07763671875, + "learning_rate": 0.0013667405575943838, + "loss": 0.0747, + "step": 39402 + }, + { + "epoch": 0.34203696148470936, + "grad_norm": 0.298828125, + "learning_rate": 0.0013667121322012258, + "loss": 0.1289, + "step": 39403 + }, + { + "epoch": 0.3420456419649135, + "grad_norm": 0.2001953125, + "learning_rate": 0.0013666837065163956, + "loss": 0.0674, + "step": 39404 + }, + { + "epoch": 0.3420543224451177, + "grad_norm": 0.087890625, + "learning_rate": 0.0013666552805399238, + "loss": 0.0928, + "step": 39405 + }, + { + "epoch": 0.3420630029253218, + "grad_norm": 0.60546875, + "learning_rate": 0.001366626854271842, + "loss": 0.105, + "step": 39406 + }, + { + "epoch": 0.342071683405526, + "grad_norm": 0.3984375, + "learning_rate": 0.001366598427712181, + "loss": 0.0972, + "step": 39407 + }, + { + "epoch": 0.34208036388573015, + "grad_norm": 0.443359375, + "learning_rate": 0.0013665700008609718, + "loss": 0.1025, + "step": 39408 + }, + { + "epoch": 0.34208904436593435, + "grad_norm": 0.515625, + "learning_rate": 0.001366541573718246, + "loss": 0.2051, + "step": 39409 + }, + { + "epoch": 0.3420977248461385, + "grad_norm": 0.625, + "learning_rate": 0.0013665131462840343, + "loss": 0.1152, + "step": 39410 + }, + { + "epoch": 0.3421064053263427, + "grad_norm": 0.90625, + "learning_rate": 0.0013664847185583678, + "loss": 0.1289, + "step": 39411 + }, + { + "epoch": 0.3421150858065468, + "grad_norm": 0.302734375, + "learning_rate": 0.0013664562905412775, + "loss": 0.0981, + "step": 39412 + }, + { + "epoch": 0.342123766286751, + "grad_norm": 0.322265625, + "learning_rate": 0.0013664278622327946, + "loss": 0.123, + "step": 39413 + }, + { + "epoch": 0.34213244676695515, + "grad_norm": 0.34765625, + "learning_rate": 0.0013663994336329502, + "loss": 0.125, + "step": 39414 + }, + { + "epoch": 0.34214112724715934, + "grad_norm": 0.291015625, + "learning_rate": 0.0013663710047417755, + "loss": 0.0947, + "step": 39415 + }, + { + "epoch": 0.3421498077273635, + "grad_norm": 0.1572265625, + "learning_rate": 0.0013663425755593012, + "loss": 0.0869, + "step": 39416 + }, + { + "epoch": 0.34215848820756767, + "grad_norm": 0.16015625, + "learning_rate": 0.0013663141460855588, + "loss": 0.1133, + "step": 39417 + }, + { + "epoch": 0.3421671686877718, + "grad_norm": 0.50390625, + "learning_rate": 0.0013662857163205793, + "loss": 0.0889, + "step": 39418 + }, + { + "epoch": 0.342175849167976, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013662572862643938, + "loss": 0.1055, + "step": 39419 + }, + { + "epoch": 0.34218452964818014, + "grad_norm": 0.251953125, + "learning_rate": 0.0013662288559170334, + "loss": 0.0757, + "step": 39420 + }, + { + "epoch": 0.34219321012838433, + "grad_norm": 0.08544921875, + "learning_rate": 0.001366200425278529, + "loss": 0.082, + "step": 39421 + }, + { + "epoch": 0.34220189060858847, + "grad_norm": 0.15625, + "learning_rate": 0.0013661719943489118, + "loss": 0.1562, + "step": 39422 + }, + { + "epoch": 0.34221057108879266, + "grad_norm": 0.625, + "learning_rate": 0.001366143563128213, + "loss": 0.1123, + "step": 39423 + }, + { + "epoch": 0.3422192515689968, + "grad_norm": 0.279296875, + "learning_rate": 0.0013661151316164634, + "loss": 0.0791, + "step": 39424 + }, + { + "epoch": 0.342227932049201, + "grad_norm": 0.330078125, + "learning_rate": 0.0013660866998136943, + "loss": 0.1064, + "step": 39425 + }, + { + "epoch": 0.3422366125294051, + "grad_norm": 0.2041015625, + "learning_rate": 0.001366058267719937, + "loss": 0.1006, + "step": 39426 + }, + { + "epoch": 0.3422452930096093, + "grad_norm": 0.19921875, + "learning_rate": 0.0013660298353352222, + "loss": 0.0703, + "step": 39427 + }, + { + "epoch": 0.34225397348981346, + "grad_norm": 0.0869140625, + "learning_rate": 0.0013660014026595812, + "loss": 0.083, + "step": 39428 + }, + { + "epoch": 0.3422626539700176, + "grad_norm": 0.0859375, + "learning_rate": 0.001365972969693045, + "loss": 0.0742, + "step": 39429 + }, + { + "epoch": 0.3422713344502218, + "grad_norm": 0.09814453125, + "learning_rate": 0.0013659445364356448, + "loss": 0.0986, + "step": 39430 + }, + { + "epoch": 0.3422800149304259, + "grad_norm": 0.263671875, + "learning_rate": 0.001365916102887412, + "loss": 0.1455, + "step": 39431 + }, + { + "epoch": 0.3422886954106301, + "grad_norm": 0.73046875, + "learning_rate": 0.0013658876690483771, + "loss": 0.1318, + "step": 39432 + }, + { + "epoch": 0.34229737589083425, + "grad_norm": 0.2353515625, + "learning_rate": 0.001365859234918571, + "loss": 0.1045, + "step": 39433 + }, + { + "epoch": 0.34230605637103845, + "grad_norm": 0.12158203125, + "learning_rate": 0.0013658308004980257, + "loss": 0.0874, + "step": 39434 + }, + { + "epoch": 0.3423147368512426, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013658023657867715, + "loss": 0.0732, + "step": 39435 + }, + { + "epoch": 0.3423234173314468, + "grad_norm": 0.9296875, + "learning_rate": 0.0013657739307848403, + "loss": 0.1602, + "step": 39436 + }, + { + "epoch": 0.3423320978116509, + "grad_norm": 0.08935546875, + "learning_rate": 0.0013657454954922625, + "loss": 0.1055, + "step": 39437 + }, + { + "epoch": 0.3423407782918551, + "grad_norm": 0.19921875, + "learning_rate": 0.0013657170599090695, + "loss": 0.0713, + "step": 39438 + }, + { + "epoch": 0.34234945877205925, + "grad_norm": 0.12109375, + "learning_rate": 0.001365688624035292, + "loss": 0.0859, + "step": 39439 + }, + { + "epoch": 0.34235813925226344, + "grad_norm": 0.275390625, + "learning_rate": 0.0013656601878709617, + "loss": 0.1533, + "step": 39440 + }, + { + "epoch": 0.3423668197324676, + "grad_norm": 0.33203125, + "learning_rate": 0.0013656317514161096, + "loss": 0.0879, + "step": 39441 + }, + { + "epoch": 0.34237550021267177, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013656033146707662, + "loss": 0.0835, + "step": 39442 + }, + { + "epoch": 0.3423841806928759, + "grad_norm": 0.236328125, + "learning_rate": 0.0013655748776349632, + "loss": 0.0957, + "step": 39443 + }, + { + "epoch": 0.3423928611730801, + "grad_norm": 0.166015625, + "learning_rate": 0.0013655464403087313, + "loss": 0.1094, + "step": 39444 + }, + { + "epoch": 0.34240154165328424, + "grad_norm": 0.1044921875, + "learning_rate": 0.0013655180026921022, + "loss": 0.127, + "step": 39445 + }, + { + "epoch": 0.34241022213348843, + "grad_norm": 0.1123046875, + "learning_rate": 0.0013654895647851062, + "loss": 0.0708, + "step": 39446 + }, + { + "epoch": 0.34241890261369257, + "grad_norm": 0.44921875, + "learning_rate": 0.001365461126587775, + "loss": 0.0825, + "step": 39447 + }, + { + "epoch": 0.34242758309389676, + "grad_norm": 0.12255859375, + "learning_rate": 0.0013654326881001396, + "loss": 0.0972, + "step": 39448 + }, + { + "epoch": 0.3424362635741009, + "grad_norm": 0.703125, + "learning_rate": 0.001365404249322231, + "loss": 0.0898, + "step": 39449 + }, + { + "epoch": 0.3424449440543051, + "grad_norm": 0.173828125, + "learning_rate": 0.0013653758102540798, + "loss": 0.0781, + "step": 39450 + }, + { + "epoch": 0.3424536245345092, + "grad_norm": 0.23046875, + "learning_rate": 0.0013653473708957183, + "loss": 0.1118, + "step": 39451 + }, + { + "epoch": 0.3424623050147134, + "grad_norm": 0.267578125, + "learning_rate": 0.0013653189312471765, + "loss": 0.0918, + "step": 39452 + }, + { + "epoch": 0.34247098549491756, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013652904913084858, + "loss": 0.0889, + "step": 39453 + }, + { + "epoch": 0.34247966597512175, + "grad_norm": 0.94140625, + "learning_rate": 0.0013652620510796775, + "loss": 0.124, + "step": 39454 + }, + { + "epoch": 0.3424883464553259, + "grad_norm": 1.0703125, + "learning_rate": 0.001365233610560783, + "loss": 0.127, + "step": 39455 + }, + { + "epoch": 0.3424970269355301, + "grad_norm": 0.251953125, + "learning_rate": 0.0013652051697518326, + "loss": 0.1074, + "step": 39456 + }, + { + "epoch": 0.3425057074157342, + "grad_norm": 0.478515625, + "learning_rate": 0.0013651767286528577, + "loss": 0.1074, + "step": 39457 + }, + { + "epoch": 0.3425143878959384, + "grad_norm": 0.203125, + "learning_rate": 0.00136514828726389, + "loss": 0.0991, + "step": 39458 + }, + { + "epoch": 0.34252306837614255, + "grad_norm": 0.09423828125, + "learning_rate": 0.0013651198455849598, + "loss": 0.0732, + "step": 39459 + }, + { + "epoch": 0.34253174885634674, + "grad_norm": 0.66015625, + "learning_rate": 0.0013650914036160986, + "loss": 0.127, + "step": 39460 + }, + { + "epoch": 0.3425404293365509, + "grad_norm": 0.2265625, + "learning_rate": 0.0013650629613573373, + "loss": 0.1426, + "step": 39461 + }, + { + "epoch": 0.34254910981675507, + "grad_norm": 0.2001953125, + "learning_rate": 0.0013650345188087074, + "loss": 0.1084, + "step": 39462 + }, + { + "epoch": 0.3425577902969592, + "grad_norm": 0.3046875, + "learning_rate": 0.0013650060759702394, + "loss": 0.0693, + "step": 39463 + }, + { + "epoch": 0.3425664707771634, + "grad_norm": 0.0966796875, + "learning_rate": 0.0013649776328419649, + "loss": 0.0835, + "step": 39464 + }, + { + "epoch": 0.34257515125736754, + "grad_norm": 0.259765625, + "learning_rate": 0.001364949189423915, + "loss": 0.0811, + "step": 39465 + }, + { + "epoch": 0.34258383173757173, + "grad_norm": 0.408203125, + "learning_rate": 0.0013649207457161203, + "loss": 0.085, + "step": 39466 + }, + { + "epoch": 0.34259251221777587, + "grad_norm": 0.1474609375, + "learning_rate": 0.0013648923017186125, + "loss": 0.1094, + "step": 39467 + }, + { + "epoch": 0.34260119269798006, + "grad_norm": 0.09033203125, + "learning_rate": 0.0013648638574314225, + "loss": 0.1104, + "step": 39468 + }, + { + "epoch": 0.3426098731781842, + "grad_norm": 0.1904296875, + "learning_rate": 0.001364835412854581, + "loss": 0.1367, + "step": 39469 + }, + { + "epoch": 0.3426185536583884, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013648069679881199, + "loss": 0.1279, + "step": 39470 + }, + { + "epoch": 0.34262723413859253, + "grad_norm": 0.173828125, + "learning_rate": 0.00136477852283207, + "loss": 0.0947, + "step": 39471 + }, + { + "epoch": 0.3426359146187967, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013647500773864617, + "loss": 0.0898, + "step": 39472 + }, + { + "epoch": 0.34264459509900086, + "grad_norm": 0.2890625, + "learning_rate": 0.001364721631651327, + "loss": 0.0776, + "step": 39473 + }, + { + "epoch": 0.34265327557920505, + "grad_norm": 0.40625, + "learning_rate": 0.0013646931856266967, + "loss": 0.1436, + "step": 39474 + }, + { + "epoch": 0.3426619560594092, + "grad_norm": 0.1748046875, + "learning_rate": 0.0013646647393126016, + "loss": 0.1094, + "step": 39475 + }, + { + "epoch": 0.3426706365396134, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013646362927090734, + "loss": 0.1221, + "step": 39476 + }, + { + "epoch": 0.3426793170198175, + "grad_norm": 0.0966796875, + "learning_rate": 0.0013646078458161434, + "loss": 0.0903, + "step": 39477 + }, + { + "epoch": 0.3426879975000217, + "grad_norm": 0.6796875, + "learning_rate": 0.0013645793986338415, + "loss": 0.1211, + "step": 39478 + }, + { + "epoch": 0.34269667798022585, + "grad_norm": 0.294921875, + "learning_rate": 0.0013645509511621997, + "loss": 0.1123, + "step": 39479 + }, + { + "epoch": 0.34270535846043004, + "grad_norm": 0.11376953125, + "learning_rate": 0.0013645225034012495, + "loss": 0.0869, + "step": 39480 + }, + { + "epoch": 0.3427140389406342, + "grad_norm": 0.3203125, + "learning_rate": 0.0013644940553510208, + "loss": 0.0962, + "step": 39481 + }, + { + "epoch": 0.3427227194208384, + "grad_norm": 0.40625, + "learning_rate": 0.0013644656070115454, + "loss": 0.123, + "step": 39482 + }, + { + "epoch": 0.3427313999010425, + "grad_norm": 0.58203125, + "learning_rate": 0.0013644371583828546, + "loss": 0.0894, + "step": 39483 + }, + { + "epoch": 0.3427400803812467, + "grad_norm": 0.166015625, + "learning_rate": 0.0013644087094649795, + "loss": 0.1123, + "step": 39484 + }, + { + "epoch": 0.34274876086145084, + "grad_norm": 0.263671875, + "learning_rate": 0.0013643802602579507, + "loss": 0.0649, + "step": 39485 + }, + { + "epoch": 0.34275744134165503, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013643518107618, + "loss": 0.1035, + "step": 39486 + }, + { + "epoch": 0.34276612182185917, + "grad_norm": 0.23046875, + "learning_rate": 0.0013643233609765576, + "loss": 0.1094, + "step": 39487 + }, + { + "epoch": 0.34277480230206336, + "grad_norm": 0.265625, + "learning_rate": 0.0013642949109022556, + "loss": 0.1426, + "step": 39488 + }, + { + "epoch": 0.3427834827822675, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013642664605389243, + "loss": 0.127, + "step": 39489 + }, + { + "epoch": 0.3427921632624717, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013642380098865953, + "loss": 0.0732, + "step": 39490 + }, + { + "epoch": 0.34280084374267583, + "grad_norm": 0.72265625, + "learning_rate": 0.0013642095589452997, + "loss": 0.1035, + "step": 39491 + }, + { + "epoch": 0.34280952422288, + "grad_norm": 0.0908203125, + "learning_rate": 0.001364181107715068, + "loss": 0.1543, + "step": 39492 + }, + { + "epoch": 0.34281820470308416, + "grad_norm": 0.111328125, + "learning_rate": 0.0013641526561959326, + "loss": 0.1533, + "step": 39493 + }, + { + "epoch": 0.34282688518328835, + "grad_norm": 0.373046875, + "learning_rate": 0.0013641242043879234, + "loss": 0.0957, + "step": 39494 + }, + { + "epoch": 0.3428355656634925, + "grad_norm": 0.2177734375, + "learning_rate": 0.0013640957522910717, + "loss": 0.0796, + "step": 39495 + }, + { + "epoch": 0.3428442461436967, + "grad_norm": 0.09326171875, + "learning_rate": 0.0013640672999054096, + "loss": 0.1211, + "step": 39496 + }, + { + "epoch": 0.3428529266239008, + "grad_norm": 0.5390625, + "learning_rate": 0.001364038847230967, + "loss": 0.0869, + "step": 39497 + }, + { + "epoch": 0.342861607104105, + "grad_norm": 0.26953125, + "learning_rate": 0.0013640103942677752, + "loss": 0.127, + "step": 39498 + }, + { + "epoch": 0.34287028758430915, + "grad_norm": 0.322265625, + "learning_rate": 0.001363981941015866, + "loss": 0.0903, + "step": 39499 + }, + { + "epoch": 0.34287896806451335, + "grad_norm": 0.205078125, + "learning_rate": 0.00136395348747527, + "loss": 0.0879, + "step": 39500 + }, + { + "epoch": 0.3428876485447175, + "grad_norm": 0.11865234375, + "learning_rate": 0.0013639250336460185, + "loss": 0.0859, + "step": 39501 + }, + { + "epoch": 0.3428963290249217, + "grad_norm": 0.15234375, + "learning_rate": 0.0013638965795281423, + "loss": 0.1123, + "step": 39502 + }, + { + "epoch": 0.3429050095051258, + "grad_norm": 0.1591796875, + "learning_rate": 0.0013638681251216729, + "loss": 0.1147, + "step": 39503 + }, + { + "epoch": 0.34291368998533, + "grad_norm": 0.08203125, + "learning_rate": 0.0013638396704266414, + "loss": 0.085, + "step": 39504 + }, + { + "epoch": 0.34292237046553414, + "grad_norm": 0.5625, + "learning_rate": 0.0013638112154430787, + "loss": 0.1172, + "step": 39505 + }, + { + "epoch": 0.34293105094573834, + "grad_norm": 0.142578125, + "learning_rate": 0.001363782760171016, + "loss": 0.1162, + "step": 39506 + }, + { + "epoch": 0.3429397314259425, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013637543046104846, + "loss": 0.0825, + "step": 39507 + }, + { + "epoch": 0.34294841190614667, + "grad_norm": 0.1767578125, + "learning_rate": 0.0013637258487615154, + "loss": 0.0869, + "step": 39508 + }, + { + "epoch": 0.3429570923863508, + "grad_norm": 0.498046875, + "learning_rate": 0.0013636973926241397, + "loss": 0.1108, + "step": 39509 + }, + { + "epoch": 0.342965772866555, + "grad_norm": 0.10791015625, + "learning_rate": 0.001363668936198388, + "loss": 0.0884, + "step": 39510 + }, + { + "epoch": 0.34297445334675913, + "grad_norm": 0.341796875, + "learning_rate": 0.0013636404794842923, + "loss": 0.1123, + "step": 39511 + }, + { + "epoch": 0.3429831338269633, + "grad_norm": 0.609375, + "learning_rate": 0.001363612022481883, + "loss": 0.0859, + "step": 39512 + }, + { + "epoch": 0.34299181430716746, + "grad_norm": 0.3828125, + "learning_rate": 0.001363583565191192, + "loss": 0.123, + "step": 39513 + }, + { + "epoch": 0.34300049478737166, + "grad_norm": 0.279296875, + "learning_rate": 0.0013635551076122496, + "loss": 0.083, + "step": 39514 + }, + { + "epoch": 0.3430091752675758, + "grad_norm": 0.1142578125, + "learning_rate": 0.0013635266497450875, + "loss": 0.0879, + "step": 39515 + }, + { + "epoch": 0.34301785574778, + "grad_norm": 0.234375, + "learning_rate": 0.0013634981915897366, + "loss": 0.083, + "step": 39516 + }, + { + "epoch": 0.3430265362279841, + "grad_norm": 0.1416015625, + "learning_rate": 0.0013634697331462282, + "loss": 0.1602, + "step": 39517 + }, + { + "epoch": 0.3430352167081883, + "grad_norm": 0.28515625, + "learning_rate": 0.0013634412744145932, + "loss": 0.1021, + "step": 39518 + }, + { + "epoch": 0.34304389718839245, + "grad_norm": 0.59375, + "learning_rate": 0.0013634128153948626, + "loss": 0.0986, + "step": 39519 + }, + { + "epoch": 0.34305257766859665, + "grad_norm": 1.1953125, + "learning_rate": 0.0013633843560870676, + "loss": 0.1123, + "step": 39520 + }, + { + "epoch": 0.3430612581488008, + "grad_norm": 0.177734375, + "learning_rate": 0.0013633558964912397, + "loss": 0.1357, + "step": 39521 + }, + { + "epoch": 0.343069938629005, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013633274366074096, + "loss": 0.1631, + "step": 39522 + }, + { + "epoch": 0.3430786191092091, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013632989764356088, + "loss": 0.1084, + "step": 39523 + }, + { + "epoch": 0.3430872995894133, + "grad_norm": 0.29296875, + "learning_rate": 0.001363270515975868, + "loss": 0.1133, + "step": 39524 + }, + { + "epoch": 0.34309598006961745, + "grad_norm": 0.56640625, + "learning_rate": 0.0013632420552282182, + "loss": 0.104, + "step": 39525 + }, + { + "epoch": 0.34310466054982164, + "grad_norm": 0.212890625, + "learning_rate": 0.0013632135941926914, + "loss": 0.1113, + "step": 39526 + }, + { + "epoch": 0.3431133410300258, + "grad_norm": 0.330078125, + "learning_rate": 0.0013631851328693178, + "loss": 0.0918, + "step": 39527 + }, + { + "epoch": 0.34312202151022997, + "grad_norm": 0.138671875, + "learning_rate": 0.0013631566712581289, + "loss": 0.106, + "step": 39528 + }, + { + "epoch": 0.3431307019904341, + "grad_norm": 0.76953125, + "learning_rate": 0.001363128209359156, + "loss": 0.1436, + "step": 39529 + }, + { + "epoch": 0.3431393824706383, + "grad_norm": 0.259765625, + "learning_rate": 0.00136309974717243, + "loss": 0.1191, + "step": 39530 + }, + { + "epoch": 0.34314806295084244, + "grad_norm": 0.1259765625, + "learning_rate": 0.001363071284697982, + "loss": 0.1299, + "step": 39531 + }, + { + "epoch": 0.34315674343104663, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013630428219358435, + "loss": 0.1221, + "step": 39532 + }, + { + "epoch": 0.34316542391125077, + "grad_norm": 0.11328125, + "learning_rate": 0.0013630143588860449, + "loss": 0.0869, + "step": 39533 + }, + { + "epoch": 0.34317410439145496, + "grad_norm": 0.09326171875, + "learning_rate": 0.0013629858955486178, + "loss": 0.0762, + "step": 39534 + }, + { + "epoch": 0.3431827848716591, + "grad_norm": 0.1484375, + "learning_rate": 0.0013629574319235936, + "loss": 0.1504, + "step": 39535 + }, + { + "epoch": 0.3431914653518633, + "grad_norm": 0.41015625, + "learning_rate": 0.0013629289680110027, + "loss": 0.1133, + "step": 39536 + }, + { + "epoch": 0.3432001458320674, + "grad_norm": 0.2080078125, + "learning_rate": 0.001362900503810877, + "loss": 0.084, + "step": 39537 + }, + { + "epoch": 0.3432088263122716, + "grad_norm": 0.58984375, + "learning_rate": 0.0013628720393232471, + "loss": 0.0903, + "step": 39538 + }, + { + "epoch": 0.34321750679247576, + "grad_norm": 0.240234375, + "learning_rate": 0.0013628435745481444, + "loss": 0.1216, + "step": 39539 + }, + { + "epoch": 0.34322618727267995, + "grad_norm": 0.28125, + "learning_rate": 0.0013628151094856, + "loss": 0.0947, + "step": 39540 + }, + { + "epoch": 0.3432348677528841, + "grad_norm": 0.185546875, + "learning_rate": 0.0013627866441356443, + "loss": 0.0767, + "step": 39541 + }, + { + "epoch": 0.3432435482330883, + "grad_norm": 0.4609375, + "learning_rate": 0.0013627581784983094, + "loss": 0.1016, + "step": 39542 + }, + { + "epoch": 0.3432522287132924, + "grad_norm": 0.359375, + "learning_rate": 0.0013627297125736264, + "loss": 0.084, + "step": 39543 + }, + { + "epoch": 0.3432609091934966, + "grad_norm": 0.162109375, + "learning_rate": 0.0013627012463616262, + "loss": 0.1094, + "step": 39544 + }, + { + "epoch": 0.34326958967370075, + "grad_norm": 0.341796875, + "learning_rate": 0.0013626727798623395, + "loss": 0.1387, + "step": 39545 + }, + { + "epoch": 0.34327827015390494, + "grad_norm": 0.298828125, + "learning_rate": 0.0013626443130757978, + "loss": 0.1406, + "step": 39546 + }, + { + "epoch": 0.3432869506341091, + "grad_norm": 0.51953125, + "learning_rate": 0.0013626158460020323, + "loss": 0.1406, + "step": 39547 + }, + { + "epoch": 0.34329563111431327, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013625873786410744, + "loss": 0.1064, + "step": 39548 + }, + { + "epoch": 0.3433043115945174, + "grad_norm": 0.08837890625, + "learning_rate": 0.0013625589109929543, + "loss": 0.1035, + "step": 39549 + }, + { + "epoch": 0.3433129920747216, + "grad_norm": 0.26953125, + "learning_rate": 0.0013625304430577038, + "loss": 0.1406, + "step": 39550 + }, + { + "epoch": 0.34332167255492574, + "grad_norm": 0.412109375, + "learning_rate": 0.0013625019748353545, + "loss": 0.165, + "step": 39551 + }, + { + "epoch": 0.3433303530351299, + "grad_norm": 0.73046875, + "learning_rate": 0.0013624735063259366, + "loss": 0.103, + "step": 39552 + }, + { + "epoch": 0.34333903351533407, + "grad_norm": 0.353515625, + "learning_rate": 0.0013624450375294816, + "loss": 0.0986, + "step": 39553 + }, + { + "epoch": 0.3433477139955382, + "grad_norm": 0.84375, + "learning_rate": 0.0013624165684460207, + "loss": 0.0786, + "step": 39554 + }, + { + "epoch": 0.3433563944757424, + "grad_norm": 0.2021484375, + "learning_rate": 0.001362388099075585, + "loss": 0.0845, + "step": 39555 + }, + { + "epoch": 0.34336507495594654, + "grad_norm": 0.326171875, + "learning_rate": 0.0013623596294182058, + "loss": 0.1211, + "step": 39556 + }, + { + "epoch": 0.34337375543615073, + "grad_norm": 0.15234375, + "learning_rate": 0.001362331159473914, + "loss": 0.1689, + "step": 39557 + }, + { + "epoch": 0.34338243591635487, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013623026892427403, + "loss": 0.0986, + "step": 39558 + }, + { + "epoch": 0.34339111639655906, + "grad_norm": 0.337890625, + "learning_rate": 0.0013622742187247168, + "loss": 0.0923, + "step": 39559 + }, + { + "epoch": 0.3433997968767632, + "grad_norm": 0.90625, + "learning_rate": 0.0013622457479198738, + "loss": 0.1123, + "step": 39560 + }, + { + "epoch": 0.3434084773569674, + "grad_norm": 0.4296875, + "learning_rate": 0.0013622172768282433, + "loss": 0.0972, + "step": 39561 + }, + { + "epoch": 0.3434171578371715, + "grad_norm": 0.201171875, + "learning_rate": 0.0013621888054498555, + "loss": 0.1465, + "step": 39562 + }, + { + "epoch": 0.3434258383173757, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013621603337847419, + "loss": 0.1143, + "step": 39563 + }, + { + "epoch": 0.34343451879757986, + "grad_norm": 0.1513671875, + "learning_rate": 0.0013621318618329342, + "loss": 0.0796, + "step": 39564 + }, + { + "epoch": 0.34344319927778405, + "grad_norm": 0.515625, + "learning_rate": 0.0013621033895944627, + "loss": 0.0938, + "step": 39565 + }, + { + "epoch": 0.3434518797579882, + "grad_norm": 0.341796875, + "learning_rate": 0.0013620749170693588, + "loss": 0.0884, + "step": 39566 + }, + { + "epoch": 0.3434605602381924, + "grad_norm": 0.251953125, + "learning_rate": 0.0013620464442576537, + "loss": 0.0957, + "step": 39567 + }, + { + "epoch": 0.3434692407183965, + "grad_norm": 0.1083984375, + "learning_rate": 0.0013620179711593787, + "loss": 0.1055, + "step": 39568 + }, + { + "epoch": 0.3434779211986007, + "grad_norm": 0.11572265625, + "learning_rate": 0.0013619894977745647, + "loss": 0.0859, + "step": 39569 + }, + { + "epoch": 0.34348660167880485, + "grad_norm": 0.078125, + "learning_rate": 0.0013619610241032431, + "loss": 0.0859, + "step": 39570 + }, + { + "epoch": 0.34349528215900904, + "grad_norm": 0.474609375, + "learning_rate": 0.0013619325501454446, + "loss": 0.0688, + "step": 39571 + }, + { + "epoch": 0.3435039626392132, + "grad_norm": 0.1064453125, + "learning_rate": 0.0013619040759012006, + "loss": 0.1318, + "step": 39572 + }, + { + "epoch": 0.34351264311941737, + "grad_norm": 0.25, + "learning_rate": 0.0013618756013705424, + "loss": 0.1221, + "step": 39573 + }, + { + "epoch": 0.3435213235996215, + "grad_norm": 0.322265625, + "learning_rate": 0.0013618471265535012, + "loss": 0.0864, + "step": 39574 + }, + { + "epoch": 0.3435300040798257, + "grad_norm": 0.267578125, + "learning_rate": 0.0013618186514501077, + "loss": 0.1113, + "step": 39575 + }, + { + "epoch": 0.34353868456002984, + "grad_norm": 0.58984375, + "learning_rate": 0.001361790176060393, + "loss": 0.1328, + "step": 39576 + }, + { + "epoch": 0.34354736504023403, + "grad_norm": 0.236328125, + "learning_rate": 0.001361761700384389, + "loss": 0.0894, + "step": 39577 + }, + { + "epoch": 0.34355604552043817, + "grad_norm": 0.049072265625, + "learning_rate": 0.0013617332244221256, + "loss": 0.0537, + "step": 39578 + }, + { + "epoch": 0.34356472600064236, + "grad_norm": 0.259765625, + "learning_rate": 0.0013617047481736353, + "loss": 0.0918, + "step": 39579 + }, + { + "epoch": 0.3435734064808465, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013616762716389483, + "loss": 0.0781, + "step": 39580 + }, + { + "epoch": 0.3435820869610507, + "grad_norm": 0.171875, + "learning_rate": 0.0013616477948180964, + "loss": 0.0933, + "step": 39581 + }, + { + "epoch": 0.34359076744125483, + "grad_norm": 0.421875, + "learning_rate": 0.0013616193177111104, + "loss": 0.127, + "step": 39582 + }, + { + "epoch": 0.343599447921459, + "grad_norm": 2.015625, + "learning_rate": 0.001361590840318021, + "loss": 0.1982, + "step": 39583 + }, + { + "epoch": 0.34360812840166316, + "grad_norm": 0.166015625, + "learning_rate": 0.00136156236263886, + "loss": 0.0952, + "step": 39584 + }, + { + "epoch": 0.34361680888186735, + "grad_norm": 0.83984375, + "learning_rate": 0.0013615338846736584, + "loss": 0.1553, + "step": 39585 + }, + { + "epoch": 0.3436254893620715, + "grad_norm": 0.0634765625, + "learning_rate": 0.0013615054064224474, + "loss": 0.0884, + "step": 39586 + }, + { + "epoch": 0.3436341698422757, + "grad_norm": 0.65234375, + "learning_rate": 0.001361476927885258, + "loss": 0.126, + "step": 39587 + }, + { + "epoch": 0.3436428503224798, + "grad_norm": 0.71484375, + "learning_rate": 0.0013614484490621207, + "loss": 0.1348, + "step": 39588 + }, + { + "epoch": 0.343651530802684, + "grad_norm": 0.07763671875, + "learning_rate": 0.0013614199699530677, + "loss": 0.0762, + "step": 39589 + }, + { + "epoch": 0.34366021128288815, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013613914905581303, + "loss": 0.0908, + "step": 39590 + }, + { + "epoch": 0.34366889176309234, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013613630108773387, + "loss": 0.1025, + "step": 39591 + }, + { + "epoch": 0.3436775722432965, + "grad_norm": 0.09130859375, + "learning_rate": 0.0013613345309107243, + "loss": 0.1035, + "step": 39592 + }, + { + "epoch": 0.3436862527235007, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013613060506583183, + "loss": 0.0884, + "step": 39593 + }, + { + "epoch": 0.3436949332037048, + "grad_norm": 0.138671875, + "learning_rate": 0.0013612775701201522, + "loss": 0.0981, + "step": 39594 + }, + { + "epoch": 0.343703613683909, + "grad_norm": 0.16796875, + "learning_rate": 0.0013612490892962567, + "loss": 0.0801, + "step": 39595 + }, + { + "epoch": 0.34371229416411314, + "grad_norm": 0.5703125, + "learning_rate": 0.0013612206081866636, + "loss": 0.1416, + "step": 39596 + }, + { + "epoch": 0.34372097464431733, + "grad_norm": 0.1015625, + "learning_rate": 0.0013611921267914031, + "loss": 0.0986, + "step": 39597 + }, + { + "epoch": 0.34372965512452147, + "grad_norm": 0.5, + "learning_rate": 0.0013611636451105068, + "loss": 0.1348, + "step": 39598 + }, + { + "epoch": 0.34373833560472566, + "grad_norm": 0.55078125, + "learning_rate": 0.001361135163144006, + "loss": 0.1348, + "step": 39599 + }, + { + "epoch": 0.3437470160849298, + "grad_norm": 0.431640625, + "learning_rate": 0.0013611066808919318, + "loss": 0.1465, + "step": 39600 + }, + { + "epoch": 0.343755696565134, + "grad_norm": 0.123046875, + "learning_rate": 0.0013610781983543153, + "loss": 0.0952, + "step": 39601 + }, + { + "epoch": 0.34376437704533813, + "grad_norm": 0.06884765625, + "learning_rate": 0.0013610497155311873, + "loss": 0.0718, + "step": 39602 + }, + { + "epoch": 0.3437730575255423, + "grad_norm": 0.322265625, + "learning_rate": 0.0013610212324225792, + "loss": 0.1387, + "step": 39603 + }, + { + "epoch": 0.34378173800574646, + "grad_norm": 0.423828125, + "learning_rate": 0.0013609927490285223, + "loss": 0.1436, + "step": 39604 + }, + { + "epoch": 0.34379041848595066, + "grad_norm": 0.45703125, + "learning_rate": 0.001360964265349048, + "loss": 0.1504, + "step": 39605 + }, + { + "epoch": 0.3437990989661548, + "grad_norm": 0.3125, + "learning_rate": 0.0013609357813841865, + "loss": 0.1475, + "step": 39606 + }, + { + "epoch": 0.343807779446359, + "grad_norm": 0.82421875, + "learning_rate": 0.00136090729713397, + "loss": 0.1191, + "step": 39607 + }, + { + "epoch": 0.3438164599265631, + "grad_norm": 0.7421875, + "learning_rate": 0.0013608788125984291, + "loss": 0.1235, + "step": 39608 + }, + { + "epoch": 0.3438251404067673, + "grad_norm": 0.271484375, + "learning_rate": 0.001360850327777595, + "loss": 0.0742, + "step": 39609 + }, + { + "epoch": 0.34383382088697145, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013608218426714989, + "loss": 0.1147, + "step": 39610 + }, + { + "epoch": 0.34384250136717565, + "grad_norm": 0.07470703125, + "learning_rate": 0.001360793357280172, + "loss": 0.0688, + "step": 39611 + }, + { + "epoch": 0.3438511818473798, + "grad_norm": 0.1357421875, + "learning_rate": 0.0013607648716036454, + "loss": 0.123, + "step": 39612 + }, + { + "epoch": 0.343859862327584, + "grad_norm": 0.134765625, + "learning_rate": 0.0013607363856419504, + "loss": 0.0815, + "step": 39613 + }, + { + "epoch": 0.3438685428077881, + "grad_norm": 0.140625, + "learning_rate": 0.0013607078993951177, + "loss": 0.1006, + "step": 39614 + }, + { + "epoch": 0.3438772232879923, + "grad_norm": 0.376953125, + "learning_rate": 0.0013606794128631788, + "loss": 0.1123, + "step": 39615 + }, + { + "epoch": 0.34388590376819644, + "grad_norm": 0.48046875, + "learning_rate": 0.001360650926046165, + "loss": 0.1562, + "step": 39616 + }, + { + "epoch": 0.34389458424840064, + "grad_norm": 0.400390625, + "learning_rate": 0.001360622438944107, + "loss": 0.0947, + "step": 39617 + }, + { + "epoch": 0.3439032647286048, + "grad_norm": 0.412109375, + "learning_rate": 0.0013605939515570365, + "loss": 0.0806, + "step": 39618 + }, + { + "epoch": 0.34391194520880897, + "grad_norm": 0.1455078125, + "learning_rate": 0.0013605654638849842, + "loss": 0.1123, + "step": 39619 + }, + { + "epoch": 0.3439206256890131, + "grad_norm": 0.25, + "learning_rate": 0.0013605369759279815, + "loss": 0.0962, + "step": 39620 + }, + { + "epoch": 0.3439293061692173, + "grad_norm": 0.24609375, + "learning_rate": 0.0013605084876860596, + "loss": 0.0957, + "step": 39621 + }, + { + "epoch": 0.34393798664942143, + "grad_norm": 0.5390625, + "learning_rate": 0.0013604799991592496, + "loss": 0.104, + "step": 39622 + }, + { + "epoch": 0.3439466671296256, + "grad_norm": 0.54296875, + "learning_rate": 0.0013604515103475823, + "loss": 0.1064, + "step": 39623 + }, + { + "epoch": 0.34395534760982976, + "grad_norm": 0.369140625, + "learning_rate": 0.0013604230212510891, + "loss": 0.0947, + "step": 39624 + }, + { + "epoch": 0.34396402809003396, + "grad_norm": 0.28515625, + "learning_rate": 0.0013603945318698015, + "loss": 0.126, + "step": 39625 + }, + { + "epoch": 0.3439727085702381, + "grad_norm": 0.373046875, + "learning_rate": 0.00136036604220375, + "loss": 0.1445, + "step": 39626 + }, + { + "epoch": 0.3439813890504423, + "grad_norm": 0.271484375, + "learning_rate": 0.001360337552252966, + "loss": 0.1191, + "step": 39627 + }, + { + "epoch": 0.3439900695306464, + "grad_norm": 0.255859375, + "learning_rate": 0.001360309062017481, + "loss": 0.1035, + "step": 39628 + }, + { + "epoch": 0.3439987500108506, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013602805714973264, + "loss": 0.0869, + "step": 39629 + }, + { + "epoch": 0.34400743049105476, + "grad_norm": 0.255859375, + "learning_rate": 0.0013602520806925325, + "loss": 0.084, + "step": 39630 + }, + { + "epoch": 0.34401611097125895, + "grad_norm": 0.451171875, + "learning_rate": 0.0013602235896031309, + "loss": 0.0859, + "step": 39631 + }, + { + "epoch": 0.3440247914514631, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013601950982291525, + "loss": 0.0752, + "step": 39632 + }, + { + "epoch": 0.3440334719316673, + "grad_norm": 0.76171875, + "learning_rate": 0.0013601666065706287, + "loss": 0.1387, + "step": 39633 + }, + { + "epoch": 0.3440421524118714, + "grad_norm": 0.380859375, + "learning_rate": 0.0013601381146275907, + "loss": 0.0732, + "step": 39634 + }, + { + "epoch": 0.3440508328920756, + "grad_norm": 0.10693359375, + "learning_rate": 0.0013601096224000694, + "loss": 0.1108, + "step": 39635 + }, + { + "epoch": 0.34405951337227975, + "grad_norm": 0.76953125, + "learning_rate": 0.0013600811298880961, + "loss": 0.0977, + "step": 39636 + }, + { + "epoch": 0.34406819385248394, + "grad_norm": 0.4609375, + "learning_rate": 0.0013600526370917021, + "loss": 0.1602, + "step": 39637 + }, + { + "epoch": 0.3440768743326881, + "grad_norm": 0.115234375, + "learning_rate": 0.0013600241440109185, + "loss": 0.1016, + "step": 39638 + }, + { + "epoch": 0.34408555481289227, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013599956506457763, + "loss": 0.1025, + "step": 39639 + }, + { + "epoch": 0.3440942352930964, + "grad_norm": 0.326171875, + "learning_rate": 0.001359967156996307, + "loss": 0.0879, + "step": 39640 + }, + { + "epoch": 0.3441029157733006, + "grad_norm": 0.79296875, + "learning_rate": 0.0013599386630625413, + "loss": 0.1157, + "step": 39641 + }, + { + "epoch": 0.34411159625350474, + "grad_norm": 0.34375, + "learning_rate": 0.0013599101688445106, + "loss": 0.1045, + "step": 39642 + }, + { + "epoch": 0.34412027673370893, + "grad_norm": 0.09765625, + "learning_rate": 0.0013598816743422462, + "loss": 0.1162, + "step": 39643 + }, + { + "epoch": 0.34412895721391307, + "grad_norm": 0.15234375, + "learning_rate": 0.001359853179555779, + "loss": 0.1128, + "step": 39644 + }, + { + "epoch": 0.34413763769411726, + "grad_norm": 0.42578125, + "learning_rate": 0.0013598246844851402, + "loss": 0.0889, + "step": 39645 + }, + { + "epoch": 0.3441463181743214, + "grad_norm": 0.357421875, + "learning_rate": 0.0013597961891303608, + "loss": 0.1523, + "step": 39646 + }, + { + "epoch": 0.3441549986545256, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013597676934914726, + "loss": 0.0981, + "step": 39647 + }, + { + "epoch": 0.3441636791347297, + "grad_norm": 0.09521484375, + "learning_rate": 0.001359739197568506, + "loss": 0.0996, + "step": 39648 + }, + { + "epoch": 0.3441723596149339, + "grad_norm": 0.1767578125, + "learning_rate": 0.0013597107013614928, + "loss": 0.1426, + "step": 39649 + }, + { + "epoch": 0.34418104009513806, + "grad_norm": 0.1689453125, + "learning_rate": 0.001359682204870464, + "loss": 0.1094, + "step": 39650 + }, + { + "epoch": 0.34418972057534225, + "grad_norm": 0.85546875, + "learning_rate": 0.0013596537080954505, + "loss": 0.126, + "step": 39651 + }, + { + "epoch": 0.3441984010555464, + "grad_norm": 0.5703125, + "learning_rate": 0.0013596252110364836, + "loss": 0.1045, + "step": 39652 + }, + { + "epoch": 0.3442070815357506, + "grad_norm": 0.11572265625, + "learning_rate": 0.0013595967136935945, + "loss": 0.1079, + "step": 39653 + }, + { + "epoch": 0.3442157620159547, + "grad_norm": 0.671875, + "learning_rate": 0.0013595682160668142, + "loss": 0.1006, + "step": 39654 + }, + { + "epoch": 0.3442244424961589, + "grad_norm": 0.08984375, + "learning_rate": 0.0013595397181561737, + "loss": 0.0776, + "step": 39655 + }, + { + "epoch": 0.34423312297636305, + "grad_norm": 0.37890625, + "learning_rate": 0.0013595112199617052, + "loss": 0.1201, + "step": 39656 + }, + { + "epoch": 0.34424180345656724, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013594827214834387, + "loss": 0.0845, + "step": 39657 + }, + { + "epoch": 0.3442504839367714, + "grad_norm": 0.162109375, + "learning_rate": 0.0013594542227214057, + "loss": 0.1318, + "step": 39658 + }, + { + "epoch": 0.34425916441697557, + "grad_norm": 0.326171875, + "learning_rate": 0.0013594257236756375, + "loss": 0.1133, + "step": 39659 + }, + { + "epoch": 0.3442678448971797, + "grad_norm": 0.09814453125, + "learning_rate": 0.0013593972243461656, + "loss": 0.0884, + "step": 39660 + }, + { + "epoch": 0.3442765253773839, + "grad_norm": 0.06982421875, + "learning_rate": 0.0013593687247330207, + "loss": 0.0771, + "step": 39661 + }, + { + "epoch": 0.34428520585758804, + "grad_norm": 1.1484375, + "learning_rate": 0.0013593402248362337, + "loss": 0.1953, + "step": 39662 + }, + { + "epoch": 0.34429388633779223, + "grad_norm": 0.470703125, + "learning_rate": 0.0013593117246558364, + "loss": 0.1001, + "step": 39663 + }, + { + "epoch": 0.34430256681799637, + "grad_norm": 1.515625, + "learning_rate": 0.0013592832241918595, + "loss": 0.1113, + "step": 39664 + }, + { + "epoch": 0.34431124729820056, + "grad_norm": 0.234375, + "learning_rate": 0.0013592547234443344, + "loss": 0.1035, + "step": 39665 + }, + { + "epoch": 0.3443199277784047, + "grad_norm": 0.13671875, + "learning_rate": 0.0013592262224132923, + "loss": 0.0986, + "step": 39666 + }, + { + "epoch": 0.3443286082586089, + "grad_norm": 0.12060546875, + "learning_rate": 0.001359197721098764, + "loss": 0.0918, + "step": 39667 + }, + { + "epoch": 0.34433728873881303, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013591692195007816, + "loss": 0.0693, + "step": 39668 + }, + { + "epoch": 0.3443459692190172, + "grad_norm": 0.515625, + "learning_rate": 0.0013591407176193754, + "loss": 0.0952, + "step": 39669 + }, + { + "epoch": 0.34435464969922136, + "grad_norm": 0.94921875, + "learning_rate": 0.0013591122154545767, + "loss": 0.1094, + "step": 39670 + }, + { + "epoch": 0.34436333017942555, + "grad_norm": 0.453125, + "learning_rate": 0.0013590837130064166, + "loss": 0.1279, + "step": 39671 + }, + { + "epoch": 0.3443720106596297, + "grad_norm": 0.357421875, + "learning_rate": 0.0013590552102749267, + "loss": 0.1016, + "step": 39672 + }, + { + "epoch": 0.3443806911398339, + "grad_norm": 0.388671875, + "learning_rate": 0.001359026707260138, + "loss": 0.0884, + "step": 39673 + }, + { + "epoch": 0.344389371620038, + "grad_norm": 0.7265625, + "learning_rate": 0.0013589982039620813, + "loss": 0.0942, + "step": 39674 + }, + { + "epoch": 0.34439805210024216, + "grad_norm": 0.52734375, + "learning_rate": 0.001358969700380788, + "loss": 0.1245, + "step": 39675 + }, + { + "epoch": 0.34440673258044635, + "grad_norm": 0.6015625, + "learning_rate": 0.0013589411965162896, + "loss": 0.0728, + "step": 39676 + }, + { + "epoch": 0.3444154130606505, + "grad_norm": 0.41015625, + "learning_rate": 0.001358912692368617, + "loss": 0.1084, + "step": 39677 + }, + { + "epoch": 0.3444240935408547, + "grad_norm": 0.8984375, + "learning_rate": 0.0013588841879378013, + "loss": 0.1729, + "step": 39678 + }, + { + "epoch": 0.3444327740210588, + "grad_norm": 0.34765625, + "learning_rate": 0.001358855683223874, + "loss": 0.0972, + "step": 39679 + }, + { + "epoch": 0.344441454501263, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013588271782268655, + "loss": 0.1494, + "step": 39680 + }, + { + "epoch": 0.34445013498146715, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013587986729468077, + "loss": 0.0947, + "step": 39681 + }, + { + "epoch": 0.34445881546167134, + "grad_norm": 0.11328125, + "learning_rate": 0.0013587701673837317, + "loss": 0.1338, + "step": 39682 + }, + { + "epoch": 0.3444674959418755, + "grad_norm": 0.185546875, + "learning_rate": 0.0013587416615376683, + "loss": 0.0869, + "step": 39683 + }, + { + "epoch": 0.34447617642207967, + "grad_norm": 0.1669921875, + "learning_rate": 0.001358713155408649, + "loss": 0.0825, + "step": 39684 + }, + { + "epoch": 0.3444848569022838, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013586846489967049, + "loss": 0.1133, + "step": 39685 + }, + { + "epoch": 0.344493537382488, + "grad_norm": 0.283203125, + "learning_rate": 0.001358656142301867, + "loss": 0.1279, + "step": 39686 + }, + { + "epoch": 0.34450221786269214, + "grad_norm": 0.8828125, + "learning_rate": 0.0013586276353241668, + "loss": 0.1543, + "step": 39687 + }, + { + "epoch": 0.34451089834289633, + "grad_norm": 0.126953125, + "learning_rate": 0.0013585991280636353, + "loss": 0.1045, + "step": 39688 + }, + { + "epoch": 0.34451957882310047, + "grad_norm": 0.11767578125, + "learning_rate": 0.0013585706205203037, + "loss": 0.127, + "step": 39689 + }, + { + "epoch": 0.34452825930330466, + "grad_norm": 0.193359375, + "learning_rate": 0.001358542112694203, + "loss": 0.0527, + "step": 39690 + }, + { + "epoch": 0.3445369397835088, + "grad_norm": 0.1162109375, + "learning_rate": 0.001358513604585365, + "loss": 0.1006, + "step": 39691 + }, + { + "epoch": 0.344545620263713, + "grad_norm": 0.193359375, + "learning_rate": 0.0013584850961938198, + "loss": 0.1025, + "step": 39692 + }, + { + "epoch": 0.34455430074391713, + "grad_norm": 0.087890625, + "learning_rate": 0.0013584565875195995, + "loss": 0.0791, + "step": 39693 + }, + { + "epoch": 0.3445629812241213, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013584280785627347, + "loss": 0.0732, + "step": 39694 + }, + { + "epoch": 0.34457166170432546, + "grad_norm": 0.5234375, + "learning_rate": 0.001358399569323257, + "loss": 0.1191, + "step": 39695 + }, + { + "epoch": 0.34458034218452965, + "grad_norm": 0.2216796875, + "learning_rate": 0.0013583710598011976, + "loss": 0.0967, + "step": 39696 + }, + { + "epoch": 0.3445890226647338, + "grad_norm": 0.55078125, + "learning_rate": 0.0013583425499965873, + "loss": 0.1357, + "step": 39697 + }, + { + "epoch": 0.344597703144938, + "grad_norm": 0.796875, + "learning_rate": 0.0013583140399094575, + "loss": 0.1221, + "step": 39698 + }, + { + "epoch": 0.3446063836251421, + "grad_norm": 0.111328125, + "learning_rate": 0.0013582855295398396, + "loss": 0.0967, + "step": 39699 + }, + { + "epoch": 0.3446150641053463, + "grad_norm": 0.62890625, + "learning_rate": 0.0013582570188877643, + "loss": 0.1069, + "step": 39700 + }, + { + "epoch": 0.34462374458555045, + "grad_norm": 0.267578125, + "learning_rate": 0.0013582285079532627, + "loss": 0.1094, + "step": 39701 + }, + { + "epoch": 0.34463242506575464, + "grad_norm": 0.1572265625, + "learning_rate": 0.0013581999967363667, + "loss": 0.0845, + "step": 39702 + }, + { + "epoch": 0.3446411055459588, + "grad_norm": 0.11962890625, + "learning_rate": 0.0013581714852371068, + "loss": 0.1289, + "step": 39703 + }, + { + "epoch": 0.344649786026163, + "grad_norm": 0.384765625, + "learning_rate": 0.0013581429734555147, + "loss": 0.1113, + "step": 39704 + }, + { + "epoch": 0.3446584665063671, + "grad_norm": 0.2890625, + "learning_rate": 0.001358114461391621, + "loss": 0.0928, + "step": 39705 + }, + { + "epoch": 0.3446671469865713, + "grad_norm": 0.55859375, + "learning_rate": 0.0013580859490454577, + "loss": 0.1074, + "step": 39706 + }, + { + "epoch": 0.34467582746677544, + "grad_norm": 0.1123046875, + "learning_rate": 0.001358057436417055, + "loss": 0.1216, + "step": 39707 + }, + { + "epoch": 0.34468450794697963, + "grad_norm": 0.0888671875, + "learning_rate": 0.001358028923506445, + "loss": 0.0845, + "step": 39708 + }, + { + "epoch": 0.34469318842718377, + "grad_norm": 0.455078125, + "learning_rate": 0.001358000410313658, + "loss": 0.1123, + "step": 39709 + }, + { + "epoch": 0.34470186890738796, + "grad_norm": 0.482421875, + "learning_rate": 0.0013579718968387258, + "loss": 0.0957, + "step": 39710 + }, + { + "epoch": 0.3447105493875921, + "grad_norm": 0.11328125, + "learning_rate": 0.0013579433830816794, + "loss": 0.1387, + "step": 39711 + }, + { + "epoch": 0.3447192298677963, + "grad_norm": 0.21875, + "learning_rate": 0.0013579148690425502, + "loss": 0.0947, + "step": 39712 + }, + { + "epoch": 0.34472791034800043, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013578863547213686, + "loss": 0.1055, + "step": 39713 + }, + { + "epoch": 0.3447365908282046, + "grad_norm": 0.427734375, + "learning_rate": 0.001357857840118167, + "loss": 0.1128, + "step": 39714 + }, + { + "epoch": 0.34474527130840876, + "grad_norm": 0.2109375, + "learning_rate": 0.0013578293252329756, + "loss": 0.1211, + "step": 39715 + }, + { + "epoch": 0.34475395178861296, + "grad_norm": 0.1220703125, + "learning_rate": 0.0013578008100658262, + "loss": 0.1235, + "step": 39716 + }, + { + "epoch": 0.3447626322688171, + "grad_norm": 0.6640625, + "learning_rate": 0.0013577722946167495, + "loss": 0.1465, + "step": 39717 + }, + { + "epoch": 0.3447713127490213, + "grad_norm": 0.076171875, + "learning_rate": 0.0013577437788857766, + "loss": 0.0854, + "step": 39718 + }, + { + "epoch": 0.3447799932292254, + "grad_norm": 0.3828125, + "learning_rate": 0.0013577152628729394, + "loss": 0.0996, + "step": 39719 + }, + { + "epoch": 0.3447886737094296, + "grad_norm": 0.6796875, + "learning_rate": 0.0013576867465782687, + "loss": 0.0879, + "step": 39720 + }, + { + "epoch": 0.34479735418963375, + "grad_norm": 0.328125, + "learning_rate": 0.0013576582300017955, + "loss": 0.1191, + "step": 39721 + }, + { + "epoch": 0.34480603466983795, + "grad_norm": 0.126953125, + "learning_rate": 0.0013576297131435511, + "loss": 0.1191, + "step": 39722 + }, + { + "epoch": 0.3448147151500421, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013576011960035668, + "loss": 0.0884, + "step": 39723 + }, + { + "epoch": 0.3448233956302463, + "grad_norm": 0.09228515625, + "learning_rate": 0.0013575726785818735, + "loss": 0.0791, + "step": 39724 + }, + { + "epoch": 0.3448320761104504, + "grad_norm": 0.4375, + "learning_rate": 0.0013575441608785026, + "loss": 0.1172, + "step": 39725 + }, + { + "epoch": 0.3448407565906546, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013575156428934855, + "loss": 0.1426, + "step": 39726 + }, + { + "epoch": 0.34484943707085874, + "grad_norm": 0.1416015625, + "learning_rate": 0.001357487124626853, + "loss": 0.0952, + "step": 39727 + }, + { + "epoch": 0.34485811755106294, + "grad_norm": 0.173828125, + "learning_rate": 0.0013574586060786367, + "loss": 0.1172, + "step": 39728 + }, + { + "epoch": 0.3448667980312671, + "grad_norm": 0.059326171875, + "learning_rate": 0.0013574300872488676, + "loss": 0.052, + "step": 39729 + }, + { + "epoch": 0.34487547851147127, + "grad_norm": 0.0732421875, + "learning_rate": 0.0013574015681375767, + "loss": 0.0718, + "step": 39730 + }, + { + "epoch": 0.3448841589916754, + "grad_norm": 0.08251953125, + "learning_rate": 0.001357373048744795, + "loss": 0.0918, + "step": 39731 + }, + { + "epoch": 0.3448928394718796, + "grad_norm": 0.091796875, + "learning_rate": 0.0013573445290705542, + "loss": 0.0918, + "step": 39732 + }, + { + "epoch": 0.34490151995208373, + "grad_norm": 0.1982421875, + "learning_rate": 0.001357316009114885, + "loss": 0.0918, + "step": 39733 + }, + { + "epoch": 0.34491020043228793, + "grad_norm": 0.51171875, + "learning_rate": 0.0013572874888778193, + "loss": 0.1016, + "step": 39734 + }, + { + "epoch": 0.34491888091249207, + "grad_norm": 0.294921875, + "learning_rate": 0.001357258968359388, + "loss": 0.0996, + "step": 39735 + }, + { + "epoch": 0.34492756139269626, + "grad_norm": 0.67578125, + "learning_rate": 0.0013572304475596216, + "loss": 0.1133, + "step": 39736 + }, + { + "epoch": 0.3449362418729004, + "grad_norm": 0.5546875, + "learning_rate": 0.0013572019264785522, + "loss": 0.082, + "step": 39737 + }, + { + "epoch": 0.3449449223531046, + "grad_norm": 0.19921875, + "learning_rate": 0.0013571734051162107, + "loss": 0.0884, + "step": 39738 + }, + { + "epoch": 0.3449536028333087, + "grad_norm": 0.1787109375, + "learning_rate": 0.0013571448834726284, + "loss": 0.1162, + "step": 39739 + }, + { + "epoch": 0.3449622833135129, + "grad_norm": 0.40234375, + "learning_rate": 0.0013571163615478357, + "loss": 0.1064, + "step": 39740 + }, + { + "epoch": 0.34497096379371706, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013570878393418648, + "loss": 0.0967, + "step": 39741 + }, + { + "epoch": 0.34497964427392125, + "grad_norm": 0.4765625, + "learning_rate": 0.0013570593168547463, + "loss": 0.0947, + "step": 39742 + }, + { + "epoch": 0.3449883247541254, + "grad_norm": 0.11376953125, + "learning_rate": 0.0013570307940865118, + "loss": 0.084, + "step": 39743 + }, + { + "epoch": 0.3449970052343296, + "grad_norm": 0.2236328125, + "learning_rate": 0.001357002271037192, + "loss": 0.0825, + "step": 39744 + }, + { + "epoch": 0.3450056857145337, + "grad_norm": 0.466796875, + "learning_rate": 0.0013569737477068188, + "loss": 0.1045, + "step": 39745 + }, + { + "epoch": 0.3450143661947379, + "grad_norm": 0.5, + "learning_rate": 0.0013569452240954228, + "loss": 0.1021, + "step": 39746 + }, + { + "epoch": 0.34502304667494205, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013569167002030352, + "loss": 0.0776, + "step": 39747 + }, + { + "epoch": 0.34503172715514624, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013568881760296877, + "loss": 0.0967, + "step": 39748 + }, + { + "epoch": 0.3450404076353504, + "grad_norm": 0.318359375, + "learning_rate": 0.0013568596515754108, + "loss": 0.0996, + "step": 39749 + }, + { + "epoch": 0.34504908811555457, + "grad_norm": 0.193359375, + "learning_rate": 0.0013568311268402362, + "loss": 0.1152, + "step": 39750 + }, + { + "epoch": 0.3450577685957587, + "grad_norm": 0.09130859375, + "learning_rate": 0.0013568026018241947, + "loss": 0.0889, + "step": 39751 + }, + { + "epoch": 0.3450664490759629, + "grad_norm": 0.5234375, + "learning_rate": 0.001356774076527318, + "loss": 0.1484, + "step": 39752 + }, + { + "epoch": 0.34507512955616704, + "grad_norm": 0.427734375, + "learning_rate": 0.001356745550949637, + "loss": 0.123, + "step": 39753 + }, + { + "epoch": 0.34508381003637123, + "grad_norm": 0.1181640625, + "learning_rate": 0.001356717025091183, + "loss": 0.0957, + "step": 39754 + }, + { + "epoch": 0.34509249051657537, + "grad_norm": 0.0849609375, + "learning_rate": 0.0013566884989519871, + "loss": 0.0771, + "step": 39755 + }, + { + "epoch": 0.34510117099677956, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013566599725320806, + "loss": 0.1045, + "step": 39756 + }, + { + "epoch": 0.3451098514769837, + "grad_norm": 0.314453125, + "learning_rate": 0.0013566314458314942, + "loss": 0.1279, + "step": 39757 + }, + { + "epoch": 0.3451185319571879, + "grad_norm": 0.107421875, + "learning_rate": 0.0013566029188502597, + "loss": 0.1006, + "step": 39758 + }, + { + "epoch": 0.34512721243739203, + "grad_norm": 0.21875, + "learning_rate": 0.0013565743915884083, + "loss": 0.1084, + "step": 39759 + }, + { + "epoch": 0.3451358929175962, + "grad_norm": 0.181640625, + "learning_rate": 0.0013565458640459706, + "loss": 0.1543, + "step": 39760 + }, + { + "epoch": 0.34514457339780036, + "grad_norm": 0.390625, + "learning_rate": 0.0013565173362229785, + "loss": 0.1016, + "step": 39761 + }, + { + "epoch": 0.34515325387800455, + "grad_norm": 0.12109375, + "learning_rate": 0.001356488808119463, + "loss": 0.0815, + "step": 39762 + }, + { + "epoch": 0.3451619343582087, + "grad_norm": 0.55859375, + "learning_rate": 0.0013564602797354547, + "loss": 0.127, + "step": 39763 + }, + { + "epoch": 0.3451706148384129, + "grad_norm": 0.0966796875, + "learning_rate": 0.0013564317510709857, + "loss": 0.0981, + "step": 39764 + }, + { + "epoch": 0.345179295318617, + "grad_norm": 0.302734375, + "learning_rate": 0.0013564032221260872, + "loss": 0.0957, + "step": 39765 + }, + { + "epoch": 0.3451879757988212, + "grad_norm": 0.1611328125, + "learning_rate": 0.001356374692900789, + "loss": 0.0908, + "step": 39766 + }, + { + "epoch": 0.34519665627902535, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013563461633951238, + "loss": 0.1064, + "step": 39767 + }, + { + "epoch": 0.34520533675922954, + "grad_norm": 0.2119140625, + "learning_rate": 0.0013563176336091225, + "loss": 0.0845, + "step": 39768 + }, + { + "epoch": 0.3452140172394337, + "grad_norm": 0.8046875, + "learning_rate": 0.001356289103542816, + "loss": 0.0894, + "step": 39769 + }, + { + "epoch": 0.34522269771963787, + "grad_norm": 0.1630859375, + "learning_rate": 0.0013562605731962355, + "loss": 0.1465, + "step": 39770 + }, + { + "epoch": 0.345231378199842, + "grad_norm": 0.27734375, + "learning_rate": 0.001356232042569412, + "loss": 0.0732, + "step": 39771 + }, + { + "epoch": 0.3452400586800462, + "grad_norm": 0.2216796875, + "learning_rate": 0.0013562035116623772, + "loss": 0.1035, + "step": 39772 + }, + { + "epoch": 0.34524873916025034, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013561749804751621, + "loss": 0.1143, + "step": 39773 + }, + { + "epoch": 0.34525741964045453, + "grad_norm": 0.474609375, + "learning_rate": 0.001356146449007798, + "loss": 0.1504, + "step": 39774 + }, + { + "epoch": 0.34526610012065867, + "grad_norm": 0.62109375, + "learning_rate": 0.0013561179172603159, + "loss": 0.1328, + "step": 39775 + }, + { + "epoch": 0.34527478060086286, + "grad_norm": 0.48046875, + "learning_rate": 0.0013560893852327472, + "loss": 0.168, + "step": 39776 + }, + { + "epoch": 0.345283461081067, + "grad_norm": 0.48828125, + "learning_rate": 0.001356060852925123, + "loss": 0.1387, + "step": 39777 + }, + { + "epoch": 0.3452921415612712, + "grad_norm": 0.10595703125, + "learning_rate": 0.0013560323203374745, + "loss": 0.1074, + "step": 39778 + }, + { + "epoch": 0.34530082204147533, + "grad_norm": 0.75390625, + "learning_rate": 0.0013560037874698326, + "loss": 0.1348, + "step": 39779 + }, + { + "epoch": 0.3453095025216795, + "grad_norm": 0.408203125, + "learning_rate": 0.0013559752543222288, + "loss": 0.1406, + "step": 39780 + }, + { + "epoch": 0.34531818300188366, + "grad_norm": 0.625, + "learning_rate": 0.0013559467208946947, + "loss": 0.127, + "step": 39781 + }, + { + "epoch": 0.34532686348208785, + "grad_norm": 0.25390625, + "learning_rate": 0.0013559181871872607, + "loss": 0.0898, + "step": 39782 + }, + { + "epoch": 0.345335543962292, + "grad_norm": 0.1591796875, + "learning_rate": 0.0013558896531999586, + "loss": 0.1289, + "step": 39783 + }, + { + "epoch": 0.3453442244424962, + "grad_norm": 0.134765625, + "learning_rate": 0.0013558611189328196, + "loss": 0.0996, + "step": 39784 + }, + { + "epoch": 0.3453529049227003, + "grad_norm": 0.458984375, + "learning_rate": 0.0013558325843858747, + "loss": 0.0972, + "step": 39785 + }, + { + "epoch": 0.3453615854029045, + "grad_norm": 0.26953125, + "learning_rate": 0.0013558040495591553, + "loss": 0.0977, + "step": 39786 + }, + { + "epoch": 0.34537026588310865, + "grad_norm": 0.1455078125, + "learning_rate": 0.001355775514452692, + "loss": 0.1064, + "step": 39787 + }, + { + "epoch": 0.34537894636331284, + "grad_norm": 0.255859375, + "learning_rate": 0.0013557469790665163, + "loss": 0.0996, + "step": 39788 + }, + { + "epoch": 0.345387626843517, + "grad_norm": 0.2294921875, + "learning_rate": 0.00135571844340066, + "loss": 0.123, + "step": 39789 + }, + { + "epoch": 0.3453963073237212, + "grad_norm": 0.1171875, + "learning_rate": 0.0013556899074551535, + "loss": 0.1016, + "step": 39790 + }, + { + "epoch": 0.3454049878039253, + "grad_norm": 0.20703125, + "learning_rate": 0.0013556613712300287, + "loss": 0.0757, + "step": 39791 + }, + { + "epoch": 0.3454136682841295, + "grad_norm": 0.31640625, + "learning_rate": 0.0013556328347253163, + "loss": 0.0884, + "step": 39792 + }, + { + "epoch": 0.34542234876433364, + "grad_norm": 0.52734375, + "learning_rate": 0.0013556042979410478, + "loss": 0.0801, + "step": 39793 + }, + { + "epoch": 0.34543102924453783, + "grad_norm": 0.08984375, + "learning_rate": 0.0013555757608772541, + "loss": 0.0752, + "step": 39794 + }, + { + "epoch": 0.345439709724742, + "grad_norm": 0.6328125, + "learning_rate": 0.0013555472235339664, + "loss": 0.0981, + "step": 39795 + }, + { + "epoch": 0.34544839020494617, + "grad_norm": 0.255859375, + "learning_rate": 0.0013555186859112165, + "loss": 0.1191, + "step": 39796 + }, + { + "epoch": 0.3454570706851503, + "grad_norm": 0.30859375, + "learning_rate": 0.0013554901480090351, + "loss": 0.1504, + "step": 39797 + }, + { + "epoch": 0.34546575116535444, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013554616098274534, + "loss": 0.1318, + "step": 39798 + }, + { + "epoch": 0.34547443164555863, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013554330713665028, + "loss": 0.1006, + "step": 39799 + }, + { + "epoch": 0.34548311212576277, + "grad_norm": 0.11865234375, + "learning_rate": 0.0013554045326262146, + "loss": 0.1064, + "step": 39800 + }, + { + "epoch": 0.34549179260596696, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013553759936066195, + "loss": 0.0967, + "step": 39801 + }, + { + "epoch": 0.3455004730861711, + "grad_norm": 0.240234375, + "learning_rate": 0.0013553474543077492, + "loss": 0.1367, + "step": 39802 + }, + { + "epoch": 0.3455091535663753, + "grad_norm": 0.158203125, + "learning_rate": 0.0013553189147296348, + "loss": 0.123, + "step": 39803 + }, + { + "epoch": 0.34551783404657943, + "grad_norm": 0.2060546875, + "learning_rate": 0.0013552903748723073, + "loss": 0.0894, + "step": 39804 + }, + { + "epoch": 0.3455265145267836, + "grad_norm": 0.07666015625, + "learning_rate": 0.0013552618347357982, + "loss": 0.0879, + "step": 39805 + }, + { + "epoch": 0.34553519500698776, + "grad_norm": 0.5546875, + "learning_rate": 0.0013552332943201387, + "loss": 0.0801, + "step": 39806 + }, + { + "epoch": 0.34554387548719195, + "grad_norm": 0.55078125, + "learning_rate": 0.00135520475362536, + "loss": 0.1406, + "step": 39807 + }, + { + "epoch": 0.3455525559673961, + "grad_norm": 0.44140625, + "learning_rate": 0.0013551762126514928, + "loss": 0.085, + "step": 39808 + }, + { + "epoch": 0.3455612364476003, + "grad_norm": 0.1982421875, + "learning_rate": 0.001355147671398569, + "loss": 0.0884, + "step": 39809 + }, + { + "epoch": 0.3455699169278044, + "grad_norm": 0.146484375, + "learning_rate": 0.0013551191298666195, + "loss": 0.0977, + "step": 39810 + }, + { + "epoch": 0.3455785974080086, + "grad_norm": 0.1005859375, + "learning_rate": 0.0013550905880556753, + "loss": 0.1338, + "step": 39811 + }, + { + "epoch": 0.34558727788821275, + "grad_norm": 1.40625, + "learning_rate": 0.0013550620459657685, + "loss": 0.1147, + "step": 39812 + }, + { + "epoch": 0.34559595836841694, + "grad_norm": 0.6875, + "learning_rate": 0.001355033503596929, + "loss": 0.1348, + "step": 39813 + }, + { + "epoch": 0.3456046388486211, + "grad_norm": 0.06787109375, + "learning_rate": 0.0013550049609491893, + "loss": 0.0737, + "step": 39814 + }, + { + "epoch": 0.3456133193288253, + "grad_norm": 0.267578125, + "learning_rate": 0.0013549764180225796, + "loss": 0.125, + "step": 39815 + }, + { + "epoch": 0.3456219998090294, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013549478748171317, + "loss": 0.0938, + "step": 39816 + }, + { + "epoch": 0.3456306802892336, + "grad_norm": 0.8359375, + "learning_rate": 0.0013549193313328765, + "loss": 0.1206, + "step": 39817 + }, + { + "epoch": 0.34563936076943774, + "grad_norm": 0.34765625, + "learning_rate": 0.0013548907875698454, + "loss": 0.1221, + "step": 39818 + }, + { + "epoch": 0.34564804124964194, + "grad_norm": 0.490234375, + "learning_rate": 0.0013548622435280695, + "loss": 0.0786, + "step": 39819 + }, + { + "epoch": 0.3456567217298461, + "grad_norm": 0.427734375, + "learning_rate": 0.0013548336992075802, + "loss": 0.0874, + "step": 39820 + }, + { + "epoch": 0.34566540221005027, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013548051546084085, + "loss": 0.1113, + "step": 39821 + }, + { + "epoch": 0.3456740826902544, + "grad_norm": 0.2353515625, + "learning_rate": 0.0013547766097305855, + "loss": 0.0938, + "step": 39822 + }, + { + "epoch": 0.3456827631704586, + "grad_norm": 0.1533203125, + "learning_rate": 0.001354748064574143, + "loss": 0.1045, + "step": 39823 + }, + { + "epoch": 0.34569144365066273, + "grad_norm": 0.21875, + "learning_rate": 0.001354719519139112, + "loss": 0.1172, + "step": 39824 + }, + { + "epoch": 0.3457001241308669, + "grad_norm": 0.2392578125, + "learning_rate": 0.0013546909734255233, + "loss": 0.1094, + "step": 39825 + }, + { + "epoch": 0.34570880461107106, + "grad_norm": 0.169921875, + "learning_rate": 0.001354662427433408, + "loss": 0.082, + "step": 39826 + }, + { + "epoch": 0.34571748509127526, + "grad_norm": 1.0078125, + "learning_rate": 0.0013546338811627984, + "loss": 0.1406, + "step": 39827 + }, + { + "epoch": 0.3457261655714794, + "grad_norm": 0.23828125, + "learning_rate": 0.0013546053346137244, + "loss": 0.1377, + "step": 39828 + }, + { + "epoch": 0.3457348460516836, + "grad_norm": 0.10302734375, + "learning_rate": 0.0013545767877862183, + "loss": 0.0859, + "step": 39829 + }, + { + "epoch": 0.3457435265318877, + "grad_norm": 0.115234375, + "learning_rate": 0.0013545482406803107, + "loss": 0.1084, + "step": 39830 + }, + { + "epoch": 0.3457522070120919, + "grad_norm": 0.140625, + "learning_rate": 0.0013545196932960326, + "loss": 0.0771, + "step": 39831 + }, + { + "epoch": 0.34576088749229605, + "grad_norm": 0.60546875, + "learning_rate": 0.0013544911456334161, + "loss": 0.127, + "step": 39832 + }, + { + "epoch": 0.34576956797250025, + "grad_norm": 0.609375, + "learning_rate": 0.0013544625976924917, + "loss": 0.085, + "step": 39833 + }, + { + "epoch": 0.3457782484527044, + "grad_norm": 0.263671875, + "learning_rate": 0.001354434049473291, + "loss": 0.0947, + "step": 39834 + }, + { + "epoch": 0.3457869289329086, + "grad_norm": 0.2080078125, + "learning_rate": 0.001354405500975845, + "loss": 0.1182, + "step": 39835 + }, + { + "epoch": 0.3457956094131127, + "grad_norm": 0.1904296875, + "learning_rate": 0.0013543769522001848, + "loss": 0.0942, + "step": 39836 + }, + { + "epoch": 0.3458042898933169, + "grad_norm": 0.265625, + "learning_rate": 0.0013543484031463418, + "loss": 0.1016, + "step": 39837 + }, + { + "epoch": 0.34581297037352104, + "grad_norm": 0.40625, + "learning_rate": 0.001354319853814347, + "loss": 0.1104, + "step": 39838 + }, + { + "epoch": 0.34582165085372524, + "grad_norm": 1.0234375, + "learning_rate": 0.001354291304204232, + "loss": 0.1279, + "step": 39839 + }, + { + "epoch": 0.3458303313339294, + "grad_norm": 0.4375, + "learning_rate": 0.0013542627543160282, + "loss": 0.1006, + "step": 39840 + }, + { + "epoch": 0.34583901181413357, + "grad_norm": 0.310546875, + "learning_rate": 0.0013542342041497663, + "loss": 0.0728, + "step": 39841 + }, + { + "epoch": 0.3458476922943377, + "grad_norm": 0.22265625, + "learning_rate": 0.0013542056537054774, + "loss": 0.1123, + "step": 39842 + }, + { + "epoch": 0.3458563727745419, + "grad_norm": 0.154296875, + "learning_rate": 0.0013541771029831934, + "loss": 0.0967, + "step": 39843 + }, + { + "epoch": 0.34586505325474604, + "grad_norm": 0.287109375, + "learning_rate": 0.001354148551982945, + "loss": 0.1309, + "step": 39844 + }, + { + "epoch": 0.34587373373495023, + "grad_norm": 0.1220703125, + "learning_rate": 0.0013541200007047634, + "loss": 0.0884, + "step": 39845 + }, + { + "epoch": 0.34588241421515437, + "grad_norm": 0.1875, + "learning_rate": 0.0013540914491486802, + "loss": 0.0825, + "step": 39846 + }, + { + "epoch": 0.34589109469535856, + "grad_norm": 0.232421875, + "learning_rate": 0.001354062897314726, + "loss": 0.1104, + "step": 39847 + }, + { + "epoch": 0.3458997751755627, + "grad_norm": 0.255859375, + "learning_rate": 0.0013540343452029327, + "loss": 0.124, + "step": 39848 + }, + { + "epoch": 0.3459084556557669, + "grad_norm": 0.2451171875, + "learning_rate": 0.0013540057928133314, + "loss": 0.082, + "step": 39849 + }, + { + "epoch": 0.345917136135971, + "grad_norm": 0.63671875, + "learning_rate": 0.0013539772401459532, + "loss": 0.085, + "step": 39850 + }, + { + "epoch": 0.3459258166161752, + "grad_norm": 0.267578125, + "learning_rate": 0.0013539486872008292, + "loss": 0.1133, + "step": 39851 + }, + { + "epoch": 0.34593449709637936, + "grad_norm": 0.6328125, + "learning_rate": 0.0013539201339779906, + "loss": 0.0708, + "step": 39852 + }, + { + "epoch": 0.34594317757658355, + "grad_norm": 0.333984375, + "learning_rate": 0.001353891580477469, + "loss": 0.1089, + "step": 39853 + }, + { + "epoch": 0.3459518580567877, + "grad_norm": 0.921875, + "learning_rate": 0.0013538630266992952, + "loss": 0.1162, + "step": 39854 + }, + { + "epoch": 0.3459605385369919, + "grad_norm": 0.349609375, + "learning_rate": 0.0013538344726435009, + "loss": 0.0801, + "step": 39855 + }, + { + "epoch": 0.345969219017196, + "grad_norm": 0.66015625, + "learning_rate": 0.0013538059183101163, + "loss": 0.0918, + "step": 39856 + }, + { + "epoch": 0.3459778994974002, + "grad_norm": 0.91796875, + "learning_rate": 0.0013537773636991739, + "loss": 0.1133, + "step": 39857 + }, + { + "epoch": 0.34598657997760435, + "grad_norm": 0.1513671875, + "learning_rate": 0.0013537488088107047, + "loss": 0.0938, + "step": 39858 + }, + { + "epoch": 0.34599526045780854, + "grad_norm": 0.27734375, + "learning_rate": 0.001353720253644739, + "loss": 0.1089, + "step": 39859 + }, + { + "epoch": 0.3460039409380127, + "grad_norm": 0.265625, + "learning_rate": 0.001353691698201309, + "loss": 0.0801, + "step": 39860 + }, + { + "epoch": 0.34601262141821687, + "grad_norm": 0.1767578125, + "learning_rate": 0.0013536631424804454, + "loss": 0.083, + "step": 39861 + }, + { + "epoch": 0.346021301898421, + "grad_norm": 0.296875, + "learning_rate": 0.0013536345864821797, + "loss": 0.1729, + "step": 39862 + }, + { + "epoch": 0.3460299823786252, + "grad_norm": 0.111328125, + "learning_rate": 0.0013536060302065432, + "loss": 0.1279, + "step": 39863 + }, + { + "epoch": 0.34603866285882934, + "grad_norm": 0.2421875, + "learning_rate": 0.0013535774736535666, + "loss": 0.064, + "step": 39864 + }, + { + "epoch": 0.34604734333903353, + "grad_norm": 0.12451171875, + "learning_rate": 0.0013535489168232818, + "loss": 0.0825, + "step": 39865 + }, + { + "epoch": 0.34605602381923767, + "grad_norm": 0.34765625, + "learning_rate": 0.0013535203597157194, + "loss": 0.1016, + "step": 39866 + }, + { + "epoch": 0.34606470429944186, + "grad_norm": 0.271484375, + "learning_rate": 0.001353491802330911, + "loss": 0.0757, + "step": 39867 + }, + { + "epoch": 0.346073384779646, + "grad_norm": 0.57421875, + "learning_rate": 0.0013534632446688879, + "loss": 0.1094, + "step": 39868 + }, + { + "epoch": 0.3460820652598502, + "grad_norm": 0.28515625, + "learning_rate": 0.0013534346867296813, + "loss": 0.166, + "step": 39869 + }, + { + "epoch": 0.34609074574005433, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013534061285133223, + "loss": 0.123, + "step": 39870 + }, + { + "epoch": 0.3460994262202585, + "grad_norm": 0.296875, + "learning_rate": 0.001353377570019842, + "loss": 0.1016, + "step": 39871 + }, + { + "epoch": 0.34610810670046266, + "grad_norm": 0.28125, + "learning_rate": 0.001353349011249272, + "loss": 0.1426, + "step": 39872 + }, + { + "epoch": 0.34611678718066685, + "grad_norm": 0.10400390625, + "learning_rate": 0.0013533204522016432, + "loss": 0.0771, + "step": 39873 + }, + { + "epoch": 0.346125467660871, + "grad_norm": 1.03125, + "learning_rate": 0.001353291892876987, + "loss": 0.5586, + "step": 39874 + }, + { + "epoch": 0.3461341481410752, + "grad_norm": 0.283203125, + "learning_rate": 0.0013532633332753344, + "loss": 0.0991, + "step": 39875 + }, + { + "epoch": 0.3461428286212793, + "grad_norm": 0.62109375, + "learning_rate": 0.0013532347733967167, + "loss": 0.1123, + "step": 39876 + }, + { + "epoch": 0.3461515091014835, + "grad_norm": 0.515625, + "learning_rate": 0.0013532062132411659, + "loss": 0.1377, + "step": 39877 + }, + { + "epoch": 0.34616018958168765, + "grad_norm": 0.16796875, + "learning_rate": 0.001353177652808712, + "loss": 0.0757, + "step": 39878 + }, + { + "epoch": 0.34616887006189184, + "grad_norm": 0.349609375, + "learning_rate": 0.0013531490920993872, + "loss": 0.103, + "step": 39879 + }, + { + "epoch": 0.346177550542096, + "grad_norm": 0.16796875, + "learning_rate": 0.0013531205311132222, + "loss": 0.1172, + "step": 39880 + }, + { + "epoch": 0.3461862310223002, + "grad_norm": 0.96484375, + "learning_rate": 0.0013530919698502483, + "loss": 0.0981, + "step": 39881 + }, + { + "epoch": 0.3461949115025043, + "grad_norm": 0.11572265625, + "learning_rate": 0.0013530634083104972, + "loss": 0.0991, + "step": 39882 + }, + { + "epoch": 0.3462035919827085, + "grad_norm": 0.2265625, + "learning_rate": 0.0013530348464939992, + "loss": 0.0957, + "step": 39883 + }, + { + "epoch": 0.34621227246291264, + "grad_norm": 0.1904296875, + "learning_rate": 0.0013530062844007867, + "loss": 0.0869, + "step": 39884 + }, + { + "epoch": 0.34622095294311683, + "grad_norm": 0.263671875, + "learning_rate": 0.0013529777220308897, + "loss": 0.0918, + "step": 39885 + }, + { + "epoch": 0.34622963342332097, + "grad_norm": 0.2578125, + "learning_rate": 0.0013529491593843404, + "loss": 0.1221, + "step": 39886 + }, + { + "epoch": 0.34623831390352516, + "grad_norm": 0.09033203125, + "learning_rate": 0.0013529205964611696, + "loss": 0.0884, + "step": 39887 + }, + { + "epoch": 0.3462469943837293, + "grad_norm": 0.35546875, + "learning_rate": 0.0013528920332614087, + "loss": 0.1992, + "step": 39888 + }, + { + "epoch": 0.3462556748639335, + "grad_norm": 0.57421875, + "learning_rate": 0.001352863469785089, + "loss": 0.0659, + "step": 39889 + }, + { + "epoch": 0.34626435534413763, + "grad_norm": 0.21875, + "learning_rate": 0.0013528349060322414, + "loss": 0.1162, + "step": 39890 + }, + { + "epoch": 0.3462730358243418, + "grad_norm": 0.248046875, + "learning_rate": 0.0013528063420028974, + "loss": 0.0962, + "step": 39891 + }, + { + "epoch": 0.34628171630454596, + "grad_norm": 0.302734375, + "learning_rate": 0.0013527777776970883, + "loss": 0.0864, + "step": 39892 + }, + { + "epoch": 0.34629039678475015, + "grad_norm": 0.1845703125, + "learning_rate": 0.001352749213114845, + "loss": 0.1133, + "step": 39893 + }, + { + "epoch": 0.3462990772649543, + "grad_norm": 0.361328125, + "learning_rate": 0.0013527206482561993, + "loss": 0.1211, + "step": 39894 + }, + { + "epoch": 0.3463077577451585, + "grad_norm": 0.298828125, + "learning_rate": 0.0013526920831211818, + "loss": 0.166, + "step": 39895 + }, + { + "epoch": 0.3463164382253626, + "grad_norm": 0.234375, + "learning_rate": 0.001352663517709824, + "loss": 0.1084, + "step": 39896 + }, + { + "epoch": 0.3463251187055668, + "grad_norm": 0.2109375, + "learning_rate": 0.0013526349520221572, + "loss": 0.1025, + "step": 39897 + }, + { + "epoch": 0.34633379918577095, + "grad_norm": 0.2197265625, + "learning_rate": 0.0013526063860582126, + "loss": 0.0962, + "step": 39898 + }, + { + "epoch": 0.34634247966597514, + "grad_norm": 0.4375, + "learning_rate": 0.0013525778198180218, + "loss": 0.1553, + "step": 39899 + }, + { + "epoch": 0.3463511601461793, + "grad_norm": 0.328125, + "learning_rate": 0.0013525492533016153, + "loss": 0.0894, + "step": 39900 + }, + { + "epoch": 0.3463598406263835, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013525206865090248, + "loss": 0.0923, + "step": 39901 + }, + { + "epoch": 0.3463685211065876, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013524921194402817, + "loss": 0.0776, + "step": 39902 + }, + { + "epoch": 0.3463772015867918, + "grad_norm": 0.224609375, + "learning_rate": 0.001352463552095417, + "loss": 0.1094, + "step": 39903 + }, + { + "epoch": 0.34638588206699594, + "grad_norm": 0.23046875, + "learning_rate": 0.0013524349844744616, + "loss": 0.0898, + "step": 39904 + }, + { + "epoch": 0.34639456254720014, + "grad_norm": 0.458984375, + "learning_rate": 0.0013524064165774472, + "loss": 0.1855, + "step": 39905 + }, + { + "epoch": 0.3464032430274043, + "grad_norm": 0.37890625, + "learning_rate": 0.0013523778484044052, + "loss": 0.1099, + "step": 39906 + }, + { + "epoch": 0.34641192350760847, + "grad_norm": 0.142578125, + "learning_rate": 0.0013523492799553665, + "loss": 0.1055, + "step": 39907 + }, + { + "epoch": 0.3464206039878126, + "grad_norm": 0.4140625, + "learning_rate": 0.0013523207112303623, + "loss": 0.1104, + "step": 39908 + }, + { + "epoch": 0.3464292844680168, + "grad_norm": 0.173828125, + "learning_rate": 0.0013522921422294241, + "loss": 0.0801, + "step": 39909 + }, + { + "epoch": 0.34643796494822093, + "grad_norm": 0.1484375, + "learning_rate": 0.0013522635729525828, + "loss": 0.085, + "step": 39910 + }, + { + "epoch": 0.3464466454284251, + "grad_norm": 0.1796875, + "learning_rate": 0.00135223500339987, + "loss": 0.1118, + "step": 39911 + }, + { + "epoch": 0.34645532590862926, + "grad_norm": 0.60546875, + "learning_rate": 0.001352206433571317, + "loss": 0.1152, + "step": 39912 + }, + { + "epoch": 0.34646400638883346, + "grad_norm": 0.396484375, + "learning_rate": 0.0013521778634669544, + "loss": 0.1006, + "step": 39913 + }, + { + "epoch": 0.3464726868690376, + "grad_norm": 0.119140625, + "learning_rate": 0.001352149293086814, + "loss": 0.1011, + "step": 39914 + }, + { + "epoch": 0.3464813673492418, + "grad_norm": 0.7734375, + "learning_rate": 0.0013521207224309268, + "loss": 0.1377, + "step": 39915 + }, + { + "epoch": 0.3464900478294459, + "grad_norm": 0.369140625, + "learning_rate": 0.0013520921514993242, + "loss": 0.063, + "step": 39916 + }, + { + "epoch": 0.3464987283096501, + "grad_norm": 0.380859375, + "learning_rate": 0.0013520635802920376, + "loss": 0.0957, + "step": 39917 + }, + { + "epoch": 0.34650740878985425, + "grad_norm": 0.1181640625, + "learning_rate": 0.0013520350088090981, + "loss": 0.0723, + "step": 39918 + }, + { + "epoch": 0.34651608927005845, + "grad_norm": 0.099609375, + "learning_rate": 0.001352006437050537, + "loss": 0.126, + "step": 39919 + }, + { + "epoch": 0.3465247697502626, + "grad_norm": 0.380859375, + "learning_rate": 0.0013519778650163852, + "loss": 0.1289, + "step": 39920 + }, + { + "epoch": 0.3465334502304667, + "grad_norm": 0.421875, + "learning_rate": 0.0013519492927066743, + "loss": 0.0938, + "step": 39921 + }, + { + "epoch": 0.3465421307106709, + "grad_norm": 0.58984375, + "learning_rate": 0.0013519207201214355, + "loss": 0.0859, + "step": 39922 + }, + { + "epoch": 0.34655081119087505, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013518921472606997, + "loss": 0.1006, + "step": 39923 + }, + { + "epoch": 0.34655949167107925, + "grad_norm": 0.61328125, + "learning_rate": 0.0013518635741244984, + "loss": 0.1289, + "step": 39924 + }, + { + "epoch": 0.3465681721512834, + "grad_norm": 0.15625, + "learning_rate": 0.0013518350007128632, + "loss": 0.0747, + "step": 39925 + }, + { + "epoch": 0.3465768526314876, + "grad_norm": 0.58203125, + "learning_rate": 0.0013518064270258247, + "loss": 0.0923, + "step": 39926 + }, + { + "epoch": 0.3465855331116917, + "grad_norm": 0.87890625, + "learning_rate": 0.0013517778530634148, + "loss": 0.1699, + "step": 39927 + }, + { + "epoch": 0.3465942135918959, + "grad_norm": 0.431640625, + "learning_rate": 0.0013517492788256646, + "loss": 0.0928, + "step": 39928 + }, + { + "epoch": 0.34660289407210004, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013517207043126048, + "loss": 0.0981, + "step": 39929 + }, + { + "epoch": 0.34661157455230424, + "grad_norm": 0.349609375, + "learning_rate": 0.001351692129524267, + "loss": 0.1455, + "step": 39930 + }, + { + "epoch": 0.3466202550325084, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013516635544606826, + "loss": 0.0898, + "step": 39931 + }, + { + "epoch": 0.34662893551271257, + "grad_norm": 0.3046875, + "learning_rate": 0.0013516349791218827, + "loss": 0.0781, + "step": 39932 + }, + { + "epoch": 0.3466376159929167, + "grad_norm": 0.2177734375, + "learning_rate": 0.0013516064035078983, + "loss": 0.1133, + "step": 39933 + }, + { + "epoch": 0.3466462964731209, + "grad_norm": 0.26953125, + "learning_rate": 0.0013515778276187614, + "loss": 0.0977, + "step": 39934 + }, + { + "epoch": 0.34665497695332503, + "grad_norm": 0.228515625, + "learning_rate": 0.0013515492514545025, + "loss": 0.1138, + "step": 39935 + }, + { + "epoch": 0.3466636574335292, + "grad_norm": 0.265625, + "learning_rate": 0.0013515206750151529, + "loss": 0.1162, + "step": 39936 + }, + { + "epoch": 0.34667233791373336, + "grad_norm": 0.40625, + "learning_rate": 0.0013514920983007446, + "loss": 0.0967, + "step": 39937 + }, + { + "epoch": 0.34668101839393756, + "grad_norm": 0.302734375, + "learning_rate": 0.001351463521311308, + "loss": 0.0869, + "step": 39938 + }, + { + "epoch": 0.3466896988741417, + "grad_norm": 0.8125, + "learning_rate": 0.0013514349440468746, + "loss": 0.1147, + "step": 39939 + }, + { + "epoch": 0.3466983793543459, + "grad_norm": 1.1484375, + "learning_rate": 0.001351406366507476, + "loss": 0.1426, + "step": 39940 + }, + { + "epoch": 0.34670705983455, + "grad_norm": 0.35546875, + "learning_rate": 0.0013513777886931429, + "loss": 0.1064, + "step": 39941 + }, + { + "epoch": 0.3467157403147542, + "grad_norm": 0.384765625, + "learning_rate": 0.001351349210603907, + "loss": 0.1196, + "step": 39942 + }, + { + "epoch": 0.34672442079495835, + "grad_norm": 0.734375, + "learning_rate": 0.0013513206322397992, + "loss": 0.1289, + "step": 39943 + }, + { + "epoch": 0.34673310127516255, + "grad_norm": 0.373046875, + "learning_rate": 0.001351292053600851, + "loss": 0.1045, + "step": 39944 + }, + { + "epoch": 0.3467417817553667, + "grad_norm": 0.1728515625, + "learning_rate": 0.0013512634746870938, + "loss": 0.083, + "step": 39945 + }, + { + "epoch": 0.3467504622355709, + "grad_norm": 0.1015625, + "learning_rate": 0.0013512348954985585, + "loss": 0.0928, + "step": 39946 + }, + { + "epoch": 0.346759142715775, + "grad_norm": 0.119140625, + "learning_rate": 0.0013512063160352763, + "loss": 0.0986, + "step": 39947 + }, + { + "epoch": 0.3467678231959792, + "grad_norm": 0.490234375, + "learning_rate": 0.001351177736297279, + "loss": 0.1074, + "step": 39948 + }, + { + "epoch": 0.34677650367618335, + "grad_norm": 0.30078125, + "learning_rate": 0.0013511491562845973, + "loss": 0.1133, + "step": 39949 + }, + { + "epoch": 0.34678518415638754, + "grad_norm": 0.1953125, + "learning_rate": 0.0013511205759972626, + "loss": 0.1025, + "step": 39950 + }, + { + "epoch": 0.3467938646365917, + "grad_norm": 0.1357421875, + "learning_rate": 0.0013510919954353065, + "loss": 0.0913, + "step": 39951 + }, + { + "epoch": 0.34680254511679587, + "grad_norm": 0.2060546875, + "learning_rate": 0.0013510634145987596, + "loss": 0.1001, + "step": 39952 + }, + { + "epoch": 0.346811225597, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013510348334876536, + "loss": 0.085, + "step": 39953 + }, + { + "epoch": 0.3468199060772042, + "grad_norm": 0.412109375, + "learning_rate": 0.00135100625210202, + "loss": 0.1113, + "step": 39954 + }, + { + "epoch": 0.34682858655740834, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013509776704418892, + "loss": 0.0879, + "step": 39955 + }, + { + "epoch": 0.34683726703761253, + "grad_norm": 0.091796875, + "learning_rate": 0.0013509490885072932, + "loss": 0.1182, + "step": 39956 + }, + { + "epoch": 0.34684594751781667, + "grad_norm": 0.306640625, + "learning_rate": 0.0013509205062982633, + "loss": 0.0889, + "step": 39957 + }, + { + "epoch": 0.34685462799802086, + "grad_norm": 0.392578125, + "learning_rate": 0.0013508919238148305, + "loss": 0.0864, + "step": 39958 + }, + { + "epoch": 0.346863308478225, + "grad_norm": 0.9921875, + "learning_rate": 0.0013508633410570258, + "loss": 0.1079, + "step": 39959 + }, + { + "epoch": 0.3468719889584292, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013508347580248807, + "loss": 0.082, + "step": 39960 + }, + { + "epoch": 0.3468806694386333, + "grad_norm": 1.3359375, + "learning_rate": 0.0013508061747184268, + "loss": 0.1084, + "step": 39961 + }, + { + "epoch": 0.3468893499188375, + "grad_norm": 0.35546875, + "learning_rate": 0.0013507775911376946, + "loss": 0.1182, + "step": 39962 + }, + { + "epoch": 0.34689803039904166, + "grad_norm": 0.09619140625, + "learning_rate": 0.001350749007282716, + "loss": 0.1079, + "step": 39963 + }, + { + "epoch": 0.34690671087924585, + "grad_norm": 0.265625, + "learning_rate": 0.0013507204231535221, + "loss": 0.105, + "step": 39964 + }, + { + "epoch": 0.34691539135945, + "grad_norm": 0.14453125, + "learning_rate": 0.0013506918387501442, + "loss": 0.1045, + "step": 39965 + }, + { + "epoch": 0.3469240718396542, + "grad_norm": 0.58984375, + "learning_rate": 0.0013506632540726132, + "loss": 0.1113, + "step": 39966 + }, + { + "epoch": 0.3469327523198583, + "grad_norm": 0.265625, + "learning_rate": 0.001350634669120961, + "loss": 0.0718, + "step": 39967 + }, + { + "epoch": 0.3469414328000625, + "grad_norm": 0.53515625, + "learning_rate": 0.0013506060838952183, + "loss": 0.1592, + "step": 39968 + }, + { + "epoch": 0.34695011328026665, + "grad_norm": 0.51171875, + "learning_rate": 0.0013505774983954165, + "loss": 0.0918, + "step": 39969 + }, + { + "epoch": 0.34695879376047084, + "grad_norm": 0.4765625, + "learning_rate": 0.001350548912621587, + "loss": 0.1143, + "step": 39970 + }, + { + "epoch": 0.346967474240675, + "grad_norm": 0.07421875, + "learning_rate": 0.0013505203265737609, + "loss": 0.0869, + "step": 39971 + }, + { + "epoch": 0.34697615472087917, + "grad_norm": 0.212890625, + "learning_rate": 0.0013504917402519694, + "loss": 0.084, + "step": 39972 + }, + { + "epoch": 0.3469848352010833, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013504631536562439, + "loss": 0.0864, + "step": 39973 + }, + { + "epoch": 0.3469935156812875, + "grad_norm": 0.95703125, + "learning_rate": 0.001350434566786616, + "loss": 0.1006, + "step": 39974 + }, + { + "epoch": 0.34700219616149164, + "grad_norm": 0.58203125, + "learning_rate": 0.0013504059796431165, + "loss": 0.1006, + "step": 39975 + }, + { + "epoch": 0.34701087664169583, + "grad_norm": 0.216796875, + "learning_rate": 0.0013503773922257766, + "loss": 0.1543, + "step": 39976 + }, + { + "epoch": 0.34701955712189997, + "grad_norm": 0.43359375, + "learning_rate": 0.0013503488045346279, + "loss": 0.1895, + "step": 39977 + }, + { + "epoch": 0.34702823760210416, + "grad_norm": 0.203125, + "learning_rate": 0.0013503202165697015, + "loss": 0.0615, + "step": 39978 + }, + { + "epoch": 0.3470369180823083, + "grad_norm": 0.09619140625, + "learning_rate": 0.0013502916283310287, + "loss": 0.0977, + "step": 39979 + }, + { + "epoch": 0.3470455985625125, + "grad_norm": 0.55078125, + "learning_rate": 0.0013502630398186406, + "loss": 0.1768, + "step": 39980 + }, + { + "epoch": 0.34705427904271663, + "grad_norm": 0.107421875, + "learning_rate": 0.0013502344510325686, + "loss": 0.1006, + "step": 39981 + }, + { + "epoch": 0.3470629595229208, + "grad_norm": 0.37890625, + "learning_rate": 0.001350205861972844, + "loss": 0.1079, + "step": 39982 + }, + { + "epoch": 0.34707164000312496, + "grad_norm": 0.1328125, + "learning_rate": 0.0013501772726394981, + "loss": 0.1162, + "step": 39983 + }, + { + "epoch": 0.34708032048332915, + "grad_norm": 0.189453125, + "learning_rate": 0.001350148683032562, + "loss": 0.0645, + "step": 39984 + }, + { + "epoch": 0.3470890009635333, + "grad_norm": 0.212890625, + "learning_rate": 0.0013501200931520672, + "loss": 0.1377, + "step": 39985 + }, + { + "epoch": 0.3470976814437375, + "grad_norm": 0.4375, + "learning_rate": 0.0013500915029980446, + "loss": 0.0967, + "step": 39986 + }, + { + "epoch": 0.3471063619239416, + "grad_norm": 0.640625, + "learning_rate": 0.0013500629125705258, + "loss": 0.1016, + "step": 39987 + }, + { + "epoch": 0.3471150424041458, + "grad_norm": 0.5625, + "learning_rate": 0.001350034321869542, + "loss": 0.1143, + "step": 39988 + }, + { + "epoch": 0.34712372288434995, + "grad_norm": 0.0859375, + "learning_rate": 0.0013500057308951248, + "loss": 0.0608, + "step": 39989 + }, + { + "epoch": 0.34713240336455414, + "grad_norm": 0.2119140625, + "learning_rate": 0.0013499771396473044, + "loss": 0.0889, + "step": 39990 + }, + { + "epoch": 0.3471410838447583, + "grad_norm": 0.10986328125, + "learning_rate": 0.0013499485481261134, + "loss": 0.1069, + "step": 39991 + }, + { + "epoch": 0.3471497643249625, + "grad_norm": 0.6796875, + "learning_rate": 0.0013499199563315818, + "loss": 0.0938, + "step": 39992 + }, + { + "epoch": 0.3471584448051666, + "grad_norm": 0.314453125, + "learning_rate": 0.0013498913642637418, + "loss": 0.126, + "step": 39993 + }, + { + "epoch": 0.3471671252853708, + "grad_norm": 0.408203125, + "learning_rate": 0.0013498627719226244, + "loss": 0.1016, + "step": 39994 + }, + { + "epoch": 0.34717580576557494, + "grad_norm": 0.08544921875, + "learning_rate": 0.0013498341793082608, + "loss": 0.1206, + "step": 39995 + }, + { + "epoch": 0.34718448624577913, + "grad_norm": 0.48828125, + "learning_rate": 0.001349805586420682, + "loss": 0.1064, + "step": 39996 + }, + { + "epoch": 0.34719316672598327, + "grad_norm": 0.107421875, + "learning_rate": 0.00134977699325992, + "loss": 0.0762, + "step": 39997 + }, + { + "epoch": 0.34720184720618746, + "grad_norm": 0.29296875, + "learning_rate": 0.0013497483998260054, + "loss": 0.0957, + "step": 39998 + }, + { + "epoch": 0.3472105276863916, + "grad_norm": 0.27734375, + "learning_rate": 0.0013497198061189696, + "loss": 0.1299, + "step": 39999 + }, + { + "epoch": 0.3472192081665958, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013496912121388446, + "loss": 0.0752, + "step": 40000 + }, + { + "epoch": 0.34722788864679993, + "grad_norm": 0.234375, + "learning_rate": 0.0013496626178856601, + "loss": 0.1001, + "step": 40001 + }, + { + "epoch": 0.3472365691270041, + "grad_norm": 0.50390625, + "learning_rate": 0.0013496340233594489, + "loss": 0.1064, + "step": 40002 + }, + { + "epoch": 0.34724524960720826, + "grad_norm": 0.44140625, + "learning_rate": 0.0013496054285602418, + "loss": 0.2148, + "step": 40003 + }, + { + "epoch": 0.34725393008741245, + "grad_norm": 0.201171875, + "learning_rate": 0.0013495768334880693, + "loss": 0.0737, + "step": 40004 + }, + { + "epoch": 0.3472626105676166, + "grad_norm": 0.1943359375, + "learning_rate": 0.001349548238142964, + "loss": 0.0986, + "step": 40005 + }, + { + "epoch": 0.3472712910478208, + "grad_norm": 0.32421875, + "learning_rate": 0.0013495196425249564, + "loss": 0.0801, + "step": 40006 + }, + { + "epoch": 0.3472799715280249, + "grad_norm": 0.5859375, + "learning_rate": 0.0013494910466340774, + "loss": 0.0923, + "step": 40007 + }, + { + "epoch": 0.3472886520082291, + "grad_norm": 0.1689453125, + "learning_rate": 0.001349462450470359, + "loss": 0.084, + "step": 40008 + }, + { + "epoch": 0.34729733248843325, + "grad_norm": 0.12451171875, + "learning_rate": 0.0013494338540338323, + "loss": 0.1123, + "step": 40009 + }, + { + "epoch": 0.34730601296863745, + "grad_norm": 0.14453125, + "learning_rate": 0.0013494052573245284, + "loss": 0.1318, + "step": 40010 + }, + { + "epoch": 0.3473146934488416, + "grad_norm": 0.142578125, + "learning_rate": 0.0013493766603424786, + "loss": 0.1157, + "step": 40011 + }, + { + "epoch": 0.3473233739290458, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013493480630877143, + "loss": 0.0908, + "step": 40012 + }, + { + "epoch": 0.3473320544092499, + "grad_norm": 0.341796875, + "learning_rate": 0.0013493194655602669, + "loss": 0.0825, + "step": 40013 + }, + { + "epoch": 0.3473407348894541, + "grad_norm": 0.232421875, + "learning_rate": 0.0013492908677601675, + "loss": 0.0947, + "step": 40014 + }, + { + "epoch": 0.34734941536965824, + "grad_norm": 0.6328125, + "learning_rate": 0.001349262269687447, + "loss": 0.0947, + "step": 40015 + }, + { + "epoch": 0.34735809584986244, + "grad_norm": 0.451171875, + "learning_rate": 0.001349233671342137, + "loss": 0.0879, + "step": 40016 + }, + { + "epoch": 0.3473667763300666, + "grad_norm": 0.419921875, + "learning_rate": 0.0013492050727242691, + "loss": 0.1143, + "step": 40017 + }, + { + "epoch": 0.34737545681027077, + "grad_norm": 0.146484375, + "learning_rate": 0.0013491764738338744, + "loss": 0.1196, + "step": 40018 + }, + { + "epoch": 0.3473841372904749, + "grad_norm": 0.376953125, + "learning_rate": 0.0013491478746709838, + "loss": 0.104, + "step": 40019 + }, + { + "epoch": 0.3473928177706791, + "grad_norm": 0.83203125, + "learning_rate": 0.0013491192752356287, + "loss": 0.1387, + "step": 40020 + }, + { + "epoch": 0.34740149825088323, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013490906755278405, + "loss": 0.1221, + "step": 40021 + }, + { + "epoch": 0.3474101787310874, + "grad_norm": 0.19921875, + "learning_rate": 0.0013490620755476509, + "loss": 0.1016, + "step": 40022 + }, + { + "epoch": 0.34741885921129156, + "grad_norm": 0.2216796875, + "learning_rate": 0.0013490334752950903, + "loss": 0.0879, + "step": 40023 + }, + { + "epoch": 0.34742753969149576, + "grad_norm": 0.2255859375, + "learning_rate": 0.0013490048747701905, + "loss": 0.0913, + "step": 40024 + }, + { + "epoch": 0.3474362201716999, + "grad_norm": 0.1689453125, + "learning_rate": 0.001348976273972983, + "loss": 0.1133, + "step": 40025 + }, + { + "epoch": 0.3474449006519041, + "grad_norm": 0.3125, + "learning_rate": 0.0013489476729034986, + "loss": 0.0688, + "step": 40026 + }, + { + "epoch": 0.3474535811321082, + "grad_norm": 0.11962890625, + "learning_rate": 0.001348919071561769, + "loss": 0.1016, + "step": 40027 + }, + { + "epoch": 0.3474622616123124, + "grad_norm": 0.0966796875, + "learning_rate": 0.0013488904699478247, + "loss": 0.1177, + "step": 40028 + }, + { + "epoch": 0.34747094209251655, + "grad_norm": 0.083984375, + "learning_rate": 0.001348861868061698, + "loss": 0.083, + "step": 40029 + }, + { + "epoch": 0.34747962257272075, + "grad_norm": 0.16015625, + "learning_rate": 0.0013488332659034193, + "loss": 0.1094, + "step": 40030 + }, + { + "epoch": 0.3474883030529249, + "grad_norm": 0.1474609375, + "learning_rate": 0.0013488046634730208, + "loss": 0.0967, + "step": 40031 + }, + { + "epoch": 0.3474969835331291, + "grad_norm": 0.1640625, + "learning_rate": 0.0013487760607705328, + "loss": 0.1128, + "step": 40032 + }, + { + "epoch": 0.3475056640133332, + "grad_norm": 0.0712890625, + "learning_rate": 0.0013487474577959875, + "loss": 0.0938, + "step": 40033 + }, + { + "epoch": 0.3475143444935374, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013487188545494152, + "loss": 0.1006, + "step": 40034 + }, + { + "epoch": 0.34752302497374155, + "grad_norm": 0.3125, + "learning_rate": 0.001348690251030848, + "loss": 0.0645, + "step": 40035 + }, + { + "epoch": 0.34753170545394574, + "grad_norm": 0.10791015625, + "learning_rate": 0.0013486616472403168, + "loss": 0.1152, + "step": 40036 + }, + { + "epoch": 0.3475403859341499, + "grad_norm": 0.12109375, + "learning_rate": 0.0013486330431778531, + "loss": 0.0918, + "step": 40037 + }, + { + "epoch": 0.34754906641435407, + "grad_norm": 0.5234375, + "learning_rate": 0.001348604438843488, + "loss": 0.0854, + "step": 40038 + }, + { + "epoch": 0.3475577468945582, + "grad_norm": 0.126953125, + "learning_rate": 0.0013485758342372527, + "loss": 0.0713, + "step": 40039 + }, + { + "epoch": 0.3475664273747624, + "grad_norm": 0.5703125, + "learning_rate": 0.0013485472293591784, + "loss": 0.1328, + "step": 40040 + }, + { + "epoch": 0.34757510785496654, + "grad_norm": 0.318359375, + "learning_rate": 0.0013485186242092968, + "loss": 0.0898, + "step": 40041 + }, + { + "epoch": 0.34758378833517073, + "grad_norm": 0.21875, + "learning_rate": 0.0013484900187876391, + "loss": 0.0781, + "step": 40042 + }, + { + "epoch": 0.34759246881537487, + "grad_norm": 0.390625, + "learning_rate": 0.0013484614130942364, + "loss": 0.1143, + "step": 40043 + }, + { + "epoch": 0.347601149295579, + "grad_norm": 0.275390625, + "learning_rate": 0.0013484328071291202, + "loss": 0.0815, + "step": 40044 + }, + { + "epoch": 0.3476098297757832, + "grad_norm": 0.83203125, + "learning_rate": 0.001348404200892321, + "loss": 0.1055, + "step": 40045 + }, + { + "epoch": 0.34761851025598733, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013483755943838713, + "loss": 0.0938, + "step": 40046 + }, + { + "epoch": 0.3476271907361915, + "grad_norm": 0.22265625, + "learning_rate": 0.0013483469876038015, + "loss": 0.0859, + "step": 40047 + }, + { + "epoch": 0.34763587121639566, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013483183805521432, + "loss": 0.1309, + "step": 40048 + }, + { + "epoch": 0.34764455169659986, + "grad_norm": 0.494140625, + "learning_rate": 0.0013482897732289276, + "loss": 0.1196, + "step": 40049 + }, + { + "epoch": 0.347653232176804, + "grad_norm": 0.1708984375, + "learning_rate": 0.0013482611656341864, + "loss": 0.085, + "step": 40050 + }, + { + "epoch": 0.3476619126570082, + "grad_norm": 0.185546875, + "learning_rate": 0.0013482325577679499, + "loss": 0.0781, + "step": 40051 + }, + { + "epoch": 0.3476705931372123, + "grad_norm": 0.1826171875, + "learning_rate": 0.0013482039496302504, + "loss": 0.0757, + "step": 40052 + }, + { + "epoch": 0.3476792736174165, + "grad_norm": 0.375, + "learning_rate": 0.001348175341221119, + "loss": 0.1348, + "step": 40053 + }, + { + "epoch": 0.34768795409762066, + "grad_norm": 0.259765625, + "learning_rate": 0.0013481467325405863, + "loss": 0.1084, + "step": 40054 + }, + { + "epoch": 0.34769663457782485, + "grad_norm": 0.375, + "learning_rate": 0.0013481181235886844, + "loss": 0.126, + "step": 40055 + }, + { + "epoch": 0.347705315058029, + "grad_norm": 0.138671875, + "learning_rate": 0.0013480895143654441, + "loss": 0.1846, + "step": 40056 + }, + { + "epoch": 0.3477139955382332, + "grad_norm": 0.076171875, + "learning_rate": 0.0013480609048708966, + "loss": 0.0996, + "step": 40057 + }, + { + "epoch": 0.3477226760184373, + "grad_norm": 0.1708984375, + "learning_rate": 0.001348032295105074, + "loss": 0.1025, + "step": 40058 + }, + { + "epoch": 0.3477313564986415, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013480036850680066, + "loss": 0.1426, + "step": 40059 + }, + { + "epoch": 0.34774003697884565, + "grad_norm": 0.546875, + "learning_rate": 0.001347975074759726, + "loss": 0.0879, + "step": 40060 + }, + { + "epoch": 0.34774871745904984, + "grad_norm": 0.09130859375, + "learning_rate": 0.001347946464180264, + "loss": 0.1016, + "step": 40061 + }, + { + "epoch": 0.347757397939254, + "grad_norm": 0.134765625, + "learning_rate": 0.001347917853329651, + "loss": 0.0908, + "step": 40062 + }, + { + "epoch": 0.34776607841945817, + "grad_norm": 1.28125, + "learning_rate": 0.0013478892422079192, + "loss": 0.1309, + "step": 40063 + }, + { + "epoch": 0.3477747588996623, + "grad_norm": 0.330078125, + "learning_rate": 0.0013478606308150991, + "loss": 0.0898, + "step": 40064 + }, + { + "epoch": 0.3477834393798665, + "grad_norm": 0.271484375, + "learning_rate": 0.0013478320191512225, + "loss": 0.1113, + "step": 40065 + }, + { + "epoch": 0.34779211986007064, + "grad_norm": 0.1904296875, + "learning_rate": 0.0013478034072163206, + "loss": 0.0977, + "step": 40066 + }, + { + "epoch": 0.34780080034027483, + "grad_norm": 0.462890625, + "learning_rate": 0.0013477747950104244, + "loss": 0.0928, + "step": 40067 + }, + { + "epoch": 0.34780948082047897, + "grad_norm": 0.33984375, + "learning_rate": 0.0013477461825335654, + "loss": 0.0869, + "step": 40068 + }, + { + "epoch": 0.34781816130068316, + "grad_norm": 0.318359375, + "learning_rate": 0.001347717569785775, + "loss": 0.1104, + "step": 40069 + }, + { + "epoch": 0.3478268417808873, + "grad_norm": 0.380859375, + "learning_rate": 0.0013476889567670845, + "loss": 0.166, + "step": 40070 + }, + { + "epoch": 0.3478355222610915, + "grad_norm": 0.12890625, + "learning_rate": 0.001347660343477525, + "loss": 0.0972, + "step": 40071 + }, + { + "epoch": 0.3478442027412956, + "grad_norm": 0.67578125, + "learning_rate": 0.001347631729917128, + "loss": 0.1572, + "step": 40072 + }, + { + "epoch": 0.3478528832214998, + "grad_norm": 0.462890625, + "learning_rate": 0.0013476031160859245, + "loss": 0.0605, + "step": 40073 + }, + { + "epoch": 0.34786156370170396, + "grad_norm": 0.4375, + "learning_rate": 0.001347574501983946, + "loss": 0.123, + "step": 40074 + }, + { + "epoch": 0.34787024418190815, + "grad_norm": 0.34375, + "learning_rate": 0.0013475458876112236, + "loss": 0.0938, + "step": 40075 + }, + { + "epoch": 0.3478789246621123, + "grad_norm": 0.1328125, + "learning_rate": 0.001347517272967789, + "loss": 0.084, + "step": 40076 + }, + { + "epoch": 0.3478876051423165, + "grad_norm": 0.09033203125, + "learning_rate": 0.001347488658053673, + "loss": 0.0869, + "step": 40077 + }, + { + "epoch": 0.3478962856225206, + "grad_norm": 0.703125, + "learning_rate": 0.001347460042868907, + "loss": 0.1641, + "step": 40078 + }, + { + "epoch": 0.3479049661027248, + "grad_norm": 0.099609375, + "learning_rate": 0.0013474314274135228, + "loss": 0.0688, + "step": 40079 + }, + { + "epoch": 0.34791364658292895, + "grad_norm": 0.1494140625, + "learning_rate": 0.001347402811687551, + "loss": 0.1221, + "step": 40080 + }, + { + "epoch": 0.34792232706313314, + "grad_norm": 0.10986328125, + "learning_rate": 0.0013473741956910232, + "loss": 0.1074, + "step": 40081 + }, + { + "epoch": 0.3479310075433373, + "grad_norm": 0.26171875, + "learning_rate": 0.001347345579423971, + "loss": 0.1582, + "step": 40082 + }, + { + "epoch": 0.34793968802354147, + "grad_norm": 0.173828125, + "learning_rate": 0.0013473169628864252, + "loss": 0.123, + "step": 40083 + }, + { + "epoch": 0.3479483685037456, + "grad_norm": 0.6796875, + "learning_rate": 0.0013472883460784177, + "loss": 0.1289, + "step": 40084 + }, + { + "epoch": 0.3479570489839498, + "grad_norm": 0.212890625, + "learning_rate": 0.0013472597289999788, + "loss": 0.0928, + "step": 40085 + }, + { + "epoch": 0.34796572946415394, + "grad_norm": 0.365234375, + "learning_rate": 0.0013472311116511403, + "loss": 0.0815, + "step": 40086 + }, + { + "epoch": 0.34797440994435813, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013472024940319339, + "loss": 0.0889, + "step": 40087 + }, + { + "epoch": 0.34798309042456227, + "grad_norm": 0.392578125, + "learning_rate": 0.0013471738761423905, + "loss": 0.0938, + "step": 40088 + }, + { + "epoch": 0.34799177090476646, + "grad_norm": 0.11083984375, + "learning_rate": 0.001347145257982542, + "loss": 0.0972, + "step": 40089 + }, + { + "epoch": 0.3480004513849706, + "grad_norm": 0.314453125, + "learning_rate": 0.0013471166395524182, + "loss": 0.0928, + "step": 40090 + }, + { + "epoch": 0.3480091318651748, + "grad_norm": 0.146484375, + "learning_rate": 0.0013470880208520518, + "loss": 0.1797, + "step": 40091 + }, + { + "epoch": 0.34801781234537893, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013470594018814737, + "loss": 0.1182, + "step": 40092 + }, + { + "epoch": 0.3480264928255831, + "grad_norm": 0.89453125, + "learning_rate": 0.0013470307826407152, + "loss": 0.0874, + "step": 40093 + }, + { + "epoch": 0.34803517330578726, + "grad_norm": 0.208984375, + "learning_rate": 0.0013470021631298071, + "loss": 0.1406, + "step": 40094 + }, + { + "epoch": 0.34804385378599145, + "grad_norm": 0.1796875, + "learning_rate": 0.0013469735433487818, + "loss": 0.124, + "step": 40095 + }, + { + "epoch": 0.3480525342661956, + "grad_norm": 0.3515625, + "learning_rate": 0.0013469449232976695, + "loss": 0.0908, + "step": 40096 + }, + { + "epoch": 0.3480612147463998, + "grad_norm": 0.30859375, + "learning_rate": 0.0013469163029765018, + "loss": 0.0986, + "step": 40097 + }, + { + "epoch": 0.3480698952266039, + "grad_norm": 0.30078125, + "learning_rate": 0.0013468876823853108, + "loss": 0.1182, + "step": 40098 + }, + { + "epoch": 0.3480785757068081, + "grad_norm": 0.80859375, + "learning_rate": 0.0013468590615241265, + "loss": 0.1177, + "step": 40099 + }, + { + "epoch": 0.34808725618701225, + "grad_norm": 0.6328125, + "learning_rate": 0.0013468304403929812, + "loss": 0.1875, + "step": 40100 + }, + { + "epoch": 0.34809593666721644, + "grad_norm": 0.404296875, + "learning_rate": 0.0013468018189919057, + "loss": 0.1133, + "step": 40101 + }, + { + "epoch": 0.3481046171474206, + "grad_norm": 0.361328125, + "learning_rate": 0.0013467731973209316, + "loss": 0.0972, + "step": 40102 + }, + { + "epoch": 0.3481132976276248, + "grad_norm": 0.103515625, + "learning_rate": 0.0013467445753800897, + "loss": 0.1523, + "step": 40103 + }, + { + "epoch": 0.3481219781078289, + "grad_norm": 0.51953125, + "learning_rate": 0.001346715953169412, + "loss": 0.1123, + "step": 40104 + }, + { + "epoch": 0.3481306585880331, + "grad_norm": 0.482421875, + "learning_rate": 0.0013466873306889294, + "loss": 0.0947, + "step": 40105 + }, + { + "epoch": 0.34813933906823724, + "grad_norm": 0.376953125, + "learning_rate": 0.0013466587079386729, + "loss": 0.0991, + "step": 40106 + }, + { + "epoch": 0.34814801954844143, + "grad_norm": 0.4375, + "learning_rate": 0.0013466300849186744, + "loss": 0.0986, + "step": 40107 + }, + { + "epoch": 0.34815670002864557, + "grad_norm": 0.494140625, + "learning_rate": 0.0013466014616289649, + "loss": 0.0752, + "step": 40108 + }, + { + "epoch": 0.34816538050884976, + "grad_norm": 0.41796875, + "learning_rate": 0.0013465728380695756, + "loss": 0.0942, + "step": 40109 + }, + { + "epoch": 0.3481740609890539, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013465442142405382, + "loss": 0.1025, + "step": 40110 + }, + { + "epoch": 0.3481827414692581, + "grad_norm": 0.37890625, + "learning_rate": 0.0013465155901418837, + "loss": 0.0894, + "step": 40111 + }, + { + "epoch": 0.34819142194946223, + "grad_norm": 0.11328125, + "learning_rate": 0.0013464869657736434, + "loss": 0.1318, + "step": 40112 + }, + { + "epoch": 0.3482001024296664, + "grad_norm": 0.58203125, + "learning_rate": 0.0013464583411358485, + "loss": 0.1641, + "step": 40113 + }, + { + "epoch": 0.34820878290987056, + "grad_norm": 0.341796875, + "learning_rate": 0.001346429716228531, + "loss": 0.0947, + "step": 40114 + }, + { + "epoch": 0.34821746339007476, + "grad_norm": 0.18359375, + "learning_rate": 0.001346401091051721, + "loss": 0.085, + "step": 40115 + }, + { + "epoch": 0.3482261438702789, + "grad_norm": 0.328125, + "learning_rate": 0.001346372465605451, + "loss": 0.0913, + "step": 40116 + }, + { + "epoch": 0.3482348243504831, + "grad_norm": 0.1630859375, + "learning_rate": 0.0013463438398897513, + "loss": 0.1123, + "step": 40117 + }, + { + "epoch": 0.3482435048306872, + "grad_norm": 1.203125, + "learning_rate": 0.001346315213904654, + "loss": 0.3711, + "step": 40118 + }, + { + "epoch": 0.3482521853108914, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013462865876501902, + "loss": 0.104, + "step": 40119 + }, + { + "epoch": 0.34826086579109555, + "grad_norm": 0.609375, + "learning_rate": 0.0013462579611263907, + "loss": 0.0977, + "step": 40120 + }, + { + "epoch": 0.34826954627129975, + "grad_norm": 0.11376953125, + "learning_rate": 0.0013462293343332874, + "loss": 0.0723, + "step": 40121 + }, + { + "epoch": 0.3482782267515039, + "grad_norm": 0.546875, + "learning_rate": 0.0013462007072709114, + "loss": 0.0811, + "step": 40122 + }, + { + "epoch": 0.3482869072317081, + "grad_norm": 0.205078125, + "learning_rate": 0.001346172079939294, + "loss": 0.083, + "step": 40123 + }, + { + "epoch": 0.3482955877119122, + "grad_norm": 0.142578125, + "learning_rate": 0.0013461434523384665, + "loss": 0.0986, + "step": 40124 + }, + { + "epoch": 0.3483042681921164, + "grad_norm": 0.3984375, + "learning_rate": 0.0013461148244684603, + "loss": 0.1113, + "step": 40125 + }, + { + "epoch": 0.34831294867232054, + "grad_norm": 0.255859375, + "learning_rate": 0.0013460861963293064, + "loss": 0.1211, + "step": 40126 + }, + { + "epoch": 0.34832162915252474, + "grad_norm": 0.515625, + "learning_rate": 0.0013460575679210365, + "loss": 0.1484, + "step": 40127 + }, + { + "epoch": 0.3483303096327289, + "grad_norm": 0.400390625, + "learning_rate": 0.0013460289392436817, + "loss": 0.1104, + "step": 40128 + }, + { + "epoch": 0.34833899011293307, + "grad_norm": 0.396484375, + "learning_rate": 0.001346000310297273, + "loss": 0.1084, + "step": 40129 + }, + { + "epoch": 0.3483476705931372, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013459716810818427, + "loss": 0.1719, + "step": 40130 + }, + { + "epoch": 0.3483563510733414, + "grad_norm": 0.2421875, + "learning_rate": 0.0013459430515974213, + "loss": 0.0635, + "step": 40131 + }, + { + "epoch": 0.34836503155354553, + "grad_norm": 0.0712890625, + "learning_rate": 0.0013459144218440403, + "loss": 0.0732, + "step": 40132 + }, + { + "epoch": 0.3483737120337497, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013458857918217306, + "loss": 0.0913, + "step": 40133 + }, + { + "epoch": 0.34838239251395386, + "grad_norm": 0.09521484375, + "learning_rate": 0.0013458571615305242, + "loss": 0.0645, + "step": 40134 + }, + { + "epoch": 0.34839107299415806, + "grad_norm": 0.11328125, + "learning_rate": 0.001345828530970452, + "loss": 0.1079, + "step": 40135 + }, + { + "epoch": 0.3483997534743622, + "grad_norm": 1.8828125, + "learning_rate": 0.0013457999001415455, + "loss": 0.1484, + "step": 40136 + }, + { + "epoch": 0.3484084339545664, + "grad_norm": 0.3515625, + "learning_rate": 0.0013457712690438358, + "loss": 0.0986, + "step": 40137 + }, + { + "epoch": 0.3484171144347705, + "grad_norm": 0.65234375, + "learning_rate": 0.0013457426376773543, + "loss": 0.1006, + "step": 40138 + }, + { + "epoch": 0.3484257949149747, + "grad_norm": 0.1044921875, + "learning_rate": 0.0013457140060421327, + "loss": 0.1079, + "step": 40139 + }, + { + "epoch": 0.34843447539517886, + "grad_norm": 0.27734375, + "learning_rate": 0.0013456853741382016, + "loss": 0.105, + "step": 40140 + }, + { + "epoch": 0.34844315587538305, + "grad_norm": 0.466796875, + "learning_rate": 0.0013456567419655927, + "loss": 0.1172, + "step": 40141 + }, + { + "epoch": 0.3484518363555872, + "grad_norm": 0.546875, + "learning_rate": 0.0013456281095243375, + "loss": 0.1055, + "step": 40142 + }, + { + "epoch": 0.3484605168357914, + "grad_norm": 0.3671875, + "learning_rate": 0.001345599476814467, + "loss": 0.0938, + "step": 40143 + }, + { + "epoch": 0.3484691973159955, + "grad_norm": 0.08447265625, + "learning_rate": 0.0013455708438360125, + "loss": 0.0684, + "step": 40144 + }, + { + "epoch": 0.3484778777961997, + "grad_norm": 0.69921875, + "learning_rate": 0.0013455422105890054, + "loss": 0.0811, + "step": 40145 + }, + { + "epoch": 0.34848655827640385, + "grad_norm": 0.326171875, + "learning_rate": 0.001345513577073477, + "loss": 0.1143, + "step": 40146 + }, + { + "epoch": 0.34849523875660804, + "grad_norm": 0.10400390625, + "learning_rate": 0.0013454849432894588, + "loss": 0.1016, + "step": 40147 + }, + { + "epoch": 0.3485039192368122, + "grad_norm": 0.166015625, + "learning_rate": 0.0013454563092369818, + "loss": 0.0957, + "step": 40148 + }, + { + "epoch": 0.34851259971701637, + "grad_norm": 0.2890625, + "learning_rate": 0.0013454276749160776, + "loss": 0.1426, + "step": 40149 + }, + { + "epoch": 0.3485212801972205, + "grad_norm": 0.55859375, + "learning_rate": 0.0013453990403267775, + "loss": 0.0791, + "step": 40150 + }, + { + "epoch": 0.3485299606774247, + "grad_norm": 1.0078125, + "learning_rate": 0.0013453704054691123, + "loss": 0.1396, + "step": 40151 + }, + { + "epoch": 0.34853864115762884, + "grad_norm": 1.21875, + "learning_rate": 0.001345341770343114, + "loss": 0.1299, + "step": 40152 + }, + { + "epoch": 0.34854732163783303, + "grad_norm": 0.1142578125, + "learning_rate": 0.0013453131349488137, + "loss": 0.0903, + "step": 40153 + }, + { + "epoch": 0.34855600211803717, + "grad_norm": 0.25, + "learning_rate": 0.0013452844992862423, + "loss": 0.1182, + "step": 40154 + }, + { + "epoch": 0.34856468259824136, + "grad_norm": 0.91015625, + "learning_rate": 0.0013452558633554315, + "loss": 0.1094, + "step": 40155 + }, + { + "epoch": 0.3485733630784455, + "grad_norm": 0.3203125, + "learning_rate": 0.001345227227156413, + "loss": 0.0674, + "step": 40156 + }, + { + "epoch": 0.3485820435586497, + "grad_norm": 0.228515625, + "learning_rate": 0.0013451985906892175, + "loss": 0.0889, + "step": 40157 + }, + { + "epoch": 0.3485907240388538, + "grad_norm": 0.515625, + "learning_rate": 0.0013451699539538764, + "loss": 0.085, + "step": 40158 + }, + { + "epoch": 0.348599404519058, + "grad_norm": 0.1259765625, + "learning_rate": 0.001345141316950421, + "loss": 0.1152, + "step": 40159 + }, + { + "epoch": 0.34860808499926216, + "grad_norm": 0.71875, + "learning_rate": 0.001345112679678883, + "loss": 0.085, + "step": 40160 + }, + { + "epoch": 0.34861676547946635, + "grad_norm": 0.208984375, + "learning_rate": 0.0013450840421392934, + "loss": 0.0923, + "step": 40161 + }, + { + "epoch": 0.3486254459596705, + "grad_norm": 0.7734375, + "learning_rate": 0.0013450554043316833, + "loss": 0.1025, + "step": 40162 + }, + { + "epoch": 0.3486341264398747, + "grad_norm": 0.7578125, + "learning_rate": 0.0013450267662560847, + "loss": 0.1216, + "step": 40163 + }, + { + "epoch": 0.3486428069200788, + "grad_norm": 0.470703125, + "learning_rate": 0.0013449981279125283, + "loss": 0.0898, + "step": 40164 + }, + { + "epoch": 0.348651487400283, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013449694893010453, + "loss": 0.1133, + "step": 40165 + }, + { + "epoch": 0.34866016788048715, + "grad_norm": 0.353515625, + "learning_rate": 0.0013449408504216677, + "loss": 0.0801, + "step": 40166 + }, + { + "epoch": 0.3486688483606913, + "grad_norm": 0.17578125, + "learning_rate": 0.0013449122112744267, + "loss": 0.1167, + "step": 40167 + }, + { + "epoch": 0.3486775288408955, + "grad_norm": 0.17578125, + "learning_rate": 0.0013448835718593531, + "loss": 0.0981, + "step": 40168 + }, + { + "epoch": 0.3486862093210996, + "grad_norm": 0.5625, + "learning_rate": 0.0013448549321764787, + "loss": 0.1118, + "step": 40169 + }, + { + "epoch": 0.3486948898013038, + "grad_norm": 0.470703125, + "learning_rate": 0.0013448262922258344, + "loss": 0.1035, + "step": 40170 + }, + { + "epoch": 0.34870357028150795, + "grad_norm": 0.400390625, + "learning_rate": 0.001344797652007452, + "loss": 0.1084, + "step": 40171 + }, + { + "epoch": 0.34871225076171214, + "grad_norm": 0.98046875, + "learning_rate": 0.0013447690115213622, + "loss": 0.0947, + "step": 40172 + }, + { + "epoch": 0.3487209312419163, + "grad_norm": 0.09033203125, + "learning_rate": 0.0013447403707675968, + "loss": 0.083, + "step": 40173 + }, + { + "epoch": 0.34872961172212047, + "grad_norm": 0.5390625, + "learning_rate": 0.001344711729746187, + "loss": 0.1348, + "step": 40174 + }, + { + "epoch": 0.3487382922023246, + "grad_norm": 0.126953125, + "learning_rate": 0.0013446830884571643, + "loss": 0.0767, + "step": 40175 + }, + { + "epoch": 0.3487469726825288, + "grad_norm": 0.1884765625, + "learning_rate": 0.00134465444690056, + "loss": 0.0801, + "step": 40176 + }, + { + "epoch": 0.34875565316273294, + "grad_norm": 0.86328125, + "learning_rate": 0.001344625805076405, + "loss": 0.1055, + "step": 40177 + }, + { + "epoch": 0.34876433364293713, + "grad_norm": 0.369140625, + "learning_rate": 0.0013445971629847309, + "loss": 0.1162, + "step": 40178 + }, + { + "epoch": 0.34877301412314127, + "grad_norm": 0.1337890625, + "learning_rate": 0.001344568520625569, + "loss": 0.1426, + "step": 40179 + }, + { + "epoch": 0.34878169460334546, + "grad_norm": 0.6796875, + "learning_rate": 0.0013445398779989507, + "loss": 0.0859, + "step": 40180 + }, + { + "epoch": 0.3487903750835496, + "grad_norm": 0.56640625, + "learning_rate": 0.0013445112351049072, + "loss": 0.0811, + "step": 40181 + }, + { + "epoch": 0.3487990555637538, + "grad_norm": 0.1591796875, + "learning_rate": 0.00134448259194347, + "loss": 0.1172, + "step": 40182 + }, + { + "epoch": 0.3488077360439579, + "grad_norm": 0.1337890625, + "learning_rate": 0.00134445394851467, + "loss": 0.1006, + "step": 40183 + }, + { + "epoch": 0.3488164165241621, + "grad_norm": 0.27734375, + "learning_rate": 0.0013444253048185393, + "loss": 0.0767, + "step": 40184 + }, + { + "epoch": 0.34882509700436626, + "grad_norm": 0.365234375, + "learning_rate": 0.0013443966608551084, + "loss": 0.1318, + "step": 40185 + }, + { + "epoch": 0.34883377748457045, + "grad_norm": 0.08837890625, + "learning_rate": 0.0013443680166244092, + "loss": 0.0747, + "step": 40186 + }, + { + "epoch": 0.3488424579647746, + "grad_norm": 0.6171875, + "learning_rate": 0.001344339372126473, + "loss": 0.123, + "step": 40187 + }, + { + "epoch": 0.3488511384449788, + "grad_norm": 1.15625, + "learning_rate": 0.0013443107273613307, + "loss": 0.0996, + "step": 40188 + }, + { + "epoch": 0.3488598189251829, + "grad_norm": 0.134765625, + "learning_rate": 0.0013442820823290136, + "loss": 0.1113, + "step": 40189 + }, + { + "epoch": 0.3488684994053871, + "grad_norm": 0.259765625, + "learning_rate": 0.0013442534370295535, + "loss": 0.0835, + "step": 40190 + }, + { + "epoch": 0.34887717988559125, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013442247914629817, + "loss": 0.0811, + "step": 40191 + }, + { + "epoch": 0.34888586036579544, + "grad_norm": 0.51953125, + "learning_rate": 0.0013441961456293289, + "loss": 0.124, + "step": 40192 + }, + { + "epoch": 0.3488945408459996, + "grad_norm": 0.291015625, + "learning_rate": 0.0013441674995286273, + "loss": 0.0884, + "step": 40193 + }, + { + "epoch": 0.34890322132620377, + "grad_norm": 0.7109375, + "learning_rate": 0.0013441388531609073, + "loss": 0.0835, + "step": 40194 + }, + { + "epoch": 0.3489119018064079, + "grad_norm": 0.61328125, + "learning_rate": 0.0013441102065262012, + "loss": 0.0889, + "step": 40195 + }, + { + "epoch": 0.3489205822866121, + "grad_norm": 0.3984375, + "learning_rate": 0.0013440815596245399, + "loss": 0.1074, + "step": 40196 + }, + { + "epoch": 0.34892926276681624, + "grad_norm": 0.0869140625, + "learning_rate": 0.0013440529124559544, + "loss": 0.0747, + "step": 40197 + }, + { + "epoch": 0.34893794324702043, + "grad_norm": 0.263671875, + "learning_rate": 0.0013440242650204764, + "loss": 0.1357, + "step": 40198 + }, + { + "epoch": 0.34894662372722457, + "grad_norm": 0.126953125, + "learning_rate": 0.0013439956173181369, + "loss": 0.1035, + "step": 40199 + }, + { + "epoch": 0.34895530420742876, + "grad_norm": 0.25, + "learning_rate": 0.0013439669693489679, + "loss": 0.082, + "step": 40200 + }, + { + "epoch": 0.3489639846876329, + "grad_norm": 0.2109375, + "learning_rate": 0.0013439383211130002, + "loss": 0.1182, + "step": 40201 + }, + { + "epoch": 0.3489726651678371, + "grad_norm": 0.36328125, + "learning_rate": 0.0013439096726102647, + "loss": 0.0781, + "step": 40202 + }, + { + "epoch": 0.34898134564804123, + "grad_norm": 0.37109375, + "learning_rate": 0.0013438810238407934, + "loss": 0.0879, + "step": 40203 + }, + { + "epoch": 0.3489900261282454, + "grad_norm": 0.119140625, + "learning_rate": 0.0013438523748046179, + "loss": 0.0742, + "step": 40204 + }, + { + "epoch": 0.34899870660844956, + "grad_norm": 0.375, + "learning_rate": 0.001343823725501769, + "loss": 0.0933, + "step": 40205 + }, + { + "epoch": 0.34900738708865375, + "grad_norm": 0.140625, + "learning_rate": 0.0013437950759322779, + "loss": 0.1416, + "step": 40206 + }, + { + "epoch": 0.3490160675688579, + "grad_norm": 0.1025390625, + "learning_rate": 0.001343766426096176, + "loss": 0.0796, + "step": 40207 + }, + { + "epoch": 0.3490247480490621, + "grad_norm": 0.11474609375, + "learning_rate": 0.0013437377759934952, + "loss": 0.1367, + "step": 40208 + }, + { + "epoch": 0.3490334285292662, + "grad_norm": 0.55859375, + "learning_rate": 0.0013437091256242664, + "loss": 0.1206, + "step": 40209 + }, + { + "epoch": 0.3490421090094704, + "grad_norm": 0.169921875, + "learning_rate": 0.001343680474988521, + "loss": 0.1279, + "step": 40210 + }, + { + "epoch": 0.34905078948967455, + "grad_norm": 0.67578125, + "learning_rate": 0.00134365182408629, + "loss": 0.1025, + "step": 40211 + }, + { + "epoch": 0.34905946996987874, + "grad_norm": 0.29296875, + "learning_rate": 0.001343623172917605, + "loss": 0.0903, + "step": 40212 + }, + { + "epoch": 0.3490681504500829, + "grad_norm": 0.515625, + "learning_rate": 0.0013435945214824978, + "loss": 0.1055, + "step": 40213 + }, + { + "epoch": 0.3490768309302871, + "grad_norm": 0.193359375, + "learning_rate": 0.001343565869780999, + "loss": 0.0869, + "step": 40214 + }, + { + "epoch": 0.3490855114104912, + "grad_norm": 0.59765625, + "learning_rate": 0.0013435372178131403, + "loss": 0.0859, + "step": 40215 + }, + { + "epoch": 0.3490941918906954, + "grad_norm": 0.365234375, + "learning_rate": 0.001343508565578953, + "loss": 0.1094, + "step": 40216 + }, + { + "epoch": 0.34910287237089954, + "grad_norm": 0.07861328125, + "learning_rate": 0.0013434799130784686, + "loss": 0.0796, + "step": 40217 + }, + { + "epoch": 0.34911155285110373, + "grad_norm": 0.2255859375, + "learning_rate": 0.0013434512603117178, + "loss": 0.0767, + "step": 40218 + }, + { + "epoch": 0.34912023333130787, + "grad_norm": 0.5625, + "learning_rate": 0.0013434226072787327, + "loss": 0.084, + "step": 40219 + }, + { + "epoch": 0.34912891381151206, + "grad_norm": 0.119140625, + "learning_rate": 0.0013433939539795438, + "loss": 0.1045, + "step": 40220 + }, + { + "epoch": 0.3491375942917162, + "grad_norm": 0.2216796875, + "learning_rate": 0.0013433653004141833, + "loss": 0.0684, + "step": 40221 + }, + { + "epoch": 0.3491462747719204, + "grad_norm": 0.2177734375, + "learning_rate": 0.001343336646582682, + "loss": 0.0977, + "step": 40222 + }, + { + "epoch": 0.34915495525212453, + "grad_norm": 0.515625, + "learning_rate": 0.0013433079924850715, + "loss": 0.1328, + "step": 40223 + }, + { + "epoch": 0.3491636357323287, + "grad_norm": 0.33203125, + "learning_rate": 0.001343279338121383, + "loss": 0.0947, + "step": 40224 + }, + { + "epoch": 0.34917231621253286, + "grad_norm": 0.130859375, + "learning_rate": 0.001343250683491648, + "loss": 0.1289, + "step": 40225 + }, + { + "epoch": 0.34918099669273706, + "grad_norm": 0.08349609375, + "learning_rate": 0.0013432220285958976, + "loss": 0.0889, + "step": 40226 + }, + { + "epoch": 0.3491896771729412, + "grad_norm": 0.435546875, + "learning_rate": 0.0013431933734341633, + "loss": 0.0986, + "step": 40227 + }, + { + "epoch": 0.3491983576531454, + "grad_norm": 0.859375, + "learning_rate": 0.001343164718006476, + "loss": 0.0991, + "step": 40228 + }, + { + "epoch": 0.3492070381333495, + "grad_norm": 0.138671875, + "learning_rate": 0.001343136062312868, + "loss": 0.0913, + "step": 40229 + }, + { + "epoch": 0.3492157186135537, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013431074063533699, + "loss": 0.0942, + "step": 40230 + }, + { + "epoch": 0.34922439909375785, + "grad_norm": 0.2294921875, + "learning_rate": 0.001343078750128013, + "loss": 0.0811, + "step": 40231 + }, + { + "epoch": 0.34923307957396205, + "grad_norm": 0.181640625, + "learning_rate": 0.0013430500936368291, + "loss": 0.0947, + "step": 40232 + }, + { + "epoch": 0.3492417600541662, + "grad_norm": 0.7109375, + "learning_rate": 0.0013430214368798487, + "loss": 0.1289, + "step": 40233 + }, + { + "epoch": 0.3492504405343704, + "grad_norm": 0.08642578125, + "learning_rate": 0.0013429927798571041, + "loss": 0.1016, + "step": 40234 + }, + { + "epoch": 0.3492591210145745, + "grad_norm": 0.14453125, + "learning_rate": 0.0013429641225686266, + "loss": 0.1055, + "step": 40235 + }, + { + "epoch": 0.3492678014947787, + "grad_norm": 0.326171875, + "learning_rate": 0.001342935465014447, + "loss": 0.1104, + "step": 40236 + }, + { + "epoch": 0.34927648197498284, + "grad_norm": 0.353515625, + "learning_rate": 0.0013429068071945964, + "loss": 0.1025, + "step": 40237 + }, + { + "epoch": 0.34928516245518704, + "grad_norm": 0.734375, + "learning_rate": 0.0013428781491091068, + "loss": 0.2461, + "step": 40238 + }, + { + "epoch": 0.3492938429353912, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013428494907580094, + "loss": 0.1367, + "step": 40239 + }, + { + "epoch": 0.34930252341559537, + "grad_norm": 0.1826171875, + "learning_rate": 0.0013428208321413352, + "loss": 0.1201, + "step": 40240 + }, + { + "epoch": 0.3493112038957995, + "grad_norm": 0.3671875, + "learning_rate": 0.0013427921732591162, + "loss": 0.1172, + "step": 40241 + }, + { + "epoch": 0.3493198843760037, + "grad_norm": 0.12060546875, + "learning_rate": 0.0013427635141113829, + "loss": 0.1201, + "step": 40242 + }, + { + "epoch": 0.34932856485620783, + "grad_norm": 0.69140625, + "learning_rate": 0.0013427348546981672, + "loss": 0.0918, + "step": 40243 + }, + { + "epoch": 0.34933724533641203, + "grad_norm": 0.240234375, + "learning_rate": 0.0013427061950195007, + "loss": 0.104, + "step": 40244 + }, + { + "epoch": 0.34934592581661617, + "grad_norm": 0.1396484375, + "learning_rate": 0.001342677535075414, + "loss": 0.1299, + "step": 40245 + }, + { + "epoch": 0.34935460629682036, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013426488748659387, + "loss": 0.1118, + "step": 40246 + }, + { + "epoch": 0.3493632867770245, + "grad_norm": 0.09912109375, + "learning_rate": 0.0013426202143911063, + "loss": 0.1357, + "step": 40247 + }, + { + "epoch": 0.3493719672572287, + "grad_norm": 0.33203125, + "learning_rate": 0.0013425915536509484, + "loss": 0.1045, + "step": 40248 + }, + { + "epoch": 0.3493806477374328, + "grad_norm": 0.1796875, + "learning_rate": 0.0013425628926454956, + "loss": 0.0928, + "step": 40249 + }, + { + "epoch": 0.349389328217637, + "grad_norm": 0.103515625, + "learning_rate": 0.0013425342313747798, + "loss": 0.0967, + "step": 40250 + }, + { + "epoch": 0.34939800869784116, + "grad_norm": 0.431640625, + "learning_rate": 0.0013425055698388323, + "loss": 0.0942, + "step": 40251 + }, + { + "epoch": 0.34940668917804535, + "grad_norm": 0.177734375, + "learning_rate": 0.001342476908037684, + "loss": 0.1357, + "step": 40252 + }, + { + "epoch": 0.3494153696582495, + "grad_norm": 0.322265625, + "learning_rate": 0.0013424482459713672, + "loss": 0.1079, + "step": 40253 + }, + { + "epoch": 0.3494240501384537, + "grad_norm": 0.134765625, + "learning_rate": 0.0013424195836399123, + "loss": 0.1201, + "step": 40254 + }, + { + "epoch": 0.3494327306186578, + "grad_norm": 0.11083984375, + "learning_rate": 0.001342390921043351, + "loss": 0.0903, + "step": 40255 + }, + { + "epoch": 0.349441411098862, + "grad_norm": 0.35546875, + "learning_rate": 0.0013423622581817147, + "loss": 0.1523, + "step": 40256 + }, + { + "epoch": 0.34945009157906615, + "grad_norm": 0.66015625, + "learning_rate": 0.0013423335950550346, + "loss": 0.1069, + "step": 40257 + }, + { + "epoch": 0.34945877205927034, + "grad_norm": 0.53125, + "learning_rate": 0.0013423049316633423, + "loss": 0.084, + "step": 40258 + }, + { + "epoch": 0.3494674525394745, + "grad_norm": 0.15234375, + "learning_rate": 0.0013422762680066684, + "loss": 0.0894, + "step": 40259 + }, + { + "epoch": 0.34947613301967867, + "grad_norm": 0.1123046875, + "learning_rate": 0.0013422476040850454, + "loss": 0.0977, + "step": 40260 + }, + { + "epoch": 0.3494848134998828, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013422189398985037, + "loss": 0.0942, + "step": 40261 + }, + { + "epoch": 0.349493493980087, + "grad_norm": 0.451171875, + "learning_rate": 0.0013421902754470751, + "loss": 0.0962, + "step": 40262 + }, + { + "epoch": 0.34950217446029114, + "grad_norm": 0.279296875, + "learning_rate": 0.001342161610730791, + "loss": 0.1094, + "step": 40263 + }, + { + "epoch": 0.34951085494049533, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013421329457496826, + "loss": 0.1216, + "step": 40264 + }, + { + "epoch": 0.34951953542069947, + "grad_norm": 0.9296875, + "learning_rate": 0.001342104280503781, + "loss": 0.1123, + "step": 40265 + }, + { + "epoch": 0.34952821590090366, + "grad_norm": 0.271484375, + "learning_rate": 0.0013420756149931184, + "loss": 0.1143, + "step": 40266 + }, + { + "epoch": 0.3495368963811078, + "grad_norm": 0.369140625, + "learning_rate": 0.001342046949217725, + "loss": 0.0854, + "step": 40267 + }, + { + "epoch": 0.349545576861312, + "grad_norm": 0.19140625, + "learning_rate": 0.0013420182831776325, + "loss": 0.1367, + "step": 40268 + }, + { + "epoch": 0.34955425734151613, + "grad_norm": 0.177734375, + "learning_rate": 0.001341989616872873, + "loss": 0.1113, + "step": 40269 + }, + { + "epoch": 0.3495629378217203, + "grad_norm": 0.181640625, + "learning_rate": 0.001341960950303477, + "loss": 0.0947, + "step": 40270 + }, + { + "epoch": 0.34957161830192446, + "grad_norm": 0.166015625, + "learning_rate": 0.0013419322834694763, + "loss": 0.1133, + "step": 40271 + }, + { + "epoch": 0.34958029878212865, + "grad_norm": 0.1123046875, + "learning_rate": 0.001341903616370902, + "loss": 0.0947, + "step": 40272 + }, + { + "epoch": 0.3495889792623328, + "grad_norm": 0.41796875, + "learning_rate": 0.0013418749490077856, + "loss": 0.1221, + "step": 40273 + }, + { + "epoch": 0.349597659742537, + "grad_norm": 0.2451171875, + "learning_rate": 0.0013418462813801582, + "loss": 0.0718, + "step": 40274 + }, + { + "epoch": 0.3496063402227411, + "grad_norm": 0.384765625, + "learning_rate": 0.0013418176134880515, + "loss": 0.1006, + "step": 40275 + }, + { + "epoch": 0.3496150207029453, + "grad_norm": 0.400390625, + "learning_rate": 0.0013417889453314967, + "loss": 0.1396, + "step": 40276 + }, + { + "epoch": 0.34962370118314945, + "grad_norm": 0.2177734375, + "learning_rate": 0.0013417602769105251, + "loss": 0.125, + "step": 40277 + }, + { + "epoch": 0.34963238166335364, + "grad_norm": 0.1416015625, + "learning_rate": 0.001341731608225168, + "loss": 0.0708, + "step": 40278 + }, + { + "epoch": 0.3496410621435578, + "grad_norm": 0.5703125, + "learning_rate": 0.0013417029392754568, + "loss": 0.1055, + "step": 40279 + }, + { + "epoch": 0.34964974262376197, + "grad_norm": 0.14453125, + "learning_rate": 0.001341674270061423, + "loss": 0.125, + "step": 40280 + }, + { + "epoch": 0.3496584231039661, + "grad_norm": 0.1015625, + "learning_rate": 0.0013416456005830976, + "loss": 0.0732, + "step": 40281 + }, + { + "epoch": 0.3496671035841703, + "grad_norm": 0.1748046875, + "learning_rate": 0.0013416169308405129, + "loss": 0.1191, + "step": 40282 + }, + { + "epoch": 0.34967578406437444, + "grad_norm": 0.1904296875, + "learning_rate": 0.0013415882608336992, + "loss": 0.1562, + "step": 40283 + }, + { + "epoch": 0.34968446454457863, + "grad_norm": 0.11669921875, + "learning_rate": 0.0013415595905626881, + "loss": 0.0811, + "step": 40284 + }, + { + "epoch": 0.34969314502478277, + "grad_norm": 0.11474609375, + "learning_rate": 0.001341530920027511, + "loss": 0.1094, + "step": 40285 + }, + { + "epoch": 0.34970182550498696, + "grad_norm": 0.0751953125, + "learning_rate": 0.0013415022492281993, + "loss": 0.0562, + "step": 40286 + }, + { + "epoch": 0.3497105059851911, + "grad_norm": 0.265625, + "learning_rate": 0.0013414735781647846, + "loss": 0.0679, + "step": 40287 + }, + { + "epoch": 0.3497191864653953, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013414449068372978, + "loss": 0.0991, + "step": 40288 + }, + { + "epoch": 0.34972786694559943, + "grad_norm": 0.38671875, + "learning_rate": 0.0013414162352457706, + "loss": 0.1367, + "step": 40289 + }, + { + "epoch": 0.34973654742580357, + "grad_norm": 0.05908203125, + "learning_rate": 0.001341387563390234, + "loss": 0.0762, + "step": 40290 + }, + { + "epoch": 0.34974522790600776, + "grad_norm": 0.1064453125, + "learning_rate": 0.0013413588912707202, + "loss": 0.125, + "step": 40291 + }, + { + "epoch": 0.3497539083862119, + "grad_norm": 0.1572265625, + "learning_rate": 0.0013413302188872594, + "loss": 0.1357, + "step": 40292 + }, + { + "epoch": 0.3497625888664161, + "grad_norm": 0.10888671875, + "learning_rate": 0.0013413015462398837, + "loss": 0.1084, + "step": 40293 + }, + { + "epoch": 0.34977126934662023, + "grad_norm": 0.0908203125, + "learning_rate": 0.0013412728733286242, + "loss": 0.1016, + "step": 40294 + }, + { + "epoch": 0.3497799498268244, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013412442001535124, + "loss": 0.1309, + "step": 40295 + }, + { + "epoch": 0.34978863030702856, + "grad_norm": 0.88671875, + "learning_rate": 0.0013412155267145795, + "loss": 0.1406, + "step": 40296 + }, + { + "epoch": 0.34979731078723275, + "grad_norm": 0.212890625, + "learning_rate": 0.0013411868530118567, + "loss": 0.0967, + "step": 40297 + }, + { + "epoch": 0.3498059912674369, + "grad_norm": 0.1513671875, + "learning_rate": 0.001341158179045376, + "loss": 0.0898, + "step": 40298 + }, + { + "epoch": 0.3498146717476411, + "grad_norm": 0.09521484375, + "learning_rate": 0.001341129504815168, + "loss": 0.0938, + "step": 40299 + }, + { + "epoch": 0.3498233522278452, + "grad_norm": 0.1435546875, + "learning_rate": 0.0013411008303212647, + "loss": 0.1162, + "step": 40300 + }, + { + "epoch": 0.3498320327080494, + "grad_norm": 0.166015625, + "learning_rate": 0.0013410721555636967, + "loss": 0.0815, + "step": 40301 + }, + { + "epoch": 0.34984071318825355, + "grad_norm": 0.326171875, + "learning_rate": 0.001341043480542496, + "loss": 0.1562, + "step": 40302 + }, + { + "epoch": 0.34984939366845774, + "grad_norm": 0.28515625, + "learning_rate": 0.001341014805257694, + "loss": 0.0908, + "step": 40303 + }, + { + "epoch": 0.3498580741486619, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013409861297093218, + "loss": 0.0781, + "step": 40304 + }, + { + "epoch": 0.3498667546288661, + "grad_norm": 0.2333984375, + "learning_rate": 0.0013409574538974108, + "loss": 0.1006, + "step": 40305 + }, + { + "epoch": 0.3498754351090702, + "grad_norm": 0.447265625, + "learning_rate": 0.0013409287778219918, + "loss": 0.1025, + "step": 40306 + }, + { + "epoch": 0.3498841155892744, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013409001014830974, + "loss": 0.1006, + "step": 40307 + }, + { + "epoch": 0.34989279606947854, + "grad_norm": 0.4453125, + "learning_rate": 0.001340871424880758, + "loss": 0.0747, + "step": 40308 + }, + { + "epoch": 0.34990147654968273, + "grad_norm": 0.625, + "learning_rate": 0.001340842748015005, + "loss": 0.1523, + "step": 40309 + }, + { + "epoch": 0.34991015702988687, + "grad_norm": 0.2314453125, + "learning_rate": 0.00134081407088587, + "loss": 0.0879, + "step": 40310 + }, + { + "epoch": 0.34991883751009106, + "grad_norm": 0.150390625, + "learning_rate": 0.0013407853934933847, + "loss": 0.1377, + "step": 40311 + }, + { + "epoch": 0.3499275179902952, + "grad_norm": 0.287109375, + "learning_rate": 0.0013407567158375802, + "loss": 0.0669, + "step": 40312 + }, + { + "epoch": 0.3499361984704994, + "grad_norm": 0.73046875, + "learning_rate": 0.0013407280379184874, + "loss": 0.084, + "step": 40313 + }, + { + "epoch": 0.34994487895070353, + "grad_norm": 0.55078125, + "learning_rate": 0.0013406993597361383, + "loss": 0.1406, + "step": 40314 + }, + { + "epoch": 0.3499535594309077, + "grad_norm": 0.16796875, + "learning_rate": 0.001340670681290564, + "loss": 0.0845, + "step": 40315 + }, + { + "epoch": 0.34996223991111186, + "grad_norm": 0.48828125, + "learning_rate": 0.0013406420025817959, + "loss": 0.1025, + "step": 40316 + }, + { + "epoch": 0.34997092039131605, + "grad_norm": 0.65625, + "learning_rate": 0.001340613323609865, + "loss": 0.1494, + "step": 40317 + }, + { + "epoch": 0.3499796008715202, + "grad_norm": 0.380859375, + "learning_rate": 0.0013405846443748028, + "loss": 0.1167, + "step": 40318 + }, + { + "epoch": 0.3499882813517244, + "grad_norm": 0.42578125, + "learning_rate": 0.0013405559648766415, + "loss": 0.1943, + "step": 40319 + }, + { + "epoch": 0.3499969618319285, + "grad_norm": 0.578125, + "learning_rate": 0.0013405272851154113, + "loss": 0.1201, + "step": 40320 + }, + { + "epoch": 0.3500056423121327, + "grad_norm": 0.279296875, + "learning_rate": 0.0013404986050911446, + "loss": 0.0967, + "step": 40321 + }, + { + "epoch": 0.35001432279233685, + "grad_norm": 0.2578125, + "learning_rate": 0.0013404699248038717, + "loss": 0.0786, + "step": 40322 + }, + { + "epoch": 0.35002300327254104, + "grad_norm": 0.1953125, + "learning_rate": 0.0013404412442536246, + "loss": 0.0684, + "step": 40323 + }, + { + "epoch": 0.3500316837527452, + "grad_norm": 0.1708984375, + "learning_rate": 0.0013404125634404347, + "loss": 0.1152, + "step": 40324 + }, + { + "epoch": 0.3500403642329494, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013403838823643334, + "loss": 0.0967, + "step": 40325 + }, + { + "epoch": 0.3500490447131535, + "grad_norm": 0.1865234375, + "learning_rate": 0.0013403552010253519, + "loss": 0.127, + "step": 40326 + }, + { + "epoch": 0.3500577251933577, + "grad_norm": 0.78125, + "learning_rate": 0.001340326519423521, + "loss": 0.1006, + "step": 40327 + }, + { + "epoch": 0.35006640567356184, + "grad_norm": 0.85546875, + "learning_rate": 0.001340297837558873, + "loss": 0.1196, + "step": 40328 + }, + { + "epoch": 0.35007508615376604, + "grad_norm": 0.2451171875, + "learning_rate": 0.001340269155431439, + "loss": 0.0884, + "step": 40329 + }, + { + "epoch": 0.3500837666339702, + "grad_norm": 0.08984375, + "learning_rate": 0.0013402404730412502, + "loss": 0.085, + "step": 40330 + }, + { + "epoch": 0.35009244711417437, + "grad_norm": 0.1826171875, + "learning_rate": 0.0013402117903883381, + "loss": 0.124, + "step": 40331 + }, + { + "epoch": 0.3501011275943785, + "grad_norm": 0.0908203125, + "learning_rate": 0.001340183107472734, + "loss": 0.0845, + "step": 40332 + }, + { + "epoch": 0.3501098080745827, + "grad_norm": 0.86328125, + "learning_rate": 0.001340154424294469, + "loss": 0.1465, + "step": 40333 + }, + { + "epoch": 0.35011848855478683, + "grad_norm": 0.12060546875, + "learning_rate": 0.001340125740853575, + "loss": 0.0889, + "step": 40334 + }, + { + "epoch": 0.350127169034991, + "grad_norm": 0.17578125, + "learning_rate": 0.001340097057150083, + "loss": 0.1328, + "step": 40335 + }, + { + "epoch": 0.35013584951519516, + "grad_norm": 0.48828125, + "learning_rate": 0.0013400683731840243, + "loss": 0.0918, + "step": 40336 + }, + { + "epoch": 0.35014452999539936, + "grad_norm": 0.2265625, + "learning_rate": 0.0013400396889554306, + "loss": 0.0869, + "step": 40337 + }, + { + "epoch": 0.3501532104756035, + "grad_norm": 0.470703125, + "learning_rate": 0.001340011004464333, + "loss": 0.1064, + "step": 40338 + }, + { + "epoch": 0.3501618909558077, + "grad_norm": 0.08984375, + "learning_rate": 0.0013399823197107633, + "loss": 0.082, + "step": 40339 + }, + { + "epoch": 0.3501705714360118, + "grad_norm": 0.79296875, + "learning_rate": 0.0013399536346947523, + "loss": 0.0752, + "step": 40340 + }, + { + "epoch": 0.350179251916216, + "grad_norm": 0.08837890625, + "learning_rate": 0.0013399249494163315, + "loss": 0.0771, + "step": 40341 + }, + { + "epoch": 0.35018793239642015, + "grad_norm": 0.17578125, + "learning_rate": 0.0013398962638755325, + "loss": 0.0918, + "step": 40342 + }, + { + "epoch": 0.35019661287662435, + "grad_norm": 0.185546875, + "learning_rate": 0.0013398675780723866, + "loss": 0.1318, + "step": 40343 + }, + { + "epoch": 0.3502052933568285, + "grad_norm": 0.3828125, + "learning_rate": 0.0013398388920069252, + "loss": 0.1328, + "step": 40344 + }, + { + "epoch": 0.3502139738370327, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013398102056791793, + "loss": 0.126, + "step": 40345 + }, + { + "epoch": 0.3502226543172368, + "grad_norm": 0.1640625, + "learning_rate": 0.0013397815190891807, + "loss": 0.0918, + "step": 40346 + }, + { + "epoch": 0.350231334797441, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013397528322369606, + "loss": 0.0967, + "step": 40347 + }, + { + "epoch": 0.35024001527764514, + "grad_norm": 0.07275390625, + "learning_rate": 0.0013397241451225505, + "loss": 0.0664, + "step": 40348 + }, + { + "epoch": 0.35024869575784934, + "grad_norm": 0.2265625, + "learning_rate": 0.0013396954577459814, + "loss": 0.1152, + "step": 40349 + }, + { + "epoch": 0.3502573762380535, + "grad_norm": 0.146484375, + "learning_rate": 0.0013396667701072852, + "loss": 0.0835, + "step": 40350 + }, + { + "epoch": 0.35026605671825767, + "grad_norm": 0.119140625, + "learning_rate": 0.0013396380822064933, + "loss": 0.1387, + "step": 40351 + }, + { + "epoch": 0.3502747371984618, + "grad_norm": 0.451171875, + "learning_rate": 0.0013396093940436363, + "loss": 0.1108, + "step": 40352 + }, + { + "epoch": 0.350283417678666, + "grad_norm": 0.3828125, + "learning_rate": 0.0013395807056187463, + "loss": 0.104, + "step": 40353 + }, + { + "epoch": 0.35029209815887014, + "grad_norm": 0.578125, + "learning_rate": 0.0013395520169318545, + "loss": 0.1147, + "step": 40354 + }, + { + "epoch": 0.35030077863907433, + "grad_norm": 0.123046875, + "learning_rate": 0.001339523327982992, + "loss": 0.1206, + "step": 40355 + }, + { + "epoch": 0.35030945911927847, + "grad_norm": 0.09765625, + "learning_rate": 0.0013394946387721906, + "loss": 0.1133, + "step": 40356 + }, + { + "epoch": 0.35031813959948266, + "grad_norm": 0.322265625, + "learning_rate": 0.0013394659492994812, + "loss": 0.1064, + "step": 40357 + }, + { + "epoch": 0.3503268200796868, + "grad_norm": 1.0, + "learning_rate": 0.0013394372595648956, + "loss": 0.1309, + "step": 40358 + }, + { + "epoch": 0.350335500559891, + "grad_norm": 0.4140625, + "learning_rate": 0.001339408569568465, + "loss": 0.1221, + "step": 40359 + }, + { + "epoch": 0.3503441810400951, + "grad_norm": 0.201171875, + "learning_rate": 0.001339379879310221, + "loss": 0.126, + "step": 40360 + }, + { + "epoch": 0.3503528615202993, + "grad_norm": 0.142578125, + "learning_rate": 0.0013393511887901945, + "loss": 0.0957, + "step": 40361 + }, + { + "epoch": 0.35036154200050346, + "grad_norm": 0.5234375, + "learning_rate": 0.0013393224980084173, + "loss": 0.1167, + "step": 40362 + }, + { + "epoch": 0.35037022248070765, + "grad_norm": 0.142578125, + "learning_rate": 0.0013392938069649203, + "loss": 0.1357, + "step": 40363 + }, + { + "epoch": 0.3503789029609118, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013392651156597356, + "loss": 0.0815, + "step": 40364 + }, + { + "epoch": 0.350387583441116, + "grad_norm": 0.44140625, + "learning_rate": 0.0013392364240928938, + "loss": 0.0869, + "step": 40365 + }, + { + "epoch": 0.3503962639213201, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013392077322644269, + "loss": 0.1006, + "step": 40366 + }, + { + "epoch": 0.3504049444015243, + "grad_norm": 0.23828125, + "learning_rate": 0.0013391790401743656, + "loss": 0.0688, + "step": 40367 + }, + { + "epoch": 0.35041362488172845, + "grad_norm": 0.255859375, + "learning_rate": 0.0013391503478227423, + "loss": 0.106, + "step": 40368 + }, + { + "epoch": 0.35042230536193264, + "grad_norm": 0.333984375, + "learning_rate": 0.0013391216552095873, + "loss": 0.1113, + "step": 40369 + }, + { + "epoch": 0.3504309858421368, + "grad_norm": 0.55859375, + "learning_rate": 0.0013390929623349332, + "loss": 0.0918, + "step": 40370 + }, + { + "epoch": 0.35043966632234097, + "grad_norm": 0.337890625, + "learning_rate": 0.0013390642691988096, + "loss": 0.1426, + "step": 40371 + }, + { + "epoch": 0.3504483468025451, + "grad_norm": 0.369140625, + "learning_rate": 0.0013390355758012495, + "loss": 0.1108, + "step": 40372 + }, + { + "epoch": 0.3504570272827493, + "grad_norm": 0.34375, + "learning_rate": 0.0013390068821422835, + "loss": 0.1191, + "step": 40373 + }, + { + "epoch": 0.35046570776295344, + "grad_norm": 0.115234375, + "learning_rate": 0.0013389781882219435, + "loss": 0.0898, + "step": 40374 + }, + { + "epoch": 0.35047438824315763, + "grad_norm": 0.1416015625, + "learning_rate": 0.00133894949404026, + "loss": 0.1309, + "step": 40375 + }, + { + "epoch": 0.35048306872336177, + "grad_norm": 0.53515625, + "learning_rate": 0.001338920799597265, + "loss": 0.0781, + "step": 40376 + }, + { + "epoch": 0.35049174920356596, + "grad_norm": 0.14453125, + "learning_rate": 0.0013388921048929903, + "loss": 0.1367, + "step": 40377 + }, + { + "epoch": 0.3505004296837701, + "grad_norm": 0.11669921875, + "learning_rate": 0.0013388634099274664, + "loss": 0.0957, + "step": 40378 + }, + { + "epoch": 0.3505091101639743, + "grad_norm": 0.208984375, + "learning_rate": 0.0013388347147007252, + "loss": 0.125, + "step": 40379 + }, + { + "epoch": 0.35051779064417843, + "grad_norm": 0.08642578125, + "learning_rate": 0.0013388060192127977, + "loss": 0.0684, + "step": 40380 + }, + { + "epoch": 0.3505264711243826, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013387773234637156, + "loss": 0.1172, + "step": 40381 + }, + { + "epoch": 0.35053515160458676, + "grad_norm": 0.408203125, + "learning_rate": 0.0013387486274535107, + "loss": 0.1128, + "step": 40382 + }, + { + "epoch": 0.35054383208479095, + "grad_norm": 0.08349609375, + "learning_rate": 0.0013387199311822133, + "loss": 0.0762, + "step": 40383 + }, + { + "epoch": 0.3505525125649951, + "grad_norm": 0.271484375, + "learning_rate": 0.0013386912346498555, + "loss": 0.1211, + "step": 40384 + }, + { + "epoch": 0.3505611930451993, + "grad_norm": 0.298828125, + "learning_rate": 0.0013386625378564686, + "loss": 0.1387, + "step": 40385 + }, + { + "epoch": 0.3505698735254034, + "grad_norm": 0.3984375, + "learning_rate": 0.0013386338408020839, + "loss": 0.1289, + "step": 40386 + }, + { + "epoch": 0.3505785540056076, + "grad_norm": 0.412109375, + "learning_rate": 0.001338605143486733, + "loss": 0.0889, + "step": 40387 + }, + { + "epoch": 0.35058723448581175, + "grad_norm": 0.06884765625, + "learning_rate": 0.0013385764459104468, + "loss": 0.1001, + "step": 40388 + }, + { + "epoch": 0.35059591496601594, + "grad_norm": 0.109375, + "learning_rate": 0.0013385477480732572, + "loss": 0.1172, + "step": 40389 + }, + { + "epoch": 0.3506045954462201, + "grad_norm": 0.10546875, + "learning_rate": 0.0013385190499751955, + "loss": 0.0825, + "step": 40390 + }, + { + "epoch": 0.3506132759264243, + "grad_norm": 0.326171875, + "learning_rate": 0.001338490351616293, + "loss": 0.1631, + "step": 40391 + }, + { + "epoch": 0.3506219564066284, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013384616529965808, + "loss": 0.104, + "step": 40392 + }, + { + "epoch": 0.3506306368868326, + "grad_norm": 0.20703125, + "learning_rate": 0.0013384329541160902, + "loss": 0.0947, + "step": 40393 + }, + { + "epoch": 0.35063931736703674, + "grad_norm": 0.0859375, + "learning_rate": 0.0013384042549748536, + "loss": 0.1079, + "step": 40394 + }, + { + "epoch": 0.35064799784724093, + "grad_norm": 0.1669921875, + "learning_rate": 0.001338375555572901, + "loss": 0.1523, + "step": 40395 + }, + { + "epoch": 0.35065667832744507, + "grad_norm": 0.150390625, + "learning_rate": 0.001338346855910265, + "loss": 0.1006, + "step": 40396 + }, + { + "epoch": 0.35066535880764926, + "grad_norm": 0.2421875, + "learning_rate": 0.0013383181559869762, + "loss": 0.0898, + "step": 40397 + }, + { + "epoch": 0.3506740392878534, + "grad_norm": 2.1875, + "learning_rate": 0.0013382894558030662, + "loss": 0.1416, + "step": 40398 + }, + { + "epoch": 0.3506827197680576, + "grad_norm": 0.3671875, + "learning_rate": 0.0013382607553585667, + "loss": 0.0869, + "step": 40399 + }, + { + "epoch": 0.35069140024826173, + "grad_norm": 0.08251953125, + "learning_rate": 0.0013382320546535085, + "loss": 0.0815, + "step": 40400 + }, + { + "epoch": 0.3507000807284659, + "grad_norm": 0.2001953125, + "learning_rate": 0.0013382033536879234, + "loss": 0.1162, + "step": 40401 + }, + { + "epoch": 0.35070876120867006, + "grad_norm": 0.2470703125, + "learning_rate": 0.001338174652461843, + "loss": 0.104, + "step": 40402 + }, + { + "epoch": 0.35071744168887425, + "grad_norm": 0.1748046875, + "learning_rate": 0.001338145950975298, + "loss": 0.1221, + "step": 40403 + }, + { + "epoch": 0.3507261221690784, + "grad_norm": 0.08056640625, + "learning_rate": 0.0013381172492283203, + "loss": 0.0664, + "step": 40404 + }, + { + "epoch": 0.3507348026492826, + "grad_norm": 0.984375, + "learning_rate": 0.001338088547220941, + "loss": 0.1797, + "step": 40405 + }, + { + "epoch": 0.3507434831294867, + "grad_norm": 0.302734375, + "learning_rate": 0.0013380598449531918, + "loss": 0.1123, + "step": 40406 + }, + { + "epoch": 0.3507521636096909, + "grad_norm": 0.25390625, + "learning_rate": 0.001338031142425104, + "loss": 0.1016, + "step": 40407 + }, + { + "epoch": 0.35076084408989505, + "grad_norm": 0.46484375, + "learning_rate": 0.001338002439636709, + "loss": 0.1016, + "step": 40408 + }, + { + "epoch": 0.35076952457009924, + "grad_norm": 0.251953125, + "learning_rate": 0.0013379737365880379, + "loss": 0.0879, + "step": 40409 + }, + { + "epoch": 0.3507782050503034, + "grad_norm": 0.34375, + "learning_rate": 0.0013379450332791222, + "loss": 0.1094, + "step": 40410 + }, + { + "epoch": 0.3507868855305076, + "grad_norm": 0.2001953125, + "learning_rate": 0.0013379163297099937, + "loss": 0.1211, + "step": 40411 + }, + { + "epoch": 0.3507955660107117, + "grad_norm": 0.2021484375, + "learning_rate": 0.001337887625880683, + "loss": 0.0815, + "step": 40412 + }, + { + "epoch": 0.35080424649091585, + "grad_norm": 0.72265625, + "learning_rate": 0.0013378589217912225, + "loss": 0.1211, + "step": 40413 + }, + { + "epoch": 0.35081292697112004, + "grad_norm": 0.15625, + "learning_rate": 0.0013378302174416425, + "loss": 0.0708, + "step": 40414 + }, + { + "epoch": 0.3508216074513242, + "grad_norm": 0.76171875, + "learning_rate": 0.0013378015128319755, + "loss": 0.1172, + "step": 40415 + }, + { + "epoch": 0.3508302879315284, + "grad_norm": 0.15625, + "learning_rate": 0.0013377728079622519, + "loss": 0.0957, + "step": 40416 + }, + { + "epoch": 0.3508389684117325, + "grad_norm": 0.2275390625, + "learning_rate": 0.001337744102832504, + "loss": 0.0898, + "step": 40417 + }, + { + "epoch": 0.3508476488919367, + "grad_norm": 0.240234375, + "learning_rate": 0.0013377153974427625, + "loss": 0.0918, + "step": 40418 + }, + { + "epoch": 0.35085632937214084, + "grad_norm": 0.291015625, + "learning_rate": 0.0013376866917930589, + "loss": 0.0752, + "step": 40419 + }, + { + "epoch": 0.35086500985234503, + "grad_norm": 0.412109375, + "learning_rate": 0.0013376579858834247, + "loss": 0.1104, + "step": 40420 + }, + { + "epoch": 0.35087369033254917, + "grad_norm": 0.65234375, + "learning_rate": 0.0013376292797138915, + "loss": 0.0977, + "step": 40421 + }, + { + "epoch": 0.35088237081275336, + "grad_norm": 0.1474609375, + "learning_rate": 0.0013376005732844902, + "loss": 0.1621, + "step": 40422 + }, + { + "epoch": 0.3508910512929575, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013375718665952527, + "loss": 0.1172, + "step": 40423 + }, + { + "epoch": 0.3508997317731617, + "grad_norm": 0.318359375, + "learning_rate": 0.0013375431596462102, + "loss": 0.0918, + "step": 40424 + }, + { + "epoch": 0.35090841225336583, + "grad_norm": 0.1162109375, + "learning_rate": 0.0013375144524373941, + "loss": 0.1118, + "step": 40425 + }, + { + "epoch": 0.35091709273357, + "grad_norm": 0.26953125, + "learning_rate": 0.0013374857449688358, + "loss": 0.0977, + "step": 40426 + }, + { + "epoch": 0.35092577321377416, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013374570372405663, + "loss": 0.0938, + "step": 40427 + }, + { + "epoch": 0.35093445369397835, + "grad_norm": 0.1181640625, + "learning_rate": 0.0013374283292526177, + "loss": 0.0879, + "step": 40428 + }, + { + "epoch": 0.3509431341741825, + "grad_norm": 0.23046875, + "learning_rate": 0.001337399621005021, + "loss": 0.1074, + "step": 40429 + }, + { + "epoch": 0.3509518146543867, + "grad_norm": 0.26171875, + "learning_rate": 0.0013373709124978078, + "loss": 0.0942, + "step": 40430 + }, + { + "epoch": 0.3509604951345908, + "grad_norm": 0.091796875, + "learning_rate": 0.0013373422037310089, + "loss": 0.1279, + "step": 40431 + }, + { + "epoch": 0.350969175614795, + "grad_norm": 0.234375, + "learning_rate": 0.001337313494704656, + "loss": 0.0947, + "step": 40432 + }, + { + "epoch": 0.35097785609499915, + "grad_norm": 0.55859375, + "learning_rate": 0.001337284785418781, + "loss": 0.1064, + "step": 40433 + }, + { + "epoch": 0.35098653657520335, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013372560758734152, + "loss": 0.0981, + "step": 40434 + }, + { + "epoch": 0.3509952170554075, + "grad_norm": 0.349609375, + "learning_rate": 0.0013372273660685891, + "loss": 0.1084, + "step": 40435 + }, + { + "epoch": 0.3510038975356117, + "grad_norm": 0.53125, + "learning_rate": 0.001337198656004335, + "loss": 0.1992, + "step": 40436 + }, + { + "epoch": 0.3510125780158158, + "grad_norm": 0.140625, + "learning_rate": 0.001337169945680684, + "loss": 0.1182, + "step": 40437 + }, + { + "epoch": 0.35102125849602, + "grad_norm": 0.11572265625, + "learning_rate": 0.0013371412350976677, + "loss": 0.0972, + "step": 40438 + }, + { + "epoch": 0.35102993897622414, + "grad_norm": 0.2001953125, + "learning_rate": 0.0013371125242553174, + "loss": 0.082, + "step": 40439 + }, + { + "epoch": 0.35103861945642834, + "grad_norm": 0.216796875, + "learning_rate": 0.001337083813153664, + "loss": 0.103, + "step": 40440 + }, + { + "epoch": 0.3510472999366325, + "grad_norm": 0.62890625, + "learning_rate": 0.0013370551017927396, + "loss": 0.0942, + "step": 40441 + }, + { + "epoch": 0.35105598041683667, + "grad_norm": 0.470703125, + "learning_rate": 0.001337026390172575, + "loss": 0.1211, + "step": 40442 + }, + { + "epoch": 0.3510646608970408, + "grad_norm": 0.302734375, + "learning_rate": 0.0013369976782932021, + "loss": 0.1094, + "step": 40443 + }, + { + "epoch": 0.351073341377245, + "grad_norm": 0.1708984375, + "learning_rate": 0.001336968966154652, + "loss": 0.1162, + "step": 40444 + }, + { + "epoch": 0.35108202185744913, + "grad_norm": 0.248046875, + "learning_rate": 0.0013369402537569562, + "loss": 0.0942, + "step": 40445 + }, + { + "epoch": 0.3510907023376533, + "grad_norm": 0.328125, + "learning_rate": 0.0013369115411001463, + "loss": 0.0918, + "step": 40446 + }, + { + "epoch": 0.35109938281785746, + "grad_norm": 0.59375, + "learning_rate": 0.0013368828281842533, + "loss": 0.2051, + "step": 40447 + }, + { + "epoch": 0.35110806329806166, + "grad_norm": 0.1787109375, + "learning_rate": 0.0013368541150093089, + "loss": 0.1113, + "step": 40448 + }, + { + "epoch": 0.3511167437782658, + "grad_norm": 0.10888671875, + "learning_rate": 0.0013368254015753442, + "loss": 0.126, + "step": 40449 + }, + { + "epoch": 0.35112542425847, + "grad_norm": 0.130859375, + "learning_rate": 0.0013367966878823911, + "loss": 0.1006, + "step": 40450 + }, + { + "epoch": 0.3511341047386741, + "grad_norm": 0.267578125, + "learning_rate": 0.0013367679739304805, + "loss": 0.1079, + "step": 40451 + }, + { + "epoch": 0.3511427852188783, + "grad_norm": 0.302734375, + "learning_rate": 0.001336739259719644, + "loss": 0.1992, + "step": 40452 + }, + { + "epoch": 0.35115146569908245, + "grad_norm": 0.2451171875, + "learning_rate": 0.001336710545249913, + "loss": 0.1338, + "step": 40453 + }, + { + "epoch": 0.35116014617928665, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013366818305213192, + "loss": 0.1157, + "step": 40454 + }, + { + "epoch": 0.3511688266594908, + "grad_norm": 0.1064453125, + "learning_rate": 0.0013366531155338931, + "loss": 0.1074, + "step": 40455 + }, + { + "epoch": 0.351177507139695, + "grad_norm": 0.18359375, + "learning_rate": 0.0013366244002876676, + "loss": 0.0747, + "step": 40456 + }, + { + "epoch": 0.3511861876198991, + "grad_norm": 2.4375, + "learning_rate": 0.0013365956847826723, + "loss": 0.2891, + "step": 40457 + }, + { + "epoch": 0.3511948681001033, + "grad_norm": 0.1005859375, + "learning_rate": 0.0013365669690189401, + "loss": 0.1279, + "step": 40458 + }, + { + "epoch": 0.35120354858030745, + "grad_norm": 0.302734375, + "learning_rate": 0.0013365382529965015, + "loss": 0.0591, + "step": 40459 + }, + { + "epoch": 0.35121222906051164, + "grad_norm": 0.65625, + "learning_rate": 0.0013365095367153885, + "loss": 0.1099, + "step": 40460 + }, + { + "epoch": 0.3512209095407158, + "grad_norm": 0.4453125, + "learning_rate": 0.001336480820175632, + "loss": 0.123, + "step": 40461 + }, + { + "epoch": 0.35122959002091997, + "grad_norm": 0.1328125, + "learning_rate": 0.0013364521033772637, + "loss": 0.063, + "step": 40462 + }, + { + "epoch": 0.3512382705011241, + "grad_norm": 0.1865234375, + "learning_rate": 0.0013364233863203147, + "loss": 0.0776, + "step": 40463 + }, + { + "epoch": 0.3512469509813283, + "grad_norm": 0.416015625, + "learning_rate": 0.001336394669004817, + "loss": 0.0894, + "step": 40464 + }, + { + "epoch": 0.35125563146153244, + "grad_norm": 0.2578125, + "learning_rate": 0.0013363659514308017, + "loss": 0.0845, + "step": 40465 + }, + { + "epoch": 0.35126431194173663, + "grad_norm": 0.0966796875, + "learning_rate": 0.0013363372335983, + "loss": 0.0742, + "step": 40466 + }, + { + "epoch": 0.35127299242194077, + "grad_norm": 1.140625, + "learning_rate": 0.0013363085155073434, + "loss": 0.168, + "step": 40467 + }, + { + "epoch": 0.35128167290214496, + "grad_norm": 0.1435546875, + "learning_rate": 0.0013362797971579632, + "loss": 0.083, + "step": 40468 + }, + { + "epoch": 0.3512903533823491, + "grad_norm": 0.31640625, + "learning_rate": 0.0013362510785501914, + "loss": 0.0771, + "step": 40469 + }, + { + "epoch": 0.3512990338625533, + "grad_norm": 0.6796875, + "learning_rate": 0.0013362223596840586, + "loss": 0.1738, + "step": 40470 + }, + { + "epoch": 0.3513077143427574, + "grad_norm": 0.0859375, + "learning_rate": 0.0013361936405595965, + "loss": 0.0679, + "step": 40471 + }, + { + "epoch": 0.3513163948229616, + "grad_norm": 0.30078125, + "learning_rate": 0.0013361649211768368, + "loss": 0.1162, + "step": 40472 + }, + { + "epoch": 0.35132507530316576, + "grad_norm": 0.25390625, + "learning_rate": 0.0013361362015358107, + "loss": 0.1289, + "step": 40473 + }, + { + "epoch": 0.35133375578336995, + "grad_norm": 0.205078125, + "learning_rate": 0.0013361074816365494, + "loss": 0.1279, + "step": 40474 + }, + { + "epoch": 0.3513424362635741, + "grad_norm": 0.134765625, + "learning_rate": 0.0013360787614790845, + "loss": 0.123, + "step": 40475 + }, + { + "epoch": 0.3513511167437783, + "grad_norm": 0.7109375, + "learning_rate": 0.0013360500410634479, + "loss": 0.1211, + "step": 40476 + }, + { + "epoch": 0.3513597972239824, + "grad_norm": 0.6796875, + "learning_rate": 0.0013360213203896703, + "loss": 0.1045, + "step": 40477 + }, + { + "epoch": 0.3513684777041866, + "grad_norm": 0.08544921875, + "learning_rate": 0.0013359925994577831, + "loss": 0.0903, + "step": 40478 + }, + { + "epoch": 0.35137715818439075, + "grad_norm": 0.466796875, + "learning_rate": 0.001335963878267818, + "loss": 0.082, + "step": 40479 + }, + { + "epoch": 0.35138583866459494, + "grad_norm": 0.345703125, + "learning_rate": 0.0013359351568198064, + "loss": 0.0801, + "step": 40480 + }, + { + "epoch": 0.3513945191447991, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013359064351137794, + "loss": 0.0913, + "step": 40481 + }, + { + "epoch": 0.35140319962500327, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013358777131497693, + "loss": 0.0894, + "step": 40482 + }, + { + "epoch": 0.3514118801052074, + "grad_norm": 0.15625, + "learning_rate": 0.0013358489909278061, + "loss": 0.1104, + "step": 40483 + }, + { + "epoch": 0.3514205605854116, + "grad_norm": 0.21875, + "learning_rate": 0.0013358202684479227, + "loss": 0.1123, + "step": 40484 + }, + { + "epoch": 0.35142924106561574, + "grad_norm": 0.1875, + "learning_rate": 0.0013357915457101494, + "loss": 0.1602, + "step": 40485 + }, + { + "epoch": 0.35143792154581993, + "grad_norm": 0.92578125, + "learning_rate": 0.0013357628227145181, + "loss": 0.1338, + "step": 40486 + }, + { + "epoch": 0.35144660202602407, + "grad_norm": 0.25, + "learning_rate": 0.0013357340994610604, + "loss": 0.0645, + "step": 40487 + }, + { + "epoch": 0.35145528250622826, + "grad_norm": 0.23046875, + "learning_rate": 0.0013357053759498068, + "loss": 0.1162, + "step": 40488 + }, + { + "epoch": 0.3514639629864324, + "grad_norm": 0.8125, + "learning_rate": 0.0013356766521807897, + "loss": 0.0801, + "step": 40489 + }, + { + "epoch": 0.3514726434666366, + "grad_norm": 0.48828125, + "learning_rate": 0.0013356479281540403, + "loss": 0.1006, + "step": 40490 + }, + { + "epoch": 0.35148132394684073, + "grad_norm": 0.55859375, + "learning_rate": 0.0013356192038695895, + "loss": 0.0957, + "step": 40491 + }, + { + "epoch": 0.3514900044270449, + "grad_norm": 0.2392578125, + "learning_rate": 0.0013355904793274694, + "loss": 0.1143, + "step": 40492 + }, + { + "epoch": 0.35149868490724906, + "grad_norm": 0.287109375, + "learning_rate": 0.001335561754527711, + "loss": 0.1367, + "step": 40493 + }, + { + "epoch": 0.35150736538745325, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013355330294703457, + "loss": 0.0767, + "step": 40494 + }, + { + "epoch": 0.3515160458676574, + "grad_norm": 0.1689453125, + "learning_rate": 0.0013355043041554055, + "loss": 0.1143, + "step": 40495 + }, + { + "epoch": 0.3515247263478616, + "grad_norm": 0.5859375, + "learning_rate": 0.0013354755785829208, + "loss": 0.1201, + "step": 40496 + }, + { + "epoch": 0.3515334068280657, + "grad_norm": 0.06396484375, + "learning_rate": 0.0013354468527529235, + "loss": 0.0718, + "step": 40497 + }, + { + "epoch": 0.3515420873082699, + "grad_norm": 0.53125, + "learning_rate": 0.0013354181266654454, + "loss": 0.0957, + "step": 40498 + }, + { + "epoch": 0.35155076778847405, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013353894003205174, + "loss": 0.1064, + "step": 40499 + }, + { + "epoch": 0.35155944826867824, + "grad_norm": 0.6171875, + "learning_rate": 0.001335360673718171, + "loss": 0.1045, + "step": 40500 + }, + { + "epoch": 0.3515681287488824, + "grad_norm": 0.203125, + "learning_rate": 0.0013353319468584377, + "loss": 0.1084, + "step": 40501 + }, + { + "epoch": 0.3515768092290866, + "grad_norm": 0.1318359375, + "learning_rate": 0.001335303219741349, + "loss": 0.1235, + "step": 40502 + }, + { + "epoch": 0.3515854897092907, + "grad_norm": 0.234375, + "learning_rate": 0.0013352744923669365, + "loss": 0.0957, + "step": 40503 + }, + { + "epoch": 0.3515941701894949, + "grad_norm": 0.54296875, + "learning_rate": 0.0013352457647352312, + "loss": 0.1074, + "step": 40504 + }, + { + "epoch": 0.35160285066969904, + "grad_norm": 0.470703125, + "learning_rate": 0.0013352170368462644, + "loss": 0.0903, + "step": 40505 + }, + { + "epoch": 0.35161153114990323, + "grad_norm": 0.2890625, + "learning_rate": 0.001335188308700068, + "loss": 0.1191, + "step": 40506 + }, + { + "epoch": 0.35162021163010737, + "grad_norm": 0.625, + "learning_rate": 0.0013351595802966734, + "loss": 0.0938, + "step": 40507 + }, + { + "epoch": 0.35162889211031156, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013351308516361114, + "loss": 0.1914, + "step": 40508 + }, + { + "epoch": 0.3516375725905157, + "grad_norm": 0.42578125, + "learning_rate": 0.001335102122718414, + "loss": 0.1318, + "step": 40509 + }, + { + "epoch": 0.3516462530707199, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013350733935436122, + "loss": 0.1016, + "step": 40510 + }, + { + "epoch": 0.35165493355092403, + "grad_norm": 0.142578125, + "learning_rate": 0.001335044664111738, + "loss": 0.0967, + "step": 40511 + }, + { + "epoch": 0.3516636140311282, + "grad_norm": 0.36328125, + "learning_rate": 0.0013350159344228225, + "loss": 0.0869, + "step": 40512 + }, + { + "epoch": 0.35167229451133236, + "grad_norm": 0.162109375, + "learning_rate": 0.0013349872044768972, + "loss": 0.0874, + "step": 40513 + }, + { + "epoch": 0.35168097499153655, + "grad_norm": 0.28515625, + "learning_rate": 0.0013349584742739931, + "loss": 0.0918, + "step": 40514 + }, + { + "epoch": 0.3516896554717407, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013349297438141421, + "loss": 0.0776, + "step": 40515 + }, + { + "epoch": 0.3516983359519449, + "grad_norm": 0.40234375, + "learning_rate": 0.0013349010130973755, + "loss": 0.1201, + "step": 40516 + }, + { + "epoch": 0.351707016432149, + "grad_norm": 0.14453125, + "learning_rate": 0.0013348722821237247, + "loss": 0.1162, + "step": 40517 + }, + { + "epoch": 0.3517156969123532, + "grad_norm": 0.400390625, + "learning_rate": 0.0013348435508932209, + "loss": 0.1035, + "step": 40518 + }, + { + "epoch": 0.35172437739255735, + "grad_norm": 0.17578125, + "learning_rate": 0.0013348148194058957, + "loss": 0.0986, + "step": 40519 + }, + { + "epoch": 0.35173305787276155, + "grad_norm": 0.455078125, + "learning_rate": 0.0013347860876617807, + "loss": 0.1069, + "step": 40520 + }, + { + "epoch": 0.3517417383529657, + "grad_norm": 0.416015625, + "learning_rate": 0.0013347573556609074, + "loss": 0.0933, + "step": 40521 + }, + { + "epoch": 0.3517504188331699, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013347286234033065, + "loss": 0.1182, + "step": 40522 + }, + { + "epoch": 0.351759099313374, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013346998908890103, + "loss": 0.1045, + "step": 40523 + }, + { + "epoch": 0.3517677797935782, + "grad_norm": 0.330078125, + "learning_rate": 0.0013346711581180496, + "loss": 0.125, + "step": 40524 + }, + { + "epoch": 0.35177646027378234, + "grad_norm": 0.392578125, + "learning_rate": 0.0013346424250904564, + "loss": 0.1182, + "step": 40525 + }, + { + "epoch": 0.35178514075398654, + "grad_norm": 0.255859375, + "learning_rate": 0.0013346136918062613, + "loss": 0.1279, + "step": 40526 + }, + { + "epoch": 0.3517938212341907, + "grad_norm": 0.84765625, + "learning_rate": 0.0013345849582654962, + "loss": 0.1099, + "step": 40527 + }, + { + "epoch": 0.35180250171439487, + "grad_norm": 0.255859375, + "learning_rate": 0.0013345562244681927, + "loss": 0.0889, + "step": 40528 + }, + { + "epoch": 0.351811182194599, + "grad_norm": 0.208984375, + "learning_rate": 0.0013345274904143823, + "loss": 0.0884, + "step": 40529 + }, + { + "epoch": 0.3518198626748032, + "grad_norm": 0.32421875, + "learning_rate": 0.0013344987561040957, + "loss": 0.0762, + "step": 40530 + }, + { + "epoch": 0.35182854315500733, + "grad_norm": 0.1904296875, + "learning_rate": 0.001334470021537365, + "loss": 0.0972, + "step": 40531 + }, + { + "epoch": 0.3518372236352115, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013344412867142212, + "loss": 0.1221, + "step": 40532 + }, + { + "epoch": 0.35184590411541566, + "grad_norm": 0.1318359375, + "learning_rate": 0.0013344125516346966, + "loss": 0.0933, + "step": 40533 + }, + { + "epoch": 0.35185458459561986, + "grad_norm": 0.26171875, + "learning_rate": 0.0013343838162988213, + "loss": 0.1104, + "step": 40534 + }, + { + "epoch": 0.351863265075824, + "grad_norm": 0.333984375, + "learning_rate": 0.0013343550807066273, + "loss": 0.1123, + "step": 40535 + }, + { + "epoch": 0.35187194555602813, + "grad_norm": 0.318359375, + "learning_rate": 0.0013343263448581464, + "loss": 0.1748, + "step": 40536 + }, + { + "epoch": 0.3518806260362323, + "grad_norm": 0.1435546875, + "learning_rate": 0.0013342976087534098, + "loss": 0.0894, + "step": 40537 + }, + { + "epoch": 0.35188930651643646, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013342688723924489, + "loss": 0.0889, + "step": 40538 + }, + { + "epoch": 0.35189798699664065, + "grad_norm": 0.35546875, + "learning_rate": 0.0013342401357752946, + "loss": 0.1201, + "step": 40539 + }, + { + "epoch": 0.3519066674768448, + "grad_norm": 0.146484375, + "learning_rate": 0.001334211398901979, + "loss": 0.1006, + "step": 40540 + }, + { + "epoch": 0.351915347957049, + "grad_norm": 0.1787109375, + "learning_rate": 0.0013341826617725336, + "loss": 0.1074, + "step": 40541 + }, + { + "epoch": 0.3519240284372531, + "grad_norm": 0.3984375, + "learning_rate": 0.0013341539243869895, + "loss": 0.1797, + "step": 40542 + }, + { + "epoch": 0.3519327089174573, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013341251867453783, + "loss": 0.1133, + "step": 40543 + }, + { + "epoch": 0.35194138939766145, + "grad_norm": 0.11865234375, + "learning_rate": 0.0013340964488477308, + "loss": 0.1104, + "step": 40544 + }, + { + "epoch": 0.35195006987786565, + "grad_norm": 0.083984375, + "learning_rate": 0.0013340677106940795, + "loss": 0.0986, + "step": 40545 + }, + { + "epoch": 0.3519587503580698, + "grad_norm": 0.1689453125, + "learning_rate": 0.001334038972284455, + "loss": 0.1069, + "step": 40546 + }, + { + "epoch": 0.351967430838274, + "grad_norm": 0.3046875, + "learning_rate": 0.0013340102336188888, + "loss": 0.127, + "step": 40547 + }, + { + "epoch": 0.3519761113184781, + "grad_norm": 0.107421875, + "learning_rate": 0.001333981494697413, + "loss": 0.0874, + "step": 40548 + }, + { + "epoch": 0.3519847917986823, + "grad_norm": 0.11376953125, + "learning_rate": 0.0013339527555200582, + "loss": 0.1147, + "step": 40549 + }, + { + "epoch": 0.35199347227888644, + "grad_norm": 0.12890625, + "learning_rate": 0.0013339240160868565, + "loss": 0.0869, + "step": 40550 + }, + { + "epoch": 0.35200215275909064, + "grad_norm": 0.4453125, + "learning_rate": 0.0013338952763978385, + "loss": 0.1084, + "step": 40551 + }, + { + "epoch": 0.3520108332392948, + "grad_norm": 0.1943359375, + "learning_rate": 0.0013338665364530365, + "loss": 0.0986, + "step": 40552 + }, + { + "epoch": 0.35201951371949897, + "grad_norm": 0.462890625, + "learning_rate": 0.0013338377962524815, + "loss": 0.1162, + "step": 40553 + }, + { + "epoch": 0.3520281941997031, + "grad_norm": 0.177734375, + "learning_rate": 0.001333809055796205, + "loss": 0.1416, + "step": 40554 + }, + { + "epoch": 0.3520368746799073, + "grad_norm": 0.212890625, + "learning_rate": 0.0013337803150842386, + "loss": 0.0957, + "step": 40555 + }, + { + "epoch": 0.35204555516011143, + "grad_norm": 0.431640625, + "learning_rate": 0.0013337515741166136, + "loss": 0.1377, + "step": 40556 + }, + { + "epoch": 0.3520542356403156, + "grad_norm": 0.337890625, + "learning_rate": 0.001333722832893361, + "loss": 0.1035, + "step": 40557 + }, + { + "epoch": 0.35206291612051976, + "grad_norm": 1.234375, + "learning_rate": 0.0013336940914145125, + "loss": 0.0854, + "step": 40558 + }, + { + "epoch": 0.35207159660072396, + "grad_norm": 0.10302734375, + "learning_rate": 0.0013336653496801001, + "loss": 0.1025, + "step": 40559 + }, + { + "epoch": 0.3520802770809281, + "grad_norm": 0.09228515625, + "learning_rate": 0.0013336366076901547, + "loss": 0.106, + "step": 40560 + }, + { + "epoch": 0.3520889575611323, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013336078654447075, + "loss": 0.1055, + "step": 40561 + }, + { + "epoch": 0.3520976380413364, + "grad_norm": 0.51953125, + "learning_rate": 0.0013335791229437907, + "loss": 0.1221, + "step": 40562 + }, + { + "epoch": 0.3521063185215406, + "grad_norm": 0.376953125, + "learning_rate": 0.001333550380187435, + "loss": 0.1157, + "step": 40563 + }, + { + "epoch": 0.35211499900174476, + "grad_norm": 0.7734375, + "learning_rate": 0.0013335216371756725, + "loss": 0.1445, + "step": 40564 + }, + { + "epoch": 0.35212367948194895, + "grad_norm": 0.53125, + "learning_rate": 0.0013334928939085337, + "loss": 0.1123, + "step": 40565 + }, + { + "epoch": 0.3521323599621531, + "grad_norm": 0.7578125, + "learning_rate": 0.001333464150386051, + "loss": 0.1025, + "step": 40566 + }, + { + "epoch": 0.3521410404423573, + "grad_norm": 0.384765625, + "learning_rate": 0.001333435406608255, + "loss": 0.103, + "step": 40567 + }, + { + "epoch": 0.3521497209225614, + "grad_norm": 0.302734375, + "learning_rate": 0.001333406662575178, + "loss": 0.0986, + "step": 40568 + }, + { + "epoch": 0.3521584014027656, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013333779182868506, + "loss": 0.0703, + "step": 40569 + }, + { + "epoch": 0.35216708188296975, + "grad_norm": 0.134765625, + "learning_rate": 0.0013333491737433048, + "loss": 0.0938, + "step": 40570 + }, + { + "epoch": 0.35217576236317394, + "grad_norm": 1.171875, + "learning_rate": 0.0013333204289445717, + "loss": 0.3965, + "step": 40571 + }, + { + "epoch": 0.3521844428433781, + "grad_norm": 0.451171875, + "learning_rate": 0.0013332916838906831, + "loss": 0.0854, + "step": 40572 + }, + { + "epoch": 0.35219312332358227, + "grad_norm": 0.1513671875, + "learning_rate": 0.0013332629385816703, + "loss": 0.1006, + "step": 40573 + }, + { + "epoch": 0.3522018038037864, + "grad_norm": 0.93359375, + "learning_rate": 0.0013332341930175644, + "loss": 0.1309, + "step": 40574 + }, + { + "epoch": 0.3522104842839906, + "grad_norm": 0.37109375, + "learning_rate": 0.0013332054471983975, + "loss": 0.1484, + "step": 40575 + }, + { + "epoch": 0.35221916476419474, + "grad_norm": 0.26953125, + "learning_rate": 0.0013331767011242003, + "loss": 0.0908, + "step": 40576 + }, + { + "epoch": 0.35222784524439893, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013331479547950044, + "loss": 0.0996, + "step": 40577 + }, + { + "epoch": 0.35223652572460307, + "grad_norm": 0.10400390625, + "learning_rate": 0.0013331192082108412, + "loss": 0.1055, + "step": 40578 + }, + { + "epoch": 0.35224520620480726, + "grad_norm": 0.640625, + "learning_rate": 0.001333090461371743, + "loss": 0.0986, + "step": 40579 + }, + { + "epoch": 0.3522538866850114, + "grad_norm": 0.53125, + "learning_rate": 0.0013330617142777402, + "loss": 0.084, + "step": 40580 + }, + { + "epoch": 0.3522625671652156, + "grad_norm": 0.48046875, + "learning_rate": 0.0013330329669288647, + "loss": 0.1064, + "step": 40581 + }, + { + "epoch": 0.3522712476454197, + "grad_norm": 0.0908203125, + "learning_rate": 0.001333004219325148, + "loss": 0.084, + "step": 40582 + }, + { + "epoch": 0.3522799281256239, + "grad_norm": 0.09033203125, + "learning_rate": 0.001332975471466621, + "loss": 0.104, + "step": 40583 + }, + { + "epoch": 0.35228860860582806, + "grad_norm": 0.333984375, + "learning_rate": 0.0013329467233533156, + "loss": 0.0933, + "step": 40584 + }, + { + "epoch": 0.35229728908603225, + "grad_norm": 0.3359375, + "learning_rate": 0.0013329179749852636, + "loss": 0.1006, + "step": 40585 + }, + { + "epoch": 0.3523059695662364, + "grad_norm": 0.181640625, + "learning_rate": 0.0013328892263624957, + "loss": 0.1055, + "step": 40586 + }, + { + "epoch": 0.3523146500464406, + "grad_norm": 0.53125, + "learning_rate": 0.0013328604774850434, + "loss": 0.125, + "step": 40587 + }, + { + "epoch": 0.3523233305266447, + "grad_norm": 0.365234375, + "learning_rate": 0.0013328317283529388, + "loss": 0.0806, + "step": 40588 + }, + { + "epoch": 0.3523320110068489, + "grad_norm": 0.7421875, + "learning_rate": 0.001332802978966213, + "loss": 0.1079, + "step": 40589 + }, + { + "epoch": 0.35234069148705305, + "grad_norm": 0.07763671875, + "learning_rate": 0.001332774229324897, + "loss": 0.0947, + "step": 40590 + }, + { + "epoch": 0.35234937196725724, + "grad_norm": 0.57421875, + "learning_rate": 0.0013327454794290225, + "loss": 0.1934, + "step": 40591 + }, + { + "epoch": 0.3523580524474614, + "grad_norm": 0.369140625, + "learning_rate": 0.0013327167292786213, + "loss": 0.0884, + "step": 40592 + }, + { + "epoch": 0.35236673292766557, + "grad_norm": 0.470703125, + "learning_rate": 0.0013326879788737246, + "loss": 0.0874, + "step": 40593 + }, + { + "epoch": 0.3523754134078697, + "grad_norm": 0.298828125, + "learning_rate": 0.0013326592282143642, + "loss": 0.0869, + "step": 40594 + }, + { + "epoch": 0.3523840938880739, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013326304773005704, + "loss": 0.063, + "step": 40595 + }, + { + "epoch": 0.35239277436827804, + "grad_norm": 0.1357421875, + "learning_rate": 0.0013326017261323757, + "loss": 0.0752, + "step": 40596 + }, + { + "epoch": 0.35240145484848223, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013325729747098113, + "loss": 0.0889, + "step": 40597 + }, + { + "epoch": 0.35241013532868637, + "grad_norm": 0.302734375, + "learning_rate": 0.0013325442230329083, + "loss": 0.0859, + "step": 40598 + }, + { + "epoch": 0.35241881580889056, + "grad_norm": 0.49609375, + "learning_rate": 0.001332515471101699, + "loss": 0.0718, + "step": 40599 + }, + { + "epoch": 0.3524274962890947, + "grad_norm": 0.486328125, + "learning_rate": 0.0013324867189162138, + "loss": 0.1367, + "step": 40600 + }, + { + "epoch": 0.3524361767692989, + "grad_norm": 0.10986328125, + "learning_rate": 0.0013324579664764848, + "loss": 0.0996, + "step": 40601 + }, + { + "epoch": 0.35244485724950303, + "grad_norm": 0.2734375, + "learning_rate": 0.0013324292137825433, + "loss": 0.0576, + "step": 40602 + }, + { + "epoch": 0.3524535377297072, + "grad_norm": 0.5, + "learning_rate": 0.0013324004608344206, + "loss": 0.0869, + "step": 40603 + }, + { + "epoch": 0.35246221820991136, + "grad_norm": 0.169921875, + "learning_rate": 0.0013323717076321484, + "loss": 0.0947, + "step": 40604 + }, + { + "epoch": 0.35247089869011555, + "grad_norm": 0.2041015625, + "learning_rate": 0.001332342954175758, + "loss": 0.0977, + "step": 40605 + }, + { + "epoch": 0.3524795791703197, + "grad_norm": 0.369140625, + "learning_rate": 0.0013323142004652807, + "loss": 0.1025, + "step": 40606 + }, + { + "epoch": 0.3524882596505239, + "grad_norm": 0.283203125, + "learning_rate": 0.0013322854465007482, + "loss": 0.0908, + "step": 40607 + }, + { + "epoch": 0.352496940130728, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013322566922821914, + "loss": 0.0742, + "step": 40608 + }, + { + "epoch": 0.3525056206109322, + "grad_norm": 0.091796875, + "learning_rate": 0.0013322279378096425, + "loss": 0.1182, + "step": 40609 + }, + { + "epoch": 0.35251430109113635, + "grad_norm": 0.1318359375, + "learning_rate": 0.001332199183083133, + "loss": 0.0854, + "step": 40610 + }, + { + "epoch": 0.35252298157134054, + "grad_norm": 0.291015625, + "learning_rate": 0.0013321704281026936, + "loss": 0.1147, + "step": 40611 + }, + { + "epoch": 0.3525316620515447, + "grad_norm": 0.455078125, + "learning_rate": 0.0013321416728683559, + "loss": 0.1357, + "step": 40612 + }, + { + "epoch": 0.3525403425317489, + "grad_norm": 0.1474609375, + "learning_rate": 0.0013321129173801515, + "loss": 0.1533, + "step": 40613 + }, + { + "epoch": 0.352549023011953, + "grad_norm": 0.1328125, + "learning_rate": 0.0013320841616381125, + "loss": 0.1113, + "step": 40614 + }, + { + "epoch": 0.3525577034921572, + "grad_norm": 0.447265625, + "learning_rate": 0.0013320554056422693, + "loss": 0.083, + "step": 40615 + }, + { + "epoch": 0.35256638397236134, + "grad_norm": 0.11474609375, + "learning_rate": 0.0013320266493926538, + "loss": 0.083, + "step": 40616 + }, + { + "epoch": 0.35257506445256553, + "grad_norm": 0.14453125, + "learning_rate": 0.0013319978928892976, + "loss": 0.1367, + "step": 40617 + }, + { + "epoch": 0.35258374493276967, + "grad_norm": 0.578125, + "learning_rate": 0.001331969136132232, + "loss": 0.1025, + "step": 40618 + }, + { + "epoch": 0.35259242541297386, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013319403791214883, + "loss": 0.1299, + "step": 40619 + }, + { + "epoch": 0.352601105893178, + "grad_norm": 0.41015625, + "learning_rate": 0.0013319116218570984, + "loss": 0.0986, + "step": 40620 + }, + { + "epoch": 0.3526097863733822, + "grad_norm": 0.322265625, + "learning_rate": 0.001331882864339093, + "loss": 0.1035, + "step": 40621 + }, + { + "epoch": 0.35261846685358633, + "grad_norm": 0.5234375, + "learning_rate": 0.0013318541065675043, + "loss": 0.1045, + "step": 40622 + }, + { + "epoch": 0.3526271473337905, + "grad_norm": 0.234375, + "learning_rate": 0.0013318253485423632, + "loss": 0.1133, + "step": 40623 + }, + { + "epoch": 0.35263582781399466, + "grad_norm": 0.4375, + "learning_rate": 0.0013317965902637013, + "loss": 0.1328, + "step": 40624 + }, + { + "epoch": 0.35264450829419886, + "grad_norm": 0.40625, + "learning_rate": 0.0013317678317315506, + "loss": 0.0981, + "step": 40625 + }, + { + "epoch": 0.352653188774403, + "grad_norm": 0.103515625, + "learning_rate": 0.0013317390729459416, + "loss": 0.166, + "step": 40626 + }, + { + "epoch": 0.3526618692546072, + "grad_norm": 0.484375, + "learning_rate": 0.0013317103139069064, + "loss": 0.1055, + "step": 40627 + }, + { + "epoch": 0.3526705497348113, + "grad_norm": 0.318359375, + "learning_rate": 0.0013316815546144762, + "loss": 0.1074, + "step": 40628 + }, + { + "epoch": 0.3526792302150155, + "grad_norm": 0.0771484375, + "learning_rate": 0.001331652795068683, + "loss": 0.1191, + "step": 40629 + }, + { + "epoch": 0.35268791069521965, + "grad_norm": 0.18359375, + "learning_rate": 0.0013316240352695572, + "loss": 0.1035, + "step": 40630 + }, + { + "epoch": 0.35269659117542385, + "grad_norm": 0.146484375, + "learning_rate": 0.0013315952752171313, + "loss": 0.1104, + "step": 40631 + }, + { + "epoch": 0.352705271655628, + "grad_norm": 0.33984375, + "learning_rate": 0.001331566514911436, + "loss": 0.1191, + "step": 40632 + }, + { + "epoch": 0.3527139521358322, + "grad_norm": 0.9140625, + "learning_rate": 0.001331537754352503, + "loss": 0.1348, + "step": 40633 + }, + { + "epoch": 0.3527226326160363, + "grad_norm": 0.56640625, + "learning_rate": 0.0013315089935403638, + "loss": 0.1143, + "step": 40634 + }, + { + "epoch": 0.3527313130962405, + "grad_norm": 0.103515625, + "learning_rate": 0.00133148023247505, + "loss": 0.1309, + "step": 40635 + }, + { + "epoch": 0.35273999357644464, + "grad_norm": 0.8984375, + "learning_rate": 0.001331451471156593, + "loss": 0.2793, + "step": 40636 + }, + { + "epoch": 0.35274867405664884, + "grad_norm": 0.2255859375, + "learning_rate": 0.0013314227095850238, + "loss": 0.1138, + "step": 40637 + }, + { + "epoch": 0.352757354536853, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013313939477603745, + "loss": 0.0737, + "step": 40638 + }, + { + "epoch": 0.35276603501705717, + "grad_norm": 0.162109375, + "learning_rate": 0.001331365185682676, + "loss": 0.1021, + "step": 40639 + }, + { + "epoch": 0.3527747154972613, + "grad_norm": 0.177734375, + "learning_rate": 0.0013313364233519603, + "loss": 0.0947, + "step": 40640 + }, + { + "epoch": 0.3527833959774655, + "grad_norm": 0.45703125, + "learning_rate": 0.0013313076607682584, + "loss": 0.125, + "step": 40641 + }, + { + "epoch": 0.35279207645766963, + "grad_norm": 0.09033203125, + "learning_rate": 0.0013312788979316019, + "loss": 0.062, + "step": 40642 + }, + { + "epoch": 0.3528007569378738, + "grad_norm": 0.10888671875, + "learning_rate": 0.001331250134842022, + "loss": 0.0977, + "step": 40643 + }, + { + "epoch": 0.35280943741807796, + "grad_norm": 0.244140625, + "learning_rate": 0.0013312213714995508, + "loss": 0.1035, + "step": 40644 + }, + { + "epoch": 0.35281811789828216, + "grad_norm": 0.107421875, + "learning_rate": 0.001331192607904219, + "loss": 0.0996, + "step": 40645 + }, + { + "epoch": 0.3528267983784863, + "grad_norm": 1.2578125, + "learning_rate": 0.0013311638440560592, + "loss": 0.1162, + "step": 40646 + }, + { + "epoch": 0.3528354788586905, + "grad_norm": 0.28125, + "learning_rate": 0.0013311350799551014, + "loss": 0.1143, + "step": 40647 + }, + { + "epoch": 0.3528441593388946, + "grad_norm": 0.365234375, + "learning_rate": 0.001331106315601378, + "loss": 0.0898, + "step": 40648 + }, + { + "epoch": 0.3528528398190988, + "grad_norm": 0.267578125, + "learning_rate": 0.0013310775509949204, + "loss": 0.1104, + "step": 40649 + }, + { + "epoch": 0.35286152029930296, + "grad_norm": 0.271484375, + "learning_rate": 0.0013310487861357596, + "loss": 0.1309, + "step": 40650 + }, + { + "epoch": 0.35287020077950715, + "grad_norm": 0.1083984375, + "learning_rate": 0.0013310200210239274, + "loss": 0.0938, + "step": 40651 + }, + { + "epoch": 0.3528788812597113, + "grad_norm": 0.2060546875, + "learning_rate": 0.001330991255659455, + "loss": 0.0928, + "step": 40652 + }, + { + "epoch": 0.3528875617399155, + "grad_norm": 0.51953125, + "learning_rate": 0.0013309624900423743, + "loss": 0.0728, + "step": 40653 + }, + { + "epoch": 0.3528962422201196, + "grad_norm": 0.57421875, + "learning_rate": 0.0013309337241727164, + "loss": 0.125, + "step": 40654 + }, + { + "epoch": 0.3529049227003238, + "grad_norm": 0.2890625, + "learning_rate": 0.0013309049580505126, + "loss": 0.0786, + "step": 40655 + }, + { + "epoch": 0.35291360318052795, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013308761916757947, + "loss": 0.0923, + "step": 40656 + }, + { + "epoch": 0.35292228366073214, + "grad_norm": 0.4921875, + "learning_rate": 0.0013308474250485942, + "loss": 0.1094, + "step": 40657 + }, + { + "epoch": 0.3529309641409363, + "grad_norm": 0.265625, + "learning_rate": 0.0013308186581689428, + "loss": 0.1143, + "step": 40658 + }, + { + "epoch": 0.35293964462114047, + "grad_norm": 0.09423828125, + "learning_rate": 0.001330789891036871, + "loss": 0.1055, + "step": 40659 + }, + { + "epoch": 0.3529483251013446, + "grad_norm": 0.2021484375, + "learning_rate": 0.0013307611236524108, + "loss": 0.0957, + "step": 40660 + }, + { + "epoch": 0.35295700558154874, + "grad_norm": 0.57421875, + "learning_rate": 0.0013307323560155942, + "loss": 0.1226, + "step": 40661 + }, + { + "epoch": 0.35296568606175294, + "grad_norm": 0.12890625, + "learning_rate": 0.001330703588126452, + "loss": 0.0625, + "step": 40662 + }, + { + "epoch": 0.3529743665419571, + "grad_norm": 0.71875, + "learning_rate": 0.0013306748199850157, + "loss": 0.1221, + "step": 40663 + }, + { + "epoch": 0.35298304702216127, + "grad_norm": 0.2333984375, + "learning_rate": 0.0013306460515913167, + "loss": 0.0996, + "step": 40664 + }, + { + "epoch": 0.3529917275023654, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013306172829453868, + "loss": 0.1006, + "step": 40665 + }, + { + "epoch": 0.3530004079825696, + "grad_norm": 0.388671875, + "learning_rate": 0.0013305885140472574, + "loss": 0.0752, + "step": 40666 + }, + { + "epoch": 0.35300908846277373, + "grad_norm": 0.38671875, + "learning_rate": 0.00133055974489696, + "loss": 0.0703, + "step": 40667 + }, + { + "epoch": 0.3530177689429779, + "grad_norm": 0.4140625, + "learning_rate": 0.0013305309754945258, + "loss": 0.0786, + "step": 40668 + }, + { + "epoch": 0.35302644942318206, + "grad_norm": 0.423828125, + "learning_rate": 0.001330502205839986, + "loss": 0.1016, + "step": 40669 + }, + { + "epoch": 0.35303512990338626, + "grad_norm": 0.11279296875, + "learning_rate": 0.001330473435933373, + "loss": 0.0957, + "step": 40670 + }, + { + "epoch": 0.3530438103835904, + "grad_norm": 0.18359375, + "learning_rate": 0.0013304446657747175, + "loss": 0.0786, + "step": 40671 + }, + { + "epoch": 0.3530524908637946, + "grad_norm": 0.4140625, + "learning_rate": 0.0013304158953640513, + "loss": 0.084, + "step": 40672 + }, + { + "epoch": 0.3530611713439987, + "grad_norm": 0.2001953125, + "learning_rate": 0.0013303871247014054, + "loss": 0.0762, + "step": 40673 + }, + { + "epoch": 0.3530698518242029, + "grad_norm": 0.138671875, + "learning_rate": 0.0013303583537868117, + "loss": 0.1006, + "step": 40674 + }, + { + "epoch": 0.35307853230440706, + "grad_norm": 0.384765625, + "learning_rate": 0.0013303295826203017, + "loss": 0.1147, + "step": 40675 + }, + { + "epoch": 0.35308721278461125, + "grad_norm": 0.447265625, + "learning_rate": 0.0013303008112019068, + "loss": 0.1016, + "step": 40676 + }, + { + "epoch": 0.3530958932648154, + "grad_norm": 0.17578125, + "learning_rate": 0.0013302720395316585, + "loss": 0.0815, + "step": 40677 + }, + { + "epoch": 0.3531045737450196, + "grad_norm": 0.98828125, + "learning_rate": 0.0013302432676095878, + "loss": 0.0708, + "step": 40678 + }, + { + "epoch": 0.3531132542252237, + "grad_norm": 0.138671875, + "learning_rate": 0.0013302144954357267, + "loss": 0.1162, + "step": 40679 + }, + { + "epoch": 0.3531219347054279, + "grad_norm": 0.6484375, + "learning_rate": 0.0013301857230101067, + "loss": 0.1201, + "step": 40680 + }, + { + "epoch": 0.35313061518563205, + "grad_norm": 0.255859375, + "learning_rate": 0.0013301569503327585, + "loss": 0.1094, + "step": 40681 + }, + { + "epoch": 0.35313929566583624, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013301281774037146, + "loss": 0.106, + "step": 40682 + }, + { + "epoch": 0.3531479761460404, + "grad_norm": 0.400390625, + "learning_rate": 0.0013300994042230057, + "loss": 0.1543, + "step": 40683 + }, + { + "epoch": 0.35315665662624457, + "grad_norm": 0.10400390625, + "learning_rate": 0.0013300706307906639, + "loss": 0.085, + "step": 40684 + }, + { + "epoch": 0.3531653371064487, + "grad_norm": 0.44140625, + "learning_rate": 0.00133004185710672, + "loss": 0.0776, + "step": 40685 + }, + { + "epoch": 0.3531740175866529, + "grad_norm": 0.85546875, + "learning_rate": 0.0013300130831712057, + "loss": 0.1367, + "step": 40686 + }, + { + "epoch": 0.35318269806685704, + "grad_norm": 0.1962890625, + "learning_rate": 0.0013299843089841527, + "loss": 0.1621, + "step": 40687 + }, + { + "epoch": 0.35319137854706123, + "grad_norm": 0.0888671875, + "learning_rate": 0.0013299555345455925, + "loss": 0.0801, + "step": 40688 + }, + { + "epoch": 0.35320005902726537, + "grad_norm": 0.302734375, + "learning_rate": 0.0013299267598555562, + "loss": 0.1089, + "step": 40689 + }, + { + "epoch": 0.35320873950746956, + "grad_norm": 0.185546875, + "learning_rate": 0.0013298979849140755, + "loss": 0.1001, + "step": 40690 + }, + { + "epoch": 0.3532174199876737, + "grad_norm": 0.482421875, + "learning_rate": 0.0013298692097211818, + "loss": 0.0981, + "step": 40691 + }, + { + "epoch": 0.3532261004678779, + "grad_norm": 0.10986328125, + "learning_rate": 0.0013298404342769064, + "loss": 0.105, + "step": 40692 + }, + { + "epoch": 0.35323478094808203, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013298116585812813, + "loss": 0.0879, + "step": 40693 + }, + { + "epoch": 0.3532434614282862, + "grad_norm": 0.1044921875, + "learning_rate": 0.0013297828826343374, + "loss": 0.0742, + "step": 40694 + }, + { + "epoch": 0.35325214190849036, + "grad_norm": 0.52734375, + "learning_rate": 0.0013297541064361063, + "loss": 0.0908, + "step": 40695 + }, + { + "epoch": 0.35326082238869455, + "grad_norm": 0.1083984375, + "learning_rate": 0.00132972532998662, + "loss": 0.105, + "step": 40696 + }, + { + "epoch": 0.3532695028688987, + "grad_norm": 0.15234375, + "learning_rate": 0.001329696553285909, + "loss": 0.1084, + "step": 40697 + }, + { + "epoch": 0.3532781833491029, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013296677763340056, + "loss": 0.0703, + "step": 40698 + }, + { + "epoch": 0.353286863829307, + "grad_norm": 0.1708984375, + "learning_rate": 0.0013296389991309407, + "loss": 0.1104, + "step": 40699 + }, + { + "epoch": 0.3532955443095112, + "grad_norm": 0.205078125, + "learning_rate": 0.0013296102216767466, + "loss": 0.1143, + "step": 40700 + }, + { + "epoch": 0.35330422478971535, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013295814439714538, + "loss": 0.0723, + "step": 40701 + }, + { + "epoch": 0.35331290526991954, + "grad_norm": 2.4375, + "learning_rate": 0.001329552666015094, + "loss": 0.2119, + "step": 40702 + }, + { + "epoch": 0.3533215857501237, + "grad_norm": 0.140625, + "learning_rate": 0.001329523887807699, + "loss": 0.1055, + "step": 40703 + }, + { + "epoch": 0.35333026623032787, + "grad_norm": 0.43359375, + "learning_rate": 0.0013294951093493004, + "loss": 0.0869, + "step": 40704 + }, + { + "epoch": 0.353338946710532, + "grad_norm": 0.125, + "learning_rate": 0.0013294663306399291, + "loss": 0.0815, + "step": 40705 + }, + { + "epoch": 0.3533476271907362, + "grad_norm": 0.2890625, + "learning_rate": 0.001329437551679617, + "loss": 0.0874, + "step": 40706 + }, + { + "epoch": 0.35335630767094034, + "grad_norm": 0.142578125, + "learning_rate": 0.0013294087724683957, + "loss": 0.1367, + "step": 40707 + }, + { + "epoch": 0.35336498815114453, + "grad_norm": 0.330078125, + "learning_rate": 0.001329379993006296, + "loss": 0.0942, + "step": 40708 + }, + { + "epoch": 0.35337366863134867, + "grad_norm": 0.28515625, + "learning_rate": 0.0013293512132933499, + "loss": 0.1396, + "step": 40709 + }, + { + "epoch": 0.35338234911155286, + "grad_norm": 0.44140625, + "learning_rate": 0.0013293224333295888, + "loss": 0.1143, + "step": 40710 + }, + { + "epoch": 0.353391029591757, + "grad_norm": 0.2412109375, + "learning_rate": 0.0013292936531150439, + "loss": 0.1035, + "step": 40711 + }, + { + "epoch": 0.3533997100719612, + "grad_norm": 0.142578125, + "learning_rate": 0.0013292648726497469, + "loss": 0.124, + "step": 40712 + }, + { + "epoch": 0.35340839055216533, + "grad_norm": 0.271484375, + "learning_rate": 0.0013292360919337296, + "loss": 0.0869, + "step": 40713 + }, + { + "epoch": 0.3534170710323695, + "grad_norm": 0.1962890625, + "learning_rate": 0.0013292073109670234, + "loss": 0.0859, + "step": 40714 + }, + { + "epoch": 0.35342575151257366, + "grad_norm": 0.162109375, + "learning_rate": 0.0013291785297496592, + "loss": 0.0879, + "step": 40715 + }, + { + "epoch": 0.35343443199277785, + "grad_norm": 0.3828125, + "learning_rate": 0.0013291497482816686, + "loss": 0.0801, + "step": 40716 + }, + { + "epoch": 0.353443112472982, + "grad_norm": 0.2431640625, + "learning_rate": 0.0013291209665630835, + "loss": 0.1416, + "step": 40717 + }, + { + "epoch": 0.3534517929531862, + "grad_norm": 0.216796875, + "learning_rate": 0.0013290921845939352, + "loss": 0.2256, + "step": 40718 + }, + { + "epoch": 0.3534604734333903, + "grad_norm": 0.138671875, + "learning_rate": 0.001329063402374255, + "loss": 0.0791, + "step": 40719 + }, + { + "epoch": 0.3534691539135945, + "grad_norm": 0.1357421875, + "learning_rate": 0.0013290346199040747, + "loss": 0.082, + "step": 40720 + }, + { + "epoch": 0.35347783439379865, + "grad_norm": 0.30078125, + "learning_rate": 0.0013290058371834253, + "loss": 0.1113, + "step": 40721 + }, + { + "epoch": 0.35348651487400284, + "grad_norm": 0.126953125, + "learning_rate": 0.0013289770542123385, + "loss": 0.1064, + "step": 40722 + }, + { + "epoch": 0.353495195354207, + "grad_norm": 0.255859375, + "learning_rate": 0.001328948270990846, + "loss": 0.125, + "step": 40723 + }, + { + "epoch": 0.3535038758344112, + "grad_norm": 0.2265625, + "learning_rate": 0.0013289194875189794, + "loss": 0.084, + "step": 40724 + }, + { + "epoch": 0.3535125563146153, + "grad_norm": 0.1044921875, + "learning_rate": 0.0013288907037967695, + "loss": 0.124, + "step": 40725 + }, + { + "epoch": 0.3535212367948195, + "grad_norm": 0.5390625, + "learning_rate": 0.0013288619198242482, + "loss": 0.1357, + "step": 40726 + }, + { + "epoch": 0.35352991727502364, + "grad_norm": 0.076171875, + "learning_rate": 0.0013288331356014471, + "loss": 0.0952, + "step": 40727 + }, + { + "epoch": 0.35353859775522783, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013288043511283977, + "loss": 0.1143, + "step": 40728 + }, + { + "epoch": 0.35354727823543197, + "grad_norm": 0.203125, + "learning_rate": 0.0013287755664051308, + "loss": 0.1543, + "step": 40729 + }, + { + "epoch": 0.35355595871563616, + "grad_norm": 0.09765625, + "learning_rate": 0.0013287467814316787, + "loss": 0.1309, + "step": 40730 + }, + { + "epoch": 0.3535646391958403, + "grad_norm": 0.103515625, + "learning_rate": 0.0013287179962080725, + "loss": 0.0762, + "step": 40731 + }, + { + "epoch": 0.3535733196760445, + "grad_norm": 0.11328125, + "learning_rate": 0.0013286892107343435, + "loss": 0.1045, + "step": 40732 + }, + { + "epoch": 0.35358200015624863, + "grad_norm": 0.1142578125, + "learning_rate": 0.001328660425010524, + "loss": 0.1299, + "step": 40733 + }, + { + "epoch": 0.3535906806364528, + "grad_norm": 0.8046875, + "learning_rate": 0.0013286316390366442, + "loss": 0.1191, + "step": 40734 + }, + { + "epoch": 0.35359936111665696, + "grad_norm": 0.275390625, + "learning_rate": 0.0013286028528127368, + "loss": 0.1289, + "step": 40735 + }, + { + "epoch": 0.35360804159686116, + "grad_norm": 0.267578125, + "learning_rate": 0.0013285740663388323, + "loss": 0.0806, + "step": 40736 + }, + { + "epoch": 0.3536167220770653, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013285452796149633, + "loss": 0.0771, + "step": 40737 + }, + { + "epoch": 0.3536254025572695, + "grad_norm": 0.82421875, + "learning_rate": 0.00132851649264116, + "loss": 0.1211, + "step": 40738 + }, + { + "epoch": 0.3536340830374736, + "grad_norm": 0.1748046875, + "learning_rate": 0.0013284877054174546, + "loss": 0.0859, + "step": 40739 + }, + { + "epoch": 0.3536427635176778, + "grad_norm": 0.123046875, + "learning_rate": 0.0013284589179438787, + "loss": 0.0874, + "step": 40740 + }, + { + "epoch": 0.35365144399788195, + "grad_norm": 0.271484375, + "learning_rate": 0.0013284301302204633, + "loss": 0.0933, + "step": 40741 + }, + { + "epoch": 0.35366012447808615, + "grad_norm": 0.1328125, + "learning_rate": 0.0013284013422472404, + "loss": 0.1011, + "step": 40742 + }, + { + "epoch": 0.3536688049582903, + "grad_norm": 0.45703125, + "learning_rate": 0.0013283725540242412, + "loss": 0.0889, + "step": 40743 + }, + { + "epoch": 0.3536774854384945, + "grad_norm": 0.49609375, + "learning_rate": 0.001328343765551497, + "loss": 0.1621, + "step": 40744 + }, + { + "epoch": 0.3536861659186986, + "grad_norm": 0.1015625, + "learning_rate": 0.0013283149768290397, + "loss": 0.1216, + "step": 40745 + }, + { + "epoch": 0.3536948463989028, + "grad_norm": 0.416015625, + "learning_rate": 0.0013282861878569003, + "loss": 0.126, + "step": 40746 + }, + { + "epoch": 0.35370352687910694, + "grad_norm": 0.25, + "learning_rate": 0.0013282573986351108, + "loss": 0.127, + "step": 40747 + }, + { + "epoch": 0.35371220735931114, + "grad_norm": 0.1044921875, + "learning_rate": 0.0013282286091637024, + "loss": 0.0898, + "step": 40748 + }, + { + "epoch": 0.3537208878395153, + "grad_norm": 0.17578125, + "learning_rate": 0.0013281998194427066, + "loss": 0.127, + "step": 40749 + }, + { + "epoch": 0.35372956831971947, + "grad_norm": 0.322265625, + "learning_rate": 0.0013281710294721547, + "loss": 0.1211, + "step": 40750 + }, + { + "epoch": 0.3537382487999236, + "grad_norm": 0.158203125, + "learning_rate": 0.0013281422392520785, + "loss": 0.0703, + "step": 40751 + }, + { + "epoch": 0.3537469292801278, + "grad_norm": 0.12109375, + "learning_rate": 0.0013281134487825095, + "loss": 0.0928, + "step": 40752 + }, + { + "epoch": 0.35375560976033193, + "grad_norm": 0.62890625, + "learning_rate": 0.0013280846580634791, + "loss": 0.1035, + "step": 40753 + }, + { + "epoch": 0.35376429024053613, + "grad_norm": 0.1669921875, + "learning_rate": 0.001328055867095019, + "loss": 0.062, + "step": 40754 + }, + { + "epoch": 0.35377297072074027, + "grad_norm": 0.259765625, + "learning_rate": 0.0013280270758771599, + "loss": 0.1523, + "step": 40755 + }, + { + "epoch": 0.35378165120094446, + "grad_norm": 0.69140625, + "learning_rate": 0.0013279982844099341, + "loss": 0.1055, + "step": 40756 + }, + { + "epoch": 0.3537903316811486, + "grad_norm": 0.10009765625, + "learning_rate": 0.0013279694926933727, + "loss": 0.1089, + "step": 40757 + }, + { + "epoch": 0.3537990121613528, + "grad_norm": 0.205078125, + "learning_rate": 0.0013279407007275074, + "loss": 0.1138, + "step": 40758 + }, + { + "epoch": 0.3538076926415569, + "grad_norm": 0.1455078125, + "learning_rate": 0.0013279119085123693, + "loss": 0.0874, + "step": 40759 + }, + { + "epoch": 0.3538163731217611, + "grad_norm": 0.2275390625, + "learning_rate": 0.00132788311604799, + "loss": 0.1089, + "step": 40760 + }, + { + "epoch": 0.35382505360196526, + "grad_norm": 0.15625, + "learning_rate": 0.0013278543233344018, + "loss": 0.127, + "step": 40761 + }, + { + "epoch": 0.35383373408216945, + "grad_norm": 0.32421875, + "learning_rate": 0.0013278255303716352, + "loss": 0.1367, + "step": 40762 + }, + { + "epoch": 0.3538424145623736, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013277967371597222, + "loss": 0.1162, + "step": 40763 + }, + { + "epoch": 0.3538510950425778, + "grad_norm": 1.2421875, + "learning_rate": 0.0013277679436986937, + "loss": 0.1943, + "step": 40764 + }, + { + "epoch": 0.3538597755227819, + "grad_norm": 0.578125, + "learning_rate": 0.001327739149988582, + "loss": 0.1104, + "step": 40765 + }, + { + "epoch": 0.3538684560029861, + "grad_norm": 0.2734375, + "learning_rate": 0.0013277103560294181, + "loss": 0.0732, + "step": 40766 + }, + { + "epoch": 0.35387713648319025, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013276815618212335, + "loss": 0.0874, + "step": 40767 + }, + { + "epoch": 0.35388581696339444, + "grad_norm": 0.34765625, + "learning_rate": 0.0013276527673640598, + "loss": 0.0752, + "step": 40768 + }, + { + "epoch": 0.3538944974435986, + "grad_norm": 0.19140625, + "learning_rate": 0.0013276239726579282, + "loss": 0.0874, + "step": 40769 + }, + { + "epoch": 0.35390317792380277, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013275951777028709, + "loss": 0.0947, + "step": 40770 + }, + { + "epoch": 0.3539118584040069, + "grad_norm": 0.490234375, + "learning_rate": 0.0013275663824989187, + "loss": 0.1045, + "step": 40771 + }, + { + "epoch": 0.3539205388842111, + "grad_norm": 0.173828125, + "learning_rate": 0.0013275375870461033, + "loss": 0.1133, + "step": 40772 + }, + { + "epoch": 0.35392921936441524, + "grad_norm": 0.49609375, + "learning_rate": 0.0013275087913444561, + "loss": 0.1074, + "step": 40773 + }, + { + "epoch": 0.35393789984461943, + "grad_norm": 0.494140625, + "learning_rate": 0.001327479995394009, + "loss": 0.1357, + "step": 40774 + }, + { + "epoch": 0.35394658032482357, + "grad_norm": 0.1318359375, + "learning_rate": 0.0013274511991947931, + "loss": 0.105, + "step": 40775 + }, + { + "epoch": 0.35395526080502776, + "grad_norm": 0.365234375, + "learning_rate": 0.00132742240274684, + "loss": 0.0986, + "step": 40776 + }, + { + "epoch": 0.3539639412852319, + "grad_norm": 0.1826171875, + "learning_rate": 0.001327393606050181, + "loss": 0.0972, + "step": 40777 + }, + { + "epoch": 0.3539726217654361, + "grad_norm": 0.1533203125, + "learning_rate": 0.001327364809104848, + "loss": 0.0742, + "step": 40778 + }, + { + "epoch": 0.35398130224564023, + "grad_norm": 0.263671875, + "learning_rate": 0.001327336011910872, + "loss": 0.1416, + "step": 40779 + }, + { + "epoch": 0.3539899827258444, + "grad_norm": 0.099609375, + "learning_rate": 0.0013273072144682852, + "loss": 0.1016, + "step": 40780 + }, + { + "epoch": 0.35399866320604856, + "grad_norm": 0.310546875, + "learning_rate": 0.001327278416777118, + "loss": 0.1104, + "step": 40781 + }, + { + "epoch": 0.35400734368625275, + "grad_norm": 0.52734375, + "learning_rate": 0.0013272496188374027, + "loss": 0.1436, + "step": 40782 + }, + { + "epoch": 0.3540160241664569, + "grad_norm": 0.8046875, + "learning_rate": 0.001327220820649171, + "loss": 0.1309, + "step": 40783 + }, + { + "epoch": 0.354024704646661, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013271920222124541, + "loss": 0.0859, + "step": 40784 + }, + { + "epoch": 0.3540333851268652, + "grad_norm": 0.65234375, + "learning_rate": 0.001327163223527283, + "loss": 0.0874, + "step": 40785 + }, + { + "epoch": 0.35404206560706936, + "grad_norm": 0.19921875, + "learning_rate": 0.0013271344245936898, + "loss": 0.1162, + "step": 40786 + }, + { + "epoch": 0.35405074608727355, + "grad_norm": 0.1435546875, + "learning_rate": 0.0013271056254117058, + "loss": 0.084, + "step": 40787 + }, + { + "epoch": 0.3540594265674777, + "grad_norm": 0.248046875, + "learning_rate": 0.0013270768259813625, + "loss": 0.165, + "step": 40788 + }, + { + "epoch": 0.3540681070476819, + "grad_norm": 0.11865234375, + "learning_rate": 0.001327048026302691, + "loss": 0.0898, + "step": 40789 + }, + { + "epoch": 0.354076787527886, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013270192263757238, + "loss": 0.0737, + "step": 40790 + }, + { + "epoch": 0.3540854680080902, + "grad_norm": 0.49609375, + "learning_rate": 0.0013269904262004915, + "loss": 0.0986, + "step": 40791 + }, + { + "epoch": 0.35409414848829435, + "grad_norm": 0.345703125, + "learning_rate": 0.0013269616257770259, + "loss": 0.1074, + "step": 40792 + }, + { + "epoch": 0.35410282896849854, + "grad_norm": 0.33203125, + "learning_rate": 0.0013269328251053587, + "loss": 0.0938, + "step": 40793 + }, + { + "epoch": 0.3541115094487027, + "grad_norm": 0.79296875, + "learning_rate": 0.0013269040241855208, + "loss": 0.2158, + "step": 40794 + }, + { + "epoch": 0.35412018992890687, + "grad_norm": 1.0234375, + "learning_rate": 0.0013268752230175442, + "loss": 0.1279, + "step": 40795 + }, + { + "epoch": 0.354128870409111, + "grad_norm": 0.439453125, + "learning_rate": 0.0013268464216014604, + "loss": 0.1348, + "step": 40796 + }, + { + "epoch": 0.3541375508893152, + "grad_norm": 0.17578125, + "learning_rate": 0.0013268176199373006, + "loss": 0.084, + "step": 40797 + }, + { + "epoch": 0.35414623136951934, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013267888180250966, + "loss": 0.1289, + "step": 40798 + }, + { + "epoch": 0.35415491184972353, + "grad_norm": 0.255859375, + "learning_rate": 0.0013267600158648797, + "loss": 0.1123, + "step": 40799 + }, + { + "epoch": 0.35416359232992767, + "grad_norm": 0.224609375, + "learning_rate": 0.0013267312134566814, + "loss": 0.127, + "step": 40800 + }, + { + "epoch": 0.35417227281013186, + "grad_norm": 0.1689453125, + "learning_rate": 0.0013267024108005335, + "loss": 0.1523, + "step": 40801 + }, + { + "epoch": 0.354180953290336, + "grad_norm": 0.07958984375, + "learning_rate": 0.0013266736078964669, + "loss": 0.0869, + "step": 40802 + }, + { + "epoch": 0.3541896337705402, + "grad_norm": 0.1796875, + "learning_rate": 0.0013266448047445135, + "loss": 0.1377, + "step": 40803 + }, + { + "epoch": 0.35419831425074433, + "grad_norm": 0.1640625, + "learning_rate": 0.001326616001344705, + "loss": 0.062, + "step": 40804 + }, + { + "epoch": 0.3542069947309485, + "grad_norm": 0.09814453125, + "learning_rate": 0.0013265871976970722, + "loss": 0.0903, + "step": 40805 + }, + { + "epoch": 0.35421567521115266, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013265583938016474, + "loss": 0.0918, + "step": 40806 + }, + { + "epoch": 0.35422435569135685, + "grad_norm": 0.10205078125, + "learning_rate": 0.0013265295896584617, + "loss": 0.0806, + "step": 40807 + }, + { + "epoch": 0.354233036171561, + "grad_norm": 0.2890625, + "learning_rate": 0.0013265007852675463, + "loss": 0.0981, + "step": 40808 + }, + { + "epoch": 0.3542417166517652, + "grad_norm": 0.4375, + "learning_rate": 0.0013264719806289338, + "loss": 0.1021, + "step": 40809 + }, + { + "epoch": 0.3542503971319693, + "grad_norm": 0.1640625, + "learning_rate": 0.0013264431757426542, + "loss": 0.0923, + "step": 40810 + }, + { + "epoch": 0.3542590776121735, + "grad_norm": 0.640625, + "learning_rate": 0.0013264143706087402, + "loss": 0.0962, + "step": 40811 + }, + { + "epoch": 0.35426775809237765, + "grad_norm": 0.166015625, + "learning_rate": 0.0013263855652272227, + "loss": 0.0713, + "step": 40812 + }, + { + "epoch": 0.35427643857258184, + "grad_norm": 0.455078125, + "learning_rate": 0.0013263567595981334, + "loss": 0.0962, + "step": 40813 + }, + { + "epoch": 0.354285119052786, + "grad_norm": 0.0908203125, + "learning_rate": 0.0013263279537215037, + "loss": 0.1094, + "step": 40814 + }, + { + "epoch": 0.3542937995329902, + "grad_norm": 0.28125, + "learning_rate": 0.0013262991475973653, + "loss": 0.1377, + "step": 40815 + }, + { + "epoch": 0.3543024800131943, + "grad_norm": 0.412109375, + "learning_rate": 0.0013262703412257493, + "loss": 0.0928, + "step": 40816 + }, + { + "epoch": 0.3543111604933985, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013262415346066872, + "loss": 0.0947, + "step": 40817 + }, + { + "epoch": 0.35431984097360264, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013262127277402112, + "loss": 0.1562, + "step": 40818 + }, + { + "epoch": 0.35432852145380683, + "grad_norm": 0.0830078125, + "learning_rate": 0.0013261839206263524, + "loss": 0.0859, + "step": 40819 + }, + { + "epoch": 0.35433720193401097, + "grad_norm": 0.3515625, + "learning_rate": 0.0013261551132651421, + "loss": 0.0713, + "step": 40820 + }, + { + "epoch": 0.35434588241421516, + "grad_norm": 0.88671875, + "learning_rate": 0.0013261263056566117, + "loss": 0.1758, + "step": 40821 + }, + { + "epoch": 0.3543545628944193, + "grad_norm": 0.259765625, + "learning_rate": 0.0013260974978007937, + "loss": 0.1069, + "step": 40822 + }, + { + "epoch": 0.3543632433746235, + "grad_norm": 0.10888671875, + "learning_rate": 0.0013260686896977183, + "loss": 0.0918, + "step": 40823 + }, + { + "epoch": 0.35437192385482763, + "grad_norm": 0.251953125, + "learning_rate": 0.001326039881347418, + "loss": 0.0752, + "step": 40824 + }, + { + "epoch": 0.3543806043350318, + "grad_norm": 0.2392578125, + "learning_rate": 0.0013260110727499233, + "loss": 0.0928, + "step": 40825 + }, + { + "epoch": 0.35438928481523596, + "grad_norm": 0.341796875, + "learning_rate": 0.0013259822639052669, + "loss": 0.0825, + "step": 40826 + }, + { + "epoch": 0.35439796529544015, + "grad_norm": 0.1904296875, + "learning_rate": 0.001325953454813479, + "loss": 0.0986, + "step": 40827 + }, + { + "epoch": 0.3544066457756443, + "grad_norm": 0.37890625, + "learning_rate": 0.0013259246454745924, + "loss": 0.1064, + "step": 40828 + }, + { + "epoch": 0.3544153262558485, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013258958358886375, + "loss": 0.0791, + "step": 40829 + }, + { + "epoch": 0.3544240067360526, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013258670260556465, + "loss": 0.1406, + "step": 40830 + }, + { + "epoch": 0.3544326872162568, + "grad_norm": 1.0390625, + "learning_rate": 0.0013258382159756507, + "loss": 0.1357, + "step": 40831 + }, + { + "epoch": 0.35444136769646095, + "grad_norm": 0.30859375, + "learning_rate": 0.001325809405648682, + "loss": 0.0986, + "step": 40832 + }, + { + "epoch": 0.35445004817666514, + "grad_norm": 0.58203125, + "learning_rate": 0.0013257805950747713, + "loss": 0.1084, + "step": 40833 + }, + { + "epoch": 0.3544587286568693, + "grad_norm": 0.16015625, + "learning_rate": 0.0013257517842539502, + "loss": 0.0884, + "step": 40834 + }, + { + "epoch": 0.3544674091370735, + "grad_norm": 0.099609375, + "learning_rate": 0.0013257229731862507, + "loss": 0.0796, + "step": 40835 + }, + { + "epoch": 0.3544760896172776, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013256941618717034, + "loss": 0.1074, + "step": 40836 + }, + { + "epoch": 0.3544847700974818, + "grad_norm": 0.349609375, + "learning_rate": 0.0013256653503103404, + "loss": 0.1235, + "step": 40837 + }, + { + "epoch": 0.35449345057768594, + "grad_norm": 0.35546875, + "learning_rate": 0.0013256365385021934, + "loss": 0.1104, + "step": 40838 + }, + { + "epoch": 0.35450213105789014, + "grad_norm": 0.08447265625, + "learning_rate": 0.0013256077264472937, + "loss": 0.1035, + "step": 40839 + }, + { + "epoch": 0.3545108115380943, + "grad_norm": 0.0947265625, + "learning_rate": 0.0013255789141456728, + "loss": 0.1377, + "step": 40840 + }, + { + "epoch": 0.35451949201829847, + "grad_norm": 0.22265625, + "learning_rate": 0.001325550101597362, + "loss": 0.0854, + "step": 40841 + }, + { + "epoch": 0.3545281724985026, + "grad_norm": 0.396484375, + "learning_rate": 0.0013255212888023933, + "loss": 0.0854, + "step": 40842 + }, + { + "epoch": 0.3545368529787068, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013254924757607977, + "loss": 0.1787, + "step": 40843 + }, + { + "epoch": 0.35454553345891093, + "grad_norm": 0.1953125, + "learning_rate": 0.0013254636624726072, + "loss": 0.0918, + "step": 40844 + }, + { + "epoch": 0.3545542139391151, + "grad_norm": 0.60546875, + "learning_rate": 0.0013254348489378524, + "loss": 0.1143, + "step": 40845 + }, + { + "epoch": 0.35456289441931926, + "grad_norm": 0.12451171875, + "learning_rate": 0.001325406035156566, + "loss": 0.1011, + "step": 40846 + }, + { + "epoch": 0.35457157489952346, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013253772211287787, + "loss": 0.123, + "step": 40847 + }, + { + "epoch": 0.3545802553797276, + "grad_norm": 0.30859375, + "learning_rate": 0.0013253484068545221, + "loss": 0.0981, + "step": 40848 + }, + { + "epoch": 0.3545889358599318, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013253195923338284, + "loss": 0.1377, + "step": 40849 + }, + { + "epoch": 0.3545976163401359, + "grad_norm": 0.087890625, + "learning_rate": 0.0013252907775667283, + "loss": 0.0977, + "step": 40850 + }, + { + "epoch": 0.3546062968203401, + "grad_norm": 0.3203125, + "learning_rate": 0.0013252619625532531, + "loss": 0.1367, + "step": 40851 + }, + { + "epoch": 0.35461497730054425, + "grad_norm": 0.275390625, + "learning_rate": 0.0013252331472934356, + "loss": 0.1221, + "step": 40852 + }, + { + "epoch": 0.35462365778074845, + "grad_norm": 0.357421875, + "learning_rate": 0.0013252043317873061, + "loss": 0.1064, + "step": 40853 + }, + { + "epoch": 0.3546323382609526, + "grad_norm": 0.158203125, + "learning_rate": 0.0013251755160348965, + "loss": 0.1143, + "step": 40854 + }, + { + "epoch": 0.3546410187411568, + "grad_norm": 0.232421875, + "learning_rate": 0.0013251467000362385, + "loss": 0.0947, + "step": 40855 + }, + { + "epoch": 0.3546496992213609, + "grad_norm": 0.0927734375, + "learning_rate": 0.0013251178837913632, + "loss": 0.0752, + "step": 40856 + }, + { + "epoch": 0.3546583797015651, + "grad_norm": 0.478515625, + "learning_rate": 0.0013250890673003027, + "loss": 0.0869, + "step": 40857 + }, + { + "epoch": 0.35466706018176924, + "grad_norm": 0.275390625, + "learning_rate": 0.001325060250563088, + "loss": 0.1035, + "step": 40858 + }, + { + "epoch": 0.35467574066197344, + "grad_norm": 0.35546875, + "learning_rate": 0.0013250314335797506, + "loss": 0.1357, + "step": 40859 + }, + { + "epoch": 0.3546844211421776, + "grad_norm": 0.1513671875, + "learning_rate": 0.0013250026163503226, + "loss": 0.1064, + "step": 40860 + }, + { + "epoch": 0.35469310162238177, + "grad_norm": 0.107421875, + "learning_rate": 0.001324973798874835, + "loss": 0.1094, + "step": 40861 + }, + { + "epoch": 0.3547017821025859, + "grad_norm": 0.2099609375, + "learning_rate": 0.0013249449811533194, + "loss": 0.082, + "step": 40862 + }, + { + "epoch": 0.3547104625827901, + "grad_norm": 0.52734375, + "learning_rate": 0.0013249161631858075, + "loss": 0.0757, + "step": 40863 + }, + { + "epoch": 0.35471914306299424, + "grad_norm": 0.287109375, + "learning_rate": 0.0013248873449723305, + "loss": 0.0854, + "step": 40864 + }, + { + "epoch": 0.35472782354319843, + "grad_norm": 0.271484375, + "learning_rate": 0.0013248585265129198, + "loss": 0.1104, + "step": 40865 + }, + { + "epoch": 0.35473650402340257, + "grad_norm": 0.470703125, + "learning_rate": 0.0013248297078076075, + "loss": 0.0894, + "step": 40866 + }, + { + "epoch": 0.35474518450360676, + "grad_norm": 0.357421875, + "learning_rate": 0.0013248008888564248, + "loss": 0.0894, + "step": 40867 + }, + { + "epoch": 0.3547538649838109, + "grad_norm": 0.080078125, + "learning_rate": 0.001324772069659403, + "loss": 0.0918, + "step": 40868 + }, + { + "epoch": 0.3547625454640151, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013247432502165742, + "loss": 0.0977, + "step": 40869 + }, + { + "epoch": 0.3547712259442192, + "grad_norm": 0.16015625, + "learning_rate": 0.0013247144305279695, + "loss": 0.1162, + "step": 40870 + }, + { + "epoch": 0.3547799064244234, + "grad_norm": 2.578125, + "learning_rate": 0.0013246856105936205, + "loss": 0.1504, + "step": 40871 + }, + { + "epoch": 0.35478858690462756, + "grad_norm": 0.162109375, + "learning_rate": 0.0013246567904135584, + "loss": 0.0811, + "step": 40872 + }, + { + "epoch": 0.35479726738483175, + "grad_norm": 0.3515625, + "learning_rate": 0.001324627969987815, + "loss": 0.1123, + "step": 40873 + }, + { + "epoch": 0.3548059478650359, + "grad_norm": 0.123046875, + "learning_rate": 0.001324599149316422, + "loss": 0.0957, + "step": 40874 + }, + { + "epoch": 0.3548146283452401, + "grad_norm": 0.09814453125, + "learning_rate": 0.001324570328399411, + "loss": 0.0845, + "step": 40875 + }, + { + "epoch": 0.3548233088254442, + "grad_norm": 0.123046875, + "learning_rate": 0.0013245415072368127, + "loss": 0.083, + "step": 40876 + }, + { + "epoch": 0.3548319893056484, + "grad_norm": 0.0810546875, + "learning_rate": 0.0013245126858286596, + "loss": 0.1074, + "step": 40877 + }, + { + "epoch": 0.35484066978585255, + "grad_norm": 0.291015625, + "learning_rate": 0.0013244838641749827, + "loss": 0.1494, + "step": 40878 + }, + { + "epoch": 0.35484935026605674, + "grad_norm": 0.126953125, + "learning_rate": 0.0013244550422758137, + "loss": 0.0649, + "step": 40879 + }, + { + "epoch": 0.3548580307462609, + "grad_norm": 0.095703125, + "learning_rate": 0.0013244262201311838, + "loss": 0.0854, + "step": 40880 + }, + { + "epoch": 0.35486671122646507, + "grad_norm": 0.08544921875, + "learning_rate": 0.001324397397741125, + "loss": 0.0889, + "step": 40881 + }, + { + "epoch": 0.3548753917066692, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013243685751056684, + "loss": 0.0918, + "step": 40882 + }, + { + "epoch": 0.3548840721868734, + "grad_norm": 0.45703125, + "learning_rate": 0.001324339752224846, + "loss": 0.0918, + "step": 40883 + }, + { + "epoch": 0.35489275266707754, + "grad_norm": 0.1591796875, + "learning_rate": 0.0013243109290986886, + "loss": 0.1172, + "step": 40884 + }, + { + "epoch": 0.35490143314728173, + "grad_norm": 0.0927734375, + "learning_rate": 0.0013242821057272282, + "loss": 0.1074, + "step": 40885 + }, + { + "epoch": 0.35491011362748587, + "grad_norm": 0.392578125, + "learning_rate": 0.0013242532821104965, + "loss": 0.1128, + "step": 40886 + }, + { + "epoch": 0.35491879410769006, + "grad_norm": 0.6015625, + "learning_rate": 0.0013242244582485246, + "loss": 0.1143, + "step": 40887 + }, + { + "epoch": 0.3549274745878942, + "grad_norm": 0.703125, + "learning_rate": 0.0013241956341413443, + "loss": 0.1309, + "step": 40888 + }, + { + "epoch": 0.3549361550680984, + "grad_norm": 0.55078125, + "learning_rate": 0.001324166809788987, + "loss": 0.1172, + "step": 40889 + }, + { + "epoch": 0.35494483554830253, + "grad_norm": 0.25390625, + "learning_rate": 0.0013241379851914841, + "loss": 0.1016, + "step": 40890 + }, + { + "epoch": 0.3549535160285067, + "grad_norm": 0.56640625, + "learning_rate": 0.0013241091603488676, + "loss": 0.0986, + "step": 40891 + }, + { + "epoch": 0.35496219650871086, + "grad_norm": 0.203125, + "learning_rate": 0.0013240803352611685, + "loss": 0.0708, + "step": 40892 + }, + { + "epoch": 0.35497087698891505, + "grad_norm": 1.15625, + "learning_rate": 0.0013240515099284184, + "loss": 0.1025, + "step": 40893 + }, + { + "epoch": 0.3549795574691192, + "grad_norm": 0.66015625, + "learning_rate": 0.0013240226843506489, + "loss": 0.0972, + "step": 40894 + }, + { + "epoch": 0.3549882379493234, + "grad_norm": 0.205078125, + "learning_rate": 0.0013239938585278919, + "loss": 0.1143, + "step": 40895 + }, + { + "epoch": 0.3549969184295275, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013239650324601785, + "loss": 0.0791, + "step": 40896 + }, + { + "epoch": 0.3550055989097317, + "grad_norm": 0.224609375, + "learning_rate": 0.0013239362061475402, + "loss": 0.1162, + "step": 40897 + }, + { + "epoch": 0.35501427938993585, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013239073795900085, + "loss": 0.1152, + "step": 40898 + }, + { + "epoch": 0.35502295987014004, + "grad_norm": 0.2392578125, + "learning_rate": 0.001323878552787615, + "loss": 0.1035, + "step": 40899 + }, + { + "epoch": 0.3550316403503442, + "grad_norm": 0.609375, + "learning_rate": 0.0013238497257403916, + "loss": 0.1562, + "step": 40900 + }, + { + "epoch": 0.3550403208305484, + "grad_norm": 0.08740234375, + "learning_rate": 0.0013238208984483696, + "loss": 0.0942, + "step": 40901 + }, + { + "epoch": 0.3550490013107525, + "grad_norm": 0.259765625, + "learning_rate": 0.0013237920709115798, + "loss": 0.124, + "step": 40902 + }, + { + "epoch": 0.3550576817909567, + "grad_norm": 0.19140625, + "learning_rate": 0.001323763243130055, + "loss": 0.1084, + "step": 40903 + }, + { + "epoch": 0.35506636227116084, + "grad_norm": 0.1455078125, + "learning_rate": 0.0013237344151038255, + "loss": 0.0942, + "step": 40904 + }, + { + "epoch": 0.35507504275136503, + "grad_norm": 0.08984375, + "learning_rate": 0.001323705586832924, + "loss": 0.0825, + "step": 40905 + }, + { + "epoch": 0.35508372323156917, + "grad_norm": 0.5546875, + "learning_rate": 0.001323676758317381, + "loss": 0.1055, + "step": 40906 + }, + { + "epoch": 0.3550924037117733, + "grad_norm": 0.2451171875, + "learning_rate": 0.0013236479295572286, + "loss": 0.1279, + "step": 40907 + }, + { + "epoch": 0.3551010841919775, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013236191005524982, + "loss": 0.1162, + "step": 40908 + }, + { + "epoch": 0.35510976467218164, + "grad_norm": 0.74609375, + "learning_rate": 0.0013235902713032216, + "loss": 0.0688, + "step": 40909 + }, + { + "epoch": 0.35511844515238583, + "grad_norm": 0.169921875, + "learning_rate": 0.0013235614418094298, + "loss": 0.1719, + "step": 40910 + }, + { + "epoch": 0.35512712563258997, + "grad_norm": 0.90234375, + "learning_rate": 0.0013235326120711543, + "loss": 0.1396, + "step": 40911 + }, + { + "epoch": 0.35513580611279416, + "grad_norm": 0.11279296875, + "learning_rate": 0.001323503782088427, + "loss": 0.0928, + "step": 40912 + }, + { + "epoch": 0.3551444865929983, + "grad_norm": 0.259765625, + "learning_rate": 0.0013234749518612795, + "loss": 0.1201, + "step": 40913 + }, + { + "epoch": 0.3551531670732025, + "grad_norm": 0.283203125, + "learning_rate": 0.0013234461213897431, + "loss": 0.082, + "step": 40914 + }, + { + "epoch": 0.35516184755340663, + "grad_norm": 0.119140625, + "learning_rate": 0.0013234172906738493, + "loss": 0.0942, + "step": 40915 + }, + { + "epoch": 0.3551705280336108, + "grad_norm": 0.1435546875, + "learning_rate": 0.0013233884597136296, + "loss": 0.0859, + "step": 40916 + }, + { + "epoch": 0.35517920851381496, + "grad_norm": 0.09033203125, + "learning_rate": 0.0013233596285091161, + "loss": 0.0742, + "step": 40917 + }, + { + "epoch": 0.35518788899401915, + "grad_norm": 0.15234375, + "learning_rate": 0.0013233307970603396, + "loss": 0.1055, + "step": 40918 + }, + { + "epoch": 0.3551965694742233, + "grad_norm": 0.263671875, + "learning_rate": 0.0013233019653673317, + "loss": 0.1099, + "step": 40919 + }, + { + "epoch": 0.3552052499544275, + "grad_norm": 0.07275390625, + "learning_rate": 0.0013232731334301242, + "loss": 0.0669, + "step": 40920 + }, + { + "epoch": 0.3552139304346316, + "grad_norm": 0.20703125, + "learning_rate": 0.001323244301248749, + "loss": 0.1123, + "step": 40921 + }, + { + "epoch": 0.3552226109148358, + "grad_norm": 0.328125, + "learning_rate": 0.0013232154688232365, + "loss": 0.1328, + "step": 40922 + }, + { + "epoch": 0.35523129139503995, + "grad_norm": 0.66796875, + "learning_rate": 0.0013231866361536194, + "loss": 0.1465, + "step": 40923 + }, + { + "epoch": 0.35523997187524414, + "grad_norm": 0.150390625, + "learning_rate": 0.0013231578032399283, + "loss": 0.0869, + "step": 40924 + }, + { + "epoch": 0.3552486523554483, + "grad_norm": 1.046875, + "learning_rate": 0.0013231289700821956, + "loss": 0.1162, + "step": 40925 + }, + { + "epoch": 0.3552573328356525, + "grad_norm": 0.77734375, + "learning_rate": 0.0013231001366804523, + "loss": 0.1152, + "step": 40926 + }, + { + "epoch": 0.3552660133158566, + "grad_norm": 0.1572265625, + "learning_rate": 0.00132307130303473, + "loss": 0.125, + "step": 40927 + }, + { + "epoch": 0.3552746937960608, + "grad_norm": 0.07177734375, + "learning_rate": 0.0013230424691450604, + "loss": 0.0723, + "step": 40928 + }, + { + "epoch": 0.35528337427626494, + "grad_norm": 0.326171875, + "learning_rate": 0.0013230136350114747, + "loss": 0.1748, + "step": 40929 + }, + { + "epoch": 0.35529205475646913, + "grad_norm": 0.318359375, + "learning_rate": 0.0013229848006340048, + "loss": 0.1152, + "step": 40930 + }, + { + "epoch": 0.35530073523667327, + "grad_norm": 0.0732421875, + "learning_rate": 0.0013229559660126822, + "loss": 0.0786, + "step": 40931 + }, + { + "epoch": 0.35530941571687746, + "grad_norm": 0.0927734375, + "learning_rate": 0.0013229271311475381, + "loss": 0.1064, + "step": 40932 + }, + { + "epoch": 0.3553180961970816, + "grad_norm": 0.21875, + "learning_rate": 0.001322898296038604, + "loss": 0.0898, + "step": 40933 + }, + { + "epoch": 0.3553267766772858, + "grad_norm": 0.2333984375, + "learning_rate": 0.001322869460685912, + "loss": 0.103, + "step": 40934 + }, + { + "epoch": 0.35533545715748993, + "grad_norm": 0.07958984375, + "learning_rate": 0.0013228406250894932, + "loss": 0.082, + "step": 40935 + }, + { + "epoch": 0.3553441376376941, + "grad_norm": 0.095703125, + "learning_rate": 0.0013228117892493795, + "loss": 0.125, + "step": 40936 + }, + { + "epoch": 0.35535281811789826, + "grad_norm": 0.07080078125, + "learning_rate": 0.001322782953165602, + "loss": 0.0703, + "step": 40937 + }, + { + "epoch": 0.35536149859810245, + "grad_norm": 0.1328125, + "learning_rate": 0.0013227541168381923, + "loss": 0.0825, + "step": 40938 + }, + { + "epoch": 0.3553701790783066, + "grad_norm": 0.205078125, + "learning_rate": 0.0013227252802671821, + "loss": 0.1074, + "step": 40939 + }, + { + "epoch": 0.3553788595585108, + "grad_norm": 0.1259765625, + "learning_rate": 0.001322696443452603, + "loss": 0.1143, + "step": 40940 + }, + { + "epoch": 0.3553875400387149, + "grad_norm": 0.1875, + "learning_rate": 0.0013226676063944863, + "loss": 0.0757, + "step": 40941 + }, + { + "epoch": 0.3553962205189191, + "grad_norm": 0.130859375, + "learning_rate": 0.0013226387690928635, + "loss": 0.1123, + "step": 40942 + }, + { + "epoch": 0.35540490099912325, + "grad_norm": 0.341796875, + "learning_rate": 0.0013226099315477664, + "loss": 0.0991, + "step": 40943 + }, + { + "epoch": 0.35541358147932745, + "grad_norm": 0.396484375, + "learning_rate": 0.0013225810937592267, + "loss": 0.1543, + "step": 40944 + }, + { + "epoch": 0.3554222619595316, + "grad_norm": 0.625, + "learning_rate": 0.0013225522557272754, + "loss": 0.1084, + "step": 40945 + }, + { + "epoch": 0.3554309424397358, + "grad_norm": 0.0810546875, + "learning_rate": 0.0013225234174519443, + "loss": 0.085, + "step": 40946 + }, + { + "epoch": 0.3554396229199399, + "grad_norm": 0.53125, + "learning_rate": 0.001322494578933265, + "loss": 0.0806, + "step": 40947 + }, + { + "epoch": 0.3554483034001441, + "grad_norm": 0.609375, + "learning_rate": 0.001322465740171269, + "loss": 0.1094, + "step": 40948 + }, + { + "epoch": 0.35545698388034824, + "grad_norm": 0.15234375, + "learning_rate": 0.0013224369011659875, + "loss": 0.0952, + "step": 40949 + }, + { + "epoch": 0.35546566436055244, + "grad_norm": 0.09765625, + "learning_rate": 0.0013224080619174525, + "loss": 0.1084, + "step": 40950 + }, + { + "epoch": 0.3554743448407566, + "grad_norm": 0.404296875, + "learning_rate": 0.0013223792224256957, + "loss": 0.0796, + "step": 40951 + }, + { + "epoch": 0.35548302532096077, + "grad_norm": 0.2490234375, + "learning_rate": 0.001322350382690748, + "loss": 0.0952, + "step": 40952 + }, + { + "epoch": 0.3554917058011649, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013223215427126413, + "loss": 0.1157, + "step": 40953 + }, + { + "epoch": 0.3555003862813691, + "grad_norm": 0.1494140625, + "learning_rate": 0.0013222927024914071, + "loss": 0.0752, + "step": 40954 + }, + { + "epoch": 0.35550906676157323, + "grad_norm": 0.373046875, + "learning_rate": 0.001322263862027077, + "loss": 0.0947, + "step": 40955 + }, + { + "epoch": 0.3555177472417774, + "grad_norm": 0.291015625, + "learning_rate": 0.0013222350213196827, + "loss": 0.1074, + "step": 40956 + }, + { + "epoch": 0.35552642772198156, + "grad_norm": 0.3203125, + "learning_rate": 0.0013222061803692554, + "loss": 0.127, + "step": 40957 + }, + { + "epoch": 0.35553510820218576, + "grad_norm": 0.1474609375, + "learning_rate": 0.0013221773391758264, + "loss": 0.1523, + "step": 40958 + }, + { + "epoch": 0.3555437886823899, + "grad_norm": 0.4140625, + "learning_rate": 0.001322148497739428, + "loss": 0.1201, + "step": 40959 + }, + { + "epoch": 0.3555524691625941, + "grad_norm": 0.375, + "learning_rate": 0.0013221196560600913, + "loss": 0.0859, + "step": 40960 + }, + { + "epoch": 0.3555611496427982, + "grad_norm": 0.43359375, + "learning_rate": 0.0013220908141378476, + "loss": 0.0928, + "step": 40961 + }, + { + "epoch": 0.3555698301230024, + "grad_norm": 0.189453125, + "learning_rate": 0.001322061971972729, + "loss": 0.0791, + "step": 40962 + }, + { + "epoch": 0.35557851060320655, + "grad_norm": 0.2490234375, + "learning_rate": 0.0013220331295647665, + "loss": 0.0898, + "step": 40963 + }, + { + "epoch": 0.35558719108341075, + "grad_norm": 0.294921875, + "learning_rate": 0.001322004286913992, + "loss": 0.0977, + "step": 40964 + }, + { + "epoch": 0.3555958715636149, + "grad_norm": 0.3046875, + "learning_rate": 0.0013219754440204371, + "loss": 0.0898, + "step": 40965 + }, + { + "epoch": 0.3556045520438191, + "grad_norm": 0.69921875, + "learning_rate": 0.0013219466008841333, + "loss": 0.1777, + "step": 40966 + }, + { + "epoch": 0.3556132325240232, + "grad_norm": 0.92578125, + "learning_rate": 0.0013219177575051115, + "loss": 0.1143, + "step": 40967 + }, + { + "epoch": 0.3556219130042274, + "grad_norm": 0.158203125, + "learning_rate": 0.0013218889138834043, + "loss": 0.0967, + "step": 40968 + }, + { + "epoch": 0.35563059348443155, + "grad_norm": 0.107421875, + "learning_rate": 0.0013218600700190429, + "loss": 0.0879, + "step": 40969 + }, + { + "epoch": 0.35563927396463574, + "grad_norm": 0.2734375, + "learning_rate": 0.001321831225912058, + "loss": 0.1719, + "step": 40970 + }, + { + "epoch": 0.3556479544448399, + "grad_norm": 0.234375, + "learning_rate": 0.0013218023815624823, + "loss": 0.0742, + "step": 40971 + }, + { + "epoch": 0.35565663492504407, + "grad_norm": 0.3671875, + "learning_rate": 0.0013217735369703465, + "loss": 0.0728, + "step": 40972 + }, + { + "epoch": 0.3556653154052482, + "grad_norm": 0.19921875, + "learning_rate": 0.0013217446921356827, + "loss": 0.1123, + "step": 40973 + }, + { + "epoch": 0.3556739958854524, + "grad_norm": 0.1689453125, + "learning_rate": 0.0013217158470585223, + "loss": 0.0986, + "step": 40974 + }, + { + "epoch": 0.35568267636565654, + "grad_norm": 0.5, + "learning_rate": 0.0013216870017388963, + "loss": 0.123, + "step": 40975 + }, + { + "epoch": 0.35569135684586073, + "grad_norm": 0.25390625, + "learning_rate": 0.0013216581561768372, + "loss": 0.0669, + "step": 40976 + }, + { + "epoch": 0.35570003732606487, + "grad_norm": 0.353515625, + "learning_rate": 0.0013216293103723762, + "loss": 0.1455, + "step": 40977 + }, + { + "epoch": 0.35570871780626906, + "grad_norm": 0.189453125, + "learning_rate": 0.0013216004643255444, + "loss": 0.1494, + "step": 40978 + }, + { + "epoch": 0.3557173982864732, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013215716180363737, + "loss": 0.0737, + "step": 40979 + }, + { + "epoch": 0.3557260787666774, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013215427715048958, + "loss": 0.1162, + "step": 40980 + }, + { + "epoch": 0.3557347592468815, + "grad_norm": 0.69921875, + "learning_rate": 0.0013215139247311416, + "loss": 0.1064, + "step": 40981 + }, + { + "epoch": 0.3557434397270857, + "grad_norm": 0.166015625, + "learning_rate": 0.0013214850777151434, + "loss": 0.1167, + "step": 40982 + }, + { + "epoch": 0.35575212020728986, + "grad_norm": 0.09033203125, + "learning_rate": 0.0013214562304569323, + "loss": 0.0864, + "step": 40983 + }, + { + "epoch": 0.35576080068749405, + "grad_norm": 0.1953125, + "learning_rate": 0.0013214273829565404, + "loss": 0.0854, + "step": 40984 + }, + { + "epoch": 0.3557694811676982, + "grad_norm": 0.134765625, + "learning_rate": 0.0013213985352139987, + "loss": 0.0811, + "step": 40985 + }, + { + "epoch": 0.3557781616479024, + "grad_norm": 0.486328125, + "learning_rate": 0.0013213696872293385, + "loss": 0.0845, + "step": 40986 + }, + { + "epoch": 0.3557868421281065, + "grad_norm": 1.0546875, + "learning_rate": 0.0013213408390025923, + "loss": 0.1367, + "step": 40987 + }, + { + "epoch": 0.3557955226083107, + "grad_norm": 0.09326171875, + "learning_rate": 0.001321311990533791, + "loss": 0.0918, + "step": 40988 + }, + { + "epoch": 0.35580420308851485, + "grad_norm": 0.3203125, + "learning_rate": 0.0013212831418229656, + "loss": 0.1016, + "step": 40989 + }, + { + "epoch": 0.35581288356871904, + "grad_norm": 0.255859375, + "learning_rate": 0.0013212542928701485, + "loss": 0.0957, + "step": 40990 + }, + { + "epoch": 0.3558215640489232, + "grad_norm": 0.2578125, + "learning_rate": 0.0013212254436753713, + "loss": 0.123, + "step": 40991 + }, + { + "epoch": 0.35583024452912737, + "grad_norm": 0.41015625, + "learning_rate": 0.0013211965942386654, + "loss": 0.1426, + "step": 40992 + }, + { + "epoch": 0.3558389250093315, + "grad_norm": 0.2294921875, + "learning_rate": 0.0013211677445600619, + "loss": 0.1196, + "step": 40993 + }, + { + "epoch": 0.3558476054895357, + "grad_norm": 0.4296875, + "learning_rate": 0.0013211388946395928, + "loss": 0.126, + "step": 40994 + }, + { + "epoch": 0.35585628596973984, + "grad_norm": 0.384765625, + "learning_rate": 0.0013211100444772896, + "loss": 0.167, + "step": 40995 + }, + { + "epoch": 0.35586496644994403, + "grad_norm": 0.25, + "learning_rate": 0.001321081194073184, + "loss": 0.1104, + "step": 40996 + }, + { + "epoch": 0.35587364693014817, + "grad_norm": 0.3828125, + "learning_rate": 0.0013210523434273069, + "loss": 0.0898, + "step": 40997 + }, + { + "epoch": 0.35588232741035236, + "grad_norm": 0.341796875, + "learning_rate": 0.0013210234925396904, + "loss": 0.0859, + "step": 40998 + }, + { + "epoch": 0.3558910078905565, + "grad_norm": 0.39453125, + "learning_rate": 0.0013209946414103658, + "loss": 0.1094, + "step": 40999 + }, + { + "epoch": 0.3558996883707607, + "grad_norm": 0.111328125, + "learning_rate": 0.001320965790039365, + "loss": 0.1025, + "step": 41000 + }, + { + "epoch": 0.35590836885096483, + "grad_norm": 0.48828125, + "learning_rate": 0.0013209369384267193, + "loss": 0.1299, + "step": 41001 + }, + { + "epoch": 0.355917049331169, + "grad_norm": 0.09765625, + "learning_rate": 0.0013209080865724601, + "loss": 0.0688, + "step": 41002 + }, + { + "epoch": 0.35592572981137316, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013208792344766194, + "loss": 0.0938, + "step": 41003 + }, + { + "epoch": 0.35593441029157735, + "grad_norm": 0.251953125, + "learning_rate": 0.0013208503821392284, + "loss": 0.127, + "step": 41004 + }, + { + "epoch": 0.3559430907717815, + "grad_norm": 0.1689453125, + "learning_rate": 0.0013208215295603188, + "loss": 0.1289, + "step": 41005 + }, + { + "epoch": 0.3559517712519857, + "grad_norm": 0.28515625, + "learning_rate": 0.0013207926767399218, + "loss": 0.1128, + "step": 41006 + }, + { + "epoch": 0.3559604517321898, + "grad_norm": 0.12158203125, + "learning_rate": 0.0013207638236780697, + "loss": 0.084, + "step": 41007 + }, + { + "epoch": 0.355969132212394, + "grad_norm": 0.2177734375, + "learning_rate": 0.0013207349703747934, + "loss": 0.1011, + "step": 41008 + }, + { + "epoch": 0.35597781269259815, + "grad_norm": 0.076171875, + "learning_rate": 0.0013207061168301244, + "loss": 0.0952, + "step": 41009 + }, + { + "epoch": 0.35598649317280234, + "grad_norm": 0.4296875, + "learning_rate": 0.0013206772630440947, + "loss": 0.1143, + "step": 41010 + }, + { + "epoch": 0.3559951736530065, + "grad_norm": 0.251953125, + "learning_rate": 0.0013206484090167356, + "loss": 0.0996, + "step": 41011 + }, + { + "epoch": 0.3560038541332107, + "grad_norm": 0.255859375, + "learning_rate": 0.001320619554748079, + "loss": 0.1162, + "step": 41012 + }, + { + "epoch": 0.3560125346134148, + "grad_norm": 0.46484375, + "learning_rate": 0.0013205907002381558, + "loss": 0.1152, + "step": 41013 + }, + { + "epoch": 0.356021215093619, + "grad_norm": 0.80859375, + "learning_rate": 0.0013205618454869984, + "loss": 0.1055, + "step": 41014 + }, + { + "epoch": 0.35602989557382314, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013205329904946372, + "loss": 0.083, + "step": 41015 + }, + { + "epoch": 0.35603857605402733, + "grad_norm": 0.2890625, + "learning_rate": 0.001320504135261105, + "loss": 0.0781, + "step": 41016 + }, + { + "epoch": 0.35604725653423147, + "grad_norm": 0.25, + "learning_rate": 0.0013204752797864325, + "loss": 0.1328, + "step": 41017 + }, + { + "epoch": 0.35605593701443566, + "grad_norm": 0.19140625, + "learning_rate": 0.0013204464240706515, + "loss": 0.0918, + "step": 41018 + }, + { + "epoch": 0.3560646174946398, + "grad_norm": 0.890625, + "learning_rate": 0.0013204175681137938, + "loss": 0.0884, + "step": 41019 + }, + { + "epoch": 0.356073297974844, + "grad_norm": 0.384765625, + "learning_rate": 0.0013203887119158906, + "loss": 0.1211, + "step": 41020 + }, + { + "epoch": 0.35608197845504813, + "grad_norm": 0.55859375, + "learning_rate": 0.0013203598554769739, + "loss": 0.1211, + "step": 41021 + }, + { + "epoch": 0.3560906589352523, + "grad_norm": 0.11328125, + "learning_rate": 0.0013203309987970748, + "loss": 0.1162, + "step": 41022 + }, + { + "epoch": 0.35609933941545646, + "grad_norm": 1.1640625, + "learning_rate": 0.001320302141876225, + "loss": 0.1416, + "step": 41023 + }, + { + "epoch": 0.35610801989566065, + "grad_norm": 0.0810546875, + "learning_rate": 0.001320273284714456, + "loss": 0.082, + "step": 41024 + }, + { + "epoch": 0.3561167003758648, + "grad_norm": 0.5234375, + "learning_rate": 0.0013202444273117998, + "loss": 0.1147, + "step": 41025 + }, + { + "epoch": 0.356125380856069, + "grad_norm": 0.1220703125, + "learning_rate": 0.0013202155696682874, + "loss": 0.1055, + "step": 41026 + }, + { + "epoch": 0.3561340613362731, + "grad_norm": 0.11083984375, + "learning_rate": 0.0013201867117839502, + "loss": 0.1211, + "step": 41027 + }, + { + "epoch": 0.3561427418164773, + "grad_norm": 0.27734375, + "learning_rate": 0.001320157853658821, + "loss": 0.0854, + "step": 41028 + }, + { + "epoch": 0.35615142229668145, + "grad_norm": 0.14453125, + "learning_rate": 0.0013201289952929295, + "loss": 0.1035, + "step": 41029 + }, + { + "epoch": 0.3561601027768856, + "grad_norm": 0.365234375, + "learning_rate": 0.0013201001366863086, + "loss": 0.0898, + "step": 41030 + }, + { + "epoch": 0.3561687832570898, + "grad_norm": 0.1357421875, + "learning_rate": 0.0013200712778389896, + "loss": 0.084, + "step": 41031 + }, + { + "epoch": 0.3561774637372939, + "grad_norm": 0.6171875, + "learning_rate": 0.001320042418751004, + "loss": 0.0996, + "step": 41032 + }, + { + "epoch": 0.3561861442174981, + "grad_norm": 0.78125, + "learning_rate": 0.0013200135594223833, + "loss": 0.0898, + "step": 41033 + }, + { + "epoch": 0.35619482469770225, + "grad_norm": 0.1484375, + "learning_rate": 0.001319984699853159, + "loss": 0.0811, + "step": 41034 + }, + { + "epoch": 0.35620350517790644, + "grad_norm": 0.158203125, + "learning_rate": 0.0013199558400433626, + "loss": 0.0938, + "step": 41035 + }, + { + "epoch": 0.3562121856581106, + "grad_norm": 0.515625, + "learning_rate": 0.0013199269799930259, + "loss": 0.1099, + "step": 41036 + }, + { + "epoch": 0.3562208661383148, + "grad_norm": 0.25390625, + "learning_rate": 0.0013198981197021804, + "loss": 0.1113, + "step": 41037 + }, + { + "epoch": 0.3562295466185189, + "grad_norm": 0.32421875, + "learning_rate": 0.0013198692591708577, + "loss": 0.1211, + "step": 41038 + }, + { + "epoch": 0.3562382270987231, + "grad_norm": 0.69140625, + "learning_rate": 0.0013198403983990892, + "loss": 0.082, + "step": 41039 + }, + { + "epoch": 0.35624690757892724, + "grad_norm": 0.50390625, + "learning_rate": 0.0013198115373869066, + "loss": 0.1104, + "step": 41040 + }, + { + "epoch": 0.35625558805913143, + "grad_norm": 0.451171875, + "learning_rate": 0.0013197826761343413, + "loss": 0.1182, + "step": 41041 + }, + { + "epoch": 0.35626426853933557, + "grad_norm": 0.08935546875, + "learning_rate": 0.001319753814641425, + "loss": 0.1001, + "step": 41042 + }, + { + "epoch": 0.35627294901953976, + "grad_norm": 0.396484375, + "learning_rate": 0.0013197249529081892, + "loss": 0.0854, + "step": 41043 + }, + { + "epoch": 0.3562816294997439, + "grad_norm": 0.09912109375, + "learning_rate": 0.0013196960909346656, + "loss": 0.1074, + "step": 41044 + }, + { + "epoch": 0.3562903099799481, + "grad_norm": 0.490234375, + "learning_rate": 0.0013196672287208856, + "loss": 0.1455, + "step": 41045 + }, + { + "epoch": 0.35629899046015223, + "grad_norm": 0.1044921875, + "learning_rate": 0.0013196383662668807, + "loss": 0.0991, + "step": 41046 + }, + { + "epoch": 0.3563076709403564, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013196095035726827, + "loss": 0.0859, + "step": 41047 + }, + { + "epoch": 0.35631635142056056, + "grad_norm": 0.259765625, + "learning_rate": 0.001319580640638323, + "loss": 0.1084, + "step": 41048 + }, + { + "epoch": 0.35632503190076475, + "grad_norm": 0.173828125, + "learning_rate": 0.0013195517774638333, + "loss": 0.0801, + "step": 41049 + }, + { + "epoch": 0.3563337123809689, + "grad_norm": 0.40625, + "learning_rate": 0.001319522914049245, + "loss": 0.0889, + "step": 41050 + }, + { + "epoch": 0.3563423928611731, + "grad_norm": 0.203125, + "learning_rate": 0.0013194940503945898, + "loss": 0.0942, + "step": 41051 + }, + { + "epoch": 0.3563510733413772, + "grad_norm": 0.48046875, + "learning_rate": 0.0013194651864998992, + "loss": 0.1016, + "step": 41052 + }, + { + "epoch": 0.3563597538215814, + "grad_norm": 0.228515625, + "learning_rate": 0.0013194363223652048, + "loss": 0.1064, + "step": 41053 + }, + { + "epoch": 0.35636843430178555, + "grad_norm": 0.19921875, + "learning_rate": 0.001319407457990538, + "loss": 0.0996, + "step": 41054 + }, + { + "epoch": 0.35637711478198975, + "grad_norm": 0.69921875, + "learning_rate": 0.0013193785933759306, + "loss": 0.1162, + "step": 41055 + }, + { + "epoch": 0.3563857952621939, + "grad_norm": 0.224609375, + "learning_rate": 0.0013193497285214142, + "loss": 0.1172, + "step": 41056 + }, + { + "epoch": 0.3563944757423981, + "grad_norm": 0.8828125, + "learning_rate": 0.00131932086342702, + "loss": 0.1104, + "step": 41057 + }, + { + "epoch": 0.3564031562226022, + "grad_norm": 0.251953125, + "learning_rate": 0.0013192919980927796, + "loss": 0.1084, + "step": 41058 + }, + { + "epoch": 0.3564118367028064, + "grad_norm": 0.14453125, + "learning_rate": 0.001319263132518725, + "loss": 0.0996, + "step": 41059 + }, + { + "epoch": 0.35642051718301054, + "grad_norm": 0.193359375, + "learning_rate": 0.0013192342667048879, + "loss": 0.0898, + "step": 41060 + }, + { + "epoch": 0.35642919766321474, + "grad_norm": 0.427734375, + "learning_rate": 0.001319205400651299, + "loss": 0.1348, + "step": 41061 + }, + { + "epoch": 0.3564378781434189, + "grad_norm": 0.10009765625, + "learning_rate": 0.0013191765343579904, + "loss": 0.1396, + "step": 41062 + }, + { + "epoch": 0.35644655862362307, + "grad_norm": 0.318359375, + "learning_rate": 0.0013191476678249938, + "loss": 0.0947, + "step": 41063 + }, + { + "epoch": 0.3564552391038272, + "grad_norm": 0.478515625, + "learning_rate": 0.0013191188010523408, + "loss": 0.1245, + "step": 41064 + }, + { + "epoch": 0.3564639195840314, + "grad_norm": 0.2216796875, + "learning_rate": 0.0013190899340400626, + "loss": 0.0923, + "step": 41065 + }, + { + "epoch": 0.35647260006423553, + "grad_norm": 0.2392578125, + "learning_rate": 0.0013190610667881907, + "loss": 0.0898, + "step": 41066 + }, + { + "epoch": 0.3564812805444397, + "grad_norm": 0.08984375, + "learning_rate": 0.001319032199296757, + "loss": 0.0688, + "step": 41067 + }, + { + "epoch": 0.35648996102464386, + "grad_norm": 0.16796875, + "learning_rate": 0.0013190033315657933, + "loss": 0.1328, + "step": 41068 + }, + { + "epoch": 0.35649864150484806, + "grad_norm": 0.1962890625, + "learning_rate": 0.0013189744635953306, + "loss": 0.0796, + "step": 41069 + }, + { + "epoch": 0.3565073219850522, + "grad_norm": 0.54296875, + "learning_rate": 0.0013189455953854007, + "loss": 0.1016, + "step": 41070 + }, + { + "epoch": 0.3565160024652564, + "grad_norm": 0.306640625, + "learning_rate": 0.001318916726936035, + "loss": 0.0835, + "step": 41071 + }, + { + "epoch": 0.3565246829454605, + "grad_norm": 0.66015625, + "learning_rate": 0.0013188878582472657, + "loss": 0.0776, + "step": 41072 + }, + { + "epoch": 0.3565333634256647, + "grad_norm": 0.3046875, + "learning_rate": 0.0013188589893191238, + "loss": 0.1099, + "step": 41073 + }, + { + "epoch": 0.35654204390586886, + "grad_norm": 0.2890625, + "learning_rate": 0.001318830120151641, + "loss": 0.1143, + "step": 41074 + }, + { + "epoch": 0.35655072438607305, + "grad_norm": 0.087890625, + "learning_rate": 0.0013188012507448486, + "loss": 0.0947, + "step": 41075 + }, + { + "epoch": 0.3565594048662772, + "grad_norm": 0.26953125, + "learning_rate": 0.0013187723810987786, + "loss": 0.1074, + "step": 41076 + }, + { + "epoch": 0.3565680853464814, + "grad_norm": 0.453125, + "learning_rate": 0.0013187435112134622, + "loss": 0.0996, + "step": 41077 + }, + { + "epoch": 0.3565767658266855, + "grad_norm": 0.3203125, + "learning_rate": 0.0013187146410889314, + "loss": 0.1543, + "step": 41078 + }, + { + "epoch": 0.3565854463068897, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013186857707252178, + "loss": 0.1064, + "step": 41079 + }, + { + "epoch": 0.35659412678709385, + "grad_norm": 0.12890625, + "learning_rate": 0.0013186569001223526, + "loss": 0.0835, + "step": 41080 + }, + { + "epoch": 0.35660280726729804, + "grad_norm": 0.65625, + "learning_rate": 0.0013186280292803673, + "loss": 0.1719, + "step": 41081 + }, + { + "epoch": 0.3566114877475022, + "grad_norm": 0.166015625, + "learning_rate": 0.0013185991581992937, + "loss": 0.0962, + "step": 41082 + }, + { + "epoch": 0.35662016822770637, + "grad_norm": 0.66015625, + "learning_rate": 0.0013185702868791637, + "loss": 0.1172, + "step": 41083 + }, + { + "epoch": 0.3566288487079105, + "grad_norm": 0.55859375, + "learning_rate": 0.001318541415320008, + "loss": 0.0933, + "step": 41084 + }, + { + "epoch": 0.3566375291881147, + "grad_norm": 0.423828125, + "learning_rate": 0.001318512543521859, + "loss": 0.085, + "step": 41085 + }, + { + "epoch": 0.35664620966831884, + "grad_norm": 0.59375, + "learning_rate": 0.0013184836714847476, + "loss": 0.126, + "step": 41086 + }, + { + "epoch": 0.35665489014852303, + "grad_norm": 0.388671875, + "learning_rate": 0.001318454799208706, + "loss": 0.1113, + "step": 41087 + }, + { + "epoch": 0.35666357062872717, + "grad_norm": 0.142578125, + "learning_rate": 0.0013184259266937658, + "loss": 0.1143, + "step": 41088 + }, + { + "epoch": 0.35667225110893136, + "grad_norm": 0.2734375, + "learning_rate": 0.001318397053939958, + "loss": 0.083, + "step": 41089 + }, + { + "epoch": 0.3566809315891355, + "grad_norm": 0.11474609375, + "learning_rate": 0.0013183681809473145, + "loss": 0.0967, + "step": 41090 + }, + { + "epoch": 0.3566896120693397, + "grad_norm": 0.318359375, + "learning_rate": 0.0013183393077158666, + "loss": 0.1157, + "step": 41091 + }, + { + "epoch": 0.3566982925495438, + "grad_norm": 0.369140625, + "learning_rate": 0.0013183104342456467, + "loss": 0.083, + "step": 41092 + }, + { + "epoch": 0.356706973029748, + "grad_norm": 0.21484375, + "learning_rate": 0.0013182815605366851, + "loss": 0.123, + "step": 41093 + }, + { + "epoch": 0.35671565350995216, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013182526865890146, + "loss": 0.0854, + "step": 41094 + }, + { + "epoch": 0.35672433399015635, + "grad_norm": 0.51953125, + "learning_rate": 0.001318223812402666, + "loss": 0.1143, + "step": 41095 + }, + { + "epoch": 0.3567330144703605, + "grad_norm": 0.12451171875, + "learning_rate": 0.0013181949379776712, + "loss": 0.1055, + "step": 41096 + }, + { + "epoch": 0.3567416949505647, + "grad_norm": 0.48046875, + "learning_rate": 0.0013181660633140612, + "loss": 0.0874, + "step": 41097 + }, + { + "epoch": 0.3567503754307688, + "grad_norm": 0.080078125, + "learning_rate": 0.0013181371884118686, + "loss": 0.1025, + "step": 41098 + }, + { + "epoch": 0.356759055910973, + "grad_norm": 0.2216796875, + "learning_rate": 0.0013181083132711247, + "loss": 0.0972, + "step": 41099 + }, + { + "epoch": 0.35676773639117715, + "grad_norm": 0.251953125, + "learning_rate": 0.0013180794378918604, + "loss": 0.0908, + "step": 41100 + }, + { + "epoch": 0.35677641687138134, + "grad_norm": 0.67578125, + "learning_rate": 0.0013180505622741078, + "loss": 0.1074, + "step": 41101 + }, + { + "epoch": 0.3567850973515855, + "grad_norm": 0.3046875, + "learning_rate": 0.0013180216864178984, + "loss": 0.0835, + "step": 41102 + }, + { + "epoch": 0.35679377783178967, + "grad_norm": 0.169921875, + "learning_rate": 0.0013179928103232636, + "loss": 0.0586, + "step": 41103 + }, + { + "epoch": 0.3568024583119938, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013179639339902353, + "loss": 0.1104, + "step": 41104 + }, + { + "epoch": 0.356811138792198, + "grad_norm": 0.337890625, + "learning_rate": 0.001317935057418845, + "loss": 0.0781, + "step": 41105 + }, + { + "epoch": 0.35681981927240214, + "grad_norm": 0.291015625, + "learning_rate": 0.001317906180609124, + "loss": 0.1113, + "step": 41106 + }, + { + "epoch": 0.35682849975260633, + "grad_norm": 0.2109375, + "learning_rate": 0.0013178773035611043, + "loss": 0.1318, + "step": 41107 + }, + { + "epoch": 0.35683718023281047, + "grad_norm": 0.2099609375, + "learning_rate": 0.001317848426274817, + "loss": 0.0869, + "step": 41108 + }, + { + "epoch": 0.35684586071301466, + "grad_norm": 0.07763671875, + "learning_rate": 0.0013178195487502941, + "loss": 0.0835, + "step": 41109 + }, + { + "epoch": 0.3568545411932188, + "grad_norm": 0.54296875, + "learning_rate": 0.0013177906709875668, + "loss": 0.1328, + "step": 41110 + }, + { + "epoch": 0.356863221673423, + "grad_norm": 0.51171875, + "learning_rate": 0.0013177617929866672, + "loss": 0.0918, + "step": 41111 + }, + { + "epoch": 0.35687190215362713, + "grad_norm": 0.10302734375, + "learning_rate": 0.0013177329147476267, + "loss": 0.0903, + "step": 41112 + }, + { + "epoch": 0.3568805826338313, + "grad_norm": 0.73828125, + "learning_rate": 0.0013177040362704765, + "loss": 0.0889, + "step": 41113 + }, + { + "epoch": 0.35688926311403546, + "grad_norm": 0.1640625, + "learning_rate": 0.001317675157555248, + "loss": 0.0908, + "step": 41114 + }, + { + "epoch": 0.35689794359423965, + "grad_norm": 0.318359375, + "learning_rate": 0.0013176462786019735, + "loss": 0.1064, + "step": 41115 + }, + { + "epoch": 0.3569066240744438, + "grad_norm": 0.291015625, + "learning_rate": 0.0013176173994106847, + "loss": 0.0986, + "step": 41116 + }, + { + "epoch": 0.356915304554648, + "grad_norm": 0.265625, + "learning_rate": 0.0013175885199814127, + "loss": 0.1064, + "step": 41117 + }, + { + "epoch": 0.3569239850348521, + "grad_norm": 0.10546875, + "learning_rate": 0.0013175596403141889, + "loss": 0.0693, + "step": 41118 + }, + { + "epoch": 0.3569326655150563, + "grad_norm": 0.09423828125, + "learning_rate": 0.0013175307604090451, + "loss": 0.1104, + "step": 41119 + }, + { + "epoch": 0.35694134599526045, + "grad_norm": 0.1689453125, + "learning_rate": 0.001317501880266013, + "loss": 0.0889, + "step": 41120 + }, + { + "epoch": 0.35695002647546464, + "grad_norm": 0.16015625, + "learning_rate": 0.0013174729998851242, + "loss": 0.0942, + "step": 41121 + }, + { + "epoch": 0.3569587069556688, + "grad_norm": 0.1474609375, + "learning_rate": 0.0013174441192664104, + "loss": 0.4062, + "step": 41122 + }, + { + "epoch": 0.356967387435873, + "grad_norm": 0.134765625, + "learning_rate": 0.0013174152384099027, + "loss": 0.0942, + "step": 41123 + }, + { + "epoch": 0.3569760679160771, + "grad_norm": 0.240234375, + "learning_rate": 0.0013173863573156328, + "loss": 0.1074, + "step": 41124 + }, + { + "epoch": 0.3569847483962813, + "grad_norm": 0.12890625, + "learning_rate": 0.0013173574759836328, + "loss": 0.0762, + "step": 41125 + }, + { + "epoch": 0.35699342887648544, + "grad_norm": 0.09375, + "learning_rate": 0.0013173285944139336, + "loss": 0.0972, + "step": 41126 + }, + { + "epoch": 0.35700210935668963, + "grad_norm": 0.1318359375, + "learning_rate": 0.0013172997126065676, + "loss": 0.1465, + "step": 41127 + }, + { + "epoch": 0.35701078983689377, + "grad_norm": 0.296875, + "learning_rate": 0.0013172708305615654, + "loss": 0.165, + "step": 41128 + }, + { + "epoch": 0.35701947031709796, + "grad_norm": 0.376953125, + "learning_rate": 0.0013172419482789595, + "loss": 0.0986, + "step": 41129 + }, + { + "epoch": 0.3570281507973021, + "grad_norm": 0.23828125, + "learning_rate": 0.0013172130657587808, + "loss": 0.1348, + "step": 41130 + }, + { + "epoch": 0.3570368312775063, + "grad_norm": 0.244140625, + "learning_rate": 0.001317184183001061, + "loss": 0.105, + "step": 41131 + }, + { + "epoch": 0.35704551175771043, + "grad_norm": 0.283203125, + "learning_rate": 0.0013171553000058322, + "loss": 0.0972, + "step": 41132 + }, + { + "epoch": 0.3570541922379146, + "grad_norm": 0.380859375, + "learning_rate": 0.0013171264167731256, + "loss": 0.1025, + "step": 41133 + }, + { + "epoch": 0.35706287271811876, + "grad_norm": 0.7734375, + "learning_rate": 0.0013170975333029727, + "loss": 0.3789, + "step": 41134 + }, + { + "epoch": 0.35707155319832296, + "grad_norm": 0.3125, + "learning_rate": 0.0013170686495954053, + "loss": 0.125, + "step": 41135 + }, + { + "epoch": 0.3570802336785271, + "grad_norm": 0.19140625, + "learning_rate": 0.0013170397656504544, + "loss": 0.0869, + "step": 41136 + }, + { + "epoch": 0.3570889141587313, + "grad_norm": 0.41015625, + "learning_rate": 0.0013170108814681526, + "loss": 0.127, + "step": 41137 + }, + { + "epoch": 0.3570975946389354, + "grad_norm": 0.37109375, + "learning_rate": 0.0013169819970485308, + "loss": 0.1006, + "step": 41138 + }, + { + "epoch": 0.3571062751191396, + "grad_norm": 0.1513671875, + "learning_rate": 0.001316953112391621, + "loss": 0.0908, + "step": 41139 + }, + { + "epoch": 0.35711495559934375, + "grad_norm": 0.0966796875, + "learning_rate": 0.001316924227497454, + "loss": 0.0723, + "step": 41140 + }, + { + "epoch": 0.35712363607954795, + "grad_norm": 0.11962890625, + "learning_rate": 0.0013168953423660624, + "loss": 0.1035, + "step": 41141 + }, + { + "epoch": 0.3571323165597521, + "grad_norm": 0.130859375, + "learning_rate": 0.001316866456997477, + "loss": 0.1094, + "step": 41142 + }, + { + "epoch": 0.3571409970399563, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013168375713917297, + "loss": 0.1494, + "step": 41143 + }, + { + "epoch": 0.3571496775201604, + "grad_norm": 0.15625, + "learning_rate": 0.001316808685548852, + "loss": 0.0898, + "step": 41144 + }, + { + "epoch": 0.3571583580003646, + "grad_norm": 0.158203125, + "learning_rate": 0.001316779799468876, + "loss": 0.126, + "step": 41145 + }, + { + "epoch": 0.35716703848056874, + "grad_norm": 0.15234375, + "learning_rate": 0.001316750913151833, + "loss": 0.1162, + "step": 41146 + }, + { + "epoch": 0.35717571896077294, + "grad_norm": 0.08984375, + "learning_rate": 0.001316722026597754, + "loss": 0.0786, + "step": 41147 + }, + { + "epoch": 0.3571843994409771, + "grad_norm": 0.61328125, + "learning_rate": 0.0013166931398066713, + "loss": 0.0884, + "step": 41148 + }, + { + "epoch": 0.35719307992118127, + "grad_norm": 0.6640625, + "learning_rate": 0.001316664252778616, + "loss": 0.1064, + "step": 41149 + }, + { + "epoch": 0.3572017604013854, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013166353655136198, + "loss": 0.0928, + "step": 41150 + }, + { + "epoch": 0.3572104408815896, + "grad_norm": 0.921875, + "learning_rate": 0.0013166064780117152, + "loss": 0.2695, + "step": 41151 + }, + { + "epoch": 0.35721912136179373, + "grad_norm": 0.7421875, + "learning_rate": 0.0013165775902729323, + "loss": 0.0845, + "step": 41152 + }, + { + "epoch": 0.35722780184199787, + "grad_norm": 0.330078125, + "learning_rate": 0.0013165487022973033, + "loss": 0.0869, + "step": 41153 + }, + { + "epoch": 0.35723648232220206, + "grad_norm": 0.2333984375, + "learning_rate": 0.0013165198140848602, + "loss": 0.1348, + "step": 41154 + }, + { + "epoch": 0.3572451628024062, + "grad_norm": 0.09033203125, + "learning_rate": 0.0013164909256356344, + "loss": 0.082, + "step": 41155 + }, + { + "epoch": 0.3572538432826104, + "grad_norm": 0.302734375, + "learning_rate": 0.001316462036949657, + "loss": 0.0688, + "step": 41156 + }, + { + "epoch": 0.35726252376281453, + "grad_norm": 0.5546875, + "learning_rate": 0.0013164331480269601, + "loss": 0.0684, + "step": 41157 + }, + { + "epoch": 0.3572712042430187, + "grad_norm": 0.11376953125, + "learning_rate": 0.0013164042588675756, + "loss": 0.1113, + "step": 41158 + }, + { + "epoch": 0.35727988472322286, + "grad_norm": 0.1318359375, + "learning_rate": 0.0013163753694715342, + "loss": 0.0928, + "step": 41159 + }, + { + "epoch": 0.35728856520342706, + "grad_norm": 0.107421875, + "learning_rate": 0.001316346479838868, + "loss": 0.0908, + "step": 41160 + }, + { + "epoch": 0.3572972456836312, + "grad_norm": 0.458984375, + "learning_rate": 0.001316317589969609, + "loss": 0.1177, + "step": 41161 + }, + { + "epoch": 0.3573059261638354, + "grad_norm": 1.6953125, + "learning_rate": 0.0013162886998637875, + "loss": 0.1582, + "step": 41162 + }, + { + "epoch": 0.3573146066440395, + "grad_norm": 0.220703125, + "learning_rate": 0.0013162598095214364, + "loss": 0.124, + "step": 41163 + }, + { + "epoch": 0.3573232871242437, + "grad_norm": 0.2890625, + "learning_rate": 0.0013162309189425868, + "loss": 0.1279, + "step": 41164 + }, + { + "epoch": 0.35733196760444785, + "grad_norm": 0.8515625, + "learning_rate": 0.00131620202812727, + "loss": 0.1143, + "step": 41165 + }, + { + "epoch": 0.35734064808465205, + "grad_norm": 0.09619140625, + "learning_rate": 0.0013161731370755186, + "loss": 0.1064, + "step": 41166 + }, + { + "epoch": 0.3573493285648562, + "grad_norm": 0.609375, + "learning_rate": 0.001316144245787363, + "loss": 0.1436, + "step": 41167 + }, + { + "epoch": 0.3573580090450604, + "grad_norm": 0.365234375, + "learning_rate": 0.0013161153542628355, + "loss": 0.082, + "step": 41168 + }, + { + "epoch": 0.3573666895252645, + "grad_norm": 0.66015625, + "learning_rate": 0.0013160864625019674, + "loss": 0.0947, + "step": 41169 + }, + { + "epoch": 0.3573753700054687, + "grad_norm": 0.0693359375, + "learning_rate": 0.0013160575705047904, + "loss": 0.0898, + "step": 41170 + }, + { + "epoch": 0.35738405048567284, + "grad_norm": 0.326171875, + "learning_rate": 0.001316028678271336, + "loss": 0.1309, + "step": 41171 + }, + { + "epoch": 0.35739273096587704, + "grad_norm": 0.15625, + "learning_rate": 0.001315999785801636, + "loss": 0.124, + "step": 41172 + }, + { + "epoch": 0.3574014114460812, + "grad_norm": 0.14453125, + "learning_rate": 0.0013159708930957218, + "loss": 0.103, + "step": 41173 + }, + { + "epoch": 0.35741009192628537, + "grad_norm": 0.373046875, + "learning_rate": 0.001315942000153625, + "loss": 0.1113, + "step": 41174 + }, + { + "epoch": 0.3574187724064895, + "grad_norm": 0.173828125, + "learning_rate": 0.0013159131069753773, + "loss": 0.1318, + "step": 41175 + }, + { + "epoch": 0.3574274528866937, + "grad_norm": 0.08203125, + "learning_rate": 0.0013158842135610105, + "loss": 0.082, + "step": 41176 + }, + { + "epoch": 0.35743613336689783, + "grad_norm": 0.1171875, + "learning_rate": 0.001315855319910556, + "loss": 0.1055, + "step": 41177 + }, + { + "epoch": 0.357444813847102, + "grad_norm": 0.1318359375, + "learning_rate": 0.0013158264260240454, + "loss": 0.1289, + "step": 41178 + }, + { + "epoch": 0.35745349432730616, + "grad_norm": 0.30859375, + "learning_rate": 0.0013157975319015097, + "loss": 0.125, + "step": 41179 + }, + { + "epoch": 0.35746217480751036, + "grad_norm": 0.3671875, + "learning_rate": 0.0013157686375429817, + "loss": 0.1377, + "step": 41180 + }, + { + "epoch": 0.3574708552877145, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013157397429484918, + "loss": 0.1011, + "step": 41181 + }, + { + "epoch": 0.3574795357679187, + "grad_norm": 0.1279296875, + "learning_rate": 0.0013157108481180724, + "loss": 0.084, + "step": 41182 + }, + { + "epoch": 0.3574882162481228, + "grad_norm": 2.15625, + "learning_rate": 0.0013156819530517547, + "loss": 0.3945, + "step": 41183 + }, + { + "epoch": 0.357496896728327, + "grad_norm": 0.138671875, + "learning_rate": 0.0013156530577495711, + "loss": 0.1328, + "step": 41184 + }, + { + "epoch": 0.35750557720853116, + "grad_norm": 0.43359375, + "learning_rate": 0.001315624162211552, + "loss": 0.1377, + "step": 41185 + }, + { + "epoch": 0.35751425768873535, + "grad_norm": 0.416015625, + "learning_rate": 0.0013155952664377297, + "loss": 0.0996, + "step": 41186 + }, + { + "epoch": 0.3575229381689395, + "grad_norm": 0.6796875, + "learning_rate": 0.0013155663704281359, + "loss": 0.0874, + "step": 41187 + }, + { + "epoch": 0.3575316186491437, + "grad_norm": 0.357421875, + "learning_rate": 0.0013155374741828014, + "loss": 0.1406, + "step": 41188 + }, + { + "epoch": 0.3575402991293478, + "grad_norm": 0.439453125, + "learning_rate": 0.0013155085777017586, + "loss": 0.1021, + "step": 41189 + }, + { + "epoch": 0.357548979609552, + "grad_norm": 0.6953125, + "learning_rate": 0.001315479680985039, + "loss": 0.2051, + "step": 41190 + }, + { + "epoch": 0.35755766008975615, + "grad_norm": 0.255859375, + "learning_rate": 0.001315450784032674, + "loss": 0.0601, + "step": 41191 + }, + { + "epoch": 0.35756634056996034, + "grad_norm": 0.1962890625, + "learning_rate": 0.001315421886844695, + "loss": 0.0869, + "step": 41192 + }, + { + "epoch": 0.3575750210501645, + "grad_norm": 0.1494140625, + "learning_rate": 0.001315392989421134, + "loss": 0.0864, + "step": 41193 + }, + { + "epoch": 0.35758370153036867, + "grad_norm": 0.1884765625, + "learning_rate": 0.001315364091762023, + "loss": 0.1172, + "step": 41194 + }, + { + "epoch": 0.3575923820105728, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013153351938673924, + "loss": 0.082, + "step": 41195 + }, + { + "epoch": 0.357601062490777, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013153062957372745, + "loss": 0.1162, + "step": 41196 + }, + { + "epoch": 0.35760974297098114, + "grad_norm": 0.345703125, + "learning_rate": 0.0013152773973717012, + "loss": 0.0654, + "step": 41197 + }, + { + "epoch": 0.35761842345118533, + "grad_norm": 0.09326171875, + "learning_rate": 0.0013152484987707035, + "loss": 0.1191, + "step": 41198 + }, + { + "epoch": 0.35762710393138947, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013152195999343134, + "loss": 0.0771, + "step": 41199 + }, + { + "epoch": 0.35763578441159366, + "grad_norm": 2.609375, + "learning_rate": 0.0013151907008625626, + "loss": 0.2383, + "step": 41200 + }, + { + "epoch": 0.3576444648917978, + "grad_norm": 0.46484375, + "learning_rate": 0.0013151618015554817, + "loss": 0.1162, + "step": 41201 + }, + { + "epoch": 0.357653145372002, + "grad_norm": 0.515625, + "learning_rate": 0.001315132902013104, + "loss": 0.0986, + "step": 41202 + }, + { + "epoch": 0.35766182585220613, + "grad_norm": 0.2060546875, + "learning_rate": 0.0013151040022354597, + "loss": 0.1104, + "step": 41203 + }, + { + "epoch": 0.3576705063324103, + "grad_norm": 0.33203125, + "learning_rate": 0.0013150751022225813, + "loss": 0.082, + "step": 41204 + }, + { + "epoch": 0.35767918681261446, + "grad_norm": 0.396484375, + "learning_rate": 0.0013150462019744995, + "loss": 0.1289, + "step": 41205 + }, + { + "epoch": 0.35768786729281865, + "grad_norm": 0.150390625, + "learning_rate": 0.0013150173014912468, + "loss": 0.0864, + "step": 41206 + }, + { + "epoch": 0.3576965477730228, + "grad_norm": 0.6796875, + "learning_rate": 0.0013149884007728544, + "loss": 0.1084, + "step": 41207 + }, + { + "epoch": 0.357705228253227, + "grad_norm": 0.22265625, + "learning_rate": 0.0013149594998193538, + "loss": 0.0923, + "step": 41208 + }, + { + "epoch": 0.3577139087334311, + "grad_norm": 0.6796875, + "learning_rate": 0.0013149305986307762, + "loss": 0.125, + "step": 41209 + }, + { + "epoch": 0.3577225892136353, + "grad_norm": 0.27734375, + "learning_rate": 0.0013149016972071543, + "loss": 0.1113, + "step": 41210 + }, + { + "epoch": 0.35773126969383945, + "grad_norm": 0.349609375, + "learning_rate": 0.0013148727955485191, + "loss": 0.1328, + "step": 41211 + }, + { + "epoch": 0.35773995017404364, + "grad_norm": 0.2890625, + "learning_rate": 0.0013148438936549023, + "loss": 0.1035, + "step": 41212 + }, + { + "epoch": 0.3577486306542478, + "grad_norm": 0.345703125, + "learning_rate": 0.0013148149915263353, + "loss": 0.1221, + "step": 41213 + }, + { + "epoch": 0.35775731113445197, + "grad_norm": 0.208984375, + "learning_rate": 0.0013147860891628497, + "loss": 0.0923, + "step": 41214 + }, + { + "epoch": 0.3577659916146561, + "grad_norm": 0.12353515625, + "learning_rate": 0.0013147571865644776, + "loss": 0.0776, + "step": 41215 + }, + { + "epoch": 0.3577746720948603, + "grad_norm": 0.62890625, + "learning_rate": 0.0013147282837312501, + "loss": 0.1191, + "step": 41216 + }, + { + "epoch": 0.35778335257506444, + "grad_norm": 0.322265625, + "learning_rate": 0.001314699380663199, + "loss": 0.1016, + "step": 41217 + }, + { + "epoch": 0.35779203305526863, + "grad_norm": 0.73828125, + "learning_rate": 0.0013146704773603557, + "loss": 0.1182, + "step": 41218 + }, + { + "epoch": 0.35780071353547277, + "grad_norm": 0.36328125, + "learning_rate": 0.0013146415738227522, + "loss": 0.0854, + "step": 41219 + }, + { + "epoch": 0.35780939401567696, + "grad_norm": 0.0966796875, + "learning_rate": 0.00131461267005042, + "loss": 0.0825, + "step": 41220 + }, + { + "epoch": 0.3578180744958811, + "grad_norm": 0.30078125, + "learning_rate": 0.0013145837660433902, + "loss": 0.1011, + "step": 41221 + }, + { + "epoch": 0.3578267549760853, + "grad_norm": 0.291015625, + "learning_rate": 0.0013145548618016952, + "loss": 0.1079, + "step": 41222 + }, + { + "epoch": 0.35783543545628943, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013145259573253659, + "loss": 0.0923, + "step": 41223 + }, + { + "epoch": 0.3578441159364936, + "grad_norm": 0.16015625, + "learning_rate": 0.0013144970526144347, + "loss": 0.0801, + "step": 41224 + }, + { + "epoch": 0.35785279641669776, + "grad_norm": 0.3203125, + "learning_rate": 0.0013144681476689325, + "loss": 0.106, + "step": 41225 + }, + { + "epoch": 0.35786147689690195, + "grad_norm": 0.318359375, + "learning_rate": 0.001314439242488891, + "loss": 0.0933, + "step": 41226 + }, + { + "epoch": 0.3578701573771061, + "grad_norm": 0.15234375, + "learning_rate": 0.0013144103370743424, + "loss": 0.1309, + "step": 41227 + }, + { + "epoch": 0.3578788378573103, + "grad_norm": 0.5390625, + "learning_rate": 0.0013143814314253174, + "loss": 0.127, + "step": 41228 + }, + { + "epoch": 0.3578875183375144, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013143525255418483, + "loss": 0.0996, + "step": 41229 + }, + { + "epoch": 0.3578961988177186, + "grad_norm": 0.388671875, + "learning_rate": 0.0013143236194239664, + "loss": 0.0977, + "step": 41230 + }, + { + "epoch": 0.35790487929792275, + "grad_norm": 0.0966796875, + "learning_rate": 0.0013142947130717034, + "loss": 0.1074, + "step": 41231 + }, + { + "epoch": 0.35791355977812694, + "grad_norm": 0.181640625, + "learning_rate": 0.001314265806485091, + "loss": 0.166, + "step": 41232 + }, + { + "epoch": 0.3579222402583311, + "grad_norm": 0.3515625, + "learning_rate": 0.0013142368996641608, + "loss": 0.0776, + "step": 41233 + }, + { + "epoch": 0.3579309207385353, + "grad_norm": 0.64453125, + "learning_rate": 0.0013142079926089442, + "loss": 0.0991, + "step": 41234 + }, + { + "epoch": 0.3579396012187394, + "grad_norm": 0.1923828125, + "learning_rate": 0.001314179085319473, + "loss": 0.1006, + "step": 41235 + }, + { + "epoch": 0.3579482816989436, + "grad_norm": 0.1298828125, + "learning_rate": 0.001314150177795779, + "loss": 0.1162, + "step": 41236 + }, + { + "epoch": 0.35795696217914774, + "grad_norm": 0.2109375, + "learning_rate": 0.0013141212700378932, + "loss": 0.1021, + "step": 41237 + }, + { + "epoch": 0.35796564265935193, + "grad_norm": 0.609375, + "learning_rate": 0.001314092362045848, + "loss": 0.1006, + "step": 41238 + }, + { + "epoch": 0.35797432313955607, + "grad_norm": 0.27734375, + "learning_rate": 0.0013140634538196742, + "loss": 0.1021, + "step": 41239 + }, + { + "epoch": 0.35798300361976026, + "grad_norm": 0.42578125, + "learning_rate": 0.001314034545359404, + "loss": 0.1064, + "step": 41240 + }, + { + "epoch": 0.3579916840999644, + "grad_norm": 0.53515625, + "learning_rate": 0.001314005636665069, + "loss": 0.0869, + "step": 41241 + }, + { + "epoch": 0.3580003645801686, + "grad_norm": 0.458984375, + "learning_rate": 0.0013139767277367004, + "loss": 0.1123, + "step": 41242 + }, + { + "epoch": 0.35800904506037273, + "grad_norm": 0.244140625, + "learning_rate": 0.0013139478185743303, + "loss": 0.1084, + "step": 41243 + }, + { + "epoch": 0.3580177255405769, + "grad_norm": 0.47265625, + "learning_rate": 0.00131391890917799, + "loss": 0.0884, + "step": 41244 + }, + { + "epoch": 0.35802640602078106, + "grad_norm": 0.38671875, + "learning_rate": 0.001313889999547711, + "loss": 0.1055, + "step": 41245 + }, + { + "epoch": 0.35803508650098526, + "grad_norm": 0.326171875, + "learning_rate": 0.0013138610896835255, + "loss": 0.082, + "step": 41246 + }, + { + "epoch": 0.3580437669811894, + "grad_norm": 0.53125, + "learning_rate": 0.0013138321795854644, + "loss": 0.1025, + "step": 41247 + }, + { + "epoch": 0.3580524474613936, + "grad_norm": 0.1640625, + "learning_rate": 0.0013138032692535597, + "loss": 0.1221, + "step": 41248 + }, + { + "epoch": 0.3580611279415977, + "grad_norm": 1.328125, + "learning_rate": 0.0013137743586878429, + "loss": 0.2695, + "step": 41249 + }, + { + "epoch": 0.3580698084218019, + "grad_norm": 0.39453125, + "learning_rate": 0.0013137454478883456, + "loss": 0.1367, + "step": 41250 + }, + { + "epoch": 0.35807848890200605, + "grad_norm": 0.1240234375, + "learning_rate": 0.0013137165368550998, + "loss": 0.0977, + "step": 41251 + }, + { + "epoch": 0.35808716938221025, + "grad_norm": 0.41796875, + "learning_rate": 0.0013136876255881364, + "loss": 0.1055, + "step": 41252 + }, + { + "epoch": 0.3580958498624144, + "grad_norm": 0.2119140625, + "learning_rate": 0.001313658714087488, + "loss": 0.0952, + "step": 41253 + }, + { + "epoch": 0.3581045303426186, + "grad_norm": 0.51171875, + "learning_rate": 0.0013136298023531854, + "loss": 0.1211, + "step": 41254 + }, + { + "epoch": 0.3581132108228227, + "grad_norm": 0.1650390625, + "learning_rate": 0.0013136008903852604, + "loss": 0.1016, + "step": 41255 + }, + { + "epoch": 0.3581218913030269, + "grad_norm": 0.11083984375, + "learning_rate": 0.0013135719781837446, + "loss": 0.0898, + "step": 41256 + }, + { + "epoch": 0.35813057178323104, + "grad_norm": 0.283203125, + "learning_rate": 0.0013135430657486694, + "loss": 0.0776, + "step": 41257 + }, + { + "epoch": 0.35813925226343524, + "grad_norm": 0.1220703125, + "learning_rate": 0.0013135141530800673, + "loss": 0.1025, + "step": 41258 + }, + { + "epoch": 0.3581479327436394, + "grad_norm": 0.25, + "learning_rate": 0.0013134852401779692, + "loss": 0.1157, + "step": 41259 + }, + { + "epoch": 0.35815661322384357, + "grad_norm": 0.07861328125, + "learning_rate": 0.0013134563270424068, + "loss": 0.0776, + "step": 41260 + }, + { + "epoch": 0.3581652937040477, + "grad_norm": 0.26953125, + "learning_rate": 0.0013134274136734115, + "loss": 0.0908, + "step": 41261 + }, + { + "epoch": 0.3581739741842519, + "grad_norm": 1.40625, + "learning_rate": 0.0013133985000710153, + "loss": 0.2148, + "step": 41262 + }, + { + "epoch": 0.35818265466445603, + "grad_norm": 0.1015625, + "learning_rate": 0.00131336958623525, + "loss": 0.1006, + "step": 41263 + }, + { + "epoch": 0.35819133514466023, + "grad_norm": 0.07421875, + "learning_rate": 0.0013133406721661468, + "loss": 0.0986, + "step": 41264 + }, + { + "epoch": 0.35820001562486437, + "grad_norm": 0.2236328125, + "learning_rate": 0.0013133117578637372, + "loss": 0.1025, + "step": 41265 + }, + { + "epoch": 0.35820869610506856, + "grad_norm": 0.212890625, + "learning_rate": 0.0013132828433280533, + "loss": 0.124, + "step": 41266 + }, + { + "epoch": 0.3582173765852727, + "grad_norm": 0.11865234375, + "learning_rate": 0.0013132539285591262, + "loss": 0.1025, + "step": 41267 + }, + { + "epoch": 0.3582260570654769, + "grad_norm": 0.09130859375, + "learning_rate": 0.0013132250135569881, + "loss": 0.1113, + "step": 41268 + }, + { + "epoch": 0.358234737545681, + "grad_norm": 0.2734375, + "learning_rate": 0.0013131960983216702, + "loss": 0.1211, + "step": 41269 + }, + { + "epoch": 0.3582434180258852, + "grad_norm": 0.5859375, + "learning_rate": 0.0013131671828532042, + "loss": 0.1357, + "step": 41270 + }, + { + "epoch": 0.35825209850608936, + "grad_norm": 0.36328125, + "learning_rate": 0.0013131382671516222, + "loss": 0.0918, + "step": 41271 + }, + { + "epoch": 0.35826077898629355, + "grad_norm": 0.296875, + "learning_rate": 0.001313109351216955, + "loss": 0.127, + "step": 41272 + }, + { + "epoch": 0.3582694594664977, + "grad_norm": 0.16015625, + "learning_rate": 0.0013130804350492346, + "loss": 0.1211, + "step": 41273 + }, + { + "epoch": 0.3582781399467019, + "grad_norm": 0.07861328125, + "learning_rate": 0.0013130515186484928, + "loss": 0.0771, + "step": 41274 + }, + { + "epoch": 0.358286820426906, + "grad_norm": 0.71484375, + "learning_rate": 0.0013130226020147608, + "loss": 0.0625, + "step": 41275 + }, + { + "epoch": 0.35829550090711015, + "grad_norm": 0.2216796875, + "learning_rate": 0.001312993685148071, + "loss": 0.1289, + "step": 41276 + }, + { + "epoch": 0.35830418138731435, + "grad_norm": 0.2412109375, + "learning_rate": 0.001312964768048454, + "loss": 0.1348, + "step": 41277 + }, + { + "epoch": 0.3583128618675185, + "grad_norm": 0.671875, + "learning_rate": 0.001312935850715942, + "loss": 0.1143, + "step": 41278 + }, + { + "epoch": 0.3583215423477227, + "grad_norm": 0.81640625, + "learning_rate": 0.0013129069331505668, + "loss": 0.1211, + "step": 41279 + }, + { + "epoch": 0.3583302228279268, + "grad_norm": 0.546875, + "learning_rate": 0.0013128780153523595, + "loss": 0.1104, + "step": 41280 + }, + { + "epoch": 0.358338903308131, + "grad_norm": 0.349609375, + "learning_rate": 0.0013128490973213522, + "loss": 0.1133, + "step": 41281 + }, + { + "epoch": 0.35834758378833514, + "grad_norm": 0.13671875, + "learning_rate": 0.0013128201790575764, + "loss": 0.082, + "step": 41282 + }, + { + "epoch": 0.35835626426853934, + "grad_norm": 0.078125, + "learning_rate": 0.0013127912605610634, + "loss": 0.0967, + "step": 41283 + }, + { + "epoch": 0.3583649447487435, + "grad_norm": 0.18359375, + "learning_rate": 0.0013127623418318453, + "loss": 0.1084, + "step": 41284 + }, + { + "epoch": 0.35837362522894767, + "grad_norm": 0.09375, + "learning_rate": 0.0013127334228699534, + "loss": 0.0811, + "step": 41285 + }, + { + "epoch": 0.3583823057091518, + "grad_norm": 0.224609375, + "learning_rate": 0.0013127045036754193, + "loss": 0.0811, + "step": 41286 + }, + { + "epoch": 0.358390986189356, + "grad_norm": 0.06640625, + "learning_rate": 0.0013126755842482749, + "loss": 0.0806, + "step": 41287 + }, + { + "epoch": 0.35839966666956014, + "grad_norm": 0.19921875, + "learning_rate": 0.0013126466645885515, + "loss": 0.0723, + "step": 41288 + }, + { + "epoch": 0.35840834714976433, + "grad_norm": 0.6328125, + "learning_rate": 0.001312617744696281, + "loss": 0.1201, + "step": 41289 + }, + { + "epoch": 0.35841702762996847, + "grad_norm": 0.13671875, + "learning_rate": 0.0013125888245714952, + "loss": 0.124, + "step": 41290 + }, + { + "epoch": 0.35842570811017266, + "grad_norm": 0.380859375, + "learning_rate": 0.001312559904214225, + "loss": 0.1182, + "step": 41291 + }, + { + "epoch": 0.3584343885903768, + "grad_norm": 0.1689453125, + "learning_rate": 0.001312530983624503, + "loss": 0.0879, + "step": 41292 + }, + { + "epoch": 0.358443069070581, + "grad_norm": 0.34375, + "learning_rate": 0.00131250206280236, + "loss": 0.0845, + "step": 41293 + }, + { + "epoch": 0.3584517495507851, + "grad_norm": 0.3359375, + "learning_rate": 0.0013124731417478279, + "loss": 0.126, + "step": 41294 + }, + { + "epoch": 0.3584604300309893, + "grad_norm": 0.474609375, + "learning_rate": 0.0013124442204609386, + "loss": 0.0947, + "step": 41295 + }, + { + "epoch": 0.35846911051119346, + "grad_norm": 0.103515625, + "learning_rate": 0.001312415298941723, + "loss": 0.1152, + "step": 41296 + }, + { + "epoch": 0.35847779099139765, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013123863771902135, + "loss": 0.1338, + "step": 41297 + }, + { + "epoch": 0.3584864714716018, + "grad_norm": 0.41796875, + "learning_rate": 0.0013123574552064417, + "loss": 0.1152, + "step": 41298 + }, + { + "epoch": 0.358495151951806, + "grad_norm": 0.33203125, + "learning_rate": 0.0013123285329904387, + "loss": 0.123, + "step": 41299 + }, + { + "epoch": 0.3585038324320101, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013122996105422362, + "loss": 0.084, + "step": 41300 + }, + { + "epoch": 0.3585125129122143, + "grad_norm": 0.154296875, + "learning_rate": 0.0013122706878618664, + "loss": 0.0898, + "step": 41301 + }, + { + "epoch": 0.35852119339241845, + "grad_norm": 0.173828125, + "learning_rate": 0.0013122417649493606, + "loss": 0.1118, + "step": 41302 + }, + { + "epoch": 0.35852987387262264, + "grad_norm": 0.1904296875, + "learning_rate": 0.0013122128418047504, + "loss": 0.0649, + "step": 41303 + }, + { + "epoch": 0.3585385543528268, + "grad_norm": 0.373046875, + "learning_rate": 0.0013121839184280673, + "loss": 0.0952, + "step": 41304 + }, + { + "epoch": 0.35854723483303097, + "grad_norm": 1.3359375, + "learning_rate": 0.001312154994819343, + "loss": 0.2168, + "step": 41305 + }, + { + "epoch": 0.3585559153132351, + "grad_norm": 0.22265625, + "learning_rate": 0.0013121260709786089, + "loss": 0.1133, + "step": 41306 + }, + { + "epoch": 0.3585645957934393, + "grad_norm": 0.298828125, + "learning_rate": 0.0013120971469058975, + "loss": 0.0742, + "step": 41307 + }, + { + "epoch": 0.35857327627364344, + "grad_norm": 0.310546875, + "learning_rate": 0.0013120682226012395, + "loss": 0.0869, + "step": 41308 + }, + { + "epoch": 0.35858195675384763, + "grad_norm": 0.181640625, + "learning_rate": 0.001312039298064667, + "loss": 0.0898, + "step": 41309 + }, + { + "epoch": 0.35859063723405177, + "grad_norm": 0.0751953125, + "learning_rate": 0.0013120103732962114, + "loss": 0.0908, + "step": 41310 + }, + { + "epoch": 0.35859931771425596, + "grad_norm": 0.392578125, + "learning_rate": 0.0013119814482959048, + "loss": 0.1836, + "step": 41311 + }, + { + "epoch": 0.3586079981944601, + "grad_norm": 0.0966796875, + "learning_rate": 0.0013119525230637782, + "loss": 0.085, + "step": 41312 + }, + { + "epoch": 0.3586166786746643, + "grad_norm": 0.40625, + "learning_rate": 0.0013119235975998633, + "loss": 0.1104, + "step": 41313 + }, + { + "epoch": 0.35862535915486843, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013118946719041923, + "loss": 0.1001, + "step": 41314 + }, + { + "epoch": 0.3586340396350726, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013118657459767961, + "loss": 0.1006, + "step": 41315 + }, + { + "epoch": 0.35864272011527676, + "grad_norm": 0.11279296875, + "learning_rate": 0.0013118368198177068, + "loss": 0.0762, + "step": 41316 + }, + { + "epoch": 0.35865140059548095, + "grad_norm": 0.267578125, + "learning_rate": 0.001311807893426956, + "loss": 0.1025, + "step": 41317 + }, + { + "epoch": 0.3586600810756851, + "grad_norm": 0.8984375, + "learning_rate": 0.0013117789668045754, + "loss": 0.168, + "step": 41318 + }, + { + "epoch": 0.3586687615558893, + "grad_norm": 0.279296875, + "learning_rate": 0.0013117500399505964, + "loss": 0.1367, + "step": 41319 + }, + { + "epoch": 0.3586774420360934, + "grad_norm": 0.310546875, + "learning_rate": 0.0013117211128650507, + "loss": 0.1201, + "step": 41320 + }, + { + "epoch": 0.3586861225162976, + "grad_norm": 0.6875, + "learning_rate": 0.0013116921855479704, + "loss": 0.1099, + "step": 41321 + }, + { + "epoch": 0.35869480299650175, + "grad_norm": 0.2470703125, + "learning_rate": 0.0013116632579993862, + "loss": 0.1328, + "step": 41322 + }, + { + "epoch": 0.35870348347670594, + "grad_norm": 0.22265625, + "learning_rate": 0.0013116343302193303, + "loss": 0.1387, + "step": 41323 + }, + { + "epoch": 0.3587121639569101, + "grad_norm": 0.1298828125, + "learning_rate": 0.0013116054022078345, + "loss": 0.0957, + "step": 41324 + }, + { + "epoch": 0.3587208444371143, + "grad_norm": 0.13671875, + "learning_rate": 0.0013115764739649299, + "loss": 0.0933, + "step": 41325 + }, + { + "epoch": 0.3587295249173184, + "grad_norm": 0.068359375, + "learning_rate": 0.0013115475454906485, + "loss": 0.0615, + "step": 41326 + }, + { + "epoch": 0.3587382053975226, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013115186167850222, + "loss": 0.0986, + "step": 41327 + }, + { + "epoch": 0.35874688587772674, + "grad_norm": 0.3203125, + "learning_rate": 0.001311489687848082, + "loss": 0.1133, + "step": 41328 + }, + { + "epoch": 0.35875556635793093, + "grad_norm": 0.33203125, + "learning_rate": 0.00131146075867986, + "loss": 0.1631, + "step": 41329 + }, + { + "epoch": 0.35876424683813507, + "grad_norm": 0.3125, + "learning_rate": 0.0013114318292803875, + "loss": 0.1357, + "step": 41330 + }, + { + "epoch": 0.35877292731833926, + "grad_norm": 0.1904296875, + "learning_rate": 0.0013114028996496965, + "loss": 0.1211, + "step": 41331 + }, + { + "epoch": 0.3587816077985434, + "grad_norm": 0.32421875, + "learning_rate": 0.0013113739697878186, + "loss": 0.0854, + "step": 41332 + }, + { + "epoch": 0.3587902882787476, + "grad_norm": 0.423828125, + "learning_rate": 0.0013113450396947852, + "loss": 0.1094, + "step": 41333 + }, + { + "epoch": 0.35879896875895173, + "grad_norm": 0.3203125, + "learning_rate": 0.0013113161093706279, + "loss": 0.0825, + "step": 41334 + }, + { + "epoch": 0.3588076492391559, + "grad_norm": 0.41015625, + "learning_rate": 0.0013112871788153787, + "loss": 0.1235, + "step": 41335 + }, + { + "epoch": 0.35881632971936006, + "grad_norm": 0.484375, + "learning_rate": 0.001311258248029069, + "loss": 0.0854, + "step": 41336 + }, + { + "epoch": 0.35882501019956425, + "grad_norm": 1.5546875, + "learning_rate": 0.0013112293170117301, + "loss": 0.1465, + "step": 41337 + }, + { + "epoch": 0.3588336906797684, + "grad_norm": 0.5390625, + "learning_rate": 0.0013112003857633943, + "loss": 0.0801, + "step": 41338 + }, + { + "epoch": 0.3588423711599726, + "grad_norm": 0.2236328125, + "learning_rate": 0.0013111714542840926, + "loss": 0.103, + "step": 41339 + }, + { + "epoch": 0.3588510516401767, + "grad_norm": 0.259765625, + "learning_rate": 0.0013111425225738576, + "loss": 0.1182, + "step": 41340 + }, + { + "epoch": 0.3588597321203809, + "grad_norm": 0.15234375, + "learning_rate": 0.0013111135906327199, + "loss": 0.0894, + "step": 41341 + }, + { + "epoch": 0.35886841260058505, + "grad_norm": 0.1630859375, + "learning_rate": 0.0013110846584607114, + "loss": 0.0801, + "step": 41342 + }, + { + "epoch": 0.35887709308078924, + "grad_norm": 0.6953125, + "learning_rate": 0.0013110557260578642, + "loss": 0.1387, + "step": 41343 + }, + { + "epoch": 0.3588857735609934, + "grad_norm": 0.09423828125, + "learning_rate": 0.0013110267934242094, + "loss": 0.0957, + "step": 41344 + }, + { + "epoch": 0.3588944540411976, + "grad_norm": 0.5390625, + "learning_rate": 0.001310997860559779, + "loss": 0.1309, + "step": 41345 + }, + { + "epoch": 0.3589031345214017, + "grad_norm": 0.50390625, + "learning_rate": 0.0013109689274646047, + "loss": 0.1406, + "step": 41346 + }, + { + "epoch": 0.3589118150016059, + "grad_norm": 0.53125, + "learning_rate": 0.0013109399941387176, + "loss": 0.0859, + "step": 41347 + }, + { + "epoch": 0.35892049548181004, + "grad_norm": 0.09716796875, + "learning_rate": 0.0013109110605821496, + "loss": 0.0928, + "step": 41348 + }, + { + "epoch": 0.35892917596201424, + "grad_norm": 0.1591796875, + "learning_rate": 0.0013108821267949328, + "loss": 0.0786, + "step": 41349 + }, + { + "epoch": 0.3589378564422184, + "grad_norm": 0.20703125, + "learning_rate": 0.0013108531927770986, + "loss": 0.1338, + "step": 41350 + }, + { + "epoch": 0.35894653692242257, + "grad_norm": 0.107421875, + "learning_rate": 0.0013108242585286782, + "loss": 0.1426, + "step": 41351 + }, + { + "epoch": 0.3589552174026267, + "grad_norm": 0.14453125, + "learning_rate": 0.0013107953240497035, + "loss": 0.0952, + "step": 41352 + }, + { + "epoch": 0.3589638978828309, + "grad_norm": 0.2314453125, + "learning_rate": 0.0013107663893402064, + "loss": 0.124, + "step": 41353 + }, + { + "epoch": 0.35897257836303503, + "grad_norm": 0.263671875, + "learning_rate": 0.0013107374544002182, + "loss": 0.0801, + "step": 41354 + }, + { + "epoch": 0.3589812588432392, + "grad_norm": 0.478515625, + "learning_rate": 0.0013107085192297708, + "loss": 0.0908, + "step": 41355 + }, + { + "epoch": 0.35898993932344336, + "grad_norm": 0.1064453125, + "learning_rate": 0.0013106795838288955, + "loss": 0.0977, + "step": 41356 + }, + { + "epoch": 0.35899861980364756, + "grad_norm": 0.1796875, + "learning_rate": 0.0013106506481976245, + "loss": 0.1104, + "step": 41357 + }, + { + "epoch": 0.3590073002838517, + "grad_norm": 0.4140625, + "learning_rate": 0.001310621712335989, + "loss": 0.1089, + "step": 41358 + }, + { + "epoch": 0.3590159807640559, + "grad_norm": 0.251953125, + "learning_rate": 0.0013105927762440208, + "loss": 0.1182, + "step": 41359 + }, + { + "epoch": 0.35902466124426, + "grad_norm": 0.126953125, + "learning_rate": 0.0013105638399217513, + "loss": 0.125, + "step": 41360 + }, + { + "epoch": 0.3590333417244642, + "grad_norm": 0.333984375, + "learning_rate": 0.0013105349033692127, + "loss": 0.0859, + "step": 41361 + }, + { + "epoch": 0.35904202220466835, + "grad_norm": 0.765625, + "learning_rate": 0.001310505966586436, + "loss": 0.0913, + "step": 41362 + }, + { + "epoch": 0.35905070268487255, + "grad_norm": 0.173828125, + "learning_rate": 0.0013104770295734535, + "loss": 0.0752, + "step": 41363 + }, + { + "epoch": 0.3590593831650767, + "grad_norm": 0.126953125, + "learning_rate": 0.0013104480923302956, + "loss": 0.0986, + "step": 41364 + }, + { + "epoch": 0.3590680636452809, + "grad_norm": 0.2578125, + "learning_rate": 0.0013104191548569954, + "loss": 0.126, + "step": 41365 + }, + { + "epoch": 0.359076744125485, + "grad_norm": 0.25390625, + "learning_rate": 0.0013103902171535842, + "loss": 0.0889, + "step": 41366 + }, + { + "epoch": 0.3590854246056892, + "grad_norm": 0.5859375, + "learning_rate": 0.0013103612792200931, + "loss": 0.1816, + "step": 41367 + }, + { + "epoch": 0.35909410508589334, + "grad_norm": 0.38671875, + "learning_rate": 0.0013103323410565545, + "loss": 0.0977, + "step": 41368 + }, + { + "epoch": 0.35910278556609754, + "grad_norm": 0.30859375, + "learning_rate": 0.001310303402662999, + "loss": 0.1299, + "step": 41369 + }, + { + "epoch": 0.3591114660463017, + "grad_norm": 0.1640625, + "learning_rate": 0.0013102744640394594, + "loss": 0.1211, + "step": 41370 + }, + { + "epoch": 0.35912014652650587, + "grad_norm": 0.3671875, + "learning_rate": 0.0013102455251859664, + "loss": 0.0752, + "step": 41371 + }, + { + "epoch": 0.35912882700671, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013102165861025522, + "loss": 0.1465, + "step": 41372 + }, + { + "epoch": 0.3591375074869142, + "grad_norm": 0.296875, + "learning_rate": 0.0013101876467892482, + "loss": 0.085, + "step": 41373 + }, + { + "epoch": 0.35914618796711834, + "grad_norm": 0.94140625, + "learning_rate": 0.0013101587072460862, + "loss": 0.1475, + "step": 41374 + }, + { + "epoch": 0.35915486844732253, + "grad_norm": 0.1337890625, + "learning_rate": 0.001310129767473098, + "loss": 0.1064, + "step": 41375 + }, + { + "epoch": 0.35916354892752667, + "grad_norm": 0.99609375, + "learning_rate": 0.001310100827470315, + "loss": 0.0933, + "step": 41376 + }, + { + "epoch": 0.35917222940773086, + "grad_norm": 0.2197265625, + "learning_rate": 0.0013100718872377688, + "loss": 0.1045, + "step": 41377 + }, + { + "epoch": 0.359180909887935, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013100429467754909, + "loss": 0.0913, + "step": 41378 + }, + { + "epoch": 0.3591895903681392, + "grad_norm": 0.248046875, + "learning_rate": 0.0013100140060835134, + "loss": 0.0981, + "step": 41379 + }, + { + "epoch": 0.3591982708483433, + "grad_norm": 0.2294921875, + "learning_rate": 0.001309985065161868, + "loss": 0.0933, + "step": 41380 + }, + { + "epoch": 0.3592069513285475, + "grad_norm": 0.0947265625, + "learning_rate": 0.001309956124010586, + "loss": 0.0938, + "step": 41381 + }, + { + "epoch": 0.35921563180875166, + "grad_norm": 0.47265625, + "learning_rate": 0.0013099271826296987, + "loss": 0.1001, + "step": 41382 + }, + { + "epoch": 0.35922431228895585, + "grad_norm": 0.54296875, + "learning_rate": 0.0013098982410192385, + "loss": 0.1377, + "step": 41383 + }, + { + "epoch": 0.35923299276916, + "grad_norm": 0.11083984375, + "learning_rate": 0.0013098692991792367, + "loss": 0.1309, + "step": 41384 + }, + { + "epoch": 0.3592416732493642, + "grad_norm": 0.302734375, + "learning_rate": 0.001309840357109725, + "loss": 0.1299, + "step": 41385 + }, + { + "epoch": 0.3592503537295683, + "grad_norm": 0.228515625, + "learning_rate": 0.001309811414810735, + "loss": 0.0957, + "step": 41386 + }, + { + "epoch": 0.3592590342097725, + "grad_norm": 0.345703125, + "learning_rate": 0.0013097824722822983, + "loss": 0.1357, + "step": 41387 + }, + { + "epoch": 0.35926771468997665, + "grad_norm": 0.173828125, + "learning_rate": 0.001309753529524447, + "loss": 0.0845, + "step": 41388 + }, + { + "epoch": 0.35927639517018084, + "grad_norm": 0.390625, + "learning_rate": 0.001309724586537212, + "loss": 0.0762, + "step": 41389 + }, + { + "epoch": 0.359285075650385, + "grad_norm": 0.25390625, + "learning_rate": 0.0013096956433206258, + "loss": 0.0977, + "step": 41390 + }, + { + "epoch": 0.35929375613058917, + "grad_norm": 0.0712890625, + "learning_rate": 0.001309666699874719, + "loss": 0.0679, + "step": 41391 + }, + { + "epoch": 0.3593024366107933, + "grad_norm": 0.30859375, + "learning_rate": 0.0013096377561995242, + "loss": 0.1045, + "step": 41392 + }, + { + "epoch": 0.3593111170909975, + "grad_norm": 0.103515625, + "learning_rate": 0.0013096088122950724, + "loss": 0.1133, + "step": 41393 + }, + { + "epoch": 0.35931979757120164, + "grad_norm": 0.6484375, + "learning_rate": 0.0013095798681613962, + "loss": 0.1924, + "step": 41394 + }, + { + "epoch": 0.35932847805140583, + "grad_norm": 0.1533203125, + "learning_rate": 0.001309550923798526, + "loss": 0.0815, + "step": 41395 + }, + { + "epoch": 0.35933715853160997, + "grad_norm": 0.341796875, + "learning_rate": 0.0013095219792064942, + "loss": 0.1367, + "step": 41396 + }, + { + "epoch": 0.35934583901181416, + "grad_norm": 0.15234375, + "learning_rate": 0.0013094930343853326, + "loss": 0.1094, + "step": 41397 + }, + { + "epoch": 0.3593545194920183, + "grad_norm": 0.234375, + "learning_rate": 0.0013094640893350723, + "loss": 0.0537, + "step": 41398 + }, + { + "epoch": 0.35936319997222244, + "grad_norm": 0.1484375, + "learning_rate": 0.001309435144055745, + "loss": 0.1396, + "step": 41399 + }, + { + "epoch": 0.35937188045242663, + "grad_norm": 0.1630859375, + "learning_rate": 0.001309406198547383, + "loss": 0.1211, + "step": 41400 + }, + { + "epoch": 0.35938056093263077, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013093772528100173, + "loss": 0.123, + "step": 41401 + }, + { + "epoch": 0.35938924141283496, + "grad_norm": 2.84375, + "learning_rate": 0.0013093483068436795, + "loss": 0.5703, + "step": 41402 + }, + { + "epoch": 0.3593979218930391, + "grad_norm": 0.58984375, + "learning_rate": 0.0013093193606484022, + "loss": 0.1201, + "step": 41403 + }, + { + "epoch": 0.3594066023732433, + "grad_norm": 0.408203125, + "learning_rate": 0.001309290414224216, + "loss": 0.0923, + "step": 41404 + }, + { + "epoch": 0.3594152828534474, + "grad_norm": 0.3828125, + "learning_rate": 0.001309261467571153, + "loss": 0.0996, + "step": 41405 + }, + { + "epoch": 0.3594239633336516, + "grad_norm": 0.205078125, + "learning_rate": 0.0013092325206892448, + "loss": 0.0806, + "step": 41406 + }, + { + "epoch": 0.35943264381385576, + "grad_norm": 1.96875, + "learning_rate": 0.001309203573578523, + "loss": 0.3047, + "step": 41407 + }, + { + "epoch": 0.35944132429405995, + "grad_norm": 5.0, + "learning_rate": 0.0013091746262390194, + "loss": 0.2598, + "step": 41408 + }, + { + "epoch": 0.3594500047742641, + "grad_norm": 0.455078125, + "learning_rate": 0.0013091456786707657, + "loss": 0.0708, + "step": 41409 + }, + { + "epoch": 0.3594586852544683, + "grad_norm": 0.09912109375, + "learning_rate": 0.0013091167308737935, + "loss": 0.1484, + "step": 41410 + }, + { + "epoch": 0.3594673657346724, + "grad_norm": 0.60546875, + "learning_rate": 0.0013090877828481338, + "loss": 0.1011, + "step": 41411 + }, + { + "epoch": 0.3594760462148766, + "grad_norm": 0.16015625, + "learning_rate": 0.0013090588345938193, + "loss": 0.1152, + "step": 41412 + }, + { + "epoch": 0.35948472669508075, + "grad_norm": 0.35546875, + "learning_rate": 0.0013090298861108808, + "loss": 0.1387, + "step": 41413 + }, + { + "epoch": 0.35949340717528494, + "grad_norm": 0.267578125, + "learning_rate": 0.0013090009373993507, + "loss": 0.0913, + "step": 41414 + }, + { + "epoch": 0.3595020876554891, + "grad_norm": 0.259765625, + "learning_rate": 0.0013089719884592605, + "loss": 0.0723, + "step": 41415 + }, + { + "epoch": 0.35951076813569327, + "grad_norm": 0.08642578125, + "learning_rate": 0.0013089430392906416, + "loss": 0.0928, + "step": 41416 + }, + { + "epoch": 0.3595194486158974, + "grad_norm": 0.369140625, + "learning_rate": 0.0013089140898935258, + "loss": 0.084, + "step": 41417 + }, + { + "epoch": 0.3595281290961016, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013088851402679443, + "loss": 0.124, + "step": 41418 + }, + { + "epoch": 0.35953680957630574, + "grad_norm": 0.1669921875, + "learning_rate": 0.0013088561904139294, + "loss": 0.1064, + "step": 41419 + }, + { + "epoch": 0.35954549005650993, + "grad_norm": 0.671875, + "learning_rate": 0.0013088272403315125, + "loss": 0.0913, + "step": 41420 + }, + { + "epoch": 0.35955417053671407, + "grad_norm": 0.1142578125, + "learning_rate": 0.001308798290020725, + "loss": 0.123, + "step": 41421 + }, + { + "epoch": 0.35956285101691826, + "grad_norm": 1.0078125, + "learning_rate": 0.0013087693394815989, + "loss": 0.0942, + "step": 41422 + }, + { + "epoch": 0.3595715314971224, + "grad_norm": 0.1337890625, + "learning_rate": 0.001308740388714166, + "loss": 0.126, + "step": 41423 + }, + { + "epoch": 0.3595802119773266, + "grad_norm": 0.3671875, + "learning_rate": 0.0013087114377184578, + "loss": 0.1196, + "step": 41424 + }, + { + "epoch": 0.35958889245753073, + "grad_norm": 0.17578125, + "learning_rate": 0.0013086824864945055, + "loss": 0.1055, + "step": 41425 + }, + { + "epoch": 0.3595975729377349, + "grad_norm": 0.2392578125, + "learning_rate": 0.0013086535350423417, + "loss": 0.0894, + "step": 41426 + }, + { + "epoch": 0.35960625341793906, + "grad_norm": 0.216796875, + "learning_rate": 0.0013086245833619971, + "loss": 0.1377, + "step": 41427 + }, + { + "epoch": 0.35961493389814325, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013085956314535041, + "loss": 0.0771, + "step": 41428 + }, + { + "epoch": 0.3596236143783474, + "grad_norm": 0.1767578125, + "learning_rate": 0.001308566679316894, + "loss": 0.0942, + "step": 41429 + }, + { + "epoch": 0.3596322948585516, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013085377269521983, + "loss": 0.0908, + "step": 41430 + }, + { + "epoch": 0.3596409753387557, + "grad_norm": 0.134765625, + "learning_rate": 0.0013085087743594489, + "loss": 0.1055, + "step": 41431 + }, + { + "epoch": 0.3596496558189599, + "grad_norm": 0.25390625, + "learning_rate": 0.0013084798215386776, + "loss": 0.0791, + "step": 41432 + }, + { + "epoch": 0.35965833629916405, + "grad_norm": 0.68359375, + "learning_rate": 0.001308450868489916, + "loss": 0.0928, + "step": 41433 + }, + { + "epoch": 0.35966701677936824, + "grad_norm": 0.70703125, + "learning_rate": 0.0013084219152131953, + "loss": 0.1221, + "step": 41434 + }, + { + "epoch": 0.3596756972595724, + "grad_norm": 0.2138671875, + "learning_rate": 0.0013083929617085477, + "loss": 0.0781, + "step": 41435 + }, + { + "epoch": 0.3596843777397766, + "grad_norm": 0.189453125, + "learning_rate": 0.0013083640079760049, + "loss": 0.123, + "step": 41436 + }, + { + "epoch": 0.3596930582199807, + "grad_norm": 0.16015625, + "learning_rate": 0.001308335054015598, + "loss": 0.124, + "step": 41437 + }, + { + "epoch": 0.3597017387001849, + "grad_norm": 0.099609375, + "learning_rate": 0.001308306099827359, + "loss": 0.0947, + "step": 41438 + }, + { + "epoch": 0.35971041918038904, + "grad_norm": 0.2734375, + "learning_rate": 0.00130827714541132, + "loss": 0.0742, + "step": 41439 + }, + { + "epoch": 0.35971909966059323, + "grad_norm": 0.39453125, + "learning_rate": 0.0013082481907675122, + "loss": 0.0918, + "step": 41440 + }, + { + "epoch": 0.35972778014079737, + "grad_norm": 0.177734375, + "learning_rate": 0.001308219235895967, + "loss": 0.166, + "step": 41441 + }, + { + "epoch": 0.35973646062100156, + "grad_norm": 1.0703125, + "learning_rate": 0.0013081902807967165, + "loss": 0.0781, + "step": 41442 + }, + { + "epoch": 0.3597451411012057, + "grad_norm": 0.1357421875, + "learning_rate": 0.001308161325469792, + "loss": 0.0938, + "step": 41443 + }, + { + "epoch": 0.3597538215814099, + "grad_norm": 0.13671875, + "learning_rate": 0.0013081323699152257, + "loss": 0.0659, + "step": 41444 + }, + { + "epoch": 0.35976250206161403, + "grad_norm": 0.11181640625, + "learning_rate": 0.0013081034141330487, + "loss": 0.1162, + "step": 41445 + }, + { + "epoch": 0.3597711825418182, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013080744581232934, + "loss": 0.0869, + "step": 41446 + }, + { + "epoch": 0.35977986302202236, + "grad_norm": 0.1640625, + "learning_rate": 0.0013080455018859908, + "loss": 0.0967, + "step": 41447 + }, + { + "epoch": 0.35978854350222655, + "grad_norm": 0.06787109375, + "learning_rate": 0.0013080165454211726, + "loss": 0.0898, + "step": 41448 + }, + { + "epoch": 0.3597972239824307, + "grad_norm": 0.2451171875, + "learning_rate": 0.0013079875887288708, + "loss": 0.1455, + "step": 41449 + }, + { + "epoch": 0.3598059044626349, + "grad_norm": 0.16015625, + "learning_rate": 0.0013079586318091169, + "loss": 0.0977, + "step": 41450 + }, + { + "epoch": 0.359814584942839, + "grad_norm": 0.2353515625, + "learning_rate": 0.001307929674661942, + "loss": 0.1172, + "step": 41451 + }, + { + "epoch": 0.3598232654230432, + "grad_norm": 0.3203125, + "learning_rate": 0.0013079007172873789, + "loss": 0.126, + "step": 41452 + }, + { + "epoch": 0.35983194590324735, + "grad_norm": 0.287109375, + "learning_rate": 0.0013078717596854586, + "loss": 0.1201, + "step": 41453 + }, + { + "epoch": 0.35984062638345155, + "grad_norm": 0.14453125, + "learning_rate": 0.0013078428018562132, + "loss": 0.166, + "step": 41454 + }, + { + "epoch": 0.3598493068636557, + "grad_norm": 0.28125, + "learning_rate": 0.0013078138437996737, + "loss": 0.0874, + "step": 41455 + }, + { + "epoch": 0.3598579873438599, + "grad_norm": 2.0625, + "learning_rate": 0.001307784885515872, + "loss": 0.125, + "step": 41456 + }, + { + "epoch": 0.359866667824064, + "grad_norm": 0.10009765625, + "learning_rate": 0.00130775592700484, + "loss": 0.123, + "step": 41457 + }, + { + "epoch": 0.3598753483042682, + "grad_norm": 0.212890625, + "learning_rate": 0.0013077269682666094, + "loss": 0.0933, + "step": 41458 + }, + { + "epoch": 0.35988402878447234, + "grad_norm": 0.21484375, + "learning_rate": 0.0013076980093012114, + "loss": 0.0928, + "step": 41459 + }, + { + "epoch": 0.35989270926467654, + "grad_norm": 0.48828125, + "learning_rate": 0.001307669050108678, + "loss": 0.123, + "step": 41460 + }, + { + "epoch": 0.3599013897448807, + "grad_norm": 0.330078125, + "learning_rate": 0.0013076400906890408, + "loss": 0.0713, + "step": 41461 + }, + { + "epoch": 0.35991007022508487, + "grad_norm": 0.76953125, + "learning_rate": 0.0013076111310423317, + "loss": 0.1138, + "step": 41462 + }, + { + "epoch": 0.359918750705289, + "grad_norm": 0.61328125, + "learning_rate": 0.0013075821711685822, + "loss": 0.0933, + "step": 41463 + }, + { + "epoch": 0.3599274311854932, + "grad_norm": 0.228515625, + "learning_rate": 0.0013075532110678242, + "loss": 0.1328, + "step": 41464 + }, + { + "epoch": 0.35993611166569733, + "grad_norm": 0.15625, + "learning_rate": 0.0013075242507400886, + "loss": 0.0781, + "step": 41465 + }, + { + "epoch": 0.3599447921459015, + "grad_norm": 0.16015625, + "learning_rate": 0.001307495290185408, + "loss": 0.0859, + "step": 41466 + }, + { + "epoch": 0.35995347262610566, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013074663294038134, + "loss": 0.1123, + "step": 41467 + }, + { + "epoch": 0.35996215310630986, + "grad_norm": 0.5390625, + "learning_rate": 0.001307437368395337, + "loss": 0.0986, + "step": 41468 + }, + { + "epoch": 0.359970833586514, + "grad_norm": 0.11376953125, + "learning_rate": 0.00130740840716001, + "loss": 0.0771, + "step": 41469 + }, + { + "epoch": 0.3599795140667182, + "grad_norm": 0.404296875, + "learning_rate": 0.0013073794456978643, + "loss": 0.082, + "step": 41470 + }, + { + "epoch": 0.3599881945469223, + "grad_norm": 0.51953125, + "learning_rate": 0.0013073504840089316, + "loss": 0.1328, + "step": 41471 + }, + { + "epoch": 0.3599968750271265, + "grad_norm": 0.1396484375, + "learning_rate": 0.0013073215220932436, + "loss": 0.1074, + "step": 41472 + }, + { + "epoch": 0.36000555550733065, + "grad_norm": 0.400390625, + "learning_rate": 0.0013072925599508316, + "loss": 0.0874, + "step": 41473 + }, + { + "epoch": 0.36001423598753485, + "grad_norm": 0.8125, + "learning_rate": 0.0013072635975817277, + "loss": 0.125, + "step": 41474 + }, + { + "epoch": 0.360022916467739, + "grad_norm": 0.337890625, + "learning_rate": 0.0013072346349859638, + "loss": 0.1099, + "step": 41475 + }, + { + "epoch": 0.3600315969479432, + "grad_norm": 0.130859375, + "learning_rate": 0.001307205672163571, + "loss": 0.1045, + "step": 41476 + }, + { + "epoch": 0.3600402774281473, + "grad_norm": 0.1376953125, + "learning_rate": 0.001307176709114581, + "loss": 0.0967, + "step": 41477 + }, + { + "epoch": 0.3600489579083515, + "grad_norm": 0.2412109375, + "learning_rate": 0.001307147745839026, + "loss": 0.0806, + "step": 41478 + }, + { + "epoch": 0.36005763838855565, + "grad_norm": 0.09619140625, + "learning_rate": 0.0013071187823369371, + "loss": 0.1162, + "step": 41479 + }, + { + "epoch": 0.36006631886875984, + "grad_norm": 0.6015625, + "learning_rate": 0.0013070898186083463, + "loss": 0.1113, + "step": 41480 + }, + { + "epoch": 0.360074999348964, + "grad_norm": 0.271484375, + "learning_rate": 0.001307060854653285, + "loss": 0.1172, + "step": 41481 + }, + { + "epoch": 0.36008367982916817, + "grad_norm": 0.1826171875, + "learning_rate": 0.001307031890471785, + "loss": 0.0884, + "step": 41482 + }, + { + "epoch": 0.3600923603093723, + "grad_norm": 0.158203125, + "learning_rate": 0.0013070029260638784, + "loss": 0.1396, + "step": 41483 + }, + { + "epoch": 0.3601010407895765, + "grad_norm": 0.11328125, + "learning_rate": 0.0013069739614295964, + "loss": 0.1074, + "step": 41484 + }, + { + "epoch": 0.36010972126978064, + "grad_norm": 0.1337890625, + "learning_rate": 0.0013069449965689708, + "loss": 0.1221, + "step": 41485 + }, + { + "epoch": 0.36011840174998483, + "grad_norm": 0.34375, + "learning_rate": 0.0013069160314820329, + "loss": 0.126, + "step": 41486 + }, + { + "epoch": 0.36012708223018897, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013068870661688153, + "loss": 0.1074, + "step": 41487 + }, + { + "epoch": 0.36013576271039316, + "grad_norm": 0.3046875, + "learning_rate": 0.001306858100629349, + "loss": 0.1328, + "step": 41488 + }, + { + "epoch": 0.3601444431905973, + "grad_norm": 0.4140625, + "learning_rate": 0.0013068291348636656, + "loss": 0.0903, + "step": 41489 + }, + { + "epoch": 0.3601531236708015, + "grad_norm": 0.2431640625, + "learning_rate": 0.0013068001688717966, + "loss": 0.127, + "step": 41490 + }, + { + "epoch": 0.3601618041510056, + "grad_norm": 0.26171875, + "learning_rate": 0.0013067712026537746, + "loss": 0.1245, + "step": 41491 + }, + { + "epoch": 0.3601704846312098, + "grad_norm": 0.2373046875, + "learning_rate": 0.0013067422362096307, + "loss": 0.1035, + "step": 41492 + }, + { + "epoch": 0.36017916511141396, + "grad_norm": 0.421875, + "learning_rate": 0.0013067132695393966, + "loss": 0.1562, + "step": 41493 + }, + { + "epoch": 0.36018784559161815, + "grad_norm": 0.07568359375, + "learning_rate": 0.0013066843026431035, + "loss": 0.1128, + "step": 41494 + }, + { + "epoch": 0.3601965260718223, + "grad_norm": 0.0986328125, + "learning_rate": 0.0013066553355207842, + "loss": 0.0752, + "step": 41495 + }, + { + "epoch": 0.3602052065520265, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013066263681724693, + "loss": 0.1016, + "step": 41496 + }, + { + "epoch": 0.3602138870322306, + "grad_norm": 0.1669921875, + "learning_rate": 0.001306597400598191, + "loss": 0.0928, + "step": 41497 + }, + { + "epoch": 0.3602225675124348, + "grad_norm": 0.47265625, + "learning_rate": 0.0013065684327979808, + "loss": 0.1143, + "step": 41498 + }, + { + "epoch": 0.36023124799263895, + "grad_norm": 0.50390625, + "learning_rate": 0.0013065394647718705, + "loss": 0.0791, + "step": 41499 + }, + { + "epoch": 0.36023992847284314, + "grad_norm": 0.142578125, + "learning_rate": 0.0013065104965198917, + "loss": 0.1797, + "step": 41500 + }, + { + "epoch": 0.3602486089530473, + "grad_norm": 0.2734375, + "learning_rate": 0.0013064815280420766, + "loss": 0.126, + "step": 41501 + }, + { + "epoch": 0.36025728943325147, + "grad_norm": 0.326171875, + "learning_rate": 0.001306452559338456, + "loss": 0.0796, + "step": 41502 + }, + { + "epoch": 0.3602659699134556, + "grad_norm": 0.5859375, + "learning_rate": 0.0013064235904090618, + "loss": 0.1143, + "step": 41503 + }, + { + "epoch": 0.3602746503936598, + "grad_norm": 0.25, + "learning_rate": 0.0013063946212539263, + "loss": 0.0854, + "step": 41504 + }, + { + "epoch": 0.36028333087386394, + "grad_norm": 0.3125, + "learning_rate": 0.0013063656518730806, + "loss": 0.1167, + "step": 41505 + }, + { + "epoch": 0.36029201135406813, + "grad_norm": 0.08203125, + "learning_rate": 0.0013063366822665566, + "loss": 0.0723, + "step": 41506 + }, + { + "epoch": 0.36030069183427227, + "grad_norm": 0.25390625, + "learning_rate": 0.0013063077124343856, + "loss": 0.0806, + "step": 41507 + }, + { + "epoch": 0.36030937231447646, + "grad_norm": 0.109375, + "learning_rate": 0.0013062787423765996, + "loss": 0.0923, + "step": 41508 + }, + { + "epoch": 0.3603180527946806, + "grad_norm": 0.2333984375, + "learning_rate": 0.0013062497720932305, + "loss": 0.0713, + "step": 41509 + }, + { + "epoch": 0.3603267332748848, + "grad_norm": 1.1015625, + "learning_rate": 0.0013062208015843096, + "loss": 0.0923, + "step": 41510 + }, + { + "epoch": 0.36033541375508893, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013061918308498689, + "loss": 0.1006, + "step": 41511 + }, + { + "epoch": 0.3603440942352931, + "grad_norm": 0.353515625, + "learning_rate": 0.0013061628598899397, + "loss": 0.0952, + "step": 41512 + }, + { + "epoch": 0.36035277471549726, + "grad_norm": 0.484375, + "learning_rate": 0.0013061338887045539, + "loss": 0.0845, + "step": 41513 + }, + { + "epoch": 0.36036145519570145, + "grad_norm": 0.1103515625, + "learning_rate": 0.0013061049172937436, + "loss": 0.082, + "step": 41514 + }, + { + "epoch": 0.3603701356759056, + "grad_norm": 0.56640625, + "learning_rate": 0.0013060759456575395, + "loss": 0.0938, + "step": 41515 + }, + { + "epoch": 0.3603788161561098, + "grad_norm": 0.2236328125, + "learning_rate": 0.0013060469737959741, + "loss": 0.1328, + "step": 41516 + }, + { + "epoch": 0.3603874966363139, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013060180017090788, + "loss": 0.127, + "step": 41517 + }, + { + "epoch": 0.3603961771165181, + "grad_norm": 0.1201171875, + "learning_rate": 0.0013059890293968853, + "loss": 0.127, + "step": 41518 + }, + { + "epoch": 0.36040485759672225, + "grad_norm": 0.26953125, + "learning_rate": 0.0013059600568594255, + "loss": 0.0835, + "step": 41519 + }, + { + "epoch": 0.36041353807692644, + "grad_norm": 0.5234375, + "learning_rate": 0.0013059310840967304, + "loss": 0.123, + "step": 41520 + }, + { + "epoch": 0.3604222185571306, + "grad_norm": 0.1611328125, + "learning_rate": 0.0013059021111088325, + "loss": 0.082, + "step": 41521 + }, + { + "epoch": 0.3604308990373347, + "grad_norm": 0.07080078125, + "learning_rate": 0.001305873137895763, + "loss": 0.0942, + "step": 41522 + }, + { + "epoch": 0.3604395795175389, + "grad_norm": 0.158203125, + "learning_rate": 0.0013058441644575541, + "loss": 0.0947, + "step": 41523 + }, + { + "epoch": 0.36044825999774305, + "grad_norm": 0.357421875, + "learning_rate": 0.0013058151907942368, + "loss": 0.1143, + "step": 41524 + }, + { + "epoch": 0.36045694047794724, + "grad_norm": 0.359375, + "learning_rate": 0.001305786216905843, + "loss": 0.083, + "step": 41525 + }, + { + "epoch": 0.3604656209581514, + "grad_norm": 0.103515625, + "learning_rate": 0.0013057572427924047, + "loss": 0.1216, + "step": 41526 + }, + { + "epoch": 0.36047430143835557, + "grad_norm": 0.279296875, + "learning_rate": 0.001305728268453953, + "loss": 0.1094, + "step": 41527 + }, + { + "epoch": 0.3604829819185597, + "grad_norm": 0.119140625, + "learning_rate": 0.0013056992938905204, + "loss": 0.0825, + "step": 41528 + }, + { + "epoch": 0.3604916623987639, + "grad_norm": 0.3984375, + "learning_rate": 0.001305670319102138, + "loss": 0.1221, + "step": 41529 + }, + { + "epoch": 0.36050034287896804, + "grad_norm": 0.1591796875, + "learning_rate": 0.0013056413440888377, + "loss": 0.1016, + "step": 41530 + }, + { + "epoch": 0.36050902335917223, + "grad_norm": 0.1689453125, + "learning_rate": 0.001305612368850651, + "loss": 0.0879, + "step": 41531 + }, + { + "epoch": 0.36051770383937637, + "grad_norm": 0.6640625, + "learning_rate": 0.0013055833933876097, + "loss": 0.1084, + "step": 41532 + }, + { + "epoch": 0.36052638431958056, + "grad_norm": 3.03125, + "learning_rate": 0.0013055544176997457, + "loss": 0.4355, + "step": 41533 + }, + { + "epoch": 0.3605350647997847, + "grad_norm": 0.3828125, + "learning_rate": 0.0013055254417870903, + "loss": 0.1055, + "step": 41534 + }, + { + "epoch": 0.3605437452799889, + "grad_norm": 0.2216796875, + "learning_rate": 0.0013054964656496752, + "loss": 0.1162, + "step": 41535 + }, + { + "epoch": 0.36055242576019303, + "grad_norm": 0.40625, + "learning_rate": 0.0013054674892875324, + "loss": 0.127, + "step": 41536 + }, + { + "epoch": 0.3605611062403972, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013054385127006934, + "loss": 0.1484, + "step": 41537 + }, + { + "epoch": 0.36056978672060136, + "grad_norm": 0.1982421875, + "learning_rate": 0.00130540953588919, + "loss": 0.0938, + "step": 41538 + }, + { + "epoch": 0.36057846720080555, + "grad_norm": 0.35546875, + "learning_rate": 0.0013053805588530538, + "loss": 0.0781, + "step": 41539 + }, + { + "epoch": 0.3605871476810097, + "grad_norm": 0.9375, + "learning_rate": 0.0013053515815923168, + "loss": 0.0986, + "step": 41540 + }, + { + "epoch": 0.3605958281612139, + "grad_norm": 0.43359375, + "learning_rate": 0.0013053226041070101, + "loss": 0.1084, + "step": 41541 + }, + { + "epoch": 0.360604508641418, + "grad_norm": 0.0859375, + "learning_rate": 0.0013052936263971654, + "loss": 0.0908, + "step": 41542 + }, + { + "epoch": 0.3606131891216222, + "grad_norm": 0.1337890625, + "learning_rate": 0.001305264648462815, + "loss": 0.0806, + "step": 41543 + }, + { + "epoch": 0.36062186960182635, + "grad_norm": 0.462890625, + "learning_rate": 0.0013052356703039903, + "loss": 0.1465, + "step": 41544 + }, + { + "epoch": 0.36063055008203054, + "grad_norm": 0.228515625, + "learning_rate": 0.0013052066919207231, + "loss": 0.1406, + "step": 41545 + }, + { + "epoch": 0.3606392305622347, + "grad_norm": 0.09228515625, + "learning_rate": 0.0013051777133130445, + "loss": 0.1035, + "step": 41546 + }, + { + "epoch": 0.3606479110424389, + "grad_norm": 0.5859375, + "learning_rate": 0.0013051487344809868, + "loss": 0.1045, + "step": 41547 + }, + { + "epoch": 0.360656591522643, + "grad_norm": 0.11474609375, + "learning_rate": 0.0013051197554245816, + "loss": 0.0908, + "step": 41548 + }, + { + "epoch": 0.3606652720028472, + "grad_norm": 0.61328125, + "learning_rate": 0.0013050907761438606, + "loss": 0.0967, + "step": 41549 + }, + { + "epoch": 0.36067395248305134, + "grad_norm": 0.22265625, + "learning_rate": 0.0013050617966388553, + "loss": 0.104, + "step": 41550 + }, + { + "epoch": 0.36068263296325553, + "grad_norm": 0.1416015625, + "learning_rate": 0.0013050328169095975, + "loss": 0.0845, + "step": 41551 + }, + { + "epoch": 0.36069131344345967, + "grad_norm": 0.2294921875, + "learning_rate": 0.001305003836956119, + "loss": 0.0703, + "step": 41552 + }, + { + "epoch": 0.36069999392366386, + "grad_norm": 0.1259765625, + "learning_rate": 0.0013049748567784513, + "loss": 0.0762, + "step": 41553 + }, + { + "epoch": 0.360708674403868, + "grad_norm": 0.08740234375, + "learning_rate": 0.0013049458763766263, + "loss": 0.0674, + "step": 41554 + }, + { + "epoch": 0.3607173548840722, + "grad_norm": 0.70703125, + "learning_rate": 0.0013049168957506752, + "loss": 0.1221, + "step": 41555 + }, + { + "epoch": 0.36072603536427633, + "grad_norm": 0.59765625, + "learning_rate": 0.0013048879149006303, + "loss": 0.0967, + "step": 41556 + }, + { + "epoch": 0.3607347158444805, + "grad_norm": 0.326171875, + "learning_rate": 0.0013048589338265232, + "loss": 0.1104, + "step": 41557 + }, + { + "epoch": 0.36074339632468466, + "grad_norm": 0.2265625, + "learning_rate": 0.0013048299525283851, + "loss": 0.1084, + "step": 41558 + }, + { + "epoch": 0.36075207680488885, + "grad_norm": 0.1328125, + "learning_rate": 0.0013048009710062483, + "loss": 0.0913, + "step": 41559 + }, + { + "epoch": 0.360760757285093, + "grad_norm": 0.21484375, + "learning_rate": 0.0013047719892601442, + "loss": 0.1143, + "step": 41560 + }, + { + "epoch": 0.3607694377652972, + "grad_norm": 0.29296875, + "learning_rate": 0.0013047430072901046, + "loss": 0.1211, + "step": 41561 + }, + { + "epoch": 0.3607781182455013, + "grad_norm": 0.4765625, + "learning_rate": 0.0013047140250961608, + "loss": 0.0996, + "step": 41562 + }, + { + "epoch": 0.3607867987257055, + "grad_norm": 0.23046875, + "learning_rate": 0.001304685042678345, + "loss": 0.0742, + "step": 41563 + }, + { + "epoch": 0.36079547920590965, + "grad_norm": 0.3984375, + "learning_rate": 0.0013046560600366891, + "loss": 0.1069, + "step": 41564 + }, + { + "epoch": 0.36080415968611385, + "grad_norm": 0.0947265625, + "learning_rate": 0.001304627077171224, + "loss": 0.1016, + "step": 41565 + }, + { + "epoch": 0.360812840166318, + "grad_norm": 0.08837890625, + "learning_rate": 0.001304598094081982, + "loss": 0.0869, + "step": 41566 + }, + { + "epoch": 0.3608215206465222, + "grad_norm": 0.455078125, + "learning_rate": 0.0013045691107689945, + "loss": 0.0923, + "step": 41567 + }, + { + "epoch": 0.3608302011267263, + "grad_norm": 0.44921875, + "learning_rate": 0.001304540127232293, + "loss": 0.0884, + "step": 41568 + }, + { + "epoch": 0.3608388816069305, + "grad_norm": 0.474609375, + "learning_rate": 0.0013045111434719099, + "loss": 0.1084, + "step": 41569 + }, + { + "epoch": 0.36084756208713464, + "grad_norm": 0.306640625, + "learning_rate": 0.0013044821594878765, + "loss": 0.0967, + "step": 41570 + }, + { + "epoch": 0.36085624256733884, + "grad_norm": 0.224609375, + "learning_rate": 0.0013044531752802243, + "loss": 0.085, + "step": 41571 + }, + { + "epoch": 0.360864923047543, + "grad_norm": 0.380859375, + "learning_rate": 0.0013044241908489848, + "loss": 0.1162, + "step": 41572 + }, + { + "epoch": 0.36087360352774717, + "grad_norm": 0.0927734375, + "learning_rate": 0.0013043952061941907, + "loss": 0.1084, + "step": 41573 + }, + { + "epoch": 0.3608822840079513, + "grad_norm": 0.48046875, + "learning_rate": 0.001304366221315873, + "loss": 0.1064, + "step": 41574 + }, + { + "epoch": 0.3608909644881555, + "grad_norm": 0.56640625, + "learning_rate": 0.0013043372362140633, + "loss": 0.0962, + "step": 41575 + }, + { + "epoch": 0.36089964496835963, + "grad_norm": 0.16015625, + "learning_rate": 0.0013043082508887936, + "loss": 0.0942, + "step": 41576 + }, + { + "epoch": 0.3609083254485638, + "grad_norm": 0.2353515625, + "learning_rate": 0.001304279265340095, + "loss": 0.0708, + "step": 41577 + }, + { + "epoch": 0.36091700592876796, + "grad_norm": 0.318359375, + "learning_rate": 0.0013042502795680001, + "loss": 0.2168, + "step": 41578 + }, + { + "epoch": 0.36092568640897216, + "grad_norm": 0.275390625, + "learning_rate": 0.0013042212935725401, + "loss": 0.1162, + "step": 41579 + }, + { + "epoch": 0.3609343668891763, + "grad_norm": 0.28125, + "learning_rate": 0.0013041923073537469, + "loss": 0.0986, + "step": 41580 + }, + { + "epoch": 0.3609430473693805, + "grad_norm": 0.353515625, + "learning_rate": 0.001304163320911652, + "loss": 0.1416, + "step": 41581 + }, + { + "epoch": 0.3609517278495846, + "grad_norm": 0.142578125, + "learning_rate": 0.001304134334246287, + "loss": 0.0967, + "step": 41582 + }, + { + "epoch": 0.3609604083297888, + "grad_norm": 0.1533203125, + "learning_rate": 0.001304105347357684, + "loss": 0.0835, + "step": 41583 + }, + { + "epoch": 0.36096908880999296, + "grad_norm": 0.1083984375, + "learning_rate": 0.001304076360245874, + "loss": 0.1074, + "step": 41584 + }, + { + "epoch": 0.36097776929019715, + "grad_norm": 0.2109375, + "learning_rate": 0.0013040473729108897, + "loss": 0.1016, + "step": 41585 + }, + { + "epoch": 0.3609864497704013, + "grad_norm": 0.44921875, + "learning_rate": 0.001304018385352762, + "loss": 0.0913, + "step": 41586 + }, + { + "epoch": 0.3609951302506055, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013039893975715226, + "loss": 0.1123, + "step": 41587 + }, + { + "epoch": 0.3610038107308096, + "grad_norm": 0.1865234375, + "learning_rate": 0.0013039604095672038, + "loss": 0.0928, + "step": 41588 + }, + { + "epoch": 0.3610124912110138, + "grad_norm": 0.1015625, + "learning_rate": 0.001303931421339837, + "loss": 0.0996, + "step": 41589 + }, + { + "epoch": 0.36102117169121795, + "grad_norm": 0.283203125, + "learning_rate": 0.0013039024328894535, + "loss": 0.1133, + "step": 41590 + }, + { + "epoch": 0.36102985217142214, + "grad_norm": 0.140625, + "learning_rate": 0.0013038734442160859, + "loss": 0.083, + "step": 41591 + }, + { + "epoch": 0.3610385326516263, + "grad_norm": 0.29296875, + "learning_rate": 0.001303844455319765, + "loss": 0.0938, + "step": 41592 + }, + { + "epoch": 0.36104721313183047, + "grad_norm": 0.400390625, + "learning_rate": 0.0013038154662005231, + "loss": 0.1514, + "step": 41593 + }, + { + "epoch": 0.3610558936120346, + "grad_norm": 0.1455078125, + "learning_rate": 0.0013037864768583915, + "loss": 0.1123, + "step": 41594 + }, + { + "epoch": 0.3610645740922388, + "grad_norm": 0.2314453125, + "learning_rate": 0.0013037574872934018, + "loss": 0.0859, + "step": 41595 + }, + { + "epoch": 0.36107325457244294, + "grad_norm": 0.2158203125, + "learning_rate": 0.0013037284975055863, + "loss": 0.1309, + "step": 41596 + }, + { + "epoch": 0.36108193505264713, + "grad_norm": 0.25, + "learning_rate": 0.0013036995074949766, + "loss": 0.0737, + "step": 41597 + }, + { + "epoch": 0.36109061553285127, + "grad_norm": 0.08251953125, + "learning_rate": 0.0013036705172616034, + "loss": 0.0894, + "step": 41598 + }, + { + "epoch": 0.36109929601305546, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013036415268055, + "loss": 0.0986, + "step": 41599 + }, + { + "epoch": 0.3611079764932596, + "grad_norm": 0.11279296875, + "learning_rate": 0.001303612536126697, + "loss": 0.0801, + "step": 41600 + }, + { + "epoch": 0.3611166569734638, + "grad_norm": 0.45703125, + "learning_rate": 0.0013035835452252262, + "loss": 0.1211, + "step": 41601 + }, + { + "epoch": 0.3611253374536679, + "grad_norm": 0.09619140625, + "learning_rate": 0.00130355455410112, + "loss": 0.0889, + "step": 41602 + }, + { + "epoch": 0.3611340179338721, + "grad_norm": 0.36328125, + "learning_rate": 0.001303525562754409, + "loss": 0.1191, + "step": 41603 + }, + { + "epoch": 0.36114269841407626, + "grad_norm": 0.26171875, + "learning_rate": 0.0013034965711851256, + "loss": 0.0977, + "step": 41604 + }, + { + "epoch": 0.36115137889428045, + "grad_norm": 0.578125, + "learning_rate": 0.0013034675793933018, + "loss": 0.1147, + "step": 41605 + }, + { + "epoch": 0.3611600593744846, + "grad_norm": 0.08984375, + "learning_rate": 0.0013034385873789687, + "loss": 0.0771, + "step": 41606 + }, + { + "epoch": 0.3611687398546888, + "grad_norm": 0.21484375, + "learning_rate": 0.0013034095951421582, + "loss": 0.0806, + "step": 41607 + }, + { + "epoch": 0.3611774203348929, + "grad_norm": 0.59765625, + "learning_rate": 0.0013033806026829019, + "loss": 0.1367, + "step": 41608 + }, + { + "epoch": 0.3611861008150971, + "grad_norm": 1.65625, + "learning_rate": 0.0013033516100012317, + "loss": 0.1719, + "step": 41609 + }, + { + "epoch": 0.36119478129530125, + "grad_norm": 0.09375, + "learning_rate": 0.0013033226170971796, + "loss": 0.1245, + "step": 41610 + }, + { + "epoch": 0.36120346177550544, + "grad_norm": 0.1689453125, + "learning_rate": 0.0013032936239707764, + "loss": 0.1016, + "step": 41611 + }, + { + "epoch": 0.3612121422557096, + "grad_norm": 0.13671875, + "learning_rate": 0.0013032646306220545, + "loss": 0.0947, + "step": 41612 + }, + { + "epoch": 0.36122082273591377, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013032356370510458, + "loss": 0.1152, + "step": 41613 + }, + { + "epoch": 0.3612295032161179, + "grad_norm": 0.49609375, + "learning_rate": 0.0013032066432577811, + "loss": 0.1001, + "step": 41614 + }, + { + "epoch": 0.3612381836963221, + "grad_norm": 0.236328125, + "learning_rate": 0.001303177649242293, + "loss": 0.1133, + "step": 41615 + }, + { + "epoch": 0.36124686417652624, + "grad_norm": 0.546875, + "learning_rate": 0.001303148655004613, + "loss": 0.0815, + "step": 41616 + }, + { + "epoch": 0.36125554465673043, + "grad_norm": 0.337890625, + "learning_rate": 0.0013031196605447725, + "loss": 0.1367, + "step": 41617 + }, + { + "epoch": 0.36126422513693457, + "grad_norm": 0.08154296875, + "learning_rate": 0.0013030906658628035, + "loss": 0.125, + "step": 41618 + }, + { + "epoch": 0.36127290561713876, + "grad_norm": 0.310546875, + "learning_rate": 0.0013030616709587375, + "loss": 0.1338, + "step": 41619 + }, + { + "epoch": 0.3612815860973429, + "grad_norm": 0.1669921875, + "learning_rate": 0.001303032675832606, + "loss": 0.0903, + "step": 41620 + }, + { + "epoch": 0.3612902665775471, + "grad_norm": 0.19140625, + "learning_rate": 0.0013030036804844416, + "loss": 0.1104, + "step": 41621 + }, + { + "epoch": 0.36129894705775123, + "grad_norm": 0.10009765625, + "learning_rate": 0.0013029746849142752, + "loss": 0.0996, + "step": 41622 + }, + { + "epoch": 0.3613076275379554, + "grad_norm": 0.177734375, + "learning_rate": 0.0013029456891221385, + "loss": 0.0913, + "step": 41623 + }, + { + "epoch": 0.36131630801815956, + "grad_norm": 0.177734375, + "learning_rate": 0.0013029166931080637, + "loss": 0.0986, + "step": 41624 + }, + { + "epoch": 0.36132498849836375, + "grad_norm": 0.26171875, + "learning_rate": 0.001302887696872082, + "loss": 0.0791, + "step": 41625 + }, + { + "epoch": 0.3613336689785679, + "grad_norm": 1.328125, + "learning_rate": 0.0013028587004142256, + "loss": 0.3125, + "step": 41626 + }, + { + "epoch": 0.3613423494587721, + "grad_norm": 0.0927734375, + "learning_rate": 0.001302829703734526, + "loss": 0.1172, + "step": 41627 + }, + { + "epoch": 0.3613510299389762, + "grad_norm": 0.083984375, + "learning_rate": 0.0013028007068330149, + "loss": 0.1279, + "step": 41628 + }, + { + "epoch": 0.3613597104191804, + "grad_norm": 0.1982421875, + "learning_rate": 0.0013027717097097238, + "loss": 0.0845, + "step": 41629 + }, + { + "epoch": 0.36136839089938455, + "grad_norm": 0.1806640625, + "learning_rate": 0.0013027427123646846, + "loss": 0.1025, + "step": 41630 + }, + { + "epoch": 0.36137707137958874, + "grad_norm": 0.306640625, + "learning_rate": 0.0013027137147979292, + "loss": 0.0967, + "step": 41631 + }, + { + "epoch": 0.3613857518597929, + "grad_norm": 0.5, + "learning_rate": 0.001302684717009489, + "loss": 0.126, + "step": 41632 + }, + { + "epoch": 0.3613944323399971, + "grad_norm": 0.19921875, + "learning_rate": 0.001302655718999396, + "loss": 0.0596, + "step": 41633 + }, + { + "epoch": 0.3614031128202012, + "grad_norm": 0.076171875, + "learning_rate": 0.0013026267207676815, + "loss": 0.0576, + "step": 41634 + }, + { + "epoch": 0.3614117933004054, + "grad_norm": 0.1875, + "learning_rate": 0.001302597722314378, + "loss": 0.1318, + "step": 41635 + }, + { + "epoch": 0.36142047378060954, + "grad_norm": 0.376953125, + "learning_rate": 0.001302568723639516, + "loss": 0.106, + "step": 41636 + }, + { + "epoch": 0.36142915426081373, + "grad_norm": 0.15625, + "learning_rate": 0.001302539724743128, + "loss": 0.084, + "step": 41637 + }, + { + "epoch": 0.36143783474101787, + "grad_norm": 0.294921875, + "learning_rate": 0.0013025107256252459, + "loss": 0.0728, + "step": 41638 + }, + { + "epoch": 0.36144651522122206, + "grad_norm": 0.29296875, + "learning_rate": 0.001302481726285901, + "loss": 0.105, + "step": 41639 + }, + { + "epoch": 0.3614551957014262, + "grad_norm": 0.54296875, + "learning_rate": 0.0013024527267251251, + "loss": 0.085, + "step": 41640 + }, + { + "epoch": 0.3614638761816304, + "grad_norm": 0.3984375, + "learning_rate": 0.0013024237269429499, + "loss": 0.1465, + "step": 41641 + }, + { + "epoch": 0.36147255666183453, + "grad_norm": 0.3125, + "learning_rate": 0.0013023947269394072, + "loss": 0.0918, + "step": 41642 + }, + { + "epoch": 0.3614812371420387, + "grad_norm": 0.1484375, + "learning_rate": 0.0013023657267145286, + "loss": 0.1074, + "step": 41643 + }, + { + "epoch": 0.36148991762224286, + "grad_norm": 0.318359375, + "learning_rate": 0.001302336726268346, + "loss": 0.1084, + "step": 41644 + }, + { + "epoch": 0.361498598102447, + "grad_norm": 0.1533203125, + "learning_rate": 0.0013023077256008912, + "loss": 0.0728, + "step": 41645 + }, + { + "epoch": 0.3615072785826512, + "grad_norm": 0.140625, + "learning_rate": 0.0013022787247121952, + "loss": 0.1064, + "step": 41646 + }, + { + "epoch": 0.36151595906285533, + "grad_norm": 0.380859375, + "learning_rate": 0.0013022497236022906, + "loss": 0.1001, + "step": 41647 + }, + { + "epoch": 0.3615246395430595, + "grad_norm": 0.0830078125, + "learning_rate": 0.0013022207222712088, + "loss": 0.0928, + "step": 41648 + }, + { + "epoch": 0.36153332002326366, + "grad_norm": 0.279296875, + "learning_rate": 0.001302191720718981, + "loss": 0.0884, + "step": 41649 + }, + { + "epoch": 0.36154200050346785, + "grad_norm": 0.443359375, + "learning_rate": 0.0013021627189456397, + "loss": 0.1338, + "step": 41650 + }, + { + "epoch": 0.361550680983672, + "grad_norm": 0.6484375, + "learning_rate": 0.001302133716951216, + "loss": 0.1104, + "step": 41651 + }, + { + "epoch": 0.3615593614638762, + "grad_norm": 0.25390625, + "learning_rate": 0.0013021047147357423, + "loss": 0.0913, + "step": 41652 + }, + { + "epoch": 0.3615680419440803, + "grad_norm": 0.2109375, + "learning_rate": 0.00130207571229925, + "loss": 0.1157, + "step": 41653 + }, + { + "epoch": 0.3615767224242845, + "grad_norm": 0.27734375, + "learning_rate": 0.0013020467096417703, + "loss": 0.083, + "step": 41654 + }, + { + "epoch": 0.36158540290448865, + "grad_norm": 0.30078125, + "learning_rate": 0.0013020177067633357, + "loss": 0.1113, + "step": 41655 + }, + { + "epoch": 0.36159408338469284, + "grad_norm": 0.275390625, + "learning_rate": 0.0013019887036639776, + "loss": 0.1133, + "step": 41656 + }, + { + "epoch": 0.361602763864897, + "grad_norm": 0.2236328125, + "learning_rate": 0.0013019597003437273, + "loss": 0.1016, + "step": 41657 + }, + { + "epoch": 0.3616114443451012, + "grad_norm": 0.2451171875, + "learning_rate": 0.0013019306968026173, + "loss": 0.0703, + "step": 41658 + }, + { + "epoch": 0.3616201248253053, + "grad_norm": 0.07421875, + "learning_rate": 0.0013019016930406785, + "loss": 0.104, + "step": 41659 + }, + { + "epoch": 0.3616288053055095, + "grad_norm": 0.1865234375, + "learning_rate": 0.0013018726890579432, + "loss": 0.0918, + "step": 41660 + }, + { + "epoch": 0.36163748578571364, + "grad_norm": 0.27734375, + "learning_rate": 0.0013018436848544431, + "loss": 0.0986, + "step": 41661 + }, + { + "epoch": 0.36164616626591783, + "grad_norm": 0.07470703125, + "learning_rate": 0.0013018146804302099, + "loss": 0.1035, + "step": 41662 + }, + { + "epoch": 0.36165484674612197, + "grad_norm": 0.65234375, + "learning_rate": 0.001301785675785275, + "loss": 0.125, + "step": 41663 + }, + { + "epoch": 0.36166352722632616, + "grad_norm": 0.36328125, + "learning_rate": 0.0013017566709196701, + "loss": 0.0977, + "step": 41664 + }, + { + "epoch": 0.3616722077065303, + "grad_norm": 0.107421875, + "learning_rate": 0.0013017276658334274, + "loss": 0.106, + "step": 41665 + }, + { + "epoch": 0.3616808881867345, + "grad_norm": 0.10302734375, + "learning_rate": 0.0013016986605265783, + "loss": 0.1064, + "step": 41666 + }, + { + "epoch": 0.36168956866693863, + "grad_norm": 0.15234375, + "learning_rate": 0.0013016696549991546, + "loss": 0.0835, + "step": 41667 + }, + { + "epoch": 0.3616982491471428, + "grad_norm": 0.1328125, + "learning_rate": 0.0013016406492511881, + "loss": 0.1025, + "step": 41668 + }, + { + "epoch": 0.36170692962734696, + "grad_norm": 0.4140625, + "learning_rate": 0.0013016116432827103, + "loss": 0.1006, + "step": 41669 + }, + { + "epoch": 0.36171561010755116, + "grad_norm": 0.294921875, + "learning_rate": 0.001301582637093753, + "loss": 0.0713, + "step": 41670 + }, + { + "epoch": 0.3617242905877553, + "grad_norm": 0.4453125, + "learning_rate": 0.001301553630684348, + "loss": 0.1006, + "step": 41671 + }, + { + "epoch": 0.3617329710679595, + "grad_norm": 0.349609375, + "learning_rate": 0.0013015246240545269, + "loss": 0.1221, + "step": 41672 + }, + { + "epoch": 0.3617416515481636, + "grad_norm": 0.111328125, + "learning_rate": 0.0013014956172043217, + "loss": 0.1001, + "step": 41673 + }, + { + "epoch": 0.3617503320283678, + "grad_norm": 0.349609375, + "learning_rate": 0.0013014666101337639, + "loss": 0.1201, + "step": 41674 + }, + { + "epoch": 0.36175901250857195, + "grad_norm": 0.259765625, + "learning_rate": 0.0013014376028428851, + "loss": 0.0903, + "step": 41675 + }, + { + "epoch": 0.36176769298877615, + "grad_norm": 0.1767578125, + "learning_rate": 0.001301408595331717, + "loss": 0.0698, + "step": 41676 + }, + { + "epoch": 0.3617763734689803, + "grad_norm": 0.119140625, + "learning_rate": 0.001301379587600292, + "loss": 0.1504, + "step": 41677 + }, + { + "epoch": 0.3617850539491845, + "grad_norm": 0.482421875, + "learning_rate": 0.001301350579648641, + "loss": 0.125, + "step": 41678 + }, + { + "epoch": 0.3617937344293886, + "grad_norm": 0.18359375, + "learning_rate": 0.001301321571476796, + "loss": 0.1738, + "step": 41679 + }, + { + "epoch": 0.3618024149095928, + "grad_norm": 0.1376953125, + "learning_rate": 0.0013012925630847888, + "loss": 0.1318, + "step": 41680 + }, + { + "epoch": 0.36181109538979694, + "grad_norm": 0.6171875, + "learning_rate": 0.001301263554472651, + "loss": 0.0806, + "step": 41681 + }, + { + "epoch": 0.36181977587000114, + "grad_norm": 0.10498046875, + "learning_rate": 0.0013012345456404144, + "loss": 0.104, + "step": 41682 + }, + { + "epoch": 0.3618284563502053, + "grad_norm": 0.1025390625, + "learning_rate": 0.0013012055365881111, + "loss": 0.1064, + "step": 41683 + }, + { + "epoch": 0.36183713683040947, + "grad_norm": 0.28515625, + "learning_rate": 0.001301176527315772, + "loss": 0.1885, + "step": 41684 + }, + { + "epoch": 0.3618458173106136, + "grad_norm": 0.1845703125, + "learning_rate": 0.0013011475178234294, + "loss": 0.0957, + "step": 41685 + }, + { + "epoch": 0.3618544977908178, + "grad_norm": 0.07373046875, + "learning_rate": 0.0013011185081111148, + "loss": 0.0786, + "step": 41686 + }, + { + "epoch": 0.36186317827102193, + "grad_norm": 0.27734375, + "learning_rate": 0.0013010894981788602, + "loss": 0.0874, + "step": 41687 + }, + { + "epoch": 0.3618718587512261, + "grad_norm": 0.0869140625, + "learning_rate": 0.0013010604880266973, + "loss": 0.0913, + "step": 41688 + }, + { + "epoch": 0.36188053923143026, + "grad_norm": 0.51171875, + "learning_rate": 0.0013010314776546572, + "loss": 0.1377, + "step": 41689 + }, + { + "epoch": 0.36188921971163446, + "grad_norm": 0.173828125, + "learning_rate": 0.0013010024670627725, + "loss": 0.0913, + "step": 41690 + }, + { + "epoch": 0.3618979001918386, + "grad_norm": 0.1962890625, + "learning_rate": 0.0013009734562510742, + "loss": 0.1079, + "step": 41691 + }, + { + "epoch": 0.3619065806720428, + "grad_norm": 0.09423828125, + "learning_rate": 0.0013009444452195948, + "loss": 0.1035, + "step": 41692 + }, + { + "epoch": 0.3619152611522469, + "grad_norm": 0.09619140625, + "learning_rate": 0.0013009154339683649, + "loss": 0.105, + "step": 41693 + }, + { + "epoch": 0.3619239416324511, + "grad_norm": 0.1708984375, + "learning_rate": 0.0013008864224974173, + "loss": 0.0874, + "step": 41694 + }, + { + "epoch": 0.36193262211265526, + "grad_norm": 0.09375, + "learning_rate": 0.0013008574108067834, + "loss": 0.1064, + "step": 41695 + }, + { + "epoch": 0.36194130259285945, + "grad_norm": 0.255859375, + "learning_rate": 0.001300828398896495, + "loss": 0.1426, + "step": 41696 + }, + { + "epoch": 0.3619499830730636, + "grad_norm": 0.1552734375, + "learning_rate": 0.0013007993867665834, + "loss": 0.0742, + "step": 41697 + }, + { + "epoch": 0.3619586635532678, + "grad_norm": 0.1513671875, + "learning_rate": 0.0013007703744170806, + "loss": 0.0947, + "step": 41698 + }, + { + "epoch": 0.3619673440334719, + "grad_norm": 1.203125, + "learning_rate": 0.0013007413618480186, + "loss": 0.1367, + "step": 41699 + }, + { + "epoch": 0.3619760245136761, + "grad_norm": 0.6640625, + "learning_rate": 0.0013007123490594284, + "loss": 0.0811, + "step": 41700 + }, + { + "epoch": 0.36198470499388025, + "grad_norm": 0.2734375, + "learning_rate": 0.0013006833360513425, + "loss": 0.1001, + "step": 41701 + }, + { + "epoch": 0.36199338547408444, + "grad_norm": 0.185546875, + "learning_rate": 0.001300654322823792, + "loss": 0.1016, + "step": 41702 + }, + { + "epoch": 0.3620020659542886, + "grad_norm": 0.1953125, + "learning_rate": 0.0013006253093768095, + "loss": 0.1484, + "step": 41703 + }, + { + "epoch": 0.36201074643449277, + "grad_norm": 0.203125, + "learning_rate": 0.001300596295710426, + "loss": 0.1299, + "step": 41704 + }, + { + "epoch": 0.3620194269146969, + "grad_norm": 0.2041015625, + "learning_rate": 0.0013005672818246733, + "loss": 0.0942, + "step": 41705 + }, + { + "epoch": 0.3620281073949011, + "grad_norm": 0.330078125, + "learning_rate": 0.0013005382677195832, + "loss": 0.1367, + "step": 41706 + }, + { + "epoch": 0.36203678787510524, + "grad_norm": 0.1884765625, + "learning_rate": 0.0013005092533951875, + "loss": 0.0703, + "step": 41707 + }, + { + "epoch": 0.36204546835530943, + "grad_norm": 0.10107421875, + "learning_rate": 0.0013004802388515177, + "loss": 0.1133, + "step": 41708 + }, + { + "epoch": 0.36205414883551357, + "grad_norm": 0.291015625, + "learning_rate": 0.001300451224088606, + "loss": 0.168, + "step": 41709 + }, + { + "epoch": 0.36206282931571776, + "grad_norm": 0.224609375, + "learning_rate": 0.001300422209106484, + "loss": 0.0859, + "step": 41710 + }, + { + "epoch": 0.3620715097959219, + "grad_norm": 0.31640625, + "learning_rate": 0.0013003931939051824, + "loss": 0.0962, + "step": 41711 + }, + { + "epoch": 0.3620801902761261, + "grad_norm": 0.34765625, + "learning_rate": 0.0013003641784847348, + "loss": 0.0874, + "step": 41712 + }, + { + "epoch": 0.36208887075633023, + "grad_norm": 0.60546875, + "learning_rate": 0.0013003351628451717, + "loss": 0.0928, + "step": 41713 + }, + { + "epoch": 0.3620975512365344, + "grad_norm": 0.154296875, + "learning_rate": 0.0013003061469865252, + "loss": 0.0903, + "step": 41714 + }, + { + "epoch": 0.36210623171673856, + "grad_norm": 0.22265625, + "learning_rate": 0.0013002771309088266, + "loss": 0.1211, + "step": 41715 + }, + { + "epoch": 0.36211491219694275, + "grad_norm": 0.10302734375, + "learning_rate": 0.001300248114612108, + "loss": 0.0791, + "step": 41716 + }, + { + "epoch": 0.3621235926771469, + "grad_norm": 0.166015625, + "learning_rate": 0.001300219098096401, + "loss": 0.1182, + "step": 41717 + }, + { + "epoch": 0.3621322731573511, + "grad_norm": 0.451171875, + "learning_rate": 0.0013001900813617378, + "loss": 0.0962, + "step": 41718 + }, + { + "epoch": 0.3621409536375552, + "grad_norm": 0.408203125, + "learning_rate": 0.0013001610644081492, + "loss": 0.0981, + "step": 41719 + }, + { + "epoch": 0.3621496341177594, + "grad_norm": 0.90234375, + "learning_rate": 0.0013001320472356675, + "loss": 0.1367, + "step": 41720 + }, + { + "epoch": 0.36215831459796355, + "grad_norm": 0.083984375, + "learning_rate": 0.0013001030298443246, + "loss": 0.0664, + "step": 41721 + }, + { + "epoch": 0.36216699507816774, + "grad_norm": 0.267578125, + "learning_rate": 0.0013000740122341523, + "loss": 0.0869, + "step": 41722 + }, + { + "epoch": 0.3621756755583719, + "grad_norm": 0.330078125, + "learning_rate": 0.0013000449944051817, + "loss": 0.0991, + "step": 41723 + }, + { + "epoch": 0.36218435603857607, + "grad_norm": 0.0732421875, + "learning_rate": 0.0013000159763574448, + "loss": 0.0674, + "step": 41724 + }, + { + "epoch": 0.3621930365187802, + "grad_norm": 0.294921875, + "learning_rate": 0.0012999869580909737, + "loss": 0.0811, + "step": 41725 + }, + { + "epoch": 0.3622017169989844, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012999579396057999, + "loss": 0.1045, + "step": 41726 + }, + { + "epoch": 0.36221039747918854, + "grad_norm": 0.2421875, + "learning_rate": 0.001299928920901955, + "loss": 0.1035, + "step": 41727 + }, + { + "epoch": 0.36221907795939273, + "grad_norm": 0.298828125, + "learning_rate": 0.0012998999019794711, + "loss": 0.1216, + "step": 41728 + }, + { + "epoch": 0.36222775843959687, + "grad_norm": 0.294921875, + "learning_rate": 0.0012998708828383789, + "loss": 0.0674, + "step": 41729 + }, + { + "epoch": 0.36223643891980106, + "grad_norm": 0.60546875, + "learning_rate": 0.0012998418634787116, + "loss": 0.1045, + "step": 41730 + }, + { + "epoch": 0.3622451194000052, + "grad_norm": 0.53125, + "learning_rate": 0.0012998128439005, + "loss": 0.0981, + "step": 41731 + }, + { + "epoch": 0.3622537998802094, + "grad_norm": 0.5078125, + "learning_rate": 0.001299783824103776, + "loss": 0.1084, + "step": 41732 + }, + { + "epoch": 0.36226248036041353, + "grad_norm": 0.17578125, + "learning_rate": 0.0012997548040885717, + "loss": 0.0781, + "step": 41733 + }, + { + "epoch": 0.3622711608406177, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012997257838549185, + "loss": 0.1104, + "step": 41734 + }, + { + "epoch": 0.36227984132082186, + "grad_norm": 0.298828125, + "learning_rate": 0.0012996967634028482, + "loss": 0.1182, + "step": 41735 + }, + { + "epoch": 0.36228852180102605, + "grad_norm": 0.94140625, + "learning_rate": 0.0012996677427323923, + "loss": 0.0586, + "step": 41736 + }, + { + "epoch": 0.3622972022812302, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012996387218435828, + "loss": 0.0889, + "step": 41737 + }, + { + "epoch": 0.3623058827614344, + "grad_norm": 0.16796875, + "learning_rate": 0.0012996097007364513, + "loss": 0.1406, + "step": 41738 + }, + { + "epoch": 0.3623145632416385, + "grad_norm": 0.6796875, + "learning_rate": 0.00129958067941103, + "loss": 0.1221, + "step": 41739 + }, + { + "epoch": 0.3623232437218427, + "grad_norm": 0.51953125, + "learning_rate": 0.00129955165786735, + "loss": 0.1045, + "step": 41740 + }, + { + "epoch": 0.36233192420204685, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012995226361054433, + "loss": 0.103, + "step": 41741 + }, + { + "epoch": 0.36234060468225104, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012994936141253418, + "loss": 0.0752, + "step": 41742 + }, + { + "epoch": 0.3623492851624552, + "grad_norm": 0.1630859375, + "learning_rate": 0.001299464591927077, + "loss": 0.0728, + "step": 41743 + }, + { + "epoch": 0.3623579656426594, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012994355695106806, + "loss": 0.1084, + "step": 41744 + }, + { + "epoch": 0.3623666461228635, + "grad_norm": 0.33984375, + "learning_rate": 0.0012994065468761847, + "loss": 0.085, + "step": 41745 + }, + { + "epoch": 0.3623753266030677, + "grad_norm": 0.271484375, + "learning_rate": 0.0012993775240236205, + "loss": 0.0991, + "step": 41746 + }, + { + "epoch": 0.36238400708327184, + "grad_norm": 0.271484375, + "learning_rate": 0.0012993485009530205, + "loss": 0.0762, + "step": 41747 + }, + { + "epoch": 0.36239268756347603, + "grad_norm": 0.09521484375, + "learning_rate": 0.0012993194776644154, + "loss": 0.0801, + "step": 41748 + }, + { + "epoch": 0.36240136804368017, + "grad_norm": 0.27734375, + "learning_rate": 0.0012992904541578377, + "loss": 0.0933, + "step": 41749 + }, + { + "epoch": 0.36241004852388436, + "grad_norm": 0.23046875, + "learning_rate": 0.001299261430433319, + "loss": 0.0933, + "step": 41750 + }, + { + "epoch": 0.3624187290040885, + "grad_norm": 0.291015625, + "learning_rate": 0.0012992324064908909, + "loss": 0.0938, + "step": 41751 + }, + { + "epoch": 0.3624274094842927, + "grad_norm": 0.181640625, + "learning_rate": 0.0012992033823305855, + "loss": 0.1157, + "step": 41752 + }, + { + "epoch": 0.36243608996449683, + "grad_norm": 0.2236328125, + "learning_rate": 0.0012991743579524341, + "loss": 0.104, + "step": 41753 + }, + { + "epoch": 0.362444770444701, + "grad_norm": 0.76953125, + "learning_rate": 0.0012991453333564688, + "loss": 0.0898, + "step": 41754 + }, + { + "epoch": 0.36245345092490516, + "grad_norm": 0.353515625, + "learning_rate": 0.0012991163085427208, + "loss": 0.1494, + "step": 41755 + }, + { + "epoch": 0.36246213140510936, + "grad_norm": 0.10791015625, + "learning_rate": 0.0012990872835112227, + "loss": 0.1128, + "step": 41756 + }, + { + "epoch": 0.3624708118853135, + "grad_norm": 0.283203125, + "learning_rate": 0.0012990582582620052, + "loss": 0.1021, + "step": 41757 + }, + { + "epoch": 0.3624794923655177, + "grad_norm": 0.55078125, + "learning_rate": 0.0012990292327951008, + "loss": 0.0967, + "step": 41758 + }, + { + "epoch": 0.3624881728457218, + "grad_norm": 0.23046875, + "learning_rate": 0.0012990002071105412, + "loss": 0.0693, + "step": 41759 + }, + { + "epoch": 0.362496853325926, + "grad_norm": 0.171875, + "learning_rate": 0.0012989711812083577, + "loss": 0.0869, + "step": 41760 + }, + { + "epoch": 0.36250553380613015, + "grad_norm": 0.3046875, + "learning_rate": 0.0012989421550885824, + "loss": 0.0898, + "step": 41761 + }, + { + "epoch": 0.36251421428633435, + "grad_norm": 0.212890625, + "learning_rate": 0.0012989131287512467, + "loss": 0.0845, + "step": 41762 + }, + { + "epoch": 0.3625228947665385, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012988841021963828, + "loss": 0.1157, + "step": 41763 + }, + { + "epoch": 0.3625315752467427, + "grad_norm": 0.12060546875, + "learning_rate": 0.0012988550754240227, + "loss": 0.1177, + "step": 41764 + }, + { + "epoch": 0.3625402557269468, + "grad_norm": 0.51953125, + "learning_rate": 0.001298826048434197, + "loss": 0.0801, + "step": 41765 + }, + { + "epoch": 0.362548936207151, + "grad_norm": 0.48828125, + "learning_rate": 0.0012987970212269384, + "loss": 0.1035, + "step": 41766 + }, + { + "epoch": 0.36255761668735514, + "grad_norm": 0.2265625, + "learning_rate": 0.0012987679938022785, + "loss": 0.1157, + "step": 41767 + }, + { + "epoch": 0.36256629716755934, + "grad_norm": 0.337890625, + "learning_rate": 0.0012987389661602484, + "loss": 0.0791, + "step": 41768 + }, + { + "epoch": 0.3625749776477635, + "grad_norm": 0.5859375, + "learning_rate": 0.0012987099383008809, + "loss": 0.1113, + "step": 41769 + }, + { + "epoch": 0.3625836581279676, + "grad_norm": 0.1318359375, + "learning_rate": 0.001298680910224207, + "loss": 0.1055, + "step": 41770 + }, + { + "epoch": 0.3625923386081718, + "grad_norm": 0.095703125, + "learning_rate": 0.0012986518819302585, + "loss": 0.1191, + "step": 41771 + }, + { + "epoch": 0.36260101908837594, + "grad_norm": 0.5, + "learning_rate": 0.0012986228534190673, + "loss": 0.1143, + "step": 41772 + }, + { + "epoch": 0.36260969956858013, + "grad_norm": 0.22265625, + "learning_rate": 0.0012985938246906656, + "loss": 0.168, + "step": 41773 + }, + { + "epoch": 0.3626183800487843, + "grad_norm": 0.076171875, + "learning_rate": 0.0012985647957450843, + "loss": 0.0811, + "step": 41774 + }, + { + "epoch": 0.36262706052898847, + "grad_norm": 0.396484375, + "learning_rate": 0.0012985357665823553, + "loss": 0.127, + "step": 41775 + }, + { + "epoch": 0.3626357410091926, + "grad_norm": 0.08544921875, + "learning_rate": 0.0012985067372025113, + "loss": 0.0967, + "step": 41776 + }, + { + "epoch": 0.3626444214893968, + "grad_norm": 0.1650390625, + "learning_rate": 0.0012984777076055826, + "loss": 0.0845, + "step": 41777 + }, + { + "epoch": 0.36265310196960093, + "grad_norm": 0.1494140625, + "learning_rate": 0.001298448677791602, + "loss": 0.0913, + "step": 41778 + }, + { + "epoch": 0.3626617824498051, + "grad_norm": 0.474609375, + "learning_rate": 0.0012984196477606007, + "loss": 0.1123, + "step": 41779 + }, + { + "epoch": 0.36267046293000926, + "grad_norm": 0.486328125, + "learning_rate": 0.0012983906175126107, + "loss": 0.085, + "step": 41780 + }, + { + "epoch": 0.36267914341021346, + "grad_norm": 0.1787109375, + "learning_rate": 0.001298361587047664, + "loss": 0.127, + "step": 41781 + }, + { + "epoch": 0.3626878238904176, + "grad_norm": 0.6015625, + "learning_rate": 0.0012983325563657915, + "loss": 0.1445, + "step": 41782 + }, + { + "epoch": 0.3626965043706218, + "grad_norm": 0.240234375, + "learning_rate": 0.0012983035254670262, + "loss": 0.0903, + "step": 41783 + }, + { + "epoch": 0.3627051848508259, + "grad_norm": 0.11181640625, + "learning_rate": 0.0012982744943513984, + "loss": 0.0977, + "step": 41784 + }, + { + "epoch": 0.3627138653310301, + "grad_norm": 0.390625, + "learning_rate": 0.001298245463018941, + "loss": 0.0811, + "step": 41785 + }, + { + "epoch": 0.36272254581123425, + "grad_norm": 0.1591796875, + "learning_rate": 0.001298216431469685, + "loss": 0.1387, + "step": 41786 + }, + { + "epoch": 0.36273122629143845, + "grad_norm": 0.51953125, + "learning_rate": 0.0012981873997036633, + "loss": 0.1226, + "step": 41787 + }, + { + "epoch": 0.3627399067716426, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012981583677209058, + "loss": 0.1094, + "step": 41788 + }, + { + "epoch": 0.3627485872518468, + "grad_norm": 0.08056640625, + "learning_rate": 0.001298129335521446, + "loss": 0.1006, + "step": 41789 + }, + { + "epoch": 0.3627572677320509, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012981003031053149, + "loss": 0.0693, + "step": 41790 + }, + { + "epoch": 0.3627659482122551, + "grad_norm": 0.400390625, + "learning_rate": 0.0012980712704725441, + "loss": 0.207, + "step": 41791 + }, + { + "epoch": 0.36277462869245924, + "grad_norm": 0.1611328125, + "learning_rate": 0.0012980422376231656, + "loss": 0.085, + "step": 41792 + }, + { + "epoch": 0.36278330917266344, + "grad_norm": 0.46484375, + "learning_rate": 0.001298013204557211, + "loss": 0.0938, + "step": 41793 + }, + { + "epoch": 0.3627919896528676, + "grad_norm": 0.439453125, + "learning_rate": 0.001297984171274712, + "loss": 0.0962, + "step": 41794 + }, + { + "epoch": 0.36280067013307177, + "grad_norm": 0.408203125, + "learning_rate": 0.001297955137775701, + "loss": 0.125, + "step": 41795 + }, + { + "epoch": 0.3628093506132759, + "grad_norm": 0.337890625, + "learning_rate": 0.0012979261040602086, + "loss": 0.0869, + "step": 41796 + }, + { + "epoch": 0.3628180310934801, + "grad_norm": 0.099609375, + "learning_rate": 0.0012978970701282675, + "loss": 0.1201, + "step": 41797 + }, + { + "epoch": 0.36282671157368424, + "grad_norm": 0.244140625, + "learning_rate": 0.0012978680359799091, + "loss": 0.0947, + "step": 41798 + }, + { + "epoch": 0.36283539205388843, + "grad_norm": 0.57421875, + "learning_rate": 0.0012978390016151653, + "loss": 0.083, + "step": 41799 + }, + { + "epoch": 0.36284407253409257, + "grad_norm": 0.7734375, + "learning_rate": 0.0012978099670340677, + "loss": 0.0938, + "step": 41800 + }, + { + "epoch": 0.36285275301429676, + "grad_norm": 0.15625, + "learning_rate": 0.0012977809322366478, + "loss": 0.1504, + "step": 41801 + }, + { + "epoch": 0.3628614334945009, + "grad_norm": 0.1953125, + "learning_rate": 0.001297751897222938, + "loss": 0.0864, + "step": 41802 + }, + { + "epoch": 0.3628701139747051, + "grad_norm": 0.462890625, + "learning_rate": 0.0012977228619929697, + "loss": 0.1025, + "step": 41803 + }, + { + "epoch": 0.3628787944549092, + "grad_norm": 0.51171875, + "learning_rate": 0.0012976938265467745, + "loss": 0.0967, + "step": 41804 + }, + { + "epoch": 0.3628874749351134, + "grad_norm": 0.09375, + "learning_rate": 0.0012976647908843845, + "loss": 0.0864, + "step": 41805 + }, + { + "epoch": 0.36289615541531756, + "grad_norm": 0.25, + "learning_rate": 0.0012976357550058307, + "loss": 0.1348, + "step": 41806 + }, + { + "epoch": 0.36290483589552175, + "grad_norm": 0.447265625, + "learning_rate": 0.0012976067189111459, + "loss": 0.085, + "step": 41807 + }, + { + "epoch": 0.3629135163757259, + "grad_norm": 0.2021484375, + "learning_rate": 0.0012975776826003613, + "loss": 0.0771, + "step": 41808 + }, + { + "epoch": 0.3629221968559301, + "grad_norm": 0.12060546875, + "learning_rate": 0.0012975486460735086, + "loss": 0.0781, + "step": 41809 + }, + { + "epoch": 0.3629308773361342, + "grad_norm": 0.66796875, + "learning_rate": 0.0012975196093306198, + "loss": 0.1201, + "step": 41810 + }, + { + "epoch": 0.3629395578163384, + "grad_norm": 0.369140625, + "learning_rate": 0.0012974905723717265, + "loss": 0.1123, + "step": 41811 + }, + { + "epoch": 0.36294823829654255, + "grad_norm": 0.404296875, + "learning_rate": 0.0012974615351968603, + "loss": 0.1289, + "step": 41812 + }, + { + "epoch": 0.36295691877674674, + "grad_norm": 0.30859375, + "learning_rate": 0.0012974324978060534, + "loss": 0.1094, + "step": 41813 + }, + { + "epoch": 0.3629655992569509, + "grad_norm": 0.150390625, + "learning_rate": 0.0012974034601993371, + "loss": 0.1006, + "step": 41814 + }, + { + "epoch": 0.36297427973715507, + "grad_norm": 0.31640625, + "learning_rate": 0.0012973744223767434, + "loss": 0.1035, + "step": 41815 + }, + { + "epoch": 0.3629829602173592, + "grad_norm": 0.37890625, + "learning_rate": 0.001297345384338304, + "loss": 0.0684, + "step": 41816 + }, + { + "epoch": 0.3629916406975634, + "grad_norm": 0.21484375, + "learning_rate": 0.0012973163460840507, + "loss": 0.0928, + "step": 41817 + }, + { + "epoch": 0.36300032117776754, + "grad_norm": 0.138671875, + "learning_rate": 0.0012972873076140153, + "loss": 0.1035, + "step": 41818 + }, + { + "epoch": 0.36300900165797173, + "grad_norm": 0.1005859375, + "learning_rate": 0.001297258268928229, + "loss": 0.0938, + "step": 41819 + }, + { + "epoch": 0.36301768213817587, + "grad_norm": 0.546875, + "learning_rate": 0.0012972292300267242, + "loss": 0.0859, + "step": 41820 + }, + { + "epoch": 0.36302636261838006, + "grad_norm": 0.298828125, + "learning_rate": 0.001297200190909533, + "loss": 0.0933, + "step": 41821 + }, + { + "epoch": 0.3630350430985842, + "grad_norm": 0.11669921875, + "learning_rate": 0.0012971711515766865, + "loss": 0.1367, + "step": 41822 + }, + { + "epoch": 0.3630437235787884, + "grad_norm": 0.349609375, + "learning_rate": 0.0012971421120282162, + "loss": 0.168, + "step": 41823 + }, + { + "epoch": 0.36305240405899253, + "grad_norm": 0.427734375, + "learning_rate": 0.0012971130722641542, + "loss": 0.1118, + "step": 41824 + }, + { + "epoch": 0.3630610845391967, + "grad_norm": 0.26953125, + "learning_rate": 0.0012970840322845325, + "loss": 0.0991, + "step": 41825 + }, + { + "epoch": 0.36306976501940086, + "grad_norm": 0.5078125, + "learning_rate": 0.0012970549920893826, + "loss": 0.125, + "step": 41826 + }, + { + "epoch": 0.36307844549960505, + "grad_norm": 0.8203125, + "learning_rate": 0.001297025951678736, + "loss": 0.0977, + "step": 41827 + }, + { + "epoch": 0.3630871259798092, + "grad_norm": 0.93359375, + "learning_rate": 0.0012969969110526256, + "loss": 0.1465, + "step": 41828 + }, + { + "epoch": 0.3630958064600134, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012969678702110818, + "loss": 0.1045, + "step": 41829 + }, + { + "epoch": 0.3631044869402175, + "grad_norm": 0.42578125, + "learning_rate": 0.0012969388291541371, + "loss": 0.0898, + "step": 41830 + }, + { + "epoch": 0.3631131674204217, + "grad_norm": 0.259765625, + "learning_rate": 0.001296909787881823, + "loss": 0.0898, + "step": 41831 + }, + { + "epoch": 0.36312184790062585, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012968807463941713, + "loss": 0.1172, + "step": 41832 + }, + { + "epoch": 0.36313052838083004, + "grad_norm": 0.06396484375, + "learning_rate": 0.0012968517046912137, + "loss": 0.0825, + "step": 41833 + }, + { + "epoch": 0.3631392088610342, + "grad_norm": 0.287109375, + "learning_rate": 0.001296822662772982, + "loss": 0.0898, + "step": 41834 + }, + { + "epoch": 0.3631478893412384, + "grad_norm": 0.1240234375, + "learning_rate": 0.0012967936206395082, + "loss": 0.0728, + "step": 41835 + }, + { + "epoch": 0.3631565698214425, + "grad_norm": 0.2080078125, + "learning_rate": 0.0012967645782908236, + "loss": 0.1069, + "step": 41836 + }, + { + "epoch": 0.3631652503016467, + "grad_norm": 0.330078125, + "learning_rate": 0.0012967355357269604, + "loss": 0.0957, + "step": 41837 + }, + { + "epoch": 0.36317393078185084, + "grad_norm": 0.337890625, + "learning_rate": 0.0012967064929479502, + "loss": 0.1084, + "step": 41838 + }, + { + "epoch": 0.36318261126205503, + "grad_norm": 0.275390625, + "learning_rate": 0.0012966774499538246, + "loss": 0.0742, + "step": 41839 + }, + { + "epoch": 0.36319129174225917, + "grad_norm": 0.25, + "learning_rate": 0.0012966484067446153, + "loss": 0.1162, + "step": 41840 + }, + { + "epoch": 0.36319997222246336, + "grad_norm": 0.125, + "learning_rate": 0.0012966193633203544, + "loss": 0.1074, + "step": 41841 + }, + { + "epoch": 0.3632086527026675, + "grad_norm": 0.18359375, + "learning_rate": 0.001296590319681074, + "loss": 0.0908, + "step": 41842 + }, + { + "epoch": 0.3632173331828717, + "grad_norm": 0.1103515625, + "learning_rate": 0.0012965612758268049, + "loss": 0.1108, + "step": 41843 + }, + { + "epoch": 0.36322601366307583, + "grad_norm": 0.322265625, + "learning_rate": 0.0012965322317575795, + "loss": 0.0747, + "step": 41844 + }, + { + "epoch": 0.36323469414328, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012965031874734292, + "loss": 0.1328, + "step": 41845 + }, + { + "epoch": 0.36324337462348416, + "grad_norm": 0.1630859375, + "learning_rate": 0.001296474142974386, + "loss": 0.1245, + "step": 41846 + }, + { + "epoch": 0.36325205510368835, + "grad_norm": 0.2255859375, + "learning_rate": 0.0012964450982604818, + "loss": 0.1357, + "step": 41847 + }, + { + "epoch": 0.3632607355838925, + "grad_norm": 0.83984375, + "learning_rate": 0.0012964160533317482, + "loss": 0.1172, + "step": 41848 + }, + { + "epoch": 0.3632694160640967, + "grad_norm": 0.2373046875, + "learning_rate": 0.001296387008188217, + "loss": 0.0967, + "step": 41849 + }, + { + "epoch": 0.3632780965443008, + "grad_norm": 0.083984375, + "learning_rate": 0.0012963579628299197, + "loss": 0.0845, + "step": 41850 + }, + { + "epoch": 0.363286777024505, + "grad_norm": 0.2158203125, + "learning_rate": 0.0012963289172568883, + "loss": 0.0957, + "step": 41851 + }, + { + "epoch": 0.36329545750470915, + "grad_norm": 0.5078125, + "learning_rate": 0.001296299871469155, + "loss": 0.084, + "step": 41852 + }, + { + "epoch": 0.36330413798491334, + "grad_norm": 0.08349609375, + "learning_rate": 0.0012962708254667503, + "loss": 0.083, + "step": 41853 + }, + { + "epoch": 0.3633128184651175, + "grad_norm": 0.478515625, + "learning_rate": 0.0012962417792497071, + "loss": 0.1177, + "step": 41854 + }, + { + "epoch": 0.3633214989453217, + "grad_norm": 0.10009765625, + "learning_rate": 0.001296212732818057, + "loss": 0.0884, + "step": 41855 + }, + { + "epoch": 0.3633301794255258, + "grad_norm": 0.08984375, + "learning_rate": 0.0012961836861718318, + "loss": 0.0679, + "step": 41856 + }, + { + "epoch": 0.36333885990573, + "grad_norm": 0.07275390625, + "learning_rate": 0.0012961546393110625, + "loss": 0.084, + "step": 41857 + }, + { + "epoch": 0.36334754038593414, + "grad_norm": 0.072265625, + "learning_rate": 0.0012961255922357819, + "loss": 0.0825, + "step": 41858 + }, + { + "epoch": 0.36335622086613834, + "grad_norm": 0.28515625, + "learning_rate": 0.001296096544946021, + "loss": 0.0869, + "step": 41859 + }, + { + "epoch": 0.3633649013463425, + "grad_norm": 0.25, + "learning_rate": 0.0012960674974418125, + "loss": 0.1055, + "step": 41860 + }, + { + "epoch": 0.36337358182654667, + "grad_norm": 0.470703125, + "learning_rate": 0.0012960384497231869, + "loss": 0.1025, + "step": 41861 + }, + { + "epoch": 0.3633822623067508, + "grad_norm": 0.294921875, + "learning_rate": 0.0012960094017901767, + "loss": 0.0869, + "step": 41862 + }, + { + "epoch": 0.363390942786955, + "grad_norm": 0.49609375, + "learning_rate": 0.0012959803536428136, + "loss": 0.1426, + "step": 41863 + }, + { + "epoch": 0.36339962326715913, + "grad_norm": 0.1298828125, + "learning_rate": 0.0012959513052811294, + "loss": 0.0889, + "step": 41864 + }, + { + "epoch": 0.3634083037473633, + "grad_norm": 0.5546875, + "learning_rate": 0.0012959222567051555, + "loss": 0.1021, + "step": 41865 + }, + { + "epoch": 0.36341698422756746, + "grad_norm": 0.58984375, + "learning_rate": 0.0012958932079149242, + "loss": 0.1758, + "step": 41866 + }, + { + "epoch": 0.36342566470777166, + "grad_norm": 0.400390625, + "learning_rate": 0.0012958641589104671, + "loss": 0.1387, + "step": 41867 + }, + { + "epoch": 0.3634343451879758, + "grad_norm": 0.1748046875, + "learning_rate": 0.001295835109691816, + "loss": 0.1514, + "step": 41868 + }, + { + "epoch": 0.36344302566818, + "grad_norm": 0.1220703125, + "learning_rate": 0.0012958060602590025, + "loss": 0.0703, + "step": 41869 + }, + { + "epoch": 0.3634517061483841, + "grad_norm": 0.1083984375, + "learning_rate": 0.0012957770106120582, + "loss": 0.0884, + "step": 41870 + }, + { + "epoch": 0.3634603866285883, + "grad_norm": 0.421875, + "learning_rate": 0.0012957479607510155, + "loss": 0.1118, + "step": 41871 + }, + { + "epoch": 0.36346906710879245, + "grad_norm": 0.0703125, + "learning_rate": 0.0012957189106759053, + "loss": 0.0811, + "step": 41872 + }, + { + "epoch": 0.36347774758899665, + "grad_norm": 0.15234375, + "learning_rate": 0.0012956898603867601, + "loss": 0.1035, + "step": 41873 + }, + { + "epoch": 0.3634864280692008, + "grad_norm": 0.0986328125, + "learning_rate": 0.001295660809883611, + "loss": 0.1045, + "step": 41874 + }, + { + "epoch": 0.363495108549405, + "grad_norm": 0.10595703125, + "learning_rate": 0.0012956317591664907, + "loss": 0.0972, + "step": 41875 + }, + { + "epoch": 0.3635037890296091, + "grad_norm": 0.419921875, + "learning_rate": 0.0012956027082354305, + "loss": 0.0947, + "step": 41876 + }, + { + "epoch": 0.3635124695098133, + "grad_norm": 0.1005859375, + "learning_rate": 0.0012955736570904619, + "loss": 0.0732, + "step": 41877 + }, + { + "epoch": 0.36352114999001744, + "grad_norm": 0.11083984375, + "learning_rate": 0.001295544605731617, + "loss": 0.0957, + "step": 41878 + }, + { + "epoch": 0.36352983047022164, + "grad_norm": 0.1357421875, + "learning_rate": 0.001295515554158927, + "loss": 0.084, + "step": 41879 + }, + { + "epoch": 0.3635385109504258, + "grad_norm": 0.201171875, + "learning_rate": 0.0012954865023724245, + "loss": 0.1074, + "step": 41880 + }, + { + "epoch": 0.36354719143062997, + "grad_norm": 0.65234375, + "learning_rate": 0.001295457450372141, + "loss": 0.106, + "step": 41881 + }, + { + "epoch": 0.3635558719108341, + "grad_norm": 0.0830078125, + "learning_rate": 0.001295428398158108, + "loss": 0.0869, + "step": 41882 + }, + { + "epoch": 0.3635645523910383, + "grad_norm": 0.388671875, + "learning_rate": 0.0012953993457303571, + "loss": 0.1445, + "step": 41883 + }, + { + "epoch": 0.36357323287124244, + "grad_norm": 0.09375, + "learning_rate": 0.001295370293088921, + "loss": 0.084, + "step": 41884 + }, + { + "epoch": 0.36358191335144663, + "grad_norm": 0.55078125, + "learning_rate": 0.0012953412402338306, + "loss": 0.1406, + "step": 41885 + }, + { + "epoch": 0.36359059383165077, + "grad_norm": 0.287109375, + "learning_rate": 0.0012953121871651183, + "loss": 0.1113, + "step": 41886 + }, + { + "epoch": 0.36359927431185496, + "grad_norm": 0.34765625, + "learning_rate": 0.0012952831338828156, + "loss": 0.0674, + "step": 41887 + }, + { + "epoch": 0.3636079547920591, + "grad_norm": 0.8359375, + "learning_rate": 0.0012952540803869534, + "loss": 0.0938, + "step": 41888 + }, + { + "epoch": 0.3636166352722633, + "grad_norm": 0.1171875, + "learning_rate": 0.001295225026677565, + "loss": 0.0791, + "step": 41889 + }, + { + "epoch": 0.3636253157524674, + "grad_norm": 0.345703125, + "learning_rate": 0.001295195972754681, + "loss": 0.1035, + "step": 41890 + }, + { + "epoch": 0.3636339962326716, + "grad_norm": 0.70703125, + "learning_rate": 0.001295166918618334, + "loss": 0.0913, + "step": 41891 + }, + { + "epoch": 0.36364267671287576, + "grad_norm": 0.25390625, + "learning_rate": 0.001295137864268555, + "loss": 0.1196, + "step": 41892 + }, + { + "epoch": 0.3636513571930799, + "grad_norm": 0.0859375, + "learning_rate": 0.0012951088097053762, + "loss": 0.0947, + "step": 41893 + }, + { + "epoch": 0.3636600376732841, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012950797549288295, + "loss": 0.063, + "step": 41894 + }, + { + "epoch": 0.3636687181534882, + "grad_norm": 0.189453125, + "learning_rate": 0.0012950506999389467, + "loss": 0.0977, + "step": 41895 + }, + { + "epoch": 0.3636773986336924, + "grad_norm": 0.2265625, + "learning_rate": 0.001295021644735759, + "loss": 0.1172, + "step": 41896 + }, + { + "epoch": 0.36368607911389655, + "grad_norm": 0.318359375, + "learning_rate": 0.0012949925893192987, + "loss": 0.1074, + "step": 41897 + }, + { + "epoch": 0.36369475959410075, + "grad_norm": 0.20703125, + "learning_rate": 0.0012949635336895976, + "loss": 0.0889, + "step": 41898 + }, + { + "epoch": 0.3637034400743049, + "grad_norm": 0.6953125, + "learning_rate": 0.001294934477846687, + "loss": 0.1006, + "step": 41899 + }, + { + "epoch": 0.3637121205545091, + "grad_norm": 0.30859375, + "learning_rate": 0.0012949054217905993, + "loss": 0.1289, + "step": 41900 + }, + { + "epoch": 0.3637208010347132, + "grad_norm": 0.3515625, + "learning_rate": 0.0012948763655213657, + "loss": 0.1152, + "step": 41901 + }, + { + "epoch": 0.3637294815149174, + "grad_norm": 0.76171875, + "learning_rate": 0.0012948473090390183, + "loss": 0.123, + "step": 41902 + }, + { + "epoch": 0.36373816199512155, + "grad_norm": 0.400390625, + "learning_rate": 0.001294818252343589, + "loss": 0.1226, + "step": 41903 + }, + { + "epoch": 0.36374684247532574, + "grad_norm": 0.255859375, + "learning_rate": 0.0012947891954351093, + "loss": 0.1235, + "step": 41904 + }, + { + "epoch": 0.3637555229555299, + "grad_norm": 0.23046875, + "learning_rate": 0.001294760138313611, + "loss": 0.0879, + "step": 41905 + }, + { + "epoch": 0.36376420343573407, + "grad_norm": 0.107421875, + "learning_rate": 0.0012947310809791259, + "loss": 0.0962, + "step": 41906 + }, + { + "epoch": 0.3637728839159382, + "grad_norm": 0.7109375, + "learning_rate": 0.001294702023431686, + "loss": 0.083, + "step": 41907 + }, + { + "epoch": 0.3637815643961424, + "grad_norm": 0.46875, + "learning_rate": 0.0012946729656713228, + "loss": 0.1309, + "step": 41908 + }, + { + "epoch": 0.36379024487634654, + "grad_norm": 0.6875, + "learning_rate": 0.0012946439076980678, + "loss": 0.1309, + "step": 41909 + }, + { + "epoch": 0.36379892535655073, + "grad_norm": 0.1650390625, + "learning_rate": 0.0012946148495119535, + "loss": 0.1079, + "step": 41910 + }, + { + "epoch": 0.36380760583675487, + "grad_norm": 0.28125, + "learning_rate": 0.0012945857911130113, + "loss": 0.0562, + "step": 41911 + }, + { + "epoch": 0.36381628631695906, + "grad_norm": 0.1142578125, + "learning_rate": 0.001294556732501273, + "loss": 0.1348, + "step": 41912 + }, + { + "epoch": 0.3638249667971632, + "grad_norm": 0.10791015625, + "learning_rate": 0.0012945276736767706, + "loss": 0.1104, + "step": 41913 + }, + { + "epoch": 0.3638336472773674, + "grad_norm": 0.267578125, + "learning_rate": 0.0012944986146395353, + "loss": 0.1016, + "step": 41914 + }, + { + "epoch": 0.3638423277575715, + "grad_norm": 0.169921875, + "learning_rate": 0.0012944695553895992, + "loss": 0.084, + "step": 41915 + }, + { + "epoch": 0.3638510082377757, + "grad_norm": 0.296875, + "learning_rate": 0.0012944404959269943, + "loss": 0.125, + "step": 41916 + }, + { + "epoch": 0.36385968871797986, + "grad_norm": 0.2197265625, + "learning_rate": 0.0012944114362517525, + "loss": 0.0928, + "step": 41917 + }, + { + "epoch": 0.36386836919818405, + "grad_norm": 0.478515625, + "learning_rate": 0.0012943823763639048, + "loss": 0.1475, + "step": 41918 + }, + { + "epoch": 0.3638770496783882, + "grad_norm": 0.337890625, + "learning_rate": 0.0012943533162634838, + "loss": 0.1211, + "step": 41919 + }, + { + "epoch": 0.3638857301585924, + "grad_norm": 0.314453125, + "learning_rate": 0.0012943242559505209, + "loss": 0.082, + "step": 41920 + }, + { + "epoch": 0.3638944106387965, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012942951954250477, + "loss": 0.0879, + "step": 41921 + }, + { + "epoch": 0.3639030911190007, + "grad_norm": 0.5390625, + "learning_rate": 0.0012942661346870962, + "loss": 0.1045, + "step": 41922 + }, + { + "epoch": 0.36391177159920485, + "grad_norm": 0.328125, + "learning_rate": 0.001294237073736698, + "loss": 0.1147, + "step": 41923 + }, + { + "epoch": 0.36392045207940904, + "grad_norm": 0.08154296875, + "learning_rate": 0.0012942080125738855, + "loss": 0.0986, + "step": 41924 + }, + { + "epoch": 0.3639291325596132, + "grad_norm": 0.427734375, + "learning_rate": 0.0012941789511986897, + "loss": 0.1074, + "step": 41925 + }, + { + "epoch": 0.36393781303981737, + "grad_norm": 0.1494140625, + "learning_rate": 0.001294149889611143, + "loss": 0.1143, + "step": 41926 + }, + { + "epoch": 0.3639464935200215, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012941208278112765, + "loss": 0.1699, + "step": 41927 + }, + { + "epoch": 0.3639551740002257, + "grad_norm": 0.54296875, + "learning_rate": 0.001294091765799123, + "loss": 0.1504, + "step": 41928 + }, + { + "epoch": 0.36396385448042984, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012940627035747132, + "loss": 0.123, + "step": 41929 + }, + { + "epoch": 0.36397253496063403, + "grad_norm": 0.255859375, + "learning_rate": 0.0012940336411380792, + "loss": 0.123, + "step": 41930 + }, + { + "epoch": 0.36398121544083817, + "grad_norm": 0.2294921875, + "learning_rate": 0.001294004578489253, + "loss": 0.1641, + "step": 41931 + }, + { + "epoch": 0.36398989592104236, + "grad_norm": 0.380859375, + "learning_rate": 0.0012939755156282666, + "loss": 0.0947, + "step": 41932 + }, + { + "epoch": 0.3639985764012465, + "grad_norm": 0.150390625, + "learning_rate": 0.0012939464525551514, + "loss": 0.1289, + "step": 41933 + }, + { + "epoch": 0.3640072568814507, + "grad_norm": 0.08740234375, + "learning_rate": 0.0012939173892699393, + "loss": 0.0884, + "step": 41934 + }, + { + "epoch": 0.36401593736165483, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012938883257726619, + "loss": 0.0996, + "step": 41935 + }, + { + "epoch": 0.364024617841859, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012938592620633513, + "loss": 0.0913, + "step": 41936 + }, + { + "epoch": 0.36403329832206316, + "grad_norm": 0.373046875, + "learning_rate": 0.001293830198142039, + "loss": 0.1885, + "step": 41937 + }, + { + "epoch": 0.36404197880226735, + "grad_norm": 0.294921875, + "learning_rate": 0.001293801134008757, + "loss": 0.0977, + "step": 41938 + }, + { + "epoch": 0.3640506592824715, + "grad_norm": 0.361328125, + "learning_rate": 0.001293772069663537, + "loss": 0.0996, + "step": 41939 + }, + { + "epoch": 0.3640593397626757, + "grad_norm": 0.240234375, + "learning_rate": 0.0012937430051064105, + "loss": 0.0762, + "step": 41940 + }, + { + "epoch": 0.3640680202428798, + "grad_norm": 0.28515625, + "learning_rate": 0.0012937139403374099, + "loss": 0.1191, + "step": 41941 + }, + { + "epoch": 0.364076700723084, + "grad_norm": 0.1015625, + "learning_rate": 0.0012936848753565663, + "loss": 0.103, + "step": 41942 + }, + { + "epoch": 0.36408538120328815, + "grad_norm": 0.322265625, + "learning_rate": 0.0012936558101639122, + "loss": 0.0913, + "step": 41943 + }, + { + "epoch": 0.36409406168349234, + "grad_norm": 0.33203125, + "learning_rate": 0.0012936267447594787, + "loss": 0.1104, + "step": 41944 + }, + { + "epoch": 0.3641027421636965, + "grad_norm": 0.515625, + "learning_rate": 0.0012935976791432983, + "loss": 0.0957, + "step": 41945 + }, + { + "epoch": 0.3641114226439007, + "grad_norm": 0.171875, + "learning_rate": 0.001293568613315402, + "loss": 0.0977, + "step": 41946 + }, + { + "epoch": 0.3641201031241048, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012935395472758226, + "loss": 0.0923, + "step": 41947 + }, + { + "epoch": 0.364128783604309, + "grad_norm": 0.2421875, + "learning_rate": 0.0012935104810245905, + "loss": 0.0859, + "step": 41948 + }, + { + "epoch": 0.36413746408451314, + "grad_norm": 0.50390625, + "learning_rate": 0.0012934814145617384, + "loss": 0.1084, + "step": 41949 + }, + { + "epoch": 0.36414614456471733, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012934523478872982, + "loss": 0.1089, + "step": 41950 + }, + { + "epoch": 0.36415482504492147, + "grad_norm": 0.439453125, + "learning_rate": 0.0012934232810013011, + "loss": 0.106, + "step": 41951 + }, + { + "epoch": 0.36416350552512566, + "grad_norm": 0.123046875, + "learning_rate": 0.0012933942139037795, + "loss": 0.0942, + "step": 41952 + }, + { + "epoch": 0.3641721860053298, + "grad_norm": 0.53125, + "learning_rate": 0.0012933651465947646, + "loss": 0.1172, + "step": 41953 + }, + { + "epoch": 0.364180866485534, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012933360790742887, + "loss": 0.0859, + "step": 41954 + }, + { + "epoch": 0.36418954696573813, + "grad_norm": 0.279296875, + "learning_rate": 0.0012933070113423835, + "loss": 0.1064, + "step": 41955 + }, + { + "epoch": 0.3641982274459423, + "grad_norm": 0.1884765625, + "learning_rate": 0.0012932779433990804, + "loss": 0.125, + "step": 41956 + }, + { + "epoch": 0.36420690792614646, + "grad_norm": 0.255859375, + "learning_rate": 0.0012932488752444115, + "loss": 0.0991, + "step": 41957 + }, + { + "epoch": 0.36421558840635065, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012932198068784087, + "loss": 0.0869, + "step": 41958 + }, + { + "epoch": 0.3642242688865548, + "grad_norm": 0.185546875, + "learning_rate": 0.0012931907383011032, + "loss": 0.1143, + "step": 41959 + }, + { + "epoch": 0.364232949366759, + "grad_norm": 0.66015625, + "learning_rate": 0.0012931616695125276, + "loss": 0.0898, + "step": 41960 + }, + { + "epoch": 0.3642416298469631, + "grad_norm": 0.23828125, + "learning_rate": 0.0012931326005127133, + "loss": 0.0801, + "step": 41961 + }, + { + "epoch": 0.3642503103271673, + "grad_norm": 0.0947265625, + "learning_rate": 0.001293103531301692, + "loss": 0.0747, + "step": 41962 + }, + { + "epoch": 0.36425899080737145, + "grad_norm": 0.5390625, + "learning_rate": 0.0012930744618794955, + "loss": 0.1641, + "step": 41963 + }, + { + "epoch": 0.36426767128757565, + "grad_norm": 0.380859375, + "learning_rate": 0.0012930453922461557, + "loss": 0.1309, + "step": 41964 + }, + { + "epoch": 0.3642763517677798, + "grad_norm": 0.765625, + "learning_rate": 0.0012930163224017044, + "loss": 0.0972, + "step": 41965 + }, + { + "epoch": 0.364285032247984, + "grad_norm": 0.1826171875, + "learning_rate": 0.0012929872523461735, + "loss": 0.1084, + "step": 41966 + }, + { + "epoch": 0.3642937127281881, + "grad_norm": 0.1943359375, + "learning_rate": 0.0012929581820795945, + "loss": 0.0923, + "step": 41967 + }, + { + "epoch": 0.3643023932083923, + "grad_norm": 0.34765625, + "learning_rate": 0.0012929291116019993, + "loss": 0.0815, + "step": 41968 + }, + { + "epoch": 0.36431107368859644, + "grad_norm": 0.375, + "learning_rate": 0.0012929000409134198, + "loss": 0.0977, + "step": 41969 + }, + { + "epoch": 0.36431975416880064, + "grad_norm": 0.302734375, + "learning_rate": 0.0012928709700138876, + "loss": 0.1172, + "step": 41970 + }, + { + "epoch": 0.3643284346490048, + "grad_norm": 0.1962890625, + "learning_rate": 0.0012928418989034346, + "loss": 0.0854, + "step": 41971 + }, + { + "epoch": 0.36433711512920897, + "grad_norm": 0.0908203125, + "learning_rate": 0.0012928128275820927, + "loss": 0.1494, + "step": 41972 + }, + { + "epoch": 0.3643457956094131, + "grad_norm": 0.3125, + "learning_rate": 0.0012927837560498935, + "loss": 0.0762, + "step": 41973 + }, + { + "epoch": 0.3643544760896173, + "grad_norm": 0.119140625, + "learning_rate": 0.0012927546843068689, + "loss": 0.1084, + "step": 41974 + }, + { + "epoch": 0.36436315656982143, + "grad_norm": 0.271484375, + "learning_rate": 0.0012927256123530509, + "loss": 0.1118, + "step": 41975 + }, + { + "epoch": 0.3643718370500256, + "grad_norm": 0.302734375, + "learning_rate": 0.0012926965401884709, + "loss": 0.0815, + "step": 41976 + }, + { + "epoch": 0.36438051753022976, + "grad_norm": 0.1728515625, + "learning_rate": 0.001292667467813161, + "loss": 0.1045, + "step": 41977 + }, + { + "epoch": 0.36438919801043396, + "grad_norm": 0.72265625, + "learning_rate": 0.0012926383952271525, + "loss": 0.1162, + "step": 41978 + }, + { + "epoch": 0.3643978784906381, + "grad_norm": 0.34375, + "learning_rate": 0.0012926093224304776, + "loss": 0.1011, + "step": 41979 + }, + { + "epoch": 0.3644065589708423, + "grad_norm": 0.1669921875, + "learning_rate": 0.0012925802494231683, + "loss": 0.0752, + "step": 41980 + }, + { + "epoch": 0.3644152394510464, + "grad_norm": 0.1767578125, + "learning_rate": 0.0012925511762052558, + "loss": 0.1025, + "step": 41981 + }, + { + "epoch": 0.3644239199312506, + "grad_norm": 0.50390625, + "learning_rate": 0.0012925221027767727, + "loss": 0.1006, + "step": 41982 + }, + { + "epoch": 0.36443260041145475, + "grad_norm": 0.17578125, + "learning_rate": 0.00129249302913775, + "loss": 0.1094, + "step": 41983 + }, + { + "epoch": 0.36444128089165895, + "grad_norm": 0.359375, + "learning_rate": 0.0012924639552882198, + "loss": 0.0781, + "step": 41984 + }, + { + "epoch": 0.3644499613718631, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012924348812282137, + "loss": 0.0776, + "step": 41985 + }, + { + "epoch": 0.3644586418520673, + "grad_norm": 0.11572265625, + "learning_rate": 0.0012924058069577642, + "loss": 0.1001, + "step": 41986 + }, + { + "epoch": 0.3644673223322714, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012923767324769025, + "loss": 0.0806, + "step": 41987 + }, + { + "epoch": 0.3644760028124756, + "grad_norm": 0.62109375, + "learning_rate": 0.00129234765778566, + "loss": 0.1504, + "step": 41988 + }, + { + "epoch": 0.36448468329267975, + "grad_norm": 0.36328125, + "learning_rate": 0.0012923185828840696, + "loss": 0.1211, + "step": 41989 + }, + { + "epoch": 0.36449336377288394, + "grad_norm": 0.1015625, + "learning_rate": 0.0012922895077721623, + "loss": 0.125, + "step": 41990 + }, + { + "epoch": 0.3645020442530881, + "grad_norm": 0.7578125, + "learning_rate": 0.0012922604324499699, + "loss": 0.0967, + "step": 41991 + }, + { + "epoch": 0.36451072473329227, + "grad_norm": 0.458984375, + "learning_rate": 0.0012922313569175246, + "loss": 0.1016, + "step": 41992 + }, + { + "epoch": 0.3645194052134964, + "grad_norm": 0.337890625, + "learning_rate": 0.001292202281174858, + "loss": 0.0942, + "step": 41993 + }, + { + "epoch": 0.3645280856937006, + "grad_norm": 0.69140625, + "learning_rate": 0.001292173205222002, + "loss": 0.083, + "step": 41994 + }, + { + "epoch": 0.36453676617390474, + "grad_norm": 0.263671875, + "learning_rate": 0.001292144129058988, + "loss": 0.0845, + "step": 41995 + }, + { + "epoch": 0.36454544665410893, + "grad_norm": 0.259765625, + "learning_rate": 0.0012921150526858476, + "loss": 0.082, + "step": 41996 + }, + { + "epoch": 0.36455412713431307, + "grad_norm": 0.1796875, + "learning_rate": 0.0012920859761026138, + "loss": 0.1172, + "step": 41997 + }, + { + "epoch": 0.36456280761451726, + "grad_norm": 0.0947265625, + "learning_rate": 0.001292056899309317, + "loss": 0.1055, + "step": 41998 + }, + { + "epoch": 0.3645714880947214, + "grad_norm": 0.4296875, + "learning_rate": 0.0012920278223059905, + "loss": 0.1245, + "step": 41999 + }, + { + "epoch": 0.3645801685749256, + "grad_norm": 0.181640625, + "learning_rate": 0.0012919987450926646, + "loss": 0.1045, + "step": 42000 + }, + { + "epoch": 0.3645888490551297, + "grad_norm": 0.2265625, + "learning_rate": 0.0012919696676693723, + "loss": 0.0977, + "step": 42001 + }, + { + "epoch": 0.3645975295353339, + "grad_norm": 0.59765625, + "learning_rate": 0.0012919405900361446, + "loss": 0.1064, + "step": 42002 + }, + { + "epoch": 0.36460621001553806, + "grad_norm": 0.75, + "learning_rate": 0.0012919115121930135, + "loss": 0.1289, + "step": 42003 + }, + { + "epoch": 0.36461489049574225, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012918824341400109, + "loss": 0.0928, + "step": 42004 + }, + { + "epoch": 0.3646235709759464, + "grad_norm": 0.73046875, + "learning_rate": 0.0012918533558771684, + "loss": 0.0889, + "step": 42005 + }, + { + "epoch": 0.3646322514561506, + "grad_norm": 0.23828125, + "learning_rate": 0.0012918242774045182, + "loss": 0.0957, + "step": 42006 + }, + { + "epoch": 0.3646409319363547, + "grad_norm": 0.51171875, + "learning_rate": 0.0012917951987220914, + "loss": 0.1069, + "step": 42007 + }, + { + "epoch": 0.3646496124165589, + "grad_norm": 0.185546875, + "learning_rate": 0.0012917661198299207, + "loss": 0.1079, + "step": 42008 + }, + { + "epoch": 0.36465829289676305, + "grad_norm": 0.1640625, + "learning_rate": 0.0012917370407280373, + "loss": 0.0913, + "step": 42009 + }, + { + "epoch": 0.36466697337696724, + "grad_norm": 0.07666015625, + "learning_rate": 0.0012917079614164731, + "loss": 0.0781, + "step": 42010 + }, + { + "epoch": 0.3646756538571714, + "grad_norm": 0.11181640625, + "learning_rate": 0.00129167888189526, + "loss": 0.0889, + "step": 42011 + }, + { + "epoch": 0.36468433433737557, + "grad_norm": 0.50390625, + "learning_rate": 0.0012916498021644301, + "loss": 0.1309, + "step": 42012 + }, + { + "epoch": 0.3646930148175797, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012916207222240144, + "loss": 0.0938, + "step": 42013 + }, + { + "epoch": 0.3647016952977839, + "grad_norm": 0.3359375, + "learning_rate": 0.0012915916420740453, + "loss": 0.0781, + "step": 42014 + }, + { + "epoch": 0.36471037577798804, + "grad_norm": 0.32421875, + "learning_rate": 0.0012915625617145546, + "loss": 0.0684, + "step": 42015 + }, + { + "epoch": 0.3647190562581922, + "grad_norm": 0.462890625, + "learning_rate": 0.0012915334811455738, + "loss": 0.103, + "step": 42016 + }, + { + "epoch": 0.36472773673839637, + "grad_norm": 0.1357421875, + "learning_rate": 0.0012915044003671348, + "loss": 0.103, + "step": 42017 + }, + { + "epoch": 0.3647364172186005, + "grad_norm": 0.166015625, + "learning_rate": 0.0012914753193792694, + "loss": 0.084, + "step": 42018 + }, + { + "epoch": 0.3647450976988047, + "grad_norm": 0.291015625, + "learning_rate": 0.0012914462381820098, + "loss": 0.0898, + "step": 42019 + }, + { + "epoch": 0.36475377817900884, + "grad_norm": 0.09423828125, + "learning_rate": 0.0012914171567753874, + "loss": 0.1094, + "step": 42020 + }, + { + "epoch": 0.36476245865921303, + "grad_norm": 0.353515625, + "learning_rate": 0.001291388075159434, + "loss": 0.0767, + "step": 42021 + }, + { + "epoch": 0.36477113913941717, + "grad_norm": 0.5078125, + "learning_rate": 0.0012913589933341814, + "loss": 0.0942, + "step": 42022 + }, + { + "epoch": 0.36477981961962136, + "grad_norm": 0.341796875, + "learning_rate": 0.0012913299112996615, + "loss": 0.1094, + "step": 42023 + }, + { + "epoch": 0.3647885000998255, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012913008290559064, + "loss": 0.0894, + "step": 42024 + }, + { + "epoch": 0.3647971805800297, + "grad_norm": 0.1708984375, + "learning_rate": 0.001291271746602947, + "loss": 0.1309, + "step": 42025 + }, + { + "epoch": 0.3648058610602338, + "grad_norm": 0.130859375, + "learning_rate": 0.001291242663940816, + "loss": 0.0869, + "step": 42026 + }, + { + "epoch": 0.364814541540438, + "grad_norm": 0.462890625, + "learning_rate": 0.0012912135810695447, + "loss": 0.1128, + "step": 42027 + }, + { + "epoch": 0.36482322202064216, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012911844979891654, + "loss": 0.1055, + "step": 42028 + }, + { + "epoch": 0.36483190250084635, + "grad_norm": 0.51953125, + "learning_rate": 0.0012911554146997095, + "loss": 0.0889, + "step": 42029 + }, + { + "epoch": 0.3648405829810505, + "grad_norm": 0.12890625, + "learning_rate": 0.0012911263312012087, + "loss": 0.0923, + "step": 42030 + }, + { + "epoch": 0.3648492634612547, + "grad_norm": 0.4765625, + "learning_rate": 0.0012910972474936952, + "loss": 0.0986, + "step": 42031 + }, + { + "epoch": 0.3648579439414588, + "grad_norm": 0.310546875, + "learning_rate": 0.001291068163577201, + "loss": 0.1289, + "step": 42032 + }, + { + "epoch": 0.364866624421663, + "grad_norm": 0.51953125, + "learning_rate": 0.001291039079451757, + "loss": 0.0732, + "step": 42033 + }, + { + "epoch": 0.36487530490186715, + "grad_norm": 0.474609375, + "learning_rate": 0.001291009995117396, + "loss": 0.1235, + "step": 42034 + }, + { + "epoch": 0.36488398538207134, + "grad_norm": 0.095703125, + "learning_rate": 0.0012909809105741486, + "loss": 0.0669, + "step": 42035 + }, + { + "epoch": 0.3648926658622755, + "grad_norm": 0.431640625, + "learning_rate": 0.001290951825822048, + "loss": 0.1514, + "step": 42036 + }, + { + "epoch": 0.36490134634247967, + "grad_norm": 0.12451171875, + "learning_rate": 0.001290922740861125, + "loss": 0.0898, + "step": 42037 + }, + { + "epoch": 0.3649100268226838, + "grad_norm": 0.130859375, + "learning_rate": 0.001290893655691412, + "loss": 0.1016, + "step": 42038 + }, + { + "epoch": 0.364918707302888, + "grad_norm": 0.10595703125, + "learning_rate": 0.0012908645703129402, + "loss": 0.0898, + "step": 42039 + }, + { + "epoch": 0.36492738778309214, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012908354847257421, + "loss": 0.124, + "step": 42040 + }, + { + "epoch": 0.36493606826329633, + "grad_norm": 0.1103515625, + "learning_rate": 0.0012908063989298493, + "loss": 0.0742, + "step": 42041 + }, + { + "epoch": 0.36494474874350047, + "grad_norm": 0.357421875, + "learning_rate": 0.0012907773129252934, + "loss": 0.0737, + "step": 42042 + }, + { + "epoch": 0.36495342922370466, + "grad_norm": 0.5, + "learning_rate": 0.001290748226712106, + "loss": 0.1543, + "step": 42043 + }, + { + "epoch": 0.3649621097039088, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012907191402903194, + "loss": 0.0918, + "step": 42044 + }, + { + "epoch": 0.364970790184113, + "grad_norm": 0.271484375, + "learning_rate": 0.0012906900536599653, + "loss": 0.1279, + "step": 42045 + }, + { + "epoch": 0.36497947066431713, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012906609668210754, + "loss": 0.1172, + "step": 42046 + }, + { + "epoch": 0.3649881511445213, + "grad_norm": 0.51953125, + "learning_rate": 0.0012906318797736814, + "loss": 0.1309, + "step": 42047 + }, + { + "epoch": 0.36499683162472546, + "grad_norm": 0.1103515625, + "learning_rate": 0.0012906027925178151, + "loss": 0.0884, + "step": 42048 + }, + { + "epoch": 0.36500551210492965, + "grad_norm": 0.1728515625, + "learning_rate": 0.001290573705053509, + "loss": 0.0986, + "step": 42049 + }, + { + "epoch": 0.3650141925851338, + "grad_norm": 0.44921875, + "learning_rate": 0.0012905446173807939, + "loss": 0.0942, + "step": 42050 + }, + { + "epoch": 0.365022873065338, + "grad_norm": 0.494140625, + "learning_rate": 0.0012905155294997022, + "loss": 0.0693, + "step": 42051 + }, + { + "epoch": 0.3650315535455421, + "grad_norm": 0.09326171875, + "learning_rate": 0.0012904864414102655, + "loss": 0.0781, + "step": 42052 + }, + { + "epoch": 0.3650402340257463, + "grad_norm": 0.345703125, + "learning_rate": 0.0012904573531125156, + "loss": 0.0835, + "step": 42053 + }, + { + "epoch": 0.36504891450595045, + "grad_norm": 0.1630859375, + "learning_rate": 0.0012904282646064849, + "loss": 0.0806, + "step": 42054 + }, + { + "epoch": 0.36505759498615464, + "grad_norm": 0.3125, + "learning_rate": 0.0012903991758922042, + "loss": 0.0664, + "step": 42055 + }, + { + "epoch": 0.3650662754663588, + "grad_norm": 0.3203125, + "learning_rate": 0.0012903700869697056, + "loss": 0.0879, + "step": 42056 + }, + { + "epoch": 0.365074955946563, + "grad_norm": 0.1220703125, + "learning_rate": 0.0012903409978390218, + "loss": 0.0962, + "step": 42057 + }, + { + "epoch": 0.3650836364267671, + "grad_norm": 0.314453125, + "learning_rate": 0.0012903119085001839, + "loss": 0.1338, + "step": 42058 + }, + { + "epoch": 0.3650923169069713, + "grad_norm": 0.69921875, + "learning_rate": 0.0012902828189532234, + "loss": 0.1387, + "step": 42059 + }, + { + "epoch": 0.36510099738717544, + "grad_norm": 0.3515625, + "learning_rate": 0.0012902537291981729, + "loss": 0.1641, + "step": 42060 + }, + { + "epoch": 0.36510967786737963, + "grad_norm": 0.380859375, + "learning_rate": 0.0012902246392350633, + "loss": 0.127, + "step": 42061 + }, + { + "epoch": 0.36511835834758377, + "grad_norm": 0.1572265625, + "learning_rate": 0.001290195549063927, + "loss": 0.1465, + "step": 42062 + }, + { + "epoch": 0.36512703882778796, + "grad_norm": 0.330078125, + "learning_rate": 0.0012901664586847958, + "loss": 0.1172, + "step": 42063 + }, + { + "epoch": 0.3651357193079921, + "grad_norm": 0.455078125, + "learning_rate": 0.0012901373680977014, + "loss": 0.1318, + "step": 42064 + }, + { + "epoch": 0.3651443997881963, + "grad_norm": 0.0771484375, + "learning_rate": 0.0012901082773026757, + "loss": 0.0713, + "step": 42065 + }, + { + "epoch": 0.36515308026840043, + "grad_norm": 0.57421875, + "learning_rate": 0.0012900791862997501, + "loss": 0.083, + "step": 42066 + }, + { + "epoch": 0.3651617607486046, + "grad_norm": 0.55078125, + "learning_rate": 0.0012900500950889572, + "loss": 0.1387, + "step": 42067 + }, + { + "epoch": 0.36517044122880876, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012900210036703283, + "loss": 0.123, + "step": 42068 + }, + { + "epoch": 0.36517912170901295, + "grad_norm": 0.1376953125, + "learning_rate": 0.001289991912043895, + "loss": 0.1152, + "step": 42069 + }, + { + "epoch": 0.3651878021892171, + "grad_norm": 0.1201171875, + "learning_rate": 0.0012899628202096896, + "loss": 0.0977, + "step": 42070 + }, + { + "epoch": 0.3651964826694213, + "grad_norm": 0.09521484375, + "learning_rate": 0.0012899337281677438, + "loss": 0.0581, + "step": 42071 + }, + { + "epoch": 0.3652051631496254, + "grad_norm": 0.361328125, + "learning_rate": 0.0012899046359180892, + "loss": 0.1172, + "step": 42072 + }, + { + "epoch": 0.3652138436298296, + "grad_norm": 0.26171875, + "learning_rate": 0.0012898755434607577, + "loss": 0.1113, + "step": 42073 + }, + { + "epoch": 0.36522252411003375, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012898464507957812, + "loss": 0.0732, + "step": 42074 + }, + { + "epoch": 0.36523120459023795, + "grad_norm": 0.35546875, + "learning_rate": 0.0012898173579231913, + "loss": 0.0811, + "step": 42075 + }, + { + "epoch": 0.3652398850704421, + "grad_norm": 0.0947265625, + "learning_rate": 0.0012897882648430204, + "loss": 0.0762, + "step": 42076 + }, + { + "epoch": 0.3652485655506463, + "grad_norm": 0.24609375, + "learning_rate": 0.0012897591715552996, + "loss": 0.0991, + "step": 42077 + }, + { + "epoch": 0.3652572460308504, + "grad_norm": 0.21875, + "learning_rate": 0.0012897300780600612, + "loss": 0.1113, + "step": 42078 + }, + { + "epoch": 0.3652659265110546, + "grad_norm": 0.40234375, + "learning_rate": 0.001289700984357337, + "loss": 0.1211, + "step": 42079 + }, + { + "epoch": 0.36527460699125874, + "grad_norm": 0.11279296875, + "learning_rate": 0.0012896718904471583, + "loss": 0.0918, + "step": 42080 + }, + { + "epoch": 0.36528328747146294, + "grad_norm": 0.291015625, + "learning_rate": 0.0012896427963295573, + "loss": 0.054, + "step": 42081 + }, + { + "epoch": 0.3652919679516671, + "grad_norm": 0.205078125, + "learning_rate": 0.001289613702004566, + "loss": 0.083, + "step": 42082 + }, + { + "epoch": 0.36530064843187127, + "grad_norm": 0.07958984375, + "learning_rate": 0.0012895846074722156, + "loss": 0.1079, + "step": 42083 + }, + { + "epoch": 0.3653093289120754, + "grad_norm": 0.1943359375, + "learning_rate": 0.0012895555127325387, + "loss": 0.0894, + "step": 42084 + }, + { + "epoch": 0.3653180093922796, + "grad_norm": 0.6875, + "learning_rate": 0.0012895264177855666, + "loss": 0.1211, + "step": 42085 + }, + { + "epoch": 0.36532668987248373, + "grad_norm": 0.314453125, + "learning_rate": 0.001289497322631331, + "loss": 0.1396, + "step": 42086 + }, + { + "epoch": 0.3653353703526879, + "grad_norm": 0.65625, + "learning_rate": 0.0012894682272698646, + "loss": 0.106, + "step": 42087 + }, + { + "epoch": 0.36534405083289206, + "grad_norm": 0.419921875, + "learning_rate": 0.0012894391317011982, + "loss": 0.0977, + "step": 42088 + }, + { + "epoch": 0.36535273131309626, + "grad_norm": 0.11767578125, + "learning_rate": 0.0012894100359253642, + "loss": 0.1025, + "step": 42089 + }, + { + "epoch": 0.3653614117933004, + "grad_norm": 0.12890625, + "learning_rate": 0.0012893809399423942, + "loss": 0.1201, + "step": 42090 + }, + { + "epoch": 0.3653700922735046, + "grad_norm": 0.154296875, + "learning_rate": 0.0012893518437523197, + "loss": 0.1426, + "step": 42091 + }, + { + "epoch": 0.3653787727537087, + "grad_norm": 0.50390625, + "learning_rate": 0.0012893227473551733, + "loss": 0.1245, + "step": 42092 + }, + { + "epoch": 0.3653874532339129, + "grad_norm": 0.2041015625, + "learning_rate": 0.001289293650750986, + "loss": 0.1108, + "step": 42093 + }, + { + "epoch": 0.36539613371411706, + "grad_norm": 0.17578125, + "learning_rate": 0.0012892645539397903, + "loss": 0.1348, + "step": 42094 + }, + { + "epoch": 0.36540481419432125, + "grad_norm": 1.1640625, + "learning_rate": 0.0012892354569216174, + "loss": 0.1885, + "step": 42095 + }, + { + "epoch": 0.3654134946745254, + "grad_norm": 0.240234375, + "learning_rate": 0.0012892063596965, + "loss": 0.0723, + "step": 42096 + }, + { + "epoch": 0.3654221751547296, + "grad_norm": 0.359375, + "learning_rate": 0.001289177262264469, + "loss": 0.0723, + "step": 42097 + }, + { + "epoch": 0.3654308556349337, + "grad_norm": 0.53515625, + "learning_rate": 0.0012891481646255568, + "loss": 0.1045, + "step": 42098 + }, + { + "epoch": 0.3654395361151379, + "grad_norm": 0.46875, + "learning_rate": 0.0012891190667797948, + "loss": 0.1064, + "step": 42099 + }, + { + "epoch": 0.36544821659534205, + "grad_norm": 0.1298828125, + "learning_rate": 0.001289089968727215, + "loss": 0.125, + "step": 42100 + }, + { + "epoch": 0.36545689707554624, + "grad_norm": 0.259765625, + "learning_rate": 0.0012890608704678496, + "loss": 0.1172, + "step": 42101 + }, + { + "epoch": 0.3654655775557504, + "grad_norm": 0.3203125, + "learning_rate": 0.0012890317720017294, + "loss": 0.1367, + "step": 42102 + }, + { + "epoch": 0.36547425803595457, + "grad_norm": 0.09521484375, + "learning_rate": 0.0012890026733288875, + "loss": 0.0586, + "step": 42103 + }, + { + "epoch": 0.3654829385161587, + "grad_norm": 0.134765625, + "learning_rate": 0.0012889735744493547, + "loss": 0.1074, + "step": 42104 + }, + { + "epoch": 0.3654916189963629, + "grad_norm": 0.15625, + "learning_rate": 0.0012889444753631633, + "loss": 0.0977, + "step": 42105 + }, + { + "epoch": 0.36550029947656704, + "grad_norm": 0.166015625, + "learning_rate": 0.0012889153760703455, + "loss": 0.0942, + "step": 42106 + }, + { + "epoch": 0.36550897995677123, + "grad_norm": 0.9140625, + "learning_rate": 0.0012888862765709323, + "loss": 0.1025, + "step": 42107 + }, + { + "epoch": 0.36551766043697537, + "grad_norm": 0.09130859375, + "learning_rate": 0.0012888571768649559, + "loss": 0.1021, + "step": 42108 + }, + { + "epoch": 0.36552634091717956, + "grad_norm": 0.34375, + "learning_rate": 0.0012888280769524484, + "loss": 0.1094, + "step": 42109 + }, + { + "epoch": 0.3655350213973837, + "grad_norm": 0.306640625, + "learning_rate": 0.0012887989768334412, + "loss": 0.0864, + "step": 42110 + }, + { + "epoch": 0.3655437018775879, + "grad_norm": 0.54296875, + "learning_rate": 0.0012887698765079662, + "loss": 0.1514, + "step": 42111 + }, + { + "epoch": 0.365552382357792, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012887407759760554, + "loss": 0.1309, + "step": 42112 + }, + { + "epoch": 0.3655610628379962, + "grad_norm": 0.2412109375, + "learning_rate": 0.0012887116752377404, + "loss": 0.0732, + "step": 42113 + }, + { + "epoch": 0.36556974331820036, + "grad_norm": 0.55859375, + "learning_rate": 0.0012886825742930532, + "loss": 0.1084, + "step": 42114 + }, + { + "epoch": 0.36557842379840455, + "grad_norm": 0.38671875, + "learning_rate": 0.0012886534731420256, + "loss": 0.0806, + "step": 42115 + }, + { + "epoch": 0.3655871042786087, + "grad_norm": 0.2431640625, + "learning_rate": 0.0012886243717846896, + "loss": 0.1084, + "step": 42116 + }, + { + "epoch": 0.3655957847588129, + "grad_norm": 0.84765625, + "learning_rate": 0.0012885952702210764, + "loss": 0.1621, + "step": 42117 + }, + { + "epoch": 0.365604465239017, + "grad_norm": 0.51953125, + "learning_rate": 0.0012885661684512184, + "loss": 0.1006, + "step": 42118 + }, + { + "epoch": 0.3656131457192212, + "grad_norm": 0.15234375, + "learning_rate": 0.0012885370664751474, + "loss": 0.0991, + "step": 42119 + }, + { + "epoch": 0.36562182619942535, + "grad_norm": 0.224609375, + "learning_rate": 0.0012885079642928953, + "loss": 0.0859, + "step": 42120 + }, + { + "epoch": 0.36563050667962954, + "grad_norm": 0.099609375, + "learning_rate": 0.0012884788619044933, + "loss": 0.0771, + "step": 42121 + }, + { + "epoch": 0.3656391871598337, + "grad_norm": 0.109375, + "learning_rate": 0.0012884497593099737, + "loss": 0.0874, + "step": 42122 + }, + { + "epoch": 0.36564786764003787, + "grad_norm": 0.2734375, + "learning_rate": 0.0012884206565093683, + "loss": 0.1152, + "step": 42123 + }, + { + "epoch": 0.365656548120242, + "grad_norm": 0.4765625, + "learning_rate": 0.0012883915535027092, + "loss": 0.1426, + "step": 42124 + }, + { + "epoch": 0.3656652286004462, + "grad_norm": 0.197265625, + "learning_rate": 0.0012883624502900278, + "loss": 0.0908, + "step": 42125 + }, + { + "epoch": 0.36567390908065034, + "grad_norm": 0.150390625, + "learning_rate": 0.0012883333468713559, + "loss": 0.1094, + "step": 42126 + }, + { + "epoch": 0.36568258956085453, + "grad_norm": 0.1220703125, + "learning_rate": 0.0012883042432467256, + "loss": 0.084, + "step": 42127 + }, + { + "epoch": 0.36569127004105867, + "grad_norm": 0.361328125, + "learning_rate": 0.0012882751394161687, + "loss": 0.1279, + "step": 42128 + }, + { + "epoch": 0.36569995052126286, + "grad_norm": 0.9140625, + "learning_rate": 0.0012882460353797168, + "loss": 0.1846, + "step": 42129 + }, + { + "epoch": 0.365708631001467, + "grad_norm": 0.1875, + "learning_rate": 0.0012882169311374018, + "loss": 0.0879, + "step": 42130 + }, + { + "epoch": 0.3657173114816712, + "grad_norm": 1.0625, + "learning_rate": 0.0012881878266892556, + "loss": 0.1299, + "step": 42131 + }, + { + "epoch": 0.36572599196187533, + "grad_norm": 0.1572265625, + "learning_rate": 0.00128815872203531, + "loss": 0.124, + "step": 42132 + }, + { + "epoch": 0.3657346724420795, + "grad_norm": 0.28515625, + "learning_rate": 0.001288129617175597, + "loss": 0.0884, + "step": 42133 + }, + { + "epoch": 0.36574335292228366, + "grad_norm": 0.193359375, + "learning_rate": 0.0012881005121101478, + "loss": 0.1191, + "step": 42134 + }, + { + "epoch": 0.36575203340248785, + "grad_norm": 0.2138671875, + "learning_rate": 0.0012880714068389954, + "loss": 0.1523, + "step": 42135 + }, + { + "epoch": 0.365760713882692, + "grad_norm": 0.310546875, + "learning_rate": 0.0012880423013621706, + "loss": 0.1201, + "step": 42136 + }, + { + "epoch": 0.3657693943628962, + "grad_norm": 0.373046875, + "learning_rate": 0.0012880131956797057, + "loss": 0.123, + "step": 42137 + }, + { + "epoch": 0.3657780748431003, + "grad_norm": 0.5703125, + "learning_rate": 0.0012879840897916319, + "loss": 0.0767, + "step": 42138 + }, + { + "epoch": 0.36578675532330446, + "grad_norm": 0.2431640625, + "learning_rate": 0.0012879549836979822, + "loss": 0.0996, + "step": 42139 + }, + { + "epoch": 0.36579543580350865, + "grad_norm": 0.435546875, + "learning_rate": 0.0012879258773987876, + "loss": 0.0757, + "step": 42140 + }, + { + "epoch": 0.3658041162837128, + "grad_norm": 0.498046875, + "learning_rate": 0.0012878967708940797, + "loss": 0.1113, + "step": 42141 + }, + { + "epoch": 0.365812796763917, + "grad_norm": 0.0869140625, + "learning_rate": 0.001287867664183891, + "loss": 0.1426, + "step": 42142 + }, + { + "epoch": 0.3658214772441211, + "grad_norm": 0.296875, + "learning_rate": 0.001287838557268253, + "loss": 0.0703, + "step": 42143 + }, + { + "epoch": 0.3658301577243253, + "grad_norm": 0.11376953125, + "learning_rate": 0.0012878094501471977, + "loss": 0.0757, + "step": 42144 + }, + { + "epoch": 0.36583883820452945, + "grad_norm": 0.267578125, + "learning_rate": 0.001287780342820757, + "loss": 0.0859, + "step": 42145 + }, + { + "epoch": 0.36584751868473364, + "grad_norm": 0.0927734375, + "learning_rate": 0.0012877512352889623, + "loss": 0.0928, + "step": 42146 + }, + { + "epoch": 0.3658561991649378, + "grad_norm": 0.19921875, + "learning_rate": 0.0012877221275518452, + "loss": 0.0649, + "step": 42147 + }, + { + "epoch": 0.36586487964514197, + "grad_norm": 0.240234375, + "learning_rate": 0.0012876930196094385, + "loss": 0.0889, + "step": 42148 + }, + { + "epoch": 0.3658735601253461, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012876639114617733, + "loss": 0.0869, + "step": 42149 + }, + { + "epoch": 0.3658822406055503, + "grad_norm": 0.2578125, + "learning_rate": 0.001287634803108882, + "loss": 0.166, + "step": 42150 + }, + { + "epoch": 0.36589092108575444, + "grad_norm": 0.40234375, + "learning_rate": 0.0012876056945507956, + "loss": 0.1108, + "step": 42151 + }, + { + "epoch": 0.36589960156595863, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012875765857875469, + "loss": 0.0952, + "step": 42152 + }, + { + "epoch": 0.36590828204616277, + "grad_norm": 0.345703125, + "learning_rate": 0.001287547476819167, + "loss": 0.0869, + "step": 42153 + }, + { + "epoch": 0.36591696252636696, + "grad_norm": 0.1884765625, + "learning_rate": 0.001287518367645688, + "loss": 0.0864, + "step": 42154 + }, + { + "epoch": 0.3659256430065711, + "grad_norm": 0.2265625, + "learning_rate": 0.0012874892582671422, + "loss": 0.1108, + "step": 42155 + }, + { + "epoch": 0.3659343234867753, + "grad_norm": 0.28125, + "learning_rate": 0.0012874601486835605, + "loss": 0.123, + "step": 42156 + }, + { + "epoch": 0.36594300396697943, + "grad_norm": 0.427734375, + "learning_rate": 0.0012874310388949753, + "loss": 0.1064, + "step": 42157 + }, + { + "epoch": 0.3659516844471836, + "grad_norm": 0.3203125, + "learning_rate": 0.0012874019289014183, + "loss": 0.0703, + "step": 42158 + }, + { + "epoch": 0.36596036492738776, + "grad_norm": 0.30078125, + "learning_rate": 0.0012873728187029216, + "loss": 0.0884, + "step": 42159 + }, + { + "epoch": 0.36596904540759195, + "grad_norm": 0.18359375, + "learning_rate": 0.0012873437082995163, + "loss": 0.0898, + "step": 42160 + }, + { + "epoch": 0.3659777258877961, + "grad_norm": 0.2041015625, + "learning_rate": 0.001287314597691235, + "loss": 0.1147, + "step": 42161 + }, + { + "epoch": 0.3659864063680003, + "grad_norm": 0.1142578125, + "learning_rate": 0.0012872854868781094, + "loss": 0.106, + "step": 42162 + }, + { + "epoch": 0.3659950868482044, + "grad_norm": 0.091796875, + "learning_rate": 0.0012872563758601712, + "loss": 0.0796, + "step": 42163 + }, + { + "epoch": 0.3660037673284086, + "grad_norm": 0.220703125, + "learning_rate": 0.0012872272646374519, + "loss": 0.0728, + "step": 42164 + }, + { + "epoch": 0.36601244780861275, + "grad_norm": 0.62109375, + "learning_rate": 0.0012871981532099842, + "loss": 0.3008, + "step": 42165 + }, + { + "epoch": 0.36602112828881694, + "grad_norm": 0.11376953125, + "learning_rate": 0.001287169041577799, + "loss": 0.1025, + "step": 42166 + }, + { + "epoch": 0.3660298087690211, + "grad_norm": 0.66015625, + "learning_rate": 0.0012871399297409289, + "loss": 0.2266, + "step": 42167 + }, + { + "epoch": 0.3660384892492253, + "grad_norm": 0.271484375, + "learning_rate": 0.001287110817699405, + "loss": 0.0859, + "step": 42168 + }, + { + "epoch": 0.3660471697294294, + "grad_norm": 0.18359375, + "learning_rate": 0.0012870817054532594, + "loss": 0.1074, + "step": 42169 + }, + { + "epoch": 0.3660558502096336, + "grad_norm": 0.12353515625, + "learning_rate": 0.0012870525930025248, + "loss": 0.1309, + "step": 42170 + }, + { + "epoch": 0.36606453068983774, + "grad_norm": 0.1455078125, + "learning_rate": 0.0012870234803472316, + "loss": 0.105, + "step": 42171 + }, + { + "epoch": 0.36607321117004193, + "grad_norm": 0.61328125, + "learning_rate": 0.0012869943674874125, + "loss": 0.0918, + "step": 42172 + }, + { + "epoch": 0.36608189165024607, + "grad_norm": 0.1103515625, + "learning_rate": 0.0012869652544230993, + "loss": 0.0957, + "step": 42173 + }, + { + "epoch": 0.36609057213045026, + "grad_norm": 0.09765625, + "learning_rate": 0.0012869361411543237, + "loss": 0.1367, + "step": 42174 + }, + { + "epoch": 0.3660992526106544, + "grad_norm": 0.466796875, + "learning_rate": 0.0012869070276811175, + "loss": 0.0752, + "step": 42175 + }, + { + "epoch": 0.3661079330908586, + "grad_norm": 0.08984375, + "learning_rate": 0.0012868779140035125, + "loss": 0.0952, + "step": 42176 + }, + { + "epoch": 0.36611661357106273, + "grad_norm": 0.11181640625, + "learning_rate": 0.0012868488001215406, + "loss": 0.0898, + "step": 42177 + }, + { + "epoch": 0.3661252940512669, + "grad_norm": 0.40625, + "learning_rate": 0.0012868196860352339, + "loss": 0.1445, + "step": 42178 + }, + { + "epoch": 0.36613397453147106, + "grad_norm": 0.08642578125, + "learning_rate": 0.001286790571744624, + "loss": 0.1172, + "step": 42179 + }, + { + "epoch": 0.36614265501167526, + "grad_norm": 0.150390625, + "learning_rate": 0.0012867614572497424, + "loss": 0.0908, + "step": 42180 + }, + { + "epoch": 0.3661513354918794, + "grad_norm": 0.357421875, + "learning_rate": 0.0012867323425506214, + "loss": 0.1011, + "step": 42181 + }, + { + "epoch": 0.3661600159720836, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012867032276472927, + "loss": 0.1328, + "step": 42182 + }, + { + "epoch": 0.3661686964522877, + "grad_norm": 0.212890625, + "learning_rate": 0.0012866741125397885, + "loss": 0.1118, + "step": 42183 + }, + { + "epoch": 0.3661773769324919, + "grad_norm": 0.125, + "learning_rate": 0.0012866449972281399, + "loss": 0.0845, + "step": 42184 + }, + { + "epoch": 0.36618605741269605, + "grad_norm": 0.189453125, + "learning_rate": 0.0012866158817123794, + "loss": 0.1074, + "step": 42185 + }, + { + "epoch": 0.36619473789290025, + "grad_norm": 0.337890625, + "learning_rate": 0.0012865867659925382, + "loss": 0.1055, + "step": 42186 + }, + { + "epoch": 0.3662034183731044, + "grad_norm": 0.173828125, + "learning_rate": 0.0012865576500686489, + "loss": 0.0889, + "step": 42187 + }, + { + "epoch": 0.3662120988533086, + "grad_norm": 0.1943359375, + "learning_rate": 0.001286528533940743, + "loss": 0.083, + "step": 42188 + }, + { + "epoch": 0.3662207793335127, + "grad_norm": 0.37109375, + "learning_rate": 0.0012864994176088518, + "loss": 0.1045, + "step": 42189 + }, + { + "epoch": 0.3662294598137169, + "grad_norm": 0.38671875, + "learning_rate": 0.001286470301073008, + "loss": 0.0884, + "step": 42190 + }, + { + "epoch": 0.36623814029392104, + "grad_norm": 0.462890625, + "learning_rate": 0.001286441184333243, + "loss": 0.084, + "step": 42191 + }, + { + "epoch": 0.36624682077412524, + "grad_norm": 0.37109375, + "learning_rate": 0.001286412067389589, + "loss": 0.0752, + "step": 42192 + }, + { + "epoch": 0.3662555012543294, + "grad_norm": 0.220703125, + "learning_rate": 0.0012863829502420773, + "loss": 0.1328, + "step": 42193 + }, + { + "epoch": 0.36626418173453357, + "grad_norm": 0.11572265625, + "learning_rate": 0.00128635383289074, + "loss": 0.1406, + "step": 42194 + }, + { + "epoch": 0.3662728622147377, + "grad_norm": 0.111328125, + "learning_rate": 0.0012863247153356088, + "loss": 0.125, + "step": 42195 + }, + { + "epoch": 0.3662815426949419, + "grad_norm": 0.2431640625, + "learning_rate": 0.0012862955975767158, + "loss": 0.1172, + "step": 42196 + }, + { + "epoch": 0.36629022317514603, + "grad_norm": 0.609375, + "learning_rate": 0.0012862664796140931, + "loss": 0.1138, + "step": 42197 + }, + { + "epoch": 0.3662989036553502, + "grad_norm": 0.447265625, + "learning_rate": 0.0012862373614477716, + "loss": 0.1143, + "step": 42198 + }, + { + "epoch": 0.36630758413555436, + "grad_norm": 0.2099609375, + "learning_rate": 0.001286208243077784, + "loss": 0.1016, + "step": 42199 + }, + { + "epoch": 0.36631626461575856, + "grad_norm": 0.09326171875, + "learning_rate": 0.001286179124504162, + "loss": 0.0811, + "step": 42200 + }, + { + "epoch": 0.3663249450959627, + "grad_norm": 0.193359375, + "learning_rate": 0.0012861500057269372, + "loss": 0.1143, + "step": 42201 + }, + { + "epoch": 0.3663336255761669, + "grad_norm": 0.1875, + "learning_rate": 0.0012861208867461413, + "loss": 0.1064, + "step": 42202 + }, + { + "epoch": 0.366342306056371, + "grad_norm": 0.369140625, + "learning_rate": 0.0012860917675618067, + "loss": 0.1113, + "step": 42203 + }, + { + "epoch": 0.3663509865365752, + "grad_norm": 0.169921875, + "learning_rate": 0.0012860626481739647, + "loss": 0.1245, + "step": 42204 + }, + { + "epoch": 0.36635966701677936, + "grad_norm": 0.103515625, + "learning_rate": 0.001286033528582648, + "loss": 0.0801, + "step": 42205 + }, + { + "epoch": 0.36636834749698355, + "grad_norm": 0.2353515625, + "learning_rate": 0.0012860044087878873, + "loss": 0.125, + "step": 42206 + }, + { + "epoch": 0.3663770279771877, + "grad_norm": 0.419921875, + "learning_rate": 0.001285975288789715, + "loss": 0.083, + "step": 42207 + }, + { + "epoch": 0.3663857084573919, + "grad_norm": 0.08935546875, + "learning_rate": 0.0012859461685881628, + "loss": 0.0957, + "step": 42208 + }, + { + "epoch": 0.366394388937596, + "grad_norm": 0.37109375, + "learning_rate": 0.0012859170481832626, + "loss": 0.123, + "step": 42209 + }, + { + "epoch": 0.3664030694178002, + "grad_norm": 0.1416015625, + "learning_rate": 0.0012858879275750468, + "loss": 0.0688, + "step": 42210 + }, + { + "epoch": 0.36641174989800435, + "grad_norm": 0.5078125, + "learning_rate": 0.0012858588067635466, + "loss": 0.1201, + "step": 42211 + }, + { + "epoch": 0.36642043037820854, + "grad_norm": 0.119140625, + "learning_rate": 0.0012858296857487935, + "loss": 0.0879, + "step": 42212 + }, + { + "epoch": 0.3664291108584127, + "grad_norm": 0.32421875, + "learning_rate": 0.0012858005645308204, + "loss": 0.1235, + "step": 42213 + }, + { + "epoch": 0.36643779133861687, + "grad_norm": 0.20703125, + "learning_rate": 0.0012857714431096586, + "loss": 0.1035, + "step": 42214 + }, + { + "epoch": 0.366446471818821, + "grad_norm": 0.30078125, + "learning_rate": 0.00128574232148534, + "loss": 0.125, + "step": 42215 + }, + { + "epoch": 0.3664551522990252, + "grad_norm": 0.123046875, + "learning_rate": 0.0012857131996578959, + "loss": 0.1162, + "step": 42216 + }, + { + "epoch": 0.36646383277922934, + "grad_norm": 0.2890625, + "learning_rate": 0.0012856840776273588, + "loss": 0.1348, + "step": 42217 + }, + { + "epoch": 0.36647251325943353, + "grad_norm": 0.119140625, + "learning_rate": 0.0012856549553937604, + "loss": 0.1318, + "step": 42218 + }, + { + "epoch": 0.36648119373963767, + "grad_norm": 0.2138671875, + "learning_rate": 0.0012856258329571324, + "loss": 0.1045, + "step": 42219 + }, + { + "epoch": 0.36648987421984186, + "grad_norm": 0.1220703125, + "learning_rate": 0.001285596710317507, + "loss": 0.1211, + "step": 42220 + }, + { + "epoch": 0.366498554700046, + "grad_norm": 0.1552734375, + "learning_rate": 0.001285567587474916, + "loss": 0.0771, + "step": 42221 + }, + { + "epoch": 0.3665072351802502, + "grad_norm": 0.07421875, + "learning_rate": 0.001285538464429391, + "loss": 0.0913, + "step": 42222 + }, + { + "epoch": 0.36651591566045433, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012855093411809637, + "loss": 0.1465, + "step": 42223 + }, + { + "epoch": 0.3665245961406585, + "grad_norm": 0.67578125, + "learning_rate": 0.0012854802177296664, + "loss": 0.104, + "step": 42224 + }, + { + "epoch": 0.36653327662086266, + "grad_norm": 1.078125, + "learning_rate": 0.0012854510940755303, + "loss": 0.1904, + "step": 42225 + }, + { + "epoch": 0.36654195710106685, + "grad_norm": 0.259765625, + "learning_rate": 0.0012854219702185881, + "loss": 0.0996, + "step": 42226 + }, + { + "epoch": 0.366550637581271, + "grad_norm": 0.55078125, + "learning_rate": 0.0012853928461588713, + "loss": 0.1094, + "step": 42227 + }, + { + "epoch": 0.3665593180614752, + "grad_norm": 0.2236328125, + "learning_rate": 0.0012853637218964112, + "loss": 0.0791, + "step": 42228 + }, + { + "epoch": 0.3665679985416793, + "grad_norm": 0.359375, + "learning_rate": 0.0012853345974312403, + "loss": 0.0874, + "step": 42229 + }, + { + "epoch": 0.3665766790218835, + "grad_norm": 0.244140625, + "learning_rate": 0.0012853054727633903, + "loss": 0.166, + "step": 42230 + }, + { + "epoch": 0.36658535950208765, + "grad_norm": 0.357421875, + "learning_rate": 0.0012852763478928932, + "loss": 0.1064, + "step": 42231 + }, + { + "epoch": 0.36659403998229184, + "grad_norm": 0.765625, + "learning_rate": 0.0012852472228197806, + "loss": 0.5273, + "step": 42232 + }, + { + "epoch": 0.366602720462496, + "grad_norm": 0.2109375, + "learning_rate": 0.0012852180975440842, + "loss": 0.1108, + "step": 42233 + }, + { + "epoch": 0.36661140094270017, + "grad_norm": 0.15234375, + "learning_rate": 0.0012851889720658365, + "loss": 0.1055, + "step": 42234 + }, + { + "epoch": 0.3666200814229043, + "grad_norm": 0.453125, + "learning_rate": 0.0012851598463850687, + "loss": 0.1006, + "step": 42235 + }, + { + "epoch": 0.3666287619031085, + "grad_norm": 0.251953125, + "learning_rate": 0.0012851307205018127, + "loss": 0.1035, + "step": 42236 + }, + { + "epoch": 0.36663744238331264, + "grad_norm": 0.091796875, + "learning_rate": 0.0012851015944161006, + "loss": 0.1299, + "step": 42237 + }, + { + "epoch": 0.36664612286351683, + "grad_norm": 0.2490234375, + "learning_rate": 0.0012850724681279639, + "loss": 0.1416, + "step": 42238 + }, + { + "epoch": 0.36665480334372097, + "grad_norm": 0.11572265625, + "learning_rate": 0.0012850433416374352, + "loss": 0.1318, + "step": 42239 + }, + { + "epoch": 0.36666348382392516, + "grad_norm": 0.095703125, + "learning_rate": 0.0012850142149445458, + "loss": 0.084, + "step": 42240 + }, + { + "epoch": 0.3666721643041293, + "grad_norm": 0.10791015625, + "learning_rate": 0.0012849850880493276, + "loss": 0.1084, + "step": 42241 + }, + { + "epoch": 0.3666808447843335, + "grad_norm": 0.6328125, + "learning_rate": 0.0012849559609518124, + "loss": 0.1216, + "step": 42242 + }, + { + "epoch": 0.36668952526453763, + "grad_norm": 0.431640625, + "learning_rate": 0.0012849268336520324, + "loss": 0.1191, + "step": 42243 + }, + { + "epoch": 0.3666982057447418, + "grad_norm": 0.9921875, + "learning_rate": 0.0012848977061500188, + "loss": 0.0879, + "step": 42244 + }, + { + "epoch": 0.36670688622494596, + "grad_norm": 0.140625, + "learning_rate": 0.001284868578445804, + "loss": 0.1172, + "step": 42245 + }, + { + "epoch": 0.36671556670515015, + "grad_norm": 0.224609375, + "learning_rate": 0.0012848394505394199, + "loss": 0.1084, + "step": 42246 + }, + { + "epoch": 0.3667242471853543, + "grad_norm": 0.08837890625, + "learning_rate": 0.0012848103224308981, + "loss": 0.1118, + "step": 42247 + }, + { + "epoch": 0.3667329276655585, + "grad_norm": 0.169921875, + "learning_rate": 0.0012847811941202702, + "loss": 0.1104, + "step": 42248 + }, + { + "epoch": 0.3667416081457626, + "grad_norm": 0.09765625, + "learning_rate": 0.0012847520656075683, + "loss": 0.1079, + "step": 42249 + }, + { + "epoch": 0.3667502886259668, + "grad_norm": 0.27734375, + "learning_rate": 0.001284722936892825, + "loss": 0.2832, + "step": 42250 + }, + { + "epoch": 0.36675896910617095, + "grad_norm": 0.2421875, + "learning_rate": 0.0012846938079760708, + "loss": 0.085, + "step": 42251 + }, + { + "epoch": 0.36676764958637514, + "grad_norm": 0.251953125, + "learning_rate": 0.0012846646788573386, + "loss": 0.0957, + "step": 42252 + }, + { + "epoch": 0.3667763300665793, + "grad_norm": 0.15234375, + "learning_rate": 0.0012846355495366598, + "loss": 0.1211, + "step": 42253 + }, + { + "epoch": 0.3667850105467835, + "grad_norm": 0.12890625, + "learning_rate": 0.0012846064200140663, + "loss": 0.1094, + "step": 42254 + }, + { + "epoch": 0.3667936910269876, + "grad_norm": 0.2890625, + "learning_rate": 0.00128457729028959, + "loss": 0.1143, + "step": 42255 + }, + { + "epoch": 0.3668023715071918, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012845481603632628, + "loss": 0.1099, + "step": 42256 + }, + { + "epoch": 0.36681105198739594, + "grad_norm": 0.3984375, + "learning_rate": 0.001284519030235116, + "loss": 0.0801, + "step": 42257 + }, + { + "epoch": 0.36681973246760013, + "grad_norm": 0.10888671875, + "learning_rate": 0.001284489899905183, + "loss": 0.0996, + "step": 42258 + }, + { + "epoch": 0.36682841294780427, + "grad_norm": 0.251953125, + "learning_rate": 0.0012844607693734936, + "loss": 0.1133, + "step": 42259 + }, + { + "epoch": 0.36683709342800846, + "grad_norm": 0.126953125, + "learning_rate": 0.001284431638640081, + "loss": 0.0742, + "step": 42260 + }, + { + "epoch": 0.3668457739082126, + "grad_norm": 0.56640625, + "learning_rate": 0.001284402507704977, + "loss": 0.082, + "step": 42261 + }, + { + "epoch": 0.36685445438841674, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012843733765682131, + "loss": 0.1143, + "step": 42262 + }, + { + "epoch": 0.36686313486862093, + "grad_norm": 0.16796875, + "learning_rate": 0.0012843442452298212, + "loss": 0.127, + "step": 42263 + }, + { + "epoch": 0.36687181534882507, + "grad_norm": 0.2294921875, + "learning_rate": 0.001284315113689833, + "loss": 0.085, + "step": 42264 + }, + { + "epoch": 0.36688049582902926, + "grad_norm": 0.20703125, + "learning_rate": 0.001284285981948281, + "loss": 0.1108, + "step": 42265 + }, + { + "epoch": 0.3668891763092334, + "grad_norm": 0.1162109375, + "learning_rate": 0.0012842568500051961, + "loss": 0.1143, + "step": 42266 + }, + { + "epoch": 0.3668978567894376, + "grad_norm": 0.220703125, + "learning_rate": 0.001284227717860611, + "loss": 0.1001, + "step": 42267 + }, + { + "epoch": 0.36690653726964173, + "grad_norm": 0.1435546875, + "learning_rate": 0.001284198585514557, + "loss": 0.0854, + "step": 42268 + }, + { + "epoch": 0.3669152177498459, + "grad_norm": 0.0712890625, + "learning_rate": 0.0012841694529670664, + "loss": 0.0771, + "step": 42269 + }, + { + "epoch": 0.36692389823005006, + "grad_norm": 0.33203125, + "learning_rate": 0.001284140320218171, + "loss": 0.0928, + "step": 42270 + }, + { + "epoch": 0.36693257871025425, + "grad_norm": 0.3984375, + "learning_rate": 0.0012841111872679022, + "loss": 0.1147, + "step": 42271 + }, + { + "epoch": 0.3669412591904584, + "grad_norm": 0.66796875, + "learning_rate": 0.0012840820541162922, + "loss": 0.0996, + "step": 42272 + }, + { + "epoch": 0.3669499396706626, + "grad_norm": 0.83203125, + "learning_rate": 0.001284052920763373, + "loss": 0.0972, + "step": 42273 + }, + { + "epoch": 0.3669586201508667, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012840237872091761, + "loss": 0.0962, + "step": 42274 + }, + { + "epoch": 0.3669673006310709, + "grad_norm": 0.55859375, + "learning_rate": 0.001283994653453734, + "loss": 0.0913, + "step": 42275 + }, + { + "epoch": 0.36697598111127505, + "grad_norm": 0.48828125, + "learning_rate": 0.0012839655194970777, + "loss": 0.1582, + "step": 42276 + }, + { + "epoch": 0.36698466159147924, + "grad_norm": 0.3203125, + "learning_rate": 0.0012839363853392393, + "loss": 0.106, + "step": 42277 + }, + { + "epoch": 0.3669933420716834, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012839072509802513, + "loss": 0.0913, + "step": 42278 + }, + { + "epoch": 0.3670020225518876, + "grad_norm": 0.3125, + "learning_rate": 0.0012838781164201448, + "loss": 0.1108, + "step": 42279 + }, + { + "epoch": 0.3670107030320917, + "grad_norm": 0.6015625, + "learning_rate": 0.001283848981658952, + "loss": 0.1328, + "step": 42280 + }, + { + "epoch": 0.3670193835122959, + "grad_norm": 0.2490234375, + "learning_rate": 0.0012838198466967046, + "loss": 0.0845, + "step": 42281 + }, + { + "epoch": 0.36702806399250004, + "grad_norm": 0.734375, + "learning_rate": 0.0012837907115334348, + "loss": 0.1279, + "step": 42282 + }, + { + "epoch": 0.36703674447270423, + "grad_norm": 0.65234375, + "learning_rate": 0.001283761576169174, + "loss": 0.0977, + "step": 42283 + }, + { + "epoch": 0.3670454249529084, + "grad_norm": 0.373046875, + "learning_rate": 0.0012837324406039547, + "loss": 0.103, + "step": 42284 + }, + { + "epoch": 0.36705410543311257, + "grad_norm": 0.54296875, + "learning_rate": 0.001283703304837808, + "loss": 0.1138, + "step": 42285 + }, + { + "epoch": 0.3670627859133167, + "grad_norm": 0.2109375, + "learning_rate": 0.0012836741688707662, + "loss": 0.104, + "step": 42286 + }, + { + "epoch": 0.3670714663935209, + "grad_norm": 0.2578125, + "learning_rate": 0.001283645032702861, + "loss": 0.1553, + "step": 42287 + }, + { + "epoch": 0.36708014687372503, + "grad_norm": 0.1591796875, + "learning_rate": 0.0012836158963341249, + "loss": 0.1226, + "step": 42288 + }, + { + "epoch": 0.3670888273539292, + "grad_norm": 0.62109375, + "learning_rate": 0.0012835867597645887, + "loss": 0.0957, + "step": 42289 + }, + { + "epoch": 0.36709750783413336, + "grad_norm": 0.1591796875, + "learning_rate": 0.0012835576229942847, + "loss": 0.0986, + "step": 42290 + }, + { + "epoch": 0.36710618831433756, + "grad_norm": 0.408203125, + "learning_rate": 0.0012835284860232452, + "loss": 0.1025, + "step": 42291 + }, + { + "epoch": 0.3671148687945417, + "grad_norm": 0.1572265625, + "learning_rate": 0.0012834993488515017, + "loss": 0.1289, + "step": 42292 + }, + { + "epoch": 0.3671235492747459, + "grad_norm": 0.125, + "learning_rate": 0.0012834702114790856, + "loss": 0.127, + "step": 42293 + }, + { + "epoch": 0.36713222975495, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012834410739060297, + "loss": 0.0796, + "step": 42294 + }, + { + "epoch": 0.3671409102351542, + "grad_norm": 0.25, + "learning_rate": 0.0012834119361323651, + "loss": 0.1064, + "step": 42295 + }, + { + "epoch": 0.36714959071535835, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012833827981581238, + "loss": 0.125, + "step": 42296 + }, + { + "epoch": 0.36715827119556255, + "grad_norm": 0.150390625, + "learning_rate": 0.0012833536599833385, + "loss": 0.0996, + "step": 42297 + }, + { + "epoch": 0.3671669516757667, + "grad_norm": 0.3203125, + "learning_rate": 0.0012833245216080397, + "loss": 0.1021, + "step": 42298 + }, + { + "epoch": 0.3671756321559709, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012832953830322605, + "loss": 0.0874, + "step": 42299 + }, + { + "epoch": 0.367184312636175, + "grad_norm": 0.228515625, + "learning_rate": 0.001283266244256032, + "loss": 0.0771, + "step": 42300 + }, + { + "epoch": 0.3671929931163792, + "grad_norm": 1.109375, + "learning_rate": 0.0012832371052793863, + "loss": 0.1064, + "step": 42301 + }, + { + "epoch": 0.36720167359658334, + "grad_norm": 0.75390625, + "learning_rate": 0.0012832079661023552, + "loss": 0.0903, + "step": 42302 + }, + { + "epoch": 0.36721035407678754, + "grad_norm": 0.7109375, + "learning_rate": 0.0012831788267249707, + "loss": 0.1055, + "step": 42303 + }, + { + "epoch": 0.3672190345569917, + "grad_norm": 0.1640625, + "learning_rate": 0.0012831496871472646, + "loss": 0.1006, + "step": 42304 + }, + { + "epoch": 0.36722771503719587, + "grad_norm": 0.123046875, + "learning_rate": 0.0012831205473692686, + "loss": 0.127, + "step": 42305 + }, + { + "epoch": 0.3672363955174, + "grad_norm": 0.326171875, + "learning_rate": 0.0012830914073910147, + "loss": 0.1221, + "step": 42306 + }, + { + "epoch": 0.3672450759976042, + "grad_norm": 0.1630859375, + "learning_rate": 0.0012830622672125346, + "loss": 0.1064, + "step": 42307 + }, + { + "epoch": 0.36725375647780834, + "grad_norm": 0.1201171875, + "learning_rate": 0.0012830331268338612, + "loss": 0.1523, + "step": 42308 + }, + { + "epoch": 0.36726243695801253, + "grad_norm": 0.1650390625, + "learning_rate": 0.001283003986255025, + "loss": 0.0986, + "step": 42309 + }, + { + "epoch": 0.36727111743821667, + "grad_norm": 0.09521484375, + "learning_rate": 0.0012829748454760582, + "loss": 0.1021, + "step": 42310 + }, + { + "epoch": 0.36727979791842086, + "grad_norm": 0.138671875, + "learning_rate": 0.001282945704496993, + "loss": 0.0898, + "step": 42311 + }, + { + "epoch": 0.367288478398625, + "grad_norm": 0.2314453125, + "learning_rate": 0.0012829165633178612, + "loss": 0.1299, + "step": 42312 + }, + { + "epoch": 0.3672971588788292, + "grad_norm": 0.76171875, + "learning_rate": 0.0012828874219386948, + "loss": 0.1152, + "step": 42313 + }, + { + "epoch": 0.3673058393590333, + "grad_norm": 0.62890625, + "learning_rate": 0.001282858280359525, + "loss": 0.103, + "step": 42314 + }, + { + "epoch": 0.3673145198392375, + "grad_norm": 0.197265625, + "learning_rate": 0.0012828291385803842, + "loss": 0.1162, + "step": 42315 + }, + { + "epoch": 0.36732320031944166, + "grad_norm": 0.388671875, + "learning_rate": 0.0012827999966013044, + "loss": 0.0933, + "step": 42316 + }, + { + "epoch": 0.36733188079964585, + "grad_norm": 0.412109375, + "learning_rate": 0.0012827708544223173, + "loss": 0.1221, + "step": 42317 + }, + { + "epoch": 0.36734056127985, + "grad_norm": 0.0947265625, + "learning_rate": 0.0012827417120434546, + "loss": 0.1201, + "step": 42318 + }, + { + "epoch": 0.3673492417600542, + "grad_norm": 0.15625, + "learning_rate": 0.0012827125694647487, + "loss": 0.167, + "step": 42319 + }, + { + "epoch": 0.3673579222402583, + "grad_norm": 0.82421875, + "learning_rate": 0.0012826834266862304, + "loss": 0.1855, + "step": 42320 + }, + { + "epoch": 0.3673666027204625, + "grad_norm": 0.05419921875, + "learning_rate": 0.0012826542837079326, + "loss": 0.0874, + "step": 42321 + }, + { + "epoch": 0.36737528320066665, + "grad_norm": 0.51171875, + "learning_rate": 0.001282625140529887, + "loss": 0.0996, + "step": 42322 + }, + { + "epoch": 0.36738396368087084, + "grad_norm": 0.62109375, + "learning_rate": 0.0012825959971521248, + "loss": 0.1279, + "step": 42323 + }, + { + "epoch": 0.367392644161075, + "grad_norm": 0.1728515625, + "learning_rate": 0.0012825668535746788, + "loss": 0.1152, + "step": 42324 + }, + { + "epoch": 0.36740132464127917, + "grad_norm": 0.3046875, + "learning_rate": 0.0012825377097975801, + "loss": 0.103, + "step": 42325 + }, + { + "epoch": 0.3674100051214833, + "grad_norm": 0.291015625, + "learning_rate": 0.0012825085658208612, + "loss": 0.1001, + "step": 42326 + }, + { + "epoch": 0.3674186856016875, + "grad_norm": 0.470703125, + "learning_rate": 0.0012824794216445537, + "loss": 0.2178, + "step": 42327 + }, + { + "epoch": 0.36742736608189164, + "grad_norm": 0.29296875, + "learning_rate": 0.0012824502772686892, + "loss": 0.1025, + "step": 42328 + }, + { + "epoch": 0.36743604656209583, + "grad_norm": 0.08447265625, + "learning_rate": 0.0012824211326932999, + "loss": 0.0938, + "step": 42329 + }, + { + "epoch": 0.36744472704229997, + "grad_norm": 1.03125, + "learning_rate": 0.0012823919879184178, + "loss": 0.0884, + "step": 42330 + }, + { + "epoch": 0.36745340752250416, + "grad_norm": 0.173828125, + "learning_rate": 0.0012823628429440743, + "loss": 0.1172, + "step": 42331 + }, + { + "epoch": 0.3674620880027083, + "grad_norm": 0.412109375, + "learning_rate": 0.0012823336977703017, + "loss": 0.1123, + "step": 42332 + }, + { + "epoch": 0.3674707684829125, + "grad_norm": 0.1923828125, + "learning_rate": 0.0012823045523971317, + "loss": 0.0967, + "step": 42333 + }, + { + "epoch": 0.36747944896311663, + "grad_norm": 0.357421875, + "learning_rate": 0.001282275406824596, + "loss": 0.0908, + "step": 42334 + }, + { + "epoch": 0.3674881294433208, + "grad_norm": 0.443359375, + "learning_rate": 0.0012822462610527269, + "loss": 0.082, + "step": 42335 + }, + { + "epoch": 0.36749680992352496, + "grad_norm": 0.32421875, + "learning_rate": 0.0012822171150815558, + "loss": 0.1094, + "step": 42336 + }, + { + "epoch": 0.36750549040372915, + "grad_norm": 0.2392578125, + "learning_rate": 0.001282187968911115, + "loss": 0.0957, + "step": 42337 + }, + { + "epoch": 0.3675141708839333, + "grad_norm": 0.18359375, + "learning_rate": 0.0012821588225414358, + "loss": 0.123, + "step": 42338 + }, + { + "epoch": 0.3675228513641375, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012821296759725508, + "loss": 0.0767, + "step": 42339 + }, + { + "epoch": 0.3675315318443416, + "grad_norm": 0.0849609375, + "learning_rate": 0.0012821005292044916, + "loss": 0.1089, + "step": 42340 + }, + { + "epoch": 0.3675402123245458, + "grad_norm": 0.48828125, + "learning_rate": 0.0012820713822372899, + "loss": 0.1133, + "step": 42341 + }, + { + "epoch": 0.36754889280474995, + "grad_norm": 0.87890625, + "learning_rate": 0.0012820422350709776, + "loss": 0.0879, + "step": 42342 + }, + { + "epoch": 0.36755757328495414, + "grad_norm": 0.28515625, + "learning_rate": 0.0012820130877055867, + "loss": 0.083, + "step": 42343 + }, + { + "epoch": 0.3675662537651583, + "grad_norm": 0.107421875, + "learning_rate": 0.0012819839401411487, + "loss": 0.1104, + "step": 42344 + }, + { + "epoch": 0.3675749342453625, + "grad_norm": 0.48046875, + "learning_rate": 0.001281954792377696, + "loss": 0.0957, + "step": 42345 + }, + { + "epoch": 0.3675836147255666, + "grad_norm": 0.640625, + "learning_rate": 0.0012819256444152605, + "loss": 0.1133, + "step": 42346 + }, + { + "epoch": 0.3675922952057708, + "grad_norm": 0.48828125, + "learning_rate": 0.0012818964962538737, + "loss": 0.1064, + "step": 42347 + }, + { + "epoch": 0.36760097568597494, + "grad_norm": 0.455078125, + "learning_rate": 0.0012818673478935678, + "loss": 0.1309, + "step": 42348 + }, + { + "epoch": 0.36760965616617913, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012818381993343743, + "loss": 0.0918, + "step": 42349 + }, + { + "epoch": 0.36761833664638327, + "grad_norm": 0.1220703125, + "learning_rate": 0.001281809050576325, + "loss": 0.0894, + "step": 42350 + }, + { + "epoch": 0.36762701712658746, + "grad_norm": 0.125, + "learning_rate": 0.0012817799016194525, + "loss": 0.1621, + "step": 42351 + }, + { + "epoch": 0.3676356976067916, + "grad_norm": 0.55078125, + "learning_rate": 0.0012817507524637884, + "loss": 0.0928, + "step": 42352 + }, + { + "epoch": 0.3676443780869958, + "grad_norm": 0.859375, + "learning_rate": 0.0012817216031093639, + "loss": 0.1201, + "step": 42353 + }, + { + "epoch": 0.36765305856719993, + "grad_norm": 0.32421875, + "learning_rate": 0.0012816924535562115, + "loss": 0.083, + "step": 42354 + }, + { + "epoch": 0.3676617390474041, + "grad_norm": 0.119140625, + "learning_rate": 0.0012816633038043631, + "loss": 0.0781, + "step": 42355 + }, + { + "epoch": 0.36767041952760826, + "grad_norm": 0.33984375, + "learning_rate": 0.0012816341538538503, + "loss": 0.0742, + "step": 42356 + }, + { + "epoch": 0.36767910000781245, + "grad_norm": 0.103515625, + "learning_rate": 0.0012816050037047053, + "loss": 0.0806, + "step": 42357 + }, + { + "epoch": 0.3676877804880166, + "grad_norm": 0.52734375, + "learning_rate": 0.0012815758533569596, + "loss": 0.1011, + "step": 42358 + }, + { + "epoch": 0.3676964609682208, + "grad_norm": 0.189453125, + "learning_rate": 0.0012815467028106453, + "loss": 0.0688, + "step": 42359 + }, + { + "epoch": 0.3677051414484249, + "grad_norm": 0.2236328125, + "learning_rate": 0.0012815175520657946, + "loss": 0.0737, + "step": 42360 + }, + { + "epoch": 0.3677138219286291, + "grad_norm": 0.2294921875, + "learning_rate": 0.0012814884011224388, + "loss": 0.1631, + "step": 42361 + }, + { + "epoch": 0.36772250240883325, + "grad_norm": 0.34375, + "learning_rate": 0.0012814592499806098, + "loss": 0.0967, + "step": 42362 + }, + { + "epoch": 0.36773118288903744, + "grad_norm": 0.298828125, + "learning_rate": 0.0012814300986403395, + "loss": 0.0952, + "step": 42363 + }, + { + "epoch": 0.3677398633692416, + "grad_norm": 0.32421875, + "learning_rate": 0.0012814009471016606, + "loss": 0.0952, + "step": 42364 + }, + { + "epoch": 0.3677485438494458, + "grad_norm": 0.578125, + "learning_rate": 0.0012813717953646042, + "loss": 0.1406, + "step": 42365 + }, + { + "epoch": 0.3677572243296499, + "grad_norm": 0.30859375, + "learning_rate": 0.0012813426434292023, + "loss": 0.1123, + "step": 42366 + }, + { + "epoch": 0.3677659048098541, + "grad_norm": 0.494140625, + "learning_rate": 0.0012813134912954866, + "loss": 0.0889, + "step": 42367 + }, + { + "epoch": 0.36777458529005824, + "grad_norm": 0.2373046875, + "learning_rate": 0.0012812843389634892, + "loss": 0.1357, + "step": 42368 + }, + { + "epoch": 0.36778326577026244, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012812551864332421, + "loss": 0.0947, + "step": 42369 + }, + { + "epoch": 0.3677919462504666, + "grad_norm": 0.5234375, + "learning_rate": 0.0012812260337047771, + "loss": 0.1162, + "step": 42370 + }, + { + "epoch": 0.36780062673067077, + "grad_norm": 0.1298828125, + "learning_rate": 0.001281196880778126, + "loss": 0.0967, + "step": 42371 + }, + { + "epoch": 0.3678093072108749, + "grad_norm": 0.8203125, + "learning_rate": 0.001281167727653321, + "loss": 0.1504, + "step": 42372 + }, + { + "epoch": 0.3678179876910791, + "grad_norm": 0.421875, + "learning_rate": 0.0012811385743303932, + "loss": 0.0923, + "step": 42373 + }, + { + "epoch": 0.36782666817128323, + "grad_norm": 0.26171875, + "learning_rate": 0.001281109420809375, + "loss": 0.0894, + "step": 42374 + }, + { + "epoch": 0.3678353486514874, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012810802670902986, + "loss": 0.0986, + "step": 42375 + }, + { + "epoch": 0.36784402913169156, + "grad_norm": 0.25390625, + "learning_rate": 0.001281051113173195, + "loss": 0.126, + "step": 42376 + }, + { + "epoch": 0.36785270961189576, + "grad_norm": 0.103515625, + "learning_rate": 0.001281021959058097, + "loss": 0.1069, + "step": 42377 + }, + { + "epoch": 0.3678613900920999, + "grad_norm": 0.169921875, + "learning_rate": 0.0012809928047450363, + "loss": 0.0703, + "step": 42378 + }, + { + "epoch": 0.3678700705723041, + "grad_norm": 0.2890625, + "learning_rate": 0.0012809636502340445, + "loss": 0.0835, + "step": 42379 + }, + { + "epoch": 0.3678787510525082, + "grad_norm": 0.5234375, + "learning_rate": 0.001280934495525153, + "loss": 0.0718, + "step": 42380 + }, + { + "epoch": 0.3678874315327124, + "grad_norm": 0.259765625, + "learning_rate": 0.001280905340618395, + "loss": 0.0894, + "step": 42381 + }, + { + "epoch": 0.36789611201291655, + "grad_norm": 0.208984375, + "learning_rate": 0.0012808761855138012, + "loss": 0.0986, + "step": 42382 + }, + { + "epoch": 0.36790479249312075, + "grad_norm": 0.1162109375, + "learning_rate": 0.0012808470302114042, + "loss": 0.1011, + "step": 42383 + }, + { + "epoch": 0.3679134729733249, + "grad_norm": 0.75, + "learning_rate": 0.0012808178747112354, + "loss": 0.0752, + "step": 42384 + }, + { + "epoch": 0.367922153453529, + "grad_norm": 0.1123046875, + "learning_rate": 0.0012807887190133267, + "loss": 0.1064, + "step": 42385 + }, + { + "epoch": 0.3679308339337332, + "grad_norm": 0.36328125, + "learning_rate": 0.0012807595631177107, + "loss": 0.1953, + "step": 42386 + }, + { + "epoch": 0.36793951441393735, + "grad_norm": 0.1279296875, + "learning_rate": 0.0012807304070244186, + "loss": 0.1172, + "step": 42387 + }, + { + "epoch": 0.36794819489414154, + "grad_norm": 0.384765625, + "learning_rate": 0.0012807012507334823, + "loss": 0.0986, + "step": 42388 + }, + { + "epoch": 0.3679568753743457, + "grad_norm": 0.1669921875, + "learning_rate": 0.0012806720942449337, + "loss": 0.083, + "step": 42389 + }, + { + "epoch": 0.3679655558545499, + "grad_norm": 0.453125, + "learning_rate": 0.001280642937558805, + "loss": 0.0776, + "step": 42390 + }, + { + "epoch": 0.367974236334754, + "grad_norm": 0.71875, + "learning_rate": 0.0012806137806751277, + "loss": 0.1611, + "step": 42391 + }, + { + "epoch": 0.3679829168149582, + "grad_norm": 0.1904296875, + "learning_rate": 0.001280584623593934, + "loss": 0.1099, + "step": 42392 + }, + { + "epoch": 0.36799159729516234, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012805554663152558, + "loss": 0.085, + "step": 42393 + }, + { + "epoch": 0.36800027777536654, + "grad_norm": 0.083984375, + "learning_rate": 0.0012805263088391248, + "loss": 0.1025, + "step": 42394 + }, + { + "epoch": 0.3680089582555707, + "grad_norm": 0.302734375, + "learning_rate": 0.001280497151165573, + "loss": 0.1079, + "step": 42395 + }, + { + "epoch": 0.36801763873577487, + "grad_norm": 0.07373046875, + "learning_rate": 0.001280467993294632, + "loss": 0.083, + "step": 42396 + }, + { + "epoch": 0.368026319215979, + "grad_norm": 0.470703125, + "learning_rate": 0.0012804388352263341, + "loss": 0.0771, + "step": 42397 + }, + { + "epoch": 0.3680349996961832, + "grad_norm": 0.09228515625, + "learning_rate": 0.001280409676960711, + "loss": 0.0767, + "step": 42398 + }, + { + "epoch": 0.36804368017638733, + "grad_norm": 0.2734375, + "learning_rate": 0.0012803805184977946, + "loss": 0.1069, + "step": 42399 + }, + { + "epoch": 0.3680523606565915, + "grad_norm": 0.93359375, + "learning_rate": 0.0012803513598376166, + "loss": 0.2041, + "step": 42400 + }, + { + "epoch": 0.36806104113679566, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012803222009802092, + "loss": 0.0732, + "step": 42401 + }, + { + "epoch": 0.36806972161699986, + "grad_norm": 1.9375, + "learning_rate": 0.0012802930419256039, + "loss": 0.1006, + "step": 42402 + }, + { + "epoch": 0.368078402097204, + "grad_norm": 0.11328125, + "learning_rate": 0.0012802638826738331, + "loss": 0.1211, + "step": 42403 + }, + { + "epoch": 0.3680870825774082, + "grad_norm": 0.259765625, + "learning_rate": 0.0012802347232249286, + "loss": 0.0781, + "step": 42404 + }, + { + "epoch": 0.3680957630576123, + "grad_norm": 0.13671875, + "learning_rate": 0.0012802055635789218, + "loss": 0.0977, + "step": 42405 + }, + { + "epoch": 0.3681044435378165, + "grad_norm": 0.80078125, + "learning_rate": 0.001280176403735845, + "loss": 0.1318, + "step": 42406 + }, + { + "epoch": 0.36811312401802065, + "grad_norm": 0.2080078125, + "learning_rate": 0.00128014724369573, + "loss": 0.083, + "step": 42407 + }, + { + "epoch": 0.36812180449822485, + "grad_norm": 0.244140625, + "learning_rate": 0.0012801180834586087, + "loss": 0.1123, + "step": 42408 + }, + { + "epoch": 0.368130484978429, + "grad_norm": 0.27734375, + "learning_rate": 0.0012800889230245131, + "loss": 0.0908, + "step": 42409 + }, + { + "epoch": 0.3681391654586332, + "grad_norm": 0.123046875, + "learning_rate": 0.0012800597623934746, + "loss": 0.167, + "step": 42410 + }, + { + "epoch": 0.3681478459388373, + "grad_norm": 0.439453125, + "learning_rate": 0.0012800306015655256, + "loss": 0.085, + "step": 42411 + }, + { + "epoch": 0.3681565264190415, + "grad_norm": 0.53515625, + "learning_rate": 0.001280001440540698, + "loss": 0.1797, + "step": 42412 + }, + { + "epoch": 0.36816520689924565, + "grad_norm": 0.55859375, + "learning_rate": 0.0012799722793190235, + "loss": 0.1216, + "step": 42413 + }, + { + "epoch": 0.36817388737944984, + "grad_norm": 0.158203125, + "learning_rate": 0.001279943117900534, + "loss": 0.0889, + "step": 42414 + }, + { + "epoch": 0.368182567859654, + "grad_norm": 0.7734375, + "learning_rate": 0.001279913956285261, + "loss": 0.125, + "step": 42415 + }, + { + "epoch": 0.36819124833985817, + "grad_norm": 0.1787109375, + "learning_rate": 0.0012798847944732374, + "loss": 0.0996, + "step": 42416 + }, + { + "epoch": 0.3681999288200623, + "grad_norm": 0.0966796875, + "learning_rate": 0.0012798556324644943, + "loss": 0.1143, + "step": 42417 + }, + { + "epoch": 0.3682086093002665, + "grad_norm": 0.1572265625, + "learning_rate": 0.0012798264702590639, + "loss": 0.0977, + "step": 42418 + }, + { + "epoch": 0.36821728978047064, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012797973078569776, + "loss": 0.1206, + "step": 42419 + }, + { + "epoch": 0.36822597026067483, + "grad_norm": 0.388671875, + "learning_rate": 0.0012797681452582676, + "loss": 0.1504, + "step": 42420 + }, + { + "epoch": 0.36823465074087897, + "grad_norm": 0.07763671875, + "learning_rate": 0.0012797389824629661, + "loss": 0.082, + "step": 42421 + }, + { + "epoch": 0.36824333122108316, + "grad_norm": 0.369140625, + "learning_rate": 0.001279709819471105, + "loss": 0.1797, + "step": 42422 + }, + { + "epoch": 0.3682520117012873, + "grad_norm": 0.41015625, + "learning_rate": 0.001279680656282716, + "loss": 0.0996, + "step": 42423 + }, + { + "epoch": 0.3682606921814915, + "grad_norm": 0.443359375, + "learning_rate": 0.0012796514928978305, + "loss": 0.0718, + "step": 42424 + }, + { + "epoch": 0.3682693726616956, + "grad_norm": 0.14453125, + "learning_rate": 0.001279622329316481, + "loss": 0.1167, + "step": 42425 + }, + { + "epoch": 0.3682780531418998, + "grad_norm": 0.3203125, + "learning_rate": 0.001279593165538699, + "loss": 0.1025, + "step": 42426 + }, + { + "epoch": 0.36828673362210396, + "grad_norm": 0.1787109375, + "learning_rate": 0.0012795640015645171, + "loss": 0.1138, + "step": 42427 + }, + { + "epoch": 0.36829541410230815, + "grad_norm": 0.208984375, + "learning_rate": 0.0012795348373939664, + "loss": 0.1387, + "step": 42428 + }, + { + "epoch": 0.3683040945825123, + "grad_norm": 0.10009765625, + "learning_rate": 0.0012795056730270792, + "loss": 0.0933, + "step": 42429 + }, + { + "epoch": 0.3683127750627165, + "grad_norm": 0.203125, + "learning_rate": 0.0012794765084638868, + "loss": 0.0962, + "step": 42430 + }, + { + "epoch": 0.3683214555429206, + "grad_norm": 0.421875, + "learning_rate": 0.001279447343704422, + "loss": 0.1211, + "step": 42431 + }, + { + "epoch": 0.3683301360231248, + "grad_norm": 0.1494140625, + "learning_rate": 0.0012794181787487162, + "loss": 0.0825, + "step": 42432 + }, + { + "epoch": 0.36833881650332895, + "grad_norm": 0.4375, + "learning_rate": 0.0012793890135968016, + "loss": 0.1108, + "step": 42433 + }, + { + "epoch": 0.36834749698353314, + "grad_norm": 0.10205078125, + "learning_rate": 0.0012793598482487099, + "loss": 0.0903, + "step": 42434 + }, + { + "epoch": 0.3683561774637373, + "grad_norm": 0.6484375, + "learning_rate": 0.0012793306827044728, + "loss": 0.1045, + "step": 42435 + }, + { + "epoch": 0.36836485794394147, + "grad_norm": 0.3203125, + "learning_rate": 0.0012793015169641222, + "loss": 0.1035, + "step": 42436 + }, + { + "epoch": 0.3683735384241456, + "grad_norm": 0.310546875, + "learning_rate": 0.0012792723510276903, + "loss": 0.1035, + "step": 42437 + }, + { + "epoch": 0.3683822189043498, + "grad_norm": 0.48828125, + "learning_rate": 0.001279243184895209, + "loss": 0.1172, + "step": 42438 + }, + { + "epoch": 0.36839089938455394, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012792140185667099, + "loss": 0.0962, + "step": 42439 + }, + { + "epoch": 0.36839957986475813, + "grad_norm": 0.1796875, + "learning_rate": 0.0012791848520422248, + "loss": 0.0796, + "step": 42440 + }, + { + "epoch": 0.36840826034496227, + "grad_norm": 0.12451171875, + "learning_rate": 0.001279155685321786, + "loss": 0.1367, + "step": 42441 + }, + { + "epoch": 0.36841694082516646, + "grad_norm": 0.076171875, + "learning_rate": 0.0012791265184054256, + "loss": 0.0918, + "step": 42442 + }, + { + "epoch": 0.3684256213053706, + "grad_norm": 0.2021484375, + "learning_rate": 0.0012790973512931748, + "loss": 0.0996, + "step": 42443 + }, + { + "epoch": 0.3684343017855748, + "grad_norm": 0.345703125, + "learning_rate": 0.0012790681839850656, + "loss": 0.1006, + "step": 42444 + }, + { + "epoch": 0.36844298226577893, + "grad_norm": 0.3125, + "learning_rate": 0.0012790390164811305, + "loss": 0.0776, + "step": 42445 + }, + { + "epoch": 0.3684516627459831, + "grad_norm": 0.177734375, + "learning_rate": 0.001279009848781401, + "loss": 0.1045, + "step": 42446 + }, + { + "epoch": 0.36846034322618726, + "grad_norm": 0.130859375, + "learning_rate": 0.0012789806808859092, + "loss": 0.0884, + "step": 42447 + }, + { + "epoch": 0.36846902370639145, + "grad_norm": 0.08203125, + "learning_rate": 0.0012789515127946862, + "loss": 0.0869, + "step": 42448 + }, + { + "epoch": 0.3684777041865956, + "grad_norm": 0.0927734375, + "learning_rate": 0.001278922344507765, + "loss": 0.0947, + "step": 42449 + }, + { + "epoch": 0.3684863846667998, + "grad_norm": 0.443359375, + "learning_rate": 0.0012788931760251768, + "loss": 0.125, + "step": 42450 + }, + { + "epoch": 0.3684950651470039, + "grad_norm": 0.1904296875, + "learning_rate": 0.0012788640073469538, + "loss": 0.1201, + "step": 42451 + }, + { + "epoch": 0.3685037456272081, + "grad_norm": 0.53515625, + "learning_rate": 0.001278834838473128, + "loss": 0.127, + "step": 42452 + }, + { + "epoch": 0.36851242610741225, + "grad_norm": 0.09765625, + "learning_rate": 0.0012788056694037308, + "loss": 0.1025, + "step": 42453 + }, + { + "epoch": 0.36852110658761644, + "grad_norm": 0.146484375, + "learning_rate": 0.0012787765001387942, + "loss": 0.0952, + "step": 42454 + }, + { + "epoch": 0.3685297870678206, + "grad_norm": 0.287109375, + "learning_rate": 0.0012787473306783508, + "loss": 0.0967, + "step": 42455 + }, + { + "epoch": 0.3685384675480248, + "grad_norm": 0.5, + "learning_rate": 0.0012787181610224318, + "loss": 0.1328, + "step": 42456 + }, + { + "epoch": 0.3685471480282289, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012786889911710695, + "loss": 0.0815, + "step": 42457 + }, + { + "epoch": 0.3685558285084331, + "grad_norm": 0.1611328125, + "learning_rate": 0.0012786598211242952, + "loss": 0.0986, + "step": 42458 + }, + { + "epoch": 0.36856450898863724, + "grad_norm": 0.56640625, + "learning_rate": 0.0012786306508821414, + "loss": 0.1094, + "step": 42459 + }, + { + "epoch": 0.36857318946884143, + "grad_norm": 0.26171875, + "learning_rate": 0.0012786014804446399, + "loss": 0.0889, + "step": 42460 + }, + { + "epoch": 0.36858186994904557, + "grad_norm": 0.271484375, + "learning_rate": 0.0012785723098118226, + "loss": 0.0991, + "step": 42461 + }, + { + "epoch": 0.36859055042924976, + "grad_norm": 0.10498046875, + "learning_rate": 0.001278543138983721, + "loss": 0.1099, + "step": 42462 + }, + { + "epoch": 0.3685992309094539, + "grad_norm": 0.388671875, + "learning_rate": 0.0012785139679603675, + "loss": 0.0708, + "step": 42463 + }, + { + "epoch": 0.3686079113896581, + "grad_norm": 0.0888671875, + "learning_rate": 0.0012784847967417942, + "loss": 0.0737, + "step": 42464 + }, + { + "epoch": 0.36861659186986223, + "grad_norm": 0.423828125, + "learning_rate": 0.0012784556253280317, + "loss": 0.0806, + "step": 42465 + }, + { + "epoch": 0.3686252723500664, + "grad_norm": 0.1767578125, + "learning_rate": 0.0012784264537191135, + "loss": 0.1011, + "step": 42466 + }, + { + "epoch": 0.36863395283027056, + "grad_norm": 0.2734375, + "learning_rate": 0.0012783972819150706, + "loss": 0.1104, + "step": 42467 + }, + { + "epoch": 0.36864263331047475, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012783681099159349, + "loss": 0.083, + "step": 42468 + }, + { + "epoch": 0.3686513137906789, + "grad_norm": 0.08154296875, + "learning_rate": 0.0012783389377217388, + "loss": 0.084, + "step": 42469 + }, + { + "epoch": 0.3686599942708831, + "grad_norm": 0.671875, + "learning_rate": 0.0012783097653325138, + "loss": 0.1084, + "step": 42470 + }, + { + "epoch": 0.3686686747510872, + "grad_norm": 0.2431640625, + "learning_rate": 0.001278280592748292, + "loss": 0.0894, + "step": 42471 + }, + { + "epoch": 0.3686773552312914, + "grad_norm": 0.216796875, + "learning_rate": 0.001278251419969105, + "loss": 0.1084, + "step": 42472 + }, + { + "epoch": 0.36868603571149555, + "grad_norm": 0.640625, + "learning_rate": 0.0012782222469949853, + "loss": 0.1055, + "step": 42473 + }, + { + "epoch": 0.36869471619169975, + "grad_norm": 0.0947265625, + "learning_rate": 0.0012781930738259642, + "loss": 0.0615, + "step": 42474 + }, + { + "epoch": 0.3687033966719039, + "grad_norm": 0.1484375, + "learning_rate": 0.0012781639004620737, + "loss": 0.0664, + "step": 42475 + }, + { + "epoch": 0.3687120771521081, + "grad_norm": 0.65625, + "learning_rate": 0.0012781347269033463, + "loss": 0.1196, + "step": 42476 + }, + { + "epoch": 0.3687207576323122, + "grad_norm": 0.703125, + "learning_rate": 0.0012781055531498131, + "loss": 0.0596, + "step": 42477 + }, + { + "epoch": 0.3687294381125164, + "grad_norm": 0.330078125, + "learning_rate": 0.0012780763792015063, + "loss": 0.1035, + "step": 42478 + }, + { + "epoch": 0.36873811859272054, + "grad_norm": 0.44140625, + "learning_rate": 0.0012780472050584579, + "loss": 0.0796, + "step": 42479 + }, + { + "epoch": 0.36874679907292474, + "grad_norm": 0.2734375, + "learning_rate": 0.0012780180307206998, + "loss": 0.1226, + "step": 42480 + }, + { + "epoch": 0.3687554795531289, + "grad_norm": 0.2373046875, + "learning_rate": 0.0012779888561882638, + "loss": 0.0669, + "step": 42481 + }, + { + "epoch": 0.36876416003333307, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012779596814611817, + "loss": 0.1299, + "step": 42482 + }, + { + "epoch": 0.3687728405135372, + "grad_norm": 0.138671875, + "learning_rate": 0.0012779305065394862, + "loss": 0.1143, + "step": 42483 + }, + { + "epoch": 0.3687815209937414, + "grad_norm": 0.12109375, + "learning_rate": 0.001277901331423208, + "loss": 0.1172, + "step": 42484 + }, + { + "epoch": 0.36879020147394553, + "grad_norm": 0.140625, + "learning_rate": 0.0012778721561123795, + "loss": 0.0659, + "step": 42485 + }, + { + "epoch": 0.3687988819541497, + "grad_norm": 0.1787109375, + "learning_rate": 0.0012778429806070332, + "loss": 0.0957, + "step": 42486 + }, + { + "epoch": 0.36880756243435386, + "grad_norm": 0.2314453125, + "learning_rate": 0.0012778138049072, + "loss": 0.085, + "step": 42487 + }, + { + "epoch": 0.36881624291455806, + "grad_norm": 0.78125, + "learning_rate": 0.0012777846290129124, + "loss": 0.1016, + "step": 42488 + }, + { + "epoch": 0.3688249233947622, + "grad_norm": 0.4765625, + "learning_rate": 0.001277755452924202, + "loss": 0.0894, + "step": 42489 + }, + { + "epoch": 0.3688336038749664, + "grad_norm": 0.42578125, + "learning_rate": 0.0012777262766411016, + "loss": 0.0825, + "step": 42490 + }, + { + "epoch": 0.3688422843551705, + "grad_norm": 0.1376953125, + "learning_rate": 0.001277697100163642, + "loss": 0.1118, + "step": 42491 + }, + { + "epoch": 0.3688509648353747, + "grad_norm": 0.1064453125, + "learning_rate": 0.0012776679234918556, + "loss": 0.0825, + "step": 42492 + }, + { + "epoch": 0.36885964531557885, + "grad_norm": 0.353515625, + "learning_rate": 0.001277638746625774, + "loss": 0.0933, + "step": 42493 + }, + { + "epoch": 0.36886832579578305, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012776095695654294, + "loss": 0.1426, + "step": 42494 + }, + { + "epoch": 0.3688770062759872, + "grad_norm": 0.671875, + "learning_rate": 0.0012775803923108539, + "loss": 0.0815, + "step": 42495 + }, + { + "epoch": 0.3688856867561914, + "grad_norm": 0.1787109375, + "learning_rate": 0.001277551214862079, + "loss": 0.0815, + "step": 42496 + }, + { + "epoch": 0.3688943672363955, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012775220372191367, + "loss": 0.1143, + "step": 42497 + }, + { + "epoch": 0.3689030477165997, + "grad_norm": 0.10546875, + "learning_rate": 0.0012774928593820589, + "loss": 0.0645, + "step": 42498 + }, + { + "epoch": 0.36891172819680385, + "grad_norm": 0.1484375, + "learning_rate": 0.0012774636813508778, + "loss": 0.1069, + "step": 42499 + }, + { + "epoch": 0.36892040867700804, + "grad_norm": 0.6875, + "learning_rate": 0.001277434503125625, + "loss": 0.1387, + "step": 42500 + }, + { + "epoch": 0.3689290891572122, + "grad_norm": 0.55859375, + "learning_rate": 0.0012774053247063323, + "loss": 0.0786, + "step": 42501 + }, + { + "epoch": 0.36893776963741637, + "grad_norm": 0.404296875, + "learning_rate": 0.001277376146093032, + "loss": 0.0859, + "step": 42502 + }, + { + "epoch": 0.3689464501176205, + "grad_norm": 0.09423828125, + "learning_rate": 0.0012773469672857558, + "loss": 0.0791, + "step": 42503 + }, + { + "epoch": 0.3689551305978247, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012773177882845358, + "loss": 0.0918, + "step": 42504 + }, + { + "epoch": 0.36896381107802884, + "grad_norm": 0.1142578125, + "learning_rate": 0.0012772886090894034, + "loss": 0.1118, + "step": 42505 + }, + { + "epoch": 0.36897249155823303, + "grad_norm": 0.37890625, + "learning_rate": 0.0012772594297003907, + "loss": 0.1162, + "step": 42506 + }, + { + "epoch": 0.36898117203843717, + "grad_norm": 1.015625, + "learning_rate": 0.00127723025011753, + "loss": 0.0791, + "step": 42507 + }, + { + "epoch": 0.3689898525186413, + "grad_norm": 0.5, + "learning_rate": 0.001277201070340853, + "loss": 0.1396, + "step": 42508 + }, + { + "epoch": 0.3689985329988455, + "grad_norm": 0.8515625, + "learning_rate": 0.001277171890370392, + "loss": 0.0996, + "step": 42509 + }, + { + "epoch": 0.36900721347904963, + "grad_norm": 0.443359375, + "learning_rate": 0.0012771427102061777, + "loss": 0.1172, + "step": 42510 + }, + { + "epoch": 0.3690158939592538, + "grad_norm": 0.283203125, + "learning_rate": 0.001277113529848243, + "loss": 0.1016, + "step": 42511 + }, + { + "epoch": 0.36902457443945796, + "grad_norm": 0.365234375, + "learning_rate": 0.00127708434929662, + "loss": 0.123, + "step": 42512 + }, + { + "epoch": 0.36903325491966216, + "grad_norm": 0.353515625, + "learning_rate": 0.0012770551685513398, + "loss": 0.085, + "step": 42513 + }, + { + "epoch": 0.3690419353998663, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012770259876124348, + "loss": 0.1006, + "step": 42514 + }, + { + "epoch": 0.3690506158800705, + "grad_norm": 0.228515625, + "learning_rate": 0.0012769968064799372, + "loss": 0.1182, + "step": 42515 + }, + { + "epoch": 0.3690592963602746, + "grad_norm": 0.1806640625, + "learning_rate": 0.001276967625153878, + "loss": 0.0781, + "step": 42516 + }, + { + "epoch": 0.3690679768404788, + "grad_norm": 0.142578125, + "learning_rate": 0.00127693844363429, + "loss": 0.0688, + "step": 42517 + }, + { + "epoch": 0.36907665732068295, + "grad_norm": 0.11328125, + "learning_rate": 0.001276909261921205, + "loss": 0.0732, + "step": 42518 + }, + { + "epoch": 0.36908533780088715, + "grad_norm": 0.419921875, + "learning_rate": 0.001276880080014654, + "loss": 0.084, + "step": 42519 + }, + { + "epoch": 0.3690940182810913, + "grad_norm": 0.35546875, + "learning_rate": 0.0012768508979146703, + "loss": 0.0928, + "step": 42520 + }, + { + "epoch": 0.3691026987612955, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012768217156212847, + "loss": 0.1006, + "step": 42521 + }, + { + "epoch": 0.3691113792414996, + "grad_norm": 0.42578125, + "learning_rate": 0.0012767925331345295, + "loss": 0.0723, + "step": 42522 + }, + { + "epoch": 0.3691200597217038, + "grad_norm": 0.10498046875, + "learning_rate": 0.001276763350454437, + "loss": 0.0884, + "step": 42523 + }, + { + "epoch": 0.36912874020190795, + "grad_norm": 0.40234375, + "learning_rate": 0.0012767341675810385, + "loss": 0.1777, + "step": 42524 + }, + { + "epoch": 0.36913742068211214, + "grad_norm": 0.671875, + "learning_rate": 0.001276704984514366, + "loss": 0.0957, + "step": 42525 + }, + { + "epoch": 0.3691461011623163, + "grad_norm": 0.240234375, + "learning_rate": 0.001276675801254452, + "loss": 0.1001, + "step": 42526 + }, + { + "epoch": 0.36915478164252047, + "grad_norm": 0.328125, + "learning_rate": 0.0012766466178013274, + "loss": 0.082, + "step": 42527 + }, + { + "epoch": 0.3691634621227246, + "grad_norm": 1.2734375, + "learning_rate": 0.0012766174341550254, + "loss": 0.1602, + "step": 42528 + }, + { + "epoch": 0.3691721426029288, + "grad_norm": 0.34375, + "learning_rate": 0.0012765882503155768, + "loss": 0.1426, + "step": 42529 + }, + { + "epoch": 0.36918082308313294, + "grad_norm": 0.185546875, + "learning_rate": 0.0012765590662830143, + "loss": 0.1084, + "step": 42530 + }, + { + "epoch": 0.36918950356333713, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012765298820573694, + "loss": 0.0967, + "step": 42531 + }, + { + "epoch": 0.36919818404354127, + "grad_norm": 0.2294921875, + "learning_rate": 0.0012765006976386737, + "loss": 0.0986, + "step": 42532 + }, + { + "epoch": 0.36920686452374546, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012764715130269598, + "loss": 0.0913, + "step": 42533 + }, + { + "epoch": 0.3692155450039496, + "grad_norm": 0.19921875, + "learning_rate": 0.0012764423282222593, + "loss": 0.1162, + "step": 42534 + }, + { + "epoch": 0.3692242254841538, + "grad_norm": 0.2578125, + "learning_rate": 0.0012764131432246037, + "loss": 0.0996, + "step": 42535 + }, + { + "epoch": 0.3692329059643579, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012763839580340254, + "loss": 0.1211, + "step": 42536 + }, + { + "epoch": 0.3692415864445621, + "grad_norm": 0.45703125, + "learning_rate": 0.001276354772650557, + "loss": 0.1426, + "step": 42537 + }, + { + "epoch": 0.36925026692476626, + "grad_norm": 0.06689453125, + "learning_rate": 0.001276325587074229, + "loss": 0.0649, + "step": 42538 + }, + { + "epoch": 0.36925894740497045, + "grad_norm": 0.189453125, + "learning_rate": 0.0012762964013050741, + "loss": 0.0952, + "step": 42539 + }, + { + "epoch": 0.3692676278851746, + "grad_norm": 0.1875, + "learning_rate": 0.0012762672153431244, + "loss": 0.0908, + "step": 42540 + }, + { + "epoch": 0.3692763083653788, + "grad_norm": 0.1728515625, + "learning_rate": 0.0012762380291884113, + "loss": 0.0835, + "step": 42541 + }, + { + "epoch": 0.3692849888455829, + "grad_norm": 0.5, + "learning_rate": 0.0012762088428409668, + "loss": 0.0918, + "step": 42542 + }, + { + "epoch": 0.3692936693257871, + "grad_norm": 0.1162109375, + "learning_rate": 0.0012761796563008233, + "loss": 0.124, + "step": 42543 + }, + { + "epoch": 0.36930234980599125, + "grad_norm": 0.35546875, + "learning_rate": 0.0012761504695680118, + "loss": 0.0933, + "step": 42544 + }, + { + "epoch": 0.36931103028619544, + "grad_norm": 0.33984375, + "learning_rate": 0.0012761212826425652, + "loss": 0.1309, + "step": 42545 + }, + { + "epoch": 0.3693197107663996, + "grad_norm": 0.5703125, + "learning_rate": 0.001276092095524515, + "loss": 0.0835, + "step": 42546 + }, + { + "epoch": 0.36932839124660377, + "grad_norm": 0.59375, + "learning_rate": 0.0012760629082138933, + "loss": 0.0811, + "step": 42547 + }, + { + "epoch": 0.3693370717268079, + "grad_norm": 0.48828125, + "learning_rate": 0.0012760337207107317, + "loss": 0.1064, + "step": 42548 + }, + { + "epoch": 0.3693457522070121, + "grad_norm": 0.19140625, + "learning_rate": 0.001276004533015062, + "loss": 0.0962, + "step": 42549 + }, + { + "epoch": 0.36935443268721624, + "grad_norm": 0.1328125, + "learning_rate": 0.0012759753451269168, + "loss": 0.1045, + "step": 42550 + }, + { + "epoch": 0.36936311316742043, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012759461570463274, + "loss": 0.0771, + "step": 42551 + }, + { + "epoch": 0.36937179364762457, + "grad_norm": 0.609375, + "learning_rate": 0.0012759169687733259, + "loss": 0.1104, + "step": 42552 + }, + { + "epoch": 0.36938047412782876, + "grad_norm": 0.10791015625, + "learning_rate": 0.0012758877803079442, + "loss": 0.0645, + "step": 42553 + }, + { + "epoch": 0.3693891546080329, + "grad_norm": 0.51953125, + "learning_rate": 0.0012758585916502144, + "loss": 0.1162, + "step": 42554 + }, + { + "epoch": 0.3693978350882371, + "grad_norm": 0.57421875, + "learning_rate": 0.001275829402800168, + "loss": 0.0957, + "step": 42555 + }, + { + "epoch": 0.36940651556844123, + "grad_norm": 0.326171875, + "learning_rate": 0.0012758002137578377, + "loss": 0.126, + "step": 42556 + }, + { + "epoch": 0.3694151960486454, + "grad_norm": 0.25390625, + "learning_rate": 0.0012757710245232547, + "loss": 0.1494, + "step": 42557 + }, + { + "epoch": 0.36942387652884956, + "grad_norm": 0.396484375, + "learning_rate": 0.0012757418350964511, + "loss": 0.0977, + "step": 42558 + }, + { + "epoch": 0.36943255700905375, + "grad_norm": 0.14453125, + "learning_rate": 0.0012757126454774588, + "loss": 0.1318, + "step": 42559 + }, + { + "epoch": 0.3694412374892579, + "grad_norm": 0.1259765625, + "learning_rate": 0.00127568345566631, + "loss": 0.1143, + "step": 42560 + }, + { + "epoch": 0.3694499179694621, + "grad_norm": 0.130859375, + "learning_rate": 0.0012756542656630365, + "loss": 0.083, + "step": 42561 + }, + { + "epoch": 0.3694585984496662, + "grad_norm": 0.1123046875, + "learning_rate": 0.0012756250754676697, + "loss": 0.0811, + "step": 42562 + }, + { + "epoch": 0.3694672789298704, + "grad_norm": 1.0859375, + "learning_rate": 0.0012755958850802422, + "loss": 0.1104, + "step": 42563 + }, + { + "epoch": 0.36947595941007455, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012755666945007856, + "loss": 0.1211, + "step": 42564 + }, + { + "epoch": 0.36948463989027874, + "grad_norm": 0.38671875, + "learning_rate": 0.0012755375037293319, + "loss": 0.0938, + "step": 42565 + }, + { + "epoch": 0.3694933203704829, + "grad_norm": 0.78125, + "learning_rate": 0.001275508312765913, + "loss": 0.0957, + "step": 42566 + }, + { + "epoch": 0.3695020008506871, + "grad_norm": 0.263671875, + "learning_rate": 0.001275479121610561, + "loss": 0.1172, + "step": 42567 + }, + { + "epoch": 0.3695106813308912, + "grad_norm": 0.1650390625, + "learning_rate": 0.001275449930263308, + "loss": 0.0952, + "step": 42568 + }, + { + "epoch": 0.3695193618110954, + "grad_norm": 0.1953125, + "learning_rate": 0.0012754207387241849, + "loss": 0.1504, + "step": 42569 + }, + { + "epoch": 0.36952804229129954, + "grad_norm": 0.06982421875, + "learning_rate": 0.0012753915469932244, + "loss": 0.0708, + "step": 42570 + }, + { + "epoch": 0.36953672277150373, + "grad_norm": 0.46484375, + "learning_rate": 0.0012753623550704586, + "loss": 0.0977, + "step": 42571 + }, + { + "epoch": 0.36954540325170787, + "grad_norm": 0.255859375, + "learning_rate": 0.0012753331629559192, + "loss": 0.1484, + "step": 42572 + }, + { + "epoch": 0.36955408373191206, + "grad_norm": 0.1953125, + "learning_rate": 0.0012753039706496381, + "loss": 0.0928, + "step": 42573 + }, + { + "epoch": 0.3695627642121162, + "grad_norm": 0.10400390625, + "learning_rate": 0.0012752747781516471, + "loss": 0.0742, + "step": 42574 + }, + { + "epoch": 0.3695714446923204, + "grad_norm": 0.2236328125, + "learning_rate": 0.001275245585461978, + "loss": 0.1523, + "step": 42575 + }, + { + "epoch": 0.36958012517252453, + "grad_norm": 0.244140625, + "learning_rate": 0.0012752163925806635, + "loss": 0.1162, + "step": 42576 + }, + { + "epoch": 0.3695888056527287, + "grad_norm": 0.17578125, + "learning_rate": 0.0012751871995077347, + "loss": 0.0923, + "step": 42577 + }, + { + "epoch": 0.36959748613293286, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012751580062432238, + "loss": 0.1035, + "step": 42578 + }, + { + "epoch": 0.36960616661313705, + "grad_norm": 0.1494140625, + "learning_rate": 0.0012751288127871626, + "loss": 0.1245, + "step": 42579 + }, + { + "epoch": 0.3696148470933412, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012750996191395834, + "loss": 0.1289, + "step": 42580 + }, + { + "epoch": 0.3696235275735454, + "grad_norm": 0.42578125, + "learning_rate": 0.0012750704253005178, + "loss": 0.1113, + "step": 42581 + }, + { + "epoch": 0.3696322080537495, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012750412312699977, + "loss": 0.0801, + "step": 42582 + }, + { + "epoch": 0.3696408885339537, + "grad_norm": 0.275390625, + "learning_rate": 0.0012750120370480554, + "loss": 0.0771, + "step": 42583 + }, + { + "epoch": 0.36964956901415785, + "grad_norm": 0.255859375, + "learning_rate": 0.0012749828426347222, + "loss": 0.1299, + "step": 42584 + }, + { + "epoch": 0.36965824949436205, + "grad_norm": 0.365234375, + "learning_rate": 0.0012749536480300308, + "loss": 0.1123, + "step": 42585 + }, + { + "epoch": 0.3696669299745662, + "grad_norm": 0.11767578125, + "learning_rate": 0.0012749244532340126, + "loss": 0.126, + "step": 42586 + }, + { + "epoch": 0.3696756104547704, + "grad_norm": 0.5234375, + "learning_rate": 0.0012748952582466995, + "loss": 0.0864, + "step": 42587 + }, + { + "epoch": 0.3696842909349745, + "grad_norm": 0.361328125, + "learning_rate": 0.0012748660630681236, + "loss": 0.1377, + "step": 42588 + }, + { + "epoch": 0.3696929714151787, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012748368676983167, + "loss": 0.0806, + "step": 42589 + }, + { + "epoch": 0.36970165189538284, + "grad_norm": 0.2255859375, + "learning_rate": 0.0012748076721373113, + "loss": 0.1064, + "step": 42590 + }, + { + "epoch": 0.36971033237558704, + "grad_norm": 0.275390625, + "learning_rate": 0.0012747784763851384, + "loss": 0.0776, + "step": 42591 + }, + { + "epoch": 0.3697190128557912, + "grad_norm": 0.322265625, + "learning_rate": 0.0012747492804418303, + "loss": 0.0933, + "step": 42592 + }, + { + "epoch": 0.36972769333599537, + "grad_norm": 0.3125, + "learning_rate": 0.0012747200843074191, + "loss": 0.1055, + "step": 42593 + }, + { + "epoch": 0.3697363738161995, + "grad_norm": 0.9921875, + "learning_rate": 0.001274690887981937, + "loss": 0.4219, + "step": 42594 + }, + { + "epoch": 0.3697450542964037, + "grad_norm": 0.2099609375, + "learning_rate": 0.0012746616914654152, + "loss": 0.1816, + "step": 42595 + }, + { + "epoch": 0.36975373477660783, + "grad_norm": 0.07666015625, + "learning_rate": 0.001274632494757886, + "loss": 0.0654, + "step": 42596 + }, + { + "epoch": 0.369762415256812, + "grad_norm": 0.27734375, + "learning_rate": 0.0012746032978593815, + "loss": 0.0615, + "step": 42597 + }, + { + "epoch": 0.36977109573701616, + "grad_norm": 0.287109375, + "learning_rate": 0.0012745741007699337, + "loss": 0.1089, + "step": 42598 + }, + { + "epoch": 0.36977977621722036, + "grad_norm": 0.373046875, + "learning_rate": 0.001274544903489574, + "loss": 0.0879, + "step": 42599 + }, + { + "epoch": 0.3697884566974245, + "grad_norm": 0.134765625, + "learning_rate": 0.0012745157060183347, + "loss": 0.0688, + "step": 42600 + }, + { + "epoch": 0.3697971371776287, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012744865083562475, + "loss": 0.1084, + "step": 42601 + }, + { + "epoch": 0.3698058176578328, + "grad_norm": 0.6640625, + "learning_rate": 0.0012744573105033445, + "loss": 0.1484, + "step": 42602 + }, + { + "epoch": 0.369814498138037, + "grad_norm": 0.1328125, + "learning_rate": 0.0012744281124596577, + "loss": 0.1045, + "step": 42603 + }, + { + "epoch": 0.36982317861824116, + "grad_norm": 0.76953125, + "learning_rate": 0.001274398914225219, + "loss": 0.1143, + "step": 42604 + }, + { + "epoch": 0.36983185909844535, + "grad_norm": 0.3203125, + "learning_rate": 0.0012743697158000599, + "loss": 0.0718, + "step": 42605 + }, + { + "epoch": 0.3698405395786495, + "grad_norm": 0.1142578125, + "learning_rate": 0.0012743405171842129, + "loss": 0.0859, + "step": 42606 + }, + { + "epoch": 0.3698492200588537, + "grad_norm": 0.5078125, + "learning_rate": 0.00127431131837771, + "loss": 0.0908, + "step": 42607 + }, + { + "epoch": 0.3698579005390578, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012742821193805825, + "loss": 0.126, + "step": 42608 + }, + { + "epoch": 0.369866581019262, + "grad_norm": 0.55859375, + "learning_rate": 0.0012742529201928627, + "loss": 0.1133, + "step": 42609 + }, + { + "epoch": 0.36987526149946615, + "grad_norm": 0.0791015625, + "learning_rate": 0.0012742237208145828, + "loss": 0.0801, + "step": 42610 + }, + { + "epoch": 0.36988394197967034, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012741945212457744, + "loss": 0.0996, + "step": 42611 + }, + { + "epoch": 0.3698926224598745, + "grad_norm": 0.25, + "learning_rate": 0.0012741653214864695, + "loss": 0.1133, + "step": 42612 + }, + { + "epoch": 0.36990130294007867, + "grad_norm": 0.142578125, + "learning_rate": 0.0012741361215366997, + "loss": 0.1021, + "step": 42613 + }, + { + "epoch": 0.3699099834202828, + "grad_norm": 0.08837890625, + "learning_rate": 0.0012741069213964976, + "loss": 0.082, + "step": 42614 + }, + { + "epoch": 0.369918663900487, + "grad_norm": 0.091796875, + "learning_rate": 0.0012740777210658945, + "loss": 0.0894, + "step": 42615 + }, + { + "epoch": 0.36992734438069114, + "grad_norm": 0.23046875, + "learning_rate": 0.0012740485205449228, + "loss": 0.1123, + "step": 42616 + }, + { + "epoch": 0.36993602486089533, + "grad_norm": 0.322265625, + "learning_rate": 0.0012740193198336144, + "loss": 0.0854, + "step": 42617 + }, + { + "epoch": 0.36994470534109947, + "grad_norm": 0.08544921875, + "learning_rate": 0.001273990118932001, + "loss": 0.0791, + "step": 42618 + }, + { + "epoch": 0.36995338582130366, + "grad_norm": 0.181640625, + "learning_rate": 0.0012739609178401147, + "loss": 0.0684, + "step": 42619 + }, + { + "epoch": 0.3699620663015078, + "grad_norm": 0.421875, + "learning_rate": 0.0012739317165579868, + "loss": 0.0977, + "step": 42620 + }, + { + "epoch": 0.369970746781712, + "grad_norm": 0.435546875, + "learning_rate": 0.00127390251508565, + "loss": 0.0889, + "step": 42621 + }, + { + "epoch": 0.3699794272619161, + "grad_norm": 0.13671875, + "learning_rate": 0.0012738733134231364, + "loss": 0.0718, + "step": 42622 + }, + { + "epoch": 0.3699881077421203, + "grad_norm": 0.193359375, + "learning_rate": 0.0012738441115704774, + "loss": 0.1357, + "step": 42623 + }, + { + "epoch": 0.36999678822232446, + "grad_norm": 0.6015625, + "learning_rate": 0.001273814909527705, + "loss": 0.1143, + "step": 42624 + }, + { + "epoch": 0.37000546870252865, + "grad_norm": 0.671875, + "learning_rate": 0.0012737857072948513, + "loss": 0.1064, + "step": 42625 + }, + { + "epoch": 0.3700141491827328, + "grad_norm": 0.78125, + "learning_rate": 0.0012737565048719483, + "loss": 0.1025, + "step": 42626 + }, + { + "epoch": 0.370022829662937, + "grad_norm": 0.12109375, + "learning_rate": 0.0012737273022590274, + "loss": 0.0991, + "step": 42627 + }, + { + "epoch": 0.3700315101431411, + "grad_norm": 0.2275390625, + "learning_rate": 0.0012736980994561213, + "loss": 0.085, + "step": 42628 + }, + { + "epoch": 0.3700401906233453, + "grad_norm": 0.296875, + "learning_rate": 0.0012736688964632614, + "loss": 0.1494, + "step": 42629 + }, + { + "epoch": 0.37004887110354945, + "grad_norm": 0.138671875, + "learning_rate": 0.00127363969328048, + "loss": 0.1064, + "step": 42630 + }, + { + "epoch": 0.3700575515837536, + "grad_norm": 0.0947265625, + "learning_rate": 0.0012736104899078087, + "loss": 0.0889, + "step": 42631 + }, + { + "epoch": 0.3700662320639578, + "grad_norm": 0.30859375, + "learning_rate": 0.0012735812863452793, + "loss": 0.1816, + "step": 42632 + }, + { + "epoch": 0.3700749125441619, + "grad_norm": 0.71875, + "learning_rate": 0.0012735520825929243, + "loss": 0.1133, + "step": 42633 + }, + { + "epoch": 0.3700835930243661, + "grad_norm": 0.546875, + "learning_rate": 0.0012735228786507756, + "loss": 0.1045, + "step": 42634 + }, + { + "epoch": 0.37009227350457025, + "grad_norm": 0.14453125, + "learning_rate": 0.0012734936745188645, + "loss": 0.1045, + "step": 42635 + }, + { + "epoch": 0.37010095398477444, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012734644701972236, + "loss": 0.1445, + "step": 42636 + }, + { + "epoch": 0.3701096344649786, + "grad_norm": 0.494140625, + "learning_rate": 0.0012734352656858844, + "loss": 0.0947, + "step": 42637 + }, + { + "epoch": 0.37011831494518277, + "grad_norm": 0.154296875, + "learning_rate": 0.001273406060984879, + "loss": 0.1201, + "step": 42638 + }, + { + "epoch": 0.3701269954253869, + "grad_norm": 0.34375, + "learning_rate": 0.0012733768560942394, + "loss": 0.127, + "step": 42639 + }, + { + "epoch": 0.3701356759055911, + "grad_norm": 0.3125, + "learning_rate": 0.0012733476510139972, + "loss": 0.1084, + "step": 42640 + }, + { + "epoch": 0.37014435638579524, + "grad_norm": 0.5078125, + "learning_rate": 0.0012733184457441847, + "loss": 0.1406, + "step": 42641 + }, + { + "epoch": 0.37015303686599943, + "grad_norm": 0.0869140625, + "learning_rate": 0.0012732892402848338, + "loss": 0.0796, + "step": 42642 + }, + { + "epoch": 0.37016171734620357, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012732600346359766, + "loss": 0.0918, + "step": 42643 + }, + { + "epoch": 0.37017039782640776, + "grad_norm": 0.6015625, + "learning_rate": 0.0012732308287976446, + "loss": 0.1235, + "step": 42644 + }, + { + "epoch": 0.3701790783066119, + "grad_norm": 0.166015625, + "learning_rate": 0.0012732016227698702, + "loss": 0.1069, + "step": 42645 + }, + { + "epoch": 0.3701877587868161, + "grad_norm": 0.78125, + "learning_rate": 0.001273172416552685, + "loss": 0.0918, + "step": 42646 + }, + { + "epoch": 0.3701964392670202, + "grad_norm": 0.58203125, + "learning_rate": 0.001273143210146121, + "loss": 0.0703, + "step": 42647 + }, + { + "epoch": 0.3702051197472244, + "grad_norm": 0.337890625, + "learning_rate": 0.0012731140035502101, + "loss": 0.0693, + "step": 42648 + }, + { + "epoch": 0.37021380022742856, + "grad_norm": 0.34375, + "learning_rate": 0.0012730847967649846, + "loss": 0.0791, + "step": 42649 + }, + { + "epoch": 0.37022248070763275, + "grad_norm": 0.4921875, + "learning_rate": 0.001273055589790476, + "loss": 0.0845, + "step": 42650 + }, + { + "epoch": 0.3702311611878369, + "grad_norm": 0.1943359375, + "learning_rate": 0.0012730263826267164, + "loss": 0.0864, + "step": 42651 + }, + { + "epoch": 0.3702398416680411, + "grad_norm": 0.0830078125, + "learning_rate": 0.0012729971752737377, + "loss": 0.0864, + "step": 42652 + }, + { + "epoch": 0.3702485221482452, + "grad_norm": 0.0927734375, + "learning_rate": 0.001272967967731572, + "loss": 0.0957, + "step": 42653 + }, + { + "epoch": 0.3702572026284494, + "grad_norm": 0.447265625, + "learning_rate": 0.001272938760000251, + "loss": 0.1738, + "step": 42654 + }, + { + "epoch": 0.37026588310865355, + "grad_norm": 0.30078125, + "learning_rate": 0.0012729095520798068, + "loss": 0.0957, + "step": 42655 + }, + { + "epoch": 0.37027456358885774, + "grad_norm": 0.12158203125, + "learning_rate": 0.0012728803439702716, + "loss": 0.0732, + "step": 42656 + }, + { + "epoch": 0.3702832440690619, + "grad_norm": 0.0859375, + "learning_rate": 0.0012728511356716766, + "loss": 0.1104, + "step": 42657 + }, + { + "epoch": 0.37029192454926607, + "grad_norm": 0.1630859375, + "learning_rate": 0.0012728219271840545, + "loss": 0.126, + "step": 42658 + }, + { + "epoch": 0.3703006050294702, + "grad_norm": 0.337890625, + "learning_rate": 0.0012727927185074367, + "loss": 0.0952, + "step": 42659 + }, + { + "epoch": 0.3703092855096744, + "grad_norm": 0.49609375, + "learning_rate": 0.0012727635096418557, + "loss": 0.0854, + "step": 42660 + }, + { + "epoch": 0.37031796598987854, + "grad_norm": 0.177734375, + "learning_rate": 0.0012727343005873428, + "loss": 0.1006, + "step": 42661 + }, + { + "epoch": 0.37032664647008273, + "grad_norm": 0.267578125, + "learning_rate": 0.0012727050913439305, + "loss": 0.0898, + "step": 42662 + }, + { + "epoch": 0.37033532695028687, + "grad_norm": 1.796875, + "learning_rate": 0.0012726758819116505, + "loss": 0.3652, + "step": 42663 + }, + { + "epoch": 0.37034400743049106, + "grad_norm": 0.057373046875, + "learning_rate": 0.0012726466722905345, + "loss": 0.0654, + "step": 42664 + }, + { + "epoch": 0.3703526879106952, + "grad_norm": 0.1669921875, + "learning_rate": 0.001272617462480615, + "loss": 0.0674, + "step": 42665 + }, + { + "epoch": 0.3703613683908994, + "grad_norm": 0.263671875, + "learning_rate": 0.0012725882524819235, + "loss": 0.104, + "step": 42666 + }, + { + "epoch": 0.37037004887110353, + "grad_norm": 0.51171875, + "learning_rate": 0.001272559042294492, + "loss": 0.1201, + "step": 42667 + }, + { + "epoch": 0.3703787293513077, + "grad_norm": 0.51953125, + "learning_rate": 0.0012725298319183527, + "loss": 0.0762, + "step": 42668 + }, + { + "epoch": 0.37038740983151186, + "grad_norm": 0.0849609375, + "learning_rate": 0.0012725006213535372, + "loss": 0.125, + "step": 42669 + }, + { + "epoch": 0.37039609031171605, + "grad_norm": 0.14453125, + "learning_rate": 0.0012724714106000777, + "loss": 0.0991, + "step": 42670 + }, + { + "epoch": 0.3704047707919202, + "grad_norm": 0.125, + "learning_rate": 0.001272442199658006, + "loss": 0.1553, + "step": 42671 + }, + { + "epoch": 0.3704134512721244, + "grad_norm": 0.13671875, + "learning_rate": 0.001272412988527354, + "loss": 0.1123, + "step": 42672 + }, + { + "epoch": 0.3704221317523285, + "grad_norm": 0.3125, + "learning_rate": 0.0012723837772081541, + "loss": 0.0952, + "step": 42673 + }, + { + "epoch": 0.3704308122325327, + "grad_norm": 0.416015625, + "learning_rate": 0.0012723545657004375, + "loss": 0.0908, + "step": 42674 + }, + { + "epoch": 0.37043949271273685, + "grad_norm": 0.083984375, + "learning_rate": 0.0012723253540042367, + "loss": 0.0732, + "step": 42675 + }, + { + "epoch": 0.37044817319294104, + "grad_norm": 0.08642578125, + "learning_rate": 0.0012722961421195835, + "loss": 0.084, + "step": 42676 + }, + { + "epoch": 0.3704568536731452, + "grad_norm": 0.6328125, + "learning_rate": 0.0012722669300465103, + "loss": 0.127, + "step": 42677 + }, + { + "epoch": 0.3704655341533494, + "grad_norm": 0.1142578125, + "learning_rate": 0.0012722377177850476, + "loss": 0.1094, + "step": 42678 + }, + { + "epoch": 0.3704742146335535, + "grad_norm": 0.07568359375, + "learning_rate": 0.001272208505335229, + "loss": 0.0874, + "step": 42679 + }, + { + "epoch": 0.3704828951137577, + "grad_norm": 1.046875, + "learning_rate": 0.0012721792926970853, + "loss": 0.0908, + "step": 42680 + }, + { + "epoch": 0.37049157559396184, + "grad_norm": 0.30078125, + "learning_rate": 0.0012721500798706493, + "loss": 0.0811, + "step": 42681 + }, + { + "epoch": 0.37050025607416603, + "grad_norm": 0.162109375, + "learning_rate": 0.0012721208668559525, + "loss": 0.1172, + "step": 42682 + }, + { + "epoch": 0.37050893655437017, + "grad_norm": 2.75, + "learning_rate": 0.0012720916536530268, + "loss": 0.2559, + "step": 42683 + }, + { + "epoch": 0.37051761703457436, + "grad_norm": 0.166015625, + "learning_rate": 0.0012720624402619046, + "loss": 0.126, + "step": 42684 + }, + { + "epoch": 0.3705262975147785, + "grad_norm": 0.2734375, + "learning_rate": 0.001272033226682617, + "loss": 0.1025, + "step": 42685 + }, + { + "epoch": 0.3705349779949827, + "grad_norm": 0.431640625, + "learning_rate": 0.0012720040129151969, + "loss": 0.1113, + "step": 42686 + }, + { + "epoch": 0.37054365847518683, + "grad_norm": 0.64453125, + "learning_rate": 0.0012719747989596753, + "loss": 0.085, + "step": 42687 + }, + { + "epoch": 0.370552338955391, + "grad_norm": 0.2412109375, + "learning_rate": 0.0012719455848160851, + "loss": 0.0815, + "step": 42688 + }, + { + "epoch": 0.37056101943559516, + "grad_norm": 0.263671875, + "learning_rate": 0.0012719163704844572, + "loss": 0.1338, + "step": 42689 + }, + { + "epoch": 0.37056969991579936, + "grad_norm": 0.193359375, + "learning_rate": 0.0012718871559648248, + "loss": 0.1172, + "step": 42690 + }, + { + "epoch": 0.3705783803960035, + "grad_norm": 0.62890625, + "learning_rate": 0.0012718579412572186, + "loss": 0.1211, + "step": 42691 + }, + { + "epoch": 0.3705870608762077, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012718287263616718, + "loss": 0.0869, + "step": 42692 + }, + { + "epoch": 0.3705957413564118, + "grad_norm": 0.2470703125, + "learning_rate": 0.0012717995112782152, + "loss": 0.1143, + "step": 42693 + }, + { + "epoch": 0.370604421836616, + "grad_norm": 0.2578125, + "learning_rate": 0.0012717702960068813, + "loss": 0.1074, + "step": 42694 + }, + { + "epoch": 0.37061310231682015, + "grad_norm": 0.2412109375, + "learning_rate": 0.0012717410805477022, + "loss": 0.0674, + "step": 42695 + }, + { + "epoch": 0.37062178279702435, + "grad_norm": 0.296875, + "learning_rate": 0.0012717118649007095, + "loss": 0.0918, + "step": 42696 + }, + { + "epoch": 0.3706304632772285, + "grad_norm": 0.07666015625, + "learning_rate": 0.0012716826490659354, + "loss": 0.0771, + "step": 42697 + }, + { + "epoch": 0.3706391437574327, + "grad_norm": 0.19140625, + "learning_rate": 0.0012716534330434112, + "loss": 0.1113, + "step": 42698 + }, + { + "epoch": 0.3706478242376368, + "grad_norm": 0.283203125, + "learning_rate": 0.00127162421683317, + "loss": 0.0747, + "step": 42699 + }, + { + "epoch": 0.370656504717841, + "grad_norm": 1.5859375, + "learning_rate": 0.001271595000435243, + "loss": 0.1582, + "step": 42700 + }, + { + "epoch": 0.37066518519804514, + "grad_norm": 0.1591796875, + "learning_rate": 0.0012715657838496622, + "loss": 0.1162, + "step": 42701 + }, + { + "epoch": 0.37067386567824934, + "grad_norm": 0.09130859375, + "learning_rate": 0.0012715365670764596, + "loss": 0.0938, + "step": 42702 + }, + { + "epoch": 0.3706825461584535, + "grad_norm": 0.1884765625, + "learning_rate": 0.0012715073501156674, + "loss": 0.0923, + "step": 42703 + }, + { + "epoch": 0.37069122663865767, + "grad_norm": 1.6015625, + "learning_rate": 0.001271478132967317, + "loss": 0.2852, + "step": 42704 + }, + { + "epoch": 0.3706999071188618, + "grad_norm": 0.255859375, + "learning_rate": 0.001271448915631441, + "loss": 0.0918, + "step": 42705 + }, + { + "epoch": 0.370708587599066, + "grad_norm": 0.08203125, + "learning_rate": 0.0012714196981080711, + "loss": 0.0952, + "step": 42706 + }, + { + "epoch": 0.37071726807927013, + "grad_norm": 0.267578125, + "learning_rate": 0.001271390480397239, + "loss": 0.1211, + "step": 42707 + }, + { + "epoch": 0.3707259485594743, + "grad_norm": 0.353515625, + "learning_rate": 0.0012713612624989767, + "loss": 0.0898, + "step": 42708 + }, + { + "epoch": 0.37073462903967846, + "grad_norm": 0.1181640625, + "learning_rate": 0.0012713320444133164, + "loss": 0.0894, + "step": 42709 + }, + { + "epoch": 0.37074330951988266, + "grad_norm": 0.1337890625, + "learning_rate": 0.00127130282614029, + "loss": 0.1011, + "step": 42710 + }, + { + "epoch": 0.3707519900000868, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012712736076799297, + "loss": 0.0933, + "step": 42711 + }, + { + "epoch": 0.370760670480291, + "grad_norm": 0.130859375, + "learning_rate": 0.0012712443890322666, + "loss": 0.1064, + "step": 42712 + }, + { + "epoch": 0.3707693509604951, + "grad_norm": 0.2119140625, + "learning_rate": 0.0012712151701973332, + "loss": 0.0991, + "step": 42713 + }, + { + "epoch": 0.3707780314406993, + "grad_norm": 0.291015625, + "learning_rate": 0.001271185951175162, + "loss": 0.0928, + "step": 42714 + }, + { + "epoch": 0.37078671192090346, + "grad_norm": 0.234375, + "learning_rate": 0.0012711567319657845, + "loss": 0.1035, + "step": 42715 + }, + { + "epoch": 0.37079539240110765, + "grad_norm": 0.2197265625, + "learning_rate": 0.001271127512569232, + "loss": 0.104, + "step": 42716 + }, + { + "epoch": 0.3708040728813118, + "grad_norm": 0.07177734375, + "learning_rate": 0.0012710982929855373, + "loss": 0.0815, + "step": 42717 + }, + { + "epoch": 0.370812753361516, + "grad_norm": 0.2109375, + "learning_rate": 0.001271069073214732, + "loss": 0.1118, + "step": 42718 + }, + { + "epoch": 0.3708214338417201, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012710398532568478, + "loss": 0.0713, + "step": 42719 + }, + { + "epoch": 0.3708301143219243, + "grad_norm": 0.2275390625, + "learning_rate": 0.0012710106331119179, + "loss": 0.1055, + "step": 42720 + }, + { + "epoch": 0.37083879480212845, + "grad_norm": 0.30078125, + "learning_rate": 0.0012709814127799725, + "loss": 0.0903, + "step": 42721 + }, + { + "epoch": 0.37084747528233264, + "grad_norm": 0.130859375, + "learning_rate": 0.0012709521922610446, + "loss": 0.1035, + "step": 42722 + }, + { + "epoch": 0.3708561557625368, + "grad_norm": 0.6875, + "learning_rate": 0.0012709229715551662, + "loss": 0.125, + "step": 42723 + }, + { + "epoch": 0.37086483624274097, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012708937506623688, + "loss": 0.0947, + "step": 42724 + }, + { + "epoch": 0.3708735167229451, + "grad_norm": 0.359375, + "learning_rate": 0.0012708645295826848, + "loss": 0.0718, + "step": 42725 + }, + { + "epoch": 0.3708821972031493, + "grad_norm": 0.1650390625, + "learning_rate": 0.0012708353083161455, + "loss": 0.1123, + "step": 42726 + }, + { + "epoch": 0.37089087768335344, + "grad_norm": 0.54296875, + "learning_rate": 0.0012708060868627834, + "loss": 0.085, + "step": 42727 + }, + { + "epoch": 0.37089955816355763, + "grad_norm": 0.546875, + "learning_rate": 0.0012707768652226305, + "loss": 0.0981, + "step": 42728 + }, + { + "epoch": 0.37090823864376177, + "grad_norm": 0.18359375, + "learning_rate": 0.0012707476433957187, + "loss": 0.1084, + "step": 42729 + }, + { + "epoch": 0.37091691912396596, + "grad_norm": 0.322265625, + "learning_rate": 0.0012707184213820793, + "loss": 0.1021, + "step": 42730 + }, + { + "epoch": 0.3709255996041701, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012706891991817453, + "loss": 0.126, + "step": 42731 + }, + { + "epoch": 0.3709342800843743, + "grad_norm": 0.11181640625, + "learning_rate": 0.0012706599767947479, + "loss": 0.0679, + "step": 42732 + }, + { + "epoch": 0.37094296056457843, + "grad_norm": 0.06640625, + "learning_rate": 0.0012706307542211196, + "loss": 0.0718, + "step": 42733 + }, + { + "epoch": 0.3709516410447826, + "grad_norm": 0.205078125, + "learning_rate": 0.001270601531460892, + "loss": 0.0898, + "step": 42734 + }, + { + "epoch": 0.37096032152498676, + "grad_norm": 0.322265625, + "learning_rate": 0.0012705723085140968, + "loss": 0.1289, + "step": 42735 + }, + { + "epoch": 0.37096900200519095, + "grad_norm": 0.08056640625, + "learning_rate": 0.0012705430853807668, + "loss": 0.0938, + "step": 42736 + }, + { + "epoch": 0.3709776824853951, + "grad_norm": 1.015625, + "learning_rate": 0.0012705138620609329, + "loss": 0.1299, + "step": 42737 + }, + { + "epoch": 0.3709863629655993, + "grad_norm": 0.146484375, + "learning_rate": 0.001270484638554628, + "loss": 0.1123, + "step": 42738 + }, + { + "epoch": 0.3709950434458034, + "grad_norm": 0.1435546875, + "learning_rate": 0.001270455414861883, + "loss": 0.1206, + "step": 42739 + }, + { + "epoch": 0.3710037239260076, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012704261909827312, + "loss": 0.0811, + "step": 42740 + }, + { + "epoch": 0.37101240440621175, + "grad_norm": 0.185546875, + "learning_rate": 0.0012703969669172039, + "loss": 0.0977, + "step": 42741 + }, + { + "epoch": 0.37102108488641594, + "grad_norm": 0.298828125, + "learning_rate": 0.0012703677426653328, + "loss": 0.0933, + "step": 42742 + }, + { + "epoch": 0.3710297653666201, + "grad_norm": 0.10546875, + "learning_rate": 0.00127033851822715, + "loss": 0.0781, + "step": 42743 + }, + { + "epoch": 0.37103844584682427, + "grad_norm": 0.30859375, + "learning_rate": 0.0012703092936026878, + "loss": 0.1123, + "step": 42744 + }, + { + "epoch": 0.3710471263270284, + "grad_norm": 0.1787109375, + "learning_rate": 0.001270280068791978, + "loss": 0.1387, + "step": 42745 + }, + { + "epoch": 0.3710558068072326, + "grad_norm": 0.115234375, + "learning_rate": 0.0012702508437950522, + "loss": 0.0996, + "step": 42746 + }, + { + "epoch": 0.37106448728743674, + "grad_norm": 0.3203125, + "learning_rate": 0.0012702216186119427, + "loss": 0.0767, + "step": 42747 + }, + { + "epoch": 0.37107316776764093, + "grad_norm": 0.337890625, + "learning_rate": 0.0012701923932426813, + "loss": 0.0752, + "step": 42748 + }, + { + "epoch": 0.37108184824784507, + "grad_norm": 0.28125, + "learning_rate": 0.0012701631676873002, + "loss": 0.0864, + "step": 42749 + }, + { + "epoch": 0.37109052872804926, + "grad_norm": 0.267578125, + "learning_rate": 0.0012701339419458313, + "loss": 0.1572, + "step": 42750 + }, + { + "epoch": 0.3710992092082534, + "grad_norm": 0.6640625, + "learning_rate": 0.0012701047160183063, + "loss": 0.1104, + "step": 42751 + }, + { + "epoch": 0.3711078896884576, + "grad_norm": 0.1279296875, + "learning_rate": 0.0012700754899047575, + "loss": 0.1221, + "step": 42752 + }, + { + "epoch": 0.37111657016866173, + "grad_norm": 0.34765625, + "learning_rate": 0.0012700462636052166, + "loss": 0.0942, + "step": 42753 + }, + { + "epoch": 0.37112525064886587, + "grad_norm": 0.330078125, + "learning_rate": 0.0012700170371197157, + "loss": 0.0898, + "step": 42754 + }, + { + "epoch": 0.37113393112907006, + "grad_norm": 0.078125, + "learning_rate": 0.0012699878104482868, + "loss": 0.0815, + "step": 42755 + }, + { + "epoch": 0.3711426116092742, + "grad_norm": 0.1796875, + "learning_rate": 0.0012699585835909614, + "loss": 0.0825, + "step": 42756 + }, + { + "epoch": 0.3711512920894784, + "grad_norm": 0.173828125, + "learning_rate": 0.0012699293565477721, + "loss": 0.082, + "step": 42757 + }, + { + "epoch": 0.37115997256968253, + "grad_norm": 0.373046875, + "learning_rate": 0.0012699001293187506, + "loss": 0.1074, + "step": 42758 + }, + { + "epoch": 0.3711686530498867, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012698709019039291, + "loss": 0.1406, + "step": 42759 + }, + { + "epoch": 0.37117733353009086, + "grad_norm": 0.265625, + "learning_rate": 0.0012698416743033391, + "loss": 0.0811, + "step": 42760 + }, + { + "epoch": 0.37118601401029505, + "grad_norm": 0.39453125, + "learning_rate": 0.0012698124465170128, + "loss": 0.0864, + "step": 42761 + }, + { + "epoch": 0.3711946944904992, + "grad_norm": 0.376953125, + "learning_rate": 0.0012697832185449823, + "loss": 0.1084, + "step": 42762 + }, + { + "epoch": 0.3712033749707034, + "grad_norm": 0.15625, + "learning_rate": 0.0012697539903872795, + "loss": 0.1113, + "step": 42763 + }, + { + "epoch": 0.3712120554509075, + "grad_norm": 0.59765625, + "learning_rate": 0.001269724762043936, + "loss": 0.1245, + "step": 42764 + }, + { + "epoch": 0.3712207359311117, + "grad_norm": 0.5546875, + "learning_rate": 0.001269695533514984, + "loss": 0.1152, + "step": 42765 + }, + { + "epoch": 0.37122941641131585, + "grad_norm": 0.337890625, + "learning_rate": 0.0012696663048004555, + "loss": 0.0928, + "step": 42766 + }, + { + "epoch": 0.37123809689152004, + "grad_norm": 0.1826171875, + "learning_rate": 0.0012696370759003829, + "loss": 0.0815, + "step": 42767 + }, + { + "epoch": 0.3712467773717242, + "grad_norm": 0.86328125, + "learning_rate": 0.0012696078468147974, + "loss": 0.105, + "step": 42768 + }, + { + "epoch": 0.37125545785192837, + "grad_norm": 0.279296875, + "learning_rate": 0.0012695786175437312, + "loss": 0.1089, + "step": 42769 + }, + { + "epoch": 0.3712641383321325, + "grad_norm": 0.306640625, + "learning_rate": 0.0012695493880872167, + "loss": 0.1064, + "step": 42770 + }, + { + "epoch": 0.3712728188123367, + "grad_norm": 0.439453125, + "learning_rate": 0.0012695201584452856, + "loss": 0.0688, + "step": 42771 + }, + { + "epoch": 0.37128149929254084, + "grad_norm": 0.3671875, + "learning_rate": 0.0012694909286179697, + "loss": 0.0811, + "step": 42772 + }, + { + "epoch": 0.37129017977274503, + "grad_norm": 0.2373046875, + "learning_rate": 0.001269461698605301, + "loss": 0.1006, + "step": 42773 + }, + { + "epoch": 0.37129886025294917, + "grad_norm": 0.7265625, + "learning_rate": 0.0012694324684073112, + "loss": 0.166, + "step": 42774 + }, + { + "epoch": 0.37130754073315336, + "grad_norm": 0.65625, + "learning_rate": 0.0012694032380240329, + "loss": 0.127, + "step": 42775 + }, + { + "epoch": 0.3713162212133575, + "grad_norm": 0.73828125, + "learning_rate": 0.0012693740074554978, + "loss": 0.2246, + "step": 42776 + }, + { + "epoch": 0.3713249016935617, + "grad_norm": 0.80859375, + "learning_rate": 0.0012693447767017378, + "loss": 0.1504, + "step": 42777 + }, + { + "epoch": 0.37133358217376583, + "grad_norm": 0.58984375, + "learning_rate": 0.0012693155457627848, + "loss": 0.1221, + "step": 42778 + }, + { + "epoch": 0.37134226265397, + "grad_norm": 0.73828125, + "learning_rate": 0.0012692863146386711, + "loss": 0.0703, + "step": 42779 + }, + { + "epoch": 0.37135094313417416, + "grad_norm": 0.609375, + "learning_rate": 0.0012692570833294284, + "loss": 0.1338, + "step": 42780 + }, + { + "epoch": 0.37135962361437835, + "grad_norm": 0.11083984375, + "learning_rate": 0.0012692278518350883, + "loss": 0.1064, + "step": 42781 + }, + { + "epoch": 0.3713683040945825, + "grad_norm": 0.0966796875, + "learning_rate": 0.0012691986201556836, + "loss": 0.1064, + "step": 42782 + }, + { + "epoch": 0.3713769845747867, + "grad_norm": 0.140625, + "learning_rate": 0.0012691693882912456, + "loss": 0.0747, + "step": 42783 + }, + { + "epoch": 0.3713856650549908, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012691401562418066, + "loss": 0.1162, + "step": 42784 + }, + { + "epoch": 0.371394345535195, + "grad_norm": 0.62890625, + "learning_rate": 0.0012691109240073983, + "loss": 0.1133, + "step": 42785 + }, + { + "epoch": 0.37140302601539915, + "grad_norm": 0.173828125, + "learning_rate": 0.0012690816915880531, + "loss": 0.1099, + "step": 42786 + }, + { + "epoch": 0.37141170649560334, + "grad_norm": 0.50390625, + "learning_rate": 0.0012690524589838027, + "loss": 0.0996, + "step": 42787 + }, + { + "epoch": 0.3714203869758075, + "grad_norm": 0.470703125, + "learning_rate": 0.0012690232261946789, + "loss": 0.1367, + "step": 42788 + }, + { + "epoch": 0.3714290674560117, + "grad_norm": 0.11767578125, + "learning_rate": 0.001268993993220714, + "loss": 0.0986, + "step": 42789 + }, + { + "epoch": 0.3714377479362158, + "grad_norm": 0.1982421875, + "learning_rate": 0.0012689647600619396, + "loss": 0.0918, + "step": 42790 + }, + { + "epoch": 0.37144642841642, + "grad_norm": 0.5859375, + "learning_rate": 0.001268935526718388, + "loss": 0.1348, + "step": 42791 + }, + { + "epoch": 0.37145510889662414, + "grad_norm": 0.11083984375, + "learning_rate": 0.001268906293190091, + "loss": 0.0918, + "step": 42792 + }, + { + "epoch": 0.37146378937682833, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012688770594770807, + "loss": 0.0942, + "step": 42793 + }, + { + "epoch": 0.3714724698570325, + "grad_norm": 0.265625, + "learning_rate": 0.0012688478255793887, + "loss": 0.1172, + "step": 42794 + }, + { + "epoch": 0.37148115033723667, + "grad_norm": 0.326171875, + "learning_rate": 0.0012688185914970474, + "loss": 0.1504, + "step": 42795 + }, + { + "epoch": 0.3714898308174408, + "grad_norm": 0.251953125, + "learning_rate": 0.0012687893572300888, + "loss": 0.1162, + "step": 42796 + }, + { + "epoch": 0.371498511297645, + "grad_norm": 0.82421875, + "learning_rate": 0.0012687601227785446, + "loss": 0.1562, + "step": 42797 + }, + { + "epoch": 0.37150719177784913, + "grad_norm": 0.51171875, + "learning_rate": 0.001268730888142447, + "loss": 0.0698, + "step": 42798 + }, + { + "epoch": 0.3715158722580533, + "grad_norm": 0.12451171875, + "learning_rate": 0.0012687016533218275, + "loss": 0.1162, + "step": 42799 + }, + { + "epoch": 0.37152455273825746, + "grad_norm": 0.37109375, + "learning_rate": 0.001268672418316719, + "loss": 0.0693, + "step": 42800 + }, + { + "epoch": 0.37153323321846166, + "grad_norm": 0.28125, + "learning_rate": 0.0012686431831271524, + "loss": 0.1128, + "step": 42801 + }, + { + "epoch": 0.3715419136986658, + "grad_norm": 0.443359375, + "learning_rate": 0.0012686139477531603, + "loss": 0.1523, + "step": 42802 + }, + { + "epoch": 0.37155059417887, + "grad_norm": 0.17578125, + "learning_rate": 0.001268584712194774, + "loss": 0.1182, + "step": 42803 + }, + { + "epoch": 0.3715592746590741, + "grad_norm": 0.373046875, + "learning_rate": 0.001268555476452027, + "loss": 0.0898, + "step": 42804 + }, + { + "epoch": 0.3715679551392783, + "grad_norm": 0.5546875, + "learning_rate": 0.0012685262405249494, + "loss": 0.1318, + "step": 42805 + }, + { + "epoch": 0.37157663561948245, + "grad_norm": 3.15625, + "learning_rate": 0.0012684970044135743, + "loss": 0.2578, + "step": 42806 + }, + { + "epoch": 0.37158531609968665, + "grad_norm": 0.24609375, + "learning_rate": 0.0012684677681179335, + "loss": 0.0815, + "step": 42807 + }, + { + "epoch": 0.3715939965798908, + "grad_norm": 0.296875, + "learning_rate": 0.0012684385316380587, + "loss": 0.1504, + "step": 42808 + }, + { + "epoch": 0.371602677060095, + "grad_norm": 0.6953125, + "learning_rate": 0.0012684092949739824, + "loss": 0.0923, + "step": 42809 + }, + { + "epoch": 0.3716113575402991, + "grad_norm": 0.1787109375, + "learning_rate": 0.001268380058125736, + "loss": 0.1104, + "step": 42810 + }, + { + "epoch": 0.3716200380205033, + "grad_norm": 0.39453125, + "learning_rate": 0.0012683508210933516, + "loss": 0.1309, + "step": 42811 + }, + { + "epoch": 0.37162871850070744, + "grad_norm": 0.302734375, + "learning_rate": 0.0012683215838768613, + "loss": 0.0825, + "step": 42812 + }, + { + "epoch": 0.37163739898091164, + "grad_norm": 0.373046875, + "learning_rate": 0.0012682923464762972, + "loss": 0.1055, + "step": 42813 + }, + { + "epoch": 0.3716460794611158, + "grad_norm": 0.484375, + "learning_rate": 0.001268263108891691, + "loss": 0.1543, + "step": 42814 + }, + { + "epoch": 0.37165475994131997, + "grad_norm": 0.203125, + "learning_rate": 0.001268233871123075, + "loss": 0.0928, + "step": 42815 + }, + { + "epoch": 0.3716634404215241, + "grad_norm": 0.333984375, + "learning_rate": 0.0012682046331704807, + "loss": 0.1074, + "step": 42816 + }, + { + "epoch": 0.3716721209017283, + "grad_norm": 0.193359375, + "learning_rate": 0.0012681753950339403, + "loss": 0.0947, + "step": 42817 + }, + { + "epoch": 0.37168080138193244, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012681461567134859, + "loss": 0.1279, + "step": 42818 + }, + { + "epoch": 0.37168948186213663, + "grad_norm": 0.6484375, + "learning_rate": 0.0012681169182091497, + "loss": 0.0986, + "step": 42819 + }, + { + "epoch": 0.37169816234234077, + "grad_norm": 0.208984375, + "learning_rate": 0.0012680876795209634, + "loss": 0.0991, + "step": 42820 + }, + { + "epoch": 0.37170684282254496, + "grad_norm": 0.2421875, + "learning_rate": 0.0012680584406489582, + "loss": 0.0898, + "step": 42821 + }, + { + "epoch": 0.3717155233027491, + "grad_norm": 0.1728515625, + "learning_rate": 0.0012680292015931675, + "loss": 0.0752, + "step": 42822 + }, + { + "epoch": 0.3717242037829533, + "grad_norm": 0.1640625, + "learning_rate": 0.001267999962353622, + "loss": 0.0991, + "step": 42823 + }, + { + "epoch": 0.3717328842631574, + "grad_norm": 0.3671875, + "learning_rate": 0.001267970722930355, + "loss": 0.1016, + "step": 42824 + }, + { + "epoch": 0.3717415647433616, + "grad_norm": 0.271484375, + "learning_rate": 0.001267941483323397, + "loss": 0.0864, + "step": 42825 + }, + { + "epoch": 0.37175024522356576, + "grad_norm": 0.1728515625, + "learning_rate": 0.001267912243532781, + "loss": 0.0869, + "step": 42826 + }, + { + "epoch": 0.37175892570376995, + "grad_norm": 0.224609375, + "learning_rate": 0.001267883003558539, + "loss": 0.083, + "step": 42827 + }, + { + "epoch": 0.3717676061839741, + "grad_norm": 0.1220703125, + "learning_rate": 0.0012678537634007025, + "loss": 0.126, + "step": 42828 + }, + { + "epoch": 0.3717762866641783, + "grad_norm": 0.5703125, + "learning_rate": 0.0012678245230593034, + "loss": 0.1191, + "step": 42829 + }, + { + "epoch": 0.3717849671443824, + "grad_norm": 0.2451171875, + "learning_rate": 0.001267795282534374, + "loss": 0.0884, + "step": 42830 + }, + { + "epoch": 0.3717936476245866, + "grad_norm": 0.2119140625, + "learning_rate": 0.0012677660418259466, + "loss": 0.0776, + "step": 42831 + }, + { + "epoch": 0.37180232810479075, + "grad_norm": 0.453125, + "learning_rate": 0.0012677368009340524, + "loss": 0.0884, + "step": 42832 + }, + { + "epoch": 0.37181100858499494, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012677075598587237, + "loss": 0.0742, + "step": 42833 + }, + { + "epoch": 0.3718196890651991, + "grad_norm": 0.5859375, + "learning_rate": 0.0012676783185999927, + "loss": 0.1348, + "step": 42834 + }, + { + "epoch": 0.37182836954540327, + "grad_norm": 0.328125, + "learning_rate": 0.0012676490771578912, + "loss": 0.1836, + "step": 42835 + }, + { + "epoch": 0.3718370500256074, + "grad_norm": 0.287109375, + "learning_rate": 0.0012676198355324514, + "loss": 0.0918, + "step": 42836 + }, + { + "epoch": 0.3718457305058116, + "grad_norm": 0.1201171875, + "learning_rate": 0.0012675905937237048, + "loss": 0.0996, + "step": 42837 + }, + { + "epoch": 0.37185441098601574, + "grad_norm": 0.5390625, + "learning_rate": 0.0012675613517316834, + "loss": 0.0967, + "step": 42838 + }, + { + "epoch": 0.37186309146621993, + "grad_norm": 0.08642578125, + "learning_rate": 0.00126753210955642, + "loss": 0.0776, + "step": 42839 + }, + { + "epoch": 0.37187177194642407, + "grad_norm": 0.388671875, + "learning_rate": 0.0012675028671979456, + "loss": 0.083, + "step": 42840 + }, + { + "epoch": 0.37188045242662826, + "grad_norm": 0.72265625, + "learning_rate": 0.0012674736246562927, + "loss": 0.0933, + "step": 42841 + }, + { + "epoch": 0.3718891329068324, + "grad_norm": 0.326171875, + "learning_rate": 0.0012674443819314931, + "loss": 0.1377, + "step": 42842 + }, + { + "epoch": 0.3718978133870366, + "grad_norm": 0.287109375, + "learning_rate": 0.0012674151390235787, + "loss": 0.0742, + "step": 42843 + }, + { + "epoch": 0.37190649386724073, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012673858959325818, + "loss": 0.0972, + "step": 42844 + }, + { + "epoch": 0.3719151743474449, + "grad_norm": 0.134765625, + "learning_rate": 0.0012673566526585345, + "loss": 0.0967, + "step": 42845 + }, + { + "epoch": 0.37192385482764906, + "grad_norm": 0.1875, + "learning_rate": 0.0012673274092014682, + "loss": 0.0874, + "step": 42846 + }, + { + "epoch": 0.37193253530785325, + "grad_norm": 0.1962890625, + "learning_rate": 0.001267298165561415, + "loss": 0.1245, + "step": 42847 + }, + { + "epoch": 0.3719412157880574, + "grad_norm": 0.427734375, + "learning_rate": 0.0012672689217384072, + "loss": 0.0757, + "step": 42848 + }, + { + "epoch": 0.3719498962682616, + "grad_norm": 0.6171875, + "learning_rate": 0.0012672396777324766, + "loss": 0.0967, + "step": 42849 + }, + { + "epoch": 0.3719585767484657, + "grad_norm": 0.26953125, + "learning_rate": 0.0012672104335436554, + "loss": 0.0728, + "step": 42850 + }, + { + "epoch": 0.3719672572286699, + "grad_norm": 0.2197265625, + "learning_rate": 0.0012671811891719752, + "loss": 0.123, + "step": 42851 + }, + { + "epoch": 0.37197593770887405, + "grad_norm": 0.451171875, + "learning_rate": 0.001267151944617468, + "loss": 0.1006, + "step": 42852 + }, + { + "epoch": 0.37198461818907824, + "grad_norm": 0.2392578125, + "learning_rate": 0.0012671226998801658, + "loss": 0.0693, + "step": 42853 + }, + { + "epoch": 0.3719932986692824, + "grad_norm": 0.4921875, + "learning_rate": 0.0012670934549601014, + "loss": 0.0884, + "step": 42854 + }, + { + "epoch": 0.3720019791494866, + "grad_norm": 0.20703125, + "learning_rate": 0.0012670642098573058, + "loss": 0.1387, + "step": 42855 + }, + { + "epoch": 0.3720106596296907, + "grad_norm": 0.77734375, + "learning_rate": 0.0012670349645718109, + "loss": 0.1182, + "step": 42856 + }, + { + "epoch": 0.3720193401098949, + "grad_norm": 0.30859375, + "learning_rate": 0.0012670057191036498, + "loss": 0.1016, + "step": 42857 + }, + { + "epoch": 0.37202802059009904, + "grad_norm": 0.31640625, + "learning_rate": 0.0012669764734528533, + "loss": 0.127, + "step": 42858 + }, + { + "epoch": 0.37203670107030323, + "grad_norm": 0.29296875, + "learning_rate": 0.0012669472276194537, + "loss": 0.084, + "step": 42859 + }, + { + "epoch": 0.37204538155050737, + "grad_norm": 2.328125, + "learning_rate": 0.0012669179816034832, + "loss": 0.1406, + "step": 42860 + }, + { + "epoch": 0.37205406203071156, + "grad_norm": 0.75, + "learning_rate": 0.0012668887354049738, + "loss": 0.0889, + "step": 42861 + }, + { + "epoch": 0.3720627425109157, + "grad_norm": 0.1181640625, + "learning_rate": 0.0012668594890239576, + "loss": 0.1025, + "step": 42862 + }, + { + "epoch": 0.3720714229911199, + "grad_norm": 0.1767578125, + "learning_rate": 0.001266830242460466, + "loss": 0.2637, + "step": 42863 + }, + { + "epoch": 0.37208010347132403, + "grad_norm": 0.11376953125, + "learning_rate": 0.0012668009957145314, + "loss": 0.1133, + "step": 42864 + }, + { + "epoch": 0.3720887839515282, + "grad_norm": 0.361328125, + "learning_rate": 0.0012667717487861861, + "loss": 0.1152, + "step": 42865 + }, + { + "epoch": 0.37209746443173236, + "grad_norm": 0.263671875, + "learning_rate": 0.0012667425016754615, + "loss": 0.0869, + "step": 42866 + }, + { + "epoch": 0.37210614491193655, + "grad_norm": 1.1640625, + "learning_rate": 0.0012667132543823898, + "loss": 0.1162, + "step": 42867 + }, + { + "epoch": 0.3721148253921407, + "grad_norm": 0.1396484375, + "learning_rate": 0.001266684006907003, + "loss": 0.1182, + "step": 42868 + }, + { + "epoch": 0.3721235058723449, + "grad_norm": 0.73828125, + "learning_rate": 0.0012666547592493332, + "loss": 0.0752, + "step": 42869 + }, + { + "epoch": 0.372132186352549, + "grad_norm": 0.2236328125, + "learning_rate": 0.001266625511409412, + "loss": 0.0811, + "step": 42870 + }, + { + "epoch": 0.3721408668327532, + "grad_norm": 0.154296875, + "learning_rate": 0.0012665962633872715, + "loss": 0.0977, + "step": 42871 + }, + { + "epoch": 0.37214954731295735, + "grad_norm": 0.1572265625, + "learning_rate": 0.0012665670151829442, + "loss": 0.082, + "step": 42872 + }, + { + "epoch": 0.37215822779316154, + "grad_norm": 0.63671875, + "learning_rate": 0.0012665377667964615, + "loss": 0.1079, + "step": 42873 + }, + { + "epoch": 0.3721669082733657, + "grad_norm": 0.57421875, + "learning_rate": 0.0012665085182278558, + "loss": 0.0879, + "step": 42874 + }, + { + "epoch": 0.3721755887535699, + "grad_norm": 1.0078125, + "learning_rate": 0.001266479269477159, + "loss": 0.1182, + "step": 42875 + }, + { + "epoch": 0.372184269233774, + "grad_norm": 0.41015625, + "learning_rate": 0.0012664500205444027, + "loss": 0.1201, + "step": 42876 + }, + { + "epoch": 0.37219294971397815, + "grad_norm": 0.466796875, + "learning_rate": 0.001266420771429619, + "loss": 0.084, + "step": 42877 + }, + { + "epoch": 0.37220163019418234, + "grad_norm": 0.12890625, + "learning_rate": 0.0012663915221328404, + "loss": 0.1025, + "step": 42878 + }, + { + "epoch": 0.3722103106743865, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012663622726540983, + "loss": 0.0889, + "step": 42879 + }, + { + "epoch": 0.3722189911545907, + "grad_norm": 0.4375, + "learning_rate": 0.001266333022993425, + "loss": 0.1094, + "step": 42880 + }, + { + "epoch": 0.3722276716347948, + "grad_norm": 0.177734375, + "learning_rate": 0.0012663037731508525, + "loss": 0.0859, + "step": 42881 + }, + { + "epoch": 0.372236352114999, + "grad_norm": 0.1484375, + "learning_rate": 0.0012662745231264124, + "loss": 0.1436, + "step": 42882 + }, + { + "epoch": 0.37224503259520314, + "grad_norm": 0.62890625, + "learning_rate": 0.0012662452729201373, + "loss": 0.1094, + "step": 42883 + }, + { + "epoch": 0.37225371307540733, + "grad_norm": 0.228515625, + "learning_rate": 0.0012662160225320588, + "loss": 0.0796, + "step": 42884 + }, + { + "epoch": 0.37226239355561147, + "grad_norm": 0.24609375, + "learning_rate": 0.0012661867719622089, + "loss": 0.0957, + "step": 42885 + }, + { + "epoch": 0.37227107403581566, + "grad_norm": 0.82421875, + "learning_rate": 0.0012661575212106198, + "loss": 0.1172, + "step": 42886 + }, + { + "epoch": 0.3722797545160198, + "grad_norm": 0.0927734375, + "learning_rate": 0.001266128270277323, + "loss": 0.0762, + "step": 42887 + }, + { + "epoch": 0.372288434996224, + "grad_norm": 0.10986328125, + "learning_rate": 0.001266099019162351, + "loss": 0.1045, + "step": 42888 + }, + { + "epoch": 0.37229711547642813, + "grad_norm": 0.099609375, + "learning_rate": 0.0012660697678657357, + "loss": 0.1172, + "step": 42889 + }, + { + "epoch": 0.3723057959566323, + "grad_norm": 0.2373046875, + "learning_rate": 0.0012660405163875089, + "loss": 0.0859, + "step": 42890 + }, + { + "epoch": 0.37231447643683646, + "grad_norm": 0.11279296875, + "learning_rate": 0.0012660112647277027, + "loss": 0.0752, + "step": 42891 + }, + { + "epoch": 0.37232315691704065, + "grad_norm": 0.421875, + "learning_rate": 0.001265982012886349, + "loss": 0.0903, + "step": 42892 + }, + { + "epoch": 0.3723318373972448, + "grad_norm": 0.92578125, + "learning_rate": 0.00126595276086348, + "loss": 0.2002, + "step": 42893 + }, + { + "epoch": 0.372340517877449, + "grad_norm": 0.10546875, + "learning_rate": 0.0012659235086591274, + "loss": 0.1279, + "step": 42894 + }, + { + "epoch": 0.3723491983576531, + "grad_norm": 0.416015625, + "learning_rate": 0.0012658942562733236, + "loss": 0.1934, + "step": 42895 + }, + { + "epoch": 0.3723578788378573, + "grad_norm": 0.291015625, + "learning_rate": 0.0012658650037061002, + "loss": 0.0918, + "step": 42896 + }, + { + "epoch": 0.37236655931806145, + "grad_norm": 0.2421875, + "learning_rate": 0.0012658357509574894, + "loss": 0.1064, + "step": 42897 + }, + { + "epoch": 0.37237523979826564, + "grad_norm": 0.2890625, + "learning_rate": 0.001265806498027523, + "loss": 0.1147, + "step": 42898 + }, + { + "epoch": 0.3723839202784698, + "grad_norm": 0.376953125, + "learning_rate": 0.0012657772449162327, + "loss": 0.1074, + "step": 42899 + }, + { + "epoch": 0.372392600758674, + "grad_norm": 0.42578125, + "learning_rate": 0.0012657479916236515, + "loss": 0.1533, + "step": 42900 + }, + { + "epoch": 0.3724012812388781, + "grad_norm": 0.482421875, + "learning_rate": 0.0012657187381498103, + "loss": 0.0996, + "step": 42901 + }, + { + "epoch": 0.3724099617190823, + "grad_norm": 0.1689453125, + "learning_rate": 0.001265689484494742, + "loss": 0.1006, + "step": 42902 + }, + { + "epoch": 0.37241864219928644, + "grad_norm": 0.0888671875, + "learning_rate": 0.001265660230658478, + "loss": 0.064, + "step": 42903 + }, + { + "epoch": 0.37242732267949064, + "grad_norm": 0.0986328125, + "learning_rate": 0.0012656309766410506, + "loss": 0.0947, + "step": 42904 + }, + { + "epoch": 0.3724360031596948, + "grad_norm": 0.146484375, + "learning_rate": 0.0012656017224424916, + "loss": 0.0757, + "step": 42905 + }, + { + "epoch": 0.37244468363989897, + "grad_norm": 0.439453125, + "learning_rate": 0.001265572468062833, + "loss": 0.1084, + "step": 42906 + }, + { + "epoch": 0.3724533641201031, + "grad_norm": 0.287109375, + "learning_rate": 0.0012655432135021069, + "loss": 0.1426, + "step": 42907 + }, + { + "epoch": 0.3724620446003073, + "grad_norm": 0.482421875, + "learning_rate": 0.001265513958760345, + "loss": 0.0938, + "step": 42908 + }, + { + "epoch": 0.37247072508051143, + "grad_norm": 0.388671875, + "learning_rate": 0.00126548470383758, + "loss": 0.1055, + "step": 42909 + }, + { + "epoch": 0.3724794055607156, + "grad_norm": 0.376953125, + "learning_rate": 0.001265455448733843, + "loss": 0.0947, + "step": 42910 + }, + { + "epoch": 0.37248808604091976, + "grad_norm": 0.220703125, + "learning_rate": 0.0012654261934491665, + "loss": 0.1074, + "step": 42911 + }, + { + "epoch": 0.37249676652112396, + "grad_norm": 0.291015625, + "learning_rate": 0.001265396937983582, + "loss": 0.1221, + "step": 42912 + }, + { + "epoch": 0.3725054470013281, + "grad_norm": 0.248046875, + "learning_rate": 0.0012653676823371227, + "loss": 0.1475, + "step": 42913 + }, + { + "epoch": 0.3725141274815323, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012653384265098194, + "loss": 0.0962, + "step": 42914 + }, + { + "epoch": 0.3725228079617364, + "grad_norm": 0.0966796875, + "learning_rate": 0.0012653091705017044, + "loss": 0.1367, + "step": 42915 + }, + { + "epoch": 0.3725314884419406, + "grad_norm": 0.119140625, + "learning_rate": 0.0012652799143128097, + "loss": 0.127, + "step": 42916 + }, + { + "epoch": 0.37254016892214475, + "grad_norm": 0.11279296875, + "learning_rate": 0.0012652506579431675, + "loss": 0.0674, + "step": 42917 + }, + { + "epoch": 0.37254884940234895, + "grad_norm": 0.1083984375, + "learning_rate": 0.0012652214013928098, + "loss": 0.1162, + "step": 42918 + }, + { + "epoch": 0.3725575298825531, + "grad_norm": 0.1357421875, + "learning_rate": 0.0012651921446617683, + "loss": 0.0933, + "step": 42919 + }, + { + "epoch": 0.3725662103627573, + "grad_norm": 0.55078125, + "learning_rate": 0.001265162887750075, + "loss": 0.1201, + "step": 42920 + }, + { + "epoch": 0.3725748908429614, + "grad_norm": 0.34375, + "learning_rate": 0.0012651336306577622, + "loss": 0.0918, + "step": 42921 + }, + { + "epoch": 0.3725835713231656, + "grad_norm": 0.294921875, + "learning_rate": 0.0012651043733848618, + "loss": 0.0825, + "step": 42922 + }, + { + "epoch": 0.37259225180336975, + "grad_norm": 0.1357421875, + "learning_rate": 0.001265075115931406, + "loss": 0.0913, + "step": 42923 + }, + { + "epoch": 0.37260093228357394, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012650458582974262, + "loss": 0.0781, + "step": 42924 + }, + { + "epoch": 0.3726096127637781, + "grad_norm": 0.39453125, + "learning_rate": 0.0012650166004829547, + "loss": 0.085, + "step": 42925 + }, + { + "epoch": 0.37261829324398227, + "grad_norm": 0.1572265625, + "learning_rate": 0.0012649873424880236, + "loss": 0.0986, + "step": 42926 + }, + { + "epoch": 0.3726269737241864, + "grad_norm": 0.201171875, + "learning_rate": 0.001264958084312665, + "loss": 0.0898, + "step": 42927 + }, + { + "epoch": 0.3726356542043906, + "grad_norm": 0.6328125, + "learning_rate": 0.0012649288259569105, + "loss": 0.085, + "step": 42928 + }, + { + "epoch": 0.37264433468459474, + "grad_norm": 0.4609375, + "learning_rate": 0.0012648995674207923, + "loss": 0.1279, + "step": 42929 + }, + { + "epoch": 0.37265301516479893, + "grad_norm": 0.4765625, + "learning_rate": 0.0012648703087043422, + "loss": 0.25, + "step": 42930 + }, + { + "epoch": 0.37266169564500307, + "grad_norm": 0.1298828125, + "learning_rate": 0.0012648410498075928, + "loss": 0.1074, + "step": 42931 + }, + { + "epoch": 0.37267037612520726, + "grad_norm": 0.138671875, + "learning_rate": 0.0012648117907305757, + "loss": 0.1172, + "step": 42932 + }, + { + "epoch": 0.3726790566054114, + "grad_norm": 0.189453125, + "learning_rate": 0.0012647825314733225, + "loss": 0.1152, + "step": 42933 + }, + { + "epoch": 0.3726877370856156, + "grad_norm": 0.203125, + "learning_rate": 0.001264753272035866, + "loss": 0.0967, + "step": 42934 + }, + { + "epoch": 0.3726964175658197, + "grad_norm": 0.11767578125, + "learning_rate": 0.0012647240124182377, + "loss": 0.0898, + "step": 42935 + }, + { + "epoch": 0.3727050980460239, + "grad_norm": 0.25390625, + "learning_rate": 0.0012646947526204697, + "loss": 0.0894, + "step": 42936 + }, + { + "epoch": 0.37271377852622806, + "grad_norm": 0.119140625, + "learning_rate": 0.0012646654926425941, + "loss": 0.0825, + "step": 42937 + }, + { + "epoch": 0.37272245900643225, + "grad_norm": 6.8125, + "learning_rate": 0.0012646362324846426, + "loss": 0.3359, + "step": 42938 + }, + { + "epoch": 0.3727311394866364, + "grad_norm": 0.353515625, + "learning_rate": 0.0012646069721466474, + "loss": 0.1001, + "step": 42939 + }, + { + "epoch": 0.3727398199668406, + "grad_norm": 0.134765625, + "learning_rate": 0.0012645777116286406, + "loss": 0.1138, + "step": 42940 + }, + { + "epoch": 0.3727485004470447, + "grad_norm": 0.11962890625, + "learning_rate": 0.0012645484509306542, + "loss": 0.127, + "step": 42941 + }, + { + "epoch": 0.3727571809272489, + "grad_norm": 0.123046875, + "learning_rate": 0.0012645191900527196, + "loss": 0.1245, + "step": 42942 + }, + { + "epoch": 0.37276586140745305, + "grad_norm": 0.333984375, + "learning_rate": 0.0012644899289948697, + "loss": 0.1182, + "step": 42943 + }, + { + "epoch": 0.37277454188765724, + "grad_norm": 0.26171875, + "learning_rate": 0.001264460667757136, + "loss": 0.1357, + "step": 42944 + }, + { + "epoch": 0.3727832223678614, + "grad_norm": 0.138671875, + "learning_rate": 0.0012644314063395507, + "loss": 0.0884, + "step": 42945 + }, + { + "epoch": 0.37279190284806557, + "grad_norm": 0.26953125, + "learning_rate": 0.0012644021447421453, + "loss": 0.1201, + "step": 42946 + }, + { + "epoch": 0.3728005833282697, + "grad_norm": 0.2080078125, + "learning_rate": 0.0012643728829649526, + "loss": 0.1094, + "step": 42947 + }, + { + "epoch": 0.3728092638084739, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012643436210080037, + "loss": 0.1377, + "step": 42948 + }, + { + "epoch": 0.37281794428867804, + "grad_norm": 0.2255859375, + "learning_rate": 0.0012643143588713314, + "loss": 0.0996, + "step": 42949 + }, + { + "epoch": 0.37282662476888223, + "grad_norm": 0.19140625, + "learning_rate": 0.0012642850965549676, + "loss": 0.0981, + "step": 42950 + }, + { + "epoch": 0.37283530524908637, + "grad_norm": 0.1962890625, + "learning_rate": 0.0012642558340589436, + "loss": 0.1133, + "step": 42951 + }, + { + "epoch": 0.37284398572929056, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012642265713832923, + "loss": 0.1011, + "step": 42952 + }, + { + "epoch": 0.3728526662094947, + "grad_norm": 0.2412109375, + "learning_rate": 0.0012641973085280454, + "loss": 0.0898, + "step": 42953 + }, + { + "epoch": 0.3728613466896989, + "grad_norm": 0.3125, + "learning_rate": 0.0012641680454932346, + "loss": 0.0771, + "step": 42954 + }, + { + "epoch": 0.37287002716990303, + "grad_norm": 0.271484375, + "learning_rate": 0.001264138782278892, + "loss": 0.0776, + "step": 42955 + }, + { + "epoch": 0.3728787076501072, + "grad_norm": 0.208984375, + "learning_rate": 0.0012641095188850498, + "loss": 0.0996, + "step": 42956 + }, + { + "epoch": 0.37288738813031136, + "grad_norm": 0.1953125, + "learning_rate": 0.0012640802553117397, + "loss": 0.0884, + "step": 42957 + }, + { + "epoch": 0.37289606861051555, + "grad_norm": 0.2197265625, + "learning_rate": 0.001264050991558994, + "loss": 0.1436, + "step": 42958 + }, + { + "epoch": 0.3729047490907197, + "grad_norm": 0.3515625, + "learning_rate": 0.0012640217276268443, + "loss": 0.1064, + "step": 42959 + }, + { + "epoch": 0.3729134295709239, + "grad_norm": 0.28515625, + "learning_rate": 0.0012639924635153234, + "loss": 0.1035, + "step": 42960 + }, + { + "epoch": 0.372922110051128, + "grad_norm": 0.234375, + "learning_rate": 0.0012639631992244628, + "loss": 0.0928, + "step": 42961 + }, + { + "epoch": 0.3729307905313322, + "grad_norm": 0.435546875, + "learning_rate": 0.0012639339347542942, + "loss": 0.0947, + "step": 42962 + }, + { + "epoch": 0.37293947101153635, + "grad_norm": 0.439453125, + "learning_rate": 0.0012639046701048502, + "loss": 0.1016, + "step": 42963 + }, + { + "epoch": 0.37294815149174054, + "grad_norm": 0.42578125, + "learning_rate": 0.0012638754052761623, + "loss": 0.1172, + "step": 42964 + }, + { + "epoch": 0.3729568319719447, + "grad_norm": 0.337890625, + "learning_rate": 0.0012638461402682628, + "loss": 0.0928, + "step": 42965 + }, + { + "epoch": 0.3729655124521489, + "grad_norm": 0.123046875, + "learning_rate": 0.0012638168750811836, + "loss": 0.1055, + "step": 42966 + }, + { + "epoch": 0.372974192932353, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012637876097149566, + "loss": 0.1123, + "step": 42967 + }, + { + "epoch": 0.3729828734125572, + "grad_norm": 0.55859375, + "learning_rate": 0.001263758344169614, + "loss": 0.0977, + "step": 42968 + }, + { + "epoch": 0.37299155389276134, + "grad_norm": 0.07470703125, + "learning_rate": 0.0012637290784451877, + "loss": 0.0713, + "step": 42969 + }, + { + "epoch": 0.37300023437296553, + "grad_norm": 0.33984375, + "learning_rate": 0.00126369981254171, + "loss": 0.0742, + "step": 42970 + }, + { + "epoch": 0.37300891485316967, + "grad_norm": 0.376953125, + "learning_rate": 0.0012636705464592127, + "loss": 0.0996, + "step": 42971 + }, + { + "epoch": 0.37301759533337386, + "grad_norm": 0.2294921875, + "learning_rate": 0.0012636412801977272, + "loss": 0.1367, + "step": 42972 + }, + { + "epoch": 0.373026275813578, + "grad_norm": 0.310546875, + "learning_rate": 0.0012636120137572867, + "loss": 0.1211, + "step": 42973 + }, + { + "epoch": 0.3730349562937822, + "grad_norm": 0.255859375, + "learning_rate": 0.0012635827471379221, + "loss": 0.1094, + "step": 42974 + }, + { + "epoch": 0.37304363677398633, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012635534803396662, + "loss": 0.0898, + "step": 42975 + }, + { + "epoch": 0.3730523172541905, + "grad_norm": 0.4140625, + "learning_rate": 0.0012635242133625503, + "loss": 0.0674, + "step": 42976 + }, + { + "epoch": 0.37306099773439466, + "grad_norm": 0.251953125, + "learning_rate": 0.0012634949462066068, + "loss": 0.1602, + "step": 42977 + }, + { + "epoch": 0.37306967821459885, + "grad_norm": 0.482421875, + "learning_rate": 0.0012634656788718679, + "loss": 0.1182, + "step": 42978 + }, + { + "epoch": 0.373078358694803, + "grad_norm": 0.11767578125, + "learning_rate": 0.0012634364113583657, + "loss": 0.0952, + "step": 42979 + }, + { + "epoch": 0.3730870391750072, + "grad_norm": 0.392578125, + "learning_rate": 0.0012634071436661313, + "loss": 0.1201, + "step": 42980 + }, + { + "epoch": 0.3730957196552113, + "grad_norm": 0.07763671875, + "learning_rate": 0.0012633778757951974, + "loss": 0.0654, + "step": 42981 + }, + { + "epoch": 0.3731044001354155, + "grad_norm": 1.25, + "learning_rate": 0.0012633486077455963, + "loss": 0.1934, + "step": 42982 + }, + { + "epoch": 0.37311308061561965, + "grad_norm": 0.455078125, + "learning_rate": 0.0012633193395173593, + "loss": 0.123, + "step": 42983 + }, + { + "epoch": 0.37312176109582385, + "grad_norm": 0.08837890625, + "learning_rate": 0.0012632900711105188, + "loss": 0.0801, + "step": 42984 + }, + { + "epoch": 0.373130441576028, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012632608025251067, + "loss": 0.1094, + "step": 42985 + }, + { + "epoch": 0.3731391220562322, + "grad_norm": 0.2265625, + "learning_rate": 0.0012632315337611551, + "loss": 0.0938, + "step": 42986 + }, + { + "epoch": 0.3731478025364363, + "grad_norm": 0.15625, + "learning_rate": 0.0012632022648186958, + "loss": 0.1123, + "step": 42987 + }, + { + "epoch": 0.3731564830166405, + "grad_norm": 0.083984375, + "learning_rate": 0.0012631729956977611, + "loss": 0.0664, + "step": 42988 + }, + { + "epoch": 0.37316516349684464, + "grad_norm": 0.1640625, + "learning_rate": 0.0012631437263983828, + "loss": 0.1074, + "step": 42989 + }, + { + "epoch": 0.37317384397704884, + "grad_norm": 0.115234375, + "learning_rate": 0.0012631144569205928, + "loss": 0.0898, + "step": 42990 + }, + { + "epoch": 0.373182524457253, + "grad_norm": 0.466796875, + "learning_rate": 0.0012630851872644235, + "loss": 0.1025, + "step": 42991 + }, + { + "epoch": 0.37319120493745717, + "grad_norm": 0.298828125, + "learning_rate": 0.001263055917429907, + "loss": 0.1074, + "step": 42992 + }, + { + "epoch": 0.3731998854176613, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012630266474170746, + "loss": 0.1138, + "step": 42993 + }, + { + "epoch": 0.3732085658978655, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012629973772259589, + "loss": 0.1025, + "step": 42994 + }, + { + "epoch": 0.37321724637806963, + "grad_norm": 0.265625, + "learning_rate": 0.0012629681068565914, + "loss": 0.1211, + "step": 42995 + }, + { + "epoch": 0.3732259268582738, + "grad_norm": 0.10791015625, + "learning_rate": 0.0012629388363090046, + "loss": 0.103, + "step": 42996 + }, + { + "epoch": 0.37323460733847796, + "grad_norm": 0.126953125, + "learning_rate": 0.0012629095655832304, + "loss": 0.0781, + "step": 42997 + }, + { + "epoch": 0.37324328781868216, + "grad_norm": 0.11279296875, + "learning_rate": 0.0012628802946793006, + "loss": 0.1221, + "step": 42998 + }, + { + "epoch": 0.3732519682988863, + "grad_norm": 0.23828125, + "learning_rate": 0.0012628510235972474, + "loss": 0.0806, + "step": 42999 + }, + { + "epoch": 0.3732606487790905, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012628217523371031, + "loss": 0.0801, + "step": 43000 + }, + { + "epoch": 0.3732693292592946, + "grad_norm": 0.087890625, + "learning_rate": 0.001262792480898899, + "loss": 0.1045, + "step": 43001 + }, + { + "epoch": 0.37327800973949876, + "grad_norm": 0.65234375, + "learning_rate": 0.0012627632092826674, + "loss": 0.1455, + "step": 43002 + }, + { + "epoch": 0.37328669021970295, + "grad_norm": 0.1240234375, + "learning_rate": 0.0012627339374884407, + "loss": 0.1006, + "step": 43003 + }, + { + "epoch": 0.3732953706999071, + "grad_norm": 0.314453125, + "learning_rate": 0.0012627046655162503, + "loss": 0.1143, + "step": 43004 + }, + { + "epoch": 0.3733040511801113, + "grad_norm": 0.57421875, + "learning_rate": 0.0012626753933661287, + "loss": 0.1216, + "step": 43005 + }, + { + "epoch": 0.3733127316603154, + "grad_norm": 0.1806640625, + "learning_rate": 0.001262646121038108, + "loss": 0.1094, + "step": 43006 + }, + { + "epoch": 0.3733214121405196, + "grad_norm": 0.2451171875, + "learning_rate": 0.0012626168485322196, + "loss": 0.0991, + "step": 43007 + }, + { + "epoch": 0.37333009262072375, + "grad_norm": 0.1630859375, + "learning_rate": 0.0012625875758484962, + "loss": 0.1338, + "step": 43008 + }, + { + "epoch": 0.37333877310092795, + "grad_norm": 0.1357421875, + "learning_rate": 0.001262558302986969, + "loss": 0.1309, + "step": 43009 + }, + { + "epoch": 0.3733474535811321, + "grad_norm": 0.6875, + "learning_rate": 0.0012625290299476707, + "loss": 0.0967, + "step": 43010 + }, + { + "epoch": 0.3733561340613363, + "grad_norm": 0.2138671875, + "learning_rate": 0.001262499756730633, + "loss": 0.1289, + "step": 43011 + }, + { + "epoch": 0.3733648145415404, + "grad_norm": 0.2119140625, + "learning_rate": 0.0012624704833358884, + "loss": 0.1064, + "step": 43012 + }, + { + "epoch": 0.3733734950217446, + "grad_norm": 0.08740234375, + "learning_rate": 0.0012624412097634683, + "loss": 0.0981, + "step": 43013 + }, + { + "epoch": 0.37338217550194874, + "grad_norm": 0.1220703125, + "learning_rate": 0.0012624119360134048, + "loss": 0.0767, + "step": 43014 + }, + { + "epoch": 0.37339085598215294, + "grad_norm": 0.498046875, + "learning_rate": 0.0012623826620857303, + "loss": 0.1094, + "step": 43015 + }, + { + "epoch": 0.3733995364623571, + "grad_norm": 0.349609375, + "learning_rate": 0.001262353387980476, + "loss": 0.1006, + "step": 43016 + }, + { + "epoch": 0.37340821694256127, + "grad_norm": 0.2333984375, + "learning_rate": 0.0012623241136976748, + "loss": 0.1348, + "step": 43017 + }, + { + "epoch": 0.3734168974227654, + "grad_norm": 0.14453125, + "learning_rate": 0.0012622948392373588, + "loss": 0.0854, + "step": 43018 + }, + { + "epoch": 0.3734255779029696, + "grad_norm": 0.68359375, + "learning_rate": 0.0012622655645995592, + "loss": 0.0762, + "step": 43019 + }, + { + "epoch": 0.37343425838317373, + "grad_norm": 0.330078125, + "learning_rate": 0.0012622362897843082, + "loss": 0.1084, + "step": 43020 + }, + { + "epoch": 0.3734429388633779, + "grad_norm": 0.158203125, + "learning_rate": 0.0012622070147916384, + "loss": 0.0923, + "step": 43021 + }, + { + "epoch": 0.37345161934358206, + "grad_norm": 0.451171875, + "learning_rate": 0.0012621777396215814, + "loss": 0.1211, + "step": 43022 + }, + { + "epoch": 0.37346029982378626, + "grad_norm": 0.38671875, + "learning_rate": 0.0012621484642741693, + "loss": 0.1543, + "step": 43023 + }, + { + "epoch": 0.3734689803039904, + "grad_norm": 0.52734375, + "learning_rate": 0.001262119188749434, + "loss": 0.0991, + "step": 43024 + }, + { + "epoch": 0.3734776607841946, + "grad_norm": 0.8046875, + "learning_rate": 0.0012620899130474073, + "loss": 0.0923, + "step": 43025 + }, + { + "epoch": 0.3734863412643987, + "grad_norm": 0.2451171875, + "learning_rate": 0.0012620606371681218, + "loss": 0.1113, + "step": 43026 + }, + { + "epoch": 0.3734950217446029, + "grad_norm": 0.134765625, + "learning_rate": 0.0012620313611116094, + "loss": 0.1055, + "step": 43027 + }, + { + "epoch": 0.37350370222480705, + "grad_norm": 0.1484375, + "learning_rate": 0.0012620020848779017, + "loss": 0.1089, + "step": 43028 + }, + { + "epoch": 0.37351238270501125, + "grad_norm": 0.1240234375, + "learning_rate": 0.001261972808467031, + "loss": 0.1016, + "step": 43029 + }, + { + "epoch": 0.3735210631852154, + "grad_norm": 0.65625, + "learning_rate": 0.001261943531879029, + "loss": 0.1016, + "step": 43030 + }, + { + "epoch": 0.3735297436654196, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012619142551139285, + "loss": 0.1157, + "step": 43031 + }, + { + "epoch": 0.3735384241456237, + "grad_norm": 0.63671875, + "learning_rate": 0.0012618849781717608, + "loss": 0.1426, + "step": 43032 + }, + { + "epoch": 0.3735471046258279, + "grad_norm": 0.318359375, + "learning_rate": 0.0012618557010525577, + "loss": 0.0781, + "step": 43033 + }, + { + "epoch": 0.37355578510603205, + "grad_norm": 0.23046875, + "learning_rate": 0.0012618264237563522, + "loss": 0.125, + "step": 43034 + }, + { + "epoch": 0.37356446558623624, + "grad_norm": 0.2314453125, + "learning_rate": 0.0012617971462831757, + "loss": 0.0874, + "step": 43035 + }, + { + "epoch": 0.3735731460664404, + "grad_norm": 0.1162109375, + "learning_rate": 0.0012617678686330598, + "loss": 0.1143, + "step": 43036 + }, + { + "epoch": 0.37358182654664457, + "grad_norm": 0.212890625, + "learning_rate": 0.0012617385908060372, + "loss": 0.0928, + "step": 43037 + }, + { + "epoch": 0.3735905070268487, + "grad_norm": 0.083984375, + "learning_rate": 0.00126170931280214, + "loss": 0.0977, + "step": 43038 + }, + { + "epoch": 0.3735991875070529, + "grad_norm": 0.357421875, + "learning_rate": 0.0012616800346214, + "loss": 0.1162, + "step": 43039 + }, + { + "epoch": 0.37360786798725704, + "grad_norm": 0.46875, + "learning_rate": 0.0012616507562638489, + "loss": 0.1172, + "step": 43040 + }, + { + "epoch": 0.37361654846746123, + "grad_norm": 0.85546875, + "learning_rate": 0.001261621477729519, + "loss": 0.1719, + "step": 43041 + }, + { + "epoch": 0.37362522894766537, + "grad_norm": 0.08935546875, + "learning_rate": 0.001261592199018442, + "loss": 0.0947, + "step": 43042 + }, + { + "epoch": 0.37363390942786956, + "grad_norm": 0.6640625, + "learning_rate": 0.0012615629201306506, + "loss": 0.1328, + "step": 43043 + }, + { + "epoch": 0.3736425899080737, + "grad_norm": 0.1669921875, + "learning_rate": 0.001261533641066176, + "loss": 0.1201, + "step": 43044 + }, + { + "epoch": 0.3736512703882779, + "grad_norm": 0.443359375, + "learning_rate": 0.001261504361825051, + "loss": 0.1133, + "step": 43045 + }, + { + "epoch": 0.373659950868482, + "grad_norm": 0.1650390625, + "learning_rate": 0.001261475082407307, + "loss": 0.0679, + "step": 43046 + }, + { + "epoch": 0.3736686313486862, + "grad_norm": 0.08203125, + "learning_rate": 0.0012614458028129766, + "loss": 0.1045, + "step": 43047 + }, + { + "epoch": 0.37367731182889036, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012614165230420912, + "loss": 0.1055, + "step": 43048 + }, + { + "epoch": 0.37368599230909455, + "grad_norm": 0.09375, + "learning_rate": 0.0012613872430946836, + "loss": 0.0806, + "step": 43049 + }, + { + "epoch": 0.3736946727892987, + "grad_norm": 0.64453125, + "learning_rate": 0.0012613579629707848, + "loss": 0.1055, + "step": 43050 + }, + { + "epoch": 0.3737033532695029, + "grad_norm": 0.1923828125, + "learning_rate": 0.0012613286826704275, + "loss": 0.1162, + "step": 43051 + }, + { + "epoch": 0.373712033749707, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012612994021936434, + "loss": 0.0991, + "step": 43052 + }, + { + "epoch": 0.3737207142299112, + "grad_norm": 0.130859375, + "learning_rate": 0.001261270121540465, + "loss": 0.0791, + "step": 43053 + }, + { + "epoch": 0.37372939471011535, + "grad_norm": 0.1943359375, + "learning_rate": 0.0012612408407109241, + "loss": 0.1143, + "step": 43054 + }, + { + "epoch": 0.37373807519031954, + "grad_norm": 0.5390625, + "learning_rate": 0.0012612115597050522, + "loss": 0.0757, + "step": 43055 + }, + { + "epoch": 0.3737467556705237, + "grad_norm": 0.189453125, + "learning_rate": 0.001261182278522882, + "loss": 0.0996, + "step": 43056 + }, + { + "epoch": 0.37375543615072787, + "grad_norm": 0.400390625, + "learning_rate": 0.0012611529971644454, + "loss": 0.0781, + "step": 43057 + }, + { + "epoch": 0.373764116630932, + "grad_norm": 0.0927734375, + "learning_rate": 0.0012611237156297743, + "loss": 0.1182, + "step": 43058 + }, + { + "epoch": 0.3737727971111362, + "grad_norm": 0.53125, + "learning_rate": 0.0012610944339189005, + "loss": 0.0996, + "step": 43059 + }, + { + "epoch": 0.37378147759134034, + "grad_norm": 0.6015625, + "learning_rate": 0.0012610651520318565, + "loss": 0.1045, + "step": 43060 + }, + { + "epoch": 0.37379015807154453, + "grad_norm": 0.26171875, + "learning_rate": 0.001261035869968674, + "loss": 0.0967, + "step": 43061 + }, + { + "epoch": 0.37379883855174867, + "grad_norm": 1.09375, + "learning_rate": 0.0012610065877293848, + "loss": 0.1221, + "step": 43062 + }, + { + "epoch": 0.37380751903195286, + "grad_norm": 0.310546875, + "learning_rate": 0.0012609773053140214, + "loss": 0.1064, + "step": 43063 + }, + { + "epoch": 0.373816199512157, + "grad_norm": 0.2490234375, + "learning_rate": 0.0012609480227226157, + "loss": 0.1221, + "step": 43064 + }, + { + "epoch": 0.3738248799923612, + "grad_norm": 0.4296875, + "learning_rate": 0.0012609187399551997, + "loss": 0.0854, + "step": 43065 + }, + { + "epoch": 0.37383356047256533, + "grad_norm": 0.10693359375, + "learning_rate": 0.0012608894570118054, + "loss": 0.1089, + "step": 43066 + }, + { + "epoch": 0.3738422409527695, + "grad_norm": 0.7265625, + "learning_rate": 0.0012608601738924648, + "loss": 0.1318, + "step": 43067 + }, + { + "epoch": 0.37385092143297366, + "grad_norm": 0.78515625, + "learning_rate": 0.0012608308905972098, + "loss": 0.1523, + "step": 43068 + }, + { + "epoch": 0.37385960191317785, + "grad_norm": 0.828125, + "learning_rate": 0.0012608016071260727, + "loss": 0.1387, + "step": 43069 + }, + { + "epoch": 0.373868282393382, + "grad_norm": 0.32421875, + "learning_rate": 0.0012607723234790854, + "loss": 0.1006, + "step": 43070 + }, + { + "epoch": 0.3738769628735862, + "grad_norm": 0.119140625, + "learning_rate": 0.0012607430396562797, + "loss": 0.1084, + "step": 43071 + }, + { + "epoch": 0.3738856433537903, + "grad_norm": 0.1884765625, + "learning_rate": 0.001260713755657688, + "loss": 0.0703, + "step": 43072 + }, + { + "epoch": 0.3738943238339945, + "grad_norm": 0.2216796875, + "learning_rate": 0.001260684471483342, + "loss": 0.1445, + "step": 43073 + }, + { + "epoch": 0.37390300431419865, + "grad_norm": 0.1728515625, + "learning_rate": 0.001260655187133274, + "loss": 0.0903, + "step": 43074 + }, + { + "epoch": 0.37391168479440284, + "grad_norm": 0.310546875, + "learning_rate": 0.0012606259026075159, + "loss": 0.127, + "step": 43075 + }, + { + "epoch": 0.373920365274607, + "grad_norm": 0.12109375, + "learning_rate": 0.0012605966179060997, + "loss": 0.1177, + "step": 43076 + }, + { + "epoch": 0.3739290457548112, + "grad_norm": 0.4765625, + "learning_rate": 0.0012605673330290572, + "loss": 0.0791, + "step": 43077 + }, + { + "epoch": 0.3739377262350153, + "grad_norm": 0.224609375, + "learning_rate": 0.0012605380479764213, + "loss": 0.1064, + "step": 43078 + }, + { + "epoch": 0.3739464067152195, + "grad_norm": 0.333984375, + "learning_rate": 0.0012605087627482231, + "loss": 0.1123, + "step": 43079 + }, + { + "epoch": 0.37395508719542364, + "grad_norm": 0.197265625, + "learning_rate": 0.0012604794773444946, + "loss": 0.0791, + "step": 43080 + }, + { + "epoch": 0.37396376767562783, + "grad_norm": 0.2392578125, + "learning_rate": 0.0012604501917652684, + "loss": 0.0967, + "step": 43081 + }, + { + "epoch": 0.37397244815583197, + "grad_norm": 0.3359375, + "learning_rate": 0.0012604209060105762, + "loss": 0.0957, + "step": 43082 + }, + { + "epoch": 0.37398112863603616, + "grad_norm": 0.81640625, + "learning_rate": 0.0012603916200804501, + "loss": 0.1328, + "step": 43083 + }, + { + "epoch": 0.3739898091162403, + "grad_norm": 0.06689453125, + "learning_rate": 0.0012603623339749225, + "loss": 0.1064, + "step": 43084 + }, + { + "epoch": 0.3739984895964445, + "grad_norm": 0.359375, + "learning_rate": 0.0012603330476940245, + "loss": 0.1309, + "step": 43085 + }, + { + "epoch": 0.37400717007664863, + "grad_norm": 0.271484375, + "learning_rate": 0.0012603037612377891, + "loss": 0.0771, + "step": 43086 + }, + { + "epoch": 0.3740158505568528, + "grad_norm": 0.28515625, + "learning_rate": 0.001260274474606248, + "loss": 0.0791, + "step": 43087 + }, + { + "epoch": 0.37402453103705696, + "grad_norm": 0.60546875, + "learning_rate": 0.0012602451877994328, + "loss": 0.1973, + "step": 43088 + }, + { + "epoch": 0.37403321151726115, + "grad_norm": 0.36328125, + "learning_rate": 0.0012602159008173758, + "loss": 0.1328, + "step": 43089 + }, + { + "epoch": 0.3740418919974653, + "grad_norm": 0.23046875, + "learning_rate": 0.0012601866136601097, + "loss": 0.1133, + "step": 43090 + }, + { + "epoch": 0.3740505724776695, + "grad_norm": 1.1953125, + "learning_rate": 0.0012601573263276652, + "loss": 0.1211, + "step": 43091 + }, + { + "epoch": 0.3740592529578736, + "grad_norm": 0.78125, + "learning_rate": 0.0012601280388200755, + "loss": 0.0923, + "step": 43092 + }, + { + "epoch": 0.3740679334380778, + "grad_norm": 0.1806640625, + "learning_rate": 0.001260098751137372, + "loss": 0.0723, + "step": 43093 + }, + { + "epoch": 0.37407661391828195, + "grad_norm": 0.1240234375, + "learning_rate": 0.001260069463279587, + "loss": 0.127, + "step": 43094 + }, + { + "epoch": 0.37408529439848615, + "grad_norm": 0.169921875, + "learning_rate": 0.0012600401752467525, + "loss": 0.1328, + "step": 43095 + }, + { + "epoch": 0.3740939748786903, + "grad_norm": 0.09765625, + "learning_rate": 0.0012600108870389004, + "loss": 0.085, + "step": 43096 + }, + { + "epoch": 0.3741026553588945, + "grad_norm": 0.08544921875, + "learning_rate": 0.0012599815986560627, + "loss": 0.0903, + "step": 43097 + }, + { + "epoch": 0.3741113358390986, + "grad_norm": 0.30859375, + "learning_rate": 0.0012599523100982714, + "loss": 0.1172, + "step": 43098 + }, + { + "epoch": 0.3741200163193028, + "grad_norm": 0.5234375, + "learning_rate": 0.0012599230213655593, + "loss": 0.1016, + "step": 43099 + }, + { + "epoch": 0.37412869679950694, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012598937324579571, + "loss": 0.125, + "step": 43100 + }, + { + "epoch": 0.37413737727971114, + "grad_norm": 0.0830078125, + "learning_rate": 0.0012598644433754979, + "loss": 0.1094, + "step": 43101 + }, + { + "epoch": 0.3741460577599153, + "grad_norm": 0.42578125, + "learning_rate": 0.001259835154118213, + "loss": 0.0972, + "step": 43102 + }, + { + "epoch": 0.37415473824011947, + "grad_norm": 0.84765625, + "learning_rate": 0.001259805864686135, + "loss": 0.1602, + "step": 43103 + }, + { + "epoch": 0.3741634187203236, + "grad_norm": 0.302734375, + "learning_rate": 0.0012597765750792958, + "loss": 0.0874, + "step": 43104 + }, + { + "epoch": 0.3741720992005278, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012597472852977274, + "loss": 0.1172, + "step": 43105 + }, + { + "epoch": 0.37418077968073193, + "grad_norm": 0.3984375, + "learning_rate": 0.0012597179953414614, + "loss": 0.1055, + "step": 43106 + }, + { + "epoch": 0.3741894601609361, + "grad_norm": 0.79296875, + "learning_rate": 0.0012596887052105305, + "loss": 0.124, + "step": 43107 + }, + { + "epoch": 0.37419814064114026, + "grad_norm": 0.171875, + "learning_rate": 0.0012596594149049665, + "loss": 0.0869, + "step": 43108 + }, + { + "epoch": 0.37420682112134446, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012596301244248013, + "loss": 0.1025, + "step": 43109 + }, + { + "epoch": 0.3742155016015486, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012596008337700667, + "loss": 0.1123, + "step": 43110 + }, + { + "epoch": 0.3742241820817528, + "grad_norm": 0.181640625, + "learning_rate": 0.0012595715429407955, + "loss": 0.0723, + "step": 43111 + }, + { + "epoch": 0.3742328625619569, + "grad_norm": 0.078125, + "learning_rate": 0.0012595422519370188, + "loss": 0.0781, + "step": 43112 + }, + { + "epoch": 0.3742415430421611, + "grad_norm": 0.10205078125, + "learning_rate": 0.0012595129607587692, + "loss": 0.0967, + "step": 43113 + }, + { + "epoch": 0.37425022352236526, + "grad_norm": 0.267578125, + "learning_rate": 0.0012594836694060788, + "loss": 0.0957, + "step": 43114 + }, + { + "epoch": 0.37425890400256945, + "grad_norm": 0.11474609375, + "learning_rate": 0.001259454377878979, + "loss": 0.126, + "step": 43115 + }, + { + "epoch": 0.3742675844827736, + "grad_norm": 0.33203125, + "learning_rate": 0.0012594250861775026, + "loss": 0.0801, + "step": 43116 + }, + { + "epoch": 0.3742762649629778, + "grad_norm": 0.2255859375, + "learning_rate": 0.0012593957943016817, + "loss": 0.0884, + "step": 43117 + }, + { + "epoch": 0.3742849454431819, + "grad_norm": 0.76171875, + "learning_rate": 0.0012593665022515477, + "loss": 0.0957, + "step": 43118 + }, + { + "epoch": 0.3742936259233861, + "grad_norm": 0.2265625, + "learning_rate": 0.0012593372100271325, + "loss": 0.1128, + "step": 43119 + }, + { + "epoch": 0.37430230640359025, + "grad_norm": 0.314453125, + "learning_rate": 0.001259307917628469, + "loss": 0.0747, + "step": 43120 + }, + { + "epoch": 0.37431098688379444, + "grad_norm": 0.423828125, + "learning_rate": 0.0012592786250555884, + "loss": 0.0942, + "step": 43121 + }, + { + "epoch": 0.3743196673639986, + "grad_norm": 0.271484375, + "learning_rate": 0.0012592493323085234, + "loss": 0.1152, + "step": 43122 + }, + { + "epoch": 0.37432834784420277, + "grad_norm": 0.1982421875, + "learning_rate": 0.0012592200393873056, + "loss": 0.1367, + "step": 43123 + }, + { + "epoch": 0.3743370283244069, + "grad_norm": 0.23046875, + "learning_rate": 0.0012591907462919671, + "loss": 0.1172, + "step": 43124 + }, + { + "epoch": 0.37434570880461104, + "grad_norm": 0.298828125, + "learning_rate": 0.00125916145302254, + "loss": 0.0962, + "step": 43125 + }, + { + "epoch": 0.37435438928481524, + "grad_norm": 0.93359375, + "learning_rate": 0.0012591321595790566, + "loss": 0.1357, + "step": 43126 + }, + { + "epoch": 0.3743630697650194, + "grad_norm": 0.64453125, + "learning_rate": 0.0012591028659615484, + "loss": 0.0986, + "step": 43127 + }, + { + "epoch": 0.37437175024522357, + "grad_norm": 0.212890625, + "learning_rate": 0.0012590735721700475, + "loss": 0.1484, + "step": 43128 + }, + { + "epoch": 0.3743804307254277, + "grad_norm": 0.177734375, + "learning_rate": 0.0012590442782045865, + "loss": 0.0938, + "step": 43129 + }, + { + "epoch": 0.3743891112056319, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012590149840651969, + "loss": 0.1006, + "step": 43130 + }, + { + "epoch": 0.37439779168583603, + "grad_norm": 0.306640625, + "learning_rate": 0.001258985689751911, + "loss": 0.0889, + "step": 43131 + }, + { + "epoch": 0.3744064721660402, + "grad_norm": 0.64453125, + "learning_rate": 0.0012589563952647605, + "loss": 0.1289, + "step": 43132 + }, + { + "epoch": 0.37441515264624436, + "grad_norm": 0.26171875, + "learning_rate": 0.0012589271006037779, + "loss": 0.1348, + "step": 43133 + }, + { + "epoch": 0.37442383312644856, + "grad_norm": 0.26953125, + "learning_rate": 0.001258897805768995, + "loss": 0.1172, + "step": 43134 + }, + { + "epoch": 0.3744325136066527, + "grad_norm": 0.09423828125, + "learning_rate": 0.001258868510760444, + "loss": 0.0879, + "step": 43135 + }, + { + "epoch": 0.3744411940868569, + "grad_norm": 0.5390625, + "learning_rate": 0.0012588392155781563, + "loss": 0.1299, + "step": 43136 + }, + { + "epoch": 0.374449874567061, + "grad_norm": 0.1005859375, + "learning_rate": 0.0012588099202221648, + "loss": 0.0908, + "step": 43137 + }, + { + "epoch": 0.3744585550472652, + "grad_norm": 0.412109375, + "learning_rate": 0.0012587806246925008, + "loss": 0.0791, + "step": 43138 + }, + { + "epoch": 0.37446723552746936, + "grad_norm": 0.1669921875, + "learning_rate": 0.0012587513289891974, + "loss": 0.0928, + "step": 43139 + }, + { + "epoch": 0.37447591600767355, + "grad_norm": 0.08642578125, + "learning_rate": 0.0012587220331122854, + "loss": 0.1001, + "step": 43140 + }, + { + "epoch": 0.3744845964878777, + "grad_norm": 0.26953125, + "learning_rate": 0.001258692737061797, + "loss": 0.0938, + "step": 43141 + }, + { + "epoch": 0.3744932769680819, + "grad_norm": 0.3203125, + "learning_rate": 0.0012586634408377653, + "loss": 0.0781, + "step": 43142 + }, + { + "epoch": 0.374501957448286, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012586341444402213, + "loss": 0.0854, + "step": 43143 + }, + { + "epoch": 0.3745106379284902, + "grad_norm": 0.33203125, + "learning_rate": 0.0012586048478691978, + "loss": 0.083, + "step": 43144 + }, + { + "epoch": 0.37451931840869435, + "grad_norm": 0.41796875, + "learning_rate": 0.0012585755511247258, + "loss": 0.0845, + "step": 43145 + }, + { + "epoch": 0.37452799888889854, + "grad_norm": 0.81640625, + "learning_rate": 0.0012585462542068384, + "loss": 0.1035, + "step": 43146 + }, + { + "epoch": 0.3745366793691027, + "grad_norm": 0.10205078125, + "learning_rate": 0.001258516957115567, + "loss": 0.0864, + "step": 43147 + }, + { + "epoch": 0.37454535984930687, + "grad_norm": 0.10986328125, + "learning_rate": 0.0012584876598509443, + "loss": 0.0859, + "step": 43148 + }, + { + "epoch": 0.374554040329511, + "grad_norm": 0.15234375, + "learning_rate": 0.0012584583624130011, + "loss": 0.1309, + "step": 43149 + }, + { + "epoch": 0.3745627208097152, + "grad_norm": 0.6953125, + "learning_rate": 0.0012584290648017704, + "loss": 0.126, + "step": 43150 + }, + { + "epoch": 0.37457140128991934, + "grad_norm": 0.1201171875, + "learning_rate": 0.0012583997670172844, + "loss": 0.1387, + "step": 43151 + }, + { + "epoch": 0.37458008177012353, + "grad_norm": 0.1103515625, + "learning_rate": 0.0012583704690595748, + "loss": 0.0815, + "step": 43152 + }, + { + "epoch": 0.37458876225032767, + "grad_norm": 0.466796875, + "learning_rate": 0.0012583411709286736, + "loss": 0.1367, + "step": 43153 + }, + { + "epoch": 0.37459744273053186, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012583118726246126, + "loss": 0.1553, + "step": 43154 + }, + { + "epoch": 0.374606123210736, + "grad_norm": 0.166015625, + "learning_rate": 0.0012582825741474242, + "loss": 0.1143, + "step": 43155 + }, + { + "epoch": 0.3746148036909402, + "grad_norm": 0.310546875, + "learning_rate": 0.0012582532754971405, + "loss": 0.0781, + "step": 43156 + }, + { + "epoch": 0.3746234841711443, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012582239766737934, + "loss": 0.124, + "step": 43157 + }, + { + "epoch": 0.3746321646513485, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012581946776774146, + "loss": 0.0908, + "step": 43158 + }, + { + "epoch": 0.37464084513155266, + "grad_norm": 0.515625, + "learning_rate": 0.0012581653785080369, + "loss": 0.106, + "step": 43159 + }, + { + "epoch": 0.37464952561175685, + "grad_norm": 0.25390625, + "learning_rate": 0.0012581360791656918, + "loss": 0.0713, + "step": 43160 + }, + { + "epoch": 0.374658206091961, + "grad_norm": 0.57421875, + "learning_rate": 0.0012581067796504114, + "loss": 0.1309, + "step": 43161 + }, + { + "epoch": 0.3746668865721652, + "grad_norm": 0.10498046875, + "learning_rate": 0.001258077479962228, + "loss": 0.0972, + "step": 43162 + }, + { + "epoch": 0.3746755670523693, + "grad_norm": 0.12890625, + "learning_rate": 0.001258048180101173, + "loss": 0.1045, + "step": 43163 + }, + { + "epoch": 0.3746842475325735, + "grad_norm": 0.345703125, + "learning_rate": 0.001258018880067279, + "loss": 0.1045, + "step": 43164 + }, + { + "epoch": 0.37469292801277765, + "grad_norm": 0.12890625, + "learning_rate": 0.0012579895798605785, + "loss": 0.0894, + "step": 43165 + }, + { + "epoch": 0.37470160849298184, + "grad_norm": 0.54296875, + "learning_rate": 0.0012579602794811023, + "loss": 0.1084, + "step": 43166 + }, + { + "epoch": 0.374710288973186, + "grad_norm": 0.3125, + "learning_rate": 0.0012579309789288835, + "loss": 0.083, + "step": 43167 + }, + { + "epoch": 0.37471896945339017, + "grad_norm": 0.10693359375, + "learning_rate": 0.0012579016782039534, + "loss": 0.1055, + "step": 43168 + }, + { + "epoch": 0.3747276499335943, + "grad_norm": 0.1005859375, + "learning_rate": 0.0012578723773063447, + "loss": 0.1113, + "step": 43169 + }, + { + "epoch": 0.3747363304137985, + "grad_norm": 0.306640625, + "learning_rate": 0.0012578430762360889, + "loss": 0.0977, + "step": 43170 + }, + { + "epoch": 0.37474501089400264, + "grad_norm": 0.251953125, + "learning_rate": 0.0012578137749932185, + "loss": 0.103, + "step": 43171 + }, + { + "epoch": 0.37475369137420683, + "grad_norm": 1.1796875, + "learning_rate": 0.001257784473577765, + "loss": 0.0898, + "step": 43172 + }, + { + "epoch": 0.37476237185441097, + "grad_norm": 0.498046875, + "learning_rate": 0.0012577551719897613, + "loss": 0.1099, + "step": 43173 + }, + { + "epoch": 0.37477105233461516, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012577258702292388, + "loss": 0.0732, + "step": 43174 + }, + { + "epoch": 0.3747797328148193, + "grad_norm": 0.72265625, + "learning_rate": 0.001257696568296229, + "loss": 0.1133, + "step": 43175 + }, + { + "epoch": 0.3747884132950235, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012576672661907652, + "loss": 0.1016, + "step": 43176 + }, + { + "epoch": 0.37479709377522763, + "grad_norm": 0.8359375, + "learning_rate": 0.001257637963912879, + "loss": 0.1338, + "step": 43177 + }, + { + "epoch": 0.3748057742554318, + "grad_norm": 0.11767578125, + "learning_rate": 0.001257608661462602, + "loss": 0.1055, + "step": 43178 + }, + { + "epoch": 0.37481445473563596, + "grad_norm": 0.197265625, + "learning_rate": 0.0012575793588399664, + "loss": 0.0977, + "step": 43179 + }, + { + "epoch": 0.37482313521584015, + "grad_norm": 0.37109375, + "learning_rate": 0.0012575500560450046, + "loss": 0.1201, + "step": 43180 + }, + { + "epoch": 0.3748318156960443, + "grad_norm": 0.396484375, + "learning_rate": 0.0012575207530777486, + "loss": 0.1201, + "step": 43181 + }, + { + "epoch": 0.3748404961762485, + "grad_norm": 0.5390625, + "learning_rate": 0.0012574914499382302, + "loss": 0.1226, + "step": 43182 + }, + { + "epoch": 0.3748491766564526, + "grad_norm": 0.19140625, + "learning_rate": 0.0012574621466264811, + "loss": 0.166, + "step": 43183 + }, + { + "epoch": 0.3748578571366568, + "grad_norm": 0.3359375, + "learning_rate": 0.0012574328431425338, + "loss": 0.0986, + "step": 43184 + }, + { + "epoch": 0.37486653761686095, + "grad_norm": 0.17578125, + "learning_rate": 0.0012574035394864206, + "loss": 0.1348, + "step": 43185 + }, + { + "epoch": 0.37487521809706514, + "grad_norm": 0.275390625, + "learning_rate": 0.0012573742356581734, + "loss": 0.1064, + "step": 43186 + }, + { + "epoch": 0.3748838985772693, + "grad_norm": 0.5859375, + "learning_rate": 0.0012573449316578239, + "loss": 0.1074, + "step": 43187 + }, + { + "epoch": 0.3748925790574735, + "grad_norm": 0.578125, + "learning_rate": 0.0012573156274854043, + "loss": 0.0693, + "step": 43188 + }, + { + "epoch": 0.3749012595376776, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012572863231409465, + "loss": 0.0938, + "step": 43189 + }, + { + "epoch": 0.3749099400178818, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012572570186244832, + "loss": 0.1348, + "step": 43190 + }, + { + "epoch": 0.37491862049808594, + "grad_norm": 0.09765625, + "learning_rate": 0.0012572277139360458, + "loss": 0.1016, + "step": 43191 + }, + { + "epoch": 0.37492730097829013, + "grad_norm": 0.421875, + "learning_rate": 0.0012571984090756665, + "loss": 0.126, + "step": 43192 + }, + { + "epoch": 0.37493598145849427, + "grad_norm": 1.0703125, + "learning_rate": 0.0012571691040433773, + "loss": 0.127, + "step": 43193 + }, + { + "epoch": 0.37494466193869846, + "grad_norm": 0.58203125, + "learning_rate": 0.0012571397988392103, + "loss": 0.0591, + "step": 43194 + }, + { + "epoch": 0.3749533424189026, + "grad_norm": 0.6640625, + "learning_rate": 0.0012571104934631981, + "loss": 0.1055, + "step": 43195 + }, + { + "epoch": 0.3749620228991068, + "grad_norm": 0.275390625, + "learning_rate": 0.0012570811879153716, + "loss": 0.1094, + "step": 43196 + }, + { + "epoch": 0.37497070337931093, + "grad_norm": 0.83203125, + "learning_rate": 0.0012570518821957634, + "loss": 0.2031, + "step": 43197 + }, + { + "epoch": 0.3749793838595151, + "grad_norm": 0.54296875, + "learning_rate": 0.0012570225763044059, + "loss": 0.0864, + "step": 43198 + }, + { + "epoch": 0.37498806433971926, + "grad_norm": 0.119140625, + "learning_rate": 0.0012569932702413311, + "loss": 0.084, + "step": 43199 + }, + { + "epoch": 0.37499674481992346, + "grad_norm": 0.578125, + "learning_rate": 0.0012569639640065704, + "loss": 0.1045, + "step": 43200 + }, + { + "epoch": 0.3750054253001276, + "grad_norm": 0.32421875, + "learning_rate": 0.0012569346576001565, + "loss": 0.1216, + "step": 43201 + }, + { + "epoch": 0.3750141057803318, + "grad_norm": 0.1767578125, + "learning_rate": 0.0012569053510221213, + "loss": 0.0933, + "step": 43202 + }, + { + "epoch": 0.3750227862605359, + "grad_norm": 0.267578125, + "learning_rate": 0.0012568760442724968, + "loss": 0.1025, + "step": 43203 + }, + { + "epoch": 0.3750314667407401, + "grad_norm": 0.271484375, + "learning_rate": 0.0012568467373513148, + "loss": 0.0713, + "step": 43204 + }, + { + "epoch": 0.37504014722094425, + "grad_norm": 0.326171875, + "learning_rate": 0.0012568174302586077, + "loss": 0.1133, + "step": 43205 + }, + { + "epoch": 0.37504882770114845, + "grad_norm": 0.0947265625, + "learning_rate": 0.0012567881229944072, + "loss": 0.0972, + "step": 43206 + }, + { + "epoch": 0.3750575081813526, + "grad_norm": 0.37109375, + "learning_rate": 0.0012567588155587455, + "loss": 0.1084, + "step": 43207 + }, + { + "epoch": 0.3750661886615568, + "grad_norm": 0.515625, + "learning_rate": 0.001256729507951655, + "loss": 0.1001, + "step": 43208 + }, + { + "epoch": 0.3750748691417609, + "grad_norm": 0.66796875, + "learning_rate": 0.0012567002001731675, + "loss": 0.1001, + "step": 43209 + }, + { + "epoch": 0.3750835496219651, + "grad_norm": 0.232421875, + "learning_rate": 0.0012566708922233146, + "loss": 0.0918, + "step": 43210 + }, + { + "epoch": 0.37509223010216924, + "grad_norm": 0.5, + "learning_rate": 0.0012566415841021293, + "loss": 0.0771, + "step": 43211 + }, + { + "epoch": 0.37510091058237344, + "grad_norm": 0.48828125, + "learning_rate": 0.0012566122758096428, + "loss": 0.0996, + "step": 43212 + }, + { + "epoch": 0.3751095910625776, + "grad_norm": 0.13671875, + "learning_rate": 0.0012565829673458876, + "loss": 0.0811, + "step": 43213 + }, + { + "epoch": 0.37511827154278177, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012565536587108954, + "loss": 0.0688, + "step": 43214 + }, + { + "epoch": 0.3751269520229859, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012565243499046987, + "loss": 0.0952, + "step": 43215 + }, + { + "epoch": 0.3751356325031901, + "grad_norm": 0.1572265625, + "learning_rate": 0.0012564950409273295, + "loss": 0.1279, + "step": 43216 + }, + { + "epoch": 0.37514431298339423, + "grad_norm": 0.703125, + "learning_rate": 0.0012564657317788192, + "loss": 0.106, + "step": 43217 + }, + { + "epoch": 0.3751529934635984, + "grad_norm": 0.22265625, + "learning_rate": 0.0012564364224592007, + "loss": 0.123, + "step": 43218 + }, + { + "epoch": 0.37516167394380256, + "grad_norm": 0.244140625, + "learning_rate": 0.0012564071129685054, + "loss": 0.0981, + "step": 43219 + }, + { + "epoch": 0.37517035442400676, + "grad_norm": 0.103515625, + "learning_rate": 0.0012563778033067659, + "loss": 0.0698, + "step": 43220 + }, + { + "epoch": 0.3751790349042109, + "grad_norm": 0.80078125, + "learning_rate": 0.0012563484934740136, + "loss": 0.0884, + "step": 43221 + }, + { + "epoch": 0.3751877153844151, + "grad_norm": 0.53125, + "learning_rate": 0.0012563191834702814, + "loss": 0.1465, + "step": 43222 + }, + { + "epoch": 0.3751963958646192, + "grad_norm": 0.29296875, + "learning_rate": 0.0012562898732956006, + "loss": 0.1006, + "step": 43223 + }, + { + "epoch": 0.3752050763448234, + "grad_norm": 0.1962890625, + "learning_rate": 0.0012562605629500036, + "loss": 0.1079, + "step": 43224 + }, + { + "epoch": 0.37521375682502756, + "grad_norm": 0.208984375, + "learning_rate": 0.0012562312524335227, + "loss": 0.0894, + "step": 43225 + }, + { + "epoch": 0.37522243730523175, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012562019417461893, + "loss": 0.1006, + "step": 43226 + }, + { + "epoch": 0.3752311177854359, + "grad_norm": 0.0966796875, + "learning_rate": 0.0012561726308880355, + "loss": 0.104, + "step": 43227 + }, + { + "epoch": 0.3752397982656401, + "grad_norm": 0.4140625, + "learning_rate": 0.0012561433198590942, + "loss": 0.1016, + "step": 43228 + }, + { + "epoch": 0.3752484787458442, + "grad_norm": 1.140625, + "learning_rate": 0.0012561140086593968, + "loss": 0.1191, + "step": 43229 + }, + { + "epoch": 0.3752571592260484, + "grad_norm": 0.298828125, + "learning_rate": 0.0012560846972889752, + "loss": 0.103, + "step": 43230 + }, + { + "epoch": 0.37526583970625255, + "grad_norm": 0.470703125, + "learning_rate": 0.0012560553857478623, + "loss": 0.0693, + "step": 43231 + }, + { + "epoch": 0.37527452018645674, + "grad_norm": 0.099609375, + "learning_rate": 0.0012560260740360893, + "loss": 0.084, + "step": 43232 + }, + { + "epoch": 0.3752832006666609, + "grad_norm": 0.287109375, + "learning_rate": 0.0012559967621536881, + "loss": 0.1143, + "step": 43233 + }, + { + "epoch": 0.37529188114686507, + "grad_norm": 0.1123046875, + "learning_rate": 0.0012559674501006917, + "loss": 0.1113, + "step": 43234 + }, + { + "epoch": 0.3753005616270692, + "grad_norm": 0.25390625, + "learning_rate": 0.0012559381378771312, + "loss": 0.1006, + "step": 43235 + }, + { + "epoch": 0.3753092421072734, + "grad_norm": 0.2275390625, + "learning_rate": 0.0012559088254830396, + "loss": 0.1045, + "step": 43236 + }, + { + "epoch": 0.37531792258747754, + "grad_norm": 0.142578125, + "learning_rate": 0.0012558795129184478, + "loss": 0.1182, + "step": 43237 + }, + { + "epoch": 0.37532660306768173, + "grad_norm": 0.60546875, + "learning_rate": 0.0012558502001833892, + "loss": 0.0913, + "step": 43238 + }, + { + "epoch": 0.37533528354788587, + "grad_norm": 0.33203125, + "learning_rate": 0.001255820887277895, + "loss": 0.0903, + "step": 43239 + }, + { + "epoch": 0.37534396402809006, + "grad_norm": 0.2734375, + "learning_rate": 0.001255791574201997, + "loss": 0.0977, + "step": 43240 + }, + { + "epoch": 0.3753526445082942, + "grad_norm": 0.1376953125, + "learning_rate": 0.001255762260955728, + "loss": 0.1113, + "step": 43241 + }, + { + "epoch": 0.3753613249884984, + "grad_norm": 0.67578125, + "learning_rate": 0.00125573294753912, + "loss": 0.1099, + "step": 43242 + }, + { + "epoch": 0.37537000546870253, + "grad_norm": 0.08349609375, + "learning_rate": 0.0012557036339522043, + "loss": 0.1064, + "step": 43243 + }, + { + "epoch": 0.3753786859489067, + "grad_norm": 0.08349609375, + "learning_rate": 0.0012556743201950138, + "loss": 0.084, + "step": 43244 + }, + { + "epoch": 0.37538736642911086, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012556450062675797, + "loss": 0.0996, + "step": 43245 + }, + { + "epoch": 0.37539604690931505, + "grad_norm": 0.2578125, + "learning_rate": 0.001255615692169935, + "loss": 0.1152, + "step": 43246 + }, + { + "epoch": 0.3754047273895192, + "grad_norm": 0.41015625, + "learning_rate": 0.001255586377902111, + "loss": 0.1484, + "step": 43247 + }, + { + "epoch": 0.3754134078697233, + "grad_norm": 0.65234375, + "learning_rate": 0.0012555570634641404, + "loss": 0.1279, + "step": 43248 + }, + { + "epoch": 0.3754220883499275, + "grad_norm": 0.486328125, + "learning_rate": 0.0012555277488560548, + "loss": 0.1074, + "step": 43249 + }, + { + "epoch": 0.37543076883013166, + "grad_norm": 0.296875, + "learning_rate": 0.0012554984340778866, + "loss": 0.0771, + "step": 43250 + }, + { + "epoch": 0.37543944931033585, + "grad_norm": 0.2001953125, + "learning_rate": 0.0012554691191296674, + "loss": 0.0781, + "step": 43251 + }, + { + "epoch": 0.37544812979054, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012554398040114296, + "loss": 0.0928, + "step": 43252 + }, + { + "epoch": 0.3754568102707442, + "grad_norm": 0.1201171875, + "learning_rate": 0.001255410488723205, + "loss": 0.0957, + "step": 43253 + }, + { + "epoch": 0.3754654907509483, + "grad_norm": 0.07763671875, + "learning_rate": 0.0012553811732650259, + "loss": 0.0723, + "step": 43254 + }, + { + "epoch": 0.3754741712311525, + "grad_norm": 0.1630859375, + "learning_rate": 0.0012553518576369244, + "loss": 0.1221, + "step": 43255 + }, + { + "epoch": 0.37548285171135665, + "grad_norm": 0.08984375, + "learning_rate": 0.0012553225418389326, + "loss": 0.0957, + "step": 43256 + }, + { + "epoch": 0.37549153219156084, + "grad_norm": 0.3671875, + "learning_rate": 0.001255293225871082, + "loss": 0.0923, + "step": 43257 + }, + { + "epoch": 0.375500212671765, + "grad_norm": 0.31640625, + "learning_rate": 0.001255263909733405, + "loss": 0.0962, + "step": 43258 + }, + { + "epoch": 0.37550889315196917, + "grad_norm": 0.10693359375, + "learning_rate": 0.001255234593425934, + "loss": 0.1006, + "step": 43259 + }, + { + "epoch": 0.3755175736321733, + "grad_norm": 0.265625, + "learning_rate": 0.001255205276948701, + "loss": 0.0879, + "step": 43260 + }, + { + "epoch": 0.3755262541123775, + "grad_norm": 0.466796875, + "learning_rate": 0.0012551759603017377, + "loss": 0.1299, + "step": 43261 + }, + { + "epoch": 0.37553493459258164, + "grad_norm": 0.1728515625, + "learning_rate": 0.001255146643485076, + "loss": 0.0806, + "step": 43262 + }, + { + "epoch": 0.37554361507278583, + "grad_norm": 0.51171875, + "learning_rate": 0.0012551173264987486, + "loss": 0.0898, + "step": 43263 + }, + { + "epoch": 0.37555229555298997, + "grad_norm": 0.337890625, + "learning_rate": 0.001255088009342787, + "loss": 0.085, + "step": 43264 + }, + { + "epoch": 0.37556097603319416, + "grad_norm": 0.3046875, + "learning_rate": 0.0012550586920172233, + "loss": 0.0957, + "step": 43265 + }, + { + "epoch": 0.3755696565133983, + "grad_norm": 0.447265625, + "learning_rate": 0.00125502937452209, + "loss": 0.0835, + "step": 43266 + }, + { + "epoch": 0.3755783369936025, + "grad_norm": 0.21875, + "learning_rate": 0.0012550000568574192, + "loss": 0.0713, + "step": 43267 + }, + { + "epoch": 0.37558701747380663, + "grad_norm": 0.25390625, + "learning_rate": 0.0012549707390232423, + "loss": 0.1152, + "step": 43268 + }, + { + "epoch": 0.3755956979540108, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012549414210195919, + "loss": 0.0889, + "step": 43269 + }, + { + "epoch": 0.37560437843421496, + "grad_norm": 0.50390625, + "learning_rate": 0.0012549121028464997, + "loss": 0.1035, + "step": 43270 + }, + { + "epoch": 0.37561305891441915, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012548827845039981, + "loss": 0.0898, + "step": 43271 + }, + { + "epoch": 0.3756217393946233, + "grad_norm": 0.13671875, + "learning_rate": 0.0012548534659921188, + "loss": 0.1484, + "step": 43272 + }, + { + "epoch": 0.3756304198748275, + "grad_norm": 0.42578125, + "learning_rate": 0.0012548241473108943, + "loss": 0.0986, + "step": 43273 + }, + { + "epoch": 0.3756391003550316, + "grad_norm": 0.1181640625, + "learning_rate": 0.0012547948284603566, + "loss": 0.082, + "step": 43274 + }, + { + "epoch": 0.3756477808352358, + "grad_norm": 0.349609375, + "learning_rate": 0.0012547655094405368, + "loss": 0.0986, + "step": 43275 + }, + { + "epoch": 0.37565646131543995, + "grad_norm": 0.2177734375, + "learning_rate": 0.0012547361902514684, + "loss": 0.1094, + "step": 43276 + }, + { + "epoch": 0.37566514179564414, + "grad_norm": 0.36328125, + "learning_rate": 0.001254706870893183, + "loss": 0.0806, + "step": 43277 + }, + { + "epoch": 0.3756738222758483, + "grad_norm": 0.0771484375, + "learning_rate": 0.0012546775513657125, + "loss": 0.0859, + "step": 43278 + }, + { + "epoch": 0.37568250275605247, + "grad_norm": 0.28515625, + "learning_rate": 0.0012546482316690883, + "loss": 0.0869, + "step": 43279 + }, + { + "epoch": 0.3756911832362566, + "grad_norm": 0.08203125, + "learning_rate": 0.0012546189118033439, + "loss": 0.0674, + "step": 43280 + }, + { + "epoch": 0.3756998637164608, + "grad_norm": 0.265625, + "learning_rate": 0.0012545895917685102, + "loss": 0.0977, + "step": 43281 + }, + { + "epoch": 0.37570854419666494, + "grad_norm": 0.25, + "learning_rate": 0.0012545602715646196, + "loss": 0.1152, + "step": 43282 + }, + { + "epoch": 0.37571722467686913, + "grad_norm": 0.2099609375, + "learning_rate": 0.0012545309511917042, + "loss": 0.126, + "step": 43283 + }, + { + "epoch": 0.37572590515707327, + "grad_norm": 0.22265625, + "learning_rate": 0.0012545016306497964, + "loss": 0.1045, + "step": 43284 + }, + { + "epoch": 0.37573458563727746, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012544723099389274, + "loss": 0.1084, + "step": 43285 + }, + { + "epoch": 0.3757432661174816, + "grad_norm": 0.0791015625, + "learning_rate": 0.00125444298905913, + "loss": 0.083, + "step": 43286 + }, + { + "epoch": 0.3757519465976858, + "grad_norm": 0.201171875, + "learning_rate": 0.001254413668010436, + "loss": 0.0869, + "step": 43287 + }, + { + "epoch": 0.37576062707788993, + "grad_norm": 0.19921875, + "learning_rate": 0.001254384346792878, + "loss": 0.1221, + "step": 43288 + }, + { + "epoch": 0.3757693075580941, + "grad_norm": 0.369140625, + "learning_rate": 0.001254355025406487, + "loss": 0.0928, + "step": 43289 + }, + { + "epoch": 0.37577798803829826, + "grad_norm": 0.32421875, + "learning_rate": 0.001254325703851296, + "loss": 0.0996, + "step": 43290 + }, + { + "epoch": 0.37578666851850245, + "grad_norm": 0.341796875, + "learning_rate": 0.0012542963821273367, + "loss": 0.0913, + "step": 43291 + }, + { + "epoch": 0.3757953489987066, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012542670602346412, + "loss": 0.1035, + "step": 43292 + }, + { + "epoch": 0.3758040294789108, + "grad_norm": 0.466796875, + "learning_rate": 0.0012542377381732416, + "loss": 0.1201, + "step": 43293 + }, + { + "epoch": 0.3758127099591149, + "grad_norm": 0.5546875, + "learning_rate": 0.0012542084159431694, + "loss": 0.1143, + "step": 43294 + }, + { + "epoch": 0.3758213904393191, + "grad_norm": 0.2080078125, + "learning_rate": 0.001254179093544458, + "loss": 0.085, + "step": 43295 + }, + { + "epoch": 0.37583007091952325, + "grad_norm": 0.57421875, + "learning_rate": 0.001254149770977138, + "loss": 0.1797, + "step": 43296 + }, + { + "epoch": 0.37583875139972744, + "grad_norm": 0.51953125, + "learning_rate": 0.0012541204482412428, + "loss": 0.0918, + "step": 43297 + }, + { + "epoch": 0.3758474318799316, + "grad_norm": 0.16796875, + "learning_rate": 0.0012540911253368031, + "loss": 0.1367, + "step": 43298 + }, + { + "epoch": 0.3758561123601358, + "grad_norm": 0.197265625, + "learning_rate": 0.0012540618022638525, + "loss": 0.1006, + "step": 43299 + }, + { + "epoch": 0.3758647928403399, + "grad_norm": 0.298828125, + "learning_rate": 0.0012540324790224213, + "loss": 0.0947, + "step": 43300 + }, + { + "epoch": 0.3758734733205441, + "grad_norm": 0.10400390625, + "learning_rate": 0.001254003155612543, + "loss": 0.1167, + "step": 43301 + }, + { + "epoch": 0.37588215380074824, + "grad_norm": 0.2470703125, + "learning_rate": 0.001253973832034249, + "loss": 0.0898, + "step": 43302 + }, + { + "epoch": 0.37589083428095243, + "grad_norm": 0.37109375, + "learning_rate": 0.0012539445082875716, + "loss": 0.1055, + "step": 43303 + }, + { + "epoch": 0.3758995147611566, + "grad_norm": 0.064453125, + "learning_rate": 0.0012539151843725426, + "loss": 0.0752, + "step": 43304 + }, + { + "epoch": 0.37590819524136077, + "grad_norm": 0.357421875, + "learning_rate": 0.0012538858602891946, + "loss": 0.0913, + "step": 43305 + }, + { + "epoch": 0.3759168757215649, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012538565360375594, + "loss": 0.0913, + "step": 43306 + }, + { + "epoch": 0.3759255562017691, + "grad_norm": 0.140625, + "learning_rate": 0.0012538272116176688, + "loss": 0.0708, + "step": 43307 + }, + { + "epoch": 0.37593423668197323, + "grad_norm": 0.7421875, + "learning_rate": 0.001253797887029555, + "loss": 0.0864, + "step": 43308 + }, + { + "epoch": 0.3759429171621774, + "grad_norm": 0.68359375, + "learning_rate": 0.0012537685622732503, + "loss": 0.1357, + "step": 43309 + }, + { + "epoch": 0.37595159764238156, + "grad_norm": 0.349609375, + "learning_rate": 0.0012537392373487863, + "loss": 0.1045, + "step": 43310 + }, + { + "epoch": 0.37596027812258576, + "grad_norm": 0.158203125, + "learning_rate": 0.0012537099122561957, + "loss": 0.1055, + "step": 43311 + }, + { + "epoch": 0.3759689586027899, + "grad_norm": 0.38671875, + "learning_rate": 0.0012536805869955102, + "loss": 0.0898, + "step": 43312 + }, + { + "epoch": 0.3759776390829941, + "grad_norm": 0.482421875, + "learning_rate": 0.001253651261566762, + "loss": 0.0791, + "step": 43313 + }, + { + "epoch": 0.3759863195631982, + "grad_norm": 0.2158203125, + "learning_rate": 0.0012536219359699828, + "loss": 0.0854, + "step": 43314 + }, + { + "epoch": 0.3759950000434024, + "grad_norm": 0.39453125, + "learning_rate": 0.0012535926102052051, + "loss": 0.0728, + "step": 43315 + }, + { + "epoch": 0.37600368052360655, + "grad_norm": 0.39453125, + "learning_rate": 0.001253563284272461, + "loss": 0.0845, + "step": 43316 + }, + { + "epoch": 0.37601236100381075, + "grad_norm": 0.09375, + "learning_rate": 0.0012535339581717822, + "loss": 0.1143, + "step": 43317 + }, + { + "epoch": 0.3760210414840149, + "grad_norm": 0.494140625, + "learning_rate": 0.0012535046319032011, + "loss": 0.0874, + "step": 43318 + }, + { + "epoch": 0.3760297219642191, + "grad_norm": 0.30078125, + "learning_rate": 0.0012534753054667493, + "loss": 0.0908, + "step": 43319 + }, + { + "epoch": 0.3760384024444232, + "grad_norm": 0.16015625, + "learning_rate": 0.0012534459788624598, + "loss": 0.0747, + "step": 43320 + }, + { + "epoch": 0.3760470829246274, + "grad_norm": 0.45703125, + "learning_rate": 0.0012534166520903637, + "loss": 0.084, + "step": 43321 + }, + { + "epoch": 0.37605576340483154, + "grad_norm": 0.953125, + "learning_rate": 0.0012533873251504935, + "loss": 0.1748, + "step": 43322 + }, + { + "epoch": 0.37606444388503574, + "grad_norm": 0.478515625, + "learning_rate": 0.001253357998042881, + "loss": 0.1309, + "step": 43323 + }, + { + "epoch": 0.3760731243652399, + "grad_norm": 0.18359375, + "learning_rate": 0.0012533286707675589, + "loss": 0.085, + "step": 43324 + }, + { + "epoch": 0.37608180484544407, + "grad_norm": 0.15625, + "learning_rate": 0.0012532993433245587, + "loss": 0.0918, + "step": 43325 + }, + { + "epoch": 0.3760904853256482, + "grad_norm": 0.26953125, + "learning_rate": 0.0012532700157139127, + "loss": 0.127, + "step": 43326 + }, + { + "epoch": 0.3760991658058524, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012532406879356528, + "loss": 0.0947, + "step": 43327 + }, + { + "epoch": 0.37610784628605654, + "grad_norm": 0.310546875, + "learning_rate": 0.0012532113599898113, + "loss": 0.1582, + "step": 43328 + }, + { + "epoch": 0.37611652676626073, + "grad_norm": 0.0751953125, + "learning_rate": 0.00125318203187642, + "loss": 0.082, + "step": 43329 + }, + { + "epoch": 0.37612520724646487, + "grad_norm": 0.328125, + "learning_rate": 0.0012531527035955115, + "loss": 0.126, + "step": 43330 + }, + { + "epoch": 0.37613388772666906, + "grad_norm": 0.41796875, + "learning_rate": 0.001253123375147117, + "loss": 0.1201, + "step": 43331 + }, + { + "epoch": 0.3761425682068732, + "grad_norm": 0.5, + "learning_rate": 0.0012530940465312692, + "loss": 0.084, + "step": 43332 + }, + { + "epoch": 0.3761512486870774, + "grad_norm": 0.126953125, + "learning_rate": 0.001253064717748, + "loss": 0.0786, + "step": 43333 + }, + { + "epoch": 0.3761599291672815, + "grad_norm": 0.1103515625, + "learning_rate": 0.001253035388797342, + "loss": 0.1182, + "step": 43334 + }, + { + "epoch": 0.3761686096474857, + "grad_norm": 0.49609375, + "learning_rate": 0.0012530060596793262, + "loss": 0.1016, + "step": 43335 + }, + { + "epoch": 0.37617729012768986, + "grad_norm": 0.455078125, + "learning_rate": 0.0012529767303939858, + "loss": 0.127, + "step": 43336 + }, + { + "epoch": 0.37618597060789405, + "grad_norm": 0.287109375, + "learning_rate": 0.001252947400941352, + "loss": 0.0996, + "step": 43337 + }, + { + "epoch": 0.3761946510880982, + "grad_norm": 0.361328125, + "learning_rate": 0.0012529180713214576, + "loss": 0.0796, + "step": 43338 + }, + { + "epoch": 0.3762033315683024, + "grad_norm": 0.1416015625, + "learning_rate": 0.001252888741534334, + "loss": 0.1084, + "step": 43339 + }, + { + "epoch": 0.3762120120485065, + "grad_norm": 0.15234375, + "learning_rate": 0.0012528594115800136, + "loss": 0.1006, + "step": 43340 + }, + { + "epoch": 0.3762206925287107, + "grad_norm": 0.44921875, + "learning_rate": 0.0012528300814585282, + "loss": 0.1108, + "step": 43341 + }, + { + "epoch": 0.37622937300891485, + "grad_norm": 0.12451171875, + "learning_rate": 0.0012528007511699104, + "loss": 0.0952, + "step": 43342 + }, + { + "epoch": 0.37623805348911904, + "grad_norm": 0.412109375, + "learning_rate": 0.001252771420714192, + "loss": 0.0579, + "step": 43343 + }, + { + "epoch": 0.3762467339693232, + "grad_norm": 0.4765625, + "learning_rate": 0.001252742090091405, + "loss": 0.0806, + "step": 43344 + }, + { + "epoch": 0.37625541444952737, + "grad_norm": 0.08642578125, + "learning_rate": 0.0012527127593015817, + "loss": 0.126, + "step": 43345 + }, + { + "epoch": 0.3762640949297315, + "grad_norm": 0.37109375, + "learning_rate": 0.0012526834283447539, + "loss": 0.1465, + "step": 43346 + }, + { + "epoch": 0.3762727754099357, + "grad_norm": 0.07373046875, + "learning_rate": 0.0012526540972209537, + "loss": 0.0767, + "step": 43347 + }, + { + "epoch": 0.37628145589013984, + "grad_norm": 0.2451171875, + "learning_rate": 0.0012526247659302133, + "loss": 0.0938, + "step": 43348 + }, + { + "epoch": 0.37629013637034403, + "grad_norm": 0.212890625, + "learning_rate": 0.0012525954344725647, + "loss": 0.1094, + "step": 43349 + }, + { + "epoch": 0.37629881685054817, + "grad_norm": 0.416015625, + "learning_rate": 0.0012525661028480403, + "loss": 0.125, + "step": 43350 + }, + { + "epoch": 0.37630749733075236, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012525367710566717, + "loss": 0.0991, + "step": 43351 + }, + { + "epoch": 0.3763161778109565, + "grad_norm": 0.275390625, + "learning_rate": 0.0012525074390984912, + "loss": 0.0791, + "step": 43352 + }, + { + "epoch": 0.3763248582911607, + "grad_norm": 0.2373046875, + "learning_rate": 0.0012524781069735307, + "loss": 0.1348, + "step": 43353 + }, + { + "epoch": 0.37633353877136483, + "grad_norm": 0.134765625, + "learning_rate": 0.0012524487746818228, + "loss": 0.0908, + "step": 43354 + }, + { + "epoch": 0.376342219251569, + "grad_norm": 0.11767578125, + "learning_rate": 0.001252419442223399, + "loss": 0.0972, + "step": 43355 + }, + { + "epoch": 0.37635089973177316, + "grad_norm": 0.0810546875, + "learning_rate": 0.0012523901095982917, + "loss": 0.0918, + "step": 43356 + }, + { + "epoch": 0.37635958021197735, + "grad_norm": 1.046875, + "learning_rate": 0.0012523607768065326, + "loss": 0.1426, + "step": 43357 + }, + { + "epoch": 0.3763682606921815, + "grad_norm": 0.1357421875, + "learning_rate": 0.0012523314438481544, + "loss": 0.1147, + "step": 43358 + }, + { + "epoch": 0.3763769411723857, + "grad_norm": 0.1220703125, + "learning_rate": 0.0012523021107231887, + "loss": 0.0986, + "step": 43359 + }, + { + "epoch": 0.3763856216525898, + "grad_norm": 0.419921875, + "learning_rate": 0.0012522727774316672, + "loss": 0.0938, + "step": 43360 + }, + { + "epoch": 0.376394302132794, + "grad_norm": 0.365234375, + "learning_rate": 0.0012522434439736232, + "loss": 0.1138, + "step": 43361 + }, + { + "epoch": 0.37640298261299815, + "grad_norm": 0.11328125, + "learning_rate": 0.0012522141103490877, + "loss": 0.1377, + "step": 43362 + }, + { + "epoch": 0.37641166309320234, + "grad_norm": 0.318359375, + "learning_rate": 0.0012521847765580932, + "loss": 0.0918, + "step": 43363 + }, + { + "epoch": 0.3764203435734065, + "grad_norm": 0.09765625, + "learning_rate": 0.0012521554426006718, + "loss": 0.105, + "step": 43364 + }, + { + "epoch": 0.3764290240536107, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012521261084768556, + "loss": 0.1143, + "step": 43365 + }, + { + "epoch": 0.3764377045338148, + "grad_norm": 0.080078125, + "learning_rate": 0.0012520967741866762, + "loss": 0.0854, + "step": 43366 + }, + { + "epoch": 0.376446385014019, + "grad_norm": 0.328125, + "learning_rate": 0.0012520674397301663, + "loss": 0.0967, + "step": 43367 + }, + { + "epoch": 0.37645506549422314, + "grad_norm": 0.0810546875, + "learning_rate": 0.0012520381051073577, + "loss": 0.0913, + "step": 43368 + }, + { + "epoch": 0.37646374597442733, + "grad_norm": 0.451171875, + "learning_rate": 0.0012520087703182825, + "loss": 0.1104, + "step": 43369 + }, + { + "epoch": 0.37647242645463147, + "grad_norm": 0.06103515625, + "learning_rate": 0.0012519794353629727, + "loss": 0.082, + "step": 43370 + }, + { + "epoch": 0.3764811069348356, + "grad_norm": 0.1015625, + "learning_rate": 0.0012519501002414605, + "loss": 0.0913, + "step": 43371 + }, + { + "epoch": 0.3764897874150398, + "grad_norm": 0.341796875, + "learning_rate": 0.001251920764953778, + "loss": 0.1108, + "step": 43372 + }, + { + "epoch": 0.37649846789524394, + "grad_norm": 0.2119140625, + "learning_rate": 0.0012518914294999573, + "loss": 0.1006, + "step": 43373 + }, + { + "epoch": 0.37650714837544813, + "grad_norm": 0.154296875, + "learning_rate": 0.0012518620938800303, + "loss": 0.1357, + "step": 43374 + }, + { + "epoch": 0.37651582885565227, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012518327580940292, + "loss": 0.0859, + "step": 43375 + }, + { + "epoch": 0.37652450933585646, + "grad_norm": 0.1494140625, + "learning_rate": 0.0012518034221419862, + "loss": 0.0967, + "step": 43376 + }, + { + "epoch": 0.3765331898160606, + "grad_norm": 0.3203125, + "learning_rate": 0.0012517740860239332, + "loss": 0.0908, + "step": 43377 + }, + { + "epoch": 0.3765418702962648, + "grad_norm": 0.28125, + "learning_rate": 0.0012517447497399023, + "loss": 0.1611, + "step": 43378 + }, + { + "epoch": 0.37655055077646893, + "grad_norm": 0.94921875, + "learning_rate": 0.0012517154132899258, + "loss": 0.1226, + "step": 43379 + }, + { + "epoch": 0.3765592312566731, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012516860766740353, + "loss": 0.1064, + "step": 43380 + }, + { + "epoch": 0.37656791173687726, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012516567398922634, + "loss": 0.0732, + "step": 43381 + }, + { + "epoch": 0.37657659221708145, + "grad_norm": 0.259765625, + "learning_rate": 0.0012516274029446418, + "loss": 0.1367, + "step": 43382 + }, + { + "epoch": 0.3765852726972856, + "grad_norm": 0.10888671875, + "learning_rate": 0.001251598065831203, + "loss": 0.104, + "step": 43383 + }, + { + "epoch": 0.3765939531774898, + "grad_norm": 0.1513671875, + "learning_rate": 0.0012515687285519785, + "loss": 0.1064, + "step": 43384 + }, + { + "epoch": 0.3766026336576939, + "grad_norm": 0.400390625, + "learning_rate": 0.0012515393911070012, + "loss": 0.1143, + "step": 43385 + }, + { + "epoch": 0.3766113141378981, + "grad_norm": 0.236328125, + "learning_rate": 0.0012515100534963024, + "loss": 0.1182, + "step": 43386 + }, + { + "epoch": 0.37661999461810225, + "grad_norm": 0.263671875, + "learning_rate": 0.0012514807157199145, + "loss": 0.1172, + "step": 43387 + }, + { + "epoch": 0.37662867509830644, + "grad_norm": 0.52734375, + "learning_rate": 0.0012514513777778693, + "loss": 0.1001, + "step": 43388 + }, + { + "epoch": 0.3766373555785106, + "grad_norm": 0.126953125, + "learning_rate": 0.0012514220396701997, + "loss": 0.0981, + "step": 43389 + }, + { + "epoch": 0.3766460360587148, + "grad_norm": 0.091796875, + "learning_rate": 0.0012513927013969369, + "loss": 0.0732, + "step": 43390 + }, + { + "epoch": 0.3766547165389189, + "grad_norm": 0.38671875, + "learning_rate": 0.0012513633629581131, + "loss": 0.0894, + "step": 43391 + }, + { + "epoch": 0.3766633970191231, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012513340243537608, + "loss": 0.1338, + "step": 43392 + }, + { + "epoch": 0.37667207749932724, + "grad_norm": 0.1337890625, + "learning_rate": 0.001251304685583912, + "loss": 0.0967, + "step": 43393 + }, + { + "epoch": 0.37668075797953143, + "grad_norm": 0.609375, + "learning_rate": 0.0012512753466485988, + "loss": 0.1504, + "step": 43394 + }, + { + "epoch": 0.37668943845973557, + "grad_norm": 0.142578125, + "learning_rate": 0.0012512460075478529, + "loss": 0.1025, + "step": 43395 + }, + { + "epoch": 0.37669811893993976, + "grad_norm": 0.251953125, + "learning_rate": 0.0012512166682817065, + "loss": 0.0664, + "step": 43396 + }, + { + "epoch": 0.3767067994201439, + "grad_norm": 0.302734375, + "learning_rate": 0.001251187328850192, + "loss": 0.1011, + "step": 43397 + }, + { + "epoch": 0.3767154799003481, + "grad_norm": 0.59765625, + "learning_rate": 0.0012511579892533415, + "loss": 0.0718, + "step": 43398 + }, + { + "epoch": 0.37672416038055223, + "grad_norm": 0.62109375, + "learning_rate": 0.0012511286494911865, + "loss": 0.1367, + "step": 43399 + }, + { + "epoch": 0.3767328408607564, + "grad_norm": 0.77734375, + "learning_rate": 0.0012510993095637597, + "loss": 0.0967, + "step": 43400 + }, + { + "epoch": 0.37674152134096056, + "grad_norm": 0.08056640625, + "learning_rate": 0.0012510699694710929, + "loss": 0.0654, + "step": 43401 + }, + { + "epoch": 0.37675020182116475, + "grad_norm": 0.25390625, + "learning_rate": 0.0012510406292132183, + "loss": 0.1113, + "step": 43402 + }, + { + "epoch": 0.3767588823013689, + "grad_norm": 0.486328125, + "learning_rate": 0.0012510112887901684, + "loss": 0.0938, + "step": 43403 + }, + { + "epoch": 0.3767675627815731, + "grad_norm": 0.154296875, + "learning_rate": 0.001250981948201974, + "loss": 0.0752, + "step": 43404 + }, + { + "epoch": 0.3767762432617772, + "grad_norm": 0.7734375, + "learning_rate": 0.0012509526074486685, + "loss": 0.0938, + "step": 43405 + }, + { + "epoch": 0.3767849237419814, + "grad_norm": 0.498046875, + "learning_rate": 0.0012509232665302834, + "loss": 0.1172, + "step": 43406 + }, + { + "epoch": 0.37679360422218555, + "grad_norm": 0.263671875, + "learning_rate": 0.001250893925446851, + "loss": 0.106, + "step": 43407 + }, + { + "epoch": 0.37680228470238974, + "grad_norm": 0.1865234375, + "learning_rate": 0.001250864584198403, + "loss": 0.0981, + "step": 43408 + }, + { + "epoch": 0.3768109651825939, + "grad_norm": 0.16796875, + "learning_rate": 0.0012508352427849717, + "loss": 0.1348, + "step": 43409 + }, + { + "epoch": 0.3768196456627981, + "grad_norm": 0.0966796875, + "learning_rate": 0.0012508059012065896, + "loss": 0.1035, + "step": 43410 + }, + { + "epoch": 0.3768283261430022, + "grad_norm": 0.255859375, + "learning_rate": 0.0012507765594632883, + "loss": 0.0618, + "step": 43411 + }, + { + "epoch": 0.3768370066232064, + "grad_norm": 1.0390625, + "learning_rate": 0.0012507472175551, + "loss": 0.1167, + "step": 43412 + }, + { + "epoch": 0.37684568710341054, + "grad_norm": 0.126953125, + "learning_rate": 0.0012507178754820568, + "loss": 0.1025, + "step": 43413 + }, + { + "epoch": 0.37685436758361474, + "grad_norm": 0.1142578125, + "learning_rate": 0.0012506885332441908, + "loss": 0.0791, + "step": 43414 + }, + { + "epoch": 0.3768630480638189, + "grad_norm": 0.294921875, + "learning_rate": 0.001250659190841534, + "loss": 0.084, + "step": 43415 + }, + { + "epoch": 0.37687172854402307, + "grad_norm": 0.142578125, + "learning_rate": 0.0012506298482741188, + "loss": 0.0791, + "step": 43416 + }, + { + "epoch": 0.3768804090242272, + "grad_norm": 0.1201171875, + "learning_rate": 0.001250600505541977, + "loss": 0.1074, + "step": 43417 + }, + { + "epoch": 0.3768890895044314, + "grad_norm": 0.63671875, + "learning_rate": 0.0012505711626451406, + "loss": 0.1118, + "step": 43418 + }, + { + "epoch": 0.37689776998463553, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012505418195836418, + "loss": 0.0623, + "step": 43419 + }, + { + "epoch": 0.3769064504648397, + "grad_norm": 0.51953125, + "learning_rate": 0.001250512476357513, + "loss": 0.0942, + "step": 43420 + }, + { + "epoch": 0.37691513094504386, + "grad_norm": 0.1123046875, + "learning_rate": 0.0012504831329667855, + "loss": 0.1016, + "step": 43421 + }, + { + "epoch": 0.37692381142524806, + "grad_norm": 0.8125, + "learning_rate": 0.0012504537894114922, + "loss": 0.1328, + "step": 43422 + }, + { + "epoch": 0.3769324919054522, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012504244456916652, + "loss": 0.084, + "step": 43423 + }, + { + "epoch": 0.3769411723856564, + "grad_norm": 0.09423828125, + "learning_rate": 0.001250395101807336, + "loss": 0.1445, + "step": 43424 + }, + { + "epoch": 0.3769498528658605, + "grad_norm": 0.10107421875, + "learning_rate": 0.0012503657577585372, + "loss": 0.0869, + "step": 43425 + }, + { + "epoch": 0.3769585333460647, + "grad_norm": 0.15234375, + "learning_rate": 0.0012503364135453003, + "loss": 0.1104, + "step": 43426 + }, + { + "epoch": 0.37696721382626885, + "grad_norm": 0.12255859375, + "learning_rate": 0.001250307069167658, + "loss": 0.0908, + "step": 43427 + }, + { + "epoch": 0.37697589430647305, + "grad_norm": 0.1201171875, + "learning_rate": 0.001250277724625642, + "loss": 0.1152, + "step": 43428 + }, + { + "epoch": 0.3769845747866772, + "grad_norm": 0.3203125, + "learning_rate": 0.0012502483799192844, + "loss": 0.1211, + "step": 43429 + }, + { + "epoch": 0.3769932552668814, + "grad_norm": 0.447265625, + "learning_rate": 0.0012502190350486178, + "loss": 0.0981, + "step": 43430 + }, + { + "epoch": 0.3770019357470855, + "grad_norm": 0.2578125, + "learning_rate": 0.0012501896900136739, + "loss": 0.1182, + "step": 43431 + }, + { + "epoch": 0.3770106162272897, + "grad_norm": 0.37890625, + "learning_rate": 0.0012501603448144846, + "loss": 0.1143, + "step": 43432 + }, + { + "epoch": 0.37701929670749385, + "grad_norm": 0.640625, + "learning_rate": 0.0012501309994510823, + "loss": 0.0928, + "step": 43433 + }, + { + "epoch": 0.37702797718769804, + "grad_norm": 0.78515625, + "learning_rate": 0.001250101653923499, + "loss": 0.1172, + "step": 43434 + }, + { + "epoch": 0.3770366576679022, + "grad_norm": 0.443359375, + "learning_rate": 0.0012500723082317668, + "loss": 0.0908, + "step": 43435 + }, + { + "epoch": 0.37704533814810637, + "grad_norm": 0.44921875, + "learning_rate": 0.0012500429623759177, + "loss": 0.1309, + "step": 43436 + }, + { + "epoch": 0.3770540186283105, + "grad_norm": 0.1904296875, + "learning_rate": 0.001250013616355984, + "loss": 0.0933, + "step": 43437 + }, + { + "epoch": 0.3770626991085147, + "grad_norm": 0.53125, + "learning_rate": 0.0012499842701719974, + "loss": 0.1094, + "step": 43438 + }, + { + "epoch": 0.37707137958871884, + "grad_norm": 0.2353515625, + "learning_rate": 0.0012499549238239903, + "loss": 0.1299, + "step": 43439 + }, + { + "epoch": 0.37708006006892303, + "grad_norm": 0.50390625, + "learning_rate": 0.0012499255773119951, + "loss": 0.1797, + "step": 43440 + }, + { + "epoch": 0.37708874054912717, + "grad_norm": 0.1171875, + "learning_rate": 0.0012498962306360434, + "loss": 0.0703, + "step": 43441 + }, + { + "epoch": 0.37709742102933136, + "grad_norm": 0.71875, + "learning_rate": 0.0012498668837961676, + "loss": 0.1245, + "step": 43442 + }, + { + "epoch": 0.3771061015095355, + "grad_norm": 0.337890625, + "learning_rate": 0.001249837536792399, + "loss": 0.0903, + "step": 43443 + }, + { + "epoch": 0.3771147819897397, + "grad_norm": 0.216796875, + "learning_rate": 0.0012498081896247708, + "loss": 0.126, + "step": 43444 + }, + { + "epoch": 0.3771234624699438, + "grad_norm": 0.2451171875, + "learning_rate": 0.0012497788422933145, + "loss": 0.0693, + "step": 43445 + }, + { + "epoch": 0.377132142950148, + "grad_norm": 0.25390625, + "learning_rate": 0.0012497494947980626, + "loss": 0.0996, + "step": 43446 + }, + { + "epoch": 0.37714082343035216, + "grad_norm": 0.28125, + "learning_rate": 0.0012497201471390466, + "loss": 0.0981, + "step": 43447 + }, + { + "epoch": 0.37714950391055635, + "grad_norm": 0.4453125, + "learning_rate": 0.0012496907993162986, + "loss": 0.1426, + "step": 43448 + }, + { + "epoch": 0.3771581843907605, + "grad_norm": 0.076171875, + "learning_rate": 0.0012496614513298515, + "loss": 0.0806, + "step": 43449 + }, + { + "epoch": 0.3771668648709647, + "grad_norm": 0.095703125, + "learning_rate": 0.0012496321031797365, + "loss": 0.0913, + "step": 43450 + }, + { + "epoch": 0.3771755453511688, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012496027548659866, + "loss": 0.126, + "step": 43451 + }, + { + "epoch": 0.377184225831373, + "grad_norm": 0.95703125, + "learning_rate": 0.0012495734063886327, + "loss": 0.1094, + "step": 43452 + }, + { + "epoch": 0.37719290631157715, + "grad_norm": 0.263671875, + "learning_rate": 0.001249544057747708, + "loss": 0.0991, + "step": 43453 + }, + { + "epoch": 0.37720158679178134, + "grad_norm": 0.193359375, + "learning_rate": 0.0012495147089432444, + "loss": 0.0776, + "step": 43454 + }, + { + "epoch": 0.3772102672719855, + "grad_norm": 0.09033203125, + "learning_rate": 0.0012494853599752736, + "loss": 0.1035, + "step": 43455 + }, + { + "epoch": 0.37721894775218967, + "grad_norm": 0.0830078125, + "learning_rate": 0.0012494560108438273, + "loss": 0.1367, + "step": 43456 + }, + { + "epoch": 0.3772276282323938, + "grad_norm": 0.392578125, + "learning_rate": 0.0012494266615489386, + "loss": 0.0859, + "step": 43457 + }, + { + "epoch": 0.377236308712598, + "grad_norm": 0.10986328125, + "learning_rate": 0.001249397312090639, + "loss": 0.1309, + "step": 43458 + }, + { + "epoch": 0.37724498919280214, + "grad_norm": 0.0810546875, + "learning_rate": 0.001249367962468961, + "loss": 0.084, + "step": 43459 + }, + { + "epoch": 0.37725366967300633, + "grad_norm": 0.224609375, + "learning_rate": 0.0012493386126839364, + "loss": 0.1064, + "step": 43460 + }, + { + "epoch": 0.37726235015321047, + "grad_norm": 0.10791015625, + "learning_rate": 0.001249309262735597, + "loss": 0.0859, + "step": 43461 + }, + { + "epoch": 0.37727103063341466, + "grad_norm": 0.51953125, + "learning_rate": 0.0012492799126239756, + "loss": 0.0972, + "step": 43462 + }, + { + "epoch": 0.3772797111136188, + "grad_norm": 0.5625, + "learning_rate": 0.0012492505623491038, + "loss": 0.1406, + "step": 43463 + }, + { + "epoch": 0.377288391593823, + "grad_norm": 0.2099609375, + "learning_rate": 0.001249221211911014, + "loss": 0.1055, + "step": 43464 + }, + { + "epoch": 0.37729707207402713, + "grad_norm": 0.1826171875, + "learning_rate": 0.001249191861309738, + "loss": 0.0737, + "step": 43465 + }, + { + "epoch": 0.3773057525542313, + "grad_norm": 0.0927734375, + "learning_rate": 0.0012491625105453077, + "loss": 0.1172, + "step": 43466 + }, + { + "epoch": 0.37731443303443546, + "grad_norm": 0.11572265625, + "learning_rate": 0.0012491331596177558, + "loss": 0.0928, + "step": 43467 + }, + { + "epoch": 0.37732311351463965, + "grad_norm": 0.228515625, + "learning_rate": 0.0012491038085271143, + "loss": 0.0859, + "step": 43468 + }, + { + "epoch": 0.3773317939948438, + "grad_norm": 0.412109375, + "learning_rate": 0.0012490744572734146, + "loss": 0.1104, + "step": 43469 + }, + { + "epoch": 0.377340474475048, + "grad_norm": 0.314453125, + "learning_rate": 0.0012490451058566898, + "loss": 0.1006, + "step": 43470 + }, + { + "epoch": 0.3773491549552521, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012490157542769716, + "loss": 0.0884, + "step": 43471 + }, + { + "epoch": 0.3773578354354563, + "grad_norm": 0.1005859375, + "learning_rate": 0.0012489864025342916, + "loss": 0.0898, + "step": 43472 + }, + { + "epoch": 0.37736651591566045, + "grad_norm": 0.81640625, + "learning_rate": 0.0012489570506286826, + "loss": 0.0972, + "step": 43473 + }, + { + "epoch": 0.37737519639586464, + "grad_norm": 0.15625, + "learning_rate": 0.001248927698560176, + "loss": 0.0986, + "step": 43474 + }, + { + "epoch": 0.3773838768760688, + "grad_norm": 0.6171875, + "learning_rate": 0.0012488983463288046, + "loss": 0.1045, + "step": 43475 + }, + { + "epoch": 0.377392557356273, + "grad_norm": 0.412109375, + "learning_rate": 0.0012488689939346002, + "loss": 0.0933, + "step": 43476 + }, + { + "epoch": 0.3774012378364771, + "grad_norm": 0.33984375, + "learning_rate": 0.0012488396413775948, + "loss": 0.0664, + "step": 43477 + }, + { + "epoch": 0.3774099183166813, + "grad_norm": 0.171875, + "learning_rate": 0.0012488102886578207, + "loss": 0.1035, + "step": 43478 + }, + { + "epoch": 0.37741859879688544, + "grad_norm": 0.2578125, + "learning_rate": 0.0012487809357753102, + "loss": 0.1396, + "step": 43479 + }, + { + "epoch": 0.37742727927708963, + "grad_norm": 0.349609375, + "learning_rate": 0.0012487515827300946, + "loss": 0.1084, + "step": 43480 + }, + { + "epoch": 0.37743595975729377, + "grad_norm": 0.3046875, + "learning_rate": 0.0012487222295222069, + "loss": 0.1367, + "step": 43481 + }, + { + "epoch": 0.37744464023749796, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012486928761516786, + "loss": 0.0977, + "step": 43482 + }, + { + "epoch": 0.3774533207177021, + "grad_norm": 0.1005859375, + "learning_rate": 0.0012486635226185419, + "loss": 0.1455, + "step": 43483 + }, + { + "epoch": 0.3774620011979063, + "grad_norm": 0.251953125, + "learning_rate": 0.0012486341689228293, + "loss": 0.0938, + "step": 43484 + }, + { + "epoch": 0.37747068167811043, + "grad_norm": 0.87109375, + "learning_rate": 0.0012486048150645722, + "loss": 0.0952, + "step": 43485 + }, + { + "epoch": 0.3774793621583146, + "grad_norm": 0.1943359375, + "learning_rate": 0.0012485754610438034, + "loss": 0.0801, + "step": 43486 + }, + { + "epoch": 0.37748804263851876, + "grad_norm": 0.212890625, + "learning_rate": 0.0012485461068605546, + "loss": 0.1221, + "step": 43487 + }, + { + "epoch": 0.37749672311872295, + "grad_norm": 0.4453125, + "learning_rate": 0.0012485167525148582, + "loss": 0.1416, + "step": 43488 + }, + { + "epoch": 0.3775054035989271, + "grad_norm": 0.1884765625, + "learning_rate": 0.001248487398006746, + "loss": 0.0981, + "step": 43489 + }, + { + "epoch": 0.3775140840791313, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012484580433362502, + "loss": 0.0996, + "step": 43490 + }, + { + "epoch": 0.3775227645593354, + "grad_norm": 0.322265625, + "learning_rate": 0.0012484286885034027, + "loss": 0.1216, + "step": 43491 + }, + { + "epoch": 0.3775314450395396, + "grad_norm": 0.134765625, + "learning_rate": 0.0012483993335082363, + "loss": 0.083, + "step": 43492 + }, + { + "epoch": 0.37754012551974375, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012483699783507822, + "loss": 0.0618, + "step": 43493 + }, + { + "epoch": 0.3775488059999479, + "grad_norm": 0.16015625, + "learning_rate": 0.0012483406230310733, + "loss": 0.1279, + "step": 43494 + }, + { + "epoch": 0.3775574864801521, + "grad_norm": 0.11376953125, + "learning_rate": 0.0012483112675491408, + "loss": 0.1084, + "step": 43495 + }, + { + "epoch": 0.3775661669603562, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012482819119050176, + "loss": 0.1396, + "step": 43496 + }, + { + "epoch": 0.3775748474405604, + "grad_norm": 0.408203125, + "learning_rate": 0.0012482525560987355, + "loss": 0.1543, + "step": 43497 + }, + { + "epoch": 0.37758352792076455, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012482232001303267, + "loss": 0.1084, + "step": 43498 + }, + { + "epoch": 0.37759220840096874, + "grad_norm": 0.1123046875, + "learning_rate": 0.0012481938439998232, + "loss": 0.0693, + "step": 43499 + }, + { + "epoch": 0.3776008888811729, + "grad_norm": 0.2373046875, + "learning_rate": 0.0012481644877072571, + "loss": 0.0928, + "step": 43500 + }, + { + "epoch": 0.3776095693613771, + "grad_norm": 0.78125, + "learning_rate": 0.0012481351312526605, + "loss": 0.0967, + "step": 43501 + }, + { + "epoch": 0.3776182498415812, + "grad_norm": 0.328125, + "learning_rate": 0.0012481057746360658, + "loss": 0.1494, + "step": 43502 + }, + { + "epoch": 0.3776269303217854, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012480764178575045, + "loss": 0.0947, + "step": 43503 + }, + { + "epoch": 0.37763561080198954, + "grad_norm": 0.302734375, + "learning_rate": 0.0012480470609170095, + "loss": 0.0908, + "step": 43504 + }, + { + "epoch": 0.37764429128219373, + "grad_norm": 0.458984375, + "learning_rate": 0.001248017703814612, + "loss": 0.1182, + "step": 43505 + }, + { + "epoch": 0.37765297176239787, + "grad_norm": 0.419921875, + "learning_rate": 0.0012479883465503448, + "loss": 0.1152, + "step": 43506 + }, + { + "epoch": 0.37766165224260206, + "grad_norm": 0.12060546875, + "learning_rate": 0.0012479589891242395, + "loss": 0.0781, + "step": 43507 + }, + { + "epoch": 0.3776703327228062, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012479296315363287, + "loss": 0.1416, + "step": 43508 + }, + { + "epoch": 0.3776790132030104, + "grad_norm": 0.28515625, + "learning_rate": 0.0012479002737866443, + "loss": 0.1309, + "step": 43509 + }, + { + "epoch": 0.37768769368321453, + "grad_norm": 0.1201171875, + "learning_rate": 0.0012478709158752184, + "loss": 0.085, + "step": 43510 + }, + { + "epoch": 0.3776963741634187, + "grad_norm": 0.1640625, + "learning_rate": 0.0012478415578020831, + "loss": 0.1064, + "step": 43511 + }, + { + "epoch": 0.37770505464362286, + "grad_norm": 0.2890625, + "learning_rate": 0.0012478121995672707, + "loss": 0.0613, + "step": 43512 + }, + { + "epoch": 0.37771373512382705, + "grad_norm": 0.392578125, + "learning_rate": 0.0012477828411708125, + "loss": 0.0776, + "step": 43513 + }, + { + "epoch": 0.3777224156040312, + "grad_norm": 0.1162109375, + "learning_rate": 0.0012477534826127415, + "loss": 0.1104, + "step": 43514 + }, + { + "epoch": 0.3777310960842354, + "grad_norm": 0.142578125, + "learning_rate": 0.0012477241238930895, + "loss": 0.0781, + "step": 43515 + }, + { + "epoch": 0.3777397765644395, + "grad_norm": 0.138671875, + "learning_rate": 0.0012476947650118888, + "loss": 0.1099, + "step": 43516 + }, + { + "epoch": 0.3777484570446437, + "grad_norm": 0.33984375, + "learning_rate": 0.001247665405969171, + "loss": 0.0723, + "step": 43517 + }, + { + "epoch": 0.37775713752484785, + "grad_norm": 0.142578125, + "learning_rate": 0.0012476360467649687, + "loss": 0.1074, + "step": 43518 + }, + { + "epoch": 0.37776581800505205, + "grad_norm": 0.173828125, + "learning_rate": 0.0012476066873993142, + "loss": 0.0698, + "step": 43519 + }, + { + "epoch": 0.3777744984852562, + "grad_norm": 0.2890625, + "learning_rate": 0.0012475773278722388, + "loss": 0.0815, + "step": 43520 + }, + { + "epoch": 0.3777831789654604, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012475479681837752, + "loss": 0.0796, + "step": 43521 + }, + { + "epoch": 0.3777918594456645, + "grad_norm": 0.359375, + "learning_rate": 0.0012475186083339552, + "loss": 0.123, + "step": 43522 + }, + { + "epoch": 0.3778005399258687, + "grad_norm": 0.203125, + "learning_rate": 0.0012474892483228114, + "loss": 0.0752, + "step": 43523 + }, + { + "epoch": 0.37780922040607284, + "grad_norm": 0.298828125, + "learning_rate": 0.0012474598881503753, + "loss": 0.0928, + "step": 43524 + }, + { + "epoch": 0.37781790088627704, + "grad_norm": 0.24609375, + "learning_rate": 0.0012474305278166791, + "loss": 0.0879, + "step": 43525 + }, + { + "epoch": 0.3778265813664812, + "grad_norm": 0.41796875, + "learning_rate": 0.0012474011673217554, + "loss": 0.0996, + "step": 43526 + }, + { + "epoch": 0.37783526184668537, + "grad_norm": 0.1513671875, + "learning_rate": 0.001247371806665636, + "loss": 0.0903, + "step": 43527 + }, + { + "epoch": 0.3778439423268895, + "grad_norm": 0.400390625, + "learning_rate": 0.0012473424458483528, + "loss": 0.0947, + "step": 43528 + }, + { + "epoch": 0.3778526228070937, + "grad_norm": 0.103515625, + "learning_rate": 0.0012473130848699384, + "loss": 0.1045, + "step": 43529 + }, + { + "epoch": 0.37786130328729783, + "grad_norm": 0.21875, + "learning_rate": 0.0012472837237304244, + "loss": 0.0679, + "step": 43530 + }, + { + "epoch": 0.377869983767502, + "grad_norm": 0.2265625, + "learning_rate": 0.0012472543624298432, + "loss": 0.0977, + "step": 43531 + }, + { + "epoch": 0.37787866424770616, + "grad_norm": 0.87890625, + "learning_rate": 0.0012472250009682268, + "loss": 0.1475, + "step": 43532 + }, + { + "epoch": 0.37788734472791036, + "grad_norm": 0.1005859375, + "learning_rate": 0.0012471956393456075, + "loss": 0.0864, + "step": 43533 + }, + { + "epoch": 0.3778960252081145, + "grad_norm": 0.1669921875, + "learning_rate": 0.0012471662775620168, + "loss": 0.0713, + "step": 43534 + }, + { + "epoch": 0.3779047056883187, + "grad_norm": 0.396484375, + "learning_rate": 0.0012471369156174877, + "loss": 0.1299, + "step": 43535 + }, + { + "epoch": 0.3779133861685228, + "grad_norm": 0.1611328125, + "learning_rate": 0.001247107553512052, + "loss": 0.0918, + "step": 43536 + }, + { + "epoch": 0.377922066648727, + "grad_norm": 0.392578125, + "learning_rate": 0.0012470781912457417, + "loss": 0.0903, + "step": 43537 + }, + { + "epoch": 0.37793074712893115, + "grad_norm": 0.84375, + "learning_rate": 0.0012470488288185884, + "loss": 0.1123, + "step": 43538 + }, + { + "epoch": 0.37793942760913535, + "grad_norm": 0.14453125, + "learning_rate": 0.001247019466230625, + "loss": 0.082, + "step": 43539 + }, + { + "epoch": 0.3779481080893395, + "grad_norm": 0.19921875, + "learning_rate": 0.0012469901034818836, + "loss": 0.1138, + "step": 43540 + }, + { + "epoch": 0.3779567885695437, + "grad_norm": 0.5703125, + "learning_rate": 0.0012469607405723958, + "loss": 0.1162, + "step": 43541 + }, + { + "epoch": 0.3779654690497478, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012469313775021937, + "loss": 0.1133, + "step": 43542 + }, + { + "epoch": 0.377974149529952, + "grad_norm": 0.39453125, + "learning_rate": 0.0012469020142713099, + "loss": 0.0957, + "step": 43543 + }, + { + "epoch": 0.37798283001015615, + "grad_norm": 0.1611328125, + "learning_rate": 0.001246872650879776, + "loss": 0.1021, + "step": 43544 + }, + { + "epoch": 0.37799151049036034, + "grad_norm": 0.5234375, + "learning_rate": 0.0012468432873276247, + "loss": 0.1055, + "step": 43545 + }, + { + "epoch": 0.3780001909705645, + "grad_norm": 0.2333984375, + "learning_rate": 0.0012468139236148877, + "loss": 0.1123, + "step": 43546 + }, + { + "epoch": 0.37800887145076867, + "grad_norm": 0.1171875, + "learning_rate": 0.001246784559741597, + "loss": 0.123, + "step": 43547 + }, + { + "epoch": 0.3780175519309728, + "grad_norm": 0.447265625, + "learning_rate": 0.0012467551957077855, + "loss": 0.1123, + "step": 43548 + }, + { + "epoch": 0.378026232411177, + "grad_norm": 0.1181640625, + "learning_rate": 0.0012467258315134841, + "loss": 0.0869, + "step": 43549 + }, + { + "epoch": 0.37803491289138114, + "grad_norm": 0.5234375, + "learning_rate": 0.001246696467158726, + "loss": 0.0957, + "step": 43550 + }, + { + "epoch": 0.37804359337158533, + "grad_norm": 0.08984375, + "learning_rate": 0.0012466671026435422, + "loss": 0.0767, + "step": 43551 + }, + { + "epoch": 0.37805227385178947, + "grad_norm": 1.109375, + "learning_rate": 0.0012466377379679661, + "loss": 0.1953, + "step": 43552 + }, + { + "epoch": 0.37806095433199366, + "grad_norm": 0.11328125, + "learning_rate": 0.0012466083731320287, + "loss": 0.1074, + "step": 43553 + }, + { + "epoch": 0.3780696348121978, + "grad_norm": 0.11572265625, + "learning_rate": 0.001246579008135763, + "loss": 0.104, + "step": 43554 + }, + { + "epoch": 0.378078315292402, + "grad_norm": 0.240234375, + "learning_rate": 0.0012465496429792005, + "loss": 0.0928, + "step": 43555 + }, + { + "epoch": 0.3780869957726061, + "grad_norm": 0.14453125, + "learning_rate": 0.0012465202776623734, + "loss": 0.1035, + "step": 43556 + }, + { + "epoch": 0.3780956762528103, + "grad_norm": 0.216796875, + "learning_rate": 0.001246490912185314, + "loss": 0.1016, + "step": 43557 + }, + { + "epoch": 0.37810435673301446, + "grad_norm": 0.28515625, + "learning_rate": 0.0012464615465480546, + "loss": 0.1055, + "step": 43558 + }, + { + "epoch": 0.37811303721321865, + "grad_norm": 0.416015625, + "learning_rate": 0.001246432180750627, + "loss": 0.0776, + "step": 43559 + }, + { + "epoch": 0.3781217176934228, + "grad_norm": 0.302734375, + "learning_rate": 0.0012464028147930629, + "loss": 0.0957, + "step": 43560 + }, + { + "epoch": 0.378130398173627, + "grad_norm": 0.197265625, + "learning_rate": 0.0012463734486753953, + "loss": 0.1221, + "step": 43561 + }, + { + "epoch": 0.3781390786538311, + "grad_norm": 0.08740234375, + "learning_rate": 0.0012463440823976557, + "loss": 0.1309, + "step": 43562 + }, + { + "epoch": 0.3781477591340353, + "grad_norm": 0.13671875, + "learning_rate": 0.0012463147159598763, + "loss": 0.1123, + "step": 43563 + }, + { + "epoch": 0.37815643961423945, + "grad_norm": 0.111328125, + "learning_rate": 0.0012462853493620896, + "loss": 0.106, + "step": 43564 + }, + { + "epoch": 0.37816512009444364, + "grad_norm": 0.22265625, + "learning_rate": 0.0012462559826043275, + "loss": 0.0996, + "step": 43565 + }, + { + "epoch": 0.3781738005746478, + "grad_norm": 0.27734375, + "learning_rate": 0.001246226615686622, + "loss": 0.084, + "step": 43566 + }, + { + "epoch": 0.37818248105485197, + "grad_norm": 0.2431640625, + "learning_rate": 0.0012461972486090052, + "loss": 0.0986, + "step": 43567 + }, + { + "epoch": 0.3781911615350561, + "grad_norm": 0.232421875, + "learning_rate": 0.0012461678813715095, + "loss": 0.1055, + "step": 43568 + }, + { + "epoch": 0.3781998420152603, + "grad_norm": 0.09765625, + "learning_rate": 0.0012461385139741663, + "loss": 0.1113, + "step": 43569 + }, + { + "epoch": 0.37820852249546444, + "grad_norm": 0.08837890625, + "learning_rate": 0.0012461091464170085, + "loss": 0.1035, + "step": 43570 + }, + { + "epoch": 0.37821720297566863, + "grad_norm": 0.1484375, + "learning_rate": 0.001246079778700068, + "loss": 0.0708, + "step": 43571 + }, + { + "epoch": 0.37822588345587277, + "grad_norm": 0.11962890625, + "learning_rate": 0.0012460504108233767, + "loss": 0.0957, + "step": 43572 + }, + { + "epoch": 0.37823456393607696, + "grad_norm": 0.158203125, + "learning_rate": 0.0012460210427869666, + "loss": 0.1348, + "step": 43573 + }, + { + "epoch": 0.3782432444162811, + "grad_norm": 0.38671875, + "learning_rate": 0.001245991674590871, + "loss": 0.0791, + "step": 43574 + }, + { + "epoch": 0.3782519248964853, + "grad_norm": 0.4921875, + "learning_rate": 0.0012459623062351201, + "loss": 0.1128, + "step": 43575 + }, + { + "epoch": 0.37826060537668943, + "grad_norm": 0.376953125, + "learning_rate": 0.0012459329377197477, + "loss": 0.0537, + "step": 43576 + }, + { + "epoch": 0.3782692858568936, + "grad_norm": 0.6015625, + "learning_rate": 0.0012459035690447848, + "loss": 0.123, + "step": 43577 + }, + { + "epoch": 0.37827796633709776, + "grad_norm": 0.310546875, + "learning_rate": 0.001245874200210264, + "loss": 0.084, + "step": 43578 + }, + { + "epoch": 0.37828664681730195, + "grad_norm": 0.15625, + "learning_rate": 0.0012458448312162177, + "loss": 0.1143, + "step": 43579 + }, + { + "epoch": 0.3782953272975061, + "grad_norm": 0.1240234375, + "learning_rate": 0.0012458154620626776, + "loss": 0.0728, + "step": 43580 + }, + { + "epoch": 0.3783040077777103, + "grad_norm": 0.16796875, + "learning_rate": 0.0012457860927496756, + "loss": 0.1084, + "step": 43581 + }, + { + "epoch": 0.3783126882579144, + "grad_norm": 0.302734375, + "learning_rate": 0.0012457567232772443, + "loss": 0.0864, + "step": 43582 + }, + { + "epoch": 0.3783213687381186, + "grad_norm": 1.0546875, + "learning_rate": 0.0012457273536454158, + "loss": 0.1016, + "step": 43583 + }, + { + "epoch": 0.37833004921832275, + "grad_norm": 0.0869140625, + "learning_rate": 0.0012456979838542217, + "loss": 0.0913, + "step": 43584 + }, + { + "epoch": 0.37833872969852694, + "grad_norm": 0.12451171875, + "learning_rate": 0.0012456686139036947, + "loss": 0.0981, + "step": 43585 + }, + { + "epoch": 0.3783474101787311, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012456392437938666, + "loss": 0.127, + "step": 43586 + }, + { + "epoch": 0.3783560906589353, + "grad_norm": 0.69921875, + "learning_rate": 0.0012456098735247696, + "loss": 0.1025, + "step": 43587 + }, + { + "epoch": 0.3783647711391394, + "grad_norm": 0.400390625, + "learning_rate": 0.001245580503096436, + "loss": 0.1289, + "step": 43588 + }, + { + "epoch": 0.3783734516193436, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012455511325088976, + "loss": 0.0762, + "step": 43589 + }, + { + "epoch": 0.37838213209954774, + "grad_norm": 0.1416015625, + "learning_rate": 0.0012455217617621866, + "loss": 0.1162, + "step": 43590 + }, + { + "epoch": 0.37839081257975193, + "grad_norm": 0.203125, + "learning_rate": 0.0012454923908563353, + "loss": 0.0952, + "step": 43591 + }, + { + "epoch": 0.37839949305995607, + "grad_norm": 0.5078125, + "learning_rate": 0.0012454630197913756, + "loss": 0.0938, + "step": 43592 + }, + { + "epoch": 0.37840817354016026, + "grad_norm": 0.27734375, + "learning_rate": 0.0012454336485673398, + "loss": 0.1152, + "step": 43593 + }, + { + "epoch": 0.3784168540203644, + "grad_norm": 0.265625, + "learning_rate": 0.0012454042771842602, + "loss": 0.0981, + "step": 43594 + }, + { + "epoch": 0.3784255345005686, + "grad_norm": 0.2890625, + "learning_rate": 0.0012453749056421683, + "loss": 0.0903, + "step": 43595 + }, + { + "epoch": 0.37843421498077273, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012453455339410968, + "loss": 0.1079, + "step": 43596 + }, + { + "epoch": 0.3784428954609769, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012453161620810777, + "loss": 0.0859, + "step": 43597 + }, + { + "epoch": 0.37845157594118106, + "grad_norm": 0.34765625, + "learning_rate": 0.001245286790062143, + "loss": 0.1138, + "step": 43598 + }, + { + "epoch": 0.37846025642138525, + "grad_norm": 0.56640625, + "learning_rate": 0.0012452574178843246, + "loss": 0.1069, + "step": 43599 + }, + { + "epoch": 0.3784689369015894, + "grad_norm": 0.173828125, + "learning_rate": 0.0012452280455476548, + "loss": 0.0957, + "step": 43600 + }, + { + "epoch": 0.3784776173817936, + "grad_norm": 0.404296875, + "learning_rate": 0.001245198673052166, + "loss": 0.0977, + "step": 43601 + }, + { + "epoch": 0.3784862978619977, + "grad_norm": 0.205078125, + "learning_rate": 0.0012451693003978901, + "loss": 0.126, + "step": 43602 + }, + { + "epoch": 0.3784949783422019, + "grad_norm": 0.13671875, + "learning_rate": 0.001245139927584859, + "loss": 0.1064, + "step": 43603 + }, + { + "epoch": 0.37850365882240605, + "grad_norm": 0.087890625, + "learning_rate": 0.0012451105546131055, + "loss": 0.0869, + "step": 43604 + }, + { + "epoch": 0.37851233930261025, + "grad_norm": 0.322265625, + "learning_rate": 0.001245081181482661, + "loss": 0.0918, + "step": 43605 + }, + { + "epoch": 0.3785210197828144, + "grad_norm": 0.1279296875, + "learning_rate": 0.001245051808193558, + "loss": 0.085, + "step": 43606 + }, + { + "epoch": 0.3785297002630186, + "grad_norm": 0.69921875, + "learning_rate": 0.0012450224347458285, + "loss": 0.0762, + "step": 43607 + }, + { + "epoch": 0.3785383807432227, + "grad_norm": 0.4453125, + "learning_rate": 0.0012449930611395047, + "loss": 0.0801, + "step": 43608 + }, + { + "epoch": 0.3785470612234269, + "grad_norm": 0.0927734375, + "learning_rate": 0.0012449636873746184, + "loss": 0.1436, + "step": 43609 + }, + { + "epoch": 0.37855574170363104, + "grad_norm": 0.44140625, + "learning_rate": 0.0012449343134512023, + "loss": 0.1157, + "step": 43610 + }, + { + "epoch": 0.37856442218383524, + "grad_norm": 0.3828125, + "learning_rate": 0.0012449049393692879, + "loss": 0.0903, + "step": 43611 + }, + { + "epoch": 0.3785731026640394, + "grad_norm": 0.67578125, + "learning_rate": 0.001244875565128908, + "loss": 0.1055, + "step": 43612 + }, + { + "epoch": 0.37858178314424357, + "grad_norm": 0.3515625, + "learning_rate": 0.0012448461907300942, + "loss": 0.1108, + "step": 43613 + }, + { + "epoch": 0.3785904636244477, + "grad_norm": 0.458984375, + "learning_rate": 0.0012448168161728788, + "loss": 0.1309, + "step": 43614 + }, + { + "epoch": 0.3785991441046519, + "grad_norm": 0.1689453125, + "learning_rate": 0.001244787441457294, + "loss": 0.0986, + "step": 43615 + }, + { + "epoch": 0.37860782458485603, + "grad_norm": 0.80078125, + "learning_rate": 0.0012447580665833715, + "loss": 0.1104, + "step": 43616 + }, + { + "epoch": 0.37861650506506017, + "grad_norm": 0.12451171875, + "learning_rate": 0.0012447286915511439, + "loss": 0.0781, + "step": 43617 + }, + { + "epoch": 0.37862518554526436, + "grad_norm": 0.36328125, + "learning_rate": 0.0012446993163606436, + "loss": 0.1099, + "step": 43618 + }, + { + "epoch": 0.3786338660254685, + "grad_norm": 0.119140625, + "learning_rate": 0.0012446699410119015, + "loss": 0.1216, + "step": 43619 + }, + { + "epoch": 0.3786425465056727, + "grad_norm": 0.1728515625, + "learning_rate": 0.0012446405655049512, + "loss": 0.1016, + "step": 43620 + }, + { + "epoch": 0.37865122698587683, + "grad_norm": 0.306640625, + "learning_rate": 0.0012446111898398236, + "loss": 0.1074, + "step": 43621 + }, + { + "epoch": 0.378659907466081, + "grad_norm": 0.1669921875, + "learning_rate": 0.001244581814016552, + "loss": 0.0913, + "step": 43622 + }, + { + "epoch": 0.37866858794628516, + "grad_norm": 0.42578125, + "learning_rate": 0.0012445524380351675, + "loss": 0.126, + "step": 43623 + }, + { + "epoch": 0.37867726842648936, + "grad_norm": 0.458984375, + "learning_rate": 0.0012445230618957027, + "loss": 0.1094, + "step": 43624 + }, + { + "epoch": 0.3786859489066935, + "grad_norm": 0.1953125, + "learning_rate": 0.0012444936855981897, + "loss": 0.0854, + "step": 43625 + }, + { + "epoch": 0.3786946293868977, + "grad_norm": 0.494140625, + "learning_rate": 0.0012444643091426607, + "loss": 0.0771, + "step": 43626 + }, + { + "epoch": 0.3787033098671018, + "grad_norm": 0.142578125, + "learning_rate": 0.0012444349325291474, + "loss": 0.1035, + "step": 43627 + }, + { + "epoch": 0.378711990347306, + "grad_norm": 0.1416015625, + "learning_rate": 0.0012444055557576826, + "loss": 0.1001, + "step": 43628 + }, + { + "epoch": 0.37872067082751015, + "grad_norm": 0.11279296875, + "learning_rate": 0.0012443761788282976, + "loss": 0.0869, + "step": 43629 + }, + { + "epoch": 0.37872935130771435, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012443468017410248, + "loss": 0.0771, + "step": 43630 + }, + { + "epoch": 0.3787380317879185, + "grad_norm": 0.4453125, + "learning_rate": 0.0012443174244958972, + "loss": 0.1455, + "step": 43631 + }, + { + "epoch": 0.3787467122681227, + "grad_norm": 0.22265625, + "learning_rate": 0.0012442880470929462, + "loss": 0.1167, + "step": 43632 + }, + { + "epoch": 0.3787553927483268, + "grad_norm": 0.169921875, + "learning_rate": 0.0012442586695322036, + "loss": 0.0791, + "step": 43633 + }, + { + "epoch": 0.378764073228531, + "grad_norm": 0.78515625, + "learning_rate": 0.001244229291813702, + "loss": 0.1201, + "step": 43634 + }, + { + "epoch": 0.37877275370873514, + "grad_norm": 0.197265625, + "learning_rate": 0.0012441999139374733, + "loss": 0.0732, + "step": 43635 + }, + { + "epoch": 0.37878143418893934, + "grad_norm": 0.10107421875, + "learning_rate": 0.00124417053590355, + "loss": 0.0767, + "step": 43636 + }, + { + "epoch": 0.3787901146691435, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012441411577119638, + "loss": 0.0967, + "step": 43637 + }, + { + "epoch": 0.37879879514934767, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012441117793627469, + "loss": 0.1064, + "step": 43638 + }, + { + "epoch": 0.3788074756295518, + "grad_norm": 0.0751953125, + "learning_rate": 0.0012440824008559314, + "loss": 0.0684, + "step": 43639 + }, + { + "epoch": 0.378816156109756, + "grad_norm": 0.337890625, + "learning_rate": 0.0012440530221915498, + "loss": 0.0874, + "step": 43640 + }, + { + "epoch": 0.37882483658996013, + "grad_norm": 0.4765625, + "learning_rate": 0.0012440236433696344, + "loss": 0.1621, + "step": 43641 + }, + { + "epoch": 0.3788335170701643, + "grad_norm": 0.75390625, + "learning_rate": 0.0012439942643902163, + "loss": 0.0967, + "step": 43642 + }, + { + "epoch": 0.37884219755036846, + "grad_norm": 0.28125, + "learning_rate": 0.0012439648852533283, + "loss": 0.0679, + "step": 43643 + }, + { + "epoch": 0.37885087803057266, + "grad_norm": 0.3046875, + "learning_rate": 0.0012439355059590026, + "loss": 0.1709, + "step": 43644 + }, + { + "epoch": 0.3788595585107768, + "grad_norm": 0.298828125, + "learning_rate": 0.0012439061265072712, + "loss": 0.1143, + "step": 43645 + }, + { + "epoch": 0.378868238990981, + "grad_norm": 0.1533203125, + "learning_rate": 0.001243876746898166, + "loss": 0.1104, + "step": 43646 + }, + { + "epoch": 0.3788769194711851, + "grad_norm": 0.28515625, + "learning_rate": 0.00124384736713172, + "loss": 0.1406, + "step": 43647 + }, + { + "epoch": 0.3788855999513893, + "grad_norm": 0.248046875, + "learning_rate": 0.001243817987207964, + "loss": 0.127, + "step": 43648 + }, + { + "epoch": 0.37889428043159346, + "grad_norm": 0.546875, + "learning_rate": 0.0012437886071269308, + "loss": 0.1045, + "step": 43649 + }, + { + "epoch": 0.37890296091179765, + "grad_norm": 0.2412109375, + "learning_rate": 0.0012437592268886531, + "loss": 0.1069, + "step": 43650 + }, + { + "epoch": 0.3789116413920018, + "grad_norm": 0.099609375, + "learning_rate": 0.0012437298464931619, + "loss": 0.1201, + "step": 43651 + }, + { + "epoch": 0.378920321872206, + "grad_norm": 0.369140625, + "learning_rate": 0.0012437004659404903, + "loss": 0.0601, + "step": 43652 + }, + { + "epoch": 0.3789290023524101, + "grad_norm": 0.08544921875, + "learning_rate": 0.0012436710852306695, + "loss": 0.0781, + "step": 43653 + }, + { + "epoch": 0.3789376828326143, + "grad_norm": 0.193359375, + "learning_rate": 0.001243641704363733, + "loss": 0.0762, + "step": 43654 + }, + { + "epoch": 0.37894636331281845, + "grad_norm": 0.39453125, + "learning_rate": 0.0012436123233397114, + "loss": 0.1016, + "step": 43655 + }, + { + "epoch": 0.37895504379302264, + "grad_norm": 0.412109375, + "learning_rate": 0.0012435829421586378, + "loss": 0.125, + "step": 43656 + }, + { + "epoch": 0.3789637242732268, + "grad_norm": 0.2412109375, + "learning_rate": 0.001243553560820544, + "loss": 0.1055, + "step": 43657 + }, + { + "epoch": 0.37897240475343097, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012435241793254622, + "loss": 0.0811, + "step": 43658 + }, + { + "epoch": 0.3789810852336351, + "grad_norm": 0.08935546875, + "learning_rate": 0.0012434947976734242, + "loss": 0.0967, + "step": 43659 + }, + { + "epoch": 0.3789897657138393, + "grad_norm": 0.18359375, + "learning_rate": 0.0012434654158644627, + "loss": 0.1445, + "step": 43660 + }, + { + "epoch": 0.37899844619404344, + "grad_norm": 0.416015625, + "learning_rate": 0.0012434360338986097, + "loss": 0.1338, + "step": 43661 + }, + { + "epoch": 0.37900712667424763, + "grad_norm": 0.16015625, + "learning_rate": 0.0012434066517758972, + "loss": 0.0713, + "step": 43662 + }, + { + "epoch": 0.37901580715445177, + "grad_norm": 0.279296875, + "learning_rate": 0.0012433772694963572, + "loss": 0.0791, + "step": 43663 + }, + { + "epoch": 0.37902448763465596, + "grad_norm": 0.4609375, + "learning_rate": 0.0012433478870600221, + "loss": 0.0889, + "step": 43664 + }, + { + "epoch": 0.3790331681148601, + "grad_norm": 0.796875, + "learning_rate": 0.001243318504466924, + "loss": 0.1177, + "step": 43665 + }, + { + "epoch": 0.3790418485950643, + "grad_norm": 0.111328125, + "learning_rate": 0.0012432891217170948, + "loss": 0.1016, + "step": 43666 + }, + { + "epoch": 0.3790505290752684, + "grad_norm": 0.09130859375, + "learning_rate": 0.0012432597388105666, + "loss": 0.1133, + "step": 43667 + }, + { + "epoch": 0.3790592095554726, + "grad_norm": 0.09423828125, + "learning_rate": 0.0012432303557473718, + "loss": 0.0874, + "step": 43668 + }, + { + "epoch": 0.37906789003567676, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012432009725275423, + "loss": 0.1094, + "step": 43669 + }, + { + "epoch": 0.37907657051588095, + "grad_norm": 0.314453125, + "learning_rate": 0.0012431715891511107, + "loss": 0.1133, + "step": 43670 + }, + { + "epoch": 0.3790852509960851, + "grad_norm": 0.17578125, + "learning_rate": 0.0012431422056181087, + "loss": 0.0889, + "step": 43671 + }, + { + "epoch": 0.3790939314762893, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012431128219285684, + "loss": 0.0977, + "step": 43672 + }, + { + "epoch": 0.3791026119564934, + "grad_norm": 0.1064453125, + "learning_rate": 0.0012430834380825222, + "loss": 0.0564, + "step": 43673 + }, + { + "epoch": 0.3791112924366976, + "grad_norm": 0.30078125, + "learning_rate": 0.001243054054080002, + "loss": 0.1055, + "step": 43674 + }, + { + "epoch": 0.37911997291690175, + "grad_norm": 0.11083984375, + "learning_rate": 0.0012430246699210402, + "loss": 0.0859, + "step": 43675 + }, + { + "epoch": 0.37912865339710594, + "grad_norm": 0.197265625, + "learning_rate": 0.0012429952856056687, + "loss": 0.123, + "step": 43676 + }, + { + "epoch": 0.3791373338773101, + "grad_norm": 0.10400390625, + "learning_rate": 0.0012429659011339196, + "loss": 0.0811, + "step": 43677 + }, + { + "epoch": 0.37914601435751427, + "grad_norm": 0.1943359375, + "learning_rate": 0.0012429365165058251, + "loss": 0.1309, + "step": 43678 + }, + { + "epoch": 0.3791546948377184, + "grad_norm": 0.5859375, + "learning_rate": 0.0012429071317214175, + "loss": 0.104, + "step": 43679 + }, + { + "epoch": 0.3791633753179226, + "grad_norm": 0.75, + "learning_rate": 0.001242877746780729, + "loss": 0.0918, + "step": 43680 + }, + { + "epoch": 0.37917205579812674, + "grad_norm": 0.455078125, + "learning_rate": 0.0012428483616837912, + "loss": 0.0972, + "step": 43681 + }, + { + "epoch": 0.37918073627833093, + "grad_norm": 0.271484375, + "learning_rate": 0.0012428189764306366, + "loss": 0.1055, + "step": 43682 + }, + { + "epoch": 0.37918941675853507, + "grad_norm": 0.326171875, + "learning_rate": 0.0012427895910212974, + "loss": 0.0801, + "step": 43683 + }, + { + "epoch": 0.37919809723873926, + "grad_norm": 0.25, + "learning_rate": 0.0012427602054558059, + "loss": 0.1025, + "step": 43684 + }, + { + "epoch": 0.3792067777189434, + "grad_norm": 0.734375, + "learning_rate": 0.0012427308197341936, + "loss": 0.0947, + "step": 43685 + }, + { + "epoch": 0.3792154581991476, + "grad_norm": 0.234375, + "learning_rate": 0.0012427014338564932, + "loss": 0.1348, + "step": 43686 + }, + { + "epoch": 0.37922413867935173, + "grad_norm": 0.0908203125, + "learning_rate": 0.0012426720478227364, + "loss": 0.0923, + "step": 43687 + }, + { + "epoch": 0.3792328191595559, + "grad_norm": 0.212890625, + "learning_rate": 0.0012426426616329559, + "loss": 0.0859, + "step": 43688 + }, + { + "epoch": 0.37924149963976006, + "grad_norm": 0.216796875, + "learning_rate": 0.0012426132752871835, + "loss": 0.0962, + "step": 43689 + }, + { + "epoch": 0.37925018011996425, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012425838887854513, + "loss": 0.1318, + "step": 43690 + }, + { + "epoch": 0.3792588606001684, + "grad_norm": 0.400390625, + "learning_rate": 0.0012425545021277913, + "loss": 0.1152, + "step": 43691 + }, + { + "epoch": 0.3792675410803726, + "grad_norm": 0.333984375, + "learning_rate": 0.0012425251153142362, + "loss": 0.0889, + "step": 43692 + }, + { + "epoch": 0.3792762215605767, + "grad_norm": 0.1494140625, + "learning_rate": 0.0012424957283448178, + "loss": 0.0874, + "step": 43693 + }, + { + "epoch": 0.3792849020407809, + "grad_norm": 1.09375, + "learning_rate": 0.0012424663412195678, + "loss": 0.0854, + "step": 43694 + }, + { + "epoch": 0.37929358252098505, + "grad_norm": 0.361328125, + "learning_rate": 0.0012424369539385189, + "loss": 0.1104, + "step": 43695 + }, + { + "epoch": 0.37930226300118924, + "grad_norm": 0.1513671875, + "learning_rate": 0.0012424075665017035, + "loss": 0.0977, + "step": 43696 + }, + { + "epoch": 0.3793109434813934, + "grad_norm": 0.51171875, + "learning_rate": 0.0012423781789091528, + "loss": 0.1445, + "step": 43697 + }, + { + "epoch": 0.3793196239615976, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012423487911608995, + "loss": 0.1016, + "step": 43698 + }, + { + "epoch": 0.3793283044418017, + "grad_norm": 0.1591796875, + "learning_rate": 0.001242319403256976, + "loss": 0.1143, + "step": 43699 + }, + { + "epoch": 0.3793369849220059, + "grad_norm": 1.4921875, + "learning_rate": 0.0012422900151974141, + "loss": 0.2656, + "step": 43700 + }, + { + "epoch": 0.37934566540221004, + "grad_norm": 0.494140625, + "learning_rate": 0.0012422606269822459, + "loss": 0.0771, + "step": 43701 + }, + { + "epoch": 0.37935434588241423, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012422312386115037, + "loss": 0.1055, + "step": 43702 + }, + { + "epoch": 0.37936302636261837, + "grad_norm": 0.515625, + "learning_rate": 0.0012422018500852195, + "loss": 0.1426, + "step": 43703 + }, + { + "epoch": 0.37937170684282256, + "grad_norm": 0.10009765625, + "learning_rate": 0.0012421724614034254, + "loss": 0.0923, + "step": 43704 + }, + { + "epoch": 0.3793803873230267, + "grad_norm": 0.19921875, + "learning_rate": 0.0012421430725661536, + "loss": 0.1279, + "step": 43705 + }, + { + "epoch": 0.3793890678032309, + "grad_norm": 0.0849609375, + "learning_rate": 0.0012421136835734361, + "loss": 0.0894, + "step": 43706 + }, + { + "epoch": 0.37939774828343503, + "grad_norm": 0.12451171875, + "learning_rate": 0.0012420842944253054, + "loss": 0.0913, + "step": 43707 + }, + { + "epoch": 0.3794064287636392, + "grad_norm": 0.22265625, + "learning_rate": 0.0012420549051217937, + "loss": 0.1147, + "step": 43708 + }, + { + "epoch": 0.37941510924384336, + "grad_norm": 0.6953125, + "learning_rate": 0.0012420255156629328, + "loss": 0.0859, + "step": 43709 + }, + { + "epoch": 0.37942378972404756, + "grad_norm": 0.640625, + "learning_rate": 0.001241996126048755, + "loss": 0.0908, + "step": 43710 + }, + { + "epoch": 0.3794324702042517, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012419667362792918, + "loss": 0.0889, + "step": 43711 + }, + { + "epoch": 0.3794411506844559, + "grad_norm": 0.26171875, + "learning_rate": 0.0012419373463545763, + "loss": 0.0957, + "step": 43712 + }, + { + "epoch": 0.37944983116466, + "grad_norm": 0.4921875, + "learning_rate": 0.0012419079562746404, + "loss": 0.1484, + "step": 43713 + }, + { + "epoch": 0.3794585116448642, + "grad_norm": 0.1298828125, + "learning_rate": 0.001241878566039516, + "loss": 0.0654, + "step": 43714 + }, + { + "epoch": 0.37946719212506835, + "grad_norm": 0.08056640625, + "learning_rate": 0.0012418491756492351, + "loss": 0.0781, + "step": 43715 + }, + { + "epoch": 0.37947587260527255, + "grad_norm": 0.22265625, + "learning_rate": 0.0012418197851038304, + "loss": 0.1104, + "step": 43716 + }, + { + "epoch": 0.3794845530854767, + "grad_norm": 1.109375, + "learning_rate": 0.0012417903944033336, + "loss": 0.1328, + "step": 43717 + }, + { + "epoch": 0.3794932335656809, + "grad_norm": 0.322265625, + "learning_rate": 0.001241761003547777, + "loss": 0.1045, + "step": 43718 + }, + { + "epoch": 0.379501914045885, + "grad_norm": 0.2099609375, + "learning_rate": 0.0012417316125371927, + "loss": 0.0947, + "step": 43719 + }, + { + "epoch": 0.3795105945260892, + "grad_norm": 0.259765625, + "learning_rate": 0.0012417022213716126, + "loss": 0.0962, + "step": 43720 + }, + { + "epoch": 0.37951927500629334, + "grad_norm": 0.11962890625, + "learning_rate": 0.0012416728300510692, + "loss": 0.0874, + "step": 43721 + }, + { + "epoch": 0.37952795548649754, + "grad_norm": 0.515625, + "learning_rate": 0.0012416434385755946, + "loss": 0.0645, + "step": 43722 + }, + { + "epoch": 0.3795366359667017, + "grad_norm": 0.265625, + "learning_rate": 0.001241614046945221, + "loss": 0.0913, + "step": 43723 + }, + { + "epoch": 0.37954531644690587, + "grad_norm": 0.474609375, + "learning_rate": 0.0012415846551599799, + "loss": 0.084, + "step": 43724 + }, + { + "epoch": 0.37955399692711, + "grad_norm": 0.13671875, + "learning_rate": 0.0012415552632199043, + "loss": 0.0928, + "step": 43725 + }, + { + "epoch": 0.3795626774073142, + "grad_norm": 0.12060546875, + "learning_rate": 0.0012415258711250261, + "loss": 0.1035, + "step": 43726 + }, + { + "epoch": 0.37957135788751833, + "grad_norm": 0.19921875, + "learning_rate": 0.001241496478875377, + "loss": 0.1143, + "step": 43727 + }, + { + "epoch": 0.3795800383677225, + "grad_norm": 0.458984375, + "learning_rate": 0.0012414670864709899, + "loss": 0.1182, + "step": 43728 + }, + { + "epoch": 0.37958871884792666, + "grad_norm": 0.88671875, + "learning_rate": 0.001241437693911896, + "loss": 0.1074, + "step": 43729 + }, + { + "epoch": 0.37959739932813086, + "grad_norm": 0.271484375, + "learning_rate": 0.0012414083011981283, + "loss": 0.0981, + "step": 43730 + }, + { + "epoch": 0.379606079808335, + "grad_norm": 0.10498046875, + "learning_rate": 0.001241378908329719, + "loss": 0.0825, + "step": 43731 + }, + { + "epoch": 0.3796147602885392, + "grad_norm": 0.1875, + "learning_rate": 0.0012413495153066993, + "loss": 0.082, + "step": 43732 + }, + { + "epoch": 0.3796234407687433, + "grad_norm": 0.57421875, + "learning_rate": 0.0012413201221291017, + "loss": 0.1162, + "step": 43733 + }, + { + "epoch": 0.3796321212489475, + "grad_norm": 0.0859375, + "learning_rate": 0.0012412907287969593, + "loss": 0.0967, + "step": 43734 + }, + { + "epoch": 0.37964080172915166, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012412613353103025, + "loss": 0.0981, + "step": 43735 + }, + { + "epoch": 0.37964948220935585, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012412319416691654, + "loss": 0.0811, + "step": 43736 + }, + { + "epoch": 0.37965816268956, + "grad_norm": 0.1787109375, + "learning_rate": 0.0012412025478735784, + "loss": 0.1387, + "step": 43737 + }, + { + "epoch": 0.3796668431697642, + "grad_norm": 0.57421875, + "learning_rate": 0.001241173153923575, + "loss": 0.0977, + "step": 43738 + }, + { + "epoch": 0.3796755236499683, + "grad_norm": 0.171875, + "learning_rate": 0.0012411437598191864, + "loss": 0.0889, + "step": 43739 + }, + { + "epoch": 0.37968420413017245, + "grad_norm": 0.10107421875, + "learning_rate": 0.0012411143655604454, + "loss": 0.1094, + "step": 43740 + }, + { + "epoch": 0.37969288461037665, + "grad_norm": 0.2431640625, + "learning_rate": 0.0012410849711473836, + "loss": 0.0889, + "step": 43741 + }, + { + "epoch": 0.3797015650905808, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012410555765800333, + "loss": 0.0752, + "step": 43742 + }, + { + "epoch": 0.379710245570785, + "grad_norm": 0.1962890625, + "learning_rate": 0.001241026181858427, + "loss": 0.1445, + "step": 43743 + }, + { + "epoch": 0.3797189260509891, + "grad_norm": 0.12060546875, + "learning_rate": 0.0012409967869825963, + "loss": 0.1035, + "step": 43744 + }, + { + "epoch": 0.3797276065311933, + "grad_norm": 0.0654296875, + "learning_rate": 0.0012409673919525739, + "loss": 0.0752, + "step": 43745 + }, + { + "epoch": 0.37973628701139744, + "grad_norm": 0.0966796875, + "learning_rate": 0.0012409379967683917, + "loss": 0.0942, + "step": 43746 + }, + { + "epoch": 0.37974496749160164, + "grad_norm": 0.58984375, + "learning_rate": 0.0012409086014300817, + "loss": 0.123, + "step": 43747 + }, + { + "epoch": 0.3797536479718058, + "grad_norm": 0.25390625, + "learning_rate": 0.001240879205937676, + "loss": 0.1133, + "step": 43748 + }, + { + "epoch": 0.37976232845200997, + "grad_norm": 0.451171875, + "learning_rate": 0.0012408498102912072, + "loss": 0.1006, + "step": 43749 + }, + { + "epoch": 0.3797710089322141, + "grad_norm": 0.1416015625, + "learning_rate": 0.001240820414490707, + "loss": 0.1177, + "step": 43750 + }, + { + "epoch": 0.3797796894124183, + "grad_norm": 0.55859375, + "learning_rate": 0.001240791018536208, + "loss": 0.0898, + "step": 43751 + }, + { + "epoch": 0.37978836989262243, + "grad_norm": 0.158203125, + "learning_rate": 0.0012407616224277419, + "loss": 0.1055, + "step": 43752 + }, + { + "epoch": 0.37979705037282663, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012407322261653408, + "loss": 0.0752, + "step": 43753 + }, + { + "epoch": 0.37980573085303077, + "grad_norm": 0.1826171875, + "learning_rate": 0.0012407028297490373, + "loss": 0.1475, + "step": 43754 + }, + { + "epoch": 0.37981441133323496, + "grad_norm": 0.1728515625, + "learning_rate": 0.0012406734331788629, + "loss": 0.0928, + "step": 43755 + }, + { + "epoch": 0.3798230918134391, + "grad_norm": 0.068359375, + "learning_rate": 0.0012406440364548506, + "loss": 0.0747, + "step": 43756 + }, + { + "epoch": 0.3798317722936433, + "grad_norm": 0.220703125, + "learning_rate": 0.0012406146395770319, + "loss": 0.0894, + "step": 43757 + }, + { + "epoch": 0.3798404527738474, + "grad_norm": 0.416015625, + "learning_rate": 0.0012405852425454395, + "loss": 0.1309, + "step": 43758 + }, + { + "epoch": 0.3798491332540516, + "grad_norm": 0.212890625, + "learning_rate": 0.0012405558453601044, + "loss": 0.1543, + "step": 43759 + }, + { + "epoch": 0.37985781373425576, + "grad_norm": 0.09423828125, + "learning_rate": 0.0012405264480210602, + "loss": 0.083, + "step": 43760 + }, + { + "epoch": 0.37986649421445995, + "grad_norm": 0.349609375, + "learning_rate": 0.0012404970505283384, + "loss": 0.1357, + "step": 43761 + }, + { + "epoch": 0.3798751746946641, + "grad_norm": 0.6484375, + "learning_rate": 0.0012404676528819708, + "loss": 0.1143, + "step": 43762 + }, + { + "epoch": 0.3798838551748683, + "grad_norm": 0.1591796875, + "learning_rate": 0.00124043825508199, + "loss": 0.1001, + "step": 43763 + }, + { + "epoch": 0.3798925356550724, + "grad_norm": 0.310546875, + "learning_rate": 0.0012404088571284278, + "loss": 0.0879, + "step": 43764 + }, + { + "epoch": 0.3799012161352766, + "grad_norm": 0.08740234375, + "learning_rate": 0.0012403794590213168, + "loss": 0.0967, + "step": 43765 + }, + { + "epoch": 0.37990989661548075, + "grad_norm": 0.1513671875, + "learning_rate": 0.001240350060760689, + "loss": 0.0698, + "step": 43766 + }, + { + "epoch": 0.37991857709568494, + "grad_norm": 0.10693359375, + "learning_rate": 0.0012403206623465763, + "loss": 0.0952, + "step": 43767 + }, + { + "epoch": 0.3799272575758891, + "grad_norm": 0.0830078125, + "learning_rate": 0.0012402912637790114, + "loss": 0.0771, + "step": 43768 + }, + { + "epoch": 0.37993593805609327, + "grad_norm": 0.5859375, + "learning_rate": 0.0012402618650580257, + "loss": 0.1191, + "step": 43769 + }, + { + "epoch": 0.3799446185362974, + "grad_norm": 0.134765625, + "learning_rate": 0.0012402324661836522, + "loss": 0.0718, + "step": 43770 + }, + { + "epoch": 0.3799532990165016, + "grad_norm": 0.2138671875, + "learning_rate": 0.0012402030671559224, + "loss": 0.1089, + "step": 43771 + }, + { + "epoch": 0.37996197949670574, + "grad_norm": 0.154296875, + "learning_rate": 0.0012401736679748684, + "loss": 0.1157, + "step": 43772 + }, + { + "epoch": 0.37997065997690993, + "grad_norm": 0.3515625, + "learning_rate": 0.0012401442686405227, + "loss": 0.0889, + "step": 43773 + }, + { + "epoch": 0.37997934045711407, + "grad_norm": 0.09765625, + "learning_rate": 0.0012401148691529172, + "loss": 0.1064, + "step": 43774 + }, + { + "epoch": 0.37998802093731826, + "grad_norm": 0.158203125, + "learning_rate": 0.0012400854695120843, + "loss": 0.1016, + "step": 43775 + }, + { + "epoch": 0.3799967014175224, + "grad_norm": 0.2333984375, + "learning_rate": 0.001240056069718056, + "loss": 0.0957, + "step": 43776 + }, + { + "epoch": 0.3800053818977266, + "grad_norm": 0.36328125, + "learning_rate": 0.0012400266697708644, + "loss": 0.1133, + "step": 43777 + }, + { + "epoch": 0.38001406237793073, + "grad_norm": 0.330078125, + "learning_rate": 0.001239997269670542, + "loss": 0.0859, + "step": 43778 + }, + { + "epoch": 0.3800227428581349, + "grad_norm": 0.10107421875, + "learning_rate": 0.0012399678694171206, + "loss": 0.1055, + "step": 43779 + }, + { + "epoch": 0.38003142333833906, + "grad_norm": 0.2373046875, + "learning_rate": 0.0012399384690106323, + "loss": 0.0645, + "step": 43780 + }, + { + "epoch": 0.38004010381854325, + "grad_norm": 0.94140625, + "learning_rate": 0.0012399090684511096, + "loss": 0.1221, + "step": 43781 + }, + { + "epoch": 0.3800487842987474, + "grad_norm": 0.52734375, + "learning_rate": 0.0012398796677385844, + "loss": 0.1348, + "step": 43782 + }, + { + "epoch": 0.3800574647789516, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012398502668730888, + "loss": 0.1035, + "step": 43783 + }, + { + "epoch": 0.3800661452591557, + "grad_norm": 0.10009765625, + "learning_rate": 0.001239820865854655, + "loss": 0.0913, + "step": 43784 + }, + { + "epoch": 0.3800748257393599, + "grad_norm": 0.1103515625, + "learning_rate": 0.0012397914646833153, + "loss": 0.1152, + "step": 43785 + }, + { + "epoch": 0.38008350621956405, + "grad_norm": 0.83984375, + "learning_rate": 0.0012397620633591018, + "loss": 0.1758, + "step": 43786 + }, + { + "epoch": 0.38009218669976824, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012397326618820467, + "loss": 0.1035, + "step": 43787 + }, + { + "epoch": 0.3801008671799724, + "grad_norm": 0.298828125, + "learning_rate": 0.0012397032602521818, + "loss": 0.084, + "step": 43788 + }, + { + "epoch": 0.38010954766017657, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012396738584695395, + "loss": 0.082, + "step": 43789 + }, + { + "epoch": 0.3801182281403807, + "grad_norm": 0.126953125, + "learning_rate": 0.001239644456534152, + "loss": 0.1426, + "step": 43790 + }, + { + "epoch": 0.3801269086205849, + "grad_norm": 0.5234375, + "learning_rate": 0.0012396150544460515, + "loss": 0.1245, + "step": 43791 + }, + { + "epoch": 0.38013558910078904, + "grad_norm": 0.12353515625, + "learning_rate": 0.0012395856522052701, + "loss": 0.1328, + "step": 43792 + }, + { + "epoch": 0.38014426958099323, + "grad_norm": 0.1923828125, + "learning_rate": 0.00123955624981184, + "loss": 0.0786, + "step": 43793 + }, + { + "epoch": 0.38015295006119737, + "grad_norm": 0.28125, + "learning_rate": 0.0012395268472657932, + "loss": 0.085, + "step": 43794 + }, + { + "epoch": 0.38016163054140156, + "grad_norm": 0.18359375, + "learning_rate": 0.001239497444567162, + "loss": 0.0889, + "step": 43795 + }, + { + "epoch": 0.3801703110216057, + "grad_norm": 0.13671875, + "learning_rate": 0.0012394680417159782, + "loss": 0.167, + "step": 43796 + }, + { + "epoch": 0.3801789915018099, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012394386387122747, + "loss": 0.0874, + "step": 43797 + }, + { + "epoch": 0.38018767198201403, + "grad_norm": 0.7109375, + "learning_rate": 0.0012394092355560828, + "loss": 0.1143, + "step": 43798 + }, + { + "epoch": 0.3801963524622182, + "grad_norm": 0.365234375, + "learning_rate": 0.0012393798322474351, + "loss": 0.0996, + "step": 43799 + }, + { + "epoch": 0.38020503294242236, + "grad_norm": 0.08056640625, + "learning_rate": 0.0012393504287863642, + "loss": 0.0996, + "step": 43800 + }, + { + "epoch": 0.38021371342262655, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012393210251729016, + "loss": 0.0791, + "step": 43801 + }, + { + "epoch": 0.3802223939028307, + "grad_norm": 0.1025390625, + "learning_rate": 0.001239291621407079, + "loss": 0.123, + "step": 43802 + }, + { + "epoch": 0.3802310743830349, + "grad_norm": 0.1884765625, + "learning_rate": 0.0012392622174889297, + "loss": 0.0762, + "step": 43803 + }, + { + "epoch": 0.380239754863239, + "grad_norm": 0.435546875, + "learning_rate": 0.0012392328134184852, + "loss": 0.1221, + "step": 43804 + }, + { + "epoch": 0.3802484353434432, + "grad_norm": 0.09619140625, + "learning_rate": 0.0012392034091957779, + "loss": 0.0938, + "step": 43805 + }, + { + "epoch": 0.38025711582364735, + "grad_norm": 1.375, + "learning_rate": 0.0012391740048208395, + "loss": 0.125, + "step": 43806 + }, + { + "epoch": 0.38026579630385154, + "grad_norm": 0.1015625, + "learning_rate": 0.0012391446002937027, + "loss": 0.1113, + "step": 43807 + }, + { + "epoch": 0.3802744767840557, + "grad_norm": 0.455078125, + "learning_rate": 0.0012391151956143995, + "loss": 0.1279, + "step": 43808 + }, + { + "epoch": 0.3802831572642599, + "grad_norm": 0.177734375, + "learning_rate": 0.0012390857907829623, + "loss": 0.1104, + "step": 43809 + }, + { + "epoch": 0.380291837744464, + "grad_norm": 0.08154296875, + "learning_rate": 0.0012390563857994229, + "loss": 0.0859, + "step": 43810 + }, + { + "epoch": 0.3803005182246682, + "grad_norm": 0.11474609375, + "learning_rate": 0.001239026980663813, + "loss": 0.1045, + "step": 43811 + }, + { + "epoch": 0.38030919870487234, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012389975753761655, + "loss": 0.1016, + "step": 43812 + }, + { + "epoch": 0.38031787918507654, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012389681699365126, + "loss": 0.1104, + "step": 43813 + }, + { + "epoch": 0.3803265596652807, + "grad_norm": 0.294921875, + "learning_rate": 0.001238938764344886, + "loss": 0.1152, + "step": 43814 + }, + { + "epoch": 0.38033524014548487, + "grad_norm": 0.193359375, + "learning_rate": 0.001238909358601318, + "loss": 0.1582, + "step": 43815 + }, + { + "epoch": 0.380343920625689, + "grad_norm": 0.69921875, + "learning_rate": 0.0012388799527058407, + "loss": 0.0923, + "step": 43816 + }, + { + "epoch": 0.3803526011058932, + "grad_norm": 0.1201171875, + "learning_rate": 0.001238850546658487, + "loss": 0.1001, + "step": 43817 + }, + { + "epoch": 0.38036128158609733, + "grad_norm": 0.486328125, + "learning_rate": 0.001238821140459288, + "loss": 0.0767, + "step": 43818 + }, + { + "epoch": 0.3803699620663015, + "grad_norm": 1.296875, + "learning_rate": 0.0012387917341082758, + "loss": 0.1177, + "step": 43819 + }, + { + "epoch": 0.38037864254650566, + "grad_norm": 0.1943359375, + "learning_rate": 0.0012387623276054838, + "loss": 0.0859, + "step": 43820 + }, + { + "epoch": 0.38038732302670986, + "grad_norm": 0.357421875, + "learning_rate": 0.0012387329209509433, + "loss": 0.0854, + "step": 43821 + }, + { + "epoch": 0.380396003506914, + "grad_norm": 0.515625, + "learning_rate": 0.0012387035141446863, + "loss": 0.0864, + "step": 43822 + }, + { + "epoch": 0.3804046839871182, + "grad_norm": 0.169921875, + "learning_rate": 0.0012386741071867453, + "loss": 0.061, + "step": 43823 + }, + { + "epoch": 0.3804133644673223, + "grad_norm": 0.1328125, + "learning_rate": 0.0012386447000771524, + "loss": 0.0825, + "step": 43824 + }, + { + "epoch": 0.3804220449475265, + "grad_norm": 0.294921875, + "learning_rate": 0.0012386152928159397, + "loss": 0.1045, + "step": 43825 + }, + { + "epoch": 0.38043072542773065, + "grad_norm": 0.51953125, + "learning_rate": 0.0012385858854031396, + "loss": 0.084, + "step": 43826 + }, + { + "epoch": 0.38043940590793485, + "grad_norm": 0.1435546875, + "learning_rate": 0.001238556477838784, + "loss": 0.1138, + "step": 43827 + }, + { + "epoch": 0.380448086388139, + "grad_norm": 0.33203125, + "learning_rate": 0.0012385270701229048, + "loss": 0.0791, + "step": 43828 + }, + { + "epoch": 0.3804567668683432, + "grad_norm": 0.423828125, + "learning_rate": 0.0012384976622555346, + "loss": 0.0718, + "step": 43829 + }, + { + "epoch": 0.3804654473485473, + "grad_norm": 0.291015625, + "learning_rate": 0.0012384682542367056, + "loss": 0.1396, + "step": 43830 + }, + { + "epoch": 0.3804741278287515, + "grad_norm": 0.298828125, + "learning_rate": 0.0012384388460664494, + "loss": 0.0933, + "step": 43831 + }, + { + "epoch": 0.38048280830895564, + "grad_norm": 0.2734375, + "learning_rate": 0.001238409437744799, + "loss": 0.1016, + "step": 43832 + }, + { + "epoch": 0.38049148878915984, + "grad_norm": 0.328125, + "learning_rate": 0.0012383800292717862, + "loss": 0.1025, + "step": 43833 + }, + { + "epoch": 0.380500169269364, + "grad_norm": 0.2490234375, + "learning_rate": 0.0012383506206474426, + "loss": 0.0713, + "step": 43834 + }, + { + "epoch": 0.38050884974956817, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012383212118718015, + "loss": 0.124, + "step": 43835 + }, + { + "epoch": 0.3805175302297723, + "grad_norm": 0.1650390625, + "learning_rate": 0.001238291802944894, + "loss": 0.1147, + "step": 43836 + }, + { + "epoch": 0.3805262107099765, + "grad_norm": 0.244140625, + "learning_rate": 0.0012382623938667525, + "loss": 0.0908, + "step": 43837 + }, + { + "epoch": 0.38053489119018064, + "grad_norm": 0.3125, + "learning_rate": 0.0012382329846374097, + "loss": 0.0884, + "step": 43838 + }, + { + "epoch": 0.38054357167038483, + "grad_norm": 0.302734375, + "learning_rate": 0.001238203575256897, + "loss": 0.1045, + "step": 43839 + }, + { + "epoch": 0.38055225215058897, + "grad_norm": 0.75390625, + "learning_rate": 0.001238174165725247, + "loss": 0.125, + "step": 43840 + }, + { + "epoch": 0.38056093263079316, + "grad_norm": 0.1552734375, + "learning_rate": 0.0012381447560424921, + "loss": 0.1299, + "step": 43841 + }, + { + "epoch": 0.3805696131109973, + "grad_norm": 0.232421875, + "learning_rate": 0.0012381153462086639, + "loss": 0.0972, + "step": 43842 + }, + { + "epoch": 0.3805782935912015, + "grad_norm": 0.158203125, + "learning_rate": 0.001238085936223795, + "loss": 0.1001, + "step": 43843 + }, + { + "epoch": 0.3805869740714056, + "grad_norm": 0.2392578125, + "learning_rate": 0.0012380565260879173, + "loss": 0.0928, + "step": 43844 + }, + { + "epoch": 0.3805956545516098, + "grad_norm": 0.166015625, + "learning_rate": 0.0012380271158010632, + "loss": 0.1128, + "step": 43845 + }, + { + "epoch": 0.38060433503181396, + "grad_norm": 0.146484375, + "learning_rate": 0.0012379977053632646, + "loss": 0.1094, + "step": 43846 + }, + { + "epoch": 0.38061301551201815, + "grad_norm": 0.08837890625, + "learning_rate": 0.0012379682947745538, + "loss": 0.1084, + "step": 43847 + }, + { + "epoch": 0.3806216959922223, + "grad_norm": 0.11328125, + "learning_rate": 0.001237938884034963, + "loss": 0.082, + "step": 43848 + }, + { + "epoch": 0.3806303764724265, + "grad_norm": 0.3515625, + "learning_rate": 0.0012379094731445241, + "loss": 0.1025, + "step": 43849 + }, + { + "epoch": 0.3806390569526306, + "grad_norm": 0.6953125, + "learning_rate": 0.0012378800621032698, + "loss": 0.082, + "step": 43850 + }, + { + "epoch": 0.3806477374328348, + "grad_norm": 0.255859375, + "learning_rate": 0.0012378506509112318, + "loss": 0.1172, + "step": 43851 + }, + { + "epoch": 0.38065641791303895, + "grad_norm": 0.23046875, + "learning_rate": 0.0012378212395684422, + "loss": 0.1152, + "step": 43852 + }, + { + "epoch": 0.38066509839324314, + "grad_norm": 0.74609375, + "learning_rate": 0.0012377918280749337, + "loss": 0.1084, + "step": 43853 + }, + { + "epoch": 0.3806737788734473, + "grad_norm": 0.107421875, + "learning_rate": 0.0012377624164307377, + "loss": 0.1113, + "step": 43854 + }, + { + "epoch": 0.38068245935365147, + "grad_norm": 0.208984375, + "learning_rate": 0.0012377330046358871, + "loss": 0.1094, + "step": 43855 + }, + { + "epoch": 0.3806911398338556, + "grad_norm": 0.65234375, + "learning_rate": 0.0012377035926904138, + "loss": 0.1328, + "step": 43856 + }, + { + "epoch": 0.3806998203140598, + "grad_norm": 0.1650390625, + "learning_rate": 0.0012376741805943498, + "loss": 0.084, + "step": 43857 + }, + { + "epoch": 0.38070850079426394, + "grad_norm": 0.11181640625, + "learning_rate": 0.0012376447683477274, + "loss": 0.1016, + "step": 43858 + }, + { + "epoch": 0.38071718127446813, + "grad_norm": 0.4296875, + "learning_rate": 0.001237615355950579, + "loss": 0.0894, + "step": 43859 + }, + { + "epoch": 0.38072586175467227, + "grad_norm": 0.53515625, + "learning_rate": 0.0012375859434029359, + "loss": 0.1216, + "step": 43860 + }, + { + "epoch": 0.38073454223487646, + "grad_norm": 0.22265625, + "learning_rate": 0.0012375565307048313, + "loss": 0.0986, + "step": 43861 + }, + { + "epoch": 0.3807432227150806, + "grad_norm": 0.5, + "learning_rate": 0.001237527117856297, + "loss": 0.1094, + "step": 43862 + }, + { + "epoch": 0.38075190319528474, + "grad_norm": 0.333984375, + "learning_rate": 0.001237497704857365, + "loss": 0.1084, + "step": 43863 + }, + { + "epoch": 0.38076058367548893, + "grad_norm": 0.283203125, + "learning_rate": 0.0012374682917080675, + "loss": 0.1143, + "step": 43864 + }, + { + "epoch": 0.38076926415569307, + "grad_norm": 0.51171875, + "learning_rate": 0.001237438878408437, + "loss": 0.0835, + "step": 43865 + }, + { + "epoch": 0.38077794463589726, + "grad_norm": 0.484375, + "learning_rate": 0.0012374094649585052, + "loss": 0.1196, + "step": 43866 + }, + { + "epoch": 0.3807866251161014, + "grad_norm": 0.123046875, + "learning_rate": 0.0012373800513583045, + "loss": 0.1064, + "step": 43867 + }, + { + "epoch": 0.3807953055963056, + "grad_norm": 0.1279296875, + "learning_rate": 0.0012373506376078672, + "loss": 0.0898, + "step": 43868 + }, + { + "epoch": 0.3808039860765097, + "grad_norm": 0.20703125, + "learning_rate": 0.0012373212237072248, + "loss": 0.0898, + "step": 43869 + }, + { + "epoch": 0.3808126665567139, + "grad_norm": 0.345703125, + "learning_rate": 0.0012372918096564104, + "loss": 0.1016, + "step": 43870 + }, + { + "epoch": 0.38082134703691806, + "grad_norm": 0.076171875, + "learning_rate": 0.0012372623954554555, + "loss": 0.085, + "step": 43871 + }, + { + "epoch": 0.38083002751712225, + "grad_norm": 0.12255859375, + "learning_rate": 0.001237232981104393, + "loss": 0.1777, + "step": 43872 + }, + { + "epoch": 0.3808387079973264, + "grad_norm": 0.134765625, + "learning_rate": 0.0012372035666032542, + "loss": 0.1455, + "step": 43873 + }, + { + "epoch": 0.3808473884775306, + "grad_norm": 0.404296875, + "learning_rate": 0.0012371741519520715, + "loss": 0.085, + "step": 43874 + }, + { + "epoch": 0.3808560689577347, + "grad_norm": 0.11962890625, + "learning_rate": 0.0012371447371508778, + "loss": 0.1143, + "step": 43875 + }, + { + "epoch": 0.3808647494379389, + "grad_norm": 0.18359375, + "learning_rate": 0.001237115322199704, + "loss": 0.0593, + "step": 43876 + }, + { + "epoch": 0.38087342991814305, + "grad_norm": 1.4296875, + "learning_rate": 0.001237085907098583, + "loss": 0.1523, + "step": 43877 + }, + { + "epoch": 0.38088211039834724, + "grad_norm": 0.2490234375, + "learning_rate": 0.0012370564918475475, + "loss": 0.0962, + "step": 43878 + }, + { + "epoch": 0.3808907908785514, + "grad_norm": 0.2373046875, + "learning_rate": 0.0012370270764466285, + "loss": 0.0791, + "step": 43879 + }, + { + "epoch": 0.38089947135875557, + "grad_norm": 0.1416015625, + "learning_rate": 0.0012369976608958588, + "loss": 0.1104, + "step": 43880 + }, + { + "epoch": 0.3809081518389597, + "grad_norm": 0.1298828125, + "learning_rate": 0.0012369682451952705, + "loss": 0.0977, + "step": 43881 + }, + { + "epoch": 0.3809168323191639, + "grad_norm": 0.373046875, + "learning_rate": 0.0012369388293448965, + "loss": 0.1177, + "step": 43882 + }, + { + "epoch": 0.38092551279936804, + "grad_norm": 0.419921875, + "learning_rate": 0.0012369094133447675, + "loss": 0.1445, + "step": 43883 + }, + { + "epoch": 0.38093419327957223, + "grad_norm": 0.38671875, + "learning_rate": 0.0012368799971949165, + "loss": 0.0967, + "step": 43884 + }, + { + "epoch": 0.38094287375977637, + "grad_norm": 0.1787109375, + "learning_rate": 0.001236850580895376, + "loss": 0.0938, + "step": 43885 + }, + { + "epoch": 0.38095155423998056, + "grad_norm": 0.333984375, + "learning_rate": 0.0012368211644461773, + "loss": 0.0747, + "step": 43886 + }, + { + "epoch": 0.3809602347201847, + "grad_norm": 0.0966796875, + "learning_rate": 0.0012367917478473534, + "loss": 0.0942, + "step": 43887 + }, + { + "epoch": 0.3809689152003889, + "grad_norm": 0.115234375, + "learning_rate": 0.0012367623310989358, + "loss": 0.1064, + "step": 43888 + }, + { + "epoch": 0.38097759568059303, + "grad_norm": 0.16015625, + "learning_rate": 0.001236732914200957, + "loss": 0.1084, + "step": 43889 + }, + { + "epoch": 0.3809862761607972, + "grad_norm": 0.2236328125, + "learning_rate": 0.0012367034971534493, + "loss": 0.0913, + "step": 43890 + }, + { + "epoch": 0.38099495664100136, + "grad_norm": 0.3671875, + "learning_rate": 0.0012366740799564446, + "loss": 0.1162, + "step": 43891 + }, + { + "epoch": 0.38100363712120555, + "grad_norm": 0.2177734375, + "learning_rate": 0.0012366446626099752, + "loss": 0.0918, + "step": 43892 + }, + { + "epoch": 0.3810123176014097, + "grad_norm": 0.326171875, + "learning_rate": 0.001236615245114073, + "loss": 0.1157, + "step": 43893 + }, + { + "epoch": 0.3810209980816139, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012365858274687711, + "loss": 0.0742, + "step": 43894 + }, + { + "epoch": 0.381029678561818, + "grad_norm": 0.1416015625, + "learning_rate": 0.0012365564096741006, + "loss": 0.1011, + "step": 43895 + }, + { + "epoch": 0.3810383590420222, + "grad_norm": 0.07470703125, + "learning_rate": 0.001236526991730094, + "loss": 0.1055, + "step": 43896 + }, + { + "epoch": 0.38104703952222635, + "grad_norm": 0.71484375, + "learning_rate": 0.0012364975736367833, + "loss": 0.1309, + "step": 43897 + }, + { + "epoch": 0.38105572000243054, + "grad_norm": 0.130859375, + "learning_rate": 0.0012364681553942015, + "loss": 0.0918, + "step": 43898 + }, + { + "epoch": 0.3810644004826347, + "grad_norm": 0.310546875, + "learning_rate": 0.0012364387370023798, + "loss": 0.0913, + "step": 43899 + }, + { + "epoch": 0.3810730809628389, + "grad_norm": 0.126953125, + "learning_rate": 0.0012364093184613508, + "loss": 0.125, + "step": 43900 + }, + { + "epoch": 0.381081761443043, + "grad_norm": 0.447265625, + "learning_rate": 0.0012363798997711464, + "loss": 0.1016, + "step": 43901 + }, + { + "epoch": 0.3810904419232472, + "grad_norm": 0.62109375, + "learning_rate": 0.0012363504809317992, + "loss": 0.1201, + "step": 43902 + }, + { + "epoch": 0.38109912240345134, + "grad_norm": 0.298828125, + "learning_rate": 0.0012363210619433412, + "loss": 0.1025, + "step": 43903 + }, + { + "epoch": 0.38110780288365553, + "grad_norm": 0.322265625, + "learning_rate": 0.0012362916428058047, + "loss": 0.0806, + "step": 43904 + }, + { + "epoch": 0.38111648336385967, + "grad_norm": 0.421875, + "learning_rate": 0.0012362622235192215, + "loss": 0.1025, + "step": 43905 + }, + { + "epoch": 0.38112516384406386, + "grad_norm": 0.66796875, + "learning_rate": 0.001236232804083624, + "loss": 0.0898, + "step": 43906 + }, + { + "epoch": 0.381133844324268, + "grad_norm": 0.15234375, + "learning_rate": 0.001236203384499044, + "loss": 0.1001, + "step": 43907 + }, + { + "epoch": 0.3811425248044722, + "grad_norm": 0.2578125, + "learning_rate": 0.0012361739647655146, + "loss": 0.0884, + "step": 43908 + }, + { + "epoch": 0.38115120528467633, + "grad_norm": 0.080078125, + "learning_rate": 0.001236144544883067, + "loss": 0.103, + "step": 43909 + }, + { + "epoch": 0.3811598857648805, + "grad_norm": 0.61328125, + "learning_rate": 0.0012361151248517339, + "loss": 0.0796, + "step": 43910 + }, + { + "epoch": 0.38116856624508466, + "grad_norm": 0.1357421875, + "learning_rate": 0.0012360857046715475, + "loss": 0.0918, + "step": 43911 + }, + { + "epoch": 0.38117724672528885, + "grad_norm": 0.1494140625, + "learning_rate": 0.0012360562843425399, + "loss": 0.1377, + "step": 43912 + }, + { + "epoch": 0.381185927205493, + "grad_norm": 0.53515625, + "learning_rate": 0.0012360268638647426, + "loss": 0.0767, + "step": 43913 + }, + { + "epoch": 0.3811946076856972, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012359974432381892, + "loss": 0.1143, + "step": 43914 + }, + { + "epoch": 0.3812032881659013, + "grad_norm": 0.419921875, + "learning_rate": 0.0012359680224629105, + "loss": 0.1133, + "step": 43915 + }, + { + "epoch": 0.3812119686461055, + "grad_norm": 0.12890625, + "learning_rate": 0.0012359386015389391, + "loss": 0.0737, + "step": 43916 + }, + { + "epoch": 0.38122064912630965, + "grad_norm": 0.28515625, + "learning_rate": 0.0012359091804663076, + "loss": 0.1079, + "step": 43917 + }, + { + "epoch": 0.38122932960651384, + "grad_norm": 0.515625, + "learning_rate": 0.0012358797592450477, + "loss": 0.0996, + "step": 43918 + }, + { + "epoch": 0.381238010086718, + "grad_norm": 0.443359375, + "learning_rate": 0.0012358503378751917, + "loss": 0.0776, + "step": 43919 + }, + { + "epoch": 0.3812466905669222, + "grad_norm": 0.296875, + "learning_rate": 0.001235820916356772, + "loss": 0.1177, + "step": 43920 + }, + { + "epoch": 0.3812553710471263, + "grad_norm": 0.076171875, + "learning_rate": 0.0012357914946898205, + "loss": 0.0615, + "step": 43921 + }, + { + "epoch": 0.3812640515273305, + "grad_norm": 0.6171875, + "learning_rate": 0.0012357620728743696, + "loss": 0.0669, + "step": 43922 + }, + { + "epoch": 0.38127273200753464, + "grad_norm": 0.2734375, + "learning_rate": 0.0012357326509104512, + "loss": 0.1309, + "step": 43923 + }, + { + "epoch": 0.38128141248773884, + "grad_norm": 0.9453125, + "learning_rate": 0.0012357032287980973, + "loss": 0.0762, + "step": 43924 + }, + { + "epoch": 0.381290092967943, + "grad_norm": 0.80859375, + "learning_rate": 0.0012356738065373409, + "loss": 0.1172, + "step": 43925 + }, + { + "epoch": 0.38129877344814717, + "grad_norm": 0.107421875, + "learning_rate": 0.0012356443841282138, + "loss": 0.0811, + "step": 43926 + }, + { + "epoch": 0.3813074539283513, + "grad_norm": 0.130859375, + "learning_rate": 0.0012356149615707474, + "loss": 0.0723, + "step": 43927 + }, + { + "epoch": 0.3813161344085555, + "grad_norm": 0.1962890625, + "learning_rate": 0.0012355855388649745, + "loss": 0.0908, + "step": 43928 + }, + { + "epoch": 0.38132481488875963, + "grad_norm": 0.1787109375, + "learning_rate": 0.001235556116010928, + "loss": 0.0908, + "step": 43929 + }, + { + "epoch": 0.3813334953689638, + "grad_norm": 0.134765625, + "learning_rate": 0.0012355266930086388, + "loss": 0.0796, + "step": 43930 + }, + { + "epoch": 0.38134217584916796, + "grad_norm": 0.28515625, + "learning_rate": 0.0012354972698581399, + "loss": 0.1348, + "step": 43931 + }, + { + "epoch": 0.38135085632937216, + "grad_norm": 0.16796875, + "learning_rate": 0.001235467846559463, + "loss": 0.1201, + "step": 43932 + }, + { + "epoch": 0.3813595368095763, + "grad_norm": 0.322265625, + "learning_rate": 0.0012354384231126408, + "loss": 0.0801, + "step": 43933 + }, + { + "epoch": 0.3813682172897805, + "grad_norm": 0.6171875, + "learning_rate": 0.0012354089995177048, + "loss": 0.1436, + "step": 43934 + }, + { + "epoch": 0.3813768977699846, + "grad_norm": 0.1865234375, + "learning_rate": 0.001235379575774688, + "loss": 0.1064, + "step": 43935 + }, + { + "epoch": 0.3813855782501888, + "grad_norm": 0.251953125, + "learning_rate": 0.0012353501518836216, + "loss": 0.0889, + "step": 43936 + }, + { + "epoch": 0.38139425873039295, + "grad_norm": 0.220703125, + "learning_rate": 0.0012353207278445386, + "loss": 0.0781, + "step": 43937 + }, + { + "epoch": 0.38140293921059715, + "grad_norm": 0.4140625, + "learning_rate": 0.001235291303657471, + "loss": 0.0889, + "step": 43938 + }, + { + "epoch": 0.3814116196908013, + "grad_norm": 0.2734375, + "learning_rate": 0.0012352618793224509, + "loss": 0.0811, + "step": 43939 + }, + { + "epoch": 0.3814203001710055, + "grad_norm": 0.25, + "learning_rate": 0.0012352324548395102, + "loss": 0.123, + "step": 43940 + }, + { + "epoch": 0.3814289806512096, + "grad_norm": 0.388671875, + "learning_rate": 0.0012352030302086815, + "loss": 0.0806, + "step": 43941 + }, + { + "epoch": 0.3814376611314138, + "grad_norm": 0.30078125, + "learning_rate": 0.0012351736054299967, + "loss": 0.0942, + "step": 43942 + }, + { + "epoch": 0.38144634161161795, + "grad_norm": 0.08251953125, + "learning_rate": 0.001235144180503488, + "loss": 0.0581, + "step": 43943 + }, + { + "epoch": 0.38145502209182214, + "grad_norm": 0.40625, + "learning_rate": 0.001235114755429188, + "loss": 0.1025, + "step": 43944 + }, + { + "epoch": 0.3814637025720263, + "grad_norm": 0.6796875, + "learning_rate": 0.001235085330207128, + "loss": 0.0996, + "step": 43945 + }, + { + "epoch": 0.38147238305223047, + "grad_norm": 0.396484375, + "learning_rate": 0.001235055904837341, + "loss": 0.0928, + "step": 43946 + }, + { + "epoch": 0.3814810635324346, + "grad_norm": 0.4296875, + "learning_rate": 0.001235026479319859, + "loss": 0.0879, + "step": 43947 + }, + { + "epoch": 0.3814897440126388, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012349970536547137, + "loss": 0.0908, + "step": 43948 + }, + { + "epoch": 0.38149842449284294, + "grad_norm": 1.8203125, + "learning_rate": 0.001234967627841938, + "loss": 0.332, + "step": 43949 + }, + { + "epoch": 0.38150710497304713, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012349382018815636, + "loss": 0.1123, + "step": 43950 + }, + { + "epoch": 0.38151578545325127, + "grad_norm": 0.52734375, + "learning_rate": 0.001234908775773623, + "loss": 0.0898, + "step": 43951 + }, + { + "epoch": 0.38152446593345546, + "grad_norm": 0.1240234375, + "learning_rate": 0.001234879349518148, + "loss": 0.0889, + "step": 43952 + }, + { + "epoch": 0.3815331464136596, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012348499231151708, + "loss": 0.0791, + "step": 43953 + }, + { + "epoch": 0.3815418268938638, + "grad_norm": 0.07373046875, + "learning_rate": 0.001234820496564724, + "loss": 0.0498, + "step": 43954 + }, + { + "epoch": 0.3815505073740679, + "grad_norm": 0.224609375, + "learning_rate": 0.0012347910698668396, + "loss": 0.1055, + "step": 43955 + }, + { + "epoch": 0.3815591878542721, + "grad_norm": 0.353515625, + "learning_rate": 0.0012347616430215497, + "loss": 0.0718, + "step": 43956 + }, + { + "epoch": 0.38156786833447626, + "grad_norm": 0.2265625, + "learning_rate": 0.001234732216028886, + "loss": 0.085, + "step": 43957 + }, + { + "epoch": 0.38157654881468045, + "grad_norm": 0.38671875, + "learning_rate": 0.0012347027888888816, + "loss": 0.1309, + "step": 43958 + }, + { + "epoch": 0.3815852292948846, + "grad_norm": 0.232421875, + "learning_rate": 0.0012346733616015683, + "loss": 0.0747, + "step": 43959 + }, + { + "epoch": 0.3815939097750888, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012346439341669783, + "loss": 0.1084, + "step": 43960 + }, + { + "epoch": 0.3816025902552929, + "grad_norm": 0.5078125, + "learning_rate": 0.0012346145065851438, + "loss": 0.1016, + "step": 43961 + }, + { + "epoch": 0.3816112707354971, + "grad_norm": 0.671875, + "learning_rate": 0.0012345850788560962, + "loss": 0.1035, + "step": 43962 + }, + { + "epoch": 0.38161995121570125, + "grad_norm": 0.1923828125, + "learning_rate": 0.001234555650979869, + "loss": 0.0898, + "step": 43963 + }, + { + "epoch": 0.38162863169590544, + "grad_norm": 0.142578125, + "learning_rate": 0.0012345262229564937, + "loss": 0.104, + "step": 43964 + }, + { + "epoch": 0.3816373121761096, + "grad_norm": 0.1640625, + "learning_rate": 0.0012344967947860023, + "loss": 0.123, + "step": 43965 + }, + { + "epoch": 0.38164599265631377, + "grad_norm": 0.40625, + "learning_rate": 0.0012344673664684272, + "loss": 0.083, + "step": 43966 + }, + { + "epoch": 0.3816546731365179, + "grad_norm": 0.11279296875, + "learning_rate": 0.0012344379380038008, + "loss": 0.1182, + "step": 43967 + }, + { + "epoch": 0.3816633536167221, + "grad_norm": 0.337890625, + "learning_rate": 0.0012344085093921554, + "loss": 0.105, + "step": 43968 + }, + { + "epoch": 0.38167203409692624, + "grad_norm": 0.310546875, + "learning_rate": 0.0012343790806335223, + "loss": 0.1074, + "step": 43969 + }, + { + "epoch": 0.38168071457713043, + "grad_norm": 5.03125, + "learning_rate": 0.0012343496517279347, + "loss": 0.3477, + "step": 43970 + }, + { + "epoch": 0.38168939505733457, + "grad_norm": 0.26171875, + "learning_rate": 0.001234320222675424, + "loss": 0.0723, + "step": 43971 + }, + { + "epoch": 0.38169807553753876, + "grad_norm": 0.18359375, + "learning_rate": 0.001234290793476023, + "loss": 0.1006, + "step": 43972 + }, + { + "epoch": 0.3817067560177429, + "grad_norm": 0.3671875, + "learning_rate": 0.0012342613641297636, + "loss": 0.0928, + "step": 43973 + }, + { + "epoch": 0.3817154364979471, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012342319346366774, + "loss": 0.1104, + "step": 43974 + }, + { + "epoch": 0.38172411697815123, + "grad_norm": 0.416015625, + "learning_rate": 0.001234202504996798, + "loss": 0.0923, + "step": 43975 + }, + { + "epoch": 0.3817327974583554, + "grad_norm": 0.107421875, + "learning_rate": 0.0012341730752101563, + "loss": 0.0986, + "step": 43976 + }, + { + "epoch": 0.38174147793855956, + "grad_norm": 0.921875, + "learning_rate": 0.001234143645276785, + "loss": 0.0986, + "step": 43977 + }, + { + "epoch": 0.38175015841876375, + "grad_norm": 0.11962890625, + "learning_rate": 0.0012341142151967163, + "loss": 0.1084, + "step": 43978 + }, + { + "epoch": 0.3817588388989679, + "grad_norm": 0.2001953125, + "learning_rate": 0.0012340847849699821, + "loss": 0.1045, + "step": 43979 + }, + { + "epoch": 0.3817675193791721, + "grad_norm": 0.146484375, + "learning_rate": 0.001234055354596615, + "loss": 0.1055, + "step": 43980 + }, + { + "epoch": 0.3817761998593762, + "grad_norm": 0.59375, + "learning_rate": 0.001234025924076647, + "loss": 0.1309, + "step": 43981 + }, + { + "epoch": 0.3817848803395804, + "grad_norm": 0.82421875, + "learning_rate": 0.0012339964934101103, + "loss": 0.0845, + "step": 43982 + }, + { + "epoch": 0.38179356081978455, + "grad_norm": 0.62890625, + "learning_rate": 0.001233967062597037, + "loss": 0.0713, + "step": 43983 + }, + { + "epoch": 0.38180224129998874, + "grad_norm": 0.3671875, + "learning_rate": 0.0012339376316374592, + "loss": 0.0879, + "step": 43984 + }, + { + "epoch": 0.3818109217801929, + "grad_norm": 0.74609375, + "learning_rate": 0.0012339082005314091, + "loss": 0.1396, + "step": 43985 + }, + { + "epoch": 0.381819602260397, + "grad_norm": 0.12060546875, + "learning_rate": 0.001233878769278919, + "loss": 0.1113, + "step": 43986 + }, + { + "epoch": 0.3818282827406012, + "grad_norm": 0.06591796875, + "learning_rate": 0.0012338493378800215, + "loss": 0.0488, + "step": 43987 + }, + { + "epoch": 0.38183696322080535, + "grad_norm": 0.61328125, + "learning_rate": 0.001233819906334748, + "loss": 0.1055, + "step": 43988 + }, + { + "epoch": 0.38184564370100954, + "grad_norm": 0.1826171875, + "learning_rate": 0.0012337904746431315, + "loss": 0.082, + "step": 43989 + }, + { + "epoch": 0.3818543241812137, + "grad_norm": 0.373046875, + "learning_rate": 0.0012337610428052033, + "loss": 0.1133, + "step": 43990 + }, + { + "epoch": 0.38186300466141787, + "grad_norm": 0.11376953125, + "learning_rate": 0.0012337316108209965, + "loss": 0.1016, + "step": 43991 + }, + { + "epoch": 0.381871685141622, + "grad_norm": 0.1044921875, + "learning_rate": 0.0012337021786905425, + "loss": 0.1177, + "step": 43992 + }, + { + "epoch": 0.3818803656218262, + "grad_norm": 0.169921875, + "learning_rate": 0.0012336727464138737, + "loss": 0.0967, + "step": 43993 + }, + { + "epoch": 0.38188904610203034, + "grad_norm": 0.8125, + "learning_rate": 0.0012336433139910224, + "loss": 0.167, + "step": 43994 + }, + { + "epoch": 0.38189772658223453, + "grad_norm": 0.0966796875, + "learning_rate": 0.001233613881422021, + "loss": 0.0781, + "step": 43995 + }, + { + "epoch": 0.38190640706243867, + "grad_norm": 0.072265625, + "learning_rate": 0.0012335844487069015, + "loss": 0.123, + "step": 43996 + }, + { + "epoch": 0.38191508754264286, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012335550158456957, + "loss": 0.0947, + "step": 43997 + }, + { + "epoch": 0.381923768022847, + "grad_norm": 0.373046875, + "learning_rate": 0.0012335255828384367, + "loss": 0.1416, + "step": 43998 + }, + { + "epoch": 0.3819324485030512, + "grad_norm": 0.79296875, + "learning_rate": 0.0012334961496851555, + "loss": 0.1162, + "step": 43999 + }, + { + "epoch": 0.38194112898325533, + "grad_norm": 0.07568359375, + "learning_rate": 0.0012334667163858854, + "loss": 0.0962, + "step": 44000 + }, + { + "epoch": 0.3819498094634595, + "grad_norm": 0.07666015625, + "learning_rate": 0.0012334372829406577, + "loss": 0.0952, + "step": 44001 + }, + { + "epoch": 0.38195848994366366, + "grad_norm": 0.146484375, + "learning_rate": 0.0012334078493495054, + "loss": 0.1025, + "step": 44002 + }, + { + "epoch": 0.38196717042386785, + "grad_norm": 0.166015625, + "learning_rate": 0.00123337841561246, + "loss": 0.1055, + "step": 44003 + }, + { + "epoch": 0.381975850904072, + "grad_norm": 0.12060546875, + "learning_rate": 0.001233348981729554, + "loss": 0.0645, + "step": 44004 + }, + { + "epoch": 0.3819845313842762, + "grad_norm": 0.447265625, + "learning_rate": 0.0012333195477008197, + "loss": 0.0898, + "step": 44005 + }, + { + "epoch": 0.3819932118644803, + "grad_norm": 0.6875, + "learning_rate": 0.0012332901135262888, + "loss": 0.1074, + "step": 44006 + }, + { + "epoch": 0.3820018923446845, + "grad_norm": 0.1611328125, + "learning_rate": 0.0012332606792059943, + "loss": 0.0894, + "step": 44007 + }, + { + "epoch": 0.38201057282488865, + "grad_norm": 0.185546875, + "learning_rate": 0.0012332312447399679, + "loss": 0.1016, + "step": 44008 + }, + { + "epoch": 0.38201925330509284, + "grad_norm": 0.1484375, + "learning_rate": 0.0012332018101282414, + "loss": 0.123, + "step": 44009 + }, + { + "epoch": 0.382027933785297, + "grad_norm": 0.09326171875, + "learning_rate": 0.0012331723753708477, + "loss": 0.0879, + "step": 44010 + }, + { + "epoch": 0.3820366142655012, + "grad_norm": 0.69140625, + "learning_rate": 0.0012331429404678186, + "loss": 0.0625, + "step": 44011 + }, + { + "epoch": 0.3820452947457053, + "grad_norm": 0.271484375, + "learning_rate": 0.0012331135054191866, + "loss": 0.0962, + "step": 44012 + }, + { + "epoch": 0.3820539752259095, + "grad_norm": 0.70703125, + "learning_rate": 0.0012330840702249836, + "loss": 0.0898, + "step": 44013 + }, + { + "epoch": 0.38206265570611364, + "grad_norm": 0.11474609375, + "learning_rate": 0.0012330546348852413, + "loss": 0.0869, + "step": 44014 + }, + { + "epoch": 0.38207133618631783, + "grad_norm": 0.5625, + "learning_rate": 0.0012330251993999928, + "loss": 0.0811, + "step": 44015 + }, + { + "epoch": 0.38208001666652197, + "grad_norm": 0.13671875, + "learning_rate": 0.0012329957637692702, + "loss": 0.1338, + "step": 44016 + }, + { + "epoch": 0.38208869714672616, + "grad_norm": 0.193359375, + "learning_rate": 0.0012329663279931054, + "loss": 0.1455, + "step": 44017 + }, + { + "epoch": 0.3820973776269303, + "grad_norm": 0.18359375, + "learning_rate": 0.0012329368920715302, + "loss": 0.1001, + "step": 44018 + }, + { + "epoch": 0.3821060581071345, + "grad_norm": 0.22265625, + "learning_rate": 0.0012329074560045777, + "loss": 0.0859, + "step": 44019 + }, + { + "epoch": 0.38211473858733863, + "grad_norm": 0.2314453125, + "learning_rate": 0.0012328780197922795, + "loss": 0.1221, + "step": 44020 + }, + { + "epoch": 0.3821234190675428, + "grad_norm": 0.1123046875, + "learning_rate": 0.001232848583434668, + "loss": 0.0933, + "step": 44021 + }, + { + "epoch": 0.38213209954774696, + "grad_norm": 0.34375, + "learning_rate": 0.001232819146931775, + "loss": 0.0894, + "step": 44022 + }, + { + "epoch": 0.38214078002795115, + "grad_norm": 0.328125, + "learning_rate": 0.0012327897102836329, + "loss": 0.1113, + "step": 44023 + }, + { + "epoch": 0.3821494605081553, + "grad_norm": 0.1904296875, + "learning_rate": 0.0012327602734902741, + "loss": 0.0747, + "step": 44024 + }, + { + "epoch": 0.3821581409883595, + "grad_norm": 0.2734375, + "learning_rate": 0.0012327308365517305, + "loss": 0.1299, + "step": 44025 + }, + { + "epoch": 0.3821668214685636, + "grad_norm": 0.421875, + "learning_rate": 0.0012327013994680347, + "loss": 0.1074, + "step": 44026 + }, + { + "epoch": 0.3821755019487678, + "grad_norm": 0.11474609375, + "learning_rate": 0.0012326719622392187, + "loss": 0.085, + "step": 44027 + }, + { + "epoch": 0.38218418242897195, + "grad_norm": 0.259765625, + "learning_rate": 0.0012326425248653146, + "loss": 0.1328, + "step": 44028 + }, + { + "epoch": 0.38219286290917615, + "grad_norm": 0.31640625, + "learning_rate": 0.0012326130873463543, + "loss": 0.0752, + "step": 44029 + }, + { + "epoch": 0.3822015433893803, + "grad_norm": 0.2734375, + "learning_rate": 0.0012325836496823708, + "loss": 0.1064, + "step": 44030 + }, + { + "epoch": 0.3822102238695845, + "grad_norm": 0.302734375, + "learning_rate": 0.0012325542118733952, + "loss": 0.1279, + "step": 44031 + }, + { + "epoch": 0.3822189043497886, + "grad_norm": 0.283203125, + "learning_rate": 0.0012325247739194609, + "loss": 0.1099, + "step": 44032 + }, + { + "epoch": 0.3822275848299928, + "grad_norm": 0.423828125, + "learning_rate": 0.0012324953358205993, + "loss": 0.106, + "step": 44033 + }, + { + "epoch": 0.38223626531019694, + "grad_norm": 0.26171875, + "learning_rate": 0.0012324658975768428, + "loss": 0.0967, + "step": 44034 + }, + { + "epoch": 0.38224494579040114, + "grad_norm": 0.2177734375, + "learning_rate": 0.0012324364591882234, + "loss": 0.1152, + "step": 44035 + }, + { + "epoch": 0.3822536262706053, + "grad_norm": 0.36328125, + "learning_rate": 0.0012324070206547737, + "loss": 0.0972, + "step": 44036 + }, + { + "epoch": 0.38226230675080947, + "grad_norm": 0.224609375, + "learning_rate": 0.0012323775819765258, + "loss": 0.0762, + "step": 44037 + }, + { + "epoch": 0.3822709872310136, + "grad_norm": 0.259765625, + "learning_rate": 0.0012323481431535113, + "loss": 0.123, + "step": 44038 + }, + { + "epoch": 0.3822796677112178, + "grad_norm": 0.5234375, + "learning_rate": 0.0012323187041857633, + "loss": 0.0923, + "step": 44039 + }, + { + "epoch": 0.38228834819142193, + "grad_norm": 0.546875, + "learning_rate": 0.0012322892650733133, + "loss": 0.1108, + "step": 44040 + }, + { + "epoch": 0.3822970286716261, + "grad_norm": 0.296875, + "learning_rate": 0.001232259825816194, + "loss": 0.0615, + "step": 44041 + }, + { + "epoch": 0.38230570915183026, + "grad_norm": 0.455078125, + "learning_rate": 0.0012322303864144369, + "loss": 0.1328, + "step": 44042 + }, + { + "epoch": 0.38231438963203446, + "grad_norm": 0.1572265625, + "learning_rate": 0.0012322009468680748, + "loss": 0.127, + "step": 44043 + }, + { + "epoch": 0.3823230701122386, + "grad_norm": 0.42578125, + "learning_rate": 0.0012321715071771397, + "loss": 0.1123, + "step": 44044 + }, + { + "epoch": 0.3823317505924428, + "grad_norm": 0.87109375, + "learning_rate": 0.0012321420673416642, + "loss": 0.1426, + "step": 44045 + }, + { + "epoch": 0.3823404310726469, + "grad_norm": 0.3359375, + "learning_rate": 0.00123211262736168, + "loss": 0.1128, + "step": 44046 + }, + { + "epoch": 0.3823491115528511, + "grad_norm": 0.19140625, + "learning_rate": 0.0012320831872372194, + "loss": 0.1182, + "step": 44047 + }, + { + "epoch": 0.38235779203305525, + "grad_norm": 0.384765625, + "learning_rate": 0.0012320537469683141, + "loss": 0.0977, + "step": 44048 + }, + { + "epoch": 0.38236647251325945, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012320243065549975, + "loss": 0.1006, + "step": 44049 + }, + { + "epoch": 0.3823751529934636, + "grad_norm": 0.1650390625, + "learning_rate": 0.001231994865997301, + "loss": 0.1074, + "step": 44050 + }, + { + "epoch": 0.3823838334736678, + "grad_norm": 0.4375, + "learning_rate": 0.0012319654252952565, + "loss": 0.1172, + "step": 44051 + }, + { + "epoch": 0.3823925139538719, + "grad_norm": 0.423828125, + "learning_rate": 0.001231935984448897, + "loss": 0.0581, + "step": 44052 + }, + { + "epoch": 0.3824011944340761, + "grad_norm": 0.158203125, + "learning_rate": 0.0012319065434582537, + "loss": 0.125, + "step": 44053 + }, + { + "epoch": 0.38240987491428025, + "grad_norm": 0.44140625, + "learning_rate": 0.00123187710232336, + "loss": 0.0669, + "step": 44054 + }, + { + "epoch": 0.38241855539448444, + "grad_norm": 0.453125, + "learning_rate": 0.0012318476610442472, + "loss": 0.1011, + "step": 44055 + }, + { + "epoch": 0.3824272358746886, + "grad_norm": 0.162109375, + "learning_rate": 0.001231818219620948, + "loss": 0.1289, + "step": 44056 + }, + { + "epoch": 0.38243591635489277, + "grad_norm": 0.28125, + "learning_rate": 0.0012317887780534942, + "loss": 0.0928, + "step": 44057 + }, + { + "epoch": 0.3824445968350969, + "grad_norm": 0.162109375, + "learning_rate": 0.0012317593363419179, + "loss": 0.0957, + "step": 44058 + }, + { + "epoch": 0.3824532773153011, + "grad_norm": 0.1201171875, + "learning_rate": 0.001231729894486252, + "loss": 0.1143, + "step": 44059 + }, + { + "epoch": 0.38246195779550524, + "grad_norm": 0.154296875, + "learning_rate": 0.0012317004524865284, + "loss": 0.1055, + "step": 44060 + }, + { + "epoch": 0.38247063827570943, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012316710103427789, + "loss": 0.1035, + "step": 44061 + }, + { + "epoch": 0.38247931875591357, + "grad_norm": 0.057861328125, + "learning_rate": 0.0012316415680550354, + "loss": 0.0674, + "step": 44062 + }, + { + "epoch": 0.38248799923611776, + "grad_norm": 0.466796875, + "learning_rate": 0.0012316121256233314, + "loss": 0.1035, + "step": 44063 + }, + { + "epoch": 0.3824966797163219, + "grad_norm": 0.12109375, + "learning_rate": 0.0012315826830476984, + "loss": 0.0972, + "step": 44064 + }, + { + "epoch": 0.3825053601965261, + "grad_norm": 0.3984375, + "learning_rate": 0.0012315532403281684, + "loss": 0.127, + "step": 44065 + }, + { + "epoch": 0.3825140406767302, + "grad_norm": 0.12158203125, + "learning_rate": 0.0012315237974647732, + "loss": 0.0811, + "step": 44066 + }, + { + "epoch": 0.3825227211569344, + "grad_norm": 0.302734375, + "learning_rate": 0.0012314943544575463, + "loss": 0.1211, + "step": 44067 + }, + { + "epoch": 0.38253140163713856, + "grad_norm": 0.125, + "learning_rate": 0.001231464911306519, + "loss": 0.0894, + "step": 44068 + }, + { + "epoch": 0.38254008211734275, + "grad_norm": 0.2451171875, + "learning_rate": 0.0012314354680117235, + "loss": 0.0645, + "step": 44069 + }, + { + "epoch": 0.3825487625975469, + "grad_norm": 0.314453125, + "learning_rate": 0.001231406024573192, + "loss": 0.1221, + "step": 44070 + }, + { + "epoch": 0.3825574430777511, + "grad_norm": 0.10546875, + "learning_rate": 0.001231376580990957, + "loss": 0.1035, + "step": 44071 + }, + { + "epoch": 0.3825661235579552, + "grad_norm": 0.1826171875, + "learning_rate": 0.0012313471372650505, + "loss": 0.123, + "step": 44072 + }, + { + "epoch": 0.3825748040381594, + "grad_norm": 0.60546875, + "learning_rate": 0.001231317693395505, + "loss": 0.0996, + "step": 44073 + }, + { + "epoch": 0.38258348451836355, + "grad_norm": 0.18359375, + "learning_rate": 0.001231288249382352, + "loss": 0.0659, + "step": 44074 + }, + { + "epoch": 0.38259216499856774, + "grad_norm": 0.48046875, + "learning_rate": 0.0012312588052256245, + "loss": 0.126, + "step": 44075 + }, + { + "epoch": 0.3826008454787719, + "grad_norm": 0.400390625, + "learning_rate": 0.0012312293609253544, + "loss": 0.0889, + "step": 44076 + }, + { + "epoch": 0.38260952595897607, + "grad_norm": 0.4296875, + "learning_rate": 0.0012311999164815737, + "loss": 0.1157, + "step": 44077 + }, + { + "epoch": 0.3826182064391802, + "grad_norm": 0.56640625, + "learning_rate": 0.0012311704718943148, + "loss": 0.2324, + "step": 44078 + }, + { + "epoch": 0.3826268869193844, + "grad_norm": 0.322265625, + "learning_rate": 0.0012311410271636097, + "loss": 0.0977, + "step": 44079 + }, + { + "epoch": 0.38263556739958854, + "grad_norm": 0.400390625, + "learning_rate": 0.001231111582289491, + "loss": 0.1016, + "step": 44080 + }, + { + "epoch": 0.38264424787979273, + "grad_norm": 0.41015625, + "learning_rate": 0.00123108213727199, + "loss": 0.0845, + "step": 44081 + }, + { + "epoch": 0.38265292835999687, + "grad_norm": 0.279296875, + "learning_rate": 0.0012310526921111402, + "loss": 0.1416, + "step": 44082 + }, + { + "epoch": 0.38266160884020106, + "grad_norm": 0.24609375, + "learning_rate": 0.0012310232468069726, + "loss": 0.1533, + "step": 44083 + }, + { + "epoch": 0.3826702893204052, + "grad_norm": 0.373046875, + "learning_rate": 0.0012309938013595205, + "loss": 0.0869, + "step": 44084 + }, + { + "epoch": 0.3826789698006094, + "grad_norm": 0.2021484375, + "learning_rate": 0.0012309643557688154, + "loss": 0.0898, + "step": 44085 + }, + { + "epoch": 0.38268765028081353, + "grad_norm": 0.470703125, + "learning_rate": 0.0012309349100348896, + "loss": 0.0767, + "step": 44086 + }, + { + "epoch": 0.3826963307610177, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012309054641577752, + "loss": 0.0713, + "step": 44087 + }, + { + "epoch": 0.38270501124122186, + "grad_norm": 0.66796875, + "learning_rate": 0.0012308760181375046, + "loss": 0.1104, + "step": 44088 + }, + { + "epoch": 0.38271369172142605, + "grad_norm": 0.5703125, + "learning_rate": 0.0012308465719741105, + "loss": 0.084, + "step": 44089 + }, + { + "epoch": 0.3827223722016302, + "grad_norm": 0.197265625, + "learning_rate": 0.0012308171256676238, + "loss": 0.0962, + "step": 44090 + }, + { + "epoch": 0.3827310526818344, + "grad_norm": 0.1474609375, + "learning_rate": 0.001230787679218078, + "loss": 0.1279, + "step": 44091 + }, + { + "epoch": 0.3827397331620385, + "grad_norm": 0.06884765625, + "learning_rate": 0.0012307582326255043, + "loss": 0.0781, + "step": 44092 + }, + { + "epoch": 0.3827484136422427, + "grad_norm": 0.40234375, + "learning_rate": 0.0012307287858899353, + "loss": 0.1138, + "step": 44093 + }, + { + "epoch": 0.38275709412244685, + "grad_norm": 0.1630859375, + "learning_rate": 0.0012306993390114038, + "loss": 0.0747, + "step": 44094 + }, + { + "epoch": 0.38276577460265104, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012306698919899413, + "loss": 0.0864, + "step": 44095 + }, + { + "epoch": 0.3827744550828552, + "grad_norm": 0.3203125, + "learning_rate": 0.0012306404448255799, + "loss": 0.1035, + "step": 44096 + }, + { + "epoch": 0.3827831355630594, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012306109975183525, + "loss": 0.0933, + "step": 44097 + }, + { + "epoch": 0.3827918160432635, + "grad_norm": 0.087890625, + "learning_rate": 0.0012305815500682905, + "loss": 0.0835, + "step": 44098 + }, + { + "epoch": 0.3828004965234677, + "grad_norm": 0.1689453125, + "learning_rate": 0.0012305521024754266, + "loss": 0.1084, + "step": 44099 + }, + { + "epoch": 0.38280917700367184, + "grad_norm": 0.146484375, + "learning_rate": 0.0012305226547397927, + "loss": 0.1123, + "step": 44100 + }, + { + "epoch": 0.38281785748387603, + "grad_norm": 0.25, + "learning_rate": 0.0012304932068614215, + "loss": 0.1338, + "step": 44101 + }, + { + "epoch": 0.38282653796408017, + "grad_norm": 0.431640625, + "learning_rate": 0.0012304637588403447, + "loss": 0.0859, + "step": 44102 + }, + { + "epoch": 0.38283521844428436, + "grad_norm": 1.1171875, + "learning_rate": 0.001230434310676595, + "loss": 0.1094, + "step": 44103 + }, + { + "epoch": 0.3828438989244885, + "grad_norm": 0.287109375, + "learning_rate": 0.0012304048623702037, + "loss": 0.1021, + "step": 44104 + }, + { + "epoch": 0.3828525794046927, + "grad_norm": 2.234375, + "learning_rate": 0.0012303754139212043, + "loss": 0.3027, + "step": 44105 + }, + { + "epoch": 0.38286125988489683, + "grad_norm": 0.302734375, + "learning_rate": 0.0012303459653296275, + "loss": 0.125, + "step": 44106 + }, + { + "epoch": 0.382869940365101, + "grad_norm": 0.30078125, + "learning_rate": 0.001230316516595507, + "loss": 0.1162, + "step": 44107 + }, + { + "epoch": 0.38287862084530516, + "grad_norm": 0.1064453125, + "learning_rate": 0.0012302870677188742, + "loss": 0.0752, + "step": 44108 + }, + { + "epoch": 0.38288730132550935, + "grad_norm": 0.30078125, + "learning_rate": 0.0012302576186997611, + "loss": 0.1182, + "step": 44109 + }, + { + "epoch": 0.3828959818057135, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012302281695382002, + "loss": 0.085, + "step": 44110 + }, + { + "epoch": 0.38290466228591763, + "grad_norm": 0.31640625, + "learning_rate": 0.001230198720234224, + "loss": 0.0781, + "step": 44111 + }, + { + "epoch": 0.3829133427661218, + "grad_norm": 0.146484375, + "learning_rate": 0.0012301692707878647, + "loss": 0.1123, + "step": 44112 + }, + { + "epoch": 0.38292202324632596, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012301398211991537, + "loss": 0.0889, + "step": 44113 + }, + { + "epoch": 0.38293070372653015, + "grad_norm": 0.1552734375, + "learning_rate": 0.001230110371468124, + "loss": 0.0781, + "step": 44114 + }, + { + "epoch": 0.3829393842067343, + "grad_norm": 0.10546875, + "learning_rate": 0.0012300809215948076, + "loss": 0.1328, + "step": 44115 + }, + { + "epoch": 0.3829480646869385, + "grad_norm": 0.177734375, + "learning_rate": 0.0012300514715792368, + "loss": 0.0913, + "step": 44116 + }, + { + "epoch": 0.3829567451671426, + "grad_norm": 0.1640625, + "learning_rate": 0.0012300220214214435, + "loss": 0.0947, + "step": 44117 + }, + { + "epoch": 0.3829654256473468, + "grad_norm": 0.28125, + "learning_rate": 0.0012299925711214599, + "loss": 0.1064, + "step": 44118 + }, + { + "epoch": 0.38297410612755095, + "grad_norm": 0.2197265625, + "learning_rate": 0.0012299631206793183, + "loss": 0.1006, + "step": 44119 + }, + { + "epoch": 0.38298278660775514, + "grad_norm": 0.23828125, + "learning_rate": 0.0012299336700950514, + "loss": 0.0981, + "step": 44120 + }, + { + "epoch": 0.3829914670879593, + "grad_norm": 0.353515625, + "learning_rate": 0.0012299042193686907, + "loss": 0.1069, + "step": 44121 + }, + { + "epoch": 0.3830001475681635, + "grad_norm": 0.18359375, + "learning_rate": 0.0012298747685002686, + "loss": 0.064, + "step": 44122 + }, + { + "epoch": 0.3830088280483676, + "grad_norm": 0.70703125, + "learning_rate": 0.0012298453174898179, + "loss": 0.1221, + "step": 44123 + }, + { + "epoch": 0.3830175085285718, + "grad_norm": 0.173828125, + "learning_rate": 0.0012298158663373697, + "loss": 0.0879, + "step": 44124 + }, + { + "epoch": 0.38302618900877594, + "grad_norm": 0.09912109375, + "learning_rate": 0.0012297864150429573, + "loss": 0.1123, + "step": 44125 + }, + { + "epoch": 0.38303486948898013, + "grad_norm": 0.232421875, + "learning_rate": 0.001229756963606612, + "loss": 0.0938, + "step": 44126 + }, + { + "epoch": 0.38304354996918427, + "grad_norm": 0.337890625, + "learning_rate": 0.0012297275120283669, + "loss": 0.1182, + "step": 44127 + }, + { + "epoch": 0.38305223044938846, + "grad_norm": 0.3125, + "learning_rate": 0.0012296980603082534, + "loss": 0.1055, + "step": 44128 + }, + { + "epoch": 0.3830609109295926, + "grad_norm": 0.1552734375, + "learning_rate": 0.001229668608446304, + "loss": 0.1157, + "step": 44129 + }, + { + "epoch": 0.3830695914097968, + "grad_norm": 0.08447265625, + "learning_rate": 0.001229639156442551, + "loss": 0.1074, + "step": 44130 + }, + { + "epoch": 0.38307827189000093, + "grad_norm": 0.1650390625, + "learning_rate": 0.001229609704297027, + "loss": 0.0928, + "step": 44131 + }, + { + "epoch": 0.3830869523702051, + "grad_norm": 0.3359375, + "learning_rate": 0.0012295802520097634, + "loss": 0.1025, + "step": 44132 + }, + { + "epoch": 0.38309563285040926, + "grad_norm": 0.06787109375, + "learning_rate": 0.0012295507995807931, + "loss": 0.0835, + "step": 44133 + }, + { + "epoch": 0.38310431333061346, + "grad_norm": 0.1083984375, + "learning_rate": 0.0012295213470101475, + "loss": 0.0796, + "step": 44134 + }, + { + "epoch": 0.3831129938108176, + "grad_norm": 0.216796875, + "learning_rate": 0.0012294918942978596, + "loss": 0.1113, + "step": 44135 + }, + { + "epoch": 0.3831216742910218, + "grad_norm": 0.3984375, + "learning_rate": 0.0012294624414439612, + "loss": 0.0874, + "step": 44136 + }, + { + "epoch": 0.3831303547712259, + "grad_norm": 0.609375, + "learning_rate": 0.0012294329884484849, + "loss": 0.1367, + "step": 44137 + }, + { + "epoch": 0.3831390352514301, + "grad_norm": 0.31640625, + "learning_rate": 0.001229403535311462, + "loss": 0.0688, + "step": 44138 + }, + { + "epoch": 0.38314771573163425, + "grad_norm": 0.34765625, + "learning_rate": 0.0012293740820329257, + "loss": 0.1201, + "step": 44139 + }, + { + "epoch": 0.38315639621183845, + "grad_norm": 0.11279296875, + "learning_rate": 0.0012293446286129079, + "loss": 0.0674, + "step": 44140 + }, + { + "epoch": 0.3831650766920426, + "grad_norm": 0.328125, + "learning_rate": 0.0012293151750514411, + "loss": 0.1328, + "step": 44141 + }, + { + "epoch": 0.3831737571722468, + "grad_norm": 0.18359375, + "learning_rate": 0.0012292857213485567, + "loss": 0.1113, + "step": 44142 + }, + { + "epoch": 0.3831824376524509, + "grad_norm": 0.330078125, + "learning_rate": 0.0012292562675042874, + "loss": 0.1074, + "step": 44143 + }, + { + "epoch": 0.3831911181326551, + "grad_norm": 0.65625, + "learning_rate": 0.0012292268135186656, + "loss": 0.0713, + "step": 44144 + }, + { + "epoch": 0.38319979861285924, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012291973593917232, + "loss": 0.0972, + "step": 44145 + }, + { + "epoch": 0.38320847909306344, + "grad_norm": 0.078125, + "learning_rate": 0.0012291679051234926, + "loss": 0.1113, + "step": 44146 + }, + { + "epoch": 0.3832171595732676, + "grad_norm": 0.1884765625, + "learning_rate": 0.0012291384507140055, + "loss": 0.084, + "step": 44147 + }, + { + "epoch": 0.38322584005347177, + "grad_norm": 0.2421875, + "learning_rate": 0.0012291089961632949, + "loss": 0.1064, + "step": 44148 + }, + { + "epoch": 0.3832345205336759, + "grad_norm": 0.73828125, + "learning_rate": 0.0012290795414713924, + "loss": 0.0889, + "step": 44149 + }, + { + "epoch": 0.3832432010138801, + "grad_norm": 0.19921875, + "learning_rate": 0.0012290500866383308, + "loss": 0.1021, + "step": 44150 + }, + { + "epoch": 0.38325188149408423, + "grad_norm": 0.41015625, + "learning_rate": 0.0012290206316641417, + "loss": 0.1855, + "step": 44151 + }, + { + "epoch": 0.3832605619742884, + "grad_norm": 0.53125, + "learning_rate": 0.0012289911765488578, + "loss": 0.0952, + "step": 44152 + }, + { + "epoch": 0.38326924245449256, + "grad_norm": 0.87890625, + "learning_rate": 0.0012289617212925107, + "loss": 0.1177, + "step": 44153 + }, + { + "epoch": 0.38327792293469676, + "grad_norm": 0.19140625, + "learning_rate": 0.0012289322658951335, + "loss": 0.1123, + "step": 44154 + }, + { + "epoch": 0.3832866034149009, + "grad_norm": 0.11474609375, + "learning_rate": 0.0012289028103567576, + "loss": 0.1196, + "step": 44155 + }, + { + "epoch": 0.3832952838951051, + "grad_norm": 0.1640625, + "learning_rate": 0.0012288733546774156, + "loss": 0.1138, + "step": 44156 + }, + { + "epoch": 0.3833039643753092, + "grad_norm": 0.150390625, + "learning_rate": 0.0012288438988571398, + "loss": 0.0898, + "step": 44157 + }, + { + "epoch": 0.3833126448555134, + "grad_norm": 0.41796875, + "learning_rate": 0.0012288144428959618, + "loss": 0.0918, + "step": 44158 + }, + { + "epoch": 0.38332132533571756, + "grad_norm": 0.267578125, + "learning_rate": 0.0012287849867939145, + "loss": 0.1235, + "step": 44159 + }, + { + "epoch": 0.38333000581592175, + "grad_norm": 0.1982421875, + "learning_rate": 0.00122875553055103, + "loss": 0.1348, + "step": 44160 + }, + { + "epoch": 0.3833386862961259, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012287260741673403, + "loss": 0.0938, + "step": 44161 + }, + { + "epoch": 0.3833473667763301, + "grad_norm": 0.498046875, + "learning_rate": 0.0012286966176428776, + "loss": 0.082, + "step": 44162 + }, + { + "epoch": 0.3833560472565342, + "grad_norm": 0.1005859375, + "learning_rate": 0.0012286671609776745, + "loss": 0.1123, + "step": 44163 + }, + { + "epoch": 0.3833647277367384, + "grad_norm": 0.2412109375, + "learning_rate": 0.0012286377041717627, + "loss": 0.1104, + "step": 44164 + }, + { + "epoch": 0.38337340821694255, + "grad_norm": 0.37890625, + "learning_rate": 0.0012286082472251745, + "loss": 0.1133, + "step": 44165 + }, + { + "epoch": 0.38338208869714674, + "grad_norm": 0.19921875, + "learning_rate": 0.0012285787901379426, + "loss": 0.1084, + "step": 44166 + }, + { + "epoch": 0.3833907691773509, + "grad_norm": 0.498046875, + "learning_rate": 0.0012285493329100986, + "loss": 0.127, + "step": 44167 + }, + { + "epoch": 0.38339944965755507, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012285198755416751, + "loss": 0.0718, + "step": 44168 + }, + { + "epoch": 0.3834081301377592, + "grad_norm": 0.294921875, + "learning_rate": 0.001228490418032704, + "loss": 0.0684, + "step": 44169 + }, + { + "epoch": 0.3834168106179634, + "grad_norm": 0.306640625, + "learning_rate": 0.001228460960383218, + "loss": 0.0933, + "step": 44170 + }, + { + "epoch": 0.38342549109816754, + "grad_norm": 0.416015625, + "learning_rate": 0.0012284315025932488, + "loss": 0.0713, + "step": 44171 + }, + { + "epoch": 0.38343417157837173, + "grad_norm": 0.16796875, + "learning_rate": 0.001228402044662829, + "loss": 0.0864, + "step": 44172 + }, + { + "epoch": 0.38344285205857587, + "grad_norm": 0.087890625, + "learning_rate": 0.0012283725865919906, + "loss": 0.1011, + "step": 44173 + }, + { + "epoch": 0.38345153253878006, + "grad_norm": 0.421875, + "learning_rate": 0.001228343128380766, + "loss": 0.0732, + "step": 44174 + }, + { + "epoch": 0.3834602130189842, + "grad_norm": 0.275390625, + "learning_rate": 0.001228313670029187, + "loss": 0.083, + "step": 44175 + }, + { + "epoch": 0.3834688934991884, + "grad_norm": 0.1591796875, + "learning_rate": 0.001228284211537286, + "loss": 0.0962, + "step": 44176 + }, + { + "epoch": 0.3834775739793925, + "grad_norm": 0.1728515625, + "learning_rate": 0.0012282547529050957, + "loss": 0.0913, + "step": 44177 + }, + { + "epoch": 0.3834862544595967, + "grad_norm": 0.65234375, + "learning_rate": 0.0012282252941326478, + "loss": 0.1279, + "step": 44178 + }, + { + "epoch": 0.38349493493980086, + "grad_norm": 0.2255859375, + "learning_rate": 0.0012281958352199744, + "loss": 0.1196, + "step": 44179 + }, + { + "epoch": 0.38350361542000505, + "grad_norm": 0.439453125, + "learning_rate": 0.0012281663761671085, + "loss": 0.1396, + "step": 44180 + }, + { + "epoch": 0.3835122959002092, + "grad_norm": 0.08544921875, + "learning_rate": 0.0012281369169740814, + "loss": 0.085, + "step": 44181 + }, + { + "epoch": 0.3835209763804134, + "grad_norm": 0.5703125, + "learning_rate": 0.0012281074576409258, + "loss": 0.0732, + "step": 44182 + }, + { + "epoch": 0.3835296568606175, + "grad_norm": 0.1845703125, + "learning_rate": 0.0012280779981676737, + "loss": 0.1138, + "step": 44183 + }, + { + "epoch": 0.3835383373408217, + "grad_norm": 0.150390625, + "learning_rate": 0.0012280485385543576, + "loss": 0.207, + "step": 44184 + }, + { + "epoch": 0.38354701782102585, + "grad_norm": 0.75390625, + "learning_rate": 0.0012280190788010095, + "loss": 0.0918, + "step": 44185 + }, + { + "epoch": 0.38355569830123004, + "grad_norm": 0.318359375, + "learning_rate": 0.0012279896189076615, + "loss": 0.0947, + "step": 44186 + }, + { + "epoch": 0.3835643787814342, + "grad_norm": 0.212890625, + "learning_rate": 0.001227960158874346, + "loss": 0.127, + "step": 44187 + }, + { + "epoch": 0.38357305926163837, + "grad_norm": 0.34375, + "learning_rate": 0.0012279306987010952, + "loss": 0.1631, + "step": 44188 + }, + { + "epoch": 0.3835817397418425, + "grad_norm": 0.08154296875, + "learning_rate": 0.0012279012383879413, + "loss": 0.0737, + "step": 44189 + }, + { + "epoch": 0.3835904202220467, + "grad_norm": 0.6328125, + "learning_rate": 0.001227871777934917, + "loss": 0.1279, + "step": 44190 + }, + { + "epoch": 0.38359910070225084, + "grad_norm": 0.56640625, + "learning_rate": 0.0012278423173420533, + "loss": 0.0869, + "step": 44191 + }, + { + "epoch": 0.38360778118245503, + "grad_norm": 0.400390625, + "learning_rate": 0.0012278128566093835, + "loss": 0.1143, + "step": 44192 + }, + { + "epoch": 0.38361646166265917, + "grad_norm": 0.1318359375, + "learning_rate": 0.0012277833957369396, + "loss": 0.0776, + "step": 44193 + }, + { + "epoch": 0.38362514214286336, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012277539347247538, + "loss": 0.0879, + "step": 44194 + }, + { + "epoch": 0.3836338226230675, + "grad_norm": 0.392578125, + "learning_rate": 0.0012277244735728577, + "loss": 0.0918, + "step": 44195 + }, + { + "epoch": 0.3836425031032717, + "grad_norm": 0.1728515625, + "learning_rate": 0.0012276950122812843, + "loss": 0.1221, + "step": 44196 + }, + { + "epoch": 0.38365118358347583, + "grad_norm": 0.388671875, + "learning_rate": 0.0012276655508500656, + "loss": 0.1162, + "step": 44197 + }, + { + "epoch": 0.38365986406368, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012276360892792338, + "loss": 0.1602, + "step": 44198 + }, + { + "epoch": 0.38366854454388416, + "grad_norm": 0.1416015625, + "learning_rate": 0.0012276066275688211, + "loss": 0.0728, + "step": 44199 + }, + { + "epoch": 0.38367722502408835, + "grad_norm": 0.427734375, + "learning_rate": 0.0012275771657188595, + "loss": 0.1211, + "step": 44200 + }, + { + "epoch": 0.3836859055042925, + "grad_norm": 0.0791015625, + "learning_rate": 0.001227547703729382, + "loss": 0.0996, + "step": 44201 + }, + { + "epoch": 0.3836945859844967, + "grad_norm": 0.56640625, + "learning_rate": 0.0012275182416004197, + "loss": 0.0996, + "step": 44202 + }, + { + "epoch": 0.3837032664647008, + "grad_norm": 0.1923828125, + "learning_rate": 0.0012274887793320055, + "loss": 0.1113, + "step": 44203 + }, + { + "epoch": 0.383711946944905, + "grad_norm": 0.2158203125, + "learning_rate": 0.0012274593169241716, + "loss": 0.083, + "step": 44204 + }, + { + "epoch": 0.38372062742510915, + "grad_norm": 0.384765625, + "learning_rate": 0.00122742985437695, + "loss": 0.0933, + "step": 44205 + }, + { + "epoch": 0.38372930790531334, + "grad_norm": 0.3359375, + "learning_rate": 0.0012274003916903726, + "loss": 0.0835, + "step": 44206 + }, + { + "epoch": 0.3837379883855175, + "grad_norm": 0.85546875, + "learning_rate": 0.0012273709288644725, + "loss": 0.1748, + "step": 44207 + }, + { + "epoch": 0.3837466688657217, + "grad_norm": 0.14453125, + "learning_rate": 0.0012273414658992815, + "loss": 0.0996, + "step": 44208 + }, + { + "epoch": 0.3837553493459258, + "grad_norm": 0.244140625, + "learning_rate": 0.0012273120027948318, + "loss": 0.1074, + "step": 44209 + }, + { + "epoch": 0.38376402982613, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012272825395511556, + "loss": 0.0728, + "step": 44210 + }, + { + "epoch": 0.38377271030633414, + "grad_norm": 0.2236328125, + "learning_rate": 0.001227253076168285, + "loss": 0.0693, + "step": 44211 + }, + { + "epoch": 0.38378139078653833, + "grad_norm": 0.11376953125, + "learning_rate": 0.0012272236126462525, + "loss": 0.1104, + "step": 44212 + }, + { + "epoch": 0.38379007126674247, + "grad_norm": 0.138671875, + "learning_rate": 0.00122719414898509, + "loss": 0.0737, + "step": 44213 + }, + { + "epoch": 0.38379875174694666, + "grad_norm": 0.28515625, + "learning_rate": 0.0012271646851848299, + "loss": 0.0825, + "step": 44214 + }, + { + "epoch": 0.3838074322271508, + "grad_norm": 0.5546875, + "learning_rate": 0.0012271352212455044, + "loss": 0.0884, + "step": 44215 + }, + { + "epoch": 0.383816112707355, + "grad_norm": 0.25390625, + "learning_rate": 0.001227105757167146, + "loss": 0.0869, + "step": 44216 + }, + { + "epoch": 0.38382479318755913, + "grad_norm": 0.85546875, + "learning_rate": 0.0012270762929497863, + "loss": 0.1543, + "step": 44217 + }, + { + "epoch": 0.3838334736677633, + "grad_norm": 0.2158203125, + "learning_rate": 0.001227046828593458, + "loss": 0.1074, + "step": 44218 + }, + { + "epoch": 0.38384215414796746, + "grad_norm": 0.259765625, + "learning_rate": 0.0012270173640981932, + "loss": 0.1357, + "step": 44219 + }, + { + "epoch": 0.38385083462817166, + "grad_norm": 0.306640625, + "learning_rate": 0.0012269878994640242, + "loss": 0.106, + "step": 44220 + }, + { + "epoch": 0.3838595151083758, + "grad_norm": 0.1279296875, + "learning_rate": 0.001226958434690983, + "loss": 0.1182, + "step": 44221 + }, + { + "epoch": 0.38386819558858, + "grad_norm": 0.2080078125, + "learning_rate": 0.0012269289697791023, + "loss": 0.1143, + "step": 44222 + }, + { + "epoch": 0.3838768760687841, + "grad_norm": 0.07470703125, + "learning_rate": 0.0012268995047284136, + "loss": 0.0835, + "step": 44223 + }, + { + "epoch": 0.3838855565489883, + "grad_norm": 0.52734375, + "learning_rate": 0.0012268700395389495, + "loss": 0.0991, + "step": 44224 + }, + { + "epoch": 0.38389423702919245, + "grad_norm": 0.404296875, + "learning_rate": 0.0012268405742107426, + "loss": 0.1021, + "step": 44225 + }, + { + "epoch": 0.38390291750939665, + "grad_norm": 0.458984375, + "learning_rate": 0.0012268111087438242, + "loss": 0.127, + "step": 44226 + }, + { + "epoch": 0.3839115979896008, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012267816431382275, + "loss": 0.1113, + "step": 44227 + }, + { + "epoch": 0.383920278469805, + "grad_norm": 0.1923828125, + "learning_rate": 0.0012267521773939843, + "loss": 0.1108, + "step": 44228 + }, + { + "epoch": 0.3839289589500091, + "grad_norm": 0.232421875, + "learning_rate": 0.001226722711511127, + "loss": 0.0928, + "step": 44229 + }, + { + "epoch": 0.3839376394302133, + "grad_norm": 0.5234375, + "learning_rate": 0.001226693245489687, + "loss": 0.084, + "step": 44230 + }, + { + "epoch": 0.38394631991041744, + "grad_norm": 0.08984375, + "learning_rate": 0.0012266637793296978, + "loss": 0.1025, + "step": 44231 + }, + { + "epoch": 0.38395500039062164, + "grad_norm": 0.25390625, + "learning_rate": 0.0012266343130311907, + "loss": 0.0742, + "step": 44232 + }, + { + "epoch": 0.3839636808708258, + "grad_norm": 0.10400390625, + "learning_rate": 0.0012266048465941985, + "loss": 0.1348, + "step": 44233 + }, + { + "epoch": 0.3839723613510299, + "grad_norm": 0.244140625, + "learning_rate": 0.0012265753800187525, + "loss": 0.1182, + "step": 44234 + }, + { + "epoch": 0.3839810418312341, + "grad_norm": 1.8359375, + "learning_rate": 0.001226545913304886, + "loss": 0.1143, + "step": 44235 + }, + { + "epoch": 0.38398972231143824, + "grad_norm": 0.41015625, + "learning_rate": 0.001226516446452631, + "loss": 0.0894, + "step": 44236 + }, + { + "epoch": 0.38399840279164243, + "grad_norm": 0.578125, + "learning_rate": 0.0012264869794620192, + "loss": 0.1045, + "step": 44237 + }, + { + "epoch": 0.38400708327184657, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012264575123330833, + "loss": 0.0952, + "step": 44238 + }, + { + "epoch": 0.38401576375205076, + "grad_norm": 0.1796875, + "learning_rate": 0.0012264280450658552, + "loss": 0.0996, + "step": 44239 + }, + { + "epoch": 0.3840244442322549, + "grad_norm": 0.306640625, + "learning_rate": 0.0012263985776603673, + "loss": 0.0864, + "step": 44240 + }, + { + "epoch": 0.3840331247124591, + "grad_norm": 0.09765625, + "learning_rate": 0.0012263691101166518, + "loss": 0.0908, + "step": 44241 + }, + { + "epoch": 0.38404180519266323, + "grad_norm": 0.400390625, + "learning_rate": 0.0012263396424347412, + "loss": 0.0889, + "step": 44242 + }, + { + "epoch": 0.3840504856728674, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012263101746146671, + "loss": 0.1387, + "step": 44243 + }, + { + "epoch": 0.38405916615307156, + "grad_norm": 0.171875, + "learning_rate": 0.0012262807066564622, + "loss": 0.0815, + "step": 44244 + }, + { + "epoch": 0.38406784663327576, + "grad_norm": 0.11767578125, + "learning_rate": 0.0012262512385601586, + "loss": 0.0864, + "step": 44245 + }, + { + "epoch": 0.3840765271134799, + "grad_norm": 0.67578125, + "learning_rate": 0.001226221770325789, + "loss": 0.0664, + "step": 44246 + }, + { + "epoch": 0.3840852075936841, + "grad_norm": 0.298828125, + "learning_rate": 0.0012261923019533845, + "loss": 0.123, + "step": 44247 + }, + { + "epoch": 0.3840938880738882, + "grad_norm": 1.0625, + "learning_rate": 0.0012261628334429783, + "loss": 0.1113, + "step": 44248 + }, + { + "epoch": 0.3841025685540924, + "grad_norm": 0.318359375, + "learning_rate": 0.0012261333647946024, + "loss": 0.0913, + "step": 44249 + }, + { + "epoch": 0.38411124903429655, + "grad_norm": 0.17578125, + "learning_rate": 0.0012261038960082891, + "loss": 0.0967, + "step": 44250 + }, + { + "epoch": 0.38411992951450075, + "grad_norm": 0.2001953125, + "learning_rate": 0.00122607442708407, + "loss": 0.124, + "step": 44251 + }, + { + "epoch": 0.3841286099947049, + "grad_norm": 0.1259765625, + "learning_rate": 0.001226044958021978, + "loss": 0.0864, + "step": 44252 + }, + { + "epoch": 0.3841372904749091, + "grad_norm": 0.06494140625, + "learning_rate": 0.0012260154888220453, + "loss": 0.0737, + "step": 44253 + }, + { + "epoch": 0.3841459709551132, + "grad_norm": 0.12060546875, + "learning_rate": 0.0012259860194843038, + "loss": 0.0942, + "step": 44254 + }, + { + "epoch": 0.3841546514353174, + "grad_norm": 0.41015625, + "learning_rate": 0.001225956550008786, + "loss": 0.0928, + "step": 44255 + }, + { + "epoch": 0.38416333191552154, + "grad_norm": 0.1591796875, + "learning_rate": 0.0012259270803955235, + "loss": 0.0791, + "step": 44256 + }, + { + "epoch": 0.38417201239572574, + "grad_norm": 0.1494140625, + "learning_rate": 0.0012258976106445498, + "loss": 0.1074, + "step": 44257 + }, + { + "epoch": 0.3841806928759299, + "grad_norm": 0.77734375, + "learning_rate": 0.001225868140755896, + "loss": 0.124, + "step": 44258 + }, + { + "epoch": 0.38418937335613407, + "grad_norm": 0.275390625, + "learning_rate": 0.001225838670729595, + "loss": 0.0957, + "step": 44259 + }, + { + "epoch": 0.3841980538363382, + "grad_norm": 0.080078125, + "learning_rate": 0.0012258092005656782, + "loss": 0.0776, + "step": 44260 + }, + { + "epoch": 0.3842067343165424, + "grad_norm": 0.0712890625, + "learning_rate": 0.0012257797302641787, + "loss": 0.0732, + "step": 44261 + }, + { + "epoch": 0.38421541479674653, + "grad_norm": 0.10546875, + "learning_rate": 0.0012257502598251283, + "loss": 0.0986, + "step": 44262 + }, + { + "epoch": 0.38422409527695073, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012257207892485589, + "loss": 0.1367, + "step": 44263 + }, + { + "epoch": 0.38423277575715487, + "grad_norm": 0.306640625, + "learning_rate": 0.0012256913185345036, + "loss": 0.1562, + "step": 44264 + }, + { + "epoch": 0.38424145623735906, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012256618476829942, + "loss": 0.0918, + "step": 44265 + }, + { + "epoch": 0.3842501367175632, + "grad_norm": 0.13671875, + "learning_rate": 0.0012256323766940626, + "loss": 0.1221, + "step": 44266 + }, + { + "epoch": 0.3842588171977674, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012256029055677416, + "loss": 0.0938, + "step": 44267 + }, + { + "epoch": 0.3842674976779715, + "grad_norm": 0.279296875, + "learning_rate": 0.001225573434304063, + "loss": 0.0977, + "step": 44268 + }, + { + "epoch": 0.3842761781581757, + "grad_norm": 0.1650390625, + "learning_rate": 0.001225543962903059, + "loss": 0.082, + "step": 44269 + }, + { + "epoch": 0.38428485863837986, + "grad_norm": 10.4375, + "learning_rate": 0.0012255144913647625, + "loss": 0.3105, + "step": 44270 + }, + { + "epoch": 0.38429353911858405, + "grad_norm": 0.66015625, + "learning_rate": 0.001225485019689205, + "loss": 0.1006, + "step": 44271 + }, + { + "epoch": 0.3843022195987882, + "grad_norm": 0.205078125, + "learning_rate": 0.0012254555478764188, + "loss": 0.1074, + "step": 44272 + }, + { + "epoch": 0.3843109000789924, + "grad_norm": 0.0859375, + "learning_rate": 0.0012254260759264363, + "loss": 0.1055, + "step": 44273 + }, + { + "epoch": 0.3843195805591965, + "grad_norm": 0.29296875, + "learning_rate": 0.0012253966038392897, + "loss": 0.1006, + "step": 44274 + }, + { + "epoch": 0.3843282610394007, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012253671316150114, + "loss": 0.1211, + "step": 44275 + }, + { + "epoch": 0.38433694151960485, + "grad_norm": 0.15234375, + "learning_rate": 0.0012253376592536336, + "loss": 0.0913, + "step": 44276 + }, + { + "epoch": 0.38434562199980904, + "grad_norm": 0.06494140625, + "learning_rate": 0.0012253081867551884, + "loss": 0.0708, + "step": 44277 + }, + { + "epoch": 0.3843543024800132, + "grad_norm": 0.388671875, + "learning_rate": 0.001225278714119708, + "loss": 0.1104, + "step": 44278 + }, + { + "epoch": 0.38436298296021737, + "grad_norm": 0.3671875, + "learning_rate": 0.0012252492413472246, + "loss": 0.0918, + "step": 44279 + }, + { + "epoch": 0.3843716634404215, + "grad_norm": 0.390625, + "learning_rate": 0.0012252197684377707, + "loss": 0.1133, + "step": 44280 + }, + { + "epoch": 0.3843803439206257, + "grad_norm": 0.09130859375, + "learning_rate": 0.0012251902953913782, + "loss": 0.0728, + "step": 44281 + }, + { + "epoch": 0.38438902440082984, + "grad_norm": 0.251953125, + "learning_rate": 0.0012251608222080793, + "loss": 0.1143, + "step": 44282 + }, + { + "epoch": 0.38439770488103403, + "grad_norm": 0.2158203125, + "learning_rate": 0.0012251313488879064, + "loss": 0.0806, + "step": 44283 + }, + { + "epoch": 0.38440638536123817, + "grad_norm": 1.0234375, + "learning_rate": 0.001225101875430892, + "loss": 0.0908, + "step": 44284 + }, + { + "epoch": 0.38441506584144236, + "grad_norm": 0.1357421875, + "learning_rate": 0.001225072401837068, + "loss": 0.1055, + "step": 44285 + }, + { + "epoch": 0.3844237463216465, + "grad_norm": 0.302734375, + "learning_rate": 0.0012250429281064667, + "loss": 0.1289, + "step": 44286 + }, + { + "epoch": 0.3844324268018507, + "grad_norm": 0.171875, + "learning_rate": 0.0012250134542391203, + "loss": 0.0796, + "step": 44287 + }, + { + "epoch": 0.38444110728205483, + "grad_norm": 0.271484375, + "learning_rate": 0.001224983980235061, + "loss": 0.1152, + "step": 44288 + }, + { + "epoch": 0.384449787762259, + "grad_norm": 0.3359375, + "learning_rate": 0.0012249545060943213, + "loss": 0.1992, + "step": 44289 + }, + { + "epoch": 0.38445846824246316, + "grad_norm": 0.1201171875, + "learning_rate": 0.001224925031816933, + "loss": 0.0771, + "step": 44290 + }, + { + "epoch": 0.38446714872266735, + "grad_norm": 0.2080078125, + "learning_rate": 0.0012248955574029289, + "loss": 0.0923, + "step": 44291 + }, + { + "epoch": 0.3844758292028715, + "grad_norm": 0.384765625, + "learning_rate": 0.0012248660828523405, + "loss": 0.1123, + "step": 44292 + }, + { + "epoch": 0.3844845096830757, + "grad_norm": 0.43359375, + "learning_rate": 0.0012248366081652005, + "loss": 0.1006, + "step": 44293 + }, + { + "epoch": 0.3844931901632798, + "grad_norm": 0.5, + "learning_rate": 0.0012248071333415412, + "loss": 0.0884, + "step": 44294 + }, + { + "epoch": 0.384501870643484, + "grad_norm": 0.29296875, + "learning_rate": 0.0012247776583813946, + "loss": 0.124, + "step": 44295 + }, + { + "epoch": 0.38451055112368815, + "grad_norm": 0.2890625, + "learning_rate": 0.001224748183284793, + "loss": 0.0933, + "step": 44296 + }, + { + "epoch": 0.38451923160389234, + "grad_norm": 0.498046875, + "learning_rate": 0.001224718708051769, + "loss": 0.1177, + "step": 44297 + }, + { + "epoch": 0.3845279120840965, + "grad_norm": 0.65234375, + "learning_rate": 0.0012246892326823542, + "loss": 0.0752, + "step": 44298 + }, + { + "epoch": 0.38453659256430067, + "grad_norm": 0.1640625, + "learning_rate": 0.0012246597571765809, + "loss": 0.064, + "step": 44299 + }, + { + "epoch": 0.3845452730445048, + "grad_norm": 0.0927734375, + "learning_rate": 0.001224630281534482, + "loss": 0.1123, + "step": 44300 + }, + { + "epoch": 0.384553953524709, + "grad_norm": 0.318359375, + "learning_rate": 0.001224600805756089, + "loss": 0.1094, + "step": 44301 + }, + { + "epoch": 0.38456263400491314, + "grad_norm": 0.3046875, + "learning_rate": 0.0012245713298414346, + "loss": 0.1211, + "step": 44302 + }, + { + "epoch": 0.38457131448511733, + "grad_norm": 0.228515625, + "learning_rate": 0.0012245418537905506, + "loss": 0.1084, + "step": 44303 + }, + { + "epoch": 0.38457999496532147, + "grad_norm": 0.05322265625, + "learning_rate": 0.0012245123776034696, + "loss": 0.0679, + "step": 44304 + }, + { + "epoch": 0.38458867544552566, + "grad_norm": 0.478515625, + "learning_rate": 0.001224482901280224, + "loss": 0.1299, + "step": 44305 + }, + { + "epoch": 0.3845973559257298, + "grad_norm": 0.498046875, + "learning_rate": 0.0012244534248208457, + "loss": 0.1191, + "step": 44306 + }, + { + "epoch": 0.384606036405934, + "grad_norm": 0.275390625, + "learning_rate": 0.001224423948225367, + "loss": 0.1133, + "step": 44307 + }, + { + "epoch": 0.38461471688613813, + "grad_norm": 0.7109375, + "learning_rate": 0.00122439447149382, + "loss": 0.0781, + "step": 44308 + }, + { + "epoch": 0.3846233973663423, + "grad_norm": 0.0859375, + "learning_rate": 0.0012243649946262372, + "loss": 0.0713, + "step": 44309 + }, + { + "epoch": 0.38463207784654646, + "grad_norm": 0.5546875, + "learning_rate": 0.0012243355176226505, + "loss": 0.0981, + "step": 44310 + }, + { + "epoch": 0.38464075832675065, + "grad_norm": 0.21875, + "learning_rate": 0.0012243060404830927, + "loss": 0.0938, + "step": 44311 + }, + { + "epoch": 0.3846494388069548, + "grad_norm": 0.1484375, + "learning_rate": 0.0012242765632075952, + "loss": 0.1001, + "step": 44312 + }, + { + "epoch": 0.384658119287159, + "grad_norm": 0.1904296875, + "learning_rate": 0.001224247085796191, + "loss": 0.1113, + "step": 44313 + }, + { + "epoch": 0.3846667997673631, + "grad_norm": 0.07958984375, + "learning_rate": 0.001224217608248912, + "loss": 0.123, + "step": 44314 + }, + { + "epoch": 0.3846754802475673, + "grad_norm": 0.435546875, + "learning_rate": 0.0012241881305657908, + "loss": 0.1064, + "step": 44315 + }, + { + "epoch": 0.38468416072777145, + "grad_norm": 0.2021484375, + "learning_rate": 0.0012241586527468589, + "loss": 0.1035, + "step": 44316 + }, + { + "epoch": 0.38469284120797564, + "grad_norm": 0.55078125, + "learning_rate": 0.001224129174792149, + "loss": 0.105, + "step": 44317 + }, + { + "epoch": 0.3847015216881798, + "grad_norm": 0.484375, + "learning_rate": 0.0012240996967016935, + "loss": 0.1006, + "step": 44318 + }, + { + "epoch": 0.384710202168384, + "grad_norm": 0.08154296875, + "learning_rate": 0.0012240702184755246, + "loss": 0.0859, + "step": 44319 + }, + { + "epoch": 0.3847188826485881, + "grad_norm": 0.1328125, + "learning_rate": 0.0012240407401136737, + "loss": 0.1152, + "step": 44320 + }, + { + "epoch": 0.3847275631287923, + "grad_norm": 0.103515625, + "learning_rate": 0.0012240112616161744, + "loss": 0.1133, + "step": 44321 + }, + { + "epoch": 0.38473624360899644, + "grad_norm": 0.07177734375, + "learning_rate": 0.001223981782983058, + "loss": 0.084, + "step": 44322 + }, + { + "epoch": 0.38474492408920064, + "grad_norm": 0.4296875, + "learning_rate": 0.0012239523042143568, + "loss": 0.0928, + "step": 44323 + }, + { + "epoch": 0.3847536045694048, + "grad_norm": 0.0849609375, + "learning_rate": 0.0012239228253101036, + "loss": 0.0723, + "step": 44324 + }, + { + "epoch": 0.38476228504960897, + "grad_norm": 0.5390625, + "learning_rate": 0.0012238933462703299, + "loss": 0.1045, + "step": 44325 + }, + { + "epoch": 0.3847709655298131, + "grad_norm": 0.291015625, + "learning_rate": 0.0012238638670950684, + "loss": 0.0854, + "step": 44326 + }, + { + "epoch": 0.3847796460100173, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012238343877843513, + "loss": 0.0986, + "step": 44327 + }, + { + "epoch": 0.38478832649022143, + "grad_norm": 0.1728515625, + "learning_rate": 0.001223804908338211, + "loss": 0.1064, + "step": 44328 + }, + { + "epoch": 0.3847970069704256, + "grad_norm": 0.189453125, + "learning_rate": 0.0012237754287566791, + "loss": 0.0752, + "step": 44329 + }, + { + "epoch": 0.38480568745062976, + "grad_norm": 0.318359375, + "learning_rate": 0.0012237459490397884, + "loss": 0.1289, + "step": 44330 + }, + { + "epoch": 0.38481436793083396, + "grad_norm": 0.24609375, + "learning_rate": 0.0012237164691875709, + "loss": 0.105, + "step": 44331 + }, + { + "epoch": 0.3848230484110381, + "grad_norm": 0.2421875, + "learning_rate": 0.0012236869892000592, + "loss": 0.1621, + "step": 44332 + }, + { + "epoch": 0.3848317288912423, + "grad_norm": 0.8671875, + "learning_rate": 0.001223657509077285, + "loss": 0.1182, + "step": 44333 + }, + { + "epoch": 0.3848404093714464, + "grad_norm": 0.1357421875, + "learning_rate": 0.001223628028819281, + "loss": 0.0986, + "step": 44334 + }, + { + "epoch": 0.3848490898516506, + "grad_norm": 0.111328125, + "learning_rate": 0.0012235985484260793, + "loss": 0.106, + "step": 44335 + }, + { + "epoch": 0.38485777033185475, + "grad_norm": 0.4296875, + "learning_rate": 0.0012235690678977118, + "loss": 0.1484, + "step": 44336 + }, + { + "epoch": 0.38486645081205895, + "grad_norm": 0.177734375, + "learning_rate": 0.0012235395872342112, + "loss": 0.0815, + "step": 44337 + }, + { + "epoch": 0.3848751312922631, + "grad_norm": 0.150390625, + "learning_rate": 0.0012235101064356096, + "loss": 0.0889, + "step": 44338 + }, + { + "epoch": 0.3848838117724673, + "grad_norm": 0.07958984375, + "learning_rate": 0.001223480625501939, + "loss": 0.0884, + "step": 44339 + }, + { + "epoch": 0.3848924922526714, + "grad_norm": 0.4375, + "learning_rate": 0.0012234511444332318, + "loss": 0.1025, + "step": 44340 + }, + { + "epoch": 0.3849011727328756, + "grad_norm": 0.361328125, + "learning_rate": 0.0012234216632295207, + "loss": 0.0864, + "step": 44341 + }, + { + "epoch": 0.38490985321307974, + "grad_norm": 0.0927734375, + "learning_rate": 0.001223392181890837, + "loss": 0.1143, + "step": 44342 + }, + { + "epoch": 0.38491853369328394, + "grad_norm": 0.20703125, + "learning_rate": 0.0012233627004172138, + "loss": 0.1475, + "step": 44343 + }, + { + "epoch": 0.3849272141734881, + "grad_norm": 0.251953125, + "learning_rate": 0.0012233332188086833, + "loss": 0.1045, + "step": 44344 + }, + { + "epoch": 0.38493589465369227, + "grad_norm": 0.2080078125, + "learning_rate": 0.001223303737065277, + "loss": 0.0859, + "step": 44345 + }, + { + "epoch": 0.3849445751338964, + "grad_norm": 0.0771484375, + "learning_rate": 0.0012232742551870276, + "loss": 0.0776, + "step": 44346 + }, + { + "epoch": 0.3849532556141006, + "grad_norm": 0.203125, + "learning_rate": 0.0012232447731739676, + "loss": 0.1035, + "step": 44347 + }, + { + "epoch": 0.38496193609430474, + "grad_norm": 0.36328125, + "learning_rate": 0.0012232152910261287, + "loss": 0.0859, + "step": 44348 + }, + { + "epoch": 0.38497061657450893, + "grad_norm": 0.1181640625, + "learning_rate": 0.0012231858087435434, + "loss": 0.0859, + "step": 44349 + }, + { + "epoch": 0.38497929705471307, + "grad_norm": 0.115234375, + "learning_rate": 0.0012231563263262442, + "loss": 0.082, + "step": 44350 + }, + { + "epoch": 0.38498797753491726, + "grad_norm": 0.8515625, + "learning_rate": 0.0012231268437742626, + "loss": 0.1157, + "step": 44351 + }, + { + "epoch": 0.3849966580151214, + "grad_norm": 0.1826171875, + "learning_rate": 0.0012230973610876316, + "loss": 0.1289, + "step": 44352 + }, + { + "epoch": 0.3850053384953256, + "grad_norm": 0.08935546875, + "learning_rate": 0.0012230678782663837, + "loss": 0.1113, + "step": 44353 + }, + { + "epoch": 0.3850140189755297, + "grad_norm": 0.515625, + "learning_rate": 0.00122303839531055, + "loss": 0.1006, + "step": 44354 + }, + { + "epoch": 0.3850226994557339, + "grad_norm": 0.1142578125, + "learning_rate": 0.0012230089122201634, + "loss": 0.1157, + "step": 44355 + }, + { + "epoch": 0.38503137993593806, + "grad_norm": 0.142578125, + "learning_rate": 0.0012229794289952563, + "loss": 0.0903, + "step": 44356 + }, + { + "epoch": 0.3850400604161422, + "grad_norm": 0.13671875, + "learning_rate": 0.0012229499456358607, + "loss": 0.1348, + "step": 44357 + }, + { + "epoch": 0.3850487408963464, + "grad_norm": 0.3046875, + "learning_rate": 0.0012229204621420089, + "loss": 0.1123, + "step": 44358 + }, + { + "epoch": 0.3850574213765505, + "grad_norm": 0.0771484375, + "learning_rate": 0.001222890978513733, + "loss": 0.0845, + "step": 44359 + }, + { + "epoch": 0.3850661018567547, + "grad_norm": 0.1982421875, + "learning_rate": 0.0012228614947510652, + "loss": 0.1074, + "step": 44360 + }, + { + "epoch": 0.38507478233695885, + "grad_norm": 0.1904296875, + "learning_rate": 0.0012228320108540382, + "loss": 0.1318, + "step": 44361 + }, + { + "epoch": 0.38508346281716305, + "grad_norm": 0.2451171875, + "learning_rate": 0.001222802526822684, + "loss": 0.1211, + "step": 44362 + }, + { + "epoch": 0.3850921432973672, + "grad_norm": 0.1630859375, + "learning_rate": 0.001222773042657035, + "loss": 0.0718, + "step": 44363 + }, + { + "epoch": 0.3851008237775714, + "grad_norm": 0.416015625, + "learning_rate": 0.0012227435583571228, + "loss": 0.0859, + "step": 44364 + }, + { + "epoch": 0.3851095042577755, + "grad_norm": 0.73828125, + "learning_rate": 0.0012227140739229803, + "loss": 0.0767, + "step": 44365 + }, + { + "epoch": 0.3851181847379797, + "grad_norm": 0.10205078125, + "learning_rate": 0.0012226845893546395, + "loss": 0.0786, + "step": 44366 + }, + { + "epoch": 0.38512686521818384, + "grad_norm": 0.330078125, + "learning_rate": 0.0012226551046521329, + "loss": 0.0781, + "step": 44367 + }, + { + "epoch": 0.38513554569838804, + "grad_norm": 0.2734375, + "learning_rate": 0.0012226256198154919, + "loss": 0.0767, + "step": 44368 + }, + { + "epoch": 0.3851442261785922, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012225961348447498, + "loss": 0.0928, + "step": 44369 + }, + { + "epoch": 0.38515290665879637, + "grad_norm": 0.74609375, + "learning_rate": 0.0012225666497399383, + "loss": 0.1113, + "step": 44370 + }, + { + "epoch": 0.3851615871390005, + "grad_norm": 0.171875, + "learning_rate": 0.00122253716450109, + "loss": 0.0981, + "step": 44371 + }, + { + "epoch": 0.3851702676192047, + "grad_norm": 0.5234375, + "learning_rate": 0.0012225076791282366, + "loss": 0.1221, + "step": 44372 + }, + { + "epoch": 0.38517894809940884, + "grad_norm": 0.0869140625, + "learning_rate": 0.0012224781936214107, + "loss": 0.0884, + "step": 44373 + }, + { + "epoch": 0.38518762857961303, + "grad_norm": 0.73046875, + "learning_rate": 0.0012224487079806445, + "loss": 0.0815, + "step": 44374 + }, + { + "epoch": 0.38519630905981717, + "grad_norm": 0.46484375, + "learning_rate": 0.0012224192222059704, + "loss": 0.0962, + "step": 44375 + }, + { + "epoch": 0.38520498954002136, + "grad_norm": 0.197265625, + "learning_rate": 0.0012223897362974205, + "loss": 0.084, + "step": 44376 + }, + { + "epoch": 0.3852136700202255, + "grad_norm": 0.1962890625, + "learning_rate": 0.0012223602502550268, + "loss": 0.0986, + "step": 44377 + }, + { + "epoch": 0.3852223505004297, + "grad_norm": 0.7421875, + "learning_rate": 0.001222330764078822, + "loss": 0.1055, + "step": 44378 + }, + { + "epoch": 0.3852310309806338, + "grad_norm": 0.1513671875, + "learning_rate": 0.0012223012777688377, + "loss": 0.1426, + "step": 44379 + }, + { + "epoch": 0.385239711460838, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012222717913251068, + "loss": 0.0747, + "step": 44380 + }, + { + "epoch": 0.38524839194104216, + "grad_norm": 0.515625, + "learning_rate": 0.0012222423047476615, + "loss": 0.1348, + "step": 44381 + }, + { + "epoch": 0.38525707242124635, + "grad_norm": 0.3671875, + "learning_rate": 0.0012222128180365336, + "loss": 0.0996, + "step": 44382 + }, + { + "epoch": 0.3852657529014505, + "grad_norm": 0.203125, + "learning_rate": 0.0012221833311917556, + "loss": 0.1377, + "step": 44383 + }, + { + "epoch": 0.3852744333816547, + "grad_norm": 0.28515625, + "learning_rate": 0.00122215384421336, + "loss": 0.1113, + "step": 44384 + }, + { + "epoch": 0.3852831138618588, + "grad_norm": 0.0859375, + "learning_rate": 0.0012221243571013784, + "loss": 0.1201, + "step": 44385 + }, + { + "epoch": 0.385291794342063, + "grad_norm": 0.357421875, + "learning_rate": 0.0012220948698558438, + "loss": 0.085, + "step": 44386 + }, + { + "epoch": 0.38530047482226715, + "grad_norm": 1.109375, + "learning_rate": 0.001222065382476788, + "loss": 0.1055, + "step": 44387 + }, + { + "epoch": 0.38530915530247134, + "grad_norm": 0.1064453125, + "learning_rate": 0.001222035894964243, + "loss": 0.1133, + "step": 44388 + }, + { + "epoch": 0.3853178357826755, + "grad_norm": 0.228515625, + "learning_rate": 0.001222006407318242, + "loss": 0.0771, + "step": 44389 + }, + { + "epoch": 0.38532651626287967, + "grad_norm": 0.058349609375, + "learning_rate": 0.001221976919538816, + "loss": 0.0522, + "step": 44390 + }, + { + "epoch": 0.3853351967430838, + "grad_norm": 0.30078125, + "learning_rate": 0.001221947431625998, + "loss": 0.0869, + "step": 44391 + }, + { + "epoch": 0.385343877223288, + "grad_norm": 0.2490234375, + "learning_rate": 0.0012219179435798205, + "loss": 0.0791, + "step": 44392 + }, + { + "epoch": 0.38535255770349214, + "grad_norm": 0.1884765625, + "learning_rate": 0.001221888455400315, + "loss": 0.0703, + "step": 44393 + }, + { + "epoch": 0.38536123818369633, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012218589670875142, + "loss": 0.1084, + "step": 44394 + }, + { + "epoch": 0.38536991866390047, + "grad_norm": 0.10986328125, + "learning_rate": 0.0012218294786414505, + "loss": 0.0874, + "step": 44395 + }, + { + "epoch": 0.38537859914410466, + "grad_norm": 0.73046875, + "learning_rate": 0.001221799990062156, + "loss": 0.1143, + "step": 44396 + }, + { + "epoch": 0.3853872796243088, + "grad_norm": 0.1630859375, + "learning_rate": 0.0012217705013496624, + "loss": 0.0859, + "step": 44397 + }, + { + "epoch": 0.385395960104513, + "grad_norm": 0.400390625, + "learning_rate": 0.0012217410125040028, + "loss": 0.0879, + "step": 44398 + }, + { + "epoch": 0.38540464058471713, + "grad_norm": 0.21875, + "learning_rate": 0.0012217115235252083, + "loss": 0.1006, + "step": 44399 + }, + { + "epoch": 0.3854133210649213, + "grad_norm": 0.26953125, + "learning_rate": 0.0012216820344133129, + "loss": 0.1328, + "step": 44400 + }, + { + "epoch": 0.38542200154512546, + "grad_norm": 0.0927734375, + "learning_rate": 0.0012216525451683473, + "loss": 0.0845, + "step": 44401 + }, + { + "epoch": 0.38543068202532965, + "grad_norm": 0.1611328125, + "learning_rate": 0.0012216230557903446, + "loss": 0.1172, + "step": 44402 + }, + { + "epoch": 0.3854393625055338, + "grad_norm": 0.318359375, + "learning_rate": 0.0012215935662793363, + "loss": 0.1006, + "step": 44403 + }, + { + "epoch": 0.385448042985738, + "grad_norm": 0.3515625, + "learning_rate": 0.0012215640766353555, + "loss": 0.0889, + "step": 44404 + }, + { + "epoch": 0.3854567234659421, + "grad_norm": 0.9375, + "learning_rate": 0.001221534586858434, + "loss": 0.1465, + "step": 44405 + }, + { + "epoch": 0.3854654039461463, + "grad_norm": 0.66796875, + "learning_rate": 0.001221505096948604, + "loss": 0.1123, + "step": 44406 + }, + { + "epoch": 0.38547408442635045, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012214756069058978, + "loss": 0.0767, + "step": 44407 + }, + { + "epoch": 0.38548276490655464, + "grad_norm": 0.2314453125, + "learning_rate": 0.0012214461167303478, + "loss": 0.1025, + "step": 44408 + }, + { + "epoch": 0.3854914453867588, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012214166264219862, + "loss": 0.1074, + "step": 44409 + }, + { + "epoch": 0.385500125866963, + "grad_norm": 0.150390625, + "learning_rate": 0.001221387135980845, + "loss": 0.0781, + "step": 44410 + }, + { + "epoch": 0.3855088063471671, + "grad_norm": 0.6484375, + "learning_rate": 0.0012213576454069568, + "loss": 0.1025, + "step": 44411 + }, + { + "epoch": 0.3855174868273713, + "grad_norm": 0.1640625, + "learning_rate": 0.001221328154700354, + "loss": 0.0869, + "step": 44412 + }, + { + "epoch": 0.38552616730757544, + "grad_norm": 0.43359375, + "learning_rate": 0.001221298663861068, + "loss": 0.1016, + "step": 44413 + }, + { + "epoch": 0.38553484778777963, + "grad_norm": 1.0703125, + "learning_rate": 0.001221269172889132, + "loss": 0.0869, + "step": 44414 + }, + { + "epoch": 0.38554352826798377, + "grad_norm": 0.107421875, + "learning_rate": 0.0012212396817845776, + "loss": 0.0825, + "step": 44415 + }, + { + "epoch": 0.38555220874818796, + "grad_norm": 0.3046875, + "learning_rate": 0.0012212101905474373, + "loss": 0.0869, + "step": 44416 + }, + { + "epoch": 0.3855608892283921, + "grad_norm": 0.1328125, + "learning_rate": 0.0012211806991777434, + "loss": 0.0859, + "step": 44417 + }, + { + "epoch": 0.3855695697085963, + "grad_norm": 0.484375, + "learning_rate": 0.0012211512076755282, + "loss": 0.0791, + "step": 44418 + }, + { + "epoch": 0.38557825018880043, + "grad_norm": 0.09814453125, + "learning_rate": 0.001221121716040824, + "loss": 0.1006, + "step": 44419 + }, + { + "epoch": 0.3855869306690046, + "grad_norm": 0.4140625, + "learning_rate": 0.0012210922242736626, + "loss": 0.0962, + "step": 44420 + }, + { + "epoch": 0.38559561114920876, + "grad_norm": 0.3046875, + "learning_rate": 0.001221062732374077, + "loss": 0.1074, + "step": 44421 + }, + { + "epoch": 0.38560429162941295, + "grad_norm": 0.08740234375, + "learning_rate": 0.0012210332403420985, + "loss": 0.0908, + "step": 44422 + }, + { + "epoch": 0.3856129721096171, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012210037481777602, + "loss": 0.1631, + "step": 44423 + }, + { + "epoch": 0.3856216525898213, + "grad_norm": 0.2158203125, + "learning_rate": 0.001220974255881094, + "loss": 0.1182, + "step": 44424 + }, + { + "epoch": 0.3856303330700254, + "grad_norm": 0.3046875, + "learning_rate": 0.001220944763452132, + "loss": 0.082, + "step": 44425 + }, + { + "epoch": 0.3856390135502296, + "grad_norm": 0.40234375, + "learning_rate": 0.0012209152708909066, + "loss": 0.126, + "step": 44426 + }, + { + "epoch": 0.38564769403043375, + "grad_norm": 0.15625, + "learning_rate": 0.0012208857781974504, + "loss": 0.0723, + "step": 44427 + }, + { + "epoch": 0.38565637451063794, + "grad_norm": 0.71484375, + "learning_rate": 0.001220856285371795, + "loss": 0.1074, + "step": 44428 + }, + { + "epoch": 0.3856650549908421, + "grad_norm": 0.10595703125, + "learning_rate": 0.0012208267924139731, + "loss": 0.0679, + "step": 44429 + }, + { + "epoch": 0.3856737354710463, + "grad_norm": 0.1298828125, + "learning_rate": 0.0012207972993240168, + "loss": 0.0928, + "step": 44430 + }, + { + "epoch": 0.3856824159512504, + "grad_norm": 0.3203125, + "learning_rate": 0.0012207678061019584, + "loss": 0.1328, + "step": 44431 + }, + { + "epoch": 0.3856910964314546, + "grad_norm": 0.2138671875, + "learning_rate": 0.0012207383127478303, + "loss": 0.1001, + "step": 44432 + }, + { + "epoch": 0.38569977691165874, + "grad_norm": 0.52734375, + "learning_rate": 0.0012207088192616644, + "loss": 0.1104, + "step": 44433 + }, + { + "epoch": 0.38570845739186294, + "grad_norm": 0.267578125, + "learning_rate": 0.0012206793256434932, + "loss": 0.0859, + "step": 44434 + }, + { + "epoch": 0.3857171378720671, + "grad_norm": 0.1455078125, + "learning_rate": 0.0012206498318933488, + "loss": 0.104, + "step": 44435 + }, + { + "epoch": 0.38572581835227127, + "grad_norm": 0.56640625, + "learning_rate": 0.0012206203380112637, + "loss": 0.1133, + "step": 44436 + }, + { + "epoch": 0.3857344988324754, + "grad_norm": 0.1484375, + "learning_rate": 0.00122059084399727, + "loss": 0.1572, + "step": 44437 + }, + { + "epoch": 0.3857431793126796, + "grad_norm": 0.6953125, + "learning_rate": 0.0012205613498513998, + "loss": 0.125, + "step": 44438 + }, + { + "epoch": 0.38575185979288373, + "grad_norm": 0.14453125, + "learning_rate": 0.0012205318555736856, + "loss": 0.0908, + "step": 44439 + }, + { + "epoch": 0.3857605402730879, + "grad_norm": 0.59765625, + "learning_rate": 0.0012205023611641598, + "loss": 0.1318, + "step": 44440 + }, + { + "epoch": 0.38576922075329206, + "grad_norm": 0.5859375, + "learning_rate": 0.0012204728666228544, + "loss": 0.1235, + "step": 44441 + }, + { + "epoch": 0.38577790123349626, + "grad_norm": 0.138671875, + "learning_rate": 0.0012204433719498013, + "loss": 0.0977, + "step": 44442 + }, + { + "epoch": 0.3857865817137004, + "grad_norm": 0.11669921875, + "learning_rate": 0.0012204138771450335, + "loss": 0.0864, + "step": 44443 + }, + { + "epoch": 0.3857952621939046, + "grad_norm": 0.158203125, + "learning_rate": 0.0012203843822085828, + "loss": 0.1201, + "step": 44444 + }, + { + "epoch": 0.3858039426741087, + "grad_norm": 0.78125, + "learning_rate": 0.0012203548871404814, + "loss": 0.0977, + "step": 44445 + }, + { + "epoch": 0.3858126231543129, + "grad_norm": 0.19140625, + "learning_rate": 0.0012203253919407617, + "loss": 0.0942, + "step": 44446 + }, + { + "epoch": 0.38582130363451705, + "grad_norm": 0.232421875, + "learning_rate": 0.001220295896609456, + "loss": 0.1045, + "step": 44447 + }, + { + "epoch": 0.38582998411472125, + "grad_norm": 0.08251953125, + "learning_rate": 0.0012202664011465969, + "loss": 0.1069, + "step": 44448 + }, + { + "epoch": 0.3858386645949254, + "grad_norm": 0.80078125, + "learning_rate": 0.001220236905552216, + "loss": 0.0938, + "step": 44449 + }, + { + "epoch": 0.3858473450751296, + "grad_norm": 0.91015625, + "learning_rate": 0.0012202074098263459, + "loss": 0.124, + "step": 44450 + }, + { + "epoch": 0.3858560255553337, + "grad_norm": 0.34765625, + "learning_rate": 0.0012201779139690185, + "loss": 0.1172, + "step": 44451 + }, + { + "epoch": 0.3858647060355379, + "grad_norm": 0.5, + "learning_rate": 0.0012201484179802666, + "loss": 0.0747, + "step": 44452 + }, + { + "epoch": 0.38587338651574205, + "grad_norm": 0.361328125, + "learning_rate": 0.0012201189218601224, + "loss": 0.1582, + "step": 44453 + }, + { + "epoch": 0.38588206699594624, + "grad_norm": 0.1904296875, + "learning_rate": 0.0012200894256086173, + "loss": 0.0869, + "step": 44454 + }, + { + "epoch": 0.3858907474761504, + "grad_norm": 0.6953125, + "learning_rate": 0.0012200599292257848, + "loss": 0.0918, + "step": 44455 + }, + { + "epoch": 0.38589942795635457, + "grad_norm": 0.330078125, + "learning_rate": 0.001220030432711656, + "loss": 0.1006, + "step": 44456 + }, + { + "epoch": 0.3859081084365587, + "grad_norm": 0.10107421875, + "learning_rate": 0.0012200009360662644, + "loss": 0.1006, + "step": 44457 + }, + { + "epoch": 0.3859167889167629, + "grad_norm": 0.1328125, + "learning_rate": 0.0012199714392896414, + "loss": 0.1108, + "step": 44458 + }, + { + "epoch": 0.38592546939696704, + "grad_norm": 0.6953125, + "learning_rate": 0.001219941942381819, + "loss": 0.1953, + "step": 44459 + }, + { + "epoch": 0.38593414987717123, + "grad_norm": 0.2333984375, + "learning_rate": 0.0012199124453428304, + "loss": 0.1045, + "step": 44460 + }, + { + "epoch": 0.38594283035737537, + "grad_norm": 0.43359375, + "learning_rate": 0.001219882948172707, + "loss": 0.0957, + "step": 44461 + }, + { + "epoch": 0.38595151083757956, + "grad_norm": 0.11572265625, + "learning_rate": 0.0012198534508714816, + "loss": 0.1182, + "step": 44462 + }, + { + "epoch": 0.3859601913177837, + "grad_norm": 0.1357421875, + "learning_rate": 0.001219823953439186, + "loss": 0.1279, + "step": 44463 + }, + { + "epoch": 0.3859688717979879, + "grad_norm": 0.2080078125, + "learning_rate": 0.0012197944558758529, + "loss": 0.1235, + "step": 44464 + }, + { + "epoch": 0.385977552278192, + "grad_norm": 0.6484375, + "learning_rate": 0.0012197649581815145, + "loss": 0.0767, + "step": 44465 + }, + { + "epoch": 0.3859862327583962, + "grad_norm": 0.267578125, + "learning_rate": 0.0012197354603562028, + "loss": 0.1055, + "step": 44466 + }, + { + "epoch": 0.38599491323860036, + "grad_norm": 0.51171875, + "learning_rate": 0.0012197059623999505, + "loss": 0.0947, + "step": 44467 + }, + { + "epoch": 0.38600359371880455, + "grad_norm": 0.8515625, + "learning_rate": 0.0012196764643127892, + "loss": 0.1445, + "step": 44468 + }, + { + "epoch": 0.3860122741990087, + "grad_norm": 0.498046875, + "learning_rate": 0.0012196469660947515, + "loss": 0.0879, + "step": 44469 + }, + { + "epoch": 0.3860209546792129, + "grad_norm": 0.419921875, + "learning_rate": 0.00121961746774587, + "loss": 0.0977, + "step": 44470 + }, + { + "epoch": 0.386029635159417, + "grad_norm": 0.306640625, + "learning_rate": 0.0012195879692661765, + "loss": 0.1221, + "step": 44471 + }, + { + "epoch": 0.3860383156396212, + "grad_norm": 0.10986328125, + "learning_rate": 0.001219558470655703, + "loss": 0.104, + "step": 44472 + }, + { + "epoch": 0.38604699611982535, + "grad_norm": 4.875, + "learning_rate": 0.0012195289719144826, + "loss": 0.7227, + "step": 44473 + }, + { + "epoch": 0.38605567660002954, + "grad_norm": 0.158203125, + "learning_rate": 0.0012194994730425467, + "loss": 0.1016, + "step": 44474 + }, + { + "epoch": 0.3860643570802337, + "grad_norm": 0.2294921875, + "learning_rate": 0.0012194699740399285, + "loss": 0.0859, + "step": 44475 + }, + { + "epoch": 0.38607303756043787, + "grad_norm": 0.43359375, + "learning_rate": 0.0012194404749066592, + "loss": 0.0986, + "step": 44476 + }, + { + "epoch": 0.386081718040642, + "grad_norm": 0.154296875, + "learning_rate": 0.001219410975642772, + "loss": 0.1099, + "step": 44477 + }, + { + "epoch": 0.3860903985208462, + "grad_norm": 0.421875, + "learning_rate": 0.0012193814762482987, + "loss": 0.1152, + "step": 44478 + }, + { + "epoch": 0.38609907900105034, + "grad_norm": 0.310546875, + "learning_rate": 0.0012193519767232717, + "loss": 0.1152, + "step": 44479 + }, + { + "epoch": 0.3861077594812545, + "grad_norm": 0.0888671875, + "learning_rate": 0.0012193224770677231, + "loss": 0.0996, + "step": 44480 + }, + { + "epoch": 0.38611643996145867, + "grad_norm": 0.267578125, + "learning_rate": 0.0012192929772816851, + "loss": 0.0898, + "step": 44481 + }, + { + "epoch": 0.3861251204416628, + "grad_norm": 0.29296875, + "learning_rate": 0.00121926347736519, + "loss": 0.0752, + "step": 44482 + }, + { + "epoch": 0.386133800921867, + "grad_norm": 0.7734375, + "learning_rate": 0.0012192339773182705, + "loss": 0.166, + "step": 44483 + }, + { + "epoch": 0.38614248140207114, + "grad_norm": 0.447265625, + "learning_rate": 0.0012192044771409582, + "loss": 0.1201, + "step": 44484 + }, + { + "epoch": 0.38615116188227533, + "grad_norm": 0.322265625, + "learning_rate": 0.0012191749768332857, + "loss": 0.0776, + "step": 44485 + }, + { + "epoch": 0.38615984236247947, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012191454763952855, + "loss": 0.1025, + "step": 44486 + }, + { + "epoch": 0.38616852284268366, + "grad_norm": 0.53515625, + "learning_rate": 0.0012191159758269893, + "loss": 0.0918, + "step": 44487 + }, + { + "epoch": 0.3861772033228878, + "grad_norm": 0.236328125, + "learning_rate": 0.00121908647512843, + "loss": 0.084, + "step": 44488 + }, + { + "epoch": 0.386185883803092, + "grad_norm": 0.1015625, + "learning_rate": 0.001219056974299639, + "loss": 0.127, + "step": 44489 + }, + { + "epoch": 0.3861945642832961, + "grad_norm": 0.46484375, + "learning_rate": 0.0012190274733406494, + "loss": 0.0938, + "step": 44490 + }, + { + "epoch": 0.3862032447635003, + "grad_norm": 0.076171875, + "learning_rate": 0.0012189979722514932, + "loss": 0.0879, + "step": 44491 + }, + { + "epoch": 0.38621192524370446, + "grad_norm": 0.07958984375, + "learning_rate": 0.0012189684710322026, + "loss": 0.083, + "step": 44492 + }, + { + "epoch": 0.38622060572390865, + "grad_norm": 0.373046875, + "learning_rate": 0.0012189389696828099, + "loss": 0.0879, + "step": 44493 + }, + { + "epoch": 0.3862292862041128, + "grad_norm": 0.193359375, + "learning_rate": 0.001218909468203347, + "loss": 0.1191, + "step": 44494 + }, + { + "epoch": 0.386237966684317, + "grad_norm": 0.1474609375, + "learning_rate": 0.0012188799665938468, + "loss": 0.1201, + "step": 44495 + }, + { + "epoch": 0.3862466471645211, + "grad_norm": 0.46875, + "learning_rate": 0.0012188504648543413, + "loss": 0.0947, + "step": 44496 + }, + { + "epoch": 0.3862553276447253, + "grad_norm": 0.30859375, + "learning_rate": 0.0012188209629848627, + "loss": 0.0757, + "step": 44497 + }, + { + "epoch": 0.38626400812492945, + "grad_norm": 0.1787109375, + "learning_rate": 0.0012187914609854428, + "loss": 0.0737, + "step": 44498 + }, + { + "epoch": 0.38627268860513364, + "grad_norm": 0.1650390625, + "learning_rate": 0.0012187619588561148, + "loss": 0.0986, + "step": 44499 + }, + { + "epoch": 0.3862813690853378, + "grad_norm": 0.08935546875, + "learning_rate": 0.0012187324565969104, + "loss": 0.1084, + "step": 44500 + }, + { + "epoch": 0.38629004956554197, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012187029542078624, + "loss": 0.1211, + "step": 44501 + }, + { + "epoch": 0.3862987300457461, + "grad_norm": 0.466796875, + "learning_rate": 0.001218673451689002, + "loss": 0.0952, + "step": 44502 + }, + { + "epoch": 0.3863074105259503, + "grad_norm": 0.11865234375, + "learning_rate": 0.0012186439490403623, + "loss": 0.1289, + "step": 44503 + }, + { + "epoch": 0.38631609100615444, + "grad_norm": 0.2119140625, + "learning_rate": 0.0012186144462619753, + "loss": 0.104, + "step": 44504 + }, + { + "epoch": 0.38632477148635863, + "grad_norm": 0.494140625, + "learning_rate": 0.0012185849433538733, + "loss": 0.0796, + "step": 44505 + }, + { + "epoch": 0.38633345196656277, + "grad_norm": 0.10546875, + "learning_rate": 0.0012185554403160888, + "loss": 0.0762, + "step": 44506 + }, + { + "epoch": 0.38634213244676696, + "grad_norm": 0.10693359375, + "learning_rate": 0.0012185259371486536, + "loss": 0.0708, + "step": 44507 + }, + { + "epoch": 0.3863508129269711, + "grad_norm": 0.44140625, + "learning_rate": 0.0012184964338516004, + "loss": 0.0918, + "step": 44508 + }, + { + "epoch": 0.3863594934071753, + "grad_norm": 0.076171875, + "learning_rate": 0.0012184669304249613, + "loss": 0.1025, + "step": 44509 + }, + { + "epoch": 0.38636817388737943, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012184374268687686, + "loss": 0.0732, + "step": 44510 + }, + { + "epoch": 0.3863768543675836, + "grad_norm": 0.11328125, + "learning_rate": 0.0012184079231830542, + "loss": 0.0977, + "step": 44511 + }, + { + "epoch": 0.38638553484778776, + "grad_norm": 0.349609375, + "learning_rate": 0.0012183784193678512, + "loss": 0.127, + "step": 44512 + }, + { + "epoch": 0.38639421532799195, + "grad_norm": 0.96875, + "learning_rate": 0.0012183489154231908, + "loss": 0.1074, + "step": 44513 + }, + { + "epoch": 0.3864028958081961, + "grad_norm": 0.162109375, + "learning_rate": 0.0012183194113491059, + "loss": 0.0713, + "step": 44514 + }, + { + "epoch": 0.3864115762884003, + "grad_norm": 0.46875, + "learning_rate": 0.0012182899071456288, + "loss": 0.0879, + "step": 44515 + }, + { + "epoch": 0.3864202567686044, + "grad_norm": 0.2197265625, + "learning_rate": 0.0012182604028127917, + "loss": 0.1128, + "step": 44516 + }, + { + "epoch": 0.3864289372488086, + "grad_norm": 0.146484375, + "learning_rate": 0.0012182308983506268, + "loss": 0.1069, + "step": 44517 + }, + { + "epoch": 0.38643761772901275, + "grad_norm": 0.2890625, + "learning_rate": 0.0012182013937591663, + "loss": 0.1035, + "step": 44518 + }, + { + "epoch": 0.38644629820921694, + "grad_norm": 0.4375, + "learning_rate": 0.0012181718890384426, + "loss": 0.0913, + "step": 44519 + }, + { + "epoch": 0.3864549786894211, + "grad_norm": 0.94921875, + "learning_rate": 0.0012181423841884876, + "loss": 0.0898, + "step": 44520 + }, + { + "epoch": 0.3864636591696253, + "grad_norm": 0.30078125, + "learning_rate": 0.0012181128792093342, + "loss": 0.0967, + "step": 44521 + }, + { + "epoch": 0.3864723396498294, + "grad_norm": 0.404296875, + "learning_rate": 0.001218083374101014, + "loss": 0.1055, + "step": 44522 + }, + { + "epoch": 0.3864810201300336, + "grad_norm": 0.1953125, + "learning_rate": 0.0012180538688635599, + "loss": 0.0952, + "step": 44523 + }, + { + "epoch": 0.38648970061023774, + "grad_norm": 0.16796875, + "learning_rate": 0.0012180243634970036, + "loss": 0.1914, + "step": 44524 + }, + { + "epoch": 0.38649838109044193, + "grad_norm": 0.0869140625, + "learning_rate": 0.001217994858001378, + "loss": 0.1099, + "step": 44525 + }, + { + "epoch": 0.38650706157064607, + "grad_norm": 0.1982421875, + "learning_rate": 0.001217965352376715, + "loss": 0.0688, + "step": 44526 + }, + { + "epoch": 0.38651574205085026, + "grad_norm": 0.408203125, + "learning_rate": 0.0012179358466230465, + "loss": 0.1689, + "step": 44527 + }, + { + "epoch": 0.3865244225310544, + "grad_norm": 0.298828125, + "learning_rate": 0.0012179063407404052, + "loss": 0.1279, + "step": 44528 + }, + { + "epoch": 0.3865331030112586, + "grad_norm": 0.2353515625, + "learning_rate": 0.0012178768347288233, + "loss": 0.0732, + "step": 44529 + }, + { + "epoch": 0.38654178349146273, + "grad_norm": 0.166015625, + "learning_rate": 0.0012178473285883332, + "loss": 0.1133, + "step": 44530 + }, + { + "epoch": 0.3865504639716669, + "grad_norm": 0.26171875, + "learning_rate": 0.001217817822318967, + "loss": 0.1309, + "step": 44531 + }, + { + "epoch": 0.38655914445187106, + "grad_norm": 0.34375, + "learning_rate": 0.001217788315920757, + "loss": 0.1035, + "step": 44532 + }, + { + "epoch": 0.38656782493207525, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012177588093937352, + "loss": 0.1152, + "step": 44533 + }, + { + "epoch": 0.3865765054122794, + "grad_norm": 0.1455078125, + "learning_rate": 0.0012177293027379346, + "loss": 0.083, + "step": 44534 + }, + { + "epoch": 0.3865851858924836, + "grad_norm": 0.609375, + "learning_rate": 0.0012176997959533868, + "loss": 0.1416, + "step": 44535 + }, + { + "epoch": 0.3865938663726877, + "grad_norm": 0.193359375, + "learning_rate": 0.0012176702890401244, + "loss": 0.1064, + "step": 44536 + }, + { + "epoch": 0.3866025468528919, + "grad_norm": 0.41015625, + "learning_rate": 0.0012176407819981792, + "loss": 0.0728, + "step": 44537 + }, + { + "epoch": 0.38661122733309605, + "grad_norm": 0.625, + "learning_rate": 0.001217611274827584, + "loss": 0.1245, + "step": 44538 + }, + { + "epoch": 0.38661990781330025, + "grad_norm": 0.392578125, + "learning_rate": 0.0012175817675283709, + "loss": 0.0928, + "step": 44539 + }, + { + "epoch": 0.3866285882935044, + "grad_norm": 0.2734375, + "learning_rate": 0.0012175522601005722, + "loss": 0.1631, + "step": 44540 + }, + { + "epoch": 0.3866372687737086, + "grad_norm": 0.10498046875, + "learning_rate": 0.00121752275254422, + "loss": 0.0967, + "step": 44541 + }, + { + "epoch": 0.3866459492539127, + "grad_norm": 0.40234375, + "learning_rate": 0.0012174932448593466, + "loss": 0.0874, + "step": 44542 + }, + { + "epoch": 0.3866546297341169, + "grad_norm": 0.427734375, + "learning_rate": 0.0012174637370459846, + "loss": 0.0923, + "step": 44543 + }, + { + "epoch": 0.38666331021432104, + "grad_norm": 1.109375, + "learning_rate": 0.001217434229104166, + "loss": 0.0947, + "step": 44544 + }, + { + "epoch": 0.38667199069452524, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012174047210339227, + "loss": 0.1582, + "step": 44545 + }, + { + "epoch": 0.3866806711747294, + "grad_norm": 0.1240234375, + "learning_rate": 0.0012173752128352877, + "loss": 0.1045, + "step": 44546 + }, + { + "epoch": 0.38668935165493357, + "grad_norm": 0.421875, + "learning_rate": 0.0012173457045082932, + "loss": 0.0957, + "step": 44547 + }, + { + "epoch": 0.3866980321351377, + "grad_norm": 0.08154296875, + "learning_rate": 0.001217316196052971, + "loss": 0.0996, + "step": 44548 + }, + { + "epoch": 0.3867067126153419, + "grad_norm": 0.6796875, + "learning_rate": 0.0012172866874693533, + "loss": 0.0771, + "step": 44549 + }, + { + "epoch": 0.38671539309554603, + "grad_norm": 0.96875, + "learning_rate": 0.0012172571787574726, + "loss": 0.1055, + "step": 44550 + }, + { + "epoch": 0.3867240735757502, + "grad_norm": 0.06298828125, + "learning_rate": 0.0012172276699173614, + "loss": 0.0845, + "step": 44551 + }, + { + "epoch": 0.38673275405595436, + "grad_norm": 0.1484375, + "learning_rate": 0.0012171981609490517, + "loss": 0.0874, + "step": 44552 + }, + { + "epoch": 0.38674143453615856, + "grad_norm": 0.08984375, + "learning_rate": 0.0012171686518525763, + "loss": 0.1133, + "step": 44553 + }, + { + "epoch": 0.3867501150163627, + "grad_norm": 0.3203125, + "learning_rate": 0.0012171391426279664, + "loss": 0.1289, + "step": 44554 + }, + { + "epoch": 0.3867587954965669, + "grad_norm": 0.41015625, + "learning_rate": 0.0012171096332752552, + "loss": 0.1387, + "step": 44555 + }, + { + "epoch": 0.386767475976771, + "grad_norm": 0.10986328125, + "learning_rate": 0.0012170801237944746, + "loss": 0.0869, + "step": 44556 + }, + { + "epoch": 0.3867761564569752, + "grad_norm": 0.392578125, + "learning_rate": 0.001217050614185657, + "loss": 0.123, + "step": 44557 + }, + { + "epoch": 0.38678483693717935, + "grad_norm": 0.2109375, + "learning_rate": 0.0012170211044488343, + "loss": 0.0767, + "step": 44558 + }, + { + "epoch": 0.38679351741738355, + "grad_norm": 0.330078125, + "learning_rate": 0.0012169915945840397, + "loss": 0.1074, + "step": 44559 + }, + { + "epoch": 0.3868021978975877, + "grad_norm": 0.267578125, + "learning_rate": 0.0012169620845913045, + "loss": 0.1279, + "step": 44560 + }, + { + "epoch": 0.3868108783777919, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012169325744706611, + "loss": 0.1084, + "step": 44561 + }, + { + "epoch": 0.386819558857996, + "grad_norm": 0.27734375, + "learning_rate": 0.0012169030642221424, + "loss": 0.1289, + "step": 44562 + }, + { + "epoch": 0.3868282393382002, + "grad_norm": 0.21484375, + "learning_rate": 0.00121687355384578, + "loss": 0.0957, + "step": 44563 + }, + { + "epoch": 0.38683691981840435, + "grad_norm": 0.337890625, + "learning_rate": 0.0012168440433416065, + "loss": 0.1123, + "step": 44564 + }, + { + "epoch": 0.38684560029860854, + "grad_norm": 0.130859375, + "learning_rate": 0.001216814532709654, + "loss": 0.0879, + "step": 44565 + }, + { + "epoch": 0.3868542807788127, + "grad_norm": 0.0869140625, + "learning_rate": 0.0012167850219499553, + "loss": 0.1064, + "step": 44566 + }, + { + "epoch": 0.38686296125901687, + "grad_norm": 0.12109375, + "learning_rate": 0.0012167555110625416, + "loss": 0.1182, + "step": 44567 + }, + { + "epoch": 0.386871641739221, + "grad_norm": 0.427734375, + "learning_rate": 0.0012167260000474462, + "loss": 0.0938, + "step": 44568 + }, + { + "epoch": 0.3868803222194252, + "grad_norm": 0.1298828125, + "learning_rate": 0.0012166964889047012, + "loss": 0.0757, + "step": 44569 + }, + { + "epoch": 0.38688900269962934, + "grad_norm": 0.232421875, + "learning_rate": 0.0012166669776343384, + "loss": 0.1182, + "step": 44570 + }, + { + "epoch": 0.38689768317983353, + "grad_norm": 0.427734375, + "learning_rate": 0.0012166374662363902, + "loss": 0.0996, + "step": 44571 + }, + { + "epoch": 0.38690636366003767, + "grad_norm": 0.32421875, + "learning_rate": 0.0012166079547108893, + "loss": 0.124, + "step": 44572 + }, + { + "epoch": 0.38691504414024186, + "grad_norm": 0.2421875, + "learning_rate": 0.0012165784430578674, + "loss": 0.0913, + "step": 44573 + }, + { + "epoch": 0.386923724620446, + "grad_norm": 0.384765625, + "learning_rate": 0.0012165489312773578, + "loss": 0.1396, + "step": 44574 + }, + { + "epoch": 0.3869324051006502, + "grad_norm": 0.53125, + "learning_rate": 0.0012165194193693913, + "loss": 0.1201, + "step": 44575 + }, + { + "epoch": 0.3869410855808543, + "grad_norm": 0.435546875, + "learning_rate": 0.001216489907334001, + "loss": 0.2461, + "step": 44576 + }, + { + "epoch": 0.3869497660610585, + "grad_norm": 0.51953125, + "learning_rate": 0.0012164603951712195, + "loss": 0.1465, + "step": 44577 + }, + { + "epoch": 0.38695844654126266, + "grad_norm": 0.28125, + "learning_rate": 0.0012164308828810785, + "loss": 0.1216, + "step": 44578 + }, + { + "epoch": 0.38696712702146685, + "grad_norm": 0.1767578125, + "learning_rate": 0.0012164013704636099, + "loss": 0.0718, + "step": 44579 + }, + { + "epoch": 0.386975807501671, + "grad_norm": 0.55859375, + "learning_rate": 0.0012163718579188472, + "loss": 0.0811, + "step": 44580 + }, + { + "epoch": 0.3869844879818752, + "grad_norm": 0.2490234375, + "learning_rate": 0.0012163423452468216, + "loss": 0.1089, + "step": 44581 + }, + { + "epoch": 0.3869931684620793, + "grad_norm": 0.10791015625, + "learning_rate": 0.001216312832447566, + "loss": 0.0737, + "step": 44582 + }, + { + "epoch": 0.3870018489422835, + "grad_norm": 0.337890625, + "learning_rate": 0.0012162833195211123, + "loss": 0.0742, + "step": 44583 + }, + { + "epoch": 0.38701052942248765, + "grad_norm": 0.06787109375, + "learning_rate": 0.0012162538064674927, + "loss": 0.062, + "step": 44584 + }, + { + "epoch": 0.38701920990269184, + "grad_norm": 0.67578125, + "learning_rate": 0.00121622429328674, + "loss": 0.1582, + "step": 44585 + }, + { + "epoch": 0.387027890382896, + "grad_norm": 0.349609375, + "learning_rate": 0.0012161947799788863, + "loss": 0.1348, + "step": 44586 + }, + { + "epoch": 0.38703657086310017, + "grad_norm": 0.09423828125, + "learning_rate": 0.0012161652665439633, + "loss": 0.084, + "step": 44587 + }, + { + "epoch": 0.3870452513433043, + "grad_norm": 0.208984375, + "learning_rate": 0.0012161357529820041, + "loss": 0.1133, + "step": 44588 + }, + { + "epoch": 0.3870539318235085, + "grad_norm": 0.6953125, + "learning_rate": 0.0012161062392930402, + "loss": 0.1426, + "step": 44589 + }, + { + "epoch": 0.38706261230371264, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012160767254771044, + "loss": 0.0991, + "step": 44590 + }, + { + "epoch": 0.38707129278391683, + "grad_norm": 0.2451171875, + "learning_rate": 0.0012160472115342288, + "loss": 0.1084, + "step": 44591 + }, + { + "epoch": 0.38707997326412097, + "grad_norm": 0.19921875, + "learning_rate": 0.0012160176974644458, + "loss": 0.123, + "step": 44592 + }, + { + "epoch": 0.38708865374432516, + "grad_norm": 0.1083984375, + "learning_rate": 0.0012159881832677878, + "loss": 0.0957, + "step": 44593 + }, + { + "epoch": 0.3870973342245293, + "grad_norm": 0.10302734375, + "learning_rate": 0.0012159586689442865, + "loss": 0.0654, + "step": 44594 + }, + { + "epoch": 0.3871060147047335, + "grad_norm": 0.86328125, + "learning_rate": 0.0012159291544939748, + "loss": 0.1133, + "step": 44595 + }, + { + "epoch": 0.38711469518493763, + "grad_norm": 0.5078125, + "learning_rate": 0.0012158996399168848, + "loss": 0.1572, + "step": 44596 + }, + { + "epoch": 0.3871233756651418, + "grad_norm": 0.2890625, + "learning_rate": 0.0012158701252130482, + "loss": 0.1123, + "step": 44597 + }, + { + "epoch": 0.38713205614534596, + "grad_norm": 0.482421875, + "learning_rate": 0.0012158406103824982, + "loss": 0.1289, + "step": 44598 + }, + { + "epoch": 0.38714073662555015, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012158110954252664, + "loss": 0.1104, + "step": 44599 + }, + { + "epoch": 0.3871494171057543, + "grad_norm": 0.1884765625, + "learning_rate": 0.0012157815803413854, + "loss": 0.1113, + "step": 44600 + }, + { + "epoch": 0.3871580975859585, + "grad_norm": 0.3671875, + "learning_rate": 0.0012157520651308875, + "loss": 0.1338, + "step": 44601 + }, + { + "epoch": 0.3871667780661626, + "grad_norm": 0.1533203125, + "learning_rate": 0.001215722549793805, + "loss": 0.1045, + "step": 44602 + }, + { + "epoch": 0.38717545854636676, + "grad_norm": 0.45703125, + "learning_rate": 0.0012156930343301697, + "loss": 0.084, + "step": 44603 + }, + { + "epoch": 0.38718413902657095, + "grad_norm": 0.330078125, + "learning_rate": 0.0012156635187400146, + "loss": 0.1035, + "step": 44604 + }, + { + "epoch": 0.3871928195067751, + "grad_norm": 0.41015625, + "learning_rate": 0.0012156340030233714, + "loss": 0.0903, + "step": 44605 + }, + { + "epoch": 0.3872014999869793, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012156044871802728, + "loss": 0.0981, + "step": 44606 + }, + { + "epoch": 0.3872101804671834, + "grad_norm": 0.3125, + "learning_rate": 0.0012155749712107506, + "loss": 0.0771, + "step": 44607 + }, + { + "epoch": 0.3872188609473876, + "grad_norm": 0.44921875, + "learning_rate": 0.0012155454551148375, + "loss": 0.0996, + "step": 44608 + }, + { + "epoch": 0.38722754142759175, + "grad_norm": 0.29296875, + "learning_rate": 0.0012155159388925654, + "loss": 0.0811, + "step": 44609 + }, + { + "epoch": 0.38723622190779594, + "grad_norm": 0.439453125, + "learning_rate": 0.001215486422543967, + "loss": 0.1172, + "step": 44610 + }, + { + "epoch": 0.3872449023880001, + "grad_norm": 1.03125, + "learning_rate": 0.0012154569060690743, + "loss": 0.082, + "step": 44611 + }, + { + "epoch": 0.38725358286820427, + "grad_norm": 0.7265625, + "learning_rate": 0.00121542738946792, + "loss": 0.126, + "step": 44612 + }, + { + "epoch": 0.3872622633484084, + "grad_norm": 0.2099609375, + "learning_rate": 0.0012153978727405357, + "loss": 0.085, + "step": 44613 + }, + { + "epoch": 0.3872709438286126, + "grad_norm": 0.2001953125, + "learning_rate": 0.001215368355886954, + "loss": 0.0947, + "step": 44614 + }, + { + "epoch": 0.38727962430881674, + "grad_norm": 0.11767578125, + "learning_rate": 0.0012153388389072072, + "loss": 0.1162, + "step": 44615 + }, + { + "epoch": 0.38728830478902093, + "grad_norm": 0.150390625, + "learning_rate": 0.001215309321801328, + "loss": 0.1157, + "step": 44616 + }, + { + "epoch": 0.38729698526922507, + "grad_norm": 0.45703125, + "learning_rate": 0.0012152798045693479, + "loss": 0.1201, + "step": 44617 + }, + { + "epoch": 0.38730566574942926, + "grad_norm": 0.4921875, + "learning_rate": 0.0012152502872112994, + "loss": 0.0859, + "step": 44618 + }, + { + "epoch": 0.3873143462296334, + "grad_norm": 0.1484375, + "learning_rate": 0.001215220769727215, + "loss": 0.1016, + "step": 44619 + }, + { + "epoch": 0.3873230267098376, + "grad_norm": 0.228515625, + "learning_rate": 0.001215191252117127, + "loss": 0.0732, + "step": 44620 + }, + { + "epoch": 0.38733170719004173, + "grad_norm": 0.09716796875, + "learning_rate": 0.0012151617343810677, + "loss": 0.1475, + "step": 44621 + }, + { + "epoch": 0.3873403876702459, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012151322165190692, + "loss": 0.082, + "step": 44622 + }, + { + "epoch": 0.38734906815045006, + "grad_norm": 0.35546875, + "learning_rate": 0.0012151026985311635, + "loss": 0.1455, + "step": 44623 + }, + { + "epoch": 0.38735774863065425, + "grad_norm": 0.376953125, + "learning_rate": 0.0012150731804173836, + "loss": 0.1592, + "step": 44624 + }, + { + "epoch": 0.3873664291108584, + "grad_norm": 0.15625, + "learning_rate": 0.0012150436621777615, + "loss": 0.123, + "step": 44625 + }, + { + "epoch": 0.3873751095910626, + "grad_norm": 0.134765625, + "learning_rate": 0.0012150141438123292, + "loss": 0.0771, + "step": 44626 + }, + { + "epoch": 0.3873837900712667, + "grad_norm": 0.3984375, + "learning_rate": 0.0012149846253211188, + "loss": 0.1201, + "step": 44627 + }, + { + "epoch": 0.3873924705514709, + "grad_norm": 0.55078125, + "learning_rate": 0.0012149551067041633, + "loss": 0.0967, + "step": 44628 + }, + { + "epoch": 0.38740115103167505, + "grad_norm": 0.1748046875, + "learning_rate": 0.0012149255879614948, + "loss": 0.0889, + "step": 44629 + }, + { + "epoch": 0.38740983151187924, + "grad_norm": 0.091796875, + "learning_rate": 0.001214896069093145, + "loss": 0.1318, + "step": 44630 + }, + { + "epoch": 0.3874185119920834, + "grad_norm": 0.455078125, + "learning_rate": 0.0012148665500991468, + "loss": 0.1064, + "step": 44631 + }, + { + "epoch": 0.3874271924722876, + "grad_norm": 0.4140625, + "learning_rate": 0.001214837030979532, + "loss": 0.103, + "step": 44632 + }, + { + "epoch": 0.3874358729524917, + "grad_norm": 0.1865234375, + "learning_rate": 0.0012148075117343336, + "loss": 0.0718, + "step": 44633 + }, + { + "epoch": 0.3874445534326959, + "grad_norm": 0.1396484375, + "learning_rate": 0.0012147779923635832, + "loss": 0.0898, + "step": 44634 + }, + { + "epoch": 0.38745323391290004, + "grad_norm": 0.1357421875, + "learning_rate": 0.0012147484728673135, + "loss": 0.1074, + "step": 44635 + }, + { + "epoch": 0.38746191439310423, + "grad_norm": 0.19921875, + "learning_rate": 0.0012147189532455563, + "loss": 0.0977, + "step": 44636 + }, + { + "epoch": 0.38747059487330837, + "grad_norm": 0.67578125, + "learning_rate": 0.0012146894334983446, + "loss": 0.124, + "step": 44637 + }, + { + "epoch": 0.38747927535351256, + "grad_norm": 0.271484375, + "learning_rate": 0.0012146599136257095, + "loss": 0.1045, + "step": 44638 + }, + { + "epoch": 0.3874879558337167, + "grad_norm": 0.248046875, + "learning_rate": 0.0012146303936276847, + "loss": 0.0679, + "step": 44639 + }, + { + "epoch": 0.3874966363139209, + "grad_norm": 1.9296875, + "learning_rate": 0.0012146008735043016, + "loss": 0.4277, + "step": 44640 + }, + { + "epoch": 0.38750531679412503, + "grad_norm": 0.07958984375, + "learning_rate": 0.0012145713532555928, + "loss": 0.085, + "step": 44641 + }, + { + "epoch": 0.3875139972743292, + "grad_norm": 0.359375, + "learning_rate": 0.0012145418328815903, + "loss": 0.1084, + "step": 44642 + }, + { + "epoch": 0.38752267775453336, + "grad_norm": 0.283203125, + "learning_rate": 0.0012145123123823267, + "loss": 0.1729, + "step": 44643 + }, + { + "epoch": 0.38753135823473756, + "grad_norm": 0.126953125, + "learning_rate": 0.0012144827917578342, + "loss": 0.1211, + "step": 44644 + }, + { + "epoch": 0.3875400387149417, + "grad_norm": 0.19140625, + "learning_rate": 0.0012144532710081448, + "loss": 0.1006, + "step": 44645 + }, + { + "epoch": 0.3875487191951459, + "grad_norm": 0.2021484375, + "learning_rate": 0.0012144237501332913, + "loss": 0.0918, + "step": 44646 + }, + { + "epoch": 0.38755739967535, + "grad_norm": 0.8828125, + "learning_rate": 0.0012143942291333056, + "loss": 0.0879, + "step": 44647 + }, + { + "epoch": 0.3875660801555542, + "grad_norm": 0.296875, + "learning_rate": 0.00121436470800822, + "loss": 0.0781, + "step": 44648 + }, + { + "epoch": 0.38757476063575835, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012143351867580669, + "loss": 0.0752, + "step": 44649 + }, + { + "epoch": 0.38758344111596255, + "grad_norm": 0.38671875, + "learning_rate": 0.0012143056653828787, + "loss": 0.0933, + "step": 44650 + }, + { + "epoch": 0.3875921215961667, + "grad_norm": 0.64453125, + "learning_rate": 0.0012142761438826877, + "loss": 0.0879, + "step": 44651 + }, + { + "epoch": 0.3876008020763709, + "grad_norm": 0.2275390625, + "learning_rate": 0.0012142466222575257, + "loss": 0.0967, + "step": 44652 + }, + { + "epoch": 0.387609482556575, + "grad_norm": 0.2080078125, + "learning_rate": 0.0012142171005074252, + "loss": 0.1484, + "step": 44653 + }, + { + "epoch": 0.3876181630367792, + "grad_norm": 0.0888671875, + "learning_rate": 0.001214187578632419, + "loss": 0.0923, + "step": 44654 + }, + { + "epoch": 0.38762684351698334, + "grad_norm": 0.11669921875, + "learning_rate": 0.001214158056632539, + "loss": 0.1084, + "step": 44655 + }, + { + "epoch": 0.38763552399718754, + "grad_norm": 0.55859375, + "learning_rate": 0.001214128534507817, + "loss": 0.1055, + "step": 44656 + }, + { + "epoch": 0.3876442044773917, + "grad_norm": 0.1220703125, + "learning_rate": 0.001214099012258286, + "loss": 0.0869, + "step": 44657 + }, + { + "epoch": 0.38765288495759587, + "grad_norm": 0.380859375, + "learning_rate": 0.001214069489883978, + "loss": 0.0806, + "step": 44658 + }, + { + "epoch": 0.3876615654378, + "grad_norm": 0.29296875, + "learning_rate": 0.0012140399673849253, + "loss": 0.0752, + "step": 44659 + }, + { + "epoch": 0.3876702459180042, + "grad_norm": 0.3046875, + "learning_rate": 0.0012140104447611604, + "loss": 0.0947, + "step": 44660 + }, + { + "epoch": 0.38767892639820833, + "grad_norm": 0.1767578125, + "learning_rate": 0.0012139809220127154, + "loss": 0.0669, + "step": 44661 + }, + { + "epoch": 0.3876876068784125, + "grad_norm": 0.2353515625, + "learning_rate": 0.0012139513991396222, + "loss": 0.0732, + "step": 44662 + }, + { + "epoch": 0.38769628735861666, + "grad_norm": 0.279296875, + "learning_rate": 0.0012139218761419138, + "loss": 0.1357, + "step": 44663 + }, + { + "epoch": 0.38770496783882086, + "grad_norm": 0.24609375, + "learning_rate": 0.001213892353019622, + "loss": 0.1113, + "step": 44664 + }, + { + "epoch": 0.387713648319025, + "grad_norm": 0.138671875, + "learning_rate": 0.0012138628297727794, + "loss": 0.0884, + "step": 44665 + }, + { + "epoch": 0.3877223287992292, + "grad_norm": 1.3046875, + "learning_rate": 0.0012138333064014178, + "loss": 0.1465, + "step": 44666 + }, + { + "epoch": 0.3877310092794333, + "grad_norm": 0.52734375, + "learning_rate": 0.0012138037829055699, + "loss": 0.1582, + "step": 44667 + }, + { + "epoch": 0.3877396897596375, + "grad_norm": 0.1259765625, + "learning_rate": 0.001213774259285268, + "loss": 0.1279, + "step": 44668 + }, + { + "epoch": 0.38774837023984166, + "grad_norm": 0.091796875, + "learning_rate": 0.0012137447355405444, + "loss": 0.1138, + "step": 44669 + }, + { + "epoch": 0.38775705072004585, + "grad_norm": 0.46875, + "learning_rate": 0.0012137152116714312, + "loss": 0.1089, + "step": 44670 + }, + { + "epoch": 0.38776573120025, + "grad_norm": 0.3671875, + "learning_rate": 0.0012136856876779605, + "loss": 0.0991, + "step": 44671 + }, + { + "epoch": 0.3877744116804542, + "grad_norm": 0.09521484375, + "learning_rate": 0.0012136561635601652, + "loss": 0.1279, + "step": 44672 + }, + { + "epoch": 0.3877830921606583, + "grad_norm": 0.220703125, + "learning_rate": 0.001213626639318077, + "loss": 0.127, + "step": 44673 + }, + { + "epoch": 0.3877917726408625, + "grad_norm": 0.201171875, + "learning_rate": 0.0012135971149517285, + "loss": 0.1191, + "step": 44674 + }, + { + "epoch": 0.38780045312106665, + "grad_norm": 0.1123046875, + "learning_rate": 0.001213567590461152, + "loss": 0.1318, + "step": 44675 + }, + { + "epoch": 0.38780913360127084, + "grad_norm": 0.150390625, + "learning_rate": 0.0012135380658463792, + "loss": 0.0854, + "step": 44676 + }, + { + "epoch": 0.387817814081475, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012135085411074437, + "loss": 0.1025, + "step": 44677 + }, + { + "epoch": 0.38782649456167917, + "grad_norm": 0.51953125, + "learning_rate": 0.0012134790162443762, + "loss": 0.1006, + "step": 44678 + }, + { + "epoch": 0.3878351750418833, + "grad_norm": 0.349609375, + "learning_rate": 0.00121344949125721, + "loss": 0.1055, + "step": 44679 + }, + { + "epoch": 0.3878438555220875, + "grad_norm": 0.6484375, + "learning_rate": 0.0012134199661459773, + "loss": 0.1348, + "step": 44680 + }, + { + "epoch": 0.38785253600229164, + "grad_norm": 0.498046875, + "learning_rate": 0.0012133904409107102, + "loss": 0.1201, + "step": 44681 + }, + { + "epoch": 0.38786121648249583, + "grad_norm": 0.28125, + "learning_rate": 0.001213360915551441, + "loss": 0.1289, + "step": 44682 + }, + { + "epoch": 0.38786989696269997, + "grad_norm": 0.197265625, + "learning_rate": 0.001213331390068202, + "loss": 0.0889, + "step": 44683 + }, + { + "epoch": 0.38787857744290416, + "grad_norm": 0.17578125, + "learning_rate": 0.0012133018644610253, + "loss": 0.0933, + "step": 44684 + }, + { + "epoch": 0.3878872579231083, + "grad_norm": 0.1181640625, + "learning_rate": 0.0012132723387299437, + "loss": 0.0952, + "step": 44685 + }, + { + "epoch": 0.3878959384033125, + "grad_norm": 0.314453125, + "learning_rate": 0.001213242812874989, + "loss": 0.1172, + "step": 44686 + }, + { + "epoch": 0.3879046188835166, + "grad_norm": 0.12109375, + "learning_rate": 0.0012132132868961935, + "loss": 0.0713, + "step": 44687 + }, + { + "epoch": 0.3879132993637208, + "grad_norm": 0.27734375, + "learning_rate": 0.0012131837607935896, + "loss": 0.1245, + "step": 44688 + }, + { + "epoch": 0.38792197984392496, + "grad_norm": 1.2734375, + "learning_rate": 0.00121315423456721, + "loss": 0.1064, + "step": 44689 + }, + { + "epoch": 0.38793066032412915, + "grad_norm": 0.498046875, + "learning_rate": 0.0012131247082170864, + "loss": 0.0996, + "step": 44690 + }, + { + "epoch": 0.3879393408043333, + "grad_norm": 0.44140625, + "learning_rate": 0.0012130951817432517, + "loss": 0.106, + "step": 44691 + }, + { + "epoch": 0.3879480212845375, + "grad_norm": 0.55078125, + "learning_rate": 0.0012130656551457373, + "loss": 0.1562, + "step": 44692 + }, + { + "epoch": 0.3879567017647416, + "grad_norm": 0.287109375, + "learning_rate": 0.0012130361284245763, + "loss": 0.1211, + "step": 44693 + }, + { + "epoch": 0.3879653822449458, + "grad_norm": 0.2275390625, + "learning_rate": 0.0012130066015798007, + "loss": 0.1016, + "step": 44694 + }, + { + "epoch": 0.38797406272514995, + "grad_norm": 0.251953125, + "learning_rate": 0.0012129770746114425, + "loss": 0.084, + "step": 44695 + }, + { + "epoch": 0.38798274320535414, + "grad_norm": 0.50390625, + "learning_rate": 0.0012129475475195343, + "loss": 0.1123, + "step": 44696 + }, + { + "epoch": 0.3879914236855583, + "grad_norm": 0.08251953125, + "learning_rate": 0.0012129180203041086, + "loss": 0.0625, + "step": 44697 + }, + { + "epoch": 0.38800010416576247, + "grad_norm": 1.4375, + "learning_rate": 0.0012128884929651974, + "loss": 0.0957, + "step": 44698 + }, + { + "epoch": 0.3880087846459666, + "grad_norm": 0.328125, + "learning_rate": 0.001212858965502833, + "loss": 0.1152, + "step": 44699 + }, + { + "epoch": 0.3880174651261708, + "grad_norm": 0.625, + "learning_rate": 0.0012128294379170478, + "loss": 0.1162, + "step": 44700 + }, + { + "epoch": 0.38802614560637494, + "grad_norm": 0.1015625, + "learning_rate": 0.001212799910207874, + "loss": 0.1011, + "step": 44701 + }, + { + "epoch": 0.38803482608657913, + "grad_norm": 0.26953125, + "learning_rate": 0.0012127703823753438, + "loss": 0.1069, + "step": 44702 + }, + { + "epoch": 0.38804350656678327, + "grad_norm": 0.130859375, + "learning_rate": 0.0012127408544194897, + "loss": 0.1094, + "step": 44703 + }, + { + "epoch": 0.38805218704698746, + "grad_norm": 0.25, + "learning_rate": 0.0012127113263403439, + "loss": 0.0879, + "step": 44704 + }, + { + "epoch": 0.3880608675271916, + "grad_norm": 0.62890625, + "learning_rate": 0.0012126817981379385, + "loss": 0.0991, + "step": 44705 + }, + { + "epoch": 0.3880695480073958, + "grad_norm": 0.169921875, + "learning_rate": 0.0012126522698123062, + "loss": 0.1113, + "step": 44706 + }, + { + "epoch": 0.38807822848759993, + "grad_norm": 0.1298828125, + "learning_rate": 0.0012126227413634786, + "loss": 0.1055, + "step": 44707 + }, + { + "epoch": 0.3880869089678041, + "grad_norm": 0.3203125, + "learning_rate": 0.0012125932127914892, + "loss": 0.0796, + "step": 44708 + }, + { + "epoch": 0.38809558944800826, + "grad_norm": 0.09814453125, + "learning_rate": 0.0012125636840963692, + "loss": 0.0869, + "step": 44709 + }, + { + "epoch": 0.38810426992821245, + "grad_norm": 0.09228515625, + "learning_rate": 0.0012125341552781512, + "loss": 0.0728, + "step": 44710 + }, + { + "epoch": 0.3881129504084166, + "grad_norm": 0.259765625, + "learning_rate": 0.0012125046263368675, + "loss": 0.1387, + "step": 44711 + }, + { + "epoch": 0.3881216308886208, + "grad_norm": 0.546875, + "learning_rate": 0.0012124750972725507, + "loss": 0.0967, + "step": 44712 + }, + { + "epoch": 0.3881303113688249, + "grad_norm": 0.1875, + "learning_rate": 0.0012124455680852326, + "loss": 0.1357, + "step": 44713 + }, + { + "epoch": 0.3881389918490291, + "grad_norm": 0.24609375, + "learning_rate": 0.0012124160387749457, + "loss": 0.0825, + "step": 44714 + }, + { + "epoch": 0.38814767232923325, + "grad_norm": 0.373046875, + "learning_rate": 0.0012123865093417224, + "loss": 0.0708, + "step": 44715 + }, + { + "epoch": 0.38815635280943744, + "grad_norm": 0.1669921875, + "learning_rate": 0.0012123569797855948, + "loss": 0.0728, + "step": 44716 + }, + { + "epoch": 0.3881650332896416, + "grad_norm": 0.1171875, + "learning_rate": 0.0012123274501065955, + "loss": 0.1006, + "step": 44717 + }, + { + "epoch": 0.3881737137698458, + "grad_norm": 0.15234375, + "learning_rate": 0.0012122979203047564, + "loss": 0.1021, + "step": 44718 + }, + { + "epoch": 0.3881823942500499, + "grad_norm": 0.35546875, + "learning_rate": 0.0012122683903801103, + "loss": 0.1152, + "step": 44719 + }, + { + "epoch": 0.3881910747302541, + "grad_norm": 0.419921875, + "learning_rate": 0.0012122388603326888, + "loss": 0.1328, + "step": 44720 + }, + { + "epoch": 0.38819975521045824, + "grad_norm": 0.3046875, + "learning_rate": 0.0012122093301625246, + "loss": 0.1582, + "step": 44721 + }, + { + "epoch": 0.38820843569066243, + "grad_norm": 0.5234375, + "learning_rate": 0.0012121797998696502, + "loss": 0.1235, + "step": 44722 + }, + { + "epoch": 0.38821711617086657, + "grad_norm": 0.08251953125, + "learning_rate": 0.0012121502694540974, + "loss": 0.0728, + "step": 44723 + }, + { + "epoch": 0.38822579665107076, + "grad_norm": 0.1513671875, + "learning_rate": 0.0012121207389158989, + "loss": 0.0884, + "step": 44724 + }, + { + "epoch": 0.3882344771312749, + "grad_norm": 0.482421875, + "learning_rate": 0.0012120912082550868, + "loss": 0.2598, + "step": 44725 + }, + { + "epoch": 0.38824315761147904, + "grad_norm": 0.375, + "learning_rate": 0.0012120616774716935, + "loss": 0.0981, + "step": 44726 + }, + { + "epoch": 0.38825183809168323, + "grad_norm": 0.234375, + "learning_rate": 0.001212032146565751, + "loss": 0.0908, + "step": 44727 + }, + { + "epoch": 0.38826051857188737, + "grad_norm": 0.265625, + "learning_rate": 0.001212002615537292, + "loss": 0.1113, + "step": 44728 + }, + { + "epoch": 0.38826919905209156, + "grad_norm": 0.07177734375, + "learning_rate": 0.0012119730843863485, + "loss": 0.0732, + "step": 44729 + }, + { + "epoch": 0.3882778795322957, + "grad_norm": 0.5, + "learning_rate": 0.0012119435531129533, + "loss": 0.083, + "step": 44730 + }, + { + "epoch": 0.3882865600124999, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012119140217171378, + "loss": 0.0757, + "step": 44731 + }, + { + "epoch": 0.38829524049270403, + "grad_norm": 0.27734375, + "learning_rate": 0.0012118844901989351, + "loss": 0.0977, + "step": 44732 + }, + { + "epoch": 0.3883039209729082, + "grad_norm": 0.9296875, + "learning_rate": 0.001211854958558377, + "loss": 0.1089, + "step": 44733 + }, + { + "epoch": 0.38831260145311236, + "grad_norm": 0.19921875, + "learning_rate": 0.0012118254267954962, + "loss": 0.1001, + "step": 44734 + }, + { + "epoch": 0.38832128193331655, + "grad_norm": 0.123046875, + "learning_rate": 0.001211795894910325, + "loss": 0.1006, + "step": 44735 + }, + { + "epoch": 0.3883299624135207, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012117663629028952, + "loss": 0.0991, + "step": 44736 + }, + { + "epoch": 0.3883386428937249, + "grad_norm": 0.10791015625, + "learning_rate": 0.0012117368307732394, + "loss": 0.125, + "step": 44737 + }, + { + "epoch": 0.388347323373929, + "grad_norm": 0.1796875, + "learning_rate": 0.00121170729852139, + "loss": 0.0664, + "step": 44738 + }, + { + "epoch": 0.3883560038541332, + "grad_norm": 0.318359375, + "learning_rate": 0.0012116777661473792, + "loss": 0.1113, + "step": 44739 + }, + { + "epoch": 0.38836468433433735, + "grad_norm": 0.435546875, + "learning_rate": 0.0012116482336512388, + "loss": 0.0728, + "step": 44740 + }, + { + "epoch": 0.38837336481454154, + "grad_norm": 0.2275390625, + "learning_rate": 0.001211618701033002, + "loss": 0.126, + "step": 44741 + }, + { + "epoch": 0.3883820452947457, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012115891682927006, + "loss": 0.1221, + "step": 44742 + }, + { + "epoch": 0.3883907257749499, + "grad_norm": 0.06884765625, + "learning_rate": 0.001211559635430367, + "loss": 0.082, + "step": 44743 + }, + { + "epoch": 0.388399406255154, + "grad_norm": 0.369140625, + "learning_rate": 0.0012115301024460332, + "loss": 0.1348, + "step": 44744 + }, + { + "epoch": 0.3884080867353582, + "grad_norm": 0.486328125, + "learning_rate": 0.0012115005693397322, + "loss": 0.0752, + "step": 44745 + }, + { + "epoch": 0.38841676721556234, + "grad_norm": 0.150390625, + "learning_rate": 0.0012114710361114955, + "loss": 0.0884, + "step": 44746 + }, + { + "epoch": 0.38842544769576653, + "grad_norm": 0.24609375, + "learning_rate": 0.0012114415027613562, + "loss": 0.124, + "step": 44747 + }, + { + "epoch": 0.38843412817597067, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012114119692893455, + "loss": 0.064, + "step": 44748 + }, + { + "epoch": 0.38844280865617486, + "grad_norm": 0.189453125, + "learning_rate": 0.0012113824356954968, + "loss": 0.0771, + "step": 44749 + }, + { + "epoch": 0.388451489136379, + "grad_norm": 0.21875, + "learning_rate": 0.001211352901979842, + "loss": 0.0879, + "step": 44750 + }, + { + "epoch": 0.3884601696165832, + "grad_norm": 0.259765625, + "learning_rate": 0.0012113233681424129, + "loss": 0.0801, + "step": 44751 + }, + { + "epoch": 0.38846885009678733, + "grad_norm": 0.07861328125, + "learning_rate": 0.0012112938341832427, + "loss": 0.0742, + "step": 44752 + }, + { + "epoch": 0.3884775305769915, + "grad_norm": 0.08154296875, + "learning_rate": 0.0012112643001023628, + "loss": 0.1123, + "step": 44753 + }, + { + "epoch": 0.38848621105719566, + "grad_norm": 0.1328125, + "learning_rate": 0.001211234765899806, + "loss": 0.0762, + "step": 44754 + }, + { + "epoch": 0.38849489153739986, + "grad_norm": 0.29296875, + "learning_rate": 0.0012112052315756047, + "loss": 0.0825, + "step": 44755 + }, + { + "epoch": 0.388503572017604, + "grad_norm": 0.345703125, + "learning_rate": 0.001211175697129791, + "loss": 0.0903, + "step": 44756 + }, + { + "epoch": 0.3885122524978082, + "grad_norm": 0.10498046875, + "learning_rate": 0.001211146162562397, + "loss": 0.1055, + "step": 44757 + }, + { + "epoch": 0.3885209329780123, + "grad_norm": 0.431640625, + "learning_rate": 0.0012111166278734558, + "loss": 0.1104, + "step": 44758 + }, + { + "epoch": 0.3885296134582165, + "grad_norm": 0.10498046875, + "learning_rate": 0.001211087093062999, + "loss": 0.0957, + "step": 44759 + }, + { + "epoch": 0.38853829393842065, + "grad_norm": 0.203125, + "learning_rate": 0.0012110575581310587, + "loss": 0.0913, + "step": 44760 + }, + { + "epoch": 0.38854697441862485, + "grad_norm": 0.51953125, + "learning_rate": 0.0012110280230776677, + "loss": 0.1191, + "step": 44761 + }, + { + "epoch": 0.388555654898829, + "grad_norm": 0.1201171875, + "learning_rate": 0.0012109984879028581, + "loss": 0.0991, + "step": 44762 + }, + { + "epoch": 0.3885643353790332, + "grad_norm": 0.49609375, + "learning_rate": 0.001210968952606662, + "loss": 0.1245, + "step": 44763 + }, + { + "epoch": 0.3885730158592373, + "grad_norm": 0.265625, + "learning_rate": 0.0012109394171891119, + "loss": 0.1338, + "step": 44764 + }, + { + "epoch": 0.3885816963394415, + "grad_norm": 0.396484375, + "learning_rate": 0.0012109098816502407, + "loss": 0.0879, + "step": 44765 + }, + { + "epoch": 0.38859037681964564, + "grad_norm": 0.2890625, + "learning_rate": 0.0012108803459900796, + "loss": 0.0996, + "step": 44766 + }, + { + "epoch": 0.38859905729984984, + "grad_norm": 0.150390625, + "learning_rate": 0.0012108508102086615, + "loss": 0.1211, + "step": 44767 + }, + { + "epoch": 0.388607737780054, + "grad_norm": 0.07958984375, + "learning_rate": 0.0012108212743060188, + "loss": 0.0723, + "step": 44768 + }, + { + "epoch": 0.38861641826025817, + "grad_norm": 0.57421875, + "learning_rate": 0.0012107917382821836, + "loss": 0.1133, + "step": 44769 + }, + { + "epoch": 0.3886250987404623, + "grad_norm": 0.2890625, + "learning_rate": 0.001210762202137188, + "loss": 0.1475, + "step": 44770 + }, + { + "epoch": 0.3886337792206665, + "grad_norm": 0.6875, + "learning_rate": 0.0012107326658710646, + "loss": 0.0991, + "step": 44771 + }, + { + "epoch": 0.38864245970087063, + "grad_norm": 0.287109375, + "learning_rate": 0.0012107031294838456, + "loss": 0.127, + "step": 44772 + }, + { + "epoch": 0.38865114018107483, + "grad_norm": 0.072265625, + "learning_rate": 0.0012106735929755634, + "loss": 0.082, + "step": 44773 + }, + { + "epoch": 0.38865982066127897, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012106440563462502, + "loss": 0.167, + "step": 44774 + }, + { + "epoch": 0.38866850114148316, + "grad_norm": 0.3671875, + "learning_rate": 0.0012106145195959384, + "loss": 0.0942, + "step": 44775 + }, + { + "epoch": 0.3886771816216873, + "grad_norm": 0.416015625, + "learning_rate": 0.00121058498272466, + "loss": 0.1299, + "step": 44776 + }, + { + "epoch": 0.3886858621018915, + "grad_norm": 0.8046875, + "learning_rate": 0.001210555445732448, + "loss": 0.1387, + "step": 44777 + }, + { + "epoch": 0.3886945425820956, + "grad_norm": 0.640625, + "learning_rate": 0.001210525908619334, + "loss": 0.1445, + "step": 44778 + }, + { + "epoch": 0.3887032230622998, + "grad_norm": 0.2392578125, + "learning_rate": 0.0012104963713853501, + "loss": 0.0835, + "step": 44779 + }, + { + "epoch": 0.38871190354250396, + "grad_norm": 0.130859375, + "learning_rate": 0.0012104668340305294, + "loss": 0.0806, + "step": 44780 + }, + { + "epoch": 0.38872058402270815, + "grad_norm": 0.486328125, + "learning_rate": 0.0012104372965549037, + "loss": 0.0796, + "step": 44781 + }, + { + "epoch": 0.3887292645029123, + "grad_norm": 0.380859375, + "learning_rate": 0.0012104077589585056, + "loss": 0.0776, + "step": 44782 + }, + { + "epoch": 0.3887379449831165, + "grad_norm": 0.326171875, + "learning_rate": 0.001210378221241367, + "loss": 0.1406, + "step": 44783 + }, + { + "epoch": 0.3887466254633206, + "grad_norm": 0.69921875, + "learning_rate": 0.0012103486834035206, + "loss": 0.1455, + "step": 44784 + }, + { + "epoch": 0.3887553059435248, + "grad_norm": 0.13671875, + "learning_rate": 0.0012103191454449986, + "loss": 0.0869, + "step": 44785 + }, + { + "epoch": 0.38876398642372895, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012102896073658332, + "loss": 0.104, + "step": 44786 + }, + { + "epoch": 0.38877266690393314, + "grad_norm": 0.380859375, + "learning_rate": 0.0012102600691660567, + "loss": 0.1172, + "step": 44787 + }, + { + "epoch": 0.3887813473841373, + "grad_norm": 0.10009765625, + "learning_rate": 0.0012102305308457012, + "loss": 0.0967, + "step": 44788 + }, + { + "epoch": 0.38879002786434147, + "grad_norm": 0.59375, + "learning_rate": 0.0012102009924047996, + "loss": 0.0898, + "step": 44789 + }, + { + "epoch": 0.3887987083445456, + "grad_norm": 0.11474609375, + "learning_rate": 0.0012101714538433842, + "loss": 0.0938, + "step": 44790 + }, + { + "epoch": 0.3888073888247498, + "grad_norm": 0.06494140625, + "learning_rate": 0.0012101419151614862, + "loss": 0.0806, + "step": 44791 + }, + { + "epoch": 0.38881606930495394, + "grad_norm": 0.16796875, + "learning_rate": 0.001210112376359139, + "loss": 0.1138, + "step": 44792 + }, + { + "epoch": 0.38882474978515813, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012100828374363743, + "loss": 0.0752, + "step": 44793 + }, + { + "epoch": 0.38883343026536227, + "grad_norm": 0.2890625, + "learning_rate": 0.0012100532983932248, + "loss": 0.1064, + "step": 44794 + }, + { + "epoch": 0.38884211074556646, + "grad_norm": 0.07470703125, + "learning_rate": 0.0012100237592297229, + "loss": 0.0688, + "step": 44795 + }, + { + "epoch": 0.3888507912257706, + "grad_norm": 0.1201171875, + "learning_rate": 0.0012099942199459005, + "loss": 0.0928, + "step": 44796 + }, + { + "epoch": 0.3888594717059748, + "grad_norm": 0.400390625, + "learning_rate": 0.00120996468054179, + "loss": 0.0918, + "step": 44797 + }, + { + "epoch": 0.38886815218617893, + "grad_norm": 0.2431640625, + "learning_rate": 0.0012099351410174242, + "loss": 0.0879, + "step": 44798 + }, + { + "epoch": 0.3888768326663831, + "grad_norm": 0.267578125, + "learning_rate": 0.0012099056013728344, + "loss": 0.0625, + "step": 44799 + }, + { + "epoch": 0.38888551314658726, + "grad_norm": 0.2001953125, + "learning_rate": 0.0012098760616080539, + "loss": 0.0757, + "step": 44800 + }, + { + "epoch": 0.38889419362679145, + "grad_norm": 0.419921875, + "learning_rate": 0.001209846521723114, + "loss": 0.0845, + "step": 44801 + }, + { + "epoch": 0.3889028741069956, + "grad_norm": 0.78125, + "learning_rate": 0.0012098169817180483, + "loss": 0.0938, + "step": 44802 + }, + { + "epoch": 0.3889115545871998, + "grad_norm": 0.69921875, + "learning_rate": 0.001209787441592888, + "loss": 0.1143, + "step": 44803 + }, + { + "epoch": 0.3889202350674039, + "grad_norm": 0.458984375, + "learning_rate": 0.001209757901347666, + "loss": 0.0933, + "step": 44804 + }, + { + "epoch": 0.3889289155476081, + "grad_norm": 0.2353515625, + "learning_rate": 0.0012097283609824143, + "loss": 0.1133, + "step": 44805 + }, + { + "epoch": 0.38893759602781225, + "grad_norm": 0.455078125, + "learning_rate": 0.0012096988204971656, + "loss": 0.0952, + "step": 44806 + }, + { + "epoch": 0.38894627650801644, + "grad_norm": 0.34375, + "learning_rate": 0.0012096692798919517, + "loss": 0.0947, + "step": 44807 + }, + { + "epoch": 0.3889549569882206, + "grad_norm": 0.3359375, + "learning_rate": 0.001209639739166805, + "loss": 0.0728, + "step": 44808 + }, + { + "epoch": 0.38896363746842477, + "grad_norm": 0.5625, + "learning_rate": 0.0012096101983217577, + "loss": 0.1001, + "step": 44809 + }, + { + "epoch": 0.3889723179486289, + "grad_norm": 0.384765625, + "learning_rate": 0.001209580657356843, + "loss": 0.1094, + "step": 44810 + }, + { + "epoch": 0.3889809984288331, + "grad_norm": 0.1123046875, + "learning_rate": 0.0012095511162720921, + "loss": 0.0908, + "step": 44811 + }, + { + "epoch": 0.38898967890903724, + "grad_norm": 0.54296875, + "learning_rate": 0.001209521575067538, + "loss": 0.0977, + "step": 44812 + }, + { + "epoch": 0.38899835938924143, + "grad_norm": 0.259765625, + "learning_rate": 0.0012094920337432122, + "loss": 0.0957, + "step": 44813 + }, + { + "epoch": 0.38900703986944557, + "grad_norm": 0.1064453125, + "learning_rate": 0.001209462492299148, + "loss": 0.0884, + "step": 44814 + }, + { + "epoch": 0.38901572034964976, + "grad_norm": 0.291015625, + "learning_rate": 0.0012094329507353774, + "loss": 0.1045, + "step": 44815 + }, + { + "epoch": 0.3890244008298539, + "grad_norm": 0.2412109375, + "learning_rate": 0.0012094034090519323, + "loss": 0.1123, + "step": 44816 + }, + { + "epoch": 0.3890330813100581, + "grad_norm": 0.44140625, + "learning_rate": 0.0012093738672488456, + "loss": 0.0688, + "step": 44817 + }, + { + "epoch": 0.38904176179026223, + "grad_norm": 0.3046875, + "learning_rate": 0.0012093443253261486, + "loss": 0.0786, + "step": 44818 + }, + { + "epoch": 0.3890504422704664, + "grad_norm": 0.2265625, + "learning_rate": 0.001209314783283875, + "loss": 0.0576, + "step": 44819 + }, + { + "epoch": 0.38905912275067056, + "grad_norm": 0.2392578125, + "learning_rate": 0.0012092852411220557, + "loss": 0.0786, + "step": 44820 + }, + { + "epoch": 0.38906780323087475, + "grad_norm": 0.1796875, + "learning_rate": 0.0012092556988407244, + "loss": 0.1172, + "step": 44821 + }, + { + "epoch": 0.3890764837110789, + "grad_norm": 0.091796875, + "learning_rate": 0.0012092261564399123, + "loss": 0.1309, + "step": 44822 + }, + { + "epoch": 0.3890851641912831, + "grad_norm": 0.353515625, + "learning_rate": 0.0012091966139196521, + "loss": 0.082, + "step": 44823 + }, + { + "epoch": 0.3890938446714872, + "grad_norm": 1.453125, + "learning_rate": 0.0012091670712799764, + "loss": 0.1328, + "step": 44824 + }, + { + "epoch": 0.3891025251516914, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012091375285209175, + "loss": 0.1006, + "step": 44825 + }, + { + "epoch": 0.38911120563189555, + "grad_norm": 0.130859375, + "learning_rate": 0.0012091079856425067, + "loss": 0.1011, + "step": 44826 + }, + { + "epoch": 0.38911988611209974, + "grad_norm": 0.162109375, + "learning_rate": 0.0012090784426447773, + "loss": 0.1108, + "step": 44827 + }, + { + "epoch": 0.3891285665923039, + "grad_norm": 0.421875, + "learning_rate": 0.0012090488995277618, + "loss": 0.1006, + "step": 44828 + }, + { + "epoch": 0.3891372470725081, + "grad_norm": 0.162109375, + "learning_rate": 0.0012090193562914915, + "loss": 0.084, + "step": 44829 + }, + { + "epoch": 0.3891459275527122, + "grad_norm": 0.4921875, + "learning_rate": 0.0012089898129359995, + "loss": 0.1006, + "step": 44830 + }, + { + "epoch": 0.3891546080329164, + "grad_norm": 0.328125, + "learning_rate": 0.0012089602694613179, + "loss": 0.1094, + "step": 44831 + }, + { + "epoch": 0.38916328851312054, + "grad_norm": 0.2451171875, + "learning_rate": 0.0012089307258674787, + "loss": 0.2109, + "step": 44832 + }, + { + "epoch": 0.38917196899332474, + "grad_norm": 0.30859375, + "learning_rate": 0.0012089011821545149, + "loss": 0.1318, + "step": 44833 + }, + { + "epoch": 0.3891806494735289, + "grad_norm": 0.61328125, + "learning_rate": 0.0012088716383224583, + "loss": 0.0854, + "step": 44834 + }, + { + "epoch": 0.38918932995373307, + "grad_norm": 0.51171875, + "learning_rate": 0.001208842094371341, + "loss": 0.1133, + "step": 44835 + }, + { + "epoch": 0.3891980104339372, + "grad_norm": 0.1943359375, + "learning_rate": 0.001208812550301196, + "loss": 0.1035, + "step": 44836 + }, + { + "epoch": 0.3892066909141414, + "grad_norm": 0.302734375, + "learning_rate": 0.001208783006112055, + "loss": 0.0918, + "step": 44837 + }, + { + "epoch": 0.38921537139434553, + "grad_norm": 0.421875, + "learning_rate": 0.0012087534618039507, + "loss": 0.0947, + "step": 44838 + }, + { + "epoch": 0.3892240518745497, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012087239173769154, + "loss": 0.1602, + "step": 44839 + }, + { + "epoch": 0.38923273235475386, + "grad_norm": 0.10205078125, + "learning_rate": 0.0012086943728309807, + "loss": 0.0879, + "step": 44840 + }, + { + "epoch": 0.38924141283495806, + "grad_norm": 0.2275390625, + "learning_rate": 0.0012086648281661799, + "loss": 0.0718, + "step": 44841 + }, + { + "epoch": 0.3892500933151622, + "grad_norm": 0.97265625, + "learning_rate": 0.0012086352833825449, + "loss": 0.1201, + "step": 44842 + }, + { + "epoch": 0.3892587737953664, + "grad_norm": 0.57421875, + "learning_rate": 0.0012086057384801078, + "loss": 0.1211, + "step": 44843 + }, + { + "epoch": 0.3892674542755705, + "grad_norm": 0.1708984375, + "learning_rate": 0.0012085761934589013, + "loss": 0.0703, + "step": 44844 + }, + { + "epoch": 0.3892761347557747, + "grad_norm": 0.37109375, + "learning_rate": 0.0012085466483189571, + "loss": 0.085, + "step": 44845 + }, + { + "epoch": 0.38928481523597885, + "grad_norm": 0.2236328125, + "learning_rate": 0.0012085171030603084, + "loss": 0.1074, + "step": 44846 + }, + { + "epoch": 0.38929349571618305, + "grad_norm": 0.18359375, + "learning_rate": 0.0012084875576829869, + "loss": 0.0801, + "step": 44847 + }, + { + "epoch": 0.3893021761963872, + "grad_norm": 0.25, + "learning_rate": 0.0012084580121870248, + "loss": 0.0723, + "step": 44848 + }, + { + "epoch": 0.3893108566765913, + "grad_norm": 0.3125, + "learning_rate": 0.0012084284665724546, + "loss": 0.1021, + "step": 44849 + }, + { + "epoch": 0.3893195371567955, + "grad_norm": 0.310546875, + "learning_rate": 0.001208398920839309, + "loss": 0.0889, + "step": 44850 + }, + { + "epoch": 0.38932821763699965, + "grad_norm": 0.330078125, + "learning_rate": 0.0012083693749876199, + "loss": 0.1084, + "step": 44851 + }, + { + "epoch": 0.38933689811720384, + "grad_norm": 0.1875, + "learning_rate": 0.0012083398290174195, + "loss": 0.0869, + "step": 44852 + }, + { + "epoch": 0.389345578597408, + "grad_norm": 0.5390625, + "learning_rate": 0.0012083102829287405, + "loss": 0.0918, + "step": 44853 + }, + { + "epoch": 0.3893542590776122, + "grad_norm": 0.1787109375, + "learning_rate": 0.0012082807367216149, + "loss": 0.1201, + "step": 44854 + }, + { + "epoch": 0.3893629395578163, + "grad_norm": 0.447265625, + "learning_rate": 0.001208251190396075, + "loss": 0.1089, + "step": 44855 + }, + { + "epoch": 0.3893716200380205, + "grad_norm": 0.162109375, + "learning_rate": 0.0012082216439521533, + "loss": 0.0938, + "step": 44856 + }, + { + "epoch": 0.38938030051822464, + "grad_norm": 0.1708984375, + "learning_rate": 0.001208192097389882, + "loss": 0.0742, + "step": 44857 + }, + { + "epoch": 0.38938898099842884, + "grad_norm": 0.95703125, + "learning_rate": 0.0012081625507092935, + "loss": 0.0889, + "step": 44858 + }, + { + "epoch": 0.389397661478633, + "grad_norm": 0.076171875, + "learning_rate": 0.00120813300391042, + "loss": 0.0791, + "step": 44859 + }, + { + "epoch": 0.38940634195883717, + "grad_norm": 0.478515625, + "learning_rate": 0.0012081034569932942, + "loss": 0.0957, + "step": 44860 + }, + { + "epoch": 0.3894150224390413, + "grad_norm": 0.703125, + "learning_rate": 0.0012080739099579478, + "loss": 0.0884, + "step": 44861 + }, + { + "epoch": 0.3894237029192455, + "grad_norm": 0.103515625, + "learning_rate": 0.0012080443628044134, + "loss": 0.084, + "step": 44862 + }, + { + "epoch": 0.38943238339944963, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012080148155327233, + "loss": 0.0708, + "step": 44863 + }, + { + "epoch": 0.3894410638796538, + "grad_norm": 0.11474609375, + "learning_rate": 0.00120798526814291, + "loss": 0.0752, + "step": 44864 + }, + { + "epoch": 0.38944974435985796, + "grad_norm": 0.2109375, + "learning_rate": 0.0012079557206350056, + "loss": 0.0698, + "step": 44865 + }, + { + "epoch": 0.38945842484006216, + "grad_norm": 0.412109375, + "learning_rate": 0.001207926173009042, + "loss": 0.1221, + "step": 44866 + }, + { + "epoch": 0.3894671053202663, + "grad_norm": 0.2041015625, + "learning_rate": 0.0012078966252650526, + "loss": 0.1182, + "step": 44867 + }, + { + "epoch": 0.3894757858004705, + "grad_norm": 1.3359375, + "learning_rate": 0.0012078670774030686, + "loss": 0.1113, + "step": 44868 + }, + { + "epoch": 0.3894844662806746, + "grad_norm": 0.2578125, + "learning_rate": 0.0012078375294231228, + "loss": 0.0801, + "step": 44869 + }, + { + "epoch": 0.3894931467608788, + "grad_norm": 0.44140625, + "learning_rate": 0.0012078079813252476, + "loss": 0.103, + "step": 44870 + }, + { + "epoch": 0.38950182724108295, + "grad_norm": 0.337890625, + "learning_rate": 0.0012077784331094754, + "loss": 0.0869, + "step": 44871 + }, + { + "epoch": 0.38951050772128715, + "grad_norm": 0.1123046875, + "learning_rate": 0.0012077488847758381, + "loss": 0.1211, + "step": 44872 + }, + { + "epoch": 0.3895191882014913, + "grad_norm": 0.11669921875, + "learning_rate": 0.0012077193363243686, + "loss": 0.1069, + "step": 44873 + }, + { + "epoch": 0.3895278686816955, + "grad_norm": 0.2109375, + "learning_rate": 0.0012076897877550984, + "loss": 0.1348, + "step": 44874 + }, + { + "epoch": 0.3895365491618996, + "grad_norm": 0.1787109375, + "learning_rate": 0.0012076602390680606, + "loss": 0.0771, + "step": 44875 + }, + { + "epoch": 0.3895452296421038, + "grad_norm": 0.11669921875, + "learning_rate": 0.001207630690263287, + "loss": 0.1167, + "step": 44876 + }, + { + "epoch": 0.38955391012230794, + "grad_norm": 0.369140625, + "learning_rate": 0.00120760114134081, + "loss": 0.1641, + "step": 44877 + }, + { + "epoch": 0.38956259060251214, + "grad_norm": 0.08642578125, + "learning_rate": 0.001207571592300662, + "loss": 0.104, + "step": 44878 + }, + { + "epoch": 0.3895712710827163, + "grad_norm": 0.357421875, + "learning_rate": 0.0012075420431428756, + "loss": 0.0623, + "step": 44879 + }, + { + "epoch": 0.38957995156292047, + "grad_norm": 0.6875, + "learning_rate": 0.0012075124938674827, + "loss": 0.0977, + "step": 44880 + }, + { + "epoch": 0.3895886320431246, + "grad_norm": 0.251953125, + "learning_rate": 0.0012074829444745157, + "loss": 0.0786, + "step": 44881 + }, + { + "epoch": 0.3895973125233288, + "grad_norm": 0.1201171875, + "learning_rate": 0.001207453394964007, + "loss": 0.0732, + "step": 44882 + }, + { + "epoch": 0.38960599300353294, + "grad_norm": 0.12451171875, + "learning_rate": 0.001207423845335989, + "loss": 0.0757, + "step": 44883 + }, + { + "epoch": 0.38961467348373713, + "grad_norm": 0.6796875, + "learning_rate": 0.0012073942955904938, + "loss": 0.0972, + "step": 44884 + }, + { + "epoch": 0.38962335396394127, + "grad_norm": 0.3515625, + "learning_rate": 0.0012073647457275538, + "loss": 0.0913, + "step": 44885 + }, + { + "epoch": 0.38963203444414546, + "grad_norm": 0.33984375, + "learning_rate": 0.0012073351957472011, + "loss": 0.0835, + "step": 44886 + }, + { + "epoch": 0.3896407149243496, + "grad_norm": 0.19921875, + "learning_rate": 0.0012073056456494687, + "loss": 0.0718, + "step": 44887 + }, + { + "epoch": 0.3896493954045538, + "grad_norm": 0.3125, + "learning_rate": 0.0012072760954343882, + "loss": 0.1172, + "step": 44888 + }, + { + "epoch": 0.3896580758847579, + "grad_norm": 0.42578125, + "learning_rate": 0.0012072465451019921, + "loss": 0.1167, + "step": 44889 + }, + { + "epoch": 0.3896667563649621, + "grad_norm": 0.08544921875, + "learning_rate": 0.001207216994652313, + "loss": 0.0952, + "step": 44890 + }, + { + "epoch": 0.38967543684516626, + "grad_norm": 0.3125, + "learning_rate": 0.001207187444085383, + "loss": 0.0898, + "step": 44891 + }, + { + "epoch": 0.38968411732537045, + "grad_norm": 0.271484375, + "learning_rate": 0.0012071578934012342, + "loss": 0.0894, + "step": 44892 + }, + { + "epoch": 0.3896927978055746, + "grad_norm": 0.482421875, + "learning_rate": 0.0012071283425998994, + "loss": 0.1123, + "step": 44893 + }, + { + "epoch": 0.3897014782857788, + "grad_norm": 0.126953125, + "learning_rate": 0.0012070987916814105, + "loss": 0.1152, + "step": 44894 + }, + { + "epoch": 0.3897101587659829, + "grad_norm": 0.126953125, + "learning_rate": 0.0012070692406457998, + "loss": 0.0874, + "step": 44895 + }, + { + "epoch": 0.3897188392461871, + "grad_norm": 0.166015625, + "learning_rate": 0.0012070396894931, + "loss": 0.0957, + "step": 44896 + }, + { + "epoch": 0.38972751972639125, + "grad_norm": 1.0, + "learning_rate": 0.001207010138223343, + "loss": 0.126, + "step": 44897 + }, + { + "epoch": 0.38973620020659544, + "grad_norm": 0.74609375, + "learning_rate": 0.0012069805868365616, + "loss": 0.0796, + "step": 44898 + }, + { + "epoch": 0.3897448806867996, + "grad_norm": 0.27734375, + "learning_rate": 0.0012069510353327876, + "loss": 0.1152, + "step": 44899 + }, + { + "epoch": 0.38975356116700377, + "grad_norm": 0.60546875, + "learning_rate": 0.0012069214837120536, + "loss": 0.0962, + "step": 44900 + }, + { + "epoch": 0.3897622416472079, + "grad_norm": 0.29296875, + "learning_rate": 0.001206891931974392, + "loss": 0.1045, + "step": 44901 + }, + { + "epoch": 0.3897709221274121, + "grad_norm": 0.2265625, + "learning_rate": 0.001206862380119835, + "loss": 0.1201, + "step": 44902 + }, + { + "epoch": 0.38977960260761624, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012068328281484152, + "loss": 0.0908, + "step": 44903 + }, + { + "epoch": 0.38978828308782043, + "grad_norm": 0.2890625, + "learning_rate": 0.0012068032760601639, + "loss": 0.1309, + "step": 44904 + }, + { + "epoch": 0.38979696356802457, + "grad_norm": 0.54296875, + "learning_rate": 0.0012067737238551147, + "loss": 0.1157, + "step": 44905 + }, + { + "epoch": 0.38980564404822876, + "grad_norm": 0.484375, + "learning_rate": 0.001206744171533299, + "loss": 0.0986, + "step": 44906 + }, + { + "epoch": 0.3898143245284329, + "grad_norm": 0.28515625, + "learning_rate": 0.0012067146190947499, + "loss": 0.1011, + "step": 44907 + }, + { + "epoch": 0.3898230050086371, + "grad_norm": 0.349609375, + "learning_rate": 0.0012066850665394988, + "loss": 0.0898, + "step": 44908 + }, + { + "epoch": 0.38983168548884123, + "grad_norm": 0.291015625, + "learning_rate": 0.0012066555138675788, + "loss": 0.0752, + "step": 44909 + }, + { + "epoch": 0.3898403659690454, + "grad_norm": 0.083984375, + "learning_rate": 0.001206625961079022, + "loss": 0.0889, + "step": 44910 + }, + { + "epoch": 0.38984904644924956, + "grad_norm": 0.283203125, + "learning_rate": 0.0012065964081738605, + "loss": 0.0918, + "step": 44911 + }, + { + "epoch": 0.38985772692945375, + "grad_norm": 0.265625, + "learning_rate": 0.001206566855152127, + "loss": 0.0933, + "step": 44912 + }, + { + "epoch": 0.3898664074096579, + "grad_norm": 0.349609375, + "learning_rate": 0.0012065373020138533, + "loss": 0.1318, + "step": 44913 + }, + { + "epoch": 0.3898750878898621, + "grad_norm": 0.171875, + "learning_rate": 0.001206507748759072, + "loss": 0.1011, + "step": 44914 + }, + { + "epoch": 0.3898837683700662, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012064781953878157, + "loss": 0.1182, + "step": 44915 + }, + { + "epoch": 0.3898924488502704, + "grad_norm": 0.193359375, + "learning_rate": 0.0012064486419001161, + "loss": 0.0723, + "step": 44916 + }, + { + "epoch": 0.38990112933047455, + "grad_norm": 0.1982421875, + "learning_rate": 0.001206419088296006, + "loss": 0.0947, + "step": 44917 + }, + { + "epoch": 0.38990980981067874, + "grad_norm": 0.2177734375, + "learning_rate": 0.0012063895345755177, + "loss": 0.0825, + "step": 44918 + }, + { + "epoch": 0.3899184902908829, + "grad_norm": 0.12451171875, + "learning_rate": 0.0012063599807386834, + "loss": 0.0918, + "step": 44919 + }, + { + "epoch": 0.3899271707710871, + "grad_norm": 0.11181640625, + "learning_rate": 0.0012063304267855355, + "loss": 0.1035, + "step": 44920 + }, + { + "epoch": 0.3899358512512912, + "grad_norm": 0.11962890625, + "learning_rate": 0.001206300872716106, + "loss": 0.0923, + "step": 44921 + }, + { + "epoch": 0.3899445317314954, + "grad_norm": 0.162109375, + "learning_rate": 0.0012062713185304278, + "loss": 0.1167, + "step": 44922 + }, + { + "epoch": 0.38995321221169954, + "grad_norm": 0.263671875, + "learning_rate": 0.0012062417642285323, + "loss": 0.0874, + "step": 44923 + }, + { + "epoch": 0.38996189269190373, + "grad_norm": 0.2060546875, + "learning_rate": 0.001206212209810453, + "loss": 0.0801, + "step": 44924 + }, + { + "epoch": 0.38997057317210787, + "grad_norm": 0.11328125, + "learning_rate": 0.0012061826552762212, + "loss": 0.1328, + "step": 44925 + }, + { + "epoch": 0.38997925365231206, + "grad_norm": 0.134765625, + "learning_rate": 0.0012061531006258698, + "loss": 0.1064, + "step": 44926 + }, + { + "epoch": 0.3899879341325162, + "grad_norm": 0.1943359375, + "learning_rate": 0.001206123545859431, + "loss": 0.0864, + "step": 44927 + }, + { + "epoch": 0.3899966146127204, + "grad_norm": 0.40625, + "learning_rate": 0.001206093990976937, + "loss": 0.0898, + "step": 44928 + }, + { + "epoch": 0.39000529509292453, + "grad_norm": 0.455078125, + "learning_rate": 0.0012060644359784202, + "loss": 0.207, + "step": 44929 + }, + { + "epoch": 0.3900139755731287, + "grad_norm": 0.1494140625, + "learning_rate": 0.001206034880863913, + "loss": 0.1006, + "step": 44930 + }, + { + "epoch": 0.39002265605333286, + "grad_norm": 0.234375, + "learning_rate": 0.0012060053256334477, + "loss": 0.0869, + "step": 44931 + }, + { + "epoch": 0.39003133653353705, + "grad_norm": 0.0712890625, + "learning_rate": 0.0012059757702870566, + "loss": 0.0483, + "step": 44932 + }, + { + "epoch": 0.3900400170137412, + "grad_norm": 0.48046875, + "learning_rate": 0.001205946214824772, + "loss": 0.1104, + "step": 44933 + }, + { + "epoch": 0.3900486974939454, + "grad_norm": 0.54296875, + "learning_rate": 0.001205916659246626, + "loss": 0.0918, + "step": 44934 + }, + { + "epoch": 0.3900573779741495, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012058871035526514, + "loss": 0.1221, + "step": 44935 + }, + { + "epoch": 0.3900660584543537, + "grad_norm": 0.3828125, + "learning_rate": 0.00120585754774288, + "loss": 0.082, + "step": 44936 + }, + { + "epoch": 0.39007473893455785, + "grad_norm": 0.271484375, + "learning_rate": 0.0012058279918173444, + "loss": 0.0845, + "step": 44937 + }, + { + "epoch": 0.39008341941476204, + "grad_norm": 0.29296875, + "learning_rate": 0.0012057984357760772, + "loss": 0.105, + "step": 44938 + }, + { + "epoch": 0.3900920998949662, + "grad_norm": 0.1708984375, + "learning_rate": 0.00120576887961911, + "loss": 0.0908, + "step": 44939 + }, + { + "epoch": 0.3901007803751704, + "grad_norm": 0.373046875, + "learning_rate": 0.0012057393233464758, + "loss": 0.104, + "step": 44940 + }, + { + "epoch": 0.3901094608553745, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012057097669582067, + "loss": 0.1216, + "step": 44941 + }, + { + "epoch": 0.3901181413355787, + "grad_norm": 0.34375, + "learning_rate": 0.001205680210454335, + "loss": 0.0889, + "step": 44942 + }, + { + "epoch": 0.39012682181578284, + "grad_norm": 0.197265625, + "learning_rate": 0.0012056506538348929, + "loss": 0.1475, + "step": 44943 + }, + { + "epoch": 0.39013550229598704, + "grad_norm": 0.10498046875, + "learning_rate": 0.0012056210970999126, + "loss": 0.1445, + "step": 44944 + }, + { + "epoch": 0.3901441827761912, + "grad_norm": 0.2421875, + "learning_rate": 0.0012055915402494266, + "loss": 0.0771, + "step": 44945 + }, + { + "epoch": 0.39015286325639537, + "grad_norm": 0.26953125, + "learning_rate": 0.0012055619832834676, + "loss": 0.1309, + "step": 44946 + }, + { + "epoch": 0.3901615437365995, + "grad_norm": 0.80078125, + "learning_rate": 0.0012055324262020676, + "loss": 0.1338, + "step": 44947 + }, + { + "epoch": 0.3901702242168037, + "grad_norm": 0.10888671875, + "learning_rate": 0.0012055028690052593, + "loss": 0.0811, + "step": 44948 + }, + { + "epoch": 0.39017890469700783, + "grad_norm": 0.10498046875, + "learning_rate": 0.001205473311693074, + "loss": 0.0845, + "step": 44949 + }, + { + "epoch": 0.390187585177212, + "grad_norm": 0.52734375, + "learning_rate": 0.001205443754265545, + "loss": 0.1162, + "step": 44950 + }, + { + "epoch": 0.39019626565741616, + "grad_norm": 0.171875, + "learning_rate": 0.0012054141967227045, + "loss": 0.0898, + "step": 44951 + }, + { + "epoch": 0.39020494613762036, + "grad_norm": 0.1728515625, + "learning_rate": 0.001205384639064584, + "loss": 0.1055, + "step": 44952 + }, + { + "epoch": 0.3902136266178245, + "grad_norm": 0.19921875, + "learning_rate": 0.0012053550812912166, + "loss": 0.0991, + "step": 44953 + }, + { + "epoch": 0.3902223070980287, + "grad_norm": 0.291015625, + "learning_rate": 0.0012053255234026347, + "loss": 0.1035, + "step": 44954 + }, + { + "epoch": 0.3902309875782328, + "grad_norm": 0.267578125, + "learning_rate": 0.0012052959653988703, + "loss": 0.082, + "step": 44955 + }, + { + "epoch": 0.390239668058437, + "grad_norm": 0.43359375, + "learning_rate": 0.0012052664072799558, + "loss": 0.0947, + "step": 44956 + }, + { + "epoch": 0.39024834853864115, + "grad_norm": 0.330078125, + "learning_rate": 0.0012052368490459236, + "loss": 0.0913, + "step": 44957 + }, + { + "epoch": 0.39025702901884535, + "grad_norm": 0.1201171875, + "learning_rate": 0.001205207290696806, + "loss": 0.0776, + "step": 44958 + }, + { + "epoch": 0.3902657094990495, + "grad_norm": 0.203125, + "learning_rate": 0.0012051777322326354, + "loss": 0.0996, + "step": 44959 + }, + { + "epoch": 0.3902743899792537, + "grad_norm": 0.19140625, + "learning_rate": 0.0012051481736534435, + "loss": 0.0977, + "step": 44960 + }, + { + "epoch": 0.3902830704594578, + "grad_norm": 0.63671875, + "learning_rate": 0.0012051186149592637, + "loss": 0.0684, + "step": 44961 + }, + { + "epoch": 0.390291750939662, + "grad_norm": 0.1494140625, + "learning_rate": 0.0012050890561501276, + "loss": 0.0864, + "step": 44962 + }, + { + "epoch": 0.39030043141986615, + "grad_norm": 0.58984375, + "learning_rate": 0.0012050594972260675, + "loss": 0.1064, + "step": 44963 + }, + { + "epoch": 0.39030911190007034, + "grad_norm": 0.2021484375, + "learning_rate": 0.001205029938187116, + "loss": 0.0977, + "step": 44964 + }, + { + "epoch": 0.3903177923802745, + "grad_norm": 0.61328125, + "learning_rate": 0.0012050003790333053, + "loss": 0.1348, + "step": 44965 + }, + { + "epoch": 0.39032647286047867, + "grad_norm": 0.07666015625, + "learning_rate": 0.0012049708197646681, + "loss": 0.063, + "step": 44966 + }, + { + "epoch": 0.3903351533406828, + "grad_norm": 0.33203125, + "learning_rate": 0.001204941260381236, + "loss": 0.125, + "step": 44967 + }, + { + "epoch": 0.390343833820887, + "grad_norm": 0.1669921875, + "learning_rate": 0.0012049117008830418, + "loss": 0.0835, + "step": 44968 + }, + { + "epoch": 0.39035251430109114, + "grad_norm": 0.14453125, + "learning_rate": 0.0012048821412701178, + "loss": 0.084, + "step": 44969 + }, + { + "epoch": 0.39036119478129533, + "grad_norm": 0.1767578125, + "learning_rate": 0.0012048525815424962, + "loss": 0.0938, + "step": 44970 + }, + { + "epoch": 0.39036987526149947, + "grad_norm": 0.1572265625, + "learning_rate": 0.0012048230217002096, + "loss": 0.1035, + "step": 44971 + }, + { + "epoch": 0.3903785557417036, + "grad_norm": 0.1552734375, + "learning_rate": 0.00120479346174329, + "loss": 0.1348, + "step": 44972 + }, + { + "epoch": 0.3903872362219078, + "grad_norm": 0.263671875, + "learning_rate": 0.0012047639016717698, + "loss": 0.1377, + "step": 44973 + }, + { + "epoch": 0.39039591670211193, + "grad_norm": 0.58984375, + "learning_rate": 0.0012047343414856814, + "loss": 0.1177, + "step": 44974 + }, + { + "epoch": 0.3904045971823161, + "grad_norm": 0.408203125, + "learning_rate": 0.001204704781185057, + "loss": 0.0928, + "step": 44975 + }, + { + "epoch": 0.39041327766252026, + "grad_norm": 0.6015625, + "learning_rate": 0.001204675220769929, + "loss": 0.1504, + "step": 44976 + }, + { + "epoch": 0.39042195814272446, + "grad_norm": 0.40625, + "learning_rate": 0.0012046456602403297, + "loss": 0.0947, + "step": 44977 + }, + { + "epoch": 0.3904306386229286, + "grad_norm": 0.94921875, + "learning_rate": 0.0012046160995962919, + "loss": 0.0879, + "step": 44978 + }, + { + "epoch": 0.3904393191031328, + "grad_norm": 0.10009765625, + "learning_rate": 0.001204586538837847, + "loss": 0.0972, + "step": 44979 + }, + { + "epoch": 0.3904479995833369, + "grad_norm": 0.1435546875, + "learning_rate": 0.0012045569779650283, + "loss": 0.0801, + "step": 44980 + }, + { + "epoch": 0.3904566800635411, + "grad_norm": 0.318359375, + "learning_rate": 0.0012045274169778675, + "loss": 0.0938, + "step": 44981 + }, + { + "epoch": 0.39046536054374525, + "grad_norm": 0.1826171875, + "learning_rate": 0.0012044978558763968, + "loss": 0.1221, + "step": 44982 + }, + { + "epoch": 0.39047404102394945, + "grad_norm": 0.3828125, + "learning_rate": 0.0012044682946606492, + "loss": 0.1309, + "step": 44983 + }, + { + "epoch": 0.3904827215041536, + "grad_norm": 0.212890625, + "learning_rate": 0.0012044387333306563, + "loss": 0.126, + "step": 44984 + }, + { + "epoch": 0.3904914019843578, + "grad_norm": 0.23046875, + "learning_rate": 0.0012044091718864509, + "loss": 0.0879, + "step": 44985 + }, + { + "epoch": 0.3905000824645619, + "grad_norm": 0.8359375, + "learning_rate": 0.0012043796103280652, + "loss": 0.1045, + "step": 44986 + }, + { + "epoch": 0.3905087629447661, + "grad_norm": 0.11572265625, + "learning_rate": 0.0012043500486555317, + "loss": 0.1152, + "step": 44987 + }, + { + "epoch": 0.39051744342497025, + "grad_norm": 0.486328125, + "learning_rate": 0.0012043204868688825, + "loss": 0.1021, + "step": 44988 + }, + { + "epoch": 0.39052612390517444, + "grad_norm": 0.2431640625, + "learning_rate": 0.00120429092496815, + "loss": 0.0947, + "step": 44989 + }, + { + "epoch": 0.3905348043853786, + "grad_norm": 0.59375, + "learning_rate": 0.001204261362953366, + "loss": 0.1094, + "step": 44990 + }, + { + "epoch": 0.39054348486558277, + "grad_norm": 0.10205078125, + "learning_rate": 0.001204231800824564, + "loss": 0.1152, + "step": 44991 + }, + { + "epoch": 0.3905521653457869, + "grad_norm": 0.267578125, + "learning_rate": 0.0012042022385817756, + "loss": 0.1011, + "step": 44992 + }, + { + "epoch": 0.3905608458259911, + "grad_norm": 0.353515625, + "learning_rate": 0.0012041726762250327, + "loss": 0.1113, + "step": 44993 + }, + { + "epoch": 0.39056952630619524, + "grad_norm": 0.36328125, + "learning_rate": 0.0012041431137543685, + "loss": 0.1162, + "step": 44994 + }, + { + "epoch": 0.39057820678639943, + "grad_norm": 0.21484375, + "learning_rate": 0.001204113551169815, + "loss": 0.1064, + "step": 44995 + }, + { + "epoch": 0.39058688726660357, + "grad_norm": 0.06591796875, + "learning_rate": 0.0012040839884714042, + "loss": 0.0791, + "step": 44996 + }, + { + "epoch": 0.39059556774680776, + "grad_norm": 0.298828125, + "learning_rate": 0.001204054425659169, + "loss": 0.1201, + "step": 44997 + }, + { + "epoch": 0.3906042482270119, + "grad_norm": 0.265625, + "learning_rate": 0.0012040248627331414, + "loss": 0.082, + "step": 44998 + }, + { + "epoch": 0.3906129287072161, + "grad_norm": 0.33203125, + "learning_rate": 0.0012039952996933537, + "loss": 0.1074, + "step": 44999 + }, + { + "epoch": 0.3906216091874202, + "grad_norm": 0.2255859375, + "learning_rate": 0.0012039657365398384, + "loss": 0.1235, + "step": 45000 + }, + { + "epoch": 0.3906302896676244, + "grad_norm": 0.0927734375, + "learning_rate": 0.0012039361732726277, + "loss": 0.0752, + "step": 45001 + }, + { + "epoch": 0.39063897014782856, + "grad_norm": 0.26171875, + "learning_rate": 0.0012039066098917537, + "loss": 0.1113, + "step": 45002 + }, + { + "epoch": 0.39064765062803275, + "grad_norm": 0.259765625, + "learning_rate": 0.001203877046397249, + "loss": 0.0957, + "step": 45003 + }, + { + "epoch": 0.3906563311082369, + "grad_norm": 0.158203125, + "learning_rate": 0.0012038474827891464, + "loss": 0.1133, + "step": 45004 + }, + { + "epoch": 0.3906650115884411, + "grad_norm": 0.318359375, + "learning_rate": 0.0012038179190674773, + "loss": 0.1309, + "step": 45005 + }, + { + "epoch": 0.3906736920686452, + "grad_norm": 0.326171875, + "learning_rate": 0.0012037883552322747, + "loss": 0.0889, + "step": 45006 + }, + { + "epoch": 0.3906823725488494, + "grad_norm": 0.294921875, + "learning_rate": 0.001203758791283571, + "loss": 0.0908, + "step": 45007 + }, + { + "epoch": 0.39069105302905355, + "grad_norm": 0.2060546875, + "learning_rate": 0.0012037292272213977, + "loss": 0.0889, + "step": 45008 + }, + { + "epoch": 0.39069973350925774, + "grad_norm": 0.333984375, + "learning_rate": 0.001203699663045788, + "loss": 0.0957, + "step": 45009 + }, + { + "epoch": 0.3907084139894619, + "grad_norm": 0.076171875, + "learning_rate": 0.0012036700987567738, + "loss": 0.0845, + "step": 45010 + }, + { + "epoch": 0.39071709446966607, + "grad_norm": 0.1884765625, + "learning_rate": 0.0012036405343543875, + "loss": 0.0654, + "step": 45011 + }, + { + "epoch": 0.3907257749498702, + "grad_norm": 0.40234375, + "learning_rate": 0.0012036109698386616, + "loss": 0.1455, + "step": 45012 + }, + { + "epoch": 0.3907344554300744, + "grad_norm": 0.87890625, + "learning_rate": 0.0012035814052096278, + "loss": 0.0869, + "step": 45013 + }, + { + "epoch": 0.39074313591027854, + "grad_norm": 0.154296875, + "learning_rate": 0.0012035518404673196, + "loss": 0.1152, + "step": 45014 + }, + { + "epoch": 0.39075181639048273, + "grad_norm": 0.27734375, + "learning_rate": 0.0012035222756117685, + "loss": 0.0947, + "step": 45015 + }, + { + "epoch": 0.39076049687068687, + "grad_norm": 0.80078125, + "learning_rate": 0.0012034927106430067, + "loss": 0.1162, + "step": 45016 + }, + { + "epoch": 0.39076917735089106, + "grad_norm": 0.2890625, + "learning_rate": 0.0012034631455610672, + "loss": 0.1318, + "step": 45017 + }, + { + "epoch": 0.3907778578310952, + "grad_norm": 0.28125, + "learning_rate": 0.001203433580365982, + "loss": 0.0967, + "step": 45018 + }, + { + "epoch": 0.3907865383112994, + "grad_norm": 0.37109375, + "learning_rate": 0.0012034040150577829, + "loss": 0.1377, + "step": 45019 + }, + { + "epoch": 0.39079521879150353, + "grad_norm": 0.2041015625, + "learning_rate": 0.001203374449636503, + "loss": 0.1021, + "step": 45020 + }, + { + "epoch": 0.3908038992717077, + "grad_norm": 0.236328125, + "learning_rate": 0.0012033448841021744, + "loss": 0.0977, + "step": 45021 + }, + { + "epoch": 0.39081257975191186, + "grad_norm": 0.146484375, + "learning_rate": 0.0012033153184548293, + "loss": 0.0874, + "step": 45022 + }, + { + "epoch": 0.39082126023211605, + "grad_norm": 0.19921875, + "learning_rate": 0.0012032857526945, + "loss": 0.0898, + "step": 45023 + }, + { + "epoch": 0.3908299407123202, + "grad_norm": 0.091796875, + "learning_rate": 0.0012032561868212193, + "loss": 0.0962, + "step": 45024 + }, + { + "epoch": 0.3908386211925244, + "grad_norm": 0.1357421875, + "learning_rate": 0.0012032266208350192, + "loss": 0.1006, + "step": 45025 + }, + { + "epoch": 0.3908473016727285, + "grad_norm": 0.384765625, + "learning_rate": 0.0012031970547359317, + "loss": 0.0908, + "step": 45026 + }, + { + "epoch": 0.3908559821529327, + "grad_norm": 0.1982421875, + "learning_rate": 0.0012031674885239897, + "loss": 0.1094, + "step": 45027 + }, + { + "epoch": 0.39086466263313685, + "grad_norm": 0.298828125, + "learning_rate": 0.0012031379221992251, + "loss": 0.0742, + "step": 45028 + }, + { + "epoch": 0.39087334311334104, + "grad_norm": 0.470703125, + "learning_rate": 0.0012031083557616704, + "loss": 0.123, + "step": 45029 + }, + { + "epoch": 0.3908820235935452, + "grad_norm": 0.2177734375, + "learning_rate": 0.0012030787892113582, + "loss": 0.063, + "step": 45030 + }, + { + "epoch": 0.3908907040737494, + "grad_norm": 0.341796875, + "learning_rate": 0.0012030492225483204, + "loss": 0.1094, + "step": 45031 + }, + { + "epoch": 0.3908993845539535, + "grad_norm": 0.138671875, + "learning_rate": 0.0012030196557725893, + "loss": 0.1191, + "step": 45032 + }, + { + "epoch": 0.3909080650341577, + "grad_norm": 0.39453125, + "learning_rate": 0.0012029900888841977, + "loss": 0.1104, + "step": 45033 + }, + { + "epoch": 0.39091674551436184, + "grad_norm": 0.0830078125, + "learning_rate": 0.0012029605218831776, + "loss": 0.0928, + "step": 45034 + }, + { + "epoch": 0.39092542599456603, + "grad_norm": 0.11474609375, + "learning_rate": 0.0012029309547695614, + "loss": 0.0918, + "step": 45035 + }, + { + "epoch": 0.39093410647477017, + "grad_norm": 0.294921875, + "learning_rate": 0.0012029013875433816, + "loss": 0.1318, + "step": 45036 + }, + { + "epoch": 0.39094278695497436, + "grad_norm": 0.443359375, + "learning_rate": 0.0012028718202046708, + "loss": 0.0781, + "step": 45037 + }, + { + "epoch": 0.3909514674351785, + "grad_norm": 0.349609375, + "learning_rate": 0.00120284225275346, + "loss": 0.1064, + "step": 45038 + }, + { + "epoch": 0.3909601479153827, + "grad_norm": 0.2119140625, + "learning_rate": 0.0012028126851897829, + "loss": 0.085, + "step": 45039 + }, + { + "epoch": 0.39096882839558683, + "grad_norm": 0.12451171875, + "learning_rate": 0.0012027831175136716, + "loss": 0.0835, + "step": 45040 + }, + { + "epoch": 0.390977508875791, + "grad_norm": 0.140625, + "learning_rate": 0.0012027535497251577, + "loss": 0.0601, + "step": 45041 + }, + { + "epoch": 0.39098618935599516, + "grad_norm": 0.1533203125, + "learning_rate": 0.0012027239818242746, + "loss": 0.0938, + "step": 45042 + }, + { + "epoch": 0.39099486983619935, + "grad_norm": 0.1708984375, + "learning_rate": 0.001202694413811054, + "loss": 0.1128, + "step": 45043 + }, + { + "epoch": 0.3910035503164035, + "grad_norm": 0.10400390625, + "learning_rate": 0.001202664845685528, + "loss": 0.0938, + "step": 45044 + }, + { + "epoch": 0.3910122307966077, + "grad_norm": 0.0576171875, + "learning_rate": 0.0012026352774477298, + "loss": 0.064, + "step": 45045 + }, + { + "epoch": 0.3910209112768118, + "grad_norm": 0.6875, + "learning_rate": 0.001202605709097691, + "loss": 0.1279, + "step": 45046 + }, + { + "epoch": 0.391029591757016, + "grad_norm": 0.28515625, + "learning_rate": 0.001202576140635444, + "loss": 0.0811, + "step": 45047 + }, + { + "epoch": 0.39103827223722015, + "grad_norm": 0.1796875, + "learning_rate": 0.001202546572061021, + "loss": 0.0957, + "step": 45048 + }, + { + "epoch": 0.39104695271742435, + "grad_norm": 0.3359375, + "learning_rate": 0.001202517003374455, + "loss": 0.0928, + "step": 45049 + }, + { + "epoch": 0.3910556331976285, + "grad_norm": 0.109375, + "learning_rate": 0.0012024874345757776, + "loss": 0.0996, + "step": 45050 + }, + { + "epoch": 0.3910643136778327, + "grad_norm": 0.08935546875, + "learning_rate": 0.0012024578656650217, + "loss": 0.0874, + "step": 45051 + }, + { + "epoch": 0.3910729941580368, + "grad_norm": 0.158203125, + "learning_rate": 0.0012024282966422194, + "loss": 0.1279, + "step": 45052 + }, + { + "epoch": 0.391081674638241, + "grad_norm": 0.31640625, + "learning_rate": 0.0012023987275074036, + "loss": 0.0923, + "step": 45053 + }, + { + "epoch": 0.39109035511844514, + "grad_norm": 0.40625, + "learning_rate": 0.0012023691582606054, + "loss": 0.1162, + "step": 45054 + }, + { + "epoch": 0.39109903559864934, + "grad_norm": 0.33984375, + "learning_rate": 0.001202339588901858, + "loss": 0.0991, + "step": 45055 + }, + { + "epoch": 0.3911077160788535, + "grad_norm": 0.1171875, + "learning_rate": 0.0012023100194311935, + "loss": 0.1001, + "step": 45056 + }, + { + "epoch": 0.39111639655905767, + "grad_norm": 0.427734375, + "learning_rate": 0.0012022804498486447, + "loss": 0.1367, + "step": 45057 + }, + { + "epoch": 0.3911250770392618, + "grad_norm": 0.2216796875, + "learning_rate": 0.0012022508801542433, + "loss": 0.0864, + "step": 45058 + }, + { + "epoch": 0.391133757519466, + "grad_norm": 0.091796875, + "learning_rate": 0.0012022213103480214, + "loss": 0.082, + "step": 45059 + }, + { + "epoch": 0.39114243799967013, + "grad_norm": 0.177734375, + "learning_rate": 0.0012021917404300125, + "loss": 0.1348, + "step": 45060 + }, + { + "epoch": 0.3911511184798743, + "grad_norm": 0.34765625, + "learning_rate": 0.0012021621704002476, + "loss": 0.0981, + "step": 45061 + }, + { + "epoch": 0.39115979896007846, + "grad_norm": 0.146484375, + "learning_rate": 0.0012021326002587603, + "loss": 0.0732, + "step": 45062 + }, + { + "epoch": 0.39116847944028266, + "grad_norm": 0.279296875, + "learning_rate": 0.001202103030005582, + "loss": 0.1162, + "step": 45063 + }, + { + "epoch": 0.3911771599204868, + "grad_norm": 0.12890625, + "learning_rate": 0.0012020734596407455, + "loss": 0.0762, + "step": 45064 + }, + { + "epoch": 0.391185840400691, + "grad_norm": 0.18359375, + "learning_rate": 0.0012020438891642829, + "loss": 0.1006, + "step": 45065 + }, + { + "epoch": 0.3911945208808951, + "grad_norm": 0.453125, + "learning_rate": 0.0012020143185762268, + "loss": 0.062, + "step": 45066 + }, + { + "epoch": 0.3912032013610993, + "grad_norm": 0.5859375, + "learning_rate": 0.0012019847478766093, + "loss": 0.0938, + "step": 45067 + }, + { + "epoch": 0.39121188184130345, + "grad_norm": 0.431640625, + "learning_rate": 0.0012019551770654625, + "loss": 0.0757, + "step": 45068 + }, + { + "epoch": 0.39122056232150765, + "grad_norm": 0.44140625, + "learning_rate": 0.0012019256061428195, + "loss": 0.0703, + "step": 45069 + }, + { + "epoch": 0.3912292428017118, + "grad_norm": 0.412109375, + "learning_rate": 0.0012018960351087118, + "loss": 0.1279, + "step": 45070 + }, + { + "epoch": 0.391237923281916, + "grad_norm": 0.60546875, + "learning_rate": 0.0012018664639631724, + "loss": 0.1436, + "step": 45071 + }, + { + "epoch": 0.3912466037621201, + "grad_norm": 0.21875, + "learning_rate": 0.0012018368927062336, + "loss": 0.0996, + "step": 45072 + }, + { + "epoch": 0.3912552842423243, + "grad_norm": 0.453125, + "learning_rate": 0.0012018073213379268, + "loss": 0.1201, + "step": 45073 + }, + { + "epoch": 0.39126396472252845, + "grad_norm": 0.474609375, + "learning_rate": 0.0012017777498582855, + "loss": 0.0977, + "step": 45074 + }, + { + "epoch": 0.39127264520273264, + "grad_norm": 0.1640625, + "learning_rate": 0.0012017481782673415, + "loss": 0.0933, + "step": 45075 + }, + { + "epoch": 0.3912813256829368, + "grad_norm": 0.31640625, + "learning_rate": 0.0012017186065651276, + "loss": 0.1738, + "step": 45076 + }, + { + "epoch": 0.39129000616314097, + "grad_norm": 0.51953125, + "learning_rate": 0.0012016890347516752, + "loss": 0.0972, + "step": 45077 + }, + { + "epoch": 0.3912986866433451, + "grad_norm": 0.2578125, + "learning_rate": 0.0012016594628270173, + "loss": 0.0947, + "step": 45078 + }, + { + "epoch": 0.3913073671235493, + "grad_norm": 0.263671875, + "learning_rate": 0.0012016298907911863, + "loss": 0.0693, + "step": 45079 + }, + { + "epoch": 0.39131604760375344, + "grad_norm": 0.38671875, + "learning_rate": 0.0012016003186442142, + "loss": 0.0938, + "step": 45080 + }, + { + "epoch": 0.39132472808395763, + "grad_norm": 0.330078125, + "learning_rate": 0.0012015707463861333, + "loss": 0.0664, + "step": 45081 + }, + { + "epoch": 0.39133340856416177, + "grad_norm": 0.5625, + "learning_rate": 0.0012015411740169768, + "loss": 0.1543, + "step": 45082 + }, + { + "epoch": 0.39134208904436596, + "grad_norm": 0.376953125, + "learning_rate": 0.001201511601536776, + "loss": 0.1113, + "step": 45083 + }, + { + "epoch": 0.3913507695245701, + "grad_norm": 0.08056640625, + "learning_rate": 0.0012014820289455639, + "loss": 0.0815, + "step": 45084 + }, + { + "epoch": 0.3913594500047743, + "grad_norm": 0.12255859375, + "learning_rate": 0.0012014524562433723, + "loss": 0.1367, + "step": 45085 + }, + { + "epoch": 0.3913681304849784, + "grad_norm": 0.15625, + "learning_rate": 0.0012014228834302339, + "loss": 0.1094, + "step": 45086 + }, + { + "epoch": 0.3913768109651826, + "grad_norm": 0.361328125, + "learning_rate": 0.001201393310506181, + "loss": 0.1025, + "step": 45087 + }, + { + "epoch": 0.39138549144538676, + "grad_norm": 0.146484375, + "learning_rate": 0.0012013637374712457, + "loss": 0.0791, + "step": 45088 + }, + { + "epoch": 0.39139417192559095, + "grad_norm": 0.427734375, + "learning_rate": 0.0012013341643254609, + "loss": 0.1465, + "step": 45089 + }, + { + "epoch": 0.3914028524057951, + "grad_norm": 0.19921875, + "learning_rate": 0.0012013045910688582, + "loss": 0.123, + "step": 45090 + }, + { + "epoch": 0.3914115328859993, + "grad_norm": 0.07470703125, + "learning_rate": 0.0012012750177014704, + "loss": 0.0913, + "step": 45091 + }, + { + "epoch": 0.3914202133662034, + "grad_norm": 0.3515625, + "learning_rate": 0.00120124544422333, + "loss": 0.082, + "step": 45092 + }, + { + "epoch": 0.3914288938464076, + "grad_norm": 0.185546875, + "learning_rate": 0.001201215870634469, + "loss": 0.0942, + "step": 45093 + }, + { + "epoch": 0.39143757432661175, + "grad_norm": 0.271484375, + "learning_rate": 0.00120118629693492, + "loss": 0.0928, + "step": 45094 + }, + { + "epoch": 0.3914462548068159, + "grad_norm": 0.08349609375, + "learning_rate": 0.001201156723124715, + "loss": 0.0771, + "step": 45095 + }, + { + "epoch": 0.3914549352870201, + "grad_norm": 0.4375, + "learning_rate": 0.0012011271492038864, + "loss": 0.0967, + "step": 45096 + }, + { + "epoch": 0.3914636157672242, + "grad_norm": 0.04443359375, + "learning_rate": 0.001201097575172467, + "loss": 0.0449, + "step": 45097 + }, + { + "epoch": 0.3914722962474284, + "grad_norm": 0.1953125, + "learning_rate": 0.0012010680010304886, + "loss": 0.0967, + "step": 45098 + }, + { + "epoch": 0.39148097672763255, + "grad_norm": 0.0771484375, + "learning_rate": 0.0012010384267779836, + "loss": 0.0674, + "step": 45099 + }, + { + "epoch": 0.39148965720783674, + "grad_norm": 0.40625, + "learning_rate": 0.0012010088524149847, + "loss": 0.1484, + "step": 45100 + }, + { + "epoch": 0.3914983376880409, + "grad_norm": 0.26953125, + "learning_rate": 0.001200979277941524, + "loss": 0.1201, + "step": 45101 + }, + { + "epoch": 0.39150701816824507, + "grad_norm": 0.306640625, + "learning_rate": 0.001200949703357634, + "loss": 0.1367, + "step": 45102 + }, + { + "epoch": 0.3915156986484492, + "grad_norm": 0.490234375, + "learning_rate": 0.0012009201286633468, + "loss": 0.1021, + "step": 45103 + }, + { + "epoch": 0.3915243791286534, + "grad_norm": 0.396484375, + "learning_rate": 0.0012008905538586946, + "loss": 0.1055, + "step": 45104 + }, + { + "epoch": 0.39153305960885754, + "grad_norm": 0.1376953125, + "learning_rate": 0.0012008609789437106, + "loss": 0.0825, + "step": 45105 + }, + { + "epoch": 0.39154174008906173, + "grad_norm": 1.015625, + "learning_rate": 0.0012008314039184263, + "loss": 0.1152, + "step": 45106 + }, + { + "epoch": 0.39155042056926587, + "grad_norm": 0.34765625, + "learning_rate": 0.0012008018287828742, + "loss": 0.1074, + "step": 45107 + }, + { + "epoch": 0.39155910104947006, + "grad_norm": 0.27734375, + "learning_rate": 0.0012007722535370864, + "loss": 0.0645, + "step": 45108 + }, + { + "epoch": 0.3915677815296742, + "grad_norm": 0.73046875, + "learning_rate": 0.001200742678181096, + "loss": 0.127, + "step": 45109 + }, + { + "epoch": 0.3915764620098784, + "grad_norm": 0.1455078125, + "learning_rate": 0.001200713102714935, + "loss": 0.0859, + "step": 45110 + }, + { + "epoch": 0.3915851424900825, + "grad_norm": 0.1259765625, + "learning_rate": 0.0012006835271386354, + "loss": 0.123, + "step": 45111 + }, + { + "epoch": 0.3915938229702867, + "grad_norm": 0.49609375, + "learning_rate": 0.0012006539514522298, + "loss": 0.0767, + "step": 45112 + }, + { + "epoch": 0.39160250345049086, + "grad_norm": 0.416015625, + "learning_rate": 0.0012006243756557508, + "loss": 0.0908, + "step": 45113 + }, + { + "epoch": 0.39161118393069505, + "grad_norm": 0.486328125, + "learning_rate": 0.0012005947997492303, + "loss": 0.1069, + "step": 45114 + }, + { + "epoch": 0.3916198644108992, + "grad_norm": 0.125, + "learning_rate": 0.0012005652237327012, + "loss": 0.0674, + "step": 45115 + }, + { + "epoch": 0.3916285448911034, + "grad_norm": 0.6171875, + "learning_rate": 0.001200535647606195, + "loss": 0.083, + "step": 45116 + }, + { + "epoch": 0.3916372253713075, + "grad_norm": 0.08056640625, + "learning_rate": 0.0012005060713697446, + "loss": 0.0752, + "step": 45117 + }, + { + "epoch": 0.3916459058515117, + "grad_norm": 0.416015625, + "learning_rate": 0.0012004764950233823, + "loss": 0.0825, + "step": 45118 + }, + { + "epoch": 0.39165458633171585, + "grad_norm": 0.1337890625, + "learning_rate": 0.0012004469185671407, + "loss": 0.1016, + "step": 45119 + }, + { + "epoch": 0.39166326681192004, + "grad_norm": 0.54296875, + "learning_rate": 0.0012004173420010517, + "loss": 0.0732, + "step": 45120 + }, + { + "epoch": 0.3916719472921242, + "grad_norm": 0.421875, + "learning_rate": 0.0012003877653251477, + "loss": 0.0762, + "step": 45121 + }, + { + "epoch": 0.39168062777232837, + "grad_norm": 0.1806640625, + "learning_rate": 0.0012003581885394613, + "loss": 0.0996, + "step": 45122 + }, + { + "epoch": 0.3916893082525325, + "grad_norm": 0.154296875, + "learning_rate": 0.0012003286116440246, + "loss": 0.1445, + "step": 45123 + }, + { + "epoch": 0.3916979887327367, + "grad_norm": 0.1025390625, + "learning_rate": 0.0012002990346388698, + "loss": 0.1089, + "step": 45124 + }, + { + "epoch": 0.39170666921294084, + "grad_norm": 0.306640625, + "learning_rate": 0.0012002694575240298, + "loss": 0.0879, + "step": 45125 + }, + { + "epoch": 0.39171534969314503, + "grad_norm": 0.326171875, + "learning_rate": 0.0012002398802995367, + "loss": 0.0781, + "step": 45126 + }, + { + "epoch": 0.39172403017334917, + "grad_norm": 0.291015625, + "learning_rate": 0.0012002103029654222, + "loss": 0.1309, + "step": 45127 + }, + { + "epoch": 0.39173271065355336, + "grad_norm": 0.314453125, + "learning_rate": 0.0012001807255217199, + "loss": 0.1147, + "step": 45128 + }, + { + "epoch": 0.3917413911337575, + "grad_norm": 0.11962890625, + "learning_rate": 0.001200151147968461, + "loss": 0.0977, + "step": 45129 + }, + { + "epoch": 0.3917500716139617, + "grad_norm": 0.12109375, + "learning_rate": 0.0012001215703056787, + "loss": 0.085, + "step": 45130 + }, + { + "epoch": 0.39175875209416583, + "grad_norm": 0.53515625, + "learning_rate": 0.0012000919925334048, + "loss": 0.1309, + "step": 45131 + }, + { + "epoch": 0.39176743257437, + "grad_norm": 0.890625, + "learning_rate": 0.001200062414651672, + "loss": 0.0991, + "step": 45132 + }, + { + "epoch": 0.39177611305457416, + "grad_norm": 0.279296875, + "learning_rate": 0.0012000328366605119, + "loss": 0.0825, + "step": 45133 + }, + { + "epoch": 0.39178479353477835, + "grad_norm": 0.76953125, + "learning_rate": 0.0012000032585599576, + "loss": 0.103, + "step": 45134 + }, + { + "epoch": 0.3917934740149825, + "grad_norm": 0.28515625, + "learning_rate": 0.0011999736803500417, + "loss": 0.1758, + "step": 45135 + }, + { + "epoch": 0.3918021544951867, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011999441020307956, + "loss": 0.1011, + "step": 45136 + }, + { + "epoch": 0.3918108349753908, + "grad_norm": 0.98046875, + "learning_rate": 0.0011999145236022523, + "loss": 0.1484, + "step": 45137 + }, + { + "epoch": 0.391819515455595, + "grad_norm": 0.142578125, + "learning_rate": 0.0011998849450644438, + "loss": 0.125, + "step": 45138 + }, + { + "epoch": 0.39182819593579915, + "grad_norm": 0.375, + "learning_rate": 0.0011998553664174028, + "loss": 0.085, + "step": 45139 + }, + { + "epoch": 0.39183687641600334, + "grad_norm": 0.12353515625, + "learning_rate": 0.0011998257876611614, + "loss": 0.1074, + "step": 45140 + }, + { + "epoch": 0.3918455568962075, + "grad_norm": 0.310546875, + "learning_rate": 0.0011997962087957523, + "loss": 0.0884, + "step": 45141 + }, + { + "epoch": 0.3918542373764117, + "grad_norm": 0.49609375, + "learning_rate": 0.001199766629821207, + "loss": 0.1162, + "step": 45142 + }, + { + "epoch": 0.3918629178566158, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011997370507375589, + "loss": 0.0786, + "step": 45143 + }, + { + "epoch": 0.39187159833682, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011997074715448398, + "loss": 0.1191, + "step": 45144 + }, + { + "epoch": 0.39188027881702414, + "grad_norm": 0.08837890625, + "learning_rate": 0.001199677892243082, + "loss": 0.1279, + "step": 45145 + }, + { + "epoch": 0.39188895929722833, + "grad_norm": 1.03125, + "learning_rate": 0.001199648312832318, + "loss": 0.1187, + "step": 45146 + }, + { + "epoch": 0.39189763977743247, + "grad_norm": 0.326171875, + "learning_rate": 0.0011996187333125799, + "loss": 0.1079, + "step": 45147 + }, + { + "epoch": 0.39190632025763666, + "grad_norm": 0.15234375, + "learning_rate": 0.0011995891536839003, + "loss": 0.0825, + "step": 45148 + }, + { + "epoch": 0.3919150007378408, + "grad_norm": 0.11572265625, + "learning_rate": 0.0011995595739463117, + "loss": 0.0806, + "step": 45149 + }, + { + "epoch": 0.391923681218045, + "grad_norm": 0.125, + "learning_rate": 0.001199529994099846, + "loss": 0.0811, + "step": 45150 + }, + { + "epoch": 0.39193236169824913, + "grad_norm": 0.59765625, + "learning_rate": 0.001199500414144536, + "loss": 0.0864, + "step": 45151 + }, + { + "epoch": 0.3919410421784533, + "grad_norm": 0.2001953125, + "learning_rate": 0.001199470834080414, + "loss": 0.1182, + "step": 45152 + }, + { + "epoch": 0.39194972265865746, + "grad_norm": 0.3828125, + "learning_rate": 0.001199441253907512, + "loss": 0.0859, + "step": 45153 + }, + { + "epoch": 0.39195840313886166, + "grad_norm": 0.7890625, + "learning_rate": 0.0011994116736258624, + "loss": 0.1201, + "step": 45154 + }, + { + "epoch": 0.3919670836190658, + "grad_norm": 0.2080078125, + "learning_rate": 0.0011993820932354978, + "loss": 0.0591, + "step": 45155 + }, + { + "epoch": 0.39197576409927, + "grad_norm": 0.306640625, + "learning_rate": 0.0011993525127364504, + "loss": 0.0664, + "step": 45156 + }, + { + "epoch": 0.3919844445794741, + "grad_norm": 0.234375, + "learning_rate": 0.0011993229321287527, + "loss": 0.1143, + "step": 45157 + }, + { + "epoch": 0.3919931250596783, + "grad_norm": 0.06591796875, + "learning_rate": 0.0011992933514124368, + "loss": 0.0579, + "step": 45158 + }, + { + "epoch": 0.39200180553988245, + "grad_norm": 0.09228515625, + "learning_rate": 0.0011992637705875353, + "loss": 0.0703, + "step": 45159 + }, + { + "epoch": 0.39201048602008665, + "grad_norm": 0.408203125, + "learning_rate": 0.0011992341896540804, + "loss": 0.0728, + "step": 45160 + }, + { + "epoch": 0.3920191665002908, + "grad_norm": 0.625, + "learning_rate": 0.0011992046086121044, + "loss": 0.1191, + "step": 45161 + }, + { + "epoch": 0.392027846980495, + "grad_norm": 0.2080078125, + "learning_rate": 0.0011991750274616398, + "loss": 0.1045, + "step": 45162 + }, + { + "epoch": 0.3920365274606991, + "grad_norm": 0.08056640625, + "learning_rate": 0.0011991454462027189, + "loss": 0.0864, + "step": 45163 + }, + { + "epoch": 0.3920452079409033, + "grad_norm": 0.1552734375, + "learning_rate": 0.0011991158648353738, + "loss": 0.0757, + "step": 45164 + }, + { + "epoch": 0.39205388842110744, + "grad_norm": 0.2275390625, + "learning_rate": 0.0011990862833596374, + "loss": 0.1455, + "step": 45165 + }, + { + "epoch": 0.39206256890131164, + "grad_norm": 0.345703125, + "learning_rate": 0.0011990567017755415, + "loss": 0.1011, + "step": 45166 + }, + { + "epoch": 0.3920712493815158, + "grad_norm": 0.291015625, + "learning_rate": 0.0011990271200831189, + "loss": 0.106, + "step": 45167 + }, + { + "epoch": 0.39207992986171997, + "grad_norm": 1.203125, + "learning_rate": 0.0011989975382824015, + "loss": 0.1436, + "step": 45168 + }, + { + "epoch": 0.3920886103419241, + "grad_norm": 0.07373046875, + "learning_rate": 0.001198967956373422, + "loss": 0.0747, + "step": 45169 + }, + { + "epoch": 0.3920972908221283, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011989383743562128, + "loss": 0.1094, + "step": 45170 + }, + { + "epoch": 0.39210597130233243, + "grad_norm": 0.365234375, + "learning_rate": 0.001198908792230806, + "loss": 0.0732, + "step": 45171 + }, + { + "epoch": 0.3921146517825366, + "grad_norm": 0.1845703125, + "learning_rate": 0.0011988792099972337, + "loss": 0.0869, + "step": 45172 + }, + { + "epoch": 0.39212333226274076, + "grad_norm": 0.337890625, + "learning_rate": 0.001198849627655529, + "loss": 0.1113, + "step": 45173 + }, + { + "epoch": 0.39213201274294496, + "grad_norm": 1.09375, + "learning_rate": 0.0011988200452057237, + "loss": 0.1123, + "step": 45174 + }, + { + "epoch": 0.3921406932231491, + "grad_norm": 0.259765625, + "learning_rate": 0.00119879046264785, + "loss": 0.1191, + "step": 45175 + }, + { + "epoch": 0.3921493737033533, + "grad_norm": 0.1591796875, + "learning_rate": 0.0011987608799819407, + "loss": 0.0742, + "step": 45176 + }, + { + "epoch": 0.3921580541835574, + "grad_norm": 0.2197265625, + "learning_rate": 0.0011987312972080281, + "loss": 0.0728, + "step": 45177 + }, + { + "epoch": 0.3921667346637616, + "grad_norm": 0.15625, + "learning_rate": 0.0011987017143261444, + "loss": 0.084, + "step": 45178 + }, + { + "epoch": 0.39217541514396576, + "grad_norm": 0.0927734375, + "learning_rate": 0.001198672131336322, + "loss": 0.1055, + "step": 45179 + }, + { + "epoch": 0.39218409562416995, + "grad_norm": 0.375, + "learning_rate": 0.0011986425482385934, + "loss": 0.0942, + "step": 45180 + }, + { + "epoch": 0.3921927761043741, + "grad_norm": 0.60546875, + "learning_rate": 0.0011986129650329905, + "loss": 0.1201, + "step": 45181 + }, + { + "epoch": 0.3922014565845783, + "grad_norm": 0.091796875, + "learning_rate": 0.001198583381719546, + "loss": 0.1094, + "step": 45182 + }, + { + "epoch": 0.3922101370647824, + "grad_norm": 0.21875, + "learning_rate": 0.0011985537982982927, + "loss": 0.166, + "step": 45183 + }, + { + "epoch": 0.3922188175449866, + "grad_norm": 0.1484375, + "learning_rate": 0.0011985242147692617, + "loss": 0.0796, + "step": 45184 + }, + { + "epoch": 0.39222749802519075, + "grad_norm": 0.546875, + "learning_rate": 0.001198494631132486, + "loss": 0.0825, + "step": 45185 + }, + { + "epoch": 0.39223617850539494, + "grad_norm": 0.74609375, + "learning_rate": 0.0011984650473879986, + "loss": 0.1094, + "step": 45186 + }, + { + "epoch": 0.3922448589855991, + "grad_norm": 0.30078125, + "learning_rate": 0.0011984354635358311, + "loss": 0.0791, + "step": 45187 + }, + { + "epoch": 0.39225353946580327, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011984058795760162, + "loss": 0.0811, + "step": 45188 + }, + { + "epoch": 0.3922622199460074, + "grad_norm": 0.388671875, + "learning_rate": 0.001198376295508586, + "loss": 0.0815, + "step": 45189 + }, + { + "epoch": 0.3922709004262116, + "grad_norm": 0.3828125, + "learning_rate": 0.0011983467113335729, + "loss": 0.0967, + "step": 45190 + }, + { + "epoch": 0.39227958090641574, + "grad_norm": 0.53515625, + "learning_rate": 0.0011983171270510096, + "loss": 0.0801, + "step": 45191 + }, + { + "epoch": 0.39228826138661993, + "grad_norm": 0.30859375, + "learning_rate": 0.001198287542660928, + "loss": 0.0962, + "step": 45192 + }, + { + "epoch": 0.39229694186682407, + "grad_norm": 0.54296875, + "learning_rate": 0.0011982579581633604, + "loss": 0.1025, + "step": 45193 + }, + { + "epoch": 0.39230562234702826, + "grad_norm": 0.294921875, + "learning_rate": 0.0011982283735583394, + "loss": 0.1738, + "step": 45194 + }, + { + "epoch": 0.3923143028272324, + "grad_norm": 0.515625, + "learning_rate": 0.0011981987888458974, + "loss": 0.0874, + "step": 45195 + }, + { + "epoch": 0.3923229833074366, + "grad_norm": 0.486328125, + "learning_rate": 0.0011981692040260669, + "loss": 0.1426, + "step": 45196 + }, + { + "epoch": 0.3923316637876407, + "grad_norm": 0.11328125, + "learning_rate": 0.00119813961909888, + "loss": 0.1035, + "step": 45197 + }, + { + "epoch": 0.3923403442678449, + "grad_norm": 0.2333984375, + "learning_rate": 0.001198110034064369, + "loss": 0.1001, + "step": 45198 + }, + { + "epoch": 0.39234902474804906, + "grad_norm": 0.365234375, + "learning_rate": 0.0011980804489225662, + "loss": 0.1445, + "step": 45199 + }, + { + "epoch": 0.39235770522825325, + "grad_norm": 0.57421875, + "learning_rate": 0.0011980508636735044, + "loss": 0.1055, + "step": 45200 + }, + { + "epoch": 0.3923663857084574, + "grad_norm": 0.27734375, + "learning_rate": 0.0011980212783172153, + "loss": 0.0918, + "step": 45201 + }, + { + "epoch": 0.3923750661886616, + "grad_norm": 0.3984375, + "learning_rate": 0.0011979916928537317, + "loss": 0.084, + "step": 45202 + }, + { + "epoch": 0.3923837466688657, + "grad_norm": 0.11279296875, + "learning_rate": 0.0011979621072830862, + "loss": 0.0776, + "step": 45203 + }, + { + "epoch": 0.3923924271490699, + "grad_norm": 0.447265625, + "learning_rate": 0.0011979325216053102, + "loss": 0.1201, + "step": 45204 + }, + { + "epoch": 0.39240110762927405, + "grad_norm": 0.078125, + "learning_rate": 0.0011979029358204372, + "loss": 0.085, + "step": 45205 + }, + { + "epoch": 0.39240978810947824, + "grad_norm": 0.390625, + "learning_rate": 0.0011978733499284986, + "loss": 0.0762, + "step": 45206 + }, + { + "epoch": 0.3924184685896824, + "grad_norm": 0.65234375, + "learning_rate": 0.0011978437639295274, + "loss": 0.1543, + "step": 45207 + }, + { + "epoch": 0.39242714906988657, + "grad_norm": 0.11962890625, + "learning_rate": 0.0011978141778235557, + "loss": 0.1064, + "step": 45208 + }, + { + "epoch": 0.3924358295500907, + "grad_norm": 0.71484375, + "learning_rate": 0.001197784591610616, + "loss": 0.1006, + "step": 45209 + }, + { + "epoch": 0.3924445100302949, + "grad_norm": 0.34765625, + "learning_rate": 0.0011977550052907407, + "loss": 0.0864, + "step": 45210 + }, + { + "epoch": 0.39245319051049904, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011977254188639615, + "loss": 0.0825, + "step": 45211 + }, + { + "epoch": 0.39246187099070323, + "grad_norm": 0.96484375, + "learning_rate": 0.0011976958323303118, + "loss": 0.0879, + "step": 45212 + }, + { + "epoch": 0.39247055147090737, + "grad_norm": 0.365234375, + "learning_rate": 0.0011976662456898227, + "loss": 0.105, + "step": 45213 + }, + { + "epoch": 0.39247923195111156, + "grad_norm": 0.1162109375, + "learning_rate": 0.001197636658942528, + "loss": 0.0737, + "step": 45214 + }, + { + "epoch": 0.3924879124313157, + "grad_norm": 0.47265625, + "learning_rate": 0.0011976070720884587, + "loss": 0.0938, + "step": 45215 + }, + { + "epoch": 0.3924965929115199, + "grad_norm": 0.11669921875, + "learning_rate": 0.001197577485127648, + "loss": 0.0947, + "step": 45216 + }, + { + "epoch": 0.39250527339172403, + "grad_norm": 0.09765625, + "learning_rate": 0.001197547898060128, + "loss": 0.0874, + "step": 45217 + }, + { + "epoch": 0.39251395387192817, + "grad_norm": 0.5546875, + "learning_rate": 0.0011975183108859314, + "loss": 0.1445, + "step": 45218 + }, + { + "epoch": 0.39252263435213236, + "grad_norm": 0.0908203125, + "learning_rate": 0.00119748872360509, + "loss": 0.0918, + "step": 45219 + }, + { + "epoch": 0.3925313148323365, + "grad_norm": 0.4921875, + "learning_rate": 0.0011974591362176365, + "loss": 0.1973, + "step": 45220 + }, + { + "epoch": 0.3925399953125407, + "grad_norm": 0.87109375, + "learning_rate": 0.001197429548723603, + "loss": 0.1104, + "step": 45221 + }, + { + "epoch": 0.39254867579274483, + "grad_norm": 0.283203125, + "learning_rate": 0.001197399961123022, + "loss": 0.0962, + "step": 45222 + }, + { + "epoch": 0.392557356272949, + "grad_norm": 0.2158203125, + "learning_rate": 0.001197370373415926, + "loss": 0.1245, + "step": 45223 + }, + { + "epoch": 0.39256603675315316, + "grad_norm": 0.205078125, + "learning_rate": 0.001197340785602347, + "loss": 0.0815, + "step": 45224 + }, + { + "epoch": 0.39257471723335735, + "grad_norm": 0.44921875, + "learning_rate": 0.0011973111976823177, + "loss": 0.0879, + "step": 45225 + }, + { + "epoch": 0.3925833977135615, + "grad_norm": 0.208984375, + "learning_rate": 0.0011972816096558706, + "loss": 0.0698, + "step": 45226 + }, + { + "epoch": 0.3925920781937657, + "grad_norm": 0.46875, + "learning_rate": 0.0011972520215230373, + "loss": 0.1162, + "step": 45227 + }, + { + "epoch": 0.3926007586739698, + "grad_norm": 0.98828125, + "learning_rate": 0.0011972224332838508, + "loss": 0.105, + "step": 45228 + }, + { + "epoch": 0.392609439154174, + "grad_norm": 0.703125, + "learning_rate": 0.0011971928449383436, + "loss": 0.1191, + "step": 45229 + }, + { + "epoch": 0.39261811963437815, + "grad_norm": 0.1044921875, + "learning_rate": 0.0011971632564865477, + "loss": 0.1025, + "step": 45230 + }, + { + "epoch": 0.39262680011458234, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011971336679284955, + "loss": 0.0732, + "step": 45231 + }, + { + "epoch": 0.3926354805947865, + "grad_norm": 0.095703125, + "learning_rate": 0.0011971040792642192, + "loss": 0.0903, + "step": 45232 + }, + { + "epoch": 0.39264416107499067, + "grad_norm": 0.275390625, + "learning_rate": 0.001197074490493751, + "loss": 0.1035, + "step": 45233 + }, + { + "epoch": 0.3926528415551948, + "grad_norm": 0.6171875, + "learning_rate": 0.0011970449016171242, + "loss": 0.1133, + "step": 45234 + }, + { + "epoch": 0.392661522035399, + "grad_norm": 0.30078125, + "learning_rate": 0.0011970153126343705, + "loss": 0.1133, + "step": 45235 + }, + { + "epoch": 0.39267020251560314, + "grad_norm": 0.4921875, + "learning_rate": 0.0011969857235455221, + "loss": 0.0938, + "step": 45236 + }, + { + "epoch": 0.39267888299580733, + "grad_norm": 0.39453125, + "learning_rate": 0.0011969561343506117, + "loss": 0.0938, + "step": 45237 + }, + { + "epoch": 0.39268756347601147, + "grad_norm": 0.828125, + "learning_rate": 0.0011969265450496713, + "loss": 0.1436, + "step": 45238 + }, + { + "epoch": 0.39269624395621566, + "grad_norm": 0.6640625, + "learning_rate": 0.0011968969556427338, + "loss": 0.25, + "step": 45239 + }, + { + "epoch": 0.3927049244364198, + "grad_norm": 0.130859375, + "learning_rate": 0.0011968673661298312, + "loss": 0.0918, + "step": 45240 + }, + { + "epoch": 0.392713604916624, + "grad_norm": 0.201171875, + "learning_rate": 0.0011968377765109958, + "loss": 0.1152, + "step": 45241 + }, + { + "epoch": 0.39272228539682813, + "grad_norm": 0.34765625, + "learning_rate": 0.0011968081867862602, + "loss": 0.0991, + "step": 45242 + }, + { + "epoch": 0.3927309658770323, + "grad_norm": 0.09765625, + "learning_rate": 0.0011967785969556565, + "loss": 0.0801, + "step": 45243 + }, + { + "epoch": 0.39273964635723646, + "grad_norm": 0.146484375, + "learning_rate": 0.001196749007019217, + "loss": 0.1094, + "step": 45244 + }, + { + "epoch": 0.39274832683744065, + "grad_norm": 0.73828125, + "learning_rate": 0.0011967194169769744, + "loss": 0.0908, + "step": 45245 + }, + { + "epoch": 0.3927570073176448, + "grad_norm": 0.76171875, + "learning_rate": 0.001196689826828961, + "loss": 0.1016, + "step": 45246 + }, + { + "epoch": 0.392765687797849, + "grad_norm": 0.08251953125, + "learning_rate": 0.0011966602365752089, + "loss": 0.0903, + "step": 45247 + }, + { + "epoch": 0.3927743682780531, + "grad_norm": 0.43359375, + "learning_rate": 0.0011966306462157508, + "loss": 0.1104, + "step": 45248 + }, + { + "epoch": 0.3927830487582573, + "grad_norm": 0.240234375, + "learning_rate": 0.0011966010557506188, + "loss": 0.1021, + "step": 45249 + }, + { + "epoch": 0.39279172923846145, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011965714651798452, + "loss": 0.1006, + "step": 45250 + }, + { + "epoch": 0.39280040971866564, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011965418745034627, + "loss": 0.1016, + "step": 45251 + }, + { + "epoch": 0.3928090901988698, + "grad_norm": 0.35546875, + "learning_rate": 0.0011965122837215032, + "loss": 0.0928, + "step": 45252 + }, + { + "epoch": 0.392817770679074, + "grad_norm": 0.10107421875, + "learning_rate": 0.0011964826928339994, + "loss": 0.1289, + "step": 45253 + }, + { + "epoch": 0.3928264511592781, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011964531018409837, + "loss": 0.0913, + "step": 45254 + }, + { + "epoch": 0.3928351316394823, + "grad_norm": 0.765625, + "learning_rate": 0.0011964235107424883, + "loss": 0.1514, + "step": 45255 + }, + { + "epoch": 0.39284381211968644, + "grad_norm": 0.546875, + "learning_rate": 0.0011963939195385456, + "loss": 0.1045, + "step": 45256 + }, + { + "epoch": 0.39285249259989063, + "grad_norm": 0.08056640625, + "learning_rate": 0.001196364328229188, + "loss": 0.1211, + "step": 45257 + }, + { + "epoch": 0.39286117308009477, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011963347368144476, + "loss": 0.085, + "step": 45258 + }, + { + "epoch": 0.39286985356029896, + "grad_norm": 0.263671875, + "learning_rate": 0.0011963051452943572, + "loss": 0.0962, + "step": 45259 + }, + { + "epoch": 0.3928785340405031, + "grad_norm": 0.166015625, + "learning_rate": 0.001196275553668949, + "loss": 0.1016, + "step": 45260 + }, + { + "epoch": 0.3928872145207073, + "grad_norm": 0.21875, + "learning_rate": 0.001196245961938255, + "loss": 0.0859, + "step": 45261 + }, + { + "epoch": 0.39289589500091143, + "grad_norm": 0.306640625, + "learning_rate": 0.0011962163701023081, + "loss": 0.0923, + "step": 45262 + }, + { + "epoch": 0.3929045754811156, + "grad_norm": 0.80859375, + "learning_rate": 0.0011961867781611405, + "loss": 0.0923, + "step": 45263 + }, + { + "epoch": 0.39291325596131976, + "grad_norm": 0.283203125, + "learning_rate": 0.0011961571861147842, + "loss": 0.0869, + "step": 45264 + }, + { + "epoch": 0.39292193644152396, + "grad_norm": 0.212890625, + "learning_rate": 0.0011961275939632722, + "loss": 0.0894, + "step": 45265 + }, + { + "epoch": 0.3929306169217281, + "grad_norm": 0.3359375, + "learning_rate": 0.0011960980017066364, + "loss": 0.1172, + "step": 45266 + }, + { + "epoch": 0.3929392974019323, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011960684093449089, + "loss": 0.0967, + "step": 45267 + }, + { + "epoch": 0.3929479778821364, + "grad_norm": 0.5390625, + "learning_rate": 0.0011960388168781228, + "loss": 0.1514, + "step": 45268 + }, + { + "epoch": 0.3929566583623406, + "grad_norm": 0.3984375, + "learning_rate": 0.0011960092243063101, + "loss": 0.0977, + "step": 45269 + }, + { + "epoch": 0.39296533884254475, + "grad_norm": 0.21484375, + "learning_rate": 0.001195979631629503, + "loss": 0.1021, + "step": 45270 + }, + { + "epoch": 0.39297401932274895, + "grad_norm": 0.474609375, + "learning_rate": 0.001195950038847734, + "loss": 0.0991, + "step": 45271 + }, + { + "epoch": 0.3929826998029531, + "grad_norm": 0.103515625, + "learning_rate": 0.0011959204459610355, + "loss": 0.1001, + "step": 45272 + }, + { + "epoch": 0.3929913802831573, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011958908529694399, + "loss": 0.1289, + "step": 45273 + }, + { + "epoch": 0.3930000607633614, + "grad_norm": 0.67578125, + "learning_rate": 0.0011958612598729794, + "loss": 0.1245, + "step": 45274 + }, + { + "epoch": 0.3930087412435656, + "grad_norm": 0.21875, + "learning_rate": 0.0011958316666716869, + "loss": 0.0791, + "step": 45275 + }, + { + "epoch": 0.39301742172376974, + "grad_norm": 0.79296875, + "learning_rate": 0.001195802073365594, + "loss": 0.1309, + "step": 45276 + }, + { + "epoch": 0.39302610220397394, + "grad_norm": 0.09716796875, + "learning_rate": 0.0011957724799547332, + "loss": 0.0752, + "step": 45277 + }, + { + "epoch": 0.3930347826841781, + "grad_norm": 0.181640625, + "learning_rate": 0.0011957428864391372, + "loss": 0.127, + "step": 45278 + }, + { + "epoch": 0.39304346316438227, + "grad_norm": 0.154296875, + "learning_rate": 0.0011957132928188383, + "loss": 0.0664, + "step": 45279 + }, + { + "epoch": 0.3930521436445864, + "grad_norm": 0.28125, + "learning_rate": 0.0011956836990938689, + "loss": 0.0986, + "step": 45280 + }, + { + "epoch": 0.3930608241247906, + "grad_norm": 0.310546875, + "learning_rate": 0.0011956541052642608, + "loss": 0.1045, + "step": 45281 + }, + { + "epoch": 0.39306950460499473, + "grad_norm": 0.0849609375, + "learning_rate": 0.001195624511330047, + "loss": 0.1016, + "step": 45282 + }, + { + "epoch": 0.39307818508519893, + "grad_norm": 0.263671875, + "learning_rate": 0.00119559491729126, + "loss": 0.1079, + "step": 45283 + }, + { + "epoch": 0.39308686556540307, + "grad_norm": 0.357421875, + "learning_rate": 0.0011955653231479316, + "loss": 0.0674, + "step": 45284 + }, + { + "epoch": 0.39309554604560726, + "grad_norm": 1.21875, + "learning_rate": 0.0011955357289000943, + "loss": 0.1416, + "step": 45285 + }, + { + "epoch": 0.3931042265258114, + "grad_norm": 0.126953125, + "learning_rate": 0.0011955061345477807, + "loss": 0.0972, + "step": 45286 + }, + { + "epoch": 0.3931129070060156, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011954765400910233, + "loss": 0.0874, + "step": 45287 + }, + { + "epoch": 0.3931215874862197, + "grad_norm": 0.2890625, + "learning_rate": 0.0011954469455298536, + "loss": 0.125, + "step": 45288 + }, + { + "epoch": 0.3931302679664239, + "grad_norm": 0.291015625, + "learning_rate": 0.001195417350864305, + "loss": 0.103, + "step": 45289 + }, + { + "epoch": 0.39313894844662806, + "grad_norm": 0.08935546875, + "learning_rate": 0.0011953877560944089, + "loss": 0.1064, + "step": 45290 + }, + { + "epoch": 0.39314762892683225, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011953581612201987, + "loss": 0.1104, + "step": 45291 + }, + { + "epoch": 0.3931563094070364, + "grad_norm": 0.14453125, + "learning_rate": 0.0011953285662417061, + "loss": 0.0898, + "step": 45292 + }, + { + "epoch": 0.3931649898872406, + "grad_norm": 0.2275390625, + "learning_rate": 0.0011952989711589634, + "loss": 0.0972, + "step": 45293 + }, + { + "epoch": 0.3931736703674447, + "grad_norm": 0.5625, + "learning_rate": 0.0011952693759720034, + "loss": 0.1133, + "step": 45294 + }, + { + "epoch": 0.3931823508476489, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011952397806808584, + "loss": 0.0767, + "step": 45295 + }, + { + "epoch": 0.39319103132785305, + "grad_norm": 0.4296875, + "learning_rate": 0.0011952101852855603, + "loss": 0.0752, + "step": 45296 + }, + { + "epoch": 0.39319971180805724, + "grad_norm": 0.173828125, + "learning_rate": 0.001195180589786142, + "loss": 0.1094, + "step": 45297 + }, + { + "epoch": 0.3932083922882614, + "grad_norm": 0.279296875, + "learning_rate": 0.0011951509941826356, + "loss": 0.1113, + "step": 45298 + }, + { + "epoch": 0.39321707276846557, + "grad_norm": 0.197265625, + "learning_rate": 0.0011951213984750734, + "loss": 0.1279, + "step": 45299 + }, + { + "epoch": 0.3932257532486697, + "grad_norm": 0.42578125, + "learning_rate": 0.0011950918026634879, + "loss": 0.1025, + "step": 45300 + }, + { + "epoch": 0.3932344337288739, + "grad_norm": 0.255859375, + "learning_rate": 0.0011950622067479113, + "loss": 0.123, + "step": 45301 + }, + { + "epoch": 0.39324311420907804, + "grad_norm": 0.53125, + "learning_rate": 0.0011950326107283762, + "loss": 0.1143, + "step": 45302 + }, + { + "epoch": 0.39325179468928223, + "grad_norm": 0.1865234375, + "learning_rate": 0.001195003014604915, + "loss": 0.0693, + "step": 45303 + }, + { + "epoch": 0.39326047516948637, + "grad_norm": 0.1484375, + "learning_rate": 0.0011949734183775598, + "loss": 0.1338, + "step": 45304 + }, + { + "epoch": 0.39326915564969056, + "grad_norm": 0.2421875, + "learning_rate": 0.001194943822046343, + "loss": 0.1006, + "step": 45305 + }, + { + "epoch": 0.3932778361298947, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011949142256112973, + "loss": 0.0771, + "step": 45306 + }, + { + "epoch": 0.3932865166100989, + "grad_norm": 0.3203125, + "learning_rate": 0.0011948846290724547, + "loss": 0.0957, + "step": 45307 + }, + { + "epoch": 0.39329519709030303, + "grad_norm": 0.369140625, + "learning_rate": 0.0011948550324298477, + "loss": 0.1064, + "step": 45308 + }, + { + "epoch": 0.3933038775705072, + "grad_norm": 0.3203125, + "learning_rate": 0.0011948254356835086, + "loss": 0.0854, + "step": 45309 + }, + { + "epoch": 0.39331255805071136, + "grad_norm": 0.31640625, + "learning_rate": 0.00119479583883347, + "loss": 0.1553, + "step": 45310 + }, + { + "epoch": 0.39332123853091555, + "grad_norm": 0.53125, + "learning_rate": 0.0011947662418797637, + "loss": 0.1191, + "step": 45311 + }, + { + "epoch": 0.3933299190111197, + "grad_norm": 0.421875, + "learning_rate": 0.001194736644822423, + "loss": 0.0986, + "step": 45312 + }, + { + "epoch": 0.3933385994913239, + "grad_norm": 0.6171875, + "learning_rate": 0.0011947070476614797, + "loss": 0.0913, + "step": 45313 + }, + { + "epoch": 0.393347279971528, + "grad_norm": 0.0986328125, + "learning_rate": 0.001194677450396966, + "loss": 0.1133, + "step": 45314 + }, + { + "epoch": 0.3933559604517322, + "grad_norm": 0.244140625, + "learning_rate": 0.0011946478530289143, + "loss": 0.104, + "step": 45315 + }, + { + "epoch": 0.39336464093193635, + "grad_norm": 0.24609375, + "learning_rate": 0.0011946182555573573, + "loss": 0.083, + "step": 45316 + }, + { + "epoch": 0.39337332141214054, + "grad_norm": 0.1806640625, + "learning_rate": 0.0011945886579823272, + "loss": 0.0874, + "step": 45317 + }, + { + "epoch": 0.3933820018923447, + "grad_norm": 0.431640625, + "learning_rate": 0.001194559060303856, + "loss": 0.0767, + "step": 45318 + }, + { + "epoch": 0.39339068237254887, + "grad_norm": 0.390625, + "learning_rate": 0.0011945294625219768, + "loss": 0.167, + "step": 45319 + }, + { + "epoch": 0.393399362852753, + "grad_norm": 0.314453125, + "learning_rate": 0.0011944998646367215, + "loss": 0.0708, + "step": 45320 + }, + { + "epoch": 0.3934080433329572, + "grad_norm": 0.384765625, + "learning_rate": 0.0011944702666481228, + "loss": 0.085, + "step": 45321 + }, + { + "epoch": 0.39341672381316134, + "grad_norm": 0.154296875, + "learning_rate": 0.0011944406685562125, + "loss": 0.1152, + "step": 45322 + }, + { + "epoch": 0.39342540429336553, + "grad_norm": 0.166015625, + "learning_rate": 0.0011944110703610237, + "loss": 0.1279, + "step": 45323 + }, + { + "epoch": 0.39343408477356967, + "grad_norm": 0.2001953125, + "learning_rate": 0.001194381472062588, + "loss": 0.0903, + "step": 45324 + }, + { + "epoch": 0.39344276525377386, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011943518736609385, + "loss": 0.1113, + "step": 45325 + }, + { + "epoch": 0.393451445733978, + "grad_norm": 0.236328125, + "learning_rate": 0.001194322275156107, + "loss": 0.0991, + "step": 45326 + }, + { + "epoch": 0.3934601262141822, + "grad_norm": 0.30859375, + "learning_rate": 0.001194292676548126, + "loss": 0.0972, + "step": 45327 + }, + { + "epoch": 0.39346880669438633, + "grad_norm": 0.2099609375, + "learning_rate": 0.0011942630778370277, + "loss": 0.1079, + "step": 45328 + }, + { + "epoch": 0.3934774871745905, + "grad_norm": 0.1982421875, + "learning_rate": 0.0011942334790228453, + "loss": 0.085, + "step": 45329 + }, + { + "epoch": 0.39348616765479466, + "grad_norm": 0.33203125, + "learning_rate": 0.0011942038801056104, + "loss": 0.1113, + "step": 45330 + }, + { + "epoch": 0.39349484813499885, + "grad_norm": 0.1533203125, + "learning_rate": 0.0011941742810853556, + "loss": 0.0952, + "step": 45331 + }, + { + "epoch": 0.393503528615203, + "grad_norm": 0.1650390625, + "learning_rate": 0.0011941446819621127, + "loss": 0.125, + "step": 45332 + }, + { + "epoch": 0.3935122090954072, + "grad_norm": 0.53125, + "learning_rate": 0.001194115082735915, + "loss": 0.1064, + "step": 45333 + }, + { + "epoch": 0.3935208895756113, + "grad_norm": 0.1259765625, + "learning_rate": 0.0011940854834067947, + "loss": 0.1045, + "step": 45334 + }, + { + "epoch": 0.3935295700558155, + "grad_norm": 0.25390625, + "learning_rate": 0.0011940558839747836, + "loss": 0.0996, + "step": 45335 + }, + { + "epoch": 0.39353825053601965, + "grad_norm": 0.37109375, + "learning_rate": 0.0011940262844399143, + "loss": 0.1177, + "step": 45336 + }, + { + "epoch": 0.39354693101622384, + "grad_norm": 0.56640625, + "learning_rate": 0.0011939966848022196, + "loss": 0.0762, + "step": 45337 + }, + { + "epoch": 0.393555611496428, + "grad_norm": 0.140625, + "learning_rate": 0.0011939670850617313, + "loss": 0.0801, + "step": 45338 + }, + { + "epoch": 0.3935642919766322, + "grad_norm": 0.75390625, + "learning_rate": 0.001193937485218482, + "loss": 0.0977, + "step": 45339 + }, + { + "epoch": 0.3935729724568363, + "grad_norm": 0.201171875, + "learning_rate": 0.001193907885272504, + "loss": 0.1079, + "step": 45340 + }, + { + "epoch": 0.3935816529370405, + "grad_norm": 0.71875, + "learning_rate": 0.0011938782852238302, + "loss": 0.0898, + "step": 45341 + }, + { + "epoch": 0.39359033341724464, + "grad_norm": 0.087890625, + "learning_rate": 0.0011938486850724922, + "loss": 0.0923, + "step": 45342 + }, + { + "epoch": 0.3935990138974488, + "grad_norm": 0.5625, + "learning_rate": 0.001193819084818523, + "loss": 0.127, + "step": 45343 + }, + { + "epoch": 0.393607694377653, + "grad_norm": 0.333984375, + "learning_rate": 0.001193789484461954, + "loss": 0.084, + "step": 45344 + }, + { + "epoch": 0.3936163748578571, + "grad_norm": 0.185546875, + "learning_rate": 0.0011937598840028186, + "loss": 0.1162, + "step": 45345 + }, + { + "epoch": 0.3936250553380613, + "grad_norm": 0.345703125, + "learning_rate": 0.001193730283441149, + "loss": 0.1021, + "step": 45346 + }, + { + "epoch": 0.39363373581826544, + "grad_norm": 0.2158203125, + "learning_rate": 0.0011937006827769771, + "loss": 0.127, + "step": 45347 + }, + { + "epoch": 0.39364241629846963, + "grad_norm": 0.197265625, + "learning_rate": 0.0011936710820103357, + "loss": 0.1309, + "step": 45348 + }, + { + "epoch": 0.39365109677867377, + "grad_norm": 0.8046875, + "learning_rate": 0.0011936414811412567, + "loss": 0.1289, + "step": 45349 + }, + { + "epoch": 0.39365977725887796, + "grad_norm": 0.37890625, + "learning_rate": 0.0011936118801697728, + "loss": 0.124, + "step": 45350 + }, + { + "epoch": 0.3936684577390821, + "grad_norm": 0.349609375, + "learning_rate": 0.0011935822790959168, + "loss": 0.0713, + "step": 45351 + }, + { + "epoch": 0.3936771382192863, + "grad_norm": 0.2890625, + "learning_rate": 0.0011935526779197204, + "loss": 0.1221, + "step": 45352 + }, + { + "epoch": 0.39368581869949043, + "grad_norm": 0.099609375, + "learning_rate": 0.0011935230766412159, + "loss": 0.0894, + "step": 45353 + }, + { + "epoch": 0.3936944991796946, + "grad_norm": 0.43359375, + "learning_rate": 0.0011934934752604366, + "loss": 0.1016, + "step": 45354 + }, + { + "epoch": 0.39370317965989876, + "grad_norm": 0.1982421875, + "learning_rate": 0.001193463873777414, + "loss": 0.1533, + "step": 45355 + }, + { + "epoch": 0.39371186014010295, + "grad_norm": 0.357421875, + "learning_rate": 0.0011934342721921804, + "loss": 0.0889, + "step": 45356 + }, + { + "epoch": 0.3937205406203071, + "grad_norm": 0.240234375, + "learning_rate": 0.0011934046705047686, + "loss": 0.0962, + "step": 45357 + }, + { + "epoch": 0.3937292211005113, + "grad_norm": 0.4296875, + "learning_rate": 0.0011933750687152108, + "loss": 0.1226, + "step": 45358 + }, + { + "epoch": 0.3937379015807154, + "grad_norm": 0.126953125, + "learning_rate": 0.0011933454668235396, + "loss": 0.1055, + "step": 45359 + }, + { + "epoch": 0.3937465820609196, + "grad_norm": 0.181640625, + "learning_rate": 0.0011933158648297873, + "loss": 0.1689, + "step": 45360 + }, + { + "epoch": 0.39375526254112375, + "grad_norm": 0.27734375, + "learning_rate": 0.001193286262733986, + "loss": 0.1328, + "step": 45361 + }, + { + "epoch": 0.39376394302132794, + "grad_norm": 0.474609375, + "learning_rate": 0.0011932566605361683, + "loss": 0.0825, + "step": 45362 + }, + { + "epoch": 0.3937726235015321, + "grad_norm": 0.26171875, + "learning_rate": 0.0011932270582363667, + "loss": 0.0864, + "step": 45363 + }, + { + "epoch": 0.3937813039817363, + "grad_norm": 0.318359375, + "learning_rate": 0.0011931974558346133, + "loss": 0.0898, + "step": 45364 + }, + { + "epoch": 0.3937899844619404, + "grad_norm": 0.271484375, + "learning_rate": 0.0011931678533309404, + "loss": 0.0781, + "step": 45365 + }, + { + "epoch": 0.3937986649421446, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011931382507253804, + "loss": 0.1162, + "step": 45366 + }, + { + "epoch": 0.39380734542234874, + "grad_norm": 0.259765625, + "learning_rate": 0.0011931086480179661, + "loss": 0.083, + "step": 45367 + }, + { + "epoch": 0.39381602590255294, + "grad_norm": 0.1640625, + "learning_rate": 0.0011930790452087292, + "loss": 0.0859, + "step": 45368 + }, + { + "epoch": 0.3938247063827571, + "grad_norm": 0.1728515625, + "learning_rate": 0.0011930494422977029, + "loss": 0.1064, + "step": 45369 + }, + { + "epoch": 0.39383338686296127, + "grad_norm": 0.2890625, + "learning_rate": 0.0011930198392849191, + "loss": 0.0967, + "step": 45370 + }, + { + "epoch": 0.3938420673431654, + "grad_norm": 0.2734375, + "learning_rate": 0.0011929902361704097, + "loss": 0.0654, + "step": 45371 + }, + { + "epoch": 0.3938507478233696, + "grad_norm": 0.69140625, + "learning_rate": 0.0011929606329542082, + "loss": 0.082, + "step": 45372 + }, + { + "epoch": 0.39385942830357373, + "grad_norm": 0.48828125, + "learning_rate": 0.001192931029636346, + "loss": 0.1768, + "step": 45373 + }, + { + "epoch": 0.3938681087837779, + "grad_norm": 0.984375, + "learning_rate": 0.0011929014262168562, + "loss": 0.0908, + "step": 45374 + }, + { + "epoch": 0.39387678926398206, + "grad_norm": 0.74609375, + "learning_rate": 0.0011928718226957703, + "loss": 0.1426, + "step": 45375 + }, + { + "epoch": 0.39388546974418626, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011928422190731214, + "loss": 0.1035, + "step": 45376 + }, + { + "epoch": 0.3938941502243904, + "grad_norm": 0.451171875, + "learning_rate": 0.0011928126153489416, + "loss": 0.1289, + "step": 45377 + }, + { + "epoch": 0.3939028307045946, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011927830115232635, + "loss": 0.082, + "step": 45378 + }, + { + "epoch": 0.3939115111847987, + "grad_norm": 0.1484375, + "learning_rate": 0.001192753407596119, + "loss": 0.0894, + "step": 45379 + }, + { + "epoch": 0.3939201916650029, + "grad_norm": 0.515625, + "learning_rate": 0.0011927238035675408, + "loss": 0.1328, + "step": 45380 + }, + { + "epoch": 0.39392887214520705, + "grad_norm": 1.046875, + "learning_rate": 0.0011926941994375613, + "loss": 0.1113, + "step": 45381 + }, + { + "epoch": 0.39393755262541125, + "grad_norm": 0.310546875, + "learning_rate": 0.001192664595206213, + "loss": 0.0967, + "step": 45382 + }, + { + "epoch": 0.3939462331056154, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011926349908735281, + "loss": 0.0645, + "step": 45383 + }, + { + "epoch": 0.3939549135858196, + "grad_norm": 0.353515625, + "learning_rate": 0.0011926053864395386, + "loss": 0.0957, + "step": 45384 + }, + { + "epoch": 0.3939635940660237, + "grad_norm": 0.408203125, + "learning_rate": 0.0011925757819042774, + "loss": 0.1143, + "step": 45385 + }, + { + "epoch": 0.3939722745462279, + "grad_norm": 0.490234375, + "learning_rate": 0.0011925461772677767, + "loss": 0.127, + "step": 45386 + }, + { + "epoch": 0.39398095502643204, + "grad_norm": 0.087890625, + "learning_rate": 0.0011925165725300692, + "loss": 0.1084, + "step": 45387 + }, + { + "epoch": 0.39398963550663624, + "grad_norm": 0.890625, + "learning_rate": 0.0011924869676911865, + "loss": 0.1167, + "step": 45388 + }, + { + "epoch": 0.3939983159868404, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011924573627511616, + "loss": 0.0791, + "step": 45389 + }, + { + "epoch": 0.39400699646704457, + "grad_norm": 0.1318359375, + "learning_rate": 0.001192427757710027, + "loss": 0.1094, + "step": 45390 + }, + { + "epoch": 0.3940156769472487, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011923981525678146, + "loss": 0.1006, + "step": 45391 + }, + { + "epoch": 0.3940243574274529, + "grad_norm": 0.10205078125, + "learning_rate": 0.0011923685473245566, + "loss": 0.0947, + "step": 45392 + }, + { + "epoch": 0.39403303790765704, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011923389419802863, + "loss": 0.127, + "step": 45393 + }, + { + "epoch": 0.39404171838786123, + "grad_norm": 0.2890625, + "learning_rate": 0.0011923093365350354, + "loss": 0.0913, + "step": 45394 + }, + { + "epoch": 0.39405039886806537, + "grad_norm": 0.09033203125, + "learning_rate": 0.0011922797309888362, + "loss": 0.1006, + "step": 45395 + }, + { + "epoch": 0.39405907934826956, + "grad_norm": 0.71484375, + "learning_rate": 0.0011922501253417215, + "loss": 0.085, + "step": 45396 + }, + { + "epoch": 0.3940677598284737, + "grad_norm": 0.359375, + "learning_rate": 0.0011922205195937232, + "loss": 0.0986, + "step": 45397 + }, + { + "epoch": 0.3940764403086779, + "grad_norm": 0.11572265625, + "learning_rate": 0.0011921909137448741, + "loss": 0.0889, + "step": 45398 + }, + { + "epoch": 0.394085120788882, + "grad_norm": 0.087890625, + "learning_rate": 0.0011921613077952064, + "loss": 0.0962, + "step": 45399 + }, + { + "epoch": 0.3940938012690862, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011921317017447523, + "loss": 0.0625, + "step": 45400 + }, + { + "epoch": 0.39410248174929036, + "grad_norm": 0.271484375, + "learning_rate": 0.0011921020955935446, + "loss": 0.0625, + "step": 45401 + }, + { + "epoch": 0.39411116222949455, + "grad_norm": 0.65234375, + "learning_rate": 0.0011920724893416154, + "loss": 0.1089, + "step": 45402 + }, + { + "epoch": 0.3941198427096987, + "grad_norm": 0.146484375, + "learning_rate": 0.0011920428829889972, + "loss": 0.0752, + "step": 45403 + }, + { + "epoch": 0.3941285231899029, + "grad_norm": 0.828125, + "learning_rate": 0.001192013276535722, + "loss": 0.0957, + "step": 45404 + }, + { + "epoch": 0.394137203670107, + "grad_norm": 0.53515625, + "learning_rate": 0.0011919836699818226, + "loss": 0.1553, + "step": 45405 + }, + { + "epoch": 0.3941458841503112, + "grad_norm": 0.390625, + "learning_rate": 0.0011919540633273309, + "loss": 0.1357, + "step": 45406 + }, + { + "epoch": 0.39415456463051535, + "grad_norm": 0.51953125, + "learning_rate": 0.0011919244565722802, + "loss": 0.1094, + "step": 45407 + }, + { + "epoch": 0.39416324511071954, + "grad_norm": 0.287109375, + "learning_rate": 0.001191894849716702, + "loss": 0.0928, + "step": 45408 + }, + { + "epoch": 0.3941719255909237, + "grad_norm": 0.72265625, + "learning_rate": 0.001191865242760629, + "loss": 0.1211, + "step": 45409 + }, + { + "epoch": 0.39418060607112787, + "grad_norm": 0.4609375, + "learning_rate": 0.0011918356357040936, + "loss": 0.0923, + "step": 45410 + }, + { + "epoch": 0.394189286551332, + "grad_norm": 0.2109375, + "learning_rate": 0.0011918060285471279, + "loss": 0.0928, + "step": 45411 + }, + { + "epoch": 0.3941979670315362, + "grad_norm": 0.09326171875, + "learning_rate": 0.0011917764212897649, + "loss": 0.0996, + "step": 45412 + }, + { + "epoch": 0.39420664751174034, + "grad_norm": 0.33203125, + "learning_rate": 0.0011917468139320364, + "loss": 0.0933, + "step": 45413 + }, + { + "epoch": 0.39421532799194453, + "grad_norm": 0.1162109375, + "learning_rate": 0.001191717206473975, + "loss": 0.0903, + "step": 45414 + }, + { + "epoch": 0.39422400847214867, + "grad_norm": 0.4140625, + "learning_rate": 0.0011916875989156127, + "loss": 0.1279, + "step": 45415 + }, + { + "epoch": 0.39423268895235286, + "grad_norm": 0.7421875, + "learning_rate": 0.0011916579912569826, + "loss": 0.1221, + "step": 45416 + }, + { + "epoch": 0.394241369432557, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011916283834981169, + "loss": 0.1309, + "step": 45417 + }, + { + "epoch": 0.3942500499127612, + "grad_norm": 0.1494140625, + "learning_rate": 0.001191598775639047, + "loss": 0.1221, + "step": 45418 + }, + { + "epoch": 0.39425873039296533, + "grad_norm": 0.193359375, + "learning_rate": 0.0011915691676798066, + "loss": 0.126, + "step": 45419 + }, + { + "epoch": 0.3942674108731695, + "grad_norm": 0.59765625, + "learning_rate": 0.0011915395596204276, + "loss": 0.1025, + "step": 45420 + }, + { + "epoch": 0.39427609135337366, + "grad_norm": 0.205078125, + "learning_rate": 0.0011915099514609423, + "loss": 0.1128, + "step": 45421 + }, + { + "epoch": 0.39428477183357785, + "grad_norm": 0.2021484375, + "learning_rate": 0.001191480343201383, + "loss": 0.0869, + "step": 45422 + }, + { + "epoch": 0.394293452313782, + "grad_norm": 0.3671875, + "learning_rate": 0.001191450734841782, + "loss": 0.1826, + "step": 45423 + }, + { + "epoch": 0.3943021327939862, + "grad_norm": 0.08984375, + "learning_rate": 0.0011914211263821719, + "loss": 0.0625, + "step": 45424 + }, + { + "epoch": 0.3943108132741903, + "grad_norm": 0.232421875, + "learning_rate": 0.0011913915178225851, + "loss": 0.0835, + "step": 45425 + }, + { + "epoch": 0.3943194937543945, + "grad_norm": 1.4453125, + "learning_rate": 0.001191361909163054, + "loss": 0.1719, + "step": 45426 + }, + { + "epoch": 0.39432817423459865, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011913323004036106, + "loss": 0.1094, + "step": 45427 + }, + { + "epoch": 0.39433685471480284, + "grad_norm": 0.23046875, + "learning_rate": 0.001191302691544288, + "loss": 0.0981, + "step": 45428 + }, + { + "epoch": 0.394345535195007, + "grad_norm": 0.1533203125, + "learning_rate": 0.001191273082585118, + "loss": 0.082, + "step": 45429 + }, + { + "epoch": 0.3943542156752112, + "grad_norm": 0.416015625, + "learning_rate": 0.0011912434735261333, + "loss": 0.0781, + "step": 45430 + }, + { + "epoch": 0.3943628961554153, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011912138643673657, + "loss": 0.061, + "step": 45431 + }, + { + "epoch": 0.3943715766356195, + "grad_norm": 0.115234375, + "learning_rate": 0.0011911842551088482, + "loss": 0.1113, + "step": 45432 + }, + { + "epoch": 0.39438025711582364, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011911546457506127, + "loss": 0.0898, + "step": 45433 + }, + { + "epoch": 0.39438893759602783, + "grad_norm": 0.546875, + "learning_rate": 0.0011911250362926923, + "loss": 0.1211, + "step": 45434 + }, + { + "epoch": 0.39439761807623197, + "grad_norm": 0.7578125, + "learning_rate": 0.0011910954267351187, + "loss": 0.0913, + "step": 45435 + }, + { + "epoch": 0.39440629855643616, + "grad_norm": 0.9765625, + "learning_rate": 0.0011910658170779242, + "loss": 0.1211, + "step": 45436 + }, + { + "epoch": 0.3944149790366403, + "grad_norm": 0.421875, + "learning_rate": 0.001191036207321142, + "loss": 0.1436, + "step": 45437 + }, + { + "epoch": 0.3944236595168445, + "grad_norm": 0.515625, + "learning_rate": 0.0011910065974648036, + "loss": 0.124, + "step": 45438 + }, + { + "epoch": 0.39443233999704863, + "grad_norm": 0.5390625, + "learning_rate": 0.0011909769875089423, + "loss": 0.106, + "step": 45439 + }, + { + "epoch": 0.3944410204772528, + "grad_norm": 0.2080078125, + "learning_rate": 0.0011909473774535894, + "loss": 0.1104, + "step": 45440 + }, + { + "epoch": 0.39444970095745696, + "grad_norm": 0.126953125, + "learning_rate": 0.001190917767298778, + "loss": 0.0942, + "step": 45441 + }, + { + "epoch": 0.39445838143766115, + "grad_norm": 0.435546875, + "learning_rate": 0.0011908881570445403, + "loss": 0.1299, + "step": 45442 + }, + { + "epoch": 0.3944670619178653, + "grad_norm": 0.5078125, + "learning_rate": 0.0011908585466909086, + "loss": 0.1572, + "step": 45443 + }, + { + "epoch": 0.3944757423980695, + "grad_norm": 0.37109375, + "learning_rate": 0.0011908289362379154, + "loss": 0.082, + "step": 45444 + }, + { + "epoch": 0.3944844228782736, + "grad_norm": 0.162109375, + "learning_rate": 0.0011907993256855928, + "loss": 0.1279, + "step": 45445 + }, + { + "epoch": 0.3944931033584778, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011907697150339738, + "loss": 0.0918, + "step": 45446 + }, + { + "epoch": 0.39450178383868195, + "grad_norm": 0.283203125, + "learning_rate": 0.0011907401042830902, + "loss": 0.1406, + "step": 45447 + }, + { + "epoch": 0.39451046431888614, + "grad_norm": 0.265625, + "learning_rate": 0.0011907104934329748, + "loss": 0.0889, + "step": 45448 + }, + { + "epoch": 0.3945191447990903, + "grad_norm": 0.095703125, + "learning_rate": 0.0011906808824836597, + "loss": 0.1279, + "step": 45449 + }, + { + "epoch": 0.3945278252792945, + "grad_norm": 0.306640625, + "learning_rate": 0.0011906512714351773, + "loss": 0.0762, + "step": 45450 + }, + { + "epoch": 0.3945365057594986, + "grad_norm": 0.1611328125, + "learning_rate": 0.00119062166028756, + "loss": 0.125, + "step": 45451 + }, + { + "epoch": 0.3945451862397028, + "grad_norm": 0.3359375, + "learning_rate": 0.0011905920490408405, + "loss": 0.1118, + "step": 45452 + }, + { + "epoch": 0.39455386671990694, + "grad_norm": 0.345703125, + "learning_rate": 0.0011905624376950504, + "loss": 0.0967, + "step": 45453 + }, + { + "epoch": 0.39456254720011114, + "grad_norm": 0.11962890625, + "learning_rate": 0.0011905328262502225, + "loss": 0.0664, + "step": 45454 + }, + { + "epoch": 0.3945712276803153, + "grad_norm": 0.5078125, + "learning_rate": 0.0011905032147063897, + "loss": 0.1211, + "step": 45455 + }, + { + "epoch": 0.39457990816051947, + "grad_norm": 0.50390625, + "learning_rate": 0.001190473603063584, + "loss": 0.1016, + "step": 45456 + }, + { + "epoch": 0.3945885886407236, + "grad_norm": 0.51171875, + "learning_rate": 0.0011904439913218371, + "loss": 0.1113, + "step": 45457 + }, + { + "epoch": 0.3945972691209278, + "grad_norm": 0.19921875, + "learning_rate": 0.001190414379481183, + "loss": 0.0859, + "step": 45458 + }, + { + "epoch": 0.39460594960113193, + "grad_norm": 0.138671875, + "learning_rate": 0.0011903847675416525, + "loss": 0.1289, + "step": 45459 + }, + { + "epoch": 0.3946146300813361, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011903551555032786, + "loss": 0.0923, + "step": 45460 + }, + { + "epoch": 0.39462331056154026, + "grad_norm": 0.54296875, + "learning_rate": 0.0011903255433660933, + "loss": 0.166, + "step": 45461 + }, + { + "epoch": 0.39463199104174446, + "grad_norm": 0.478515625, + "learning_rate": 0.0011902959311301299, + "loss": 0.0859, + "step": 45462 + }, + { + "epoch": 0.3946406715219486, + "grad_norm": 0.1123046875, + "learning_rate": 0.00119026631879542, + "loss": 0.105, + "step": 45463 + }, + { + "epoch": 0.3946493520021528, + "grad_norm": 0.09716796875, + "learning_rate": 0.0011902367063619961, + "loss": 0.125, + "step": 45464 + }, + { + "epoch": 0.3946580324823569, + "grad_norm": 0.0654296875, + "learning_rate": 0.0011902070938298913, + "loss": 0.0669, + "step": 45465 + }, + { + "epoch": 0.39466671296256106, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011901774811991367, + "loss": 0.125, + "step": 45466 + }, + { + "epoch": 0.39467539344276525, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011901478684697656, + "loss": 0.0898, + "step": 45467 + }, + { + "epoch": 0.3946840739229694, + "grad_norm": 0.23828125, + "learning_rate": 0.0011901182556418102, + "loss": 0.1172, + "step": 45468 + }, + { + "epoch": 0.3946927544031736, + "grad_norm": 0.466796875, + "learning_rate": 0.0011900886427153026, + "loss": 0.0776, + "step": 45469 + }, + { + "epoch": 0.3947014348833777, + "grad_norm": 0.173828125, + "learning_rate": 0.0011900590296902757, + "loss": 0.1079, + "step": 45470 + }, + { + "epoch": 0.3947101153635819, + "grad_norm": 0.09130859375, + "learning_rate": 0.0011900294165667613, + "loss": 0.0942, + "step": 45471 + }, + { + "epoch": 0.39471879584378605, + "grad_norm": 0.07666015625, + "learning_rate": 0.0011899998033447926, + "loss": 0.0918, + "step": 45472 + }, + { + "epoch": 0.39472747632399025, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011899701900244006, + "loss": 0.0918, + "step": 45473 + }, + { + "epoch": 0.3947361568041944, + "grad_norm": 0.49609375, + "learning_rate": 0.0011899405766056194, + "loss": 0.0742, + "step": 45474 + }, + { + "epoch": 0.3947448372843986, + "grad_norm": 0.12255859375, + "learning_rate": 0.00118991096308848, + "loss": 0.0874, + "step": 45475 + }, + { + "epoch": 0.3947535177646027, + "grad_norm": 0.25, + "learning_rate": 0.0011898813494730157, + "loss": 0.0996, + "step": 45476 + }, + { + "epoch": 0.3947621982448069, + "grad_norm": 0.1875, + "learning_rate": 0.0011898517357592582, + "loss": 0.0908, + "step": 45477 + }, + { + "epoch": 0.39477087872501104, + "grad_norm": 1.046875, + "learning_rate": 0.0011898221219472404, + "loss": 0.1206, + "step": 45478 + }, + { + "epoch": 0.39477955920521524, + "grad_norm": 0.0908203125, + "learning_rate": 0.0011897925080369945, + "loss": 0.083, + "step": 45479 + }, + { + "epoch": 0.3947882396854194, + "grad_norm": 0.26171875, + "learning_rate": 0.0011897628940285528, + "loss": 0.1553, + "step": 45480 + }, + { + "epoch": 0.39479692016562357, + "grad_norm": 0.1650390625, + "learning_rate": 0.0011897332799219477, + "loss": 0.0781, + "step": 45481 + }, + { + "epoch": 0.3948056006458277, + "grad_norm": 0.2158203125, + "learning_rate": 0.0011897036657172115, + "loss": 0.124, + "step": 45482 + }, + { + "epoch": 0.3948142811260319, + "grad_norm": 0.1962890625, + "learning_rate": 0.0011896740514143766, + "loss": 0.1001, + "step": 45483 + }, + { + "epoch": 0.39482296160623603, + "grad_norm": 0.10888671875, + "learning_rate": 0.001189644437013476, + "loss": 0.0781, + "step": 45484 + }, + { + "epoch": 0.3948316420864402, + "grad_norm": 0.1240234375, + "learning_rate": 0.0011896148225145416, + "loss": 0.1152, + "step": 45485 + }, + { + "epoch": 0.39484032256664436, + "grad_norm": 0.1552734375, + "learning_rate": 0.0011895852079176054, + "loss": 0.1191, + "step": 45486 + }, + { + "epoch": 0.39484900304684856, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011895555932227003, + "loss": 0.0938, + "step": 45487 + }, + { + "epoch": 0.3948576835270527, + "grad_norm": 0.125, + "learning_rate": 0.0011895259784298587, + "loss": 0.0977, + "step": 45488 + }, + { + "epoch": 0.3948663640072569, + "grad_norm": 0.130859375, + "learning_rate": 0.0011894963635391128, + "loss": 0.0791, + "step": 45489 + }, + { + "epoch": 0.394875044487461, + "grad_norm": 0.357421875, + "learning_rate": 0.001189466748550495, + "loss": 0.0825, + "step": 45490 + }, + { + "epoch": 0.3948837249676652, + "grad_norm": 0.6328125, + "learning_rate": 0.0011894371334640374, + "loss": 0.1455, + "step": 45491 + }, + { + "epoch": 0.39489240544786935, + "grad_norm": 0.57421875, + "learning_rate": 0.001189407518279773, + "loss": 0.1221, + "step": 45492 + }, + { + "epoch": 0.39490108592807355, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011893779029977334, + "loss": 0.0854, + "step": 45493 + }, + { + "epoch": 0.3949097664082777, + "grad_norm": 0.212890625, + "learning_rate": 0.0011893482876179523, + "loss": 0.0938, + "step": 45494 + }, + { + "epoch": 0.3949184468884819, + "grad_norm": 0.271484375, + "learning_rate": 0.0011893186721404604, + "loss": 0.0903, + "step": 45495 + }, + { + "epoch": 0.394927127368686, + "grad_norm": 0.494140625, + "learning_rate": 0.0011892890565652916, + "loss": 0.1133, + "step": 45496 + }, + { + "epoch": 0.3949358078488902, + "grad_norm": 0.28515625, + "learning_rate": 0.0011892594408924775, + "loss": 0.1191, + "step": 45497 + }, + { + "epoch": 0.39494448832909435, + "grad_norm": 0.400390625, + "learning_rate": 0.0011892298251220503, + "loss": 0.1113, + "step": 45498 + }, + { + "epoch": 0.39495316880929854, + "grad_norm": 0.447265625, + "learning_rate": 0.001189200209254043, + "loss": 0.1338, + "step": 45499 + }, + { + "epoch": 0.3949618492895027, + "grad_norm": 0.142578125, + "learning_rate": 0.0011891705932884876, + "loss": 0.0732, + "step": 45500 + }, + { + "epoch": 0.39497052976970687, + "grad_norm": 0.74609375, + "learning_rate": 0.0011891409772254167, + "loss": 0.1123, + "step": 45501 + }, + { + "epoch": 0.394979210249911, + "grad_norm": 0.50390625, + "learning_rate": 0.0011891113610648621, + "loss": 0.0933, + "step": 45502 + }, + { + "epoch": 0.3949878907301152, + "grad_norm": 0.1962890625, + "learning_rate": 0.001189081744806857, + "loss": 0.0781, + "step": 45503 + }, + { + "epoch": 0.39499657121031934, + "grad_norm": 0.271484375, + "learning_rate": 0.0011890521284514334, + "loss": 0.1001, + "step": 45504 + }, + { + "epoch": 0.39500525169052353, + "grad_norm": 0.80859375, + "learning_rate": 0.001189022511998624, + "loss": 0.0815, + "step": 45505 + }, + { + "epoch": 0.39501393217072767, + "grad_norm": 0.248046875, + "learning_rate": 0.0011889928954484609, + "loss": 0.1621, + "step": 45506 + }, + { + "epoch": 0.39502261265093186, + "grad_norm": 0.275390625, + "learning_rate": 0.0011889632788009761, + "loss": 0.0986, + "step": 45507 + }, + { + "epoch": 0.395031293131136, + "grad_norm": 3.109375, + "learning_rate": 0.0011889336620562025, + "loss": 0.2256, + "step": 45508 + }, + { + "epoch": 0.3950399736113402, + "grad_norm": 0.8203125, + "learning_rate": 0.0011889040452141726, + "loss": 0.1396, + "step": 45509 + }, + { + "epoch": 0.3950486540915443, + "grad_norm": 0.2490234375, + "learning_rate": 0.0011888744282749185, + "loss": 0.0796, + "step": 45510 + }, + { + "epoch": 0.3950573345717485, + "grad_norm": 0.142578125, + "learning_rate": 0.0011888448112384723, + "loss": 0.1104, + "step": 45511 + }, + { + "epoch": 0.39506601505195266, + "grad_norm": 0.875, + "learning_rate": 0.0011888151941048673, + "loss": 0.125, + "step": 45512 + }, + { + "epoch": 0.39507469553215685, + "grad_norm": 0.70703125, + "learning_rate": 0.0011887855768741352, + "loss": 0.0928, + "step": 45513 + }, + { + "epoch": 0.395083376012361, + "grad_norm": 0.349609375, + "learning_rate": 0.0011887559595463082, + "loss": 0.1167, + "step": 45514 + }, + { + "epoch": 0.3950920564925652, + "grad_norm": 0.0791015625, + "learning_rate": 0.0011887263421214195, + "loss": 0.0723, + "step": 45515 + }, + { + "epoch": 0.3951007369727693, + "grad_norm": 0.24609375, + "learning_rate": 0.0011886967245995006, + "loss": 0.0791, + "step": 45516 + }, + { + "epoch": 0.3951094174529735, + "grad_norm": 0.423828125, + "learning_rate": 0.0011886671069805846, + "loss": 0.0669, + "step": 45517 + }, + { + "epoch": 0.39511809793317765, + "grad_norm": 0.13671875, + "learning_rate": 0.0011886374892647032, + "loss": 0.0972, + "step": 45518 + }, + { + "epoch": 0.39512677841338184, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011886078714518895, + "loss": 0.1025, + "step": 45519 + }, + { + "epoch": 0.395135458893586, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011885782535421751, + "loss": 0.0913, + "step": 45520 + }, + { + "epoch": 0.39514413937379017, + "grad_norm": 0.1611328125, + "learning_rate": 0.0011885486355355938, + "loss": 0.0913, + "step": 45521 + }, + { + "epoch": 0.3951528198539943, + "grad_norm": 0.1953125, + "learning_rate": 0.001188519017432176, + "loss": 0.1011, + "step": 45522 + }, + { + "epoch": 0.3951615003341985, + "grad_norm": 0.1533203125, + "learning_rate": 0.0011884893992319557, + "loss": 0.1021, + "step": 45523 + }, + { + "epoch": 0.39517018081440264, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011884597809349647, + "loss": 0.1006, + "step": 45524 + }, + { + "epoch": 0.39517886129460683, + "grad_norm": 0.56640625, + "learning_rate": 0.0011884301625412353, + "loss": 0.0918, + "step": 45525 + }, + { + "epoch": 0.39518754177481097, + "grad_norm": 0.734375, + "learning_rate": 0.0011884005440508, + "loss": 0.0957, + "step": 45526 + }, + { + "epoch": 0.39519622225501516, + "grad_norm": 0.451171875, + "learning_rate": 0.0011883709254636911, + "loss": 0.0879, + "step": 45527 + }, + { + "epoch": 0.3952049027352193, + "grad_norm": 0.126953125, + "learning_rate": 0.0011883413067799414, + "loss": 0.1309, + "step": 45528 + }, + { + "epoch": 0.3952135832154235, + "grad_norm": 0.06982421875, + "learning_rate": 0.0011883116879995827, + "loss": 0.0684, + "step": 45529 + }, + { + "epoch": 0.39522226369562763, + "grad_norm": 0.123046875, + "learning_rate": 0.0011882820691226476, + "loss": 0.0708, + "step": 45530 + }, + { + "epoch": 0.3952309441758318, + "grad_norm": 0.236328125, + "learning_rate": 0.0011882524501491687, + "loss": 0.0933, + "step": 45531 + }, + { + "epoch": 0.39523962465603596, + "grad_norm": 0.08447265625, + "learning_rate": 0.0011882228310791782, + "loss": 0.1328, + "step": 45532 + }, + { + "epoch": 0.39524830513624015, + "grad_norm": 0.330078125, + "learning_rate": 0.0011881932119127089, + "loss": 0.0723, + "step": 45533 + }, + { + "epoch": 0.3952569856164443, + "grad_norm": 0.1611328125, + "learning_rate": 0.0011881635926497924, + "loss": 0.1138, + "step": 45534 + }, + { + "epoch": 0.3952656660966485, + "grad_norm": 0.13671875, + "learning_rate": 0.0011881339732904615, + "loss": 0.082, + "step": 45535 + }, + { + "epoch": 0.3952743465768526, + "grad_norm": 0.1796875, + "learning_rate": 0.001188104353834749, + "loss": 0.1357, + "step": 45536 + }, + { + "epoch": 0.3952830270570568, + "grad_norm": 0.419921875, + "learning_rate": 0.0011880747342826866, + "loss": 0.0786, + "step": 45537 + }, + { + "epoch": 0.39529170753726095, + "grad_norm": 0.2890625, + "learning_rate": 0.0011880451146343072, + "loss": 0.0967, + "step": 45538 + }, + { + "epoch": 0.39530038801746514, + "grad_norm": 0.275390625, + "learning_rate": 0.0011880154948896427, + "loss": 0.1084, + "step": 45539 + }, + { + "epoch": 0.3953090684976693, + "grad_norm": 0.345703125, + "learning_rate": 0.0011879858750487257, + "loss": 0.1025, + "step": 45540 + }, + { + "epoch": 0.3953177489778735, + "grad_norm": 0.345703125, + "learning_rate": 0.0011879562551115887, + "loss": 0.1025, + "step": 45541 + }, + { + "epoch": 0.3953264294580776, + "grad_norm": 0.2109375, + "learning_rate": 0.0011879266350782645, + "loss": 0.1064, + "step": 45542 + }, + { + "epoch": 0.3953351099382818, + "grad_norm": 0.18359375, + "learning_rate": 0.0011878970149487847, + "loss": 0.0713, + "step": 45543 + }, + { + "epoch": 0.39534379041848594, + "grad_norm": 0.337890625, + "learning_rate": 0.001187867394723182, + "loss": 0.125, + "step": 45544 + }, + { + "epoch": 0.39535247089869013, + "grad_norm": 0.2412109375, + "learning_rate": 0.001187837774401489, + "loss": 0.0967, + "step": 45545 + }, + { + "epoch": 0.39536115137889427, + "grad_norm": 0.271484375, + "learning_rate": 0.0011878081539837381, + "loss": 0.1182, + "step": 45546 + }, + { + "epoch": 0.39536983185909846, + "grad_norm": 0.169921875, + "learning_rate": 0.0011877785334699614, + "loss": 0.1221, + "step": 45547 + }, + { + "epoch": 0.3953785123393026, + "grad_norm": 0.138671875, + "learning_rate": 0.0011877489128601913, + "loss": 0.1006, + "step": 45548 + }, + { + "epoch": 0.3953871928195068, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011877192921544602, + "loss": 0.1162, + "step": 45549 + }, + { + "epoch": 0.39539587329971093, + "grad_norm": 0.447265625, + "learning_rate": 0.0011876896713528006, + "loss": 0.1396, + "step": 45550 + }, + { + "epoch": 0.3954045537799151, + "grad_norm": 0.1591796875, + "learning_rate": 0.0011876600504552453, + "loss": 0.1084, + "step": 45551 + }, + { + "epoch": 0.39541323426011926, + "grad_norm": 0.3671875, + "learning_rate": 0.0011876304294618255, + "loss": 0.0815, + "step": 45552 + }, + { + "epoch": 0.39542191474032345, + "grad_norm": 0.171875, + "learning_rate": 0.0011876008083725753, + "loss": 0.0781, + "step": 45553 + }, + { + "epoch": 0.3954305952205276, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011875711871875256, + "loss": 0.0811, + "step": 45554 + }, + { + "epoch": 0.3954392757007318, + "grad_norm": 0.345703125, + "learning_rate": 0.0011875415659067098, + "loss": 0.106, + "step": 45555 + }, + { + "epoch": 0.3954479561809359, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011875119445301596, + "loss": 0.0781, + "step": 45556 + }, + { + "epoch": 0.3954566366611401, + "grad_norm": 0.1875, + "learning_rate": 0.0011874823230579078, + "loss": 0.0977, + "step": 45557 + }, + { + "epoch": 0.39546531714134425, + "grad_norm": 0.6328125, + "learning_rate": 0.0011874527014899862, + "loss": 0.1426, + "step": 45558 + }, + { + "epoch": 0.39547399762154845, + "grad_norm": 0.69921875, + "learning_rate": 0.0011874230798264278, + "loss": 0.103, + "step": 45559 + }, + { + "epoch": 0.3954826781017526, + "grad_norm": 0.263671875, + "learning_rate": 0.0011873934580672652, + "loss": 0.0928, + "step": 45560 + }, + { + "epoch": 0.3954913585819568, + "grad_norm": 0.72265625, + "learning_rate": 0.00118736383621253, + "loss": 0.0889, + "step": 45561 + }, + { + "epoch": 0.3955000390621609, + "grad_norm": 0.34765625, + "learning_rate": 0.0011873342142622555, + "loss": 0.0845, + "step": 45562 + }, + { + "epoch": 0.3955087195423651, + "grad_norm": 0.376953125, + "learning_rate": 0.001187304592216473, + "loss": 0.0981, + "step": 45563 + }, + { + "epoch": 0.39551740002256924, + "grad_norm": 0.375, + "learning_rate": 0.0011872749700752162, + "loss": 0.0869, + "step": 45564 + }, + { + "epoch": 0.39552608050277344, + "grad_norm": 0.14453125, + "learning_rate": 0.0011872453478385164, + "loss": 0.1133, + "step": 45565 + }, + { + "epoch": 0.3955347609829776, + "grad_norm": 0.2197265625, + "learning_rate": 0.0011872157255064062, + "loss": 0.1069, + "step": 45566 + }, + { + "epoch": 0.39554344146318177, + "grad_norm": 0.0927734375, + "learning_rate": 0.0011871861030789187, + "loss": 0.0986, + "step": 45567 + }, + { + "epoch": 0.3955521219433859, + "grad_norm": 0.15625, + "learning_rate": 0.0011871564805560854, + "loss": 0.1484, + "step": 45568 + }, + { + "epoch": 0.3955608024235901, + "grad_norm": 0.625, + "learning_rate": 0.0011871268579379394, + "loss": 0.1318, + "step": 45569 + }, + { + "epoch": 0.39556948290379423, + "grad_norm": 0.6015625, + "learning_rate": 0.0011870972352245122, + "loss": 0.1021, + "step": 45570 + }, + { + "epoch": 0.3955781633839984, + "grad_norm": 0.373046875, + "learning_rate": 0.0011870676124158372, + "loss": 0.0811, + "step": 45571 + }, + { + "epoch": 0.39558684386420256, + "grad_norm": 1.140625, + "learning_rate": 0.0011870379895119465, + "loss": 0.125, + "step": 45572 + }, + { + "epoch": 0.39559552434440676, + "grad_norm": 0.34375, + "learning_rate": 0.0011870083665128722, + "loss": 0.0815, + "step": 45573 + }, + { + "epoch": 0.3956042048246109, + "grad_norm": 0.142578125, + "learning_rate": 0.0011869787434186467, + "loss": 0.1064, + "step": 45574 + }, + { + "epoch": 0.3956128853048151, + "grad_norm": 0.08740234375, + "learning_rate": 0.0011869491202293025, + "loss": 0.1006, + "step": 45575 + }, + { + "epoch": 0.3956215657850192, + "grad_norm": 0.232421875, + "learning_rate": 0.0011869194969448727, + "loss": 0.0913, + "step": 45576 + }, + { + "epoch": 0.3956302462652234, + "grad_norm": 0.306640625, + "learning_rate": 0.0011868898735653886, + "loss": 0.0806, + "step": 45577 + }, + { + "epoch": 0.39563892674542755, + "grad_norm": 0.81640625, + "learning_rate": 0.0011868602500908828, + "loss": 0.1055, + "step": 45578 + }, + { + "epoch": 0.39564760722563175, + "grad_norm": 0.75, + "learning_rate": 0.0011868306265213882, + "loss": 0.1055, + "step": 45579 + }, + { + "epoch": 0.3956562877058359, + "grad_norm": 0.140625, + "learning_rate": 0.0011868010028569368, + "loss": 0.0623, + "step": 45580 + }, + { + "epoch": 0.3956649681860401, + "grad_norm": 0.5390625, + "learning_rate": 0.0011867713790975614, + "loss": 0.0811, + "step": 45581 + }, + { + "epoch": 0.3956736486662442, + "grad_norm": 0.72265625, + "learning_rate": 0.0011867417552432937, + "loss": 0.1001, + "step": 45582 + }, + { + "epoch": 0.3956823291464484, + "grad_norm": 0.27734375, + "learning_rate": 0.0011867121312941669, + "loss": 0.0928, + "step": 45583 + }, + { + "epoch": 0.39569100962665255, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011866825072502129, + "loss": 0.1113, + "step": 45584 + }, + { + "epoch": 0.39569969010685674, + "grad_norm": 0.09521484375, + "learning_rate": 0.0011866528831114642, + "loss": 0.0972, + "step": 45585 + }, + { + "epoch": 0.3957083705870609, + "grad_norm": 0.3515625, + "learning_rate": 0.0011866232588779533, + "loss": 0.1201, + "step": 45586 + }, + { + "epoch": 0.39571705106726507, + "grad_norm": 0.154296875, + "learning_rate": 0.0011865936345497121, + "loss": 0.1025, + "step": 45587 + }, + { + "epoch": 0.3957257315474692, + "grad_norm": 0.212890625, + "learning_rate": 0.0011865640101267737, + "loss": 0.0957, + "step": 45588 + }, + { + "epoch": 0.39573441202767334, + "grad_norm": 0.271484375, + "learning_rate": 0.0011865343856091704, + "loss": 0.1118, + "step": 45589 + }, + { + "epoch": 0.39574309250787754, + "grad_norm": 0.232421875, + "learning_rate": 0.0011865047609969344, + "loss": 0.1094, + "step": 45590 + }, + { + "epoch": 0.3957517729880817, + "grad_norm": 0.12890625, + "learning_rate": 0.0011864751362900977, + "loss": 0.1748, + "step": 45591 + }, + { + "epoch": 0.39576045346828587, + "grad_norm": 0.1025390625, + "learning_rate": 0.0011864455114886931, + "loss": 0.1104, + "step": 45592 + }, + { + "epoch": 0.39576913394849, + "grad_norm": 0.16796875, + "learning_rate": 0.0011864158865927535, + "loss": 0.1055, + "step": 45593 + }, + { + "epoch": 0.3957778144286942, + "grad_norm": 0.439453125, + "learning_rate": 0.0011863862616023103, + "loss": 0.0938, + "step": 45594 + }, + { + "epoch": 0.39578649490889833, + "grad_norm": 0.12353515625, + "learning_rate": 0.0011863566365173963, + "loss": 0.0728, + "step": 45595 + }, + { + "epoch": 0.3957951753891025, + "grad_norm": 0.224609375, + "learning_rate": 0.0011863270113380447, + "loss": 0.0928, + "step": 45596 + }, + { + "epoch": 0.39580385586930666, + "grad_norm": 0.515625, + "learning_rate": 0.0011862973860642864, + "loss": 0.0942, + "step": 45597 + }, + { + "epoch": 0.39581253634951086, + "grad_norm": 0.73828125, + "learning_rate": 0.001186267760696155, + "loss": 0.1074, + "step": 45598 + }, + { + "epoch": 0.395821216829715, + "grad_norm": 0.353515625, + "learning_rate": 0.0011862381352336824, + "loss": 0.1001, + "step": 45599 + }, + { + "epoch": 0.3958298973099192, + "grad_norm": 0.1103515625, + "learning_rate": 0.001186208509676901, + "loss": 0.1113, + "step": 45600 + }, + { + "epoch": 0.3958385777901233, + "grad_norm": 0.11474609375, + "learning_rate": 0.0011861788840258432, + "loss": 0.0908, + "step": 45601 + }, + { + "epoch": 0.3958472582703275, + "grad_norm": 0.099609375, + "learning_rate": 0.0011861492582805415, + "loss": 0.0923, + "step": 45602 + }, + { + "epoch": 0.39585593875053166, + "grad_norm": 0.095703125, + "learning_rate": 0.0011861196324410284, + "loss": 0.0986, + "step": 45603 + }, + { + "epoch": 0.39586461923073585, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011860900065073358, + "loss": 0.1348, + "step": 45604 + }, + { + "epoch": 0.39587329971094, + "grad_norm": 0.1328125, + "learning_rate": 0.001186060380479497, + "loss": 0.085, + "step": 45605 + }, + { + "epoch": 0.3958819801911442, + "grad_norm": 0.478515625, + "learning_rate": 0.0011860307543575437, + "loss": 0.1348, + "step": 45606 + }, + { + "epoch": 0.3958906606713483, + "grad_norm": 0.73046875, + "learning_rate": 0.0011860011281415084, + "loss": 0.0898, + "step": 45607 + }, + { + "epoch": 0.3958993411515525, + "grad_norm": 0.5234375, + "learning_rate": 0.0011859715018314235, + "loss": 0.0884, + "step": 45608 + }, + { + "epoch": 0.39590802163175665, + "grad_norm": 0.232421875, + "learning_rate": 0.0011859418754273214, + "loss": 0.127, + "step": 45609 + }, + { + "epoch": 0.39591670211196084, + "grad_norm": 0.1064453125, + "learning_rate": 0.0011859122489292348, + "loss": 0.0908, + "step": 45610 + }, + { + "epoch": 0.395925382592165, + "grad_norm": 0.09912109375, + "learning_rate": 0.0011858826223371955, + "loss": 0.1211, + "step": 45611 + }, + { + "epoch": 0.39593406307236917, + "grad_norm": 0.15234375, + "learning_rate": 0.0011858529956512364, + "loss": 0.127, + "step": 45612 + }, + { + "epoch": 0.3959427435525733, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011858233688713897, + "loss": 0.0938, + "step": 45613 + }, + { + "epoch": 0.3959514240327775, + "grad_norm": 0.0625, + "learning_rate": 0.0011857937419976882, + "loss": 0.0645, + "step": 45614 + }, + { + "epoch": 0.39596010451298164, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011857641150301637, + "loss": 0.1055, + "step": 45615 + }, + { + "epoch": 0.39596878499318583, + "grad_norm": 0.5234375, + "learning_rate": 0.0011857344879688488, + "loss": 0.1104, + "step": 45616 + }, + { + "epoch": 0.39597746547338997, + "grad_norm": 0.10205078125, + "learning_rate": 0.0011857048608137758, + "loss": 0.0664, + "step": 45617 + }, + { + "epoch": 0.39598614595359416, + "grad_norm": 0.4921875, + "learning_rate": 0.0011856752335649777, + "loss": 0.1201, + "step": 45618 + }, + { + "epoch": 0.3959948264337983, + "grad_norm": 1.40625, + "learning_rate": 0.0011856456062224862, + "loss": 0.1855, + "step": 45619 + }, + { + "epoch": 0.3960035069140025, + "grad_norm": 0.6015625, + "learning_rate": 0.0011856159787863338, + "loss": 0.0752, + "step": 45620 + }, + { + "epoch": 0.3960121873942066, + "grad_norm": 0.51953125, + "learning_rate": 0.0011855863512565533, + "loss": 0.2295, + "step": 45621 + }, + { + "epoch": 0.3960208678744108, + "grad_norm": 1.6484375, + "learning_rate": 0.0011855567236331768, + "loss": 0.2158, + "step": 45622 + }, + { + "epoch": 0.39602954835461496, + "grad_norm": 0.17578125, + "learning_rate": 0.0011855270959162368, + "loss": 0.1055, + "step": 45623 + }, + { + "epoch": 0.39603822883481915, + "grad_norm": 0.14453125, + "learning_rate": 0.0011854974681057657, + "loss": 0.1191, + "step": 45624 + }, + { + "epoch": 0.3960469093150233, + "grad_norm": 0.13671875, + "learning_rate": 0.0011854678402017955, + "loss": 0.0972, + "step": 45625 + }, + { + "epoch": 0.3960555897952275, + "grad_norm": 0.3046875, + "learning_rate": 0.0011854382122043591, + "loss": 0.0757, + "step": 45626 + }, + { + "epoch": 0.3960642702754316, + "grad_norm": 0.13671875, + "learning_rate": 0.001185408584113489, + "loss": 0.1055, + "step": 45627 + }, + { + "epoch": 0.3960729507556358, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011853789559292172, + "loss": 0.1152, + "step": 45628 + }, + { + "epoch": 0.39608163123583995, + "grad_norm": 0.58203125, + "learning_rate": 0.0011853493276515763, + "loss": 0.105, + "step": 45629 + }, + { + "epoch": 0.39609031171604414, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011853196992805985, + "loss": 0.0894, + "step": 45630 + }, + { + "epoch": 0.3960989921962483, + "grad_norm": 0.4296875, + "learning_rate": 0.0011852900708163167, + "loss": 0.103, + "step": 45631 + }, + { + "epoch": 0.39610767267645247, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011852604422587626, + "loss": 0.105, + "step": 45632 + }, + { + "epoch": 0.3961163531566566, + "grad_norm": 0.5859375, + "learning_rate": 0.0011852308136079692, + "loss": 0.0972, + "step": 45633 + }, + { + "epoch": 0.3961250336368608, + "grad_norm": 0.1337890625, + "learning_rate": 0.0011852011848639684, + "loss": 0.0771, + "step": 45634 + }, + { + "epoch": 0.39613371411706494, + "grad_norm": 0.65625, + "learning_rate": 0.001185171556026793, + "loss": 0.1206, + "step": 45635 + }, + { + "epoch": 0.39614239459726913, + "grad_norm": 0.427734375, + "learning_rate": 0.0011851419270964751, + "loss": 0.127, + "step": 45636 + }, + { + "epoch": 0.39615107507747327, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011851122980730477, + "loss": 0.0957, + "step": 45637 + }, + { + "epoch": 0.39615975555767746, + "grad_norm": 0.1796875, + "learning_rate": 0.0011850826689565426, + "loss": 0.124, + "step": 45638 + }, + { + "epoch": 0.3961684360378816, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011850530397469923, + "loss": 0.1299, + "step": 45639 + }, + { + "epoch": 0.3961771165180858, + "grad_norm": 0.328125, + "learning_rate": 0.001185023410444429, + "loss": 0.0933, + "step": 45640 + }, + { + "epoch": 0.39618579699828993, + "grad_norm": 0.296875, + "learning_rate": 0.0011849937810488858, + "loss": 0.1289, + "step": 45641 + }, + { + "epoch": 0.3961944774784941, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011849641515603948, + "loss": 0.0996, + "step": 45642 + }, + { + "epoch": 0.39620315795869826, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011849345219789877, + "loss": 0.1445, + "step": 45643 + }, + { + "epoch": 0.39621183843890245, + "grad_norm": 0.234375, + "learning_rate": 0.0011849048923046979, + "loss": 0.1064, + "step": 45644 + }, + { + "epoch": 0.3962205189191066, + "grad_norm": 0.109375, + "learning_rate": 0.0011848752625375572, + "loss": 0.1055, + "step": 45645 + }, + { + "epoch": 0.3962291993993108, + "grad_norm": 0.71875, + "learning_rate": 0.0011848456326775982, + "loss": 0.1729, + "step": 45646 + }, + { + "epoch": 0.3962378798795149, + "grad_norm": 0.203125, + "learning_rate": 0.0011848160027248533, + "loss": 0.1128, + "step": 45647 + }, + { + "epoch": 0.3962465603597191, + "grad_norm": 0.16015625, + "learning_rate": 0.0011847863726793548, + "loss": 0.0776, + "step": 45648 + }, + { + "epoch": 0.39625524083992325, + "grad_norm": 0.2890625, + "learning_rate": 0.0011847567425411357, + "loss": 0.1074, + "step": 45649 + }, + { + "epoch": 0.39626392132012744, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011847271123102275, + "loss": 0.1348, + "step": 45650 + }, + { + "epoch": 0.3962726018003316, + "grad_norm": 0.322265625, + "learning_rate": 0.0011846974819866632, + "loss": 0.0781, + "step": 45651 + }, + { + "epoch": 0.3962812822805358, + "grad_norm": 0.42578125, + "learning_rate": 0.0011846678515704747, + "loss": 0.0859, + "step": 45652 + }, + { + "epoch": 0.3962899627607399, + "grad_norm": 0.310546875, + "learning_rate": 0.0011846382210616948, + "loss": 0.1196, + "step": 45653 + }, + { + "epoch": 0.3962986432409441, + "grad_norm": 0.158203125, + "learning_rate": 0.0011846085904603561, + "loss": 0.0894, + "step": 45654 + }, + { + "epoch": 0.39630732372114824, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011845789597664902, + "loss": 0.1064, + "step": 45655 + }, + { + "epoch": 0.39631600420135243, + "grad_norm": 0.2197265625, + "learning_rate": 0.0011845493289801303, + "loss": 0.1006, + "step": 45656 + }, + { + "epoch": 0.39632468468155657, + "grad_norm": 0.34375, + "learning_rate": 0.0011845196981013087, + "loss": 0.1035, + "step": 45657 + }, + { + "epoch": 0.39633336516176076, + "grad_norm": 0.07568359375, + "learning_rate": 0.0011844900671300574, + "loss": 0.0933, + "step": 45658 + }, + { + "epoch": 0.3963420456419649, + "grad_norm": 0.34765625, + "learning_rate": 0.0011844604360664094, + "loss": 0.0869, + "step": 45659 + }, + { + "epoch": 0.3963507261221691, + "grad_norm": 0.498046875, + "learning_rate": 0.001184430804910396, + "loss": 0.1426, + "step": 45660 + }, + { + "epoch": 0.39635940660237323, + "grad_norm": 0.6796875, + "learning_rate": 0.0011844011736620507, + "loss": 0.1001, + "step": 45661 + }, + { + "epoch": 0.3963680870825774, + "grad_norm": 0.1005859375, + "learning_rate": 0.001184371542321406, + "loss": 0.1289, + "step": 45662 + }, + { + "epoch": 0.39637676756278156, + "grad_norm": 0.5390625, + "learning_rate": 0.001184341910888493, + "loss": 0.1328, + "step": 45663 + }, + { + "epoch": 0.39638544804298576, + "grad_norm": 0.2177734375, + "learning_rate": 0.0011843122793633457, + "loss": 0.1182, + "step": 45664 + }, + { + "epoch": 0.3963941285231899, + "grad_norm": 0.12890625, + "learning_rate": 0.0011842826477459953, + "loss": 0.1074, + "step": 45665 + }, + { + "epoch": 0.3964028090033941, + "grad_norm": 0.1083984375, + "learning_rate": 0.0011842530160364748, + "loss": 0.0913, + "step": 45666 + }, + { + "epoch": 0.3964114894835982, + "grad_norm": 0.08251953125, + "learning_rate": 0.0011842233842348166, + "loss": 0.0991, + "step": 45667 + }, + { + "epoch": 0.3964201699638024, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011841937523410528, + "loss": 0.1099, + "step": 45668 + }, + { + "epoch": 0.39642885044400655, + "grad_norm": 0.24609375, + "learning_rate": 0.001184164120355216, + "loss": 0.0835, + "step": 45669 + }, + { + "epoch": 0.39643753092421075, + "grad_norm": 0.263671875, + "learning_rate": 0.0011841344882773388, + "loss": 0.0928, + "step": 45670 + }, + { + "epoch": 0.3964462114044149, + "grad_norm": 0.78125, + "learning_rate": 0.0011841048561074534, + "loss": 0.1416, + "step": 45671 + }, + { + "epoch": 0.3964548918846191, + "grad_norm": 0.625, + "learning_rate": 0.001184075223845592, + "loss": 0.1016, + "step": 45672 + }, + { + "epoch": 0.3964635723648232, + "grad_norm": 0.171875, + "learning_rate": 0.001184045591491787, + "loss": 0.1006, + "step": 45673 + }, + { + "epoch": 0.3964722528450274, + "grad_norm": 0.53515625, + "learning_rate": 0.0011840159590460716, + "loss": 0.167, + "step": 45674 + }, + { + "epoch": 0.39648093332523154, + "grad_norm": 1.7890625, + "learning_rate": 0.0011839863265084769, + "loss": 0.1582, + "step": 45675 + }, + { + "epoch": 0.39648961380543574, + "grad_norm": 0.52734375, + "learning_rate": 0.0011839566938790362, + "loss": 0.0908, + "step": 45676 + }, + { + "epoch": 0.3964982942856399, + "grad_norm": 0.283203125, + "learning_rate": 0.0011839270611577822, + "loss": 0.0869, + "step": 45677 + }, + { + "epoch": 0.39650697476584407, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011838974283447461, + "loss": 0.1338, + "step": 45678 + }, + { + "epoch": 0.3965156552460482, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011838677954399616, + "loss": 0.0977, + "step": 45679 + }, + { + "epoch": 0.3965243357262524, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011838381624434604, + "loss": 0.105, + "step": 45680 + }, + { + "epoch": 0.39653301620645653, + "grad_norm": 0.10205078125, + "learning_rate": 0.001183808529355275, + "loss": 0.0835, + "step": 45681 + }, + { + "epoch": 0.3965416966866607, + "grad_norm": 0.419921875, + "learning_rate": 0.001183778896175438, + "loss": 0.1328, + "step": 45682 + }, + { + "epoch": 0.39655037716686486, + "grad_norm": 0.189453125, + "learning_rate": 0.0011837492629039814, + "loss": 0.0986, + "step": 45683 + }, + { + "epoch": 0.39655905764706906, + "grad_norm": 0.28125, + "learning_rate": 0.001183719629540938, + "loss": 0.1147, + "step": 45684 + }, + { + "epoch": 0.3965677381272732, + "grad_norm": 0.3515625, + "learning_rate": 0.00118368999608634, + "loss": 0.1016, + "step": 45685 + }, + { + "epoch": 0.3965764186074774, + "grad_norm": 0.2470703125, + "learning_rate": 0.0011836603625402199, + "loss": 0.1152, + "step": 45686 + }, + { + "epoch": 0.3965850990876815, + "grad_norm": 0.1337890625, + "learning_rate": 0.0011836307289026103, + "loss": 0.0718, + "step": 45687 + }, + { + "epoch": 0.3965937795678857, + "grad_norm": 0.47265625, + "learning_rate": 0.0011836010951735431, + "loss": 0.1216, + "step": 45688 + }, + { + "epoch": 0.39660246004808986, + "grad_norm": 0.52734375, + "learning_rate": 0.001183571461353051, + "loss": 0.0938, + "step": 45689 + }, + { + "epoch": 0.39661114052829405, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011835418274411668, + "loss": 0.0762, + "step": 45690 + }, + { + "epoch": 0.3966198210084982, + "grad_norm": 0.10546875, + "learning_rate": 0.001183512193437922, + "loss": 0.1533, + "step": 45691 + }, + { + "epoch": 0.3966285014887024, + "grad_norm": 0.2734375, + "learning_rate": 0.0011834825593433499, + "loss": 0.0918, + "step": 45692 + }, + { + "epoch": 0.3966371819689065, + "grad_norm": 0.1220703125, + "learning_rate": 0.001183452925157482, + "loss": 0.0898, + "step": 45693 + }, + { + "epoch": 0.3966458624491107, + "grad_norm": 0.1943359375, + "learning_rate": 0.001183423290880352, + "loss": 0.0967, + "step": 45694 + }, + { + "epoch": 0.39665454292931485, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011833936565119909, + "loss": 0.1172, + "step": 45695 + }, + { + "epoch": 0.39666322340951904, + "grad_norm": 0.1982421875, + "learning_rate": 0.0011833640220524323, + "loss": 0.0811, + "step": 45696 + }, + { + "epoch": 0.3966719038897232, + "grad_norm": 0.306640625, + "learning_rate": 0.0011833343875017074, + "loss": 0.1118, + "step": 45697 + }, + { + "epoch": 0.39668058436992737, + "grad_norm": 0.412109375, + "learning_rate": 0.00118330475285985, + "loss": 0.0889, + "step": 45698 + }, + { + "epoch": 0.3966892648501315, + "grad_norm": 0.189453125, + "learning_rate": 0.001183275118126891, + "loss": 0.1157, + "step": 45699 + }, + { + "epoch": 0.3966979453303357, + "grad_norm": 0.33984375, + "learning_rate": 0.0011832454833028642, + "loss": 0.1094, + "step": 45700 + }, + { + "epoch": 0.39670662581053984, + "grad_norm": 0.111328125, + "learning_rate": 0.0011832158483878012, + "loss": 0.1055, + "step": 45701 + }, + { + "epoch": 0.39671530629074403, + "grad_norm": 0.77734375, + "learning_rate": 0.0011831862133817347, + "loss": 0.1016, + "step": 45702 + }, + { + "epoch": 0.39672398677094817, + "grad_norm": 0.21484375, + "learning_rate": 0.0011831565782846967, + "loss": 0.1143, + "step": 45703 + }, + { + "epoch": 0.39673266725115236, + "grad_norm": 0.091796875, + "learning_rate": 0.00118312694309672, + "loss": 0.0776, + "step": 45704 + }, + { + "epoch": 0.3967413477313565, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011830973078178371, + "loss": 0.1309, + "step": 45705 + }, + { + "epoch": 0.3967500282115607, + "grad_norm": 0.3984375, + "learning_rate": 0.0011830676724480802, + "loss": 0.0742, + "step": 45706 + }, + { + "epoch": 0.3967587086917648, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011830380369874818, + "loss": 0.0894, + "step": 45707 + }, + { + "epoch": 0.396767389171969, + "grad_norm": 0.08740234375, + "learning_rate": 0.001183008401436074, + "loss": 0.083, + "step": 45708 + }, + { + "epoch": 0.39677606965217316, + "grad_norm": 0.9609375, + "learning_rate": 0.0011829787657938896, + "loss": 0.0874, + "step": 45709 + }, + { + "epoch": 0.39678475013237735, + "grad_norm": 0.228515625, + "learning_rate": 0.001182949130060961, + "loss": 0.1309, + "step": 45710 + }, + { + "epoch": 0.3967934306125815, + "grad_norm": 0.07373046875, + "learning_rate": 0.0011829194942373205, + "loss": 0.0747, + "step": 45711 + }, + { + "epoch": 0.3968021110927856, + "grad_norm": 0.150390625, + "learning_rate": 0.001182889858323, + "loss": 0.0977, + "step": 45712 + }, + { + "epoch": 0.3968107915729898, + "grad_norm": 0.455078125, + "learning_rate": 0.0011828602223180327, + "loss": 0.1084, + "step": 45713 + }, + { + "epoch": 0.39681947205319396, + "grad_norm": 0.181640625, + "learning_rate": 0.001182830586222451, + "loss": 0.127, + "step": 45714 + }, + { + "epoch": 0.39682815253339815, + "grad_norm": 0.07958984375, + "learning_rate": 0.0011828009500362865, + "loss": 0.1226, + "step": 45715 + }, + { + "epoch": 0.3968368330136023, + "grad_norm": 0.091796875, + "learning_rate": 0.0011827713137595724, + "loss": 0.1128, + "step": 45716 + }, + { + "epoch": 0.3968455134938065, + "grad_norm": 0.515625, + "learning_rate": 0.0011827416773923407, + "loss": 0.1006, + "step": 45717 + }, + { + "epoch": 0.3968541939740106, + "grad_norm": 0.267578125, + "learning_rate": 0.0011827120409346242, + "loss": 0.1196, + "step": 45718 + }, + { + "epoch": 0.3968628744542148, + "grad_norm": 0.267578125, + "learning_rate": 0.001182682404386455, + "loss": 0.0957, + "step": 45719 + }, + { + "epoch": 0.39687155493441895, + "grad_norm": 0.08837890625, + "learning_rate": 0.0011826527677478655, + "loss": 0.103, + "step": 45720 + }, + { + "epoch": 0.39688023541462314, + "grad_norm": 0.435546875, + "learning_rate": 0.0011826231310188883, + "loss": 0.1006, + "step": 45721 + }, + { + "epoch": 0.3968889158948273, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011825934941995554, + "loss": 0.1201, + "step": 45722 + }, + { + "epoch": 0.39689759637503147, + "grad_norm": 0.28125, + "learning_rate": 0.0011825638572898994, + "loss": 0.1133, + "step": 45723 + }, + { + "epoch": 0.3969062768552356, + "grad_norm": 0.1337890625, + "learning_rate": 0.0011825342202899532, + "loss": 0.0742, + "step": 45724 + }, + { + "epoch": 0.3969149573354398, + "grad_norm": 0.197265625, + "learning_rate": 0.0011825045831997485, + "loss": 0.0957, + "step": 45725 + }, + { + "epoch": 0.39692363781564394, + "grad_norm": 0.10107421875, + "learning_rate": 0.0011824749460193182, + "loss": 0.1035, + "step": 45726 + }, + { + "epoch": 0.39693231829584813, + "grad_norm": 0.4453125, + "learning_rate": 0.0011824453087486946, + "loss": 0.0942, + "step": 45727 + }, + { + "epoch": 0.39694099877605227, + "grad_norm": 0.26953125, + "learning_rate": 0.0011824156713879101, + "loss": 0.0889, + "step": 45728 + }, + { + "epoch": 0.39694967925625646, + "grad_norm": 1.2109375, + "learning_rate": 0.0011823860339369968, + "loss": 0.125, + "step": 45729 + }, + { + "epoch": 0.3969583597364606, + "grad_norm": 0.19140625, + "learning_rate": 0.0011823563963959875, + "loss": 0.0688, + "step": 45730 + }, + { + "epoch": 0.3969670402166648, + "grad_norm": 0.2001953125, + "learning_rate": 0.0011823267587649144, + "loss": 0.1143, + "step": 45731 + }, + { + "epoch": 0.39697572069686893, + "grad_norm": 0.228515625, + "learning_rate": 0.0011822971210438103, + "loss": 0.1089, + "step": 45732 + }, + { + "epoch": 0.3969844011770731, + "grad_norm": 0.6875, + "learning_rate": 0.0011822674832327072, + "loss": 0.1357, + "step": 45733 + }, + { + "epoch": 0.39699308165727726, + "grad_norm": 0.083984375, + "learning_rate": 0.0011822378453316372, + "loss": 0.1001, + "step": 45734 + }, + { + "epoch": 0.39700176213748145, + "grad_norm": 0.111328125, + "learning_rate": 0.0011822082073406335, + "loss": 0.0679, + "step": 45735 + }, + { + "epoch": 0.3970104426176856, + "grad_norm": 0.1884765625, + "learning_rate": 0.001182178569259728, + "loss": 0.1123, + "step": 45736 + }, + { + "epoch": 0.3970191230978898, + "grad_norm": 0.474609375, + "learning_rate": 0.0011821489310889534, + "loss": 0.1426, + "step": 45737 + }, + { + "epoch": 0.3970278035780939, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011821192928283418, + "loss": 0.0669, + "step": 45738 + }, + { + "epoch": 0.3970364840582981, + "grad_norm": 0.0693359375, + "learning_rate": 0.0011820896544779257, + "loss": 0.064, + "step": 45739 + }, + { + "epoch": 0.39704516453850225, + "grad_norm": 0.130859375, + "learning_rate": 0.001182060016037738, + "loss": 0.0957, + "step": 45740 + }, + { + "epoch": 0.39705384501870644, + "grad_norm": 0.375, + "learning_rate": 0.0011820303775078103, + "loss": 0.1318, + "step": 45741 + }, + { + "epoch": 0.3970625254989106, + "grad_norm": 0.318359375, + "learning_rate": 0.0011820007388881756, + "loss": 0.0898, + "step": 45742 + }, + { + "epoch": 0.39707120597911477, + "grad_norm": 0.298828125, + "learning_rate": 0.001181971100178866, + "loss": 0.0898, + "step": 45743 + }, + { + "epoch": 0.3970798864593189, + "grad_norm": 0.2470703125, + "learning_rate": 0.0011819414613799142, + "loss": 0.1416, + "step": 45744 + }, + { + "epoch": 0.3970885669395231, + "grad_norm": 0.365234375, + "learning_rate": 0.0011819118224913525, + "loss": 0.1025, + "step": 45745 + }, + { + "epoch": 0.39709724741972724, + "grad_norm": 0.1904296875, + "learning_rate": 0.0011818821835132133, + "loss": 0.1021, + "step": 45746 + }, + { + "epoch": 0.39710592789993143, + "grad_norm": 0.275390625, + "learning_rate": 0.0011818525444455285, + "loss": 0.0947, + "step": 45747 + }, + { + "epoch": 0.39711460838013557, + "grad_norm": 0.1806640625, + "learning_rate": 0.0011818229052883314, + "loss": 0.0957, + "step": 45748 + }, + { + "epoch": 0.39712328886033976, + "grad_norm": 0.4375, + "learning_rate": 0.001181793266041654, + "loss": 0.1177, + "step": 45749 + }, + { + "epoch": 0.3971319693405439, + "grad_norm": 0.578125, + "learning_rate": 0.0011817636267055286, + "loss": 0.0908, + "step": 45750 + }, + { + "epoch": 0.3971406498207481, + "grad_norm": 0.55078125, + "learning_rate": 0.0011817339872799875, + "loss": 0.0869, + "step": 45751 + }, + { + "epoch": 0.39714933030095223, + "grad_norm": 0.201171875, + "learning_rate": 0.0011817043477650635, + "loss": 0.1777, + "step": 45752 + }, + { + "epoch": 0.3971580107811564, + "grad_norm": 0.1591796875, + "learning_rate": 0.001181674708160789, + "loss": 0.061, + "step": 45753 + }, + { + "epoch": 0.39716669126136056, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011816450684671962, + "loss": 0.1162, + "step": 45754 + }, + { + "epoch": 0.39717537174156475, + "grad_norm": 0.162109375, + "learning_rate": 0.0011816154286843174, + "loss": 0.0693, + "step": 45755 + }, + { + "epoch": 0.3971840522217689, + "grad_norm": 0.38671875, + "learning_rate": 0.0011815857888121855, + "loss": 0.1216, + "step": 45756 + }, + { + "epoch": 0.3971927327019731, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011815561488508323, + "loss": 0.0869, + "step": 45757 + }, + { + "epoch": 0.3972014131821772, + "grad_norm": 0.1875, + "learning_rate": 0.001181526508800291, + "loss": 0.1426, + "step": 45758 + }, + { + "epoch": 0.3972100936623814, + "grad_norm": 0.4375, + "learning_rate": 0.001181496868660593, + "loss": 0.1475, + "step": 45759 + }, + { + "epoch": 0.39721877414258555, + "grad_norm": 0.3203125, + "learning_rate": 0.0011814672284317718, + "loss": 0.1445, + "step": 45760 + }, + { + "epoch": 0.39722745462278974, + "grad_norm": 0.58203125, + "learning_rate": 0.0011814375881138586, + "loss": 0.1133, + "step": 45761 + }, + { + "epoch": 0.3972361351029939, + "grad_norm": 0.3984375, + "learning_rate": 0.001181407947706887, + "loss": 0.0874, + "step": 45762 + }, + { + "epoch": 0.3972448155831981, + "grad_norm": 0.5859375, + "learning_rate": 0.0011813783072108887, + "loss": 0.1387, + "step": 45763 + }, + { + "epoch": 0.3972534960634022, + "grad_norm": 0.1650390625, + "learning_rate": 0.0011813486666258962, + "loss": 0.0908, + "step": 45764 + }, + { + "epoch": 0.3972621765436064, + "grad_norm": 0.66015625, + "learning_rate": 0.0011813190259519419, + "loss": 0.0889, + "step": 45765 + }, + { + "epoch": 0.39727085702381054, + "grad_norm": 0.310546875, + "learning_rate": 0.001181289385189059, + "loss": 0.0933, + "step": 45766 + }, + { + "epoch": 0.39727953750401473, + "grad_norm": 0.09765625, + "learning_rate": 0.0011812597443372785, + "loss": 0.0938, + "step": 45767 + }, + { + "epoch": 0.39728821798421887, + "grad_norm": 0.46875, + "learning_rate": 0.001181230103396634, + "loss": 0.2168, + "step": 45768 + }, + { + "epoch": 0.39729689846442307, + "grad_norm": 0.44921875, + "learning_rate": 0.0011812004623671575, + "loss": 0.1113, + "step": 45769 + }, + { + "epoch": 0.3973055789446272, + "grad_norm": 0.111328125, + "learning_rate": 0.001181170821248881, + "loss": 0.0608, + "step": 45770 + }, + { + "epoch": 0.3973142594248314, + "grad_norm": 0.2333984375, + "learning_rate": 0.0011811411800418377, + "loss": 0.127, + "step": 45771 + }, + { + "epoch": 0.39732293990503553, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011811115387460594, + "loss": 0.1387, + "step": 45772 + }, + { + "epoch": 0.3973316203852397, + "grad_norm": 0.54296875, + "learning_rate": 0.0011810818973615788, + "loss": 0.0757, + "step": 45773 + }, + { + "epoch": 0.39734030086544386, + "grad_norm": 0.34765625, + "learning_rate": 0.0011810522558884282, + "loss": 0.0947, + "step": 45774 + }, + { + "epoch": 0.39734898134564806, + "grad_norm": 0.671875, + "learning_rate": 0.0011810226143266404, + "loss": 0.0835, + "step": 45775 + }, + { + "epoch": 0.3973576618258522, + "grad_norm": 0.30078125, + "learning_rate": 0.0011809929726762471, + "loss": 0.0737, + "step": 45776 + }, + { + "epoch": 0.3973663423060564, + "grad_norm": 0.19921875, + "learning_rate": 0.0011809633309372814, + "loss": 0.125, + "step": 45777 + }, + { + "epoch": 0.3973750227862605, + "grad_norm": 0.2392578125, + "learning_rate": 0.001180933689109775, + "loss": 0.1152, + "step": 45778 + }, + { + "epoch": 0.3973837032664647, + "grad_norm": 0.53515625, + "learning_rate": 0.001180904047193761, + "loss": 0.0693, + "step": 45779 + }, + { + "epoch": 0.39739238374666885, + "grad_norm": 0.32421875, + "learning_rate": 0.0011808744051892716, + "loss": 0.1006, + "step": 45780 + }, + { + "epoch": 0.39740106422687305, + "grad_norm": 0.1474609375, + "learning_rate": 0.001180844763096339, + "loss": 0.0977, + "step": 45781 + }, + { + "epoch": 0.3974097447070772, + "grad_norm": 0.4609375, + "learning_rate": 0.0011808151209149959, + "loss": 0.0898, + "step": 45782 + }, + { + "epoch": 0.3974184251872814, + "grad_norm": 0.283203125, + "learning_rate": 0.0011807854786452748, + "loss": 0.1011, + "step": 45783 + }, + { + "epoch": 0.3974271056674855, + "grad_norm": 0.482421875, + "learning_rate": 0.0011807558362872076, + "loss": 0.1221, + "step": 45784 + }, + { + "epoch": 0.3974357861476897, + "grad_norm": 0.13671875, + "learning_rate": 0.0011807261938408273, + "loss": 0.0903, + "step": 45785 + }, + { + "epoch": 0.39744446662789384, + "grad_norm": 0.53515625, + "learning_rate": 0.0011806965513061656, + "loss": 0.1484, + "step": 45786 + }, + { + "epoch": 0.39745314710809804, + "grad_norm": 0.2001953125, + "learning_rate": 0.0011806669086832557, + "loss": 0.1465, + "step": 45787 + }, + { + "epoch": 0.3974618275883022, + "grad_norm": 0.9375, + "learning_rate": 0.0011806372659721297, + "loss": 0.1328, + "step": 45788 + }, + { + "epoch": 0.39747050806850637, + "grad_norm": 0.53515625, + "learning_rate": 0.00118060762317282, + "loss": 0.1001, + "step": 45789 + }, + { + "epoch": 0.3974791885487105, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011805779802853587, + "loss": 0.106, + "step": 45790 + }, + { + "epoch": 0.3974878690289147, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011805483373097789, + "loss": 0.1074, + "step": 45791 + }, + { + "epoch": 0.39749654950911883, + "grad_norm": 0.08154296875, + "learning_rate": 0.0011805186942461126, + "loss": 0.085, + "step": 45792 + }, + { + "epoch": 0.39750522998932303, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011804890510943922, + "loss": 0.0967, + "step": 45793 + }, + { + "epoch": 0.39751391046952717, + "grad_norm": 0.423828125, + "learning_rate": 0.00118045940785465, + "loss": 0.1318, + "step": 45794 + }, + { + "epoch": 0.39752259094973136, + "grad_norm": 0.16796875, + "learning_rate": 0.0011804297645269188, + "loss": 0.1055, + "step": 45795 + }, + { + "epoch": 0.3975312714299355, + "grad_norm": 0.51171875, + "learning_rate": 0.0011804001211112307, + "loss": 0.1279, + "step": 45796 + }, + { + "epoch": 0.3975399519101397, + "grad_norm": 0.376953125, + "learning_rate": 0.0011803704776076186, + "loss": 0.0986, + "step": 45797 + }, + { + "epoch": 0.3975486323903438, + "grad_norm": 0.13671875, + "learning_rate": 0.0011803408340161142, + "loss": 0.1006, + "step": 45798 + }, + { + "epoch": 0.397557312870548, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011803111903367503, + "loss": 0.0767, + "step": 45799 + }, + { + "epoch": 0.39756599335075216, + "grad_norm": 0.19140625, + "learning_rate": 0.0011802815465695594, + "loss": 0.1211, + "step": 45800 + }, + { + "epoch": 0.39757467383095635, + "grad_norm": 0.5703125, + "learning_rate": 0.0011802519027145737, + "loss": 0.1348, + "step": 45801 + }, + { + "epoch": 0.3975833543111605, + "grad_norm": 0.2890625, + "learning_rate": 0.0011802222587718259, + "loss": 0.1162, + "step": 45802 + }, + { + "epoch": 0.3975920347913647, + "grad_norm": 0.15234375, + "learning_rate": 0.0011801926147413477, + "loss": 0.0913, + "step": 45803 + }, + { + "epoch": 0.3976007152715688, + "grad_norm": 0.59375, + "learning_rate": 0.0011801629706231728, + "loss": 0.1006, + "step": 45804 + }, + { + "epoch": 0.397609395751773, + "grad_norm": 0.478515625, + "learning_rate": 0.0011801333264173325, + "loss": 0.0898, + "step": 45805 + }, + { + "epoch": 0.39761807623197715, + "grad_norm": 0.5859375, + "learning_rate": 0.0011801036821238596, + "loss": 0.1191, + "step": 45806 + }, + { + "epoch": 0.39762675671218134, + "grad_norm": 0.578125, + "learning_rate": 0.0011800740377427865, + "loss": 0.0742, + "step": 45807 + }, + { + "epoch": 0.3976354371923855, + "grad_norm": 0.328125, + "learning_rate": 0.0011800443932741458, + "loss": 0.0869, + "step": 45808 + }, + { + "epoch": 0.39764411767258967, + "grad_norm": 0.57421875, + "learning_rate": 0.0011800147487179696, + "loss": 0.0996, + "step": 45809 + }, + { + "epoch": 0.3976527981527938, + "grad_norm": 0.345703125, + "learning_rate": 0.0011799851040742903, + "loss": 0.084, + "step": 45810 + }, + { + "epoch": 0.397661478632998, + "grad_norm": 0.373046875, + "learning_rate": 0.0011799554593431407, + "loss": 0.124, + "step": 45811 + }, + { + "epoch": 0.39767015911320214, + "grad_norm": 0.51953125, + "learning_rate": 0.0011799258145245531, + "loss": 0.0952, + "step": 45812 + }, + { + "epoch": 0.39767883959340633, + "grad_norm": 0.1435546875, + "learning_rate": 0.00117989616961856, + "loss": 0.0991, + "step": 45813 + }, + { + "epoch": 0.39768752007361047, + "grad_norm": 0.43359375, + "learning_rate": 0.0011798665246251934, + "loss": 0.0928, + "step": 45814 + }, + { + "epoch": 0.39769620055381466, + "grad_norm": 0.796875, + "learning_rate": 0.001179836879544486, + "loss": 0.1445, + "step": 45815 + }, + { + "epoch": 0.3977048810340188, + "grad_norm": 0.11328125, + "learning_rate": 0.0011798072343764698, + "loss": 0.0894, + "step": 45816 + }, + { + "epoch": 0.397713561514223, + "grad_norm": 0.765625, + "learning_rate": 0.0011797775891211782, + "loss": 0.1045, + "step": 45817 + }, + { + "epoch": 0.39772224199442713, + "grad_norm": 0.6640625, + "learning_rate": 0.0011797479437786428, + "loss": 0.209, + "step": 45818 + }, + { + "epoch": 0.3977309224746313, + "grad_norm": 0.32421875, + "learning_rate": 0.0011797182983488963, + "loss": 0.1157, + "step": 45819 + }, + { + "epoch": 0.39773960295483546, + "grad_norm": 0.26171875, + "learning_rate": 0.0011796886528319707, + "loss": 0.1064, + "step": 45820 + }, + { + "epoch": 0.39774828343503965, + "grad_norm": 0.400390625, + "learning_rate": 0.0011796590072278993, + "loss": 0.0972, + "step": 45821 + }, + { + "epoch": 0.3977569639152438, + "grad_norm": 0.271484375, + "learning_rate": 0.001179629361536714, + "loss": 0.1338, + "step": 45822 + }, + { + "epoch": 0.397765644395448, + "grad_norm": 0.1552734375, + "learning_rate": 0.0011795997157584468, + "loss": 0.1934, + "step": 45823 + }, + { + "epoch": 0.3977743248756521, + "grad_norm": 0.53515625, + "learning_rate": 0.0011795700698931308, + "loss": 0.085, + "step": 45824 + }, + { + "epoch": 0.3977830053558563, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011795404239407982, + "loss": 0.0928, + "step": 45825 + }, + { + "epoch": 0.39779168583606045, + "grad_norm": 0.11181640625, + "learning_rate": 0.0011795107779014812, + "loss": 0.0737, + "step": 45826 + }, + { + "epoch": 0.39780036631626464, + "grad_norm": 0.71875, + "learning_rate": 0.0011794811317752127, + "loss": 0.0894, + "step": 45827 + }, + { + "epoch": 0.3978090467964688, + "grad_norm": 0.6015625, + "learning_rate": 0.0011794514855620246, + "loss": 0.1006, + "step": 45828 + }, + { + "epoch": 0.39781772727667297, + "grad_norm": 0.52734375, + "learning_rate": 0.0011794218392619496, + "loss": 0.1138, + "step": 45829 + }, + { + "epoch": 0.3978264077568771, + "grad_norm": 1.0546875, + "learning_rate": 0.0011793921928750202, + "loss": 0.1118, + "step": 45830 + }, + { + "epoch": 0.3978350882370813, + "grad_norm": 0.2099609375, + "learning_rate": 0.0011793625464012686, + "loss": 0.1113, + "step": 45831 + }, + { + "epoch": 0.39784376871728544, + "grad_norm": 0.11962890625, + "learning_rate": 0.0011793328998407273, + "loss": 0.083, + "step": 45832 + }, + { + "epoch": 0.39785244919748963, + "grad_norm": 0.201171875, + "learning_rate": 0.0011793032531934287, + "loss": 0.104, + "step": 45833 + }, + { + "epoch": 0.39786112967769377, + "grad_norm": 0.314453125, + "learning_rate": 0.0011792736064594055, + "loss": 0.124, + "step": 45834 + }, + { + "epoch": 0.3978698101578979, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011792439596386894, + "loss": 0.1504, + "step": 45835 + }, + { + "epoch": 0.3978784906381021, + "grad_norm": 0.0966796875, + "learning_rate": 0.0011792143127313138, + "loss": 0.105, + "step": 45836 + }, + { + "epoch": 0.39788717111830624, + "grad_norm": 0.625, + "learning_rate": 0.0011791846657373101, + "loss": 0.1089, + "step": 45837 + }, + { + "epoch": 0.39789585159851043, + "grad_norm": 0.62109375, + "learning_rate": 0.0011791550186567115, + "loss": 0.1396, + "step": 45838 + }, + { + "epoch": 0.39790453207871457, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011791253714895503, + "loss": 0.0591, + "step": 45839 + }, + { + "epoch": 0.39791321255891876, + "grad_norm": 0.208984375, + "learning_rate": 0.0011790957242358584, + "loss": 0.1328, + "step": 45840 + }, + { + "epoch": 0.3979218930391229, + "grad_norm": 0.109375, + "learning_rate": 0.0011790660768956692, + "loss": 0.0815, + "step": 45841 + }, + { + "epoch": 0.3979305735193271, + "grad_norm": 0.52734375, + "learning_rate": 0.0011790364294690138, + "loss": 0.0898, + "step": 45842 + }, + { + "epoch": 0.39793925399953123, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011790067819559258, + "loss": 0.1143, + "step": 45843 + }, + { + "epoch": 0.3979479344797354, + "grad_norm": 0.1171875, + "learning_rate": 0.0011789771343564373, + "loss": 0.0786, + "step": 45844 + }, + { + "epoch": 0.39795661495993956, + "grad_norm": 0.51171875, + "learning_rate": 0.0011789474866705802, + "loss": 0.1143, + "step": 45845 + }, + { + "epoch": 0.39796529544014375, + "grad_norm": 0.6640625, + "learning_rate": 0.0011789178388983876, + "loss": 0.1357, + "step": 45846 + }, + { + "epoch": 0.3979739759203479, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011788881910398915, + "loss": 0.1133, + "step": 45847 + }, + { + "epoch": 0.3979826564005521, + "grad_norm": 2.03125, + "learning_rate": 0.0011788585430951245, + "loss": 0.0977, + "step": 45848 + }, + { + "epoch": 0.3979913368807562, + "grad_norm": 0.203125, + "learning_rate": 0.0011788288950641188, + "loss": 0.1309, + "step": 45849 + }, + { + "epoch": 0.3980000173609604, + "grad_norm": 0.1962890625, + "learning_rate": 0.0011787992469469073, + "loss": 0.105, + "step": 45850 + }, + { + "epoch": 0.39800869784116455, + "grad_norm": 0.2373046875, + "learning_rate": 0.001178769598743522, + "loss": 0.1318, + "step": 45851 + }, + { + "epoch": 0.39801737832136874, + "grad_norm": 0.15234375, + "learning_rate": 0.0011787399504539955, + "loss": 0.1279, + "step": 45852 + }, + { + "epoch": 0.3980260588015729, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011787103020783603, + "loss": 0.1172, + "step": 45853 + }, + { + "epoch": 0.3980347392817771, + "grad_norm": 0.80859375, + "learning_rate": 0.0011786806536166485, + "loss": 0.1128, + "step": 45854 + }, + { + "epoch": 0.3980434197619812, + "grad_norm": 0.10009765625, + "learning_rate": 0.0011786510050688925, + "loss": 0.1152, + "step": 45855 + }, + { + "epoch": 0.3980521002421854, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011786213564351255, + "loss": 0.1572, + "step": 45856 + }, + { + "epoch": 0.39806078072238954, + "grad_norm": 0.7265625, + "learning_rate": 0.001178591707715379, + "loss": 0.0957, + "step": 45857 + }, + { + "epoch": 0.39806946120259373, + "grad_norm": 0.1337890625, + "learning_rate": 0.001178562058909686, + "loss": 0.1016, + "step": 45858 + }, + { + "epoch": 0.39807814168279787, + "grad_norm": 0.0927734375, + "learning_rate": 0.0011785324100180782, + "loss": 0.125, + "step": 45859 + }, + { + "epoch": 0.39808682216300206, + "grad_norm": 0.361328125, + "learning_rate": 0.0011785027610405892, + "loss": 0.0928, + "step": 45860 + }, + { + "epoch": 0.3980955026432062, + "grad_norm": 0.341796875, + "learning_rate": 0.0011784731119772506, + "loss": 0.0776, + "step": 45861 + }, + { + "epoch": 0.3981041831234104, + "grad_norm": 1.046875, + "learning_rate": 0.0011784434628280946, + "loss": 0.1035, + "step": 45862 + }, + { + "epoch": 0.39811286360361453, + "grad_norm": 0.515625, + "learning_rate": 0.0011784138135931541, + "loss": 0.0767, + "step": 45863 + }, + { + "epoch": 0.3981215440838187, + "grad_norm": 0.26171875, + "learning_rate": 0.0011783841642724617, + "loss": 0.0879, + "step": 45864 + }, + { + "epoch": 0.39813022456402286, + "grad_norm": 0.34765625, + "learning_rate": 0.0011783545148660495, + "loss": 0.1162, + "step": 45865 + }, + { + "epoch": 0.39813890504422705, + "grad_norm": 0.142578125, + "learning_rate": 0.0011783248653739497, + "loss": 0.1064, + "step": 45866 + }, + { + "epoch": 0.3981475855244312, + "grad_norm": 0.369140625, + "learning_rate": 0.0011782952157961956, + "loss": 0.1465, + "step": 45867 + }, + { + "epoch": 0.3981562660046354, + "grad_norm": 0.076171875, + "learning_rate": 0.0011782655661328183, + "loss": 0.0898, + "step": 45868 + }, + { + "epoch": 0.3981649464848395, + "grad_norm": 0.64453125, + "learning_rate": 0.0011782359163838513, + "loss": 0.1191, + "step": 45869 + }, + { + "epoch": 0.3981736269650437, + "grad_norm": 0.353515625, + "learning_rate": 0.0011782062665493267, + "loss": 0.1016, + "step": 45870 + }, + { + "epoch": 0.39818230744524785, + "grad_norm": 0.130859375, + "learning_rate": 0.0011781766166292771, + "loss": 0.0977, + "step": 45871 + }, + { + "epoch": 0.39819098792545204, + "grad_norm": 1.1875, + "learning_rate": 0.0011781469666237345, + "loss": 0.1089, + "step": 45872 + }, + { + "epoch": 0.3981996684056562, + "grad_norm": 0.5234375, + "learning_rate": 0.0011781173165327315, + "loss": 0.1104, + "step": 45873 + }, + { + "epoch": 0.3982083488858604, + "grad_norm": 0.1943359375, + "learning_rate": 0.0011780876663563006, + "loss": 0.1011, + "step": 45874 + }, + { + "epoch": 0.3982170293660645, + "grad_norm": 0.3203125, + "learning_rate": 0.0011780580160944743, + "loss": 0.1035, + "step": 45875 + }, + { + "epoch": 0.3982257098462687, + "grad_norm": 0.166015625, + "learning_rate": 0.001178028365747285, + "loss": 0.126, + "step": 45876 + }, + { + "epoch": 0.39823439032647284, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011779987153147646, + "loss": 0.0981, + "step": 45877 + }, + { + "epoch": 0.39824307080667704, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011779690647969465, + "loss": 0.0947, + "step": 45878 + }, + { + "epoch": 0.3982517512868812, + "grad_norm": 0.451171875, + "learning_rate": 0.0011779394141938624, + "loss": 0.1147, + "step": 45879 + }, + { + "epoch": 0.39826043176708537, + "grad_norm": 0.5625, + "learning_rate": 0.0011779097635055452, + "loss": 0.1221, + "step": 45880 + }, + { + "epoch": 0.3982691122472895, + "grad_norm": 0.671875, + "learning_rate": 0.0011778801127320265, + "loss": 0.1367, + "step": 45881 + }, + { + "epoch": 0.3982777927274937, + "grad_norm": 0.1875, + "learning_rate": 0.0011778504618733397, + "loss": 0.123, + "step": 45882 + }, + { + "epoch": 0.39828647320769783, + "grad_norm": 0.12890625, + "learning_rate": 0.0011778208109295165, + "loss": 0.1055, + "step": 45883 + }, + { + "epoch": 0.398295153687902, + "grad_norm": 0.1416015625, + "learning_rate": 0.0011777911599005902, + "loss": 0.1167, + "step": 45884 + }, + { + "epoch": 0.39830383416810616, + "grad_norm": 0.1611328125, + "learning_rate": 0.0011777615087865921, + "loss": 0.1177, + "step": 45885 + }, + { + "epoch": 0.39831251464831036, + "grad_norm": 0.11083984375, + "learning_rate": 0.001177731857587555, + "loss": 0.1035, + "step": 45886 + }, + { + "epoch": 0.3983211951285145, + "grad_norm": 0.11669921875, + "learning_rate": 0.001177702206303512, + "loss": 0.0957, + "step": 45887 + }, + { + "epoch": 0.3983298756087187, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011776725549344949, + "loss": 0.0952, + "step": 45888 + }, + { + "epoch": 0.3983385560889228, + "grad_norm": 0.11474609375, + "learning_rate": 0.001177642903480536, + "loss": 0.1123, + "step": 45889 + }, + { + "epoch": 0.398347236569127, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011776132519416681, + "loss": 0.1143, + "step": 45890 + }, + { + "epoch": 0.39835591704933115, + "grad_norm": 2.484375, + "learning_rate": 0.001177583600317924, + "loss": 0.3984, + "step": 45891 + }, + { + "epoch": 0.39836459752953535, + "grad_norm": 0.283203125, + "learning_rate": 0.0011775539486093352, + "loss": 0.061, + "step": 45892 + }, + { + "epoch": 0.3983732780097395, + "grad_norm": 0.228515625, + "learning_rate": 0.0011775242968159348, + "loss": 0.0659, + "step": 45893 + }, + { + "epoch": 0.3983819584899437, + "grad_norm": 0.392578125, + "learning_rate": 0.0011774946449377547, + "loss": 0.1328, + "step": 45894 + }, + { + "epoch": 0.3983906389701478, + "grad_norm": 0.298828125, + "learning_rate": 0.0011774649929748278, + "loss": 0.0908, + "step": 45895 + }, + { + "epoch": 0.398399319450352, + "grad_norm": 0.1611328125, + "learning_rate": 0.0011774353409271863, + "loss": 0.1299, + "step": 45896 + }, + { + "epoch": 0.39840799993055614, + "grad_norm": 0.9296875, + "learning_rate": 0.0011774056887948628, + "loss": 0.0908, + "step": 45897 + }, + { + "epoch": 0.39841668041076034, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011773760365778893, + "loss": 0.1182, + "step": 45898 + }, + { + "epoch": 0.3984253608909645, + "grad_norm": 0.111328125, + "learning_rate": 0.0011773463842762988, + "loss": 0.0791, + "step": 45899 + }, + { + "epoch": 0.39843404137116867, + "grad_norm": 0.13671875, + "learning_rate": 0.0011773167318901232, + "loss": 0.085, + "step": 45900 + }, + { + "epoch": 0.3984427218513728, + "grad_norm": 0.6328125, + "learning_rate": 0.0011772870794193955, + "loss": 0.1099, + "step": 45901 + }, + { + "epoch": 0.398451402331577, + "grad_norm": 0.1787109375, + "learning_rate": 0.0011772574268641477, + "loss": 0.1074, + "step": 45902 + }, + { + "epoch": 0.39846008281178114, + "grad_norm": 0.5625, + "learning_rate": 0.001177227774224412, + "loss": 0.0996, + "step": 45903 + }, + { + "epoch": 0.39846876329198533, + "grad_norm": 0.2431640625, + "learning_rate": 0.0011771981215002217, + "loss": 0.1289, + "step": 45904 + }, + { + "epoch": 0.39847744377218947, + "grad_norm": 0.48828125, + "learning_rate": 0.0011771684686916084, + "loss": 0.1094, + "step": 45905 + }, + { + "epoch": 0.39848612425239366, + "grad_norm": 0.3125, + "learning_rate": 0.001177138815798605, + "loss": 0.1084, + "step": 45906 + }, + { + "epoch": 0.3984948047325978, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011771091628212437, + "loss": 0.1011, + "step": 45907 + }, + { + "epoch": 0.398503485212802, + "grad_norm": 1.765625, + "learning_rate": 0.0011770795097595568, + "loss": 0.3047, + "step": 45908 + }, + { + "epoch": 0.3985121656930061, + "grad_norm": 0.7265625, + "learning_rate": 0.001177049856613577, + "loss": 0.0923, + "step": 45909 + }, + { + "epoch": 0.3985208461732103, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011770202033833368, + "loss": 0.1201, + "step": 45910 + }, + { + "epoch": 0.39852952665341446, + "grad_norm": 0.265625, + "learning_rate": 0.001176990550068868, + "loss": 0.1318, + "step": 45911 + }, + { + "epoch": 0.39853820713361865, + "grad_norm": 0.2421875, + "learning_rate": 0.001176960896670204, + "loss": 0.127, + "step": 45912 + }, + { + "epoch": 0.3985468876138228, + "grad_norm": 0.451171875, + "learning_rate": 0.0011769312431873767, + "loss": 0.0786, + "step": 45913 + }, + { + "epoch": 0.398555568094027, + "grad_norm": 0.384765625, + "learning_rate": 0.001176901589620418, + "loss": 0.1016, + "step": 45914 + }, + { + "epoch": 0.3985642485742311, + "grad_norm": 0.1787109375, + "learning_rate": 0.0011768719359693614, + "loss": 0.0942, + "step": 45915 + }, + { + "epoch": 0.3985729290544353, + "grad_norm": 0.0751953125, + "learning_rate": 0.0011768422822342386, + "loss": 0.1074, + "step": 45916 + }, + { + "epoch": 0.39858160953463945, + "grad_norm": 0.38671875, + "learning_rate": 0.0011768126284150825, + "loss": 0.1245, + "step": 45917 + }, + { + "epoch": 0.39859029001484364, + "grad_norm": 0.216796875, + "learning_rate": 0.0011767829745119248, + "loss": 0.0654, + "step": 45918 + }, + { + "epoch": 0.3985989704950478, + "grad_norm": 0.365234375, + "learning_rate": 0.001176753320524799, + "loss": 0.0977, + "step": 45919 + }, + { + "epoch": 0.39860765097525197, + "grad_norm": 0.07861328125, + "learning_rate": 0.0011767236664537361, + "loss": 0.082, + "step": 45920 + }, + { + "epoch": 0.3986163314554561, + "grad_norm": 0.279296875, + "learning_rate": 0.0011766940122987701, + "loss": 0.1104, + "step": 45921 + }, + { + "epoch": 0.3986250119356603, + "grad_norm": 0.67578125, + "learning_rate": 0.0011766643580599324, + "loss": 0.0981, + "step": 45922 + }, + { + "epoch": 0.39863369241586444, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011766347037372555, + "loss": 0.1426, + "step": 45923 + }, + { + "epoch": 0.39864237289606863, + "grad_norm": 0.1474609375, + "learning_rate": 0.0011766050493307726, + "loss": 0.0752, + "step": 45924 + }, + { + "epoch": 0.39865105337627277, + "grad_norm": 0.255859375, + "learning_rate": 0.001176575394840515, + "loss": 0.1196, + "step": 45925 + }, + { + "epoch": 0.39865973385647696, + "grad_norm": 0.734375, + "learning_rate": 0.001176545740266516, + "loss": 0.1143, + "step": 45926 + }, + { + "epoch": 0.3986684143366811, + "grad_norm": 0.349609375, + "learning_rate": 0.0011765160856088076, + "loss": 0.0845, + "step": 45927 + }, + { + "epoch": 0.3986770948168853, + "grad_norm": 0.0966796875, + "learning_rate": 0.0011764864308674223, + "loss": 0.1318, + "step": 45928 + }, + { + "epoch": 0.39868577529708943, + "grad_norm": 0.458984375, + "learning_rate": 0.0011764567760423929, + "loss": 0.085, + "step": 45929 + }, + { + "epoch": 0.3986944557772936, + "grad_norm": 0.1748046875, + "learning_rate": 0.0011764271211337513, + "loss": 0.0923, + "step": 45930 + }, + { + "epoch": 0.39870313625749776, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011763974661415301, + "loss": 0.0923, + "step": 45931 + }, + { + "epoch": 0.39871181673770195, + "grad_norm": 0.1533203125, + "learning_rate": 0.0011763678110657619, + "loss": 0.0986, + "step": 45932 + }, + { + "epoch": 0.3987204972179061, + "grad_norm": 0.11474609375, + "learning_rate": 0.001176338155906479, + "loss": 0.0571, + "step": 45933 + }, + { + "epoch": 0.3987291776981103, + "grad_norm": 0.2216796875, + "learning_rate": 0.0011763085006637137, + "loss": 0.0796, + "step": 45934 + }, + { + "epoch": 0.3987378581783144, + "grad_norm": 0.1103515625, + "learning_rate": 0.0011762788453374986, + "loss": 0.0874, + "step": 45935 + }, + { + "epoch": 0.3987465386585186, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011762491899278663, + "loss": 0.0884, + "step": 45936 + }, + { + "epoch": 0.39875521913872275, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011762195344348489, + "loss": 0.166, + "step": 45937 + }, + { + "epoch": 0.39876389961892694, + "grad_norm": 0.349609375, + "learning_rate": 0.0011761898788584788, + "loss": 0.1006, + "step": 45938 + }, + { + "epoch": 0.3987725800991311, + "grad_norm": 0.10107421875, + "learning_rate": 0.001176160223198789, + "loss": 0.0996, + "step": 45939 + }, + { + "epoch": 0.3987812605793353, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011761305674558111, + "loss": 0.1001, + "step": 45940 + }, + { + "epoch": 0.3987899410595394, + "grad_norm": 0.81640625, + "learning_rate": 0.001176100911629578, + "loss": 0.1187, + "step": 45941 + }, + { + "epoch": 0.3987986215397436, + "grad_norm": 0.326171875, + "learning_rate": 0.0011760712557201223, + "loss": 0.0845, + "step": 45942 + }, + { + "epoch": 0.39880730201994774, + "grad_norm": 0.173828125, + "learning_rate": 0.001176041599727476, + "loss": 0.103, + "step": 45943 + }, + { + "epoch": 0.39881598250015193, + "grad_norm": 0.099609375, + "learning_rate": 0.001176011943651672, + "loss": 0.082, + "step": 45944 + }, + { + "epoch": 0.39882466298035607, + "grad_norm": 0.31640625, + "learning_rate": 0.0011759822874927422, + "loss": 0.1006, + "step": 45945 + }, + { + "epoch": 0.39883334346056026, + "grad_norm": 0.279296875, + "learning_rate": 0.0011759526312507195, + "loss": 0.1357, + "step": 45946 + }, + { + "epoch": 0.3988420239407644, + "grad_norm": 0.380859375, + "learning_rate": 0.0011759229749256361, + "loss": 0.0972, + "step": 45947 + }, + { + "epoch": 0.3988507044209686, + "grad_norm": 0.291015625, + "learning_rate": 0.001175893318517525, + "loss": 0.1016, + "step": 45948 + }, + { + "epoch": 0.39885938490117273, + "grad_norm": 0.5546875, + "learning_rate": 0.0011758636620264173, + "loss": 0.1172, + "step": 45949 + }, + { + "epoch": 0.3988680653813769, + "grad_norm": 0.087890625, + "learning_rate": 0.0011758340054523464, + "loss": 0.0757, + "step": 45950 + }, + { + "epoch": 0.39887674586158106, + "grad_norm": 0.1357421875, + "learning_rate": 0.001175804348795345, + "loss": 0.1035, + "step": 45951 + }, + { + "epoch": 0.39888542634178525, + "grad_norm": 0.38671875, + "learning_rate": 0.0011757746920554447, + "loss": 0.1572, + "step": 45952 + }, + { + "epoch": 0.3988941068219894, + "grad_norm": 0.25, + "learning_rate": 0.0011757450352326785, + "loss": 0.123, + "step": 45953 + }, + { + "epoch": 0.3989027873021936, + "grad_norm": 0.095703125, + "learning_rate": 0.0011757153783270784, + "loss": 0.1006, + "step": 45954 + }, + { + "epoch": 0.3989114677823977, + "grad_norm": 0.47265625, + "learning_rate": 0.0011756857213386773, + "loss": 0.104, + "step": 45955 + }, + { + "epoch": 0.3989201482626019, + "grad_norm": 0.109375, + "learning_rate": 0.0011756560642675075, + "loss": 0.1191, + "step": 45956 + }, + { + "epoch": 0.39892882874280605, + "grad_norm": 0.27734375, + "learning_rate": 0.0011756264071136017, + "loss": 0.1108, + "step": 45957 + }, + { + "epoch": 0.3989375092230102, + "grad_norm": 0.146484375, + "learning_rate": 0.0011755967498769913, + "loss": 0.1084, + "step": 45958 + }, + { + "epoch": 0.3989461897032144, + "grad_norm": 0.244140625, + "learning_rate": 0.0011755670925577097, + "loss": 0.1118, + "step": 45959 + }, + { + "epoch": 0.3989548701834185, + "grad_norm": 0.7421875, + "learning_rate": 0.001175537435155789, + "loss": 0.1367, + "step": 45960 + }, + { + "epoch": 0.3989635506636227, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011755077776712621, + "loss": 0.1094, + "step": 45961 + }, + { + "epoch": 0.39897223114382685, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011754781201041605, + "loss": 0.083, + "step": 45962 + }, + { + "epoch": 0.39898091162403104, + "grad_norm": 0.26171875, + "learning_rate": 0.0011754484624545173, + "loss": 0.1172, + "step": 45963 + }, + { + "epoch": 0.3989895921042352, + "grad_norm": 0.1201171875, + "learning_rate": 0.0011754188047223648, + "loss": 0.1133, + "step": 45964 + }, + { + "epoch": 0.3989982725844394, + "grad_norm": 0.59375, + "learning_rate": 0.0011753891469077358, + "loss": 0.1455, + "step": 45965 + }, + { + "epoch": 0.3990069530646435, + "grad_norm": 0.263671875, + "learning_rate": 0.001175359489010662, + "loss": 0.1025, + "step": 45966 + }, + { + "epoch": 0.3990156335448477, + "grad_norm": 0.16015625, + "learning_rate": 0.0011753298310311763, + "loss": 0.1089, + "step": 45967 + }, + { + "epoch": 0.39902431402505184, + "grad_norm": 0.2109375, + "learning_rate": 0.0011753001729693112, + "loss": 0.1006, + "step": 45968 + }, + { + "epoch": 0.39903299450525603, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011752705148250985, + "loss": 0.0605, + "step": 45969 + }, + { + "epoch": 0.39904167498546017, + "grad_norm": 0.341796875, + "learning_rate": 0.0011752408565985717, + "loss": 0.1104, + "step": 45970 + }, + { + "epoch": 0.39905035546566436, + "grad_norm": 0.466796875, + "learning_rate": 0.001175211198289762, + "loss": 0.0977, + "step": 45971 + }, + { + "epoch": 0.3990590359458685, + "grad_norm": 0.10009765625, + "learning_rate": 0.0011751815398987028, + "loss": 0.127, + "step": 45972 + }, + { + "epoch": 0.3990677164260727, + "grad_norm": 0.75, + "learning_rate": 0.001175151881425426, + "loss": 0.1709, + "step": 45973 + }, + { + "epoch": 0.39907639690627683, + "grad_norm": 0.546875, + "learning_rate": 0.0011751222228699645, + "loss": 0.1123, + "step": 45974 + }, + { + "epoch": 0.399085077386481, + "grad_norm": 0.412109375, + "learning_rate": 0.0011750925642323505, + "loss": 0.0889, + "step": 45975 + }, + { + "epoch": 0.39909375786668516, + "grad_norm": 0.287109375, + "learning_rate": 0.001175062905512616, + "loss": 0.1543, + "step": 45976 + }, + { + "epoch": 0.39910243834688935, + "grad_norm": 0.4140625, + "learning_rate": 0.0011750332467107942, + "loss": 0.1387, + "step": 45977 + }, + { + "epoch": 0.3991111188270935, + "grad_norm": 0.19140625, + "learning_rate": 0.0011750035878269171, + "loss": 0.1562, + "step": 45978 + }, + { + "epoch": 0.3991197993072977, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011749739288610171, + "loss": 0.1006, + "step": 45979 + }, + { + "epoch": 0.3991284797875018, + "grad_norm": 0.255859375, + "learning_rate": 0.0011749442698131267, + "loss": 0.0845, + "step": 45980 + }, + { + "epoch": 0.399137160267706, + "grad_norm": 0.361328125, + "learning_rate": 0.0011749146106832787, + "loss": 0.1025, + "step": 45981 + }, + { + "epoch": 0.39914584074791015, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011748849514715046, + "loss": 0.0952, + "step": 45982 + }, + { + "epoch": 0.39915452122811435, + "grad_norm": 0.40234375, + "learning_rate": 0.001174855292177838, + "loss": 0.1318, + "step": 45983 + }, + { + "epoch": 0.3991632017083185, + "grad_norm": 0.84765625, + "learning_rate": 0.0011748256328023103, + "loss": 0.1416, + "step": 45984 + }, + { + "epoch": 0.3991718821885227, + "grad_norm": 0.314453125, + "learning_rate": 0.001174795973344955, + "loss": 0.1021, + "step": 45985 + }, + { + "epoch": 0.3991805626687268, + "grad_norm": 0.8359375, + "learning_rate": 0.0011747663138058037, + "loss": 0.1367, + "step": 45986 + }, + { + "epoch": 0.399189243148931, + "grad_norm": 0.57421875, + "learning_rate": 0.0011747366541848892, + "loss": 0.1191, + "step": 45987 + }, + { + "epoch": 0.39919792362913514, + "grad_norm": 0.50390625, + "learning_rate": 0.0011747069944822437, + "loss": 0.082, + "step": 45988 + }, + { + "epoch": 0.39920660410933934, + "grad_norm": 0.59375, + "learning_rate": 0.0011746773346978994, + "loss": 0.0898, + "step": 45989 + }, + { + "epoch": 0.3992152845895435, + "grad_norm": 0.4765625, + "learning_rate": 0.0011746476748318896, + "loss": 0.1152, + "step": 45990 + }, + { + "epoch": 0.39922396506974767, + "grad_norm": 0.19140625, + "learning_rate": 0.0011746180148842458, + "loss": 0.0884, + "step": 45991 + }, + { + "epoch": 0.3992326455499518, + "grad_norm": 0.2265625, + "learning_rate": 0.0011745883548550011, + "loss": 0.1152, + "step": 45992 + }, + { + "epoch": 0.399241326030156, + "grad_norm": 0.2001953125, + "learning_rate": 0.0011745586947441876, + "loss": 0.0972, + "step": 45993 + }, + { + "epoch": 0.39925000651036013, + "grad_norm": 1.2109375, + "learning_rate": 0.0011745290345518378, + "loss": 0.5625, + "step": 45994 + }, + { + "epoch": 0.3992586869905643, + "grad_norm": 0.296875, + "learning_rate": 0.0011744993742779845, + "loss": 0.0815, + "step": 45995 + }, + { + "epoch": 0.39926736747076846, + "grad_norm": 0.373046875, + "learning_rate": 0.0011744697139226595, + "loss": 0.2578, + "step": 45996 + }, + { + "epoch": 0.39927604795097266, + "grad_norm": 0.138671875, + "learning_rate": 0.0011744400534858957, + "loss": 0.1172, + "step": 45997 + }, + { + "epoch": 0.3992847284311768, + "grad_norm": 0.4765625, + "learning_rate": 0.0011744103929677253, + "loss": 0.0962, + "step": 45998 + }, + { + "epoch": 0.399293408911381, + "grad_norm": 0.140625, + "learning_rate": 0.0011743807323681807, + "loss": 0.1367, + "step": 45999 + }, + { + "epoch": 0.3993020893915851, + "grad_norm": 0.71875, + "learning_rate": 0.0011743510716872945, + "loss": 0.0859, + "step": 46000 + }, + { + "epoch": 0.3993107698717893, + "grad_norm": 0.6484375, + "learning_rate": 0.0011743214109250992, + "loss": 0.125, + "step": 46001 + }, + { + "epoch": 0.39931945035199345, + "grad_norm": 0.20703125, + "learning_rate": 0.001174291750081627, + "loss": 0.1055, + "step": 46002 + }, + { + "epoch": 0.39932813083219765, + "grad_norm": 0.396484375, + "learning_rate": 0.0011742620891569107, + "loss": 0.0918, + "step": 46003 + }, + { + "epoch": 0.3993368113124018, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011742324281509822, + "loss": 0.0918, + "step": 46004 + }, + { + "epoch": 0.399345491792606, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011742027670638745, + "loss": 0.1104, + "step": 46005 + }, + { + "epoch": 0.3993541722728101, + "grad_norm": 0.21484375, + "learning_rate": 0.0011741731058956195, + "loss": 0.1123, + "step": 46006 + }, + { + "epoch": 0.3993628527530143, + "grad_norm": 0.57421875, + "learning_rate": 0.00117414344464625, + "loss": 0.1089, + "step": 46007 + }, + { + "epoch": 0.39937153323321845, + "grad_norm": 0.279296875, + "learning_rate": 0.0011741137833157984, + "loss": 0.0981, + "step": 46008 + }, + { + "epoch": 0.39938021371342264, + "grad_norm": 0.79296875, + "learning_rate": 0.0011740841219042971, + "loss": 0.0869, + "step": 46009 + }, + { + "epoch": 0.3993888941936268, + "grad_norm": 0.1181640625, + "learning_rate": 0.0011740544604117783, + "loss": 0.0923, + "step": 46010 + }, + { + "epoch": 0.39939757467383097, + "grad_norm": 0.53125, + "learning_rate": 0.0011740247988382747, + "loss": 0.1221, + "step": 46011 + }, + { + "epoch": 0.3994062551540351, + "grad_norm": 0.7109375, + "learning_rate": 0.001173995137183819, + "loss": 0.1299, + "step": 46012 + }, + { + "epoch": 0.3994149356342393, + "grad_norm": 0.341796875, + "learning_rate": 0.001173965475448443, + "loss": 0.062, + "step": 46013 + }, + { + "epoch": 0.39942361611444344, + "grad_norm": 0.2490234375, + "learning_rate": 0.0011739358136321797, + "loss": 0.104, + "step": 46014 + }, + { + "epoch": 0.39943229659464763, + "grad_norm": 0.259765625, + "learning_rate": 0.001173906151735061, + "loss": 0.1367, + "step": 46015 + }, + { + "epoch": 0.39944097707485177, + "grad_norm": 1.1640625, + "learning_rate": 0.0011738764897571202, + "loss": 0.3691, + "step": 46016 + }, + { + "epoch": 0.39944965755505596, + "grad_norm": 0.1552734375, + "learning_rate": 0.0011738468276983887, + "loss": 0.0781, + "step": 46017 + }, + { + "epoch": 0.3994583380352601, + "grad_norm": 0.48828125, + "learning_rate": 0.0011738171655588997, + "loss": 0.1895, + "step": 46018 + }, + { + "epoch": 0.3994670185154643, + "grad_norm": 0.23828125, + "learning_rate": 0.0011737875033386849, + "loss": 0.1006, + "step": 46019 + }, + { + "epoch": 0.3994756989956684, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011737578410377773, + "loss": 0.0962, + "step": 46020 + }, + { + "epoch": 0.3994843794758726, + "grad_norm": 0.12109375, + "learning_rate": 0.0011737281786562098, + "loss": 0.0742, + "step": 46021 + }, + { + "epoch": 0.39949305995607676, + "grad_norm": 0.1328125, + "learning_rate": 0.0011736985161940137, + "loss": 0.0923, + "step": 46022 + }, + { + "epoch": 0.39950174043628095, + "grad_norm": 0.298828125, + "learning_rate": 0.0011736688536512223, + "loss": 0.0903, + "step": 46023 + }, + { + "epoch": 0.3995104209164851, + "grad_norm": 0.439453125, + "learning_rate": 0.0011736391910278675, + "loss": 0.0918, + "step": 46024 + }, + { + "epoch": 0.3995191013966893, + "grad_norm": 0.087890625, + "learning_rate": 0.0011736095283239826, + "loss": 0.0786, + "step": 46025 + }, + { + "epoch": 0.3995277818768934, + "grad_norm": 0.4140625, + "learning_rate": 0.0011735798655395989, + "loss": 0.0972, + "step": 46026 + }, + { + "epoch": 0.3995364623570976, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011735502026747496, + "loss": 0.084, + "step": 46027 + }, + { + "epoch": 0.39954514283730175, + "grad_norm": 0.1123046875, + "learning_rate": 0.0011735205397294664, + "loss": 0.082, + "step": 46028 + }, + { + "epoch": 0.39955382331750594, + "grad_norm": 0.21484375, + "learning_rate": 0.0011734908767037826, + "loss": 0.0918, + "step": 46029 + }, + { + "epoch": 0.3995625037977101, + "grad_norm": 0.2490234375, + "learning_rate": 0.0011734612135977302, + "loss": 0.1123, + "step": 46030 + }, + { + "epoch": 0.39957118427791427, + "grad_norm": 0.08984375, + "learning_rate": 0.0011734315504113419, + "loss": 0.0928, + "step": 46031 + }, + { + "epoch": 0.3995798647581184, + "grad_norm": 0.98828125, + "learning_rate": 0.00117340188714465, + "loss": 0.2891, + "step": 46032 + }, + { + "epoch": 0.3995885452383226, + "grad_norm": 0.2158203125, + "learning_rate": 0.0011733722237976865, + "loss": 0.1182, + "step": 46033 + }, + { + "epoch": 0.39959722571852674, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011733425603704846, + "loss": 0.1006, + "step": 46034 + }, + { + "epoch": 0.39960590619873093, + "grad_norm": 0.267578125, + "learning_rate": 0.0011733128968630763, + "loss": 0.0732, + "step": 46035 + }, + { + "epoch": 0.39961458667893507, + "grad_norm": 0.154296875, + "learning_rate": 0.0011732832332754942, + "loss": 0.1055, + "step": 46036 + }, + { + "epoch": 0.39962326715913926, + "grad_norm": 0.2275390625, + "learning_rate": 0.0011732535696077704, + "loss": 0.0947, + "step": 46037 + }, + { + "epoch": 0.3996319476393434, + "grad_norm": 0.31640625, + "learning_rate": 0.0011732239058599378, + "loss": 0.0933, + "step": 46038 + }, + { + "epoch": 0.3996406281195476, + "grad_norm": 0.52734375, + "learning_rate": 0.0011731942420320283, + "loss": 0.1108, + "step": 46039 + }, + { + "epoch": 0.39964930859975173, + "grad_norm": 0.4609375, + "learning_rate": 0.0011731645781240753, + "loss": 0.0786, + "step": 46040 + }, + { + "epoch": 0.3996579890799559, + "grad_norm": 0.1181640625, + "learning_rate": 0.00117313491413611, + "loss": 0.1191, + "step": 46041 + }, + { + "epoch": 0.39966666956016006, + "grad_norm": 0.2431640625, + "learning_rate": 0.0011731052500681659, + "loss": 0.1406, + "step": 46042 + }, + { + "epoch": 0.39967535004036425, + "grad_norm": 0.08447265625, + "learning_rate": 0.001173075585920275, + "loss": 0.084, + "step": 46043 + }, + { + "epoch": 0.3996840305205684, + "grad_norm": 0.314453125, + "learning_rate": 0.0011730459216924695, + "loss": 0.0938, + "step": 46044 + }, + { + "epoch": 0.3996927110007726, + "grad_norm": 0.435546875, + "learning_rate": 0.0011730162573847817, + "loss": 0.1162, + "step": 46045 + }, + { + "epoch": 0.3997013914809767, + "grad_norm": 0.55859375, + "learning_rate": 0.0011729865929972453, + "loss": 0.1104, + "step": 46046 + }, + { + "epoch": 0.3997100719611809, + "grad_norm": 0.5859375, + "learning_rate": 0.0011729569285298914, + "loss": 0.1279, + "step": 46047 + }, + { + "epoch": 0.39971875244138505, + "grad_norm": 0.33203125, + "learning_rate": 0.0011729272639827526, + "loss": 0.1426, + "step": 46048 + }, + { + "epoch": 0.39972743292158924, + "grad_norm": 0.34765625, + "learning_rate": 0.0011728975993558623, + "loss": 0.1123, + "step": 46049 + }, + { + "epoch": 0.3997361134017934, + "grad_norm": 0.92578125, + "learning_rate": 0.0011728679346492516, + "loss": 0.1309, + "step": 46050 + }, + { + "epoch": 0.3997447938819976, + "grad_norm": 0.1279296875, + "learning_rate": 0.001172838269862954, + "loss": 0.1064, + "step": 46051 + }, + { + "epoch": 0.3997534743622017, + "grad_norm": 0.07373046875, + "learning_rate": 0.0011728086049970015, + "loss": 0.0635, + "step": 46052 + }, + { + "epoch": 0.3997621548424059, + "grad_norm": 0.294921875, + "learning_rate": 0.0011727789400514266, + "loss": 0.085, + "step": 46053 + }, + { + "epoch": 0.39977083532261004, + "grad_norm": 0.234375, + "learning_rate": 0.0011727492750262618, + "loss": 0.1187, + "step": 46054 + }, + { + "epoch": 0.39977951580281423, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011727196099215395, + "loss": 0.0986, + "step": 46055 + }, + { + "epoch": 0.39978819628301837, + "grad_norm": 0.267578125, + "learning_rate": 0.0011726899447372923, + "loss": 0.0967, + "step": 46056 + }, + { + "epoch": 0.39979687676322256, + "grad_norm": 0.11376953125, + "learning_rate": 0.001172660279473552, + "loss": 0.0693, + "step": 46057 + }, + { + "epoch": 0.3998055572434267, + "grad_norm": 0.302734375, + "learning_rate": 0.0011726306141303519, + "loss": 0.1689, + "step": 46058 + }, + { + "epoch": 0.3998142377236309, + "grad_norm": 0.2158203125, + "learning_rate": 0.0011726009487077237, + "loss": 0.0776, + "step": 46059 + }, + { + "epoch": 0.39982291820383503, + "grad_norm": 0.47265625, + "learning_rate": 0.0011725712832057005, + "loss": 0.0947, + "step": 46060 + }, + { + "epoch": 0.3998315986840392, + "grad_norm": 0.27734375, + "learning_rate": 0.0011725416176243144, + "loss": 0.1084, + "step": 46061 + }, + { + "epoch": 0.39984027916424336, + "grad_norm": 0.359375, + "learning_rate": 0.0011725119519635976, + "loss": 0.1084, + "step": 46062 + }, + { + "epoch": 0.39984895964444755, + "grad_norm": 0.671875, + "learning_rate": 0.0011724822862235831, + "loss": 0.1196, + "step": 46063 + }, + { + "epoch": 0.3998576401246517, + "grad_norm": 0.1357421875, + "learning_rate": 0.001172452620404303, + "loss": 0.0923, + "step": 46064 + }, + { + "epoch": 0.3998663206048559, + "grad_norm": 0.68359375, + "learning_rate": 0.0011724229545057897, + "loss": 0.1025, + "step": 46065 + }, + { + "epoch": 0.39987500108506, + "grad_norm": 0.32421875, + "learning_rate": 0.0011723932885280759, + "loss": 0.0845, + "step": 46066 + }, + { + "epoch": 0.3998836815652642, + "grad_norm": 0.4765625, + "learning_rate": 0.001172363622471194, + "loss": 0.1279, + "step": 46067 + }, + { + "epoch": 0.39989236204546835, + "grad_norm": 4.0, + "learning_rate": 0.001172333956335176, + "loss": 0.2256, + "step": 46068 + }, + { + "epoch": 0.39990104252567255, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011723042901200548, + "loss": 0.1162, + "step": 46069 + }, + { + "epoch": 0.3999097230058767, + "grad_norm": 0.2333984375, + "learning_rate": 0.0011722746238258628, + "loss": 0.0762, + "step": 46070 + }, + { + "epoch": 0.3999184034860809, + "grad_norm": 0.16796875, + "learning_rate": 0.0011722449574526324, + "loss": 0.0801, + "step": 46071 + }, + { + "epoch": 0.399927083966285, + "grad_norm": 0.30078125, + "learning_rate": 0.0011722152910003958, + "loss": 0.083, + "step": 46072 + }, + { + "epoch": 0.3999357644464892, + "grad_norm": 0.2099609375, + "learning_rate": 0.001172185624469186, + "loss": 0.1064, + "step": 46073 + }, + { + "epoch": 0.39994444492669334, + "grad_norm": 0.267578125, + "learning_rate": 0.0011721559578590347, + "loss": 0.0505, + "step": 46074 + }, + { + "epoch": 0.39995312540689754, + "grad_norm": 0.69140625, + "learning_rate": 0.0011721262911699745, + "loss": 0.1201, + "step": 46075 + }, + { + "epoch": 0.3999618058871017, + "grad_norm": 0.51953125, + "learning_rate": 0.0011720966244020388, + "loss": 0.0703, + "step": 46076 + }, + { + "epoch": 0.39997048636730587, + "grad_norm": 0.349609375, + "learning_rate": 0.0011720669575552589, + "loss": 0.1201, + "step": 46077 + }, + { + "epoch": 0.39997916684751, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011720372906296676, + "loss": 0.1074, + "step": 46078 + }, + { + "epoch": 0.3999878473277142, + "grad_norm": 0.11376953125, + "learning_rate": 0.0011720076236252975, + "loss": 0.1191, + "step": 46079 + }, + { + "epoch": 0.39999652780791833, + "grad_norm": 0.8671875, + "learning_rate": 0.001171977956542181, + "loss": 0.1221, + "step": 46080 + }, + { + "epoch": 0.40000520828812247, + "grad_norm": 0.265625, + "learning_rate": 0.0011719482893803503, + "loss": 0.0981, + "step": 46081 + }, + { + "epoch": 0.40001388876832666, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011719186221398384, + "loss": 0.0776, + "step": 46082 + }, + { + "epoch": 0.4000225692485308, + "grad_norm": 0.2021484375, + "learning_rate": 0.001171888954820677, + "loss": 0.1191, + "step": 46083 + }, + { + "epoch": 0.400031249728735, + "grad_norm": 0.33984375, + "learning_rate": 0.001171859287422899, + "loss": 0.0884, + "step": 46084 + }, + { + "epoch": 0.40003993020893913, + "grad_norm": 0.419921875, + "learning_rate": 0.0011718296199465371, + "loss": 0.1045, + "step": 46085 + }, + { + "epoch": 0.4000486106891433, + "grad_norm": 0.55078125, + "learning_rate": 0.001171799952391623, + "loss": 0.0747, + "step": 46086 + }, + { + "epoch": 0.40005729116934746, + "grad_norm": 0.1875, + "learning_rate": 0.0011717702847581896, + "loss": 0.0811, + "step": 46087 + }, + { + "epoch": 0.40006597164955165, + "grad_norm": 0.2099609375, + "learning_rate": 0.0011717406170462692, + "loss": 0.1182, + "step": 46088 + }, + { + "epoch": 0.4000746521297558, + "grad_norm": 0.1552734375, + "learning_rate": 0.0011717109492558947, + "loss": 0.1348, + "step": 46089 + }, + { + "epoch": 0.40008333260996, + "grad_norm": 0.1376953125, + "learning_rate": 0.001171681281387098, + "loss": 0.1602, + "step": 46090 + }, + { + "epoch": 0.4000920130901641, + "grad_norm": 0.2041015625, + "learning_rate": 0.001171651613439912, + "loss": 0.0923, + "step": 46091 + }, + { + "epoch": 0.4001006935703683, + "grad_norm": 0.93359375, + "learning_rate": 0.0011716219454143684, + "loss": 0.1162, + "step": 46092 + }, + { + "epoch": 0.40010937405057245, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011715922773105002, + "loss": 0.0903, + "step": 46093 + }, + { + "epoch": 0.40011805453077665, + "grad_norm": 0.64453125, + "learning_rate": 0.00117156260912834, + "loss": 0.1426, + "step": 46094 + }, + { + "epoch": 0.4001267350109808, + "grad_norm": 0.09521484375, + "learning_rate": 0.0011715329408679201, + "loss": 0.084, + "step": 46095 + }, + { + "epoch": 0.400135415491185, + "grad_norm": 0.0947265625, + "learning_rate": 0.0011715032725292725, + "loss": 0.0874, + "step": 46096 + }, + { + "epoch": 0.4001440959713891, + "grad_norm": 0.2734375, + "learning_rate": 0.00117147360411243, + "loss": 0.127, + "step": 46097 + }, + { + "epoch": 0.4001527764515933, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011714439356174252, + "loss": 0.1089, + "step": 46098 + }, + { + "epoch": 0.40016145693179744, + "grad_norm": 0.23828125, + "learning_rate": 0.0011714142670442905, + "loss": 0.1035, + "step": 46099 + }, + { + "epoch": 0.40017013741200164, + "grad_norm": 0.29296875, + "learning_rate": 0.001171384598393058, + "loss": 0.0923, + "step": 46100 + }, + { + "epoch": 0.4001788178922058, + "grad_norm": 0.47265625, + "learning_rate": 0.0011713549296637601, + "loss": 0.0806, + "step": 46101 + }, + { + "epoch": 0.40018749837240997, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011713252608564302, + "loss": 0.1099, + "step": 46102 + }, + { + "epoch": 0.4001961788526141, + "grad_norm": 0.1025390625, + "learning_rate": 0.0011712955919711, + "loss": 0.0938, + "step": 46103 + }, + { + "epoch": 0.4002048593328183, + "grad_norm": 0.75, + "learning_rate": 0.0011712659230078018, + "loss": 0.1699, + "step": 46104 + }, + { + "epoch": 0.40021353981302243, + "grad_norm": 0.2412109375, + "learning_rate": 0.001171236253966568, + "loss": 0.0732, + "step": 46105 + }, + { + "epoch": 0.4002222202932266, + "grad_norm": 0.78515625, + "learning_rate": 0.0011712065848474315, + "loss": 0.1187, + "step": 46106 + }, + { + "epoch": 0.40023090077343076, + "grad_norm": 0.228515625, + "learning_rate": 0.0011711769156504243, + "loss": 0.0962, + "step": 46107 + }, + { + "epoch": 0.40023958125363496, + "grad_norm": 0.1240234375, + "learning_rate": 0.0011711472463755796, + "loss": 0.0684, + "step": 46108 + }, + { + "epoch": 0.4002482617338391, + "grad_norm": 0.19140625, + "learning_rate": 0.0011711175770229292, + "loss": 0.1011, + "step": 46109 + }, + { + "epoch": 0.4002569422140433, + "grad_norm": 0.341796875, + "learning_rate": 0.0011710879075925054, + "loss": 0.0771, + "step": 46110 + }, + { + "epoch": 0.4002656226942474, + "grad_norm": 0.0693359375, + "learning_rate": 0.001171058238084341, + "loss": 0.0698, + "step": 46111 + }, + { + "epoch": 0.4002743031744516, + "grad_norm": 0.6875, + "learning_rate": 0.001171028568498469, + "loss": 0.1021, + "step": 46112 + }, + { + "epoch": 0.40028298365465576, + "grad_norm": 0.1083984375, + "learning_rate": 0.0011709988988349207, + "loss": 0.0938, + "step": 46113 + }, + { + "epoch": 0.40029166413485995, + "grad_norm": 0.447265625, + "learning_rate": 0.001170969229093729, + "loss": 0.1108, + "step": 46114 + }, + { + "epoch": 0.4003003446150641, + "grad_norm": 0.06494140625, + "learning_rate": 0.0011709395592749265, + "loss": 0.0898, + "step": 46115 + }, + { + "epoch": 0.4003090250952683, + "grad_norm": 0.267578125, + "learning_rate": 0.0011709098893785456, + "loss": 0.0713, + "step": 46116 + }, + { + "epoch": 0.4003177055754724, + "grad_norm": 0.275390625, + "learning_rate": 0.0011708802194046184, + "loss": 0.1533, + "step": 46117 + }, + { + "epoch": 0.4003263860556766, + "grad_norm": 0.306640625, + "learning_rate": 0.001170850549353178, + "loss": 0.0742, + "step": 46118 + }, + { + "epoch": 0.40033506653588075, + "grad_norm": 0.181640625, + "learning_rate": 0.0011708208792242565, + "loss": 0.1167, + "step": 46119 + }, + { + "epoch": 0.40034374701608494, + "grad_norm": 0.6484375, + "learning_rate": 0.0011707912090178866, + "loss": 0.0894, + "step": 46120 + }, + { + "epoch": 0.4003524274962891, + "grad_norm": 0.67578125, + "learning_rate": 0.0011707615387341, + "loss": 0.0835, + "step": 46121 + }, + { + "epoch": 0.40036110797649327, + "grad_norm": 0.61328125, + "learning_rate": 0.00117073186837293, + "loss": 0.1367, + "step": 46122 + }, + { + "epoch": 0.4003697884566974, + "grad_norm": 0.58984375, + "learning_rate": 0.0011707021979344083, + "loss": 0.1035, + "step": 46123 + }, + { + "epoch": 0.4003784689369016, + "grad_norm": 0.19140625, + "learning_rate": 0.001170672527418568, + "loss": 0.0786, + "step": 46124 + }, + { + "epoch": 0.40038714941710574, + "grad_norm": 0.419921875, + "learning_rate": 0.0011706428568254413, + "loss": 0.0996, + "step": 46125 + }, + { + "epoch": 0.40039582989730993, + "grad_norm": 0.9140625, + "learning_rate": 0.0011706131861550606, + "loss": 0.0908, + "step": 46126 + }, + { + "epoch": 0.40040451037751407, + "grad_norm": 0.29296875, + "learning_rate": 0.0011705835154074582, + "loss": 0.085, + "step": 46127 + }, + { + "epoch": 0.40041319085771826, + "grad_norm": 0.203125, + "learning_rate": 0.0011705538445826672, + "loss": 0.1279, + "step": 46128 + }, + { + "epoch": 0.4004218713379224, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011705241736807192, + "loss": 0.0967, + "step": 46129 + }, + { + "epoch": 0.4004305518181266, + "grad_norm": 0.212890625, + "learning_rate": 0.0011704945027016472, + "loss": 0.1035, + "step": 46130 + }, + { + "epoch": 0.4004392322983307, + "grad_norm": 0.267578125, + "learning_rate": 0.0011704648316454831, + "loss": 0.084, + "step": 46131 + }, + { + "epoch": 0.4004479127785349, + "grad_norm": 0.1171875, + "learning_rate": 0.00117043516051226, + "loss": 0.1074, + "step": 46132 + }, + { + "epoch": 0.40045659325873906, + "grad_norm": 0.125, + "learning_rate": 0.0011704054893020102, + "loss": 0.1143, + "step": 46133 + }, + { + "epoch": 0.40046527373894325, + "grad_norm": 0.166015625, + "learning_rate": 0.001170375818014766, + "loss": 0.0977, + "step": 46134 + }, + { + "epoch": 0.4004739542191474, + "grad_norm": 0.0986328125, + "learning_rate": 0.0011703461466505596, + "loss": 0.0742, + "step": 46135 + }, + { + "epoch": 0.4004826346993516, + "grad_norm": 0.25390625, + "learning_rate": 0.0011703164752094237, + "loss": 0.124, + "step": 46136 + }, + { + "epoch": 0.4004913151795557, + "grad_norm": 0.423828125, + "learning_rate": 0.001170286803691391, + "loss": 0.0908, + "step": 46137 + }, + { + "epoch": 0.4004999956597599, + "grad_norm": 0.251953125, + "learning_rate": 0.0011702571320964938, + "loss": 0.0908, + "step": 46138 + }, + { + "epoch": 0.40050867613996405, + "grad_norm": 1.078125, + "learning_rate": 0.0011702274604247643, + "loss": 0.1157, + "step": 46139 + }, + { + "epoch": 0.40051735662016824, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011701977886762348, + "loss": 0.1206, + "step": 46140 + }, + { + "epoch": 0.4005260371003724, + "grad_norm": 0.166015625, + "learning_rate": 0.0011701681168509383, + "loss": 0.1338, + "step": 46141 + }, + { + "epoch": 0.40053471758057657, + "grad_norm": 0.6484375, + "learning_rate": 0.0011701384449489072, + "loss": 0.0898, + "step": 46142 + }, + { + "epoch": 0.4005433980607807, + "grad_norm": 0.390625, + "learning_rate": 0.0011701087729701735, + "loss": 0.1084, + "step": 46143 + }, + { + "epoch": 0.4005520785409849, + "grad_norm": 0.2265625, + "learning_rate": 0.0011700791009147697, + "loss": 0.1191, + "step": 46144 + }, + { + "epoch": 0.40056075902118904, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011700494287827288, + "loss": 0.0894, + "step": 46145 + }, + { + "epoch": 0.40056943950139323, + "grad_norm": 0.36328125, + "learning_rate": 0.001170019756574083, + "loss": 0.1123, + "step": 46146 + }, + { + "epoch": 0.40057811998159737, + "grad_norm": 0.267578125, + "learning_rate": 0.0011699900842888644, + "loss": 0.1182, + "step": 46147 + }, + { + "epoch": 0.40058680046180156, + "grad_norm": 0.11376953125, + "learning_rate": 0.001169960411927106, + "loss": 0.083, + "step": 46148 + }, + { + "epoch": 0.4005954809420057, + "grad_norm": 0.1806640625, + "learning_rate": 0.0011699307394888395, + "loss": 0.0952, + "step": 46149 + }, + { + "epoch": 0.4006041614222099, + "grad_norm": 0.4765625, + "learning_rate": 0.001169901066974098, + "loss": 0.1201, + "step": 46150 + }, + { + "epoch": 0.40061284190241403, + "grad_norm": 0.82421875, + "learning_rate": 0.0011698713943829138, + "loss": 0.0967, + "step": 46151 + }, + { + "epoch": 0.4006215223826182, + "grad_norm": 0.6015625, + "learning_rate": 0.0011698417217153195, + "loss": 0.1016, + "step": 46152 + }, + { + "epoch": 0.40063020286282236, + "grad_norm": 0.640625, + "learning_rate": 0.0011698120489713468, + "loss": 0.1309, + "step": 46153 + }, + { + "epoch": 0.40063888334302655, + "grad_norm": 0.138671875, + "learning_rate": 0.001169782376151029, + "loss": 0.1162, + "step": 46154 + }, + { + "epoch": 0.4006475638232307, + "grad_norm": 0.1923828125, + "learning_rate": 0.001169752703254398, + "loss": 0.0913, + "step": 46155 + }, + { + "epoch": 0.4006562443034349, + "grad_norm": 0.62890625, + "learning_rate": 0.001169723030281487, + "loss": 0.1074, + "step": 46156 + }, + { + "epoch": 0.400664924783639, + "grad_norm": 0.67578125, + "learning_rate": 0.0011696933572323276, + "loss": 0.1172, + "step": 46157 + }, + { + "epoch": 0.4006736052638432, + "grad_norm": 0.1875, + "learning_rate": 0.0011696636841069525, + "loss": 0.1226, + "step": 46158 + }, + { + "epoch": 0.40068228574404735, + "grad_norm": 0.0888671875, + "learning_rate": 0.0011696340109053946, + "loss": 0.1162, + "step": 46159 + }, + { + "epoch": 0.40069096622425154, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011696043376276859, + "loss": 0.0986, + "step": 46160 + }, + { + "epoch": 0.4006996467044557, + "grad_norm": 0.11328125, + "learning_rate": 0.0011695746642738586, + "loss": 0.0713, + "step": 46161 + }, + { + "epoch": 0.4007083271846599, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011695449908439457, + "loss": 0.0742, + "step": 46162 + }, + { + "epoch": 0.400717007664864, + "grad_norm": 0.2197265625, + "learning_rate": 0.0011695153173379796, + "loss": 0.0869, + "step": 46163 + }, + { + "epoch": 0.4007256881450682, + "grad_norm": 0.150390625, + "learning_rate": 0.0011694856437559921, + "loss": 0.0874, + "step": 46164 + }, + { + "epoch": 0.40073436862527234, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011694559700980167, + "loss": 0.0972, + "step": 46165 + }, + { + "epoch": 0.40074304910547653, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011694262963640849, + "loss": 0.1084, + "step": 46166 + }, + { + "epoch": 0.40075172958568067, + "grad_norm": 0.09033203125, + "learning_rate": 0.0011693966225542298, + "loss": 0.0645, + "step": 46167 + }, + { + "epoch": 0.40076041006588486, + "grad_norm": 0.193359375, + "learning_rate": 0.0011693669486684836, + "loss": 0.1025, + "step": 46168 + }, + { + "epoch": 0.400769090546089, + "grad_norm": 0.26171875, + "learning_rate": 0.0011693372747068787, + "loss": 0.126, + "step": 46169 + }, + { + "epoch": 0.4007777710262932, + "grad_norm": 0.23828125, + "learning_rate": 0.0011693076006694474, + "loss": 0.0928, + "step": 46170 + }, + { + "epoch": 0.40078645150649733, + "grad_norm": 0.478515625, + "learning_rate": 0.0011692779265562226, + "loss": 0.0864, + "step": 46171 + }, + { + "epoch": 0.4007951319867015, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011692482523672366, + "loss": 0.0737, + "step": 46172 + }, + { + "epoch": 0.40080381246690566, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011692185781025212, + "loss": 0.1152, + "step": 46173 + }, + { + "epoch": 0.40081249294710986, + "grad_norm": 0.19140625, + "learning_rate": 0.00116918890376211, + "loss": 0.1084, + "step": 46174 + }, + { + "epoch": 0.400821173427314, + "grad_norm": 0.314453125, + "learning_rate": 0.0011691592293460345, + "loss": 0.1108, + "step": 46175 + }, + { + "epoch": 0.4008298539075182, + "grad_norm": 0.640625, + "learning_rate": 0.0011691295548543277, + "loss": 0.1328, + "step": 46176 + }, + { + "epoch": 0.4008385343877223, + "grad_norm": 0.369140625, + "learning_rate": 0.0011690998802870221, + "loss": 0.0889, + "step": 46177 + }, + { + "epoch": 0.4008472148679265, + "grad_norm": 0.138671875, + "learning_rate": 0.0011690702056441494, + "loss": 0.0703, + "step": 46178 + }, + { + "epoch": 0.40085589534813065, + "grad_norm": 0.33984375, + "learning_rate": 0.0011690405309257425, + "loss": 0.0679, + "step": 46179 + }, + { + "epoch": 0.40086457582833485, + "grad_norm": 0.11083984375, + "learning_rate": 0.0011690108561318345, + "loss": 0.0889, + "step": 46180 + }, + { + "epoch": 0.400873256308539, + "grad_norm": 0.35546875, + "learning_rate": 0.0011689811812624568, + "loss": 0.1367, + "step": 46181 + }, + { + "epoch": 0.4008819367887432, + "grad_norm": 0.486328125, + "learning_rate": 0.0011689515063176426, + "loss": 0.1035, + "step": 46182 + }, + { + "epoch": 0.4008906172689473, + "grad_norm": 0.447265625, + "learning_rate": 0.001168921831297424, + "loss": 0.0928, + "step": 46183 + }, + { + "epoch": 0.4008992977491515, + "grad_norm": 0.416015625, + "learning_rate": 0.0011688921562018333, + "loss": 0.0796, + "step": 46184 + }, + { + "epoch": 0.40090797822935564, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011688624810309032, + "loss": 0.1123, + "step": 46185 + }, + { + "epoch": 0.40091665870955984, + "grad_norm": 0.234375, + "learning_rate": 0.0011688328057846665, + "loss": 0.1074, + "step": 46186 + }, + { + "epoch": 0.400925339189764, + "grad_norm": 0.384765625, + "learning_rate": 0.0011688031304631548, + "loss": 0.1104, + "step": 46187 + }, + { + "epoch": 0.40093401966996817, + "grad_norm": 0.1328125, + "learning_rate": 0.0011687734550664013, + "loss": 0.0703, + "step": 46188 + }, + { + "epoch": 0.4009427001501723, + "grad_norm": 0.33984375, + "learning_rate": 0.0011687437795944381, + "loss": 0.0977, + "step": 46189 + }, + { + "epoch": 0.4009513806303765, + "grad_norm": 0.734375, + "learning_rate": 0.0011687141040472979, + "loss": 0.0923, + "step": 46190 + }, + { + "epoch": 0.40096006111058063, + "grad_norm": 0.3203125, + "learning_rate": 0.001168684428425013, + "loss": 0.1289, + "step": 46191 + }, + { + "epoch": 0.4009687415907848, + "grad_norm": 0.12353515625, + "learning_rate": 0.0011686547527276156, + "loss": 0.125, + "step": 46192 + }, + { + "epoch": 0.40097742207098896, + "grad_norm": 0.220703125, + "learning_rate": 0.0011686250769551384, + "loss": 0.1079, + "step": 46193 + }, + { + "epoch": 0.40098610255119316, + "grad_norm": 0.107421875, + "learning_rate": 0.0011685954011076138, + "loss": 0.1289, + "step": 46194 + }, + { + "epoch": 0.4009947830313973, + "grad_norm": 0.07958984375, + "learning_rate": 0.0011685657251850746, + "loss": 0.082, + "step": 46195 + }, + { + "epoch": 0.4010034635116015, + "grad_norm": 0.220703125, + "learning_rate": 0.0011685360491875527, + "loss": 0.0933, + "step": 46196 + }, + { + "epoch": 0.4010121439918056, + "grad_norm": 0.13671875, + "learning_rate": 0.001168506373115081, + "loss": 0.1104, + "step": 46197 + }, + { + "epoch": 0.4010208244720098, + "grad_norm": 0.45703125, + "learning_rate": 0.0011684766969676918, + "loss": 0.0864, + "step": 46198 + }, + { + "epoch": 0.40102950495221396, + "grad_norm": 0.306640625, + "learning_rate": 0.0011684470207454175, + "loss": 0.1011, + "step": 46199 + }, + { + "epoch": 0.40103818543241815, + "grad_norm": 0.1748046875, + "learning_rate": 0.0011684173444482901, + "loss": 0.1235, + "step": 46200 + }, + { + "epoch": 0.4010468659126223, + "grad_norm": 0.43359375, + "learning_rate": 0.0011683876680763428, + "loss": 0.0947, + "step": 46201 + }, + { + "epoch": 0.4010555463928265, + "grad_norm": 0.62109375, + "learning_rate": 0.001168357991629608, + "loss": 0.0884, + "step": 46202 + }, + { + "epoch": 0.4010642268730306, + "grad_norm": 0.2353515625, + "learning_rate": 0.0011683283151081177, + "loss": 0.105, + "step": 46203 + }, + { + "epoch": 0.40107290735323475, + "grad_norm": 0.310546875, + "learning_rate": 0.0011682986385119047, + "loss": 0.1191, + "step": 46204 + }, + { + "epoch": 0.40108158783343895, + "grad_norm": 0.357421875, + "learning_rate": 0.0011682689618410012, + "loss": 0.1074, + "step": 46205 + }, + { + "epoch": 0.4010902683136431, + "grad_norm": 0.54296875, + "learning_rate": 0.00116823928509544, + "loss": 0.1133, + "step": 46206 + }, + { + "epoch": 0.4010989487938473, + "grad_norm": 0.1474609375, + "learning_rate": 0.0011682096082752532, + "loss": 0.0723, + "step": 46207 + }, + { + "epoch": 0.4011076292740514, + "grad_norm": 0.3671875, + "learning_rate": 0.0011681799313804735, + "loss": 0.1455, + "step": 46208 + }, + { + "epoch": 0.4011163097542556, + "grad_norm": 0.40625, + "learning_rate": 0.0011681502544111332, + "loss": 0.1357, + "step": 46209 + }, + { + "epoch": 0.40112499023445974, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011681205773672648, + "loss": 0.124, + "step": 46210 + }, + { + "epoch": 0.40113367071466394, + "grad_norm": 0.201171875, + "learning_rate": 0.0011680909002489005, + "loss": 0.0957, + "step": 46211 + }, + { + "epoch": 0.4011423511948681, + "grad_norm": 0.234375, + "learning_rate": 0.0011680612230560731, + "loss": 0.1079, + "step": 46212 + }, + { + "epoch": 0.40115103167507227, + "grad_norm": 0.435546875, + "learning_rate": 0.0011680315457888151, + "loss": 0.1001, + "step": 46213 + }, + { + "epoch": 0.4011597121552764, + "grad_norm": 0.609375, + "learning_rate": 0.001168001868447159, + "loss": 0.1328, + "step": 46214 + }, + { + "epoch": 0.4011683926354806, + "grad_norm": 0.236328125, + "learning_rate": 0.0011679721910311366, + "loss": 0.0938, + "step": 46215 + }, + { + "epoch": 0.40117707311568473, + "grad_norm": 0.37109375, + "learning_rate": 0.0011679425135407815, + "loss": 0.0928, + "step": 46216 + }, + { + "epoch": 0.4011857535958889, + "grad_norm": 0.51171875, + "learning_rate": 0.0011679128359761252, + "loss": 0.1108, + "step": 46217 + }, + { + "epoch": 0.40119443407609306, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011678831583372003, + "loss": 0.1113, + "step": 46218 + }, + { + "epoch": 0.40120311455629726, + "grad_norm": 0.294921875, + "learning_rate": 0.0011678534806240394, + "loss": 0.1387, + "step": 46219 + }, + { + "epoch": 0.4012117950365014, + "grad_norm": 0.0927734375, + "learning_rate": 0.001167823802836675, + "loss": 0.1143, + "step": 46220 + }, + { + "epoch": 0.4012204755167056, + "grad_norm": 0.1611328125, + "learning_rate": 0.0011677941249751396, + "loss": 0.0894, + "step": 46221 + }, + { + "epoch": 0.4012291559969097, + "grad_norm": 0.09912109375, + "learning_rate": 0.0011677644470394655, + "loss": 0.0864, + "step": 46222 + }, + { + "epoch": 0.4012378364771139, + "grad_norm": 0.8125, + "learning_rate": 0.0011677347690296851, + "loss": 0.1172, + "step": 46223 + }, + { + "epoch": 0.40124651695731806, + "grad_norm": 0.1875, + "learning_rate": 0.0011677050909458314, + "loss": 0.1289, + "step": 46224 + }, + { + "epoch": 0.40125519743752225, + "grad_norm": 0.0576171875, + "learning_rate": 0.0011676754127879362, + "loss": 0.0786, + "step": 46225 + }, + { + "epoch": 0.4012638779177264, + "grad_norm": 0.177734375, + "learning_rate": 0.0011676457345560318, + "loss": 0.1187, + "step": 46226 + }, + { + "epoch": 0.4012725583979306, + "grad_norm": 0.1103515625, + "learning_rate": 0.0011676160562501512, + "loss": 0.1064, + "step": 46227 + }, + { + "epoch": 0.4012812388781347, + "grad_norm": 0.21484375, + "learning_rate": 0.001167586377870327, + "loss": 0.126, + "step": 46228 + }, + { + "epoch": 0.4012899193583389, + "grad_norm": 0.314453125, + "learning_rate": 0.0011675566994165912, + "loss": 0.0996, + "step": 46229 + }, + { + "epoch": 0.40129859983854305, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011675270208889763, + "loss": 0.0806, + "step": 46230 + }, + { + "epoch": 0.40130728031874724, + "grad_norm": 0.201171875, + "learning_rate": 0.001167497342287515, + "loss": 0.0972, + "step": 46231 + }, + { + "epoch": 0.4013159607989514, + "grad_norm": 0.271484375, + "learning_rate": 0.0011674676636122398, + "loss": 0.1182, + "step": 46232 + }, + { + "epoch": 0.40132464127915557, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011674379848631827, + "loss": 0.0977, + "step": 46233 + }, + { + "epoch": 0.4013333217593597, + "grad_norm": 0.2265625, + "learning_rate": 0.0011674083060403764, + "loss": 0.1006, + "step": 46234 + }, + { + "epoch": 0.4013420022395639, + "grad_norm": 0.2177734375, + "learning_rate": 0.0011673786271438535, + "loss": 0.0879, + "step": 46235 + }, + { + "epoch": 0.40135068271976804, + "grad_norm": 0.41015625, + "learning_rate": 0.0011673489481736463, + "loss": 0.0928, + "step": 46236 + }, + { + "epoch": 0.40135936319997223, + "grad_norm": 0.50390625, + "learning_rate": 0.0011673192691297877, + "loss": 0.0889, + "step": 46237 + }, + { + "epoch": 0.40136804368017637, + "grad_norm": 0.193359375, + "learning_rate": 0.0011672895900123094, + "loss": 0.1094, + "step": 46238 + }, + { + "epoch": 0.40137672416038056, + "grad_norm": 0.40625, + "learning_rate": 0.0011672599108212439, + "loss": 0.0957, + "step": 46239 + }, + { + "epoch": 0.4013854046405847, + "grad_norm": 0.326171875, + "learning_rate": 0.0011672302315566244, + "loss": 0.085, + "step": 46240 + }, + { + "epoch": 0.4013940851207889, + "grad_norm": 0.234375, + "learning_rate": 0.0011672005522184825, + "loss": 0.0972, + "step": 46241 + }, + { + "epoch": 0.40140276560099303, + "grad_norm": 0.10107421875, + "learning_rate": 0.0011671708728068517, + "loss": 0.0737, + "step": 46242 + }, + { + "epoch": 0.4014114460811972, + "grad_norm": 0.10302734375, + "learning_rate": 0.0011671411933217636, + "loss": 0.0967, + "step": 46243 + }, + { + "epoch": 0.40142012656140136, + "grad_norm": 0.24609375, + "learning_rate": 0.0011671115137632507, + "loss": 0.0869, + "step": 46244 + }, + { + "epoch": 0.40142880704160555, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011670818341313455, + "loss": 0.0752, + "step": 46245 + }, + { + "epoch": 0.4014374875218097, + "grad_norm": 0.5, + "learning_rate": 0.0011670521544260813, + "loss": 0.1167, + "step": 46246 + }, + { + "epoch": 0.4014461680020139, + "grad_norm": 0.33984375, + "learning_rate": 0.0011670224746474893, + "loss": 0.0815, + "step": 46247 + }, + { + "epoch": 0.401454848482218, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011669927947956028, + "loss": 0.0703, + "step": 46248 + }, + { + "epoch": 0.4014635289624222, + "grad_norm": 0.220703125, + "learning_rate": 0.0011669631148704539, + "loss": 0.0854, + "step": 46249 + }, + { + "epoch": 0.40147220944262635, + "grad_norm": 0.1220703125, + "learning_rate": 0.001166933434872075, + "loss": 0.0957, + "step": 46250 + }, + { + "epoch": 0.40148088992283054, + "grad_norm": 0.099609375, + "learning_rate": 0.001166903754800499, + "loss": 0.125, + "step": 46251 + }, + { + "epoch": 0.4014895704030347, + "grad_norm": 0.29296875, + "learning_rate": 0.0011668740746557577, + "loss": 0.0801, + "step": 46252 + }, + { + "epoch": 0.40149825088323887, + "grad_norm": 0.609375, + "learning_rate": 0.0011668443944378845, + "loss": 0.1279, + "step": 46253 + }, + { + "epoch": 0.401506931363443, + "grad_norm": 0.07421875, + "learning_rate": 0.0011668147141469108, + "loss": 0.0874, + "step": 46254 + }, + { + "epoch": 0.4015156118436472, + "grad_norm": 0.30078125, + "learning_rate": 0.0011667850337828696, + "loss": 0.0967, + "step": 46255 + }, + { + "epoch": 0.40152429232385134, + "grad_norm": 0.08154296875, + "learning_rate": 0.0011667553533457934, + "loss": 0.0854, + "step": 46256 + }, + { + "epoch": 0.40153297280405553, + "grad_norm": 0.54296875, + "learning_rate": 0.0011667256728357142, + "loss": 0.1157, + "step": 46257 + }, + { + "epoch": 0.40154165328425967, + "grad_norm": 0.09130859375, + "learning_rate": 0.0011666959922526652, + "loss": 0.0913, + "step": 46258 + }, + { + "epoch": 0.40155033376446386, + "grad_norm": 0.28515625, + "learning_rate": 0.0011666663115966784, + "loss": 0.0986, + "step": 46259 + }, + { + "epoch": 0.401559014244668, + "grad_norm": 0.1005859375, + "learning_rate": 0.0011666366308677864, + "loss": 0.0977, + "step": 46260 + }, + { + "epoch": 0.4015676947248722, + "grad_norm": 0.58984375, + "learning_rate": 0.001166606950066021, + "loss": 0.0889, + "step": 46261 + }, + { + "epoch": 0.40157637520507633, + "grad_norm": 0.1572265625, + "learning_rate": 0.001166577269191416, + "loss": 0.084, + "step": 46262 + }, + { + "epoch": 0.4015850556852805, + "grad_norm": 0.57421875, + "learning_rate": 0.0011665475882440025, + "loss": 0.1045, + "step": 46263 + }, + { + "epoch": 0.40159373616548466, + "grad_norm": 0.20703125, + "learning_rate": 0.001166517907223814, + "loss": 0.1045, + "step": 46264 + }, + { + "epoch": 0.40160241664568885, + "grad_norm": 0.126953125, + "learning_rate": 0.0011664882261308823, + "loss": 0.0942, + "step": 46265 + }, + { + "epoch": 0.401611097125893, + "grad_norm": 0.11767578125, + "learning_rate": 0.00116645854496524, + "loss": 0.1221, + "step": 46266 + }, + { + "epoch": 0.4016197776060972, + "grad_norm": 0.12109375, + "learning_rate": 0.00116642886372692, + "loss": 0.1094, + "step": 46267 + }, + { + "epoch": 0.4016284580863013, + "grad_norm": 0.13671875, + "learning_rate": 0.0011663991824159541, + "loss": 0.1099, + "step": 46268 + }, + { + "epoch": 0.4016371385665055, + "grad_norm": 0.189453125, + "learning_rate": 0.0011663695010323749, + "loss": 0.1094, + "step": 46269 + }, + { + "epoch": 0.40164581904670965, + "grad_norm": 0.1103515625, + "learning_rate": 0.0011663398195762154, + "loss": 0.0801, + "step": 46270 + }, + { + "epoch": 0.40165449952691384, + "grad_norm": 0.1064453125, + "learning_rate": 0.0011663101380475075, + "loss": 0.1064, + "step": 46271 + }, + { + "epoch": 0.401663180007118, + "grad_norm": 0.125, + "learning_rate": 0.001166280456446284, + "loss": 0.1055, + "step": 46272 + }, + { + "epoch": 0.4016718604873222, + "grad_norm": 0.369140625, + "learning_rate": 0.001166250774772577, + "loss": 0.1494, + "step": 46273 + }, + { + "epoch": 0.4016805409675263, + "grad_norm": 0.828125, + "learning_rate": 0.0011662210930264189, + "loss": 0.1221, + "step": 46274 + }, + { + "epoch": 0.4016892214477305, + "grad_norm": 0.302734375, + "learning_rate": 0.0011661914112078427, + "loss": 0.1191, + "step": 46275 + }, + { + "epoch": 0.40169790192793464, + "grad_norm": 0.228515625, + "learning_rate": 0.0011661617293168805, + "loss": 0.125, + "step": 46276 + }, + { + "epoch": 0.40170658240813883, + "grad_norm": 0.369140625, + "learning_rate": 0.0011661320473535651, + "loss": 0.1484, + "step": 46277 + }, + { + "epoch": 0.40171526288834297, + "grad_norm": 0.46484375, + "learning_rate": 0.001166102365317928, + "loss": 0.1211, + "step": 46278 + }, + { + "epoch": 0.40172394336854717, + "grad_norm": 0.96484375, + "learning_rate": 0.001166072683210003, + "loss": 0.1025, + "step": 46279 + }, + { + "epoch": 0.4017326238487513, + "grad_norm": 0.1171875, + "learning_rate": 0.0011660430010298217, + "loss": 0.1196, + "step": 46280 + }, + { + "epoch": 0.4017413043289555, + "grad_norm": 0.224609375, + "learning_rate": 0.0011660133187774166, + "loss": 0.124, + "step": 46281 + }, + { + "epoch": 0.40174998480915963, + "grad_norm": 0.328125, + "learning_rate": 0.0011659836364528206, + "loss": 0.0781, + "step": 46282 + }, + { + "epoch": 0.4017586652893638, + "grad_norm": 0.1748046875, + "learning_rate": 0.0011659539540560657, + "loss": 0.083, + "step": 46283 + }, + { + "epoch": 0.40176734576956796, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011659242715871846, + "loss": 0.0811, + "step": 46284 + }, + { + "epoch": 0.40177602624977216, + "grad_norm": 0.09912109375, + "learning_rate": 0.0011658945890462098, + "loss": 0.0713, + "step": 46285 + }, + { + "epoch": 0.4017847067299763, + "grad_norm": 0.2421875, + "learning_rate": 0.001165864906433174, + "loss": 0.082, + "step": 46286 + }, + { + "epoch": 0.4017933872101805, + "grad_norm": 0.435546875, + "learning_rate": 0.0011658352237481084, + "loss": 0.1191, + "step": 46287 + }, + { + "epoch": 0.4018020676903846, + "grad_norm": 0.44140625, + "learning_rate": 0.0011658055409910473, + "loss": 0.0732, + "step": 46288 + }, + { + "epoch": 0.4018107481705888, + "grad_norm": 0.51171875, + "learning_rate": 0.0011657758581620214, + "loss": 0.126, + "step": 46289 + }, + { + "epoch": 0.40181942865079295, + "grad_norm": 0.306640625, + "learning_rate": 0.0011657461752610648, + "loss": 0.061, + "step": 46290 + }, + { + "epoch": 0.40182810913099715, + "grad_norm": 0.296875, + "learning_rate": 0.0011657164922882087, + "loss": 0.0688, + "step": 46291 + }, + { + "epoch": 0.4018367896112013, + "grad_norm": 0.3125, + "learning_rate": 0.0011656868092434862, + "loss": 0.1094, + "step": 46292 + }, + { + "epoch": 0.4018454700914055, + "grad_norm": 0.1806640625, + "learning_rate": 0.0011656571261269295, + "loss": 0.0796, + "step": 46293 + }, + { + "epoch": 0.4018541505716096, + "grad_norm": 0.6171875, + "learning_rate": 0.0011656274429385713, + "loss": 0.1084, + "step": 46294 + }, + { + "epoch": 0.4018628310518138, + "grad_norm": 0.447265625, + "learning_rate": 0.001165597759678444, + "loss": 0.0942, + "step": 46295 + }, + { + "epoch": 0.40187151153201794, + "grad_norm": 0.470703125, + "learning_rate": 0.00116556807634658, + "loss": 0.0928, + "step": 46296 + }, + { + "epoch": 0.40188019201222214, + "grad_norm": 0.31640625, + "learning_rate": 0.0011655383929430113, + "loss": 0.0718, + "step": 46297 + }, + { + "epoch": 0.4018888724924263, + "grad_norm": 0.2373046875, + "learning_rate": 0.001165508709467771, + "loss": 0.0962, + "step": 46298 + }, + { + "epoch": 0.40189755297263047, + "grad_norm": 0.146484375, + "learning_rate": 0.0011654790259208914, + "loss": 0.1172, + "step": 46299 + }, + { + "epoch": 0.4019062334528346, + "grad_norm": 0.42578125, + "learning_rate": 0.0011654493423024048, + "loss": 0.0815, + "step": 46300 + }, + { + "epoch": 0.4019149139330388, + "grad_norm": 0.259765625, + "learning_rate": 0.001165419658612344, + "loss": 0.0928, + "step": 46301 + }, + { + "epoch": 0.40192359441324294, + "grad_norm": 0.5078125, + "learning_rate": 0.0011653899748507411, + "loss": 0.1318, + "step": 46302 + }, + { + "epoch": 0.40193227489344713, + "grad_norm": 0.10546875, + "learning_rate": 0.001165360291017629, + "loss": 0.0996, + "step": 46303 + }, + { + "epoch": 0.40194095537365127, + "grad_norm": 0.404296875, + "learning_rate": 0.0011653306071130393, + "loss": 0.0854, + "step": 46304 + }, + { + "epoch": 0.40194963585385546, + "grad_norm": 0.1953125, + "learning_rate": 0.0011653009231370056, + "loss": 0.1377, + "step": 46305 + }, + { + "epoch": 0.4019583163340596, + "grad_norm": 0.310546875, + "learning_rate": 0.0011652712390895594, + "loss": 0.0776, + "step": 46306 + }, + { + "epoch": 0.4019669968142638, + "grad_norm": 0.177734375, + "learning_rate": 0.0011652415549707339, + "loss": 0.0684, + "step": 46307 + }, + { + "epoch": 0.4019756772944679, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011652118707805607, + "loss": 0.0811, + "step": 46308 + }, + { + "epoch": 0.4019843577746721, + "grad_norm": 0.412109375, + "learning_rate": 0.0011651821865190733, + "loss": 0.0898, + "step": 46309 + }, + { + "epoch": 0.40199303825487626, + "grad_norm": 0.453125, + "learning_rate": 0.0011651525021863034, + "loss": 0.0747, + "step": 46310 + }, + { + "epoch": 0.40200171873508045, + "grad_norm": 0.1416015625, + "learning_rate": 0.0011651228177822837, + "loss": 0.1182, + "step": 46311 + }, + { + "epoch": 0.4020103992152846, + "grad_norm": 0.2255859375, + "learning_rate": 0.001165093133307047, + "loss": 0.0762, + "step": 46312 + }, + { + "epoch": 0.4020190796954888, + "grad_norm": 0.330078125, + "learning_rate": 0.0011650634487606248, + "loss": 0.1123, + "step": 46313 + }, + { + "epoch": 0.4020277601756929, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011650337641430507, + "loss": 0.0898, + "step": 46314 + }, + { + "epoch": 0.4020364406558971, + "grad_norm": 0.35546875, + "learning_rate": 0.0011650040794543566, + "loss": 0.1245, + "step": 46315 + }, + { + "epoch": 0.40204512113610125, + "grad_norm": 0.357421875, + "learning_rate": 0.0011649743946945748, + "loss": 0.1045, + "step": 46316 + }, + { + "epoch": 0.40205380161630544, + "grad_norm": 0.255859375, + "learning_rate": 0.001164944709863738, + "loss": 0.127, + "step": 46317 + }, + { + "epoch": 0.4020624820965096, + "grad_norm": 0.146484375, + "learning_rate": 0.0011649150249618786, + "loss": 0.1108, + "step": 46318 + }, + { + "epoch": 0.40207116257671377, + "grad_norm": 0.26171875, + "learning_rate": 0.0011648853399890292, + "loss": 0.1484, + "step": 46319 + }, + { + "epoch": 0.4020798430569179, + "grad_norm": 0.271484375, + "learning_rate": 0.001164855654945222, + "loss": 0.0952, + "step": 46320 + }, + { + "epoch": 0.4020885235371221, + "grad_norm": 0.19921875, + "learning_rate": 0.00116482596983049, + "loss": 0.083, + "step": 46321 + }, + { + "epoch": 0.40209720401732624, + "grad_norm": 0.330078125, + "learning_rate": 0.001164796284644865, + "loss": 0.1187, + "step": 46322 + }, + { + "epoch": 0.40210588449753043, + "grad_norm": 0.21875, + "learning_rate": 0.00116476659938838, + "loss": 0.0811, + "step": 46323 + }, + { + "epoch": 0.40211456497773457, + "grad_norm": 0.353515625, + "learning_rate": 0.001164736914061067, + "loss": 0.0869, + "step": 46324 + }, + { + "epoch": 0.40212324545793876, + "grad_norm": 0.15234375, + "learning_rate": 0.001164707228662959, + "loss": 0.0786, + "step": 46325 + }, + { + "epoch": 0.4021319259381429, + "grad_norm": 0.5703125, + "learning_rate": 0.0011646775431940878, + "loss": 0.1001, + "step": 46326 + }, + { + "epoch": 0.40214060641834704, + "grad_norm": 0.53125, + "learning_rate": 0.001164647857654486, + "loss": 0.168, + "step": 46327 + }, + { + "epoch": 0.40214928689855123, + "grad_norm": 0.1630859375, + "learning_rate": 0.001164618172044187, + "loss": 0.1328, + "step": 46328 + }, + { + "epoch": 0.40215796737875537, + "grad_norm": 0.27734375, + "learning_rate": 0.001164588486363222, + "loss": 0.0767, + "step": 46329 + }, + { + "epoch": 0.40216664785895956, + "grad_norm": 0.10302734375, + "learning_rate": 0.0011645588006116239, + "loss": 0.1162, + "step": 46330 + }, + { + "epoch": 0.4021753283391637, + "grad_norm": 0.171875, + "learning_rate": 0.0011645291147894257, + "loss": 0.0879, + "step": 46331 + }, + { + "epoch": 0.4021840088193679, + "grad_norm": 0.22265625, + "learning_rate": 0.0011644994288966593, + "loss": 0.1182, + "step": 46332 + }, + { + "epoch": 0.402192689299572, + "grad_norm": 0.2421875, + "learning_rate": 0.0011644697429333572, + "loss": 0.1133, + "step": 46333 + }, + { + "epoch": 0.4022013697797762, + "grad_norm": 0.1181640625, + "learning_rate": 0.001164440056899552, + "loss": 0.1157, + "step": 46334 + }, + { + "epoch": 0.40221005025998036, + "grad_norm": 0.20703125, + "learning_rate": 0.001164410370795276, + "loss": 0.0898, + "step": 46335 + }, + { + "epoch": 0.40221873074018455, + "grad_norm": 0.30078125, + "learning_rate": 0.001164380684620562, + "loss": 0.1348, + "step": 46336 + }, + { + "epoch": 0.4022274112203887, + "grad_norm": 0.244140625, + "learning_rate": 0.0011643509983754422, + "loss": 0.1035, + "step": 46337 + }, + { + "epoch": 0.4022360917005929, + "grad_norm": 0.306640625, + "learning_rate": 0.001164321312059949, + "loss": 0.1021, + "step": 46338 + }, + { + "epoch": 0.402244772180797, + "grad_norm": 0.1162109375, + "learning_rate": 0.001164291625674115, + "loss": 0.1221, + "step": 46339 + }, + { + "epoch": 0.4022534526610012, + "grad_norm": 0.126953125, + "learning_rate": 0.0011642619392179725, + "loss": 0.1084, + "step": 46340 + }, + { + "epoch": 0.40226213314120535, + "grad_norm": 0.1025390625, + "learning_rate": 0.0011642322526915546, + "loss": 0.1055, + "step": 46341 + }, + { + "epoch": 0.40227081362140954, + "grad_norm": 0.09765625, + "learning_rate": 0.0011642025660948928, + "loss": 0.0776, + "step": 46342 + }, + { + "epoch": 0.4022794941016137, + "grad_norm": 0.263671875, + "learning_rate": 0.0011641728794280202, + "loss": 0.1084, + "step": 46343 + }, + { + "epoch": 0.40228817458181787, + "grad_norm": 0.36328125, + "learning_rate": 0.001164143192690969, + "loss": 0.0732, + "step": 46344 + }, + { + "epoch": 0.402296855062022, + "grad_norm": 0.126953125, + "learning_rate": 0.0011641135058837722, + "loss": 0.0981, + "step": 46345 + }, + { + "epoch": 0.4023055355422262, + "grad_norm": 0.1484375, + "learning_rate": 0.0011640838190064614, + "loss": 0.127, + "step": 46346 + }, + { + "epoch": 0.40231421602243034, + "grad_norm": 0.2578125, + "learning_rate": 0.0011640541320590695, + "loss": 0.082, + "step": 46347 + }, + { + "epoch": 0.40232289650263453, + "grad_norm": 0.27734375, + "learning_rate": 0.001164024445041629, + "loss": 0.127, + "step": 46348 + }, + { + "epoch": 0.40233157698283867, + "grad_norm": 0.291015625, + "learning_rate": 0.0011639947579541723, + "loss": 0.1113, + "step": 46349 + }, + { + "epoch": 0.40234025746304286, + "grad_norm": 0.318359375, + "learning_rate": 0.0011639650707967323, + "loss": 0.1187, + "step": 46350 + }, + { + "epoch": 0.402348937943247, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011639353835693404, + "loss": 0.0728, + "step": 46351 + }, + { + "epoch": 0.4023576184234512, + "grad_norm": 0.1376953125, + "learning_rate": 0.00116390569627203, + "loss": 0.0938, + "step": 46352 + }, + { + "epoch": 0.40236629890365533, + "grad_norm": 0.255859375, + "learning_rate": 0.0011638760089048336, + "loss": 0.1064, + "step": 46353 + }, + { + "epoch": 0.4023749793838595, + "grad_norm": 0.37890625, + "learning_rate": 0.0011638463214677832, + "loss": 0.1099, + "step": 46354 + }, + { + "epoch": 0.40238365986406366, + "grad_norm": 0.2265625, + "learning_rate": 0.0011638166339609114, + "loss": 0.1611, + "step": 46355 + }, + { + "epoch": 0.40239234034426785, + "grad_norm": 0.109375, + "learning_rate": 0.0011637869463842505, + "loss": 0.0854, + "step": 46356 + }, + { + "epoch": 0.402401020824472, + "grad_norm": 0.365234375, + "learning_rate": 0.001163757258737833, + "loss": 0.124, + "step": 46357 + }, + { + "epoch": 0.4024097013046762, + "grad_norm": 0.2470703125, + "learning_rate": 0.001163727571021692, + "loss": 0.0957, + "step": 46358 + }, + { + "epoch": 0.4024183817848803, + "grad_norm": 0.158203125, + "learning_rate": 0.0011636978832358594, + "loss": 0.1309, + "step": 46359 + }, + { + "epoch": 0.4024270622650845, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011636681953803673, + "loss": 0.1035, + "step": 46360 + }, + { + "epoch": 0.40243574274528865, + "grad_norm": 0.298828125, + "learning_rate": 0.001163638507455249, + "loss": 0.0933, + "step": 46361 + }, + { + "epoch": 0.40244442322549284, + "grad_norm": 0.337890625, + "learning_rate": 0.0011636088194605368, + "loss": 0.1147, + "step": 46362 + }, + { + "epoch": 0.402453103705697, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011635791313962627, + "loss": 0.0967, + "step": 46363 + }, + { + "epoch": 0.4024617841859012, + "grad_norm": 0.173828125, + "learning_rate": 0.0011635494432624593, + "loss": 0.1387, + "step": 46364 + }, + { + "epoch": 0.4024704646661053, + "grad_norm": 0.48046875, + "learning_rate": 0.0011635197550591593, + "loss": 0.1226, + "step": 46365 + }, + { + "epoch": 0.4024791451463095, + "grad_norm": 0.1904296875, + "learning_rate": 0.001163490066786395, + "loss": 0.0854, + "step": 46366 + }, + { + "epoch": 0.40248782562651364, + "grad_norm": 0.455078125, + "learning_rate": 0.001163460378444199, + "loss": 0.1182, + "step": 46367 + }, + { + "epoch": 0.40249650610671783, + "grad_norm": 0.87109375, + "learning_rate": 0.0011634306900326038, + "loss": 0.0918, + "step": 46368 + }, + { + "epoch": 0.40250518658692197, + "grad_norm": 0.5078125, + "learning_rate": 0.0011634010015516415, + "loss": 0.1025, + "step": 46369 + }, + { + "epoch": 0.40251386706712616, + "grad_norm": 0.416015625, + "learning_rate": 0.0011633713130013447, + "loss": 0.1182, + "step": 46370 + }, + { + "epoch": 0.4025225475473303, + "grad_norm": 0.3359375, + "learning_rate": 0.0011633416243817464, + "loss": 0.0991, + "step": 46371 + }, + { + "epoch": 0.4025312280275345, + "grad_norm": 0.181640625, + "learning_rate": 0.0011633119356928784, + "loss": 0.1045, + "step": 46372 + }, + { + "epoch": 0.40253990850773863, + "grad_norm": 0.921875, + "learning_rate": 0.0011632822469347737, + "loss": 0.1104, + "step": 46373 + }, + { + "epoch": 0.4025485889879428, + "grad_norm": 0.447265625, + "learning_rate": 0.0011632525581074641, + "loss": 0.0845, + "step": 46374 + }, + { + "epoch": 0.40255726946814696, + "grad_norm": 0.328125, + "learning_rate": 0.0011632228692109825, + "loss": 0.1445, + "step": 46375 + }, + { + "epoch": 0.40256594994835115, + "grad_norm": 0.29296875, + "learning_rate": 0.0011631931802453615, + "loss": 0.0879, + "step": 46376 + }, + { + "epoch": 0.4025746304285553, + "grad_norm": 0.080078125, + "learning_rate": 0.0011631634912106332, + "loss": 0.0928, + "step": 46377 + }, + { + "epoch": 0.4025833109087595, + "grad_norm": 0.2578125, + "learning_rate": 0.0011631338021068302, + "loss": 0.0854, + "step": 46378 + }, + { + "epoch": 0.4025919913889636, + "grad_norm": 0.412109375, + "learning_rate": 0.0011631041129339856, + "loss": 0.0728, + "step": 46379 + }, + { + "epoch": 0.4026006718691678, + "grad_norm": 1.046875, + "learning_rate": 0.0011630744236921307, + "loss": 0.1084, + "step": 46380 + }, + { + "epoch": 0.40260935234937195, + "grad_norm": 0.189453125, + "learning_rate": 0.0011630447343812988, + "loss": 0.0918, + "step": 46381 + }, + { + "epoch": 0.40261803282957614, + "grad_norm": 0.61328125, + "learning_rate": 0.0011630150450015221, + "loss": 0.0938, + "step": 46382 + }, + { + "epoch": 0.4026267133097803, + "grad_norm": 0.2001953125, + "learning_rate": 0.001162985355552833, + "loss": 0.0801, + "step": 46383 + }, + { + "epoch": 0.4026353937899845, + "grad_norm": 0.2041015625, + "learning_rate": 0.0011629556660352643, + "loss": 0.1147, + "step": 46384 + }, + { + "epoch": 0.4026440742701886, + "grad_norm": 0.466796875, + "learning_rate": 0.0011629259764488478, + "loss": 0.0767, + "step": 46385 + }, + { + "epoch": 0.4026527547503928, + "grad_norm": 0.451171875, + "learning_rate": 0.0011628962867936165, + "loss": 0.1025, + "step": 46386 + }, + { + "epoch": 0.40266143523059694, + "grad_norm": 0.0859375, + "learning_rate": 0.001162866597069603, + "loss": 0.1182, + "step": 46387 + }, + { + "epoch": 0.40267011571080114, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011628369072768394, + "loss": 0.1138, + "step": 46388 + }, + { + "epoch": 0.4026787961910053, + "grad_norm": 0.2470703125, + "learning_rate": 0.0011628072174153586, + "loss": 0.0898, + "step": 46389 + }, + { + "epoch": 0.40268747667120947, + "grad_norm": 0.185546875, + "learning_rate": 0.0011627775274851927, + "loss": 0.0713, + "step": 46390 + }, + { + "epoch": 0.4026961571514136, + "grad_norm": 0.443359375, + "learning_rate": 0.0011627478374863739, + "loss": 0.124, + "step": 46391 + }, + { + "epoch": 0.4027048376316178, + "grad_norm": 0.287109375, + "learning_rate": 0.0011627181474189352, + "loss": 0.1006, + "step": 46392 + }, + { + "epoch": 0.40271351811182193, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011626884572829089, + "loss": 0.1406, + "step": 46393 + }, + { + "epoch": 0.4027221985920261, + "grad_norm": 0.09912109375, + "learning_rate": 0.001162658767078327, + "loss": 0.1196, + "step": 46394 + }, + { + "epoch": 0.40273087907223026, + "grad_norm": 0.130859375, + "learning_rate": 0.0011626290768052229, + "loss": 0.0791, + "step": 46395 + }, + { + "epoch": 0.40273955955243446, + "grad_norm": 0.92578125, + "learning_rate": 0.0011625993864636285, + "loss": 0.106, + "step": 46396 + }, + { + "epoch": 0.4027482400326386, + "grad_norm": 0.1884765625, + "learning_rate": 0.0011625696960535766, + "loss": 0.0967, + "step": 46397 + }, + { + "epoch": 0.4027569205128428, + "grad_norm": 0.1005859375, + "learning_rate": 0.001162540005575099, + "loss": 0.0879, + "step": 46398 + }, + { + "epoch": 0.4027656009930469, + "grad_norm": 0.13671875, + "learning_rate": 0.0011625103150282286, + "loss": 0.1279, + "step": 46399 + }, + { + "epoch": 0.4027742814732511, + "grad_norm": 0.28515625, + "learning_rate": 0.0011624806244129981, + "loss": 0.1396, + "step": 46400 + }, + { + "epoch": 0.40278296195345525, + "grad_norm": 0.279296875, + "learning_rate": 0.0011624509337294397, + "loss": 0.0869, + "step": 46401 + }, + { + "epoch": 0.40279164243365945, + "grad_norm": 0.279296875, + "learning_rate": 0.001162421242977586, + "loss": 0.0957, + "step": 46402 + }, + { + "epoch": 0.4028003229138636, + "grad_norm": 0.419921875, + "learning_rate": 0.001162391552157469, + "loss": 0.1025, + "step": 46403 + }, + { + "epoch": 0.4028090033940678, + "grad_norm": 0.2001953125, + "learning_rate": 0.0011623618612691216, + "loss": 0.0981, + "step": 46404 + }, + { + "epoch": 0.4028176838742719, + "grad_norm": 0.263671875, + "learning_rate": 0.0011623321703125764, + "loss": 0.1221, + "step": 46405 + }, + { + "epoch": 0.4028263643544761, + "grad_norm": 0.2578125, + "learning_rate": 0.0011623024792878657, + "loss": 0.1602, + "step": 46406 + }, + { + "epoch": 0.40283504483468024, + "grad_norm": 0.240234375, + "learning_rate": 0.0011622727881950218, + "loss": 0.0801, + "step": 46407 + }, + { + "epoch": 0.40284372531488444, + "grad_norm": 0.76953125, + "learning_rate": 0.0011622430970340775, + "loss": 0.3555, + "step": 46408 + }, + { + "epoch": 0.4028524057950886, + "grad_norm": 0.51171875, + "learning_rate": 0.001162213405805065, + "loss": 0.1348, + "step": 46409 + }, + { + "epoch": 0.40286108627529277, + "grad_norm": 0.1640625, + "learning_rate": 0.0011621837145080168, + "loss": 0.1055, + "step": 46410 + }, + { + "epoch": 0.4028697667554969, + "grad_norm": 0.33203125, + "learning_rate": 0.0011621540231429655, + "loss": 0.1504, + "step": 46411 + }, + { + "epoch": 0.4028784472357011, + "grad_norm": 0.40234375, + "learning_rate": 0.0011621243317099432, + "loss": 0.1084, + "step": 46412 + }, + { + "epoch": 0.40288712771590524, + "grad_norm": 0.1640625, + "learning_rate": 0.0011620946402089829, + "loss": 0.0967, + "step": 46413 + }, + { + "epoch": 0.40289580819610943, + "grad_norm": 0.33203125, + "learning_rate": 0.001162064948640117, + "loss": 0.0593, + "step": 46414 + }, + { + "epoch": 0.40290448867631357, + "grad_norm": 0.146484375, + "learning_rate": 0.0011620352570033776, + "loss": 0.1226, + "step": 46415 + }, + { + "epoch": 0.40291316915651776, + "grad_norm": 0.12890625, + "learning_rate": 0.0011620055652987972, + "loss": 0.0933, + "step": 46416 + }, + { + "epoch": 0.4029218496367219, + "grad_norm": 0.08251953125, + "learning_rate": 0.0011619758735264087, + "loss": 0.1289, + "step": 46417 + }, + { + "epoch": 0.4029305301169261, + "grad_norm": 0.181640625, + "learning_rate": 0.0011619461816862442, + "loss": 0.0864, + "step": 46418 + }, + { + "epoch": 0.4029392105971302, + "grad_norm": 0.21484375, + "learning_rate": 0.0011619164897783366, + "loss": 0.0986, + "step": 46419 + }, + { + "epoch": 0.4029478910773344, + "grad_norm": 0.0810546875, + "learning_rate": 0.001161886797802718, + "loss": 0.0806, + "step": 46420 + }, + { + "epoch": 0.40295657155753856, + "grad_norm": 0.41796875, + "learning_rate": 0.0011618571057594204, + "loss": 0.103, + "step": 46421 + }, + { + "epoch": 0.40296525203774275, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011618274136484773, + "loss": 0.1074, + "step": 46422 + }, + { + "epoch": 0.4029739325179469, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011617977214699203, + "loss": 0.0898, + "step": 46423 + }, + { + "epoch": 0.4029826129981511, + "grad_norm": 0.482421875, + "learning_rate": 0.0011617680292237828, + "loss": 0.104, + "step": 46424 + }, + { + "epoch": 0.4029912934783552, + "grad_norm": 0.26171875, + "learning_rate": 0.0011617383369100958, + "loss": 0.105, + "step": 46425 + }, + { + "epoch": 0.4029999739585594, + "grad_norm": 0.205078125, + "learning_rate": 0.0011617086445288936, + "loss": 0.1191, + "step": 46426 + }, + { + "epoch": 0.40300865443876355, + "grad_norm": 0.189453125, + "learning_rate": 0.0011616789520802074, + "loss": 0.1338, + "step": 46427 + }, + { + "epoch": 0.40301733491896774, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011616492595640702, + "loss": 0.1172, + "step": 46428 + }, + { + "epoch": 0.4030260153991719, + "grad_norm": 0.322265625, + "learning_rate": 0.0011616195669805138, + "loss": 0.1094, + "step": 46429 + }, + { + "epoch": 0.40303469587937607, + "grad_norm": 0.349609375, + "learning_rate": 0.0011615898743295716, + "loss": 0.0928, + "step": 46430 + }, + { + "epoch": 0.4030433763595802, + "grad_norm": 0.318359375, + "learning_rate": 0.0011615601816112756, + "loss": 0.084, + "step": 46431 + }, + { + "epoch": 0.4030520568397844, + "grad_norm": 0.6640625, + "learning_rate": 0.001161530488825658, + "loss": 0.1021, + "step": 46432 + }, + { + "epoch": 0.40306073731998854, + "grad_norm": 0.1298828125, + "learning_rate": 0.001161500795972752, + "loss": 0.1094, + "step": 46433 + }, + { + "epoch": 0.40306941780019273, + "grad_norm": 0.333984375, + "learning_rate": 0.0011614711030525893, + "loss": 0.0742, + "step": 46434 + }, + { + "epoch": 0.40307809828039687, + "grad_norm": 0.294921875, + "learning_rate": 0.001161441410065203, + "loss": 0.1123, + "step": 46435 + }, + { + "epoch": 0.40308677876060106, + "grad_norm": 0.146484375, + "learning_rate": 0.0011614117170106251, + "loss": 0.0947, + "step": 46436 + }, + { + "epoch": 0.4030954592408052, + "grad_norm": 0.111328125, + "learning_rate": 0.0011613820238888886, + "loss": 0.0908, + "step": 46437 + }, + { + "epoch": 0.4031041397210094, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011613523307000254, + "loss": 0.1113, + "step": 46438 + }, + { + "epoch": 0.40311282020121353, + "grad_norm": 0.09521484375, + "learning_rate": 0.001161322637444068, + "loss": 0.0967, + "step": 46439 + }, + { + "epoch": 0.4031215006814177, + "grad_norm": 0.255859375, + "learning_rate": 0.0011612929441210493, + "loss": 0.1094, + "step": 46440 + }, + { + "epoch": 0.40313018116162186, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011612632507310018, + "loss": 0.1045, + "step": 46441 + }, + { + "epoch": 0.40313886164182605, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011612335572739575, + "loss": 0.0752, + "step": 46442 + }, + { + "epoch": 0.4031475421220302, + "grad_norm": 0.455078125, + "learning_rate": 0.0011612038637499487, + "loss": 0.1162, + "step": 46443 + }, + { + "epoch": 0.4031562226022344, + "grad_norm": 0.1982421875, + "learning_rate": 0.001161174170159009, + "loss": 0.1016, + "step": 46444 + }, + { + "epoch": 0.4031649030824385, + "grad_norm": 0.138671875, + "learning_rate": 0.0011611444765011697, + "loss": 0.1143, + "step": 46445 + }, + { + "epoch": 0.4031735835626427, + "grad_norm": 0.62109375, + "learning_rate": 0.0011611147827764639, + "loss": 0.0698, + "step": 46446 + }, + { + "epoch": 0.40318226404284685, + "grad_norm": 0.302734375, + "learning_rate": 0.0011610850889849237, + "loss": 0.0996, + "step": 46447 + }, + { + "epoch": 0.40319094452305104, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011610553951265819, + "loss": 0.1094, + "step": 46448 + }, + { + "epoch": 0.4031996250032552, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011610257012014708, + "loss": 0.0981, + "step": 46449 + }, + { + "epoch": 0.4032083054834594, + "grad_norm": 0.1484375, + "learning_rate": 0.0011609960072096232, + "loss": 0.1099, + "step": 46450 + }, + { + "epoch": 0.4032169859636635, + "grad_norm": 0.09765625, + "learning_rate": 0.0011609663131510707, + "loss": 0.1006, + "step": 46451 + }, + { + "epoch": 0.40322566644386765, + "grad_norm": 3.125, + "learning_rate": 0.0011609366190258466, + "loss": 0.4023, + "step": 46452 + }, + { + "epoch": 0.40323434692407184, + "grad_norm": 0.2294921875, + "learning_rate": 0.0011609069248339832, + "loss": 0.0825, + "step": 46453 + }, + { + "epoch": 0.403243027404276, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011608772305755132, + "loss": 0.1016, + "step": 46454 + }, + { + "epoch": 0.40325170788448017, + "grad_norm": 0.134765625, + "learning_rate": 0.0011608475362504683, + "loss": 0.1191, + "step": 46455 + }, + { + "epoch": 0.4032603883646843, + "grad_norm": 0.5546875, + "learning_rate": 0.0011608178418588814, + "loss": 0.1396, + "step": 46456 + }, + { + "epoch": 0.4032690688448885, + "grad_norm": 0.287109375, + "learning_rate": 0.0011607881474007852, + "loss": 0.1016, + "step": 46457 + }, + { + "epoch": 0.40327774932509264, + "grad_norm": 0.328125, + "learning_rate": 0.0011607584528762122, + "loss": 0.0947, + "step": 46458 + }, + { + "epoch": 0.40328642980529683, + "grad_norm": 0.169921875, + "learning_rate": 0.0011607287582851948, + "loss": 0.1108, + "step": 46459 + }, + { + "epoch": 0.40329511028550097, + "grad_norm": 0.40234375, + "learning_rate": 0.0011606990636277648, + "loss": 0.0962, + "step": 46460 + }, + { + "epoch": 0.40330379076570516, + "grad_norm": 0.53515625, + "learning_rate": 0.0011606693689039556, + "loss": 0.1289, + "step": 46461 + }, + { + "epoch": 0.4033124712459093, + "grad_norm": 0.271484375, + "learning_rate": 0.001160639674113799, + "loss": 0.1226, + "step": 46462 + }, + { + "epoch": 0.4033211517261135, + "grad_norm": 0.1884765625, + "learning_rate": 0.0011606099792573282, + "loss": 0.1396, + "step": 46463 + }, + { + "epoch": 0.40332983220631763, + "grad_norm": 0.220703125, + "learning_rate": 0.0011605802843345748, + "loss": 0.0894, + "step": 46464 + }, + { + "epoch": 0.4033385126865218, + "grad_norm": 0.09423828125, + "learning_rate": 0.001160550589345572, + "loss": 0.085, + "step": 46465 + }, + { + "epoch": 0.40334719316672596, + "grad_norm": 0.166015625, + "learning_rate": 0.0011605208942903516, + "loss": 0.1221, + "step": 46466 + }, + { + "epoch": 0.40335587364693015, + "grad_norm": 0.4453125, + "learning_rate": 0.0011604911991689468, + "loss": 0.1152, + "step": 46467 + }, + { + "epoch": 0.4033645541271343, + "grad_norm": 0.181640625, + "learning_rate": 0.0011604615039813898, + "loss": 0.1367, + "step": 46468 + }, + { + "epoch": 0.4033732346073385, + "grad_norm": 0.2451171875, + "learning_rate": 0.0011604318087277126, + "loss": 0.0928, + "step": 46469 + }, + { + "epoch": 0.4033819150875426, + "grad_norm": 0.39453125, + "learning_rate": 0.0011604021134079484, + "loss": 0.0967, + "step": 46470 + }, + { + "epoch": 0.4033905955677468, + "grad_norm": 0.8359375, + "learning_rate": 0.001160372418022129, + "loss": 0.084, + "step": 46471 + }, + { + "epoch": 0.40339927604795095, + "grad_norm": 0.13671875, + "learning_rate": 0.0011603427225702879, + "loss": 0.1289, + "step": 46472 + }, + { + "epoch": 0.40340795652815514, + "grad_norm": 0.2734375, + "learning_rate": 0.0011603130270524565, + "loss": 0.1025, + "step": 46473 + }, + { + "epoch": 0.4034166370083593, + "grad_norm": 0.294921875, + "learning_rate": 0.0011602833314686676, + "loss": 0.1582, + "step": 46474 + }, + { + "epoch": 0.4034253174885635, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011602536358189539, + "loss": 0.0942, + "step": 46475 + }, + { + "epoch": 0.4034339979687676, + "grad_norm": 0.2490234375, + "learning_rate": 0.0011602239401033478, + "loss": 0.0967, + "step": 46476 + }, + { + "epoch": 0.4034426784489718, + "grad_norm": 0.203125, + "learning_rate": 0.0011601942443218813, + "loss": 0.1074, + "step": 46477 + }, + { + "epoch": 0.40345135892917594, + "grad_norm": 0.357421875, + "learning_rate": 0.001160164548474588, + "loss": 0.0977, + "step": 46478 + }, + { + "epoch": 0.40346003940938013, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011601348525614992, + "loss": 0.1104, + "step": 46479 + }, + { + "epoch": 0.40346871988958427, + "grad_norm": 1.265625, + "learning_rate": 0.0011601051565826477, + "loss": 0.1182, + "step": 46480 + }, + { + "epoch": 0.40347740036978846, + "grad_norm": 0.0830078125, + "learning_rate": 0.0011600754605380664, + "loss": 0.0947, + "step": 46481 + }, + { + "epoch": 0.4034860808499926, + "grad_norm": 0.380859375, + "learning_rate": 0.0011600457644277873, + "loss": 0.0933, + "step": 46482 + }, + { + "epoch": 0.4034947613301968, + "grad_norm": 0.1669921875, + "learning_rate": 0.001160016068251843, + "loss": 0.1104, + "step": 46483 + }, + { + "epoch": 0.40350344181040093, + "grad_norm": 0.279296875, + "learning_rate": 0.0011599863720102661, + "loss": 0.082, + "step": 46484 + }, + { + "epoch": 0.4035121222906051, + "grad_norm": 0.15234375, + "learning_rate": 0.0011599566757030894, + "loss": 0.0801, + "step": 46485 + }, + { + "epoch": 0.40352080277080926, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011599269793303442, + "loss": 0.1016, + "step": 46486 + }, + { + "epoch": 0.40352948325101345, + "grad_norm": 0.671875, + "learning_rate": 0.0011598972828920644, + "loss": 0.1318, + "step": 46487 + }, + { + "epoch": 0.4035381637312176, + "grad_norm": 0.443359375, + "learning_rate": 0.0011598675863882814, + "loss": 0.123, + "step": 46488 + }, + { + "epoch": 0.4035468442114218, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011598378898190285, + "loss": 0.0693, + "step": 46489 + }, + { + "epoch": 0.4035555246916259, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011598081931843377, + "loss": 0.0811, + "step": 46490 + }, + { + "epoch": 0.4035642051718301, + "grad_norm": 0.08642578125, + "learning_rate": 0.001159778496484241, + "loss": 0.0908, + "step": 46491 + }, + { + "epoch": 0.40357288565203425, + "grad_norm": 0.1640625, + "learning_rate": 0.0011597487997187718, + "loss": 0.1025, + "step": 46492 + }, + { + "epoch": 0.40358156613223845, + "grad_norm": 0.0732421875, + "learning_rate": 0.0011597191028879625, + "loss": 0.0752, + "step": 46493 + }, + { + "epoch": 0.4035902466124426, + "grad_norm": 0.125, + "learning_rate": 0.0011596894059918452, + "loss": 0.083, + "step": 46494 + }, + { + "epoch": 0.4035989270926468, + "grad_norm": 0.080078125, + "learning_rate": 0.0011596597090304524, + "loss": 0.0869, + "step": 46495 + }, + { + "epoch": 0.4036076075728509, + "grad_norm": 0.154296875, + "learning_rate": 0.0011596300120038166, + "loss": 0.0825, + "step": 46496 + }, + { + "epoch": 0.4036162880530551, + "grad_norm": 0.34765625, + "learning_rate": 0.0011596003149119702, + "loss": 0.1221, + "step": 46497 + }, + { + "epoch": 0.40362496853325924, + "grad_norm": 0.1123046875, + "learning_rate": 0.0011595706177549461, + "loss": 0.0879, + "step": 46498 + }, + { + "epoch": 0.40363364901346344, + "grad_norm": 0.2353515625, + "learning_rate": 0.001159540920532776, + "loss": 0.084, + "step": 46499 + }, + { + "epoch": 0.4036423294936676, + "grad_norm": 0.138671875, + "learning_rate": 0.001159511223245493, + "loss": 0.104, + "step": 46500 + }, + { + "epoch": 0.40365100997387177, + "grad_norm": 0.28515625, + "learning_rate": 0.0011594815258931296, + "loss": 0.1367, + "step": 46501 + }, + { + "epoch": 0.4036596904540759, + "grad_norm": 0.609375, + "learning_rate": 0.001159451828475718, + "loss": 0.2949, + "step": 46502 + }, + { + "epoch": 0.4036683709342801, + "grad_norm": 0.408203125, + "learning_rate": 0.0011594221309932907, + "loss": 0.1182, + "step": 46503 + }, + { + "epoch": 0.40367705141448423, + "grad_norm": 0.208984375, + "learning_rate": 0.0011593924334458805, + "loss": 0.0884, + "step": 46504 + }, + { + "epoch": 0.4036857318946884, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011593627358335195, + "loss": 0.1064, + "step": 46505 + }, + { + "epoch": 0.40369441237489256, + "grad_norm": 0.23828125, + "learning_rate": 0.0011593330381562401, + "loss": 0.1592, + "step": 46506 + }, + { + "epoch": 0.40370309285509676, + "grad_norm": 0.205078125, + "learning_rate": 0.0011593033404140752, + "loss": 0.1211, + "step": 46507 + }, + { + "epoch": 0.4037117733353009, + "grad_norm": 0.515625, + "learning_rate": 0.0011592736426070571, + "loss": 0.1025, + "step": 46508 + }, + { + "epoch": 0.4037204538155051, + "grad_norm": 0.09765625, + "learning_rate": 0.001159243944735218, + "loss": 0.0615, + "step": 46509 + }, + { + "epoch": 0.4037291342957092, + "grad_norm": 0.103515625, + "learning_rate": 0.0011592142467985909, + "loss": 0.0771, + "step": 46510 + }, + { + "epoch": 0.4037378147759134, + "grad_norm": 0.3046875, + "learning_rate": 0.0011591845487972077, + "loss": 0.1079, + "step": 46511 + }, + { + "epoch": 0.40374649525611755, + "grad_norm": 0.451171875, + "learning_rate": 0.0011591548507311011, + "loss": 0.0791, + "step": 46512 + }, + { + "epoch": 0.40375517573632175, + "grad_norm": 0.07275390625, + "learning_rate": 0.0011591251526003041, + "loss": 0.0698, + "step": 46513 + }, + { + "epoch": 0.4037638562165259, + "grad_norm": 0.169921875, + "learning_rate": 0.0011590954544048483, + "loss": 0.1162, + "step": 46514 + }, + { + "epoch": 0.4037725366967301, + "grad_norm": 0.2275390625, + "learning_rate": 0.0011590657561447672, + "loss": 0.105, + "step": 46515 + }, + { + "epoch": 0.4037812171769342, + "grad_norm": 0.138671875, + "learning_rate": 0.001159036057820092, + "loss": 0.1182, + "step": 46516 + }, + { + "epoch": 0.4037898976571384, + "grad_norm": 0.333984375, + "learning_rate": 0.0011590063594308561, + "loss": 0.0752, + "step": 46517 + }, + { + "epoch": 0.40379857813734255, + "grad_norm": 0.2138671875, + "learning_rate": 0.001158976660977092, + "loss": 0.1133, + "step": 46518 + }, + { + "epoch": 0.40380725861754674, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011589469624588313, + "loss": 0.0908, + "step": 46519 + }, + { + "epoch": 0.4038159390977509, + "grad_norm": 0.28515625, + "learning_rate": 0.0011589172638761074, + "loss": 0.0962, + "step": 46520 + }, + { + "epoch": 0.40382461957795507, + "grad_norm": 0.9609375, + "learning_rate": 0.0011588875652289526, + "loss": 0.1064, + "step": 46521 + }, + { + "epoch": 0.4038333000581592, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011588578665173992, + "loss": 0.1011, + "step": 46522 + }, + { + "epoch": 0.4038419805383634, + "grad_norm": 0.15234375, + "learning_rate": 0.0011588281677414795, + "loss": 0.1182, + "step": 46523 + }, + { + "epoch": 0.40385066101856754, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011587984689012264, + "loss": 0.0703, + "step": 46524 + }, + { + "epoch": 0.40385934149877173, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011587687699966722, + "loss": 0.0884, + "step": 46525 + }, + { + "epoch": 0.40386802197897587, + "grad_norm": 0.11376953125, + "learning_rate": 0.0011587390710278494, + "loss": 0.1006, + "step": 46526 + }, + { + "epoch": 0.40387670245918006, + "grad_norm": 0.298828125, + "learning_rate": 0.0011587093719947904, + "loss": 0.1426, + "step": 46527 + }, + { + "epoch": 0.4038853829393842, + "grad_norm": 0.51953125, + "learning_rate": 0.0011586796728975275, + "loss": 0.1309, + "step": 46528 + }, + { + "epoch": 0.4038940634195884, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011586499737360936, + "loss": 0.0986, + "step": 46529 + }, + { + "epoch": 0.4039027438997925, + "grad_norm": 0.24609375, + "learning_rate": 0.001158620274510521, + "loss": 0.0889, + "step": 46530 + }, + { + "epoch": 0.4039114243799967, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011585905752208418, + "loss": 0.1377, + "step": 46531 + }, + { + "epoch": 0.40392010486020086, + "grad_norm": 0.1953125, + "learning_rate": 0.0011585608758670895, + "loss": 0.1235, + "step": 46532 + }, + { + "epoch": 0.40392878534040505, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011585311764492952, + "loss": 0.0801, + "step": 46533 + }, + { + "epoch": 0.4039374658206092, + "grad_norm": 0.66015625, + "learning_rate": 0.0011585014769674923, + "loss": 0.1475, + "step": 46534 + }, + { + "epoch": 0.4039461463008134, + "grad_norm": 0.451171875, + "learning_rate": 0.0011584717774217133, + "loss": 0.1514, + "step": 46535 + }, + { + "epoch": 0.4039548267810175, + "grad_norm": 0.41796875, + "learning_rate": 0.0011584420778119901, + "loss": 0.1309, + "step": 46536 + }, + { + "epoch": 0.4039635072612217, + "grad_norm": 0.095703125, + "learning_rate": 0.0011584123781383559, + "loss": 0.1221, + "step": 46537 + }, + { + "epoch": 0.40397218774142585, + "grad_norm": 0.134765625, + "learning_rate": 0.0011583826784008423, + "loss": 0.1123, + "step": 46538 + }, + { + "epoch": 0.40398086822163004, + "grad_norm": 0.6640625, + "learning_rate": 0.0011583529785994826, + "loss": 0.1162, + "step": 46539 + }, + { + "epoch": 0.4039895487018342, + "grad_norm": 0.1083984375, + "learning_rate": 0.001158323278734309, + "loss": 0.0967, + "step": 46540 + }, + { + "epoch": 0.40399822918203837, + "grad_norm": 0.11376953125, + "learning_rate": 0.001158293578805354, + "loss": 0.0894, + "step": 46541 + }, + { + "epoch": 0.4040069096622425, + "grad_norm": 0.435546875, + "learning_rate": 0.00115826387881265, + "loss": 0.0854, + "step": 46542 + }, + { + "epoch": 0.4040155901424467, + "grad_norm": 0.3515625, + "learning_rate": 0.0011582341787562294, + "loss": 0.1094, + "step": 46543 + }, + { + "epoch": 0.40402427062265084, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011582044786361248, + "loss": 0.103, + "step": 46544 + }, + { + "epoch": 0.40403295110285503, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011581747784523686, + "loss": 0.084, + "step": 46545 + }, + { + "epoch": 0.40404163158305917, + "grad_norm": 0.326171875, + "learning_rate": 0.0011581450782049934, + "loss": 0.1543, + "step": 46546 + }, + { + "epoch": 0.40405031206326336, + "grad_norm": 0.333984375, + "learning_rate": 0.0011581153778940317, + "loss": 0.1484, + "step": 46547 + }, + { + "epoch": 0.4040589925434675, + "grad_norm": 0.267578125, + "learning_rate": 0.0011580856775195156, + "loss": 0.0869, + "step": 46548 + }, + { + "epoch": 0.4040676730236717, + "grad_norm": 0.185546875, + "learning_rate": 0.001158055977081478, + "loss": 0.0713, + "step": 46549 + }, + { + "epoch": 0.40407635350387583, + "grad_norm": 0.2421875, + "learning_rate": 0.0011580262765799511, + "loss": 0.0986, + "step": 46550 + }, + { + "epoch": 0.40408503398408, + "grad_norm": 0.162109375, + "learning_rate": 0.0011579965760149678, + "loss": 0.0967, + "step": 46551 + }, + { + "epoch": 0.40409371446428416, + "grad_norm": 0.86328125, + "learning_rate": 0.0011579668753865606, + "loss": 0.1211, + "step": 46552 + }, + { + "epoch": 0.40410239494448835, + "grad_norm": 0.98828125, + "learning_rate": 0.0011579371746947614, + "loss": 0.1328, + "step": 46553 + }, + { + "epoch": 0.4041110754246925, + "grad_norm": 0.08056640625, + "learning_rate": 0.001157907473939603, + "loss": 0.0776, + "step": 46554 + }, + { + "epoch": 0.4041197559048967, + "grad_norm": 0.259765625, + "learning_rate": 0.0011578777731211173, + "loss": 0.0781, + "step": 46555 + }, + { + "epoch": 0.4041284363851008, + "grad_norm": 0.28125, + "learning_rate": 0.0011578480722393382, + "loss": 0.0991, + "step": 46556 + }, + { + "epoch": 0.404137116865305, + "grad_norm": 0.330078125, + "learning_rate": 0.0011578183712942969, + "loss": 0.1235, + "step": 46557 + }, + { + "epoch": 0.40414579734550915, + "grad_norm": 0.126953125, + "learning_rate": 0.0011577886702860262, + "loss": 0.0972, + "step": 46558 + }, + { + "epoch": 0.40415447782571334, + "grad_norm": 0.1875, + "learning_rate": 0.0011577589692145587, + "loss": 0.1523, + "step": 46559 + }, + { + "epoch": 0.4041631583059175, + "grad_norm": 0.87109375, + "learning_rate": 0.0011577292680799269, + "loss": 0.0977, + "step": 46560 + }, + { + "epoch": 0.4041718387861217, + "grad_norm": 0.197265625, + "learning_rate": 0.0011576995668821635, + "loss": 0.1045, + "step": 46561 + }, + { + "epoch": 0.4041805192663258, + "grad_norm": 0.314453125, + "learning_rate": 0.0011576698656213004, + "loss": 0.0762, + "step": 46562 + }, + { + "epoch": 0.40418919974653, + "grad_norm": 0.35546875, + "learning_rate": 0.0011576401642973706, + "loss": 0.0947, + "step": 46563 + }, + { + "epoch": 0.40419788022673414, + "grad_norm": 0.1533203125, + "learning_rate": 0.0011576104629104065, + "loss": 0.1235, + "step": 46564 + }, + { + "epoch": 0.40420656070693833, + "grad_norm": 0.189453125, + "learning_rate": 0.0011575807614604403, + "loss": 0.0781, + "step": 46565 + }, + { + "epoch": 0.40421524118714247, + "grad_norm": 0.38671875, + "learning_rate": 0.0011575510599475048, + "loss": 0.1104, + "step": 46566 + }, + { + "epoch": 0.40422392166734666, + "grad_norm": 0.078125, + "learning_rate": 0.001157521358371632, + "loss": 0.1074, + "step": 46567 + }, + { + "epoch": 0.4042326021475508, + "grad_norm": 0.267578125, + "learning_rate": 0.0011574916567328546, + "loss": 0.0781, + "step": 46568 + }, + { + "epoch": 0.404241282627755, + "grad_norm": 0.197265625, + "learning_rate": 0.0011574619550312058, + "loss": 0.2051, + "step": 46569 + }, + { + "epoch": 0.40424996310795913, + "grad_norm": 0.162109375, + "learning_rate": 0.001157432253266717, + "loss": 0.0889, + "step": 46570 + }, + { + "epoch": 0.4042586435881633, + "grad_norm": 0.5, + "learning_rate": 0.0011574025514394213, + "loss": 0.0654, + "step": 46571 + }, + { + "epoch": 0.40426732406836746, + "grad_norm": 0.3125, + "learning_rate": 0.001157372849549351, + "loss": 0.083, + "step": 46572 + }, + { + "epoch": 0.40427600454857165, + "grad_norm": 0.181640625, + "learning_rate": 0.0011573431475965386, + "loss": 0.1123, + "step": 46573 + }, + { + "epoch": 0.4042846850287758, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011573134455810168, + "loss": 0.1113, + "step": 46574 + }, + { + "epoch": 0.40429336550897993, + "grad_norm": 0.1943359375, + "learning_rate": 0.0011572837435028177, + "loss": 0.1494, + "step": 46575 + }, + { + "epoch": 0.4043020459891841, + "grad_norm": 0.94140625, + "learning_rate": 0.0011572540413619737, + "loss": 0.1226, + "step": 46576 + }, + { + "epoch": 0.40431072646938826, + "grad_norm": 0.458984375, + "learning_rate": 0.0011572243391585178, + "loss": 0.0947, + "step": 46577 + }, + { + "epoch": 0.40431940694959245, + "grad_norm": 0.77734375, + "learning_rate": 0.0011571946368924822, + "loss": 0.1406, + "step": 46578 + }, + { + "epoch": 0.4043280874297966, + "grad_norm": 0.130859375, + "learning_rate": 0.0011571649345638994, + "loss": 0.0947, + "step": 46579 + }, + { + "epoch": 0.4043367679100008, + "grad_norm": 0.107421875, + "learning_rate": 0.0011571352321728021, + "loss": 0.126, + "step": 46580 + }, + { + "epoch": 0.4043454483902049, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011571055297192221, + "loss": 0.1035, + "step": 46581 + }, + { + "epoch": 0.4043541288704091, + "grad_norm": 0.515625, + "learning_rate": 0.0011570758272031927, + "loss": 0.0869, + "step": 46582 + }, + { + "epoch": 0.40436280935061325, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011570461246247462, + "loss": 0.0928, + "step": 46583 + }, + { + "epoch": 0.40437148983081744, + "grad_norm": 0.90234375, + "learning_rate": 0.0011570164219839147, + "loss": 0.1279, + "step": 46584 + }, + { + "epoch": 0.4043801703110216, + "grad_norm": 0.6484375, + "learning_rate": 0.0011569867192807307, + "loss": 0.1094, + "step": 46585 + }, + { + "epoch": 0.4043888507912258, + "grad_norm": 0.265625, + "learning_rate": 0.001156957016515227, + "loss": 0.0752, + "step": 46586 + }, + { + "epoch": 0.4043975312714299, + "grad_norm": 0.33984375, + "learning_rate": 0.001156927313687436, + "loss": 0.0684, + "step": 46587 + }, + { + "epoch": 0.4044062117516341, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011568976107973901, + "loss": 0.0889, + "step": 46588 + }, + { + "epoch": 0.40441489223183824, + "grad_norm": 0.53515625, + "learning_rate": 0.0011568679078451218, + "loss": 0.0669, + "step": 46589 + }, + { + "epoch": 0.40442357271204243, + "grad_norm": 0.12451171875, + "learning_rate": 0.0011568382048306638, + "loss": 0.1079, + "step": 46590 + }, + { + "epoch": 0.40443225319224657, + "grad_norm": 0.134765625, + "learning_rate": 0.0011568085017540483, + "loss": 0.0713, + "step": 46591 + }, + { + "epoch": 0.40444093367245076, + "grad_norm": 0.169921875, + "learning_rate": 0.001156778798615308, + "loss": 0.0967, + "step": 46592 + }, + { + "epoch": 0.4044496141526549, + "grad_norm": 0.6171875, + "learning_rate": 0.0011567490954144753, + "loss": 0.084, + "step": 46593 + }, + { + "epoch": 0.4044582946328591, + "grad_norm": 0.388671875, + "learning_rate": 0.0011567193921515823, + "loss": 0.0928, + "step": 46594 + }, + { + "epoch": 0.40446697511306323, + "grad_norm": 0.169921875, + "learning_rate": 0.001156689688826662, + "loss": 0.1582, + "step": 46595 + }, + { + "epoch": 0.4044756555932674, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011566599854397465, + "loss": 0.1182, + "step": 46596 + }, + { + "epoch": 0.40448433607347156, + "grad_norm": 0.65625, + "learning_rate": 0.0011566302819908687, + "loss": 0.0918, + "step": 46597 + }, + { + "epoch": 0.40449301655367575, + "grad_norm": 0.203125, + "learning_rate": 0.0011566005784800608, + "loss": 0.166, + "step": 46598 + }, + { + "epoch": 0.4045016970338799, + "grad_norm": 0.177734375, + "learning_rate": 0.0011565708749073555, + "loss": 0.1387, + "step": 46599 + }, + { + "epoch": 0.4045103775140841, + "grad_norm": 0.1982421875, + "learning_rate": 0.001156541171272785, + "loss": 0.0762, + "step": 46600 + }, + { + "epoch": 0.4045190579942882, + "grad_norm": 0.169921875, + "learning_rate": 0.0011565114675763822, + "loss": 0.1069, + "step": 46601 + }, + { + "epoch": 0.4045277384744924, + "grad_norm": 0.431640625, + "learning_rate": 0.001156481763818179, + "loss": 0.0835, + "step": 46602 + }, + { + "epoch": 0.40453641895469655, + "grad_norm": 0.18359375, + "learning_rate": 0.0011564520599982085, + "loss": 0.0928, + "step": 46603 + }, + { + "epoch": 0.40454509943490075, + "grad_norm": 0.373046875, + "learning_rate": 0.0011564223561165027, + "loss": 0.1211, + "step": 46604 + }, + { + "epoch": 0.4045537799151049, + "grad_norm": 0.3359375, + "learning_rate": 0.001156392652173094, + "loss": 0.1143, + "step": 46605 + }, + { + "epoch": 0.4045624603953091, + "grad_norm": 0.275390625, + "learning_rate": 0.0011563629481680154, + "loss": 0.085, + "step": 46606 + }, + { + "epoch": 0.4045711408755132, + "grad_norm": 0.55078125, + "learning_rate": 0.0011563332441012991, + "loss": 0.1094, + "step": 46607 + }, + { + "epoch": 0.4045798213557174, + "grad_norm": 0.166015625, + "learning_rate": 0.0011563035399729776, + "loss": 0.1089, + "step": 46608 + }, + { + "epoch": 0.40458850183592154, + "grad_norm": 0.1953125, + "learning_rate": 0.0011562738357830833, + "loss": 0.083, + "step": 46609 + }, + { + "epoch": 0.40459718231612574, + "grad_norm": 0.29296875, + "learning_rate": 0.001156244131531649, + "loss": 0.1338, + "step": 46610 + }, + { + "epoch": 0.4046058627963299, + "grad_norm": 0.43359375, + "learning_rate": 0.0011562144272187068, + "loss": 0.0957, + "step": 46611 + }, + { + "epoch": 0.40461454327653407, + "grad_norm": 0.216796875, + "learning_rate": 0.0011561847228442894, + "loss": 0.0957, + "step": 46612 + }, + { + "epoch": 0.4046232237567382, + "grad_norm": 0.26171875, + "learning_rate": 0.0011561550184084294, + "loss": 0.0889, + "step": 46613 + }, + { + "epoch": 0.4046319042369424, + "grad_norm": 0.92578125, + "learning_rate": 0.0011561253139111588, + "loss": 0.1533, + "step": 46614 + }, + { + "epoch": 0.40464058471714653, + "grad_norm": 0.055419921875, + "learning_rate": 0.0011560956093525105, + "loss": 0.0562, + "step": 46615 + }, + { + "epoch": 0.4046492651973507, + "grad_norm": 0.2890625, + "learning_rate": 0.0011560659047325167, + "loss": 0.0728, + "step": 46616 + }, + { + "epoch": 0.40465794567755486, + "grad_norm": 0.107421875, + "learning_rate": 0.0011560362000512104, + "loss": 0.0737, + "step": 46617 + }, + { + "epoch": 0.40466662615775906, + "grad_norm": 0.3125, + "learning_rate": 0.0011560064953086238, + "loss": 0.1094, + "step": 46618 + }, + { + "epoch": 0.4046753066379632, + "grad_norm": 0.2080078125, + "learning_rate": 0.0011559767905047892, + "loss": 0.1191, + "step": 46619 + }, + { + "epoch": 0.4046839871181674, + "grad_norm": 0.318359375, + "learning_rate": 0.0011559470856397388, + "loss": 0.0918, + "step": 46620 + }, + { + "epoch": 0.4046926675983715, + "grad_norm": 0.376953125, + "learning_rate": 0.001155917380713506, + "loss": 0.1128, + "step": 46621 + }, + { + "epoch": 0.4047013480785757, + "grad_norm": 0.1376953125, + "learning_rate": 0.001155887675726123, + "loss": 0.1416, + "step": 46622 + }, + { + "epoch": 0.40471002855877986, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011558579706776218, + "loss": 0.0742, + "step": 46623 + }, + { + "epoch": 0.40471870903898405, + "grad_norm": 0.232421875, + "learning_rate": 0.001155828265568035, + "loss": 0.0869, + "step": 46624 + }, + { + "epoch": 0.4047273895191882, + "grad_norm": 0.3515625, + "learning_rate": 0.0011557985603973954, + "loss": 0.1357, + "step": 46625 + }, + { + "epoch": 0.4047360699993924, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011557688551657353, + "loss": 0.1001, + "step": 46626 + }, + { + "epoch": 0.4047447504795965, + "grad_norm": 0.796875, + "learning_rate": 0.001155739149873087, + "loss": 0.0801, + "step": 46627 + }, + { + "epoch": 0.4047534309598007, + "grad_norm": 0.412109375, + "learning_rate": 0.0011557094445194835, + "loss": 0.0918, + "step": 46628 + }, + { + "epoch": 0.40476211144000485, + "grad_norm": 0.26953125, + "learning_rate": 0.001155679739104957, + "loss": 0.1118, + "step": 46629 + }, + { + "epoch": 0.40477079192020904, + "grad_norm": 0.32421875, + "learning_rate": 0.0011556500336295402, + "loss": 0.0962, + "step": 46630 + }, + { + "epoch": 0.4047794724004132, + "grad_norm": 0.7578125, + "learning_rate": 0.0011556203280932648, + "loss": 0.1191, + "step": 46631 + }, + { + "epoch": 0.40478815288061737, + "grad_norm": 0.427734375, + "learning_rate": 0.0011555906224961644, + "loss": 0.0991, + "step": 46632 + }, + { + "epoch": 0.4047968333608215, + "grad_norm": 0.400390625, + "learning_rate": 0.0011555609168382704, + "loss": 0.0649, + "step": 46633 + }, + { + "epoch": 0.4048055138410257, + "grad_norm": 0.373046875, + "learning_rate": 0.001155531211119616, + "loss": 0.0898, + "step": 46634 + }, + { + "epoch": 0.40481419432122984, + "grad_norm": 0.26171875, + "learning_rate": 0.0011555015053402336, + "loss": 0.0869, + "step": 46635 + }, + { + "epoch": 0.40482287480143403, + "grad_norm": 0.357421875, + "learning_rate": 0.0011554717995001557, + "loss": 0.0801, + "step": 46636 + }, + { + "epoch": 0.40483155528163817, + "grad_norm": 0.328125, + "learning_rate": 0.0011554420935994143, + "loss": 0.0806, + "step": 46637 + }, + { + "epoch": 0.40484023576184236, + "grad_norm": 0.1416015625, + "learning_rate": 0.0011554123876380427, + "loss": 0.1143, + "step": 46638 + }, + { + "epoch": 0.4048489162420465, + "grad_norm": 0.146484375, + "learning_rate": 0.0011553826816160725, + "loss": 0.1094, + "step": 46639 + }, + { + "epoch": 0.4048575967222507, + "grad_norm": 0.12353515625, + "learning_rate": 0.001155352975533537, + "loss": 0.1104, + "step": 46640 + }, + { + "epoch": 0.4048662772024548, + "grad_norm": 0.232421875, + "learning_rate": 0.001155323269390468, + "loss": 0.0752, + "step": 46641 + }, + { + "epoch": 0.404874957682659, + "grad_norm": 0.279296875, + "learning_rate": 0.0011552935631868985, + "loss": 0.1064, + "step": 46642 + }, + { + "epoch": 0.40488363816286316, + "grad_norm": 0.474609375, + "learning_rate": 0.0011552638569228611, + "loss": 0.106, + "step": 46643 + }, + { + "epoch": 0.40489231864306735, + "grad_norm": 0.1591796875, + "learning_rate": 0.0011552341505983875, + "loss": 0.1167, + "step": 46644 + }, + { + "epoch": 0.4049009991232715, + "grad_norm": 0.1337890625, + "learning_rate": 0.0011552044442135105, + "loss": 0.1104, + "step": 46645 + }, + { + "epoch": 0.4049096796034757, + "grad_norm": 0.10791015625, + "learning_rate": 0.0011551747377682633, + "loss": 0.0791, + "step": 46646 + }, + { + "epoch": 0.4049183600836798, + "grad_norm": 0.4296875, + "learning_rate": 0.0011551450312626773, + "loss": 0.1855, + "step": 46647 + }, + { + "epoch": 0.404927040563884, + "grad_norm": 0.291015625, + "learning_rate": 0.0011551153246967861, + "loss": 0.1084, + "step": 46648 + }, + { + "epoch": 0.40493572104408815, + "grad_norm": 0.5390625, + "learning_rate": 0.0011550856180706215, + "loss": 0.1001, + "step": 46649 + }, + { + "epoch": 0.40494440152429234, + "grad_norm": 0.328125, + "learning_rate": 0.0011550559113842156, + "loss": 0.1494, + "step": 46650 + }, + { + "epoch": 0.4049530820044965, + "grad_norm": 0.78125, + "learning_rate": 0.001155026204637602, + "loss": 0.0996, + "step": 46651 + }, + { + "epoch": 0.40496176248470067, + "grad_norm": 0.126953125, + "learning_rate": 0.0011549964978308123, + "loss": 0.1318, + "step": 46652 + }, + { + "epoch": 0.4049704429649048, + "grad_norm": 0.076171875, + "learning_rate": 0.001154966790963879, + "loss": 0.1152, + "step": 46653 + }, + { + "epoch": 0.404979123445109, + "grad_norm": 0.322265625, + "learning_rate": 0.001154937084036835, + "loss": 0.1123, + "step": 46654 + }, + { + "epoch": 0.40498780392531314, + "grad_norm": 0.12060546875, + "learning_rate": 0.0011549073770497128, + "loss": 0.1104, + "step": 46655 + }, + { + "epoch": 0.40499648440551733, + "grad_norm": 0.55859375, + "learning_rate": 0.0011548776700025444, + "loss": 0.1021, + "step": 46656 + }, + { + "epoch": 0.40500516488572147, + "grad_norm": 0.3203125, + "learning_rate": 0.0011548479628953631, + "loss": 0.1152, + "step": 46657 + }, + { + "epoch": 0.40501384536592566, + "grad_norm": 0.421875, + "learning_rate": 0.0011548182557282005, + "loss": 0.0835, + "step": 46658 + }, + { + "epoch": 0.4050225258461298, + "grad_norm": 0.341796875, + "learning_rate": 0.0011547885485010895, + "loss": 0.1064, + "step": 46659 + }, + { + "epoch": 0.405031206326334, + "grad_norm": 0.546875, + "learning_rate": 0.001154758841214063, + "loss": 0.0889, + "step": 46660 + }, + { + "epoch": 0.40503988680653813, + "grad_norm": 0.4609375, + "learning_rate": 0.0011547291338671526, + "loss": 0.0918, + "step": 46661 + }, + { + "epoch": 0.4050485672867423, + "grad_norm": 0.294921875, + "learning_rate": 0.0011546994264603914, + "loss": 0.0811, + "step": 46662 + }, + { + "epoch": 0.40505724776694646, + "grad_norm": 0.2109375, + "learning_rate": 0.0011546697189938116, + "loss": 0.0791, + "step": 46663 + }, + { + "epoch": 0.40506592824715065, + "grad_norm": 0.2490234375, + "learning_rate": 0.0011546400114674459, + "loss": 0.0869, + "step": 46664 + }, + { + "epoch": 0.4050746087273548, + "grad_norm": 0.091796875, + "learning_rate": 0.0011546103038813267, + "loss": 0.1011, + "step": 46665 + }, + { + "epoch": 0.405083289207559, + "grad_norm": 0.11279296875, + "learning_rate": 0.0011545805962354864, + "loss": 0.1211, + "step": 46666 + }, + { + "epoch": 0.4050919696877631, + "grad_norm": 0.203125, + "learning_rate": 0.0011545508885299578, + "loss": 0.0967, + "step": 46667 + }, + { + "epoch": 0.4051006501679673, + "grad_norm": 0.28515625, + "learning_rate": 0.0011545211807647729, + "loss": 0.1035, + "step": 46668 + }, + { + "epoch": 0.40510933064817145, + "grad_norm": 0.65234375, + "learning_rate": 0.0011544914729399646, + "loss": 0.1348, + "step": 46669 + }, + { + "epoch": 0.40511801112837564, + "grad_norm": 0.54296875, + "learning_rate": 0.0011544617650555654, + "loss": 0.104, + "step": 46670 + }, + { + "epoch": 0.4051266916085798, + "grad_norm": 0.08544921875, + "learning_rate": 0.0011544320571116073, + "loss": 0.0713, + "step": 46671 + }, + { + "epoch": 0.405135372088784, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011544023491081232, + "loss": 0.0767, + "step": 46672 + }, + { + "epoch": 0.4051440525689881, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011543726410451456, + "loss": 0.0952, + "step": 46673 + }, + { + "epoch": 0.4051527330491923, + "grad_norm": 0.341796875, + "learning_rate": 0.001154342932922707, + "loss": 0.0938, + "step": 46674 + }, + { + "epoch": 0.40516141352939644, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011543132247408398, + "loss": 0.1377, + "step": 46675 + }, + { + "epoch": 0.40517009400960063, + "grad_norm": 0.30078125, + "learning_rate": 0.0011542835164995763, + "loss": 0.1357, + "step": 46676 + }, + { + "epoch": 0.40517877448980477, + "grad_norm": 0.265625, + "learning_rate": 0.0011542538081989492, + "loss": 0.1377, + "step": 46677 + }, + { + "epoch": 0.40518745497000896, + "grad_norm": 0.39453125, + "learning_rate": 0.001154224099838991, + "loss": 0.1069, + "step": 46678 + }, + { + "epoch": 0.4051961354502131, + "grad_norm": 0.2421875, + "learning_rate": 0.0011541943914197342, + "loss": 0.0732, + "step": 46679 + }, + { + "epoch": 0.4052048159304173, + "grad_norm": 0.55078125, + "learning_rate": 0.0011541646829412112, + "loss": 0.1128, + "step": 46680 + }, + { + "epoch": 0.40521349641062143, + "grad_norm": 0.2109375, + "learning_rate": 0.0011541349744034546, + "loss": 0.1074, + "step": 46681 + }, + { + "epoch": 0.4052221768908256, + "grad_norm": 0.150390625, + "learning_rate": 0.0011541052658064963, + "loss": 0.0918, + "step": 46682 + }, + { + "epoch": 0.40523085737102976, + "grad_norm": 0.224609375, + "learning_rate": 0.0011540755571503697, + "loss": 0.0923, + "step": 46683 + }, + { + "epoch": 0.40523953785123396, + "grad_norm": 0.09619140625, + "learning_rate": 0.001154045848435107, + "loss": 0.083, + "step": 46684 + }, + { + "epoch": 0.4052482183314381, + "grad_norm": 0.212890625, + "learning_rate": 0.0011540161396607403, + "loss": 0.0977, + "step": 46685 + }, + { + "epoch": 0.4052568988116423, + "grad_norm": 0.07080078125, + "learning_rate": 0.0011539864308273025, + "loss": 0.0674, + "step": 46686 + }, + { + "epoch": 0.4052655792918464, + "grad_norm": 0.2236328125, + "learning_rate": 0.001153956721934826, + "loss": 0.1211, + "step": 46687 + }, + { + "epoch": 0.4052742597720506, + "grad_norm": 0.2001953125, + "learning_rate": 0.0011539270129833433, + "loss": 0.0859, + "step": 46688 + }, + { + "epoch": 0.40528294025225475, + "grad_norm": 0.515625, + "learning_rate": 0.0011538973039728864, + "loss": 0.083, + "step": 46689 + }, + { + "epoch": 0.40529162073245895, + "grad_norm": 0.71875, + "learning_rate": 0.0011538675949034887, + "loss": 0.1104, + "step": 46690 + }, + { + "epoch": 0.4053003012126631, + "grad_norm": 0.392578125, + "learning_rate": 0.0011538378857751822, + "loss": 0.1436, + "step": 46691 + }, + { + "epoch": 0.4053089816928673, + "grad_norm": 0.123046875, + "learning_rate": 0.0011538081765879989, + "loss": 0.0972, + "step": 46692 + }, + { + "epoch": 0.4053176621730714, + "grad_norm": 0.203125, + "learning_rate": 0.001153778467341972, + "loss": 0.1348, + "step": 46693 + }, + { + "epoch": 0.4053263426532756, + "grad_norm": 0.484375, + "learning_rate": 0.001153748758037134, + "loss": 0.0894, + "step": 46694 + }, + { + "epoch": 0.40533502313347974, + "grad_norm": 0.50390625, + "learning_rate": 0.0011537190486735171, + "loss": 0.1084, + "step": 46695 + }, + { + "epoch": 0.40534370361368394, + "grad_norm": 0.10009765625, + "learning_rate": 0.001153689339251154, + "loss": 0.0996, + "step": 46696 + }, + { + "epoch": 0.4053523840938881, + "grad_norm": 0.1552734375, + "learning_rate": 0.0011536596297700766, + "loss": 0.1406, + "step": 46697 + }, + { + "epoch": 0.4053610645740922, + "grad_norm": 0.294921875, + "learning_rate": 0.001153629920230318, + "loss": 0.0898, + "step": 46698 + }, + { + "epoch": 0.4053697450542964, + "grad_norm": 0.484375, + "learning_rate": 0.0011536002106319109, + "loss": 0.1309, + "step": 46699 + }, + { + "epoch": 0.40537842553450054, + "grad_norm": 0.39453125, + "learning_rate": 0.001153570500974887, + "loss": 0.1299, + "step": 46700 + }, + { + "epoch": 0.40538710601470473, + "grad_norm": 0.134765625, + "learning_rate": 0.0011535407912592792, + "loss": 0.1338, + "step": 46701 + }, + { + "epoch": 0.40539578649490887, + "grad_norm": 0.298828125, + "learning_rate": 0.00115351108148512, + "loss": 0.1006, + "step": 46702 + }, + { + "epoch": 0.40540446697511306, + "grad_norm": 0.263671875, + "learning_rate": 0.0011534813716524421, + "loss": 0.1338, + "step": 46703 + }, + { + "epoch": 0.4054131474553172, + "grad_norm": 0.19921875, + "learning_rate": 0.0011534516617612775, + "loss": 0.1064, + "step": 46704 + }, + { + "epoch": 0.4054218279355214, + "grad_norm": 0.69921875, + "learning_rate": 0.0011534219518116594, + "loss": 0.1426, + "step": 46705 + }, + { + "epoch": 0.40543050841572553, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011533922418036193, + "loss": 0.0713, + "step": 46706 + }, + { + "epoch": 0.4054391888959297, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011533625317371907, + "loss": 0.1216, + "step": 46707 + }, + { + "epoch": 0.40544786937613386, + "grad_norm": 0.0869140625, + "learning_rate": 0.0011533328216124057, + "loss": 0.1064, + "step": 46708 + }, + { + "epoch": 0.40545654985633806, + "grad_norm": 0.6484375, + "learning_rate": 0.0011533031114292964, + "loss": 0.1104, + "step": 46709 + }, + { + "epoch": 0.4054652303365422, + "grad_norm": 0.0751953125, + "learning_rate": 0.0011532734011878956, + "loss": 0.064, + "step": 46710 + }, + { + "epoch": 0.4054739108167464, + "grad_norm": 0.1064453125, + "learning_rate": 0.0011532436908882361, + "loss": 0.0605, + "step": 46711 + }, + { + "epoch": 0.4054825912969505, + "grad_norm": 0.09228515625, + "learning_rate": 0.0011532139805303498, + "loss": 0.0947, + "step": 46712 + }, + { + "epoch": 0.4054912717771547, + "grad_norm": 0.3671875, + "learning_rate": 0.0011531842701142697, + "loss": 0.1035, + "step": 46713 + }, + { + "epoch": 0.40549995225735885, + "grad_norm": 0.2099609375, + "learning_rate": 0.001153154559640028, + "loss": 0.0908, + "step": 46714 + }, + { + "epoch": 0.40550863273756305, + "grad_norm": 0.189453125, + "learning_rate": 0.0011531248491076574, + "loss": 0.0742, + "step": 46715 + }, + { + "epoch": 0.4055173132177672, + "grad_norm": 0.10546875, + "learning_rate": 0.0011530951385171902, + "loss": 0.1143, + "step": 46716 + }, + { + "epoch": 0.4055259936979714, + "grad_norm": 0.1611328125, + "learning_rate": 0.001153065427868659, + "loss": 0.0933, + "step": 46717 + }, + { + "epoch": 0.4055346741781755, + "grad_norm": 0.494140625, + "learning_rate": 0.001153035717162096, + "loss": 0.0859, + "step": 46718 + }, + { + "epoch": 0.4055433546583797, + "grad_norm": 0.6484375, + "learning_rate": 0.001153006006397534, + "loss": 0.1123, + "step": 46719 + }, + { + "epoch": 0.40555203513858384, + "grad_norm": 0.298828125, + "learning_rate": 0.0011529762955750057, + "loss": 0.085, + "step": 46720 + }, + { + "epoch": 0.40556071561878804, + "grad_norm": 0.2451171875, + "learning_rate": 0.001152946584694543, + "loss": 0.083, + "step": 46721 + }, + { + "epoch": 0.4055693960989922, + "grad_norm": 0.287109375, + "learning_rate": 0.0011529168737561789, + "loss": 0.0952, + "step": 46722 + }, + { + "epoch": 0.40557807657919637, + "grad_norm": 0.107421875, + "learning_rate": 0.0011528871627599453, + "loss": 0.0874, + "step": 46723 + }, + { + "epoch": 0.4055867570594005, + "grad_norm": 0.431640625, + "learning_rate": 0.0011528574517058756, + "loss": 0.1104, + "step": 46724 + }, + { + "epoch": 0.4055954375396047, + "grad_norm": 0.103515625, + "learning_rate": 0.0011528277405940019, + "loss": 0.0596, + "step": 46725 + }, + { + "epoch": 0.40560411801980883, + "grad_norm": 0.2109375, + "learning_rate": 0.001152798029424356, + "loss": 0.1211, + "step": 46726 + }, + { + "epoch": 0.405612798500013, + "grad_norm": 0.53125, + "learning_rate": 0.0011527683181969718, + "loss": 0.1143, + "step": 46727 + }, + { + "epoch": 0.40562147898021716, + "grad_norm": 0.28515625, + "learning_rate": 0.0011527386069118799, + "loss": 0.0898, + "step": 46728 + }, + { + "epoch": 0.40563015946042136, + "grad_norm": 0.138671875, + "learning_rate": 0.0011527088955691144, + "loss": 0.1016, + "step": 46729 + }, + { + "epoch": 0.4056388399406255, + "grad_norm": 0.287109375, + "learning_rate": 0.0011526791841687074, + "loss": 0.0747, + "step": 46730 + }, + { + "epoch": 0.4056475204208297, + "grad_norm": 0.3359375, + "learning_rate": 0.001152649472710691, + "loss": 0.1099, + "step": 46731 + }, + { + "epoch": 0.4056562009010338, + "grad_norm": 0.498046875, + "learning_rate": 0.0011526197611950977, + "loss": 0.0947, + "step": 46732 + }, + { + "epoch": 0.405664881381238, + "grad_norm": 0.310546875, + "learning_rate": 0.0011525900496219606, + "loss": 0.0879, + "step": 46733 + }, + { + "epoch": 0.40567356186144216, + "grad_norm": 0.29296875, + "learning_rate": 0.001152560337991312, + "loss": 0.1299, + "step": 46734 + }, + { + "epoch": 0.40568224234164635, + "grad_norm": 0.54296875, + "learning_rate": 0.001152530626303184, + "loss": 0.083, + "step": 46735 + }, + { + "epoch": 0.4056909228218505, + "grad_norm": 0.423828125, + "learning_rate": 0.0011525009145576088, + "loss": 0.1064, + "step": 46736 + }, + { + "epoch": 0.4056996033020547, + "grad_norm": 0.255859375, + "learning_rate": 0.0011524712027546201, + "loss": 0.1426, + "step": 46737 + }, + { + "epoch": 0.4057082837822588, + "grad_norm": 0.18359375, + "learning_rate": 0.0011524414908942492, + "loss": 0.0859, + "step": 46738 + }, + { + "epoch": 0.405716964262463, + "grad_norm": 0.53125, + "learning_rate": 0.0011524117789765294, + "loss": 0.1182, + "step": 46739 + }, + { + "epoch": 0.40572564474266715, + "grad_norm": 0.142578125, + "learning_rate": 0.0011523820670014927, + "loss": 0.1216, + "step": 46740 + }, + { + "epoch": 0.40573432522287134, + "grad_norm": 0.154296875, + "learning_rate": 0.0011523523549691714, + "loss": 0.0869, + "step": 46741 + }, + { + "epoch": 0.4057430057030755, + "grad_norm": 0.1318359375, + "learning_rate": 0.001152322642879599, + "loss": 0.0864, + "step": 46742 + }, + { + "epoch": 0.40575168618327967, + "grad_norm": 0.1865234375, + "learning_rate": 0.001152292930732807, + "loss": 0.126, + "step": 46743 + }, + { + "epoch": 0.4057603666634838, + "grad_norm": 2.0, + "learning_rate": 0.0011522632185288287, + "loss": 0.2207, + "step": 46744 + }, + { + "epoch": 0.405769047143688, + "grad_norm": 0.5078125, + "learning_rate": 0.0011522335062676954, + "loss": 0.0815, + "step": 46745 + }, + { + "epoch": 0.40577772762389214, + "grad_norm": 0.2333984375, + "learning_rate": 0.0011522037939494408, + "loss": 0.1152, + "step": 46746 + }, + { + "epoch": 0.40578640810409633, + "grad_norm": 0.373046875, + "learning_rate": 0.0011521740815740968, + "loss": 0.0684, + "step": 46747 + }, + { + "epoch": 0.40579508858430047, + "grad_norm": 0.162109375, + "learning_rate": 0.0011521443691416963, + "loss": 0.0879, + "step": 46748 + }, + { + "epoch": 0.40580376906450466, + "grad_norm": 0.2001953125, + "learning_rate": 0.0011521146566522708, + "loss": 0.1328, + "step": 46749 + }, + { + "epoch": 0.4058124495447088, + "grad_norm": 0.17578125, + "learning_rate": 0.0011520849441058539, + "loss": 0.1426, + "step": 46750 + }, + { + "epoch": 0.405821130024913, + "grad_norm": 0.3203125, + "learning_rate": 0.0011520552315024777, + "loss": 0.1182, + "step": 46751 + }, + { + "epoch": 0.40582981050511713, + "grad_norm": 0.6640625, + "learning_rate": 0.0011520255188421748, + "loss": 0.1523, + "step": 46752 + }, + { + "epoch": 0.4058384909853213, + "grad_norm": 0.12451171875, + "learning_rate": 0.0011519958061249774, + "loss": 0.0903, + "step": 46753 + }, + { + "epoch": 0.40584717146552546, + "grad_norm": 0.053466796875, + "learning_rate": 0.001151966093350918, + "loss": 0.0654, + "step": 46754 + }, + { + "epoch": 0.40585585194572965, + "grad_norm": 0.671875, + "learning_rate": 0.0011519363805200297, + "loss": 0.1475, + "step": 46755 + }, + { + "epoch": 0.4058645324259338, + "grad_norm": 0.296875, + "learning_rate": 0.0011519066676323442, + "loss": 0.167, + "step": 46756 + }, + { + "epoch": 0.405873212906138, + "grad_norm": 0.53515625, + "learning_rate": 0.0011518769546878945, + "loss": 0.0977, + "step": 46757 + }, + { + "epoch": 0.4058818933863421, + "grad_norm": 0.15625, + "learning_rate": 0.0011518472416867126, + "loss": 0.1445, + "step": 46758 + }, + { + "epoch": 0.4058905738665463, + "grad_norm": 0.32421875, + "learning_rate": 0.0011518175286288318, + "loss": 0.084, + "step": 46759 + }, + { + "epoch": 0.40589925434675045, + "grad_norm": 0.166015625, + "learning_rate": 0.001151787815514284, + "loss": 0.1172, + "step": 46760 + }, + { + "epoch": 0.40590793482695464, + "grad_norm": 0.11962890625, + "learning_rate": 0.0011517581023431018, + "loss": 0.123, + "step": 46761 + }, + { + "epoch": 0.4059166153071588, + "grad_norm": 0.0869140625, + "learning_rate": 0.0011517283891153175, + "loss": 0.0938, + "step": 46762 + }, + { + "epoch": 0.40592529578736297, + "grad_norm": 0.09375, + "learning_rate": 0.0011516986758309642, + "loss": 0.0723, + "step": 46763 + }, + { + "epoch": 0.4059339762675671, + "grad_norm": 0.2333984375, + "learning_rate": 0.0011516689624900738, + "loss": 0.1416, + "step": 46764 + }, + { + "epoch": 0.4059426567477713, + "grad_norm": 0.419921875, + "learning_rate": 0.0011516392490926793, + "loss": 0.1465, + "step": 46765 + }, + { + "epoch": 0.40595133722797544, + "grad_norm": 0.396484375, + "learning_rate": 0.0011516095356388122, + "loss": 0.0859, + "step": 46766 + }, + { + "epoch": 0.40596001770817963, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011515798221285063, + "loss": 0.0996, + "step": 46767 + }, + { + "epoch": 0.40596869818838377, + "grad_norm": 0.06689453125, + "learning_rate": 0.0011515501085617933, + "loss": 0.0737, + "step": 46768 + }, + { + "epoch": 0.40597737866858796, + "grad_norm": 0.2236328125, + "learning_rate": 0.001151520394938706, + "loss": 0.1113, + "step": 46769 + }, + { + "epoch": 0.4059860591487921, + "grad_norm": 0.68359375, + "learning_rate": 0.0011514906812592767, + "loss": 0.123, + "step": 46770 + }, + { + "epoch": 0.4059947396289963, + "grad_norm": 0.373046875, + "learning_rate": 0.0011514609675235374, + "loss": 0.1138, + "step": 46771 + }, + { + "epoch": 0.40600342010920043, + "grad_norm": 0.3984375, + "learning_rate": 0.001151431253731522, + "loss": 0.0898, + "step": 46772 + }, + { + "epoch": 0.4060121005894046, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011514015398832618, + "loss": 0.0903, + "step": 46773 + }, + { + "epoch": 0.40602078106960876, + "grad_norm": 0.294921875, + "learning_rate": 0.0011513718259787897, + "loss": 0.0928, + "step": 46774 + }, + { + "epoch": 0.40602946154981295, + "grad_norm": 0.2412109375, + "learning_rate": 0.001151342112018138, + "loss": 0.0879, + "step": 46775 + }, + { + "epoch": 0.4060381420300171, + "grad_norm": 0.111328125, + "learning_rate": 0.0011513123980013396, + "loss": 0.0977, + "step": 46776 + }, + { + "epoch": 0.4060468225102213, + "grad_norm": 0.494140625, + "learning_rate": 0.0011512826839284265, + "loss": 0.0977, + "step": 46777 + }, + { + "epoch": 0.4060555029904254, + "grad_norm": 0.08740234375, + "learning_rate": 0.0011512529697994315, + "loss": 0.1055, + "step": 46778 + }, + { + "epoch": 0.4060641834706296, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011512232556143867, + "loss": 0.1123, + "step": 46779 + }, + { + "epoch": 0.40607286395083375, + "grad_norm": 0.578125, + "learning_rate": 0.0011511935413733255, + "loss": 0.1035, + "step": 46780 + }, + { + "epoch": 0.40608154443103794, + "grad_norm": 0.2041015625, + "learning_rate": 0.0011511638270762796, + "loss": 0.1484, + "step": 46781 + }, + { + "epoch": 0.4060902249112421, + "grad_norm": 0.328125, + "learning_rate": 0.0011511341127232817, + "loss": 0.1191, + "step": 46782 + }, + { + "epoch": 0.4060989053914463, + "grad_norm": 0.26171875, + "learning_rate": 0.0011511043983143644, + "loss": 0.1348, + "step": 46783 + }, + { + "epoch": 0.4061075858716504, + "grad_norm": 0.181640625, + "learning_rate": 0.0011510746838495596, + "loss": 0.0791, + "step": 46784 + }, + { + "epoch": 0.4061162663518546, + "grad_norm": 0.70703125, + "learning_rate": 0.001151044969328901, + "loss": 0.0898, + "step": 46785 + }, + { + "epoch": 0.40612494683205874, + "grad_norm": 0.1728515625, + "learning_rate": 0.00115101525475242, + "loss": 0.0723, + "step": 46786 + }, + { + "epoch": 0.40613362731226293, + "grad_norm": 0.36328125, + "learning_rate": 0.0011509855401201495, + "loss": 0.0718, + "step": 46787 + }, + { + "epoch": 0.40614230779246707, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011509558254321217, + "loss": 0.0679, + "step": 46788 + }, + { + "epoch": 0.40615098827267127, + "grad_norm": 0.0654296875, + "learning_rate": 0.0011509261106883697, + "loss": 0.0752, + "step": 46789 + }, + { + "epoch": 0.4061596687528754, + "grad_norm": 0.09326171875, + "learning_rate": 0.0011508963958889258, + "loss": 0.0908, + "step": 46790 + }, + { + "epoch": 0.4061683492330796, + "grad_norm": 0.3359375, + "learning_rate": 0.0011508666810338222, + "loss": 0.1875, + "step": 46791 + }, + { + "epoch": 0.40617702971328373, + "grad_norm": 0.2265625, + "learning_rate": 0.0011508369661230916, + "loss": 0.1162, + "step": 46792 + }, + { + "epoch": 0.4061857101934879, + "grad_norm": 0.75, + "learning_rate": 0.0011508072511567664, + "loss": 0.0918, + "step": 46793 + }, + { + "epoch": 0.40619439067369206, + "grad_norm": 0.404296875, + "learning_rate": 0.0011507775361348793, + "loss": 0.0884, + "step": 46794 + }, + { + "epoch": 0.40620307115389626, + "grad_norm": 0.162109375, + "learning_rate": 0.0011507478210574626, + "loss": 0.1279, + "step": 46795 + }, + { + "epoch": 0.4062117516341004, + "grad_norm": 0.12255859375, + "learning_rate": 0.0011507181059245487, + "loss": 0.1128, + "step": 46796 + }, + { + "epoch": 0.4062204321143046, + "grad_norm": 0.208984375, + "learning_rate": 0.0011506883907361702, + "loss": 0.1187, + "step": 46797 + }, + { + "epoch": 0.4062291125945087, + "grad_norm": 0.34765625, + "learning_rate": 0.0011506586754923597, + "loss": 0.0923, + "step": 46798 + }, + { + "epoch": 0.4062377930747129, + "grad_norm": 0.25, + "learning_rate": 0.00115062896019315, + "loss": 0.1113, + "step": 46799 + }, + { + "epoch": 0.40624647355491705, + "grad_norm": 0.26953125, + "learning_rate": 0.001150599244838573, + "loss": 0.1367, + "step": 46800 + }, + { + "epoch": 0.40625515403512125, + "grad_norm": 0.376953125, + "learning_rate": 0.0011505695294286608, + "loss": 0.1367, + "step": 46801 + }, + { + "epoch": 0.4062638345153254, + "grad_norm": 0.2734375, + "learning_rate": 0.0011505398139634473, + "loss": 0.1045, + "step": 46802 + }, + { + "epoch": 0.4062725149955296, + "grad_norm": 0.279296875, + "learning_rate": 0.001150510098442964, + "loss": 0.0767, + "step": 46803 + }, + { + "epoch": 0.4062811954757337, + "grad_norm": 0.361328125, + "learning_rate": 0.0011504803828672437, + "loss": 0.0737, + "step": 46804 + }, + { + "epoch": 0.4062898759559379, + "grad_norm": 0.8359375, + "learning_rate": 0.0011504506672363186, + "loss": 0.1387, + "step": 46805 + }, + { + "epoch": 0.40629855643614204, + "grad_norm": 0.6015625, + "learning_rate": 0.0011504209515502217, + "loss": 0.0791, + "step": 46806 + }, + { + "epoch": 0.40630723691634624, + "grad_norm": 0.08203125, + "learning_rate": 0.0011503912358089847, + "loss": 0.0889, + "step": 46807 + }, + { + "epoch": 0.4063159173965504, + "grad_norm": 0.138671875, + "learning_rate": 0.0011503615200126412, + "loss": 0.0884, + "step": 46808 + }, + { + "epoch": 0.40632459787675457, + "grad_norm": 0.54296875, + "learning_rate": 0.001150331804161223, + "loss": 0.0908, + "step": 46809 + }, + { + "epoch": 0.4063332783569587, + "grad_norm": 0.0830078125, + "learning_rate": 0.0011503020882547622, + "loss": 0.0859, + "step": 46810 + }, + { + "epoch": 0.4063419588371629, + "grad_norm": 0.294921875, + "learning_rate": 0.0011502723722932923, + "loss": 0.1328, + "step": 46811 + }, + { + "epoch": 0.40635063931736704, + "grad_norm": 0.70703125, + "learning_rate": 0.001150242656276845, + "loss": 0.209, + "step": 46812 + }, + { + "epoch": 0.40635931979757123, + "grad_norm": 0.11083984375, + "learning_rate": 0.0011502129402054533, + "loss": 0.1035, + "step": 46813 + }, + { + "epoch": 0.40636800027777537, + "grad_norm": 0.345703125, + "learning_rate": 0.0011501832240791493, + "loss": 0.0811, + "step": 46814 + }, + { + "epoch": 0.40637668075797956, + "grad_norm": 0.228515625, + "learning_rate": 0.0011501535078979658, + "loss": 0.0859, + "step": 46815 + }, + { + "epoch": 0.4063853612381837, + "grad_norm": 0.78125, + "learning_rate": 0.001150123791661935, + "loss": 0.0806, + "step": 46816 + }, + { + "epoch": 0.4063940417183879, + "grad_norm": 0.62109375, + "learning_rate": 0.0011500940753710897, + "loss": 0.1035, + "step": 46817 + }, + { + "epoch": 0.406402722198592, + "grad_norm": 0.376953125, + "learning_rate": 0.0011500643590254623, + "loss": 0.0796, + "step": 46818 + }, + { + "epoch": 0.4064114026787962, + "grad_norm": 0.173828125, + "learning_rate": 0.0011500346426250852, + "loss": 0.1309, + "step": 46819 + }, + { + "epoch": 0.40642008315900036, + "grad_norm": 0.30859375, + "learning_rate": 0.0011500049261699912, + "loss": 0.1553, + "step": 46820 + }, + { + "epoch": 0.4064287636392045, + "grad_norm": 0.310546875, + "learning_rate": 0.0011499752096602123, + "loss": 0.1387, + "step": 46821 + }, + { + "epoch": 0.4064374441194087, + "grad_norm": 0.1875, + "learning_rate": 0.0011499454930957815, + "loss": 0.0903, + "step": 46822 + }, + { + "epoch": 0.4064461245996128, + "grad_norm": 0.19140625, + "learning_rate": 0.0011499157764767308, + "loss": 0.0864, + "step": 46823 + }, + { + "epoch": 0.406454805079817, + "grad_norm": 0.08154296875, + "learning_rate": 0.001149886059803093, + "loss": 0.0947, + "step": 46824 + }, + { + "epoch": 0.40646348556002115, + "grad_norm": 0.404296875, + "learning_rate": 0.0011498563430749006, + "loss": 0.1025, + "step": 46825 + }, + { + "epoch": 0.40647216604022535, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011498266262921861, + "loss": 0.1123, + "step": 46826 + }, + { + "epoch": 0.4064808465204295, + "grad_norm": 0.142578125, + "learning_rate": 0.0011497969094549816, + "loss": 0.0698, + "step": 46827 + }, + { + "epoch": 0.4064895270006337, + "grad_norm": 0.07275390625, + "learning_rate": 0.0011497671925633204, + "loss": 0.0908, + "step": 46828 + }, + { + "epoch": 0.4064982074808378, + "grad_norm": 0.42578125, + "learning_rate": 0.0011497374756172345, + "loss": 0.0645, + "step": 46829 + }, + { + "epoch": 0.406506887961042, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011497077586167567, + "loss": 0.1553, + "step": 46830 + }, + { + "epoch": 0.40651556844124614, + "grad_norm": 1.0703125, + "learning_rate": 0.0011496780415619185, + "loss": 0.3672, + "step": 46831 + }, + { + "epoch": 0.40652424892145034, + "grad_norm": 0.177734375, + "learning_rate": 0.0011496483244527536, + "loss": 0.0986, + "step": 46832 + }, + { + "epoch": 0.4065329294016545, + "grad_norm": 2.265625, + "learning_rate": 0.001149618607289294, + "loss": 0.2383, + "step": 46833 + }, + { + "epoch": 0.40654160988185867, + "grad_norm": 0.375, + "learning_rate": 0.0011495888900715724, + "loss": 0.1572, + "step": 46834 + }, + { + "epoch": 0.4065502903620628, + "grad_norm": 0.216796875, + "learning_rate": 0.0011495591727996208, + "loss": 0.1035, + "step": 46835 + }, + { + "epoch": 0.406558970842267, + "grad_norm": 0.11279296875, + "learning_rate": 0.0011495294554734721, + "loss": 0.0908, + "step": 46836 + }, + { + "epoch": 0.40656765132247114, + "grad_norm": 2.265625, + "learning_rate": 0.0011494997380931589, + "loss": 0.2051, + "step": 46837 + }, + { + "epoch": 0.40657633180267533, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011494700206587133, + "loss": 0.0908, + "step": 46838 + }, + { + "epoch": 0.40658501228287947, + "grad_norm": 0.287109375, + "learning_rate": 0.0011494403031701683, + "loss": 0.1191, + "step": 46839 + }, + { + "epoch": 0.40659369276308366, + "grad_norm": 0.26171875, + "learning_rate": 0.0011494105856275558, + "loss": 0.125, + "step": 46840 + }, + { + "epoch": 0.4066023732432878, + "grad_norm": 0.7109375, + "learning_rate": 0.001149380868030909, + "loss": 0.1045, + "step": 46841 + }, + { + "epoch": 0.406611053723492, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011493511503802597, + "loss": 0.0562, + "step": 46842 + }, + { + "epoch": 0.4066197342036961, + "grad_norm": 0.1064453125, + "learning_rate": 0.0011493214326756412, + "loss": 0.0806, + "step": 46843 + }, + { + "epoch": 0.4066284146839003, + "grad_norm": 0.353515625, + "learning_rate": 0.001149291714917085, + "loss": 0.0957, + "step": 46844 + }, + { + "epoch": 0.40663709516410446, + "grad_norm": 0.3671875, + "learning_rate": 0.0011492619971046243, + "loss": 0.106, + "step": 46845 + }, + { + "epoch": 0.40664577564430865, + "grad_norm": 0.85546875, + "learning_rate": 0.0011492322792382914, + "loss": 0.0996, + "step": 46846 + }, + { + "epoch": 0.4066544561245128, + "grad_norm": 0.326171875, + "learning_rate": 0.0011492025613181187, + "loss": 0.0732, + "step": 46847 + }, + { + "epoch": 0.406663136604717, + "grad_norm": 0.07373046875, + "learning_rate": 0.001149172843344139, + "loss": 0.0908, + "step": 46848 + }, + { + "epoch": 0.4066718170849211, + "grad_norm": 0.38671875, + "learning_rate": 0.0011491431253163847, + "loss": 0.1016, + "step": 46849 + }, + { + "epoch": 0.4066804975651253, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011491134072348884, + "loss": 0.0669, + "step": 46850 + }, + { + "epoch": 0.40668917804532945, + "grad_norm": 0.1533203125, + "learning_rate": 0.001149083689099682, + "loss": 0.1299, + "step": 46851 + }, + { + "epoch": 0.40669785852553364, + "grad_norm": 0.150390625, + "learning_rate": 0.0011490539709107985, + "loss": 0.1094, + "step": 46852 + }, + { + "epoch": 0.4067065390057378, + "grad_norm": 0.2041015625, + "learning_rate": 0.0011490242526682702, + "loss": 0.1182, + "step": 46853 + }, + { + "epoch": 0.40671521948594197, + "grad_norm": 0.427734375, + "learning_rate": 0.00114899453437213, + "loss": 0.1094, + "step": 46854 + }, + { + "epoch": 0.4067238999661461, + "grad_norm": 0.1806640625, + "learning_rate": 0.00114896481602241, + "loss": 0.1016, + "step": 46855 + }, + { + "epoch": 0.4067325804463503, + "grad_norm": 0.10791015625, + "learning_rate": 0.0011489350976191426, + "loss": 0.1045, + "step": 46856 + }, + { + "epoch": 0.40674126092655444, + "grad_norm": 0.08740234375, + "learning_rate": 0.0011489053791623608, + "loss": 0.0957, + "step": 46857 + }, + { + "epoch": 0.40674994140675863, + "grad_norm": 0.2158203125, + "learning_rate": 0.0011488756606520968, + "loss": 0.124, + "step": 46858 + }, + { + "epoch": 0.40675862188696277, + "grad_norm": 0.205078125, + "learning_rate": 0.001148845942088383, + "loss": 0.0996, + "step": 46859 + }, + { + "epoch": 0.40676730236716696, + "grad_norm": 0.271484375, + "learning_rate": 0.0011488162234712521, + "loss": 0.0938, + "step": 46860 + }, + { + "epoch": 0.4067759828473711, + "grad_norm": 0.09912109375, + "learning_rate": 0.0011487865048007365, + "loss": 0.1016, + "step": 46861 + }, + { + "epoch": 0.4067846633275753, + "grad_norm": 0.322265625, + "learning_rate": 0.0011487567860768687, + "loss": 0.0708, + "step": 46862 + }, + { + "epoch": 0.40679334380777943, + "grad_norm": 0.171875, + "learning_rate": 0.001148727067299681, + "loss": 0.1719, + "step": 46863 + }, + { + "epoch": 0.4068020242879836, + "grad_norm": 0.0859375, + "learning_rate": 0.0011486973484692064, + "loss": 0.0713, + "step": 46864 + }, + { + "epoch": 0.40681070476818776, + "grad_norm": 0.369140625, + "learning_rate": 0.001148667629585477, + "loss": 0.1289, + "step": 46865 + }, + { + "epoch": 0.40681938524839195, + "grad_norm": 0.4453125, + "learning_rate": 0.0011486379106485252, + "loss": 0.1738, + "step": 46866 + }, + { + "epoch": 0.4068280657285961, + "grad_norm": 0.1708984375, + "learning_rate": 0.001148608191658384, + "loss": 0.0942, + "step": 46867 + }, + { + "epoch": 0.4068367462088003, + "grad_norm": 0.375, + "learning_rate": 0.0011485784726150857, + "loss": 0.0918, + "step": 46868 + }, + { + "epoch": 0.4068454266890044, + "grad_norm": 0.37890625, + "learning_rate": 0.0011485487535186625, + "loss": 0.0957, + "step": 46869 + }, + { + "epoch": 0.4068541071692086, + "grad_norm": 0.37109375, + "learning_rate": 0.001148519034369147, + "loss": 0.1162, + "step": 46870 + }, + { + "epoch": 0.40686278764941275, + "grad_norm": 0.26953125, + "learning_rate": 0.001148489315166572, + "loss": 0.0869, + "step": 46871 + }, + { + "epoch": 0.40687146812961694, + "grad_norm": 0.1962890625, + "learning_rate": 0.0011484595959109698, + "loss": 0.0884, + "step": 46872 + }, + { + "epoch": 0.4068801486098211, + "grad_norm": 0.6484375, + "learning_rate": 0.001148429876602373, + "loss": 0.0889, + "step": 46873 + }, + { + "epoch": 0.4068888290900253, + "grad_norm": 0.10791015625, + "learning_rate": 0.0011484001572408138, + "loss": 0.127, + "step": 46874 + }, + { + "epoch": 0.4068975095702294, + "grad_norm": 0.36328125, + "learning_rate": 0.0011483704378263248, + "loss": 0.0889, + "step": 46875 + }, + { + "epoch": 0.4069061900504336, + "grad_norm": 0.19921875, + "learning_rate": 0.001148340718358939, + "loss": 0.1777, + "step": 46876 + }, + { + "epoch": 0.40691487053063774, + "grad_norm": 0.46484375, + "learning_rate": 0.0011483109988386885, + "loss": 0.1143, + "step": 46877 + }, + { + "epoch": 0.40692355101084193, + "grad_norm": 0.10791015625, + "learning_rate": 0.0011482812792656056, + "loss": 0.0986, + "step": 46878 + }, + { + "epoch": 0.40693223149104607, + "grad_norm": 0.27734375, + "learning_rate": 0.0011482515596397229, + "loss": 0.1621, + "step": 46879 + }, + { + "epoch": 0.40694091197125026, + "grad_norm": 0.22265625, + "learning_rate": 0.0011482218399610734, + "loss": 0.082, + "step": 46880 + }, + { + "epoch": 0.4069495924514544, + "grad_norm": 0.255859375, + "learning_rate": 0.001148192120229689, + "loss": 0.0918, + "step": 46881 + }, + { + "epoch": 0.4069582729316586, + "grad_norm": 0.099609375, + "learning_rate": 0.0011481624004456026, + "loss": 0.084, + "step": 46882 + }, + { + "epoch": 0.40696695341186273, + "grad_norm": 0.298828125, + "learning_rate": 0.0011481326806088463, + "loss": 0.126, + "step": 46883 + }, + { + "epoch": 0.4069756338920669, + "grad_norm": 0.2734375, + "learning_rate": 0.0011481029607194526, + "loss": 0.1084, + "step": 46884 + }, + { + "epoch": 0.40698431437227106, + "grad_norm": 0.515625, + "learning_rate": 0.0011480732407774547, + "loss": 0.0967, + "step": 46885 + }, + { + "epoch": 0.40699299485247525, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011480435207828845, + "loss": 0.0933, + "step": 46886 + }, + { + "epoch": 0.4070016753326794, + "grad_norm": 0.8359375, + "learning_rate": 0.0011480138007357746, + "loss": 0.1074, + "step": 46887 + }, + { + "epoch": 0.4070103558128836, + "grad_norm": 0.46875, + "learning_rate": 0.0011479840806361574, + "loss": 0.085, + "step": 46888 + }, + { + "epoch": 0.4070190362930877, + "grad_norm": 0.474609375, + "learning_rate": 0.0011479543604840658, + "loss": 0.0884, + "step": 46889 + }, + { + "epoch": 0.4070277167732919, + "grad_norm": 0.55078125, + "learning_rate": 0.001147924640279532, + "loss": 0.0859, + "step": 46890 + }, + { + "epoch": 0.40703639725349605, + "grad_norm": 0.12890625, + "learning_rate": 0.0011478949200225883, + "loss": 0.0645, + "step": 46891 + }, + { + "epoch": 0.40704507773370024, + "grad_norm": 0.158203125, + "learning_rate": 0.0011478651997132675, + "loss": 0.1021, + "step": 46892 + }, + { + "epoch": 0.4070537582139044, + "grad_norm": 0.11279296875, + "learning_rate": 0.001147835479351602, + "loss": 0.1396, + "step": 46893 + }, + { + "epoch": 0.4070624386941086, + "grad_norm": 0.5, + "learning_rate": 0.0011478057589376246, + "loss": 0.0977, + "step": 46894 + }, + { + "epoch": 0.4070711191743127, + "grad_norm": 0.0732421875, + "learning_rate": 0.0011477760384713672, + "loss": 0.0698, + "step": 46895 + }, + { + "epoch": 0.4070797996545169, + "grad_norm": 0.828125, + "learning_rate": 0.0011477463179528627, + "loss": 0.0728, + "step": 46896 + }, + { + "epoch": 0.40708848013472104, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011477165973821438, + "loss": 0.1104, + "step": 46897 + }, + { + "epoch": 0.40709716061492524, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011476868767592427, + "loss": 0.082, + "step": 46898 + }, + { + "epoch": 0.4071058410951294, + "grad_norm": 0.08935546875, + "learning_rate": 0.0011476571560841916, + "loss": 0.0791, + "step": 46899 + }, + { + "epoch": 0.40711452157533357, + "grad_norm": 0.267578125, + "learning_rate": 0.0011476274353570235, + "loss": 0.0732, + "step": 46900 + }, + { + "epoch": 0.4071232020555377, + "grad_norm": 0.12060546875, + "learning_rate": 0.0011475977145777713, + "loss": 0.1113, + "step": 46901 + }, + { + "epoch": 0.4071318825357419, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011475679937464664, + "loss": 0.0938, + "step": 46902 + }, + { + "epoch": 0.40714056301594603, + "grad_norm": 0.55078125, + "learning_rate": 0.0011475382728631416, + "loss": 0.0889, + "step": 46903 + }, + { + "epoch": 0.4071492434961502, + "grad_norm": 0.09033203125, + "learning_rate": 0.00114750855192783, + "loss": 0.0732, + "step": 46904 + }, + { + "epoch": 0.40715792397635436, + "grad_norm": 0.162109375, + "learning_rate": 0.001147478830940564, + "loss": 0.0986, + "step": 46905 + }, + { + "epoch": 0.40716660445655856, + "grad_norm": 0.35546875, + "learning_rate": 0.0011474491099013756, + "loss": 0.127, + "step": 46906 + }, + { + "epoch": 0.4071752849367627, + "grad_norm": 0.482421875, + "learning_rate": 0.0011474193888102975, + "loss": 0.083, + "step": 46907 + }, + { + "epoch": 0.4071839654169669, + "grad_norm": 0.61328125, + "learning_rate": 0.0011473896676673625, + "loss": 0.0825, + "step": 46908 + }, + { + "epoch": 0.407192645897171, + "grad_norm": 0.578125, + "learning_rate": 0.0011473599464726025, + "loss": 0.0605, + "step": 46909 + }, + { + "epoch": 0.4072013263773752, + "grad_norm": 0.625, + "learning_rate": 0.0011473302252260507, + "loss": 0.0801, + "step": 46910 + }, + { + "epoch": 0.40721000685757935, + "grad_norm": 0.64453125, + "learning_rate": 0.0011473005039277393, + "loss": 0.1221, + "step": 46911 + }, + { + "epoch": 0.40721868733778355, + "grad_norm": 0.2099609375, + "learning_rate": 0.0011472707825777007, + "loss": 0.0718, + "step": 46912 + }, + { + "epoch": 0.4072273678179877, + "grad_norm": 0.185546875, + "learning_rate": 0.0011472410611759673, + "loss": 0.1201, + "step": 46913 + }, + { + "epoch": 0.4072360482981919, + "grad_norm": 0.201171875, + "learning_rate": 0.0011472113397225719, + "loss": 0.1084, + "step": 46914 + }, + { + "epoch": 0.407244728778396, + "grad_norm": 0.93359375, + "learning_rate": 0.001147181618217547, + "loss": 0.1177, + "step": 46915 + }, + { + "epoch": 0.4072534092586002, + "grad_norm": 0.447265625, + "learning_rate": 0.001147151896660925, + "loss": 0.1182, + "step": 46916 + }, + { + "epoch": 0.40726208973880434, + "grad_norm": 0.5390625, + "learning_rate": 0.0011471221750527383, + "loss": 0.0801, + "step": 46917 + }, + { + "epoch": 0.40727077021900854, + "grad_norm": 0.46875, + "learning_rate": 0.0011470924533930197, + "loss": 0.1162, + "step": 46918 + }, + { + "epoch": 0.4072794506992127, + "grad_norm": 0.359375, + "learning_rate": 0.0011470627316818012, + "loss": 0.0732, + "step": 46919 + }, + { + "epoch": 0.40728813117941687, + "grad_norm": 0.16015625, + "learning_rate": 0.0011470330099191157, + "loss": 0.1523, + "step": 46920 + }, + { + "epoch": 0.407296811659621, + "grad_norm": 0.283203125, + "learning_rate": 0.0011470032881049958, + "loss": 0.1172, + "step": 46921 + }, + { + "epoch": 0.4073054921398252, + "grad_norm": 0.37890625, + "learning_rate": 0.0011469735662394734, + "loss": 0.1377, + "step": 46922 + }, + { + "epoch": 0.40731417262002934, + "grad_norm": 0.408203125, + "learning_rate": 0.0011469438443225816, + "loss": 0.0894, + "step": 46923 + }, + { + "epoch": 0.40732285310023353, + "grad_norm": 0.28515625, + "learning_rate": 0.0011469141223543527, + "loss": 0.1143, + "step": 46924 + }, + { + "epoch": 0.40733153358043767, + "grad_norm": 0.6875, + "learning_rate": 0.0011468844003348193, + "loss": 0.1104, + "step": 46925 + }, + { + "epoch": 0.40734021406064186, + "grad_norm": 0.296875, + "learning_rate": 0.0011468546782640138, + "loss": 0.1074, + "step": 46926 + }, + { + "epoch": 0.407348894540846, + "grad_norm": 0.3515625, + "learning_rate": 0.0011468249561419688, + "loss": 0.1001, + "step": 46927 + }, + { + "epoch": 0.4073575750210502, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011467952339687167, + "loss": 0.124, + "step": 46928 + }, + { + "epoch": 0.4073662555012543, + "grad_norm": 0.41015625, + "learning_rate": 0.00114676551174429, + "loss": 0.127, + "step": 46929 + }, + { + "epoch": 0.4073749359814585, + "grad_norm": 0.27734375, + "learning_rate": 0.0011467357894687214, + "loss": 0.1221, + "step": 46930 + }, + { + "epoch": 0.40738361646166266, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011467060671420432, + "loss": 0.1172, + "step": 46931 + }, + { + "epoch": 0.40739229694186685, + "grad_norm": 0.2734375, + "learning_rate": 0.0011466763447642873, + "loss": 0.125, + "step": 46932 + }, + { + "epoch": 0.407400977422071, + "grad_norm": 0.1650390625, + "learning_rate": 0.0011466466223354877, + "loss": 0.0645, + "step": 46933 + }, + { + "epoch": 0.4074096579022752, + "grad_norm": 0.56640625, + "learning_rate": 0.0011466168998556757, + "loss": 0.126, + "step": 46934 + }, + { + "epoch": 0.4074183383824793, + "grad_norm": 0.41015625, + "learning_rate": 0.001146587177324884, + "loss": 0.0952, + "step": 46935 + }, + { + "epoch": 0.4074270188626835, + "grad_norm": 0.177734375, + "learning_rate": 0.0011465574547431456, + "loss": 0.125, + "step": 46936 + }, + { + "epoch": 0.40743569934288765, + "grad_norm": 0.68359375, + "learning_rate": 0.0011465277321104925, + "loss": 0.1963, + "step": 46937 + }, + { + "epoch": 0.40744437982309184, + "grad_norm": 0.37890625, + "learning_rate": 0.0011464980094269574, + "loss": 0.127, + "step": 46938 + }, + { + "epoch": 0.407453060303296, + "grad_norm": 0.28515625, + "learning_rate": 0.0011464682866925728, + "loss": 0.0728, + "step": 46939 + }, + { + "epoch": 0.40746174078350017, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011464385639073712, + "loss": 0.1201, + "step": 46940 + }, + { + "epoch": 0.4074704212637043, + "grad_norm": 0.1015625, + "learning_rate": 0.001146408841071385, + "loss": 0.0889, + "step": 46941 + }, + { + "epoch": 0.4074791017439085, + "grad_norm": 0.107421875, + "learning_rate": 0.0011463791181846469, + "loss": 0.0854, + "step": 46942 + }, + { + "epoch": 0.40748778222411264, + "grad_norm": 0.1611328125, + "learning_rate": 0.001146349395247189, + "loss": 0.1006, + "step": 46943 + }, + { + "epoch": 0.4074964627043168, + "grad_norm": 0.2216796875, + "learning_rate": 0.0011463196722590442, + "loss": 0.0986, + "step": 46944 + }, + { + "epoch": 0.40750514318452097, + "grad_norm": 0.2490234375, + "learning_rate": 0.0011462899492202451, + "loss": 0.1123, + "step": 46945 + }, + { + "epoch": 0.4075138236647251, + "grad_norm": 0.16796875, + "learning_rate": 0.0011462602261308242, + "loss": 0.1445, + "step": 46946 + }, + { + "epoch": 0.4075225041449293, + "grad_norm": 0.220703125, + "learning_rate": 0.0011462305029908135, + "loss": 0.0811, + "step": 46947 + }, + { + "epoch": 0.40753118462513344, + "grad_norm": 0.251953125, + "learning_rate": 0.0011462007798002459, + "loss": 0.1221, + "step": 46948 + }, + { + "epoch": 0.40753986510533763, + "grad_norm": 0.11572265625, + "learning_rate": 0.0011461710565591537, + "loss": 0.0986, + "step": 46949 + }, + { + "epoch": 0.40754854558554177, + "grad_norm": 0.1103515625, + "learning_rate": 0.0011461413332675696, + "loss": 0.1055, + "step": 46950 + }, + { + "epoch": 0.40755722606574596, + "grad_norm": 0.2021484375, + "learning_rate": 0.0011461116099255257, + "loss": 0.0977, + "step": 46951 + }, + { + "epoch": 0.4075659065459501, + "grad_norm": 0.515625, + "learning_rate": 0.0011460818865330553, + "loss": 0.0903, + "step": 46952 + }, + { + "epoch": 0.4075745870261543, + "grad_norm": 0.6875, + "learning_rate": 0.0011460521630901903, + "loss": 0.085, + "step": 46953 + }, + { + "epoch": 0.4075832675063584, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011460224395969634, + "loss": 0.0928, + "step": 46954 + }, + { + "epoch": 0.4075919479865626, + "grad_norm": 0.11376953125, + "learning_rate": 0.001145992716053407, + "loss": 0.1143, + "step": 46955 + }, + { + "epoch": 0.40760062846676676, + "grad_norm": 0.171875, + "learning_rate": 0.0011459629924595541, + "loss": 0.103, + "step": 46956 + }, + { + "epoch": 0.40760930894697095, + "grad_norm": 0.451171875, + "learning_rate": 0.001145933268815436, + "loss": 0.0908, + "step": 46957 + }, + { + "epoch": 0.4076179894271751, + "grad_norm": 0.12451171875, + "learning_rate": 0.0011459035451210863, + "loss": 0.1074, + "step": 46958 + }, + { + "epoch": 0.4076266699073793, + "grad_norm": 0.060302734375, + "learning_rate": 0.0011458738213765373, + "loss": 0.0737, + "step": 46959 + }, + { + "epoch": 0.4076353503875834, + "grad_norm": 0.177734375, + "learning_rate": 0.001145844097581821, + "loss": 0.0825, + "step": 46960 + }, + { + "epoch": 0.4076440308677876, + "grad_norm": 0.3984375, + "learning_rate": 0.0011458143737369704, + "loss": 0.1328, + "step": 46961 + }, + { + "epoch": 0.40765271134799175, + "grad_norm": 0.203125, + "learning_rate": 0.001145784649842018, + "loss": 0.1377, + "step": 46962 + }, + { + "epoch": 0.40766139182819594, + "grad_norm": 0.095703125, + "learning_rate": 0.0011457549258969966, + "loss": 0.105, + "step": 46963 + }, + { + "epoch": 0.4076700723084001, + "grad_norm": 0.1162109375, + "learning_rate": 0.0011457252019019377, + "loss": 0.0894, + "step": 46964 + }, + { + "epoch": 0.40767875278860427, + "grad_norm": 0.08154296875, + "learning_rate": 0.0011456954778568745, + "loss": 0.0918, + "step": 46965 + }, + { + "epoch": 0.4076874332688084, + "grad_norm": 0.96484375, + "learning_rate": 0.0011456657537618398, + "loss": 0.1162, + "step": 46966 + }, + { + "epoch": 0.4076961137490126, + "grad_norm": 0.75390625, + "learning_rate": 0.0011456360296168656, + "loss": 0.083, + "step": 46967 + }, + { + "epoch": 0.40770479422921674, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011456063054219841, + "loss": 0.1221, + "step": 46968 + }, + { + "epoch": 0.40771347470942093, + "grad_norm": 0.193359375, + "learning_rate": 0.0011455765811772285, + "loss": 0.1143, + "step": 46969 + }, + { + "epoch": 0.40772215518962507, + "grad_norm": 0.181640625, + "learning_rate": 0.0011455468568826311, + "loss": 0.1279, + "step": 46970 + }, + { + "epoch": 0.40773083566982926, + "grad_norm": 0.306640625, + "learning_rate": 0.001145517132538224, + "loss": 0.0835, + "step": 46971 + }, + { + "epoch": 0.4077395161500334, + "grad_norm": 0.2294921875, + "learning_rate": 0.0011454874081440406, + "loss": 0.0947, + "step": 46972 + }, + { + "epoch": 0.4077481966302376, + "grad_norm": 0.1962890625, + "learning_rate": 0.0011454576837001127, + "loss": 0.1226, + "step": 46973 + }, + { + "epoch": 0.40775687711044173, + "grad_norm": 0.2080078125, + "learning_rate": 0.0011454279592064726, + "loss": 0.1084, + "step": 46974 + }, + { + "epoch": 0.4077655575906459, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011453982346631534, + "loss": 0.103, + "step": 46975 + }, + { + "epoch": 0.40777423807085006, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011453685100701876, + "loss": 0.0952, + "step": 46976 + }, + { + "epoch": 0.40778291855105425, + "grad_norm": 0.1015625, + "learning_rate": 0.0011453387854276072, + "loss": 0.0908, + "step": 46977 + }, + { + "epoch": 0.4077915990312584, + "grad_norm": 0.392578125, + "learning_rate": 0.001145309060735445, + "loss": 0.0767, + "step": 46978 + }, + { + "epoch": 0.4078002795114626, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011452793359937334, + "loss": 0.103, + "step": 46979 + }, + { + "epoch": 0.4078089599916667, + "grad_norm": 0.208984375, + "learning_rate": 0.001145249611202505, + "loss": 0.125, + "step": 46980 + }, + { + "epoch": 0.4078176404718709, + "grad_norm": 0.2109375, + "learning_rate": 0.0011452198863617926, + "loss": 0.1162, + "step": 46981 + }, + { + "epoch": 0.40782632095207505, + "grad_norm": 0.1552734375, + "learning_rate": 0.0011451901614716285, + "loss": 0.0588, + "step": 46982 + }, + { + "epoch": 0.40783500143227924, + "grad_norm": 0.1240234375, + "learning_rate": 0.0011451604365320448, + "loss": 0.0918, + "step": 46983 + }, + { + "epoch": 0.4078436819124834, + "grad_norm": 0.2890625, + "learning_rate": 0.0011451307115430745, + "loss": 0.0952, + "step": 46984 + }, + { + "epoch": 0.4078523623926876, + "grad_norm": 0.158203125, + "learning_rate": 0.0011451009865047497, + "loss": 0.1758, + "step": 46985 + }, + { + "epoch": 0.4078610428728917, + "grad_norm": 0.51171875, + "learning_rate": 0.0011450712614171034, + "loss": 0.1396, + "step": 46986 + }, + { + "epoch": 0.4078697233530959, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011450415362801675, + "loss": 0.1138, + "step": 46987 + }, + { + "epoch": 0.40787840383330004, + "grad_norm": 0.1025390625, + "learning_rate": 0.0011450118110939752, + "loss": 0.123, + "step": 46988 + }, + { + "epoch": 0.40788708431350423, + "grad_norm": 0.50390625, + "learning_rate": 0.0011449820858585582, + "loss": 0.127, + "step": 46989 + }, + { + "epoch": 0.40789576479370837, + "grad_norm": 0.21875, + "learning_rate": 0.0011449523605739501, + "loss": 0.1084, + "step": 46990 + }, + { + "epoch": 0.40790444527391256, + "grad_norm": 0.419921875, + "learning_rate": 0.0011449226352401823, + "loss": 0.0889, + "step": 46991 + }, + { + "epoch": 0.4079131257541167, + "grad_norm": 0.1337890625, + "learning_rate": 0.001144892909857288, + "loss": 0.1172, + "step": 46992 + }, + { + "epoch": 0.4079218062343209, + "grad_norm": 0.228515625, + "learning_rate": 0.0011448631844252996, + "loss": 0.0854, + "step": 46993 + }, + { + "epoch": 0.40793048671452503, + "grad_norm": 0.2451171875, + "learning_rate": 0.0011448334589442492, + "loss": 0.0752, + "step": 46994 + }, + { + "epoch": 0.4079391671947292, + "grad_norm": 0.1328125, + "learning_rate": 0.00114480373341417, + "loss": 0.1045, + "step": 46995 + }, + { + "epoch": 0.40794784767493336, + "grad_norm": 0.1259765625, + "learning_rate": 0.0011447740078350936, + "loss": 0.1064, + "step": 46996 + }, + { + "epoch": 0.40795652815513755, + "grad_norm": 1.140625, + "learning_rate": 0.0011447442822070534, + "loss": 0.1104, + "step": 46997 + }, + { + "epoch": 0.4079652086353417, + "grad_norm": 0.56640625, + "learning_rate": 0.0011447145565300812, + "loss": 0.0889, + "step": 46998 + }, + { + "epoch": 0.4079738891155459, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011446848308042101, + "loss": 0.126, + "step": 46999 + }, + { + "epoch": 0.40798256959575, + "grad_norm": 0.67578125, + "learning_rate": 0.001144655105029472, + "loss": 0.1445, + "step": 47000 + }, + { + "epoch": 0.4079912500759542, + "grad_norm": 0.5625, + "learning_rate": 0.0011446253792059, + "loss": 0.1348, + "step": 47001 + }, + { + "epoch": 0.40799993055615835, + "grad_norm": 0.5859375, + "learning_rate": 0.0011445956533335264, + "loss": 0.0757, + "step": 47002 + }, + { + "epoch": 0.40800861103636255, + "grad_norm": 0.328125, + "learning_rate": 0.0011445659274123838, + "loss": 0.1133, + "step": 47003 + }, + { + "epoch": 0.4080172915165667, + "grad_norm": 0.09130859375, + "learning_rate": 0.0011445362014425042, + "loss": 0.1025, + "step": 47004 + }, + { + "epoch": 0.4080259719967709, + "grad_norm": 0.62109375, + "learning_rate": 0.001144506475423921, + "loss": 0.0654, + "step": 47005 + }, + { + "epoch": 0.408034652476975, + "grad_norm": 0.79296875, + "learning_rate": 0.0011444767493566657, + "loss": 0.0918, + "step": 47006 + }, + { + "epoch": 0.4080433329571792, + "grad_norm": 0.099609375, + "learning_rate": 0.0011444470232407715, + "loss": 0.125, + "step": 47007 + }, + { + "epoch": 0.40805201343738334, + "grad_norm": 0.306640625, + "learning_rate": 0.0011444172970762707, + "loss": 0.0991, + "step": 47008 + }, + { + "epoch": 0.40806069391758754, + "grad_norm": 0.76953125, + "learning_rate": 0.0011443875708631957, + "loss": 0.3223, + "step": 47009 + }, + { + "epoch": 0.4080693743977917, + "grad_norm": 0.345703125, + "learning_rate": 0.0011443578446015791, + "loss": 0.1191, + "step": 47010 + }, + { + "epoch": 0.40807805487799587, + "grad_norm": 0.484375, + "learning_rate": 0.0011443281182914535, + "loss": 0.0732, + "step": 47011 + }, + { + "epoch": 0.4080867353582, + "grad_norm": 0.28125, + "learning_rate": 0.0011442983919328516, + "loss": 0.1074, + "step": 47012 + }, + { + "epoch": 0.4080954158384042, + "grad_norm": 0.291015625, + "learning_rate": 0.0011442686655258052, + "loss": 0.0645, + "step": 47013 + }, + { + "epoch": 0.40810409631860833, + "grad_norm": 0.390625, + "learning_rate": 0.0011442389390703476, + "loss": 0.1084, + "step": 47014 + }, + { + "epoch": 0.4081127767988125, + "grad_norm": 0.1650390625, + "learning_rate": 0.0011442092125665106, + "loss": 0.1309, + "step": 47015 + }, + { + "epoch": 0.40812145727901666, + "grad_norm": 0.08203125, + "learning_rate": 0.0011441794860143275, + "loss": 0.0825, + "step": 47016 + }, + { + "epoch": 0.40813013775922086, + "grad_norm": 0.1953125, + "learning_rate": 0.0011441497594138301, + "loss": 0.1035, + "step": 47017 + }, + { + "epoch": 0.408138818239425, + "grad_norm": 0.455078125, + "learning_rate": 0.001144120032765051, + "loss": 0.1309, + "step": 47018 + }, + { + "epoch": 0.4081474987196292, + "grad_norm": 0.337890625, + "learning_rate": 0.0011440903060680233, + "loss": 0.0791, + "step": 47019 + }, + { + "epoch": 0.4081561791998333, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011440605793227791, + "loss": 0.1006, + "step": 47020 + }, + { + "epoch": 0.4081648596800375, + "grad_norm": 0.3203125, + "learning_rate": 0.0011440308525293506, + "loss": 0.123, + "step": 47021 + }, + { + "epoch": 0.40817354016024165, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011440011256877707, + "loss": 0.0874, + "step": 47022 + }, + { + "epoch": 0.40818222064044585, + "grad_norm": 0.27734375, + "learning_rate": 0.0011439713987980722, + "loss": 0.1064, + "step": 47023 + }, + { + "epoch": 0.40819090112065, + "grad_norm": 0.1845703125, + "learning_rate": 0.001143941671860287, + "loss": 0.1045, + "step": 47024 + }, + { + "epoch": 0.4081995816008542, + "grad_norm": 0.134765625, + "learning_rate": 0.0011439119448744478, + "loss": 0.1035, + "step": 47025 + }, + { + "epoch": 0.4082082620810583, + "grad_norm": 0.283203125, + "learning_rate": 0.001143882217840587, + "loss": 0.0889, + "step": 47026 + }, + { + "epoch": 0.4082169425612625, + "grad_norm": 0.3828125, + "learning_rate": 0.0011438524907587374, + "loss": 0.1001, + "step": 47027 + }, + { + "epoch": 0.40822562304146665, + "grad_norm": 0.283203125, + "learning_rate": 0.0011438227636289314, + "loss": 0.0928, + "step": 47028 + }, + { + "epoch": 0.40823430352167084, + "grad_norm": 0.33984375, + "learning_rate": 0.0011437930364512014, + "loss": 0.1582, + "step": 47029 + }, + { + "epoch": 0.408242984001875, + "grad_norm": 0.6171875, + "learning_rate": 0.00114376330922558, + "loss": 0.0874, + "step": 47030 + }, + { + "epoch": 0.40825166448207917, + "grad_norm": 0.07666015625, + "learning_rate": 0.0011437335819521, + "loss": 0.1006, + "step": 47031 + }, + { + "epoch": 0.4082603449622833, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011437038546307935, + "loss": 0.0879, + "step": 47032 + }, + { + "epoch": 0.4082690254424875, + "grad_norm": 0.578125, + "learning_rate": 0.0011436741272616928, + "loss": 0.1016, + "step": 47033 + }, + { + "epoch": 0.40827770592269164, + "grad_norm": 0.77734375, + "learning_rate": 0.001143644399844831, + "loss": 0.0898, + "step": 47034 + }, + { + "epoch": 0.40828638640289583, + "grad_norm": 0.09033203125, + "learning_rate": 0.0011436146723802403, + "loss": 0.1045, + "step": 47035 + }, + { + "epoch": 0.40829506688309997, + "grad_norm": 0.15234375, + "learning_rate": 0.0011435849448679534, + "loss": 0.1084, + "step": 47036 + }, + { + "epoch": 0.40830374736330416, + "grad_norm": 0.18359375, + "learning_rate": 0.0011435552173080022, + "loss": 0.0884, + "step": 47037 + }, + { + "epoch": 0.4083124278435083, + "grad_norm": 0.17578125, + "learning_rate": 0.0011435254897004202, + "loss": 0.085, + "step": 47038 + }, + { + "epoch": 0.4083211083237125, + "grad_norm": 0.72265625, + "learning_rate": 0.0011434957620452386, + "loss": 0.1069, + "step": 47039 + }, + { + "epoch": 0.4083297888039166, + "grad_norm": 0.392578125, + "learning_rate": 0.0011434660343424914, + "loss": 0.1089, + "step": 47040 + }, + { + "epoch": 0.4083384692841208, + "grad_norm": 0.2197265625, + "learning_rate": 0.0011434363065922104, + "loss": 0.1152, + "step": 47041 + }, + { + "epoch": 0.40834714976432496, + "grad_norm": 0.166015625, + "learning_rate": 0.0011434065787944282, + "loss": 0.1074, + "step": 47042 + }, + { + "epoch": 0.40835583024452915, + "grad_norm": 0.279296875, + "learning_rate": 0.0011433768509491766, + "loss": 0.104, + "step": 47043 + }, + { + "epoch": 0.4083645107247333, + "grad_norm": 0.42578125, + "learning_rate": 0.0011433471230564891, + "loss": 0.1445, + "step": 47044 + }, + { + "epoch": 0.4083731912049375, + "grad_norm": 0.212890625, + "learning_rate": 0.0011433173951163977, + "loss": 0.1328, + "step": 47045 + }, + { + "epoch": 0.4083818716851416, + "grad_norm": 0.0810546875, + "learning_rate": 0.0011432876671289354, + "loss": 0.1367, + "step": 47046 + }, + { + "epoch": 0.4083905521653458, + "grad_norm": 0.310546875, + "learning_rate": 0.001143257939094134, + "loss": 0.1162, + "step": 47047 + }, + { + "epoch": 0.40839923264554995, + "grad_norm": 0.3515625, + "learning_rate": 0.0011432282110120263, + "loss": 0.1143, + "step": 47048 + }, + { + "epoch": 0.40840791312575414, + "grad_norm": 0.388671875, + "learning_rate": 0.0011431984828826451, + "loss": 0.1064, + "step": 47049 + }, + { + "epoch": 0.4084165936059583, + "grad_norm": 0.49609375, + "learning_rate": 0.0011431687547060227, + "loss": 0.0708, + "step": 47050 + }, + { + "epoch": 0.40842527408616247, + "grad_norm": 0.150390625, + "learning_rate": 0.0011431390264821917, + "loss": 0.0869, + "step": 47051 + }, + { + "epoch": 0.4084339545663666, + "grad_norm": 0.1923828125, + "learning_rate": 0.001143109298211184, + "loss": 0.0894, + "step": 47052 + }, + { + "epoch": 0.4084426350465708, + "grad_norm": 0.1904296875, + "learning_rate": 0.0011430795698930331, + "loss": 0.1201, + "step": 47053 + }, + { + "epoch": 0.40845131552677494, + "grad_norm": 0.443359375, + "learning_rate": 0.0011430498415277712, + "loss": 0.0869, + "step": 47054 + }, + { + "epoch": 0.40845999600697913, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011430201131154302, + "loss": 0.0977, + "step": 47055 + }, + { + "epoch": 0.40846867648718327, + "grad_norm": 0.12109375, + "learning_rate": 0.0011429903846560435, + "loss": 0.0996, + "step": 47056 + }, + { + "epoch": 0.40847735696738746, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011429606561496426, + "loss": 0.0698, + "step": 47057 + }, + { + "epoch": 0.4084860374475916, + "grad_norm": 0.13671875, + "learning_rate": 0.0011429309275962608, + "loss": 0.0713, + "step": 47058 + }, + { + "epoch": 0.4084947179277958, + "grad_norm": 0.41015625, + "learning_rate": 0.0011429011989959304, + "loss": 0.1172, + "step": 47059 + }, + { + "epoch": 0.40850339840799993, + "grad_norm": 0.08935546875, + "learning_rate": 0.0011428714703486838, + "loss": 0.0791, + "step": 47060 + }, + { + "epoch": 0.4085120788882041, + "grad_norm": 0.11669921875, + "learning_rate": 0.0011428417416545539, + "loss": 0.085, + "step": 47061 + }, + { + "epoch": 0.40852075936840826, + "grad_norm": 0.248046875, + "learning_rate": 0.0011428120129135728, + "loss": 0.0859, + "step": 47062 + }, + { + "epoch": 0.40852943984861245, + "grad_norm": 0.51171875, + "learning_rate": 0.0011427822841257732, + "loss": 0.1436, + "step": 47063 + }, + { + "epoch": 0.4085381203288166, + "grad_norm": 0.30078125, + "learning_rate": 0.0011427525552911873, + "loss": 0.1094, + "step": 47064 + }, + { + "epoch": 0.4085468008090208, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011427228264098478, + "loss": 0.0967, + "step": 47065 + }, + { + "epoch": 0.4085554812892249, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011426930974817874, + "loss": 0.1084, + "step": 47066 + }, + { + "epoch": 0.40856416176942906, + "grad_norm": 0.1591796875, + "learning_rate": 0.0011426633685070384, + "loss": 0.1006, + "step": 47067 + }, + { + "epoch": 0.40857284224963325, + "grad_norm": 0.08447265625, + "learning_rate": 0.0011426336394856334, + "loss": 0.1025, + "step": 47068 + }, + { + "epoch": 0.4085815227298374, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011426039104176049, + "loss": 0.0747, + "step": 47069 + }, + { + "epoch": 0.4085902032100416, + "grad_norm": 0.08837890625, + "learning_rate": 0.0011425741813029856, + "loss": 0.1128, + "step": 47070 + }, + { + "epoch": 0.4085988836902457, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011425444521418076, + "loss": 0.1162, + "step": 47071 + }, + { + "epoch": 0.4086075641704499, + "grad_norm": 0.134765625, + "learning_rate": 0.0011425147229341037, + "loss": 0.1084, + "step": 47072 + }, + { + "epoch": 0.40861624465065405, + "grad_norm": 0.244140625, + "learning_rate": 0.0011424849936799062, + "loss": 0.1045, + "step": 47073 + }, + { + "epoch": 0.40862492513085824, + "grad_norm": 0.287109375, + "learning_rate": 0.0011424552643792478, + "loss": 0.0879, + "step": 47074 + }, + { + "epoch": 0.4086336056110624, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011424255350321612, + "loss": 0.1006, + "step": 47075 + }, + { + "epoch": 0.40864228609126657, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011423958056386782, + "loss": 0.106, + "step": 47076 + }, + { + "epoch": 0.4086509665714707, + "grad_norm": 0.189453125, + "learning_rate": 0.0011423660761988323, + "loss": 0.0815, + "step": 47077 + }, + { + "epoch": 0.4086596470516749, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011423363467126552, + "loss": 0.1226, + "step": 47078 + }, + { + "epoch": 0.40866832753187904, + "grad_norm": 0.302734375, + "learning_rate": 0.0011423066171801798, + "loss": 0.0811, + "step": 47079 + }, + { + "epoch": 0.40867700801208323, + "grad_norm": 0.439453125, + "learning_rate": 0.001142276887601439, + "loss": 0.1162, + "step": 47080 + }, + { + "epoch": 0.40868568849228737, + "grad_norm": 0.341796875, + "learning_rate": 0.0011422471579764643, + "loss": 0.0981, + "step": 47081 + }, + { + "epoch": 0.40869436897249156, + "grad_norm": 0.138671875, + "learning_rate": 0.0011422174283052884, + "loss": 0.1006, + "step": 47082 + }, + { + "epoch": 0.4087030494526957, + "grad_norm": 0.427734375, + "learning_rate": 0.0011421876985879447, + "loss": 0.0913, + "step": 47083 + }, + { + "epoch": 0.4087117299328999, + "grad_norm": 0.34375, + "learning_rate": 0.0011421579688244653, + "loss": 0.1152, + "step": 47084 + }, + { + "epoch": 0.40872041041310403, + "grad_norm": 0.287109375, + "learning_rate": 0.001142128239014882, + "loss": 0.0903, + "step": 47085 + }, + { + "epoch": 0.4087290908933082, + "grad_norm": 0.60546875, + "learning_rate": 0.0011420985091592278, + "loss": 0.1016, + "step": 47086 + }, + { + "epoch": 0.40873777137351236, + "grad_norm": 0.271484375, + "learning_rate": 0.0011420687792575359, + "loss": 0.0791, + "step": 47087 + }, + { + "epoch": 0.40874645185371655, + "grad_norm": 0.443359375, + "learning_rate": 0.001142039049309838, + "loss": 0.1079, + "step": 47088 + }, + { + "epoch": 0.4087551323339207, + "grad_norm": 0.388671875, + "learning_rate": 0.0011420093193161669, + "loss": 0.1143, + "step": 47089 + }, + { + "epoch": 0.4087638128141249, + "grad_norm": 0.3125, + "learning_rate": 0.001141979589276555, + "loss": 0.124, + "step": 47090 + }, + { + "epoch": 0.408772493294329, + "grad_norm": 0.130859375, + "learning_rate": 0.001141949859191035, + "loss": 0.0967, + "step": 47091 + }, + { + "epoch": 0.4087811737745332, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011419201290596389, + "loss": 0.1055, + "step": 47092 + }, + { + "epoch": 0.40878985425473735, + "grad_norm": 0.154296875, + "learning_rate": 0.0011418903988823998, + "loss": 0.1006, + "step": 47093 + }, + { + "epoch": 0.40879853473494154, + "grad_norm": 0.70703125, + "learning_rate": 0.0011418606686593499, + "loss": 0.0845, + "step": 47094 + }, + { + "epoch": 0.4088072152151457, + "grad_norm": 0.1396484375, + "learning_rate": 0.001141830938390522, + "loss": 0.0698, + "step": 47095 + }, + { + "epoch": 0.4088158956953499, + "grad_norm": 0.1689453125, + "learning_rate": 0.001141801208075948, + "loss": 0.0791, + "step": 47096 + }, + { + "epoch": 0.408824576175554, + "grad_norm": 0.2353515625, + "learning_rate": 0.001141771477715661, + "loss": 0.0942, + "step": 47097 + }, + { + "epoch": 0.4088332566557582, + "grad_norm": 0.49609375, + "learning_rate": 0.0011417417473096937, + "loss": 0.0889, + "step": 47098 + }, + { + "epoch": 0.40884193713596234, + "grad_norm": 0.53515625, + "learning_rate": 0.0011417120168580777, + "loss": 0.0703, + "step": 47099 + }, + { + "epoch": 0.40885061761616653, + "grad_norm": 0.09375, + "learning_rate": 0.0011416822863608465, + "loss": 0.0854, + "step": 47100 + }, + { + "epoch": 0.40885929809637067, + "grad_norm": 0.46875, + "learning_rate": 0.001141652555818032, + "loss": 0.1196, + "step": 47101 + }, + { + "epoch": 0.40886797857657486, + "grad_norm": 0.1767578125, + "learning_rate": 0.001141622825229667, + "loss": 0.0679, + "step": 47102 + }, + { + "epoch": 0.408876659056779, + "grad_norm": 0.26171875, + "learning_rate": 0.0011415930945957839, + "loss": 0.0684, + "step": 47103 + }, + { + "epoch": 0.4088853395369832, + "grad_norm": 0.287109375, + "learning_rate": 0.001141563363916415, + "loss": 0.0913, + "step": 47104 + }, + { + "epoch": 0.40889402001718733, + "grad_norm": 0.2353515625, + "learning_rate": 0.0011415336331915928, + "loss": 0.1182, + "step": 47105 + }, + { + "epoch": 0.4089027004973915, + "grad_norm": 0.09765625, + "learning_rate": 0.0011415039024213506, + "loss": 0.0874, + "step": 47106 + }, + { + "epoch": 0.40891138097759566, + "grad_norm": 0.17578125, + "learning_rate": 0.0011414741716057203, + "loss": 0.0967, + "step": 47107 + }, + { + "epoch": 0.40892006145779985, + "grad_norm": 0.43359375, + "learning_rate": 0.0011414444407447343, + "loss": 0.1309, + "step": 47108 + }, + { + "epoch": 0.408928741938004, + "grad_norm": 0.291015625, + "learning_rate": 0.0011414147098384253, + "loss": 0.1162, + "step": 47109 + }, + { + "epoch": 0.4089374224182082, + "grad_norm": 0.2216796875, + "learning_rate": 0.0011413849788868258, + "loss": 0.1147, + "step": 47110 + }, + { + "epoch": 0.4089461028984123, + "grad_norm": 0.11865234375, + "learning_rate": 0.0011413552478899682, + "loss": 0.0908, + "step": 47111 + }, + { + "epoch": 0.4089547833786165, + "grad_norm": 0.15625, + "learning_rate": 0.0011413255168478853, + "loss": 0.0996, + "step": 47112 + }, + { + "epoch": 0.40896346385882065, + "grad_norm": 0.44140625, + "learning_rate": 0.0011412957857606094, + "loss": 0.1602, + "step": 47113 + }, + { + "epoch": 0.40897214433902485, + "grad_norm": 0.197265625, + "learning_rate": 0.001141266054628173, + "loss": 0.0752, + "step": 47114 + }, + { + "epoch": 0.408980824819229, + "grad_norm": 0.193359375, + "learning_rate": 0.0011412363234506086, + "loss": 0.0991, + "step": 47115 + }, + { + "epoch": 0.4089895052994332, + "grad_norm": 0.18359375, + "learning_rate": 0.001141206592227949, + "loss": 0.0962, + "step": 47116 + }, + { + "epoch": 0.4089981857796373, + "grad_norm": 0.294921875, + "learning_rate": 0.0011411768609602265, + "loss": 0.1206, + "step": 47117 + }, + { + "epoch": 0.4090068662598415, + "grad_norm": 0.57421875, + "learning_rate": 0.0011411471296474732, + "loss": 0.1191, + "step": 47118 + }, + { + "epoch": 0.40901554674004564, + "grad_norm": 0.08544921875, + "learning_rate": 0.0011411173982897225, + "loss": 0.1035, + "step": 47119 + }, + { + "epoch": 0.40902422722024984, + "grad_norm": 0.232421875, + "learning_rate": 0.0011410876668870062, + "loss": 0.1182, + "step": 47120 + }, + { + "epoch": 0.409032907700454, + "grad_norm": 0.12060546875, + "learning_rate": 0.0011410579354393568, + "loss": 0.0918, + "step": 47121 + }, + { + "epoch": 0.40904158818065817, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011410282039468075, + "loss": 0.0918, + "step": 47122 + }, + { + "epoch": 0.4090502686608623, + "grad_norm": 0.11669921875, + "learning_rate": 0.0011409984724093903, + "loss": 0.0737, + "step": 47123 + }, + { + "epoch": 0.4090589491410665, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011409687408271375, + "loss": 0.1055, + "step": 47124 + }, + { + "epoch": 0.40906762962127063, + "grad_norm": 0.48046875, + "learning_rate": 0.001140939009200082, + "loss": 0.0942, + "step": 47125 + }, + { + "epoch": 0.4090763101014748, + "grad_norm": 0.140625, + "learning_rate": 0.0011409092775282563, + "loss": 0.1143, + "step": 47126 + }, + { + "epoch": 0.40908499058167896, + "grad_norm": 0.1201171875, + "learning_rate": 0.0011408795458116927, + "loss": 0.0967, + "step": 47127 + }, + { + "epoch": 0.40909367106188316, + "grad_norm": 0.31640625, + "learning_rate": 0.0011408498140504243, + "loss": 0.082, + "step": 47128 + }, + { + "epoch": 0.4091023515420873, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011408200822444826, + "loss": 0.0898, + "step": 47129 + }, + { + "epoch": 0.4091110320222915, + "grad_norm": 0.0947265625, + "learning_rate": 0.0011407903503939008, + "loss": 0.0972, + "step": 47130 + }, + { + "epoch": 0.4091197125024956, + "grad_norm": 0.279296875, + "learning_rate": 0.0011407606184987117, + "loss": 0.0894, + "step": 47131 + }, + { + "epoch": 0.4091283929826998, + "grad_norm": 0.14453125, + "learning_rate": 0.0011407308865589469, + "loss": 0.1128, + "step": 47132 + }, + { + "epoch": 0.40913707346290396, + "grad_norm": 0.373046875, + "learning_rate": 0.0011407011545746394, + "loss": 0.083, + "step": 47133 + }, + { + "epoch": 0.40914575394310815, + "grad_norm": 0.197265625, + "learning_rate": 0.001140671422545822, + "loss": 0.1562, + "step": 47134 + }, + { + "epoch": 0.4091544344233123, + "grad_norm": 0.2265625, + "learning_rate": 0.0011406416904725268, + "loss": 0.1143, + "step": 47135 + }, + { + "epoch": 0.4091631149035165, + "grad_norm": 0.375, + "learning_rate": 0.0011406119583547865, + "loss": 0.0928, + "step": 47136 + }, + { + "epoch": 0.4091717953837206, + "grad_norm": 0.373046875, + "learning_rate": 0.0011405822261926336, + "loss": 0.0879, + "step": 47137 + }, + { + "epoch": 0.4091804758639248, + "grad_norm": 0.5390625, + "learning_rate": 0.0011405524939861005, + "loss": 0.1074, + "step": 47138 + }, + { + "epoch": 0.40918915634412895, + "grad_norm": 0.423828125, + "learning_rate": 0.00114052276173522, + "loss": 0.0928, + "step": 47139 + }, + { + "epoch": 0.40919783682433314, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011404930294400244, + "loss": 0.0752, + "step": 47140 + }, + { + "epoch": 0.4092065173045373, + "grad_norm": 0.24609375, + "learning_rate": 0.0011404632971005462, + "loss": 0.0796, + "step": 47141 + }, + { + "epoch": 0.40921519778474147, + "grad_norm": 0.189453125, + "learning_rate": 0.0011404335647168175, + "loss": 0.0869, + "step": 47142 + }, + { + "epoch": 0.4092238782649456, + "grad_norm": 0.138671875, + "learning_rate": 0.0011404038322888716, + "loss": 0.1084, + "step": 47143 + }, + { + "epoch": 0.4092325587451498, + "grad_norm": 0.11474609375, + "learning_rate": 0.0011403740998167408, + "loss": 0.084, + "step": 47144 + }, + { + "epoch": 0.40924123922535394, + "grad_norm": 0.216796875, + "learning_rate": 0.0011403443673004574, + "loss": 0.0889, + "step": 47145 + }, + { + "epoch": 0.40924991970555813, + "grad_norm": 0.1435546875, + "learning_rate": 0.001140314634740054, + "loss": 0.0869, + "step": 47146 + }, + { + "epoch": 0.40925860018576227, + "grad_norm": 0.1533203125, + "learning_rate": 0.001140284902135563, + "loss": 0.1182, + "step": 47147 + }, + { + "epoch": 0.40926728066596646, + "grad_norm": 0.169921875, + "learning_rate": 0.0011402551694870173, + "loss": 0.1138, + "step": 47148 + }, + { + "epoch": 0.4092759611461706, + "grad_norm": 0.2080078125, + "learning_rate": 0.001140225436794449, + "loss": 0.085, + "step": 47149 + }, + { + "epoch": 0.4092846416263748, + "grad_norm": 0.17578125, + "learning_rate": 0.001140195704057891, + "loss": 0.0981, + "step": 47150 + }, + { + "epoch": 0.4092933221065789, + "grad_norm": 0.171875, + "learning_rate": 0.001140165971277375, + "loss": 0.1318, + "step": 47151 + }, + { + "epoch": 0.4093020025867831, + "grad_norm": 0.189453125, + "learning_rate": 0.0011401362384529348, + "loss": 0.1279, + "step": 47152 + }, + { + "epoch": 0.40931068306698726, + "grad_norm": 0.224609375, + "learning_rate": 0.0011401065055846016, + "loss": 0.1211, + "step": 47153 + }, + { + "epoch": 0.40931936354719145, + "grad_norm": 0.6640625, + "learning_rate": 0.0011400767726724088, + "loss": 0.168, + "step": 47154 + }, + { + "epoch": 0.4093280440273956, + "grad_norm": 0.455078125, + "learning_rate": 0.0011400470397163888, + "loss": 0.0874, + "step": 47155 + }, + { + "epoch": 0.4093367245075998, + "grad_norm": 0.09765625, + "learning_rate": 0.0011400173067165737, + "loss": 0.1084, + "step": 47156 + }, + { + "epoch": 0.4093454049878039, + "grad_norm": 0.45703125, + "learning_rate": 0.0011399875736729963, + "loss": 0.1016, + "step": 47157 + }, + { + "epoch": 0.4093540854680081, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011399578405856892, + "loss": 0.0801, + "step": 47158 + }, + { + "epoch": 0.40936276594821225, + "grad_norm": 0.359375, + "learning_rate": 0.001139928107454685, + "loss": 0.084, + "step": 47159 + }, + { + "epoch": 0.40937144642841644, + "grad_norm": 0.330078125, + "learning_rate": 0.0011398983742800155, + "loss": 0.1094, + "step": 47160 + }, + { + "epoch": 0.4093801269086206, + "grad_norm": 0.07958984375, + "learning_rate": 0.001139868641061714, + "loss": 0.0771, + "step": 47161 + }, + { + "epoch": 0.40938880738882477, + "grad_norm": 0.1474609375, + "learning_rate": 0.001139838907799813, + "loss": 0.0703, + "step": 47162 + }, + { + "epoch": 0.4093974878690289, + "grad_norm": 0.38671875, + "learning_rate": 0.0011398091744943444, + "loss": 0.1445, + "step": 47163 + }, + { + "epoch": 0.4094061683492331, + "grad_norm": 0.41796875, + "learning_rate": 0.001139779441145341, + "loss": 0.1309, + "step": 47164 + }, + { + "epoch": 0.40941484882943724, + "grad_norm": 0.43359375, + "learning_rate": 0.001139749707752836, + "loss": 0.0674, + "step": 47165 + }, + { + "epoch": 0.40942352930964143, + "grad_norm": 0.09228515625, + "learning_rate": 0.0011397199743168609, + "loss": 0.0654, + "step": 47166 + }, + { + "epoch": 0.40943220978984557, + "grad_norm": 0.451171875, + "learning_rate": 0.0011396902408374488, + "loss": 0.1177, + "step": 47167 + }, + { + "epoch": 0.40944089027004976, + "grad_norm": 0.265625, + "learning_rate": 0.0011396605073146317, + "loss": 0.1011, + "step": 47168 + }, + { + "epoch": 0.4094495707502539, + "grad_norm": 0.1533203125, + "learning_rate": 0.001139630773748443, + "loss": 0.0869, + "step": 47169 + }, + { + "epoch": 0.4094582512304581, + "grad_norm": 0.3203125, + "learning_rate": 0.0011396010401389144, + "loss": 0.1108, + "step": 47170 + }, + { + "epoch": 0.40946693171066223, + "grad_norm": 0.5234375, + "learning_rate": 0.0011395713064860785, + "loss": 0.085, + "step": 47171 + }, + { + "epoch": 0.4094756121908664, + "grad_norm": 0.2138671875, + "learning_rate": 0.001139541572789968, + "loss": 0.1035, + "step": 47172 + }, + { + "epoch": 0.40948429267107056, + "grad_norm": 0.1162109375, + "learning_rate": 0.0011395118390506156, + "loss": 0.0737, + "step": 47173 + }, + { + "epoch": 0.40949297315127475, + "grad_norm": 0.0830078125, + "learning_rate": 0.0011394821052680536, + "loss": 0.0732, + "step": 47174 + }, + { + "epoch": 0.4095016536314789, + "grad_norm": 0.4296875, + "learning_rate": 0.001139452371442315, + "loss": 0.0957, + "step": 47175 + }, + { + "epoch": 0.4095103341116831, + "grad_norm": 0.10791015625, + "learning_rate": 0.0011394226375734315, + "loss": 0.1211, + "step": 47176 + }, + { + "epoch": 0.4095190145918872, + "grad_norm": 0.212890625, + "learning_rate": 0.001139392903661436, + "loss": 0.0713, + "step": 47177 + }, + { + "epoch": 0.4095276950720914, + "grad_norm": 0.185546875, + "learning_rate": 0.001139363169706361, + "loss": 0.1025, + "step": 47178 + }, + { + "epoch": 0.40953637555229555, + "grad_norm": 0.33984375, + "learning_rate": 0.001139333435708239, + "loss": 0.1309, + "step": 47179 + }, + { + "epoch": 0.40954505603249974, + "grad_norm": 0.40234375, + "learning_rate": 0.0011393037016671028, + "loss": 0.0913, + "step": 47180 + }, + { + "epoch": 0.4095537365127039, + "grad_norm": 0.59765625, + "learning_rate": 0.0011392739675829841, + "loss": 0.1172, + "step": 47181 + }, + { + "epoch": 0.4095624169929081, + "grad_norm": 0.5859375, + "learning_rate": 0.0011392442334559163, + "loss": 0.0972, + "step": 47182 + }, + { + "epoch": 0.4095710974731122, + "grad_norm": 0.6015625, + "learning_rate": 0.0011392144992859318, + "loss": 0.0884, + "step": 47183 + }, + { + "epoch": 0.4095797779533164, + "grad_norm": 0.267578125, + "learning_rate": 0.001139184765073063, + "loss": 0.1279, + "step": 47184 + }, + { + "epoch": 0.40958845843352054, + "grad_norm": 0.16796875, + "learning_rate": 0.001139155030817342, + "loss": 0.0991, + "step": 47185 + }, + { + "epoch": 0.40959713891372473, + "grad_norm": 0.431640625, + "learning_rate": 0.0011391252965188014, + "loss": 0.0996, + "step": 47186 + }, + { + "epoch": 0.40960581939392887, + "grad_norm": 0.58984375, + "learning_rate": 0.0011390955621774746, + "loss": 0.1167, + "step": 47187 + }, + { + "epoch": 0.40961449987413306, + "grad_norm": 0.7734375, + "learning_rate": 0.0011390658277933932, + "loss": 0.0869, + "step": 47188 + }, + { + "epoch": 0.4096231803543372, + "grad_norm": 0.734375, + "learning_rate": 0.00113903609336659, + "loss": 0.0986, + "step": 47189 + }, + { + "epoch": 0.40963186083454134, + "grad_norm": 0.1103515625, + "learning_rate": 0.001139006358897097, + "loss": 0.1055, + "step": 47190 + }, + { + "epoch": 0.40964054131474553, + "grad_norm": 0.197265625, + "learning_rate": 0.001138976624384948, + "loss": 0.1318, + "step": 47191 + }, + { + "epoch": 0.40964922179494967, + "grad_norm": 0.62890625, + "learning_rate": 0.0011389468898301742, + "loss": 0.0908, + "step": 47192 + }, + { + "epoch": 0.40965790227515386, + "grad_norm": 0.12353515625, + "learning_rate": 0.0011389171552328092, + "loss": 0.1191, + "step": 47193 + }, + { + "epoch": 0.409666582755358, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011388874205928846, + "loss": 0.082, + "step": 47194 + }, + { + "epoch": 0.4096752632355622, + "grad_norm": 0.2490234375, + "learning_rate": 0.0011388576859104333, + "loss": 0.0757, + "step": 47195 + }, + { + "epoch": 0.40968394371576633, + "grad_norm": 0.447265625, + "learning_rate": 0.0011388279511854883, + "loss": 0.1025, + "step": 47196 + }, + { + "epoch": 0.4096926241959705, + "grad_norm": 0.14453125, + "learning_rate": 0.001138798216418081, + "loss": 0.1123, + "step": 47197 + }, + { + "epoch": 0.40970130467617466, + "grad_norm": 0.119140625, + "learning_rate": 0.0011387684816082451, + "loss": 0.0859, + "step": 47198 + }, + { + "epoch": 0.40970998515637885, + "grad_norm": 0.103515625, + "learning_rate": 0.0011387387467560119, + "loss": 0.1143, + "step": 47199 + }, + { + "epoch": 0.409718665636583, + "grad_norm": 0.29296875, + "learning_rate": 0.0011387090118614154, + "loss": 0.124, + "step": 47200 + }, + { + "epoch": 0.4097273461167872, + "grad_norm": 0.6015625, + "learning_rate": 0.0011386792769244867, + "loss": 0.1064, + "step": 47201 + }, + { + "epoch": 0.4097360265969913, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011386495419452591, + "loss": 0.1006, + "step": 47202 + }, + { + "epoch": 0.4097447070771955, + "grad_norm": 0.27734375, + "learning_rate": 0.0011386198069237648, + "loss": 0.1191, + "step": 47203 + }, + { + "epoch": 0.40975338755739965, + "grad_norm": 0.09765625, + "learning_rate": 0.0011385900718600367, + "loss": 0.0859, + "step": 47204 + }, + { + "epoch": 0.40976206803760384, + "grad_norm": 0.1142578125, + "learning_rate": 0.001138560336754107, + "loss": 0.1367, + "step": 47205 + }, + { + "epoch": 0.409770748517808, + "grad_norm": 0.2333984375, + "learning_rate": 0.0011385306016060085, + "loss": 0.1104, + "step": 47206 + }, + { + "epoch": 0.4097794289980122, + "grad_norm": 0.4375, + "learning_rate": 0.001138500866415773, + "loss": 0.1182, + "step": 47207 + }, + { + "epoch": 0.4097881094782163, + "grad_norm": 0.3046875, + "learning_rate": 0.001138471131183434, + "loss": 0.0884, + "step": 47208 + }, + { + "epoch": 0.4097967899584205, + "grad_norm": 0.7265625, + "learning_rate": 0.0011384413959090235, + "loss": 0.0947, + "step": 47209 + }, + { + "epoch": 0.40980547043862464, + "grad_norm": 0.490234375, + "learning_rate": 0.001138411660592574, + "loss": 0.0854, + "step": 47210 + }, + { + "epoch": 0.40981415091882883, + "grad_norm": 0.1416015625, + "learning_rate": 0.0011383819252341178, + "loss": 0.1426, + "step": 47211 + }, + { + "epoch": 0.40982283139903297, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011383521898336879, + "loss": 0.0928, + "step": 47212 + }, + { + "epoch": 0.40983151187923716, + "grad_norm": 0.265625, + "learning_rate": 0.0011383224543913167, + "loss": 0.0771, + "step": 47213 + }, + { + "epoch": 0.4098401923594413, + "grad_norm": 0.78515625, + "learning_rate": 0.0011382927189070367, + "loss": 0.0996, + "step": 47214 + }, + { + "epoch": 0.4098488728396455, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011382629833808801, + "loss": 0.1074, + "step": 47215 + }, + { + "epoch": 0.40985755331984963, + "grad_norm": 0.47265625, + "learning_rate": 0.0011382332478128798, + "loss": 0.0879, + "step": 47216 + }, + { + "epoch": 0.4098662338000538, + "grad_norm": 0.421875, + "learning_rate": 0.0011382035122030683, + "loss": 0.1699, + "step": 47217 + }, + { + "epoch": 0.40987491428025796, + "grad_norm": 0.232421875, + "learning_rate": 0.0011381737765514782, + "loss": 0.123, + "step": 47218 + }, + { + "epoch": 0.40988359476046216, + "grad_norm": 0.1650390625, + "learning_rate": 0.001138144040858141, + "loss": 0.1504, + "step": 47219 + }, + { + "epoch": 0.4098922752406663, + "grad_norm": 0.25390625, + "learning_rate": 0.0011381143051230908, + "loss": 0.0781, + "step": 47220 + }, + { + "epoch": 0.4099009557208705, + "grad_norm": 0.384765625, + "learning_rate": 0.0011380845693463593, + "loss": 0.0747, + "step": 47221 + }, + { + "epoch": 0.4099096362010746, + "grad_norm": 0.279296875, + "learning_rate": 0.0011380548335279787, + "loss": 0.0801, + "step": 47222 + }, + { + "epoch": 0.4099183166812788, + "grad_norm": 0.33984375, + "learning_rate": 0.0011380250976679824, + "loss": 0.0757, + "step": 47223 + }, + { + "epoch": 0.40992699716148295, + "grad_norm": 0.1611328125, + "learning_rate": 0.0011379953617664024, + "loss": 0.0693, + "step": 47224 + }, + { + "epoch": 0.40993567764168715, + "grad_norm": 0.375, + "learning_rate": 0.0011379656258232707, + "loss": 0.1221, + "step": 47225 + }, + { + "epoch": 0.4099443581218913, + "grad_norm": 0.466796875, + "learning_rate": 0.0011379358898386208, + "loss": 0.123, + "step": 47226 + }, + { + "epoch": 0.4099530386020955, + "grad_norm": 0.1982421875, + "learning_rate": 0.0011379061538124846, + "loss": 0.0942, + "step": 47227 + }, + { + "epoch": 0.4099617190822996, + "grad_norm": 0.267578125, + "learning_rate": 0.001137876417744895, + "loss": 0.0928, + "step": 47228 + }, + { + "epoch": 0.4099703995625038, + "grad_norm": 0.36328125, + "learning_rate": 0.0011378466816358843, + "loss": 0.1289, + "step": 47229 + }, + { + "epoch": 0.40997908004270794, + "grad_norm": 0.23046875, + "learning_rate": 0.0011378169454854847, + "loss": 0.0918, + "step": 47230 + }, + { + "epoch": 0.40998776052291214, + "grad_norm": 0.46875, + "learning_rate": 0.0011377872092937294, + "loss": 0.0991, + "step": 47231 + }, + { + "epoch": 0.4099964410031163, + "grad_norm": 0.30078125, + "learning_rate": 0.0011377574730606504, + "loss": 0.0947, + "step": 47232 + }, + { + "epoch": 0.41000512148332047, + "grad_norm": 0.345703125, + "learning_rate": 0.0011377277367862805, + "loss": 0.0894, + "step": 47233 + }, + { + "epoch": 0.4100138019635246, + "grad_norm": 0.7578125, + "learning_rate": 0.001137698000470652, + "loss": 0.1045, + "step": 47234 + }, + { + "epoch": 0.4100224824437288, + "grad_norm": 1.375, + "learning_rate": 0.0011376682641137978, + "loss": 0.1113, + "step": 47235 + }, + { + "epoch": 0.41003116292393293, + "grad_norm": 0.279296875, + "learning_rate": 0.00113763852771575, + "loss": 0.1445, + "step": 47236 + }, + { + "epoch": 0.4100398434041371, + "grad_norm": 0.2109375, + "learning_rate": 0.0011376087912765413, + "loss": 0.1094, + "step": 47237 + }, + { + "epoch": 0.41004852388434126, + "grad_norm": 0.09521484375, + "learning_rate": 0.001137579054796204, + "loss": 0.0986, + "step": 47238 + }, + { + "epoch": 0.41005720436454546, + "grad_norm": 0.69921875, + "learning_rate": 0.0011375493182747708, + "loss": 0.1191, + "step": 47239 + }, + { + "epoch": 0.4100658848447496, + "grad_norm": 0.0712890625, + "learning_rate": 0.0011375195817122745, + "loss": 0.1064, + "step": 47240 + }, + { + "epoch": 0.4100745653249538, + "grad_norm": 0.13671875, + "learning_rate": 0.0011374898451087473, + "loss": 0.1494, + "step": 47241 + }, + { + "epoch": 0.4100832458051579, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011374601084642218, + "loss": 0.0908, + "step": 47242 + }, + { + "epoch": 0.4100919262853621, + "grad_norm": 0.1259765625, + "learning_rate": 0.0011374303717787302, + "loss": 0.084, + "step": 47243 + }, + { + "epoch": 0.41010060676556626, + "grad_norm": 0.51171875, + "learning_rate": 0.0011374006350523057, + "loss": 0.0898, + "step": 47244 + }, + { + "epoch": 0.41010928724577045, + "grad_norm": 0.267578125, + "learning_rate": 0.00113737089828498, + "loss": 0.0972, + "step": 47245 + }, + { + "epoch": 0.4101179677259746, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011373411614767862, + "loss": 0.1074, + "step": 47246 + }, + { + "epoch": 0.4101266482061788, + "grad_norm": 0.1513671875, + "learning_rate": 0.001137311424627757, + "loss": 0.0986, + "step": 47247 + }, + { + "epoch": 0.4101353286863829, + "grad_norm": 0.400390625, + "learning_rate": 0.0011372816877379243, + "loss": 0.0952, + "step": 47248 + }, + { + "epoch": 0.4101440091665871, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011372519508073206, + "loss": 0.1475, + "step": 47249 + }, + { + "epoch": 0.41015268964679125, + "grad_norm": 0.22265625, + "learning_rate": 0.001137222213835979, + "loss": 0.0928, + "step": 47250 + }, + { + "epoch": 0.41016137012699544, + "grad_norm": 0.41015625, + "learning_rate": 0.0011371924768239319, + "loss": 0.105, + "step": 47251 + }, + { + "epoch": 0.4101700506071996, + "grad_norm": 0.2265625, + "learning_rate": 0.0011371627397712117, + "loss": 0.127, + "step": 47252 + }, + { + "epoch": 0.41017873108740377, + "grad_norm": 0.21484375, + "learning_rate": 0.0011371330026778506, + "loss": 0.0791, + "step": 47253 + }, + { + "epoch": 0.4101874115676079, + "grad_norm": 0.65625, + "learning_rate": 0.0011371032655438818, + "loss": 0.1182, + "step": 47254 + }, + { + "epoch": 0.4101960920478121, + "grad_norm": 0.78125, + "learning_rate": 0.001137073528369337, + "loss": 0.1084, + "step": 47255 + }, + { + "epoch": 0.41020477252801624, + "grad_norm": 1.09375, + "learning_rate": 0.0011370437911542494, + "loss": 0.1309, + "step": 47256 + }, + { + "epoch": 0.41021345300822043, + "grad_norm": 0.443359375, + "learning_rate": 0.0011370140538986514, + "loss": 0.082, + "step": 47257 + }, + { + "epoch": 0.41022213348842457, + "grad_norm": 0.3359375, + "learning_rate": 0.0011369843166025748, + "loss": 0.1084, + "step": 47258 + }, + { + "epoch": 0.41023081396862876, + "grad_norm": 0.2734375, + "learning_rate": 0.0011369545792660532, + "loss": 0.1182, + "step": 47259 + }, + { + "epoch": 0.4102394944488329, + "grad_norm": 0.234375, + "learning_rate": 0.0011369248418891182, + "loss": 0.1465, + "step": 47260 + }, + { + "epoch": 0.4102481749290371, + "grad_norm": 0.1328125, + "learning_rate": 0.0011368951044718033, + "loss": 0.1143, + "step": 47261 + }, + { + "epoch": 0.41025685540924123, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011368653670141404, + "loss": 0.1201, + "step": 47262 + }, + { + "epoch": 0.4102655358894454, + "grad_norm": 1.015625, + "learning_rate": 0.0011368356295161617, + "loss": 0.3262, + "step": 47263 + }, + { + "epoch": 0.41027421636964956, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011368058919779005, + "loss": 0.0903, + "step": 47264 + }, + { + "epoch": 0.41028289684985375, + "grad_norm": 0.171875, + "learning_rate": 0.0011367761543993888, + "loss": 0.0889, + "step": 47265 + }, + { + "epoch": 0.4102915773300579, + "grad_norm": 0.263671875, + "learning_rate": 0.0011367464167806591, + "loss": 0.1074, + "step": 47266 + }, + { + "epoch": 0.4103002578102621, + "grad_norm": 0.267578125, + "learning_rate": 0.0011367166791217442, + "loss": 0.1279, + "step": 47267 + }, + { + "epoch": 0.4103089382904662, + "grad_norm": 0.388671875, + "learning_rate": 0.0011366869414226764, + "loss": 0.1348, + "step": 47268 + }, + { + "epoch": 0.4103176187706704, + "grad_norm": 0.283203125, + "learning_rate": 0.0011366572036834884, + "loss": 0.125, + "step": 47269 + }, + { + "epoch": 0.41032629925087455, + "grad_norm": 0.41796875, + "learning_rate": 0.0011366274659042126, + "loss": 0.1006, + "step": 47270 + }, + { + "epoch": 0.41033497973107874, + "grad_norm": 0.138671875, + "learning_rate": 0.0011365977280848817, + "loss": 0.1113, + "step": 47271 + }, + { + "epoch": 0.4103436602112829, + "grad_norm": 0.73828125, + "learning_rate": 0.0011365679902255277, + "loss": 0.0942, + "step": 47272 + }, + { + "epoch": 0.41035234069148707, + "grad_norm": 0.58203125, + "learning_rate": 0.0011365382523261841, + "loss": 0.1006, + "step": 47273 + }, + { + "epoch": 0.4103610211716912, + "grad_norm": 0.2890625, + "learning_rate": 0.0011365085143868823, + "loss": 0.126, + "step": 47274 + }, + { + "epoch": 0.4103697016518954, + "grad_norm": 0.1083984375, + "learning_rate": 0.0011364787764076556, + "loss": 0.103, + "step": 47275 + }, + { + "epoch": 0.41037838213209954, + "grad_norm": 0.23828125, + "learning_rate": 0.001136449038388536, + "loss": 0.1055, + "step": 47276 + }, + { + "epoch": 0.41038706261230373, + "grad_norm": 0.78515625, + "learning_rate": 0.0011364193003295564, + "loss": 0.0908, + "step": 47277 + }, + { + "epoch": 0.41039574309250787, + "grad_norm": 0.169921875, + "learning_rate": 0.0011363895622307492, + "loss": 0.0791, + "step": 47278 + }, + { + "epoch": 0.41040442357271206, + "grad_norm": 0.1767578125, + "learning_rate": 0.001136359824092147, + "loss": 0.082, + "step": 47279 + }, + { + "epoch": 0.4104131040529162, + "grad_norm": 0.58203125, + "learning_rate": 0.0011363300859137822, + "loss": 0.0928, + "step": 47280 + }, + { + "epoch": 0.4104217845331204, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011363003476956872, + "loss": 0.0664, + "step": 47281 + }, + { + "epoch": 0.41043046501332453, + "grad_norm": 0.11083984375, + "learning_rate": 0.0011362706094378949, + "loss": 0.1084, + "step": 47282 + }, + { + "epoch": 0.4104391454935287, + "grad_norm": 0.5078125, + "learning_rate": 0.0011362408711404377, + "loss": 0.1006, + "step": 47283 + }, + { + "epoch": 0.41044782597373286, + "grad_norm": 0.189453125, + "learning_rate": 0.0011362111328033478, + "loss": 0.1143, + "step": 47284 + }, + { + "epoch": 0.41045650645393705, + "grad_norm": 0.330078125, + "learning_rate": 0.001136181394426658, + "loss": 0.0679, + "step": 47285 + }, + { + "epoch": 0.4104651869341412, + "grad_norm": 0.1650390625, + "learning_rate": 0.001136151656010401, + "loss": 0.1309, + "step": 47286 + }, + { + "epoch": 0.4104738674143454, + "grad_norm": 0.357421875, + "learning_rate": 0.0011361219175546084, + "loss": 0.1006, + "step": 47287 + }, + { + "epoch": 0.4104825478945495, + "grad_norm": 0.212890625, + "learning_rate": 0.0011360921790593143, + "loss": 0.1514, + "step": 47288 + }, + { + "epoch": 0.4104912283747537, + "grad_norm": 0.462890625, + "learning_rate": 0.0011360624405245497, + "loss": 0.1797, + "step": 47289 + }, + { + "epoch": 0.41049990885495785, + "grad_norm": 0.2080078125, + "learning_rate": 0.001136032701950348, + "loss": 0.0952, + "step": 47290 + }, + { + "epoch": 0.41050858933516204, + "grad_norm": 0.103515625, + "learning_rate": 0.0011360029633367414, + "loss": 0.0981, + "step": 47291 + }, + { + "epoch": 0.4105172698153662, + "grad_norm": 0.57421875, + "learning_rate": 0.0011359732246837627, + "loss": 0.0767, + "step": 47292 + }, + { + "epoch": 0.4105259502955704, + "grad_norm": 0.43359375, + "learning_rate": 0.001135943485991444, + "loss": 0.1035, + "step": 47293 + }, + { + "epoch": 0.4105346307757745, + "grad_norm": 1.140625, + "learning_rate": 0.0011359137472598184, + "loss": 0.1318, + "step": 47294 + }, + { + "epoch": 0.4105433112559787, + "grad_norm": 0.2470703125, + "learning_rate": 0.0011358840084889177, + "loss": 0.1465, + "step": 47295 + }, + { + "epoch": 0.41055199173618284, + "grad_norm": 0.1376953125, + "learning_rate": 0.001135854269678775, + "loss": 0.1133, + "step": 47296 + }, + { + "epoch": 0.41056067221638703, + "grad_norm": 0.1650390625, + "learning_rate": 0.0011358245308294224, + "loss": 0.0879, + "step": 47297 + }, + { + "epoch": 0.41056935269659117, + "grad_norm": 0.1064453125, + "learning_rate": 0.0011357947919408927, + "loss": 0.0645, + "step": 47298 + }, + { + "epoch": 0.41057803317679537, + "grad_norm": 0.345703125, + "learning_rate": 0.0011357650530132183, + "loss": 0.1055, + "step": 47299 + }, + { + "epoch": 0.4105867136569995, + "grad_norm": 0.4765625, + "learning_rate": 0.001135735314046432, + "loss": 0.1113, + "step": 47300 + }, + { + "epoch": 0.4105953941372037, + "grad_norm": 0.76171875, + "learning_rate": 0.0011357055750405658, + "loss": 0.0903, + "step": 47301 + }, + { + "epoch": 0.41060407461740783, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011356758359956527, + "loss": 0.0874, + "step": 47302 + }, + { + "epoch": 0.410612755097612, + "grad_norm": 0.125, + "learning_rate": 0.001135646096911725, + "loss": 0.0791, + "step": 47303 + }, + { + "epoch": 0.41062143557781616, + "grad_norm": 0.271484375, + "learning_rate": 0.0011356163577888154, + "loss": 0.1152, + "step": 47304 + }, + { + "epoch": 0.41063011605802036, + "grad_norm": 0.404296875, + "learning_rate": 0.0011355866186269564, + "loss": 0.0918, + "step": 47305 + }, + { + "epoch": 0.4106387965382245, + "grad_norm": 0.72265625, + "learning_rate": 0.00113555687942618, + "loss": 0.1099, + "step": 47306 + }, + { + "epoch": 0.4106474770184287, + "grad_norm": 0.490234375, + "learning_rate": 0.001135527140186519, + "loss": 0.1895, + "step": 47307 + }, + { + "epoch": 0.4106561574986328, + "grad_norm": 0.173828125, + "learning_rate": 0.0011354974009080065, + "loss": 0.0996, + "step": 47308 + }, + { + "epoch": 0.410664837978837, + "grad_norm": 0.65234375, + "learning_rate": 0.0011354676615906744, + "loss": 0.1011, + "step": 47309 + }, + { + "epoch": 0.41067351845904115, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011354379222345554, + "loss": 0.1016, + "step": 47310 + }, + { + "epoch": 0.41068219893924535, + "grad_norm": 0.263671875, + "learning_rate": 0.0011354081828396823, + "loss": 0.0869, + "step": 47311 + }, + { + "epoch": 0.4106908794194495, + "grad_norm": 0.91015625, + "learning_rate": 0.001135378443406087, + "loss": 0.1079, + "step": 47312 + }, + { + "epoch": 0.4106995598996536, + "grad_norm": 0.2333984375, + "learning_rate": 0.0011353487039338023, + "loss": 0.0791, + "step": 47313 + }, + { + "epoch": 0.4107082403798578, + "grad_norm": 0.1181640625, + "learning_rate": 0.0011353189644228612, + "loss": 0.0957, + "step": 47314 + }, + { + "epoch": 0.41071692086006195, + "grad_norm": 0.28515625, + "learning_rate": 0.0011352892248732954, + "loss": 0.0703, + "step": 47315 + }, + { + "epoch": 0.41072560134026614, + "grad_norm": 0.24609375, + "learning_rate": 0.0011352594852851378, + "loss": 0.0693, + "step": 47316 + }, + { + "epoch": 0.4107342818204703, + "grad_norm": 1.3515625, + "learning_rate": 0.0011352297456584215, + "loss": 0.1416, + "step": 47317 + }, + { + "epoch": 0.4107429623006745, + "grad_norm": 0.2412109375, + "learning_rate": 0.001135200005993178, + "loss": 0.0938, + "step": 47318 + }, + { + "epoch": 0.4107516427808786, + "grad_norm": 0.15234375, + "learning_rate": 0.0011351702662894406, + "loss": 0.0845, + "step": 47319 + }, + { + "epoch": 0.4107603232610828, + "grad_norm": 0.318359375, + "learning_rate": 0.0011351405265472411, + "loss": 0.0801, + "step": 47320 + }, + { + "epoch": 0.41076900374128694, + "grad_norm": 0.21875, + "learning_rate": 0.0011351107867666127, + "loss": 0.084, + "step": 47321 + }, + { + "epoch": 0.41077768422149114, + "grad_norm": 0.10400390625, + "learning_rate": 0.0011350810469475878, + "loss": 0.085, + "step": 47322 + }, + { + "epoch": 0.4107863647016953, + "grad_norm": 0.796875, + "learning_rate": 0.001135051307090199, + "loss": 0.1055, + "step": 47323 + }, + { + "epoch": 0.41079504518189947, + "grad_norm": 0.421875, + "learning_rate": 0.0011350215671944782, + "loss": 0.1108, + "step": 47324 + }, + { + "epoch": 0.4108037256621036, + "grad_norm": 0.39453125, + "learning_rate": 0.0011349918272604584, + "loss": 0.0825, + "step": 47325 + }, + { + "epoch": 0.4108124061423078, + "grad_norm": 0.2138671875, + "learning_rate": 0.0011349620872881721, + "loss": 0.0791, + "step": 47326 + }, + { + "epoch": 0.41082108662251193, + "grad_norm": 0.1318359375, + "learning_rate": 0.001134932347277652, + "loss": 0.1162, + "step": 47327 + }, + { + "epoch": 0.4108297671027161, + "grad_norm": 0.271484375, + "learning_rate": 0.0011349026072289296, + "loss": 0.1035, + "step": 47328 + }, + { + "epoch": 0.41083844758292026, + "grad_norm": 0.177734375, + "learning_rate": 0.001134872867142039, + "loss": 0.1025, + "step": 47329 + }, + { + "epoch": 0.41084712806312446, + "grad_norm": 0.07275390625, + "learning_rate": 0.0011348431270170119, + "loss": 0.0859, + "step": 47330 + }, + { + "epoch": 0.4108558085433286, + "grad_norm": 0.455078125, + "learning_rate": 0.0011348133868538809, + "loss": 0.0913, + "step": 47331 + }, + { + "epoch": 0.4108644890235328, + "grad_norm": 0.1728515625, + "learning_rate": 0.0011347836466526783, + "loss": 0.0996, + "step": 47332 + }, + { + "epoch": 0.4108731695037369, + "grad_norm": 0.390625, + "learning_rate": 0.001134753906413437, + "loss": 0.125, + "step": 47333 + }, + { + "epoch": 0.4108818499839411, + "grad_norm": 0.193359375, + "learning_rate": 0.0011347241661361892, + "loss": 0.0654, + "step": 47334 + }, + { + "epoch": 0.41089053046414525, + "grad_norm": 0.23046875, + "learning_rate": 0.0011346944258209676, + "loss": 0.1387, + "step": 47335 + }, + { + "epoch": 0.41089921094434945, + "grad_norm": 0.22265625, + "learning_rate": 0.0011346646854678043, + "loss": 0.1182, + "step": 47336 + }, + { + "epoch": 0.4109078914245536, + "grad_norm": 0.55078125, + "learning_rate": 0.0011346349450767328, + "loss": 0.2188, + "step": 47337 + }, + { + "epoch": 0.4109165719047578, + "grad_norm": 0.34375, + "learning_rate": 0.0011346052046477847, + "loss": 0.0908, + "step": 47338 + }, + { + "epoch": 0.4109252523849619, + "grad_norm": 0.201171875, + "learning_rate": 0.0011345754641809932, + "loss": 0.1006, + "step": 47339 + }, + { + "epoch": 0.4109339328651661, + "grad_norm": 0.515625, + "learning_rate": 0.0011345457236763906, + "loss": 0.1226, + "step": 47340 + }, + { + "epoch": 0.41094261334537024, + "grad_norm": 0.12109375, + "learning_rate": 0.0011345159831340089, + "loss": 0.064, + "step": 47341 + }, + { + "epoch": 0.41095129382557444, + "grad_norm": 0.263671875, + "learning_rate": 0.0011344862425538811, + "loss": 0.0654, + "step": 47342 + }, + { + "epoch": 0.4109599743057786, + "grad_norm": 0.10302734375, + "learning_rate": 0.00113445650193604, + "loss": 0.1021, + "step": 47343 + }, + { + "epoch": 0.41096865478598277, + "grad_norm": 0.076171875, + "learning_rate": 0.0011344267612805174, + "loss": 0.1279, + "step": 47344 + }, + { + "epoch": 0.4109773352661869, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011343970205873464, + "loss": 0.1074, + "step": 47345 + }, + { + "epoch": 0.4109860157463911, + "grad_norm": 0.126953125, + "learning_rate": 0.0011343672798565592, + "loss": 0.1025, + "step": 47346 + }, + { + "epoch": 0.41099469622659524, + "grad_norm": 0.6171875, + "learning_rate": 0.0011343375390881884, + "loss": 0.1357, + "step": 47347 + }, + { + "epoch": 0.41100337670679943, + "grad_norm": 0.3203125, + "learning_rate": 0.0011343077982822669, + "loss": 0.0991, + "step": 47348 + }, + { + "epoch": 0.41101205718700357, + "grad_norm": 0.12158203125, + "learning_rate": 0.001134278057438827, + "loss": 0.1582, + "step": 47349 + }, + { + "epoch": 0.41102073766720776, + "grad_norm": 0.330078125, + "learning_rate": 0.0011342483165579006, + "loss": 0.0967, + "step": 47350 + }, + { + "epoch": 0.4110294181474119, + "grad_norm": 0.29296875, + "learning_rate": 0.001134218575639521, + "loss": 0.0938, + "step": 47351 + }, + { + "epoch": 0.4110380986276161, + "grad_norm": 0.16796875, + "learning_rate": 0.0011341888346837205, + "loss": 0.085, + "step": 47352 + }, + { + "epoch": 0.4110467791078202, + "grad_norm": 0.376953125, + "learning_rate": 0.0011341590936905317, + "loss": 0.0762, + "step": 47353 + }, + { + "epoch": 0.4110554595880244, + "grad_norm": 0.1201171875, + "learning_rate": 0.0011341293526599867, + "loss": 0.085, + "step": 47354 + }, + { + "epoch": 0.41106414006822856, + "grad_norm": 0.1640625, + "learning_rate": 0.0011340996115921185, + "loss": 0.0654, + "step": 47355 + }, + { + "epoch": 0.41107282054843275, + "grad_norm": 0.330078125, + "learning_rate": 0.0011340698704869596, + "loss": 0.1025, + "step": 47356 + }, + { + "epoch": 0.4110815010286369, + "grad_norm": 0.091796875, + "learning_rate": 0.001134040129344543, + "loss": 0.1074, + "step": 47357 + }, + { + "epoch": 0.4110901815088411, + "grad_norm": 0.10400390625, + "learning_rate": 0.0011340103881648997, + "loss": 0.0791, + "step": 47358 + }, + { + "epoch": 0.4110988619890452, + "grad_norm": 0.66015625, + "learning_rate": 0.0011339806469480634, + "loss": 0.1074, + "step": 47359 + }, + { + "epoch": 0.4111075424692494, + "grad_norm": 0.173828125, + "learning_rate": 0.0011339509056940665, + "loss": 0.1543, + "step": 47360 + }, + { + "epoch": 0.41111622294945355, + "grad_norm": 1.0390625, + "learning_rate": 0.0011339211644029413, + "loss": 0.0903, + "step": 47361 + }, + { + "epoch": 0.41112490342965774, + "grad_norm": 0.375, + "learning_rate": 0.0011338914230747206, + "loss": 0.1113, + "step": 47362 + }, + { + "epoch": 0.4111335839098619, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011338616817094362, + "loss": 0.0854, + "step": 47363 + }, + { + "epoch": 0.41114226439006607, + "grad_norm": 0.1474609375, + "learning_rate": 0.0011338319403071215, + "loss": 0.0835, + "step": 47364 + }, + { + "epoch": 0.4111509448702702, + "grad_norm": 0.5078125, + "learning_rate": 0.0011338021988678089, + "loss": 0.0991, + "step": 47365 + }, + { + "epoch": 0.4111596253504744, + "grad_norm": 0.6875, + "learning_rate": 0.0011337724573915306, + "loss": 0.0864, + "step": 47366 + }, + { + "epoch": 0.41116830583067854, + "grad_norm": 0.1259765625, + "learning_rate": 0.0011337427158783192, + "loss": 0.1221, + "step": 47367 + }, + { + "epoch": 0.41117698631088273, + "grad_norm": 0.56640625, + "learning_rate": 0.0011337129743282074, + "loss": 0.0771, + "step": 47368 + }, + { + "epoch": 0.41118566679108687, + "grad_norm": 0.271484375, + "learning_rate": 0.0011336832327412277, + "loss": 0.0991, + "step": 47369 + }, + { + "epoch": 0.41119434727129106, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011336534911174122, + "loss": 0.125, + "step": 47370 + }, + { + "epoch": 0.4112030277514952, + "grad_norm": 0.484375, + "learning_rate": 0.0011336237494567938, + "loss": 0.0962, + "step": 47371 + }, + { + "epoch": 0.4112117082316994, + "grad_norm": 0.416015625, + "learning_rate": 0.0011335940077594052, + "loss": 0.0977, + "step": 47372 + }, + { + "epoch": 0.41122038871190353, + "grad_norm": 0.546875, + "learning_rate": 0.0011335642660252785, + "loss": 0.127, + "step": 47373 + }, + { + "epoch": 0.4112290691921077, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011335345242544463, + "loss": 0.0723, + "step": 47374 + }, + { + "epoch": 0.41123774967231186, + "grad_norm": 0.18359375, + "learning_rate": 0.0011335047824469415, + "loss": 0.0967, + "step": 47375 + }, + { + "epoch": 0.41124643015251605, + "grad_norm": 0.0673828125, + "learning_rate": 0.0011334750406027962, + "loss": 0.0898, + "step": 47376 + }, + { + "epoch": 0.4112551106327202, + "grad_norm": 0.1953125, + "learning_rate": 0.0011334452987220433, + "loss": 0.1611, + "step": 47377 + }, + { + "epoch": 0.4112637911129244, + "grad_norm": 0.14453125, + "learning_rate": 0.001133415556804715, + "loss": 0.0874, + "step": 47378 + }, + { + "epoch": 0.4112724715931285, + "grad_norm": 0.248046875, + "learning_rate": 0.0011333858148508443, + "loss": 0.0684, + "step": 47379 + }, + { + "epoch": 0.4112811520733327, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011333560728604625, + "loss": 0.0664, + "step": 47380 + }, + { + "epoch": 0.41128983255353685, + "grad_norm": 0.279296875, + "learning_rate": 0.0011333263308336038, + "loss": 0.1377, + "step": 47381 + }, + { + "epoch": 0.41129851303374104, + "grad_norm": 0.267578125, + "learning_rate": 0.0011332965887702997, + "loss": 0.0918, + "step": 47382 + }, + { + "epoch": 0.4113071935139452, + "grad_norm": 0.2265625, + "learning_rate": 0.0011332668466705828, + "loss": 0.0825, + "step": 47383 + }, + { + "epoch": 0.4113158739941494, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011332371045344857, + "loss": 0.1523, + "step": 47384 + }, + { + "epoch": 0.4113245544743535, + "grad_norm": 0.2021484375, + "learning_rate": 0.0011332073623620414, + "loss": 0.0938, + "step": 47385 + }, + { + "epoch": 0.4113332349545577, + "grad_norm": 0.1650390625, + "learning_rate": 0.001133177620153282, + "loss": 0.0957, + "step": 47386 + }, + { + "epoch": 0.41134191543476184, + "grad_norm": 0.5234375, + "learning_rate": 0.0011331478779082395, + "loss": 0.1211, + "step": 47387 + }, + { + "epoch": 0.41135059591496603, + "grad_norm": 0.4375, + "learning_rate": 0.0011331181356269478, + "loss": 0.1064, + "step": 47388 + }, + { + "epoch": 0.41135927639517017, + "grad_norm": 0.55078125, + "learning_rate": 0.0011330883933094377, + "loss": 0.0894, + "step": 47389 + }, + { + "epoch": 0.41136795687537436, + "grad_norm": 0.119140625, + "learning_rate": 0.0011330586509557432, + "loss": 0.1099, + "step": 47390 + }, + { + "epoch": 0.4113766373555785, + "grad_norm": 1.2265625, + "learning_rate": 0.0011330289085658962, + "loss": 0.2002, + "step": 47391 + }, + { + "epoch": 0.4113853178357827, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011329991661399293, + "loss": 0.0874, + "step": 47392 + }, + { + "epoch": 0.41139399831598683, + "grad_norm": 0.150390625, + "learning_rate": 0.0011329694236778748, + "loss": 0.0825, + "step": 47393 + }, + { + "epoch": 0.411402678796191, + "grad_norm": 0.60546875, + "learning_rate": 0.0011329396811797657, + "loss": 0.1099, + "step": 47394 + }, + { + "epoch": 0.41141135927639516, + "grad_norm": 0.23046875, + "learning_rate": 0.0011329099386456342, + "loss": 0.1338, + "step": 47395 + }, + { + "epoch": 0.41142003975659935, + "grad_norm": 0.134765625, + "learning_rate": 0.0011328801960755128, + "loss": 0.0894, + "step": 47396 + }, + { + "epoch": 0.4114287202368035, + "grad_norm": 0.0673828125, + "learning_rate": 0.001132850453469434, + "loss": 0.0815, + "step": 47397 + }, + { + "epoch": 0.4114374007170077, + "grad_norm": 0.455078125, + "learning_rate": 0.0011328207108274306, + "loss": 0.106, + "step": 47398 + }, + { + "epoch": 0.4114460811972118, + "grad_norm": 0.2275390625, + "learning_rate": 0.001132790968149535, + "loss": 0.1387, + "step": 47399 + }, + { + "epoch": 0.411454761677416, + "grad_norm": 0.39453125, + "learning_rate": 0.0011327612254357798, + "loss": 0.0913, + "step": 47400 + }, + { + "epoch": 0.41146344215762015, + "grad_norm": 0.09375, + "learning_rate": 0.0011327314826861969, + "loss": 0.0972, + "step": 47401 + }, + { + "epoch": 0.41147212263782434, + "grad_norm": 0.486328125, + "learning_rate": 0.00113270173990082, + "loss": 0.1123, + "step": 47402 + }, + { + "epoch": 0.4114808031180285, + "grad_norm": 0.4140625, + "learning_rate": 0.0011326719970796803, + "loss": 0.1182, + "step": 47403 + }, + { + "epoch": 0.4114894835982327, + "grad_norm": 0.263671875, + "learning_rate": 0.0011326422542228113, + "loss": 0.1025, + "step": 47404 + }, + { + "epoch": 0.4114981640784368, + "grad_norm": 0.17578125, + "learning_rate": 0.0011326125113302454, + "loss": 0.1416, + "step": 47405 + }, + { + "epoch": 0.411506844558641, + "grad_norm": 0.169921875, + "learning_rate": 0.0011325827684020148, + "loss": 0.0762, + "step": 47406 + }, + { + "epoch": 0.41151552503884514, + "grad_norm": 0.13671875, + "learning_rate": 0.0011325530254381521, + "loss": 0.0693, + "step": 47407 + }, + { + "epoch": 0.41152420551904934, + "grad_norm": 0.263671875, + "learning_rate": 0.0011325232824386901, + "loss": 0.0957, + "step": 47408 + }, + { + "epoch": 0.4115328859992535, + "grad_norm": 0.625, + "learning_rate": 0.001132493539403661, + "loss": 0.106, + "step": 47409 + }, + { + "epoch": 0.41154156647945767, + "grad_norm": 0.10302734375, + "learning_rate": 0.0011324637963330973, + "loss": 0.1035, + "step": 47410 + }, + { + "epoch": 0.4115502469596618, + "grad_norm": 0.251953125, + "learning_rate": 0.001132434053227032, + "loss": 0.0864, + "step": 47411 + }, + { + "epoch": 0.411558927439866, + "grad_norm": 0.08984375, + "learning_rate": 0.001132404310085497, + "loss": 0.0967, + "step": 47412 + }, + { + "epoch": 0.41156760792007013, + "grad_norm": 0.14453125, + "learning_rate": 0.0011323745669085256, + "loss": 0.1299, + "step": 47413 + }, + { + "epoch": 0.4115762884002743, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011323448236961492, + "loss": 0.1504, + "step": 47414 + }, + { + "epoch": 0.41158496888047846, + "grad_norm": 0.11376953125, + "learning_rate": 0.0011323150804484015, + "loss": 0.0981, + "step": 47415 + }, + { + "epoch": 0.41159364936068266, + "grad_norm": 0.29296875, + "learning_rate": 0.0011322853371653144, + "loss": 0.0718, + "step": 47416 + }, + { + "epoch": 0.4116023298408868, + "grad_norm": 0.2197265625, + "learning_rate": 0.0011322555938469204, + "loss": 0.104, + "step": 47417 + }, + { + "epoch": 0.411611010321091, + "grad_norm": 0.0947265625, + "learning_rate": 0.0011322258504932524, + "loss": 0.0825, + "step": 47418 + }, + { + "epoch": 0.4116196908012951, + "grad_norm": 0.29296875, + "learning_rate": 0.0011321961071043423, + "loss": 0.1084, + "step": 47419 + }, + { + "epoch": 0.4116283712814993, + "grad_norm": 0.5390625, + "learning_rate": 0.0011321663636802235, + "loss": 0.1074, + "step": 47420 + }, + { + "epoch": 0.41163705176170345, + "grad_norm": 0.69140625, + "learning_rate": 0.0011321366202209275, + "loss": 0.0786, + "step": 47421 + }, + { + "epoch": 0.41164573224190765, + "grad_norm": 0.232421875, + "learning_rate": 0.0011321068767264878, + "loss": 0.0732, + "step": 47422 + }, + { + "epoch": 0.4116544127221118, + "grad_norm": 0.37109375, + "learning_rate": 0.0011320771331969361, + "loss": 0.0752, + "step": 47423 + }, + { + "epoch": 0.411663093202316, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011320473896323058, + "loss": 0.0835, + "step": 47424 + }, + { + "epoch": 0.4116717736825201, + "grad_norm": 0.07861328125, + "learning_rate": 0.001132017646032629, + "loss": 0.0786, + "step": 47425 + }, + { + "epoch": 0.4116804541627243, + "grad_norm": 0.11767578125, + "learning_rate": 0.001131987902397938, + "loss": 0.1201, + "step": 47426 + }, + { + "epoch": 0.41168913464292844, + "grad_norm": 0.1123046875, + "learning_rate": 0.0011319581587282655, + "loss": 0.104, + "step": 47427 + }, + { + "epoch": 0.41169781512313264, + "grad_norm": 1.9921875, + "learning_rate": 0.0011319284150236439, + "loss": 0.209, + "step": 47428 + }, + { + "epoch": 0.4117064956033368, + "grad_norm": 0.3984375, + "learning_rate": 0.001131898671284106, + "loss": 0.0938, + "step": 47429 + }, + { + "epoch": 0.41171517608354097, + "grad_norm": 0.318359375, + "learning_rate": 0.0011318689275096844, + "loss": 0.0933, + "step": 47430 + }, + { + "epoch": 0.4117238565637451, + "grad_norm": 0.271484375, + "learning_rate": 0.001131839183700411, + "loss": 0.1182, + "step": 47431 + }, + { + "epoch": 0.4117325370439493, + "grad_norm": 0.10205078125, + "learning_rate": 0.0011318094398563188, + "loss": 0.0972, + "step": 47432 + }, + { + "epoch": 0.41174121752415344, + "grad_norm": 0.349609375, + "learning_rate": 0.0011317796959774403, + "loss": 0.1064, + "step": 47433 + }, + { + "epoch": 0.41174989800435763, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011317499520638083, + "loss": 0.0977, + "step": 47434 + }, + { + "epoch": 0.41175857848456177, + "grad_norm": 0.267578125, + "learning_rate": 0.001131720208115455, + "loss": 0.0957, + "step": 47435 + }, + { + "epoch": 0.4117672589647659, + "grad_norm": 0.0986328125, + "learning_rate": 0.0011316904641324127, + "loss": 0.0864, + "step": 47436 + }, + { + "epoch": 0.4117759394449701, + "grad_norm": 0.1806640625, + "learning_rate": 0.001131660720114714, + "loss": 0.0879, + "step": 47437 + }, + { + "epoch": 0.41178461992517423, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011316309760623922, + "loss": 0.1396, + "step": 47438 + }, + { + "epoch": 0.4117933004053784, + "grad_norm": 0.0859375, + "learning_rate": 0.0011316012319754788, + "loss": 0.0791, + "step": 47439 + }, + { + "epoch": 0.41180198088558256, + "grad_norm": 0.34765625, + "learning_rate": 0.0011315714878540068, + "loss": 0.0747, + "step": 47440 + }, + { + "epoch": 0.41181066136578676, + "grad_norm": 0.27734375, + "learning_rate": 0.0011315417436980086, + "loss": 0.1348, + "step": 47441 + }, + { + "epoch": 0.4118193418459909, + "grad_norm": 0.11669921875, + "learning_rate": 0.001131511999507517, + "loss": 0.0967, + "step": 47442 + }, + { + "epoch": 0.4118280223261951, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011314822552825645, + "loss": 0.0981, + "step": 47443 + }, + { + "epoch": 0.4118367028063992, + "grad_norm": 0.30859375, + "learning_rate": 0.0011314525110231836, + "loss": 0.1128, + "step": 47444 + }, + { + "epoch": 0.4118453832866034, + "grad_norm": 0.23046875, + "learning_rate": 0.001131422766729406, + "loss": 0.1094, + "step": 47445 + }, + { + "epoch": 0.41185406376680755, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011313930224012656, + "loss": 0.1035, + "step": 47446 + }, + { + "epoch": 0.41186274424701175, + "grad_norm": 0.08935546875, + "learning_rate": 0.0011313632780387943, + "loss": 0.0957, + "step": 47447 + }, + { + "epoch": 0.4118714247272159, + "grad_norm": 0.08447265625, + "learning_rate": 0.0011313335336420242, + "loss": 0.0684, + "step": 47448 + }, + { + "epoch": 0.4118801052074201, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011313037892109886, + "loss": 0.1089, + "step": 47449 + }, + { + "epoch": 0.4118887856876242, + "grad_norm": 0.328125, + "learning_rate": 0.0011312740447457193, + "loss": 0.0859, + "step": 47450 + }, + { + "epoch": 0.4118974661678284, + "grad_norm": 0.09521484375, + "learning_rate": 0.001131244300246249, + "loss": 0.0874, + "step": 47451 + }, + { + "epoch": 0.41190614664803255, + "grad_norm": 0.56640625, + "learning_rate": 0.0011312145557126109, + "loss": 0.1396, + "step": 47452 + }, + { + "epoch": 0.41191482712823674, + "grad_norm": 0.67578125, + "learning_rate": 0.0011311848111448368, + "loss": 0.1494, + "step": 47453 + }, + { + "epoch": 0.4119235076084409, + "grad_norm": 0.2578125, + "learning_rate": 0.0011311550665429593, + "loss": 0.1328, + "step": 47454 + }, + { + "epoch": 0.41193218808864507, + "grad_norm": 0.10205078125, + "learning_rate": 0.0011311253219070114, + "loss": 0.1035, + "step": 47455 + }, + { + "epoch": 0.4119408685688492, + "grad_norm": 0.11328125, + "learning_rate": 0.0011310955772370253, + "loss": 0.1138, + "step": 47456 + }, + { + "epoch": 0.4119495490490534, + "grad_norm": 0.083984375, + "learning_rate": 0.0011310658325330335, + "loss": 0.1143, + "step": 47457 + }, + { + "epoch": 0.41195822952925754, + "grad_norm": 0.11669921875, + "learning_rate": 0.0011310360877950683, + "loss": 0.083, + "step": 47458 + }, + { + "epoch": 0.41196691000946173, + "grad_norm": 0.14453125, + "learning_rate": 0.0011310063430231628, + "loss": 0.0664, + "step": 47459 + }, + { + "epoch": 0.41197559048966587, + "grad_norm": 0.15625, + "learning_rate": 0.0011309765982173488, + "loss": 0.1025, + "step": 47460 + }, + { + "epoch": 0.41198427096987006, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011309468533776596, + "loss": 0.1167, + "step": 47461 + }, + { + "epoch": 0.4119929514500742, + "grad_norm": 0.169921875, + "learning_rate": 0.0011309171085041275, + "loss": 0.0747, + "step": 47462 + }, + { + "epoch": 0.4120016319302784, + "grad_norm": 0.177734375, + "learning_rate": 0.0011308873635967847, + "loss": 0.0879, + "step": 47463 + }, + { + "epoch": 0.4120103124104825, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011308576186556641, + "loss": 0.1118, + "step": 47464 + }, + { + "epoch": 0.4120189928906867, + "grad_norm": 0.09033203125, + "learning_rate": 0.001130827873680798, + "loss": 0.0996, + "step": 47465 + }, + { + "epoch": 0.41202767337089086, + "grad_norm": 0.14453125, + "learning_rate": 0.001130798128672219, + "loss": 0.104, + "step": 47466 + }, + { + "epoch": 0.41203635385109505, + "grad_norm": 0.2431640625, + "learning_rate": 0.0011307683836299595, + "loss": 0.1006, + "step": 47467 + }, + { + "epoch": 0.4120450343312992, + "grad_norm": 0.314453125, + "learning_rate": 0.0011307386385540524, + "loss": 0.0713, + "step": 47468 + }, + { + "epoch": 0.4120537148115034, + "grad_norm": 0.1884765625, + "learning_rate": 0.0011307088934445297, + "loss": 0.0913, + "step": 47469 + }, + { + "epoch": 0.4120623952917075, + "grad_norm": 0.29296875, + "learning_rate": 0.0011306791483014244, + "loss": 0.1172, + "step": 47470 + }, + { + "epoch": 0.4120710757719117, + "grad_norm": 0.31640625, + "learning_rate": 0.0011306494031247687, + "loss": 0.1104, + "step": 47471 + }, + { + "epoch": 0.41207975625211585, + "grad_norm": 0.1298828125, + "learning_rate": 0.0011306196579145954, + "loss": 0.0889, + "step": 47472 + }, + { + "epoch": 0.41208843673232004, + "grad_norm": 0.142578125, + "learning_rate": 0.001130589912670937, + "loss": 0.085, + "step": 47473 + }, + { + "epoch": 0.4120971172125242, + "grad_norm": 0.09716796875, + "learning_rate": 0.0011305601673938259, + "loss": 0.0859, + "step": 47474 + }, + { + "epoch": 0.41210579769272837, + "grad_norm": 0.244140625, + "learning_rate": 0.0011305304220832948, + "loss": 0.1069, + "step": 47475 + }, + { + "epoch": 0.4121144781729325, + "grad_norm": 0.1640625, + "learning_rate": 0.0011305006767393757, + "loss": 0.1338, + "step": 47476 + }, + { + "epoch": 0.4121231586531367, + "grad_norm": 0.482421875, + "learning_rate": 0.0011304709313621018, + "loss": 0.0996, + "step": 47477 + }, + { + "epoch": 0.41213183913334084, + "grad_norm": 0.1845703125, + "learning_rate": 0.0011304411859515053, + "loss": 0.085, + "step": 47478 + }, + { + "epoch": 0.41214051961354503, + "grad_norm": 0.369140625, + "learning_rate": 0.0011304114405076187, + "loss": 0.125, + "step": 47479 + }, + { + "epoch": 0.41214920009374917, + "grad_norm": 0.162109375, + "learning_rate": 0.0011303816950304743, + "loss": 0.1001, + "step": 47480 + }, + { + "epoch": 0.41215788057395336, + "grad_norm": 0.259765625, + "learning_rate": 0.0011303519495201055, + "loss": 0.0977, + "step": 47481 + }, + { + "epoch": 0.4121665610541575, + "grad_norm": 0.365234375, + "learning_rate": 0.0011303222039765442, + "loss": 0.1338, + "step": 47482 + }, + { + "epoch": 0.4121752415343617, + "grad_norm": 0.5546875, + "learning_rate": 0.0011302924583998228, + "loss": 0.0947, + "step": 47483 + }, + { + "epoch": 0.41218392201456583, + "grad_norm": 0.2041015625, + "learning_rate": 0.001130262712789974, + "loss": 0.1104, + "step": 47484 + }, + { + "epoch": 0.41219260249477, + "grad_norm": 0.416015625, + "learning_rate": 0.0011302329671470303, + "loss": 0.0845, + "step": 47485 + }, + { + "epoch": 0.41220128297497416, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011302032214710244, + "loss": 0.0586, + "step": 47486 + }, + { + "epoch": 0.41220996345517835, + "grad_norm": 0.263671875, + "learning_rate": 0.0011301734757619888, + "loss": 0.0859, + "step": 47487 + }, + { + "epoch": 0.4122186439353825, + "grad_norm": 0.353515625, + "learning_rate": 0.0011301437300199558, + "loss": 0.1108, + "step": 47488 + }, + { + "epoch": 0.4122273244155867, + "grad_norm": 0.1396484375, + "learning_rate": 0.001130113984244958, + "loss": 0.0869, + "step": 47489 + }, + { + "epoch": 0.4122360048957908, + "grad_norm": 0.298828125, + "learning_rate": 0.0011300842384370283, + "loss": 0.1074, + "step": 47490 + }, + { + "epoch": 0.412244685375995, + "grad_norm": 0.287109375, + "learning_rate": 0.0011300544925961986, + "loss": 0.1328, + "step": 47491 + }, + { + "epoch": 0.41225336585619915, + "grad_norm": 0.419921875, + "learning_rate": 0.0011300247467225018, + "loss": 0.1172, + "step": 47492 + }, + { + "epoch": 0.41226204633640334, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011299950008159705, + "loss": 0.1123, + "step": 47493 + }, + { + "epoch": 0.4122707268166075, + "grad_norm": 0.423828125, + "learning_rate": 0.001129965254876637, + "loss": 0.0854, + "step": 47494 + }, + { + "epoch": 0.4122794072968117, + "grad_norm": 0.302734375, + "learning_rate": 0.0011299355089045343, + "loss": 0.0874, + "step": 47495 + }, + { + "epoch": 0.4122880877770158, + "grad_norm": 0.478515625, + "learning_rate": 0.0011299057628996942, + "loss": 0.0938, + "step": 47496 + }, + { + "epoch": 0.41229676825722, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011298760168621498, + "loss": 0.1001, + "step": 47497 + }, + { + "epoch": 0.41230544873742414, + "grad_norm": 0.1513671875, + "learning_rate": 0.0011298462707919335, + "loss": 0.1113, + "step": 47498 + }, + { + "epoch": 0.41231412921762833, + "grad_norm": 0.267578125, + "learning_rate": 0.0011298165246890772, + "loss": 0.1143, + "step": 47499 + }, + { + "epoch": 0.41232280969783247, + "grad_norm": 0.482421875, + "learning_rate": 0.0011297867785536146, + "loss": 0.1455, + "step": 47500 + }, + { + "epoch": 0.41233149017803666, + "grad_norm": 0.89453125, + "learning_rate": 0.0011297570323855774, + "loss": 0.1123, + "step": 47501 + }, + { + "epoch": 0.4123401706582408, + "grad_norm": 0.119140625, + "learning_rate": 0.0011297272861849985, + "loss": 0.0996, + "step": 47502 + }, + { + "epoch": 0.412348851138445, + "grad_norm": 0.322265625, + "learning_rate": 0.00112969753995191, + "loss": 0.0986, + "step": 47503 + }, + { + "epoch": 0.41235753161864913, + "grad_norm": 1.0390625, + "learning_rate": 0.0011296677936863454, + "loss": 0.1016, + "step": 47504 + }, + { + "epoch": 0.4123662120988533, + "grad_norm": 0.142578125, + "learning_rate": 0.0011296380473883357, + "loss": 0.0898, + "step": 47505 + }, + { + "epoch": 0.41237489257905746, + "grad_norm": 0.193359375, + "learning_rate": 0.0011296083010579148, + "loss": 0.0732, + "step": 47506 + }, + { + "epoch": 0.41238357305926165, + "grad_norm": 0.6640625, + "learning_rate": 0.0011295785546951147, + "loss": 0.1172, + "step": 47507 + }, + { + "epoch": 0.4123922535394658, + "grad_norm": 0.1474609375, + "learning_rate": 0.0011295488082999677, + "loss": 0.125, + "step": 47508 + }, + { + "epoch": 0.41240093401967, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011295190618725065, + "loss": 0.1006, + "step": 47509 + }, + { + "epoch": 0.4124096144998741, + "grad_norm": 0.408203125, + "learning_rate": 0.0011294893154127636, + "loss": 0.0723, + "step": 47510 + }, + { + "epoch": 0.4124182949800783, + "grad_norm": 0.20703125, + "learning_rate": 0.001129459568920772, + "loss": 0.0635, + "step": 47511 + }, + { + "epoch": 0.41242697546028245, + "grad_norm": 0.146484375, + "learning_rate": 0.0011294298223965638, + "loss": 0.0947, + "step": 47512 + }, + { + "epoch": 0.41243565594048665, + "grad_norm": 0.33984375, + "learning_rate": 0.0011294000758401714, + "loss": 0.1387, + "step": 47513 + }, + { + "epoch": 0.4124443364206908, + "grad_norm": 0.0859375, + "learning_rate": 0.0011293703292516277, + "loss": 0.0742, + "step": 47514 + }, + { + "epoch": 0.412453016900895, + "grad_norm": 0.62890625, + "learning_rate": 0.0011293405826309651, + "loss": 0.0698, + "step": 47515 + }, + { + "epoch": 0.4124616973810991, + "grad_norm": 0.734375, + "learning_rate": 0.001129310835978216, + "loss": 0.1484, + "step": 47516 + }, + { + "epoch": 0.4124703778613033, + "grad_norm": 0.59375, + "learning_rate": 0.001129281089293413, + "loss": 0.064, + "step": 47517 + }, + { + "epoch": 0.41247905834150744, + "grad_norm": 0.2080078125, + "learning_rate": 0.0011292513425765881, + "loss": 0.123, + "step": 47518 + }, + { + "epoch": 0.41248773882171164, + "grad_norm": 0.408203125, + "learning_rate": 0.0011292215958277753, + "loss": 0.1221, + "step": 47519 + }, + { + "epoch": 0.4124964193019158, + "grad_norm": 0.1328125, + "learning_rate": 0.0011291918490470058, + "loss": 0.0854, + "step": 47520 + }, + { + "epoch": 0.41250509978211997, + "grad_norm": 0.1240234375, + "learning_rate": 0.0011291621022343123, + "loss": 0.0757, + "step": 47521 + }, + { + "epoch": 0.4125137802623241, + "grad_norm": 0.07373046875, + "learning_rate": 0.0011291323553897282, + "loss": 0.0986, + "step": 47522 + }, + { + "epoch": 0.4125224607425283, + "grad_norm": 0.51171875, + "learning_rate": 0.0011291026085132848, + "loss": 0.1016, + "step": 47523 + }, + { + "epoch": 0.41253114122273243, + "grad_norm": 0.20703125, + "learning_rate": 0.0011290728616050155, + "loss": 0.1006, + "step": 47524 + }, + { + "epoch": 0.4125398217029366, + "grad_norm": 0.71875, + "learning_rate": 0.0011290431146649524, + "loss": 0.1099, + "step": 47525 + }, + { + "epoch": 0.41254850218314076, + "grad_norm": 0.1669921875, + "learning_rate": 0.001129013367693128, + "loss": 0.1094, + "step": 47526 + }, + { + "epoch": 0.41255718266334496, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011289836206895753, + "loss": 0.0811, + "step": 47527 + }, + { + "epoch": 0.4125658631435491, + "grad_norm": 0.1123046875, + "learning_rate": 0.0011289538736543266, + "loss": 0.0703, + "step": 47528 + }, + { + "epoch": 0.4125745436237533, + "grad_norm": 0.08984375, + "learning_rate": 0.001128924126587414, + "loss": 0.083, + "step": 47529 + }, + { + "epoch": 0.4125832241039574, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011288943794888712, + "loss": 0.1089, + "step": 47530 + }, + { + "epoch": 0.4125919045841616, + "grad_norm": 0.498046875, + "learning_rate": 0.001128864632358729, + "loss": 0.125, + "step": 47531 + }, + { + "epoch": 0.41260058506436575, + "grad_norm": 0.111328125, + "learning_rate": 0.0011288348851970216, + "loss": 0.1216, + "step": 47532 + }, + { + "epoch": 0.41260926554456995, + "grad_norm": 0.080078125, + "learning_rate": 0.0011288051380037806, + "loss": 0.0859, + "step": 47533 + }, + { + "epoch": 0.4126179460247741, + "grad_norm": 0.1171875, + "learning_rate": 0.0011287753907790385, + "loss": 0.0933, + "step": 47534 + }, + { + "epoch": 0.4126266265049783, + "grad_norm": 0.267578125, + "learning_rate": 0.0011287456435228283, + "loss": 0.0957, + "step": 47535 + }, + { + "epoch": 0.4126353069851824, + "grad_norm": 0.07373046875, + "learning_rate": 0.0011287158962351822, + "loss": 0.0737, + "step": 47536 + }, + { + "epoch": 0.4126439874653866, + "grad_norm": 0.21484375, + "learning_rate": 0.001128686148916133, + "loss": 0.1221, + "step": 47537 + }, + { + "epoch": 0.41265266794559075, + "grad_norm": 0.59375, + "learning_rate": 0.0011286564015657126, + "loss": 0.1079, + "step": 47538 + }, + { + "epoch": 0.41266134842579494, + "grad_norm": 0.412109375, + "learning_rate": 0.0011286266541839542, + "loss": 0.0898, + "step": 47539 + }, + { + "epoch": 0.4126700289059991, + "grad_norm": 0.64453125, + "learning_rate": 0.0011285969067708903, + "loss": 0.1045, + "step": 47540 + }, + { + "epoch": 0.41267870938620327, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011285671593265533, + "loss": 0.0796, + "step": 47541 + }, + { + "epoch": 0.4126873898664074, + "grad_norm": 0.365234375, + "learning_rate": 0.0011285374118509754, + "loss": 0.1279, + "step": 47542 + }, + { + "epoch": 0.4126960703466116, + "grad_norm": 0.08935546875, + "learning_rate": 0.0011285076643441896, + "loss": 0.0801, + "step": 47543 + }, + { + "epoch": 0.41270475082681574, + "grad_norm": 0.1923828125, + "learning_rate": 0.001128477916806228, + "loss": 0.1396, + "step": 47544 + }, + { + "epoch": 0.41271343130701993, + "grad_norm": 0.6171875, + "learning_rate": 0.0011284481692371239, + "loss": 0.0967, + "step": 47545 + }, + { + "epoch": 0.41272211178722407, + "grad_norm": 0.37109375, + "learning_rate": 0.0011284184216369087, + "loss": 0.1416, + "step": 47546 + }, + { + "epoch": 0.41273079226742826, + "grad_norm": 0.4375, + "learning_rate": 0.001128388674005616, + "loss": 0.064, + "step": 47547 + }, + { + "epoch": 0.4127394727476324, + "grad_norm": 0.27734375, + "learning_rate": 0.0011283589263432776, + "loss": 0.1133, + "step": 47548 + }, + { + "epoch": 0.4127481532278366, + "grad_norm": 0.09326171875, + "learning_rate": 0.0011283291786499264, + "loss": 0.1235, + "step": 47549 + }, + { + "epoch": 0.4127568337080407, + "grad_norm": 0.2255859375, + "learning_rate": 0.001128299430925595, + "loss": 0.0825, + "step": 47550 + }, + { + "epoch": 0.4127655141882449, + "grad_norm": 0.171875, + "learning_rate": 0.0011282696831703153, + "loss": 0.0947, + "step": 47551 + }, + { + "epoch": 0.41277419466844906, + "grad_norm": 0.2294921875, + "learning_rate": 0.0011282399353841211, + "loss": 0.1143, + "step": 47552 + }, + { + "epoch": 0.41278287514865325, + "grad_norm": 0.2109375, + "learning_rate": 0.0011282101875670434, + "loss": 0.0703, + "step": 47553 + }, + { + "epoch": 0.4127915556288574, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011281804397191158, + "loss": 0.0977, + "step": 47554 + }, + { + "epoch": 0.4128002361090616, + "grad_norm": 0.462890625, + "learning_rate": 0.0011281506918403704, + "loss": 0.0928, + "step": 47555 + }, + { + "epoch": 0.4128089165892657, + "grad_norm": 0.296875, + "learning_rate": 0.0011281209439308398, + "loss": 0.0732, + "step": 47556 + }, + { + "epoch": 0.4128175970694699, + "grad_norm": 0.306640625, + "learning_rate": 0.0011280911959905564, + "loss": 0.0801, + "step": 47557 + }, + { + "epoch": 0.41282627754967405, + "grad_norm": 0.0771484375, + "learning_rate": 0.001128061448019553, + "loss": 0.0811, + "step": 47558 + }, + { + "epoch": 0.4128349580298782, + "grad_norm": 0.1201171875, + "learning_rate": 0.0011280317000178623, + "loss": 0.0923, + "step": 47559 + }, + { + "epoch": 0.4128436385100824, + "grad_norm": 0.310546875, + "learning_rate": 0.0011280019519855165, + "loss": 0.1113, + "step": 47560 + }, + { + "epoch": 0.4128523189902865, + "grad_norm": 0.4140625, + "learning_rate": 0.0011279722039225482, + "loss": 0.1484, + "step": 47561 + }, + { + "epoch": 0.4128609994704907, + "grad_norm": 0.5859375, + "learning_rate": 0.0011279424558289894, + "loss": 0.082, + "step": 47562 + }, + { + "epoch": 0.41286967995069485, + "grad_norm": 0.3671875, + "learning_rate": 0.0011279127077048737, + "loss": 0.0781, + "step": 47563 + }, + { + "epoch": 0.41287836043089904, + "grad_norm": 0.2314453125, + "learning_rate": 0.0011278829595502329, + "loss": 0.0947, + "step": 47564 + }, + { + "epoch": 0.4128870409111032, + "grad_norm": 0.95703125, + "learning_rate": 0.0011278532113650995, + "loss": 0.125, + "step": 47565 + }, + { + "epoch": 0.41289572139130737, + "grad_norm": 0.1396484375, + "learning_rate": 0.0011278234631495064, + "loss": 0.0664, + "step": 47566 + }, + { + "epoch": 0.4129044018715115, + "grad_norm": 1.875, + "learning_rate": 0.001127793714903486, + "loss": 0.1514, + "step": 47567 + }, + { + "epoch": 0.4129130823517157, + "grad_norm": 0.267578125, + "learning_rate": 0.0011277639666270709, + "loss": 0.0771, + "step": 47568 + }, + { + "epoch": 0.41292176283191984, + "grad_norm": 0.18359375, + "learning_rate": 0.0011277342183202935, + "loss": 0.0776, + "step": 47569 + }, + { + "epoch": 0.41293044331212403, + "grad_norm": 0.376953125, + "learning_rate": 0.0011277044699831863, + "loss": 0.1094, + "step": 47570 + }, + { + "epoch": 0.41293912379232817, + "grad_norm": 0.080078125, + "learning_rate": 0.001127674721615782, + "loss": 0.0747, + "step": 47571 + }, + { + "epoch": 0.41294780427253236, + "grad_norm": 0.44921875, + "learning_rate": 0.001127644973218113, + "loss": 0.1152, + "step": 47572 + }, + { + "epoch": 0.4129564847527365, + "grad_norm": 0.1005859375, + "learning_rate": 0.001127615224790212, + "loss": 0.1182, + "step": 47573 + }, + { + "epoch": 0.4129651652329407, + "grad_norm": 0.6328125, + "learning_rate": 0.001127585476332111, + "loss": 0.0874, + "step": 47574 + }, + { + "epoch": 0.4129738457131448, + "grad_norm": 1.015625, + "learning_rate": 0.0011275557278438435, + "loss": 0.1611, + "step": 47575 + }, + { + "epoch": 0.412982526193349, + "grad_norm": 0.11669921875, + "learning_rate": 0.0011275259793254412, + "loss": 0.0869, + "step": 47576 + }, + { + "epoch": 0.41299120667355316, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011274962307769368, + "loss": 0.0894, + "step": 47577 + }, + { + "epoch": 0.41299988715375735, + "grad_norm": 0.08251953125, + "learning_rate": 0.0011274664821983632, + "loss": 0.1074, + "step": 47578 + }, + { + "epoch": 0.4130085676339615, + "grad_norm": 0.279296875, + "learning_rate": 0.0011274367335897522, + "loss": 0.1182, + "step": 47579 + }, + { + "epoch": 0.4130172481141657, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011274069849511372, + "loss": 0.1211, + "step": 47580 + }, + { + "epoch": 0.4130259285943698, + "grad_norm": 0.1162109375, + "learning_rate": 0.0011273772362825503, + "loss": 0.0771, + "step": 47581 + }, + { + "epoch": 0.413034609074574, + "grad_norm": 0.435546875, + "learning_rate": 0.0011273474875840243, + "loss": 0.1113, + "step": 47582 + }, + { + "epoch": 0.41304328955477815, + "grad_norm": 0.201171875, + "learning_rate": 0.001127317738855591, + "loss": 0.0776, + "step": 47583 + }, + { + "epoch": 0.41305197003498234, + "grad_norm": 0.26953125, + "learning_rate": 0.0011272879900972835, + "loss": 0.1206, + "step": 47584 + }, + { + "epoch": 0.4130606505151865, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011272582413091343, + "loss": 0.1191, + "step": 47585 + }, + { + "epoch": 0.41306933099539067, + "grad_norm": 0.1298828125, + "learning_rate": 0.001127228492491176, + "loss": 0.1133, + "step": 47586 + }, + { + "epoch": 0.4130780114755948, + "grad_norm": 0.10498046875, + "learning_rate": 0.001127198743643441, + "loss": 0.1123, + "step": 47587 + }, + { + "epoch": 0.413086691955799, + "grad_norm": 0.251953125, + "learning_rate": 0.0011271689947659619, + "loss": 0.1167, + "step": 47588 + }, + { + "epoch": 0.41309537243600314, + "grad_norm": 0.1279296875, + "learning_rate": 0.001127139245858771, + "loss": 0.0894, + "step": 47589 + }, + { + "epoch": 0.41310405291620733, + "grad_norm": 0.6953125, + "learning_rate": 0.0011271094969219013, + "loss": 0.1514, + "step": 47590 + }, + { + "epoch": 0.41311273339641147, + "grad_norm": 0.146484375, + "learning_rate": 0.0011270797479553851, + "loss": 0.0903, + "step": 47591 + }, + { + "epoch": 0.41312141387661566, + "grad_norm": 0.07177734375, + "learning_rate": 0.0011270499989592547, + "loss": 0.085, + "step": 47592 + }, + { + "epoch": 0.4131300943568198, + "grad_norm": 0.322265625, + "learning_rate": 0.0011270202499335429, + "loss": 0.0669, + "step": 47593 + }, + { + "epoch": 0.413138774837024, + "grad_norm": 0.2294921875, + "learning_rate": 0.0011269905008782822, + "loss": 0.0996, + "step": 47594 + }, + { + "epoch": 0.41314745531722813, + "grad_norm": 0.107421875, + "learning_rate": 0.001126960751793505, + "loss": 0.1196, + "step": 47595 + }, + { + "epoch": 0.4131561357974323, + "grad_norm": 0.08544921875, + "learning_rate": 0.0011269310026792438, + "loss": 0.082, + "step": 47596 + }, + { + "epoch": 0.41316481627763646, + "grad_norm": 0.4296875, + "learning_rate": 0.0011269012535355314, + "loss": 0.0947, + "step": 47597 + }, + { + "epoch": 0.41317349675784065, + "grad_norm": 0.8125, + "learning_rate": 0.0011268715043624006, + "loss": 0.0977, + "step": 47598 + }, + { + "epoch": 0.4131821772380448, + "grad_norm": 0.08056640625, + "learning_rate": 0.0011268417551598831, + "loss": 0.1094, + "step": 47599 + }, + { + "epoch": 0.413190857718249, + "grad_norm": 0.2177734375, + "learning_rate": 0.001126812005928012, + "loss": 0.0923, + "step": 47600 + }, + { + "epoch": 0.4131995381984531, + "grad_norm": 0.1591796875, + "learning_rate": 0.0011267822566668193, + "loss": 0.0708, + "step": 47601 + }, + { + "epoch": 0.4132082186786573, + "grad_norm": 0.32421875, + "learning_rate": 0.0011267525073763386, + "loss": 0.0996, + "step": 47602 + }, + { + "epoch": 0.41321689915886145, + "grad_norm": 0.435546875, + "learning_rate": 0.0011267227580566011, + "loss": 0.1826, + "step": 47603 + }, + { + "epoch": 0.41322557963906564, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011266930087076403, + "loss": 0.0762, + "step": 47604 + }, + { + "epoch": 0.4132342601192698, + "grad_norm": 0.5, + "learning_rate": 0.0011266632593294882, + "loss": 0.0977, + "step": 47605 + }, + { + "epoch": 0.413242940599474, + "grad_norm": 0.150390625, + "learning_rate": 0.0011266335099221778, + "loss": 0.0654, + "step": 47606 + }, + { + "epoch": 0.4132516210796781, + "grad_norm": 0.30859375, + "learning_rate": 0.0011266037604857415, + "loss": 0.0967, + "step": 47607 + }, + { + "epoch": 0.4132603015598823, + "grad_norm": 0.482421875, + "learning_rate": 0.0011265740110202116, + "loss": 0.124, + "step": 47608 + }, + { + "epoch": 0.41326898204008644, + "grad_norm": 0.439453125, + "learning_rate": 0.0011265442615256208, + "loss": 0.1289, + "step": 47609 + }, + { + "epoch": 0.41327766252029063, + "grad_norm": 0.134765625, + "learning_rate": 0.0011265145120020015, + "loss": 0.0908, + "step": 47610 + }, + { + "epoch": 0.41328634300049477, + "grad_norm": 0.703125, + "learning_rate": 0.0011264847624493863, + "loss": 0.1045, + "step": 47611 + }, + { + "epoch": 0.41329502348069896, + "grad_norm": 0.1767578125, + "learning_rate": 0.001126455012867808, + "loss": 0.1035, + "step": 47612 + }, + { + "epoch": 0.4133037039609031, + "grad_norm": 0.197265625, + "learning_rate": 0.0011264252632572985, + "loss": 0.1494, + "step": 47613 + }, + { + "epoch": 0.4133123844411073, + "grad_norm": 0.11376953125, + "learning_rate": 0.001126395513617891, + "loss": 0.085, + "step": 47614 + }, + { + "epoch": 0.41332106492131143, + "grad_norm": 0.287109375, + "learning_rate": 0.0011263657639496176, + "loss": 0.084, + "step": 47615 + }, + { + "epoch": 0.4133297454015156, + "grad_norm": 0.232421875, + "learning_rate": 0.0011263360142525112, + "loss": 0.1182, + "step": 47616 + }, + { + "epoch": 0.41333842588171976, + "grad_norm": 0.470703125, + "learning_rate": 0.0011263062645266043, + "loss": 0.0786, + "step": 47617 + }, + { + "epoch": 0.41334710636192395, + "grad_norm": 0.486328125, + "learning_rate": 0.0011262765147719287, + "loss": 0.082, + "step": 47618 + }, + { + "epoch": 0.4133557868421281, + "grad_norm": 0.30078125, + "learning_rate": 0.0011262467649885179, + "loss": 0.0806, + "step": 47619 + }, + { + "epoch": 0.4133644673223323, + "grad_norm": 0.06298828125, + "learning_rate": 0.001126217015176404, + "loss": 0.0732, + "step": 47620 + }, + { + "epoch": 0.4133731478025364, + "grad_norm": 0.31640625, + "learning_rate": 0.0011261872653356196, + "loss": 0.1162, + "step": 47621 + }, + { + "epoch": 0.4133818282827406, + "grad_norm": 0.2099609375, + "learning_rate": 0.0011261575154661969, + "loss": 0.0654, + "step": 47622 + }, + { + "epoch": 0.41339050876294475, + "grad_norm": 0.58984375, + "learning_rate": 0.0011261277655681687, + "loss": 0.1035, + "step": 47623 + }, + { + "epoch": 0.41339918924314895, + "grad_norm": 0.39453125, + "learning_rate": 0.0011260980156415682, + "loss": 0.0938, + "step": 47624 + }, + { + "epoch": 0.4134078697233531, + "grad_norm": 0.11083984375, + "learning_rate": 0.0011260682656864267, + "loss": 0.0918, + "step": 47625 + }, + { + "epoch": 0.4134165502035573, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011260385157027776, + "loss": 0.0728, + "step": 47626 + }, + { + "epoch": 0.4134252306837614, + "grad_norm": 0.37890625, + "learning_rate": 0.001126008765690653, + "loss": 0.0967, + "step": 47627 + }, + { + "epoch": 0.4134339111639656, + "grad_norm": 0.439453125, + "learning_rate": 0.001125979015650086, + "loss": 0.1055, + "step": 47628 + }, + { + "epoch": 0.41344259164416974, + "grad_norm": 0.6796875, + "learning_rate": 0.0011259492655811085, + "loss": 0.1182, + "step": 47629 + }, + { + "epoch": 0.41345127212437394, + "grad_norm": 0.6953125, + "learning_rate": 0.0011259195154837533, + "loss": 0.1406, + "step": 47630 + }, + { + "epoch": 0.4134599526045781, + "grad_norm": 0.07958984375, + "learning_rate": 0.0011258897653580527, + "loss": 0.0972, + "step": 47631 + }, + { + "epoch": 0.41346863308478227, + "grad_norm": 0.16015625, + "learning_rate": 0.0011258600152040396, + "loss": 0.1104, + "step": 47632 + }, + { + "epoch": 0.4134773135649864, + "grad_norm": 0.5859375, + "learning_rate": 0.0011258302650217463, + "loss": 0.104, + "step": 47633 + }, + { + "epoch": 0.4134859940451906, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011258005148112053, + "loss": 0.1113, + "step": 47634 + }, + { + "epoch": 0.41349467452539473, + "grad_norm": 0.115234375, + "learning_rate": 0.0011257707645724493, + "loss": 0.0752, + "step": 47635 + }, + { + "epoch": 0.4135033550055989, + "grad_norm": 0.37109375, + "learning_rate": 0.0011257410143055112, + "loss": 0.1162, + "step": 47636 + }, + { + "epoch": 0.41351203548580306, + "grad_norm": 0.2265625, + "learning_rate": 0.0011257112640104228, + "loss": 0.1177, + "step": 47637 + }, + { + "epoch": 0.41352071596600726, + "grad_norm": 0.263671875, + "learning_rate": 0.001125681513687217, + "loss": 0.0981, + "step": 47638 + }, + { + "epoch": 0.4135293964462114, + "grad_norm": 0.37890625, + "learning_rate": 0.001125651763335926, + "loss": 0.1216, + "step": 47639 + }, + { + "epoch": 0.4135380769264156, + "grad_norm": 0.10302734375, + "learning_rate": 0.001125622012956583, + "loss": 0.0908, + "step": 47640 + }, + { + "epoch": 0.4135467574066197, + "grad_norm": 0.263671875, + "learning_rate": 0.0011255922625492201, + "loss": 0.0854, + "step": 47641 + }, + { + "epoch": 0.4135554378868239, + "grad_norm": 0.2294921875, + "learning_rate": 0.0011255625121138696, + "loss": 0.1123, + "step": 47642 + }, + { + "epoch": 0.41356411836702806, + "grad_norm": 0.16015625, + "learning_rate": 0.0011255327616505647, + "loss": 0.1562, + "step": 47643 + }, + { + "epoch": 0.41357279884723225, + "grad_norm": 0.15234375, + "learning_rate": 0.0011255030111593373, + "loss": 0.0815, + "step": 47644 + }, + { + "epoch": 0.4135814793274364, + "grad_norm": 0.255859375, + "learning_rate": 0.0011254732606402202, + "loss": 0.0903, + "step": 47645 + }, + { + "epoch": 0.4135901598076406, + "grad_norm": 0.0830078125, + "learning_rate": 0.0011254435100932461, + "loss": 0.0811, + "step": 47646 + }, + { + "epoch": 0.4135988402878447, + "grad_norm": 0.08544921875, + "learning_rate": 0.0011254137595184473, + "loss": 0.0996, + "step": 47647 + }, + { + "epoch": 0.4136075207680489, + "grad_norm": 0.373046875, + "learning_rate": 0.0011253840089158562, + "loss": 0.1465, + "step": 47648 + }, + { + "epoch": 0.41361620124825305, + "grad_norm": 0.59375, + "learning_rate": 0.0011253542582855057, + "loss": 0.0635, + "step": 47649 + }, + { + "epoch": 0.41362488172845724, + "grad_norm": 0.443359375, + "learning_rate": 0.0011253245076274282, + "loss": 0.0996, + "step": 47650 + }, + { + "epoch": 0.4136335622086614, + "grad_norm": 0.09423828125, + "learning_rate": 0.001125294756941656, + "loss": 0.1543, + "step": 47651 + }, + { + "epoch": 0.41364224268886557, + "grad_norm": 0.345703125, + "learning_rate": 0.0011252650062282222, + "loss": 0.1523, + "step": 47652 + }, + { + "epoch": 0.4136509231690697, + "grad_norm": 0.318359375, + "learning_rate": 0.0011252352554871584, + "loss": 0.0859, + "step": 47653 + }, + { + "epoch": 0.4136596036492739, + "grad_norm": 0.56640625, + "learning_rate": 0.0011252055047184981, + "loss": 0.1299, + "step": 47654 + }, + { + "epoch": 0.41366828412947804, + "grad_norm": 0.640625, + "learning_rate": 0.0011251757539222735, + "loss": 0.1104, + "step": 47655 + }, + { + "epoch": 0.41367696460968223, + "grad_norm": 0.2138671875, + "learning_rate": 0.001125146003098517, + "loss": 0.1191, + "step": 47656 + }, + { + "epoch": 0.41368564508988637, + "grad_norm": 0.494140625, + "learning_rate": 0.001125116252247261, + "loss": 0.083, + "step": 47657 + }, + { + "epoch": 0.41369432557009056, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011250865013685385, + "loss": 0.106, + "step": 47658 + }, + { + "epoch": 0.4137030060502947, + "grad_norm": 0.259765625, + "learning_rate": 0.0011250567504623818, + "loss": 0.0566, + "step": 47659 + }, + { + "epoch": 0.4137116865304989, + "grad_norm": 0.1298828125, + "learning_rate": 0.001125026999528823, + "loss": 0.0913, + "step": 47660 + }, + { + "epoch": 0.413720367010703, + "grad_norm": 0.498046875, + "learning_rate": 0.0011249972485678954, + "loss": 0.1138, + "step": 47661 + }, + { + "epoch": 0.4137290474909072, + "grad_norm": 0.16796875, + "learning_rate": 0.001124967497579631, + "loss": 0.0854, + "step": 47662 + }, + { + "epoch": 0.41373772797111136, + "grad_norm": 0.43359375, + "learning_rate": 0.0011249377465640626, + "loss": 0.084, + "step": 47663 + }, + { + "epoch": 0.41374640845131555, + "grad_norm": 0.251953125, + "learning_rate": 0.0011249079955212228, + "loss": 0.0732, + "step": 47664 + }, + { + "epoch": 0.4137550889315197, + "grad_norm": 0.671875, + "learning_rate": 0.0011248782444511437, + "loss": 0.1143, + "step": 47665 + }, + { + "epoch": 0.4137637694117239, + "grad_norm": 0.58203125, + "learning_rate": 0.0011248484933538585, + "loss": 0.0879, + "step": 47666 + }, + { + "epoch": 0.413772449891928, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011248187422293993, + "loss": 0.0693, + "step": 47667 + }, + { + "epoch": 0.4137811303721322, + "grad_norm": 0.6171875, + "learning_rate": 0.0011247889910777989, + "loss": 0.1021, + "step": 47668 + }, + { + "epoch": 0.41378981085233635, + "grad_norm": 0.09326171875, + "learning_rate": 0.0011247592398990892, + "loss": 0.083, + "step": 47669 + }, + { + "epoch": 0.41379849133254054, + "grad_norm": 0.271484375, + "learning_rate": 0.0011247294886933034, + "loss": 0.1162, + "step": 47670 + }, + { + "epoch": 0.4138071718127447, + "grad_norm": 0.61328125, + "learning_rate": 0.0011246997374604732, + "loss": 0.1143, + "step": 47671 + }, + { + "epoch": 0.41381585229294887, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011246699862006325, + "loss": 0.0869, + "step": 47672 + }, + { + "epoch": 0.413824532773153, + "grad_norm": 0.1162109375, + "learning_rate": 0.0011246402349138127, + "loss": 0.0879, + "step": 47673 + }, + { + "epoch": 0.4138332132533572, + "grad_norm": 0.1884765625, + "learning_rate": 0.0011246104836000466, + "loss": 0.0967, + "step": 47674 + }, + { + "epoch": 0.41384189373356134, + "grad_norm": 0.08056640625, + "learning_rate": 0.001124580732259367, + "loss": 0.0898, + "step": 47675 + }, + { + "epoch": 0.41385057421376553, + "grad_norm": 0.14453125, + "learning_rate": 0.0011245509808918066, + "loss": 0.1074, + "step": 47676 + }, + { + "epoch": 0.41385925469396967, + "grad_norm": 0.150390625, + "learning_rate": 0.0011245212294973973, + "loss": 0.1201, + "step": 47677 + }, + { + "epoch": 0.41386793517417386, + "grad_norm": 0.345703125, + "learning_rate": 0.001124491478076172, + "loss": 0.1167, + "step": 47678 + }, + { + "epoch": 0.413876615654378, + "grad_norm": 0.83203125, + "learning_rate": 0.0011244617266281634, + "loss": 0.1162, + "step": 47679 + }, + { + "epoch": 0.4138852961345822, + "grad_norm": 0.40234375, + "learning_rate": 0.0011244319751534035, + "loss": 0.0811, + "step": 47680 + }, + { + "epoch": 0.41389397661478633, + "grad_norm": 0.34375, + "learning_rate": 0.0011244022236519251, + "loss": 0.1504, + "step": 47681 + }, + { + "epoch": 0.4139026570949905, + "grad_norm": 0.244140625, + "learning_rate": 0.001124372472123761, + "loss": 0.1016, + "step": 47682 + }, + { + "epoch": 0.41391133757519466, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011243427205689435, + "loss": 0.1099, + "step": 47683 + }, + { + "epoch": 0.4139200180553988, + "grad_norm": 0.357421875, + "learning_rate": 0.0011243129689875052, + "loss": 0.084, + "step": 47684 + }, + { + "epoch": 0.413928698535603, + "grad_norm": 0.34375, + "learning_rate": 0.0011242832173794786, + "loss": 0.0923, + "step": 47685 + }, + { + "epoch": 0.4139373790158071, + "grad_norm": 0.55859375, + "learning_rate": 0.0011242534657448962, + "loss": 0.0957, + "step": 47686 + }, + { + "epoch": 0.4139460594960113, + "grad_norm": 0.310546875, + "learning_rate": 0.0011242237140837903, + "loss": 0.1089, + "step": 47687 + }, + { + "epoch": 0.41395473997621546, + "grad_norm": 0.45703125, + "learning_rate": 0.0011241939623961943, + "loss": 0.1123, + "step": 47688 + }, + { + "epoch": 0.41396342045641965, + "grad_norm": 0.2421875, + "learning_rate": 0.00112416421068214, + "loss": 0.1162, + "step": 47689 + }, + { + "epoch": 0.4139721009366238, + "grad_norm": 0.56640625, + "learning_rate": 0.0011241344589416596, + "loss": 0.1377, + "step": 47690 + }, + { + "epoch": 0.413980781416828, + "grad_norm": 0.298828125, + "learning_rate": 0.0011241047071747863, + "loss": 0.1309, + "step": 47691 + }, + { + "epoch": 0.4139894618970321, + "grad_norm": 0.330078125, + "learning_rate": 0.0011240749553815526, + "loss": 0.0933, + "step": 47692 + }, + { + "epoch": 0.4139981423772363, + "grad_norm": 0.267578125, + "learning_rate": 0.001124045203561991, + "loss": 0.0903, + "step": 47693 + }, + { + "epoch": 0.41400682285744045, + "grad_norm": 0.11328125, + "learning_rate": 0.0011240154517161338, + "loss": 0.1221, + "step": 47694 + }, + { + "epoch": 0.41401550333764464, + "grad_norm": 0.2265625, + "learning_rate": 0.0011239856998440136, + "loss": 0.083, + "step": 47695 + }, + { + "epoch": 0.4140241838178488, + "grad_norm": 0.1181640625, + "learning_rate": 0.0011239559479456632, + "loss": 0.0874, + "step": 47696 + }, + { + "epoch": 0.41403286429805297, + "grad_norm": 0.08935546875, + "learning_rate": 0.0011239261960211146, + "loss": 0.0791, + "step": 47697 + }, + { + "epoch": 0.4140415447782571, + "grad_norm": 1.1875, + "learning_rate": 0.001123896444070401, + "loss": 0.1143, + "step": 47698 + }, + { + "epoch": 0.4140502252584613, + "grad_norm": 0.12255859375, + "learning_rate": 0.0011238666920935544, + "loss": 0.103, + "step": 47699 + }, + { + "epoch": 0.41405890573866544, + "grad_norm": 0.1123046875, + "learning_rate": 0.0011238369400906076, + "loss": 0.1123, + "step": 47700 + }, + { + "epoch": 0.41406758621886963, + "grad_norm": 0.1552734375, + "learning_rate": 0.001123807188061593, + "loss": 0.0845, + "step": 47701 + }, + { + "epoch": 0.41407626669907377, + "grad_norm": 0.07763671875, + "learning_rate": 0.0011237774360065434, + "loss": 0.0947, + "step": 47702 + }, + { + "epoch": 0.41408494717927796, + "grad_norm": 0.22265625, + "learning_rate": 0.0011237476839254911, + "loss": 0.1123, + "step": 47703 + }, + { + "epoch": 0.4140936276594821, + "grad_norm": 0.1162109375, + "learning_rate": 0.0011237179318184685, + "loss": 0.0898, + "step": 47704 + }, + { + "epoch": 0.4141023081396863, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011236881796855087, + "loss": 0.1128, + "step": 47705 + }, + { + "epoch": 0.41411098861989043, + "grad_norm": 0.10009765625, + "learning_rate": 0.0011236584275266437, + "loss": 0.1128, + "step": 47706 + }, + { + "epoch": 0.4141196691000946, + "grad_norm": 0.55859375, + "learning_rate": 0.0011236286753419063, + "loss": 0.0752, + "step": 47707 + }, + { + "epoch": 0.41412834958029876, + "grad_norm": 0.33203125, + "learning_rate": 0.001123598923131329, + "loss": 0.123, + "step": 47708 + }, + { + "epoch": 0.41413703006050295, + "grad_norm": 0.10791015625, + "learning_rate": 0.0011235691708949436, + "loss": 0.1123, + "step": 47709 + }, + { + "epoch": 0.4141457105407071, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011235394186327836, + "loss": 0.0942, + "step": 47710 + }, + { + "epoch": 0.4141543910209113, + "grad_norm": 0.1015625, + "learning_rate": 0.0011235096663448816, + "loss": 0.1167, + "step": 47711 + }, + { + "epoch": 0.4141630715011154, + "grad_norm": 0.515625, + "learning_rate": 0.0011234799140312697, + "loss": 0.1006, + "step": 47712 + }, + { + "epoch": 0.4141717519813196, + "grad_norm": 0.486328125, + "learning_rate": 0.0011234501616919801, + "loss": 0.0903, + "step": 47713 + }, + { + "epoch": 0.41418043246152375, + "grad_norm": 0.419921875, + "learning_rate": 0.0011234204093270464, + "loss": 0.1191, + "step": 47714 + }, + { + "epoch": 0.41418911294172794, + "grad_norm": 0.298828125, + "learning_rate": 0.0011233906569365, + "loss": 0.0938, + "step": 47715 + }, + { + "epoch": 0.4141977934219321, + "grad_norm": 0.71484375, + "learning_rate": 0.0011233609045203743, + "loss": 0.0981, + "step": 47716 + }, + { + "epoch": 0.4142064739021363, + "grad_norm": 0.1279296875, + "learning_rate": 0.001123331152078701, + "loss": 0.0737, + "step": 47717 + }, + { + "epoch": 0.4142151543823404, + "grad_norm": 0.83203125, + "learning_rate": 0.0011233013996115135, + "loss": 0.1152, + "step": 47718 + }, + { + "epoch": 0.4142238348625446, + "grad_norm": 0.5703125, + "learning_rate": 0.0011232716471188437, + "loss": 0.0845, + "step": 47719 + }, + { + "epoch": 0.41423251534274874, + "grad_norm": 0.1474609375, + "learning_rate": 0.0011232418946007245, + "loss": 0.0986, + "step": 47720 + }, + { + "epoch": 0.41424119582295293, + "grad_norm": 0.33203125, + "learning_rate": 0.0011232121420571884, + "loss": 0.1084, + "step": 47721 + }, + { + "epoch": 0.41424987630315707, + "grad_norm": 0.169921875, + "learning_rate": 0.0011231823894882675, + "loss": 0.0889, + "step": 47722 + }, + { + "epoch": 0.41425855678336126, + "grad_norm": 0.150390625, + "learning_rate": 0.001123152636893995, + "loss": 0.0854, + "step": 47723 + }, + { + "epoch": 0.4142672372635654, + "grad_norm": 0.1083984375, + "learning_rate": 0.0011231228842744033, + "loss": 0.0942, + "step": 47724 + }, + { + "epoch": 0.4142759177437696, + "grad_norm": 0.2060546875, + "learning_rate": 0.0011230931316295243, + "loss": 0.0898, + "step": 47725 + }, + { + "epoch": 0.41428459822397373, + "grad_norm": 0.1767578125, + "learning_rate": 0.001123063378959391, + "loss": 0.1162, + "step": 47726 + }, + { + "epoch": 0.4142932787041779, + "grad_norm": 0.177734375, + "learning_rate": 0.0011230336262640364, + "loss": 0.0947, + "step": 47727 + }, + { + "epoch": 0.41430195918438206, + "grad_norm": 0.33984375, + "learning_rate": 0.0011230038735434921, + "loss": 0.0811, + "step": 47728 + }, + { + "epoch": 0.41431063966458626, + "grad_norm": 0.2177734375, + "learning_rate": 0.0011229741207977916, + "loss": 0.0967, + "step": 47729 + }, + { + "epoch": 0.4143193201447904, + "grad_norm": 3.765625, + "learning_rate": 0.0011229443680269664, + "loss": 0.3203, + "step": 47730 + }, + { + "epoch": 0.4143280006249946, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011229146152310502, + "loss": 0.1011, + "step": 47731 + }, + { + "epoch": 0.4143366811051987, + "grad_norm": 0.1875, + "learning_rate": 0.0011228848624100746, + "loss": 0.0928, + "step": 47732 + }, + { + "epoch": 0.4143453615854029, + "grad_norm": 0.1220703125, + "learning_rate": 0.0011228551095640723, + "loss": 0.0947, + "step": 47733 + }, + { + "epoch": 0.41435404206560705, + "grad_norm": 0.30078125, + "learning_rate": 0.001122825356693076, + "loss": 0.0742, + "step": 47734 + }, + { + "epoch": 0.41436272254581125, + "grad_norm": 0.11572265625, + "learning_rate": 0.0011227956037971182, + "loss": 0.1206, + "step": 47735 + }, + { + "epoch": 0.4143714030260154, + "grad_norm": 0.298828125, + "learning_rate": 0.0011227658508762316, + "loss": 0.0991, + "step": 47736 + }, + { + "epoch": 0.4143800835062196, + "grad_norm": 0.287109375, + "learning_rate": 0.0011227360979304492, + "loss": 0.1035, + "step": 47737 + }, + { + "epoch": 0.4143887639864237, + "grad_norm": 0.0791015625, + "learning_rate": 0.001122706344959802, + "loss": 0.0708, + "step": 47738 + }, + { + "epoch": 0.4143974444666279, + "grad_norm": 0.18359375, + "learning_rate": 0.0011226765919643237, + "loss": 0.0884, + "step": 47739 + }, + { + "epoch": 0.41440612494683204, + "grad_norm": 0.09423828125, + "learning_rate": 0.001122646838944047, + "loss": 0.0776, + "step": 47740 + }, + { + "epoch": 0.41441480542703624, + "grad_norm": 0.5546875, + "learning_rate": 0.0011226170858990039, + "loss": 0.1895, + "step": 47741 + }, + { + "epoch": 0.4144234859072404, + "grad_norm": 0.294921875, + "learning_rate": 0.001122587332829227, + "loss": 0.1299, + "step": 47742 + }, + { + "epoch": 0.41443216638744457, + "grad_norm": 0.65625, + "learning_rate": 0.0011225575797347489, + "loss": 0.0938, + "step": 47743 + }, + { + "epoch": 0.4144408468676487, + "grad_norm": 0.33203125, + "learning_rate": 0.0011225278266156023, + "loss": 0.1035, + "step": 47744 + }, + { + "epoch": 0.4144495273478529, + "grad_norm": 0.345703125, + "learning_rate": 0.0011224980734718195, + "loss": 0.0791, + "step": 47745 + }, + { + "epoch": 0.41445820782805703, + "grad_norm": 0.326171875, + "learning_rate": 0.0011224683203034335, + "loss": 0.126, + "step": 47746 + }, + { + "epoch": 0.4144668883082612, + "grad_norm": 0.353515625, + "learning_rate": 0.001122438567110476, + "loss": 0.0933, + "step": 47747 + }, + { + "epoch": 0.41447556878846536, + "grad_norm": 0.2890625, + "learning_rate": 0.00112240881389298, + "loss": 0.1069, + "step": 47748 + }, + { + "epoch": 0.41448424926866956, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011223790606509783, + "loss": 0.124, + "step": 47749 + }, + { + "epoch": 0.4144929297488737, + "grad_norm": 0.19921875, + "learning_rate": 0.0011223493073845032, + "loss": 0.1133, + "step": 47750 + }, + { + "epoch": 0.4145016102290779, + "grad_norm": 0.2578125, + "learning_rate": 0.001122319554093587, + "loss": 0.1118, + "step": 47751 + }, + { + "epoch": 0.414510290709282, + "grad_norm": 0.09228515625, + "learning_rate": 0.0011222898007782628, + "loss": 0.1074, + "step": 47752 + }, + { + "epoch": 0.4145189711894862, + "grad_norm": 0.326171875, + "learning_rate": 0.0011222600474385628, + "loss": 0.1152, + "step": 47753 + }, + { + "epoch": 0.41452765166969036, + "grad_norm": 0.259765625, + "learning_rate": 0.0011222302940745192, + "loss": 0.1426, + "step": 47754 + }, + { + "epoch": 0.41453633214989455, + "grad_norm": 0.388671875, + "learning_rate": 0.0011222005406861652, + "loss": 0.1152, + "step": 47755 + }, + { + "epoch": 0.4145450126300987, + "grad_norm": 0.3125, + "learning_rate": 0.0011221707872735328, + "loss": 0.1406, + "step": 47756 + }, + { + "epoch": 0.4145536931103029, + "grad_norm": 0.26171875, + "learning_rate": 0.0011221410338366552, + "loss": 0.1621, + "step": 47757 + }, + { + "epoch": 0.414562373590507, + "grad_norm": 0.294921875, + "learning_rate": 0.0011221112803755638, + "loss": 0.1011, + "step": 47758 + }, + { + "epoch": 0.4145710540707112, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011220815268902925, + "loss": 0.1543, + "step": 47759 + }, + { + "epoch": 0.41457973455091535, + "grad_norm": 0.8046875, + "learning_rate": 0.0011220517733808726, + "loss": 0.0908, + "step": 47760 + }, + { + "epoch": 0.41458841503111954, + "grad_norm": 0.396484375, + "learning_rate": 0.0011220220198473376, + "loss": 0.0635, + "step": 47761 + }, + { + "epoch": 0.4145970955113237, + "grad_norm": 0.466796875, + "learning_rate": 0.0011219922662897195, + "loss": 0.1182, + "step": 47762 + }, + { + "epoch": 0.41460577599152787, + "grad_norm": 0.0732421875, + "learning_rate": 0.0011219625127080514, + "loss": 0.083, + "step": 47763 + }, + { + "epoch": 0.414614456471732, + "grad_norm": 0.1259765625, + "learning_rate": 0.001121932759102365, + "loss": 0.0908, + "step": 47764 + }, + { + "epoch": 0.4146231369519362, + "grad_norm": 0.296875, + "learning_rate": 0.001121903005472693, + "loss": 0.1064, + "step": 47765 + }, + { + "epoch": 0.41463181743214034, + "grad_norm": 0.369140625, + "learning_rate": 0.0011218732518190686, + "loss": 0.1133, + "step": 47766 + }, + { + "epoch": 0.41464049791234453, + "grad_norm": 0.23046875, + "learning_rate": 0.0011218434981415239, + "loss": 0.083, + "step": 47767 + }, + { + "epoch": 0.41464917839254867, + "grad_norm": 0.140625, + "learning_rate": 0.0011218137444400915, + "loss": 0.0957, + "step": 47768 + }, + { + "epoch": 0.41465785887275286, + "grad_norm": 0.19140625, + "learning_rate": 0.0011217839907148037, + "loss": 0.1133, + "step": 47769 + }, + { + "epoch": 0.414666539352957, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011217542369656934, + "loss": 0.1172, + "step": 47770 + }, + { + "epoch": 0.4146752198331612, + "grad_norm": 0.3515625, + "learning_rate": 0.001121724483192793, + "loss": 0.1309, + "step": 47771 + }, + { + "epoch": 0.41468390031336533, + "grad_norm": 0.51953125, + "learning_rate": 0.0011216947293961352, + "loss": 0.0967, + "step": 47772 + }, + { + "epoch": 0.4146925807935695, + "grad_norm": 0.107421875, + "learning_rate": 0.001121664975575752, + "loss": 0.1143, + "step": 47773 + }, + { + "epoch": 0.41470126127377366, + "grad_norm": 0.20703125, + "learning_rate": 0.0011216352217316768, + "loss": 0.0908, + "step": 47774 + }, + { + "epoch": 0.41470994175397785, + "grad_norm": 0.1728515625, + "learning_rate": 0.0011216054678639413, + "loss": 0.1221, + "step": 47775 + }, + { + "epoch": 0.414718622234182, + "grad_norm": 0.058837890625, + "learning_rate": 0.0011215757139725784, + "loss": 0.0737, + "step": 47776 + }, + { + "epoch": 0.4147273027143862, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011215459600576207, + "loss": 0.1143, + "step": 47777 + }, + { + "epoch": 0.4147359831945903, + "grad_norm": 0.1982421875, + "learning_rate": 0.0011215162061191004, + "loss": 0.0742, + "step": 47778 + }, + { + "epoch": 0.4147446636747945, + "grad_norm": 0.10205078125, + "learning_rate": 0.0011214864521570506, + "loss": 0.1152, + "step": 47779 + }, + { + "epoch": 0.41475334415499865, + "grad_norm": 0.326171875, + "learning_rate": 0.0011214566981715035, + "loss": 0.1162, + "step": 47780 + }, + { + "epoch": 0.41476202463520284, + "grad_norm": 0.2392578125, + "learning_rate": 0.001121426944162492, + "loss": 0.1084, + "step": 47781 + }, + { + "epoch": 0.414770705115407, + "grad_norm": 0.2353515625, + "learning_rate": 0.0011213971901300477, + "loss": 0.1187, + "step": 47782 + }, + { + "epoch": 0.41477938559561117, + "grad_norm": 0.5234375, + "learning_rate": 0.0011213674360742042, + "loss": 0.1387, + "step": 47783 + }, + { + "epoch": 0.4147880660758153, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011213376819949935, + "loss": 0.0591, + "step": 47784 + }, + { + "epoch": 0.4147967465560195, + "grad_norm": 0.330078125, + "learning_rate": 0.001121307927892448, + "loss": 0.0938, + "step": 47785 + }, + { + "epoch": 0.41480542703622364, + "grad_norm": 0.291015625, + "learning_rate": 0.0011212781737666004, + "loss": 0.0869, + "step": 47786 + }, + { + "epoch": 0.41481410751642783, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011212484196174834, + "loss": 0.106, + "step": 47787 + }, + { + "epoch": 0.41482278799663197, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011212186654451298, + "loss": 0.1162, + "step": 47788 + }, + { + "epoch": 0.41483146847683616, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011211889112495716, + "loss": 0.0811, + "step": 47789 + }, + { + "epoch": 0.4148401489570403, + "grad_norm": 0.478515625, + "learning_rate": 0.0011211591570308415, + "loss": 0.1221, + "step": 47790 + }, + { + "epoch": 0.4148488294372445, + "grad_norm": 0.55078125, + "learning_rate": 0.001121129402788972, + "loss": 0.1396, + "step": 47791 + }, + { + "epoch": 0.41485750991744863, + "grad_norm": 0.2470703125, + "learning_rate": 0.001121099648523996, + "loss": 0.0732, + "step": 47792 + }, + { + "epoch": 0.4148661903976528, + "grad_norm": 0.4453125, + "learning_rate": 0.0011210698942359452, + "loss": 0.1021, + "step": 47793 + }, + { + "epoch": 0.41487487087785696, + "grad_norm": 0.326171875, + "learning_rate": 0.0011210401399248534, + "loss": 0.1445, + "step": 47794 + }, + { + "epoch": 0.41488355135806115, + "grad_norm": 0.703125, + "learning_rate": 0.0011210103855907516, + "loss": 0.1055, + "step": 47795 + }, + { + "epoch": 0.4148922318382653, + "grad_norm": 0.0927734375, + "learning_rate": 0.0011209806312336733, + "loss": 0.1069, + "step": 47796 + }, + { + "epoch": 0.4149009123184695, + "grad_norm": 0.21484375, + "learning_rate": 0.0011209508768536514, + "loss": 0.1157, + "step": 47797 + }, + { + "epoch": 0.4149095927986736, + "grad_norm": 0.1162109375, + "learning_rate": 0.001120921122450718, + "loss": 0.0903, + "step": 47798 + }, + { + "epoch": 0.4149182732788778, + "grad_norm": 0.5234375, + "learning_rate": 0.001120891368024905, + "loss": 0.0967, + "step": 47799 + }, + { + "epoch": 0.41492695375908195, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011208616135762457, + "loss": 0.0996, + "step": 47800 + }, + { + "epoch": 0.41493563423928614, + "grad_norm": 0.3046875, + "learning_rate": 0.0011208318591047726, + "loss": 0.0986, + "step": 47801 + }, + { + "epoch": 0.4149443147194903, + "grad_norm": 0.2734375, + "learning_rate": 0.0011208021046105181, + "loss": 0.0703, + "step": 47802 + }, + { + "epoch": 0.4149529951996945, + "grad_norm": 0.310546875, + "learning_rate": 0.0011207723500935144, + "loss": 0.1191, + "step": 47803 + }, + { + "epoch": 0.4149616756798986, + "grad_norm": 0.34765625, + "learning_rate": 0.0011207425955537947, + "loss": 0.0957, + "step": 47804 + }, + { + "epoch": 0.4149703561601028, + "grad_norm": 0.18359375, + "learning_rate": 0.0011207128409913912, + "loss": 0.1055, + "step": 47805 + }, + { + "epoch": 0.41497903664030694, + "grad_norm": 0.1884765625, + "learning_rate": 0.0011206830864063364, + "loss": 0.1191, + "step": 47806 + }, + { + "epoch": 0.4149877171205111, + "grad_norm": 0.1376953125, + "learning_rate": 0.0011206533317986629, + "loss": 0.1035, + "step": 47807 + }, + { + "epoch": 0.41499639760071527, + "grad_norm": 0.283203125, + "learning_rate": 0.001120623577168403, + "loss": 0.0737, + "step": 47808 + }, + { + "epoch": 0.4150050780809194, + "grad_norm": 0.53515625, + "learning_rate": 0.0011205938225155895, + "loss": 0.0986, + "step": 47809 + }, + { + "epoch": 0.4150137585611236, + "grad_norm": 0.326171875, + "learning_rate": 0.0011205640678402552, + "loss": 0.1611, + "step": 47810 + }, + { + "epoch": 0.41502243904132774, + "grad_norm": 0.484375, + "learning_rate": 0.0011205343131424324, + "loss": 0.1582, + "step": 47811 + }, + { + "epoch": 0.41503111952153193, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011205045584221532, + "loss": 0.0762, + "step": 47812 + }, + { + "epoch": 0.41503980000173607, + "grad_norm": 0.0986328125, + "learning_rate": 0.0011204748036794509, + "loss": 0.0771, + "step": 47813 + }, + { + "epoch": 0.41504848048194026, + "grad_norm": 0.1328125, + "learning_rate": 0.0011204450489143575, + "loss": 0.0698, + "step": 47814 + }, + { + "epoch": 0.4150571609621444, + "grad_norm": 0.1181640625, + "learning_rate": 0.0011204152941269054, + "loss": 0.0977, + "step": 47815 + }, + { + "epoch": 0.4150658414423486, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011203855393171278, + "loss": 0.1113, + "step": 47816 + }, + { + "epoch": 0.41507452192255273, + "grad_norm": 0.53515625, + "learning_rate": 0.001120355784485057, + "loss": 0.1387, + "step": 47817 + }, + { + "epoch": 0.4150832024027569, + "grad_norm": 0.423828125, + "learning_rate": 0.001120326029630725, + "loss": 0.0815, + "step": 47818 + }, + { + "epoch": 0.41509188288296106, + "grad_norm": 0.328125, + "learning_rate": 0.0011202962747541652, + "loss": 0.0859, + "step": 47819 + }, + { + "epoch": 0.41510056336316525, + "grad_norm": 0.1259765625, + "learning_rate": 0.0011202665198554093, + "loss": 0.0947, + "step": 47820 + }, + { + "epoch": 0.4151092438433694, + "grad_norm": 0.08447265625, + "learning_rate": 0.0011202367649344905, + "loss": 0.0991, + "step": 47821 + }, + { + "epoch": 0.4151179243235736, + "grad_norm": 0.1494140625, + "learning_rate": 0.001120207009991441, + "loss": 0.1201, + "step": 47822 + }, + { + "epoch": 0.4151266048037777, + "grad_norm": 0.203125, + "learning_rate": 0.0011201772550262935, + "loss": 0.0684, + "step": 47823 + }, + { + "epoch": 0.4151352852839819, + "grad_norm": 0.59375, + "learning_rate": 0.0011201475000390804, + "loss": 0.1377, + "step": 47824 + }, + { + "epoch": 0.41514396576418605, + "grad_norm": 0.283203125, + "learning_rate": 0.0011201177450298342, + "loss": 0.0688, + "step": 47825 + }, + { + "epoch": 0.41515264624439024, + "grad_norm": 0.314453125, + "learning_rate": 0.0011200879899985873, + "loss": 0.1035, + "step": 47826 + }, + { + "epoch": 0.4151613267245944, + "grad_norm": 0.8359375, + "learning_rate": 0.001120058234945373, + "loss": 0.1084, + "step": 47827 + }, + { + "epoch": 0.4151700072047986, + "grad_norm": 0.2431640625, + "learning_rate": 0.0011200284798702231, + "loss": 0.0708, + "step": 47828 + }, + { + "epoch": 0.4151786876850027, + "grad_norm": 0.126953125, + "learning_rate": 0.0011199987247731704, + "loss": 0.0942, + "step": 47829 + }, + { + "epoch": 0.4151873681652069, + "grad_norm": 0.2265625, + "learning_rate": 0.0011199689696542471, + "loss": 0.0889, + "step": 47830 + }, + { + "epoch": 0.41519604864541104, + "grad_norm": 0.28125, + "learning_rate": 0.0011199392145134863, + "loss": 0.0732, + "step": 47831 + }, + { + "epoch": 0.41520472912561524, + "grad_norm": 0.2470703125, + "learning_rate": 0.0011199094593509205, + "loss": 0.126, + "step": 47832 + }, + { + "epoch": 0.4152134096058194, + "grad_norm": 0.19140625, + "learning_rate": 0.0011198797041665817, + "loss": 0.1309, + "step": 47833 + }, + { + "epoch": 0.41522209008602357, + "grad_norm": 0.30859375, + "learning_rate": 0.0011198499489605026, + "loss": 0.127, + "step": 47834 + }, + { + "epoch": 0.4152307705662277, + "grad_norm": 0.11669921875, + "learning_rate": 0.0011198201937327162, + "loss": 0.0859, + "step": 47835 + }, + { + "epoch": 0.4152394510464319, + "grad_norm": 0.2158203125, + "learning_rate": 0.0011197904384832548, + "loss": 0.1201, + "step": 47836 + }, + { + "epoch": 0.41524813152663603, + "grad_norm": 0.09228515625, + "learning_rate": 0.0011197606832121507, + "loss": 0.0767, + "step": 47837 + }, + { + "epoch": 0.4152568120068402, + "grad_norm": 0.6484375, + "learning_rate": 0.0011197309279194366, + "loss": 0.125, + "step": 47838 + }, + { + "epoch": 0.41526549248704436, + "grad_norm": 0.380859375, + "learning_rate": 0.0011197011726051449, + "loss": 0.0898, + "step": 47839 + }, + { + "epoch": 0.41527417296724856, + "grad_norm": 0.78125, + "learning_rate": 0.0011196714172693088, + "loss": 0.1436, + "step": 47840 + }, + { + "epoch": 0.4152828534474527, + "grad_norm": 0.16015625, + "learning_rate": 0.00111964166191196, + "loss": 0.1357, + "step": 47841 + }, + { + "epoch": 0.4152915339276569, + "grad_norm": 0.14453125, + "learning_rate": 0.001119611906533131, + "loss": 0.0957, + "step": 47842 + }, + { + "epoch": 0.415300214407861, + "grad_norm": 0.10693359375, + "learning_rate": 0.0011195821511328554, + "loss": 0.1074, + "step": 47843 + }, + { + "epoch": 0.4153088948880652, + "grad_norm": 0.1943359375, + "learning_rate": 0.0011195523957111646, + "loss": 0.1006, + "step": 47844 + }, + { + "epoch": 0.41531757536826935, + "grad_norm": 0.19140625, + "learning_rate": 0.0011195226402680918, + "loss": 0.1523, + "step": 47845 + }, + { + "epoch": 0.41532625584847355, + "grad_norm": 0.27734375, + "learning_rate": 0.0011194928848036695, + "loss": 0.0977, + "step": 47846 + }, + { + "epoch": 0.4153349363286777, + "grad_norm": 0.275390625, + "learning_rate": 0.0011194631293179296, + "loss": 0.1758, + "step": 47847 + }, + { + "epoch": 0.4153436168088819, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011194333738109054, + "loss": 0.0938, + "step": 47848 + }, + { + "epoch": 0.415352297289086, + "grad_norm": 0.2158203125, + "learning_rate": 0.0011194036182826292, + "loss": 0.0879, + "step": 47849 + }, + { + "epoch": 0.4153609777692902, + "grad_norm": 0.6953125, + "learning_rate": 0.0011193738627331335, + "loss": 0.1104, + "step": 47850 + }, + { + "epoch": 0.41536965824949434, + "grad_norm": 0.16015625, + "learning_rate": 0.0011193441071624506, + "loss": 0.1104, + "step": 47851 + }, + { + "epoch": 0.41537833872969854, + "grad_norm": 0.3125, + "learning_rate": 0.0011193143515706135, + "loss": 0.0977, + "step": 47852 + }, + { + "epoch": 0.4153870192099027, + "grad_norm": 0.11083984375, + "learning_rate": 0.0011192845959576542, + "loss": 0.0957, + "step": 47853 + }, + { + "epoch": 0.41539569969010687, + "grad_norm": 0.341796875, + "learning_rate": 0.001119254840323606, + "loss": 0.1016, + "step": 47854 + }, + { + "epoch": 0.415404380170311, + "grad_norm": 0.2373046875, + "learning_rate": 0.001119225084668501, + "loss": 0.0791, + "step": 47855 + }, + { + "epoch": 0.4154130606505152, + "grad_norm": 0.205078125, + "learning_rate": 0.0011191953289923712, + "loss": 0.1406, + "step": 47856 + }, + { + "epoch": 0.41542174113071934, + "grad_norm": 1.0390625, + "learning_rate": 0.0011191655732952501, + "loss": 0.1406, + "step": 47857 + }, + { + "epoch": 0.41543042161092353, + "grad_norm": 0.236328125, + "learning_rate": 0.00111913581757717, + "loss": 0.0928, + "step": 47858 + }, + { + "epoch": 0.41543910209112767, + "grad_norm": 0.458984375, + "learning_rate": 0.001119106061838163, + "loss": 0.1221, + "step": 47859 + }, + { + "epoch": 0.41544778257133186, + "grad_norm": 0.44921875, + "learning_rate": 0.0011190763060782616, + "loss": 0.0679, + "step": 47860 + }, + { + "epoch": 0.415456463051536, + "grad_norm": 0.177734375, + "learning_rate": 0.001119046550297499, + "loss": 0.1152, + "step": 47861 + }, + { + "epoch": 0.4154651435317402, + "grad_norm": 0.095703125, + "learning_rate": 0.0011190167944959072, + "loss": 0.0859, + "step": 47862 + }, + { + "epoch": 0.4154738240119443, + "grad_norm": 0.291015625, + "learning_rate": 0.0011189870386735191, + "loss": 0.0957, + "step": 47863 + }, + { + "epoch": 0.4154825044921485, + "grad_norm": 0.5390625, + "learning_rate": 0.0011189572828303668, + "loss": 0.1001, + "step": 47864 + }, + { + "epoch": 0.41549118497235266, + "grad_norm": 0.142578125, + "learning_rate": 0.0011189275269664833, + "loss": 0.1011, + "step": 47865 + }, + { + "epoch": 0.41549986545255685, + "grad_norm": 0.1923828125, + "learning_rate": 0.001118897771081901, + "loss": 0.1162, + "step": 47866 + }, + { + "epoch": 0.415508545932761, + "grad_norm": 0.796875, + "learning_rate": 0.0011188680151766525, + "loss": 0.373, + "step": 47867 + }, + { + "epoch": 0.4155172264129652, + "grad_norm": 0.69921875, + "learning_rate": 0.0011188382592507698, + "loss": 0.1099, + "step": 47868 + }, + { + "epoch": 0.4155259068931693, + "grad_norm": 0.45703125, + "learning_rate": 0.0011188085033042863, + "loss": 0.0835, + "step": 47869 + }, + { + "epoch": 0.4155345873733735, + "grad_norm": 0.5, + "learning_rate": 0.001118778747337234, + "loss": 0.1123, + "step": 47870 + }, + { + "epoch": 0.41554326785357765, + "grad_norm": 0.10400390625, + "learning_rate": 0.0011187489913496453, + "loss": 0.1035, + "step": 47871 + }, + { + "epoch": 0.41555194833378184, + "grad_norm": 0.115234375, + "learning_rate": 0.001118719235341553, + "loss": 0.0889, + "step": 47872 + }, + { + "epoch": 0.415560628813986, + "grad_norm": 0.2041015625, + "learning_rate": 0.0011186894793129897, + "loss": 0.123, + "step": 47873 + }, + { + "epoch": 0.41556930929419017, + "grad_norm": 1.0703125, + "learning_rate": 0.001118659723263988, + "loss": 0.1689, + "step": 47874 + }, + { + "epoch": 0.4155779897743943, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011186299671945804, + "loss": 0.0977, + "step": 47875 + }, + { + "epoch": 0.4155866702545985, + "grad_norm": 0.0986328125, + "learning_rate": 0.001118600211104799, + "loss": 0.1211, + "step": 47876 + }, + { + "epoch": 0.41559535073480264, + "grad_norm": 0.27734375, + "learning_rate": 0.0011185704549946766, + "loss": 0.1152, + "step": 47877 + }, + { + "epoch": 0.41560403121500683, + "grad_norm": 0.3125, + "learning_rate": 0.0011185406988642463, + "loss": 0.1582, + "step": 47878 + }, + { + "epoch": 0.41561271169521097, + "grad_norm": 1.203125, + "learning_rate": 0.0011185109427135401, + "loss": 0.3301, + "step": 47879 + }, + { + "epoch": 0.41562139217541516, + "grad_norm": 0.287109375, + "learning_rate": 0.0011184811865425903, + "loss": 0.0996, + "step": 47880 + }, + { + "epoch": 0.4156300726556193, + "grad_norm": 0.1044921875, + "learning_rate": 0.00111845143035143, + "loss": 0.0942, + "step": 47881 + }, + { + "epoch": 0.4156387531358235, + "grad_norm": 0.08837890625, + "learning_rate": 0.0011184216741400915, + "loss": 0.1025, + "step": 47882 + }, + { + "epoch": 0.41564743361602763, + "grad_norm": 0.2451171875, + "learning_rate": 0.001118391917908607, + "loss": 0.0967, + "step": 47883 + }, + { + "epoch": 0.4156561140962318, + "grad_norm": 0.279296875, + "learning_rate": 0.0011183621616570097, + "loss": 0.1465, + "step": 47884 + }, + { + "epoch": 0.41566479457643596, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011183324053853317, + "loss": 0.1309, + "step": 47885 + }, + { + "epoch": 0.41567347505664015, + "grad_norm": 0.08544921875, + "learning_rate": 0.0011183026490936057, + "loss": 0.0977, + "step": 47886 + }, + { + "epoch": 0.4156821555368443, + "grad_norm": 0.23046875, + "learning_rate": 0.0011182728927818643, + "loss": 0.105, + "step": 47887 + }, + { + "epoch": 0.4156908360170485, + "grad_norm": 0.10693359375, + "learning_rate": 0.0011182431364501397, + "loss": 0.0781, + "step": 47888 + }, + { + "epoch": 0.4156995164972526, + "grad_norm": 0.173828125, + "learning_rate": 0.0011182133800984651, + "loss": 0.0889, + "step": 47889 + }, + { + "epoch": 0.4157081969774568, + "grad_norm": 0.1337890625, + "learning_rate": 0.001118183623726872, + "loss": 0.1074, + "step": 47890 + }, + { + "epoch": 0.41571687745766095, + "grad_norm": 0.1337890625, + "learning_rate": 0.001118153867335394, + "loss": 0.083, + "step": 47891 + }, + { + "epoch": 0.41572555793786514, + "grad_norm": 0.142578125, + "learning_rate": 0.001118124110924063, + "loss": 0.0781, + "step": 47892 + }, + { + "epoch": 0.4157342384180693, + "grad_norm": 0.54296875, + "learning_rate": 0.0011180943544929118, + "loss": 0.1074, + "step": 47893 + }, + { + "epoch": 0.4157429188982735, + "grad_norm": 0.44921875, + "learning_rate": 0.0011180645980419727, + "loss": 0.1011, + "step": 47894 + }, + { + "epoch": 0.4157515993784776, + "grad_norm": 0.09716796875, + "learning_rate": 0.001118034841571279, + "loss": 0.1191, + "step": 47895 + }, + { + "epoch": 0.4157602798586818, + "grad_norm": 0.46875, + "learning_rate": 0.0011180050850808623, + "loss": 0.0908, + "step": 47896 + }, + { + "epoch": 0.41576896033888594, + "grad_norm": 0.337890625, + "learning_rate": 0.0011179753285707557, + "loss": 0.1084, + "step": 47897 + }, + { + "epoch": 0.41577764081909013, + "grad_norm": 0.404296875, + "learning_rate": 0.0011179455720409911, + "loss": 0.0781, + "step": 47898 + }, + { + "epoch": 0.41578632129929427, + "grad_norm": 0.359375, + "learning_rate": 0.0011179158154916021, + "loss": 0.1211, + "step": 47899 + }, + { + "epoch": 0.41579500177949846, + "grad_norm": 0.482421875, + "learning_rate": 0.00111788605892262, + "loss": 0.0947, + "step": 47900 + }, + { + "epoch": 0.4158036822597026, + "grad_norm": 0.0751953125, + "learning_rate": 0.001117856302334078, + "loss": 0.1128, + "step": 47901 + }, + { + "epoch": 0.4158123627399068, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011178265457260092, + "loss": 0.0879, + "step": 47902 + }, + { + "epoch": 0.41582104322011093, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011177967890984451, + "loss": 0.1035, + "step": 47903 + }, + { + "epoch": 0.4158297237003151, + "grad_norm": 0.212890625, + "learning_rate": 0.0011177670324514191, + "loss": 0.0713, + "step": 47904 + }, + { + "epoch": 0.41583840418051926, + "grad_norm": 0.283203125, + "learning_rate": 0.0011177372757849632, + "loss": 0.082, + "step": 47905 + }, + { + "epoch": 0.41584708466072345, + "grad_norm": 0.08349609375, + "learning_rate": 0.00111770751909911, + "loss": 0.1523, + "step": 47906 + }, + { + "epoch": 0.4158557651409276, + "grad_norm": 0.30859375, + "learning_rate": 0.001117677762393892, + "loss": 0.0981, + "step": 47907 + }, + { + "epoch": 0.4158644456211318, + "grad_norm": 0.09765625, + "learning_rate": 0.001117648005669342, + "loss": 0.0903, + "step": 47908 + }, + { + "epoch": 0.4158731261013359, + "grad_norm": 0.12060546875, + "learning_rate": 0.0011176182489254925, + "loss": 0.083, + "step": 47909 + }, + { + "epoch": 0.4158818065815401, + "grad_norm": 0.4140625, + "learning_rate": 0.0011175884921623758, + "loss": 0.0908, + "step": 47910 + }, + { + "epoch": 0.41589048706174425, + "grad_norm": 0.87109375, + "learning_rate": 0.0011175587353800245, + "loss": 0.0869, + "step": 47911 + }, + { + "epoch": 0.41589916754194844, + "grad_norm": 0.314453125, + "learning_rate": 0.0011175289785784713, + "loss": 0.1338, + "step": 47912 + }, + { + "epoch": 0.4159078480221526, + "grad_norm": 0.08740234375, + "learning_rate": 0.0011174992217577485, + "loss": 0.0923, + "step": 47913 + }, + { + "epoch": 0.4159165285023568, + "grad_norm": 0.55859375, + "learning_rate": 0.0011174694649178892, + "loss": 0.1221, + "step": 47914 + }, + { + "epoch": 0.4159252089825609, + "grad_norm": 0.185546875, + "learning_rate": 0.0011174397080589255, + "loss": 0.0986, + "step": 47915 + }, + { + "epoch": 0.4159338894627651, + "grad_norm": 0.12451171875, + "learning_rate": 0.0011174099511808897, + "loss": 0.1289, + "step": 47916 + }, + { + "epoch": 0.41594256994296924, + "grad_norm": 0.3828125, + "learning_rate": 0.0011173801942838148, + "loss": 0.0889, + "step": 47917 + }, + { + "epoch": 0.41595125042317344, + "grad_norm": 0.251953125, + "learning_rate": 0.0011173504373677334, + "loss": 0.1045, + "step": 47918 + }, + { + "epoch": 0.4159599309033776, + "grad_norm": 0.142578125, + "learning_rate": 0.0011173206804326776, + "loss": 0.1084, + "step": 47919 + }, + { + "epoch": 0.41596861138358177, + "grad_norm": 0.189453125, + "learning_rate": 0.0011172909234786802, + "loss": 0.0884, + "step": 47920 + }, + { + "epoch": 0.4159772918637859, + "grad_norm": 0.36328125, + "learning_rate": 0.0011172611665057734, + "loss": 0.0923, + "step": 47921 + }, + { + "epoch": 0.4159859723439901, + "grad_norm": 0.52734375, + "learning_rate": 0.0011172314095139904, + "loss": 0.0977, + "step": 47922 + }, + { + "epoch": 0.41599465282419423, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011172016525033636, + "loss": 0.106, + "step": 47923 + }, + { + "epoch": 0.4160033333043984, + "grad_norm": 0.2470703125, + "learning_rate": 0.001117171895473925, + "loss": 0.0884, + "step": 47924 + }, + { + "epoch": 0.41601201378460256, + "grad_norm": 0.2265625, + "learning_rate": 0.0011171421384257075, + "loss": 0.0938, + "step": 47925 + }, + { + "epoch": 0.41602069426480676, + "grad_norm": 0.31640625, + "learning_rate": 0.0011171123813587436, + "loss": 0.0908, + "step": 47926 + }, + { + "epoch": 0.4160293747450109, + "grad_norm": 0.173828125, + "learning_rate": 0.001117082624273066, + "loss": 0.0835, + "step": 47927 + }, + { + "epoch": 0.4160380552252151, + "grad_norm": 0.09326171875, + "learning_rate": 0.001117052867168707, + "loss": 0.0967, + "step": 47928 + }, + { + "epoch": 0.4160467357054192, + "grad_norm": 0.166015625, + "learning_rate": 0.001117023110045699, + "loss": 0.1533, + "step": 47929 + }, + { + "epoch": 0.41605541618562336, + "grad_norm": 0.74609375, + "learning_rate": 0.0011169933529040753, + "loss": 0.0908, + "step": 47930 + }, + { + "epoch": 0.41606409666582755, + "grad_norm": 0.1416015625, + "learning_rate": 0.0011169635957438675, + "loss": 0.1133, + "step": 47931 + }, + { + "epoch": 0.4160727771460317, + "grad_norm": 0.30078125, + "learning_rate": 0.0011169338385651087, + "loss": 0.1035, + "step": 47932 + }, + { + "epoch": 0.4160814576262359, + "grad_norm": 0.306640625, + "learning_rate": 0.001116904081367831, + "loss": 0.0776, + "step": 47933 + }, + { + "epoch": 0.41609013810644, + "grad_norm": 0.244140625, + "learning_rate": 0.0011168743241520677, + "loss": 0.0996, + "step": 47934 + }, + { + "epoch": 0.4160988185866442, + "grad_norm": 0.263671875, + "learning_rate": 0.0011168445669178509, + "loss": 0.1367, + "step": 47935 + }, + { + "epoch": 0.41610749906684835, + "grad_norm": 0.1484375, + "learning_rate": 0.0011168148096652132, + "loss": 0.1094, + "step": 47936 + }, + { + "epoch": 0.41611617954705254, + "grad_norm": 0.244140625, + "learning_rate": 0.001116785052394187, + "loss": 0.0942, + "step": 47937 + }, + { + "epoch": 0.4161248600272567, + "grad_norm": 0.734375, + "learning_rate": 0.0011167552951048046, + "loss": 0.0977, + "step": 47938 + }, + { + "epoch": 0.4161335405074609, + "grad_norm": 0.12158203125, + "learning_rate": 0.001116725537797099, + "loss": 0.1025, + "step": 47939 + }, + { + "epoch": 0.416142220987665, + "grad_norm": 0.103515625, + "learning_rate": 0.0011166957804711027, + "loss": 0.0825, + "step": 47940 + }, + { + "epoch": 0.4161509014678692, + "grad_norm": 0.1416015625, + "learning_rate": 0.0011166660231268482, + "loss": 0.1118, + "step": 47941 + }, + { + "epoch": 0.41615958194807334, + "grad_norm": 0.400390625, + "learning_rate": 0.0011166362657643676, + "loss": 0.1162, + "step": 47942 + }, + { + "epoch": 0.41616826242827754, + "grad_norm": 0.3203125, + "learning_rate": 0.0011166065083836943, + "loss": 0.1123, + "step": 47943 + }, + { + "epoch": 0.4161769429084817, + "grad_norm": 0.10400390625, + "learning_rate": 0.0011165767509848604, + "loss": 0.0747, + "step": 47944 + }, + { + "epoch": 0.41618562338868587, + "grad_norm": 0.1884765625, + "learning_rate": 0.001116546993567898, + "loss": 0.0894, + "step": 47945 + }, + { + "epoch": 0.41619430386889, + "grad_norm": 0.38671875, + "learning_rate": 0.00111651723613284, + "loss": 0.0664, + "step": 47946 + }, + { + "epoch": 0.4162029843490942, + "grad_norm": 0.19140625, + "learning_rate": 0.0011164874786797198, + "loss": 0.0991, + "step": 47947 + }, + { + "epoch": 0.41621166482929833, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011164577212085685, + "loss": 0.1064, + "step": 47948 + }, + { + "epoch": 0.4162203453095025, + "grad_norm": 0.5546875, + "learning_rate": 0.0011164279637194193, + "loss": 0.123, + "step": 47949 + }, + { + "epoch": 0.41622902578970666, + "grad_norm": 0.53125, + "learning_rate": 0.0011163982062123048, + "loss": 0.1475, + "step": 47950 + }, + { + "epoch": 0.41623770626991086, + "grad_norm": 0.326171875, + "learning_rate": 0.0011163684486872574, + "loss": 0.0957, + "step": 47951 + }, + { + "epoch": 0.416246386750115, + "grad_norm": 0.43359375, + "learning_rate": 0.0011163386911443099, + "loss": 0.1748, + "step": 47952 + }, + { + "epoch": 0.4162550672303192, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011163089335834945, + "loss": 0.1035, + "step": 47953 + }, + { + "epoch": 0.4162637477105233, + "grad_norm": 0.123046875, + "learning_rate": 0.001116279176004844, + "loss": 0.1289, + "step": 47954 + }, + { + "epoch": 0.4162724281907275, + "grad_norm": 0.458984375, + "learning_rate": 0.0011162494184083905, + "loss": 0.1152, + "step": 47955 + }, + { + "epoch": 0.41628110867093165, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011162196607941675, + "loss": 0.061, + "step": 47956 + }, + { + "epoch": 0.41628978915113585, + "grad_norm": 0.0810546875, + "learning_rate": 0.0011161899031622066, + "loss": 0.0776, + "step": 47957 + }, + { + "epoch": 0.41629846963134, + "grad_norm": 0.59375, + "learning_rate": 0.0011161601455125405, + "loss": 0.0835, + "step": 47958 + }, + { + "epoch": 0.4163071501115442, + "grad_norm": 2.734375, + "learning_rate": 0.001116130387845202, + "loss": 0.2363, + "step": 47959 + }, + { + "epoch": 0.4163158305917483, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011161006301602233, + "loss": 0.0732, + "step": 47960 + }, + { + "epoch": 0.4163245110719525, + "grad_norm": 0.150390625, + "learning_rate": 0.0011160708724576379, + "loss": 0.1035, + "step": 47961 + }, + { + "epoch": 0.41633319155215665, + "grad_norm": 0.12158203125, + "learning_rate": 0.0011160411147374772, + "loss": 0.0859, + "step": 47962 + }, + { + "epoch": 0.41634187203236084, + "grad_norm": 0.357421875, + "learning_rate": 0.001116011356999774, + "loss": 0.125, + "step": 47963 + }, + { + "epoch": 0.416350552512565, + "grad_norm": 0.1640625, + "learning_rate": 0.0011159815992445614, + "loss": 0.0864, + "step": 47964 + }, + { + "epoch": 0.41635923299276917, + "grad_norm": 0.21875, + "learning_rate": 0.0011159518414718714, + "loss": 0.0825, + "step": 47965 + }, + { + "epoch": 0.4163679134729733, + "grad_norm": 0.3046875, + "learning_rate": 0.0011159220836817368, + "loss": 0.0874, + "step": 47966 + }, + { + "epoch": 0.4163765939531775, + "grad_norm": 0.390625, + "learning_rate": 0.0011158923258741899, + "loss": 0.0933, + "step": 47967 + }, + { + "epoch": 0.41638527443338164, + "grad_norm": 0.23828125, + "learning_rate": 0.0011158625680492635, + "loss": 0.0889, + "step": 47968 + }, + { + "epoch": 0.41639395491358583, + "grad_norm": 0.1201171875, + "learning_rate": 0.0011158328102069896, + "loss": 0.0781, + "step": 47969 + }, + { + "epoch": 0.41640263539378997, + "grad_norm": 0.38671875, + "learning_rate": 0.0011158030523474017, + "loss": 0.103, + "step": 47970 + }, + { + "epoch": 0.41641131587399416, + "grad_norm": 0.224609375, + "learning_rate": 0.001115773294470532, + "loss": 0.1143, + "step": 47971 + }, + { + "epoch": 0.4164199963541983, + "grad_norm": 0.306640625, + "learning_rate": 0.0011157435365764122, + "loss": 0.0864, + "step": 47972 + }, + { + "epoch": 0.4164286768344025, + "grad_norm": 0.12158203125, + "learning_rate": 0.001115713778665076, + "loss": 0.1143, + "step": 47973 + }, + { + "epoch": 0.4164373573146066, + "grad_norm": 0.69921875, + "learning_rate": 0.0011156840207365553, + "loss": 0.0898, + "step": 47974 + }, + { + "epoch": 0.4164460377948108, + "grad_norm": 0.51953125, + "learning_rate": 0.0011156542627908828, + "loss": 0.1113, + "step": 47975 + }, + { + "epoch": 0.41645471827501496, + "grad_norm": 0.08251953125, + "learning_rate": 0.001115624504828091, + "loss": 0.082, + "step": 47976 + }, + { + "epoch": 0.41646339875521915, + "grad_norm": 0.1025390625, + "learning_rate": 0.0011155947468482123, + "loss": 0.0562, + "step": 47977 + }, + { + "epoch": 0.4164720792354233, + "grad_norm": 0.08642578125, + "learning_rate": 0.0011155649888512796, + "loss": 0.1328, + "step": 47978 + }, + { + "epoch": 0.4164807597156275, + "grad_norm": 0.267578125, + "learning_rate": 0.0011155352308373253, + "loss": 0.0762, + "step": 47979 + }, + { + "epoch": 0.4164894401958316, + "grad_norm": 0.080078125, + "learning_rate": 0.001115505472806382, + "loss": 0.0811, + "step": 47980 + }, + { + "epoch": 0.4164981206760358, + "grad_norm": 0.0859375, + "learning_rate": 0.0011154757147584819, + "loss": 0.1167, + "step": 47981 + }, + { + "epoch": 0.41650680115623995, + "grad_norm": 0.2890625, + "learning_rate": 0.0011154459566936582, + "loss": 0.0693, + "step": 47982 + }, + { + "epoch": 0.41651548163644414, + "grad_norm": 0.087890625, + "learning_rate": 0.0011154161986119426, + "loss": 0.0938, + "step": 47983 + }, + { + "epoch": 0.4165241621166483, + "grad_norm": 0.3203125, + "learning_rate": 0.0011153864405133683, + "loss": 0.0913, + "step": 47984 + }, + { + "epoch": 0.41653284259685247, + "grad_norm": 0.27734375, + "learning_rate": 0.0011153566823979672, + "loss": 0.0894, + "step": 47985 + }, + { + "epoch": 0.4165415230770566, + "grad_norm": 0.4296875, + "learning_rate": 0.001115326924265773, + "loss": 0.0791, + "step": 47986 + }, + { + "epoch": 0.4165502035572608, + "grad_norm": 0.09521484375, + "learning_rate": 0.001115297166116817, + "loss": 0.0898, + "step": 47987 + }, + { + "epoch": 0.41655888403746494, + "grad_norm": 0.1689453125, + "learning_rate": 0.0011152674079511322, + "loss": 0.104, + "step": 47988 + }, + { + "epoch": 0.41656756451766913, + "grad_norm": 0.1259765625, + "learning_rate": 0.001115237649768751, + "loss": 0.0767, + "step": 47989 + }, + { + "epoch": 0.41657624499787327, + "grad_norm": 0.1337890625, + "learning_rate": 0.0011152078915697067, + "loss": 0.124, + "step": 47990 + }, + { + "epoch": 0.41658492547807746, + "grad_norm": 0.25390625, + "learning_rate": 0.0011151781333540312, + "loss": 0.1235, + "step": 47991 + }, + { + "epoch": 0.4165936059582816, + "grad_norm": 0.130859375, + "learning_rate": 0.001115148375121757, + "loss": 0.1123, + "step": 47992 + }, + { + "epoch": 0.4166022864384858, + "grad_norm": 0.09912109375, + "learning_rate": 0.001115118616872917, + "loss": 0.0854, + "step": 47993 + }, + { + "epoch": 0.41661096691868993, + "grad_norm": 0.34765625, + "learning_rate": 0.001115088858607543, + "loss": 0.0967, + "step": 47994 + }, + { + "epoch": 0.4166196473988941, + "grad_norm": 0.21875, + "learning_rate": 0.0011150591003256682, + "loss": 0.0713, + "step": 47995 + }, + { + "epoch": 0.41662832787909826, + "grad_norm": 0.220703125, + "learning_rate": 0.0011150293420273253, + "loss": 0.1226, + "step": 47996 + }, + { + "epoch": 0.41663700835930245, + "grad_norm": 0.15625, + "learning_rate": 0.0011149995837125464, + "loss": 0.1074, + "step": 47997 + }, + { + "epoch": 0.4166456888395066, + "grad_norm": 0.41015625, + "learning_rate": 0.001114969825381364, + "loss": 0.1201, + "step": 47998 + }, + { + "epoch": 0.4166543693197108, + "grad_norm": 0.1083984375, + "learning_rate": 0.001114940067033811, + "loss": 0.0854, + "step": 47999 + }, + { + "epoch": 0.4166630497999149, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011149103086699197, + "loss": 0.1045, + "step": 48000 + }, + { + "epoch": 0.4166717302801191, + "grad_norm": 0.46484375, + "learning_rate": 0.0011148805502897228, + "loss": 0.1035, + "step": 48001 + }, + { + "epoch": 0.41668041076032325, + "grad_norm": 0.76953125, + "learning_rate": 0.0011148507918932527, + "loss": 0.1426, + "step": 48002 + }, + { + "epoch": 0.41668909124052744, + "grad_norm": 0.3203125, + "learning_rate": 0.0011148210334805417, + "loss": 0.0986, + "step": 48003 + }, + { + "epoch": 0.4166977717207316, + "grad_norm": 0.1884765625, + "learning_rate": 0.0011147912750516231, + "loss": 0.0645, + "step": 48004 + }, + { + "epoch": 0.4167064522009358, + "grad_norm": 0.294921875, + "learning_rate": 0.001114761516606529, + "loss": 0.0645, + "step": 48005 + }, + { + "epoch": 0.4167151326811399, + "grad_norm": 0.8046875, + "learning_rate": 0.0011147317581452917, + "loss": 0.0791, + "step": 48006 + }, + { + "epoch": 0.4167238131613441, + "grad_norm": 0.36328125, + "learning_rate": 0.001114701999667944, + "loss": 0.1113, + "step": 48007 + }, + { + "epoch": 0.41673249364154824, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011146722411745182, + "loss": 0.1074, + "step": 48008 + }, + { + "epoch": 0.41674117412175243, + "grad_norm": 0.197265625, + "learning_rate": 0.0011146424826650471, + "loss": 0.0864, + "step": 48009 + }, + { + "epoch": 0.41674985460195657, + "grad_norm": 0.60546875, + "learning_rate": 0.0011146127241395637, + "loss": 0.1074, + "step": 48010 + }, + { + "epoch": 0.41675853508216076, + "grad_norm": 0.875, + "learning_rate": 0.0011145829655980996, + "loss": 0.1201, + "step": 48011 + }, + { + "epoch": 0.4167672155623649, + "grad_norm": 0.1455078125, + "learning_rate": 0.0011145532070406879, + "loss": 0.1001, + "step": 48012 + }, + { + "epoch": 0.4167758960425691, + "grad_norm": 0.20703125, + "learning_rate": 0.0011145234484673612, + "loss": 0.0957, + "step": 48013 + }, + { + "epoch": 0.41678457652277323, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011144936898781516, + "loss": 0.0923, + "step": 48014 + }, + { + "epoch": 0.4167932570029774, + "grad_norm": 0.0830078125, + "learning_rate": 0.001114463931273092, + "loss": 0.0967, + "step": 48015 + }, + { + "epoch": 0.41680193748318156, + "grad_norm": 0.1943359375, + "learning_rate": 0.0011144341726522149, + "loss": 0.0859, + "step": 48016 + }, + { + "epoch": 0.41681061796338575, + "grad_norm": 0.185546875, + "learning_rate": 0.0011144044140155528, + "loss": 0.0771, + "step": 48017 + }, + { + "epoch": 0.4168192984435899, + "grad_norm": 0.1982421875, + "learning_rate": 0.001114374655363138, + "loss": 0.0991, + "step": 48018 + }, + { + "epoch": 0.4168279789237941, + "grad_norm": 0.1787109375, + "learning_rate": 0.0011143448966950038, + "loss": 0.0913, + "step": 48019 + }, + { + "epoch": 0.4168366594039982, + "grad_norm": 0.29296875, + "learning_rate": 0.0011143151380111817, + "loss": 0.0767, + "step": 48020 + }, + { + "epoch": 0.4168453398842024, + "grad_norm": 0.09375, + "learning_rate": 0.001114285379311705, + "loss": 0.0996, + "step": 48021 + }, + { + "epoch": 0.41685402036440655, + "grad_norm": 0.400390625, + "learning_rate": 0.0011142556205966064, + "loss": 0.1211, + "step": 48022 + }, + { + "epoch": 0.41686270084461075, + "grad_norm": 0.453125, + "learning_rate": 0.0011142258618659178, + "loss": 0.1543, + "step": 48023 + }, + { + "epoch": 0.4168713813248149, + "grad_norm": 0.07470703125, + "learning_rate": 0.0011141961031196717, + "loss": 0.0566, + "step": 48024 + }, + { + "epoch": 0.4168800618050191, + "grad_norm": 0.1162109375, + "learning_rate": 0.0011141663443579013, + "loss": 0.0947, + "step": 48025 + }, + { + "epoch": 0.4168887422852232, + "grad_norm": 0.26171875, + "learning_rate": 0.0011141365855806386, + "loss": 0.0947, + "step": 48026 + }, + { + "epoch": 0.4168974227654274, + "grad_norm": 0.271484375, + "learning_rate": 0.0011141068267879165, + "loss": 0.0752, + "step": 48027 + }, + { + "epoch": 0.41690610324563154, + "grad_norm": 0.12109375, + "learning_rate": 0.0011140770679797672, + "loss": 0.1191, + "step": 48028 + }, + { + "epoch": 0.41691478372583574, + "grad_norm": 0.271484375, + "learning_rate": 0.0011140473091562238, + "loss": 0.1133, + "step": 48029 + }, + { + "epoch": 0.4169234642060399, + "grad_norm": 0.10302734375, + "learning_rate": 0.0011140175503173183, + "loss": 0.1357, + "step": 48030 + }, + { + "epoch": 0.41693214468624407, + "grad_norm": 0.255859375, + "learning_rate": 0.001113987791463083, + "loss": 0.1016, + "step": 48031 + }, + { + "epoch": 0.4169408251664482, + "grad_norm": 0.2099609375, + "learning_rate": 0.0011139580325935517, + "loss": 0.0767, + "step": 48032 + }, + { + "epoch": 0.4169495056466524, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011139282737087555, + "loss": 0.0986, + "step": 48033 + }, + { + "epoch": 0.41695818612685653, + "grad_norm": 0.294921875, + "learning_rate": 0.0011138985148087277, + "loss": 0.1064, + "step": 48034 + }, + { + "epoch": 0.4169668666070607, + "grad_norm": 0.349609375, + "learning_rate": 0.0011138687558935009, + "loss": 0.0771, + "step": 48035 + }, + { + "epoch": 0.41697554708726486, + "grad_norm": 0.337890625, + "learning_rate": 0.001113838996963107, + "loss": 0.0913, + "step": 48036 + }, + { + "epoch": 0.41698422756746906, + "grad_norm": 0.33984375, + "learning_rate": 0.0011138092380175788, + "loss": 0.0835, + "step": 48037 + }, + { + "epoch": 0.4169929080476732, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011137794790569496, + "loss": 0.0718, + "step": 48038 + }, + { + "epoch": 0.4170015885278774, + "grad_norm": 0.6171875, + "learning_rate": 0.0011137497200812512, + "loss": 0.1113, + "step": 48039 + }, + { + "epoch": 0.4170102690080815, + "grad_norm": 0.32421875, + "learning_rate": 0.0011137199610905164, + "loss": 0.0835, + "step": 48040 + }, + { + "epoch": 0.4170189494882857, + "grad_norm": 0.244140625, + "learning_rate": 0.0011136902020847772, + "loss": 0.1465, + "step": 48041 + }, + { + "epoch": 0.41702762996848985, + "grad_norm": 0.1025390625, + "learning_rate": 0.0011136604430640672, + "loss": 0.083, + "step": 48042 + }, + { + "epoch": 0.41703631044869405, + "grad_norm": 0.328125, + "learning_rate": 0.0011136306840284181, + "loss": 0.0796, + "step": 48043 + }, + { + "epoch": 0.4170449909288982, + "grad_norm": 0.265625, + "learning_rate": 0.001113600924977863, + "loss": 0.0889, + "step": 48044 + }, + { + "epoch": 0.4170536714091024, + "grad_norm": 0.11083984375, + "learning_rate": 0.0011135711659124334, + "loss": 0.0933, + "step": 48045 + }, + { + "epoch": 0.4170623518893065, + "grad_norm": 0.0771484375, + "learning_rate": 0.001113541406832163, + "loss": 0.0908, + "step": 48046 + }, + { + "epoch": 0.4170710323695107, + "grad_norm": 0.2197265625, + "learning_rate": 0.001113511647737084, + "loss": 0.1367, + "step": 48047 + }, + { + "epoch": 0.41707971284971485, + "grad_norm": 0.59375, + "learning_rate": 0.0011134818886272288, + "loss": 0.0918, + "step": 48048 + }, + { + "epoch": 0.41708839332991904, + "grad_norm": 0.244140625, + "learning_rate": 0.00111345212950263, + "loss": 0.1289, + "step": 48049 + }, + { + "epoch": 0.4170970738101232, + "grad_norm": 0.283203125, + "learning_rate": 0.00111342237036332, + "loss": 0.1016, + "step": 48050 + }, + { + "epoch": 0.41710575429032737, + "grad_norm": 0.322265625, + "learning_rate": 0.0011133926112093318, + "loss": 0.0972, + "step": 48051 + }, + { + "epoch": 0.4171144347705315, + "grad_norm": 0.400390625, + "learning_rate": 0.0011133628520406974, + "loss": 0.1221, + "step": 48052 + }, + { + "epoch": 0.41712311525073564, + "grad_norm": 0.32421875, + "learning_rate": 0.0011133330928574498, + "loss": 0.082, + "step": 48053 + }, + { + "epoch": 0.41713179573093984, + "grad_norm": 0.80859375, + "learning_rate": 0.001113303333659621, + "loss": 0.0713, + "step": 48054 + }, + { + "epoch": 0.417140476211144, + "grad_norm": 0.62109375, + "learning_rate": 0.0011132735744472442, + "loss": 0.1162, + "step": 48055 + }, + { + "epoch": 0.41714915669134817, + "grad_norm": 0.06494140625, + "learning_rate": 0.0011132438152203512, + "loss": 0.0776, + "step": 48056 + }, + { + "epoch": 0.4171578371715523, + "grad_norm": 0.447265625, + "learning_rate": 0.0011132140559789755, + "loss": 0.1143, + "step": 48057 + }, + { + "epoch": 0.4171665176517565, + "grad_norm": 0.35546875, + "learning_rate": 0.001113184296723149, + "loss": 0.0923, + "step": 48058 + }, + { + "epoch": 0.41717519813196063, + "grad_norm": 0.58203125, + "learning_rate": 0.0011131545374529041, + "loss": 0.1348, + "step": 48059 + }, + { + "epoch": 0.4171838786121648, + "grad_norm": 0.154296875, + "learning_rate": 0.0011131247781682736, + "loss": 0.1494, + "step": 48060 + }, + { + "epoch": 0.41719255909236896, + "grad_norm": 0.1884765625, + "learning_rate": 0.0011130950188692904, + "loss": 0.125, + "step": 48061 + }, + { + "epoch": 0.41720123957257316, + "grad_norm": 0.1728515625, + "learning_rate": 0.0011130652595559864, + "loss": 0.125, + "step": 48062 + }, + { + "epoch": 0.4172099200527773, + "grad_norm": 0.353515625, + "learning_rate": 0.0011130355002283944, + "loss": 0.1504, + "step": 48063 + }, + { + "epoch": 0.4172186005329815, + "grad_norm": 0.1513671875, + "learning_rate": 0.001113005740886547, + "loss": 0.0864, + "step": 48064 + }, + { + "epoch": 0.4172272810131856, + "grad_norm": 0.25390625, + "learning_rate": 0.0011129759815304765, + "loss": 0.0654, + "step": 48065 + }, + { + "epoch": 0.4172359614933898, + "grad_norm": 0.421875, + "learning_rate": 0.0011129462221602163, + "loss": 0.1611, + "step": 48066 + }, + { + "epoch": 0.41724464197359395, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011129164627757978, + "loss": 0.0693, + "step": 48067 + }, + { + "epoch": 0.41725332245379815, + "grad_norm": 0.07666015625, + "learning_rate": 0.001112886703377254, + "loss": 0.0703, + "step": 48068 + }, + { + "epoch": 0.4172620029340023, + "grad_norm": 0.140625, + "learning_rate": 0.0011128569439646178, + "loss": 0.0845, + "step": 48069 + }, + { + "epoch": 0.4172706834142065, + "grad_norm": 0.1728515625, + "learning_rate": 0.0011128271845379212, + "loss": 0.0618, + "step": 48070 + }, + { + "epoch": 0.4172793638944106, + "grad_norm": 0.2734375, + "learning_rate": 0.0011127974250971972, + "loss": 0.0728, + "step": 48071 + }, + { + "epoch": 0.4172880443746148, + "grad_norm": 0.5859375, + "learning_rate": 0.001112767665642478, + "loss": 0.0835, + "step": 48072 + }, + { + "epoch": 0.41729672485481895, + "grad_norm": 0.1630859375, + "learning_rate": 0.0011127379061737963, + "loss": 0.1094, + "step": 48073 + }, + { + "epoch": 0.41730540533502314, + "grad_norm": 0.392578125, + "learning_rate": 0.0011127081466911844, + "loss": 0.1221, + "step": 48074 + }, + { + "epoch": 0.4173140858152273, + "grad_norm": 0.5, + "learning_rate": 0.0011126783871946754, + "loss": 0.1094, + "step": 48075 + }, + { + "epoch": 0.41732276629543147, + "grad_norm": 0.2236328125, + "learning_rate": 0.0011126486276843012, + "loss": 0.0996, + "step": 48076 + }, + { + "epoch": 0.4173314467756356, + "grad_norm": 0.49609375, + "learning_rate": 0.0011126188681600951, + "loss": 0.0767, + "step": 48077 + }, + { + "epoch": 0.4173401272558398, + "grad_norm": 0.435546875, + "learning_rate": 0.0011125891086220887, + "loss": 0.0942, + "step": 48078 + }, + { + "epoch": 0.41734880773604394, + "grad_norm": 0.14453125, + "learning_rate": 0.0011125593490703153, + "loss": 0.0737, + "step": 48079 + }, + { + "epoch": 0.41735748821624813, + "grad_norm": 0.83984375, + "learning_rate": 0.001112529589504807, + "loss": 0.0913, + "step": 48080 + }, + { + "epoch": 0.41736616869645227, + "grad_norm": 0.388671875, + "learning_rate": 0.0011124998299255969, + "loss": 0.0864, + "step": 48081 + }, + { + "epoch": 0.41737484917665646, + "grad_norm": 0.5625, + "learning_rate": 0.0011124700703327168, + "loss": 0.0791, + "step": 48082 + }, + { + "epoch": 0.4173835296568606, + "grad_norm": 0.35546875, + "learning_rate": 0.0011124403107261997, + "loss": 0.0889, + "step": 48083 + }, + { + "epoch": 0.4173922101370648, + "grad_norm": 0.36328125, + "learning_rate": 0.0011124105511060782, + "loss": 0.0864, + "step": 48084 + }, + { + "epoch": 0.4174008906172689, + "grad_norm": 0.80859375, + "learning_rate": 0.0011123807914723846, + "loss": 0.1113, + "step": 48085 + }, + { + "epoch": 0.4174095710974731, + "grad_norm": 0.275390625, + "learning_rate": 0.0011123510318251516, + "loss": 0.1191, + "step": 48086 + }, + { + "epoch": 0.41741825157767726, + "grad_norm": 0.130859375, + "learning_rate": 0.0011123212721644118, + "loss": 0.0801, + "step": 48087 + }, + { + "epoch": 0.41742693205788145, + "grad_norm": 0.0517578125, + "learning_rate": 0.0011122915124901975, + "loss": 0.0664, + "step": 48088 + }, + { + "epoch": 0.4174356125380856, + "grad_norm": 0.279296875, + "learning_rate": 0.0011122617528025415, + "loss": 0.0762, + "step": 48089 + }, + { + "epoch": 0.4174442930182898, + "grad_norm": 0.515625, + "learning_rate": 0.0011122319931014763, + "loss": 0.0928, + "step": 48090 + }, + { + "epoch": 0.4174529734984939, + "grad_norm": 0.06298828125, + "learning_rate": 0.001112202233387034, + "loss": 0.0718, + "step": 48091 + }, + { + "epoch": 0.4174616539786981, + "grad_norm": 0.09423828125, + "learning_rate": 0.0011121724736592477, + "loss": 0.1143, + "step": 48092 + }, + { + "epoch": 0.41747033445890225, + "grad_norm": 0.171875, + "learning_rate": 0.0011121427139181498, + "loss": 0.0786, + "step": 48093 + }, + { + "epoch": 0.41747901493910644, + "grad_norm": 0.11572265625, + "learning_rate": 0.0011121129541637727, + "loss": 0.1299, + "step": 48094 + }, + { + "epoch": 0.4174876954193106, + "grad_norm": 0.181640625, + "learning_rate": 0.0011120831943961494, + "loss": 0.0933, + "step": 48095 + }, + { + "epoch": 0.41749637589951477, + "grad_norm": 0.10107421875, + "learning_rate": 0.0011120534346153116, + "loss": 0.1099, + "step": 48096 + }, + { + "epoch": 0.4175050563797189, + "grad_norm": 0.46484375, + "learning_rate": 0.0011120236748212926, + "loss": 0.0796, + "step": 48097 + }, + { + "epoch": 0.4175137368599231, + "grad_norm": 0.68359375, + "learning_rate": 0.0011119939150141247, + "loss": 0.0996, + "step": 48098 + }, + { + "epoch": 0.41752241734012724, + "grad_norm": 0.4921875, + "learning_rate": 0.0011119641551938406, + "loss": 0.103, + "step": 48099 + }, + { + "epoch": 0.41753109782033143, + "grad_norm": 0.17578125, + "learning_rate": 0.0011119343953604726, + "loss": 0.0938, + "step": 48100 + }, + { + "epoch": 0.41753977830053557, + "grad_norm": 0.439453125, + "learning_rate": 0.0011119046355140532, + "loss": 0.0718, + "step": 48101 + }, + { + "epoch": 0.41754845878073976, + "grad_norm": 0.34375, + "learning_rate": 0.0011118748756546151, + "loss": 0.0786, + "step": 48102 + }, + { + "epoch": 0.4175571392609439, + "grad_norm": 0.48046875, + "learning_rate": 0.0011118451157821907, + "loss": 0.1011, + "step": 48103 + }, + { + "epoch": 0.4175658197411481, + "grad_norm": 0.193359375, + "learning_rate": 0.0011118153558968126, + "loss": 0.1318, + "step": 48104 + }, + { + "epoch": 0.41757450022135223, + "grad_norm": 0.263671875, + "learning_rate": 0.001111785595998514, + "loss": 0.0957, + "step": 48105 + }, + { + "epoch": 0.4175831807015564, + "grad_norm": 0.22265625, + "learning_rate": 0.0011117558360873264, + "loss": 0.1201, + "step": 48106 + }, + { + "epoch": 0.41759186118176056, + "grad_norm": 0.07275390625, + "learning_rate": 0.0011117260761632827, + "loss": 0.0771, + "step": 48107 + }, + { + "epoch": 0.41760054166196475, + "grad_norm": 0.294921875, + "learning_rate": 0.0011116963162264158, + "loss": 0.127, + "step": 48108 + }, + { + "epoch": 0.4176092221421689, + "grad_norm": 0.0966796875, + "learning_rate": 0.0011116665562767578, + "loss": 0.0962, + "step": 48109 + }, + { + "epoch": 0.4176179026223731, + "grad_norm": 0.1328125, + "learning_rate": 0.0011116367963143416, + "loss": 0.1064, + "step": 48110 + }, + { + "epoch": 0.4176265831025772, + "grad_norm": 0.091796875, + "learning_rate": 0.0011116070363391994, + "loss": 0.1104, + "step": 48111 + }, + { + "epoch": 0.4176352635827814, + "grad_norm": 0.349609375, + "learning_rate": 0.0011115772763513638, + "loss": 0.2207, + "step": 48112 + }, + { + "epoch": 0.41764394406298555, + "grad_norm": 0.578125, + "learning_rate": 0.0011115475163508677, + "loss": 0.0977, + "step": 48113 + }, + { + "epoch": 0.41765262454318974, + "grad_norm": 0.2421875, + "learning_rate": 0.0011115177563377435, + "loss": 0.0977, + "step": 48114 + }, + { + "epoch": 0.4176613050233939, + "grad_norm": 0.224609375, + "learning_rate": 0.0011114879963120233, + "loss": 0.0752, + "step": 48115 + }, + { + "epoch": 0.4176699855035981, + "grad_norm": 0.4609375, + "learning_rate": 0.0011114582362737404, + "loss": 0.0791, + "step": 48116 + }, + { + "epoch": 0.4176786659838022, + "grad_norm": 0.26171875, + "learning_rate": 0.0011114284762229268, + "loss": 0.1406, + "step": 48117 + }, + { + "epoch": 0.4176873464640064, + "grad_norm": 0.19140625, + "learning_rate": 0.0011113987161596153, + "loss": 0.1001, + "step": 48118 + }, + { + "epoch": 0.41769602694421054, + "grad_norm": 0.1044921875, + "learning_rate": 0.001111368956083838, + "loss": 0.0835, + "step": 48119 + }, + { + "epoch": 0.41770470742441473, + "grad_norm": 0.302734375, + "learning_rate": 0.0011113391959956277, + "loss": 0.0684, + "step": 48120 + }, + { + "epoch": 0.41771338790461887, + "grad_norm": 0.2001953125, + "learning_rate": 0.0011113094358950176, + "loss": 0.1006, + "step": 48121 + }, + { + "epoch": 0.41772206838482306, + "grad_norm": 0.123046875, + "learning_rate": 0.0011112796757820392, + "loss": 0.1172, + "step": 48122 + }, + { + "epoch": 0.4177307488650272, + "grad_norm": 0.36328125, + "learning_rate": 0.0011112499156567255, + "loss": 0.125, + "step": 48123 + }, + { + "epoch": 0.4177394293452314, + "grad_norm": 0.8203125, + "learning_rate": 0.0011112201555191095, + "loss": 0.1211, + "step": 48124 + }, + { + "epoch": 0.41774810982543553, + "grad_norm": 0.298828125, + "learning_rate": 0.001111190395369223, + "loss": 0.0762, + "step": 48125 + }, + { + "epoch": 0.4177567903056397, + "grad_norm": 0.3828125, + "learning_rate": 0.0011111606352070989, + "loss": 0.1133, + "step": 48126 + }, + { + "epoch": 0.41776547078584386, + "grad_norm": 0.17578125, + "learning_rate": 0.0011111308750327699, + "loss": 0.125, + "step": 48127 + }, + { + "epoch": 0.41777415126604805, + "grad_norm": 0.375, + "learning_rate": 0.0011111011148462679, + "loss": 0.0869, + "step": 48128 + }, + { + "epoch": 0.4177828317462522, + "grad_norm": 0.1796875, + "learning_rate": 0.001111071354647626, + "loss": 0.1426, + "step": 48129 + }, + { + "epoch": 0.4177915122264564, + "grad_norm": 0.10546875, + "learning_rate": 0.001111041594436877, + "loss": 0.1133, + "step": 48130 + }, + { + "epoch": 0.4178001927066605, + "grad_norm": 0.419921875, + "learning_rate": 0.0011110118342140527, + "loss": 0.1289, + "step": 48131 + }, + { + "epoch": 0.4178088731868647, + "grad_norm": 0.1015625, + "learning_rate": 0.0011109820739791862, + "loss": 0.1133, + "step": 48132 + }, + { + "epoch": 0.41781755366706885, + "grad_norm": 0.189453125, + "learning_rate": 0.0011109523137323096, + "loss": 0.1064, + "step": 48133 + }, + { + "epoch": 0.41782623414727305, + "grad_norm": 0.2265625, + "learning_rate": 0.0011109225534734562, + "loss": 0.168, + "step": 48134 + }, + { + "epoch": 0.4178349146274772, + "grad_norm": 0.1298828125, + "learning_rate": 0.001110892793202658, + "loss": 0.1182, + "step": 48135 + }, + { + "epoch": 0.4178435951076814, + "grad_norm": 0.07421875, + "learning_rate": 0.0011108630329199468, + "loss": 0.0664, + "step": 48136 + }, + { + "epoch": 0.4178522755878855, + "grad_norm": 0.1591796875, + "learning_rate": 0.0011108332726253566, + "loss": 0.0723, + "step": 48137 + }, + { + "epoch": 0.4178609560680897, + "grad_norm": 0.0888671875, + "learning_rate": 0.0011108035123189195, + "loss": 0.1094, + "step": 48138 + }, + { + "epoch": 0.41786963654829384, + "grad_norm": 0.169921875, + "learning_rate": 0.0011107737520006675, + "loss": 0.0977, + "step": 48139 + }, + { + "epoch": 0.41787831702849804, + "grad_norm": 0.267578125, + "learning_rate": 0.0011107439916706338, + "loss": 0.1118, + "step": 48140 + }, + { + "epoch": 0.4178869975087022, + "grad_norm": 0.33203125, + "learning_rate": 0.0011107142313288502, + "loss": 0.0845, + "step": 48141 + }, + { + "epoch": 0.41789567798890637, + "grad_norm": 0.0869140625, + "learning_rate": 0.0011106844709753499, + "loss": 0.0933, + "step": 48142 + }, + { + "epoch": 0.4179043584691105, + "grad_norm": 0.38671875, + "learning_rate": 0.0011106547106101653, + "loss": 0.0933, + "step": 48143 + }, + { + "epoch": 0.4179130389493147, + "grad_norm": 0.4375, + "learning_rate": 0.0011106249502333288, + "loss": 0.1533, + "step": 48144 + }, + { + "epoch": 0.41792171942951883, + "grad_norm": 0.2109375, + "learning_rate": 0.0011105951898448727, + "loss": 0.0742, + "step": 48145 + }, + { + "epoch": 0.417930399909723, + "grad_norm": 0.2177734375, + "learning_rate": 0.0011105654294448304, + "loss": 0.1348, + "step": 48146 + }, + { + "epoch": 0.41793908038992716, + "grad_norm": 0.1982421875, + "learning_rate": 0.0011105356690332337, + "loss": 0.0811, + "step": 48147 + }, + { + "epoch": 0.41794776087013136, + "grad_norm": 0.5234375, + "learning_rate": 0.0011105059086101152, + "loss": 0.0845, + "step": 48148 + }, + { + "epoch": 0.4179564413503355, + "grad_norm": 0.17578125, + "learning_rate": 0.0011104761481755079, + "loss": 0.0952, + "step": 48149 + }, + { + "epoch": 0.4179651218305397, + "grad_norm": 0.7265625, + "learning_rate": 0.0011104463877294438, + "loss": 0.0908, + "step": 48150 + }, + { + "epoch": 0.4179738023107438, + "grad_norm": 0.15234375, + "learning_rate": 0.0011104166272719557, + "loss": 0.1191, + "step": 48151 + }, + { + "epoch": 0.417982482790948, + "grad_norm": 0.2138671875, + "learning_rate": 0.001110386866803076, + "loss": 0.1045, + "step": 48152 + }, + { + "epoch": 0.41799116327115216, + "grad_norm": 0.51171875, + "learning_rate": 0.0011103571063228377, + "loss": 0.0996, + "step": 48153 + }, + { + "epoch": 0.41799984375135635, + "grad_norm": 0.259765625, + "learning_rate": 0.0011103273458312728, + "loss": 0.105, + "step": 48154 + }, + { + "epoch": 0.4180085242315605, + "grad_norm": 0.25390625, + "learning_rate": 0.001110297585328414, + "loss": 0.0879, + "step": 48155 + }, + { + "epoch": 0.4180172047117647, + "grad_norm": 0.322265625, + "learning_rate": 0.0011102678248142944, + "loss": 0.0986, + "step": 48156 + }, + { + "epoch": 0.4180258851919688, + "grad_norm": 0.1025390625, + "learning_rate": 0.0011102380642889456, + "loss": 0.1016, + "step": 48157 + }, + { + "epoch": 0.418034565672173, + "grad_norm": 0.1171875, + "learning_rate": 0.0011102083037524007, + "loss": 0.0635, + "step": 48158 + }, + { + "epoch": 0.41804324615237715, + "grad_norm": 0.78125, + "learning_rate": 0.0011101785432046922, + "loss": 0.1465, + "step": 48159 + }, + { + "epoch": 0.41805192663258134, + "grad_norm": 0.2255859375, + "learning_rate": 0.0011101487826458525, + "loss": 0.1011, + "step": 48160 + }, + { + "epoch": 0.4180606071127855, + "grad_norm": 0.06982421875, + "learning_rate": 0.0011101190220759144, + "loss": 0.0835, + "step": 48161 + }, + { + "epoch": 0.41806928759298967, + "grad_norm": 0.45703125, + "learning_rate": 0.0011100892614949098, + "loss": 0.1006, + "step": 48162 + }, + { + "epoch": 0.4180779680731938, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011100595009028724, + "loss": 0.1069, + "step": 48163 + }, + { + "epoch": 0.418086648553398, + "grad_norm": 0.208984375, + "learning_rate": 0.0011100297402998338, + "loss": 0.1191, + "step": 48164 + }, + { + "epoch": 0.41809532903360214, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011099999796858268, + "loss": 0.1084, + "step": 48165 + }, + { + "epoch": 0.41810400951380633, + "grad_norm": 0.12060546875, + "learning_rate": 0.0011099702190608843, + "loss": 0.1113, + "step": 48166 + }, + { + "epoch": 0.41811268999401047, + "grad_norm": 0.1787109375, + "learning_rate": 0.001109940458425038, + "loss": 0.0786, + "step": 48167 + }, + { + "epoch": 0.41812137047421466, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011099106977783214, + "loss": 0.0835, + "step": 48168 + }, + { + "epoch": 0.4181300509544188, + "grad_norm": 0.234375, + "learning_rate": 0.0011098809371207663, + "loss": 0.0669, + "step": 48169 + }, + { + "epoch": 0.418138731434623, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011098511764524054, + "loss": 0.124, + "step": 48170 + }, + { + "epoch": 0.4181474119148271, + "grad_norm": 0.236328125, + "learning_rate": 0.0011098214157732717, + "loss": 0.1465, + "step": 48171 + }, + { + "epoch": 0.4181560923950313, + "grad_norm": 0.14453125, + "learning_rate": 0.0011097916550833976, + "loss": 0.125, + "step": 48172 + }, + { + "epoch": 0.41816477287523546, + "grad_norm": 0.396484375, + "learning_rate": 0.0011097618943828151, + "loss": 0.1133, + "step": 48173 + }, + { + "epoch": 0.41817345335543965, + "grad_norm": 0.55859375, + "learning_rate": 0.0011097321336715578, + "loss": 0.124, + "step": 48174 + }, + { + "epoch": 0.4181821338356438, + "grad_norm": 0.1513671875, + "learning_rate": 0.001109702372949657, + "loss": 0.0908, + "step": 48175 + }, + { + "epoch": 0.4181908143158479, + "grad_norm": 0.11328125, + "learning_rate": 0.0011096726122171458, + "loss": 0.0928, + "step": 48176 + }, + { + "epoch": 0.4181994947960521, + "grad_norm": 0.11767578125, + "learning_rate": 0.0011096428514740572, + "loss": 0.0698, + "step": 48177 + }, + { + "epoch": 0.41820817527625626, + "grad_norm": 0.0927734375, + "learning_rate": 0.001109613090720423, + "loss": 0.0781, + "step": 48178 + }, + { + "epoch": 0.41821685575646045, + "grad_norm": 0.400390625, + "learning_rate": 0.001109583329956276, + "loss": 0.1484, + "step": 48179 + }, + { + "epoch": 0.4182255362366646, + "grad_norm": 0.15234375, + "learning_rate": 0.001109553569181649, + "loss": 0.0889, + "step": 48180 + }, + { + "epoch": 0.4182342167168688, + "grad_norm": 0.322265625, + "learning_rate": 0.0011095238083965741, + "loss": 0.0869, + "step": 48181 + }, + { + "epoch": 0.4182428971970729, + "grad_norm": 0.07568359375, + "learning_rate": 0.0011094940476010845, + "loss": 0.0864, + "step": 48182 + }, + { + "epoch": 0.4182515776772771, + "grad_norm": 0.275390625, + "learning_rate": 0.0011094642867952123, + "loss": 0.0854, + "step": 48183 + }, + { + "epoch": 0.41826025815748125, + "grad_norm": 0.15234375, + "learning_rate": 0.0011094345259789898, + "loss": 0.0918, + "step": 48184 + }, + { + "epoch": 0.41826893863768544, + "grad_norm": 0.1123046875, + "learning_rate": 0.00110940476515245, + "loss": 0.0747, + "step": 48185 + }, + { + "epoch": 0.4182776191178896, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011093750043156254, + "loss": 0.0977, + "step": 48186 + }, + { + "epoch": 0.41828629959809377, + "grad_norm": 0.34765625, + "learning_rate": 0.0011093452434685486, + "loss": 0.1108, + "step": 48187 + }, + { + "epoch": 0.4182949800782979, + "grad_norm": 0.2431640625, + "learning_rate": 0.0011093154826112514, + "loss": 0.1074, + "step": 48188 + }, + { + "epoch": 0.4183036605585021, + "grad_norm": 0.11474609375, + "learning_rate": 0.0011092857217437674, + "loss": 0.0762, + "step": 48189 + }, + { + "epoch": 0.41831234103870624, + "grad_norm": 0.11865234375, + "learning_rate": 0.0011092559608661282, + "loss": 0.1221, + "step": 48190 + }, + { + "epoch": 0.41832102151891043, + "grad_norm": 0.4375, + "learning_rate": 0.0011092261999783672, + "loss": 0.1216, + "step": 48191 + }, + { + "epoch": 0.41832970199911457, + "grad_norm": 0.51171875, + "learning_rate": 0.0011091964390805167, + "loss": 0.1133, + "step": 48192 + }, + { + "epoch": 0.41833838247931876, + "grad_norm": 0.123046875, + "learning_rate": 0.0011091666781726085, + "loss": 0.082, + "step": 48193 + }, + { + "epoch": 0.4183470629595229, + "grad_norm": 0.1953125, + "learning_rate": 0.0011091369172546764, + "loss": 0.0879, + "step": 48194 + }, + { + "epoch": 0.4183557434397271, + "grad_norm": 0.2294921875, + "learning_rate": 0.0011091071563267521, + "loss": 0.0889, + "step": 48195 + }, + { + "epoch": 0.4183644239199312, + "grad_norm": 0.361328125, + "learning_rate": 0.0011090773953888683, + "loss": 0.1133, + "step": 48196 + }, + { + "epoch": 0.4183731044001354, + "grad_norm": 0.388671875, + "learning_rate": 0.0011090476344410576, + "loss": 0.1426, + "step": 48197 + }, + { + "epoch": 0.41838178488033956, + "grad_norm": 0.09716796875, + "learning_rate": 0.0011090178734833528, + "loss": 0.1289, + "step": 48198 + }, + { + "epoch": 0.41839046536054375, + "grad_norm": 0.333984375, + "learning_rate": 0.0011089881125157857, + "loss": 0.1348, + "step": 48199 + }, + { + "epoch": 0.4183991458407479, + "grad_norm": 0.181640625, + "learning_rate": 0.00110895835153839, + "loss": 0.0918, + "step": 48200 + }, + { + "epoch": 0.4184078263209521, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011089285905511968, + "loss": 0.0957, + "step": 48201 + }, + { + "epoch": 0.4184165068011562, + "grad_norm": 0.0625, + "learning_rate": 0.0011088988295542398, + "loss": 0.0525, + "step": 48202 + }, + { + "epoch": 0.4184251872813604, + "grad_norm": 0.216796875, + "learning_rate": 0.0011088690685475513, + "loss": 0.1108, + "step": 48203 + }, + { + "epoch": 0.41843386776156455, + "grad_norm": 0.5390625, + "learning_rate": 0.0011088393075311636, + "loss": 0.0962, + "step": 48204 + }, + { + "epoch": 0.41844254824176874, + "grad_norm": 0.3984375, + "learning_rate": 0.001108809546505109, + "loss": 0.0928, + "step": 48205 + }, + { + "epoch": 0.4184512287219729, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011087797854694211, + "loss": 0.0942, + "step": 48206 + }, + { + "epoch": 0.41845990920217707, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011087500244241315, + "loss": 0.0986, + "step": 48207 + }, + { + "epoch": 0.4184685896823812, + "grad_norm": 0.1923828125, + "learning_rate": 0.001108720263369273, + "loss": 0.1069, + "step": 48208 + }, + { + "epoch": 0.4184772701625854, + "grad_norm": 0.10693359375, + "learning_rate": 0.0011086905023048779, + "loss": 0.0933, + "step": 48209 + }, + { + "epoch": 0.41848595064278954, + "grad_norm": 0.57421875, + "learning_rate": 0.0011086607412309792, + "loss": 0.1143, + "step": 48210 + }, + { + "epoch": 0.41849463112299373, + "grad_norm": 0.83203125, + "learning_rate": 0.0011086309801476093, + "loss": 0.125, + "step": 48211 + }, + { + "epoch": 0.41850331160319787, + "grad_norm": 0.416015625, + "learning_rate": 0.0011086012190548004, + "loss": 0.1104, + "step": 48212 + }, + { + "epoch": 0.41851199208340206, + "grad_norm": 0.404296875, + "learning_rate": 0.0011085714579525858, + "loss": 0.0879, + "step": 48213 + }, + { + "epoch": 0.4185206725636062, + "grad_norm": 0.08203125, + "learning_rate": 0.0011085416968409972, + "loss": 0.1104, + "step": 48214 + }, + { + "epoch": 0.4185293530438104, + "grad_norm": 0.423828125, + "learning_rate": 0.0011085119357200678, + "loss": 0.0972, + "step": 48215 + }, + { + "epoch": 0.41853803352401453, + "grad_norm": 0.099609375, + "learning_rate": 0.0011084821745898298, + "loss": 0.1118, + "step": 48216 + }, + { + "epoch": 0.4185467140042187, + "grad_norm": 0.140625, + "learning_rate": 0.0011084524134503159, + "loss": 0.0928, + "step": 48217 + }, + { + "epoch": 0.41855539448442286, + "grad_norm": 0.1650390625, + "learning_rate": 0.0011084226523015582, + "loss": 0.1133, + "step": 48218 + }, + { + "epoch": 0.41856407496462705, + "grad_norm": 0.173828125, + "learning_rate": 0.0011083928911435898, + "loss": 0.127, + "step": 48219 + }, + { + "epoch": 0.4185727554448312, + "grad_norm": 0.265625, + "learning_rate": 0.0011083631299764432, + "loss": 0.1328, + "step": 48220 + }, + { + "epoch": 0.4185814359250354, + "grad_norm": 0.55078125, + "learning_rate": 0.0011083333688001508, + "loss": 0.1113, + "step": 48221 + }, + { + "epoch": 0.4185901164052395, + "grad_norm": 0.5390625, + "learning_rate": 0.0011083036076147451, + "loss": 0.085, + "step": 48222 + }, + { + "epoch": 0.4185987968854437, + "grad_norm": 0.0810546875, + "learning_rate": 0.0011082738464202584, + "loss": 0.082, + "step": 48223 + }, + { + "epoch": 0.41860747736564785, + "grad_norm": 0.375, + "learning_rate": 0.0011082440852167239, + "loss": 0.106, + "step": 48224 + }, + { + "epoch": 0.41861615784585204, + "grad_norm": 0.10400390625, + "learning_rate": 0.0011082143240041737, + "loss": 0.1021, + "step": 48225 + }, + { + "epoch": 0.4186248383260562, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011081845627826405, + "loss": 0.1035, + "step": 48226 + }, + { + "epoch": 0.4186335188062604, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011081548015521568, + "loss": 0.0913, + "step": 48227 + }, + { + "epoch": 0.4186421992864645, + "grad_norm": 0.1728515625, + "learning_rate": 0.001108125040312755, + "loss": 0.0933, + "step": 48228 + }, + { + "epoch": 0.4186508797666687, + "grad_norm": 0.099609375, + "learning_rate": 0.0011080952790644676, + "loss": 0.0869, + "step": 48229 + }, + { + "epoch": 0.41865956024687284, + "grad_norm": 0.134765625, + "learning_rate": 0.0011080655178073277, + "loss": 0.124, + "step": 48230 + }, + { + "epoch": 0.41866824072707703, + "grad_norm": 0.296875, + "learning_rate": 0.0011080357565413671, + "loss": 0.0889, + "step": 48231 + }, + { + "epoch": 0.41867692120728117, + "grad_norm": 0.796875, + "learning_rate": 0.0011080059952666194, + "loss": 0.103, + "step": 48232 + }, + { + "epoch": 0.41868560168748536, + "grad_norm": 0.734375, + "learning_rate": 0.001107976233983116, + "loss": 0.0879, + "step": 48233 + }, + { + "epoch": 0.4186942821676895, + "grad_norm": 0.34765625, + "learning_rate": 0.00110794647269089, + "loss": 0.0996, + "step": 48234 + }, + { + "epoch": 0.4187029626478937, + "grad_norm": 0.2177734375, + "learning_rate": 0.0011079167113899736, + "loss": 0.0967, + "step": 48235 + }, + { + "epoch": 0.41871164312809783, + "grad_norm": 0.169921875, + "learning_rate": 0.0011078869500803995, + "loss": 0.0952, + "step": 48236 + }, + { + "epoch": 0.418720323608302, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011078571887622008, + "loss": 0.0879, + "step": 48237 + }, + { + "epoch": 0.41872900408850616, + "grad_norm": 0.4765625, + "learning_rate": 0.0011078274274354094, + "loss": 0.1182, + "step": 48238 + }, + { + "epoch": 0.41873768456871036, + "grad_norm": 0.54296875, + "learning_rate": 0.0011077976661000581, + "loss": 0.0952, + "step": 48239 + }, + { + "epoch": 0.4187463650489145, + "grad_norm": 0.51953125, + "learning_rate": 0.0011077679047561792, + "loss": 0.0723, + "step": 48240 + }, + { + "epoch": 0.4187550455291187, + "grad_norm": 0.1005859375, + "learning_rate": 0.0011077381434038058, + "loss": 0.1035, + "step": 48241 + }, + { + "epoch": 0.4187637260093228, + "grad_norm": 0.51171875, + "learning_rate": 0.00110770838204297, + "loss": 0.0996, + "step": 48242 + }, + { + "epoch": 0.418772406489527, + "grad_norm": 0.107421875, + "learning_rate": 0.0011076786206737042, + "loss": 0.0957, + "step": 48243 + }, + { + "epoch": 0.41878108696973115, + "grad_norm": 0.3203125, + "learning_rate": 0.0011076488592960414, + "loss": 0.0635, + "step": 48244 + }, + { + "epoch": 0.41878976744993535, + "grad_norm": 0.3671875, + "learning_rate": 0.0011076190979100139, + "loss": 0.0593, + "step": 48245 + }, + { + "epoch": 0.4187984479301395, + "grad_norm": 0.08447265625, + "learning_rate": 0.0011075893365156543, + "loss": 0.0903, + "step": 48246 + }, + { + "epoch": 0.4188071284103437, + "grad_norm": 0.259765625, + "learning_rate": 0.0011075595751129945, + "loss": 0.0996, + "step": 48247 + }, + { + "epoch": 0.4188158088905478, + "grad_norm": 0.109375, + "learning_rate": 0.0011075298137020685, + "loss": 0.1143, + "step": 48248 + }, + { + "epoch": 0.418824489370752, + "grad_norm": 0.146484375, + "learning_rate": 0.0011075000522829076, + "loss": 0.0737, + "step": 48249 + }, + { + "epoch": 0.41883316985095614, + "grad_norm": 0.107421875, + "learning_rate": 0.0011074702908555447, + "loss": 0.0825, + "step": 48250 + }, + { + "epoch": 0.41884185033116034, + "grad_norm": 0.087890625, + "learning_rate": 0.0011074405294200125, + "loss": 0.0996, + "step": 48251 + }, + { + "epoch": 0.4188505308113645, + "grad_norm": 0.5859375, + "learning_rate": 0.0011074107679763436, + "loss": 0.0908, + "step": 48252 + }, + { + "epoch": 0.41885921129156867, + "grad_norm": 0.2353515625, + "learning_rate": 0.0011073810065245701, + "loss": 0.0703, + "step": 48253 + }, + { + "epoch": 0.4188678917717728, + "grad_norm": 0.2734375, + "learning_rate": 0.0011073512450647251, + "loss": 0.1113, + "step": 48254 + }, + { + "epoch": 0.418876572251977, + "grad_norm": 0.10986328125, + "learning_rate": 0.0011073214835968408, + "loss": 0.0664, + "step": 48255 + }, + { + "epoch": 0.41888525273218113, + "grad_norm": 0.35546875, + "learning_rate": 0.0011072917221209497, + "loss": 0.1416, + "step": 48256 + }, + { + "epoch": 0.4188939332123853, + "grad_norm": 0.83203125, + "learning_rate": 0.0011072619606370849, + "loss": 0.1016, + "step": 48257 + }, + { + "epoch": 0.41890261369258946, + "grad_norm": 0.35546875, + "learning_rate": 0.0011072321991452778, + "loss": 0.0952, + "step": 48258 + }, + { + "epoch": 0.41891129417279366, + "grad_norm": 0.328125, + "learning_rate": 0.0011072024376455622, + "loss": 0.1289, + "step": 48259 + }, + { + "epoch": 0.4189199746529978, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011071726761379703, + "loss": 0.0913, + "step": 48260 + }, + { + "epoch": 0.418928655133202, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011071429146225342, + "loss": 0.1309, + "step": 48261 + }, + { + "epoch": 0.4189373356134061, + "grad_norm": 0.53125, + "learning_rate": 0.0011071131530992866, + "loss": 0.1108, + "step": 48262 + }, + { + "epoch": 0.4189460160936103, + "grad_norm": 0.087890625, + "learning_rate": 0.0011070833915682603, + "loss": 0.1025, + "step": 48263 + }, + { + "epoch": 0.41895469657381446, + "grad_norm": 0.2216796875, + "learning_rate": 0.0011070536300294878, + "loss": 0.0825, + "step": 48264 + }, + { + "epoch": 0.41896337705401865, + "grad_norm": 0.1162109375, + "learning_rate": 0.0011070238684830016, + "loss": 0.1182, + "step": 48265 + }, + { + "epoch": 0.4189720575342228, + "grad_norm": 0.185546875, + "learning_rate": 0.0011069941069288339, + "loss": 0.0952, + "step": 48266 + }, + { + "epoch": 0.418980738014427, + "grad_norm": 0.16796875, + "learning_rate": 0.0011069643453670175, + "loss": 0.0879, + "step": 48267 + }, + { + "epoch": 0.4189894184946311, + "grad_norm": 0.09130859375, + "learning_rate": 0.0011069345837975855, + "loss": 0.127, + "step": 48268 + }, + { + "epoch": 0.4189980989748353, + "grad_norm": 0.51953125, + "learning_rate": 0.0011069048222205695, + "loss": 0.166, + "step": 48269 + }, + { + "epoch": 0.41900677945503945, + "grad_norm": 0.111328125, + "learning_rate": 0.0011068750606360025, + "loss": 0.084, + "step": 48270 + }, + { + "epoch": 0.41901545993524364, + "grad_norm": 0.294921875, + "learning_rate": 0.0011068452990439172, + "loss": 0.1504, + "step": 48271 + }, + { + "epoch": 0.4190241404154478, + "grad_norm": 0.134765625, + "learning_rate": 0.0011068155374443463, + "loss": 0.0972, + "step": 48272 + }, + { + "epoch": 0.41903282089565197, + "grad_norm": 0.41015625, + "learning_rate": 0.0011067857758373214, + "loss": 0.1045, + "step": 48273 + }, + { + "epoch": 0.4190415013758561, + "grad_norm": 0.52734375, + "learning_rate": 0.0011067560142228763, + "loss": 0.1074, + "step": 48274 + }, + { + "epoch": 0.4190501818560603, + "grad_norm": 0.228515625, + "learning_rate": 0.0011067262526010424, + "loss": 0.0635, + "step": 48275 + }, + { + "epoch": 0.41905886233626444, + "grad_norm": 0.1669921875, + "learning_rate": 0.0011066964909718527, + "loss": 0.0796, + "step": 48276 + }, + { + "epoch": 0.41906754281646863, + "grad_norm": 0.8125, + "learning_rate": 0.0011066667293353402, + "loss": 0.1289, + "step": 48277 + }, + { + "epoch": 0.41907622329667277, + "grad_norm": 0.408203125, + "learning_rate": 0.0011066369676915372, + "loss": 0.0957, + "step": 48278 + }, + { + "epoch": 0.41908490377687696, + "grad_norm": 0.26171875, + "learning_rate": 0.0011066072060404758, + "loss": 0.1123, + "step": 48279 + }, + { + "epoch": 0.4190935842570811, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011065774443821887, + "loss": 0.1172, + "step": 48280 + }, + { + "epoch": 0.4191022647372853, + "grad_norm": 0.1416015625, + "learning_rate": 0.001106547682716709, + "loss": 0.0996, + "step": 48281 + }, + { + "epoch": 0.41911094521748943, + "grad_norm": 0.12890625, + "learning_rate": 0.0011065179210440687, + "loss": 0.1074, + "step": 48282 + }, + { + "epoch": 0.4191196256976936, + "grad_norm": 0.2890625, + "learning_rate": 0.0011064881593643003, + "loss": 0.085, + "step": 48283 + }, + { + "epoch": 0.41912830617789776, + "grad_norm": 0.11865234375, + "learning_rate": 0.0011064583976774368, + "loss": 0.0781, + "step": 48284 + }, + { + "epoch": 0.41913698665810195, + "grad_norm": 0.2578125, + "learning_rate": 0.0011064286359835102, + "loss": 0.1348, + "step": 48285 + }, + { + "epoch": 0.4191456671383061, + "grad_norm": 0.33984375, + "learning_rate": 0.0011063988742825537, + "loss": 0.1021, + "step": 48286 + }, + { + "epoch": 0.4191543476185103, + "grad_norm": 1.140625, + "learning_rate": 0.0011063691125745994, + "loss": 0.0938, + "step": 48287 + }, + { + "epoch": 0.4191630280987144, + "grad_norm": 0.70703125, + "learning_rate": 0.0011063393508596797, + "loss": 0.125, + "step": 48288 + }, + { + "epoch": 0.4191717085789186, + "grad_norm": 0.337890625, + "learning_rate": 0.0011063095891378277, + "loss": 0.1504, + "step": 48289 + }, + { + "epoch": 0.41918038905912275, + "grad_norm": 0.2734375, + "learning_rate": 0.0011062798274090754, + "loss": 0.1035, + "step": 48290 + }, + { + "epoch": 0.41918906953932694, + "grad_norm": 0.72265625, + "learning_rate": 0.001106250065673456, + "loss": 0.1484, + "step": 48291 + }, + { + "epoch": 0.4191977500195311, + "grad_norm": 0.58203125, + "learning_rate": 0.001106220303931001, + "loss": 0.0747, + "step": 48292 + }, + { + "epoch": 0.41920643049973527, + "grad_norm": 0.072265625, + "learning_rate": 0.001106190542181744, + "loss": 0.0967, + "step": 48293 + }, + { + "epoch": 0.4192151109799394, + "grad_norm": 0.10400390625, + "learning_rate": 0.001106160780425717, + "loss": 0.0762, + "step": 48294 + }, + { + "epoch": 0.4192237914601436, + "grad_norm": 0.52734375, + "learning_rate": 0.0011061310186629525, + "loss": 0.082, + "step": 48295 + }, + { + "epoch": 0.41923247194034774, + "grad_norm": 0.138671875, + "learning_rate": 0.0011061012568934832, + "loss": 0.0986, + "step": 48296 + }, + { + "epoch": 0.41924115242055193, + "grad_norm": 0.189453125, + "learning_rate": 0.0011060714951173418, + "loss": 0.1104, + "step": 48297 + }, + { + "epoch": 0.41924983290075607, + "grad_norm": 0.11328125, + "learning_rate": 0.0011060417333345607, + "loss": 0.0933, + "step": 48298 + }, + { + "epoch": 0.4192585133809602, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011060119715451727, + "loss": 0.0894, + "step": 48299 + }, + { + "epoch": 0.4192671938611644, + "grad_norm": 0.498046875, + "learning_rate": 0.00110598220974921, + "loss": 0.0806, + "step": 48300 + }, + { + "epoch": 0.41927587434136854, + "grad_norm": 0.42578125, + "learning_rate": 0.001105952447946705, + "loss": 0.0942, + "step": 48301 + }, + { + "epoch": 0.41928455482157273, + "grad_norm": 0.138671875, + "learning_rate": 0.0011059226861376906, + "loss": 0.0947, + "step": 48302 + }, + { + "epoch": 0.41929323530177687, + "grad_norm": 0.140625, + "learning_rate": 0.0011058929243221993, + "loss": 0.1318, + "step": 48303 + }, + { + "epoch": 0.41930191578198106, + "grad_norm": 0.173828125, + "learning_rate": 0.0011058631625002634, + "loss": 0.0972, + "step": 48304 + }, + { + "epoch": 0.4193105962621852, + "grad_norm": 0.1435546875, + "learning_rate": 0.0011058334006719154, + "loss": 0.1387, + "step": 48305 + }, + { + "epoch": 0.4193192767423894, + "grad_norm": 0.283203125, + "learning_rate": 0.0011058036388371887, + "loss": 0.0967, + "step": 48306 + }, + { + "epoch": 0.41932795722259353, + "grad_norm": 0.2265625, + "learning_rate": 0.0011057738769961148, + "loss": 0.1177, + "step": 48307 + }, + { + "epoch": 0.4193366377027977, + "grad_norm": 0.44921875, + "learning_rate": 0.001105744115148727, + "loss": 0.082, + "step": 48308 + }, + { + "epoch": 0.41934531818300186, + "grad_norm": 0.146484375, + "learning_rate": 0.001105714353295057, + "loss": 0.0664, + "step": 48309 + }, + { + "epoch": 0.41935399866320605, + "grad_norm": 0.06884765625, + "learning_rate": 0.0011056845914351385, + "loss": 0.0669, + "step": 48310 + }, + { + "epoch": 0.4193626791434102, + "grad_norm": 0.10595703125, + "learning_rate": 0.0011056548295690033, + "loss": 0.1016, + "step": 48311 + }, + { + "epoch": 0.4193713596236144, + "grad_norm": 0.390625, + "learning_rate": 0.0011056250676966836, + "loss": 0.1074, + "step": 48312 + }, + { + "epoch": 0.4193800401038185, + "grad_norm": 0.271484375, + "learning_rate": 0.0011055953058182128, + "loss": 0.0942, + "step": 48313 + }, + { + "epoch": 0.4193887205840227, + "grad_norm": 0.439453125, + "learning_rate": 0.0011055655439336232, + "loss": 0.1016, + "step": 48314 + }, + { + "epoch": 0.41939740106422685, + "grad_norm": 0.21875, + "learning_rate": 0.0011055357820429468, + "loss": 0.083, + "step": 48315 + }, + { + "epoch": 0.41940608154443104, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011055060201462165, + "loss": 0.1318, + "step": 48316 + }, + { + "epoch": 0.4194147620246352, + "grad_norm": 0.37109375, + "learning_rate": 0.0011054762582434654, + "loss": 0.123, + "step": 48317 + }, + { + "epoch": 0.41942344250483937, + "grad_norm": 1.0234375, + "learning_rate": 0.001105446496334725, + "loss": 0.1084, + "step": 48318 + }, + { + "epoch": 0.4194321229850435, + "grad_norm": 0.10009765625, + "learning_rate": 0.001105416734420029, + "loss": 0.0781, + "step": 48319 + }, + { + "epoch": 0.4194408034652477, + "grad_norm": 0.2197265625, + "learning_rate": 0.001105386972499409, + "loss": 0.1152, + "step": 48320 + }, + { + "epoch": 0.41944948394545184, + "grad_norm": 0.1943359375, + "learning_rate": 0.001105357210572898, + "loss": 0.0928, + "step": 48321 + }, + { + "epoch": 0.41945816442565603, + "grad_norm": 0.06787109375, + "learning_rate": 0.001105327448640528, + "loss": 0.0664, + "step": 48322 + }, + { + "epoch": 0.41946684490586017, + "grad_norm": 0.1181640625, + "learning_rate": 0.0011052976867023325, + "loss": 0.1367, + "step": 48323 + }, + { + "epoch": 0.41947552538606436, + "grad_norm": 0.2041015625, + "learning_rate": 0.0011052679247583433, + "loss": 0.1045, + "step": 48324 + }, + { + "epoch": 0.4194842058662685, + "grad_norm": 0.310546875, + "learning_rate": 0.0011052381628085931, + "loss": 0.1069, + "step": 48325 + }, + { + "epoch": 0.4194928863464727, + "grad_norm": 0.390625, + "learning_rate": 0.0011052084008531147, + "loss": 0.0923, + "step": 48326 + }, + { + "epoch": 0.41950156682667683, + "grad_norm": 0.1357421875, + "learning_rate": 0.0011051786388919403, + "loss": 0.125, + "step": 48327 + }, + { + "epoch": 0.419510247306881, + "grad_norm": 0.23828125, + "learning_rate": 0.001105148876925103, + "loss": 0.1226, + "step": 48328 + }, + { + "epoch": 0.41951892778708516, + "grad_norm": 0.38671875, + "learning_rate": 0.0011051191149526348, + "loss": 0.0933, + "step": 48329 + }, + { + "epoch": 0.41952760826728935, + "grad_norm": 0.0908203125, + "learning_rate": 0.0011050893529745683, + "loss": 0.0977, + "step": 48330 + }, + { + "epoch": 0.4195362887474935, + "grad_norm": 0.07373046875, + "learning_rate": 0.001105059590990936, + "loss": 0.0752, + "step": 48331 + }, + { + "epoch": 0.4195449692276977, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011050298290017709, + "loss": 0.0991, + "step": 48332 + }, + { + "epoch": 0.4195536497079018, + "grad_norm": 0.1064453125, + "learning_rate": 0.001105000067007105, + "loss": 0.1084, + "step": 48333 + }, + { + "epoch": 0.419562330188106, + "grad_norm": 0.23046875, + "learning_rate": 0.0011049703050069711, + "loss": 0.1348, + "step": 48334 + }, + { + "epoch": 0.41957101066831015, + "grad_norm": 0.28125, + "learning_rate": 0.0011049405430014019, + "loss": 0.1113, + "step": 48335 + }, + { + "epoch": 0.41957969114851434, + "grad_norm": 0.10888671875, + "learning_rate": 0.0011049107809904298, + "loss": 0.1729, + "step": 48336 + }, + { + "epoch": 0.4195883716287185, + "grad_norm": 0.70703125, + "learning_rate": 0.0011048810189740875, + "loss": 0.0991, + "step": 48337 + }, + { + "epoch": 0.4195970521089227, + "grad_norm": 0.1396484375, + "learning_rate": 0.0011048512569524073, + "loss": 0.0977, + "step": 48338 + }, + { + "epoch": 0.4196057325891268, + "grad_norm": 0.138671875, + "learning_rate": 0.0011048214949254216, + "loss": 0.1118, + "step": 48339 + }, + { + "epoch": 0.419614413069331, + "grad_norm": 0.0673828125, + "learning_rate": 0.0011047917328931632, + "loss": 0.0918, + "step": 48340 + }, + { + "epoch": 0.41962309354953514, + "grad_norm": 0.232421875, + "learning_rate": 0.0011047619708556649, + "loss": 0.1113, + "step": 48341 + }, + { + "epoch": 0.41963177402973934, + "grad_norm": 0.357421875, + "learning_rate": 0.001104732208812959, + "loss": 0.1523, + "step": 48342 + }, + { + "epoch": 0.4196404545099435, + "grad_norm": 0.1875, + "learning_rate": 0.0011047024467650778, + "loss": 0.0825, + "step": 48343 + }, + { + "epoch": 0.41964913499014767, + "grad_norm": 0.0986328125, + "learning_rate": 0.0011046726847120539, + "loss": 0.0986, + "step": 48344 + }, + { + "epoch": 0.4196578154703518, + "grad_norm": 0.146484375, + "learning_rate": 0.0011046429226539204, + "loss": 0.0811, + "step": 48345 + }, + { + "epoch": 0.419666495950556, + "grad_norm": 0.2431640625, + "learning_rate": 0.0011046131605907095, + "loss": 0.0762, + "step": 48346 + }, + { + "epoch": 0.41967517643076013, + "grad_norm": 0.26953125, + "learning_rate": 0.0011045833985224537, + "loss": 0.0908, + "step": 48347 + }, + { + "epoch": 0.4196838569109643, + "grad_norm": 0.1142578125, + "learning_rate": 0.0011045536364491851, + "loss": 0.0952, + "step": 48348 + }, + { + "epoch": 0.41969253739116846, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011045238743709372, + "loss": 0.125, + "step": 48349 + }, + { + "epoch": 0.41970121787137266, + "grad_norm": 0.28125, + "learning_rate": 0.001104494112287742, + "loss": 0.0869, + "step": 48350 + }, + { + "epoch": 0.4197098983515768, + "grad_norm": 0.2041015625, + "learning_rate": 0.001104464350199632, + "loss": 0.127, + "step": 48351 + }, + { + "epoch": 0.419718578831781, + "grad_norm": 0.177734375, + "learning_rate": 0.0011044345881066398, + "loss": 0.106, + "step": 48352 + }, + { + "epoch": 0.4197272593119851, + "grad_norm": 0.322265625, + "learning_rate": 0.0011044048260087978, + "loss": 0.0635, + "step": 48353 + }, + { + "epoch": 0.4197359397921893, + "grad_norm": 0.1611328125, + "learning_rate": 0.0011043750639061392, + "loss": 0.0708, + "step": 48354 + }, + { + "epoch": 0.41974462027239345, + "grad_norm": 0.68359375, + "learning_rate": 0.0011043453017986957, + "loss": 0.0762, + "step": 48355 + }, + { + "epoch": 0.41975330075259765, + "grad_norm": 0.1083984375, + "learning_rate": 0.0011043155396865006, + "loss": 0.1367, + "step": 48356 + }, + { + "epoch": 0.4197619812328018, + "grad_norm": 0.91015625, + "learning_rate": 0.0011042857775695857, + "loss": 0.1992, + "step": 48357 + }, + { + "epoch": 0.419770661713006, + "grad_norm": 0.1533203125, + "learning_rate": 0.0011042560154479843, + "loss": 0.0947, + "step": 48358 + }, + { + "epoch": 0.4197793421932101, + "grad_norm": 0.11962890625, + "learning_rate": 0.0011042262533217286, + "loss": 0.1035, + "step": 48359 + }, + { + "epoch": 0.4197880226734143, + "grad_norm": 0.09619140625, + "learning_rate": 0.0011041964911908508, + "loss": 0.1099, + "step": 48360 + }, + { + "epoch": 0.41979670315361844, + "grad_norm": 0.0966796875, + "learning_rate": 0.0011041667290553842, + "loss": 0.1089, + "step": 48361 + }, + { + "epoch": 0.41980538363382264, + "grad_norm": 0.1318359375, + "learning_rate": 0.0011041369669153604, + "loss": 0.0752, + "step": 48362 + }, + { + "epoch": 0.4198140641140268, + "grad_norm": 0.2255859375, + "learning_rate": 0.001104107204770813, + "loss": 0.0703, + "step": 48363 + }, + { + "epoch": 0.41982274459423097, + "grad_norm": 0.5625, + "learning_rate": 0.0011040774426217738, + "loss": 0.1162, + "step": 48364 + }, + { + "epoch": 0.4198314250744351, + "grad_norm": 0.244140625, + "learning_rate": 0.0011040476804682754, + "loss": 0.085, + "step": 48365 + }, + { + "epoch": 0.4198401055546393, + "grad_norm": 0.138671875, + "learning_rate": 0.0011040179183103506, + "loss": 0.0957, + "step": 48366 + }, + { + "epoch": 0.41984878603484344, + "grad_norm": 0.13671875, + "learning_rate": 0.001103988156148032, + "loss": 0.0684, + "step": 48367 + }, + { + "epoch": 0.41985746651504763, + "grad_norm": 0.12890625, + "learning_rate": 0.0011039583939813523, + "loss": 0.0996, + "step": 48368 + }, + { + "epoch": 0.41986614699525177, + "grad_norm": 0.119140625, + "learning_rate": 0.0011039286318103435, + "loss": 0.0986, + "step": 48369 + }, + { + "epoch": 0.41987482747545596, + "grad_norm": 0.65234375, + "learning_rate": 0.001103898869635038, + "loss": 0.1367, + "step": 48370 + }, + { + "epoch": 0.4198835079556601, + "grad_norm": 0.126953125, + "learning_rate": 0.0011038691074554692, + "loss": 0.082, + "step": 48371 + }, + { + "epoch": 0.4198921884358643, + "grad_norm": 0.18359375, + "learning_rate": 0.0011038393452716688, + "loss": 0.1396, + "step": 48372 + }, + { + "epoch": 0.4199008689160684, + "grad_norm": 0.08349609375, + "learning_rate": 0.0011038095830836704, + "loss": 0.1035, + "step": 48373 + }, + { + "epoch": 0.4199095493962726, + "grad_norm": 0.09814453125, + "learning_rate": 0.0011037798208915052, + "loss": 0.0923, + "step": 48374 + }, + { + "epoch": 0.41991822987647676, + "grad_norm": 0.49609375, + "learning_rate": 0.001103750058695207, + "loss": 0.1084, + "step": 48375 + }, + { + "epoch": 0.41992691035668095, + "grad_norm": 0.310546875, + "learning_rate": 0.0011037202964948077, + "loss": 0.1011, + "step": 48376 + }, + { + "epoch": 0.4199355908368851, + "grad_norm": 0.1904296875, + "learning_rate": 0.0011036905342903399, + "loss": 0.0942, + "step": 48377 + }, + { + "epoch": 0.4199442713170893, + "grad_norm": 0.26953125, + "learning_rate": 0.0011036607720818358, + "loss": 0.1133, + "step": 48378 + }, + { + "epoch": 0.4199529517972934, + "grad_norm": 0.08251953125, + "learning_rate": 0.0011036310098693285, + "loss": 0.0757, + "step": 48379 + }, + { + "epoch": 0.4199616322774976, + "grad_norm": 0.40234375, + "learning_rate": 0.0011036012476528508, + "loss": 0.0786, + "step": 48380 + }, + { + "epoch": 0.41997031275770175, + "grad_norm": 0.1728515625, + "learning_rate": 0.0011035714854324343, + "loss": 0.1172, + "step": 48381 + }, + { + "epoch": 0.41997899323790594, + "grad_norm": 0.197265625, + "learning_rate": 0.0011035417232081125, + "loss": 0.1582, + "step": 48382 + }, + { + "epoch": 0.4199876737181101, + "grad_norm": 0.10498046875, + "learning_rate": 0.0011035119609799171, + "loss": 0.1211, + "step": 48383 + }, + { + "epoch": 0.41999635419831427, + "grad_norm": 0.279296875, + "learning_rate": 0.0011034821987478815, + "loss": 0.0947, + "step": 48384 + }, + { + "epoch": 0.4200050346785184, + "grad_norm": 0.263671875, + "learning_rate": 0.0011034524365120374, + "loss": 0.0615, + "step": 48385 + }, + { + "epoch": 0.4200137151587226, + "grad_norm": 0.390625, + "learning_rate": 0.0011034226742724185, + "loss": 0.0728, + "step": 48386 + }, + { + "epoch": 0.42002239563892674, + "grad_norm": 1.7109375, + "learning_rate": 0.0011033929120290557, + "loss": 0.2715, + "step": 48387 + }, + { + "epoch": 0.42003107611913093, + "grad_norm": 0.326171875, + "learning_rate": 0.0011033631497819833, + "loss": 0.1162, + "step": 48388 + }, + { + "epoch": 0.42003975659933507, + "grad_norm": 0.279296875, + "learning_rate": 0.0011033333875312323, + "loss": 0.0889, + "step": 48389 + }, + { + "epoch": 0.42004843707953926, + "grad_norm": 0.486328125, + "learning_rate": 0.0011033036252768363, + "loss": 0.1104, + "step": 48390 + }, + { + "epoch": 0.4200571175597434, + "grad_norm": 0.263671875, + "learning_rate": 0.0011032738630188275, + "loss": 0.0942, + "step": 48391 + }, + { + "epoch": 0.4200657980399476, + "grad_norm": 0.08544921875, + "learning_rate": 0.0011032441007572382, + "loss": 0.1035, + "step": 48392 + }, + { + "epoch": 0.42007447852015173, + "grad_norm": 0.197265625, + "learning_rate": 0.0011032143384921015, + "loss": 0.1094, + "step": 48393 + }, + { + "epoch": 0.4200831590003559, + "grad_norm": 0.71875, + "learning_rate": 0.0011031845762234495, + "loss": 0.0947, + "step": 48394 + }, + { + "epoch": 0.42009183948056006, + "grad_norm": 0.83203125, + "learning_rate": 0.001103154813951315, + "loss": 0.2344, + "step": 48395 + }, + { + "epoch": 0.42010051996076425, + "grad_norm": 0.09326171875, + "learning_rate": 0.0011031250516757305, + "loss": 0.127, + "step": 48396 + }, + { + "epoch": 0.4201092004409684, + "grad_norm": 0.427734375, + "learning_rate": 0.0011030952893967283, + "loss": 0.1064, + "step": 48397 + }, + { + "epoch": 0.4201178809211726, + "grad_norm": 0.107421875, + "learning_rate": 0.0011030655271143412, + "loss": 0.0918, + "step": 48398 + }, + { + "epoch": 0.4201265614013767, + "grad_norm": 0.287109375, + "learning_rate": 0.001103035764828602, + "loss": 0.0874, + "step": 48399 + }, + { + "epoch": 0.4201352418815809, + "grad_norm": 0.75, + "learning_rate": 0.0011030060025395423, + "loss": 0.249, + "step": 48400 + }, + { + "epoch": 0.42014392236178505, + "grad_norm": 0.23046875, + "learning_rate": 0.0011029762402471957, + "loss": 0.0718, + "step": 48401 + }, + { + "epoch": 0.42015260284198924, + "grad_norm": 0.06689453125, + "learning_rate": 0.0011029464779515944, + "loss": 0.0596, + "step": 48402 + }, + { + "epoch": 0.4201612833221934, + "grad_norm": 0.408203125, + "learning_rate": 0.0011029167156527706, + "loss": 0.0786, + "step": 48403 + }, + { + "epoch": 0.4201699638023976, + "grad_norm": 0.90234375, + "learning_rate": 0.001102886953350757, + "loss": 0.2012, + "step": 48404 + }, + { + "epoch": 0.4201786442826017, + "grad_norm": 0.8515625, + "learning_rate": 0.0011028571910455867, + "loss": 0.1338, + "step": 48405 + }, + { + "epoch": 0.4201873247628059, + "grad_norm": 0.2080078125, + "learning_rate": 0.0011028274287372918, + "loss": 0.085, + "step": 48406 + }, + { + "epoch": 0.42019600524301004, + "grad_norm": 0.30078125, + "learning_rate": 0.0011027976664259047, + "loss": 0.1064, + "step": 48407 + }, + { + "epoch": 0.42020468572321423, + "grad_norm": 0.2373046875, + "learning_rate": 0.0011027679041114582, + "loss": 0.0923, + "step": 48408 + }, + { + "epoch": 0.42021336620341837, + "grad_norm": 0.07275390625, + "learning_rate": 0.0011027381417939846, + "loss": 0.064, + "step": 48409 + }, + { + "epoch": 0.42022204668362256, + "grad_norm": 0.09228515625, + "learning_rate": 0.0011027083794735164, + "loss": 0.1001, + "step": 48410 + }, + { + "epoch": 0.4202307271638267, + "grad_norm": 0.123046875, + "learning_rate": 0.0011026786171500868, + "loss": 0.0879, + "step": 48411 + }, + { + "epoch": 0.4202394076440309, + "grad_norm": 0.13671875, + "learning_rate": 0.001102648854823728, + "loss": 0.1079, + "step": 48412 + }, + { + "epoch": 0.42024808812423503, + "grad_norm": 0.4609375, + "learning_rate": 0.001102619092494472, + "loss": 0.1641, + "step": 48413 + }, + { + "epoch": 0.4202567686044392, + "grad_norm": 0.296875, + "learning_rate": 0.001102589330162352, + "loss": 0.0811, + "step": 48414 + }, + { + "epoch": 0.42026544908464336, + "grad_norm": 0.373046875, + "learning_rate": 0.0011025595678274003, + "loss": 0.0957, + "step": 48415 + }, + { + "epoch": 0.42027412956484755, + "grad_norm": 0.1943359375, + "learning_rate": 0.0011025298054896495, + "loss": 0.0728, + "step": 48416 + }, + { + "epoch": 0.4202828100450517, + "grad_norm": 0.10693359375, + "learning_rate": 0.0011025000431491322, + "loss": 0.1006, + "step": 48417 + }, + { + "epoch": 0.4202914905252559, + "grad_norm": 0.1552734375, + "learning_rate": 0.001102470280805881, + "loss": 0.1074, + "step": 48418 + }, + { + "epoch": 0.42030017100546, + "grad_norm": 0.404296875, + "learning_rate": 0.0011024405184599283, + "loss": 0.1201, + "step": 48419 + }, + { + "epoch": 0.4203088514856642, + "grad_norm": 5.46875, + "learning_rate": 0.0011024107561113064, + "loss": 0.3164, + "step": 48420 + }, + { + "epoch": 0.42031753196586835, + "grad_norm": 0.23828125, + "learning_rate": 0.0011023809937600485, + "loss": 0.1152, + "step": 48421 + }, + { + "epoch": 0.4203262124460725, + "grad_norm": 0.1171875, + "learning_rate": 0.0011023512314061866, + "loss": 0.0635, + "step": 48422 + }, + { + "epoch": 0.4203348929262767, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011023214690497533, + "loss": 0.0938, + "step": 48423 + }, + { + "epoch": 0.4203435734064808, + "grad_norm": 0.26953125, + "learning_rate": 0.0011022917066907817, + "loss": 0.1523, + "step": 48424 + }, + { + "epoch": 0.420352253886685, + "grad_norm": 0.1494140625, + "learning_rate": 0.0011022619443293036, + "loss": 0.1279, + "step": 48425 + }, + { + "epoch": 0.42036093436688915, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011022321819653515, + "loss": 0.0986, + "step": 48426 + }, + { + "epoch": 0.42036961484709334, + "grad_norm": 3.1875, + "learning_rate": 0.0011022024195989591, + "loss": 0.1719, + "step": 48427 + }, + { + "epoch": 0.4203782953272975, + "grad_norm": 0.2275390625, + "learning_rate": 0.0011021726572301579, + "loss": 0.1357, + "step": 48428 + }, + { + "epoch": 0.4203869758075017, + "grad_norm": 0.2265625, + "learning_rate": 0.0011021428948589805, + "loss": 0.105, + "step": 48429 + }, + { + "epoch": 0.4203956562877058, + "grad_norm": 0.1865234375, + "learning_rate": 0.0011021131324854595, + "loss": 0.1143, + "step": 48430 + }, + { + "epoch": 0.42040433676791, + "grad_norm": 0.20703125, + "learning_rate": 0.001102083370109628, + "loss": 0.0898, + "step": 48431 + }, + { + "epoch": 0.42041301724811414, + "grad_norm": 0.2197265625, + "learning_rate": 0.0011020536077315183, + "loss": 0.1104, + "step": 48432 + }, + { + "epoch": 0.42042169772831833, + "grad_norm": 0.0869140625, + "learning_rate": 0.0011020238453511625, + "loss": 0.0908, + "step": 48433 + }, + { + "epoch": 0.42043037820852247, + "grad_norm": 0.181640625, + "learning_rate": 0.0011019940829685934, + "loss": 0.1143, + "step": 48434 + }, + { + "epoch": 0.42043905868872666, + "grad_norm": 0.14453125, + "learning_rate": 0.0011019643205838437, + "loss": 0.0552, + "step": 48435 + }, + { + "epoch": 0.4204477391689308, + "grad_norm": 0.18359375, + "learning_rate": 0.001101934558196946, + "loss": 0.0869, + "step": 48436 + }, + { + "epoch": 0.420456419649135, + "grad_norm": 0.75390625, + "learning_rate": 0.0011019047958079325, + "loss": 0.0952, + "step": 48437 + }, + { + "epoch": 0.42046510012933913, + "grad_norm": 0.34375, + "learning_rate": 0.001101875033416836, + "loss": 0.083, + "step": 48438 + }, + { + "epoch": 0.4204737806095433, + "grad_norm": 0.255859375, + "learning_rate": 0.0011018452710236891, + "loss": 0.0889, + "step": 48439 + }, + { + "epoch": 0.42048246108974746, + "grad_norm": 0.298828125, + "learning_rate": 0.0011018155086285238, + "loss": 0.1025, + "step": 48440 + }, + { + "epoch": 0.42049114156995165, + "grad_norm": 0.490234375, + "learning_rate": 0.0011017857462313734, + "loss": 0.1211, + "step": 48441 + }, + { + "epoch": 0.4204998220501558, + "grad_norm": 0.294921875, + "learning_rate": 0.0011017559838322705, + "loss": 0.0791, + "step": 48442 + }, + { + "epoch": 0.42050850253036, + "grad_norm": 0.75, + "learning_rate": 0.0011017262214312467, + "loss": 0.103, + "step": 48443 + }, + { + "epoch": 0.4205171830105641, + "grad_norm": 0.4609375, + "learning_rate": 0.0011016964590283354, + "loss": 0.1104, + "step": 48444 + }, + { + "epoch": 0.4205258634907683, + "grad_norm": 0.26953125, + "learning_rate": 0.0011016666966235688, + "loss": 0.0903, + "step": 48445 + }, + { + "epoch": 0.42053454397097245, + "grad_norm": 0.1064453125, + "learning_rate": 0.0011016369342169798, + "loss": 0.0928, + "step": 48446 + }, + { + "epoch": 0.42054322445117664, + "grad_norm": 0.1826171875, + "learning_rate": 0.0011016071718086, + "loss": 0.1094, + "step": 48447 + }, + { + "epoch": 0.4205519049313808, + "grad_norm": 0.296875, + "learning_rate": 0.0011015774093984633, + "loss": 0.125, + "step": 48448 + }, + { + "epoch": 0.420560585411585, + "grad_norm": 0.2265625, + "learning_rate": 0.0011015476469866013, + "loss": 0.1113, + "step": 48449 + }, + { + "epoch": 0.4205692658917891, + "grad_norm": 0.08447265625, + "learning_rate": 0.001101517884573047, + "loss": 0.0747, + "step": 48450 + }, + { + "epoch": 0.4205779463719933, + "grad_norm": 0.369140625, + "learning_rate": 0.0011014881221578325, + "loss": 0.0811, + "step": 48451 + }, + { + "epoch": 0.42058662685219744, + "grad_norm": 0.13671875, + "learning_rate": 0.0011014583597409907, + "loss": 0.1035, + "step": 48452 + }, + { + "epoch": 0.42059530733240164, + "grad_norm": 0.703125, + "learning_rate": 0.001101428597322554, + "loss": 0.0811, + "step": 48453 + }, + { + "epoch": 0.4206039878126058, + "grad_norm": 0.369140625, + "learning_rate": 0.0011013988349025553, + "loss": 0.0938, + "step": 48454 + }, + { + "epoch": 0.42061266829280997, + "grad_norm": 0.5234375, + "learning_rate": 0.0011013690724810267, + "loss": 0.0938, + "step": 48455 + }, + { + "epoch": 0.4206213487730141, + "grad_norm": 0.5703125, + "learning_rate": 0.0011013393100580008, + "loss": 0.0967, + "step": 48456 + }, + { + "epoch": 0.4206300292532183, + "grad_norm": 0.58984375, + "learning_rate": 0.0011013095476335101, + "loss": 0.0732, + "step": 48457 + }, + { + "epoch": 0.42063870973342243, + "grad_norm": 0.2431640625, + "learning_rate": 0.0011012797852075872, + "loss": 0.1113, + "step": 48458 + }, + { + "epoch": 0.4206473902136266, + "grad_norm": 0.1337890625, + "learning_rate": 0.0011012500227802653, + "loss": 0.0918, + "step": 48459 + }, + { + "epoch": 0.42065607069383076, + "grad_norm": 0.498046875, + "learning_rate": 0.0011012202603515762, + "loss": 0.1357, + "step": 48460 + }, + { + "epoch": 0.42066475117403496, + "grad_norm": 0.259765625, + "learning_rate": 0.0011011904979215527, + "loss": 0.0708, + "step": 48461 + }, + { + "epoch": 0.4206734316542391, + "grad_norm": 0.1767578125, + "learning_rate": 0.0011011607354902268, + "loss": 0.1025, + "step": 48462 + }, + { + "epoch": 0.4206821121344433, + "grad_norm": 0.181640625, + "learning_rate": 0.0011011309730576324, + "loss": 0.165, + "step": 48463 + }, + { + "epoch": 0.4206907926146474, + "grad_norm": 0.32421875, + "learning_rate": 0.0011011012106238007, + "loss": 0.125, + "step": 48464 + }, + { + "epoch": 0.4206994730948516, + "grad_norm": 0.09228515625, + "learning_rate": 0.0011010714481887644, + "loss": 0.0825, + "step": 48465 + }, + { + "epoch": 0.42070815357505575, + "grad_norm": 0.2734375, + "learning_rate": 0.0011010416857525568, + "loss": 0.1729, + "step": 48466 + }, + { + "epoch": 0.42071683405525995, + "grad_norm": 0.0986328125, + "learning_rate": 0.0011010119233152101, + "loss": 0.1045, + "step": 48467 + }, + { + "epoch": 0.4207255145354641, + "grad_norm": 0.1064453125, + "learning_rate": 0.0011009821608767565, + "loss": 0.1357, + "step": 48468 + }, + { + "epoch": 0.4207341950156683, + "grad_norm": 0.439453125, + "learning_rate": 0.0011009523984372286, + "loss": 0.126, + "step": 48469 + }, + { + "epoch": 0.4207428754958724, + "grad_norm": 0.53125, + "learning_rate": 0.0011009226359966595, + "loss": 0.167, + "step": 48470 + }, + { + "epoch": 0.4207515559760766, + "grad_norm": 0.31640625, + "learning_rate": 0.0011008928735550818, + "loss": 0.1074, + "step": 48471 + }, + { + "epoch": 0.42076023645628075, + "grad_norm": 0.1279296875, + "learning_rate": 0.0011008631111125273, + "loss": 0.1113, + "step": 48472 + }, + { + "epoch": 0.42076891693648494, + "grad_norm": 0.0849609375, + "learning_rate": 0.0011008333486690288, + "loss": 0.0894, + "step": 48473 + }, + { + "epoch": 0.4207775974166891, + "grad_norm": 0.12158203125, + "learning_rate": 0.001100803586224619, + "loss": 0.1045, + "step": 48474 + }, + { + "epoch": 0.42078627789689327, + "grad_norm": 0.1572265625, + "learning_rate": 0.0011007738237793308, + "loss": 0.0879, + "step": 48475 + }, + { + "epoch": 0.4207949583770974, + "grad_norm": 0.263671875, + "learning_rate": 0.0011007440613331958, + "loss": 0.1035, + "step": 48476 + }, + { + "epoch": 0.4208036388573016, + "grad_norm": 0.76953125, + "learning_rate": 0.0011007142988862474, + "loss": 0.1016, + "step": 48477 + }, + { + "epoch": 0.42081231933750574, + "grad_norm": 0.27734375, + "learning_rate": 0.0011006845364385177, + "loss": 0.1084, + "step": 48478 + }, + { + "epoch": 0.42082099981770993, + "grad_norm": 0.2451171875, + "learning_rate": 0.0011006547739900393, + "loss": 0.1133, + "step": 48479 + }, + { + "epoch": 0.42082968029791407, + "grad_norm": 0.2119140625, + "learning_rate": 0.0011006250115408451, + "loss": 0.084, + "step": 48480 + }, + { + "epoch": 0.42083836077811826, + "grad_norm": 0.46484375, + "learning_rate": 0.0011005952490909677, + "loss": 0.104, + "step": 48481 + }, + { + "epoch": 0.4208470412583224, + "grad_norm": 0.08544921875, + "learning_rate": 0.0011005654866404386, + "loss": 0.0879, + "step": 48482 + }, + { + "epoch": 0.4208557217385266, + "grad_norm": 0.1474609375, + "learning_rate": 0.0011005357241892918, + "loss": 0.0659, + "step": 48483 + }, + { + "epoch": 0.4208644022187307, + "grad_norm": 0.265625, + "learning_rate": 0.0011005059617375588, + "loss": 0.1123, + "step": 48484 + }, + { + "epoch": 0.4208730826989349, + "grad_norm": 0.1083984375, + "learning_rate": 0.0011004761992852725, + "loss": 0.1084, + "step": 48485 + }, + { + "epoch": 0.42088176317913906, + "grad_norm": 0.421875, + "learning_rate": 0.001100446436832465, + "loss": 0.0742, + "step": 48486 + }, + { + "epoch": 0.42089044365934325, + "grad_norm": 0.10546875, + "learning_rate": 0.0011004166743791702, + "loss": 0.0977, + "step": 48487 + }, + { + "epoch": 0.4208991241395474, + "grad_norm": 0.431640625, + "learning_rate": 0.0011003869119254191, + "loss": 0.0908, + "step": 48488 + }, + { + "epoch": 0.4209078046197516, + "grad_norm": 0.1923828125, + "learning_rate": 0.001100357149471245, + "loss": 0.123, + "step": 48489 + }, + { + "epoch": 0.4209164850999557, + "grad_norm": 0.7421875, + "learning_rate": 0.0011003273870166803, + "loss": 0.1123, + "step": 48490 + }, + { + "epoch": 0.4209251655801599, + "grad_norm": 0.87890625, + "learning_rate": 0.0011002976245617574, + "loss": 0.1211, + "step": 48491 + }, + { + "epoch": 0.42093384606036405, + "grad_norm": 0.1201171875, + "learning_rate": 0.0011002678621065097, + "loss": 0.1064, + "step": 48492 + }, + { + "epoch": 0.42094252654056824, + "grad_norm": 0.1923828125, + "learning_rate": 0.0011002380996509684, + "loss": 0.1748, + "step": 48493 + }, + { + "epoch": 0.4209512070207724, + "grad_norm": 0.107421875, + "learning_rate": 0.0011002083371951673, + "loss": 0.1602, + "step": 48494 + }, + { + "epoch": 0.42095988750097657, + "grad_norm": 0.62890625, + "learning_rate": 0.0011001785747391378, + "loss": 0.0835, + "step": 48495 + }, + { + "epoch": 0.4209685679811807, + "grad_norm": 0.37890625, + "learning_rate": 0.001100148812282913, + "loss": 0.0864, + "step": 48496 + }, + { + "epoch": 0.4209772484613849, + "grad_norm": 0.08984375, + "learning_rate": 0.0011001190498265256, + "loss": 0.0894, + "step": 48497 + }, + { + "epoch": 0.42098592894158904, + "grad_norm": 0.201171875, + "learning_rate": 0.001100089287370008, + "loss": 0.1416, + "step": 48498 + }, + { + "epoch": 0.42099460942179323, + "grad_norm": 0.671875, + "learning_rate": 0.001100059524913393, + "loss": 0.0732, + "step": 48499 + }, + { + "epoch": 0.42100328990199737, + "grad_norm": 0.1962890625, + "learning_rate": 0.0011000297624567128, + "loss": 0.1221, + "step": 48500 + }, + { + "epoch": 0.42101197038220156, + "grad_norm": 0.1708984375, + "learning_rate": 0.0011, + "loss": 0.0698, + "step": 48501 + }, + { + "epoch": 0.4210206508624057, + "grad_norm": 0.11669921875, + "learning_rate": 0.0010999702375432874, + "loss": 0.1006, + "step": 48502 + }, + { + "epoch": 0.4210293313426099, + "grad_norm": 0.498046875, + "learning_rate": 0.0010999404750866072, + "loss": 0.1387, + "step": 48503 + }, + { + "epoch": 0.42103801182281403, + "grad_norm": 0.12890625, + "learning_rate": 0.001099910712629992, + "loss": 0.1387, + "step": 48504 + }, + { + "epoch": 0.4210466923030182, + "grad_norm": 0.1484375, + "learning_rate": 0.0010998809501734745, + "loss": 0.1309, + "step": 48505 + }, + { + "epoch": 0.42105537278322236, + "grad_norm": 0.09716796875, + "learning_rate": 0.001099851187717087, + "loss": 0.1104, + "step": 48506 + }, + { + "epoch": 0.42106405326342655, + "grad_norm": 0.08984375, + "learning_rate": 0.0010998214252608625, + "loss": 0.0908, + "step": 48507 + }, + { + "epoch": 0.4210727337436307, + "grad_norm": 0.4375, + "learning_rate": 0.001099791662804833, + "loss": 0.1553, + "step": 48508 + }, + { + "epoch": 0.4210814142238349, + "grad_norm": 0.5546875, + "learning_rate": 0.0010997619003490318, + "loss": 0.1201, + "step": 48509 + }, + { + "epoch": 0.421090094704039, + "grad_norm": 0.15234375, + "learning_rate": 0.0010997321378934909, + "loss": 0.126, + "step": 48510 + }, + { + "epoch": 0.4210987751842432, + "grad_norm": 0.19140625, + "learning_rate": 0.0010997023754382427, + "loss": 0.0996, + "step": 48511 + }, + { + "epoch": 0.42110745566444735, + "grad_norm": 0.30859375, + "learning_rate": 0.00109967261298332, + "loss": 0.0903, + "step": 48512 + }, + { + "epoch": 0.42111613614465154, + "grad_norm": 0.1806640625, + "learning_rate": 0.0010996428505287553, + "loss": 0.1011, + "step": 48513 + }, + { + "epoch": 0.4211248166248557, + "grad_norm": 0.10498046875, + "learning_rate": 0.001099613088074581, + "loss": 0.1152, + "step": 48514 + }, + { + "epoch": 0.4211334971050599, + "grad_norm": 0.166015625, + "learning_rate": 0.0010995833256208301, + "loss": 0.0811, + "step": 48515 + }, + { + "epoch": 0.421142177585264, + "grad_norm": 0.3828125, + "learning_rate": 0.0010995535631675348, + "loss": 0.0918, + "step": 48516 + }, + { + "epoch": 0.4211508580654682, + "grad_norm": 0.345703125, + "learning_rate": 0.0010995238007147276, + "loss": 0.1025, + "step": 48517 + }, + { + "epoch": 0.42115953854567234, + "grad_norm": 0.3359375, + "learning_rate": 0.0010994940382624417, + "loss": 0.0894, + "step": 48518 + }, + { + "epoch": 0.42116821902587653, + "grad_norm": 0.296875, + "learning_rate": 0.0010994642758107086, + "loss": 0.1143, + "step": 48519 + }, + { + "epoch": 0.42117689950608067, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010994345133595615, + "loss": 0.1182, + "step": 48520 + }, + { + "epoch": 0.42118557998628486, + "grad_norm": 0.09033203125, + "learning_rate": 0.0010994047509090329, + "loss": 0.104, + "step": 48521 + }, + { + "epoch": 0.421194260466489, + "grad_norm": 0.154296875, + "learning_rate": 0.001099374988459155, + "loss": 0.083, + "step": 48522 + }, + { + "epoch": 0.4212029409466932, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010993452260099606, + "loss": 0.123, + "step": 48523 + }, + { + "epoch": 0.42121162142689733, + "grad_norm": 0.1103515625, + "learning_rate": 0.0010993154635614824, + "loss": 0.0986, + "step": 48524 + }, + { + "epoch": 0.4212203019071015, + "grad_norm": 0.337890625, + "learning_rate": 0.001099285701113753, + "loss": 0.0859, + "step": 48525 + }, + { + "epoch": 0.42122898238730566, + "grad_norm": 0.306640625, + "learning_rate": 0.0010992559386668044, + "loss": 0.0781, + "step": 48526 + }, + { + "epoch": 0.42123766286750985, + "grad_norm": 0.1357421875, + "learning_rate": 0.0010992261762206696, + "loss": 0.0923, + "step": 48527 + }, + { + "epoch": 0.421246343347714, + "grad_norm": 0.17578125, + "learning_rate": 0.0010991964137753811, + "loss": 0.0996, + "step": 48528 + }, + { + "epoch": 0.4212550238279182, + "grad_norm": 0.251953125, + "learning_rate": 0.0010991666513309715, + "loss": 0.1152, + "step": 48529 + }, + { + "epoch": 0.4212637043081223, + "grad_norm": 0.361328125, + "learning_rate": 0.0010991368888874733, + "loss": 0.1006, + "step": 48530 + }, + { + "epoch": 0.4212723847883265, + "grad_norm": 0.09619140625, + "learning_rate": 0.0010991071264449188, + "loss": 0.0791, + "step": 48531 + }, + { + "epoch": 0.42128106526853065, + "grad_norm": 0.125, + "learning_rate": 0.0010990773640033402, + "loss": 0.1108, + "step": 48532 + }, + { + "epoch": 0.42128974574873485, + "grad_norm": 0.07275390625, + "learning_rate": 0.0010990476015627713, + "loss": 0.0708, + "step": 48533 + }, + { + "epoch": 0.421298426228939, + "grad_norm": 0.427734375, + "learning_rate": 0.0010990178391232436, + "loss": 0.0859, + "step": 48534 + }, + { + "epoch": 0.4213071067091432, + "grad_norm": 0.3984375, + "learning_rate": 0.0010989880766847902, + "loss": 0.1387, + "step": 48535 + }, + { + "epoch": 0.4213157871893473, + "grad_norm": 0.09716796875, + "learning_rate": 0.0010989583142474433, + "loss": 0.1074, + "step": 48536 + }, + { + "epoch": 0.4213244676695515, + "grad_norm": 0.1796875, + "learning_rate": 0.0010989285518112357, + "loss": 0.0898, + "step": 48537 + }, + { + "epoch": 0.42133314814975564, + "grad_norm": 0.2158203125, + "learning_rate": 0.0010988987893761999, + "loss": 0.0796, + "step": 48538 + }, + { + "epoch": 0.42134182862995984, + "grad_norm": 0.08642578125, + "learning_rate": 0.0010988690269423682, + "loss": 0.0913, + "step": 48539 + }, + { + "epoch": 0.421350509110164, + "grad_norm": 0.234375, + "learning_rate": 0.001098839264509773, + "loss": 0.1113, + "step": 48540 + }, + { + "epoch": 0.42135918959036817, + "grad_norm": 0.296875, + "learning_rate": 0.0010988095020784475, + "loss": 0.0918, + "step": 48541 + }, + { + "epoch": 0.4213678700705723, + "grad_norm": 0.09521484375, + "learning_rate": 0.0010987797396484241, + "loss": 0.0767, + "step": 48542 + }, + { + "epoch": 0.4213765505507765, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010987499772197346, + "loss": 0.0732, + "step": 48543 + }, + { + "epoch": 0.42138523103098063, + "grad_norm": 0.234375, + "learning_rate": 0.0010987202147924127, + "loss": 0.0762, + "step": 48544 + }, + { + "epoch": 0.42139391151118477, + "grad_norm": 0.43359375, + "learning_rate": 0.00109869045236649, + "loss": 0.0859, + "step": 48545 + }, + { + "epoch": 0.42140259199138896, + "grad_norm": 0.68359375, + "learning_rate": 0.0010986606899419996, + "loss": 0.1016, + "step": 48546 + }, + { + "epoch": 0.4214112724715931, + "grad_norm": 0.259765625, + "learning_rate": 0.001098630927518974, + "loss": 0.0898, + "step": 48547 + }, + { + "epoch": 0.4214199529517973, + "grad_norm": 0.25, + "learning_rate": 0.0010986011650974453, + "loss": 0.1138, + "step": 48548 + }, + { + "epoch": 0.42142863343200143, + "grad_norm": 0.203125, + "learning_rate": 0.001098571402677446, + "loss": 0.0601, + "step": 48549 + }, + { + "epoch": 0.4214373139122056, + "grad_norm": 0.2734375, + "learning_rate": 0.0010985416402590094, + "loss": 0.125, + "step": 48550 + }, + { + "epoch": 0.42144599439240976, + "grad_norm": 0.5078125, + "learning_rate": 0.0010985118778421676, + "loss": 0.0938, + "step": 48551 + }, + { + "epoch": 0.42145467487261395, + "grad_norm": 0.1064453125, + "learning_rate": 0.0010984821154269534, + "loss": 0.1367, + "step": 48552 + }, + { + "epoch": 0.4214633553528181, + "grad_norm": 0.5390625, + "learning_rate": 0.0010984523530133989, + "loss": 0.1133, + "step": 48553 + }, + { + "epoch": 0.4214720358330223, + "grad_norm": 0.2353515625, + "learning_rate": 0.0010984225906015368, + "loss": 0.0942, + "step": 48554 + }, + { + "epoch": 0.4214807163132264, + "grad_norm": 0.318359375, + "learning_rate": 0.0010983928281914, + "loss": 0.1084, + "step": 48555 + }, + { + "epoch": 0.4214893967934306, + "grad_norm": 0.0927734375, + "learning_rate": 0.0010983630657830205, + "loss": 0.1094, + "step": 48556 + }, + { + "epoch": 0.42149807727363475, + "grad_norm": 0.2265625, + "learning_rate": 0.0010983333033764316, + "loss": 0.124, + "step": 48557 + }, + { + "epoch": 0.42150675775383895, + "grad_norm": 0.201171875, + "learning_rate": 0.0010983035409716647, + "loss": 0.0801, + "step": 48558 + }, + { + "epoch": 0.4215154382340431, + "grad_norm": 0.06689453125, + "learning_rate": 0.0010982737785687534, + "loss": 0.0776, + "step": 48559 + }, + { + "epoch": 0.4215241187142473, + "grad_norm": 0.125, + "learning_rate": 0.0010982440161677299, + "loss": 0.1143, + "step": 48560 + }, + { + "epoch": 0.4215327991944514, + "grad_norm": 0.2451171875, + "learning_rate": 0.0010982142537686267, + "loss": 0.0835, + "step": 48561 + }, + { + "epoch": 0.4215414796746556, + "grad_norm": 0.39453125, + "learning_rate": 0.0010981844913714762, + "loss": 0.125, + "step": 48562 + }, + { + "epoch": 0.42155016015485974, + "grad_norm": 0.5234375, + "learning_rate": 0.0010981547289763112, + "loss": 0.1016, + "step": 48563 + }, + { + "epoch": 0.42155884063506394, + "grad_norm": 0.30859375, + "learning_rate": 0.0010981249665831642, + "loss": 0.1221, + "step": 48564 + }, + { + "epoch": 0.4215675211152681, + "grad_norm": 0.15234375, + "learning_rate": 0.001098095204192068, + "loss": 0.123, + "step": 48565 + }, + { + "epoch": 0.42157620159547227, + "grad_norm": 0.255859375, + "learning_rate": 0.001098065441803054, + "loss": 0.084, + "step": 48566 + }, + { + "epoch": 0.4215848820756764, + "grad_norm": 0.80078125, + "learning_rate": 0.0010980356794161564, + "loss": 0.0703, + "step": 48567 + }, + { + "epoch": 0.4215935625558806, + "grad_norm": 0.08447265625, + "learning_rate": 0.0010980059170314067, + "loss": 0.0796, + "step": 48568 + }, + { + "epoch": 0.42160224303608473, + "grad_norm": 0.71484375, + "learning_rate": 0.0010979761546488379, + "loss": 0.0928, + "step": 48569 + }, + { + "epoch": 0.4216109235162889, + "grad_norm": 0.17578125, + "learning_rate": 0.001097946392268482, + "loss": 0.1289, + "step": 48570 + }, + { + "epoch": 0.42161960399649306, + "grad_norm": 0.330078125, + "learning_rate": 0.0010979166298903723, + "loss": 0.1152, + "step": 48571 + }, + { + "epoch": 0.42162828447669726, + "grad_norm": 0.1083984375, + "learning_rate": 0.0010978868675145406, + "loss": 0.1094, + "step": 48572 + }, + { + "epoch": 0.4216369649569014, + "grad_norm": 0.1357421875, + "learning_rate": 0.0010978571051410198, + "loss": 0.0723, + "step": 48573 + }, + { + "epoch": 0.4216456454371056, + "grad_norm": 0.0986328125, + "learning_rate": 0.0010978273427698427, + "loss": 0.0674, + "step": 48574 + }, + { + "epoch": 0.4216543259173097, + "grad_norm": 0.0732421875, + "learning_rate": 0.001097797580401041, + "loss": 0.0757, + "step": 48575 + }, + { + "epoch": 0.4216630063975139, + "grad_norm": 0.06494140625, + "learning_rate": 0.0010977678180346482, + "loss": 0.0737, + "step": 48576 + }, + { + "epoch": 0.42167168687771805, + "grad_norm": 0.07373046875, + "learning_rate": 0.0010977380556706966, + "loss": 0.0933, + "step": 48577 + }, + { + "epoch": 0.42168036735792225, + "grad_norm": 0.115234375, + "learning_rate": 0.0010977082933092187, + "loss": 0.1191, + "step": 48578 + }, + { + "epoch": 0.4216890478381264, + "grad_norm": 0.2578125, + "learning_rate": 0.0010976785309502468, + "loss": 0.0737, + "step": 48579 + }, + { + "epoch": 0.4216977283183306, + "grad_norm": 0.376953125, + "learning_rate": 0.0010976487685938137, + "loss": 0.0996, + "step": 48580 + }, + { + "epoch": 0.4217064087985347, + "grad_norm": 0.44921875, + "learning_rate": 0.0010976190062399521, + "loss": 0.0596, + "step": 48581 + }, + { + "epoch": 0.4217150892787389, + "grad_norm": 0.302734375, + "learning_rate": 0.001097589243888694, + "loss": 0.0791, + "step": 48582 + }, + { + "epoch": 0.42172376975894305, + "grad_norm": 0.263671875, + "learning_rate": 0.0010975594815400723, + "loss": 0.1094, + "step": 48583 + }, + { + "epoch": 0.42173245023914724, + "grad_norm": 0.1943359375, + "learning_rate": 0.001097529719194119, + "loss": 0.1001, + "step": 48584 + }, + { + "epoch": 0.4217411307193514, + "grad_norm": 0.10302734375, + "learning_rate": 0.001097499956850868, + "loss": 0.1094, + "step": 48585 + }, + { + "epoch": 0.42174981119955557, + "grad_norm": 0.30859375, + "learning_rate": 0.0010974701945103504, + "loss": 0.1211, + "step": 48586 + }, + { + "epoch": 0.4217584916797597, + "grad_norm": 0.2578125, + "learning_rate": 0.0010974404321725997, + "loss": 0.1045, + "step": 48587 + }, + { + "epoch": 0.4217671721599639, + "grad_norm": 0.453125, + "learning_rate": 0.0010974106698376483, + "loss": 0.0947, + "step": 48588 + }, + { + "epoch": 0.42177585264016804, + "grad_norm": 0.2275390625, + "learning_rate": 0.0010973809075055284, + "loss": 0.0879, + "step": 48589 + }, + { + "epoch": 0.42178453312037223, + "grad_norm": 0.130859375, + "learning_rate": 0.0010973511451762726, + "loss": 0.1162, + "step": 48590 + }, + { + "epoch": 0.42179321360057637, + "grad_norm": 0.1826171875, + "learning_rate": 0.0010973213828499133, + "loss": 0.0972, + "step": 48591 + }, + { + "epoch": 0.42180189408078056, + "grad_norm": 0.435546875, + "learning_rate": 0.0010972916205264835, + "loss": 0.126, + "step": 48592 + }, + { + "epoch": 0.4218105745609847, + "grad_norm": 0.248046875, + "learning_rate": 0.0010972618582060156, + "loss": 0.0957, + "step": 48593 + }, + { + "epoch": 0.4218192550411889, + "grad_norm": 0.1025390625, + "learning_rate": 0.001097232095888542, + "loss": 0.1074, + "step": 48594 + }, + { + "epoch": 0.421827935521393, + "grad_norm": 0.27734375, + "learning_rate": 0.0010972023335740955, + "loss": 0.0781, + "step": 48595 + }, + { + "epoch": 0.4218366160015972, + "grad_norm": 0.1767578125, + "learning_rate": 0.0010971725712627086, + "loss": 0.0947, + "step": 48596 + }, + { + "epoch": 0.42184529648180136, + "grad_norm": 0.125, + "learning_rate": 0.0010971428089544136, + "loss": 0.0996, + "step": 48597 + }, + { + "epoch": 0.42185397696200555, + "grad_norm": 0.1611328125, + "learning_rate": 0.001097113046649243, + "loss": 0.0913, + "step": 48598 + }, + { + "epoch": 0.4218626574422097, + "grad_norm": 0.361328125, + "learning_rate": 0.0010970832843472298, + "loss": 0.0957, + "step": 48599 + }, + { + "epoch": 0.4218713379224139, + "grad_norm": 0.34765625, + "learning_rate": 0.001097053522048406, + "loss": 0.0918, + "step": 48600 + }, + { + "epoch": 0.421880018402618, + "grad_norm": 0.1552734375, + "learning_rate": 0.0010970237597528044, + "loss": 0.0835, + "step": 48601 + }, + { + "epoch": 0.4218886988828222, + "grad_norm": 0.7421875, + "learning_rate": 0.0010969939974604576, + "loss": 0.127, + "step": 48602 + }, + { + "epoch": 0.42189737936302635, + "grad_norm": 0.470703125, + "learning_rate": 0.0010969642351713982, + "loss": 0.064, + "step": 48603 + }, + { + "epoch": 0.42190605984323054, + "grad_norm": 0.435546875, + "learning_rate": 0.001096934472885659, + "loss": 0.1035, + "step": 48604 + }, + { + "epoch": 0.4219147403234347, + "grad_norm": 0.53125, + "learning_rate": 0.0010969047106032719, + "loss": 0.0796, + "step": 48605 + }, + { + "epoch": 0.42192342080363887, + "grad_norm": 0.296875, + "learning_rate": 0.0010968749483242697, + "loss": 0.0986, + "step": 48606 + }, + { + "epoch": 0.421932101283843, + "grad_norm": 0.12060546875, + "learning_rate": 0.0010968451860486853, + "loss": 0.0767, + "step": 48607 + }, + { + "epoch": 0.4219407817640472, + "grad_norm": 0.1796875, + "learning_rate": 0.0010968154237765509, + "loss": 0.0874, + "step": 48608 + }, + { + "epoch": 0.42194946224425134, + "grad_norm": 0.2236328125, + "learning_rate": 0.0010967856615078989, + "loss": 0.127, + "step": 48609 + }, + { + "epoch": 0.42195814272445553, + "grad_norm": 0.150390625, + "learning_rate": 0.0010967558992427617, + "loss": 0.1035, + "step": 48610 + }, + { + "epoch": 0.42196682320465967, + "grad_norm": 0.16796875, + "learning_rate": 0.0010967261369811728, + "loss": 0.1094, + "step": 48611 + }, + { + "epoch": 0.42197550368486386, + "grad_norm": 0.384765625, + "learning_rate": 0.001096696374723164, + "loss": 0.0859, + "step": 48612 + }, + { + "epoch": 0.421984184165068, + "grad_norm": 0.1337890625, + "learning_rate": 0.0010966666124687678, + "loss": 0.0923, + "step": 48613 + }, + { + "epoch": 0.4219928646452722, + "grad_norm": 0.8515625, + "learning_rate": 0.0010966368502180173, + "loss": 0.1074, + "step": 48614 + }, + { + "epoch": 0.42200154512547633, + "grad_norm": 0.5234375, + "learning_rate": 0.0010966070879709444, + "loss": 0.1484, + "step": 48615 + }, + { + "epoch": 0.4220102256056805, + "grad_norm": 0.10400390625, + "learning_rate": 0.0010965773257275821, + "loss": 0.0977, + "step": 48616 + }, + { + "epoch": 0.42201890608588466, + "grad_norm": 0.2001953125, + "learning_rate": 0.0010965475634879627, + "loss": 0.1133, + "step": 48617 + }, + { + "epoch": 0.42202758656608885, + "grad_norm": 0.169921875, + "learning_rate": 0.0010965178012521187, + "loss": 0.0938, + "step": 48618 + }, + { + "epoch": 0.422036267046293, + "grad_norm": 0.09716796875, + "learning_rate": 0.0010964880390200828, + "loss": 0.0767, + "step": 48619 + }, + { + "epoch": 0.4220449475264972, + "grad_norm": 0.64453125, + "learning_rate": 0.0010964582767918878, + "loss": 0.082, + "step": 48620 + }, + { + "epoch": 0.4220536280067013, + "grad_norm": 0.412109375, + "learning_rate": 0.0010964285145675658, + "loss": 0.0703, + "step": 48621 + }, + { + "epoch": 0.4220623084869055, + "grad_norm": 0.083984375, + "learning_rate": 0.0010963987523471494, + "loss": 0.0889, + "step": 48622 + }, + { + "epoch": 0.42207098896710965, + "grad_norm": 0.470703125, + "learning_rate": 0.0010963689901306716, + "loss": 0.0815, + "step": 48623 + }, + { + "epoch": 0.42207966944731384, + "grad_norm": 0.18359375, + "learning_rate": 0.0010963392279181644, + "loss": 0.0781, + "step": 48624 + }, + { + "epoch": 0.422088349927518, + "grad_norm": 0.3515625, + "learning_rate": 0.0010963094657096607, + "loss": 0.124, + "step": 48625 + }, + { + "epoch": 0.4220970304077222, + "grad_norm": 0.373046875, + "learning_rate": 0.0010962797035051928, + "loss": 0.0649, + "step": 48626 + }, + { + "epoch": 0.4221057108879263, + "grad_norm": 0.126953125, + "learning_rate": 0.001096249941304793, + "loss": 0.0933, + "step": 48627 + }, + { + "epoch": 0.4221143913681305, + "grad_norm": 0.10498046875, + "learning_rate": 0.0010962201791084947, + "loss": 0.0957, + "step": 48628 + }, + { + "epoch": 0.42212307184833464, + "grad_norm": 0.212890625, + "learning_rate": 0.00109619041691633, + "loss": 0.0996, + "step": 48629 + }, + { + "epoch": 0.42213175232853883, + "grad_norm": 0.146484375, + "learning_rate": 0.0010961606547283311, + "loss": 0.0732, + "step": 48630 + }, + { + "epoch": 0.42214043280874297, + "grad_norm": 0.203125, + "learning_rate": 0.001096130892544531, + "loss": 0.1191, + "step": 48631 + }, + { + "epoch": 0.42214911328894716, + "grad_norm": 0.08935546875, + "learning_rate": 0.0010961011303649622, + "loss": 0.0811, + "step": 48632 + }, + { + "epoch": 0.4221577937691513, + "grad_norm": 0.2158203125, + "learning_rate": 0.001096071368189657, + "loss": 0.1211, + "step": 48633 + }, + { + "epoch": 0.4221664742493555, + "grad_norm": 0.515625, + "learning_rate": 0.0010960416060186483, + "loss": 0.0903, + "step": 48634 + }, + { + "epoch": 0.42217515472955963, + "grad_norm": 0.10009765625, + "learning_rate": 0.001096011843851968, + "loss": 0.1406, + "step": 48635 + }, + { + "epoch": 0.4221838352097638, + "grad_norm": 0.1650390625, + "learning_rate": 0.0010959820816896493, + "loss": 0.1289, + "step": 48636 + }, + { + "epoch": 0.42219251568996796, + "grad_norm": 0.263671875, + "learning_rate": 0.0010959523195317247, + "loss": 0.0986, + "step": 48637 + }, + { + "epoch": 0.42220119617017215, + "grad_norm": 0.09521484375, + "learning_rate": 0.0010959225573782265, + "loss": 0.0791, + "step": 48638 + }, + { + "epoch": 0.4222098766503763, + "grad_norm": 0.408203125, + "learning_rate": 0.0010958927952291872, + "loss": 0.0996, + "step": 48639 + }, + { + "epoch": 0.4222185571305805, + "grad_norm": 0.396484375, + "learning_rate": 0.0010958630330846395, + "loss": 0.0986, + "step": 48640 + }, + { + "epoch": 0.4222272376107846, + "grad_norm": 0.2734375, + "learning_rate": 0.0010958332709446162, + "loss": 0.1001, + "step": 48641 + }, + { + "epoch": 0.4222359180909888, + "grad_norm": 0.2216796875, + "learning_rate": 0.0010958035088091493, + "loss": 0.1387, + "step": 48642 + }, + { + "epoch": 0.42224459857119295, + "grad_norm": 0.26171875, + "learning_rate": 0.0010957737466782717, + "loss": 0.1064, + "step": 48643 + }, + { + "epoch": 0.42225327905139715, + "grad_norm": 0.376953125, + "learning_rate": 0.0010957439845520156, + "loss": 0.1465, + "step": 48644 + }, + { + "epoch": 0.4222619595316013, + "grad_norm": 0.1943359375, + "learning_rate": 0.0010957142224304142, + "loss": 0.0986, + "step": 48645 + }, + { + "epoch": 0.4222706400118055, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010956844603134998, + "loss": 0.0791, + "step": 48646 + }, + { + "epoch": 0.4222793204920096, + "grad_norm": 0.734375, + "learning_rate": 0.0010956546982013044, + "loss": 0.1309, + "step": 48647 + }, + { + "epoch": 0.4222880009722138, + "grad_norm": 0.46484375, + "learning_rate": 0.0010956249360938612, + "loss": 0.0977, + "step": 48648 + }, + { + "epoch": 0.42229668145241794, + "grad_norm": 0.5078125, + "learning_rate": 0.0010955951739912025, + "loss": 0.0977, + "step": 48649 + }, + { + "epoch": 0.42230536193262214, + "grad_norm": 0.208984375, + "learning_rate": 0.0010955654118933607, + "loss": 0.1289, + "step": 48650 + }, + { + "epoch": 0.4223140424128263, + "grad_norm": 0.443359375, + "learning_rate": 0.0010955356498003686, + "loss": 0.0742, + "step": 48651 + }, + { + "epoch": 0.42232272289303047, + "grad_norm": 0.494140625, + "learning_rate": 0.0010955058877122586, + "loss": 0.1172, + "step": 48652 + }, + { + "epoch": 0.4223314033732346, + "grad_norm": 0.2734375, + "learning_rate": 0.0010954761256290629, + "loss": 0.1289, + "step": 48653 + }, + { + "epoch": 0.4223400838534388, + "grad_norm": 0.240234375, + "learning_rate": 0.001095446363550815, + "loss": 0.083, + "step": 48654 + }, + { + "epoch": 0.42234876433364293, + "grad_norm": 0.1201171875, + "learning_rate": 0.001095416601477547, + "loss": 0.1035, + "step": 48655 + }, + { + "epoch": 0.4223574448138471, + "grad_norm": 0.060546875, + "learning_rate": 0.001095386839409291, + "loss": 0.0625, + "step": 48656 + }, + { + "epoch": 0.42236612529405126, + "grad_norm": 0.3984375, + "learning_rate": 0.00109535707734608, + "loss": 0.085, + "step": 48657 + }, + { + "epoch": 0.42237480577425546, + "grad_norm": 0.412109375, + "learning_rate": 0.0010953273152879462, + "loss": 0.1011, + "step": 48658 + }, + { + "epoch": 0.4223834862544596, + "grad_norm": 0.490234375, + "learning_rate": 0.0010952975532349226, + "loss": 0.0605, + "step": 48659 + }, + { + "epoch": 0.4223921667346638, + "grad_norm": 0.181640625, + "learning_rate": 0.0010952677911870415, + "loss": 0.1396, + "step": 48660 + }, + { + "epoch": 0.4224008472148679, + "grad_norm": 0.4375, + "learning_rate": 0.0010952380291443353, + "loss": 0.1289, + "step": 48661 + }, + { + "epoch": 0.4224095276950721, + "grad_norm": 0.2177734375, + "learning_rate": 0.0010952082671068367, + "loss": 0.1211, + "step": 48662 + }, + { + "epoch": 0.42241820817527626, + "grad_norm": 0.0888671875, + "learning_rate": 0.0010951785050745783, + "loss": 0.084, + "step": 48663 + }, + { + "epoch": 0.42242688865548045, + "grad_norm": 0.9921875, + "learning_rate": 0.001095148743047593, + "loss": 0.0913, + "step": 48664 + }, + { + "epoch": 0.4224355691356846, + "grad_norm": 0.08251953125, + "learning_rate": 0.0010951189810259128, + "loss": 0.1108, + "step": 48665 + }, + { + "epoch": 0.4224442496158888, + "grad_norm": 0.345703125, + "learning_rate": 0.0010950892190095705, + "loss": 0.0884, + "step": 48666 + }, + { + "epoch": 0.4224529300960929, + "grad_norm": 0.451171875, + "learning_rate": 0.0010950594569985983, + "loss": 0.1162, + "step": 48667 + }, + { + "epoch": 0.42246161057629705, + "grad_norm": 0.62890625, + "learning_rate": 0.001095029694993029, + "loss": 0.0752, + "step": 48668 + }, + { + "epoch": 0.42247029105650125, + "grad_norm": 0.32421875, + "learning_rate": 0.0010949999329928953, + "loss": 0.1182, + "step": 48669 + }, + { + "epoch": 0.4224789715367054, + "grad_norm": 0.1474609375, + "learning_rate": 0.0010949701709982295, + "loss": 0.1211, + "step": 48670 + }, + { + "epoch": 0.4224876520169096, + "grad_norm": 0.73828125, + "learning_rate": 0.001094940409009064, + "loss": 0.1348, + "step": 48671 + }, + { + "epoch": 0.4224963324971137, + "grad_norm": 0.341796875, + "learning_rate": 0.0010949106470254319, + "loss": 0.125, + "step": 48672 + }, + { + "epoch": 0.4225050129773179, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010948808850473656, + "loss": 0.085, + "step": 48673 + }, + { + "epoch": 0.42251369345752204, + "grad_norm": 0.1201171875, + "learning_rate": 0.0010948511230748975, + "loss": 0.1191, + "step": 48674 + }, + { + "epoch": 0.42252237393772624, + "grad_norm": 0.1884765625, + "learning_rate": 0.0010948213611080598, + "loss": 0.085, + "step": 48675 + }, + { + "epoch": 0.4225310544179304, + "grad_norm": 0.34375, + "learning_rate": 0.0010947915991468854, + "loss": 0.1279, + "step": 48676 + }, + { + "epoch": 0.42253973489813457, + "grad_norm": 0.095703125, + "learning_rate": 0.0010947618371914072, + "loss": 0.1138, + "step": 48677 + }, + { + "epoch": 0.4225484153783387, + "grad_norm": 0.208984375, + "learning_rate": 0.001094732075241657, + "loss": 0.0928, + "step": 48678 + }, + { + "epoch": 0.4225570958585429, + "grad_norm": 0.279296875, + "learning_rate": 0.0010947023132976677, + "loss": 0.0874, + "step": 48679 + }, + { + "epoch": 0.42256577633874703, + "grad_norm": 0.578125, + "learning_rate": 0.001094672551359472, + "loss": 0.1318, + "step": 48680 + }, + { + "epoch": 0.4225744568189512, + "grad_norm": 0.205078125, + "learning_rate": 0.0010946427894271023, + "loss": 0.0923, + "step": 48681 + }, + { + "epoch": 0.42258313729915536, + "grad_norm": 0.59375, + "learning_rate": 0.0010946130275005911, + "loss": 0.0986, + "step": 48682 + }, + { + "epoch": 0.42259181777935956, + "grad_norm": 0.63671875, + "learning_rate": 0.0010945832655799715, + "loss": 0.1816, + "step": 48683 + }, + { + "epoch": 0.4226004982595637, + "grad_norm": 0.134765625, + "learning_rate": 0.001094553503665275, + "loss": 0.0972, + "step": 48684 + }, + { + "epoch": 0.4226091787397679, + "grad_norm": 0.1728515625, + "learning_rate": 0.0010945237417565351, + "loss": 0.1064, + "step": 48685 + }, + { + "epoch": 0.422617859219972, + "grad_norm": 0.314453125, + "learning_rate": 0.0010944939798537836, + "loss": 0.1318, + "step": 48686 + }, + { + "epoch": 0.4226265397001762, + "grad_norm": 0.255859375, + "learning_rate": 0.0010944642179570533, + "loss": 0.1064, + "step": 48687 + }, + { + "epoch": 0.42263522018038036, + "grad_norm": 1.0078125, + "learning_rate": 0.001094434456066377, + "loss": 0.123, + "step": 48688 + }, + { + "epoch": 0.42264390066058455, + "grad_norm": 0.36328125, + "learning_rate": 0.0010944046941817873, + "loss": 0.1113, + "step": 48689 + }, + { + "epoch": 0.4226525811407887, + "grad_norm": 0.376953125, + "learning_rate": 0.0010943749323033163, + "loss": 0.1191, + "step": 48690 + }, + { + "epoch": 0.4226612616209929, + "grad_norm": 0.1533203125, + "learning_rate": 0.001094345170430997, + "loss": 0.0674, + "step": 48691 + }, + { + "epoch": 0.422669942101197, + "grad_norm": 0.35546875, + "learning_rate": 0.0010943154085648616, + "loss": 0.1484, + "step": 48692 + }, + { + "epoch": 0.4226786225814012, + "grad_norm": 0.154296875, + "learning_rate": 0.0010942856467049428, + "loss": 0.0752, + "step": 48693 + }, + { + "epoch": 0.42268730306160535, + "grad_norm": 0.2041015625, + "learning_rate": 0.0010942558848512733, + "loss": 0.1069, + "step": 48694 + }, + { + "epoch": 0.42269598354180954, + "grad_norm": 0.158203125, + "learning_rate": 0.0010942261230038855, + "loss": 0.0938, + "step": 48695 + }, + { + "epoch": 0.4227046640220137, + "grad_norm": 0.4609375, + "learning_rate": 0.0010941963611628114, + "loss": 0.0659, + "step": 48696 + }, + { + "epoch": 0.42271334450221787, + "grad_norm": 0.08447265625, + "learning_rate": 0.0010941665993280845, + "loss": 0.1006, + "step": 48697 + }, + { + "epoch": 0.422722024982422, + "grad_norm": 0.162109375, + "learning_rate": 0.001094136837499737, + "loss": 0.1055, + "step": 48698 + }, + { + "epoch": 0.4227307054626262, + "grad_norm": 0.294921875, + "learning_rate": 0.001094107075677801, + "loss": 0.104, + "step": 48699 + }, + { + "epoch": 0.42273938594283034, + "grad_norm": 0.1494140625, + "learning_rate": 0.0010940773138623098, + "loss": 0.0933, + "step": 48700 + }, + { + "epoch": 0.42274806642303453, + "grad_norm": 0.6171875, + "learning_rate": 0.0010940475520532953, + "loss": 0.1113, + "step": 48701 + }, + { + "epoch": 0.42275674690323867, + "grad_norm": 0.546875, + "learning_rate": 0.0010940177902507907, + "loss": 0.0654, + "step": 48702 + }, + { + "epoch": 0.42276542738344286, + "grad_norm": 0.306640625, + "learning_rate": 0.0010939880284548279, + "loss": 0.083, + "step": 48703 + }, + { + "epoch": 0.422774107863647, + "grad_norm": 0.5234375, + "learning_rate": 0.0010939582666654394, + "loss": 0.083, + "step": 48704 + }, + { + "epoch": 0.4227827883438512, + "grad_norm": 0.255859375, + "learning_rate": 0.0010939285048826581, + "loss": 0.1367, + "step": 48705 + }, + { + "epoch": 0.4227914688240553, + "grad_norm": 0.09765625, + "learning_rate": 0.001093898743106517, + "loss": 0.1426, + "step": 48706 + }, + { + "epoch": 0.4228001493042595, + "grad_norm": 0.203125, + "learning_rate": 0.0010938689813370477, + "loss": 0.0879, + "step": 48707 + }, + { + "epoch": 0.42280882978446366, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010938392195742833, + "loss": 0.1055, + "step": 48708 + }, + { + "epoch": 0.42281751026466785, + "grad_norm": 0.150390625, + "learning_rate": 0.0010938094578182562, + "loss": 0.1289, + "step": 48709 + }, + { + "epoch": 0.422826190744872, + "grad_norm": 0.1923828125, + "learning_rate": 0.0010937796960689991, + "loss": 0.1211, + "step": 48710 + }, + { + "epoch": 0.4228348712250762, + "grad_norm": 0.36328125, + "learning_rate": 0.0010937499343265445, + "loss": 0.0986, + "step": 48711 + }, + { + "epoch": 0.4228435517052803, + "grad_norm": 0.1767578125, + "learning_rate": 0.001093720172590925, + "loss": 0.1025, + "step": 48712 + }, + { + "epoch": 0.4228522321854845, + "grad_norm": 0.1416015625, + "learning_rate": 0.0010936904108621724, + "loss": 0.0879, + "step": 48713 + }, + { + "epoch": 0.42286091266568865, + "grad_norm": 0.283203125, + "learning_rate": 0.0010936606491403203, + "loss": 0.0977, + "step": 48714 + }, + { + "epoch": 0.42286959314589284, + "grad_norm": 0.2373046875, + "learning_rate": 0.001093630887425401, + "loss": 0.0996, + "step": 48715 + }, + { + "epoch": 0.422878273626097, + "grad_norm": 0.392578125, + "learning_rate": 0.0010936011257174465, + "loss": 0.0654, + "step": 48716 + }, + { + "epoch": 0.42288695410630117, + "grad_norm": 0.125, + "learning_rate": 0.00109357136401649, + "loss": 0.1074, + "step": 48717 + }, + { + "epoch": 0.4228956345865053, + "grad_norm": 0.09228515625, + "learning_rate": 0.0010935416023225633, + "loss": 0.1328, + "step": 48718 + }, + { + "epoch": 0.4229043150667095, + "grad_norm": 0.447265625, + "learning_rate": 0.0010935118406356999, + "loss": 0.1021, + "step": 48719 + }, + { + "epoch": 0.42291299554691364, + "grad_norm": 0.1318359375, + "learning_rate": 0.0010934820789559316, + "loss": 0.0854, + "step": 48720 + }, + { + "epoch": 0.42292167602711783, + "grad_norm": 0.390625, + "learning_rate": 0.0010934523172832916, + "loss": 0.1045, + "step": 48721 + }, + { + "epoch": 0.42293035650732197, + "grad_norm": 0.2119140625, + "learning_rate": 0.0010934225556178112, + "loss": 0.1934, + "step": 48722 + }, + { + "epoch": 0.42293903698752616, + "grad_norm": 0.2353515625, + "learning_rate": 0.0010933927939595245, + "loss": 0.1084, + "step": 48723 + }, + { + "epoch": 0.4229477174677303, + "grad_norm": 0.515625, + "learning_rate": 0.0010933630323084632, + "loss": 0.1187, + "step": 48724 + }, + { + "epoch": 0.4229563979479345, + "grad_norm": 0.10693359375, + "learning_rate": 0.0010933332706646599, + "loss": 0.1055, + "step": 48725 + }, + { + "epoch": 0.42296507842813863, + "grad_norm": 0.134765625, + "learning_rate": 0.0010933035090281474, + "loss": 0.0835, + "step": 48726 + }, + { + "epoch": 0.4229737589083428, + "grad_norm": 0.31640625, + "learning_rate": 0.001093273747398958, + "loss": 0.1035, + "step": 48727 + }, + { + "epoch": 0.42298243938854696, + "grad_norm": 0.10302734375, + "learning_rate": 0.0010932439857771242, + "loss": 0.0942, + "step": 48728 + }, + { + "epoch": 0.42299111986875115, + "grad_norm": 0.263671875, + "learning_rate": 0.0010932142241626788, + "loss": 0.0654, + "step": 48729 + }, + { + "epoch": 0.4229998003489553, + "grad_norm": 0.2080078125, + "learning_rate": 0.0010931844625556542, + "loss": 0.084, + "step": 48730 + }, + { + "epoch": 0.4230084808291595, + "grad_norm": 0.431640625, + "learning_rate": 0.0010931547009560827, + "loss": 0.1021, + "step": 48731 + }, + { + "epoch": 0.4230171613093636, + "grad_norm": 0.2890625, + "learning_rate": 0.0010931249393639974, + "loss": 0.1025, + "step": 48732 + }, + { + "epoch": 0.4230258417895678, + "grad_norm": 0.212890625, + "learning_rate": 0.0010930951777794306, + "loss": 0.0996, + "step": 48733 + }, + { + "epoch": 0.42303452226977195, + "grad_norm": 0.1533203125, + "learning_rate": 0.001093065416202415, + "loss": 0.064, + "step": 48734 + }, + { + "epoch": 0.42304320274997614, + "grad_norm": 0.1162109375, + "learning_rate": 0.0010930356546329828, + "loss": 0.1104, + "step": 48735 + }, + { + "epoch": 0.4230518832301803, + "grad_norm": 0.482421875, + "learning_rate": 0.0010930058930711665, + "loss": 0.1055, + "step": 48736 + }, + { + "epoch": 0.4230605637103845, + "grad_norm": 0.37890625, + "learning_rate": 0.0010929761315169987, + "loss": 0.1064, + "step": 48737 + }, + { + "epoch": 0.4230692441905886, + "grad_norm": 0.2275390625, + "learning_rate": 0.0010929463699705124, + "loss": 0.1406, + "step": 48738 + }, + { + "epoch": 0.4230779246707928, + "grad_norm": 0.1787109375, + "learning_rate": 0.0010929166084317397, + "loss": 0.1064, + "step": 48739 + }, + { + "epoch": 0.42308660515099694, + "grad_norm": 0.67578125, + "learning_rate": 0.0010928868469007133, + "loss": 0.1045, + "step": 48740 + }, + { + "epoch": 0.42309528563120113, + "grad_norm": 0.19140625, + "learning_rate": 0.001092857085377466, + "loss": 0.1191, + "step": 48741 + }, + { + "epoch": 0.42310396611140527, + "grad_norm": 0.10400390625, + "learning_rate": 0.0010928273238620299, + "loss": 0.1187, + "step": 48742 + }, + { + "epoch": 0.42311264659160946, + "grad_norm": 0.376953125, + "learning_rate": 0.001092797562354438, + "loss": 0.1118, + "step": 48743 + }, + { + "epoch": 0.4231213270718136, + "grad_norm": 0.4296875, + "learning_rate": 0.001092767800854722, + "loss": 0.1201, + "step": 48744 + }, + { + "epoch": 0.4231300075520178, + "grad_norm": 1.1015625, + "learning_rate": 0.0010927380393629155, + "loss": 0.0835, + "step": 48745 + }, + { + "epoch": 0.42313868803222193, + "grad_norm": 0.28515625, + "learning_rate": 0.0010927082778790505, + "loss": 0.0879, + "step": 48746 + }, + { + "epoch": 0.4231473685124261, + "grad_norm": 0.326171875, + "learning_rate": 0.0010926785164031595, + "loss": 0.127, + "step": 48747 + }, + { + "epoch": 0.42315604899263026, + "grad_norm": 0.134765625, + "learning_rate": 0.001092648754935275, + "loss": 0.0967, + "step": 48748 + }, + { + "epoch": 0.42316472947283446, + "grad_norm": 0.1171875, + "learning_rate": 0.0010926189934754298, + "loss": 0.0874, + "step": 48749 + }, + { + "epoch": 0.4231734099530386, + "grad_norm": 0.140625, + "learning_rate": 0.0010925892320236565, + "loss": 0.051, + "step": 48750 + }, + { + "epoch": 0.4231820904332428, + "grad_norm": 0.11181640625, + "learning_rate": 0.0010925594705799876, + "loss": 0.0986, + "step": 48751 + }, + { + "epoch": 0.4231907709134469, + "grad_norm": 0.380859375, + "learning_rate": 0.0010925297091444555, + "loss": 0.0957, + "step": 48752 + }, + { + "epoch": 0.4231994513936511, + "grad_norm": 0.1650390625, + "learning_rate": 0.001092499947717093, + "loss": 0.1035, + "step": 48753 + }, + { + "epoch": 0.42320813187385525, + "grad_norm": 0.52734375, + "learning_rate": 0.001092470186297932, + "loss": 0.1094, + "step": 48754 + }, + { + "epoch": 0.42321681235405945, + "grad_norm": 0.0859375, + "learning_rate": 0.0010924404248870054, + "loss": 0.1084, + "step": 48755 + }, + { + "epoch": 0.4232254928342636, + "grad_norm": 0.1875, + "learning_rate": 0.0010924106634843463, + "loss": 0.1035, + "step": 48756 + }, + { + "epoch": 0.4232341733144678, + "grad_norm": 0.4296875, + "learning_rate": 0.0010923809020899863, + "loss": 0.0806, + "step": 48757 + }, + { + "epoch": 0.4232428537946719, + "grad_norm": 0.267578125, + "learning_rate": 0.0010923511407039588, + "loss": 0.082, + "step": 48758 + }, + { + "epoch": 0.4232515342748761, + "grad_norm": 0.1904296875, + "learning_rate": 0.001092321379326296, + "loss": 0.0767, + "step": 48759 + }, + { + "epoch": 0.42326021475508024, + "grad_norm": 0.2138671875, + "learning_rate": 0.0010922916179570304, + "loss": 0.0898, + "step": 48760 + }, + { + "epoch": 0.42326889523528444, + "grad_norm": 0.388671875, + "learning_rate": 0.0010922618565961946, + "loss": 0.0957, + "step": 48761 + }, + { + "epoch": 0.4232775757154886, + "grad_norm": 0.291015625, + "learning_rate": 0.001092232095243821, + "loss": 0.1377, + "step": 48762 + }, + { + "epoch": 0.42328625619569277, + "grad_norm": 0.2578125, + "learning_rate": 0.0010922023338999422, + "loss": 0.106, + "step": 48763 + }, + { + "epoch": 0.4232949366758969, + "grad_norm": 0.83203125, + "learning_rate": 0.0010921725725645908, + "loss": 0.082, + "step": 48764 + }, + { + "epoch": 0.4233036171561011, + "grad_norm": 0.298828125, + "learning_rate": 0.0010921428112377993, + "loss": 0.0957, + "step": 48765 + }, + { + "epoch": 0.42331229763630523, + "grad_norm": 0.431640625, + "learning_rate": 0.0010921130499196006, + "loss": 0.1074, + "step": 48766 + }, + { + "epoch": 0.4233209781165094, + "grad_norm": 0.306640625, + "learning_rate": 0.0010920832886100265, + "loss": 0.1108, + "step": 48767 + }, + { + "epoch": 0.42332965859671357, + "grad_norm": 0.08642578125, + "learning_rate": 0.0010920535273091104, + "loss": 0.0859, + "step": 48768 + }, + { + "epoch": 0.42333833907691776, + "grad_norm": 0.09765625, + "learning_rate": 0.0010920237660168844, + "loss": 0.0771, + "step": 48769 + }, + { + "epoch": 0.4233470195571219, + "grad_norm": 0.0908203125, + "learning_rate": 0.001091994004733381, + "loss": 0.1055, + "step": 48770 + }, + { + "epoch": 0.4233557000373261, + "grad_norm": 0.07275390625, + "learning_rate": 0.001091964243458633, + "loss": 0.0898, + "step": 48771 + }, + { + "epoch": 0.4233643805175302, + "grad_norm": 0.2158203125, + "learning_rate": 0.0010919344821926726, + "loss": 0.1279, + "step": 48772 + }, + { + "epoch": 0.4233730609977344, + "grad_norm": 0.1298828125, + "learning_rate": 0.0010919047209355323, + "loss": 0.0835, + "step": 48773 + }, + { + "epoch": 0.42338174147793856, + "grad_norm": 0.302734375, + "learning_rate": 0.001091874959687245, + "loss": 0.125, + "step": 48774 + }, + { + "epoch": 0.42339042195814275, + "grad_norm": 0.08349609375, + "learning_rate": 0.0010918451984478434, + "loss": 0.1099, + "step": 48775 + }, + { + "epoch": 0.4233991024383469, + "grad_norm": 0.8515625, + "learning_rate": 0.0010918154372173596, + "loss": 0.1191, + "step": 48776 + }, + { + "epoch": 0.4234077829185511, + "grad_norm": 0.1376953125, + "learning_rate": 0.0010917856759958264, + "loss": 0.0874, + "step": 48777 + }, + { + "epoch": 0.4234164633987552, + "grad_norm": 0.2236328125, + "learning_rate": 0.0010917559147832763, + "loss": 0.083, + "step": 48778 + }, + { + "epoch": 0.4234251438789594, + "grad_norm": 0.09521484375, + "learning_rate": 0.001091726153579742, + "loss": 0.0977, + "step": 48779 + }, + { + "epoch": 0.42343382435916355, + "grad_norm": 0.25390625, + "learning_rate": 0.0010916963923852554, + "loss": 0.0693, + "step": 48780 + }, + { + "epoch": 0.42344250483936774, + "grad_norm": 0.201171875, + "learning_rate": 0.0010916666311998498, + "loss": 0.0996, + "step": 48781 + }, + { + "epoch": 0.4234511853195719, + "grad_norm": 0.2333984375, + "learning_rate": 0.0010916368700235569, + "loss": 0.0933, + "step": 48782 + }, + { + "epoch": 0.42345986579977607, + "grad_norm": 0.119140625, + "learning_rate": 0.0010916071088564103, + "loss": 0.1064, + "step": 48783 + }, + { + "epoch": 0.4234685462799802, + "grad_norm": 0.423828125, + "learning_rate": 0.001091577347698442, + "loss": 0.124, + "step": 48784 + }, + { + "epoch": 0.4234772267601844, + "grad_norm": 0.2080078125, + "learning_rate": 0.0010915475865496845, + "loss": 0.1211, + "step": 48785 + }, + { + "epoch": 0.42348590724038854, + "grad_norm": 0.67578125, + "learning_rate": 0.0010915178254101705, + "loss": 0.1045, + "step": 48786 + }, + { + "epoch": 0.42349458772059273, + "grad_norm": 0.12890625, + "learning_rate": 0.0010914880642799326, + "loss": 0.0884, + "step": 48787 + }, + { + "epoch": 0.42350326820079687, + "grad_norm": 0.1123046875, + "learning_rate": 0.001091458303159003, + "loss": 0.0903, + "step": 48788 + }, + { + "epoch": 0.42351194868100106, + "grad_norm": 0.0712890625, + "learning_rate": 0.0010914285420474145, + "loss": 0.1157, + "step": 48789 + }, + { + "epoch": 0.4235206291612052, + "grad_norm": 0.19140625, + "learning_rate": 0.0010913987809451997, + "loss": 0.0942, + "step": 48790 + }, + { + "epoch": 0.4235293096414094, + "grad_norm": 0.1396484375, + "learning_rate": 0.0010913690198523908, + "loss": 0.1104, + "step": 48791 + }, + { + "epoch": 0.42353799012161353, + "grad_norm": 0.140625, + "learning_rate": 0.001091339258769021, + "loss": 0.1084, + "step": 48792 + }, + { + "epoch": 0.42354667060181767, + "grad_norm": 0.66796875, + "learning_rate": 0.0010913094976951223, + "loss": 0.1211, + "step": 48793 + }, + { + "epoch": 0.42355535108202186, + "grad_norm": 0.06884765625, + "learning_rate": 0.0010912797366307275, + "loss": 0.0918, + "step": 48794 + }, + { + "epoch": 0.423564031562226, + "grad_norm": 0.89453125, + "learning_rate": 0.0010912499755758687, + "loss": 0.0928, + "step": 48795 + }, + { + "epoch": 0.4235727120424302, + "grad_norm": 0.52734375, + "learning_rate": 0.0010912202145305792, + "loss": 0.0898, + "step": 48796 + }, + { + "epoch": 0.4235813925226343, + "grad_norm": 0.419921875, + "learning_rate": 0.0010911904534948909, + "loss": 0.1152, + "step": 48797 + }, + { + "epoch": 0.4235900730028385, + "grad_norm": 0.263671875, + "learning_rate": 0.001091160692468837, + "loss": 0.083, + "step": 48798 + }, + { + "epoch": 0.42359875348304266, + "grad_norm": 0.1943359375, + "learning_rate": 0.001091130931452449, + "loss": 0.0913, + "step": 48799 + }, + { + "epoch": 0.42360743396324685, + "grad_norm": 0.150390625, + "learning_rate": 0.0010911011704457603, + "loss": 0.0825, + "step": 48800 + }, + { + "epoch": 0.423616114443451, + "grad_norm": 0.53125, + "learning_rate": 0.0010910714094488033, + "loss": 0.1328, + "step": 48801 + }, + { + "epoch": 0.4236247949236552, + "grad_norm": 0.330078125, + "learning_rate": 0.0010910416484616106, + "loss": 0.1035, + "step": 48802 + }, + { + "epoch": 0.4236334754038593, + "grad_norm": 0.21484375, + "learning_rate": 0.0010910118874842146, + "loss": 0.1157, + "step": 48803 + }, + { + "epoch": 0.4236421558840635, + "grad_norm": 0.10693359375, + "learning_rate": 0.0010909821265166475, + "loss": 0.0918, + "step": 48804 + }, + { + "epoch": 0.42365083636426765, + "grad_norm": 0.259765625, + "learning_rate": 0.0010909523655589426, + "loss": 0.0947, + "step": 48805 + }, + { + "epoch": 0.42365951684447184, + "grad_norm": 0.125, + "learning_rate": 0.0010909226046111318, + "loss": 0.0698, + "step": 48806 + }, + { + "epoch": 0.423668197324676, + "grad_norm": 0.330078125, + "learning_rate": 0.0010908928436732482, + "loss": 0.0703, + "step": 48807 + }, + { + "epoch": 0.42367687780488017, + "grad_norm": 0.314453125, + "learning_rate": 0.0010908630827453237, + "loss": 0.0771, + "step": 48808 + }, + { + "epoch": 0.4236855582850843, + "grad_norm": 1.0, + "learning_rate": 0.0010908333218273914, + "loss": 0.1133, + "step": 48809 + }, + { + "epoch": 0.4236942387652885, + "grad_norm": 0.09765625, + "learning_rate": 0.0010908035609194836, + "loss": 0.0806, + "step": 48810 + }, + { + "epoch": 0.42370291924549264, + "grad_norm": 0.236328125, + "learning_rate": 0.0010907738000216329, + "loss": 0.1611, + "step": 48811 + }, + { + "epoch": 0.42371159972569683, + "grad_norm": 0.298828125, + "learning_rate": 0.0010907440391338722, + "loss": 0.0859, + "step": 48812 + }, + { + "epoch": 0.42372028020590097, + "grad_norm": 0.53515625, + "learning_rate": 0.001090714278256233, + "loss": 0.1367, + "step": 48813 + }, + { + "epoch": 0.42372896068610516, + "grad_norm": 0.52734375, + "learning_rate": 0.0010906845173887488, + "loss": 0.0874, + "step": 48814 + }, + { + "epoch": 0.4237376411663093, + "grad_norm": 0.173828125, + "learning_rate": 0.001090654756531452, + "loss": 0.1021, + "step": 48815 + }, + { + "epoch": 0.4237463216465135, + "grad_norm": 0.47265625, + "learning_rate": 0.0010906249956843751, + "loss": 0.0781, + "step": 48816 + }, + { + "epoch": 0.42375500212671763, + "grad_norm": 0.5859375, + "learning_rate": 0.00109059523484755, + "loss": 0.0796, + "step": 48817 + }, + { + "epoch": 0.4237636826069218, + "grad_norm": 0.29296875, + "learning_rate": 0.0010905654740210103, + "loss": 0.1079, + "step": 48818 + }, + { + "epoch": 0.42377236308712596, + "grad_norm": 0.271484375, + "learning_rate": 0.0010905357132047879, + "loss": 0.0781, + "step": 48819 + }, + { + "epoch": 0.42378104356733015, + "grad_norm": 0.0791015625, + "learning_rate": 0.0010905059523989159, + "loss": 0.1484, + "step": 48820 + }, + { + "epoch": 0.4237897240475343, + "grad_norm": 0.09423828125, + "learning_rate": 0.001090476191603426, + "loss": 0.0884, + "step": 48821 + }, + { + "epoch": 0.4237984045277385, + "grad_norm": 0.177734375, + "learning_rate": 0.0010904464308183514, + "loss": 0.0986, + "step": 48822 + }, + { + "epoch": 0.4238070850079426, + "grad_norm": 0.2451171875, + "learning_rate": 0.0010904166700437241, + "loss": 0.1328, + "step": 48823 + }, + { + "epoch": 0.4238157654881468, + "grad_norm": 0.46875, + "learning_rate": 0.0010903869092795775, + "loss": 0.1206, + "step": 48824 + }, + { + "epoch": 0.42382444596835095, + "grad_norm": 0.404296875, + "learning_rate": 0.0010903571485259434, + "loss": 0.0762, + "step": 48825 + }, + { + "epoch": 0.42383312644855514, + "grad_norm": 0.3203125, + "learning_rate": 0.001090327387782854, + "loss": 0.0913, + "step": 48826 + }, + { + "epoch": 0.4238418069287593, + "grad_norm": 0.205078125, + "learning_rate": 0.0010902976270503431, + "loss": 0.0928, + "step": 48827 + }, + { + "epoch": 0.42385048740896347, + "grad_norm": 0.1767578125, + "learning_rate": 0.0010902678663284426, + "loss": 0.1045, + "step": 48828 + }, + { + "epoch": 0.4238591678891676, + "grad_norm": 0.09716796875, + "learning_rate": 0.0010902381056171848, + "loss": 0.0981, + "step": 48829 + }, + { + "epoch": 0.4238678483693718, + "grad_norm": 0.197265625, + "learning_rate": 0.0010902083449166027, + "loss": 0.0742, + "step": 48830 + }, + { + "epoch": 0.42387652884957594, + "grad_norm": 0.09521484375, + "learning_rate": 0.0010901785842267284, + "loss": 0.0737, + "step": 48831 + }, + { + "epoch": 0.42388520932978013, + "grad_norm": 1.1953125, + "learning_rate": 0.0010901488235475945, + "loss": 0.1777, + "step": 48832 + }, + { + "epoch": 0.42389388980998427, + "grad_norm": 0.10693359375, + "learning_rate": 0.001090119062879234, + "loss": 0.0801, + "step": 48833 + }, + { + "epoch": 0.42390257029018846, + "grad_norm": 0.1298828125, + "learning_rate": 0.001090089302221679, + "loss": 0.0923, + "step": 48834 + }, + { + "epoch": 0.4239112507703926, + "grad_norm": 0.11279296875, + "learning_rate": 0.001090059541574962, + "loss": 0.0786, + "step": 48835 + }, + { + "epoch": 0.4239199312505968, + "grad_norm": 0.1142578125, + "learning_rate": 0.001090029780939116, + "loss": 0.0815, + "step": 48836 + }, + { + "epoch": 0.42392861173080093, + "grad_norm": 0.10693359375, + "learning_rate": 0.0010900000203141733, + "loss": 0.1108, + "step": 48837 + }, + { + "epoch": 0.4239372922110051, + "grad_norm": 1.015625, + "learning_rate": 0.0010899702597001663, + "loss": 0.1523, + "step": 48838 + }, + { + "epoch": 0.42394597269120926, + "grad_norm": 0.27734375, + "learning_rate": 0.0010899404990971277, + "loss": 0.104, + "step": 48839 + }, + { + "epoch": 0.42395465317141345, + "grad_norm": 0.130859375, + "learning_rate": 0.0010899107385050903, + "loss": 0.0786, + "step": 48840 + }, + { + "epoch": 0.4239633336516176, + "grad_norm": 0.6953125, + "learning_rate": 0.001089880977924086, + "loss": 0.0947, + "step": 48841 + }, + { + "epoch": 0.4239720141318218, + "grad_norm": 0.314453125, + "learning_rate": 0.0010898512173541477, + "loss": 0.0708, + "step": 48842 + }, + { + "epoch": 0.4239806946120259, + "grad_norm": 0.291015625, + "learning_rate": 0.0010898214567953081, + "loss": 0.1074, + "step": 48843 + }, + { + "epoch": 0.4239893750922301, + "grad_norm": 0.2392578125, + "learning_rate": 0.0010897916962475994, + "loss": 0.1016, + "step": 48844 + }, + { + "epoch": 0.42399805557243425, + "grad_norm": 0.126953125, + "learning_rate": 0.0010897619357110547, + "loss": 0.1123, + "step": 48845 + }, + { + "epoch": 0.42400673605263844, + "grad_norm": 0.228515625, + "learning_rate": 0.001089732175185706, + "loss": 0.0815, + "step": 48846 + }, + { + "epoch": 0.4240154165328426, + "grad_norm": 0.1650390625, + "learning_rate": 0.0010897024146715862, + "loss": 0.1416, + "step": 48847 + }, + { + "epoch": 0.4240240970130468, + "grad_norm": 0.08447265625, + "learning_rate": 0.0010896726541687277, + "loss": 0.1045, + "step": 48848 + }, + { + "epoch": 0.4240327774932509, + "grad_norm": 0.12890625, + "learning_rate": 0.0010896428936771627, + "loss": 0.1001, + "step": 48849 + }, + { + "epoch": 0.4240414579734551, + "grad_norm": 0.1025390625, + "learning_rate": 0.001089613133196924, + "loss": 0.0562, + "step": 48850 + }, + { + "epoch": 0.42405013845365924, + "grad_norm": 0.103515625, + "learning_rate": 0.0010895833727280447, + "loss": 0.0908, + "step": 48851 + }, + { + "epoch": 0.42405881893386344, + "grad_norm": 0.11572265625, + "learning_rate": 0.0010895536122705566, + "loss": 0.0894, + "step": 48852 + }, + { + "epoch": 0.4240674994140676, + "grad_norm": 0.490234375, + "learning_rate": 0.0010895238518244922, + "loss": 0.0898, + "step": 48853 + }, + { + "epoch": 0.42407617989427177, + "grad_norm": 0.12353515625, + "learning_rate": 0.0010894940913898847, + "loss": 0.0796, + "step": 48854 + }, + { + "epoch": 0.4240848603744759, + "grad_norm": 0.1435546875, + "learning_rate": 0.0010894643309667666, + "loss": 0.1201, + "step": 48855 + }, + { + "epoch": 0.4240935408546801, + "grad_norm": 0.1640625, + "learning_rate": 0.0010894345705551698, + "loss": 0.1318, + "step": 48856 + }, + { + "epoch": 0.42410222133488423, + "grad_norm": 0.0966796875, + "learning_rate": 0.0010894048101551272, + "loss": 0.1152, + "step": 48857 + }, + { + "epoch": 0.4241109018150884, + "grad_norm": 0.1044921875, + "learning_rate": 0.0010893750497666718, + "loss": 0.1191, + "step": 48858 + }, + { + "epoch": 0.42411958229529256, + "grad_norm": 0.11474609375, + "learning_rate": 0.0010893452893898353, + "loss": 0.1328, + "step": 48859 + }, + { + "epoch": 0.42412826277549676, + "grad_norm": 0.318359375, + "learning_rate": 0.0010893155290246502, + "loss": 0.0938, + "step": 48860 + }, + { + "epoch": 0.4241369432557009, + "grad_norm": 0.2255859375, + "learning_rate": 0.00108928576867115, + "loss": 0.0996, + "step": 48861 + }, + { + "epoch": 0.4241456237359051, + "grad_norm": 0.1494140625, + "learning_rate": 0.0010892560083293663, + "loss": 0.1377, + "step": 48862 + }, + { + "epoch": 0.4241543042161092, + "grad_norm": 0.2119140625, + "learning_rate": 0.0010892262479993324, + "loss": 0.1348, + "step": 48863 + }, + { + "epoch": 0.4241629846963134, + "grad_norm": 0.361328125, + "learning_rate": 0.0010891964876810808, + "loss": 0.1006, + "step": 48864 + }, + { + "epoch": 0.42417166517651755, + "grad_norm": 0.11572265625, + "learning_rate": 0.0010891667273746435, + "loss": 0.1025, + "step": 48865 + }, + { + "epoch": 0.42418034565672175, + "grad_norm": 0.1025390625, + "learning_rate": 0.0010891369670800533, + "loss": 0.0947, + "step": 48866 + }, + { + "epoch": 0.4241890261369259, + "grad_norm": 0.251953125, + "learning_rate": 0.0010891072067973427, + "loss": 0.0723, + "step": 48867 + }, + { + "epoch": 0.4241977066171301, + "grad_norm": 0.259765625, + "learning_rate": 0.0010890774465265442, + "loss": 0.084, + "step": 48868 + }, + { + "epoch": 0.4242063870973342, + "grad_norm": 0.439453125, + "learning_rate": 0.0010890476862676903, + "loss": 0.0928, + "step": 48869 + }, + { + "epoch": 0.4242150675775384, + "grad_norm": 0.1455078125, + "learning_rate": 0.0010890179260208141, + "loss": 0.1133, + "step": 48870 + }, + { + "epoch": 0.42422374805774254, + "grad_norm": 0.2041015625, + "learning_rate": 0.0010889881657859474, + "loss": 0.0825, + "step": 48871 + }, + { + "epoch": 0.42423242853794674, + "grad_norm": 0.86328125, + "learning_rate": 0.0010889584055631231, + "loss": 0.1025, + "step": 48872 + }, + { + "epoch": 0.4242411090181509, + "grad_norm": 0.478515625, + "learning_rate": 0.001088928645352374, + "loss": 0.1523, + "step": 48873 + }, + { + "epoch": 0.42424978949835507, + "grad_norm": 0.12890625, + "learning_rate": 0.0010888988851537325, + "loss": 0.1074, + "step": 48874 + }, + { + "epoch": 0.4242584699785592, + "grad_norm": 0.6171875, + "learning_rate": 0.0010888691249672307, + "loss": 0.0898, + "step": 48875 + }, + { + "epoch": 0.4242671504587634, + "grad_norm": 0.11181640625, + "learning_rate": 0.0010888393647929015, + "loss": 0.0898, + "step": 48876 + }, + { + "epoch": 0.42427583093896754, + "grad_norm": 0.26953125, + "learning_rate": 0.0010888096046307771, + "loss": 0.0938, + "step": 48877 + }, + { + "epoch": 0.42428451141917173, + "grad_norm": 0.4375, + "learning_rate": 0.0010887798444808908, + "loss": 0.0967, + "step": 48878 + }, + { + "epoch": 0.42429319189937587, + "grad_norm": 0.466796875, + "learning_rate": 0.0010887500843432744, + "loss": 0.0977, + "step": 48879 + }, + { + "epoch": 0.42430187237958006, + "grad_norm": 0.49609375, + "learning_rate": 0.0010887203242179611, + "loss": 0.1387, + "step": 48880 + }, + { + "epoch": 0.4243105528597842, + "grad_norm": 1.28125, + "learning_rate": 0.0010886905641049828, + "loss": 0.1289, + "step": 48881 + }, + { + "epoch": 0.4243192333399884, + "grad_norm": 0.09912109375, + "learning_rate": 0.0010886608040043722, + "loss": 0.0879, + "step": 48882 + }, + { + "epoch": 0.4243279138201925, + "grad_norm": 0.296875, + "learning_rate": 0.0010886310439161624, + "loss": 0.1162, + "step": 48883 + }, + { + "epoch": 0.4243365943003967, + "grad_norm": 0.3046875, + "learning_rate": 0.0010886012838403855, + "loss": 0.1069, + "step": 48884 + }, + { + "epoch": 0.42434527478060086, + "grad_norm": 0.08203125, + "learning_rate": 0.0010885715237770738, + "loss": 0.0654, + "step": 48885 + }, + { + "epoch": 0.42435395526080505, + "grad_norm": 0.1865234375, + "learning_rate": 0.0010885417637262597, + "loss": 0.0996, + "step": 48886 + }, + { + "epoch": 0.4243626357410092, + "grad_norm": 0.16015625, + "learning_rate": 0.0010885120036879768, + "loss": 0.125, + "step": 48887 + }, + { + "epoch": 0.4243713162212134, + "grad_norm": 0.265625, + "learning_rate": 0.0010884822436622568, + "loss": 0.1016, + "step": 48888 + }, + { + "epoch": 0.4243799967014175, + "grad_norm": 0.19921875, + "learning_rate": 0.0010884524836491324, + "loss": 0.0986, + "step": 48889 + }, + { + "epoch": 0.4243886771816217, + "grad_norm": 0.205078125, + "learning_rate": 0.0010884227236486362, + "loss": 0.084, + "step": 48890 + }, + { + "epoch": 0.42439735766182585, + "grad_norm": 0.1015625, + "learning_rate": 0.001088392963660801, + "loss": 0.0977, + "step": 48891 + }, + { + "epoch": 0.42440603814203004, + "grad_norm": 0.482421875, + "learning_rate": 0.0010883632036856588, + "loss": 0.083, + "step": 48892 + }, + { + "epoch": 0.4244147186222342, + "grad_norm": 0.1396484375, + "learning_rate": 0.0010883334437232425, + "loss": 0.1035, + "step": 48893 + }, + { + "epoch": 0.42442339910243837, + "grad_norm": 0.58203125, + "learning_rate": 0.0010883036837735847, + "loss": 0.0908, + "step": 48894 + }, + { + "epoch": 0.4244320795826425, + "grad_norm": 0.89453125, + "learning_rate": 0.0010882739238367173, + "loss": 0.1221, + "step": 48895 + }, + { + "epoch": 0.4244407600628467, + "grad_norm": 0.51171875, + "learning_rate": 0.001088244163912674, + "loss": 0.1172, + "step": 48896 + }, + { + "epoch": 0.42444944054305084, + "grad_norm": 0.3203125, + "learning_rate": 0.0010882144040014862, + "loss": 0.1465, + "step": 48897 + }, + { + "epoch": 0.42445812102325503, + "grad_norm": 0.314453125, + "learning_rate": 0.0010881846441031873, + "loss": 0.0869, + "step": 48898 + }, + { + "epoch": 0.42446680150345917, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010881548842178095, + "loss": 0.1465, + "step": 48899 + }, + { + "epoch": 0.42447548198366336, + "grad_norm": 0.2890625, + "learning_rate": 0.0010881251243453852, + "loss": 0.082, + "step": 48900 + }, + { + "epoch": 0.4244841624638675, + "grad_norm": 0.5078125, + "learning_rate": 0.0010880953644859471, + "loss": 0.123, + "step": 48901 + }, + { + "epoch": 0.4244928429440717, + "grad_norm": 0.08837890625, + "learning_rate": 0.001088065604639528, + "loss": 0.085, + "step": 48902 + }, + { + "epoch": 0.42450152342427583, + "grad_norm": 0.294921875, + "learning_rate": 0.0010880358448061593, + "loss": 0.0859, + "step": 48903 + }, + { + "epoch": 0.42451020390448, + "grad_norm": 0.158203125, + "learning_rate": 0.0010880060849858752, + "loss": 0.1318, + "step": 48904 + }, + { + "epoch": 0.42451888438468416, + "grad_norm": 0.380859375, + "learning_rate": 0.0010879763251787074, + "loss": 0.0898, + "step": 48905 + }, + { + "epoch": 0.42452756486488835, + "grad_norm": 0.1337890625, + "learning_rate": 0.0010879465653846885, + "loss": 0.0918, + "step": 48906 + }, + { + "epoch": 0.4245362453450925, + "grad_norm": 0.267578125, + "learning_rate": 0.001087916805603851, + "loss": 0.0771, + "step": 48907 + }, + { + "epoch": 0.4245449258252967, + "grad_norm": 0.1748046875, + "learning_rate": 0.0010878870458362274, + "loss": 0.051, + "step": 48908 + }, + { + "epoch": 0.4245536063055008, + "grad_norm": 0.26953125, + "learning_rate": 0.0010878572860818505, + "loss": 0.0688, + "step": 48909 + }, + { + "epoch": 0.424562286785705, + "grad_norm": 0.359375, + "learning_rate": 0.0010878275263407526, + "loss": 0.1396, + "step": 48910 + }, + { + "epoch": 0.42457096726590915, + "grad_norm": 0.326171875, + "learning_rate": 0.0010877977666129664, + "loss": 0.0977, + "step": 48911 + }, + { + "epoch": 0.42457964774611334, + "grad_norm": 0.75390625, + "learning_rate": 0.001087768006898524, + "loss": 0.1172, + "step": 48912 + }, + { + "epoch": 0.4245883282263175, + "grad_norm": 0.09619140625, + "learning_rate": 0.0010877382471974587, + "loss": 0.0977, + "step": 48913 + }, + { + "epoch": 0.4245970087065217, + "grad_norm": 0.455078125, + "learning_rate": 0.0010877084875098027, + "loss": 0.1201, + "step": 48914 + }, + { + "epoch": 0.4246056891867258, + "grad_norm": 0.11181640625, + "learning_rate": 0.0010876787278355884, + "loss": 0.1128, + "step": 48915 + }, + { + "epoch": 0.42461436966692995, + "grad_norm": 0.326171875, + "learning_rate": 0.0010876489681748485, + "loss": 0.1177, + "step": 48916 + }, + { + "epoch": 0.42462305014713414, + "grad_norm": 0.1552734375, + "learning_rate": 0.0010876192085276157, + "loss": 0.1055, + "step": 48917 + }, + { + "epoch": 0.4246317306273383, + "grad_norm": 0.32421875, + "learning_rate": 0.001087589448893922, + "loss": 0.0874, + "step": 48918 + }, + { + "epoch": 0.42464041110754247, + "grad_norm": 0.208984375, + "learning_rate": 0.0010875596892738005, + "loss": 0.127, + "step": 48919 + }, + { + "epoch": 0.4246490915877466, + "grad_norm": 0.69921875, + "learning_rate": 0.0010875299296672833, + "loss": 0.0913, + "step": 48920 + }, + { + "epoch": 0.4246577720679508, + "grad_norm": 0.609375, + "learning_rate": 0.001087500170074403, + "loss": 0.1182, + "step": 48921 + }, + { + "epoch": 0.42466645254815494, + "grad_norm": 0.52734375, + "learning_rate": 0.0010874704104951928, + "loss": 0.0894, + "step": 48922 + }, + { + "epoch": 0.42467513302835913, + "grad_norm": 0.244140625, + "learning_rate": 0.0010874406509296848, + "loss": 0.0879, + "step": 48923 + }, + { + "epoch": 0.42468381350856327, + "grad_norm": 0.2138671875, + "learning_rate": 0.0010874108913779114, + "loss": 0.0928, + "step": 48924 + }, + { + "epoch": 0.42469249398876746, + "grad_norm": 0.75, + "learning_rate": 0.0010873811318399054, + "loss": 0.1582, + "step": 48925 + }, + { + "epoch": 0.4247011744689716, + "grad_norm": 0.455078125, + "learning_rate": 0.001087351372315699, + "loss": 0.1167, + "step": 48926 + }, + { + "epoch": 0.4247098549491758, + "grad_norm": 0.099609375, + "learning_rate": 0.0010873216128053247, + "loss": 0.1035, + "step": 48927 + }, + { + "epoch": 0.42471853542937993, + "grad_norm": 0.1494140625, + "learning_rate": 0.0010872918533088155, + "loss": 0.1069, + "step": 48928 + }, + { + "epoch": 0.4247272159095841, + "grad_norm": 0.11669921875, + "learning_rate": 0.0010872620938262038, + "loss": 0.0972, + "step": 48929 + }, + { + "epoch": 0.42473589638978826, + "grad_norm": 0.330078125, + "learning_rate": 0.0010872323343575222, + "loss": 0.0894, + "step": 48930 + }, + { + "epoch": 0.42474457686999245, + "grad_norm": 0.1962890625, + "learning_rate": 0.0010872025749028029, + "loss": 0.1377, + "step": 48931 + }, + { + "epoch": 0.4247532573501966, + "grad_norm": 0.09423828125, + "learning_rate": 0.001087172815462079, + "loss": 0.0894, + "step": 48932 + }, + { + "epoch": 0.4247619378304008, + "grad_norm": 0.091796875, + "learning_rate": 0.0010871430560353825, + "loss": 0.0913, + "step": 48933 + }, + { + "epoch": 0.4247706183106049, + "grad_norm": 0.2265625, + "learning_rate": 0.0010871132966227461, + "loss": 0.1011, + "step": 48934 + }, + { + "epoch": 0.4247792987908091, + "grad_norm": 0.1650390625, + "learning_rate": 0.0010870835372242028, + "loss": 0.0869, + "step": 48935 + }, + { + "epoch": 0.42478797927101325, + "grad_norm": 0.412109375, + "learning_rate": 0.0010870537778397845, + "loss": 0.0908, + "step": 48936 + }, + { + "epoch": 0.42479665975121744, + "grad_norm": 1.2109375, + "learning_rate": 0.0010870240184695235, + "loss": 0.1387, + "step": 48937 + }, + { + "epoch": 0.4248053402314216, + "grad_norm": 0.212890625, + "learning_rate": 0.0010869942591134531, + "loss": 0.1226, + "step": 48938 + }, + { + "epoch": 0.4248140207116258, + "grad_norm": 0.55078125, + "learning_rate": 0.001086964499771606, + "loss": 0.1021, + "step": 48939 + }, + { + "epoch": 0.4248227011918299, + "grad_norm": 0.375, + "learning_rate": 0.0010869347404440137, + "loss": 0.1016, + "step": 48940 + }, + { + "epoch": 0.4248313816720341, + "grad_norm": 0.84765625, + "learning_rate": 0.00108690498113071, + "loss": 0.1445, + "step": 48941 + }, + { + "epoch": 0.42484006215223824, + "grad_norm": 0.41796875, + "learning_rate": 0.0010868752218317263, + "loss": 0.0942, + "step": 48942 + }, + { + "epoch": 0.42484874263244243, + "grad_norm": 0.2470703125, + "learning_rate": 0.0010868454625470962, + "loss": 0.1182, + "step": 48943 + }, + { + "epoch": 0.42485742311264657, + "grad_norm": 0.11328125, + "learning_rate": 0.0010868157032768516, + "loss": 0.0967, + "step": 48944 + }, + { + "epoch": 0.42486610359285076, + "grad_norm": 0.1630859375, + "learning_rate": 0.0010867859440210248, + "loss": 0.084, + "step": 48945 + }, + { + "epoch": 0.4248747840730549, + "grad_norm": 0.51953125, + "learning_rate": 0.0010867561847796488, + "loss": 0.1011, + "step": 48946 + }, + { + "epoch": 0.4248834645532591, + "grad_norm": 0.1494140625, + "learning_rate": 0.0010867264255527561, + "loss": 0.082, + "step": 48947 + }, + { + "epoch": 0.42489214503346323, + "grad_norm": 0.44140625, + "learning_rate": 0.001086696666340379, + "loss": 0.1211, + "step": 48948 + }, + { + "epoch": 0.4249008255136674, + "grad_norm": 0.1552734375, + "learning_rate": 0.0010866669071425503, + "loss": 0.0991, + "step": 48949 + }, + { + "epoch": 0.42490950599387156, + "grad_norm": 0.1015625, + "learning_rate": 0.0010866371479593027, + "loss": 0.1211, + "step": 48950 + }, + { + "epoch": 0.42491818647407575, + "grad_norm": 0.33203125, + "learning_rate": 0.0010866073887906683, + "loss": 0.0957, + "step": 48951 + }, + { + "epoch": 0.4249268669542799, + "grad_norm": 0.2333984375, + "learning_rate": 0.00108657762963668, + "loss": 0.0889, + "step": 48952 + }, + { + "epoch": 0.4249355474344841, + "grad_norm": 0.201171875, + "learning_rate": 0.0010865478704973704, + "loss": 0.0522, + "step": 48953 + }, + { + "epoch": 0.4249442279146882, + "grad_norm": 0.208984375, + "learning_rate": 0.0010865181113727715, + "loss": 0.0908, + "step": 48954 + }, + { + "epoch": 0.4249529083948924, + "grad_norm": 2.046875, + "learning_rate": 0.0010864883522629161, + "loss": 0.209, + "step": 48955 + }, + { + "epoch": 0.42496158887509655, + "grad_norm": 0.640625, + "learning_rate": 0.0010864585931678371, + "loss": 0.1167, + "step": 48956 + }, + { + "epoch": 0.42497026935530074, + "grad_norm": 0.1240234375, + "learning_rate": 0.0010864288340875668, + "loss": 0.0669, + "step": 48957 + }, + { + "epoch": 0.4249789498355049, + "grad_norm": 0.11767578125, + "learning_rate": 0.0010863990750221372, + "loss": 0.1572, + "step": 48958 + }, + { + "epoch": 0.4249876303157091, + "grad_norm": 0.71484375, + "learning_rate": 0.0010863693159715822, + "loss": 0.1016, + "step": 48959 + }, + { + "epoch": 0.4249963107959132, + "grad_norm": 0.1142578125, + "learning_rate": 0.0010863395569359329, + "loss": 0.105, + "step": 48960 + }, + { + "epoch": 0.4250049912761174, + "grad_norm": 0.5703125, + "learning_rate": 0.0010863097979152228, + "loss": 0.1543, + "step": 48961 + }, + { + "epoch": 0.42501367175632154, + "grad_norm": 0.1259765625, + "learning_rate": 0.001086280038909484, + "loss": 0.125, + "step": 48962 + }, + { + "epoch": 0.42502235223652574, + "grad_norm": 0.1142578125, + "learning_rate": 0.0010862502799187492, + "loss": 0.083, + "step": 48963 + }, + { + "epoch": 0.4250310327167299, + "grad_norm": 0.29296875, + "learning_rate": 0.0010862205209430504, + "loss": 0.1196, + "step": 48964 + }, + { + "epoch": 0.42503971319693407, + "grad_norm": 0.1845703125, + "learning_rate": 0.001086190761982421, + "loss": 0.1221, + "step": 48965 + }, + { + "epoch": 0.4250483936771382, + "grad_norm": 0.09912109375, + "learning_rate": 0.0010861610030368934, + "loss": 0.1006, + "step": 48966 + }, + { + "epoch": 0.4250570741573424, + "grad_norm": 0.400390625, + "learning_rate": 0.0010861312441064995, + "loss": 0.0874, + "step": 48967 + }, + { + "epoch": 0.42506575463754653, + "grad_norm": 0.10400390625, + "learning_rate": 0.0010861014851912725, + "loss": 0.0752, + "step": 48968 + }, + { + "epoch": 0.4250744351177507, + "grad_norm": 0.93359375, + "learning_rate": 0.0010860717262912447, + "loss": 0.0972, + "step": 48969 + }, + { + "epoch": 0.42508311559795486, + "grad_norm": 0.447265625, + "learning_rate": 0.0010860419674064488, + "loss": 0.123, + "step": 48970 + }, + { + "epoch": 0.42509179607815906, + "grad_norm": 0.111328125, + "learning_rate": 0.001086012208536917, + "loss": 0.1025, + "step": 48971 + }, + { + "epoch": 0.4251004765583632, + "grad_norm": 0.3515625, + "learning_rate": 0.0010859824496826822, + "loss": 0.0771, + "step": 48972 + }, + { + "epoch": 0.4251091570385674, + "grad_norm": 0.27734375, + "learning_rate": 0.0010859526908437763, + "loss": 0.1025, + "step": 48973 + }, + { + "epoch": 0.4251178375187715, + "grad_norm": 0.2578125, + "learning_rate": 0.001085922932020233, + "loss": 0.0591, + "step": 48974 + }, + { + "epoch": 0.4251265179989757, + "grad_norm": 0.08447265625, + "learning_rate": 0.0010858931732120837, + "loss": 0.0791, + "step": 48975 + }, + { + "epoch": 0.42513519847917985, + "grad_norm": 0.150390625, + "learning_rate": 0.0010858634144193615, + "loss": 0.0957, + "step": 48976 + }, + { + "epoch": 0.42514387895938405, + "grad_norm": 0.10986328125, + "learning_rate": 0.001085833655642099, + "loss": 0.1396, + "step": 48977 + }, + { + "epoch": 0.4251525594395882, + "grad_norm": 0.15234375, + "learning_rate": 0.0010858038968803285, + "loss": 0.0933, + "step": 48978 + }, + { + "epoch": 0.4251612399197924, + "grad_norm": 0.2158203125, + "learning_rate": 0.0010857741381340826, + "loss": 0.1006, + "step": 48979 + }, + { + "epoch": 0.4251699203999965, + "grad_norm": 0.44140625, + "learning_rate": 0.0010857443794033942, + "loss": 0.0928, + "step": 48980 + }, + { + "epoch": 0.4251786008802007, + "grad_norm": 0.1337890625, + "learning_rate": 0.0010857146206882948, + "loss": 0.1289, + "step": 48981 + }, + { + "epoch": 0.42518728136040485, + "grad_norm": 0.12890625, + "learning_rate": 0.0010856848619888182, + "loss": 0.0747, + "step": 48982 + }, + { + "epoch": 0.42519596184060904, + "grad_norm": 0.1923828125, + "learning_rate": 0.0010856551033049963, + "loss": 0.1758, + "step": 48983 + }, + { + "epoch": 0.4252046423208132, + "grad_norm": 0.53515625, + "learning_rate": 0.001085625344636862, + "loss": 0.1465, + "step": 48984 + }, + { + "epoch": 0.42521332280101737, + "grad_norm": 0.5078125, + "learning_rate": 0.0010855955859844476, + "loss": 0.1113, + "step": 48985 + }, + { + "epoch": 0.4252220032812215, + "grad_norm": 0.1904296875, + "learning_rate": 0.0010855658273477853, + "loss": 0.0801, + "step": 48986 + }, + { + "epoch": 0.4252306837614257, + "grad_norm": 0.08056640625, + "learning_rate": 0.0010855360687269083, + "loss": 0.082, + "step": 48987 + }, + { + "epoch": 0.42523936424162984, + "grad_norm": 0.1630859375, + "learning_rate": 0.0010855063101218487, + "loss": 0.0908, + "step": 48988 + }, + { + "epoch": 0.42524804472183403, + "grad_norm": 0.484375, + "learning_rate": 0.0010854765515326394, + "loss": 0.0928, + "step": 48989 + }, + { + "epoch": 0.42525672520203817, + "grad_norm": 0.10205078125, + "learning_rate": 0.001085446792959312, + "loss": 0.0908, + "step": 48990 + }, + { + "epoch": 0.42526540568224236, + "grad_norm": 0.0654296875, + "learning_rate": 0.0010854170344019006, + "loss": 0.0923, + "step": 48991 + }, + { + "epoch": 0.4252740861624465, + "grad_norm": 0.19921875, + "learning_rate": 0.0010853872758604366, + "loss": 0.0972, + "step": 48992 + }, + { + "epoch": 0.4252827666426507, + "grad_norm": 0.1484375, + "learning_rate": 0.001085357517334953, + "loss": 0.1221, + "step": 48993 + }, + { + "epoch": 0.4252914471228548, + "grad_norm": 0.3515625, + "learning_rate": 0.0010853277588254821, + "loss": 0.1016, + "step": 48994 + }, + { + "epoch": 0.425300127603059, + "grad_norm": 0.14453125, + "learning_rate": 0.0010852980003320564, + "loss": 0.0869, + "step": 48995 + }, + { + "epoch": 0.42530880808326316, + "grad_norm": 0.87890625, + "learning_rate": 0.0010852682418547087, + "loss": 0.1943, + "step": 48996 + }, + { + "epoch": 0.42531748856346735, + "grad_norm": 0.359375, + "learning_rate": 0.0010852384833934714, + "loss": 0.1172, + "step": 48997 + }, + { + "epoch": 0.4253261690436715, + "grad_norm": 0.1083984375, + "learning_rate": 0.0010852087249483768, + "loss": 0.0908, + "step": 48998 + }, + { + "epoch": 0.4253348495238757, + "grad_norm": 0.10595703125, + "learning_rate": 0.0010851789665194582, + "loss": 0.082, + "step": 48999 + }, + { + "epoch": 0.4253435300040798, + "grad_norm": 0.333984375, + "learning_rate": 0.0010851492081067476, + "loss": 0.0986, + "step": 49000 + }, + { + "epoch": 0.425352210484284, + "grad_norm": 0.251953125, + "learning_rate": 0.0010851194497102773, + "loss": 0.0986, + "step": 49001 + }, + { + "epoch": 0.42536089096448815, + "grad_norm": 0.55078125, + "learning_rate": 0.0010850896913300804, + "loss": 0.1045, + "step": 49002 + }, + { + "epoch": 0.42536957144469234, + "grad_norm": 0.2734375, + "learning_rate": 0.0010850599329661893, + "loss": 0.0996, + "step": 49003 + }, + { + "epoch": 0.4253782519248965, + "grad_norm": 0.6015625, + "learning_rate": 0.0010850301746186365, + "loss": 0.1309, + "step": 49004 + }, + { + "epoch": 0.42538693240510067, + "grad_norm": 0.3125, + "learning_rate": 0.0010850004162874538, + "loss": 0.0869, + "step": 49005 + }, + { + "epoch": 0.4253956128853048, + "grad_norm": 0.32421875, + "learning_rate": 0.001084970657972675, + "loss": 0.1357, + "step": 49006 + }, + { + "epoch": 0.425404293365509, + "grad_norm": 0.169921875, + "learning_rate": 0.0010849408996743317, + "loss": 0.062, + "step": 49007 + }, + { + "epoch": 0.42541297384571314, + "grad_norm": 0.12353515625, + "learning_rate": 0.001084911141392457, + "loss": 0.0864, + "step": 49008 + }, + { + "epoch": 0.42542165432591733, + "grad_norm": 0.65625, + "learning_rate": 0.0010848813831270832, + "loss": 0.103, + "step": 49009 + }, + { + "epoch": 0.42543033480612147, + "grad_norm": 0.08740234375, + "learning_rate": 0.001084851624878243, + "loss": 0.0762, + "step": 49010 + }, + { + "epoch": 0.42543901528632566, + "grad_norm": 0.4609375, + "learning_rate": 0.001084821866645969, + "loss": 0.0938, + "step": 49011 + }, + { + "epoch": 0.4254476957665298, + "grad_norm": 0.1689453125, + "learning_rate": 0.0010847921084302934, + "loss": 0.0815, + "step": 49012 + }, + { + "epoch": 0.425456376246734, + "grad_norm": 0.3046875, + "learning_rate": 0.001084762350231249, + "loss": 0.0752, + "step": 49013 + }, + { + "epoch": 0.42546505672693813, + "grad_norm": 0.287109375, + "learning_rate": 0.001084732592048868, + "loss": 0.0923, + "step": 49014 + }, + { + "epoch": 0.4254737372071423, + "grad_norm": 0.095703125, + "learning_rate": 0.0010847028338831835, + "loss": 0.0947, + "step": 49015 + }, + { + "epoch": 0.42548241768734646, + "grad_norm": 0.240234375, + "learning_rate": 0.0010846730757342275, + "loss": 0.0898, + "step": 49016 + }, + { + "epoch": 0.42549109816755065, + "grad_norm": 0.65625, + "learning_rate": 0.0010846433176020327, + "loss": 0.0996, + "step": 49017 + }, + { + "epoch": 0.4254997786477548, + "grad_norm": 0.1533203125, + "learning_rate": 0.001084613559486632, + "loss": 0.1465, + "step": 49018 + }, + { + "epoch": 0.425508459127959, + "grad_norm": 0.306640625, + "learning_rate": 0.0010845838013880578, + "loss": 0.0898, + "step": 49019 + }, + { + "epoch": 0.4255171396081631, + "grad_norm": 0.29296875, + "learning_rate": 0.0010845540433063424, + "loss": 0.1113, + "step": 49020 + }, + { + "epoch": 0.4255258200883673, + "grad_norm": 0.41796875, + "learning_rate": 0.0010845242852415185, + "loss": 0.0898, + "step": 49021 + }, + { + "epoch": 0.42553450056857145, + "grad_norm": 0.09814453125, + "learning_rate": 0.0010844945271936185, + "loss": 0.0981, + "step": 49022 + }, + { + "epoch": 0.42554318104877564, + "grad_norm": 0.2470703125, + "learning_rate": 0.0010844647691626748, + "loss": 0.0845, + "step": 49023 + }, + { + "epoch": 0.4255518615289798, + "grad_norm": 0.10205078125, + "learning_rate": 0.0010844350111487205, + "loss": 0.0854, + "step": 49024 + }, + { + "epoch": 0.425560542009184, + "grad_norm": 0.203125, + "learning_rate": 0.0010844052531517878, + "loss": 0.1064, + "step": 49025 + }, + { + "epoch": 0.4255692224893881, + "grad_norm": 0.2265625, + "learning_rate": 0.0010843754951719092, + "loss": 0.0859, + "step": 49026 + }, + { + "epoch": 0.4255779029695923, + "grad_norm": 0.20703125, + "learning_rate": 0.0010843457372091176, + "loss": 0.1025, + "step": 49027 + }, + { + "epoch": 0.42558658344979644, + "grad_norm": 0.291015625, + "learning_rate": 0.001084315979263445, + "loss": 0.124, + "step": 49028 + }, + { + "epoch": 0.42559526393000063, + "grad_norm": 0.185546875, + "learning_rate": 0.0010842862213349242, + "loss": 0.1123, + "step": 49029 + }, + { + "epoch": 0.42560394441020477, + "grad_norm": 0.1787109375, + "learning_rate": 0.0010842564634235882, + "loss": 0.0781, + "step": 49030 + }, + { + "epoch": 0.42561262489040896, + "grad_norm": 0.10595703125, + "learning_rate": 0.0010842267055294686, + "loss": 0.0767, + "step": 49031 + }, + { + "epoch": 0.4256213053706131, + "grad_norm": 0.23828125, + "learning_rate": 0.0010841969476525983, + "loss": 0.1177, + "step": 49032 + }, + { + "epoch": 0.4256299858508173, + "grad_norm": 0.6015625, + "learning_rate": 0.0010841671897930103, + "loss": 0.0967, + "step": 49033 + }, + { + "epoch": 0.42563866633102143, + "grad_norm": 0.3203125, + "learning_rate": 0.0010841374319507369, + "loss": 0.0898, + "step": 49034 + }, + { + "epoch": 0.4256473468112256, + "grad_norm": 0.10107421875, + "learning_rate": 0.0010841076741258102, + "loss": 0.1084, + "step": 49035 + }, + { + "epoch": 0.42565602729142976, + "grad_norm": 0.138671875, + "learning_rate": 0.0010840779163182634, + "loss": 0.084, + "step": 49036 + }, + { + "epoch": 0.42566470777163395, + "grad_norm": 0.392578125, + "learning_rate": 0.0010840481585281288, + "loss": 0.1079, + "step": 49037 + }, + { + "epoch": 0.4256733882518381, + "grad_norm": 0.10009765625, + "learning_rate": 0.001084018400755439, + "loss": 0.1182, + "step": 49038 + }, + { + "epoch": 0.42568206873204223, + "grad_norm": 0.11474609375, + "learning_rate": 0.0010839886430002261, + "loss": 0.0986, + "step": 49039 + }, + { + "epoch": 0.4256907492122464, + "grad_norm": 2.453125, + "learning_rate": 0.0010839588852625233, + "loss": 0.2812, + "step": 49040 + }, + { + "epoch": 0.42569942969245056, + "grad_norm": 0.1181640625, + "learning_rate": 0.0010839291275423627, + "loss": 0.0986, + "step": 49041 + }, + { + "epoch": 0.42570811017265475, + "grad_norm": 0.2412109375, + "learning_rate": 0.0010838993698397766, + "loss": 0.1396, + "step": 49042 + }, + { + "epoch": 0.4257167906528589, + "grad_norm": 0.154296875, + "learning_rate": 0.001083869612154798, + "loss": 0.0908, + "step": 49043 + }, + { + "epoch": 0.4257254711330631, + "grad_norm": 0.2197265625, + "learning_rate": 0.0010838398544874596, + "loss": 0.1484, + "step": 49044 + }, + { + "epoch": 0.4257341516132672, + "grad_norm": 0.07763671875, + "learning_rate": 0.0010838100968377938, + "loss": 0.0781, + "step": 49045 + }, + { + "epoch": 0.4257428320934714, + "grad_norm": 0.484375, + "learning_rate": 0.0010837803392058328, + "loss": 0.1006, + "step": 49046 + }, + { + "epoch": 0.42575151257367555, + "grad_norm": 0.244140625, + "learning_rate": 0.0010837505815916096, + "loss": 0.1016, + "step": 49047 + }, + { + "epoch": 0.42576019305387974, + "grad_norm": 0.3203125, + "learning_rate": 0.0010837208239951564, + "loss": 0.0918, + "step": 49048 + }, + { + "epoch": 0.4257688735340839, + "grad_norm": 0.1953125, + "learning_rate": 0.0010836910664165059, + "loss": 0.0591, + "step": 49049 + }, + { + "epoch": 0.4257775540142881, + "grad_norm": 0.86328125, + "learning_rate": 0.0010836613088556902, + "loss": 0.1133, + "step": 49050 + }, + { + "epoch": 0.4257862344944922, + "grad_norm": 0.08154296875, + "learning_rate": 0.0010836315513127427, + "loss": 0.0781, + "step": 49051 + }, + { + "epoch": 0.4257949149746964, + "grad_norm": 0.294921875, + "learning_rate": 0.0010836017937876953, + "loss": 0.082, + "step": 49052 + }, + { + "epoch": 0.42580359545490054, + "grad_norm": 0.259765625, + "learning_rate": 0.001083572036280581, + "loss": 0.0996, + "step": 49053 + }, + { + "epoch": 0.42581227593510473, + "grad_norm": 0.1943359375, + "learning_rate": 0.0010835422787914318, + "loss": 0.0898, + "step": 49054 + }, + { + "epoch": 0.42582095641530887, + "grad_norm": 0.15625, + "learning_rate": 0.0010835125213202808, + "loss": 0.1211, + "step": 49055 + }, + { + "epoch": 0.42582963689551306, + "grad_norm": 0.67578125, + "learning_rate": 0.00108348276386716, + "loss": 0.0811, + "step": 49056 + }, + { + "epoch": 0.4258383173757172, + "grad_norm": 0.294921875, + "learning_rate": 0.0010834530064321023, + "loss": 0.1123, + "step": 49057 + }, + { + "epoch": 0.4258469978559214, + "grad_norm": 0.2470703125, + "learning_rate": 0.0010834232490151403, + "loss": 0.062, + "step": 49058 + }, + { + "epoch": 0.42585567833612553, + "grad_norm": 0.302734375, + "learning_rate": 0.0010833934916163058, + "loss": 0.1445, + "step": 49059 + }, + { + "epoch": 0.4258643588163297, + "grad_norm": 0.07080078125, + "learning_rate": 0.0010833637342356323, + "loss": 0.0742, + "step": 49060 + }, + { + "epoch": 0.42587303929653386, + "grad_norm": 0.2177734375, + "learning_rate": 0.0010833339768731522, + "loss": 0.1289, + "step": 49061 + }, + { + "epoch": 0.42588171977673805, + "grad_norm": 0.1416015625, + "learning_rate": 0.0010833042195288977, + "loss": 0.1436, + "step": 49062 + }, + { + "epoch": 0.4258904002569422, + "grad_norm": 0.359375, + "learning_rate": 0.0010832744622029011, + "loss": 0.1084, + "step": 49063 + }, + { + "epoch": 0.4258990807371464, + "grad_norm": 0.33984375, + "learning_rate": 0.0010832447048951957, + "loss": 0.1035, + "step": 49064 + }, + { + "epoch": 0.4259077612173505, + "grad_norm": 0.181640625, + "learning_rate": 0.0010832149476058136, + "loss": 0.0981, + "step": 49065 + }, + { + "epoch": 0.4259164416975547, + "grad_norm": 0.154296875, + "learning_rate": 0.0010831851903347874, + "loss": 0.0996, + "step": 49066 + }, + { + "epoch": 0.42592512217775885, + "grad_norm": 0.115234375, + "learning_rate": 0.0010831554330821493, + "loss": 0.1055, + "step": 49067 + }, + { + "epoch": 0.42593380265796305, + "grad_norm": 0.2451171875, + "learning_rate": 0.0010831256758479322, + "loss": 0.1099, + "step": 49068 + }, + { + "epoch": 0.4259424831381672, + "grad_norm": 0.173828125, + "learning_rate": 0.0010830959186321689, + "loss": 0.0933, + "step": 49069 + }, + { + "epoch": 0.4259511636183714, + "grad_norm": 0.890625, + "learning_rate": 0.0010830661614348916, + "loss": 0.3242, + "step": 49070 + }, + { + "epoch": 0.4259598440985755, + "grad_norm": 0.1865234375, + "learning_rate": 0.0010830364042561328, + "loss": 0.0801, + "step": 49071 + }, + { + "epoch": 0.4259685245787797, + "grad_norm": 0.197265625, + "learning_rate": 0.001083006647095925, + "loss": 0.0918, + "step": 49072 + }, + { + "epoch": 0.42597720505898384, + "grad_norm": 0.404296875, + "learning_rate": 0.001082976889954301, + "loss": 0.0874, + "step": 49073 + }, + { + "epoch": 0.42598588553918804, + "grad_norm": 0.2294921875, + "learning_rate": 0.0010829471328312934, + "loss": 0.0791, + "step": 49074 + }, + { + "epoch": 0.4259945660193922, + "grad_norm": 0.55078125, + "learning_rate": 0.0010829173757269347, + "loss": 0.0972, + "step": 49075 + }, + { + "epoch": 0.42600324649959637, + "grad_norm": 0.216796875, + "learning_rate": 0.0010828876186412563, + "loss": 0.1025, + "step": 49076 + }, + { + "epoch": 0.4260119269798005, + "grad_norm": 0.173828125, + "learning_rate": 0.0010828578615742926, + "loss": 0.0918, + "step": 49077 + }, + { + "epoch": 0.4260206074600047, + "grad_norm": 0.130859375, + "learning_rate": 0.0010828281045260752, + "loss": 0.0645, + "step": 49078 + }, + { + "epoch": 0.42602928794020883, + "grad_norm": 0.2021484375, + "learning_rate": 0.0010827983474966368, + "loss": 0.1973, + "step": 49079 + }, + { + "epoch": 0.426037968420413, + "grad_norm": 0.236328125, + "learning_rate": 0.00108276859048601, + "loss": 0.0879, + "step": 49080 + }, + { + "epoch": 0.42604664890061716, + "grad_norm": 0.07080078125, + "learning_rate": 0.0010827388334942267, + "loss": 0.0723, + "step": 49081 + }, + { + "epoch": 0.42605532938082136, + "grad_norm": 0.1474609375, + "learning_rate": 0.0010827090765213202, + "loss": 0.0938, + "step": 49082 + }, + { + "epoch": 0.4260640098610255, + "grad_norm": 0.2333984375, + "learning_rate": 0.0010826793195673227, + "loss": 0.1011, + "step": 49083 + }, + { + "epoch": 0.4260726903412297, + "grad_norm": 0.171875, + "learning_rate": 0.001082649562632267, + "loss": 0.1055, + "step": 49084 + }, + { + "epoch": 0.4260813708214338, + "grad_norm": 0.09033203125, + "learning_rate": 0.0010826198057161851, + "loss": 0.0771, + "step": 49085 + }, + { + "epoch": 0.426090051301638, + "grad_norm": 0.5, + "learning_rate": 0.0010825900488191102, + "loss": 0.1172, + "step": 49086 + }, + { + "epoch": 0.42609873178184215, + "grad_norm": 0.16796875, + "learning_rate": 0.0010825602919410746, + "loss": 0.1011, + "step": 49087 + }, + { + "epoch": 0.42610741226204635, + "grad_norm": 0.2373046875, + "learning_rate": 0.0010825305350821112, + "loss": 0.1196, + "step": 49088 + }, + { + "epoch": 0.4261160927422505, + "grad_norm": 0.51953125, + "learning_rate": 0.0010825007782422516, + "loss": 0.1182, + "step": 49089 + }, + { + "epoch": 0.4261247732224547, + "grad_norm": 0.1572265625, + "learning_rate": 0.001082471021421529, + "loss": 0.1113, + "step": 49090 + }, + { + "epoch": 0.4261334537026588, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010824412646199756, + "loss": 0.1328, + "step": 49091 + }, + { + "epoch": 0.426142134182863, + "grad_norm": 0.8125, + "learning_rate": 0.0010824115078376247, + "loss": 0.0889, + "step": 49092 + }, + { + "epoch": 0.42615081466306715, + "grad_norm": 0.46875, + "learning_rate": 0.001082381751074508, + "loss": 0.0928, + "step": 49093 + }, + { + "epoch": 0.42615949514327134, + "grad_norm": 0.296875, + "learning_rate": 0.001082351994330658, + "loss": 0.0996, + "step": 49094 + }, + { + "epoch": 0.4261681756234755, + "grad_norm": 0.10009765625, + "learning_rate": 0.0010823222376061081, + "loss": 0.0825, + "step": 49095 + }, + { + "epoch": 0.42617685610367967, + "grad_norm": 0.921875, + "learning_rate": 0.0010822924809008902, + "loss": 0.1387, + "step": 49096 + }, + { + "epoch": 0.4261855365838838, + "grad_norm": 0.0869140625, + "learning_rate": 0.0010822627242150372, + "loss": 0.0815, + "step": 49097 + }, + { + "epoch": 0.426194217064088, + "grad_norm": 0.54296875, + "learning_rate": 0.0010822329675485812, + "loss": 0.1025, + "step": 49098 + }, + { + "epoch": 0.42620289754429214, + "grad_norm": 0.376953125, + "learning_rate": 0.001082203210901555, + "loss": 0.1074, + "step": 49099 + }, + { + "epoch": 0.42621157802449633, + "grad_norm": 0.1796875, + "learning_rate": 0.001082173454273991, + "loss": 0.0908, + "step": 49100 + }, + { + "epoch": 0.42622025850470047, + "grad_norm": 0.2353515625, + "learning_rate": 0.0010821436976659218, + "loss": 0.0801, + "step": 49101 + }, + { + "epoch": 0.42622893898490466, + "grad_norm": 0.61328125, + "learning_rate": 0.0010821139410773799, + "loss": 0.0854, + "step": 49102 + }, + { + "epoch": 0.4262376194651088, + "grad_norm": 0.28515625, + "learning_rate": 0.0010820841845083982, + "loss": 0.124, + "step": 49103 + }, + { + "epoch": 0.426246299945313, + "grad_norm": 0.0947265625, + "learning_rate": 0.0010820544279590088, + "loss": 0.105, + "step": 49104 + }, + { + "epoch": 0.4262549804255171, + "grad_norm": 0.10205078125, + "learning_rate": 0.0010820246714292445, + "loss": 0.1436, + "step": 49105 + }, + { + "epoch": 0.4262636609057213, + "grad_norm": 0.1708984375, + "learning_rate": 0.001081994914919138, + "loss": 0.0669, + "step": 49106 + }, + { + "epoch": 0.42627234138592546, + "grad_norm": 0.10205078125, + "learning_rate": 0.0010819651584287214, + "loss": 0.1177, + "step": 49107 + }, + { + "epoch": 0.42628102186612965, + "grad_norm": 0.236328125, + "learning_rate": 0.0010819354019580272, + "loss": 0.0957, + "step": 49108 + }, + { + "epoch": 0.4262897023463338, + "grad_norm": 0.4765625, + "learning_rate": 0.0010819056455070885, + "loss": 0.0957, + "step": 49109 + }, + { + "epoch": 0.426298382826538, + "grad_norm": 0.375, + "learning_rate": 0.0010818758890759372, + "loss": 0.0718, + "step": 49110 + }, + { + "epoch": 0.4263070633067421, + "grad_norm": 0.44921875, + "learning_rate": 0.001081846132664606, + "loss": 0.1309, + "step": 49111 + }, + { + "epoch": 0.4263157437869463, + "grad_norm": 0.376953125, + "learning_rate": 0.0010818163762731279, + "loss": 0.0762, + "step": 49112 + }, + { + "epoch": 0.42632442426715045, + "grad_norm": 0.232421875, + "learning_rate": 0.0010817866199015352, + "loss": 0.084, + "step": 49113 + }, + { + "epoch": 0.42633310474735464, + "grad_norm": 0.1630859375, + "learning_rate": 0.0010817568635498602, + "loss": 0.0972, + "step": 49114 + }, + { + "epoch": 0.4263417852275588, + "grad_norm": 0.44140625, + "learning_rate": 0.001081727107218136, + "loss": 0.1602, + "step": 49115 + }, + { + "epoch": 0.42635046570776297, + "grad_norm": 0.162109375, + "learning_rate": 0.0010816973509063946, + "loss": 0.1406, + "step": 49116 + }, + { + "epoch": 0.4263591461879671, + "grad_norm": 0.2138671875, + "learning_rate": 0.0010816675946146686, + "loss": 0.1021, + "step": 49117 + }, + { + "epoch": 0.4263678266681713, + "grad_norm": 0.314453125, + "learning_rate": 0.0010816378383429908, + "loss": 0.0742, + "step": 49118 + }, + { + "epoch": 0.42637650714837544, + "grad_norm": 0.2158203125, + "learning_rate": 0.0010816080820913929, + "loss": 0.0933, + "step": 49119 + }, + { + "epoch": 0.42638518762857963, + "grad_norm": 0.1171875, + "learning_rate": 0.0010815783258599089, + "loss": 0.0986, + "step": 49120 + }, + { + "epoch": 0.42639386810878377, + "grad_norm": 0.205078125, + "learning_rate": 0.0010815485696485704, + "loss": 0.1113, + "step": 49121 + }, + { + "epoch": 0.42640254858898796, + "grad_norm": 0.154296875, + "learning_rate": 0.0010815188134574098, + "loss": 0.1074, + "step": 49122 + }, + { + "epoch": 0.4264112290691921, + "grad_norm": 0.22265625, + "learning_rate": 0.0010814890572864602, + "loss": 0.0908, + "step": 49123 + }, + { + "epoch": 0.4264199095493963, + "grad_norm": 0.65234375, + "learning_rate": 0.0010814593011357538, + "loss": 0.1104, + "step": 49124 + }, + { + "epoch": 0.42642859002960043, + "grad_norm": 0.2578125, + "learning_rate": 0.0010814295450053235, + "loss": 0.0742, + "step": 49125 + }, + { + "epoch": 0.4264372705098046, + "grad_norm": 0.1318359375, + "learning_rate": 0.0010813997888952014, + "loss": 0.0854, + "step": 49126 + }, + { + "epoch": 0.42644595099000876, + "grad_norm": 0.185546875, + "learning_rate": 0.0010813700328054202, + "loss": 0.0825, + "step": 49127 + }, + { + "epoch": 0.42645463147021295, + "grad_norm": 0.62109375, + "learning_rate": 0.001081340276736012, + "loss": 0.0928, + "step": 49128 + }, + { + "epoch": 0.4264633119504171, + "grad_norm": 0.546875, + "learning_rate": 0.0010813105206870105, + "loss": 0.0781, + "step": 49129 + }, + { + "epoch": 0.4264719924306213, + "grad_norm": 0.306640625, + "learning_rate": 0.0010812807646584472, + "loss": 0.084, + "step": 49130 + }, + { + "epoch": 0.4264806729108254, + "grad_norm": 0.419921875, + "learning_rate": 0.0010812510086503548, + "loss": 0.0654, + "step": 49131 + }, + { + "epoch": 0.4264893533910296, + "grad_norm": 0.55859375, + "learning_rate": 0.0010812212526627662, + "loss": 0.1006, + "step": 49132 + }, + { + "epoch": 0.42649803387123375, + "grad_norm": 0.482421875, + "learning_rate": 0.0010811914966957139, + "loss": 0.1123, + "step": 49133 + }, + { + "epoch": 0.42650671435143794, + "grad_norm": 0.271484375, + "learning_rate": 0.0010811617407492304, + "loss": 0.1338, + "step": 49134 + }, + { + "epoch": 0.4265153948316421, + "grad_norm": 0.46484375, + "learning_rate": 0.001081131984823348, + "loss": 0.1436, + "step": 49135 + }, + { + "epoch": 0.4265240753118463, + "grad_norm": 0.1298828125, + "learning_rate": 0.0010811022289180993, + "loss": 0.0938, + "step": 49136 + }, + { + "epoch": 0.4265327557920504, + "grad_norm": 0.373046875, + "learning_rate": 0.0010810724730335166, + "loss": 0.1162, + "step": 49137 + }, + { + "epoch": 0.4265414362722546, + "grad_norm": 0.3515625, + "learning_rate": 0.001081042717169633, + "loss": 0.1016, + "step": 49138 + }, + { + "epoch": 0.42655011675245874, + "grad_norm": 0.6171875, + "learning_rate": 0.001081012961326481, + "loss": 0.1553, + "step": 49139 + }, + { + "epoch": 0.42655879723266293, + "grad_norm": 0.216796875, + "learning_rate": 0.001080983205504093, + "loss": 0.0864, + "step": 49140 + }, + { + "epoch": 0.42656747771286707, + "grad_norm": 0.1376953125, + "learning_rate": 0.0010809534497025013, + "loss": 0.0898, + "step": 49141 + }, + { + "epoch": 0.42657615819307126, + "grad_norm": 0.154296875, + "learning_rate": 0.0010809236939217386, + "loss": 0.0713, + "step": 49142 + }, + { + "epoch": 0.4265848386732754, + "grad_norm": 0.2392578125, + "learning_rate": 0.0010808939381618374, + "loss": 0.1758, + "step": 49143 + }, + { + "epoch": 0.4265935191534796, + "grad_norm": 0.08642578125, + "learning_rate": 0.0010808641824228305, + "loss": 0.125, + "step": 49144 + }, + { + "epoch": 0.42660219963368373, + "grad_norm": 0.85546875, + "learning_rate": 0.0010808344267047498, + "loss": 0.207, + "step": 49145 + }, + { + "epoch": 0.4266108801138879, + "grad_norm": 0.10498046875, + "learning_rate": 0.0010808046710076287, + "loss": 0.063, + "step": 49146 + }, + { + "epoch": 0.42661956059409206, + "grad_norm": 0.12060546875, + "learning_rate": 0.0010807749153314992, + "loss": 0.1143, + "step": 49147 + }, + { + "epoch": 0.42662824107429625, + "grad_norm": 0.2314453125, + "learning_rate": 0.0010807451596763942, + "loss": 0.1094, + "step": 49148 + }, + { + "epoch": 0.4266369215545004, + "grad_norm": 0.318359375, + "learning_rate": 0.001080715404042346, + "loss": 0.1152, + "step": 49149 + }, + { + "epoch": 0.4266456020347046, + "grad_norm": 0.2333984375, + "learning_rate": 0.0010806856484293869, + "loss": 0.085, + "step": 49150 + }, + { + "epoch": 0.4266542825149087, + "grad_norm": 0.5625, + "learning_rate": 0.0010806558928375496, + "loss": 0.1357, + "step": 49151 + }, + { + "epoch": 0.4266629629951129, + "grad_norm": 0.154296875, + "learning_rate": 0.0010806261372668668, + "loss": 0.126, + "step": 49152 + }, + { + "epoch": 0.42667164347531705, + "grad_norm": 0.2275390625, + "learning_rate": 0.0010805963817173711, + "loss": 0.085, + "step": 49153 + }, + { + "epoch": 0.42668032395552125, + "grad_norm": 0.2333984375, + "learning_rate": 0.0010805666261890945, + "loss": 0.1084, + "step": 49154 + }, + { + "epoch": 0.4266890044357254, + "grad_norm": 0.1787109375, + "learning_rate": 0.0010805368706820705, + "loss": 0.0977, + "step": 49155 + }, + { + "epoch": 0.4266976849159296, + "grad_norm": 0.1455078125, + "learning_rate": 0.0010805071151963309, + "loss": 0.1436, + "step": 49156 + }, + { + "epoch": 0.4267063653961337, + "grad_norm": 0.4765625, + "learning_rate": 0.0010804773597319084, + "loss": 0.1406, + "step": 49157 + }, + { + "epoch": 0.4267150458763379, + "grad_norm": 0.08154296875, + "learning_rate": 0.0010804476042888356, + "loss": 0.0962, + "step": 49158 + }, + { + "epoch": 0.42672372635654204, + "grad_norm": 0.1484375, + "learning_rate": 0.0010804178488671447, + "loss": 0.1069, + "step": 49159 + }, + { + "epoch": 0.42673240683674624, + "grad_norm": 0.142578125, + "learning_rate": 0.001080388093466869, + "loss": 0.0962, + "step": 49160 + }, + { + "epoch": 0.4267410873169504, + "grad_norm": 0.5625, + "learning_rate": 0.0010803583380880404, + "loss": 0.1167, + "step": 49161 + }, + { + "epoch": 0.4267497677971545, + "grad_norm": 0.1025390625, + "learning_rate": 0.0010803285827306916, + "loss": 0.0889, + "step": 49162 + }, + { + "epoch": 0.4267584482773587, + "grad_norm": 0.2099609375, + "learning_rate": 0.001080298827394855, + "loss": 0.0581, + "step": 49163 + }, + { + "epoch": 0.42676712875756284, + "grad_norm": 0.359375, + "learning_rate": 0.0010802690720805638, + "loss": 0.1309, + "step": 49164 + }, + { + "epoch": 0.42677580923776703, + "grad_norm": 0.4609375, + "learning_rate": 0.0010802393167878494, + "loss": 0.0752, + "step": 49165 + }, + { + "epoch": 0.42678448971797117, + "grad_norm": 0.427734375, + "learning_rate": 0.0010802095615167456, + "loss": 0.0889, + "step": 49166 + }, + { + "epoch": 0.42679317019817536, + "grad_norm": 0.234375, + "learning_rate": 0.0010801798062672842, + "loss": 0.0752, + "step": 49167 + }, + { + "epoch": 0.4268018506783795, + "grad_norm": 0.578125, + "learning_rate": 0.0010801500510394975, + "loss": 0.1455, + "step": 49168 + }, + { + "epoch": 0.4268105311585837, + "grad_norm": 0.1884765625, + "learning_rate": 0.0010801202958334187, + "loss": 0.0967, + "step": 49169 + }, + { + "epoch": 0.42681921163878783, + "grad_norm": 0.52734375, + "learning_rate": 0.0010800905406490799, + "loss": 0.1133, + "step": 49170 + }, + { + "epoch": 0.426827892118992, + "grad_norm": 0.1884765625, + "learning_rate": 0.0010800607854865136, + "loss": 0.0957, + "step": 49171 + }, + { + "epoch": 0.42683657259919616, + "grad_norm": 0.3046875, + "learning_rate": 0.0010800310303457528, + "loss": 0.0874, + "step": 49172 + }, + { + "epoch": 0.42684525307940036, + "grad_norm": 0.298828125, + "learning_rate": 0.0010800012752268297, + "loss": 0.0811, + "step": 49173 + }, + { + "epoch": 0.4268539335596045, + "grad_norm": 0.7109375, + "learning_rate": 0.001079971520129777, + "loss": 0.082, + "step": 49174 + }, + { + "epoch": 0.4268626140398087, + "grad_norm": 0.19140625, + "learning_rate": 0.0010799417650546273, + "loss": 0.1016, + "step": 49175 + }, + { + "epoch": 0.4268712945200128, + "grad_norm": 0.0732421875, + "learning_rate": 0.0010799120100014128, + "loss": 0.0654, + "step": 49176 + }, + { + "epoch": 0.426879975000217, + "grad_norm": 0.64453125, + "learning_rate": 0.0010798822549701662, + "loss": 0.125, + "step": 49177 + }, + { + "epoch": 0.42688865548042115, + "grad_norm": 0.53125, + "learning_rate": 0.00107985249996092, + "loss": 0.0859, + "step": 49178 + }, + { + "epoch": 0.42689733596062535, + "grad_norm": 0.15625, + "learning_rate": 0.001079822744973707, + "loss": 0.082, + "step": 49179 + }, + { + "epoch": 0.4269060164408295, + "grad_norm": 0.103515625, + "learning_rate": 0.001079792990008559, + "loss": 0.1152, + "step": 49180 + }, + { + "epoch": 0.4269146969210337, + "grad_norm": 0.359375, + "learning_rate": 0.0010797632350655095, + "loss": 0.1299, + "step": 49181 + }, + { + "epoch": 0.4269233774012378, + "grad_norm": 0.44140625, + "learning_rate": 0.0010797334801445908, + "loss": 0.1504, + "step": 49182 + }, + { + "epoch": 0.426932057881442, + "grad_norm": 0.248046875, + "learning_rate": 0.001079703725245835, + "loss": 0.1328, + "step": 49183 + }, + { + "epoch": 0.42694073836164614, + "grad_norm": 0.21484375, + "learning_rate": 0.0010796739703692751, + "loss": 0.1113, + "step": 49184 + }, + { + "epoch": 0.42694941884185034, + "grad_norm": 0.36328125, + "learning_rate": 0.0010796442155149436, + "loss": 0.0991, + "step": 49185 + }, + { + "epoch": 0.4269580993220545, + "grad_norm": 0.1142578125, + "learning_rate": 0.0010796144606828727, + "loss": 0.0806, + "step": 49186 + }, + { + "epoch": 0.42696677980225867, + "grad_norm": 0.51171875, + "learning_rate": 0.0010795847058730948, + "loss": 0.0996, + "step": 49187 + }, + { + "epoch": 0.4269754602824628, + "grad_norm": 0.08984375, + "learning_rate": 0.001079554951085643, + "loss": 0.0957, + "step": 49188 + }, + { + "epoch": 0.426984140762667, + "grad_norm": 0.6171875, + "learning_rate": 0.0010795251963205492, + "loss": 0.1299, + "step": 49189 + }, + { + "epoch": 0.42699282124287113, + "grad_norm": 0.07421875, + "learning_rate": 0.0010794954415778465, + "loss": 0.1021, + "step": 49190 + }, + { + "epoch": 0.4270015017230753, + "grad_norm": 0.578125, + "learning_rate": 0.001079465686857568, + "loss": 0.0801, + "step": 49191 + }, + { + "epoch": 0.42701018220327946, + "grad_norm": 0.2890625, + "learning_rate": 0.001079435932159745, + "loss": 0.084, + "step": 49192 + }, + { + "epoch": 0.42701886268348366, + "grad_norm": 0.1640625, + "learning_rate": 0.0010794061774844104, + "loss": 0.0918, + "step": 49193 + }, + { + "epoch": 0.4270275431636878, + "grad_norm": 0.0751953125, + "learning_rate": 0.0010793764228315972, + "loss": 0.0742, + "step": 49194 + }, + { + "epoch": 0.427036223643892, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010793466682013375, + "loss": 0.0771, + "step": 49195 + }, + { + "epoch": 0.4270449041240961, + "grad_norm": 0.08740234375, + "learning_rate": 0.0010793169135936637, + "loss": 0.0854, + "step": 49196 + }, + { + "epoch": 0.4270535846043003, + "grad_norm": 0.1669921875, + "learning_rate": 0.0010792871590086088, + "loss": 0.106, + "step": 49197 + }, + { + "epoch": 0.42706226508450446, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010792574044462056, + "loss": 0.1021, + "step": 49198 + }, + { + "epoch": 0.42707094556470865, + "grad_norm": 0.5234375, + "learning_rate": 0.0010792276499064855, + "loss": 0.0684, + "step": 49199 + }, + { + "epoch": 0.4270796260449128, + "grad_norm": 0.337890625, + "learning_rate": 0.001079197895389482, + "loss": 0.2266, + "step": 49200 + }, + { + "epoch": 0.427088306525117, + "grad_norm": 0.19921875, + "learning_rate": 0.0010791681408952277, + "loss": 0.0757, + "step": 49201 + }, + { + "epoch": 0.4270969870053211, + "grad_norm": 0.09423828125, + "learning_rate": 0.0010791383864237544, + "loss": 0.0791, + "step": 49202 + }, + { + "epoch": 0.4271056674855253, + "grad_norm": 0.11279296875, + "learning_rate": 0.0010791086319750954, + "loss": 0.126, + "step": 49203 + }, + { + "epoch": 0.42711434796572945, + "grad_norm": 0.1591796875, + "learning_rate": 0.0010790788775492828, + "loss": 0.0977, + "step": 49204 + }, + { + "epoch": 0.42712302844593364, + "grad_norm": 0.14453125, + "learning_rate": 0.001079049123146349, + "loss": 0.0952, + "step": 49205 + }, + { + "epoch": 0.4271317089261378, + "grad_norm": 0.126953125, + "learning_rate": 0.0010790193687663266, + "loss": 0.1123, + "step": 49206 + }, + { + "epoch": 0.42714038940634197, + "grad_norm": 0.10107421875, + "learning_rate": 0.0010789896144092486, + "loss": 0.1016, + "step": 49207 + }, + { + "epoch": 0.4271490698865461, + "grad_norm": 0.1318359375, + "learning_rate": 0.001078959860075147, + "loss": 0.1289, + "step": 49208 + }, + { + "epoch": 0.4271577503667503, + "grad_norm": 0.62109375, + "learning_rate": 0.0010789301057640547, + "loss": 0.1021, + "step": 49209 + }, + { + "epoch": 0.42716643084695444, + "grad_norm": 0.2041015625, + "learning_rate": 0.0010789003514760046, + "loss": 0.1104, + "step": 49210 + }, + { + "epoch": 0.42717511132715863, + "grad_norm": 0.1640625, + "learning_rate": 0.0010788705972110282, + "loss": 0.0977, + "step": 49211 + }, + { + "epoch": 0.42718379180736277, + "grad_norm": 0.78515625, + "learning_rate": 0.0010788408429691588, + "loss": 0.1494, + "step": 49212 + }, + { + "epoch": 0.42719247228756696, + "grad_norm": 0.1904296875, + "learning_rate": 0.0010788110887504287, + "loss": 0.0879, + "step": 49213 + }, + { + "epoch": 0.4272011527677711, + "grad_norm": 0.07275390625, + "learning_rate": 0.0010787813345548708, + "loss": 0.125, + "step": 49214 + }, + { + "epoch": 0.4272098332479753, + "grad_norm": 0.41796875, + "learning_rate": 0.0010787515803825165, + "loss": 0.0801, + "step": 49215 + }, + { + "epoch": 0.4272185137281794, + "grad_norm": 0.322265625, + "learning_rate": 0.0010787218262333996, + "loss": 0.0801, + "step": 49216 + }, + { + "epoch": 0.4272271942083836, + "grad_norm": 0.390625, + "learning_rate": 0.0010786920721075522, + "loss": 0.1074, + "step": 49217 + }, + { + "epoch": 0.42723587468858776, + "grad_norm": 0.08447265625, + "learning_rate": 0.0010786623180050069, + "loss": 0.0752, + "step": 49218 + }, + { + "epoch": 0.42724455516879195, + "grad_norm": 0.27734375, + "learning_rate": 0.0010786325639257962, + "loss": 0.1162, + "step": 49219 + }, + { + "epoch": 0.4272532356489961, + "grad_norm": 0.169921875, + "learning_rate": 0.0010786028098699525, + "loss": 0.1143, + "step": 49220 + }, + { + "epoch": 0.4272619161292003, + "grad_norm": 0.3671875, + "learning_rate": 0.0010785730558375087, + "loss": 0.1025, + "step": 49221 + }, + { + "epoch": 0.4272705966094044, + "grad_norm": 0.6015625, + "learning_rate": 0.0010785433018284966, + "loss": 0.1172, + "step": 49222 + }, + { + "epoch": 0.4272792770896086, + "grad_norm": 0.42578125, + "learning_rate": 0.0010785135478429493, + "loss": 0.124, + "step": 49223 + }, + { + "epoch": 0.42728795756981275, + "grad_norm": 0.275390625, + "learning_rate": 0.0010784837938808995, + "loss": 0.1138, + "step": 49224 + }, + { + "epoch": 0.42729663805001694, + "grad_norm": 0.6875, + "learning_rate": 0.0010784540399423794, + "loss": 0.0879, + "step": 49225 + }, + { + "epoch": 0.4273053185302211, + "grad_norm": 0.08544921875, + "learning_rate": 0.0010784242860274217, + "loss": 0.0825, + "step": 49226 + }, + { + "epoch": 0.42731399901042527, + "grad_norm": 0.0947265625, + "learning_rate": 0.0010783945321360588, + "loss": 0.1162, + "step": 49227 + }, + { + "epoch": 0.4273226794906294, + "grad_norm": 0.0947265625, + "learning_rate": 0.0010783647782683235, + "loss": 0.0986, + "step": 49228 + }, + { + "epoch": 0.4273313599708336, + "grad_norm": 0.3984375, + "learning_rate": 0.001078335024424248, + "loss": 0.0601, + "step": 49229 + }, + { + "epoch": 0.42734004045103774, + "grad_norm": 0.26953125, + "learning_rate": 0.0010783052706038652, + "loss": 0.0962, + "step": 49230 + }, + { + "epoch": 0.42734872093124193, + "grad_norm": 0.31640625, + "learning_rate": 0.0010782755168072071, + "loss": 0.1113, + "step": 49231 + }, + { + "epoch": 0.42735740141144607, + "grad_norm": 0.1826171875, + "learning_rate": 0.0010782457630343066, + "loss": 0.0742, + "step": 49232 + }, + { + "epoch": 0.42736608189165026, + "grad_norm": 0.4921875, + "learning_rate": 0.0010782160092851962, + "loss": 0.085, + "step": 49233 + }, + { + "epoch": 0.4273747623718544, + "grad_norm": 0.396484375, + "learning_rate": 0.0010781862555599086, + "loss": 0.0977, + "step": 49234 + }, + { + "epoch": 0.4273834428520586, + "grad_norm": 0.154296875, + "learning_rate": 0.0010781565018584765, + "loss": 0.1055, + "step": 49235 + }, + { + "epoch": 0.42739212333226273, + "grad_norm": 0.419921875, + "learning_rate": 0.0010781267481809315, + "loss": 0.0918, + "step": 49236 + }, + { + "epoch": 0.4274008038124669, + "grad_norm": 0.251953125, + "learning_rate": 0.001078096994527307, + "loss": 0.0791, + "step": 49237 + }, + { + "epoch": 0.42740948429267106, + "grad_norm": 0.203125, + "learning_rate": 0.0010780672408976354, + "loss": 0.0933, + "step": 49238 + }, + { + "epoch": 0.42741816477287525, + "grad_norm": 0.4375, + "learning_rate": 0.0010780374872919492, + "loss": 0.123, + "step": 49239 + }, + { + "epoch": 0.4274268452530794, + "grad_norm": 0.1728515625, + "learning_rate": 0.0010780077337102804, + "loss": 0.1099, + "step": 49240 + }, + { + "epoch": 0.4274355257332836, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010779779801526625, + "loss": 0.1201, + "step": 49241 + }, + { + "epoch": 0.4274442062134877, + "grad_norm": 0.359375, + "learning_rate": 0.0010779482266191273, + "loss": 0.0786, + "step": 49242 + }, + { + "epoch": 0.4274528866936919, + "grad_norm": 0.103515625, + "learning_rate": 0.0010779184731097079, + "loss": 0.0669, + "step": 49243 + }, + { + "epoch": 0.42746156717389605, + "grad_norm": 0.1298828125, + "learning_rate": 0.0010778887196244361, + "loss": 0.1084, + "step": 49244 + }, + { + "epoch": 0.42747024765410024, + "grad_norm": 0.1123046875, + "learning_rate": 0.0010778589661633452, + "loss": 0.1055, + "step": 49245 + }, + { + "epoch": 0.4274789281343044, + "grad_norm": 0.375, + "learning_rate": 0.0010778292127264673, + "loss": 0.1079, + "step": 49246 + }, + { + "epoch": 0.4274876086145086, + "grad_norm": 0.298828125, + "learning_rate": 0.0010777994593138352, + "loss": 0.1157, + "step": 49247 + }, + { + "epoch": 0.4274962890947127, + "grad_norm": 0.81640625, + "learning_rate": 0.001077769705925481, + "loss": 0.1069, + "step": 49248 + }, + { + "epoch": 0.4275049695749169, + "grad_norm": 0.380859375, + "learning_rate": 0.0010777399525614374, + "loss": 0.0835, + "step": 49249 + }, + { + "epoch": 0.42751365005512104, + "grad_norm": 0.109375, + "learning_rate": 0.0010777101992217374, + "loss": 0.0977, + "step": 49250 + }, + { + "epoch": 0.42752233053532523, + "grad_norm": 0.16015625, + "learning_rate": 0.0010776804459064128, + "loss": 0.1138, + "step": 49251 + }, + { + "epoch": 0.42753101101552937, + "grad_norm": 0.46875, + "learning_rate": 0.001077650692615497, + "loss": 0.1045, + "step": 49252 + }, + { + "epoch": 0.42753969149573356, + "grad_norm": 0.3203125, + "learning_rate": 0.001077620939349022, + "loss": 0.1094, + "step": 49253 + }, + { + "epoch": 0.4275483719759377, + "grad_norm": 0.337890625, + "learning_rate": 0.0010775911861070201, + "loss": 0.0967, + "step": 49254 + }, + { + "epoch": 0.4275570524561419, + "grad_norm": 0.2353515625, + "learning_rate": 0.0010775614328895245, + "loss": 0.1162, + "step": 49255 + }, + { + "epoch": 0.42756573293634603, + "grad_norm": 0.146484375, + "learning_rate": 0.001077531679696567, + "loss": 0.0898, + "step": 49256 + }, + { + "epoch": 0.4275744134165502, + "grad_norm": 0.1279296875, + "learning_rate": 0.0010775019265281808, + "loss": 0.0947, + "step": 49257 + }, + { + "epoch": 0.42758309389675436, + "grad_norm": 0.07470703125, + "learning_rate": 0.0010774721733843977, + "loss": 0.0972, + "step": 49258 + }, + { + "epoch": 0.42759177437695856, + "grad_norm": 0.283203125, + "learning_rate": 0.001077442420265251, + "loss": 0.0684, + "step": 49259 + }, + { + "epoch": 0.4276004548571627, + "grad_norm": 0.44921875, + "learning_rate": 0.001077412667170773, + "loss": 0.124, + "step": 49260 + }, + { + "epoch": 0.4276091353373669, + "grad_norm": 0.244140625, + "learning_rate": 0.0010773829141009963, + "loss": 0.1064, + "step": 49261 + }, + { + "epoch": 0.427617815817571, + "grad_norm": 0.4375, + "learning_rate": 0.0010773531610559534, + "loss": 0.1094, + "step": 49262 + }, + { + "epoch": 0.4276264962977752, + "grad_norm": 2.125, + "learning_rate": 0.0010773234080356765, + "loss": 0.4629, + "step": 49263 + }, + { + "epoch": 0.42763517677797935, + "grad_norm": 0.05322265625, + "learning_rate": 0.001077293655040198, + "loss": 0.0571, + "step": 49264 + }, + { + "epoch": 0.42764385725818355, + "grad_norm": 0.28125, + "learning_rate": 0.0010772639020695514, + "loss": 0.0713, + "step": 49265 + }, + { + "epoch": 0.4276525377383877, + "grad_norm": 0.169921875, + "learning_rate": 0.0010772341491237683, + "loss": 0.106, + "step": 49266 + }, + { + "epoch": 0.4276612182185919, + "grad_norm": 0.1298828125, + "learning_rate": 0.0010772043962028817, + "loss": 0.0986, + "step": 49267 + }, + { + "epoch": 0.427669898698796, + "grad_norm": 0.173828125, + "learning_rate": 0.001077174643306924, + "loss": 0.0762, + "step": 49268 + }, + { + "epoch": 0.4276785791790002, + "grad_norm": 0.1220703125, + "learning_rate": 0.001077144890435928, + "loss": 0.1143, + "step": 49269 + }, + { + "epoch": 0.42768725965920434, + "grad_norm": 0.91796875, + "learning_rate": 0.001077115137589926, + "loss": 0.1855, + "step": 49270 + }, + { + "epoch": 0.42769594013940854, + "grad_norm": 1.1796875, + "learning_rate": 0.0010770853847689504, + "loss": 0.2012, + "step": 49271 + }, + { + "epoch": 0.4277046206196127, + "grad_norm": 0.11865234375, + "learning_rate": 0.0010770556319730337, + "loss": 0.1328, + "step": 49272 + }, + { + "epoch": 0.42771330109981687, + "grad_norm": 0.169921875, + "learning_rate": 0.0010770258792022088, + "loss": 0.124, + "step": 49273 + }, + { + "epoch": 0.427721981580021, + "grad_norm": 0.1845703125, + "learning_rate": 0.001076996126456508, + "loss": 0.1147, + "step": 49274 + }, + { + "epoch": 0.4277306620602252, + "grad_norm": 0.44921875, + "learning_rate": 0.0010769663737359638, + "loss": 0.0874, + "step": 49275 + }, + { + "epoch": 0.42773934254042933, + "grad_norm": 0.65625, + "learning_rate": 0.0010769366210406088, + "loss": 0.1484, + "step": 49276 + }, + { + "epoch": 0.4277480230206335, + "grad_norm": 0.359375, + "learning_rate": 0.0010769068683704759, + "loss": 0.1143, + "step": 49277 + }, + { + "epoch": 0.42775670350083767, + "grad_norm": 0.326171875, + "learning_rate": 0.001076877115725597, + "loss": 0.1152, + "step": 49278 + }, + { + "epoch": 0.42776538398104186, + "grad_norm": 1.2421875, + "learning_rate": 0.0010768473631060052, + "loss": 0.123, + "step": 49279 + }, + { + "epoch": 0.427774064461246, + "grad_norm": 0.32421875, + "learning_rate": 0.0010768176105117326, + "loss": 0.1084, + "step": 49280 + }, + { + "epoch": 0.4277827449414502, + "grad_norm": 0.13671875, + "learning_rate": 0.0010767878579428121, + "loss": 0.0947, + "step": 49281 + }, + { + "epoch": 0.4277914254216543, + "grad_norm": 0.216796875, + "learning_rate": 0.0010767581053992758, + "loss": 0.1216, + "step": 49282 + }, + { + "epoch": 0.4278001059018585, + "grad_norm": 0.138671875, + "learning_rate": 0.0010767283528811567, + "loss": 0.1099, + "step": 49283 + }, + { + "epoch": 0.42780878638206266, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010766986003884868, + "loss": 0.1133, + "step": 49284 + }, + { + "epoch": 0.4278174668622668, + "grad_norm": 0.294921875, + "learning_rate": 0.001076668847921299, + "loss": 0.0918, + "step": 49285 + }, + { + "epoch": 0.427826147342471, + "grad_norm": 0.1650390625, + "learning_rate": 0.0010766390954796258, + "loss": 0.0967, + "step": 49286 + }, + { + "epoch": 0.4278348278226751, + "grad_norm": 0.2158203125, + "learning_rate": 0.0010766093430635, + "loss": 0.0796, + "step": 49287 + }, + { + "epoch": 0.4278435083028793, + "grad_norm": 0.150390625, + "learning_rate": 0.001076579590672954, + "loss": 0.0933, + "step": 49288 + }, + { + "epoch": 0.42785218878308345, + "grad_norm": 0.2216796875, + "learning_rate": 0.0010765498383080198, + "loss": 0.0845, + "step": 49289 + }, + { + "epoch": 0.42786086926328765, + "grad_norm": 0.1416015625, + "learning_rate": 0.0010765200859687307, + "loss": 0.105, + "step": 49290 + }, + { + "epoch": 0.4278695497434918, + "grad_norm": 0.26953125, + "learning_rate": 0.0010764903336551188, + "loss": 0.0693, + "step": 49291 + }, + { + "epoch": 0.427878230223696, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010764605813672163, + "loss": 0.0952, + "step": 49292 + }, + { + "epoch": 0.4278869107039001, + "grad_norm": 0.146484375, + "learning_rate": 0.0010764308291050567, + "loss": 0.1006, + "step": 49293 + }, + { + "epoch": 0.4278955911841043, + "grad_norm": 0.302734375, + "learning_rate": 0.0010764010768686714, + "loss": 0.1162, + "step": 49294 + }, + { + "epoch": 0.42790427166430844, + "grad_norm": 0.625, + "learning_rate": 0.001076371324658094, + "loss": 0.0928, + "step": 49295 + }, + { + "epoch": 0.42791295214451264, + "grad_norm": 0.146484375, + "learning_rate": 0.0010763415724733565, + "loss": 0.1035, + "step": 49296 + }, + { + "epoch": 0.4279216326247168, + "grad_norm": 0.11279296875, + "learning_rate": 0.0010763118203144914, + "loss": 0.1143, + "step": 49297 + }, + { + "epoch": 0.42793031310492097, + "grad_norm": 0.3203125, + "learning_rate": 0.0010762820681815316, + "loss": 0.0898, + "step": 49298 + }, + { + "epoch": 0.4279389935851251, + "grad_norm": 0.625, + "learning_rate": 0.0010762523160745092, + "loss": 0.0737, + "step": 49299 + }, + { + "epoch": 0.4279476740653293, + "grad_norm": 0.154296875, + "learning_rate": 0.0010762225639934571, + "loss": 0.1001, + "step": 49300 + }, + { + "epoch": 0.42795635454553344, + "grad_norm": 0.08642578125, + "learning_rate": 0.001076192811938407, + "loss": 0.0933, + "step": 49301 + }, + { + "epoch": 0.42796503502573763, + "grad_norm": 0.10009765625, + "learning_rate": 0.0010761630599093925, + "loss": 0.0952, + "step": 49302 + }, + { + "epoch": 0.42797371550594177, + "grad_norm": 0.466796875, + "learning_rate": 0.0010761333079064457, + "loss": 0.0718, + "step": 49303 + }, + { + "epoch": 0.42798239598614596, + "grad_norm": 0.298828125, + "learning_rate": 0.0010761035559295991, + "loss": 0.0732, + "step": 49304 + }, + { + "epoch": 0.4279910764663501, + "grad_norm": 0.07421875, + "learning_rate": 0.0010760738039788855, + "loss": 0.0918, + "step": 49305 + }, + { + "epoch": 0.4279997569465543, + "grad_norm": 0.2275390625, + "learning_rate": 0.0010760440520543371, + "loss": 0.0928, + "step": 49306 + }, + { + "epoch": 0.4280084374267584, + "grad_norm": 0.294921875, + "learning_rate": 0.0010760143001559866, + "loss": 0.0884, + "step": 49307 + }, + { + "epoch": 0.4280171179069626, + "grad_norm": 0.2333984375, + "learning_rate": 0.0010759845482838665, + "loss": 0.0703, + "step": 49308 + }, + { + "epoch": 0.42802579838716676, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010759547964380094, + "loss": 0.1079, + "step": 49309 + }, + { + "epoch": 0.42803447886737095, + "grad_norm": 0.14453125, + "learning_rate": 0.0010759250446184473, + "loss": 0.0986, + "step": 49310 + }, + { + "epoch": 0.4280431593475751, + "grad_norm": 0.333984375, + "learning_rate": 0.0010758952928252136, + "loss": 0.1055, + "step": 49311 + }, + { + "epoch": 0.4280518398277793, + "grad_norm": 0.34375, + "learning_rate": 0.0010758655410583408, + "loss": 0.1143, + "step": 49312 + }, + { + "epoch": 0.4280605203079834, + "grad_norm": 0.0634765625, + "learning_rate": 0.0010758357893178602, + "loss": 0.0588, + "step": 49313 + }, + { + "epoch": 0.4280692007881876, + "grad_norm": 0.208984375, + "learning_rate": 0.0010758060376038058, + "loss": 0.0918, + "step": 49314 + }, + { + "epoch": 0.42807788126839175, + "grad_norm": 0.1328125, + "learning_rate": 0.0010757762859162096, + "loss": 0.1299, + "step": 49315 + }, + { + "epoch": 0.42808656174859594, + "grad_norm": 0.10546875, + "learning_rate": 0.0010757465342551042, + "loss": 0.1011, + "step": 49316 + }, + { + "epoch": 0.4280952422288001, + "grad_norm": 0.4609375, + "learning_rate": 0.0010757167826205218, + "loss": 0.0996, + "step": 49317 + }, + { + "epoch": 0.42810392270900427, + "grad_norm": 0.67578125, + "learning_rate": 0.001075687031012495, + "loss": 0.0894, + "step": 49318 + }, + { + "epoch": 0.4281126031892084, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010756572794310567, + "loss": 0.1133, + "step": 49319 + }, + { + "epoch": 0.4281212836694126, + "grad_norm": 0.107421875, + "learning_rate": 0.001075627527876239, + "loss": 0.0654, + "step": 49320 + }, + { + "epoch": 0.42812996414961674, + "grad_norm": 0.099609375, + "learning_rate": 0.0010755977763480752, + "loss": 0.0957, + "step": 49321 + }, + { + "epoch": 0.42813864462982093, + "grad_norm": 0.7734375, + "learning_rate": 0.0010755680248465966, + "loss": 0.127, + "step": 49322 + }, + { + "epoch": 0.42814732511002507, + "grad_norm": 0.271484375, + "learning_rate": 0.0010755382733718371, + "loss": 0.1094, + "step": 49323 + }, + { + "epoch": 0.42815600559022926, + "grad_norm": 0.224609375, + "learning_rate": 0.0010755085219238284, + "loss": 0.1069, + "step": 49324 + }, + { + "epoch": 0.4281646860704334, + "grad_norm": 0.09033203125, + "learning_rate": 0.001075478770502603, + "loss": 0.084, + "step": 49325 + }, + { + "epoch": 0.4281733665506376, + "grad_norm": 0.1962890625, + "learning_rate": 0.0010754490191081938, + "loss": 0.1123, + "step": 49326 + }, + { + "epoch": 0.42818204703084173, + "grad_norm": 0.08740234375, + "learning_rate": 0.0010754192677406328, + "loss": 0.0894, + "step": 49327 + }, + { + "epoch": 0.4281907275110459, + "grad_norm": 0.15625, + "learning_rate": 0.0010753895163999531, + "loss": 0.0967, + "step": 49328 + }, + { + "epoch": 0.42819940799125006, + "grad_norm": 0.4453125, + "learning_rate": 0.0010753597650861875, + "loss": 0.0957, + "step": 49329 + }, + { + "epoch": 0.42820808847145425, + "grad_norm": 0.259765625, + "learning_rate": 0.0010753300137993679, + "loss": 0.1006, + "step": 49330 + }, + { + "epoch": 0.4282167689516584, + "grad_norm": 0.158203125, + "learning_rate": 0.001075300262539527, + "loss": 0.1387, + "step": 49331 + }, + { + "epoch": 0.4282254494318626, + "grad_norm": 0.1171875, + "learning_rate": 0.0010752705113066972, + "loss": 0.1011, + "step": 49332 + }, + { + "epoch": 0.4282341299120667, + "grad_norm": 0.11083984375, + "learning_rate": 0.0010752407601009113, + "loss": 0.1045, + "step": 49333 + }, + { + "epoch": 0.4282428103922709, + "grad_norm": 0.365234375, + "learning_rate": 0.0010752110089222017, + "loss": 0.1104, + "step": 49334 + }, + { + "epoch": 0.42825149087247505, + "grad_norm": 0.36328125, + "learning_rate": 0.001075181257770601, + "loss": 0.1025, + "step": 49335 + }, + { + "epoch": 0.42826017135267924, + "grad_norm": 0.10498046875, + "learning_rate": 0.0010751515066461414, + "loss": 0.0786, + "step": 49336 + }, + { + "epoch": 0.4282688518328834, + "grad_norm": 0.12255859375, + "learning_rate": 0.0010751217555488562, + "loss": 0.1104, + "step": 49337 + }, + { + "epoch": 0.42827753231308757, + "grad_norm": 0.2158203125, + "learning_rate": 0.001075092004478777, + "loss": 0.1221, + "step": 49338 + }, + { + "epoch": 0.4282862127932917, + "grad_norm": 0.8203125, + "learning_rate": 0.0010750622534359373, + "loss": 0.123, + "step": 49339 + }, + { + "epoch": 0.4282948932734959, + "grad_norm": 0.0869140625, + "learning_rate": 0.001075032502420369, + "loss": 0.0771, + "step": 49340 + }, + { + "epoch": 0.42830357375370004, + "grad_norm": 0.58203125, + "learning_rate": 0.0010750027514321047, + "loss": 0.0942, + "step": 49341 + }, + { + "epoch": 0.42831225423390423, + "grad_norm": 0.2109375, + "learning_rate": 0.001074973000471177, + "loss": 0.0742, + "step": 49342 + }, + { + "epoch": 0.42832093471410837, + "grad_norm": 0.33203125, + "learning_rate": 0.0010749432495376187, + "loss": 0.082, + "step": 49343 + }, + { + "epoch": 0.42832961519431256, + "grad_norm": 0.322265625, + "learning_rate": 0.0010749134986314615, + "loss": 0.1543, + "step": 49344 + }, + { + "epoch": 0.4283382956745167, + "grad_norm": 0.134765625, + "learning_rate": 0.0010748837477527389, + "loss": 0.0737, + "step": 49345 + }, + { + "epoch": 0.4283469761547209, + "grad_norm": 0.1328125, + "learning_rate": 0.0010748539969014832, + "loss": 0.1123, + "step": 49346 + }, + { + "epoch": 0.42835565663492503, + "grad_norm": 0.2734375, + "learning_rate": 0.0010748242460777268, + "loss": 0.1191, + "step": 49347 + }, + { + "epoch": 0.4283643371151292, + "grad_norm": 0.0771484375, + "learning_rate": 0.001074794495281502, + "loss": 0.0869, + "step": 49348 + }, + { + "epoch": 0.42837301759533336, + "grad_norm": 0.1923828125, + "learning_rate": 0.0010747647445128418, + "loss": 0.125, + "step": 49349 + }, + { + "epoch": 0.42838169807553755, + "grad_norm": 0.1630859375, + "learning_rate": 0.0010747349937717781, + "loss": 0.1123, + "step": 49350 + }, + { + "epoch": 0.4283903785557417, + "grad_norm": 0.140625, + "learning_rate": 0.001074705243058344, + "loss": 0.1113, + "step": 49351 + }, + { + "epoch": 0.4283990590359459, + "grad_norm": 0.2138671875, + "learning_rate": 0.0010746754923725722, + "loss": 0.1768, + "step": 49352 + }, + { + "epoch": 0.42840773951615, + "grad_norm": 0.1884765625, + "learning_rate": 0.0010746457417144945, + "loss": 0.6328, + "step": 49353 + }, + { + "epoch": 0.4284164199963542, + "grad_norm": 0.3203125, + "learning_rate": 0.0010746159910841437, + "loss": 0.1006, + "step": 49354 + }, + { + "epoch": 0.42842510047655835, + "grad_norm": 0.1357421875, + "learning_rate": 0.001074586240481553, + "loss": 0.1416, + "step": 49355 + }, + { + "epoch": 0.42843378095676254, + "grad_norm": 0.57421875, + "learning_rate": 0.0010745564899067542, + "loss": 0.1094, + "step": 49356 + }, + { + "epoch": 0.4284424614369667, + "grad_norm": 0.181640625, + "learning_rate": 0.00107452673935978, + "loss": 0.1094, + "step": 49357 + }, + { + "epoch": 0.4284511419171709, + "grad_norm": 0.11669921875, + "learning_rate": 0.001074496988840663, + "loss": 0.0972, + "step": 49358 + }, + { + "epoch": 0.428459822397375, + "grad_norm": 0.1083984375, + "learning_rate": 0.0010744672383494358, + "loss": 0.1055, + "step": 49359 + }, + { + "epoch": 0.4284685028775792, + "grad_norm": 0.1748046875, + "learning_rate": 0.0010744374878861305, + "loss": 0.0996, + "step": 49360 + }, + { + "epoch": 0.42847718335778334, + "grad_norm": 0.318359375, + "learning_rate": 0.0010744077374507802, + "loss": 0.123, + "step": 49361 + }, + { + "epoch": 0.42848586383798754, + "grad_norm": 0.447265625, + "learning_rate": 0.001074377987043417, + "loss": 0.0928, + "step": 49362 + }, + { + "epoch": 0.4284945443181917, + "grad_norm": 0.130859375, + "learning_rate": 0.001074348236664074, + "loss": 0.0845, + "step": 49363 + }, + { + "epoch": 0.42850322479839587, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010743184863127833, + "loss": 0.0776, + "step": 49364 + }, + { + "epoch": 0.4285119052786, + "grad_norm": 0.29296875, + "learning_rate": 0.0010742887359895775, + "loss": 0.1406, + "step": 49365 + }, + { + "epoch": 0.4285205857588042, + "grad_norm": 0.1982421875, + "learning_rate": 0.0010742589856944892, + "loss": 0.0845, + "step": 49366 + }, + { + "epoch": 0.42852926623900833, + "grad_norm": 0.1083984375, + "learning_rate": 0.0010742292354275506, + "loss": 0.1133, + "step": 49367 + }, + { + "epoch": 0.4285379467192125, + "grad_norm": 0.142578125, + "learning_rate": 0.0010741994851887948, + "loss": 0.0889, + "step": 49368 + }, + { + "epoch": 0.42854662719941666, + "grad_norm": 0.470703125, + "learning_rate": 0.001074169734978254, + "loss": 0.1094, + "step": 49369 + }, + { + "epoch": 0.42855530767962086, + "grad_norm": 0.5, + "learning_rate": 0.0010741399847959605, + "loss": 0.1006, + "step": 49370 + }, + { + "epoch": 0.428563988159825, + "grad_norm": 0.4921875, + "learning_rate": 0.0010741102346419477, + "loss": 0.0884, + "step": 49371 + }, + { + "epoch": 0.4285726686400292, + "grad_norm": 0.09765625, + "learning_rate": 0.0010740804845162468, + "loss": 0.0854, + "step": 49372 + }, + { + "epoch": 0.4285813491202333, + "grad_norm": 0.625, + "learning_rate": 0.0010740507344188918, + "loss": 0.1099, + "step": 49373 + }, + { + "epoch": 0.4285900296004375, + "grad_norm": 0.2216796875, + "learning_rate": 0.0010740209843499143, + "loss": 0.0835, + "step": 49374 + }, + { + "epoch": 0.42859871008064165, + "grad_norm": 0.19921875, + "learning_rate": 0.001073991234309347, + "loss": 0.1094, + "step": 49375 + }, + { + "epoch": 0.42860739056084585, + "grad_norm": 0.59765625, + "learning_rate": 0.0010739614842972228, + "loss": 0.1113, + "step": 49376 + }, + { + "epoch": 0.42861607104105, + "grad_norm": 0.27734375, + "learning_rate": 0.0010739317343135735, + "loss": 0.1045, + "step": 49377 + }, + { + "epoch": 0.4286247515212542, + "grad_norm": 0.240234375, + "learning_rate": 0.0010739019843584322, + "loss": 0.0996, + "step": 49378 + }, + { + "epoch": 0.4286334320014583, + "grad_norm": 0.255859375, + "learning_rate": 0.0010738722344318312, + "loss": 0.105, + "step": 49379 + }, + { + "epoch": 0.4286421124816625, + "grad_norm": 0.1328125, + "learning_rate": 0.0010738424845338033, + "loss": 0.1016, + "step": 49380 + }, + { + "epoch": 0.42865079296186664, + "grad_norm": 0.140625, + "learning_rate": 0.0010738127346643807, + "loss": 0.0791, + "step": 49381 + }, + { + "epoch": 0.42865947344207084, + "grad_norm": 0.6640625, + "learning_rate": 0.0010737829848235963, + "loss": 0.1328, + "step": 49382 + }, + { + "epoch": 0.428668153922275, + "grad_norm": 0.2060546875, + "learning_rate": 0.0010737532350114823, + "loss": 0.0957, + "step": 49383 + }, + { + "epoch": 0.42867683440247917, + "grad_norm": 0.10595703125, + "learning_rate": 0.0010737234852280714, + "loss": 0.1064, + "step": 49384 + }, + { + "epoch": 0.4286855148826833, + "grad_norm": 0.1025390625, + "learning_rate": 0.0010736937354733963, + "loss": 0.0977, + "step": 49385 + }, + { + "epoch": 0.4286941953628875, + "grad_norm": 0.11669921875, + "learning_rate": 0.0010736639857474892, + "loss": 0.1562, + "step": 49386 + }, + { + "epoch": 0.42870287584309164, + "grad_norm": 0.11572265625, + "learning_rate": 0.0010736342360503823, + "loss": 0.1055, + "step": 49387 + }, + { + "epoch": 0.42871155632329583, + "grad_norm": 0.15625, + "learning_rate": 0.0010736044863821092, + "loss": 0.1011, + "step": 49388 + }, + { + "epoch": 0.42872023680349997, + "grad_norm": 0.154296875, + "learning_rate": 0.0010735747367427017, + "loss": 0.1113, + "step": 49389 + }, + { + "epoch": 0.42872891728370416, + "grad_norm": 0.83984375, + "learning_rate": 0.0010735449871321922, + "loss": 0.0801, + "step": 49390 + }, + { + "epoch": 0.4287375977639083, + "grad_norm": 0.298828125, + "learning_rate": 0.0010735152375506139, + "loss": 0.1045, + "step": 49391 + }, + { + "epoch": 0.4287462782441125, + "grad_norm": 0.263671875, + "learning_rate": 0.0010734854879979987, + "loss": 0.1079, + "step": 49392 + }, + { + "epoch": 0.4287549587243166, + "grad_norm": 0.11328125, + "learning_rate": 0.0010734557384743796, + "loss": 0.0645, + "step": 49393 + }, + { + "epoch": 0.4287636392045208, + "grad_norm": 0.6640625, + "learning_rate": 0.0010734259889797887, + "loss": 0.1357, + "step": 49394 + }, + { + "epoch": 0.42877231968472496, + "grad_norm": 0.1982421875, + "learning_rate": 0.0010733962395142588, + "loss": 0.1445, + "step": 49395 + }, + { + "epoch": 0.42878100016492915, + "grad_norm": 0.12109375, + "learning_rate": 0.0010733664900778222, + "loss": 0.063, + "step": 49396 + }, + { + "epoch": 0.4287896806451333, + "grad_norm": 0.40625, + "learning_rate": 0.001073336740670512, + "loss": 0.1138, + "step": 49397 + }, + { + "epoch": 0.4287983611253375, + "grad_norm": 0.62890625, + "learning_rate": 0.00107330699129236, + "loss": 0.1484, + "step": 49398 + }, + { + "epoch": 0.4288070416055416, + "grad_norm": 0.361328125, + "learning_rate": 0.0010732772419433992, + "loss": 0.1133, + "step": 49399 + }, + { + "epoch": 0.4288157220857458, + "grad_norm": 0.296875, + "learning_rate": 0.001073247492623662, + "loss": 0.0703, + "step": 49400 + }, + { + "epoch": 0.42882440256594995, + "grad_norm": 0.19140625, + "learning_rate": 0.0010732177433331808, + "loss": 0.1235, + "step": 49401 + }, + { + "epoch": 0.42883308304615414, + "grad_norm": 0.45703125, + "learning_rate": 0.0010731879940719885, + "loss": 0.125, + "step": 49402 + }, + { + "epoch": 0.4288417635263583, + "grad_norm": 0.52734375, + "learning_rate": 0.0010731582448401174, + "loss": 0.1084, + "step": 49403 + }, + { + "epoch": 0.42885044400656247, + "grad_norm": 0.89453125, + "learning_rate": 0.0010731284956376, + "loss": 0.1064, + "step": 49404 + }, + { + "epoch": 0.4288591244867666, + "grad_norm": 0.53125, + "learning_rate": 0.0010730987464644686, + "loss": 0.1172, + "step": 49405 + }, + { + "epoch": 0.4288678049669708, + "grad_norm": 0.12158203125, + "learning_rate": 0.0010730689973207563, + "loss": 0.1138, + "step": 49406 + }, + { + "epoch": 0.42887648544717494, + "grad_norm": 0.134765625, + "learning_rate": 0.0010730392482064953, + "loss": 0.0737, + "step": 49407 + }, + { + "epoch": 0.4288851659273791, + "grad_norm": 1.4765625, + "learning_rate": 0.0010730094991217181, + "loss": 0.373, + "step": 49408 + }, + { + "epoch": 0.42889384640758327, + "grad_norm": 0.138671875, + "learning_rate": 0.0010729797500664572, + "loss": 0.0781, + "step": 49409 + }, + { + "epoch": 0.4289025268877874, + "grad_norm": 0.181640625, + "learning_rate": 0.0010729500010407456, + "loss": 0.1001, + "step": 49410 + }, + { + "epoch": 0.4289112073679916, + "grad_norm": 0.54296875, + "learning_rate": 0.0010729202520446152, + "loss": 0.0835, + "step": 49411 + }, + { + "epoch": 0.42891988784819574, + "grad_norm": 0.318359375, + "learning_rate": 0.0010728905030780989, + "loss": 0.0854, + "step": 49412 + }, + { + "epoch": 0.42892856832839993, + "grad_norm": 0.5, + "learning_rate": 0.001072860754141229, + "loss": 0.0967, + "step": 49413 + }, + { + "epoch": 0.42893724880860407, + "grad_norm": 0.6796875, + "learning_rate": 0.0010728310052340382, + "loss": 0.0664, + "step": 49414 + }, + { + "epoch": 0.42894592928880826, + "grad_norm": 0.177734375, + "learning_rate": 0.001072801256356559, + "loss": 0.1201, + "step": 49415 + }, + { + "epoch": 0.4289546097690124, + "grad_norm": 0.27734375, + "learning_rate": 0.0010727715075088242, + "loss": 0.1167, + "step": 49416 + }, + { + "epoch": 0.4289632902492166, + "grad_norm": 0.1533203125, + "learning_rate": 0.0010727417586908659, + "loss": 0.1069, + "step": 49417 + }, + { + "epoch": 0.4289719707294207, + "grad_norm": 0.2236328125, + "learning_rate": 0.0010727120099027169, + "loss": 0.0918, + "step": 49418 + }, + { + "epoch": 0.4289806512096249, + "grad_norm": 0.244140625, + "learning_rate": 0.0010726822611444094, + "loss": 0.125, + "step": 49419 + }, + { + "epoch": 0.42898933168982906, + "grad_norm": 0.1982421875, + "learning_rate": 0.0010726525124159765, + "loss": 0.0757, + "step": 49420 + }, + { + "epoch": 0.42899801217003325, + "grad_norm": 0.1865234375, + "learning_rate": 0.0010726227637174502, + "loss": 0.0923, + "step": 49421 + }, + { + "epoch": 0.4290066926502374, + "grad_norm": 0.119140625, + "learning_rate": 0.0010725930150488627, + "loss": 0.1006, + "step": 49422 + }, + { + "epoch": 0.4290153731304416, + "grad_norm": 0.1201171875, + "learning_rate": 0.0010725632664102477, + "loss": 0.0791, + "step": 49423 + }, + { + "epoch": 0.4290240536106457, + "grad_norm": 0.20703125, + "learning_rate": 0.0010725335178016372, + "loss": 0.0781, + "step": 49424 + }, + { + "epoch": 0.4290327340908499, + "grad_norm": 0.6796875, + "learning_rate": 0.0010725037692230635, + "loss": 0.1768, + "step": 49425 + }, + { + "epoch": 0.42904141457105405, + "grad_norm": 0.3359375, + "learning_rate": 0.0010724740206745592, + "loss": 0.0889, + "step": 49426 + }, + { + "epoch": 0.42905009505125824, + "grad_norm": 0.1005859375, + "learning_rate": 0.0010724442721561566, + "loss": 0.1113, + "step": 49427 + }, + { + "epoch": 0.4290587755314624, + "grad_norm": 0.2353515625, + "learning_rate": 0.001072414523667889, + "loss": 0.0977, + "step": 49428 + }, + { + "epoch": 0.42906745601166657, + "grad_norm": 0.2373046875, + "learning_rate": 0.0010723847752097883, + "loss": 0.1118, + "step": 49429 + }, + { + "epoch": 0.4290761364918707, + "grad_norm": 0.28125, + "learning_rate": 0.0010723550267818874, + "loss": 0.1089, + "step": 49430 + }, + { + "epoch": 0.4290848169720749, + "grad_norm": 0.23828125, + "learning_rate": 0.0010723252783842179, + "loss": 0.1069, + "step": 49431 + }, + { + "epoch": 0.42909349745227904, + "grad_norm": 0.330078125, + "learning_rate": 0.0010722955300168139, + "loss": 0.0742, + "step": 49432 + }, + { + "epoch": 0.42910217793248323, + "grad_norm": 0.7578125, + "learning_rate": 0.0010722657816797066, + "loss": 0.1089, + "step": 49433 + }, + { + "epoch": 0.42911085841268737, + "grad_norm": 0.416015625, + "learning_rate": 0.0010722360333729293, + "loss": 0.1035, + "step": 49434 + }, + { + "epoch": 0.42911953889289156, + "grad_norm": 0.244140625, + "learning_rate": 0.0010722062850965141, + "loss": 0.1025, + "step": 49435 + }, + { + "epoch": 0.4291282193730957, + "grad_norm": 0.2041015625, + "learning_rate": 0.001072176536850494, + "loss": 0.1182, + "step": 49436 + }, + { + "epoch": 0.4291368998532999, + "grad_norm": 0.486328125, + "learning_rate": 0.0010721467886349008, + "loss": 0.0923, + "step": 49437 + }, + { + "epoch": 0.42914558033350403, + "grad_norm": 0.078125, + "learning_rate": 0.0010721170404497675, + "loss": 0.0874, + "step": 49438 + }, + { + "epoch": 0.4291542608137082, + "grad_norm": 0.337890625, + "learning_rate": 0.0010720872922951265, + "loss": 0.1338, + "step": 49439 + }, + { + "epoch": 0.42916294129391236, + "grad_norm": 0.1376953125, + "learning_rate": 0.0010720575441710105, + "loss": 0.1045, + "step": 49440 + }, + { + "epoch": 0.42917162177411655, + "grad_norm": 0.490234375, + "learning_rate": 0.0010720277960774524, + "loss": 0.1045, + "step": 49441 + }, + { + "epoch": 0.4291803022543207, + "grad_norm": 0.359375, + "learning_rate": 0.0010719980480144837, + "loss": 0.0898, + "step": 49442 + }, + { + "epoch": 0.4291889827345249, + "grad_norm": 0.59765625, + "learning_rate": 0.0010719682999821378, + "loss": 0.1533, + "step": 49443 + }, + { + "epoch": 0.429197663214729, + "grad_norm": 0.69140625, + "learning_rate": 0.0010719385519804472, + "loss": 0.0845, + "step": 49444 + }, + { + "epoch": 0.4292063436949332, + "grad_norm": 0.478515625, + "learning_rate": 0.0010719088040094438, + "loss": 0.1055, + "step": 49445 + }, + { + "epoch": 0.42921502417513735, + "grad_norm": 0.328125, + "learning_rate": 0.0010718790560691606, + "loss": 0.1016, + "step": 49446 + }, + { + "epoch": 0.42922370465534154, + "grad_norm": 0.09912109375, + "learning_rate": 0.00107184930815963, + "loss": 0.085, + "step": 49447 + }, + { + "epoch": 0.4292323851355457, + "grad_norm": 0.0830078125, + "learning_rate": 0.0010718195602808845, + "loss": 0.0918, + "step": 49448 + }, + { + "epoch": 0.4292410656157499, + "grad_norm": 0.291015625, + "learning_rate": 0.0010717898124329565, + "loss": 0.1436, + "step": 49449 + }, + { + "epoch": 0.429249746095954, + "grad_norm": 0.1513671875, + "learning_rate": 0.0010717600646158792, + "loss": 0.1016, + "step": 49450 + }, + { + "epoch": 0.4292584265761582, + "grad_norm": 0.1123046875, + "learning_rate": 0.0010717303168296846, + "loss": 0.1143, + "step": 49451 + }, + { + "epoch": 0.42926710705636234, + "grad_norm": 0.126953125, + "learning_rate": 0.0010717005690744053, + "loss": 0.1426, + "step": 49452 + }, + { + "epoch": 0.42927578753656653, + "grad_norm": 0.220703125, + "learning_rate": 0.0010716708213500737, + "loss": 0.105, + "step": 49453 + }, + { + "epoch": 0.42928446801677067, + "grad_norm": 0.37890625, + "learning_rate": 0.0010716410736567226, + "loss": 0.1016, + "step": 49454 + }, + { + "epoch": 0.42929314849697486, + "grad_norm": 0.34765625, + "learning_rate": 0.0010716113259943844, + "loss": 0.0781, + "step": 49455 + }, + { + "epoch": 0.429301828977179, + "grad_norm": 0.1337890625, + "learning_rate": 0.0010715815783630914, + "loss": 0.0786, + "step": 49456 + }, + { + "epoch": 0.4293105094573832, + "grad_norm": 0.671875, + "learning_rate": 0.0010715518307628765, + "loss": 0.1904, + "step": 49457 + }, + { + "epoch": 0.42931918993758733, + "grad_norm": 0.3203125, + "learning_rate": 0.0010715220831937718, + "loss": 0.0918, + "step": 49458 + }, + { + "epoch": 0.4293278704177915, + "grad_norm": 0.1669921875, + "learning_rate": 0.0010714923356558105, + "loss": 0.1045, + "step": 49459 + }, + { + "epoch": 0.42933655089799566, + "grad_norm": 0.07861328125, + "learning_rate": 0.0010714625881490247, + "loss": 0.0918, + "step": 49460 + }, + { + "epoch": 0.42934523137819985, + "grad_norm": 0.1318359375, + "learning_rate": 0.001071432840673447, + "loss": 0.0679, + "step": 49461 + }, + { + "epoch": 0.429353911858404, + "grad_norm": 0.56640625, + "learning_rate": 0.0010714030932291098, + "loss": 0.1138, + "step": 49462 + }, + { + "epoch": 0.4293625923386082, + "grad_norm": 0.26953125, + "learning_rate": 0.001071373345816046, + "loss": 0.1357, + "step": 49463 + }, + { + "epoch": 0.4293712728188123, + "grad_norm": 0.115234375, + "learning_rate": 0.0010713435984342878, + "loss": 0.1147, + "step": 49464 + }, + { + "epoch": 0.4293799532990165, + "grad_norm": 0.1455078125, + "learning_rate": 0.0010713138510838674, + "loss": 0.0967, + "step": 49465 + }, + { + "epoch": 0.42938863377922065, + "grad_norm": 0.072265625, + "learning_rate": 0.001071284103764818, + "loss": 0.0889, + "step": 49466 + }, + { + "epoch": 0.42939731425942484, + "grad_norm": 0.3828125, + "learning_rate": 0.001071254356477172, + "loss": 0.0986, + "step": 49467 + }, + { + "epoch": 0.429405994739629, + "grad_norm": 0.5859375, + "learning_rate": 0.0010712246092209616, + "loss": 0.0938, + "step": 49468 + }, + { + "epoch": 0.4294146752198332, + "grad_norm": 0.28125, + "learning_rate": 0.0010711948619962198, + "loss": 0.0972, + "step": 49469 + }, + { + "epoch": 0.4294233557000373, + "grad_norm": 0.15625, + "learning_rate": 0.0010711651148029785, + "loss": 0.0947, + "step": 49470 + }, + { + "epoch": 0.4294320361802415, + "grad_norm": 0.11572265625, + "learning_rate": 0.001071135367641271, + "loss": 0.0845, + "step": 49471 + }, + { + "epoch": 0.42944071666044564, + "grad_norm": 0.123046875, + "learning_rate": 0.0010711056205111294, + "loss": 0.0967, + "step": 49472 + }, + { + "epoch": 0.42944939714064984, + "grad_norm": 0.1767578125, + "learning_rate": 0.001071075873412586, + "loss": 0.0991, + "step": 49473 + }, + { + "epoch": 0.429458077620854, + "grad_norm": 0.1162109375, + "learning_rate": 0.0010710461263456735, + "loss": 0.0732, + "step": 49474 + }, + { + "epoch": 0.42946675810105817, + "grad_norm": 0.205078125, + "learning_rate": 0.0010710163793104248, + "loss": 0.1201, + "step": 49475 + }, + { + "epoch": 0.4294754385812623, + "grad_norm": 0.48828125, + "learning_rate": 0.001070986632306872, + "loss": 0.0762, + "step": 49476 + }, + { + "epoch": 0.4294841190614665, + "grad_norm": 0.23828125, + "learning_rate": 0.0010709568853350478, + "loss": 0.1064, + "step": 49477 + }, + { + "epoch": 0.42949279954167063, + "grad_norm": 0.08740234375, + "learning_rate": 0.0010709271383949849, + "loss": 0.1104, + "step": 49478 + }, + { + "epoch": 0.4295014800218748, + "grad_norm": 0.171875, + "learning_rate": 0.0010708973914867156, + "loss": 0.0898, + "step": 49479 + }, + { + "epoch": 0.42951016050207896, + "grad_norm": 0.109375, + "learning_rate": 0.0010708676446102724, + "loss": 0.0938, + "step": 49480 + }, + { + "epoch": 0.42951884098228316, + "grad_norm": 0.1416015625, + "learning_rate": 0.0010708378977656878, + "loss": 0.0957, + "step": 49481 + }, + { + "epoch": 0.4295275214624873, + "grad_norm": 0.11962890625, + "learning_rate": 0.0010708081509529944, + "loss": 0.124, + "step": 49482 + }, + { + "epoch": 0.4295362019426915, + "grad_norm": 0.09619140625, + "learning_rate": 0.0010707784041722249, + "loss": 0.0898, + "step": 49483 + }, + { + "epoch": 0.4295448824228956, + "grad_norm": 0.1728515625, + "learning_rate": 0.0010707486574234118, + "loss": 0.125, + "step": 49484 + }, + { + "epoch": 0.4295535629030998, + "grad_norm": 0.2119140625, + "learning_rate": 0.0010707189107065874, + "loss": 0.0713, + "step": 49485 + }, + { + "epoch": 0.42956224338330395, + "grad_norm": 0.490234375, + "learning_rate": 0.001070689164021784, + "loss": 0.0815, + "step": 49486 + }, + { + "epoch": 0.42957092386350815, + "grad_norm": 0.099609375, + "learning_rate": 0.0010706594173690352, + "loss": 0.0903, + "step": 49487 + }, + { + "epoch": 0.4295796043437123, + "grad_norm": 0.51953125, + "learning_rate": 0.0010706296707483725, + "loss": 0.0977, + "step": 49488 + }, + { + "epoch": 0.4295882848239165, + "grad_norm": 0.11865234375, + "learning_rate": 0.0010705999241598287, + "loss": 0.126, + "step": 49489 + }, + { + "epoch": 0.4295969653041206, + "grad_norm": 0.2373046875, + "learning_rate": 0.0010705701776034364, + "loss": 0.104, + "step": 49490 + }, + { + "epoch": 0.4296056457843248, + "grad_norm": 0.08935546875, + "learning_rate": 0.001070540431079228, + "loss": 0.1055, + "step": 49491 + }, + { + "epoch": 0.42961432626452895, + "grad_norm": 0.6953125, + "learning_rate": 0.0010705106845872363, + "loss": 0.1104, + "step": 49492 + }, + { + "epoch": 0.42962300674473314, + "grad_norm": 0.2060546875, + "learning_rate": 0.0010704809381274936, + "loss": 0.1172, + "step": 49493 + }, + { + "epoch": 0.4296316872249373, + "grad_norm": 0.1748046875, + "learning_rate": 0.0010704511917000328, + "loss": 0.0898, + "step": 49494 + }, + { + "epoch": 0.42964036770514147, + "grad_norm": 0.75, + "learning_rate": 0.0010704214453048856, + "loss": 0.1475, + "step": 49495 + }, + { + "epoch": 0.4296490481853456, + "grad_norm": 0.28515625, + "learning_rate": 0.0010703916989420854, + "loss": 0.0723, + "step": 49496 + }, + { + "epoch": 0.4296577286655498, + "grad_norm": 0.1455078125, + "learning_rate": 0.0010703619526116646, + "loss": 0.0732, + "step": 49497 + }, + { + "epoch": 0.42966640914575394, + "grad_norm": 0.1875, + "learning_rate": 0.0010703322063136552, + "loss": 0.1001, + "step": 49498 + }, + { + "epoch": 0.42967508962595813, + "grad_norm": 0.10107421875, + "learning_rate": 0.00107030246004809, + "loss": 0.1123, + "step": 49499 + }, + { + "epoch": 0.42968377010616227, + "grad_norm": 0.1181640625, + "learning_rate": 0.0010702727138150014, + "loss": 0.1094, + "step": 49500 + }, + { + "epoch": 0.42969245058636646, + "grad_norm": 0.28125, + "learning_rate": 0.0010702429676144225, + "loss": 0.104, + "step": 49501 + }, + { + "epoch": 0.4297011310665706, + "grad_norm": 0.259765625, + "learning_rate": 0.0010702132214463855, + "loss": 0.1094, + "step": 49502 + }, + { + "epoch": 0.4297098115467748, + "grad_norm": 0.484375, + "learning_rate": 0.001070183475310923, + "loss": 0.085, + "step": 49503 + }, + { + "epoch": 0.4297184920269789, + "grad_norm": 0.416015625, + "learning_rate": 0.0010701537292080667, + "loss": 0.1641, + "step": 49504 + }, + { + "epoch": 0.4297271725071831, + "grad_norm": 0.1865234375, + "learning_rate": 0.0010701239831378505, + "loss": 0.0859, + "step": 49505 + }, + { + "epoch": 0.42973585298738726, + "grad_norm": 0.0927734375, + "learning_rate": 0.001070094237100306, + "loss": 0.1035, + "step": 49506 + }, + { + "epoch": 0.42974453346759145, + "grad_norm": 0.11376953125, + "learning_rate": 0.0010700644910954663, + "loss": 0.0889, + "step": 49507 + }, + { + "epoch": 0.4297532139477956, + "grad_norm": 0.53515625, + "learning_rate": 0.001070034745123363, + "loss": 0.126, + "step": 49508 + }, + { + "epoch": 0.4297618944279998, + "grad_norm": 0.330078125, + "learning_rate": 0.0010700049991840294, + "loss": 0.1006, + "step": 49509 + }, + { + "epoch": 0.4297705749082039, + "grad_norm": 0.2353515625, + "learning_rate": 0.0010699752532774983, + "loss": 0.0596, + "step": 49510 + }, + { + "epoch": 0.4297792553884081, + "grad_norm": 0.10546875, + "learning_rate": 0.0010699455074038016, + "loss": 0.0928, + "step": 49511 + }, + { + "epoch": 0.42978793586861225, + "grad_norm": 0.146484375, + "learning_rate": 0.001069915761562972, + "loss": 0.126, + "step": 49512 + }, + { + "epoch": 0.42979661634881644, + "grad_norm": 0.0888671875, + "learning_rate": 0.0010698860157550424, + "loss": 0.1133, + "step": 49513 + }, + { + "epoch": 0.4298052968290206, + "grad_norm": 0.2314453125, + "learning_rate": 0.0010698562699800445, + "loss": 0.1289, + "step": 49514 + }, + { + "epoch": 0.42981397730922477, + "grad_norm": 0.380859375, + "learning_rate": 0.0010698265242380114, + "loss": 0.1289, + "step": 49515 + }, + { + "epoch": 0.4298226577894289, + "grad_norm": 0.1767578125, + "learning_rate": 0.001069796778528976, + "loss": 0.1562, + "step": 49516 + }, + { + "epoch": 0.4298313382696331, + "grad_norm": 0.7734375, + "learning_rate": 0.0010697670328529696, + "loss": 0.0801, + "step": 49517 + }, + { + "epoch": 0.42984001874983724, + "grad_norm": 0.4375, + "learning_rate": 0.0010697372872100262, + "loss": 0.1221, + "step": 49518 + }, + { + "epoch": 0.42984869923004143, + "grad_norm": 0.2578125, + "learning_rate": 0.0010697075416001773, + "loss": 0.082, + "step": 49519 + }, + { + "epoch": 0.42985737971024557, + "grad_norm": 0.21875, + "learning_rate": 0.0010696777960234561, + "loss": 0.0791, + "step": 49520 + }, + { + "epoch": 0.42986606019044976, + "grad_norm": 0.1337890625, + "learning_rate": 0.0010696480504798947, + "loss": 0.083, + "step": 49521 + }, + { + "epoch": 0.4298747406706539, + "grad_norm": 0.359375, + "learning_rate": 0.0010696183049695256, + "loss": 0.0986, + "step": 49522 + }, + { + "epoch": 0.4298834211508581, + "grad_norm": 0.111328125, + "learning_rate": 0.0010695885594923815, + "loss": 0.1113, + "step": 49523 + }, + { + "epoch": 0.42989210163106223, + "grad_norm": 0.1650390625, + "learning_rate": 0.001069558814048495, + "loss": 0.0757, + "step": 49524 + }, + { + "epoch": 0.4299007821112664, + "grad_norm": 0.259765625, + "learning_rate": 0.0010695290686378984, + "loss": 0.0693, + "step": 49525 + }, + { + "epoch": 0.42990946259147056, + "grad_norm": 0.1220703125, + "learning_rate": 0.0010694993232606242, + "loss": 0.1182, + "step": 49526 + }, + { + "epoch": 0.42991814307167475, + "grad_norm": 0.384765625, + "learning_rate": 0.0010694695779167054, + "loss": 0.1035, + "step": 49527 + }, + { + "epoch": 0.4299268235518789, + "grad_norm": 0.12451171875, + "learning_rate": 0.001069439832606174, + "loss": 0.0762, + "step": 49528 + }, + { + "epoch": 0.4299355040320831, + "grad_norm": 0.6484375, + "learning_rate": 0.001069410087329063, + "loss": 0.1328, + "step": 49529 + }, + { + "epoch": 0.4299441845122872, + "grad_norm": 0.41796875, + "learning_rate": 0.0010693803420854047, + "loss": 0.1108, + "step": 49530 + }, + { + "epoch": 0.42995286499249136, + "grad_norm": 0.48828125, + "learning_rate": 0.0010693505968752313, + "loss": 0.1162, + "step": 49531 + }, + { + "epoch": 0.42996154547269555, + "grad_norm": 0.11767578125, + "learning_rate": 0.0010693208516985757, + "loss": 0.083, + "step": 49532 + }, + { + "epoch": 0.4299702259528997, + "grad_norm": 0.28125, + "learning_rate": 0.0010692911065554704, + "loss": 0.0752, + "step": 49533 + }, + { + "epoch": 0.4299789064331039, + "grad_norm": 0.419921875, + "learning_rate": 0.0010692613614459478, + "loss": 0.106, + "step": 49534 + }, + { + "epoch": 0.429987586913308, + "grad_norm": 0.390625, + "learning_rate": 0.0010692316163700404, + "loss": 0.0767, + "step": 49535 + }, + { + "epoch": 0.4299962673935122, + "grad_norm": 0.380859375, + "learning_rate": 0.001069201871327781, + "loss": 0.1104, + "step": 49536 + }, + { + "epoch": 0.43000494787371635, + "grad_norm": 0.29296875, + "learning_rate": 0.001069172126319202, + "loss": 0.1758, + "step": 49537 + }, + { + "epoch": 0.43001362835392054, + "grad_norm": 0.1162109375, + "learning_rate": 0.0010691423813443362, + "loss": 0.0898, + "step": 49538 + }, + { + "epoch": 0.4300223088341247, + "grad_norm": 0.09814453125, + "learning_rate": 0.0010691126364032156, + "loss": 0.1152, + "step": 49539 + }, + { + "epoch": 0.43003098931432887, + "grad_norm": 0.216796875, + "learning_rate": 0.0010690828914958728, + "loss": 0.1338, + "step": 49540 + }, + { + "epoch": 0.430039669794533, + "grad_norm": 0.5390625, + "learning_rate": 0.0010690531466223405, + "loss": 0.0879, + "step": 49541 + }, + { + "epoch": 0.4300483502747372, + "grad_norm": 0.1484375, + "learning_rate": 0.0010690234017826513, + "loss": 0.083, + "step": 49542 + }, + { + "epoch": 0.43005703075494134, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010689936569768376, + "loss": 0.1025, + "step": 49543 + }, + { + "epoch": 0.43006571123514553, + "grad_norm": 0.177734375, + "learning_rate": 0.0010689639122049316, + "loss": 0.1074, + "step": 49544 + }, + { + "epoch": 0.43007439171534967, + "grad_norm": 0.478515625, + "learning_rate": 0.0010689341674669668, + "loss": 0.1006, + "step": 49545 + }, + { + "epoch": 0.43008307219555386, + "grad_norm": 0.375, + "learning_rate": 0.0010689044227629749, + "loss": 0.0898, + "step": 49546 + }, + { + "epoch": 0.430091752675758, + "grad_norm": 0.2373046875, + "learning_rate": 0.0010688746780929887, + "loss": 0.1133, + "step": 49547 + }, + { + "epoch": 0.4301004331559622, + "grad_norm": 0.154296875, + "learning_rate": 0.0010688449334570408, + "loss": 0.1504, + "step": 49548 + }, + { + "epoch": 0.43010911363616633, + "grad_norm": 0.7109375, + "learning_rate": 0.0010688151888551636, + "loss": 0.0864, + "step": 49549 + }, + { + "epoch": 0.4301177941163705, + "grad_norm": 0.2578125, + "learning_rate": 0.0010687854442873897, + "loss": 0.0742, + "step": 49550 + }, + { + "epoch": 0.43012647459657466, + "grad_norm": 0.58984375, + "learning_rate": 0.001068755699753751, + "loss": 0.3906, + "step": 49551 + }, + { + "epoch": 0.43013515507677885, + "grad_norm": 0.4140625, + "learning_rate": 0.001068725955254281, + "loss": 0.1025, + "step": 49552 + }, + { + "epoch": 0.430143835556983, + "grad_norm": 0.1572265625, + "learning_rate": 0.001068696210789012, + "loss": 0.1001, + "step": 49553 + }, + { + "epoch": 0.4301525160371872, + "grad_norm": 0.41015625, + "learning_rate": 0.001068666466357976, + "loss": 0.1064, + "step": 49554 + }, + { + "epoch": 0.4301611965173913, + "grad_norm": 0.2109375, + "learning_rate": 0.0010686367219612059, + "loss": 0.0762, + "step": 49555 + }, + { + "epoch": 0.4301698769975955, + "grad_norm": 0.1591796875, + "learning_rate": 0.0010686069775987343, + "loss": 0.1016, + "step": 49556 + }, + { + "epoch": 0.43017855747779965, + "grad_norm": 0.130859375, + "learning_rate": 0.0010685772332705938, + "loss": 0.1113, + "step": 49557 + }, + { + "epoch": 0.43018723795800384, + "grad_norm": 0.1396484375, + "learning_rate": 0.001068547488976817, + "loss": 0.0898, + "step": 49558 + }, + { + "epoch": 0.430195918438208, + "grad_norm": 0.8203125, + "learning_rate": 0.0010685177447174357, + "loss": 0.2539, + "step": 49559 + }, + { + "epoch": 0.4302045989184122, + "grad_norm": 0.1728515625, + "learning_rate": 0.0010684880004924828, + "loss": 0.0957, + "step": 49560 + }, + { + "epoch": 0.4302132793986163, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010684582563019913, + "loss": 0.085, + "step": 49561 + }, + { + "epoch": 0.4302219598788205, + "grad_norm": 0.326171875, + "learning_rate": 0.0010684285121459933, + "loss": 0.1172, + "step": 49562 + }, + { + "epoch": 0.43023064035902464, + "grad_norm": 0.140625, + "learning_rate": 0.0010683987680245213, + "loss": 0.0776, + "step": 49563 + }, + { + "epoch": 0.43023932083922883, + "grad_norm": 0.1025390625, + "learning_rate": 0.0010683690239376082, + "loss": 0.124, + "step": 49564 + }, + { + "epoch": 0.43024800131943297, + "grad_norm": 0.1123046875, + "learning_rate": 0.001068339279885286, + "loss": 0.083, + "step": 49565 + }, + { + "epoch": 0.43025668179963716, + "grad_norm": 0.1630859375, + "learning_rate": 0.0010683095358675879, + "loss": 0.0693, + "step": 49566 + }, + { + "epoch": 0.4302653622798413, + "grad_norm": 0.3359375, + "learning_rate": 0.0010682797918845457, + "loss": 0.0962, + "step": 49567 + }, + { + "epoch": 0.4302740427600455, + "grad_norm": 0.458984375, + "learning_rate": 0.0010682500479361923, + "loss": 0.0977, + "step": 49568 + }, + { + "epoch": 0.43028272324024963, + "grad_norm": 0.07861328125, + "learning_rate": 0.0010682203040225594, + "loss": 0.0796, + "step": 49569 + }, + { + "epoch": 0.4302914037204538, + "grad_norm": 0.10400390625, + "learning_rate": 0.0010681905601436813, + "loss": 0.0977, + "step": 49570 + }, + { + "epoch": 0.43030008420065796, + "grad_norm": 0.283203125, + "learning_rate": 0.0010681608162995892, + "loss": 0.0928, + "step": 49571 + }, + { + "epoch": 0.43030876468086215, + "grad_norm": 0.1728515625, + "learning_rate": 0.001068131072490316, + "loss": 0.1162, + "step": 49572 + }, + { + "epoch": 0.4303174451610663, + "grad_norm": 0.1865234375, + "learning_rate": 0.0010681013287158943, + "loss": 0.0845, + "step": 49573 + }, + { + "epoch": 0.4303261256412705, + "grad_norm": 0.251953125, + "learning_rate": 0.0010680715849763563, + "loss": 0.0864, + "step": 49574 + }, + { + "epoch": 0.4303348061214746, + "grad_norm": 0.08935546875, + "learning_rate": 0.001068041841271735, + "loss": 0.0767, + "step": 49575 + }, + { + "epoch": 0.4303434866016788, + "grad_norm": 0.291015625, + "learning_rate": 0.0010680120976020624, + "loss": 0.0986, + "step": 49576 + }, + { + "epoch": 0.43035216708188295, + "grad_norm": 0.291015625, + "learning_rate": 0.0010679823539673713, + "loss": 0.0669, + "step": 49577 + }, + { + "epoch": 0.43036084756208715, + "grad_norm": 0.1669921875, + "learning_rate": 0.0010679526103676942, + "loss": 0.1621, + "step": 49578 + }, + { + "epoch": 0.4303695280422913, + "grad_norm": 0.11181640625, + "learning_rate": 0.0010679228668030636, + "loss": 0.1084, + "step": 49579 + }, + { + "epoch": 0.4303782085224955, + "grad_norm": 0.384765625, + "learning_rate": 0.0010678931232735124, + "loss": 0.0957, + "step": 49580 + }, + { + "epoch": 0.4303868890026996, + "grad_norm": 0.2451171875, + "learning_rate": 0.0010678633797790726, + "loss": 0.0752, + "step": 49581 + }, + { + "epoch": 0.4303955694829038, + "grad_norm": 0.10791015625, + "learning_rate": 0.0010678336363197768, + "loss": 0.0918, + "step": 49582 + }, + { + "epoch": 0.43040424996310794, + "grad_norm": 0.2578125, + "learning_rate": 0.001067803892895658, + "loss": 0.1152, + "step": 49583 + }, + { + "epoch": 0.43041293044331214, + "grad_norm": 0.4609375, + "learning_rate": 0.0010677741495067482, + "loss": 0.1572, + "step": 49584 + }, + { + "epoch": 0.4304216109235163, + "grad_norm": 0.125, + "learning_rate": 0.00106774440615308, + "loss": 0.0889, + "step": 49585 + }, + { + "epoch": 0.43043029140372047, + "grad_norm": 0.208984375, + "learning_rate": 0.0010677146628346856, + "loss": 0.0835, + "step": 49586 + }, + { + "epoch": 0.4304389718839246, + "grad_norm": 0.28125, + "learning_rate": 0.0010676849195515987, + "loss": 0.103, + "step": 49587 + }, + { + "epoch": 0.4304476523641288, + "grad_norm": 0.5390625, + "learning_rate": 0.001067655176303851, + "loss": 0.0864, + "step": 49588 + }, + { + "epoch": 0.43045633284433293, + "grad_norm": 0.06884765625, + "learning_rate": 0.001067625433091475, + "loss": 0.0645, + "step": 49589 + }, + { + "epoch": 0.4304650133245371, + "grad_norm": 0.07958984375, + "learning_rate": 0.001067595689914503, + "loss": 0.0679, + "step": 49590 + }, + { + "epoch": 0.43047369380474126, + "grad_norm": 0.30859375, + "learning_rate": 0.001067565946772968, + "loss": 0.084, + "step": 49591 + }, + { + "epoch": 0.43048237428494546, + "grad_norm": 0.439453125, + "learning_rate": 0.0010675362036669028, + "loss": 0.1084, + "step": 49592 + }, + { + "epoch": 0.4304910547651496, + "grad_norm": 0.154296875, + "learning_rate": 0.0010675064605963395, + "loss": 0.0879, + "step": 49593 + }, + { + "epoch": 0.4304997352453538, + "grad_norm": 0.3828125, + "learning_rate": 0.0010674767175613102, + "loss": 0.1196, + "step": 49594 + }, + { + "epoch": 0.4305084157255579, + "grad_norm": 0.1396484375, + "learning_rate": 0.0010674469745618478, + "loss": 0.0688, + "step": 49595 + }, + { + "epoch": 0.4305170962057621, + "grad_norm": 0.58984375, + "learning_rate": 0.0010674172315979853, + "loss": 0.1162, + "step": 49596 + }, + { + "epoch": 0.43052577668596625, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010673874886697548, + "loss": 0.0991, + "step": 49597 + }, + { + "epoch": 0.43053445716617045, + "grad_norm": 0.205078125, + "learning_rate": 0.0010673577457771888, + "loss": 0.1045, + "step": 49598 + }, + { + "epoch": 0.4305431376463746, + "grad_norm": 0.30078125, + "learning_rate": 0.0010673280029203198, + "loss": 0.1064, + "step": 49599 + }, + { + "epoch": 0.4305518181265788, + "grad_norm": 0.26953125, + "learning_rate": 0.0010672982600991804, + "loss": 0.1138, + "step": 49600 + }, + { + "epoch": 0.4305604986067829, + "grad_norm": 0.578125, + "learning_rate": 0.0010672685173138032, + "loss": 0.1387, + "step": 49601 + }, + { + "epoch": 0.4305691790869871, + "grad_norm": 0.0927734375, + "learning_rate": 0.0010672387745642208, + "loss": 0.1328, + "step": 49602 + }, + { + "epoch": 0.43057785956719125, + "grad_norm": 0.875, + "learning_rate": 0.001067209031850465, + "loss": 0.0962, + "step": 49603 + }, + { + "epoch": 0.43058654004739544, + "grad_norm": 0.33984375, + "learning_rate": 0.0010671792891725693, + "loss": 0.106, + "step": 49604 + }, + { + "epoch": 0.4305952205275996, + "grad_norm": 0.314453125, + "learning_rate": 0.0010671495465305661, + "loss": 0.0645, + "step": 49605 + }, + { + "epoch": 0.43060390100780377, + "grad_norm": 0.11962890625, + "learning_rate": 0.0010671198039244874, + "loss": 0.085, + "step": 49606 + }, + { + "epoch": 0.4306125814880079, + "grad_norm": 0.4296875, + "learning_rate": 0.0010670900613543662, + "loss": 0.1523, + "step": 49607 + }, + { + "epoch": 0.4306212619682121, + "grad_norm": 0.31640625, + "learning_rate": 0.0010670603188202346, + "loss": 0.1099, + "step": 49608 + }, + { + "epoch": 0.43062994244841624, + "grad_norm": 0.130859375, + "learning_rate": 0.0010670305763221253, + "loss": 0.1055, + "step": 49609 + }, + { + "epoch": 0.43063862292862043, + "grad_norm": 0.26171875, + "learning_rate": 0.0010670008338600709, + "loss": 0.0933, + "step": 49610 + }, + { + "epoch": 0.43064730340882457, + "grad_norm": 0.2158203125, + "learning_rate": 0.0010669710914341039, + "loss": 0.1201, + "step": 49611 + }, + { + "epoch": 0.43065598388902876, + "grad_norm": 0.412109375, + "learning_rate": 0.0010669413490442567, + "loss": 0.082, + "step": 49612 + }, + { + "epoch": 0.4306646643692329, + "grad_norm": 0.365234375, + "learning_rate": 0.0010669116066905622, + "loss": 0.0908, + "step": 49613 + }, + { + "epoch": 0.4306733448494371, + "grad_norm": 0.1572265625, + "learning_rate": 0.0010668818643730525, + "loss": 0.0718, + "step": 49614 + }, + { + "epoch": 0.4306820253296412, + "grad_norm": 1.6015625, + "learning_rate": 0.0010668521220917604, + "loss": 0.1611, + "step": 49615 + }, + { + "epoch": 0.4306907058098454, + "grad_norm": 0.2255859375, + "learning_rate": 0.0010668223798467186, + "loss": 0.1006, + "step": 49616 + }, + { + "epoch": 0.43069938629004956, + "grad_norm": 0.1572265625, + "learning_rate": 0.001066792637637959, + "loss": 0.0874, + "step": 49617 + }, + { + "epoch": 0.43070806677025375, + "grad_norm": 0.6875, + "learning_rate": 0.0010667628954655144, + "loss": 0.0908, + "step": 49618 + }, + { + "epoch": 0.4307167472504579, + "grad_norm": 0.76171875, + "learning_rate": 0.0010667331533294174, + "loss": 0.1162, + "step": 49619 + }, + { + "epoch": 0.4307254277306621, + "grad_norm": 0.1513671875, + "learning_rate": 0.0010667034112297007, + "loss": 0.1123, + "step": 49620 + }, + { + "epoch": 0.4307341082108662, + "grad_norm": 0.287109375, + "learning_rate": 0.0010666736691663965, + "loss": 0.0801, + "step": 49621 + }, + { + "epoch": 0.4307427886910704, + "grad_norm": 0.515625, + "learning_rate": 0.0010666439271395374, + "loss": 0.0938, + "step": 49622 + }, + { + "epoch": 0.43075146917127455, + "grad_norm": 0.095703125, + "learning_rate": 0.0010666141851491562, + "loss": 0.0693, + "step": 49623 + }, + { + "epoch": 0.43076014965147874, + "grad_norm": 0.35546875, + "learning_rate": 0.0010665844431952851, + "loss": 0.1201, + "step": 49624 + }, + { + "epoch": 0.4307688301316829, + "grad_norm": 0.2470703125, + "learning_rate": 0.001066554701277957, + "loss": 0.1016, + "step": 49625 + }, + { + "epoch": 0.43077751061188707, + "grad_norm": 0.376953125, + "learning_rate": 0.001066524959397204, + "loss": 0.0791, + "step": 49626 + }, + { + "epoch": 0.4307861910920912, + "grad_norm": 0.359375, + "learning_rate": 0.0010664952175530589, + "loss": 0.0869, + "step": 49627 + }, + { + "epoch": 0.4307948715722954, + "grad_norm": 0.3515625, + "learning_rate": 0.0010664654757455538, + "loss": 0.1035, + "step": 49628 + }, + { + "epoch": 0.43080355205249954, + "grad_norm": 0.61328125, + "learning_rate": 0.0010664357339747216, + "loss": 0.1875, + "step": 49629 + }, + { + "epoch": 0.43081223253270373, + "grad_norm": 0.142578125, + "learning_rate": 0.0010664059922405951, + "loss": 0.123, + "step": 49630 + }, + { + "epoch": 0.43082091301290787, + "grad_norm": 0.09130859375, + "learning_rate": 0.0010663762505432061, + "loss": 0.0879, + "step": 49631 + }, + { + "epoch": 0.43082959349311206, + "grad_norm": 0.27734375, + "learning_rate": 0.001066346508882588, + "loss": 0.1406, + "step": 49632 + }, + { + "epoch": 0.4308382739733162, + "grad_norm": 0.12109375, + "learning_rate": 0.0010663167672587729, + "loss": 0.0967, + "step": 49633 + }, + { + "epoch": 0.4308469544535204, + "grad_norm": 0.09619140625, + "learning_rate": 0.001066287025671793, + "loss": 0.1309, + "step": 49634 + }, + { + "epoch": 0.43085563493372453, + "grad_norm": 0.1455078125, + "learning_rate": 0.0010662572841216812, + "loss": 0.0645, + "step": 49635 + }, + { + "epoch": 0.4308643154139287, + "grad_norm": 0.1396484375, + "learning_rate": 0.0010662275426084698, + "loss": 0.0879, + "step": 49636 + }, + { + "epoch": 0.43087299589413286, + "grad_norm": 0.10693359375, + "learning_rate": 0.0010661978011321912, + "loss": 0.0918, + "step": 49637 + }, + { + "epoch": 0.43088167637433705, + "grad_norm": 0.103515625, + "learning_rate": 0.0010661680596928784, + "loss": 0.1045, + "step": 49638 + }, + { + "epoch": 0.4308903568545412, + "grad_norm": 0.15234375, + "learning_rate": 0.0010661383182905637, + "loss": 0.0811, + "step": 49639 + }, + { + "epoch": 0.4308990373347454, + "grad_norm": 0.54296875, + "learning_rate": 0.0010661085769252795, + "loss": 0.0879, + "step": 49640 + }, + { + "epoch": 0.4309077178149495, + "grad_norm": 0.34375, + "learning_rate": 0.001066078835597059, + "loss": 0.1055, + "step": 49641 + }, + { + "epoch": 0.4309163982951537, + "grad_norm": 0.265625, + "learning_rate": 0.0010660490943059338, + "loss": 0.1172, + "step": 49642 + }, + { + "epoch": 0.43092507877535785, + "grad_norm": 0.5234375, + "learning_rate": 0.001066019353051937, + "loss": 0.1211, + "step": 49643 + }, + { + "epoch": 0.43093375925556204, + "grad_norm": 0.12060546875, + "learning_rate": 0.0010659896118351006, + "loss": 0.0889, + "step": 49644 + }, + { + "epoch": 0.4309424397357662, + "grad_norm": 0.3046875, + "learning_rate": 0.0010659598706554579, + "loss": 0.0981, + "step": 49645 + }, + { + "epoch": 0.4309511202159704, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010659301295130406, + "loss": 0.0898, + "step": 49646 + }, + { + "epoch": 0.4309598006961745, + "grad_norm": 0.15625, + "learning_rate": 0.0010659003884078812, + "loss": 0.0693, + "step": 49647 + }, + { + "epoch": 0.4309684811763787, + "grad_norm": 0.14453125, + "learning_rate": 0.0010658706473400132, + "loss": 0.0889, + "step": 49648 + }, + { + "epoch": 0.43097716165658284, + "grad_norm": 0.1884765625, + "learning_rate": 0.0010658409063094684, + "loss": 0.0747, + "step": 49649 + }, + { + "epoch": 0.43098584213678703, + "grad_norm": 0.267578125, + "learning_rate": 0.0010658111653162796, + "loss": 0.1064, + "step": 49650 + }, + { + "epoch": 0.43099452261699117, + "grad_norm": 0.16796875, + "learning_rate": 0.001065781424360479, + "loss": 0.1426, + "step": 49651 + }, + { + "epoch": 0.43100320309719536, + "grad_norm": 0.1416015625, + "learning_rate": 0.0010657516834420998, + "loss": 0.0825, + "step": 49652 + }, + { + "epoch": 0.4310118835773995, + "grad_norm": 0.1484375, + "learning_rate": 0.0010657219425611738, + "loss": 0.0952, + "step": 49653 + }, + { + "epoch": 0.43102056405760364, + "grad_norm": 0.150390625, + "learning_rate": 0.0010656922017177335, + "loss": 0.0918, + "step": 49654 + }, + { + "epoch": 0.43102924453780783, + "grad_norm": 0.12890625, + "learning_rate": 0.0010656624609118115, + "loss": 0.1367, + "step": 49655 + }, + { + "epoch": 0.43103792501801197, + "grad_norm": 0.61328125, + "learning_rate": 0.0010656327201434407, + "loss": 0.1113, + "step": 49656 + }, + { + "epoch": 0.43104660549821616, + "grad_norm": 0.171875, + "learning_rate": 0.0010656029794126538, + "loss": 0.126, + "step": 49657 + }, + { + "epoch": 0.4310552859784203, + "grad_norm": 0.1015625, + "learning_rate": 0.001065573238719483, + "loss": 0.0635, + "step": 49658 + }, + { + "epoch": 0.4310639664586245, + "grad_norm": 0.294921875, + "learning_rate": 0.0010655434980639603, + "loss": 0.0913, + "step": 49659 + }, + { + "epoch": 0.43107264693882863, + "grad_norm": 0.470703125, + "learning_rate": 0.0010655137574461192, + "loss": 0.1104, + "step": 49660 + }, + { + "epoch": 0.4310813274190328, + "grad_norm": 0.25, + "learning_rate": 0.0010654840168659915, + "loss": 0.1328, + "step": 49661 + }, + { + "epoch": 0.43109000789923696, + "grad_norm": 0.130859375, + "learning_rate": 0.00106545427632361, + "loss": 0.0835, + "step": 49662 + }, + { + "epoch": 0.43109868837944115, + "grad_norm": 0.24609375, + "learning_rate": 0.0010654245358190071, + "loss": 0.1113, + "step": 49663 + }, + { + "epoch": 0.4311073688596453, + "grad_norm": 0.1962890625, + "learning_rate": 0.0010653947953522152, + "loss": 0.1318, + "step": 49664 + }, + { + "epoch": 0.4311160493398495, + "grad_norm": 0.1650390625, + "learning_rate": 0.0010653650549232675, + "loss": 0.1113, + "step": 49665 + }, + { + "epoch": 0.4311247298200536, + "grad_norm": 0.1015625, + "learning_rate": 0.0010653353145321956, + "loss": 0.0623, + "step": 49666 + }, + { + "epoch": 0.4311334103002578, + "grad_norm": 0.06689453125, + "learning_rate": 0.001065305574179033, + "loss": 0.0781, + "step": 49667 + }, + { + "epoch": 0.43114209078046195, + "grad_norm": 0.625, + "learning_rate": 0.0010652758338638112, + "loss": 0.1055, + "step": 49668 + }, + { + "epoch": 0.43115077126066614, + "grad_norm": 0.1064453125, + "learning_rate": 0.0010652460935865636, + "loss": 0.0752, + "step": 49669 + }, + { + "epoch": 0.4311594517408703, + "grad_norm": 0.345703125, + "learning_rate": 0.001065216353347322, + "loss": 0.1426, + "step": 49670 + }, + { + "epoch": 0.4311681322210745, + "grad_norm": 0.16796875, + "learning_rate": 0.0010651866131461197, + "loss": 0.0854, + "step": 49671 + }, + { + "epoch": 0.4311768127012786, + "grad_norm": 0.640625, + "learning_rate": 0.0010651568729829887, + "loss": 0.0991, + "step": 49672 + }, + { + "epoch": 0.4311854931814828, + "grad_norm": 0.189453125, + "learning_rate": 0.001065127132857961, + "loss": 0.0815, + "step": 49673 + }, + { + "epoch": 0.43119417366168694, + "grad_norm": 0.296875, + "learning_rate": 0.0010650973927710703, + "loss": 0.1299, + "step": 49674 + }, + { + "epoch": 0.43120285414189113, + "grad_norm": 0.306640625, + "learning_rate": 0.0010650676527223487, + "loss": 0.1074, + "step": 49675 + }, + { + "epoch": 0.43121153462209527, + "grad_norm": 0.3046875, + "learning_rate": 0.0010650379127118284, + "loss": 0.0815, + "step": 49676 + }, + { + "epoch": 0.43122021510229946, + "grad_norm": 0.06689453125, + "learning_rate": 0.001065008172739542, + "loss": 0.0713, + "step": 49677 + }, + { + "epoch": 0.4312288955825036, + "grad_norm": 0.185546875, + "learning_rate": 0.0010649784328055222, + "loss": 0.1338, + "step": 49678 + }, + { + "epoch": 0.4312375760627078, + "grad_norm": 0.10400390625, + "learning_rate": 0.0010649486929098016, + "loss": 0.1177, + "step": 49679 + }, + { + "epoch": 0.43124625654291193, + "grad_norm": 0.42578125, + "learning_rate": 0.0010649189530524126, + "loss": 0.1465, + "step": 49680 + }, + { + "epoch": 0.4312549370231161, + "grad_norm": 0.2578125, + "learning_rate": 0.0010648892132333872, + "loss": 0.1104, + "step": 49681 + }, + { + "epoch": 0.43126361750332026, + "grad_norm": 0.68359375, + "learning_rate": 0.0010648594734527588, + "loss": 0.1201, + "step": 49682 + }, + { + "epoch": 0.43127229798352446, + "grad_norm": 0.43359375, + "learning_rate": 0.0010648297337105597, + "loss": 0.1211, + "step": 49683 + }, + { + "epoch": 0.4312809784637286, + "grad_norm": 0.234375, + "learning_rate": 0.0010647999940068221, + "loss": 0.0889, + "step": 49684 + }, + { + "epoch": 0.4312896589439328, + "grad_norm": 0.29296875, + "learning_rate": 0.0010647702543415789, + "loss": 0.1211, + "step": 49685 + }, + { + "epoch": 0.4312983394241369, + "grad_norm": 0.74609375, + "learning_rate": 0.0010647405147148623, + "loss": 0.1045, + "step": 49686 + }, + { + "epoch": 0.4313070199043411, + "grad_norm": 0.1806640625, + "learning_rate": 0.0010647107751267048, + "loss": 0.123, + "step": 49687 + }, + { + "epoch": 0.43131570038454525, + "grad_norm": 0.1650390625, + "learning_rate": 0.001064681035577139, + "loss": 0.1064, + "step": 49688 + }, + { + "epoch": 0.43132438086474945, + "grad_norm": 0.2021484375, + "learning_rate": 0.0010646512960661978, + "loss": 0.0811, + "step": 49689 + }, + { + "epoch": 0.4313330613449536, + "grad_norm": 0.11279296875, + "learning_rate": 0.001064621556593913, + "loss": 0.0801, + "step": 49690 + }, + { + "epoch": 0.4313417418251578, + "grad_norm": 0.412109375, + "learning_rate": 0.0010645918171603179, + "loss": 0.064, + "step": 49691 + }, + { + "epoch": 0.4313504223053619, + "grad_norm": 0.1376953125, + "learning_rate": 0.0010645620777654445, + "loss": 0.123, + "step": 49692 + }, + { + "epoch": 0.4313591027855661, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010645323384093257, + "loss": 0.1074, + "step": 49693 + }, + { + "epoch": 0.43136778326577024, + "grad_norm": 0.427734375, + "learning_rate": 0.0010645025990919936, + "loss": 0.1113, + "step": 49694 + }, + { + "epoch": 0.43137646374597444, + "grad_norm": 0.09228515625, + "learning_rate": 0.0010644728598134812, + "loss": 0.083, + "step": 49695 + }, + { + "epoch": 0.4313851442261786, + "grad_norm": 0.1396484375, + "learning_rate": 0.0010644431205738202, + "loss": 0.0723, + "step": 49696 + }, + { + "epoch": 0.43139382470638277, + "grad_norm": 0.73046875, + "learning_rate": 0.0010644133813730442, + "loss": 0.1211, + "step": 49697 + }, + { + "epoch": 0.4314025051865869, + "grad_norm": 0.1875, + "learning_rate": 0.001064383642211185, + "loss": 0.1006, + "step": 49698 + }, + { + "epoch": 0.4314111856667911, + "grad_norm": 0.1455078125, + "learning_rate": 0.001064353903088275, + "loss": 0.0737, + "step": 49699 + }, + { + "epoch": 0.43141986614699523, + "grad_norm": 0.162109375, + "learning_rate": 0.0010643241640043474, + "loss": 0.0967, + "step": 49700 + }, + { + "epoch": 0.4314285466271994, + "grad_norm": 0.287109375, + "learning_rate": 0.001064294424959434, + "loss": 0.1172, + "step": 49701 + }, + { + "epoch": 0.43143722710740356, + "grad_norm": 0.11474609375, + "learning_rate": 0.0010642646859535683, + "loss": 0.1084, + "step": 49702 + }, + { + "epoch": 0.43144590758760776, + "grad_norm": 0.8046875, + "learning_rate": 0.001064234946986782, + "loss": 0.1406, + "step": 49703 + }, + { + "epoch": 0.4314545880678119, + "grad_norm": 0.1337890625, + "learning_rate": 0.0010642052080591078, + "loss": 0.1172, + "step": 49704 + }, + { + "epoch": 0.4314632685480161, + "grad_norm": 0.1455078125, + "learning_rate": 0.001064175469170578, + "loss": 0.1143, + "step": 49705 + }, + { + "epoch": 0.4314719490282202, + "grad_norm": 0.10498046875, + "learning_rate": 0.0010641457303212254, + "loss": 0.0986, + "step": 49706 + }, + { + "epoch": 0.4314806295084244, + "grad_norm": 0.412109375, + "learning_rate": 0.0010641159915110825, + "loss": 0.0991, + "step": 49707 + }, + { + "epoch": 0.43148930998862856, + "grad_norm": 0.357421875, + "learning_rate": 0.0010640862527401818, + "loss": 0.1011, + "step": 49708 + }, + { + "epoch": 0.43149799046883275, + "grad_norm": 0.10791015625, + "learning_rate": 0.001064056514008556, + "loss": 0.1113, + "step": 49709 + }, + { + "epoch": 0.4315066709490369, + "grad_norm": 0.2294921875, + "learning_rate": 0.0010640267753162375, + "loss": 0.0742, + "step": 49710 + }, + { + "epoch": 0.4315153514292411, + "grad_norm": 0.099609375, + "learning_rate": 0.0010639970366632585, + "loss": 0.1128, + "step": 49711 + }, + { + "epoch": 0.4315240319094452, + "grad_norm": 0.11279296875, + "learning_rate": 0.001063967298049652, + "loss": 0.166, + "step": 49712 + }, + { + "epoch": 0.4315327123896494, + "grad_norm": 0.400390625, + "learning_rate": 0.0010639375594754506, + "loss": 0.123, + "step": 49713 + }, + { + "epoch": 0.43154139286985355, + "grad_norm": 0.30078125, + "learning_rate": 0.001063907820940686, + "loss": 0.103, + "step": 49714 + }, + { + "epoch": 0.43155007335005774, + "grad_norm": 0.44921875, + "learning_rate": 0.0010638780824453915, + "loss": 0.0938, + "step": 49715 + }, + { + "epoch": 0.4315587538302619, + "grad_norm": 0.267578125, + "learning_rate": 0.0010638483439895994, + "loss": 0.1104, + "step": 49716 + }, + { + "epoch": 0.43156743431046607, + "grad_norm": 0.341796875, + "learning_rate": 0.0010638186055733421, + "loss": 0.0928, + "step": 49717 + }, + { + "epoch": 0.4315761147906702, + "grad_norm": 0.318359375, + "learning_rate": 0.0010637888671966524, + "loss": 0.1377, + "step": 49718 + }, + { + "epoch": 0.4315847952708744, + "grad_norm": 0.20703125, + "learning_rate": 0.0010637591288595624, + "loss": 0.0874, + "step": 49719 + }, + { + "epoch": 0.43159347575107854, + "grad_norm": 0.380859375, + "learning_rate": 0.0010637293905621052, + "loss": 0.0859, + "step": 49720 + }, + { + "epoch": 0.43160215623128273, + "grad_norm": 0.578125, + "learning_rate": 0.001063699652304313, + "loss": 0.0635, + "step": 49721 + }, + { + "epoch": 0.43161083671148687, + "grad_norm": 0.35546875, + "learning_rate": 0.0010636699140862183, + "loss": 0.0713, + "step": 49722 + }, + { + "epoch": 0.43161951719169106, + "grad_norm": 0.359375, + "learning_rate": 0.0010636401759078534, + "loss": 0.1475, + "step": 49723 + }, + { + "epoch": 0.4316281976718952, + "grad_norm": 0.18359375, + "learning_rate": 0.0010636104377692507, + "loss": 0.1104, + "step": 49724 + }, + { + "epoch": 0.4316368781520994, + "grad_norm": 0.1416015625, + "learning_rate": 0.0010635806996704437, + "loss": 0.083, + "step": 49725 + }, + { + "epoch": 0.4316455586323035, + "grad_norm": 0.62109375, + "learning_rate": 0.0010635509616114642, + "loss": 0.1699, + "step": 49726 + }, + { + "epoch": 0.4316542391125077, + "grad_norm": 0.263671875, + "learning_rate": 0.0010635212235923447, + "loss": 0.1475, + "step": 49727 + }, + { + "epoch": 0.43166291959271186, + "grad_norm": 0.11474609375, + "learning_rate": 0.0010634914856131178, + "loss": 0.0737, + "step": 49728 + }, + { + "epoch": 0.43167160007291605, + "grad_norm": 0.091796875, + "learning_rate": 0.0010634617476738164, + "loss": 0.083, + "step": 49729 + }, + { + "epoch": 0.4316802805531202, + "grad_norm": 0.1328125, + "learning_rate": 0.0010634320097744725, + "loss": 0.1289, + "step": 49730 + }, + { + "epoch": 0.4316889610333244, + "grad_norm": 0.369140625, + "learning_rate": 0.0010634022719151186, + "loss": 0.1074, + "step": 49731 + }, + { + "epoch": 0.4316976415135285, + "grad_norm": 0.09912109375, + "learning_rate": 0.001063372534095788, + "loss": 0.0811, + "step": 49732 + }, + { + "epoch": 0.4317063219937327, + "grad_norm": 0.1630859375, + "learning_rate": 0.0010633427963165118, + "loss": 0.0854, + "step": 49733 + }, + { + "epoch": 0.43171500247393685, + "grad_norm": 0.2099609375, + "learning_rate": 0.0010633130585773237, + "loss": 0.1006, + "step": 49734 + }, + { + "epoch": 0.43172368295414104, + "grad_norm": 0.302734375, + "learning_rate": 0.0010632833208782562, + "loss": 0.0952, + "step": 49735 + }, + { + "epoch": 0.4317323634343452, + "grad_norm": 0.55859375, + "learning_rate": 0.001063253583219341, + "loss": 0.0889, + "step": 49736 + }, + { + "epoch": 0.43174104391454937, + "grad_norm": 0.1083984375, + "learning_rate": 0.0010632238456006113, + "loss": 0.0986, + "step": 49737 + }, + { + "epoch": 0.4317497243947535, + "grad_norm": 0.158203125, + "learning_rate": 0.0010631941080220997, + "loss": 0.1426, + "step": 49738 + }, + { + "epoch": 0.4317584048749577, + "grad_norm": 0.341796875, + "learning_rate": 0.0010631643704838386, + "loss": 0.1152, + "step": 49739 + }, + { + "epoch": 0.43176708535516184, + "grad_norm": 0.087890625, + "learning_rate": 0.00106313463298586, + "loss": 0.0811, + "step": 49740 + }, + { + "epoch": 0.43177576583536603, + "grad_norm": 0.291015625, + "learning_rate": 0.001063104895528197, + "loss": 0.0762, + "step": 49741 + }, + { + "epoch": 0.43178444631557017, + "grad_norm": 0.322265625, + "learning_rate": 0.0010630751581108815, + "loss": 0.1504, + "step": 49742 + }, + { + "epoch": 0.43179312679577436, + "grad_norm": 0.5234375, + "learning_rate": 0.001063045420733947, + "loss": 0.0791, + "step": 49743 + }, + { + "epoch": 0.4318018072759785, + "grad_norm": 0.1796875, + "learning_rate": 0.0010630156833974253, + "loss": 0.0801, + "step": 49744 + }, + { + "epoch": 0.4318104877561827, + "grad_norm": 0.193359375, + "learning_rate": 0.001062985946101349, + "loss": 0.0986, + "step": 49745 + }, + { + "epoch": 0.43181916823638683, + "grad_norm": 0.50390625, + "learning_rate": 0.001062956208845751, + "loss": 0.0864, + "step": 49746 + }, + { + "epoch": 0.431827848716591, + "grad_norm": 0.294921875, + "learning_rate": 0.0010629264716306632, + "loss": 0.1094, + "step": 49747 + }, + { + "epoch": 0.43183652919679516, + "grad_norm": 0.3125, + "learning_rate": 0.0010628967344561188, + "loss": 0.125, + "step": 49748 + }, + { + "epoch": 0.43184520967699935, + "grad_norm": 0.240234375, + "learning_rate": 0.0010628669973221495, + "loss": 0.1001, + "step": 49749 + }, + { + "epoch": 0.4318538901572035, + "grad_norm": 0.1005859375, + "learning_rate": 0.0010628372602287885, + "loss": 0.0977, + "step": 49750 + }, + { + "epoch": 0.4318625706374077, + "grad_norm": 0.07275390625, + "learning_rate": 0.001062807523176068, + "loss": 0.0781, + "step": 49751 + }, + { + "epoch": 0.4318712511176118, + "grad_norm": 0.1494140625, + "learning_rate": 0.0010627777861640208, + "loss": 0.0923, + "step": 49752 + }, + { + "epoch": 0.431879931597816, + "grad_norm": 0.1904296875, + "learning_rate": 0.0010627480491926795, + "loss": 0.0957, + "step": 49753 + }, + { + "epoch": 0.43188861207802015, + "grad_norm": 0.15625, + "learning_rate": 0.001062718312262076, + "loss": 0.1172, + "step": 49754 + }, + { + "epoch": 0.43189729255822434, + "grad_norm": 0.060791015625, + "learning_rate": 0.0010626885753722433, + "loss": 0.0674, + "step": 49755 + }, + { + "epoch": 0.4319059730384285, + "grad_norm": 0.32421875, + "learning_rate": 0.001062658838523214, + "loss": 0.0752, + "step": 49756 + }, + { + "epoch": 0.4319146535186327, + "grad_norm": 0.373046875, + "learning_rate": 0.0010626291017150204, + "loss": 0.0645, + "step": 49757 + }, + { + "epoch": 0.4319233339988368, + "grad_norm": 0.333984375, + "learning_rate": 0.001062599364947695, + "loss": 0.1123, + "step": 49758 + }, + { + "epoch": 0.431932014479041, + "grad_norm": 0.6484375, + "learning_rate": 0.00106256962822127, + "loss": 0.0859, + "step": 49759 + }, + { + "epoch": 0.43194069495924514, + "grad_norm": 0.80859375, + "learning_rate": 0.0010625398915357785, + "loss": 0.0664, + "step": 49760 + }, + { + "epoch": 0.43194937543944933, + "grad_norm": 0.12451171875, + "learning_rate": 0.0010625101548912528, + "loss": 0.082, + "step": 49761 + }, + { + "epoch": 0.43195805591965347, + "grad_norm": 0.1259765625, + "learning_rate": 0.0010624804182877256, + "loss": 0.0952, + "step": 49762 + }, + { + "epoch": 0.43196673639985766, + "grad_norm": 0.2236328125, + "learning_rate": 0.0010624506817252294, + "loss": 0.0859, + "step": 49763 + }, + { + "epoch": 0.4319754168800618, + "grad_norm": 0.1181640625, + "learning_rate": 0.0010624209452037962, + "loss": 0.1187, + "step": 49764 + }, + { + "epoch": 0.431984097360266, + "grad_norm": 0.1982421875, + "learning_rate": 0.001062391208723459, + "loss": 0.0781, + "step": 49765 + }, + { + "epoch": 0.43199277784047013, + "grad_norm": 0.318359375, + "learning_rate": 0.0010623614722842503, + "loss": 0.0967, + "step": 49766 + }, + { + "epoch": 0.4320014583206743, + "grad_norm": 0.103515625, + "learning_rate": 0.0010623317358862026, + "loss": 0.0908, + "step": 49767 + }, + { + "epoch": 0.43201013880087846, + "grad_norm": 0.3359375, + "learning_rate": 0.0010623019995293478, + "loss": 0.0854, + "step": 49768 + }, + { + "epoch": 0.43201881928108266, + "grad_norm": 0.0693359375, + "learning_rate": 0.0010622722632137197, + "loss": 0.0913, + "step": 49769 + }, + { + "epoch": 0.4320274997612868, + "grad_norm": 0.455078125, + "learning_rate": 0.0010622425269393497, + "loss": 0.1074, + "step": 49770 + }, + { + "epoch": 0.432036180241491, + "grad_norm": 0.146484375, + "learning_rate": 0.001062212790706271, + "loss": 0.085, + "step": 49771 + }, + { + "epoch": 0.4320448607216951, + "grad_norm": 0.35546875, + "learning_rate": 0.0010621830545145155, + "loss": 0.0947, + "step": 49772 + }, + { + "epoch": 0.4320535412018993, + "grad_norm": 0.13671875, + "learning_rate": 0.0010621533183641158, + "loss": 0.0952, + "step": 49773 + }, + { + "epoch": 0.43206222168210345, + "grad_norm": 0.380859375, + "learning_rate": 0.0010621235822551051, + "loss": 0.1211, + "step": 49774 + }, + { + "epoch": 0.43207090216230765, + "grad_norm": 0.330078125, + "learning_rate": 0.0010620938461875155, + "loss": 0.0957, + "step": 49775 + }, + { + "epoch": 0.4320795826425118, + "grad_norm": 0.28515625, + "learning_rate": 0.001062064110161379, + "loss": 0.0845, + "step": 49776 + }, + { + "epoch": 0.4320882631227159, + "grad_norm": 0.251953125, + "learning_rate": 0.0010620343741767294, + "loss": 0.082, + "step": 49777 + }, + { + "epoch": 0.4320969436029201, + "grad_norm": 0.447265625, + "learning_rate": 0.001062004638233598, + "loss": 0.1055, + "step": 49778 + }, + { + "epoch": 0.43210562408312425, + "grad_norm": 0.421875, + "learning_rate": 0.0010619749023320177, + "loss": 0.126, + "step": 49779 + }, + { + "epoch": 0.43211430456332844, + "grad_norm": 0.26953125, + "learning_rate": 0.0010619451664720212, + "loss": 0.0894, + "step": 49780 + }, + { + "epoch": 0.4321229850435326, + "grad_norm": 0.2109375, + "learning_rate": 0.001061915430653641, + "loss": 0.0654, + "step": 49781 + }, + { + "epoch": 0.4321316655237368, + "grad_norm": 0.1953125, + "learning_rate": 0.0010618856948769093, + "loss": 0.1162, + "step": 49782 + }, + { + "epoch": 0.4321403460039409, + "grad_norm": 0.80078125, + "learning_rate": 0.001061855959141859, + "loss": 0.0898, + "step": 49783 + }, + { + "epoch": 0.4321490264841451, + "grad_norm": 0.11669921875, + "learning_rate": 0.0010618262234485226, + "loss": 0.082, + "step": 49784 + }, + { + "epoch": 0.43215770696434924, + "grad_norm": 0.36328125, + "learning_rate": 0.0010617964877969318, + "loss": 0.1006, + "step": 49785 + }, + { + "epoch": 0.43216638744455343, + "grad_norm": 0.3515625, + "learning_rate": 0.0010617667521871201, + "loss": 0.1543, + "step": 49786 + }, + { + "epoch": 0.43217506792475757, + "grad_norm": 0.2353515625, + "learning_rate": 0.00106173701661912, + "loss": 0.1309, + "step": 49787 + }, + { + "epoch": 0.43218374840496177, + "grad_norm": 0.4375, + "learning_rate": 0.0010617072810929639, + "loss": 0.0986, + "step": 49788 + }, + { + "epoch": 0.4321924288851659, + "grad_norm": 0.3359375, + "learning_rate": 0.0010616775456086835, + "loss": 0.1001, + "step": 49789 + }, + { + "epoch": 0.4322011093653701, + "grad_norm": 0.208984375, + "learning_rate": 0.0010616478101663125, + "loss": 0.106, + "step": 49790 + }, + { + "epoch": 0.43220978984557423, + "grad_norm": 0.19140625, + "learning_rate": 0.0010616180747658826, + "loss": 0.1211, + "step": 49791 + }, + { + "epoch": 0.4322184703257784, + "grad_norm": 0.57421875, + "learning_rate": 0.0010615883394074267, + "loss": 0.1211, + "step": 49792 + }, + { + "epoch": 0.43222715080598256, + "grad_norm": 0.1572265625, + "learning_rate": 0.001061558604090977, + "loss": 0.1191, + "step": 49793 + }, + { + "epoch": 0.43223583128618676, + "grad_norm": 0.46875, + "learning_rate": 0.0010615288688165662, + "loss": 0.0835, + "step": 49794 + }, + { + "epoch": 0.4322445117663909, + "grad_norm": 0.81640625, + "learning_rate": 0.001061499133584227, + "loss": 0.0874, + "step": 49795 + }, + { + "epoch": 0.4322531922465951, + "grad_norm": 0.33984375, + "learning_rate": 0.0010614693983939918, + "loss": 0.0903, + "step": 49796 + }, + { + "epoch": 0.4322618727267992, + "grad_norm": 0.1484375, + "learning_rate": 0.001061439663245893, + "loss": 0.0898, + "step": 49797 + }, + { + "epoch": 0.4322705532070034, + "grad_norm": 0.45703125, + "learning_rate": 0.0010614099281399637, + "loss": 0.0986, + "step": 49798 + }, + { + "epoch": 0.43227923368720755, + "grad_norm": 0.265625, + "learning_rate": 0.0010613801930762356, + "loss": 0.0918, + "step": 49799 + }, + { + "epoch": 0.43228791416741175, + "grad_norm": 0.232421875, + "learning_rate": 0.0010613504580547412, + "loss": 0.127, + "step": 49800 + }, + { + "epoch": 0.4322965946476159, + "grad_norm": 0.083984375, + "learning_rate": 0.0010613207230755135, + "loss": 0.0898, + "step": 49801 + }, + { + "epoch": 0.4323052751278201, + "grad_norm": 0.1640625, + "learning_rate": 0.0010612909881385847, + "loss": 0.0918, + "step": 49802 + }, + { + "epoch": 0.4323139556080242, + "grad_norm": 0.2490234375, + "learning_rate": 0.001061261253243988, + "loss": 0.0869, + "step": 49803 + }, + { + "epoch": 0.4323226360882284, + "grad_norm": 0.251953125, + "learning_rate": 0.0010612315183917552, + "loss": 0.1011, + "step": 49804 + }, + { + "epoch": 0.43233131656843254, + "grad_norm": 0.25390625, + "learning_rate": 0.001061201783581919, + "loss": 0.1094, + "step": 49805 + }, + { + "epoch": 0.43233999704863674, + "grad_norm": 0.11376953125, + "learning_rate": 0.001061172048814512, + "loss": 0.0947, + "step": 49806 + }, + { + "epoch": 0.4323486775288409, + "grad_norm": 0.055908203125, + "learning_rate": 0.0010611423140895666, + "loss": 0.084, + "step": 49807 + }, + { + "epoch": 0.43235735800904507, + "grad_norm": 0.2099609375, + "learning_rate": 0.0010611125794071157, + "loss": 0.0928, + "step": 49808 + }, + { + "epoch": 0.4323660384892492, + "grad_norm": 0.8828125, + "learning_rate": 0.0010610828447671912, + "loss": 0.1245, + "step": 49809 + }, + { + "epoch": 0.4323747189694534, + "grad_norm": 0.498046875, + "learning_rate": 0.0010610531101698257, + "loss": 0.082, + "step": 49810 + }, + { + "epoch": 0.43238339944965754, + "grad_norm": 0.2216796875, + "learning_rate": 0.0010610233756150522, + "loss": 0.1182, + "step": 49811 + }, + { + "epoch": 0.43239207992986173, + "grad_norm": 0.3125, + "learning_rate": 0.0010609936411029028, + "loss": 0.1123, + "step": 49812 + }, + { + "epoch": 0.43240076041006587, + "grad_norm": 0.52734375, + "learning_rate": 0.0010609639066334103, + "loss": 0.1406, + "step": 49813 + }, + { + "epoch": 0.43240944089027006, + "grad_norm": 0.2060546875, + "learning_rate": 0.001060934172206607, + "loss": 0.0947, + "step": 49814 + }, + { + "epoch": 0.4324181213704742, + "grad_norm": 0.10400390625, + "learning_rate": 0.0010609044378225257, + "loss": 0.0728, + "step": 49815 + }, + { + "epoch": 0.4324268018506784, + "grad_norm": 0.291015625, + "learning_rate": 0.0010608747034811987, + "loss": 0.0908, + "step": 49816 + }, + { + "epoch": 0.4324354823308825, + "grad_norm": 0.46484375, + "learning_rate": 0.0010608449691826584, + "loss": 0.1543, + "step": 49817 + }, + { + "epoch": 0.4324441628110867, + "grad_norm": 0.1962890625, + "learning_rate": 0.0010608152349269376, + "loss": 0.1133, + "step": 49818 + }, + { + "epoch": 0.43245284329129086, + "grad_norm": 0.330078125, + "learning_rate": 0.0010607855007140686, + "loss": 0.1328, + "step": 49819 + }, + { + "epoch": 0.43246152377149505, + "grad_norm": 0.2412109375, + "learning_rate": 0.0010607557665440837, + "loss": 0.0996, + "step": 49820 + }, + { + "epoch": 0.4324702042516992, + "grad_norm": 0.73828125, + "learning_rate": 0.0010607260324170158, + "loss": 0.0884, + "step": 49821 + }, + { + "epoch": 0.4324788847319034, + "grad_norm": 0.10888671875, + "learning_rate": 0.0010606962983328973, + "loss": 0.1055, + "step": 49822 + }, + { + "epoch": 0.4324875652121075, + "grad_norm": 0.6640625, + "learning_rate": 0.001060666564291761, + "loss": 0.0654, + "step": 49823 + }, + { + "epoch": 0.4324962456923117, + "grad_norm": 0.40625, + "learning_rate": 0.0010606368302936392, + "loss": 0.1104, + "step": 49824 + }, + { + "epoch": 0.43250492617251585, + "grad_norm": 0.1123046875, + "learning_rate": 0.0010606070963385644, + "loss": 0.1416, + "step": 49825 + }, + { + "epoch": 0.43251360665272004, + "grad_norm": 0.140625, + "learning_rate": 0.0010605773624265688, + "loss": 0.0908, + "step": 49826 + }, + { + "epoch": 0.4325222871329242, + "grad_norm": 0.57421875, + "learning_rate": 0.0010605476285576853, + "loss": 0.1484, + "step": 49827 + }, + { + "epoch": 0.43253096761312837, + "grad_norm": 0.240234375, + "learning_rate": 0.0010605178947319463, + "loss": 0.0791, + "step": 49828 + }, + { + "epoch": 0.4325396480933325, + "grad_norm": 0.11865234375, + "learning_rate": 0.0010604881609493843, + "loss": 0.1113, + "step": 49829 + }, + { + "epoch": 0.4325483285735367, + "grad_norm": 0.2080078125, + "learning_rate": 0.001060458427210032, + "loss": 0.1006, + "step": 49830 + }, + { + "epoch": 0.43255700905374084, + "grad_norm": 0.1533203125, + "learning_rate": 0.0010604286935139218, + "loss": 0.0889, + "step": 49831 + }, + { + "epoch": 0.43256568953394503, + "grad_norm": 0.158203125, + "learning_rate": 0.001060398959861086, + "loss": 0.0801, + "step": 49832 + }, + { + "epoch": 0.43257437001414917, + "grad_norm": 0.08837890625, + "learning_rate": 0.0010603692262515574, + "loss": 0.0835, + "step": 49833 + }, + { + "epoch": 0.43258305049435336, + "grad_norm": 0.48046875, + "learning_rate": 0.0010603394926853684, + "loss": 0.0991, + "step": 49834 + }, + { + "epoch": 0.4325917309745575, + "grad_norm": 0.056884765625, + "learning_rate": 0.0010603097591625518, + "loss": 0.0884, + "step": 49835 + }, + { + "epoch": 0.4326004114547617, + "grad_norm": 0.259765625, + "learning_rate": 0.0010602800256831395, + "loss": 0.0835, + "step": 49836 + }, + { + "epoch": 0.43260909193496583, + "grad_norm": 0.296875, + "learning_rate": 0.0010602502922471642, + "loss": 0.083, + "step": 49837 + }, + { + "epoch": 0.43261777241517, + "grad_norm": 0.1201171875, + "learning_rate": 0.0010602205588546587, + "loss": 0.0913, + "step": 49838 + }, + { + "epoch": 0.43262645289537416, + "grad_norm": 0.96875, + "learning_rate": 0.0010601908255056555, + "loss": 0.0811, + "step": 49839 + }, + { + "epoch": 0.43263513337557835, + "grad_norm": 0.1455078125, + "learning_rate": 0.0010601610922001874, + "loss": 0.0684, + "step": 49840 + }, + { + "epoch": 0.4326438138557825, + "grad_norm": 0.62109375, + "learning_rate": 0.001060131358938286, + "loss": 0.0938, + "step": 49841 + }, + { + "epoch": 0.4326524943359867, + "grad_norm": 0.3671875, + "learning_rate": 0.0010601016257199844, + "loss": 0.126, + "step": 49842 + }, + { + "epoch": 0.4326611748161908, + "grad_norm": 0.369140625, + "learning_rate": 0.0010600718925453155, + "loss": 0.0972, + "step": 49843 + }, + { + "epoch": 0.432669855296395, + "grad_norm": 0.21875, + "learning_rate": 0.001060042159414311, + "loss": 0.0991, + "step": 49844 + }, + { + "epoch": 0.43267853577659915, + "grad_norm": 0.11279296875, + "learning_rate": 0.0010600124263270036, + "loss": 0.1128, + "step": 49845 + }, + { + "epoch": 0.43268721625680334, + "grad_norm": 0.32421875, + "learning_rate": 0.0010599826932834264, + "loss": 0.1123, + "step": 49846 + }, + { + "epoch": 0.4326958967370075, + "grad_norm": 0.09716796875, + "learning_rate": 0.0010599529602836114, + "loss": 0.0767, + "step": 49847 + }, + { + "epoch": 0.43270457721721167, + "grad_norm": 0.474609375, + "learning_rate": 0.0010599232273275914, + "loss": 0.0884, + "step": 49848 + }, + { + "epoch": 0.4327132576974158, + "grad_norm": 0.08935546875, + "learning_rate": 0.0010598934944153987, + "loss": 0.1084, + "step": 49849 + }, + { + "epoch": 0.43272193817762, + "grad_norm": 0.1171875, + "learning_rate": 0.0010598637615470655, + "loss": 0.0649, + "step": 49850 + }, + { + "epoch": 0.43273061865782414, + "grad_norm": 0.322265625, + "learning_rate": 0.001059834028722625, + "loss": 0.0728, + "step": 49851 + }, + { + "epoch": 0.43273929913802833, + "grad_norm": 0.1142578125, + "learning_rate": 0.0010598042959421095, + "loss": 0.0664, + "step": 49852 + }, + { + "epoch": 0.43274797961823247, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010597745632055513, + "loss": 0.1279, + "step": 49853 + }, + { + "epoch": 0.43275666009843666, + "grad_norm": 0.11376953125, + "learning_rate": 0.0010597448305129827, + "loss": 0.0908, + "step": 49854 + }, + { + "epoch": 0.4327653405786408, + "grad_norm": 0.498046875, + "learning_rate": 0.001059715097864437, + "loss": 0.0898, + "step": 49855 + }, + { + "epoch": 0.432774021058845, + "grad_norm": 0.11181640625, + "learning_rate": 0.001059685365259946, + "loss": 0.1025, + "step": 49856 + }, + { + "epoch": 0.43278270153904913, + "grad_norm": 0.1904296875, + "learning_rate": 0.0010596556326995428, + "loss": 0.083, + "step": 49857 + }, + { + "epoch": 0.4327913820192533, + "grad_norm": 0.2060546875, + "learning_rate": 0.0010596259001832596, + "loss": 0.0894, + "step": 49858 + }, + { + "epoch": 0.43280006249945746, + "grad_norm": 0.2431640625, + "learning_rate": 0.0010595961677111283, + "loss": 0.1104, + "step": 49859 + }, + { + "epoch": 0.43280874297966165, + "grad_norm": 0.2490234375, + "learning_rate": 0.0010595664352831826, + "loss": 0.1299, + "step": 49860 + }, + { + "epoch": 0.4328174234598658, + "grad_norm": 0.470703125, + "learning_rate": 0.0010595367028994542, + "loss": 0.1216, + "step": 49861 + }, + { + "epoch": 0.43282610394007, + "grad_norm": 0.072265625, + "learning_rate": 0.0010595069705599763, + "loss": 0.0703, + "step": 49862 + }, + { + "epoch": 0.4328347844202741, + "grad_norm": 0.5546875, + "learning_rate": 0.0010594772382647802, + "loss": 0.0723, + "step": 49863 + }, + { + "epoch": 0.4328434649004783, + "grad_norm": 0.07470703125, + "learning_rate": 0.0010594475060138994, + "loss": 0.0957, + "step": 49864 + }, + { + "epoch": 0.43285214538068245, + "grad_norm": 0.47265625, + "learning_rate": 0.0010594177738073665, + "loss": 0.1133, + "step": 49865 + }, + { + "epoch": 0.43286082586088664, + "grad_norm": 2.0625, + "learning_rate": 0.0010593880416452138, + "loss": 0.2949, + "step": 49866 + }, + { + "epoch": 0.4328695063410908, + "grad_norm": 0.19140625, + "learning_rate": 0.0010593583095274736, + "loss": 0.1006, + "step": 49867 + }, + { + "epoch": 0.432878186821295, + "grad_norm": 0.431640625, + "learning_rate": 0.0010593285774541782, + "loss": 0.123, + "step": 49868 + }, + { + "epoch": 0.4328868673014991, + "grad_norm": 0.44921875, + "learning_rate": 0.0010592988454253608, + "loss": 0.1079, + "step": 49869 + }, + { + "epoch": 0.4328955477817033, + "grad_norm": 0.2333984375, + "learning_rate": 0.0010592691134410535, + "loss": 0.0957, + "step": 49870 + }, + { + "epoch": 0.43290422826190744, + "grad_norm": 0.103515625, + "learning_rate": 0.0010592393815012887, + "loss": 0.0771, + "step": 49871 + }, + { + "epoch": 0.43291290874211164, + "grad_norm": 0.10498046875, + "learning_rate": 0.0010592096496060991, + "loss": 0.0762, + "step": 49872 + }, + { + "epoch": 0.4329215892223158, + "grad_norm": 0.2314453125, + "learning_rate": 0.0010591799177555176, + "loss": 0.1113, + "step": 49873 + }, + { + "epoch": 0.43293026970251997, + "grad_norm": 0.29296875, + "learning_rate": 0.0010591501859495763, + "loss": 0.0869, + "step": 49874 + }, + { + "epoch": 0.4329389501827241, + "grad_norm": 0.30078125, + "learning_rate": 0.0010591204541883074, + "loss": 0.0952, + "step": 49875 + }, + { + "epoch": 0.4329476306629283, + "grad_norm": 0.263671875, + "learning_rate": 0.0010590907224717438, + "loss": 0.1602, + "step": 49876 + }, + { + "epoch": 0.43295631114313243, + "grad_norm": 0.515625, + "learning_rate": 0.0010590609907999181, + "loss": 0.0903, + "step": 49877 + }, + { + "epoch": 0.4329649916233366, + "grad_norm": 0.396484375, + "learning_rate": 0.0010590312591728628, + "loss": 0.0859, + "step": 49878 + }, + { + "epoch": 0.43297367210354076, + "grad_norm": 0.173828125, + "learning_rate": 0.0010590015275906103, + "loss": 0.1172, + "step": 49879 + }, + { + "epoch": 0.43298235258374496, + "grad_norm": 0.1552734375, + "learning_rate": 0.0010589717960531927, + "loss": 0.0908, + "step": 49880 + }, + { + "epoch": 0.4329910330639491, + "grad_norm": 0.796875, + "learning_rate": 0.001058942064560643, + "loss": 0.1226, + "step": 49881 + }, + { + "epoch": 0.4329997135441533, + "grad_norm": 0.197265625, + "learning_rate": 0.001058912333112994, + "loss": 0.0664, + "step": 49882 + }, + { + "epoch": 0.4330083940243574, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010588826017102777, + "loss": 0.083, + "step": 49883 + }, + { + "epoch": 0.4330170745045616, + "grad_norm": 0.1044921875, + "learning_rate": 0.001058852870352527, + "loss": 0.0938, + "step": 49884 + }, + { + "epoch": 0.43302575498476575, + "grad_norm": 0.349609375, + "learning_rate": 0.001058823139039774, + "loss": 0.1162, + "step": 49885 + }, + { + "epoch": 0.43303443546496995, + "grad_norm": 0.1826171875, + "learning_rate": 0.0010587934077720513, + "loss": 0.1089, + "step": 49886 + }, + { + "epoch": 0.4330431159451741, + "grad_norm": 0.7421875, + "learning_rate": 0.0010587636765493915, + "loss": 0.1089, + "step": 49887 + }, + { + "epoch": 0.4330517964253783, + "grad_norm": 0.10546875, + "learning_rate": 0.0010587339453718271, + "loss": 0.0815, + "step": 49888 + }, + { + "epoch": 0.4330604769055824, + "grad_norm": 0.515625, + "learning_rate": 0.0010587042142393907, + "loss": 0.0996, + "step": 49889 + }, + { + "epoch": 0.4330691573857866, + "grad_norm": 0.462890625, + "learning_rate": 0.0010586744831521146, + "loss": 0.1118, + "step": 49890 + }, + { + "epoch": 0.43307783786599074, + "grad_norm": 0.1708984375, + "learning_rate": 0.0010586447521100317, + "loss": 0.0962, + "step": 49891 + }, + { + "epoch": 0.43308651834619494, + "grad_norm": 0.166015625, + "learning_rate": 0.0010586150211131745, + "loss": 0.0728, + "step": 49892 + }, + { + "epoch": 0.4330951988263991, + "grad_norm": 0.328125, + "learning_rate": 0.0010585852901615748, + "loss": 0.0825, + "step": 49893 + }, + { + "epoch": 0.43310387930660327, + "grad_norm": 0.29296875, + "learning_rate": 0.001058555559255266, + "loss": 0.0845, + "step": 49894 + }, + { + "epoch": 0.4331125597868074, + "grad_norm": 0.146484375, + "learning_rate": 0.00105852582839428, + "loss": 0.083, + "step": 49895 + }, + { + "epoch": 0.4331212402670116, + "grad_norm": 0.119140625, + "learning_rate": 0.0010584960975786497, + "loss": 0.0884, + "step": 49896 + }, + { + "epoch": 0.43312992074721574, + "grad_norm": 0.103515625, + "learning_rate": 0.0010584663668084069, + "loss": 0.0588, + "step": 49897 + }, + { + "epoch": 0.43313860122741993, + "grad_norm": 0.2890625, + "learning_rate": 0.0010584366360835854, + "loss": 0.1035, + "step": 49898 + }, + { + "epoch": 0.43314728170762407, + "grad_norm": 0.1376953125, + "learning_rate": 0.0010584069054042162, + "loss": 0.0981, + "step": 49899 + }, + { + "epoch": 0.4331559621878282, + "grad_norm": 0.17578125, + "learning_rate": 0.0010583771747703333, + "loss": 0.1318, + "step": 49900 + }, + { + "epoch": 0.4331646426680324, + "grad_norm": 0.125, + "learning_rate": 0.0010583474441819682, + "loss": 0.1123, + "step": 49901 + }, + { + "epoch": 0.43317332314823653, + "grad_norm": 0.1611328125, + "learning_rate": 0.0010583177136391537, + "loss": 0.1147, + "step": 49902 + }, + { + "epoch": 0.4331820036284407, + "grad_norm": 0.38671875, + "learning_rate": 0.0010582879831419225, + "loss": 0.083, + "step": 49903 + }, + { + "epoch": 0.43319068410864486, + "grad_norm": 0.2314453125, + "learning_rate": 0.001058258252690307, + "loss": 0.0776, + "step": 49904 + }, + { + "epoch": 0.43319936458884906, + "grad_norm": 0.462890625, + "learning_rate": 0.0010582285222843393, + "loss": 0.0903, + "step": 49905 + }, + { + "epoch": 0.4332080450690532, + "grad_norm": 0.2578125, + "learning_rate": 0.0010581987919240522, + "loss": 0.0898, + "step": 49906 + }, + { + "epoch": 0.4332167255492574, + "grad_norm": 0.228515625, + "learning_rate": 0.0010581690616094784, + "loss": 0.0752, + "step": 49907 + }, + { + "epoch": 0.4332254060294615, + "grad_norm": 0.23046875, + "learning_rate": 0.0010581393313406505, + "loss": 0.0874, + "step": 49908 + }, + { + "epoch": 0.4332340865096657, + "grad_norm": 0.26171875, + "learning_rate": 0.0010581096011176004, + "loss": 0.0781, + "step": 49909 + }, + { + "epoch": 0.43324276698986985, + "grad_norm": 0.5234375, + "learning_rate": 0.0010580798709403615, + "loss": 0.0928, + "step": 49910 + }, + { + "epoch": 0.43325144747007405, + "grad_norm": 0.083984375, + "learning_rate": 0.0010580501408089654, + "loss": 0.1055, + "step": 49911 + }, + { + "epoch": 0.4332601279502782, + "grad_norm": 0.515625, + "learning_rate": 0.0010580204107234455, + "loss": 0.085, + "step": 49912 + }, + { + "epoch": 0.4332688084304824, + "grad_norm": 0.11181640625, + "learning_rate": 0.0010579906806838335, + "loss": 0.0869, + "step": 49913 + }, + { + "epoch": 0.4332774889106865, + "grad_norm": 0.41796875, + "learning_rate": 0.0010579609506901624, + "loss": 0.1143, + "step": 49914 + }, + { + "epoch": 0.4332861693908907, + "grad_norm": 0.1328125, + "learning_rate": 0.001057931220742464, + "loss": 0.1006, + "step": 49915 + }, + { + "epoch": 0.43329484987109484, + "grad_norm": 0.71875, + "learning_rate": 0.001057901490840772, + "loss": 0.1025, + "step": 49916 + }, + { + "epoch": 0.43330353035129904, + "grad_norm": 0.12060546875, + "learning_rate": 0.0010578717609851182, + "loss": 0.1279, + "step": 49917 + }, + { + "epoch": 0.4333122108315032, + "grad_norm": 0.376953125, + "learning_rate": 0.0010578420311755353, + "loss": 0.0811, + "step": 49918 + }, + { + "epoch": 0.43332089131170737, + "grad_norm": 0.17578125, + "learning_rate": 0.0010578123014120555, + "loss": 0.0957, + "step": 49919 + }, + { + "epoch": 0.4333295717919115, + "grad_norm": 0.197265625, + "learning_rate": 0.001057782571694712, + "loss": 0.1079, + "step": 49920 + }, + { + "epoch": 0.4333382522721157, + "grad_norm": 0.0810546875, + "learning_rate": 0.0010577528420235363, + "loss": 0.1348, + "step": 49921 + }, + { + "epoch": 0.43334693275231984, + "grad_norm": 0.482421875, + "learning_rate": 0.0010577231123985618, + "loss": 0.0879, + "step": 49922 + }, + { + "epoch": 0.43335561323252403, + "grad_norm": 0.1748046875, + "learning_rate": 0.00105769338281982, + "loss": 0.1084, + "step": 49923 + }, + { + "epoch": 0.43336429371272817, + "grad_norm": 0.1064453125, + "learning_rate": 0.0010576636532873447, + "loss": 0.1191, + "step": 49924 + }, + { + "epoch": 0.43337297419293236, + "grad_norm": 0.10498046875, + "learning_rate": 0.0010576339238011678, + "loss": 0.0967, + "step": 49925 + }, + { + "epoch": 0.4333816546731365, + "grad_norm": 0.25390625, + "learning_rate": 0.0010576041943613217, + "loss": 0.0981, + "step": 49926 + }, + { + "epoch": 0.4333903351533407, + "grad_norm": 0.07275390625, + "learning_rate": 0.001057574464967839, + "loss": 0.0884, + "step": 49927 + }, + { + "epoch": 0.4333990156335448, + "grad_norm": 0.140625, + "learning_rate": 0.0010575447356207521, + "loss": 0.0986, + "step": 49928 + }, + { + "epoch": 0.433407696113749, + "grad_norm": 0.107421875, + "learning_rate": 0.001057515006320094, + "loss": 0.0742, + "step": 49929 + }, + { + "epoch": 0.43341637659395316, + "grad_norm": 0.2109375, + "learning_rate": 0.0010574852770658966, + "loss": 0.1289, + "step": 49930 + }, + { + "epoch": 0.43342505707415735, + "grad_norm": 0.2275390625, + "learning_rate": 0.001057455547858193, + "loss": 0.0537, + "step": 49931 + }, + { + "epoch": 0.4334337375543615, + "grad_norm": 0.130859375, + "learning_rate": 0.0010574258186970146, + "loss": 0.0928, + "step": 49932 + }, + { + "epoch": 0.4334424180345657, + "grad_norm": 0.26953125, + "learning_rate": 0.001057396089582395, + "loss": 0.1162, + "step": 49933 + }, + { + "epoch": 0.4334510985147698, + "grad_norm": 0.1435546875, + "learning_rate": 0.001057366360514367, + "loss": 0.0947, + "step": 49934 + }, + { + "epoch": 0.433459778994974, + "grad_norm": 0.67578125, + "learning_rate": 0.0010573366314929617, + "loss": 0.1094, + "step": 49935 + }, + { + "epoch": 0.43346845947517815, + "grad_norm": 0.09326171875, + "learning_rate": 0.001057306902518213, + "loss": 0.0962, + "step": 49936 + }, + { + "epoch": 0.43347713995538234, + "grad_norm": 0.486328125, + "learning_rate": 0.0010572771735901526, + "loss": 0.1504, + "step": 49937 + }, + { + "epoch": 0.4334858204355865, + "grad_norm": 0.380859375, + "learning_rate": 0.0010572474447088133, + "loss": 0.0903, + "step": 49938 + }, + { + "epoch": 0.43349450091579067, + "grad_norm": 0.1357421875, + "learning_rate": 0.0010572177158742276, + "loss": 0.0811, + "step": 49939 + }, + { + "epoch": 0.4335031813959948, + "grad_norm": 0.0859375, + "learning_rate": 0.0010571879870864278, + "loss": 0.0879, + "step": 49940 + }, + { + "epoch": 0.433511861876199, + "grad_norm": 0.1435546875, + "learning_rate": 0.0010571582583454463, + "loss": 0.0957, + "step": 49941 + }, + { + "epoch": 0.43352054235640314, + "grad_norm": 0.380859375, + "learning_rate": 0.0010571285296513161, + "loss": 0.1338, + "step": 49942 + }, + { + "epoch": 0.43352922283660733, + "grad_norm": 0.435546875, + "learning_rate": 0.0010570988010040697, + "loss": 0.1138, + "step": 49943 + }, + { + "epoch": 0.43353790331681147, + "grad_norm": 0.400390625, + "learning_rate": 0.0010570690724037396, + "loss": 0.1367, + "step": 49944 + }, + { + "epoch": 0.43354658379701566, + "grad_norm": 0.1015625, + "learning_rate": 0.0010570393438503578, + "loss": 0.0962, + "step": 49945 + }, + { + "epoch": 0.4335552642772198, + "grad_norm": 0.36328125, + "learning_rate": 0.001057009615343957, + "loss": 0.1089, + "step": 49946 + }, + { + "epoch": 0.433563944757424, + "grad_norm": 0.1494140625, + "learning_rate": 0.0010569798868845701, + "loss": 0.0986, + "step": 49947 + }, + { + "epoch": 0.43357262523762813, + "grad_norm": 0.1884765625, + "learning_rate": 0.0010569501584722291, + "loss": 0.1045, + "step": 49948 + }, + { + "epoch": 0.4335813057178323, + "grad_norm": 0.435546875, + "learning_rate": 0.0010569204301069668, + "loss": 0.0898, + "step": 49949 + }, + { + "epoch": 0.43358998619803646, + "grad_norm": 0.341796875, + "learning_rate": 0.0010568907017888159, + "loss": 0.1064, + "step": 49950 + }, + { + "epoch": 0.43359866667824065, + "grad_norm": 0.330078125, + "learning_rate": 0.0010568609735178084, + "loss": 0.1167, + "step": 49951 + }, + { + "epoch": 0.4336073471584448, + "grad_norm": 0.1201171875, + "learning_rate": 0.0010568312452939775, + "loss": 0.1025, + "step": 49952 + }, + { + "epoch": 0.433616027638649, + "grad_norm": 0.1474609375, + "learning_rate": 0.001056801517117355, + "loss": 0.0874, + "step": 49953 + }, + { + "epoch": 0.4336247081188531, + "grad_norm": 0.8828125, + "learning_rate": 0.0010567717889879737, + "loss": 0.1367, + "step": 49954 + }, + { + "epoch": 0.4336333885990573, + "grad_norm": 0.5, + "learning_rate": 0.0010567420609058661, + "loss": 0.1406, + "step": 49955 + }, + { + "epoch": 0.43364206907926145, + "grad_norm": 0.2080078125, + "learning_rate": 0.0010567123328710651, + "loss": 0.0894, + "step": 49956 + }, + { + "epoch": 0.43365074955946564, + "grad_norm": 0.1953125, + "learning_rate": 0.0010566826048836024, + "loss": 0.1338, + "step": 49957 + }, + { + "epoch": 0.4336594300396698, + "grad_norm": 0.5390625, + "learning_rate": 0.0010566528769435108, + "loss": 0.0859, + "step": 49958 + }, + { + "epoch": 0.433668110519874, + "grad_norm": 0.078125, + "learning_rate": 0.0010566231490508234, + "loss": 0.1133, + "step": 49959 + }, + { + "epoch": 0.4336767910000781, + "grad_norm": 0.1689453125, + "learning_rate": 0.0010565934212055724, + "loss": 0.1104, + "step": 49960 + }, + { + "epoch": 0.4336854714802823, + "grad_norm": 0.1953125, + "learning_rate": 0.00105656369340779, + "loss": 0.1035, + "step": 49961 + }, + { + "epoch": 0.43369415196048644, + "grad_norm": 0.275390625, + "learning_rate": 0.0010565339656575087, + "loss": 0.126, + "step": 49962 + }, + { + "epoch": 0.43370283244069063, + "grad_norm": 0.27734375, + "learning_rate": 0.0010565042379547614, + "loss": 0.1045, + "step": 49963 + }, + { + "epoch": 0.43371151292089477, + "grad_norm": 0.1552734375, + "learning_rate": 0.00105647451029958, + "loss": 0.126, + "step": 49964 + }, + { + "epoch": 0.43372019340109896, + "grad_norm": 0.515625, + "learning_rate": 0.0010564447826919982, + "loss": 0.168, + "step": 49965 + }, + { + "epoch": 0.4337288738813031, + "grad_norm": 0.203125, + "learning_rate": 0.0010564150551320471, + "loss": 0.1602, + "step": 49966 + }, + { + "epoch": 0.4337375543615073, + "grad_norm": 0.306640625, + "learning_rate": 0.0010563853276197598, + "loss": 0.1592, + "step": 49967 + }, + { + "epoch": 0.43374623484171143, + "grad_norm": 0.08349609375, + "learning_rate": 0.001056355600155169, + "loss": 0.0835, + "step": 49968 + }, + { + "epoch": 0.4337549153219156, + "grad_norm": 0.1279296875, + "learning_rate": 0.0010563258727383073, + "loss": 0.0947, + "step": 49969 + }, + { + "epoch": 0.43376359580211976, + "grad_norm": 0.6328125, + "learning_rate": 0.0010562961453692069, + "loss": 0.2139, + "step": 49970 + }, + { + "epoch": 0.43377227628232395, + "grad_norm": 0.1533203125, + "learning_rate": 0.0010562664180479005, + "loss": 0.1377, + "step": 49971 + }, + { + "epoch": 0.4337809567625281, + "grad_norm": 0.212890625, + "learning_rate": 0.00105623669077442, + "loss": 0.1113, + "step": 49972 + }, + { + "epoch": 0.4337896372427323, + "grad_norm": 0.66796875, + "learning_rate": 0.001056206963548799, + "loss": 0.1152, + "step": 49973 + }, + { + "epoch": 0.4337983177229364, + "grad_norm": 0.2177734375, + "learning_rate": 0.001056177236371069, + "loss": 0.1318, + "step": 49974 + }, + { + "epoch": 0.4338069982031406, + "grad_norm": 0.1298828125, + "learning_rate": 0.0010561475092412628, + "loss": 0.0698, + "step": 49975 + }, + { + "epoch": 0.43381567868334475, + "grad_norm": 0.283203125, + "learning_rate": 0.001056117782159413, + "loss": 0.126, + "step": 49976 + }, + { + "epoch": 0.43382435916354894, + "grad_norm": 0.09521484375, + "learning_rate": 0.0010560880551255525, + "loss": 0.1245, + "step": 49977 + }, + { + "epoch": 0.4338330396437531, + "grad_norm": 0.384765625, + "learning_rate": 0.0010560583281397134, + "loss": 0.1094, + "step": 49978 + }, + { + "epoch": 0.4338417201239573, + "grad_norm": 0.21484375, + "learning_rate": 0.0010560286012019282, + "loss": 0.0786, + "step": 49979 + }, + { + "epoch": 0.4338504006041614, + "grad_norm": 0.412109375, + "learning_rate": 0.0010559988743122292, + "loss": 0.1162, + "step": 49980 + }, + { + "epoch": 0.4338590810843656, + "grad_norm": 0.072265625, + "learning_rate": 0.0010559691474706495, + "loss": 0.0786, + "step": 49981 + }, + { + "epoch": 0.43386776156456974, + "grad_norm": 0.453125, + "learning_rate": 0.0010559394206772215, + "loss": 0.1123, + "step": 49982 + }, + { + "epoch": 0.43387644204477394, + "grad_norm": 0.08642578125, + "learning_rate": 0.001055909693931977, + "loss": 0.054, + "step": 49983 + }, + { + "epoch": 0.4338851225249781, + "grad_norm": 0.287109375, + "learning_rate": 0.0010558799672349488, + "loss": 0.0918, + "step": 49984 + }, + { + "epoch": 0.43389380300518227, + "grad_norm": 0.39453125, + "learning_rate": 0.00105585024058617, + "loss": 0.1064, + "step": 49985 + }, + { + "epoch": 0.4339024834853864, + "grad_norm": 0.93359375, + "learning_rate": 0.0010558205139856727, + "loss": 0.0991, + "step": 49986 + }, + { + "epoch": 0.4339111639655906, + "grad_norm": 0.12255859375, + "learning_rate": 0.0010557907874334893, + "loss": 0.085, + "step": 49987 + }, + { + "epoch": 0.43391984444579473, + "grad_norm": 0.07763671875, + "learning_rate": 0.0010557610609296527, + "loss": 0.0771, + "step": 49988 + }, + { + "epoch": 0.4339285249259989, + "grad_norm": 0.09765625, + "learning_rate": 0.001055731334474195, + "loss": 0.1021, + "step": 49989 + }, + { + "epoch": 0.43393720540620306, + "grad_norm": 0.9609375, + "learning_rate": 0.001055701608067149, + "loss": 0.1807, + "step": 49990 + }, + { + "epoch": 0.43394588588640726, + "grad_norm": 1.1484375, + "learning_rate": 0.0010556718817085469, + "loss": 0.1123, + "step": 49991 + }, + { + "epoch": 0.4339545663666114, + "grad_norm": 0.248046875, + "learning_rate": 0.0010556421553984208, + "loss": 0.0737, + "step": 49992 + }, + { + "epoch": 0.4339632468468156, + "grad_norm": 0.353515625, + "learning_rate": 0.0010556124291368043, + "loss": 0.0889, + "step": 49993 + }, + { + "epoch": 0.4339719273270197, + "grad_norm": 1.1171875, + "learning_rate": 0.0010555827029237294, + "loss": 0.1445, + "step": 49994 + }, + { + "epoch": 0.4339806078072239, + "grad_norm": 0.125, + "learning_rate": 0.0010555529767592286, + "loss": 0.0854, + "step": 49995 + }, + { + "epoch": 0.43398928828742805, + "grad_norm": 0.69140625, + "learning_rate": 0.0010555232506433343, + "loss": 0.1157, + "step": 49996 + }, + { + "epoch": 0.43399796876763225, + "grad_norm": 0.1455078125, + "learning_rate": 0.0010554935245760794, + "loss": 0.1064, + "step": 49997 + }, + { + "epoch": 0.4340066492478364, + "grad_norm": 0.294921875, + "learning_rate": 0.001055463798557496, + "loss": 0.0938, + "step": 49998 + }, + { + "epoch": 0.4340153297280406, + "grad_norm": 0.1435546875, + "learning_rate": 0.0010554340725876165, + "loss": 0.1113, + "step": 49999 + }, + { + "epoch": 0.4340240102082447, + "grad_norm": 0.5234375, + "learning_rate": 0.0010554043466664738, + "loss": 0.0613, + "step": 50000 + } + ], + "logging_steps": 1, + "max_steps": 96000, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 5000, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.1221338733472563e+18, + "train_batch_size": 12, + "trial_name": null, + "trial_params": null +}